Retinoid X receptor agonist elevation of serum triglycerides in rats by potentiation of retinoic acid receptor agonist induction or by action as single agents.

Hypertriglyceridemia is a major side-effect of retinoid therapy in humans. We previously reported that agonists for the retinoic acid receptors (RARs), but not the retinoid X receptors (RXRs), elevate serum triglycerides in male Fischer rats, and that, surprisingly, the RAR/RXR pan-agonists 9-cis-retinoic acid and AGN 191659 [(E)-5-[2-(5,6,7,8-tetrahydro-3,5,5,8,8-pentamethyl-2-naphthyl)propen-1-yl]-2-thiophenecarboxylic acid] induce 2- to 3-fold higher levels of serum triglycerides than the RAR-selective agonists alone. We have now demonstrated that hypertriglyceridemia induced by an RAR agonist, AGN 190121 [4-[4-(2',6',6'-trimethylcyclohex-1-enyl)-but-1-yne-3-enyl]benzoic acid], is substantially potentiated by the RXR-selective agonists AGN 191701 [(E) 2-[2-(5,6,7,8-tetrahydro-3,5,5,8,8-pentamethyl-2-naphthyl)propen-1-yl]-4-thiophene-carboxylic acid] and AGN 192849 [(3,5,5,8,8,-pentamethyl-5,6,7,8-tetrahydronaphthalen-2-yl) (5 carboxypyrid-2-yl)sulfide] in a dose-dependent manner. RXR-specific retinoids, as previously reported, had no independent effect on serum triglycerides when tested at 24 hr after final dosing, but did elicit a reversible hypertriglyceridemia at 2.5 and 5 hr. This induction of serum triglycerides could not be blocked by the potent RAR-specific antagonist AGN 193109 [4-[(5,6-dihydro-5,5-dimethyl-8-(4-methylphenyl)-2-naphthalenyl)-ethynyl] benzoic acid]. The RXR ligand-induced hypertriglyceridemia was independent of the effect of feeding or fasting. The relative potencies of RXR-specific retinoids for acute triglyceride elevation (AGN 194204 [3,7-dimethyl-6S,7S-methano-7-[1,1,4,4-tetramethyl-1,2,3,4-tetrahydronaphth-7-yl] 2(E),4(E) heptadienoic acid] > AGN 192849 approximately AGN 191701) approximately correlated with potencies in the activation of the RXR receptors. The RAR/RXR pan-agonist effect included >50% inhibition of total heparin-releasable lipase activity in serum, consistent with inhibition of lipase-mediated triglyceride disposal. These data also indicate that RAR and RXR ligands can act synergistically to induce hypertriglyceridemia through distinct mechanisms of action.

Phenylcyclohexene and phenylcyclohexadiene substituted compounds having retinoid antagonist activity.

Retinoids are natural and synthetic analogues of the hormone retinoic acid. Systemic retinoid agonist therapy is usually associated with toxic side effects, such as mucocutaneous toxicity, which may be alleviated by the use of topical retinoid antagonists. We report the synthesis and biological activity of a new series of potent, RAR-specific antagonists substituted with phenylcyclohexene and phenylcyclohexadiene groups.

Synthesis and biological activity of high-affinity retinoic acid receptor antagonists.

This article reports the synthesis and biological activity of new high affinity retinioic acid receptor (RAR) antagonists. The effect of introducing heteroatoms in the bicyclic ring system of the potent dihydronaphthalene RAR antagonist 8, and the variation of the pendant aromatic group on the ability of these compounds to function as RAR antagonists is discussed. The use of binding, transcriptional, and in vivo assays revealed that the 2,2-dimethylthiochromene analogue 59, and the 2,2-dimethylchromene derivative 85, were the most effective in blocking retinoid agonist induced activity.

A new class of RAR subtype selective retinoids: correlation of pharmacological effects with receptor activity.

The synthesis and biological activity of a series of structurally related retinoids with different RAR subtype selectivities are described. These retinoids bind to all three RAR subtypes but in functional transactivation assays, they show RARbeta or RARbeta,gamma selectivity with weak RARalpha activity. The subtype selectivity of these retinoids was found to correlate with their efficacy (ODC inhibition) and toxicity (topical irritation and teratogenicity) profiles. The degree of RARgamma transactivation activity correlates with their topical toxicity and teratogenicity as measured by the inhibition of chondrogenesis. Of the RARbeta selective retinoids reported here, retinoid 12 is the most promising, as it is completely devoid of two common retinoid related toxicities, namely topical irritation and teratogenesis.

