A” ‘_. w . >- “‘1‘?"- ..»V.. “A..- Vq’-.~,...--.v. . .. .. .. m , s. .. 9.0.1 K 5.. 3% . .1... D" i. 00 N RA AAA ORDINA" ION N“? 613. H). 100 THE I. .w c loner: ;E DIAMMINEPLATINU-M ( 4 5 8 WV? AND ITS” COMPO a . NA WITH ‘ Thesis for the.” . 5A 'sm-x vgflPh i rg‘e UN! .8 De MICHIGAN STATE ‘ MANSY ~ 1R REBEL-{Ml}? ALY S 1972' ; f... '22, . . . fulfih. .._.. $2.... I . ‘ . , . . .. V. .I . . . . . . , A / .. . . y . . . . . .1 n .1 . . a . . 1L. . . AA . A A . ..y . . fi‘r. . f2 . . ...r. . A v... . . . .T . ;. .a 1,. .A V I .: . LIBRARY 5 Michigan State Univwny This is to certify that the V thesis entitled THE INTERACTION OF THE CIS— AND I§A§§7DICHLORODIAMMINE— PLATINUM(II) COORDINATION COMPLEXES WITH DNA AND ITS COMPONENTS IN VITRO 2 presented by Sill: Abdel-Lntif Aly Mansy has been accepted towards fulfillment of the requirements for Phi D. gjkgmwin__§122hlfiifis 30,me ”r" , MCDAC. 1491974 0-7639 mac-:mq ABSTRACT THE INTERACTION OF THE CIS- AND I§Afl§7DICHLORO- DIAMMINEPLATINUM Pt1-—————53> Pt2 (l) where Pto is the initial platinum species; Pt1 species which is the active species for DNA inhibition but less so for is a slowly increasingly RNA and protein inhibition; Ptz, which is the equilibrium with Pt is 1’ responsible for RNA and protein synthesis inhibition. Their suggestion for the slow formation rate of Pt1, and the low equilibrium constant for Pt2 production, would eXplain the selective synthesis inhibition 10 of DNA, and the slower rate of RNA and protein synthesis inhibition. Renshaw et al.3 found that an appreciable amount of the labelled platinum compound was bound to the protein as well as to the nucleic acids. Table 1 shows almost equal distribution of platinum complex compounds between the nucleic acids and cytoplasmic proteins. The selectivity of one or more species produced from the anti tumor platinum complexes for DNA inhibition raises some important questions. What ligand changes occur in this species that the other species lack, and is the selectivity due to these changes? Does the geometry and the stereochemistry of this species determine its interaction with DNA, RNA. and proteins? Are these changed ligands more kinetically active? The kinetics of the substitution of the ligand groups of Pt(II)(NH3)2C12 with different ligands in aqueous solution will shed some light on the reactivity of the platinum species towards the reactive groups of the nucleic acids and the proteins. Substitution reactions of square planar complexes are best explained in terms of a biomolecular displacement mechanism.27’28 "20 11+ .. m3x“+ + Y ——> mg + x (2) A two-term rate law is followed: Rate-kllMA3Xfi] + kleA3Xn+l [Y’] (3) k1 - first-order rate constant k2 - second-order rate constant 11 For the case of an excess of Y, a pseudo - first order condition is obtained: k - k + k (4) A plot of k0 3 versus [Y] 13 linear with an intercept of k1 for the b reagent independent path, and a slape of k for the reagent path (see 2 below). Basolo et al.29 studied the rate of substitution of the chlorines of gig: and 55225: dichlorodiammineplatinum(II) in aqueous solution with a variety of nucleOphilic reagents. QigrPt(NH3)2C12 dissolved in water for some time, comes to an equilibrium within a few days in which 452 of the original coordinated chlorine is ionic as indicated in equations 5 and 6; while the Eggggrisomer + - cis-Pt(NH3)2C12 + H 0 -————=>- cis-Pt(NH3)2(H20) C1 + C1 (5) 2 cis-Pt(NH3)2C12 + OH -————=*- cis-Pt(NH3)2(OH)2 + 2Cl (6) on standing in solution for several weeks did not form any ionic chlo— ride. When the tgggg_complex was treated with alkali all of the chlo- rine was replaced. The rate of the reactions which are independent of the concentration of the entering reagent is insensitive to the net charge on the platinum complexes. gig_and‘£:ggg - Pt(NH3)2C12, Pt(NH3)3Cl+, gigrPth(NH3)2Cl+, and PtCl:-, all react at the same rate within a factor of about three. Thus, the 8N1 mechanism is not compa- tible with the coulombic effects. A square planar complex will have two further groups coordinated above and below the plane at a greater distance than the four primary ligands.3o-.33 These groups may be solvent 12 or solute molecules . MLA2X+Y————>MLA2Y+ X (7) The substitution may occur via two different pathways. Path I Patu II (Solvent dependent) (Solvent independent) 5 M \ A/?\x “/E x s 5 +x -x slow +x1 -x slow 11 S 5 Y ~e—-— / AL’/’ "\.5 ‘A;/' 3 “e S A» S l s 5 \fi/ \ :/ a .M 5 (products) 5 13 This scheme explains the stereospecificity of the reactions, the effect of the s-bonding group on those trans to it, the reactivity and second order kinetics of n-bonding reagents, the small influence of electrosta- tic factors on the rates, and the second order kinetics for some non- bonding reagents such as NH3. Both bond making and bond breaking are of comparable importance in the reaction of Pt(II)(NH3)2Cl2 with nucleo- philic agents. Basolo et al.29 found that the rate of release of Cl- from the interaction of transmisomer with aniline or pyridine (see equations 8-11) was approximately twice that of the hydroxide ion or glycine. k1 -————‘. + _ Pt(NH3)2C12 + H20 ‘____ Pt(NH3)2(H20)C1 + Cl (8) k2 k Pt(NH) (H 0)Cl+ + Y 3 P (NH ) YC1+ + H 0 9 3 2 2 ’ t 3 2 2 ( ) kt. Pt(NH ) YCl+ H 0 'T“—“‘ Pt(NH ) Y(H 0)++ + Cl- (10) 3 2 2 ‘= ' ' 3 2 2 k 5 k6 ++ ++ P NH + - P H + H t( 3)2Y(H20) Y ————=> t(N 3)2Y2 20 (11) Thus the kinetics of Pt(II)(NH3)2C12 is in 8N2 (lim) caterorv, and the rates show the behavior of displacement mechanism. The nature and con- centration of the entering groups exert a major effect on the rates of the reaction. According to the combined<3- and n-trans effect theory, which suggests that good trans activators are strongly bonded to the metal, the leaving groups, which are high in the trans effect series, are slowly replaced. Therpoor trans effect groups, NH and 0H’, are 3 very difficult to replace in Pt(II) complex,34 since the amount of 14 covalent bonding into the as orbital must be small for electronegative ligands because of the high energy of the up orbital of platinum. The bond strength is determined mainly by as and op orbitals, and by ionic interactions. The nucleOphilic agent is governed by its polarizability. The order of nucleOphilic reactivity of agents to Pt(II) had been esta- blished as follows:29’35 thiourea> I") 8011‘) N3) N02) Py> aniline> olefin) NH? 01‘) 1129011- Thus the species that could be formed from Pt(II)(NH3)2Cl2 in aqueous solution would be Pt(II)(NH3)2(H20)+C1 and/or Pt(II)(NH3)2(H20);+ and both will interact with DNA, RNA and more strongly with proteins, according to the nucleOphilic reactivity scheme above. Proteins will be a stronger nucleOphilic agent than DNA and RNA due to the presence of the sulphur bonds. The hypothesis of Howle et al.26 and Harder et al.22 of two kinds of Pt species, one of which interacts with DNA, RNA and protein generally, and the other with DNA specifically, does not agree with the chemistry of the square planar platinum(II) complexes. Inhibition of DNA synthesis could be due to the intra- and or inter- strand linkage with gingt(II)(NH3)2C12. The cross linking between the two strands of DNA with gigrPt(II)(NH3)2C12 was investigated, and compared with the classical antitumor, bifunctional, alkyalting agents.36 Roberts et al.37 develOped a new method for demon- strating interstrand cross-links in DNA (hela cells) treated with mustard 15 gas. One strand of DNA bore a density label "heavy" and other strand was "light". The cross linking between these two strands produced a "hybrid" species. This hybrid species contains equal amounts of heavy and light DNA which will show a peak in the center between the heavy and the light single strands in 03012 gradient centrifugation. To calculate the percentage of cross-linked DNA, the heavy stranded DNA was radioactively labeled and the hybrid was assayed as a percentage of the heavy DNA peak. The cross-linking of DNA was similarly observed treated cells, and it is approximately of the 2 same order as of the mustard gas treated cells. It is worth noting that in Ola-Pt(II)(NH3)ZCI Roberts et al.36 did not observe cross—linking in the trans-Pt(II)(NH3)2C12 0 treated cells, which may indicate that the distance 3.3A between the two active groups of the cis-isomer plays a role in cross-linking the two strands. The reacgive centers in DNA Nucleic acids are polymeric molecules which are built up by the extended repetition of a few fundamental units called nucleotides. Each nucleotide is composed of nitrogenous purine or pyrimidine base, a sugar residue and a phosphate group. There are two kinds of nucleic acids, ribonucleic acid (RNA), whose sugar is D—ribose, and deoxyribonucleic acid (DNA) whose sugar is 2-deoxy-D—ribose. Both nucleic acids contain two purines and two pyrim- idines. The two purines, adenine and guanine and the pyrimidine cyto- sine are common to both types of acids. RNA contains the pyrimidine uracil and DNA contains the pyrimidine, thymine. The chemical preper- ties of these heterogeneous bases depend on the distribution of the 16 electronic indices (electrical charges of the n-electrons, bond orders and free valencies). The MO calculations indicate that all the carbon atoms are positively charged and since the platinum complex compounds used in this research are electrOphilic agents, we will not discuss the properties of the carbon atoms. The nitrogen atoms of the purine and pyrimidine bases can be classified into three groups. 