The different cleavage DNA sequence specificity explains the camptothecin resistance of the human topoisomerase I Glu418Lys mutant

Abstract

Yeast cells expressing the Glu418Lys human topoisomerase I mutant display a camptothecin resistance that slowly decreases as a function of time. Molecular characterization of the single steps of the catalytic cycle of the purified mutant indicates that it has a relaxation activity identical to the wild-type protein but a different DNA sequence specificity for the cleavage sites when compared to the wild-type enzyme, as assayed on several substrates. In particular the mutant has a low specificity for CPT sensitive cleavable sites. In fact, the mutant has, at variance of the wild-type enzyme, a reduced preference for cleavage sites having a thymine base in position -1 of the scissile strand. This preference, together with the strict requirement for a thymine base in position -1 for an efficient camptothecin binding, explains the temporary camptothecin resistance of the yeast cell expressing the mutant and points out the importance of the DNA sequence in the binding of the camptothecin drug.

Figure 1

Glu418Lys is partially resistant to CPT in vivo. (A) Exponentially growing cells in dextrose, transformed with YEpGAL1-wild-type, YEpGAL1-Glu418Lys or YEpGAL1 (vector), serially 10-fold diluted starting from an A₅₉₅ of 0.3; 5 µl, and spotted onto selective media in the presence of dextrose (left) or dextrose plus 0.5 µg/ml CPT (middle), or induced with galactose (right). (B) Number of colonies relative to that obtained at time 0 plotted against time for wild-type, (circle), Glu418Lys mutant (square), Ala653Pro mutant (diamond) and vector (triangle). Exponentially growing cells in dextrose transformed with YCpGAL1-wild-type, YCpGAL1-Glu418Lys, YCpGAL1-Ala653Pro or YCpGAL1 (Vector) were diluited 1:100 into selective medium containing 2% raffinose. After induction with 2% galactose cells were treated with 50 µM CPT or 1% Me₂SO. At various time point aliquots were serially diluted and plated onto selective media containing 2% dextrose. Three different plates were averaged for each data point, and the error bars indicate the standard deviation of the individual values from the mean.

Figure 2

Cleavage/religation equilibrium of the CL25/CP25 fully duplex DNA substrate. Gel electrophoresis of the products coming from the incubation of the wild-type topoisomerase I with the [γ-³²P] end-labelled duplex DNA, shown at the top of the figure in the absence (lane 1) and presence (lanes 2–6) of increasing amount of CPT. The arrow at the DNA sequence indicates the CL1 site preferred by the wild-type protein. Lanes 7 and 8–12 show the same experiment with the Glu418Lys mutant. The asterisk indicates the band corresponding to the CL1 site.

Figure 3

Suicide cleavage experiment with the CL14/CP25 substrate for the wild-type and Glu418Lys mutant. (A) Time course of the suicide cleavage reaction carried out with the substrate described on the top of the figure. CL1 and CL2 identify the cleaved complexes at the site indicated by the arrow and the asterisk, respectively. Lane C, no protein added. (B) Percentage of the DNA substrate cleaved at the CL1 or CL2 site for the wild-type and the Glu418Lys mutant.

Figure 4

Religation experiment with the R13 oligonucleotide for the wild-type and Glu418Lys topoisomerase I. (A) Time course of the religation experiment between the R13 substrate and the wild-type or Glu418Lys covalent complexes, in the absence or presence of 100 µM CPT. The R13 oligonucleotide is selectively religated to the CL2 site but not to the CL1 site. (B) Percentage of the remaining covalent complex 2 plotted at different time for the wild-type and Glu418Lys (black and red lines, respectively), in absence (full line) and presence (dashed line) of CPT. Three different religation experiments were averaged for each data point, and the error bars indicate the standard deviation of the individual values from the mean.

Figure 5

Religation experiment with the R11 oligonucleotide for the wild-type and Glu418Lys topoisomerase I. (A) Time course of the religation experiment between the R11 substrate and the wild-type or Glu418Lys covalent complexes, in the absence or presence of 100 µM CPT. The R11 oligonucleotide is selectively religated to the CL1 site but not to the CL2 site. (B) Percentage of the remaining covalent complex 1 plotted at different time for the wild-type and Glu418Lys (black and red lines, respectively), in absence (full line) and presence (dashed line) of CPT. Note the difference in scale between this figure and Figure 4. Three different religation experiments were averaged for each data point, and the error bars indicate the SD of the individual values from the mean.

Figure 6

Suicide cleavage experiment with the CL14-A/CP25-T or the CL14-G/CP25-C substrates for the wild-type and Glu418Lys mutant. Left side: time course of the suicide cleavage reaction carried out with the CL14-A/CP25-T substrate, described on the top of the figure. CL1 and CL2 identify the cleaved complexes at the site indicated by the arrow and the asterisk, respectively. Lane C, no protein added. Right side: time course of the suicide cleavage reaction carried out with the CL14-G/CP25-C substrate, described on the top of the figure.

Figure 7

Cleavage/religation equilibrium of a 900 bp fully duplex DNA substrate. Gel electrophoresis of the products coming from the incubation of a 900 bp ³²P-end-labelled DNA duplex with the wild-type topoisomerase I (lanes 1–5) or Glu418Lys mutant (lanes 6–10), as a function of time, in the absence (lanes 5 and 10) or in the presence of 1 µM CPT (lanes 1–4 and 6–9). Control, no enzyme added (C). Arrow indicates the preferred cleavage site.

Figure 8

Human topoisomerase I structure in complex with duplex DNA. The main chain of the 417–423 loop, contacting the intact strand of the lip 1 (represented in red), is shown in yellow. The mutated Glu418 and the catalytic Tyr723 residues are represented in ball and stick, in green and yellow colours, respectively. The −1 base of the scissile strand, strictly required for CPT binding, is coloured in blue; the −1 base of the intact strand, that contacts the 417–423 loop, is represented in purple. All other enzyme residues and DNA bases are represented in grey. Core subdomain III is partial transparent.