Phosphorylation of serine 1106 in the catalytic domain of topoisomerase II alpha regulates enzymatic activity and drug sensitivity.

Topoisomerases alter DNA topology and are vital for the maintenance of genomic integrity. Topoisomerases I and II are also targets for widely used antitumor agents. We demonstrated previously that in the human leukemia cell line, HL-60, resistance to topoisomerase (topo) II-targeting drugs such as etoposide is associated with site-specific hypophosphorylation of topo II alpha. This effect can be mimicked in sensitive cells treated with the intracellular Ca(2+) chelator, 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA-AM). Here we identify Ser-1106 as a major phosphorylation site in the catalytic domain of topo II alpha. This site lies within the consensus sequence for the acidotrophic kinases, casein kinase I and casein kinase II. Mutation of serine 1106 to alanine (S1106A) abrogates phosphorylation of phosphopeptides that were found to be hypophosphorylated in resistant HL-60 cells or sensitive cells treated with BAPTA-AM. Purified topo II alpha containing a S1106A substitution is 4-fold less active than wild type topo II alpha in decatenating kinetoplast DNA and also exhibits a 2-4-fold decrease in the level of etoposide-stabilized DNA cleavable complex formation. Saccharomyces cerevisiae (JN394t2-4) cells expressing S1106A mutant topo II alpha protein are more resistant to the cytotoxic effects of etoposide or amsacrine. These results demonstrate that Ca(2+)-regulated phosphorylation of Ser-1106 in the catalytic domain of topo II alpha modulates the enzymatic activity of this protein and sensitivity to topo II-targeting drugs.

Topoisomerases alter DNA topology and are vital for the maintenance of genomic integrity. Topoisomerases I and II are also targets for widely used antitumor agents. We demonstrated previously that in the human leukemia cell line, HL-60, resistance to topoisomerase (topo)

II-targeting drugs such as etoposide is associated with site-specific hypophosphorylation of topo II␣. This effect can be mimicked in sensitive cells treated with the intracellular Ca 2؉ chelator, 1,2-bis(2-aminophenoxy)ethane-N,N,N,N-tetraacetic acid (BAPTA-AM).
Here we identify Ser-1106 as a major phosphorylation site in the catalytic domain of topo II␣. This site lies within the consensus sequence for the acidotrophic kinases, casein kinase I and casein kinase II. Mutation of serine 1106 to alanine (S1106A) abrogates phosphorylation of phosphopeptides that were found to be hypophosphorylated in resistant HL-60 cells or sensitive cells treated with BAPTA-AM. Purified topo II␣ containing a S1106A substitution is 4-fold less active than wild type topo II␣ in decatenating kinetoplast DNA and also exhibits a 2-4-fold decrease in the level of etoposidestabilized DNA cleavable complex formation. Saccharomyces cerevisiae (JN394t2-4) cells expressing S1106A mutant topo II␣ protein are more resistant to the cytotoxic effects of etoposide or amsacrine. These results demonstrate that Ca 2؉ -regulated phosphorylation of Ser-1106 in the catalytic domain of topo II␣ modulates the enzymatic activity of this protein and sensitivity to topo II-targeting drugs.
Topoisomerases alter DNA topology for the efficient processing of genetic material (1)(2)(3). These enzymes play a pivotal role in the maintenance of genomic integrity and are essential for many chromosomal functions including DNA replication and recombination, transcription, and chromosome segregation (1)(2)(3). Topoisomerases regulate various cellular processes by relaxing and untangling the intertwined strands of DNA. Two major categories of topoisomerases, type I and type II, have been characterized. The type II enzymes, which consist of two highly homologous isoforms in humans, topoisomerase (topo) 1 II␣ and topo II␤ with molecular masses of 170 and 180 kDa, respectively, catalyze the ATP-dependent transport of one intact DNA double helix through another by creating a transient double-stranded break (1).
The essential nature of topo II␣ during cell proliferation and its ability to cleave DNA in a reversible manner makes topo II␣ an ideal target for agents that poison the enzyme (4,5). In the presence of DNA-damaging topo II-targeting drugs, topo II is converted to a nuclease that irreversibly cleaves duplex DNA. The "poisoning" of the enzyme is via the trapping of the transient reaction intermediate, termed a cleavable complex, which is composed of topo II bound covalently to the 5Ј end of the cleaved DNA strands. This leads to DNA damage, apoptosis, genomic instability, and cell death. Several clinically effective cancer chemotherapeutic agents, daunorubicin, doxorubicin (DOX), amsacrine (m-AMSA), and etoposide (VP-16), stabilize the topo II-DNA cleavable complex and prevent religation of DNA (5). However, a major factor limiting success of topodirected chemotherapeutic regimens is the development of resistance to topo II-targeting drugs, which is frequently observed clinically and in tumor model systems. Thus, an understanding of the mechanism(s) that lead to development of resistance to drugs that poison topo II is essential for improving the therapeutic potential of these agents.
