Urinary metabolites of 3,3-dimethyl-1-phenyltriazene

Chem.-Biol. Interactions, 14 (1976) 301-311 0 Elsevier Scientific Publishing Company, Amsterdam

- Printed in The Netherlands

URINARY METABOLITES OF 3,3-DIMETHYL-l-PHENYLTRIAZENE

G.F. KOLAR and J. SCHLESIGER Institute (F.R.G.) (Received (Accepted

for Toxicology

and Chemotherapy,

German Cancer Research

Centre. Heidelberg

March 3rd, 1976) March 20th. 1976)

SUMMARY

Urinary metabolites excreted after a subcutaneous injection of 3,3dimethyl-l-[ 14C] phenyltriazene (DM[ 1-14C] PT) to rats accounted for 82% of the applied radioactivity. We have isolated aniline (l-2%), 2-hydroxyaniline (5-7%), 3-hydroxyaniline (about 1%) and 4-hydroxyaniline (3137%) from ethyl acetate extracts of acid-hydrolysed urine. UV spectrometric determination of 4-hydroxyaniline, using the indophenol reaction, showed that the most abundant metabolite accounted for 56 to 61% of the applied dose. We have also demonstrated the excretion of metabolites containing the intact triazene structure (0.9-1.1%) by cold acid cleavage of these compounds, followed by coupling of the released arenediazonium cations with IV-ethyl-1-naphthylamine (EN). The coloured derivatives of these metabolites, 4-benzeneazo-IV-ethyl-l-naphthylamine (BAEN) (0.6-0.7%), 4-( 2hydroxybenzeneazo)~IV-ethyl-1-napthylamine (2-HO-BAEN) (0.02%) and 4-(4.hydroxybenzeneazo)~Methyl-l-naphthylamine (4-HO-BAEN) (0.30.4%) were isolated. The identification of BAEN as the principal azo derivative of the excreted triazene metabolites is in full agreement with the proposed in tivo activation of 3,3dimethyl-l-phenyltriazene (DMPT) to a carcinogenic methylating agent. The hydroxylation of the methyl group at N-3 yields the corresponding aminol, some of vhich is covalently bonded to a water-sohlble compound.

Abbreviations: BAEN, 4-benzeneazo-N-ethyl-1-naphthylamine; DMPT, 3,3-dimethyl-l[‘4C]DMPT, 3,3phenyltriazene; DM[ l-14C]PT, 3,3-dimethyl-1-[ l- 14C]phenyltriazene; [ “C ] dimethyl-l-phenyltriazene; EN, N-ethyl-l -naphthylamine; 2-HO-BAEN, 4.( 2hydtoxybenzeneazo)-EN; 3-HO-BAEN, 4-(3.hydroxybenzeneazo)-EN; 4-HO-BAEN, 4-(4hydrol ybenzeneazo)-EN; MPT, 3.methyl-1-phenyltriazene.

301

INTRODUCTION

The acyclic DMPT, Fig. 2, was prepared by Elks and Hey [l J in 1943 but its versatile biolo~c~ activity was not imm~ia~ly recognized. Subsequently, Clarke et al. [2] discovered the inhibitory activity of DMPT against mouse sarcoma 1SO in 1955 and Druckrey et al. reported its carcinogenic [3] and teratogenic [4f potency in rats in 1967. The mutagenicity of DMPT was described in 1971 by Vogel f5], Gng and de Serres [6] and by Fahrig [?I, using ~ro~oph~l~mel~~og~s~er,~e~rospor~ crassa and ~cc~u~o~yces cereuisiae, respectively. It has generally been accepted that the biological activity of DMPT predom~~tly depends on the formation of reactive intermediates by enzymic activation. Since DMPT is demethylated by rat-liver microsomes in vitro [8,9], it has been suggested that the carcinogenic [lo) and the mutagenic ill] activity of this compound could be due to conversion into MPT which is a known methylating agent [12]. Krilger et al. [lS] have demonstrated the formation of N-7 methylguanine in she liver RNA and DNA of rats that had received an intraperitone~ injection of [“C~DMPT. Enzymic degradation of 3,SdialkyLl-aryltriazenes to the corresponding monoalkyl analogues has also been implicated as an essential step for the expression of tumourinhibitory activity by that class of compounds [14,15]. Because the metabolic fate of DMPT or related triazenes is unknown, the suggestion that MPT p~icipates in c~cinogenesis awaits ex~~rnent~ confirmation. It is also not known whether the same intermediates and mechanisms are responsible for all known modes of biological activity exhibited by DMPT. In this paper we report our studies on the urinary metabolites of DMPT in rats. ‘The most abundant products of the biotransformation were modified anilines but, surprisingly, metabolites which contained the intact triazene structure were also detected. Since such labile metabolites are difficult to isolate, we have cleaved them with cold acid and trapped the released arenediazonium cations by coupling with EN to yield stable azo derivatives. In order to identify the derived 4-areneazo-N-ethyl-1-naphthylamines, we had to synthesize several hitherto unreported model compounds. MATERIALS

