University of South Florida Scholar Commons
Graduate Theses and Dissertations Graduate School
5-3-2008 Analysis of the Role of bHLH/PAS Proteins in Aryl Hydrocarbon Receptor Signaling Edward J. Dougherty University of South Florida
Follow this and additional works at: https://scholarcommons.usf.edu/etd Part of the American Studies Commons
Scholar Commons Citation Dougherty, Edward J., "Analysis of the Role of bHLH/PAS Proteins in Aryl Hydrocarbon Receptor Signaling" (2008). Graduate Theses and Dissertations. https://scholarcommons.usf.edu/etd/218
This Dissertation is brought to you for free and open access by the Graduate School at Scholar Commons. It has been accepted for inclusion in Graduate Theses and Dissertations by an authorized administrator of Scholar Commons. For more information, please contact [email protected]. Analysis of the Role of bHLH/PAS Proteins in Aryl Hydrocarbon Receptor Signaling
by
Edward J. Dougherty
A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Biology College of Arts and Sciences University of South Florida
Major Professor: Richard S. Pollenz, Ph.D. Brian T. Livingston, Ph.D. Kristina H. Schmidt, Ph.D. Robert L. Potter, Ph.D.
Date of Approval: May 3, 2008
Keywords: AHR, ARNT, ARNT2, XAP2, Dioxin, Signal transduction, Xenobiotic metabolism
© Copyright 2008, Edward John Dougherty
Dedication
This manuscript is dedicated to Richard S. Pollenz, Ph.D. who taught me that perseverance pays off (eventually) and that science can be fulfilling (eventually). I would also like to thank my wife, Amy A.B. Tatem, for being extremely supportive and coming to the laboratory with me on late nights and weekends to watch me culture cells and my parents for helping me to attain the education necessary to reach this point. I would also like to thank the remaining members of both my families for all of their support over the years. Acknowledgements
I would also like to acknowledge the members of my committee: Brian T.
Livingston, Ph.D., Kristina H. Schmidt, Ph.D., and Robert L. Potter, Ph.D. as well as my coworkers: Sarah E. Wilson, Gary T. ZeRuth, Robert Buzzeo, and Jesal Popat who together have made working at the University of South Florida a wonderful experience and who have provided valuable input over the past four years.
Table of Contents
List of Tables v
List of Figures vi
List of Acronyms xi
Abstract xiv
Chapter 1: Introduction 1
1.1 AHR and ARNT History 1 1.2 AHR and ARNT Domains 8 1.3 AHR Proteins 15 1.4 AHR-Mediated Signaling 21 1.5 AHR Associated Proteins 23 1.6 Degradation of the AHR 25 1.7 Consequences of AHR Degradation or Overexpression 27 1.8 Mammalian ARNT Isoforms 28 1.9 ARNT Isoforms from Other Species 34 1.10 ARNT-Dependent Signaling 36 1.10.1 AHR Signaling 36 1.10.2 Hypoxic Signaling 38 1.10.3 SIM1/SIM2 Interactions 39 1.10.4 Circadian Signaling 41 1.10.5 AINT Interactions 42 1.10.6 ARNT Homodimer Interactions 43 1.11 ARNT Interactions with Transcriptional Repressors 45 1.11.1 Aryl Hydrocarbon Receptor Repressor Interactions 45 1.11.2 NPAS Interactions 46 1.11.3 Necdin Interactions 46 1.12 AHR and ARNT Levels 47 1.13 ARNT Crosstalk 48 1.14 AHR and ARNT Defective Hepatoma Cell Lines 51 1.15 Knockout Animals 52 1.16 Toxicity of TCDD 56 1.17 Role of AHR and ARNT in Mediating TCDD Toxicity 58
i Chapter 2: Generation and Characterization of Stable Lines Expressing Different Species of AHR 60
2.1 Rationale for Stable Cell Lines Expressing Different Species of AHR 60 2.2 Generation of LA-I lines Expressing AHRb-1 or AHRb-2 61 2.3 Reduced Association of XAP2 with Ahb-2 Receptors Expressed in the Hepa-1 Background Mimics That of Endogenously Expressed Ahb-2 Receptors 68 2.4 Ahb-2 Receptor Expressed in the Hepa-1 Background Exhibits Dynamic Nucleocytoplasmic Shuttling 71 2.5 Exogenous Expression of XAP2 Does Not Affect the Subcellular Localization of the Ahb-1 or the Ahb-2 Receptor 75 2.6 Degradation of the Mouse Ahb-2 Receptor in the Hepa-1 Background Varies From the Ahb-1 80 2.7 Time Course of CYP1A1 Induction is Unaltered by Ah Receptor Type 83 2.8 AHb2 Stable Line Conclusions 86
Chapter 3: Generation and Characterization of Stable Lines Expressing Ah-Receptors Deficient in DNA-Binding or ARNT Dimerization 88
3.1 Rationale for AHR Mutant Lines 88 3.2 Generation of LA-I Lines Expressing AHR DNA Binding Mutants and ARNT Dimerization Mutants 92 3.3 Characterization of AHR Mutants Deficient in DNA Binding and ARNT Dimerization 94 3.4 Ligand-Dependent Degradation of the AHR Requires DNA Binding 104 3.5 Ligand-Independent Degradation of the AHR Does Not Require DNA Binding 108 3.6 Impact of NH-Terminal Tags on AHR Degradation 110
Chapter 4: Functional Analysis of ARNT2 in AHR-mediated Signal Transduction 117 4.1 Rationale for Evaluating ARNT2 Function 117 4.2 ARNT2 Does Not Function to the Same Level as ARNT in AHR-Mediated Signaling 118 4.3 Inability of ARNT2 to Function in the Regulation of CYP1A1 is Not a Result of Proline 352 126 4.4 In Vitro Synthesized ARNT and ARNT2 Appear to Exhibit Equivalent Ability to Dimerize With the AHR and Bind DNA 128 4.5 ARNT2 Can Out-Compete ARNT for Association with the Liganded AHR 137
ii 4.6 The Ability of ARNT2 to Associate with the AHR and Bind DNA is Not Dependent on AHR Concentration or ARNT Protein Concentration 139 4.7 The Ability of ARNT2 to Associate with the AHR and Bind DNA May be XRE-Specific 144 4.8 DNA Binding Ability of AHR•ARNT2 Heterodimers Appears to be Ligand Dependent 146 4.9 The Ability of ARNT or ARNT2 to Dimerize with the AHR in Vitro May be Receptor Species Dependent 153 4.10 The Ability of ARNT or ARNT2 to Dimerize with the AHR in Vitro Does Not Appear to Differ When Expressed in Vivo 157 4.11 Inhibition of ARNT by ARNT2 in AHR-Mediated Signaling 159 4.12 ARNT2 Does Not Function to the Same Level as ARNT in Endogenous AHR-Mediated Signaling 163 4.13 Analysis of ARNT2 Function When Endogenously Expressed 166 4.14 ARNT2 from Nuclear Extracts Fails to Bind XREs 169 4.15 Function of ARNT Versus ARNT2 in Hypoxic Signaling 177 4.16 Evaluation of Potential ARNT and ARNT2 Homodimers 178
Chapter 5: Implications and Future Directions 183
5.1 Implications of AHRb-2 Studies 183 5.2 Implications of AHR Degradation Studies 190 5.3 Implications of ARNT2 Studies 198
Chapter 6: Materials and Methods 212
6.1 Materials 212 6.2 Buffers 212 6.3 Cells and Growth Conditions 213 6.4 Antibodies 213 6.5 Generation of Expression Constructs 214 6.6 In Vitro Expression of Protein 218 6.7 Transient Transfection 218 6.8 Viral Transfections and Selection 219 6.9 Luciferase Reporter Studies 220 6.10 RNA Interference 220 6.11 Immunofluorescence Staining and Microscopy 221 6.12 Preparation of Total Cell/Tissue Lysates 222 6.13 Preparation of Cytosol and Nuclear Extracts 222 6.14 Western Blot Analysis and Quantification of Protein 223 6.15 In Vitro Activation of AHR•ARNT Complexes and Electrophoretic Mobility Shift Assays 224 6.16 Immunoprecipitations 226 6.17 Statistical Analysis 226
iii References 227
Bibliography 264
Appendices 265 Appendix A: Endogenous AHRb-2 Data 266 Appendix B: Hypoxia qRT-PCR Primer Sets 274
About the Author End Page
iv
List of Tables
Table 1.1 Summary of ARNT cloning 5
Table 1.2 Amino acid identity comparisons of murine ARNT types and across ARNT isoform domains 6
Table 1.3 Amino acid changes among the first 805 amino acids of murine AHR alleles 18
Table 1.4 mRNA expression for mARNT and mARNT2 in murine peripheral organs and central nervous system at postnatal day 1.5 33
v
List of Figures
Figure 1.1 Overview of known cellular pathways requiring ARNT or ARNT2 in mammals 7
Figure 1.2 Domain structures of the AHR (b1 allele) and ARNT 13
Figure 1.3 Protein sequence alignment of various AHR c-termini 17
Figure 1.4 AHR protein dendogram 20
Figure 1.5 Schematic of the AHR signaling pathway 22
Figure 1.6 Homology between ARNT and ARNT2 domains 32
Figure 1.7 ARNT protein dendogram 35
Figure 1.8 Structures of common tetra-chlorinated- congeners for various halogenated aromatic hydrocarbons 57
Figure 2.1 Invitrogen plasmid maps 62
Figure 2.2 Western analysis of AHR constructs following transfection into AHR-deficient PT67 viral packaging cells 64
Figure 2.3 Schematic for viral transfection 65
Figure 2.4 Western analysis of stable line expression of target AHR constructs 66
Figure 2.5 Analysis of AHR and XAP2 expression and association in stable cell lines expressing Ahb-1 or Ahb-2 receptors 70
Figure 2.6 Subcellular localization of AHR in AHWT and AHb2 cells exposed to TCDD or LMB 74
Figure 2.7 Association of the Ahb-2 receptor with XAP2 in cells expressing increased levels of XAP2 77
Figure 2.8 Quantification of association of the Ahb-2 receptor with XAP2 in cells expressing increased levels of XAP2 78 vi b-2 Figure 2.9 Localization of Ah receptor in AHb2 cells expressing hXAP2 79
Figure 2.10 Western blot analysis of AHR in stable cell lines expressing Ahb-1 or Ahb-2 receptors exposed to TCDD 82
Figure 2.11 Western blot analysis of CYP1A1 protein in stable cell lines expressing Ah b-1 or Ahb-2 receptors following exposure to TCDD 84
Figure 2.12 Analysis of dose-response for CYP1A1 induction in stable cell lines expressing Ahb-1 or Ahb-2 receptor 85
Figure 3.1 Schematic of the AHR ligand-dependent and ligand-independent 90
Figure 3.2 Schematic of AHR mutations for R39A and ∆NES AHR mutant stable lines. 93
Figure 3.3 Western analysis of AHRtr, R39A, and ∆NES AHR mutants 95
Figure 3.4 Western analysis of R39A and ∆NES stable lines 96
Figure 3.5 Time course of ligand-induced degradation of the AHRWT and trAHR 97
Figure 3.6 Overday or overnight ligand-induced degradation of the AHWT, R39A, or ∆NES stable lines 99
Figure 3.7 Analysis of TCDD-induced luciferase activity in LA-I, AHWT, R39A, and ∆NES stable lines 101
Figure 3.8 Subcellular localization of the ∆NES and R39A AHR 102
Figure 3.9 Immunoprecipitation analysis of ARNT association in ∆NES and R39A stable lines 103
Figure 3.10 Analysis of accumulation of AHR and ARNT in cytosolic and nuclear extracts for AHWT, R39A, and ∆NES stable lines 105
Figure 3.11 Impact of actinomycin D on TCDD-induced degradation of AHRWT and ∆NES stable lines 107
Figure 3.12 Geldanamycin-induced degradation of wild-type Hepa-1 cells, AHWT, R39A, or ∆NES stable lines 109
vii Figure 3.13 Invitrogen plasmid maps for generating constructs with NH- terminal tags 111
Figure 3.14 Characterization of HM- and GFP-tagged AHR stable lines 113
Figure 3.15 Ligand-induced degradation in HM- and GFP-tagged AHR stable lines 114
Figure 3.16 Ligand-induced CYP1A1 induction in HM- and GFP-tagged AHR stable lines 115 Figure 3.17 Geldanamycin-induced degradation of the HM- and GFP-tagged AHR 116
Figure 4.1 Protein schematic of V5-ARNT and V5-ARNT2 constructs 120
Figure 4.2 Analysis of TCDD-induced luciferase activity in LA-II cells transfected with ARNT or ARNT2 121
Figure 4.3 Induction of CYP1A1 protein in cells expressing ARNT or ARNT2 123
Figure 4.4 Induction of CYP1A1 protein in ARNT or ARNT2 stable lines 125
Figure 4.5 Induction of CYP1A1 protein in cells expressing ARNT, ARNT2, or ARNT2-H 127
Figure 4.6 Schematic for evaluation of DNA binding potential of ARNT and ARNT2 using in vitro synthesized and activated samples 129
Figure 4.7 DNA binding of AHR•ARNT and AHR•ARNT2 heterodimers 131
Figure 4.8 Schematic for evaluation of DNA binding potential of a mixture of ARNT and ARNT2 using in vitro synthesized and activated samples 133
Figure 4.9 DNA binding of AHR•ARNT and AHR•ARNT2 heterodimers in the presence of both ARNT and ARNT2 134
Figure 4.10 Association of ARNT and ARNT2 with AHR 136
Figure 4.11 Effect of ARNT2 concentration of the formation of AHR•ARNT complexes 138
viii Figure 4.12 DNA binding of AHR•ARNT and AHR•ARNT2 heterodimers in the presence of limiting AHR 141
Figure 4.13 Effect of target protein concentration on the formation of AHR• ARNT and AHR•ARNT2 heterodimers 143
Figure 4.14 Analysis of XRE sequence on DNA binding ability of AHR• ARNT and AHR•ARNT2 heterodimers 145
Figure 4.15 Effect of 3-MC on the DNA binding ability of AHR•ARNT and AHR•ARNT2 complexes 148
Figure 4.16 Effect of different ligands on the DNA binding ability 149
Figure 4.17 Association of ARNT and ARNT2 with BAP or TCDD activated AHR 152 Figure 4.18 Analysis of AHR species on the formation of TCDD or BAP dependent AHR•ARNT or AHR•ARNT2 heterodimers 155
Figure 4.19 DNA binding of ARNT or ARNT2 expressed in cell culture 158
Figure 4.20 Impact of ARNT2 expression on AHR-mediated signaling 161
Figure 4.21 Impact of ARNT2 expression on AHR-mediated signaling in cell culture 162
Figure 4.22 ARNT and ARNT2 protein expression in tissues and cells 165
Figure 4.23 Reduction of endogenous ARNT or ARNT2 by siRNA knockdown in hRPE cells 167
Figure 4.24 DNA binding of NRK and hRPE nuclear extracts 170
Figure 4.25 DNA binding of NRK cytosolic extracts 172
Figure 4.26 DNA binding of Hepa-1 nuclear extracts 174
Figure 4.27 Immunoprecipitation analysis of NRK and WT Hepa-1 cells expressing ARNT2 176
Figure 4.28 Analysis of the role of ARNT2 in hypoxia 179
Figure 5.1 Hypothetical model for lack of ARNT2 function in vivo 203
ix Figure A-1 Analysis of AHR and XAP2 expression and association in Hepa-1, A7, and C2C12 cells 267
Figure A-2 Subcellular localization of AHR in Hepa-1, A7, C2C12, and 10T1/2 cells exposed to TCDD or LMB 268
Figure A-3 Reduction of endogenous XAP2 by siRNA knockdown in Hepa-1 cells 269
Figure A-4 Subcellular localization of AHR in cells with reduced levels of XAP2 270
Figure A-5 Association of endogenous XAP2 with AHR in Hepa-1 cells transfected with hXAP2 expression vectors 271
Figure A-6 Localization of endogenous Ahb-1 receptors in Hepa-1 cells expressing hXAP2 272
Figure A-7 Western blot analysis of AHR in Hepa-1, A7, and C2C12 cells exposed to TCDD 273
x
List of Acronyms
3MC, 3-methylcholanthrene AD, actinomycin-D AHR, aryl hydrocarbon receptor AHRR, aryl hydrocarbon receptor repressor ARNT, aryl hydrocarbon receptor nuclear translocator ARNT2, aryl hydrocarbon receptor nuclear translocator isoform 2 ARNTL, ARNT-like protein BAP, benzo[a]pyrene bHLH, basic helix-loop-helix domain BMAL, brain and muscle ARNT-like protein CAR, constitutively active AHR ChIP, chromatin immunoprecipitation CHIP, carboxyl terminus of hsc70-interacting protein CHX, cycloheximide CLIF, cycle-like factor CME, central midline enhancer CMV, cytomegalovirus CoCl2, cobalt chloride CUL4, cullin 4b CYP1A1, cytochrome P450 1A1 CyP40, cyclophilin 40 DDB1, DNA damage binding protein 1 DFO/DFX, desferrioxamine DMEM, Dulbecco’s modified Eagle’s medium DORV, double outlet right ventricle DRE, dioxin response element E-box, enhancer box ECL, enhanced chemiluminescence ED, embryonic dat EMSA, electrophoretic mobility shift assay ER, estrogen receptor EROD, 7-ethoxycoumarin O-demethylase FBS, fetal bovine serum FCS, fetal calf serum FK506, Tacrolimus or Fujimycin FK, FK506 binding domain GA, geldanamycin
xi GAM-HRP, goat-anti-mouse IgG conjugated to hydrogen peroxidase GAR-HRP, goat-anti-rabbit IgG conjugated to hydrogen peroxidase GAR-RHO, goat-anti-rabbit IgG conjugated to hydrogen rhodamine GD, gestational day GFP, green fluorescent protein GR, glucocorticoid receptor GST, glutathione S-transferase HAH, halogenated aromatic hydrocarbons HIF, hypoxia-inducible factor HLF, hypoxia-like factor HM, hexahistidine tag with Xpress epitope (Invitrogen) HRE, hypoxia response element Hsp90, heat shock protein 90 IP, immunoprecipitation ITE, 2-(1'H-indole-3'-carbonyl)-thiazole-4-carboxylic acid methyl ester kD/kDa, kilodalton LMB, leptomycin B Me2SO, dimethyl sulfoxide MOP, modulator of PAS NES, nuclear export sequence NLS, nuclear localization sequence PAGE, polyacrylamide electrophoresis PAH, polycyclic aromatic hyrocarbons PAS, PER/ARNT/SIM homology domain PBS, phosphate buffered saline PCR, polymerase chain reaction PD, postnatal day PER, period protein Pi, Pre-immune PPI, peptidylprolyl isomerase domain PVN, paraventricular nuclei qRT-PCR, quantitative real time polymerase chain reaction SDS, sodium dodecyl sulfate SE, standard error of the mean SIM, single-minded protein SON, supraoptic nuclei SRC, steroid receptor coactivator TAD, transactivation domain TCDD, 2,3,7,8-tetrachlorodibenzo-p-dioxin TNT, transcription and translation product TPR, tetratricopeptide repeat TTBS, Tris buffered saline with Tween 20 V5, epitope present on the P and V proteins of the paramyxovirus of simian virus 5 (SV5) VSD, ventricular septal defect WT, wild-type
xii XAP2, hepatitis B virus X-associated protein XRE, xenobiotic response element
xiii
Analysis of the Role of bHLH/PAS Proteins in Aryl Hydrocarbon Receptor Signaling
Edward J. Dougherty
ABSTRACT
The aryl hydrocarbon receptor (AHR) is a basic helix-loop-helix PER/ARNT/SIM
(bHLH-PAS) transcription factor that binds ligands typified by 2,3,7,8- tetracholordibenzo-p-dioxin, translocates to the nucleus, dimerizes with the aryl hydrocarbon nuclear translocator (ARNT) and associates with specific cis xenobiotic response elements to activate transcription of genes involved with xenobiotic metabolism.
AHR-mediated signal transduction has been evaluated thoroughly in the C57BL/6J mouse model system. This model system, however, may not be the most accurate model for human comparisons as the AHRb-1 allele carried by C57BL/6J contains a point mutation that prematurely truncates the receptor at 805 amino acids, while the AHRb-2, rat, and human AHR all contain an additional 42-45 amino acids at their carboxy- terminus that have 70% identity. This carboxy-terminal region could be functionally significant and the analysis of AHR-mediated signal transduction in the rat, human, or other mouse strains may better represent the physiology of the AHR pathway.
ARNT is another member of the bHLH-PAS family of proteins that is essential in several distinct signal transduction pathways mediated by its dimerization with a variety of bHLH-PAS proteins. Several isoforms of ARNT have been identified in mammalian
xiv and aquatic species. While ARNT and ARNT2 exhibit >90% amino acid identity in the
bHLH and PAS domains, gene knock-out of either ARNT or ARNT2 results in
embryonic/perinatal lethality characterized by distinct phenotypes. This suggests that
neither protein can compensate fully for the loss of the other. Since overlapping tissue
specific expression of ARNT and ARNT2 does exist, but neither ARNT can compensate
fully for loss of the other, this suggests that the two proteins have distinct functions in the
presence of various dimerization partners. Thus, the focus of these studies is to examine the discrepancies between the rat, human, or AHRb-2 possessing the extended carboxy-
terminal region and that of the AHRb-1 and also to examine the role of both ARNT and
ARNT2 during AHR-mediated signal transduction.
xv
Chapter One
Introduction
1.1 AHR and ARNT History
The aryl hydrocarbon receptor (AHR) was initially described as a heritable simple autosomal dominant trait, designated the Ah locus that conferred the ability to induce aryl hydrocarbon hydroxylases by polycyclic aromatic hydrocarbons in the C57BL/6N mouse
(Gielen et al., 1972). Response to polycyclic hydrocarbons, as measured in vitro by increased rate of formation of benzo[a]pyrene metabolites through an increase in aryl hydrocarbon hydroxylase activity had been determined by administration of aromatic hydrocarbon ligands such as 3-methylcholanthrene and benzo[a]pyrene. This response was not seen in the DBA/2N, DBA/2J, AKR/N, or NZW/BLN mice, now known to carry the AHRd low affinity ligand binding allele (Poland and Glover, 1974; Poland et al.,
1976; Poland et al., 1994). It was later described that 2, 3, 7, 8-tetrachlorodibenzo-p-
dioxin (TCDD) was approximately 30,000 times as potent as 3-methylcholanthrene as an
inducer for hepatic aryl hydrocarbon hydroxylase activity in the rat and could elicit
induction of aryl hydrocarbon hydroxylase activity in the previously “nonresponsive”
DBA/2N mice as well as in the "responsive" C57BL/6J mice (Gielen and Nebert, 1972).
Intraperitoneal or topical administration of TCDD to mice also led to an induction of aryl hydrocarbon hydroxylase activity and the formation of new cytochrome P1-450 in the
1 liver, bowel, lung, kidney and skin as well as induction of 7-ethoxycoumarin O-
demethylase (EROD) activity in the liver and kidney and hepatic p-nitroanisole O-
demethylase and 3-methyl-4-methylaminoazobenzene N-demethylase activity to a similar
magnitude in either the "nonresponsive" or “responsive” mice (Poland and Glover, 1974).
Thus, it was demonstrated that the genetically nonresponsive mice had the structural and
regulatory genes necessary for aryl hydrocarbon hydroxylase induction, but possessed a
defect that failed to recognize less potent inducers.
The regulatory product of the Ah locus was then shown to be a soluble protein that specifically bound [3H]-TCDD with high affinity with rank-ordered binding affinities
for various ligands that correlated to their relative potencies to induce aryl hydrocarbon
hydroxylase activity (Legraverend et al., 1982; Poland et al., 1976). Though the AHR
was initially described as cytosolic, it has since been shown that the subcellular location
of the endogenous complex may be cell/receptor dependant and the AHR has been found
in the cytoplasm and nucleus and may exhibit dynamic nucleocytoplasmic shuttling
(Pollenz and Dougherty, 2005; Pollenz et al., 1994). In any case, there is no apparent
difference in the ability of the AHR to serve as a ligand-activated transcription factor
from either subcellular compartment.
While the AHR has been shown to be activated by a variety of polycyclic and
halogenated hydrocarbons, the identity of an endogenous ligand remains controversial,
though several endogenous and alternate exogenous ligands have been suggested
including: 2-(1'H-indole-3'-carbonyl)-thiazole-4-carboxylic acid methyl ester (ITE), a
compound found predominantly in the preliminary testes and lung, equine estrogen
equilenin, tryptophan products, dietary polyphenols, prostaglandins, indirubin and indigo,
2 biliverdin and bilirubin (Adachi et al., 2001; Gouedard et al., 2004; Henry et al., 2006;
Jinno et al., 2006; Oberg et al., 2005; Phelan et al., 1998a; Seidel et al., 2001; Song et al.,
2002; respectively). However, while these endogenous compounds have been shown to bind the AHR, little has been done to assess whether the AHR functionally binds these compounds in vivo.
The aryl hydrocarbon receptor nuclear translocator (ARNT) protein was initially proposed to function in the nuclear translocation of the aryl hydrocarbon receptor (AHR) protein. This assumption was based on the observation that mutant hepatoma cells deficient in AHR-mediated induction of the target gene, cytochrome P450 1A1
(CYP1A1), did not exhibit an increase of AHR in nuclear extracts following treatment with ligand as was seen in wild-type cells, and this lack of nuclear accumulation of AHR in response to ligand could be rescued by genomic DNA coding for ARNT (Hankinson,
1979; Hoffman et al., 1991; Legraverend et al., 1982; Okey et al., 1980; Reyes et al.,
1992; Whitlock and Galeazzi, 1984). These studies also established that ARNT was a basic helix-loop-helix protein, as was the AHR, and that ARNT served as a heterodimerization partner for the AHR and was a necessary component of the DNA binding AHR complex (Burbach et al., 1992; Ema et al., 1992; Hoffman et al., 1991;
Reisz-Porszasz et al., 1994; Reyes et al., 1992). Using immunohistochemical techniques, it was later determined that ARNT was constitutively nuclear and played no role in ligand-mediated translocation of the AHR since nuclear translocation of the AHR occurred following ligand binding even in ARNT deficient cells, and that this translocation was not detected in nuclear extracts since the AHR lacked the ability to bind
DNA in the absence of ARNT (Holmes and Pollenz, 1997; Pollenz et al., 1994).
3 Since the initial cloning of AHR and ARNT, it has been demonstrated that the
both proteins are members of the class VII helix-loop-helix superfamily of transcriptional
regulators, which are classified by the presence of a basic region prior to the HLH
domain (bHLH) as well as the PER/ARNT/SIM (PAS) domain (Sablitzky 2005). These
bHLH/PAS proteins can be further divided into two main groups: one which contains the
ARNT proteins: ARNT (HIF-1β), ARNT2, BMAL1 (ARNT3, MOP3, JAP3, ARNTL1,
TIC), and BMAL2 (ARNT4, ARNTL2, MOP9) and one which contains the hypoxia-
inducible factor 1 family (HIF-1α, HIF-2α, HIF-3α), the hypoxia-like factor (HLF), the
single-minded proteins (SIM1, SIM2), and the AHR (Barrow et al., 2002; Drutel et al.,
1996; Hirose et al., 1996; Hogenesch et al., 1997; Hogenesch et al., 2000; Ikeda and
Nomura, 1997; Ikeda et al., 2000; Maemura et al., 2000; Okano et al., 2001; Takahata et
al., 1998; Wolting and McGlade, 1998) (Tables 1.1, 1.2). These proteins generally
appear to function through heterodimerization of members of the first group with
members of the second group, though homodimerization of ARNT has also been
suggested. Together, these proteins are involved with the regulation of a variety of
signaling pathways ranging from angiogenesis, vasculogenesis, xenobiotic metabolism,
and hypoxic response to developmental pathways wherein each dimer associates with
specific cis acting DNA elements to regulate target genes (Crews, 1998; Ema et al.,
1996a; Ema et al., 1997; Furness et al., 2007; Kewley et al., 2004; Whitelaw et al., 1993)
(Figure 1.1).
Interestingly, while many of ARNT's interactions with various dimerization
partners have been established both in vitro and in vivo, as have the more recently
4
Table 1.1: Summary of ARNT cloning.
ARNT Type, the ARNT group to which each clone has the highest identity; Clone, the original name ascribed to the individual cDNA during the cloning of each gene; Species, the origin of the cDNA library or EST database used in the generation of probes. * Tian,H., Russell,D.W. and McKnight,S.L., Unpublished Direct Submission to NCBI database 10-JUN-1996.
5
Table 1.2: Amino acid identity comparisons of murine ARNT types and across ARNT isoform domains.
Numbers represent the percent identities between compared proteins using the CLUSTALW (Slow/Accurate, Gonnet) method (MEGALIGN, DNAstar, Madison, WI) with a PAM250 residue weight table. A, Identity among murine ARNT types. Accession numbers were: mARNT (U10325), mARNT2 (D63644), mARNT3 (AB014494), mARNT4 (AY005163). B, Identity among murine ARNT protein domains.
6
Figure 1.1: Overview of known cellular pathways requiring ARNT or ARNT2 in mammals. Protein•protein interactions between the ARNT proteins and their dimerization partners that have been functionally and biochemically established both in vitro and in vivo are indicated by solid thick arrows, while interactions that have been established either in vitro or in vivo are indicated by solid thin lines, and presumed interactions are shown with dashed arrows. Also shown are the gene regulatory sequences to which the heterodimers/homodimers bind: XRE: xenobiotic response element; HRE: Hypoxia response element; CME: central midline enhancer, with the bolded sequence representing the portion of the regulatory sequence that is recognized by ARNT/2. Below these sequences are the some of the target genes of the heterodimers/homodimers that have been assessed and their overall physiological role.
7 discovered ARNT3 and ARNT4 proteins’ roles in circadian rhythm regulation with
Clock, the physiological roles of ARNT2 remain less clear.
1.2 AHR and ARNT Domains
These proteins are modular, and their common characteristics are the bHLH domains and the PAS domains that are distal to the bHLH motif. Through mutation and deletion analyses, the basic domain of these proteins has been shown to be critical for
DNA binding whereby 2-4 basic residues appear to make direct contact with DNA (in the
AHR and ARNT respectively) while other basic residues may make contact with the phosphodiester backbone (Bacsi and Hankinson, 1996). In the AHR, tyrosine 9 (Y9) and arginine 39 (R39) appear to be essential for AHR contact with the DNA, while other residues of the basic region maintain α-helical structure and make contact with the phosphodiester backbone (Bacsi and Hankinson, 1996). Interestingly, the basic region of
ARNT more closely resembles the basic region found in leucine zipper-HLH proteins such as TFEB, USF, c-Myc and Max than the basic region of the AHR (Sogawa et al.,
1995). This basic region has been shown to be critical for binding to the E-box half site
GTG, whereby during AHR signaling, four residues are critical for contact with DNA
(H94, E98, Rl0l, and R102), which correspond in position to essential residues found in the bHLH proteins USF and MAX and are conserved across all characterized ARNT proteins (Bacsi and Hankinson, 1996; ED, unpublished observations). Though these mutational analyses focused on XRE binding ability of the AHR•ARNT complex as the endpoint and have not included other ARNT types (ARNT2/3/4), each ARNT protein appears to associate with the GTG half-site regardless of its dimerization partner, while
8 the partner provides the specificity for enhancer binding (Ebert et al., 1995; Fujisawa-
Sehara et al., 1986; Hapgood et al., 1989; Paul and Ferl, 1991; Semenza et al., 1994;
Swanson et al., 1995; Wharton and Crews, 1993; Wharton et al., 1994).
The HLH motif consists of two amphipathic α-helices connected by a variable loop and appears to serve as a surface for protein-protein interactions and is also responsible for positioning the α-helix of the basic domain within the major groove of B-
DNA to allow for specific interactions between the DNA binding protein and the target response elements (Anthony-Cahill et al., 1992). Deletion of either helix has been
shown to abolish dimerization (Reisz-Porszasz et al., 1994).
The PAS domain, named for the first three proteins found to exhibit this conserved sequence (PER/ARNT/SIM) is found in numerous proteins that are involved with detection of environmental cues and stress response (Gu et al., 2000). However,
while this domain has a high level of identity between the ARNT and ARNT2 proteins,
there is less than 20% identity in this region between the PER, ARNT, and SIM proteins
themselves (Gu et al., 2000; Huang et al., 1993). Therefore, the PAS domain is characterized by an approximately 100 amino acid hydrophobic-rich region, which is further characterized by a highly degenerate ~50 amino acid repeat with an invariant phenylalanine, histidine, and aspartic acid at positions 1, 41, and 44 respectively within the PAS domain repeat (Huang et al., 1993; Nambu et al., 1996). Both the AHR and all
ARNT proteins appear to contain two PAS domains termed PAS A and PAS B. In each case, the PAS domain appears to provide specificity for dimerization, may also provide the specificity for the protein binding of the adjacent HLH domain, and may enable transcription following DNA binding (Chapman-Smith et al., 2004; Dolwick et al., 1993;
9 Jain et al., 1994; Lindebro et al., 1995; Pongratz et al., 1998; Reisz-Porszasz et al., 1994;
Whitelaw et al., 1994; Zelzer et al., 1997). In addition, it has been suggested that the
PAS A domain is critical for DNA binding and protein-protein interactions and
contributes directly to AHR•ARNT XRE binding (Chapman-Smith et al., 2004;
Chapman-Smith and Whitelaw, 2006; Pongratz et al., 1998). Similarly, the PAS B
domain has been implicated in contributing to the heterodimerization potential and stability of ARNT and HIF-1α/HIF-2α through interactions occurring via the PAS B
central β-sheet, and mutations in this region have been suggested to affect the transcriptional ability of the overall heterodimer (Card et al., 2005; Erbel et al., 2003).
For the AHR, the PAS domain appears to have additional functions including Hsp90-,
XAP2-, and ligand-binding (Dolwick et al., 1993; Jain et al., 1994; Reisz-Porszasz et al.,
1994; Whitelaw et al., 1994). It has also been suggested that the PAS domain for the
AHR contains a repressor region, since Ah receptor lacking this region exhibits
constitutive activity (CAR mutant), though it is now known that it is the presence of
Hsp90 bound to the receptor by the PAS domain that prevents DNA binding from
occurring (Antonsson et al., 1995; Heid et al., 2000; Whitelaw et al., 1994).
The AHR and ARNT proteins also each contain a putative nuclear localization
site (NLS). While the AHR may be cytoplasmic, nuclear, or shuttling through the
cytoplasm and nucleus based on the receptor species, ARNT appears to be constitutively
nuclear (Pollenz and Dougherty, 2005; Pollenz et al., 1994). In the AHR, the NLS is a single basic region (RKRRKPVQK) located at the NH-terminus that is capable of recognizing the importin proteins necessary for nuclear translocation following ligand
binding, however, the AHRb-1 appears to exist in a conformation whereby the NLS is
10 exposed only after a conformational change following ligand binding, resulting in the nuclear import of the AHR (Holmes and Pollenz, 1997).
The AHR also contains a leucine-rich nuclear export site (NES) located within the
HLH helix 2 region amino acids 63-71 (LDKLSVLRL). Leucine to alanine scanning
mutagenesis of leucines 66 and 71 of the putative AHR NES revealed that alanine
substitutions at these residues resulted in a protein with reduced function in dimerization
to ARNT and binding to DNA (Pollenz and Barbour, 2000). In these same studies,
mutagenesis of leucine 69 did not impact function with regard to ARNT dimerization or
DNA binding, yet resulted in a protein that was incapable of nuclear export suggesting
that AHR defective in nuclear export can remain functional during ligand mediated gene
induction.
The ARNT proteins also each contain a putative nuclear localization site (NLS)
and appear to be localized to the nucleus, though it has been suggested that partial
cytoplasmic localization may be occurring in certain cell types or during developmental periods (Eguchi et al., 1997; Holmes and Pollenz, 1997; Ikeda and Nomura, 1997; Ikeda et al., 2000; Jain et al., 1998; Pollenz et al., 1994; Sadek et al., 2000; Schoenhard et al.,
2002). In the ARNT and ARNT2 proteins, this NLS is bipartite, consisting of two basic domains separated by an acidic region spacer (RXXKRRSGXDFDDEXXXXXKFXR)
that are capable of interacting with the proteins of the nuclear pore targeting complex for
efficient nuclear localization (Eguchi et al., 1997). Deletion of this region, however,
results in a cytosolic protein that remains capable of interacting with the AHR (Holmes
and Pollenz, 1997; Song and Pollenz, 2003). ARNT3 and ARNT4 both contain a short
basic NLS (RKRK), though the ARNT3 protein also appears to contain two nuclear
11 export sites (NES) located in the PAS domains that allow for dynamic nucleocytoplasmic shuttling of this ARNT isoform (Ikeda et al., 2000; Kwon et al., 2006; Schoenhard et al.,
2002).
Deletion and mutational analysis has also defined regions in the carboxy-terminus of the AHR that are important in transactivation including an acidic domain, a proline- rich domain, and a serine-rich region as well as a glutamine rich region in ARNT (Jain et al., 1994; Ko et al., 1996; Ko et al., 1997; Li et al., 1994; Pollenz et al., 2005; Whitelaw et al., 1994) (Figure 1.2). In the AHR, loss of the proline and serine rich regions by truncation reduces the transactivation ability of the AHR as measured by a marked reduction of CYP1A1 protein following treatment with TCDD compared to full length
AHR, while additional loss of the acidic region ablated all apparent transactivation ability
(Pollenz et al., 2005).
The carboxy-terminal portion of the ARNT, ARNT2, and ARNT3 proteins also appears to contain a transactivation domain and a glutamine-rich region has been demonstrated to have functional significance in both ARNT and ARNT2, while ARNT3 appears to have an acidic region important to its transactivation potential when partnered to HIF-1α (Corton et al., 1996; Li et al., 1994; Takahata et al., 1998). In contrast, the carboxy-terminal portion of ARNT4 has been shown to be dispensable for luciferase reporter transcription in the presence of Clock (Schoenhard et al., 2002). In the case of
ARNT, reports of the functional role of the carboxy-terminal portion as a transactivation domain during AHR signaling have shown conflicting results. Studies performed by Li et al. (1994) demonstrated that deletion of the glutamine-rich region of the ARNT
12
Figure 1.2: Domain structures of the AHR (b1 allele) and ARNT. Thin horizontal lines indicate domain regions; bold lines indicate functional regions for the AHR. NLS, nuclear localization signal; NES, nuclear export signal; b, basic region; HLH, helix-loop- helix; PAS A and PAS B, PER/ARNT/SIM homology regions; PAC, PAS associated carboxy-terminal region; TAD, transactivation domain. Numbers under each box represent amino acids.
13
transactivation domain (TAD) by truncation resulted in decreased CAT activity
controlled by a TCDD-responsive enhancer when ARNT deficient (c4) hepatoma cells
were cotransfected with ARNT and reporter vectors. In contrast, studies by Corton et al.
(1996) using a Gal4 hybrid approach demonstrated that loss of the ARNT TAD through
truncation resulted in no change in overall a CAT activity reporter under the control of
multiple Gal4 upstream activating sequences. Thus, in contrast to the studies by Li et al.,
the studies by Corton et al. suggest that the ARNT TAD has minimal impact on
transactivation of the AHR•ARNT complex, and suggest that during AHR signaling, the
AHR itself provides the functional TAD within the AHR•ARNT complex while the
ARNT TAD is silent (Corton et al., 1996; Li et al., 1994).
However, the studies by Corton et al. (1996) also suggested that the strength of
the ARNT TAD may be cell specific, since the ARNT TAD appeared to exhibit the same properties as the well-characterized VP16 TAD in COS-1 cells, but exhibited ~10 fold weaker activity in hepatoma ARNT deficient (c4) cells. In separate studies, the AHR and
ARNT TAD showed equal transactivation ability in Hepa-1, CHO, and COS-7 cells,
while the ARNT TAD appeared to be much weaker than the AHR TAD in HepG2 cells
when recombined with the glucocorticoid receptor DNA binding domain and used to
drive reporter activity controlled by a glucocorticoid receptor responsive enhancer
(Whitelaw et al., 1994). Since the studies by Li et al. (1994) also employed the ARNT
deficient c4 cells, cell-specificity of the ARNT TAD cannot explain the discrepancies
between these and other studies and since all studies described herein employed
transiently transfected cells with no immunohistochemistry to support population wide
14 transfection, more thorough studies were later conducted in this area using stably transfected cells (Ko et al., 1996). Using retrovirally transfected truncated ARNT cDNA in ARNT deficient (BPrc1, LA-II) mouse hepatoma cells, these studies showed that in vivo, the ARNT TAD is dispensable for TCDD-induced CYP1A1 induction and that the
AHR when dimerized with ARNT supplies the dominant TAD (Ko et al., 1996; Whitlock and Galeazzi, 1984). While reporter studies have also been performed using ARNT2-4, similar evaluations using stably transfected cells have not yet been performed using these
ARNT proteins nor have such studies examined the contribution of the ARNT TAD in other ARNT requiring pathways (Cowden and Simon, 2002; Hirose et al., 1996;
Takahata et al., 1998).
1.3 AHR Proteins
AHR-mediated signal transduction has been evaluated thoroughly in the
C57BL/6J mouse model system that carries the AHRb-1 allele. While this allele has been the most studied, it may not be the most appropriate choice for study as it is clearly the outlier amongst the AHR alleles as well as amongst Ah receptors from other species including the rat and human. The AHRb-1 is found in the C57, C58, and the MA/My strains where it exists as a ~95kDa (805 amino acid) cytosolic receptor, which does not appear to be shuttling through the nucleus endogenously (Poland et al., 1994; Pollenz and
Dougherty, 2005). The AHRb-2 found in the BALB/cBy and C3H strains exists as a
~104kDa (848 amino acid) receptor that is primarily nuclear, though capable of nucleocytoplasmic shuttling. Additionally, the AHRb-3 is a~105kDa (883 amino acid) receptor allele found in mus caroli, mus spretus, and the MOLF/Ei strains, while the
15 AHRd allele found in the AKR, DBA/2, and 129 strains is a ~104kDa (848 amino acid)
receptor. Collectively, these alleles differ by only 8 point mutations in the initial 805
amino acid open reading frame and by the presence of additional 43-78 amino acids in
the carboxy-terminus of the Ahb2, Ahb3, and Ahd receptors. Of these 8 point mutations
found amongst the murine AHR alleles, the AHRb-1 accounts for more than half, resulting
in a receptor that binds Hsp90, XAP2, and ligand more avidly and is more stable than the
other mouse allelic receptors (Figure 1.3, Table 1.3). These discrepancies support the
idea that the AHRb-1 is the outlier. The amino terminal halves of these AHR alleles are
identical except for 1 amino acid change in the second PAS box (isoleucine to methionine
at position 324) with the remaining mutations present in the carboxy-terminus.
Interestingly, these mutations are generally outside of the characterized functional
domains with the exception of the PAS box mutation (324) and a serine to asparagine
mutation in the acidic portion of the transactivation domain, yet these receptors have a
varying degree of activity following ligand binding. Though the AHRd differs from the
AHRb-2 by only three amino acids, it is nonresponsive to low affinity ligands, unlike the
AHRb-2. This loss of function has been attributed to the alanine to valine mutation at
position 375, which occurs immediately distal to the ligand-binding domain (Poland et
al., 1994). Furthermore, the AHRb-1 is truncated when compared with Ah receptors
found in other mammalian species, containing a point mutation that prematurely truncates
the receptor at 805 amino acids, while the AHRb-2, rat, and human AHR all contain an additional 42-45 amino acids at their carboxy-terminus that have 70% identity (Figure
1.3). As this additional carboxy-terminal sequence is present in the other AHR alleles as
16
Figure 1.3: Protein sequence alignment of various AHR c-termini. Carboxy-terminal alignment of AHR proteins from amino acid position 800 to termination in Mus musculus C57BL/6J (Ahb1 allele); Mus musculus BALB/c (Ahb2 allele); Mus musculus DBA/2J (Ahd allele); the chicken Gallus gallus; Homo sapiens, and Rattus norvegicus. Proteins were aligned by using the CLUSTALW (Slow/Accurate, Gonnet) method (MEGALIGN, DNAstar, Madison, WI) using a PAM250 residue weight table.
17
Table 1.3: Amino acid changes among the first 805 amino acids of murine AHR alleles.
Full length AHR proteins were aligned from the murine strains: C57BL/6J (Ahb-1 allele); C3H (Ahb-2 allele); MOLF/Ei (Ahb-3 allele); and DBA/2J (Ahd allele). Proteins were aligned by using the CLUSTALW (Slow/Accurate, Gonnet) method (MEGALIGN, DNAstar, Madison, WI) using a PAM250 residue weight table. Point mutations occurring in the first 805 amino acids between proteins are shown in bold. Numbers above each column represent the amino acid location within the open reading frame.
18 well as across species and is reasonably conserved in sequence, it is reasonable to assume
that this sequence may be functionally relevant.
Several orthologs of the AHR have also been identified in other mammalian, invertebrate, aquatic and avian species including Mesocricetus auratus, Gallus gallus,
Cavia porcellus, Cricetulus griseus, the rabbit Oryctolagus cuniculus, the seal Phoca
siberica, the North Atlantic right whale Eubalaena glacialis, the clam Mya arenaria, the
zebra mussel Dreissena polymorpha, the sea lamprey Petromyzon marinus, the common
tern Sterna hirundo, the common cormorant Phalacrocorax carbo, the Asian malaria
mosquito Anopheles stephensi, Drosophila menologaster, and Carnorhabditis elegans
(Bennett et al., 1996; Butler et al., 2001; Emmons et al., 1999; Hahn et al., 2004; Jensen
and Hahn, 2001; Karchner et al., 1999; Kim and Hahn, 2002; Korkalainen et al., 2001;
Meyer et al., 2003; Powell-Coffman et al., 1998; Roy and Wirgin, 1997; Satoh et al.,
2003; Takahashi et al., 1996; Tanguay et al., 1999; Walker et al., 2000; Yasui et al.,
2004; Figure 1.4). Many more AHR proteins have also been predicted in silico based on
genomic sequencing data (www.ensembl.org), though have not yet been evaluated in
vivo. Interestingly, the function of the AHR in response to xenobiotics appears to be
well-conserved across many of these species. However, the AHR proteins from several
invertebrate species appear to be developmentally regulated, lacking constitutive
expression and exhibiting a lack of ligand binding (Butler et al., 2001; Emmons et al.,
1999; Powell-Coffman et al., 1998).
19 Mus musculus C3H Mus musculus CBA/J Mus musculus BALB Mus musculus A/J Mus musculus CAST Mus musculus DBA/2J Mus musculus SJL/J Mus musculus MOLF/Ei Mus spicilegus Pancevo Mus spretus Mus Musculus C57Bl Mus caroli Rattus norvegicus Cricetulus griseus Mesocricetus auratus Eubalaena glacialis Megaptera novaeangliae Delphinapterus leucas Phoca siberica Homo sapiens Cavia porcellus Oryctolagus cuniculus Phalacrocorax carbo Sterna hirundo Gallus gallus Xenopus laevis Danio rerio 1B Petromyzon marinus Oncohynchus mykiss A Oncorhynchus mykiss B Salmo salar 2B Microgadus tomcod Danio rerio 2 Aedes aegypti Anopheles stephensi Drosophila melanogaster (SS) Dreissena polymorpha Mya arenaria Caenorhabditis elegans 1 Ciona intestinalis 140.3 140 120 100 80 60 40 20 0
Amino acid substitutions x100
Figure 1.4: AHR protein dendogram. Full-length proteins were aligned by using the CLUSTALW (Slow/Accurate, Gonnet) method (MEGALIGN, DNAstar, Madison, WI) using a PAM250 residue weight table. The horizontal distance to the subclusters corresponds to degree of amino acid substitutions among members. References are as follows: (Abnet et al., 1999; Bennett et al., 1996; Butler et al., 2001; Carver et al., 1994a; Emmons et al., 1999; Hahn et al., 2004; Jensen and Hahn, 2001; Karchner et al., 1999; Kim and Hahn, 2002; Korkalainen et al., 2001; Meyer et al., 2003a; Nene et al., 2007; Powell-Coffman et al., 1998; Roy and Wirgin, 1997; Satoh et al., 2003; Takahashi et al., 1996; Tanguay et al., 1999; Walker et al., 2000; Yasui et al., 2004)
20 1.4 AHR-Mediated Signaling
The current model of AHR-mediated signal transduction in mammals
hypothesizes that the AHR in its unliganded state exists in a ~280 kDa multimeric
complex consisting of two molecules of Hsp90, the immunophilin-like protein XAP2
(XAP, AIP, ARA9), and p23 (Chen and Perdew, 1994; Gu et al., 2000; Petrulis and
Perdew, 2002; Pollenz et al., 2002; Whitlock, 1999). Once activated by ligand, the AHR
translocates to the nucleus where it dissociates from the Hsp90 complex and forms a
dimer with ARNT via its bHLH domains by an unknown regulatory step that may
involve a phosphorylation event on the AHR following nuclear import (Heid et al., 2000;
Pollenz et al., 1994; Pongratz et al., 1998). The Hsp90-free AHR•ARNT complex is then
capable of binding the non-canonical E-box like consensus xenobiotic response element
(XRE, DRE, AHRE, ARE) 5'-T(T/A)GCGTG-3', associating with numerous coactivators such as members of the 160 family (SRC-1, SRC-2, SRC-3), RIP140, CoCoA, and p300, and regulating the transcription of phase I metabolizing enzymes including the monooxygenase CYP1A1, NAD(P)H:quinone oxidoreductase, alcohol dehydrogenase,
and aldehyde dehydrogenase, along with phase II conjugation enzymes including
glutathione-S-transferase Ya, UDP-glucoronosyltransferase, and N-acetyltransferases that
are important in the metabolism of drugs, steroids, and toxins (Denison et al., 1988;
Hankinson, 2005; Hapgood et al., 1989). The AHR is then targeted for degradation by an
unknown mechanism that terminates at the 26-S proteasome (Pollenz, 1996; Pollenz,
2002). It must be noted, however, that this description is simplified and there are several
unknowns throughout this pathway such as the regulation of each step in the pathway, the
21
Figure 1.5: Schematic of the AHR signaling pathway. In brief, ligands such as halogenated aromatic hydrocarbons (HAHs), polychlorinated biphenyls (PCBs), polyaromatic hydrocarbons (PAHs) enter a cell, bind to the AHR complex leading to a conformational change of the receptor, which then translocates to the nucleus and heterodimerizes with ARNT. The AHR•ARNT complex then associates with xenobiotic response elements with the core sequence 5’-TNGCGTG to regulate phase I and phase II metabolizing enzymes, typified by CYP1A1. The AHR is then targeted for degradation via the 26S proteasome.
22 involvement of specific coactivators and coregulators and the precise mechanism behind
the degradation of the AHR (Figure 1.5).
1.5 AHR Associated Proteins
The heat shock protein 90 family (Hsp90a and Hsp90b in the human, Hsp84 and
Hsp86 in the mouse) is a group of highly conserved and highly expressed chaperone
proteins implicated in the maintenance of proper protein folding and prevention of
protein aggregation with a variety of intracellular receptors (Hickey et al., 1989; Moore et
al., 1989). The AHR appears to have equal affinities for either Hsp90 type (Chen and
Perdew, 1994; Czar et al., 1994), and this association is essential for agonist-induced
AHR signaling in vivo since Hsp90's association with the AHR, as well as with the
glucocorticoid receptor (GR) and MyoD, has been correlated to the receptor's ability to
maintain proper conformation and to bind ligand (Carver et al., 1994b; Evans, 1989;
Hutchison et al., 1994; Schmidt et al., 1996; Shue and Kohtz, 1994; Whitelaw et al.,
1995). Preceding ligand binding, Hsp90 appears to hold the AHR in a conformation
capable of binding ligand while repressing the ability of the receptor to bind DNA.
However, it has also been suggested that Hsp90 is necessary only for the initial folding of
the AHR, since in the presence of salt conditions that result in the disassociation of the
existing AHR from the Hsp90 complex leads to Hsp90-free AHR that remains capable of binding ligand (Heid et al., 2000; Phelan et al., 1998b; Whitelaw et al., 1994). The affinity of Hsp90 for the AHR has been correlated to the phosphorylation status of three serine residues within Hsp90 (S225 and S254 for Hsp90b and S230 in Hsp90a
[equivalent to S225 of Hsp90b]) that occur within the charged linker domain that
23 associates with the AHR (Chen and Perdew, 1994; Ogiso et al., 2004). Interestingly, in
these studies, mutation of the potential phosphorylation sites S225 and S254 to alanines
increased both the affinity of AHR for Hsp90 as well as the transactivation activity of the
AHR as measured by a luciferase reporter following treatment with 3-methycholanthrene
compared to glutamine substitutions at these residues, suggesting that phosphorylation
status of Hsp90 may play a role in modulating AHR signaling (Ogiso et al., 2004).
XAP2 is a ~37kDa protein that contains three regions of homology to the
tetratricopeptide (TPR) motif found in the FKBP12 and FKBP59 and FKBP52
immunophilins, which act as molecular chaperones in steroid receptor signaling, targeting
steroid receptors to the nucleus. Unlike immunophilins, however, XAP2 does not
associate with immunosuppressive drugs like FK506 (Tacrolimus, Fujimycin), but
appears to associate with the AHR through this drug-binding (FK) domain and with
Hsp90 via its TPR domain (Carver et al., 1998; Ma and Whitlock, 1997). Also, XAP2
does not appear to be involved in maturation of the AHR, but can modulate AHR activity in that overexpression of XAP2 has been suggested to increase response to aryl hydrocarbon hydroxylase inducers without appearing to alter the affinity of the AHR for
TCDD (Carver et al., 1998; Ma and Whitlock, 1997). Reduction of XAP2 by siRNA appears to result in the transformation of the AHR from a statically cytoplasmic receptor
to one capable of undergoing dynamic nucleocytoplasmic shuttling (Pollenz et al., 2006).
Furthermore, following ligand activation of the AHRb-1, XAP2 remains associated with
the receptor following nuclear translocation and may serve to inhibit transformation of
the AHR complex to the ARNT dimerized form (Pollenz and Dougherty, 2005; Pollenz et
al., 2005). Phosphorylation of XAP2 has not been implicated in modulation of AHR
24 function; however, mutation of S53 to alanine appears to inhibit the ability of XAP2 to
enter the nucleus (Dull et al., 2002).
P23 is a 23-kilodalton Hsp90 binding protein that appears to serve several roles in
AHR signaling including stabilization of the AHR-Hsp90 complex, assistance of nuclear
translocation, and recruitment of XAP2 to the AHR complex (Kazlauskas et al., 2001).
Similar to its role in glucocorticoid receptor signaling, p23 appears to be important for
proper maturation of the latent AHR complex, allowing for a stable complex capable of
binding ligand, heterodimerizing with ARNT and subsequently binding DNA. These
hypotheses stemmed from observations demonstrating that loss of p23 from crude
reticulocyte extracts resulted in a diminished DNA binding shift in electrophoretic
mobility shift assays, while the presence of p23 appeared to enhance AHR•ARNT
heterodimerization in an Hsp90-dependent manner (Shetty et al., 2003). This
enhancement by p23 was also seen by others and therefore, p23 has been suggested to be
an important enhancer of AHR signaling, though ultimately not necessary for AHR
function (Cox and Miller, 2003). Similar results were also seen with another Hsp90
binding protein cyclophilin 40 (CyP40), suggesting that other proteins may also be involved with the formation/stability of the latent AHR complex (Shetty et al., 2004).
