Skip to main content
NCBI home page
Carcinogenesis logoLink to Carcinogenesis
. 2008 Sep 26;29(12):2425–2431. doi: 10.1093/carcin/bgn225

Association of hsp90 to the hTERT promoter is necessary for hTERT expression in human oral cancer cells

Reuben H Kim 1,2,3, Roy Kim 1, Wei Chen 1, Shen Hu 1,2, Ki-Hyuk Shin 1,2,3, No-Hee Park 1,2,3,4, Mo K Kang 1,2,3,*
PMCID: PMC2639246  PMID: 18820283

Abstract

Enhanced expression of human telomerase reverse transcriptase (hTERT) occurs frequently during cellular immortalization. The current study was undertaken to determine the mechanism regulating the hTERT promoter activity during cellular immortalization of human oral keratinocytes. Normal human oral keratinocytes (NHOKs) were immortalized with Bmi-1 and the E6 oncoprotein of human papillomavirus type 16 to establish the telomerase-positive HOK-Bmi-1/E6 cell line. Using DNA–protein-binding assay, we found that heat shock protein 90 (hsp90) physically interacts with the hTERT promoter in vitro. The hsp90 interaction with the promoter was detected more strongly in the telomerase-positive HOK-Bmi-1/E6 cells compared with that in senescing NHOK. Chromatin immunoprecipitation confirmed the in vivo interaction between hsp90 and the hTERT promoter in SCC4 cells, a telomerase-positive oral cancer cell line, but not in the NHOK. To determine the physiological significance of this interaction, SCC4 cells were exposed to geldanamycin (GA), a competitive inhibitor of hsp90. GA exposure led to decrease in telomerase activity, hTERT promoter activity and hTERT messenger RNA expression in SCC4 cells, even in the absence of de novo protein synthesis. Also, it abolished the in vivo interaction of the hTERT promoter region with hsp90 but not with Sp1 or c-Myc. These results indicate that physical interaction between hsp90 and the hTERT promoter occurs in telomerase-positive cells but not in normal human cells and is necessary for the enhanced hTERT expression and telomerase activity in cancer cells.

Introduction

Carcinogenesis is a multistep process that progresses with sequential escape from the tumor-suppressive mechanisms, including senescence and crisis. Cellular senescence occurs as a result of telomere shortening or extrinsic stress (1). Loss of telomere length during senescence mimics cellular DNA damage response, and continued replication beyond senescence leads to genetic catastrophe and massive cell death during crisis (2). Aberrant cells can escape from crisis by activation of telomerase, which performs de novo synthesis of telomeres and protects the telomeric ends by capping (3,4). Accordingly, the vast majority of human cancers demonstrate readily detectable telomerase activity, whereas most normal cells lack the enzyme activity (5). Telomerase activation occurs during cellular immortalization primarily in cells escaping from the crisis checkpoint (6). Telomerase activity is also required for cancer cell survival, suggesting a possible anticancer therapy by targeting telomerase (7). Therefore, detailed understanding of telomerase regulation is needed to elucidate the mechanism of carcinogenesis and to innovate a novel approach against cancer.

Telomerase activity is closely associated with the expression of human telomerase reverse transcriptase, hTERT, but not the RNA template of telomerase (8,9). Attempts to understand the mode of telomerase activation during carcinogenesis has thus been focused on the regulation of the hTERT expression and the cognate promoter activity. The hTERT promoter activity has been specifically associated with cancers and used widely for transgene expression selectively in cancer cells but not in telomerase-negative normal cells (10). The proximal hTERT promoter has been partially characterized to encompass numerous putative binding sites for transcriptional regulators, including c-Myc and Sp1 (11). Subsequent studies have shown the regulatory effects of c-Myc/Max heterodimers, HIF-2α, Cbfa1, AP-2β and Sp1/Sp3 on the hTERT promoter activity presumably through direct physical interaction with the DNA element (12–14). However, the detailed mode of hTERT promoter activation during immortalization and carcinogenesis remains to be elucidated.

The current study was undertaken to identify the cellular factors that physically interact with the hTERT promoter specifically in cancer but not in normal cells. We developed and utilized the promoter magnetic precipitation (PMP) assay coupled with mass spectrometry (MS) to identify putative transregulatory proteins that physically associate with the hTERT promoter in vitro. Using this approach, we found that heat shock protein 90 (hsp90) interacts with the hTERT promoter preferentially in the immortalized and cancerous oral epithelial cells compared with senescing cultures of normal human oral keratinocytes (NHOKs), which do not express telomerase. Hsp90 is an adenosine triphosphate-dependent molecular chaperone that binds to numerous client proteins, such as mutant p53, Raf-1 and Akt, HER2/Neu (Erb2) (15–18). It also forms a stable complex with human telomerase and facilitates synthesis of telomeres from telomerase activity, adding its important role in regulation of telomerase activity at the post-translational level (19–21). Sustained activity of hsp90 is closely linked to cancer cell survival and resistance to apoptosis (22), making hsp90 a novel and efficacious target for anticancer chemotherapy (23,24). Thus, involvement of hsp90 in cancer cell viability is typically attributed to its chaperonic activities preferentially for the proteins distinctly implicated in oncogenesis.

In this report, we provide evidence for transcriptional involvement of hsp90 at the hTERT promoter during immortalization of NHOK. Interaction between hsp90 and the hTERT promoter in vivo was confirmed by chromatin immunoprecipitation (ChIP) in SCC4 cells, a telomerase-positive oral squamous cell carcinoma (OSCC) cells, but not in senescing NHOK. This interaction was found to be necessary for the maximal hTERT promoter activity in cancer cells since exposure to geldanamycin (GA), a competitive inhibitor of hsp90 (25), resulted in diminution of the hTERT promoter activity, hTERT messenger RNA (mRNA) expression and telomerase activity in SCC4 cells. These effects of GA are presumed to be associated specifically with inhibition of hsp90 activity because it caused reduction in hTERT mRNA expression in the absence of de novo protein synthesis. Also, GA abolished the in vivo interaction between hsp90 and the hTERT promoter while other known hTERT promoter-binding factors, such as Sp1 and c-Myc, remained unaffected. These results indicate a new mechanism by which hsp90 contributes to carcinogenesis.