AND METHODS

Chemicalsand model compounds (a) DM[ 1J4C] PT [ 161 was prepared from [ 1-14C]aniline hydrochloride f6.6 mg, 0.05 mmole, 500 I;1Ci, spec. act. 77 mCi~g, F~bwerke Hoechst, Frankfurt/M., F.R.G.) which wasdiluted with inactive compound (252.6 mg, 1.95 mmole), diazotised and coupled with dimethyl~ine hydrochloride (815.5 mg, 10 mmole) in the presence of excess sodium carbonate. The crude product was extracted with ether and purified by repeated distillation under reduced pressure. The yield of DM~l-~4C~PT, spec. act. 245 pCi/ mmole, was 266 mg (89% of the theoretical value}. The ~diochemic~ purity of the product was established by scanning of 302

thin-layer chromatograms (LB 2723 Thin-layer Scanner II, R. Berthold, Wildbad, F.R.G.) and was found to be better than 99%. (b) DMPT was prepared by described methods [ 1,171. The redistilled product was a pale yellow liquid b.p. 115/14 mm [I] (125-127/19 mm), uv: x,,, 223 nm, log E 4.12 which was homogeneous on TLC (R, 0.84 in toluene-acetonz, 3 : 1 v/v, on precoated silica gel plates F-254, Merck A.G., Darmstadt, F.R.G.). The prurity of the compound was confirmed by its MS fragmentation pattern and by relative intensity of the principal ions which agreed with the values reported for an analytical sample of DMPT 1181. (c) 2-, 3- and 4-Hydroxyanilines (Merck A.G., Darmstadt, F.R.G.) were recrystallised from aqueous ethanol and their purity was confirmed by thinlayer chromatography. EN (Fluka A.G., Buchs, Switzerland) was purified by chromatography on alumina (Activity grade III, Woelm A.G., Eschwege, F.R.G.) in petroleum ether (60-70), followed by distillation under reduced pressure (b.p. 169/15 mm). EN hydrochloride was recrystallized twice from hot water (m.p. 216.5-217, [ 191 193). (d) BAEN and the isomeric 2-HO-BAEN, 3-HO-BAEN and 4-HO-BAEN were prepared by coupling the corresponding arenediazonium fluoborates [20] with EN. EN hydrochloride (10.38 g, 0.05 mole) was dissolved in 400 ml of 0.1 N HCl and ethanol (3 : 7 v/v). The filtered solution was cooled to 0°C and the stoichiometric amount (0.05 mole) of the required arenediazonium fluoborate was added in portions over 10 min with mechanical stirring. The reaction was stirred t jr an additional 30 min after which it was placed in a refrigerator to complete the crystallization of the formed azo derivatives. 4-Areneazo-Nethyl-1-naphthylamine hydrochlorides were recrystallized from 0.1 N HCl and ethanol mixture (3 : 7 v/v) and isolated in yields ranging between 60-80% of the theory. The free bases of the azo derivatives were extracted with ether and purified by crystallization from cyclohexane-ether mixtures.

m.p. BAEN[ 21,221 68-71 P-HO-BAEN 111-112 3-HO-BAEN 159-l 60 I-HO-BAEN 198-199

RI; in solvent

Hydrochloride

4-Areneazo-N-ethyl-l-naphthylamine Amax

log E

m.p.