1.6 Degradation of the AHR
Agonist binding to the AHR results in degradation of the receptor following DNA
binding, but not of the complex’s components: ARNT, XAP2, or Hsp90 (Cioffi et al.,
2002; Giannone et al., 1998; Pollenz, 2002; Roman and Peterson, 1998; Roman et al.,
1998). Following TCDD binding, the AHRb-1 is rapidly depleted by >60-80% within 4-6
25 hours of treatment in numerous cell culture models and does not return to basal levels as long as ligand is present in the media (Pollenz, 1996; Reick et al., 1994). This
degradation is even more rapid with the Ahb-2, human or rat receptors, depleting by 90-
100% within 2 hours of treatment (Pollenz and Dougherty, 2005). In either case, this
degradation has been demonstrated to occur via the 26S proteasome complex since pre-
treatment with the proteasome inhibitors MG-132 or lactacystin prior to treatment with
agonist blocks degradation of the receptor, while pre-treatment with inhibitors of calpain, serine, or cysteine proteases nor lysosomal proteases cannot (Davarinos and Pollenz,
1999; Ma and Baldwin, 2000; Wentworth et al., 2004).
As such, several studies have suggested that the AHR is ubiquitinated, though none demonstrated definitive evidence of such events and at this time no E3 ligase has been demonstrated to be involved in the degradation of the AHR (Ciechanover, 2005;
Kazlauskas, 2000; Ma and Baldwin, 2000). Furthermore, ligand-dependant degradation can also be blocked by pre-treatment with the transcription inhibitor actinomycin D (AD) or the translation inhibitor cycloheximide (CHX) without affecting nuclear localization or
DNA binding of the receptor, suggesting that both active transcription and translation are necessary for ligand-induced degradation of the AHR (Pollenz et al., 2005).
Carboxy-terminally truncated Ah receptors as well as those defective in DNA binding or ARNT dimerization similarly exhibit a low level of degradation following treatment with TCDD albeit a much lower level of degradation than that seen in wild type; however, this degradation cannot be blocked by either AD or CHX, suggesting that this loss is not representative of the typical degradation seen following ligand binding with the wild-type receptor (Pollenz et al., 2005).
26 It also appears, however, that the AHR is capable of being degraded by a second,
distinct pathway that does not require ligand binding. This pathway is typified by
treatment with geldanamycin (GA), a benzoquinone ansamycin capable of binding to the
ATP-binding pocket of Hsp90 and thereby likely altering the conformation of the Hsp90
associated AHR to allow for nuclear translocation of the receptor and its subsequent
degradation without disruption of the AHR complex itself and, importantly, without
DNA binding or subsequent gene induction (Chen et al., 1997; Meyer et al., 2003b; Song
and Pollenz, 2002). This second pathway is characterized by a rapid and robust
degradation profile (>80% within 1 hour) that also appears to occur via the 26S
proteasome and can be blocked by MG-132 or lactacystin (Song and Pollenz, 2002).
Though treatment with GA results in a rapid nuclear translocation of the AHR as does
ligand-binding, further studies suggest this mode of degradation can occur in either the
cytoplasm or nucleus (Song and Pollenz, 2002). Additionally, degradation of the AHR
via GA cannot be blocked by treatment with either AD or CHX, suggesting that multiple
mechanisms exist for the degradation of the receptor, though both terminate at the 26S
proteasome (Pollenz et al., 2005).
1.7 Consequences of AHR Degradation or Overexpression
Studies have shown that a single oral dose of TCDD can lead to sustained
depletion of AHR proteins in the liver, spleen, thymus, and lung in vivo and that such a
depletion correlates with reduction in TCDD-mediated reporter gene expression in mammalian culture cells following a second dose of TCDD (Pollenz et al., 1998). The importance of this loss is underscored by the variety of physiological defects seen in
27 AHR-/- mice (Andreola et al., 1997; Fernandez-Salguero et al., 1995; Gonzalez et al.,
1995; McDonnell et al., 1996). It is important to note that many of the phenotypes seen
in TCDD-treated mice are similar to those reported for AHR-/- mice that have not been
exposed to TCDD. Thus, this loss of AHR may contribute to some of the biological
effects of xenobiotics. Degradation of the AHR, therefore, which may be used as a
means of attenuating the transcriptional response, can have significant repercussions on
future signaling ability of the AHR pool as well as biological implications.
Conversely, blockage of degradation by CHX appears to result in potentiation of gene induction (superinduction), whereby genes regulated by the AHR are induced to a higher level and for a longer period of time (Ma and Baldwin, 2000; Pollenz and Barbour,
2000). Induction of CYP1A1 protein in response to TCDD has also been implicated in some subsets of the physiological defects associated with TCDD toxicity such as pericardial edema and reduced blood flow (Teraoka et al., 2003). Furthermore, constitutive activation of the AHR in the CAR mutant lacking a ligand-binding domain led to proliferation of stomach tumors and reduced life span (Andersson et al., 2002).
Together, these data suggest a role for the gene products of AHR-dependant signaling in mediating the toxic effects of TCDD.
1.8 Mammalian ARNT Isoforms
Several isoforms of ARNT have been identified in mammalian species and have been grouped into four ARNTs: ARNT (HIF-1B), ARNT2, ARNT3 (BMAL1, MOP3,
JAP3, ARNTL1, TIC) and ARNT4 (BMAL2, ARNTL2, MOP9, CLIF) (Table 1.1).
ARNT appears to be ubiquitously expressed, while ARNT2 has been described primarily
28 in the kidneys and central nervous system (Abbott and Probst, 1995; Aitola and Pelto-
Huikko, 2003; Hirose et al., 1996; Jain et al., 1998). A more recent study, however, has demonstrated that mRNA for both ARNT and ARNT2 is co-localized in many murine
peripheral organs and neuronally derived tissue, suggesting that the distribution of
ARNT2 may not be as restricted as previously described (Aitola and Pelto-Huikko,
2003). ARNT3 and ARNT4, on the other hand, were initially characterized by Northern
blot analysis as being expressed only in the skeletal muscle, heart, and brain, with high expression in the suprachiasmatic nucleus (SCN) (Hogenesch et al., 1997; Ikeda and
Nomura, 1997; Ikeda et al., 2000; Okano et al., 2001; Takahata et al., 1998), though other studies have reported more ubiquitous expression of ARNT3 including high levels detectable in the liver, kidney, lung, skeletal muscle pancreas, stomach, and testis, and a varied expression of ARNT4 differing between splice variants (Ikeda et al., 2000;
Schoenhard et al., 2002; Wolting and McGlade, 1998). Given that the ARNT3 and
ARNT4 proteins appear to be primarily involved with circadian rhythm signaling and are expressed in an oscillatory manner, it is likely that expression would be seen not only in the SCN, which acts as the central pacemaker, but also in the heart, kidney, liver, and other tissues that have peripheral clocks (Kohsaka et al., 2007; Maemura et al., 2007).
Additionally, since the expression of these proteins oscillates, it is expected that detection
of these proteins in vivo would differ based on the central clock time. Therefore, studies
were carried out to examine the time course of expression of both ARNT3 and ARNT4 in the zebrafish in several tissues (Cermakian et al., 2000). These studies revealed that mRNA for both ARNT3 and ARNT4 can be detected in the zebrafish brain, pineal gland, eye, heart, liver, spleen, and testis depending on the physiological time point evaluated,
29 suggesting a ubiquitous, but oscillatory expression pattern for these proteins in vivo that was asynchronous between ARNT3 and ARNT4. Since it has been suggested that expression of ARNT also oscillates on a circadian rhythm in several tissues, similar studies to those performed on ARNT3 and ARNT4 should be considered for ARNT and
ARNT2 to fully evaluate tissue distribution of these proteins in vivo (Garrett and
Gasiewicz, 2006; Richardson et al., 1998).
In contrast to mammalian systems, expression of ARNT2 is markedly different amongst several fishes and avian species; however, ARNT2 in Xenopus has also been characterized as being highly expressed in the brain and kidneys with low or absent levels in other tissues as has been described in mammalian systems. In contrast, the marine teleost, Fundulus heteroclitus, appears to predominantly express a form of ARNT that is
ARNT2-like (fhARNT2), resembling mouse ARNT2 (83%) more highly than mouse
ARNT (63%), and surprisingly, more highly than rainbow trout ARNT (rtARNTb, 54%) as well (Powell et al., 1999). These studies showed that fhARNT2 was detectable in the liver, gill, ovary, and brain and was the only detectable form of ARNT (Powell et al.,
2000; Powell et al., 1999). Similarly, studies identifying expression of ARNT proteins in the zebrafish, Danio rerio, have identified an ARNT2 (zfARNT2b) that had a high level of identity when compared to fhARNT2 (80%) and mouse ARNT2 (82%), but had low identity with mouse ARNT (59%) and rtARNTb (52%; Tanguay et al., 2000).
Expression of zfARNT2b was detected to high levels in the brain, eye, gill, skin, and skeletal muscle with low or undetectable levels in the liver, heart, and kidney (Andreasen et al., 2002; Tanguay et al., 2000). ARNT2 has also been described as being the predominant ARNT form in the common cormorant, Phalacrocorax carbo with high
30 levels of mRNA detected in the liver, kidney, brain, muscle, heart, lung, spleen,
testis/ovary, eye, stomach, and intestine (Lee et al., 2007). Taken together, these studies
suggest a different functional role of ARNT2 in these species.
It is important to note, however, that the majority of studies examining the tissue
distribution of the ARNT proteins have focused primarily on mRNA expression patterns
through Northern blot analysis or whole mount in situ hybridization with little
examination of endogenous protein expression and more thorough protein studies are
missing from the current literature. Furthermore, numerous splice variants have been identified for each ARNT type and appear to exhibit different tissue distributions, which may also confound these characterizations (Drutel et al., 1996; Ikeda and Nomura, 1997;
Korkalainen et al., 2003; Pollenz et al., 1996; Prasch et al., 2006; Tanguay et al., 2000;
Wang et al., 1998; Wilson et al., 1997; Yu et al., 1999).
Interestingly, while ARNT and ARNT2 exhibit >90% and >80% amino acid identity between the bHLH and PAS domains respectively (Figure 1.6), gene knock-out of either ARNT or ARNT2 results in embryonic/perinatal lethality characterized by distinct phenotypes. This suggests that neither protein can compensate fully for the loss of the other. The lack of compensation may be due to differences in tissue specific
expression, discrepancies in the level of protein expression, or biochemical differences
that influence gene activation. While ARNT is known to have a more ubiquitous
expression pattern than ARNT2, mRNA for the two genes is clearly co-expressed to
some degree in many tissues, though the ratio of ARNT:ARNT2 proteins in these tissues
is unknown (Table 1.4). Since overlapping tissue specific expression of ARNT and
31
Figure 1.6: Homology between ARNT and ARNT2 domains. Thin lines indicate domain regions. NLS, nuclear localization signal; b, basic region; HLH, helix-loop-helix; PAS A and PAS B, PER/ARNT/SIM homology regions; TAD, transactivation domain. Numbers under each box represent amino acids. Thick lines represent regions of homology with the numbers representing the percent identity.
32
Table 1.4: mRNA expression for mARNT and mARNT2 in murine peripheral organs and central nervous system at postnatal day 1.5. (Adapted from: Aitola and Pelto-Huikko, 2003).
+ indicates the level of mRNA expression; +, low but positive signal; ++, weak positive signal, +++, strong positive signal.
33 ARNT2 does appear to exist, but neither ARNT can compensate fully for loss of the
other, this suggests that the two proteins have distinct functions in the presence of various
dimerization partners. Furthermore, the co-expression of ARNT and ARNT2 and the
potential impact each may have on the other in terms of signal transduction has not been
fully explored to date.
1.9 ARNT Isoforms from Other Species
Several orthologs of the ARNT proteins have also been identified in other
mammalian, invertebrate, aquatic and avian species including Drosophila melanogaster,
Caenorhabditis elegans, Xenopus laevis, the yellow fever mosquito Aedes aegypti, the common cormorant Phalacrocorax carbo, the chicken Gallus gallus, the domestic cow
Bos taurus, the rabbit Oryctolagus cuniculus, the crustacean Daphnia magna, the killifish
Fundulus heteroclitus, the zebrafish Danio rerio, and the rainbow trout Oncorhynchus mykiss (Lee et al., 2007; Nene et al., 2007; Pollenz et al., 1996; Powell-Coffman et al.,
1998; Powell et al., 1999; Prasch et al., 2006; Rowatt et al., 2003; Smith et al., 2001;
Sonnenfeld et al., 1997; Takahashi et al., 1996; Tanguay et al., 2000; Tokishita et al.,
2006; Walker et al., 2000). Many more ARNT proteins have also been predicted in silico based on genomic sequencing data (www.ensembl.org), though have not yet been evaluated in vivo. Phylogenetically, the ARNT proteins appear to be divided into two
major clades: one containing the BMAL type ARNT3 and ARNT4 proteins and one
containing the ARNT and ARNT2 type proteins (Figure 1.7). Consistent with the
evolutionary timeline, the mammalian, avian, and amphibian ARNT proteins grouped
34
Amino acid substitutions x100
Figure 1.7: ARNT protein dendogram. Full-length proteins were aligned by using the CLUSTALW (Slow/Accurate, Gonnet) method (MEGALIGN, DNAstar, Madison, WI) using a PAM250 residue weight table. The horizontal distance to the subclusters corresponds to degree of amino acid substitutions among members. Cavia porcellus ARNT (AB263100) and Mesocricetus auratus ARNT (AB263099) sequences were unpublished direct submissions to the NCBI database by Kawanishi,M., Sakamoto,M., Shimohara,C. and Yagi,T. 19-JUN-2006. Mammalian, avian, and amphibian ARNT proteins grouped together, as did the ARNT2, ARNT3, and ARNT4 proteins. ARNT proteins from piscine and invertebrate species also primarily grouped together as ancestors to the ARNT/ARNT2 split. Exceptions (Daphnia magna ARNT2 and Antheraea pernyi ARNT) are denoted by an asterisk.
35 together, while the ARNT proteins from piscine and invertebrate species also primarily
grouped together as ancestors to the ARNT/ARNT2 event. The ARNT2 proteins also
grouped together as did the BMAL type ARNT3 and ARNT4 proteins with few
exceptions.
1.10 ARNT-Dependent Signaling
The ARNT protein was initially proposed to function in the nuclear translocation
of the AHR protein (Hoffman et al., 1991; Reyes et al., 1992). It was later determined
that ARNT was constitutively nuclear, playing no role in the ligand mediated
translocation of the AHR (Holmes and Pollenz, 1997; Pollenz et al., 1994). ARNT is
now known to act as a dimerization partner essential in several distinct signal
transduction pathways each of which are mediated by ARNT's dimerization with a variety of bHLH-PAS proteins such as the aryl hydrocarbon nuclear receptor (AHR), the
hypoxia-inducible factor 1 family (HIF-1α, HIF-2α, HIF-3α), the hypoxia-like factor
(HLF), the single minded proteins (SIM1 and SIM2), or ARNT itself; however, each dimer associates with specific cis acting DNA elements to regulate genes (Figure 1.1).
Thus, ARNT is considered a master regulator required for response to hypoxia,
angiogenesis, xenobiotics, and in various developmental pathways.
1.10.1 AHR Signaling
In response to xenobiotics typified by 2,3,7,8 tetrachlorodibenzo-p-dioxin, ARNT
dimerizes with the liganded AHR and binds the asymmetric xenobiotic response element
(XRE: 5'T T/A GCGTG) enhancer sequences to regulate a battery of phase I and II
36 metabolizing enzymes that include CYP1A1, glutathione-S-transferase Ya, and
NAD(P)H:quinone oxidoreductase as previously described (Hapgood et al., 1989;
Hoffman et al., 1991; Ma, 2001; Ramadoss et al., 2005; Safe, 2001).
Interestingly, analysis of the role of ARNT2 in AHR-mediated signaling is less clear and studies examining the ability of ARNT2 to function during AHR signaling have been sparse and conflicting. The initial evaluation of the mouse ARNT2 clone included
XRE driven luciferase reporter studies, in which it was determined that ARNT2 appeared to be able to compensate for loss of ARNT in AHR signaling (Hirose et al., 1996).
However, recently Sekine et al. (2006) have hypothesized that ARNT2 does not dimerize with the AHR because it contains a proline and not a histidine residue at amino acid 352 within the PAS B domain. This hypothesis is based on the observation that all characterized mammalian ARNT2 proteins contain a proline residue in the PAS B domain while the homologous position in ARNT proteins contains a histidine residue.
Therefore, studies were initiated in which the histidine at amino acid 378 in the PAS B region of ARNT was mutated to a proline causing the ARNT to have greatly reduced ability to function in AHR-mediated signaling that was similar to that observed with wild type ARNT2. Interestingly, while the histidine to proline mutation in the Sekine et al. study affected the transcriptional ability of ARNT•AHR heterodimers, the transcriptional ability of ARNT•HIF-1α heterodimers remained unaffected, although the ARNT interactions of either AHR or HIF-1α appear to be mediated by the same domains.
However, the studies did not directly evaluate the role of the P352 in ARNT2 function by converting it a histidine and producing a protein capable of functioning in AHR-mediated signaling. Thus, the mechanism that underlies the lack of ARNT2 function in AHR-
37 mediated signaling in vivo is currently undefined. Further studies assessing the potential
roles of ARNT2 in AHR signaling should be considered.
The BMAL type ARNT proteins do not appear to be involved with AHR
signaling. Although, ARNT3 was initially demonstrated to coimmunoprecipitate with the
AHR using in vitro expressed and activated proteins, it was also determined that ARNT3
did not appear to associate with the AHR in a yeast two-hybrid system and also did not
appear to be able to substitute for ARNT in ARNT-deficient c4 hepatoma cells to
regulate an XRE controlled reporter (Hogenesch et al., 1997; Takahata et al., 1998).
1.10.2 Hypoxic Signaling
Under hypoxic conditions, ARNT binds with the bHLH/PAS protein hypoxia- inducible factor (HIF) -1α, -2α, -3α, which are constitutively degraded under non- hypoxic conditions, to mediate gene regulation through binding to hypoxia response elements (HRE: 5'T/GACGTG) in specific genes involved with vascularization and glucose transport such as erythropoetin (EPO), vascular endothelial growth factor
(VEGF), and glucose-transporter-1 (GLUT-1) (Maxwell et al., 1997; Poellinger and
Johnson, 2004; Semenza, 2007; Semenza et al., 1997; Wood et al., 1996). However, since the angiogenic deficits seen in HIF-/- animals are more severe than those seen in
ARNT-/- animals, it is likely that other protein(s) can partially substitute for loss of ARNT during hypoxic signaling (Iyer et al., 1998; Ryan et al., 1998).
Thus, the suggestion that ARNT2 may be this protein is provided by evidence that
HRE controlled genes such as VEGF are still able to be regulated during development in areas where ARNT2 mRNA is known to be present, such as the neural tube (Maltepe et
38 al., 1997). Furthermore, ARNT2 has been shown to be able to substitute for ARNT in
regulation of a HRE controlled reporter in ARNT-deficient cells (Sekine et al., 2006).
ARNT3 was originally demonstrated to associate with HIF-1α and HLF by yeast-
two hybrid screens and gel shift assays and could substitute for ARNT in ARNT-
deficient c4 hepatoma cells to regulate an HRE controlled reporter (Hogenesch et al.,
1997; Takahata et al., 1998). However, it was later shown that ARNT3 did not appear to
be participating in hypoxic response since ARNT3 could not complement hypoxia-
mediated signaling in ARNT-/- ES cells and ARNT3-/- embryos lacked angiogenic defects
similar to those seen in ARNT-/- , ARNT2-/- or HIF1-/- animals, though the ubiquitous
expression pattern of ARNT would likely limit potential angiogenic defects resulting
through loss of ARNT3 since ARNT would be present to complement hypoxic signaling
(Cowden and Simon, 2002; Kozak et al., 1997; Maltepe et al., 2000; Maltepe et al., 1997;
Ryan et al., 1998). Similarly, ARNT4 has been shown to be competent to participate in hypoxic signaling in Hep3B cells by regulating an HRE controlled reporter (Hogenesch
et al., 2000). Further biochemical studies using endogenously expressed proteins should be carried out to continue to assess the potential interactions of ARNT2-4 with the HIF family of proteins.
1.10.3 SIM1/SIM2 Interactions
The single-minded protein, first identified in Drosophila, is another bHLH/PAS transcription factor that controls the development of the neurons and glia in the central nervous system by regulating target genes that repress neuroectodermal genes and activate CNS midline genes (Nambu et al., 1991; Sonnenfeld et al., 1997; Thomas et al.,
39 1988). The mammalian orthologs Sim1 and Sim2 are also expressed in the developing
CNS and appear to similarly regulate mammalian CNS development by regulating genes controlled by central midline enhancing elements (CME: G/ATACGTGA; Michaud and
Fan, 1997; Nambu et al., 1991; Sonnenfeld et al., 1997; Swanson et al., 1995; Wharton and Crews, 1993). Since all ARNT proteins appear to be expressed in the brain along with the Sim proteins (Sim1, Sim2), all ARNT proteins would appear to have the potential to interact with Sim in vivo. Interestingly, ARNT, ARNT2, and ARNT3 have all been shown to interact with Sim1 by co-immunoprecipitation, while the potential interactions of ARNT4 with Sim remain unexplored (Michaud et al., 2000).
In Drosophila, mutations in ARNT (tango) result in defects in CNS midline and tracheal development (Sonnenfeld et al., 1997). In the mouse, the interactions of ARNT with SIM1 are suggested by neural tube defects seen in ARNT knockout animals and have been supported with biochemical analyses including co-immunoprecipitation, yeast two-hybrid, and electrophoretic mobility shift assays using CME oligos (Ema et al.,
1996; Kozak et al., 1997; Probst et al., 1997). Furthermore, expression of Sim1 appears to inhibit TCDD-dependent reporter activity in Hepa-1 cells as well as the DNA binding activity of the AHR•ARNT complex in vitro, suggesting that Sim1 is able to sequester
ARNT away from the AHR (Probst et al., 1997). ARNT•Sim2 interactions have also been shown via co-immunoprecipitation and CME reporter assays, though these dimers lack transcriptional ability, and Sim2 appears to function as a dominant negative regulator (Moffett and Pelletier, 2000; Moffett et al., 1997).
Along these lines, the interactions of ARNT2 with SIM are implied by the similar phenotype of SIM-/- and ARNT2-/- animals in the failure to develop specific
40 neuroendocrine lineages in the paraventricular (PVN) and supraoptic nuclei (SON) of the
hypothalamus (Hosoya et al., 2001; Michaud et al., 2000; Wines et al., 1998). In support of this, ARNT2 and Sim1 expression appears to be co-localized in the PVN/SON, while
ARNT levels are low or absent (Michaud et al., 2000). Furthermore, these studies also demonstrated that both Sim1 and ARNT2 appear to function in maintaining the expression of Brn2 in the PVN/SON, which in turn, controls the differentiation of cells expressing vasopressin, oxytocin, and corticotropin-releasing hormones. Thus, since
ARNT2 and Sim1 are co-localized spatially and temporally, exhibit similar phenotypes, and appear to control the same developmental processes, it is likely that they are partners in vivo, while interactions of other ARNT proteins with Sim1 in other tissues or with
Sim2 remain unexplored possibilities.
1.10.4 Circadian Signaling
The BMAL type ARNT proteins, ARNT3 and ARNT4, are currently believed to be primarily involved in the maintenance of circadian rhythms since targeted disruption of ARNT3 in the mouse results in the abolishment of circadian rhythm maintenance under constant darkness (Bunger et al., 2000). During Clock/ARNT3/4 signaling, Clock and ARNT3/4 when temporally co-expressed form a heterodimer that associates with E- box response elements (5’CACGTG) to regulate target genes involved with circadian signaling such as the period (Per) family of genes, the cryptochrome (Cry) genes, and vasopressin, leading to gradual accumulation of the Per protein in the cytoplasm. In turn, Per, along with timeless (Tim) interact with Cry to inhibit transcription by the
Clock/ARNT3/4 complex as a negative feedback loop. The interactions of both ARNT3
41 and ARNT4 with Clock have been well established biochemically and several excellent reviews detail these studies and the Clock pathway (Gekakis et al., 1998; Gu et al., 2000;
Maemura et al., 2007; Sangoram et al., 1998). As a result of these analyses, ARNT3 and
ARNT4 are not believed to be involved with the same array of signaling pathways as
ARNT and ARNT2.
1.10.5 AINT Interactions
While interactions with ARNT proteins with Sim and HIF proteins are critical for proper development of the CNS and vasculature, other proteins have been suggested to interact with ARNT proteins during specific developmental time points. One of these proteins, the ARNT interacting protein (AINT, TACC3) is a coiled-coil/PAS protein that has been shown to interact with both ARNT and ARNT2 in yeast and in vitro (Sadek et al., 2000). These studies showed that overexpression of AINT appears to lead to non- nuclearization of ARNT and an apparent augmentation of response to hypoxia as measured by HRE controlled luciferase reporters in Hepa-1 cells transfected with naked vector or AINT. More recently, it was demonstrated that expression of AINT in vivo coincided with high levels of HIF-1α in the neuroepithelium of the neural tube and was highly expressed in several areas of rapid cellular proliferation in the developing mouse as well as in the ovaries and testes of the adult, suggesting that AINT may be expressed as a means to augment cellular proliferation possibly through increasing the speed/strength of hypoxic signaling (Aitola et al., 2003). Non-nuclearization of ARNT could also serve to inhibit constitutive ARNT homodimer regulation of genes, which
42 have also been implicated in the regulation of cellular differentiation and proliferation,
and such studies have yet to be explored.
Another protein suggested to interact with ARNT during specific developmental
time points is the hypoxia-like factor (HLF) (Ema et al., 1997). In contrast to the hypoxia
inducible factors, this novel member of the HIF family appears to be expressed under
normoxic conditions to high levels in the lung, heart, and liver and appears to regulate
expression of VEGF in an ARNT dependent manner via the previously described HRE.
1.10.6 ARNT Homodimer Interactions
ARNT has also been implicated in serving as a transcriptional regulator in a
homodimeric form, based on early studies in which it was demonstrated that during size
exclusion high performance lipid chromatography of Sf9 whole cell lysates, ARNT was eluted as a single peak with a molecular mass of 205 kDa (Sogawa et al., 1995). These studies also demonstrated that ARNT could form homodimers via the HLH/PAS domains, had affinity for the core sequence of the adenovirus major late promoter (MLP) containing the canonical E-box (5’CACGTG) sequence, and could drive expression of E- box controlled reporters when produced in vitro similar to the activities of other E-box binding proteins MyoD, Max, and USF. Further studies showed similar results, again suggesting that ARNT homodimers could associate with the E-box sequence and could competitively displace the c-Myc/Max heterodimer from binding to the E-box, though the ability of ARNT to homodimerize may be dependent upon ARNT phosphorylation status
(Antonsson et al., 1995; Huffman et al., 2001; Levine and Perdew, 2001; Levine and
Perdew, 2002; Swanson et al., 1995; Swanson and Yang, 1999). However, physiological
43 target genes for ARNT homodimers remained undescribed until recently when microarrays examining differential gene regulation between wild-type Hepa-1 cells and
their ARNT-deficient variants revealed 27 genes whose expression was upregulated by
≥1.5-fold in the wild-type Hepa-1 cells, 15 of which had confirmed E-box sequences in
the 5’ promoter region, including BCL2/adenovirus E1B 19 kDa interacting protein 1
(NIP3), Serine (or cysteine) protease inhibitor clade E number 1 (PAI1), and N-Myc
downstream regulated-like (NDR1) (Wang et al., 2006). Additionally, ARNT
homodimers have been implicated in the partial regulation of CYP2A5 through chromatin immunoprecipitation, electrophoretic mobility shift assays, a liver-specific
ARNT knockout mouse showing reduced CYP2A5 levels, and mutational analyses in the
E-box region of the CYP2A5 promoter, though AHR•ARNT dimers have also been suggested to participate in the regulation of CYP2A5 expression as well as PAI1
(Arpiainen et al., 2007; Arpiainen et al., 2005; Son and Rozman, 2002). In contrast, multiple yeast two-hybrid experiments have indicated only weak or absent interactions for ARNT homodimers and other studies have failed to detect ARNT homodimerization
(Hirose et al., 1996; Reisz-Porszasz et al., 1994). Similarly, the interactions of ARNT2-4 as homodimers have been examined through yeast two-hybrid systems and have showed only weak interactions at best (Hirose et al., 1996; Takahata et al., 1998). Thus, the potential role of ARNT as a homodimeric transcriptional regulator remains controversial and though such studies prove difficult, attempts must be made to further explore the possibility under physiological conditions.
44 1.11 ARNT Interactions with Transcriptional Repressors
Along with ARNT’s interactions with several transcription factors and co- activators, ARNT proteins also appear to associate with transcriptional repressors such as the aryl hydrocarbon receptor repressor, NPAS, and necdin to prevent further PAS protein signaling. Such interactions have not been well-defined and may be important feedback or negative regulators of several of the aforementioned pathways.
1.11.1 Aryl Hydrocarbon Receptor Repressor Interactions
ARNT has also been implicated in binding to the aryl hydrocarbon receptor repressor (AHRR), which is upregulated by liganded-AHR•ARNT dimers and subsequently appears to be involved with negatively influencing AHR signaling by associating with ARNT and binding XREs, thereby preventing association of
AHR•ARNT dimers (Baba et al., 2001; Kikuchi et al., 2003; Mimura et al., 1997).
Interestingly, this repression does not appear to require DNA binding by the
AHRR•ARNT complex and cannot be rescued by additional ARNT expression, suggesting that AHRR repression of AHR signaling does not occur through sequestration of ARNT, but through alternative protein-protein interactions (Evans et al., 2008).
Recently, it has been suggested that AHRR plays a role as a tumor suppressor since silencing of AHRR in several human malignant tissues led to increased cell growth, while exogenous expression of AHRR led to diminished growth, diminished angiogenesis, and appeared to confer resistance to apoptotic signals (Zudaire et al., 2008). AHRR null mice, however, exhibited a seemingly contradictory delayed response to carcinogenesis in the skin following exposure to benzo[a]pyrene along with increased CYP1A1 mRNA
45 induction in the skin, stomach, and spleen compared to wild-type AHRR mice (Hosoya et
al., 2008). Thus, the role of the interaction of AHRR with ARNT remains controversial.
Future studies should continue to address the potential physiological role of the AHRR
and should address whether AHRR can interact with other ARNT proteins.
The ARNT proteins have also been implicated in binding to several repressors
1.11.2 NPAS Interactions
ARNT and ARNT2 have been implicated in binding to the neuronal PAS domain
protein 1 (NPAS), a bHLH-PAS transcriptional repressor expressed in the mouse brain
following organogenesis that is thought to be involved with nervous system development
and morphogenesis in the murine lung by regulating genes such as erythropoeitin and
tyrosine hydroxylase that are involved in hypoxic response (Levesque et al., 2007;
Ohsawa et al., 2005; Teh et al., 2006). In these studies, NPAS was shown to be
cytoplasmic, entering the nucleus via the NLS of ARNT following heterodimerization,
resulting in binding to enhancer regions and subsequent transrepression of gene targets
(Teh et al., 2006). In contrast, NPAS2 expressed in the mammalian forebrain appears to
be CLOCK-like and can interact with ARNT3 to drive expression of PER and CRY,
showing no transrepression (Franken et al., 2006; Reick et al., 2001; Rutter et al., 2001).
1.11.3 Necdin Interactions
Necdin is a member of the melanoma antigen (MAGE) protein family that is
thought to be a growth suppressor expressed predominantly in postmitotic neurons
(Kuwako et al., 2004; Maruyama et al., 1991). Recently, a study has implicated necdin as
46 a transcriptional repressor involved with modulation of ARNT2•SIM1 and ARNT2•HIF-
1α complexes (Friedman and Fan, 2007). In this study, necdin could be co-precipitated with ARNT2 or HIF-1α and appeared to interact with the bHLH of ARNT2 or HIF-1α via a 115 residue region in its amino-terminus, resulting in a dramatic repression of reporter activity controlled by CMEs. Interestingly, interaction of necdin with ARNT or
ARNT3 was not seen in cell culture, suggesting that necdin may specifically act to repress the transactivation of ARNT2 containing complexes. Since studies examining the impact of necdin on other bHLH-PAS proteins are relatively non-existent, this avenue should be continued as well as the potential repression of target genes by necdin in other
ARNT2 requiring pathways.
1.12 AHR and ARNT Levels
The levels of both AHR and ARNT are extremely important to the AHR-signaling pathway as it has been characterized. Calculations of the concentration of AHR and
ARNT protein in continuous cell culture lines from a variety of species and tissues revealed that the levels of ARNT remained relatively consistent between lines, ranging from approximately 13,979 molecules of ARNT/cell in MDCK canine kidney cells (94 fmol/mg lysate) up to 33,445 molecules of ARNT/cell in Hepa-1 mouse hepatoma cells
(231 fmol/mg lysate; Holmes and Pollenz, 1997). In contrast, AHR levels in those lines showed a much greater range of expression, varying from 4,763 molecules AHR/cell in the H4IIE rat hepatoma cells (69.4 fmol/mg lysate) up to 323,004 molecules of AHR/cell
(2231 fmol/mg lysate; Holmes and Pollenz, 1997).
47 Recent studies have examined the critical importance of these levels by
modulating the expression of these proteins by siRNA (RSP, unpublished). While a
moderate reduction of ARNT levels by siRNA in Hepa-1 cells has little effect on
CYP1A1 induction following TCDD treatment, a similar reduction of AHR levels by
siRNA leads to a marked reduction of CYP1A1 levels following TCDD treatment.
Similarly, overexpression of AHR in Hepa-1 cells can lead to an increased response in
CYP1A1 induction at the mRNA and protein level following TCDD treatment,
suggesting that the levels of AHR may be more limiting than the levels of ARNT, even
though there appears to be ~10x more AHR than ARNT in these lines.
1.13 ARNT Crosstalk
Since the ARNT proteins appear to be capable of dimerizing with a variety of
partners in vivo and appear to be expressed at relatively low levels, it is possible that
signaling in one pathway requiring ARNT would potentially disrupt/inhibit simultaneous
signaling of other ARNT requiring pathways (Holmes and Pollenz, 1997). Several
studies have addressed these questions, but have reported conflicting results (Chan et al.,
1999; Chilov et al., 2001; Gradin et al., 1996; Lee et al., 2006; Pollenz et al., 1999; Qu et al., 2007; Safe and Wormke, 2003; Woods and Whitelaw, 2002).
Early studies on hypoxia/AHR crosstalk demonstrated that AHR signaling appeared to be reduced following hypoxic signaling induction by cobalt chlorine
(CoCl2), a known inducer of hypoxia-inducible factor 1α (Gradin et al., 1996). In these studies, HepG2 cells transfected with an XRE controlled luciferase reporter showed reduced luciferase activity when treated with both CoCl2 and TCDF, a known AHR
48 ligand, in comparison to those treated with TCDF alone (Gradin et al., 1996).
Electrophoretic mobility shift assays (EMSA) of nuclear extracts from TCDF treated
HepG2 cells also showed XRE binding that was reduced when the cells were pretreated for increasing time periods with CoCl2 prior to TCDF treatment, suggesting decreased recruitment of AHR•ARNT heterodimers to XRE enhancers under hypoxic conditions and similar results were seen in EMSAs combining various amounts of in vitro produced
AHR, ARNT, and HIF-1a in the presence of β-naphthoflavone (Chan et al., 1999).
Additionally, coimmunoprecipitation studies using in vitro translated AHR, ARNT, or
HIF-1a demonstrated that coprecipitation of the AHR with ARNT antiserum following
TCDD treatment was reduced in a concentration dependent manner when HIF-1α was added to the activation (Gradin et al., 1996). However, in neither of these studies was
ARNT used to rescue AHR signaling during hypoxia.
In contrast, studies by Pollenz et al. demonstrated that the physiological functional interference between hypoxia and AHR signaling does not appear to occur through competition for ARNT (Pollenz et al., 1999). In these studies, hypoxia did not affect the concentration or localization of ARNT, and the ARNT protein sequestered during hypoxia was returned to the cellular ARNT pool following hypoxia-driven signaling. Furthermore, under conditions of physiological hypoxia (1% O2), only a small fraction of the total cellular ARNT pool (12-15% in Hepa-1 and H4IIE cells respectively) was sequestered, suggesting that the remaining ARNT would remain free for other
ARNT requiring pathways assuming that the entire protein pool is accessible for dimerization. Additionally, the formation of AHR•ARNT heterodimers as measured by electrophoretic mobility shift assays of nuclear extracts from four distinct TCDD treated
49 cell lines (Hepa-1, HepG2, MCF, or H4IIE) remained unchanged in comparison to normoxic conditions. Furthermore, these studies indicated that while the level of TCDD
inducible CYP1A1 protein was decreased under hypoxic conditions, the level of
CYP1A1 mRNA remained similar regardless of hypoxia, suggesting that reductions in
CYP1A1 protein levels were distal to the induction of CYP1A1. Since cellular protein
synthesis is generally reduced in hypoxic cells, it is likely that reductions in CYP1A1
protein occurred as a non-specific cellular response to hypoxia rather than a mechanistic
response resulting from loss of available ARNT.
More recent studies on AHR/HIF crosstalk suggest that AHR/HIF crosstalk is
limited to genes with enhancer regions containing specific regulatory motifs (Lee et al.,
2006). Using global gene expression patterns derived from high-density oligonucleotide
arrays of Hep3B cells treated with Me2SO, CoCl2, TCDD, or TCDD and CoCl2, Lee et al.
identified only 33 target genes affected by co-treatment with TCDD and CoCl2 whose expression appeared to be modulated by crosstalk. However, of these 33 genes, 35% were actually upregulated by co-treatment with TCDD and CoCl2, while 38% were
downregulated by co-treatment with TCDD and CoCl2, and 33% were differentially
regulated through co-treatment with TCDD and CoCl2 in comparison with either TCDD
or CoCl2 alone. Upon evaluating these target gene promoters, the authors found no
correlation between the types or number of HRE or XRE response elements in the
putative regulatory regions of these genes and the apparent crosstalk that was observed
and, instead, determined that the incidence of serum response factor regulatory elements
(SRE) was more predictive of crosstalk than the number of HRE or XRE sites. Given
that the majority of genes evaluated appeared to be unaffected by co-treatment with
50 TCDD and CoCl2 compared with either treatment alone, it is unlikely that treatment with either TCDD or CoCl2 resulted in loss of ARNT to the other signaling pathway since such a loss would likely have a more global effect on gene regulation by AHR or HIF.
Furthermore, the co-treatment with TCDD and CoCl2 resulted in an additive effect in
35% of the 33 affected genes. Taken together along with the data presented by Pollenz et al., these data suggest that in a physiological setting, AHR/HIF crosstalk is limited in scope and does not appear to occur through competition for ARNT (Lee et al., 2006;
Pollenz et al., 1999).
In addition to direct competition for ARNT, crosstalk of ARNT-requiring pathways may also be occurring indirectly through regulation of bHLH/PAS proteins by another bHLH/PAS pathway. For example, hypoxia has been implicated in the modulation of expression of PER and CLOCK in the mouse brain leading to increased expression of both PER1 and CLOCK and, similarly, Sim has been shown to attenuate regulation of EPO during hypoxia (Chilov et al., 2001; Woods and Whitelaw, 2002).
1.14 AHR and ARNT Defective Hepatoma Cell Lines
Six benzo[a]pyrene mutant clones (c1-c6) all exhibiting decreased or undetectable levels of aryl hydrocarbon hydroxylase inducibility, sparing them from the toxic metabolites of reactive benzo[a]pyrene intermediates, were isolated from the Hepa1-c1c7 mouse hepatoma parental line and characterized (Hankinson et al., 1979; Hankinson,
1979; Legraverend et al., 1982). Through somatic cell hybridization, five of these mutations were found to be recessive and belong to three complementation groups (c1 and c5, c2 and c6, and c4), while another was found to be dominant (c3). The c2 and c6
51 clones were shown to be AHR-deficient (<10% wt) and the c4 clone was later determined
to be ARNT-deficient (Hankinson, 1979; Hoffman et al., 1991; Legraverend et al., 1982;
Pollenz et al., 1994). At the same time, another laboratory group utilized the mouse
Hepa1-c1c7 cell line to select populations defective in aryl hydrocarbon hydroxylase
activity (Whitlock and Galeazzi, 1984). Similarly, their LA-I variant was shown to be
AHR-deficient, while the LA-II variant was later shown to be ARNT-deficient (Pollenz
et al., 1994; Reyes et al., 1992). Since re-introduction of AHR cDNA is known to restore
TCDD mediated CYP1A1 induction in the c2 and LA-I variants, these variants will be
used to explore the function of other AHRs (Legraverend et al., 1982). Similarly, re-
introduction of ARNT cDNA to the c4 or LA-II variants has been shown to restore some
level of function in terms of AHR- and hypoxia- mediated signaling (Maxwell et al.,
1997; Pollenz et al., 1994). It is therefore expected that the re-introduction of various
ARNT constructs, and possibly, ARNT2 should complement AHR and hypoxia signaling
and would thereby provide a common genetic background to evaluate ARNT and
ARNT2 function.
1.15 Knockout Animals
An important tool for examining protein function is targeted disruption of genes
through homologous recombination using targeting vectors. With the use of this technique, null mice have been generated for the AHR, ARNT, and ARNT2
independently. Two separate laboratories have generated AHR null mice (Gonzalez et
al., 1995; Schmidt et al., 1996). The resulting AHR-/- mice were demonstrated to be
viable, though they exhibited a multitude of physiological changes and developmental
52 abnormalities including: reduced liver weight with pronounced fibrosis in the portal tract,
glycogen depletion, eosinophilia, reproductive defects, and a retarded growth rate
(Fernandez-Salguero et al., 1995; Schmidt et al., 1996). A 40-50% neonatal lethality and
depressed immune systems were also seen in mice generated by the Fernandez-Salguero
group, but neither was seen in those of Schmidt et al., (1996). As both studies were done
in C57BL/6J mice, differences between the two are not attributable to genetic
background; rather, these differences are more likely to be attributable to partial allelic
inactivation, which was addressed by the Schimdt et al. (1996) group, but not in the study
by Fernandez-Salguero et al. (1995). In either case, mice lacking the AHR protein were
unable to induce wild-type levels of CYP1A1 in response to TCDD administration.
Additionally, the Fernandez-Salguero AHR-/- mice exhibited a significant decrease in
CYP1A2 and ΜGT*06, which are constitutively expressed in the wild-type mouse, indicating that the AHR is responsible for controlling the basal levels of these enzymes as well as for controlling induction of other drug metabolizing enzymes in response to ligand (Fernandez-Salguero et al., 1995). Interestingly, the disruption of the Ah locus also led to protection from the harmful effects of TCDD, benzo[a]pyrene, polychlorinated biphenyls, and polybrominated biphenyls including protection against thymic atrophy, lesions, and disruption of thyroid hormone homeostasis (Fernandez-Salguero et al., 1996;
Nishimura et al., 2005).
Null mice have also been generated for the ARNT, ARNT2, and ARNT3 proteins
independently. Interestingly, while the ARNT and ARNT2 proteins possess a high
degree of homology in the functional domain regions, the phenotypes of either null
animal differ significantly. For ARNT, two separate laboratories have generated null
53 mice (Kozak et al., 1997; Maltepe et al., 1997). In both groups, the knockout of ARNT proved to be fatal, with animals dying in utero between embryonic day (ED) 9.5 and
10.5. As such, the ARNT null animals were examined prior to 10.5 days of gestation in an attempt to reveal the underlying cause of death. In the Maltepe line, ARNT-/- animals exhibited defective angiogenesis of the yolk sac and branchial arches, stunted development and embryo wasting and were generally distinguishable from their heterozygous ARNT littermates by ED 8.5-10.5. The cause of lethality in these animals was attributed to the formation of hypoxic/nutrient-deprived cells resulting from an inability to promote vascularization for the increasing tissue mass during organogenesis due to a loss of ARNT. To support these data, ARNT-/- animals were shown to have less
VEGF mRNA in the yolk sac as well as in the embryo overall, with high VEGF mRNA appearing to be restricted to areas known to be colocalized with ARNT2 mRNA. In contrast, animals of the Kozak line were indistinguishable from their littermates at ED
8.5, exhibiting normal vasculogenesis in the yolk sac and embryo. By ED 9.5, however,
60% of ARNT-/- animals exhibited developmental delay and/or abnormal phenotypes
(100% by ED 10.5). These abnormalities included: neural tube closure defects, forebrain hyperplasia, delayed rotation of the embryo, placental hemorrhaging, and visceral arch abnormalities, with no defects seen in yolk sac circulation. However, they postulated that the cause of lethality lay with the failure of the embryonic component of the placenta to vascularize preventing further development, possibly as a result of loss of ARNT in necessary hypoxic response. These data are supported from evidence that other knockouts in which lethality occurs at ED 10.5 were attributed to improper placental development (Guillemot et al., 1994; Gurtner et al., 1995). Loss of ARNT has also been
54 linked to pancreatic-islet dysfunction in type II diabetes (Czech, 2006; Gunton et al.,
2005; Levisetti and Polonsky, 2005).
ARNT2 knockouts have also been examined by three independent laboratories
(Hosoya et al., 2001; Keith et al., 2001; Michaud et al., 2000; Wines et al., 1998). In all
groups, the knockout of ARNT2 was non-embryonic lethal; instead, ARNT2-/- animals died perinatally within two weeks of birth (Hosoya et al., 2001). Loss of ARNT2 was correlated to deficiencies in the formation of specific neuroendocrine lineages in the hypothalamus and altered expression of Brn2 in the hypothalamus similar to SIM-/-
animals, and also exhibited impaired regulation of HIF-1 target genes and thymic defects
(Keith et al., 2001; Wines et al., 1998). Attempted double knockouts of ARNT and
ARNT2 in the mouse were unsuccessful and out of 67 embryos resulting from a crossing
of heterozygous mutant ARNT-/+, ARNT2-/+ animals embryos, no embryos were found to be homozygous double mutants by ED 8.5, and interestingly, only a single embryo was found for the ARNT-/+, ARNT2-/- or ARNT-/-, ARNT2-/+ genotype (Keith et al., 2001).
This suggests that in the mouse, two wild-type alleles of either ARNT or ARNT2 are
required to prevent resorptions of the embryo and that ARNT and ARNT2 may have
overlapping functions during early development prior to ED 8.5.
Conversely, knockout of ARNT3 does not appear to elicit defects in early
development, but does lead to increased mortality after 26 weeks of age, abolishment of
circadian rhythm maintenance under constant darkness, and progressive noninflammatory
arthropathy evidenced by ossification of tendons and ligaments (Bunger et al., 2005;
Bunger et al., 2000). In another, independent, ARNT3 knockout mouse, deficiency in
ARNT3 led to reduced levels of B cells in the peripheral blood, spleen and bone marrow
55 and also led to increased mortality around 6 months of age, but was correlated to elevated
levels of serum glutamic oxaocetic transaminase, serum glutamic pyruvic transaminase,
and blood urea nitrogen, which are indicators of liver, heart, and kidney damage (Sun et al., 2006a; Sun et al., 2006b). However, pathological assessment of these tissues revealed no obvious differences between wild-type and ARNT3-/- mice, suggesting a possible loss of function in these tissues rather than tissue damage. An ARNT4 knockout mouse has not yet been evaluated.
1.16 Toxicity of TCDD
Halogenated aromatic hydrocarbons are widespread environmental contaminants that are generally chemically stable, lipophilic, and resistant to degradation. As such, they tend to persist in the environment where they may bioconcentrate and bioaccumulate through food webs. While some of these hydrocarbons are directly manufactured such as polybrominated biphenyls, some are formed as contaminants in the manufacture of other commercial products. TCDD, for example, is formed as a contaminant in the synthesis of
2, 4, 5-trichlorophenol, which is used in the commercial synthesis of 2, 4, 5- trichlorophenoxyacetic acid (2, 4, 5-T), a widespread herbicide and defoliant.
Of the common types, chlorinated dibenzo-p-dioxins, dibenzofurans, azo(xy)benzenes, naphthalenes, biphenyls and brominated biphenyls, each has produced incidents of poisoning of industrial workers, members of the general population, or farm
workers (Poland and Knutson, 1982). All of these aromatic hydrocarbons share similar
chemical structures and produce similar toxic responses varying in potency (Figure 1.8).
56
Figure 1.8: Structures of common tetra-chlorinated- congeners for various halogenated aromatic hydrocarbons. Shown above are several of the chemical structures for various AHR ligands. A, halogenated hydrocarbons. The toxic isomers are halogenated at 3 or 4 of the lateral ring positions with at least 1 ring position that is unsubstituted. Adapted from: (Poland and Knutson 1982). B, polycyclic hydrocarbons. C, potential endogenous (indirubin) and dietary (indole-3-carbinol) ligands.
57 The toxic isomers are halogenated at 3 or 4 of the lateral ring positions with at least 1 ring
position that is unsubstituted (Poland and Knutson, 1982).
Toxicity resulting from exposure to polycyclic hydrocarbons is mediated by their
high affinity and saturable binding to the AHR and involves a broad range of adaptive
and toxic responses including the induction of xenobiotic metabolizing enzymes, such as
wasting syndrome, tumor production, thymic involution, hepatotoxicity, skin disorders,
gastric lesions, and alterations in endocrine homeostasis (Poland and Knutson, 1982;
Sutter et al., 1994). Exposure to TCDD has also been associated with urinary tract
hyperplasia, subcutaneous edema, decreased spermatogenesis, decreased testicular
weight, degeneration of the seminiferous tubules, fetal death and resorptions, fetal
wastage, and malformations (Kremer et al., 1994; Nebert et al., 1993; Okey et al., 1994;
Safe et al., 1998).
1.17 Role of AHR and ARNT in Mediating TCDD Toxicity
Recent studies using antisense morpholino oligonucleotides to knock down levels of the zebrafish AHR (zfAHR2) protein in vivo have revealed that the AHR is required for mediating many endpoints of TCDD toxicity, exhibiting protection against TCDD- induced toxic endpoints such as reduced blood flow, pericardial edema, and reductions in lower-jaw growth (Bello et al., 2004; Carney et al., 2004; Prasch et al., 2003; Teraoka et al., 2003). Reductions of non-functional isoforms of the AHR (zfAHR1) showed no such
protections. Similarly, zfARNT1 morphants show complete protection against TCDD-
induced pericardial edema and show partial protection against reduced peripheral blood
58 flow and lower-jaw growth deficiencies (Prasch et al., 2006). These results have also
been confirmed with a mammalian model system (Walisser et al., 2004).
Interestingly, antisense morpholinos against zfARNT2 indicate that reduced or absent levels of ARNT2 provide no protection against TCDD-induced developmental toxicity (Prasch et al., 2004a; Prasch et al., 2004b). In these experiments, both wild-type and zfARNT2-/- animals exhibited similar responses to TCDD as measured by the toxic
endpoints previously described. Taken together, these data suggest that disruption of
AHR-mediated signaling provides protection against many of the molecular and
physiological TCDD-induced toxic endpoints and that the AHR and ARNT1 are essential
for TCDD toxicity. Furthermore, these data also suggest that zfARNT2 is not functional
in TCDD-mediated AHR signaling.
While AHR-mediated signaling may be required for many endpoints of TCDD-
induced toxicity, the mechanism(s) for this toxicity is still unknown. Since the AHR and
ARNT are transcription factors, it is likely that increased signaling leads to an alteration
of gene expression resulting in toxicity. However, while zfCYP1A1 is the well-
characterized of the known TCDD regulated genes, zfCYP1A1 morphants exhibit no
protection against endpoints of TCDD-mediated toxicity (Carney et al., 2004). Thus,
genes involved in mediating such toxicity are as yet unknown.
59
Chapter Two
Generation and Characterization of Stable Lines Expressing Different Species of AHR
2.1 Rationale for Stable Lines Expressing Different Species of AHR
The aryl hydrocarbon receptor (AHR) is a basic helix-loop-helix PER/ARNT/SIM
(bHLH-PAS) transcription factor that binds ligands typified by 2,3,7,8- tetracholordibenzo-p-dioxin (TCDD), translocates to the nucleus, dimerizes with the aryl hydrocarbon nuclear translocator (ARNT) and associates with specific cis acting xenobiotic response elements (XRE) to activate transcription of genes involved with xenobiotic metabolism and is subsequently degraded. AHR-mediated signal transduction has been evaluated primarily in the C57BL/6J mouse model system. This model system, however, may not be the most accurate model for human comparisons as the AHRb-1 allele carried by C57BL/6J contains a point mutation that prematurely truncates the receptor at 805 amino acids, while the AHR found in other murine strains as well as in the rat and human all contain an additional 42-45 amino acids at their carboxy-terminus that have 70% identity (Figure 1.3, Table 1.3). This carboxy-terminal region could be functionally significant and the analysis of AHR-mediated signal transduction in the rat, human, or other mouse strains may better represent the physiology of the AHR pathway.
Since the Ah receptors found in mice carrying the Ahb-2 allele, like those found in
the rat and human, possess a carboxy-terminal region that is more typical across species
60 than the truncated b1 allele, these Ah receptors were chosen to be evaluated in an attempt
to evaluate a possible functional role for this additional carboxy-terminal sequence. The
ability to directly compare the function of these Ah receptors versus that of the AHRb-1 in a genetically identical background was essential for delineating possible roles of this sequence, since studies using an endogenously expressed AHRb-2 would need to be performed in a different cell line than the murine Hepa-1 line used to examine the AHRb-1
function and therefore could have confounding results due to other genetic variation
between these lines. Furthermore, the AHRb-2 allele was highly useful in that antibodies
currently in use by the Pollenz laboratory for the detection of the AHRb-1 allow
equivalent detection of both proteins since these antibodies were generated against an
amino-terminal portion of the AHRb-1 which shares 100% identity with the AHRb-2.
To formally investigate protein-protein associations and degradation patterns of the AHRb-2 receptor, and whether the differences seen were due to the receptor itself or
other cellular factors, stable lines were generated exploiting the LA-I, AHR deficient cells as described in Chapter Six. The use of a genetically identical background to
examine the function and degradation rate of the AHRb-2 in comparison with the AHRb-1 provided significant insights into these biological properties of the AHR contributed by this functional region.
