The use of histone deacetylase inhibitors (HDACI), a promising new class of antineoplastic agents, in combination with cytotoxic agents, such as ionizing radiation and anticancer drugs, has been attracting attention. In this study, we found that sodium butyrate (SB), a widely studied HDACI, remarkably enhanced the cell killing effect of psoralen plus UVA (PUVA) in several cancer cell lines, including skin melanoma. Although a single treatment with PUVA or SB did not greatly affect cell survival, combined treatment with SB and PUVA induced marked apoptosis within 24 hours. The SB-induced augmentation of the cell killing effect was more dramatic in combination with PUVA than with anticancer drugs. The number of double-strand breaks that formed during the repair of PUVA-induced interstrand cross-links (ICL) in chromosomal DNA was significantly reduced in SB-pretreated cells, suggesting that the ability to repair ICL was attenuated by SB. In addition, the incorporation of bromodeoxyuridine and the formation of repair foci of proliferating cell nuclear antigen after PUVA treatment, associated with nucleotide excision repair (NER) in the removal of ICL, were not observed in SB-pretreated cells. Furthermore, the repair kinetics of UV-induced cyclobutane pyrimidine dimers (well-known photolesions repaired by NER) were much slower in SB-pretreated cells than in untreated cells. These results indicated that the enhanced cell killing effect of PUVA by SB was attributable to an attenuated ability to repair DNA and, especially, dysfunctional NER. [Cancer Res 2009;69(8):3492–500]

Histone deacetylase (HDAC) inhibitors (HDACI) have been attracting attention as a promising new class of antineoplastic agents for the treatment of solid and hematologic malignancies (13). HDAC is an enzyme that regulates the acetylation status of core nucleosomal histones by catalyzing the removal of acetyl groups from their amino-terminal residue. Inhibition of HDAC results in the accumulation of acetylated histones by histone acetyltransferase. It weakens histone-DNA interactions and consequently increases the accessibility of transcription factors to DNA, which causes the altered expression of several sets of proteins involved in regulating cell cycle, differentiation, and apoptosis (4). For instance, proapoptotic (e.g., Bak, Bax, and Bim) and antiapoptotic (e.g., Bcl-2, Bcl-XL, XIAP, and Mcl-1) factors are up-regulated and down-regulated by HDACI and seem to play an important role in their antitumor activity (5).

Recently, the use of HDACI in combination with cytotoxic agents to treat cancer has been a focus of attention (6, 7). Enhancement of the cell killing effects of radiation and anticancer drugs by various HDACI [sodium butyrate (SB), trichostatin A (TSA), suberoylanilide hydroxamic acid, valproic acid, etc.], due to a marked induction of apoptosis, has been reported (613). However, the degree of enhancement is highly dependent on the combination of HDACI and cytotoxic agent and on cell type. In addition, the molecular mechanisms, by which HDACI improves the cell killing effect of cytotoxic agents, are not fully understood. At present, efforts are under way to determine which combinations of HDACI and cytotoxic agents are most suitable for cancer therapy and elucidate the mechanisms underlying the enhanced cell killing effect.

Psoralens are tricyclic aromatic compounds with phototherapeutic activities. Due to their planar structure, psoralens can intercalate into DNA at alternating pyrimidine-purine sites, in particular, at TA sites (14). After UVA (320–400 nm) irradiation, the intercalated complex becomes activated, resulting in the formation of covalent interstrand cross-links (ICL) with pyrimidines (15). At present, psoralen plus UVA (PUVA) is used for the treatment of several hyperproliferative skin diseases, such as psoriasis and cutaneous T-cell lymphoma (CTCL; ref. 16). PUVA-induced ICL in chromosomal DNA induces an antiproliferative effect convenient for skin therapy. Some studies found that PUVA showed a cell killing effect (apoptosis) in cultured lymphoma cell lines (17, 18); however, the application of PUVA to melanoma and carcinoma has not been realized because the cell killing power is not as strong as that of anticancer drugs and radiation.

In this study, we found that SB markedly enhanced the apoptotic cell killing effect of PUVA. The degree to which the cytotoxicity was enhanced by SB pretreatment was much greater for PUVA treatment than for general anticancer drugs. Furthermore, we clarified that SB attenuated DNA repair, especially nucleotide excision repair (NER), which potentially contributed to the augmented cell killing effect of PUVA.

Cells and cell culture conditions. The cell lines SK-MEL-28 and G-361, derived from malignant human skin melanoma, were purchased from Health Science Research Resources Bank. The cell line of a squamous cell carcinoma of human skin (HSC-1) was purchased from Japanese Collection of Research Bioresources. The cells were maintained in MEM (for SK-MEL-28 and G-361) or DMEM (for HSC-1) supplemented with fetal bovine serum [FBS; 10% (SK-MEL28 and G-361) or 20% (HSC-1)] and 100 units/mL of penicillin/streptomycin at 37°C in an atmosphere of 5% CO2.

Combined treatment with SB and PUVA. The cells treated with SB (0.1–3 mmol/L) for specific periods (4–24 h) were subjected to PUVA treatment. The cells were cultured with 8-methoxypsoralen (8-MOP; 10−7–10−5 mol/L) for 1 h in a medium containing SB and irradiated with several doses of UVA (1–5 J/cm2) immediately after changing to a SB-free medium. The emission characteristics of the UVA lamp were described previously (19). The irradiance at the sample level was about 1.5 mW/cm2. It takes about 1 h for the irradiation of 5 J/cm2 UVA. The combined treatment with SB and PUVA described above is henceforth called “SB-PUVA.” During UVA irradiation, control cells and cells treated with SB only (these cells were blocked out from UVA with an aluminum foil) were kept in the same conditions as SB-PUVA–treated cells.

