Skip to main content
PMC home page
Nucleic Acids Research logoLink to Nucleic Acids Research
. 2003 Dec 15;31(24):7255–7263. doi: 10.1093/nar/gkg927

Residues 190–210 of human topoisomerase I are required for enzyme activity in vivo but not in vitro

Morten O Christensen 1, Hans U Barthelmes 1, Fritz Boege 1, Christian Mielke 1,*
PMCID: PMC291868  PMID: 14654701

Abstract

DNA-topoisomerase I (topo I) unwinds the DNA- double helix by cutting one strand and allowing rotation of the other. In vitro, this function does not require the N-terminal domain of the enzyme, which is believed to regulate cellular properties. To assess this role, we studied the cellular distribution and mobility of green fluorescent protein-chimera of human topo I lacking either the entire N-terminal domain or a portion of it. We find that topo I truncated up to position 210 is not stabilized by camptothecin in covalent DNA-complexes inside a living cell, whereas in vitro it retains full DNA-relaxation activity, and is targeted by camptothecin in the usual manner. This difference is not shared with a fragment lacking the N-terminal domain up to position 190, indicating that residues 190–210 play a crucial role for the activity of the enzyme in its physiological environment, but not in vitro. Since it is impossible to discriminate in vivo whether this region is required for topo I to form covalent DNA intermediates in the cell, or just for camptothecin to bind and stabilize such complexes, we could not explain precisely these cellular observations. However, inactivity in vivo of the enzyme lacking this region is indicated by a lesser cytotoxicity.

INTRODUCTION

DNA-topoisomerase I (topo I) changes the pitch of DNA-double helices by cutting one DNA strand and allowing rotation of the complementary strand in a controlled manner. This mechanism is well understood (1). It plays an established role in the release of torsion stress adjacent to replication forks (2) and transcription complexes (3), and may also be involved in DNA-recombination (4), DNA-damage (5), DNA-repair (6–10) and chromosome organization (11). It is unclear, how inside a living cell the enzyme is assigned to these diverse tasks and how they are orchestrated with complex nuclear processes and higher-order chromatin structure.

According to X-ray crystallography (12) and biochemical analyses (13,14) topo I is organized into four domains: N-terminal domain (Met1-Gly214), core (Ile215-Ala635), linker (Pro636-Lys712) and C-terminal domain (Gln713-Phe765). The last three domains constitute the minimal active unit of the enzyme. Together, they form a clamp-like structure carrying out the basic type I topoisomerase reactions, such as relaxation of supercoiled plasmid-DNA (13,15). The function of the N-terminal domain is less clear, because this portion has eluded structural analysis until recently (13,14). It is not essential for DNA-relaxation activity in vitro (16–18). On the other hand, evolution has preserved a similar, though mainly genetically divergent domain in all type IB topoisomerases (19), suggesting that it is essential—albeit in a manner not readily apparent in vitro or delineated by sequence motifs.

Some functions of the N-terminal domain are known: it serves as a docking place for interacting proteins (9,20–23). It contains nuclear localization signals (NLSs) (24,25) and putative target sequences for protein phosphorylation (26,27). It influences the distribution of the enzyme in the cell and in the cell nucleus (9,21,28,29). These features suggest that the N-terminal domain provides some kind of a handle by which the catalytic machinery of topo I is controlled and directed in vivo. More recent evidence suggests that the N-terminal domain could also be involved in the catalytic process itself. It has been observed that a fragment lacking more or less the entire N-terminal domain (amino acids 1–206) has catalytic properties in vitro that are slightly different from full-length topo I (18), whereas this is not the case for a fragment lacking at the N-terminal end 16 amino acids less (17). Consistent with this, two recent X-ray crystallography studies extending the structure until Gln201 revealed interactions between Trp203 and His346, and between Trp205, Trp206 and the flexible hinge, which may be important for proper closing of the enzyme around its DNA substrate (30,31). Thus, a region within the N-terminal domain of human topo I spanning approximately residues 190–210 seems to play a role in the mechanism of DNA-catalysis, although it is not part of the genetically conserved minimal active unit of the enzyme. On the other hand, the covalent topo I-DNA cleavage complex is equally stabilized by camptothecin in vitro, irrespective of the presence or absence of the first 206 residues (18).

Here, we address the question of what role this region plays in the functioning of human topo I inside a living cell. For this purpose we removed, at the N-terminal end, either 190 or 210 amino acid residues, and established human cell lines expressing bio-fluorescent chimera of these truncated enzymes. We used a bicistronic expression vector (32), conferring stable and permanent expression of the fluorescent enzymes at levels similar to that of endogenous topo I. Recently this approach has allowed us to investigate localization, mobility and DNA-interactions of full-length human topo I in substructures of the living cell nucleus (33). By this same approach, we show here that human topo I lacking amino acids 190–210 does not interact in a catalytic manner with the DNA inside a living cell, although it is active in vitro.

MATERIALS AND METHODS

Constructs and cell culture

Cell lines supporting stable expression of fusion proteins of green fluorescent protein (GFP) and variants of topo I followed published protocols (28,33). Briefly, GFP-chimera of full-length human topo I, of the N-terminally truncated fragments topo I190–765, topo I210–765 and of the active site mutant topo IPhe723 were stably expressed in the human embryonal kidney cell line 293 (German Collection of Micro organisms and Cell Cultures, Braunschweig, Germany) using the bicistronic expression vector pMC-2P (32), in which the translational initiation of the selection marker puromycin-N-acetyl transferase in the second cistron is mediated by an internal ribosome entry site from poliovirus. Since the N-terminally truncated versions of topo I lack a NLS, PCR-primers were supplemented with the NLS of SV40 (PKKKRKV), thus generating amino acid sequences YK-RWTRPPKKKRKVPP-K190K191 and YK-RWTRPPKKKRKVPP-R210Y211 for GFP–topo I190–765 and GFP–topo I210–765, respectively. (The last amino acids of GFP and the first ones of topo I are indicated in bold letters.) Cells grown in DMEM supplemented with Glutamax-I (Gibco BRL/Life Technologies, Karlsruhe, Germany) were transfected using Lipofectamine (Gibco BRL/Life Technologies, Karlsruhe, Germany). Stable transgenic cell lines were selected after 2 days with 0.35 µg ml–1 puromycin and maintained under selection. We have ascertained previously (28,33) that (i) the GFP-chimera were not heavily over-expressed, (ii) the chimeric genes were not rearranged, (iii) green fluorescence of the cells could be unambiguously assigned to constitutive expression of the intended proteins and (iv) dose response to camptothecin was equal for GFP–topo I and endogenous topo I.