2.2 Generation of LA-I Lines Expressing the Ahb-1 or Ahb-2 Receptor
The first step in these studies was the ligation of the various coding regions from
each cDNA into base and retroviral vectors, as detailed in Chapter Six (Invitrogen; Figure
2.1). The base vector, pcDNA 3.1 allowed cloned products to be produced through in
61
Figure 2.1: Invitrogen plasmid maps. Target genes are expressed from the cytomegalovirus immediate early (CMV IE) promoter and the plasmids contain both neomycin (G418) and ampicillin resistance used for antibiotic selection in eukaryotic and prokaryotic cells respectively. Top: pcDNA 3.1 was used as the base vector for most constructs and the T7 promoter exploited for the production of in vitro transcribed and translated proteins. This construct also contains a CMV promoter for in vivo expression located upstream of the viral T7 (not shown). Bottom: pQCXIN and pLNCX2 are retroviral vectors optimized to yield high titers. In pQCXIN, the possibility of promoter interference is minimized by expressing the gene of interest along with the neomycin resistance gene as a bicistronic transcript by means of an internal ribosomal entry site (IRES). Following transfection of either retroviral vector into packaging cells, the vector provides the viral packaging signal (Ψ+), the target gene, and the drug-resistance marker. Vectors were obtained from the manufacturer (Invitrogen; Carlsbad, California).
62 vitro transcription and translation reactions. The initial retroviral vector, pLNCX2, was chosen as it allows each of the proteins to be stably integrated into a common genetic background under the control of the same promoter. PCR was used to amplify the cDNA coding region as well as to add specific restriction endonuclease sites that were exploited for the ligation of the cDNA into the multiple cloning site of pLNCX2. Alternatively, retroviral constructs were also prepared using pQCXIN, a retroviral vector derived from pLNCX2 containing an internal ribosomal entry site for bicistronic expression of the neomycin resistance and target genes (Invitrogen, Figure 2.1). Using this vector allows for a greater percentage of positive clones following infection with competent virus since expression of the neomycin resistance gene for selection in eukaryotic cells is coincident with expression of the target cDNA.
Since neither pLNCX2 nor pQXCIN has an in vitro promoter, retroviral constructs were sequenced and tested by transient transfection into the AHR-deficient
LA-I Hepa-1 variant or in PT67 viral packaging cells derived from an NIH 3T3 line, then analyzed by SDS-PAGE and evaluated by Western blotting (Figure 2.2). Western analysis also validated that the molecular mass of each protein is correct. In either case, following transfection of viral packaging cells, viral media was collected and used to infect target cells as shown in Figure 2.3 (see also Chapter Six).
Stable lines surviving a two-week selection in neomycin were then analyzed for expression of the target genes by SDS-PAGE and Western blotting (Figure 2.4). While these studies were intended to generate stable lines in the Hepa-1 background expressing either the wild-type AHR, the Ahb-2, the human AHR, or the rat AHR, numerous colonies surviving neomycin selection following infection with human or rat AHR viral media
63 A
B
Figure 2.2: Western analysis of the AHR constructs following transfection into AHR-deficient PT67 viral packaging cells. Equal amounts of total cell lysates from independent clones were resolved by SDS-PAGE, blotted, and stained with A-1A anti- AHR IgG (1.0 µg/ml) and -actin IgG (1:1000). Reactivity was visualized by ECL with GAR-HRP IgG (1:10,000). A, total cell lysates from PT67 viral packaging cells transfected with Ahb-2 or Rat AHR. B, total cell lysates from PT67 cells transfected with Ahb-1, Ahb-2, rat AHR, or human AHR. WT, wild-type Hepa-1 cells; PT67, non- transfected viral packaging cells; B1, PT67 cells transiently expressing the Ahb-1 receptor; B2, PT67 cells transiently expressing the Ahb-2 receptor; RAT, PT67 cells transiently expressing rat AHR; HU, PT67 cells transiently expressing human AHR. HU TNT, in vitro transcribed and translated human AHR used as a control. Note that in each case PT67 cells express detectable levels of the target AHR.
64
Figure 2.3: Schematic for viral transfection. (Adapted from Clontech user manual PT3132-1). The viral genes gag, pol and env are required for viral production and are absent from the retroviral vector, but are integrated into the genome of the viral packaging cells. Following transfection of the retroviral vector into the packaging cells, the vector provides the viral packaging signal (Ψ+), the target gene, and the drug- resistance marker, while the packaging cells provide gag, pol and env to generate infection-competent, replication-incompetent virus, which is collected from the cell culture media. This viral media is then used to infect target cells, which are selected for infection by culturing in the presence of neomycin. Resultant colonies are subsequently evaluated for expression of the target gene.
65 A B
Figure 2.4: Western analysis of stable line expression of target AHR constructs. Equal amounts of total cell lysates from independent clones were resolved by SDS-PAGE, blotted, and stained with A-1A anti-AHR IgG (1.0 µg/ml) and -actin IgG (1:1000). Reactivity was visualized by ECL with GAR-HRP IgG (1:10,000). A, LA-I clones surviving selection after infection with Ahb-2, a truncated AHR (AHRtr), or rat AHR. WT, LA-I clones stably expressing the Ahb-1 receptor; B2, LA-I clones stably expressing the Ahb-2 receptor; AHRtr, amino-terminally truncated AHR; PT67, non-transfected viral packaging cell line; RAT, LA-I clones surviving selection after infection with rat AHR. B, LA-I clones surviving selection after infection with human AHR (Hu). Arrows indicate clones that were chosen for future analysis. Note that each lane represents an independent clone. Also note that several independent clones for both the rat and human AHR failed to express the AHR following selection.
66 failed to express the desired target genes or expressed these genes at very low levels, though the constructs contained the correct sequence and transiently expressed the target
genes following transient transfections into LA-I cells (Figure 2.2). Repeated attempts to
generate the lines by altering the number of rounds of infection with competent virus, the
confluency of target cells, or by using different lots of competent virus, all resulted in
colonies that failed to express the target genes at levels detectable with the A1-A AHR
antibody (Figure 2.4). Since the Ahb-1 and Ahb-2 lines were easily created, these lines
became the focus for the following studies.
Stable lines expressing the Ahb-2 were selected and those that expressed the
receptor to the same physiological level as the AHRb-1 in WT Hepa-1 cells were chosen
for future study and are referred to as AHb2 cells (Figure 2.4). The same process was also
completed for the AHRb-1 whereby the receptor was re-introduced into the LA-I AHR
deficient line. This line, termed hereafter as AHWT, was used as the control for the
following studies, rather than WT Hepa-1, since these cells have also undergone selection
and this line is therefore more comparable to other generated stable lines. Since the
expression of the AHRb-1 and AHRb-2 were achieved using retroviral vectors which insert
the target genes directly into the genome, it was possible that insertion of the target genes
could disrupt the expression of other gene(s) and thus confound the biochemical analyses
of AHR function. Therefore, all studies were performed using at least three independent
AHWT or AHb2 clones to increase the likelihood that the biochemical properties of the
AHR being evaluated were not resulting from altered expression of other genes impacting
the AHR signaling pathway.
67 2.3 Reduced Association of XAP2 with Ahb-2 Receptors Expressed in the Hepa-1
Background Mimics that of Endogenously Expressed Ahb-2 Receptors
It has been proposed that XAP2 functions in AHR-mediated signal transduction
by influencing the stability, shuttling behavior and expression of the AHR (Bell and
Poland, 2000; Berg and Pongratz, 2002; Carver and Bradfield, 1997; Kazlauskas et al.,
2002; LaPres et al., 2000; Lees et al., 2003; Ma and Whitlock, 1997; Meyer and Perdew,
1999; Meyer et al., 2000; Meyer et al., 1998; Petrulis et al., 2003; Ramadoss and Perdew,
2005). However, these conclusions have been based primarily on the analysis of AHR
and XAP2 in transient transfection systems. Thus, there has been minimal information on
the function of endogenous XAP2 especially as it relates to interactions with AHR
proteins from species other than the murine C57BL strain. Interestingly, initial studies
carried out on mouse Hepa-1c1c7 cells, which express the C57BL Ahb-1 receptor, mouse
C2C12 myoblasts, which express the Ahb-2 receptor, and rat smooth muscle cells (A7) suggested that AHR proteins from non-C57BL species exhibited a reduced association with XAP2. This observation was based on evidence that while Western blot analysis of total cell lysates from these lines showed equal levels of endogenous hsp90, p23, and
XAP2 protein and similar levels of AHR protein, cytosol produced from these lines and
immunoprecipitated with AHR antibodies as detailed in Chapter Six showed that the
level of XAP2 co-precipitated with the AHR in the C2C12 and A7 cells was greatly
reduced in comparison to the Hepa-1 cells (Figure A-1). In repeated experiments, the amount of XAP2 co-precipitated with AHR IgG from A7 and C2C12 cells averaged 15 ±
7% and 19 ± 6% respectively of the level co-precipitated from Hepa-1 cells. Since the
overall level of AHR precipitated from the C2C12 or A7 cells was similar to that from the
68 Hepa-1, and the total cellular levels of AHR and XAP2 were similar, these results
suggested that the differences in XAP2 association between the Hepa-1 cells and the
C2C12 or A7 cells may be related to the AHR itself, rather than cellular context.
Furthermore, recent studies likewise suggest that the human AHR is associated with
reduced levels of XAP2 when transiently expressed in COS cells (Ramadoss and Perdew,
2005). However, since other cellular factors could be contributing to the reduced association of AHR and XAP2 in these lines by competing for binding with either protein or blocking interaction sites, it was essential to explore these protein•protein interactions in a common genetic background.
Thus, to further explore the reduced association of XAP2 seen in the C2C12 cells carrying the Ahb-2 allele and to determine more definitively whether this reduced association between the non-C57BL/6J AHR and XAP2 was receptor or cell specific,
studies were carried out in the previously described AHWT and AHb2 stable cell lines.
Figure 2.5 A shows the expression of AHR, hsp90, XAP2, and actin in the parental LA-I
Hepa-1 variant, the AHWT, and the AHb2 lines. Importantly, the level of AHR expressed
in each of the stable cell lines approximates that of the endogenous AHR in wild type
Hepa-1 cells, and the levels of hsp90 and XAP2 in total cell lysates are also similar. To assess the composition of the unliganded AHR complex, cytosol was generated from two
independent AHb2 cell lines as well as from the AHWT line and immunoprecipitated with
antibodies specific to the AHR as previously described. As shown for the endogenous
b-1 Ah receptor in Hepa-1 cells, immunoprecipitation of the AHR from the AHWT cells co- precipitated a high level of XAP2 (Figure 2.5 B). In contrast, as shown for the
69
Figure 2.5: Analysis of AHR and XAP2 expression and association in stable cell b-1 b-2 b-1 lines expressing Ah or Ah receptors. Stable cell lines expressing the Ah (AHWT) b-2 or Ah (AHb2) were generated as detailed in Chapter Six. A, equal amounts of total cell lysates from LA-I (LA), Hepa-1 (He), AHWT (WT), or AHb2 (b2) cells were resolved by SDS-PAGE, blotted, and stained with A-1A rabbit IgG (1.0 µg/ml), -actin rabbit IgG (1:1000), hsp90 rabbit IgG (1:500), or XAP2 mouse IgG1 (1:750). Reactivity was visualized by ECL with GAR-HRP or GAM-HRP IgG (1:10,000). B, cytosol was prepared from AHWT (WT) or two independent AHb2 (b2) cell lines as detailed. 800 µg of cytosol was precipitated with either affinity-pure A1-A IgG (5 µg) or affinity-pure preimmune rabbit IgG (5 µg) along with Protein A/G-agarose (25 µl) for 2.5 h at 4 °C with rocking. Pellets were washed three times for 5 min each with TTBS supplemented with sodium molybdate (20 mM) and then boiled in 30 µl of SDS sample buffer. 15 µg of cytosol (input) or 15 µl of the eluted protein were resolved by SDS-PAGE, blotted, and stained with A-1A IgG (1.0 µg/ml) or XAP2 mouse IgG1 (1:750). Reactivity was visualized by ECL with GAR-HRP or GAM-HRP IgG (1:10,000). Ah, precipitated with A1-A IgG; Pi, precipitated with preimmune IgG. The numbers under the XAP2 blot represent the percentage of XAP in relation to the level in the AHWT cell line (100%). The precipitated IgG band is shown to demonstrate the uniformity of the precipitation across all samples. 70 endogenous rat and Ahb-2 receptors, immunoprecipitation of the AHR from both of the
AHb2 cell lines precipitated significantly lower levels of endogenous XAP2 than the
AHWT (compare Figures 2.5 B and A-1 B). Indeed, over several experiments, the average
amount of XAP2 co-precipitated from the AHb2 cells was 21 ± 8% of the level co-
precipitated from AHWT, which highly mimics the 19 ± 6% seen in the C2C12 cells
endogenously expressing the Ahb-2 receptor. Taken together, these studies clearly
indicate that even in the same genetic background as the Ahb-1, the Ahb-2 receptor exhibits
a reduced association with XAP2. Thus, the species of receptor, not the cellular context
of its expression, appears to determine the level of association with XAP2. Furthermore,
since the Ahb-1 and the Ahb-2 exhibit 99% identity of the first 805 amino acids, this reduced association is likely to be related to the extended carboxy-terminal region found in most species. However, since XAP2 appears to associate with amino acids 380-419 of the PAS B domain of AHR which share 100% identity between the Ahb-1 and Ahb-2
(Meyer and Perdew, 1999) and not through a carboxy-terminal region of AHR, this reduced association is likely due to a conformational change in the AHR resulting from the extended carboxy terminus rather than sequence specific differences.
2.4 Ahb-2 Receptor Expressed in the Hepa-1 Background Exhibits Nucleocytoplasmic
Shuttling
The endogenous Ahb-1 receptor is predominantly cytoplasmic in Hepa-1 cells
when evaluated by immunohistochemistry or confocal microscopy (Pollenz, 1996;
Pollenz et al., 1994), though exogenous expression of the Ahb-1 receptor in HeLa or COS
cells is nuclear unless XAP2 is also expressed in the cells (Berg and Pongratz, 2002;
71 Kazlauskas et al., 2000; Kazlauskas et al., 2002; Petrulis et al., 2000; Petrulis et al., 2003).
Thus, it has been hypothesized that XAP2 inhibits nuclear localization of the unliganded
AHR complex. Previous studies have demonstrated that the endogenous Ahb-1 receptor
in untreated Hepa-1 exhibits a predominantly cytoplasmic location that becomes strongly nuclear following exposure to ligand (Pollenz, 1996; Pollenz et al., 1994). In contrast,
the endogenous AHR proteins expressed in A7, C2C12, or 10T1/2 embryonic fibroblasts, which express the rat (A7) or Ahb-2 receptor, exhibit a subcellular localization in untreated
cells that is both cytoplasmic and nuclear, but becomes predominantly nuclear after a 60- min exposure to TCDD (Figure A-2). Importantly, exposure of any of these cell lines to the CRM-1 nuclear export inhibitor leptomycin B (LMB) for 2 or 4 h results in an accumulation of the AHR in the nucleus that is not seen in Hepa-1 cells expressing the
Ahb-1 (Figure A-2). LMB-induced nuclear localization of the endogenous AHR was also
observed in the MCF7 human breast cancer cell line (Wentworth et al., 2004). Thus,
these previous findings show an LMB-induced accumulation of nuclear AHR that is
suggestive of rat and mouse Ahb-2 receptors that are dynamically shuttling between through the nucleus with a periodicity of 90–180 min.
Since these results indicated that the subcellular localization of the AHR and its
ability to shuttle through the nucleus was dependent on the species and strain of AHR
examined and since these findings correlated to a reduced level of XAP2 associated with
the AHR complex, it was again pertinent to assess whether the lack of shuttling seen in
the Hepa-1 was related to its genetic background or whether these biochemical properties
were specific to the AHR expressed in the cell. To evaluate this question, AHWT and
AHb2 cells were propagated on glass coverslips, exposed to TCDD or LMB, and then
72 fixed and stained for AHR protein. As was previously shown, the location of the Ahb-1
receptor in the AHWT cells was predominantly cytoplasmic and became nuclear following
1 h of TCDD treatment (Figure 2.6). In addition, as observed in the wild-type Hepa-1
lines, 4 h of LMB exposure did not influence the location of the Ahb-1 receptor. In
contrast, the unliganded Ahb-2 receptor exhibited a subcellular localization that was both
cytoplasmic and nuclear that became more strongly nuclear in the presence of LMB
(Figure 2.8, lower panels). Since the Ahb-2 receptor showed a predominately nuclear
localization prior to LMB treatment, the change in nuclear fluorescence intensity was
measured by the average level of fluorescence among a population of nuclei as detailed in
Chapter Six. In the experiment shown, the average level of fluorescence in control nuclei was 106 ± 10 (across a population of 100 cells), whereas the LMB-treated nuclei
averaged 165 ± 12 (also across a population of 100 cells). Thus, these results show that
nucleocytoplasmic shuttling of an AHR can occur in the Hepa-1 cell line, but that this
shuttling is dependent on the species of AHR that is expressed.
Furthermore, this lack of nucleocytoplasmic shuttling in the Hepa-1 line expressing the Ahb-1 receptor can be directly correlated to the high level of XAP2
association seen with this receptor species. Other studies performed in this laboratory
have demonstrated that reductions in XAP2 by siRNA treatment in the Hepa-1 line result
in an AHR that is associated with reduced levels of XAP2 similar to the situation seen in
cells expressing other AHR species (Figure A-3). Additionally, while these cells with
reduced XAP2 show the same cytoplasmic distribution of the AHR as cells transfected
with control siRNA, there is a marked increase in the nuclear localization of the AHR when cells transfected with siXAP2 are treated with LMB for 4 h, which was not
73
Figure 2.6: Subcellular localization of AHR in AHWT and AHb2 cells exposed to TCDD or LMB. Cells were grown on glass coverslips exposed to the compounds detailed below and then fixed as detailed previously (Holmes and Pollenz, 1997; Pollenz, 1996; Pollenz et al., 1994). Coverslips were incubated with A-1 IgG (1.0 µg/ml) and visualized with GAR-Rhodamine IgG (1:400). A, B, D, and E, AHWT cells exposed to Me2SO (0.1%) (A) for 1 h, TCDD (2 nM) for 1 h (B), methanol (0.5%) for 4 h (D), or LMB (20 nM) for 4 h (E). F–I, AHb2 cells exposed to Me2SO (0.1%) for 1 h (F), TCDD (2 nM) for 1 h (G); methanol (0.5%) for 4 h (H), or LMB (20 nM) for 4 h (I). C, the parental LA-I cell line stained and photographed under the identical conditions to the AHWT and AHb2 cells. All panels were exposed for identical times. Bar (A), 10 µm.
74 observed in LMB-treated cells transfected with siCON (Figure A-4). Together, these results suggest that when endogenous XAP2 is reduced in the Hepa-1 cell line, the endogenous AHR exhibits a reduced association with XAP2 that is similar to that seen in the rat and Ahb-2 receptor and, importantly, that such a reduction appears to result in a receptor capable of undergoing dynamic nucleocytoplasmic shuttling. Thus, the high levels of XAP2 associated with the Ahb-1 appear to prevent shuttling of this AHR species.
Since dynamic nucleocytoplasmic shuttling has been shown to be important level of control for many transcriptional regulators such as members of the E2F family, NFκB, and mdm2 (Cartwright and Helin, 2000), transcriptional regulation of some target genes may differ between the C57BL/6J mouse and other species.
2.5 Exogenous Expression of XAP2 Does Not Affect the Subcellular Localization of the
Ahb-1 or the Ahb-2 Receptor
Previous results from this laboratory further suggest that the entire pool of Ahb-1 receptor complexes in Hepa-1 cells is associated with XAP2 since exogenous expression of XAP2 in the Hepa-1 line results in an equivalent level of association between the Ahb-1 receptor and XAP2, though there was a large increase in cellular expression of XAP2
(Figure A-5). Additionally, when exogenous XAP2 is expressed in the Hepa-1 line, no significant alterations were seen in the localization of the Ahb-1 receptor (Figure A-6).
Likewise, transgenic animals exhibiting a hepatocyte specific increase in XAP2 expression levels also fail to exhibit increased levels of Ahb-1•XAP2 association
(Hollingshead et al., 2006). Since the Ahb-2 receptor and rat AHR appeared to be associated with a lesser degree of XAP2, even in the presence of similar cellular levels of
75 XAP2, these receptor complexes are either not all associated with XAP2, unlike the Ahb-1, or there is a reduced affinity of XAP2 for the Ahb-2 receptor leading to a weaker
association of these proteins and possible loss during immunoprecipitation analysis.
Therefore, studies were also carried out to determine whether increased expression of
XAP2 in lines expressing the Ahb-2 receptor would increase the level of association
between the Ahb-2 and XAP2 and/or inhibit nucleocytoplasmic shuttling of the Ahb-2
receptor. To establish whether Ahb-2 complexes could be generated with increased XAP2
association in the presence of exogenous XAP2 expression, AHb2 cells were transfected as above, lysed and the AHR immunoprecipitated with A-1A anti-AHR IgG. The
samples were then evaluated for AHR and XAP2 protein by Western blotting. A
representative experiment is shown in Figure 2.14 and quantified results presented in
Figure 2.7. The input samples demonstrate that AHb2 cells transfected with XAP2 exhibited an increase in the total cellular level of XAP2 compared with cells transfected
with control vector, and that such an increase in XAP2 resulted in increased association
of the Ahb-2 receptor with XAP2 (Figures 2.7 and 2.8). Indeed, when the AHR was
immunoprecipitated from these samples, the ratio between the level of AHR and the total
amount of XAP2 associated with the endogenous Ahb-2 complex was similar to that seen
in the Ahb-1 (Figure 2.8).
To establish whether these Ahb-2 complexes exhibiting increased XAP2
association would be exhibit an altered subcellular distribution, cells were transfected
with control vector or pCI-hXAP2 as detailed in Chapter Six and allowed to recover for
48 h, and subsequently harvested and analyzed by Western blotting or fixed onto
coverslips and stained for AHR or XAP2 (Figure 2.9 A). Western blot analysis of these
76
Figure 2.7: Association of the Ahb-2 receptor with XAP2 in cells expressing increased levels of XAP2. Hepa-1 (WT) or AHb2 stable lines were transfected with pCI-hXAP2 or control vector pcDNA3.1 as detailed in Chapter Six. After 24 h, populations of cells were harvested, and cytosol was generated for immunoprecipitation experiments. 600 µg of cytosol from the indicated samples was precipitated in duplicate with either AHR (AHR- IgG) or preimmune IgG (Pi-IgG) as detailed in Chapter Six. Each of the precipitated samples as well as 15 µg of cytosol (input) was resolved by SDS-PAGE and blotted. Blots were stained with either 1.0 µg/ml A-1A IgG or XAP2 mouse IgG1 (1:750), and reactivity was visualized by ECL with GAR-HRP or GAM-HRP IgG (1:10,000). mXAP2, endogenous mouse XAP2; hXAP2, exogenous human XAP2; and AHR (AHR). The IgG bands are presented to show the consistency of the precipitations. C, samples from cells transfected with pcDNA3.1; X, samples from cells transfected with pCI-hXAP2. The precipitated IgG band is shown to demonstrate the uniformity of the precipitation across all samples.
77
Figure 2.8: Quantification of association of the Ahb-2 receptor with XAP2 in cells expressing increased levels of XAP2. Computer densitometry was used to determine the relative level of AHR or XAP2 protein present in the precipitated samples from Figure 2.13. Each column represents the relative densitometry units of an individual band and can be used to evaluate differences in the ratio of XAP2/AHR in the different samples. However, because of the difference in the sensitivity of each antibody for its target protein, the ratio does not represent the absolute number of protein molecules. Note that the XAP2/AHR ratio is essentially between the WT and WT+XAP2 samples, while the relative amount of XAP2 increases in the b2 samples transfected with XAP2 to a XAP2/AHR ratio similar to WT levels. end, endogenous expression; ex, exogenous expression; tot, total expression levels. 78
b-2 Figure 2.9: Localization of Ah receptor in AHb2 cells expressing hXAP. AHb2 populations were transfected with pCI-hXAP2 (XAP2) or control vector pcDNA3.1 (CON) as detailed in Chapter Six. A, Western blot of AHR and XAP2 expression in total cell lysates prepared from transfected cells. Each of the samples was resolved by SDS- PAGE and blotted. Blots were stained with either 1.0 µg/ml A-1A anti-AHR IgG, anti-β actin IgG (1:1000), or XAP2 mouse IgG (1:750), and reactivity was visualized by ECL with GAR-HRP (1:10,000; AHR, Actin) or GAM-HRP IgG (1:10,000; XAP2). B, Duplicate populations of transfected cells were fixed and stained for the AHR with 1.0 µg/ml A-1A AHR IgG or hXAP2 IgG and visualized with GAR-RHO IgG (1:400; AHR) or GAM-FITC (1:400; XAP2). Scale bar, 10 µm.
79 samples revealed an increased level of total cellular XAP2 following transfection.
Unexpectedly, however, immunohistochemical staining of the Ahb-2 receptor in cells
transiently transfected with XAP2 failed to exhibit any changes in AHR localization from
control transfected cells (Figure 2.9 B). However, it is important to note that while the
total level of XAP2 association is these transfected cells increased relative to endogenous
Ahb-2•XAP2 association, this level remained lower than that seen with the Ahb-1. It is
noteworthy, though, that while 100% of the population of Ahb-1 complexes appeared to
be associated with XAP2 and Ahb-2 complexes appeared to be endogenously associated
with less XAP2, Ahb-2 complexes could be “forced” to associate with increased levels of
XAP2 when the cellular expression of XAP2 was greatly increased. Since an increased
association with the Ahb-2 receptor could be achieved, further studies should be
performed to evaluate the potential affect of this increased association of the Ahb-2
receptor with XAP2. Transgenic hepatocyte specific overexpression of XAP2 in the C3H
mouse expressing the Ahb-2 receptor, as has been performed in the C57BL/6J mouse
(Hollingshead et al., 2006), may be particularly insightful and may serve as a better model for examining the potential impact of XAP2 overexpression in model systems where the AHR is endogenously associated with lower levels of XAP2.
2.6 Degradation of the Mouse Ahb-2 Receptor in Hepa-1 Background Varies from the
Ahb-1 Receptor
The AHR proteins expressed in the A7 and C2C12 cells also exhibit an increased
susceptibility to TCDD-induced degradation wherein C2C12 and A7 cells treated with
TCDD for 0–120 min results in a dramatic reduction of cellular AHR protein by the 120-
80 min time point (Figure A-7). Indeed, in the C2C12 lines, AHR protein was not
detectable at the 90-min time point, and in A7 cells, the AHR protein decreased by 88%
of control levels by the 120-min time point. Additionally, previous reports have shown
that the AHR was also reduced by greater than 90% following 120 min of TCDD
exposure in the 10T1/2 cell line (Pollenz et al., 1994; Richter et al., 2001). In contrast,
TCDD-induced degradation of the Ahb-1 receptor exhibits a degradation profile of lesser
magnitude whereby AHR protein in this line is only reduced by 30–40% after 120 min.
Collectively, these results indicate that the endogenous Ahb-2 and rat receptors are
degraded to near completion within 2 h of ligand exposure and highlight another striking
difference in the response of these AHR proteins in comparison to the Ahb-1 receptor.
Since the AHR expressed in the A7 and C2C12 cells exhibited a different time course
and magnitude of degradation from the AHR expressed in the Hepa-1 cells, it was
therefore pertinent to assess the degradation of the AHR in the AHb2 cell line in
comparison to the AHWT line. To evaluate the degradation pattern of each receptor, both
lines were compared in a time course of degradation following ligand binding. Two different time course experiments were completed. In the first experiment, cells were treated with TCDD for 0, 1, 2, 4, or 6 hours and total cell lysates were harvested and
evaluated by SDS-PAGE and Western blotting for levels of AHR, CYP1A1, and actin.
Results from a representative experiment are given in Figure 2.10. In the second
experiment, cells were treated with TCDD for 0, 30, 60, 90, or 120 minutes and total cell
lysates were again harvested and evaluated by SDS-PAGE and Western blotting for
levels of AHR, CYP1A1, and actin. Results from a representative experiment are given
in Figure 2.10 B inset. In both time courses evaluated, TCDD-induced degradation in the
81
Figure 2.10: Western blot analysis of AHR in stable cell lines expressing Ahb-1 or b-2 Ah receptors exposed to TCDD. A, AHWT and AHb2 cells were treated with Me2SO (0.1%) for 6 h or TCDD (2 nM) for the indicated times. Total cell lysates were prepared, and equal amounts of protein were resolved by SDS-PAGE, blotted, and stained with A- 1A IgG (1.0 µg/ml) and actin IgG (1:1000). Reactivity was visualized by ECL with GAR-HRP (1:10,000), and protein bands were quantified and normalized as detailed (Holmes and Pollenz, 1997; Pollenz, 1996; Pollenz et al., 1994). B, the levels of AHR protein in the AHWT and AHb2 samples were divided by the corresponding level of actin, and the average ± S.E. of the three independent samples was plotted as the percentage of time 0 control. The inset shows the time course of degradation over 120 min. *, statistically different from the control siRNA; p < 0.001.
82
AHb2 line was more rapid and proceeded to a greater magnitude than that in the AHWT.
Indeed, the Ahb-2 receptor was reduced by 52% after 2 h and >85% after 6 h of TCDD
exposure compared with 28 and 74% for the Ahb-1 receptor. Overall, while the Ahb-2 receptor expressed in the AHb2 line exhibited a reduced degradation rate than that
expressed in the C2C12 line, the increased magnitude of degradation seen in the AHb2
line was more representative of that observed in the C2C12 cell line which endogenously
expresses the AHRb-2 or the rat A7 cell line than the Hepa-1 (compare Figures 2.10 and
A-7).
2.7 Time Course of CYP1A1 Induction is Unaltered by Ah Receptor Type
Since nucleocytoplasmic shuttling and a more rapid degradation profile could be
impacting target gene induction by the Ahb-2 compared to the Ahb-1 following ligand
binding, the ability of each receptor to induce CYP1A1 was assessed. LA-I cells not
expressing any exogenous AHR were also examined to insure that the levels of CYP1A1 protein were not influenced by the presence of any endogenous receptor in the LA-I line.
Figure 2.11 shows that CYP1A1 was detectable within 2 h of TCDD exposure in both cells lines and increased proportionally over time. To determine whether there were any
quantitative differences in the induction of CYP1A1 protein in the two cell lines, cells
were treated with TCDD for 16 h, and identical levels of total cell lysates were evaluated by quantitative Western blotting. Interestingly, the level of CYP1A1 protein induced
after 16 h was not statistically different between the AHWT and AHb2 cell lines although
the Ahb-2 receptor was associated with less XAP2 and degraded with a more rapid profile
(Figures A-1, 2.5, A-7, and 2.10). Thus, in this model system, a reduction in the level of
83
Figure 2.11: Western blot analysis of CYP1A1 protein in stable cell lines expressing b-1 b-2 Ah or Ah receptors following exposure to TCDD. A, AHWT and AHb2 cells were treated with Me2SO (0.1%) for 6 h or TCDD (2 nM) for the indicated times. Total cell lysates were prepared, and equal amounts of protein were resolved by SDS-PAGE, blotted, and stained with rat CYP1A1 IgG (1.0 µg/ml) and actin IgG (1:1000). Reactivity was visualized by ECL with GAR-HRP (1:10,000). B, LA-I, AHWT, and AHb2 cells were treated with Me2SO (0.1%) or TCDD (2 nM) for 16 h. Total cell lysates were prepared, and equal amounts of protein were resolved by SDS-PAGE, blotted, and stained with rat CYP1A1 IgG (1.0 µg/ml) and actin IgG (1:1000). Reactivity was visualized by ECL with GAR-HRP (1:10,000), and protein bands were quantified and normalized as detailed (Holmes and Pollenz, 1997; Pollenz, 1996; Pollenz et al., 1994). C, the levels of CYP1A1 protein in the LA-I, AHWT, and AHb2 samples were divided by the corresponding level of actin, and the average ± S.E. of the three independent samples was potted as normalized densitometry units. Con, average density of Me2SO-treated cells. 84
Figure 2.12: Analysis of dose-response for CYP1A1 induction in stable cell lines b-1 b-2 expressing Ah or Ah receptors. A, Triplicate plates of AHWT or AHb2 cells were dosed with 1, 10, 100, or 1000 pM TCDD for 6 hr at 37° as detailed in Chapter Six. Total cell lysates were prepared, and equal amounts of protein were resolved by SDS- PAGE, blotted, and stained with A-1A IgG (1.0 µg/ml), rat CYP1A1 IgG (1.0 µg/ml), and actin IgG (1:1000). Reactivity was visualized by ECL with GAR-HRP (1:10,000), and protein bands were quantified and normalized as detailed (Holmes and Pollenz, 1997; Pollenz, 1996; Pollenz et al., 1994). B, the levels of CYP1A1 protein in the AHWT and AHb2 samples were divided by the corresponding level of actin, and the average of the three independent samples was plotted compared to the time 0 control. Each data point represents the average of the triplicate normalized samples for each TCDD concentration. Note that the EC50 for CYP1A1 induction in both AHWT or AHb2 cells is similar an close to the expected 30-50 pM range.
85 XAP2 in the core AHR complex did not appear to impact induction of the endogenous
CYP1A1 gene. To confirm these CYP1A1 induction levels, dose-response studies were
performed using graded doses of TCDD on cell lines in culture. Following the
incubation, cells were harvested and evaluated for CYP1A1 protein and actin expression
by quantitative Western blotting as detailed previously. Representative results are shown
in Figure 2.12. From these studies, it was determined that the AHWT and AHb2 lines
exhibited a similar EC50 (~40 pM) for CYP1A1 induction in response to TCDD that was
within the 30-50 pM published range (Pollenz, 1996). WT Hepa-1 cells were included as
a control, and exhibited similar results as those obtained using the AHWT line (data not
shown). Together, these data further support the hypothesis that properties of the
receptor itself rather than other cellular factors are mostly responsible for subcellular localization and degradation following ligand binding.
2.8 AHb2 Stable Line Conclusions
When considered in the context of the results presented in Figures 2.5-2.19, these
findings collectively show that the Ahb-2 receptor when in the same C57BL/6J genetic
background as the Ahb-1 exhibits distinct susceptibility to ligand exposure and
nucleocytoplasmic shuttling when compared with the Ahb-1 receptor. Additionally, these
biochemical properties of the Ahb-2 receptor mimic those of the Ahb-2 endogenously
expressed in the C2C12 cell line. Furthermore, these studies indicate that these
differences are likely related to the receptor-dependent reduced association of the Ahb-2 receptor with XAP2 rather than cellular context. Studies focusing on the function of the
86 AHR in the C57BL/6Jmouse or Hepa-1 cells are therefore not universal descriptions of
AHR function and may not be entirely representative of receptor function in the human.
87
Chapter Three
Generation and Characterization of Stable Lines Expressing Ah-Receptors Deficient in
DNA-Binding or ARNT Dimerization
3.1 Rationale for AHR Mutant Lines
Agonist binding to the AHR results in degradation of the receptor following DNA
binding, but not of the complex’s components: ARNT, XAP2, p23, or Hsp90 (Cioffi et al., 2002; Giannone et al., 1998; Pollenz, 2002; Roman and Peterson, 1998; Roman et al.,
1998). While it has been well-established that the AHR is targeted for degradation following ligand activation, the mechanism of this event remains unclear (Dale and
Eltom, 2006; Davarinos and Pollenz, 1999; Giannone et al., 1998; Ma, 2007; Ma and
Baldwin, 2000; Ma et al., 2000; Morales and Perdew, 2007; Ohtake et al., 2007; Poland and Glover, 1988; Pollenz, 2002; Pollenz, 2007; Pollenz and Dougherty, 2005; Pollenz et al., 2005; Song and Pollenz, 2002; Song and Pollenz, 2003; Wentworth et al., 2004).
Following TCDD binding, the AHRb-1 is rapidly depleted by 80-95% within 4-6 hours of
treatment in numerous cell culture models and does not return to basal levels as long as ligand is present in the media (Pollenz, 1996; Reick et al., 1994). This degradation is even more rapid with the Ahb-2, human or rat receptors, depleting by 90-100% within 2
hours of treatment (Pollenz and Dougherty, 2005). In either case, this degradation has
been demonstrated to occur via the 26S proteasome complex since pre-treatment with the
88 proteasome inhibitors MG-132 or lactacystin prior to treatment with agonist blocks
degradation of the receptor, while pre-treatment with inhibitors of calpain, serine, or cysteine proteases or lysosomal proteases does not inhibit the degradation response
(Davarinos and Pollenz, 1999; Ma and Baldwin, 2000; Wentworth et al., 2004).
Furthermore, ligand-dependant degradation can also be blocked by pre-treatment with the
transcription inhibitor actinomycin D (AD) or the translation inhibitor cycloheximide
(CHX) without affecting nuclear localization or DNA binding of the receptor, suggesting
that both active transcription and translation are necessary for ligand-induced degradation
of the AHR (Pollenz et al., 2005). A summary of the potential pathways leading to AHR
degradation is given in Figure 3.1.
Thus, it was pertinent to assess whether ligand-induced degradation of the AHR,
which appears to require active transcription and translation, required active transcription
of the AHR•ARNT complex itself. If DNA binding is necessary for degradation of the
receptor by the agonist-dependent pathway, than mutations or truncations of the DNA
binding domain should impact the ability of the receptor to be degraded. Similarly, if a
region responsible for dimerization with ARNT were mutated, DNA binding would be
prevented by loss of ARNT heterodimerization and should impact degradation of the
receptor.
To formally investigate the impact that DNA binding and ARNT dimerization
mutations would have on the degradation of the AHR, and whether the differences in
degradation were due to the receptor itself or other cellular factors, stable lines were
generated exploiting the LA-I, AHR deficient cells as described in Chapter Six. The use
89
90
Figure 3.1: Schematic of the AHR ligand-dependent and ligand-independent degradation pathways. In brief, ligands such as halogenated aromatic hydrocarbons (HAHs), polychlorinated biphenyls (PCBs), polyaromatic hydrocarbons (PAHs) enter a cell, bind to the AHR complex leading to a conformational change of the receptor, which then translocates to the nucleus and heterodimerizes with ARNT. The AHR•ARNT complex then associates with xenobiotic response elements with the core sequence 5’- T(T/A)GCGTG to regulate phase I and phase II metabolizing enzymes, typified by CYP1A1. The AHR is then targeted for degradation via the 26S proteasome in the nucleus (Roberts and Whitelaw, 1999), but degradation can also occur following nuclear export via CRM-1 (dashed line; Davarinos and Pollenz, 1999). In ligand-independent degradation, hsp90 inhibitors such as geldanamycin and radicicol bind to Hsp90, resulting in nuclear accumulation of the AHR and subsequent degradation via the 26S- proteasome without DNA binding or ARNT dimerization (Chen et al., 1997); however, this degradation is independent of nuclear import and can occur in the cytoplasm (dashed line; Song and Pollenz, 2002). Note that while ligand-dependant degradation can be blocked by pre-treatment with the transcription inhibitor actinomycin D (AD) or the translation inhibitor cycloheximide (CHX), degradation of the AHR via GA cannot be blocked by treatment with either AD or CHX (Pollenz et al., 2005).
91 of a genetically identical background to examine the function and degradation rate of the
mutant receptors in comparison with the wild-type AHRb-1 provided significant insights
into these questions.
3.2 Generation of LA-I Lines Expressing AHR DNA Binding Mutants and ARNT
Dimerization Mutants
The first step in these studies was the evaluation of potential sites for mutation or truncation in the AHR that would disrupt DNA binding or ARNT dimerization as desired.
The AHR basic domain spans amino acid residues 13-39 and appears to make 2 amino acid base contacts with DNA using tyrosine 9 and arginine 39 (Bacsi and Hankinson,
1996; Fukunaga et al., 1995). Based on this data, a construct was generated in which the amino-terminal portion of the AHR was truncated resulting in the loss of the first 40 amino acids (trAHR). This construct was generated and stably integrated into the LA-I
line as described in Chapter Six. Additionally, a second construct was generated in which the AHR arginine 39 residue was mutated to an alanine, which has previously been
shown to reduce the DNA binding potential of the AHR to approximately 1±1% versus wild-type receptor (Bacsi and Hankinson, 1996; Figure 3.2). For the ARNT dimerization mutants, a construct which had previously been evaluated by this laboratory (Pollenz and
Barbour, 2000) was used, wherein two leucine residues within a portion of the spacer region between the PAS-A and PAS-B domains of the AHR, thought to contain a second nuclear export site (NES), were mutated to alanines leading to an inability to dimerize with ARNT (AHR-L69/71A, referred to as AHR∆NES; Figure 3.2).
92
Figure 3.2: Schematic of AHR mutations for R39A and ∆NES AHR mutant stable lines. AHR protein domain schematic indicating the amino acid positions of the mutagenized residues and the locations of the characterized domains. Thin horizontal lines indicate functional domain regions. NLS, nuclear localization sequence; NES, nuclear export sequence; bHLH, basic-helix-loop-helix; PAS, PER/ARNT/SIM homology regions; PAC, PAS-associated c-terminal region; TAD, transactivation domain; P/S/T-rich, proline, serine and threonine-rich sequence. Numbers underneath represent amino acid positions. Note that while the Ahb-1 receptor is shown as an example, the R39 and the L69/71 are conserved among all characterized murine AHR alleles.
93 Constructs were then tested by transient transfection into the AHR-deficient LA-I Hepa-1 variant or in PT67 viral packaging cells, then analyzed by SDS-PAGE and evaluated by
Western blotting (Figure 3.3 A and B). Western analysis also validated that the
molecular mass of each protein was correct. Following transfection of viral packaging
cells, viral media was collected and used to infect target cells and stable lines surviving a
two-week selection in neomycin were then analyzed for expression of the target genes by
SDS-PAGE and Western blotting (Figure 3.3 C and D). Stable lines expressing the trAHR, the R39A, and the ∆NES AHR proteins were then selected and those that
expressed the receptor to the same physiological level as the AHRb-1 in WT Hepa-1 cells
were chosen for future study. Total cell lysates of each stable line revealed no
differences in total cellular levels of AHR, ARNT or Hsp90 (Figure 3.4). Note that while
the cellular content of the trAHR line is not shown, no differences were observed.
3.3 Characterization of AHR Mutants Deficient in DNA Binding and ARNT Dimerization
Stable cell lines expressing the trAHR were then treated with TCDD for 0, 1, 2, 4, or 6 hours and total cell lysates harvested, assayed for total protein content, and evaluated by SDS-PAGE and Western blotting for levels of AHR, CYP1A1, and actin. Results from a representative experiment are given in Figure 3.5. However, there was no apparent difference in the rate of degradation between the AHRb-1 and the trAHR receptors when expressed in Hepa-1, suggesting that DNA binding of the AHR was not relevant to the ability of the liganded receptor to undergo degradation. However, the trAHR, which was localized to the cytoplasm as expected, continued to become predominately nuclear in the presence of ligand, even though the NLS was removed in
94
Figure 3.3: Western analysis of AHRtr, R39A, and ∆NES AHR mutants. Equal amounts of total cell lysates from independent samples were resolved by SDS-PAGE, blotted, and stained with A-1A rabbit IgG (1.0 µg/ml) and -actin rabbit IgG (1:1000). Reactivity was visualized by ECL with GAR-HRP IgG (1:10,000). A, LA-I cells transiently expressing the AHRtr compared to WT cells endogenously expressing the Ahb-1 receptor. B, PT67 cells transiently stably expressing the R39A or ∆NES receptor compared to several AHR controls. C, D, Stable line analysis of the AHRtr, R39A, and ∆NES AHR constructs. Arrows indicate examples of clones used for subsequent analysis based on similar expression levels of the AHR in WT Hepa-1 cells.
95
Figure 3.4: Western analysis of R39A and ∆NES AHR mutant stable lines. Stable cell lines expressing the R39A AHR, the ∆NES AHR, and the wild-type AHR (A3) were generated as detailed in Chapter Six. Equal amounts of total cell lysates from LA-I, Hepa-1 (WT), ∆NES, R39A, or A3 cells were resolved by SDS-PAGE, blotted, and stained with A-1A rabbit IgG (1.0 µg/ml), -actin rabbit IgG (1:1000), or hsp90 rabbit IgG (1:500). Reactivity was visualized by ECL with GAR-HRP IgG (1:10,000).
96
Figure 3.5: Time course of ligand-induced degradation of AHRWT and trAHR. A, The indicated cell lines were treated with 0.5% Me2SO (time 0 control), or TCDD (5nM) for the indicated times (hr) at 37°C. Dosing was staggered so that all cells were harvested at the same time. Equal amounts of total cell lysates were resolved by SDS- PAGE, blotted and stained with A-1A anti-AHR IgG (1.0μg/ml), anti-P450 (1:750) and anti-β actin (1:1000). Reactivity was visualized by ECL with GAR-HRP (1:10,000). B, The level of AHR protein at each time point was divided by the corresponding level of actin for normalization purposes and the average ± S.E. of the three independent samples plotted as a function of the vehicle control cells (100%).
97 the truncation and this removal confirmed by DNA sequencing analysis (data not shown).
Thus, the truncated AHR did not appear to be functioning as expected. However, the
trAHR did appear to be unable to induce CYP1A1 protein in the presence of ligand
(Figure 3.5). Since the truncation of 40 amino acids from the NH-terminus of the AHR
could be affecting the folding of the AH receptor or the interaction of the AHR with other proteins, leading to seemingly confounding data, an alternate approach was taken to evaluate the degradation of a full-length AHR deficient in DNA binding.
For this approach, an AHR construct was generated in which the arginine 39 residue was mutated into an alanine (R39A) as previously described. As before, the functional ability of the R39A mutant was characterized and the degradation pattern assessed. Since the R39A mutant had already been thoroughly characterized in its deficiency to bind DNA when coexpressed with ARNT and treated with TCDD (Bacsi and Hankinson, 1996), studies were initiated to confirm the lack of function in vivo by evaluating the ability of the R39A stable line to induce CYP1A1 following treatment with TCDD. A representative experiment is given in Figure 3.6. Similar to the trAHR construct, the absence of CYP1A1 induction in Hepa-1 cells stably expressing the R39A
AHR following 6 or 16 hrs of TCDD treatment indicates the inability of the mutant AHR to bind DNA. Since this study only evaluated the ability of either construct to induce a single gene, studies were initiated to further support the hypothesis that neither the R39A nor the ∆NES AHR constructs were able to function in target gene induction,. For these studies, stable lines expressing the wild-type AHR, the R39A, or the ∆NES AHR constructs were transfected with the XRE controlled GudLuc 1.1 luciferase reporter and pSV-β-Galactosidase as detailed in Chapter Six. A representative experiment is given in
98
Figure 3.6: Overday or overnight ligand-induced degradation of the AHRwt, R39A, or ∆NES stable lines. The indicated cell lines were treated with 0.5% Me2SO (time 0 control), or TCDD (5nM) for 6 (+) or 16 hours (++) at 37°C. Dosing was staggered so that all cells were harvested at the same time. A, Equal amounts of total cell lysates were resolved by SDS-PAGE, blotted and stained with A-1A anti-AHR IgG (1.0μg/ml), anti- P450 (1:250; Santa Cruz) and anti- β actin (1:1000). Reactivity was visualized by ECL with GAR-HRP (1:10,000). B, Computer densitometry was used to determine the relative level of AHR protein present in the Me2SO or TCDD samples presented on the blot in A for the AHRWT and R39A stable lines. The level of AHR present was normalized to the level of actin and each column represents the relative densitometry units of an individual band and shows that the degradation of the R39A AHR is minimal compared to wild-type AHR. 99 Figure 3.7. While stable lines expressing the wild-type AHR were capable of inducing
very high levels of luciferase activity following TCDD treatment, neither the R39A nor
the ∆NES lines were capable of driving luciferase expression above that of the AHR-
deficient LA-I parental line. Indeed, these lines exhibited TCDD-dependent luciferase
activity that was >40-fold lower than that of the lines expressing wild-type AHR.
However, unlike the trAHR, the R39A AHR remained predominately cytoplasmic and exhibited nuclear accumulation only following 1 hr of TCDD treatment (Figure 3.8).
Since the R39A AHR should be capable of dimerizing with ARNT, it is expected that such a nuclear accumulation should occur in the presence of TCDD. Interestingly, the
∆NES lines appeared to express a receptor that was predominately nuclear even in the absence of TCDD (Figure 3.8 middle panels). However, since association of the AHR with XAP2 appears to require the PAS B/ligand binding domain of the AHR (Meyer and
Perdew, 1999), which partly overlaps with the sequence removed in the ∆NES construct, it is possible that the ∆NES AHR would exhibit a reduced association with XAP2, and would therefore be capable of shuttling through the nucleus as seen with the Ahb-2 receptor in Chapter Two. However, this hypothesis has not yet been tested.
To confirm that the R39A AHR remained capable of dimerizing with ARNT and to further confirm that the ∆NES AHR could not, cytosol was prepared from R39A or
∆NES cell lines treated with 0.5% Me2SO or 10μM TCDD and subjected to
immunoprecipitation analysis using anti-AHR antibodies as detailed in Chapter Six
(Figure 3.9). In these studies, ARNT was specifically immunoprecipitated in the TCDD
treated R39A cell lines and failed to precipitate with AHR antibodies in the Me2SO
treated samples or when pre-immune antibodies were used. Furthermore, cytosol
100
Figure 3.7: Analysis of TCDD-induced luciferase activity in LA-I, AHWT, R39A, and ∆NES stable lines. The indicated cell lines were transfected with pSV- β -Galactosidase along with GudLuc 1.1 and treated with Me2SO (0.5%) or TCDD (5nM) for 6 hours. Luciferase activity and β-Galactosidase activity were measured as previously described in Chapter Six and (Zeruth and Pollenz, 2007). All luciferase values were normalized to β - Galactosidase. 101
Figure 3.8: Subcellular localization of the ∆NES and R39A AHR. Cells were grown on glass coverslips, exposed to Me2SO (0.5%) or TCDD (5nm) for 1 hr and then fixed as detailed in Chapter Six. Coverslips were incubated with anti-AHR A-1 IgG (2.8μg/ml) and visualized with GAR-RHO IgG (1:400). The top row shows control LA-I, ∆NES, or the R39A stable lines exposed to Me2SO. The bottom row shows those cells exposed to TCDD (5nM) for 1 hour (+). Each set of panels was exposed for identical times.
102
Figure 3.9: Immunoprecipitation analysis of ARNT association in ∆NES and R39A stable lines. Cytosol was prepared from ∆NES and R39A lines treated with 0.5% Me2SO (-) or 10uM TCDD (+) as detailed in Chapter Six. 800μg of cytosol was precipitated with either high affinity anti-AHR A-1A IgG (5μg; A) or affinity pure pre- immune rabbit IgG (5μg; Pi) along with protein A-G agarose (25ul) for 2.5 hours at 4°C with rocking. Pellets were washed 3 times for 5 minutes each with TTBS supplemented with sodium molybdate (20mM) and then boiled in 30ul SDS sample buffer. 15μg cytosol (input) or 15ul of the eluted protein were resolved by SDS-PAGE, blotted, and stained for ARNT (R1 IgG at 1μg/ml). Reactivity was measured by ECL with GAR-HRP (1:10,000). A, precipitated with ARNT; Pi, precipitated with pre-immune IgG. Note the precipitation of ARNT in the treated R39A samples that is absent in the ∆NES samples. The precipitated IgG band is shown to demonstrate the uniformity of the precipitation across all samples.
103 prepared from 0.5% Me2SO or 10μM TCDD ∆NES stable lines confirmed the lack of
ARNT dimerization in this mutant since ARNT was not co-precipitated with the AHR in
any sample type (Figure 3.9). Thus, while the R39A AHR remained capable of
associating with ARNT, the mutant receptor failed to function in CYP1A1 induction,
while the ∆NES AHR was neither capable of dimerizing with ARNT or binding DNA.
Additionally, cytosolic and nuclear extracts prepared from the wild-type, R39A or
∆NES stable lines treated with 0.5% Me2SO or 10nM TCDD revealed that while the
AHR and ARNT co-accumulated in nuclear extracts of cell lines expressing the wild-type
AHR, in cells expressing AHR proteins deficient in DNA binding or ARNT dimerization,
ARNT failed to accumulate in nuclear extracts following TCDD treatment (Figure 3.10).
While ARNT is a nuclear protein, in its latent state, ARNT is not tightly associated with
nuclear structures and therefore has a tendency to “leak” from nuclear extracts into the
cytosolic compartment (Pollenz et al., 1994). Thus, this lack of nuclear accumulation is a
result of the failure of ARNT to be recruited to bind DNA rather than being indicative of
nuclear ARNT. Taken together, the collective results shown in Figures 3.4-3.8 reveal
that the R39A and ∆NES stable lines express Ah receptor that is deficient in either DNA
binding (R39A) or ARNT dimerization (∆NES), but in each case remains otherwise functional. Thus the ability of the AHR receptor to be degraded under these conditions can now be examined.
3.4 Ligand-Dependent Degradation of the AHR Requires DNA Binding
To initially examine the degradation profile of these constructs, stable cell lines expressing the R39A or ∆NES mutation were treated for 6 or 16 hours with TCDD and
104
Figure 3.10: Analysis of accumulation of AHR and ARNT in cytosolic and nuclear extracts from AHWT, R39A, and ∆NES stable lines. The indicated stable cell lines were treated with Me2SO (0.5%; -) or TCDD (10nM; +) for 1 h at 37°C. Cytosolic (C) and nuclear extracts (N) were prepared as detailed in Chapter Six. Equal amounts of the indicated cytosolic and nuclear extracts were combined with 2X gel sample buffer and resolved by SDS-PAGE, blotted and stained with anti-AHR A-1A IgG (1.0μg/ml) or anti- ARNT R1 IgG (1.0 μg/ml). Reactivity was visualized by ECL with GAR-HRP (1:10,000). The bold arrow on the bottom indicates the accumulation of both AHR and ARNT in the nuclear extracts from TCDD treated AHRWT cells.
105 the resulting protein levels analyzed by SDS-PAGE and Western blotting and compared
to those seen in the AHWT line (Figure 3.11). Interestingly, neither line exhibited >30%
degradation of the AHR at either 6 or 16 hours. To confirm this lack of degradation, 4
and 5 hour time points of TCDD treatment were also examined with neither point
exhibiting >30% reduction in AHR (data not shown). Taken together, this suggests that
the DNA-binding event is itself necessary for ligand-induced degradation of the AHR and
by preventing DNA binding either directly or by inhibiting dimerization with ARNT, the
ability of the AHR to degrade is limited.
Since a modest level of degradation of the AHR was seen in both the R39A and
∆NES lines in the presence of TCDD, it was also pertinent to assess the mechanism of
this degradation and whether this small level of degradation was occurring in a ligand-
dependent manner similar to loss of the wild-type AHR; albeit to a lesser magnitude. To
determine if this small loss of mutant receptor was occurring via the ligand-dependant or
ligand-independent pathway, further studies were performed using actinomycin D (AD),
a known blocker of ligand-induced degradation of the AHR (Pollenz et al., 2005).