Several different mechanisms, including post-translational modification via phosphorylation, regulate topo II activity. Site-specific phosphorylation of topo II␣, which occurs in a cell cycle phase-dependent manner (6 -9), also modulates drug sensitivity to topo II-targeting drugs (10,11). Whereas Takano et al. (12) reported hyperphosphorylation of topo II␣ in etoposideresistant cells, most other studies (9, 10, 13) have demonstrated a correlation between hypophosphorylation of this enzyme and resistance to the topo II-targeting drugs, DOX, m-AMSA, or VP-16. Hypophosphorylation of topo II␣ is observed in resistant cells in the absence or presence of multidrug resistance gene (MDR1) overexpression and in cells expressing decreased levels of protein kinase C (9,10,13). Decreasing intracellular calcium transients, which mimics the resistant phenotype, also results in site-specific hypophosphorylation of topo II␣ (9,10). A majority of the phosphorylation sites in topo II␣ protein is located within the C-terminal region (3). Casein kinase II has been recognized as the major kinase interacting with and phosphorylating several sites in topo II␣, including Ser-1342, Ser-1376, Ser-1469, and Ser-1524 in human topo II␣ (14 -21). In addition to casein kinase II, protein kinase C has been shown to phosphorylate Ser-29 (8) and a proline-directed kinase phosphorylates Ser-1212, Ser-1246, Ser-1353, Ser-1360, and Ser-1392 (9).
Despite the extensive knowledge of sites in topo II␣ that are phosphorylated, very little is known about the significance of these phosphorylation sites in regulating topo II␣ function. In this study, we employed a combination of in vivo phosphorylation and proteomic approaches to identify the site-specific phosphorylation of topo II␣, which affects enzymatic activity and confers resistance to topo II-targeting drugs. Our results demonstrate that Ser-1106 is a major phosphorylation site in the catalytic domain of topo II␣, hypophosphorylation of which correlates with drug resistance in the human leukemia cell line, HL-60. Mutation of serine 1106 to alanine leads to a decrease in the enzymatic activity and sensitivity to topo IItargeting drugs, thereby establishing the functional significance of the Ser-1106 phosphorylation site in topo II␣.

EXPERIMENTAL PROCEDURES
Materials-The topo II-targeting drugs VP-16 and m-AMSA were obtained from Sigma and the NCI, National Institutes of Health, respectively. Stock solutions of these drugs were prepared in dimethyl sulfoxide (Sigma) and stored frozen at Ϫ20 or 4°C. 1,2-Bis(2-aminophenoxy)ethane-N,N,NЈ,NЈ-tetraacetic acid (BAPTA-AM) was obtained from Calbiochem. 5Ј-Fluoroorotic acid was obtained from Sigma.
Cell Culture-Cultures of wild type HL-60 cells (HL-60/S) were maintained in RPMI 1640 supplemented with 10% fetal bovine serum and 2 mM L-glutamine (BioWhittaker, Gaithersburg, MD) at 37°C in a humidified 5% CO 2 plus 95% air atmosphere. The resistant subline of HL-60 developed by culturing the wild type cells in increasing concentrations of 0.025-0.05 g/ml DOX has been described previously (10). Following in vitro selection, the DOX-resistant subline (HL-60/R), which is 40-fold resistant to the cytotoxic effects of VP-16 compared with the HL-60/S cells, was routinely cultured in the absence of DOX. Under these conditions the drug-resistant phenotype of HL-60/R cells was not altered following in vitro culture for at least 3 months.
Human Topoisomerase II␣ Plasmids-Two plasmids, pHT212 (22) and pYEpWob6 (23), containing wild type human topo II␣ sequences were employed. The plasmid pHT212 also contains a c-myc and hexahistidine tag in the C terminus (22), and the plasmid pYEpWob6 has a URA3 selection marker (23). For mutating Ser-1106 to alanine (S1106A) in topo II␣ in these plasmids, site-directed mutagenesis was carried out using the Quikchange Site-directed mutagenesis kit (Stratagene Inc., La Jolla, CA). The primers used for mutation are as follows: 5Ј-CA GAT GAA GAA GAA AAT GAA GAG GCT GAC AAC GAA AAG GAA ACT G-3Ј and 5Ј-C AGT TTC CTT TTC GTT GTC AGC CTC TTC ATT TTC TTC TTC ATC TG-3Ј. The site-directed mutagenesis of S1106A was confirmed by PCR and DNA sequence analysis.
