410 Biochimica et Biophysica Acta, 740 (1983) 410-416 Elsevier

BBA91245

REACTION OF APURINIC/APYRIMIDINIC SITES WITH I t4CIMETHOXYAMINE

A METHOD FOR THE QUANTITATIVE ASSAY OF AP SITES IN DNA *

MYRIAM TALPAERT-BORLE and MICHEL LIUZZI

Laboratory of Biochemistry, Biology Group Ispra, D.G, XII, C.E.C, Joint Research Centre, 1-21020 [spra (VA) (Italy)

(Received April 22nd, 1983)

Key words: DNA damage," AP site," Methoxyamine

A simple and rapid method is described for the determination of AP (apurinic/apyrimidinic) sites in DNA. The method involves the reaction of [ 14C]methoxyamine with the aldehyde group present in the deoxyribose moiety after a base loss. Studies with alkylated-depurinated DNA and with uracil-containing polyde- oxyribonucleotides depyrimidinated by uraciI-DNA glycosylase show that methoxyamine reacts with both apurinic and apyrimidinic sites in a rapid and exhaustive way. Under standard conditions (30-min incubation with 5 mM methoxyamine at 37°C, pH 7.2) untreated DNA is almost unreactive and the [ t4C]metboxyamine incorporation in DNA is proportional to the number of AP sites. Since the methoxyamine reaction is free from any degradative effect on DNA, AP sites may be estimated from a simple determination of the acid-insoluble radioactivity.

Introduction

AP (apurinic/apyrimidinic) sites are very com- mon DNA lesions that can be produced in differ- ent ways [1,2]. They can result from the sponta- neous hydrolysis, even at normal pH, of the glyco- syl bond between deoxyribose and purines [3] and, at a slower rate, between deoxyribose and pyrimi- dines [4]. Some base modifications responsible for the weakening of the glycosyl bond - such as alkylation of purines [5], saturation of C5-C 6 bond of pyrimidines [6], fragmentation of the hetero- cyclic ring [7] - accelerate their appearance. Fur- thermore, some of the damaged bases are en- zymatically excised by specific DNA glycosylases [2,8] leading also to the formation of AP sites. Finally, AP sites can be generated in DNA by

* European Patent pending, application No. P 33 10 563.41/2064 DE.

some antitumor antibiotics such as bleomycin known to depress DNA synthesis [9].

Intact AP sites have so far been assayed indi- rectly; the assay is based on the determination of strand breaks after chemical or enzymatic nicking of a phosphoester bond adjacent to the AP sites. This paper describes a method for the direct assay of all AP sites - intact or already associated with strand breaks - based on the reaction of the aldehyde group with [14C]methoxyamine.

Materials and Methods

Materials. Methoxyamine hydrochloride was obtained from Serva; methylmethanesulfonate from ICN pharmaceuticals Inc. Escherichia coil DNA was purchased from Worthington Biochemi- cal Corporation; E. coil [3H]DNA (0.29 #Ci/#g), [14C]methoxyamine hydroch lor ide (3.13 #Ci/#mol) and omnifluor from New England Nuclear Corporation. Insta-Gel solution was ob-

0167-4781/83/$03.00 © Elsevier Science Publishers B.V.

tained from Packard Instrument Company, and glass fiber paper of type G F / C from Whatman. All other chemicals were analytical grade prod- ucts.

Uracil-DNA glycosylase. The glycosylase from calf thymus was prepared as already described [10].

Uracil-containing polydeoxyribonucleotides. (dA)230. (dT, dU)230 (dT : dU = 9) and (dA)230. (dT, [3H]dU)z30 (150 #Ci/#mol of uracil; dT :dU = 9) were synthesized according to the general procedure previously reported [11].

Polydeoxyribonucleotide with apyrimidinic sites. Apyrimidinic sites were introduced in (dA)230. (dT, [3H]dU)230 with dT: dU = 9 by uracil-DNA glycosylase according to Talpaert-Borl6 and Liuzzi [12]. The reaction was carried out at a concentra- tion of 0.56 mM as total nucleotides with 4.5 enzyme units/ml. After a 3-h incubation at 37°C, the release of [3H]uracil was almost complete. The reaction was stopped and the polydeoxyribonuc- leotide containing about 50 AP sites per 1000 nucleotides was isolated.

