Inhibition of Histone Deacetylase Increases Cytotoxicity to Anticancer Drugs Targeting DNA

Cells and the Treatments of Cells.

The human glioblastoma cell line D54 (13) was provided by Drs. Eugenie S. Kleinerman and Federick F. Lang (The University of Texas M. D. Anderson Cancer Center, Houston, TX); the U118 cells, the breast cancer MCF-7 cells, and the normal breast (MCF-12F) and intestinal (FHs 74 Int) epithelial cells were purchased from American Type Culture Collection (ATCC, Manassas, VA). The glioblastomas cells were grown in MEM containing 10% FBS (Life Technologies, Inc., Rockville, MD). The MCF-7 cells were grown in RPMI 1640 containing 10% FBS. The MCF-12F cells were grown in a 1:1 mixture of DMEM and Ham’s F-12 containing 20 ng/ml epidermal growth factor, 100 ng/ml cholera toxin, 0.01 mg/ml insulin, 500 ng/ml hydrocortisone, and 5% Chelex-treated horse serum. The FHs 74 Int cells were grown in Hybri-Care media (ATCC) containing 10% FBS. The human colon carcinoma RKO cells were obtained from Dr. Michael Kastan (St. Jude Children’s Research Hospital, Memphis, TN). The RKO-E6 cells were provided by Al Fornace (National Cancer Institute, Bethesda, MD). The RKO and RKO-E6 cells were grown in RPMI 1640 containing 10% FBS in the absence of antibiotics. Clonogenic survival assays were performed in the presence of different drug concentrations. The cells were plated at 2–3 × 10² cells/60-mm-diameter Petri dish and exposed to either 10 ng/ml TSA (Sigma, St. Louis, MO) or 1.25 μm SAHA (Biomol, Plymouth Meeting, PA) for 4 h or left untreated before exposure to the anticancer drugs. After removing the HDAC inhibitors, the cells were exposed to increasing amounts of VP-16 (0–100 μm), ellipticine (0–25 μm), doxorubicin (0–10 μm), cisplatin (0–10 μg/ml), CPT (0–10 μm), cyclophosphamide (0–10 mm), or 5-FU (0–10 μm) for 1 h. The anticancer drugs (all from Sigma) were then washed away, and the cells were allowed to grow for 2 weeks. The colonies were fixed with methanol-acetic acid (3:1), stained with crystal violet (0.4%), and counted. The survival fraction was determined by dividing the number of colonies formed in the presence of the drugs by the number of colonies formed in the untreated control cells. The plating efficiency of the untreated control cells was around 50%. Each dose was done in triplicate, and the experiments were repeated at least twice.

MTT Assays.

The cells were plated at 1 × 10⁴ cells/24-well plate, and 72 h after the indicated treatments, the cells were incubated with the tetrazolium salt MTT (0.25 mg/ml in PBS) for 3 h. The media were then removed, and DMSO (1 ml of 100%) was added to solubilize the MTT-formazan product. The amount of violet crystals reflecting cellular growth and viability was determined by absorbance at 540 nm. Each sample was assayed at least three times. The results were expressed as a percentage of MTT reduction in treated samples compared with untreated sample (100%) for the drug alone or compared with TSA-treated sample (94% of untreated sample) for TSA plus anticancer drug-treated samples.

Western Blot Analyses.

Whole cell lysates were separated on 10–15% SDS-PAGE and transferred to nitrocellulose membrane with a semidry blotting apparatus (Bio-Rad, Richmond, CA). The acid-soluble histone proteins were extracted as described below. The blots were incubated with either anti-Topo II monoclonal antibody (Oncogene, Cambridge, MA) at a 1:1000 dilution, anti-acetylated H4 rabbit polyclonal antibody (Upstate Biotechnology, Lake Placid, NY) at a 1:2000 dilution, anti-p53 monoclonal antibody (Pab421; Oncogene) at a 1:1000 dilution, anti-p21 rabbit polyclonal antibody (PharMingen, San Diego, CA) at a 1:500 dilution, anti-Gadd45 rabbit polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz Biotechnology, CA) at a 1:500 dilution, anti-PCNA monoclonal antibody (Santa Cruz Biotechnology) at a 1:1000 dilution, or anti-actin polyclonal antibody (Sigma) at a 1:5000 dilution for 2 h at room temperature. After washing, the blots were reacted with their respective secondary antibody at a 1:5000 dilution and detected with enhanced chemiluminescence reagents (Amersham) according to the supplier’s protocol. For Topo II detection, nuclear proteins were extracted and used instead of whole cell lysate.

