Genetic Regulation of Aryl Hydrocarbon Hydroxylase Induction

In the inbred aryl hydroxylase by polycyclic hydrocarbons increases rapidly in the neonatal this response to these pharmacological, exogenous compounds does not appear in the DBA/2N, or NZB/BLN strains. The develop-ment of the constitutive hydroxylase activity in the neonatal period is the same in all four strains. Small differences in pH optima, substrate affinities, relative thermolability, or the do not for the large genetic difference in the response to polycyclic hydrocarbons. The hydrocarbons the


SUMMARY
In the inbred C57BL/6N mouse, aryl hydrocarbon hydroxylase induction by polycyclic hydrocarbons increases rapidly in the neonatal period; this response to these pharmacological, exogenous compounds does not appear in the DBA/2N, NZW/BLN, or NZB/BLN strains.
The development of the constitutive hydroxylase activity in the neonatal period is the same in all four strains.
Small differences in pH optima, substrate affinities, relative thermolability, or benzo[a]pyrene metabolism, which might have affected the enzyme assay, do not account for the large genetic difference in the response to polycyclic hydrocarbons.
The hydroxylase induction by aromatic hydrocarbons is inherited as a simple autosomal dominant trait, designated the ah locus.
In the individual mouse that is genetically responsive to polycyclic hydrocarbons, the hydroxylase activity is induced as an all-or-none phenomenon in all tissues which regularly contain the polycyclic hydrocarbon-inducible enzyme.
The magnitude of induction of the hydroxylase is 4-to S-fold in liver, and ranges from 5-to more than SO-fold in kidney, bowel, lung, and skin.
The formation of a spectrally distinct carbon monoxide-binding cytochrome from liver microsomes is associated with the polycyclic hydrocarbon-inducible enzyme activity and is not seen in the liver of mice which are genetically nonresponsive to aromatic hydrocarbons.
Phenobarbital induces the hepatic oxygenase to similar levels (i.e. about 2-fold) in all four strains.
We have been studying (l-14) the mechanisms regulating microsomal enzyme induction from two fundamental points of interest.
First, the stimulation of enzyme activity in response to a foreign, pharmacological stimulus such as a polycyclic hydrocarbon or a drug serves as an interesting experimental * The first paper in this series has appeared (1). Presented in part at the Svmuosium on Pharmacogenetics (2) system for studying genetic expression in eukaryotic cells and for comparing this experimental model with other systems in which endogenous compounds cause enzyme induction.
Second, these membrane-bound mixed-function (15) oxygenases are important for the oxidative metabolism of most drugs, polycylic hydrocarbons, and insecticides, as well as lipophilic endogenous substrates (16) such as steroids, bilirubin, indoles, thyroxine. sympathomimetic amines, hemin (17), and fatty acids (18), Thus, the day-to-day fluctuations in the levels of these oxidases influence the intensity and duration of drug action and the rate of metabolism of chemical carcinogens (16), insecticides (19), toxic chemicals (20)) and numerous normal body substrates (16).
This NADPHlinked enzyme has as an active site for its oxidative function the CO-binding cytochrome P450 (al), so named because the reduced form of this pigment upon combination with CO has a Soret. maximum at about 450 nm. This hydroxylase activity in liver microsomes from rats treated with MC can be reconstituted from three components: flavoprotein, lipid, and a cytochrome-containing fraction; substrate specificity is presumably determined by the type of cytochrome comprising the active oxidative site (22). There is growing evidence (1,2,7, P4s0 consists of a mixture of at least two hemoproteins. Furthermore, a direct correlation exists between the formation of high spin iron-containing cytochrome P 450 and aryl hydrocarbon hydroxylase induction in mouse liver by polycyclic hydrocarbons (2). It is conceivable that an incidental increase in this type of cytochrome P450 occurs during the 1'Sstimulated nonspecific proliferation of the hepatic smooth endoplasmic reticulum and that this incidental increase explains the induction of the hydroxylnse activity by PB. Although the polycyclic hydrocarbon-inducible hydroxylase activity exists in most extrahepatic tissues as well (5, 10, 29%31), it is not known whether or not a multicomponent membrane-bound euzyme system similar to that in liver is responsible for polycyclic hydrocarbon hydroxylation in any of these other tissues.
