*Laboratory of Developmental Pharmacology, National Institute of Child Health and Human. Development, National Institutes of Health, Bethesda, Maryland ...
Copyright 0 1984 by the Genetics Society of America
ARYL HYDROCARBON HYDROXYLASE INDUCTION BY BENZO[a]ANTHRACENE: REGULATORY GENE LOCALIZED TO T H E DISTAL PORTION OF MOUSE CHROMOSOME 17 CATHERINE LEGRAVEREND,*" SIRPA 0.KARENLAMPI,*.~ SANFORD W. BIGELOW,* PETER A. LALLEY? CHRISTINE A. KOZAK? JAMES E. WOMACKP AND DANIEL W. NEBERT*
*Laboratory of Developmental Pharmacology, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland 20205; +Biology Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37830; $Laboratory of Viral Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland 20205; and $Institute on Comparative Medicine, College of Veterinarian Medicine, Texas A U M University, College Station, Texas 77843 Manuscript received February 2, 1984 Accepted March 15, 1984 ABSTRACT
Aryl hydrocarbon (benzo[a]pyrene) hydroxylase inducibility by benzo[a]anthracene was studied in 29 somatic cell hybrid clones, developed by fusing mouse spleen or peritoneal cells from four different inbred strains with hypoxanthine phosphoribosyltransferasedeficient Chinese hamster E36 cells. Karyotype analysis plus 25 markers assigned to 16 autosomes and the X chromosome were examined. In 28 of the 29 clones, the presence or absence of inducibility is associated with the presence or absence, respectively, of mouse chromosome I7.-Liver microsomal aryl hydrocarbon hydroxylase induction by 3-methylcholanthrene or benzo[a]anthracene was assessed in appropriate backcrosses with the Mus musculus molossinus, M. m. castaneus, MOR/Cv, PL/ J, SM/J and DBA/2J inbred strains and in 13 NX8 recombinant inbred lines. Twenty-seven biochemical genetic markers representing all but four autosomes were tested for possible linkage with the hydroxylase inducibility, and no linkage was found. The hepatic Ah receptor was quantitated in 26 BXD recombinant inbred lines; the Ah phenotype did not match exactly any of the more than 70 genes with established strain distribution patterns representing 12 autosomes and at least five unlinked markers.-It is concluded that a major gene controlling aryl hydrocarbon hydroxylase inducibility by benzo[a]anthracene is located on chromosome 17. Because there is no significant linkage with any of three biochemical markers in the upper third of the chromosome, we Abbreviations include: TCDD, 2,3,7,8-tetrachlorodibenzo-pdioxin; ['HITCDD, ['H-l,6]2,3,7,8-tetrachIorodibenzo-pdioxin: AHH, aryl hydrocarbon (benzo[a]pyrene) hydroxylase (EC 1.14.14.1); hprt-, lacking hypoxanthine phosphoribosyltransferase activity; tk-, lacking thymidine kinase activity. The term P-450 designates any or all forms of microsomal cytochrome P-450 (multisubstrate monooxygenases). Mouse PI-450 and Ps-450 are defined as those forms of polycyclic hydrocarbon-induced P-450 with the highest turnover number for induced AHH activity and acetanilide 4-hydroxylase activity, respectively. AHH activity represents benzo[a]pyrene monooxygenation; although quinones. diols and tetraols are also formed, the AHH assay is principally a measure of phenols, the 3-hydroxy- and 9-hydroxybenzo[a]pyrene accounting for more than 80% of the fluorescence mea1978). sured (NEBERT I Present address: EF Lab Oy, P.O. Box 82, 0081 1 , Helsinki 81, Finland. Present address: Laboratory of Biomedical and Environmental Sciences, University of California, Los Angeles, California 90024. Genetics 107: 447-461. July, 1984
448
C. LEGRAVEREND ET AL.
conclude that the inducibility gene is located in the distal 40% of mouse chromosome 17. Whether this trait represents the Ah locus, i.e., the gene encoding the cytosolic Ah receptor, will require further study.
