Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The DNA topoisomerases (topo) alter DNA topology for the processing of genetic material and are key targets for the clinically important antineoplastic agents amsacrine (m-AMSA) and etoposide (VP-l6) (Chen and Liu, 1994; Watt and Hickson, 1994; Wang, 1996). Although topo I exists as a single 97-kDa protein, topo II has an - and -isoform with molecular mass of 170 and 180 kDa, respectively (Watt and Hickson, 1994; Wang, 1996). Although a substantial literature exists on the regulation, function, and drug interactions of the -isoform, the role of the -isoform has received substantially less attention (Chen and Liu, 1994; Watt and Hickson, 1994; Wang, 1996).

A role for topo II in cell differentiation was originally proposed by Woessner et al. (1990, 1991). More recently, the altered regulation of topo II and the pronounced up-regulation of topo II during all trans-retinoic-induced differentiation of human leukemia HL-60 cells has been reported (Aoyama et al., 1998b). Previous studies (Cornarotti et al., 1996; Dereuddre et al., 1997) have suggested that the - and -isoforms of topo II may represent distinct targets that govern differential sensitivity to drugs that poison the enzyme. A more recent study (Herzog et al., 1998) has provided evidence that high levels of resistance to m-AMSA in a subline of HL-60 is correlative with the absence of detectable levels of the topo II protein. A notable observation of this study was the finding that the absence of topo II does not interfere with cell proliferation (Herzog et al., 1998).

HL-60 cells that exhibit increased resistance to doxorubicin (DOX) have been isolated (Ganapathi et al., 1996a). Although these resistant cells exhibit decreased drug accumulation due to overexpression of P-glycoprotein, the expression of resistance to topo II poisons is correlative not with intracellular drug levels but with decreased drug-stabilized topo II-DNA cleavable complex formation (Ganapathi et al., 1996a; Aoyama et al., 1998) and functional alterations in topo II. The reduced DNA damage also has been found to be related to site-specific hypophosphorylation of topo II (Aoyama et al., 1998). Although these changes in drug accumulation and hypophosphorylation of topo II are possibly linked, the differential sensitivity to m-AMSA compared with VP-16 remains unexplained in the HL-60 cells. In the present study, we have investigated the regulation of topo II as well as the differential sensitivity of the - and -isoforms to m-AMSA and VP-16 with isoform-specific antisera in sensitive (HL-60/S) and DOX-resistant (HL-60/DOX0.05) HL-60 cells. Results suggest that unlike topo II, both overall and site-specific hyperphosphorylation of topo II is observed in the HL-60/DOX0.05 cells. Furthermore, in both the HL-60/S and HL-60/DOX0.05 cells, m-AMSA was >2-fold more effective than VP-16 in topo II-stabilized DNA cleavable complex formation (based on band depletion experiments) and cytotoxicity in a soft-agar colony assay.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

The wild-type HL-60 (HL-60/S) cells were obtained from Dr. Andrew Yen, College of Veterinary Medicine, Cornell University, Ithaca, NY. Cultures of HL-60/S cells were maintained in RPMI 1640 supplemented with 10% fetal bovine serum and 2 mM L-glutamine (BioWhittaker, Walkersville, MD) at 37°C in a humidified 5% CO₂ plus 95% air atmosphere. The resistant subline of HL-60 developed by culturing the wild-type cells in increasing concentrations of 0.025 to 0.05 µg/ml DOX has been described previously (Ganapathi et al., 1996b). The DOX-resistant subline (HL-60/DOX0.05) was maintained in the absence of DOX during experimentation. Doubling time in vitro of the HL-60/S and HL-60/DOX0.05 cells was 18 to 20 h.

The enrichment of cells in G₁, S, and G₂ + M phases of the cell cycle was carried out by centrifugal elutriation (Hengstschlager et al., 1997) in a J2-21 centrifuge equipped with a JE-6 rotor (Beckman Coulter, Fullerton, CA). Briefly, cells (2 × 10⁸) were loaded at a rotor speed of 2000 and 1875 rpm for the HL-60/S and HL-60/DOX0.05 cells, respectively. Fractions were collected with incremental increases in flow rate. The cells in each fraction were analyzed for cell cycle phase distribution by flow cytometry (Kawamura et al., 1996). Fractions containing cells in the cell cycle phase of interest were pooled and used for experiments on topo II protein levels and phosphorylation.