x may

logi

4

5

6

471 501 474 469

4.36 4.51 4.36 4.37

180-181 171-172 179-180 180-181

538 558 541 575

4.68 4.5s 4.66 4.63

0.17 0.51 0.03 0.05

0.46 0.83 0.14 0.30

0.81 0. 0.47 0.62

(e) Precoated plates (Silica gel F-254, Merck A.G., Darmstadt, F.R.G.) were used for thin-layer chromatography in the following solvents: For the separation of aniline metabolites: (1) toluene; (2) toluene acid (60 : 39 : 1 v/v); (3) toluene=-acetone (3 : 1 v/v). For 4-areneazo-N-ethyl-1-nuphthylamines: (4) toluene-acetic acid (9 : 1 v/v): 303

(5) toluene-acetic acid-acetone (8.5 : 1 : 0.5 v/v); (6) tolueneacetic acidacetone (6 : 1 : 1 v/v). Silica gel (0.05-6.2 mm, Merck A.G., Darmstadt, F.R.G.) for neutral alumina (activity grade III, Woelm A.G., Eschwege, F.R.G.) were used as solid supports for column chromatography. Animal experiments

Male Sprague-Dawley rats weighing 300-350 g, kept on “#romin@‘” diet, were used in all experiments. Unless stated otherwise, the e>;iperimental and control animals were housed in Metabowl Mark IV cages f#‘lencons of Hemel Mempstead, Hertfordshire, U.K.). To avoid unnecessary f :ontamination, the animals received water ad libitum but no food during I’,he experiments. (a) Metabolic experiment with DM[1-‘4C] PT (245 pCi/mmole; 149.2 mg/kg): three rats (845 g) were subcutaneously injected with 69.4 pCi (126 mg, 0.8~15mmole) DM[ 1-14C]PT. The excreted radioactivity was determined by scintillation counting (Mark II, Chicago Nuclear, U.S.A.) using “Instagel” counting fluid. The results were corrected by internal standardization and the counting efficiency was 83%. The 4t) h urines were combined, made up to 100 ml and 25 ml portions (14.2 MCi), acidified with cont. HCl(5 ml), were hydrolysed by heating on a water bath for 1 h. The hydrolysates were neutralized with ammonia and extracted with ethyl acetate (5 X 25 ml). The organic phase contained 33% of the urinary radioactivity. The dried extract (MgS04) was evaporated and chromatographed on three TLC plates in solvent l), revealing a mobile zone at RF 0.17 that cochromatographed with caniline. The retained radioactive material was eluted with acet,one and rechromatographed in solvent (2); this treatment detected radioactive zones at RF 0.56,0.46 and at RF 0.20-6.29, corresponding to 2-hydroxyaniline, 3-hydroxyaniline and 4-hydroxyaniline, respectively. (b) Isolation and identification of modified anilines: Groups of ten rats were injected with 298.4 mg (2 mmole)/kg of DMPT dissolved in edible oil. The combined urine, excreted ir 72 h (about 300 ml) was acidified with cont. HCl (40 ml) and hydrolysed as already described. The ethyl acetate extract was purified by passing through a 1.5 X 20 cm column of silica gel and the metabolites were separated by thin-layer chromatography, as described, using the isolated labelled metabolites as markers. The isolated metabolites were identified by cochromatography with authentic samples, diazotisation and subsequent coupling with EN followed by identification of the azo derivative and by the indophenol reaction, when applicable, as aniline (l-2%), 2-hydroxyaniline (5-7%), 3-hydroxyaniline (about 1%) and as 4-hydroxyaniline (31-36%). (c) Calorimetric determination of 4-hydroxyaniline by the indophenol reaction [23,24] : 0.5-5 ml urine was acidified with an equal volume of cont. HCl and the solution hydrolysed by heating on a water bath for 1 h under a stream of nitrogen. The reaction was clarified by centrifugation 304