A new class of potent RAR antagonists: dihydroanthracenyl, benzochromenyl and benzothiochromenyl retinoids.

The synthesis and biological activity of a novel series of tricyclic retinoic acid receptor antagonists are described. These compounds bind with high affinity to the RARs and are potent antagonists of retinoid function in in vitro and in vivo systems.

Receptor specificity of retinoid-induced epidermal hyperplasia: effect of RXR-selective agonists and correlation with topical irritation.

Retinoid induction of epidermal hyperplasia was investigated in hairless mice with synthetic ligands for the retinoic acid (RAR) and retinoid X (RXR) nuclear receptors. Induction of hyperplasia by all-trans retinoic acid and the RAR-specific retinoids TTNPB, tazarotene and AGN 190121 varied over a wide range (ED50 = 0.2-100 nmol/animal in three daily applications). Potency of induction was not directly correlated to receptor-binding affinity, but specificity of action could be demonstrated by inhibition with the high-affinity antagonist of the RARs, AGN 193109. Although RAR is functionally complexed with RXR in vivo, RXR-selective compounds have only weak potency in induction of hyperplasia. The ED50 value of the RXR-selective AGN 191701 was 600 nmol/animal compared with an ED50 value of 0.2 nmol for the structurally similar RAR-selective TTNPB. SR11237 and SR11217, also RXR-selective, each have an ED50 value of >1000 nmol. Unlike RAR-specific retinoids, RXR-selective retinoids cause only very mild skin flaking at high doses. Relative potencies for cumulative topical irritation (flaking and abrasion) of both RAR and RXR ligands were well correlated with epidermal hyperplasia. These data are consistent with RXR as a silent partner in the RAR-RXR heterodimer in skin.

Mitogenic effect of retinoid X receptor agonists in rat liver.

(E)-2-[2-(5,6,7,8-Tetrahydro-3,5,5,8,8-pentamethyl-2-naphthyl) propen-1-yl]-4-thiophenecarboxylic acid (AGN 191701) and other retinoid X receptor (RXR)-selective agonists were observed to cause hepatomegaly in rats. The purpose of the present study was to understand the biochemical basis of RXR agonist-induced hepatomegaly. Male Fischer rats were implanted s.c. with osmotic pumps containing 5-bromo-2'-deoxyuridine (BrdU) and treated by gavage with 0,60, or 180 mumol/kg/day of AGN 191701 for 3 days. AGN 191701 caused dose-dependent hepatomegaly in the absence of hepatic necrosis and necrosis and increased hepatocyte BrdU labeling index (LI). To determine if AGN 191701-induced hepatic hyperplasia was sustained, rats were treated by gavage with 60 mumol/kg of AGN 191701 for up to 7 days and exposed to BrdU via osmotic pump on days 1-3 or on days 6-8. Hepatocyte L1 and mitotic index were increased only in rats exposed to BrdU on days 1-3, indicating that AGN 191701-induced hepatocyte proliferation was transient. The receptor specificity of this mitogenic effect was tested by co-treating rats for 2 days with various retinoids and BrdU. 2-(5,6,7,8-Tetrahydro-5,5,8,8-tetramethyl-2-naphthyl)-2-(4-carboxylph enyl)-1,3-dioxolane (SR11237), an RXR-selective agonist, and (E)-5-[2-(5,6,7,8-tetrahydro-3,5,5,8-pentamethyl-2-naphthyl)propen -1-yl]-2-thiophenecarboxylic acid (AGN 191659), a retinoic acid receptor (RAR)/RXR pan-agonist, both increased hepatocyte LI. Two RAR-selective agonist, all-trans-retinoic acid and (E)-4-[2-(5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2-naphthyl)propen -1-yl] benzoic acid (TTNPB), did not affect hepatocyte LI. To determine if RXR agonists have biochemical effects in common with a peroxisome proliferator, various endpoints were measured 24 hr after two daily treatments with AGN 191701, SR11237, or clofibrate. While all three compounds induced hepatic acyl CoA oxidase activity, only clofibrate increased hepatic carnitine acyl transferase activity and lowered serum triglycerides. Taken together, these data show that RXR-selective agonists but not RAR-selective agonists cause hepatomegaly accompanied by hepatocyte mitogenesis in rats. The fact that RXR agonist have some biological effects distinct from RAR agonists and clofibrate suggests that RXR-selective agonists may have unique therapeutic applications.

