- Email: [email protected]
Incomplete correlation of 2,3,7,8-tetrachlorodibenzo-p-dioxin hepatotoxicity with Ah phenotype in mice
TOXICOLOGY
AND
APPLIED
incomplete
J . B . GREIG,*‘~ *MRC
PHARMACOLOGY
74, 17-25 (1984)
Correlation of 2,3,7,8-Tetrachlorodibenzo-p-dioxin Hepatotoxicity with Ah Phenotype in Mice’ J. E. FRANCIS,*
S. J. E. KAY,*
Toxicology Unit and tMRC Experimental Laboratories, Woodmansteme Road,
Received
August
D. P. LOVELL,~,~
Embryology Carshalton.
2, 1983; accepted
AND A. G. SMITH*
and Teratology Unit, Medical Research Surrey SMS 4EF. United Kingdom
January
Council
3, 1984
Incomplete Correlation of 2,3,7,8-Tetrachlorodibenzo-p-dioxin Hepatotoxicity with Ah Phenotype in Mice. GREIG, J. B., FRANCIS, J. E., KAY, S. J. E., JJXELL, D. P., AND SMITH, A. G. (1984). Toxicol. Appl. Pharmacol., 74, 17-25. Pretreatment of male mice of the inbred strains A2G, BALB/c, C57BL/lO, and AKR with iron dextran synergized the action of a single dose of 2,3,7,8-tetmc~orodibenzo-pdioxin (TCDD, 75 &) in causing hepatic porphyria and necrosis 35 days later. There was no effect in DBA/2 mice. Increased porphyrin levels were associated with decreased hepatic activity of uroporphyrinogen decarboxylase. Iron alone had no effect on porphyrin levels or decarboxylase activity. In male BALB/c mice given TCDD alone there was a delay in the onset of porphyria. Female BALB/c, AKR, and AKR X DBA/Z Fr mice were more resistant to the porphyrinogenic effect of TCDD than males. Development of porphyria did not correlate with Ah phenotype of the mice. The inheritance of sensitivity to TCDD in crosses of the AKR and DBA/Z strains, both Ah nonresponsive, was studied by a biometrical genetic analysis. The inheritances of increased porphyrin levels and of increased plasma activity of enzymes indicative of hepatic necrosis were both complex. Segregation of alleles at more than one locus was required to explain the data. A lack of correlation of porphyrins with plasma enzyme levels in the Fz generation suggested that the expression of these traits was determined independently. Genes other than Ah influence the development of TCDD-induced hepatotoxicity in mice
Tukey et al., 1982a,b). Strains in which MC induces AHH, e.g., C57BL/6, are termed Ah responsive4 (Nebert et al., 1982; Poland et al., 1974). In other strains, typically DBA/2, this response is minimal and receptor levels are
In mice, alleles of the Ah locus influence the liver cytosolic levels of a receptor protein capable of binding 3-methylcholanthrene (MC) and preeminently, 2,3,7,8-tetrachlorodibenzop-dioxin (TCDD) (Gasiewicz and Neal, 1982; Hannah et al., 1981; Poland et al., 1976). The receptor acts in the induction of cytochrome PI-450 synthesis and polycyclic aromatic hydrocarbon hydroxylase (AHH) activity (Greenlee and Poland, 1979; Okey et al., 1979;
4 The Ah phenotype of inbred strains of mice can be assessedin vivo and in vitro by several methods. These will not necessarily distinguish between strains which have been shown by breeding tests apparently to differ at the Ah locus. We have therefore avoided the use of allele symbols and follow the usage of Nebert et al. (1982) in categorizing asAh responsive those strains which (a) possess high hepatic levels of the TCDD receptor protein, (b) following treatment with MC or Bnaphthoflavone have highly elevated levels of hepatic cytochrome P&50 and associated AHH-type enzymes, or (c) are sensitive to the ulcerative action of 7, I2-dimethylbenzanthracene on the skin. Ah-nonresponsive mice lack these attributes.
’ Presented in part at the 9th Mammalian Biochemical Genetics Workshop, November 24-25th, 1982, London, UK. ’ To whom correspondence should be addressed. 3 Present address: British Industrial Biological Research Association, Woodmansteme Road, Carshalton, Surrey SM5 4DS, UK. 17
0041-008X/84
$3.00
Copyright 0 1984 by Academic Press. Inc. All rights of reproduction in any form reserved.
18
GREIG ET AL.
undetectable (Okey et al., 1979; Poland et al., 1976). In crosses of these two strains the Ahresponsive trait segregates as if determined by an autosomal, dominant allele (Nebert et al., 1972; Thomas et al., 1972). Data from different crosses have indicated the involvement of other alleles and loci (Robinson et al., 1974; Thomas and Hutton, 1973). TCDD induces AHH activity and elicits several toxic effects in mice with either Ahresponsive or nonresponsive phenotypes. However, in each case it appears that the dose required to produce the effect in an Ah-nonresponsive mouse is approximately lo-fold greater than that needed for a responsive animal. By using crosses and backcrosses of C57BL/6 and DBA/2 mice, it has been shown that sensitivity to TCDD-induced thymic atrophy (Poland and Glover, 1980a) and hepatic porphyria (Jones and Sweeney, 1980), or to induction of skin lesions of hairless (hr/ hr) congenics by a TCDD analog (Knutson and Poland, 1982), all segregate with the Ahresponsive phenotype. Additionally, the strain distribution pattern of the teratogenic effects of TCDD, with one exception (Poland and Glover, 1980a), and immune system disturbances (Vecchi et al., 1983) appear to correlate with the Ah phenotype of the strains. Data from studies of DBA/2 mice given either single (Smith et al., 198 1) or multiple (Jones and Sweeney, 1980) doses of TCDD also suggest that the LD50 in this nonresponsive strain is at least fivefold greater than the values recorded for the C57BL/6 (Jones and Greig, 1975; Vos et al., 1974) and C57BL/lO (Smith et al., 1981) strains. In this paper we show that the Ah phenotype of mice is not the sole factor which determines the hepatotoxicity of TCDD. Additional genetic loci and the sex of the animals are involved. Also, we extend observations on the role of iron status in TCDD toxicity (Jones et al., 1981). METHODS Mice were bred in these laboratories. Subline designations of the inbred strains are A2G/Lac, BALB/
cLac, C57BL/lOScSn//Tox, AKR/Nimr//Tox, and DBA/ ZNimr//Tox. In crosses the maternal genotype is listed first. Tap water and Breeder Diet, No. 3 Expanded,’ were available ad libitum throughout. All the inbred strains have been monitored at various biochemical loci and were found to possess the correct alleles. In addition, the Ah locus was checked by an adaptation of the zoxazolamine paralysis time procedure of Nebert et al. (1982), and the strains were shown to have the correct responsive or nonresponsive phenotype. When iron pretreatment was used, randomized groups of mice aged 36 to 57 days were given a sc injection of either Imferon6 (0.25 ml, 12.5 mg Fe) or an equal volume of the vehicle, aqueous dextran C (200 mg/ml, a gift from Fisons Ltd). Seven days later each animal was given either TCDD (75 &kg, 7.5 &ml in corn oil) or corn oil (10 ml/kg) po. The preparation of the TCDD is described elsewhere (Greig et a/., 1973). The mice were then immediately transferred to a negative-pressure, plastic-film isolator where they were housed in plastic boxes on stainless-steel grid floors over paper-lined trays. Additional safety precautions are described in Smith et al. (198 1). At the appropriate time after TCDD administration mice were weighed and were killed by cervical dislocation or with CO;!. In the latter instance heart blood was collected in heparinized syringes. Livers were removed, washed, and weighed, the median lobes were analyzed for porphyrins and uroporphyrinogen decarboxylase (EC 4.1.1.37) activity (Smith et a/., 198 1) and the left lateral lobes were analyzed for lipids (Jones and Greig, 1975) or total iron (Smith et al.. 198 1). Plasma alanine aminotransferase (ALT, EC 2.6.1.2) and sorbitol dehydrogenase (L-iditol dehydrogenase, SDH, EC 1.I. I. 14) activities were assayed, in low volume cuvettes, with kits.’ SDH was measured at 25°C on the day blood was obtained, ALT was measured at 30°C within 3 days of collection, after storage at 0 to 5°C. The significance of difference between groups was assessedby Student’s f test. The biometrical genetic analysis used standard methods (Mather and Jinks, 1982).
RESULTS Influence of Iron on TCDD-Induced in Ah-Responsive Mice
Porphyria
The hepatic porphyria induced in mice by TCDD and other polychlorinated aromatic compounds is associated with reduced activity s Purchased from Special Diet Services Ltd., Witham, Essex, UK. 6 Iron dextran BP bought from Fisons Limited, Loughborough, UK. ’ Purchased from Sigma London, Poole, Dorset, UK.