1. Ring nitrogens which don't have a hydrogen atom attached (pyridine-type) 2, Ring nitrogens which have a hydrogen atom attached (pyrrole- type) 3. Nitrogen of extracyclic amino groups. I. Nitrogens of the pyridine type: The nitrogen atoms in heterocylic compounds contribute one n-elec— tron to the pool of the mobile electrons, and, due to the conjugation in the molecules, they carry an excess of electrons. They are negatively charged and thus they are the centers of reactivity of the purines and pyrimidines in the case of electrOphilic agents attach. Protonation and alkylation of the purine and pyrimidine bases occur at the nitrogens38-40 of the pyridine type. Consequently we may expect the electrOphilic platinum species (hydrolysed products from Pt(II)(NH3)2Cl2 to combine with the purine and pyrimidine bases at the ‘1 tried to find a unportonated ring nitrogen. Pullman and Pullman relation between the electronic charge of the ring nitrogen and its basicity, but they met many difficulties. In 1958, by using the self- consistent field molecular-orbital method, Nakajina and Pullman"2 showed that in polynitrogen heterocylics, the electronic charge of a ring nitrogen cannot be taken as a measure of its basicity. Protonation and other reactions involving the lone-pair electrons of the heterocylic 17 nitrogen depend not only on electronic charge of the nitrogens but also on the values of the coulombic integrals between the "lone-pair" elec- trons of this atom and the other n-electrons of the molecule. pKa = B + I Q (ll/pp) lip p B - constant characteristic of a family of substances. Q - the net charge of atom p P (ll/pp)- coulomb integral between an electron l of the lone pair of nitrogen and the electrons of atom p B, is obtained by a semi-empirical method, comparing the experimental and theoretical values of the basicities of a series of related compounds. Calculating the coulombic integral of every site, we can predict the most basic site for interaction with the electrophilic agents as illustrated in Table 5. The greater the absolute values of the summation of the coulombic integral, the greater the pKa, and, as indicated in Table 5, N of adenine is of the highest basicity and the electrOphilic attack 1 should occur at this site; while for guanosine, N is of the highest 7 basicity for this interaction. These theoretical results agree with 44-48 the exPerimental data for the cases of protonation and alkylation. Pyrimidine, (VIl)49’50 shows weak basicity (pKa - 1.31) while pyridine, (VI) shows a higher value (pKa - 5.2). The insertion of the electron- attracting second nuclear nitrogen atom in the pyridine ring depletes the n-electrons from the first nitrogen atom thus decreasing its basic- 51,52 ity. Pyrimidine could be correlated to B-nitrOpyridine, (V), (pKa - 0.8), as a nitro-group is a strong electron attracting group. 18 Table 5. The theoretical values of the coulombic integral of the dif- ferent sites of the purine and pyrimidine bases 2 Qp(11/pp) (ev) 1gp pKa Compound -51 £3 E? Experimental Purine -1.52 -0.97 -0.32 2.4 Adenine -l.9l -l.72 —0.42 4.2 Guanine -l.ll -1&§l_ 3.3 Hypoxanthine -0.49 -1;12. 2.0 Xanthine -_(_)__._§_2_ 0 . 8 Cytosine —2 25 Figure 2. The basicity of the pyridine derivatives. VI, pyridine; VII, pyrimidine; VIII, B—nitro-pyridine; IX, 4-methylpyrimidine; X, 4,6—dimethy1pyrimidine, XI, 4-methoxy-pyrimidine; XII, S-hydro- xypyrimidine. 19 v: \m vm cu3 ca: 8‘3 o" / / u/ "0 1 . 1;) (I IX x XI XII Figure 2 20 Insertion of electron-donating groups, e.g. methyl, methoxy or hydroxy-groups, will increase the charge layer on the nuclear nitrogen 53 atom and the basicity increases. 4-Methy1pyrimidine, Ix. has a pKa - 2.0; 4,6-dimethylpyrimidine,54 X pKa - 2.8; 4-methoxypyrimidine,55 XI has a pKa - 2.5; and 5-hydroxypyrimidine,56 XII has a pKa = 1.87. 2. Ring nitrogens with a hydrogen atom attached: A very interesting phenomenon appears in the case of hydroxy pyrim- idines. An increase or decrease of the basicity of the nuclear nitro- gen depends on the position of the hydroxyl-group in the ring. Figure 3 shows the change of the pKa of the nuclear nitrogen of the ring. In aqueous solution 4-HydroxypyrimidineS7 XIII and 2-hydroxvpvrimidine58 XIV exist as cyclic amide tautomers, and the strength of a ring nitrogen atom is considerably weakened when it becomes part of such a cyclic amide. The enhanced basic strength of the hydroxypyrimidines is due to the nitrogen atom that is not involved as amide. When both nitrogen atoms are invol- ved e.g. uracil XV and thymine XVI, the basic strength draps to zero. Insertion of an amino-group at different positions shows a great change of pKa's of the nuclear nitrogen. The pKa of S-aminOpyrimidine59 is 2.8, which is of the expected order compared to a pKa - 1.31 of the 49 unsubstituted pyrimidine. A large increase of the pKa's of 2- and 4- 49 (3.54 and 5.71) is due to the presence of the amino aminOpyrimidine groups which allow more resonance in the cations, as shown in the (next page The 4-amin0pyrimidine shows a higher pKa than that of the 2-aminopyrimi— dine due to the preferred p-quinonoid form that contributes to its cationic hybrid than that of the o-quinonoid form of the 2-isomer. The basic strength of the substituted pyrimidine depends upon the distribu— tion of the double bonds in the ring. Figure 3. The effect of the hydroxy-group on the basicity of the pyrimidine ring. XIII, 4-hydroxypyrimidine; XIV, 2—hydroxypyrimidine; XV, Uracil; XVI, thymine. 21 155.13 H”? XIII XIV O O /"\ /u c": "" 1 "J: . a»): . :. xv xvu Figure 3 I'll: 1/\\ .{* I -———I; II 22 I :z+ I: N J / :2 an: Redistribution of the double bond by substitution in the ring removes the base-strengthening resonance. 4-Amino—1,2-dihydro-l- methyl-2-0x0pyrimidine (l-Methyl cytosine)6o has a pKa a 4.57 which is below that of 4-amino—pyrimidine. The basic strength is greatly diminished in the case of 4-amino-, 1,6 dihydrO-l-methyl 6-ox0pyrimi~ dine51 lI/l‘ Ofl;\aa . 4..., (pKa - 4.57) (pKa - 0.98) the cause of which is not understood. NH, A 1,1», Cit: (pKa - 0.98) 23 Uracil and thymine have very low pRa's while the basicity of N3 of the cytosine is of the order of adenine, as the theoretical calculation predicts. In a polynucleotide (doubly stranded), like DNA, the involvement of some of the ring nitrogens in hydrogen bonding decreases the value of the theoretical predictions. N1 of adenine is hydrogen bonded with N1 of thymine, while the N1 and N of adenine are available for chemical 7 reactions. N1 of cytosine is hydrogen bonded with N consequently N 1 of guanine and of cytosine is blocked, and N of guanine is available 1 7 for interaction with electrOphilic agents. Alkylation of DNA and RNA lead to the alkylation of the guanosine61 at N7. 3. Nitrogen of extracyclic amino groups: I 3'"? t ' NH, '1: \n \> "5'; MK) MN. N T H2N A... OAN/ a I It Adenosine Guanosine Cytosine H fill/5 Isocytosine 24 The amino groups of adenosine and cytosine react readily with alkyl groups forming N-alkyl derivatives.62 Guanine and isocytosine do not so react. Freenkel-Conrat63 showed that formaldehyde (HCHO) is a specific reagent for attacking the amino groups of the purine and pyrimidine bases. In studying the interaction of formaldehyde with DNA and RNA, Staehelin64 showed that RNA was much more reactive than DNA, and that the reactions occurs with the amino groups that are not engaged in hydro- gen bonds. The reaction of formaldehydewith the free amino groups forms a Schiff's base63 and/or secondary condensation64 with other active hydrogen atoms. HCHO + HzN-Base _____a. CHi=N-Base (l3) HCHO + 2H2N-Base ____€,. Base-g-CHz-fi-Base (14) The reaction of formaldehyde with two free amine bases corresponds to the formation of more complex bridges by bifunctional alkylating agents of the nitrogen mustard type.65 The reactivity of adenine, cytosine, and the inertness of guanine and isocytosine in the reaction with HCHO or any alkylating agents could be interpreted in terms of the electronic characteristics of the amino groups. Table 6 summarizes the ebctronic characters of the amino groups. Among the purines and pyrimidines of nucleic acids we find that the N atom of the NH2 group of adenine has the highest electroniccharge and the greatest free valence. We expect that this base is the most reactive one in the reaction with electrOphilic reagents. Guanine and cytosine have the same order of electronic charge on the NH2 group, but the free valence of cytosine is higher than that of guanine and less than that of 25 Table 6. The electronic characteristics of the amino groups of the purine and pyrimidine bases N. atom of the amino group Compound ghgggg_ Free Valence Adenine 1.810 0.960 Guanine 1.803 0.933 Cytosine 1.803 0.944 Isocytosine 1.800 0.925 2,6—Diaminopurine N2 - 1.829 0.978 N6 - 1.810 0.959 Adenine (in AT pair) 1.763 0.908 Cytosine (in CC pair with 2-H bonds) 1.751 0.828 Guanine (in CC pair with 2-H bonds) 1.809 0.943 Guanine (in SC pair with 3-H bonds) 1.773 0.906 Cytosine (in CC pair with 3-H bonds) 1.751 0.828 26 adenine, which explains why the reactivity of the amino group of cyto- sine is less than that of adenine and of guanine, which is inert. Adenine Cytosine Guanine reactive less reactive inert 0.960 0.944 0.933 The order of decreasing reactivity of the amino groups of the bases is shown above. The lowest value of a free base, which is guanine is 0.933. These bases, however, form hydrogen bonds with their complimen- tary bases and the electronic charges and free valencies change. Thus the electronic charge of the amino group of the adenine in an