Experimental procedures

Cells and cell culture

Primary NHOK was prepared from keratinized oral epithelial tissues according to the methods described earlier (26). Isolated single cells were seeded onto collagen-treated flasks and cultured in keratinocyte growth medium containing low level of calcium (0.15 mM) and a supplementary bullet kit (Lonza, Allendale, NJ). The primary NHOK was maintained until the cells arrived at the senescence stage, being passaged at every 60–70% confluency level. Actively proliferating NHOK was serially infected with retroviral vectors expressing full-length Bmi-1 or human papillomavirus type 16 E6 to establish HOK-Bmi-1/E6 cells, which were serially passaged to establish immortalized cell populations. Detailed methods for retroviral vector construction, infection, selection and cellular immortalization of NHOK can be found in our previous report (6). SCC4, SCC9 and SCC15 cells were cultured as described elsewhere (27).

PMP–MS analysis of the hTERT promoter-binding proteins

Biotinylated hTERT promoter (568 bp from −544 to +5 with flanking sequences) was polymerase chain reaction (PCR) amplified using the forward primer 5′-CTCCGTCCTCCCCTTCAC-3′ and the reverse primer 5′-ATCGATCAGCGCTGCCTGAAACTC-3′ with biotin conjugated on 5′ end of the forward primer. The hTERT promoter fragments were separated in 1% agarose gel and then purified using the gel extraction kit (Qiagen, Valencia, CA). The hTERT promoter fragment (0.5 μg) was preincubated with 40 μl of Dynabeads (Invitrogen, Carlsbad, CA) for 3 h. Nuclear extracts were incubated with the hTERT promoter fragment for 3 h in the binding buffer (0.01% non-iodet P-40 10 mM Tris at pH 7.5, 50 mM KCl, 5 mM MgCl2, 2 mM dithiothreitol (DTT) and 20% glycerol). The protein–DNA complexes were washed three times with washing buffer, eluted by heating the samples at 94°C and loaded onto 10–20% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) gradient gel. The gel was stained with Sypro Ruby (Bio-Rad, Hercules, CA), and the image was taken using ChemiDoc XRS System (Bio-Rad). The bands of interest were excised and subjected to the protein identification.

For the regional analysis of the hTERT promoter, the full-length promoter was amplified into six overlapping regions using specific primer sets. The sequence of the primers and PCR conditions can be available upon request. The PMP analysis of the different hTERT promoter regions was performed as described above using the nuclear extracts of SCC4 cells. The promoter-binding proteins were precipitated by magnetic fields and separated by SDS–PAGE. Western blotting was then performed to detect the abundance of hsp90, hsp90α, hsp90β and Sp1 among the promoter-binding protein complexes. The following antibodies were used: hsp90 (Axxora, LLC, San Diego, CA); hsp90α and hsp90β (Stressgen, Victoria, BC Canada) and Sp1 (Santa Cruz Biotechnology, Santa Cruz, CA).

In-gel digestion and liquid chromatography–tandem mass spectrometry identification

Proteins in each gel bands were reduced with 10 mM DTT (1 h at 60°C), alkylated with 50 mM iodoacetamide (45 min at 45°C) and finally digested with 10 ng trypsin at 37°C for overnight. The resulting peptides were extracted with 50% acetonitrile/1% trifluoroacetic acid and subsequently identified by liquid chromatography–tandem mass spectrometry.

Liquid chromatography–tandem mass spectrometry analysis of peptides was performed using capillary liquid chromatography (LC Packings, Sunnyvale, CA) with quadrupole time-of-flight MS (Applied Biosystems, QSTAR XL, Foster City, CA). The samples were first loaded onto a LC Packings PepMap C18 precolumn (300 μm × 1 mm; particle size 5 μm) and then injected onto a LC Packings PepMap C18 column (75 μm × 150 mm; particle size 5 μm) for nano-liquid chromatography separation at a flow rate of 180 nl/min. The eluents used for liquid chromatography separation were (i) 0.1% formic acid and (ii) 95% acetonitrile/0.1% formic acid and 1%/min gradient. The acquired MS/MS data were searched against the International Protein Index human protein database using Mascot (Matrix Science, Boston, MA) database searching engine. Positive protein identification was based on standard Mascot criteria for statistical analysis of liquid chromatography–tandem mass spectrometry data.

Western blotting

To determine the level of hsp90 expression in normal and cancerous oral epithelial cells, whole-cell extracts from NHOK cultures expressing Bmi-1 and/or E6 were fractionated by SDS–PAGE and transferred to Immobilon protein membrane (Millipore, Billerica, MA). Immobilized membrane was incubated with primary antibodies, i.e. hsp90 (Axxora, LLC), lamin B1 (H-90) and β-actin (I-19, Santa Cruz Biotechnology), and probed with the respective secondary antibodies.

Telomeric repeat amplification protocol assay

Telomerase activity was determined by the TRAP-eze Telomerase Detection Kit (Chemicon, Temecula, CA) according to the manufacturer's guidelines. Cell lysate was extracted using 1× cholamidopropyl-dimethylammonium-propanesulfonate buffer, aliquoted and frozen in liquid nitrogen for subsequent telomeric repeat amplification protocol assay.

The telomerase reaction mixture was prepared by adding the cell lysate containing 1 μg of cellular protein to 24 μl solution comprising 1× telomeric repeat amplification protocol reaction buffer. The telomerase reaction product was allowed for 30 min at 30°C and amplified by PCR using a DNA Thermal Cycler (Perkin-Elmer, Foster City, CA). The following conditions were used for PCR: 32 cycles at 94°C for 30 s and 59°C for 30 s, followed by one delayed extension cycle at 72°C for 10 min. The PCR products were electrophoresed in 12.5% non-denaturing polyacrylamide gel in 1× Tris–borate ethylenediaminetetraacetic acid for 90 min at 60 W. After drying the gel, the radioactive signal was detected by Phosphor Imager (Molecular Dynamics, Sunnyvale, CA).

