|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2011 flu ob 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0004949
Cuccinello, Sarah Elizabeth.
Analysis of ahr expression and stability in a recombinant yeast model system
h [electronic resource] /
by Sarah Elizabeth Cuccinello.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 160 pages.
(Ph.D.)--University of South Florida, 2011.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
ABSTRACT: The aryl hydrocarbon receptor (Ahr) and the aryl hydrocarbon receptor nuclear translocator (Arnt) are well characterized bHLH-PAS transcription factors shown to regulate expression of xenobiotic metabolism genes. Extensive study has shown that upon treatment with certain aromatic hydrocarbons, mammalian cells rapidly activate the Ahr signaling pathway in order to stimulate gene expression and attempt to metabolize the xenobiotic compounds. It has been shown that after DNA-binding, the Ahr but not the Arnt protein, is quickly eliminated from the nuclear compartment thereby attenuating the dose of gene regulation administered by the Ahr*Arnt transcription factor complex. Previous studies have implicated involvement of the 26S proteasome complex in the degradation process, but the exact identity of the intermediary proteins and/or ligases remains to be defined. Identification and characterization of the protein(s) involved in degrading the receptor is essential for understanding the signaling pathway in its entirety including the mechanism for regulating the genetic response to Ahr ligands. The model organism, Saccharomyces cerevisiae, was used in order to characterize the Ahr signaling pathway and degradation mechanism in a more simplified cellular setting in which the major processes required for growth and development are conserved. First, the AHR and ARNT cDNAs were stably inserted into the yeast genome such that protein expression was inducible. A time course of induction demonstrated detectable levels of Ahr and Arnt proteins via western blotting while protein function was confirmed by detection of ligand-dependent reporter activity in an expressor strain carrying the pLXRE5-Z beta-galactosidase reporter plasmid. Additionally, a rapid reduction in protein levels was observed upon turning off the inducible GAL1 promoter located upstream of both AHR and ARNT cDNAs. Studies in mammalian cell culture have demonstrated that disrupting receptor chaperoning results in rapid Ahr protein turnover, as demonstrated by treatment with Hsp90 inhibitors. In order to determine if reduced Ahr protein expression in the yeast system was attributed to improper chaperoning of the exogenous protein; human heat shock proteins were constitutively expressed from yeast expression vectors in the Ahr and Arnt expressing strains, but did not confer any effect on Ahr stability when protein levels were evaluated by western blotting. Additionally, a strain of yeast was constructed such that the gene encoding the cell-wall protein, ERG6, was deleted from the yeast genome to allow for permeation of proteasome inhibitors. Treatment of this strain with proteasome inhibitors blocked the receptor degradation, therefore implicating the 26S proteasome in Ahr degradation when expressed exogenously in yeast.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Pollenz, Richard S.
x Molecular Biology
t USF Electronic Theses and Dissertations.
Analysis of Ahr Expression and Stability in a Recombinant Yeast Model System by Sarah Elizabeth Cuccinello A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Cell Biology, Microbiology, and Molecular Biology College of Arts and Sciences University of South Florida Major Professor: Richard S. Pollenz, Ph.D. Kristina H. Schmidt, Ph.D. Jessica L. Moore, Ph.D. James Garey, Ph.D. Date of Approval: March 28, 2011 Keywords: Ah receptor, Arnt, transcription factor, protein degradation, Saccharomyces cerevisia e Copyright 2011 Sarah Cuccinello
DEDICA T ION To my ever supportive family for their unending patience and understa nding I would specifically like to dedicate this dissertation to my grandfather, Don Wilson, my father, Tom Wilson, and my husband, A.J. Cuccinello. These men have been wonderful supporters of my educational endeavors and I wish to honor and thank them. ! ! ! ! !
ACKNOWLEDGEMENTS This work could not have been completed without the mentorship and guidance o f Dr. Richard Pollenz. Thank you to my committee members, Dr. Kristina Schmidt Dr. Jessica Moore, and Dr. James Garey I would also like to acknow ledge Dr. Fiona Cra wford for her continued support.
i TABLE OF CONTENTS LIST OF T A BLES ................................ ................................ ................................ .............. iv LIST OF FIGURES ................................ ................................ ................................ ........... v ABSTRACT ................................ ................................ ................................ ..................... v iii CHAPTER ONE : INTRODUCTION ................................ ................................ .................. 1 Basic helix loop helix Superfamily of Transcription Factor s ................................ .. 1 bHLH Subfamily ................................ ................................ ......................... 1 bHLH Zip Subfamily ................................ ................................ .................. 2 bHLH PAS Subfamily ................................ ................................ ................ 2 Classification of bHLH PAS Members ................................ ....................... 2 Ahr Mediated Gene Regulation ................................ ................................ ............. 3 Functional Domains of Ahr ................................ ................................ ........ 3 Cytosolic Unliganded Complex ................................ ................................ .. 5 Ligand Interactions with Ahr ................................ ................................ ...... 8 Nuclear Impo rt ................................ ................................ ........................... 9 Heterodimerization with Arnt and DNA binding ................................ ......... 9 Transcription Activation of Drug Metabolizing Enzymes ......................... 10 Nuclear Export ................................ ................................ ......................... 12 Degradation of the Aryl Hydrocarbon Receptor ................................ .................. 12 Ahr Half life ................................ ................................ .............................. 12 Ligand mediated Ahr Degradation ................................ ........................... 13 Ligan d independent Ahr Degradation ................................ ..................... 18 Suspected ligases involved in Ahr Degradation ................................ ...... 19 Degradation Mechanisms ................................ ................................ ................... 20 Ubiquitin Proteasome Pathway ................................ ............................... 20 Calpain Family of Proteases ................................ ................................ .... 25 Endogenous and Recombinant Expressi on of bHLH Proteins in Yeast ............. 26 Endo gen ous bHLH Proteins in Yeast ................................ ...................... 26 Expression of Heterologous Proteins in Yeast ................................ ........ 27 Expression of Mammalian bHLH Proteins in Yeast ................................ 27 Ahr Studies in Yeast ................................ ................................ ................ 29 Summary of System and Specific Aims ................................ .............................. 32 CHAPTER TWO : GENERATION OF AHR AND ARNT EXPRESSING STRAINS ......... 35 Experimental Question and Rationale ................................ ................................ 35
ii General Strategy ................................ ................................ ..................... 36 Construction of Ahr and Arnt Expressing Strains ................................ ................ 37 Cloning of Selectable Marker Cassettes into AHR and AR N T vectors ................................ ................................ ......................... 37 PCR Amplification of cDNAs and Adjacent Marker Cassettes ................ 3 8 Transformation Procedure ................................ ................................ ....... 39 Integration of the Inducible Promoter ................................ ...................... 4 1 Generation of the Double Knock In Strain ................................ ............... 43 CHAPTER THREE : CHARACTERIZATION OF AHR AND ARNT PROTEIN EXPRESSION IN YEAST STRAINS AND VALIDATION OF YEAST MODEL ............... 52 Experimental Quest ion and Rationale ................................ ................................ 52 General Strategy ................................ ................................ ..................... 53 Detection of Ahr and Arnt Protei n Expression in Yeast Strains .......................... 54 Induction of Ahr and Arnt Proteins in Single Knock In Yeast Strains ................................ ................................ .......................... 55 Induction of Ahr and Arnt Proteins in the Double Knock In Yeast Strains ................................ ................................ .......................... 58 Vali dation of the Yeast Model ................................ ................................ ............. 60 Reporter Assay Method ................................ ................................ ........... 61 Reporter Activation Follow ing Treatment with Ahr Ligand ....................... 62 Gene ration of a Functional Strain ................................ ............................ 69 Assessment of Ahr Protein Turnover in Yeast ................................ .................... 70 Stability of the Ahr in the Absence of Ligand ................................ ........... 70 Optimization of Sample Preparation for Ahr Stability .............................. 78 Preparation of Soluble Protein Fractions ................................ ................. 83 Analysis of Ligand Indepen dent Ahr Degradation in Yeast ..................... 87 Impact of Human Hsp90 Express ion on Ahr Stability in Yeast ................ 90 Effect of Ahr Ligand Treat ment in Yeast Strains ................................ ..... 93 Mechanis m of Ahr Degradation in Yeast ................................ ........................... 100 Generation of a Permeable Yeast Strain ................................ ............... 101 Ahr Protein Stability with Pr oteasome Inhibitor Treatment .................... 105 Sum mary of Experimental Results ................................ ................................ .... 110 CHAPTER FOUR : CONCLUSIONS, IMPLICA TIONS, AND FUTURE DIRECTIONS .. 112 Conclusions and Implications ................................ ................................ ............ 112 Expression of Ahr and Arnt under the GAL1 Pro moter is Detectable in Yeast ................................ ................................ .... 112 Mammalian Ahr and Arnt Pro teins are Functi onal when Expressed in Yeast ................................ ................................ ...................... 113 Ahr and Arnt Proteins Levels ar e Red uced in Y east Over Time ........... 114 Overexpression of Chaperone Proteins does not impact the l oss of Ahr and Arnt Proteins ................................ ............................ 114 Ligand Mediated Degradation of Ahr was not Detectable in Yeast ....... 115
iii Ahr and Arnt Protein Degradation is Reversed with Pr oteasome Inhibitor Treatment ................................ ................................ ..... 116 Hypothesis for Ahr and Ar nt Protein Turnover in Yeast ......................... 116 Future Directions ................................ ................................ .............................. 117 Further Analysis of Ahr and Arnt in Yeast ................................ ............. 117 Analysis of Ahr and Arnt Stabi lity in E3 Knock out Strains .................... 118 Overexpression of Mammalian E3 Ligas es in the Recom binant Strain ................................ ................................ ......................... 118 Analysis of Ahr and Arnt isoforms using a Yeast Model ........................ 119 CHAPTER FIVE : MATERIALS AND METHODS ................................ .......................... 121 Materials ................................ ................................ ................................ ............ 121 Buffers ................................ ................................ ................................ ... 121 Reagents and Chemicals ................................ ................................ ...... 121 Yeast Media ................................ ................................ ........................... 122 Plasmids ................................ ................................ ................................ 122 Antibodies ................................ ................................ .............................. 122 Methods ................................ ................................ ................................ ............. 123 Yeast Transformations ................................ ................................ .......... 123 Random Sporulation ................................ ................................ .............. 123 Activation and Repression of the Galactose Inducible Promoter .......... 124 Preparation of Soluble Protein Fractions ................................ ............... 125 Preparation of TCA Precipitated Protein Samples ................................ 125 Western Blotting ................................ ................................ .................... 126 Reporter Analysis ................................ ................................ .................. 127 LITERATURE CITED ................................ ................................ ................................ .... 128 APPENDICES ................................ ................................ ................................ ............... 141 Appendix A: Table of Strains Gener ated ................................ ........................... 142 Ap pendix B: Additional Figures ................................ ................................ ......... 143 Appendix C: E3 Knockout Preliminary Results ................................ ................. 144
iv LIST OF TABLES TABLE 2.1: Mating Type Genotyping of Crossed Strains ................................ ............... 46 TABLE 2.2: Genotyping and Mating Type Following Random Sporulation ..................... 48 TABLE 3.0: Percentage of galactose added to yeast cultures to minimized induction of Ahr and Arnt protein levels and reduce cellular stress ............. 7 9 TABLE 5.1: Sources for Plasmids ................................ ................................ ................. 122 TABLE A1 : List of Strains Generated ................................ ................................ ........... 1 42 TABLE A2: List of Yeast Genes Involved in Prote asome Mediated Degra dation ........ 144
v LIST OF FIGURES FIGURE 1.1: Ahr and Arnt Functional Domains ................................ .............................. 4 FIGURE 1.2: Canonical Ahr Signaling Pathway ................................ ............................. 6 FIGURE 1.3: Mechanisms of Ahr Degradation ................................ ............................. 14 FIGURE 1.4: Ubiquitin Proteasome Pathway ................................ .............................. 23 FIGURE 2.1: Plasmid maps depicting PCR primers and the result of PCR amplification ................................ ................................ ............................. 38 FIGURE 2.2: Transformed yeast colonies on selective media and c onfirmation of cDNA integration ................................ ................................ ................ 40 FIGURE 2.3: Growth Curves and Doubling Times for KHSY1538 and KHSY1541 strains ................................ ................................ .................. 42 FIGURE 2 .4 : Mating type g eno typing for AHR/ARNT Crossing ................................ ... 45 FIGURE 2.5: Genotyping of haploid spores ................................ ................................ .. 47 FIGURE 2.6: Mating type genotyping after sporulation of Ah r and Arnt expressing strains ................................ ................................ .................... 47 FIGURE 2.7: Plasmid map for pLXRE5 Z reporter plasmid ................................ ......... 49 FIGURE 2.8: Growth Curve a nd Doubling Time for KHSY1566 ................................ ... 51 FIGURE 3.1: Western blot analysis of Ahr and Arnt protein expression in yeast ......... 57 FIGURE 3.2: Induction of Ahr and Arnt proteins under the galactos e inducible promoter in yeast ................................ ................................ ..................... 59
vi FIGURE 3.3: Analysis of galactosidase activity in KHSY1565 and KHSY1 566 following TCDD treatment ................................ ................................ ...... 64 FIGURE 3.4: Analysis of g alactosidase acti vity in KHSY1565 and KHSY 1566 following treatment w ith several known Ahr ligands ................................ 66 FIGURE 3.5: Dose response analysis of KHSY1566 with TCDD and !NF .................. 68 FIGURE 3.6 : Predicted pattern of protein detection when an inducible promoter is tur ned off and the recombinant protein is stable ................................ .. 73 FIGURE 3.7 : Detection of A hr protein degradation in yeast ................................ ......... 74 FIGURE 3.8: Detection of Arnt protein degradation in yeast ................................ ........ 76 FIGURE 3.9: Ahr protein expression is reduced with decreased levels of galactose ................................ ................................ ................................ 80 FIGURE 3.10: Activation of the GAL promoter with 2% or 0.05% galactose and s ubsequently turning off the promoter with glucose causes rapid Ahr turnover in yeast ................................ ................................ ............... 81 FIGURE 3.11 : Analysis of galactosidase activity in KHSY1566 following activation of t he GAL1 promoter with decreasing amounts of galactose ................................ ................................ ................................ 83 FIGURE 3.12: Detection of Ahr protein in soluble protein fractions ................................ 85 FIGURE 3.13: Turnover of Ahr and Arnt in samples prepared using trichloroacetic acid or the soluble protein extraction met h od when the inducible promoter is turned off ................................ ................ 86 FIGURE 3.14: Effect of Hsp90 inhibitors on Ahr and Arnt expressing yeast strain ........ 90 FIGURE 3.15 : Human Hsp90 vector maps and transformation of KHSY1547 with each plasmid ................................ ................................ ........................... 91 FIGURE 3.16 : Ahr turnover is eviden t in strains overexpressing human heat shock proteins ................................ ................................ ......................... 92 FIGURE 3.17 : Induction of Ahr protein with galactose and simultaneous treatment with TCDD revealed equal levels of Ahr protein ..................... 95 FIGURE 3.18 : Effect of ligand treatment on Ahr protein in induced c ultures .................. 97
vii FIGURE 3.19: Ahr protein levels after TCDD treatment in TCA precipitated and soluble protein fraction samples ................................ .............................. 98 FIGURE 3.20 : Effect of TCDD on Ahr and Arnt protein aft er the promoter is turned off ................................ ................................ ............................... 100 FIGURE 3.21: Selection of !erg6 Clones for use in Proteasome Inhibitor Studies ...... 102 FIGURE 3.22 : Effect of MG132 on the doubling time of the parental and permeable strains ................................ ................................ .................. 104 FIGURE 3.23 : Effect of MG132 on Ahr Degradation in Yeast ................................ ........... 107 FIGURE 3.24 : Ahr protein levels shift towards predicted levels with MG132 treatment ................................ ................................ ............................... 108 FIGURE 3.25 : Effect of MG132 on Arnt Degradation in Yeast ................................ ..... 109 FIGURE 3.26 : Arnt protein levels shift towards predicted levels with MG132 treatment ................................ ................................ ............................... 110 FIGURE A .1: Reduction of Ahr protein levels in yeast upon addition of glucose to the growth medium ................................ ................................ ............... 143 FIGURE A.2: Levels of Arnt protein expressed in the KHSY1541 strain are reduced following addition of glucose to the growth medium ................ 143 FIGURE A3: Ahr protein is degraded in the !ubr1 strain ................................ ........... 144
viii ABSTRACT Th e aryl hydrocarbon receptor (Ahr ) and the aryl hydrocarbon rec eptor nuclear translocator (Arnt ) are well characterized bHLH PAS transcription factors shown to regulate expression of xenobiotic metabolism genes. Extensive study has shown that upon treat ment with certain aromatic hydrocarbons, mammalia n cells rapidly activate the Ahr signaling pathway in order to stimulate gene expression and attempt to metabolize the xenobiotic compounds. It has been shown that after DNA b inding, the Ahr but not the Arnt protein, is quickly eliminated from the nuclear compartment thereby attenuating the dose of gene regulat ion administered by the AhrArnt transcription factor complex. Previous studies have implicated involvement of the 26S proteasome complex in the degra dation process, but the exact identity of the intermediary proteins and/or ligases remains to be defined. Identification and characterization of the protein(s) involved in degrading the receptor is essential for understanding the signaling pathway in its entirety including the mechanism for regula ting the genetic response to Ahr ligands. The model organism, Saccharomyces cerevisiae was used in order to characterize the Ahr signaling pathway and degradation mechanism in a more simplified cellular setting i n which the major processes required for growth and development are conserved First, the AHR and ARNT cDNAs were stably inserted into the yeast geno me such that protein expression was induc ible A time course of induction demon strated detectable levels of Ahr and Arnt proteins via western blotting while protein function was confirmed by detection of ligand dependent reporter activity in an
ix expressor strain carrying the pLXRE5 Z beta galactosidase reporter plasmid. Additionally, a rapid reduction in prot ein levels was observed upon turning off the inducible GAL1 promo ter located upstream of both AHR and ARNT cDNAs. Studies in mammalian cell culture have demonstrated that disrupting receptor chaperoning results in rapid Ahr protein turnover, as demonstrated by treatment with Hsp90 i nhibitors In o rder to determine if reduced Ahr protein ex pression in the yeast system was attributed to improper chaperoning of the exogenous protein; human heat shock protein s were constitutive ly expressed from yeast expression vector s in the Ahr and Arnt expressing strains, but did not confer any effect on Ahr stability when protein levels were evaluated by western blotting. Additionally, a strain of yeast was constructed such that the gene en coding the cell wall protein, ERG6 was deleted from the yeast genome to allow for permeation of proteasome inhibitors. T reatment of this strain with proteasome inhibitors blocked the receptor degradation, therefore implicating the 26S proteasome in Ahr d egradation when expressed exogenously in yeast.
1 CHAPTER ONE: INTRODUCTION Basic helix loop helix Superfamily of Transcription Factors The basic helix loop helix (bHLH) family of transcription factors is composed of over 240 genes involved in the regulation of many developmental and physiological processes ( reviewed by Massari and Murre 2000). Members of this family encode a basic DNA binding region and an adjacent helix loop helix dimerization domain and interact wi th DNA subsequent to homodimerization or heterodimerization with other bHLH proteins. bHLH pro teins can be divided into 3 subfamilies, the first sub family members encode a bHLH domain, the second have a bHLH domain and an adjacent leucine zipper, and the third have a bHLH domain and a flanking PAS (PER, ARNT, SIM) domain ( reviewed by Kewley et al 2004 ) bHLH Subfamily Members of the first bHLH subfamily include M yoD N euroD and S cl and have an NH 2 terminal bHLH domain. These proteins are involved in regulation of myogenesis, neurogenesis, and hematopoiesis and are found to be constitutively expressed but restricted to particular tissues and stages in development. They are incapable of forming homodimers and instead form heterodimers with other bHLH proteins and bind to the canonical E box element, 5' CANNTG 3' l ocated upstream of target genes (reviewed by Massari and Murre 2000 and Kewley et al 20 0 4).