(6000 rpm) for 10 min, and supernatant was concentrated in a rotary film evaporator below 40. The residue was taken up in 0.01 N HCl (4 ml) and chromatographed on a G-10 Sephadex column (25.5 X 1.5 cm) which was duted with 0.01 N HCI. The fractions lo-22 (46.8 ml) were made up to 100 ml with 0.01 N HCl and portions (l-4 ml) of this solution were used for the determination. The concentration of 4-hydroxyphenylquinoneimine, developed in aqueous sodium carbonate after 25 min, was determined by UV absorption at 630 nm. (d) Conversion of triazene metabolites into 4-areneazo-Nethyl-l-naphthylamines: EN reagent: NE hydrochloride (700 mg, 3.3 mmole) was dissolved in cont. HCl(65 ml) and ethanol (285 ml). Ten rats were subcuta-\eously injected with 298.4 mg (2 mmole)/kg of DMPT and the urine excreted by each pair of animals was collected in a 250ml conical flask, wrapped in aluminium foil, containing 35 ml of the EN reagent. The progress of excretion was followed by increase in the formation of azo derivatives by absorption at 545 nm and expressed as concentration of BAEN. The assay showed that 65%, 87% and 90% of triazene metabolites, capable of diazo coupling, were excreted in 24 h, 48 h and in 72 h respectively. Triazene metabolites, excreted in 72 h, accounted for 0.9-1.1% of the applied dose. (e) Isolation and identification of 4-areneazo-Methyl-1-naphthylamine derivatives: The combined coloured solutions (about 800 ml) were neutralized to pH 8 with ammonia, the azo derivatives were extracted with ethyl acetate (3 X 250 ml) and the dried organic phase (MgS04) was evaporated at reduced pressure below 40. The residue of the formed azo derivatives was dissolved in toluene and adsorbed on a 2 X 30 cm column of neutral alumina. The derivative A was elutr,; with toluene after which the derivative B was removed with tolueneacetnne (8 : 2 v/v). The strongly adsorbed derivative C was extracted from the extruded adsorbent with ethanolic HCl. After purification of the pigments by thin-layer chromatography in solvents (4), (5) or (6) and recrystal’lisation from petroleum ether (60-70), the structures of derivatives A, B and C were confirmed by mass spectrometry as BAEN (0.6-O.$%), 2-OH-BAEN (0.02%) and as 4-OH-BAEN (0.3-0.4%). RESUL’I’S

Metabolic experiments with DM[l-14]PT in rats showed that 82% of the injected radioactivity was excreted in 48 h urine and that most of the labelled metabolites were retained at the origin after thin-layer chromatography. On the other hand, thin-layer chromatograms of the ethyl acetatesoluble constituents from urine hydrolysates revealed mobile zones which cochromatographed with aniline (solvent 1) and with the three isomeric hgdroxyanilines (solvents 2 and 3). The detected metabolites were isolated from hydrolysed urine of ten rats, injected with 298.4 mg (2 mmole)/kg of DMPT. The extracted constituents were purified by passing through a column of silica gel and separated by

305

thin-layer chromatography. The first development in solvent 1 separated a non-polar metabolite at RF 0.14-0.17 which was identified as aniline. Bechromatography of the retained material in sclvent 2 separated additional metabolites at RF 0.56, 0.46 and at RF 0.20-4.29which were identified as 2-hydroxyaniline, Q-hydroxyaniline and as 4-hydroxyaniline. The structures and yields of the isolated aniline metabolites are given in Fig. 1. Since the most abundant metabolite of DMPT, 4_hydroxyaniline, is a sensitive compound which deteriorates rapidly during isolation and chromatography, its concentration in urine hydrolysates was also determined by UV spectroscopy using the indophenol reaction. The calorimetric assay showed that the yield of the excreted hydroxyanihne conjugates was higher and accounted for 56-61% of the applied DMPT dose. The metabolites containing the intact triazene structure ‘were converted into stable 4-~~eneazoðyl-l-naphthylamine derivatives by a direct collection of the urine into acidified EN reagent. The sequence of reactions implicated in the cleavage of these metabolites into arenediazonium cations and their coupling with the passive component is shown in Fig. 2. The areneazo derivatives, which were obtained from triazene metabolites excreted by ten rats injected with 298.4 mg (2 mmoles)/kg of DMPT, were extracted with ethyl acetat-e at pH 8 and separated by column chromatography on alumina. The coloured derivatives were purified by thin-layer chromatography with authentic model compounds as BAEN, 2-HO-BAEN and as 4-HOBAEN. The structures and yields of the isolated 4-areneazo-N-ethyl-lnaphthylamines are given in Fig. 3. The structures of the coloured 4-areneazo-IV-ethyl-1-naphthylamines were confirmed by mass spectrometry and spectra of BAEN and of 4-HO-BAEN are shown in Fig. 4.

