Lymphokine-Activated Killer T-Cell-Originated Protein Kinase Phosphorylation of Histone H2AX Prevents Arsenite-Induced Apoptosis in RPMI7951 Melanoma Cells

Purpose: Arsenic is a valuable therapeutic tool in cancer treatment. Lymphokine-activated killer T-cell-originated protein kinase (TOPK) is highly expressed in cancer cells, but its specific function is still unknown. We investigated the role of TOPK in arsenic-induced apoptosis in RPMI7951 human melanoma cells.

Experimental Design: Expression of TOPK was evaluated in different melanoma cell lines, and liquid chromatography-tandem mass spectrometry analysis was used to identify proteins binding with TOPK. Immunofluorescence, Western blot, and flow cytometry were used to assess the effect of arsenic on TOPK, histone H2AX, and apoptosis in RPMI7951 cells.

Results: Melanoma cell lines expressing high levels of TOPK were more resistant to arsenite (As³⁺)-induced apoptosis. As³⁺ treatment induced phosphorylation of TOPK and histone H2AX in RPMI7951 human melanoma cells. Liquid chromatography-tandem mass spectrometry results indicated that TOPK could bind with histone H2AX, and in vitro and in vivo assays confirmed that TOPK binds with and phosphorylates histone H2AX. As³⁺ treatment caused phosphorylation of TOPK, which colocalized with phosphorylated histone H2AX in the nucleus. TOPK small interfering RNA cells exhibited a decreased phosphorylation of histone H2AX with As³⁺ treatment. As³⁺-induced apoptosis was decreased in H2AX^−/− cells but increased in TOPK small interfering RNA cells.

Conclusions: TOPK binds with histone H2AX and inhibits As³⁺-induced apoptosis through phosphorylation of histone H2AX. Melanoma cell lines with high levels of TOPK are more resistant to As³⁺-induced apoptosis. Therefore, inhibition of TOPK activity combined with As³⁺ treatment may be helpful in the treatment of melanomas.

Lymphokine-activated killer T-cell-originated protein kinase (TOPK) is a novel mitotic protein kinase that is highly expressed only in various cancers, such as leukemia, myeloma, and lymphoma, and its expression has been correlated with the malignant potential of these tumors (1–4). TOPK is phosphorylated and active only during mitosis (2) and has been shown to phosphorylate the p38 mitogen-activated protein kinase but not extracellular signal-regulated kinases or c-Jun NH₂-terminal kinases (JNK; ref. 1). Recent reports showed that TOPK is up-regulated in murine myeloma cells by an interleukin-6-mediated protein-protein interaction between TOPK and Raf-A (4, 5). In addition, TOPK is involved in cell cycle regulation and can be a substrate of cdc2/cyclin B (2, 6). TOPK expression has also been shown to be up-regulated during the G₂ to M phase transition, where the cdc2/cyclin B complex plays an important role (6). Thr⁹ is an important phosphorylation site of TOPK because, when Thr⁹ was substituted to Ala, the binding ability of TOPK to cdc2/cyclin B was decreased (6). The cell cycle–specific transcription factors E2F and cyclic AMP-responsive element binding protein/activating transcription factor are critical regulators of TOPK expression during growth arrest in leukemia cells (7). Thus, TOPK may be a potential target for chemotherapeutic or chemopreventive compounds.

The effectiveness of many therapeutic approaches, including γ-irradiation and chemotherapeutic drug treatment, has been proposed to be associated with the reactivation of apoptosis in cancer cells. Arsenic is a paradoxical compound that induces global changes in gene expression and cell signal transduction pathways in different types of cells (8, 9). Dose-dependent effects of arsenic in a tissue-specific manner have enabled arsenic to be used for the effective treatment of certain types of cancers, including leukemia and myeloma, through its induction of apoptosis (10, 11). Arsenic treatment has been shown to induce the up-regulation of cyclin B1 and activates the cdc2/cyclin B1 complex in mitotic cells (12) and therefore could have an effect on TOPK and its potential binding partners.