Semiquantitative reverse transcription–PCR

Total RNA was isolated from the cultured cells using Trizol™ reagent (Invitrogen) and was subjected to RNases-free DNase I digestion at 37°C for 2 h to eliminate any genomic DNA contamination. DNA-free total RNA (5 μg) was dissolved in 15 μl diethylpyrocarbonate-treated water, and the reverse transcription reaction was performed in first-strand buffer (Invitrogen) containing 300 U Superscript II (Invitrogen); 10 mM DTT and 0.5 μg random hexamer (Promega Corporation, Madison, WI) and 125 μM deoxynucleoside triphosphates. The annealing reaction was carried out for 5 min at 65°C, and complementary DNA synthesis was performed for 2 h at 37°C, followed by incubation for 15 min at 70°C to stop the enzyme reaction. The reverse transcription product was diluted with 70 μl H2O. The following primers were used for PCR amplification: glyceraldehyde 3-phosphate dehydrogenase (GAPDH) primers, 5′-GACCCCATTGACCTCAAC-3′ (forward) and 5′-CTTCTCCATGGTGGTGAAGA-3′ (reverse); hTERT primers, 5′-GCCTGAGCTGTACTTTGTCAA-3′ (forward) and 5′-CGCAAACAGCTTGTTCTCCATGTC-3′ (reverse); proliferating cell nuclear antigen (PCNA) primers, 5′-AGGGCTCCATCCTCAAGAAG-3′ (forward) and 5′-CTCCTGGTTTGGTGCTTCAA-3′ (reverse) and transcription factor II D (TFIID) primers, 5′-CCACAGCCTATTCAGAAC-3′ (forward) and 5′-GCTCCTGTGCACACCAT-3′ (reverse) (28). PCR amplification was quantiated using Scion Image software (National Institutes of Health, Bethesda, MD).

Analysis of the hTERT promoter activity

A pGL3B-TRTP containing the 1670 bp fragment (−1665 to +5) of the hTERT promoter upstream of the firefly luciferase gene in the pGL3-basic (Promega Corporation), kindly provided by Dr J.Carl Barrett (National Institute of Environmental Health Science, Bethesda, MD), was used for the hTERT reporter assay. Prior to transfection, a six-well plate with 5 × 104 cells per well were plated and cultured for 24 h. The pGL3B-TRTP vector (1 μg per well) was introduced into the cells using Lipofectin reagent (Invitrogen). To control the differences in transfection efficiencies, pRL-SV40 plasmid (0.001 μg per well) containing Renilla luciferase gene under SV40 enhancer/promoter was also introduced into the cells. After 24 h post-transfection, the cells were exposed to 0, 0.01, 0.1 or 1 μM GA for 24 h and harvested by trypsinization. The cell lysates were prepared using Dual-Luciferase Reporter assay system (Promega Corporation). Luciferase activities from triplicated samples were measured using a luminometer.

ChIP assay

ChIP assay was performed using the ChIP assay kit (Upstate Technology, Lake Placid, NY) according to the manufacturer's instructions. Briefly, cells were fixed at room temperature for 10 min in the culture medium containing 1% formaldehyde. Cells were harvested and sonicated with four bursts of 10 s each followed by 1 min rest in ice. The sonicated samples were pre-cleared with protein G-agarose to reduce non-specific binding and incubated with the antibodies overnight. The following antibodies were used: hsp90 (Axxora, LLC); hsp90α and hsp90β (Stressgen) and c-myc and Sp1 (Santa Cruz Biotechnology). Protein–DNA complexes were then washed with washing buffers, and the cross-links were reversed by incubating at 65°C for 6 h. The samples were then treated with RNase A for 30 min at 37°C followed by proteinase K treatment at 45°C for 2 h. DNA fragments were eluted using PCR purification kit (Qiagen). To amplify the hTERT promoter fragment, 2 μl was used for PCR amplification with primers (forward, 5′-GGCCGGGCTCCCAGTGGATTC-3′ and reverse, 5′-CAGCGGGGAGCGCGCGGCATCG-3′).

Results

Hsp90 interacts with the hTERT promoter in vitro preferentially in immortalized and cancerous oral epithelial cells

In order to identify the cellular factors that interact with the hTERT promoter region, we developed PMP–MS assay, which allows for high throughput and global screening of hTERT promoter-binding proteins. Primary NHOK was infected serially with retroviral vectors capable of expressing Bmi-1 (RV-Bmi-1) and E6 of human papillomavirus type 16 (RV-16E6), which cooperate to immortalize NHOK (6). The cells infected with the two retroviral vectors were named HOK-Bmi-1/E6 and maintained until they were immortalized. Cellular extracts from senescing NHOK at population doublings (PDs) 20, HOK-Bmi-1/E6 at PD 36 (pre-crisis) and HOK-Bmi-1/E6 at PD 73 (post-crisis) were isolated and incubated with the 549 bp proximal region of the hTERT promoter (11) immobilized onto Dynabeads®. The promoter-bound proteins were fractionated by SDS–PAGE and visualized by Coomasie staining (Figure 1A).

Fig. 1.

Fig. 1.

Open in a new tab

Hsp90 associates with the hTERT promoter in telomerase-expressing cells. (A) Whole-cell extracts from senescing NHOK at PD 20, HOK-Bmi-1/E6 (PD 36, pre-crisis) and HOK-Bmi-1/E6 (PD 73, post-crisis) were mixed with the proximal hTERT promoter DNA immobilized onto Dynabeads®. The negative control indicates a sample incubated with Dynabeads® without the hTERT promoter. The protein–DNA complex was enriched and isolated by magnetic field and separated by SDS–PAGE. The proteins that coprecipitate with the hTERT promoter were detected in gel by Coomasie staining. Individual bands were excised and identified by MS as indicated in the figure. Some proteins were identified by the theoretical molecular mass (*), which may be different from the observed electrophoretic migration pattern. (B) ChIP was performed with a senescing culture of NHOK at PD 18 and rapidly proliferating SCC4, SCC9 and RKO cells. The cells were exposed to formalin for 10 min to cross-link the DNA–protein complexes. After fragmentation of genomic DNA by sonication, the DNA–protein complexes were precipitated by hsp90 antibody. The presence of the hTERT promoter element was determined by PCR. (C) Whole-cell extracts from actively proliferating (PD 14) or senescent (PD 20) NHOK, HOK-Bmi-1/E6 (PDs 43 and 79, representing pre- and post-crisis cells, respectively), SCC4, SCC9 and SCC15 cells were obtained and fractionated by SDS–PAGE. Intracellular hsp90 protein level was determined by western blotting. (D) Nuclear and cytosolic proteins were isolated from senescent NHOK (PD 20), SCC4 and SCC9. Western blotting was performed for hsp90. Lamin B1 was used as nuclear loading control.

