Biochemical Pllarnlac~logb, \:t~l 24. p p 1793 1797 Pcrginlon Picss. 197> Printed in (]teat Biitain

INVESTIGATION ON THE OXIDATIVE N-DEMETHYLATION OF ARYL TRIAZENES IN VITRO TtlLLI() GIRALDI,* CARL() NISIt and GIANNI SAVA* *lstituto di Farmacologia and +Islituto di Chimica Farmaceutica, Universita di Trieste, Trieste 1-34100, Italy (Receil:ed 6 January 1975: accepted 26 March 19751

Abstract The oxidative N-demethylation of several p-substituted phenyl triazenes has been determined ill t'itro under carefully controlled conditions. Three of these compounds had been reported not to be appreciably dcmethylated, one of them still possessing antitumour activity. The per cent dcmethylation figures obtained for the tested compounds ranged from about 2(I to 60 per cent and was not dependent on the nature of the solvent used to keep the drug in solution. The enzymatic reaction proved to be strongly inhibited by diazonium cations either generated by spontaneous hydrolysis of the drug or added to the incubation mixture. The reaction is also inhibited at wu'ious degrees by the monomethyl derivatives produced by dcmethylation of the dimethyl triazenes. The inhibition by diazonium and monomethyl derivatives can not account for the observed time-course of demethylation, which occurs for all the tested compounds only in the tirsl minutes of incubation. The p~siblc occurrence of other reaction(s), competing with the demethylation process, the impossibility of correctly expressing kinetic parameters for demethylation, and the possible rclcwmce of these obscrwitions Jot the activity of triazenes in riro. are pointed out. Dialkyl triazenes arc a class of compounds endowed ted [ 13, 15], although having biological activity [ 13]. with interesting biological activities. 5-(3,3-dimethyl- This discrepancy has prompted a reinvestigation of 1-triazeno)-imidazole-4-carboxamide (DIC) has been the N-demethylation reaction, perforlned under carereported to have marked inhibitory activity against fully controlled conditions. The restllts we have L 1210 leukemia and other rodent neoplasms[1,2]. obtained suggest that either the diazo derivatives It also proved to be clinically effective, and is cur- formed by hydrolysis of the parent triazenes, or the rently employed in the therapy of malignant melano- monomethyl derivatives produced enzymatically, ma [3]. Its mechanism of action is essentially un- might inhibit the N-demethylation process in citro. known [4]. DIC undergoes hydrolysis to 5-diazo-im- They further indicate that side reactions occurring, idazole-4-carboxamide and dimethylamine[5], the catalyzed by microsomal enzymes, might chemically diazo derivative having much smaller antitumor acti- modify the substrate, thereby affecting the main provity than DIC itself[2]. This makes unlikely the pos- cess of oxidative N-demethylation. sibility that the drug exerts its effects only by acting as an antimetabolite [4], i.e. via the diazo derivative, which has been shown to inhibit the purine biosynMATERIALS AND METHODS thesis in L:itro [6]. Another modification of DIC concerns its oxidative N-demethylation by liver [7] and The abbreviations and the references for the syntumour [8] microsomes to the corresponding mono- thesis of the used triazenes are reported in Table 1. methyl derivative, which decomposes to 5-amino-imThe purity of these substances was checked by eleidazole-4-carboxamide, nitrogen, and methyl mental analysis and thin layer chromatography, using cations [9]. The formation of methyl cations suggests Kieselgel G F 254 (Merck) and methanol ethyl acethat the activity mechanism of DIC involves an alky- tate ligroin (3:2: 1) or occasionally aluminium oxide lation of cellular components. G F 2 5 4 Type E (Merck) and ethyl acetate hexane This possibility is supported by the reports on the (1:4). In particular, the purity of the samples of DMbiological activity of 3,3-dimethyl-l-phenyl triazene CH 3 and D M - O C H 3 used was confirmed also by and its derivatives, for which a selective inhibitory their n.m.r, spectra: ~$ TMS in CDCI3, 230 (s, 3H, activity on purine biosynthesis is unlikely, due to the ArCHs), 322 (s, 6H, NCHs), 7-22 (sym. m., 4H, ArH) presence of an aromatic ring instead of the imidazole for D M - C H > and (5 TMS, 3-07 (s, 6H, NCHz), 3'57 moiety in their molecule. These compounds have been (s, 3H, OCH3), 7'20 (sym. m., 4H, ArH) for DMshown to possess carcinogenic[10] and mutagenic OCH s. For DM-OCH3, the half-life at pH 7.4 and effects [11], and exhibit an inhibitory activity against 3 7 was determined. Since hydrolysis of this substance rodent tumours [12, 13]. Also, these molecules un- gives a very stable diazonium cation, a direct spectrodergo hydrolysis to the corresponding aromatic dia- photometrical determination has been preferred to zonlum compounds [14] and microsomal oxidative the usual procedures for determining the rate conN-demethylation, as described for DIC [13,15]. At stant [14]. Thus, 0.1 ml of an approximately 4 × least for the effects at systemic level, the latter path10 3 M solution of this substance m absolute ethanol way has been given the greatest importance [16]. was diluted to 10ml with a prewarmed buffer phosIn contrast with this conclusion, some molecules phate 0-1 M solution (pH 7"4). and the absorbances with minute differences in the substituents in the aro- were timely determined at 317nm. Two isosbestic matic ring, have been reported not to be demethyla- points, at 293 and 327 nm. were observed. Linear 1793

