Human Small Cell Lung Cancer NYH Cells Selected for Resistance to the Bisdioxopiperazine Topoisomerase II Catalytic Inhibitor ICRF-187 Demonstrate a Functional R162Q Mutation in the Walker A Consensus ATP Binding Domain of the α Isoform¹

DNA topoisomerase II is an essential nuclear enzyme that carries out topological changes in DNA and plays important roles in DNA replication, transcription, and chromosome condensation (1, 2). The enzyme exists in two isoforms: namely, a M_r 170,000 form termed topoisomerase IIα and a M_r 180,000 form termed topoisomerase IIβ. The homodimeric enzyme has a complex catalytic cycle that includes an ATP-driven conformational change in its NH₂-terminal clamp, leading to capture of the transported double-stranded DNA, cleavage of the gate DNA strand, passage of the transported DNA strand, and subsequent religation of the gate strand, followed by ATP hydrolysis and opening of the clamp, thus making it ready for another cycle (1, 2). The mechanism of coupling the ATP hydrolysis with DNA strand passage is not fully understood and is not tight in eukaryotic topoisomerases because some ATP hydrolysis occurs without a strand passage event (3). Furthermore, there appear to be species differences because human topoisomerase II is more efficient in coupling ATP hydrolysis to strand passage than is the yeast enzyme (4).

Topoisomerase II is one of the most important clinical targets for anticancer drugs (5). Thus, the anthracyclines doxorubicin (Adriamycin) and daunorubicin, the epipodophyllotoxins etoposide and teniposide, and the aminoacridine m-AMSA³ all act at the stage in the catalytic cycle in which the gate DNA strand has been cleaved. These drugs all stabilize an enzyme-DNA cleavable complex that leads to irreversible DNA breaks and, ultimately, cell death. Cleavable complex stabilizing drugs are termed poisons because they convert the essential enzyme into one that damages the cell (6). However, other drugs exist that act on the topoisomerase II catalytic cycle without stabilizing cleavable complexes, i.e., they do not lead to DNA breaks. This second class of drugs has been identified as topoisomerase II catalytic inhibitors. One of the most studied of these have been the bisdioxopiperazine compounds, which were first described as acting on topoisomerase II in 1991 (7, 8). These compounds have at least two mechanisms of action on eukaryote topoisomerase II: namely, locking the enzyme in its closed clamp form when ATP is present and inhibiting the enzyme’s ATPase activity (9, 10). We have recently described a CHO cell line that is resistant to the bisdioxopiperazine ICRF-159 which carries a functional Y49F mutation (human Y50F) in the NH₂-terminal clamp part of the enzyme (11). To further investigate how bisdioxopiperazines exert their catalytic inhibition, we exposed human small cell lung cancer NYH cells to increasing concentrations of ICRF-187 until they were 5.5-fold resistant to the drug in a clonogenic assay (NYH/187). Here, we describe a functional mutation discovered in NYH/187 cells, R162Q, which is the first drug-induced mutation in the Walker A consensus ATP binding site in eukaryote topoisomerase IIα. This mutation causes the enzyme to be nearly catalytically inactive at <0.25 mm ATP and confers resistance to bisdioxopiperazines but not to the clinically important topoisomerase II poisons etoposide and m-AMSA. The mutation also causes collateral hypersensitivity to merbarone, another catalytic topoisomerase II inhibitor. Characterization of the mutant enzyme illuminates several functional aspects of the ATP binding site in topoisomerase II and the action of topoisomerase II inhibitors.