Melanomas develop through well-defined morphologic and histologic stages that involve the loss of cell proliferation control, the acquisition of invasiveness, and, ultimately, the acquisition of metastatic potential (13). Melanoma cells generally express chromosome instability during their development (14). Histone H2AX has been shown to prevent aberrant repair of both programmed and general DNA breakage (15). H2AX deficiency decreases genomic stability and increases tumor susceptibility of normal cells and tissues (16). In addition, a lack of histone H2AX causes genomic instability in mice (17). Here, we have investigated a functional relationship and interaction of TOPK and histone H2AX in mediating arsenite (As³⁺)-induced apoptosis in melanoma cells. Results indicated that As³⁺-induced phosphorylation of TOPK and TOPK directly phosphorylated histone H2AX, resulting in the inhibition of As³⁺-induced apoptosis in RPMI7951 melanoma cells. This suggested that TOPK could be a potential target for chemotherapeutic treatment of melanomas.

Reagents and antibodies. As³⁺ and anti-β-actin were from Sigma (St. Louis, MO); NE-PER Nuclear and Cytoplasmic Extraction Reagents were from Pierce Biotechnology (Rockville, IL). Stock solutions of As³⁺ (10 mmol/L) in 0.1% DMSO were stored at −20°C. PDZ-binding kinase/TOPK, phosphorylated PDZ-binding kinase/TOPK (Thr⁹) antibodies, and active glutathione S-transferase-TOPK were from Cell Signaling Technology, Inc. (Beverly, MA); anti-histone H2AX, anti-phosphorylated H2AX, and histone H2AX (recombinant protein expressed in Escherichia coli) were from Upstate Biotechnology, Inc. (Lake Placid, NY); anti-hemagglutinin (HA) probe (F-7) was from Santa Cruz Biotechnology (Santa Cruz, CA).

Cell culture. SK-MEL5 and SK-MEL28 human malignant melanoma and SK-MEL31 and RPMI7951 human malignant melanoma epithelial-like cell lines were from the American Type Culture Collection (Manassas, VA). Human melanoma cell lines were cultured in MEM containing 10% fetal bovine serum (FBS; 15% FBS for SK-MEL31), 2 mmol/L l-glutamine, and 25 μg/mL gentamicin, 0.1 mmol/L nonessential amino acids, and 1 mmol/L sodium pyruvate (SK-MEL31 also required 1.5 g/L sodium bicarbonate) at 37°C in humidified air with 5% CO₂. Mouse epidermal JB6 promotion-sensitive Cl41 cells were cultured in MEM containing 5% FBS, 2 mmol/L l-glutamine, and 25 μg/mL gentamicin at 37°C in humidified air with 5% CO₂. H2AX^+/+ and H2AX^−/− cell lines were a gift from Dr. Andre Nussenzweig (Experimental Immunology Branch, National Cancer Institute, NIH, Bethesda, MD) and were cultured in DMEM supplemented with 10% FBS, 2 mmol/L l-glutamine, and 25 μg/mL gentamicin at 37°C in humidified air with 5% CO₂.

Generation of the TOPK-overexpressing cell line. pc-DNA3, pcDNA3-HA-TOPK, or mutant pcDNA3-HA-TOPK-T9A (gifts from Dr. J. Abe, Department of Pathology, Division of Molecular Pathology, Ehime University School of Medicine, Toh-on, Ehime, Japan; ref. 6) was transfected into JB6 Cl41 cells using SuperFect (Qiagen, Valencia, CA). Transfected cells were selected in MEM containing 5% FBS and G418 (800 μg/mL) for 2 weeks. Stable cell lines were maintained in MEM containing 5% FBS and 200 μg/mL G418.

Small interfering RNA preparation and vector construction. Two pairs of hairpin small interfering RNA (siRNA) oligonucleotides, containing BamHI and HindIII sites, were designed as described previously (18). Hairpin siRNA template oligonucleotides were chemically synthesized, deprotected, and gel purified by Sigma-Genosys (Woodlands, TX). The TOPK siRNA sequence target sequences were aligned to the genome database in a BLAST search to ensure sequences without significant homology to other genes. The sense siRNA template sequence for TOPK was 5′-GATCCGAGGTTTGTCTCATTCTCCTTCAAGAGAGGAGAATGAGACAAACCTCTTTTTTGGAAA-3′ and the antisense siRNA template sequence was 5′-AGCTTTTCCAAAAAAGAGGTTTGTCTCATTCTCCTCTCTTGAAGGAGAATGAGACAAACCTCG-3′. The sense and antisense oligonucleotides were annealed and cloned into the pSilencer 3.1-H1 neo vector (Ambion, Austin, TX) at the BamHI and HindIII sites as described by the manufacturer. A scrambled siRNA with a sequence lacking significant homology to the mouse, human, or rat genome database was used as the control or mock siRNA. The resulting pSilencer 3.1-H1-siRNA plasmids were transfected into RPMI7951 cells, and the stable cell lines were obtained by G418 screening. These TOPK siRNA cell lines were cultured in MEM supplemented with 10% FBS, 2 mmol/L l-glutamine, 25 μg/mL gentamicin, 0.1 mmol/L nonessential amino acids, 1 mmol/L sodium pyruvate, and 200 μg/mL G418.