Metabolic Labeling with [ 32 P]Orthophosphoric Acid-HL-60 or BJ201 yeast cells were metabolically labeled with [ 32 P]orthophosphoric acid to obtain in vivo 32 P-labeled phosphorylated topo II␣ protein. Log phase cultures of HL-60/S or HL-60/R cells were first incubated in phosphate-free RPMI supplemented with 10% dialyzed fetal bovine serum and 2 mM L-glutamine for 1 h at 37°C. Cells were then labeled with 100 Ci/ml of carrier-free [ 32 P]orthophosphoric acid (PerkinElmer Life Sciences) for an additional 2 h. In experiments involving treatment with BAPTA-AM, HL-60/S cells were incubated with 20 M BAPTA-AM during the 2-h labeling period. The BJ201 yeast cells, expressing WT or S1106A-MT human topo II␣ protein, were incubated overnight at 30°C with shaking (250 rpm) in synthetic dropout liquid medium lacking leucine. The overnight cultures were transferred to YPDA without phosphate medium and incubated at 30°C with shaking to a cell density corresponding to 0.6 units (A 600 ). Following centrifugation, cells were resuspended into 20 ml of YPDA medium without phosphate containing 5 mCi of [ 32 P]orthophosphoric acid and incubated at 30°C for 1 h with shaking.
Purification of Human Topo II␣-HL-60 cells were lysed at 4°C for 30 min in RIPA buffer (50 mM Tris-HCl, pH 8.0, 425 mM NaCl, 1% Nonidet P-40, 0.5% deoxycholate, 0.1% SDS, and 10 mM 2-mercaptoethanol) supplemented with a mixture of proteases and phosphatase inhibitors (10,11). The lysate was centrifuged at 100,000 ϫ g for 30 min, and topo II␣ present in the supernatant was incubated overnight at 4°C with a topo II␣-specific polyclonal antibody (25) and protein A-agarose. The antigen-antibody complex was dissociated in lithium dodecyl sulfate sample buffer (Invitrogen), and topo II␣ was purified by gel electrophoresis in Tris acetate gels. Following transfer to nitrocellulose membrane (0.45 M), the topo II␣ protein band, visualized by staining with 0.025% Coomassie Brilliant Blue R-250, was excised and used for phosphopeptide mapping.
Recombinant topo II␣ from yeast cells was isolated after freezing the cells in liquid nitrogen. The frozen cell pellet stored at Ϫ80°C was lysed by gentle rotation at 4°C for 40 min in 2-3 volumes of Y-PER lysis buffer (Pierce) supplemented with 40 mM imidazole, 20 mM ␤-mercaptoethanol, and a mixture of protease and phosphatase inhibitors. The cell lysate was centrifuged at 13,000 ϫ g for 10 min at 4°C, and topo II␣ protein in the supernatant was purified by Ni 2ϩ -nitrilotriacetic acid (Ni-NTA)-agarose column (Qiagen, Valencia, CA) chromatography. After the column was washed twice with wash buffer (40 mM imidazole, 20 mM sodium phosphate, 0.5 M NaCl, pH 7.4), the protein was eluted with six successive 1-ml portions of the elution buffer (200 mM imidazole, 20 mM sodium phosphate, 0.5 M NaCl, pH 7.4). The fractions from eluates 2-4 were pooled and concentrated at 3000 ϫ g for 15 min using an Amicon Ultra-4, 30K NMWL concentrator (Millipore Inc., Milford, MA). The retentate was recovered and stored at Ϫ20°C in 40% glycerol. Total protein content was determined using the Bio-Rad protein assay reagent (Bio-Rad), and purity was assessed by SDS-PAGE followed by staining with Coomassie Brilliant Blue R-250. To determine the relative amount of WT and S1106A-MT protein present in purified preparations, Western blot analysis was also carried out using polyclonal topo II␣-specific antibody and horseradish peroxidase-labeled secondary antibody.
Phosphopeptide Mapping of Topo II␣-32 P-Labeled topo II␣ from HL-60 or BJ201 cells was purified by SDS-PAGE and transferred to nitrocellulose membrane as described above. The 170-kDa band of topo II␣ on the nitrocellulose membrane, identified by staining with Coomassie Brilliant Blue R-250, was excised and used for proteolysis by CNBr or trypsin. CNBr digestion was carried out at 47°C for 90 min in the presence of 250 -400 l of a solution containing 160 g/ml CNBr in 70% formic acid. Following the incubation, the peptides released into the supernatant were concentrated by evaporation in a Savant Speed-Vac and separated by one-dimensional gel electrophoresis on Tris-Tricine peptide gels. The separated peptides were transferred to a PVDF membrane and subjected to autoradiography or Cyclone image analysis (PerkinElmer Life Sciences) to compare the phosphopeptide profiles of topo II␣ obtained from HL-60 cells treated under different conditions or yeast cells transformed with WT or S1106A-MT topo II␣.
For tryptic digestion, the 170-kDa topo II protein bound to the nitrocellulose membrane was incubated overnight with L-1-tosylamido-2phenylethyl chloromethyl ketone-trypsin (0.5 g) in 50 l of 1% ammonium bicarbonate. Two-dimensional phosphopeptide mapping (11,18,26) was carried out by electrophoresis with pH 1.9 buffer in the horizontal dimension and chromatography in the vertical dimension using phosphochromatography buffer (n-butyl alcohol/pyridine/acetic acid/ deionized water, 6.4:5:1:4, v/v)). Image density of the phosphopeptides was quantified with a Cyclone imager.