Lack of involvement of retinoic acid receptor alpha in retinoid-induced skin irritation in hairless mice.

It has been proposed that RAR gamma, the major retinoic acid receptor (RAR) subtype in skin, mediates retinoid-induced skin irritation. However, RAR alpha is also found in skin, and its role in retinoid-induced skin irritation has not been tested. In this study, RAR subtype-specific agonists and antagonists were used to test the possible contribution of RAR alpha to retinoid-induced skin irritation. Female hairless mice were treated topically on the dorsal skin for 5 days with various retinoids over a 2-log dose range, and cutaneous toxicity was scored by semiquantitative visual observations of skin flaking and abrasions daily up to 3 days post-treatment. Three RAR alpha-selective agonists were > or = 100-fold less potent as skin irritants than the structurally-related RAR pan-agonist, TTNPB. Skin irritation potency decreased in the following order: TTNPB > > Am580 > AGN 193835 > > 193836 and correlated with RAR beta and/or RAR gamma binding affinity rather than RAR alpha binding affinity. TTNPB-induced skin irritation was blocked in a dose-dependent fashion by co-treatment with the RAR pan-antagonist AGN 193109 but was not blocked by co-treatment with the RAR alpha-specific antagonist AGN 194301. In contrast, skin irritation induced by the RAR alpha-selective agonist AGN 193835 was almost completely blocked by co-treatment with AGN 193644, an RAR beta/gamma-selective antagonist. These data demonstrate that RAR alpha is not significantly involved in mediating retinoid-induced skin irritation in mice and suggest that RAR alpha-selective agonists may have reduced mucocutaneous side effects relative to other retinoids.

Retinoid-induced epiphyseal plate closure in guinea pigs.

Vitamin A and its derivatives (retinoids) have been known to cause premature epiphyseal closure in humans as an unwanted side effect of chronic treatment. The purpose of the present study was to determine if guinea pigs could serve as an animal model of retinoid-induced epiphyseal plate closure, and to utilize this model to study the mechanism. Weanling male Hartley guinea pigs were treated ip via osmotic pump for up to 14 days with vehicle or 0.50 to 5.5 mg/kg/day of the retinoic acid receptor (RAR)-selective agonist AGN 190121. Histopathological examination of the proximal tibia of AGN 190121-treated guinea pigs revealed a dose-dependent disruption of the epiphyseal plate. The natural retinoids all-trans-retinoic acid and 13-cis-retinoic acid also induced epiphyseal plate closure in guinea pigs when administered by ip injection for 10 days. Prominent histological features of retinoid-induced epiphyseal closure included the loss of basophilic staining in the extracellular matrix of epiphyseal plate chondrocytes and the invasion of the epiphyseal plate by osteoclasts. To determine if the epiphyseal closure detected histologically was reversible, guinea pigs were treated for 6 days with the RAR-selective agonist (E)-4[2-(5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2-naphthyl)propen- 1-y1]benzoic acid (TTNPB) or vehicle, and groups of guinea pigs were euthanized on Day 7 or 57. TTNPB but not vehicle treatment caused histological evidence of epiphyseal closure at both time points, and significant bone elongation between Day 7 and Day 57 was detected only in vehicle-treated animals. Epiphyseal closure and other toxic effects of TTNPB were blocked by cotreatment of guinea pigs with a fivefold molar excess of AGN 193109, an RAR antagonist. Taken together, these data demonstrate the utility of the guinea pig as an animal model of retinoid-induced epiphyseal closure and suggest that RAR activation is necessary and sufficient for this activity.

Retinoid-induced hypertriglyceridemia in rats is mediated by retinoic acid receptors.