Interestingly, the 30% loss of the mutant AHR proteins exhibited in the presence of
TCDD was unable to be blocked by pretreatment of cells with AD prior to dosing with
TCDD in either the ∆NES or R39A line (results for R39A not shown), though neither
was it able to fully block degradation in the AHWT line, which still exhibited a >15% loss
of receptor (Figure 3.11). Thus, while the minimal loss of the receptor in the ∆NES and
R39A occurred in a ligand-dependent manner, it did not appear to be occurring via the same ligand-induced mechanism seen in the AHWT line. However, this assessment is
difficult since pretreatment with AD did not completely inhibit degradation of the AHR
106
Figure 3.11: Impact of actinomycin D on TCDD-induced degradation of AHRwt and ∆NES stable lines. Indicated cell lines (AHRwt, top; or AHR ∆NES, bottom) were treated normally with 0.5% Me2SO (CON) or TCDD (5nM) for 4 hours (TCDD), with the AD (0.3uM) inhibitor for 5 hours (AD), or pre-treated with AD for 1 hour followed by treatment with TCDD (5nM) for an additional 4 hours (AD/TCDD). A, Equal amounts of total cell lysates were resolved by SDS-PAGE, blotted and stained with A-1A anti-AHR IgG (1.0μg/ml) and anti-β actin (1:1000). Reactivity was visualized by ECL with GAR-HRP (1:10,000). B, Computer densitometry was used to determine the relative level of AHR protein present in the each sample treatment type presented on the blot in A for the AHRWT and ∆NES stable lines. All AHR levels were normalized to actin controls. 107 in the AHWT line, and, therefore, the low level of degradation seen in the AHWT line may
be occurring by the same mechanism of that of the AHR in the ∆NES and R39A lines.
3.5 Ligand-Independent Degradation of the AHR Does Not Require DNA Binding
Ligand-independent degradation of the AHR is typified by treatment with
geldanamycin (GA), a benzoquinone ansamycin capable of binding to the ATP-binding
pocket of Hsp90 and thereby likely altering the conformation of the Hsp90 associated
AHR to allow for nuclear translocation of the receptor and its subsequent degradation
without disruption of the AHR complex itself and, importantly, without DNA binding or
subsequent gene induction (Chen et al., 1997; Meyer et al., 2003b; Song and Pollenz,
2002). This pathway is characterized by a rapid and robust degradation profile (>80%
within 1 hour) that also appears to occur via the 26S proteasome and can be blocked by
MG-132 or lactacystin (Song and Pollenz, 2002). Additionally, degradation of the AHR
via GA cannot be blocked by treatment with either AD or CHX, suggesting that multiple
mechanisms exist for the degradation of the receptor, though both terminate at the 26S
proteasome (Pollenz et al., 2005).
Since ligand-independent degradation of the AHR induced by GA results in
degradation of the AHR without DNA binding, it was expected that AHR protein
deficient in DNA binding or ARNT dimerization would continue to exhibit the same time
course and magnitude of GA-induced degradation as wild-type receptor. Thus, AHWT,
R39A, and ∆NES stable lines were treated with 200uM GA for 0, 60, or 120 min and the level of AHR protein assessed and compared to that of Hepa-1 lines endogenously expressing the Ahb-1. Results from a typical experiment are presented in Figure 3.12. In
108
Figure 3.12: Geldanamycin-induced degradation of wild-type Hepa-1 cells, AHWT, R39A, or ∆NES stable lines. The indicated cell lines were treated with 0.5% Me2SO (time 0 control), or Geldanamycin (200nM) for 60 or 120 min at 37°C. Dosing was staggered so that all cells were harvested at the same time. A, Equal amounts of total cell lysates were resolved by SDS-PAGE, blotted and stained with A-1A anti-AHR IgG (1.0μg/ml) and anti-β actin (1:1000). Reactivity was visualized by ECL with GAR-HRP (1:10,000). B, Computer densitometry was used to determine the relative level of AHR protein present in the each sample treatment type presented on the blot in A for the Hepa- 1, AHRWT, R39A and ∆NES stable lines. All AHR levels were normalized to actin controls. Hepa-1, grey diamonds; AHRWT, open circles; ∆NES, black triangles R39A, white squares.
109 contrast to the minimal degradation of the R39A or ∆NES AHR in the presence of
TCDD, in the presence of GA, the R39A and ∆NES AHR proteins exhibited a similar
time course and magnitude of degradation over 120 minutes in a ligand-independent
manner. As expected, total AHR levels in all cell types were reduced by >70% within 60
min and by nearly 100% within 120 min (Figure 3.12). The similarity of both the time
course and magnitude of this degradation in the absence of TCDD suggests that the ligand-independent mechanism of AHR degradation induced by GA remains intact regardless of the ability of the receptor to function as a transcription factor. Therefore,
GA-induced degradation of the AHR does not appear to require DNA binding of the
AHR. Furthermore, these studies indicate that the lack of degradation of the AHR in the
R39A and ∆NES is not a result of an inability of the mutant AHR to be degraded.
3.6 Impact of NH-Terminal Tags on AHR Degradation
During the creation of stable lines expressing amino-terminally tagged AHR proteins to be used in another set of studies, it became apparent that the presence of these tags was altering the degradation rate of the AHR following ligand binding. Figure 3.13 and 3.14 illustrates the schematic of the tagged AHR proteins as well as their expression
levels in several cell lines stably expressing the untagged AHR (AHWT), His Max tagged
AHR (HMAHR; a Hexahistidine tag with Xpress epitope), or GFP-tagged AHR (GFPAHR; green fluorescent protein), which were generated as detailed in Chapter Six. After dosing these lines, or other independent stable lines expressing either tagged AHR, with TCDD for 0, 2, 4, 6, 8 or 16 hours, the amino-terminally tagged AHR proteins exhibited a marked reduction in the overall level of degradation when compared to stable lines
110
Figure 3.13: Invitrogen plasmid maps for generating constructs with NH-terminal tags. Target genes are expressed from the cytomegalovirus immediate early (CMV IE) promoter and the plasmids contain both neomycin (G418) and ampicillin resistance used for antibiotic selection in eukaryotic and prokaryotic cells respectively. Top: Constructs were generated with NH-terminal EYFP tags by cloning products in frame with the EYFP sequence present on the EYFP fusion vector. Bottom: Constructs were generated with NH-terminal HisMax tags by TOPO cloning products in frame with the 6xHis/Xpress sequence present on the pcDNA 4/HisMax vector. Vectors were obtained from the manufacturer (Invitrogen; Carlsbad, California).
111 expressing the untagged AHR (Figure 3.15). Indeed, while the AHWT exhibited an approximately 70-80% reduction in total AHR levels following 6 hours of TCDD TCDD treatment, which appeared to recover to nearly time 0 levels by 16 hours, indicating a lesser magnitude of degradation for these receptor types. Interestingly, while the overall magnitude of degradation in the HMAHR and GFPAHR lines was reduced compared to
untagged AHR, these receptors exhibited a greater level of degradation following TCDD
treatment than the non-functional AHR mutants (R39A and ∆NES), and maintained the
ability to induce CYP1A1 to nearly the same level as wild-type AHR (Figure 3.16).
Thus, the magnitude of ligand-induced degradation of the AHR can be affected without
creating non-functional AHR types and functional ability of the receptor is not the sole
determinant for AHR degradation.
Furthermore, treatment with GA revealed no difference in the time course or
magnitude of the degradation of the AHR in lines expressing the untagged AHR, the
HMAHR, or the GFPAHR with all three lines exhibiting a >80% reduction in total AHR
levels after 120 min of GA-treatment, similar to the results obtained in the previously
detailed studies (Figure 3.17). These results again indicate a different mechanism for the
ligand-induced and ligand-independent degradation of the AHR and further suggest that
the mechanism for ligand-independent degradation may have little to do with the AHR
itself, since numerous AHR constructs differing in size and in functional ability all
exhibit similar magnitudes of GA-induced degradation.
112
Figure 3.14: Characterization of HM- and GFP-tagged AHR stable lines. A, Schematic showing location and size of the HisMax and GFP tags. B, Equal amounts of total cell lysates from independent clones were resolved by SDS-PAGE, blotted, and stained with A-1A anti-AHR rabbit IgG (1.0 µg/ml) and anti- -actin rabbit IgG (1:1000). Reactivity was visualized by ECL with GAR-HRP IgG (1:10,000). LA-I, mutant Hepa-1 cells deficient in AHR expression; WT, wild-type Hepa-1 cells; AHWT, LA-I clones b-1 stably expressing the Ah receptor; HMAH, LA-I clones stably expressing the HisMax tagged AHR; GFPAH, LA-I clones stably expressing the GFP tagged AHR. Note that each lane represents an independent clone.
113 A TCDD Exposure (hr) 024816
AHRWT AHR Actin
AHR HMAHR Actin
AHR GFPAHR Actin B
AHRWT HMAHR
GFPAHR
100
80 AHR Protein 60
(% of time 0) 40
20 Normalized
0 0246 8101214 16
TCDD exposure (hours)
Figure 3.15: Ligand-induced AHR degradation in HM- and GFP-tagged AHR stable lines. A, Stable cell lines expressing the untagged AHR, the HisMax tagged AHR, or the GFP-tagged AHR were dosed with Me2SO (0.5%, time 0 control) or 10 nM TCDD for 2, 4, 6, 8 or 16 hours. Equal amounts of total cell lysates from independent samples were resolved by SDS-PAGE, blotted, and stained with A-1A anti-AHR rabbit IgG (1.0 µg/ml) and anti- -actin rabbit IgG (1:1000). Reactivity was visualized by ECL with GAR-HRP IgG (1:10,000). B, The amount of AHR protein at each time point was analyzed by computer densitometry of the blots shown in A as detailed in Chapter Six. Results were normalized to the level of actin and are expressed as the percentage of protein seen at b-1 time 0 ± SEM. AHWT, LA-I clones stably expressing the Ah receptor; HMAH, LA-I clones stably expressing the HisMax tagged AHR; GFPAH, LA-I clones stably expressing the GFP tagged AHR. Note that each lane represents an independent clone. treatment, both the HMAHR and GFPAHR exhibited a nadir of 42% AHR after 4 hours of 114
Figure 3.16: Ligand-induced CYP1A1 induction in HM- and GFP-tagged AHR stable lines. A, Stable cell lines expressing the untagged AHR (A), the HisMax tagged AHR (H), or the GFP-tagged AHR (G) were dosed with Me2SO (0.5%, time 0 control) or 10 nM TCDD for 2, 4, 8 or 16 hours (+). Equal amounts of total cell lysates from independent samples were resolved by SDS-PAGE, blotted, and stained with A-1A anti- AHR rabbit IgG (1.0 µg/ml), anti-CYP1A1 IgG (1:750) and anti- -actin rabbit IgG (1:1000). Reactivity was visualized by ECL with GAR-HRP IgG (1:10,000). B, The amount of CYP1A1 protein after TCDD treatment was analyzed by computer densitometry of the blots shown in A as detailed in Chapter Six. Results were normalized to the level of actin and are expressed as the percentage of protein compared to the untagged AHR control.
115 A GA Exposure (min) 030 60 90 120 AHR AHR Actin AHR GFPAHR Actin AHR HMAHR Actin B AHR GFP-AHR
100 HM-AHR
80
60
(% of time 0) 40
Normalized AHR Protein 20
0 0 30 60 90 120
GA Exposure (min)
Figure 3.17: Geldanamycin-induced degradation of the HM- and GFP-tagged AHR. The indicated cell lines were treated with 0.5% Me2SO (time 0 control), or Geldanamycin (200nM) for 30, 60, 90 or 120 min at 37°C. Dosing was staggered so that all cells were harvested at the same time. A, Equal amounts of total cell lysates were resolved by SDS- PAGE, blotted and stained with A-1A anti-AHR IgG (1.0μg/ml) and anti-β actin (1:1000). Reactivity was visualized by ECL with GAR-HRP (1:10,000). B, Computer densitometry was used to determine the relative level of AHR protein present in the each sample treatment type presented on the blot in A. All AHR levels were normalized to actin controls.
116
Chapter Four
Functional Analysis of ARNT2 in AHR-mediated Signal Transduction
4.1 Rationale for Evaluating ARNT2 Function
ARNT is a member of the basic-helix loop-helix PER/ARNT/SIM (bHLH/PAS)
protein family that is involved in mediating numerous developmental and response
pathways (Crews, 1998; Furness et al., 2007; Kewley et al., 2004). Several isoforms of
ARNT have been identified in mammalian and aquatic species and are termed: ARNT
(HIF-1B), ARNT2, ARNT3 (BMAL1, ARNTL1, MOP3, JAP3), and ARNT4 (BMAL2,
ARNTL2, MOP9). ARNT and ARNT2 possess a 95% amino acid identity within the bHLH and >80% amino acid identity PAS A and B domains that are known to be involved in DNA binding and heterodimerization (Pongratz et al., 1998; Reisz-Porszasz et al., 1994). ARNT appears to be ubiquitously expressed in nearly all cell types in various species (Abbott, 1995; Aitola and Pelto-Huikko, 2003; Hirose et al., 1996;
Holmes and Pollenz, 1997; Kozak et al., 1997; Sojka et al., 2000), while ARNT2 was initially classified a being expressed primarily in the brain and kidney (Hirose et al.,
1996). However, while ARNT is known to have a more ubiquitous expression pattern than ARNT2, mRNA for the two genes is co-expressed to some degree during mouse development and in most adult tissues (Aitola and Pelto-Huikko, 2003). It is interesting, therefore, that gene knock-out of either ARNT or ARNT2 in mice, results in embryonic
117 or perinatal lethality that is characterized by distinct phenotypes (Keith et al., 2001;
Kozak et al., 1997; Maltepe et al., 1997). These findings have led to the hypothesis that the ARNT and ARNT2 proteins have distinct functions in the presence of various dimerization partners and are not fully capable of complementing each other, despite the high level of amino acid identity. But while much is known about the expression of
ARNT at the protein level and its ability to dimerize with various bHLH/PAS partners, the level of ARNT2 protein expression in cells and tissues and well as its interactions with other proteins are less defined (Crews, 1998; Furness et al., 2007; Kewley et al.,
2004; Reisz-Porszasz et al., 1994) (See Chapter One).
Unfortunately, the limited number of studies that have evaluated biochemical and molecular differences in ARNT and ARNT2 function and whether they compete for dimerization partners report conflicting results (Hirose et al., 1996; Sekine et al., 2006).
In addition, there are no studies that have evaluated the co-expression of ARNT and
ARNT2 protein in cell lines or tissues. In an effort to address these questions, studies were initiated to investigate the ability of defined concentrations of ARNT and ARNT2 to interact with the AHR and bind DNA in vivo and in vitro, and to isolate model systems in which both proteins were expressed physiologically.
4.2 ARNT2 Does Not Function to the Same Level as ARNT in AHR-Mediated Signaling
It has been reported that in comparison with ARNT, the ARNT2 isoform exhibits a reduced ability to complement AHR signaling in the induction of XRE driven luciferase reporter activity in response to treatment with 3-Methylcholanthrene (3-MC) (Sekine et al., 2006). To directly compare the ability of ARNT2 to substitute for ARNT in AHR-
118 mediated signaling of an endogenous gene (CYP1A1), several ARNT/ARNT2 constructs
were prepared. Initially, ARNT and ARNT2 full length cDNAs were cloned into either
pcDNA 3.0(-) with or without a V5 epitope tag at the NH-terminal of the full length
ARNT or ARNT2 protein (Figure 4.1). Since antibodies raised against ARNT and
ARNT2 are differently specific, the presence of a common NH-terminal tag allowed each
ARNT to be detected with the same specificity by the same antibody. These constructs
were then transiently transfected into the LA-II ARNT deficient cell line and their ability
to complement the defect in these cells was analyzed. To determine whether or not each
ARNT was able to form a DNA binding species with AHR capable of transactivation of
XRE controlled genes, luciferase reporter studies were completed using these constructs.
In this case, the ARNT or ARNT2 lines were transfected with an XRE controlled luciferase reporter construct along with β-galactosidase, a transfection efficiency control.
Results from a typical experiment are shown in Figure 4.2. In these experiments, LA-II
cells transfected with naked vector were unable to complement AHR-mediated signaling
in response to TCDD as expected. Additionally, when the LA-II cells were transfected
with ARNT alone, high basal levels of luciferase activity were seen that became even
greater in the presence of TCDD. However, when mARNT2 was transfected into the
LA-II cells, both the basal levels as well as the levels of luciferase activity in response to
TCDD were significantly reduced to <17% of those levels seen with ARNT alone. These
data indicate that ARNT2 is not as well able to complement the defect in the ARNT-
deficient line.
119
Figure 4.1: Protein schematic of V5-ARNT and V5-ARNT2 constructs. The 14 amino acid V5 epitope (GKPIPNPLLGLDST) was cloned directly 5' to the ARNT sequence onto each of the ARNT proteins into pcDNA 3.1 (-) downstream of a T7 promoter along with a consensus Kozak sequence. The original ARNT start codon was removed to prevent possible transcription of a clam ARNT lacking the V5 tag.
120
Figure 4.2: Analysis of TCDD-induced luciferase activity in LA-II cells transfected with ARNT or ARNT2. LA-II cells were transfected with identical amounts of ARNT or ARNT2 vector and with pSV- β -Galactosidase along with GudLuc 1.1 and treated with Me2SO (0.5%) or TCDD (5nM) for 6 hours. Luciferase activity and β- Galactosidase activity were measured as previously described in Chapter Six and (Zeruth and Pollenz, 2007). All luciferase values were normalized to β -Galactosidase. LA-II = parental cells transfected with naked vector; ARNT = LA-II cells expressing V5-ARNT; ARNT2 = LA-II cells expressing V5-ARNT2. 121 To assess this data on a more physiological scale, it was pertinent to determine whether or not each ARNT was able to form a DNA binding species with AHR capable of transactivation of the endogenous CYP1A1 protein induction in response to treatment with TCDD as well as to assess the relative levels of ARNT and ARNT2 in each sample type. Therefore, ARNT and ARNT2 constructs possessing a common NH-terminal V5 epitope tag were separately transfected into LA-II cells, treated with TCDD and the resulting cell lysates analyzed by SDS-PAGE and Western blotting for target protein expression. Representative results are given in Figure 4.3. As expected, the LA-II cells were unable to induce CYP1A1 protein in response to TCDD; however, they were able to do so when ARNT was reintroduced by transfection as has been previously shown
(Hoffman et al., 1991; Whitlock and Galeazzi, 1984). In contrast, ARNT2 appeared to be unable to functionally substitute for ARNT during AHR-mediated induction of CYP1A1 protein in response to TCDD even though ARNT and ARNT2 were expressed to similar levels as assessed by computer densitometry quantification of V5 staining. This inefficiency in complementing loss of ARNT was also mimicked when untagged ARNT proteins were used, when ARNT2 was expressed at a greater level than ARNT, and when other AHR ligands were used (EJD, unpublished observations).
To confirm that the inability of ARNT2 to substitute for ARNT was not a result of poor transfection efficiency resulting in subpopulations of LA-II cells expressing high levels of ARNT2 while others expressed little or no ARNT2, yielding an apparent overall similar level of ARNT2 with less apparent CYP1A1 induction, immunohistochemistry was performed on samples similar to those used in Figure 4.3 A. Resultant microscopy images are given in Figure 4.3 B. In these studies, it is apparent that while ARNT and
122
Figure 4.3: Induction of CYP1A1 protein in cells expressing ARNT or ARNT2. A, LA-II cells were transfected with the indicated constructs as detailed in Chapter Six and treated with Me2SO (0.5%) or TCDD (2nM) for 6 hours at 37°C. Equal amounts of total cell lysates from triplicate plates were resolved by SDS-PAGE, blotted and stained with A-1A IgG (1.0μg/ml), anti-V5 IgG (1:500), anti-CYP1A1 (1:200) as well as anti-β-actin (1:1000). Reactivity was visualized by ECL with GAR-HRP (1:10,000) or GAM-HRP (1:10,000). LA-II = parental cells transfected with naked vector; V5-ARNT = LA-II cells expressing V5-ARNT; V5-ARNT2 = LA-II cells expressing V5-ARNT2. B, LA-II cells were propagated on glass coverslips and transfected and treated as detailed in A. Fixed slips were stained with anti-ARNT IgG (0.5μg/ml), anti-ARNT2 IgG (0.5μg/ml), or anti- CYP1A1 (1:100) and reactivity was visualized using GAR-RHO (1:400). All panels that were stained with the same antibodies were photographed for identical times. ARNT) LA-II cells transfected with ARNT and stained for ARNT. ARNT+TC CYP) LA-II cells transfected with ARNT, treated with TCDD or Me2SO (CON) and stained for CYP1A1. Arrowheads indicate cells expressing cytoplasmic CYP1A1. ARNT2) LA-II cells transfected with ARNT2 and stained for ARNT2. ARNT+TC CYP) LA-II cells transfected with ARNT2, treated with TCDD or Me2SO (CON) and stained for CYP1A1.
123 ARNT2 were not transfected with 100% efficiency, a similar percentage of cells were transfected with ARNT2 as was seen with ARNT; however, while the ability of ARNT transfected cells to induce CYP1A1 in response to TCDD was robust, no induction of
CYP1A1 was seen in any cell in the population transfected with ARNT2. LA-II cells transfected with naked vector showed no apparent staining for ARNT, ARNT2, or
CYP1A1 in the presence or absence of TCDD (data not shown). These data further suggest that in cell culture, ARNT2 is not able to substitute for ARNT in the TCDD- dependent induction of CYP1A1.
To further assess this question independently from transiently transfected cell lines, stable cell lines expressing ARNT or ARNT2 were generated using the ARNT deficient
LA-II line as detailed in Chapter Six. The creation of these lines allowed for populations of cells clonally expressing either ARNT or ARNT2, allowing for a comparison of the ability of either protein in function in the AHR mediated induction of CYP1A1 following
TCDD treatment. Thus, each cell line was plated onto triplicate 35mm cell culture plates, dosed with 0.5% Me2SO or 10nM TCDD for 2, 4, or 6 hrs, harvested, and assessed for total protein content. Identical amounts of total cellular protein were then assessed by
SDS-PAGE and Western blotting for target protein expression relative to wild-type
Hepa-1 cells. Results from an experiment are shown in Figure 4.4. In these studies, the
Hepa-1 cells endogenously expressing ARNT show a robust induction of CYP1A1 protein at 2, 4 and 6 hrs of TCDD treatment. Similarly, LA-II stable lines expressing
ARNT were capable of inducing CYP1A1 protein in response to TCDD treatment at all examined time points. While the level of induction in this cell lines was lower than that
124
Figure 4.4: Induction of CYP1A1 protein in ARNT or ARNT2 stable lines. Equal amounts of total cell lysates from the indicated cell lines treated with 0.5% Me2SO or 10nM TCDD for 2, 4, or 6 hrs were resolved by SDS-PAGE, blotted and stained with A- 1A IgG (1.0μg/ml), anti-V5 IgG (1:500), anti-CYP1A1 (1:200) as well as anti-β-actin (1:1000). Reactivity was visualized by ECL with GAR-HRP (1:10,000) or GAM-HRP (1:10,000). ARNT, LA-II line stably expressing ARNT; ARNT2, LA-II line stably expressing ARNT2.
125 seen in the Hepa-1 cells, the level of ARNT was also lower. This is expected since a
reduction in ARNT can lead to a reduction in CYP1A1 induction (RSP, unpublished
observations). In contrast, cell lines stably expressing ARNT2 remained incapable of inducing CYP1A1 at any of the examined time points, confirming that ARNT2 cannot
functionally substitute for ARNT in this situation.
4.3 Inability of ARNT2 to Function in the Regulation of CYP1A1 is Not a Result of
Proline 352
Within the PAS B motif of all characterized ARNT2 proteins, there is a proline
residue in a position where all characterized vertebrate ARNT proteins contain a histidine.
A recent report has suggested that ARNT2 does not function in AHR signaling in vivo
due to the presence of this proline at amino acid 352 within the PAS B domain of the
murine ARNT2 protein (Sekine et al., 2006). This assessment was based on the
observation that mutant ARNT containing the homologous proline showed a reduced
ability to function in the induction of CYP1A1, similar to the results obtained for ARNT2.
However, this hypothesis was not examined directly since an ARNT2 construct
possessing a histidine substitution was not evaluated for a gain-of-function study.
Furthermore, homology modeling places this residue outside of the central β-sheet that is
the characterized functional region for the PAS B domain and also suggests that the
three-dimensional structure of the ARNT and ARNT2 PAS domains are similar (Figure
4.5 A). Therefore, to directly address whether this residue is responsible for loss of
function in ARNT2, mutagenesis was utilized to change the proline at amino acid 352 to
a histidine in the V5-tagged ARNT2 cDNA (ARNT2-H). Studies were designed to
126
Figure 4.5: Induction of CYP1A1 protein in cells expressing ARNT, ARNT2, or ARNT2-H. A, Homology modeling of ARNT and ARNT2 PAS domains showing the position of the swapped histidine and proline residues within the PAS B domain of ARNT and ARNT2 prepared by Pandini et al. (2007). B, LA-II cells were transfected with the indicated expression constructs as detailed in Chapter Six and treated with DMSO (0.5%) or TCDD (2nM) for 6 hours at 37°C. Equal amounts of total cell lysates from triplicate plates were resolved by SDS-PAGE, blotted and stained with anti-V5 IgG (1:500), anti-CYP1A1 (1:200) as well as anti-β-actin (1:1000). Reactivity was visualized by ECL using GAR-HRP (1:10,000) or GAM-HRP (1:10,000). Each lane represents an independent sample. LA-II, parental cells transfected with naked vector.
127 assess the ability of ARNT2-H to restore AHR-mediated induction of the endogenous
CYP1A1 gene in the LA-II cell line as detailed for ARNT2 above. The results in Figure
4.5 B show that ARNT, ARNT2 and ARNT2-H were expressed to similar levels in the
LA-II cells, but as before, only cells expressing ARNT were capable of inducing endogenous CYP1A1. In addition, immunofluorescence microscopy confirmed that all proteins were expressed in the nucleus and only cells expressing ARNT expressed
CYP1A1 protein (data not shown). Thus, these studies show that ARNT2 does not appear to function to the same level of ARNT in the induction of the endogenous
CYP1A1 gene in cell culture and the lack of function is not due to the presence of a proline at amino acid 352.
4.4 In Vitro Synthesized ARNT and ARNT2 Appear to Exhibit Equivalent Ability to
Dimerize with the AHR and Bind DNA
To determine whether the reduced ability of ARNT2 to complement AHR signaling was occurring at the level of DNA binding, electrophoretic mobility shift assays (EMSA) were performed using either in vitro translated ARNT or ARNT2 along with mouse AHR
(Ahb-1). Equal amounts of ARNT and ARNT2 protein as determined by quantified computer densitometry analysis of Western blotting against the common NH-terminal V5 epitope were mixed with AHR and the samples incubated in the presence of TCDD or
Me2SO for 2 h. Antibodies against ARNT, ARNT2, AHR or pre-immune antibodies were then added to aliquots of the original TCDD-activated samples to supershift specific proteins (Figure 4.6). This technique prevents proteins specific to the antibody being used from entering the gel matrix as a result of the increased size of the antibody-bound
128
Figure 4.6: Schematic for evaluation of DNA binding potential of ARNT and ARNT2 using in vitro synthesized and activated samples. In vitro synthesized V5- ARNT, V5-ARNT2, and AHR were produced as detailed in Chapter Six and a portion of each sample was denatured and evaluated for ARNT, ARNT2, or AHR protein by Western analysis using either anti-V5 IgG (1:500) or A-1A anti-AHR IgG (1ug/ml) and visualized by ECL with GAM-HRP IgG (1:10,000; V5) or GAR-HRP IG (1:10,000; AHR). The intensity of the resultant bands from the Western analysis were quantified by computer densitometry as detailed in Chapter Six and equal amounts of either ARNT protein were combined with unprogrammed reticulocyte and AHR protein and incubated in the presence of DMSO (0.5%, vehicle) or TCDD (100 nM, ligand) for 2 h at 30°C. Following activation, aliquots of each sample were combined with EMSA sample buffer and antibodies as described in Chapter Six and in figure legends. These aliquots were then subjected to both EMSA and further Western analysis. 129 protein complex. The formation of AHR•ARNT•XRE and AHR•ARNT2•XRE complexes were then evaluated by EMSA by probing the samples with a 32-P labeled
XRE from the murine CYP1A1 promoter (Figure 4.7). In both cases (AHR and ARNT or AHR and ARNT2), mixtures of both proteins formed a TCDD-dependent DNA binding shift that was not visible in samples activated with the Me2SO vehicle suggesting that the evident shift was a result of activation of the AHR (Figure 4.7, compare lanes 1-2 and 7-8). Furthermore, this study demonstrated that the TCDD-dependent DNA binding shift depended on the presence of both the AHR as well as an ARNT protein (either
ARNT or ARNT2) and the DNA binding shifts corresponded to the relative molecular mass of AHR•ARNT and AHR•ARNT2 complexes. In the TCDD-activated samples containing AHR and ARNT, addition of antibodies against either AHR or ARNT resulted in a supershifting of complexes containing these proteins, essentially ablating DNA binding. In contrast, antibodies against ARNT2 in the AHR/ARNT samples or pre- immune antibodies have no effect on the shift, demonstrating that the shift depends on the presence of both the ARNT and AHR proteins. Similarly, in the TCDD-activated samples containing AHR and ARNT2, addition of antibodies against either AHR or
ARNT2 resulted in a supershifting of complexes containing these proteins, essentially ablating DNA binding. In contrast, antibodies against ARNT in the AHR/ARNT2 samples or pre-immune antibodies have no effect on the shift, demonstrating that this shift depends on the presence of both the ARNT2 and AHR proteins. Thus, these studies show that both ARNT and ARNT2 were capable of forming TCDD-dependent DNA binding shifts that depended on the presence of AHR protein. Furthermore, the intensity of the DNA binding shift was similar when equal amounts of either ARNT protein was
130
Figure 4.7: DNA binding of AHR•ARNT and AHR•ARNT2 heterodimers. In vitro expressed AHR was combined with either ARNT or ARNT2 into a stock sample and then aliquoted and incubated in the presence of DMSO (0.5%) or TCDD (100 nM) for 2 h at 30°C. A, Aliquots from the stock samples were mixed with 32P-labeled XRE oligonucleotides and protein•DNA complexes resolved as detailed in Chapter Six. In some samples, 50ng of IgG specific to ARNT (A1), ARNT2 (A2), AHR (Ah) or preimmune IgG (IG) were included in the binding reaction prior to loading on the gel. The location of the specific AHR•ARNT•XRE and AHR•ARNT2•XRE complex and free XRE are indicated. B, Aliquots of the activation reactions utilized in the EMSA were denatured and evaluated for ARNT, ARNT2 or AHR protein by Western analysis using either anti-V5 IgG (1:500) or anti-AHR IgG (1ug/ml) and visualized by ECL with GAM- HRP IgG (1:10,000; V5) or GAR-HRP IG (1:10,000; AHR). Note that the concentration of both V5-ARNT and V5-ARNT2 in each sample is similar. C, control samples; T, samples activated with TCDD.
131 co-expressed with the AHR, suggesting that ARNT and ARNT2 have an equal ability to
associate with the AHR and bind DNA when expressed in vitro (Figure 4.7, B).
To examine whether this would hold true when both ARNT and ARNT2 were co-
expressed along with the AHR in equal ratios, another EMSA was performed using a
mixture of ARNT and ARNT2 proteins along with in vitro synthesized AHR (Figure 4.8).
In this experiment, a mixture of both ARNT proteins with the AHR again resulted in the formation of a DNA binding shift that was TCDD-dependent (Figure 4.9 A).
Furthermore, this shift resulted from equal expression of both ARNT proteins and appeared to require both proteins as well as the AHR. Interestingly, the addition of antibodies against ARNT in the mixed sample, resulted in a DNA binding shift that was approximately 51% of the intensity of the duplicate sample with TCDD alone when quantified by computer densitometry, suggesting that ARNT containing complexes
account for approximately 49% of the total DNA binding shift (Figure 4.9 A, lane 3, C).
Similarly, the addition of antibodies against ARNT2 yielded a shift that was 50% of the
intensity of the control, suggesting that ARNT2 containing complexes also account for
approximately 50% of the total shift (Figure 4.9 A, lane 4, C). In contrast, addition of
antibodies against the AHR ablated all evidence of DNA binding, further suggesting that
the DNA binding shift seen in each lane is indeed representative of ARNT or
ARNT2•AHR heterodimers, while addition of pre-immune IgG had no effect on DNA
binding. Furthermore, these differences in DNA binding were not attributable to the
amount of protein in each sample as revealed by quantitative Western blotting performed
on a portion of the exact samples used for the EMSA since each EMSA sample contained
approximately equal amounts of both ARNT and ARNT2 as determined by V5 staining
132
Figure 4.8: Schematic for evaluation of DNA binding potential of a mixture of ARNT and ARNT2 using in vitro synthesized and activated samples. In vitro synthesized V5-ARNT, V5-ARNT2, and AHR were produced as detailed in Chapter Six and a portion of each sample was denatured and evaluated for ARNT, ARNT2, or AHR protein by Western analysis using either anti-V5 IgG (1:500) or A-1A anti-AHR IgG (1ug/ml) and visualized by ECL with GAM-HRP IgG (1:10,000; V5) or GAR-HRP IG (1:10,000; AHR). The intensity of the resultant bands from the Western analysis were quantified by computer densitometry as detailed in Chapter Six and equal amounts of both ARNT and ARNT2 proteins were combined with unprogrammed reticulocyte and AHR protein and incubated in the presence of DMSO (0.5%, vehicle) or TCDD (100 nM, ligand) for 2 h at 30°C. Following activation, aliquots of each sample were combined with EMSA sample buffer and antibodies as described in Chapter Six and in figure legends. These aliquots were then subjected to both EMSA, further Western analysis, and immunoprecipitation analyses.
133
Figure 4.9: DNA binding of AHR•ARNT and AHR•ARNT2 heterodimers in the presence of both ARNT and ARNT2. In vitro expressed AHR was combined with equal amounts of both ARNT and ARNT2 into a stock sample and then aliquoted and incubated in the presence of DMSO (0.5%) or TCDD (100 nM) for 2 h at 30°C. A, Aliquots from the stock samples were mixed with 32P-labeled XRE oligonucleotides and protein•DNA complexes resolved as detailed in Chapter Six. In some samples, 50ng of IgG specific to ARNT (A1), ARNT2 (A2), AHR (Ah) or preimmune IgG (IG) were included in the binding reaction prior to loading on the gel. B, Aliquots of the activation reactions utilized in C were denatured and evaluated for ARNT expressing using anti-V5 IgG (1:500) and visualized by ECL with GAM-HRP IgG (1:10,000; V5). Note that the level of ARNT and ARNT2 was similar. C) The intensity of the shifted bands from several different EMSA experiments were quantified by computer densitometry as detailed in Chapter Six. Results are plotted as the mean +/- SE of each shifted band with the samples containing preimmune IgG set to 100%. IG, samples incubated with preimmune IgG; A1, samples incubated with anti-ARNT IgG; A2, samples incubated with anti-ARNT2. Numbers on the bottom indicate the relative intensity compared to samples containing preimmune IgG. *, Statistically different from samples incubated with IgG; p<0 134 (Figure 4.9 B). Additionally, each sample also contained approximately equal amounts
of AHR (data not shown). Taken together, this suggests that both ARNT and ARNT2 are
approximately equally able to heterodimerize with the AHR and bind XREs when
synthesized in vitro and expressed at identical levels.
While ARNT and ARNT2 appeared to exhibit an equal ability to contribute to a
DNA binding shift that required AHR, it had not yet been demonstrated directly that this
shift was occurring through dimerization of the ARNT or ARNT2 proteins with the AHR.
Instead, the EMSA analyses identified that the DNA binding shift depended upon ligand-
activated AHR as well as the presence of an ARNT protein. Therefore, it was pertinent to
assess the direct interaction of these proteins through alternate methods. To establish
that activated AHR complexes could be generated that were associated with either ARNT
or ARNT2, in vitro expressed AHR and equal amounts of ARNT and ARNT2 were co-
incubated and activated as detailed previously, and immunoprecipitation studies carried
out using anti-AHR IgG. A representative experiment is shown in Figure 4.10. The input
samples demonstrate that the in vitro expressed proteins exhibited equal amounts of AHR
protein and also expressed equal amounts of both ARNT and ARNT2 as assessed by
quantitative Western blotting against the V5 epitope. As expected, when the AHR was
immunoprecipitated from these samples in the absence of TCDD, the AHR precipitated efficiently, but neither ARNT protein was co-precipitated with the AHR (Figure 4.10 lane
3). In the TCDD activated samples, the AHR was precipitated to the same level as before; however, in this sample both ARNT and ARNT2 were brought down using anti-AHR
IgG (Figure 4.10 compare lanes 3 and 5). Furthermore, the level of ARNT that was co- precipitated with the AHR was equal to that of the ARNT2 that was precipitated,
135
Figure 4.10: Association of ARNT and ARNT2 with AHR. In vitro expressed AHR, ARNT, and ARNT2 were combined into a stock sample and then equally split and incubated in the presence of DMSO (0.5%) or TCDD (100 nM) for 2 h at 30°C. Equal amounts of each sample were then incubated with 1 μg anti-AHR IgG (Ah) or pre- immune IgG (Pi) for 1 hr at 4oC. The samples were then precipitated with protein A/G agarose beads, washed with TTBS and the boiled in the presence of 1x gel sample buffer. Equal amounts of sample were resolved by SDS-PAGE, blotted and stained with either anti-V5 IgG (1:500) or anti-AHR IgG (1ug/ml) and visualized by ECL with GAM-HRP IgG (1:10,000; V5) or GAR-HRP IG (1:10,000; AH). The precipitated IgG band (lanes 3-6) is shown to demonstrate the uniformity of the precipitation across all samples.
136 demonstrating that in the presence of both ARNT and ARNT2, the AHR was equally able
to dimerize with either protein in a ligand-dependent manner. This held true in non-
mixed ARNT samples as well, when ARNT or ARNT2 protein co-precipitated with the
AHR individually (data not shown). Thus, these studies demonstrate that both ARNT isoforms can form TCDD-dependent AHR heterodimers when expressed in vitro.
Additionally, these studies support the ability of these dimers to bind DNA, since
portions of the samples used for immunoprecipitation analyses were also used in EMSA
experiments that mimicked the data previously shown (data not shown).
4.5 ARNT2 Can Out-Compete ARNT for Association with the Liganded AHR
Since in vitro synthesized ARNT2 appeared to be equally able to heterodimerize
with the AHR and bind DNA, the effect of increasing amounts of ARNT2 on the ability
of ARNT to heterodimerize with the AHR was also evaluated. In vitro synthesized
ARNT was mixed with AHR and ARNT2 was added in a 1:1, 1:3, or 1:11 ratio versus
ARNT, in each case keeping the total level of ARNT proteins constant. The resulting
samples were then incubated in the presence of TCDD or Me2SO for 2 h. Again, antibodies against ARNT, ARNT2, AHR or pre-immune antibodies were then added to duplicate portions of the 1:11 protein mix in the presence of TCDD to prevent target proteins from entering the gel matrix. The formation of AHR•ARNT•XRE and
AHR•ARNT2•XRE complexes were then evaluated by EMSA. Representative results shown in Figure 4.11 indicate that as previously shown, ARNT•AHR heterodimers formed a TCDD-dependent DNA binding shift and that addition of ARNT2 to this mixture increased the intensity of the DNA binding shift, reaching saturation by the 1:11
137
Figure 4.11: Effect of ARNT2 concentration on the formation of AHR•ARNT complexes. In vitro expressed AHR was combined with the indicated ratios of ARNT and ARNT2, the samples equally split and incubated in the presence of DMSO (0.5%) or TCDD (100 nM) for 2 h at 30°C. A) Aliquots of the activated reactions were denatured and evaluated for ARNT, ARNT2 or AHR protein by Western analysis using either anti- V5 IgG (1:500) or A-1A (1ug/ml) and visualized by ECL with GAM-HRP IgG (1:10,000; V5) or GAR-HRP IG (1:10,000; AHR). B) The exact samples visualized in A were mixed with 32P-labeled XRE oligonucleotides and protein•DNA complexes resolved as detailed in Chapter Six. In some samples, 50ng of IgG specific to ARNT (A1), ARNT2 (A2), or preimmune IgG (IG) were included in the binding reaction prior to loading on the gel (lanes 11-14). The location of the specific AHR•ARNT•XRE and AHR•ARNT2•XRE complex and free XRE are indicated. Numbers indicate the relative level of ARNT or ARNT2 that were used in the activation reaction. 138 sample. Western blotting of the EMSA samples confirmed the relative ratios of both
ARNT proteins as determined by computer densitometry analysis of V5 staining. In the
1:3 and 1:11 ratio sample containing increased levels of ARNT2, the majority of
heterodimers appeared to be ARNT2•AHR dimers, though ARNT remained in this
sample, as evidenced by the addition of antibodies against each protein. While addition
of ARNT antibodies to the 1:11 ratio sample had no apparent effect on the overall
intensity of the DNA binding shift, addition of antibodies against ARNT2 or AHR
instead ablated all DNA binding. These results suggest that ARNT2 is indeed able to
associate with the AHR and furthermore can out-compete ARNT for AHR dimerization.
4.6 The Ability of ARNT2 to Associate with the AHR and Bind DNA is Not Dependent on
AHR Concentration or ARNT Protein Concentration
Since studies examining the ability of either ARNT protein to associate with the
AHR have shown conflicting results (Dougherty and Pollenz 2007; Hirose et al., 1996;
Sekine et al., 2006), it was important to assess possible reasons for the discrepancies
between these studies and in those contained in this report. If ARNT and ARNT2 truly
differed in their ability to associate with the AHR and bind DNA in vitro in contrast to the results presented in Figures 4.7-4.11, what conditions being used in our studies might lead to the apparent equal ability of these proteins to function in DNA binding?
One such condition that could alter the apparent ability of these proteins to function during the EMSA assay is if the level of AHR or ARNT being used in the studies. Since these studies used in vitro synthesized proteins, it is possible that high levels of expression of the ARNT proteins or the AHR led to an association of these proteins that
139 would not seen if the proteins were expressed at more physiological levels. This idea was
the basis for the explanation of the discrepancies between the ability of ARNT2 to
associate with the AHR that was seen by Hirose et al. (1996), versus the apparent
inability of these proteins to associate in the studies performed by Sekine et al. (2006).
However, while this claim was made by Sekine et al. (2006), it remained untested. Thus,
it was important to assess whether the concentration of either the AHR or the ARNT
proteins played a role in the ability/inability of these proteins to associate in vitro.
To begin to evaluate this question, limiting amounts of in vitro expressed AHR
were combined with equal ratios of either ARNT or ARNT2 into stock samples and then
aliquoted and incubated in the presence of DMSO (0.5%) or TCDD (100 nM) for 2 h at
30°C and tested for target protein expression by Western blotting (Figure 4.12 A). In
these studies, the level of AHR was altered while the level of ARNT and ARNT2 was
held constant. Aliquots from the stock samples were then mixed with 32P-labeled XRE
oligonucleotides and protein•DNA complexes resolved as detailed in Chapter Six.
Representative results are given in Figure 4.12 B. As the level of AHR increased in the
EMSA samples, so did the intensity of the ligand-dependent DNA binding shift, suggesting that the level of AHR being used in these studies was a limiting factor for
ARNT or ARNT2. Since the level of DNA binding increased in the presence of increased AHR, and since the intensity of the DNA binding shift was similar between
ARNT and ARNT2 when co-expressed with similar levels of AHR, overexpression of the
AHR could be ruled out as a factor influencing the association of these proteins.
140
Figure 4.12: DNA binding of AHR•ARNT and AHR•ARNT2 heterodimers in the presence of limiting AHR. Limiting amounts of in vitro expressed AHR were combined with either ARNT and ARNT2 in equal ratios into stock samples and then aliquoted and incubated in the presence of DMSO (0.5%) or TCDD (100 nM) for 2 h at 30°C. A, Aliquots of the activation reactions utilized in B were denatured and evaluated for ARNT and AHR expression using anti-V5 IgG (1:500) or A-1A anti-AHR IgG (1μg/ml) and visualized by ECL with GAR-HRP IgG (1:10,000; AHR) or GAM-HRP IgG (1:10,000; V5). Note that the level of ARNT and ARNT2 was similar, while the level of AHR was altered. B, Aliquots from the stock samples were mixed with 32P-labeled XRE oligonucleotides and protein•DNA complexes resolved as detailed in Chapter Six.
141 Since the ability of ARNT2 to associate with the AHR and bind DNA could also be
related to possible overexpression of ARNT2 relative to the AHR yielding artifactual
results, it was important to assess whether dilution of each of the ARNT proteins relative
to each other would alter the results obtained in the assay. Therefore, another EMSA was
performed to directly compare the ability of each ARNT to associate with the AHR and
bind DNA at reduced levels. To analyze this, increasing amounts of either in vitro
synthesized ARNT or ARNT2 were incubated with identical amounts of AHR in the
presence of TCDD or Me2SO for 2 h and the formation of AHR•ARNT•XRE and
AHR•ARNT2•XRE complexes evaluated by EMSA as previously described.
Comparable dilutions were then examined for DNA binding relative to the amount of
ARNT or ARNT2. In each dilution set, both ARNTs were able to heterodimerize with the AHR and bind DNA in a TCDD-dependent fashion (Figure 4.13 A). Furthermore, quantification of the DNA binding shift by densitometry revealed that the intensity of the
DNA binding shift was similar between comparable ARNT2 dilution and ARNT dilution sets exhibiting 78% binding in the lowest ARNT2 dilution, 106% binding in the middle set, and 114% binding at the highest concentration of ARNT2 in comparison with the
ARNT dilutions (Figure 4.13 B). Likewise, at each dilution, the relative amounts of
ARNT and ARNT2 were similar in range as determined by quantified Western blotting of the EMSA samples stained against the V5 epitope, though there was 1.9, 1.3, and 1.3 times more ARNT2 than ARNT respectively (Figure 4.13 B). No significant differences in the levels of AHR between samples were noted (Figure 4.13 A). Since the lowest dilution set showed 90% more ARNT2 than ARNT, yet exhibited only 78% binding relative to the ARNT sample, there may have been a slight difference in affinity of
142
Figure 4.13: Effect of target protein concentration on the formation of AHR•ARNT and AHR•ARNT2 heterodimers. In vitro expressed AHR was mixed with increasing concentrations of ARNT or ARNT2, the samples equally split and incubated in the presence of DMSO (0.5%) or TCDD (100 nM) for 2 h at 30°C. A) A portion of the activated samples were denatured and evaluated for ARNT and ARNT2 or AHR expression by Western analysis using anti-V5 IgG (1:500) or anti-AHR IgG (1ug/ml), respectively. Reactive bands were visualized using GAM-HRP IgG (1:10,000) and ECL. The remaining samples were mixed with 32P-labeled XRE oligonucleotides and protein•DNA complexes resolved as detailed in Chapter Six. The location of the specific AHR•ARNT•XRE and AHR•ARNT2•XRE complex and free XRE are indicated. B) The relative level of ARNT and ARNT2 protein (left) and DNA binding (right) was determined using computer densitometry of the bands shown in the Western blot of A. The relative intensity of the shifted bands from the EMSA in A was determined using computer densitometry. 143 ARNT2 for the AHR in comparison with ARNT at low levels of ARNT proteins.
However, throughout the evaluated points, the expression of the ARNT proteins and the relative level of DNA binding was essentially equal, and under no conditions was
ARNT2 unable to associate with the AHR and bind XREs. Therefore, in contrast to the
statement posed by Sekine et al. (2006), high expression levels of these proteins do not
appear to play a role in their association.
4.7 The Ability of ARNT2 to Associate with the AHR and Bind DNA May be XRE-Specific
The murine CYP1A1 promoter has been demonstrated to possess five functional
putative XREs ranging from -489 to -1218 nucleotides upstream of the transcriptional
start site and have been designated as XRE A, B, D, E, and F (Lusska et al., 1993). In
those studies, each of these XREs was shown to exhibit varying levels of CAT activity
when used to drive reporter constructs while XRE C appeared to be non-responsive
(Figure 4.14 A). Since the previous studies described herein (Figures 4.7-4.13) had used
only XRE D as a probe for DNA binding, it was possible that the ability of each ARNT
to bind DNA was related to the XRE being used and would not be mimicked with other
CYP1A1 XREs. To evaluate this possibility, each XRE as described by Lusska et al.
(1993) was labeled with 32P, quantified by spectrophotometry and scintillation counting, normalized to the number of counts/minute and used as a probe against aliquots of single
samples containing in vitro activated AHR•ARNT or AHR•ARNT2 dimers as previously
described. Typical results are shown in Figure 4.14 B. Each XRE showed an
approximately equal ability to associate with AHR•ARNT or AHR•ARNT2 dimers with
the exception of XRE A, which appeared to show a greater affinity for AHR•ARNT than
144
Figure 4.14: Analysis of XRE sequence on the DNA binding ability of AHR•ARNT or AHR•ARNT2 heterodimers. A, Schematic representing the mouse CYP1A1 promoter XRE locations and relative activity levels as described by Lusska et al. (1993). Position numbers are nucleotide positions relative to the transcriptional start site. B, AHR and ARNT or ARNT2 were expressed in vitro and mixed together, incubated in the presence of Me2SO (-; 0.5%) or TCDD (170 nM) for 2 h at 30°C and a portion of each sample analyzed by EMSA, as detailed in Chapter Six. The location of the specific AHR•ARNT•XRE and AHR•ARNT2•XRE complex are indicated. C, Equal volumes of 2x gel sample buffer were added to the remaining portion of each EMSA sample and the samples used for Western blotting. Duplicate Western blots were stained with anti-V5 IgG (1:500) and visualized by ECL with GAM-HRP IgG (1:10,000).
145 AHR•ARNT2 dimers. It is important to note, however, that in the studies by Lusska et al.
(1993), XRE A appeared to contribute the least amount of functional activity towards the induction of CAT activity driven by this XRE alone in comparison with other functional
CYP1A1 XREs. Additionally, in repeated experiments, this discrepancy appeared to be less extreme than represented by this individual EMSA. In the event that AHR•ARNT/2 dimers had a preference for the non-responsive XRE C, the ability of this XRE to bind either ARNT or AHR•ARNT2 dimers was also examined and no discrepancy found between binding (data not shown). Again, these differences were not attributable to variations in the levels of ARNT vs. ARNT2 and no significant differences in the levels of AHR between samples were noted (Figure 4.14 C, data not shown). Thus, while the ability of AHR•ARNT/2 heterodimers to bind DNA may be XRE specific, it is unlikely that increased affinity or a lack thereof of either dimer for individual XREs of the
CYP1A1 promoter plays a role in the inefficiency of ARNT2 in substituting for ARNT in
AHR-mediated regulation of CYP1A1 since XREs B, C, D, E, and F that contribute
>90% of the functional regulation of CYP1A1 all show an essentially equivalent ability to be bound by either heterodimer.
4.8 DNA Binding Ability of AHR•ARNT2 Heterodimers Appears to be Ligand Dependent
Since previous studies had examined the ability of ARNT2 to associate with the
AHR using 3-methylcholanthrene (3-MC) rather than TCDD and seen disparities between ARNT and ARNT2•AHR dimerization, and since little is known in regard to the structure of the liganded AHR, it was pertinent to assess whether these disparities were related to the ligand being used (Hirose et al., 1996; Sekine et al., 2006). To examine this
146 possibility, identical pools of AHR and ARNT or AHR and ARNT2 were activated using either TCDD, 3-MC, or benzo[a]pyrene (BAP) and the formation of AHR•ARNT•XRE and AHR•ARNT2•XRE complexes evaluated by EMSA as previously described. In the first set of experiments, ARNT or ARNT2 were co-expressed individually with the AHR, activated in the presence of Me2SO, TCDD, or 3-MC, and analyzed by EMSA (Figure
4.15). As previously seen, activation of the EMSA samples with TCDD yielded a DNA binding complex in both the AHR•ARNT and AHR•ARNT2 samples that exhibited a similar intensity (Figure 4.15 B). Interestingly, activation of the EMSA samples with 3-
MC resulted in a DNA binding shift that was similar in intensity to the AHR•ARNT
TCDD activated sample, but was significantly (p<0.05) reduced in the AHR•ARNT2 sample over the course of three experiments, though the levels of ARNT and AHR proteins remained constant (Figure 4.15 B, data not shown).
In a mixture experiment similar to Figure 4.9, samples containing equal levels of both ARNT and ARNT2 activated with either TCDD, 3-MC, or BAP exhibited similar discrepancies (Figure 4.16). As previously seen, in a sample containing equal levels of both ARNT and ARNT2, activation with TCDD yielded a DNA binding complex that consisted of approximately 46% ARNT•AHR dimers and 44-54% ARNT2•AHR dimers
(Figure 4.16 A, B). In contrast, though ARNT and ARNT2 were equally expressed, activation by either 3-MC or BAP yielded a DNA binding complex that exhibited 24-
36% binding when antibodies against ARNT were included following activation and 61-
64% binding when antibodies against ARNT2 were included, suggesting that in the case of 3-MC, 64% of the DNA binding shift was attributable to ARNT•AHR dimers and, in the case of BAP 76% were likewise attributed to ARNT•AHR dimers (Figure 4.16 B, D).
147
Figure 4.15: Effect of 3-MC on the DNA binding of AHR•ARNT and AHR•ARNT2 complexes. A, In vitro expressed AHR was combined with either ARNT or ARNT2 into a stock sample and then aliquoted and incubated in the presence of DMSO (0.5%), TCDD (100 nM) or 3-MC (54µM) for 2 h at 30°C. Aliquots from the stock samples were mixed with 32P-labeled XRE oligonucleotides and protein•DNA complexes resolved as detailed in Chapter Six. The location of the specific AHR•ARNT•XRE and AHR•ARNT2•XRE complexes are indicated. Free XRE was run off the bottom of the gel so that the difference in migration of the complexes could be observed. B, Aliquots of the activation reactions utilized in A were denatured and evaluated for ARNT and ARNT2 protein by Western analysis using anti-V5 IgG (1:500) and visualized by ECL with GAM-HRP IgG (1:10,000). Note that the concentration of both V5-ARNT and V5- ARNT2 in each sample is similar. T, samples activated with TCDD; M, samples activated with 3-MC. The intensity of the shifted bands from several different EMSA experiments were quantified by computer densitometry as detailed in Chapter Six. Results are plotted as the mean +/- SE of the 3-MC or TCDD activated samples. *, statistically different from the ARNT sample activated with 3-MC; p<0.05.
148
149
Figure 4.16. Effect of different ligands on the DNA binding of AHR•ARNT and AHR•ARNT2 complexes. In vitro expressed AHR, ARNT, and ARNT2 were combined into a stock sample and then equally split and incubated in the presence of the specific ligands for 2 h at 30°C as detailed below. A, Samples were incubated with DMSO (0.5%), TCDD (100nM) or 3-MC (54µM) and then mixed with 32P-labeled XRE oligonucleotides and protein•DNA complexes resolved as detailed in Chapter Six. In some samples, 50ng of IgG specific to ARNT (A1), ARNT2 (A2), AHR (AH) or preimmune IgG were included in the binding reaction prior to loading on the gel. The location of the specific AHR•ARNT•XRE and AHR•ARNT2•XRE complex and free XRE are indicated. B, Aliquots of the stock activation mixture utilized in A were denatured and evaluated for ARNT and ARNT2 expression by Western analysis. The Western blot was stained with anti-V5 IgG (1:500) and visualized by ECL with GAM- HRP IgG (1:10,000). The intensity of the shifted bands from the EMSA in A was quantified by computer densitometry as detailed in Chapter Six. Results are plotted with the samples containing preimmune IgG set to 100%. IG = samples incubated with preimmune IgG; A1 = samples incubated with anti-ARNT IgG; A2 = samples incubated with anti-ARNT2. Numbers on the bottom indicate the relative intensity compared to samples containing preimmune IgG. C, Samples were incubated with DMSO (0.5%), TCDD (100nM) or BAP (17µM) and EMSA performed as detailed in A. D, Aliquots of the stock activation mixture utilized in C were denatured and evaluated for ARNT and ARNT2 expression by Western analysis as indicated in B. The intensity of the shifted bands from the EMSA in C was quantified by computer densitometry as detailed in Chapter Six. Results are plotted with the samples containing preimmune IgG set to 100%. IG, samples incubated with preimmune IgG; A1, samples incubated with anti-ARNT IgG; A2, samples incubated with anti-ARNT2. Numbers on the bottom indicate the relative intensity compared to samples containing preimmune IgG.