ICRF-187 (Cardioxane; Chiron, Amsterdam, the Netherlands), vincristine (Lilly, Copenhagen, Denmark), aclarubicin (Lundbeck, Taastrap, Denmark), novobiocin (Sigma-Aldrich, Copenhagen, Denmark), daunorubicin (Rhône Poulenc, Holte, Denmark), and doxorubicin (Adriamycin; Pharmacia Upjohn, Copenhagen, Denmark) were dissolved in sterile water immediately prior to use. Etoposide and cisplatin (Bristol-Myers Squibb, Lyngby, Denmark) were in solution for infusion. 1-β-d-Arabinofuranosylcytosine (Pharmacia Upjohn) was dissolved in benzyl alcohol, camptothecin (Sigma) was dissolved in DMSO, and carmustine [1,3-bis(2-chloroethyl)-1-nitrosourea; Bristol-Myers Squibb] was dissolved in 10% (v/v) ethanol in sterile water. Merbarone and fostriecin were gifts from the Drug Synthesis and Chemistry Branch, Developmental Therapeutics Program, Division of Cancer Treatment, National Cancer Institute, NIH (Bethesda, MD), and were dissolved in DMSO and sterile water, respectively. The drugs were diluted with tissue culture medium to 300× final concentrations, portioned into single-use aliquots, frozen on ethanol-dry ice, and stored at −80°C.

Human small cell lung cancer NYH cells were selected in increasing concentrations of ICRF-187 for 45 passages, to a final concentration of 1.0 mm drug, and were termed NYH/187.

A 3-week clonogenic assay using soft agar on a sheep RBC feeder layer was used with continuous drug incubation, and IC₅₀s were computed by linear regression analysis (12).

Alkaline filter elution assay for measuring DNA SSBs was performed as described previously (13), with minor modifications (11).

Western blots and band depletion assays were performed as described previously (11) using either a monoclonal antibody for the α isoform (Cambridge Research Biochemicals, Cheshire, United Kingdom) or a polyclonal antibody for the β isoform (Bio-Trend, Cologne, Germany). Visualization was performed using the Amersham (Amersham, United Kingdom) chemiluminescence kit according to the manufacturer’s instructions.

cDNA derived from 10 ng of mRNA was mixed with 25 pmol of primer pairs, 2.5 units of Thermoprime plus DNA polymerase (Advanced Biotechnologies, Surrey, United Kingdom), and 0.2 mm dNTP in PCR buffer IV [20 mm (NH₄)₂SO₄, 75 mm Tris-HCl, 0.01% Tween 20, and 1.5 mm MgCl₂; Advanced Biotechnologies]. PCR was carried out in 100 μl using the GeneAmp PCR system 2400 (Perkin-Elmer, Foster City, CA), with an initial denaturation for 5 min at 94°C, followed by 40 cycles of denaturation for 30 s at 94°C, annealing for 30 s at 58°C, and extension for 45 s at 72°C. To verify that PCR products were of the correct length, 10 μl of PCR products were electrophoresed in TAE buffer (40 mm Tris-acetate-1 mm EDTA) in a 4% agarose-0.5% ethidium bromide gel. When only one band was present, the PCR product was purified with QIAquick PCR purification kit (Qiagen, Hilden, Germany), and the concentration and purity were determined by spectrophotometry. Genomic DNA was purified using the TRIzol reagent from Life Technologies, Inc. (Gaithersburg, MD). Sequencing of the PCR products was performed using the ABI Prism Dye Terminator Cycle Sequencing kit (Perkin-Elmer) using AmpliTaq DNA polymerase (Perkin-Elmer). Sequencing was performed from both 5′ and 3′ ends for two or three times each on two different mRNA extractions using the primers described previously (14).

