Chem. Res. Toxicol. 1994, 7, 344-351
344
Activation of the Liver Carcinogen 2-Nitropropane by Aryl Sulfotransferase Rama S. Sodum,* Ock Soon Sohn, Guo Nie, and Emerich S. Fiala' Division of Biochemical Pharmacology, American Health Foundation, 1 Dana Road, Valhalla, New York 10595 Received December 8,1993"
8-Aminoguanine had previously been identified as one of the nucleic acid base modifications produced in livers of rats by treatment with the hepatocarcinogen 2-nitropropane (2-NP), and a hypothetical mechanism of activation of 2-NP to hydroxylamine-0-sulfonate or acetate that would lead to NH2+, an aminating species, was proposed [Sodum e t al. (1993) Chem. Res. Toxicol. 6,269-2761. We now present in vivo and in vitro experimental evidence for the activation of 2-NP to an aminating species by rat liver aryl sulfotransferase. Pretreatment of rats with the aryl sulfotransferase inhibitors pentachlorophenol or 2,6-dichloro-4-nitrophenolsignificantly decreased the levels of liver nucleic acid modifications produced by 2-NP treatment. Furthermore, partially purified rat liver aryl sulfotransferase was shown t o activate 2-NP and 2-NP nitronate in vitro at neutral pH and 37 "C, to a reactive species that aminated guanosine at the C8 position. This activation was dependent on the presence of the enzyme, its specific cofactor adenosine 3'-phosphate 5'-phosphosulfate, and mercaptoethanol. As in the case of the in vivo studies, pentachlorophenol and 2,6-dichloro-4-nitrophenolinhibited the in vitro formation of 8aminoguanosine and 8-oxoguanosine. The corresponding primary nitroalkane, 1-nitropropane, which is not mutagenic and does not appear to be carcinogenic, was not a substrate for aryl sulfotransferase in the in vitro amination of guanosine.
Introduction 2-Nitropropane(2-NP),' an important industrial chemical (I) and a constituent of cigarette smoke (21, is a hepatocarcinogenin male Sprague-Dawley rats ( 3 , 4 )and induces characteristic base modifications in rat liver DNA and RNA (5-7). In contrast, the primary nitroalkane 1-nitropropane (1-NP),which is not mutagenic (8,9) and is apparently not carcinogenic (3),does not produce these modifications to a significant extent (6). One of the modifications produced by 2-NP in rat liver nucleic acids has recently been identified as the C8amination of guanine; another is an increase in the level of 8-oxoguanine (5,IO). Other nucleic acid modifications, designated as DX1 in DNA and RX1 in RNA (5, IO), remain unidentified. Several other secondary nitroalkanes, namely, 2-nitrobutane, 3-nitropentane, 2-nitrooctane,and 2-nitroheptane, but not the corresponding primary nitroalkanes, also induce a pattern of nucleic acid modifications in rat liver that is qualitatively very similar to that produced by 2-NP (6). In order to explain the 2-NP-induced formation of 8-aminoguaninein rat liver nucleic acids, we have proposed, in an earlier report (lo), a mechanism whereby 2-NP is metabolically converted to an aminating species. As illustrated in Scheme 1, the first step in the proposed mechanism is the dissociation of the acidic proton on carbon 2 of 2-NP followed by delocalization of the negative charge to give 2-NP nitronate (11). This can occur under
* Address correspondence to either of these authors at the American Health Foundation, 1Dana Rd., Valhalla, NY 10595. Tel: 914-789-7136; F s : 914-592-6317. *Abstract published in Aduance ACS Abstracts, April 1, 1994. Abbreviations: 2-NP, 2-nitropropane; 1-NP, 1-nitropropane; PCP, pentachlorophenol; DCNP, 2,6-dichloro-4-nitrophenol; PG, propylene glycol; EC, electrochemical detection; N-OH-2-AAF,N-hydroxy-2-(acetylamino)fluorene; PAPS, adenosine 3'-phosphate 5'-phosphosulfate.
Scheme 1. Proposed 2-NP Activation Mechanism
\
/
deoxvaenation and sulfation
I V
.ic = o +1 /
CH3
-
(30-SO3H
CH
H-N
H '
+ N H ~
VI
+
HSO~
VI1 nucleiclacids
~
8-Aminoguanine
physiologicalconditions since the pKa of 2-NP is 7.63 (11). Central to the proposed schemeis the enzymatic conversion of the nitronate anion to an 0-sulfonate or acetate, either before or after its deoxygenation,in a sulfotransferase- or acyltransferase-catalyzed reaction. Removal of one of the oxygens of the nitronate could be mediated enzymatically or nonenzymatically. If sulfotransferase is involved in the activation of 2-NP, then the resulting conjugate IV could react with water and dissociate t o acetone and hydroxylamine-0-sulfonic acid. The latter is a known aminating reagent which can give the highly reactive unsubstituted nitrenium ion, NH2+, capable of aminating biological molecules (10, 12).
O S ~ ~ - ~ ~ ~ ~ I ~ ~ I ~ I O I -0O1994 ~ ~ American ~ ~ O ~ . Chemical ~ O I O Society
Sulfotransferase Activation of 2-Nitropropane
In this article we report that the aryl sulfotransferase inhibitors pentachlorophenol (PCP) and 2,6-dichloro-4nitrophenol (DCNP) (13, 14) strongly suppress base modifications induced in rat liver RNA and DNA by administration of 2-NP. These results indicate that a sulfation step catalyzed by aryl sulfotransferase is indeed involved in the activation of 2-NP to a species capable of modifying nucleic acids in vivo. We further corroborate this conclusion by demonstrating that partially purified rat liver aryl sulfotransferase, in the presence of 2-NP nitronate, adenosine 3'-phosphate 5'-phosphosulfate (PAPS),and mercaptoethanol, catalyzes the formation of 8-aminoguanosine and 8-oxoguanosine from guanosine in vitro.