TCDD HEPATOTOXICITY
of uroporphyrinogen decarboxylase. It has been shown that iron deficiency protects Ahresponsive mice from the hepatotoxic effects of TCDD (Jones et al., 1981). With the aim of maximizing the porphyrinogenic effects of a single dose of TCDD, mice were pretreated with Imferon. The dose used resulted in a 25fold increase of liver iron content in both male and female C57BL/10 mice (data not presented). Iron had a synergistic action on the development of porphyria, assessedas hepatic
19
AND Ah PHENOTYPE
porphyrin levels 5 weeks after TCDD dosage, in males of the A2G, BALB/c, and C57BL/ 10 Ah-responsive strains (Table 1). The high levels of porphyrins seen in A2G mice given TCDD alone are raised by the iron overload; also, both C57BL/ 10 and BALB/c mice show large increases in hepatic porphyrin levels if given Imferon. The administration of iron alone did not cause any clinically significant elevation of hepatic porphyrin levels (Table 1). Addition-
TABLE 1 EFFECT OF IRON AND TCDD ON HEPATIC PORPHYRINSAND PLASMA ENZYMES OF MALE MICE’ Treatment
Hepatic porphyrins
Plasma enzymes (III/liter)
Ah
Genotype
phenotype b
Iron
TCDD
nmol/g tissue
N
ALT
SDH
N
+ + -
0.67 f 0.01 748 f 60’ 915 f 33’
3 5 5
39.3 + 5.6 117 + 8.3’ 374 + 42’
33.8 IL 4.9 39.3 + 2.9 21.8 +- 3.8d
3 5 5
0.74 + 0.03 0.84 _t 0.08 12.4 + 10 431 + 115d
5 5 10 11
NDf ND 59.6 + 5.3 561 2 2W
ND ND 38.7 + 2.1 158t48’
5 5
0.46 0.64 303 826
0.02 0.06 42’ 75<
5 9 10 11
ND 65.8 f 15 143 f 25’ 1080 f 170’
ND 42.5 + 4.5 169 f 38’ 202+31
4 4 5
A2G A2G A2G
R R R
BALB/c BALB/c BALB/c BALB/c
R R R R
C57BLjlO C57BL/lO C57BLj IO C57BL/IO
R R R R
+ + + + +
AKR AKR AKR
NR NR NR
+ +
DBA/2 DBA/2 DBA/Z
NR NR NR
AKR X DBA/2 F, AKR X DBA/2 F, DBA/Z x AKR F,
1.08 + 0.05 23.1 f 5.0’ 622 f 45’
3 11 10
79.1 + 13 76.1 k 24 341 + 81’
45.7 + 8.9 51.2 f 11 273 k 5gd
3 5 4
+ +
+ + + + + + + +
0.55 AT0.01 0.98 + 0.03c 1.08 _+0.04
5 6 6
30.0 + 1.4 29.3 f 1.3 40.7 k 2.2d
24.0 -+ 1.7 27.8 t 0.9 35.0 _t 2.7’
5 6 6
NR NR
+
+ +
85.2 + 28 219 AC30d
6 13
ND 94.4 + 18
ND 40.6 f 7.0
7
NR
+
+
103 k 42
4
174 + 31
95.4 k 13
4
f f f k
o Mice were pretreated with Imferon or dextran C and dosed with TCDD or corn oil according to the schedule in the Methods section. They were killed for assay of porphyrins and plasma enzymes 35 days after TCDD dosage. Values are X + SE. Significance of difference Between groups is indicated for the folIowing comparisons: the group given iron + TCDD compared with that given TCDD alone; the group given TCDD compared with a group not given TCDD. b Mice are classified as R: Ah responsive (see footnote 4 to the text) or NR: Ah nonresponsive. rp -=c0.001. d.LJ< 0.01. ‘p -=z0.05. ‘Not determined.
20
GREIG ET AL.
ally, neither in the C57BL/lO nor in the BALB/c strain did it decrease the activity of hepatic uroporphyrinogen decarboxylase (Fig. 1). Increases in hepatic porphyrins in TCDDtreated groups correlated with the degree of reduction of uroporphyrinogen decarboxylase activity in the same groups (Fig. 1). The relative lack of response of BALB/c mice to TCDD alone was unexpected, and the cause was investigated. The time course of the development of porphyria in male and female BALB/c mice given the single, standard dose of TCDD was measured. Data in Fig. 2 show that, in this strain, maximum porphyria in the males did not occur until at least 11 weeks after dosing. There was no development of porphyria in females. Interaction of Iron and TCDD in Ah-Nonresponsive Mice Mice of the Ah-nonresponsive DBA/2 strain are highly resistant to the porphyrinogenic ac-
-
iv=
5
5
5
C57B L I 10
5
5
5
10 10
BAtB Ic
3
5
4
AKR
1. Influence of iron and TCDD on levels of hepatic uroporphyrinogen decarboxylase in inbred strains of mice. Male mice were dosed according to the schedule described in the Methods section. Empty histogram cells represent mice receiving dextran C and corn oil, stippled cells those given Imferon and corn oil, hatched cells those given Dextran C and TCDD, and cross-hatched cells those given Imferon and TCDD. Twenty-one days (C57BL/lO) or thirty-five days (BALB/c and AKR) after TCDD dosage, mice were killed and portions of the liver assayed for porphyrins and uroporphyrinogen decarboxylase activity. Values are 1+ SE. Figures in cells are group mean porphyrin levels (nmol/g). Significance of difference between groups is indicated as l p c 0.05, tp i 0.001. The same comparisons are made as in footnote a to Table 1. FIG.
I
2
4 Weeks
6 after
a
10
12
dosing
FIG. 2. Onset of porphyria in TCDD-treated male (m) and female (0) BALB/c mice. Groups of two to six mice were dosed with TCDD (75 &kg, po) and were killed for hepatic porphyrin analysis after the appropriate interval. Values are X + 1 SE. Error bars not shown are smaller than the points. Male and female values overlap up to 4 weeks.
tion of large single (Smith et al., 1981) or weekly (Jones and Sweeney, 1980) doses of TCDD. Even pretreatment with Imferon did not potentiate the effect of TCDD (Table 1). Other strains of Ah-nonresponsive mice might be expected to show a similar lack of response, but surprisingly there was no such finding. AKR males given TCDD alone showed a distinct elevation of porphyrin levels; with iron pretreatment the levels approached those found in the Ah-responsive C57BL/lO strain (Table 1). As with the Ah-responsive mice there was a clear relationship between decrease in uroporphyrinogen decarboxylase activity and the degree of porphyria (Fig. 1). Reciprocal Fr hybrid males of the DBA/2 and ARR Ah-nonresponsive strains became porphyric, like the AKR males. There was no evidence of any maternal effect. Although the DBA/2 and AKR strains are both Ah nonresponsive, a variant nonresponsive allele of this locus has been detected on one occasion in the AKR/N subline (Robinson et al., 1974). If the observed difference in the effects of TCDD in these nonresponsive strains
TCDD HEPATOTOXICITY
were due to variation at the Ah locus then the sensitivity to TCDD should be inherited in a simple Mendelian fashion. Before testing this proposition it was necessary to consider other factors that might influence the action of TCDD. Sex DiJserences in the Development phyria following TCDD Dosage
of Por-
The data in Fig. 2 show that female BALB/c mice, unlike the males, did not respond to TCDD within the 1 l-week experimental period. With C57BL/lO females, either given TCDD alone or TCDD following Imferon pretreatment, the hepatic porphyrin levels were 60% higher than the males (Table 2). This finding differs from previous results (Smith et al., 1981) and may be caused by alteration in animal housing and the period of exposure to TCDD. DBA/2 females were as unaffected by either TCDD or TCDD + Imferon as were males (data not presented). In the AKR strain, as well as in AKR X DBA/ TABLE 2 EFFECT OF IRON AND TCDD ON HEPATIC PORPHYRINSOF FEMALE MICE~
Treatment Genotype AKR AKR AKR X DBA/2F, AKR X DBA/2F, C57BL/‘lO C57BL/lO
Hepatic porphyrins (% of male mean values)
Iron
TCDD
X + SE
N
+ + +
+ + + + + +
3.7 f 0.30b 0.12 + 0.01” 0.68 + O.Ogd
4 5 8 10 6 6
0.50 t- 0.18’
165 f 14’ 163 + lob
’ The mice were pretreated with Imferon or dextran C and dosed with TCDD according to the schedule in the Methods section. They were killed for assayof porphyrins 35 days after TCDD dosage. Absolute values of the equivalent male treatment groups are those of Table 1. b-d Significance of the difference between the male and female groups is indicated as: “p < 0.001, ‘p < 0.005, “p < 0.02.