Extraction of Acid-Soluble Proteins.

For detection of acetylated histone H4, acid-soluble proteins were extracted with the following procedure, as recommended by Upstate Biotechnology. Briefly, the cells were pelleted down, washed with PBS, and resuspended in 5–10 volumes of cell pellets in histone lysis buffer [10 mm HEPES (pH 7.9), 1.5 mm MgCl₂, 10 mm KCl, 5 mm DTT, and 1.5 mm phenylmethylsulfonyl fluoride]. Sulfuric acid was then added to a final concentration of 0.2 m in polypropylene tubes on ice. After 30 min, the tubes were centrifuged at 11,000 × g for 10 min at 4°C, and the acid-soluble fraction was dialyzed against 0.1 mm acetic acid twice for 1–2 h each time. The lysates were dialyzed again three more times against distilled water for 1, 3, and 16 h, respectively. Anti-acetyl histone H4 antibody was used for the detection of histone acetylation.

Apoptosis Detection.

Fragmented DNA was detected by DNA ladder formation and end-labeling of free 3′-OH. The suicide track DNA ladder isolation kit (Oncogene) was used as recommended by the manufacturer. The cells were treated as indicated in the colony formation assays, and fragmented DNA was extracted 24 h after treatments. The DNA was run on a 1.5% agarose gel and stained with ethidium bromide. Detection of the free 3′-OH was performed with the DeadEnd fluorometric TUNEL system (Promega, Madison, WI) according to the manufacturer’s protocol, 24 h after the indicated treatments. Fluorescence was detected with an inverted fluorescence microscope (Nikon TE 200, HG-100W mercury lamp).

Caspase 3 assay (Promega) was performed according to the manufacturer’s recommendation. The assay was performed in triplicate on 50 μg of cellular extracts. The specific caspase 3 inhibitor Ac-DEVD-CHO was used. Caspase activity was measured 12 h after treatment, and fluorescence of the cleaved substrate was read on a Wallac Victor-1420 microtiter plate fluorometer reader.

Nuclear Extracts and Topo II Assay.

Cells were washed with ice-cold PBS and lysed by incubating them for 10 min in isotonic buffer (150 mm NaCl, 5 mm MgCl₂, 2 mm KH₂PO₄, 1 mm EGTA, 10% glycerol, 0.1 mm DTT, and 1 mm phenylmethylsulfonyl fluoride) containing 0.3% Triton X-100. The nuclear proteins were then extracted for 30 min in lysis buffer containing 350 mm NaCl. The extracts were centrifuged at 20,000 × g at 4°C. The protein content of the supernatant was quantified using the Bradford method (Bio-Rad). The Topo II decatenation activity was measured with a Topo II assay kit (TopoGEN, Colombus, OH) as recommended by the manufacturer. Kinetoplast DNA was used as a substrate, and the Topo II source was provided by the different nuclear extracts.

DNA-Protein Cross-Links.

The DNA-protein cross-links were measured by alkaline elution as described previously (14 , 15) . The cells were metabolically labeled with [¹⁴C]thymidine (0.04 μCi/ml) for 36–48 h at 37°C. The cells were then chased with fresh media for at least 4 h before drug treatment. The cells were then treated with TSA and VP-16 as described in the text.

p53 Sequencing.