Structure

:

@%

Metabolite

RF /

:

Anlline

0.14 - 0.17

Yield

1

P/O):

l-2

2- Hydroxyaniline

0.46

“OeNH2

Fig. 1. Aniline (2 mmoles)/kg acetone-acetic

306

4-Hydroxyaniline

0.20-0.29

2

2

About

1

31-36

metabolites isolated from hydrolysed urine of rats injected with 298.4 mg of 3,3-dimethyl-1-phenyltriazene. Solvent 1: toluene; solvent 2: tolueneacid (60 : 39 : 1 v/v).

X3-Dimethyl-l-phrnyltriazene

Biotransformation

N+N/N

of the phenyl ring

/CH3

‘CH3 b) Conjugation

Ho Protonation at N3 I NI’O
HO

of the N3 bond

Cleavage

- HN:iE;

N2-

I .,_o”‘”

Fig. 2. Conversion

‘ri-“Hz-C”

of

triazene

metabolites

into

4-areneazo-N-ethyl-1-naphthylamines.

The mass spectrum of BAEN showed the expected molecular ion at m/e 275 (100%) which lost 29 mass units (C,H,) to give a peak at m/e 246 (5.3%). The main fragmentation of the molecule involved the loss of 105 mass units, extruded probably as a Ph-Nz radical, to yield a peak at m/e 170 (86%), corresponding to the cation C12H12N which is characteristic for the fragmentation of 4-areneazo-IV-ethyl-1-naphthylamines. The spectra of isomerit 2-HO-BAEN and 4-HO-BAEN revealed analogous cracking patterns. They contained base peaks at m/e 170 (100%) with molecular ions at m/e

Structure:

Azo Derivative’

RF /

SNQNqJIpH5

Solvent:

Yield W.):

4-Benreneazo-N-ethyl-1-naphthylamine

0.46

5

0.6 - 0.7

4-(2-HydroxybenZeneazo)-N-ethyl-l-naDhthylamIne

063

5

0.0 2

4-(4-Hydronybenteneazo)-N-ethyl-1-nophthylam1ne

0.30

5

0.3-04

.n

Fig. 3. Areneazo-IV-ethyl-I-naphthylamine triazene metabolites.

derivatives

obtained

by conversion

of excreted

307

0,-,

25

N=N

/H 8\/ ‘W-b

50

75

,-,

N

iD0

125

175

150

A-’ 'C2'-k

i I

7Y

; 200

I

225

I

250

I

275

Fig. 4. Mass spectra of 4-benzeneazo-N-ethyl-1-naphthylamine benzeneazo)-N-ethyl-l-naphthylamine.

291 (96% and 63%), respectively, in full agreement one hydroxyl group int!:, the phenyl ring.

I 300 m/o

and

of

4-($-hydroxy-

with the introduction

of

DISCUSSION

DMPT was rapidly metabolised in rats since about 69% of radioactivity, injected as DM[‘“] PT, was excreted in 24 h urine; additional 13% of radioactivity was excreted irl 48 h urine. Similarly, the formation of metabolites containing the intact triazene structure was completed within 48 h, as demonstrated by the time-course of the transcoupling reaction. These results tie in good agreement with the findings of Kleihues et al. [25] who showed that about 40% of radioactivity, injected as [14C]DMPT, was exhaled in 15 h and also that methylation of DNA and RNA bases by DMPT in vivo was completed in about the same time. The most abundant metabolites of DMPT which we isolated from urine hydrolysates were ring-hydroxylated anilines (38-46%). Because these metabolites are labile compounds, the concentration of 4-hydroxyaniline was also determined by the indophenol reaction. The calorimetric assay showed that the amount of 4-hydroxyaniline accounted for 56-61% of the injected DMPT dose and that, if similar correction for losses due to oxidation was applied to all isolated aniline metabolites, at least 70% of the injected carcinogen was recovered as hydroxyaniline conjugates. Since the unhydrolysed urine gave a negative indophenol reaction [23,24], which is a specific test for the detection of 4-hydroxyaniline, it can be concluded that 4hydroxyaniline and possibly other phenolic metabolites were excreted entirely in conjugated form. The isolation of modified anilines in high yield is not surprising and that re-