N-terminal Edman Sequencing and Mass Spectrometry-CNBr phosphopeptides were separated by gel electrophoresis, transferred to a PVDF membrane, and stained with Coomassie Brilliant Blue R-250. The stained bands were excised and subjected to N-terminal Edman sequencing (27,28). The N-terminal sequencing of the peptide fragments was carried out on a Procise, model 492, protein sequencer (Applied Biosystems, Foster City, CA) fitted with a 140c microgradient system, 785A programmable absorbance detector, and a 610A (version 2.1) data analysis system. The internal peptide sequences were searched and identified by BLASTP program of the NCBI. To determine whether the sequenced peptides were phosphorylated, the membrane was autoradiographed before and after cutting out the stained bands.
LC-tandem MS analysis was carried out following tryptic digestion of the 170-kDa topo II␣ protein or CNBr peptides of interest, which were excised from SDS-polyacrylamide gels (29,30). This procedure was accomplished using a Finnigan LCQ-Deca ion trap mass spectrometry system equipped with an electrospray ion source interfaced to a 10cm ϫ 50-m inner diameter C18 capillary high pressure liquid chromatography column. The digests were analyzed by LC-tandem MS with data-dependent acquisition methods to map as many peptides as possible from the protein. In these analyses, the LC-tandem MS data were searched versus the topoisomerase sequence using the program Sequest. Phosphopeptides were recognized by combinations of characteristic additions of 80-Da multiples (depending on the degree of phosphorylation) to peptide molecular weights and/or characteristic neutral losses of 98 Da for H 3 PO 4 from phosphoserine and phosphothreonine in the collisionally induced dissociation (CID) spectra. The MALDI-TOF analyses were carried out using a Micromass TofSpec 2E matrix-assisted laser desorption ionization-time-of-flight (MALDI-TOF) mass spectrometry system. Digests desalted using ZipTips were eluted with a solution of ␣-cyano-4-hydroxycinnamic acid as the matrix before spotting on the target plate.
The data were analyzed using CID spectra to search NCBI nonredundant data base with the search program TurboSequest. Spectra from samples analyzed by MALDI-TOF were internally calibrated using trypsin autolysis peptides, giving mass accuracy that was generally better than 25 ppm. These spectra were used for data base searches with the program MASCOT.
Decatenation of Kinetoplast DNA (kDNA) by Topo II␣-Topo II␣ enzymatic activity was assayed by measuring the decatenation (31) of kDNA (Topogen, Columbus, OH). A standard assay carried out in a total volume of 20 l included 50 mM Tris-HCl, pH 7.9, 88 mM KCl, 10 mM MgCl 2 , 0.5 mM EDTA, 10 mM ATP, 10 mM dithiothreitol, 100 g/ml bovine serum albumin, and 300 ng of kDNA. The reaction mixture containing varying amounts of WT or S1106A-MT topo II␣ was incubated at 37°C, and at timed intervals the reaction was stopped by the addition of 5 l of stop solution (5% SDS, 25% Ficoll, and 0.05% bromphenol blue). The samples were resolved by electrophoresis at 115 V using a 1% agarose gel in Tris acetate EDTA buffer. Following electrophoresis, the gel was stained with ethidium bromide and photographed under UV illumination, and the amount of decatenated minicircles of kDNA was quantified using an AlphaInnotech Image analyzer (Al-phaInnotech Corp., San Leandro, CA).
Precipitation of Covalent Topo II␣-DNA Complex-Formation of covalent topo II-DNA complex by WT or S1106A-MT topo II␣ was determined by the precipitation of 3Ј-end-labeled 32 P-pcDNA3 in a cell-free system as described by Zwelling et al. (32). Briefly, pcDNA3 was linearized with EcoRI and 3Ј-end-labeled with [ 32 P]dATP. The 3Ј-endlabeled pcDNA3 was incubated with 10 -200 ng of WT or S1106A-MT topo II␣ in the absence or presence of 1 mM ATP. To determine drugstabilized DNA-cleavable complex formation in the presence of 1 mM ATP, 25 or 100 ng of WT or S1106A-MT topo II␣ was incubated with 1-100 M VP-16 for 30 min at 37°C (10,11,32). The reaction was stopped by the addition of SDS, and the protein-DNA complex was precipitated by the addition of KCl (10,11,32). The precipitate was washed twice with 100 mM KCl and dissolved in water at 65°C, and the solution was added to Ecolume (ICN Pharmaceutical, Costa Mesa, CA). The level of radiolabel was then determined using a liquid scintillation counter.