AND Ah PHENOTYPE
21
2 Fr hybrids, the females responded far less than did the males to the porphyrinogenic action of TCDD (Table 2). For this reason only males were used in subsequent studies of the inheritance of the TCDD-induced porphyria. Other Aspects of TCDD-Induced age
Liver Dam-
Cellular necrosis and lipid accumulation are two additional forms of liver damage found in TCDD-treated mice (Goldstein et al., 1973; Jones et al., I98 1; Jones and Greig, 1975). In AKR and DBA/2 mice, the administration of TCDD following iron pretreatment resulted in less than 70% elevation of liver lipid content 5 weeks later. Since this change occurred in both strains and was far lower than the values found in C57BL/lO mice (data not presented), it was not considered to be a useful marker of liver damage in crosses of the AKR and DBA/2 strains. As indicators of liver necrosis, the plasma levels of ALT and SDH were measured when the mice were killed (Seefeld et al., 1980). In a separate experiment, with male and female A2G mice killed 4 weeks after treatment with TCDD (75 &kg, PO), elevated plasma ALT was found in animals in which there was histological evidence of liver cell necrosis. The histological changes included paracentral necrotic foci, tinctorial changes in the hematoxylin and eosin staining of centrilobular hepatocytes, fatty changes in the midzonal region, and glycogen accumulation in the periportal zone (unpublished data, cf. Jones et al., 1981). There was association of increased plasma ALT levels with increased porphyria among the inbred strains and also, except for the A2G strain, increased SDH was correlated with elevated ALT levels. Accordingly, both these enzymes were assayed in studies of the genetics of the TCDD-induced liver damage. There was no significant difference in plasma ALT levels of C57BL/ 10 male mice given either Imferon or dextran 1 week previously (mean ID/liter ? SE (N): 32.5 1- 2.1 (7) and 33.7 * 1.2 (6), respectively). However,
22
GREIG ET AL.
Imferon potentiated the necrotic action of TCDD, as it did the porphyria (Table 1). Inheritance of TCDD-Induced Liver Damage in Ah-Nonresponsive Mice
The induction of liver damage by TCDD in iron-pretreated mice was studied in parental, F, , F2, and backcross generations from the AKR and DBA/2 strains. The means and SD of these generations for measures of hepatic porphyrins, plasma ALT, and plasma SDH are shown in Table 3. The data were transformed to make the variances of inbred strain and F, hybrid values homogeneous. A biometrical genetic analysis on the generation means showed that the inheritance of the variation in each of the three measures was explained by an additive-dominance model of gene action. The porphyrin data were transformed to log-log values (Fig. 3a). On this scale there was considerable genetic, as opposed to environmental, variation in the crosses between the two inbred strains, with incomplete dom-
inance for porphyric phenotypes. There were discontinuities in the distribution of porphyrin values, but there was no clear evidence of a single gene effect. The values, or transformed values, of plasma ALT and SDH were highly correlated (p < 0.001) in each of the segregating generations, and showed a similar pattern of distribution and mode of inheritance. There was incomplete dominance for low plasma levels of both enzymes after the data had been transformed to logarithms (Figs. 3b and c). The pattern of values in the crosses could not be explained by the segregation of allelic differences at a single gene locus. Hepatic porphyrin levels were significantly correlated with plasma enzyme levels in individual animals of the backcross generations (for ALT: in the AKR backcross N = 40, r = 0.396, p < 0.02; in the DBA/2 backcross N = 42, r = 0.67 1, p < 0.001). There was no significant correlation in the F2 generation (N = 69, r = 0.158, p > 0.1). Similar results were obtained either with the SDH values or with the transformed values (data not presented).
TABLE 3 GENERATION MEANS OF MEASURES OF LIVER DAMAGE IN AKR X DBAR CROSSESGIVEN TCDD” Plasma enzymes (W/liter) Generation
N
AKR DBAI2 FI’ be W W
10 6 17 40 42 69
Hepatic porphyrins (nmol/g tissue) 622.4 1.1 191.4 211.4 28.7 46.7
+ + + + f +
142.9 0.1 112.5 135.1 48.9 81.3
ALT 340.5 40.7 123.4 130.3 44.4 52.3
+ + f + f f
162.3b 5.4 64.4d 84.6 24.3 28.9
SDH 27‘3.2 35.0 60.5 101.1 36.5 33.8
f + + + + +
117.gb 6.7 34.3d 73.4 20.7 15.4
’ All mice were males and received Imferon and TCDD according to the schedule in the Methods section. They were killed for assayof porphyrins and plasma enzymes 35 days after TCDD dosage. Values are X + SD. The values for the parental strains are those of Table 1 but are included here to facilitate comparisons. b Based on four values. ’ In the absence of a maternal effect the F, cross includes AKR X DBA/2 and DBA/2 X AKR Fr hybrids. d Based on 11 values. ’ Backcross to the AKR strain. Includes the progeny of (AKR X DBA/2 F,) X AKR and AKR X (DBA/Z X AKR F,) matings. ‘Backcross to the DBA/2 strain. Includes the progeny of (DBA/Z X ARR F,) X DBA/Z and DBA/2 X (DBA/Z X ARR F,) matings. gIntercross of AKR X DBA/Z F, hybrids.
TCDD HEPATOTOXICITY
F*
23
AND Ah PHENOTYPE
‘“,~9~j&T,l~,~fj
-0.6 -9.4-0.2
0
0.2 0.4 0.6
Hepetic porphyrins
1.2 1.5 1.8 2.1 Plasma ALT
2.4 2.7
log lIU/ll
1.2
1.5 1.8 2.1 2.4 2.7
Plasma SDH
log (IUII)
log log ( nmollg + 1) FIG. 3. Frequency distribution of transformed values of (a) hepatic porphyrins, (b) plasma ALT, and (c) plasma SDH in male offspring of AKR and DBA/2 crosses. Treatment of the mice and descriptions of the matings used to produce the different generations are as in the footnotes to Table 3.
DISCUSSION The synergistic action of an iron overload, given as Imferon, on TCDD-induced porphyria in mice is in agreement with observations of the effect of dietary manipulation of iron status on multiple dose administration of TCDD (Jones et al., 1981). An additional parallel can be found in the influence of Imferon on the porphyria induced by hexachlorobenzene in mice (Smith and Francis, 1983). The effect of iron is independent of the Ah phenotype of the strains and is maximal in BALB/c and AKR mice. Another effect, also apparently independent of Ah phenotype, is the sex difference in the action of TCDD, either with or without iron pretreatment. In three strains the observation of a greater reduction of hepatic uroporphyrinogen decarboxylase activity in groups with more elevated porphyrin levels supports previous results which showed that the decrease of this enzyme activity preceded accumulation of
porphyrins (Smith et al., 1981). Further, since iron alone has no effect on either the enzyme or porphyrin levels, it must be potentiating the action of TCDD at an early step, one leading to the decreased decarboxylase activity. The strain distribution pattern of the porphyric response of inbred mice to TCDD dosage alone was different from that which might have been expected if this response were controlled solely by the Ah locus. Two anomalous strains were found, BALB/c and AKR. However, the minimal effect of TCDD in the BALB/c strain is caused by a delayed response, rather than by a total failure to develop porphyria. More striking was the finding of TCDD-induced cell necrosis and hepatic porphyria in the Ah-nonresponsive AKB strain. The patterns of inheritance of these two traits, induction of porphyria and hepatocellular necrosis, in crosses, backcrosses, and intercrosses of the TCDD-resistant DBA/2 strain with AKB mice were different. Each trait was clearly determined by more than one gene. It
24
GREIG ET AL.
