Mechanistic Modeling of Rodent Liver Tumor Promotion at Low Levels of Exposure: An Example Related to Dose-Response Relationships for 2,3,7,8-Tetrachlorodibenzo-p-dioxin

Melvin E. Andersen, Ph.D., DABT, CIH

The K.S. Crump Division

ICF Kaiser Engineers Inc.

PO Box 14348

Research Triangle Park, NC 27709

Tel: 919/547-1723

Fax: 919/547-1710

Email: andersenme@aol.com



Rory B. Conolly, ScD

Chemical Industry Institute of Toxicology

Six Davis Drive

Research Triangle Park, NC 27709

Tel: 919/558-1330

Fax: 919/558-1300

Email: rconolly@ciit.org

INTRODUCTION

Mechanistic dose-response models link exposure, tissue dosimetry, biochemical responses, and alterations in cell and tissue function into quantitative mathematical descriptions of the development of adverse responses. Although relatively few mechanistic models have been successfully applied to gain understanding of biological effects at low exposure levels, noteworthy advances have occurred in physiologically based pharmacokinetic modeling and in biologically-structured mutation-cell growth models for tumor promotion and cancer. These two types of models have been combined in our recent work developing a mechanistic model for the effects of TCDD on liver tumor promotion in rats. Our combined model attempts to integrate many of the biochemical and cellular effects of TCDD in the liver. This overview describes the range of hepatic effects associated TCDD, outlines our mechanistic dose response model, emphasizes the continuing challenges in verifying the biological basis of this or any other mechanistic model, and examines the inferences gleaned about expected responses at low doses.

HEPATIC EFFECTS

TCDD is a persistent, ubiquitous environmental contaminant that produces a characteristic group of biological effects including liver cancer in female rats (Kociba et al., 1978; NTP, 1982). Our mechanistic model for TCDD-induced liver promotion attempts to explain in a quantitative manner the relationship between various precursor responses and liver tumors. The endpoint examined in the model was the growth of enzyme altered foci (EAF) in hepatic initiation-promotion studies. The critical aspects of any mechanistic model for EAF production and subsequent tumor promotion are the manner in which liver cells mutate to form premalignant phenotypes and the growth characteristics of mutated cells that endow them with a selective advantage over normal cells.

After ingestion into the body and delivery to the liver via the bloodstream, TCDD increases the concentrations of a number of hepatic enzymes involved in cellular metabolism. The most extensively studied molecular response is induction of two cytochrome P450 enzymes, CYP1A1 and CYP1A2. Enzyme induction is mediated via the interaction of TCDD with the aryl hydrocarbon (Ah) receptor. The inactive (Ah) receptor is an aggregate of several proteins. Two of these are heat shock proteins (Perdew, 1992), involved in maintaining the receptor in a proper form for binding to TCDD. A third protein, originally designated p50 based on its molecular weight (Tai et al., 1992), has been characterized as c-Src protein kinase (Enan and Matsumura, 1996). Binding of TCDD to the Ah receptor leads to dissociation of both the Ah receptor and the c-Src kinase from the aggregate. The Ah-TCDD complex dimerizes with another protein, the aryl hydrocarbon nuclear translocator, known as ahnt. The heterodimer formed from these 2 proteins has a high affinity for specific DNA-response elements (Reyes et al., 1992) and acts as a transcriptional regulatory protein. The activation of gene transcription by heterodimers containing the Ah-receptor-TCDD complex is a necessary, although not sufficient, step for all of the downstream effects of TCDD.

The rat liver consists of acinar subunits. These acinar units are defined with respect to the arrangement of portal triads ­ i.e., the portal vein, hepatic artery, and bile ductules ­ and the centrilobular vein that drains the acinar units (see Figure 1). Enzyme induction in the liver is not homogeneous within these acinar structures (Bars and Elcombe, 1991). At lower dose rates, averaging 3.5 and 10.7 ng TCDD/kg/day for 90 days, induction occurred preferentially in the centrilobular region and progressively moved outward to the mid-zonal and then the periportal areas as daily dose rate increased up to 125 ng/kg/day. The individual hepatocytes were either fully induced or remained in a basal state, i.e., there was a sharp boundary between areas of induced cells and areas with no induced cells (Tritscher et al., 1992). This heterogeneity of enzyme induction within the liver acinar unit is a prominent characteristic of TCDD and other liver tumor promoters and should be accountedfor explicitly in the mechanistic model for tumor promotion.