1794

T. GIRAt.DI. ('. NISl and G. SAVA

Table 1. Triazene derivatives and diazonium salts used. abbreviations of their chcmical names and rcli~rcnces Ik>r lhcir synthesis //R'

R

N=-:N--N "\R"

Abbreviation DM-H DM-OCH3 DM-CH 3 DM-COOH DM-COOEt DM-CONH e MMCH3 MMCOOEt

R

R'

R"

H OCH3 CH 3 ('()OH COOCeH 5 CONH: CH3 COOCeH s

CH 3

(I'H 3

I-phenyl-3,3-dimethyltriazcne

CH3 CH 3 CH.~ CH 3 CH~ H H

CHs ('H: CH~ CH.~ CHs CH3 CH~

I -p-mcthoxyphenyl-3,3-dimefllyhriazcnc I -p-tolyl-3,3-dimelhyltriazcne

2[ 22 23

l-I>-carboxyphenyl-3.3-dimethyltriazcnc

24

R

('heroical mlmcs

1-p-carbethoxyphenyl-3,3-dimethyltriazcne

13

l-p-carboxamidophcnyl-3.3-dimcthyltriazenc

25

1-p-tolyl-3-n'mt hyltriazene I -p-carbcthoxyphenyl-3-mct hylt riazcne

26 *

',:--N e BF., Cult.

-I"R OCH3 CONH,_

u.v.~,'~:~, 311 nm u,,...... .,,,d,~,,,,,~ 257nm

Rclcfcncus

m.p. 142dec. m.p. 114 dec.

Analyses

Found

C

H

N

("