To assess whether the R162Q mutation found in NYH/ICRF topoisomerase IIα contributes to the acquired bisdioxopiperazine resistance, the corresponding point mutation was introduced into the episomal expression vector for human topoisomerase IIα pMJ1 by oligonucleotide-directed mutagenesis, as described previously (11), using a QuikChange site-directed mutagenesis kit from Stratagene (La Jolla, CA) with the two mutagenic primers R162Q-SN and R162Q-ASN (Table 1). The existence of the R162Q mutation in two independent preparations of the resulting plasmid pMJ1-R162Q was verified by automated dye termination cycle sequencing using the primers R162Q-SEQ-F and R162Q-SEQ-R (Table 1). The pMJ1 and pMJ1-R162Q plasmids were introduced into the temperature sensitive hyperpermeable RAD52-deficient yeast strain JN394t2-4 (MATa ura-52 leu2 trp1 his7 ade1-2 ISE2 rad52::LEU2 top2-4; Ref. 15) using a modified lithium acetate method with single-stranded salmon sperm DNA as carrier. Establishment and propagation of transformed yeast clones, test for complementation of the temperature-sensitive t2-4 TOP2 allele by expression of wt and mutant human topoisomerase IIα at the nonpermissive temperature 34°C, and determination of drug sensitivity in clonogenic assay was performed as described previously (11). All experiments were performed at least twice. Representative data are shown.

To overexpress and purify the mutant enzyme, we used a modification of a protocol described previously (16). The R162Q mutation was introduced in human topoisomerase IIα cDNA in the expression vector pGALhTOP2, in which the expression of human topoisomerase IIα is controlled by a strong galactose-inducible GAL promoter, using the R162Q-SN and R162Q-ASN mutagenic primers, resulting in pGALhTOP2-R162Q. The presence of the mutation was confirmed by sequencing. Both pGALhTOP2 and pGALhTOP2-R162Q were transformed to the protease-deficient yeast strain JelΔTOP1 (trp1, leu2, ura-52, pbr1-1122, pep4-3, Δhis3::PGAL10-GAL4, TOP1::LEU2). Transformed cells were inoculated in 70 ml of SC-URA medium containing 3% glycerol, 2% lactic acid, and 2% glucose and were grown overnight at 30°C to an A₆₀₀ of 1.5–3.0. The cultures were then diluted 1:100 into SC-URA medium containing no glucose supplemented with 3% glycerol and 2% lactic acid and were grown to an A₆₀₀ of ∼0.6–0.8. At this point, 2% galactose was added to induce the expression of human topoisomerase lIα in the yeast cells, and the cells were grown for an additional 12–18 h, harvested by centrifugation at 3000 × g for 15 min at 4°C, and washed with 5 mm Tris-HCl (pH 8.0)-10 mm EDTA. Finally, the cells were quick-frozen in liquid nitrogen and stored at −80° C.