HE murine Ah locus provides an interesting experimental model for studyT ing gene expression (reviewed in EISENet al. 1983; NEBERT al. 1983). The Ah gene encodes a cytosolic receptor that regulates the induction of et
several drug-metabolizing enzymes by foreign chemicals such as benzo[a]anthracene, 3-methylcholanthrene and TCDD. Recently, the sequence of events during the induction process has become much better understood, following the isolation and characterization of the et al. mouse PI-450 cDNA (NEGISHIet al. 1981) and genomic (NAKAMURA 1983) clones. It is now clear, for example, that relatively planar foreign chemicals bind to the cytosolic Ah receptor with an avidity that is in proportion to the potency of the inducer to stimulate PI-450 mRNA (TUKEY, NEBERTand NEGISHI 1981; TUKEY, NEGISHI and NEBERT 1982), the immunoprecipitable membrane-bound P1-450 protein (NEGISHIand NEBERT1979; NEGISHIet al. 1981) and AHH activity (POLAND,GLOVERand KENDE 1976; OKEYet al. 1979). Translocation of the inducer-receptor complex into the nucleus is temperature dependent (OKEYet al. 1980). Appearance of the inducer-receptor complex in the nucleus correlates very well with PI-450 mRNA induction (TUKEY et al. 1982). Maximal increases in an intranuclear high molecular weight precursor mRNA precede by several hours the maximal increases in cytoplasmic P1-450 (23s) mRNA (TUKEY, NEBERTand NEGISHI1981). Increases in PI-450 mRNA and Ps-450 mRNA have been shown to be the direct result of transcriptional activation of both of these genes (GONZALEZ, TUKEY and NEBERT 1984). T h e Ah locus thus represents a complicated interaction of a group of genes activated by an inducer-receptor complex. It would be advantageous to know the chromosomal localization of the Ah gene. For example, the P1-450 and Ps450 genes are located on mouse chromosome 9 (TUKEY, LALLEYand NEBERT 1984); is the major gene that regulates these two P-450 proteins tightly linked? In this report, we have studied 29 somatic cell mouse X hamster hybrid clones, backcross progeny derived from six inbred mouse strains and 39 recombinant inbred lines of two types. From these data, we assign a major gene controlling AHH inducibility by benzo[a]anthracene to the distal portion of mouse chromosome 17. EXPERIMENTAL PROCEDURES
Hybrid clones: Sources of the materials and the techniques employed are included in the following references: KOZAK,NICHOLSand RUDDLE ( 1 975); MINNA, MARSHALL and SHAFFER-BERMAN (1975); FRANCKE et al. (1977); LALLEY, FRANKE and MINNA (1978); LALLEY, MINNA and FRANKE (1978); KOZAK and ROWE(1979); and OKEYet al. (1979). For the 15 EBS hybrids, mouse X hamster somatic cell hybrids segregating mouse chromosomes were prepared by fusing Chinese hamster E36 (hprt-) cells to BALB/cJ mouse spleen cells (MINNA,MARSHALLand SHAFFER-BERMAN 1975). Mouse peritoneal cells were selected for thymidine kinase deficiency (tk-) and fused with
AHH INDUCIBILITY ON MOUSE CHROMOSOME 17
449
the hprt- Chinese hamster E36 cells (KOZAK,NICHOLS and RUDDLE1975; KOZAKand ROWE1979). For making the BM34, BM20, BM3, BMlPH, BM8E, BMJC, BM4, BE7-2-9-4 and BE7-2-9-1 hybrid lines (KOZAKand ROWE1979), BALB/cJ peritoneal cells were used. A/J peritoneal cells were used for preparing the MA5A, MA13, MA151 and MA6D6 hybrids; C3H/HeJ peritoneal cells were used for developing the CHI 1 hybrid line (KOZAK,NICHOLSand RUDDLE1975). Biochemical markers and karotype analysis: Each hybrid clone was analyzed for the expression of 25 enzymes representing linkage groups assigned to 16 of the 19 autosomes and the X chromosome (LALLEY, FRANKE and MINNA 1978; LALLEY, MINNAand FRANKE 1978). The biochemical markers included: Pep-3, Ak-1, Sdh-1, Acp-2, Pgd-1, Pgm-2, Pep-7, Pgm-1, Tpi-I, Ldh-1, Gpi-1, Gr-1, Aprt, Mod-I, Mpi-1, Pep-2, Hk-1, Glk, Acp-1, Es-IO, Sod-1, Glo-1, Pep-1, Got-1 and Hprt. Trypsin/Giemsa et al. 1977; LALLEY, banding