H

N

37-88 _~5.78

3-18 2.57

12.62 17.88

38.02 35.73

3.40 2.%_

12.carbethoxyphcn~ldiazonmln tctrafluoborate and acqucous mclhylaminc: rccrystal, fl-om hcxanc. m.p. 83dcc.. u.v.~'d',~.,., 311 nm. Analysis calculatcd: C 57.96: H. 6.32: N. 20.28: found: C. 57.85: H. 6.23: N. 2(I.00. + Prepared according to a reported procedure [24] and rccrystal, from methanol and washed with ether. regression analysis applied to 29 spectrophotometrical data. lOglo (E, - E 0, relevant to a range of 60 rain, gave the equation loglo (E, - E~) -0"3457 - 0"01756 t. The root mean square deviation of the observed points from the interpolating line was _+0-01414. Oxidative N-demethylation in dlro was determined essentially as described by Preussmann et al. [15] and Audette et al. [13]. Sprague Dawley rats weighing about 250g were given drinking water containing phenobarbitone 0.5 g:'l. for 3 days. Following sacnlice by cervical dislocation, the washed livers were homogenized by a Potter-Elvehjem homogenizer at 4 in 4 vol of K phosphate buffer 0.1 M, pH 7.4. Both the supernatant obtained by a 20-min centrifugation at 10,000q, or a microsomal preparation were used. Microsomes were prepared by centrifuging the 10,000-q supernatant for 1 hr at 100,(X)0q and resuspending the pellets in a glass homogenizer to the original volume with the phosphate buffer. To measure demethylation, the reaction mixture contained I-9ml 0-I M K phosphate buffer (pH 7.4), 1-5 ml 10,000-g supernatant (about 60rag protein) or microsomal suspension (about 15 mg proteink 01 ml dimethylsulphoxide (DMSO) containing 0"5/xmolcs of the drug (unless otherwise stated), 1001zmoles semicarbazide, 25,umoles MgCI> 21,moles NADP, 25 izmoles glucose-6-phosphate, 2.5 units glucose-6phosphate dehydrogenase (EC 1.1.1.49) type XV from Sigma (final vol 4 ml). At the end of the incubation, carried out at 37 in air for 60 rain (unless otherwise indicated), the reaction was stopped by adding l ml 207{i (w/v) ZnSO,, and I ml satured Ba(OH) 2 solution, and, after centrifugation, the Nash colorimetric assay for formaldehyde was performed on the supernatant [17]. Occasionally, the assay was carried out on formaldehyde separated by distillation. For each determination, a

blank containing the appropriate solvent and no drug, and standards containing known amounts of formaldehyde, were run. The demethylating capacity of the system used was checked by using aminopyrinc as substrate, and proved to be constant for at least 1 hr. Protein was measured according to the biuret method, using bovine serum albumin as standard. RESUI/rS AND I)ISCt S S I O N

Data reported in Table 2 show that all the triazenes tested were demethylated, although at different degrees, by both liver postmitocondrial supernatant and microsomes. MM-CH3 and dimethylanline did not react, thus excluding the possibility that the formaldehyde assayed originated from the demethylatcd triazene or from the demethylamine formed by hydrolysis of the drug. The results concerning D M - C O O H , DM-CH3 and D M - C O O E t are in disagreement with those reported by Audette ut al.[13] and Preussmann el al.[15]. who were unable to show any demethylation for these triazenes, Even by following their procedure exactl}. i.e. using ethanol instead of DMSO, and performing the colorimctric reaction on distilled formaldehyde. we found a percentage of demethylation (see Tables 2 and 3) ranging from about 20 to 60 per cent. Some hint to explain the observed discrepancy derives flom the experiments described in Table 3. When the triazenes were preincubatcd with either 10,(X)0-g supernatant or microsome before adding the NADPH-generating system, there was a decrease in the per cent of demcthylation. This effect is less c~ident with the 10,(X)()-# supernatant than with microsomes, which provides a 4-fold purified :V-dcmethylation system. Since the inhibitor~ effect of preincubations is greater the shorter the h a l f life of the drugs

1795

Oxidative N-demethylation of triazenes Table 2. Demethylation of substituted aryl triazenes by rat liver 10,000-~t supernatant or microsomes Drug

10,000-,0 supernatant

DM-H DM-OCH 3 DM-CONH: DM-CH 3 DM-COO H DM-COOEt HN(CH3) e MM-CH 3

46.8 55.7 48.6 58.5 22.6 26"7 1"9 4"2

Microsomes

+ 4.05 (3) + 7.17(31 _+ 5.15(4} + 4.74(6) ( 1) + 5'1 [4) (2) (2)

34-6 + 5.05 (3) 44-2* 32.2 + 5.35(3) 42.5 (1) 60.0 + 4.8I (6)85-4* 18-9 ( 1) 20"1 (1) 0"7

(2t

Each wdue is the mean + S.E., expressed as per cent of the drug demethylated in 1 hr (number of determinations in parentheses). * The formaldehyde assay was performed after distillation as described by Audette et al.[ 13], on half scale.