Other Responses:

Other hepatic responses to TCDD are intermediate steps between the initial transcriptional activation and tumor promotion. Concentrations of porphyrins are increased in the liver, an increase that correlates with CYP1A2 induction (van Birgelen et al., 1996). Oxidative stress associated with this porphyria is expected to lead to cytotoxicity. Increases in cell division rates throughout the liver and cellular toxicity, defined as cytoplasmic vacuolation, fatty changes, bile duct hyperplasia, and pigment accumulation in Kupfer cells, were observed at the end of a 90-day study (Maronpot et al., 1993).

Cell Labeling Index/Regional Cell Proliferation:

Treatment of rats with TCDD alters the regional distribution of dividing cells within the liver acinus. In an accelerated dosing scenario designed to bring rats to steady-state in two weeks, histological analysis of liver showed a marked shift in the pattern of hepatocyte proliferation within the acinar structures of the liver. In control rats, proliferation occurred randomly throughout the acinus (Fox et al., 1993). Proliferating cells were concentrated in the periportal area in the TCDD treated rats. Compared to control animals, the cell division rates in the mid-zonal and centrilobular regions appeared to be reduced. In an initiation-promotion study (Maronpot et al., 1993), cell division preferentially occurred in the periportal area in the groups of rats treated with TCDD at doses of 10, 35, and 125 ng/kg/day. Although these regional effects were not noted in the diethylnitreosamine (DEN)-initiated rats treated with TCDD, there was a decrease in total labeling in DEN-initiated livers treated with 3.5 ng/kg/day compared to controls.

After 70% partial hepatectomy, the rat liver undergoes a period of enhanced cell division, returning to normal size over several days. Twenty four hours after partial hepatectomy, 61% of the hepatocytes in control animals had been recruited in the cell cycle. With TCDD, only 41% of the hepatocytes were in the cell cycle at a similar time after partial hepatectomy (Bauman et al., 1995). In general, the body of research with TCDD is consistent with mito-suppression at lower doses, a shift of cell division to the periportal region at low-to-moderate doses, and increased cell division rates and toxicity throughout the liver at higher doses.

Tumor-Promotion Studies:

In tumor promotion studies, young animals are treated with an initiator, i.e., a compound that will alter DNA structure and increase the probability of mutations during cell replication, during periods of enhanced cell division. Cell division can be initiated by necrotizing doses of DEN, by partial hepatectomy, or by using young, rapidly growing neonatal animals. Pitot et al. (1987) dosed rats with 10 mg DEN/kg 24 hr after a 70% partial hepatectomy. The rats were then treated with TCDD for 6 months and the livers examined for the number and size of so-called enzyme altered foci (EAFs). These EAF stain differently than do normal liver cells and can be visualized and counted. They are presumed to arise from specific mutated cells that derive a growth advantage during treatment with the promoter. TCDD was a potent promoter, causing an increase in the number of foci per unit volume of liver and in the amount of the liver occupied by foci at daily doses of 100 ng TCDD/kg. The dose response for these effects appeared to be U-shaped (Figure 2); however, control and treated animals were killed at different times confounding interpretation of the point for the zero TCDD dose group. In another liver tumor promotion study (Stinchcombe et al., 1995), apoptotic rates, a measure of cell death, were much lower in TCDD-treated rats than they were in foci from control animals.

MECHANISTIC MODELS FOR TUMOR PROMOTION

Mechanistic models for TCDD-induced liver tumor promotion were developed in which DEN-initiation or TCDD-treatment was assumed to produce a single type of initiated cell whose growth characteristics were then altered by TCDD (Moolgavkar et al., 1996; Portier et al., 1996). This model structure is referred to here as a 'one-cell' model for TCDD liver tumor promotion. This model could only be fitted to available clone growth data when the combined TCDD-DEN treatment was assumed to have initiating potential, acting to increase the mutation rate of hepatocytes during the prolonged treatment with TCDD. This conclusion of TCDD-induced initiation from the 'one-cell' model is difficult to reconcile with the fact that TCDD has been tested in multiple in vivo and in vitro test systems for mutagenicity. With only two exceptions (Rogers et al., 1982; Yang et al., 1992), the mutagenicity tests were negative. In these single cell promotion models, all biological effects were considered monotonic functions of the TCDD dose rates. Thus, these descriptions did not use liver tissue dosimetry or enzyme induction estimates from pharmacokinetic models to provide dosimeters for the initiation-promotion model.