Identification of proteins binding with TOPK by liquid chromatography-tandem mass spectrometry analysis. The topk gene was amplified by PCR and then cloned into pET-46 using a pET-46 Ek/LIC kit (Novagen, Inc., Madison, WI). His-TOPK was purified from BL21 (DE3) cells (Novagen). His-TOPK (0.5 mg) was used for binding with 400 μL Ni-NTA agarose beads (Qiagen, Hilden, Germany). Then, a lysate (10 mg) of RPMI7951 cells was incubated with His-TOPK beads at 4°C overnight. The TOPK-binding proteins were eluted with 50% acetonitrile (Fisher Biotech, Fair Lawn, NJ). Approximately 12 μg of protein eluted from the His-TOPK beads were digested in solution with sequencing grade-modified trypsin (Promega, Madison, WI) according to the manufacturer's protocol. The sample was purified using a C18 SepPak cartridge (Waters Institute, Milford, MA) according to the manufacturer's directions and then speed vacuumed to dryness. The peptide mixture was reconstituted with loading buffer (98:2, H₂O/acetonitrile, 0.1% formic acid) and analyzed by capillary liquid chromatography-tandem mass spectrometry (LC-MS/MS; refs. 19, 20). Product ion mass spectra were searched against the nonredundant database⁴

of National Center for Biotechnology Information (April 20, 2005; total 1,182,676 protein sequences), interpreted using the Pro ID version 1.1 software (Applied Biosystems, Inc., Foster City, CA), which used the Interrogator algorithm for scoring peptide/protein candidates (21), and results were verified by manual interpretation.

SDS-PAGE and Western blotting. Cell lines (7 × 10⁵) were cultured in their respective medium for 12 to 15 hours in 10-cm-diameter dishes to 70% to 80% confluence. Cells were treated with As³⁺ and harvested after 24 hours with 200 μL of radioimmunoprecipitation assay buffer (1× PBS, 1% NP40, 0.5% sodium deoxycholate, 0.1% SDS, 1 mmol/L Na₃VO₄, 1 mmol/L aprotinin, 1 mmol/L phenylmethylsulfonyl fluoride). The quantity of protein was determined by the Bradford method (22). The samples (30-50 μg protein) with 5× SDS were loaded into 10% to 15% SDS polyacrylamide gel for electrophoresis and subsequently transferred onto an Immobilon-P transfer membrane (Millipore, Chelmsford, MA). Antibody-bound proteins were detected by chemiluminescence (Enhanced Chemifluorescence, Amersham Pharmacia Biotech, Piscataway, NJ) and analyzed using the Storm 840 Scanner (Molecular Dynamics, Sunnyvale, CA). Untreated cell samples were used as negative controls.

In vitro kinase assays. Samples containing recombinant histone H2AX expressed in E. coli were incubated at 30°C for 30 minutes with active glutathione S-transferase-TOPK in 10× kinase buffer A [50 mmol/L Tris-HCl (pH 7.5), 10 mmol/L MgCl₂, 1 mmol/L EGTA, 1 mmol/L DTT, 0.01% Brij 35; Cell Signaling Technology] containing 200 μmol/L ATP. The reactions were stopped by adding 5× SDS sample buffer. Then, phosphorylation of H2AX (Ser¹³⁹) was analyzed by Western blot using a phosphorylated H2AX (Ser¹³⁹) antibody. Phosphorylation of histone H2AX by JNK1 was used as a positive control. For some experiments, equal protein loading was verified by silver staining for histone H2AX.