Table 3. Effects of preincubation on the demethylation of aryl triazenes, dissolved either in DMSO or ethanol

Drug DM-H DM-CH 3 DM-OCH 3 DM-COOH DM-COOEt DM-CONH= HN(CH3)2

Solvent

10,000-,q supernatant ( - )* (+ )

EtOH DM SO EtOH DMSO EtOH DMSO EtOH EtOH EtOH EtOH

68"4 45- l 75-1 64"6 42-2 42-6 20-5 22'4 628 13

Microsomes

56"6 45-0 45-3 58"4 16"2 21"7

(- )

(+ )

4l "8 29.6 59"7 56-1 26-1 26-9

31"0 28.9 18"7 19"2 1"65 0"70

0.5/lmoles of the drug in 100/d. of DMSO or l()ld, of ethanol (EtOH) were added to 1-4 (1.49) ml of the phosphate buffer and, when indicated, preincubated for 30rain at 37. After completion of the reaction mixture, the per cent demethylation was determined after 60 rain. * ( - ): no preincubation: ( + ) preincubation). is [14], it is likely that the d i a z o n i u m salts, formed by s p o n t a n e o u s hydrolysis of the triazenes, are responsible for this i m p a i r m e n t o f the e n z y m e system. Evidence in favour o f this h y p o t h e s i s also derives from the e x p e r i m e n t s s h o w n in Fig. 1, w h e r e the direct effect of two d i a z o n i u m salts has been tested on the N - d e m e t h y l a t i o n o f D M - C O N H > Both diaz o n i u m salts tested exerted a n inhibitory activity, a l t h o u g h o f different degree, the inhibition being m o r e p r o n o u n c e d w h e n using m i c r o s o m e s t h a n with 10,000-,q s u p e r n a t a n t . T h e results so far described w o u l d suggest that a partial hydrolysis o f the drug before the e x p e r i m e n t s could seriously affect a following assay. In addition, the d i a z o n i u m generated during the assay' by hydrolysis o f short lived molecules could affect the final results, particularly in m i c r o s o mal p r e p a r a t i o n s . The f o r m a t i o n o f d i a z o n i u m salt can not fully a c c o u n t however, for the o b s e r v e d inhibitory effects: in fact, the rate o f d e m e t h y l a t i o n o f D M - C O O E t and D M - C H 3, m o r e stable than D M - O C H 3 * , also levels off after a few m i n u t e s o f i n c u b a t i o n (Fig. 2). To und e r s t a n d this result bettcr, we have studied the d e m thylation in 10-min incubations, following an expo* For this substance the half-life at pH 7"4 and 37 is 17"14 min: the reported valuc at pH 7 and 3 7 was 11.0, and for DM-CH 3 and D M - C O O E t was 37.0 and 5-0 x 10'* mm respectively [ 14].

sure o f the e n z y m e system to either the dimethyl triazenes itself or to the c o r r e s p o n d i n g m o n o m e t h y l triazenes. The results in Table 4 show that with a triazene with a m o d e r a t e rate o f d e m e t h y l a t i o n , such as D M C O O E t , a previous i n c u b a t i o n of either lO,O00-g IO0

50

_

O..,-...----

Q ---O

I

[

12-5

25

I

62.5

Concn, p.M Fig. 1. Effects of two diazonium salts on the demethylation of DM-CONH2. Each value represents the per cent demethylation of 125 izM D M - C O N H 2 in 1 hr, in the presence of different concentrations of p-methoxyphenyldiazonmm tetrafluoborate (0 ©l or p-carboxamidophenyldiazonium tetrafluoborate (liD), by 10,000-g supernatant (O ~1 or microsomes ( e l i ) .