A-T pair decreases to 1.763 and the free valence to 0.908. Cytosine, in a GC pair with two hydrogen bonds, has a free valence 0.828. The amino group of the guanine in the pair has a free valence of 0.943. Guanine Adenine Cytosine (G-C pair) (A—T pair) (G-C pair) 2-H bonds 2-H bonds 0.943 0.908 0.828 Adenine Guanine Cytosine (A-T pair) (G-C pair) (G—C pair) 3—H bonds 3-H bonds 0.908 0.906 0.828 27 From the free valence calculation one would expect that in native DNA the amino-groups of the purines and pyrimidines will be inert for any reaction. In partially denatured DNA, the amino group of the guanine will have the highest free valence and a reaction at this amino group will occur. When the hydrogen bonds of an adenine-thymine pair are broken, the free amino group of adenine will have a high free valence for interaction. Also the free amino group of cytosine will be avail- able for interaction with electrOphilic reagents. Interaction of Metal ions with Nucleic acids 0n polynucleotides, there are two main sites of interaction with metal ions, the phosphate moieties of the ribose phosphate backbone 67’68 The reaction with 69-73 a and the elctron donor groups on the bases. phosphate leads to stabilization of the ordered structure nd consequently, an increase of the melting temperature (Tm) with increase of metal concentration. Reaction with the bases destabilizes the ordered structures, which leads to denaturation and a decrease67 of Tm. Using the phenomenon of variation of Tm with varying concentration of metal ions, the interaction can be divided into two classes. Most mono- valent ions and some divalent cations bind to the negatively charged 2+ 2+), sugar phosphate backbone, (Na+, Mn , and Co while metal ions like 2+ Cd , Cu2+ and Pb2+ bind preferentially to the heterocylic bases. Bind- ing of metal ions on polynucleotides is not an all or nothing interaction.7 Mg(II)> Co(II)> Ni(II)> Mn(II> Zn(II> Cd(1V)> Cu(II> Ag(I)> HgIII) The metal ion Mg(II) at the left of the series above increases the Tm of 28 DNA and cannot induce any renaturation of the double helix. Co, Ni, and Mn cause partial denaturation, while Cu(II) at the right of the series produces almost complete renaturation of the double helix. The affinity of the metal ions for the base to that of the phosphate increa— ses from left to right. Cu+2 which binds mainly to the nucleoside bases can bind also to the phosphate”.80 depending upon the ionic strength and the concentration of the free metal ions in solution. Davidson et al.“"82 studied the binding of Hg(II) by DNA and its components at an acid pH. Hg++ forms linear complexes of the type L—-Hg--L, where L is the ligand (nucleoside base) in the presence of excess of ligand. With excess of Hg++ there is a 1:1 complex. Simpson83 determined the site of mercury binding at various pH values using UV difference spectra technqiue. The interaction of the CHBHgOH or CH3Hg+ with the nitrogen site depends on the pH, which determines which species predominates. Also the pKa of the nitrogen site determines whether the site is protonated or not e.g. atiH2KIH21._ CH3Hg+ + NH ;—_.—‘* CHBHgN + 11* (17) the reaction is favored as the pH increases. at 3K LN) (pH >11KIH2) 011311; + N {—3 CH3HgN+ (13) the reaction is pH independent. As the different nitrogen sites of the nucleosides were blocked by a proton (at different pH's) or by a methyl group (nucleoside derivative) Simpson was able to determine the site of interaction and its formation constants with CH HgOH and Hg(OH)2. He found that methyl mercuric ions 3 bind at N7 of guanosine and inosine. Mercuration occurs at N1 of aden- osine and N1 of cytidine. X-ray diffraction study84 showed that copper chloride reacts with cytosine at N forming 1:2 complex. Methyl mer- l curic ions reacted with the amino groups of adenine, cytidine and guano- sine at pH 10-11. The binding of the Hg-ions at ”If which is protonated, of uridine does not agree with the x-ray diffraction study of Uracil- mercuric chloride complexes by Sundaralingam.85 X-ray studies show that the binding is at 0 and the combination is with HgCl rather than with A 2 ng+. Nuclear magnetic resonance studies of metal ions bound with nucleo- sides, nucleotides and polynucleotides show that the site of binding of the metal to the purine or pyrimidine bases changes with the location 30 of the phosphate group on the sugar ring. Capper binds preferentially to adenosine rather than to the ribose. The unchanged sharp resonance peak of the 6-amino group of adenosine in the presence86 of Cu(II), and the broadening of the H-8 and H-2 protons of the adenine ring may indicate that Cu(II) does not bind to the 6-amino-group and that binding is at the nitrogen atoms neighboring these positions (N7, N or N3). 1 Addition of Cu(II) to 5-AMP broadens preferentially the H-8 proton resonance than H-2, indicating that binding at N is favored to N of 7 l 87 The effect was observed in the interaction of Cu(II) with the ring. 3' -AMP. The addition of Cu(II) to 2' -AMP broadens the H-8 and H-2 peaks simultaneously. The lower H-l' (ribose proton) resonance broaden- ing of 5' - and 3' -AMP and the broadening of H-l' resonance along with the base protons of 2' -AMP,, indicates binding at N , and Cu(II) may 1 form a chelate structure between the phosphate group at the 2' - posi— tion and N3. In poly A, H-8 broadening was maximal and the interaction was at the N7 position. In 5' -IMP, the H—8 and H-2 resonance were broadened simultaneously,88 indicating the binding of Cu(II) near both protons, at N7 and 06 or at N7 and N1. In poly I, Cu(II) produced preferential broadening of H-8 rather than H-2 proton resonance. Thus binding is also at N7 as in poly A. Binding of Cu(II) to poly C is at the N1 posi- tion88 as 131the monomer.86 Cu(II) coordinates to the ribose hydroxyls of uridine, UMP, and poly U, and as these are being occupied. binding will be on the uracil ring in the vicinity of N1 and C5° Poly A, po1y u did not show any interaction with Cu(II) at room temperature. As melt- ing occurs in the presence of Cu(II), H-8 proton resonance of poly A was mostly broadened; H-2 was less broadened, and H—5 of poly U was the 31 least effected. Cu(II) binds to N of poly A and N to a lesser extent, 7 l and the least to N1 of poly U. Approach to the problem. It is believed that the active antitumor coordination complexes of platinum exert their biological effect by producing a primary lesion in cellular DNA. It is now necessary to investigate the nature of the interaction between these complexes and DNA. Both interstrand cross- 1inking and intrastrand dimer formation have been suggested as the bio— logically significant lesion. We have attempted here to investigate these interactions, using both simple nucleosides and oligomers and polymers of nuclear bases, to determine all possible classes of reac- tions with the platinum coordination complexes. In order to assess the biologically significant interactions out of the total class of permissable reactions, we have used the simple fact that gigrdichlorodiammineplatinum(II) exhibits marked antitumor activity against a variety of tumor systems, while its isomer, trans-Pt(II)(NH3)2Cl2 is inactive. Therefore, only those reactions occurring with the gig complex--but not with the £ggp§_complex will be significant. This assumes that the trgpg complex has the same fate in an animal system as the gi§_complex up to the point of DNA attach. While this is not yet completely proven, there are indications (Hoeschele and Van Camp, personal communication)that this is so. II. EXPERIMENTAL All materials were obtained from Sigma Biochemicals and MILES LABORATORIES were of highest purity and were used without further puri- fication. Monomers Adenosine, sigma grade; l-methyladenosine; cytidine (pfs); 3-Iethy1cytidine methosulfate(l-ribofuranosvl-J-metnvlcvtosine) 3-methylcytidine will be named l-methylcytidine in this thesis to agree with the numbering of the purine ring; thymidine (pfs), sigma grade; uridine (pfs), sigma grade; guanosine (pfs), sigma grade; l-methylguano- sine (pfs); 7-methylguanosine, grade II: 90-95%, crystalline; inosine (pfs), sigma grade; l-methylinosine; 7-methylinosine (pfs), grade II: 85-95%, were obtained from Sigma Biochemicals. Dimers Adenylyl (3' + 5') adenosine (ApA), free acid (pfs), adenylyl (2'., 5') cytidine (ApC), free acid (pfs), were also obtained from Sigma Biochemicals. Polymers The following synthetic polynucleotides (white fibrous), were obtained from MILES LABORATORIES. Polyadenylic acid (poly A), potassium; polycytidylic acid (poly C), potassium; polyguanylic acid (poly G), sodium; and polyuridylic acid (poly U). Ammonium , did not obtain olignucleotides, nucleotides, nucleo- sides and inorganic salts. 32 33 Polyadenylate-polyuridylate (poly A. poly U) had a base ratio A:U of 1:1.3, Tm 42.0°C, and hyperchromicity of 25.3). Polydeoxyadenylate- thymidylate, Na salt [poly d(A—T). poly d(A-T)] had a Tm 61.0°C, and hyperchromicity of 49.52. Deoxyribonucleic acids (DNA) were obtained from Sigma Biochemicals. Calfthymus DNA, type 1: sodium salt (pfs) G-C content is 422; E. Coli DNA, type VIII (pfs) G-C content is 502; Micrococcus Lysodeikticus DNA, type X (pfs), G-C content is 702. The g_i_s_-- and _t_r__a_n.s isomers of Pt(II)(NH3)2Cl2 were prepared in this laboratory by Dr. J. Hoeschele as described elsewhere (Lebedinskii and Korabel'nik, 1947; Kauffman, 1963); the purity was verified by their Laser-Raman spectra (Figure 4 )44. Each nucleoside in solution with various concentrations of Pt(II)(NH3)2C12, were kept in the dark at 37’C to avoid any possible photo-reactions of platinum complexes. The spectra of the solutions were measured at room temperature using a l-cm cell in a Cary 15 Spectrophotometer until no further change of O.D. occured (equilibrium was usually reached within two weeks). pH study One ml of lmJ‘of each nucleoside was diluted to 10 ml with 0.7, 0.07, and 0.001 M HClOa, and with 10-5, 10—4’ 10-2, 10"1 N NaOH to give -2. 