Chem.-Biol. Interactions, 11 (1975) 535-544
0 Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands
STUDIES ON LUNG TUMOURS III. OXIDATIVE METABOLISM OF DIMETHYLNITROSAMINE RODENT AND HUMAN LUNG TISSUE
L. DEN ENGELSE, M. GEBBINK
AND
535
BY
P. EMMELOT
Chemical Carcinogenesis Unit, Department of Biochemistry, Antoni van Leeuwenhoek-Laboratory, The Netherlands Cancer Institute, Amsterdam (The Netherlands)
(Received February 2&h, 1975) (Revision received April 29th, 1975) (Accepted May 3Oth, 1975)
SUMMARY
The oxidative metabolism of the carcinogen dimethylnitrosamine (DMN) was studied in mouse, rat, hamster and human respiratory tissue. [W]DMN was purified by Dowex-1-bisulfite column chromatography to remove a contaminant (probably [W]formaldehyde) interfering with the enzyme assay. Since formaldehyde and methyl carbonium ions - yielding methanol with water - are considered to be the primary products of DMN metabolism, tissue slices were assayed for the production of [l%]CO2 from W-labelled methanol, formaldehyde, formate, and DMN. Oxidation of formaldehyde to formate was not, but oxidation of formate to CO2 was very much rate-limiting. This rate-limiting step was circumvented by introducing quantitative chemical oxidation of formate to CO2 by mercury(II)chloride following the enzymic reaction. Since oxidation of methanol to CO2 proved to be insignificant, production of CO2 from DMN by lung tissue enzymes and HgClz may serve as a parameter for N-demethylating activity and the production of the suspected carcinogenically active methyl carbonium ions. The DMN-N-demethylating activities of lung tissue slices of two mouse strains with widely different susceptibilities to formation of lung adenomas by DMN differed significantly, but the difference seemed too small to explain the divergence in tumourigenic response. The enzymatic activities decreased in hamster bronchus, hamster trachea, hamster lung, GRSJA mouse lung, C3Hf/A mouse lung, human lung, Sprague-Dawley rat lung, in that order. The reported resistance of the hamster Abbreviation: DMN, dimethylnitrosamine.
536 respiratory system to tumour induction by DMN may therefore not be due to poor DMN-N-demethylating capacity.
m-rRODUCTION
Although many carcinogens need to be enzymatically activated to manifest their carcinogenic properties, the extent of enzymatic activation or binding of activated intermediates to cellular macromolecules generally does not correlate with the degree of susceptibility to these compounds 1~2.This is in line with the hypothesis of cancer as a multi-stage process in which a number of successive factors/events (e.g., metabolic (in)activation, interaction with DNA, repair processes, hormonal environment, antigenic structure of tumour cells) determine whether a tumour will ultimately develop or not. Low levels of metabolic activation of carcinogens have been thought largely responsible for resistance to N-acetylaminofluorene in the guinea pig3, aflatoxin in mice4, and DMN in the Syrian golden hamsteF8. This investigation concerns the metabolism of DMN in the rodent respiratory system. It is shown that, in contradistinction to reports by other author+s, the remarkable resistance of the hamster respiratory tract to DMN carcinogenesis cannot be ascribed to the very low degree or even the lack of DMN metabolism in the hamster respiratory system. We also investigated whether and if so to what extent the lungs of C3Hf/A and GRS/A male mice were able to N-demethylate DMN. In previous experiments we had found that a single injection of DMN resulted in the formation of high numbers of lung adenomas in GRS/A ma!e mice whereas male C3Hf/A mice were not very susceptible to lung tumour formation by this carcinogenl. Whereas the DMN-N-demethylating activity of liver can be measured by the colorimetrical assay of formaldehyde produced from DMN by isolated microsomes, production of [tJC]CO2 from [t%]DMN by tissue slices is used in the case of low activities such as may be expected in the lung. Because CO2 production correctly reflects DMN metabolism only when the metabolic conversion of formaldehyde via fcrmate to COz is not a rate-limiting step, we first tested the conversion of [‘%Iformaldehyde and [*4C]formate to [tJC]COz by mouse dung slices. The enzyme capacity to convert formate to COz proved to be totally insufficient, independent o,f the amount of substrate. Only after introducing chemical oxidation of formate by mercury(II)chlo could CO2 production be used as a measure of DMN metabolism. Regarding the variety of nitrosamines and other carcinogens which are capable of inducing tumours of the hamster respiratory system, the resistance of this species to DMN8Jo-1a is remarkable. Since hamster lung slices might, like mouse lung slices, be poor oxidizers of formate, the DMN-IV-demethylating activities of hamster lung, primary bronchus and trachea were determined using chemical oxidation of formate. Human lung and bronchus were also tested for their capacity to metabolize
537 DMN, it having been suggested that nitrosamines might contribute to the process of human respiratory tract tumourigenesisl4Js. MATERIALS
AND METHODS
Chemicals
[rJC]DMN (8 mCi/mmole) was prepared according to Dutton and Heath16 and purified by Dowex-1 in the bisulfite cycle to remove contaminating [r4C]formaldehyder7 interfering in the assay for enzymatic DMN-N-demethylation by conversion to [14C]C02. [14C]Methanol (4 mCi/mmole) and [r‘C]formaldehyde (10 mC /mmole) were obtained from New England Nuclear (Boston, Mass., U.S.A.). [t%]Sodium formate (54 mCi/mmole) and [%]sodium bicarbonate (0.1 mCi/mmole) were purchased from the Radiochemical Centre (Amersham, U.K.). Animals
Conventional animals were used. Male mice of the inbred strains CSHf/A and GRS/A (formerly1 called C3Hf and GR) were maintained as described previously* and used for enzymatic experiments at the age of 13-19 weeks. Male SpragueDawley rats from the breeding colony of our Institute were used at the age of 13-14 weeks. Syrian golden hamsters (males) were purchased from the TNO-Central Institute for the breeding of laboratory animals (Zeist, The Netherlands), and used when 8-10 weeks old. Prior to all experiments the animals were subjected for at least 10 days to a light-dark regimen (start of the 12-h light period at 07.00 a.m.). In all rodent experiments tissues were removed between 10.30 and 12.00 h and used as rapidly as possible. Human respiratory
tissue
Freshly removed human respiratory tissue was obtained from patients lobectomized after diagnosis of primary iung neoplasia. The period between removal and enzymatic incubation amounted to maximally 2 h. As judged macroscopically, lung tissue assayed for DMN metabolism was tumour-free. In two cases tumour tissue was included in the enzyme assay. Enzymatic
experiments