Retinoids in clinical use today are known to induce hypertriglyceridemia as one of their major side effects. The purpose of the present study was to determine, in an appropriate animal model, if retinoid-induced hypertriglyceridemia is mediated by retinoic acid receptors (RARs) and/or by retinoid X receptors (RXRs). Oral gavage of male Fischer rats with 13-cis-retinoic acid for 6 days caused a rapid and sustained increase in serum triglycerides that was reversible within 4 days posttreatment. In subsequent experiments, rats were treated by gavage once daily for 3 days with various retinoids, and serum triglyceride levels were determined 24 hr after the last treatment without fasting. All-trans- and 13-cis-retinoic acid, which can be converted to both RAR and RXR agonists, and 9-cis-retinoic acid, an RAR/RXR pan-agonist, caused dose-dependent increases in serum triglycerides at doses that did not cause weight loss or mucocutaneous toxicity. Ro 13-6298 and AGN 190121, two RAR-specific agonists, caused dose-dependent increases in serum triglycerides, although Ro 13-6298 only induced hypertriglyceridemia at weight-suppressive doses. Two RXR-selective agonists, LG100268 and AGN 191701, failed to induce hypertriglyceridemia or weight loss up to the highest doses tested. A structural isomer of AGN 190121 that does not activate RARs or RXRs, AGN 190727, did not induce hypertriglyceridemia. Hypertriglyceridemia induced by AGN 190121 was significantly inhibited by cotreatment with an RAR-selective antagonist, AGN 193109. Taken together, these data provide strong evidence that retinoid-induced hypertriglyceridemia is mediated, at least in part, by RARs. These data also suggest that RXR-specific agonists may have reduced potential to induce hypertriglyceridemia relative to RAR-active retinoids.

Cell proliferation and regulation of negative growth factors in mouse liver foci.

Foci of altered hepatocytes are preneoplastic lesions capable of progressing to hepatocellular carcinomas. To characterize the growth of preneoplastic hepatic lesions, size of hepatic foci was analyzed with regard to growth factor regulation and hepatocyte proliferation in focal and non-focal hepatocytes. Twelve-day-old female B6C3F1 mice were initiated with a single dose of the potent mutagen N-nitrosodiethylamine (DEN) (5 mg/kg body weight). Beginning at 6 weeks of age, mice were exposed for 16 weeks to 2038 p.p.m. unleaded gasoline (UG) vapor or 1 p.p.m. ethinyl estradiol (EE) in the diet. Analysis of hepatic foci demonstrated that UG significantly increased, but EE significantly decreased the size of DEN-initiated foci. Hepatic labeling index (LI), as measured by the incorporation of 5-bromo-2'-deoxyuridine, was similar in non-focal hepatocytes at 16 weeks in all groups (0.4-0.8%) and greatly increased in hepatic foci. Hepatocyte LI was significantly increased in DEN/UG foci (29%, n = 41) and significantly decreased in DEN/EE foci (6%, n = 23) relative to DEN/control focal hepatocytes (18%, n = 25). The mean LI of foci correlated with the focal size differences observed in the treatment groups. Immunohistochemical analysis with antibodies directed to the negative growth regulator transforming growth factor-beta 1 (TGF-beta1) demonstrated a consistent decrease of TGF-beta 1 in DEN/Ct and DEN/UG hepatic foci relative to non-lesion hepatocytes. Similar results were seen with mannose 6-phosphate/insulin-like growth factor-II receptor (M6P/IGF-II R), which facilitates activation of latent TGF-beta 1. In contrast, only 50% of DEN/EE foci had decreased levels of TGF-beta 1 and M6P/IGF-II R relative to non-focal hepatocytes. These data suggest that proliferative responses observed in hepatic foci may be correlated with foci size. In contrast, chemically induced proliferative responses in non-focal hepatocytes after subchronic exposure cannot necessarily be used to predict proliferative effects in preneoplastic cell populations. Furthermore, these studies suggest that hepatic foci may occur by M6P/IGF-II R enhancing activation of latent TGF-beta 1 in non-focal hepatocytes but not in the focal hepatocytes, thereby affording focal hepatocytes a selective growth advantage.

Specific antagonist of retinoid toxicity in mice.