Life cell imaging

For confocal imaging and for measurements of fluorescence recovery after photobleaching (FRAP), we used a Zeiss LSM 510 inverted confocal laser scanning microscope. A CO2-controlled on-stage heated cell chamber and a heated 63x/1.4 NA oil-immersion objective were used, thus allowing culturing of the cells under the microscope at 37°C. For FRAP measurements, fluorescent images of a single optical section were taken at 1.6 s time intervals before (n = 5) and after bleaching of a circular area at 20 mW nominal laser power with three iterations. Imaging scans were acquired with the laser power attenuated to 0.1–1% of the bleaching intensity. For a quantitative analysis of FRAP, fluorescence intensities of the bleached region and the entire cell nucleus were measured at each time point. Data were corrected for extra-cellular background intensity and for the overall loss in total intensity as a result of the bleach pulse itself and of the imaging scans. The relative intensity of the bleached area Irel was calculated according to (34). We ascertained that FRAP results were independent of the expression level of GFP–topo I, in as much as we failed to detect any differences in FRAP between individual cells exhibiting strong or weak GFP fluorescence within a cell population expressing the same construct.

Protein purification

Endogenous topo I and GFP–topo I210–765 from the same batch of transfected 293 cells were purified as described in (35,36) and summarized in Figure 2A. Endogenous topo I was adsorbed from nuclear extracts (330 mM NaCl) to Ni-NTA–Sepharose (Quiagen, Hilden, Germany), eluted from there with 200 mM imidazol, then adsorbed to phenyl–Sepharose (Amersham-Pharmacia-Biotech, Freiburg, Germany) in the presence of 0.9 M ammonium sulfate, and finally eluted from the latter resin with buffer lacking ammonium sulfate. The enzyme was thus purified to apparent homogeneity and had the expected size (Fig. 2B, lane 3). GFP–topo I210–765 present in the same nuclear extracts was separated from endogenous topo I, as it did not bind to Ni-NTA–Sepharose. It was recovered from the effluent of the Ni-NTA column by hydrophobic interaction chromatography (as described above) and further purified by adsorption to heparin–Sepharose (Amersham-Pharmacia-Biotech, Freiburg, Germany), followed by elution with 800 mM NaCl. In the final eluate endogenous topo I or topo II were absent and GFP–topo I210–765 constituted ∼70% of the protein (Fig. 2B, lane 2).

Figure 2.

Figure 2

Figure 2

Open in a new tab

In vitro-assessment of DNA-relaxation and camptothecin-induced DNA-cleavage by GFP–topo I210–765 and endogenous topo I purified from the same batch of cells. (A) Strategy of purification. (B) Characterization of purified enzymes. Lane 1: recombinant full-length human topo I produced in yeast. Lane 2: GFP–topo I210–765. Lane 3: Endogenous topo I. (Top) SDS-gel electrophoresis (7.5% polyacrylamide gels) followed by protein staining with silver nitrate. (Middle) Western-blotting and immuno-staining with topo I antibodies. (Bottom) Western-blotting and immuno-staining with GFP antibodies. (C) Characterization of purified enzymes. Western-blotting and immuno-staining with topo I antibodies. Lane 1: recombinant full-length human topo I produced in yeast. Lane 2: GFP–topo I210–765. Lane 3: Endogenous topo I. Lane 4: Mixture of GFP–topo I210–765 and endogenous topo I. (D) DNA-relaxation: supercoiled pUC18 plasmid-DNA (300 ng) was either reacted with increasing amounts of recombinant topo I, endogenous topo I or GFP–topo I210–765 (the wedge indicates serial 1: two dilutions of the enzymes starting with 3 ng for a fixed time of 20 min (left), or it was reacted with a fixed amount (1 ng) of the enzymes for various time periods (right). DNA-electrophoresis was in the absence of ethidium bromide in order to separate supercoiled (SC) and relaxed (RL) plasmid forms. (E) Salt optimum for DNA-relaxation. Similar reactions as in (D) were done with fixed amounts of recombinant topo I, endogenous topo I or GFP–topo I210–765 and increasing KCl concentrations (wedges indicate 50, 75, 100, 125, 150, 200, 300, 400 mM KCl) for a fixed time of 20 min. Bars indicate increasing amounts (0.25, 0.5, 1 ng) of enzyme added to the reactions. (F) Salt extractability in vivo. Nuclei from 293 cells expressing full-length GFP–topo I (top) or GFP–topo I210–765 (bottom) were sequentially extracted with increasing amounts of NaCl, and extracted proteins were subjected to GFP-directed immuno-blotting. The wedge indicates increasing amounts of salt (100, 150, 200, 250, 300, 400 mM NaCl) in the extraction buffer. (G) DNA-cleavage: similar reactions as in (D) were done in the absence (left) or presence (right) of 10 µM camptothecin (CpT). Wedges indicate increasing amounts (5, 20, 40, 80 ng) of enzyme added to the reactions. DNA-electrophoresis was in the presence of ethidium bromide in order to separate closed (SC, RL) and nicked (OC) plasmid forms. (H) Camptothecin sensitivity: similar reactions as in (G) were done with a fixed amount (40 ng) of purified topo I in the presence of increasing concentrations of camptothecin (0.1, 0.5, 1, 2, 5, 10 and 20 µM are indicated by wedges). (I) Salt stability: similar reactions as in (G) were done in the presence of increasing amounts of KCl (50, 100, 150 mM) or KGlu (33, 66, 100 mM). In the latter case, KGlu also served as a buffer substance. After adjustment with KOH to pH 7.9 the ratio of K to Glu was 2:1. Thus, KCl (top) and KGlu (bottom) are compared under the wedges at comparable ionic strength.