A chronological sequence of events occurs during the process of hydroxylase induction in mammalian cell culture by polycyclic hydrocarbons (2, 6-9, 11). The rapid entry of polycyclic hydrocarbons into the cell is independent of temperature (S), and synthesis of an induction-specific RNA takes place during the first 20 mill (9). The oxygenase induction (6,9) and the appearance of a new spectrally distinct CO-binding cytochrome (7) are concomitantly dependent upon translation involving this RNA species. The parent polycyclic hydrocarbon molecule is metabolized by the control enzyme and subsequently by the induced hydroxylase as well; this process results in metabolites covalently bound to cellular material (8,9) and water-soluble derivatives excreted into the growth medium (8). More recently (14) we found that the primary action of PB, as well as that of polycyclic hydrocarbons, on aryl hydrocarbon hydroxylase induction may be transcriptional" and that wit.h either inducer there is also a secondary effect at the posttranslational level in which the regular rate of decay of the induced enzyme is retarded.
In the previous paper (1) of this series, we found differences between C57BL/6N and DBA/BK fetal cells in culture with respect to this chronological sequence of events. Thus, compared with fetal cultures from the C57BL/6X mouse, similar cell cultures from the DBA/2N mouse show (a) a relative lack of inducible hydroxylase activity in response to polycyclic hydrocarbons, (b) a diminished formation of the new spectrally distinct CO-binding cytochrome, and (c) indirect evidence for a. decreased expression of induction-specific RNA.
On the other hand, we found (1) no differences between these two cell types in the rat,e of uptake of polycyclio hydrocarbons by these cells, in the gross binding of polycyclic hydrocarbons to subcellular fractions, or in the rate of degradation of the induced hydroxylase activities.
By examining various tissues from the inbred C57BL/6N, DBA/2N, NZW/BLN, and NZB/BLN mouse strains and from offspring of the appropriate genetic crosses between these inbred strains, we show in this report that aryl hydrocarbon hydroxylase induction and formation of the spectrally distinct CO-binding hepxtic cytochrome in response to polycyclic hydrocarbons follolv simple Rlendelian genetics.
* Both processes are inhibited if actinomycin D is added simultaneously with the inducer initially.
Whereas the inducers may be stimulating the rate of synthesis of specific mRNh species, these studies do not in fact distinguish between a mechanism whereby the inducers ;Lct to amDlifv specific genes or to allow transpdrt or stabilization of (oth&wiscj rapidly degraded inductiorr-specific RNA which is continuously synthesized yet rapidly degraded within the nucleus. i.e. an automatic day-night (16 hours to 8 hours) cycle and avoidance of exposure to pharmacologically active compounds such as cigarette smoke and insecticides.
Except where indicated, the mice ranged between 3 and 10 weeks of age at the time the hydroxylase specific activities and cytochrome Ptsa content were deternlined.
Among inbred mice of the same strain, we found no statistically significant difference (p > .05) in the control, MC-inducible or PB-inducible enzyme or hepatic hemoprotein levels between male and female littermates or betlveen a-week-old and lo-week-old animals of the same genetic constitution.
MC-treated mice were injected once intraperitoneally with 80 mg of MC in corn oil per kg of body weight 24 hours before death; control mice received corn oil only. PB-breated mice received intraperitoneally SO mg of PB in normal (O.gOc/,) saline per kg of body weight on each of 3 successive days before sacrifice. All experiments were begun at approximately the same hour of the day.