AGN 193109 was recently identified as a potent retinoic acid receptor (RAR) antagonist in vitro. The purpose of the present study was to determine if AGN 193109 functions as an RAR antagonist in vivo and thus could prevent and/or treat retinoid toxicity. Female hairless mice were treated topically for 5 consecutive days with the synthetic retinoic acid receptor agonist (E)-4-[2-(5,6,7,8- tetrahydro-5,5,8,8-tetramethyl-2-naphthyl)propen-1-yl]benzoic acid (TTNPB) alone or in the presence of a 1-, 4-, or 16-fold molar excess of AGN 193109. TTNPB caused skin flaking, skin abrasions, and splenomegaly, and these effects were blocked in a dose-dependent fashion by AGN 193109 cotreatment. In the same model, AGN 193109 also decreased topical irritation induced by the natural RAR agonist, all-trans-retinoic acid. To determine if topical AGN 193109 could block toxicity induced by an oral retinoid, mice were treated by gavage with TTNPB (0.75 mumol/kg/day) and topically with 0, 0.3, or 1.2 mumol/kg/day of AGN 193109 for 4 days. TTNPB treatment alone caused cutaneous irritation and weight loss, and these effects were inhibited by AGN 193109 cotreatment. To determine if AGN 193109 could be used to treat preexisting retinoid toxicity, mice were pretreated topically on Days 1-2 with TTNPB (0.72 mumol/kg/day) and then treated topically on Days 3-5 with 0, 1.44, 7.2, or 36.0 mumol/kg of AGN 193109. TTNPB pretreatment caused precipitous weight loss and, in the absence of AGN 193109 intervention, 60% mortality. AGN 193109 treatment at all dose levels significantly accelerated recovery of body weight and prevented death in TTNPB-intoxicated mice. These data demonstrate that AGN 193109 is a potent RAR antagonist and a potential antidote of retinoid intoxication in vivo. In addition to potential clinical applications in the prevention and treatment of retinoid toxicity, AGN 193109 should provide a powerful experimental tool for the elucidation of retinoid biology.

Promotion of hepatic preneoplastic lesions in male B6C3F1 mice by unleaded gasoline.

In previous studies, unleaded gasoline (UG) vapor was found to be a liver tumor promoter and hepatocarcinogen in female mice, but UG was not a hepatocarcinogen in male mice. However, UG vapor had similar transient mitogenic effects in nonlesioned liver of both male and female mice under the conditions of the cancer bioassay. We used an initiation-promotion protocol to determine whether UG vapor acts as a liver tumor promoter in male mice and to examine proliferative effects that may be critical to tumor development. Twelve-day-old male B6C3F1 mice were injected with N-nitrosodiethylamine (DEN; 5 mg/kg, intraperitoneally) or vehicle. Starting at 5-7 weeks of age, mice were exposed by inhalation 6 hr/day, 5 days/week for 16 weeks to 0 or 2046 ppm of PS-6 blend UG. UG treatment caused a significant 2.3-fold increase in the number of macroscopic hepatic masses in DEN-initiated mice, whereas no macroscopic masses were observed in non-initiated mice. Altered hepatic foci (AHF), which were predominantly basophilic in phenotype, were found almost exclusively in DEN-initiated mice. UG treatment significantly increased both the mean volume (threefold) and the volume fraction (twofold) of the AHF without increasing the number of AHF per unit area. UG also induced hepatic pentoxyresorufin-O-dealkylase (PROD) activity, a marker of CYP2B, by more than 12-fold over control with or without DEN cotreatment. To study hepatocyte proliferative effects of UG, we treated mice with 5-bromo-2'-deoxyuridine (BrdU) via osmotic pump for 3 days before necropsy and measured hepatocyte BrdU labeling index (LI) in AHF and nonlesioned liver.(ABSTRACT TRUNCATED AT 250 WORDS)

Identification of hepatic mitogenic and cytochrome P-450-inducing fractions of unleaded gasoline in B6C3F1 mice.