is recognized that the values obtained from the biometrical genetic analysis, between one and two genes for each trait, are probably underestimates (Mather and Jinks, 1982). The work of Jones and Sweeney (1980) showed that in crosses of the Ah-responsive C57BL/6 and Ah-nonresponsive DBA/2 mice the susceptibility to TCDD-induced porphyria segregated with the Ah locus, behaving as if determined by a single gene. Our demonstration of a complex inheritance of susceptibility to TCDD in crosses of the AKR and DBA/2 strains does not confhct with these results. The C57BL/6 and DBA/2 strains may be monomorphic at those loci which determine the different responses of AKR and DBA/2 mice. In such circumstances only the effect of variation at the Ah locus is seen in the C57BL/6 X DBA/2 crosses. Correlations of measures of hepatic porphyria and cell necrosis were significant in the AKR X DBA/2 backcross generations, but, in the F2 animals, with their greater genetic variability, the correlation broke down. This finding suggests that there is independent assortment of the genes controlling expression of these traits. It has been suggested that the porphyria induced by TCDD is dependent on its metabolic activation (Elder and Sheppard, 1982; Sinclair and Granick, 1974), although TCDD met&&es isolated from dog bile are nontoxic (Poiger et al., 1982). However, it is possible that an intermediate metabolite is capable of reacting with uroporphyrinogen decarboxylase. The inheritance of variation of basal (Thomas et al,, 1972) and MC-inducible (Nebert et al., 1982; Robinson et al., 1974) AHH levels in AKR X DBA/2 crosses, like that of TCDD-induced hepatotoxicity, is complex. Also, the MC-inducible activities of several AHH-type enzymes show considerable variability in HS (heterogeneic stock) mice (Lang et al., 1981). Those alleles which render AKR mice susceptible to the hepatotoxic effects of TCDD might thus control the induced levels of that AHH-type enzyme which activates TCDD to a toxic metabolite. An alternative
possibility is that they control the activity of enzymes involved in hepatic lipid peroxidation (Sweeney, 1982; Stohs et al., 1983). This latter explanation would be compatible with the role of iron, which does not potentiate the cutaneous lesions or thymic atrophy caused by TCDD (Jones et al., 1981). Although, at low doses of TCDD, some aspects of its toxicity have been shown to be limited by the absence of the Ah receptor protein (Poland and Glover, 198Ob), we conclude that, at higher doses, the expression of genes other than Ah influences the porphyrinogenic and necrotic effects of TCDD in the liver of mice. REFERENCES ELDER, G. H., AND SHEPPARD, D. M. (1982). Immu-
noreactive uroporphyrinogen decarboxylaseis unchangedin porphyria caused by TCDD and hexachlorobenzene. B&hem. Biophys. Res. Commun. 109, 113120. GASIEWICZ, T. A., AND NEAL, R. A. (1982). The examination and quantitation of tissue cytosolic receptors for 2,3,7,8-tetrachlorodibenzo-p-dioxin using hydroxylapatite. Anal. Biochem. 124, I- 11. GOLDSTEIN, J. A., HICKMAN, P., BERGMAN, H., AND VOS, J. G. (1973). Hepatic porphyria induced by 2,3,7,8tetrachlorodibenzo-p-dioxin in the mouse. Res. Commun. Chem. Pathol. Pharmacol. 6, 9 19-928. GREENLEE, W. F., AND POLAND, A. (1979). Nuclear uptake of 2,3,7,8-tetrachlorodibenzopdioxin in C57BL/ 65 and DBA/ZJ mice. .I. Biol. Chem. 254,9814-9821. GREIG, J. B., JONES,G., BUTLER, W. H., AND BARNES, J. M. (1973). Toxic effectsof 2,3,7,8-tetmchlorodibenzo p-dioxin. Food Cosmet. Toxicol. 11, 585-595. HANNAH, R. R., NEBERT, D. W., AND EISEN,H. J. (1981). Regulatory gene product of the Ah complex. J. Biol. Chem. 256,4584-4590. JONES,G., AND GREIG, J. B. (1975). Pathological changes in the liver of mice given 2,3,7,8-tetrachloroclibenzop-dioxin. Experientia 31, 1315-1317. JONES,K. G., COLE, F. M., AND SWEENEY,G. D. (1981). The role of iron in the toxicity of 2,3,7,8-tetrachlorodibenzo-(p)-dioxin (TCDD). Toxicol. Appl. Pharmacol. 61,74-88.
JONES,K. G., AND SWEENEY,G. D. (1980). Dependence of the porphyrinogenic effect of 2,3,7,8-tetrncbIorodibenzo(p)dioxin upon inheritance of aryl hydrocarbon hydroxylase responsiveness. Toxicol. Appl. Pharmacol. 53, 42-49.
KNUTSON, J. C., AND POLAND, A. (1982). Response of
TCDD HEPATOTOXICITY murine epidermis to 2,3,7,8-tetrachlorodibenzo-p dioxin: Interaction of the Ah and hr loci. Cell 30,225234.
LANG, M. A., GIELEN, J. E., AND NEBERT, D. W. (198 1). Genetic evidence for many unique liver microsomal P450-mediated monooxygenase activities in heterogeneic stock mice. J. BioI. Chem. 256, 12068-12075. MATHER, K., AND JINKS, J. L. (1982). Biometrical Genetics, 3rd Edition. Chapman & Hall Ltd., London. NEBERT, D. W., GOUJON,F. M., AND GIELEN, J. E. (1972). Aryl hydrocarbon hydroxylase induction by polycyclic hydrocarbons: Simple autosomal dominant trait in the mouse. Nature New Biol. 236, 107-l 10. NEBERT, D. W., NEGISHI,M., LANG, M. A., HJELMELAND, L. M., AND EISEN, H. J. (1982). The Ah locus, a multigene family necessary for survival in a chemically adverse environment: Comparison with the immune system. Adv. Genet. 21, 1-51. OKEY, A. B., BONDY, G. P., MASON, M. E., KAHL, G. F., EISEN, H. J., GUENTHNER, T. M., AND NEBERT, D. W. ( 1979). Regulatory gene product of the Ah locus. J. Biol. Chem. 254, 11636-I 1648. POIGER, H., BUSER, H.-R., WEBER, H., ZWEIFEL, U., AND SCHLATTER, CH. (1982). Structure elucidation of mammalian TCDD-metabolites. Experientia 38,484486. POLAND,
A., AND GLOVER, E. (1980a). 2,3,7,8-tetrachlorodibenzopdioxin: Segregation of toxicity with the Ah locus. Mol. Pharmacol. 17, 86-94. POLAND, A., AND GLOVER, E. (1980b). 2,3,7,8-tetrachlorodiheuzo-pdioxin: studies on the mechauism of action. In The Scientific Basis of Toxicity Assessment (H. R. W&hi, ed.), pp. 223-239, Elsevier/North-Holland Biomed. Press, Amsterdam. POLAND, A., GLOVER, E., AND KENDE, A. S. (1976). Stereospecific, high affinity binding of 2,3,7,8-tetrachlorodibenzo-pdioxin by hepatic cytosol. J. Biol. Chem. 251, 4936-4946.
POLAND, A. P., GLOVER, E., ROBINSON, J. R., AND NEBERT, D. W. ( 1974). Genetic expression of aryl hydrocarbon hydroxylase activity. J. Biol. Chem. 249,55995606.
ROBINSON, J. R., CONSIDINE, N., AND NEBERT, D. W. (1974). Genetic expression of aryl hydrocarbon hydroxylase induction. J. Biol. Chem. 249, 5851-5859. SEEFELD,M. D., ALBRECHT, R. M., GILCHRIST, K. W., AND PETERSON,R. E. ( 1980). Blood clearance tests for detecting 2,3,7,8-tetrachlorodibenzo-p-dioxin hepato-
AND Ah PHENOTYPE
25
toxicity in rats and rabbits. Arch. Environ. Toxicol.
Contam.
9, 3 17-327.
SINCLAIR, P. R., AND GRANICK, S. (1974). Uroporphyrin formation induced by chlorinated hydrocarbons (lindane, polychlorinated biphenyls, tetrachIorodihenzo-p dioxin). Requirements for endogenous iron, protein synthesis and drug-metabolizing activity. Biochem. Biophys. Res. Commun. 61, 124-133. SMITH, A. G., AND FRANCIS, J. E. (1983). Synergism of iron and hexachloroheuzene inhibits hepatic uroporphyrinogen decarboxylase in inbred mice. Biochem. J. 214,909-9
13.
SMITH, A. G., FRANCIS,J. E., KAY, S. J. E., AND GREIG, J. B. (1981). Hepatic toxicity and uroporphyrinogen decarboxylase activity following a single dose of 2,3,7,8tetrachlorodibenzo-Hioxin to mice Biochem. Pharmacol. 30,2825-2830. STOHS, S. J., WAN,
M. Q., AND MURRAY, W. J. (1983). Lipid peroxidation as a possible cause of TCDD toxicity. Biochem.
Biophys.
Res. Commun.
111, 854-859.
SWEENEY,G. D. (1982). The heme biosynthetic pathway in the prediction of haloaromatic hydrocarbon toxicity. Adv. Pharmacol. Therapeut. II. 5, 147-159. THOMAS, P. E., AND Hu-I-~oN, J. J. (1973). Genetics of aryl hydrocarbon hydroxylase induction in mice: Additive inheritance in crosses between C3H/HeJ and DBA/ZJ. B&hem. Genet. 8,249-257. THOMAS, P. E., KOURI, R. E., AND HUTTON, J. J. (1972). The genetics of aryl hydrocarbon hydroxylase induction in mice: A single gene difference between C57BL/6J and DBA/ZJ. B&hem. Genet. 6, 157-168. TXJKEY,R. H., NEGISHI, M., AND NEBERT, D. W. (1982a). Quantitation of hepatic cytochrome Pi-450 mRNA with the use of a cloned DNA probe. Mol. Pharmacol. 22, 779-786.
TUKEY, R. H., HANNAH, R. R., NEGISHI, M., NEBERT, D. W., AND EISEN, H. J. (1982b). The Ah locus: Correlation of intranuclear appearance of inducer-receptor complex with induction ofcytochrome Pa450 mRNA. Cell 31, 275-284. VECCHI, A., SIRONI, M., CANEGRATI, M. A., RECCHIA, M., AND GARAI-~INI, S. (1983). Immunosuppressive effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin in strains of mice with different susceptibility to induction of aryl hydrocarbon hydroxylase. Toxicol. Appl. Pharmacol. 68,434-44 1. Vos, J. G., MOORE, J. A., AND ZINKL, J. G. (I 974). Toxicity of 2,3,7,8-tetmchlorodibenzo-p-dioxin (TCDD) in C57BL/6 mice. Toxicol. Appl. Pharmacol. 29,229-24 1,
AND
APPLIED
incomplete
J . B . GREIG,*‘~ *MRC
PHARMACOLOGY
74, 17-25 (1984)
Correlation of 2,3,7,8-Tetrachlorodibenzo-p-dioxin Hepatotoxicity with Ah Phenotype in Mice’ J. E. FRANCIS,*
S. J. E. KAY,*
Toxicology Unit and tMRC Experimental Laboratories, Woodmansteme Road,
Received
August
D. P. LOVELL,~,~
Embryology Carshalton.