An Alternative Promotion Model:

Negative selection is the condition under which the growth regulatory environment in an organ suppresses division rates of normal cells and encourages growth and clonal expansion of mutated cells. Jirtle et al. (1991) proposed a negative selection model to explain hepatic tumor promotion by phenobarbital (PB). In this mechanism, PB first provides a mito-stimulatory stimulus that is counteracted by elaboration of mitosuppressant growth factors. Specific cells, mutated at loci that render them insensitive to the mito-inhibitory environment, derive a growth advantage compared to normal cells. Colonies of these cells expand and form foci that may eventually progress to tumors. With PB, transforming growth factor b1 is the inhibitory growth factor. Mutations in the cell surface receptor that binds and activates TGF-b1 render specific initiated cells unresponsive to TGF-b1 (Jirtle et al., 1994). This negative selection mechanism may be broadly applicable to a diverse group of liver tumor promoters, including dioxin (Andersen et al., 1995). The proliferative and mito-inhibitory factors at work for different compounds will very likely vary from promoter to promoter without altering the basic underlying mechanism.

Many of the effects of TCDD on cell division rates in the liver are consistent with the presence of a mito- inhibitory growth environment (Fox et al., 1993; Maronpot et al., 1993; Bauman et al., 1995). In addition, several of the biological effects, including tumor incidence (Kociba et al., 1978), promotion (Pitot et al., 1987), and cell proliferation (Maronpot et al., 1993), appeared to show U-shaped dose response behaviors. These U-shaped dose response curves led us to consider alternative clone growth models and motivated our development of a quantitative negative selection model for TCDD-related clonal growth (Conolly and Andersen, 1997).

Physiologically-Based Regional Induction Models:

TCDD-dose metrics for use in the clonal growth model should be based on the concentrations of TCDD in liver or on the extent and location of induction within the liver. PB-PK models for TCDD that included gene induction were developed in the late 1980's and early 1990's (Leung et al., 1988; Leung et al., 1990; Andersen et al., 1993; Kedderis et al., 1993; Kohn et al., 1993). These pharmacokinetic and gene induction models were only recently extended to consider the regional nature of the induction together with the observation of the sharp boundary between induced and non-induced regions. To describe regional induction, the liver was divided into 5 sub- compartments whose structure was defined based on geometric considerations with respect to a hexagonal acinus (Andersen et al., 1997b). The volumes of the 5 regions in the liver acinus were, respectively, 13.5, 25.5, 33.9, 20.3, and 6.8 percent of total for the subcompartments going in order from the periportal to the centrilobular regions (Figure 1).

Enzyme induction was modeled by an increased rate of transcription of CYP1A1 and CYP1A2 m- RNA in proportion to the occupancy of Dioxin Response Elements (DREs) on DNA by the Ah-TCDD complex. The relationship between the concentration of the Ah-TCDD complex and the rate of transcription was:

To create regional induction with sharp boundaries between induced and non-induced regions, the model derived binding affinity of DREs for the Ah-TCDD complex (Kdi) varied between the 5-compartments in the liver. This geometric induction model was capable of simulating all available data on total induction of m-RNA and CYP1A family proteins and on the regional patterns of induction. The successful set of parameters had Hill-coefficients of 4 or greater (Andersen et al., 1997a). The Kdi values varied by a factor of 3 between the sub- compartments. The large 'n' value indicates that the dose response for induction at very low doses is expected to be highly non-linear. A representative visual depiction of a regional simulation of induced protein is provided in Figure 1. Since the response in individual cells is an all-or-none function, the prediction of the percent induction in the liver is equivalent to a prediction of the proportion of cells activated to a dioxin-responding phenotype. The possible relationship between switching cells abruptly from the off-to-on state and autoregulation of Ah receptors has recently been discussed (Andersen and Barton, 1998).

Figure 1.

Hexagonal Acinar Structures in the Liver: The geometric model describes the regional organization of the acinus with hexagonal acinar structures delineated by the portal triads located at the 6-corners of the hexagons and the central vein, located in the center of the acinus. This model predicts regional induction by assuming different binding affinities between the Ah- receptor-TCDD complex and DRE DNA binding sites in each subcompartment. The extent of induction is calculated from the PB-PK model (Andersen et al., 1997a,b). The proportion induction is used to calculate color intensity for the region (bottom panel). The high 'n' values in equation (1) lead to a situation where there is partial induction in one subcompartment, the midzonal subcompartment in the bottom panel. Some zones, such as the two central areas, are fully induced, while other zones, such as the periportal subcompartments, remain completely un-induced.



