A 5′-Phosphodiester Group Attached to Deoxyguanosine does not Accelerate the Hydrolysis of cis-[PtCl(NH3)2(dGuo)]+

A 5′-Phosphodiester Group Attached to Deoxyguanosine does not Accelerate the Hydrolysis of cis-[PtCl(NH3)2(dGuo)]+

Please download to get full document.

View again

of 12
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.

Kids & Toys

Publish on:

Views: 3 | Pages: 12

Extension: PDF | Download: 0

  A 5’-PHOSPHODIESTER GROUP ATTACHED TO DEOXYGUANOSINE DOES NOTACCELERATE THE HYDROLYSIS OF cis- [PtCI(NH)2(dGuo)] + Tiphaine Weber, Franck Legendre, Veronika Novozamsky and Jif Kozelka* Laboratoire de Chimie et Biochimie Pharmacologiques et Toxicologiques URACNRS 400, 45, rue des Saints-Pres, 75270 Paris Cedex 06, France Fax: +331 42 86 21 75; E-mail: kozelka@citi2.fr AbstractThe influence of the methylphosphoestergroup on the reversible reaction shown below was studied. Evaluation of the rate constants forthe systemdepicted as well as for the analogous equilibriuminvolving the nucleoside deoxyguanosine showed that whereas thechlorideanation is slowed down by the presence of the methylphosphoester group, the hydrolysis rate constant is not significantly altered. This result speaks against a catalytic role of the 5 -phosphodiester group in the hydrolysis of cisplatin monoadducts with DNA, as suggested previously (Kozelka   Barre, Chem. Eur. J. 1997, 3, 1405-1409). NH3 Gu + CI /   OH OMe kci kH20 NH 3 0  0 OH OMeThe reaction of Me-5’-dGMP- and dGuo with the diaqua form of cisplatin, cis-[Pt(NH3)2(H20)2] 2 in 0.1 M NaC104 was also investigated and the corresponding rate constants determined. The phosphodiester group accelerates the replacement of the first H20 ligand 10 times, and that of the second H20 ligand ~2 times. 1. Introduction A number of experimental observations have indicated that the antitumor activity of cisplatin (cis- [PtC12(NH3)2]) is related to  N damage caused by covalent platinum binding to nucleobases [1,2]. The cytotoxic effect is generally ascribed to the major cisplatin-DNA adducts, the 1.,2-GG and 1,2-AG intrastrand diadducts, which represent -85 of all adducts formed upon cisplatin-DNA interaction in vivoas well asafter  N platination under certain in vitro conditions [3]. The assumption that the 1,2-intrastrand crosslinks are at thesrcin of antitumor activity is based, on one hand, on the finding that in E. Coli, these adducts are indeed cytotoxic [4,5], and on the other hand, on the correlation between the levels of the 1,2- diadducts detected in the white blood cells of cisplatin-treated patients and the patients response to the treatment,as observed by Reed et al. [6]. However, a causalrelationship between the1,2-intrastrand crosslinks and anticancer activity in humans has never been demonstrated. We havepreviouslyhypothesized that the cytotoxic effect of cisplatin could be related to the platinum-DNA monoadducts [7,8]. Our hypothesis was based on the reasoning that the monoadducs, bearing a labile ligand, could  lure repair proteins and fix them in a covalent DNA-Pt-protein complex. Such a ternary complex formation withthe recognition part of the UvrABC excinuclease  N repair system of E. Coli was indeed observed in an experiment by Lambert et al. [9]. Another indication that the monoadducts could be related to antitumor activity was the finding that asymmetrical cis-diaminedichloro complexes with bulky substituents, forming with  N long-lived monoadducts, showed enhanced cytotoxicity [7,9]. Recently, Natile s and Farrell s groups reported