All steps were carried out at 4°C. Twenty to 30 g of frozen cells obtained from a 6-liter culture were used for the purification of mutant and wt human topoisomerase IIα. The cells were disrupted in buffer 1 [50 mm Tris-HCl (pH 7.7), 1 mm EDTA, 1 mm EGTA, 10% glycerol, 1 mm PMSF, 5 mm DTT, 0.5 μg/ml leupeptin, and 1 μg/ml pepstatin A) and 25 mm KCl using acid-washed glass beads (Sigma) in a Bead Beater (Biospec), and cell debris was removed by centrifugation for 15 min at 27,000 × g. Lysates were diluted to obtain a protein concentration of 5 mg/ml in buffer 1 plus 25 mm KCl. Next, nucleic acids and protein-nucleic acid complexes were precipitated by slowly adding 10% polymin P (Sigma) to a concentration of 0.2%, followed by stirring for 30 min and centrifugation at 7000 × g for 10 min. The pellets were washed with 90 ml of buffer 1 plus 150 mm KCl by stirring for 25 min, followed by centrifugation as described above, and resuspended in 90 ml of buffer 1 plus 750 mm KCl by stirring for 25 min, followed by centrifugation at 7000 × g for 10 min. The supernatant was set aside, and the pellets were resuspended again with buffer 1 plus 750 mm KCl followed by stirring for 25 min and centrifugation at 7000 × g for 10 min. Supernatants were combined and brought to 35% saturation with the addition of (NH₄)₂SO₄ and stirring for 30 min. After centrifugation for 25 min at 22,000 × g, the supernatant was brought to 65% saturation using (NH₄)₂SO₄, followed by stirring for 30 min and centrifugation at 22,000 × g for 25 min. Next, the pellets were resuspended in buffer 1 plus 15 mm sodium phosphate (pH 7.7) and 1 mm 2-mercaptoethanol, using ∼2.5 ml per g of cells. The conductivity of the solution was adjusted to a little less than the conductivity of H-KP 150 [15 mm sodium phosphate (pH 7.7), 10% glycerol, 150 mm potassium phosphate (pH 7.7), 1 m PMSF, 0.5 mm DTT, and 0 mm Na₂S₂O₅] by additional dilution when this was required. The sample was loaded on a hydroxyapatite (Bio-Rad) column (height = 25 cm, 10 diameter = 1.5 cm), which was then eluted with a linear salt gradient of 125 ml of H-KP 150 and 125 ml of H-KP 600 [same composition as H-PK 150, but with 600 mm potassium phosphate (pH 7.7)] at a flow rate of 50 ml/h. The fraction collector was set at 80 drops/fraction. The fractions were assessed on a 7.5% SDS gel, and the fractions containing human topoisomerase IIα were pooled. The fraction pool was diluted with buffer 1 so that its conductivity was slightly less than P200 [15 mm sodium phosphate (pH 7.7), 1 mm EDTA, 1 mm EGTA, 10% glycerol, 200 mm KCl, 0.5 mm DTT, and 0.33 mm PMSF] and was then loaded on to a phosphor cellulose (Whatman) column (height = 5 cm, diameter = 1 cm), which was eluted with P750 [50 mm Tris (pH 7.7), 5 mm DTT, 1 mm EDTA, 750 mm KCl, and 30% glycerol] at a flow rate of 50 ml/h at 40 drops/fraction. The fractions were assessed with a BSA standard curve to find the fraction containing the human topoisomerase IIα, aliquoted, and stored in liquid nitrogen. The purity of human topoisomerase IIα was checked by gel electrophoresis and Coomassie blue staining.

Topoisomerase II catalytic activity was measured by kDNA decatenation. ³H-labeled or unlabeled kDNA was isolated from Crithidia fasciculata (American Type Culture Collection, Manassas, VA) as described previously (17). Briefly, purified wt or R162Q mutant topoisomerase IIα was incubated with 0.2 μg of kDNA for 15 min at 37°C in a final volume of 20 μl in a buffer containing [50 mm Tris-Cl (pH 8), 120 KCl, 10 mm MgCl₂, 1.0 mm ATP, 0.5 mm DTT, and 30 μg/ml BSA]. After addition of stop buffer/loading dye mix (5% sarkosyl, 0.0025% bromphenol blue, and 25% glycerol), samples were loaded on 1% agarose-0.5% ethidium bromide gels and run in Tris-boric acid-disodium EDTA buffer at 100 V for ∼50 min. Loading wells were cut out, and scintillation was counted or, alternatively, bands were measured by densitometry.

The following were mixed together: 100 ng of topoisomerase II, either wt or mutant; topoisomerase II assay buffer [50 mm Tris (pH 8.0), 8 mm MgCl₂, 1 mm EDTA, 7 mm β-mercaptoethanol, 150 mm KCl, and 100 mg/ml acetylated BSA]; ATP at 0, 5, 10, 15, 20, or 30 μm; and 20 ng of supercoiled pUC18 substrate. The reaction was carried out for 30 min at 37°C and stopped with 10 mm EDTA (final concentration). The DNA samples were analyzed by electrophoresis on 1% agarose gels.