Incubations were carried out at 37” in Warburg respiratory vessels under an atmosphere of 95 % OL~5% COz. The standard reaction mixture (total volume 2 ml) contained NaCl (8 mg), KC1 (0.4 mg), MgS04.7 Hz0 (2 mg), glucose (3.6 mg), sodium pyruvate (2.2 mg), phosphate buffer (5 mM, pH 7.5) and streptomycin sulfate (0.1 mg). To determine the influence of pH on the production of [14C]CG2from [i4C]DMN, 5 rnfid phosphate was replaced by citric acid, KH2PG4, Tris, KC1 (all 1.125 mM) and sodium tetraborate (0.56 mM) adjusted with NaOH or HCI. [14C]C02 was absorbed in 0.1 ml 4 N KOH on a Whatman GF/C glass fibrc disk in the centre well. KOH was used as a trapping agent because it permitted nearly quantitative evaporation of unchanged [14C]DMN, which became absorbed to the disk, by heating
538 in VUC~O. Enzymatic experiments were stopped after 60 min by the addition of 0.30 ml 3 M phosphoric acid (adjusted to pH 3.0 with NaOH); [14C]COQabsorption was completed by subsequent incubation at 37” for 15 min. When chemical oxidation of formate to CO2 had to be carried auf, 250 mg HgCl2 was added to each vessel after stopping the enzymatic reaction followed by 40 min incubation at 85” (c$ ref. 9). In all experiments controls without tissue ([Xlsubstrate with or without HgCl2) were included and used to correct the tissue data. The glass fibre disks were collected and absorbed [14C]DMN was removed by heating at 80-90” in IUCUOduring 60 min (in some experiments disks were dried 60 min under infra-red radiation). The dried papier was immediately placed in a toluene-omnifluor scintillator and counted in a Nuclear Chicago Mark II liquid scintillation counter with an efficiency of 62.8%. After removal of the glass fibre disk, the total protein content of each vessel was determined with the Folin phenol reagent 1s. In a series of control experiments on the efficiency if formate oxidation by HgCl2,20 ~1 hyamine hydroxide (1 nli in methanol) was placed on an 8 x 20 mm strip of Whatman No. 1 filter paper. The filter paper was transferred to a counting vial containing 1 ml water and 15 ml scintillation solvent after HattorilQ. RESULTS
Conversiorzof formate into CO2 In a typical experiment C3Hf/A lung slices were incubated with 1.5 nmoles [l‘%]formate. Only 3.8 % of the added radioactivity was recovered as COz, indicating that only 0.062 nmole formate had been oxidized to COz. Using HgClQ at least 98.4% of added formate was oxidized to CO2 whereas total recovery of radioactivity in the centre well amounted to 89, f 1 y0 (hyamine hydroxide on filter paper) and 92 f 3 % (KOH on glass fibre disk) respectively. These figures correspond closely to those of control experiments in which the efficiency of [lJC]CO2 absorption (15 min, 37’) after addition of phosphoric acid to an alkaline solution of [l”C] sodium bicarbonate was found to be 92 f 3 %. As shown in Fig. 1 the percentage of formate recovered as CO2 was independent of the formate concentration up to 1.66 nmole/ vessel. Con version offormaldehyde intoformate It was found that a quantity of mouse, rat, or hamster lung slices, corresponding to 5-l 5 mg protein, oxidized at least 90 % of the added formaldehyde (1.O nmole) to formate (60. min; standard incubation conditions; chemical oxidation of formate by HgClz). The corresponding figure for human lung slices was about 60%. Conversion of methanol into formaldehyde Because part of the [l%]COQ might be derived from methyl carbonium ions, yielding methanol after reaction with water, the conversion of methanol into formaldehyde was measured. Incubation of 1.5 nmoles [l%]methanol with rodent lung slices (60 min; standard incubation conditions; chemical oxidation of formate by
539
B
I
I
1
I
Xl
100
150
200
dpm formata added ; x lo3 Fig. I. Oxidation of [Wlformate by mercury(W chloride. Recovery of radioactivity on the glass fibre disk in the centre well. Each point represents the mean of 2 experiments; individual experiments are indicated by the corresponding horizontal bars.
HgCls) indicated that only minimal amounts of methanol were converted into formaldehyde (GRS./A: 2.9 %; C3Hf/A: I .6’;/,; Sprague-Dawley : 1.2 %; Syrian golden hamster: < 0.1%). Owing to these low figures the [r%]CO2 production from [tJC]DMN by lung tissue may be interpreted as resulting from formaldehyde directly produced by DMN-N-demethylation. DMN-
N-derrvthylatiow csperinwnts; standard conditions
Standard conditions for [W]CO2 production from [W]DMN were established for male C3Hf/A lung slices and applied to all species. A pH optimum between pH 7.0 and 10.0 was found by incubating 100 nmoles [tW]DMN with an amount of C3Hf/A lung slices corresponding to 5-15 mg protein in the standard reaction mixture in which phosphate buffer was replaced by universal buffer (pH 2-12; see Enzymatic c~sperinwnts). pH-Dependency in the pH 7-10 region was small and a standard pH of 7.5 was selected for further experiments. The production of labelled CO2 was roughly proportional to the protein concentration of up to I5 mg protein per incubation. Although [t4C]COs production from labelled DMN (100 nmoles) by C3Hf/A lung slices (5-13 mg protein/incubation; pH 7.5) was still observed after 60 min, incubation experiments were generally terminated after 60 min. As shown previouslytr, complete substrate saturation with C3HfiA male mouse lung slices could be obtained by adding increasing amounts of unhbelled DMN to 100 nmoles [t%]DMN. This is important because no complete substrate saturation could
540
be obtained with crude [~~C~DMN preparations, i.e., preparations not purified by Dowex-1-bisuhite column chromatography to remove contaminating [r%]formaldehyder’. Although substrate saturation of the DMN-IV-demethylating enzyme system was obtained at 20-50 m&f DMN, a standard concentration of 50 PM DMN was preferred for two reasons. Firstly, the latter concentration is comparable to concentrations in body fluids after injection of low doses (about 3.5 mg/kg body wt.) of DMN into ex~rimental animals; secondly, substrate saturation levels would consume huge amounts of labelled substrate and lead to low and therefore less accurate 1% measurements. As a consequence the potential DMN-IV-demethylating activity (at least in C3Hf/A male mouse lung &es) may be much higher than that observed at a DMN concentration of 50 @I#. No significant DMN-~.demethyla~ing activities were found in blood or in preheated C3Hf/A lung slices (10 min/lOO”). ~~N-N-de~~et~y~ati~g activities in the rodent respiratory system The results of the enzymatic assays are summarized in Table I. Determinations with and without HgCla oxidation were carried out and showed for all rodent species a 10 to 20-fold r~uction in [r*C]COa prod~lction from ~14C~DMN when oxidation by HgCla was omitted from the assay. We regard this as evidence that the metabolic conversion of formate into CO2 is rate-limiting not only in mouse but also in rat and hamster respiratory tissue. DMN metabolism in rodents was found to be highest in hamster bronchus followed by hamster trachea, hamster lung, GRS/A mouse lung, C3Hf/A mouse lung and Sprague-Dawley rat lung in that order. Differences between hamster lung, bronchus and trachea were not statistically significant. DMN-N-demethylating activities in kumur~respiratory tissue Results of enzyme assays are given in Table II. Since material was obtained from patients with primary lung tumours and since the presence of a local tumour
TABLE I METABOLISM OF
[“CIDMN
BY RODENT
RESPIRATORY
TISSUES
-
Species
S&dP
Tissx?
