Indian Journal of ChemistryVol. 35A, October 1996, pp. 856-860

Oxidation of pyrimidine and purine base components of nucleic acids byhexavalent manganese in aqueous alkaline medium: A kinetic and

mechanistic study

Koya Venkateswara Rao, Mundra Thirupathi Rao & Mundra Adinarayana*

Department of Chemistry, Osmania University, Hyderabad 500 007

Received 19 December 1995; revised 23 Apri11996

Oxidation of pyrimidine bases viz, uracil, thymine and cytosine and purine bases viz., adenine andguanine by MnOl- in aqueous alkali shows first order dependence on [MnOl-] and fractional orderdependence on [nucleic acid base]. The effect of ionic strength is negligible on the rate of oxidation. 5,6-Dihydroxylated compounds have been found to be the products of oxidation in the case of pyrimidineswhile purines yield the 5-formamidopyriInidines. A mechanism is proposed involving formation of a5-membered cyclic complex between the substrate and MnOl- in a fast step which subsequently dis-proportionates to give the final products. The formation constants of the intermediate complex and therates of its disproportionation have been calculated from the derived rate law. The rate of oxidation ofpyrimidines is in the order: thymine> uracil> cytosine while in the case of purines the order is guanine> adenine.

Loss of biological activity in living cells caused byionising radiation is believed to be the result ofdamage to DNA!. The OH radical produced byionising radiation react base or sugar moiety ofDNN. The base radicals which are generated bythe reaction of OH radicals in DNA might ab-stract hydrogen from the sugar resulting in thetransfer of radical site to the sugar", The sugarradicals thus produced cause either base release orstand breaks or both in DNA which are severetype of damages to DNA in living cells.

Oxidants like OS04 have been found to .reactpredominantly with the thymine moiety of ~X-174single stranded DNA leading to hydroxylation ofthymine moiety without any significant strandbreaks. The loss of biological activity in ~X-174DNA was attributed to the hydroxylation of thy-mine in DNN. Burton and Riley' and Beer et aL6studied the reaction of mono- and oligonucleo-tides with OS04 and observed that thymine moietyreadily reacts with OS04 while cytosine moiety isless reactive. The purine moieties in mononucleo-tides, namely, adenine and guanine are found tobe resistant to oxidation by OS04' It was also re-ported that OsO 4 reacts predominantly with thethymine moiety of denatured DNA, but does notreact with the same in native DNA. Recently,

there has been significant effort directed towardsthe selective oxidation of DNA by metal com-plexes7• Significant progress has been made inachieving selectivity by altering the binding specif-icity of the complex by appropriate tailoring of theligands, leading to shape-selective cleavage's". Ef-forts have also been directed towards achievingselectivity by altering the reactivity of the metalcomplex'P'". Mn04- is reported'v'? to selectivelydegrade thymine residues in DNA. Extensive de-gradation results in oxidation of cytosine residuesalso but the purine bases are not affected undermild conditions. Hexavalent manganese oxidationsare milder compared to those of heptavalent man-ganese under otherwise similar conditions'v ".Manganate ion, being similar to permanganate instructure, is reported to react with organic com-pounds containing double bonds in a similar fash-ion!". In fact we have reported the oxidation ofuracil and thymine by Mn(VI)16 and the rates ofoxidation were found to be lower compared tothose with permanganate. In an attempt to under-stand the effects of Mn(VI) on nucleic acids, herewe report the kinetics of oxidation of cytosine,adenine and guanine by barium manganate in al-kaline medium. For comparison we have incorpo-rated the data on the oxidation of uracil and thy-mine also.

RAO et al: KINETICS OF OXIDATION OF NUCLEIC ACIDS BY Mn(VI)

Materials and MethodsBarium manganate and its stock solutions were

prepared as per procedure reported earlier". Thesolution of BaMnO 4 is stable and homogeneous in1.0 mol dm - 3 KOH and the solutions were alwaysprepared fresh just before use. Uracil, thymine,cytosine, adenine and guanine were of Sigma makeand used as received.