pUC18 plasmid was cleaved with EcoRl and labeled with [α-³²P]dATP. Three hundred fifty ng of DNA substrate were mixed with equal amount of purified wt or R162Q topoisomerase IIα, Top2 reaction buffer (20 mm Tris-HCl, 10 mm MgCl₂, 1 mm ATP, 1 mm EDTA, 1 mm DTT, 150 mm KCl, and 30 μg/ml acetylated BSA), and the drug to be tested (18) and was incubated at 37°C for 10 min. The reaction was stopped by adding Stop solution [1.25% SDS, 5 mm EDTA (pH 8.0), and 0.4 mg/ml salmon sperm DNA] and 250 μl of 325 mm KCl. After 10 min on ice and centrifugation at 5700 × g for 10 min at 4°C. The supernatant was aspirated, and the pellet was resuspended in 1 ml of wash solution [10 mm Tris-HCl (pH 8.0), 100 mm KCl, 1 mm EDTA, and 1 mm salmon sperm]. Samples were heated for 10 min at 65°C, followed by 10 min on ice and centrifugation as described above. The washing procedure was repeated twice, and finally, the pellet was resuspended in 400 μl of water. One hundred μl of sample and 4 ml of scintillation buffer were mixed, and the amount of covalent complexes was determined by scintillation counting.

NYH and the ICRF-187 resistant subline NYH/ICRF had doubling times of 20 and 20.5 h, respectively, and were morphologically similar, although NYH/187 cells were slightly more flattened. Using reverse transcriptase-PCR, we detected no overexpression of MDR1, MRP, or LRP (multidrug resistance 1, multidrug resistance protein, and lung resistance-related protein genes, respectively) mRNA (data not shown). NYH/187 cells were 5.5-fold resistant to ICRF-187 (Fig. 1 and Table 2) and slightly hypersensitive to the topoisomerase II poisons doxorubicin and etoposide, which correlates with their slight (∼20%) increase in topoisomerase IIα level (see below) and the increased induction of DNA breaks by R162Q enzyme (see Fig. 9). No change in sensitivity to the topoisomerase I-directed drug camptothecin or the tubulin-destabilizing agent vincristine was observed (Table 2). Only minor differences in sensitivity were observed to the catalytic inhibitors aclarubicin, novobiocin, and fostriecin. However, a collateral sensitivity to the catalytic inhibitor merbarone was seen in NYH/187 cells. When NYH and NYH/187 cells were coincubated with 20 μm etoposide and increasing concentrations of ICRF-187, protection of cytotoxicity was observed in both cell lines (Fig. 1). This is in contrast to a similar experiment performed in CHO and CHO/159–1 cells, in which no protection was conveyed by ICRF-187 in the resistant line (Fig. 1 and Ref. (11)).

NYH/187 cells had an ∼25% increase in topoisomerase IIα level, with no change in the amount of the β isoform compared to NYH cells (data not shown). In band depletion experiments, NYH cells showed a dose-dependent decrease caused by ICRF-187 of both topoisomerase IIα and IIβ, whereas in NYH/187 cells, this was only seen in the β isoform (Fig. 2). Thus, the α isoform in NYH/187 is only changed with respect to its ability to be trapped by ICRF-187 onto its DNA substrate. Conversely, etoposide was able to band deplete both isoforms in both NYH and NYH/187 cells (Fig. 2), attesting to the specificity of the change toward ICRF-187 in NYH/187 cells.

Etoposide-induced DNA SSBs were slightly increased in NYH/187 compared to NYH cells, which is consistent with both the increased level of topoisomerase IIα in the former as well as the higher amount of DNA breaks induced by etoposide by purified R162Q enzyme (see below). ICRF-187 was only able to prevent etoposide-induced SSBs in NYH and not in NYH/187 cells (Fig. 3). In contrast, the intercalating catalytic inhibitor aclarubicin (19) was able to inhibit etoposide induced SSBs in both NYH and NYH/187 cells (Fig. 3). Again, this is consistent with the specificity of the resistance toward ICRF-187 in NYH/187 cells.

Sequencing of the entire topoisomerase IIα cDNA in NYH/187 cells revealed a homozygous G→A point mutation at nucleotide 485, leading to a R162Q conversion in the Walker A consensus ATP binding site (residues 161–166 in the α isoform, Table 3). Sequencing of the corresponding region in NYH cDNA showed the wt sequence only. Sequencing of genomic DNA from NYH/187 cells covering the residues of the Walker A site confirmed that the G→A mutation is homozygous because the wt sequence was not detected.