Rat
Sprague-Dawley C3Hf/A GRS/A Syrian ;{olden Syrian golden Syrian golden
Iliilg
Nouse
Mouse Hamster Hamster Hamster
lung lung IllIlg
trachea bronchus
ptI?&s ~~~~~~1~ pro?e6/60 nlW
5.0
& l.OC
34.2 & Il.6 56.6 ri: 12.4 100.4 & 47.4 110.0 It 32.2” 149.0 f 45.4
i 0.001
: 0.01 *: 0.05 not significant not significant
B Males were used in all experiments. b Under standard condttiom. c Alifigures(exceptd) representmean 4 standard deviation of 8 determinations (8 animals). d Mean 41standard deviationof 4 determinations (4 animals).
541 TABLE I I PRODUCTION OF [‘T]co,
FROM
[‘4C]DMN
Patient no.
Sex
Age (years)
1
0
61
BY WUMAN LUNG TISSUE
Tumour
Alveolar
ptnoles COz/mg proteinl60 mit?
cell
carcinoma 2
cf
52
squtimous cell
3
d
69
carcinoma !Squamouscell carcinoma
lS.lb 9.4 7.8 10.8 12.5 19.0
a Under standard conditions. b All figures represent single experiments.
may influence metabolic processes in non-tumourous parts of the lobe concerned, our results require a cautious interpretation. Human lung slices metabolized DMN at a rate significantly higher than rat lung (P < 0.001; Student’s ( test). Comparison of human and rat liver slices”O showed higher DMN-ALdemethylating activities in the rat liver (Wistar, female). Production of [*%Z]CO2from [‘%]DMN by slices from both human bronchi (1.7 f 0.7 pmoles COz/mg protein/60 min) and squamous cell carcinomas (1.5 f 0.3 pmoles COz/mg protein/60 min) were not significantly different from control values. DISCUSSION
The production of [l%]CO2 from [%]DMN by lung slices was studied. Whereas in C3Hf/A male mouse liver slices the primary product of DMN metabolism (formaldehyde) is converted nearly quantitatively into formate and, subsequently, into carbon dioxide (not illustrated), the capacity of mouse lung slices to produce [l%]CO2 from [laC]formaldehyde was found to be insufficient even for the low amounts of formaldehyde arising during DMN metabolism. This was shown to be due to a very low oxidation rate of formate. Chemical oxidation of formate by mercury(II)chlorideS was therefore carried out after enzymatic incubation. By this method at least 98.4% of added formate was oxidized to CO2. In all further experiments there was a IO-fold (or greater) increase in [l%]CO2 production by postincubation with HgCls; this was taken as evidence that the enzymatic formatecarbon dioxide conversion was rate-limiting. Parenthetically, the finding that respiratory tissue slices produced IO-20 times more formate than CO2 from DMN suggests that the reaction mechanism proposed by Kriiger ef ~I.21 in which CO2 instead of formaldehyde is directly split from the oxidized DMN molecule (giving rise to more CO2 than formate) does not apply to lung tissue. The difference between the DMN-IV-demethylating activities of GRSlA and C3Hf/A lung slices is small, showing that the susceptibility of these strains to lung adenoma formation after administration of DMN is not correlated with DMN-N-
542 demethylating activities. The relatively high DMN-N-demethylating activity of hamster lung slices as reported in this paper (about 0.75 % of 1% added recovered as [14C]COz under standard conditions) contrasts with results obtained by Montesano and Magee6*22. These authors reported a very low DMN metabolism in hamster lung slices (0.07% of 14C added recovered as [W]COs) and concluded that this finding might explain the resistance of the hamster respiratory system to tumour formation by DMN. The dramatic rise in [W]COs production in the presence of HgCls as reported in this paper for lung tissue including that of the hamster suggests that the low DMN-N-demethylating activity as reported by these author&s2 may well be due to the insufficient capacity for formate oxidation on the part of hamster lung slices in their experiments. Moreover, in a pilot experiment we measured the methylation of hamster lung DNA (7-methylguanine formation) after [t%]DMN injection and found it to be only slightly lower than the values previously obtained for mouse lung DNA’. Since in vitro metabolism of DMN by hamster lung slices is relatively high as compared to that by mouse lung slices (Table I) this (preliminary) result suggests that the amount of 7-methylguanine in lung DNA after injection of DMN into the intact animal depends equally on other factors such as the rate of carcinogen disappearance from the circulation or the in riro rate at which DNA-carcinogen interaction products disappear from cellular DNA. Our results indicate that hamster lung DNA is methylated to an appreciable extent by DMN and that factors other than metabolic activation determine the resistance of the hamster respiratory tract to DMN. Since tumours in the hamster respiratory system were observed not only after administration of diethylnitrosamine23, di-,z-propylnitrosamine24, dibutylnitrosaminees, and a number of cyclic nitrosamines ~26 but also after administration of the methylating agent methylnitrosourea 27, the resistance to DMN seems to stem from a highly specific mechanism. This specificity might be related either to the specific pattern of DNA alkylation by DMNs*, resulting in very rapid repair, or to the high toxicity of DMN giving rise to cell death or to (temporary) loss of proliferating capacity. In the rat lung only low DMN-N-demethylating activities were found. This might be due to the relative resistance of rat lung to DMN tumourigenesis (see ref. 6), though factors other than metabolic activation may play a more decisive role here. Metabolism of DMN by human lung tissue indicates that one of the requirements for tumour formation by DMN in this organ is satisfied. The low and insignificant [r4C]COa production from [WJDMN by human bronchus slices contrasts with activities observed in the hamster respiratory system. Since this result may be related to patients’ age and to the presence of respiratory neoplasia no definite conclusion with regard to DMN metabolism by the bronchiolar epithelium of young and healthy men and women can be drawn at this moment. The absence of significant DMN metabolism in lung tumour slices is in agreement with earlier reports on rat hepatoma DMN N-demethylating activitieP. Only further investigations can show whether or not nitrosamines contribute to human respiratory system tumourigenesis.