Immunofluorescence assay. To determine the translocation ability of phosphorylated TOPK and phosphorylated H2AX, RPMI7951 melanoma cells (5 × 10⁵) treated or not treated with 2.5 μmol/L As³⁺ were incubated for 24 hours. Cells were fixed in 4% paraformaldehyde and incubated with anti-phosphorylated TOPK and anti-phosphorylated H2AX and then with either FITC-conjugated secondary antibody or Texas red–conjugated secondary antibody (Invitrogen, Carlsbad, CA). Samples were analyzed with a fluorescence microscope system (Leica, Mannheim, Germany).

Isolation of histone H2AX. Histones were extracted from As³⁺-treated cells by disrupting cells with NETN buffer [150 mmol/L NaCl, 1 mmol/L EDTA, 20 mmol/L Tris (pH 8.0), 0.5% nonionic detergent Igepal CA630 (NP40); Sigma]. The insoluble fraction was pelleted for 5 minutes in a microcentrifuge (8,400 rpm). Nuclei were extracted with 0.1 N HCl to isolate total histones. Samples were precipitated with 1 mol/L Tris-HCl (pH 8.0) and then resuspended in double-distilled water (23).

Flow cytometry analysis. Apoptosis induced by As³⁺ was determined using the Annexin V-FITC Apoptosis Detection kit (Medical & Biological Laboratories, Nagoya, Japan) according to the protocol provided. Briefly, cells were trypsinized, washed once with MEM containing serum, and incubated with Annexin V–conjugated FITC. Apoptosis was analyzed using a flow cytometer (FACSCalibur, Becton Dickinson, Franklin Lakes, NJ).

Statistical analysis. Comparisons were made using one-way ANOVA, and data are expressed as mean ± SD of three to four independent experiments. Differences were considered significant with P < 0.05.

TOPK expression in human melanoma cell lines. TOPK is expressed in a wide range of cancers, including leukemia, myeloma, and lymphoma (1–4, 7). However, the expression of TOPK in melanoma cells is yet to be determined. Thus the expression of TOPK was compared in the mouse epidermal JB6 Cl41 skin cell line and in four different human malignant melanoma cell lines, SK-MEL28, SK-MEL31, SK-MEL5, and RPMI7951, which are routinely used in our laboratory. Results indicated that the expression of TOPK in the SK-MEL28 and RPMI7951 cell lines was highest compared with SK-MEL31, SK-MEL5, or JB6 Cl41 cells (Fig. 1). Based on these results, we chose the RPMI7951 melanoma cell line to study TOPK function and to identify potential binding partners.

Identification of TOPK-binding partners by LC-MS/MS. First, a His-TOPK protein was generated and was found to bind strongly with Ni-NTA agarose beads. His-TOPK was confirmed to remain bound to the beads even after elution with 50% acetonitrile (data not shown). Therefore, His-TOPK-Ni-NTA agarose (His-TOPK) beads could be used for immobilizing TOPK-binding proteins for subsequent identification. RPMI7951 cell lysate was incubated with Ni-NTA agarose beads as a negative control (Fig. 2A, lane 1) or Ni-NTA-His-TOPK beads (Fig. 2A, lane 2). After washing with PBS, proteins binding with His-TOPK beads were visualized with silver staining after elution with 50% acetonitrile (Fig. 2A). LC-MS/MS analysis (24) identified 26 individual proteins that could bind with TOPK (data not shown). A fragment composed of 19 amino acids, VTIAQGGVLPNIQAVLLPK, was identified with a Pro ID confidence of 99% as a peptide from the family of H2A proteins (Fig. 2B). The human genome contains 16 genes that encode for H2A peptides classified as H2A variants (25). The identified peptide sequence was 100% homologous to the sequences of 12 genes of the 16 H2A histone family members, including the important H2A variant, H2AX. A role for H2AX phosphorylation has been shown in DNA repair, cell cycle checkpoint regulation, regulation of gene recombination events, and tumor suppression (26). The LC-MS/MS result suggested that TOPK could bind with histone H2AX. Thus, we focused on elucidating the function and the physiologic significance of the interaction of TOPK and histone H2AX.