2, 1983; accepted
AND A. G. SMITH*
and Teratology Unit, Medical Research Surrey SMS 4EF. United Kingdom
January
Council
3, 1984
Incomplete Correlation of 2,3,7,8-Tetrachlorodibenzo-p-dioxin Hepatotoxicity with Ah Phenotype in Mice. GREIG, J. B., FRANCIS, J. E., KAY, S. J. E., JJXELL, D. P., AND SMITH, A. G. (1984). Toxicol. Appl. Pharmacol., 74, 17-25. Pretreatment of male mice of the inbred strains A2G, BALB/c, C57BL/lO, and AKR with iron dextran synergized the action of a single dose of 2,3,7,8-tetmc~orodibenzo-pdioxin (TCDD, 75 &) in causing hepatic porphyria and necrosis 35 days later. There was no effect in DBA/2 mice. Increased porphyrin levels were associated with decreased hepatic activity of uroporphyrinogen decarboxylase. Iron alone had no effect on porphyrin levels or decarboxylase activity. In male BALB/c mice given TCDD alone there was a delay in the onset of porphyria. Female BALB/c, AKR, and AKR X DBA/Z Fr mice were more resistant to the porphyrinogenic effect of TCDD than males. Development of porphyria did not correlate with Ah phenotype of the mice. The inheritance of sensitivity to TCDD in crosses of the AKR and DBA/Z strains, both Ah nonresponsive, was studied by a biometrical genetic analysis. The inheritances of increased porphyrin levels and of increased plasma activity of enzymes indicative of hepatic necrosis were both complex. Segregation of alleles at more than one locus was required to explain the data. A lack of correlation of porphyrins with plasma enzyme levels in the Fz generation suggested that the expression of these traits was determined independently. Genes other than Ah influence the development of TCDD-induced hepatotoxicity in mice
Tukey et al., 1982a,b). Strains in which MC induces AHH, e.g., C57BL/6, are termed Ah responsive4 (Nebert et al., 1982; Poland et al., 1974). In other strains, typically DBA/2, this response is minimal and receptor levels are
In mice, alleles of the Ah locus influence the liver cytosolic levels of a receptor protein capable of binding 3-methylcholanthrene (MC) and preeminently, 2,3,7,8-tetrachlorodibenzop-dioxin (TCDD) (Gasiewicz and Neal, 1982; Hannah et al., 1981; Poland et al., 1976). The receptor acts in the induction of cytochrome PI-450 synthesis and polycyclic aromatic hydrocarbon hydroxylase (AHH) activity (Greenlee and Poland, 1979; Okey et al., 1979;
4 The Ah phenotype of inbred strains of mice can be assessedin vivo and in vitro by several methods. These will not necessarily distinguish between strains which have been shown by breeding tests apparently to differ at the Ah locus. We have therefore avoided the use of allele symbols and follow the usage of Nebert et al. (1982) in categorizing asAh responsive those strains which (a) possess high hepatic levels of the TCDD receptor protein, (b) following treatment with MC or Bnaphthoflavone have highly elevated levels of hepatic cytochrome P&50 and associated AHH-type enzymes, or (c) are sensitive to the ulcerative action of 7, I2-dimethylbenzanthracene on the skin. Ah-nonresponsive mice lack these attributes.
’ Presented in part at the 9th Mammalian Biochemical Genetics Workshop, November 24-25th, 1982, London, UK. ’ To whom correspondence should be addressed. 3 Present address: British Industrial Biological Research Association, Woodmansteme Road, Carshalton, Surrey SM5 4DS, UK. 17
0041-008X/84
$3.00
Copyright 0 1984 by Academic Press. Inc. All rights of reproduction in any form reserved.
18
GREIG ET AL.
undetectable (Okey et al., 1979; Poland et al., 1976). In crosses of these two strains the Ahresponsive trait segregates as if determined by an autosomal, dominant allele (Nebert et al., 1972; Thomas et al., 1972). Data from different crosses have indicated the involvement of other alleles and loci (Robinson et al., 1974; Thomas and Hutton, 1973). TCDD induces AHH activity and elicits several toxic effects in mice with either Ahresponsive or nonresponsive phenotypes. However, in each case it appears that the dose required to produce the effect in an Ah-nonresponsive mouse is approximately lo-fold greater than that needed for a responsive animal. By using crosses and backcrosses of C57BL/6 and DBA/2 mice, it has been shown that sensitivity to TCDD-induced thymic atrophy (Poland and Glover, 1980a) and hepatic porphyria (Jones and Sweeney, 1980), or to induction of skin lesions of hairless (hr/ hr) congenics by a TCDD analog (Knutson and Poland, 1982), all segregate with the Ahresponsive phenotype. Additionally, the strain distribution pattern of the teratogenic effects of TCDD, with one exception (Poland and Glover, 1980a), and immune system disturbances (Vecchi et al., 1983) appear to correlate with the Ah phenotype of the strains. Data from studies of DBA/2 mice given either single (Smith et al., 198 1) or multiple (Jones and Sweeney, 1980) doses of TCDD also suggest that the LD50 in this nonresponsive strain is at least fivefold greater than the values recorded for the C57BL/6 (Jones and Greig, 1975; Vos et al., 1974) and C57BL/lO (Smith et al., 1981) strains. In this paper we show that the Ah phenotype of mice is not the sole factor which determines the hepatotoxicity of TCDD. Additional genetic loci and the sex of the animals are involved. Also, we extend observations on the role of iron status in TCDD toxicity (Jones et al., 1981). METHODS Mice were bred in these laboratories. Subline designations of the inbred strains are A2G/Lac, BALB/
cLac, C57BL/lOScSn//Tox, AKR/Nimr//Tox, and DBA/ ZNimr//Tox. In crosses the maternal genotype is listed first. Tap water and Breeder Diet, No. 3 Expanded,’ were available ad libitum throughout. All the inbred strains have been monitored at various biochemical loci and were found to possess the correct alleles. In addition, the Ah locus was checked by an adaptation of the zoxazolamine paralysis time procedure of Nebert et al. (1982), and the strains were shown to have the correct responsive or nonresponsive phenotype. When iron pretreatment was used, randomized groups of mice aged 36 to 57 days were given a sc injection of either Imferon6 (0.25 ml, 12.5 mg Fe) or an equal volume of the vehicle, aqueous dextran C (200 mg/ml, a gift from Fisons Ltd). Seven days later each animal was given either TCDD (75 &kg, 7.5 &ml in corn oil) or corn oil (10 ml/kg) po. The preparation of the TCDD is described elsewhere (Greig et a/., 1973). The mice were then immediately transferred to a negative-pressure, plastic-film isolator where they were housed in plastic boxes on stainless-steel grid floors over paper-lined trays. Additional safety precautions are described in Smith et al. (198 1). At the appropriate time after TCDD administration mice were weighed and were killed by cervical dislocation or with CO;!. In the latter instance heart blood was collected in heparinized syringes. Livers were removed, washed, and weighed, the median lobes were analyzed for porphyrins and uroporphyrinogen decarboxylase (EC 4.1.1.37) activity (Smith et a/., 198 1) and the left lateral lobes were analyzed for lipids (Jones and Greig, 1975) or total iron (Smith et al.. 198 1). Plasma alanine aminotransferase (ALT, EC 2.6.1.2) and sorbitol dehydrogenase (L-iditol dehydrogenase, SDH, EC 1.I. I. 14) activities were assayed, in low volume cuvettes, with kits.’ SDH was measured at 25°C on the day blood was obtained, ALT was measured at 30°C within 3 days of collection, after storage at 0 to 5°C. The significance of difference between groups was assessedby Student’s f test. The biometrical genetic analysis used standard methods (Mather and Jinks, 1982).
RESULTS Influence of Iron on TCDD-Induced in Ah-Responsive Mice
Porphyria
The hepatic porphyria induced in mice by TCDD and other polychlorinated aromatic compounds is associated with reduced activity s Purchased from Special Diet Services Ltd., Witham, Essex, UK. 6 Iron dextran BP bought from Fisons Limited, Loughborough, UK. ’ Purchased from Sigma London, Poole, Dorset, UK.
