BIOLOGICAL MASS SPECTROMETRY, VOL. 20, 665-668 (1991)
Direct Observation of the Alkylation Products of Deoxyguanosine and DNA by Fast Atom Bombardment Mass Spectrometry Grigorii V. Andrievsky Kharkov Research Institute for Therapy, UkrSSR Ministry of Health, 2 Postyshev Ave., Kharkov, USSR
Leonid F. Sukhodub, Tatyana, L. Pyatigorskaya, Oleg, A. Boryak, Olga Yu. Limanskaya and Vadim S. Shelkovskyt Institute for Low Temperature Physics and Engineering, UkrSSR Academy of Sciences, 47 Lenin Ave., Kharkov 310164, USSR
By the methods of fast atom bombardment (FAB) mass spectrometry, thin-layer chromatography and ultraviolet absorption spectroscopy adducts have heen studied which are formed by an antitumour alkylating drug thiotepa both in a model system, containing only deoxyguanosine (dGuo), and in DNA. Analysis of the model reaction mixture (dGuo + thiotepa) by FAB mass spectrometry permitted observation of adducts dCuo thiotepa, 2dGuo thiotepa, and also the products of their further modification in solution, which occurs by hydrolysis of the glycosidic bond and also by opening of the imidazole ring. In the case of DNA FAB mass spectrometry made it possible to characterize adducts of thiotepa with guanosine (Gua) and adenosine (Ade) without their preliminary purification. The site of alkylation of Gua in both dGuo and DNA is N7, and that of Ade in DNA is N3. The application of the results to the study of the molecular mechanism of the antitumour action of thiotepa is discussed.
INTRODUCTION
A number of investigations have been carried out on the interaction of a well-known antitumour alkylating drug, N,N’,N”-triethylenethiophosphoramide(thiotepa, A; see Fig. 2) with DNAi4 and its components.” By the methods of heat denaturation, electron microscopy2 and vi~cometry,~ the kinetics of alkylation in oitro and alterations of the DNA secondary structure induced by thiotepa have been studied. As for the molecular mechanisms of interaction, those were investigated only at a monomer level. Ultraviolet (UV) ab~orption~.~.’ and nuclear magnetic resonance (NMR)4*6data obtained with dGMP, GMP or dGuo indicated that the principal site of alkylation in Gua was the N7 position. However, it appeared that thiotepa produced a very complex reaction mixture, from which only one compound-N7-monothiotepa derivative of dGMP (GMP)--could be purified and identified. The methods applied did not show the products with modified bases in the case of other nucleotides (nucleosides) owing to low yields of reaction products and some other peculiarities of thiotepa-containing systems. An analysis of DNA alkylation by thiotepa (as well as by other reagents containing aziridine groups) is further complicated by the destruction of the reaction products during acid hydrolysis, by impeding enzymatic hydrolysis due to the interaction of free aziridine groups with nucleases’ and by difficulties connected with purifi-
t Author to whom correspondence should be addressed. 0 Ukrainian Branch of the All Union Copyright Agency
0 1991 by John Wiley & Sons, Ltd.
cation of individual products from an extremely complex reaction mixture. During recent years soft ionization mass spectrometry has been applied in a continuously increasing area of nucleic acid research, including interactions of nucleic acid components with In particular, some sites of alkylation of all four DNA bases by thiotepa were identified by field ionization mass spectrometry using a set of methylated bases.’ In the present paper we show that the use of fast atom bombardment (FAB) mass spectrometry facilitates considerably the analysis of the products of the reaction between DNA or its components and thiotepa, omitting the isolation of individual products. The high sensitivity of the method permitted deduction of a detailed scheme of the formation and further hydrolysis of dGuo thiotepa adducts in solution and also identification for the first time of Gua- and Ade- adducts formed in DNA alkylated by thiotepa. EXPERIMENTAL Chemicals Thiotepa was supplied by the OIaynsky ChemicoPharmaceutical Works, Riga, USSR, and recrystallized twice from benzenehexane mixtures.” dGuo was purchased from Serva, Germany. DNA from chicken blood (Reanal, Hungary) was additionally purified by centrifugation at 105000 x g for 2 h, followed by dialysis against 0.01 M NaCl. Received 31 October 1990 Revised 8 July 1991
G . V. ANDRIEVSKY ET AL.