Immunoblotting

We analyzed either nuclear extracts, or whole cell lysates. Whole cell lysates were prepared by adding to cells suspended in PBS an equal volume of 2-fold lysis buffer [25 mM Tris–HCl, pH 6.8, 10% SDS, 8 M urea, 20% glycerol, 0.04% bromophenol blue, 10 mM 4-(2-aminoethyl)-benzenesulfonyl fluoride, 1 mM PMSF, 20 µg ml–1 aprotinin, 10 µg ml–1 pepstatin A]. Material equivalent to 5 × 105 cells was then applied to each slot of an SDS-gel (8% polyacrylamide). After electrophoresis, proteins were electro-blotted onto PVDF-membranes (Immobilon P, Millipore, Bedford, MD). The membranes were subsequently blocked with PBS, containing 2% BSA, 0.05% Tween 20, and then incubated for 1 h with the primary antibodies diluted with the same buffer. Topo I was stained with rabbit peptide antibodies against the last 18 C-terminal amino acid residues of the human enzyme (Genosys, Cambridge, UK). GFP was stained with mouse monoclonal antibodies (Clontech, Heidelberg, Germany). After washing, the filters were incubated for 1 h with HRP-conjugated goat anti-rabbit or goat anti-mouse antibodies diluted with PBS containing 2% BSA and 0.1% Tween 20. Following extensive washing with the same buffer, labelled protein bands were finally visualized with the ECL Plus system (Amersham Pharmacia Biotech, Freiburg, Germany). For immuno-band depletion, cells were first incubated with camptothecin (20 min) and the same concentration of drug was also added to the lysis buffer. Immuno-band depletion assays with nuclear extracts were carried out under the same conditions as plasmid-relaxation assays (see next paragraph), with the exception that 0.1 mg/ml calf thymus DNA was added to the reaction and the reaction product was applied to an SDS-protein gel instead of an agarose DNA-gel.

Sequential salt extraction of nuclei

Exponentially growing cells were harvested, sedimented (600 g, 3 min), carefully resuspended (5 × 109 cells/3 ml) in lysis buffer [10% sucrose, 15 mM HEPES pH 7.5, 0.5 mM EGTA, 60 mM KCl, 15 mM NaCl, 30 µg ml–1 spermine, 7.5 µg ml–1 spermidine, 5 µg ml–1 leupeptin, 5 µg ml–1 pepstatin A, 10 µg ml–1 aprotinin, 1 mM DTT, 1 mM 4-(2-aminoethyl)-benzenesulfonyl fluoride, 1 mM PMSF], and incubated for 5 min on ice to allow for cell swelling. Cells were then lysed by addition of Triton-X-100 to a final concentration of 0.5%, followed by a brief incubation (2 min on ice). Nuclei were then purified by centrifugation across a 20% sucrose-cushion (1500 g, 15 min, 4°C), washed once with lysis buffer, and finally extracted (5 × 109 nuclei/200 µl) with 100 mM NaCl in extraction buffer [50 mM HEPES pH 7.5, 0.5 mM EDTA, 50 µg ml–1 leupeptin, 5 µg ml–1 pepstatin A, 1 mM benzamidine, 10 µg ml–1 aprotinin, 1 mM DTT, 1 mM 4-(2-aminoethyl)-benzenesulfonyl fluoride, 1 mM PMSF, 10% glycerol]. Nuclear remnant and extracted proteins were separated by centrifugation (2000 g, 5 min, 4°C). Nuclear remnant from the first extraction step was washed once in 100 mM NaCl in extraction buffer and then extracted again with 150 mM NaCl in extraction buffer, and again separated by centrifugation. This procedure was repeated with 200, 250, 300 and 400 mM NaCl. Extracted proteins from each extraction step were subjected to SDS–PAGE and GFP-directed immuno-blotting.

RESULTS

In vivo, topo I lacking the entire N-terminal domain (topo I210–765) does not respond to camptothecin, whereas topo I lacking 20 residues less (topo I190–765) does

To examine the physiological role of residues 190–210, we evaluated the cellular distribution of various GFP–topo I-chimera. Figure 1A shows in the middle representative fluorescent images of cells at interphase expressing GFP-chimera of full-length human topo I (row 1, GFP–topo I), of the active site mutant topo IPhe723 (row 2, GFP–topo IPhe723), of topo I190–765 lacking a major portion of the N-terminal domain (row 3, GFP–topo I190–765), and of topo I210–765 lacking the entire N-terminal domain (row 4, GFP–topo I210–765), which is considered the minimal portion of the enzyme retaining full activity in vitro. With the exception of the active site mutant GFP–topo IPhe723 (row 2), which resides preferentially in the nucleoplasm due to its prolonged binding to genomic DNA (33), all these constructs accumulated in the nucleoli. This conflicts with previous data (25) claiming the N-terminal domain to be required for nucleolar localization—an issue addressed in a recently published work (28).

Figure 1.

Figure 1

Open in a new tab

Distribution, mobility and camptothecin sensitivity of various topo I-fragments, expressed as GFP-chimera in 293 cells. (A) Fluorescent images of cells at interphase expressing GFP-chimera of full-length human topo I, the active site mutant topo IPhe723, and the N-terminally truncated enzymes topo I190–765 and topo I210–765 are shown before (middle) and after (right) exposure to camptothecin (20 µM, 20 min). Corresponding images obtained by transmitted light are shown on the left. (B) FRAP curves determined in the nucleoplasm of cells treated with 20 µM camptothecin for 20 min (closed triangles) or not (open triangles). The lines represent all data points, some of which are highlighted by symbols. The actual bleach spots are indicated in the corresponding confocal images of GFP-fluorescence of the cells, shown at the top of the curves.