substantial in vitro cytotoxicity and in vivo antitumor activity of trans-[PtCl2L2] complexes, with L being E-imino-ether [10,11] or quinoline [12,13]; these complexes are characterized by relatively long-lived monoadducs with  N [14-16]. There is conclusive evidence showing that cis-[PtC12(NH3)2] does not react with  N directly but through a solvent-assisted pathway [17-19]. Less clear is whether the monoaquated cisplatin form, cis- [PtCI(NH3)2(H20)] + (1), or the diaqua complex, cis-[Pt(NH3)2(H20)2] 2+ (2), is the major species undergoing the  N platination reaction. Whereas  N monoadductsformed upon the reaction with 2,  Vol. 6, No. 1, 1999 A 5 ’-Phsphdiester Group Attached To Deoxyguanosine DoesNot Accelerate the Hydrolysis of cis-[PtCl(NH3)2(dGuo] + bearing an aqua ligand, can rearrange to diadducts  chelates) directly, the chloro-monoadducts resulting from a reaction between DN and 1 have to be hydrolyzed to aqua-monoadducts before further ligand substitution by a nucleobase 19-21]. Thus, the chloro-monoadducts have considerably longer lifetimes than the aqua- monoadducts [19,22-25]. The rate ofhydrolysis of the former has been shown to depend on the adjacent bases, in particular on the base 5 to theplatinated guanine [26]. We have recently suggested that this sequence-dependence could be due to a catalytic action of the phosphodiester group flanking the monoadduct from the 5 side. Either a nucleophilic catalysis [27] or a general base catalysis are conceivable mechanisms; in both cases, the exact positioning of the phosphodiestergroup, which depends on the nature of the adjacent bases, would be expected to determine the hydrolysis rate. + 2+ Pt Pt HN   Cl HN   OH: 1 2 In order to test the possibleinvolvementof the 5’-phosphodiester group in the hydrolysis mechanism, we have studied in this work the reversible chlorideanation of the mononucleotide complex cis- [Pt(NH3)2(H20)(Me-5’-dGMP)]   3) [equation  1)]. Thecomplexes 3 and 4 are models for respectively aqua-andchloro-monoadducts formed between cisplatin and DNA. The hydrolysis and chlorideanation rate constants, kH20 and kc1 were compared to those determined in a parallel studyof the analogous reaction  2),   cis-[Pt(NH3)2(H20)(Me-5’-dGMP)]   +  I cis-[PtCI(NH3)2(Me-5’-dGMP)] + H20  1) 3 kH20 4 kc1 cis-[Pt(NH3)2(H20)(dGuo)] 2+ + CI knzo 5 cis-[PtCl(NH3)2(dGuo)] + + H20  2) where thenucleotide Me-5 -dGMP was replaced by the nucleoside deoxyguanosine. We were thus able to quantify the influence of the phosphodiester group on kH20 and kcl In addition, we havefollowed the formation of 3from 2 and Me-5 -dGMP, and that of 5 from 2 and deoxyguanosine,and determined the appropriate rate constants. The reaction between Me-5 -dGMP and 2 and the reversible anation of 3  Eq.1) were followedusing 1H NMR, whereas the analogous reactions with dGuo were analyzed by means ofreverse-phase HPLC. In the NMR-monitored runs, pH changes during the reactions were accounted for mathematically, while in the reactions followed by HPLC, the pH was kept constant. Advantages and disadvantagesofboth methods are discussed. 2. Experimental cis-[Pt(ND3)2(D20)2] 2+  2) was prepared by dissolving cis-[Pt(NO3)2(NH3)2] [28,29] in D20. 2 - deoxyguanosine (dGuo) was purchasedfromSigma.2’-deoxyguanosine-5’-monophosphate-methylester (Me-5’-dGMP) was prepared as ammonium salt by an adaptation of the method of Miller et al. [30]. 