Alkylated-depurinated DNA. Unlabelled and tri- tium-labelled E. coli DNA were alkylated with 0.3 M methylmethanesulfonate and partially de- purinated by heating at 50°C for 6 h. The resulting DNAs contained approx. 1 apurinic site/20 nucleotides [13]. Treatment with 0.2 M NaOH yielded an acid-soluble fraction of 35 %.

Determination of acid-soluble radioactivity. To 10-/~1 aliquots of radioactive DNA solutions, 100 #1 calf thymus DNA (200 #g) in 0.15 M NaCI/15 mM sodium citrate (pH 7.0) and 220 /tl 7.5% perchloric acid were added. After shaking, the mixtures were left 10 min in ice and centrifuged at 12 000 × g for 10 re_in. A portion of the super- natant was made up to 0.4 ml with water, supple- mented with 4 ml Insta-Gel and counted in a liquid scintillation spectrometer.

Reaction with [14C]methoxyamine. The reaction mixture in 0.1 M sodium borate, pH 7.2, contained 100-200 /~g/ml DNA or synthetic polyde- oxyribonucleotide and 5 mM [24C]methoxyamine. The incubation was carried out at 37°C for 30 min.

Determination of acid-insoluble radioactivity. Aliquots of the reaction mixtures were spotted on glass fiber disks which were immediately dipped in

411

ice-cold 1 M HC1. The free [14C]methoxyamine was separated from the [14 C]methoxyamine bound to the acid-insoluble DNA by washing the disks five times with 10 ml of 1 M HCI for each disk. After three rinsings with ethanol, the disks were dried, put in 4 ml toluene containing 16 mg omnifluor and counted in a liquid scintillation spectrometer. The values of [Z4C]methoxyamine bound to the acid-insoluble DNA were corrected taking into account the retention of free [t4C]methoxyamine on the disks. The retention amounted to 0.05~ of the total radioactivity put on the disk.

Results

Reaction of methoxyamine with alkylated-de- purinated DNA

Alkylated-depurinated DNA contains some groups which fix methoxyamine in a reaction de- pending on pH (Fig. 1). The reaction occurs mainly in the acidic pH range and drops sharply at pH values greater than 8. On the contrary, metho- xyamine does not react significantly with un- treated DNA unless in the pH range of 3-6 where it is slightly incorporated. Thus, pH 7.2 was cho- sen in order to have the lowest ratio of the aspe- cific retention of [14C]methoxyamine to the label- ling of specific reactive sites in alkylated-de- purinated DNA.

Fig. 2 represents the amount of [14C]metho- xyamine taken up by untreated and alkylated-de- purinated DNAs as a function of the reagent concentration after a 30-min incubation at 37°C, pH 7.2. The incorporation of [14C]methoxyamine in untreated DNA remains negligible even with increasing reagent concentrations. Alkylated-de- purinated DNA retains a significant amount of radioactivity which depends on the reagent con- centration. When [14C]methoxyamine concentra- tion approaches 5 mM, the incorporation curve levels off. It thus seems that the reactive sites specific to alkylated-depurinated DNA are pro- gressively occupied. At a 5 mM concentration, these sites have taken up 57 molecules of [t4C]methoxyamine per 1000 DNA nucleotides.

The aspecific retention of free [14C]methoxya- mine on the glass fiber disks used to measure the incorporation of the radioactive reagent in acid-in- soluble material increases continuously with the

412

6 O

= ~ 2 0

° ~ °""'~o--o.o o.o~.~." o\ 0 \

I - - ~ ~ : ~ _ . l - - J 2 4 6 8

pH

O \ O \

10 12

Fig. 1. Influence of pH on the reaction of [~4C]methoxyamine with alkylated-depurinated DNA. The reaction mixtures con- tained 200 /.~g/ml of either untreated (o) or alkylated-de- p urinated (O) DNA. They were incubated with 5 m M [~4C]methoxyamine at 37°C for 30 rain. The indicated pH values were reached after extensive dialysis against l0 m M NaC1/10 m M potassium phosphate at various pH values or after the addition of HCI or NaOH. The incorporation of [J'~C]methoxyarnine in DNA was measured as described.