Here, we will concentrate on the effects of camptothecin on the nuclear distribution and the mobility of these constructs, which are demonstrated in Figure 1A, right and Figure 1B, respectively. For the understanding of these data one must keep in mind that camptothecin binds and stabilizes covalent catalytic DNA-intermediates of topo I (37). Therefore, it renders GFP-chimera of topo I less mobile in the cell. This can be measured by FRAP. Such experiments have revealed recently that camptothecin targets topo I preferentially in the nucleoplasm, thus shifting the enzyme’s nuclear distribution equilibrium away from the nucleoli (33). Both effects [nucleolar de-localization (38,39) and nucleoplasmic immobilization (33)] require topo I to be active in the cell nucleus. Therefore, the two phenomena can be used as criteria for judging topo I cleavage activity in vivo. The effect of camptothecin on the localization of the various topo I constructs can be gauged by comparing the middle and right columns of Figure 1A, where GFP-fluorescence of the same set of cells is shown before and after exposure to the drug. Figure 1B shows corresponding FRAP curves determined in the nucleoplasm of cells treated with camptothecin (closed symbols) or not (open symbols). It is apparent that full-length GFP–topo I became depleted from the nucleoli in response to camptothecin, and that FRAP was much faster without camptothecin than with the drug present. This confirms our previous observation that camptothecin renders topo I less mobile in the nucleoplasm, and thus induces nucleolar depletion of the enzyme (33). In contrast, GFP–topo IPhe723 was not affected by camptothecin with respect to its localization, and FRAP curves of the active site mutant were virtually the same with and without the drug. This observation fits the fact that in lack of the active site tyrosine, GFP–topo IPhe723 does not form covalent DNA-intermediates to be targeted by camptothecin. The construct with a partial truncation of the N-terminal domain (GFP–topo I190–765) was affected by camptothecin in the same fashion as the full-length enzyme. It was depleted from the nucleoli and became less mobile in the nucleoplasm. Similar results were also obtained with constructs lacking even smaller portions of the N-terminal domain (not shown). In contrast, the construct lacking the entire N-terminal domain (GFP–topo I210–765) was hardly affected by camptothecin. It was not depleted from the nucleoli by camptothecin, and its mobility in the nucleoplasm was almost the same with or without the drug.

In vitro, GFP–topo I210–765 has the same DNA-relaxation activity and the same sensitivity to camptothecin as full-length topo I at physiologically relevant salt concentrations

These findings suggested two alternative interpretations: (i) GFP–topo I210–765 is not performing cleavage; (ii) GFP–topo I210–765 is actively cleaving DNA, but resistant to camptothecin. To discriminate between these possibilities, we first determined whether GFP–topo I210–765 was actively cleaving DNA in vitro. In vivo, GFP–topo I190–765 and GFP–topo I have the same sensitivity towards camptothecin (Fig. 1B). Moreover, GFP–topo I and endogenous topo I are known to respond equally well to camptothecin in vivo (33). Thus, it is reasonable to assume that GFP–topo I190–765 behaves similar to endogenous topo I, at least with respect to camptothecin. Therefore, endogenous topo I is a suitable control for the in vitro assessment of the cleavage activity of GFP–topo I210–765. Following this strategy, we purified from the same batch of nuclear extract GFP–topo I210–765 and endogenous topo I, and compared the two enzyme varieties with respect to DNA-relaxation activity in vitro. Figure 2A outlines the purification strategy and Figures 2B and C demonstrate the outcome with respect to enzyme purity. The top panel of Figure 2B shows the result of SDS-gel electrophoresis followed by protein staining. It can be seen that endogenous topo I (lane 3) was purified to apparent homogeneity similar to recombinant human topo I produced in yeast (lane 1), whereas GFP–topoI210–765 (lane 2) was ∼60–70% pure. Immuno-blotting with topo I antibodies (Fig. 2B, middle) demonstrates that the major bands were indeed topo I, that the three preparations had a similar concentration of topo I, and that the enzymes were mostly not degraded. GFP–topo I210–765 (lane 2) appeared to be marginally larger than endogenous topo I. This feature was employed to exclude a contamination with endogenous topo I. First, we extended electrophoresis time to separate GFP–topo I210–765 and endogenous topo I more clearly (Fig. 2C, compare lanes 2 and 3). Secondly, we mixed GFP–topo I210–765 and endogenous topo I and demonstrated that we could separate the two proteins (Fig. 2C, lane 4). Thus, we would have detected a significant contamination of GFP–topo I210–765 with endogenous topo I by a double band. It should be noted that proteolysis in the linker domain would not show up in these blots, because they were probed with a serum against a C-terminal peptide of topo I. However, such fragments would reconstitute the so-called Core + C-terminus-topo I known to have a reduced sensitivity to camptothecin (16). Thus, they would not interfere with the camptothecin assays shown below, because they would be less active and not a likely cause of false positive results. Immunoblotting with GFP antibodies (Fig. 2B, bottom) shows that GFP–topoI210–765 (lane 2) was almost entirely separated from endogenous topo I (lane 3), which contained only insignificant contaminations of GFP-positive material. By immunoblotting with topo II-antibodies we have also ascertained that none of these preparations was contaminated with type II topoisomerases (not shown). Figure 2D shows a comparison of the in vitro DNA-relaxation activity of the enzymes, determined by the limiting dilution of DNA-relaxation endpoints (left) and by relaxation kinetics obtained with equivalent enzyme amounts (right). By both approaches, GFP–topo I210–765 (bottom) had the same in vitro activity as endogenous full-length topo I prepared from the same batch of nuclear extract (middle), whereas recombinant human topo I produced in yeast (top) was slightly less active, which represents an established finding (36). Thus, GFP–topo I210–765 relaxes supercoiled DNA as efficient as endogenous topo I under low salt conditions in vitro. However, it has been suggested that the N-terminal domain is important for DNA binding, and that topo I lacking the first 206 amino acids has a lower salt optimum than full-length enzyme in vitro (18), raising the possibility that GFP–topo I210–765 is inactive at physiological salt concentrations. Figure 2E shows a comparison of the in vitro DNA-relaxation activity of the enzymes in the presence of increasing concentrations of KCl (50–400 mM). Salt optima were determined at low (left), medium (middle) or high (right) enzyme concentrations. At high enzyme concentrations, recombinant human topo I (top) and endogenous full-length topo I (middle) exhibited the same relaxation activity. At low enzyme concentrations, both exhibited their optimum at 200 mM KCl. In contrast, GFP–topo I210–765 (bottom) was less active at high salt concentrations and had its optimum at 150 mM KCl. These observations are consistent with the notion that the N-terminal domain plays a role in DNA binding. To further corroborate this finding, we compared the salt stability of the binding to nuclear structures between GFP–topo I210–765 and GFP–topo I. Figure 2F shows a sequential salt extraction (100–400 mM NaCl) of nuclei from 293 cells expressing either GFP–topo I (top) or GFP–topo I210–765 (bottom). Fitting the data shown in Figure 2B, the nuclear binding of GFP–topo I210–765 was less salt stable than that of full-length GFP–topo I. The major fraction of GFP–topo I210–765 was extracted with 250–300 mM NaCl, whereas GFP–topo I remained stably bound until 300 mM NaCl. In conclusion, we find that topo I lacking the entire N-terminal domain is more salt sensitive and better extractable from the nuclear structure than the full-length enzyme. However, these alterations are not apparent at salt concentrations encountered inside a living cell under physiological conditions, and, therefore, should not play a crucial role.