TCDD HEPATOTOXICITY
of uroporphyrinogen decarboxylase. It has been shown that iron deficiency protects Ahresponsive mice from the hepatotoxic effects of TCDD (Jones et al., 1981). With the aim of maximizing the porphyrinogenic effects of a single dose of TCDD, mice were pretreated with Imferon. The dose used resulted in a 25fold increase of liver iron content in both male and female C57BL/10 mice (data not presented). Iron had a synergistic action on the development of porphyria, assessedas hepatic
19
AND Ah PHENOTYPE
porphyrin levels 5 weeks after TCDD dosage, in males of the A2G, BALB/c, and C57BL/ 10 Ah-responsive strains (Table 1). The high levels of porphyrins seen in A2G mice given TCDD alone are raised by the iron overload; also, both C57BL/ 10 and BALB/c mice show large increases in hepatic porphyrin levels if given Imferon. The administration of iron alone did not cause any clinically significant elevation of hepatic porphyrin levels (Table 1). Addition-
TABLE 1 EFFECT OF IRON AND TCDD ON HEPATIC PORPHYRINSAND PLASMA ENZYMES OF MALE MICE’ Treatment
Hepatic porphyrins
Plasma enzymes (III/liter)
Ah
Genotype
phenotype b
Iron
TCDD
nmol/g tissue
N
ALT
SDH
N
+ + -
0.67 f 0.01 748 f 60’ 915 f 33’
3 5 5
39.3 + 5.6 117 + 8.3’ 374 + 42’
33.8 IL 4.9 39.3 + 2.9 21.8 +- 3.8d
3 5 5
0.74 + 0.03 0.84 _t 0.08 12.4 + 10 431 + 115d
5 5 10 11
NDf ND 59.6 + 5.3 561 2 2W
ND ND 38.7 + 2.1 158t48’
5 5
0.46 0.64 303 826
0.02 0.06 42’ 75<
5 9 10 11
ND 65.8 f 15 143 f 25’ 1080 f 170’
ND 42.5 + 4.5 169 f 38’ 202+31
4 4 5
A2G A2G A2G
R R R
BALB/c BALB/c BALB/c BALB/c
R R R R
C57BLjlO C57BL/lO C57BLj IO C57BL/IO
R R R R
+ + + + +
AKR AKR AKR
NR NR NR
+ +
DBA/2 DBA/2 DBA/Z
NR NR NR
AKR X DBA/2 F, AKR X DBA/2 F, DBA/Z x AKR F,
1.08 + 0.05 23.1 f 5.0’ 622 f 45’
3 11 10
79.1 + 13 76.1 k 24 341 + 81’
45.7 + 8.9 51.2 f 11 273 k 5gd
3 5 4
+ +
+ + + + + + + +
0.55 AT0.01 0.98 + 0.03c 1.08 _+0.04
5 6 6
30.0 + 1.4 29.3 f 1.3 40.7 k 2.2d
24.0 -+ 1.7 27.8 t 0.9 35.0 _t 2.7’
5 6 6
NR NR
+
+ +
85.2 + 28 219 AC30d
6 13
ND 94.4 + 18
ND 40.6 f 7.0
7
NR
+
+
103 k 42
4
174 + 31
95.4 k 13
4
f f f k
o Mice were pretreated with Imferon or dextran C and dosed with TCDD or corn oil according to the schedule in the Methods section. They were killed for assay of porphyrins and plasma enzymes 35 days after TCDD dosage. Values are X + SE. Significance of difference Between groups is indicated for the folIowing comparisons: the group given iron + TCDD compared with that given TCDD alone; the group given TCDD compared with a group not given TCDD. b Mice are classified as R: Ah responsive (see footnote 4 to the text) or NR: Ah nonresponsive. rp -=c0.001. d.LJ< 0.01. ‘p -=z0.05. ‘Not determined.
20
GREIG ET AL.
ally, neither in the C57BL/lO nor in the BALB/c strain did it decrease the activity of hepatic uroporphyrinogen decarboxylase (Fig. 1). Increases in hepatic porphyrins in TCDDtreated groups correlated with the degree of reduction of uroporphyrinogen decarboxylase activity in the same groups (Fig. 1). The relative lack of response of BALB/c mice to TCDD alone was unexpected, and the cause was investigated. The time course of the development of porphyria in male and female BALB/c mice given the single, standard dose of TCDD was measured. Data in Fig. 2 show that, in this strain, maximum porphyria in the males did not occur until at least 11 weeks after dosing. There was no development of porphyria in females. Interaction of Iron and TCDD in Ah-Nonresponsive Mice Mice of the Ah-nonresponsive DBA/2 strain are highly resistant to the porphyrinogenic ac-
-
iv=
5
5
5
C57B L I 10
5
5
5
10 10
BAtB Ic
3
5
4
AKR
1. Influence of iron and TCDD on levels of hepatic uroporphyrinogen decarboxylase in inbred strains of mice. Male mice were dosed according to the schedule described in the Methods section. Empty histogram cells represent mice receiving dextran C and corn oil, stippled cells those given Imferon and corn oil, hatched cells those given Dextran C and TCDD, and cross-hatched cells those given Imferon and TCDD. Twenty-one days (C57BL/lO) or thirty-five days (BALB/c and AKR) after TCDD dosage, mice were killed and portions of the liver assayed for porphyrins and uroporphyrinogen decarboxylase activity. Values are 1+ SE. Figures in cells are group mean porphyrin levels (nmol/g). Significance of difference between groups is indicated as l p c 0.05, tp i 0.001. The same comparisons are made as in footnote a to Table 1. FIG.
I
2
4 Weeks
6 after
a
10
12
dosing
FIG. 2. Onset of porphyria in TCDD-treated male (m) and female (0) BALB/c mice. Groups of two to six mice were dosed with TCDD (75 &kg, po) and were killed for hepatic porphyrin analysis after the appropriate interval. Values are X + 1 SE. Error bars not shown are smaller than the points. Male and female values overlap up to 4 weeks.
tion of large single (Smith et al., 1981) or weekly (Jones and Sweeney, 1980) doses of TCDD. Even pretreatment with Imferon did not potentiate the effect of TCDD (Table 1). Other strains of Ah-nonresponsive mice might be expected to show a similar lack of response, but surprisingly there was no such finding. AKR males given TCDD alone showed a distinct elevation of porphyrin levels; with iron pretreatment the levels approached those found in the Ah-responsive C57BL/lO strain (Table 1). As with the Ah-responsive mice there was a clear relationship between decrease in uroporphyrinogen decarboxylase activity and the degree of porphyria (Fig. 1). Reciprocal Fr hybrid males of the DBA/2 and ARR Ah-nonresponsive strains became porphyric, like the AKR males. There was no evidence of any maternal effect. Although the DBA/2 and AKR strains are both Ah nonresponsive, a variant nonresponsive allele of this locus has been detected on one occasion in the AKR/N subline (Robinson et al., 1974). If the observed difference in the effects of TCDD in these nonresponsive strains
TCDD HEPATOTOXICITY
were due to variation at the Ah locus then the sensitivity to TCDD should be inherited in a simple Mendelian fashion. Before testing this proposition it was necessary to consider other factors that might influence the action of TCDD. Sex DiJserences in the Development phyria following TCDD Dosage
of Por-
The data in Fig. 2 show that female BALB/c mice, unlike the males, did not respond to TCDD within the 1 l-week experimental period. With C57BL/lO females, either given TCDD alone or TCDD following Imferon pretreatment, the hepatic porphyrin levels were 60% higher than the males (Table 2). This finding differs from previous results (Smith et al., 1981) and may be caused by alteration in animal housing and the period of exposure to TCDD. DBA/2 females were as unaffected by either TCDD or TCDD + Imferon as were males (data not presented). In the AKR strain, as well as in AKR X DBA/ TABLE 2 EFFECT OF IRON AND TCDD ON HEPATIC PORPHYRINSOF FEMALE MICE~
Treatment Genotype AKR AKR AKR X DBA/2F, AKR X DBA/2F, C57BL/‘lO C57BL/lO
Hepatic porphyrins (% of male mean values)
Iron
TCDD
X + SE
N
+ + +
+ + + + + +
3.7 f 0.30b 0.12 + 0.01” 0.68 + O.Ogd
4 5 8 10 6 6
0.50 t- 0.18’
165 f 14’ 163 + lob
’ The mice were pretreated with Imferon or dextran C and dosed with TCDD according to the schedule in the Methods section. They were killed for assayof porphyrins 35 days after TCDD dosage. Absolute values of the equivalent male treatment groups are those of Table 1. b-d Significance of the difference between the male and female groups is indicated as: “p < 0.001, ‘p < 0.005, “p < 0.02.