In the kinetic runs, the reaction was quenchedat different times by adding 5 em? aliquots to amixture containing 10 em" of 1.0 mol dm - 3 aceticacid and 10.0 em? of 5% Kl solution. The un-reacted Mn(VI) was then estimated by titrating theliberated 12 against standard Na2S203 solution.The results obtained by this method of analysiswere found to agree well with those obtained byspectrophotometric method, where the unreactedMn(VI) was estimated at regular intervals of timeby measuring the absorbance at 600 nm E = 2000dm! mol "! em -1. The spectrophotometric mea-surements were carried out on a Hitachi U-1000spectrophotometer. The HPLC system used foranalysis of the products includes Shimadzu LC 6AD equipment with a dual piston-pump system, asolvent programmer .and universal loop injector.Prepacked octadecylsilyl silica gel ODS Nucleosilsemi preparative column (25 cm x 0.46 CM, meanparticle size 5 ILm)was used. The column effluentswere monitored at 220 nm, 260 nm or 280 nmusing variable wavelength detector equipped witha 8 ILlflow cell. The chromatographic runs wererecorded on a Shimadzu model 221-25412 recor-der ala chart speed of 5 ern/min.

The buffer used in the gradient elution experi-ments consisted of 50 mM KH2P04 aqueous solu-tion adjusted to pH 5.6 with phosphoric acid. Be-fore use the phosphate buffer was filtered througha millipore type HA 0.45 ILmmembrane filter. Allmobile phases were degassed using a vacuumpump. The solvent flow rate was kept constant at0.5 ml/min. The pressure of the pump was main-tained at 30 bar. All the HPLC runs were carriedout at ambient temperature.

Pyrimidines, viz., uracil, thymine and cytosine,and purines, viz., adenine -and guanine (1 x 10- 3mol dm-3) were allowed to react with MnOl- inalkaline medium under kinetic conditions. Whenthe green colour of the solution disappeared, thereaction mixture was filtered through a WhatmanNo. 40 filter paper to remove Mn02. The reactionmixture after filtration was neutralised with Heland the pH was brought to 6.0. This solution wasthen injected into the HPLC system.

The unreacted uracil, thymine and cytosinewere detected at their A max 260 nm with different

857

..~o

'"o

tC)

+on

t,,0.,'

~oo 0

0-5 oe

0-' 0-1

0-3 0-6

/0-2 0-5

0-' 0400 40 60 120

0-20 0-40 0-60 0-60

020 0-40 0- 60 0-60

0-9

0-8

0-7

0-6

L-_L-_L-_~_-L -.J 0-5

160 ~ Tim!,min (A)

1-00-':' 3+ 'og [Cytosine) (8)

1-00_ 3+ 'og [Adenino)(C)

Fig. l-(A) Order in [MnOl-] in the oxidation of uracil byMnOl- in aqueous alkaline medium[uracil] = 5.0 x 10-3 mol dm-3; [KOH] = 1.0 mol dm-3;

temp. = 313 K; plot of 4+ log[MnOl-] versus time

(B) Order in [cytosine] in the oxidation of cytosine byMnOl- in aqueous alkaline medium[MnO~-]=7.20xlO-4 mol dm" '; [KOH]=l.O moldm-3; temp=323 K; plot of 5+logkobs versus

3 + log [cytosine]

(C) Order in [adenine] in the oxidation of adenine byMnOl- in aqueous alkaline medium[MnOl-] = 2.37 x 10-4 mol dm-3; [KOH] = 1.0 moldm - J; temp. = 333 K; plot of 5 + log kObs versus

3 + log[adenine]

retention times while their products were detectedat 220 nm. The retention times of the productsare comparable with those of the authentic sam-ples of the 5,6-dihydroxylated products of uracil,thymine and cytosine.

The unreacted adenine and guanine were de-tected at 260 nm while their products were detect-ed at 280 nm with different retention times. Theeluent collected at 280 nm was tested for thepresence of formaldehyde using chromotropic acidwhich yielded a violet colour. On the basis of thisanalysis, the product in the oxidation of purines ofMnOl- may most probably be 4,6-diamino-5-formamidopyrimidine and 2,6-diamino-4-hydroxy-5-formamidopyrimidine with adenineand guanine respectively.