760 mg Dicyclohexylcarbodiimide (DCC, Aldrich) wereadded to a suspension of 5 -dGMP  free acid, Sigma, 250 mg) in 100 mL of methanol (Carlo Erba, pro analysis) and the mixture was stirred 48h at room temperature. The solvent was evaporatedunder vacuum to ~3 mL. DCC and its by-productswere precipitated by addition of 50 mL of water and filtered off in two steps: first, using filter paper, and second, passing through a D4 glass frit  filtering over frit directly congests the frit The filtrate was extractedwith 3x20 mL of cyclohexane,evaporated under vacuum to ~10 mL and the rest lyophilised to dryness. The colorless microcrystalline material obtained was dissolved in ~3 mL D20 and the solution passedthrough a column filled with~0.5 mL of a Dowex R)-50 resin (Sigma) charged with NH4 + and rinsed with D20. The purity of thecollected fractions was checked using 1H NMR. The fractions with satisfactory purity were assembled and lyophilized. The final product was stored underargon at -32 C.   thermoanalysis revealed a 5.73 weight loss in the temperature range between 30 and120 C, witha maximum rate at 70 C, and a 15.13 weight loss between120 and 250 C, with a maximum rate   170 C. The first step was completely reversible when the samplewas keptunder ambient atmosphere, and was attributed to 1.5 equivs, of  adsorbed H20  theor.: 6.54 ).During the second, irreversible step, the sample turned black, indicating decomposition. Anal.  deuterated sample): Calcd. for C HI DsN607 P 1.5 D20: C, 31.74; N, 20.19 . Found: C 31.10; N, 19.48 .  Tiphaine Weber et al. Metal-Based Drugs The samples for the kinetic runs to befollowed by NMR were prepared by weighing all quantities  including the liquid components) using a semimicro balance precision +_0.01 mg). All reactions were carried out in 0.1 M NaC104 at 20 C. The H NMR spectra were recorded on a Bruker ARX 250 spectrometer with 3-trimethylsilyl 2,2,4,4-D4)propionate as reference. The HDO peak was suppressed by means of presaturation. The reaction between 2 and deoxyguanosine was followedusing the HPLC-basedmethod described by Gonnet et al. [31 ]. The samples withdrawn at different time intervals were quenchedby addition of KC1 in excess and by cooling down to liquid nitrogen temperature. The pH was kept within 4.5+_0.1 by addition of HC104. The HPLC analysis was performed using a Beckman 126 pump coupled to a Beckman diode array detector 168and a System Gold V810 integrator. The system was connected to a Rheodyn 7125 valve.   cation exchange HPLC column Nucleosil SA, 250 x 4.6 mm, ID 5 mm  Colochrom, France) was employed, the mobile phase wassodium NaC104 0.25 M  pH 4.4 adjusted by HC104) for 15 minutes and a gradient to 0.5 M for 30 minutes, flow rate mL/min; column temperature 50C. The detection wavelength of 258 nm was that of the quasi-isosbestic point of theoverall reaction. The elttion times increased with increasing positive charge of the eluted species, i.e., deoxyguanosine < cis-[PtCl NH3)z dGuo)]   6 < cis- [Pt NH3)z dGuo)2] 2 . For the kinetic .nalysisof thereversible hydrolysisof 6, the same reaction was carried out with 2 in twofold excess over deoxyguanosine, so that theyield of ti was maximized. After onehour reaction time at room temperature, the same volume of saturated NaC1 solution was added to thereaction mixture in order to replace all aqua ligands with chloride. The mixture was subjected to cation exchange HPLC separationusing the same conditions as described above, and6 collected at the outlet was immediately cooled to liquid nitrogen temperature. This solution, which was 0.5 M in NaC104, was subsequently used to follow theestablishment of the