543 ACKNOWLELXXMENTS
Partly supported by the Resarch Fund of the Scientific Advisory Committee on Smoking and Health. This Fund was established by the Cigarette Industry Foundation, The Hague, The Netherlands. Thanks are due to Dr. A. M. Griindemann and his colleagues of the Department of Thorax Surgery, Onze Lieve Vrouwe Gasthuis, Amsterdam, for providing human lung tissue and to Mr. E. Philippus for skilled technical assistance.
REFERENCFS I L. den Engek, P. A. J. RentwIzen and P. Emmelot, Studies on lung tumours, 1. Methylation of deoxyribonucleic acid and tumour formation following administration of dimethylnitrosamine to mioe, Che/n..-Biol. Infefucf., I (1969/1970) 395-406. 2 L. den Engelse, The formation of methylated bases in DNA by dimethylnitrosamine and its relation to differences in the formation of tumours in the livers of GR and C3Hf mice, C’herpr.-Biol. Ittteract., 8 (1974) 329-338. 3 E. C. Miller. J. A. Miller and M. Enomoto. The comparative carcinogenicities of 2-acetylaminofluorene and its N-hydroxy metabolitc in mice, hamsters, and guinea pigs, Cancer Res., 24 (1964) 2018-2032. 4 R. S. Portman, K. M. Plowman and T. C. Campbell, Aflatoxin metabolism by liver microsomal preparations of two different species, BiochpIIl. Biophys. Res. Cor~mun., 33 (1968) 711-715. 5 R. Montesano, Systemic carcinogens (IV-nitroso compounds) and synergistic or additive effects in respiratory carcinogenesis, Twtori, 56 (1970) 335-344. 6 R. Montesano and P. N. Magee, Comparative metabolism in vitro of nitrosamines in various animal species including man, in R. Montesano, L. Tomatis and W. Davis (Eds.), Chemical Curcittogenesis Essays, IARC Sc. Publ. IO, WHO-IARC, 1974, p. 39-56. 7 H. Bartsch, C. Malaveille and R. Montesano, in vitro metabolism and microsome-mediated mutagenicity of dialkvlnitrosamines in rat, hamster, and mouse tissues, Cancer Res., 35 (1975) 644-651. 8 F. Stenback, A. Ferrero, R. Montesano and P. Shubik, Synergistic effect of ferric oxide on dimethylnitrosamine carcinogenesis in the Syrian golden hamster, Z. Krebsforsch., 79 (1973) 31-38. 9 S. F. Yang, A simple method for assay of 1Gformic acid, Anal. Biochem., 32 (1969) 519-521. IO L. Tomatis, P. N. Magee and P. Shubik, Induction of liver tumors in the Syrian golden hamster by feeding dimethylnitrosamine, J. Nut/. Cancer Ins:., 33 (1964) 341-345. 1 I K. McD. Herrold, Histogenesis of malignant liver tumors induced by dimethylnitrosamine. An experimental study in Syrian hamsters, J. Nufl. Cancer bat., 39 (1967) 1099-l I 1I. I2 W. Dontenwill, Experimental studies on the organotropic effect of nitrosamines in the respiratory tract, Food Cowret. Toxicol., 6 (1968) 571. 13 H. Haas, U. Mohr and F. W. Kriiger. Comparative studies with different doses of N-nitrosomorpholine, N-nitrosopiperidine, N-nitrosomethylurea and dimethylnitrosamine in Syrian golden hamsters, J. Nat/. Cower Inst., 51 (1973) 1295-l 301. 14 W. Liiinsky, H. W. Taylor, C. Snyder and P. Nettesheim, Malignant tumours of liver and lung in rats fed aminopyrine or heptamethyleneimine together with nitrite, Nature, 244 (1973) 176-178. I5 Anonymous, Environmental nitrosamines, turrcel, 11 (1973) 1243-1244. I6 A. Dutton and D. F. Heath, The preparation of [lXI]dimethylamine and [lJCldimethylnitrosaminc. J. Chnrr. Ser., (1956) l892-.1893. 17 L. den Engelse, M. Gcbbink and E. Philippus, [lVI]Formaldehyde: a possible contaminant of (‘,‘C]dimethylnitrosamine, Chem-Biol. /&racr., I I (1975) 133-137. I8 0. H. Lo\hry. N. J. Rosebrough, A. L. Farr and R. J. Randall, Protein measurement with the Folin phenol reagent, J. Biol. Chen~., 193 (1951) 265-275. I9 T. Hattori, H. Aoki. 1. Matsuzaki, B. Maruo and H. Takahashi, Liquid scintillation counting of HQucleic acids, Ad. Chent., 37 (1965) 159-160. 20 R. Montesano and P. N. Magee, Metabolism of dimethylnitrosamine by human liver slices in vitro, N&di e, 228 (1970) 173-l 74.
544 21 F. W. Kri.iger, H. Osswald, G. Walker and E. Schelten, investigations on the organotropy of alkylating activity of dimethylnitrosamine (DMNA) and nitroso-methyl-urea (NMH) in G’iyo, 2. Kreb@rsch. 74 (1970)434-447. 22 R. Montesano and P. N. Magee, Metabolism of nitrosamines by rat and hamster tissue s ices in vitro, Proc. Am. Ass. Cancer Res., 12 (1971) 14. 23 V. J. Feron, P. Emmelot and T. Vossenaar, Lower respiratory tract tumours in Syrian golden hamsters after intratracheal instillations of diethylnitrosamine alone and with ferric oxide, Eur. J. Cancer, 8 (I 972) 445419. 24 P. Pour, F. W. Krikger, A. Cardesa, J. Althoff and U. Mohr, Carcinogenic effect of di-n-propylnitrosamine in Syrian golden hamsters, J. Nad Cllncer Inst., 51 (1973! 1019-1027. 25 J. Althoff, Simultaneous tumors in the respiratory system and urinary bladder of Syrian golden hamsters, Z. Krebs/brsch., 82 (1974) 153-l 58. 26 J. Althoff, R Wilson, A. Cardesa and P. Pour, Comparative studies of neoplastic response to a single dose of nitroso compounds, 3. The e&t of IV-nitrosopiperidine and IV-nitrosomorpholinc in Syrian golden hamsters, Z. Krebsforsch., 81 (1974) 251-259. 27 H. Schreiber, K. Schreiber and D. H. Martin, Experimental tumor induction in a circumscribed region of the hamster trachea: correlation of histology and exfoliative cytology, J. NatI. Cuncer Ins,., 54 (1975) 187-197. 28 R. Ramanathan, D. S. R. Sarma, S. Rajalakshmi and E. Farber, Non-random methylation of rat liver chromatin DNA by dimethylnitrosamine in VI’VO, Proc. Am. Ass. Cancer Res., 16 (1975) abstr. 36. 29 J. A. J. Brouwers and P. Emmelot, Microsomal N-demethylation and the effect of the hepatic carcinogen dimethylnitrosamine on amino acid incorporation into the proteins of rat livers and hepatomas, Exprl. Cell Res., 19 (1960) 467-474.
