462
Chem. Res. Toxicol. 1991,4, 462-466
between DNA and the enantiomers of anti-benzo(a)pyrene7 , s [a,j]anthracene and (h)-7-methyl-trans-3,4-dihydrosy-anti-l,2diol-9,lO-epoxide. Carcinogenesis 4, 211-215. epoxy-l,2,3,etetrahydrodibenz[a,j]ant~acene. Chem. Res. Toricol. 2,341-348. (47) Clarke, D. A., Elion, G.B., Hitchings, G.H., and Stock, C. C. (1958) Structure-activity relationehips among purines related to (46) Subbiah, A,, Islam, S. A., and Neidle, S. (1983) Molecular modeling studies on the non-covalent intercalative interactions 6-mercaptopurine. Cancer Res. 18, 445-456.
Hydroxyethylatlon of Hemoglobin by I-( 2-Chloroethy1)-I-nitrosoureas Eric Bailey,? Peter B. Farmer,*pt Yuk-Sim Tang,t Helen Vangikar,t Andrew Gray,*J Dean Slee,* Robert M. J. Ings,j D. Bruce Campbell,* J. Gordon McVie,g and Ria DubbelmanII MRC Toxicology Unit, Woodmansterne Road, Carshalton, Surrey SM5 4EF, U.K., Servier Research and Development Ltd., Fulmer Hall, Windmill Road, Fulmer, Slough SL3 6HH, U.K., Cancer Research Campaign, 2 Carlton House Terrace, London SWl Y 5AR, U.K.,and Antoni Van Leeuwenhoekhuis, Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands Received January 31, 1991
Following treatment of cancer patients with three l-(2-chloroethyl)-l-nitrosoureas, hemoglobin was isolated and analyzed by GC-MS for N-(2-hydroxyethyl)-N-teerminal valine. This alkylated amino acid was liberated as a (pentafluorophenyl) thiohydantoin from the hemoglobin by a modified Fdman degradation procedure. Following intravenous infusion of fotemustine [diethyl [1-[3-(2-chloroethyl)-3-nitrosoureido]ethylphosphonate] (ca. 90 mg/m2) the levels of (hydroxyethy1)valine in two patients increased steadily, reaching a peak of ca. 300 pmol/g after 6 h. In a further five patients receiving fotemustine (LOO mg/m2) alkylation levels 24 h after treatment ranged from 132 to 1524 pmol/g of globin. Treatment with TCNU [1-(2-~hloroethy1)-3-[2[ (dimethylamino)sulfonyl]ethyl]-1-nitrosourea]or ACNU [ 1-[ (4-amino-2-methylpyrimidin-5yl)methyl]-3-(2-chloroethyl)-3-nitrosourea] resulted in similar increases in (hydroxyethy1)valine in hemoglobin, although the amounts (as with fotemustine) showed considerable interindividual variation. It appears that the measurement of (hydroxyethy1)valine in hemoglobin may be a suitable monitor of exposure to hydroxyethylating agents during (chloroethy1)nitrosourea chemotherapy.
I ntroductlon Measurement of hemoglobin alkylation products (adducts) is now an established method for monitoring human exposure to environmental or occupational alkylating carcinogens (1-4). As hemoglobin adducts are generally stable, and in many cases have the same lifetime in vivo as the hemoprotein, the technique holds the advantage that exposure measurements may still be carried out, even if a period of time has elapsed since the exposure occurred. Also in cases of chronic exposure adducts accumulate in erythrocytes, giving an integral measure of exposure over a prolonged time period. A further advantage of the technique is that hemoglobin is very readily accessible from exposed individuals, allowing large amounts of hemoglobin (50 mg-1 g) to be used for the analysis. Exceptional sensitivity may then be achieved, for example, down to 5-10 pg of adduct/g of human hemoglobin for aromatic amine adducts (5,6). Experiments in animal model systems have *To whom correspondence should be addressed. 'MRC Toxicology Unit. Servier Research and Development Ltd. Present address: Fisons Research and Development, Bakewell Road, Loughborough, Leiceetershire LE1 1 ORH,U.K. Cancer Research Campaign. Netherlands Cancer Institute.