666
Akylation of dGuo
Mass spectrometry
dGuo (28.2 mg, 0.1 mmol) and thiotepa (94.5 mg, 0.5 mmol) were incubated in a water-acetone mixture (1 :1, 3 ml) at 70-75 "C for 2.5 h. In the course of reaction the pH increased from 6.8 to 7.8 and when the reaction was over it was adjusted to 6.0 with 0.1 M HC1. Acetone was removed by evaporation of the mixture under vacuum at 40"C, then the mixture was diluted with water, and excess thiotepa was washed with benzenechloroform (1 : 1,5 x 5 ml) mixture.
FAB mass spectra were obtained with an MI-120IE mass spectrometer (Electron Works, USSR) in a glycerol matrix at an Ar primary beam energy of 3.5 keV.
Akylation of DNA Forty millilitres of the DNA solution in 0.01 M NaCl with DNA concentration 1.43 mg ml-' were mixed with 10 ml of an aqueous solution of 600 mg thiotepa (the final molar ratio of thiotepa to DNA nucleotides being 20) and incubated at 37°C for 140 h. (The pH increased from 6.6 up to 7.8 during the incubation time.) The solution was then evaporated under vacuum at 40°C up to a volume of 4-5 ml, excess thiotepa was removed as described above and the evaporation was repeated up to 3 ml. Chromatography Thin-layer chromatography (TLC) was performed on silica gel plates (Silufol UV-254, Kavalier, Czechoslovakia) using n-butanol-isopropanol-acetic acid-H,O (65 :10: 10:30; solvent 'a'), n-butanolacetone-H,O (2 : 1 :1; solvent 'b') or isopropanol-25% aqueous ammonia (3:2, solvent 'c'). A Sephadex G-25 (Pharmacia Fine Chemicals, Sweden) column (26 x 2.5) was eluted with water at a flow rate of 1.5 ml min-'. The optical density of the eluate was monitored at 260 nm.
U V absorption spectroscopy Spectra of the materials extracted from the spots on TLC plates with water were recorded using a Specord UV VIS spectrophotometer (Carl Zeiss, Germany) at pH 1,7 and 12.
I,% 1
RESULTS Interaction of thiotepa with dGuo Separation of the reaction mixture (dGuo + thiotepa) by TLC in different solvents gave, besides the main spot of the starting material (dGuo), two luminescent spots (R, 0.40 and O.Ol),t UV spectra being characteristic of N7-alkyl derivatives of dGuo,6-8.12 and nonluminescent spots (R, 0.78), UV spectra being characteristic of N7-alkyl derivatives of and (R, 0.27 and 0.lo), UV spectra being characteristic of N7-alkyl derivatives of dGuo, in which imidazole rings were opened in alkaline media.7.8.'2 The total yield of the reaction products was about 1.5%. Thus, the spectral data indicate that under the employed reaction conditions Gua in dGuo is alkylated only at the N7-position. Figure 1 shows a typical FAB mass spectrum of the total reaction mixture (dGuo + thiotepa). The mass spectral data (m/z values for the peaks and their isotopic distribution) together with the UV absorption results made it possible to elucidate the structure of the reaction products and the pathways of their further hydrolysis in solution (Fig. 2). From the scheme shown in Fig. 2 it can be seen that the primary product of the reaction, C, is unstable under the employed conditions of alkylation. Its transformation occurs by two main pathways: by hydrolysis of the glycosidic bond with the formation of product D, and by opening of the imidazole ring (which is facilitated by the increase of pH of the reaction mixture from 6.8 to 7.8 during its incubation) with the formation of R, values are given for system 'a'; in this particular case, in order to improve the separation of the spots, chromatography was conducted continuously for 20 h.
=1
I 150
+
250
I .
[B+Hj
300
,100---------1
+
350450
550
GOOAA
Figure 1 . FAB mass spectrum of a total reaction mixture (dGuo + thiotepa).