During immortalization, HOK-Bmi-1/E6 cells do not express any detectable telomerase activity until the cells overcome crisis (6). Thus, the proteins that differentially bind the hTERT promoter between the pre- and post-crisis HOK-Bmi-1/E6 may represent the proteins involved in the hTERT/telomerase activation during crisis. For this reason, we identified the proteins that were differentially associated with the hTERT promoter in the pre- and post-crisis cells. One of the bands (∼100 kDa) strongly present in the post-crisis cells but not in the pre-crisis or the parental NHOK corresponded to hsp90. We identified the protein by using in-gel tryptic digestion and subsequent MS/MS analysis of the resulting peptides using highly accurate quadrupole time-of-flight MS. Mascot database searching revealed five peptides matched with a total protein score of 213, indicating that the protein was confidently identified. We also identified the other proteins that coprecipitated with the hTERT promoter by the PMP assay. These proteins included pyruvate carboxylase mitochondrial precursor, 70 kDa protein (the number indicating theoretical mass), M6a methyltransferase, tubulin α-3, tubulin β-2, 50 kDa protein and 48 kDa protein (Figure 1A).

We confirmed the binding of hsp90 onto the hTERT promoter region in vivo by ChIP assay using the cell lysates of OSCC cells (SCC4 and SCC9) and RKO cells (human colorectal cancer cells). These OSCC cells express high levels of telomerase activity (29). Senescing NHOK (PD 18), which do not express telomerase activity (29,30), were included for comparison. The ChIP assay revealed amplification of the hTERT promoter sequences in the immunoprecipitates of SCC4, SCC9 and RKO cells but not of senescing NHOK (Figure 1B). These results confirm in vivo binding of hsp90 to the hTERT promoter sequences in telomerase-positive OSCC cells.

We next sought the possibility that the differential binding patterns of hsp90 to the hTERT promoter in the pre- and post-crisis cells resulted from altered expression and localization of hps90 in cells. This possibility was tested by comparing the level of hsp90 expression in normal and cancerous oral epithelial cells showing differential telomerase activity. As shown in Figure 1C, whole-cell extracts of proliferating (PD 14) and senescent (PD 20) NHOK, HOK-Bmi-1/E6, SCC4, SCC9 and SCC15 contained very similar levels of hsp90 protein. When we compared the level of hsp90 in nuclear and cytosolic fractions of NHOK, SCC9 and SCC15, we did not find any significant difference in the level of hsp90 expression although cytosolic hsp90 appears to be more abundant than that of the nuclear fractions in all tested samples (Figure 1D). Interestingly, a unique signal representing a smaller isoform of hsp90 was noted in the cytosolic fractions of the OSCC cells but not of NHOK. These data indicate that the subcellular localization and the expression level of hsp90 are not associated with its preferential binding to the hTERT promoter in telomerase-expressing cells.

Binding pattern of Hsp90 to the hTERT promoter region

To further characterize the association between hsp90 and hTERT promoter, we dissected the 549 bp hTERT fragment into six overlapping regions and determined the degree of hsp90 binding to each region by PMP and ChIP analyses (Figure 2A). The PMP analysis using three different antibodies against hsp90 consistently revealed the binding of hsp90 to the two different regions of hTERT promoter. The strongest interaction occurred in the region 5 (from −465 to −341), whereas a moderate level of interaction was noted in the regions 2 and 1 (both covering from −188 to +5). Region 3 (from −260 to −157) showed little or no detectable binding of hsp90. We compared this pattern of hsp90 binding to that of Sp1, a transcription factor known to interact with the hTERT promoter (31). Interestingly, Sp1 also demonstrated the binding pattern similar to that of hsp90 (Figure 2B). Similar binding pattern along the hTERT promoter was also observed by the ChIP analysis that revealed the hsp90 interaction with the hTERT promoter regions in vivo (Figure 2C). These results indicate that hsp90 associates with the hTERT promoter at multiple sites that resemble those of Sp1.

Fig. 2.

Fig. 2.

Open in a new tab

Binding pattern of hsp90 to the hTERT promoter. (A) The hTERT promoter (−544 to +5) was amplified into six overlapping fragments using the primer sets indicated. (B) PMP was performed with the hTERT promoter fragments (regions 1 to 6) using nuclear extracts from SCC4 cells. The precipitated proteins were then transferred onto the membranes and probed for the presence of hsp90, hsp90α, hsp90β and Sp1 by western blotting. (C) ChIP was performed with SCC4 cells for hsp90 using the same primer sets as shown in (A). The upper panel shows the amplification using the immunoprecipitates, and the lower panel shows the input control.

Chemical inhibition of hsp90 results in decrease of the hTERT promoter activity, mRNA expression level and telomerase activity in OSCC cells

To determine the physiological roles of hsp90 in regulation of hTERT expression and telomerase activity, we utilized GA, a specific inhibitor of hsp90 (25), to perturb the normal function of hsp90 in SCC4 cells. SCC4 cells were treated with varying doses of GA (0, 0.01, 0.1 and 1 μM) for 24 h or at 0.1 μM for varying duration of exposure (0, 6, 12 and 24 h). hTERT mRNA level was determined by semiquantitative reverse transcription–PCR (Figure 3A), which showed reduced hTERT mRNA expression in dose- and time-dependent manner upon treatment with GA. To determine the specificity of the GA-mediated inhibition of hTERT expression, we compared the relative abundance of hTERT mRNA to those of TFIID, a component of general transcriptional machinery (32), PCNA, a cell proliferation marker (33), or GAPDH. Exposure of cells to GA resulted in preferential reduction in the hTERT mRNA expression with respect to those of TFIID, PCNA or GAPDH in dose- and time-dependent manner. This result does not support the possibility that the reduced hTERT mRNA expression after GA treatment is a consequence of non-specific cytotoxicity of GA. Also, since GA exposure did not alter the expression level of PCNA, the reduced hTERT mRNA expression does not appear to result from altered cell proliferation during the first 24 h of exposure.