was used to identify each of the mouse chromosomes (FRANCKE FRANKE and MINNA 1978). Twenty metaphases per hybrid were scored for their mouse chromosomal complement with the use of standard nomenclature (NESBITI and FRANCKE 1973). Induction of AHH activity in culture: Culture conditions were the same as those previously described for somatic cell hybrids (BENEDICT, NEBERTand THOMPSON 1972), except that the fetal calf serum concentration was 5%. Benzo[a]anthracene (13 @M)was added for 24 or 48 hr during logarithmic growth, and the higher AHH activity of these two time points was recorded as the maximal hydroxylase value. In several experiments, the values at 72 hr were never higher than those at 48 or 24 hr. Inbred mouse studies: Liver microsomal AHH inducibility was determined in the Mus musculus molossinus, M . m. castaneus, MOR/Cv, PL/J and SM/J inbred strains. Each of these strains, known to have the high-affinity Ah receptor (NEBERTet al. 1982), was crossed with DBA/2J, known to have the poor-affinity receptor (NEBERTet al. 1982), and the F1 hybrids were backcrossed to the DBA/2J parent. Progeny from these backcrosses allowed linkage analysis of Ah with 22 biochemical genetic markers on 13 different chromosomes: ldh-1, Pep-3, Amy-1, Gpd-1, Pgm-1, Gus, Mor-I, Es8, Mod-2, Hbb, Es-1, Es-2, Mpi-I, Apk, Es-3, N p - I , Es-IO, Gpt-1, Gdc-1, Glo-1, Neu-1, and Got-1. Between 40 and 108 3-methylcholanthrene- or benzo[a]anthracene-treated offspring of each backcross were examined for each of the markers. Thirteen NX8 recombinant inbred lines (DATTAet al. 1982) were also studied for AHH inducibility by 3-methylcholanthrene. Treatment of each individual mouse consisted of a single intraperitoneal dose [3-methylcholanthrene (200 mg/kg) or benzo[a]anthracene (500 mg/kg)] 48 hr before killing. Liver microsomes were prepared (ATLASet al. 1977). Hydroxylase assay: One unit of AHH activity is defined (NEBERT1978) as that amount of enzyme catalyzing in 1 min at 37" the formation of hydroxylated product causing fluorescence equivalent to that of 1 pmol of 3-hydroxybenzo[a]pyrene recrystallized standard. Specific activity is expressed in units per milligram of cellular protein for the cell cultures and per milligram of microsomal protein for intact mouse liver. Ah receptor assay by sucrose density gradient analysis: All procedures were similar to those described (OKEYet al. 1979). Liver from an individual mouse was homogenized in HEDG buffer (3 ml/g of liver). HEDG buffer is composed of 25 mM N-2-hydroxyethylpiperazine-N'-2-ethane-sulfonic acid, 1.5 mM ethylenediaminetetraacetic acid, 1 mM dithiothreitol and 10% glycerol (v/v), pH 7.6. Dextran-charcoal solutions contained 5mg of charcoal and 0.5 mg of dextran per ml of HEDG buffer. Liver cytosol (105,000 X g X 1 hr supernatant) from individual mice was isolated. One milliliter of cytosol was incubated with 1 nM ['HITCDD in the absence or presence of 100 nM nonlabeled TCDD for 1 hr at 4". Dextran-coated charcoal was used to remove unbound ['HITCDD, following which the sample was centrifuged in a vertical rotor (TSUI and OKEY1981) at 235,000 X g for 2 hr at 2" in a linear 5-20% sucrose density gradient. Twenty-five fractions from the gradient were each counted by scintillation spectrometry. Radioactivity, in those three to six fractions in which the peak could be completely displaced with a 100-fold excess of nonlabeled TCDD, was used to quantitate the receptor in femtomoles per milligram of cytosolic protein. Each receptor determination performed is the average of duplicate centrifuge tubes and never varied by more than 10%. At least three mice of each of 26 BXD recombinant inbred lines and the C57B1/6J and DBA/2J progenitors (The Jackson Laboratory, Bar Harbor, Maine) were examined individually; the coefficient of variance (standard deviation + mean) was always less than 0.11 for each line.