*
demonstrated that in all cases studied the presence of a hemoglobin adduct indicates the presence of nucleic acid adducts, although the quantity ratio varies according to the alkylating agent considered. Dose-response relationships for hemoglobin adduct formation are also linear for most alkylating agents over wide dose ranges ( 1 , 3 ) . Thus the measurement of hemoglobin adducts may be used as an indicator of circulating dose of alkylating species, and as an indirect measure of nucleic acid damage cause by these species. Surprisingly, there is remarkably little knowledge on the covalent bound reaction products formed between alkylating anticancer cytostatics and hemoglobin, and consequently measurement of adduct formation has not until now been applied for the monitoring of such circulating alkylating agents during anticancer treatment. We have therefore now explored the value of this technique, using as an example investigation of therapy with a variety of l-(2-chloroethyl)-l-nitrosoureas.Such nitrosoureas are unstable molecules and decompose spontaneously in vivo and in vitro to yield two reactive species, a chloroethanediazohydroxide ion and an isocyanate (Figure 1). The isocyanate is involved in a variety of protein carbamoylation reactions and enzyme inhibitions [see review by Kohn (7)].Notably, it inhibits DNA replication, RNA strand scission steps, and the rejoining of DNA single-
0893-228x/91/2704-0462$02.50/0@ 1991 American Chemical Society
Hydroxyethylation of Hemoglobin by Nitrosoureas M/F
subject
F M
1
2 3
F F M M F
4 5 6 7 8 9
M F
M M
10 11 12
F
M M
13 14
Chem. Res. Toxicol., Vol. 4, No. 4, 1991 468
Table I. Diagnosis, Drug Treatment, and Response of Patients smoker tumor drug" dose, mg/m2 nrc ovarian carcinoma I1 87.5 I1 95.2 nr prostate adenocarcinoma no melanoma I1 100 I1 100 nr melanoma I1 100 no melanoma Yes melanoma I1 100 I1 100 no melanoma IV 75 no soft tissue Yes synoviasarcoma IV 75 *d sarcoma IV 75 Yes non-small cell lung IV 100 nr colon I11 130 nr adenocolon I11 130 no colon I11 130
responseb PR P CR NC P CR NC P P P D P P PR
I1 = fotemustine, I11 = TCNU, 1'. = ACNU. * CR = complete response, NC = no change, P = progression, D = early death, PR = partial response. e nr = not recorded. Occasional. 0 N=O II
I
RNH-C-N-CH&HzCI
----t
R-N=C=O
+
CICHzCHz-N=N-OH
1 [CICH2CH2It
-R BCNU Fotemustine TCNU ACNU
CICHzCHzCH3
I
(EtO)zNP(OICH-
1
MezNSOzCHzCHz-
III
N;qHzIp
+N CH3 CCNU
0
"
Figure 1. Decomposition of l-(2-chloroethy1)-l-nitrosoureas. [Although established for BCNU (I) and CCNU (V), the degradation scheme for II-IV is less well studied.]