Immediately upon exsanguination of the animal, the minced tissues from each individual mouse were separately washed free of blood in ice-cold 0.15 M KCl-0.25 RI potassium phosphate buffer, pH 7.25. Tissue homogenates of t,he bo\T-el, lung, kidney, and skin at concentrations between 3 and 10 mg of protein per ml were prepared for the enzyme assay. Liver homogenates were centnfuged at 15,000 x g for 15 min, and the supernatant fraction from this was recentrifuged at 78,000 x g for 90 min. The surface of the microsomal pellet was washed several times. The microsomes were then suspended in 30% glycerol-O.25 31 potassium phosphate buffer, $1 7.25, and immediately used concomitantly for the determinations of hydroxylase activity and cytochrome I'd50 content from each individual mouse.
Enxylne Assay-The determination of hydroxylase activity and tissue protein concentration in duplicate was similar to that (2,3,8) previously descr,betl.
For determining the hepatic enzyme activity, the l.OO-ml react.ion mixture included 50 prnoles of I)ot:l&unl phoxphate buffer, pI1 7.2, 0.36 pmole of NADPH, 0.39 b nlole of ?;Al>H, 600 pg of bovine herun albumin, 3 pmoles crosomal protein), and 80 nmolea of the substrate benzo[a]pyrene added in 40 ~1 of methanol just prior to the IO-min incubation. We found (2) that when low arnounts of protein are present in the assay mixture, additional protein in the form of albumin aids the solubility of the benzo[a]p)-rene and therefore the reproducibility of the assay. For assaying the nonhepatic enzyme activity, the l.OO-ml reaction misture contained 50 pmoles of potassium phosphate buffer, pH 7.5, 0.36 pmole of NADPH, 0.39 pmole of NADH, 0.10 ml of tissue homogenate (containing 0.30 to 1.0 mg of protein), and 80 nmoles of benzo [u]pyrene added in 40 ~1 of methanol just prior to the 30-min incubation.
Following incubation at 37", addition of cold acetone, and extraction with hexane (3), the alkali-extractable metabolites were examined with an Iminco-Bowman model 4-8202 SPF recording spectrophotofluorometer (American Instrument Company, Baxter Laboratories, Silver Spring, Maryland) ; fluorescence corresponding to 3-h!tlroxyberlzo[cr]p\;reire has an activation peak at 396 nm and a11 emission maximum at 522 nm (3). The fluorescence of a blank sample, to which benzo[a]pyrene had been added after the incubation and addition of acetone, was subtracted from the fluorescence of each esperimental sample. One unit of aryl hydrocarbon hydroxylase activity has been defined (7) as that amount of enzyrne catalyzing the formation per min at 37" of hydrosylated product causing fluorescence equivalent, to that of 1 pmole of 3-hydroxybenzo[a]pyrene. Protein concentrations were determined by a slight modification of the method as described by Lowry et al. (32), with crystalline bovine serum albumin as the standard.
The limit of sensitivity for the assayed specific hydrosylase activit,y is about 0.10 unit per mg of protein, and duplicat,e determinations normally vary less than 10% (2). III separate experiments the optimal l)H conditions for the hydrosylase activity from different tissue sources were determined. In other separate esl)eriments the substrate [3H]berlzo[a]pyrene was used and the tot,al amount of alkali-extractable radioactivity was measured; 0.10. or 0.20-ml aliquots of the 1 N NaOH were dissolved in 1 ml of Suclear Chicago Solubilizer, and the solution was mixed with 10 1111 of a toluene scintillation mixture and count,ed in a Packard scintlllat,ion counter.
SpectrophotometrK~Difference spectra of turbid microsomal fractiolis were measrtred as previously described (7,21) in l-cm cuvettes at room trrnperature in a Shimadzu model MPS-50L multipurpose recording spectrophotometers (American Instrument Company, Baxter Laboratories, Silver Spring, %Iaryland). Wave length measurements were standardized by the use of a holmiunl oxide crystal.
The method of Oniura and. Sato (21) for determining the concentrations of CO-billding microsomal cytochromes was used. The experimental sample was saturated with CO by bubbling the gas gently for 1.3 to 30 set, and a base-line for the oxidized fractions in both sample and reference cuvettes was established. The suspensions in both cuvettes were then reduced by minimal amounts of solid K\a&Oa.