1796

f. GIRAII)I. ('. Nlsl and (i. SAVA

Table 4. bfl'ccts of preincubation with triazcnc dcmativcs on the subsequent dcmcthylation of ~111 added dimclh~l triazcnc 10.000-~/ supcrnata nt Incubations 0 I()min II) 2()rain ",, Demclh\lalion DM-OOEI DM-('OOEl

17-I 14.7 26.2

I)M-C'OOEt DM-(OOFt

DM-('Hs I)M-CH 3 MM-('H3 DM-OCH.~ DM-OCtt.~

M icrosomcs [ncubalions 0 Illl/lill I(l 20rain I)M-COO L t DM-('OOI!I MM-COOEt MM-('OOIt MM-('OO[ t } DM-(OOI!l +

DM-('O()t{I I)M-( 'OOFI I)M-('()()[1

".. l)emclh,.lalion

16-4 17.s a4"2 2.0 175 17.5

63.5 30-6

DM-('H 3 DM-('H s I)M-('H3

77. I

17.9 50.3 3 I.N 70"9

DM-O('H 3 I)M-()('H 3

The Iriazcnes wcrc added to Ihc reaction mixture, containing the coihctors and incubated at 37, either :~t time 1) or 10 rain: the rcaction was stopped 10 rain after the last addition. IOO



,A

L

!

I

5

20

60

Time,

mln

Fig. 2. fime-course of dcmcthylation of substituted afyl triazcnes by 10,O00-g stlpernatant. Each value represents the per cent demetlolation of 1).5 itmoles of DM-COOEt II, I)M-('H,~ II, DM-O('H~ at. at Ihc time indicated.

undergo oxidativc N-demethylation in rim,, supports the hypothesis put forward by several authors [16, 18 20] that this metabolic pathway is gcnerall,~ involved in the activity of these c o m p o u n d s in fifo. As lhr as the time-course of dcmethylatiou in rilro is concerned, fine lack of a complete block of the enzymatic system due to diazonium or monometlnyl triazene derivatives, observed in these cxperiments, cannot explain the occurrence of dcmethylalion only in the first minutes of incubation (Fig. 2). O t h e r enzymatic reaction(s) are thus likely to occur on the intact drug, competing with the dcmcthylation process, and transtbrming the triazene into dcriwltives uueapablc of tmdergoing demcthylation. All these data indicate the impossibility' of expressing kinetics figures correctly related to the capacity of triazenes to behave as substrate lbr microsomal demethylation. They also indicate the need tbr a more detailed exmnination of the reactions occurring in ri/ro by hepatic microsomal preparations, and their relevance with the activity of this class of drugs in rilo.

supcrnatant or microsomcs with the drug itself or with its monomethyl derivative is without effect. With D M - O C H > two additions of the drug to the samc assay system give a demethylation per cent, which is somehow lower than the n u m b e r calculated by summing up the demethylation per cent obtained with single additions. Finally, with D M - ( H 3 the results are more complicated. First, the prcincubation of the enzyme system alone produces a 50 per cent rcduction of the demethylation of the drug subsequently added. Further, a premcubation of the 10.(X)0-¢] supernatant with either the drug itself or its monometh\,l deriwttive results in an inhibition of its demcthylating capacit3, corresponding respectively to 84 and 57 per cent of the controls. In conclusion, the data described so far cannot explain the discrepancy between the demethylation figures for DM-CHa, D M - C O O H and D M - C O O E t obtained by us and by other authors. The fact that three triazene deriwltives, already reported to possess antitulnour activity in rim. have now been shown to

4ckmm'ledgument The authors would like to thank I)r. T. A. Connors lbr lhc ollcotiragcmcnl and suggestions given to one of them (T.G.) during a silt) ill his departn)enl.