The episomal expression vector for human topoisomerase IIα in yeast has been described elsewhere (15). Briefly, this vector includes the entire coding region of human topoisomerase IIα under the control of the yeast topoisomerase l promoter, as well as an URA3 marker, a yeast origin of replication and a yeast centromere for the introduction and maintenance of the plasmid in yeast. Both pMJ1- and pMJ1-R162Q-transformed cells were able to grow at 34°C, whereas cells transformed with the vector plasmid yCP50 used for the construction of pMJ1 were only able to grow at 25°C, demonstrating complementation of the conditional TOP2 allele at the nonpermissive temperature by the recombinant human mutant enzyme (data not shown). The JN394t2-4 strain also comprises a dominant drug permeability mutation ISE2, which increases the sensitivity toward a number of topoisomerase II-directed drugs, including the bisdioxopiperazines (20). At 34°C, JN394t2-4 cells expressing the R162Q mutant enzyme displayed resistance to both ICRF-187 and ICRF-193 compared to cells expressing the wt enzyme (Fig. 4). In contrast, cells expressing the mutant enzyme were slightly hypersensitive toward the topoisomerase II poison m-AMSA (Fig. 4), whereas no change was observed toward etoposide (Fig. 4). This result demonstrates that the R162Q mutation is specific for bisdioxopiperazines and confirms the functional significance of this mutation in the acquired bisdioxopiperazine resistance.

Using the kDNA decatenation assay, the recombinant R162Q enzyme showed 20–25% decreased activity compared to wt (Fig. 5 A) at 1 mm ATP. To address the question of whether there was a difference in the ATP requirements, we performed decatenation experiments with equipotent recombinant enzyme levels and decreasing ATP concentrations. At <0.25 mm ATP, the R162Q enzyme revealed a marked decrease in decatenation activity compared to wt (Fig. 5 C). A similar result was reached when using the DNA relaxation assay (Fig. 6), where 15 μm ATP on wt enzyme equaled 30 μm on the R162Q mutant.

When wt or R162Q enzyme was coincubated with ICRF-187 at 1 mm ATP in a kDNA decatenation assay, no change in sensitivity was observed (Fig. 7 A). However, at subsaturation ATP levels of 0.125 mm (Fig. 7 B) and 0.0315 mm (Fig. 7 C), the sensitivity of R162Q enzyme to ICRF-187 enzyme compared to wt was clearly reduced, the difference being most marked at the lowest ATP level. A further observation is that the sensitivity of wt enzyme to ICRF-187 is also susceptible to decreases in ATP levels, e.g., the lack of inhibition at 12.5 μ m at 0.0315 mm ATP compared to 0.125 mm (Fig. 7 B and C).

We observed hypersensitivity to merbarone in R162Q (Fig. 8 A), which is consistent with results from the clonogenic assay in whole cells (Table 2). However, no change in the catalytic sensitivity to drugs in R162Q compared to wt enzyme was detected with m-AMSA (Fig. 8 B), etoposide, aclarubicin, fostriecin, or novobiocin (data not shown).

When comparing the ability of wt and R162Q enzyme to form cleavable complexes in the presence of the poisons etoposide and m-AMSA, we observed that both compounds appeared to be more efficacious on R162Q enzyme (Fig. 9). This is most noticeable with m-AMSA with its characteristic bell-shaped dose-response curve, which is typical for DNA-intercalating agents.

Since the discovery in 1991 that bisdioxopiperazine anticancer agents act on topoisomerase II (7, 8), they have proved to be highly specific catalytic inhibitors of this enzyme. This drug class has, therefore, received much attention, with regard to both its direct effect on the enzyme (9, 10, 21, 22) and its ability to modulate the effect of topoisomerase II poisons both in vitro (23, 24) and in vivo (25), the latter leading to the possibility of using bisdioxopiperazine combinations with topoisomerase II poisons to target tumors in pharmacological sanctuaries (26).