10 M nucleoside at the different pH's. Cis- and trans-Pt(II)(NH3)2Cl2 were dissolved in 0.1 “M NaClOl‘ 3 produce a solution of 10- M. Addition of equal volumes of nucleoside to platinum complex allowed the reaction to proceed at various pH's. as shown on the next page: Figure 4. Raman spectra of cis- and trans-Pt(II)(NH3)2Cl The 2. spectra were recorded using solid samples in capillary tubes, a 100 on He-Ne laser and a spectrOphotometer, spex 1400 Double Mono- chromotor and ITT-FN-l30 "Star-tracker" photomultiplier, capable of less than 1 cm"1 resolution. 34 a shaman Tau >< q 00- N0. #1 q DON con ,2' s—r'r ~ g. --'~- .' on: *N» tun Woumzzzetg ‘— ALISNSLNI 35 Egg 2!. l 2 ml of nucleoside in 0.7 M HC104 + 2 m1 of Cis-, trans-Pt* 0.81 2 .. .. .. .. «0.07 n .. +" " " " " " 1.61 3 v " H " " 0.001 M " + " " " " " " 3.27 4 n n " " " 0.1 M Neelo4 + " " " ” " " 5.58 5 n n n v " 10‘5 N NaOH + " " " " " " 5-95 6 n n n H " 10" N " + " " " " " " 8.86 7 n n H H " 10-3 N " + " " " " " " 10.46 8 n n n H " 10'2 N " + " " " " " " 11.31 9 n H H H " 10'1 N " + " " " " " " 12.12 *Pt I Pt(II)(NH3)2C12 Determipagion of number of Ligands of nucleoside per platinum complex: The composition of the formed nucleoside-platinum complex usually one species, in solution was determined by the variation method, "Job's‘ method".90 For more than one species the compositions were determined by the Voshburgh and C00per teChnique,104 mettle: A mole fraction of 2 x 10-4M of each nucleoside, X, dissolved in 1, 2, 3, a, s, 6, 7, 8, and 9 m1 is diluted to 10 ml with 2 x 10‘“M Pt(II)(NH3)2012. Thus (M) + (L) - C, where (M), (L), C are the molar concentrations of Pt—complex, nucleoside complex and constant respec- tively. The absorption spectrum of the formed complex is maximal when the metal and the ligand are brought together in the same ratio in which they exist in the complex. In the case of more than one species, the absorption spectra of 1:1, 1:2, 1:3 mixtures of M:L were recorded. If there are three species 36 in solution, the wavelength at which 1:1 and 1:2 have the same extinc- tion coefficient would be suitable for identifying the ML species, and that at which 1:2 and 1:3 have the same extinction coefficient would be suitalbe for the ML2 species. _petermination of the sites of bindigggiand its formation constants: The site of the interaction of gig: and iigpngt(II)(NH3)2C12 with nucleosides was identified at different pH's. and with nucleosides methyl-substituted at different sites. The formation constants at these sites were measured using Simpson's technique.83 A series of spectra of nucleosides with varying concentrations of gig: and iigpngt(II)(NH3)2C12, at a certain pH where there is only one species to be expected were measured. The ratio of the platinum metal to the nucleoside varied from 0.1 to 8 in the cases of adenosine and l-methyl adenosine, and from 0.05 to 4 in the cases of guanosine, l-methyl guanosine and cytidine. The change in absorption at the wavelength of maximum change as measured by difference spectra using four l-cm cell, was plotted against log[Pt(II)(NH3)2C12] and a sigmoidal titration curve was fitted to the points. See next page for calculation of the formation constant. Calculation of (8) [ML] ' [L] ' [M] - [Lo] ["0] At the midpoint therefore 37 the formation constant: M + L.-—-“__ ML (19) f 2.1—. (20) [M] [L] concentration of the formed complex compound free ligand in solution free metal complex in solution total concentration of ligand in solution total concentration of metal complex in solution of the curve we have the following: [LM] - 1/2 [L0] - [L] (21) Kf " __[_L_]_.... ' .1... (22) [L] [M1 [M] [M] - [fix] - 1/2 [L0] (23) where [Mx] - [M0] corresponds to the midpoint of the curve K - 1 (24) [MK] - 1/2 [L0] 38 (b) M + 2L (:3. ML2 (25) [L12 [M] At the midpoint: ZIML2] - 1/2 L0 therefore [MLZ] - 1/4 L0 (27) [L] - L0 - 2(1/4 L0) - 1/2 L0 (28) [M] - Mx - 1/4 L0 (29) Kf _ 1/4 [L0] . i74 [L0] (30) T [1/2 L01 [8, - 1/4 L0] [1/4 L31 [MK] - l/4[Lol 1 (31) 7< I therefore [Lo] [Mx] - 1/4 [L0] 39 There are many sites on the mononucleoside where the gig: and ggggg platinum species may interact. Both gig: and ggggg: isomers can inter- act bifunctionally with the azanitrogen and the amino groups. Since the lgig: and ggggg: isomers have dramatically different biological effects, differences in the modes of binding of the isomers are of particular interest. The gig: isomer has an additional mode of binding, not avail- able to the ggggg isomer, by forming a bi-dentate chelation via N and 7 the 6—amino-group of adenosine; and the N and the 4-amino group of cy— 3 tidine. The evidence (see below) that the gig:platinum species can bind to the 6-amino group of adenosine and the 6-amino-group of cytidine, led to the investigation of the possibility that two adenosines or one adeno- sine and one cytidine or two cytidines moieties could be linked together by the gig:isomer. The inter—base separation in DNA is close to 3.4A, and as the distance between the two cis-ligands of square planar platinum compounds is about 3.3A, the linking of two bases by one platinum complex is stereochemically possible. The purine dimer A 3' p 5' A in solution has a separation between the 6-amino groups of 3 to 4 A. The ApA, ApU and the platinum isomers were dissolved in 0.1 M NaC104 at a molar ratio of 1:1 (dimer to Pt) and incubated in the dark at 37°C until no further change was seen. The circular dichrosim spectra were measured on a JASCO CD machine fitted with a thermostatted cell holder. The temperature could be varied over the range 5° to 45°C. Kinetics of polynucleotides and DNA - Platinum Compiex Igieraction One ml of DNA, (Calf Thymus, 382 CC; E. Coli, $02 GC; M. Lysodeik- ticus, 721 CC; 0.3mM p) in 0.1 M NaC104 was mixed with one m1 0.3mM cf freshly prepared cis-, and trans-Pt(II)(NH3)2Cl2 in a one cm cell fitted 40 into a thermostatted holder in the Cary 15 at 37°, 51.5° and 61.5°C. The rate of the reaction was measured by the difference spectra technique. Both the sample and reference compartment were held at the same tempera- ture. The reaction was followed with time until precipitation occurred. The kinetics of interaction of the single stranded poly A, poly U, poly G, poly C and the double stranded poly A. poly U, and d(A-T),d(A-T) with the platinum isomers were also followed at the same temperatures and under the same conditions as for DNA-platinum interactions, using the same technique. III. RESULTS The effect of theng on the binding,site of nucleosides to the cis- and trans-Pt(II)(NH3)29i2. The interaction between the isomers of Pt(II)(NH 3)2C12 and nucleo- sides, substituted nucleosides, synthetic polynucleotides and DNA was carried out in a medium of 0.1 M NaClOa until no further changes occur- red in the absorption spectra. The medium provides the high salt con- centration necessary to maintain the stability of the DNA. Perchlorate ions do not bind the metal complexes so the medium will not compete with the DNA for the platinum complexes. Aquation, with loss of chloride ions occur rapidly under these conditions, so that the reactions will occur via the aquated species. Figure 5 shows the hydrolysis of gig: and giggg:Pt(II)(NH3)2C12 in 0.1 M NaClO4 overnight. The pH of the solution was not controlled since a buffer may introduce potential ligands for the metal complexes. In those experiments the pH of the solutions fell in the range 5 to 7, except for the case of the pH dependent studies where the pH was adjusted by the addition of cone. HClO4 or NaOH and measured at the time the spectra was run. As thegig:Pt(II)(NH3)2Cl2 can interact either monofunctionally or bifunctionally, such as chelation with a neighboring group, while the ‘ggggg:isomer is a mono-functional agent (Figure 6), a comparison of the difference spectra of the two isomers will reveal the nature of binding with nucleosides, dinucleotides, and polynucleotides. 41 Figure 5. Absorption spectra of the Pt(II)(NH3)2Cl2 isomers. The spectra of 1 mM cis- and trans- isomers in 0.1 M NaClO4 using 5—cm. cell, were recorded on cary 15, a, freshly prepared solution; b, solution left overnight at 37°C. 42 m ouowam $1.5QO zofiamOmmq KNI '1‘ 2 R R (ii) Bl-FUNCTIONAL BINDING (gig) I“: HO—Pt—NH3 "(”3 NH, HZIO HN—FIH—NH, N’ ' N> __, N< | N> + 211,0 3N N N N I I R R Figure 6 ad th 511 Pee the he 44 (A) Monomers Adenosine Figure 7 shows the binding of the gig:Pt(II)(NH3)2C12, and trans-Pt(II)(NH3)2C12, with adenosine (ratio of Pt Metal/base = 10) at different pH's. At a pH around 1, the gigrisomer with adenosine shows a peak of A280nm which is similar to that of the giggg:isomer with adenosine at the same pH. This suggests that both isomers interact with adenosine monofunctionally at a low pH. At a pH 5-7, in addition to the peak of Azgonm, the gig:isomer exhibits a shoulder extending from Azaonm, to A350nm. The giggg:isomer shows a peak only at A nm. The peak at 2280nm could be assigned to the same 280 mono-functional interaction of the gig; and the giggg:isomers with adeno- sine as at the low pH. The appearance of the shoulder in the case of the gig:isomer only suggests the possibility of the chelation with the neighbor amino group. At a pH around 11 both isomers did not show any interaction with adenosine, and no precipitation was observed. The binding of both isomers with adenosine could be at N or N of l 7 the purine ring. l-Methyladenosine At a pH around 1, both the cis- and the trans-isomer show a weak binding, while at a pH around 3, the cis-isomer shows a broad band, its peak at Azgonm, which extends to A350nm (Figure 8), which could be due to the binding at N and chelation with the neighbor 6-NH2 group of the pur- 7 ine ring. The trans-isomer shows a sharp peak at Azaonm and therefore, Figure 7. Difference spectra of the Pt(II)(NH adenosine at various pH values. 