Experimental Section Materials. Beckman Ultrasphere columns,Ultrasphere ODS guard columns, and Optisil 10 SCX columns were purchased from Phenomenex (Torrence, CA). Emulphor was purchased from GAF Chemical Corp. (Wayne, NJ). 8-Aminoguanosine, PAPS, and propylene glycol (PG) were purchased from Sigma Chemical Co. (St. Louis,MO). 8-Aminoguanine,8-oxoguanosine, 8-oxodeoxyguanosine, and 8-aminodeoxyguanosine were synthesized as described previously(10). Affi-GelBlue was obtained from Bio-Rad Laboratories (Richmond, CA). N-Hydroxy-2(acety1amino)fluorene(N-OH-2-AAF) was obtained from the Midwest Research Institute (Kansas City, MO). 2-NP, 1-NP, and PCP were purchased from Aldrich Chemical Co. (Milwaukee, WI). DCNP was purchased from ICN Biochemicals Inc. (Costa Mesa, CA). The nitroalkanes were purified by distillation, and purity was verified by NMR spectral data and capillary GC. Male F344 rats, body weights 170-190 g, were obtained from Charles River Laboratories (Kingston, NY). HPLC Systems. They were as follows: system 1, reversephase HPLC using two 0.46- X 25-cm Ultrasphere ODS columns in series, with a 0.46- X 4.6-cm Ultrasphere guard column eluted with 12.5 mM citric acid and 25 mM sodium acetate buffer, 9.5 % methanol (pH 5.1); system 2, reverse-phase HPLC on a 0.46- X 25-cm Ultrasphere ODS column protected by a 0.46- X 4.6-cm Ultrasphere ODS guard column eluted with 12.5 mM citric acid and 25 mM sodium acetate, 5% methanol (pH 5.1); system 3, cation-exchangeHPLC using two 0.46- X 25-cm Optisill0 SCX columns in series with a guard column of a 3-cm Aquapore CX300 cation-exchangecartridge (BrownleeLabs, Inc., Santa Clara, CA), eluted with 12.5 mM sodium citrate and 25 mM sodium acetate (pH 5.1); system 4, same as system 2 except methanol concentration changed to 0.5%. In all cases effluents were monitored with UV detectors at 253 nm and electrochemical (EC) detectors at a potential of +600 mV vs the Ag/AgCl/3 M NaClreference electrode. The flow rates were 1mL/min. Details of the instrumentation were given earlier (10). Pretreatments of Animals with Sulfotransferase Inhibitors. F344 rata were pretreated ip with the aryl sulfotransferase inhibitors DCNP or PCP at doses of 26 or 39 pmol/kg, respectively, or with the alcohol sulfotransferase inhibitor PG at a dose of 1 mL/kg. PCP and DCNP were administered as solutions in corn oil (1mL of corn oil/kg body wt). 2-NP, at a dose of 1.12 mmol/ kg, was administered ip in a vehicle of 10% aqueous Emulphor 2 h after pretreatment with inhibitors. Control animals were pretreated with corn oil alone followed by ip treatment with 1.12 mmol/kg 2-NP 2 h later. Additional controls consisted of rats pretreated with either corn oil, PCP, DCNP, or PG and treated 2 h later with Emulphor-HzO. All animals were killed 18h after the 2-NP (or Emulphor-HzO) treatments. Livers were excised, frozen in liquid nitrogen, and stored at -80 "C. DNA and RNA were isolated, hydrolyzed, and analyzed by HPLC-EC as previously described (10). Isolation and Partial Purification of Rat Liver Aryl Sulfotransferase. For the partial purification of aryl sulfotransferase from livers of male F344 rats, the procedure of
Chem. Res. Toxicol., Vol. 7,No. 3, 1994 345 Sekura et al. (15) was followed up to, and including, the stages of ammonium sulfate precipitation and overnight dialysis. The dialyzed enzyme preparation was stored at -80 "C in 2.5 mM sodium phosphate buffer containing 250 mM sucrose and 3 mM mercaptoethanol (pH 6.9). Protein concentration was 2.5-3.0 mg/mL as determined by the Lowry method (16). Aryl sulfotransferase was assayed with 2-naphthol as substrate at pH 7.4 as described by Sekura et al. (15). For assays of activity using 2-naphthol, the reaction was allowed to proceed for 10 min. Sulfation of the substrate was determined by measuring the absorbance (651 nm) of the ion pair formed by the product with methylene blue. One unit of activity is defined as the amount of enzyme catalyzing the formation of 1 nmol of 2-naphthol-Osulfonate/min. Enzyme preparations from the livers of male F344 rats typically gave about 17 units/mg of protein. Enzyme activity with 2-naphthol as substrate did not change even after 3 months of storage at -80 "C. Activation of 2-NP and 2-NP Nitronate in Vitro by Rat Liver Aryl Sulfotransferase. A lOmM stock solution of 2-NP was prepared by dissolving 0.89 mg of the nitroalkane in 1 mL of 0.1 M sodium phosphate buffer (pH 7.0). A stock solution of 100mM 2-NP nitronate solutionwas freshlyprepared by addition of 0.11 mL of 1N NaOH to 0.89 mL of 0.1 M sodium phosphate buffer solution (pH 6.9) containing 8.9 mg of 2-NP. A 10-pL aliquot of this nitronate stock solution or a 100-pL portion of the parent nitroalkane stock solution was added to the in vitro sulfotransferase reaction mixtures to give a 2 mM final concentration of substrate. Reaction mixtures of 0.5-mL total volume contained a mixed buffer of 0.1 M sodium phosphate and 0.1 M Tris (pH 7.0) that was saturated at 25 "C with guanosine (final concentration of guanosine, 2-3 mM), 2 mM 2-NP or 2-NP nitronate, 8mM mercaptoethanol(40pL of a 0.1 M stock solution in water), 1.0 mM PAPS, and 5 units of enzyme. The enzyme and PAPS, in that order, were added last. The reaction mixtures were then shaken gently at 37 "C for 90 min. The same procedure was used for the in vitro reactions using 1-NP or 1-NP nitronate as substrate. The stock solutions of 10or lOOmM 1-NPnitronate were prepared as described above for 2-NP and its nitronate. Followingincubation, the reaction mixture was passed through a Sep-Pak cartridge (Waters ChromatographyProducts, Milford, MA), to remove mercaptoethanol and unreacted nitronate, both of which are electrochemicallyactive and would interfere in the HPLC-EC analyses. 8-Aminoguanosine and 8-oxoguanosinewere eluted from the cartridge with 60 % aqueous methanol and were determined using HPLC-EC systems 2 and 3. When 2-NP nitronate alone was incubated at 37 "C for 90 min in theTris-phosphate buffer (pH 7.0),virtually allof the substrate remained in the nitronate form during the incubation as judged from examining the UV spectrum of the nitronate before and after incubation (the nitronate of 2-NP, but not the parent nitro compound, has an absorption maximum at approximately 235 nm; 17). To confirm the requirement for mercaptoethanol in the reaction, control incubations were carried out using enzyme preparations from which mercaptoethanol was removed by passage through a Sephadex G-10 column eluted with 2.5 mM sodium phosphate and 250 mM sucrose (pH 6.9). Other control incubations were carried out in the absence of either PAPS or the enzyme. Incubationswith Sulfotransferase Inhibitors and N-OH2-AAF. Stock solutions of 1.0 and 0.1 mM PCP or DCNP were prepared in Me2SO/H20 (1:l).A 5-pL aliquot of the stock solution of PCP or DCNP or 5 pL of Me2SO/H20 (controls) was added to the reaction mixture at pH 7.3, and the reaction was initiated by addition of enzyme and PAPS. The incubations were carried out as described above. To determine the effects of N-OH-2AAF, 25 pL of either Me2SO (controls) or N-OH-2-AAF in 25 pL of MezSO (final concentration in the reaction mixture, 0.5 mM) was added to the incubations.