Viability assay. Cell viability was estimated from the metabolism of fluorescein diacetate (FDA; Wako). FDA is hydrolyzed by cytoplasmic esterases into fluorescent fluorescein in living cells. Cells were suspended in PBS containing FDA (0.1 mg/mL) and incubated for 15 min at 37°C. Propidium iodide (PI; 10 μg/mL) was added before the measurement of fluorescence intensity using a flow cytometer (FCM; Epics XL; Coulter). Double staining with FDA and PI divided the cells into three populations: (a) viable cells (FDA+, PI−), (b) dead cells (FDA−, PI+), and (c) cells undergoing apoptosis (FDA−, PI−). The assessment of apoptosis by double staining with FDA and PI is equivalent to that of using Annexin V staining, a highly specific indicator of apoptosis (20).

Apoptosis assay. Apoptosis was determined by staining with both the DNA-binding fluorochrome Hoechst 33258 to detect morphologic changes to chromatin and PI to determine the percentage of subdiploid apoptotic nuclei (sub-G1 fraction) as described previously (19). Caspase-3 activity was measured by a direct assay in cell lysates using a fluorescent substrate of the enzyme (Ac-Asp-Glu-Val-Asp-MCA; ref. 19).

Detection of DNA double-strand breaks. Double-strand breaks (DSB) were detected using a biased sinusoidal field gel electrophoresis (BSFGE) system (ATTO) as described previously (21). Cells were suspended in 1% low-melting agarose immediately after treatment and solidified. The agarose plugs were treated with 0.5 mg/mL of proteinase K and 1 mg/mL of RNase A and electrophoresed in a 0.8% agarose gel. The gel was stained with ethidium bromide and photographed.

DNA repair synthesis assay. SK-MEL-28 cells grown to confluence on 60-mm dishes or 35-mm glass base dishes were further cultured for 24 h in serum-depleted MEM. They were treated with SB (3 mmol/L) for 24 h in MEM containing 0.5% FBS. Subsequently, PUVA treatment was carried out and the cells were incubated with bromodeoxyuridin (BrdUrd; 10 μmol/L) for predetermined periods (0–30, 30–60, 60–90, and 90–120 min after PUVA treatment).

For the FCM analysis, cells fixed in by ethanol (70%) were treated with 2 N HCl for 15 min at room temperature. After blocking with 1% bovine serum albumin (BSA), they were incubated with primary antibody against BrdUrd (Santa Cruz Biotechnology; 1:200) and then with secondary antibody conjugated with FITC (Jackson Immuno Research; 1:200). The enhanced intensity of FITC based on BrdUrd uptake during each repair period was analyzed using FCM.

For immunofluorescence microscopy, cells fixed with 4% paraformaldehyde were immersed in PBS containing 0.5% Triton X-100 for 3 min at 4°C and treated with 2 N HCl [this process of DNA denaturing was carried out only for the examination of BrdUrd uptake and not for the detection of repair foci of proliferating cell nuclear antigen (PCNA)]. After blocking with 1% BSA, they were incubated with primary antibody against BrdUrd (1:200) or PCNA (Santa Cruz Biotechnology; 1:200) and then with secondary antibody conjugated with FITC (1:200). The nucleus was stained with PI (20 μg/mL). Images were acquired on a laser scanning confocal microscope (LSM510, Carl Zeiss).

Measurement of repair kinetics of cyclobutane pyrimidine dimers. For in situ detection of cyclobutane pyrimidine dimers (CPD), SK-MEL-28 cells cultured in 35-mm glass base dishes were treated with SB (3 mmol/L) for 24 h and irradiated with UVC (0.02 J/cm2; main emission wavelength, 254 nm; ATTO) through a polycarbonate isopore membrane filter (pore size, 3 μm; Millipore). After irradiation, they were incubated with SB-free MEM for a predetermined period (0–24 h) to allow the repair of photolesions. Cells were fixed with 4% paraformaldehyde, treated with 0.5% Triton X-100 in PBS, and treated with 2 N HCl. After blocking with 1% BSA, they were incubated with primary antibody against anti-CPD (1:1,500; Medical and Biological Laboratories) and then with secondary antibody conjugated with FITC. The nucleus was stained with PI (20 μg/mL). Images were acquired on a laser scanning confocal microscope.

For the detection of CPD by ELISA, cells were treated with SB (3 mmol/L) for 24 h and irradiated with UVC (0.002 J/cm2). After incubation for 24 h with SB-free MEM, genomic DNA was extracted and denatured by boiling for 15 min. The DNA (1 μg/well) was applied into a polyvinylchloride 96-well plate precoated with protamine sulfate (0.003%). After blocking with FBS (2% in PBS), the samples were incubated with the primary antibody against anti-CPD (1:2,000) and then with the biotin-F(ab')2 fragment of antimouse IgG (Zymed; 1:2,000). Subsequently, a peroxidase reaction was performed using o-phenylene diamine (0.4 mg/mL) in the presence of 0.02% H2O2. The reaction was stopped by the addition of H2SO4 (2 mol/L), and absorbance at 492 nm was measured by spectrophotometer.

Statistics. Values are means ± SD (n = 3–4). Data were analyzed with one-way ANOVA followed by Dunnett's t test for comparisons between groups. Statistical significance is represented as follows: *, P < 0.05; **, P < 0.01; ***, P < 0.001.