150 Since treatment with saturating levels of 3-MC or BAP resulted in a greater proportion of
ARNT•AHR dimers than ARNT2•AHR dimers, ARNT may have a greater affinity for 3-
MC or BAP liganded AHR than ARNT2 suggesting that the activating AHR ligand may confer the specificity of the AHR for its heterodimeric partner (likely through receptor conformational differences) or may confer the specificity of DNA binding by the
AHR•ARNT/2 dimer using this model system.
To evaluate whether the AHR ligand was altering the specificity of the AHR for either the ARNT proteins or altering the specificity of the liganded AHR•ARNT2 dimer, it was necessary to establish that AHR complexes activated with TCDD or BAP could be generated that were associated with either ARNT or ARNT2. Therefore, in vitro
expressed AHR and equal amounts of ARNT and ARNT2 were co-incubated and
activated as detailed previously, and immunoprecipitation studies carried out using anti-
AHR IgG, anti-ARNT IgG, or anti-ARNT2 IgG. A representative experiment is shown
in Figure 4.17. The input sample (lane 9) demonstrates equal amounts of both ARNT
and ARNT2 were co-expressed with the AHR. As expected, when the AHR was
immunoprecipitated from these samples in the presence of TCDD, the AHR precipitated efficiently, and both ARNT proteins were co-precipitated to equivalent levels (Figure
4.17 lane 5). Additionally, IgG against ARNT or ARNT2 specifically precipitated each
ARNT protein and also co-precipitated the AHR. Interestingly, in the BAP activated samples, the AHR was precipitated to the same level as before; however, in this sample both ARNT and ARNT2 were also brought down using anti-AHR IgG to the same levels as before (Figure 4.17 compare lanes 1 and 5). Furthermore, when portions of the same mixture used for immunoprecipitation were subjected to EMSA analysis, DNA binding
151
Figure 4.17: Association of ARNT and ARNT2 with BAP or TCDD activated AHR. In vitro expressed AHR, ARNT, and ARNT2 were combined into a stock sample and then equally split and incubated in the presence of DMSO (0.5%, -), BAP (17 μM, B) or TCDD (100 nM, T) for 2 h at 30°C. A, Equal amounts of each sample were then incubated with 1 μg anti-AHR IgG (A), 1 μg anti-ARNT IgG (1), 1 μg anti-ARNT2 IgG (2), or pre-immune IgG (Pi) for 1 hr at 4oC or saved for analysis of input (Inp). The samples were then precipitated with protein A/G agarose beads, washed with TTBS and the boiled in the presence of 1x gel sample buffer. Equal amounts of sample were resolved by SDS-PAGE, blotted and stained with either anti-V5 IgG (for ARNT and ARNT2, 1:500) or anti-AHR IgG (1ug/ml) and visualized by ECL with GAM-HRP IgG (1:10,000; V5) or GAR-HRP IG (1:10,000; AH). The precipitated IgG band is shown to demonstrate the uniformity of the precipitation across all samples. B, Portions of the immunoprecipitation reactions used in A were mixed with 32P-labeled XRE oligonucleotides and protein•DNA complexes resolved as detailed in Chapter Six.
152 shifts were seen as before. Taken together, this suggests that the ligand activating the
AHR alters the ability of the AHR•ARNT2 dimer to bind DNA, but not the ability of these complexes to form in this model system.
4.9 The Ability of ARNT or ARNT2 to Dimerize with the AHR in Vitro May be Receptor
Species Dependent
The Ahb-1 receptor found in the C57, C58, and the MA/My strains is known to exhibit several distinct characteristics from other Ah murine alleles as well as Ah receptors from other species (see Chapter Two). While the Ahb-1 receptor exists as a
~95kDa (805 amino acid) cytosolic receptor, which does not appear to be shuttling through the nucleus endogenously, the Ahb-2 receptor found in the murine BALB/cBy, A,
and C3H strains as well as the rat and human AHR exist as a ~104kDa (848 amino acid)
receptor that is primarily nuclear and exhibits dynamic nucleocytoplasmic shuttling
(Poland et al., 1994; Pollenz and Dougherty, 2005). Furthermore, the Ahb-1 is truncated
when compared with Ah receptors found in other mammalian species, containing a point
mutation that prematurely truncates the receptor at 805 amino acids, while the Ahb-2, rat, and human AHR all contain an additional 42-45 amino acids at their carboxy-terminus that have 70% identity. As this additional carboxy-terminal sequence is present in the other AHR alleles as well as across species and is reasonable conserved in sequence, it is reasonable to assume that this sequence may be functionally relevant. Therefore, it was relevant to assess the ability of ARNT or ARNT2 to heterodimerize with other Ah receptor species. This was investigated by incubating equal amounts of in vitro synthesized ARNT or ARNT2 with equal amounts of Ahb-1, Ahb-2 or rat AHR, activating
153 the receptors with TCDD or BAP and examining the DNA binding ability of each
AHR•ARNT or AHR•ARNT2 dimer by EMSA (Figure 4.18). As expected, both ARNT
and ARNT2 were able to form a TCDD- or BAP-dependant DNA binding shift with each
Ah receptor species (Figure 4.18 A). Additionally, the formation of AHR•ARNT and
AHR•ARNT2 heterodimers was essentially equal in the presence of the Ahb-1 receptor
activated with TCDD (Figure 4.18 A, compare lanes 2 and 5), while ARNT2 showed a
reduced affinity for BAP liganded AHR as seen previously (Figure 4.18 A, compare
lanes 3 and 6). In contrast, however, ARNT2 appeared to exhibit a greater affinity for rat
AHR in the presence of TCDD or BAP than ARNT as computer densitometry analysis of
the intensity of the DNA binding shift revealed a band that was 33-75% greater in
intensity in the Ahb-2•ARNT2 and Rat AHR•ARNT2 samples than in identical samples
containing ARNT (Figure 4.18 A, B). Importantly, this was not due to different levels of
ARNT2 in these samples, since the levels of ARNT, ARNT2, and AHR remained
constant across the sample types (Figure 4.18 C). Additionally, while ARNT previously
showed a greater affinity for BAP-liganded AHR than ARNT2, this did not appear to be
maintained when other Ah receptor species were used (Figures 4.16, 4.18). Instead,
while ARNT and ARNT2 exhibited similar magnitudes of binding when activated by
TCDD, but not when activated with BAP, in the presence of Ahb-1 receptor as expected,
ARNT exhibited a lesser magnitude of binding in comparison with ARNT2 in the
presence of Ahb-2 receptor or rat AHR irrespective of ligand. Similarly, both Ahb-2 receptor and rat AHR exhibited a lesser degree of binding when activated with BAP in
154
155
Figure 4.18: Analysis of AHR species on the formation of TCDD or BAP dependent AHR•ARNT or AHR•ARNT2 heterodimers. A, Ahb-1, Ahb-2, rat AHR, ARNT and ARNT2 were mixed and incubated in the presence of Me2SO (-; 0.5%), TCDD (100 nM), or BAP (17 μM) for 2 h at 30°C and a portion of each sample analyzed by EMSA, as detailed in Chapter Six. The location of the specific AHR•ARNT•XRE and AHR•ARNT2•XRE complex and free XRE are indicated. Non-specific binding is also noted by the asterisk. B, Equal volumes of 2x gel sample buffer were added to the remaining portion of each EMSA sample and the samples used for Western blotting. Duplicate Western blots were stained with either anti-V5 IgG (1:500) or with A-1A AHR IgG (1.0μg/ml) and visualized by ECL with GAM-HRP IgG (1:10,000) or GAR-HRP (1:10,000). C, The DNA binding shift from each EMSA TCDD or BAP treated lane resulting from combination of Ahb-1, Ahb-2, or rat AHR with ARNT or ARNT2 was quantified by densitometry and normalized against background. Results are plotted as the relative densitometry units for each sample.
156 the presence of either ARNT or ARNT2 in comparison to activation with TCDD, suggesting that BAP is a weaker ligand for these Ah receptor species, but not for the
Ahb-1 receptor. Together, these results suggest that ARNT2 can dimerize with several Ah receptor species and the Ahb-2 and rat AHR, which more closely resemble the human
AHR, may have a greater affinity for ARNT2 than ARNT. Thus, the ability of ARNT2 to function in other genetic backgrounds from the Hepa-1 expressing Ahb-1 should be addressed.
4.10 The Ability of ARNT or ARNT2 to Dimerize with the AHR in Vitro Does Not Appear to Differ When Expressed in Vivo
Since ARNT, ARNT2, or the AHR may have post-translational modifications that could potentially influence their ability to heterodimerize, it was important to assess whether the previously described relationships between ARNT and ARNT2•AHR dimer formation would be mimicked in vivo. Therefore, V5-ARNT and V5-ARNT2 were transfected into LA-II cells, and following a 24 hr recovery period, total cell lysates were generated, activated by TCDD or BAP, analyzed by EMSA and the resulting binding ability assessed by EMSA. Representative results are given in Figure 4.19. As seen in the in vitro synthesized protein experiments, ARNT2 remained able to associate with the
AHR and bind XREs when expressed in vivo, suggesting that no post-translational modifications are occurring in LA-II cytosol that would alter the affinity of the ARNT2 for the AHR (Figure 4.19 A). Additionally, since ARNT2 remained able to dimerize with the AHR when LA-II cytosol was incubated with ARNT2, this also suggests that the
157
Figure 4.19: DNA binding of ARNT and ARNT2 expressed in cell culture. Expression vectors for ARNT and ARNT2 were transfected in LA-II cells and cytosol produced as detailed in Chapter Six. A, Stock mixtures containing equal levels of ARNT and ARNT2 protein were incubated with DMSO (0.5%), TCDD (100nM) or BAP (17µM) for 2 h at 30°C. Samples were aliquoted and mixed with 32P-labeled XRE oligonucleotides and protein•DNA complexes resolved as detailed in Chapter Six. In some samples, 50ng of IgG specific to ARNT (A1), ARNT2 (A2), AHR (AH) or preimmune IgG were included in the binding reaction prior to loading on the gel. The location of the specific AHR•ARNT•XRE and AHR•ARNT2•XRE complex and free XRE are indicated. B, An aliquot of the stock activation mixture utilized in A were denatured and evaluated for ARNT and ARNT2 expression by Western analysis. The Western blot was stained with anti-V5 IgG (1:500) and visualized by ECL with GAM- HRP IgG (1:10,000). The intensity of the shifted bands from the EMSA in A was quantified by computer densitometry as detailed in Chapter Six. Results are plotted with the samples containing preimmune IgG set to 100%. IG, samples incubated with preimmune IgG; A1, samples incubated with anti-ARNT IgG; A2, samples incubated with anti-ARNT2. Numbers on the bottom indicate the relative intensity compared to samples containing preimmune IgG.
158 Ahb-1 does not exhibit any post-translational modifications that would alter its ability to recognize ARNT2.
As expected, ARNT also retained an apparent higher affinity for BAP liganded Ahb-
1 receptor in comparison with ARNT2 (Figure 4.19 A, B). When antibodies were
included in the EMSA reaction using ARNT or ARNT2 expressing cytosol to ablate
either ARNT or ARNT2 containing complexes, as previously detailed, ARNT and
ARNT2 again showed an apparently equal ability to bind DNA in the presence of TCDD,
but not BAP.
Furthermore, when similar studies were performed to directly compare the binding
ability of samples generated from in vitro synthesis of ARNT or ARNT2 against those
generated from transfected LA-II cytosol, the intensity of DNA binding was equal when
similar levels of ARNT proteins were compared either in vitro or in vivo (data not shown).
Importantly, these results collectively indicate that the previous results based on in vitro
synthesized ARNT, ARNT2, and AHR (Figures 4.7-4.18) do not differ from similar
studies based on in vivo expressed proteins and therefore, there are no significant post-
translational modifications that appear to alter the ability of either ARNT protein or AHR
to heterodimerize or bind XREs in this model system.
4.11 Inhibition of ARNT by ARNT2 in AHR-Mediated Signaling
Since ARNT2 appeared to be able to bind DNA in response to TCDD, yet exhibited
a decreased ability to complement loss of ARNT in LA-II cells, it became of interest to
determine if its co-expression with ARNT would impede the ability of ARNT to
complement AHR-mediated signaling, possibly through squelching of the AHR. LA-II
159 cells were therefore transfected with either: naked vector, ARNT, or ARNT with
increasing amounts of ARNT2, along with pSV-β-Galactosidase and GudLuc 1.1
plasmids and treated with Me2SO (0.5%) or TCDD (5nM) for 6 hours. Luciferase
studies were then performed as previously described. LA-II cells transfected with naked
vector were unable to complement AHR-mediated signaling in response to TCDD as
expected (Figure 4.20). Additionally, when the LA-II cells were transfected with ARNT
alone, high basal levels of luciferase activity were seen that became even greater in the presence of TCDD. However, when ARNT2 was co-expressed with ARNT in either a
2:1 or 1:1 ratio of plasmid DNA, the levels of luciferase activity in response to TCDD
were significantly reduced in comparison with ARNT alone. Importantly, the overall
level of ARNT expression remained constant as the level of ARNT2 was changed. Thus,
this suggests that ARNT2 when co-expressed with ARNT may exhibit some degree of
dominant negative activity.
To confirm whether or not this result would be mimicked in vivo, WT Hepa-1 cells
endogenously expressing ARNT were transfected with ARNT2 and correspondingly evaluated by SDS-PAGE and Western blotting. After 6 hrs of TCDD treatment, the WT cells transfected with ARNT2 exhibited an approximately 30% reduction in the level of
CYP1A1 protein in comparison to the untransfected WT cells as determined by quantitative Western blotting (Figure 4.21 A). Similarly, in a second experiment where other time points were evaluated, the WT cells transfected with mARNT2 exhibited an approximately 20% reduction in the level of CYP1A1 protein in comparison to the untransfected WT cells as determined by quantitative Western blotting at 2 and 4 hrs of
TCDD treatment (data not shown). At both 4 hours and 6 hours, this reduction was
160
Figure 4.20: Impact of ARNT2 expression on AHR-mediated signaling. LA-II cells were transfected with equal total amounts of naked vector, ARNT alone, or ARNT with increasing amounts of ARNT2 as indicated as well as with equal amounts of both pSV-β- Galactosidase and GudLuc 1.1 plasmids as described in Chapter Six. Following 24 h recovery, cells were then treated with Me2SO (0.05%) or TCDD (5nM) for 6 hours. Equal amounts of total cell lysates were resolved by SDS-PAGE, blotted and stained with mouse ARNT IgG (A1; 1 μg/ μl) or mouse ARNT2 IgG (1:500). Reactivity was visualized by ECL with GAR-HRP (1:10,000). All luciferase values were normalized to β -Galactosidase.
161
Figure 4.21: Impact of ARNT2 expression on AHR-mediated signaling in cell culture. A, Hepa-1 cells were transfected with naked vector or ARNT2 expression vectors and treated with DMSO (0.5%) or TCDD (2nM) for 6 h at 37°C. Equal amounts of total cell lysates were resolved by SDS-PAGE, blotted and stained with either anti AHR IgG (1µg/ml), anti-ARNT2 IgG (1µg/ml), anti-CYP1A1 IgG (1:200), or anti-β- actin IgG (1:1000). Reactivity was visualized by ECL with GAR-HRP (1:10,000). Each lane represents and independent sample. CYP1A1 protein expression was quantified by computer densitometry and normalized to β -actin controls. Results represent the mean +/- SE of three independent samples. *, Statistically different from TCDD treated cells that were not transfected with ARNT2; p<0.05. B, Hepa-1 cells were propagated on glass coverslips and transfected and treated as detailed in A. Fixed slips were stained with anti-ARNT IgG (0.5μg/ml; A1), anti-ARNT2 IgG (0.5μg/ml; A2), or anti-CYP1A1 (1:100) and reactivity was visualized using GAR-RHO (1:400). All panels that were stained with the same antibodies were photographed for identical times. WT, cells transfected with naked vector. WT + A2, cells transfected with ARNT2 vector.
162 statistically significant (P<0.05) and similar reductions were not seen when WT cells were transfected with excess ARNT (data not shown). Therefore, again, the co- expression of ARNT2 with ARNT appeared to be hindering the ability of ARNT to function in AHR-mediated signaling, albeit through an unknown mechanism.
4.12 ARNT2 Does Not Function to the Same Level as ARNT in Endogenous AHR-
Mediated Signaling
The previous studies suggest that ARNT2 has the potential to affect AHR-mediated signaling in a negative manner when expressed in cell culture. However, it is not known whether the conditions utilized for the various studies represent a truly physiological condition. For example, the ability of ARNT2 to impact AHR-mediated signaling and affect the function of AHR•ARNT2 complexes in vivo will be dependent on the co- expression of ARNT and ARNT2 in the same cells. Also, since transient transfections typically lead to >60% overexpression of the target protein, the inhibition of ARNT functionality in AHR-mediated signaling by ARNT2 may be artifactual based on non- physiologic expression of ARNT2. This stresses the importance of examining protein function using endogenous protein levels. Thus, to examine the possible role of ARNT2 in AHR-mediated signaling, it is necessary to also examine cells endogenously expressing ARNT2. However, it has generally been hypothesized that ARNT and
ARNT2 are not expressed in the same cells at the protein level due to the limited tissue distribution of ARNT2 (Hirose et al., 1996). Importantly, a more recent study, has described mRNA for both ARNT and ARNT2 as being co-localized in many murine peripheral organs and neuronally derived tissue, suggesting that distribution of ARNT2
163 may not be as restricted as previously described (Aitola and Pelto-Huikko, 2003). In addition, studies of ARNT2 expression have generally examined mRNA for ARNT2 with no analysis of endogenous protein expression.
In order to investigate possible interactions between ARNT and ARNT2 in a physiological setting, studies were carried out to determine the expression of ARNT and
ARNT2 protein in various murine tissues and cell culture lines. In the first set of studies,
whole tissue lysates from the brain, eye, heart, kidney, liver, lung, muscle, skin, spleen,
and thymus were prepared as described in Chapter Six and analyzed for AHR, ARNT,
and ARNT2 protein by Western blotting. Figure 4.22 shows that while ARNT protein
was detected in all the tissues evaluated as has previously been described, expression of
ARNT2 protein was detected in the brain, eye, and kidney, and was also detected at lower
expression levels in the heart, spleen and thymus (Figure 4.22 A). From these studies, it
is evident that ARNT2 expression is not limited only to the brain and kidneys, though
these tissues do exhibit the highest relative levels. Since these results do not confirm that
the ARNT and ARNT2 protein are localized to the same cells, several continuous cell
lines were purchased from ATCC and the level of ARNT and ARNT2 protein evaluated
by Western blotting. Figure 4.22 B shows that ARNT and ARNT2 protein were co
expressed in human pigmented retinal epithelial cells (ARPE-19), rat central nervous
system cells (B35), and human, mouse and rat kidney cells (A498, TCMK, NRK,
respectively). To gain insight into the ratio of ARNT and ARNT2 protein in these lines,
ARNT and ARNT2 TNT reactions containing equal amounts of ARNT and ARNT2 protein (Figure 4.22 C), were included in the experiment so that the staining with the specific ARNT and ARNT2 antibodies could be normalized against staining of the in
164
Figure 4.22: ARNT and ARNT2 protein expression in tissues and cells. A, Total cell lysates were generated from a variety of adult C57BL/6Jmouse tissues, resolved by SDS- PAGE, blotted and stained with anti-ARNT R1 IgG (1.0μg/ml) or anti-ARNT2 IgG (1:500). Antibodies were titered to react to the TNT samples with the same sensitivity. Reactivity was visualized by ECL with GAR-HRP (1:10,000). Tissues expressing both ARNT and ARNT2 proteins are noted by the asterisk. B, The indicated cell lines were purchased from ATCC, and cultured as detailed in Chapter Six. Total cell lysates evaluated for expression of both ARNT and ARNT2 using RI and ARNT2 IgG that were titered to give the same level of reactivity to the TNT samples. ARPE-19 normal rat kidney; TCMK-1, mouse kidney; A498; human kidney adenocarcinoma; Hepa-1, mouse hepatoma; TNT, in vitro synthesized ARNT or ARNT2. C, The same level of TNT samples loaded in B was stained with anti-V5 IgG (1:500). Reactivity was visualized by ECL with GAM-HRP (1:10,000). D, ARNT and ARNT2 levels were determined by computer densitometry from the blots presented in B. Results are presented as relative densitometry units.
165 vitro synthesized proteins with anti-V5 IgG. The quantified results are presented in
Figure 4.22 D. The ratio of ARNT:ARNT2 protein was variable among the different
cells with the NRK-49F rat kidney cells showing a near equal ratio (1.3:1), and the A498
cells showing the lowest (7.2:1). Thus, these studies confirm that ARNT and ARNT2 can be co-expressed in the same cell and provide novel models for the future analysis of
the physiological interactions of these proteins in AHR as well as other bHLH/PAS
signaling pathways.
4.13 Analysis of ARNT2 Function When Endogenously Expressed
Since the hRPE lines appeared to co-express both ARNT and ARNT2
endogenously as determined by SDS-PAGE and Western blotting, it was pertinent to
assess the function of ARNT2 in these lines. Importantly, these lines also express AHR
protein and CYP1A1 protein was induced in these cells in response to treatment with
TCDD (Figure 4.23, lanes 3, 4). Using short-interfering RNA (siRNA) against ARNT or
ARNT2, levels of either ARNT or ARNT2 were decreased by >80% or by >50%
respectively. Interestingly, reduction of ARNT led to complete ablation of CYP1A1
activity in response to TCDD, even though endogenous levels of ARNT2 continued to be
expressed (Figure 4.23, lanes 5-8). While this agreed with previous data that suggested
that ARNT2 was not capable of substituting for ARNT in the induction of CYP1A1 in
cell culture (Figures 4.2-4.4), it is in contrast to unpublished studies from this laboratory
that suggest that low levels of ARNT are sufficient for CYP1A1 induction (RSP).
Reduction of ARNT2, however, showed no alteration of the ability to induce
CYP1A1 in response to TCDD (Figure 4.23, lanes 9-12). This result was intriguing in
166
Figure 4.23: Reduction of endogenous ARNT or ARNT2 by siRNA knockdown in hRPE cells. A, hRPE cells were transfected with siRNA specific to ARNT (siA) or ARNT2 (siA2) or control siRNA (siCON) as detailed in Chapter Six. Forty-eight hours later, cells were dosed with Me2SO (0.5%) or TCDD (5nM) for 6 h at 37°C, and equal amounts of total cell lysates were resolved by SDS-PAGE, blotted and stained with either anti-AHR IgG (1µg/ml), anti-ARNT IgG (1µg/ml ), anti-ARNT2 IgG (1:500), anti- CYP1A1 IgG (1:200), or anti-β-actin IgG (1:1000). Reactivity was visualized by ECL with GAR-HRP (1:10,000). Each lane represents and independent sample. B, computer densitometry was used to determine the relative level of ARNT or ARNT2 protein present in the samples presented on the blot in A. Each column represents the relative densitometry units of an individual band and shows the level of reduction of each protein following siRNA treatment.
167 light of the results presented in Figures 4.20-4.21 that suggested that ARNT2 could
inhibit induction of CYP1A1 in Hepa-1 cells. Interestingly, reduction of ARNT2 by
siRNA did not appear to increase induction of CYP1A1 protein induction following
treatment with TCDD suggesting that co-expression of ARNT2 with ARNT in the hRPE lines does not appear to have an inhibitory effect on the induction of CYP1A1 in
response to TCDD. However, previous studies in this laboratory have demonstrated that
ARNT when reduced to 10-15% of its endogenous concentration in Hepa-1 cells maintains its ability to function in AHR-mediated signaling with visible induction of
CYP1A1 protein via Western blotting as a result of its endogenous expression level being
10-fold greater than that of the AHR protein (Holmes and Pollenz, 1997). In the case of
this experiment, reduction of ARNT to 10% its normal concentration led to an inability to
complement AHR-mediated signaling. Therefore, it is possible that ARNT2 has an
inhibitory effect on ARNT only when expressed to a greater level than ARNT as seen in
both the transient transfections in WT cells and following siRNA against ARNT. Thus,
the ratio of ARNT to ARNT2 may be important to AHR-signaling in cells co-expressing
both ARNT proteins. Co-siRNA treatment against both ARNT and ARNT2 may assist in
demonstrating this possibility, whereby if the above hypothesis were correct, it would be
expected that in hRPE cells reduced in both ARNT and ARNT2, the 10% of ARNT
remaining would then remain able to complement AHR-mediated signaling as a result of
loss of the repressive ARNT2 expression. This is also particularly intriguing given that
NRK cells, which express higher levels of ARNT2 than the hRPE cells, appear to be
unable to induce CYP1A1 in the presence of TCDD, although the ARNT and AHR
168 proteins are expressed. As such, reduction of ARNT2 in these lines may restore CYP1A1
induction if ARNT2 is truly inhibiting this signaling pathway.
4.14 ARNT2 from Nuclear Extracts Fails to Bind XREs
To confirm previous findings that ARNT and ARNT2 are equally capable of
heterodimerizing and binding DNA when expressed in cell culture, nuclear extracts were
prepared as detailed in Chapter Six from hRPE and NRK lines treated with Me2SO or
TCDD for 1 hr. The resultant nuclear extracts were then evaluated for target protein
expression by Western blotting and DNA binding assessed by EMSA analysis.
Representative results are shown in Figure 4.24. In striking contrast to the results
presented in Figures 4.7-4.19, these EMSA assays revealed no evidence of XRE binding
requiring ARNT2. Instead, while the typical TCDD-dependent DNA binding shift was
evident in both the NRK and hRPE lines, inclusion of IgG against either ARNT or AHR
ablated all evidence of DNA binding, while inclusion of ARNT2 IgG had no effect. This suggests that ARNT2 was not contributing to the level of DNA binding seen following
TCDD treatment in these lines, though ARNT2 could be detected in the nuclear extracts
of both lines (Figure 4.24 C, data not shown for RPE).
Since these lines exhibit a reduced level of ARNT2 in comparison to ARNT, it is
possible that the reduced level of ARNT2 was leading to an inability to visualize
AHR•ARNT2 DNA binding. To assess whether the lack of apparent ARNT2 binding
was a sensitivity issue or was specific to nuclear extracts, cytosolic extracts were also
prepared from NRK cells and in vitro activated with Me2SO or TCDD as previously
described since cytosolic extracts of Hepa-1 cells exogenously expressing ARNT2 were
169
Figure 4.24: DNA binding of NRK and hRPE nuclear extracts. Nuclear extracts from NRK and hRPE cells treated with Me2SO (0.5%, C) or TCDD (100nM, TC) for 1 hr at 37°C were produced as detailed in Chapter Six. A, B, 30µg/well of nuclear extracts from NRK (A) or RPE (B) cells were aliquoted and mixed with 32P-labeled XRE oligonucleotides and protein•DNA complexes resolved as detailed in Chapter Six. In some samples, 50ng of IgG specific to ARNT (A1), ARNT2 (A2), AHR (AH) or preimmune IgG were included in the binding reaction prior to loading on the gel. The location of the specific AHR•ARNT•XRE and AHR•ARNT2•XRE complex and free XRE are indicated. C, An aliquot of the nuclear extract mixtures utilized in A were denatured and evaluated for AHR, ARNT, and ARNT2 expression by Western analysis. The Western blot was stained with anti-AHR IgG (1µg/µl), anti-ARNT IgG (1µg/µl), or anti-ARNT2 IgG (1:250) and visualized by ECL with GAR-HRP IgG (1:10,000).
170 capable of exhibiting ARNT2-dependent DNA binding in the EMSA assay. Therefore,
NRK cytosolic extracts were evaluated for DNA binding potential by EMSA assay
analysis and representative results are shown in Figure 4.25. Interestingly, in contrast to
the results obtained for in vitro activated cytosol of Hepa-1 cells co-expressing ARNT
and ARNT2, the NRK lines did not exhibit any evidence of the formation of DNA-bound
complexes containing ARNT2 (Figure 4.25 A, lanes 3 and 4). However, when additional
ARNT2 was added to the DNA-binding reaction, the cytosolic extracts exhibited very
strong TCDD-dependent DNA binding shifts that could be supershifted with IgG against
ARNT2, but not with pre-immune IgG (Figure 4.25 B). Therefore, these studies
demonstrate that ARNT2 remains capable of forming XRE-bound complexes with the
AHR in the NRK genetic background, suggesting that protein(s) in the NRK lines is not
inhibiting the ability of ARNT2 to function in these assays. Thus, the lack of visible
ARNT2 association with DNA in Figure 4.24 and 4.25 A may be a sensitivity issue rather
than a biochemical one.
To further evaluate this, nuclear extracts were also prepared from wild-type Hepa-1
cells and wild-type Hepa-1 cells transfected with ARNT2 (WT+ARNT2), expressing
high levels of ARNT2 in comparison to the low levels seen in NRK or hRPE lines.
Western blot analysis of these nuclear extracts revealed a TCDD-dependent nuclear
accumulation of both the AHR and ARNT, and revealed that ARNT2 appeared to be
constitutively associated with nuclear structures (Figure 4.26 A), similar to results seen in
the NRK line (Figure 4.24 C). EMSA analysis of the nuclear extracts also revealed
strikingly similar results to those obtained in the NRK and hRPE lines, wherein the
typical TCDD-dependent DNA binding shift was evident in both the WT and
171
Figure 4.25: DNA binding of NRK cytosolic extracts. Cytosolic extracts from NRK cells treated with Me2SO (0.5%, C) or TCDD (100nM, TC) for 1 hr at 30°C were produced as detailed in Chapter Six. A, 60µg/well of cytosolic extracts from NRK cells were mixed with 32P-labeled XRE oligonucleotides and protein•DNA complexes resolved as detailed in Chapter Six. In some samples, 50ng of IgG specific to ARNT (A1), ARNT2 (A2), AHR (AH) or preimmune IgG were included in the binding reaction prior to loading on the gel. The location of the specific AHR•ARNT•XRE and free XRE are indicated. A, 15µg/well of cytosolic extracts from NRK cells were mixed with 32P- labeled XRE oligonucleotides and protein•DNA complexes resolved as detailed in Chapter Six. In some samples, in vitro synthesized ARNT2 (A2) was included in the binding reaction prior to loading on the gel. The location of the specific AHR•ARNT•XRE, AHR•ARNT2•XRE, and free XRE are indicated. In some samples, 50ng of IgG specific to ARNT (A1), ARNT2 (A2), AHR (AH) or preimmune IgG were included in the binding reaction prior to loading on the gel. The location of the specific AHR•ARNT•XRE is indicated.
172 WT+ARNT2 lines, but inclusion of ARNT2 IgG had no effect, while inclusion of IgG against either ARNT or AHR ablated all evidence of DNA binding. These data again suggested that ARNT2 was not contributing to the level of DNA binding seen following
TCDD treatment in the WT+ARNT2 lines (Figure 4.26 B). This was particularly intriguing since in vitro activated cytosolic extracts of Hepa-1 cells expressing both
ARNT and ARNT2 showed disparate results wherein XRE-bound complexes could be supershifted with ARNT2 IgG (Figure 4.19). However, since ARNT2 appeared to be constitutively associated with nuclear structures even in the absence of TCDD and did not show a large increase in nuclear extract protein levels following TCDD treatment, unlike the ARNT or AHR protein, it is possible that ARNT2 is inaccessible to the AHR during the transformation of the AHR complex into an AHR•ARNT/2 heterodimer. If this were true, it may be pertinent to assess whether the addition of ARNT2 to nuclear extracts from NRK cells would generate AHR•ARNT2 dimers as seen with cytosolic extracts in Figure 4.25. If excess ARNT2 led to the formation of AHR•ARNT2 dimers, this would further suggest that AHR remains capable of dimerizing with ARNT2 and binding DNA in cell culture and that ARNT2 may have an endogenous function that impairs its ability to freely associate with the ligand activated AHR in the nucleus. Thus, further studies should be performed to assess the ability of ARNT2 isolated from nuclear extracts to function during AHR signaling and to further identify possible reasons for the difference between in vitro activated cytosolic ARNT2 to dimerize with the AHR versus
ARNT2 from nuclear extracts.
173
Figure 4.26: DNA binding of Hepa-1 nuclear extracts. Nuclear extracts were generated from wild-type Hepa-1 cells transfected with naked vector (WT) and wild-type Hepa-1 cells transfected with ARNT2 (WT+A2) dosed with Me2SO (0.5%, -) or TCDD (100nM, +) for 1 hr at 37°C were produced as detailed in Chapter Six. A, An aliquot of the nuclear extract mixtures utilized in B were denatured and evaluated for AHR, ARNT, and ARNT2 expression by Western analysis. The Western blot was stained with anti- AHR IgG (1µg/µl), anti-ARNT IgG (1µg/µl), or anti-ARNT2 IgG (1:250) and visualized by ECL with GAR-HRP IgG (1:10,000). B, 30µg/well of nuclear extracts from WT or WT+A2 cells were mixed with 32P-labeled XRE oligonucleotides and protein•DNA complexes resolved as detailed in Chapter Six. In some samples, 50ng of IgG specific to ARNT (A1), ARNT2 (A2), AHR (AH) or preimmune IgG were included in the binding reaction prior to loading on the gel. The location of the specific AHR•ARNT•XRE is indicated.
174 Conversely, another possibility for the apparent lack of DNA binding in nuclear extracts from lines expressing ARNT and ARNT2 is that AHR•ARNT2 dimers are
forming in these extracts, but are unable to bind DNA. If this were the case, then
immunoprecipitation analysis of nuclear extracts should reveal the presence of such
dimers. To test this hypothesis, nuclear extracts were prepared from NRK cells and wild-
type Hepa-1 cells transfected with ARNT2 (WT+A2), and subjected to
immunoprecipitation analysis using anti-AHR IgG. Additionally, these samples were
compared to cytosolic extracts generated from both lines, since cytosolic extracts from
WT+A2 did exhibit AHR•ARNT2 dimerization and therefore, should act as a positive
control for the nuclear extracts. Representative results are presented in Figure 4.27.
Immunoprecipitation of the NRK extracts revealed that precipitation of the AHR
specifically co-precipitated ARNT in nuclear extracts of TCDD treated cells (Figure 4.27
A). In contrast, ARNT2 did not appear to co-precipitate with the AHR in any of the
NRK samples. However, since the proportion of AHR or ARNT2 precipitated in the
nuclear extracts was less than 15% of the amount seen in the input, it was again possible that sensitivity of the ARNT2 antibodies was an issue in this assay since 15% of the already low levels of ARNT2 expressed in the NRK line would be undetectable. Future
studies employing a more sensitive ECL such as SuperSignal® West Femto Maximum
Sensitivity Substrate ECL kit (Pierce, Rockford, IL) or precipitation of larger amounts of
nuclear extract could help to resolve this issue.
In contrast to the studies shown in Figure 4.27 A, immunoprecipitation of the
WT+A2 extracts revealed that precipitation of the AHR specifically co-precipitated
ARNT in nuclear extracts of TCDD treated cells, but co-precipitated ARNT2 in all of the
175
Figure 4.27: Immunoprecipitation analysis of NRK and WT Hepa-1 cells expressing ARNT2. Cytosolic extracts (CY) from NRK cells (A) or wild-type cells transfected with ARNT2 (B) were treated with Me2SO (0.5%, C) or TCDD (100nM, TC) for 1 hr at 30°C and were produced as detailed in Chapter Six. Nuclear extracts (NE) from these lines were generated from lines treated with Me2SO (0.5%, C) or TCDD (10nM, T) for 1 hr at 37°C and were produced as detailed in Chapter Six. 800ug of cytosolic extracts or 300ug of nuclear extracts were then incubated with 1 μg monoclonal anti-AHR IgG (IP) or pre- immune IgG (Pi) for 1 hr at 4oC or saved for analysis of input (Input). The samples were then precipitated with protein A/G agarose beads, washed with TTBS and the boiled in the presence of 1x gel sample buffer. Equal amounts of sample were resolved by SDS- PAGE, blotted and stained with A-1A anti-AHR IgG (1ug/ml), anti-ARNT IgG (1ug/ml), or anti-ARNT2 (1:500) and visualized by ECL with GAR-HRP IG (1:10,000). The precipitated IgG band is shown to demonstrate the uniformity of the precipitation across all samples.
176 samples evaluated with the greatest levels of AHR associated ARNT2 present in the
nuclear extracts from TCDD treated cells; however, these nuclear extracts could not bind
DNA in the EMSA assay (Figure 4.27 B, Figure 4.26). Numerous studies were
performed to examine the specificity of the AHR antibody that provide evidence that the
monoclonal AHR antibody used for these analyses would not non-specifically precipitate
the ARNT2 protein (data not shown). However, due to the very high levels of ARNT2
expression resulting from the transient transfection of the wild-type cells, it was also
possible that the interaction itself was non-specific, particularly since ARNT2 co-
precipitated with the AHR even in Me2SO treated cells. Therefore, this hypothesis is yet
unclear and future studies should continue to assess whether AHR•ARNT2 dimers are
forming that are unable to bind DNA.
4.15 Function of ARNT Versus ARNT2 in Hypoxic Signaling
To confirm that ARNT2 is functional in other signaling pathways that are believed
to involve ARNT2 and where its involvement has been supported by the previously
described knockout studies as detailed in Chapter One, the ability of ARNT2 to complement loss of ARNT in hypoxia-mediated signaling was evaluated. To evaluate the potential role of ARNT2 during hypoxia, various ARNT constructs were transfected into LA-II cells along with an HRE controlled luciferase reporter and were then subjected to physiological hypoxia for 14 hours. Representative results are given in Figure 4.28 A.
These results demonstrate a functional role for ARNT2 in hypoxia-mediated signaling and functional activity for our ARNT2 construct. In LA-II cells transfected with either murine ARNT2 or zebrafish ARNT2, high levels of HRE controlled luciferase activity
177 were seen following 16 hours of hypoxia that were similar to those seen in LA-II cells transfected with murine ARNT. To correlate these data to results obtained in cell culture,
NRK cells endogenously expressing ARNT and ARNT2 were cultured under normal conditions or in the presence of TCDD as controls or cultured in the presence of Cobalt chloride (CoCl2) or desferrioxamine (DFO), known chemical inducers of hypoxia (Cain,
1975; Myers et al., 1985; Nielsen et al., 1987; Wang and Semenza, 1993). As previously
shown, following 1 hr treatment with TCDD, ARNT but not ARNT2, accumulated in
nuclear extracts prepared from NRK cells (Figure 4.28 B). In contrast, both ARNT and
ARNT2 protein increased in nuclear extracts in response to the hypoxic inducers CoCl2 or DFO, suggesting a functional DNA binding response for both ARNT proteins during the hypoxic response. Further analysis to examine specific gene regulation requiring the
ARNT proteins as well as their ability to interact with HIF and bind HREs should be performed in the future, but, importantly, these initial results indicate that the ARNT2 used in the previously described studies is capable of functioning in the hypoxic response and therefore a lack of response during AHR signaling is not an artifact of a non- functional ARNT2 construct.
4.16 Evaluation of Potential ARNT and ARNT2 Homodimers
ARNT has also been implicated in serving as a transcriptional regulator in a homodimeric form, based on early studies in which it was demonstrated that during size exclusion HPLC of Sf9 whole cell lysates, ARNT was eluted as a single peak with a molecular mass of 205 kDa (Sogawa et al., 1995). These studies also demonstrated that
ARNT could form homodimers via the HLH/PAS domains, had affinity for the core
178
Figure 4.28: Analysis of the role of ARNT2 in hypoxia. LA-II cells were transfected with identical amounts of ARNT or ARNT2 vector and with pSV- β -Galactosidase along with an HRE-controlled luciferase reporter and cultured in 5% O2 for 14 hrs. Hypoxia responsive luciferase activity and β-Galactosidase activity were measured as previously described in Chapter Six and (Zeruth and Pollenz, 2007). All luciferase values were normalized to β -Galactosidase. LA-II = parental cells transfected with naked vector; ARNT = LA-II cells expressing ARNT; ARNT2 = LA-II cells expressing ARNT2, zfARNT2 = LA-II cells expressing ARNT2 from the zebrafish, Danio rerio. B, Cytosolic (CYTO) and nuclear extracts (NE) from wild-type Hepa-1 cells transfected with ARNT2 were cultured untreated (C), or in the presence of TCDD (2 nM, T), Cobalt Chloride (100 μM, Co), or Desferrioxamine (100 μM, D) for 1 hr at 37°C were produced as detailed in Chapter Six. Arrows below indicate accumulation of ARNT and ARNT2, but not AHR in response to hypoxia induced by Cobalt Chloride or Desferrioxamine.
179 sequence of the adenovirus major late promoter (MLP) containing the canonical E-box
(5’CACGTG) sequence, and could drive expression of E-box controlled reporters when
produced in vitro similar to the activities of other E-box binding proteins MyoD, Max,
and USF. Further studies showed similar results, again suggesting that ARNT
homodimerize, could associate with the E-box sequence and could competitively displace the c-Myc/Max heterodimer from binding to the E-box, though the ability of ARNT to homodimerize may be dependent upon ARNT phosphorylation status (Antonsson et al.,
1995; Huffman et al., 2001; Levine and Perdew, 2001; Levine and Perdew, 2002;
Swanson et al., 1995; Swanson and Yang, 1999). However, no studies have attempted to
examine ARNT homodimerization using full-length proteins expressed in cell culture.
Additionally, ARNT2 homodimerization has not been evaluated except by means of
yeast-two hybrid interactions using truncated ARNT2 proteins. Furthermore, the idea of
ARNT•ARNT2 interactions have not yet been proposed since it was believed that these
proteins are not co-expressed. Therefore, the ability of ARNT or ARNT2 to serve a
homodimeric role was analyzed as was the potential for ARNT•ARNT2 interactions.
To begin to analyze whether or not these interactions were occurring, constructs for
each ARNT were generated that possessed a HisMax (HM) or V5 epitope tag as detailed
previously and in Chapter Six. In each case, in vitro synthesized HM-ARNT was mixed
with V5-ARNT, HM-ARNT with V5-ARNT2, V5-ARNT and HM-ARNT2, or untagged
ARNT and ARNT2. Alternatively, sets of ARNT and ARNT2 were transfected into the
ARNT deficient LA-II cells. In either case, immunoprecipitation analyses were
performed using antibodies against HM, V5, ARNT, or ARNT2 to assess the potential
interactions of each protein. For example, LA-II cells were transfected with either
180 HisMax (HM) tagged ARNT (or HM-ARNT2) and V5 tagged ARNT (or V5-ARNT2) or
both ARNT and ARNT2. For the HM-ARNT•V5-ARNT and HM-ARNT2•V5-ARNT2
homodimerization studies, either ARNT or ARNT2 was immunoprecipitated using HM
or V5 antibodies and Western blot analysis of each sample stained with the opposing
antibody. For the ARNT•ARNT2 dimerization study, either ARNT or ARNT2 was
immunoprecipitated with ARNT or ARNT2 antibodies and Western blots stained for the
presumed partner. Interestingly, ARNT and ARNT2 showed little ability to
homodimerize or heterodimerize with each other (data not shown). However, it is
recognized that even in the event that such interactions were occurring, immunoprecipitation analyses would be able to resolve only a portion of these. In a
mixture of HM-ARNT and V5-ARNT, where ARNT was capable of homodimerizing, it is possible that HM-ARNT•HM-ARNT, V5-ARNT•V5-ARNT, or HM-ARNT•V5-
ARNT interactions would occur as would a portion of non-dimerized ARNT proteins in
all likelihood. Thus, the pool of HM-ARNT•V5-ARNT would likely only represent at
most ~1/3 of total ARNT complexes and therefore, detection of these dimerized species
may be limited by the sensitivity of the assay.
In a further attempt to analyze the potential formation of ARNT or ARNT2
homodimers, EMSA analyses using an E-box probe from the adenovirus major late
promoter (MLP) were performed using cytosolic or nuclear extracts from cells
endogenously expressing both ARNT proteins (RPE, NRK), wild-type Hepa-1 cells
expressing ARNT, or Hepa-1 cells transfected with ARNT2 as detailed in Chapter Six.
In these analyses, a specific shift was observed that was not seen in the presence of a
labeled mutant E-box probe, but this shift could not be supershifted with antibodies
181 against ARNT or ARNT2, suggesting the involvement of other E-box binding proteins
(data not shown). A similar lack of ARNT or ARNT2 E-box binging was also obtained using in vitro synthesized ARNT proteins or unprogrammed reticulocytes (data not shown). Again, however, it remains possible that such dimers exist yet do not associate with the adenovirus E-box as flanking nucleotides may also play a role in transcription factor DNA binding. Thus, while these studies failed to identify in vivo ARNT or
ARNT2 dimerization or the presence of ARNT•ARNT2 dimers, they emphasize the need to identify in vivo ARNT or ARNT2 homodimers as well as the inherent difficulty of performing such studies. While ARNT homodimers have been implicated in the partial regulation of CYP2A5 (Arpiainen et al., 2007), these studies focused on the regulation of
CYP2A5 in the presence or absence of ARNT, which may be impacting more than
ARNT alone.
182
Chapter Five
Implications and Future Directions
5.1 Implications of AHRb-2 Studies
The AHRb-1 allele carried by the C57BL/6J murine strain contains a point
mutation that prematurely truncates the receptor at 805 amino acids, while the AHR
found in other murine strains as well as in the rat and human all contain an additional 42-
45 amino acids at their carboxy-terminus that have 70% identity (Figure 1.3, Table 1.3).
Thus, as detailed throughout Chapters One and Two, the current model system used for the evaluation of the physiological role of the AHR, which focuses on using the C57BL mouse and hepatoma cells isolated from this murine strain to assess AHR function, may not be the most accurate model. Furthermore, the results of this report detail a possible functional role of this extended carboxy-terminal sequence, either through an altered
AHR conformation resulting from the presence of this extended sequence or through a
direct function of this region in the association of the AHR with XAP2. Collectively, the
results presented in Chapter Two also describe the impact of this extended region on
AHR subcellular localization and degradation, which differ between Ah receptor species.
Furthermore, the use of a genetically identical background to examine the function and
degradation rate of the AHRb-2 in comparison with the AHRb-1 was ideal and provided
significant insights into the biological properties of the AHR contributed by this
183 functional region, which, importantly, were mimicked in alternate cell lines endogenously expressing the AHRb-2.
Similarly, studies of XAP2 function in AHR-mediated signal transduction have
focused on the interactions of XAP2 with the Ahb-1 receptor and have also been based primarily on the analysis of XAP2 in transient transfection systems using AHR-deficient
COS-7 cells. Collectively, these studies suggested that XAP2 may influence the stability,
subcellular localization, and overall expression of the AHR (Bell and Poland, 2000; Berg
and Pongratz, 2002; Carver and Bradfield, 1997; Kazlauskas et al., 2002; LaPres et al.,
2000; Lees et al., 2003; Ma and Whitlock, 1997; Meyer and Perdew, 1999; Meyer et al.,
2000; Meyer et al., 1998; Petrulis et al., 2003; Ramadoss and Perdew, 2005). However,
there has been minimal information on the function of endogenous XAP2 as it relates to
interactions with endogenous AHR proteins from species other than the murine C57BL/6J
strain since there has been a general assumption that the functions ascribed to XAP2 for
the Ahb-1 receptor complex are universal for all other Ah receptor species as is the
function of the AHR itself. The key findings of the current report challenge this view by
showing that (i) the level of endogenous XAP2 that associates with the AHR is receptor
species specific, wherein the AHRb-1 associates with a level of XAP2 that is much greater
then other mammalian AHR proteins characterized in this report independently of its
genetic background, (ii) the level of XAP2 associating with the AHR may partially alter
the subcellular location of the AHR, wherein a high level of XAP2 association with the
Ahb-1 receptor leads to cytoplasmic retention of the latent AHR complex, while AHR
proteins exhibiting reduced association with XAP2 appear to dynamically shuttle through
the nucleus, (iii) increased XAP2 association with the AHR may decrease the rate of
184 ligand-induced degradation of the AHR, and (iv) association of XAP2 with the Ahb-1 receptor may stabilize a portion of the AHR population in that ~30% of the total cellular levels of AHR remain even in the presence of TCDD for 6 hours, whereas >90% loss of endogenous Ahb-2 receptor is achieved under the same conditions. Since these
associations are maintained even when each receptor species is expressed in the same
genetic background (ie: in the presence of the same total levels of XAP2 and AHR), these
results collectively imply that XAP2 may only impact AHR-mediated signaling in the context of the Ahb-1 receptor and suggest that the analysis of the AHR-mediated signaling via rat and mouse Ahb-2 receptors may better represent the physiology of this signal
transduction pathway across species and in the human.
Since these studies revealed biochemical differences between the Ahb-1 receptor
and AHR proteins from other mammalian species, it will be important to assess the
functional relevance of these differences, particularly in terms of gene regulation. Since
genetic variation will always complicate such studies if different cell lines or organismal
models are used, the model system of expressing each receptor to the same level in the
same genetic background using the same promoter will prove useful for such studies.
Additionally, non-mammalian Ah receptors could be placed in the same genetic context
as the Ahb-1, though it is possible that non-murine Ah receptors would exhibit differences
from the Ahb-1 or from their endogenous physiology as a result of not having their
specific species AHR pathway components. However, since such studies can always be
compared to lines endogenously expressed alternate AHR species, such studies could be
performed if care is taken to ensure that the results obtained are indeed related to AHR
function.
185 High levels of XAP2 association with the AHR have also been linked to
displacement of p23 from hsp90 (Hollingshead et al., 2004). However, these studies
focused on overexpression of cellular XAP2 in COS cells transfected with Ahb-1 receptor and an increase in XAP2 association with the AHR was minimal (Hollingshead et al.,
2004). Since the studies shown in this report suggest that 100% of the Ahb-1 receptor
pool is associated with XAP2 (Figure A-5), it is not surprising that no increase in XAP2
association was seen when XAP2 was overexpressed. Thus, such studies should be
repeated in the context of other AHR species that exhibit reduced XAP2 association and,
preferably, in cells stably expressing the AHR to reduce population differences resulting from transfection efficiency. Importantly, p23 appears to be involved with stabilization
of the latent AHR complex with Hsp90 and the subsequent ability of the AHR complex
to heterodimerize with ARNT, and is also involved in the modulation of function for
many intracellular receptors such as the glucocorticoid receptor and members of the
intracellular receptor family (Dittmar et al., 1997; Freeman et al., 2000; Shetty et al.,
2003; Wochnik et al., 2004).
In light of these studies, it could be suggested that increased XAP2 association
with the AHR would lead to displacement of p23 and subsequent destabilization of the
latent AHR complex; however, the opposite of this effect was observed. In the studies
described in Chapter Two, the Ahb-1 exhibits high XAP2 association and, yet, exhibits
increased stabilization, though the studies by Hollingshead et al. (2004) would imply that
high levels of XAP2 association with the AHR would result in the displacement of p23
and subsequent destabilization the AHR complex. Thus, it would be pertinent to reassess
the studies performed by Hollingshead et al. (2004). To evaluate this, the Ahb-1 and the
186 non-Ahb-1 receptors examined in this report, which exhibited differential association with
XAP2, should be evaluated for p23 association in the latent AHR complexes. Such
studies could easily be performed using the same techniques as those shown for
AHR•XAP2 analysis. If non-Ahb-1 receptor species exhibit increased p23 association,
these data would be supportive of XAP2’s role in displacement of p23, yet would then
appear to be contradictory of the reported role of p23 in stabilization of the AHR since
these receptor species undergo a more rapid degradation following ligand-activation.
Likewise, unless non-Ahb-1 receptor species exhibited decreased p23 association, which
would not be supportive of XAP2’s role in displacement of p23, there would be no apparent correlation between AHR•hsp90•p23 association and the role of p23 in
stabilization of the AHR. However, since Ahb-2 complexes can be generated that
associate with increased levels of XAP2 (Figures 2.7 and 2.8) compared to endogenous
b-2 Ah when XAP2 is transfected into AHb2 stable lines, these lines will be an ideal model
for directly analyzing the relationship between the association of AHR•XAP2 and that of hsp90•p23 since the effect of increasing XAP2 association with the AHR can be evaluated in the context of a receptor capable of exhibiting such an increase.
XAP2 knockout studies have also recently been performed in the C57BL/6J mouse (Lin et al., 2007). In these studies, knockout of XAP2 was embryonically lethal with most homozygous null mice dying by embryonic day 15-17 as a result of severe vascular and cardiac abnormalities including the presence of a double outlet right ventricle (DORV) and reduced blood flow to the extremities, suggesting a role for XAP2 in cardiac development. The authors of this study suggest that these data indicate a role for XAP2 outside of AHR and PPARα function. It is striking, however, that AHR
187 signaling has also been implicated in cardiac development by other studies in that i) AHR
null mice have been demonstrated to exhibit cardiac hypertrophy resulting from
thickened ventricular walls and increased cellular proliferation in the heart and ii) TCDD-
treated mice have been shown to also exhibit cardiac hypertrophy, edema, and an increased formation of ventricular septal defects (VSDs), which are highly associated with DORV (Thackaberry et al., 2003; Walker et al., 1997). Since loss of XAP2 in the
C57BL/6J mouse Hepa-1 cell line results in increased nucleocytoplasmic shuttling of the
AHR and a more rapid degradation profile, loss of XAP2 in the C57BL/6J mouse may lead to altered regulation of AHR target genes or complete depletion of the AHR in the presence of ligand and in either case, may be contributing to the formation of DORV.
Furthermore, since it appears that XAP2 may stabilize a portion of the AHR in the
C57BL/6J mouse, cardiac abnormalities seen with loss of XAP2 could correlate with results obtained through loss of the AHR in knockout or TCDD-treated mice; however,
C2C12 mice do not constitutively develop cardiac conditions though the Ahb-2 is
endogenously associated with reduced levels of XAP2. Therefore, it remains possible
that loss of XAP2 contributes to formation of DORV through disruption of AHR
signaling by destabilization of the receptor in the presence of ligand. However, since the
authors chose to evaluate loss of XAP2 in the C57BL/6J mouse, which expresses the
Ahb-1 receptor that associates with high levels of XAP2, it would be pertinent to assess
loss of XAP2 in the BALB/cBy or C3H mouse as well, which express the Ahb-2 receptor
that associates with reduced levels of XAP2. Care should also be taken that the animal
foods being used contain no AHR agonists as the presence of AHR ligands would lead to
loss of AHR and could contribute to cardiac abnormalities. Such studies may further
188 support evidence that the effects of loss of XAP2 could be mediated through AHR
signaling if different physiological abnormalities are seen in different murine strains.