Total RNA was isolated using Trizol reagents (Life Technologies, Inc.) according to the manufacturer’s protocols. Five μg of DNA-free total RNA were converted to cDNA with SuperScript II reverse transcriptase and oligo(dT)₁₅ (Life Technologies, Inc.). Equal amounts of cDNA were subsequently amplified by PCR in a 50-μl reaction volume containing 1× PCR buffer, 200 μm deoxynucleotide triphosphates, 10 μm specific primers for human p53 (sense, 5′-CAG-AAA-ACC-TAC-CAG-GGC-AGC-3′ and antisense, 5′-TTT-CTT-GCG-GAG-ATT-CTC-TTC-3′) covering the entire p53 DNA binding domain [residues ∼100 to ∼300 (16)] , and 1.25 units of Taq DNA polymerase (Promega). The PCR product was sequenced directly using the ABI PRISM Dye terminator cycle sequencing ready kit protocol with automated DNA Sequencer (Perkin-Elmer ABI PRISM model).

HDAC Inhibitors Increase the Efficiency of Anticancer Drugs.

To test our hypothesis that increased chromatin accessibility could increase the efficiency of drugs targeting the DNA or enzymes acting on the DNA, we first treated human glioblastoma D54 cells with TSA, a HDAC inhibitor, before treating the cells with VP-16 (etoposide, a Topo II inhibitor). The D54 cells were selected for this study based on their intrinsic resistance to Topo II inhibitors (17) . The dose and time of TSA treatments were selected to obtain enough histone acetylation to confer an open chromatin structure without triggering apoptosis or genome instability. Previous reports have shown that a dose of 10 ng/ml TSA is sufficient to generate di- and tri-acetylated histones (18) without causing significant apoptosis (19) . This level of acetylation correlates with increased DNase I sensitivity in chromatin (20) and thus corresponds to a more accessible, open, chromatin structure. Our data (Fig. 1A) ⇓ indicate that treatment of the cells with VP-16 alone decreased survival at every dose, with a maximum effect at 25 μm VP-16 resulting in 22% survival. However, pretreatment of the cells with TSA increased VP-16 killing efficiency at every dose, with a maximum effect seen at 25 μm VP-16, where >98% of the cells were killed. This represents a 10-fold (1-log kill) increase in VP-16 killing efficiency. To determine whether histone acetylation was a requirement to increase VP-16 efficiency, we treated the cells in the reverse order. When the cells were treated with VP-16 first and then treated with TSA, no significant difference was observed compared with VP-16 treatment alone (Fig. 1A) ⇓ . Survival in the presence of TSA alone (10 ng/ml) was 94%.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1.

A, clonogenic survival after treatment with VP-16 alone, pretreatment with 10 ng/ml TSA for 4 h before VP-16 treatment, or VP-16 treatment followed by TSA in human glioblastoma D54 cells. Colonies were counted 2 weeks after treatment. Survival is expressed as the percentage of colonies obtained with untreated cells or TSA alone. B, clonogenic survival as described in A, except that cells were treated with CPT. C, clonogenic survival as described in A, except that treatments were performed in human glioblastoma U118 cells. D, clonogenic survival assessed in human colon cancer carcinoma RKO cells as described in A.

The effect of TSA on the Topo I inhibitor CPT was also evaluated. Our data (Fig. 1B) ⇓ indicate that pretreatment of D54 cells with TSA did not significantly affect CPT killing efficiency, whereas treating the cells with CPT first followed by TSA treatments slightly decreased CPT killing efficiency. These data suggest that under the conditions used here, histone acetylation could increase the killing efficiency of Topo II inhibitors but not Topo I inhibitors in D54 cells.

To address the possibility that the TSA effect on VP-16 was cell type specific, we treated two other cell lines similarly (the human glioblastoma U118 and human colon carcinoma RKO cell lines). The data presented in Fig. 1, C and D ⇓ , indicate that pretreatment of the cells with TSA before exposure to VP-16 increases the capacity of the Topo II inhibitor to kill cells, with greater efficiency at higher doses of VP-16. For example, at a dose of 100 μm VP-16, survival was reduced 5× when the U118 cells were pretreated with TSA (Fig. 1C) ⇓ , whereas 25 μm VP-16 was sufficient to reduce RKO cell survival 3× (Fig. 1D) ⇓ . Survival in the presence of TSA alone was 100% in U118 cells and 103% in RKO cells. Our data suggest that the effects of TSA on VP-16 are not cell type specific because VP-16 killing efficiency was increased in all three cell lines by pretreatment with TSA. Treating the cells in the reverse order had no more effect than that seen in D54 cells. Therefore, acetylation of core histone by the HDAC inhibitors apparently has to precede the contact of the anticancer drugs with the DNA to obtain a synergistic effect.