SB enhances cell killing effect of PUVA. SK-MEL-28 cells were treated with SB (0.1–3 mmol/L) for a predetermined time (4–24 h; the acetylated status of histone H3 is shown in Supplementary Fig. S1) and subjected to PUVA treatment (Fig. 1). The cells were quite resistant to PUVA treatment (8-MOP, 10−5 mol/L; UVA, 5 J/cm2). In clinical treatment, a variety of 8-MOP concentrations have been used. For example, tissue concentrations of 8-MOP after topical administration [8-MOP bath therapy at 3 mg/L (1.4 × 10−5 mol/L)] were reported to reach 4.5 × 10−6 mol/L (22). SB-PUVA treatment induced drastic cell death, the extent of which was dependent on SB dose, SB treatment time, 8-MOP dose, and UVA dose (Fig. 1A). A time course study showed that survival decreased sharply from 8 to 24 hours (Fig. 1B). No synergistic effects were observed when the cells were treated in the reverse order (PUVA first, followed by SB; data not shown). TSA, another HDACI, also enhanced the cell killing effect of PUVA (Fig. 1C). SB-PUVA treatment could effectively kill other skin cancer cell lines: malignant human skin melanoma (G-361) and human skin squamous cell carcinoma (HSC-1; Fig. 1D).

Figure 1.

SB-mediated enhanced cell killing effect of PUVA. A, SK-MEL-28 cells pretreated with SB for 24 h (or 4–24 h) were subjected to PUVA (8-MOP, 10−7–10−5 mol/L; UVA, 1–5 J/cm2) treatment. Cell viability was determined by FDA assay 24 h after the treatment. B, cell viability after SB (3 mmol/L)–PUVA (8-MOP, 10−5 mol/L; UVA, 5 J/cm2) treatment was determined at the indicated time points. C, SK-MEL-28 cells pretreated with TSA (0.1–10 μmol/L) for 24 h were subjected to PUVA (8-MOP, 10−5 mol/L; UVA, 5 J/cm2). Cell viability was determined at 24 h. D, G-361 and HSC-1 cells pretreated with SB (3 mmol/L) for 24 h were subjected to PUVA (8-MOP, 10−7–10−5 mol/L; UVA, 5 J/cm2) treatment. Cell viability was determined at 24 h.

Figure 1.

SB-mediated enhanced cell killing effect of PUVA. A, SK-MEL-28 cells pretreated with SB for 24 h (or 4–24 h) were subjected to PUVA (8-MOP, 10−7–10−5 mol/L; UVA, 1–5 J/cm2) treatment. Cell viability was determined by FDA assay 24 h after the treatment. B, cell viability after SB (3 mmol/L)–PUVA (8-MOP, 10−5 mol/L; UVA, 5 J/cm2) treatment was determined at the indicated time points. C, SK-MEL-28 cells pretreated with TSA (0.1–10 μmol/L) for 24 h were subjected to PUVA (8-MOP, 10−5 mol/L; UVA, 5 J/cm2). Cell viability was determined at 24 h. D, G-361 and HSC-1 cells pretreated with SB (3 mmol/L) for 24 h were subjected to PUVA (8-MOP, 10−7–10−5 mol/L; UVA, 5 J/cm2) treatment. Cell viability was determined at 24 h.

Close modal

Human breast adenocarcinoma cells (MCF-7) and human lung adenocarcinoma epithelial cells (A549) also showed similar susceptibility to SB-PUVA treatment (Supplementary Fig. S2). Furthermore, pretreatment with SB significantly augmented the cytotoxicity induced by 5-MOP, a positional isomer of 8-MOP, plus UVA (Supplementary Fig. S3).

Combined treatment with SB and PUVA induces apoptosis. Induction of apoptosis at each indicated point in time after the SB-PUVA treatment was assessed based on double staining with FDA and PI (Fig. 2A), detection of a sub-G1 fraction (Fig. 2B), morphologic change of chromatin (Fig. 2C), and caspase-3 activity (Fig. 2D). After the treatment with SB-PUVA, the population of the cells in the FDA(+)PI(−) section time-dependently moved to FDA(−)PI(+) via FDA(−)PI(−) (Fig. 2A). A time-dependent increase in the sub-G1 fraction was also observed (Fig. 2B). Chromatin-condensed cells increased in number time-dependently after the SB-PUVA treatment but not after single treatment with PUVA or SB-UVA treatment (data not shown). Activation of caspase-3 was observed in the period from 8 to 16 hours after the treatment (Fig. 2D). Furthermore, z-vad-fmk, a pan-caspase inhibitor, clearly attenuated cell death induced by SB-PUVA treatment (data not shown). These results indicated that SB-PUVA treatment effectively induced typical caspase-dependent apoptosis in SK-MEL-28 cells.

Figure 2.

Induction of apoptosis by combined treatment with SB and PUVA. SK-MEL-28 cells pretreated with SB (3 mmol/L) for 24 h were subjected to PUVA (8-MOP, 10−5 mol/L; UVA, 5 J/cm2) treatment and further incubated for the indicated period (4–24 h). A, FCM analysis by double staining with FDA and PI. B, detection of sub-G1 fractions. C, morphologic assessment of apoptotic cells using Hoechst 33258. D, measurement of caspase-3 activity using a fluorescence substrate. **, P < 0.01; ***, P < 0.001.

Figure 2.

Induction of apoptosis by combined treatment with SB and PUVA. SK-MEL-28 cells pretreated with SB (3 mmol/L) for 24 h were subjected to PUVA (8-MOP, 10−5 mol/L; UVA, 5 J/cm2) treatment and further incubated for the indicated period (4–24 h). A, FCM analysis by double staining with FDA and PI. B, detection of sub-G1 fractions. C, morphologic assessment of apoptotic cells using Hoechst 33258. D, measurement of caspase-3 activity using a fluorescence substrate. **, P < 0.01; ***, P < 0.001.

Close modal

Enhanced cell killing effect by SB is more drastic in combination with PUVA than with anticancer drugs. SK-MEL-28 cells precultured with SB for 24 hours were treated with various anticancer drugs (bleomycin, etoposide, Adriamycin, 5-fluorouracil, paclitaxel, and vincristine) for 24 hours (Fig. 3). The dose ranges of the anticancer drugs used here were based on other papers (1113, 23) and are comparable with or higher than the maximally achieved therapeutic concentrations in humans (24). In all cases, combined treatment with SB and anticancer drugs resulted in augmented cytotoxicity compared with single treatments with anticancer drugs. The cell killing effects of combined treatment with SB and anticancer drugs were much weaker than those of SB and PUVA treatment. Similar results were also observed in G-361 and HSC-1 cells (Supplementary Fig. S4).