Unleaded gasoline (UG), a complex mixture of over 300 hydrocarbons, induced liver tumors selectively in female mice and exhibited liver tumor promoting activity. UG also induced cell proliferation and cytochrome P-450-related enzyme activities in mouse liver, properties commonly associated with liver tumor promoters. To determine if the mitogenic and/or cytochrome P-450-inducing properties of UG reside in individual fractions of UG, UG was separated into four fractions on the basis of boiling point (BP): fraction 1, BP < 66 degrees C; fraction 2, 66 degrees C < BP < 100 degrees C; fraction 3, 100 degrees C < BP < 132 degrees C; fraction 4, BP > 132 degrees C. Fractions 1 and 2 were combined to form "light UG" (BP < 100 degrees C), and fractions 3 and 4 were combined to form "heavy UG" (BP > 100 degrees C). Female B6C3F1 mice were implanted with osmotic pumps containing 5-bromo-2'-deoxyuridine (BrdU) on d 1, treated by intragastric intubation with corn oil or 3000 mg/kg/d of light, heavy, or whole UG on d 2-4, and euthanized on d 5. Pentoxyresorufin O-dealkylase (PROD) and ethoxyresorufin O-deethylase (EROD) activities were assayed in hepatic microsomes, and hepatocyte BrdU labeling index (LI) was determined in liver sections. Whole UG and heavy UG caused comparable increases in hepatic PROD and EROD activities and the hepatocyte LI. Light UG caused relatively small increases in hepatic PROD and EROD activities and did not increase the hepatocyte LI. When fractions 3 and 4 were tested separately in the above treatment protocol, both fractions strongly induced hepatic PROD and weakly induced hepatic EROD activities. However, only fraction 3 increased the hepatocyte LI. To isolate mitogenic components in fraction 3, equimolar doses of individual chemicals in fraction 3 were tested in the above treatment protocol. Toluene did not increase the hepatocyte LI, whereas 2,2,3-trimethylpentane (TMP), 2,2,4-TMP, and 2,3,4-TMP all dramatically increased the hepatocyte LI. Thus, while the hepatic cytochrome P-450-inducing activity of UG was concentrated in components of UG with BPs > 100 degrees C, this activity apparently resides in UG components with a wide range of BPs. The mitogenic activity of UG, in contrast, was highly concentrated in components of UG with BPs ranging from approximately 100 to 132 degrees C, and quite possibly in specific TMPs.

Investigation of antiestrogenic properties of unleaded gasoline in female mice.

Chronic exposure of female B6C3F1 mice to a high concentration of unleaded gasoline (UG) vapor induced liver tumors and caused uterine changes suggestive of estrogen antagonism. These effects of UG may be related, since estrogens inhibit hepatocarcinogenesis in mice. The purpose of this study was to determine if antiestrogenic properties of UG could be demonstrated in sensitive short-term assays. Competitive binding to estrogen receptors was assayed in vitro in uterine cytosols prepared from ovariectomized (OVEX) mice. UG did not inhibit specific binding of 17 beta-[3H]estradiol (E2) to uterine cytosols. To determine if UG induced estrogen metabolism, hepatocyte suspensions were prepared from female mice treated by intragastric intubation (ig) for 3 days with corn oil (control) or UG (1800 mg/kg/day). In a quantitative in vitro assay, hepatocytes isolated from UG-treated mice converted E2 and 17 alpha-ethinyl estradiol to water soluble metabolites at a three-fold faster rate than control hepatocytes. Dose-response studies confirmed the induction of E2 metabolism by UG doses as low as 600 mg/kg/day. In a 3-day uterotrophic assay, immature female mice cotreated with UG (600 or 1800 mg/kg/day, ig) and E2 (1 microgram/day, sc) had similar relative uterus weights and uterine peroxidase activity as mice cotreated with corn oil and E2. In a modified uterotrophic assay, OVEX mice treated with corn oil or UG (2400 mg/kg/day, ig) on Days 1-4 and cotreated with E2 (4 micrograms/kg/day, sc) on Days 3-4 had similar uterus weights on Day 5. Thus, while ig treatment of mice with UG induced estrogen metabolism in isolated hepatocytes, this induction did not have functional antiestrogenic consequences as measured by uterotrophic assays. These data suggest that the uterine effects caused by chronic exposure of mice to UG vapor may not be due to direct antiestrogenic effects of UG.

Interactive effects of unleaded gasoline and estrogen on liver tumor promotion in female B6C3F1 mice.