2 bHLH Zip Subfamily Similar to other bHLH proteins, bHLH Zip members bind the canonical E box cor e enhancer sequence in the promoter regions of their target genes. However, these proteins, including Myc, Max, and Mad, encode a leucine zipper dimerization domain adjacent to the bHLH region. Again, bHLH Zip proteins are constitutively active, but are only expressed in certain tissues and developmental stages and bind at E box enhancer sequences to initiate transcription (reviewed by Kewley et al 2004). The Myc/Max/Mad network of protein s functions in cell cycle control and plays a role in tumor format ion (Luscher and Larsson 19 9 9). bHLH PAS Subfamily The bHLH PAS sub family of genes is an offshoot of the larger bHLH family and its members include: Ahr, Arnt1, Arnt2, Per, Sim1, Sim2, Clock, Hif 1 ", Hif 2", Hif 3 Bmal 1, and Bmal 2 ( reviewed by Kewley et al 2004 and Crews 1998 ). Members of this family function in gene expression networks that play a role in detection and adaptation to environ mental changes, including xenobiotic metabolism, oxygen balance, and circadian rhythm mainten a nce ( Gu et a l 2000). Similar to other bHLH proteins, they possess a modular structure with specific conserved functional domains including the basic DNA binding domain and the helix loop helix domain required for protein dimerization. In addition to the bHLH domain, this sub f amily also shares a common PAS region, so named for the first three proteins identified as containing this homology; Per ( Drosophila Period protein), Arnt (Aryl hydrocarbon receptor nuclear translocator protein), and Sim ( Drosophila Single minded protein). A dditionally, most bHLH PAS members have transcription activation domains within their C termi n us. Classification of bHLH PAS Members Members of the bHLH PAS sub family are further divided into two classes ; receptors that function in a specific pathwa y are
3 classified separately from receptors that act as generic partner proteins. Ahr, Hif, and Sim are Class I bHLH PAS proteins that become transcriptionally active when coupled with a Class II bHLH PAS protein, namely, Arnt and Bmal Class I proteins a re unable to form homodimers or heterodimers with other Class I proteins while Class II bHLH PAS proteins have been isolated as homodimers and heterodimers with other members of the same class (reviewed by Kewley et al 2004) Ahr Mediated Gene Regulation The aryl hydrocarbon receptor (Ahr) has been studied extensively in the field of toxicology for its role in transcriptional activation of drug metabolizing enzymes in response to exposure to certain persistent environmental contaminants. Each step of the signal transduction pathway has been evaluated; from its latent unliganded state, through ligand binding and nuclearization, heterodimerization with its DNA bindin g partner, DNA binding and transactivation of target genes, and finally its targeted degradat i on. Functional Domains of Ahr Cloning of the Ah receptor (Burbach et al 1992) and comparative sequence analysis revealed a basic helix loop helix domain located in the amino terminus of the protein. The N terminal basic domain consists of a short sequ ence of basic residues that directly interact with the major groove of DNA in the promoter region of target genes. This interaction is facilitated by the dimerization of Ahr with its DNA binding partner Arnt along the helix loop helix region The HLH is comprised of two amphipathic alpha helices that are separated by a flexible loop. Upon dimerization, the HLH regions form a four helix bundle that then stabilizes the dimer in a DNA binding conformation (reviewed by Massari and Murre 2000).
4 Immunoprecip itation studies (Perdew 1988, Reyes et al 1992) revealed potential interactions of Ahr with the 90kDa heat shock protein (Hsp90) and the aryl hydrocarbon receptor nuclear translocator protein (Arnt), suggesting the presence of additional domains required f or interaction with these proteins. In order to identify the site for interactions with these proteins and to locate additional functional domains, a series of deletion mutants were generated and expressed in vitro (Fukunaga 1995). This study, along with a similar experiment by Soshilov and Dension (2008) revealed precisely the amino acid sequences that compose the bHLH domain, the PAS A and PAS B domains, a ligand binding domain, the site for Hsp90 binding, as well as the C terminal transactivation domain (TAD). FIGURE 1.1: Ahr and Arnt Functional Domains. Ahr and Arnt share homology in their N terminal regions, which encode conserved bHLH PAS domains that function in DNA binding and serve as a surface for heterodimerization of these two proteins. Th e bHLH of Ahr also functions in facilitating p r otein :protein interactions with Hsp90 and encodes a nuclear localization and nuclear export signal. Both Ahr and Arnt encode C terminal transactivation domains that recruit transcription factors to the promot er region of target genes when Ahr and Arnt are bound together at enhancer elements of target genes.
5 Additional studies further characterized the TAD domain, which consists of three sub domains shown to function synergistically (Ma et al 1995). Following DNA binding at enhancer sequences, certain co activators and general transcription factors are recruited to the promoter region via the C terminal t ransactivation domain s of Ahr and Arnt (Hank inson 2005) to facilitate transcription initi ation. Additionally, a putative nuclear localiz a tion signal (NLS) as well as a nuclear export signal (NES) was also identified in Ahr using sequence analysis (Ikuta et al 1998, Holmes and Pollenz 1997, Pollenz and Barbour 2000). Cytosolic Unliganded Comp lex The unliganded Ahr exists in a complex with two molecules of Hsp90 bound by the co chaperone p23, and an immunophilin like molecule called Xap2 as shown in Figure 1.2. Studies using the mouse Hepa 1 cell line, expressing the Ah b 1 isoform, revealed cytoplasmic localization of the unliganded receptor complex that is rapidly nuclearized following ligand binding (Pollenz et al 1993). However, analysis of other Ahr allelic variants, including Ah b 2 derived from mouse C 2 C 12 cells and the rat Ahr from A 7 smooth muscle cells, showed a conflicting localization pattern for the unliganded receptor complex (Pollenz and Dougherty 2005). In these cell lines, the unliganded receptor exhibits dynamic nucleocytoplasmic shuttling, however, the receptor was shown to accumulate in the nucleus following ligand binding similar to the Hepa 1 cell line. Hsp90 is a molecular chaperone whose expression is up regulated in response to cellular stress; however, in both stressed and unstressed cells Hsp90 plays a role in proper protein folding. Hsp90 interacts with newly synthesized proteins to facilitate folding, but also acts to stabilize and refold denatured proteins subjected to stress (Chen et al 2005). In this pathway, Hsp90 binds Ahr within the receptor's bHLH and PASB d omains (Figure 1.1) (Soshilov and Dension 2008); it has been suggested that this
6 interaction maintains the receptor in an inactive state by shielding the nuclear localization signal (Ikuta et al 1998). Hsp90 binding has also been implicated in maintaining Ahr in a high affinity ligand binding configuration (Pongratz et al 1992). FIGURE 1.2: Canonical Ahr signaling pathway. In the absence of ligand, Ahr predominantly exists in a cytoplasmic complex with a dimer of Hsp90, p23 and Xap2. Upon ligand binding, the latent protein complex translocates to the nucleus of the cell whereby Ahr and Arnt heterodimerize, bind XREs in the 5' regulatory region of dioxin responsive genes, and activate transcription of Phase I and Phase II xenobiotic metabolism gene s. Following DNA binding, Ahr is rapidly degraded via the 26S proteasome. The AhrHsp90 interaction is maintained following ligand binding and nuclear localization, but when this complex was stabilized with the addition of sodium molybdate, Ahr was rende red unable to interact with its DNA binding partner Arnt or transactivate target genes. This result suggested that Hsp90 is dissociated from Ahr after the receptor enters the nuclear compartment. It is also likely that the constitutively nuclear Arnt pro tein therefore displaces Hsp90 due to a higher affinity for Ahr binding (Heid et al 1999). The co chaperone protein p23 is associated with the Hsp90 dimer, as was
7 determined by immunoprecipitation studies (Kazlauskas et al 1999). This interaction occurs at the N terminal ATP binding domain of Hsp90. It was also revealed that the dissociation of p23 from Ahr requires both ligand binding and Arnt protein in an in vitro expression system. This result confirmed that activati on of the receptor from the latent complex to its transcriptionally active form occurs in the nuclear compartment such that Hsp90 and p23 are displaced in the presence of Arnt. A yeast two hybrid experiment identified an additional protein that interacts w ith the Ah receptor (Carver and Bradfield 1997, Ma and Whitlock 1997). Sequence analysis of the protein revealed three TPR (tetratricopeptide repeats) domains. TPR sequences are 34 amino acids long and form short helices to facilitate protein protein i nteractions. The N terminus of the protein contains a region of 80 amino acids with homology to the FKBP family of molecular chaperones (Carver and Bradfield 1997). Interaction of the Xap2 protein (also referred to as Aip and Ara9) with Ahr was confirme d using immunoprecipitation experiments (Ma and Whitlock 1997, Carver et al 1998). The exact function of Xap2 (hepatitis B virus X associated protein 2) in Ahr signaling is not completely understood, but it is believed to play a role in nucleocytoplasmic shuttling of the receptor (Lees et al 2003, Pollenz and Dougherty 2005, Pollenz et al 2006). Comparative analysis of Xap2 interaction with Ahr allelic variants demonstrated a relationship between Xap2 levels and nucleocytoplasmic shuttling of the unligand ed receptor. The Ah b 1 allele, expressed in the C57BL mouse and Hepa 1 cells derived from that strain, is associated with 80 90% more Xap2 protein as compared to the Ah b 2 allele expressed in C 2 C 12 cells or the rat Ahr, as determined by immunoprecipitatio n experiments. The Ah b 1 allele is localized primarily in the cytoplasm, while the Ah b 2 rat, and human Ahr demonstrate a dynamic shuttling between the nuclear and cytoplasmic compartment, this effect is attributed to reduced levels of Xap2 associated wit h the receptor (Ramadoss et al 2004, Pollenz et al 20 0 6).
8 Ligand Interactions with Ahr. The Ahr signaling pathway becomes activated upon binding of ligand to the hydrophobic ligand binding pocket within the bHLH domain of the latent receptor. Ahr ligand s are typically planar and hydrophobic molecules, most of which are synthetic compounds including halogenated aromatic hydrocarbons and polycyclic aromatic hydrocarbons. Additionally, several naturally occurring compounds have also been ide ntified as Ahr ligands. Halogenated aromatic hydrocarbons (HAHs) are the most potent Ahr ligands due to their metabolic stability; HAHs include polyhalogenated dibenzo p dioxins, dibenzofurans, azobenzenes, naphthalenes, and biphenyls. Documented effects of exposure to these compounds include teratogenic, mutagenic, and carcinogenic effects. These compounds are lipophilic and are known to bioaccumulate as a result of dietary intake of fats. The most widely studied and most potent ligand of Ahr is T C DD (2 ,3,7,8 tetrachl orodibenzo p dioxin); it is produced as a byproduct during synthesis of 2,4,5 trichlorophenoxyacetic acid, the broad spectrum herbicide. Trace amounts of chlorinated dibenzofurans, another potent ligand, are produced as a byproduct of commercial productio n of the fungicide pentachlorophenol, used in paper manufacturing. Polychlorinated biphenyls (PCBs) were manufactured in the US until 1977 for commercial use as lubricants, plasticizers, and adhesives (Poland and Knutson 1982). Polycyclic aromatic hydro carbons (PAHs) are more metabolically labile compounds and are subsequ ently less potent ligands for the Ah receptor. PAH ligands include 3 methychloranthracene (3 MC), benzo(a)pyrene (BAP), benzanthracene, and benzoflavones. These compounds are produced as a byproduct of fossil fuel combustion (reviewed by Liu et al 2008). BAP is of particular interest as it is a carcinogen that is found in cigarette smoke. PAHs are also present in some foods
9 including oils, fats, and cooked meats; in this case they are produced as a byproduct of incomplete steroid metabolism (Larsen 1995). Several endogenous compounds have been shown to bind Ahr, activate the receptor, and induce Ahr dependent gene expression. These include indoles, tetrapyroles, arachidonic acid metab olites, and several carotinoids (Adachi et al 2001, Denison and Nagy 2003). Indoles, including indigo and indirubin, are tryptophan derivatives. Bilirubin is an example tetr apyrole and lipoxin A and prostaglandin G are archidonic acid metabolites that in teract with A hr. Nuclear Import. Following ligand binding, Ahr undergoes a conformational change such that the basic nuclear localization signal (NLS) becomes exposed. The nuclear localization signal of Ahr is composed of a bipartite region spanning ami no acids 12 41 (Song and Pollenz 2003). This sequence is bound by karyopherin ", an adapter protein, that interacts with importin to mediate import via the nuclear pore complex (Pemberton and Paschal 2005). The intact AhrHsp90Xap2 complex is then rap idly translocated to the nuclear compartment (Pollenz et al 1993). Expectantly, disruption of the Ahr NLS resulted in a receptor defective for TCDD dependent nuclear import (Song and Pollenz 2003). Studies supporting this model were carried out using the Ah b 1 allelic variant which has been shown to exhibit a predominantly cytoplasmic localization in the absence of ligand. Alternatively, rat Ahr, human Ahr, and mouse Ah b 2 alleles appear to undergo dynamic nucleocytoplasmic shuttling in the absence of li gand and exposure of each of these cell type s to ligand resulted in a predominantly nuclear localization of Ahr (Pollenz and Dougherty 20 0 5). Heterodimization with Arnt and DNA binding The aryl hydrocarbon receptor nuclear translocator (Arnt) protein is a bHLH PAS protein whose expression is restricted
10 to the nuclear compartment of higher eukaryotic cells (Pollenz et al 1993). In 1992, Reyes et al identified Arnt as a component of the DNA binding form of the Ah receptor and sequence analysis suggested t hat Ahr and Arnt form a heterodimer along their bHLH PAS domains. Domain analysis using deletion mutants provided insight into some of the functional aspects of Arnt. The basic region of Arnt facilitates DNA binding, while the HLH and PAS domains both c ontribute to Ahr binding. Additionally, a C terminal glutamine rich domain is required for transcriptional activity and can function independently of the other protein domains (Li et al 1994, Reisz Porszasz et al 1994 ). DNA binding of the activated Ahr complexed with Arnt takes place at enhancer sequences located in the promoter regions of target genes. The consensus sequence for these XREs (xenobiotic response elements) differs from the traditional E box recognized by m ost members of the bHLH superfamily. The basic regions of Ahr and Arnt bind the consensus DNA sequence 5' TNGCGTG 3' (Denison et al 1988) with half site specificity such that Ahr interacts with the 5' TNGC 3' sequence and Arnt binds the 5' GTG 3' sequence (Swanson et al 1995). The CYP1A1 gene, a well studied target for Ahr mediated induction, has eight XREs located in its upstream regulatory sequence. Following AhrArnt binding, the promoter region of CYP1A 1 undergoes chromatin remodeling to allow adequa te room for binding of transcriptional machin e ry. Transcription Activation of Drug Metabolizing Enzyme s Ahr modulates the expression of a number of genes that encode phase I and phase II drug metabolizing enzymes including: cytochrome P450 1A1 (CYP1A1), CYP1A2, CYP1B1, glutathione S transf e rase Ya subunit, plasminogen activator inhibitor 2, interleukin 1, and UDP glucuronsyltransferase 1A1 (Poland and Knutson 1982).
11 Ahr and Arnt proteins convey a signal to initiate promoter occupancy and transactivation by way of their carboxyl terminal regions (Jain et al 1994, Ma et al 1995, Ko et al 1997, Whitlock 1999). Within the Ahr TAD region, exists three sub domains: an acidic region, a proline rich region, and a serine rich domain (Ma et al 1995, Ko et al 1997) The acidic domain spans amino acids 515 583, is rich in glutamic acid and aspartic acid residues (24%) and has the strongest transactivation potential of the three sub domains. Amino acids 643 805 are composed of 13% proline residues and amino acids 72 6 805 are 16% serine residues. The latter two sub domains exhibit weaker, but still independent abilities to activate transcription. Ahr is phosphorylated within the C terminal domain (Mahon and Gasiewicz 1995) likely mediated by PKC. PKC (Protein Kinase C), a serine/threonine kinase, is essential for transactivation following Ahr/Arnt binding at XREs. Subsequent to phosphorylation, Ahr is capable of recruitment and assembly of the transcription initiation complex (Chen and Tukey 1996, Long et al 1998). Transcriptional adaptors called co activators have been implicated in Ahr gene regulation (Nguyen et al 1999, Beischlag et al 2002, Hankinson 2005). In general, co activators are recruited to DNA bound transcription factors and communicate with proteins i n the core transcription initiation complex. They also facilitate transactivation by way of their histone acetyltransferase activity which confers a more relaxed chromatin structure (Beischlag et al 2002). Co activators interact with a short helical LX XLL motif located within the protein's transactivation domain (Flaveny et al 2008). Specific co activators shown to influence Ahr mediated gene regulation include: SRC 1(steroid receptor co activator 1), NCoA 2 (nuclear co activator 1), p/CIP (p300/CBP co integrator protein), p300, and CBP (CREB binding protein) (Hankinson 2005) and it is believed that there may be some combination o f co activators involved in both chromatin remodeling as well as transcription initiation eve n ts.
12 Nuclear Export. As nuclear export appears to occur following Ahr mediated transactivation of target genes, it is believed that nuclear export functions in regulation of the dose of gene induction. The Ahr nuclear export signal (NES), amino acids 63 71, is a leucine ric h sequence located within the helix 2 domain (Pollenz and Barbour 2000). The nuclear export receptor CRM 1 interacts with the NES and mediates its transfer through the nuclear pore complex (Pemberton and Paschal 2005). Expectedly, mutation of the Ahr NES resulted in accumulation of receptor in the nucleus following ligand treatment (Pollenz and Barbour 2000) that was accompanied by an increase in Ahr mediated transcription activation. D egradation of the Aryl Hydrocarbon Receptor Turnover of the Ah recept or in mammalian cells can be attributed to three cellular circumstances with each playing a di stinct role in the downstream gene regulatory effects of the receptor. First, the normal half life of the latent receptor was determined via pulse labeling H epa1 c1 c 7 cells grown in culture wherein the half life was found to be approximately 28hrs (Ma and Baldwin 2000). In addition to normal protein turnover, ligand mediated degradation of the receptor has been observed upon treatment with TCDD and similar compoun ds which thereby reduced the receptor half life to approximately 3hrs (Ma and Baldwin 2000, Pollenz 1996). Finally, ligand independent degradation of the receptor occurs following disruption of the interaction between A hr and a dimer of Hsp90 upon treatme nt with benzoquinone ansamycin antibiotics (Chen et al 1997 Song and Pollenz 2002 ). Ahr Half life. The half life of unliganded Ahr was first evaluated by Swanson and Perdew in 1993 using ligand binding to measure receptor protein levels in sucrose
13 dens ity gradients of Hepa 1 cells The results suggested that the t # for u nliganded receptor was approximately 7.7 hours. When treated with the agonist !NF, the half life of the receptor was extended to approximately 9.7 hours, while treatment with the partial antag o nist "NF yielded a half life similar to the unliganded receptor (Swanson and Perdew 1993). This result was in contrast to observed reduction in receptor levels following TCDD treatment (Prokipcak and Okey 1991). These conflicting results were attributed to variation in ligand lability such that metabolism of !NF allows receptor levels to recover more rapidly than when treated with TCDD, but is also likely due to the choice of technique. To definitively determine the receptor half life, Ma and Baldwin (2000) evaluated the Ah r b 1 t # in the presence and absen ce of TCDD using the more exact method of pulse chase labeling. I mmunoprecipitation following pulse chase with [ 35 S] methionine revealed the receptor half life to be 28 hours in the absence of ligand and 3 hours following ligand treatment. Interestingly, this type of analysis has not been carried out to verify the half life of the other Ahr mouse variants, in other species, or when expressed heterologously in other model organi s ms. Ligand mediated Ahr Degradation. Ahr degradation was first described usi ng mouse Hepa 1 ( liver hepatoma ) cells following treatment with [ 3 H] TCDD (Prokipcak and Okey 1991). Nuclear and cytoplasmic fractions revealed reduction in cytoplasmic receptor in conjunction with an increase in nuclear receptor levels within two hours o f ligand treatment. Subsequently nuclear receptor levels were reduced to background levels within six hours, with no reaccumulation in the cytoplasmic fraction. This result suggested that the protein was not simply being shuttled back to the cytoplasm af ter nuclearization, but that it was being degraded.