r I.:Ft,:r EN('Es 1, Y. F. Shcaly, J. A. M o r a g o m c r y and W. R. Lastcf ,h.. Biochem. Pharnmc. I I , 674 (1962). 2. K. Hano. A. Akashi, I. Yamamoto. S. Narumi, Z. Horii and I. Ninomiya, (;ann 56, 417 (1965). 3. S. K. Carter and W. T. Sopcr. (am'er ?i'calim, nl Rcr. !. I {1974). 4. S. K, (ku'tcr and M. A. Friedman. /;'ur. J, ('am'dr 8,

~5 /19721. 5. Y. F. Shcaly, .I. Fharm. Sci. 59, 153311970). 6. D. A. Peters and P. L. Mc(iccL ('an..1. Phr~iol. Pharmac. 46, 195 (1968). 7. J. L. Skibba, D. [). P,eal. G. Ramircz and (i. T. Br_,,an. ('am'er Res. 3{1, 147 (1970L ,'4. N. S. Mizuno and F. W. Humphrc}. ('ore'or ('hclllIll/It'l'. Rt'l~. 1>01"I I. 56, 4(~5 119721.

Oxidative N-demethylation of triazenes 9. H. T. Nagasawa, F. N. Shirota and N. S. Mizuno, Chem.-Biol. Interact. 8, 403 (1974). 10. R. Preussmann, S. lvankovic, C. Landschiitz, J. Gimmy, E. Flohr and U. Griesbach, Z. Krehslorsch. 81, 285 (1974). 11. G. F. Kolar, R. Fahrig and E. Vogel, Chem.-Biol. Interact. 9, 365 (1974). 12. J. H. Burchenal, M. K. Dagg, M. Beyer and C. C. Stock. Proc. Soc. exp. Biol. Med. 91, 398 (1956). 13. R. C. S. Audette, T. A. Connors, H. G. Mandel, K. Metal and W. C. J. Ross, Biochem. Pharmac. 22, 1855 (1973). 14. G. F. Kolar and R. Preussmann, Z. Naturlorsch. 26b, 950 (1971). 15. R. Preussmann, A. Von Hodenberg and H. Hengy, Biochem. Pharmac. 18, I (1969). 16. G. F. Kolar, in Advances i, Antimicrohial and Anti,eoplastic Chemotherapy (Eds. M. Semonsk}', M. Hejzlar, S. Mas6k), p. 121. Urban and Schwarzenberg, Miinchen (19721.

1797

17. J. Cochin and J. Axelrod, J. Pharmac. exp. Ther. 125, 105 (1959). 18. F.W. Krtiger, R. Preussmann and N. Nieplet, Biochem. Pharmac. 20, 529 (1971). 19. J. L. Skibba and G. T. Bryan. Toxic. appl. Pharmac. 18, 707 (1971). 20. F. A. Schmid and D. J. Hutchinson, Ca,cer Res. 34, 1671 (1974). 21. J. Elks and D. H. Hey, J. chem. Soc. 441 (1943). 22. N. Barton, J. W. Cook, J. D. Loudon and J. MacMillan, J. chem. Soc. 1079 (1949). 23. F. Klages and W. Mesch, Bet. 88, 388 (1955). 24. G. F. Kolar, Z. NaturJbrsch. 27b, 1183 (1972), 25. Y. Lin and T. L. Loo, J. reed. Chem. 15, 20I (1972). Y. F. Shealy, C. A. O'Dell, J. D. Clayton and C. A. Krauth, J. Pharm. Sci. 60, 1426 (1971). 26. Org. Symh., Coll. Vol. V, p, 797. Wiley, New York (1973}.

Investigation on the oxidative N-demethylation of aryl triazenes in vitro.

Biochemical Pllarnlac~logb, \:t~l 24. p p 1793 1797 Pcrginlon Picss. 197> Printed in (]teat Biitain INVESTIGATION ON THE OXIDATIVE N-DEMETHYLATION OF...
352KB Sizes 0 Downloads 0 Views