strand breaks, which may affect the repair of alkylation damage (8). However, the isocyanate is not thought to be a major contributor to antitumour activity as nitrosoureas that yield an isocyanate of low carbamoylating activity (e.g., chlorozotocin, which has an internally reactive hydroxyl group which destroys the carbamoylating function) still show activity as antitumor agents. The presence of the alkylating chloroethanediazohydroxide is however essential for antitumor activity. The interaction of this alkylating agent with nucleic acid bases yields reactive chloroethyl adducts which may further react by intramolecular nucleophilic substitution to give internally etheno-bridged bases (9). Intrastrand cross-links may also be generated by such reactions, as demonstrated, for example, by Kohn for 1,3-bis(2-chloroethyl)-l-nitrosourea(BCNU) (I) and related nitrosoureas (10). Reaction with water, to yield a hydroxyethyl derivative, though chemically quite feasible, was shown not to occur with 1-(2-chloroethyl)adenosine, which is the product formed by reacting BCNU with adenosine in vitro (11). However, the l-(Bhydroxyethy1)adenosine was in fact a major product in this reaction, presumably being formed from a hydroxyethyl carbonium ion or another reactive nitrosourea breakdown product. Interaction of alkylating agents with hemoglobin occurs predominantly at cysteine, histidine, carboxylic acid, and N-terminal valine amino acid residues. Interaction of a chloroethanediazohydroxide,or a 2-chloroethyl carbonium ion formed from it, with the N-terminal valine amino group would yield N-(2-chloroethyl)valine although a hydroxy-
ethylated adduct would also be possible, formed by mechanisms outlined above. The chloroethyl adduct would also be susceptible to hydrolysis to N-(2-hydroxyethyl)valiie (HOEtVal), by virtue of intramolecular participation by the nitrogen atom, either in vivo or during the chemical procedures that are used (see below) for investigation of such adducts. A method has been developed by Tornqvist et al. (12) and used by these workers and ourselves (13-18) for the analysis of HOEtVal adducted to hemoglobin as a measure of industrial exposure to ethylene oxide (16). Increased adduct levels were also demonstrated in tobacco smokers (17). This method is based on the use of pentafluorophenyl isothiocyanate (PFPITC) to cleave the adduct from the protein by a modified Edman degradation reaction. The formed (pentafluoropheny1)thiohydantoin derivative (HOEtVal-PFPTH) is then quantitated by selective ion recording (SIR) mass spectrometry. We have now modified this method and used it (a) to investigate if HOEtVal is detectable in hemoglobin extracts from nitrosoureatreated patients, (b) to explore the time course of ita formation and degradation, and (c) to determine if such a measure of adduct formation is of value in the monitoring of treatment.
Materlals and Methods Materials. PFPITC was purchased from Fluka (Glossop, U.K.), and N,O-bis(trimethylsily1)trifluoroacetamide(BSTFA) from Sigma (Poole, U.K.). Both were redistilled before use. Formamide (Analar, BDH, Poole, U.K.) was purified by elution through aluminum oxide (Woelm, activity grade 1). All other solvents were obtained as Analar grade (BDH) and were redietiled. HOEtVal and (hydroxypropy1)valine(HOhVal) were synthesized as described previously (12). Sep-Pak C18 cartridges were obtained from Millipore U.K. (Croxley Green, U.K.). Heptafluorobutyric anhydride was purchased from Aldrich Chemical Co. (Gillingham, U.K.). Treatment of Subjects. (a) Time Course Study. A female volunteer patient (52.4 kg) with ovarian carcinoma (subject 1, Table I) was treated by intravenous infusion over 70 min with a solution of fotemustine [diethyl [1-[3-(2-~hloroethyl)-3nitrosoureido]ethyl]phosphonate,II] in sterile 5% glume solution containing 1.5% ethanol. The total dose administered was 134 mg (2.56mg/kg, 87.5 mg/m*). A venous blood sample (20mL) was taken before dose, and further blood samples (10 mL) were taken at intervals up to 48 h. A second patient (male, 73.9 kg, subject 2), wtih prostate adenocarcinoma, was treated similarly with 176 mg of fotemustine (2.38mg/kg, 95.2 mg/m2). Blood samples were cooled and centrifuged immediately after collection and the cells frozen. The globin was prepared from the erythrocytes, after washing and lysis, by precipitation with 1% HCl in acetone and washed successively with 1% HC1 in
464 Chem. Res. Toxicol., Vol. 4, No. 4, 1991
Bailey et al.