Results and DiscussionKinetic experiments were conducted under

pseudo-first order conditions with excess of [sub-strate] over [Mn(NI)]. Under these conditions theplot of 10g[MnOl-] versus time was linear (Fig.1A), indicating first order dependence on

858 INDIAN J CHEM. SEe. A, OCTOBER 1996

Table I-Effect of [substrate] on kObS in the oxidation of uracil,thymine, cytosine or adenine by hexavalent manganese in

aqueous alkaline medium[MnOl-j= 7.2 x 10-4 mol dm ":'; [KOHj= 1.00 mol dm':

temp.=323 K103[Substrate] 105 k"h< (s -I)

mol dm-3

uracil thymine cytosine adenine2.0 5.13 11.5 1.67 3.644.0 8.93 15.4 2.62 4.796.0 12.2 20.9 3.20 5.378.0 14.3 24.8 3.52 5.96

12.0 18.9 29.0 4.56 6.6116.0 22.2 41.1 7.16

Table 2-Effect of [Mn01-] and [guanine] on initial rate in theoxidation of guanine by MnOl- in aqueous alkaline medium

[KOH] = 1.0 moldm-3;Temp.=313 K104[MnOl-] 104[guanine] lOH[initial rate](mol dm-3) (mol dm-3) (mol dm-3s-l)

3.607.2010.814.421.6

7.207.207.207.20

5.005.005.005.005.001.25

2.5010.015.0

1.82

3.675.257.0810.51.30

2.325.827.70

[Mn01-]. The pseudo-first order rate constant(kobs) values were calculated from the slopes ofsuch plots for uracil, thymine, cytosine and ade-nine (Table 1). The plot of logkobs versus Ioglsub-strate] was linear with slope equal to 0.60 for ura-cil, thymine and cytosine (Fig. 1B) whereas in thecase of adenine the slope is 0.3 (Fig. 1C). How-ever, in the case of guanine initial rate method wasused to analyse the rate as pseudo-conditionscould not be maintained due to its low solubility.The initial rates were obtained by computer analy-sis and the rate increases with increase in[Mn01-] at constant (guanine]. The plot of logiin-itial rate) versus log[Mn01-], is linear with unitslope indicating first order dependence. on[Mn01-] (Table 2). An increase in [guanine] in-creased the initial rate and the plot of log(initialrate) versus log[guanine] (Table 2) is also linearwith a slope equal to 0.60. The rate constant (kObs)for Mn01- oxidation of uracil or adenine doesnot change appreciably over a four fold change inionic strength from 0.20 to 0.80 mol dm-3• An in-crease in [KOH] also has no effect on the ob-served rate constant.

CD ••

i i2·04'0

103·0

x

0·02·0

1·0

0'0 '--_-'-_--L. __ L...-_+- -I

0'0 2000 4000 6000 8000_ '/[Guonino] l Al

0'0 50 100 15·0 200 -- 'I[Thymino] l B I

Fig. 2-(A) Search for complex formation in MnOl- oxida-tion of guanine in aqueous alkaline medium at 323 K,

[MnOl-) = 7.20 x 10-4 moldm-3;[KOHj = 1.0moldm-3;plotof10--; Kxf L'rate) versus.Lz[guanine]

(B) Search for complex formation in MnOl- oxidationof thymine in aqueous alkaline medium at 333 K, plot

of 10-3 x (II kohJ versus 1![thymine]

It is generally accepted that Mn04- in aqueoussolutions reacts rapidly with compounds contain-ing > C = C < 17-19 to form an intermediate cyclic5-membered complex, which subsequently decom-poses to give products. In the Mn04- oxidation ofnucleic acid bases it has been reported20-23 thatMn04- shows specificity towards thymine and oxi-dises it to 5,6-dihydroxy-5,6-dihydrothymine and5-hydroxy-5-methylbarbituric acid as the primaryoxidation products, although further oxidation hasbeen reported to occur under vigorous conditions.Hexavalent manganese in the form of Mn01- inaqueous alkaline solutions has been reported'v "to react with > C = C < in a similar fashion butreactions are relatively slow. Under the experi-mental conditions employed in the present studythe negligible effect of added salts on the rate ofoxidation indicates the reaction to be of eitherion-dipole or dipole-dipole type. Mn{VI) in aque-ous alkaline solution exists as MnO1- and there-fore the reaction is expected to be betweenMn01- and neutral pyrimidine molecule. Reac-tions of Mn01- with organic substrates have beenreported to involve a direct two electron transferand also one electron transfer mediated throughMnO]- (ref. 24). However, we have observed inour systems MnO1- acting as two electron trans-fer reagent'v". In the present work also, we have

RAO et al.: KINETICS OF OXIDATION OF NUCLEIC ACIDS BY Mn(VI) 859

Table 3-Disproportionation constant (k) and formation constant of the complex (K) and the activation parameters for theoxidation of pyrimidine and purine bases by MnO~- in aqueous alkaline medium at 323 K

104 X k K sit» ss:(S-I) (kJmol-l) (Jmol=K:")

(a) Pyrimidine bases:uracil 4.00thymine 6.67cytosine 0.714

(b) Purine basesadenine 0.779guanine 2.22

Error limits are standard deviations.