AND Ah PHENOTYPE
21
2 Fr hybrids, the females responded far less than did the males to the porphyrinogenic action of TCDD (Table 2). For this reason only males were used in subsequent studies of the inheritance of the TCDD-induced porphyria. Other Aspects of TCDD-Induced age
Liver Dam-
Cellular necrosis and lipid accumulation are two additional forms of liver damage found in TCDD-treated mice (Goldstein et al., 1973; Jones et al., I98 1; Jones and Greig, 1975). In AKR and DBA/2 mice, the administration of TCDD following iron pretreatment resulted in less than 70% elevation of liver lipid content 5 weeks later. Since this change occurred in both strains and was far lower than the values found in C57BL/lO mice (data not presented), it was not considered to be a useful marker of liver damage in crosses of the AKR and DBA/2 strains. As indicators of liver necrosis, the plasma levels of ALT and SDH were measured when the mice were killed (Seefeld et al., 1980). In a separate experiment, with male and female A2G mice killed 4 weeks after treatment with TCDD (75 &kg, PO), elevated plasma ALT was found in animals in which there was histological evidence of liver cell necrosis. The histological changes included paracentral necrotic foci, tinctorial changes in the hematoxylin and eosin staining of centrilobular hepatocytes, fatty changes in the midzonal region, and glycogen accumulation in the periportal zone (unpublished data, cf. Jones et al., 1981). There was association of increased plasma ALT levels with increased porphyria among the inbred strains and also, except for the A2G strain, increased SDH was correlated with elevated ALT levels. Accordingly, both these enzymes were assayed in studies of the genetics of the TCDD-induced liver damage. There was no significant difference in plasma ALT levels of C57BL/ 10 male mice given either Imferon or dextran 1 week previously (mean ID/liter ? SE (N): 32.5 1- 2.1 (7) and 33.7 * 1.2 (6), respectively). However,
22
GREIG ET AL.
Imferon potentiated the necrotic action of TCDD, as it did the porphyria (Table 1). Inheritance of TCDD-Induced Liver Damage in Ah-Nonresponsive Mice
The induction of liver damage by TCDD in iron-pretreated mice was studied in parental, F, , F2, and backcross generations from the AKR and DBA/2 strains. The means and SD of these generations for measures of hepatic porphyrins, plasma ALT, and plasma SDH are shown in Table 3. The data were transformed to make the variances of inbred strain and F, hybrid values homogeneous. A biometrical genetic analysis on the generation means showed that the inheritance of the variation in each of the three measures was explained by an additive-dominance model of gene action. The porphyrin data were transformed to log-log values (Fig. 3a). On this scale there was considerable genetic, as opposed to environmental, variation in the crosses between the two inbred strains, with incomplete dom-
inance for porphyric phenotypes. There were discontinuities in the distribution of porphyrin values, but there was no clear evidence of a single gene effect. The values, or transformed values, of plasma ALT and SDH were highly correlated (p < 0.001) in each of the segregating generations, and showed a similar pattern of distribution and mode of inheritance. There was incomplete dominance for low plasma levels of both enzymes after the data had been transformed to logarithms (Figs. 3b and c). The pattern of values in the crosses could not be explained by the segregation of allelic differences at a single gene locus. Hepatic porphyrin levels were significantly correlated with plasma enzyme levels in individual animals of the backcross generations (for ALT: in the AKR backcross N = 40, r = 0.396, p < 0.02; in the DBA/2 backcross N = 42, r = 0.67 1, p < 0.001). There was no significant correlation in the F2 generation (N = 69, r = 0.158, p > 0.1). Similar results were obtained either with the SDH values or with the transformed values (data not presented).
TABLE 3 GENERATION MEANS OF MEASURES OF LIVER DAMAGE IN AKR X DBAR CROSSESGIVEN TCDD” Plasma enzymes (W/liter) Generation
N
AKR DBAI2 FI’ be W W
10 6 17 40 42 69
Hepatic porphyrins (nmol/g tissue) 622.4 1.1 191.4 211.4 28.7 46.7
+ + + + f +
142.9 0.1 112.5 135.1 48.9 81.3
ALT 340.5 40.7 123.4 130.3 44.4 52.3
+ + f + f f
162.3b 5.4 64.4d 84.6 24.3 28.9
SDH 27‘3.2 35.0 60.5 101.1 36.5 33.8
f + + + + +
117.gb 6.7 34.3d 73.4 20.7 15.4
’ All mice were males and received Imferon and TCDD according to the schedule in the Methods section. They were killed for assayof porphyrins and plasma enzymes 35 days after TCDD dosage. Values are X + SD. The values for the parental strains are those of Table 1 but are included here to facilitate comparisons. b Based on four values. ’ In the absence of a maternal effect the F, cross includes AKR X DBA/2 and DBA/2 X AKR Fr hybrids. d Based on 11 values. ’ Backcross to the AKR strain. Includes the progeny of (AKR X DBA/2 F,) X AKR and AKR X (DBA/Z X AKR F,) matings. ‘Backcross to the DBA/2 strain. Includes the progeny of (DBA/Z X ARR F,) X DBA/Z and DBA/2 X (DBA/Z X ARR F,) matings. gIntercross of AKR X DBA/Z F, hybrids.
TCDD HEPATOTOXICITY
F*
23
AND Ah PHENOTYPE
‘“,~9~j&T,l~,~fj
-0.6 -9.4-0.2
0
0.2 0.4 0.6
Hepetic porphyrins
1.2 1.5 1.8 2.1 Plasma ALT
2.4 2.7
log lIU/ll
1.2
1.5 1.8 2.1 2.4 2.7
Plasma SDH
log (IUII)
log log ( nmollg + 1) FIG. 3. Frequency distribution of transformed values of (a) hepatic porphyrins, (b) plasma ALT, and (c) plasma SDH in male offspring of AKR and DBA/2 crosses. Treatment of the mice and descriptions of the matings used to produce the different generations are as in the footnotes to Table 3.
DISCUSSION The synergistic action of an iron overload, given as Imferon, on TCDD-induced porphyria in mice is in agreement with observations of the effect of dietary manipulation of iron status on multiple dose administration of TCDD (Jones et al., 1981). An additional parallel can be found in the influence of Imferon on the porphyria induced by hexachlorobenzene in mice (Smith and Francis, 1983). The effect of iron is independent of the Ah phenotype of the strains and is maximal in BALB/c and AKR mice. Another effect, also apparently independent of Ah phenotype, is the sex difference in the action of TCDD, either with or without iron pretreatment. In three strains the observation of a greater reduction of hepatic uroporphyrinogen decarboxylase activity in groups with more elevated porphyrin levels supports previous results which showed that the decrease of this enzyme activity preceded accumulation of
porphyrins (Smith et al., 1981). Further, since iron alone has no effect on either the enzyme or porphyrin levels, it must be potentiating the action of TCDD at an early step, one leading to the decreased decarboxylase activity. The strain distribution pattern of the porphyric response of inbred mice to TCDD dosage alone was different from that which might have been expected if this response were controlled solely by the Ah locus. Two anomalous strains were found, BALB/c and AKR. However, the minimal effect of TCDD in the BALB/c strain is caused by a delayed response, rather than by a total failure to develop porphyria. More striking was the finding of TCDD-induced cell necrosis and hepatic porphyria in the Ah-nonresponsive AKB strain. The patterns of inheritance of these two traits, induction of porphyria and hepatocellular necrosis, in crosses, backcrosses, and intercrosses of the TCDD-resistant DBA/2 strain with AKB mice were different. Each trait was clearly determined by more than one gene. It
24
GREIG ET AL.