ALKYLATION PRODUCTS OF DEOXYGUANOSINE AND DNA BY FABMS
H,N(CH,hHNP
K
.s '
L -
=
667
K
YNa,
Figure 2. Products of interaction between thiotepa and dGuo and the pathways of their further transformation in solution.
products H, I, J and K. Though products H, and H, were not observed by FAB mass spectrometry, their formation is proved by the existence of the products of their further modification: I, J and K. Similarly, product G could be formed only from product F which, however, is not observed in the mass spectra in the form of a doubly charged ion (m/z 362.5). In order to concentrate product F (presumably the luminescent material with R, 0.01 in system 'a') the following procedure was employed. Sixteen milligrams of silica gel LC 5/40 (Chemapol, Czechoslovakia) were added to the reaction mixture, which had been preliminarily washed from excess thiotepa, and it was extracted three times with solvent 'b' for TLC. The remaining adsorbed products were rapidly extracted with water and dried. Their mass spectra showed the main peak at m/z 725, which corresponds to a singly charged ion of an adduct of two dGuo molecules and one thiotepa molecule (dimer F). The loss of one charge '
:I
is consistent with the well-known behaviour of dications in FAB.I4 Interaction of thiotepa with DNA
DNA was alkylated and washed from excess thiotepa as described under Experimental. The solution of the alkylated DNA was then heated in a water bath at 100°C for 1 h, which led to a detachment of alkylated purines. The resultant mixture was applied to a Sephadex G-25 column and three main fractions were eluted. Fraction 1 (total volume Y = 82 ml, retention volume u = 85 ml) contained oligonucleotides, which were produced owing to DNA fragmentation at apurinic sites. Fractions I1 and I11 (V = 35 ml and 55 ml, u = 165 ml and 210 ml, respectively) were dried and studied by FAB mass spectrometry. The partial mass spectra, which show the main relevant peaks, are given in Fig. 3.
I. %
0
300
1 b+HI+ 34 1
...i..
300
350 M i Z
Figure 3. Fragments of FAB mass spectra of fractions II (b) and Ill (a), isolated from the reaction mixture (DNA + thiotepa) by gel filtration (see text for details).
G. V. ANDRIEVSKY ET AL.
668
Separation of fractions I1 and 111 by TLC in both cases produced only one the most intensive spot with R, in systems 'a' and 'b' 0.52 and 0.59 (fraction 11), 0.26 and 0.72 (fraction III), respectively. UV absorption spectra of the materials extracted from these spots with water were recorded and compared with known UV spectra of alkyl derivatives of Ade and d G ~ o . ~ * ' ~The , ' ' combination of FAB mass spectrometry and UV spectroscopic data indicates that the main compounds which are present in fractions I1 and I11 are the N7monothiotepa derivative of Gua (D, m/z 343, yield 8.3%) and the N3-monothiotepa derivative of Ade (L, mlz 324, yield 3.4%), respectively. It has been verified that no material with UV spectra different from those of the two main products was present in the fractions under investigation. DISCUSSION ___ - _ _ _ ~
_ .
The results of the present investigation show that FAB mass spectrometry allows detection (and, with the aid of UV absorption spectroscopy, identification) of the adducts of nucleotides and alkylating drugs directly in a complex reaction mixture, the content of an individual
product being in our case as small as 0.5% (1 pg). The method also allows detection and identification of DNA adducts with drugs, omitting the step of separation of primary individual interaction products, which may be especially difficult and even impossible in the case of drugs containing more than one aziridine group. In this work, for the first time, we managed to identify the sites of DNA alkylation by thiotepa in uitro. It has been confirmed that thiotepa reacts with the Gua residue in both dGuo and DNA only at the N7 position. The detailed analysis of the model system (dGuo + thiotepa) considerably simplified the identification of the Gua adduct with thiotepa in DNA. The results with dGuo permit also the prediction of the fate of an alkylated dGMP residue in DNA in uiuo. Besides this, the observation of dimers 2dGuo thiotepa (F) in the reaction mixture (dGuo + thiotepa) shows a possible pathway of formation of intramolecular cross-finks in DNA between two opposite Gua residues. Although the question needs further study, including molecular modelling, it may be noted that DNA cross-linking by thiotepa was also suggested earlier on the basis of indi~ type of rect data obtained in uitro' and in ~ i u o . ' * 'This lesion may be of essential importance in the molecular mechanism of antitumour action of thiotepa and other diethyleneimides of phosphoric acids.