72.2 50.0± 1.36 -139.9 ± 4.2078.4 32.6±2.93 - 205.2 ± 9.10145 27.2 ± 11.2 - 239.7 ± 34.6

428 14.3 ±0.60 - 279.9 ± 1.941500 32.3±0.06 -215.7±0.80

checked the possible involvement of MnOl- inthe reaction. MnOl- has been reported to havestrong absorption at 700 nm'", but we have notobserved any absorption at 700 nm with the reac-tion mixture and also there is no polymerisationwith added acrylamide suggesting that neitherMnOl- nor any radicals are involved in the reac-tion, both with pyrimidines and purines. The frac-tional order dependence of reaction rate on [pyri-midine] indicates that it may be involved in com-plex formation with MnO1-. In fact in the oxida-tion of uracil by permanganate in alkaline mediumformation of cyclic five membered diester typestable complex was observed using stopped flowtechnique".

Therefore, it is reasonable to assume the forma-tion of such a cyclic intermediate complex withMn01- as well. In view of the experimental re-sults obtained and the above discussion, the reac-tion Scheme 1 has been proposed for Mn01- oxi-dation of pyrimidines.

Beer et aL6 have studied the oxidation of purinemononucleotides by OsO4 and reported that thepurine bases are not affected by OS04. Similar ob-servations were also made by Iida and Hayatsu+'with permanganate. Jones and Walker-" reporteddegradation of guanine moiety in DNA by per-manganate under vigorous conditions. However,no attempt has been made on the kinetic and me-chanistic aspects of oxidation of purines by metalions. Under the experimental conditions employedthe negligible effect of added salts on the rate ofoxidation of purines by MnO1- indicates the reac-tion to be of ion-dipole type. The fractional orderdependence of rate on [purine] indicates that itmay be involved in complex formation withMn01-. The metal ion interaction studies carriedout by Chatterji and Nandf" with purines suggestthat N(7) in purines is a potential metal bindingsite. The calculated charge density and localisationenergy values of electrophilic attack at variouscarbon atoms in purine ring suggest that C(8) is

more favourable for electrophilic attack". Fromthe kinetic results, product analysis and the abovediscussion we propose that the MnO1- attacksN(7)....C(8) double bond leading to the formationof a cyclic intermediate as shown in the Scheme 2,

From the Schemes 1 and 2, similar rate lawscan be derived for the oxidation of pyrimidinesand purines by MnO1-. The rate law derived isgiven by Eq. (1).

- d[MnO;-] Kk[substrate][MnO;-]dt 1+ K[substrate]

· .. (1)

- 2.303 dlog[MnO~-] k bs = Kk[substrate]dt 0 1+ K[ substrate]

· .. (2)

It is to be pointed that a generalised law with asummation of terms in the denominator may in-dicate either a change in the rate limiting step orthat the substrate is split substantially between twoforms in a prior equilibrium. A change in the ratelimiting step would lead to changes in the rate lawas the concentrations of the reactants are varied.But, in our study, the rate law (Eq. 1) was foundto be applicable over the entire range of concentr-ations used, indicating no change in the rate limit-ing step. The other possibility that the substrate issplit substantially between two forms in a priorequilibrium is also not applicable in our systembecause of the obvious reasons that the equilibri-um involves the substrate and the reagent and notthe substrate and an acid or base. The rate law ex-plains well the first order dependence of rate on[Mn(VI)]and fractional order dependence on [sub-strate]. Taking the reciprocal of Eq. (2),we get

111-= +-kObs 1+ Kk[substrate] k

· .. (3)

860 INDIAN J CHEM. SEe. A, OCTOBER 1996

According to Eq. (3) the plot of 1/ kobs versus 1/[substrate] should be linear and was found to beso for both pyrimidines and purines [Figs 2(A) &(B)], supporting the proposed mechanism. The for-mation constant of the intermediate cyclic com-plex (K) and the disproportionation constant ofthe complex (k) have been calculated from the in-tercept and slopes of these plots (Table 3).