Drug Sensitivity Test in Yeast JN394t2-4 Strain-S. cerevisiae strain JN394 t2-4 transformed with either WT or S1106A-MT topo II␣ pYEp-Wob6 (23) plasmid was cultured at 30°C in synthetic dropout liquid medium without uracil. The cells were collected by centrifugation and resuspended in YPDA medium. After adjusting the cell number to 2 ϫ 10 6 cells/ml, cells were treated for 24 h at 35°C with VP-16 (0 -200 M) or m-AMSA (0.5-25 M) at a final concentration of 2% Me 2 SO. Following treatment, the control or treated cultures were diluted 1000-fold with sterile water and plated in triplicate using YPD agar in 100 ϫ 15-mm Petri dishes. After incubation for 3-4 days at 35°C, the number of colonies on the control and treated drug plates were counted in an AlphaInnotech image analyzer (AlphaImager TM , Alpha Innotech Corp., San Leandro, CA). We have confirmed that the JN394t2-4 yeast cells are not viable at 35°C and growth at 35°C occurs only following transformation with WT pYEpWob6 or S1106A-MT pYEpWob6 plasmid. Colony growth of the JN394t2-4 on YPDA is also similar following transformation of the WT pYEpWob6 or S1106A-MT pYEpWob6 plasmid. Topo II␣ protein levels in the JN394t2-4 cells transformed with WT pYEpWob6 or S1106A-MT pYEpWob6 plasmid were determined in cell lysates following SDS-PAGE and staining with Coomassie Brilliant Blue R-250 and Western blotting with a topo II␣-specific antibody (25).

CNBr Peptide 34 Is Hypophosphorylated in HL-60/S Cells Treated with BAPTA-AM and in HL-60/R Cells-Previously
we have demonstrated the presence of two hypophosphorylated tryptic peptides in a derivative of HL-60 cells (HL-60/R) that are resistant to the topo II-targeting drug, VP-16 (10,11). However, the location of the phosphorylation site(s) in topo II␣ was not defined. To identify the region in topo II␣ that contained these sites, we compared the one-dimensional maps of CNBr phosphopeptides generated from 32 P-labeled topo II␣ immunoprecipitated from HL-60/S cells, HL-60/S cells treated with BAPTA-AM, and HL-60/R cells. Our results demonstrated that a CNBr phosphopeptide of an approximate molecular mass of 12 kDa was hypophosphorylated in HL-60/S cells treated with BAPTA-AM and HL-60/R cells as compared with HL-60/S cells (Fig. 1). Although the decrease in phosphoryla- tion of this peptide was modest, it was consistently lower (30% when normalized to the phosphorylated peptide migrating at 25 kDa) than that observed for other peptides. The 12-kDa hypophosphorylated peptide was identified by N-terminal Edman sequencing to correspond to CNBr peptide 34 (amino acids 1041-1131) of a calculated molecular mass of 10.4 kDa. This peptide is located in the catalytic domain of topo II␣. Twodimensional phosphopeptide maps of complete tryptic digests of topo II␣ protein from HL-60/S cells or HL-60/S cells treated with BAPTA-AM revealed hypophosphorylation of two peptides (peptide 2 and 3, Fig. 1B) in HL-60/S cells treated with BAPTA-AM compared with HL-60/S cells. These two hypophosphorylated peptides are identical to those observed in HL-60/R cells and are also present in tryptic digests of peptide 34 (data not shown).
Serine 1106 Is the Major Phosphorylation Site Present in CNBr Peptide 34 -To determine the specific hypophosphorylation site in peptide 34, we performed LC-tandem mass spectrometry. This analysis led to the identification of a phosphorylated serine, Ser-1106, located in the catalytic domain of topo II␣. The Ser-1106 phosphorylation site was detected initially in the 12-kDa CNBr fragment of topo II␣ by data-dependent analysis. The CID spectrum of the peptide, VPDEEENEEpSDNEK is shown in Fig. 2. In addition, another peptide, VP-DEEENEEpSDNEKETEK, was detected. The detection of two phosphorylated peptides containing Ser-1106 (generated because of partial proteolysis) provides an explanation for the presence of two hypophosphorylated peptides in the two-dimensional tryptic phosphopeptide maps (Fig. 1B). The CID spectrum of these peptides is typical for a serine or threonine phosphopeptide, with a high abundance ion due to the loss of H 3 PO 4 and low abundance (but significant) sequence ions. Based on this spectrum, a highly specific, selected, reactionmonitoring scheme was designed and used to detect this phosphopeptide in the more complex digest of intact topo II␣.