Based on the results of the regional induction model, three dose regions were identified for use as potential dosimeters in the clone growth model. Low concentration effects of dioxin would be expected to occur for those doses that induce cells in the centri-lobular area, equivalent to a single subcompartment in the model. Moderate concentration effects would be expected for doses that induce cells in the centrilobular through the midzonal region (equivalent to three subcompartments in the model). High dose effects were anticipated for those doses that induce cells in the entire liver (equivalent to all 5 subcompartments in the model). These definitions of low, moderate and high dose ranges were carried over for use in the stochastic clonal growth model.

Clonal Growth Modeling:

In our implementation of a negative selection model, DEN-initiation coupled with partial hepatectomy is assumed to produce 2 different cell types (A and B), both of which are capable of growing out to produce enzyme altered foci in the liver. The model assumes that one of these cell types (A) responds to the negative growth environment associated with TCDD treatment. The net growth rate for this cell type, given by the cell division rate minus the death rate (AA), decreases with increasing TCDD exposure. The majority of clones observed in DEN-controls in the absence of TCDD treatment would be derived from the A-type cells. The further assumption is that the second type of cell (B) is unresponsive to the mito-inhibitory environment associated with TCDD treatment. For these cells, (BB) increases with increasing concentrations of TCDD. At the high exposure concentrations in the initiation-promotion studies, the observed clones would primarily be derived from these B-cells. This 'two-cell' model is consistent with the observed U-shaped dose response curves. It also explains TCDD promotion without assuming a mutational component for TCDD contributing to the formation of the clones over time. In addition, the model is consistent with a mitoinhibitory action of TCDD on normal cells that is lacking in the B-cell clones.

The quantitative simulations of the growth of these A and B type clones followed general methods described by Conolly and Kimbell (1994). The stochastic clone growth model was first run against data from the Pitot et al. (1987) to demonstrate that the 'two- cell' model could provide curve shapes similar to the published results (Figure 2). With the A-type cells, the cell division rate increased with dose. The ratios of the division rate in treated rats versus the rate in control rats were1, 1, 2, and 3, respectively, for daily intake rates of 0.0, 0.0001, 0.001, 0.01, 0.1 ug/kg/day. The death rate for these A-type cells increased more sharply with dose by 20, 40, 50, and 70 fold at the 4 test concentrations. The increasing death rate leads to extinction of most of the A-cell clones at the higher dose rates. The B cells had relative birth rates of 1, 1, 8, and 33 at the 4 dose rates. The death rate for B cells had relative rates compared to control of 1, 1, 1, and 0.12. The interpretation is that A cells are affected preferentially by the mitosuppression in the liver and lack mutations associated with escape from the mitosuppression. The B cells are resistant to the mitosuppression and respond at the high dose with both increased cell division rates and reduced apoptosis (Conolly and Andersen, 1997). In modeling the clone growth data, both the net difference (BB) and the absolute magnitude of the rates are important. The (BB) value determines the volume of liver occupied by the foci. The magnitude of the rates for any net difference determines the size of clones observed. When and ß are large, a small number of large clones are expected. When and ß are small, a large number of small clones are expected (Dewanji et al., 1989).

Figure 2.

Clonal Growth Modeling for TCDD: The altered foci here represent areas staining for three markers - gamma-glutamyltranspeptidase, adenosine triphosphatase and glucose-6-phosphatase (Pitot et al., 1987). The doses correspond to the daily doses in the original cancer bioassay (Kociba et al., 1978) where the incidence of liver cancer at 0.0, 0.001, 0.01, and 0.1 ug/kg/day, was 1/86, 0/50, 2/50, and 11/49, respectively. The simulations were run multiple times with a common set of parameters. The upper plot shows number of foci per ml liver; the lower plot shows volume of liver occupied by the foci. Due to the stochastic nature of the model, the predictions vary for each simulation. The value given at intervals along the curves are the data points from Pitot et al. (1987). The point at the origin for the control was model derived because the control animals in the study were held for a longer time before they were killed.



We also parameterized the clone growth model based on dosimeters obtained from the regional induction model for enzyme induction in the liver. The growth suppressant effects on A cells was modeled as if it were related to the activation of cells to the dioxin phenotype in the most sensitive section of the liver near the centrilobular region of the liver (in one subcompartment). The second region, with increasing cell division rates of B cells related to toxicity and decreasing death rates of B cells due to escape from mitoinhibition, was assumed to occur when the majority of cells in the liver became activated. This definition coincides with induction in the centrilobular through the midzonal regions (in three subcompartments). A dose metric related to total induction in the liver did not match the steepness of the foci growth curves (Figure 2) between 0.01 and 0.1 ug/kg/day. This approach, linking the areas of the liver induced with specific alterations in growth characteristics of A and B cells, provided a good description of the steep increase in response at the higher doses. However, this approach to parameter estimation was not as accurate in accounting for the region of the curve with the downward going slope at low intake rates (Conolly and Andersen, 1997).