300 7 acetone, acetone, and finally with ether (17). The globin samples were then dried and stored over PzOs in a vacuum desiccator. (b) 24-h Alkylation Study. At the Netherlands Cancer Institute, blood samples were obtained before, and 24 h after, treatment of volunteer patients (subjects 3-14) with each of three nitrosoureas, tauromustine (TCNU) [1-(2-~hloroethyl)-3-[2[ (dimethylamino)sulfonyl]ethyl]-1-nitrosourea, 1111, ACNU [ 1[ (4-amino-2-methylpyrimidin-5-yl)methyl]-3-(2-chloroethyl)-3I / nitrosourea, IV], and fotemustine (11). Details of patients and 100 f dosages are given in Table I. Globin was isolated from the blood samples as described above. Preparation of Calibration Standards. Blood labeled in vitro with ethylene oxide was prepared by incubating a 10-mL 0 I aliquot of control human blood with ethylene oxide (10 pL) ov0 2 4 6 0 10 12 ernight at 37 "C. The red cella were separated by centrifugation, and globin was prepared as described above. Globin alkylated Hours with ethylene-d4 oxide prepared previously (17) was used as inFigure 2. Time course for the production of N-(a-hydroxyternal standard for the GC-MS analyses. ethy1)valine in globin from a patient (subject 2) treated inDetermination of HOEtVal in Hemoglobin. The procedure travenously with fotemustine (11). The infusion period was 70 used was a modification of that previously published (17). The min. The alkylation level at 24 h was 384 pmol of HOEtVal/g following steps were involved: (1)Globin (50 mg) was dissolved of globin. in formamide (2 mL). (2) Internal standard (d4-hydroxyethylated globin, containing 93 pmol of HOEtVal-d,) was added. (3) HOEtVal in the calibration standard was 6.9 nmol/mg of globin. Pyridine (7 pL) and PFPITC (10 pL) were added. (4) The readion (b) Quantitation of i n Vivo Labeled Globins. With each mixture was shaken gently overnight at room temperature and batch of analyses of patient globins, four calibration samples were then heated for 90 min at 45 "C. (5)The reaction mixture was analyzed. These each contained 50 mg of control globin plus diluted with 3 mL of water and applied to SepPak C18 cartridge. d4-hydroxyethylated globin containing 93 pmol of HOEtVal-dl. [Before use the cartridge was washed with ethyl acetate (4 mL), Additionally, they contained a range of amounts of unlabeled methanol (4 mL), and finally with water (4 mL).] After sample hydroxyethylated globin (calibration standard) containing 18.7, application the cartridge was washed with water (4 mL) and 12.5,6.2, and 0 pmol of HOEtVal. The peak height ratio of m/z hexane (1 mL) and the product then eluted with ethyl acetate 440/444 (%, HOEtVal-do/HOEtVal-d,) was linearly related to (4 mL). (6) The ethyl acetate extract was dried in a rotary vacuum the amount of HOEtVal (pmol) in the added calibration globin evaporator (Savant), and the residue was dissolved in toluene (2 (e.g., peak height ratio = 0.536 [HOEtVal] 1.775, r = 0.9993). mL) and washed with 0.1 M Na2C03(1mL) and then with water (1 mL). (7) The toluene extract was dried in a rotary vacuum Results and Dlscussion evaporator, and the residue was redissolved in acetonitrile (30 pL) and BSTFA (20 pL) and heated for 30 min at 60 "C. (8)The Preliminary in vitro and in vivo (rat, mouse) experiments sample was dried and the residue redissolved in acetonitrile (30 (P. B. Farmer, B. Passingham, and B. W. Street, unpubpL) for GC-MS analysis of the trimethylsilyl ether derivative of lished data) indicated the potential of 1-(2-chloroethyl)HOEtVal-PFPTH. 