REFERENCES 1. T. G. Sjakste, N. 1. Sjakste and A. V. Lichtenstein, Biokhimia 48.1 37 (1 983). 2. T. L. Pyatigorskaya, 0. Yu. Zhilkova, L. M . Muravyova and L. F. Sukhodub, Molekulyarnaya Biologiya (Eng. Transl.) 20, 334 (1986). 3. T. L. Pyatigorskaya, 0. Yu. Zhilkova, E. S. Ruzaeva, S. V. Kornilova and L. F. Sukhodub, in Modern Problems of Experimental Chemotherapy of Tumours, Part 2, p. 92. Proceedings of the Third All Union Conference, Chernogolovka (1987). 4. A. M . Serebryany, G. V. Andrievsky, A. R. Becker, L. A. Sibeldina and 0. L. Sharova, Bioorganicheskaya Khimia 13, 786 (7987). 5. L. F. Sukhodub, V. S. Shelkovsky, M . V. Kosevich, T. L. Pysatigorskaya and 0. Yu. Zhilkova, Biomed. Environ. Mass Spectrorn. 13, 167 (1986). 6. A. M . Serebryany, G. V. Andrievsky, A. R. Becker, L. A: Sibeldina and M . 1. Povolotskaya, Bioorganicheskaya Khimia 12. 499 (1 986).
7. K. Hemminski, S. Kallama and K. Falck, Acta Pharmacol. Toxicol. 53,421 (1983). 8. G. V. Andrievsky and A. M . Serebryany, unpublished results. 9. C. Fenselau, J. Nat. Prod. 47, 21 5 (1 984). 10. D. L. Slowikovski and K. H. Schram, Nucleosides and Nucleotides 4, 309 (1 985). 11. T. L. Pyatigorskaya, 0. Yu. Zhilkova, V. S. Shelkovsky, N. M. Arkhangelova, A. I. Grizodub and L. F. Sukhodub. Biomed. Environ. MassSpectrom. 14,143 (1987). 12. C. C. Price, G. M . Caucher, P. Koneru, R. Shibakava, J. R. Sowa and M . Yamaguchi, Biochim. Biophys. Acta 166, 327 (1968). 13. E. Institoris and J. Tamas, Chem.-Biol. interact. 47. 133 (1 983). 14. 0. N. Heller, J. Yergey and R. J. Cotter, Anal. Chem. 55, 1310 (1 983). 15. J. J. McCann. T. M . Lo and D. A. Webster, Cancer Res. 31, 1573 (1971).
Direct Observation of the Alkylation Products of Deoxyguanosine and DNA by Fast Atom Bombardment Mass Spectrometry Grigorii V. Andrievsky Kharkov Research Institute for Therapy, UkrSSR Ministry of Health, 2 Postyshev Ave., Kharkov, USSR
Leonid F. Sukhodub, Tatyana, L. Pyatigorskaya, Oleg, A. Boryak, Olga Yu. Limanskaya and Vadim S. Shelkovskyt Institute for Low Temperature Physics and Engineering, UkrSSR Academy of Sciences, 47 Lenin Ave., Kharkov 310164, USSR
By the methods of fast atom bombardment (FAB) mass spectrometry, thin-layer chromatography and ultraviolet absorption spectroscopy adducts have heen studied which are formed by an antitumour alkylating drug thiotepa both in a model system, containing only deoxyguanosine (dGuo), and in DNA. Analysis of the model reaction mixture (dGuo + thiotepa) by FAB mass spectrometry permitted observation of adducts dCuo thiotepa, 2dGuo thiotepa, and also the products of their further modification in solution, which occurs by hydrolysis of the glycosidic bond and also by opening of the imidazole ring. In the case of DNA FAB mass spectrometry made it possible to characterize adducts of thiotepa with guanosine (Gua) and adenosine (Ade) without their preliminary purification. The site of alkylation of Gua in both dGuo and DNA is N7, and that of Ade in DNA is N3. The application of the results to the study of the molecular mechanism of the antitumour action of thiotepa is discussed.