Hypophosphorylated Site in CNBr Peptide 34 and in Tryptic Peptides 2 and 3 Corresponds to Ser-1106 -Because
Ser-1106 was identified as the major phosphorylation site in peptide 34, we determined the effect of mutation of this site to alanine on the phosphorylation of CNBr peptide 34 and tryptic peptides 2 and 3. Although the mammalian system was the basis of our findings, the facile expression of human topo II␣ and the absence of topo II␤ in yeast cells made this system an attractive model to carry out comparative studies of WT and S1106A-MT topo II␣ protein. BJ201 yeast cells expressing WT or S1106A-MT topo II␣ protein were metabolically labeled with [ 32 P]orthophosphoric acid, and recombinant topo II␣ protein purified from these cells was digested with CNBr or trypsin. The maps of the CNBr digests characterized by one-dimensional SDS-PAGE (Fig. 3A) showed that peptide 34 was phosphorylated only in digests of WT topo II␣ protein but not of S1106A-MT protein. Comparison of the tryptic digests (Fig. 3B) by two-dimensional phosphopeptide mapping revealed the presence of the phosphorylated peptides 2 and 3 (inset) in the WT topo II␣ protein sample but not in the S1106A-MT protein sample. This finding indicates that hypophosphorylation of CNBr peptide 34 and of tryptic peptides 2 and 3, observed in HL-60 cells resistant to topo II-targeting drugs, occurs at Ser-  3. Differential phosphorylation of CNBr peptide 34 and tryptic peptides 2 and 3 generated from WT and S1106A-MT topo II␣ protein expressed in BJ201 yeast cells. BJ201 yeast cells expressing WT or S1106A-MT protein were metabolically labeled with [ 32 P]orthophosphoric acid. The recombinant protein from these cells was purified by Ni 2ϩ -NTA column chromatography and subjected to SDS-PAGE. Following transfer to a nitrocellulose membrane, the stained topo II␣ band was excised and digested with CNBr or trypsin. A, samples of CNBr digests of WT (lane 1) or S1106A-MT (lane 2) topo II␣ were electrophoresed on SDS-polyacrylamide gels (10 -20% Tris-Tricine gels), transferred to PVDF membrane, and autoradiographed. B, samples of tryptic digests of WT or S1106A-MT topo II␣ protein were analyzed by two-dimensional phosphopeptide mapping and autoradiography.
FIG. 4. Purification of WT and S1106A-MT topo II␣ from BJ201 yeast cells. Extracts of yeast cells expressing WT or S1106A-MT topo II␣ were purified by Ni 2ϩ -NTA column chromatography. An aliquot (1.2 g) of the purified preparation of WT (lanes 1 and 3) or S1106A-MT (lanes 2 and 4) topo II␣ was subjected to SDS-PAGE and transferred to a PVDF membrane. The membrane was stained with Coomassie Brilliant Blue R-250 (lanes 1 and 2) or immunoblotted with topo II␣-specific antibody (lanes 3 and 4).
1106. It is interesting to note that the CNBr and tryptic phosphopeptide maps of human topo II␣ from HL-60 and yeast cells (Figs. 1 and 3) are similar, although the relative intensities of individual phosphopeptides varies depending on the source of the topo II␣ protein. The only major difference is the absence of a 4-kDa phosphopeptide in the CNBr maps and of phosphopeptide 1 in the tryptic maps of the recombinant protein (WT or S1106A-MT) expressed in yeast, as compared with native topo II␣ expressed in HL-60 cells. This discrepancy is due to differential migration of the C-terminal CNBr or tryptic peptide, which is larger in the recombinant protein (due to the presence of additional amino acids transcribed from linker sequences) compared with the native protein. Indeed the Cterminal CNBr peptide has a molecular mass of 4 kDa and the C-terminal tryptic peptide phosphorylated at Ser-1524 corresponds to peptide 1 (18). S1106A-MT Topo II␣ Exhibits Reduced Enzymatic Activity and Formation of Protein-DNA Cleavable Complex-We next examined the functional role of Ser-1106 phosphorylation by comparing the catalytic activity and level of drug-stabilized, DNA-cleavable complex formation of recombinant WT or S1106A-MT topo II␣. For these studies the WT and S1106A-MT topo II␣ proteins were expressed in the BJ201 yeast strain and purified by metal ion affinity chromatography using Ni 2ϩ -NTA columns. This procedure led to significant purification of topo II␣ protein, which was present in equivalent amounts in the WT and S1106-MT protein preparation, as judged by Coomassie Brilliant Blue R-250 staining and Western blot analysis (Fig. 4). The catalytic activity of equivalent amounts of purified WT or S1106A-MT topo II␣ protein was determined using 300 ng of kDNA as the substrate (31). As seen in Fig. 5, 4-fold more S1106A-MT topo II␣ protein, as compared with WT topo II␣ protein, was required to produce equivalent levels of decatenated minicircles from kDNA (Fig. 5,  A and B). Comparison of the time course of relative decatenation activity by WT and S1106A-MT topo II␣ revealed a 4-fold higher rate of reaction for WT, compared with S1106A-MT, topo II␣ protein, based on the initial slope of the reaction curves (Fig. 5C). In these experiments (Fig. 5C), to observe measurable decatenation in the linear range of the slope, 50 ng of the mutant protein and 12.5 ng of the WT protein were required.