Figure 3.

Cell killing effect of combined treatment with SB and anticancer drugs. SK-MEL-28 cells precultured with SB (3 mmol/L) for 24 h were treated with anticancer drugs (bleomycin, etoposide, Adriamycin, 5-fluorouracil, paclitaxel, and vincristin) for 24 h. Cell viability was determined by FDA assay.

Figure 3.

Cell killing effect of combined treatment with SB and anticancer drugs. SK-MEL-28 cells precultured with SB (3 mmol/L) for 24 h were treated with anticancer drugs (bleomycin, etoposide, Adriamycin, 5-fluorouracil, paclitaxel, and vincristin) for 24 h. Cell viability was determined by FDA assay.

Close modal

As treatment with HDACI, including SB, is reported to change the representation of cellular apoptosis-related proteins (5), some antiapoptotic proteins, such as c-IAP and Bcl-2, were analyzed and found to be down-regulated by SB (Supplementary Fig. S5); however, down-regulation might not be the main reason for the significant cell death induced by SB-PUVA treatment because cell killing was specific to PUVA treatment and not to the anticancer drugs (Fig. 3 and Supplementary Fig. S4). In addition, the change in cell cycle distribution after treatment with SB was independent of the cell killing effect (Supplementary Fig. S6). SB caused cells to accumulate in the G2-M phase, and nocodazole, an antimicrotublue drug, also arrested cells in G2-M; however, the drastic cell death induced by PUVA treatment was observed in SB-pretreated cells, not in nocodazole-pretreated cells. Munshi and colleagues showed that cell cycle distribution was not involved in radiosensitive effects of SB (9). Given these results, the mechanism of augmented cytotoxicity by SB-PUVA treatment might not be merely changes in the levels of cellular apoptosis-related proteins and cell cycle distribution.

SB reduces the number of DSBs formed by PUVA treatment. The DNA damage induced by PUVA treatment is mainly psoralen ICL, which is believed to be the primary cause of the PUVA-induced cell killing effect. We speculated that SB attenuated the repair efficiency of psoralen ICL, which led to a significant cell death. DSBs have been reported to be generated as intermediates in the cellular repair of psoralen ICL (25). The degree to which DSBs formed after treatment with PUVA or SB-PUVA was examined using BSFGE as an index of the efficiency of the repair of ICL (Fig. 4). Immediately after the PUVA treatment, clear DSBs were observed. Pretreatment with SB reduced the number of DSBs generated, the extent of which was dependent on SB-pretreated dose (Fig. 4A) and time (Fig. 4B). Pretreatment with TSA also reduced the generation of DSBs after the treatment with PUVA in a dose-dependent manner (Fig. 4C). In Fig. 4D, cells were pretreated with SB for 24 hours, further cultured in SB-free medium for a predetermined time, and then subjected to PUVA. According to the amount of time cultured with SB-free medium, the formation of DSBs attenuated by SB returned to the original level induced by PUVA. These results suggested that SB reduced the repair of ICL induced by PUVA.

Figure 4.

Induction of DSBs after PUVA or SB-PUVA treatment. A and B, SK-MEL-28 cells pretreated with SB (A, 0.5–3 mmol/L; B, 3 mmol/L) for the indicated period (A, 24 h; B, 2–24 h) were subjected to PUVA treatment (8-MOP, 10−5 mol/L; UVA, 5 J/cm2). C, cells pretreated with TSA (0.1–30 μmol/L) for 24 h were subjected to PUVA treatment (8-MOP, 10−5 mol/L; UVA, 5 J/cm2). D, cells pretreated with SB (3 mmol/L) for 24 h were further cultured for the indicated period (0–24 h) in SB-free medium and subjected to PUVA (8-MOP, 10−5 mol/L; UVA, 5 J/cm2). Immediately after PUVA treatment, DSBs were detected by BSFGE. Severed DNA due to DSBs accompanied by ICL repair is detected at the lower position in the gel.

Figure 4.

Induction of DSBs after PUVA or SB-PUVA treatment. A and B, SK-MEL-28 cells pretreated with SB (A, 0.5–3 mmol/L; B, 3 mmol/L) for the indicated period (A, 24 h; B, 2–24 h) were subjected to PUVA treatment (8-MOP, 10−5 mol/L; UVA, 5 J/cm2). C, cells pretreated with TSA (0.1–30 μmol/L) for 24 h were subjected to PUVA treatment (8-MOP, 10−5 mol/L; UVA, 5 J/cm2). D, cells pretreated with SB (3 mmol/L) for 24 h were further cultured for the indicated period (0–24 h) in SB-free medium and subjected to PUVA (8-MOP, 10−5 mol/L; UVA, 5 J/cm2). Immediately after PUVA treatment, DSBs were detected by BSFGE. Severed DNA due to DSBs accompanied by ICL repair is detected at the lower position in the gel.

Close modal

SB attenuates DNA repair synthesis after PUVA treatment. It has been proposed that the NER pathway is involved in the removal of posralen ICL (25). The capacity for NER in SB-treated cells after PUVA treatment was measured by DNA repair synthesis (DRS) assay. The intensity of fluorescence from incorporated BrdUrd was significantly higher in the PUVA-treated cells than untreated cells (Fig. 5A,, left) during all periods examined (0–2 h; the degree of increase in fluorescence intensity was not so different among the BrdUrd-labeled periods). On the other hand, no increase in fluorescence after PUVA treatment was observed when the cells were pretreated with SB (Fig. 5A,, right). In addition, we could detect discrete foci of BrdUrd in the nucleus due to DRS after PUVA treatment (Fig. 5B), consistent with an earlier study (26). However, such foci were hardly detectable after PUVA treatment in cells pretreated with SB. Furthermore, repair foci of PCNA in the late stages of NER (2729) were detected after PUVA treatment but not in cells pretreated with SB. These results suggested that SB reduced the capacity for NER.