FKBP12 (human FKBP51) is structurally similar to XAP2 and both proteins contain FK (FK506-binding), TPR (tetratricopeptide), and peptidylprolyl isomerase (PPI) domains, and both proteins associate with Hsp90 through the TPR domain, yet XAP2 lacks the ability to bind immunosuppressive drugs through its FK domain and also appears to lack PPI activity in which PPI interacts with dynein to regulate receptor
localization (Carver et al., 1998; Galigniana et al., 2002; Galigniana et al., 2001). Instead,
XAP2 interacts with the AHR through its FK domain (Meyer et al., 1998; Peattie et al.,
1992). It is also striking, however, that loss of the immunophilin FKBP12, which
functions in the release of Ca2+ from the sarcoplasmic reticulum in skeletal muscle, in
knockout mice is also embryonic lethal, leading to death between embryonic day 14.5
and birth, resulting from cardiac hypertrophy and VSDs (Shou et al., 1998). Furthermore,
these effects are seen only in male FKBP12 null mice, while null-females exhibit cardiac
hypertrophy only in the presence of estrogen receptor antagonists, though both sexes
exhibit deficiencies in Ca2+ release. Thus, these results suggest a role for estrogen in
protection from loss of FKBP12 induced cardiac hypertrophy. It may also then be
pertinent to assess whether female XAP2 null mice exhibit a similar lack of cardiac hypertrophy as seen in males.
Mutations in the FK domain of XAP2 (R304X), which contains an AHR binding region, have also been implicated in predisposition for pituitary adenomas (Georgitsi et al., 2007; Vierimaa et al., 2006). Since little is known regarding the function of XAP2
outside of its interaction with the AHR, future studies should also therefore be aimed at
189 further analyzing the potential physiological role(s) of XAP2, focusing on its apparent
role in cardiac development and whether XAP2 also serves a role in Ca2+ regulation since
knockout of FKBP12 resembles that of XAP2. Such studies could involve yeast two-
hybrid screens to identify XAP2 interacting proteins, conditional knockouts of XAP2 in
the heart, and Northern or microarray analysis of cardiac tissues in XAP2 and XAP2-/- animals, followed by molecular evaluation of potential gene targets.
5.2 Implications of AHR Degradation Studies
Importantly, association with XAP2 may also affect AHR degradation. Following
TCDD binding, the AHRb-1 is rapidly depleted by 80-95% within 4-6 hours of treatment
in numerous cell culture models and does not return to basal levels as long as ligand is
present in the media (Pollenz, 1996; Reick et al., 1994). As detailed in Chapters Two and
Three, this degradation is even more rapid with the Ahb-2, human or rat receptors, which
deplete by 90-100% within 2 hours of treatment. The importance of this loss is
underscored by the variety of physiological defects seen in AHR-/- mice (Andreola et al.,
1997; Fernandez-Salguero et al., 1995; Gonzalez et al., 1995; McDonnell et al., 1996)
and the evidence that a single oral dose of TCDD can lead to sustained depletion of AHR
proteins in the liver, spleen, thymus, and lung in vivo and that such a depletion correlates with reduction in TCDD-mediated reporter gene expression in mammalian culture cells following a second dose of TCDD (Pollenz et al., 1998). It is important to note that many of the phenotypes seen in TCDD-treated mice are similar to those reported for
AHR-/- mice that have not been exposed to TCDD. Thus, this loss of AHR may
contribute to some of the biological effects of xenobiotics. Degradation of the AHR,
190 therefore, which may be used as a means of attenuating the transcriptional response, can
have significant repercussions on the future signaling ability of the AHR pool as well as
biological implications. Thus, a more rapid degradation profile in animals expressing alternate Ah receptor species would likely impact the toxic effects that result from loss of
receptor, particularly since the AHRb-1 does not appear to exhibit >80% reduction in cell
culture, while the AHRb-2 and rat AHR appear to be nearly 100% degraded in response to
TCDD. Thus, in the presence of sufficient TCDD or other PAHs or HAHs, animals
expressing non Ahb-1 receptors may exhibit increased susceptibility to these toxins as a
result of total loss of receptor, particularly if the AHR has other endogenous functions
that may not be disrupted with >20% AHR remaining.
Conversely, blockage of degradation by the translation inhibitor CHX in the
context of a functional AHR appears to result in potentiation of gene induction
(superinduction), whereby genes regulated by the AHR are induced to a higher level and for a longer period of time (Ma and Baldwin, 2000; Pollenz and Barbour, 2000).
However, in stable lines expressing amino-terminally tagged AHR proteins that exhibit a
reduced magnitude of degradation, but are capable of inducing CYP1A1 (Figures 3.14
and 3.15), superinduction at the protein level is not readily apparent. In fact, the GFP-
AHR exhibits both a reduced magnitude of degradation as well as a reduced level of
CYP1A1 induction (Figure 3.15). This suggests that blockage of AHR degradation is not
sufficient for superinduction of CYP1A1, although evaluation of CYP1A1 at the mRNA
level should also be performed since the studies describing superinduction using CHX
could not evaluate CYP1A1 protein.
191 From the results presented in this report, it appears that the AHR can be degraded
through several distinct mechanisms: (i) ligand-dependant degradation that can be
blocked by pre-treatment with the transcription inhibitor AD or the translation inhibitor
CHX without affecting nuclear localization or DNA binding ability of the receptor
(Pollenz et al., 2005) that requires DNA binding by AHR•ARNT, which suggests that both active transcription and translation by AHR•ARNT are necessary for ligand-induced degradation of the AHR, (ii) ligand-dependent degradation in which carboxy-terminal truncated Ah receptors (Pollenz et al., 2005) as well as those defective in DNA binding or
ARNT dimerization (Chapter Three) similarly exhibit a very low level of degradation following treatment with TCDD; however, this degradation does not require DNA binding and cannot be blocked by either AD or CHX, suggesting that this loss is not representative of the typical degradation seen following ligand binding with the wild-type receptor, (iii) ligand-independent degradation typified by treatment with geldanamycin
(GA) that leads to nuclear translocation of the receptor and its subsequent degradation,
but cannot be blocked by treatment with either AD or CHX and occurs without disruption
of the AHR complex itself and, importantly, without DNA binding or subsequent gene
induction (Chen et al., 1997; Meyer et al., 2003b; Song and Pollenz, 2002), and (iv)
endogenous turnover of the AHR about which little is known.
Thus, the precise mechanisms involved with regulating the degradation of the
AHR are still not resolved. It is intriguing to consider that several studies have shown
that the degradation of some nuclear receptors is correlated to transactivation, DNA
binding, and the recruitment of co-activators to the promoter/enhancer region of
responsive genes (Reid et al., 2003; Salghetti et al., 2001; Verma et al., 2004), which
192 appears to be the case with ligand-induced degradation of the AHR. The strong
association of XAP2 with the Ahb-1 receptor complex (as compared with rat and Ahb-2 receptors), may slow down the transformation process and prevent a population of AHRs from interacting with ARNT, binding DNA, and being degraded as may the presence of amino-terminal tags on the AHR, though these characteristics have not been directly assessed. To assess this, it would be possible to examine the intensity of DNA binding of
in vitro expressed Ahb-1 receptor and ARNT in rabbit reticulocytes. Unprogrammed
reticulocytes can be assessed for XAP2 expression by Western blotting to determine
whether or not they express XAP2 endogenously. If so, rabbit reticulocytes could be
used from which XAP2 had been removed. This could be achieved using a Millipore
centricon filtration device followed by immunodepletion of the ~37 kDa fraction using
XAP2 IgG and protein A/G beads in a manner similar to that outlined by Shetty et al.
(2004). In either case, in vitro produced XAP2 could be added along with AHR and
ARNT in activation reactions in increasing amounts and portions of each activation
reaction removed every 15 min for 2 hours and run on an EMSA as previously described
to examine AHR•ARNT DNA binding. A portion of each sample could also be
immunoprecipitated to examine the association of XAP2 with the AHR complex. In the
presence of increasing XAP2 up to saturation of the AHR, it would be expected that the
presence of XAP2 would stabilize the AHR complex and slow the AHR•ARNT
transformation process as evidenced by a decreased intensity of DNA binding in XAP2
containing activation reactions at earlier time points of analysis. A significantly reduced
effect on the transformation ability and AHR saturation point would be expected in the
193 context of non-Ahb-1 receptors since these receptor species exhibit reduced association with XAP2.
Interestingly, recent studies also suggest that the Ahb-1 receptor remains associated
at the CYP1A1 promoter during ligand stimulation in Hepa-1 cells, whereas the human
AHR shows periods of association and dissociation at the CYP1A1 promoter in MCF-7
cells (Hestermann and Brown, 2003; Wang et al., 2004). Thus, it would be important to assess whether the human AHR in the Hepa-1 background continues to cycle on/off the
CYP1A1 promoter to determine whether these associations occur as a result of the AHR itself, another cellular factor, or the genome. These results would be especially important in light of a recent report that suggest that oscillation of a transcription factor at a promoter is linked to accessibility of binding sites and not an oscillatory recruitment of the transcription factor itself (Karpova et al., 2008). Similar studies could also be performed using chromatin immunoprecipitation (ChIP) time course analysis of TCDD
b-2 treated AHb2 cells as well as C2C12 cells to determine whether the Ah receptor also
cycles on/off the CYP1A1 promoter endogenously or when expressed in the Hepa-1
background. If these receptors also exhibit promoter association cycling, this would
correlate these results with the rapid degradation profile of these receptors.
In either case, both ligand-dependent and ligand-independent degradation of the AHR
have been demonstrated to occur via the 26S proteasome complex since pre-treatment
with the proteasome inhibitors MG-132 or lactacystin prior to treatment with agonist
blocks degradation of the receptor, while pre-treatment with inhibitors of calpain, serine
or cysteine proteases nor lysosomal proteases cannot (Davarinos and Pollenz, 1999; Ma
and Baldwin, 2000; Wentworth et al., 2004). As such, several studies have suggested
194 that the AHR is ubiquitinated, though none have demonstrated definitive evidence of
such events and at this time no specific E3 ligase has been demonstrated to be involved in
the degradation of the AHR nor has any lysine residue been implicated as a ubiquitination
site (Ciechanover, 2005; Kazlauskas, 2000; Ma and Baldwin, 2000). Additionally, in
studies performed in this laboratory, no accumulation of ubiquitinated AHR is seen under
the presence of AD or MG-132 (RSP, unpublished observations).
Recently, AHR ligand dependent, but not estradiol dependent, degradation of the
estrogen receptor (ERα) has been shown to occur via ubiquitination by a cullin 4b (cul4)
E3 ligase multiprotein complex containing the DNA damage binding protein 1 (ddb1),
and the ubiquitin conjugating enzyme ring finger protein RBX1/ROC1 (Ohtake et al.,
2007). Surprisingly, in the presence of AHR ligand, atypical cul4 multiprotein
complexes are formed that also appear to contain the AHR (Ohtake et al., 2007). These
cul4bAHR complexes then appear to be involved with AHR ligand-mediated degradation
of ERα since reduction of any cul4bAHR complex component reduces the degradation of
ERα. However, in contrast to these studies, ligand-dependent degradation of the AHR is
unaffected by reduction of cul4b or cul4b associated proteins such as ddb1 (RSP,
unpublished results). Thus, cul4 complexes do not appear to mediate the ubiquitination
of the AHR.
It has also been proposed that the carboxyl terminus of hsc70-interacting protein
(CHIP), an E3 ligase which contains a TRP domain for interaction with Hsp proteins and
a U-box domain for recruitment of ubiquitin conjugating enzymes, may play a role in
AHR degradation. This hypothesis stems from data suggesting that CHIP interacts with
Hsp70/Hsc70 or Hsp90 to promote the degradation of Hsp client proteins including the
195 glucocorticoid receptor and tau (Ballinger et al., 1999; Dickey et al., 2006; Lees et al.,
2003; Morales and Perdew, 2007). In these studies, CHIP appeared to be capable of
interacting with both Hsp90 and the AHR when endogenously expressed, expression of
CHIP resulted in a dramatic loss of AHR in COS-1 cells, and an in vitro degradation
assay resulted in ubiquitination of the AHR in the presence of CHIP (Morales and
Perdew, 2007). Interestingly, the studies of Morales et al. (2007) also suggested that
XAP2 can partially protect in vitro expressed Ahb-1 receptor from CHIP mediated
ubiquitination, but did not protect mutant AHR (Y408A) unable to bind XAP2. As
previously discussed, these results again correlate the reduced degradation profile of the
Ahb-1 receptor, which associates with high levels of XAP2, and the rapid degradation
profile of the Ahb-2 receptor, which associates with low levels of XAP2. However, in
these studies, expression of CHIP∆U-box in COS-1 cells also resulted in loss of AHR,
suggesting this ubiquitin conjugating domain was not essential for AHR turnover.
Additionally, loss of CHIP via siRNA in Hepa-1 cells or using CHIP knockout cell lines
did not impact the degradation of the AHR (Morales and Perdew, 2007; Pollenz and
Dougherty, 2005). Thus, CHIP is either not involved in the degradation of the AHR in
the context of Hepa-1 cells or a compensatory mechanism is at play wherein other E3
ligases can substitute for loss of CHIP.
In support of a compensatory mechanism, during CHIP knockout studies, only
50% of CHIP-/- mice died by postnatal day 30-35, while the remaining animals were
viable (Dickey et al., 2006). Since CHIP has been implicated as an E3 ligase responsible
for modulating the degradation of essential proteins, survival of CHIP knockout animals suggests that either a compensatory mechanism(s) exist for degradation of Hsp client
196 proteins, that CHIP alone is not essential for degradation of these proteins, or that
degradation of these proteins is not strictly necessary for survival.
Thus, studies must be continued to identify the E3 ligase or AHR lysine residue
necessary for degradation of the AHR. While it is possible to mutate each lysine residue
in the AHR singly to attempt to identify a mutation that would prevent ubiquitination,
there are 33 lysine residues in the Ahb-1 receptor and mutation of each individually would
be expensive. Additionally, any disruption of AHR function would prevent ligand-
dependent degradation of the receptor without necessarily identifying a key
ubiquitination site. Furthermore, more than one lysine residue could be a target for
ubiquitination, which would likely result in failure of this scanning mutagenesis to identify a key residue. Thus, techniques scanning for protein-protein interactions may serve better to identify an E3 ligase. As such, studies are currently underway in this laboratory to identify a ligase using yeast expression systems in Saccharomyces cerevisiae and gene knockout libraries. However, while many of the ligases involved with ubiquitination are highly conserved, yeast may lack the specific E3 ligase involved with targeting the AHR for degradation, particularly since yeast lack AHR and ARNT homologs. Yeast strain may also be important; for example, Saccharomyces cerevisiae lack cul4 while Saccharomyces pombe do not (Higa and Zhang, 2007). In a second approach, a bacterial two-hybrid screen will be employed to attempt to identify proteins interacting with the carboxy-terminal region of the AHR, where it is believed that any ubiquitination would occur. Potential gene targets from either the yeast or bacterial screen will be validated in mammalian cell culture.
197 5.3 Implications of ARNT2 Studies
AHR-mediated signal transduction may also be impacted by other cellular factors,
such as the presence/level of various cofactors or the presence/level of other potential
heterodimeric partners (ARNT proteins). The ARNT and ARNT2 proteins share > 95%
amino acid identity in the bHLH domain and exhibit >80% amino acid identity in the
PAS A and PAS B domains (Hirose et al., 1996). These domains are known to specify both protein-protein interaction and DNA binding of the various bHLH/PAS proteins
(Pongratz et al., 1998; Reisz-Porszasz et al., 1994). Because of this high level of identity
in these regions, it is striking that the knockout of either ARNT and ARNT2 in the mouse
produces different phenotypes (Hosoya et al., 2001; Keith et al., 2001; Kozak et al., 1997;
Maltepe et al., 1997), because it would appear that both proteins have the same potential
to dimerize with common bHLH/PAS partners. However, while many of the potential
partners for ARNT have been evaluated both in vivo and in vitro, the ability of ARNT2 to
associate with these heterodimeric partners is less well established. Thus, because there
is limited functional analysis of the ARNT2 protein in mammalian model systems, the
goal of the studies outlined in Chapter Four was to investigate whether ARNT and
ARNT2 could interact with a common dimerization partner in vitro and in cell culture.
Since different phenotypes could also result from differential tissue/cellular expression of
ARNT proteins or discrepancies in the level of ARNT proteins in these tissues/cells, it was also pertinent to determine the endogenous protein expression of these proteins in the mouse.
Therefore, experiments were performed to assess the ability of ARNT or ARNT2 to associate with the AHR. The analysis of this interaction was chosen since the
198 interaction of ARNT with the AHR can be easily assessed in vitro where the
stoichiometry of the expressed proteins can be precisely controlled and because there
have been discrepancies in the literature as to whether or not this interaction is capable of
occurring (Hirose et al., 1996; Sekine et al., 2006). Early studies of ARNT2•AHR
interaction suggested that ARNT2 could dimerize with the AHR, could bind XREs, and
could induce XRE controlled luciferase reporter activity when coexpressed with the AHR
in the presence of ligand (Hirose et al., 1996). In more recent studies, results from the
same laboratory suggested that the ARNT2 did not dimerize with the AHR (Sekine et al.,
2006). The results presented in Chapter Four challenge this second view, in that they i)
clearly show that both ARNT and ARNT2 can specifically interact with the AHR in a
TCDD-dependant manner in vitro, ii) that both proteins compete with each other equally to form dimers with the TCDD-bound AHR, and iii) that ARNT2 was able to out- compete ARNT for binding to the AHR when expressed in excess of ARNT.
However, the studies designed in Chapter Four also sought to determine the reason for any inequity between the previous studies and our own. A major finding of these studies was that the ability of ARNT2 to associate with the AHR appeared to decrease (in comparison to ARNT) when the AHR was activated by ligands other than
TCDD. Indeed, when the low affinity non-halogenated polyaromatic hydrocarbons BAP or 3-MC were used to activate the AHR in the in vitro assay in the presence of equal concentrations of ARNT and ARNT2, approximately two-thirds of the DNA bound complexes formed were AHR•ARNT. These findings were also observed when the in vitro assays were carried out with ARNT or ARNT2 protein expressed in cell culture lines that lacked either ARNT protein. This is an intriguing finding that suggests the
199 AHR may have a slightly different conformation when bound with different ligands.
This implies that dimerization potential may be influenced by type of ligand and this may
partially explain why recent studies suggest that ARNT2 does not interact with the AHR
(Sekine et al., 2006). However, there is little known about how the ligand binding
domain impacts the structure of the HLH or PAS regions at the molecular level.
Homology modeling is currently being used to gain insight into this important question
(Pandini et al., 2007).
In contrast to the in vitro studies, a physiological model of expression of ARNT2
using the ARNT-deficient LA-II variant of the Hepa-1 cell line showed that ARNT2 was
unable to support TCDD-mediated induction of the endogenous CYP1A1 gene.
Furthermore, not only was ARNT2 unable to support TCDD-mediated induction of the
endogenous CYP1A1 gene, expression of ARNT2 in the context of the Hepa-1 line
actually appeared to partially inhibit the positive function of ARNT in AHR-mediated
signaling by an unknown mechanism. The ability of ARNT isoforms to function as
dominant negative regulators of AHR-mediated signaling has previously been reported
for the ARNTa splice variant expressed in rainbow trout and zebrafish (Necela and
Pollenz, 1999; Necela and Pollenz, 2001; Pollenz et al., 1996). In the case of the
rtARNTa, negative function is manifested by dimerization with the AHR but lack of binding to XRE enhancers, resulting in squelching of the AHR.
A similar mechanism may explain the current results with ARNT2, as Western
blotting and EMSA studies failed to detect AHR•ARNT2 complexes in nuclear extracts
of Hepa-1 cells expressing ARNT2 protein, but immunoprecipitation analyses of these
lines revealed a possible association between ARNT2 and the AHR. Unfortunately,
200 endogenous association of AHR and ARNT2 could not be detected in cell lines
expressing both ARNT isoforms. As previously discussed, this may have arisen from a lack of sensitivity in the probing of the resultant precipitations. Future studies in this area
should therefore seek to repeat these studies using greater amounts of cytosolic and nuclear extracts and/or more sensitive ECL techniques, such as the SuperSignal® West
Femto Maximum Sensitivity Substrate ECL kit (Pierce, Rockford, IL) that allows detection of as little as 1 fg of HRP conjugated protein—approximately 5000x more sensitive than the ECL kit (Amersham, Piscataway, NJ) used to analyze the Western blots shown in Chapter Four. These techniques may allow for the detection of dimers that were previously undetectable. Additionally, to reveal whether the negative impact of
ARNT2 expression on AHR signaling stems from ARNT2•AHR dimerization, it would be pertinent to dose Hepa-1 cells expressing ARNT2 with ligands other than TCDD, such as BAP or 3-MC. Since these low-affinity ligands appear to preferentially result in the formation of AHR•ARNT DNA binding heterodimers in vitro, reducing the formation or
DNA binding ability of AHR•ARNT2 dimers, similar reductions in CYP1A1 following dosing with BAP or 3-MC may imply that these reductions are not caused by DNA binding of AHR•ARNT2 dimers, but through another undefined mechanism.
It is also possible that ARNT2 expressed in culture loses the ability to form AHR complexes capable of associating with DNA as evidenced by the lack of accumulation in nuclear lysates of TCDD treated cells expressing ARNT and ARNT2. Such a mechanism could involve the interaction with tissue specific proteins yet to be defined that are not found in cytosolic fractions used in the in vitro activation assay, the formation of homodimers that limit the pool of ARNT2 that is available, the formation of a non-
201 functional AHR•ARNT2 complex or compartmentalization or heterodimerization with
ARNT. Loss of ARNT2 function in vivo would be consistent with studies that have evaluated the function of ARNT2 in zebrafish. In these studies, it has been shown that although ARNT2 is the predominate ARNT protein in aquatic species and can associate with AHR and bind DNA in vitro, knockdown of ARNT2 levels using antisense morpholino oligonucleotides in vivo does not reduce AHR-mediated signaling in this organism. In contrast, reducing AHR or ARNT in zebrafish completely abolishes the effects of TCDD on the development of embryos indicating that zfARNT2 cannot compensate for the loss of ARNT in vivo (Bello et al., 2004; Carney et al., 2004; Prasch et al., 2004; Prasch et al., 2006; Prasch et al., 2003; Teraoka et al., 2003). However, this does not address the results demonstrating that ARNT2 expressed in culture remains capable of interacting with the AHR and binding XREs when cytosolic extracts from cells are in vitro activated (Figure 4.19). Thus, it is the presence of a component of the nuclear extract, a result of the preparation of the nuclear extract, or a result of the conditions for the in vitro activation that leads to the apparent inability of ARNT2 to function in AHR signaling.
Intriguingly, nuclear extracts prepared from cell lines expressing ARNT2 (either transiently or endogenously) exhibit a constitutive association of ARNT2 with nuclear structures in direct contrast to ARNT or AHR, which exhibit a ligand-dependent increase in nuclear extracts (Figures 4.24 and 4.26). Thus, these results suggest a nuclear role for
ARNT2 that may prevent ARNT2 from being accessible to the liganded AHR (Figure
5.1). If this were true, it may be pertinent to assess whether the addition of ARNT2 to nuclear extracts from NRK cells would generate AHR•ARNT2 dimers as seen with
202
Figure 5.1: Hypothetical model for lack of ARNT2 function in vivo. Since the majority of ARNT2 appears to be constitutively associated with nuclear structures as previously described, it is possible that ARNT2 is constitutively associated with other PAS proteins or otherwise bound to DNA as a homodimer. Such an association would potentially lead to ARNT2 being inaccessible to heterodimerization with the AHR following AHR activation.
203 cytosolic extracts in Figure 4.25. If excess ARNT2 led to the formation of AHR•ARNT2 dimers, this would further suggest that AHR remains capable of dimerizing with ARNT2 and binding DNA in cell culture and that ARNT2 may have an endogenous function that impairs its ability to freely associate with the ligand activated AHR in the nucleus. If another nuclear component is preventing the formation of AHR•ARNT2 dimers, then additional ARNT2 should eventually swamp out the competitor component allowing for heterodimerization with the AHR. Thus, further studies should be performed to assess the ability of ARNT2 isolated from nuclear extracts to function during AHR signaling and to further identify possible reasons for the difference between in vitro activated cytosolic
ARNT2 to dimerize with the AHR versus ARNT2 from nuclear extracts.
Therefore, it would also be important to assess whether protein•protein interactions or phosphorylation status of ARNT2 in nuclear extracts was impacting the ability of ARNT2 to associate with the AHR. One of the best available tools to analyze the components of protein complexes is mass spectrometry analysis, especially where the potential components of these mixtures may be previously unknown. However, the difficulty of such studies tends to stem from the availability or expense of mass spectrometry equipment and trained personnel and from the purification of the native proteins. In an attempt to assess the potential interactions of ARNT2, ARNT2 protein complexes can be precipitated from cell extracts endogenously expressing ARNT2 either by immunoprecipitation techniques as previously described or through affinity chromatography, and the components assessed by mass spectrometry. Tandem affinity purification (TAP) of proteins can be performed against proteins tagged with the “TAP” tag, which is composed of two IgG binding domains of the Staphylococcus aureus
204 protein A and a calmodulin binding peptide separated by a TEV protease cleavage site
(Puig et al., 2001). To achieve purification of ARNT2 complexes, TAP-tagged ARNT2
from total cell lysates or nuclear extracts can be run through an affinity column coated
with Staphylococcus aureus protein A IgG beads. Elution of the proteins is subsequently
triggered by treatment with TEV protease and the elutant subjected to a second round of
purification using calmodulin beads and re-eluted following treatment with EGTA. The
use of this technique will allow for high yield of protein complexes in their native states.
With the ability to generate large amounts of these purified complexes, functional studies
can be performed using these complexes and mass spectrophotometry can be employed to
identify components of ARNT2 containing complexes. Furthermore, even without mass
spectrophotometry, purified ARNT2 containing complexes can be denatured and the
protein mixtures assessed for potential interactions by Western blotting, including the
potential presence of AHR in TCDD-treated cells expressing ARNT and/or ARNT2.
Since this purification technique allows for high-yield purification, this will also increase
the sensitivity of the previously described studies. 2D-gel analysis or blue-native 2D gel analysis can also be used from these purified samples to assess the potential phosphorylation status of ARNT2 (or ARNT, AHR) to attempt to determine if such post- translational modifications may also be playing a role in the regulation of ARNT2 in vivo.
Additionally, little is known regarding the transformation of the liganded AHR complex to the AHR•ARNT dimer and whether the AHR component of an AHR•ARNT dimer can dynamically swap its partner protein is entirely unknown. Since in vitro
ARNT and ARNT2 have an equal ability to dimerize with TCDD-bound AHR, studies
could be initiated to analyze the dynamics of AHR transformation using this model
205 system. For these studies, in vitro activation reactions containing limiting AHR and
ARNT could be run for 1.5-2 hrs at 30°C as usual; however, after time is allowed for initial AHR•ARNT dimerization, ARNT2 could then be added to the activation reactions at various time points for an additional 1-2 hours and portions of each activation reaction run on an EMSA as previously described using IgG against ARNT or ARNT2 to examine
AHR•ARNT or AHR•ARNT2 DNA binding. A portion of each sample could also be immunoprecipitated to examine the association of ARNT proteins with the AHR complex.
In this manner, the question of whether preformed AHR•ARNT dimers could swap partners could be assessed, if the level of AHR•ARNT dimerization decreased over time in the presence of ARNT2. However, if this swapping was truly dynamic (ie: AHR could dimerize with ARNT, swap with ARNT2, swap again with ARNT, etc.), it would be expected that an “equilibrium” would be reached wherein 50% of the total dimers were
AHR•ARNT and 50% were AHR•ARNT2 as previously shown. In contrast, if preformed dimers could not swap partners, the level of AHR•ARNT DNA binding seen initially would remain constant across all samples, and in the presence of limiting AHR, no ARNT2 dimerization would be seen. If the AHR cannot swap partner proteins, it is certainly possible that ARNT2 does not associate with the AHR as a result of inaccessibility if ARNT2 is associated with other proteins.
Thus, the mechanism that underlies the lack of ARNT2 function in AHR- mediated signaling in vivo is currently undefined. However, Sekine et al., (2006) have hypothesized that ARNT2 does not dimerize with the AHR in vivo because it contains a proline and not a histidine residue at amino acid 352 within the PAS B domain. This hypothesis is based on studies in which the histidine at amino acid 378 in the PAS B
206 region of ARNT was mutated to a proline (thus resembling the ARNT2 PAS B domain)
causing the ARNT to have greatly reduced ability to function in AHR-mediated signaling
that was similar to that observed with wild type ARNT2. However, the studies did not
directly evaluate the role of the P352 in ARNT2 function by converting it a histidine and producing a protein capable of functioning in AHR-mediated signaling. However,
Sekine et al., (2006) have hypothesized that ARNT2 does not dimerize with the AHR in
vivo because it contains a proline and not a histidine residue at amino acid 352 within the
PAS B domain. This hypothesis is based on studies in which the histidine at amino acid
378 in the PAS B region of ARNT was mutated to a proline that caused ARNT to have
reduced ability to function in AHR-mediated signaling. However, the studies did not
directly evaluate the role of the P352 in ARNT2 function by converting it a histidine and producing a protein capable of functioning in AHR-mediated signaling. The PAS B domain has been implicated in contributing to the heterodimerization potential and stability of ARNT and HIF-1α/HIF-2α through interactions occurring via the PAS B central B-sheet, and mutations in this region have been suggested to affect the transcriptional ability of the overall heterodimer (Card et al., 2005; Erbel et al., 2003).
Importantly, mutation of P352 to histidine in the mouse ARNT2 did not cause the protein to function like ARNT in AHR-mediated signaling when transfected into LA-II cells.
This may be due to the observation that the PAS A domain is also critical for DNA binding and protein-protein interactions and contributes directly to AHR•ARNT XRE binding (Chapman-Smith et al., 2004; Pongratz et al., 1998). Thus, the findings that i)
ARNT2 containing the P352 can dimerize and form functional complexes with the AHR in vitro and ii) ARNT2-H does not restore ARNT-like function in cells, indicate that the
207 reduced function of ARNT2 in vivo is unlikely to be solely related to P352 residue and,
instead, is likely to involve regulatory mechanisms that are impacting the interaction of
ARNT2 with the AHR.
Since the PAS B proline/histidine residue does not explain the differences
between the ability of ARNT or ARNT2 to function in AHR signaling, it would also be
pertinent to assess which domain(s) would. Thus, swapping of functional domains between the ARNT proteins followed by their transfection into LA-II cells and evaluation
of CYP1A1/reporter gene induction in cell culture following treatment with ligand may reveal more information about which regions of ARNT when placed into the ARNT2 protein would allow ARNT2 to function in AHR mediated CYP1A1 induction. In doing so, it is tempting to believe that the TAD of ARNT versus that of ARNT2 would be the key region since this domain shares <42% identity between ARNT proteins, while other functional domains share >80% identity. With this in mind, it is evidently possible that
ARNT2 may be capable of binding DNA, but not capable of recruiting the necessary
cofactors for the subsequent transcription of CYP1A1. However, as described in Chapter
One, the ARNT domain appears to be dispensable for transactivation by the AHR•ARNT
complex, though studies in this laboratory indicated that truncated ARNT proteins could
not functionally substitute for full-length ARNT in LA-II cells (data not shown).
However, it is also necessary to note that so far, CYP1A1 induction or reporter
gene activity controlled by XREs from the CYP1A1 promoter have been the primary
endpoints examined. Thus, it remains possible that AHR•ARNT2 dimers function, even
in cell culture, in the regulation of other target genes, perhaps even in the regulation of a
different set of target genes than AHR•ARNT dimers. While other AHR•ARNT targets
208 have been identified, such as CYP1A2, CYP1B1, and Glutathione S-transferase-ya, these
targets are constitutively expressed and assessment of induction/repression is difficult.
Therefore, an attempt should be made to identify other target genes of AHR•ARNT or
AHR•ARNT2 dimers. Such studies could involve microarray screening followed by
Northern blotting of presumed targets, and subsequent protein evaluation.
In order to further understand the possible physiological role of ARNT2, it is also
necessary to evaluate its expression patterns and whether it is coexpressed with ARNT.
Initial studies of the expression patterns of ARNT and ARNT2 mRNA in mouse
suggested that ARNT2 had an expression pattern that was restricted to the kidneys and
central nervous system and this would limit the function of the ARNT2 protein (Hirose et
al., 1996; Jain et al., 1998). However, the studies presented in Chapter Four demonstrate
that while ARNT2 is expressed less ubiquitously than ARNT in the mouse, ARNT and
ARNT2 protein can be detected in the several tissues including the brain, eye, and kidney,
heart, spleen and thymus, and in many cell lines from the tissues indicating co-expression of these protein isoforms (Figure 4.22 A). Indeed, a more sensitive antibody may reveal
expression of ARNT2 in still other tissues. Interestingly, ARNT2 has been demonstrated to be the predominant form of ARNT in several fishes including Fundulus heteroclitus
and Danio rerio and more recently in the common cormorant, Phalacrocorax carbo (Lee
et al., 2007; Powell et al., 1999; Tanguay et al., 2000). Furthermore, ARNT2 protein
from aquatic species has also been shown to be able to associate with the liganded AHR
and bind XREs when synthesized in vitro including those from Fundulus heteroclitus,
Danio rerio and Xenopus laevis, but does not appear to function in vivo (Powell et al.,
1999; Rowatt et al., 2003; Tanguay et al., 2000). Collectively, these results reveal the
209 importance of studying the function of ARNT2 and demonstrate the ARNT2 protein is
fully capable of associating with the AHR and binding DNA, but other unknown
regulatory mechanism(s) or compartmentalization prevent these interactions from occurring. The ability to now have model cell culture lines that for the first time show the endogenous co-expression of ARNT and ARNT2 in the same cell, will be key systems to further define the function and interaction of these proteins across different signaling pathways.
While the studies described in Chapter Four have focused on the potential role of
ARNT2 in AHR signaling, the endogenous role of ARNT2 is likely to involve its dimerization with other bHLH/PAS proteins. Since the knockout of ARNT2 results in abnormalities of the hypothalamus in the failure to develop neuroendocrine lineages in the paraventricular and supraoptic nuclei, similar to the effects seen with knockout of the single-minded protein (SIM), the physiological role of ARNT2 appears to be as a dimerization partner with SIM (Hosoya et al., 2001; Michaud et al., 2000; Wines et al.,
1998). Surprisingly, however, while the evaluation of the interactions between SIM and
ARNT2 have been analyzed via immunoprecipitation, reporter gene studies, and two- hybrid screens, the target genes regulated by these heterodimers remains unclear. Since
ARNT2 appears to play an essential role in development of the hypothalamus, it is important that these potential gene targets be assessed. The first step in these studies would be to perform microarray analysis using mRNA extracted from the hypothalamic tissue from wild-type or ARNT2-/- mice between ED 8.5-PD 5, when expression of
ARNT2 appears to be critical for proper development. An attempt could then be made to
correlate these changes in gene expression between these tissues to those genes
210 containing the CME enhancer that appears to bind ARNT2•SIM dimers. Depending on
the results, further studies could be performed using Northern analyses to examine
specific gene expression changes or using quantitative real time PCR (qRT-PCR).
ARNT2 also appears to play a role in hypoxic response, yet the potential
differences between ARNT and ARNT2 in this role have not been evaluated. Since
many of the target genes regulated by HIF-1 are constitutive, particularly in cancer cell lines, an optimal means of studying gene regulation by HIF-1α•ARNT or HIF-
1α•ARNT2 dimers will be real time quantitative qRT-PCR. A compilation of commonly
used primer sets and specific antibodies used in immunofluorescence and
immunohistochemical analysis of hypoxic response in several species has been culled
from the literature base and is given in Appendix B focusing on studies that employed
cell lines/model systems available to this laboratory. Unlike gene regulation by
ARNT2•SIM dimers, a variety of gene targets have been identified for HIF-1α•ARNT
dimers. Thus, the ability of ARNT2 to functionally substitute for ARNT or to
heterodimerize with HIF-1α when co-expressed with ARNT can be evaluated using many
of the resources available already in this laboratory.
211
Chapter Six
Materials and Methods
6.1 Materials
2,3,7,8-tetracholorodibenzo-p-dioxin (TCDD) (≥98% stated chemical purity) was
obtained from Radian Corp. (Austin, TX) and was solubilized in dimethyl sulfoxide
(Me2SO). Benzo[a]pyrene (BAP) (≥96% stated chemical purity) and 3-
methylcholanthrene (3-MC) (≥98% stated chemical purity) were purchased from Sigma
(St. Louis, MO) and solubilized in dimethyl sulfoxide (Me2SO).
6.2 Buffers
Phosphate-buffered saline is 0.8% NaCl, 0.02%KCL, 0.14% Na2HPO4, 0.02%
KH2PO4, pH 7.4. Gel sample buffer (2X) is 125 mM Tris, pH 6.8, 4% SDS, 25%
glycerol, 4 mM EDTA, 20 mM dithiothreitol, 0.005% bromophenol blue. Tris-buffered
saline is 50 mM Tris and 150 mM NaCl, pH 7.5. TTBS is 50 mM Tris, 0.2% Tween 20,
300 mM NaCl, pH 7.5. TTBS+ is 50 mM Tris, 0.5% Tween 20, 300 mM NaCl, pH 7.5.
BLOTTO is 5% dry milk in TTBS. Lysis buffer (2X) is 50 mM HEPES, pH 7.4, 40 mM
sodium molybdate, 10 mM EGTA, 6 mM MgCl2, and 20% glycerol. Gel shift buffer (5X)
is 50 mM HEPES, pH 7.5, 15 mM MgCl2 and 50% glycerol. MENG is MOPS, EDTA,
NaN3, glycerol. RIPA is 50 mM Tris, pH 7.4, 150 mM NaCl, and 0.2% NP40.
212 6.3 Cells and Growth Conditions
Wild-type Hepa-1c1c7 (Hepa-1) mouse hepatoma cells, type II (LA-II) Hepa-1 variants, human ARPE-19 retinal pigmented epithelium cells, B35 rat central nervous system cells, TCMK-1 mouse kidney cells, NRK-49F rat kidney cells, and A498 human kidney cells were purchased from the American Type Culture Collection (Manassas, VA).
The Hepa-1, LA-II, and B35 cells were propagated in Dulbecco’s Modified Eagle’s
Medium (DMEM) supplemented with 1.5 g/L sodium bicarbonate and 10% fetal bovine serum. ARPE-19 cells were propagated in a medium containing a 1:1 mixture of DMEM and Ham’s F12 medium with 2.5 mM L-glutamine adjusted to contain 15 mM HEPES,
0.5 mM sodium pyruvate, and 1.2 g/L sodium bicarbonate supplemented with 10% fetal bovine serum. NRK-49F cells were propagated in Dulbecco’s Modified Eagle’s Medium
(DMEM) supplemented with 1.5 g/L sodium bicarbonate and 5% bovine calf serum.
TCMK-1 cells were propagated in Minimal Essential Medium (MEM) with 2 mM L- glutamine with Earle’s BSS adjusted to contain 1.5 g/L sodium bicarbonate, 0.1 mM non- essential amino acids and 1.0 mM sodium pyruvate with 10% fetal calf serum. A498 cells were propagated in RPMI medium with 10% fetal calf serum. All cells were passaged at 3-4 day intervals and were used in experiments during a 2-month period at approximately 70-90% confluence. For treatment regiments, TCDD was administered directly into growth medium for the indicated incubation times.
6.4 Antibodies
Specific antibodies against the AHR (A-1) and ARNT (R-1) were identical to those described previously (Holmes and Pollenz, 1997; Pollenz et al., 1994). All
213 antibodies are affinity-purified IgG fractions. Monoclonal mouse antibodies against the
V5 epitope were purchased from Invitrogen (Carlsbad, CA). Polyclonal mouse antibodies
against ARNT2 and CYP1A1 were purchased from Santa Cruz Biotechnology (Santa
Cruz, CA). Polyclonal rabbit B-actin antibodies were purchased from Sigma (St. Louis,
MO).
6.5 Generation of Expression Constructs
Numerous expression constructs were generated in order to obtain the results
described in this report. The primer sets that were used are listed below and are grouped
according to the type of construct that was generated and are listed by primer name
containing the species, target DNA sequence, direction of primer (5’ or 3’), and
restriction site if applicable. Mutagenesis primers are listed as mut for or mut rev (for
forward or reverse). TOPO cloning primers are listed as T/A. Site-directed mutagenesis
was performed using the Quikchange II XL site-directed mutagenesis kit as per the
manufacturer's protocol (Stratagene).
AHR NH-terminal tags
HM mouse AHR 5’HindIII 5’-TTTTAAGCTTACCACCATGGGGGGTTCTCATCAT HM mouse AHR 5’MluI 5’-TTTTACGCGTACCACCATGGGGGGTTCTCATCAT GFP-mouse AHR 5’StuI 5’-TTTTAGGCCTGCCACCATGGTGAGCAAGGGCGA GGAG GFP-mouse AHR 3’ClaI 5’-TTTTATCGATTCAACTCTGCACCTTGCTTAGGAA TGC
AHR constructs
mouse AHR b2-3’ HindIII 5’-CAATAAGCTTCTACAGGAATCCACCAGG human AHR-3’ HindIII 5’-CAATAAGCTTTTACCAGGAATCCACTGGATG human AHR-5’ XhoI 5’-CAATCTCGAGCACCATGAACAGCAGCAGCGCC
214 rat AHR-3’ HindIII 5’-CAATAAGCTGAGCTACAGGAATCCGCTGGG rat AHR-5’ XhoI 5’-CAATCTCGAGGCACCATGAGCAGCGGCGCC rat/b1/b2 AHR-5’ XhoI 5’-CAATCTCGAGGCACCATGAGCAGCGGCGCCA ACATC mouse trAHR 5’XhoI 5’ATATCTCGAGCCACCATGGGGCTGAACACAGAG TTAGAC mouse AHR 5’HpaI 5’-CAATGTTAACCCACCATGAGCAGCGGCGCCAA CATC mouse AHR 3’XhoI 5’-CAATCTCGAGTCAACTCTGCACCTTGCTTA mouse AHR b2 3’XhoI 5’-CAATCTCGAGCTACAGGAATCCACCAGGTGT rat AHR 3’XhoI 5’-CAATCTCGAGCTACAGGAATCCGCTGGGTGT human AHR 5’HpaI 5’-CAATGTTAACCCACCATGAACAGCAGCAGCG CCAAC human AHR 3’XhoI 5’-CAATCTCGAGTTACAGAATCCACTGGATGT mouse AHR R39A mut for 5’-CAAATCCTTCTAAGCGACACGCAGACCGGCTGA ACACAGAG mouse AHR R39A mut rev 5’-CTCTGTGTTCAGCCGGTCTGCGTGTCGCTTAGAA GGATTTG mouse AHR b2 3’XhoI 5’-CAATCTCGAGTCAGCAGCGGCGCCAACATC mouse AHR b2 5’HindIII 5’-CAATAAGCTTTCAGCAGCGGCGCCAACATC mouse AHR b2/rat 5’HindIII 5’-CAATAAGCTTATGAGCAGCGGCGCCAACATC mouse AHR b2 3’ClaI 5’-CAATATCGATCTACAGAATCCACCAGG rat AHR 3’ClaI 5’-CAATATCGATCTACAGGAATCCGCTGG human AHR 5’HindIII 5’-CAATAAGCTTATGAACAGCAGCAGCGCC human AHR 3’ClaI 5’-CAATATCGATTTACAGGAATCCACTGGATG mouse AHR 5’AgeI 5’-CAATACCGGTATGAGCAGCGGCGCCAACATC mouse AHR 3’PacI 5’-CAATTTAATTAATCAACTCTGCACCTTGCTTAG mouse AHR b2 3’PacI 5’-CAATTTAATTAACTACAGGAATCACCAGGTGTG rat AHR 3’PacI 5’-CAATTTAATTAACTACAGGAATCCGCTGGGTGT G human AHR 3’PacI 5’-CAATTTAATTAATTACAGGAATCCACTGGATG
ARNT and ARNT2 studies
mouse AHR 3’T/A 5’-TCAACTCTGACACCTTGCTTAG mouse AHR 5’T/A 5’-ATGAGCAGCGGCGCCAACATC mouse ARNT2 5’-T/A 5’-GCAACCCCGGCCGCCGTCAAC mouse ARNT2 3’-T/A 5’-CTACTCAGAAAATGGAGGGAA mouse ARNT2 3’SalI 5’-TTTGTCGACCTACTCAGAAAATGGAGGGA mouse AHR b2 3’-T/A 5’-CTACAGGAATCCACCAGGTGT mouse AHR 3’Stu 5’-CAATAGGCCTTCAACTCTGCACCTTGCTTAG mouse AHR b2 3’Stu 5’-CAATAGGCCTCTACAGAATCCACCAGGTGT mouse ARNT2 5’XhoI 5’-CAATCTCGAGACCACCATGGCAACCCCGGCCG CCGTCAAC
215 mouse ARNT2 3’BamH1mut 5’-CAATGGATCCAATTCCGAAGGCAGATGACTG mouse ARNT2 5’BamHI mut 5’-CAATGGATCCAGCCACCCTTACCCGGCTGAC mouse ARNT2 3’HindIII 5’-CAATAAGCTTCTACTCAGAAAATGGAGGGAACA TGCCCAG mouse ARNT V5 2nd half 5’-CCTAACCCTCTCCTCGGTCTCGATTCTACGGCGG CGACTACAGCTAACCCAG 5’Universal V5 half XhoI 5’-CAATCTCGAGCCACCATGGGTAAGCCTATCCCT AACCCTCTCCTCGGTCTC mouse ARNT2 V5 2nd half 5’-CCTAACCCTCTCCTCGGTCTCGATTCTACGGCAA CCCCGGCCGCCGTCAAC mouse ARNT 503 3’HindIII 5’-CAATAAGCTTCTAGCTGGCCAGCCCATCTCTTC CTG mouse ARNT 450 3’HindIII 5’-CAATAAGCTTTTACTGGCTAGAGTTCTTCACAT TGGTGTTGG mouse ARNT2 3’HindIII 5’-CAATAAGCTTCTACTCAGAAAATGGAGGGAACA TGC 5’V5 Universal AgeI 5’-CAATACCGGTCCACCATGGGTAAGCCTATCCC TAAC mouse ARNT 3’EcoR1 5’-CAATGAATTCCTATTCGGAAAAGGGGGGAAAC ATAGTTAG mouse ARNT2 3’EcoR1 5’-CAATGAATTCCTACTCAGAAAATGGAGGGAACA TGC 5’V5 Universal HindIII 5’-CAATAAGCTTGCACCATGGGTAAGCCTATCC mouse ARNT 3’ClaI 5’-CAATATCGATCTATTCGGAAAACGGTGGAAACA TAGTTAG mouse ARNT2 3’ClaI 5’-CAATATCGATCTACTCAGAAAATGGAGGGAAC ATGC V5 2nd half human ARNT 5’-CCTAACCCTCTCCTCGGTCTCGATTCTACGGC GGCGACTACTGCCAACC human ARNT 3’HindIII 5’-CAATAAGCTTCTATTCTGAAAAGGGGGGAAAC ATAGTTAG human ARNT 3’PacI 5’-CAATTTAATTAACTATTCTGAAAAGGGGGAAAC ATAGTTAG hARNT 378P mut for 5’-CTTCACTTTTGTGGATCCCCGCTGTGTGGCTA CTG hARNT 378P mut rev 5’-CAGTAGCCACACAGCGGGGATCCACAAAAGTG AAG mARNT2 352H mut for 5’-ACGTTTGTGGACCACAGATGCATCAGTGTG mARNT2 352H mut rev 5’-CACACTGATGCATCTGTGGTCCACAAACGTG
For the mouse ARNT2 construct, no full length ARNT2 cDNA was available,
however, two cDNA fragments were obtained one of which contained the carboxy-
216 terminal portion of ARNT2 (Sogawa fragment) while one contained the full length
ARNT2 with a 10 amino acid insertion contained within the carboxy-terminal region
(Simon clone). This second clone is believed to be an ARNT2 splice variant since a
similar variant in the rat possesses an identical insert in this region (Drutel et al., 1996).
To evaluate the full-length ARNT2 clone as published in GenBank, the amino terminal portion of the Simon clone was amplified using the following primers: mouse ARNT2
3’BamH1mut 5’-CAATGGATCCAATTCCGAAGGCAGATGACTG and mouse
ARNT2 5’BamHI mut 5’-CAATGGATCCAGCCACCCTTACCCGGCTGAC. The amplicon was then ligated to the carboxy-terminal portion of the Sogawa clone after mutagenesis reactions were performed to create a BamH1 site within each construct.
Using this BamH1 site, each amplified region was ligated together and cloned into pcDNA 3.1- for further studies. The full-length ARNT2 construct was then sequenced to test for accuracy of the construct.
For the V5-tagged constructs, each of the coding regions from ARNT and
ARNT2 was ligated into a pcDNA 3.1(-) vector allowing each of the full-length untagged proteins to be expressed. V5 tagged ARNTs were also generated. For these expression constructs, the V5 epitope (GKPIPNPLLGLDST) was added to the coding region of each
ARNT by sequential PCR along with XhoI and HindIII sites allowing the resultant PCR product to cloned into pcDNA 3.1(-). The following forward primers were used for
ARNT: 5'CCTAACCCTCTCCTCGGTCTCGATTCTACGGCGGCGACTACAGCTAA
CCCAG and 5'CAATCTCGAGCCACCATGGGTAAGCCTATCCCTAACCCTCTCCT
CGGTCTC. The reverse primer for ARNT was 5’CAATAAGCTTCTATTCGGAAAAG
217 GGGGGAAACATAGTTAG. For the V5 tag of ARNT2, the following forward primers were used: 5’-CCTAACCCTCTCCTCGGTCTCGATTCTACGGCAACCCCGGCCGC
CGTCAAC and 5’-CAATCTCGAGCCACCATGGGTAAGCCTATCCCTAACCCTCT
CCTCGGTCTC. The reverse ARNT2 primer was 5’-CAATAAGCTTCTACTCAGAAA
ATGGAGGGAACATGC. Transcription start and stop sites are underlined.
6.6 In Vitro Expression of Protein
Recombinant protein was produced from expression constructs using the TNT
Coupled Reticulocyte Lysate System essentially as detailed by the manufacturer
(Promega, Madison, WI). Upon completion of the 90-min reaction, a portion of the sample was combined with an equal volume of 2X gel sample buffer and boiled for 5 min for Western blotting and the remaining portion stored at -80°C for use in functional studies.
6.7 Transient Transfection
LA-II cells were seeded at 1.5-2.2x105 cells/well onto 35 mm dishes and propagated overnight. A cocktail containing 3-6 μg ARNT expression vector and 22 μl
LipofectAMINE™ (Gibco) transfection reagent was prepared in 1.3 mL of serum-free
DMEM. This volume was sufficient for transfecting 6 dishes with the same pool of
DNA/transfection reagent and the transfection was carried out according to the manufacturer’s instructions. After a 6-8 h transfection period, medium from each well was replaced with fresh medium containing FBS or an equal volume of 20% FBS medium was added to each well to yield an overall 10% FBS DMEM culture medium. In
218 either case, cells were allowed to recover for 16 h prior to experimental treatments. Cells
were then harvested from plates and processed as detailed below. WT Hepa-1 cells were
similarly transfected using 3-6 μg expression vector and 15 μl LipofectAMINE™.
6.8 Viral Infections and Selection.
For stable cell generation, 1x105 PT67 viral packaging cells were plated onto
35mm plates or 8.2x105 PT67 cells were plated onto 100mm plates and each well/plate
transiently transfected with one of the retroviral constructs. The cells were grown for 24
hours during which time, the viral packaging cells created infectious, but replication-
incompetent virus, resulting in the creation of high-titer viral media. After this time, the
viral media was harvested, sterile filtered through a 0.45-um filter and aliquots of viral
media used immediately for infection of target cells or frozen at -80°C for future
experimentation. Target cells plated at 7.5x104 – 1.0x105 cells on 35 mm plates or 6-
8x105 cells on 100mm plates in normal culture media were then washed and cultured
with 1000ul of the viral media containing limiting amounts of packaged virus. Based on
infection efficiency judged by the number of colonies formed following infection, some
repeated experiments included a second day of infection using another 1000ul of viral
media. In either case, 48 hours after the initial infection, the cells were harvested by
trypsinization and plated onto 150mm plates in the presence of 800μg/ml of G418 for
selection. This concentration has already been confirmed as being fatal to the cultured
cells being evaluated herein. After 7-14 days of growth, colonies were chosen and
harvested via sterile cloning disks onto 15mm culture plates still in the presence of G418
selection media. One disk was placed into each 15mm plate to allow for the growth of a
219 clonal population and the continued use of G418 will allow for a second round of
selection. At 90% confluence, the cells will be harvested and propagated onto 35mm
plates and allowed to continue to propagate until cell populations can be frozen in liquid
nitrogen and evaluated for expression of the gene of interest. Previous experiments in the
lab indicated that cells grown in this manner are typically a homogenous population in which >80% express the protein of interest. Populations expressing inconsistent levels of
the target protein in comparison to wild-type cells or inconsistencies in molecular mass
were discarded. If all lines exhibited identical responses, 1-2 lines were selected for the
analyses listed. Deviations among the cells resulted in characterization of other clones to
identify the consensus response.
6.9 Luciferase Reporter Studies
LA-II cells were transfected with the ARNT expression constructs as well as
pSV- B-Galactosidase and GudLuc 1.1 plasmids and treated with Me2SO or TCDD for 6
hours (Garrison and Denison, 2000). The luciferase reporter assay was then performed
according to the protocol dictated by the manufacturer (Promega, Madison, WI). All
luciferase values were then normalized to B-Galactosidase as a transfection efficiency
control.
6.10 RNA Interference
Annealed small interfering RNA (siRNA) complexes containing 21-bp regions of
identity to regions of the murine XAP2 or an annealed 21-bp RNA to sequence not present in the mouse genome (siRNA control) were purchased from Ambion (Austin, TX).
220 siRNA (final concentration of 50–100 nM) was transfected into 35-mm dishes containing
1–2 x 105/cells using LipofectAMINE™ reagent (Gibco). 36–48 h after transfection, cells were treated as detailed in the figure legends, and total cell lysates were harvested for
Western blotting. The efficiency of siRNA gene knockdown was determined by Western blotting. Transfection efficiency was monitored by microscopy using fluorescein isothiocyanate-labeled RNA (Clontech). In general, transfection efficiency in all experiments was >85%.