308

suit agrees well with the effective demethylation

of DMPT to MPT by the hepatic mixed-function oxidases [8,9] . It seems probable that hydroxyanilines are formed by secondary hydroxylation of aniline which arises by hydrolytic fission of the intermediary MPT, #-N = N-NH-CHJ * #-NH-N = N--CH+ Alternatively, the aniline metabolites may be formed by enzymic reduction and subsequent fission of the intermediary l,l-dimethyl-3-phenyltriazene. A possible objection to that mechanism is the expected cleavage of the reduced system either between N-2/N-3 bond to yield aniline, or between N-l/N-2 bond to yield phenylhydrazine. Since no hydroxy!ated phenylhydrazine, nor phenylhydrazine itself, were detected, this sequence does not seem to be the main pathway of aniline formation in vivo. The excretion of metabolites containing the intact triazene structure indicates that a portion of the injected DMPT was converted into more stable water-soluble products. An addition of excreted urine to a basic solution of 2-naphthol failed to yield any azo derivatives, proving that no free diazonium cations were eliminated. The pretreatment of the metabolites was therefore essential for the release of the areneadiazonium cations. The triazene metabolites were converted into the corresponding 4-areneazo-N-ethyl-l-naphthylamines while the conjugative groups of the phenolic hydroxyls were cleaved off during the reaction with the acidic reagent (pH 1.5). The ring-hydroxylation in both the aniline metabolites and in metabolites capable of diazo coupling had occurred predominantly at the para and ortho positions and only slightly at the meta position. The colours of the areneazo-IV-ethyl-l-naphthylamines obtained suggested their chemical structure since the protonated form of BAEN is red whereas 2-HO-BAEN and 4-HO-BAEN are violet and blue, respectively, owing to the increased number of resonance forms. The formation of BAEN from a water-soluble triazene metabolite (0.60.7%) may provide a lead to the reaction sequence occurring in activation of DMPT to a carcinogenic intermediate. Since no unchanged DMPT was detected amongst the excreted products, it is probable that the injected carcinogen was enzymatically hydroxylated at one of the methyl groups and the resulting amino1 was covalently bonded to a water-soluble compound. Similarly, 4-HO-BAEN is probably formed by transcoupling and subsequent hydrolysis of 1-(4-hydroxyphenyl)-3,3-dimethyltriazene O-conjugate. The results obtained tempt one to speculate whether these soluble triazene conjugates are only detoxication products of DMPT or whether they represent transport intermediates. These in turn may release the biologically active agent by cleaving off the conjugative groups in target organs containing specific enzymes. REFERENCES 1 J. Elks and D.H. Hey, Union of aryl nuclei, VI. Reactions with 1-aryl-3:3-dimethyltriazenes, J. Chem. Sot., (1943) 441. 2 D.A. Clarke, R.K. Barclay, C.C. Stock and C.S. Rondestvedt Jr., Triazenes as inhibitors of mouse sarcoma 180, Proc. Sot. Exp. Biol. Med., 90 (1955) 484,.