Fig. 3.

Fig. 3.

Open in a new tab

Inhibition of hsp90 results in reduction of hTERT promoter activity and mRNA expression in the presence or absence of de novo protein synthesis. (A) SCC4 cells were exposed to 0, 0.01, 0.1 or 1 μM GA for 24 h or to 0.1 μM GA for varying duration (0, 6, 12 or 24 h). Amount of hTERT mRNA was determined by semiquantitative reverse transcription–PCR. hTERT, PCNA, TFIID and GAPDH were amplified from the same complementary DNA samples. The signal strength of the hTERT amplification was normalized and plotted individually against those of TFIID and GAPDH. (B) SCC4 cells were pretreated with 10 μg/ml CHX starting at 30 min prior to exposure to GA (0.1 and 1 μM) for 24 h. The cells were harvested and reverse transcription–PCR was performed for hTERT and GAPDH amplification. Normalized hTERT signal strength was plotted. (C) SCC4 cells were transfected with pGL3B-TRTP plasmid containing firefly luciferase gene under the control of the hTERT promoter. For normalization of transfection efficiencies, the cells were also transfected with pRL-SV40 containing the Renilla luciferase gene under SV40 promoter. Subsequent to transfection, the cells were exposed to 0, 0.01, 0.1 or 1 μM GA for 24 h and harvested for luminometric analysis. The graph shows the mean of triplicated relative firefly luciferase light units normalized against those of Renilla luciferase activity. Error bar represents the standard deviation.

Since it is possible that the reduction of hTERT expression was mediated by other proteins than hsp90 in the GA-treated cells, we pretreated SCC4 cells with cycloheximide (CHX) before exposure to GA. We found that the inhibitory effects of the GA treatment on hTERT mRNA expression was similar with and without the CHX pretreatment (Figure 3B). When we measured the hTERT promoter activity in the SCC4 cells exposed to 0.01, 0.1 and 1 μM GA for 24 h, GA exposure led to a dose-dependent decrease in the hTERT promoter activity (Figure 3C). Furthermore, the decreased hTERT mRNA expression with GA exposure was mirrored in the reduced telomerase activity in the GA-exposed SCC4 cells in time- and dose-dependent manner (Figure 4). Therefore, hsp90 inhibition by GA led to decreased hTERT expression in the absence of de novo protein synthesis in SCC4 cells, resulting in reduced telomerase activity.

Fig. 4.

Fig. 4.

Open in a new tab

Telomerase activity is reduced in SCC4 cells after exposure to GA. SCC4 cells were exposed to 0, 0.01, 0.1 or 1 μM GA and harvested after 24, 48 and 72 h of drug exposure to determine telomerase activity by telomeric repeat amplification protocol assay. Band intensities per individual samples were collectively quantitated by densitometric analysis and indicated in relative values against the control group (dimethyl sulfoxide), as shown at the bottom of the figure. Internal telomeric amplification standard (ITAS).

Inhibition of hsp90 results in the loss of its association with the hTERT promoter in vivo

Based on the above results, hsp90 activity appears to be necessary for intact hTERT mRNA expression and telomerase activity in SCC4 cells. In the following experiment, we tested the possibility that the reduced hTERT mRNA expression by GA treatment resulted from the loss of hsp90 binding to the hTERT promoter. When the cells were exposed to 0.1 μM GA for 24 h, we observed ∼50% reduction in hTERT transcript level while only ∼10% of telomerase activity was reduced (Figures 3A and C and 4), owing to the long half-life of telomerase complex (34). Therefore, SCC4 cells were exposed to 0.1 μM GA for 24 h, and ChIP assay was performed to detect any changes in hsp90 binding to the promoter region. For comparison, we also included Sp1 and c-myc proteins, which are already known to interact with the hTERT promoter (35,36). After exposure to GA, the interaction between hsp90 and the hTERT promoter in vivo was completely abolished while that of Sp1 and c-myc was marginally affected (Figure 5). This result suggests that GA exposure led to reduction in the hTERT expression through loss of interaction between hsp90 and the hTERT promoter in cells.

Fig. 5.

Fig. 5.

Open in a new tab

Inhibition of hsp90 by GA exposure leads to loss of its association with the hTERT promoter complex. ChIP assay was performed with SCC4 cells cultured in the presence or absence of 0.1 μM GA for 24 h. The genomic DNAs were fractionated by sonication after formaldehyde cross-linking, and the DNA–protein complexes were precipitated using specific antibodies against hsp90, Sp1 or c-Myc. Presence of the hTERT promoter elements were determined by PCR and agarose gel electrophoresis.

Discussion

In the present study, we examined the involvement of hsp90 in the regulation of the hTERT promoter activity by physical association with the DNA–protein complex. This was accomplished by identification of hsp90 among the proteins that coprecipitated with the hTERT promoter by the PMP–MS assay, confirming the in vivo interaction between hsp90 and the hTERT promoter preferentially in cells expressing high telomerase activity and the perturbation experiments that showed reduction of the promoter activity and gene expression upon inhibition of hsp90. Use of the hsp90-specific inhibitor, GA, revealed the requirement of the physical association of hsp90 with the promoter to maintain the high level of the hTERT promoter activity and the gene expression in OSCC cells.

The significance of this study lies in that the PMP–MS assay was shown to be a useful tool to identify novel transregulatory factors for the hTERT promoter through global and high throughput screening of the DNA-binding proteins. Similar approaches such as the streptoavidin–agarose pulldown assays have successfully been used to identify the promoter-binding and transregulatory factors (13,37). These approaches including the PMP–MS assay used herein can be tailored to pulldown the cellular proteins that physically associate with any DNA element of interest. With the recent advances in proteomics, these techniques can be developed into a powerful high throughput screening of DNA–protein complexes. Furthermore, the paramagnetic Dynabeads® used for the PMP assay allows for rapid and clean processing of the protein samples without the need for centrifugation so as to enhance the reliability of the assay.