INTRODUCTION
A number of investigations have been carried out on the interaction of a well-known antitumour alkylating drug, N,N’,N”-triethylenethiophosphoramide(thiotepa, A; see Fig. 2) with DNAi4 and its components.” By the methods of heat denaturation, electron microscopy2 and vi~cometry,~ the kinetics of alkylation in oitro and alterations of the DNA secondary structure induced by thiotepa have been studied. As for the molecular mechanisms of interaction, those were investigated only at a monomer level. Ultraviolet (UV) ab~orption~.~.’ and nuclear magnetic resonance (NMR)4*6data obtained with dGMP, GMP or dGuo indicated that the principal site of alkylation in Gua was the N7 position. However, it appeared that thiotepa produced a very complex reaction mixture, from which only one compound-N7-monothiotepa derivative of dGMP (GMP)--could be purified and identified. The methods applied did not show the products with modified bases in the case of other nucleotides (nucleosides) owing to low yields of reaction products and some other peculiarities of thiotepa-containing systems. An analysis of DNA alkylation by thiotepa (as well as by other reagents containing aziridine groups) is further complicated by the destruction of the reaction products during acid hydrolysis, by impeding enzymatic hydrolysis due to the interaction of free aziridine groups with nucleases’ and by difficulties connected with purifi-
t Author to whom correspondence should be addressed. 0 Ukrainian Branch of the All Union Copyright Agency
0 1991 by John Wiley & Sons, Ltd.
cation of individual products from an extremely complex reaction mixture. During recent years soft ionization mass spectrometry has been applied in a continuously increasing area of nucleic acid research, including interactions of nucleic acid components with In particular, some sites of alkylation of all four DNA bases by thiotepa were identified by field ionization mass spectrometry using a set of methylated bases.’ In the present paper we show that the use of fast atom bombardment (FAB) mass spectrometry facilitates considerably the analysis of the products of the reaction between DNA or its components and thiotepa, omitting the isolation of individual products. The high sensitivity of the method permitted deduction of a detailed scheme of the formation and further hydrolysis of dGuo thiotepa adducts in solution and also identification for the first time of Gua- and Ade- adducts formed in DNA alkylated by thiotepa. EXPERIMENTAL Chemicals Thiotepa was supplied by the OIaynsky ChemicoPharmaceutical Works, Riga, USSR, and recrystallized twice from benzenehexane mixtures.” dGuo was purchased from Serva, Germany. DNA from chicken blood (Reanal, Hungary) was additionally purified by centrifugation at 105000 x g for 2 h, followed by dialysis against 0.01 M NaCl. Received 31 October 1990 Revised 8 July 1991
G . V. ANDRIEVSKY ET AL.
666
Akylation of dGuo
Mass spectrometry
dGuo (28.2 mg, 0.1 mmol) and thiotepa (94.5 mg, 0.5 mmol) were incubated in a water-acetone mixture (1 :1, 3 ml) at 70-75 "C for 2.5 h. In the course of reaction the pH increased from 6.8 to 7.8 and when the reaction was over it was adjusted to 6.0 with 0.1 M HC1. Acetone was removed by evaporation of the mixture under vacuum at 40"C, then the mixture was diluted with water, and excess thiotepa was washed with benzenechloroform (1 : 1,5 x 5 ml) mixture.
FAB mass spectra were obtained with an MI-120IE mass spectrometer (Electron Works, USSR) in a glycerol matrix at an Ar primary beam energy of 3.5 keV.
Akylation of DNA Forty millilitres of the DNA solution in 0.01 M NaCl with DNA concentration 1.43 mg ml-' were mixed with 10 ml of an aqueous solution of 600 mg thiotepa (the final molar ratio of thiotepa to DNA nucleotides being 20) and incubated at 37°C for 140 h. (The pH increased from 6.6 up to 7.8 during the incubation time.) The solution was then evaporated under vacuum at 40°C up to a volume of 4-5 ml, excess thiotepa was removed as described above and the evaporation was repeated up to 3 ml. Chromatography Thin-layer chromatography (TLC) was performed on silica gel plates (Silufol UV-254, Kavalier, Czechoslovakia) using n-butanol-isopropanol-acetic acid-H,O (65 :10: 10:30; solvent 'a'), n-butanolacetone-H,O (2 : 1 :1; solvent 'b') or isopropanol-25% aqueous ammonia (3:2, solvent 'c'). A Sephadex G-25 (Pharmacia Fine Chemicals, Sweden) column (26 x 2.5) was eluted with water at a flow rate of 1.5 ml min-'. The optical density of the eluate was monitored at 260 nm.
U V absorption spectroscopy Spectra of the materials extracted from the spots on TLC plates with water were recorded using a Specord UV VIS spectrophotometer (Carl Zeiss, Germany) at pH 1,7 and 12.