450
C. LEGRAVEREND ET AL.
RESULTS
AHH induction in somatic cell hybrids: Polycyclic-aromatic-inducible AHH activity is regarded as an accurate biochemical marker for determining the presence of the Ah receptor. Cell lines having receptor do not necessarily express inducible hydroxylase activity (OKEY et al. 1980; LEGRAVEREND et al. 1982). Therefore, we felt that AHH inducibility rather than Ah receptor levels is the better determinant of Ah phenotype in cultured cells. Benzo[a]anthracene was used as the inducer in all cell culture studies simply because of past history with its successful use in this laboratory as a potent inducer. We first studied somatic cell hybrids formed by the fusion of BALB/cJ spleen cells with Chinese hamster E36 cells (Table 1). The presence or absence of mouse chromosome 17 correlated with the presence or absence of AHH inducibility by benzo[a]anthracene in all 15 hybrids studied. Whenever chromosome 17 was present, the hydroxylase activity in the clone was inducible. If the hybrid clone had lost chromosome 17, AHH activity was not inducible. There were no exceptions. These data strongly suggest that AHH inducibility by benzo[a]anthracene is controlled by a gene on mouse chromosome 17. No other chromosome, or combination of chromosomes, appeared to account for the variability in AHH induction among the inducible clones. Several chromosomes were discordant in only three or four of the 15 hybrids studied, however, and we wondered whether any of these chromosomes contributed to AHH inducibility. T o confirm the chromosomal assignment, therefore, we also studied mouse X hamster somatic cell hybrids formed by the fusion of peritoneal cells derived from three different inbred mouse strains with Chinese hamster E36 cells. In these hybrids (Table 2), AHH inducibility also segregated with chromosome 17. There was one discordant clone, CH 1 1. T h e chromosomal analysis of CH11 showed that a relatively low percentage of cells (31%) had mouse chromosome 17. At the time of typing for AHH inducibility, a chromosome 17 marker (Glo-I) exhibited barely detectable mouse activity. Thus, this clone may represent a false discrepancy in which the frequency of chromosome 17 was less than the threshold level necessary for detectable AHH inducibility. Our case is substantially strengthened by the fact that one AHH-inducible hybrid, BE7-2-9-4, retained only mouse chromosome 17. Moreover, the several chromosomes showing less than 30% discordance in Table 1 exhibited discordancy rates of more than 35% in Table 2. These data further confirm the assignment of a major gene regulating AHH inducibility by benzo[a]anthracene to mouse chromosome 17. Examination of all of the somatic cell hybrid data (Tables 1 and 2) reveals that more than twice as many hybrids exhibit AHH inducibility than those that do not. T h e reason for this distribution is not clear, but this distribution parallels that seen for retention of mouse chromosome 17. AHH induction in the intact mouse: Liver microsomal AHH activity is highly inducible in M. m. molossinus, M . m. castancus, MOR/Cv, PL/J and SM/J and not inducible in DBA/2J mice. Each of the five responsive strains was crossed with DBA/2J, and the female F1 mice were backcrossed with DBA/2J males. Between 40 and 108 3-methylcholanthrene- or benzo[a]anthracene-treated
AHH INDUCIBILITY ON MOUSE CHROMOSOME I7
451
TABLE 1
Association between chromosomal retention and AHH inducibility by benro[a] anthracene in somatic cell hybrid clones formed by fusing BALB1.J spleen cells with Chinese hamster E36 cells Chromosomal retention/AHH inducibility Concordance Mouse chromosome
I
2 3 4 5 6
7 8 9 IO I1 12 13 14 I5 16 17 18 19 X
Discordance
+/+
-/-
+/-
-/+
5 9 5 4 5 5 9 6
2 1 2 5 5 5 1 5 5 2 5 2 2 3 1 2 5 4 5 0
3 3 3 0 0
3 0 3 3 2 5 3
5 1 3 6 4 4 1 3 3 5 10 2 2 4 0 3
53 29 46 40 29 29 29 21 20 53 67 33 33 40 33
0
0
1 0 5
5 4 0
0 43 27 33
7 5 0
8 8 6 9 7 10 4 6 10
0
3 0 0
% discordance
40
The 15 hybrids used in this study include EBS-1, -2, -3, -4, -5, -9, -10, -11, -15, -17, -18, -51, -63, -71 and -74. All analyses were carried o u t on parallel cultures of each hybrid clone so that the data on AHH inducibility by benzo[a]anthracene, karyotype analysis and gene marker determinations could be correlated. BALB/cJ spleen cells had maximal control and benzo[a]anthracenetreated AHH specific activities of 1.1 and 12.1, respectively. The parent E36 line had maximal control and benzo[a]anthracene-treatedAHH levels of 0.25 and 0.42, respectively. All hybrid lines considered “noninducible” exhibited maximal AHH values between 0.17 and 2.2 after benzo[a] anthracene treatment; all those considered “inducible” exhibited maximal AHH levels between 7.1 and 68 after benzo[a]anthracene treatment. Each assay was repeated a minimum of three times; coefficients of variance (standard deviation f mean) were always less than 0.22. None of the AHH values of inducible clones overlapped with any of those of the noninducible clones. Variation in AHH inducibility among various cell lines (BENEDICT et al. 1973) and among clones derived from et al. 1982) has been commonly seen; the reason for such variability one cell line (LEGRAVEREND is not known.