MODEL INFERENCES AND BIOLOGICAL VERIFICATION

The behavior of our 'two-cell' model is consistent with the clone growth results when the parameters are simply fitted to the data sets. There was a discrepancy in the fit in the low dose region using our presumptions about dose metrics from the regional induction model. This discrepancy may imply that some effects of TCDD on cell dynamics and tumor promotion occur at daily intake rates lower than the intake rates that cause significant enzyme induction. Other tumor promotion studies with phenobarbital (Pitot et al., 1987) and 2,3',4,4',5-pentachlorobiphenyl (Haag-Gronlund et al., 1997) also have weakly U-shaped responses where the downward moving portion of the clone growth dose-response curve is in a region below the intake rates associated with significant enzyme induction.

Recent insights into the function of the Ah receptor complex may be consistent with differential dose response behaviors for enzyme induction and for cell proliferative/mitosuppressant responses. The dissociation of the Ah-receptor aggregate after ligand binding releases Ah receptor-TCDD complexes and c-Src protein kinase (Enan and Matsumura, 1996). The Ah-receptor targets DREs throughout the genome to activate gene transcription of specific groups of TCDD-responsive genes. The c-Src kinase may trigger protein kinase mediated growth factor signal transduction pathways, such as activation of mitogen activated protein kinase 2 (MAP-kinase). These two response pathways with the Ah-receptor and c-Src kinase, discussed by Matsumura and colleagues (Matsumura, 1994; Enan and Matsumura, 1995; Enan and Matsumura, 1996; Matsumura et al., 1997), need not have similar dose response relationships. Both are likely to include bioamplification cascades - one due to Ah-receptor autoregulation (Andersen et al., 1998) and the other to c-Src kinase cascade networks. The tuning of these two pathways, i.e., their relative dose response relationships toward TCDD, needs to be further examined to assess the role each may play in regard to mitogenesis and mitoinhibitory signaling during hepatic promotion by TCDD.

WHAT DO WE DO WITH COMPETING MODELS?

There are currently two mechanistic models that have been proposed for probing the effects of TCDD as a tumor promoter in the low dose region. Both models can be provided with biologically plausible parameter estimates to adequately describe specific data sets. Each is based on different biological assumptions and produces somewhat novel predictions in their implementation. The 'one-cell' model leads to the assertion that TCDD, in the DEN initiated rats, acts as an initiator. The 'two-cell' model, linked more closely to liver dosimetry and regional induction, indicates that growth stimulation may occur at doses lower than the doses that induce enzymes. Neither model can be deemed correct simply because it produces predictions consistent with the data. Nor can the inferences about initiation and low dose growth control be assumed to be correct because they improve correspondence between prediction and data. The correctness of either of these models can only be established by biological studies that examine the basis of the fundamental assumptions. In our opinion, the present state of knowledge with TCDD is much more consistent with the 'two-cell' model and with complex, dose-dependent mechanisms of tumor promotion (see below). Because of this belief, we strongly support mechanistic studies that examine the nonlinearities in enzyme induction and seek to uncover the role of the c-Src kinase cascade in mitostimulation.

DOSE-DEPENDENCIES OF RESPONSE MECHANISMS

Regardless of the mathematical implementation of the mechanistic model, the quantitative description has to be augmented with a narrative of the mode of action that conveys the biological factors involved in the pathogenesis of hepatic neoplasia. Our narrative for the two-cell model is as follows. The body of data on the hepatic effects of TCDD is consistent with three distinct ranges in which different responses are predominant. Lower doses (1-5 ng/kg/day in the rats) affect a small portion of the hepatocytes (up to about 10 %) and appear to have little effect on either cell division rates or toxicity. There may be a limited degree of mitosuppression in this region (or even at slightly lower doses) that is sufficient to alter growth of A-type clones. Moderate doses (10-30 ng/kg/day) produce induction of between 20 and 40 % of the cells. Here, there is evidence of mitoinhibition with little enhancement of cell division rates, although there is a shift to a more periportal pattern of cell proliferation. Finally, at higher doses (35-125 ng/kg/day), cell replication, toxicity, and the persistence of mitoinhibition combine to enhance growth of altered cells, increase the probability of progression to more aggressive phenotypes, and produce frank carcinogenicity. The generalized increase in cell division rates at high doses is probably related to toxicity and may be associated with a role of CYP1A2 induction leading to porphyria. The interaction of multiple, high dose risk factors strongly indicate that cancer risk should decrease disproportionately as dose rates decrease.