6.11 Immunofluorescence Staining and Microscopy
All immunohistochemical procedures (cell plating, fixation, and staining) were carried out as previously described (22-24). Dry, autoclaved coverslips were charged with
0.01% sterile poly-lysine. These coverslips were then placed in 35mm cell culture plates and cells were cultured directly onto the slips in appropriate medium. The slips were then fixed with 4% paraformaldehyde (pH 7.4) and ice-cold methanol. Slips were stained using the A-1A AHR IgG (2.8μg/ml) in BSA/PBS/Histidine buffer, washed with
TTBS/TTBS+ and stained with a GAR-Rhodamine secondary antibody (1:200). Cells were observed on an Olympus IX70 microscope. On average, 15–20 fields (5–20 cells each) were evaluated on each coverslip, and 3–4 fields were photographed with a digital camera at the same exposure time to generate the raw data. Nuclear fluorescence intensities of 25–50 cells in three distinct fields of view were obtained using MicroSuite image analysis software (Olympus America Inc.).
221 6.12 Preparation of Total Cell/Tissue Lysates
After treatment, cell monolayers were washed twice with PBS and detached from plates by trypsinization (0.5% trypsin/0.5 mM EDTA). Cell pellets were then washed with PBS and suspended in 50 to 100 μl of ice-cold 2X lysis buffer supplemented with
Nonidet P-40 (0.5%), leupeptin (10 μg/ml), and aprotinin (20 μg/ml). Cell suspensions were immediately sonicated for 5 s, supplemented with phenylmethylsulfonyl fluoride
(PMSF, final concentration, 100 μM), and sonicated for an additional 5 s. For total tissue lysates, frozen tissue samples were weighed and 200-750ul of 2x lysis buffer supplemented with Nonidet P-40 (0.5%), leupeptin (10 μg/ml), and aprotinin (20 μg/ml) was added to each sample depending on tissue sample size. Samples were then sonicated for 10-20 s, supplemented with PMSF, and sonicated for an additional 10-20 s. For either total cell or total tissue lysates, following sonication, a small portion of the lysate was then removed for protein determination, and the remainder was combined with an equal volume of 2X gel sample buffer, vortexed, and immediately heated for 5 min at 100°C.
Samples were then sonicated for an additional 5 s and stored at -20°C or -70°C. Protein concentrations were determined by the Coomassie Blue Plus assay (Pierce, Rockford, IL) with bovine serum albumin as the standard.
6.13 Preparation of Cytosol and Nuclear Extracts
Cell monolayers were washed twice with PBS and detached from plates by trypsinization (0.5% trypsin/0.5 mM EDTA). Cell pellets were then washed with PBS and suspended in 1000 ul of ice-cold MENG. Lysis was carried out by homogenization in small glass dounce vessels using 30-50 strokes or by vortexing for 30 s in 1% Nonidet P-
222 40 lysis buffer. Samples were then centrifuged at 5,000 rpm for 2 min at 4°C. The
supernatant was then removed for protein determination or combined with an equal volume of 2X gel sample buffer and boiled for 5 min for Western blotting to assess
protein expression. Nuclei were then washed twice with 0% Nonidet P-40 lysis buffer
and extraction performed with MENG supplemented with 400 mM KCl for 30 min,
vortexing periodically. Samples were then centrifuged at 14,000 rpm for 15 min at 4°C.
Nuclear extracts were then dialyzed in 4L of MENG at 4°C for 2 h in Slide-A-Lyzer®
Mini Dialysis Units (Pierce) according to the manufacturer’s instructions. The
supernatant was then removed for protein determination or combined with 2X gel sample
buffer and boiled for 5 min for Western blotting to assess protein expression. The
remaining sample was stored at -80°C until analysis. Protein concentrations were
determined by the Coomassie Blue Plus assay (Pierce, Rockford, IL) with bovine serum
albumin as the standard.
6.14 Western Blot Analysis and Quantification of Protein
Protein samples were resolved by denaturing electrophoresis on discontinuous
polyacrylamide slab gels (SDS-PAGE) and were electrophoretically transferred to
nitrocellulose. Immunochemical staining was carried out with varying concentrations of
primary antibody (figure legends) in BLOTTO buffer supplemented with DL-histidine
(20 mM) for 1 hour at 22°C. Blots were washed with three changes of TTBS or TTBS+
for a total of 45 min. The blot was then incubated in BLOTTO buffer containing a
1:10,000 dilution of goat anti-rabbit- or goat anti-mouse-HRP secondary antibodies for 1
h at 22°C and washed in TTBS or TTBS+ as above. Before detection, the blots were
223 washed in PBS for 5 min. Bands were visualized with the enhanced chemiluminescence
(ECL) kit as specified by the manufacturer (Amersham Biosciences, Piscataway, NJ).
Multiple exposures of each set of samples were produced. The relative concentration of
target proteins were determined by computer analysis of the autoradiographs as detailed
previously (Pollenz, 1996; Sojka et al., 2000).
6.15 In Vitro Activation of AHR•ARNT Complexes and Electrophoretic Mobility Shift
Assays
50 ug of cytosol or approximately equal amounts of in vitro translated ARNT or
ARNT2, as determined by quantitative Western blotting, were mixed with in vitro
translated AHR and combined with 25 mM MOPS, 10 mM EDTA, and 10% glycerol buffer in a reaction of a total volume proportionate to the number of samples being
evaluated such that each sample being evaluated was taken from a pooled total reaction
and therefore represents identical samples. Each sample was then supplemented with
TCDD (170 nM), 3-MC (54 uM), BAP (1.7 uM) or Me2SO (0.5%) and incubated at 30°C for 2 h. To examine specificity of binding, 50-100 ng of antibodies against AHR, ARNT, or ARNT2 were also included in the activation reactions. For EMSA, double-stranded
fragments corresponding to the consensus XREs A, B, C, D, E and F of the murine
CYP1A1 promoter have been described previously (Shen and Whitlock, 1992). XRE
duplexes corresponding to the six XREs of the murine CYP1A1 promoter as described by
(Lusska et al., 1993) were purchased from Integrated DNA Technologies, Inc. (Coralville,
IA). Approximately 4 ng of each 32P-labeled XRE was added to each sample, and the
incubation continued for an additional 15 min at 22°C. The samples were then resolved
224 on 5% acrylamide/0.5% 45 mM Tris-borate and 1 mM EDTA gels, dried, and exposed to film. A portion of the activated samples were also combined with an equal volume of 2X gel sample buffer and boiled for 5 min for Western blotting to assess protein expression.
The relative DNA binding intensity of EMSA samples were determined by computer analysis of the EMSA autoradiographs as detailed previously (Pollenz, 1996; Sojka et al.,
2000).
Oligo duplexes used in the EMSA reaction are given below showing the two guanine residue overhang used for [32P]-dCTP labeling. Core sequences are bolded and mutated residues underlined.
mXRE A 5’-GGCCAAGCTCGCGTGAGAAGCG-3’ 3’-GGTTCGAGCGCACTCTTCGCGG-5’ mXRE B 5’-GGGCTTGGCACGCACACAGGTT-3’ 3’-CGAACCGTGCGTGTGTCCAAGG-5’ mXRE C 5’-GGGAGGCTAGCGTGCGTAAGCC-3’ 3’-CTCCGATCGCACGCATTCGGGG-5’ mXRE D wild-type 5’-GGCCGGAGTTGCGTGAGAAGAG-3’ 3’-GGCCTCAACGCACTCTTCTCGG-5’ mXRE D mutant 5’-GGCTCTTCTCGCGTAACTCCGG-3’ 3’-GAGAAGAGCGCATTGAGGCC-5’ mXRE E 5’-GGAGTGCTGTCACGCTAGCTGG-3’ 3’-TCACGACAGTGCGATCGACCGG-5’
225 mXRE F 5’-GGCCGGGTTTGCGTGCGATGCT-3’ 3’-GGCCCAAACGCACGCTACGAGG-5’
E-box wild-type 5’-GGCCCGGTCACGTGGCCTACG-3’ 3’-GGGCCAGTGCACCGGATGCGG-5’
E-box mutant 5’-GGCCCGGTCGCATGGCCTACG-3’ 3’-GGGCCAGCGTACCGGATGCGG-5’
6.16 Immunoprecipitations
For the immunoprecipitations, 30 ul of in vitro translated AHR, ARNT, or
ARNT2 were precipitated in RIPA buffer supplemented with bovine serum albumin (20
μg/ml), histidine (20 mM), 1 μg specific or pre-immune IgG and 15 ul Protein A/G agarose (Pierce) for 2 hours at 4°C. Pellets were washed with 800 ul TTBS three times for 10 minutes at 4°C and protein eluted by boiling in 30 ul SDS sample buffer. Samples were centrifuged at 14,000 rpm and the supernatant resolved by SDS-PAGE and Western blotting as described above.
6.17 Statistical Analysis
Statistical analysis was carried out using InStat software (GraphPad Software Inc.,
San Diego, CA).
226
Literature Cited
1. Abbott BD (1995) Review of the interaction between TCDD and glucocorticoids in embryonic palate. Toxicology 105(2-3):365-373.
2. Abbott BD and Probst MR (1995) Developmental expression of two members of a new class of transcription factors: II. Expression of aryl hydrocarbon receptor nuclear translocator in the C57BL/6N mouse embryo. Dev Dyn 204(2):144-155.
3. Abnet CC, Tanguay RL, Hahn ME, Heideman W and Peterson RE (1999) Two forms of aryl hydrocarbon receptor type 2 in rainbow trout (Oncorhynchus mykiss). Evidence for differential expression and enhancer specificity. J Biol Chem 274(21):15159-15166.
4. Adachi J, Mori Y, Matsui S, Takigami H, Fujino J, Kitagawa H, Miller CA, 3rd, Kato T, Saeki K and Matsuda T (2001) Indirubin and indigo are potent aryl hydrocarbon receptor ligands present in human urine. J Biol Chem 276(34):31475-31478.
5. Aitola M, Sadek CM, Gustafsson JA and Pelto-Huikko M (2003) Aint/Tacc3 is highly expressed in proliferating mouse tissues during development, spermatogenesis, and oogenesis. J Histochem Cytochem 51(4):455-469.
6. Aitola MH and Pelto-Huikko MT (2003) Expression of Arnt and Arnt2 mRNA in developing murine tissues. J Histochem Cytochem 51(1):41-54.
7. Andersson P, McGuire J, Rubio C, Gradin K, Whitelaw ML, Pettersson S, Hanberg A and Poellinger L (2002) A constitutively active dioxin/aryl hydrocarbon receptor induces stomach tumors. Proc Natl Acad Sci U S A 99(15):9990-9995.
227 8. Andreasen EA, Spitsbergen JM, Tanguay RL, Stegeman JJ, Heideman W and Peterson RE (2002) Tissue-specific expression of AHR2, ARNT2, and CYP1A in zebrafish embryos and larvae: effects of developmental stage and 2,3,7,8- tetrachlorodibenzo-p-dioxin exposure. Toxicol Sci 68(2):403-419.
9. Andreola F, Fernandez-Salguero PM, Chiantore MV, Petkovich MP, Gonzalez FJ and De Luca LM (1997) Aryl hydrocarbon receptor knockout mice (AHR-/-) exhibit liver retinoid accumulation and reduced retinoic acid metabolism. Cancer Res 57(14):2835-2838.
10. Anthony-Cahill SJ, Benfield PA, Fairman R, Wasserman ZR, Brenner SL, Stafford WF, 3rd, Altenbach C, Hubbell WL and DeGrado WF (1992) Molecular characterization of helix-loop-helix peptides. Science 255(5047):979-983.
11. Antonsson C, Arulampalam V, Whitelaw ML, Pettersson S and Poellinger L (1995) Constitutive function of the basic helix-loop-helix/PAS factor Arnt. Regulation of target promoters via the E box motif. J Biol Chem 270(23):13968- 13972.
12. Arpiainen S, Lamsa V, Pelkonen O, Yim SH, Gonzalez FJ and Hakkola J (2007) Aryl hydrocarbon receptor nuclear translocator and upstream stimulatory factor regulate Cytochrome P450 2a5 transcription through a common E-box site. J Mol Biol 369(3):640-652.
13. Arpiainen S, Raffalli-Mathieu F, Lang MA, Pelkonen O and Hakkola J (2005) Regulation of the Cyp2a5 gene involves an aryl hydrocarbon receptor-dependent pathway. Mol Pharmacol 67(4):1325-1333.
14. Baba T, Mimura J, Gradin K, Kuroiwa A, Watanabe T, Matsuda Y, Inazawa J, Sogawa K and Fujii-Kuriyama Y (2001) Structure and expression of the Ah receptor repressor gene. J Biol Chem 276(35):33101-33110.
15. Bacsi SG and Hankinson O (1996) Functional characterization of DNA-binding domains of the subunits of the heterodimeric aryl hydrocarbon receptor complex imputing novel and canonical basic helix-loop-helix protein-DNA interactions. J Biol Chem 271(15):8843-8850.
228 16. Ballinger CA, Connell P, Wu Y, Hu Z, Thompson LJ, Yin LY and Patterson C (1999) Identification of CHIP, a novel tetratricopeptide repeat-containing protein that interacts with heat shock proteins and negatively regulates chaperone functions. Mol Cell Biol 19(6):4535-4545.
17. Barrow LL, Wines ME, Romitti PA, Holdener BC and Murray JC (2002) Aryl hydrocarbon receptor nuclear translocator 2 (ARNT2): structure, gene mapping, polymorphisms, and candidate evaluation for human orofacial clefts. Teratology 66(2):85-90.
18. Bell DR and Poland A (2000) Binding of aryl hydrocarbon receptor (AhR) to AhR-interacting protein. The role of hsp90. J Biol Chem 275(46):36407-36414.
19. Bello SM, Heideman W and Peterson RE (2004) 2,3,7,8-Tetrachlorodibenzo-p- dioxin inhibits regression of the common cardinal vein in developing zebrafish. Toxicol Sci 78(2):258-266.
20. Bennett P, Ramsden DB and Williams AC (1996) Complete structural characterisation of the human aryl hydrocarbon receptor gene. Clin Mol Pathol 49(1):M12-M16.
21. Berg P and Pongratz I (2002) Two parallel pathways mediate cytoplasmic localization of the dioxin (aryl hydrocarbon) receptor. J Biol Chem 277(35):32310-32319.
22. Bollerot K, Angelier N and Coumailleau P (2001) Molecular cloning and embryonic expression of the Xenopus Arnt gene. Mech Dev 108(1-2):227-231.
23. Bunger MK, Walisser JA, Sullivan R, Manley PA, Moran SM, Kalscheur VL, Colman RJ and Bradfield CA (2005) Progressive arthropathy in mice with a targeted disruption of the Mop3/Bmal-1 locus. Genesis 41(3):122-132.
24. Bunger MK, Wilsbacher LD, Moran SM, Clendenin C, Radcliffe LA, Hogenesch JB, Simon MC, Takahashi JS and Bradfield CA (2000) Mop3 is an essential component of the master circadian pacemaker in mammals. Cell 103(7):1009- 1017.
229 25. Burbach KM, Poland A and Bradfield CA (1992) Cloning of the Ah-receptor cDNA reveals a distinctive ligand-activated transcription factor. Proc Natl Acad Sci U S A 89(17):8185-8189.
26. Butler RA, Kelley ML, Powell WH, Hahn ME and Van Beneden RJ (2001) An aryl hydrocarbon receptor (AHR) homologue from the soft-shell clam, Mya arenaria: evidence that invertebrate AHR homologues lack 2,3,7,8- tetrachlorodibenzo-p-dioxin and beta-naphthoflavone binding. Gene 278(1- 2):223-234.
27. Cain SM (1975) Oxygen delivery and utilization in dogs with a sublethal dose of cobalt chloride. J Appl Physiol 38(1):20-25.
28. Card PB, Erbel PJ and Gardner KH (2005) Structural basis of ARNT PAS-B dimerization: use of a common beta-sheet interface for hetero- and homodimerization. J Mol Biol 353(3):664-677.
29. Carney SA, Peterson RE and Heideman W (2004) 2,3,7,8-Tetrachlorodibenzo-p- dioxin activation of the aryl hydrocarbon receptor/aryl hydrocarbon receptor nuclear translocator pathway causes developmental toxicity through a CYP1A- independent mechanism in zebrafish. Mol Pharmacol 66(3):512-521.
30. Cartwright P and Helin K (2000) Nucleocytoplasmic shuttling of transcription factors. Cell Mol Life Sci 57(8-9):1193-1206.
31. Carver LA and Bradfield CA (1997) Ligand-dependent interaction of the aryl hydrocarbon receptor with a novel immunophilin homolog in vivo. J Biol Chem 272(17):11452-11456.
32. Carver LA, Hogenesch JB and Bradfield CA (1994a) Tissue specific expression of the rat Ah-receptor and ARNT mRNAs. Nucleic Acids Res 22(15):3038-3044.
33. Carver LA, Jackiw V and Bradfield CA (1994b) The 90-kDa heat shock protein is essential for Ah receptor signaling in a yeast expression system. J Biol Chem 269(48):30109-30112.
34. Carver LA, LaPres JJ, Jain S, Dunham EE and Bradfield CA (1998) Characterization of the Ah receptor-associated protein, ARA9. J Biol Chem 273(50):33580-33587. 230 35. Cermakian N, Whitmore D, Foulkes NS and Sassone-Corsi P (2000) Asynchronous oscillations of two zebrafish CLOCK partners reveal differential clock control and function. Proc Natl Acad Sci U S A 97(8):4339-4344.
36. Chan WK, Yao G, Gu YZ and Bradfield CA (1999) Cross-talk between the aryl hydrocarbon receptor and hypoxia inducible factor signaling pathways. Demonstration of competition and compensation. J Biol Chem 274(17):12115- 12123.
37. Chang DC, McWatters HG, Williams JA, Gotter AL, Levine JD and Reppert SM (2003) Constructing a feedback loop with circadian clock molecules from the silkmoth, Antheraea pernyi. J Biol Chem 278(40):38149-38158.
38. Chapman-Smith A, Lutwyche JK and Whitelaw ML (2004) Contribution of the Per/Arnt/Sim (PAS) domains to DNA binding by the basic helix-loop-helix PAS transcriptional regulators. J Biol Chem 279(7):5353-5362.
39. Chapman-Smith A and Whitelaw ML (2006) Novel DNA binding by a basic helix-loop-helix protein. The role of the dioxin receptor PAS domain. J Biol Chem 281(18):12535-12545.
40. Chen HS, Singh SS, Perdew GH. (1997) The Ah receptor is a sensitive target of geldanamycin-induced protein turnover. Arch Biochem Biophys. 348(1):190-8.
41. Chen HS and Perdew GH (1994) Subunit composition of the heteromeric cytosolic aryl hydrocarbon receptor complex. J Biol Chem 269(44):27554-27558.
42. Chilov D, Hofer T, Bauer C, Wenger RH and Gassmann M (2001) Hypoxia affects expression of circadian genes PER1 and CLOCK in mouse brain. FASEB J 15(14):2613-2622.
43. Ciechanover A (2005) The ubiquitin proteolytic system. From a vague idea, through basic mechanisms, and onto human diseases and drug targeting. Neurology.
231 44. Cioffi CC, Pollenz RS, Middleton DL, Wilson MR, Miller NW, William Clem L, Warr GW and Ross DA (2002) Oct2 transcription factor of a teleost fish: activation domains and function from an enhancer. Arch Biochem Biophys 404(1):55-61.
45. Corton JC, Moreno ES, Hovis SM, Leonard LS, Gaido KW, Joyce MM and Kennett SB (1996) Identification of a cell-specific transcription activation domain within the human Ah receptor nuclear translocator. Toxicol Appl Pharmacol 139(2):272-280.
46. Cowden KD and Simon MC (2002) The bHLH/PAS factor MOP3 does not participate in hypoxia responses. Biochem Biophys Res Commun 290(4):1228- 1236.
47. Cox MB and Miller CA, 3rd (2003) Pharmacological and genetic analysis of 90- kDa heat shock isoprotein-aryl hydrocarbon receptor complexes. Mol Pharmacol 64(6):1549-1556.
48. Crews ST (1998) Control of cell lineage-specific development and transcription by bHLH-PAS proteins. Genes Dev 12(5):607-620.
49. Czar MJ, Owens-Grillo JK, Dittmar KD, Hutchison KA, Zacharek AM, Leach KL, Deibel MR, Jr. and Pratt WB (1994) Characterization of the protein-protein interactions determining the heat shock protein (hsp90.hsp70.hsp56) heterocomplex. J Biol Chem 269(15):11155-11161.
50. Czech MP (2006) ARNT misbehavin' in diabetic beta cells. Nat Med 12(1):39-40.
51. Dale YR and Eltom SE (2006) Calpain mediates the dioxin-induced activation and down-regulation of the aryl hydrocarbon receptor. Mol Pharmacol 70(5):1481-1487.
52. Davarinos NA and Pollenz RS (1999) Aryl hydrocarbon receptor imported into the nucleus following ligand binding is rapidly degraded via the cytosplasmic proteasome following nuclear export. J Biol Chem 274(40):28708-28715.
53. Denison MS, Fisher JM and Whitlock JP, Jr. (1988) Inducible, receptor- dependent protein-DNA interactions at a dioxin-responsive transcriptional enhancer. Proc Natl Acad Sci U S A 85(8):2528-2532. 232 54. Dickey CA, Yue M, Lin WL, Dickson DW, Dunmore JH, Lee WC, Zehr C, West G, Cao S, Clark AM, Caldwell GA, Caldwell KA, Eckman C, Patterson C, Hutton M and Petrucelli L (2006) Deletion of the ubiquitin ligase CHIP leads to the accumulation, but not the aggregation, of both endogenous phospho- and caspase- 3-cleaved tau species. J Neurosci 26(26):6985-6996.
55. Dittmar KD, Demady DR, Stancato LF, Krishna P and Pratt WB (1997) Folding of the glucocorticoid receptor by the heat shock protein (hsp) 90-based chaperone machinery. The role of p23 is to stabilize receptor.hsp90 heterocomplexes formed by hsp90.p60.hsp70. J Biol Chem 272(34):21213-21220.
56. Dolwick KM, Swanson HI and Bradfield CA (1993) In vitro analysis of Ah receptor domains involved in ligand-activated DNA recognition. Proc Natl Acad Sci U S A 90(18):8566-8570.
57. Dougherty EJ and Pollenz RS (2007) Analysis of AH Receptor-ARNT and AH Receptor-ARNT2 Complexes In Vitro and in Cell Culture. Toxicol Sci. [Epub ahead of print]
58. Drutel G, Kathmann M, Heron A, Schwartz JC and Arrang JM (1996) Cloning and selective expression in brain and kidney of ARNT2 homologous to the Ah receptor nuclear translocator (ARNT). Biochem Biophys Res Commun 225(2):333-339.
59. Dull AB, Carlson DB, Petrulis JR and Perdew GH (2002) Characterization of the phosphorylation status of the hepatitis B virus X-associated protein 2. Arch Biochem Biophys 406(2):209-221.
60. Ebert BL, Firth JD and Ratcliffe PJ (1995) Hypoxia and mitochondrial inhibitors regulate expression of glucose transporter-1 via distinct Cis-acting sequences. J Biol Chem 270(49):29083-29089.
61. Eguchi H, Ikuta T, Tachibana T, Yoneda Y and Kawajiri K (1997) A nuclear localization signal of human aryl hydrocarbon receptor nuclear translocator/hypoxia-inducible factor 1beta is a novel bipartite type recognized by the two components of nuclear pore-targeting complex. J Biol Chem 272(28):17640-17647.
233 62. Ema M, Morita M, Ikawa S, Tanaka M, Matsuda Y, Gotoh O, Saijoh Y, Fujii H, Hamada H, Kikuchi Y and Fujii-Kuriyama Y (1996a) Two new members of the murine Sim gene family are transcriptional repressors and show different expression patterns during mouse embryogenesis. Mol Cell Biol 16(10):5865- 5875.
63. Ema M, Sogawa K, Watanabe N, Chujoh Y, Matsushita N, Gotoh O, Funae Y and Fujii-Kuriyama Y (1992) cDNA cloning and structure of mouse putative Ah receptor. Biochem Biophys Res Commun 184(1):246-253.
64. Ema M, Suzuki M, Morita M, Hirose K, Sogawa K, Matsuda Y, Gotoh O, Saijoh Y, Fujii H, Hamada H and Fujii-Kuriyama Y (1996b) cDNA cloning of a murine homologue of Drosophila single-minded, its mRNA expression in mouse development, and chromosome localization. Biochem Biophys Res Commun 218(2):588-594.
65. Ema M, Taya S, Yokotani N, Sogawa K, Matsuda Y and Fujii-Kuriyama Y (1997) A novel bHLH-PAS factor with close sequence similarity to hypoxia-inducible factor 1alpha regulates the VEGF expression and is potentially involved in lung and vascular development. Proc Natl Acad Sci U S A 94(9):4273-4278.
66. Emmons RB, Duncan D, Estes PA, Kiefel P, Mosher JT, Sonnenfeld M, Ward MP, Duncan I and Crews ST (1999) The spineless-aristapedia and tango bHLH- PAS proteins interact to control antennal and tarsal development in Drosophila. Development 126(17):3937-3945.
67. Erbel PJ, Card PB, Karakuzu O, Bruick RK and Gardner KH (2003) Structural basis for PAS domain heterodimerization in the basic helix--loop--helix-PAS transcription factor hypoxia-inducible factor. Proc Natl Acad Sci U S A 100(26):15504-15509.
68. Evans BR, Karchner SI, Allan LL, Pollenz RS, Tanguay RL, Jenny MJ, Sherr DH and Hahn ME (2007) Repression of aryl hydrocarbon receptor (AHR) signaling by AHR repressor (AHRR): Role of DNA binding and competition for ARNT. Mol Pharmacol.
69. Evans RM (1989) Molecular characterization of the glucocorticoid receptor. Recent Prog Horm Res 45:1-22; discussion 22-27.
234 70. Fernandez-Salguero P, Pineau T, Hilbert DM, McPhail T, Lee SS, Kimura S, Nebert DW, Rudikoff S, Ward JM and Gonzalez FJ (1995) Immune system impairment and hepatic fibrosis in mice lacking the dioxin-binding Ah receptor. Science 268(5211):722-726.
71. Fernandez-Salguero PM, Hilbert DM, Rudikoff S, Ward JM and Gonzalez FJ (1996) Aryl-hydrocarbon receptor-deficient mice are resistant to 2,3,7,8- tetrachlorodibenzo-p-dioxin-induced toxicity. Toxicol Appl Pharmacol 140(1):173-179.
72. Franken, P., Dudley, C.A., Estill, S.J., Barakat, M., Thomason, R., O'Hara, B.F., and McKnight, S.L. (2006). NPAS2 as a transcriptional regulator of non-rapid eye movement sleep: genotype and sex interactions. Proc Natl Acad Sci U S A 103(18), 7118-7123.
73. Freeman BC, Felts SJ, Toft DO and Yamamoto KR (2000) The p23 molecular chaperones act at a late step in intracellular receptor action to differentially affect ligand efficacies. Genes Dev 14(4):422-434.
74. Fujisawa-Sehara A, Sogawa K, Nishi C and Fujii-Kuriyama Y (1986) Regulatory DNA elements localized remotely upstream from the drug-metabolizing cytochrome P-450c gene. Nucleic Acids Res 14(3):1465-1477.
75. Fukunaga BN, Probst MR, Reisz-Porszasz S and Hankinson O (1995) Identification of functional domains of the aryl hydrocarbon receptor. J Biol Chem 270(49):29270-29278.
76. Furness SG, Lees MJ and Whitelaw ML (2007) The dioxin (aryl hydrocarbon) receptor as a model for adaptive responses of bHLH/PAS transcription factors. FEBS Lett 581(19):3616-3625.
77. Galigniana MD, Harrell JM, Murphy PJ, Chinkers M, Radanyi C, Renoir JM, Zhang M and Pratt WB (2002) Binding of hsp90-associated immunophilins to cytoplasmic dynein: direct binding and in vivo evidence that the peptidylprolyl isomerase domain is a dynein interaction domain. Biochemistry 41(46):13602- 13610.
235 78. Galigniana MD, Radanyi C, Renoir JM, Housley PR and Pratt WB (2001) Evidence that the peptidylprolyl isomerase domain of the hsp90-binding immunophilin FKBP52 is involved in both dynein interaction and glucocorticoid receptor movement to the nucleus. J Biol Chem 276(18):14884-14889.
79. Garrett RW and Gasiewicz TA (2006) The aryl hydrocarbon receptor agonist 2,3,7,8-tetrachlorodibenzo-p-dioxin alters the circadian rhythms, quiescence, and expression of clock genes in murine hematopoietic stem and progenitor cells. Mol Pharmacol 69(6):2076-2083.
80. Garrison PM and Denison MS (2000) Analysis of the murine AhR gene promoter. J Biochem Mol Toxicol 14(1):1-10.
81. Gekakis N, Staknis D, Nguyen HB, Davis FC, Wilsbacher LD, King DP, Takahashi JS and Weitz CJ (1998) Role of the CLOCK protein in the mammalian circadian mechanism. Science 280(5369):1564-1569.
82. Georgitsi M, Raitila A, Karhu A, Tuppurainen K, Makinen MJ, Vierimaa O, Paschke R, Saeger W, van der Luijt RB, Sane T, Robledo M, De Menis E, Weil RJ, Wasik A, Zielinski G, Lucewicz O, Lubinski J, Launonen V, Vahteristo P and Aaltonen LA (2007) Molecular diagnosis of pituitary adenoma predisposition caused by aryl hydrocarbon receptor-interacting protein gene mutations. Proc Natl Acad Sci U S A 104(10):4101-4105.
83. Giannone JV, Li W, Probst M and Okey AB (1998) Prolonged depletion of AH receptor without alteration of receptor mRNA levels after treatment of cells in culture with 2,3,7,8-tetrachlorodibenzo-p-dioxin. Biochem Pharmacol 55(4):489- 497.
84. Gielen JE and Nebert DW (1972) Aryl hydrocarbon hydroxylase induction in mammalian liver cell culture. 3. Effects of various sera, hormones, biogenic amines, and other endogenous compounds on the enzyme activity. J Biol Chem 247(23):7591-7602.
85. Gonzalez FJ, Fernandez-Salguero P, Lee SS, Pineau T and Ward JM (1995) Xenobiotic receptor knockout mice. Toxicol Lett 82-83:117-121.
86. Gouedard C, Barouki R and Morel Y (2004) Dietary polyphenols increase paraoxonase 1 gene expression by an aryl hydrocarbon receptor-dependent mechanism. Mol Cell Biol 24(12):5209-5222. 236 87. Gradin K, McGuire J, Wenger RH, Kvietikova I, fhitelaw ML, Toftgard R, Tora L, Gassmann M and Poellinger L (1996) Functional interference between hypoxia and dioxin signal transduction pathways: competition for recruitment of the Arnt transcription factor. Mol Cell Biol 16(10):5221-5231.
88. Gu YZ, Hogenesch JB and Bradfield CA (2000) The PAS superfamily: sensors of environmental and developmental signals. Annu Rev Pharmacol Toxicol 40:519- 561.
89. Guillemot F, Nagy A, Auerbach A, Rossant J and Joyner AL (1994) Essential role of Mash-2 in extraembryonic development. Nature 371(6495):333-336.
90. Gunton JE, Kulkarni RN, Yim S, Okada T, Hawthorne WJ, Tseng YH, Roberson RS, Ricordi C, O'Connell PJ, Gonzalez FJ and Kahn CR (2005) Loss of ARNT/HIF1beta mediates altered gene expression and pancreatic-islet dysfunction in human type 2 diabetes. Cell 122(3):337-349.
91. Gurtner GC, Davis V, Li H, McCoy MJ, Sharpe A and Cybulsky MI (1995) Targeted disruption of the murine VCAM1 gene: essential role of VCAM-1 in chorioallantoic fusion and placentation. Genes Dev 9(1):1-14.
92. Hahn ME, Karchner SI, Franks DG and Merson RR (2004) Aryl hydrocarbon receptor polymorphisms and dioxin resistance in Atlantic killifish (Fundulus heteroclitus). Pharmacogenetics 14(2):131-143.
93. Hankinson GJ, White WJ and Lang CM (1979) The effects of selected antimicrobials on glucose transport in the rat intestine. Lab Anim Sci 29(1):35-39.
94. Hankinson O (1979) Single-step selection of clones of a mouse hepatoma line deficient in aryl hydrocarbon hydroxylase. Proc Natl Acad Sci U S A 76(1):373- 376.
95. Hankinson O (2005) Role of coactivators in transcriptional activation by the aryl hydrocarbon receptor. Arch Biochem Biophys 433(2):379-386.
96. Hapgood J, Cuthill S, Denis M, Poellinger L and Gustafsson JA (1989) Specific protein-DNA interactions at a xenobiotic-responsive element: copurification of dioxin receptor and DNA-binding activity. Proc Natl Acad Sci U S A 86(1):60-64.
237 97. Heid SE, Pollenz RS and Swanson HI (2000) Role of heat shock protein 90 dissociation in mediating agonist-induced activation of the aryl hydrocarbon receptor. Mol Pharmacol 57(1):82-92.
98. Henry EC, Bemis JC, Henry O, Kende AS and Gasiewicz TA (2006) A potential endogenous ligand for the aryl hydrocarbon receptor has potent agonist activity in vitro and in vivo. Arch Biochem Biophys 450(1):67-77.
99. Hestermann EV and Brown M (2003) Agonist and chemopreventative ligands induce differential transcriptional cofactor recruitment by aryl hydrocarbon receptor. Mol Cell Biol 23(21):7920-7925.
100. Hickey E, Brandon SE, Smale G, Lloyd D and Weber LA (1989) Sequence and regulation of a gene encoding a human 89-kilodalton heat shock protein. Mol Cell Biol 9(6):2615-2626.
101. Higa LA and Zhang H (2007) Stealing the spotlight: CUL4-DDB1 ubiquitin ligase docks WD40-repeat proteins to destroy. Cell Div 2:5.
102. Hirose K, Morita M, Ema M, Mimura J, Hamada H, Fujii H, Saijo Y, Gotoh O, Sogawa K and Fujii-Kuriyama Y (1996) cDNA cloning and tissue-specific expression of a novel basic helix-loop-helix/PAS factor (Arnt2) with close sequence similarity to the aryl hydrocarbon receptor nuclear translocator (Arnt). Mol Cell Biol 16(4):1706-1713.
103. Hoffman EC, Reyes H, Chu FF, Sander F, Conley LH, Brooks BA and Hankinson O (1991) Cloning of a factor required for activity of the Ah (dioxin) receptor. Science 252(5008):954-958.
104. Hogenesch JB, Chan WK, Jackiw VH, Brown RC, Gu YZ, Pray-Grant M, Perdew GH and Bradfield CA (1997) Characterization of a subset of the basic-helix-loop- helix-PAS superfamily that interacts with components of the dioxin signaling pathway. J Biol Chem 272(13):8581-8593.
105. Hogenesch JB, Gu YZ, Moran SM, Shimomura K, Radcliffe LA, Takahashi JS and Bradfield CA (2000) The basic helix-loop-helix-PAS protein MOP9 is a brain-specific heterodimeric partner of circadian and hypoxia factors. J Neurosci 20(13):RC83.
238 106. Hollingshead BD, Patel RD and Perdew GH (2006) Endogenous hepatic expression of the hepatitis B virus X-associated protein 2 is adequate for maximal association with aryl hydrocarbon receptor-90-kDa heat shock protein complexes. Mol Pharmacol 70(6):2096-2107.
107. Hollingshead BD, Petrulis JR and Perdew GH (2004) The aryl hydrocarbon (Ah) receptor transcriptional regulator hepatitis B virus X-associated protein 2 antagonizes p23 binding to Ah receptor-Hsp90 complexes and is dispensable for receptor function. J Biol Chem 279(44):45652-45661.
108. Holmes JL and Pollenz RS (1997) Determination of aryl hydrocarbon receptor nuclear translocator protein concentration and subcellular localization in hepatic and nonhepatic cell culture lines: development of quantitative Western blotting protocols for calculation of aryl hydrocarbon receptor and aryl hydrocarbon receptor nuclear translocator protein in total cell lysates. Mol Pharmacol 52(2):202-211.
109. Hosoya T, Oda Y, Takahashi S, Morita M, Kawauchi S, Ema M, Yamamoto M and Fujii-Kuriyama Y (2001) Defective development of secretory neurones in the hypothalamus of Arnt2-knockout mice. Genes Cells 6(4):361-374.
110. Hosoya, T., Harada, N., Mimura, J., Motohashi, H., Takahashi, S., Nakajima, O., Morita, M., Kawauchi, S., Yamamoto, M., and Fujii-Kuriyama, Y. (2008). Inducibility of cytochrome P450 1A1 and chemical carcinogenesis by benzo[a]pyrene in AhR repressor-deficient mice. Biochem Biophys Res Commun 365 (3), 562-567.
111. Huang ZJ, Edery I and Rosbash M (1993) PAS is a dimerization domain common to Drosophila period and several transcription factors. Nature 364(6434):259-262.
112. Huffman JL, Mokashi A, Bachinger HP and Brennan RG (2001) The basic helix- loop-helix domain of the aryl hydrocarbon receptor nuclear transporter (ARNT) can oligomerize and bind E-box DNA specifically. J Biol Chem 276(44):40537- 40544.
113. Hutchison KA, Dittmar KD and Pratt WB (1994) All of the factors required for assembly of the glucocorticoid receptor into a functional heterocomplex with heat shock protein 90 are preassociated in a self-sufficient protein folding structure, a "foldosome". J Biol Chem 269(45):27894-27899.
239 114. Ikeda M and Nomura M (1997) cDNA cloning and tissue-specific expression of a novel basic helix-loop-helix/PAS protein (BMAL1) and identification of alternatively spliced variants with alternative translation initiation site usage. Biochem Biophys Res Commun 233(1):258-264.
115. Ikeda M, Yu W, Hirai M, Ebisawa T, Honma S, Yoshimura K, Honma KI and Nomura M (2000) cDNA cloning of a novel bHLH-PAS transcription factor superfamily gene, BMAL2: its mRNA expression, subcellular distribution, and chromosomal localization. Biochem Biophys Res Commun 275(2):493-502.
116. Ishikawa T, Hirayama J, Kobayashi Y and Todo T (2002) Zebrafish CRY represses transcription mediated by CLOCK-BMAL heterodimer without inhibiting its binding to DNA. Genes Cells 7(10):1073-1086.
117. Iyer NV, Leung SW and Semenza GL (1998) The human hypoxia-inducible factor 1alpha gene: HIF1A structure and evolutionary conservation. Genomics 52(2):159-165.
118. Jain S, Dolwick KM, Schmidt JV and Bradfield CA (1994) Potent transactivation domains of the Ah receptor and the Ah receptor nuclear translocator map to their carboxyl termini. J Biol Chem 269(50):31518-31524.
119. Jain S, Maltepe E, Lu MM, Simon C and Bradfield CA (1998) Expression of ARNT, ARNT2, HIF1 alpha, HIF2 alpha and Ah receptor mRNAs in the developing mouse. Mech Dev 73(1):117-123.
120. Jensen BA and Hahn ME (2001) cDNA cloning and characterization of a high affinity aryl hydrocarbon receptor in a cetacean, the beluga, Delphinapterus leucas. Toxicol Sci 64(1):41-56.
121. Jinno A, Maruyama Y, Ishizuka M, Kazusaka A, Nakamura A and Fujita S (2006) Induction of cytochrome P450-1A by the equine estrogen equilenin, a new endogenous aryl hydrocarbon receptor ligand. J Steroid Biochem Mol Biol 98(1):48-55.
122. Karchner SI, Powell WH and Hahn ME (1999) Identification and functional characterization of two highly divergent aryl hydrocarbon receptors (AHR1 and AHR2) in the teleost Fundulus heteroclitus. Evidence for a novel subfamily of ligand-binding basic helix loop helix-Per-ARNT-Sim (bHLH-PAS) factors. J Biol Chem 274(47):33814-33824. 240 123. Karpova TS, Kim MJ, Spriet C, Nalley K, Stasevich TJ, Kherrouche Z, Heliot L and McNally JG (2008) Concurrent fast and slow cycling of a transcriptional activator at an endogenous promoter. Science 319(5862):466-469.
124. Kazlauskas A (2000) A new member of an old family. Nat Cell Biol 2(5):E78-79.
125. Kazlauskas A, Poellinger L and Pongratz I (2000) The immunophilin-like protein XAP2 regulates ubiquitination and subcellular localization of the dioxin receptor. J Biol Chem 275(52):41317-41324.
126. Kazlauskas A, Poellinger L and Pongratz I (2002) Two distinct regions of the immunophilin-like protein XAP2 regulate dioxin receptor function and interaction with hsp90. J Biol Chem 277(14):11795-11801.
127. Kazlauskas A, Sundstrom S, Poellinger L and Pongratz I (2001) The hsp90 chaperone complex regulates intracellular localization of the dioxin receptor. Mol Cell Biol 21(7):2594-2607.
128. Keith B, Adelman DM and Simon MC (2001) Targeted mutation of the murine arylhydrocarbon receptor nuclear translocator 2 (Arnt2) gene reveals partial redundancy with Arnt. Proc Natl Acad Sci U S A 98(12):6692-6697.
129. Kewley RJ, Whitelaw ML and Chapman-Smith A (2004) The mammalian basic helix-loop-helix/PAS family of transcriptional regulators. Int J Biochem Cell Biol 36(2):189-204.
130. Kikuchi Y, Ohsawa S, Mimura J, Ema M, Takasaki C, Sogawa K and Fujii- Kuriyama Y (2003) Heterodimers of bHLH-PAS protein fragments derived from AhR, AhRR, and Arnt prepared by co-expression in Escherichia coli: characterization of their DNA binding activity and preparation of a DNA complex. J Biochem (Tokyo) 134(1):83-90.
131. Kim EY and Hahn ME (2002) cDNA cloning and characterization of an aryl hydrocarbon receptor from the harbor seal (Phoca vitulina): a biomarker of dioxin susceptibility? Aquat Toxicol 58(1-2):57-73.
241 132. Kim EY, Iwata H, Suda T, Tanabe S, Amano M, Miyazaki N and Petrov EA (2005) Aryl hydrocarbon receptor (AHR) and AHR nuclear translocator (ARNT) expression in Baikal seal (Pusa sibirica) and association with 2,3,7,8-TCDD toxic equivalents and CYP1 expression levels. Comp Biochem Physiol C Toxicol Pharmacol 141(3):281-291.
133. Ko HP, Okino ST, Ma Q and Whitlock JP, Jr. (1996) Dioxin-induced CYP1A1 transcription in vivo: the aromatic hydrocarbon receptor mediates transactivation, enhancer-promoter communication, and changes in chromatin structure. Mol Cell Biol 16(1):430-436.
134. Ko HP, Okino ST, Ma Q and Whitlock JP, Jr. (1997) Transactivation domains facilitate promoter occupancy for the dioxin-inducible CYP1A1 gene in vivo. Mol Cell Biol 17(7):3497-3507.
135. Kohsaka A, Laposky AD, Ramsey KM, Estrada C, Joshu C, Kobayashi Y, Turek FW and Bass J (2007) High-fat diet disrupts behavioral and molecular circadian rhythms in mice. Cell Metab 6(5):414-421.
136. Korkalainen M, Tuomisto J and Pohjanvirta R (2001) The AH receptor of the most dioxin-sensitive species, guinea pig, is highly homologous to the human AH receptor. Biochem Biophys Res Commun 285(5):1121-1129.
137. Korkalainen M, Tuomisto J and Pohjanvirta R (2003) Identification of novel splice variants of ARNT and ARNT2 in the rat. Biochem Biophys Res Commun 303(4):1095-1100.
138. Kozak KR, Abbott B and Hankinson O (1997) ARNT-deficient mice and placental differentiation. Dev Biol 191(2):297-305.
139. Kremer J, Gleichmann E and Esser C (1994) Thymic stroma exposed to arylhydrocarbon receptor-binding xenobiotics fails to support proliferation of early thymocytes but induces differentiation. J Immunol 153(6):2778-2786.
140. Kuwako, K., Taniura, H., and Yoshikawa, K. (2004). Necdin-related MAGE proteins differentially interact with the E2F1 transcription factor and the p75 neurotrophin receptor. J Biol Chem 279(3), 1703-1712.
242 141. Kwon I, Lee J, Chang SH, Jung NC, Lee BJ, Son GH, Kim K and Lee KH (2006) BMAL1 shuttling controls transactivation and degradation of the CLOCK/BMAL1 heterodimer. Mol Cell Biol 26(19):7318-7330.
142. LaPres JJ, Glover E, Dunham EE, Bunger MK and Bradfield CA (2000) ARA9 modifies agonist signaling through an increase in cytosolic aryl hydrocarbon receptor. J Biol Chem 275(9):6153-6159.
143. Lee JS, Kim EY, Iwata H and Tanabe S (2007) Molecular characterization and tissue distribution of aryl hydrocarbon receptor nuclear translocator isoforms, ARNT1 and ARNT2, and identification of novel splice variants in common cormorant (Phalacrocorax carbo). Comp Biochem Physiol C Toxicol Pharmacol 145(3):379-393.
144. Lee K, Burgoon LD, Lamb L, Dere E, Zacharewski TR, Hogenesch JB and LaPres JJ (2006) Identification and characterization of genes susceptible to transcriptional cross-talk between the hypoxia and dioxin signaling cascades. Chem Res Toxicol 19(10):1284-1293.
145. Lees MJ, Peet DJ and Whitelaw ML (2003) Defining the role for XAP2 in stabilization of the dioxin receptor. J Biol Chem 278(38):35878-35888.
146. Legraverend C, Hannah RR, Eisen HJ, Owens IS, Nebert DW and Hankinson O (1982) Regulatory gene product of the Ah locus. Characterization of receptor mutants among mouse hepatoma clones. J Biol Chem 257(11):6402-6407.
147. Levesque, B.M., Zhou, S., Shan, L., Johnston, P., Kong, Y., Degan, S., and Sunday, M.E. (2007). NPAS1 regulates branching morphogenesis in embryonic lung. Am J Respir Cell Mol Biol 36(4), 427-434.
148. Levine SL and Perdew GH (2001) Aryl hydrocarbon receptor (AhR)/AhR nuclear translocator (ARNT) activity is unaltered by phosphorylation of a periodicity/ARNT/single-minded (PAS)-region serine residue. Mol Pharmacol 59(3):557-566.
149. Levine SL and Perdew GH (2002) Okadaic acid increases ARNT homodimer transactivation potential. Cell Biol Toxicol 18(2):109-120.
243 150. Levisetti MG and Polonsky KS (2005) Diabetic pancreatic beta cells ARNT all they should be. Cell Metab 2(2):78-80.
151. Li H, Dong L and Whitlock JP, Jr. (1994) Transcriptional activation function of the mouse Ah receptor nuclear translocator. J Biol Chem 269(45):28098-28105.
152. Lin BC, Sullivan R, Lee Y, Moran S, Glover E and Bradfield CA (2007) Deletion of the aryl hydrocarbon receptor associated protein 9 leads to cardiac malformation and embryonic lethality. J Biol Chem.
153. Lindebro MC, Poellinger L and Whitelaw ML (1995) Protein-protein interaction via PAS domains: role of the PAS domain in positive and negative regulation of the bHLH/PAS dioxin receptor-Arnt transcription factor complex. EMBO J 14(14):3528-3539.
154. Lusska A, Shen E and Whitlock JP, Jr. (1993) Protein-DNA interactions at a dioxin-responsive enhancer. Analysis of six bona fide DNA-binding sites for the liganded Ah receptor. J Biol Chem 268(9):6575-6580.
155. Ma Q (2001) Induction of CYP1A1. The AhR/DRE paradigm: transcription, receptor regulation, and expanding biological roles. Curr Drug Metab 2(2):149- 164.
156. Ma Q (2007) Aryl hydrocarbon receptor degradation-promoting factor (ADPF) and the control of the xenobiotic response. Mol Interv 7(3):133-137.
157. Ma Q and Baldwin KT (2000) 2,3,7,8-tetrachlorodibenzo-p-dioxin-induced degradation of aryl hydrocarbon receptor (AhR) by the ubiquitin-proteasome pathway. Role of the transcription activaton and DNA binding of AhR. J Biol Chem 275(12):8432-8438.
158. Ma Q, Renzelli AJ, Baldwin KT and Antonini JM (2000) Superinduction of CYP1A1 gene expression. Regulation of 2,3,7, 8-tetrachlorodibenzo-p-dioxin- induced degradation of Ah receptor by cycloheximide. J Biol Chem 275(17):12676-12683.
244 159. Ma Q and Whitlock JP, Jr. (1997) A novel cytoplasmic protein that interacts with the Ah receptor, contains tetratricopeptide repeat motifs, and augments the transcriptional response to 2,3,7,8-tetrachlorodibenzo-p-dioxin. J Biol Chem 272(14):8878-8884.
160. Maemura K, de la Monte SM, Chin MT, Layne MD, Hsieh CM, Yet SF, Perrella MA and Lee ME (2000) CLIF, a novel cycle-like factor, regulates the circadian oscillation of plasminogen activator inhibitor-1 gene expression. J Biol Chem 275(47):36847-36851.
161. Maemura K, Takeda N and Nagai R (2007) Circadian rhythms in the CNS and peripheral clock disorders: role of the biological clock in cardiovascular diseases. J Pharmacol Sci 103(2):134-138.
162. Maltepe E, Keith B, Arsham AM, Brorson JR and Simon MC (2000) The role of ARNT2 in tumor angiogenesis and the neural response to hypoxia. Biochem Biophys Res Commun 273(1):231-238.
163. Maltepe E, Schmidt JV, Baunoch D, Bradfield CA and Simon MC (1997) Abnormal angiogenesis and responses to glucose and oxygen deprivation in mice lacking the protein ARNT. Nature 386(6623):403-407.
164. Maruyama, K., Usami, M., Aizawa, T., and Yoshikawa, K. (1991). A novel brain- specific mRNA encoding nuclear protein (necdin) expressed in neurally differentiated embryonal carcinoma cells. Biochem Biophys Res Commun 178(1), 291-296.
165. Maxwell PH, Dachs GU, Gleadle JM, Nicholls LG, Harris AL, Stratford IJ, Hankinson O, Pugh CW and Ratcliffe PJ (1997) Hypoxia-inducible factor-1 modulates gene expression in solid tumors and influences both angiogenesis and tumor growth. Proc Natl Acad Sci U S A 94(15):8104-8109.
166. McDonnell WM, Chensue SW, Askari FK and Moseley RH (1996) Hepatic fibrosis in Ahr-/- mice. Science 271(5246):223-224.
167. Meyer BK and Perdew GH (1999) Characterization of the AhR-hsp90-XAP2 core complex and the role of the immunophilin-related protein XAP2 in AhR stabilization. Biochemistry 38(28):8907-8917.
245 168. Meyer BK, Petrulis JR and Perdew GH (2000) Aryl hydrocarbon (Ah) receptor levels are selectively modulated by hsp90-associated immunophilin homolog XAP2. Cell Stress Chaperones 5(3):243-254.
169. Meyer BK, Pray-Grant MG, Vanden Heuvel JP and Perdew GH (1998) Hepatitis B virus X-associated protein 2 is a subunit of the unliganded aryl hydrocarbon receptor core complex and exhibits transcriptional enhancer activity. Mol Cell Biol 18(2):978-988.
170. Meyer JN, Wassenberg DM, Karchner SI, Hahn ME and Di Giulio RT (2003a) Expression and inducibility of aryl hydrocarbon receptor pathway genes in wild- caught killifish (Fundulus heteroclitus) with different contaminant-exposure histories. Environ Toxicol Chem 22(10):2337-2343.
171. Meyer P, Prodromou C, Hu B, Vaughan C, Roe SM, Panaretou B, Piper PW and Pearl LH (2003b) Structural and functional analysis of the middle segment of hsp90: implications for ATP hydrolysis and client protein and cochaperone interactions. Mol Cell 11(3):647-658.
172. Michaud J and Fan CM (1997) Single-minded--two genes, three chromosomes. Genome Res 7(6):569-571.
173. Michaud JL, DeRossi C, May NR, Holdener BC and Fan CM (2000) ARNT2 acts as the dimerization partner of SIM1 for the development of the hypothalamus. Mech Dev 90(2):253-261.
174. Mimura J, Yamashita K, Nakamura K, Morita M, Takagi TN, Nakao K, Ema M, Sogawa K, Yasuda M, Katsuki M and Fujii-Kuriyama Y (1997) Loss of teratogenic response to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in mice lacking the Ah (dioxin) receptor. Genes Cells 2(10):645-654.
175. Moffett P and Pelletier J (2000) Different transcriptional properties of mSim-1 and mSim-2. FEBS Lett 466(1):80-86.
176. Moffett P, Reece M and Pelletier J (1997) The murine Sim-2 gene product inhibits transcription by active repression and functional interference. Mol Cell Biol 17(9):4933-4947.
246 177. Moore SK, Kozak C, Robinson EA, Ullrich SJ and Appella E (1989) Murine 86- and 84-kDa heat shock proteins, cDNA sequences, chromosome assignments, and evolutionary origins. J Biol Chem 264(10):5343-5351.
178. Morales JL and Perdew GH (2007) Carboxyl terminus of hsc70-interacting protein (CHIP) can remodel mature aryl hydrocarbon receptor (AhR) complexes and mediate ubiquitination of both the AhR and the 90 kDa heat-shock protein (hsp90) in vitro. Biochemistry 46(2):610-621.
179. Murphy BA, Vick MM, Sessions DR, Cook RF and Fitzgerald BP (2006) Evidence of an oscillating peripheral clock in an equine fibroblast cell line and adipose tissue but not in peripheral blood. J Comp Physiol A Neuroethol Sens Neural Behav Physiol 192(7):743-751.
180. Myers CL, Weiss SJ, Kirsh MM and Shlafer M (1985) Involvement of hydrogen peroxide and hydroxyl radical in the 'oxygen paradox': reduction of creatine kinase release by catalase, allopurinol or deferoxamine, but not by superoxide dismutase. J Mol Cell Cardiol 17(7):675-684.
181. Nambu JR, Chen W, Hu S and Crews ST (1996) The Drosophila melanogaster similar bHLH-PAS gene encodes a protein related to human hypoxia-inducible factor 1 alpha and Drosophila single-minded. Gene 172(2):249-254.
182. Nambu JR, Lewis JO, Wharton KA, Jr. and Crews ST (1991) The Drosophila single-minded gene encodes a helix-loop-helix protein that acts as a master regulator of CNS midline development. Cell 67(6):1157-1167.
183. Nebert DW, Puga A and Vasiliou V (1993) Role of the Ah receptor and the dioxin-inducible [Ah] gene battery in toxicity, cancer, and signal transduction. Ann N Y Acad Sci 685:624-640.
184. Necela B and Pollenz RS (1999) Functional analysis of activation and repression domains of the rainbow trout aryl hydrocarbon receptor nuclear translocator (rtARNT) protein isoforms. Biochem Pharmacol 57(10):1177-1190.
185. Necela B and Pollenz RS (2001) Identification of a novel Carboxy-terminal domain involved in the negative function of the rainbow trout Ah receptor nuclear translocator protein isoform a (rtARNTa) in Ah receptor-mediated signaling. Biochem Pharmacol 62(3):307-318.