8, pH - 11.31; 9, pH - 12.12. 3)2C12 isomers with 1-6, pH 8 0.8-8.86; 7, pH - 10.46; Dr‘ 45 n ouswum emeomem 855.115 458%. mozmmmeaa 05:4 ctS‘ 00m 000 Omm .00m 00m 000 . .IITIT..-. ..-.8-1. . . . .4.....J.8- ) .00- i~01 1 #0. 0.0 ._.0 so .m0 .00 .m_|o .eo mzqoe so .00 00 m _ .8 A??? mzaozge ioo Baku wIz .NAV mm< mm< Figure 8. Difference spectra of the Pt(II)(NH3)2Cl2 isomers with l—methyladenosine at pH - 3.27. 46 w ouowfim «Eomem wozwmmnna 0 mm< 49 Guanosine The cis- and trans-isomers did not show any interaction with guano— sine or l-methylguanosine at a low pH, around 1. At a pH 5-7, N of guanosine is mainly protonated and the binding 1 will be at N7. The cis-Pt(II)(NH Cl2 will give the same spectra as 3)2 the trans-isomer when the interaction is with N7, since there is no neighboring group to chelate with. The pH dependent spectra (FigurelO) show the interaction at A nm in the acidic region and the other band 290 at A nm in the alkaline region. Figure Ilshows the similarity between 265 the gig: and g5ggg:isomers complexed with guanosine, and the gig: and giggg:platinum complex with 1-methylguanosine, which have the same peak at A290nm. There was no interaction between either of the platinum iso- mers with 7-methy1gaunosine. Thus the interaction at pH 5.58 is at N7, and both gig: and giggg:isomers act monofunctionally. Guanosine and 1—methy1guanosine did not show any interaction with either the gig: or giggg:isomers at a pH above 9. The hydrolysis of 7-methylguanosine to guanosine in the alkaline region, and to 7-methylguanine in the acidic region, hindered the study of it with the platinum isomers at various pH's. Inosine The gig: and igggg:isomers gave similar peaks at 2262nm at a low pH, around 1 (Figure12). At a pH 5—7, both isomers show a broad band which has a peak at Xz7onm. l-methylinosine, with both isomers, showed a band at 2260“” at a pH around 3 (Figure I». Thus we can assign the band of l-methylinosine with the platinum isomers at 1260nm as due to the interaction at N7. The difference spectra (Figure 13) of the cis- and trans-isomers with 7-methylinosine show a band at A285nm which is Figure 10. Difference spectra of the Pt(II)(NH3)2C12 isomers with guanosine at various pH values. 1, pH 8 0.81; 2-4, pH - 1.61—5.58, the binding is at N7(1290nm); 7, pH = 10.46, the binding is at N1 only; 8-9, pH = 11.31-12.12. rt )0 CH senses Emz-_ $4 0 ~ 85 54 due to the interaction at N1. We can conclude that the interaction of both platinum isomers with inosine at a neutral pH would be at N1, but mainly at N7. At a high pH, above 9, inosine, with both platinum isomers, shows a band at A258nm which could be due to the interaction at N1. Uridine There were no interactions observed at any pH. Thymidine There were no interactions observed at any pH. (B) Dimers Adenylyl (g)..5') AdenosinejjA3'p5'A): At a pH around 1, the cis- and trans-platinum isomers interacting with A3'p5'A gave a similar peak at A nm, which could be correlated to 280 the one which appeared at A28Onm in the case of the monomers at the same pH (Figure 14). As the peak of the monomer was assigned to the interaction at N7 of the adenosine base, we can assign the peak of the dimer as due to the interaction at N7 of each base. At a pH 5-7 the cis-isomer gives a peak at A282nm and a shoulder which extends from A3oonm to A350nm. The trans-isomer with A3'p5'A shows only a peak at A nm which is the peak that was assigned for the inter- 282 action at N7. The difference spectra of the cis- and trans—platinum isomers with A3'p5'A are similar to that of both isomers with adenosine. ‘Ve can conclude that the interaction of the gig-Pt(II)(NH C12 is at N 3'2 iind chelating with the 6-NH2 group in the case of the dimer, as in the 7 unanomer. The trans-isomer interacts at N7 in the case of the dimer, also Figure 14. Difference spectra of the Pt(II)(NH3)2Cl2 isomers with A3'p5A at various pH values. 1—7, pH = 0.81-10.46; 8, pH = 11.31. 55 0 L no mm< Figure 20. Difference spectra of the Pt(II)(NH3)2C12, Pt(II), isomers with guanosine, L, "Job's method”. The ratio of the molar Pt(II) Pt(II) + L 0.5, 5; 0.6, 6; 0.7, 7; 0.8, 8; 0.9, 9. concentration of is 0.1, 1; 0.2, 2; 0.3, 3; 0.4, 4; 63 «#:0me mozwmwmma 2.5 a 8m 0m ouowwm 05* 0.4- ' A28 , $3 0.3“ 2: ' A285 0.2" 1390‘ OJ ’ / . 1295 ooL . .. . Pic 2 4 LIO 8 6 0.7 1 0.6 t 0.5 ’ e 1275 0.4 I AABS *“0 TRANS L/Pt=l 0.3 I ° 0.2 ' A285 0.I-/ - . 0.0 l_ L l 1 l l L l 1 PI 0 2 4 6 8 IO L l0 8 6 4 Figure 24 Figure 25. Graphical determination of composition of the Pt(II)(NH3)2C12 — l-methyladenosine complex. The straight lines are the relationships expected for PtL complex. The curved lines are the experimental results. 68 NH I—ME-ADENOSINE H,,C~NK I N> ‘N AABS Figure 25 Figure 26. Graphical determination of composition of the Pt(II)(NH3)2C1 - cytidine complex. The straight lines are the 2 relationships expected for PtL complex in case of the gig: isomer and for Pt2L3 in case of the trans: isomer. The curved lines are the experimental results. AA88 AA88 09 F CYTIDINE NJ} 04‘?! P R TRANS L/Pt= l.5 0 2 4 6 8 I0 6 Figure 26 Figure 27. Graphical determination of composition of the Pt(II)(NH3)2C12 - guanosine complex. The straight lines are the relationships eXpected for Pth complex. The curved lines are the experimental results. 70 0 Z 03_ GUANOSINE HNJ‘I > ' HZNkN If (D 02}. 0 1292 R 2 1295 CIS <1 0.! - / . 1&0 . L/Pt=2 0.0 l J— l 1 l l 1 l 1 Pt 0 2 4 6 8 :0 L IO 8 6 4 2 0 0.4 03 3’3 4 02 <1 0.: 0.0 Pt 0 Figure 27 Figure 28. Graphical determination of composition of the Pt(II)(NH3)2C12-l-methylguanosine complex. The straight lines are the relationships expected for PtL complex. The curved lines 2 are the experimental results. AABS AABS 71 0.20%- 0.5. 0.10- / TRANS 0.051» I. ””32 ().C’ 1 1 4L. J_ l 41 p, o f 4 6 8 IO L l0 8 5 ‘ 2 0 Figure 28 Figure 29. Graphical determination of composition of the Pt(II)(NH3)2C12 - inosine complex. The straight lines are the relationships expected for PtL2 complex. The curved lines are the experimental results. AABS 72 0 INOSI E ~ J N 0.3~ N ”'1 | \> N '3‘ 0265 R 0.2., A270 275 o... . 08. 23c, \ L/Pt=2 0.0L. ‘1 l 1 l 1 1\ Pt 0 4 6 8 l0 L l0 8 6 4 2 0 0°5F 0.4+ 8205 0.3- . 275 \ / 0.2r- TRAth 0" ' 1280 L/Pt= 2 (l‘>-' l J. 1 .4 l n\\ P? o 2 i 6 8 l0 L l0 8 6 4 2 0 Figure 29 Figure 30. Graphical determination of composition of the Pt(II)(NH3)ZCIZ-l-methylinosine complex. The straight lines are the relationships eXpected for PtL complex. The curved lines are 2 the experimental results. 0.3 ' AABS 73 O I-ME-INOSINE HacmK'IN.) kN N :5 0.4 ' 0.3 _ AABS 0.2 ' 0.| ” 0.0 “' Pt 0 L l0 2 4 6 8 l0 2 A262 1265 1270 . \\\ TRANS \ L/Pt=2 Figure 30 74 Table 7. The stoichiometry of the gig: and trans-platinum nucleoside Couplexes in solution at a pH around 5 Nucleoside cis-Etjlljjflha trans- _ ‘. __ 12212 __ W3.) 25.12 Adenosine PtL, PtZL PtL l-Methyladenosine PtL PtL Cytidine PtL PtL, PtL2 Guanosine Pth PtL2 l-Methylguanosine Pth Pth Inosine PtL2 Pth l-Methylinosine PtL2 PtLZ 7-Methylinosine Pt L Pt L 2 2 7S lyre one species, only that contribute to the spectra. In the case of adenosine, the formed complex PtL will be further attacked to form Pt L as the concentration of the cis-Pt(II)(NH ) C1 increases. The 2 3 2 2 trans-isomer did not show this phenomenon with adenosine, but with [L] c tidine the ratio of . - 1.5 su eats the ossibilit of Y [Pt] + [L] 88 P Y the formation of a mixture of PtL and PtL2 which will hinder the measure- ment of the formation constant. The Formation constant of the cig: and trgn§:platinumégggleggide complexes After determining the existence of one species and the nature of the formed complex, the formation constant of the complex could be determined by measuring the change of the O.D. versus the log[Pt(II)(NH3)2C12] and taking the midpoint of the tiration curve as the point at which half the complex has been formed (see appendix for calculation of the formation constant). Figures 31-38 show the difference spectra of the nucleosides with varying concentrations of the cis- and trans-Pt(II)(NH3)2C1 , at a 2 pH around 5, and also the titration curves. The concentration of the free ligand and platinum complexes that did not bind with the nucleosides could be calculated knowing the total analytical concentrations and the formation constants (see Appendix). Measuring a difference spectra of the Pt, nucleoside mixture versus the free ligand and free metal gives the absorption spectra of the Pt-nucleo- side only. The absorption spectra of adenosine-, l—methyladenosine-, cytidine-, guanosine-Pt complexes are shown in Figures 30, an, 41 and 42 respectively. Table 8 summarizes the different sites of the interaction with the platinum complexes, and Table 9 lists their formation constants. Figure 31. Difference spectra of adenosine, 0.55 X lO—AM in 0.1 M NaClOa, with various concentration of the Pt(II)(NH3)2Cl 2 4 4M; 7, 0.5 x 10“; 8, 0.5 x 10“”; isomers; 5, 0.25 x 10‘ 20, 4 x lo-AM. M; 6, 0.37 X 10- 76 an muswwm I._.ms_n_ -m. LVN l smm mm<< Figure 35. Difference spectra of cytidine, 1074M in 0.1 M NaClOa, isomers, 5, 0.25 X 4 with various concentration of Pt(II)(NH3)2C1 4 2 -4 - 4 M; 7, 0.5 X 10 M; 8, 0.75 X 10 10'4M; 6, 0.37 x 10‘ M; 20, a x 10’ M. 80 dmhommm wozwmwuma ¢ecV‘ com com mm ouswwm u 4ON imN .wn 10v mmdd -4 Figure 37. Difference spectra of guanosine, 1.1 X 10 M in 0.1 M NaClOA, with various concentration of Pt(II)(NH3)2Cl2 isomers, 5, 0.25 X lO—4M; 4 4 4 6, 0.37 x 10’ M; 7, 0.5 x 10' M; 8, 0.75 x 10’ M; 20, a x 10‘4M. ABS 82 GUANO$NE 350 300 1(nm) WFFERENCE SPECTRA 250 300 350 250 1(nm) MFFERENCE SPECTRA Figure 37 Figure 38. Titration curve of guanosine with Pt(II)(NH3)2Cl isomers, 2 plotted from Figure 37. 