NON-MONOTONIC DOSE RESPONSE CURVES

Several studies have suggested the possibility of U-shaped dose-response relationships for some of the hepatic effects of TCDD, including initiation-promotion (Pitot et al., 1987), carcinogenicity (Kociba et al., 1978), and cell labeling (Maronpot et al., 1993). These U-shaped curves reflect the superposition of several different mechanisms with differential dose-dependencies. At low doses, the mechanisms of growth suppression inhibit clonal expansion of initiated cells, reducing clone growth and tumor promotion. The high dose region has mitoinhibition, toxicity, and reparative hyperplasia, all contributing to enhanced clonal growth. When we recognize that the pathogenicity of most toxic responses are composites of multiple, dose-dependent effects of the chemical on the organism, it is easy to appreciate that complex U-shaped dose response curves for adverse effects may arise from differential dose response relationships for the contributing mechanisms of pathogenesis.

SUMMARY

Mechanistic dose-response models, such as those presented here for regional enzyme induction and tumor promotion, have the potential to improve the characterization of the functional relationships between toxicity endpoints and daily intake rates and to reduce the uncertainties in risk assessments for several tumor promoters. The not infrequent observation of U-shaped dose response curves and mitostimulatory/mitoinhibitory responses to multiple tumor promoters indicate that negative selection may be a common mechanism for many rodent liver tumor promoters. Other compounds, including phenobarbital, peroxisomal proliferators, and a variety of dioxin-like compounds, such as dibenzofurans and coplanar PCBs, should be more thoroughly examined to evaluate the role of regional induction and negative selection in the pathogenesis of liver tumor promotion. With respect to TCDD, mechanistic studies of Ah receptor autoregulation and c-Src mediated mitostimulation will be instrumental in corroborating or refuting the assumptions of this 'two-cell' model for clone growth and liver tumor promotion.

REFERENCES

Andersen, M. E. and Barton, H. A. (1998). Biological Regulation of Receptor-Hormone Complex Concentrations in Relation to Dose Response Assessments for Endocrine Active Compounds. Toxicology , in press.

Andersen, M. E., Birnbaum, L. S., Barton, H. A. and Eklund, C. (1997a). Regional Hepatic CYP1A1 and CYP1A2 Induction with 2,3,7,8-tetrachlorodibenzo-p-dioxin Evaluated with a Multi-Compartment Geometric Model of Hepatic Zonation. Toxicol. Appl. Pharmacol. 144, 145-155.

Andersen, M. E., Eklund, C. R., Mills, J. J., Barton, H. A. and Birnbaum, L. S. (1997b). A Multi-Compartment Geometric Model of the Liver in Relation to Regional Induction of Cytochrome P450s. Toxicol. Appl. Pharmacol. 144, 135-144.

Andersen, M. E., Mills, J. J., Gargas, M. L., Kedderis, L., Birnbaum, L. S., Neubert, D. and Greenlee, W. F. (1993). Modeling Receptor-Mediated Processes with Dioxin: Implications for Pharmacokinetics and Risk Assessment. J. Risk Anal. 13, 25-36.

Andersen, M. E., Mills, J. J., Jirtle, R. L. and Greenlee, W. F. (1995). Negative Selection in Hepatic Tumor Promotion in Relation to Cancer Risk Assessment. Toxicology 102, 223-237.

Bars, R. G. and Elcombe, C. R. (1991). Dose-dependent Acinar Induction of Cytochromes P450 in Rat Liver: Evidence for a Differential Mechanism of Induction of P450IA1 by b-naphthoflavone and Dioxin. Biochem. J. 277, 577-580.

Bauman, J. W., Goldsworthy, T. L., Dunn, C. S. and Fox, T. R. (1995). Inhibitory Effects of 2,3,7,8-Tetrachlorodibenzo-p-dioxin on Rat Hepatocyte Proliferation Induced by 2/3 Partial Hepatectomy. Cell Prolif. 28, 437-451.