14 FIGURE 1.3: Mechanisms of Ahr degradation. Ahr degradation occurs via two distinct mechanisms. First, ligand mediated degradation occurs after several events subsequent to ligand binding. The latent Ahr complex is translocated to the nuclear compartment, heterodimerizes with Arnt, and activates transcription of genes that encode xenobiotic metabolizing enzymes. Ahr is then exported to the cytoplasm where it is degraded via the 26S proteasome. This mechanism of degradation is blocked with proteasome inhibitor treatment and when active transcription and translation are blocked in the cell. An alternative degradation occurs following treatment with Hsp90 inhibitors such as geldanamycin (GA) The dime r of Hsp90 is unable to interact with Ahr which is then translocated to the nucleus. Here, Ahr does not interact with Arnt or bind DNA, but it is rapidly degraded by the 26S proteasome in the nuclear compartment. This mechanism is also blocked by proteas ome inhibitors; however, treatment with transcription and translation blockers do not block Ahr turnover as is seen in the ligand mediated system. Additionally, neither mode of degradation is blocked with calpain inhibitor treatment. I mmunofluorescence m icroscopy and western blotting of mouse Hepa 1 cells provided visual evidence confirming ligand dependent receptor degradation (Pollenz 1996) TCDD treatment reduced the total Ahr pool by 85% within four hours of tr eatment while the amount of Arnt in the same lysates remained constant. Nuclear and cytoplasmic fractions revealed that the receptor was predominantly localized in the cytosol prior to addition of ligand and was maximally nuclear within one hour of TCDD treatment. These fractions also revealed a predominant ly nuclear localization for Arnt
15 that was unchanged in the presence or absence of TCDD. Additionally, immunofluorescence microscopy of cells dosed with ligand for various time points mirrored the localization of the receptor that was demonstrated via western blotting (Pollenz 1996 ; reviewed by Pollenz 2002 ). To confirm that loss of receptor was directly related to turnover of the protein as opposed to a block in transcription at the Ah locus, quantitative PCR was performed in order to evaluate the mRNA levels of Ahr over a seventy two hour TCDD treatmen t. Messenger RNA levels for Ahr and Arnt were unchanged throughout the time course, while induction of P450 mRNA at one hour served as a control for the experiment (Giannone et al 199 8). Certain that the observed reduction in receptor level was due to some targeted turnover of the protein itself, researchers began investigating t he mechanism responsible for Ahr degradation. Initial experiments entailed co treatment of cells in cultu re with ligand and various protease inhibitors. Multiple groups demonstrated that MG 132, a compound used to inhibit the 26S proteasome, effectively prevented degradation of the ligand activated receptor (Ma and Baldwin 2000, Roberts and Whitelaw 1999, So ng and Pollenz 2002). Additionally, calpain inhibitors were tested and unable to prevent loss of the protein (Davarinos and Pollenz 1999, Roberts and Whitelaw 1999, Ma and Baldwin 2000, Pollenz 2007). Stabilization of liganded Ahr with protease inhibitor s also led to an increase in cytochrome P450 expression levels (Ma and Baldwin 2000), further demonstrating the importance of understanding the degradation mechanism, as the removal of the receptor is the means in which the pathway is turned off. Next, re searchers investigated the c ellular compartment in which Ahr degradation was taking place. Roberts and Whitelaw (1999) generated a constitutively nuclear Ah receptor (Ahr NLS) found to have a very short half life (less than one hour). The A hr NLS turned over rapidly in the presence or absence of ligand and when
16 c omplexed with Hsp90 or with Arnt suggesting that the nuclear compartment is the site for receptor degradation. In contrast, a report by Davarinos and Pollenz (1999) suggests that Ahr degradation occurs in the cytoplasm following TCDD treatment, and nuclear import followed by nuclear export of the receptor. Treatment with leptomycin B (LMB), a fungal antibiotic that inhibits the function of nuclear export proteins, in combination with liga nd caus ed an accumulation of Ahr in the nuclear compartment and as such the ligand mediated Ahr degradation was blocked. This conclusion was further val idated through the use of an Ahr with a defective nuclear export signal (Ahr $NES). Receptor turnover i n the tr ansiently expressing Ahr $NES strain was reduce d when compared to wild type Ahr turnover, as observed via western blotting and immunofluorescence microscopy (Davarinos and Pollenz 1999). The degradation site was again evaluated in Ahr $ NLS mutants (Song and Pollenz 2003) whereby receptor turnover occurred with ligand treatment even when nuclearization was blocked. This study, along with several others (Pollenz and Dougherty 2006, Pollenz et al 2006) provided further evidence for ligand mediated Ah receptor degradation that takes place in the cytoplasm. Ahr degradation studies have come to focus largely on the exact manner in which Ahr is targeted by the ubiquitin proteasome pathway for destruction. The Ahr protein has been investigated for potential sites wherein a conformational change or post translational modification may lead to targeted degradation. Other components of the latent receptor complex as well as components of the active transcriptional activation complex have been implicated in targeting t he receptor for turnover. Sites for specific post translational signals that may trigger ligand mediated degradation were investigated. Treatment o f the constitutively nuclear Ahr (Ahr NLS) with phosphatase inhibitors produced a higher molecular weight pr otein as detected by western blotting, suggesting that the protein exists in a phosphorylated state. Ahr NLS expressing cells were treated with combinations of ligand, MG132, and phosphatase
17 inihibitor and protein samples w ere immunoprecipitated with Ahr specific antibody. Detection of ubiquitin by antibody staining was significant when cells were exposed to all three treatments such th at Ahr was activated by ligand but could not be degraded by the 26S proteasome. It was therefore suggested that phospho rylation of the receptor following nuclearization is required for ubiquitination and subsequent degradation to occur (Roberts and Whitelaw 1999). Pollenz et al (2005) evaluated the role of the transactivati on domain in ligand mediated Ahr degradation in or der to determine if transcription regulation is required prior to the degradation event. Truncation mutants were generated and stably expressed in the LA 1 Hepa 1 variant line. The AHR 500 mutant, lacking all three TAD regions, and the AHR 640 lacking two of the three TADs showed no CYP1A1 induction and reduced CYP1A1 induct ion as compared to wild type Ahr, respectively. Though transactivation was reduced, thes e Ahr s were degraded foll owing ligand treatment and degradation could be blocked with MG132 trea tment suggesting that transcriptional a ctivation is not required for Ahr turnover. Arnt's role in Ahr turnover was investigated by Pollenz (2005) using a C4 Hepa 1 variant that expr esses reduced levels of the Arnt prot ein with wild type levels of Ahr. U pon TCDD treatment, Ahr levels were slightly reduced, but the majority of the receptor pool remained. In contrast cells expressing a full complement of A rnt protein in conjun ction with normal levels of Ahr TC DD treatment greatly reduced Ahr levels as co mpared to vehicle treated controls. These results show that Ahr degradation requires dimerization with Arnt and likely also requires a DNA binding event that occurs upon dimer formation. Giannone et al (1995 and 199 8) demonstrated an Ahr stabilization effect in response to treatment with ligand and actinomycin D (AD), a transcription inhibitor. This effect was also observed in Hepa 1 cells treated with ligand and cycloheximide (CHX),
18 an inhibitor of eukaryotic translation. Blocked degradation was also observed in rat smooth muscle cells and mouse 10T1/2 embryonic fibroblasts that express the Ah b 2 allele (Pollenz et al 2005). These results suggest that expression of an a uxiliary protein with a rapid turnover is a n additional compon ent of the degradati on machiner y Ligand independent Ahr Degradation. Ligand independent degradation of the receptor occurs upon administration of Hsp90 inhibitors such as geldanamycin (GA). GA, a benzoquinone ansamycin antibiotic, prevents heterodimerization of Hsp90 wi th the receptor and thu s results in exposure of the Ahr nuclear localization signal in the amino terminus of the receptor's protein sequence. This event subsequently induces rapid nuclear localization and degradation of the receptor in the absence of ligand (Song and Pollenz 2002). Two other Hsp90 inhibitors herbimycin and novobioc in have been shown to reduce Ahr signaling in reporter assays by interacting with the ATPase/p23 binding site in the N terminus of Hsp90. The unrelated compound radicicol has also been shown to disrupt ATP and p23 binding in the N terminus of Hsp90 (Cox and Miller 2002). These studies suggest that Hsp90 plays a role in stabilizing Ahr in the latent complex. In mammalian tissues, administration of Hsp90 inhibitors causes the Ah receptor to translocate to the nucleus where it is d egraded within 3 hours of treatment, as compared to 6 hours with TCDD treatment. Interestingly, the nuclear accumulation of Ahr does not occur in conjunction with transcription activation of target genes (Song and Pollenz 2002). GA mediated degradation w as blocked when Hepa 1 cells were pre treated with MG132, providing that turnover is still mediated by the 26S proteasome although it is occurring at a much more rapid rate. Further experimentation in order to evaluate the mechanism of ligand independent d egradation demonstrated treatment with leptomycin B (LMB), which was
19 shown to block ligand mediated degradation, does not inhibit this degradation and Ahr levels are drastically reduced in cells treated with GA in combination with LMB. This effect suggest s that the degradation machinery responding to GA treatment resides in the nucleus of the cells. Importantly, cycloheximide (CHX) also d id not prevent the GA mediated degradation of the receptor (Pollenz et al 2005) suggesting that expression of an additi onal short lived protein is not essential for this mechanism of degradation and thus providing additional evidence that the ligand dependent and independent mechanism of degradation are occurring via distinct pathways. The TAD truncation mutants generated by Polle nz (2005) were evaluated for Ahr turnover with actinomycin D (AD) and cycloheximide (CHX) treatments. While it was previously mentioned that these truncated Ahr s degraded following ligand binding, it should be noted that the proteins degraded sig nificantly faste r (2.5 X) than the wild type Ahr Also, AD and CHX treatment did not block turnover of AHR 500 and AHR 640 following ligand exposure. Together, the increased rate of turnover along with the non effect observed with transcription and transla tion inhibitor treatment suggests that the TAD mutants are degrading in a fashion similar to Ah r with disrupted Hsp90 chaperoning. As demonstrated using protease inhibitors, ligand mediated as well as ligand independent activation and degradation of the a ryl hydrocarbon receptor occurs by way of proteolysis via the 26S proteasome (Ma and Baldwin 2000, Pollenz et al 2005), but as yet, the manner in which the receptor is targeted for degradation is unclear. Suspected Ligases involved in Ahr Degradation. Seve ral reports have implicated Chip ( Carboxyl t erminus of Hsp70 interacting p rotein) as the particular E3 ubiquitin liga se responsible for targeting Ahr for proteasomal degradation (Lees et al 2003, Morales and Perdew 2007). Chip is a U box dependent ubi quitin E3 ligase shown to interact with molecular chaperones and target their substrates for degradation (Jiang et
20 al 2000). With Hsp90's crucial role in receptor stability and degradation, a role for this ligase in receptor degradation seemed promising. However, several reports (Pollenz and Dougherty 2005, Morales and Perdew 2007) refuted its role as a mediator of Ahr's targeted degradation by the proteasome. While both Ahr and Hsp90 have been shown to interact with Chip via immunoprecipitation, there h as been no evidence of a function in degradation in this system. Ligand dependent Ahr degradation was observed following knock down of Chip using siRNA in Hepa 1 cells and in cells derived from a Chip knockout mouse strain, therefore suggesting that some other E3 is responsible for targeted Ahr degradation. While, disrupting this interaction in vivo does not produce an effect on ligand mediated receptor degradation, binding of Chip to Ahr or Hsp90 may plausibly play a role in degradation of misfolded rece ptor. Ahr itself has been implicated as an E3 ligase when complexed with cullin 4B protein in the presence of Ahr ligands, TCDD or 3 methylcholanthracene (Wormke et al 2003, Ohtake et al 2007). This work suggested a role for Ahr in mediating the proteasom al degradation of the estrogen receptor (ER) protein when cells were treated with certain Ahr ligands. Immunoprecipitation experiments demonstrated an interaction between the Ahrcullin4B complex and ER, while ER turnover was reduced when Ahr or cullin 4B levels were knocked down using siRNA. However, specific degradation of Ahr was unaffected when cullin 4B was knocked down using siRNA, therefore suggesting that Ah r does not likely act as an E3 ligase for its own proteolytic degradation. Degradation Mec ha n isms Ubiquitin Proteasome Pa thway The ubiquitin proteasome pathway include s a group of proteins involved in a highly selective mechanism of rapid protein turnover that is required in order to carry out essential regulatory processes (Ciechanover and Schwartz 1994, Jentsch 1992). Cell cycle control, transcriptional regulation, an d antigen
21 processing in immune response are examples of cellular events in which the degradation of proteins by the ubiquitin proteasome pathway is required (Hochstrasser 1996). One role of the ubiquitin proteasome pathway is to function in locating, iden tifying, and degrading abnormal proteins in order to eliminate them from the cellular protein pool. Aberrant proteins including truncated (Kohlmann et al 2008), misfolded (Betting and Seufert 1996), improperly chaperoned (Hayes and Dice 1996), and imprope rly post translationally modified proteins (Liu 1999) have been shown to be targeted for degradation by this mechanism. Another role of the ubiquitin proteasome pathway is its function in mediating the turnover of intact intracellular proteins. In gene ral, proteins degraded by the 26S proteasome are targeted for degradation via poly ubiquitination whereby identification and degradation of these substrates is extremely selective. Ubiquitin is a 76 amino acid peptide that is conserved among all eukaryote s and contains only three amino acid substitutions between yeast and human (Jentsch 1992). Ubiquitin is linked to the targeted protein by way of three classes of enzymes. First, the ubiquitin molecule is conjugated to the E1 ubiquitin activating enzyme b etween the ubiquitin carboxyl group and a cysteine residue located within the active site of the E1 enzyme (Ciechanover and Schwartz 1994). Ubiquitin E1 enzymes are encoded by 9 genes in humans and there are t hree known E1s in yeast. It i s believed tha t different E1s may be expressed in different cellular compartments or may interact preferentially with various E2 ligases or substrates (Hochstrasser 1996). Subsequently, ubiquitin is transferred from the E1 ubiquitin intermediate to an E2 ubiquitin comp lex. The E2 enzyme is also referred to as the ubiquitin conjugating enzyme or ubiquitin carrier protein and it becomes linked with ubiquitin at a specific cysteine residue. From here, the E2 can either donate the ubiquitin to an E3 ubiquitin ligase or th e E2 can catalyze
2 2 substrate ubiquitination (Jentsch 1992, Pahl and Baeuerle 1996). Thirteen genes encode yeast E2 enzymes and there are over 30 human E2 genes identified. E3 ubiquitin ligases have been shown to function in recognition of target proteins and recruitment of ubiquitin E2 complexes in order to catalyze formation of the iso peptide bond between ubiquitin and lysine residues of protein substrates (Ciechanover and Schwartz 1994). An exact method for classification and identification of E3 enzy mes is still imprecise but thus far, enzymes are classified as E3s if they have been shown to stimulate substrate ubiquitination when combined with the appropriate E1 and E2 enzymes and have been shown to bind both the E2 and the substrate proteins (Hochst rasser 1996). The interaction between E3s and substrate proteins are believed to be transient, thus further complicating their identification and classification. There are three subfamilies of E3 ligases that have been classified based on sequence homol ogy; the HECT domain containing E3s, the RING finger domain containing E3s, and the UBox E3s (Bernassola et al 2008). It is suggested that each subfamily has a specific set of substrates and within each subfamily there are further levels of classification resulting in even more precise targeting of substrates. It is the E3s that have been implicated in the extreme level of specificity in substrate recognition in turnover of intact cellular proteins by the 26S proteasome Lastly, E4 enzymes have more rec ently been shown to catalyze the addition of ubiquitin moieties to Lysine48 of the ubiquitin molecule that is substrate bound and form a multiubiquitin chain that is more efficiently targeted for degradation by the proteasome (Hochstrasser 1996, Koegl et a l 1999).