We then asked whether the TSA effect was specific for VP-16 or could also affect other Topo II inhibitors. The data shown in Fig. 2, A and B ⇓ , indicate that pretreatment with TSA also increases ellipticine and doxorubicin killing efficiency in U118 cells. The cytotoxicity of ellipticine was increased 3×, whereas doxorubicin killing efficiency was increased 7× at the highest doses of anticancer drugs used. Again, treating the cells in the reverse order (TSA first, followed by anticancer drugs) had no more effect than the anticancer drug alone. These data indicate that the synergistic effect of TSA is not drug specific and can be applied to a variety of Topo II inhibitors. Interestingly, among the three drugs tested in U118 cells, doxorubicin is the one that shows the most sensitivity to pretreatment with TSA. Doxorubicin is a DNA groove binder and an intercalating agent, whereas ellipticine is a pure intercalator, and VP-16 is not known to bind DNA (21) . Close interaction of doxorubicin with DNA might thus be more responsive to a looser chromatin structure.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2.

Clonogenic survival performed as described in the Fig. 1 ⇓ legend with the indicated cell lines and anticancer drugs, except that SAHA (1.25 μm) was used as a HDAC inhibitor in C and D.

The effect of SAHA, a HDAC inhibitor structurally related to TSA and currently under study in clinical trials, was also evaluated. A dose of 1.25 μm SAHA for 4 h was selected because under these conditions, SAHA can increase histone acetylation without affecting cell cycle progression in MCF-7 cells (22) . As reported previously (22) , treatment of the cells with SAHA alone had no significant effect on MCF-7 cells. Survival of 95% was consistently observed under these conditions. Like TSA, pretreatment of the cells with SAHA increased VP-16 killing efficiency. At a dose of 25 μm VP-16, survival of MCF-7 cells was reduced 4× when the cells were pretreated with SAHA (Fig. 2C) ⇓ . Treating the cells in the reverse order (VP-16 first, followed by SAHA) had no more effect than treating the cells with VP-16 alone. We also tested the efficiency of SAHA on the cross-linking agent cisplatin (Fig. 2D) ⇓ . Our data indicate that SAHA also increased the killing efficiency of cisplatin. Survival was reduced 5× when the cells were pretreated with SAHA (Fig. 2D ⇓ ; 10 μg/ml cisplatin). Treating the cells in the reverse order (cisplatin first, followed by SAHA) slightly increased (∼1.8×) cisplatin killing efficiency, but to a lesser extent than treating the cells with SAHA first. This may reflect the different action mechanism of the anticancer drugs and their interaction with DNA. The specificity of SAHA for HDAC supports our hypothesis that a more open chromatin structure can enhance the efficiency of an anticancer drug targeting DNA or enzymes acting on DNA.

We then expanded our study to two other anticancer drugs currently used in cancer treatment, 5-FU and cyclophosphamide. 5-FU is an antimetabolite that can be inserted in either RNA or DNA. The state of chromatin compaction is thus not expected to influence its efficiency. Indeed, our data indicate (Fig. 3A) ⇓ that pretreatment of MCF-7 cells with SAHA had no significant effect on 5-FU killing efficiency in these cells. The last drug that we tested is cyclophosmamide. This nitrogen mustard derivative is a bifunctional alkylating agent that causes interstand DNA cross-link. Although cyclophosmamide is a cross-linking agent like cisplatin (Fig. 2D) ⇓ , no significant effect on pretreatment with SAHA was observed in MCF-7 cells (Fig. 3B) ⇓ . This may be due to the fact that cyclophosphamide usually requires activation by hepatic microsomal P450 enzymes, or it may indicate that not all cross-linking agents could be affected by pretreatment with HDAC inhibitors.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 3.