hydrolysisequilibrium  Eq. 2 . The reaction was started by diluting 5 times with water, warming up to 20   and adjusting the pH to 4.5+0.1 by addition of HC104. The pH was kept within this range by eventual addition of NaOH. The total content in deoxyguanosine was quantified spectrophotometrically by measuring the absorbance at the .quasi-isosbestic point  258 nm), withthe molarabsorption coefficient e258 determined as 12300 M lcm   from a deoxyguanosine solution of a known concentration. The kinetics of the reversible hydrolysisof 6 to 5 was followed by withdrawing aliquots and analyzing them immediately using the same cation exchange HPLC system. Six independentexperiments were carried out. An exactlydetermined quantityof NaC1  ~1 equiv, with respect to 6) was added to the solution of 6 at the beginning of two experiments. 3. Results 3.1. Kinetics of the reaction between 2 and Me-5 -dGMP The reaction scheme is depicted in Scheme 1. The guanine H8 resonancesof the species 3  8.30<5H8<8.57; see Figure 1 and7  5H8 8.42 ppm) are downfield from that due to the free ligand  5H8 8.08 ppm); integration of the H8 peaks could therefore be used for concentration measurements. The chemical shift of H8 3), which is sensitive to the deprotonation of the aqua ligand  PKa3 7.03+_0.06 in D20 as well as to that of the guanine N1 atom  PKa N1) 9.16+_0.06 in D20  , has been utilizedas an internal pD indicator  Figure 1 s. We have followed the reactions between2and Meo5 -dGMP  L) in two different runs: i in acidicmilieu, where thereaction along kLl was preponderant,and ii in neutral medium, where a major fraction of 2 was deprotonated to 2-D and thus the pathway involving kL2 was more important. The establishment of the protolyticequilibria is, of course, rapid, therefore, the integration of NMR peaks allows only the sum [3tot] [3] + [3-D] to be determined. The differential equations forthetime- derivatives of ILl, [3tot], and [7] are shown belowalong with that for [2tot], defined as [2to t] [2] + [2-D] + [2-2D] d[L]/dt =- kLl[2] + kLz[2-D] + kBIS[3] [L]  3a d[2tot]/dt =- kLl[2 + kLZ[2.D] [L  3b d[3tot]/dt  kLl[2] + kLz[2-D] kBIS[3] [L]  3c d[7]/dt kBis[3][L]  3d Numerical integration of these differential equations yields the theoretical concentration curves. Obviously, thecalculation of the time-derivatives  Eq. 3a-d) requires the concentrations of the specificprotolytic forms, [2], [2-D], and [3], to be defined. Thiscan be achieved using two different approaches, one applicable in the acidic solution, and theother in theneutral milieu, as explained in the followingparagraphs.  Vol. 6, No. 1, 1999 A 5 -Phsphdiester Group Attached To Deoxyguanosine Does Not Accelerate the Hydrolysis of cis-[PtCl NH3)2 dGuo] + Conversionof these pK a values obtained in D20 solution to H20 according to Martin [32]yields PKa3 kL2 L D20 pKal 5.90 a PKa2 7.77 a PKa3 =7.03 b 2-2D Scheme 1. Reaction betweencis-[Pt ND3)2 D20)2] 2 2) and Me-5 -dGMP  L) in D20. a extrapolated for D20 from [32] and [33]. b determined from the titration curve in Figure 1. 5 H8) [ppm] 8,608,55 8,50 8,458,40 8,35 8,30 4 5 6 7 8 910 11 12 pD Figure 1. The H8 chemical shift of cis-[Pt ND3)2 D20) Me-5 -dGMP)] +  3) in D20 asafunction of pD. The fullline rep.resents a double-sigmoidal function 5 H8) [ppm] p + q[10 s pD)/ 1 + 10 s pD))] + r[10 t pD)/ 1 + 10 t pD))], with p 8.301 ppm, q 0.152 ppm, r 0.110 ppm, s 9.158,7.028. p, p+q, and p q r represent the H8 chemical shifts of the unprotonated, monoprotonated, and doubly protonatedforms of 3, respectively; s and are the pKa values corresponding to the  de)protonationof the guanine N1 and of the aqua ligand,respectively. 