The concentrations of the CO-binding hemoproteins were determined from the CO-difference spectra: extinction coeflic,ietlts used were 91 mM-1 cm-1 for the difference in absorption between the Soret maximum and 490 nm for the pigments absorbing maximally in the 450 nm region, and 110 1llhT' cm -I for the change in absorbance between 420 nm and 490 nm for P420. 'The absorbance of P460 below the base-line at 420 nm is about 464;. of the absorption maslmum at 450 nm (33).
In this study the determinations of cytochrome Pdb,, were linear at microsomal protein concentrations ranging between 0.80 and 3.0 mg per ml.

RESULTS
Presence or Absence of Age-dependent Response of Hydroxylase Induction by S-Xethylcholanthrene- Table  I compares the hepatic hydroxylase activity of C57BL/6N and DBA/2N mice in response to MC as a function of age. The constitutive level of this enzyme5 in either mouse strain was detectable in utero, increased markedly during the first post parbm.
week, and was maximal in the 3-week-old meanling.
The phenomenon of this "physiologic" induction of hepatic oxygenase activity occurring immediately post partum is presumably related to such clinical entities as the Gray syndrome (34) and neonatal hyperbilirubinemia, in which the newborn and especially the prernature have not yet acquired the necessary microsomal enzymes for metabolizing chloramphenicol or bilirubin. The constitutlve hydroxylase in utero appeared significantly (p < .05) earlier in the C57BL/6X mouse liver, compared with that in DBA/SN liver; however, the postnatal rise in the control hepatic enzyme activity from either strain was not significantly different. Twenty-four hours after a single intraperitoneal dose of MC to the pregnant C57BL/6N mouse, the hydroxylase activity was induced transplacentally to detectable levels as early as 9 days before partur,tion.
Induction of this enzyme in response to the polycyclic hydrocarbon also increased markedly during the first week post partum and slowly declined after the weanling period.
In the DBA/XN mouse this age-dependent response of aryl hydrorarbon hydroxylase induction by MC did not, appear.
If one compares the control and MC-inducible hydroxylase n&vities at each age (first two columns in Table I), there is a 5. to more than 50.fold difference at the various ages, indicating a poor correlation between the constitutive and inducible enzyme levels. We have also found6 in fetal rat hepatocyte cultures that the maximally inducible hydroxylase activity is not dependent upon the constitutive enzyme level.
This marked increase in hydrosylase induction by XIC as a function of age also occurred in C5713L/GN nonhepatic tissues, whereas the lack of response to MC was found in DBA/BN nonhepatic tissues. Furthermore, we found a similar complete absence of response to polycyclic hydrocarbons in the NZW/BLN and NZB/BLN strains, and MC-inducible hydrouylase activity in the National Inst,tutes of Health General-Purpose and the C3H/HEN mouse strains. On the contrary, the transplacental and post partum stimulation of this oxygenase activity in the liver by PB was similar in each of the six strains mentioned. The transplacental induction of hydrosylase activity by l~olycyclic hydrocarbons occurs in fetal tissues of the hamster or rat, and the marked rise in both constitutive and inducible oxygenase activities immediately post partum has been previously noted (5,11,35).
E$ect of pH on Hydroxylase Activities from S-Methylcholanthrew-treated and Control C67BL/6N or DBA/SN Mice-The enzyme active sites may differ in such a way that MC-inducible and MC-noninducible hydroxylase activities are different with respect to pH optima, substrate affinity, relative thermolability, or product formation.
These possible considerations, which might affect the enzyme assay, are ruled out in the following two figures in which we compare the enzymes from C57BL/BN and DBA/2N mice. Similar results were obtained with MC-inducible and MC-noninducible hydroxylase activities from the other mouse strains. Fig. 1 illustrates the effects of pH on the hydroxylase activities from hepatic and renal microsomes of MC-treated and control C57BL/6N or DBA/2N mice. A pH range of 6.9 to 7.2 was optimal for the MC-induced and control liver oxygenase from C57BL/6N mice and for the enzyme from DBA/2N liver microsomes after MC treatment.