Consequently, we considered the alternative hypothesis of GFP–topo I210–765 being resistant to camptothecin, and investigated if DNA cleavage by this enzyme was stimulated in vitro by the drug. Figure 2G demonstrates nicking of a double stranded DNA-plasmid in the presence of camptothecin as a function of topo I concentration. DNA cleavage by endogenous topo I (middle section) and GFP–topo I210–765 (right section) was significantly stimulated by camptothecin, and it can be seen clearly that in this respect the two sets of enzyme had a capacity similar to each other and also similar to recombinant human topo I produced in yeast (left section). Figure 2H demonstrates nicking of plasmid-DNA by the three sets of enzymes as a function of camptothecin concentration. It shows that GFP–topo I210–765 (right), endogenous topo I (middle) and the recombinant human enzyme (left) had a similar sensitivity to the drug in vitro. Again we needed to take into account that topo I becomes more sensitive to salt, when lacking the entire N-terminus (Fig. 2E) (18). To exclude this as a possible reason for the apparent camptothecin resistance of the enzyme inside the cell, we measured camptothecin-induced DNA-nicking in vitro in the presence of increasing concentrations of KCl (50–150 mM) or KGlu (33–100 mM), the latter resembling most closely the physiological intracellular saline milieu. Figure 2I (right section) demonstrates that in vitro DNA cleavage by GFP–topo I210–765 was not affected by KCl (top) or KGlu (bottom) within a concentration range to be encountered inside a living cell under physiological conditions. In summary, we concluded from these data (Fig. 2D–I) that in vitro and at optimal salt concentrations GFP–topo I210–765 was indistinguishable from endogenous topo I or the full-length recombinant human enzyme in as much as it was neither inactive, nor resistant to camptothecin. These in vitro-observations conform to the established finding that the N-terminal domain influences the salt sensitivity of topo I to some extent (18), but otherwise is mostly dispensable for the catalytic action of the enzyme (1), and that linkage to GFP does not appear to influence enzymatic properties of topo I210–765 in vitro. These observations are however in striking contrast to the behavior of GFP–topo I210–765 in vivo.

By a comparable assay, GFP–topo I210–765 is sensitive to camptothecin in vitro, but resistant in vivo

The most plausible hypothesis explaining the discrepancy of data obtained with GFP–topo I210–765 in vivo (Fig. 1A and B) and in vitro (Fig. 2D–I) would be that the innermost part of the N-terminal domain (i.e. the region spanning residues 190–210) is crucially required for the activity and/or the camptothecin sensitivity of the enzyme inside the living cell, whereas it is dispensable for these properties in vitro. We corroborated this notion by comparing camptothecin sensitivity of GFP–topo I210–765 in vivo and in vitro by immuno-band depletion, a biochemical assay that can be carried out in a similar manner with intact cells (i.e. in vivo) and with nuclear extracts supplemented with a surrogate of genomic DNA (i.e. in vitro). In both constellations, the enzyme will be trapped in large DNA–protein complexes upon addition of camptothecin, provided it is capable of interacting with DNA and sensitive to the drug. Thus, the active and camptothecin-sensitive fraction of the enzyme will be prevented from entering an SDS–polyacrylamide gel, and consequently will not be detectable in subsequent immuno-blotting. Figure 3 demonstrates a dose response curve obtained by treating a mixture of 293 cells expressing equal amounts of either GFP–topo I210–765 or full-length GFP–topo I with various concentrations of camptothecin, and a similar in vitro curve obtained by treating nuclear extracts of such cells substituted with calf thymus DNA. Inside the cells (Fig. 3, left), GFP–topo I210–765 (the lower band) was completely resistant to camptothecin, since it was not depleted upon exposing the cells to concentrations of the drug as high as 100 µM. In contrast, full-length GFP–topo I (the upper band) was completely lost from the blots at concentrations higher than 1 µM. Consistent with the in vivo situation in Figure 1A and B, GFP–topo I190–765 was also efficiently depleted at camptothecin concentrations >1 µM (not shown). A completely different situation was encountered in vitro (Fig. 3, right). Here, full-length GFP–topo I and GFP–topo I210–765 were both sensitive to camptothecin in a similar fashion. They both disappeared from the blot, when camptothecin was added to the assay at concentrations >1 µM. Taken together, these findings confirm that a region spanning residues 190–210 is absolutely required for topo I to be targeted by camptothecin inside the living cell, whereas it is dispensable for this property in vitro.

Figure 3.

Figure 3

Open in a new tab

Camptothecin-induced immuno-band depletion of GFP–topo I210–765 and full-length GFP–topo I in intact cells and nuclear extracts. 293 cells expressing GFP–topo I210–765 were mixed in a 1:4 ratio with 293 cells expressing full-length GFP–topo I. Living cells (left), or nuclear extracts thereof (right) were then treated for 20 min with various concentrations of camptothecin. Nuclear extracts were supplemented with 0.1 µg µl–1 of calf thymus DNA prior to the incubation. Incubations were stopped with lysis buffer and samples were subjected to GFP-directed immuno-blotting. The two outmost lanes on the left of the right panel show controls without calf thymus DNA ± camptothecin at the maximal concentration.