0 Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands
STUDIES ON LUNG TUMOURS III. OXIDATIVE METABOLISM OF DIMETHYLNITROSAMINE RODENT AND HUMAN LUNG TISSUE
L. DEN ENGELSE, M. GEBBINK
AND
535
BY
P. EMMELOT
Chemical Carcinogenesis Unit, Department of Biochemistry, Antoni van Leeuwenhoek-Laboratory, The Netherlands Cancer Institute, Amsterdam (The Netherlands)
(Received February 2&h, 1975) (Revision received April 29th, 1975) (Accepted May 3Oth, 1975)
SUMMARY
The oxidative metabolism of the carcinogen dimethylnitrosamine (DMN) was studied in mouse, rat, hamster and human respiratory tissue. [W]DMN was purified by Dowex-1-bisulfite column chromatography to remove a contaminant (probably [W]formaldehyde) interfering with the enzyme assay. Since formaldehyde and methyl carbonium ions - yielding methanol with water - are considered to be the primary products of DMN metabolism, tissue slices were assayed for the production of [l%]CO2 from W-labelled methanol, formaldehyde, formate, and DMN. Oxidation of formaldehyde to formate was not, but oxidation of formate to CO2 was very much rate-limiting. This rate-limiting step was circumvented by introducing quantitative chemical oxidation of formate to CO2 by mercury(II)chloride following the enzymic reaction. Since oxidation of methanol to CO2 proved to be insignificant, production of CO2 from DMN by lung tissue enzymes and HgClz may serve as a parameter for N-demethylating activity and the production of the suspected carcinogenically active methyl carbonium ions. The DMN-N-demethylating activities of lung tissue slices of two mouse strains with widely different susceptibilities to formation of lung adenomas by DMN differed significantly, but the difference seemed too small to explain the divergence in tumourigenic response. The enzymatic activities decreased in hamster bronchus, hamster trachea, hamster lung, GRSJA mouse lung, C3Hf/A mouse lung, human lung, Sprague-Dawley rat lung, in that order. The reported resistance of the hamster Abbreviation: DMN, dimethylnitrosamine.
536 respiratory system to tumour induction by DMN may therefore not be due to poor DMN-N-demethylating capacity.
m-rRODUCTION
Although many carcinogens need to be enzymatically activated to manifest their carcinogenic properties, the extent of enzymatic activation or binding of activated intermediates to cellular macromolecules generally does not correlate with the degree of susceptibility to these compounds 1~2.This is in line with the hypothesis of cancer as a multi-stage process in which a number of successive factors/events (e.g., metabolic (in)activation, interaction with DNA, repair processes, hormonal environment, antigenic structure of tumour cells) determine whether a tumour will ultimately develop or not. Low levels of metabolic activation of carcinogens have been thought largely responsible for resistance to N-acetylaminofluorene in the guinea pig3, aflatoxin in mice4, and DMN in the Syrian golden hamsteF8. This investigation concerns the metabolism of DMN in the rodent respiratory system. It is shown that, in contradistinction to reports by other author+s, the remarkable resistance of the hamster respiratory tract to DMN carcinogenesis cannot be ascribed to the very low degree or even the lack of DMN metabolism in the hamster respiratory system. We also investigated whether and if so to what extent the lungs of C3Hf/A and GRS/A male mice were able to N-demethylate DMN. In previous experiments we had found that a single injection of DMN resulted in the formation of high numbers of lung adenomas in GRS/A ma!e mice whereas male C3Hf/A mice were not very susceptible to lung tumour formation by this carcinogenl. Whereas the DMN-N-demethylating activity of liver can be measured by the colorimetrical assay of formaldehyde produced from DMN by isolated microsomes, production of [tJC]CO2 from [t%]DMN by tissue slices is used in the case of low activities such as may be expected in the lung. Because CO2 production correctly reflects DMN metabolism only when the metabolic conversion of formaldehyde via fcrmate to COz is not a rate-limiting step, we first tested the conversion of [‘%Iformaldehyde and [*4C]formate to [tJC]COz by mouse dung slices. The enzyme capacity to convert formate to COz proved to be totally insufficient, independent o,f the amount of substrate. Only after introducing chemical oxidation of formate by mercury(II)chlo could CO2 production be used as a measure of DMN metabolism. Regarding the variety of nitrosamines and other carcinogens which are capable of inducing tumours of the hamster respiratory system, the resistance of this species to DMN8Jo-1a is remarkable. Since hamster lung slices might, like mouse lung slices, be poor oxidizers of formate, the DMN-IV-demethylating activities of hamster lung, primary bronchus and trachea were determined using chemical oxidation of formate. Human lung and bronchus were also tested for their capacity to metabolize
537 DMN, it having been suggested that nitrosamines might contribute to the process of human respiratory tract tumourigenesisl4Js. MATERIALS
AND METHODS
Chemicals
[rJC]DMN (8 mCi/mmole) was prepared according to Dutton and Heath16 and purified by Dowex-1 in the bisulfite cycle to remove contaminating [r4C]formaldehyder7 interfering in the assay for enzymatic DMN-N-demethylation by conversion to [14C]C02. [14C]Methanol (4 mCi/mmole) and [r‘C]formaldehyde (10 mC /mmole) were obtained from New England Nuclear (Boston, Mass., U.S.A.). [t%]Sodium formate (54 mCi/mmole) and [%]sodium bicarbonate (0.1 mCi/mmole) were purchased from the Radiochemical Centre (Amersham, U.K.). Animals
Conventional animals were used. Male mice of the inbred strains CSHf/A and GRS/A (formerly1 called C3Hf and GR) were maintained as described previously* and used for enzymatic experiments at the age of 13-19 weeks. Male SpragueDawley rats from the breeding colony of our Institute were used at the age of 13-14 weeks. Syrian golden hamsters (males) were purchased from the TNO-Central Institute for the breeding of laboratory animals (Zeist, The Netherlands), and used when 8-10 weeks old. Prior to all experiments the animals were subjected for at least 10 days to a light-dark regimen (start of the 12-h light period at 07.00 a.m.). In all rodent experiments tissues were removed between 10.30 and 12.00 h and used as rapidly as possible. Human respiratory
tissue
Freshly removed human respiratory tissue was obtained from patients lobectomized after diagnosis of primary iung neoplasia. The period between removal and enzymatic incubation amounted to maximally 2 h. As judged macroscopically, lung tissue assayed for DMN metabolism was tumour-free. In two cases tumour tissue was included in the enzyme assay. Enzymatic
experiments