A pH optimum of about 6.4 was found for the constitutive hydroxylase from DBA/BN liver. The reason for this unusually low pH optimum is unknown but suggests some structural differences between the constitutive enzyme system from DBA/SN liver and the control hepatic hydroxylase from C57BL/6N or the liver oxygenase from MCtreated C57BL/6N or DBA/SN mice. Of further interest, we found this same low pH optimum of 6.4 for the constitutive enzyme from both NZW/BLN and NZB/BLN strains.
However, the small differences in pH optima for the hydroxylase activities of constitutive and MC-treated C57BL/6N and DBA/PN mice do not account for the 5-fold to more than 50.fold stimulation of enzyme activity by MC that develops in the neonatal C57BL/ 6N mouse (Table I).
The stability of hepatic cytochrome P450 is greatest between pH 7.2 and 7.5 (36). However, the CO-binding pigments cannot be compared at different pH values, since the stability of the MCinduced, PB-induced, and control forms of the cytochrome is pH-dependent (28). Thus, we chose to assay the hepatic hydroxylase activity and CO-binding hemoproteins at ~1-1 7.2 for the remainder of the results presented in this report.
A pH optimum of about 7.5 was observed for the MC-induced enzyme activity from C57BL/6N kidney microsomes. Because of the low enzyme activity in C57BL/6N kidney control microsomes and in renal microsomes from the hlC-treated or control DBA/BN mouse, the pH profile on these samples could not be accurately determined.
This higher pH optimum of the enzyme from an extrahepntic tissue is consistent with that found (2,3) in secondary cell cultures from rodent embryos, since the cells in culture are predominantly nonhepatic tissue (i.e., fibroblasts). Therefore, the estrnhepatic hydroxylase activity ~vns examined at p1-I 7.5 for all data shown in the remainder of this paper. ~4pparent K, Values and IIeat Inactivations for Ilydroxylase Activities jrom 3-Methylcholanthrene-treated and Control C57BL/ GA1-or DBJ /SKY JIice---The apparent Michaehs constants for the hepatic oxygena,<es from MC-treated and control C57BL/6N or Dl<a/2N mice ranged from 10 to 50 PM and were not significantly different from one another in several experiments.
Since the hydrorylaae artlvity apparently represents a functional membrane-bound multicomponent electron chain (22), it must be emphasized (3,13)  somes from either MC-treated or control DBA/SN mice were too low for accurately determining an apparent K, value.
We found that the constitutive hepatic enzymes from C57BL/ BN, DBA/2N, NZW/BLn', and SZB/BLS strains, the liver oxygenase activities in all four strains after MC treatment, and the MC-induced kidney hydroxylase activity in the C57BL/BK mouse were not different with respect to heat inactivation.
A 5-min preliminary treatment period of the enzyme in phosphate buffer without cofactors or substrate at 50" before the assay destroyed about one-half, and 5 min at 60" inactivated virtually all, of the hydroxylase activity in each case. This finding indicates to us that, if a multicomponent enzyme system is responsible for polycyclic hydrocarbon hydroxylation in earh case, the most thermolabile moiety is the same in earh of the basal and MCinducible hydrosylases.
In the F1 hybrids from crosses between C57BL/ 6N and any one of the DBA/2N, NZW/BLN, or NZB/BLK strains, and in offspring from the cross between inbred C57BL/ 6N and the F1 hybrid of C57BL/BN and DBA/2N parents (i.e., C/CD), the hydroxylase activity in response to MC rose about 4-fold.