The effect of VP-16 or m-AMSA on induction of topo II-mediated DNA scission was determined by measuring precipitation of the protein DNA complex by a modification of the SDS-KCl technique (Zwelling et al., 1989; Ganapathi et al., 1996a; Aoyama et al., 1998). Cells were labeled for 24 h with 0.02 to 0.04 µCi/ml of [¹⁴C]thymidine, specific activity 53 mCi/mmol (Amersham, Arlington Heights, IL). The labeled HL-60/S cells were treated with 0.1 to 5.0 µM m-AMSA or VP-16 for 1 h and processed for measuring DNA damage.

Cytotoxicity studies in vitro were carried out by a soft-agar colony assay (Ganapathi et al., 1996a,b) after exposure to m-AMSA or VP-16 for 1 h. The colony-forming efficiency of the HL-60 cells under these conditions was 29%.

Levels of 170 kDa () and 180 kDa () topo II protein in HL-60/S and HL-60/DOX0.05 cells were determined by SDS-polyacrylamide gel electrophoresis (PAGE) and immunoblotting (Ganapathi et al., 1996). Briefly, extracts from HL-60/S or HL-60/DOX0.05 cells were prepared in radioimmunoprecipitation assay buffer with whole cells or nuclei isolated in nucleus buffer supplemented with 0.3% Triton X-l00 (Ganapathi et al., 1996; Aoyama et al., 1998a,b). Protein content was determined by the Coomassie assay reagent (BioRad, Hercules, CA) and serial dilutions containing equivalent amounts of protein from HL-60/S and HL60/DOX0.05 cells were analyzed by SDS-PAGE (Ganapathi et al., 1996a; Aoyama et al., 1998a,b). After electroblotting onto nitrocellulose, the topo II protein ( and ) was detected after incubation with rabbit polyclonal antibodies that recognize either the 170-kDa (Ganapathi et al., 1996a; Turley et al., 1997; Aoyama et al., 1998b) or the 180-kDa (Turley et al., 1997) topo II followed by ¹²⁵I-goat anti-rabbit IgG. The specific enhancement by m-AMSA or VP-16 of topo II DNA-cleavable complex formation was determined by the band depletion technique (Kawamura et al., 1996). The HL-60/S or HL-60/DOX0.05 cells were treated with 0.1 to 100 µM m-AMSA or VP-16 for 1 h at 37°C. Control and treated cells (2 × 10⁶) were lysed in 2× Laemmli buffer, samples were processed by SDS-PAGE and electroblotted onto nitrocellulose (Kawamura et al., 1996), and topo II was detected with a polyclonal rabbit antibody that specifically recognizes the 170- or 180-kDa isoform followed by ¹²⁵I-goat anti-rabbit IgG (Ganapathi et al., 1996a; Turley et al., 1997; Aoyama et al., 1998). Depletion of topo II due to enhanced cleavable complex formation in treated versus control cells was quantified by a phosphorImager.

Phosphorylation of topo II or topo II in HL-60/S and HL-60/DOX0.05 cells was determined by metabolic labeling with [³²P]orthophosphoric acid (Ganapathi et al., 1996a; Aoyama et al., 1998a,b). Nuclei were isolated from the labeled cells, lysed in RIPA buffer (Ganapathi et al., 1996; Aoyama et al., 1998a,b), and topo II or topo II protein in lysates containing equivalent number of nuclei, or similar amounts of protein, was immunoprecipitated with rabbit polyclonal antibodies that recognize the 170-kDa () or 180-kDa () protein (Ganapathi et al., 1996a; Turley et al., 1997; Aoyama et al., 1998b). Details of the technique for metabolic labeling and immunoprecipitation have been reported previously (Ganapathi et al., 1996; Aoyama et al., 1998a,b). Phosphorylated topo II or topo II protein levels were determined by densitometric scanning of autoradiograms or by the use of a phosphorImager. Phosphopeptide analysis of the immunoprecipitated 180-kDa () topo II was carried out by a modification (Ganapathi et al., 1996a; Aoyama et al., 1998a,b) of the methods described previously (Boyle et al., 1991; Wells et al., 1994). Briefly, the band corresponding to the 180-kDa () topo II protein was excised from the dried, unfixed gel, and eluted with 50 mM ammonium bicarbonate, 0.1% SDS, and 0.5% 2-mercaptoethanol overnight. The protein was precipitated with 100% trichloroacetic acid and oxidized with performic acid. Protein samples were digested overnight in N-tosyl-L-phenylalanine chloromethylketone-trypsin, and the radioactivity determined by Cerenkov counting. Aliquots of the phosphopeptides reconstituted in pH 1.9 electrophoresis buffer (containing equivalent dpm) were loaded onto thin layer cellulose plates and analyzed by electrophoresis with pH 1.9 buffer in the horizontal dimension, and phosphochromatography buffer in the vertical dimension (Ganapathi et al., 1996; Aoyama et al., 1998a,b).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