We showed previously that hTERT/telomerase activation is required for cellular immortalization in cultured NHOK by overcoming crisis (6). Also, enhanced hTERT expression is observed in clinical samples during mild to moderate dysplastic cell transformation in situ (38). Thus, telomerase activation occurs very early during oral carcinogenesis. Hsp90 is one of the protein subunits of active telomerase required for the assembly of the enzyme complex in vitro and in vivo (39). A recent study showed that hsp82p, yeast homologue of hsp90, is required for binding of telomerase onto DNA and for elongation of telomeric sequences by telomerase (20). Also, hTERT was found to associate with human telomerase RNA, the RNA template, in the absence of functional hsp90 and that GA inhibits the loading of active telomerase onto telomeres (21). These studies suggest that physical association of hsp90 with hTERT promotes telomerase activity by increasing the affinity of the enzyme complex to telomeric ends. The current study revealed the second mechanism by which hsp90 enhances telomerase activity. Exposure of cells to GA led to preferential reduction of hTERT expression, whereas the mRNA levels of TFIID, PCNA and GAPDH remained unchanged (Figure 3A). Thus, the GA-mediated reduction of hTERT expression does not represent non-specific cytotoxicity of the drug. Hsp90 physically associates with the hTERT promoter complexes and enhances the promoter activity and the gene expression of hTERT. Interestingly, this occurs preferentially in telomerase-expressing cells, such as immortalized HOK-Bmi-1/E6 cells or OSCC cells, but not in senescing NHOK or pre-crisis HOK-Bmi-1/E6 cells, which do not express telomerase (6). Thus, hsp90 association with the hTERT promoter complex may, in part, be responsible for telomerase activation during cellular immortalization. The current study suggests a possible mechanism by which GA exerts its anticancer effects through inhibition of hsp90 association with the hTERT promoter, resulting in the loss of hTERT mRNA expression.

Involvement of hsp90 in transcriptional regulation has also been demonstrated in other systems such as the lipoxygenase promoter, to which hsp90 was recruited by Sp1, a positive regulator of the promoter activity (37). The authors of this study showed that hsp90 enhanced the physical association of Sp1 to the promoter, leading to enhanced lipoxygenase promoter activity. Inhibition of hsp90 with GA almost completely abolished the binding of Sp1 to the lipoxygenase promoter in vivo as revealed by the ChIP assay. In our study, the association of Sp1 to the hTERT promoter was marginally altered by the GA exposure, although it led to the complete loss of hsp90 association with the promoter. This result does not support the possibility that hsp90 modulates the affinity of Sp1 to the hTERT promoter. However, the PMP–western analysis of the hTERT promoter regions revealed a similar pattern of promoter binding between hsp90 and Sp1 (Figure 2B), suggesting that hsp90 and Sp1 may occupy similar regions along the hTERT promoter. Since Sp1 is known to be a negative regulator of hTERT promoter (31,40,41), hsp90 may impose inhibitory effects onto Sp1 in hTERT promoter regulation. Therefore, the functional and perhaps the physical interactions between hsp90 and Sp1 appear to be different for the two promoter systems and need further investigation.

One of the important questions raised from this study is how hsp90 associates with the hTERT promoter preferentially in telomerase-positive cells. We have ruled out the possibility of differential expression or subcellular localization of hsp90 between the normal and cancerous cells (Figure 1). Also, we confirmed that purified recombinant hsp90 does not directly bind the hTERT promoter (data not shown). Thus, hsp90 is probably to associate with protein complexes assembled onto the hTERT promoter through protein–protein interactions that appears to be specific for telomerase-positive cells and subjected to inhibition by GA treatment. Recent findings suggest that hTERT expression may be repressed in normal somatic cells through enhancer-like elements located on distal regions of the hTERT promoter (42,43). While the current study demonstrates binding of hsp90 to the proximal regions of the promoter, it does not rule out the possibility of the interaction between hsp90 and the distal elements to enhance hTERT expression in OSCC cells. Although binding of hsp90 to the promoter is necessary for the intact hTERT expression in OSCC cells, it is unlikely to be the sole factor necessary to enhance hTERT expression and telomerase activity. The findings raise a possibility that hsp90 is recruited to the hTERT promoter by one of its client proteins selectively present on the promoter in cancerous cells but not in normal cells. With this hypothesis, identification of such client protein of hsp90 will allow us to understand the mechanism of hTERT/telomerase activation during oral carcinogenesis.

Hsp90 is known to be a ‘cancer chaperone’ whose clientele includes numerous aberrantly expressed proteins, such as mutants, overexpressed proteins, transcription factors and the proteins involved in growth signaling pathways (44). Hsp90 in normal cells interacts with the client proteins in low-affinity and dynamic manner, but in cancer cells it tightly binds the aberrant proteins to promote the oncogenic potential of its client proteins (45). Since we found no difference in the expression levels or localization of hsp90 between normal and cancerous cells, the mechanisms underlying the possible oncogenic potential of hsp90 remain to be determined. Interestingly, we note that the hsp90 binding pattern is different between NHOK and the OSCC cells, especially in the cytoplasmic fractions (Figure 1D). The lower band present in the OSCC cells may represent hsp90N, a recently identified isoform of hsp90 demonstrating oncogenic properties (46,47). Further studies are needed to determine the identity and the association of the second hsp90 band with oral carcinogenesis.

Funding

National Institutes of Dental and Craniofacial Research (NIDCR) from National Institutes of Health (K08DE17121 to R.H.K., K02DE18959 and R01DE18295 to M.K.K. and R01DE14147 to N.-H.P.).

Acknowledgments

Conflict of Interest Statement: None declared.