Conolly, R. B. and Andersen, M. E. (1997). Quantitative Analysis of Dose-Response Data for Tetrachlorodibenzo-p-dioxin: Hepatic Foci in Rats after Diethylnitrosamine Initiation. Toxicol. Appl. Pharmacol., in review.

Conolly, R. B. and Kimbell, J. S. (1994). Computer Simulation of Cell Growth Governed by Stochastic Processes: Application to Clonal Growth Cancer Models. Toxicol. Appl. Pharmacol. 124, 284-295.

Dewanji, A., Venzon, D. J. and Moolgavkar, S. H. (1989). A Stochastic Two-Stage Model for Cancer Risk Assessment. II. The Number and Size of Premalignant Clones. Risk Anal. 9(2), 179-187.

Enan, E. and Matsumura, F. (1995). Evidence for a Second Pathway in the Action Mechanism of 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD). Significance of Ah-receptor Mediated Activation of Protein Kinase Under Cell-free Conditions. Biochem. Pharmacol. 49(2), 249-261.

Enan, E. and Matsumura, F. (1996). Identification of C-SRC as the Integral Component of the Cytosolic Ah Receptor Complex, Transducing the Signal of 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) through the Protein Phosphorylation Pathway. Biochem. Pharmacol. 52(10), 1599-1612.

Fox, T. R., Best, L. L., Goldsworthy, S. M., Mills, J. J. and Goldsworthy, T. L. (1993). Gene Expression and Cell Proliferation in Rat Liver after 2,3,7,8-tetrachlorodibenzo-p-dioxin Exposure. Cancer Res. 53(Suppl 10), 2265-2271.

Haag-Gronlund, M., Warngard, L., Flodstrom, S., Scheu, G., Kronevi, T., Ahlborg, U. G. and Fransson-Steen, R. (1997). Promotion of Altered Hepatic Foci by 2,3',4,4',5-Pentachlorobiphenyl in Sprague-Dawley Female Rats. Fundam. Appl. Toxicol. 35, 120-130.

Jirtle, R. L., Hankins, G. R., Reisenbichler, B. and Boyer, I. J. (1994). Regulation of Mannose 6-Phosphate/Insulin-Like Growth Factor-II Receptors and Transforming Growth Factor Beta During Liver Tumor Promotion with Phenobarbital. Carcinogenesis 15(8), 1473-1478.

Jirtle, R. L., Meyer, S. A. and Brockenbrough, J. S., Eds. (1991). Liver Tumor Promoter Phenobarbital: A Biphasic Modulator of Hepatocyte Proliferation. Chemically Induced Cell Proliferation: Implications for Risk Assessment. Wiley-Liss, Inc.

Kedderis, L. B., Mills, J. J., Andersen, M. E. and Birnbaum, L. S. (1993). A Physiologically Based Pharmacokinetic Model for 2,3,7,8-tetrabromodibenzo-p-dioxin (TBDD) in the Rat: Tissue Distribution and CYP1A Induction. Toxicol. Appl. Pharmacol. 121, 87-98.

Kociba, R. J., Keyes, D. G., Beyer, J. E., Carreon, R. M., Wade, C. E., Dittenber, D. A., Kalnins, R. P., Frauson, L. E., Park, C. N., Banard, S. D., Hummel, R. A. and Humiston, C. G. (1978). Results of a Two-Year Chronic Toxicity and Oncogenicity Study of 2,3,7,8-Tetrachlorodibenzo-p-dioxin in Rats. Toxicol. Appl. Pharmacol. 120, 138-154.

Kohn, M. C., Lucier, G. W., Clark, G. C., Sewall, C., Tritscher, A. and Portier, C. J. (1993). A Mechanistic Model of the Effects of Dioxin on Gene Expression in the Rat Liver. Toxicol. Appl. Pharmacol. 120, 138-154.

Leung, H. W., Ku, R. H., Paustenbach, D. J. and Andersen, M. E. (1988). A Physiologically Based Pharmacokinetic Model for 2,3,7,8-tetrachloro-p-dioxin in C57B1/6J and DBA/2J Mice. Toxicol. Lett. 42, 15-28.

Leung, H. W., Paustenbach, D. J., Murray, F. J. and Andersen, M. E. (1990). A Physiological Pharmacokinetic Description of the Tissue Distribution and Enzyme Inducing Properties of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the Rat. Toxicol. Appl. Pharmacol. 103, 399-410.