346 Chem. Res. Toxicol., Vol. 7, No. 3, 1994
Sodum et al.
1
11
0 4 nA
DX1
OT
NH, - d C w
8-
0 12 AU Q 253 n m
/
I/
8-Oxo-dGuo
T
0.2 n d OT
0 2 4 AU Q 253 n m
-1
".
I IV 1
0
1
1
1
1
1
1
1
1
1
1
1
1
10 20 ELUTION VOLUME, m L
1
1
1
,
l
30
Figure 1. Effects of pretreatment with PG, PCP, or DCNP on 2-NP-inducednucleoside modificationsin rat liver DNA. HPLCEC analyses, using system 1, of enzymatically hydrolyzed liver DNA obtained from 1,rata pretreated with corn oil and treated with Emulphor-HzO vehicle 2 h later;2, rata pretreated with PG and treated with 2-NP 2 h later, 3, rats pretreated with PCP and treated with 2-NP 2 h later; 4,rata pretreated with DCNP and treated with2-NP2 h later. The vertical dimensions of EC traces 1-4 were normalized to the deoxyguanosine content of the hydrolysates; thus, the UV profile applies to all four EC traces.
Results Effects of Sulfotransferase Inhibitors on 2-NPInduced RNA and DNA Modification in Vivo. PCP and DCNP are powerful and specific inhibitors of rat liver aryl sulfotransferase in vivo and in vitro (13,141, and PG has been reported to function as a competitive inhibitor of alcohol sulfotransferase in vivo (18). As shown in Figure 1, pretreatment of rats with 39 pmol/kg PCP significantly decreased the induction, by 2-NP, of the electrochemically active deoxynucleosidesDX1 and 8-aminodeoxyguanosine in liver DNA. With respect to corn oil-pretreated controls, the electrochemically active peak corresponding to 8oxodeoxyguanosine also increased to a lesser extent as a result of PCP pretreatment and 2-NP treatment, but the effect was not as great as noted with 8-aminodeoxyguanosine. Pretreatment with 26 pmol/kg DCNP also suppressed the induction of all of the electrochemically active modified deoxynucleosides by 2-NP, but the effect was noticeably less than that caused by PCP. In contrast, pretreatment with the alcohol sulfotransferase inhibitor PG at a dose of 1 mL/kg (approximately 13.6 mmol/kg) had no significant effect on the levels of 2-NP-induced nucleic acid modifications; these HPLC-EC profiles were essentially identical to those obtained using liver DNA from corn oil-pretreated, 2-NP-treated rats (not shown). HPLC-EC analysesof enzymatic hydrolysates of liver DNA from control animals pretreated with PCP, DCNP, or PG, treated 2 h later with Emulphor-HaO, and sacrificed 18 h later showed no detectable differences from analogous preparations obtained from the corn oil-pretreated, Emulphor-H2O-treated control rata (trace 1 in Figure 1). As illustrated by typical HPLC-EC profiles in Figure 2, pretreatments of rats with PCP or DCNP also inhibited 2-NP-induced electrochemically active base modifications
V
, A U \h Tu I
0
-1
v
10 20 ELUTION VOLUME, mL
Figure 2. Effecta of pretreatment with PG, PCP, or DCNP on 2-NP-inducednucleosidemodifications in rat liver RNA. HPLCEC analyses, using system 2, of enzymatically hydrolyzed liver RNA from 1, rata pretreated with corn oil and treated 2 h later with Emulphor-HzO; 2, rata pretreated with PG and treated 2 h later with 2-NP; 3, rata pretreated with PCP and treated 2 h later with 2-NP; 4, rata pretreated with DCNP and treated 2 h later with 2-NP. The vertical dimensions of EC traces 1-4 were normalized to guanosine content; the UV profile applies to all four EC traces. in liver RNA. Both inhibitors reduced the formation of 8-aminoguanosine, 8-oxoguanosine, and the unknown RX1. As in the case of DNA, PCP was stronger than DCNP in inhibiting the formation of these modified nucleosides; PG pretreatment had no effect. Data summarizing the results of HPLC-EC assays of 2-NP-induced DNA and RNA nucleoside modifications in livers of rata pretreated with sulfotransferase inhibitors PCP, DCNP, and PG are given in Table 1. In the case of DNA, 2-NP treatment induced a 40% increase (p < 0.05) in the 8-oxodeoxyguanosine content, measured 18 h after administration of the carcinogen. Pretreatment of the animals with PCP or DCNP limited this increase to about 10-11 % . Pretreatment with PCP inhibited the 2-NPinduced formation of DX1 and 8-aminodeoxyguanosine by 85% (p < 0.01) and 92 5% (p < 0.011, respectively, while pretreatment with DCNP decreased the formation of DX1 and 8-aminodeoxyguanosine by 49 % @ < 0.05) and 48% 03 < 0.05),respectively. Pretreatment with the alcohol sulfotransferase inhibitor PG had no apparent effect on the 2-NP-induced liver DNA modifications. While the basal (control)levels of 8-oxoguanineper mole of guanine in rat liver RNA are similar to those of DNA, as determined by HPLC-EC, the administration of 2-NP induces a much greater increase of this oxidized base in RNA than in DNA (5). As shown in Table 1, 2-NP produced an 8.7-fold increase (p< 0.01) in 8-oxoguanosine in rat liver RNA, measured 18 h after administration of the carcinogen. Pretreatment with PCP or DCNP significantly inhibited this increase by 74% (p < 0.01) or 45% 03 < 0.01), respectively. Pretreatment with PCP inhibited the 2-NP-induced formation of RX1 and 8-
Chem. Res. Toxicol., Vol. 7,No. 3, 1994 347
Sulfotransferase Activation of 2-Nitropropane
Table 1. Effects of Pretreatments with Sulfotransferase Inhibitors on 2-NP-Induced Nucleoside Modifications DNA and RNA. DNA RNA 8-0~0-dGuol DXlbI 8-NHz-dGuOI &OXO-GUO/ RXlbl pretreatment treatment 105 dGuo dGuo 105 dGuo 105 Guo Guo corn oil Emulphor 1.23 f 0.22 NDe ND 1.02 f 0.19 ND 0.53 f 0.09 2.95 f 0.54 8.87 f 0.59 0.78 f 0.16 1.72 f 0.12d 2-NP corn oil 0.05 f 0.03 0.25 f 0.23 2.30 f 0.298 0.12 f 0.058 1.35 f 0.ld 2-NP PCP 0.19 f 0.078 1.54 i 0.w 4.87 f 0.468 0.40 f 0.07f 1.37 f 0.13f 2-NP DCNP 0.57 f 0.09 2.96 f 0.55 8.98 f 0.85 0.70 f 0.16 1.73 f 0.15 PG 2-NP a