Glossary

Abbreviations

ChIP

chromatin immunoprecipitation

GA

geldanamycin

GAPDH

glyceraldehyde 3-phosphate dehydrogenase

hsp90

heat shock protein 90

hTERT

human telomerase reverse transcriptase

mRNA

messenger RNA

MS

mass spectrometry

NHOK

normal human oral keratinocyte

OSCC

oral squamous cell carcinoma

PCNA

proliferating cell nuclear antigen

PCR

polymerase chain reaction

PD

population doubling

PMP

promoter magnetic precipitation

SDS–PAGE

sodium dodecyl sulfate–polyacrylamide gel electrophoresis

TFIID

transcription factor II D

References

  • 1.Itahana K, et al. Mechanisms of cellular senescence in human and mouse cells. Biogerontology. 2004;5:1–10. doi: 10.1023/b:bgen.0000017682.96395.10. [DOI] [PubMed] [Google Scholar]
  • 2.Stewart SA, et al. Telomeres: cancer to human aging. Annu. Rev. Cell Dev. Biol. 2006;22:531–557. doi: 10.1146/annurev.cellbio.22.010305.104518. [DOI] [PubMed] [Google Scholar]
  • 3.Greider CW, et al. The telomere terminal transferase of Tetrahymena is a ribonucleoprotein enzyme with two kinds of primer specificity. Cell. 1987;51:887–898. doi: 10.1016/0092-8674(87)90576-9. [DOI] [PubMed] [Google Scholar]
  • 4.Blackburn EH, et al. Molecular manifestations and molecular determinants of telomere capping. Cold Spring Harb. Symp. Quant. Biol. 2006;65:253–263. doi: 10.1101/sqb.2000.65.253. [DOI] [PubMed] [Google Scholar]
  • 5.Kim NW, et al. Specific association of human telomerase activity with immortal cells and cancer. Science. 1994;266:2011–2015. doi: 10.1126/science.7605428. [DOI] [PubMed] [Google Scholar]
  • 6.Kim RH, et al. Bmi-1 cooperates with human papillomavirus type 16 E6 to immortalize normal human oral keratinocytes. Exp. Cell Res. 2007;313:462–472. doi: 10.1016/j.yexcr.2006.10.025. [DOI] [PubMed] [Google Scholar]
  • 7.Li S, et al. Rapid inhibition of cancer cell growth induced by lentiviral delivery and expression of mutant-template telomerase RNA and anti-telomerase short-interfering RNA. Cancer Res. 2004;64:4833–4840. doi: 10.1158/0008-5472.CAN-04-0953. [DOI] [PubMed] [Google Scholar]
  • 8.Kyo S, et al. Human telomerase reverse transcriptase as a critical determinant of telomerase activity in normal and malignant endometrial tissues. Int. J. Cancer. 1999;80:60–63. doi: 10.1002/(sici)1097-0215(19990105)80:1<60::aid-ijc12>3.0.co;2-e. [DOI] [PubMed] [Google Scholar]
  • 9.Yan P, et al. Expression of telomerase genes correlates with telomerase activity in human colorectal carcinogenesis. J. Pathol. 2001;193:21–26. doi: 10.1002/1096-9896(2000)9999:9999<::AID-PATH728>3.0.CO;2-G. [DOI] [PubMed] [Google Scholar]
  • 10.Schepelmann S, et al. Suicide gene therapy of human colon carcinoma xenografts using an armed oncolytic adenovirus expressing carboxypeptidase G2. Cancer Res. 2007;67:4949–4955. doi: 10.1158/0008-5472.CAN-07-0297. [DOI] [PubMed] [Google Scholar]
  • 11.Horikawa I, et al. Cloning and characterization of the promoter region of human telomerase reverse transcriptase gene. Cancer Res. 1999;59:826–830. [PubMed] [Google Scholar]
  • 12.Lou F, et al. The opposing effect of hypoxia-inducible factor-2α on expression of telomerase reverse transcriptase. Mol. Cancer Res. 2007;5:793–800. doi: 10.1158/1541-7786.MCR-07-0065. [DOI] [PubMed] [Google Scholar]
  • 13.Deng WG, et al. Tumor-specific activation of human telomerase reverses transcriptase promoter activity by activating enhancer-binding protein-2β in human lung cancer cells. J. Biol. Chem. 2007;282:26460–26470. doi: 10.1074/jbc.M610579200. [DOI] [PubMed] [Google Scholar]
  • 14.Isenmann S, et al. hTERT transcription is repressed by Cbfa1 in human mesenchymal stem cell populations. J. Bone Miner. Res. 2007;22:897–906. doi: 10.1359/jbmr.070308. [DOI] [PubMed] [Google Scholar]
  • 15.Csermely P, et al. The 90-kDa molecular chaperone family: structure, function, and clinical applications. A comprehensive review. Pharmacol. Ther. 1998;79:129–168. doi: 10.1016/s0163-7258(98)00013-8. [DOI] [PubMed] [Google Scholar]
  • 16.Zhao R, et al. Hsp90: a chaperone for protein folding and gene regulation. Biochem. Cell Biol. 2005;83:703–710. doi: 10.1139/o05-158. [DOI] [PubMed] [Google Scholar]
  • 17.Neckers L, et al. Heat-shock protein 90 inhibitors as novel cancer chemotherapeutic agents. Expert Opin. Emerg. Drugs. 2002;7:277–288. doi: 10.1517/14728214.7.2.277. [DOI] [PubMed] [Google Scholar]
  • 18.Whitesell L, et al. Hsp90 and the chaperoning of cancer. Nat. Rev. 2005;5:761–772. doi: 10.1038/nrc1716. [DOI] [PubMed] [Google Scholar]
  • 19.Forsythe HL, et al. Stable association of hsp90 and p23, but not hsp70, with active human telomerase. J. Biol. Chem. 2001;276:15571–15574. doi: 10.1074/jbc.C100055200. [DOI] [PubMed] [Google Scholar]
  • 20.Toogun OA, et al. The hsp90 molecular chaperone modulates multiple telomerase activities. Mol. Cell. Biol. 2008;28:457–467. doi: 10.1128/MCB.01417-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Keppler BR, et al. The biochemical role of the heat shock protein 90 chaperone complex in establishing human telomerase activity. J. Biol. Chem. 2006;281:19840–19848. doi: 10.1074/jbc.M511067200. [DOI] [PubMed] [Google Scholar]
  • 22.Lang SA, et al. Inhibition of heat shock protein 90 impairs epidermal growth factor-mediated signaling in gastric cancer cells and reduces tumor growth and vascularization in vivo. Mol. Cancer Ther. 2007;6:1123–1132. doi: 10.1158/1535-7163.MCT-06-0628. [DOI] [PubMed] [Google Scholar]
  • 23.Kim S, et al. Geldanamycin decreases Raf-1 and Akt levels and induces apoptosis in neuroblastomas. Int. J. Cancer. 2003;103:352–359. doi: 10.1002/ijc.10820. [DOI] [PubMed] [Google Scholar]