Maronpot, R. R., Foley, J. F., Takahashi, K., Goldsworthy, T., Clark, G., Tritscher, A., Portier, C. and Lucier, G. (1993). Dose Response for TCDD Promotion of Hepatocarcinogenesis in Rats Initiated with DEN: Histologic, Biochemical and Cell Proliferation Endpoints. Environ. Health Perspect. 101(7), 634-642.

Matsumura, F. (1994). How Important is the Protein Phosphorylation Pathway in the Toxic Expression of Dioxin-type Chemicals. Biochem. Pharmacol. 48(2), 215-224.

Matsumura, F., Enan, E., Dunlap, D. Y., Pinkerton, K. E. and Peake, J. (1997). Altered in vivo Toxicity of 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) in C-SRC Deficient Mice. Biochem. Pharmacol. 53(10), 1397-1404.

Moolgavkar, S. H., Luebeck, E. G., Buchmann, A. and Bock, K. W. (1996). Quantitative Analysis of Enzyme-altered Liver Foci in Rats Initiated with Diethylnitrosamine and Promoted with 2,3,7,8-Tetrachlorodibenzo-p-dioxin or 1,2,3,4,6,7,8-Heptachlorodibenzo-p-dioxin. Toxicol. Appl. Pharmacol. 138, 31-42.

NTP (1982). Bioassays of 2,3,7,8-Tetrachlorodibenzo-p-dioxin for Possible Carcinogenicity (Gavage Study).National Toxicology Program, Technical Report Series No. 102, Research Triangle Park, NC

Perdew, G. (1992). Chemical Cross-Linking of the Cytosolic and Nuclear Forms of the Ah Receptor in Hepatoma Cell Line 1C1C7. Biochem. Biophys. Res. Commun. 182, 55-63.

Pitot, H., Goldsworthy, T. L., Moran, S., Kennan, W., Glauert, H.P., Maronpot, R. R. and Campbell, H. A. (1987). A Method to Quantitate the Relative Initiating and Promoting Potencies of Hepatocarcinogenic Agents in their Dose-Response Relationships to Altered Hepatic Foci. Carcinogenesis 8, 1491-1499.

Portier, C. J., Kohn, M. C., Kopp-Schneider, A., Sherman, C. D., Maronpot, R. and Lucier, G. W. (1996). Modeling the Number and Size of Hepatic Focal Lesions Following Exposure to 2,3,7,8-TCDD. Toxicol. Appl. Pharmacol. 138, 20-30.

Reyes, H., Reisz-Porszasz, S. and Hankinson, O. (1992). Identification of the Ah Receptor Nuclear Translocator Protein (Arnt) as a Component of the DNA Binding Form of the Ah Receptor. Science 256, 1193-1195.

Rogers, A., Andersen, M. E. and Back, K. C. (1982). The Toxicity and Mutagenicity of 2,3,7,8-Tetrachlorodibenzo-p-dioxin and Similar Acting Compounds. Environmental Mutagen Society Meeting, Boston, MA.

Stinchcombe, S., Buchmann, A., Bock, K. W. and Schwarz, M. (1995). Inhibition of Apoptosis during 2,3,7,8-tetrachlorodibenzo-p-dioxin-mediated Tumour Promotion in Rat Liver. Carcinogenesis 16(6), 1271-1275.

Tai, P.-K. K., Albers, M. W., Chang, H., m Faber, L. E. and Schrieber, S. L. (1992). Association of a 59-Kilodalton Immunophilin with the Glucocorticoid Receptor Complex. Science 256, 1315-1319.

Tritscher, A. M., Goldstein, J. A., Portier, C. J., McCoy, Z., Clark, G. C. and Lucier, G. W. (1992). Dose-Response Relationships for Chronic Exposure to 2,3,7,8-Tetrachlorodibenzo-p-dioxin in a Rat Tumor Promotion Model: Quantification and Immunolocalization of CYP1A1 and CYP1A2 in the Liver. Cancer Res. 52, 3436-3442.

van Birgelen, A. P. J. M., DeVito, M. J., Akins, J. M., Ross, D. G., Diliberto, J. J. and Birnbaum, L. S. (1996). Relative Potencies of Polychlorinated Dibenzo-p-dioxins, Dibenzofurans and Biphenyls Derived from Hepatic Porphyrin Accumulation in Mice. Toxicol. Appl. Pharmacol. 138, 98-109.

Yang, J.-H., Thraves, P., Dritschilo, A. and Rhim, J. S. (1992). Neoplastic Transformation of Immortalized Human Keratinocytes by 2,3,7,8-Tetrachlorodibenzo-p-dioxin. Cancer Res. 52, 3478-3482.