6 C

"o O

6 6 0 ~ 4 6

~--JD--e e l ..(I Time(:m) _e 0 6o 120 180

Time (min /

Fig. 3. Incorporation of [~4C]methoxyamine in alkylated-de- purinated D N A as a function of time. The reaction mixtures contained 200 ~ g / m l of either untreated or alkylated-de- purinated DNA. The incubation was at 37°C, pH 7.2 with 1 m M (®) or 5 mM (O) [14C]methoxyamine for the alkylated- depurinated D N A and with 5 m M (O) for the untreated DNA. At the indicated times, the incorporation of [~4C]methoxya- mine in D N A was measured as described. Insert of Fig. 3. Number of unreacted sites (F) as a function of time. The number of unreacted sites (F) is calculated by subtracting the number of reacted sites from the number of total sites, both were estimated from the amount of [ laC]methoxyamine in- corporated. The graph represents the variation of In F with time.

'E6O

o 5

0 0 0

Methoxyamine (raM1

Fig. 2. Incorporation of [t4C]methoxyamine in alkylated-de- purinated D N A as a function of the reagent concentration. The reaction mixtures contained 200 F g / m l of either untreated or alkylated-depurinated D N A in 0.1 M borate buffer (pH 7.2). They were incubated at 37°C for 30 rain with the indicated concentrations of [14C]methoxyamine. The quantities of [34C}methoxyamine bound to untreated (e) and to alkylated- depurinated (O) D N A were measured after precipitation in acid as described.

increase in reagent concentration. To maintain the background at an acceptable level, a concentration of [lnC]methoxyamine as low as 5 mM, which allows the complete occupation of the reactive sites in a reasonable time, has to be used in the assay.

Fig. 3 shows the fixation of [14C]methoxyamine by untreated and alkylated-depurinated DNAs at

6o

E c

=E- 5

0

O

o / °

0 2~ 5b i~ - 160- % Alkylated. flepurinated ONA

Fig. 4. Incorporation of [~4C]methoxyamine as a function of the reactive sites in alkylated-depurinated DNA. The reaction mixtures contained various amounts of reactive sites obtained by mixing various proportions of untreated and alkylated-de- purinated D N A at the concentration of 115 # g / m l . The reac- tions were performed under standard conditions: 5 mM [laC]methoxyamine/0.1 M borate buffer (pH 7.2), 30 min at 37°C. The incorporation of [14C]methoxyarnine in DNA was measured as described.

413

two different reagent concentrations as a function of time (incubation 37°C, pH 7.2). A prolonged incubation with 5 mM [~4C]methoxyamine does not cause a significant labelling of untreated DNA. On the contrary, the fixation of [t4C]methoxya- mine by alkylated-depurinated DNA increased with time and the reaction velocity depends on the reagent concentration. The incorporation curves corresponding to 1 and 5 mM methoxyamine con- verge to the same maximum. This suggests that the reaction of alkylated-depurinated DNA with methoxyamine goes to completion after a suffi- ciently long incubation time. The insert of Fig. 3 gives the logarithm of the number of specific reac- tive sites which have not yet reacted at a metho- xyamine concentration of 5 mM as a function of time. There is a linear relationship the slope of which allows the determination of a kinetic con- stant of 34.4 M - 1. rain- ~.

The velocity of the reaction depends on the temperature; proceeding 8-times more rapidly at 37°C than at 0°C (data not shown).

In conclusion, a 30-min incubation at 37°C with 5 mM methoxyamine which leads to a com- plete reaction of all the reactive sites was chosen for the standard assay. Under these conditions, the

1.0(

20 r ime (rain)

0 2 4 6 24 Time {hour)

Fig. 5. Stability of the complex resulting from the methoxya- mine reaction with alkylated-depurinated DNA. Alkylated-de- purinated D N A (200/Lg/ml) was incubated with [14C]metho- xyamine under the standard conditions. The reagent excess was eliminated by extensive dialysis against 10 m M NaCI /10 mM potassium phosphate (pH 7.2). The resulting D N A contained 55 molecules (No) of []4C]methoxyamine bound per 1000 nucleotides. The incubation was at 37°C with (O) or without (e) 10 m M unlabelled methoxyamine. At the indicated times, the amount of [a4C]methoxyamine (N t) still bound to D N A was checked after precipitation in acid as described.

amount of methoxyamine bound is proportional to the number of specific reactive sites in alkylated- depurinated DNA (Fig. 4).