Figure 5.

DRS after PUVA or SB-PUVA treatment. A, SK-MEL-28 cells treated with SB (3 mmol/L) for 24 h in MEM containing 0.5% FBS were subjected to PUVA (8-MOP 10−5 mol/L, UVA 5 J/cm2 or 8-MOP 10−4 mol/L, UVA 1 J/cm2) treatment. Subsequently, they were incubated with BrdUrd (10 μmol/L) for the indicated periods (a, 0–30 min; b, 30–60 min; c, 60–90 min; d, 90–120 min) after PUVA treatment. The fluorescence intensity due to the incorporated BrdUrd was measured using FCM. As typical profiles of “untreated” and “treated with SB only,” a pattern of the cells labeled with BrdUrd during 30 to 60 min is shown. B, cells pretreated with SB (3 mmol/L) for 24 h in MEM containing 0.5% FBS were subjected to PUVA (8-MOP, 10−5 mol/L; UVA, 5 J/cm2) treatment. After culture for 1 h (labeling period with BrdUrd was from 30–60 min after irradiation), the cells were fixed. The incorporation of BrdUrd and repair foci of PCNA were visualized by immunofluorescence microscopy. Cells that showed discrete foci of BrdUrd and PCNA in the nucleus were counted.

Figure 5.

DRS after PUVA or SB-PUVA treatment. A, SK-MEL-28 cells treated with SB (3 mmol/L) for 24 h in MEM containing 0.5% FBS were subjected to PUVA (8-MOP 10−5 mol/L, UVA 5 J/cm2 or 8-MOP 10−4 mol/L, UVA 1 J/cm2) treatment. Subsequently, they were incubated with BrdUrd (10 μmol/L) for the indicated periods (a, 0–30 min; b, 30–60 min; c, 60–90 min; d, 90–120 min) after PUVA treatment. The fluorescence intensity due to the incorporated BrdUrd was measured using FCM. As typical profiles of “untreated” and “treated with SB only,” a pattern of the cells labeled with BrdUrd during 30 to 60 min is shown. B, cells pretreated with SB (3 mmol/L) for 24 h in MEM containing 0.5% FBS were subjected to PUVA (8-MOP, 10−5 mol/L; UVA, 5 J/cm2) treatment. After culture for 1 h (labeling period with BrdUrd was from 30–60 min after irradiation), the cells were fixed. The incorporation of BrdUrd and repair foci of PCNA were visualized by immunofluorescence microscopy. Cells that showed discrete foci of BrdUrd and PCNA in the nucleus were counted.

Close modal

SB attenuates repair kinetics of CPD formed by UVC irradiation. To further confirm that SB impaired NER, we investigated the kinetics of the repair of CPD formed by exposure to local UVC (Fig. 6). Immediately after irradiation (0 hour), clear CPD foci having high fluorescent intensities were detected in both SB-pretreated and untreated cells (Fig. 6A). Without pretreatment with SB, the fluorescence weakened and the outline of CPD foci became fuzzy 24 hours after the treatment. On the other hand, distinct CPD foci were still detected in SB-pretreated cells. Cells having CPD foci showing both a clear contour and high fluorescent intensity were counted as CPD-positive (Fig. 6B). CPD-positive cells decreased with repair time more rapidly in the SB-untreated cells than SB-pretreated cells. At 24 hours after UV irradiation, about 15% of the untreated cells and 60% of the SB-pretreated cells were CPD positive. In Fig. 6C, the levels of CPD 24 hours after UV irradiation were determined using ELISA. Levels were significantly higher for SB-pretreated cells than untreated cells. In addition, cytotoxicity was much enhanced when SB-pretreated cells were exposed to UVC (Supplementary Fig. S7). These results further supported our hypothesis that SB attenuates NER capacity.

Figure 6.

Repair kinetics of CPD after PUVA or SB-PUVA treatment. A, SK-MEL-28 cells treated with SB (3 mmol/L) for 24 h were partially irradiated with UVC (0.02 J/cm2) through a polycarbonate isopore membrane filter. After incubation for the indicated period (0–24 h), the cells were fixed. In situ visualization of CPD was carried out as described in Materials and Methods. B, cells having CPD foci with both a clear contour and a high fluorescent intensity were counted as CPD positive. C, cells treated with SB (3 mmol/L) for 24 h were irradiated with UVC (0.002 J/cm2, without membrane filter). After culture for 24 h, the amount of CPD in genomic DNA was examined by ELISA, as described in Materials and Methods. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 6.

Repair kinetics of CPD after PUVA or SB-PUVA treatment. A, SK-MEL-28 cells treated with SB (3 mmol/L) for 24 h were partially irradiated with UVC (0.02 J/cm2) through a polycarbonate isopore membrane filter. After incubation for the indicated period (0–24 h), the cells were fixed. In situ visualization of CPD was carried out as described in Materials and Methods. B, cells having CPD foci with both a clear contour and a high fluorescent intensity were counted as CPD positive. C, cells treated with SB (3 mmol/L) for 24 h were irradiated with UVC (0.002 J/cm2, without membrane filter). After culture for 24 h, the amount of CPD in genomic DNA was examined by ELISA, as described in Materials and Methods. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Close modal

We reported here, for the first time, that pretreatment with HDACI (SB and TSA) remarkably enhanced the cell killing effect of PUVA in several cancer cell lines. Recently, the use of HDACI, in combination with various inducers of cell death, such as ionizing radiation and anticancer drugs, has attracted considerable attention (613). Among these combinations, we believe the HDACI-PUVA treatment presented in this study to be one of the most effective.