in Rat Liver
8-NHrGu01 108 Guo
ND 0.86 f 0.08 0.11 f 0.02r 0.33 f 0.W 0.90 f 0.08
Mean f SD of 3-4 separate determinations, each of which used 1animal. b Area of EC pewarea of UV peak. e ND = not detectable. p
< 0.05 relative to corn oil-pretreatedEmulphor-treated control group. e p < 0.01 relative to corn oil-pretreatedEmulphor-treatedcontrol group. f p < 0.05 relative to corn oil-pretreated 2-NP-treated group. 8 p < 0.01 relative to corn oil-pretreated 2-NP-treated group. aminoguanosine in liver RNA by 91% (p < 0.01) and 87 % (p < 0.01), respectively, while pretreatment with DCNP inhibited the formation of RX1 and 8-aminoguanosine by 64% (p < 0.01) and 62% @ < 0.01),respectively. As in the case of DNA, pretreatment with PG had no effect on the 2-NP-induced nucleoside modifications in liver RNA. The levels of 8-oxodeoxyguanosine and 8-oxoguanosine in liver DNA and RNA, respectively, of control animals pretreated with PCP, DCNP, or PG and treated with Emulphor-HzO, the vehicle for 2-NP (results not shown), were not significantly different from those of the control animals pretreated with corn oil and treated with EmulphorHzO. Aryl SulfotransferaseActivation of 2-NP and 2-NP Nitronate to an Aminating Species in Vitro. Conclusive evidence for the ability of aryl sulfotransferase to activate 2-NP to a species capable of modifying nucleic acid bases was obtained from in vitro studies with the partially purified rat liver enzyme. Using incubation mixtures containing PAPS, mercaptoethanol, enzyme, guanosine, and 2-NP or the nitronate of 2-NP as detailed in the Experimental Section, the formation of 8aminoguanosine and an increase in the level of 8-oxoguanosine could be demonstrated (commercial preparations of guanosine normally contain 8-oxoguanosine as a minor contaminant). As illustrated in the reverse-phase HPLC-EC profiles in Figure 3A, the major product formed during the 90 min of incubation coelutes with standard 8-aminoguanosine a t approximately 24 mL. In addition, there occurs an increase in the amount of 8-oxoguanosine, eluting at approximately 21 mL. Acid hydrolysis of the material eluting at 24 mL (collected with the EC detector turned off) in methanol containing 1 N HCI (95 "C, 15 min) followedby neutralization and chromatography using HPLC system 4 yielded an electrochemically active species which coeluted with standard 8-aminoguanine (Figure 3B). Further evidence for the identity of the products of the sulfotransferase-catalyzedreaction as 8-aminoguanosine and 8-oxoguanosine is presented in Figure 4, which shows the analysis of the reaction mixture by cation-exchange HPLC-EC. In this system 8-oxoguanosine elutes early at 9-11 mL and 8-aminoguanosine, which is very basic (pK, 4.85),2elutes later than all other nucleosides (IO)at 24-26 mL. As in the case of reverse-phase HPLC, the major electrochemically active product from the in vitro incubation mixtures coelutes with 8-aminoguanosine standard in the cation-exchange HPLC system. Neither C8 amination of guanosine nor an increase in 8-oxoguanosinewas detectable when 1-NP or 1-NP nitronate was used as substrate in the aryl sulfotransferase incubation mixtures *Unpublishedreeults.
A
B
II1! ?-
f
1.5 n A or 0 30 Ab' @ 253 n m
11
I
8- NHz -G
\
f
2 0 nA
L
>' I
10
,
I
l
l
I
I
20
I
I
I
I
30
1
r l , , , , , , , , l
0
10
20
ELUTION VOLUME, mL Figure 3. (A) HPLC-EC analyses (system 2) of products of the in vitro rat liver aryl sulfotransferase-mediated amination of guanosine. 1,UV detector response (applies to all traces); 2-5, EC detector response: 2, incubation with 2-NP nitronate but in the absence of PAPS; 3, complete incubation mixture with 2-NP substituted for 2-NP nitronate; 4, complete incubation mixture using 2-NP nitronate; 5, cochromatography of producta of complete incubation 4 with 8-aminoguanosine standard. (B) Material eluting a t 25 mL from complete incubations as in HPLCEC trace 4 in (A) was collected, acid hydrolyzed, and submitted to HPLC-EC using system 4. EC trace 1, product of acid hydrolysis; EC trace 2, cochromatography of product with 8-aminoguanine.
under the same conditions as those used for the incubations with 2-NP or 2-NP nitronate. Reverse-phase HPLC analyses of the reaction mixtures obtained with 1-NP or 1-NP nitronate were equivalent to trace 2 in Figure 3A. When PAPS was omitted from the incubation mixture, no formation of 8-aminoguanosine or increase in 8oxoguanosine was observed (trace 2 in Figures 3A and 4). The same results were obtained when either the enzyme, mercaptoethanol, or the nitronate was individually omitted from otherwise complete incubation mixtures, or when boiled enzyme (100 "C for 10 min) was used (results not shown; equivalent to trace 2 in Figures 3A and 4). As shown in Figures 5 and 6, respectively, maximum product formation was observed at 2 mM substrate concentration and 90-min incubation time. The formation of 8-aminoguanosine was accompanied by a time-dependent increase in the level of 8-oxoguanosine. When the parent nitroalkane 2-NP was used as substrate in these incubations at pH 7.3, the yield of 8-aminoguanosine decreased from 0.1 nmol to approximately 0.03 nmol. The effects of pH on the in vitro activation of 2-NP nitronate by aryl sulfotransferase are shown in Table 2.