23 FIGURE 1.4: Ubiquitin Proteasome Pathway. Targeted degradation of intracellular proteins via the ubiquitin proteasome pathway is mediated by three classes of enzymes. The E1 ubiquitin activating enzyme binds ubiquitin and facilitates its transfer to the E2 ubiquitin conjugating enzyme. The E2 then donates ubiquitin to an E3 ubiquitin ligase. The E3 is typically bound to the substrate for degradation and therefore transfers ubiquitin to the target protein. E4 enzymes can function in ubiq uitin chain elongation and the ubiquitin chains are then recognized by the 19S regulatory subunit of the 26S proteasome complex. Lastly, the protein is proteolytically cleaved in the 20S catalytic domain of the proteasome. The mechanism by which substrat es are identified by these enzymes has been extensively investigated but still remains largely unclear. Some protein substrates contain se quence elements that serve as degradation signals. For example, the identity of the N terminal amino acid of protein s has been shown to correlate with its in vivo half life as mediated by the ubiquitin proteasome pathway (Varshavsky 1997). This effect is termed the "N end Rule" and suggests that arginine, lysine, histidine, phenylalanine, leucine, isoleucine, tryptopha n, and tyrosine are considered destabilizing N terminal residues in eukaryotes. The E3 ligase has binding sites for basic and bulky hydrophobic N terminal residues such that it interacts with the substrate and triggers ubiquitination. A
24 9 amino acid sequence called a destruction box has been shown to be required for turnover of cyclins that mediate cell cycle progression (King et al 1996). Another sequence that has been identified is called a PEST element since these regions are rich in proline, glut am ic acid, serine, and threonine. PEST regions have been implicated as targets for protein degradation (Rechsteiner and Rogers 1996). Finally, post translational modifications including phosphorylation of particular residues have been shown to be signal s for ubiquitination (Hochstrasser 1996, Kornitzer and Ciechanover 2000). The 26S proteasome, so named for its sedimentation coefficient, resembles a dumbbell such that the catalytic 20S proteasome appea rs as the cylindrical handle with two regulatory 19S s ubunits attached to opposite ends of the 20S subunit in an ATP dependent manner (Tanaka 1998). The 20S catalytic component of the proteasome is composed of 28 peptides, whereby f ourteen dimeric subunits are arranged in rings to form a barrel structure suc h that the catalytic sites are located on the interior of the cylinder and are shielded from the cytoplasm (Kornitzer and Ciechanover 2000). The 19S complexes, located on either end of the 20S complex have been implicated in recognition of ubiquitinated p roteins and serves as the entry point into the catalytic 20S complex (Ciechanover and Schwartz 1998). The 19S complexes contain ubiquitin chain binding proteins which bind with high affinity to ubiquitinated substrates. In addition to binding ubiquitinat ed proteins, the 19S complexes have the ability to unfold substrates in order to allow the proteins to enter the catalytic 20S co re for degradation (Kornitzer and Ciechanover 2000, Tanaka 1998). Another component of the 19S complex consists of a deubiquit inating enzyme that serves to free ubiquitin moieties from mulit ubiquitin chains for recycling (Tanaka 1998). Following deubiquitination, the protein is translocated into the 20S catalytic domain for destruction. The 20S proteasome contains six protease sites within the
25 barrel structure; these catalytic sites mediate degradation via an amino terminal threonine residue that undergoes a nucleophilic attack on a lysine residue of the target protein (Pahl and Baeuerle 1996). Substrates are subsequently cleav ed into small peptides ranging from 4 to 24 amino acids in length. Lactacystin, epoxomicin, and MG 132 are examples of inhibitors of the 26S proteasome that covalently bind to the reacting threonine residue with carboxy terminal aldehyde groups such that n ucleophilic attack is prevented (Pahl and Baeuerle 1996). Again, this pathway has been implicated in mediating ligand dependent and independent Ahr degradation in all model organisms thus far examined. At this time there are still many unanswered quest ions regarding the exact mechanism for Ahr degradation via the proteasome, most notably, the identification of the E3 ligase and its site for interaction within the receptor are still unknown entities. Additionally, receptor half life and mechanism of deg radation have yet to be analyzed in Saccharomyces cerevisiae despite a great deal of work on Ahr in this mo d el. Calpain Family of Proteases Calpains are ubiquitously expressed cysteine proteases that become activated in response to increased intracellul ar calcium levels. Specifically, calpain is activated in response to micromolar influxes of Ca 2+ while m calpain is activated with millimolar influxes of Ca 2+ (reviewed by Sorimachi et al 1997 and Goll et al 2003). Each enzyme consists of two subunits that function in Ca 2+ binding and proteolysis. There exists one calpain homologue in yeast called p83, it encodes a cysteine protease domain similar to the mammalian calpains, however it does not code for a calcium binding domain. Further analysis of th is protein is needed in order to characterize its function in yeast. One group published a report that calpains mediate Ahr turnover in a ligand dependent manner (Dale and Eltom 2006) in contrast to all of the evidence supporting a
26 26S proteasome mediated degradation event. This work was based on the hypothesis that Ahr ligand exposure causes an increase in intracellular calcium levels, thereby activating calpain proteases that subsequently degrade the receptor. These results were refuted by Pollenz (2007 ) who tested the effect of several calpain inhibitors on the Ahr signaling pathway showing no effect at the level of transactivation or degradation. Endogenous and Recombinant Expression of bHLH Proteins in Y east Endogenous bHLH Proteins in Yeast A number of genes encoding basic helix loop helix (bHLH) proteins have been identified in the yeast genome and their functions have been investigated due to their major role in gene regulation in higher eukaryotes. The yeast genome was compared to genes e ncoding mammalian and Drosophila bHLH proteins in order to identify yeast genes with potential homology (Robinson and Lopes 2000). While several genes were identified as containing bHLH domains, it should be noted that there are no homologues to any bHLH PAS transcription regulators. The first identified yeast gene shown to encode a bHLH protein, PHO4 acts as a transcription factor and activates several genes in response to phosphate starvation (Berben et al 1990). The Pho4 protein, which is encoded by t he PHO4 gene, binds to the core bHLH consensus sequence, 5' CACGTG 3', which is located upstream of several genes involved in phosphate uptake. INO2 and INO4 are yeast bHLH genes whose protein products (Ino2 and Ino4 ) have been shown to form a heterodimer both in vitro and in vivo but both are incapable of forming homodimers (Robinson and Lopes 2000). Ino2 and Ino4 bind at 5' CATGTG 3' sequences located upstream of more than 30 genes involved in phospholipid biosynthesis (Greenberg and Lopes 1996). Ot her yeast bHLH proteins have been shown to act as transcription factors in activation of genes required for glycolysis, genes involved in regulation of filamentous growth, and methionine biosynthesis. T he CBF1 ge ne encodes a bHLH protein and
27 binds a parti cular sequence located in centromeres. While this is atypical for bHLH proteins, the expression of CBF1 is required to maintain chromosomal integrity (Robinson and Lopes 2000 ) Expression of Heterologous Proteins in Yeast Saccharomyces cerevisiae can be us ed in order to characterize genes and proteins that are not endogenously expressed One major advan tage to using yeast as opposed to mammalian tissue culture for this type of analysis is that y east have the ability to homologously recombine similar sequences permanently into the genome. The double stranded DNA break repair pathway is activated when the PCR product is introduced into the cells, such that the repair mechanism searches for homology in the integrating sequence. This preference allows for integration of double stranded DNA of interest into the yeast genome in a highly directed and efficient manner (Ito et al 1983, Lorenz et al 1995). This recombination can facilitate either knocking a gene i nto the genome in a nonessential region or knocking a gene out simply by disrupting the target sequence or replacing it entir e ly. Expression of Mammalian bHLH Proteins in Yeast Saccharomyces cerevisiae has been used as a model organism in which non cons erved transcription factor signaling pathways have been reconstructed and subsequently dissected in order to evaluate various steps in gene regulation. Knock in y east strains have been generated that ex press members of the bHLH and bHLH PAS family of tran scription regulators including; Hif 1 (Braliou et al 2006), c Myc (Amati et al 1992, Escamilla Powers and Sears 2007, Hanel et al 1997), Ahr (Miller 1997, Miller et al 1998) The studies provided insight into various aspects of these signaling pathways; h owever, the majority of the
28 data analyzed consisted of reporter studies with little emphasis on analysis at the protein level. Hif 1 is a bHLH PAS receptor that mediates the cellular response to hypoxia. Braliou et al (2006) reconstituted the Hif 1 sign aling pathway in yeast by transforming yeast cells with HIF 1" and ARNT galactose inducible expression plasmids. R eporter studies confirmed Hif/Arnt heterodimerization and demonstrated transcription of a lacZ reporter by way of HRE (hypoxia response eleme nt) binding. The recombinant strain was also treated with Hsp90 inhibitors that resulted in reduced reporter activity; this effect can be attribut ed to an interaction between Hif and Hsp90 that is critical for transactivation. Finally, plasmids were cons tructed to express Hif with C terminal truncations in order to identify dominant negative mutants tha t would function in blocking Hif mediated transactivation (Braliou 2006). Importantly, these results were reported in terms of a series of reporter assays and did not evaluate the level of protein expression or the stability of the expressed proteins In the Amati et al (1992) study, the investigators generated a recombinant yeast strain expressing the mammalian c Myc and Max proteins. This study showed that c Myc requires its helix loop helix partner Max in order to interact with specific sequences of DNA containing the 5' CACGTG 3' core consensus sequence to transactivate target genes. In l ater studies Hanel et al ( 1 997) further investigated transactivation of c Myc and Max using reporters constructed with natural promoter elements. Similar to the experiments with Hif, these studies used reporters as an output for c Myc f unction in yeast and did not assess the level of protein expression or the stability of the expressed proteins. A more recent study of c Myc and Max has evaluated the stability of the proteins in a recombinant yeast model. Escamilla Powers et al (2007) u sed a c MycMax recombinant strain to evaluate th e phosphorylation state of Myc in conjunction with its stability. The results of this study suggest that Myc turnover is directly related to
29 the phosphorylation state of particular stabilizing and destabili zing residues confirming a similar effect has been demonstrated in mammalian cell culture. While Myc is a bHLH protein, it is not in the bHLH PAS family that contains the Ahr and Arn t Ahr Studies in Yeast Studies of the Ahr signaling pathway in yeas t have been used largely to elucidate the role of molecular ch aperones and co chaperones in Ahr function. For example, studies in yeast have evaluated the auxiliary proteins found in the cytoplasmic latent receptor complex such as: Hsp90 (Carver et al 199 4, Cox and Miller 2003, Whitelaw et al 1995), Xap 2 (Miller 2002), and p23 (Cox and Miller 2002, Cox and Miller 2004). Reports from these studies have demonstrat ed that the yeast homologues for Ahr chaperones generally play a conserved role in Ah receptor signaling. The first studies to assess Ahr signaling using a yeast model were published by Carver et al (1994). Human Ahr and Arnt proteins were cloned into a low copy yeast expression vector (CEN) and were constitutively expressed in a n Hsp 82 temperature sensitive yeast strain such that Hsp 82 levels could be modulated with increasing or decreasing tempera ture. Cultures were grown at the permissive and restrictive temperatures and a XRE driven lacZ reporter assay suggested that H sp 82 is an essential protein for Ahr signaling in yeast and simi larly, Hsp90 is required for Ahr signaling in higher eukaryotes. Work by Cox and Miller (2003) and Whitelaw et al ( 1995) further confirmed that Hsp 82 is an essential component of the reconstituted signaling pathway, again using the same galactosidase reporter assay in these studies. The level of Ahr and Arnt protein expression was never evaluated in these studies. In 1997, Miller generated a strain of yeast expressing human Ahr and Arnt by way of a bidirectional GAL1/GAL10 promoter. This promoter is activated with the addition of galactose to the growth medium and is turned off with the addition of glucose
30 to the media. The GAL1/GAL10 promoter was utilized to allow for greater control over protein expression levels in order to prevent toxicity due to overexpression; however, the cDNAs were cloned into a 2m high c opy number plasmid. The previous Ahr strains, expressed from plasmid encoding centromeric sequences likely contained 1 3 copies of the expression vector, while this strain carried 10 40 copies. No analysis of Ahr or Arnt protein expression was carried ou t in these studies. Interestingly, Miller observed constitutive reporter activity when the Ahr/Arnt expressing strain was transformed with the l acZ reporter plasmid. It was later suggested that the constitutive reporter activity was attributed to activat ion of Ahr via exogenous tryptophan in the culture medium (Miller et al 1998). While UV photoproducts of tryptophan have been implicated as potential ligands for Ahr (Rannug et al 1987), constitutive reporter activity was not observed in later studies whe n low copy number plasmids were used. Therefore, it is also plausible that constitutive activity resulted from gross overexpression of Ahr and Arnt proteins. Additionally, Miller (1997) grew the recombinant Ahr and Arnt strain in the presence of galactos e for approximately 24 hours prior to use in the galactosidase reporter assay. The promoter was transcriptionally active throughout the analysis and could result in excess of Ahr and Arnt proteins (Miller 1997). Interestingly, Miller later demonstrated that stable integration of Ahr and Arnt into the yeast genome conferred less constitutive activity in the reporter assay. These analyses confirm that protein expression levels should be evaluated in order to avoid overexpression that could lead to toxic effects or even aberrant results. In a follow up to the 1997 study, Miller used the same recombinant strain to evaluate t wo yeast proteins cont aining tetratricopeptide repeat domains, similar to the mammalian Xap2 protein. Kno c k out s of CPR6 and CPR7 g enes caused a significant reduction in lacZ reporter activity while transformation of a CPR7 but not a CPR6 expression vector, was able to restore Ah receptor signaling as measured by the
31 galactosidase reporter assay (M iller 2002). These results suggest that Cpr 7 is required for proper Ahr signaling and is the likely yeast Xap2 homologue; however the function of Cpr7 in yeast still remains unclear In 2002, the Miller group published a report on the co chaperone p23, demonstrating that knock out of the yeast homolog SBA1 resulted in a modest decrease in reporter activity. Given that yeast SBA1 and human p23 only have 28% sequence identity, transformation of a human p23 expressing plasmid into the SBA1 kn ock out strain recovered the Ahr signaling to approximately normal levels (Cox and Miller 2002). Although these studies neve r evaluated the level of expression of Ahr or Arnt protein in the recombinant strain, they did sugge st that the chaperones and co chaperones that are required for proper Ahr functioning in mamm a lian cells, are conserved and functional in yeast. The other type of studies that have been carried out in yeast are t ransactivation analyses that are useful studies for screening of potential ligands (Miller 1997, Miller 1999, Kawani shi 2003, Noguerol et al 2006, Ohura et al 2007, Sugihara et al 2008, Kamata et al 2009). Miller (1997) demonstrated a ligand specific dose dependent effect on transactivation as meas ured by the XRE driven reporter in a strain of yeast expressing Ahr and Arnt from a bi directional promoter The study measured reporter activity when cultures were treated with various concentrations of known Ahr ligands including TCDD (2,3,7,8 tetrachlorodibenzo p dioxin), !NF (! napthoflavone), HCB (hexachlorobenzene), B A P (benzo( a )pyrene as well as several potential Ahr ligands including tryptophan, IAA (indole acetic acid), IND (indole), I3C (indole 3 carbinol), and TA (tryptamine). In 1999, Miller demonstrated that there was an additive response when recombinant yeast were treated with a combination of aromatic and chloroaromatic hydrocarbons suggesting that there is a single li gand binding site within the Ahr protein. Kawanishi et al (2003) evaluated t he effect of !NF, TCDD, and BAP as was done
32 previously, but also ev aluated several other compounds (3 methylcholanthrene and indirubin) for Ahr transactivation using a similar recombinant strain. Additionally, the reporter activity was measured in a y east strain expressing mouse AHR and A RNT and was compared to a strain expressing the human AHR and ARNT genes to evaluate species specific variation in ligand sensitivity. The Ahr and Arnt expressing strain originally created by Miller (1997) was used as an in vivo model by several other groups in order to measure reporter a ctivity in respo nse to many other potential Ahr ligands. Noguerol (2006) tested 21 compounds that are known pollutants and antifouling pesticides. Ohura (2007) evaluated reporter activity resulting from treatment with 18 chlorinated polycyclic aromatic hydrocarbons while Sugihara (2008) tested six kno wn Ahr ligands for activity. Eighty four polychlorinated biphenyls (PCBs) with varying degrees of hydroxylation were evaluated for reporter activation; the report suggests that metabolic intermediates of PCBs have h igher binding affinities for Ahr and can therefore enhance Ahr transcriptional activity (Kamata 2009). In summary, there have been a number of important studies of Ahr signaling using recombinant yeast models. These studies clearly show the utility of yeast in the evaluation of Ahr and Arnt s ignaling and suggest that the requisite chaperones are expressed to support Ahr mediated signaling. There have been no studies that have taken a systematic approach to evaluate the level of Ahr and Arnt protein expression, degradation and stability in a r ecombinant yeast model. Summary of System and Specific Aims The aryl hydrocarbon receptor is a basic helix loop helix transcription factor shown to rapidly degrade following ligand activated transactivation of responsive genes. Ahr requir es its DNA bind ing partner (Arnt ), another bHLH transcription factor, in order to
33 associate with xenobiotic response elements in the promoter regions of target genes. The current model of Ahr signaling suggests that the hydrophobic ligand, typified by 2,3,7,8 tetrachlor odibenzo p dioxin, diffuses through the cell membrane and bin ds to the latent cytoplasmic Ahr comple x containing one molecule of Ahr a dim er of heat shock protein 90 (Hsp 90), a n immunophilin like protein (Xap 2), and the co chaperone p23 (Carver and Bradfi eld 1997, Kazlauskas et al 1999). Upon ligand binding, a conformational change exposes the nuclear localization signal such that the receptor translocates to the nuclear compartme nt and heterodimerizes with Arnt. Subsequently, the AhrArnt complex associ ates with enhancer elements found upstream of responsive genes (Denison et al 1988) including the gene encoding the drug metabolizing en zyme Cyp1a 1 (Legraverend et al 1982). Fin ally, after DNA binding, the Ahr but not the Arnt protein is quickly degraded by the 26S proteasome thereby attenuating the dose of gene regulation administered by the heterodimer (Pollenz 1996). Saccharomyces cerevisiae has been used as model organism in order to further characterize signal transduction pathways in a more simpli fied cellular setting in which the major processes required for growth and development are conserved to mammals. The yeast model has been used to evaluate Ahr signal transduction by several groups; however the majority of the published work has focused pr edominantly on determining the function of the auxiliary proteins that interact with the Ahr latent receptor complex (Carver et al 1994, Cox and Miller 2002, Cox and Miller 2003, Cox and Miller 2004, Miller 2002, Whitelaw et al 1995). Across these studies Ahr was expressed to varying degrees. The cDNA was expressed using a high copy plasmid in some studies, a low copy plasmid in others, and lastly the cDNA was stably integrated into the yeast genome. Additionally, various promoter sequences were used an d the promoter was cons t itutively active in some studies while it was inducible in others.
34 Although these reports laid the groundwork for investigation of the Ahr signaling pathway using S. cerevisiae as a model organism, almost every conclusion was made based on the results of an Ahr activated galactosidase reporter assay. The level of Ahr and Arnt protein expressed in the yeast strains was never quantified and the stability of the expressed proteins was never evaluated. Additionally, the ability of the Ahr to degrade in yeast was never explored. The overall goal of this work was to generate a recombinant yeast model that could be utilized to assess the ligand mediated degradation of the aryl hydrocarbon receptor (Ahr). The first aim was to generate novel yeast strains with mammalian AHR and ARNT cDNAs integrated into the yeast genome. The second aim was to evaluate Ahr and Arnt protein expression and evaluate the stability of these proteins when expressed heterologously in yeast. The third aim was to evaluate Ahr and Arnt function through analysis of a ligand mediated XRE driven reporter. Finally, the fourth aim was to characterize ligand dependent and independent Ahr degradation.
35 CHAPTER TWO: GENERATION OF AHR AND ARNT EXPRESSING YEAST STRAINS Experimental Question and Rationale Yeast can be used to characterize genes and proteins that they do not normally express. One major advantage to using yeast as opposed to mammalian tissue culture is that y east have the abi lity to homologously recombine similar DNA sequences permanently into their genome. The double stranded DNA break repair pathway is activated when a portion of double stranded DNA (PCR product) is introduced into the cells, such that the repair mechanism searches for homology to the yeast genome within the integrating sequence. This preference allows for the integration of the double stranded DNA of interest into the yeast genome in a highly directed and efficient manner (Ito et al 1983, Lorenz et al 1995) Recombination in yeast can therefore facilitate either knocking a gene into the genome in a nonessential region (gain of function) or knocking a gene out by disrupting the target sequence or replacing it entirely (loss of function). The aim of this port ion of the project was to integrate mammalian AHR and ARNT cDNAs into yeast strains in order to heterologously express these genes. A major advantage of this approach as compared to others was to utilize directed homologous recombination of the cDNAs into the yeast genome so that only one copy of each cDNA was integrated into the yeast genome. This method was chosen in order to mimic physiological levels of a given yeast protein and prevent overexpression and cellular
36 stress. The ultimate question asked whether or not Ahr and Arnt proteins could be detected in yeast using this type of expression system. General Strategy. For these studies, the exogenous genes will be referenced in all capital letters (i.e. AHR, ARNT). The proteins expressed from the e x ogenous genes will be referenced in upper and lowercase (i.e. Ahr, Arnt). Yeast genes will be referenced using th e standard yeast nomenclature (a ll capital letters and italics). The yeast genome does not encode AHR and ARNT homologues; therefore, yeast strains were constructed such that mouse AHR and ARNT cDNAs were stably integrated into the genome adjacent to an inducible promoter for subsequent expression analysis. Yeast strains were constructed according to a precise plan that would allow the Ahr and Arnt expressing strains to be of opposite mating types. This plan also worked to preserve the selectable markers and to prevent addition of tr yptophan to the growth medium to avoid erroneous activation of Ahr (Rannug et al 1987). In order to facilitate the uptake of the cDNAs, selectable marker cassettes were cloned into plasmids adjacent to AHR and ARNT cDNAs. The parental yeast strain s are auxotrophic for several commonly used yeast selectable markers. The strain s are therefore unable to grow in the absence of certain compounds whose production is dependent on the expression of those genes encoded by the marker cassette s The cDNAs and th eir adjacent marker cassettes were PCR amplified. The resulting PCR product was transformed into yeast cells treated with lithium cations in order to facilitate uptake of the DNA into the yeast cells This procedure is described in detail in Chapter 5. The transformed cells were spread on synthetic medium lacking the particular compound whose synthesis occurs with expression of the selectable marker. This allowed for select ion of clones that have taken up the cDNAs and adjacent markers The resulting c olonies were tested for integration of cDNAs using PCR.