TOPK binds with and phosphorylates histone H2AX in vitro. Histone H2AX was incubated with Ni-NTA agarose beads (as a negative control) or His-TOPK-Ni-NTA agarose beads at 30°C for 30 minutes. The binding interaction was analyzed by Western blot using an H2AX antibody. Results confirmed that TOPK binds with histone H2AX (Fig. 2C). Because TOPK is a kinase, we next determined whether TOPK can phosphorylate histone H2AX by using histone H2AX as substrate for active TOPK. The phosphorylation was visualized by autoradiography in the presence of [γ-³²P]ATP (Fig. 2D) or by a phosphorylated H2AX (Ser¹³⁹) antibody (Fig. 2E). Phosphorylation of H2AX by JNK1 was used as positive control (27). These results strongly indicated that TOPK binds with and phosphorylates H2AX at Ser¹³⁹ in vitro. Silver staining of H2AX, TOPK, and JNK1 was used as an internal control to verify equal protein loading.

As³⁺ induces phosphorylation of TOPK and H2AX in RPMI7951 melanoma cells. As³⁺ is a toxin with multiple effects in animal and human populations. Low concentration doses of As³⁺ have been reported to induce apoptosis of human melanoma cells (10, 11). The effect of As³⁺ on TOPK (Thr⁹) or H2AX (Ser¹³⁹) phosphorylation in RPMI7951 melanoma cells was investigated herein by using Western blot analysis. Results indicated that a strong phosphorylation of TOPK (Thr⁹) and H2AX (Ser¹³⁹) was induced in a dose-dependent manner following a 24-hour As³⁺ treatment of RPMI7951 cells (Fig. 3, top and middle top). For further experiments, cells were treated with 2.5 μmol/L As³⁺ and harvested 24 hours after treatment. No change in the nonphosphorylated levels of TOPK or H2AX expression was observed in these cell lines under these conditions with As³⁺ treatment (Fig. 3, bottom and middle bottom).

Phosphorylated TOPK and phosphorylated H2AX colocalize in the nucleus after As³⁺ treatment. Yih et al. (28) previously reported that phosphorylated H2AX was clearly detectable in the interface nuclei of CGL-2 cells treated for 24 hours with 2 μmol/L As³⁺. In the present study, RPMI7951 cells were treated with 2.5 μmol/L As³⁺ for 24 hours and then cytoplasmic and nuclear proteins were extracted and TOPK and phosphorylated TOPK were detected by Western blot (Fig. 4A). The results indicated that TOPK is located mostly in the cytoplasm with very little protein in the nucleus in the absence of As³⁺ treatment (Fig. 4A). TOPK localization to the nucleus increased dramatically with As³⁺ treatment (Fig. 4A, top), and phosphorylated TOPK was located only in the nucleus (Fig. 4A, bottom). Immunocytofluorescence analysis was used to further examine the nuclear localization and interaction of TOPK and histone H2AX (Fig. 4B). Results indicated that phosphorylated TOPK (green) and phosphorylated H2AX (red) were both localized in the nucleus after As³⁺ treatment (Fig. 4B, top and middle right). The merged result confirmed that TOPK colocalized with histone H2AX in the nucleus with As³⁺ treatment (Fig. 4B, bottom right).

TOPK siRNA-transfected cells inhibit the phosphorylation of histone H2AX. RPMI7951 cells were transfected with pSilencer 3.1-H1-topk-siRNA and selected by G418. The expression of TOPK was verified by Western blot using anti-TOPK. Results indicated that, in control siRNA mock-transfected cells, As³⁺ induced phosphorylation of TOPK (Thr⁹) and H2AX (Ser¹³⁹; Fig. 5A). On the other hand, As³⁺-induced phosphorylation of histone H2AX (Ser¹³⁹), TOPK (Thr⁹), and nonphosphorylated TOPK was dramatically decreased in TOPK siRNA-transfected cells (Fig. 5A). The expression of total H2AX or β-actin was not different between these cell lines (Fig. 5A, bottom and middle bottom). Thus, these data indicated that TOPK is required for mediating As³⁺-stimulated phosphorylation of H2AX at Ser¹³⁹.