The effects of VP-16 and m-AMSA on the cytotoxic response were determined in HL-60/S and HL-60/DOX0.05 cells with a soft-agar colony assay (Fig. 1). The data indicate that although the HL-60/DOX0.05 cells are >20-fold resistant to VP-16, the magnitude of resistance to m-AMSA is only 2-fold. These results suggest that although VP-16 and m-AMSA target topo II, the expression of resistance is markedly different depending on their affinity for intercalation with DNA.

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Cytotoxic effects of VP-16 (A) and m-AMSA (B) in HL-60/S and HL-60/DOX0.05 cells. After treatment for 1 h, cytotoxicity was determined by a soft-agar colony assay (Ganapathi et al., 1996a). Data are mean values from at least three experiments. S.D. <15%. HL- 60/S and HL-60/DOX0.05 cells were plated at a density of 1.5 × 104 and 4 × 104 cells, respectively, per 35- × 10-mm Petri dish. Colony-forming efficiency of the HL-60/S and HL-60/DOX0.05 cells was 29 and 10%, respectively.

We have previously reported that resistance to VP-16 in the HL-60/DOX0.05 cells is not due to alterations in steady-state levels of topo II protein and is correlative with the hypophosphorylation of topo II (Ganapathi et al., 1996a; Aoyama et al., 1998a). Based on the availability of topo II isoform-specific antisera, and the ability of the technique of band depletion to ascertain drug-stabilized DNA cleavable complex formation with the topo II isoforms, the specific interaction of m-AMSA and VP-16 with topo II and  was determined. Representative gels and data that summarize the percentage of depletion of each topo II isoform for HL-60/S and HL-60/DOX0.05 cells are shown in Figs. 2 and 3. It is apparent that m-AMSA is 10- to 20-fold more potent than VP-16 in depleting an equivalent amount of topo II in either HL-60/S or HL-60/DOX0.05 cells. In the HL-60/S cells, 50% depletion of topo II protein occurred at 1 and 25 µM m-AMSA and VP-16, respectively. However, <50% depletion of topo II protein was observed in the HL-60/DOX0.05 cells over the range of drug concentrations tested. The depletion of topo II protein after treatment with the topo II poisons tested was as follows: 1) in HL-60/S cells, 50% depletion was observed at 0.5 µM m-AMSA and 10 µM VP-16; and 2) a 50% depletion in HL60/DOX0.05 cells was observed at 1.0 µM m-AMSA and 50 µM VP-16. Although m-AMSA was superior to VP-16 in depleting the topo II isoforms at lower drug concentrations, maximal depletion of topo II without an early plateau was observed with VP-16. As shown in Table 1, these results on depletion of the topo II isoforms are also consistent with the formation of an equivalent level of drug-stabilized topo II-DNA cleavable complex with 10-fold lower concentrations of m-AMSA in the HL-60/S cells.

View larger version (30K): [in this window] [in a new window]   Fig. 2.   Band depletion analysis of 170-kDa () topo II in HL-60/S and HL-60/DOX0.05 cells treated with m-AMSA or VP-16 for 1 h. A, samples of control and treated cells were processed for band depletion analysis as described in Materials and Methods. Data are from a representative gel. B, phosphorImager analysis of topo II depletion by m-AMSA or VP-16 in HL-60/S and HL-60/DOX0.05 cells. Data are mean values from at least three experiments.

View larger version (31K): [in this window] [in a new window]   Fig. 3.   Band depletion analysis of 180-kDa () topo II in HL-60/S and HL-60/DOX0.05 cells treated with m-AMSA or VP-16 for 1 h. A, samples of control and treated cells were processed for band depletion analysis as described in Materials and Methods. Data are from a representative gel. B, phosphorImager analysis of topo II depletion by m-AMSA or VP-16 in HL-60/S and HL-60/DOX0.05 cells. Data are mean values from at least three experiments.