Incubations were carried out at 37” in Warburg respiratory vessels under an atmosphere of 95 % OL~5% COz. The standard reaction mixture (total volume 2 ml) contained NaCl (8 mg), KC1 (0.4 mg), MgS04.7 Hz0 (2 mg), glucose (3.6 mg), sodium pyruvate (2.2 mg), phosphate buffer (5 mM, pH 7.5) and streptomycin sulfate (0.1 mg). To determine the influence of pH on the production of [14C]CG2from [i4C]DMN, 5 rnfid phosphate was replaced by citric acid, KH2PG4, Tris, KC1 (all 1.125 mM) and sodium tetraborate (0.56 mM) adjusted with NaOH or HCI. [14C]C02 was absorbed in 0.1 ml 4 N KOH on a Whatman GF/C glass fibrc disk in the centre well. KOH was used as a trapping agent because it permitted nearly quantitative evaporation of unchanged [14C]DMN, which became absorbed to the disk, by heating
538 in VUC~O. Enzymatic experiments were stopped after 60 min by the addition of 0.30 ml 3 M phosphoric acid (adjusted to pH 3.0 with NaOH); [14C]COQabsorption was completed by subsequent incubation at 37” for 15 min. When chemical oxidation of formate to CO2 had to be carried auf, 250 mg HgCl2 was added to each vessel after stopping the enzymatic reaction followed by 40 min incubation at 85” (c$ ref. 9). In all experiments controls without tissue ([Xlsubstrate with or without HgCl2) were included and used to correct the tissue data. The glass fibre disks were collected and absorbed [14C]DMN was removed by heating at 80-90” in IUCUOduring 60 min (in some experiments disks were dried 60 min under infra-red radiation). The dried papier was immediately placed in a toluene-omnifluor scintillator and counted in a Nuclear Chicago Mark II liquid scintillation counter with an efficiency of 62.8%. After removal of the glass fibre disk, the total protein content of each vessel was determined with the Folin phenol reagent 1s. In a series of control experiments on the efficiency if formate oxidation by HgCl2,20 ~1 hyamine hydroxide (1 nli in methanol) was placed on an 8 x 20 mm strip of Whatman No. 1 filter paper. The filter paper was transferred to a counting vial containing 1 ml water and 15 ml scintillation solvent after HattorilQ. RESULTS
Conversiorzof formate into CO2 In a typical experiment C3Hf/A lung slices were incubated with 1.5 nmoles [l‘%]formate. Only 3.8 % of the added radioactivity was recovered as COz, indicating that only 0.062 nmole formate had been oxidized to COz. Using HgClQ at least 98.4% of added formate was oxidized to CO2 whereas total recovery of radioactivity in the centre well amounted to 89, f 1 y0 (hyamine hydroxide on filter paper) and 92 f 3 % (KOH on glass fibre disk) respectively. These figures correspond closely to those of control experiments in which the efficiency of [lJC]CO2 absorption (15 min, 37’) after addition of phosphoric acid to an alkaline solution of [l”C] sodium bicarbonate was found to be 92 f 3 %. As shown in Fig. 1 the percentage of formate recovered as CO2 was independent of the formate concentration up to 1.66 nmole/ vessel. Con version offormaldehyde intoformate It was found that a quantity of mouse, rat, or hamster lung slices, corresponding to 5-l 5 mg protein, oxidized at least 90 % of the added formaldehyde (1.O nmole) to formate (60. min; standard incubation conditions; chemical oxidation of formate by HgClz). The corresponding figure for human lung slices was about 60%. Conversion of methanol into formaldehyde Because part of the [l%]COQ might be derived from methyl carbonium ions, yielding methanol after reaction with water, the conversion of methanol into formaldehyde was measured. Incubation of 1.5 nmoles [l%]methanol with rodent lung slices (60 min; standard incubation conditions; chemical oxidation of formate by
539
B
I
I
1
I
Xl
100
150
200
dpm formata added ; x lo3 Fig. I. Oxidation of [Wlformate by mercury(W chloride. Recovery of radioactivity on the glass fibre disk in the centre well. Each point represents the mean of 2 experiments; individual experiments are indicated by the corresponding horizontal bars.
HgCls) indicated that only minimal amounts of methanol were converted into formaldehyde (GRS./A: 2.9 %; C3Hf/A: I .6’;/,; Sprague-Dawley : 1.2 %; Syrian golden hamster: < 0.1%). Owing to these low figures the [r%]CO2 production from [tJC]DMN by lung tissue may be interpreted as resulting from formaldehyde directly produced by DMN-N-demethylation. DMN-
N-derrvthylatiow csperinwnts; standard conditions
Standard conditions for [W]CO2 production from [W]DMN were established for male C3Hf/A lung slices and applied to all species. A pH optimum between pH 7.0 and 10.0 was found by incubating 100 nmoles [tW]DMN with an amount of C3Hf/A lung slices corresponding to 5-15 mg protein in the standard reaction mixture in which phosphate buffer was replaced by universal buffer (pH 2-12; see Enzymatic c~sperinwnts). pH-Dependency in the pH 7-10 region was small and a standard pH of 7.5 was selected for further experiments. The production of labelled CO2 was roughly proportional to the protein concentration of up to I5 mg protein per incubation. Although [t4C]COs production from labelled DMN (100 nmoles) by C3Hf/A lung slices (5-13 mg protein/incubation; pH 7.5) was still observed after 60 min, incubation experiments were generally terminated after 60 min. As shown previouslytr, complete substrate saturation with C3HfiA male mouse lung slices could be obtained by adding increasing amounts of unhbelled DMN to 100 nmoles [t%]DMN. This is important because no complete substrate saturation could
540
be obtained with crude [~~C~DMN preparations, i.e., preparations not purified by Dowex-1-bisuhite column chromatography to remove contaminating [r%]formaldehyder’. Although substrate saturation of the DMN-IV-demethylating enzyme system was obtained at 20-50 m&f DMN, a standard concentration of 50 PM DMN was preferred for two reasons. Firstly, the latter concentration is comparable to concentrations in body fluids after injection of low doses (about 3.5 mg/kg body wt.) of DMN into ex~rimental animals; secondly, substrate saturation levels would consume huge amounts of labelled substrate and lead to low and therefore less accurate 1% measurements. As a consequence the potential DMN-IV-demethylating activity (at least in C3Hf/A male mouse lung &es) may be much higher than that observed at a DMN concentration of 50 @I#. No significant DMN-~.demethyla~ing activities were found in blood or in preheated C3Hf/A lung slices (10 min/lOO”). ~~N-N-de~~et~y~ati~g activities in the rodent respiratory system The results of the enzymatic assays are summarized in Table I. Determinations with and without HgCla oxidation were carried out and showed for all rodent species a 10 to 20-fold r~uction in [r*C]COa prod~lction from ~14C~DMN when oxidation by HgCla was omitted from the assay. We regard this as evidence that the metabolic conversion of formate into CO2 is rate-limiting not only in mouse but also in rat and hamster respiratory tissue. DMN metabolism in rodents was found to be highest in hamster bronchus followed by hamster trachea, hamster lung, GRS/A mouse lung, C3Hf/A mouse lung and Sprague-Dawley rat lung in that order. Differences between hamster lung, bronchus and trachea were not statistically significant. DMN-N-demethylating activities in kumur~respiratory tissue Results of enzyme assays are given in Table II. Since material was obtained from patients with primary lung tumours and since the presence of a local tumour
TABLE I METABOLISM OF
[“CIDMN
BY RODENT
RESPIRATORY
TISSUES
-
Species
S&dP
Tissx?