Phosphorylation of histone H2AX is associated with inhibition of As³⁺-induced apoptosis. H2AX^+/+ and H2AX^−/− cell lines were treated for 24 hours with 2.5 μmol/L As³⁺, and phosphorylation of histone H2AX (Ser¹³⁹) and total histone H2AX was determined by Western blot analysis. As expected, the H2AX^−/− cell line displayed a markedly reduced expression and phosphorylation of histone H2AX compared with the H2AX^+/+ cells (Fig. 5B). However, As³⁺-induced phosphorylation of TOPK and the expression of total histone H3 or TOPK were similar in the two cell lines (Fig. 5B). Furthermore, early apoptosis was significantly less (P < 0.0001) following treatment with 2.5 μmol/L As³⁺ at all time points in H2AX^−/− cells compared with H2AX^+/+ cells (Fig. 5C, bottom right quadrants). This suggested that phosphorylation of H2AX protects cells against As³⁺-induced apoptosis.

TOPK inhibits As³⁺-induced apoptosis. Expression of TOPK was associated with decreased As³⁺-induced apoptosis. At all times after As³⁺ treatment, a significant (P < 0.0001) increase in early apoptosis was observed in TOPK siRNA cells compared with control siRNA-transfected cells (Fig. 5D). This suggested that the absence of TOPK sensitizes cells to As³⁺-induced apoptosis. Taken together, these results indicated that TOPK plays an important role in inhibiting As³⁺-induced apoptosis through its phosphorylation of histone H2AX.

As³⁺-induced apoptosis is dependent on TOPK status. We further confirmed these results by examining the effect of As³⁺ treatment on apoptosis in human melanoma cells, including SK-MEL28, SK-MEL31, SK-MEL5, and RPMI7951 cells, which have varying expression levels of TOPK. RPMI7951 and SK-MEL28 cells, which both express TOPK, showed very little induction of apoptosis by As³⁺ at all time points tested (Fig. 6A). In contrast, apoptosis was markedly induced in SK-MEL31 and SK-MEL5, which express low levels of TOPK (Fig. 6B). To further confirm that the expression level of TOPK has a major influence on whether As³⁺ can induce apoptosis, JB6 Cl41 cells, which do not express TOPK (see Fig. 1), were transfected with pc-DNA3 mock vector (JB6/Vector), pcDNA3-HA-TOPK for overexpression of TOPK, or pcDNA3-HA-TOPK-T9A for overexpression of a mutant TOPK (Tyr⁹ replaced with alanine). Expression of these plasmids was verified by Western blot using a HA antibody (Fig. 6C). Induction of apoptosis by As³⁺ in JB6/Vector cells was ∼60% after 72 hours and ∼42% in JB6-pcDNA3-HA-TOPK-T9A cells (Fig. 6D). In contrast, very little apoptosis occurred in TOPK-overexpressing cells at any time point. This suggested that suppression of TOPK activity by the mutant TOPK-T9A cells almost totally restored the apoptotic response to As³⁺. Overall, these results indicated that TOPK is critical in protecting against As³⁺-induced apoptosis and thus could be an important target in As³⁺ treatment of cancers.

TOPK is highly expressed in lymphoma cells, in myeloid leukemia cells, and in several other highly proliferative malignant cell lines derived from sarcomas, carcinomas, or myelomas of various tissue origins (3, 29). In the present study, we used LC-MS/MS analysis to confirm that His-TOPK could bind with histone H2AX. The in vitro kinase data indicated that TOPK directly interacted with and phosphorylated histone H2AX. This is significant because H2AX has been designated as the histone guardian of the genome (26). Phosphorylation of H2AX has been suggested to have two important functions in DNA repair: to promote changes in the structural configuration of chromatin and to assist in chromatin binding of repair factors. These functions may or may not be related, but both are likely to be required for efficient synapses of broken chromosome ends (26, 30). Several reports have shown that As³⁺ can induce DNA damage (31, 32). Our laboratory has focused on the elucidation of mechanisms explaining how arsenic can act both as a carcinogen and as an effective chemotherapeutic agent (9, 33, 34). Accumulating data suggested that As³⁺ may specifically induce apoptosis in certain types of tumor cells, including megakaryocytic leukemia cell lines, human breast cancer cells (32), and human myeloma cells (11, 34, 35). Furthermore, mitogen-activated protein kinases have been shown to have a very important role in mediating As³⁺-induced apoptosis (36).