247 186. Nene V, Wortman JR, Lawson D, Haas B, Kodira C, Tu ZJ, Loftus B, Xi Z, Megy K, Grabherr M, Ren Q, Zdobnov EM, Lobo NF, Campbell KS, Brown SE, Bonaldo MF, Zhu J, Sinkins SP, Hogenkamp DG, Amedeo P, Arensburger P, Atkinson PW, Bidwell S, Biedler J, Birney E, Bruggner RV, Costas J, Coy MR, Crabtree J, Crawford M, Debruyn B, Decaprio D, Eiglmeier K, Eisenstadt E, El- Dorry H, Gelbart WM, Gomes SL, Hammond M, Hannick LI, Hogan JR, Holmes MH, Jaffe D, Johnston JS, Kennedy RC, Koo H, Kravitz S, Kriventseva EV, Kulp D, Labutti K, Lee E, Li S, Lovin DD, Mao C, Mauceli E, Menck CF, Miller JR, Montgomery P, Mori A, Nascimento AL, Naveira HF, Nusbaum C, O'Leary S, Orvis J, Pertea M, Quesneville H, Reidenbach KR, Rogers YH, Roth CW, Schneider JR, Schatz M, Shumway M, Stanke M, Stinson EO, Tubio JM, Vanzee JP, Verjovski-Almeida S, Werner D, White O, Wyder S, Zeng Q, Zhao Q, Zhao Y, Hill CA, Raikhel AS, Soares MB, Knudson DL, Lee NH, Galagan J, Salzberg SL, Paulsen IT, Dimopoulos G, Collins FH, Birren B, Fraser-Liggett CM and Severson DW (2007) Genome sequence of Aedes aegypti, a major arbovirus vector. Science 316(5832):1718-1723.
187. Nielsen OJ, Schuster SJ, Kaufman R, Erslev AJ and Caro J (1987) Regulation of erythropoietin production in a human hepatoblastoma cell line. Blood 70(6):1904- 1909.
188. Nishimura N, Yonemoto J, Miyabara Y, Fujii-Kuriyama Y and Tohyama C (2005) Altered thyroxin and retinoid metabolic response to 2,3,7,8-tetrachlorodibenzo-p- dioxin in aryl hydrocarbon receptor-null mice. Arch Toxicol 79(5):260-267.
189. Oberg M, Bergander L, Hakansson H, Rannug U and Rannug A (2005) Identification of the tryptophan photoproduct 6-formylindolo[3,2-b]carbazole, in cell culture medium, as a factor that controls the background aryl hydrocarbon receptor activity. Toxicol Sci 85(2):935-943.
190. Ogiso H, Kagi N, Matsumoto E, Nishimoto M, Arai R, Shirouzu M, Mimura J, Fujii-Kuriyama Y and Yokoyama S (2004) Phosphorylation analysis of 90 kDa heat shock protein within the cytosolic arylhydrocarbon receptor complex. Biochemistry 43(49):15510-15519.
191. Ohsawa, S., Hamada, S., Kakinuma, Y., Yagi, T., and Miura, M. (2005). Novel function of neuronal PAS domain protein 1 in erythropoietin expression in neuronal cells. J Neurosci Res 79(4), 451-458.
248 192. Ohtake F, Baba A, Takada I, Okada M, Iwasaki K, Miki H, Takahashi S, Kouzmenko A, Nohara K, Chiba T, Fujii-Kuriyama Y and Kato S (2007) Dioxin receptor is a ligand-dependent E3 ubiquitin ligase. Nature 446(7135):562-566.
193. Okano T, Sasaki M and Fukada Y (2001) Cloning of mouse BMAL2 and its daily expression profile in the suprachiasmatic nucleus: a remarkable acceleration of Bmal2 sequence divergence after Bmal gene duplication. Neurosci Lett 300(2):111-114.
194. Okey AB, Bondy GP, Mason ME, Nebert DW, Forster-Gibson CJ, Muncan J and Dufresne MJ (1980) Temperature-dependent cytosol-to-nucleus translocation of the Ah receptor for 2,3,7,8-tetrachlorodibenzo-p-dioxin in continuous cell culture lines. J Biol Chem 255(23):11415-11422.
195. Okey AB, Riddick DS and Harper PA (1994) The Ah receptor: mediator of the toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and related compounds. Toxicol Lett 70(1):1-22.
196. Pandini A, Denison MS, Song Y, Soshilov AA and Bonati L (2007) Structural and functional characterization of the aryl hydrocarbon receptor ligand binding domain by homology modeling and mutational analysis. Biochemistry 46(3):696- 708.
197. Paul AL and Ferl RJ (1991) In vivo footprinting reveals unique cis-elements and different modes of hypoxic induction in maize Adh1 and Adh2. Plant Cell 3(2):159-168.
198. Peattie DA, Harding MW, Fleming MA, DeCenzo MT, Lippke JA, Livingston DJ and Benasutti M (1992) Expression and characterization of human FKBP52, an immunophilin that associates with the 90-kDa heat shock protein and is a component of steroid receptor complexes. Proc Natl Acad Sci U S A 89(22):10974-10978.
199. Petrulis JR, Hord NG and Perdew GH (2000) Subcellular localization of the aryl hydrocarbon receptor is modulated by the immunophilin homolog hepatitis B virus X-associated protein 2. J Biol Chem 275(48):37448-37453.
249 200. Petrulis JR, Kusnadi A, Ramadoss P, Hollingshead B and Perdew GH (2003) The hsp90 Co-chaperone XAP2 alters importin beta recognition of the bipartite nuclear localization signal of the Ah receptor and represses transcriptional activity. J Biol Chem 278(4):2677-2685.
201. Petrulis JR and Perdew GH (2002) The role of chaperone proteins in the aryl hydrocarbon receptor core complex. Chem Biol Interact 141(1-2):25-40.
202. Phelan D, Winter GM, Rogers WJ, Lam JC and Denison MS (1998a) Activation of the Ah receptor signal transduction pathway by bilirubin and biliverdin. Arch Biochem Biophys 357(1):155-163.
203. Phelan DM, Brackney WR and Denison MS (1998b) The Ah receptor can bind ligand in the absence of receptor-associated heat-shock protein 90. Arch Biochem Biophys 353(1):47-54.
204. Poellinger L and Johnson RS (2004) HIF-1 and hypoxic response: the plot thickens. Curr Opin Genet Dev 14(1):81-85.
205. Poland A and Glover E (1974) Comparison of 2,3,7,8-tetrachlorodibenzo-p- dioxin, a potent inducer of aryl hydrocarbon hydroxylase, with 3- methylcholanthrene. Mol Pharmacol 10(2):349-359.
206. Poland A and Glover E (1988) Ca2+-dependent proteolysis of the Ah receptor. Arch Biochem Biophys 261(1):103-111.
207. Poland A, Glover E and Kende AS (1976) Stereospecific, high affinity binding of 2,3,7,8-tetrachlorodibenzo-p-dioxin by hepatic cytosol. Evidence that the binding species is receptor for induction of aryl hydrocarbon hydroxylase. J Biol Chem 251(16):4936-4946.
208. Poland A and Knutson JC (1982) 2,3,7,8-tetrachlorodibenzo-p-dioxin and related halogenated aromatic hydrocarbons: examination of the mechanism of toxicity. Annu Rev Pharmacol Toxicol 22:517-554.
209. Poland A, Palen D and Glover E (1994) Analysis of the four alleles of the murine aryl hydrocarbon receptor. Mol Pharmacol 46(5):915-921.
250 210. Pollenz RS (1996) The aryl-hydrocarbon receptor, but not the aryl-hydrocarbon receptor nuclear translocator protein, is rapidly depleted in hepatic and nonhepatic culture cells exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Mol Pharmacol 49(3):391-398.
211. Pollenz RS (2002) The mechanism of AH receptor protein down-regulation (degradation) and its impact on AH receptor-mediated gene regulation. Chem Biol Interact 141(1-2):41-61.
212. Pollenz RS (2007) Specific blockage of ligand-induced degradation of the Ah receptor by proteasome but not calpain inhibitors in cell culture lines from different species. Biochem Pharmacol 74(1):131-143.
213. Pollenz RS and Barbour ER (2000) Analysis of the complex relationship between nuclear export and aryl hydrocarbon receptor-mediated gene regulation. Mol Cell Biol 20(16):6095-6104.
214. Pollenz RS, Davarinos NA and Shearer TP (1999) Analysis of aryl hydrocarbon receptor-mediated signaling during physiological hypoxia reveals lack of competition for the aryl hydrocarbon nuclear translocator transcription factor. Mol Pharmacol 56(6):1127-1137.
215. Pollenz RS and Dougherty EJ (2005) Redefining the role of the endogenous XAP2 and Carboxy-terminal hsp70-interacting protein on the endogenous Ah receptors expressed in mouse and rat cell lines. J Biol Chem 280(39):33346- 33356.
216. Pollenz RS, Necela B and Marks-Sojka K (2002) Analysis of rainbow trout Ah receptor protein isoforms in cell culture reveals conservation of function in Ah receptor-mediated signal transduction. Biochem Pharmacol 64(1):49-60.
217. Pollenz RS, Popat J and Dougherty EJ (2005) Role of the carboxy-terminal transactivation domain and active transcription in the ligand-induced and ligand- independent degradation of the mouse Ahb-1 receptor. Biochem Pharmacol 70(11):1623-1633.
218. Pollenz RS, Santostefano MJ, Klett E, Richardson VM, Necela B and Birnbaum LS (1998) Female Sprague-Dawley rats exposed to a single oral dose of 2,3,7,8- tetrachlorodibenzo-p-dioxin exhibit sustained depletion of aryl hydrocarbon receptor protein in liver, spleen, thymus, and lung. Toxicol Sci 42(2):117-128. 251 219. Pollenz RS, Sattler CA and Poland A (1994) The aryl hydrocarbon receptor and aryl hydrocarbon receptor nuclear translocator protein show distinct subcellular localizations in Hepa 1c1c7 cells by immunofluorescence microscopy. Mol Pharmacol 45(3):428-438.
220. Pollenz RS, Sullivan HR, Holmes J, Necela B and Peterson RE (1996) Isolation and expression of cDNAs from rainbow trout (Oncorhynchus mykiss) that encode two novel basic helix-loop-Helix/PER-ARNT-SIM (bHLH/PAS) proteins with distinct functions in the presence of the aryl hydrocarbon receptor. Evidence for alternative mRNA splicing and dominant negative activity in the bHLH/PAS family. J Biol Chem 271(48):30886-30896.
221. Pollenz RS, Wilson SE and Dougherty EJ (2006) Role of endogenous XAP2 protein on the localization and nucleocytoplasmic shuttling of the endogenous mouse Ahb-1 receptor in the presence and absence of ligand. Mol Pharmacol 70(4):1369-1379.
222. Pongratz I, Antonsson C, Whitelaw ML and Poellinger L (1998) Role of the PAS domain in regulation of dimerization and DNA binding specificity of the dioxin receptor. Mol Cell Biol 18(7):4079-4088.
223. Powell-Coffman JA, Bradfield CA and Wood WB (1998) Caenorhabditis elegans orthologs of the aryl hydrocarbon receptor and its heterodimerization partner the aryl hydrocarbon receptor nuclear translocator. Proc Natl Acad Sci U S A 95(6):2844-2849.
224. Powell WH, Bright R, Bello SM and Hahn ME (2000) Developmental and tissue- specific expression of AHR1, AHR2, and ARNT2 in dioxin-sensitive and - resistant populations of the marine fish Fundulus heteroclitus. Toxicol Sci 57(2):229-239.
225. Powell WH, Karchner SI, Bright R and Hahn ME (1999) Functional diversity of vertebrate ARNT proteins: identification of ARNT2 as the predominant form of ARNT in the marine teleost, Fundulus heteroclitus. Arch Biochem Biophys 361(1):156-163.
226. Prasch AL, Andreasen EA, Peterson RE and Heideman W (2004a) Interactions between 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and hypoxia signaling pathways in zebrafish: hypoxia decreases responses to TCDD in zebrafish embryos. Toxicol Sci 78(1):68-77.
252 227. Prasch AL, Heideman W and Peterson RE (2004b) ARNT2 is not required for TCDD developmental toxicity in zebrafish. Toxicol Sci 82(1):250-258.
228. Prasch AL, Tanguay RL, Mehta V, Heideman W and Peterson RE (2006) Identification of zebrafish ARNT1 homologs: 2,3,7,8-tetrachlorodibenzo-p-dioxin toxicity in the developing zebrafish requires ARNT1. Mol Pharmacol 69(3):776- 787.
229. Prasch AL, Teraoka H, Carney SA, Dong W, Hiraga T, Stegeman JJ, Heideman W and Peterson RE (2003) Aryl hydrocarbon receptor 2 mediates 2,3,7,8- tetrachlorodibenzo-p-dioxin developmental toxicity in zebrafish. Toxicol Sci 76(1):138-150.
230. Probst MR, Fan CM, Tessier-Lavigne M and Hankinson O (1997) Two murine homologs of the Drosophila single-minded protein that interact with the mouse aryl hydrocarbon receptor nuclear translocator protein. J Biol Chem 272(7):4451- 4457.
231. Puig O, Caspary F, Rigaut G, Rutz B, Bouveret E, Bragado-Nilsson E, Wilm M, and Seraphin B (2001) The tandem affinity purification (TAP) method: a general procedure of protein complex purification. Methods. 24:218-229.
232. Qu X, Metz RP, Porter WW, Cassone VM and Earnest DJ (2007) Disruption of Clock Gene Expression Alters Responses of the Ahr Signaling Pathway in the Mouse Mammary Gland. Mol Pharmacol.
233. Ramadoss P, Marcus C and Perdew GH (2005) Role of the aryl hydrocarbon receptor in drug metabolism. Expert Opin Drug Metab Toxicol 1(1):9-21.
234. Ramadoss P and Perdew GH (2005) The transactivation domain of the Ah receptor is a key determinant of cellular localization and ligand-independent nucleocytoplasmic shuttling properties. Biochemistry 44(33):11148-11159.
235. Reick M, Robertson RW, Pasco DS and Fagan JB (1994) Down-regulation of nuclear aryl hydrocarbon receptor DNA-binding and transactivation functions: requirement for a labile or inducible factor. Mol Cell Biol 14(9):5653-5660.
253 236. Reid G, Hubner MR, Metivier R, Brand H, Denger S, Manu D, Beaudouin J, Ellenberg J and Gannon F (2003) Cyclic, proteasome-mediated turnover of unliganded and liganded ERalpha on responsive promoters is an integral feature of estrogen signaling. Mol Cell 11(3):695-707.
237. Reick, M., Garcia, J.A., Dudley, C., and McKnight, S.L. (2001). NPAS2: an analog of clock operative in the mammalian forebrain. Science 293(5529), 506- 509.
238. Reisz-Porszasz S, Probst MR, Fukunaga BN and Hankinson O (1994) Identification of functional domains of the aryl hydrocarbon receptor nuclear translocator protein (ARNT). Mol Cell Biol 14(9):6075-6086.
239. 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(5060):1193-1195.
240. Richardson VM, Santostefano MJ and Birnbaum LS (1998) Daily cycle of bHLH- PAS proteins, Ah receptor and Arnt, in multiple tissues of female Sprague- Dawley rats. Biochem Biophys Res Commun 252(1):225-231.
241. Richter CA, Tillitt DE and Hannink M (2001) Regulation of subcellular localization of the aryl hydrocarbon receptor (AhR). Arch Biochem Biophys 389(2):207-217.
242. Roberts BJ, Whitelaw ML. (1999) Degradation of the basic helix-loop-helix/Per- ARNT-Sim homology domain dioxin receptor via the ubiquitin/proteasome pathway. J Biol Chem. 274(51):36351-6.
243. Roman BL and Peterson RE (1998) In utero and lactational exposure of the male rat to 2,3,7,8-tetrachlorodibenzo-p-dioxin impairs prostate development. 1. Effects on gene expression. Toxicol Appl Pharmacol 150(2):240-253.
244. Roman BL, Pollenz RS and Peterson RE (1998) Responsiveness of the adult male rat reproductive tract to 2,3,7,8-tetrachlorodibenzo-p-dioxin exposure: Ah receptor and ARNT expression, CYP1A1 induction, and Ah receptor down- regulation. Toxicol Appl Pharmacol 150(2):228-239.
254 245. Rowatt AJ, DePowell JJ and Powell WH (2003) ARNT gene multiplicity in amphibians: characterization of ARNT2 from the frog Xenopus laevis. J Exp Zoolog B Mol Dev Evol 300(1):48-57.
246. Roy NK and Wirgin I (1997) Characterization of the aromatic hydrocarbon receptor gene and its expression in Atlantic tomcod. Arch Biochem Biophys 344(2):373-386.
247. Rutter, J., Reick, M., Wu, L.C., and McKnight, S.L. (2001). Regulation of clock and NPAS2 DNA binding by the redox state of NAD cofactors. Science 293 (5529), 510-514.
248. Ryan HE, Lo J and Johnson RS (1998) HIF-1 alpha is required for solid tumor formation and embryonic vascularization. EMBO J 17(11):3005-3015.
249. Sablistsky, F (2005) Protein motifs: the helix-loop-helix motif. In: Encyclopedia of Life Sciences (Online). John Wiley & Sons, Chichester, UK, pp. (5 pages)
250. Sadek CM, Jalaguier S, Feeney EP, Aitola M, Damdimopoulos AE, Pelto-Huikko M and Gustafsson JA (2000) Isolation and characterization of AINT: a novel ARNT interacting protein expressed during murine embryonic development. Mech Dev 97(1-2):13-26.
251. Safe S (2001) Molecular biology of the Ah receptor and its role in carcinogenesis. Toxicol Lett 120(1-3):1-7.
252. Safe S, Wang F, Porter W, Duan R and McDougal A (1998) Ah receptor agonists as endocrine disruptors: antiestrogenic activity and mechanisms. Toxicol Lett 102- 103:343-347.
253. Safe S and Wormke M (2003) Inhibitory aryl hydrocarbon receptor-estrogen receptor alpha cross-talk and mechanisms of action. Chem Res Toxicol 16(7):807- 816.
254. Salghetti SE, Caudy AA, Chenoweth JG and Tansey WP (2001) Regulation of transcriptional activation domain function by ubiquitin. Science 293(5535):1651- 1653.
255 255. Sangoram AM, Saez L, Antoch MP, Gekakis N, Staknis D, Whiteley A, Fruechte EM, Vitaterna MH, Shimomura K, King DP, Young MW, Weitz CJ and Takahashi JS (1998) Mammalian circadian autoregulatory loop: a timeless ortholog and mPer1 interact and negatively regulate CLOCK-BMAL1-induced transcription. Neuron 21(5):1101-1113.
256. Satoh N, Satou Y, Davidson B and Levine M (2003) Ciona intestinalis: an emerging model for whole-genome analyses. Trends Genet 19(7):376-381.
257. Schmidt JV, Su GH, Reddy JK, Simon MC and Bradfield CA (1996) Characterization of a murine Ahr null allele: involvement of the Ah receptor in hepatic growth and development. Proc Natl Acad Sci U S A 93(13):6731-6736.
258. Schoenhard JA, Eren M, Johnson CH and Vaughan DE (2002) Alternative splicing yields novel BMAL2 variants: tissue distribution and functional characterization. Am J Physiol Cell Physiol 283(1):C103-114.
259. Seidel SD, Winters GM, Rogers WJ, Ziccardi MH, Li V, Keser B and Denison MS (2001) Activation of the Ah receptor signaling pathway by prostaglandins. J Biochem Mol Toxicol 15(4):187-196.
260. Sekine H, Mimura J, Yamamoto M and Fujii-Kuriyama Y (2006) Unique and overlapping transcriptional roles of arylhydrocarbon receptor nuclear translocator (Arnt) and Arnt2 in xenobiotic and hypoxic responses. J Biol Chem 281(49):37507-37516.
261. Semenza GL (2007) Hypoxia-inducible factor 1 (HIF-1) pathway. Sci STKE 2007(407):cm8.
262. Semenza GL, Agani F, Booth G, Forsythe J, Iyer N, Jiang BH, Leung S, Roe R, Wiener C and Yu A (1997) Structural and functional analysis of hypoxia- inducible factor 1. Kidney Int 51(2):553-555.
263. Semenza GL, Roth PH, Fang HM and Wang GL (1994) Transcriptional regulation of genes encoding glycolytic enzymes by hypoxia-inducible factor 1. J Biol Chem 269(38):23757-23763.
256 264. Shen ES and Whitlock JP, Jr. (1992) Protein-DNA interactions at a dioxin- responsive enhancer. Mutational analysis of the DNA-binding site for the liganded Ah receptor. J Biol Chem 267(10):6815-6819.
265. Shetty PV, Bhagwat BY and Chan WK (2003) P23 enhances the formation of the aryl hydrocarbon receptor-DNA complex. Biochem Pharmacol 65(6):941-948.
266. Shetty PV, Wang X and Chan WK (2004) CyP40, but not Hsp70, in rabbit reticulocyte lysate causes the aryl hydrocarbon receptor-DNA complex formation. Arch Biochem Biophys 429(1):42-49.
267. Shou W, Aghdasi B, Armstrong DL, Guo Q, Bao S, Charng MJ, Mathews LM, Schneider MD, Hamilton SL and Matzuk MM (1998) Cardiac defects and altered ryanodine receptor function in mice lacking FKBP12. Nature 391(6666):489-492.
268. Shue G and Kohtz DS (1994) Structural and functional aspects of basic helix- loop-helix protein folding by heat-shock protein 90. J Biol Chem 269(4):2707- 2711.
269. Smith TP, Grosse WM, Freking BA, Roberts AJ, Stone RT, Casas E, Wray JE, White J, Cho J, Fahrenkrug SC, Bennett GL, Heaton MP, Laegreid WW, Rohrer GA, Chitko-McKown CG, Pertea G, Holt I, Karamycheva S, Liang F, Quackenbush J and Keele JW (2001) Sequence evaluation of four pooled-tissue normalized bovine cDNA libraries and construction of a gene index for cattle. Genome Res 11(4):626-630.
270. Sogawa K, Nakano R, Kobayashi A, Kikuchi Y, Ohe N, Matsushita N and Fujii- Kuriyama Y (1995) Possible function of Ah receptor nuclear translocator (Arnt) homodimer in transcriptional regulation. Proc Natl Acad Sci U S A 92(6):1936- 1940.
271. Sojka KM, Kern CB and Pollenz RS (2000) Expression and subcellular localization of the aryl hydrocarbon receptor nuclear translocator (ARNT) protein in mouse and chicken over developmental time. Anat Rec 260(4):327-334.
272. Son DS and Rozman KK (2002) 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) induces plasminogen activator inhibitor-1 through an aryl hydrocarbon receptor- mediated pathway in mouse hepatoma cell lines. Arch Toxicol 76(7):404-413.
257 273. Song J, Clagett-Dame M, Peterson RE, Hahn ME, Westler WM, Sicinski RR and DeLuca HF (2002) A ligand for the aryl hydrocarbon receptor isolated from lung. Proc Natl Acad Sci U S A 99(23):14694-14699.
274. Song Z and Pollenz RS (2002) Ligand-dependent and independent modulation of aryl hydrocarbon receptor localization, degradation, and gene regulation. Mol Pharmacol 62(4):806-816.
275. Song Z and Pollenz RS (2003) Functional analysis of murine aryl hydrocarbon (AH) receptors defective in nuclear import: impact on AH receptor degradation and gene regulation. Mol Pharmacol 63(3):597-606.
276. Sonnenfeld M, Ward M, Nystrom G, Mosher J, Stahl S and Crews S (1997) The Drosophila tango gene encodes a bHLH-PAS protein that is orthologous to mammalian Arnt and controls CNS midline and tracheal development. Development 124(22):4571-4582.
277. Sun Y, Yang Z, Niu Z, Peng J, Li Q, Xiong W, Langnas AN, Ma MY and Zhao Y (2006a) MOP3, a component of the molecular clock, regulates the development of B cells. Immunology 119(4):451-460.
278. Sun Y, Yang Z, Niu Z, Wang W, Peng J, Li Q, Ma MY and Zhao Y (2006b) The mortality of MOP3 deficient mice with a systemic functional failure. J Biomed Sci 13(6):845-851.
279. Sutter TR, Tang YM, Hayes CL, Wo YY, Jabs EW, Li X, Yin H, Cody CW and Greenlee WF (1994) Complete cDNA sequence of a human dioxin-inducible mRNA identifies a new gene subfamily of cytochrome P450 that maps to chromosome 2. J Biol Chem 269(18):13092-13099.
280. Swanson HI, Chan WK and Bradfield CA (1995) DNA binding specificities and pairing rules of the Ah receptor, ARNT, and SIM proteins. J Biol Chem 270(44):26292-26302.
281. Swanson HI and Yang JH (1999) Specificity of DNA binding of the c-Myc/Max and ARNT/ARNT dimers at the CACGTG recognition site. Nucleic Acids Res 27(15):3205-3212.
258 282. Takahashi Y, Nakayama K, Shimojima T, Itoh S and Kamataki T (1996) Expression of aryl hydrocarbon receptor (AhR) and aryl hydrocarbon receptor nuclear translocator (Arnt) in adult rabbits known to be non-responsive to cytochrome P-450 1A1 (CYP1A1) inducers. Eur J Biochem 242(3):512-518.
283. Takahata S, Sogawa K, Kobayashi A, Ema M, Mimura J, Ozaki N and Fujii- Kuriyama Y (1998) Transcriptionally active heterodimer formation of an Arnt- like PAS protein, Arnt3, with HIF-1a, HLF, and clock. Biochem Biophys Res Commun 248(3):789-794.
284. Tanguay RL, Abnet CC, Heideman W and Peterson RE (1999) Cloning and characterization of the zebrafish (Danio rerio) aryl hydrocarbon receptor. Biochim Biophys Acta 1444(1):35-48.
285. Tanguay RL, Andreasen E, Heideman W and Peterson RE (2000) Identification and expression of alternatively spliced aryl hydrocarbon nuclear translocator 2 (ARNT2) cDNAs from zebrafish with distinct functions. Biochim Biophys Acta 1494(1-2):117-128.
286. Teh, C.H., Lam, K.K., Loh, C.C., Loo, J.M., Yan, T., and Lim, T.M. (2006). Neuronal PAS domain protein 1 is a transcriptional repressor and requires arylhydrocarbon nuclear translocator for its nuclear localization. J Biol Chem 281(45), 34617-34629.
287. Teraoka H, Dong W, Tsujimoto Y, Iwasa H, Endoh D, Ueno N, Stegeman JJ, Peterson RE and Hiraga T (2003) Induction of cytochrome P450 1A is required for circulation failure and edema by 2,3,7,8-tetrachlorodibenzo-p-dioxin in zebrafish. Biochem Biophys Res Commun 304(2):223-228.
288. Thackaberry EA, Bedrick EJ, Goens MB, Danielson L, Lund AK, Gabaldon D, Smith SM and Walker MK (2003) Insulin regulation in AhR-null mice: embryonic cardiac enlargement, neonatal macrosomia, and altered insulin regulation and response in pregnant and aging AhR-null females. Toxicol Sci 76(2):407-417.
289. Thomas JB, Crews ST and Goodman CS (1988) Molecular genetics of the single- minded locus: a gene involved in the development of the Drosophila nervous system. Cell 52(1):133-141.
259 290. Tokishita S, Kimura S, Mandokora Y, Kato K, Shiga Y, Takahashi Y, Ohta T and Yamagata H (2006) Tissue-specific expression of a bHLH-PAS protein homologous to ARNT during the development of crustacean Daphnia magna. Gene 376(2):231-239.
291. Verma S, Ismail A, Gao X, Fu G, Li X, O'Malley BW and Nawaz Z (2004) The ubiquitin-conjugating enzyme UBCH7 acts as a coactivator for steroid hormone receptors. Mol Cell Biol 24(19):8716-8726.
292. Vierimaa O, Georgitsi M, Lehtonen R, Vahteristo P, Kokko A, Raitila A, Tuppurainen K, Ebeling TM, Salmela PI, Paschke R, Gundogdu S, De Menis E, Makinen MJ, Launonen V, Karhu A and Aaltonen LA (2006) Pituitary adenoma predisposition caused by germline mutations in the AIP gene. Science 312(5777):1228-1230.
293. Walisser JA, Bunger MK, Glover E and Bradfield CA (2004) Gestational exposure of Ahr and Arnt hypomorphs to dioxin rescues vascular development. Proc Natl Acad Sci U S A 101(47):16677-16682.
294. Walker MK, Heid SE, Smith SM and Swanson HI (2000) Molecular characterization and developmental expression of the aryl hydrocarbon receptor from the chick embryo. Comp Biochem Physiol C Toxicol Pharmacol 126(3):305- 319.
295. Walker MK, Pollenz RS and Smith SM (1997) Expression of the aryl hydrocarbon receptor (AhR) and AhR nuclear translocator during chick cardiogenesis is consistent with 2,3,7,8-tetrachlorodibenzo-p-dioxin-induced heart defects. Toxicol Appl Pharmacol 143(2):407-419.
296. Wang F, Gao JX, Mimura J, Kobayashi A, Sogawa K and Fujii-Kuriyama Y (1998) Structure and expression of the mouse AhR nuclear translocator (mArnt) gene. J Biol Chem 273(38):24867-24873.
297. Wang F, Shi S, Zhang R and Hankinson O (2006) Identifying target genes of the aryl hydrocarbon receptor nuclear translocator (Arnt) using DNA microarray analysis. Biol Chem 387(9):1215-1218.
298. Wang GL and Semenza GL (1993) Desferrioxamine induces erythropoietin gene expression and hypoxia-inducible factor 1 DNA-binding activity: implications for models of hypoxia signal transduction. Blood 82(12):3610-3615. 260 299. Wang S, Ge K, Roeder RG and Hankinson O (2004) Role of mediator in transcriptional activation by the aryl hydrocarbon receptor. J Biol Chem 279(14):13593-13600.
300. Wentworth JN, Buzzeo R and Pollenz RS (2004) Functional characterization of aryl hydrocarbon receptor (zfAHR2) localization and degradation in zebrafish (Danio rerio). Biochem Pharmacol 67(7):1363-1372.
301. Wharton KA, Jr. and Crews ST (1993) CNS midline enhancers of the Drosophila slit and Toll genes. Mech Dev 40(3):141-154.
302. Wharton KA, Jr., Franks RG, Kasai Y and Crews ST (1994) Control of CNS midline transcription by asymmetric E-box-like elements: similarity to xenobiotic responsive regulation. Development 120(12):3563-3569.
303. Whitelaw M, Pongratz I, Wilhelmsson A, Gustafsson JA and Poellinger L (1993) Ligand-dependent recruitment of the Arnt coregulator determines DNA recognition by the dioxin receptor. Mol Cell Biol 13(4):2504-2514.
304. Whitelaw ML, Gustafsson JA and Poellinger L (1994) Identification of transactivation and repression functions of the dioxin receptor and its basic helix- loop-helix/PAS partner factor Arnt: inducible versus constitutive modes of regulation. Mol Cell Biol 14(12):8343-8355.
305. Whitelaw ML, McGuire J, Picard D, Gustafsson JA and Poellinger L (1995) Heat shock protein hsp90 regulates dioxin receptor function in vivo. Proc Natl Acad Sci U S A 92(10):4437-4441.
306. Whitlock JP, Jr. (1999) Induction of cytochrome P4501A1. Annu Rev Pharmacol Toxicol 39:103-125.
307. Whitlock JP, Jr. and Galeazzi DR (1984) 2,3,7,8-Tetrachlorodibenzo-p-dioxin receptors in wild type and variant mouse hepatoma cells. Nuclear location and strength of nuclear binding. J Biol Chem 259(2):980-985.
308. Wilson CL, Thomsen J, Hoivik DJ, Wormke MT, Stanker L, Holtzapple C and Safe SH (1997) Aryl hydrocarbon (Ah) nonresponsiveness in estrogen receptor- negative MDA-MB-231 cells is associated with expression of a variant arnt protein. Arch Biochem Biophys 346(1):65-73. 261 309. Wines ME, Tiffany AM and Holdener BC (1998) Physical localization of the mouse aryl hydrocarbon receptor nuclear translocator-2 (Arnt2) gene within the c112K deletion. Genomics 51(2):223-232.
310. Wochnik GM, Young JC, Schmidt U, Holsboer F, Hartl FU and Rein T (2004) Inhibition of GR-mediated transcription by p23 requires interaction with Hsp90. FEBS Lett 560(1-3):35-38.
311. Wolting CD and McGlade CJ (1998) Cloning and chromosomal localization of a new member of the bHLH/PAS transcription factor family. Mamm Genome 9(6):463-468.
312. Wood SM, Gleadle JM, Pugh CW, Hankinson O and Ratcliffe PJ (1996) The role of the aryl hydrocarbon receptor nuclear translocator (ARNT) in hypoxic induction of gene expression. Studies in ARNT-deficient cells. J Biol Chem 271(25):15117-15123.
313. Woods SL and Whitelaw ML (2002) Differential activities of murine single minded 1 (SIM1) and SIM2 on a hypoxic response element. Cross-talk between basic helix-loop-helix/per-Arnt-Sim homology transcription factors. J Biol Chem 277(12):10236-10243.
314. Yasui T, Kim EY, Iwata H and Tanabe S (2004) Identification of aryl hydrocarbon receptor 2 in aquatic birds; cDNA cloning of AHR1 and AHR2 and characteristics of their amino acid sequences. Mar Environ Res 58(2-5):113-118.
315. Yu W, Ikeda M, Abe H, Honma S, Ebisawa T, Yamauchi T, Honma K and Nomura M (1999) Characterization of three splice variants and genomic organization of the mouse BMAL1 gene. Biochem Biophys Res Commun 260(3):760-767.
316. Zelzer E, Wappner P and Shilo BZ (1997) The PAS domain confers target gene specificity of Drosophila bHLH/PAS proteins. Genes Dev 11(16):2079-2089.
317. Zeruth G and Pollenz RS (2007) Functional analysis of cis-regulatory regions within the dioxin-inducible CYP1A promoter/enhancer region from zebrafish (Danio rerio). Chem Biol Interact.
262 318. Zudaire, E., Cuesta, N., Murty, V., Woodson, K., Adams, L., Gonzalez, N., Martinez, A., Narayan, G., Kirsch, I., Franklin, W., et al. (2008). The aryl hydrocarbon receptor repressor is a putative tumor suppressor gene in multiple human cancers. J Clin Invest 118(2), 640-650.
263
Bibliography
1. Alberts, B (2003) Essential Cell Biology. Garland Science.
2. Boelsterli, UA (2003) Mechanistic Toxicology: The molecular basis of how chemicals disrupt biological targets. New York: Taylor & Francis, Inc.
3. Hicks, BW (2002) Green Fluorescent Protein: Applications and Protocols. Totowa, New Jersey: Humana Press.
4. Lewis, DFV (2001) Guide to Cytochromes P450: Structure and Function. New York: Taylor & Francis, Inc.
5. Nicholl, DST (2002). Genetic Engineering. United Kingdom: Cambridge University Press.
264
Appendices
265 Appendix A: Endogenous AHRb-2 data
The following figures (A-1 through A-7) were generated by Dr. Richard S.
Pollenz and come from the publications: Pollenz RS, Wilson SE, Dougherty EJ (2006)
Role of endogenous XAP2 protein on the localization and nucleocytoplasmic shuttling of the endogenous mouse Ahb-1 receptor in the presence and absence of ligand. Mol
Pharmacol. 70:1369-79 and Pollenz RS, Dougherty EJ (2005) Redefining the role of the endogenous XAP2 and Carboxy-terminal hsp70-interacting protein on the endogenous
Ah receptors expressed in mouse and rat cell lines. J Biol Chem. 280:33346-56. These figures are included since they will aid the reader in interpreting the studies performed in
Chapter Two, particularly in the comparison between endogenous Ahb-2 receptor function and that of Ahb-2 receptor in the Ahb-1 genetic background described in Chapter Two.
266 Appendix A: (Continued)
Figure A-1: Analysis of AHR and XAP2 expression and association in Hepa-1, A7, and C2C12 cells. A, equal amounts of total cell lysates from the indicated cells were resolved by SDS-PAGE, blotted, and stained with A-1A rabbit IgG (1.0 µg/ml), -actin rabbit IgG (1:1000), hsp90 rabbit IgG (1:500), XAP2 mouse IgG1 (1:750), or p23 mouse IgG (1:1000). Reactivity was visualized by ECL with GAR-HRP or GAM-HRP IgG (1:10,000). B, cytosol was prepared from Hepa-1, A7, or C2C12 cells as detailed in Chapter Six. 800 µg of cytosol was precipitated with either affinity-pure A1-A IgG (5 µg) or affinity-pure preimmune rabbit IgG (5 µg) along with Protein A/G-agarose (25 µl) for 2.5 h at 4 °C with rocking. Pellets were washed three times for 5 min each with TTBS supplemented with sodium molybdate (20 mM) and then boiled in 30 µl of SDS sample buffer. 15 µg of cytosol (input) or 15 µl of the eluted protein were resolved by SDS- PAGE, blotted, and stained with A-1A IgG (1.0 µg/ml) or XAP2 mouse IgG1 (1:750). Reactivity was visualized by ECL with GAR-HRP or GAM-HRP IgG (1:10,000). He, Hepa-1; C2, C2C12; Ah, precipitated with A1-A IgG; Pi, precipitated with preimmune IgG. The numbers under the XAP2 blot represent the percentage of XAP in relation to the level in Hepa-1 (100%). The precipitated IgG band is shown to demonstrate the uniformity of the precipitation across all samples.
267 Appendix A: (Continued)
Figure A-2: Subcellular localization of AHR in Hepa-1, A7, C2C12, and 10T1/2 cells exposed to TCDD or LMB. Cells were grown on glass coverslips exposed to the compounds detailed below and then fixed as detailed previously (Holmes and Pollenz, 1997; Pollenz, 1996; Pollenz et al., 1994). Coverslips were incubated with A-1 IgG (1.0 µg/ml) and visualized with GAR-Rhodamine IgG (1:400). A–D, Hepa-1 cells exposed to methanol (0.5%) for 4 h (A), TCDD (2 nM) for 1 h (B), LMB (20 nM) for 2 h (C),orLMBfor 4 h (D). E–H, C2C12 cells exposed to methanol (0.5%) for 4 h (E), TCDD (2 nM) for 1 h (F), LMB (20 nM) for 2 h (G), or LMB for 4 h (H). Panels I–L, A7 cells exposed to methanol (0.5%) for 4 h (I), TCDD (2 nM) for 1 h (J), LMB (20 nM) for 2 h (K), LMB for 4 h (L). M–P, 10T1/2 cells exposed to methanol (0.5%) for 4 h (M), TCDD (2 nM) for 1 h (N), LMB (20 nM) for 2 h (O),or LMB for 4 h (P). Each set of four panels was exposed for identical times. Bar (A), 10 µm.
268 Appendix A: (Continued)
Figure A-3: Reduction of endogenous XAP2 by siRNA knockdown in Hepa-1 cells. A, Hepa-1 cells were transfected with siRNA specific to XAP2 or control siRNA as detailed in Chapter Six. Forty-eight hours later, cytosol was prepared, and 600 µg was precipitated with either AHR (Ah-IgG) or preimmune IgG (Pi-IgG) as detailed in Chapter Six. Each of the precipitated samples as well as 15 µg of cytosol (input) was resolved by SDS-PAGE and blotted. Blots were stained with either 1.0 µg/ml A-1A IgG or XAP2 mouse IgG1 (1:750), and reactivity was visualized by ECL with GAR-HRP or GAM-HRP IgG (1:10,000). The IgG bands are presented to show the consistency of the precipitations. C, cells transfected with control siRNA; X, cells transfected with XAP2 siRNA. B, computer densitometry was used to determine the relative level of AHR or XAP2 protein present in the precipitated samples presented on the blot in A. Each column represents the relative densitometry units of an individual band and shows that the ratio of XAP2/AHR has changed in the siXAP samples. Ah, level of AHR protein; mX, level of XAP2 protein.
269 Appendix A: (Continued)
Figure A-4: Subcellular localization of AHR in cells with reduced levels of XAP2. Hepa-1 cells were transfected with siRNA specific to XAP2 or control siRNA along with FITC-labeled RNA (BLOCK-IT) as detailed in Chapter Six. Forty-eight hours later, cells were treated with either 0.1% methanol or 1 nM LMB for 4 h and either fixed for immunofluorescence microscopy or harvested for the preparation of total cell lysates. A, Western blot of AHR and XAP2 expression in total cell lysates prepared from cells transfected with siCON or XAP2 siRNA (siXAP). B, cells were visualized for AHR. Fixed cells were stained with 1.0 µg/ml A-1 IgG and visualized with GAR-RHO IgG (1:400). The FITC-labeled panels represent the exact fields presented to the left and illustrate the transfection efficiency of the experiment. Scale bar, 10 µm. 270 Appendix A: (Continued)
Figure A-5: Association of endogenous XAP2 with AHR in Hepa-1 cells transfected with hXAP2 expression vectors. Hepa-1 cells were transfected with pCI-hXAP2 or control vector pcDNA3.1 as detailed in Chapter Six. After 24 h, populations of cells were harvested, and cytosol was generated for immunoprecipitation experiments. A, 600 µg of cytosol from the indicated samples was precipitated in duplicate with either AHR (AHR- IgG) or preimmune IgG (Pi-IgG) as detailed in Chapter Six. Each of the precipitated samples as well as 15 µg of cytosol (input) was resolved by SDS-PAGE and blotted. Blots were stained with either 1.0 µg/ml A-1A IgG or XAP2 mouse IgG1 (1:750), and reactivity was visualized by ECL with GAR-HRP or GAM-HRP IgG (1:10,000). mXAP2, endogenous mouse XAP2; hXAP2, exogenous human XAP2; and AHR (AHR). The IgG bands are presented to show the consistency of the precipitations. C, samples from cells transfected with pcDNA3.1; X, samples from cells transfected with pCI-hXAP2. B, computer densitometry was used to determine the relative level of AHR or XAP2 protein present in the precipitated samples presented on the blot in A. Each column represents the relative densitometry units of an individual band and can be used to evaluate differences in the ratio of XAP2/AHR in the different samples. However, because of the difference in the sensitivity of each antibody for its target protein, the ratio does not represent the absolute number of protein molecules. Note that the XAP2/AHR ratio is essentially unchanged in all the samples. The precipitated IgG band is shown to demonstrate the uniformity of the precipitation across all samples.
271 Appendix A: (Continued)
Figure A-6: Localization of endogenous Ahb-1 receptor in Hepa-1 cells expressing hXAP2. Hepa-1 cells were transfected with pCI-hXAP2 along with FITC-labeled RNA (BLOCK-IT), pEYFP-XAP2-FLAG, or control vector pCDNA3.1 as detailed in Chapter Six. After 24 h, populations of cells were either harvested for the generation of total cell lysates or fixed for immunocytochemical staining. A, equal amounts of the indicated samples were resolved by SDS-PAGE and blotted. Blots were stained with either 1.0 µg/ml A-1A IgG, XAP2 mouse IgG1 (1:750), or anti-actin IgG (1:1000), and reactivity was visualized by ECL with GAR-HRP or GAM-HRP IgG (1:10,000). Con, cells transfected with pcDNA3.1; +XAP, cells transfected with pCI-hXAP2 and BLOCK-IT; +Y-XAP, cells transfected with pEYFP-hXAP2-FLAG. B, exact Hepa-1 populations that were transfected as detailed in A were fixed and stained for the AHR with 1.0 µg/ml A-1 IgG and visualized with GAR-RHO IgG (1:400). Con AHR, AHR staining pattern in cells transfected with pcDNA3.1; AHR + hXAP2, AHR staining pattern in cells transfected with pCI-hXAP2 and BLOCK-IT; FITC, FITC labeling of the exact field presented to the left to illustrate the transfection efficiency of the experiment; YFP-XAP2, pattern of the expressed EYFP-XAP2-FLAG; AHR, AHR staining pattern of the exact field presented to the left. The arrows indicate cells that are not expressing the EYFP- XAP2-FLAG protein. Scale bar, 10 µm. 272 Appendix A: (Continued)
Figure A-7: Western blot analysis of AHR in Hepa-1, A7, and C2C12 cells exposed to TCDD. A, Hepa-1, A7, and C2C12 cells were exposed to Me2SO (0.1%) for 2 h or TCDD (2 nM) for the indicated times. Equal amounts of total cell lysates were then resolved by SDS-PAGE, blotted, and stained with A-1A IgG (1.0 µg/ml) and -actin IgG (1:1000). Reactivity was visualized by ECL with GAR-HRP IgG (1:10,000), and bands were quantified and normalized as detailed (Holmes and Pollenz, 1997; Pollenz, 1996; Pollenz et al., 1994). B, the level of AHR protein at each time point was divided by the corresponding level of actin, and the average ± S.E. of the three independent samples was plotted as a function of the time 0 control. The error bars on some of the later time points are within the symbol used to show the data point. *, statistically different from the Hepa-1 273 Appendix B: Hypoxia qRT-PCR Primer sets
The following lists qRT-PCR primer sets (forward and reverse) and antibodies
that have been used in analyzing hypoxic response in several cell lines. These data were
complied from several publications, whose references are listed above each data set, and
are further subdivided by the species and cell line being examined. Amplicon sizes for
PCR products are given where known. Target genes are abbreviated as follows: HIF-1α, hypoxia-inducible factor-1 α; VEGF, vascular endothelial growth factor; NMDA-R1, N- methyl D-aspartate receptor subtype 1; eNOS, endothelial nitric oxide synthase; iNOS, inducible NOS; nNOS, neuronal NOS; ANG, angiopoietin, GLUT1, glucose transporter 1;
PGK, phosphoglycerate kinase; GluR2/3, glutamate receptor subunit; GFAP, glial fibrillary acidic protein; GAPDH, glyceraldehydes 3-phosphate dehydrogenase.
274 Appendix B: (Continued)
HUMAN (ARPE-19 cells)
Matsuda, S, Gomi, F, Katayama, T, Koyama, Y, Tohyama, M, and Yasuo Tano (2006). Induction of Connective Tissue Growth Factor in Retinal Pigment Epithelium Cells by Oxidative Stress. Jpn J Ophthalmol 50:229–234
CTGF Forward 5’-CGGCTTACCGACTGGAAGAC Reverse 5’-CGTCGGTACATACTCCACAG
VEGF 5’-CAGCGCAGCTACTGCCATCCAATCGAGA 5’-GCTTGTCACATCTGCAAGTACGTTCGTTTA human or mouse β-actin 5′-TCCTCCCTGGAGAAGAGCTA 5′-TCCTGCTTGCTGATCCACAT
275 Appendix B: (Continued)
RAT (corpus collosum from brains of rats in hypoxic environment)
Kaur C, Sivakumar V, Ang LS and Sundaresan A (2006). Hypoxic damage to the periventricular white matter in neonatal brain: role of vascular endothelial growth factor, nitric oxide and excitotoxicity. J Neurochem. 98:1200-16.
Primer Amplicon size
HIF-1α Forward: 5’-TCAAGTCAGCAACGTGGAAG Reverse: 5’-TATCGAGGCTGTGTCGACTG 198 bp
VEGF 5’-AGAAAGCCCAATGAAGTGGTG 5’-ACTCCAGGGCTTCATCATTG 177 bp
NMDA-R1 5’-AACCTGCAGAACCGCAAG 5’-GCTTGATGAGCAGGTCTATGC 333 bp eNOS 5’-TGGCAGCCCTAAGACCTATG 5’-AGTCCGAAAATGTCCTCGTG 243 bp iNOS 5’-CCTTGTTCAGCTACGCCTTC 5’-GGTATGCCCGAGTTCTTTCA 179 bp nNOS 5’-AACCTGCAGAACCGCAAG 5’-GCTTGATGAGCAGGTCTATGC 617 bp
β-actin 5’-TCATGAAGTGTGACGTTGACATCCGT 5’-CCTAGAAGCATTTGCGGTGCAGGATG 285 bp
276 Appendix B: (Continued)
RAT (retinas)
Kaur C, Sivakumar V, and Foulds WS (2006). Early response of neurons and glial cells to hypoxia in the retina. Invest Ophthalmol Vis Sci. 47:1126-41
Primer Amplicon Size
HIF-1α Forward: 5’-TCAAGTCAGCAACGTGGAAG Reverse: 5’-TATCGAGGCTGTGTCGACTG 198
VEGF 5’-AGAAAGCCCAATGAAGTGGTG 5’-ACTCCAGGGCTTCATCATTG 177
NMDA-R1 5’-AACCTGCAGAACCGCAAG 5’-GCTTGATGAGCAGGTCTATGC 333
GluR2 5’-CTATTTCCAAGGGGCGCTGAT 5’-CAGTCCAGGATTACACGCCG 539
GluR3 5’-CGCAGAGCCATCTGTGTTTA 5’-GTTGCCACACCATAGCCTTT 180
eNOS 5’-TGGCAGCCCTAAGACCTATG 5’-AGTCCGAAAATGTCCTCGTG 243
iNOS 5’-CCTTGTTCAGCTACGCCTTC 5’-GGTATGCCCGAGTTCTTTCA 179
nNOS 5’-AACCTGCAGAACCGCAAG 5’-GCTTGATGAGCAGGTCTATGC 617
β -Actin (96% identity with mouse) 5’-TCATGAAGTGTGACGTTGACATCCGT 5’-CCTAGAAGCATTTGCGGTGCAGGATG 285
277 Appendix B: (Continued)
MOUSE
Shao G, Gao C, and Lu G (2005). Alterations of hypoxia-inducible factor-1 alpha in the hippocampus of mice acutely and repeatedly exposed to hypoxia. Neurosignals 14:255– 261.
GAPDH primers: Forward 5’-CCCTTCATTGACCTCAAC-3 Reverse: 5’-TTCACACCCATCACAAAC-3
HIF-1α primers: 5’-TATAAACCTGGCAATGTCTCC-3 5’-GATGCCTTAGCAGTGGTCGT-3
278 Appendix B: (Continued)
MOUSE (Used C57Bl mice)
Kociok N, Krohne TU, Poulaki V, and Joussen AM (2006). Geldanamycin treatment reduces neovascularization in a mouse model of retinopathy of prematurity. Graefe’s Arch Clin Exp Ophthalmol. 245:258-66.
VEGF: Forward 5′-CAG CTA TTGCCG TCC GAT TGA GA-3′ and Reverse 5′-TGC TGG CTT TGG TGA GGT TTG AT-3′
ANG1: 5′-CTG ATG GAC TGG GAA GGG AAC C-3′ and 5′-CGC AGA AAT CAG CAC CGT GTA AG-3′
ANG2: 5′-GAA GGA CTG GGA AGG CAA CGA-3′ and 5′-CCA CCA GCC TCC TGA GAGCAT C-3′
GAPDH: 5′-AAC TTT GTG AAG CTC ATT TCC TGG TAT-3′ and 5′-CCT TGC TGG GCT GGG TGG T-3′
279 Appendix B: (Continued)
MOUSE (Used C57Bl mice)
Lazovic JL, Basu A, Lin H, Rothstein, RP, Krady K, Smith MB, and Levison SW (2005). Neuroinflammation and both cytotoxic and vasogenic edema are reduced in interleukin-1 type 1 receptor-deficient mice conferring neuroprotection. Stroke. 36:2226-31. iNOS Forward 5'-CCCTTCCGAAGTTTCTGGCAGCAGC-3' Reverse 5'-GGCTGTCAGAGCCTCGTGGCTTTGG-3' eNOS 5'-TTCCGGCTGCCACCTGATCCTAA-3' 5'-AACATATGTCCTTGCTCAAGGCA-3'
Cyclophilin 5'-CCATCGTGTCATCAAGGACTTCAT-3' 5'-TTGCCATCCAGCCAGGAGGTCT-3'
280 Appendix B: (Continued)
MOUSE (Used primary mouse hepatocytes)
Allena JW, Johnsonb RS, Bhatia SN (2005). Hypoxic inhibition of 3- methylcholanthrene-induced CYP1A1 expression is independent of HIF-1alpha. Toxicology Letters 155, 151–159.
VEGF Forward 5’-AGTCCCATGAAGTGATCAAGTTCA Reverse 5’-ATCCGCATGATCTGCATGG
PGK 5’-CAAATTTGATGAGAATGCCAAGACT 5’-TTCTTGCTGCTCTCAGTACCACA
GLUT-1 5’-GGGCATGTGCTTCCAGTATGT 5’-ACGAGGAGCACCGTGAAGAT
CYP1A1 5’-AAAACACGCCCGCTGTGAA 5’-TGAATCACAGGAACAGCCACC
β-Actin 5’-AGGCCCAGAGCAAGAGAGG 5’-TACATGGCTGGGGTGTTGAA
281 Appendix B: (Continued)
MOUSE: Antibodies used for immunohistochemistry and immunofluorescence
Kaur C, Sivakumar V, Ang LS and Sundaresan A (2006). Hypoxic damage to the periventricular white matter in neonatal brain: role of vascular endothelial growth factor, nitric oxide and excitotoxicity. J Neurochem. 98:1200-16.
Antibodies Dilution
VEGF Rabbit-polyclonal Santa Cruz Biotechnology (Santa Cruz, CA, USA) 1: 200
NMDA-R1 Rabbit-polyclonal Chemicon International Inc. (Temecula, CA, USA) 1: 200
eNOS Mouse-monoclonal BD Biosciences (Franklin Lakes, NJ, USA) 1: 250
iNOS Mouse-monoclonal BD Biosciences (Franklin Lakes, NJ, USA) 1: 1000
nNOS Rabbit-polyclonal BD Transduction (Franklin Lakes, NJ, USA) 1: 500
282 Appendix B: (Continued)
RAT (retinas)
Kaur C, Sivakumar V, and Foulds WS (2006). Early response of neurons and glial cells to hypoxia in the retina. Invest Ophthalmol Vis Sci. 47:1126-41
Antibodies Dilution
VEGF Rabbit polyclonal Santa Cruz Biotechnology, Santa Cruz, CA 1:200
GluR2/3 Rabbit polyclonal Chemicon International Inc., Temecula, CA 1:200
NMDA-R1 Rabbit polyclonal Chemicon International Inc., Temecula, CA 1:200 eNOS Mouse monoclonal BD Biosciences, San Diego, CA 1:250
iNOS Mouse monoclonal BD Biosciences, San Diego, CA 1:1000 nNOS Rabbit polyclonal BD Transduction Labs, San Jose, CA 1:500
GFAP Mouse monoclonal Chemicon International Inc., Temecula, CA 1:1000
Glutamine synthetase Rabbit polyclonal Sigma-Aldrich, Ann Arbor, MI 1:600
283 About the Author
Edward John Dougherty received a Bachelor’s Degree in Biology from Ursinus
College in 2001 and a M.S. in Biological Sciences from Bowling Green State University in 2004. He was a laboratory teaching assistant for Biology I and Endocrinology while in the Master’s program, and continued on as a laboratory teaching assistant in General
Physiology and Cell Biology after entering the Ph.D. program at the University of South
Florida in 2004.
While in the Ph.D. program at the University of South Florida, Edward received two consecutive second place awards for graduate research given by the Molecular
Biology Specialty Section of the Society of Toxicology as well as a Tharpe Summer
Fellowship from the Department of Biology. He has also coauthored six publications in notable journals such as the Journal of Biological Chemistry and Molecular
Pharmacology and has made several platform presentations at international meetings of the Society of Toxicology. His research on ARNT2 will be featured in Toxicological
Sciences in May 2008.
284