Because the S1106A-MT protein was enzymatically less active than the WT protein, we next determined whether these proteins also differed in their ability to form DNA-cleavable complex in the absence or presence of drugs. In the presence of 1 mM ATP, the level of topo II DNA-cleavable complex formation with the S1106A-MT topo II␣ was 2-fold reduced compared with that of the WT protein (Fig. 6A). No difference was observed in the absence of ATP (Fig. 6A). In the presence of ATP of WT or S1106A-MT topo II␣ protein were incubated with kinetoplast DNA (kDNA). A, an aliquot of the reaction mixture was electrophoresed on a 1% agarose gel to separate the kDNA substrate from decatenated (minicircles) DNA, and the DNA bands were visualized by UV illumination of ethidium bromide-stained gels. The relative intensity of the bands was determined using an AlphaInnotech image analyzer. B, the percent decatenation, calculated as the ratio of the intensity of the decatenated band to the total intensity of the substrate band plus the decatenated band, was plotted versus concentration of WT (q) or S1106A-MT (E) topo II␣. C, time course of percent relative decatenation/ng of purified WT (q) or S1106A-MT (E) topo II␣.

FIG. 6. Comparison of the level of formation of DNA-cleavable complex by WT and S1106-MT topo II.
A, varying concentrations of purified WT (closed symbol) or S1106A-MT (open symbol) topo II␣ were incubated with 32 P-labeled pcDNA3 in the absence (OE, ‚) or presence of 1 mM ATP (q, E). B, WT (f) or S1106A-MT (Ⅺ) topo II␣ protein (25 or 100 ng) was incubated with 32 P-labeled pcDNA3, 1 mM ATP, and varying concentrations of VP-16. The protein-DNA complex was precipitated by the addition of KCl, and the counts/min present in the precipitate were determined by liquid scintillation counting. and the topo II-targeting drug, VP-16, a significant (p ϭ 0.01) 2-4-fold decrease in drug-stabilized, DNA-cleavable complex formation was observed with the S1106A-MT compared with WT topo II␣ protein (Fig. 6B). This difference was seen using two different concentrations of the topoII␣ protein (25 and 100 ng as indicated above the panels in Fig. 6B).

S1106A-MT Topo II␣ Transformed in the JN394t2-4 Yeast Strain Is Resistant to the Cytotoxic Effects of the Topo IItargeting Drugs, VP-16 and m-AMSA-
To confirm that hypophosphorylation of topoII␣ at Ser-1106 confers a resistant phenotype in vivo, we tested drug sensitivity of yeast cells expressing WT or S1106A-MT topo II␣ protein. Sensitivity to two topo II-targeting drugs, VP-16 and m-AMSA, was determined in the yeast strain JN394t2-4 transformed with the pYepWob6 construct of WT or S1106A-MT human topo II␣ protein. At 35°C JN394t2-4 yeast cells do not grow unless the human recombinant topo II␣ is expressed. The survival data in Table I demonstrate that JN394t2-4 cells transformed with S1106A-MT topo II␣ are more resistant to the cytotoxic effects of different concentrations of VP-16 (up to 12-fold higher survival) and m-AMSA (up to 20-fold higher survival) than JN394t2-4 cells transformed with WT topo II␣. This survival difference is not due to differential cellular expression of the WT and S1106A-MT protein, because extracts made from identical numbers of yeast cells transformed with the pYEpWob6 construct of human WT or S1106A-MT topo II␣ express equivalent amounts of topo II␣ protein (data not shown).

DISCUSSION
Post-translational modification of topo II␣ by reversible phosphorylation is a key mechanism regulating its function. In this study, we identified a novel phosphorylation site, Ser-1106, in the catalytic domain of topo II␣. Mutation of this site to alanine leads to a reduction in the catalytic activity of topo II␣ and level of formation of enzyme-DNA-cleavable complex. This results in decreased sensitivity to topo II-targeting drugs in vivo. To our knowledge this is the first report identifying a phosphorylation site in the catalytic domain of topo II␣ that is capable of modulating its function.
Our previous observation demonstrating that specific sites in topo II␣ protein were hypophosphorylated in cells resistant to topo II-targeting drugs (10,11) provided the impetus for defining phosphorylation site(s) that were functionally relevant for topo II␣ function, in particular regulation of sensitivity to topo II-targeting drugs. By using an integrated approach involving CNBr and tryptic phosphopeptide mapping of in vivo 32 P-labeled topo II␣, N-terminal Edman sequencing of CNBr peptides, and mass spectrometry, we were initially able to locate a region in the catalytic domain of the enzyme (amino acids 1041-1131) that harbored the hypophosphorylated site. Phosphorylation of this region could also be manipulated by altering intracellular Ca 2ϩ transients using BAPTA-AM, which mimicked the resistant phenotype (10,11). Mass spectrometric analysis of this region led to the identification of Ser-1106 as a major phosphorylation site. The functional significance of this site was established by comparing the activity of WT and Ser-1106-MT topo II␣ protein in vitro and drug sensitivity of yeast cells transformed with these proteins in vivo.