TOPK, a novel mitogen-activated protein kinase kinase–like protein kinase (6), is highly expressed in tumor cells but not in normal cells, and the function of this kinase is not clear. In this study, we showed that As³⁺ induced phosphorylation of both TOPK (Thr⁹) and histone H2AX (Ser¹³⁹) and that TOPK directly phosphorylated histone H2AX (Ser¹³⁹) in vitro and in vivo. Further, Western blot and immunocytofluorescence analyses clearly showed that As³⁺ induced the accumulation of phosphorylated TOPK and phosphorylated H2AX in the nucleus of RPMI7951 cells. Another group has also shown that arsenic treatment induces the phosphorylation and nuclear accumulation of histone H2AX (28). However, our data indicated the colocalization and direct involvement of TOPK in As³⁺-induced apoptosis. The accumulation of phosphorylated H2AX in the nucleus suggested that DNA damage occurred in As³⁺-treated RPMI7951 cells. Previous data showed that, after γ-irradiation, nuclei from H2AX^+/+ and H2AX^−/− fibroblasts exhibited 0.25% and 10% nuclear fragmentation, respectively (17). Loss of H2AX has been shown to lead to increased chromosomal abnormalities, deficiencies in gene targeting, and radiation sensitivity. Furthermore, DNA repair was observed to proceed less efficiently in the absence of H2AX, resulting in increased apoptosis (17). In the present study, we showed that apoptosis in H2AX^−/− cells was increased substantially compared with H2AX^+/+ cells, and in TOPK siRNA-transfected cells, As³⁺-induced apoptosis was also markedly increased. These results suggested that TOPK phosphorylation of histone H2AX is associated with an inhibition of As³⁺-induced apoptosis in RPMI7951 melanoma cells.

Investigations using many different experimental systems suggested that arsenic may offer significant therapeutic benefits to patients across many neoplasms, and numerous clinical trials are under way in hematopoietic malignancies and solid tumors. Miller et al. (37) summarized reports of the activity of arsenic in animal models and in the clinic. Low doses of arsenic trioxide (0.06-0.2 mg/kg of body) can induce complete remission in relapsed acute promyelocytic leukemia patients, and complete remissions occurred in 11 of the 12 patients (38). Low doses of As³⁺ induced apoptosis in some (11, 32–34), but not all, cancer cell types (39–41), suggesting that a mechanism exists that is responsible for resistance to As³⁺. Using different melanoma cell lines with different levels of TOPK expression, we found that As³⁺ induces apoptosis dependent on TOPK expression level. In melanoma cells without expression of TOPK, a low concentration of As³⁺ induces rapid apoptosis—more than 25% to 50% in 48 hours. In contrast, melanoma cells expressing high levels of TOPK showed full resistance to As³⁺ following 72 hours of exposure, suggesting that melanoma cell lines expressing TOPK are more resistant to As³⁺-induced apoptosis. These various melanoma cells also display other abnormalities. For example, the p16 gene is deleted in SK-MEL5 (42) cells. Furthermore, B-raf is mutated (V599E) in SK-MEL28 and SK-MEL5 cells but not in SK-MEL31 cells (43, 44). Interestingly, the TOPK-expressing SK-MEL28 and RPMI7951 cells each express a mutant p53 (45, 46), whereas SK-MEL5 cells express wild-type p53 (45). However, additional studies are needed to specifically link the genetic background of these cell lines with resistance to As³⁺-induced apoptosis. Our study indicated that this resistance seems to be related to the activation and phosphorylation of histone H2AX by activated TOPK (Fig. 6E). In this model, cells lacking TOPK expression readily undergo apoptosis induced by low levels of As³⁺. On the other hand, in cells expressing TOPK, As³⁺ induces its phosphorylation, which leads to phosphorylation of histone H2AX and resistance to apoptosis. Thus, discovery of a small-molecule inhibitor of TOPK may be very important in combination with low doses of As³⁺ in the treatment of melanomas and other TOPK-positive cancers.

We thank Dr. Andre Nussenzweig for the H2AX^+/+ and H2AX^−/− murine embryonic fibroblasts, Dr. J. Abe for pcDNA3-HA-TOPK and pcDNA3-HA-TOPK-T9A, Todd Schuster for flow cytometry analysis, Andria Hansen for secretarial assistance, and Paul Hoversten for his help in the writing of the text for this article.