1-nitrosoureas to hydroxyethylate hemoglobin. Thus, for Gas Chromatography-Mass Spectrometry (GC-MS). example, hemoglobin isolated from rats 16 h after a single Analyses of the HOEtVal derivative were carried out by electron ip dose (50 mg/kg) of CCNU [ l-(2-chloroethyl)-3-cycloimpact SIR GC-MS using a VG 70-70 sector mass spectrometer coupled to a Carlo Erba HRGC 5160 Mega gas chromatograph hexyl-1-nitrosourea] (V) contained 742 pmol of HOEtVal/g or a VG Trio 1 quadrupole mass spectrometer coupled to a HP of globin (n = 2). A 20 mg/kg dose yielded 354 pmol of 5890 Series I1 gas chromatograph. Samples were introduced by HOEtVal/g of globin (n = 2). A similar extent of alkylsplit injection into a fused silica capillary column (25 m X 0.32 ation was seen in mice treated with fotemustine, and exmm) coated in the authors' laboratories by the procedure of Blum tensive in vitro alkylation of hemoglobin in rat blood was and Eglinton (19)with the medium polar stationary phase PS-090 also observed with CCNU, ACNU, and TCNU. These [a CH30-terminted polysiloxane prepolymer (20% diphenyl and preliminary results were considered sufficiently encour80% dimethyl substituted)]. The initial oven temperature was aging for human studies to be initiated. 80 OC for 1 min, followed by a temperature rise to 285 "C a t 30 Following intravenous infusion of fotemustine (87.5 and "C/min. SIR was carried out by monitoring the ions m/z 440 95.2 mg/m2) into two patients, the levels of HOEtVal in and 444 a t 40 ms/channel. Method of Quantitation. (a) Analysis of Calibration globin increased steadily, reaching a peak shortly after Standard. Duplicate aliquots of 2 mg of calibration standard infusion ceased (Figure 2). After 6 h the alkylation level (ethylene oxide treated globin-see above) were hydrolyzed in was 310 pmol/g of globin for subject 1and 292 pmol/g for vacuo at 110 "C for 24 h in the presence of 1 mL of 6 N HCl, 10 subject 2. Before treatment these levels were 22 and 27 pL of 1--01, and the internal standard HOPrVal(10 pg). After pmol/g, respectively. Background levels of this adduct are hydrolysis the solution was evaporated to dryness and derivatized found in human control hemoglobin (15, 17), and these by treatment with isobutyl alcohol/3 M HCl (300 pL) at 60 OC latter values fall within the normal range for nonsmokers. for 45 min, and subsequently with heptafluorobutyric anhydride Although HOEtVal levels were only measured up to 48 h (30 pL) in ethyl acetate (100 pL) at 120 "C for 10 min. Samples after dosage, there was no evidence for rapid removal of were introduced, by split injection, into a OV-1 coated fused silica capillary column (25 m X 0.32 mm). The oven temperature the adduct from globin. program was 100 "C for 1 min, followed by a temperature rise In a series of patients receiving fotemustine, five of to 140 "Cat 30 "C/min and then to 180 "C at 5 "C/min and W y which are represented in Figure 3a and Table I, alkylation to 285 OC at 40 "C/min. Quantitation of the derivatized HOEtVal levels 24 h after treatment ranged from 132 pmol/g of in the hemoglobin hydrolysate was determined by reference to globin (subject 4) to 1524 pmol/g of globin (subject 7). calibration lines established from analysis, using flame ionization Both of these patients were female and were treated for detection, of 2-mg aliquots of hydrolyzed globin mixed with 10 melanoma with a dose of 100 mg/m2 fotemustine. Thus pg of HOPrVal and 0-20 pg of HOEtVal. (HOPrVal is diastemarked variations in circulating "hydroxyethylating" reomeric, and the derivatized compound gives two peaks which species exist during similar patient treatments. Alkylation were resolved by GC.) Such calibration lines were linear using either of the HOPrVal isomers as internal standard and typically levels before treatment were in the range 41-70 pmol of had regression coefficients greater than 0.99. The amount of HOEtVal/g of globin, which is within the normal range
;
I
I
+
Hydroxyethylation of Hemoglobin by Nitrosoureas
a 1524
1
.-c
8
-cn-
300
. v,
z
i
U O h r 24hr
200
W
P n 0
SUBJECT
NUMBER
b 600
2
-0cn
500
cn 400 m
-0
300 200
EtVal-PFPTH, 133930-76-4; Val, 72-184; fotemwtine, 9211827-9; N-(2-hydroxyethyl)valine,101769-73-7.