We tested whether a concentration of unleaded gasoline (UG) vapor that was selectively hepatocarcinogenic in female mice in a chronic bioassay is antiestrogenic and whether liver tumor promotion by UG is secondary to antiestrogenicity. Twelve-day-old female C57BL/6 x C3H F1 mice (hereafter called B6C3F1) received i.p. injections of N-nitrosodiethylamine (5 mg/kg) or vehicle. Beginning at 5-7 weeks of age, mice were exposed to 0, 292, or 2056 ppm of PS-6 blend UG vapor for 6 h/day, 5 days/week for 16 weeks, 1 ppm ethinyl estradiol (EE2) in the diet, or 2056 ppm UG vapor and 1 ppm EE2 in the diet. Treatment with 2026 ppm UG but not 292 ppm UG increased relative liver weight, the number of macroscopic hepatic neoplasms, and the size and volume fraction of altered hepatic foci in N-nitrosodiethylamine-initiated mice. Treatment with 2056 ppm UG reduced relative uterus, ovary, and pituitary weights but did not change serum 17 beta-estradiol levels, uterine peroxidase activity, or uterine cytosolic estrogen receptor levels. EE2 treatment reduced the number and size of altered hepatic foci in N-nitrosodiethylamine-initiated mice, caused weight loss, anestrus, vaginal keratinization, decreased uterine peroxidase activity, and decreased uterine cytosolic estrogen receptor levels. UG/EE2 co-treatment attenuated the weight loss, anestrus, and vaginal keratinization caused by EE2 treatment alone but dramatically increased the number of macroscopic hepatic neoplasms and the size and volume fraction of altered hepatic foci as compared to UG treatment alone. Thus, in this two-stage model of carcinogenesis (a) 2056 ppm UG had antiestrogenic effects, particularly with respect to pharmacological actions of EE2; (b) 2056 ppm UG but not 292 ppm UG acted as a liver tumor promoter; (c) EE2 inhibited liver tumor promotion; and (d) EE2 strongly potentiated liver tumor promotion by UG. These data demonstrate significant individual and interactive effects of UG vapor and estrogens in liver tumor promotion in female mice.

Promotion of preneoplastic lesions and induction of CYP2B by unleaded gasoline vapor in female B6C3F1 mouse liver.

An initiation-promotion protocol was used to test the hypothesis that unleaded gasoline (UG) vapor acts as a liver tumor promoter in female mice under exposure conditions in which UG was hepatocarcinogenic in a cancer bioassay. Twelve day old female B6C3F1 mice were injected with N-nitrosodiethylamine (DEN, 5 mg/kg, i.p.) or vehicle. Starting at 5-7 weeks of age, mice were exposed by inhalation 6 h/day, 5 days/week for 13 weeks to 0 or 2039 p.p.m. of PS-6 blend UG, the same gasoline blend used in the cancer bioassay. Putative preneoplastic lesions in liver, characterized mainly as basophilic foci in H&E-stained liver sections, were found exclusively in mice treated with DEN. While similar numbers of altered hepatic foci were found in DEN-initiated mice treated with 0 or 2039 p.p.m. UG, UG treatment significantly increased both the mean volume (3.2-fold) and the volume fraction (3.6-fold) of the foci. To determine if UG induced CYP2B, a subfamily of cytochrome P450 commonly induced by liver tumor promoters in rodents, pentoxyresorufin-O-dealkylase (PROD) activity was assayed in hepatic microsomes derived from the above livers. UG vapor increased hepatic PROD activity approximately 8-fold, while increasing cytochrome P450 content only approximately 30%. To ascertain if a more recent blend of UG, API 91-1, would have similar biological effects as PS-6, female B6C3F1 mice were gavaged for 3 days with corn oil or 1800 mg/kg/day PS-6 or API 91-1 blend UG. PS-6 and API 91-1 blend UG induced similar increases in relative liver weight (approximately 25%), PROD activity (approximately 9-fold) and hepatocyte labeling index (approximately 8-fold) relative to controls. These data demonstrate that PS-6 blend UG vapor promotes preneoplastic lesions and induces CYP2B in female mouse liver under exposure conditions in which it causes liver tumors, and suggest that a more recent blend of UG may have similar effects.