Clonogenic survival performed as described in the Fig. 1 ⇓ legend with breast cancer MCF-7 cells (A and B) and normal breast epithelial MCF-12 cells (C and D) and the indicated anticancer drugs, except that SAHA (1.25 μm) was used as a HDAC inhibitor.

To determine whether pretreatments with the HDAC inhibitors were toxic to normal cells, we repeated the assays in normal human breast (MCF-12F) and intestinal (FHs 74 Int) epithelial cells. Data obtained with cisplatinum and cyclophosphamide in MCF-12 cells are shown in Fig. 3, C and D ⇓ . No significant effects on the killing efficiency of these drugs or on VP-16 and 5-FU (data not shown) were observed with pretreatments of the cells with SAHA. Similar results were obtained with the normal intestinal epithelial cells (Fig. 4) ⇓ , where no significant effect of pretreatment of the cells with TSA was observed with all of the drugs tested. These data indicate that pretreatments of cancer cells with HDAC inhibitors specifically increase the cytotoxicity of the anticancer drugs in cancer cells.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 4.

Clonogenic survival performed as described in the Fig. 1 ⇓ legend with normal epithelial intestinal FHs 74 Int cells and the indicated anticancer drugs.

Cytotoxicity was also evaluated by measuring mitochondrial metabolic activity in short-term colorimetric assays (MTT) in the two glioblastomas (D54 and U118) and the MCF-7 breast cancer cell line. Our data (data not shown) indicate that TSA or SAHA alone had no cytotoxic effect. In D54 cells, TSA pretreatment of the cells before VP-16 treatments resulted in increased cytotoxicity at all VP-16 doses. Addition of TSA after VP-16 treatments did not increase the VP-16 cytotoxic effect. Comparable effects were observed in U118 cells; these cells were more resistant to VP-16 but still responded to TSA pretreatment in a manner similar to that of D54 cells. In both D54 and U118 cells, about twice as many cells were killed at every dose of VP-16 when the cells were pretreated with TSA. Cytotoxic effects were also evaluated for CPT in D54 cells (data not shown), and as shown above with clonogenic survival (Fig. 1B) ⇓ , no significant differences were observed between the different treatments in D54 cells. Pretreatment of MCF-7 cells with SAHA (data not shown) also increased cisplatin cytotoxicity to a greater extent than treating the cells in the reverse order. For example, about 60% of the cells survived a 5 μg/ml dose of cisplatin, whereas 3× more cells were killed when the cells were pretreated with SAHA. These data also correlate well with the clonogenic results shown in Fig. 2D ⇓ .

Evaluation of Topo II Levels and Function.

Cytotoxicity of Topo II inhibitors may be affected by Topo II levels and activity. Because Topo II promoter is activated by TSA (23) , we wanted to determine whether the increase in sensitivity of Topo II inhibitors was due to increased Topo II levels. The levels of the nuclear Topo II enzymes were evaluated by Western blot analyses in D54 cells either immediately or 24 h after drug exposure (24 h). Our data (data not shown) indicate that the enzyme protein levels did not change in response to the different treatments either immediately or 24 h after the treatments. The levels of Topo II decatenation activity present in the cells under each treatment were also evaluated. Our data (data not shown) indicate that the kinetoplast catenated DNA substrate (cK) was fully decatenated in the untreated sample. Topo II decatenation activity was thus not affected by TSA treatment and only slightly inhibited by a 1-h treatment with VP-16. Twenty-four h after treatments, Topo II was fully active in all of the samples. These data indicate that the Topo II decatenation activity in D54 cells was not significantly affected by any of the treatments.

Effects of Treatments on Histone Acetylation and Endogenous Gene Expression.