DISCUSSION

We show here that a fragment of human topo I lacking the entire N-terminal domain up to position 210 is not stabilized by camptothecin in covalent DNA-complexes inside a living cell, whereas in vitro it is targeted by camptothecin in the usual manner and retains full DNA-relaxation activity at physiologically relevant salt concentrations. This difference in behavior in vivo and in vitro is not shared with a fragment lacking the N-terminal domain up to residue 190 or smaller portions of it. Thus, our findings indicate residues 190–209 of topo I to play a crucial role for the activity of the enzyme in its physiological environment. Unfortunately, topo I DNA-relaxation activity in vivo cannot be measured directly. It can only be approximated from camptothecin stabilized cleavage activity [using assays such as immuno-band depletion (40), LM-PCR (41) or tardis (42)]. Consequently, we cannot discriminate whether the apparent insensitivity of GFP–topo I210–765 to camptothecin indicated in Figure 1A and B is due to inactivity or drug resistance, and whether residues 190–209 are required for topo I to form covalent intermediates with genomic DNA, or just for camptothecin to bind and stabilize such complexes in the cell.

The latter possibility, however, seems unlikely, since all residues and portions of topo I known to confer camptothecin resistance at the cellular level are placed around the active center (43) or in the linker domain (16) and not in the N-terminal domain. In keeping with this, the X-ray crystal structure of the topo I-DNA cleavage complex (30) did not suggest a role for the N-terminal domain in the interplay between drug, DNA and topo I. Thus, resistance seems an unlikely explanation for the lack of in vivo response of topo I210–765 to camptothecin. More likely, it seems that the enzyme has a dramatically reduced activity inside the living cell, when it lacks the N-terminal domain including residues 190–210. This notion is also supported by the observation that GFP–topo I210–765 appears to be less toxic to the cells than full-length topo I, since heterologous expression of this fragment is tolerated at 4-fold higher levels (28).

This reasoning implies that a region ranging from residues 190–210 of human topo I must somehow play a direct role in the catalytic cycle, and that this role is crucial in vivo, but not or only marginally so in vitro. It has been reported that in vitro the catalytic properties of topo I lacking the N-terminal domain up to position 206 are slightly different from those of the full-length enzyme. Most notably, DNA-relaxation is more distributive and trans-ligation to external DNA-pieces is impeded (18). In keeping with this, we observed here that GFP–topo I210–765 is better salt extractable from the nuclear structure and its activity is more salt sensitive. These features could indicate a reduced binding affinity to DNA and other nuclear proteins, which suggests that removal of the entire N-terminal domain changes conformation of the enzyme in a global manner. This is plausible in the light of recent structural data proposing a physical contact of Trp203 with His346 and of Trp205, Trp206 with the hinge region of the core domain, which determines flexibility of the enzyme clamp and allows for a control of strand rotation (31). In synopsis, these findings suggest that in the absence of the entire N-terminal domain including region 190–210, flexibility of topo I could be altered. Recent reconstitution experiments have shown that the ‘cap’ of topo I (amino acid residues 175–432) complements DNA cleavage by a portion of topo I comprising residues 433–765 (‘Δcap’), which lacks catalytic activity, although it contains all of the residues known to be important for catalysis and binds DNA with an affinity similar to that of the intact enzyme. These findings suggest that, when topo I binds DNA, two opposing lobes in the enzyme clamp tightly around the DNA helix to form the pre-cleavage complex, one of them being the ‘cap’ and the other ‘Δcap’ (44). Interestingly, only part of the ‘cap’ (i.e. 210–432) seems to be required in vitro for catalytic activity. However, considering the complexity of the DNA-structures encountered by topo I within the mammalian genome, and considering the highly limited space provided in the cell nucleus for the enzyme’s interactions with these structures, one could imagine that in vivo the flexibility of the enzyme clamp might actually become a factor that limits DNA-interactions of topo I in a more general manner. Thus, it can well be that an enzyme lacking the entire N-terminal domain is altogether precluded from engaging in catalytic DNA-interactions inside the living cell because the flexibility of the enzyme clamp is improperly controlled, whereas this does not play a role, when the same enzyme fragment deals in a test tube with comparably simple substrate molecules such as DNA-plasmids or oligonucleotides. On the other hand, residues 190–210 might contain interaction domains of nuclear proteins that position topo I in the nucleus and thus facilitate its activity in living cells. This view is clearly supported by the altered extractability of the truncated enzyme shown here, and by the striking observation that all proteins shown so far to interact with topo I bind the N-terminal domain. Moreover, some of these proteins have been shown to stimulate topo I activity in vitro and thus could be required for its functioning in the living cell (9,20–23).

Acknowledgments

ACKNOWLEDGEMENTS

We are grateful to Jörg Hacker and Hilde Merkert of the Research Center for Infectious Diseases, University of Würzburg, Germany, for generously providing access to their confocal laser-scanning microscope. The clone of topo IPhe723 was a generous gift of James J. Champoux, Department of Microbiology, University of Washington, Seattle, USA. Our work has been supported by the Deutsche Forschungs gemeinschaft (Bo 910/3-1, Bo 910/4-1, GRK 639 and HA 1434/13-1).