83 mm musamm Togfzxsé on ON 0_ 04 e0 Huddefid d dddqdqq d d ddudddq d d --du trans 5.6 + - PtL, Pt2L T R l-MefiAdenosine NH cis 3.27 - + PtL “\‘3 N II | §h trans 3.27 - - PtL EN N/ I It Guanosine cis 5.6 - — Pth J? "N trans 5.6 - - Pth ;\ | lizll II I It l-Me—Guanosine cis 5.6 - - 2th CM N§ trans 5 6 - - PtL klfl/ . 2 "2!! Table 8 (cont'd.) - H - Type of Purines 2§_ 51 £7 che a complex Structural Formula tion 5.6 - + PtL2 Inosine cia 5.6 - + PtL 2 H 'N> \ N I . Ii trans l-Me-Inosine cis 5. 6 -— + PtL2 CQJIIN\ I 7 2 hi. If R. trans 5.6 - + PtL 7-Me-Inosine CH3 3.27 + - PtL * 2 "3'15 km N cis trans 3.27 + - PtzL I I: Table 8 (cont'd.) 90 -NH - Type of Pyrimidines “RE £1 E] cheize complex Structural Formula 5.12.11 Cytidine "Hz cis 5.6 + + PtL ” "' 1 trans 5.6 + — PtL2 g”? I. l—Me—Cytidine cis 3.27 - - c" NH ‘12 trans 3.27 - — ) ‘7' “I" Uridine (D cis 5.6 T** )K Her [I trans 5.6 - 0/\~ R. l-Me—Uridine CH3 0 \ ) cis 5.6 - - N | trans 5.6 — - “‘7. R. * - Me (Methyl) **- i (very weak) 91 Table 9. The formation constants of Pt(II)-Nucleoside Complexes p§_ c 8 trans Type of cogplex Adenosine 5.6 0.24xlO+5 0.64x10+5 PtL l-Me-Adenosine 3.27 0.22x10+5 0.1.1x1o+5 PtL Cytidine 5.6 0.66x10+5 ------ PtL Inosine 5.6 0.46x1o+9 0.72x10+9 PtL2 l-Me—Inosine 3.27 l.49x10+9 1.75x10+9 PtL2 Guanosine 5.6 1.06x10+9 l.06x10+9 m.2 l-Me-Guanosine 5.6 0.6xlO+9 0.6x10+9 PtL2 L - Ligand 92 The Interacgion of Pt(II)(NH 2912 with Polynucleotides Studying the interaction of the platinum species with synthetic polynucleotides may aid in understanding the interactions with nucleic acids, which contain a multiplicity of bases and base sequences. Poly Guanylic Acid Poly G (single stranded) interacts with the cis- and trans- platinum isomers in solution at 37°C giving two spectral peaks at 1 nm 290 and 1265nm. The cis- and trans-platinum isomers have been shown to interact monofunctionally with guanosine. The peak at 1 nm could be 290 assigned to the one of guanosine monomers at N7, and the other peak at 1265nm assigned to attack at N Figure 43 shows the difference spectra 1. of poly G with pip: and Egpppeplatinum isomers. Precipitation of poly G occurs at the end of the reaction when the N1 (N1 of guanosine is involved in the hydrogen bonding with the cytidine in the Watson and Crick double helix of DNA) is fully attacked. As the spectra of the supernate of the poly G-platinum complex (before complete precipitation) shows an increase in the absorption band at A nm (platinum attack at 290 N7), the interaction at N1 must decrease the solubility of poly G in solution (Figure 44). Poly Cytidylic acid The difference spectra of poly C with gig: and pgppprplatinum isomers (Figure 45) show two peaks (Azsonm, 1295nm). The similarity between the.gipr and pgppp: poly C spectra excludes the chelation between the N1 and the 6-NH2 group in case of the cis-isomer, as com- pared to the cytidine monomers. Precipitation also occurred at the end of the reaction. Figure 43. Difference spectra of the Pt(II)(NH3)2Cl isomers, 2 0.2 mM, with polyguanylic acid (poly G), 0.1 mM P. 93 m8 ounwam . /\.I\ /oo ._.o .No Figure 44. Difference spectra of the Pt(II)(NH3)2Cl2 isomers, 0.2 mM, with polyguanylic acid (poly G), 0.1 mM P, in 0.1 M NaClOa at 37°C. A, before precipitation; B, precipitation of the complex. 94 ea muswam 10 mm< < _a mmq Figure 47. Difference spectra of the Pt(II)(NH3)2Cl2 isomers, 0.2 mM, with polyuridylic acid (poly U), 0.1 mM P, in 0.1 M NaClOa at 37°C, B, end of the reaction. 98 me shaman + “freak-NH; ‘—_": *3 N a: w» . 110 I) 111 This reaction is suppressed by high concentrations of protons. At pH 4, N1 of adenosine is not protonated and the reaction will be pH independent. t W 8’ 6" it../' ."H2 '1": } \’ / \> + u o A e N l 2 "2'12N'> + .qo—H:||;_NH3—— v9“ =1 ‘ w h" At pH 6, the Pt-species exists as [Pt(II)(NH3)2(H20XOH)]+1and.the reaction will be suppressed as the OH- concentrations rise in case of reaction (a), or be pH independent, as in reaction (b). g ‘V e” 5!. a“! "“2 e v - / e’ \N’ + °" 1” "2 If": I! + HO—PIt—Nl'la g s. \ t b *0 0 e?! * ""2 / ’ a o 4? “ I ‘+- ‘2 e” | 112 U .2. C0 9 a: g z 23 g? 3 Q‘E ”_5 NE «'5 N7 ‘2... it if 15 If: / f?- U D g I 3 l 1 ’l «2.: _“HQEQFUNCT'WAL 3 1’ ¢\~,- I \HHH t5 /’ ./' I \-\B'FUNCTIONAL g I / .0, 2 '\\ 0 13’)", :H \' figerH’f"! . s. r . 0l 2 l2 Figure 55 Schematic representation of the existance of different complexes in solution at different pH's. Figure SSillustrates the interaction at different pH's. If the interaction is monofunctional, the highest concentrations of platinum-adenosine complexes (interaction at N1) will occur at pH . 3.5, 92,93 which is the pK of N of adenosine. The curve turns down at point 1 B if the OH‘ ion rather than the H20 molecule is replaced at pH 6, other- wise it will turn down at the pH - pK2 of the platinum species. The pK2 of the platinum species is 7.2 or 9.2 (point C or D). N7 of adenosine is never protonated, and the reaction will be suppressed at high pH due only to the release of OH' ions from the platinum species, [Pt(II)(NH3)2(OH)2]. 113 Interaction at the 6—NH2 of adenosine *9 ‘5 $2 4* he "N'l !l "\> + ago—mun, _.___=-. "Nm\ + it' pflilt‘ \ I do I a l H20 This reaction is suppressed at low pH, while that at pH 6 is neutral. 45‘?” 4§§g* /‘\o ""2 T“: NH ~ ._____25. so + \——...... ~~ +~° I “2° 114 At pH - 10, the reaction is pH independent. {pfib {pet éMf. ‘C 129 / _ _.__>. - Q>+ "° ”“3 W kfi' >+"’° N P "’ I " I a I: Bifunctional binding at 6-NH2_and N7 of adenosine: At pII = 1 I“! I: an" ' _— '— ’ ""2 "“3 "N ‘2‘ 3 H l __:5 fiflfla’ ._+— ’"N~ II ‘j>'+' lfl!3"P"'N"H3 « I Ifzb "'1 N T "2° I. 115 At pH - 6 ‘ fl": ""2 N" —"—NH3 N/| K) + uo-IL-fuu ——‘\————— / . 2“ lfimi ' Io '3 'fli GI qéfigl .+- if, . "2° I At pH - 10 .. _ 9— .. H 9| N": on ""2 I~‘II-I3 m "’ "°:'j,"""3 flit-HF: 1:39 "‘ I. .1 ..,. 116 Reaction at pH - 1 is suppressed by high concentrations of H+ ions, at pH - 6 it is pH independent, and at pH - 10 it is suppressed (above the pK of the [Pt(II)(NH3)2(H20)2]++. 2 . In the case of dimers like A3'p5'A and A2'p5'C, more groups are available for forming different complex compounds. \\ + illafii 117 _ Z I .. u ”3 no A 1' V !" + ”‘29 ‘ «.2 .3 . + I ~“Eu: 10 9“”: "-lwz nu; ...... 1..., g 2..-a: x . é 2..“..1'“: ”0.02. z "2 3 2 Reaction at pH - l is suppressed at low pH, while that at pH - 6, and above the pK2 of the Pt(II)-complex is enhanced. 118 Figure 55 is a schematic representation of the existence of different 3)2012 has the ability to chelate with a neighboring active group. Adenosine with complexes in solution at different pH's. Cis-Pt(II)(NH cis-Pt(II)(NH3)2C12, Figure 56 shows attack at N7 predominantly from pH 1-3.2. At pH - pK (N1) - 3.5 the interaction is at N and N7 1 (pH - 3.5-5.6). In the region of the pKl-pK2 of the Pt(II)(NH3)2(H20)i++, pH - 5.6-9.2, interaction at N1 and at N7 chelating with 6-NH2, occurs as the predominant species with [Pt(II)(NH3)2(H20)(OH)]++ a pH higher than the pK2 of Pt(II)(NH2)2(H20)2++ the interaction decreases to zero. As the trans-isomer has a pK - 4.2, pK2 - 7.3, and no possibility of 1 chelation, the interaction is monofunctional. N7 is attacked in the region pH - 0-4.2; N7 and N1 from pH - 4.2 to 7, and then a decrease in the high pH region. Cytidine follows the same path as the interactions of N and 6-NH 1 2 of adenosine, and also a chelate is formed with NHZ in the case of the gggfisomer only. Cytidinegz’93 has a pK of N3 . 4.2. Thus the gingt(II)- complex interacts at N3 and chelates with the amino group in the 6-position in the pH region 4.2 to 9. The £5§2§_isomer interacts at N3 in the same region, without chelation with the neighboring amino-group. Guanosinega’gshas a pK of N7 8 2.4 and of N1 - 9.2. At low pH's, N7 of guanosine will be protonated and the interaction at N7 will not be favored. Guanosine Reactions (see next page) Figure 56. Summary of the interactions of the Pt(II)(NH3)2Cl2 isomers with nucleosides at different pH's. 119 em shaman S .— m. . w . .. .. .. .. .N . I I. .N .m _ 8m -_ .. TIL... ....... .\ ... ZmOZ< 2 Is. PM P. ........... I‘\.\. 1” q‘ d 4 go 1 a < 4 40 4. .__ N .N Ln .n q 4 1 T 1 fl 1 4 a . q q a go mzaozmoq .N .F: Slides .. m ,I .. I 116.092 .e mmammHH. 120 At pH I 1 NH , Hé-gfjflfla $4.5 DINO H9 01;; "Nfif’ ~I!) A \ . 1" W*}.' '> ‘91:] ranking?) ‘1'“ I I H20 At pH I 6 60‘s,“?! 