37 Construction of Ahr and Arnt Expressing St r ains Cloning of Selectable Marker Cassettes into AHR and ARNT Vectors The stable integration of the AHR or ARNT cDNAs into the yeast genome required that a selectable marker cassette was incorporated as part of the transformed sequence. The pRS304 plasmid (Sikorski and Hieter 1989) containing the TRP1 marker cassette enco des a gene required for tryptophan synthesis. The pRS306 plasmid (Sikorski and Hieter 1989) containing the URA3 marker cassette encodes a gene required for the synthesis of uracil. The AHR and ARNT expression vectors and TRP1 and URA3 sequences were eva luated in order to identify particular re striction enzymes that would cleave the plasmids immediately downstream of the cDNAs but would not internally cleave the marker cassettes. For the AHR expression vector, pLNcx2 AHR b 2 (Pollenz and Dougherty 2005 ), the ClaI site is 54 base pairs downstream of the AHR cDNA and does not cut within AHR or the TRP cassette. The ARNT vector, pCDNA ARNT (Dougherty and Pollenz 2008) has an AflII restriction site located 72 b ase p airs downstream of the ARNT cDNA and this enzyme does not cleave ARNT or the URA3 marker cassette. The TRP1 cassette was PCR amplified using primers designed to anneal to the cassette but also contained overhanging ClaI restriction sites. The pLNcx2 AHR b 2 ve ctor and the TRP1 PCR product were digested with ClaI restriction enzyme. Similarly, the URA3 cassette was PCR amplified using primers with overhanging AflII restriction sites. The pCDNA ARNT vector and the URA3 PCR product were digested with AflII restr iction enzyme. The linearized vectors were treated with alkaline phosphatase in order to prevent re ligation of empty vector. Next, each vector and respective marker cassette was combined with DNA ligase. A schematic of the ligated plasmids is shown in Figure 2.1A, these plasmids were subsequently transformed into competent E. coli
38 and purified from clonal populations of bacterial cells and evaluated for direction of insert i on. PCR Amplification of cDNAs and Adjacent Marker Cassettes. Linking the cDNAs with the yeast selectable marker cassettes provides a means for selection of integrated clones whereby transformed yeast cells are plated on media lacking a particular compound required for growth Therefore, only yeast that s t ably integ rat e the cDNA and its linked marker cassette into their genome will have the ability to produce that particular compound and therefore survive and grow on the media. FIGURE 2.1: Plasmid maps depicting PCR primers and the result of PCR amplification. (A) Plasmid maps illustrate the ligation of selectable marker cassettes downstream of AHR and A RNT cDNAs. The arrows represent PCR primers designed to amplify the cDNA and the s electable marker cassette in order to produce one double stranded DNA fragment. The primers also contain 50bp overhangs with homology to the integration site within the yeast genome. (B) The PCR product from amplification of the AHR. TRP1 sequence and the ARNT. URA3 after resolving on a 1% agarose gel. This double stranded DNA was then transformed into parental yeast strains to generate recombinant yeast for expression analysis. PCR primers were designed to anneal to the 5' end of the AHR or ARNT cDNA and the 3' end of the marker cassette while also containing 50 b ase p air overhangs with sequence identity for the target recombination locus within the yeast genome on each
39 primer, see Figure 2.1A. The primers were designed such that the sequences recombined to nonessential chromosomal regions within the genome of the parental strains, specifically a region of chromosome 15 for Ahr and chromosome 4 for Arnt. The cDNAs and adjacent marker cassettes were PCR amplified and this PCR product, sho wn in Figure 2.1B served as a target for homologous recombination in the parental yeast stra i ns. Transformation Procedure. Yeast transformations were carried out according the the method of Gietz and Woods (2006) and is detailed in Chapter 5. The KHSY421 yeast strain ( MATa, ura3 52, trp1 63, his3 200, leu2 1 ) was streaked on YPD agar media and incubated at 30¡C for two days. On the third day, one colony was used to inoculate a liquid culture that was placed in a shaking incubator overnight at 30¡C. The overnight culture was used to make a 25 milliliter culture at an approximate OD 600 of 0.2. This culture was placed in the shaking incubator and grown to an OD 600 of 0.8. The cells were pelleted and combined with lithium acetate, PEG (polyethylene glycol), the AHR. TRP PCR product, and boiled/snap cooled salmon sperm DNA in order to facilitate the uptake of the DNA. Following heat shock, the cells were spread on media lacking tryptophan in order to select for yeast with the integrated cDNA and TRP marker. After two days, the plates shown in Figure 2.2A were evaluated for growth and each colony was streaked on another selective plate in order to obtain individual clones. These clones were grown overnight in liquid medium, genomic DNA was extracted (Invitrogen) and PC R was used to confirm pro per integration of the cDNA. Specifically, one primer was designed to anneal 100bp upstream of the integration site while the second primer annealed approximately 900bp inside the AHR cDNA resulting in a 1000 base pair PCR product in strains that conferred proper integration. The PCR products were resolved on a 1% agarose gel, with the results shown in Figure 2.2B. Several
40 positive clones were identified, and stored for future use. Specifically, the AHR. TRP integrated clone call ed KHSY1535 ( MATa, C hrXV: :AHR.TRP1, ura3 52, trp1!63, his3!200, leu2!1 ) was used for subsequent experiments. The ARNT strain was generated in precisely the same manner however the ARNT. URA PCR product was transformed into the KHSY422 parental strain ( MAT", ura3 52, trp1 63, his3 200, leu2 1 ). Several ARNT transformants were identified and their integration was verified using PCR (Figure 2.2B); these clones were also catalogued and stored for the next set of experiments. The ARNT. URA integrated clone, KHSY1540 ( MAT", C hrIV: :V5 ARNT.URA3, ura3 52, trp1!63, his3!200, leu2!1 ) was used in later experiments. FIGURE 2.2: Transformed yeast colonies on selective media and confirmation of cDNA integration. (A) The parental strains KHSY421 and KHSY422 were transformed with the PCR amplified cDNAs/marker cassettes AHRb2. TRP1 and ARNT. URA3 The transformed strains were selected for incorporation of the DNA using media lacking either tryptophan (421) or a compound required for uracil synthesis (422). (B) PCR confirmation for proper integration of AHR and ARNT genes. PCR amplification of g enomic DNA prepared from twelve AHR transformants and twelve ARNT transformants revealed that 9 AHRb2. TRP1 transformants and five ARNT. URA3 transformants were properly integrated into the yeast genome, as indicated by asterisks. O ne AHRb2. TRP1 and one ARN T. URA3 transformant was chosen for the next step in strain construct i on.
41 Integration of the Inducible Promoter. The galactose inducible promoter sequence was inserted in the yeast genome directly upstream of each cDNA in order to drive expression of the AHR and ARNT genes. The same experimental design was employed such that PCR amplification of the GAL1 promoter and the adjacent G418 cassette from the pFA6a PGAL1KanMx plasmid (Longtine et al 1998) resulted in double stranded PC R product with 50bp overhan ging sequence with identity to the yeast sequence immediately upstream of the AHR or ARNT integration site. This sequence was transformed into the AHR. TRP (1535) and ARNT. URA (1540) strains using the lithium acetate approach (Gietz and Woods 2006) and tra nsformed yeast were spread on media selecting for G418 resistance. Integration of the promoter was confirmed via PCR for several clones, and each positive clone was stored. The GAL .AHR. TRP clone, namely KHSY1538 ( MATa, C hrXV: : KanMX PGAL1 AHR.TRP1, ura3 52, trp1 63, his3 200, leu2 1 ) and the GAL .ARNT. URA clone, KHSY1541 ( MAT", ChrIV:: KanMX PGAL1 V5 ARNT.URA3, ura3 52, trp1 63, his3 200, leu2 1 ) were utilized in the next step of strain construction whereby they were mated in order to generate a strain e xpressing both Ahr and Arnt proteins under the inducible promoter. Next, KHSY1538 and KHSY1541 strains were streaked on medium containing glycerol as the carbon source in order to check for the petite (% ) mutation. Yeasts carrying the petite mutation ha ve little or no mitochondrial DNA and form small anaerobic colonies on non fermentable growth media containing either glycerol or ethanol (Sherman 2002). Both strains exhibited normal growth on YPG (glycerol) plates and were then used in the generation of the double knock in strain. Prior to mating the Ahr expressing KHSY1538 strain with the Arnt expressing KHSY1541 strain, both strains were further evaluated to determine if they retained a wild type doubling time of approximately 90 minutes in YPD media ( Sherman 2002).
42 Doubling times were measured according to the protocol described by Schmidt and Kolodner (2004). FIGURE 2.3: Growth Curves and Doubling Times for KHSY1538 and KHSY1541 strains. Ahr and Arnt knock in strains were evaluated for doubling time. Three colonies from each strain were grown in YPD media for eight hours with a 1ml sample measured for optical density at 600nm every hour. Optical densities were plotted versus time for samples in the exponenti al growth phase and the equation from each linear regression was used to determine the doubling time for each yeast sample. The average of three rates was taken for each strain such that the doubling time for KHSY1538 (A) is 95 minutes and the doubling ti me for KHSY1541 (B) is 102 minutes. Each strain was streaked for single colonies on YPD media and after two days, three single colonies of each strain were used to inoculate individual 2ml liquid overnight cultures. The next day, 20ml cultures were prepared at an OD 600 of 0.2 for each of the
43 six overnight cultures. These cultures were placed in a 30¡ shaking incubator, and 1ml was measured for optical density at 600nm every hour for eight hours. The optical densities, once the cultures reached exponential growth, were plotted as a semi log regression using excel such that the slope of each line represented the growth rate for each colony, as shown in Figure 2.3. The doubling time for the KHSY1538 and KHSY1541 strains were then calculated using the equation of each regression line and the average of three was taken for each strain. The Ahr knock in strain revealed an approximate doubling time of 95 minutes with a standard deviation of 2 minutes, while the Arnt knock in stra in had a doubling tim e of approximately 102 minutes with a standard deviation of 1.7 minutes. Generation of the Double Knock In Strain. The Ahr protein requires Arnt heterodimerization in order to bind XREs and therefore induce target gene expression (Reyes et al 1992), such that a strain needed to be constructed in which both Ahr and Arnt were expressed. The previous transformations were designed so that KHSY1538 and KHSY1541 would be of opposite mating types and crossing the strains could produce one strain expressing both Ahr and Arnt. Mating of two strains was carried out according to Sherman ( 2002 ) and required that one colony of each stra in, 1538 and 1541, was mixed together in water. Ten microliters of that solution was spotted onto a YPD plate and placed in the 30¡C incubator overnight. Next, the growth was streaked out for single colonies on another YPD plate and again grown overnight. The resulting colonies were genotyped to determine mating type as MATa MAT" or diploid. Results for the mating are shown in Table 2.1. To test for di ploid strains, colonies were tested for growth on synthetic media to assess complementation of auxotrophic markers. In order to do this, a lawn of strain KHSY1435 ( MATa, thr4 ) and KHSY1436 ( MAT thr4 ) wa s grown on YPD agar med ia
44 and placed in the 30¡C incubator overnight. These two haploid strains are of opposite mating types and both strains are auxotrophic for the threonine selectable marker. Single colonies o f the 1538/1541 mixture were spotted onto the KHSY1435 and KHSY1436 lawns. Since haploid str a ins ( a or ") can only mate with their opposite mating type, it was expected that any 1538 (a) colonies would mate with KHSY1436 (") and 1541 (") would similarly mate with KHSY1435 (a) to produce a/" diploid strains. Any 1538/1541 colonies that formed diploids during the initial strain mating were therefore already (a/") and were unable to mate with the lawn of KHSY1435 or KHSY1436 The following day, a replica of each plate was made on minimal medium. This ensured that only 1538/1 436 diploids and 1541/1 435 diploids could grow on the plates while any 1538/1541 diploids formed during the initial strain mating would again be unable to grow. The KHSY1435 and KHSY1436 strains are auxotrophic for the threonine biosynthesis s electable marker such that they are un able to grow on minimal medium unless mated to 1538 or 1541 and genetic complementation will sustain growth While the 1538 and 1541 strains require the addition of supplemental histidine, and leucine to grow. Forma tion of a diploid with 1 435 or 1 436 allowed the strains to grow in the absence of supplemental amino acids. Therefore, all colonies that did not grow on the minimal medium were diploids generated in the original crossing of 1538 and 1541 and these colonie s were next subjected to random spore isolation. The resulting minimal plates were evaluated for cell growth. Again, growth on neither plate suggests that the sample was already heterozygous for the MAT locus, and was therefore diploid prior to spotting o n the KHSY1435 and KHSY1436 lawns. It is those diploid strains that are useful, as they carry two sets of chromosomes and should encode both AHR and ARNT cDNAs and promoter sequences. The mating results are shown in table 2.1 below. Thirty one diploid clones were observed and nine clones were
45 of the MATa mating type. Three of the diploid strains were stored for future use and were used for random spore isolation. FIGURE 2.4 : Mating type genotyping for AHR/ARNT Crossing. (A) Forty single colonies were picked after mating KHSY15 38 with KHSY1541 strains. The colonies were spotted onto rich medium first and were subsequently spotted on a lawn of KHSY1435 and KHSY1436 (B) Growth of certain colonies on minimal media results from mating with the 1436 ( MAT ) lawn, suggesting that t hose colonies must be of the opposite mating type ( MAT a ) and are therefore haploid 1538 yeast. (C) Conversely, no growth was observed on the 1435 plate and therefore no 1 541 MAT" haploid colonies were obtained. The colonies that did not exhibit growth o n either p late is therefore a 1538/1541 diploid. Lastly, meiotic progeny of the mated strains was produced such that one haploid strain encoding both AHR and ARNT was generated via random sporulation ( Rockmill et al 1991 ) Three diploids were chosen for sporulation and grown in th e non fermentable carbon source 1% potassium acetate for 5 days. Growth in potassium acetate causes a shift from mitotic cell division to meiosis resulting in mature asci containing four haploid ascospores (Sherman 2002). These spores were isolated using the enzyme zymolase followed by sonication and were spread on rich medium for genotyping. The goal was to identify at least one clone containing chromosome 15 from the 1538 strain in combination with chromosome 4 of the 1541 st rain such that both AHR and ARNT genes could be expressed in combination.
46 TABLE 2.1: Mating T ype Genotyping of Crossed Strains Clone # MAT Clone # MAT Clone # MAT Clone # MAT 1 diploid 11 diploid 21 diploid 31 diploid 2 diploid 12 diploid 22 diploid 32 A 3 diploid 13 diploid 23 diploid 33 diploid 4 diploid 14 diploid 24 diploid 34 diploid 5 a 15 diploid 25 diploid 35 diploid 6 diploid 16 a 26 diploid 36 diploid 7 diploid 17 a 27 diploid 37 diploid 8 diploid 18 diploid 28 a 38 diploid 9 diploid 19 diploid 29 a 39 A 10 diploid 20 diploid 30 a 40 A The Ahr and Arnt expressing strains were mated in order to generate one strain expressing both proteins. After mixing one colony of each strain, the ye ast was streaked for single colonies and genotyped for mating type. Nine clones gave rise to MATa strains, while the remaining clones gave rise to diploid strains. Three diploid strains, 7, 23, and 32 were stored for future use (bolded entries). Forty e ight haploid clones wer e evaluated for AHR, ARNT, and GAL1 by way of selection for their respective auxotrophic markers. The clones were spotted on media lacking tryptophan to identify clones with AHR. TRP 1, while media lacking uracil identified clones with ARNT. URA3 Additionally, the clones were tested for G418 resistance that should occur in conjunction with the heterologously integrated GAL1 promoter. Clones that grew on all three selective plates, therefore, encoded both AHR and ARNT cDNAs and exp ression of these genes was inducible by way of the integrated GAL1 promoters. Images of the selective plates are shown below in Figure 2.5. While the random sporulation procedure generally induces spore formation in the majority of the cells, it was important to determine the mating type of the clones to ensure that a diploid is not selected for future analysis. While a diploid strain would still express both AHR and ARNT, it is not as useful because it can no l onger mate with other strains. M ating t ype genotyping was carried out as shown previously and the plates are shown in Figure 2.6.
47 FIGURE 2.5 Genotyping of h aploid spores Forty eight clones were plated on media lacking tryptophan (A), uracil (B), and media containing canavanine to test for G418 resistance (C). Growth on all three plates indicates expression of the selectable marker downstream of the cDNAs of interest. FIGURE 2.6 Mating type gen otyping after sporulation of Ahr and Arnt expressing strains. Clones were spotted on a lawn of 1 435 or 1 436 yeast and plated on minimal media in order to determine their mating type. (A) Growth on minimal media coated with a lawn of 1435 (MAT a), suggests that the colonies are of opposite mating type (MAT ") and are therefore h aploids. (B) Grow t h on the 1436 (MAT") lawn, suggests that those colonies are o f the opposite mating type (MAT a) and are also haploids. Conversely, where no growth was observed on either plate, those clones were diploid. Table 2.2 displays the genotyping results for the 48 spores that were analyzed. Several haploid clones grew on all three selective plates and were selected for future analysis; in particular clones 27, 33, and 35 were stored for use in expression analysis. Clone 33, also known as KHSY1547 ( MATa, C hr XV: : Kan MX 6 PGAL1 AHR.TRP1, ChrIV:: Kan MX6 PGAL1 V5 ARNT.URA3, ura3 52, trp1 63, his3 200, leu2 1 ) is the double knock in strain used in subsequent studies. Again, KHSY1547 was streaked on medium containing glycerol as the carbon source in order to check fo r the petite (% )
48 mutation. This strain exhibited normal growth on YPG (glycerol) plates and was then used in the generation of the reporter strain. TABLE 2.2: Genotyping and Mating type following Random Sporulation Spore # MAT ::G418 ::URA ::TRP Spore # MAT ::G418 ::URA ::TRP 1 alpha x x 25 alpha x x 2 alpha 26 diploid x x 3 alpha 27 a x x x 4 alpha x x 28 alpha x x 5 alpha x x 29 alpha x x 6 alpha x x 30 a x x 7 alpha x x 31 alpha x x 8 alpha x x 32 a x x 9 a x x 33 a x x x 10 alpha x x 34 alpha 11 alpha x x 35 a x x x 12 alpha x x x 36 a 13 diploid x x 37 alpha 14 alpha x x 38 diploid x x 15 alpha x x 39 diploid x x x 16 a x 40 a x x 17 diploid x x x 41 alpha 18 alpha x x 42 alpha x x x 19 a x x 43 a 20 diploid x x x 44 a 21 diploid x x 45 alpha x x 22 alpha x x 46 a x x 23 diploid x x x 47 alpha x x x 24 diploid x x 48 diploid This table displays the results for the 48 clones analyzed following random sporulation. Each clone was evaluated for mating type and growth on selective media in order to determine which clones express both AHR and ARNT cDNAs, express the GAL1 promoter, and are MAT a haploids. Clones 27, 33, and 35 grew on all thr ee selective plates and are MAT a haploids, and these clones were stored for future use. The last step in strain construction required the transformation of the pLXRE5 Z reporter plasmid into the double knock in as well as the AHR expressing strain (Cox and
49 Miller 2003 ). This plasmid has five xenobiotic response elements located upstream of t he lacZ gene, such that binding of Ahr/Arnt heterodimers to the XREs will induce expression of lacZ reporter gene. The plasmid also carries a LEU2 marker cassette for selection and a centromeric origin of replication that maintains the plasmid at a low co py number. FIGURE 2.7 Plasmid map for pLXRE5 Z reporter plasmid. The plasmid was transformed into the double knock in strain KHSY1547 and the AHR expressing strain KHSY15 38. The repor ter (Cox and Miller 2003) was used in these strains to evaluate transactivation of the recombinant signaling pathway. The plasmid was transformed into the double knock in strain as well as the AHR expressing strain using the lithium acet ate method described previously; however, 100ng of plasmid DNA was substituted fo r double stranded DNA. Selection was carried out on medium lacking leucine and positive transformants were selected and stored as frozen stocks. The completed strain, KHSY1566 ( MATa, C hrXV: : Kan MX 6 PGAL1 AHR.TRP1 ChrIV: :Kan MX 6 PGAL1 V5 ARNT.URA3, ura3 52, trp1 63, his3 200, leu2 1, pLXRE5 Z ) therefore expresses AHR and ARNT under the inducible GAL1 promoter and contains the pLXRE5 Z reporter plasmid. Additionally, the KHSY1565 ( MATa, C hrXV: :Kan MX 6 PGAL1 AHR.TRP1, ura3 52, trp1 63, his3 200, leu2 1, pLXRE5 Z ) strain expresses AHR under the inducible GAL1 promoter and contains the reporter plasmid. While only the double knock in strain should demonstrate XRE binding
50 capability and reporter activation, the AHR knock in strain was generated in order to confirm specific activation of the reporter as mediated by the signaling pathway as it exists in mammalian cells. KHSY1566 was streaked on medium containing glycerol as the carbon source in order to check for the petite (% ) mutation, as described prev iously. Again, this strain exhibited normal growth on YPG (glycerol) and was stored for use in future experiments. It was also of interest to determine if this strain retained a wild type doubling time of approximately 90 minutes in YPD media The cultu re was evaluated for doubling time as descr ibed for KHSY1538 and 1541 (Schmidt and Kolodner 2004) The optical densities, once the cultures reached exponential growth, were plotted as a semi log regression using excel such that the slope of each line repr esented the growth rate for each colony, as shown in Figure 2.8. The doubling time was then calculated using the equation of each regression line and the average of three was taken. The results shown in Figure 2.8 show that the KHSY1566 strain has an ap proximate doubling time of 106.7 minutes with a standard deviation of 2.8 minutes. This doubling time is slightly longer than the wild type and longer than the doubling time of the previously constructed strains. It should be noted that this strain, carr ying the reporter plasmid, is propagated in synthetic medium which is likely the reason for this observation.