A possible mechanism by which SB may enhance the cell killing effect of PUVA was the augmentation of DNA damage. The number of initial ICL induced by PUVA treatment might not be so different between SB-pretreated and untreated cells because the initial amount of CPD after UVC exposure was almost the same regardless of SB pretreatment (Fig. 6). Kim and colleagues also showed that pretreatment with TSA did not increase the overall number of DNA-protein cross-links induced by camptothecin and etoposide (11). On the other hand, the formation of DSBs as intermediates in the repair of ICL was clearly attenuated by pretreatment with SB (Fig. 4). Therefore, SB is suspected to cause the accumulation of ICL by inhibiting their repair.

The mechanism by which ICL are repaired in mammals is not well understood; however, it seems to involve several DNA repair systems, such as NER, homologous recombination (HR) repair, translesion synthesis (TLS), and the Fanconi anemia/BRCA (FA/BRCA) pathway (25, 30, 31). In our understanding based on a recently proposed mechanism for the repair of psoralen-ICL, NER is important for the early stages of the repair process: (a) recognition of ICL is probably carried out by NER proteins, such as XPE (also called DDB2 or p48), XPC-Rad23, and XPA-RPA (28, 32, 33) and (b) dual incision at both sides of the lesion is mediated by the NER excinuclease ERCC1/XPF, leading to an uncoupling of one arm of the ICL (34). Thereafter, TLS, FA/BRCA, and HR pathways would function sequentially. After the first incision (b), the resulting gap of nucleotides is filled by DNA polymerase. PCNA is essential for the repair synthesis (gap filling) stage of NER and forms chromatin-bound foci (repair foci; refs. 2729). Detection of DRS by immunostaining of BrdUrd and PCNA is often used to measure NER activity (28, 29, 35, 36). In this study, limited DRS was observed in SB-pretreated cells after PUVA treatment (Fig. 5). In addition, slower repair kinetics of CPD (Fig. 6) and much enhanced cytotoxicity (Supplementary Fig. S7) after UVC irradiation in SB-pretreated cells suggested that NER in the removal of ICL was attenuated by SB. Furthermore, the degree of UVC-induced and PUVA-induced cell death augmented by SB pretreatment was more remarkable in cell lines expressing XPA, an essential protein for NER, than in isogenic lines deficient in XPA (Supplementary Fig. S8), supporting our contention that SB attenuated DNA repair, especially the NER pathway. On the other hand, our findings are inconsistent with earlier findings that pretreatment with SB enhanced the capacity for NER (37, 38). If SB-pretreated cells have an enhanced NER capacity, the rate of their survival after UV irradiation would be high compared with that of untreated cells. However, we and others have shown that pretreatment with HDACI enhanced cell death induced by UVC (Supplementary Fig. S7; refs. 39, 40). A recent report by Kim and colleagues showed that levels of CPD remaining at 4 hours after UVC irradiation were more than twice higher in TSA-pretreated cells than in untreated cells, with significant cell death in TSA-pretreated cells (39). Further study is needed to explain this contradiction.

It is important to realize that the attenuation of NER capacity by SB was reversible. In fact, the SB-attenuated generation of DSBs during the repair of ICL was reversed with time in the SB-free medium after treatment with SB (Fig. 4D), suggesting that the capacity for NER recovered. This result reflected cell survival, that is, SK-MEL-28 cells gradually became resistant to PUVA treatment with time in the SB-free medium (Supplementary Fig. S9A). Acetylated levels of histone H3 were well correlated with the survival rate (Supplementary Fig. S9B). To achieve effective treatment, acetylation levels of histone would be a good indicator of when PUVA should be conducted.

Munshi and colleagues reported that the enhancement of radiosensitivity by HDACI, including SB, was due to a down-regulation in the expression of DSB repair proteins, especially nonhomologous end rejoining-related (NHEJ) proteins, such as Ku70, Ku80, DNA-PKcs, and Rad50 (9, 10). Furthermore, HDACI induced the suppression of BRCA1 and Rad51, suggesting the attenuation of DSB repair by HR (41, 42). In our study, the levels of several NHEJ-related (Ku70 and Ku80) and NER-related (XPA and PCNA) proteins were not significantly affected by the treatment with SB (Supplementary Fig. S10). However, importantly, down-regulation of p53 expression by SB was observed. Peltonen and colleagues also showed the suppression of p53 by TSA in several melanoma cell lines, including SK-MEL-28 cells (43). p53 recruits NER factors (XPC, XPD, XPB, and XPE) to CPD sites (4446). Furthermore, expression of XPC and XPE is regulated by p53 (46). These reports suggested that loss or disruption of p53 function may result in a significant decrease of NER capacity.

Although our final goal is to apply SB-PUVA treatment to therapy for some types of cancer, including melanoma and carcinoma, the immediate potential applications of SB-PUVA treatment would be in therapy for CTCL because PUVA and suberoylanilide hydroxamic acid (vorinostat), a HDACI, are used separately for CTCL therapy. Indeed, SB-PUVA treatment killed the CTCL cells (HUT78 and HH) and was more effective than single PUVA treatment (Supplementary Fig. S11). The enhanced cell killing effect was greater than that of SB in combination with general anticancer drugs.

We should consider the risk-benefit of SB-PUVA treatment, that is, tumor-killing activity and cancer risk are two sides of the same coin. PUVA therapy itself is associated with an increased incidence of skin cancer (47). Because SB could attenuate the repair of ICL, there is a possibility that SB increases the incidence of cancer caused by PUVA treatment. How to handle HDACI in combination with PUVA for clinical use needs to be studied further.