Differential binding of chromium(VI) and chromium(III) complexes to salmon sperm nuclei and nuclear DNA and isolated calf thymus DNA.

The binding of CrCl3.6H2O, Cr(NO3)3.9H2O, [Cr(L-His)2] (NO3)3.H2O, [Cr(L-Cys)(L-His)].3.5H2O, [Cr(L-His)(D-Pen)].H2O, Na[Cr(L-Cys)2].2H2O, K2[Cr(GS)2].3H2O, Na2-CrO4.4H2O, and Na2Cr2O7.2H2O to salmon sperm nuclei and nuclear DNA was determined. The Cr(III)-amino acid complexes and Cr(VI) exhibited significantly lower Cr-nuclei and Cr-DNA binding levels relative to the inorganic complexes CrCl3.6H2O and Cr(NO3)3.9H2O. The binding of CrCl3.6H2O, Cr(NO3)3.9H2O and Na2Cr2O7.2H2O to salmon sperm nuclei and nuclear DNA in the presence of rat lung cytosol was determined under the same conditions. For those complexes studied in both buffer and cytosol, the Cr-DNA binding levels for Cr(III) complexes were higher in buffer than in cytosol, while a relatively higher binding level was observed for Cr(VI) in cytosol than in buffer. Slightly lower nuclear protein levels were present in Cr(VI) incubations than in Cr(III) incubations with nuclei both in the presence and the absence of cytosol. The relative binding of CrCl3.6H2O, Cr(NO3)3.9H2O, [Cr(L-His)2](NO)3.H2O, [Cr(L-Cys) (L-His)].3.5H2O, [Cr(L-His)(D-Pen)].H2O, Na[Cr(L-Cys)2].2H2O and Na2CrO4.4H2O to isolated calf thymus DNA in buffer was also determined. Positively charged, labile inorganic Cr(III) complexes, CrCl3.6H2O and Cr(NO3)3.9H2O, exhibited higher binding to DNA than [Cr(L-His) (D-Pen)].H2O, and no binding to DNA was observed with Cr(VI) and the other neutral, positively and negatively charged, inert Cr(III)-amino acid complexes. Although labile aquo chromium(III) complexes are quite reactive with DNA, the reactivity of chromium(III), formed upon intracellular reduction of carcinogenic chromium(VI), toward DNA will be diminished by complexation with cellular proteins, peptides and amino acids.

Ascorbate is the principal reductant of chromium(VI) in rat lung ultrafiltrates and cytosols, and mediates chromium-DNA binding in vitro.

Chromium(VI) reductase activity was measured in ultrafiltrates of rat lung after various pretreatments in vitro at 37 degrees C and pH 7.0. Pretreatment of ultrafiltrates with L-ascorbate oxidase (EC 1.10.3.3), which specifically eliminated ascorbate, blocked approximately 95% of chromium(VI) reductase activity in ultrafiltrates. Preincubation of ultrafiltrates with heat-denatured ascorbate oxidase or the sulfhydryl-blocking agent N-ethylmaleimide (NEM) had no significant effect on Cr(VI) reductase activity. In rat lung cytosols, L-ascorbate oxidase blocked approximately 95% and NEM blocked approximately 15% of Cr(VI) reductase activity. The extent of inhibition of Cr(VI) reductase activity in cytosols by L-ascorbate oxidase was significantly decreased to approximately 75% after addition of 1.0 mM NADPH. When Cr(VI) was incubated with salmon sperm nuclei suspended in rat lung cytosol for 15 min, Cr became bound to nuclear DNA. This Cr-DNA binding was completely inhibited by preincubation of rat lung cytosols with L-ascorbate oxidase and inhibited approximately 60% by preincubation with NEM. Taken together these data suggest that ascorbate and/or ascorbate-dependent factors are the principal reductants of Cr(VI) in both ultrafiltrates and cytosols prepared from rat lung and ascorbate-dependent metabolism of Cr(VI) results in Cr binding to nuclear DNA in vitro. Although sulfhydryl-containing factors and NADPH-dependent factors only make a minor contribution to Cr(VI) reduction in rat lung cytosols, sulfhydryls may be significantly involved in the binding of Cr to nuclear DNA.