51 FIGURE 2.8: Growth Curve and Doubling Time for KHSY1566. The double knock in reporter strain was evaluated for doubling time. Three colonies were grown in YPD media for eight hours with a 1ml sample measured for optical density at 600nm every hour. Optical densities were plotted versus time for samples in the exponential growth phase and the equation from each linear regression was used to det ermine the doubling time for each yeast sample. The average of three rates was taken and revealed that the doubling time for KHSY1566 is 106.7 minutes. In summary, this set of studies demonstrates that several novel recombinant yeast strains have been p roduced that may be utilized to assess Ahr signal transduction and Ahr degradation. These strains have been validated using PCR to contain the cDNAs and reporter constructs that will recapitulate the Ahr signal transduction pathway. In addition, studies show that the integration of these constructs does not appear to impact the growth of the different strains and these results indicate that it is appropriate to pursue further analysis. The studies presented in Chapter 3 provide a comprehensive analysis o f Ahr signal transduction in several of these novel strains.
52 C HAPTER THREE: CHARACTERIZATION OF AHR AND ARNT PROTEIN EXPRESSION IN YEAST STRAINS AND VALIDATION OF YEAST MODEL Experimental Question and Rationale The central goal of these studies was to use Saccharomyces cerevisiae as a model to assess Ahr protein degradation. Therefore, after completing the construction of the Ahr and Arnt expressing yeast strains (Aim #1) ; it was essential to demonstrate that th e proteins would be detectable using western blotting and antibody staini ng. As stated in chapter one Saccharomyces cerevisiae has been used as a model for Ahr signaling however, Ahr and Arnt protein levels were not evaluated in these studies a nd reporte r assays were utilized t o measure receptor activation (Carver et al 1994, Cox and Miller 2002, Cox and Miller 2003, Cox and Miller 2004, Miller 2002, Whitelaw et al 1995). There are no studies that have evaluated recombinant Ahr and Arnt protein expression in yeast. This chapter is broken down into several sections that encompass the various aims of the study. The first section describe s the set of experiments ai med to optimize induction conditions in order to visualize the Ahr and Arnt proteins using western blotting (Aim #2). Next, the function of the Ahr and Arnt proteins was tested using the XRE reporter gene as an output (Aim #3) Once it was determined tha t the Ahr and Arnt could bind XREs and drive reporter expression, a series of experiments were conducted to assess Ahr and Arnt stability (Aim #2). Finally, studies aimed to assess the ligand dependent and independent degradation of the Ahr are presented (Aim #4).
53 General Strategy. Single knock in strains expressing Ahr (KHSY1538) and Arnt (KHSY1541) were used in the initial western blotting experiments to confirm that the protein levels were detectable prior to mating the strains. Expression of Ahr and Arnt proteins was induced with the addition of galactose to the culture medium for various time points and protein samples were prepared from the yeast cells. These samples were then used in western blotting and antibody staining using Ahr and Arnt speci fic antibodies. Upon detection of the proteins of interest in the single expression strains, the double knock in strain (KHSY1547) was generated as described in chapter two. Finally, Ahr and Arnt protein levels were evaluated whereby both proteins could be expressed in a single yeast strain. Reporter assays were carried out next. These studies were carried out to show that Ahr and Arnt were able to heterodimerize and bind the XRE containing promoter of the reporter plasmid. Several strains were produc ed to carry out these analyses. The KHSY1566 strain expressed both Ahr and Arnt and was transformed with the XRE reporter plasmid. The KHSY1565 strain expressed only Ahr and was also transformed with the XRE reporter. Both strains were tested for report er activation following treatment with TCDD or DMSO. Activation with TCDD was tested at a single time point to confirm that the strain expressing both Ahr and Arnt, and not the strain only expressing Ahr was able drive reporter activity. Later, the KHSY1 566 strain was used to test other ligands at various doses and time points. Finally, the Ahr and Arnt proteins were tested for stability in the presence of Ahr ligand since the results of the reporter assays provided evidence of a functioning Ahr signaling pathway in yeast. The half life of the proteins was evaluated first to provide a baseline for the ligand treatment studies. Both the single and double knock in strains were used to determine the half life of the Ah receptor in yeast. Since Ahr and Arnt expression is driven by an inducible promoter, expression was induced for a time period
54 and subsequently turned off to evaluate the stability of the Ahr and Arnt proteins over time. The results of the half life study led to an additional series of experi ments that involved various methods of induction and sample preparation. The stability of Ahr was tested under various conditions including; reduced galactose in the growth medium, reduced induction time, and different sample preparation methods. Finally the Ahr and Arnt expressor strains were treated with ligand under various conditions to visualize a degradation event in the yeast cells. Detection of Ahr and Arnt protein expression in yeast strains In the previous studies of Ahr and Arnt in yeast the Ahr and Arnt proteins were induced with the addition of 2% galactose to the growth medium for approximately 16 hours and then the cultures were treated with ligand for an additional 8 hours to activate Ahr (Cox and Miller 2002, Miller et al 1998, Mille r 1997) Induction of Ahr expression followed by ligand treatment allowed for heterodimerization of Ahr with Arnt and subsequent DNA binding and reporter activation. While this experimental paradigm was effective and reporter activity was detectable for the strains tested, the level of receptor protein ( having been induced for a total of 24 hours ) was not evaluated at any time point. Since these studies showed that TCDD induced galactosidase reporter could be detected Ahr and Arnt protein must have be en present in the samples tested. Therefore, we decided to mimic the 16 hour induction in the first set of studies. Western blotting studies were carried out in th e KHSY1538 and KHSY1541 strains. These strains are single recombinants and KHSY1538 express es only Ahr and KHSY1541 expresses only Arnt. This experiment was completed prior to mating the strains in order to confirm the ability of each strain to express their respective genes/proteins. The rationale of this approach was to confirm expression in the individual strains before producing the double knock in str a in
55 Induction of Ahr and Arnt proteins in single knock in yeast strains. The first set of experiments was designed for detection of Ahr and Arnt protein expression by western b lotting in the single knock in strains. The inducible GAL1 promoter was inserted upstream of the AHR and ARNT c DNAs during strain construction, as detailed in chapter two. This promoter sequence, also called UAS G is the binding site for the yeast Gal4 t ranscription factor in the presence of galactose or glycerol. When glucose is present, the UAS G is bound by the Gal80 repressor protein which acts to block transactivation (reviewed by Lohr et al 1995). Therefore, the GAL1 promoter upstream of the cDNAs is silent in the presence of glucose while Ahr and Arnt expression is activated upon addition of galactose to the growth medium. For these experiments, induction of the AHR and ARNT mRNA expression simply required the removal of glucose and the subsequent addition of galactose to liquid cultures. Expression of Ahr and Arnt proteins should begin immediately upon removal of glucose when galactose is present. In the first set of experiments, yeast cultures were grown in the presence of galactose for 16 h ou rs (overnight) because this was the typical time course used in previous reports using recombinant yeast strains (Cox and Miller 2002, Miller et al 1998, Miller 1997) Total protein was prepared from the induced cultures (2% galactose) and non induced contr ols (2% glucose) using 20% trichloroacetic acid (TCA) The TCA precipitation procedure (Wright et al 1989) is described in detail in Chapter 5. To assure that equal levels of protein were being evaluated at each time point, the culture density was quanti fied at OD 600 and the same number of cells was harvested. Briefly, e ach culture was centrifuged at 2000 rpm for 2 minutes and the medium was aspirated The cell pellets, containing 1.0 ODs of cells, were combined with 100l of 20% TCA and an equivalent volume of glass beads and were placed in a mini bead beater for 4 minutes. Each sample was transferred to a fresh tube which was then centrifuged at 14 ,000 rpm
56 for one minute. The resulting protein pellet was resuspended in 100l of SDS PAGE sample loading buffer, and was then r esolved on an SDS PAGE gel for western blotting. Once protein samples were prepared, western blotting was carried out by resolv ing e qual volumes of the yeast protein samples on a 7% SDS polyacrylamide gel The proteins were transferred to nitrocellulose using a semi dry blotting apparatus. To confirm even sample loading nitrocellulose membranes were stained with P oncea u S dye. The detection of Ahr and Arnt was carried out using polyclonal antibodies characterized by Pollenz et al (1994 ). Ahr and Arnt induction experiments were completed several times for each induction that is presented below and r epresentative western blot s a re shown in Figure 3.1 The results show that both Ahr and Arnt could be detected in the yeast strains following overnight induction. The specificity of the Ahr and Arnt band is confirmed by the lack of reactivity in the negative control and the migrati on of the reactive band from the yeast samples at the same molecular mass as the positive control. To further assess the time course for the induction of Ahr and Arnt, c ultures were grown overnight in glucose, spun down, the media was removed and the pell ets were resuspended in galactose containing media for 2 to 6 hours. Samples containing equivalent cell numbers (1.0 ODs) were harvested every two hours for six hours and protein was prepared using the 20% TCA precipitation method as described above. A re presentative western blot for this experiment is shown in Figure 3.1B The results show that Ahr protein is detected at the 2 hour time point. T he level of Ahr protein does not appear to increase above the level observed at 2 hours, even after treatment with galactose for 6 hours. It is important to note that each lane was loaded with protein that was extracted from the same number of cells. Thus, the expression level in each lane is comparable. Since expression was easily detected at 2 hours, experime nts were repeated and the level of Ahr and Arnt protein determined after 15,
57 30, 45, 60, 90, and 120 minutes of galactose treatment. Representative western blots for Ahr and Arnt expression are presented in Figure 3.1C. A representative western blot for t his set of experiments is shown in Figure 3.1B The results show that Ahr protein is detected at the 2 hour time point. Importantly, the level of Ahr protein does not appear to increase above the level observed at 2 hours, even after treatment with galac tose for 6 hours. FIGURE 3.1: Western blot analysis of Ahr and Arnt protein express ion in yeast. Ahr and Arnt proteins were detected in yeast cultures following induction with 2% galactose for 16 hours (A), 2 6 hours (B), and 15 120 minutes (C). Equal numbers of cell s were harvested at each time point and total cellular protein prepared using 20% TCA. Fifteen microliter v olumes of each sample were resolved on 7% SDS PAGE gels and transferred to nitrocellulose for western blotting. Even sample lo ading was confirmed with Ponceau S staining prior to antibody detection. Ahr and Arnt protein expression was detected using 1g/ml concentrations of polyclonal Ahr and Arnt specific antibodies and reactivity was detected using ECL reagent.
58 The results show that Ahr was detected within 30 minutes of galactose treatment. In contrast, Arnt was detected at the 15 minute time point. Ahr expression appeared to reach equilibrium within 60 minutes of galactose induction, while Arnt expression cont inued to increase up to the 120 minute time point. The difference in these results may be related to a higher level of s ensitivity of the Arnt antibody (Pollenz 1996), or differences in the turnover of the proteins. The se results also highlight the fact that yeast expressing exogenous proteins from the GAL1 pro m oter do not require long incubations with galactose to obtain detectable levels of protein. This is an important observation because long term induction and overexpression of exogenous proteins in yeast can lead to cell stress that might impact the interpretation of the experiment (Mattanovich 2004). In summary, these results show for the first time that mammalian Ahr and Arnt protein can be detected in a recombinant yeast model using a western blo tting approach. The ability to easily detect these proteins shows that further studies to assess Ahr degradation may be possible in the recombinant yeast model. Induction of Ahr and Arnt proteins double knock in yeast strain. The next set of experiments was designed to detect Ahr and Arnt proteins when co expressed in a single yeast strain. KHSY1538 and KHSY1541 strains were mated and sporulated as described in chapter two to obtain the double knock in strain, KHSY1547. KHSY1547 was streaked on YPD to produce single colonies and was placed in a 30¡ C incubator for two days. A single colony was used to inoculate an overnight culture with 2ml of liquid YPD media containing 2% glucose and was placed in a shaking 30¡ C incubator overnight. The following day, the optical density was measured at 600nm and a portion
59 of the sample was added to liquid YP medium supplemented with galactose (2% final). The induction culture was set up at an OD 600 of 0.1 in a final volume of 10ml. FIGURE 3.2 : Induction of Ahr and Arnt proteins under the galactose inducible promoter in yeast. Ahr and Arnt proteins were induced in a recombinant yeast strain with the addition of 2% galactose to the growth medium. 1.0 ODs of cells were harve sted from the culture at 1, 2, and 3 hours following addition of galactose. Prote in samples were prepared using the 20% TCA precipitation method. 15l samples were run on 7% SDS PAGE gels, protein was then transferred to nitrocellulose using a semi dry b lotter. Even sample loading was confirmed with Ponceau S staining prior to detection of Ahr and Arnt proteins using specific polyclonal antibodies. A 1g/ml concentration of Ahr antibody and a 0.5g/ml concentration of Arnt ant i body was used in conjunctio n with 1:10000 goat anti rabbit secondary antibody. For tubulin staining, a 1:1000 dilution of primary antibody was used with a 1:1000 dilution of goat anti mouse secondary antibody. ECL reagent was used for detection of protein bands. Protein expressio n was quantified and normalized using ImageJ software. This data illustrates induction of both Ah r and Arnt proteins in the KHSY 1547 recombinant yeast strain. For the experiment shown in Figure 3.2, Ahr and Arnt protein induction was carried out for 1, 2 and 3 hours in the presence of 2% galactose. This time course was based on the studies shown in Figure 3.1. Total protein was prepared from the same
60 number of cells in the induced cultures using 20% trichloroacetic acid as described Two identical SD S PAGE gels were run; one was stained for Ahr protein expression and the other was evaluated for Arnt protein. In addition to verifying even sample loading with Ponceau S staining, these blots were stained with tubulin antibody (Sigma Aldrich) as a load ing control. Protein levels were quantified and normalized using ImageJ software and were plotted using Microsoft Excel. A western blot from a representative experiment is shown in Figure 3.2A and the quantified data is shown in Figure 3.2B. Figure 3.2 shows induction of Ahr and Arnt proteins in the KHSY1547 yeast strain that is detectable by western blotting. The time course of induction is comparable between Ahr and Arnt, and it is detectable within 1 hour of induction and strongly induced at 2 and 3 hours with slight variation in staining intensity likely due to antibody specificity (Pollenz et al 1994) Figure 3.2 is a representative experiment; however Ahr and Arnt inductions were carried out multiple times using the KHSY1547 strain All experimen ts showed the inducible expression of both Ahr and Arnt that was easily detected within 1 hour. Based on these studies future experiments utilized a 2 3 hour induction time. Validation of the yeast model The previous set of experiments demonstrated d etectable levels of Ahr and Arnt protein expression i n the recombinant yeast strains (A im #2). However, the study did not demonstrate the ability of the proteins to fold and function in a manner similar to endogenous Ahr and Arnt proteins While there ar e several proteins belonging to the basic helix loop helix (bHLH) family of proteins expressed in yeast (Berben et al 1990, Greenberg and Lopes 1996, reviewed by Robinson and Lopes 2000), there are no yeast bHLH/PAS homologues for Ahr and Arnt. Thus, since these proteins are not
61 endogenously expressed in yeast it was possible that they may not function properly It was crucial to conf irm that Ahr and Arnt proteins were capable of heterodimerization and DNA binding, since these functions are central to the Ahr signaling pathway and ligand induced degradation (Pollenz 1996, Pollenz 2002, Song and Pollenz 2002). Therefore, our next aim w as to functionally evaluate the Ahr and Arnt proteins and assess the ir ability to initiate tra nscription of a reporter gene (A im #3) The double knock in strain (KHSY1547) was transformed with the pLXRE5 Z reporter plasmid to generate strain KHSY1566 as described in Chapter T wo The reporter plasmid carries five xenobiotic response elements (XREs) located upstream of the lacZ gene (Cox and Miller 2002). For proper reporter activation to occur Ahr and Arnt protein expression must be induced and the Ahr m ust be activated with ligand treatment. After ligand binding, the activated Ahr dimerizes with Arnt and the resulting heterodimer interacts with the XREs to drive transcription of galactosidase. Several strains w ere constructed as detailed in Chapter T wo. KHSY1538 was transformed with the reporter plasmid to produce KHSY1565 This strain only expresses the Ahr and was generated as a negative control for specific activation of the reporter This strain should not produce ligand inducible reporter acti vity, because it does not express Arnt and Arnt i s required for transactivation of target genes in the Ahr signaling pathway (Reyes et al 1992 ) Reporter Assay Method. The first set of reporter assays were carried out using a method s imilar to that of Miller (1997). L ater studies were performed based on the protocol of Kippert (1995) that employed the use of the detergent sodium lauroyl sarconsinate or sarcosyl. Because yeast have cell walls, Miller's protocol entailed preparation of whole cell extra cts in order to release the galactosidase enzymes into the resulting cell lysate so that reporter activity could be detected by addition of
62 substrate. However, Kippert (1995) demonstrated that the results of assays performed in this manner showed a grea t deal of variation Instead, Kippert suggested the use of the detergent sarcosyl to permeabilize the yeast cells to allow the substrate to bind galactosidase in intact cells. The data in Figure 3.3 was generated using the Miller protocol. To test for galactosidase activity in intact cells, 1ml of yeast liquid culture was removed from the larger test cultures. Each sample was centrifuged for 1 minute at 14 ,000 rpm, the mediu m was aspirated, and the cell pellet was combined with 500l of lacZ buffer (60mM Na 2 HPO 4 40mM NaH 2 PO 4 1mM MgCl 2 10mM KCL, and 0.4mg/ml ONPG (o nitrophenyl D galactoside)). The pellet was resuspended in the solution and placed in a 37¡ C water bath an d upon observation of a color change ; the reactions were stopped with the addition of 1.5M sodium carbonate. The samples were centrifuged at 14 ,000 rpm for one minute to pellet the cells and the optical density at 420nm was measured for the supernatant. The observed optical density was normalized for protein concentration and the reaction time with the substrate using the following equation; (Abs 420nm x 1000) / [(Abs 595nm) x ( reaction time in minutes)]. The resulting values, called Miller units, were p lotted using Ex c el. Reporter Activation Following Treatment with Ahr L igand To test whether the Ahr signaling pathway was functional when Ahr and Arnt were expressed exogenously in yeast, b oth the Ahr expressing reporter strain (KHSY1565) and the double knock in reporter strain (KHSY1566) were assayed for galactosidase activity. One liquid culture of each strain was prepare d from overnight cultures. Ahr expression was induced in the KHSY1565 strain with the addition of 2% galactose to the growth medi um for 2 hours. Similarly, Ahr and Arnt co expression was induced in the KHSY1566 strain with 2% galactose for 2 hours.