In summary, we found that combined treatment with HDACI (SB and TSA) and PUVA had a significant cell killing effect on several cancer cell lines. The cell killing effect of SB-PUVA treatment was much greater than that of SB in combination with general anticancer drugs. This enhanced cell killing was potentially caused by the observed attenuation of DNA repair by SB, especially NER.

No potential conflicts of interest were disclosed.

Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

Grant support: Japan Ministry of Education, Culture, Sports, Science and Technology grant-in-aid for scientific research (C) 19510071.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1
Marks PA, Miller T, Richon VM. Histone deacetylases.
Curr Opin Pharmacol
2003
;
3
:
344
–51.
2
Marks PA, Richon VM, Miller T, Kelly WK. Histone deacetylase inhibitors.
Adv Cancer Res
2004
;
91
:
137
–68.
3
Mai A, Massa S, Rotili D, et al. Histone deacetylation in epigenetics: an attractive target for anticancer therapy.
Med Res Rev
2005
;
25
:
261
–309.
4
Marks PA, Richon VM, Rifkind RA. Histone deacetylase inhibitors: inducers of differentiation or apoptosis of transformed cells.
J Natl Cancer Inst
2000
;
92
:
1210
–6.
5
Zhang XD, Gillespie SK, Borrow JM, Hersey P. The histone deacetylase inhibitor suberic bishydroxamate regulates the expression of multiple apoptotic mediators and induces mitochondria-dependent apoptosis of melanoma cells.
Mol Cancer Ther
2004
;
3
:
425
–35.
6
Karagiannis TC, El-Osta A. The paradox of histone deacetylase inhibitor-mediated modulation of cellular responses to radiation.
Cell Cycle
2006
;
5
:
288
–95.
7
Camphausen K, Tofilon PJ. Inhibition of histone deacetylation: a strategy for tumor radiosensitization.
J Clin Oncol
2007
;
25
:
4051
–6.
8
Chinnaiyan P, Vallabhaneni G, Armstrong E, Huang SM, Harari PM. Modulation of radiation response by histone deacetylase inhibition.
Int J Radiat Oncol Biol Phys
2005
;
62
:
223
–9.
9
Munshi A, Kurland JF, Nishikawa T, et al. Histone deacetylase inhibitors radiosensitize human melanoma cells by suppressing DNA repair activity.
Clin Cancer Res
2005
;
11
:
4912
–22.
10
Munshi A, Tanaka T, Hobbs ML, Tucher SL, Richon VM, Meyn RE. Vorinostat, a histone deacetylase inhibitor, enhances the response of human tumor cells to ionizing radiation through prolongation of γ-H2AX foci.
Mol Cancer Ther
2006
;
5
:
1967
–74.
11
Kim MS, Blake M, Baek JH, Kohlhagen G, Pommier Y, Carrier F. Inhibition of histone deacetylase increases cytotoxicity to anticancer drugs targeting DNA.
Cancer Res
2003
;
63
:
7291
–300.
12
Dowdy SC, Jiang S, Zhou XC, et al. Histone deacetylase inhibitors and paclitaxel cause synergistic effects on apoptosis and microtubule stabilization in papillary serous endometrial cancer cells.
Mol Cancer Ther
2006
;
5
:
2767
–76.
13
Sonnemann J, Kumar KS, Heesch S, et al. Histone deacetylase inhibitors induce cell death and enhance the susceptibility to ionizing radiation, etoposide, and TRAIL in medulloblastoma cells.
Int J Oncol
2006
;
28
:
755
–66.
14
Hearst JE, Isaacs ST, Kanne D, Rapoport H, Straub K. The reaction of the psoralens with deoxyribonucleic acid.
Q Rev Biophys
1984
;
17
:
1
–44.
15
Cimino GD, Gamper HB, Isaacs ST, Hearst JE. Psoralens as photoactive probes of nucleic acid structure and function: organic chemistry, photochemistry, and biochemistry.
Annu Rev Biochem
1985
;
54
:
1151
–93.
16
Stern RS. Psoralen and ultraviolet a light therapy for psoriasis.
N Engl J Med
2007
;
357
:
682
–90.
17
Yoo EK, Rook AH, Elenitsas R, Gasparro FP, Vowels BR. Apoptosis induction of ultraviolet light A and photochemotherapy in cutaneous T-cell Lymphoma: relevance to mechanism of therapeutic action.
J Invest Dermatol
1996
;
107
:
235
–42.
18
Canton M, Caffieri S, Dall'Acqua F, Di Lisa F. PUVA-induced apoptosis involves mitochondrial dysfunction caused by the opening of the permeability transition pore.
FEBS Lett
2002
;
522
:
168
–72.
19
Ibuki Y, Goto R. Antiapoptotic effects induced by different wavelengths of ultraviolet light.
Photochem Photobiol
2002
;
75
:
495
–502.
20
Saha NR, Usami T, Suzuki Y. A double staining flow cytometric assay for the detection of steroid induced apoptotic leucocytes in common carp (Cyprinus carpio).
Dev Comp Immunol
2003
;
27
:
351
–63.
21
Toyooka T, Ibuki Y. Co-exposure to benzo(a)pyrene and UVA induces phosphorylation of histone H2AX.
FEBS Lett
2005
;
579
:
6338
–42.
22
Tegeder I, Bräutigam L, Podda M, et al. Time course of 8-methoxypsoralen concentrations in skin and plasma after topical (bath and cream) and oral administration of 8-methoxypsoralen.
Clin Pharmacol Ther
2002
;
71
:
153
–61.
23