To explore the stability of the complex, after reacting the alkylated-depurinated DNA with [14C]methoxyamine, the free labelled reagent was dialyzed away and replaced or not by 10 mM unlabelled methoxyamine. The results are shown in Fig. 5. In the absence of unlabelled methoxya- mine, the radioactive complex is practically stable. In the presence of 10 mM methoxyamine at 37°C and pH 7.2, the alkylated-depurinated DNA pro- gressively lost the bound [~4C]methoxyamine. A semi-logarithmic plot of the results yields a straight line (insert of Fig. 5) from which a half-life of 42 min for the bound []4C]methoxyamine can be deduced, corresponding to a dissociation rate con- stant of 0.0167 rain -~. This rate constant of dis- sociation depends on the concentration of un- labelled methoxyamine (data not shown) suggest- ing that the loss of radioactivity from the al- kylated-depurinated DNA results from the sub- stitution of bound [14C]methoxyamine by un- labelled methoxyamine.

Absence of degradation of alkylated-depurinated DATA by methoxyamine

Table I shows that incubation of alkylated-de- purinated [3H]DNA with methoxyamine at 37°C, pH 7.2, does not increase the acid-soluble radioac- tivity.

TABLE I

STABILITY OF A L K Y L A T E D - D E P U R I N A T E D D N A IN THE PRESENCE OF M E T H O X Y A M I N E

Alkylated-depurinated [3H]DNA (100 / tg /ml ) was incubated at 37°C with or without the addition of 0.2 M N a O H or 5 m M methoxyamine. At the indicated times, the acid-soluble radio- activity was measured as described.

Inc. time Acid-soluble fraction

(rain) NaOH Methoxyamine

- + - +

15 1.9 36 2.2 2.6 30 - - - - 2.3 2.6 60 - - - - 2.4 2.6

120 - - - - 2.7 2.7

414

6O

0

S 0 - 0 0 k~O

.L, 0 60 120 180 240

Time (min I

Fig. 6. Reaction of [ 14 C]methoxyamine with a synthetic polymer containing apyrimidinic sites. The reaction mixtures contained 174 /~g/ml of either (dA)2a0-(dT,[3H]dU)2so treated with uracil-DNA glycosylase or untreated (dA)230' (dT, dU)2s0. The incubation was at 37°C, pH 7.2 with 1 mM (®) or 5 mM (O) [14C]methoxyamine for the polymer containing apyrimidinic sites and with 5 mM (e) for the polymer control. At the indicated times, the incorporation of [ f4 C]methoxyamine in the polymer was measured as described.

Reaction of methoxyamine with a synthetic polymer containing apyrimidinic sites

Fig. 6 shows that [14C]methoxyamine reacts with the apyrimidinic sites introduced by uracil- D N A glycosylase in a uracil-containing polyde- oxyribonucleotide. With a 5 mM reagent con- centration, the reaction is quick and reaches a maximum corresponding to the saturation of all the apyrimidinic sites after a 30-min incubation at 37°C. As for alkylated-depurinated DNA, the re- action with 1 mM methoxyamine requires a longer time to go to completion. Moreover, the uracil- containing polymer control does not retain [14C]methoxyamine. From the insert of Fig. 6 which gives the logarithm of the unreacted sites at a methoxyamine concentration of 5 mM as a function of time, a kinetic constant of 58.3 M - ] • min- I was calculated.