One way of determining the biochemical and molecular drug-target enzyme interactions is to select and study drug-resistant mutant protein. Recently, we described a bisdioxopiperazine-resistant CHO/159-1 cell line with a Y49F mutation in its topoisomerase IIα cDNA, which conferred high level resistance to ICRF-193 in human topoisomerase IIα (Y50F) transformed to the same temperature-conditional yeast system used in this study (11). Furthermore, an adjacent T48I mutation has recently been described in another bisdioxopiperazine-resistant CHO cell line (27), strongly indicating that this NH₂-terminal region is important in the interaction between topoisomerase IIα and this drug class. This would agree with bisdioxopiperazines acting on the NH₂-terminal ATPase clamp in the two-gate model of topoisomerase II proposed previously (28). However, ICRF-193 also inhibits NH₂- and COOH-terminal truncated Schizosaccharomyces pombe topoisomerase II containing residues 75–1219 (21). Furthermore, recent evidence suggests that bisdioxopiperazines can lock Drosophila topoisomerase II on circular DNA using only the core domain truncated at NH₂-terminal residue 406 (22). One possible explanation for this is that bisdioxopiperazines are able to lock both the NH₂-terminal ATPase clamp as well as closing the B′-B′ interface at the NH₂ terminus of the core domain. This is supported by the bisdioxopiperazine-induced “lock” being stable at 4 mm salt when drug was combined with ATP and full-length enzyme, whereas it was at least 50% less stable when only the core domain was used (22). Further studies are obviously needed to determine the functional biochemistry of the bisdioxopiperazine-induced enzyme clamp lock.

The phenotypes of the functional bisdioxopiperazine Y50F mutation in (11) and the present R162Q differ in at least two aspects: namely, the lack of protection by ICRF-187 to etoposide cytotoxicity only occurs in CHO/159-1 and not in NYH/187 cells (Fig. 1), and that the catalytic ability of R162Q enzyme is inhibited equally to wt at 1 mm ATP, which was not the case for Y50F (11). This indicates that these two mutations cause bisdioxopiperazine resistance by different mechanisms, and our working hypothesis as to the latter finding is that Y50F is involved in drug binding, whereas R162Q has an indirect effect via the enzyme’s interaction with ATP. The difference in protection of etoposide in clonogenic assay between Y50F and R162Q is, at present, unexplained and could be due to differences in cell cycle progression between CHO and NYH cells.