121 At pH I 10 Reaction b at pH I 6, and both reactions at pH I 10 are not favored at high pH's. From Figure56 we can conclude that the interaction with the platinum isomers in the pH region 3-9 is at N7, and above pH I 9 the interactions at both sites, N1 and N7, will be suppressed as the concen- trations of the 0H” ions increases. Inosine has the same route of interaction with the Pt isomers as guanosine, the only differeence being in the pK's.96 The pl( of N1 is 8.8 and of N7 is 1.2. The interaction of the platinum isomers with guanosine and inosine is monofunctional. The second active group of the platinum isomers could interact with another molecule forming the Pth complex. 122 AtpHI6 ”$6 91’ ‘\\o I~= as; fir "Git-NH; H2 2 fig?“ ‘51: 6 A ._+__ The formation of Pth complex is also favored in the case of the trans- isomer of the platinum species. a: ”2»: was. __ rm, “04-"? l-Vt" Thu: 2-8 u u , 123 Reactions in both isomers are suppressed at high concentrations of OH- ions in the region below the pK of the platinum species. 2 l-Methzl Guanosine Reactions: qui dy’ é‘e’é O O “'1‘ng * 2““ no '31“ NH "’Nk l 'I' ""2 g " 3 I! ll 2 “'3‘ w _______________;:s lflzhl “:;flH=H6 '-1- Interaction at the N1 and the 6-NH2 of cytidine Cytidine forms PtL complexes with the trans-isomer only; the amino group 2 in the six position chelates with the platinum in the case of the cis- isomer. See next page. 124 I! so 74' pug,“ \n an, [5” I] / \. HO \ 2” gfl A I ‘1" 2 h’x” 2° \ ' 125 A study of the interaction of dinucleotides with cis- and trans-PthIlSNHalzglz by Circular Dichroigm spectrosCOEy The two isomers of Pt(II)(NH3)zClz can bind to the 6-amino group of adenosine and cytidine forming PtL and/or Pth complexes. This led to the investigation of the possibility of the formation of the intra- and inter-strand complexes in DNA in case of the gig:isomer, and inter-strand complexes only in case of the Eggggrisomer. The inter-base separation in DNA97 is close to 3.4; and as the distance between the two gigfligands of 98 is about 3.3A, the linking of two bases square planar platinum compounds to one platinum ion is stereochemically possible. The purine dimer (A3'p5'A) provides an ideal model compound. The 6-Amino groups of the bases of ApA in solution lies vertically above one another90 with a separation of 3 to 4A. Thus the amino groups are in an ideal conformation to bind to a metal ion. Since it is the object of this study to detect all possible types of binding with all sites, a circular dichroism technique would reveal the inter-base links. Dinucleotides in solution at room temperature are in an equilibrium between the stacked form (Figure 57) and an unstacked or random conformation. The stacked conformer has a characteristic circular dichroism spectrum which is more intense than that of the unstacked form. The temperature dependence of the equilibrium between stacked and unstacked conformers has been examined by studying the temperature dependence of the CD spectrum characteristic of the stacked conformer. If the binding of gigyPt(II)(NH3)2C12 to A3'p5'A results in a cross- link, then the dimer will be locked into the stacked conformation. Such a dimer will not pass into the unstacked form as the temperature is raised, since the platinum-base complexes are stable in solution at these higher Figure 57. M NaClOa. CD spectra of A3'p5'A. Conc. dimer, 1 X 10- l. M, in 0.1 126 300 127 temperatures. The absorption spectrum of A3'p5'A, and the temperature depen- dence of its CD spectrum agreed well with previously published results. The absorption spectrum of the incubated A3'p5'A with trans-Pt(II)(NH3)2C12, molar ratio of 1:1 (Dimer to Ft), is shifted to A nm (Figure 58) and 270 the characteristic double peaked CD spectrum is abolished. Thus the binding of the transfisomer has brought about unstacking of the dimer. This effect is similar to that observed with A3'p5'A at low pH. At a pH below 3.5 (pK of N1 of adenosine),N 1is protonated and each base will carry a positive charge at pH 1 and the repulsion between the bases will lead to unstacking. Metallation at N1 of adenosine will place a positive charge on the base. If both bases of A3'p5'A are metallated, repulsion and complete unstacking will take place. In contrast to this result, the CD spectrum of the complex between the gig:isomer and A3'p5'A shows a remnant of the double peaked CD curve characteristic of a stacked conformation, and is temperature independent, confirming that this component of the solution does not arise from a small residual quantity of stacked, but unmetallated, conformer. Since most of the binding of the gig:isomer is of a mono-functional type, the strength of the CD signal is low due to the small concentration of the dimer- complex in solution. This species is most likely to be a form of A3'p5'A with both bases linked to one platinum ion via their 6-amino groups. In DNA itself the 6-amino-groups of two adenine bases are vertically above one another with a separation of 3.4A. The binding of cis-Pt(II)(NH3)2 Cl2 to two adenine moieties via their 6-amino-groups could introduce an inter— and intra-strand cross-link into DNA. Since the platinum isomers could interact at many sites with almost all the bases of DNA, the kinetics of the interaction of the two isomers with DNA and Figure 58. CD spectra of A3'p5'A with Pt(II)(NH3)2Cl2 isomers. Conc. dimer, 1 X lO-aM; dimer: Pt(II)(NH3)2C12,———-, l X 10-4M; _-_ , 1:1; _.- , 1; 3.0; -~-- ; 1:5; ---- , 1:10; in 0.1 M NaClOa. 126' is. K .\ -\Oossooo lo“... ' O {A I _ mz _CH. CH. JL ,H— \ N 0| O N N ’N\ O-FID-O H 'fi/LO/H H I H CH3 1 Figure 59 133 complexes. The gig:isomer is more likely to form an intra-strand complex than the Eggggrisomer. All the possible complexes could be formed with both the gig: and .gggggrisomers of the square planar dichlorodiammineplatinum(II). However the traggrisomer does not inhibit DNA synthesis and has no antitumor action. Further work needs to be carried out to investigate whether the trans-Pt(II)(NH Cl2 binds to DNA in the cell or not. Another area to 3’2 be investigated is to determine which complex compound is responsible for the antitumor activity in the case of the cis-platinum isomers. Additional work needs to be carried out on the kinetics of the reaction of DNA with the platinum complexes, and the isolation of these complexes by chroma- tographic techniques. BIBLIOGRAPHY 10. ll. 12. 13. 14. 15. 16. 17. BIBLIOGRAPHY Rosenberg, 8., Van Camp, L., and Krigas, T., Nature 205, 698 (1965). Rosenberg, 8., Van Camp, L., Crimley, E. 8., and Thomson, A. J., J. Biol. Chem. 242, 1347 (1967a). Renshaw, 8., and Thomson, A. J., J. Bacteriol. 24, 1915 (1967). Rosenberg, 8., Van Camp, L., Trosko, J. 8.. and Mansour, V. H., Nature 222, 385 (1969). Rosenberg, 8., Van Camp, L., Cancer Research 39, 1799 (1970). Tolley, R. w., Proc. Amer. Assoc. Cancer Res. 11, 78 (1970). Kociba, R. J., Sleight, S. D., and Rosenberg, 8., Cancer Chemo- therapy Reports (part 1) 24, 325, (1970). National Cancer Institute (NCI), Bethesda, Md, Screening summary data. (1971). Dos, T. L., Prog. EXptl. Tumor Res. 5, 157 (1964). Middleton, P. J., Brit. J. Cancer 12, 830 (1965). Hugguns, C., Braiziarelle, C.,Sulton, H. J., Exptl. Med. 109 (1959). Welsch, C. H., J. Nat. Cancer Inst. unpublished (1971). Siricia, A., Venditti, J. M., and Kline, 1., Sixty-second Annual Meeting. Proceedings of the American Association for Cancer Research 12, 4 (1971). Speer, R. J., Lapes, S., Ridgmary, H., Meyers, T. D., and Hill, J. M. Cis-Platinum Diammine Dichloride (PDD) in Combination Therapy of Leukemia LlZlO, presented at the VIIth International Congress of Chemotherapy, prague, (Abstracts 8-1.4/60) (1971). Witkena, E. M., Proc. Nat. Acad. Scie. (Wash.), 51, 1275 (1967). Adler, H. 1., and Hardigree, A. A., Rad. Res. 2;, 92 (1965). Rosenkranz, H. S., Carro, A. J., Levy, J. A., and Carr, H. 8., Biochem. BiOphys. Acta. 114, 501 (1966). 134 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 135 Young, C. H., Schochetman, C., and Kormofsky, D. H., Cancer Res. 31, 526 (1967). Setlow, R. 8., Swenson, P. A., and Carrier, w. L., Science 142, 1464 (1963). Young, C. H., and Hodas, S., Science, 146, 1172 (1964). Harder, H. C., and Rosenberg, 8., Inhibition of DNA Synthesis in Vivo in Human Amnionic AVé Cells Heated with Tumour Inhibit- h ory Platinum Compounds. ird International BiOphysics Congress of the International Union for Pure and Applied Biophysics, Cambridge, Mass. U.S.A., August 29-September 3, 31 (1969). Harder, H. C., and Rosenberg, 8., Int. J. Cancer 6, 207 (1970). Kim, J. H., Celbard, A. S., and Berez, A. C., Cancer Res. 27, 1301 (1967). Pfeiffer, S. 8., and Tolmach, L. J., Cancer Res. 21, 124 (1967). Yarbro, J. H., Cancer Res. gg, 1082 (1968). Howle, J. A., and Gale, C. R., Cis Dichlorodiammine Platinum II: Persistent and Selective Inhibition of Deoxyribonucleic Acid Synthesis in Vivo. Submitted for publication. (1970) Basolo, F., and Pearson, R. C., Progr. Inorg. Chem. 4, 318 (1962). Langford, C., and Gray, H. 8., Ligand Substitution Processes, w. A. Benjamine, New York (1966). Banerjea, D., Basolo, F., and Pearson, R., J. Am. Chem. Soc. 12, 4055 (1957). Tschugaiff, L., Compt. rend. 161, 563 (1915). Sundaram, A. K., Sandell, E. 8., J. Am. Chem. Soc. 11, 855 (1955). Malatesta, L., and Vallarimo, L., J. Chem. Soc. 1867 (1956). Harris, C. H., and Nyholm, R. S., J. Chem. Soc. 4375 (1956). Llewellyn, D. R., O'Connor, C. J., and Odell, A. L., J. Chem. Soc. 196 (1964). Gray, H. 8., J. Am. Chem. Soc. 84, 1548 (1962). Roberts, J. J., and Poscoe, J. M., Nature (in press). Ball, C. R., and Roberts, J. J., Chem. Biol. Interactions (in press). 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 136 Dekker, C. A., Ann. Rev. Biochem.