Discussion

Until now AP sites were quantified by the determination of strand breaks resulting from the rupture of an adjacent phosphoester bond. The most widely used alkaline treatment was, in a few

cases, replaced by AP endodeoxyribonuclease hy- drolysis [14]. The labelled DNA-containing AP sites is either a heterogeneous population of mole- cules extracted from E. coli, or a homogeneous one of linear T7 phage DNA, or circular and super- twisted RF-I molecules of various origins. The number of breaks introduced can be estimated in labelled E. coli DNA by the acid-soluble radioac- tivity [13], in T7 DNA by sedimentation through neutral sucrose gradient [15] and in RF-I DNA by filtration through nitrocellulose or agarose gel electrophoresis [16]. These two-step methods only determine intact AP sites, i.e., those which are not already associated with strand breaks. Some are not very sensitive: the break lrequency must be high to have a measurable acid-soluble radioactiv- ity. Others, using RF-I DNA, can explore only a very narrow range of break frequencies, while the use of T7 DNA and sedimentation analysis is long and tedious.

The objective of this work is to present a method for the direct assay of AP sites, intact or associated with breaks, by the addition of a specific labelled reagent. As the deoxyribose residue at the AP site is most often in an open form with a free aldehyde group, a specific reagent for this function was selected. Aldehyde reagents containing - N H 2 groups, were found to degrade apurinic acid and were used for this purpose. Livingston [17] noticed that degradation was caused by compounds con- taining - N H - N H 2 and not by those containing - O - N H 2. The mechanism of degradation pos- tulated by Livingston is summarized in Fig. 7: the formation in two steps of a Schiff base enhances the departure of a proton from C~ and makes the phosphodiester bond attached to C~ labile (/3- elimination). With a reagent containing - O - NH2, the first addition product is stabilized by the formation of a hydrogen bond between the - O H group which appears on C' 1 and the oxygen of the reagent (Fig. 8), preventing the loss of water lead- ing to the Schiff base [18]. Coombs and Livingston showed that [t4C]methoxyamine reacts quantita- tively with the aldehyde groups of apurinic acid. Taking advantage of this information, we investi- gated the possible use of [14C]methoxyamine for counting AP sites in double-stranded DNA.

The present work shows that alkylated-de- purinated DNA which contains apurinic sites, and

415

I I t ' I {

6 6 o o I I I I

OH2 CH2 CH2 ,.OH

o o o

I 1 ' I

Fig. 7. Hypothetical pathway for the degradation of DNA containing AP sites after reaction with an aldehyde reagent R-NH 2.

Fig. 8. Hypothetical hydrogen bond stabilization of the addition product of NH2OR on AP sites.

a synthetic double-stranded polymer containing apyrimidinic sites, fix much more [14C]metho- xyamine than the corresponding controls (un- treated DNA or the homologous polymer without apyrimidinic sites).

The alkylated-depurinated DNA seems to con- tain only one kind of specific reactive sites: a straight line is obtained when the number of specific reactive sites which have not yet reacted with methoxyamine is plotted against time of ex- posure to the reagent; on the other hand, the release of bound [t4C]methoxyamine in the pres- ence of a fixed concentration of free unlabelled methoxyamine is a simple decreasing exponential. Saturation of these specific binding sites results in the fixation of 57 molecules of [14 C]methoxyamine per 1000 nucleotides. According to Paquette et al. [13], the depurination of the alkylated DNA, un- der the conditions described in Materials and Methods, is responsible for the appearance of about 50 apurinic sites per 1000 nucleotides. It is thus possible that the specific reactive sites in alkylated-depurinated DNA are, indeed, only the apurinic sites.

With the synthetic polymer containing apyrimidinic sites, there is no doubt that the specific binding of [t4C]methoxyamine is at the apyrimidinic sites since it is the only difference with the homologous control.

Conditions were sought having the highest ratio of specific to unspecific bindings of [14C]metho- xyamine. A 30-min incubation at 37°C with 5 mM [14C]methoxyamine at pH 7.2 was chosen. Under these conditions, nearly all the specific reactive

sites in alkylated-depurinated DNA and in a syn- thetic polymer containing apyrimidinic sites react with [t4C]methoxyamine. Moreover, the reaction rates for apurinic and apyrimidinic sites, although they are in very different macromolecules, are nearly the same.

With the chosen conditions, there is a linear relationship between the number of AP sites and the amount of bound [~4C]methoxyamine. Con- firming the work of Livingston, we have shown that the reaction of methoxyamine with AP sites is not followed by the formation of a strand break so that the bound [14C]methoxyamine can be mea- sured by the simple determination of the acid-in- soluble radioactivity.