is recognized that the values obtained from the biometrical genetic analysis, between one and two genes for each trait, are probably underestimates (Mather and Jinks, 1982). The work of Jones and Sweeney (1980) showed that in crosses of the Ah-responsive C57BL/6 and Ah-nonresponsive DBA/2 mice the susceptibility to TCDD-induced porphyria segregated with the Ah locus, behaving as if determined by a single gene. Our demonstration of a complex inheritance of susceptibility to TCDD in crosses of the AKR and DBA/2 strains does not confhct with these results. The C57BL/6 and DBA/2 strains may be monomorphic at those loci which determine the different responses of AKR and DBA/2 mice. In such circumstances only the effect of variation at the Ah locus is seen in the C57BL/6 X DBA/2 crosses. Correlations of measures of hepatic porphyria and cell necrosis were significant in the AKR X DBA/2 backcross generations, but, in the F2 animals, with their greater genetic variability, the correlation broke down. This finding suggests that there is independent assortment of the genes controlling expression of these traits. It has been suggested that the porphyria induced by TCDD is dependent on its metabolic activation (Elder and Sheppard, 1982; Sinclair and Granick, 1974), although TCDD met&&es isolated from dog bile are nontoxic (Poiger et al., 1982). However, it is possible that an intermediate metabolite is capable of reacting with uroporphyrinogen decarboxylase. The inheritance of variation of basal (Thomas et al,, 1972) and MC-inducible (Nebert et al., 1982; Robinson et al., 1974) AHH levels in AKR X DBA/2 crosses, like that of TCDD-induced hepatotoxicity, is complex. Also, the MC-inducible activities of several AHH-type enzymes show considerable variability in HS (heterogeneic stock) mice (Lang et al., 1981). Those alleles which render AKR mice susceptible to the hepatotoxic effects of TCDD might thus control the induced levels of that AHH-type enzyme which activates TCDD to a toxic metabolite. An alternative
possibility is that they control the activity of enzymes involved in hepatic lipid peroxidation (Sweeney, 1982; Stohs et al., 1983). This latter explanation would be compatible with the role of iron, which does not potentiate the cutaneous lesions or thymic atrophy caused by TCDD (Jones et al., 1981). Although, at low doses of TCDD, some aspects of its toxicity have been shown to be limited by the absence of the Ah receptor protein (Poland and Glover, 198Ob), we conclude that, at higher doses, the expression of genes other than Ah influences the porphyrinogenic and necrotic effects of TCDD in the liver of mice. REFERENCES ELDER, G. H., AND SHEPPARD, D. M. (1982). Immu-
noreactive uroporphyrinogen decarboxylaseis unchangedin porphyria caused by TCDD and hexachlorobenzene. B&hem. Biophys. Res. Commun. 109, 113120. GASIEWICZ, T. A., AND NEAL, R. A. (1982). The examination and quantitation of tissue cytosolic receptors for 2,3,7,8-tetrachlorodibenzo-p-dioxin using hydroxylapatite. Anal. Biochem. 124, I- 11. GOLDSTEIN, J. A., HICKMAN, P., BERGMAN, H., AND VOS, J. G. (1973). Hepatic porphyria induced by 2,3,7,8tetrachlorodibenzo-p-dioxin in the mouse. Res. Commun. Chem. Pathol. Pharmacol. 6, 9 19-928. GREENLEE, W. F., AND POLAND, A. (1979). Nuclear uptake of 2,3,7,8-tetrachlorodibenzopdioxin in C57BL/ 65 and DBA/ZJ mice. .I. Biol. Chem. 254,9814-9821. GREIG, J. B., JONES,G., BUTLER, W. H., AND BARNES, J. M. (1973). Toxic effectsof 2,3,7,8-tetmchlorodibenzo p-dioxin. Food Cosmet. Toxicol. 11, 585-595. HANNAH, R. R., NEBERT, D. W., AND EISEN,H. J. (1981). Regulatory gene product of the Ah complex. J. Biol. Chem. 256,4584-4590. JONES,G., AND GREIG, J. B. (1975). Pathological changes in the liver of mice given 2,3,7,8-tetrachloroclibenzop-dioxin. Experientia 31, 1315-1317. JONES,K. G., COLE, F. M., AND SWEENEY,G. D. (1981). The role of iron in the toxicity of 2,3,7,8-tetrachlorodibenzo-(p)-dioxin (TCDD). Toxicol. Appl. Pharmacol. 61,74-88.
JONES,K. G., AND SWEENEY,G. D. (1980). Dependence of the porphyrinogenic effect of 2,3,7,8-tetrncbIorodibenzo(p)dioxin upon inheritance of aryl hydrocarbon hydroxylase responsiveness. Toxicol. Appl. Pharmacol. 53, 42-49.
KNUTSON, J. C., AND POLAND, A. (1982). Response of
TCDD HEPATOTOXICITY murine epidermis to 2,3,7,8-tetrachlorodibenzo-p dioxin: Interaction of the Ah and hr loci. Cell 30,225234.
LANG, M. A., GIELEN, J. E., AND NEBERT, D. W. (198 1). Genetic evidence for many unique liver microsomal P450-mediated monooxygenase activities in heterogeneic stock mice. J. BioI. Chem. 256, 12068-12075. MATHER, K., AND JINKS, J. L. (1982). Biometrical Genetics, 3rd Edition. Chapman & Hall Ltd., London. NEBERT, D. W., GOUJON,F. M., AND GIELEN, J. E. (1972). Aryl hydrocarbon hydroxylase induction by polycyclic hydrocarbons: Simple autosomal dominant trait in the mouse. Nature New Biol. 236, 107-l 10. NEBERT, D. W., NEGISHI,M., LANG, M. A., HJELMELAND, L. M., AND EISEN, H. J. (1982). The Ah locus, a multigene family necessary for survival in a chemically adverse environment: Comparison with the immune system. Adv. Genet. 21, 1-51. OKEY, A. B., BONDY, G. P., MASON, M. E., KAHL, G. F., EISEN, H. J., GUENTHNER, T. M., AND NEBERT, D. W. ( 1979). Regulatory gene product of the Ah locus. J. Biol. Chem. 254, 11636-I 1648. POIGER, H., BUSER, H.-R., WEBER, H., ZWEIFEL, U., AND SCHLATTER, CH. (1982). Structure elucidation of mammalian TCDD-metabolites. Experientia 38,484486. POLAND,
A., AND GLOVER, E. (1980a). 2,3,7,8-tetrachlorodibenzopdioxin: Segregation of toxicity with the Ah locus. Mol. Pharmacol. 17, 86-94. POLAND, A., AND GLOVER, E. (1980b). 2,3,7,8-tetrachlorodiheuzo-pdioxin: studies on the mechauism of action. In The Scientific Basis of Toxicity Assessment (H. R. W&hi, ed.), pp. 223-239, Elsevier/North-Holland Biomed. Press, Amsterdam. POLAND, A., GLOVER, E., AND KENDE, A. S. (1976). Stereospecific, high affinity binding of 2,3,7,8-tetrachlorodibenzo-pdioxin by hepatic cytosol. J. Biol. Chem. 251, 4936-4946.
POLAND, A. P., GLOVER, E., ROBINSON, J. R., AND NEBERT, D. W. ( 1974). Genetic expression of aryl hydrocarbon hydroxylase activity. J. Biol. Chem. 249,55995606.
ROBINSON, J. R., CONSIDINE, N., AND NEBERT, D. W. (1974). Genetic expression of aryl hydrocarbon hydroxylase induction. J. Biol. Chem. 249, 5851-5859. SEEFELD,M. D., ALBRECHT, R. M., GILCHRIST, K. W., AND PETERSON,R. E. ( 1980). Blood clearance tests for detecting 2,3,7,8-tetrachlorodibenzo-p-dioxin hepato-
AND Ah PHENOTYPE
25
toxicity in rats and rabbits. Arch. Environ. Toxicol.
Contam.
9, 3 17-327.
SINCLAIR, P. R., AND GRANICK, S. (1974). Uroporphyrin formation induced by chlorinated hydrocarbons (lindane, polychlorinated biphenyls, tetrachIorodihenzo-p dioxin). Requirements for endogenous iron, protein synthesis and drug-metabolizing activity. Biochem. Biophys. Res. Commun. 61, 124-133. SMITH, A. G., AND FRANCIS, J. E. (1983). Synergism of iron and hexachloroheuzene inhibits hepatic uroporphyrinogen decarboxylase in inbred mice. Biochem. J. 214,909-9
13.
SMITH, A. G., FRANCIS,J. E., KAY, S. J. E., AND GREIG, J. B. (1981). Hepatic toxicity and uroporphyrinogen decarboxylase activity following a single dose of 2,3,7,8tetrachlorodibenzo-Hioxin to mice Biochem. Pharmacol. 30,2825-2830. STOHS, S. J., WAN,
M. Q., AND MURRAY, W. J. (1983). Lipid peroxidation as a possible cause of TCDD toxicity. Biochem.
Biophys.
Res. Commun.
111, 854-859.
SWEENEY,G. D. (1982). The heme biosynthetic pathway in the prediction of haloaromatic hydrocarbon toxicity. Adv. Pharmacol. Therapeut. II. 5, 147-159. THOMAS, P. E., AND Hu-I-~oN, J. J. (1973). Genetics of aryl hydrocarbon hydroxylase induction in mice: Additive inheritance in crosses between C3H/HeJ and DBA/ZJ. B&hem. Genet. 8,249-257. THOMAS, P. E., KOURI, R. E., AND HUTTON, J. J. (1972). The genetics of aryl hydrocarbon hydroxylase induction in mice: A single gene difference between C57BL/6J and DBA/ZJ. B&hem. Genet. 6, 157-168. TXJKEY,R. H., NEGISHI, M., AND NEBERT, D. W. (1982a). Quantitation of hepatic cytochrome Pi-450 mRNA with the use of a cloned DNA probe. Mol. Pharmacol. 22, 779-786.
TUKEY, R. H., HANNAH, R. R., NEGISHI, M., NEBERT, D. W., AND EISEN, H. J. (1982b). The Ah locus: Correlation of intranuclear appearance of inducer-receptor complex with induction ofcytochrome Pa450 mRNA. Cell 31, 275-284. VECCHI, A., SIRONI, M., CANEGRATI, M. A., RECCHIA, M., AND GARAI-~INI, S. (1983). Immunosuppressive effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin in strains of mice with different susceptibility to induction of aryl hydrocarbon hydroxylase. Toxicol. Appl. Pharmacol. 68,434-44 1. Vos, J. G., MOORE, J. A., AND ZINKL, J. G. (I 974). Toxicity of 2,3,7,8-tetmchlorodibenzo-p-dioxin (TCDD) in C57BL/6 mice. Toxicol. Appl. Pharmacol. 29,229-24 1,