E p
100
References
n
a
9 10 11 SUBJECT NUMBER
C Me2NS02CH2CH2NH
-
Yo
- N - CH2CHzCl
400 TCNU
.-nC 0 0,
300
0,
1 m
zW
the use of a Sep-Pak C18 column chromatograph to isolate the HOEtVal (pentafluoropheny1)thiohydantoinfrom the reaction mixture. We have found that incorporation of thia isolation step results in essentially contamination-free SIR GC-MS traces. Whereas in the past lengthy GC temperature programs were required to ensure resolution of the trimethylsilyl ether thiohydantoin derivative from contaminants (and on occasion even tandem mass spectrometry was required to confirm the nature of the product), it is now possible to achieve a complete analysis of derivatized extracted samples in 15 min, by using a benchtop quadrupole mass spectrometer. In summary, it appears that the measurement of HOEtVal in globin may be used as a monitor of exposure to hydroxyethylating species during nitrosourea therapy. With the data currently available there does not appear to be a correlation between HOEtVal adduct levels and response to treatment. However, the greater value of this monitoring technique might be in the use of such adduct measurements in the optimization of the dose of nitrosourea that may be given to an individual; i.e., it may be possible to increase the drug dose in patients showing low adduct levels without undesirable increases in toxicity.
Acknowledgment. We thank B. Street for assistance with sample preparation and J. Lamb for mass spectrometry advice. We acknowledge the Commission of the European Communities for financial support. Registry No. TCNU, 85977-49-7; ACNU, 55661-38-6; HO-
1.
2
Chem. Res. Toxicol., Vol. 4, No. 4, 1991 465
200
100
n 0
12
13
14
SUBJECT NUMBER
Figure 3. N-(2-Hydroxyethyl)valineconcentrations in globin of patients before and 24 h after treatment with (a) fotemustine (II), (b) ACNU (III), or (c) TCNU (IV). See Table I for dosages.
for nonsmokers [average 50 pmol/g of globin, range 22-106 pmol/g of globin (17)]. It is known that subjects 3,5, and 7 were not current smokers, although subject 6 did smoke. (It is surprising therefore that the latter's "backgroundn level was only 70 pmol of HOEtVal/g of globin.) Treatment with ACNU (Figure 3b) or TCNU (Figure 3c) resulted in a similar increase in HOEtVal adduct levels in globin. Subjects 9 and 11 had elevated background levels (220 and 200 pmol of HOEtVal/g of globin, respectively), which could be accounted for on the basis of the patients' known smoking habits. The improvements made to the method of Tornqvist et al. (12) for quantifying HOEtVal in hemoglobin involved
(1) Ehrenberg, L., and Osterman-Golkar, S. (1980) Alkylation of
macromolecules for detecting mutagenic agents. Teratog., Carcinog., Mutagen. l, 105-127. (2) Farmer, P. B., and Bailey, E. (1989) Protein-carcinogen adducts in human dosimetry. Arch. Toxicol., Suppl. 13,83-90. (3) Neumann, H.-G. (1988) Biomonitoring of aromatic amines and alkylating agents by measuring hemoglobin adducts. Znt. Arch. Occup. Enuiron. Health 60, 151-155. (4) Farmer, P. B., Neumann, H.-G., and Henschler, D. (1987) Estimation of exposure of man to substances reacting covalently with macromolecules Arch. Toxicol. 