Previous studies have identified several phosphorylation sites in topo II␣, primarily in the C-terminal region. Consistent with the physiologic function of topo II␣, phosphorylation of several sites is cell cycle phase-regulated (6 -9). Although several kinases have been shown to phosphorylate topo II␣, casein kinase II has emerged as the major kinase interacting with and phosphorylating several sites, including Ser-1342, Ser-1367, Ser-1469, and Ser-1524 in human topo II␣ (14 -21). In addition to casein kinase II, protein kinase C has been shown to phosphorylate Ser-29 (8), and a proline-directed kinase phosphorylates Ser-1212, Ser-1246, Ser-1353, Ser-1360, and Ser-1392 (9). Identification of most of these sites was based on in vitro studies employing purified protein kinases. However, in vivo phosphorylation of these sites was confirmed by matching tryptic phosphopeptides generated following phosphorylation in vitro and in vivo (8,9). Despite these extensive studies, the functional significance of phosphorylation at these sites in vivo remains unclear. Although casein kinase II is capable of modulating the activity of mammalian topo II␣, the mechanism by which this occurs is thought to involve stabilization of topo II␣ but not phosphorylation per se (33,34). Indeed, it has been shown that the C-terminal domain is not important for enzymatic activity, because deletion of this region or mutation of Ser-1376 and/or -1524 does not render the enzyme inactive (34). Rather, it has been proposed that phosphorylation at these sites may be important for subcellular localization of topo II␣ (34). Furthermore, phosphorylation of Ser-29 does not affect the ATPase activity of topo II␣, which is localized to the N-terminal region and required for the final religation step in the enzymatic reaction (35).
Resistance of tumor cells to topo II-targeting drugs have primarily focused on (a) overexpression of MDR1, which encodes P-glycoprotein resulting in enhanced drug efflux, and (b) point mutations or truncation in the topo II␣ gene (5). Indeed, point mutations identified in model systems resistant to topo II-targeting drugs have been shown to confer drug resistance when tested in the JN394t2-4 yeast system (24). The functional role of site-specific phosphorylation on sensitivity of topo II␣ to DNA-cleavable complex formation by topo II-targeting drugs in vitro or in vivo has not been addressed. In general, hypophosphorylated (10,13) or hyperphosphorylated (12,36) topo II␣ has been correlated with drug insensitivity. Our results on resistance to topo II-targeting drugs in vitro and in vivo with S1106A-MT topo II␣ provide evidence for a regulatory role of site-specific phosphorylation in sensitivity to topo II-targeting drugs. Thus, hypophosphorylation of topo II␣ may be responsible for the lack of response of cancer patients to treatment with topo II-targeting drugs. Unlike S1106A-MT, the double mutant S1376A and S1524A topo II␣ enzyme that is enzymatically active (34) is not resistant to the topo II-targeting drugs VP-16 or m-AMSA when tested in the JN394t2-4 yeast system (data not shown). Ser-1106 is flanked by acidic amino acids, thereby fulfilling the consensus sequence requirement for both casein kinase I and casein kinase II (37). Thus these enzymes could serve as potential physiologic kinases regulating phosphorylation of Ser-1106. Although casein kinase II has been shown to phosphorylate several sites in the C-terminal region of topo II␣, the role of casein kinase I has not been evaluated, despite the presence of several casein kinase I consensus sites. Both casein kinase I and casein kinase II play an important role in regulating numerous cellular events. However, the presence of different isoforms of casein kinase I allows for a more diverse mechanism by which this enzyme can mediate signaling events. Of particular relevance to this study is the ability of two isoforms of casein kinase I, casein kinase I␦ and casein kinase I⑀ (but not casein kinase II), to be activated by Ca 2ϩ -dependent dephosphorylation or proteolysis (38,39). Indeed, it has been reported that Ca 2ϩ -dependent dephosphorylation of casein kinase I⑀ by calcineurin regulates phosphorylation and activation of DARP32 by metabotropic glutamate receptors in neostriatal neurons (40,41). A mechanism similar to this could provide for an explanation for Ca 2ϩ -dependent phosphorylation of Ser-1106 by casein kinase I␦ or casein kinase I⑀, which is supported by our preliminary data demonstrating decreased phosphorylation of CNBr and tryptic peptides containing Ser-1106 by two inhibitors of casein kinase I, CKI-7 and IC-261. 2 However more detailed studies are required for identifying the physiologic kinases(s) required for phosphorylation of Ser-1106.
In summary, results from the present study demonstrate the significance of Ser-1106 phosphorylation in regulating topo II␣ function, viz. decatenation of DNA and formation of drug-stabilized DNA cleavable complex. Although the S1106A mutant topo II␣ protein exhibits decreased enzymatic activity, it is able to complement growth of yeast cells. This observation suggests that the presence of adequate cellular levels of S1106A-MT topo II␣ protein can compensate for attenuation in enzymatic activity. Alternatively it is possible that other mechanisms in conjunction with Ser-1106 phosphorylation may be involved in regulating topo II␣ function. Our results implicating the importance of Ser-1106 phosphorylation in modulating drug sensitivity suggest that mechanisms upstream of Ser-1106 phosphorylation may be altered in clinical resistance to topo IItargeting drugs. Thus a future challenge will be the identification of the relevant kinase(s) and/or phosphatase(s) involved in regulating phosphorylation of this site, which could aid in the design of novel treatment strategies for tumors resistant to topo II-targeting drugs.