The reactivity of pyrimidines is found to be:thymine> uracil> cytosine. The high reactivity ofthymine could be due to the ( + I) inductive effectof the methyl group facilitating the formation ofthe cyclic intermediate, which is evident from theformation constant values (Table 3). Similar ob-servations were also made by Freeman et aL29concerning the oxidation of pyrimidines by pota-ssium permanganate. The data in Table 3 showthat guanine reacts faster than adenine.

AcknowledgementThe authors thank, Dr P. Reddanna, Reader,

Department of Animal Sciences, University of Hy-derabad, for providing facilities to carry out HPLCanalysis at their laboratory and Prof. P. Jayapra-kash Rao, Department of Chemistry, OsmaniaUniversity for helpful discussions. One of the au-thors, KVR is thankful to the UGC, New Delhifor awarding teacher fellowship.

References1 Frankenberg D, Frankenberg-Schwager M Blocher D &

Harbich R, Radiat Res, 88 (1981) 524.2 Yon Sonntag C, The chemical basis of radiation biology

(Taylor and Francis, London (1987).3 Adinarayana M, Bothe E & Schulte-Frehlinde D, Int J ra-

diat Bioi, 54 (1988) 723.4 Hariharan P V, Achey P M & Corutte P A, Radiat Res,

69 (1977) 375.5 Burton K & Riley W T, Biochem J, 98 (1966) 70.

---- -------

6 Beer M, Stem S, Carmalt D & Mohlhenrich K H, Bio-chem; 5 (1966) 2283.

7 Sigman D S, Chern C B, Annu Rev Biochem, 59 (1990)207.

8 Pyle A M, Morn T & Barton J K, J Am chem Soc, 112(1990) 9432.

9 Pyle A M, Long E C & Barton J K, J Am chem Soc, 111(1989) 4520.

10 Muller J G, Chern X, Dadiz A, Rokita S E & Burrows CJ, JAm chem Soc, 114 (1992) 6407.

11 Neyhart G A, Grover N, Smith S R, Kalsbeck W A, FairlyT A, Cory M & Thorp H H, JAm chem Soc, 115 (1993)4423.

12 Freeman F, Fluselier C 0 & Karchafski E M, TetrahedronLeft, (1975) 2133.

13 Iida S & Hayatsu H, Biochem Biophys Acta, 240 (1971)370.

14 Pati S C & Dev B R, Proc Ind Natl Sci Acad, 47 (1981)335.

15 Sreenivasulu P V, Adinarayana M, Sethuram B & Nava-neeth Rao T, IntJ chem Kinet, 17 (1985) 1017.

16 Venkateswar Rao K & Adinarayana M, Trans met Chem,17 (1992) 331.

17 Stewart R in Wiberg K B (Ed), Oxidation in organic chem-istry Part A (Academic Press, New York), 1965, p 1, 14.

18 Wiberg K B & Geer R D, J Am chem Soc, 88 (1966)5827.

19 Wiberg K B & Saegebarter K A, J chem Soc, 79 (1959)2822.

20 Hayatsu H & Ukita T, Biochim Biophys Res Common, 29(1967) 556.

21 Hayatsu H & Iida S, Tetrahedron Lett, (1969) 1031.22 IidaS & Hayatsu H, Biochim BiophysActa, 213 (1970) 1.23 Benn M H, Chatmra B & Jones A S, J chem Soc, (1956)

717.24 Donald G, Lee & Carlos F Sabastian, Can J Chern; 59

(1980) 2780.25 Freeman F & Karchefski E M, Biochim Biophys Acta, 447

(1976) 238.26 Jones A S & Walker R T, J chem Soc, (1963) 3554.27 Chatterji D & Nandi U S, J scient ind Res, 37 (1978) 449.28 Pullman B, J chem Soc, (1959) 1621.29 Freeman F, Fuselier C 0, Armsted C R, Dalton C E,

Davidson P A, Karchesfski E M, Krochman D E, JohnsonMN & Jones NK,JAm chem Soc, 103 (1981) 1154.