In offspring from the backcross between inbred DBA/ 2N or NZW/BLN mice and FL hybrids from previous crosses between C57BL/BN and DBA/BN or C57BL/6N plus NZW/BLS (i.e. CD/D, CW/W, or CW/D), distribution of the oxygenase activity after MC treatment was distinctly bimodal: in approximately one-half of the population the enzyme was induced about 4-fold; in the other half the specific activities were not significantly different from control values. In MC-treated off spring from F1 hybrids of C57BL/6S and either DBA/2N or SZWj BLN parents (i.e. Fz generation), the hepatic microsomal hydroxylase activities were again distributed in a bimodal manner: in about, three-fourths of the population the oxygenase was induced about 4-fold by MC, whereas the enzyme levels in the remaining one-fourth of the MC-treated mice were not different from the constitutive levels. We found no correlation between the extent of MC-inducible hydroxylase activity and the coat color or sex in any of the genetic crosses examined.
Alto, we observed that other polycyclic hydrocarbons similarly induce the enzyme in the same patteru among these various inbred and hybrid mice.
Therefore, the expression of aryl hydrocarbon hydroxylase induction by polycyclic hydrocarbons is inherited as :I simple autosomal dominant trait, for which we postulate the genes -~1-1/1 and   ah (39). It can be seen in Table II that if we decide that strain C57BL/GN is homozygous for Ah and the other three strains are homozygous for ah, the microsomal oxygenase activity is inducible by MC in any mouse homozygous or heterozygous for Ah. A p < ,001 was found by chi square analysis in which MCinducible hepatic hydroxylase activity was observed with the expected frequencies of 50 and 75% in the Ahah x ahah and Ahah x Ahah crosses, respectively: there were 37 mice in which the osygennse was inducible by MC arnong a total of 70 offspring from the Ahah x ahah backcrosses; among 81 Fz offspring from the various genetic crosses, MC induced the hydroxylase system in 63 of these mice. Moreover, by MC administration to pregnant mice containing fetuses of an &ah x ahah genetic backcross (39), we can detect in utero fetuses having the MC-inducible hydroxylase and those nonresponsive to MC. Also, if such fetuses are individually placed in cell culture (39), the bimodal distribution in response to polycyclic hydrocarbons in the growth medium is found.
Among the MC-treated animals there are clearly two groups of specific activities: (a) those ranging from 550 to 760 in which the means are not significantly different from the means of the constitutive enzyme activities, and (b) those ranging from 2390 to 3090 in which the hydroxylase activity is induced between 4-and &fold.
Actually, the mean value of 3090 is significantly different (p < .Ol) from the means of 2500 and 2390. However, we feel that the variations in these maximally inducible levels probably reflect small differences in the assay (e.g. pH, chemical reagents, efficiency of the spectrophotofluorometer) or in the mice (e.g. effective dose of MC absorbed, age, hormonal, nutritional, or environmental factors) for the B-month period during which these determinations were made. Two general observations which were not significantly (p > .05) different should also be noted. First, among the inbred NZW/BLN and crosses involving NZW/ BLN with the other three strains, we observed a greater variation and generally higher mean values for the constitutive hydroxylase specific activities.
Second, in the MC-treated mice in which the enzyme was noninducible the mean specific activities of the oxygenase were generally less than those of the constitutive enzyme.
For all genetic groups examined in Table II, PB induced thm hepatic hydroxylase activity to the same extent.
The mean values ranged from 1090 to 1320 and were not significantly (p > .05) different from each other, but were significantly (p < .Ol) different from those of the constitutive oxygenase levels. The magnitude of induction by PB was about 2-fold or slightly less. Hence, induction of the enzyme by polycyclic hydrocarbons is expressed in an nutosomal dominant character; however, among the strains of mice examined, this simple Uendelian expression was not found for the hydrosylase induction by PB. This result is consistent with other studies (12-14, 26-28, 40-42) indicaating that microsomal enzyme induction by PB and by polyc~yclic hydrocarbons involves proresses whirl1 are distinctly different at least at one step in t,he sequence of events. For example, the combination of P13 and a polycyclic hydrocarbon ill vioo (41) or in cell culture (12, 13) stimulates certain hepatic oxygcnases to levels which are the sum of that induced by either inducer alone.