348 Chem. Res. Toxicol., Vol. 7, No. 3,1994
Sodum et al. Table 2. Effect of pH on the Formation of 8-Aminoguanosine in the in Vitro Sulfotransferase Reaction.
1.0 n A or 0 . 4 AV 0 253 nm
....,.....,...,....._............_........,.,. 1
0
I
I
l
i
.........................................................................................
1.
3
l
I
l
I
l
I
,
I
I
I
10 20 ELUTION VOLUME, mL
1 l
30
Figure 4. Analyses by cation-exchange HPLC (system 3) of products of the in vitro aryl sulfotransferase reaction: 1, UV response; 2-4, EC response: 2, incubation mixture containing 2-NP nitronate but no PAPS, otherwise complete; 3, complete reaction mixture using 2-NP nitronate as Substrate; 4, cochromatography of 8-aminoguanosine standard with products of reaction mixture shown in trace 3. 7n
..;
15
E
'E
>
2
I
10
Q I
6
57
o ! 0
I
I
I
2
4
6
10
8
2-NP Nitronatr Concrntration. mM
[SI
Figure 5. Rate of 8-aminoguanosine formation in the in vitro aryl sulfotransferasereaction mixtures at various substrate (2NP nitronate) concentrations.
0.00
I 0
' , 20
8-NH2-Guo(nmol)
pH
8-NH2-Guo(nmol)
5.4 6.5 6.9
0.0 0.06
7.1 7.3 8.2
0.16 0.11 0.05
0.08
a The completesystem (seeExperimental Section)containing 2 - N P nitronate was incubated 90 min at 37 O C .
1.
I
pH
I
I
I
I
40
60
80
100
120
Reaction Time ( m i d
Figure 6. Time course of aryl sulfotransferase-mediatedin vitro formation of 8-aminoguanosine at 37 O C . 2-NP nitronate waa used as substrate. The maximum yield of 8-aminoguanosine was obtained a t pH 7.1; at pH 8.2 the amount of product was about 3196 of that at pH 7.1; and no reaction was observed at pH 5.4.
Effects of Sulfotransferase Inhibitors and N-Hydroxy-2-AAF on the in Vitro Formation of 8Aminoguanosine. Table 3 shows that aryl sulfotransferase inhibitors, PCP and DCNP, suppress the in vitro formation of 8-aminoguanosine. As also observed in the in vivo studies, PCP was a stronger inhibitor than DCNP. While 1 pM DCNP resulted in only a 51% inhibition, PCP a t the same concentration inhibited the formation of 8-aminoguanosine almost completely (98%). The formation of 8-oxoguanosine in the incubation system, however, was only incompletely inhibited (65 96 ) even at concentrations of PCP or DCNP as high as 10 pM. This suggests that 8-oxoguanosine may be formed by both a sulfotransferase-catalyzedreaction and a nonenzymatic pathway in this in vitro system. Control incubations in the absence of PAPS, but in the presence of PCP or DCNP, showed an increase in the level of 8-oxoguanosine (results not shown). At concentrations of 66 or 660 mM in the incubation mixture, PG showed no inhibitory effect on 8-aminoguanosine formation. Interestingly, the increase in the level of 8-oxoguanosine was suppressed completely (100%)by PG at a concentration of 660 mM, and by approximately 60 % at 66 mM concentration. It is possible that PG may be acting as a hydroxyl radical scavenger. Alternatively, taking into regard the mechanisms proposed earlier ( l o ) ,while PG may inhibit the reaction of HzO with C8 of a postulated 7-aminoguanosine intermediate, the formation of 8-aminoguanosine would not be affected, since the latter reaction may simply involve an internal rearrangement (10). At 0.5 mM concentration, N-OH2-AAF also inhibited the formation of 8-aminoguanosine completely (Table 3). Discussion From earlier studies showing that 2-NP is hepatocarcinogenic in male Sprague-Dawley rats (3, 4 ) and is mutagenic in a number of systems (8, 9, 19-24), it was evident that the compound must act through the production of some sort of DNA damage. However, until recently, no information existed concerning the type of damage or the pathways by which it was produced. In 1989, our laboratory reported that 2-NP, but not the noncarcinogenic ( 3 )and nonmutagenic ( 8 , 9 )isomer, 1-NP, caused an increase in the amounts of 8-oxoguanine in rat liver DNA and RNA and, in addition, caused the appearance of other modified nucleosides in DNA and RNA that were detectable using HPLC-EC (5). These 2-NP-induced nucleic acid modifications seem to be relevant to the hepatocarcinogenicity of 2-NP since they are almost completely absent from the rat kidney (7),which is not a target for 2-NP (31,and are present in much lower amounts in the livers of male New Zealand White rabbits (25)which, relative to male Sprague-Dawley rats, are resistant to the toxicity and hepatocarcinogenicity of 2-NP (4). Recently, we identified another of the modifications produced by 2-NP in rat liver RNA and DNA as C8
Sulfotransferase Activation of 2-Nitropropane
Chem. Res. Toxicol., Vol. 7, No. 3, 1994 349
Table 3. Effects of Sulfotransferase Inhibitors on the Aryl Sulfotransferase-MediatedFormation of 8-Amino- and 8-Oxoguanosinesin Vitro. &NHz-Guob &oxo-Guob yield (nmol) % inhibition inhibitor yield (nmol) % inhibition 0.101 f 0.010 0.055 f 0.009 none 2.0 0.098 f 0.002 PG, 660 mM 90.9 0.005 f 0.008 PG, 66 mM 0 60 0.022 t 0.003 0.120 f 0.008 0.002 f 0.003 97.7 PCP, 1pM 67.3 0.018 f 0.010 PCP, 10 pM NDC 100 60 0.022 f 0.010 DCNP, 1r M 51.3 0.032 f 0.016 41.8 0.049 f 0.010 0.005 f 0.009 94.7 60 DCNP, 10 pM 0.022 f 0.016 N-OH-2-AAF,0.5 mM ND 100 98.2 0.001 f 0.007 a Mean & SD obtained using three different liver sulfotransferase preparations from separate rats. T otal amount formed in the reaction mixture. The basal level of 8-oxo-Guo in control incubations performed without PAPS was 0.135 f 0.007 nmol; 8-NHg-Guowas undetectable in these incubations. ND = not detected.