Our previous data have demonstrated that topo II is hypophosphorylated without changes in the steady-state level of the enzyme in the HL-60/DOX0.05 cells (Ganapathi et al., 1996a). Based on the availability of antisera that efficiently immunoprecipitate the - or -isoforms of topo II, we sought to determine the potential for alterations in the protein levels and the phosphorylation of topo II in the resistant cells. As shown in Fig. 4A, the steady-state levels of topo II protein in the HL-60/S and HL-60/DOX0.05 cells were comparable. However, in contrast to the hypophosphorylation of topo II, the topo II protein is hyperphosphorylated in the HL-60/DOX0.05 cells (Fig. 4B). The hyperphosphorylation is not a consequence of the differential cellular distribution of topo II because the data in Fig. 4C demonstrate that hyperphosphorylated topo II is detectable both in whole cells and isolated nuclei. Because site-specific phosphorylation of proteins can have important regulatory functions, phosphopeptide mapping of complete tryptic digests of topo II in the HL-60/S and HL-60/DOX0.05 cells was carried out. The data in Fig. 5 demonstrate that the hyperphosphorylation of topo II in the HL60/DOX0.05 cells is site specific. Notably, eight different sites in the topo II protein are hyperphosphorylated in the HL-60/DOX0.05 cells versus HL-60/S cells.

View larger version (40K): [in this window] [in a new window]   Fig. 4.   Analysis of topo II levels by immunoblotting and immunoprecipitation in HL-60/S and HL-60/DOX0.05 cells. A, immunoblot analysis of topo II in HL-60/S (lane 1) and HL-60/DOX0.05 (lane 2) cells; B, phosphorylated topo II and topo II in HL-60/S (lanes 1 and 3) and HL-60/DOX0.05 (lanes 2 and 4); C, phosphorylated topo II from HL-60/S (lanes 1 and 3) and HL-60/DOX0.05 (lanes 2 and 4) cells or nuclei.

View larger version (80K): [in this window] [in a new window]   Fig. 5.   Representative two-dimensional tryptic phosphopeptide maps of topo II protein from HL-60/S (A) and HL-60/DOX0.05 (B) cells. The sites that are differentially phosphorylated between the HL-60/S and HL-60/DOX0.05 cells are identified by arrows.

Because the phosphorylation but not protein levels of the - and -isoforms of topo II were different in the HL-60/S and HL-60/DOX0.05 cells, the potential for such changes based on cell cycle phase were investigated. The levels of topo II and topo II protein in HL-60/S or HL-60/DOX0.05 cells enriched in the G₁, S, and G₂ + M phases of the cell cycle by centrifugal elutriation are shown in Fig. 6. In a typical experiment, the distribution of cells in the G_1, S, and G₂ + M phase was 43, 44, and 13%, respectively, in asynchronous exponential cultures of HL-60/S or HL-60/DOX0.05 cells. After elutriation the enrichment of cells in the G_1, S, and G₂ + M was 95, 82, and 68%, respectively. Because the topo II protein levels were the lowest in the G₁ phase, the increased expression in the S and G₂ + M phases were expressed relative to the G₁ phase. In both the HL-60/S and HL-60/DOX0.05 cells, the levels of topo II protein in the S and G₂ + M phases were 2- and 4-fold higher, respectively. However, a similar cell cycle phase-dependent increase in the level of topo II protein was not apparent, and represented an ~2-fold increase (compared with G₁) in the S and G₂ + M phases.

View larger version (37K): [in this window] [in a new window]   Fig. 6.   Analysis of topo II and II protein levels by immunoblotting in elutriated G1, S, and G2 + M populations in HL-60/S and HL-60/DOX0.05 cells. Data are mean values (±S.D.) from at least three experiments.

The cell cycle phase-dependent phosphorylation of topo II isoforms in HL-60/S and HL-60/DOX0.05 cells is outlined in Fig. 7. In HL-60/S cells, an increase (2-fold) compared with the G₁ phase in the phosphorylation of topo II but not topo II was observed in the S and G₂ + M phases of the cell cycle. In contrast, although an increase of a lower magnitude was observed with topo II in the S and G₂ + M phases, with the HL-60/DOX0.05 cells, the phosphorylation of topo II in the G₁, S, and G₂ + M was similar or greater than that of topo II.