309

3 H. Druckrey, S. Ivankovic and R. Preussmann, Neurotrope carcinogene Wirkung von Phenyl-dimehtyl-triazen an Ratten, Naturwissenschaften, 54 (1967) 171. 4 H. Druckrey, S. Ivankovic, R. Preussmann and U. Brunner, Teratogene Wirkung von I-Phenyl-3,3-dimethyltriazen; Erzeugung von Gaumenspalten bei BD-Ratten, Experientia, 23 (1967) 1042. 5 E. Vogel, Chemische Konstitution und mutagene Wirkung, VI. Induktion dominanter und rezcssiv-geschlechtsgebundener Letalmutation durch Arylalkyitriazene bei Drosophila melanogaster, Mutation Res., 11 (1971) 397. 6 T. Ong and F.J. de Serres, Mutagenicity of 1-phenylB,3-climethyltriazene and lphenyl-3-monomethyltriaaene in Neurospora crassa, Mutation Res., 13 (1971) 276. 7 R. Fahrig, Metabolic activation of aryldialkyltriazenes in the mouse: induction of mitotic gene conversion in Saccharomyces cerevisiae in the host-mediated assay, Mutation Res., 13 (1.971) 436. 8 R. Preussmann, A. v. Hodenberg and H. Hengy, Mechanism of carcinogenesis with 1-aryl3,3-dialkyltriazenes. Enzymatic dealkylation by rat liver microsomal fraction in vitro, Biochem. Pharmacol., 18 (1969) 1. 9 T. Giraldi, C. Nisi and G. Sava, Investigation on the oxidative N-demethylation of aryi triazenes in vitro, Biochem. Pharmacol., 24 (197 5) 1793. 10 R. Preussmann, H. Druckrey, S. Ivankovic and A. v. Hodenberg, Chemical structure and carcinogenicity of aiiphatic hydrazo, azo, and azoxy compounds and of triazenes, potential in vivo alkylating agents, Ann. N.Y. Acad. Sci., 163 (1969) 697. 11 G.F. Kolar, R. Fahrig and E. Vogel, Structure-activity dependence in some novel ring substituted 3,3dimethyl-1-phenyltriazenes. Genetic effects in Drosophila melanogaster and in Sacchasomyces cerevisiae by a direct and host-mediated assay, Chem.Biol. Interact., 9 (19’74) 365. 12 0. Dimroth, Synthesen mit Diazobenzolimid, Ber. Deut. Chem. Ges., 36 (1903) 909; Ueber eine neue Synthese von Diazoaminoverbindungen, 4. Mittheilung iiber Synthesen mit Aziden, Bar. Deut. Chem. Ges., 38 (1905) 670. 13 F.W. Kruger, R. Preussmann and N. Niepelt, Mechanism of carcinogvnesis with l-aryl3,3_diaikyltriazenes, III. In vivo methylation of RNA and DNA with l-phenyi-3,3[ “C]dimethyltriazene, Biochem. Pharmacol., 20 (1970) 529. 14 R.C.S. Audette, T.A. Connors, H.G. Mandel, K. Merai and W.C.J. Ross, Studies on the mechanism of action. of the tumour inhibitory triazenes, Biochem. Pharmacol., 22 (1973) 1855. 15 T.A. Connors, P.M. Goddard, K. Merai, W.C.J. Ross and D.E.V. Wilman, Tumour inhibitory triazenes: Structural requirements for an active metabolite, Biochem. Pharmacol., 25 (1976) 241. 16 G.F. Kolar and J. Schlesiger, unpublished results. 17 C.S. Rondestvedt and S.J. Davis, I-Aryl-3,3_dialkyltriazenes as tumour inhibitors, J. Org. Chem., 22 (1957) 200. 18 G.F. Kolar, Mass spectrometry of carcinogenic alkyltriazenes, in A. Frigerio and N. Castagnoli (Eds.), Mass Spectrometry in Biochemistry and Medicine, Raven, New York, 1974, pp. 267--275. 19 E. Knoevenagel, Ueber die katalytischen Wirkungen des Jods. Kondensation von ound P-Naphthylamin mit Alkoholen, J. Prakt. Chem., 89 (1914) 1. 20 G.F. Kolar, Synthesis of biologically active triazenes from isolable diazonium salts, 2. Naturforsch., 27b (1972) 1183. 21 0. Fischer and E. Hepp, Studien in der Indulingruppe, Ann. Chem. Pharm., 266 (1890) 233. 22 K. Eicker, Zur E nntnis der Phenazine, Ber. Deut. Chem. Ges., 23 (1890) 3803. 23 B.B. Brodie and J. Axelrod, The estimation of acetanilide and its metabolic products aniline, N-acetyl p-aminophenol and p-aminophenol (free and total conjugated) irr biological fluids and tissues, J. Pharmacol. Exptl. Ther., 94 (1948) 22. 24 S.L. Tompsett, The determination of aminophenols and aminobenzoic

310

acids. A study

of extraction procedures applicable to biological materials, Anal. Chim. Acta, 21 (1959) 555. 25 P. KIeihues, G.F. Kolar and G.P. Margison, Interaction of the carcinogen 3,3-dimethyll-phenyltriazene with nucleic acids of various rat tissues and the effect of a proteinfree diet, Cancer Res., 36 (1976) 2169.

311