63 Whole cell extracts were prepared from both strains after induction of Ahr (1565) or Ahr and Arnt (1566) for two hours. These cell l ysates were prepared prior to ligand treatment and activation of Ahr. Ahr signaling requires activation of the Ah receptor with ligand prior to Arnt heterodimerization and DNA binding (Li et al 1994, Whitelaw et al 1993). Therefore, the presence of Ahr and Arnt protein in yeast should not produce galactosidase activity. These samples were prepared to provide a baseline for reporter activity in these strains. After the three hour protein induction each culture was split in half. One sample of each st rain was treated with 20nM TCDD and the other sample was treated with DMSO as a control. The dose used in these experiments was based on a study by Miller (1999) since 20nM TCDD treatment showed maximal reporter activity in similar yeast strains. The sam ples were placed in a shaking incubator for 3 hours to allow for Ahr activation Whole cell extracts were prepared and the lysates were combined with the galactosidase substrate, ONPG. The results of the reporter study are shown in Figure 3.3. The re sults of Figure 3.3 show r eporter activity in the Ahr and Arnt expressing strain ( KHSY1566 ) after 3 hours of 20nM TCDD treatment. The level of activity was approximately eight fold higher than the time zero and DMSO (vehicle) controls. Importantly, the set of zero hour controls produced very low levels of reporter activity even after 2 hours of Ahr and Arnt expression. These results suggest that the culture conditions do not result in spurious act ivation of the Ahr pathway as has been noted in recombinant mammalian report assays ( Miller 1997 ). In addition, the strains did not produce any reporter induction when incubated for 3 hours with DMSO, the vehicle used to prepare and dilute TCDD.
64 FIGURE 3.3: Analysis of galactosidase activity in KHSY1565 and KHSY1566 following TCDD treatment Strains were induced with galactose for 3 hours, split into two equal aliquots and induced with 20nM TCDD or DMSO for an additional 3 h o urs. galactosidase activ ity was determined as detailed in the text. Note that on ly the KHSY1566 strain induced gal activity when treated with 20nM TCDD suggesting that Ahr and Arnt are capable of heterodimerizing and binding to XREs in this system. The results in Figure 3.3 were important for several reasons First, reporter activity was present in the KHSY 1566 strain and not in KHSY1565. T he repo rter was activated upon binding of Ahr and Arnt heterodimers and did not show activation when Ahr was expressed alone This is consistent with Ahr signaling in other organisms where XREs are bound by Ahr/Arnt heterodimers, but not Ahr alone (Reyes et al 1992). Also, we know that reporter activity is not present in samples treated with vehicle only, suggesting that Ahr and Arnt pr oteins are not likely heterodimerizing without first activating Ahr with ligand. These findings are consistent with the model of Ahr signaling where only the ligand bound Ahr interacts with Arnt to form the DNA binding species (Li et al 1994, Whitelaw et al 1993). Thus, they confirm that the basic components of the Ahr signaling pathway are intact in this yeast model T he receptor and its DNA binding partner are detectable and can function as transcription factors when activated with TCDD to drive an art ificial reporter
65 Interestingly, when similar studies were carried in recombinant yeast strains by other groups, a high level of reporter activit y was detectable in the vehicle treated samples. This effect was initially attributed to exogenous tryptophan present in the culture medium since tryptophan has been shown to act as an Ahr agonist and can activate the receptor ( Rannug et al 1987, Miller 1997 ). However, later reports by the same group suggested that this effect was due to overexpression of Ahr and Arnt proteins (Miller 1999). In Miller's 1997 study, Ahr and Arnt proteins were expressed on a high copy plasmid unde r a GAL1 / GAL10 bidirectional prom oter in the presence of galactose for 24 hours. Our western blotti ng experiments provided in Figure 3.1 demonstrated that Ahr and Arnt were detectable within minutes of promoter activation and suggest that 24 hours of induction may result in extremely hig h pr o tein levels or high stress to the cells that likely had reached stationary growth Additionally, experiments presented here were carried out using a yeast strain with a single copy of each cDNA integrated in the yeast genome while Miller's strains carried between 10 and 40 copies of each cDNA on a high copy plasmid (Miller 1997, 1999) This further confirms the importance of detection and analysis of exogenously expressed proteins when working in a recombinant yeast model. Since TCDD could activate the reporter construct, it was pertinent to assess the ability of other Ahr ligands to induce galactosidase in our yeast model. The aryl hydrocarbon receptor signaling pathway can be activated upon exposure to halogenated aromatic hydrocarbons (HAHs), polycyclic aromatic hydrocarbons (PAHs), flavones and several naturally occurring compounds. Here the KHSY1566 strain was evaluated for reporter activity after treatment with several Ahr ligands known to have varying degrees of Ahr binding affinity and p otenc y. TCDD, the prototypical and most potent HAH, was tested and compared to several PAHs inclu d ing benzo( a )pyrene (BAP) and 3 methylcholanthrene (3 MC) and the flavone, naphthoflavone (!NF) It was of interest
66 to determine which ligand produced the highest level of reporter activity for use in later western blotting experiments. FIGURE 3.4: Analysis of galactosidase activity in KHSY1565 and KHSY 1566 following treatment with several known Ahr ligands Strains were induced with galactose for 2 hours split into two equal aliquots and induced with TCDD(40nM), NF(10um), BAP(10uM), 3 MC(100nM), or DMSO for an additional 4 hours galactosidase activity was determined as detailed in the text. Reporter activity was measured for duplicate samples a nd normalized values were plotted using Excel. Note that NF and TCDD produced significant levels of gal activity and BAP and 3 MC produced gal levels only slightly above the DMSO treated control samples. KHSY1566 was induced with the addition of 2% galactose to the growth medium for two hours to allow for Ahr and Arnt expression. Next, the culture was split and the resulting samples were treated with either vehicle or Ahr ligands for an additional four hours. Using the cell permeabilization meth od (Kippert 1995) reporter activity was measured and normalized as previously described. The Miller units were plotted using Excel and are displayed in Figure 3.4. T he ligand doses tested in Figure 3.4 were selected based on previous studies (Miller 1999 ). Figure 3.4 shows a 25 fold induction of galactosidase activity with TCDD treatment and a 50 fold induction of gal activity with !NF treatment compared to the DMSO treated control. This high level of reporter activity in the TCDD and !NF treated cu ltures is in contrast to the BAP and 3 MC treated samples. Treatment with
67 BAP produced a 2 fold induction of gal activity and 3 MC treatment caused a 3 fold induction of gal activity compared to the DMSO treated control These results were unexpected since TCDD is known to be the most potent Ahr ligand with only pM levels required to saturate Ahr binding in mammalian cell culture (Pollenz 1996). The results show that NF, a more labile ligand produces higher gal actosidase levels. One explanation for this observation is the hydrophobic nature of TCDD, making it highly insoluble in water (reviewed by Denison and Nagy 2003). Since yeast growth medium is aqueous, it is likely that the polar nature of TCDD may have caused it to come out of solution b efore entering the yeast cells and the effective dose may actually be be lower than 40nM. Testing this hypothesis would require radio labeled TCDD and since these experiments generate mixed waste, they are not approved at USF. To further evaluate the A hr signaling pathway and better refine the ligand activation t he next set of experiments assess ed whether incre asing amounts of ligand produced dose dependent transactivation of the reporter in the recombinant yeast model. In tissue culture cell lines, the degree of transactivation of CYP1A1 is dependent on the potency of the ligand as well as the dose (Song and Pollenz 2002). Therefore, the reporter strain was used to determine whether a dose dependent increase in reporte r ac tivity could be detected This experiment would also provide information regarding TCDD solubility since it would be expected that if the TCDD remained in solution, activation of the gal reporter would reach saturation at high doses. TCDD and !NF were chosen for these studies since they induced the highest levels of galactosidase activity in the previous experiment ( Figure 3.4 ). The KHSY1566 strain was induced for Ahr and Arnt prot ein expression for 2 hours, split into several cultures, and ind ividual cultures were treated with increasing amounts of TCDD, N F, or vehicle for an additional 4 hours All samples were harvested at the same time and processed for galactosidase activity as previously described.
68 FIGURE 3.5: Dose response analys is of KHSY1566 with TCDD and !NF. Strains were induced with galactose for 2 hours split i nto two equal aliquots and induced with increasing doses of TCDD or NF for an additional 4 h o urs galactosidase activity was determined as detailed in the text. Reporter activity was measured for duplicate samples and normalized values were plotted using Excel. Note that reporter activation is evident for both treatments in a dose dependent manner. The results in Figure 3.5 are consistent with data published by other groups in yeast (Miller 1999, Sugihara et al 2008). Ahr ligand treatment induced expression of the Ahr specific reporter plasmid in a dose dependent manner. For !NF, the results show a sigmoidal d ose response curve with saturation at ~1M. The EC 50 for !NF was approximately 60nM. TCDD was tested using a range of 5nM to 80nM, and higher doses were not tested due to the insolubility of the chemical. The resulting TCDD dose response curve did not a ppear as a sigmoidal curve; instead the curve is linear and does not reach saturation, even at 80nM. A TCDD dose of 80nM is 1.5 orders of magnitude above the EC 50 shown in mammalian cell culture models (Poland and Knutson 1982, Bradfield and Poland 1988, Song and Pollenz 2002). This result suggests that TCDD did not saturate the Ah receptors present in the yeast cells and that the hydrophobicity of TCDD was causing to the compound to come out of solution or become sequestered by
69 protein or other co mponent s of the culture medium. Therefore, the effective dose of TCDD could not be calculate d Generation of a Functional Strain. Based on the data shown in the previous figures, it is suggested that the KHSY1566 yeast strain expresses a functioning Ahr sign aling pathway. The results that support this statement include: Ahr and its DNA binding partner Arnt are detectable by Western blotting ( Figures 3.1 and 3.2 ). Ahr specific activation o f the XRE driven reporter can be detected after treatment with known Ahr ligands (Figure 3.3). The se results suggest that at least some fraction of the expressed Ahr and Arnt proteins are pro p erly folded, can dimerize with each other and can associate with specific DNA elements in the yeast cells. These observations also suggested that the key components of the Ahr signaling pathway are functional when expressed in our yeast model. This information is crucial because there are several other important proteins in the Ahr signaling pathway that were not exogenously expressed in our strains. Functional Ahr signaling also requires Hsp90, Xap2, p23 and other transcriptional activators (Pongratz et al 1992, Carver and Bradfield 1997, Ma and Whitlock 1997, Kazlauskas et al 1999). Yeast homologues to the mammalian Hsp90, Xap2, p23, several co activators and other general transcription factors have been identified and appear to perform the functions of their mammalian counterparts in order for the pathway to function (Cox and Miller 2002, Cox and Miller 200 3). The final step in the Ahr signaling pathway is the proteasome mediated degradation of Ahr and this will be evaluated extensively in later experiments. From this point on the KHSY1566 strain was used to further analyze the signaling pathway when expr essed in yeast.
70 Assessment of Ahr Protein Turnover in Yeast Ahr degradation, as studied in mammalian cells, can be considered in three different scenarios First, the normal half life of the latent receptor has been reported to be approximately 28 hou rs (Ma and Baldwin 2000). This is based on the normal synthesis and destruction of the protein to maintain homeostasis. Second many reports show that the Ahr is degraded in a ligand dependent manner (reviewed by Pollenz 2002, 2010). T he receptor half l ife in the presence of ligand is reduced to approximately 3 hours (Ma and Baldwin 2000, Pollenz 1996). Finally, degradation of the receptor has been observed in a ligand independent manner following treatment with benzoquinone ansamycin antibiotics (Chen et al 1997, Song and Pollenz 2002, Pollenz et al 2005). While many groups have published data on Ahr degradation in higher eukaryotes, there have been no reports on Ahr protein expression or degradation in a yeast model. Such a model is advant ageous sinc e it would allow genetic dissection of the various proteins involved in the proces s Stability of the Ahr in the Absence of Ligand The recombinant yeast strains were created so that the different aspects of Ahr degradation could be evaluated in a model that was amenable to genetic dissection. However, initial studies suggested that ligand mediated degradation of the Ahr was not dete ctable in the recombinant strain when the Ahr was continually induced (see later sections of this chapter). Therefore, it was crutial to better understand the stability of the Ahr in the absence of ligand in the yeast model and that would allow for the us e of other experimental approaches to assess the Ahr degradation. Therefore, it was essential to assess what was happening to the level of Ahr protein after the inducible expression of the Ahr was turned off. It was
71 h ypothesized that if the expression of the Ahr is turned off it may be possible to see changes in the level of Ahr expression upon ligand treatment. The stability of the Ah receptor was evaluated in the Ahr expressing recombinant yeast strain (KHSY1538). Pulse labeling has been used to deter mine the Ahr half life in mammalian cell culture studies (Ma and Baldwin 2000); however, we did not have the ability to use a radioactive method and a simpler approach was employed here. Since Ahr and Arnt were expressed downstream of an inducible promote r, it was possible to induce protein expression for a time period and subsequently turn off the promoter to evaluate the Ahr and Arnt protein levels using western blotting. Importantly, this technique was used by others to test the half life of exogenousl y expressed proteins in yeast (Ke et al 2003). The results presented in the Ke et al (2003) study provided an additional consideration for the promoter off approach. In the Ke study (2003), levels of thymidine kinase (Thk) proteins expressed under the GA L1 promoter were evaluated via western blotting using the promoter off scheme described throughout this chapter. The results of this study showed that the Thk protein was reduced in the cultures over time once the inducible promoter was turned off. This effect was attributed to the rate of yeast cell division (approximately 90 minutes for wild type S. cerevisiae (Sherman 2002)) that will cause the recombinant protein levels to be "diluted" from the culture over time as the yeast bud and continue to propag ate in the absence of new recombinant protein expression. The quantified data from this study revealed a direct correlation between the density of the yeast culture and the level of Thk protein detected on the western blot. Therefore, it was hypothesized that a similar effect would be observed with Ahr and Arnt protein levels when the GAL1 promoter was turned off in the Ahr/Arnt recombinant strain. S. cerevisiae divide every 90 minutes (wild type) during exponential growth, and according to this theory th e number of cells that contain Ahr and Arnt protein
72 would be reduced by half after the first doubling (~90 minutes). After 180 minutes, the protein levels would be expected to be further diluted to 25%. This model assumes that the recombinant protein is essentially stable (~24 hr half life) and is not significantly degraded during the short time line of these studies. A schematic that illustrates this process is presented in Figure 3.6. Note that for this system to be amenable to the study o f Ahr degrad ation all results would fall to the left of the predicted trend line. It is also important to note that the cultures utilized in the following experiments were asynchronous and the cells appeared to persist in the lag growth phase for the initial time po ints of each experiment before reaching exponential growth. Therefore, the time line of the reduction in the level of recombinant proteins may not exactly mirror the theoretical prediction (it would likely shift the predicted trend line to the right). I t is possible that the effect of yeast cell division on the level of Ahr and Arnt proteins in the recombinant strain may confound the ability to detect a ligand mediated effect on Ahr protein levels in the promoter off paradigm. Western blotting and antib ody staining of Ahr and Arnt should reveal reduced levels of protein over time due to active cell division. Because we are looking for a ligand mediated effect on Ahr protein levels, a more rapid reduction in Ahr protein would have to be observed t han the anticipated dilution would cause (a left shift of the predicted curve). Using the promoter off method, it was hypothesized that a loss of Ahr and Arnt proteins would be observed over time and an enhanced loss of only Ahr in the presence of ligand.
73 FIGU RE 3.6 : Predicted pattern of protein detection when an inducible promoter is turned off and the recombinant protein is stable. This figure represents the effect of cell division on the detection of exogenously expressed proteins in the promoter off experi ment. This model is based on a protein that would have a normal half life of 24 hrs (highly stable). A. In this example, the inducible expression of the proteins was activated. All cells in this culture express the given proteins (dark blue). The induc ible promoter is turned off. The cells continue to divide over time and double after ~90 minutes. This culture at 90 minutes then contains double the number of cells but the same level of protein. After 180 minutes the culture doubles in size agai n with the same level of exogenous protein that was present at time zero. For the analysis, e qual number of cells are harvested at the 0, 90 and 180 minutes time points to produce equal sample loading on the SDS PAGE gels for comparison of protein levels. The prepared protein samples will contain decreasing amounts of the target protein. B. Theoretical quantification of the dilution effect. Note that a shift to the left (green line) would indicate that the recombinant protein is being degraded during the time course and to observe enhanced degradation of the recombinant protein that could be quantified, it would be essential to observe additional shifts to the left. For the studies detailed below, the double knoc k in strain KHSY1547 was utilized since both Ahr and Arnt proteins are expressed at the same time. Additional results from the single knock in strains are presented in Appendix A. Ahr and Arnt proteins were induced with galactose for two hours and transcription of AHR and A RNT mRNAs was blocked with the addition of glucose to the growth medium. Duplicate sets
74 FIGURE 3.7 : Detection of Ahr protein degradation in yeast. The KHSY1547 strain was induced with 2% galactose for two hours followed by a change to growth media cont aining 2% glucose. Glucose was added to the medium in order to transcriptionally repress the GAL1 promoter. Protein samples from an equal number of cells were prepared using 20% TCA at various time points after the promoter was turned off. A. Western bl otting and antibody staining revealed a rapid loss of Ahr protein over time. The loading control shows that equal amounts of protein were loaded in each lane B. Ahr protein bands were quantified using ImageJ software and normalized to the loading control The normalized data was plotted using Excel. Note the addition of the red dashed line that represents the predicted level of Ahr that would be present in the cells based on the culture densities shown in part C. C. Optical density of the yeast culture at the time each sample was prepared. Since the OD represents the number of cells in the culture, this can be used to produce a trend line that would show the theoretical loss of the Ahr if the reduction was solely based on dilution through cell division Note that the Ahr protein level decreased over time and the line of the graph is shifted to the left of the predicted trend line. This observation suggests that Ahr is being degraded from the cells when the promoter is turn ed off and the loss of prote in is not entirely due to dilution of proteins as the cells divid e
75 of p rotein samples were prepared at time zero and then equal numbers of cells harvested at one hour intervals. Protein samples were prepared using the TCA methods. The results for Ahr protein expression are shown in Figure 3.7A. The level of Ahr protein was quantified using ImageJ software and normalized to the loading control (Figure 3.7B). The western blotting results in Figure 3.7A show that the Ahr protein levels are rapidly reduce d after the GAL1 promoter is turned off Importantly, the plot of Ahr protein is shifted to the left of the predicted trend line (dashed line in Figure 3.7B) and suggests that the estimated t 1/2 of the Ahr in this model is ~1 h ou r. This half life is significantly shorter than what has been determined in mammalian cell culture models ( Ma and Baldwin 2000 ). This result implies that Ahr protein levels are not simply being diluted from the yeast culture, but instead, the protein is bei ng actively degraded in a non ligan d ed state and may be unstable when expressed in yeast. It is important to note that the reduction of the Ahr protein does not appear to be related to premature initiation of the Ahr pathway by endogenous ligands present in the yeast or yeast media since minimal levels of gal induction are observed in the absence of exogenous ligand in this model (Figures 3.3 3.5). Based on the results presented in figure 3.7, it was pertinent to assess the stability of Arnt in this sy stem. Arnt is predicted to be a stable protein with a half life of >20 hrs in mammalian cells (Pollenz 2002, 2010). The samples prepared in the experiment detailed in 3.7 were evaluated for Arnt and the results are presented in Figure 3.8. The results s how that Arnt protein levels are also rapidly reduced after the GAL1 promoter is turned off and the plot of Arnt protein is shifted to the left of the predicted trend line. The estimated t 1/2 for Arnt was ~2 h ou r s in this experiment Thus, both Ahr and A rnt proteins appear to be less stable when expressed in yeast compared to mammalian cells. The rapid loss of the Ahr in the absence of ligand complicated the
76 FIGURE 3.8: De tection of Arnt protein degradation in yeast. The KHSY1547 strain was induced with 2% galactose for two hours followed by a change to growth media containing 2% glucose. Glucose was added to the medium in order to transcriptionally repress the GAL1 promoter. Protein sa mples from an equal number of cells were prepared using 20% TCA at various time points after the promoter was turned off. A. Western blotting and antibody staining revealed a rapid loss of Arnt protein over time. The loading control shows that equal amou nts of protein were loaded in each lane. B. Arnt protein bands were quantified using ImageJ software and normalized to the loading control The normalized data was plotted using Excel. Note the addition of the red dashed line that repre sents the predicte d level of Arnt that would be present in the cells based on the culture densities shown in part C. C. Optical density of the yeast culture at the time each sample was prepared. Since the OD represents the number of cells in the culture, this can be used to produce a trend line that would show the theoretical loss of the Arnt if the reduction was solely based on dilution through cell division. Note that the Arnt protein level decreased over time and the line of the graph is shifted to the left of the pre dicted trend line. This observation suggests that Arnt is being degraded from the cells when the promoter is turne d off and the loss of protein is not entirely due to dilution of proteins as the cells divide
77 use of this model for assessment of Ahr degradation. The original aim of this project was to produce a model to genetically evaluate Ahr degradation in yeast in a ligand dependent manner. Thus, experiments designed to evaluate the level of Ahr protein in th e presence of ligand using the promoter off model would require the Ahr to be reduced within a matter of minutes in order to see changes from the control Thus, additional experiments were performed to determine the impact of cell growth, dilution and prot ein overexpression in an effort to determine if conditions could be identified that would result in a more stable Ahr Since the Ahr was being rapidly lost from cultures when the GAL1 promoter was was transcriptionally repressed it is possible that this was due to overexpression in the recombinant yeast strains. Importantly, similar results were also observed for the single knock in strains as shown in Appendix B. Overexpression of exogenous proteins correlates with an increase in expression of chaperone proteins in the cell in order to facilitate proper protein folding. I n cases of extreme overexpression, the cell s may be unable to produce enough chaperones to properly fold the newly synthesized proteins (Mattanovich et al 2004 ). Unfolded proteins in the cytoplasm of Saccharomyces cerevisiae have been shown to cause the unfolded protein response (UPR). UPR then detects the misfolded proteins in the cytoplasm and conjugates the substrates with ubiquitin in order to degrade the unfolded and misfolded proteins via the 26S proteasome (Eisele et al 2008, Mattanovich et al 2004). Therefore, w e chose to use the galactose inducible promoter as opposed to a constitutively active promoter ups tream of the AHR and ARNT cDNAs to avoid over expression, protein turnover, and cell stress. With the inducible promoter, could be carefully modulate the amount of protein produced in order to avoid gross over expression that could occur when using a strong, constitutively active promoter. However, w i t hout a true baseline to directy assess the number of protein molecules in each yeast cell, it was unclear whether this system was
78 expressing a level of protein that was comparable to that found in mammalian cells. We hypothesized that inducing the promo ter with less galactose, might illustrate whether overexpression was contributing to the rapid loss of Ahr. Next, it was hypothesized that the chaperone proteins (i.e H sp90) required for Ahr f olding may not be functionin g efficiently in yeast and resul t ing in rapid loss of Ahr. This hypothesis will be explored in later sections of this chapter, but did not appear to impact the stability of the Ahr Finally, it was proposed that Ahr proteins may be degrading in the yeast strains as a resu l t of an intrin si c yeast specific degradation signal within the protein. It is possible that a signal or particular sequence within the proteins is recognized by the yeast cell's degradation machinery that targets the proteins for degradation. However, there is no data ba se of degrad a tion signals in yeast that can be used to assess this question without carrying out mutagenesis of the Ahr. In addition, the sequence involved in Ahr degrad a tion in mammals is also current l y unknown. Because Ahr and Arnt have some sequence in common, there may be a deg radation signal within the bHLH PAS sequence since they share a high identity within this reg i on. Optimization of Sample Preparation for Ahr Stability. A ddition of 2% galactose to the growth medium is standard protocol for maximum activation of the GAL1 promoter in yeast (Sherman 2002), and initial studies were carried out under these conditions. It was hypothesized that reducing the amount of galactose added to the culture medium would lead to a reduct ion in the expression of Ahr and Arnt proteins L owering the amoun t of sugar, should reduce the degree of GAL1 promoter activation and produce a lower level of Ahr and Arnt expression. It was then hypothesized that lowered expression of the exogenous proteins may lead to pro duction of more stable proteins Due to the high sensitivity of the Ahr and Arnt antibodies, it was believed that the protein could still be detected even when expressed at much lower levels.