3.1.1. Reaction in acidic medium When approximately stoichiometric, exactlydetermined quantities of 2and L were mixed in an  R tube so that the initial concentrations were --4 mM, the pD of thereaction mixture remained within  Tiphaine Weber et al. Metal-Based Drugs the limits 5.3<pD<5.6 in the courseof the reaction. In these conditions, the protolytic equilibrium Ka2  Scheme 1 couldbeneglected, and dissociation of 2  Kal and of 3  Ka3 couldbe considered as the only sources of D +. The reducedsystemof equations characterized by the rate constants kLl,kL2, kBi S, and the equilibrium constants Kal and Ka3 leads to a cubic equation for [2-D]: a[2-D] 3 + b[2-D] 2 +c[2-D] + d 0  4 with a Ka3 Kal b Ka3[3tot] + Kal Ka3 + [2tot] Kal c Ka [2tot]  2Kal Ka3 d Kal2[2tot] 2 and [2tot] [2] + [2-D] [3tot] [3] + [3-D] [2-2D] 0 From thethree theoretical roots ofEq. 4 [34], the physically meaningful one was found to be that shown in Eq. 5 : [2-D] =-2Ql/2cos[ 0+4r /3] b/3a  5 with Q [ b/a 2 -3c/a]/9 0 arccos R/Q 3/2 R [2 b/a 3 9bc/a 2 + 27d/a]/54 From the balance for [D+], we have [D +] Kal [2tot]- [2-D] /[2-D] [2] [2-D][D+]/Ka [3-D] [D+]-[2-D] [3] [3tot] [3-D] For any given set of instantaneousconcentrations [L], [2tot],[3tot], and [7], we can thus express their time-derivatives  Eq. 3a-d by means of the rate constants kLl,kL2, kBi S, the equilibrium constants Kal and Ka3, and theconcentrations [L], [2tot],[3tot], and [7] themselves. 3.1.2. Reaction in neutral milieu In a secondexperiment,approximately stoichiometric, exactly determined quantities of 2and L were mixed in an N R tubewith an equimolar amount of NaOD so that the initial concentrations were -.8 mM The initial pD was 6.7 and increased to 7.7 in thecourse of the reaction. In this case,neither equilibrium of Scheme could beneglected, and a purely analytical determinationof [D/], [2], [2-D], and [3] became impossible. We then took advantage of the fact that the pD range lie in the buffer zone of 3  pKa3 7.03 in D20 , and could thus be monitoredusing the chemical shift of H8 3 Figure 1 withprecision.   fit function of the form [D +] p + q exp -r 1/2 6 t is the time in seconds , withthe coefficients  optimizedusing the program Kaleidagraph p 2.2968x 10 -8 M q 1.7583 M and r 0.014189 s /2, was found to describe very wellthe experimental [D+]-time curve  Figure 2 . The fit function  6 enabled us to convert the experimental [D+]-time curve into an analytical expression for [D+], from which the remaining concentrations required for the differential equations  3a-d could easily be determined: [2-D] [2tot]/ [D+]/Kal + Kaz/[D +] + 1 [2] [2-D][D+]/Ka [3] [D+][3tot]/ Ka3 + [D+] 3.1.3. Evaluation of the platination rate constants While in acidic medium L reacts almost exclusively with the diaquaform 2  pathway along kLl , in neutral medium 2 converts preponderantly to the aqua-hydroxo form 2-D, and the reactions of the latter  along kL2 become important. In order to obtain precise and consistent values for kLl, kL2, and kBl S, the experimental curves for [L] and [3tot] were fitted simultaneously for both reactions  Figure 3 .  Since [7] is related to [L]and [3tot] via the constant sum of all the H8 peak integrals,fitting its curve would be redundant. The optimized rate constantsare given in Tabletogether withthose determined forthe platination of 2 with deoxyguanosine  Section 3.3 .
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks