Levels of endogenous anionic brain metabolites of dopamine and serotonin increase when elevated levels of organic anions, such as 2,4-D, are present in the serum (Kim et al., 1988).

No other studies pertaining to the absorption, distribution, metabolism, and excretion of 2,4-D were identified.

Mechanism of Action

The mechanisms underlying the toxic effects of 2,4-D may involve alterations in mitochondrial bioenergetics, effects on the metabolism of xenobiotics, and/or inhibition of protein synthesis. Mechanistic studies investigating these effects are summarized below.

Jover et al. (1994) completed an extensive series of in vitro toxicity studies in which the toxic effects of several chemicals, including 2,4-D, were examined in human and rat cultured hepatocytes and in two established cell lines (HepG2 and 3T3). 2,4-D elicited a basal cytotoxic effect at high concentrations but was not considered to be particularly hepatotoxic or to exert species-specific toxicity. Palmeira et al. (1994a) reported that cytotoxicity involves alterations in mitochondrial bioenergetics. Concentration-dependent decreases in mitochondrial membrane potential and in repolarization rate were observed in isolated mitochondria. Tripathy et al. (1993) reported that 2,4-D is genotoxic in somatic and germ-line cells of Drosophila at high concentrations. The relevance of the findings in Drosophila to humans is uncertain.

2,4-D appears to modulate hepatocyte function and to influence biochemical pathways involved in drug metabolism. Palmeira et al. (1994a) reported that 2,4-D induces cell death in isolated rat hepatocytes by decreasing cellular glutathione and depleting adenine and pyridine nucleotide contents. These results are in opposition to those reported by Evangelista et al. (1993), in which fertilized hen eggs were topically treated with 3.1 mg of the 2,4-D butyl ester before starting incubation. The microsomal and cytosolic glutathione S-transferase activities remained unchanged. No significant change in reduced glutathione content between control and treated livers was observed. However, the catalase activity doubled and the glucose-6-phosphatase activity decreased by 46 percent, suggesting that the ester but not the parent compound may have effects on the metabolism of xenobiotics. Despite different effects on cellular glutathione levels, studies conducted by Palmeira et al. (1994a,b) and Evangelista et al. (1993) suggest that 2,4-D appears to modulate hepatocyte function and to influence the biochemical pathways involved in the metabolism of xenobiotics.

A significant decrease in total lipids, especially phospholipid content, was observed. In vivo studies did not reveal any changes in glutathione-S-transferase activity, although in vitro a decrease in enzymatic activity was observed. Catalase levels increased two-fold, while glucose-6-phosphatase activity decreased 46 percent. Although the relevance of the findings in chickens to humans is



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 47
--> Levels of endogenous anionic brain metabolites of dopamine and serotonin increase when elevated levels of organic anions, such as 2,4-D, are present in the serum (Kim et al., 1988). No other studies pertaining to the absorption, distribution, metabolism, and excretion of 2,4-D were identified. Mechanism of Action The mechanisms underlying the toxic effects of 2,4-D may involve alterations in mitochondrial bioenergetics, effects on the metabolism of xenobiotics, and/or inhibition of protein synthesis. Mechanistic studies investigating these effects are summarized below. Jover et al. (1994) completed an extensive series of in vitro toxicity studies in which the toxic effects of several chemicals, including 2,4-D, were examined in human and rat cultured hepatocytes and in two established cell lines (HepG2 and 3T3). 2,4-D elicited a basal cytotoxic effect at high concentrations but was not considered to be particularly hepatotoxic or to exert species-specific toxicity. Palmeira et al. (1994a) reported that cytotoxicity involves alterations in mitochondrial bioenergetics. Concentration-dependent decreases in mitochondrial membrane potential and in repolarization rate were observed in isolated mitochondria. Tripathy et al. (1993) reported that 2,4-D is genotoxic in somatic and germ-line cells of Drosophila at high concentrations. The relevance of the findings in Drosophila to humans is uncertain. 2,4-D appears to modulate hepatocyte function and to influence biochemical pathways involved in drug metabolism. Palmeira et al. (1994a) reported that 2,4-D induces cell death in isolated rat hepatocytes by decreasing cellular glutathione and depleting adenine and pyridine nucleotide contents. These results are in opposition to those reported by Evangelista et al. (1993), in which fertilized hen eggs were topically treated with 3.1 mg of the 2,4-D butyl ester before starting incubation. The microsomal and cytosolic glutathione S-transferase activities remained unchanged. No significant change in reduced glutathione content between control and treated livers was observed. However, the catalase activity doubled and the glucose-6-phosphatase activity decreased by 46 percent, suggesting that the ester but not the parent compound may have effects on the metabolism of xenobiotics. Despite different effects on cellular glutathione levels, studies conducted by Palmeira et al. (1994a,b) and Evangelista et al. (1993) suggest that 2,4-D appears to modulate hepatocyte function and to influence the biochemical pathways involved in the metabolism of xenobiotics. A significant decrease in total lipids, especially phospholipid content, was observed. In vivo studies did not reveal any changes in glutathione-S-transferase activity, although in vitro a decrease in enzymatic activity was observed. Catalase levels increased two-fold, while glucose-6-phosphatase activity decreased 46 percent. Although the relevance of the findings in chickens to humans is

OCR for page 47
--> unknown, these data provide additional support that 2,4-D may influence the biochemical pathways involved in the metabolism of xenobiotics. The effect of 2,4-D on the in vitro synthesis of proteins was studied in Chinese hamster ovary cells by Rivarola et al. (1992), who reported marked inhibition of protein synthesis in cells treated with 1 Mm of 2,4-D for 24 hours. Interestingly, this effect could be reversed by the addition of 0.1 Mm of putrescine, spermidine, or spermine, suggesting that alterations in polyamine metabolism mediate the ability of 2,4-D to interfere with protein synthesis. In a recent study, fertilized hen eggs were externally treated with a single 3.1 mg of 2,4-D (de Moro et al., 1993). The herbicide was shown to induce hypomyelination even before the period of active myelination. The DNA content in brain was increased from the fourteenth embryonic day to the first day of hatching. It is not clear if similar changes occur in mammalian species. Disease Outcomes and Mechanisms of Toxicity In this section, we summarize studies that investigated the toxic effects of 2,4-D. The mechanism of toxicity, if known, is also explained. Carcinogenicity Data prior to 1993 suggest that, in general, 2,4-D produced negative results in carcinogenicity bioassays. In support of this, Edwards et al. (1993) found no association between 2,4-D exposure and mutation of C-N-ras in the dog, suggesting that oncogenic activation of this gene does not occur. Neurotoxicity Case reports of human poisonings and studies in cats and dogs administered high doses of 2,4-D have demonstrated several central nervous system effects, including general sedation, tenseness, loss of righting reflex, motor incoordination, and coma. The animal studies suggest that the primary site of action is the cerebral cortex or the reticular formation (Dési et al., 1962a,b; Arnold et al., 1991). The acute effects of 2,4-D on the central nervous system were recently studied in male Wistar rats (Oliveira and Palermo-Neto, 1993). Behavioral, neuroanatomical, and neurochemical studies were performed. The rats were given single oral doses of 2,4-D ranging from 10 mg/kg of body weight to 300 mg/kg. These doses were chosen because they were high enough to induce sedation and impairment of motor functions but were lower than the calculated LD50 for Wistar rats (945 mg/kg). Single doses of 2,4-D were able to decrease rearing frequencies, decrease locomotion, and increase the deviation of immobility in an open-field test. The neuroanatomical and neurochemical results suggested that 2,4-D modified the functional activities of serotonergic systems within the central nervous system.

OCR for page 47
--> Other Toxicities No other studies were identified that implicated other health outcomes in laboratory animals. Toxicity Profile Update of 2,4,5-T Summary Only one toxicokinetic study and one developmental toxicity study were published on 2,4,5-T during the period 1992-95. Evidence was presented suggesting that 2,4,5-T gains access to the central nervous system and that the uptake process is energy-dependent. The ability of this herbicide to interfere with calcium homeostasis in vitro was also documented and linked to the teratogenic effects of 2,4,5-T on the early development of sea urchin eggs. The relevance of this finding to humans is not known. Toxicokinetics Kim and Pritchard (1993) examined the transport of 2,4,5-T across the blood–cerebrospinal fluid barrier using the isolated choroid plexus of the adult rabbit in vitro and ventriculocisternal perfusion in vivo. In vitro transport was effective at tissue concentrations 20 times those found in the medium after only five min of incubation with 1 µM 2,4,5-T. Uptake was energy-dependent and inhibited by ouabain, phloridzin, and several organic anions, suggesting that 2,4,5-T is a suitable substrate for the organic anion transport system of the rabbit choroid plexus. These data suggest that 2,4,5-T can gain access to the central nervous system. No other studies were identified pertaining to the absorption, distribution, metabolism, and excretion of 2,4,5-T. Disease Outcomes and Mechanisms of Toxicity In this section, we summarize studies that investigate the toxic effects of 2,4,5-T. The mechanism of toxicity, if known, is also explained. Developmental Toxicity The effects of 2,4,5-T on the early development of sea urchin eggs were investigated by Graillet and Girard (1994). Concentrations lower than 5 × 10-4 M were shown to delay the first cleavages and produce a teratogenic effect characterized by a large spectrum of structural malformations at the pluteus stage. The relevance of these findings to humans is not known. Mechanism of Toxicity 2,4,5-T has been shown to increase plasmalemmal calcium permeability of unfertilized sea urchin eggs by opening voltage-dependent calcium channels. Calcium permeability was also increased after

OCR for page 47
--> fertilization of eggs. ATP-dependent intracellular sequestration of calcium in cortices in vivo was inhibited, suggesting that the teratogenic potency of 2,4,5-T in sea urchin eggs is associated with delay in first cleavages and alterations in calcium homeostasis. Other Toxicities No other studies were identified that implicated other health outcomes in laboratory animals. Toxicity Profile Update of Cacodylic Acid Summary During the reference period, cacodylic acid was reported to induce renal lesions in rats. A series of studies indicates that high concentrations of cacodylic acid result in the formation of a toxic intermediate: the dimethylarsenic radical. Toxicokinetics No studies were identified pertaining to the absorption, distribution, metabolism, and excretion of cacodylic acid. Mechanism of Action In a series of studies conducted by Yamanaka and colleagues (1993, 1994a, 1994b), results indicated that cacodylic acid induces lung DNA damage in mice and that this effect involves formation of the dimethylarsenic peroxyl radical during metabolic processing of cacodylic acid. DNA damage was hypothesized to result from the radical's ability to induce single-strand breaks and DNA-protein crosslinks. These results were later reproduced in a human embryonic alveolar epithelial cell line where alkali-labile sites in DNA were shown to precede the formation of DNA single-strand breaks and DNA-protein crosslinks (Yamanaka et al., 1994b). However, the relevance of these data has been questioned, because the changes have only been observed at high concentrations of the parent compound. Disease Outcomes and Mechanisms of Toxicity In this section, we summarize studies that investigate the toxic effects of cacodylic acid. The mechanism of toxicity, if known, is also explained. Renal Toxicity Only one study examined the toxicological effects of cacodylic acid. In 1993, Murai et al. reported that cacodylic acid induces renal lesions in F344/DuCrj rats following oral administration. Male and female rats were administered

OCR for page 47
--> 57, 85, and 113 mg/kg of cacodylic acid for four weeks. Chemical treatment was associated with dose-related decreases in body weight and survival rates in both sexes. Mortality was higher and appeared more quickly in females than males. Histopathological analysis revealed proximal tubular degeneration and necrosis, as well as papillary necrosis, and hyperplasia of the epithelium covering the papillae. Since extensive proximal tubular necrosis was only found in dead animals of both sexes, these investigators concluded that death was indeed mediated by nephrotoxicity. Other Toxicities No other studies were identified that implicated other health outcomes in laboratory animals. Toxicity Profile Update of Picloram No studies were identified during the period 1992-1995 pertaining to the toxicokinetics, mechanism of action, or toxicity of picloram. Toxicity Profile Update of TCDD Summary TCDD elicits a diverse spectrum of sex-, strain-, age-, and species-specific effects, including carcinogenicity, immunotoxicity, reproductive/developmental toxicity, hepatotoxicity, neurotoxicity, chloracne, loss of body weight, and numerous biological responses, such as the induction of phase I and phase II drugmetabolizing enzymes and the modulation of hormone systems and factors associated with the regulation of cellular differentiation and proliferation. TCDD is slowly removed from the body. It is metabolized by liver enzymes to water-soluble derivatives that are more easily eliminated from the body than TCDD itself. The development of refined, physiologically based pharmacokinetic models provides insight into the complexity of TCDD effects in target cells. These models take into account intracellular diffusion, receptor and protein binding, and liver enzyme induction to establish the fractional distribution of the total body burden as a function of the overall body concentrations. The evidence to date continues to support the notion that the biologic effects of TCDD are often mediated by the aryl hydrocarbon receptor (AhR), a member of the ligand-activated transcription factor superfamily. The AhR requires a second protein, known as ARNT, for DNA binding capability and transcriptional activation. Cloning of the AhR and ARNT proteins have identified them as bHLH proteins. Comparisons of the murine and human AhR have demonstrated that the N-terminal but not the C-terminal region of both proteins is highly conserved. The carboxy terminal region involved in transactivation is hypervariable,

OCR for page 47
--> a feature that may account in part for the species differences in TCDD responsiveness. Evidence is beginning to accumulate that both the expression of multiple forms of the AhR in target tissues and events other than receptor binding influence the biological response to TCDD. With respect to the latter, there is evidence for ligand-independent activation of the AhR, for transcriptional independent responses (e.g., modulation of cellular kinase activities and calcium homeostasis), and data to support the view that the transcriptional regulation of the AhR-responsive genes is dictated by combinatorial interactions among proteins. It is clear that AhR-related signaling influences, and is itself influenced by, other signal transduction mechanisms at low concentrations. Signaling interactions explaining the toxic effects of TCDD may involve growth factors, free radicals, the interaction of TCDD with the estrogen transduction pathway, and protein kinases. It appears that PKC is responsible for the phosphorylation of the active complex, but it appears unlikely that PKC phosphorylation is required for ligand binding and transformation. The biologic significance of phosphorylation-related events is not yet fully understood, but it is interesting to note that intracellular protein tyrosine kinase phosphorylation is a more sensitive index of TCDD exposure than hepatic EROD induction. Toxic equivalency factors (TEFs) have been used to estimate the potential health risks associated with exposure to TCDD and complex mixtures containing structurally similar chemicals. The approach assumes linearity of the toxic response for TCDD and is based on a receptor-mediated mechanism of action. The TEF approach has come under increasing scrutiny, because it disregards potentially significant kinetic interactions, tissue-specific effects, and interactions among chemicals present in complex mixtures. These factors, in addition to dietary factors and interspecies and interindividual differences in sensitivity, may influence TCDD toxicity. Studies published during the reference period have focused primarily on the mechanisms of toxicity of TCDD. To date, the consensus is that TCDD is not genotoxic and that its ability to influence the carcinogenic process is mediated via epigenetic events such as enzyme induction, cell proliferation, apoptosis, and intracellular communication. Recent studies on the effects of TCDD and related substances on the immune system amplify earlier findings and suggest that these compounds affect primarily the T-cell arm of the immune response. The effects may result from decreased functions of the thymus gland. Direct effects of TCDD on T cells in vitro, however, have not been demonstrated, suggesting that the action of TCDD may be indirect. A number of recent studies suggest that cytokines may play an important role in regulating the immune response. It should be emphasized that very little change to the overall immune competence of the intact animal has been reported. In contrast, a number of animal studies of the reproductive and developmental

OCR for page 47
--> toxicity of TCDD suggest that developing animals may be particularly sensitive to the effects of TCDD. Specifically, male reproductive function has been reported to be altered following perinatal exposure to TCDD. In addition, experimental studies in rats of the effects of TCDD on the peripheral nervous system suggest that a single low dose of TCDD can cause a toxic polyneuropathy. Other studies have focused on the mechanism of hepatotoxicity, cardiovascular toxicity, wasting syndrome, chloracne, and renal toxicity. There is evidence that hepatotoxicity of TCDD involves AhR-dependent mechanisms. Specifically, there is evidence that the AhR receptor plays a role in the co-mitogenic action of TCDD with epidermal growth factor and in the induction of liver enzymes involved in the metabolism of xenobiotics. Acute exposures to TCDD have been correlated with effects on intermediary metabolism and hepatomegaly. The myocardium has been shown to be a target of TCDD toxicity; impairment of a cAMP-modulated contraction has been implicated. TCDD has been reported to decrease an acidic type I Keratin involved in epidermal development, leading to keratinocyte hyperproliferation and skin irritations such as chloracne. Recent evidence suggests that the inhibition of glucose transport in adipose tissue, pancreas, and brain may be one of the major contributing factors to the wasting syndrome. Finally, in vitro studies have identified glomerular mesangial cells as sensitive cellular targets. These findings are consistent with epidemiologic reports that aromatic hydrocarbons result in glomerulonephritis. Toxicokinetics Physiologically based pharmacokinetic models for the tissue distribution and enzyme-induction properties of dioxin have recently been proposed by Anderson et al. (1993) and Carrier et al. (1995a, 1995b). In the most recent model, Carrier et al. (1995a) describes the distribution kinetics of polychlorinated dibenzodioxins (PCDDs) and polychlorinated dibensofurans (PCDFs) in mammalian species. The model was designed to take into account intracellular diffusion, receptor and protein binding, and liver enzyme induction to establish the fractional distribution of the total body burden between liver and adipose tissues as a function of the overall body concentrations at any one time. In this model, the distribution in rats, monkeys, and humans was shown to follow a nonlinear pattern. In a companion study (Carrier et al., 1995b), it was shown that the liver fraction of the total body burden decreases as overall body concentration decreases, leading to lower global elimination rates and longer half-lives. The study also concluded that for a given body burden of PCDDs and PCDFs, the adipose tissue concentration is inversely proportional to the mass of the adipose tissue. A recent study by Geyer et al. (1993) concluded that total body fat content for lipophilic chemicals, including TCDD, correlates inversely with lethality and in some instances serves as a detoxication mechanism by which TCDD is removed

OCR for page 47
--> from sites of action. This interpretation is consistent with the conclusions reached for veterans of Operation Ranch Hand (Wolfe et al., 1994). The extent to which fat mobilization influences TCDD toxicity has not been completely defined but is likely to be of toxicological significance. The wasting syndrome may influence the disposition of TCDD (Weber et al., 1993), with TCDD concentrations in tissue increasing only when the relative tissue volumes decrease more rapidly than the whole body elimination rate (Roth et al., 1994). TCDD is removed slowly from the body. In humans it takes seven to ten years or more for half of the body burden to be removed (Wolfe et al., 1994). TCDD is metabolized by enzymes in the liver to form water-soluble derivatives that are more easily eliminated from the body than TCDD itself. Due to their more rapid elimination, the water-soluable derivatives of TCDD are also thought to be much less toxic to animals than TCDD itself. Mechanism of Action The mechanism by which TCDD elicits its effects is thought to be mediated by the aryl hydrocarbon receptor (AhR), a member of the ligand-activated transcription factor superfamily. The studies summarized below discuss the structural and functional aspects of the AhR, its DNA binding capability and transcriptional activation, and the biological consequences associated with the activation. This is followed by a discussion of the body of accumulating evidence (see Inconsistencies in the Receptor Model) suggesting that multiple forms of the AhR, as well as events in addition to receptor binding, influence the biological response to TCDD. This section ends with a discussion of the method used to estimate the potential health risk associated with exposure to TCDD and the factors influencing TCDD toxicity (e.g., tissue specificity, interspecies and interindividual differences in sensitivity, chemical interaction). Structural and Functional Aspects of the AhR A cDNA encoding the murine Ah receptor (Ahb-1 allele) was first isolated and characterized from wild type Hepa-1c1c7 cells (Burbach et al., 1992; Ema et al., 1992). Sequence analysis revealed three domains: a basic helix-loop-helix (BHLH) motif; a region that exhibits sequence homology with Per (a Drosophilia circadian rhythm protein) and Sim (a regulatory protein that participates in Drosopholia central nervous system development), termed the "PAS" region; and a "glutamine-rich" region. The BHLH region is believed to be involved in heterodimerization and DNA binding, the PAS region in ligand binding (Dolwick et al., 1993), and the ''glutamine-rich" region in activation (Whitlock, 1993). With respect to the BHLH region, of particular significance is the fact that the AhR itself does not bind strongly to DNA but requires another protein, the AhR nuclear translocator (ARNT) protein, for DNA binding capability and transcriptional activation (Reyes et al., 1992).

OCR for page 47
--> The AhR locus encodes the structural gene for the AhR and has recently been localized to human chromosome 7, band p21 (Le Beau et al., 1994; Ema et al., 1994). Recent studies have characterized a cDNA encoding a human AhR and shown that this protein shares many of the same structural and functional regions as the murine AhR (Dolwick et al., 1993). Comparison of the murine and human receptors has demonstrated that the amino terminal halves of these proteins are highly conserved. This high degree of homology is consistent with the region's role in ligand binding, DNA recognition, and dimerization. A number of conserved phosphorylation sites are also present in the amino terminal half of the protein and a consensus nuclear localization sequence in the center of the PAS region. In contrast, the "glutamine-rich" carboxy end of the proteins is poorly conserved and referred to as a hypervariable region (Dolwick et al., 1993). This region is believed to participate in receptor transformation to a high DNA affinity conformation and transcriptional activation and thus may account for the variability often reported within and across species. The cDNA for mouse and human ARNT have been recently cloned (Reisz-Porszasz et al., 1994; Li et al., 1994). The human ARNT cDNA encodes an 86 kDa protein which, like the AhR, contains a BHLH domain and a domain homologous to Per and Sim. ARNT itself does not bind TCDD or DNA (Whitelaw et al., 1993), but the AhR:ARNT heterodimer functions as a transcriptional enhancer of a number of genes. This is described in more detail below (see DNA Binding Capability and Transcriptional Activation). In mice, four receptor alleles have been identified that encode for the AhR (Poland et al., 1994). These alleles differ by a few point mutations in the open reading frame and by additional sequences in their carboxy ends. The AhRb-1, AhRb-2, and AhRb-3 alleles encode proteins of 95, 104, and 105 kDa, respectively. The Ahb-1 receptor is more thermostable than the Ahb-2 and is not as easily activated in vitro (Poland et al., 1994). The fourth allele, referred to as AhRd, encodes a 104 kDa protein that has a ten-fold lower affinity for agonist relative to other alleles; thus mice harboring the Ahrd allelle are much less susceptible to the biological effects of receptor ligands. A polymerase chain reaction approach has been used to determine the AhR mRNA content in several tissues of C57BL/6J and DBA/2J mice (Li et al., 1994). The highest mRNA level was found in lung, followed by heart, liver, thymus, brain, and placenta. Low levels were found in spleen, kidney, and muscle. No significant differences in mRNA levels were found between the two mouse strains. The normal cellular functions of the AhR and its role in cellular homeostasis have remained elusive for the past 20 years. The AhR binds aromatic hydrocarbons and in this manner serves as a sink for highly lipophilic molecules. This, however, is not likely to be the primary function of the protein under normal conditions. In support of this suggestion is the demonstration that loss of the AhR is associated with immune deficiency and lymphocyte loss (Fernandez-Salguero et al., 1995).

OCR for page 47
--> The exciting development of AhR-deficient "knock-out" mice by homologous recombination in embryonic stem cells will likely provide the tools to define the roles of the AhR in mammalian species (Fernandez-Salguero et al., 1995). The data in this highly publicized study implicate the AhR in the regulation of liver and immune system development. Decreased accumulation of lymphocytes in spleen and lymph node, but not in thymus, was observed in these animals. The livers of these mice were reduced in size by 50 percent and exhibited bile duct fibrosis. DNA Binding Capability and Transcription Activation The molecular mechanism of AhR-mediated responses has primarily been defined based on studies of TCDD-induced cytochrome P450-IA1 (CYP1A1) gene expression. The unbound AhR appears to be a multimeric cytosolic complex that undergoes a process of transformation following binding of a ligand, such as TCDD, to the AhR complex. Recent immunofluorescence studies have confirmed the cytosolic localization of the unoccupied AhR and the nuclear translocation after ligand binding (Pollenz and Poland, 1993). Although the molecular events associated with the transformation process have yet to be fully defined, transformation involves dissociation of two molecules of heat shock protein 90 (Hsp90). Hsp90 itself is composed of two separate gene products, hsp86 and hsp84 (Perdew et al., 1993). Hsp90 represses the intrinsic DNA binding ability of the AhR and preserves the conformation of the AhR (Pongratz et al., 1992). The dissociation of Hsp90 allows the transformed ligand-bound receptor to form a heterodimer with ARNT (Whitelaw et al., 1993). The TCDD-transformed AhR complex binds to xenobiotic responsive sequences (XREs) in the 5' flanking region of downstream genes, resulting in initiation of transcription. Analysis of the murine TCDD-responsive domain of CYP1A1 indicates that enhancer activity results from the sum of the activities of as many as six xenobiotic responsive elements (XREs), which can function independently and coordinately (We and Whitlock, 1993). Both components of the transformed AhR complex (the AhR and ARNT) directly contact DNA. It is interesting to note that the XRE recognized by the AhR/ARNT heterodimer differs from the recognition sequence for nearly all other BHLH proteins, such as MyoD, a protein known to influence determination and differentiation of muscle cells. As with other DNA binding proteins, the primary interaction of the ligand/AhR complex occurs within the core sequence, but nucleotides adjacent to the core also contribute to XRE binding (Lusska et al., 1993). Analysis of mutant XREs have revealed that substitution within the four-base-pair core sequence 5'-CGTG-3' results in loss of binding of the ligand/AhR complex. The three base pairs immediately flanking each end of the essential domain are thought to contribute less strongly to receptor binding (Shen and Whitlock, 1992). The mechanism by which the AhR complex influences transcription is still unclear, but the AhR has been reported to influence DNA bending (Elferink and Whitlock, 1990),

OCR for page 47
--> chromatin structure (Durrin and Whitlock, 1989), and nucleosome displacement (Morgan and Whitlock, 1992) and to increase promoter accessibility (Wu and Whitlock, 1992). Biological Consequences of Activation The ability of TCDD to regulate the expression of growth-related genes has recently been shown. This association is not surprising, in view of the similarities that exist between the AhR/ARNT heterodimer and the Myc/Max or MyoD/E2A growth regulatory heterodimers. TCDD induces the expression of c-jun and c-fos in Hepa-1 cells, although this effect may be cell- type-specific (Puga et al., 1992). Binding of the AhR complex to a XRE has been correlated with induction of c-Ha-ras and c-myc expression (Sadhu et al., 1993). XREs have also been identified in the human estrogen receptor (White and Gasiewicz, 1993); thus it is likely that TCDD interacts in a significant manner with the estrogen signal transduction pathway. Evidence consistent with this suggestion has recently been published (Krishnan et al., 1994). In their studies, TCDD was shown to inhibit 17-β estradiol-induced cathepsin D gene expression by targeted interaction of the nuclear AhR with imperfect XREs strategically located within the estrogen receptor-Sp1 enhancer sequence of this gene. While all of the molecular features of AhR-mediated effects have not been fully elucidated, structure activity studies using ligands of varying receptor affinity continue to support the view that most of the biologic actions of TCDD are mediated through interaction with the AhR. Inconsistencies in the Receptor Model Evidence is beginning to accumulate that multiple forms of the AhR and events in addition to receptor binding influence the biological response to TCDD. These are discussed below. Multiple Forms of AhR The expression of multiple forms of the AhR in target tissue has long been suspected, but concrete evidence in support of this view has been lacking. In 1992, Denison reported the presence of two distinct forms of the AhR, alpha and beta, in similar concentrations in cytosolic extracts of rat liver. The binding of ligand to the alpha form requires the receptor to be in its oligomeric conformation (8-10S). In contrast, ligand binding to the beta form can occur with the dissociated species (5-6S). The addition of molybdate to cytosol during homogenization stabilizes alpha against salt-dependent inactivation and subunit dissociation but does not appear to influence the overall amount of TCDD/AhR complex bound to its specific DNA recognition site. These results suggest that alpha, but not beta, is unable to transform or, alternatively, to bind to the XRE with high affinity, which raises interesting questions regarding the heterogeneity of the response to AhR ligands. Combinatorial Interactions Two DNA-binding proteins have been identified in cytosolic and nuclear extracts of mouse Hepa-1 cells, which overlap the

OCR for page 47
--> DNA-binding specificities to the AhR (Carrier et al., 1994). One of these proteins has an apparent mass of 35-40 kDa and only binds to XRE3, one of six DNA binding sites in the promoter region of the CYP1A1 gene. This protein was tentatively identified as a member of the c/EBP family of transcription factors. The second protein, purified by DNA-affinity chromatography, has an apparent molecular mass of 95 kDa and binds to a larger DNA motif that includes the XRE sequence, in XRE3 and XRE5 but not in XRE1 and XRE2. This protein is not likely to be AhR or ARNT, since it was found in receptorless and nuclear-translocation-deficient cells, as well as in cells not exposed to TCDD. In vivo methylation protection assays indicated that two G residues flanking XRE3, one of which is required for binding of the 95 kDa protein, may be protected from methylation in uninduced cells and become exposed upon TCDD treatment, suggesting that the 95 kDa protein may be constitutively bound to XRE3 and displaced by binding of the AhR complex. These results lend support to the view that the transcriptional regulation of the Ah battery of genes, and perhaps other genes influenced by TCDD, could be modulated by combinatorial interactions of the AhR complex with other transcription factors. In addition, an inhibitory factor has been identified in rat thymus; this factor interferes with the binding of the cytosolic AhR to the XRE (Kurl, 1994). Using guinea pig hepatic AhR, Swanson et al. (1993) identified four TCDD-inducible protein-DNA complexes and two distinct heteromeric DNA-binding forms. These findings are in agreement with a recent report by Okino et al. (1993). Ligand-independent Activation Evidence for the ligand-independent activation of the AhR has recently been presented by Lesca et al. (1995). Their studies were conducted to determine whether ligand-independent activation of the AhR mediates the induction of CYP1A1 by benzimidazole derivatives, agents that do not appear to bind the AhR directly. Benzimidazoles were shown to bind early and transiently to an unknown protein and to deplete the AhR in a time- and dose-dependent manner, and only in cells that express low-affinity forms of the AhR, such as those present in rabbit and human cells. In contrast, benzimidazoles were unable to induce CYP1A1 mRNA in mouse Hepa-1 cells and to deplete the high-affinity AhR form from these cells. These data are intriguing and provide evidence that the effects of AhR ligands can be influenced at multiple levels. Transcriptional-independent Responses The view that not all of the biological responses elicited by TCDD involve a transcriptional component has become increasingly accepted. This is particularly true for the ability of TCDD at low concentrations to modulate cellular kinase activities (Enan and Matsumura, 1995) and calcium homeostasis (Puga et al., 1992). However, the extent to which transcriptional independent events are mediated through the AhR or directed by receptor-independent mechanisms remains unclear. One of the first observable

OCR for page 47
--> effects of TCDD in cultured murine hepatoma cells is a rapid, transient increase in Ca2+ influx and a minor, but significant, elevation of activated, membranebound protein kinase C (PKC) (Puga et al., 1992). Increased PKC is associated with induction of several immediate early proto-oncogenes, including c-fos, jun-B, c-jun, and jun-D, and large increases in AP-1 transcription factor activity. Induction of proto-oncogene expression in hepatoma cells by TCDD is independent of AhR or ARNT (Puga et al., 1992). AhR Signaling Interactions An interesting correlation was recently established between the induction of the multidrug resistance (MDR) gene product, P-glycoprotein, and AhR signaling in primary human hepatocyte cultures (Schuetz et al., 1995). In these studies, induction profiles of mdr mRNA were compared to those for CYP1A1 mRNA. Induction of CYP1A1 mRNA was observed in hepatocyte cultures from 15 different individuals treated with TCDD or 3-methylcholanthrene. However, induction of mdr mRNA was only observed in half of the preparations treated with TCDD, suggesting that TCDD regulates mdr in humans by a mechanism distinct from the classical AhR pathway. There is a growing body of evidence that AhR-related signaling influences, and is itself influenced by, other signal transduction mechanisms at low concentrations. The toxic effects of TCDD involve disruption of various signal transduction pathways. Signaling interactions explaining the toxic effects of TCDD are discussed below. These involve growth factors and their corresponding receptors, free radicals, protein kinases, and the interaction of TCDD with the estrogen transduction pathway. Growth Factor In vivo and in vitro evidence suggests that TCDD influences epidermal growth factor (EGF), transforming growth factor (TGF) α_, TGF β_, interleukin (IL) 1β, and tumor necrosis factor (TNF), among others. TGFβ1 exerts an inhibitory effect of TCDD-induced EROD activity in cultured cells (Vogel et al., 1994). Likewise, IL-1β interferes with TCDD induction of CYP1A1 and CYP1A2 via a transcriptional mechanism (Barker et al., 1992). Down-regulation of TCDD-induced CYP1A1 activity has also been demonstrated for other cytokines or growth factors (Jeong et al., 1993). In non-transformed human keratinocytes treated with 10 nM TCDD prior to confluence, TCDD altered both the mRNA and protein concentrations of TGFα, TGFβ2, plasminogen activator inhibitor (PAI)-2, and IL-1β, regulatory proteins known to influence the cellular programming of growth and differentiation in these cells (Gaido and Maness, 1994). However, the effects of TCDD on the signal transduction cascades triggered by these factors are strongly influenced by cell type and by the degree of cellular differentiation and hormonal status. For instance, TCDD decreases binding of EGF in the livers of intact female rats but not in ovariectomized rats, suggesting that the response is dependent on estrogen action (Kohn et al., 1993).