progeny were studied for each of the six backcrosses. AHH inducibility was compared with biochemical markers on 13 autosomes. Alleles at the Ah locus assorted independently of alleles at each of the 22 other loci examined. With regard to two chromosome 17 markers, the recombinants were 30.6 f 7.1% and 32.5 2 7.4% (means f SE), respectively, for Glo-1 among 49 benzo[a] anthracene-treated progeny of the (PL X DBA)FI X DBA/2J backcross, and for Neu-1 among 40 3-methylcholanthrene-treatedprogeny of the (SM X
452
C. LEGRAVEREND E T AL.
TABLE 2 Association between chromosomal retention and AHH inducibility by benzo[a] anthracene in hybrid clones formed by fusing mouse peritoneal cells with Chinese hamster E36 cells Chromosomal retention1AHH inducibility Concordance
Mouse chromosome
+I+
I
8
2 3 4 5 6
7 2 4 4 5
7
8
8 9
4 5 4 0 7 6 5
IO I1 12 13 14 15 16
17 18 19
X
8
6 10 6
6 6
Discordance
-13 1 3 2 3 2 1 3 3
+I-
-I+
% ’ discordance
1
2
21
3 1
3 8 6 6 5 2 6 5 6 10 3 4
5 2 4
43 64 57 50 50 36 43 43 57 71 36 43 43 43 43
0 4 4 4
36 43 43
2 1
2 3 1 1
2
2
4
0
2
2
2 3 0 2 3 3 2 2
2 1 4 2 1“ 1
2 2
7.1
Karyotyping and testing with biochemical markers, similar to that described for Table 1, were carried out two to five cell passages before testing for AHH inducibility by benzo[a]anthracene. BALB/cJ peritoneal (primary) cells had average control and maximal benzo[a]anthracene-treated AHH specific activities of 1.4 and 16, respectively. The parent E36 line had maximal control and benzo[a]anthracene-treated AHH levels of 0.39 and 0.41, respectively. All hybrid lines considered “noninducible” possessed maximal AHH values between 0.34 and 0.47; all those considered “inducible” possessed maximal AHH levels between 2.1 and 6.6. Each assay were repeated a minimum of two times; coefficients of variance (standard deviation f mean) were always less than 0.18. None of the AHH values of inducible clones overlapped with any of those of the noninducible clones. The only discordant hybrid line is CHI 1 . Of interest, CHI 1 was derived from the C3H/HeJ inbred mouse strain, which has by far the lowest AHH inducibility, when compared with BALB/ CJ or A/J inbred strains (NEBERT et al. 1982).
DBA)FI X DBA/2J backcross. All other recombinations ranged between 37.5 and 62.5%. Of the 13 NX8 recombinant inbred lines (DATTAet al. 1982), ten were found to carry the Ahd allele of NZB and three carried the Ahb allele of C58. No significant linkage was uncovered in the strain distribution patterns of these lines between the Ah and 11 other genes on ten different chromosomes: Pep3, Mup-1, Gpd-1, Pgm-1, IgK, Hbb, Lap-1, Igh-C, Es-10, Gpt-1 and H-2. Ah receptor levels in BXD lines: In addition to liver AHH induction by benzo[a]anthracene or 3-methylcholanthrene, another way to determine the Ah phenotype in the intact animal is to measure the quantity of hepatic cyto-
AHH INDUCIBILITY ON MOUSE CHROMOSOME I 7
453
1.5 1.o I
0
I
0 F
X
0.5
E a
U
I
z
0
0
0
1.o
F
a a
9.0fmollma
BXD-5
LL
a w 0,
0.5
a z
3
0 m
a a
0 F
0 1.o
I
YL
0.5
0
10 15 20 5 FRACTION NUMBER
5
FIGURE1.-Representative examples of sucrose gradient analysis of the cytosolic Ah receptor in three of 26 BXD recombinant inbred lines examined. Cytosolic fractions (1 mg of protein/ml) from 5-week-old mice were incubated with I nM ['HITCDD in the absence or presence of 100 nM nonlabeled TCDD. Following dextran-charcoal treatment, the material was placed on 5-20% sucrose density gradients, centrifuged and fractionated as described in EXPERIMENTAL PROCEDURES. The amount of saturable receptor, i.e., that in which the radiolabel can be displaced by a 100-fold excess of nonlabeled TCDD, is illustrated as a stippled area and can be converted to femtomoles of receptor per milligram of cytosolic protein. To avoid clutter, the loss of the peak (when 1 nM ['HITCDD was incubated with 100 nM of nonlabeled TCDD) is not shown. BXD-25 and BXD-5 represents the B phenotype, and BXD-20 represents the D phenotype.
solic Ah receptor (Figure 1). The Ahb phenotype is known to vary between about 6 and 90 fmol of receptor per mg of cytosolic protein; the Ahd phenotype, always