Dong YB, Yang HL, Elliott MJ, McMasters KM. Adenovirus-mediated E2F-1 gene transfer sensitizes melanoma cells to apoptosis induced by topoisomerase II inhibitors.
Cancer Res
2002
;
62
:
1776
–83.
24
Markasz L, Stuber G, Vanherberghen B, et al. Effect of frequently used chemotherapeutic drugs on the cytotoxic activity of human natural killer cells.
Mol Cancer Ther
2007
;
6
:
644
–54.
25
Dronkert ML, Kanaar R. Repair of DNA interstrand cross-links.
Mutat Res
2001
;
486
:
217
–47.
26
Lee YJ, Park SJ, Ciccone SL, Kim CR, Lee SH. An in vivo analysis of MMC-induced DNA damage and its repair.
Carcinogenesis
2006
;
27
:
446
–53.
27
Costa RM, Chigaçans V, Galhardo Rda S, Carvalho H, Menck CF. The eukaryotic nucleotide excision repair pathway.
Biochimie
2003
;
85
:
1083
–99.
28
Aboussekhra A, Wood RD. Detection of nucleotide excision repair incisions in human fibroblasts by immunostaining for PCNA.
Exp Cell Res
1995
;
221
:
326
–32.
29
Toschi L, Bravo R. Changes in cyclin/proliferating cell nuclear antigen distribution during DNA repair synthesis.
J Cell Biol
1988
;
107
:
1623
–8.
30
Zhang N, Liu X, Li L, Legerski R. Double-strand breaks induce homologous recombinational repair of interstrand cross-links via cooperation of MSH2, ERCC1-XPF, REV3, and the Fanconi anemia pathway.
DNA Repair (Amst)
2007
;
6
:
1670
–8.
31
Rothfuss A, Grompe M. Repair kinetics of genomic interstrand DNA cross-links: evidence for DNA double-strand break-dependent activation of the Fanconi anemia/BRCA pathway.
Mol Cell Biol
2004
;
24
:
123
–34.
32
Thoma BS, Wakasugi M, Christensen J, Reddy MC, Vasquez KM. Human XPC-hHR23B interacts with XPA-RPA in the recognition of triplex-directed psoralen DNA interstrand crosslinks.
Nucleic Acids Res
2005
;
33
:
2993
–3001.
33
Vasquez KM, Christensen J, Li L, Finch RA, Glazer PM. Human XPA and RPA DNA repair proteins participate in specific recognition of triplex-induced helical distortions.
Proc Natl Acad Sci USA
2002
;
99
:
5848
–53.
34
Kuraoka I, Kobertz WR, Ariza RR, Biggerstaff M, Essigmann JM, Wood RD. Repair of an interstrand DNA cross-link initiated by ERCC1-XPF repair/recombination nuclease.
J Biol Chem
2000
;
275
:
26632
–6.
35
Svetlova MP, Solovjeva LV, Pleskach NA, Tomillin NV. Focal sites of DNA repair synthesis in human chromosomes.
Biochem Biophys Res Commun
1999
;
257
:
378
–83.
36
Beisker W, Hittelman WN. Measurement of the kinetics of DNA repair synthesis after uv irradiation using immunochemical staining of incorporated 5-bromo-2′-deoxyuridine and flow cytometry.
Exp Cell Res
1988
;
174
:
156
–67.
37
Ramanathan B, Smerdon MJ. Enhanced DNA repair synthesis in hyperacetylated nucleosomes.
J Biol Chem
1989
;
264
:
11026
–34.
38
Smerdon MJ, Lan SY, Calza RE, Reeves R. Sodium butyrate stimulates DNA repair in UV-irradiated normal and xeroderma pigmentosum human fibroblasts.
J Biol Chem
1982
;
257
:
13441
–7.
39
Kim MS, Baek JH, Chakravarty D, Sidransky D, Carrier F. Sensitization to UV-induced apoptosis by the histone deacetylase inhibitor trichostatin A (TSA).
Exp Cell Res
2005
;
306
:
94
–102.
40
Pani B, Babudri N, Giancotti V, Russo E. Sodium butyrate affects the cytotoxic and mutagenic response of V79 Chinese hamster cells to the genotoxic agents, daunorubicin and U.V. radiation.
Mutat Res
1984
;
140
:
175
–9.
41
Zhang Y, Carr T, Dimtchev A, Zaer N, Dritschilo A, Jung M. Attenuated DNA damage repair by trichostatin A through BRCA1 suppression.
Radiat Res
2007
;
168
:
115
–24.
42
Adimoolam S, Sirisawad M, Chen J, Thiemann P, Ford JM, Buggy JJ. HDAC inhibitor PCI-24781 decreases RAD51 expression and inhibits homologous recombination.
Proc Natl Acad Sci U S A
2007
;
104
:
19482
–7.
43
Peltonen K, Kiviharju TM, Jävinen PM, Ra R, Laiho M. Melanoma cell lines are susceptible to histone deacetylase inhibitor TSA provoked cell cycle arrest and apoptosis.
Pigment Cell Res
2005
;
18
:
196
–202.
44
Wang QE, Zhu Q, Wani MA, Wani G, Chen J, Wani AA. Tumor suppressor p53 dependent recruitment of nucleotide excision repair factors XPC and TFIIH to DNA damage.
DNA Repair (Amst)
2003
;
2
:
483
–99.
45
Chang YC, Jan KY, Cheng CA, Liao CB, Liu YC. Direct involvement of the tumor suppressor p53 in nucleotide excision repair.
DNA Repair (Amst)
2008
;
7
:
751
–61.
46
Fitch ME, Cross IV, Ford JM. p53 responsive nucleotide excision repair gene products p48 and XPC, but not p53, localize to sites of UV-irradiation-induced DNA damage, in vivo.
Carcinogenesis
2003
;
24
:
843
–50.
47
Stern RS, Nichols KT, Väkevä LH. Malignant melanoma in patients treated for psoriasis with methoxsalen (psoralen) and ultraviolet A radiation (PUVA). The PUVA follow-up study.
N Engl J Med
1997
;
336
:
1041
–5.