ATP is required for the enzyme to go back and forth between the open and closed state and only in the presence of ATP can bisdioxopiperazines access the closed form and prevent its reopening (9), although this latter statement has recently been modified as mentioned above (22). It was, thus, highly intriguing that the only point mutation in NYH/187 topoisomerase IIα was R162Q which is in the Walker A consensus ATP binding site of 161GRNGYG166 (Ref. (29); Table 3). Site-directed mutations in the corresponding Walker A site in yeast TOP2 have been performed for the middle G144 (corresponding to human G164), in which all three mutations G144I, G144P, and G144V resulted in an inactive enzyme (3), confirming the critical functional role of this residue. However, changing the upstream charged K137 or D132-134 for alanine only resulted in a slightly reduced catalytic ability of the mutant enzyme (Ref. 29; Table 3). In a similar fashion, loss of charge in the human R162Q only resulted in a minor loss of catalytic function at saturation ATP, although this residue is highly conserved in eukaryote topoisomerase II (Table 3). The precise role of R162 in human topoisomerase IIα is undetermined. However, the analogous residue in prokaryotic gyrase, L115, has been proposed to hydrogen bond with the γ-phosphate of ATP (30, 31), and it would appear a reasonable assumption that R162 has a similar function. DNA topoisomerase II has an intricate catalytic cycle in which ATP binding and hydrolysis play important roles (1, 2). Recent evidence further indicates that the enzyme has an unexpectedly complex mechanism of ATP hydrolysis in that the two bound ATP molecules are hydrolyzed sequentially thus creating a transient asymmetry in the reaction cycle (32, 33). ATP hydrolysis and DNA transport are coupled with a degree of linking that is dependent on the reaction conditions. In the absence of DNA, ATP binds slowly to topoisomerase II, which is an advantage because ATP-bound topoisomerase II has a conformation in which it cannot bind DNA. Topoisomerase rapidly hydrolyzes at least one ATP (3), and the concentration of topoisomerase II saturated with ATP changes dramatically in the 0.1–0.3 mm ATP concentration range. Thus, ATP binding is partially rate-determining at 0.1 mm ATP, whereas at 0.3 mm, ATP binding is much faster than the rate-limiting mechanism (32, 33). In accordance with these biochemical data, we found that R162Q had a dramatically decreased enzymatic activity at an ATP concentration of <0.25 mm, indicating that the mutant enzyme binds poorly to ATP, whereas at saturating concentrations of 1 mm ATP, the catalytic activity was only reduced by 20–25% (Fig. 5). The catalytic ability of R162Q is not resistant to ICRF-187 at saturation ATP levels of 1 mm (Fig. 7), which, as mentioned previously, is in contrast to the Y50F functional mutation (11). This indicates that resistance to bisdioxopiperazines caused by R162Q is due to a shift in the enzyme’s catalytic equilibrium to an early pre-ATP binding stage, in which the enzyme is in an open-clamp form (28) and is, thus, inaccessible to bisdioxopiperazines (9, 10). To further investigate this hypothesis, we performed drug inhibition assays at low ATP concentrations of 0.125 and 0.0315 mm and observed both that wt enzyme became less sensitive to ICRF-187 the lower the ATP level was and that R162Q was resistant to ICRF-187 compared to wt at both the low ATP concentrations tested (Fig. 7). With regard to the other catalytic inhibitors only merbarone showed a difference with hypersensitivity to R162Q both in clonogenic assay as well as in vitro on recombinant enzyme (Table 2 and Fig. 8). Merbarone has recently been suggested to act on topoisomerase II at the catalytic step just prior to ATP binding (34), which would agree with its hypersensitivity in R162Q being caused by the above-suggested shift to the pre-ATP-bound state.

Topoisomerase IIα has both a Walker A and a Walker B consensus ATP binding site. It is remarkable that a R450Q mutation at a similar position of first positively charged residue in the Walker B consensus site was found in a cell line selected for resistance to the topoisomerase II poison teniposide and cross-resistant to other poisons such as etoposide, doxorubicin, and m-AMSA (35). Although only studied in nuclear extracts, the R450Q mutation also appeared to increase the enzyme’s ATP requirement (36). Furthermore, the R450Q mutation in human topoisomerase IIα has also been demonstrated to confer marked resistance to both etoposide and m-AMSA in the same temperature permissive JN394t2-4 yeast system as used in this study (15), which is, thus, directly comparable to the lack of cross-resistance to etoposide and even hypersensitivity to m-AMSA in JN394t2-4 yeast cells expressing the R162Q (Fig. 4). This points to a hitherto unsuspected difference in the importance of the Walker A and B ATP binding sites in the sensitivity of the enzyme to various classes of anticancer drugs. Binding of ATP to topoisomerase II induces a conformational change in the enzyme (1, 2). However, we do not know whether different ATP molecules bind to the A and B sites simultaneously or sequentially. Were the latter the case, the conformational change could be split up into steps which were differentially sensitive to the drug classes. Thus, future studies on recombinant topoisomerase IIα carrying site-directed mutations in and around the Walker A and B sites may yield more precise biochemical information on the interaction of these important anticancer agents with their main intracellular molecular target.

We are much obliged to the Drug Synthesis and Chemistry Branch, Developmental Therapeutics Program, Division of Cancer Treatment, National Cancer Institute, NIH (Bethesda, MD), for the gifts of merbarone and fostriecin. The technical assistance of Susanne Rasmussen, Sanne Christiansen, and Annette Nielsen is highly appreciated.