60, 187-191. (5) Stillwell, W. G., Bryant, M. S., and Wishnok, J. S. (1987) GC/ MS analysis of biologically important aromatic amines. Application to human dosimetry. Biomed. Enuiron. MQSSSpec. 14, 221-227. (6) Bailey, E., Brooks, A. G., Bird, I., Farmer, P. B.,and Street, B. (1990)Monitoring exposure to 4,4'-methylene dianiline by the gas chromatography-mass spectrometry determination of adducte to hemoglobin. Anal. Biochem. 190, 175-181. (7) Kohn, K. W. (1981) Mechanistic approaches to new nitrosourea development. Recent Results Cancer Res. 76,141-152. (8) Heal, J. M., Fox, P. A., and Schein, P. S. (1979) Effect of carbamoylation on the repair of nitroaourea-induced DNA alkylation damage in L1210 cells. Cancer Res. 39,82-89. (9) Kramer, B. S., Fenselau, C. C., and Ludlum, D. B. (1974) Reaction of BCNU (1,3-bis(2-chloroethyl)-l-nitrosourea) with polycytidylic acid. Substitution of the cytosine ring. Biochem. Biophys. Res. Commun. 56,783-788. (10) Kohn, K. W. (1977) Interetrand cross-linking of DNA by 1,3bis(2-chloroethyl)-l-nitrosourea and other l-(2-haloethyl)-lnitrosoureas. Cancer Res. 37, 1450-1454. (11) Tong, W. P., and Ludlum, D. B. (1979) Mechanism of action of the nitrosoureas-111. Reaction of bis-chloroethyl nitrosourea and bis-fluoroethyl nitrosourea with adenosine. Biochem. Pharmacol. 28, 1175-1179. (12) Tornqvist, M., Mowrer, J., Jensen, S., and Ehrenberg, L. (1986) Monitoring of environmental cancer initiators through hemoglobin adducts by a modified Edman Degradation method. A n d . Biochem. 154, 255-266. (13) Tornqvist, M., Osterman-Golkar, S., Kautiainen, A., Nblund, M., Calleman, C. J., and Ehrenberg, L. (1988) Methylations in human hemoglobin. Mutat. Res. 204, 521-529.
Chem. Res. Toxicol. 1991,4,466-473
466
(14) Tornqvist, M., Kautiainen, A., Gatz, R. N., and Ehrenberg, L. (1988) Hemoglobin adducts in animals exposed to gasoline and diesel exhausts. I. Alkenes. J. Appl. Tonicol. 8, 159-170. (15) Tomqvist, M., Gustafsson, B., Kautiainen, B., Harms-Ringdahl, M., Granath, F., and Ehrenberg, L. (1989)Unsaturated lipids and intestinal bacteria as sources of endogenous production of ethane and ethylene oxide. Carcinogenesis 10,39-41. (16) Farmer, P. B., Bailey, E., Gorf, S. M., Tornqvist, M., Osterman-Golkar, S., Kautiainen, A,, and Lewis-Enright, D. P. (1986) Monitoring human exposure to ethylene oxide by the determination of haemoglobin adducts using gas chromatography-mass spectrometry. Carcinogenesis 7,637-640. (17) Bailey, E.,Brooks, A. G. F., Dollery, C. T., Farmer, P. B.,
Passingham, B. J., Sleightholm, M. A., and Yates, D. W.(1988) Hydroxyethylvaline adduct formation in haemoglobin as e biological monitor of cigarette smoke intake. Arch. Toxicol. 62, 247-253. (18) Tomqvist, M., Osterman-Golkar, S., Kautiainen, A., Jensen, S., Farmer, P. B., and Ehrenberg, L. (1986b) Tissue d m s of ethylene oxide in cigarette smokers determined from adduct levels in hemoglobin. Carcinogenesis 7, 1519-1521. (19) Blum, W.,and Eglinton, G. (1989) Preparation of high temperature stable glass capillary columns coated with PS-090 (20% diphenyl-substituted CHSO-terminated polysiloxane) a selective stationary phase for the direct analysis of metal-porphyrin complexes. High Res. Chrom. Chrom. Commun. 12, 290-293.