amination of guanine residues (10). To account for this unique base modification, we proposed (10) that 2-NP is converted in vivo to an aminating species by a series of nonenzymatic and enzymatic reactions as shown in Scheme 1. A step involving the enzymatic formation of a reactive sulfate or acetate intermediate was central to this proposed mechanism. The results presented here demonstrate clearly that aryl sulfotransferase can mediate the activation of 2-NP to a species capable of modifying nucleic acids. Pretreatment of rats with aryl sulfotransferase inhibitors, PCP or DCNP (13,141,but not the alcohol sulfotransferase inhibitor (18) PG, significantly decreased the 2-NP-induced formation of 8-aminoguanine, 8-oxoguanine,and the yet unidentified modifications, DX1 and RX1, in liver nucleic acids. The role of sulfoconjugation in the activation of 2-NP was confirmed by demonstrating that, in the presence of 2-NP or 2-NP nitronate, PAPS, and mercaptoethanol, partially purified rat liver aryl sulfotransferase can catalyze the formation of 8-aminoguanosine and 8-oxoguanosine from guanosine in vitro. This reaction did not proceed in the absence of either the enzyme or its cofactor PAPS, or the substrate 2-NP, or when boiled enzyme was used. Mercaptoethanol was also required for the in vitro formation of &amino- and 8-oxoguanosines. Omission of mercaptoethanol in the otherwise complete incubation system did not yield any detectable 8-aminoguanosine or increase in the level of 8-oxoguanosine. This experiment does not show conclusively whether deoxygenation of nitronate occurs before or after sulfonation because mercaptoethanol is required for the enzyme to catalyze sulfoconjugation (15). When the primary nitroalkane 1-NP or its nitronate was substituted for 2-NP in the reaction, no formation of 8-aminoguanosineor increase in the level of 8-oxoguanosine was detectable, indicating that 1-NP and 1-NP nitronate are not substrates of rat liver aryl sulfotransferase. These results are in accord with the differences in the toxicity, mutagenicity, and carcinogenicity of 2-NP and 1-NP (3, 5,9,19-23). The in vitro formation of 8-aminoguanosine and 8-oxoguanosine was strongly inhibited by the aryl sulfotransferase inhibitors PCP and DCNP, again reflecting accurately the conditions in vivo. The in vitro aryl sulfotransferase-catalyzed formation of 8-aminoguanosine and 8-oxoguanosine was also inhibited by the hepatocarcinogen N-OH-2-AAF, suggesting that aryl sulfotransferase IV, the enzyme that activates N-OH-2-AAF (26-281, might specifically be involved. Further purification of the enzyme and more extensive kinetic studies are necessary to rigorously test this possibility; however, additional indirect evidence for the specific involvement of sulfotransferase IV comes from
data indicating that male New Zealand White rabbits, which have significantly lower levels of hepatic aryl sulfotransferase IV than do male Sprague-Dawleyrata (29), are also much less susceptible to the induction of liver DNA and RNA modifications (25) as well as hepatocarcinogenesis ( 4 ) by 2-NP. When 2-NP was added to the incubation mixture in the nitronate form, the yield of 8-aminoguanosine was 3-5 times greater than when the nitro form of the substrate was used. Interestingly, the nitronate form of 2-NP was also found to be more mutagenic in Salmonella typhimurium than the parent nitro compound (20). 2-NP, which has been reported to have a pKa of 7.63 (II), will ionize to the nitronate form in appreciable amounts at physiological pH; however, this reaction as well as the reversion of the nitronate to the parent nitro form is unusually slow, with a pseudo-first-order rate constant of 0.15 h-l at pH 7.4 (11). It appears evident that both a relatively low PKa and a long half-life of the nitronate form are important factors in determining the suitability of a given nitroalkane as a substrate for sulfotransferasemediated activation to an aminating species. In the case of 1-NP, which is not a substrate for the sulfotransferase reaction, the PKa is 8.98 (17) and the pseudo-first-order rate constant for the equilibration of the nitronate with the nitro form has been reported as 24.2 h-I at pH 7.4 (11). The in vitro incubations using partially purified aryl sulfotransferase did not produce detectable amounts of any species with the chromatographic and electrochemical characteristics of the modified rat liver RNA (presumed) nucleoside, RX1 (5). Yet, the results shown in Figure 1 and Table 1indicate that the 2-NP-induced formation of RX1 is inhibited by PCP and DCNP to exactly the same extent as the formation of 8-aminoguanosine, implying that aryl sulfotransferase is involved in the formation of both these modified nucleosides. The explanation for this apparent discrepancy may be trivial or complex. It is of course possible that RX1 may be derived from a nucleoside other than guanosine in vivo. On the other hand, we have previously noted that RX1 and DX1 (the latter may be the structural counterpart of RX1) are not stable, even at -20 "C, once the corresponding nucleic acids are enzymatically hydrolyzed to the nucleoside level. It is therefore possible that, if derived from guanine, RX1 is stabilized by some primary or secondary structural features of the nucleic acid chains. Once these protective features are disrupted by hydrolysis, RX1 may degrade rapidly. If formed in vitro from guanosine, the species may be undetectable because its rate of degradation may equal or exceed its rate of formation. The identities of RX1 and DX1 are subjects under study in this laboratory.