Rat
Sprague-Dawley C3Hf/A GRS/A Syrian ;{olden Syrian golden Syrian golden
Iliilg
Nouse
Mouse Hamster Hamster Hamster
lung lung IllIlg
trachea bronchus
ptI?&s ~~~~~~1~ pro?e6/60 nlW
5.0
& l.OC
34.2 & Il.6 56.6 ri: 12.4 100.4 & 47.4 110.0 It 32.2” 149.0 f 45.4
i 0.001
: 0.01 *: 0.05 not significant not significant
B Males were used in all experiments. b Under standard condttiom. c Alifigures(exceptd) representmean 4 standard deviation of 8 determinations (8 animals). d Mean 41standard deviationof 4 determinations (4 animals).
541 TABLE I I PRODUCTION OF [‘T]co,
FROM
[‘4C]DMN
Patient no.
Sex
Age (years)
1
0
61
BY WUMAN LUNG TISSUE
Tumour
Alveolar
ptnoles COz/mg proteinl60 mit?
cell
carcinoma 2
cf
52
squtimous cell
3
d
69
carcinoma !Squamouscell carcinoma
lS.lb 9.4 7.8 10.8 12.5 19.0
a Under standard conditions. b All figures represent single experiments.
may influence metabolic processes in non-tumourous parts of the lobe concerned, our results require a cautious interpretation. Human lung slices metabolized DMN at a rate significantly higher than rat lung (P < 0.001; Student’s ( test). Comparison of human and rat liver slices”O showed higher DMN-ALdemethylating activities in the rat liver (Wistar, female). Production of [*%Z]CO2from [‘%]DMN by slices from both human bronchi (1.7 f 0.7 pmoles COz/mg protein/60 min) and squamous cell carcinomas (1.5 f 0.3 pmoles COz/mg protein/60 min) were not significantly different from control values. DISCUSSION
The production of [l%]CO2 from [%]DMN by lung slices was studied. Whereas in C3Hf/A male mouse liver slices the primary product of DMN metabolism (formaldehyde) is converted nearly quantitatively into formate and, subsequently, into carbon dioxide (not illustrated), the capacity of mouse lung slices to produce [l%]CO2 from [laC]formaldehyde was found to be insufficient even for the low amounts of formaldehyde arising during DMN metabolism. This was shown to be due to a very low oxidation rate of formate. Chemical oxidation of formate by mercury(II)chlorideS was therefore carried out after enzymatic incubation. By this method at least 98.4% of added formate was oxidized to CO2. In all further experiments there was a IO-fold (or greater) increase in [l%]CO2 production by postincubation with HgCls; this was taken as evidence that the enzymatic formatecarbon dioxide conversion was rate-limiting. Parenthetically, the finding that respiratory tissue slices produced IO-20 times more formate than CO2 from DMN suggests that the reaction mechanism proposed by Kriiger ef ~I.21 in which CO2 instead of formaldehyde is directly split from the oxidized DMN molecule (giving rise to more CO2 than formate) does not apply to lung tissue. The difference between the DMN-IV-demethylating activities of GRSlA and C3Hf/A lung slices is small, showing that the susceptibility of these strains to lung adenoma formation after administration of DMN is not correlated with DMN-N-
542 demethylating activities. The relatively high DMN-N-demethylating activity of hamster lung slices as reported in this paper (about 0.75 % of 1% added recovered as [14C]COz under standard conditions) contrasts with results obtained by Montesano and Magee6*22. These authors reported a very low DMN metabolism in hamster lung slices (0.07% of 14C added recovered as [W]COs) and concluded that this finding might explain the resistance of the hamster respiratory system to tumour formation by DMN. The dramatic rise in [W]COs production in the presence of HgCls as reported in this paper for lung tissue including that of the hamster suggests that the low DMN-N-demethylating activity as reported by these author&s2 may well be due to the insufficient capacity for formate oxidation on the part of hamster lung slices in their experiments. Moreover, in a pilot experiment we measured the methylation of hamster lung DNA (7-methylguanine formation) after [t%]DMN injection and found it to be only slightly lower than the values previously obtained for mouse lung DNA’. Since in vitro metabolism of DMN by hamster lung slices is relatively high as compared to that by mouse lung slices (Table I) this (preliminary) result suggests that the amount of 7-methylguanine in lung DNA after injection of DMN into the intact animal depends equally on other factors such as the rate of carcinogen disappearance from the circulation or the in riro rate at which DNA-carcinogen interaction products disappear from cellular DNA. Our results indicate that hamster lung DNA is methylated to an appreciable extent by DMN and that factors other than metabolic activation determine the resistance of the hamster respiratory tract to DMN. Since tumours in the hamster respiratory system were observed not only after administration of diethylnitrosamine23, di-,z-propylnitrosamine24, dibutylnitrosaminees, and a number of cyclic nitrosamines ~26 but also after administration of the methylating agent methylnitrosourea 27, the resistance to DMN seems to stem from a highly specific mechanism. This specificity might be related either to the specific pattern of DNA alkylation by DMNs*, resulting in very rapid repair, or to the high toxicity of DMN giving rise to cell death or to (temporary) loss of proliferating capacity. In the rat lung only low DMN-N-demethylating activities were found. This might be due to the relative resistance of rat lung to DMN tumourigenesis (see ref. 6), though factors other than metabolic activation may play a more decisive role here. Metabolism of DMN by human lung tissue indicates that one of the requirements for tumour formation by DMN in this organ is satisfied. The low and insignificant [r4C]COa production from [WJDMN by human bronchus slices contrasts with activities observed in the hamster respiratory system. Since this result may be related to patients’ age and to the presence of respiratory neoplasia no definite conclusion with regard to DMN metabolism by the bronchiolar epithelium of young and healthy men and women can be drawn at this moment. The absence of significant DMN metabolism in lung tumour slices is in agreement with earlier reports on rat hepatoma DMN N-demethylating activitieP. Only further investigations can show whether or not nitrosamines contribute to human respiratory system tumourigenesis.
543 ACKNOWLELXXMENTS
Partly supported by the Resarch Fund of the Scientific Advisory Committee on Smoking and Health. This Fund was established by the Cigarette Industry Foundation, The Hague, The Netherlands. Thanks are due to Dr. A. M. Griindemann and his colleagues of the Department of Thorax Surgery, Onze Lieve Vrouwe Gasthuis, Amsterdam, for providing human lung tissue and to Mr. E. Philippus for skilled technical assistance.