Urinary Excretion and DNA Binding of Coal Tar Components in B6C3F1 Mice following Ingestion Eric H. Weyand,* Yun Wu, and Shruti Pate1 Rutgers, The State University of New Jersey, College of Pharmacy, P.O. Box 789, Piscataway, New Jersey 08855-0789
Barbara B. Taylor and David M. Mauro META Environmental, Inc., 49 Clarendon Street, Watertown, Massachusetts 021 72 Received March 4, 1991
Urinary excretion of polycyclic aromatic hydrocarbon (PAH) metabolites and DNA binding of coal tar components in male mice were investigated following the ingestion of a coal tar adulterated diet. Male B6C3F1 mice were able to tolerate an F0927 basal gel diet which contained from 0.1 to 1% coal tar (tar weight/dry food weight) for 15 days. Mice maintained on a 0.1 and 0.2% coal tar diet had body weight gains similar to those of control animals. However, mice maintained on the 0.5 and 1.0% diet had body weight gains considerably lower than control values. Chemical-DNA adduct formation was detected and quantified in lung and forestomach tissue of animals on 0.1, 0.2, 0.5, and 1% coal tar containing diets. A dose-related effect was observed in lung DNA adduct formation while no dose effect was observed in forestomach tissue. In addition, overall adduct levels in lung tissue were considerably higher than forestomach levels for animals on the 0.5 or 1% diet. In contrast, DNA adduct levels were highest in the forestomach of animals on diets lower in coal tar content (0.1 or 0.2%). Chemical-DNA adducts of coal tar components were also evaluated for four other coal tar samples which varied in chemical composition. Mice were maintained on diets containing 0.25% of each coal tar for 15 days. Chemical-DNA adducts were detected in lung, liver, and spleen for all animals receiving these coal tar diets. DNA adduct patterns were similar while quantitative differences were observed between coal tar samples and tissue sites. Highest adduct levels were detected in lung DNA. Benzo[a]pyrene content in coal tar samples could not account for the DNA adduct levels observed with coal tar ingestion. The urinary excretion of select PAH metabolites following coal tar ingestion was evaluated by using urine collected on days 1 and 14 of diet administration. The levels of l-hydroxypyrene in urine, the major PAH metabolite detected, correlated with the pyrene content of these coal tars. These data demonstrate that coal tar components are readily bioavailable following ingestion. Introduction . _
Polycyclic aromatic hydrocarbons (PAHs)' are a large class of ubiquitous environmental pollutants that have been implicated in the etiology of several human cancers ('-3)* Human exposure to occurs through inhalation, ingestion, or topical absorption and is most often occupationally related. Since several PAHs have been determined to be potent chemical carcinogens in laboratory animals, monitoring human exposure to PAHs is essential
* To whom correspondence should be addressed.
in establishing reliable risk assessments. Biological monitoring methds such as mutagenicity of urine extracts and in urine have the detection of PAH been used to evaluate exposure to PAH using both laboratory animals and humans (4-13). Rscently, studies have evaluated the covalent binding of aromatic chemic& to of exposure to complex DNA as a mixtures, &hoket and co-workers have demonstraM that Abbreviations: B[o]P, benzo[o]pyrene; TLC,thin-layer chrometography; HPLC,high-pressure liquid chromatography; PAH,polycyclic aromatic hydrocarbon; PEI-cellulose, poly(ethylenimine)-cellulose.
0S93-228~/91/2104-0466~02.50~0 0 1991 American Chemical Society