To verify that the brief TSA treatment used here was sufficient to influence histone acetylation before VP-16 treatment, we performed Western blot analysis with histone H4 acetylated specific antibody. The data presented in Fig. 5A ⇓ indicate that histone H4 is acetylated after a 4-h exposure to TSA (10 ng/ml; Lane 2). TSA treatment either before or after VP-16 treatment (Lanes 3 and 5) can also increase histone H4 acetylation levels, whereas VP-16 treatment alone (Lane 4) had no effect. Thus, VP-16 did not prevent TSA from increasing histone H4 acetylation (Lane 3 and 5). Coomassie Blue staining of the acid-urea gel indicates that total protein loading was relatively the same in all of the lanes. As mentioned above, it is known that a dose of 10 ng/ml TSA is sufficient to generate di- and tri-acetylated histones (18) . This level of histone hyperacetylation has been shown to correlate with increased DNase I sensitivity in chromatin (20) . One can thus assume that under the conditions used here, acetylation of the core histone could confer an opening of the chromatin structure similar to what has been reported previously (18) .

• Download figure
• Open in new tab
• Download powerpoint

Fig. 5.

Western blots. A, the acid-soluble proteins from D54 cells were extracted immediately (No Delay) after the indicated treatments (10 μm VP-16, 1 h; 10 ng/ml TSA, 4 h). Ten μg were loaded on a 15% SDS-PAGE. Detection of acetylated histone H4 (ActH4) was performed as described in “Materials and Methods.” Even loading of protein was verified by Coomassie Blue staining of the acid-urea gel. Detection of p53 was performed on total cell extracts (50 μg) after the indicated treatments. Actin was used as a control for loading. B, detection of acetylated histone H4 and p53 was as described in A, except that the proteins were extracted 24 h after the indicated treatments. Detection of p21, Gadd45 (G45), PCNA, and actin was performed on 50 μg of total cell extracts as described in “Materials and Methods.” C, the same as B, except that the proteins were extracted from U118 cells.

Because TSA has been shown to increase p53 levels (24) and because p53 can mediate apoptosis (25) , we also evaluated the effects of TSA and VP-16 treatments on p53 levels in D54 cells. As shown in Fig. 5A ⇓ , treatments with a low concentration of TSA (10 ng/ml; Lane 2) or VP-16 (10 μm, Lane 4) were sufficient to increase p53 above basal level. These data are in good agreement with previous reports using higher doses of TSA and VP-16 (17 , 24) . However, when the cells were pretreated with TSA and then exposed to VP-16 (Lane 3), a much higher level of p53 was observed with the same doses of drugs. Treating the cells in the reverse order [VP-16 first, followed by TSA (Lane 5)] had no more effect than each drug alone.

It is well established that the inhibition of HDAC by TSA is transient (26) ; thus, we aimed at determining the effect of the treatments 24 h after removal of the drugs. The data presented in Fig. 5B ⇓ indicate that the levels of histone H4 acetylation were back to basal levels 24 h after exposure to the drugs. As indicated by the staining of the acid-urea gel, reduction of histone acetylation was not caused by protein loading inconsistency. The levels of p53 were still elevated after exposure to VP-16 alone (Lane 4) or TSA followed by VP-16 treatments (Lane 3) but were back to near basal levels after treatment with TSA alone (Lane 2) and VP-16 followed by TSA (Lane 5). These data suggest that brief exposure of the cells to TSA followed by VP-16 has a lasting effect on p53 levels long after the drugs have been removed and the histone acetylation levels were back to basal (hypoacetylated) levels. The effect of VP-16 alone (Lane 4) on p53 levels also seems to last longer than the combined effect of VP-16 followed by TSA (Lane 5). We then determined the effect of the treatments on Gadd45 and p21, two p53 downstream effector genes (27) . The data presented in Fig. 5B ⇓ indicate that, as reported previously, p21 protein levels are increased by TSA (24) and VP-16 (28) . Pretreatment of the cells with TSA before VP-16 exposure (Lane 3) increased p21 levels more than any of the treatments separately (Lanes 2 and 4), whereas the p21 levels were at basal levels 24 h after treating the cells in the opposite order (Lane 5). Interestingly, when the cells were analyzed immediately after treatments, the p21 levels of the sample treated with VP-16 first followed by TSA were comparable with levels seen after treatment with each drug separately (data not shown). Although induction of p21 by HDAC inhibitors is apparently p53 independent (29) , the effects and order of the drugs treatments on p21 protein levels correlate very well with p53 expression here. The levels of Gadd45 also correlated well with p21 and p53 expression. As with p53 and p21, levels of Gadd45 were highest when the cells were treated with TSA first, followed by VP-16 (Lane 3). No significant increase in Gadd45 protein levels was observed when the cells were treated for 1 h with VP-16 alone (Lane 4). This is different than what has been reported by another group (30) , where Gadd45 levels were increased after exposure to VP-16. However, in that study, the cells were treated with VP-16 for 3 h. Addition of TSA after VP-16 did not stimulate p53, p21, and Gadd45 protein levels (Lanes 5). The levels of PCNA and actin (Fig. 5B) ⇓ , as well as staining for total proteins, were measured as internal controls and were not significantly affected by any of the treatments. These data thus indicate that TSA followed by VP-16 (Lanes 3) is the most effective treatment to induce p53, p21, and Gadd45 protein levels. We also measured p53, p21, and Gadd45 protein levels in U118 cells treated similarly (Fig. 5C) ⇓ . Essentially the same results were obtained, even though the levels of p53 and p21 induction were not as high as those seen in D54 cells. Taken together, these data suggest that acetylation of the core histones has to occur before VP-16 can lock Topo II on the DNA to enhance the VP-16-induced expression of p53, p21, and Gadd45.