The reaction of [14 C]methoxyamine with specific reactive sites in alkylated-depurinated DNA goes to completion when enough time is left for the reaction. We have shown that the reaction is irre- versible: the complex cannot be hydrolyzed at pH 7.2. Subsequent incubation with unlabelled methoxyamine is however followed by a release of radioactivity: if the complex cannot be hydro- lyzed, it seems that it can be attacked by free methoxyamine, the reaction leading likely to the substitution of the bound reagent molecule.

In conclusion, labelled methoxyamine appears as an excellent and specific reagent for both apurinic and apyrimidinic sites. A radioactivity determination enables the counting of the AP s i t e s whether associated with breaks or not. This simple technique may serve to elucidate several questions concerning the formation of AP sites and their repair. Some of the possible applications may be:

416

the analysis of damage introduced in DNA by physical and chemical agents - in particular those with carcinogenic and mutagenic properties - the characterization of substrates for AP endo- deoxyribonucleases, the discovery of new DNA glycosylases and the dosage of their activity, the investigation on the mechanism of chemical or enzymatic rupture of the phosphoester bond ad- jacent to an AP site, etc.

Acknowledgements

We thank Dr. W.G. Verly for reading the manuscript. This publication is contribution 2010 of the Programme Biology, Radiation Protection and Medical Research, Directorate General XII for Research, Science and Education of the Com- mission of the European Communities. M.L. is the recipient of a fellowship from the commission of the European Communities at the Joint Research Centre of Ispra.

References

1 Lindahl, T. and Ljungquist, S. (1975) in Molecular Mecha- nisms for Repair of DNA (Hanawalt, P.C. and Setlow, R.B., eds.), Part A, pp. 31-38, Plenum Press, New York

2 Lindahl, T. (1979) in Progress in Nucleic Acids Research and Molecular Biology (Cohn, W.E., ed.), Vol. 22, pp. 135-192, Academic Press, New York

3 Lindahl, T. and Nyberg, B. (1972) Biochemistry 11, 3610-3618

4 Lindahl, T. and Karlstr6m, O. (1973) Biochemistry 12, 5151-5154

5 Zoltewicz, J.A., Clark, D.F., Sharpless, T.W. and Grahe, G. (1970) J. Am. Chem. Soc. 92, 1741-1750

6 Dunlap, B. and Cerutti, P. (1975) FEBS Lett. 51, 188-190 7 Kochetkov, N.K., Budovsky, E.I, Sverdlov, E.D., Sym-

ukova, N.A., Turchinsky, M.F. and Shibaev, V.N. (1972) in Organic Chemistry of Nucleic Acids (Kochetkov, N.K. and Budovsky, E.I., eds.), Part-B pp. 381-423, Plenum, New York

8 Caradonna, S.J. and Cheng, Y,C. (1982) Mol. Cell. Bio- chem. 46, 49-63

9 Muller, W.E.G. and Zahn, R.K. (1977) in Progress in Nucleic Acids Research and Molecular Biology (Cohn, W.E., ed.), Vol. 20, pp. 21-57, Academic Press, New York

10 Talpaert-Borl~, M., Campagnari, F. and Creissen, D.M. (1982) J. Biol. Chem. 257, 1208-1214

11 Talpaert-Borl~, M., Clerici, L. and Campagnari, F. (1979) J. Biol. Chem. 254, 6387-6391

12 Talpeart-Borl~, M. and Liuzzi, M. (1982) Eur. J. Biochem. 124, 435-440

13 Paquette, Y., Crine, P. and Verly, W.G. (1972) Can. J. Biochem. 50, 1199-1209

14 Brent, T.P., Teebor, G.W. and Duker, N.J. (1978) in DNA Repair Mechanisms (Hanawalt, P.C., Friedberg, E.C. and Fox, C.F., eds.), pp. 19-22, Academic Press, New York

15 Crine, P. and Verly, W.G. (1976) Anal. Biochem. 75, 583-595

16 Goffin, C. and Verly, W.G. (1982) Eur. J. Biochem. 127, 619-623

17 Livingston, D.C. (1964) Biochim. Biophys. Acta 87,538-540 18 Coombs, M.M. and Livingston, D.C. (1969) Biochim. Bio-

phys. Acta 174, 161-173