I,% 1
RESULTS Interaction of thiotepa with dGuo Separation of the reaction mixture (dGuo + thiotepa) by TLC in different solvents gave, besides the main spot of the starting material (dGuo), two luminescent spots (R, 0.40 and O.Ol),t UV spectra being characteristic of N7-alkyl derivatives of dGuo,6-8.12 and nonluminescent spots (R, 0.78), UV spectra being characteristic of N7-alkyl derivatives of and (R, 0.27 and 0.lo), UV spectra being characteristic of N7-alkyl derivatives of dGuo, in which imidazole rings were opened in alkaline media.7.8.'2 The total yield of the reaction products was about 1.5%. Thus, the spectral data indicate that under the employed reaction conditions Gua in dGuo is alkylated only at the N7-position. Figure 1 shows a typical FAB mass spectrum of the total reaction mixture (dGuo + thiotepa). The mass spectral data (m/z values for the peaks and their isotopic distribution) together with the UV absorption results made it possible to elucidate the structure of the reaction products and the pathways of their further hydrolysis in solution (Fig. 2). From the scheme shown in Fig. 2 it can be seen that the primary product of the reaction, C, is unstable under the employed conditions of alkylation. Its transformation occurs by two main pathways: by hydrolysis of the glycosidic bond with the formation of product D, and by opening of the imidazole ring (which is facilitated by the increase of pH of the reaction mixture from 6.8 to 7.8 during its incubation) with the formation of R, values are given for system 'a'; in this particular case, in order to improve the separation of the spots, chromatography was conducted continuously for 20 h.
=1
I 150
+
250
I .
[B+Hj
300
,100---------1
+
350450
550
GOOAA
Figure 1 . FAB mass spectrum of a total reaction mixture (dGuo + thiotepa).
ALKYLATION PRODUCTS OF DEOXYGUANOSINE AND DNA BY FABMS
H,N(CH,hHNP
K
.s '
L -
=
667
K
YNa,
Figure 2. Products of interaction between thiotepa and dGuo and the pathways of their further transformation in solution.
products H, I, J and K. Though products H, and H, were not observed by FAB mass spectrometry, their formation is proved by the existence of the products of their further modification: I, J and K. Similarly, product G could be formed only from product F which, however, is not observed in the mass spectra in the form of a doubly charged ion (m/z 362.5). In order to concentrate product F (presumably the luminescent material with R, 0.01 in system 'a') the following procedure was employed. Sixteen milligrams of silica gel LC 5/40 (Chemapol, Czechoslovakia) were added to the reaction mixture, which had been preliminarily washed from excess thiotepa, and it was extracted three times with solvent 'b' for TLC. The remaining adsorbed products were rapidly extracted with water and dried. Their mass spectra showed the main peak at m/z 725, which corresponds to a singly charged ion of an adduct of two dGuo molecules and one thiotepa molecule (dimer F). The loss of one charge '
:I
is consistent with the well-known behaviour of dications in FAB.I4 Interaction of thiotepa with DNA
DNA was alkylated and washed from excess thiotepa as described under Experimental. The solution of the alkylated DNA was then heated in a water bath at 100°C for 1 h, which led to a detachment of alkylated purines. The resultant mixture was applied to a Sephadex G-25 column and three main fractions were eluted. Fraction 1 (total volume Y = 82 ml, retention volume u = 85 ml) contained oligonucleotides, which were produced owing to DNA fragmentation at apurinic sites. Fractions I1 and I11 (V = 35 ml and 55 ml, u = 165 ml and 210 ml, respectively) were dried and studied by FAB mass spectrometry. The partial mass spectra, which show the main relevant peaks, are given in Fig. 3.
I. %
0
300
1 b+HI+ 34 1
...i..
300
350 M i Z
Figure 3. Fragments of FAB mass spectra of fractions II (b) and Ill (a), isolated from the reaction mixture (DNA + thiotepa) by gel filtration (see text for details).
G. V. ANDRIEVSKY ET AL.
668
Separation of fractions I1 and 111 by TLC in both cases produced only one the most intensive spot with R, in systems 'a' and 'b' 0.52 and 0.59 (fraction 11), 0.26 and 0.72 (fraction III), respectively. UV absorption spectra of the materials extracted from these spots with water were recorded and compared with known UV spectra of alkyl derivatives of Ade and d G ~ o . ~ * ' ~The , ' ' combination of FAB mass spectrometry and UV spectroscopic data indicates that the main compounds which are present in fractions I1 and I11 are the N7monothiotepa derivative of Gua (D, m/z 343, yield 8.3%) and the N3-monothiotepa derivative of Ade (L, mlz 324, yield 3.4%), respectively. It has been verified that no material with UV spectra different from those of the two main products was present in the fractions under investigation. DISCUSSION ___ - _ _ _ ~
_ .