Pretreatment with TSA Does Not Increase the Overall Number of DNA-Protein Cross-Links.

Because a change in cellular sensitivity to Topo inhibitors could be reflected on the levels of covalent linkages formed between the enzymes and the DNA (3) , we measured the levels of proteins cross-links formed with VP-16 and CPT in U118 cells pretreated with TSA (Fig. 6) ⇓ . Our data indicate that the levels of DNA-protein cross-links formed with each inhibitor increased in a dose-dependent manner but that the levels of DNA-protein cross-links did not change consistently when the cells were pretreated for 4 h with TSA (Fig. 6) ⇓ . The difference between the survival data (Fig. 1) ⇓ and the cross-linking data may be due to the limitation of the alkaline elution technique (14) , which can measure DNA-protein cross-links in the overall genome but will not detect selective changes.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 6.

DNA-protein cross-links produced in U118 cells with (+TSA) or without (−TSA) TSA pretreatment and 1-h exposure to (A) VP-16 or (B) CPT as measured by alkaline elution.

TSA Increases VP-16-Mediated Apoptosis.

D54 cells are resistant to VP-16 induced apoptosis (17) . Because cytotoxicity to Topo II inhibitors is often correlated with apoptosis (31) and because our data (Fig. 1) ⇓ indicate that pretreatment of D54 cells with TSA increases VP-16 cytotoxicity, we wanted to determine whether the increased cytotoxicity observed in Fig. 1 ⇓ affected apoptosis. We first measured the levels of internucleosomal DNA cleavage (DNA ladder) in each of the treatments used. Our data (data not shown) indicate that none of the treatments used, including pulse treatment with VP-16 alone and brief treatment of the cells with TSA alone or before and after VP-16 treatment, had an effect on internucleosomal DNA cleavage. However, internucleosomal cleavage occurs rather late in apoptosis stages and cannot be detected when <10–20% of the cells are apoptosing. We thus performed a more sensitive assay (DeadEnd; fluorescence TUNEL system) that allows detection of DNA strand breaks by labeling free 3′-OH termini. The data presented in Fig. 7 ⇓ indicate that brief TSA treatment with doses as low as 10 ng/ml (Fig. 7f) ⇓ can increase the levels of DNA strand breaks above the basal level in D54 cells. The 1-h pulse treatment with VP-16 alone (Fig. 7j) ⇓ did not cause sufficient DNA strand breaks to be detectable by the TUNEL assay. Pretreatment of the cells with TSA followed by VP-16 1-h pulse increased the levels of DNA stand breaks severalfold above background (Fig. 7h) ⇓ . Treating the cells in the reverse order [VP-16 first, followed by TSA (Fig. 7l) ⇓] produced similar amounts of DNA strand breaks as treatment with TSA alone (Fig. 7f) ⇓ . These data correlate well with the cytotoxicity data presented in Fig. 1 ⇓ and suggest that acetylation of the core histone before VP-16 interaction with DNA and the Topo II enzyme can increase VP-16 cytotoxicity.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 7.