‘12, 465 (1960). Brookes, P., and Lawley, P. D., J. Chem. Soc. 1348 (1962). Haines, J. A., Reese, C. 8., and Todd, L., J. Chem. Soc. 1406 (1964). Pullman, 8., and Pullman, A., Les Thiories Electroniques de la Chimie Organique, pp, 328-30, Masson, Paris, 1952. Nakajima, T., and Pullman, A., J. Chim. Phys. 793 (1958). Albert, A., Heterocylic Chemistry, The Athlone Press, London, 1959, Brookes, P., and Lawley, P. D., J. Chem. Soc. 539 (1960). Fischer, 8., Chem. Ber. 30, 2400 (1897). Jardetzky, C. D., and Jardetsky, 0., J. Am. Chem. Soc. 81, 272 (1960). Windmneller, H. C., and Kaplan, N. 0., J. Biol. Chem. 116, 2716. Brookes, P., and Lawley, P. D., J. Chem. Soc. 539 (1960). Albert, A., Coldacre, R., and Phillips, J., J. Chem. Soc. 2240 (1948). Boarland, H. P. V., and McOmie, J. F. N., J. Chem. Soc. 3716 (1952). Brown, D. J., and Harper, J. S. L., J. Chem. Soc. 1298 (1961). Bryson, A., J. Am. Chem. Soc. 81, 4871 (1960). Marshall, J. R., and Walker, J., J. Chem. Soc. 1004 (1951). Boarland, M. P. V., and McOmie, J. F. N., J. Chem. Soc. 3722 (1952). Brown, D. J., Hoerger, 8.. and Mason, S. F., J. Chem. Soc. 211 (1955). Mason, 8. F., J. Chem Soc. 674 (1958). Brown, D. J., and Short, L. N., J. Chem. Soc. 331 (1953). Brown, D. J., Nature 162, 1010 (1950). Whittaker, N., J. Chem. Soc. 1565 (1951). Kenner, C. N., Reese, F. 8., and Todd, A. R., J. Chem. Soc. 855 (1955). 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 137 Lawley, P. D., and Wallick, C. A., Chem of Ind. (Londdn) 633 (1957), Rainer, 8., and Zamenhof , S., J. Biol. Chem. 228, 475 (1957), Brookes, P., and Lawley, P. D., J. Chem. Soc. 3928 (1961), Jones, J. W., and Robins, R. K., J. Am» Chem. Soc. 85, 193 (1963). Cerduschek, E. P., Proc. Natl. Acad. Scie. U.S. 47, 950 (1961). Whitehead, C. W., and Traverso, J. J., J. Am. Chem. Soc. 21, 3971 (1960). Froenkel-Conrat, H., Biochem. et BiOphys. Acta.12, 307 (1954). Staehelin, M., Biochem. et Biophys. Acta 12, 410 (1958). Timmis, C. M., Proc. VIIth Int. Congr. of Thematology, Vol. III, 657, "II Rensiero Scientifico" Publ. Roma( 1968). Pullman, A., and Pullman, 8., Bull. Soc. Chim. France, 766 (1958); 591 (1959). Eichhorn, C. L., Nature 122, 474 (1962). Eichhorn, C. L., Aduan. Chem. 12, 378 (1966). Shack, J., Jenkns, R. T., and Thomprett, J. M., J. Biol. Chem. 203, 373 (1953). Thomas, R., Trans Faraday Soc. 29, 324 (1954). Steiner, R. F., and Beers, R. F., Biochim. BiOphys. Acta.11, 166 (1956). Fuwa, K., Wacker, W. E. C., Druyon, R., Batholomy, A. F., and Valler, B. L., Proc. Natl. Acad. Sci. U.S. 22, 1298 (1960). Dove, W. F., Davidson, N., J. Mol. Biol. 2, 467( 1962). Eichhorn, C. L., and Shin, Y. A., J. Am. Chem. Soc. 90, 7323 (1968). Cohr, M., Hughes, T. R., J. Biol. Chem. 273, 176 (1962). Eichorn, C. L., and Clark, P., Proc. Natl. Acad. Sci. U. S. 22, 586 (1965). Eisinger, J., Shulman, R. C., and Szymanski, 8. M., J. Chem. Phys. 22, 1721( 1962). Rapars, C., and Viovy,R., J. Chim. Phys. 21, 408 (1965). Brintzinger, H., Biochim. BiOphys. Acta 11, 343 (1963). Mall, H., Schneider, P. W., and Brintzinger, H. Helv. Chim. Acta 11, 1837 (1964). 138 81. Yamane, T., and Davidson, N., J. Am. Chem. Soc. 22, 2599 (1961). 82. Ferreira, R., Ben-Zvi, 8., Yamane, T., Vasilenskis, J., and Davidson, N., "Advances in the Chemistry of the Coordination Compounds", MacMillan Co., New York, N. Y. 457-62 (1961). 83. Simpson, R. 8., J. Am. Chem. Soc. 22, 2059 (1964). 84. Carrabine, J. A., Sundaralingam, M., Chemical Communication 746 (1968). 85. Carrabine, J. A., Sundaralingam, M., Biochem 12, 292 (1971). 86. Eichorn, C. L., Clark P., and Becker, E. D., Biochemistry 2, 254 (1966). 87. Bichorn, C. L., Berger, N. A., Biochemistry 12, 1847 (1971). 88. Eichorn, C. L., Berger, N. A., Biochemistry 12, 1857 (1971). 89. Hoeschele, J. D., Biophysics Department, Michigan State University, East Lansing, Michigan 48823, U.S.A. Unpublished work. 90. Job, M. P., Ann. Chem. (II) 6, 97 (1936). 91. Tucker, M. A., Colvin, C. 8., Martin, D. 8., Jr., Inorg. Chem. 2, 1373 (1963). 92. Miles, H. T., Biochem. Biophys. Acta 22, 324 (1958). 93. Miles, H. T., Bradley, R. 8., and Becker, 8. D., Science 142, 1569 (1963). 94. Jardetzky, C. D., and Jardetzky, 0., J. Am. Chem. Soc. 21, 222 (1960). 95. Miles, H. T., Howard, F. 8., and Frazier, J., Science 142, 1458 (1963); 96. Miles, H. T., Proc. Natl, Acad. Sci. U.S. 21, 791 (1961). 97. Wilkins, M. H. F., and Arnott, S., J. Mol. Biol. 11, 391 (1966). 98. Truter, M., and Milburn, C. H. W., J. Chem. Soc. (A) 19609 (1966). 99. T50, P. 0. P., Holmes, H. M., Stempol, L. M., and Kondo, N. 8., Biochem. 2, 3491 (1970). 100. Van Holde, K. 8., Brahms, J., and Michelson, A. M., J. Mol. Biol. 11, 726 (1965). 101. Brahms, J., Maurizot, J. C., Michelson, A. M., J. M01. Biol. 11, 726 (1965). 139 102. Horacek, P., and Drobnik, J., Institute of Macromolecular Chemis- try, Czechoslovak Academy of Science, Prague, Czechoslovak, unpublished work. 103. Ibid, unpublished work. 104. Vosbrugh, W. C., and Cooper, C. R., J. Am; Chem. Soc. 22, 437 (1941). APP END IX Calcu1gtion of the formation constants of the Pt(II)-nucleoside complexes. Adenosine absortivity A A ltt (1C t C10 . - as y C1 oncen ra n Extinction coefficient x cell path 8 x b E of adenosine I 14.9 x 10+3 b 1 cm A of adenosine I 0.82 C I total analytical concentration of the ligand in solution 0-82 = 0.55 x 10‘“ a 14.9 x 10+3 therefore C Figure 24 shows that the formed complex in solution of the cis- and trans-Pt(II)(NH3)2012 is PtL l . (24) [Mxl- l/2[LO] therefore Kf Kf I the formation constant of the Pt(II)-nucleoside complex Mx I concentration of the Pt(II) at the midpoint of the curve (Figure (1) cis-Pt(11)-adenosine comglex, (PtAZCis Figure 32 shows Mx 6.5 x 10"5 M 1 = 0.24 3 10+5 therefore Kf I 5 6.5 x 10‘ - 1/2 x 0.55 x 10‘“ K (PtA)Cis f a " [CA ' (PtA)cis] [CPt ’ (PtA)cis] 140 - 0.24 x 10+5 141 CA , total concentration of adenosine in solution I 0.55 x 10-4 M CPt’ total concentration of cis-Pt(II)(NH3)2012 in solution I 4 x 10-4 M Therefore (PtA) ' 0.24 X 10+5 [2.2 x 10-8 ’ 4'55 x 10-Z.(P':A)cis cis 2 + (PtA) ] cis (PtA) - 0 49 x 10" M cis ° and L I the free ligand = 0.55 x 10-4 - 0.49 x 10-4 I 0.06 x 10-4 M Cis-Pt(II) = the free cis-Pt(II) = 4 x 10'4 - 0.49 x 10‘“ . 3.51 x 10'° M (ii) Trans-Pt(I1)—adenosine complex, (PtA)T Figure 32 shows Mx = 4.2 x 10"5 M Kf = 1 - 0.69 x 10“5 -5 —5 4.2 x 10 - 2.75 x 10 (PtA)T K = = 0.69 x 10+5 CPt’ total concentration of trans-Pt(II)(NH3)2C12 in solution I 4 x 10‘4 M Therefore (ptA)T = 0.69 x 10+5[2.2 x 10‘8 - 4.55 x 10'4 (pm)T + (PtA)2] T (Pt/I)T = 0.48 x 10‘“ M 4 and the free ligand I 0.55 x 10‘“ - 0.48 x 10- I 0.07 x 10_4 M the free trans-Pt(II) I 4 x 10'“ - 0.48 x 10-4 I 3.52 x 10-4 M 142 As illustrated in the previous case, the concentration of the free ligand (L) of the following nucleosides and the free platinum species (Pt(II)) in solution, could be calculated. The total concentration of the platinum species in 4 x 10-4 M in all cases of the nucleosides. IIMethyladenosine The formed complex is of the PtL type (Figure 25) in both cases of the cis- and trans-Pt(II)(NH3)2C12. Knowing the analytical concentration of the nucleoside (0.41 x 10—410 the following physical constants were calculated. (i) cis-Pt(11)-methy1adenosine complex 5 M (obtained from Figure 34) 5 M = 6.5 x 10— X Kf . 0.22 x 10+ L . 0.05 x 10“ M g1§ePt(II) = 3.65 x 10‘“ M (ii) grans-Pt(11):methy1adgposine Mx = 4.5 x 10.5 M (obtained from Figure 34) x, - 0.41 x 10+5 L - 0.04 x 10‘“ M trans-Pt(II) I 3.63 x 10'4 M gggnosine The formed complex is of the PtL2 type in both isomers of the Pt(II)(NH3)2CIZ. Using equation 31 and knowing the 143 analytical concentration of the nucleoside (1.1 x 10"4 M ) and the following physical constants were calculated. K . 1 1 (31) f [LO] [Mx] - 174 [L0] (i) cis—Pt(I1)_guanosine comp1ex - 3.6 x 10'5 M (obtained from Figure 33) Z I +9 74‘ II 1.06 x 10 0.05 x 10“ M 1'" I Cis-Pt(II) - 2.05 x 10‘“ M (ii) trans-Ptjll):guanosine complex 5 M (obtained from Figure 38) 9 Mx a 3.6 x 10‘ Kf - 1.06 x 10* L - 0.05 x 10‘“ M [Egflg—Pt(11) = 2.05 x 10-4 M 1—Methy1gpanosine The formed complex is also of the Pth type in both isomers of the Pt(II)(NH Knowing the analytical concentration of the l. 3)2012° nucleoside (1.27 x 10- M), the formation constant of the following complexes were calculated. (1) cis-Pt(11)-methylguanosine complex M - 4.5 x 10‘5 M x 14f - 0.6 x 10+9 144 ( ii) trans-Pt(II)-methylguanosine complex -5 MK I 4.5 x 10 M Kf . 0.6 x 10+9 Inosine The formed complex is of the Pth type in both isomers of the Pt(II)(NH C1 Knowing the analytical concentration of the nucleo- 3)2 2' side( 1.097 x 10-4 M) the formation constant of the following complexes was calculated. (1) cis-253112;;nos1ne comp1ex M a 4.7 x 10"5 M X Kf - 0.46 x 10+9 (ii) trans-Pt(11)-inosine complex M I 4 x 10"5 M x xf - 0.72 x 10+9 M 1—Methy1inosine The formed complex is also of the PtL2 type in both isomers of the Pt(II)(NH3)2C1 Knowing the analytical concentration of the nucleoside 2. (0.96 x 10" M). The formation constant was calculated. (1) cis-Pt(II):methylinos1ge complex M - 3.1 x 10‘5 M X Kf - 1.49 x 10“9 145 (ii) trans-Pt(11)-methylinosine com21ex M - 3.0 x 10‘5 M x xf - 1.75 x 10+9 gxtidine The formed complex is of the PtL type in case of the cis—Pt(II)(NH3)2C12. There was more than one complex formed in case of the trans-Pt(II)-isomer and thus its formation constant could not be calculated. Knowing the analytical concentration of the nucleoside (10-4 M), the physical constants of the following complex were calculated. cis-P2(II)-qytidine complex M I 6.5 x 10-5 x Kf - 0.66 x 10+5 L - 0.07 x 10'“ a cis—Pt(II) I 3.07 x 10-4 M "‘IIIIIIIIIIIIIII“