  • 24.Weigel BJ, et al. A phase I study of 17-allylaminogeldanamycin in relapsed/refractory pediatric patients with solid tumors: a Children's Oncology Group study. Clin. Cancer Res. 2007;13:1789–1793. doi: 10.1158/1078-0432.CCR-06-2270. [DOI] [PubMed] [Google Scholar]
  • 25.Whitesell L, et al. Inhibition of heat shock protein HSP90-pp60v-src heteroprotein complex formation by benzoquinone ansamycins: essential role for stress proteins in oncogenic transformation. Proc. Natl Acad. Sci. USA. 1994;91:8324–8328. doi: 10.1073/pnas.91.18.8324. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Kang MK, et al. In vitro replication and differentiation of normal human oral keratinocytes. Exp. Cell Res. 2000;258:288–297. doi: 10.1006/excr.2000.4943. [DOI] [PubMed] [Google Scholar]
  • 27.Kang MK, et al. Elevated Bmi-1 expression is associated with dysplastic cell transformation during oral carcinogenesis and is required for cancer cell replication and survival. Br. J. Cancer. 2007;96:126–133. doi: 10.1038/sj.bjc.6603529. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Sorli CH, et al. Basal expression of cyclooxygenase-2 and nuclear factor-interleukin 6 are dominant and coordinately regulated by interleukin 1 in the pancreatic islet. Proc. Natl Acad. Sci. USA. 1998;95:1788–1793. doi: 10.1073/pnas.95.4.1788. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Kang MK, et al. Replicative senescence of normal human oral keratinocytes is associated with the loss of telomerase activity without shortening of telomeres. Cell Growth Differ. 1998;9:85–95. [PubMed] [Google Scholar]
  • 30.Kang MK, et al. Senescence occurs with hTERT repression and limited telomere shortening in human oral keratinocytes cultured with feeder cells. J. Cell. Physiol. 2004;199:364–370. doi: 10.1002/jcp.10410. [DOI] [PubMed] [Google Scholar]
  • 31.Won J, et al. Sp1 and Sp3 recruit histone deacetylase to repress transcription of human telomerase reverse transcriptase (hTERT) promoter in normal human somatic cells. J. Biol. Chem. 2002;277:38230–38238. doi: 10.1074/jbc.M206064200. [DOI] [PubMed] [Google Scholar]
  • 32.Sawadogo M, et al. Factors involved in specific transcription by human RNA polymerase II: analysis by a rapid and quantitative in vitro assay. Proc. Natl Acad. Sci. USA. 1985;82:4394–4398. doi: 10.1073/pnas.82.13.4394. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Kubben FJ, et al. Proliferating cell nuclear antigen (PCNA): a new marker to study human colonic cell proliferation. Gut. 1994;35:530–535. doi: 10.1136/gut.35.4.530. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Holt SE, et al. Lack of cell cycle regulation of telomerase activity in human cells. Proc. Natl Acad. Sci. USA. 1997;94:10687–10692. doi: 10.1073/pnas.94.20.10687. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Takakura M, et al. Cloning of human telomerase catalytic subunit (hTERT) gene promoter and identification of proximal core promoter sequences essential for transcriptional activation in immortalized and cancer cells. Cancer Res. 1999;59:551–557. [PubMed] [Google Scholar]
  • 36.Kyo S, et al. Sp1 cooperates with c-Myc to activate transcription of the human telomerase reverse transcriptase gene (hTERT) Nucleic Acids Res. 2000;28:669–677. doi: 10.1093/nar/28.3.669. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Hung JJ, et al. Hsp90alpha recruited by Sp1 is important for transcription of 12(S)-lipoxygenase in A431 cells. J. Biol. Chem. 2005;280:36283–36292. doi: 10.1074/jbc.M504904200. [DOI] [PubMed] [Google Scholar]
  • 38.Kim HR, et al. Elevated expression of hTERT is associated with dysplastic cell transformation during human oral carcinogenesis in situ. Clin. Cancer Res. 2001;7:3079–3086. [PubMed] [Google Scholar]
  • 39.Holt SE, et al. Functional requirement of p23 and Hsp90 in telomerase complexes. Genes Dev. 1999;13:817–826. doi: 10.1101/gad.13.7.817. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Racek T, et al. C-terminal p73 isoforms repress transcriptional activity of the human telomerase reverse transcriptase (hTERT) promoter. J. Biol. Chem. 2005;280:40402–40405. doi: 10.1074/jbc.C500193200. [DOI] [PubMed] [Google Scholar]
  • 41.Fujiki T, et al. TAK1 represses transcription of the human telomerase reverse transcriptase gene. Oncogene. 2007;26:5258–5266. doi: 10.1038/sj.onc.1210331. [DOI] [PubMed] [Google Scholar]
  • 42.Ducrest A-L, et al. Regulation of the human telomerase reverse transcriptase gene. Oncogene. 2002;21:541–552. doi: 10.1038/sj.onc.1205081. [DOI] [PubMed] [Google Scholar]
  • 43.Szutorisz H, et al. A chromosome 3-encoded repressor of the human telomerase reverse transcriptase (hTERT) gene controls the state of hTERT chromatin. Cancer Res. 2003;63:689–695. [PubMed] [Google Scholar]
  • 44.Neckers L. Heat shock protein 90: the cancer chaperone. J. Biosci. 2007;32:517–530. doi: 10.1007/s12038-007-0051-y. [DOI] [PubMed] [Google Scholar]
  • 45.Rodina A, et al. Selective compounds define Hsp90 as a major inhibitor of apoptosis in small-cell lung cancer. Nat. Chem. Biol. 2007;3:498–507. doi: 10.1038/nchembio.2007.10. [DOI] [PubMed] [Google Scholar]
  • 46.Sreedhar AS, et al. Hsp90 isoforms: functions, expression and clinical importance. FEBS Lett. 2004;562:11–15. doi: 10.1016/s0014-5793(04)00229-7. [DOI] [PubMed] [Google Scholar]
  • 47.Grammatikakis N, et al. The role of Hsp90N, a new member of the Hsp90 family, in signal transduction and neoplastic transformation. J. Biol. Chem. 2002;277:8312–8320. doi: 10.1074/jbc.M109200200. [DOI] [PubMed] [Google Scholar]

Articles from Carcinogenesis are provided here courtesy of Oxford University Press

RESOURCES