Detection of free 3′-OH. The D54 cells were treated as indicated. Twenty-four h later, the cells were reacted with terminal deoxynucleotidyl transferase (TdT) and fluorescein-labeled and unlabeled deoxynucleotides. The cell populations are shown in phase-contrast (a, c, e, g, i, and k) and with a fluorescein filter (b, d, f, h, j, and l).

Role of p53 in Mediating TSA-Induced Sensitization to Anticancer Drugs.

Depending on the cell lines, functional p53 can either increase or reduce the sensitivity to anticancer drugs (32) . To determine whether p53 plays a role in the TSA-induced sensitization to anticancer agents, we repeated the survival and TUNEL assays in addition to performing a caspase 3 assay in RKO-E6 cells. These cells are identical to the parental RKO cells, except that they express the viral E6 protein that degrades p53 (33) . We first verified that the levels of p53 protein in the RKO-E6 cells were reduced. The data shown in Fig. 8A ⇓ indicate that p53 levels are readily detectable in the RKO parental cell line (Lanes 1, 3, and 5) but are below detection levels in the RKO-E6 cellular extracts (Lanes 2, 4, and 6). However, the survival curves obtained with both cell lines pretreated with TSA before VP-16 are virtually identical (Fig. 8B) ⇓ . No differences between the two cell lines were noticed when the cells were treated with either VP-16 alone (data not shown) or VP-16 followed by TSA (Fig. 8C) ⇓ . Nonetheless, because clonogenic survival does not distinguish between the different cell death mechanisms and because functional p53 has been associated with apoptosis mediated by Topo II poisons (34) , we also measured the role of p53 in the TSA sensitization to VP-16-mediated apoptosis. Both RKO and RKO-E6 cells were treated and analyzed by TUNEL assays as described above for the D54 cells. Like D54 cells, the RKO cells show increased VP-16-induced apoptosis when pretreated with TSA (data not shown). On the other hand, VP-16-induced apoptosis was not increased to the same extent by TSA pretreatment in RKO-E6 cells (data not shown). These data suggest that both p53-dependent and p53-independent apoptosis are responsible for TSA sensitization to VP-16 cytotoxicity.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 8.

A, Western blot analysis of RKO (−) and RKO-E6 (+) protein extracts. The blots were reacted with either p53 or actin antibody as described in “Materials and Methods.” B, clonogenic survival assessed as described in the Fig. 1 ⇓ legend in RKO and RKO-E6 cells. Cells were pretreated with TSA (10 ng/ml) for 4 h before addition of VP-16. C, clonogenic survival was assessed as described in B, except that the cells were treated with VP-16 before addition of TSA.

As a more direct measure of apoptosis, we evaluated the activation of caspase 3 after each treatment in RKO and RKO-E6 (Fig. 9) ⇓ . The data shown in Fig. 9A ⇓ indicate that activation of caspase 3 is markedly increased when the RKO cells are pretreated with TSA. Pretreatment with TSA also increased caspase 3 activity in RKO-E6 cells, but to a lesser extent than that seen in the wild-type cell line (Fig. 9B) ⇓ . These data indicate that the increased cytotoxicity observed with pretreatment of the cells with HDAC inhibitors is mediated by increased activation of caspase 3; consequently, apoptosis, part of this effect, is p53 dependent.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 9.

Caspase 3 activation assay. The RKO and RKO-E6 cells were treated as indicated in the Fig. 1 ⇓ legend. Twelve h later, the cells were harvested, and the protein extracts (50 μg) were assayed for caspase 3 activity as described in “Materials and Methods.” Relative caspase 3 activity is measured by subtracting background (blank) and the fluorescence in the presence of the specific caspase 3 inhibitor Ac-DEDV-CHO.