View larger version (44K): [in this window] [in a new window]   Fig. 7.   Determination of phosphorylated topo II and II levels by immunoprecipitation in elutriated G1, S, and G2 + M populations in HL-60/S and HL-60/DOX0.05 cells. Data are mean values (±S.D.) from at least two experiments. Based on statistical analysis of phosphorImager data, the following differences between the phosphorylation of topo II and topo II were found to be significantly different. HL-60/DOX0.05 cells: G1 phase, topo II > topo II, p = .017 and HL-60/S cells: S phase, topo II > topo II, p = .045.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Topo II in mammalian cells is present as the - and -isoforms (Woessner et al., 1990, 1991; Turley et al., 1997). The two isoforms perform similar catalytic functions related to unknotting and decatenation of DNA, and an essential role for the -isoform in cell proliferation has been demonstrated (Watt and Hickson, 1994; Wang, 1996). Although a precise functional role for the topo II isoform is yet to be demonstrated, its altered phosphorylation during mitosis has been demonstrated (Burden and Sullivan, 1994; Kimura et al., 1994). The evidence that proliferation is not compromised in cells without detectable levels of the -isoform (Herzog et al., 1998) suggests that it is not essential for cell replication or chromosome segregation. Although the relative abundance of the topo II isoforms varies by cell type (Turley et al., 1997), the contribution of the -isoform as a determinant of antitumor drug sensitivity is unclear. Based on quantification of protein levels of topo II isoforms in cells selected for resistance to drugs that intercalate with DNA, it appears that higher levels of resistance are correlative with alterations in topo II (Harker et al., 1991; Herzog et al., 1998; Perrin et al., 1998). Although these studies have to a large extent been correlative, transfection studies (Dereuddre et al., 1997) have provided new evidence for a functional role. The data on the markedly higher depletion of topo II by m-AMSA versus VP-16 also support the differences observed in levels of resistance to these drugs in the HL-60 cells deficient in the topo II protein (Herzog et al., 1998). The depletion of topo II versus topo II at cytotoxic concentrations of m-AMSA in the HL-60/S and HL-60/DOX0.05 cells suggests that this differential interaction with the topo II isoforms is responsible for the greater cytotoxic potency of m-AMSA in vitro. The data also demonstrate that steady-state levels of topo II or  protein are unaltered in the HL-60/DOX0.05 cells. However, unlike the -isoform, the -isoform is hyperphosphorylated in the HL-60/DOX0.05 versus HL-60/S cells. It is not clear at the present time whether this change in the phosphorylation state of the -isoform is a compensatory response to changes in the -isoform, but the absence of marked differences in the cell cycle kinetics of the HL-60/S and HL60/DOX0.05 cells supports such a role. Similar to the asynchronous exponential cultures, protein levels of the topo II isoforms in elutriated G₁, S, and G₂ + M populations also were similar between the HL-60/S and HL-60/DOX0.05 cells. It is well recognized that the phosphorylation of topo II in the S and G₂ + M phases are critical for cell replication (Watt and Hickson, 1994; Wang, 1996). Although, topo II is hypophosphorylated in the HL-60/DOX0.05 cells, no major differences in cell cycle kinetics compared with the HL-60/S cells are apparent. However, a compensatory role of topo II is apparent in the similarities of cell cycle phase-dependent phosphorylation of topo II in the HL-60/DOX0.05 cells and topo II in the HL-60/S. The site-specific hyperphosphorylation of topo II in the HL-60/DOX0.05 cells is unique, given the site-specific hypophosphorylation of the -isoform in these cells. It does not appear that the same site may be involved, due to the distinctly different phosphopeptide maps of the - and -isoforms in complete tryptic digests (Aoyama et al., 1998a,b). There also appears to be some consistency in the site-specific hyperphosphorylation of topo II in the HL-60/DOX0.05 cells and in HL-60/S cells induced to differentiate with all trans-retinoic acid (Aoyama et al., 1998b).

In summary, the results from this study suggest that the differential sensitivity of HL-60 cells to m-AMSA and VP-16 is dependent on the differential interaction of these agents with the topo II isoforms. m-AMSA appears to be more effective in interacting with the -isoform at concentrations, which are at least 10-fold lower than similar concentrations of VP-16 on a molar basis and relevant to induction of a cytotoxic response. The cell cycle phase-specific, as well as site-specific, hyperphosphorylation of topo II in the HL-60/DOX0.05 cells may compensate for the hypophosphorylation of the -isoform. Furthermore, the hyperphosphorylation of topo II may be functionally involved in the enhanced interaction and reduced level of cross-resistance to m-AMSA. Studies are in progress to determine the functional role of site-specific phosphorylation of topo II in drug-stabilized DNA-cleavable complex formation, and in the long term, an understanding of the functional role of the -isoform could provide new information in exploiting its role as a therapeutic target.