360 Chem. Res. Toxicol., Vol. 7, No. 3, 1994
Previously, it had been suggested that the enzymatic denitrification of 2-NP to nitrite by liver P450 enzyme(& might lead to the formation of diazo and N-nitroso compounds capable of chemically modifying biologically important macromolecules (30). Although such a mechanism could, in part, be responsible for some of the toxic effects of 2-NP, the denitrification of 2-NP by liver cytochrome P450s is not responsible for the formation of 8-aminoguanine, 8-oxoguanine, or the other unidentified electrochemicallyactive nucleic acid modifications we have described (5, 10). These modifications, as shown in the present study, are produced through the action of aryl sulfotransferase. Moreover, preliminary data2 from our laboratory indicate that depletion of rat liver cytochrome P450 by pretreatment with cobalt protoporphyrin IX results in an increase in the rat liver nucleic acid modifications induced by 2-NP while induction of cytochrome P450 by phenobarbital pretreatment significantly decreases the levels of 2-NP-induced nucleic acid modifications. These results are not consistent with a role of cytochrome P450 in the activation of 2-NP to a species capable of producing the nucleic acid modifications that we observe (5). Sulfoconjugations catalyzed by sulfotransferases are known to occur with a wide variety of hydroxyl groupcontaining compounds, including phenols, primary and secondary alcohols,phenolic and alicyclichydroxy steroids, and hydroxylamines (31). Although the products of such reactions in vivo are generally less toxic and more watersoluble, facilitating removal from the body, sulfoconjugations can also lead to intermediates which are more reactive and toxic than the parent compounds (32,331.A classical example of the latter type of reaction is the sulfotransferase-catalyzed activation of the carcinogen N-OH-AAF to the N-0-sulfonic acid which dissociates to an electrophilic nitrenium ion that can modify DNA, RNA, and protein (26, 32, 34, 35). Since these early findings, other carcinogens of diverse chemical structures, including N-nitroso-N-( 2-hydroxyethy1)methylamine(28),&nitrosodiethanolamine (36),1-(hydroxymethy1)pyrene(37),1’hydroxysafrole (38), 7-(hydroxymethy1)-1%-methylbenz[alanthracene (39))hydroxylamines of heterocyclicamines (40),and other compounds (32) have also been shown to be activated by sulfotransferases to reactive species that can modify DNA. The present findings constitute the first report of a similar activation, by sulfotransferase, of a carcinogenic secondary nitroalkane.
Acknowledgment. This work was supported in part by Grant ES03257 from the National Institute of Environmental Health Sciences. The excellent technical assistance of Adriana Surace in parts of this work is acknowledged. References International Agency for Research on Cancer (1982) ZARC Monog r a p h on the eualuation of carcinogenic risk of chemicals to man, Some industrial chemicals and dyestuffs,\rol.29, pp 331-343, WHO, Lyon, France. Hoffmann, D. H., and Rathkamp, G. (1968) Chemical studies on tobaccosmoke. 111. Primary andsecondary nitroalkanesincigarette smoke. Beitr. Tabakforsch. 4, 124-134. Fiala, E. S., Czerniak, R., Castonguay, A., Conaway, C. C., and Rivenson, A. (1987) h a y of I-nitropropane, 2-nitropropane, 1-azoxypropane,and 2-azoxypropane for carcinogenicity by gavage in Sprague-Dawley rats. Carcinogenesis 8,1947-1949.
Sodum et al. Lewis, T. R., Ulrich, C. E., and Busey, W. M. (1979) Subchronic inhalation toxicity of nitromethane and 2-nitropropane. J.Enuiron. Pathol. T O X ~ C 2,233-249. O~. Fiala, E. S., Conaway, C. C., and Mathis, J. E. (1989)Oxidative DNA and RNA damage in the livers of Sprague-Dawleyrats treated with the hepatocarcinogen 2-nitropropane. Cancer Res. 49,5518-5522. Conaway, C. C., Guo, N., Hussain, N. S., and Fiala, E. S. (1991) Comparison of oxidative damage to rat liver DNA and RNA by primarynitroalkanes, secondarynitroalkanea,cyclopentanoneoxime, and related compounds. Cancer Res. 51, 3143-3147. Guo, N., Conaway, C. C., Husaan, N. S., and Fiala, E. S. (1990) Sex and organ differences in oxidative DNA and RNA damage due to treatment of Sprague-Dawleyrats withacetoximeor 2-nitropropane. Carcinogenesis 11, 1659-1662. Conaway, C. C., Hueeain, N. S., Way, B. M., and Fiala, E. S. (1991) Evaluation of secondary nitroalkanes, their nitronates, primary nitroalkanes, nitrocarbinole, and other aliphatic nitro compounds in the Ames Salmonella assay. Mutat. Res. 261, 197-207. Andrae, U., Homfeldt, H., Vcgl,L., Lichtmannegger,J., and Summer, K. H. (1988) 2-Nitropropane induces DNA repair synthesis in rat hepatocytes in vitro and in vivo. Carcinogenesis 9,811-815. Sodum,R. S., Guo, N., and Fiala, E. S. (1993) EAminoguanine: A base modification produced in rat liver nucleic acids by the hepatocarcinogen 2-nitropropane. Chem. Res. Toxicol. 6,269-276. Linhart, I., Gescher, A., and Goodwin, B. (1991)Investigation of the chemical basis of nitroalkane toxicity: Tautomerism and decomposition of propane 1-and 2-nitronate under physiologicalconditions. Chem.-Biol. Interact. 80, 187-201. Kawazoe, Y., and Huang, G.-F. (1972) Direct C-amination of guanoeine with hydroxylamine-0-sulfonic acid. Simplest model reaction possibly involved in DNA damage leading to carcinogeneais by N-arylhydroxylamiiea. Chem. Pharm. Bull. 20, 2073-2074. Mulder, G. J., and Scholtem, E. (1977)Phenol sulfotransferase and uridine diphosphate glucuronyltransferase from rat liver in vivo and in vitro. Biochem. J. 166,553-559. Lai, C.