Genetic Expression of Formation of Spectrally Distllnct Hepatic CO-binding Cytochrome ajler 3-Slethylcholanthrene Treatment- Table III shows the levels of hepatic CO-binding pigments in some of t,he same groups of control, MCtreated, and PB-treated mice shown in the previous table. The mean values of constitutive cytochrome P450 ranged between 700 and 950 pmoles per mg of microsomal protein and were not significantly (p > .05) different from eacah other.
The CO-binding hemoprot,ein in the inbred SZW/l3LN and the C57BL/SK-NZW/lsLN Fi hybrid appeared more I-ariable and again the mean values were generally higher than those for the other inbred and hybrid strains.
Distribution of the hemoprotein concentration in the MC-treated mice was also distinctly blmodal and followed classical Mendelian genetics. 111 those groups having 110 hyclroxylase induction in response to MC, the CO-b;nding cytochrome content was not sigmficantly (p > .05) different from constitutive levels; in mice possessing the 1ICinducible osygenase, the CO-binding pigment levels ranged between 1130 and 1710 pmoles per mg of rnicrosomal protein.
The means of the groups in this latter population are not significantly different from each other but are significantly (p < .Ol) greater than those of cytochrome P450 in the control mice. Fulthermorc, a correlation between the oxygenase induction (Table II) and the rise in cytochrome P450 content (Table  III) aillong kIC-treated mice can be clearly seen. This observation has been investigated further,7 and we found a stoichiometric relationship between the induction by MC and the increase in high-spin iron-containing CO-binding pigment in mouse liver microsoiiies (2). Polycyclic hydrocarbon administration in viva causes about a 2-rnn blue spectral shift in the Soret maximum of the reduced hemoproteinK0 complex from hepate microsomes (24-26). Similarly we found that, this liypsochrorn~c sh:ft was present in each AIC-tre:rted mouse in n-hicli the enzyme was mducible and abseiit 111 earl1 1IC-treated mouse in whic~h the hydroxylase did not respond to MC.
Prom additional studies,' we conclude that an in(.rease 111 high spin ironcontaining cytochrome Pdsa is directly related to th,s blue shift in the absorpt;on peak of the reduced herr~oprote.li-CO complex (2). Stimulation of the cytochrome PdsO content by PB was similar in all groups of inbred and hybrid mice examined and ranged between 1430 and 1950 lmoles per mg of microsomal protein.
The means of the groups within this range are not significantly (p > 7 D. W. Nebert, F. &I. Goujon, and J. E. Cielen, manuscript in preparation. .Ol) different from one another.
The levels of the CO-binding pigment induced by PB were generally, but not significantly (p > .05), greater than those induced by MC. As expected, I'B caused no spect,ral shift in the absorbance maximum of the reduced cytochrome-CO complex.
Genetic Expression of Hydroxylase /?lduction by SLIJethylcholanthrene in Kidney- Table  IV shows the enzyme levels from the kidney of the various control, MC-treated, and PB-treated inbred and hybrid mice. The mean hydroxylase specific activities from control and PB-treated mice ranged from 0.09 to 0.72 and were not significantly (p > .05) different from each other.
It is known that PB is not an effective inducer of aryl hydrocarbon hydroxylase activity in nonhepatic tissues in vivo (5) or in nonhepatic cells in culture (3). In the MC-treated mice the oxidase activity was again distributed as before: in one population the means were not different from constitutive levels, whereas in the other population the means of the specific activities ranged from 12 to 19 and were significantly (p < .Ol) different from the control values.
In mice possessing the MC-inducible hydroxylase, the specific activities in the kidney were highly variable, ranging from 3 to more than 40. Fig. 3 illustrates that if one plots the hydroxylase-specific activit,ies of the liver versus that of the kidney for MC-treated mice from Ahah x ahah genetic crosses, a bimodal distribution is again appreciated. Furthermore, there is a high correlation in the magnitude of enzyme induction between the liver and kidney.