REFERENCFS I L. den Engek, P. A. J. RentwIzen and P. Emmelot, Studies on lung tumours, 1. Methylation of deoxyribonucleic acid and tumour formation following administration of dimethylnitrosamine to mioe, Che/n..-Biol. Infefucf., I (1969/1970) 395-406. 2 L. den Engelse, The formation of methylated bases in DNA by dimethylnitrosamine and its relation to differences in the formation of tumours in the livers of GR and C3Hf mice, C’herpr.-Biol. Ittteract., 8 (1974) 329-338. 3 E. C. Miller. J. A. Miller and M. Enomoto. The comparative carcinogenicities of 2-acetylaminofluorene and its N-hydroxy metabolitc in mice, hamsters, and guinea pigs, Cancer Res., 24 (1964) 2018-2032. 4 R. S. Portman, K. M. Plowman and T. C. Campbell, Aflatoxin metabolism by liver microsomal preparations of two different species, BiochpIIl. Biophys. Res. Cor~mun., 33 (1968) 711-715. 5 R. Montesano, Systemic carcinogens (IV-nitroso compounds) and synergistic or additive effects in respiratory carcinogenesis, Twtori, 56 (1970) 335-344. 6 R. Montesano and P. N. Magee, Comparative metabolism in vitro of nitrosamines in various animal species including man, in R. Montesano, L. Tomatis and W. Davis (Eds.), Chemical Curcittogenesis Essays, IARC Sc. Publ. IO, WHO-IARC, 1974, p. 39-56. 7 H. Bartsch, C. Malaveille and R. Montesano, in vitro metabolism and microsome-mediated mutagenicity of dialkvlnitrosamines in rat, hamster, and mouse tissues, Cancer Res., 35 (1975) 644-651. 8 F. Stenback, A. Ferrero, R. Montesano and P. Shubik, Synergistic effect of ferric oxide on dimethylnitrosamine carcinogenesis in the Syrian golden hamster, Z. Krebsforsch., 79 (1973) 31-38. 9 S. F. Yang, A simple method for assay of 1Gformic acid, Anal. Biochem., 32 (1969) 519-521. IO L. Tomatis, P. N. Magee and P. Shubik, Induction of liver tumors in the Syrian golden hamster by feeding dimethylnitrosamine, J. Nut/. Cancer Ins:., 33 (1964) 341-345. 1 I K. McD. Herrold, Histogenesis of malignant liver tumors induced by dimethylnitrosamine. An experimental study in Syrian hamsters, J. Nufl. Cancer bat., 39 (1967) 1099-l I 1I. I2 W. Dontenwill, Experimental studies on the organotropic effect of nitrosamines in the respiratory tract, Food Cowret. Toxicol., 6 (1968) 571. 13 H. Haas, U. Mohr and F. W. Kriiger. Comparative studies with different doses of N-nitrosomorpholine, N-nitrosopiperidine, N-nitrosomethylurea and dimethylnitrosamine in Syrian golden hamsters, J. Nat/. Cower Inst., 51 (1973) 1295-l 301. 14 W. Liiinsky, H. W. Taylor, C. Snyder and P. Nettesheim, Malignant tumours of liver and lung in rats fed aminopyrine or heptamethyleneimine together with nitrite, Nature, 244 (1973) 176-178. I5 Anonymous, Environmental nitrosamines, turrcel, 11 (1973) 1243-1244. I6 A. Dutton and D. F. Heath, The preparation of [lXI]dimethylamine and [lJCldimethylnitrosaminc. J. Chnrr. Ser., (1956) l892-.1893. 17 L. den Engelse, M. Gcbbink and E. Philippus, [lVI]Formaldehyde: a possible contaminant of (‘,‘C]dimethylnitrosamine, Chem-Biol. /&racr., I I (1975) 133-137. I8 0. H. Lo\hry. N. J. Rosebrough, A. L. Farr and R. J. Randall, Protein measurement with the Folin phenol reagent, J. Biol. Chen~., 193 (1951) 265-275. I9 T. Hattori, H. Aoki. 1. Matsuzaki, B. Maruo and H. Takahashi, Liquid scintillation counting of HQucleic acids, Ad. Chent., 37 (1965) 159-160. 20 R. Montesano and P. N. Magee, Metabolism of dimethylnitrosamine by human liver slices in vitro, N&di e, 228 (1970) 173-l 74.
544 21 F. W. Kri.iger, H. Osswald, G. Walker and E. Schelten, investigations on the organotropy of alkylating activity of dimethylnitrosamine (DMNA) and nitroso-methyl-urea (NMH) in G’iyo, 2. Kreb@rsch. 74 (1970)434-447. 22 R. Montesano and P. N. Magee, Metabolism of nitrosamines by rat and hamster tissue s ices in vitro, Proc. Am. Ass. Cancer Res., 12 (1971) 14. 23 V. J. Feron, P. Emmelot and T. Vossenaar, Lower respiratory tract tumours in Syrian golden hamsters after intratracheal instillations of diethylnitrosamine alone and with ferric oxide, Eur. J. Cancer, 8 (I 972) 445419. 24 P. Pour, F. W. Krikger, A. Cardesa, J. Althoff and U. Mohr, Carcinogenic effect of di-n-propylnitrosamine in Syrian golden hamsters, J. Nad Cllncer Inst., 51 (1973! 1019-1027. 25 J. Althoff, Simultaneous tumors in the respiratory system and urinary bladder of Syrian golden hamsters, Z. Krebs/brsch., 82 (1974) 153-l 58. 26 J. Althoff, R Wilson, A. Cardesa and P. Pour, Comparative studies of neoplastic response to a single dose of nitroso compounds, 3. The e&t of IV-nitrosopiperidine and IV-nitrosomorpholinc in Syrian golden hamsters, Z. Krebsforsch., 81 (1974) 251-259. 27 H. Schreiber, K. Schreiber and D. H. Martin, Experimental tumor induction in a circumscribed region of the hamster trachea: correlation of histology and exfoliative cytology, J. NatI. Cuncer Ins,., 54 (1975) 187-197. 28 R. Ramanathan, D. S. R. Sarma, S. Rajalakshmi and E. Farber, Non-random methylation of rat liver chromatin DNA by dimethylnitrosamine in VI’VO, Proc. Am. Ass. Cancer Res., 16 (1975) abstr. 36. 29 J. A. J. Brouwers and P. Emmelot, Microsomal N-demethylation and the effect of the hepatic carcinogen dimethylnitrosamine on amino acid incorporation into the proteins of rat livers and hepatomas, Exprl. Cell Res., 19 (1960) 467-474.