-C., Miller, E. C., Miller, J. A., and Liem, A. (1987) Initiation of hepatocarcinogenesiain infant male B6C3F1mice by N-hydroxy2-aminofluorene or N-hydroxy-2-acetylaminofluorene depends primarily on metabolism to N-sulfoxy-2-aminofluoreneand formation of DNA-(deoxyguanosin-&yl)-2-aminofluoreneadducts. Carcinogenesis 8,471-478. (15) Sekura, R. D., Duffel, M. W., and Jakoby, W. B. (1981) Aryl sulfotransferases. Methods Enzymol. 77, 196-206. (16) Lowry, 0. H., Roeebrough, N. J., Farr, A. C., and Randall, R. J. (1951)Protein measurement with the Folin phenol reagent. J.Biol. Chem. 193, 265-275. (17) Nieleen, T. A. (1981) Nitronic acids and esters. In The chemistry of the nitro and nitroso groups (Feur, H., Ed.) Part 1,pp 349-486, Robert E. Kreiger Publishing Co., New York. (18) Kroeger-Koepke, M., Koepke, S. R., Hernandez, L., and Michejda, C. (1992) Activation of j3-hydroxyalkylnitrosine to alkylating agents: Evidence for the involvement of a sulfotransferase. Cancer Res. 52,3300-3305. (19) George, E., Burlinson, B., and Gatehouse, D. (1989)Genotoxicity of 1-and 2-nitropropane in the rat. Carcinogenesis 10, 2329-2334. (20) Fiala, E. S., Conaway, C. C., Biles, T. W., and Johnson, B. (1987) Enhanced mutagenicity of 2-nitropropane nitronate with respect to 2-nitropropane-possible involvementof free radical species. Mutat. Res. 179, 15-22. (21) Hite, M., and Skeggs, H. (1979) Mutagenic evaluation of nitroparaffins in the Salmonella typhimuriumlmammalian-microsome test and the micronucleus test. Enuiron. Mutagen. 1, 383-389. (22) Speck, W. T., Meyer, L. W., Zeiger, E., and Rosenkranz, H. S. (1986) Mutagenicity and DNA-modifyingactivity of 2-nitropropane. Mutat. Res. 104, 49-54. (23) Lofroth,G.,Nilsson,L.,andAndersen, J.R. (1986)Structure-activity relationship of nitroalkane-induced mutagenicity in the Ames Salmonella assay. B o g . Clin. Biol. Res. 209, 149-155. (24) Davies, J. E., Mynett, K., Gescher, A., and Chipman, J. K. (1993) DNA modification and repair by 2-nitropropane is extensive in hepatocytes of rats compared to those of humans and mice. Mutat. Res. 287, 157-164. (25) Fiala, E. S., Nie, G., Sodum,R. S., Conaway, C. C., and Sohn,0. S. (1993) 2-Nitropropane-induced liver DNA and RNA base modification: Differences between Sprague-Dawley rats and New Zealand White rabbits. Cancer Lett. 74,9-14. (26) DeBaun, J. R., Miller, E. C., and Miller, J. A. (1970) N-Hydroxy2-acetylaminofluorene sulfotransferase: Ita probable role in carcinogenesisand in protein-(methion-S-yl)binding in rat liver. Cancer Res. 30, 577-595. (27) Sekura, R. D., and Jakoby, W. B. (1981) Aryl sulfotransferase IV from rat liver. Arch. Biochem. Biophys. 211,352-359. (28) Wu, G. S.-C., and Straub, K. D. (1976) Purification and characterization of N-hydroxy-2-acetylaminofluorenesulfotransferase from rat liver. J . Biol. Chem. 251, 6529-6536.
Sulfotransferase Activation of 2-Nitropropane DeBaun, J. R., Rowley, J. Y., Miller, E. C., and Miller, J. A. (1968) Sulfotransferase activation of N-hydroxy-2-acetylaminofluorenein rodent livers susceptible and resistant to this carcinogen. Roc. SOC. Exp. Biol. Med. 129, 268-273. Ullrich, V., Hermann, G., and Weber, P. (1978) Nitrite formation from 2-nitro~rooanebv microsomal monooxvzenases. Biochem. -Pharmacal. i7,i301-2304. Jakobv. W. B., Sekura, R. D., Lyon, E. S., Marcus, C. J., and Wang, J.-L. (i980) Sulfotransferases. I n Enzymatic basis ojdetoxication (Jakoby, W. B., Ed.) Vol. 11,pp 199-228, Academic Press, New York. Miller, J. A., and Miller, E. C. (1986)Electrophilic sulfuric acid ester metabolites as ultimate carcinogens. In Biological Reactive Zntermediates ZII (Koscis, J. J., Jollow, D. J., Witmer, C. M., Nelson, J. O., and Snyder, R., Eds.) pp 583-595, Plenum Press, New York. (33) Miller, J. A., Surh, Y.-J., Liem, A., and Miller, E. C. (1991) Electrophilic sulfuric acid ester metabolites of hydroxymethyl aromatic hydrocarbons as precursors of hepatic benzylic DNA adducts in vivo. In Biological Reactive Intermediates IV (Witmer, C. M., Snyder, R., Jollow, D. J., Kalf,G. F., Koscis, J. J., and Sipes, I. G., Eds.) pp 555-567, Plenum Press, New York. (34) King, C. M., and Philips, B. (1968) Enzyme-catalyzed reactions of the carcinogen N-hydroxy-2-fluorenylacetamidewith nucleic acid. Science 159, 1351-1353. (35) Weisburger, J. H., Yamamoto, R. S., Williams, G. M., Matsushima, T., and Weisburger, E. K. (1972) On the sulfate ester of N-hydroxy-
Chem. Res. Toxicol., Vol. 7, No. 3, 1994 351 N-2-fluorenylacetamide as a key ultimate hepatocarcinogen in the rat. Cancer Res. 32,491-500. (36) Sterzel, W., and Eisenbrand, G. (1986)N-Nitroeodiethanolamine is activated in the rat to an ultimate genotoxic metabolite by sulfotransferase. J. Cancer Res. Clin. Oncol. 111, 20-24. (37) Surh, Y.-J., Blomquest, J. C., and Miller, J. A. (1991) Activation of 1-hydroxymethylpyreneto an electrophilicand mutagenic metabolite by rat hepatic sulfotransferase activity. In Biological Reactive Intermediates ZV (Witmer, C. M., Snyder, R., Jollow, D. J., Kalf, G. F., Koscis, J. J., and Sipes, I. G., Eds.) pp 383-391, Plenum Press, New York. (38) Boberg, E. W., Miller, E. C., Miller, J. A., Poland, A., and Liem, A. (1983) Strong evidence from studies with brachymorphic mice and pentachlorophenol that 1'-sulfooxysafrole is the major ultimate electrophilic and carcinogenic metabolite of 1'-hydroxysafrole in mouse liver. Cancer Res. 43, 5163-5173. (39) Watabe, T., Hiratsuka, A., and Ogura, K. (1987)Sulphotransferaeemediated covalent binding of the carcinogen 7,12-dihydroxymethylbenz[a]anthracene to calf thymus DNA and its inhibition by glutathione transferase. Carcinogenesis 8,445-453. (40) Davis,C. D., Schut, H. A. J., andSnydenvine,E.G. (1993)Enzymatic phase I1 activation of the N-hydroxylaminesof IQ, MeIQx and PhIP by various organs of monkeys and man. Carcinogenesis 14,20912096.