REFERENCES

  • 1.Champoux J.J. (2001) DNA TOPOISOMERASES: structure, function and mechanism. Annu. Rev. Biochem., 70, 369–413. [DOI] [PubMed] [Google Scholar]
  • 2.Tsao Y.P., Russo,A., Nyamuswa,G., Silber,R. and Liu,L.F. (1993) Interaction between replication forks and topoisomerase I-DNA cleavable complexes: studies in a cell-free SV40 DNA replication system. Cancer Res., 53, 5908–5914. [PubMed] [Google Scholar]
  • 3.Zhang H., Wang,J.C. and Liu,L.F. (1988) Involvement of DNA topoisomerase I in transcription of human ribosomal RNA genes. Proc. Natl Acad. Sci. USA, 85, 1060–1064. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Pommier Y., Jenkins,J., Kohlhagen,G. and Leteurtre,F. (1995) DNA recombinase activity of eukaryotic DNA topoisomerase I; effects of camptothecin and other inhibitors. Mutat. Res., 337, 135–145. [DOI] [PubMed] [Google Scholar]
  • 5.Pourquier P. and Pommier,Y. (2001) Topoisomerase I-mediated DNA damage. Adv. Cancer Res., 80, 189–216. [DOI] [PubMed] [Google Scholar]
  • 6.Gobert C., Bracco,L., Rossi,F., Olivier,M., Tazi,J., Lavelle,F., Larsen,A.K. and Riou,J.F. (1996) Modulation of DNA topoisomerase I activity by p53. Biochemistry, 35, 5778–5786. [DOI] [PubMed] [Google Scholar]
  • 7.Lanza A., Tornaletti,S., Rodolfo,C., Scanavini,M.C. and Pedrini,A.M. (1996) Human DNA topoisomerase I-mediated cleavages stimulated by ultraviolet light-induced DNA damage. J. Biol. Chem., 271, 6978–6986. [DOI] [PubMed] [Google Scholar]
  • 8.Mao Y., Okada,S., Chang,L.S. and Muller,M.T. (2000) p53 dependence of topoisomerase I recruitment in vivo. Cancer Res., 60, 4538–4543. [PubMed] [Google Scholar]
  • 9.Mao Y., Mehl,I.R. and Muller,M.T. (2002) Subnuclear distribution of topoisomerase I is linked to ongoing transcription and p53 status. Proc. Natl Acad. Sci. USA, 99, 1235–1240. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Stevnsner T. and Bohr,V.A. (1993) Studies on the role of topoisomerases in general, gene- and strand-specific DNA repair. Carcinogenesis, 14, 1841–1850. [DOI] [PubMed] [Google Scholar]
  • 11.Meyer K.N., Kjeldsen,E., Straub,T., Knudsen,B.K., Kikuchi,A., Hickson,I.D., Kreipe,H. and Boege,F. (1997) Cell-cycle coupled relocation of type I and II topoisomerases and modulation of catalytic enzyme activities. J. Cell Biol., 136, 775–788. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Redinbo M.R., Stewart,L., Kuhn,P., Champoux,J.J. and Hol,W.G. (1998) Crystal structures of human topoisomerase I in covalent and noncovalent complexes with DNA. Science, 279, 1504–1513. [DOI] [PubMed] [Google Scholar]
  • 13.Stewart L., Ireton,G.C. and Champoux,J.J. (1996) The domain organization of human topoisomerase I. J. Biol. Chem., 271, 7602–7608. [DOI] [PubMed] [Google Scholar]
  • 14.Stewart L., Ireton,G.C., Parker,L.H., Madden,K.R. and Champoux,J.J. (1996) Biochemical and biophysical analyses of recombinant forms of human topoisomerase I. J. Biol. Chem., 271, 7593–7601. [DOI] [PubMed] [Google Scholar]
  • 15.Stewart L., Redinbo,M.R., Qiu,X., Hol,W.G. and Champoux,J.J. (1998) A model for the mechanism of human topoisomerase I. Science, 279, 1534–1541. [DOI] [PubMed] [Google Scholar]
  • 16.Stewart L., Ireton,G.C. and Champoux,J.J. (1999) A functional linker in human topoisomerase I is required for maximum sensitivity to camptothecin in a DNA relaxation assay. J. Biol. Chem., 274, 32950–32960. [DOI] [PubMed] [Google Scholar]
  • 17.Bronstein I.B., Wynne-Jones,A., Sukhanova,A., Fleury,F., Ianoul,A., Holden,J.A., Alix,A.J., Dodson,G.G., Jardillier,J.C., Nabiev,I. et al. (1999) Expression, purification and DNA-cleavage activity of recombinant 68-kDa human topoisomerase I-target for antitumor drugs. Anticancer Res., 19, 317–327. [PubMed] [Google Scholar]
  • 18.Lisby M., Olesen,J., Skouboe,C., Krogh,B., Straub,T., Boege,F., Velmurugan,S., Martensen,P., Andersen,A., Jayaram,M. et al. (2001) Residues within the N-terminal domain of human topoisomerase I play a direct role in relaxation. J. Biol. Chem., 276, 20220–20227. [DOI] [PubMed] [Google Scholar]
  • 19.Redinbo M.R., Champoux,J.J. and Hol,W.G. (1999) Structural insights into the function of type IB topoisomerases. Curr. Opin. Struct. Biol., 9, 29–36. [DOI] [PubMed] [Google Scholar]
  • 20.Bharti A., Olson,M., Kufe,D. and Rubin,E. (1996) Identification of a nucleolin binding site in human topoisomerase I. J. Biol. Chem., 271, 1993–1997. [DOI] [PubMed] [Google Scholar]
  • 21.Edwards T., Saalem,A., Shaman,J., Dennis,T., Gerigk,C., Oliveros,E., Gartenberg,M. and Rubin,E. (2000) Role for nucleolin/Nsr1 in the cellular localization of topoisomerase I. J. Biol. Chem., 275, 36181–36188. [DOI] [PubMed] [Google Scholar]
  • 22.Haluska P., Salleem,A., Rasheed,Z., Ahmed,F., Su,E., Liu,L. and Rubin,E. (1999) Interaction between human topoisomerase I and a novel RING finger/arginine-serine protein. Nucleic Acids Res., 27, 2538–2544. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Rasheed Z.A., Saleem,A., Ravee,Y., Pandolfi,P.P. and Rubin,E.H. (2002) The topoisomerase I-binding RING protein, topors, is associated with promyelocytic leukemia nuclear bodies. Exp. Cell Res., 277, 152–160. [DOI] [PubMed] [Google Scholar]