We found similarly high correlations in the magnitude of hydroxylase induction in bowel, lung, and skin. Thus, if the hepatic hydroxylase is highly inducible by MC, so is the enzyme in the kidney, bowel, lung, and skin; if the hel)atic oxygenase does not respond to MC, then hydrosylase activity iu these other tissues is also noninducible by MC. Most likely, the extent of enzyme induct.ion is partially dependent upon how much MC is taken up by each tissue or is related to such factors as a,ge, sex, nutrition, and stress of each individual animal. Genetic Expression OJ" Hydroxylase Induction in Bowel, Lung, and Skin by S-Methylcholanthrene-Tables V and VI show the hydroxylase-specific activities in the bowel and lung of the same groups of inbred and hybrid mice. In the bowel of control and PB-treated mice, the mean values for the constitutive enzyme ranged between 0.31 and 3.0. The higher constituti\-e levels were found in the inbred NZW/BLS or in hybrids involving the KZW/BLN strain, and one of these means (i.e. 3.0) was significantly (p < .05) different from the lowest means in the control groups (e.g. 0.24 and 0.31). MC treatment produced the san1e bimodal distribution, lvith one population not different, from control levels and the other population inducible and highly var.iable. The magnitude of induction by i\IC was 30.fold tjo more than SO-fold; the means of the specific activities ranged ~'IYJIII 31 to 81. 'J'he genetic expression of tlie osygena.~e induction in the lung 1)~. 1 t(> was the same. The mean specific activities in the control qoups, MC-treated mic*e in w1lic.h the enzyme was noninduc*ible, and l'U-treated mice were higher th:u1 that for kidney or bowel, ranging between 3.0 and 7.0. We found that means of the MCinducible hydroxylase specific activities from lung ranged between 13 and 31. Similar studies with mouse skin demonstrated that the identical genetic expression also exists in that tissue: for control animals and MC-treated mice in which the enzyme is genetically noninducible the specific activities were less than 1.5, whereas MC-inducible hydroxylase activities were greater than 7 units per mg of skin homogenate l)roteill.
What cellular mechanisms are involved in the genetic regulation of this hydroxylase induction by polycyclic hydrocarbons? Which steps in the sequence of events occurring during the induction process are genetically expressed in the MC-responsive strain or are genetically suppressed in the MC-nonresponsive strains? From our studies of microsomal enzyme induction (l-14), we can exclude the possibility that changes in aryl hydrocarbon hydroxylase activity are regulated solely by the posttranslational activation or inhibition of pre-existing enzyme protein.
Also, the quantity of active enzyme rather than the kinetics constants appears to be the main factor in determining the amount of polycyclic hydrocarbon hydroxylation. Therefore, all the steps involved in normal gene expression must be considered as possibilities for explaining the genetic difference in the regulation of polycyclic hydrocarbon-inducible hydroxylase activity described in this report.
Hence, possible differences might occur during the (a) uptake and binding of inducer to receptor sites,8 (b) transcription of induction-specific RNh,g (c) nucleorytoplasmic transport of this new RNA species and translation 8 A receptor site for either PB or a polycyclic hydrocarbon is presently hypothetical.
Presumably, these inducers must bind to some cellular macromolecule so as to effect the sequence of into induction-specific protein, or (d) assembly or degradation of the microsomal membrane components.
We found that the enzyme induction by polycyclic hydrocarbons is expressed as an autosomal dominant trait and that when the individual mouse is genetically responsive to MC, induction of the hydroxylase activity is an all-or-none phenomenon occurring in every tissue regularly containing the polycyclic hydrocarbon-inducible enzyme.
To our knowledge this is the first example in mammaliatr genetics in which the induction of enzyme activity is regulat,ed by one chromosome and perhaps by genes at a single locus. Our studies with clones of mouse 3T3 cellslo are also consistent with this concept that genetic control of the hydroxylase induction by