The results of the present investigation show that FAB mass spectrometry allows detection (and, with the aid of UV absorption spectroscopy, identification) of the adducts of nucleotides and alkylating drugs directly in a complex reaction mixture, the content of an individual
product being in our case as small as 0.5% (1 pg). The method also allows detection and identification of DNA adducts with drugs, omitting the step of separation of primary individual interaction products, which may be especially difficult and even impossible in the case of drugs containing more than one aziridine group. In this work, for the first time, we managed to identify the sites of DNA alkylation by thiotepa in uitro. It has been confirmed that thiotepa reacts with the Gua residue in both dGuo and DNA only at the N7 position. The detailed analysis of the model system (dGuo + thiotepa) considerably simplified the identification of the Gua adduct with thiotepa in DNA. The results with dGuo permit also the prediction of the fate of an alkylated dGMP residue in DNA in uiuo. Besides this, the observation of dimers 2dGuo thiotepa (F) in the reaction mixture (dGuo + thiotepa) shows a possible pathway of formation of intramolecular cross-finks in DNA between two opposite Gua residues. Although the question needs further study, including molecular modelling, it may be noted that DNA cross-linking by thiotepa was also suggested earlier on the basis of indi~ type of rect data obtained in uitro' and in ~ i u o . ' * 'This lesion may be of essential importance in the molecular mechanism of antitumour action of thiotepa and other diethyleneimides of phosphoric acids.
REFERENCES 1. T. G. Sjakste, N. 1. Sjakste and A. V. Lichtenstein, Biokhimia 48.1 37 (1 983). 2. T. L. Pyatigorskaya, 0. Yu. Zhilkova, L. M . Muravyova and L. F. Sukhodub, Molekulyarnaya Biologiya (Eng. Transl.) 20, 334 (1986). 3. T. L. Pyatigorskaya, 0. Yu. Zhilkova, E. S. Ruzaeva, S. V. Kornilova and L. F. Sukhodub, in Modern Problems of Experimental Chemotherapy of Tumours, Part 2, p. 92. Proceedings of the Third All Union Conference, Chernogolovka (1987). 4. A. M . Serebryany, G. V. Andrievsky, A. R. Becker, L. A. Sibeldina and 0. L. Sharova, Bioorganicheskaya Khimia 13, 786 (7987). 5. L. F. Sukhodub, V. S. Shelkovsky, M . V. Kosevich, T. L. Pysatigorskaya and 0. Yu. Zhilkova, Biomed. Environ. Mass Spectrorn. 13, 167 (1986). 6. A. M . Serebryany, G. V. Andrievsky, A. R. Becker, L. A: Sibeldina and M . 1. Povolotskaya, Bioorganicheskaya Khimia 12. 499 (1 986).
7. K. Hemminski, S. Kallama and K. Falck, Acta Pharmacol. Toxicol. 53,421 (1983). 8. G. V. Andrievsky and A. M . Serebryany, unpublished results. 9. C. Fenselau, J. Nat. Prod. 47, 21 5 (1 984). 10. D. L. Slowikovski and K. H. Schram, Nucleosides and Nucleotides 4, 309 (1 985). 11. T. L. Pyatigorskaya, 0. Yu. Zhilkova, V. S. Shelkovsky, N. M. Arkhangelova, A. I. Grizodub and L. F. Sukhodub. Biomed. Environ. MassSpectrom. 14,143 (1987). 12. C. C. Price, G. M . Caucher, P. Koneru, R. Shibakava, J. R. Sowa and M . Yamaguchi, Biochim. Biophys. Acta 166, 327 (1968). 13. E. Institoris and J. Tamas, Chem.-Biol. interact. 47. 133 (1 983). 14. 0. N. Heller, J. Yergey and R. J. Cotter, Anal. Chem. 55, 1310 (1 983). 15. J. J. McCann. T. M . Lo and D. A. Webster, Cancer Res. 31, 1573 (1971).