  • 24.Alsner J., Svejstrup,J.Q., Kjeldsen,E., Sorensen,B.S. and Westergaard,O. (1992) Identification of an N-terminal domain of eukaryotic DNA topoisomerase I dispensable for catalytic activity but essential for in vivo function. J. Biol. Chem., 267, 12408–12411. [PubMed] [Google Scholar]
  • 25.Mo Y.Y., Wang,C. and Beck,W.T. (2000) A novel nuclear localization signal in human DNA topoisomerase I. J. Biol. Chem., 275, 41107–41113. [DOI] [PubMed] [Google Scholar]
  • 26.Pommier Y., Kerrigan,D., Hartman,K.D. and Glazer,R.I. (1990) Phosphorylation of mammalian DNA topoisomerase I and activation by protein kinase C. J. Biol. Chem., 265, 9418–9422. [PubMed] [Google Scholar]
  • 27.Samuels D.S., Shimizu,Y., Nakabayashi,T. and Shimizu,N. (1994) Phosphorylation of DNA topoisomerase I is increased during the response of mammalian cells to mitogenic stimuli. Biochim. Biophys. Acta, 1223, 77–83. [DOI] [PubMed] [Google Scholar]
  • 28.Christensen M.O., Barthelmes,H.U., Boege,F. and Mielke,C. (2002) The N-terminal domain anchors human topoisomerase I at fibrillar centers of nucleoli and nucleolar organizer regions of mitotic chromosomes. J. Biol. Chem., 277, 35932–35938. [DOI] [PubMed] [Google Scholar]
  • 29.Mo Y.Y., Wang,P. and Beck,W.T. (2000) Functional expression of human DNA topoisomerase I and its subcellular localization in HeLa cells. Exp. Cell Res., 256, 480–490. [DOI] [PubMed] [Google Scholar]
  • 30.Staker B.L., Hjerrild,K., Feese,M.D., Behnke,C.A., Burgin,A.B.,Jr and Stewart,L. (2002) The mechanism of topoisomerase I poisoning by a camptothecin analog. Proc. Natl Acad. Sci. USA, 99, 15387–15392. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Redinbo M.R., Champoux,J.J. and Hol,W.G. (2000) Novel insights into catalytic mechanism from a crystal structure of human topoisomerase I in complex with DNA. Biochemistry, 39, 6832–6840. [DOI] [PubMed] [Google Scholar]
  • 32.Mielke C., Tummler,M., Schubeler,D., von Hoegen,I. and Hauser,H. (2000) Stabilized, long-term expression of heterodimeric proteins from tricistronic mRNA. Gene, 254, 1–8. [DOI] [PubMed] [Google Scholar]
  • 33.Christensen M., Barthelmes,H., Feineis,S., Knudsen,B., Boege,F. and Mielke,C. (2002) Changes in mobility account for camptothecin induced subnuclear relocation of topoisomerase I. J. Biol. Chem., 277, 15661–15665. [DOI] [PubMed] [Google Scholar]
  • 34.Phair R.D. and Misteli,T. (2000) High mobility of proteins in the mammalian cell nucleus. Nature, 404, 604–609. [DOI] [PubMed] [Google Scholar]
  • 35.Straub T., Knudsen,B. and Boege,F. (2000) PSF/p54nrb stimulates ‘jumping’ of DNA topoisomerase I between separate DNA helices. Biochemistry, 39, 7552–7558. [DOI] [PubMed] [Google Scholar]
  • 36.Straub T., Grue,P., Uhse,A., Lisby,M., Knudsen,B.R., Tange,T., Westergaard,O. and Boege,F. (1998) The RNA-splicing factor PSF/p54 controls DNA-topoisomerase I activity by a direct interaction. J. Biol. Chem., 273, 26261–26264. [DOI] [PubMed] [Google Scholar]
  • 37.Hertzberg R.P., Caranfa,M.J. and Hecht,S.M. (1989) On the mechanism of topoisomerase I inhibition by camptothecin: evidence for binding to an enzyme-DNA complex. Biochemistry, 28, 4629–4638. [DOI] [PubMed] [Google Scholar]
  • 38.Danks M.K., Garrett,K.E., Marion,R.C. and Whipple,D.O. (1996) Subcellular redistribution of DNA topoisomerase I in anaplastic astrocytoma cells treated with topotecan. Cancer Res., 56, 1664–1673. [PubMed] [Google Scholar]
  • 39.Wadkins R.M., Danks,M.K., Horowitz,L. and Baker,S.D. (1998) Characterization of topotecan-mediated redistribution of DNA topoisomerase I by digital imaging microscopy. Exp. Cell Res., 241, 332–339. [DOI] [PubMed] [Google Scholar]
  • 40.Liu L.F. (1989) DNA topoisomerase poisons as antitumor drugs. Annu. Rev. Biochem., 58, 351–375. [DOI] [PubMed] [Google Scholar]
  • 41.Pondarre C., Strumberg,D., Fujimori,A., Torres-Leon,R. and Pommier,Y. (1997) In vivo sequencing of camptothecin-induced topoisomerase I cleavage sites in human colon carcinoma cells. Nucleic Acids Res., 25, 4111–4116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Padget K., Carr,R., Pearson,A.D., Tilby,M.J. and Austin,C.A. (2000) Camptothecin-stabilised topoisomerase I-DNA complexes in leukaemia cells visualised and quantified in situ by the TARDIS assay (trapped in agarose DNA immunostaining). Biochem. Pharmacol., 59, 629–638. [DOI] [PubMed] [Google Scholar]
  • 43.Gupta M., Fujimori,A. and Pommier,Y. (1995) Eukaryotic DNA topoisomerases I. Biochim. Biophys. Acta, 1262, 1–14. [DOI] [PubMed] [Google Scholar]
  • 44.Yang Z. and Champoux,J.J. (2002) Reconstitution of enzymatic activity by the association of the cap and catalytic domains of human topoisomerase I. J. Biol. Chem., 277, 30815–30823. [DOI] [PubMed] [Google Scholar]

Articles from Nucleic Acids Research are provided here courtesy of Oxford University Press

RESOURCES