79 In order to test this hypothesis, a series of 20ml cultures w ere prepared containing decreasing amounts of galactose. The p ercentage and volumes of galactose used to prepare the cultures are shown in Table 3.0. Each culture was prepared such that the starting cell density at OD 600 equaled 0.1. All volumes of gala ctose were brought up to 1ml with sterile water prior to addition to the culture media. The culture s were then placed in a 30¡C shaking incubator for 3 hours to allow for protein induction. While the purpose of adding galactose to the growth medium was to activate the promoter for these experiments, it also functions as the sole carbon source for the yeast to sustain cell growth. For this reason, each culture tested in Figure 3.9 was supplemented with an additional 1% sucrose. Sucrose does not affect t he GAL1 promoter; it does not activate or inhibit the promoter in the way galactose and glucose do (reviewed by Lohr et al 1995, Platt and Reece 1998). Addition of sucrose to the culture medium, therefore, allowed the cultures with the minutest amounts of galactose to continue to grow without exhausting their carbon source. TABLE 3.0: Percentage of galactose added to yeast cultures to minimized induction of Ahr and Arnt protein levels and reduce cellular stress. KHSY1547 was used in five separate cultures containing decreasing amounts of galactose. One additional culture wa s prepared as a negative control and contained only glucose as a carbon source. Each volume of galactose was brought up to one milliliter with sterile water before it was added to the culture medium.
80 Protein samples were prepared as previously described u sing 20% TCA. Two SDS PAGE gels were prepared; one was loaded with 5l of each sample and the other was loaded with 15l of each sample. The negative control, which was grown in presence of 2% glucose, was loaded in the first lane of each gel. Samples prepared from cultures grown in the presence of increasing amounts of galactose were loaded next, and a sample of protein prepared from mouse Hepa 1 cells was loaded as a positive control in the last lane. The mouse cells express the Ah b 1 allel e of the receptor and it migrates at a lower molecular mass than the Ah b 2 that is expressed in the yeast cells. The representative experiment is presented in Figure 3.9. FIGURE 3.9: Ahr protein expression is reduced with decreased levels of galactose. Six 20ml cultures were prepared using the KHSY1547 strain. The culture medium was prepared using the amounts of sugar listed in Table 3.0. After three hours of induction in a 30¡C shaking incubator, protein samples were prepared using 20% trichloroacet ic acid. SDS PAGE gels were run with either 5 or 15l of lysate and stained with anti Ahr antibodies. The western blotting results show that Ahr protein induction can be reduced by adding less galactose to the culture medium. Figure 3.9 shows Ahr expr ession is detectable in all cultures grown in galactose. The samples grown in 2%, 0.2%, 0.1%, and 0.05% galactose all showed Ahr staining when 15l of the sample was run on the SDS PAGE gel. Even the induction with 0.02% galactose, the equivalent of 10l in a 20ml culture, was enough to ind uce detectable expression of the proteins by way of the GAL1 promoter. Overall levels of Ahr
81 expression induced by 0.05% galactose were reduced approximately by half compared to 2.0% galactose This result suggests th at the addition of 2% galactose to the growth medium is not required in order to produce detectable levels of Ahr. T herefore the strains could be tested for protein stability following induction with less galactose added to the culture medium and determi ne whether reduced levels of Ahr induction can improve the protein stability and reduce the rapid loss that was observed in Figure 3.10. FIGURE 3.10: Activation of the GAL promoter with 2% or 0.05% galactose and subsequently turning off the promoter with glucose causes rapid Ahr turnover in yeast. Two separate cultures of equal cell density were prepared using the KHSY1547 strain. One culture was prepared with 2% galactos e while the other was prepared with 0.05% galactose. The western blotting results show that 2% galactose leads to a stronger activation of the promoter than 0.05% galactose. Removal of galactose and addition of glucose to the growth medium causes rapid d egradation of Ahr after induction with 2% or 0.05% galactose, suggesting that minimizing the activation of the promoter does not produce a more stable Ah receptor protein. The KHSY1547 strain was propagated and split into two aliquots. One culture was induced with the standard 2% galactose and the other was induced with 40 fold less galactose at 0.05%. Both cultures were also supplemented with an additional 1% sucrose in o rder to maintain cell viability and division, particularly in the 0.05% galactose culture. The cultures were placed in a 30¡C shaking incubator and induced for 3 hours to allow for Ahr and Arnt protein induction. After three hours, the cultures were cent rifuged at 2000 rpm for 2 minutes and the culture medium was removed. The cell pellets were washed with 1 ml of sterile water and centrifuged again. The cells were then resuspended in medium containing 2% glucose in order to turn off the GAL1
82 promoter. Samples from both cultures were harvested at 0, 30, 60, 120, and 180 minute time points after the promoter was turned off. Protein was prepared using the TCA method from an equal numbers of cells, as described. Protein samples were resolved on a 7% SDS P AGE gel and stained with anti Ahr antibodies. The resulting western blot is shown in Figure 3.10. The results show that the level of Ahr protein induced with 0.05% galactose is approximately half of the level induced with 2% galactose at time zero. Howe ver, in both experiments, the Ahr is rapidly lost from the cells once the promoter is turned off This result suggests that while reducing the induction with minimal galactose in the culture medium does result in less induction of the Ahr protein, it is do es not appear to enhance Ah receptor stability in the recombinant yeast model. The next experiment was carrie d out to test Ahr signaling when reduced amounts of galactose are used for GAL1 induction. This exp eriment was important because the results w ould demonstrate if there was a correlation betwee n the level of protein observed using western blotting and the proportion of that protein pool that was functional and could activate the XRE reporter. The KHSY1566 strain was tested for reporter activatio n after induction in 1.0%, 0.1%, and 0.01% galactose. Again, these cultures were supplemented with 1% sucrose in order prevent depletion of the yeast resources. The results of the galactosidase assay shown in Figure 3.11 demonstrate that induction of th e GAL1 inducible promoter with galactose is dose dependent. Induction with 1% galactose produced a 5 fold induction of gal activity over the uninduced and vehicle treated controls. Importantly, induction with ten fold less galactose at 0.1% revealed gal levels that were only slightly above the negative controls. This result suggests that these low amounts of galactose cannot be used in the reporter assays and higher amounts are required in order to induce expression of enough protein. When these res ults are compared to Figure 3.9 it is clear that there is some correlation
83 between the level of protein detectable in the western blots and the portion that is functional as measured by the reporter assay. FIGURE 3.11: Analysis of galactosidase acti vity in KHSY1566 following activation of the GAL 1 promoter with decreasing amounts of galactose The reporter strain was induced with 1%, 0.1%, or 0.01% galactose for 2 hours prior to dosing with 20nM TCDD for an additional 3 hours. Two controls were als o evaluated. An uninduced strain wa s treated with 20nM TCDD and an induced strain was treated with vehicle for the same time cou r se. galactosidase activity was determined as detailed in the text. S ignificant levels of gal activity were observed when the strain induced with 1% galactose was treated with 20nM TCDD. T he culture induced with 0.1% galactose still showed some level of induction, but it was significantly reduced compared to the 1% induction. Also, the 0.01% induction was below background levels. This result suggests that there is a correlation between t he level of Ahr observed on the western blots and the functional protein that is being produc e d. Preparation of Soluble Protein Frac tions The previous hyp othesis suggested that Ahr turnover in yeast may be due to overexpression of the AHR and ARNT mRNA by the GAL1 promoter. However, the theory that addition of less galactose and the subsequent reduction of protein induction did not appear to produce a more stable protein. Another possibility arose after questioning our protein preparation technique. The procedure employed in the previous western blotting experiments use d 20% trichloracetic acid to precipitate protein from the yeast cell pellets This pro cedure was useful for its simplicity and because it produced samples with a high concentration of
84 protein in a small sample volume. Ahr and Arnt proteins were detect able in samples prepared using this method, but we asked whether this sample preparation m ethod may have contributed in some way to the instability of our proteins in yeast. We asked if the AHR and ARNT mRNAs were being overexpressed such that the Ahr and Arnt proteins were being shuttled to inclusion bodies within the cells and subsequently un dergoing degradation. If this was the case, and the TCA procedure precipitated all proteins within the cell, it was possible that insoluble and/or nonfunctional proteins were precipitated along with a functional pool of protein With that, w e wondered if there may be a more stable Ahr present in the soluble protein fraction If there is a stable form of the receptor in the cells, could the soluble p roteins be isolate d away from the insoluble proteins In the next several experiments, soluble protein fractions were prepared from the double knock in strain using a phosphate buffer extraction method described in Chapter 5 One large culture of KHSY1547 was grown for several hours in glucose and was split in half so that one half was induced with 2% gala ctose and the other remained uninduced with glucose. The cultures were placed in a shaking incubator for 3 hours to allow for Ahr protein induction. After the induction period, a sample of each culture was used to determine the cell density at OD 600 1. 5 ODs of each cu l ture was removed and protein was extracted using the standard 20% TCA protocol used previously. Previous attempts to prepare whole cell lysates produced samples that were too dilute for detection of our proteins using western blotting. T herefore, the cell number was increased 10 fold and 15.0 ODs of culture was used for the soluble protein extraction procedure. The 15.0 ODs of yeast cells were spun at 2000 rpm for 2 minutes and the pellet was suspended in 1ml of cold phosphate buffer (2 0mM NaPi pH 7.7, 300mM NaoAc, 10% glycerol, protease inhibitor cocktail). The samples were then spun at 5000 rpm in a
85 refrigerated centrifuge at 4¡C for five minutes. The pellet was resuspended in 100l of cold phosphate buffer, an equal volume of glass beads were added, and the tubes were placed in a mini bead beater for 4 times for one minute each, and placed on ice for two minutes in between each. The resulting lysate was spun again at 14,000 rpm for 30 minutes at 4¡C. The resulting supernatant, cont aining presumed soluble proteins was then combined with an equal volume of 20% TCA in order to precipitate our proteins and later resuspend them using a smaller volume of SDS PAGE sample buffer. FIGURE 3.12: Detection of Ahr protein in soluble protein fractions Protein samples from galactose induced KHSY1547 cultures reveal Ahr staining following TCA precipitation or soluble protein extraction methods. Intensity of Ahr staining is much lower in samples prepared using the soluble extraction procedure especially considering that the sample contained ten times more cells to start with. Decreasing amounts of sample were loaded to determine the minimum sample volume required for detection. Figure 3.12 shows detectable levels of Ahr protein in both the t richloroacetic acid precipitation and the soluble protein extraction methods. Interestingly, 1.5 ODs of cells prepared with TCA shows much more intense staining than 15 ODs of cells prepared for soluble protein This observation may suggest that there ar e two separate Ahr protein pools within the yeast cells, one soluble fraction containing the functional receptor, and one insoluble fraction containing improperly chaperoned receptor. If this is the case, the results of Figure 3.12 imply that the soluble protein pool is indeed a very small fraction of the total Ahr when expressed in our yeast model. However, this effect could simply be due to the efficacy of the protein preparation procedure.
8 6 The stability of Ahr and Arnt was tested again using protein extracts prepared using phosphate buffer for the protein preparation procedure to obtain what was believed to be the soluble protein fraction. The double knock in strain was induced using 2% galactose for three hours before the promoter was turned off, as previously described. Protein samples were then prepared via TCA precipitation or using phosphate buffer to prepare the soluble fraction Samples were obtained at 0, 1, 2, and 3 hour time points after the promoter was turned off with the removal of gala ctose and addition of glucose to the growth medium. Figure 3.13 shows two SDS PAGE gels; one stained with anti Ahr antibody and the other stained with anti Arnt antibody. FIGURE 3.13: Turnover of Ahr and Arnt in samples prepared using trichloroacetic acid or the soluble protein extraction method when the inducible promoter is turned off. Samples were prepared using both methods to evaluate whether the proteins were being o verexpressed and forming an insoluble pool. The TCA extraction method precipitated all proteins in the yeast cells, whereas the alternative protein extraction purified the soluble protein pool only. A culture of KHSY1547 was grown in the presence of gala ctose for three hours to induce expression of Ahr and Arnt. The promoter was turned off with the addition of glucose to the culture medium, and portions of the culture were removed at 0, 1, 2, and 3 hours after the promoter was turned off. Protein sample s were prepared from 1.0 ODs of cells usi ng 20% trichloroacetic acid and soluble protein extractions were prepared from 15.0 ODs of ce l ls. Western blotting and antibody staining revealed rapid turnover of Ahr and Arnt proteins when the promoter was turned off for samples prepared via both me thods Similar to what was observed in Figure 3.12, the protein level detectable after TCA precipitation of 1.0 ODs of cells was significantly higher than the protein detected in the soluble protein extractions from 15 times more starting material. Also consistent with
87 previous experiments, Ahr and Arnt proteins rapidly turned over in the samples prepared via TCA precipitation. However, Figure 3.13 also shows the Ahr and Arnt proteins are lost in the soluble fraction when the promoter was turned off. This result implied that there may not be two separate Ahr protein pools and the TCA precipitation procedure was simply more effective at bringing down the proteins than the soluble protein extractions. Importantly, the results from the whole cell lysates show that the Ahr is still reduced in the cells and exhibits a t 1/2 of approximately 1 hr, while Arnt shows a longer t 1/2 that is consistent with the re sults presented in Figure 3.8. It is apparent from these results th at the Ahr protein is rapidly turning over in the samples prepared by either TCA precipitation or soluble protein extraction. Thus, these results suggest that the preparation of soluble protein extracts is not required. If Ahr does indeed degrade in a l igand dependent manner in these cells, observation of a measurable effect may not be possible. Thus, the next set of studies aimed to determine if the rapid reduction of Ahr was related to improper chaperoning of the receptor by the yeast H s ps. Analysis of Ligand Independent Ahr Degradation in Yeast Heat shock protein 90 (Hsp90) comprises approximately 1 2% of the cellular protein pool of eukaryotes (reviewed by Sreedhar et al 2004). There are two Hsp90 isoforms in mammals, termed Hsp90" and Hsp90!. H sp90" is expressed at low levels but is highly inducible under conditions of cell stress while Hsp90! is known to be constitutively expressed and less inducible than its counterpart (reviewed by Sreedhar et al 2004). Similarly, the HSP82 and HSC82 genes i n yeast, encode two isoforms of Hsp90 whose sequences are highly conserved (Borkovich et al 1989). It has been shown that the yeast HSP82 shows 61% amino acid identity and exhibits similar function to the human Hsp90" while the yeast HSC82 shows 49% identity to the human Hsp90! (Cox and Miller 2003).
88 Several groups have utilized S. cerevisiae to evaluate the role of Hsp90 in Ahr signaling (Carver et al 1994, Cox and Miller 2003, Whitelaw et al 1995). First, Carver (1994) used a yeast str ain in which both HSP82 and HSC82 genes were disrupted and cell viability was maintained by transformation with a plasmid expressing HSP82 at approximately 5% of the wild type level. Plasmids carrying the AHR and ARNT cDNAs were transformed into the Hsp90 deficient yeast strain. Ahr signaling was evaluated using a reporter plasmid similar to the pLXRE5 Z used in our experiments. These investigators demonstrated that Ahr signaling was not detectable in their yeast strain. However, increased expression of HSP82 from 5% to wild type levels rescued a functioning Ahr signaling pathway in the strain. This experiment confirmed the hypothesis that Hsp90 is an essential component of the Ahr signaling pathway but it also suggested that the yeast Hsp90 homologues are able to function in place of human Hsp90 in the reconstituted system. This information was critical to our experiments. These results suggested that there is enough conserved sequence identity, most likely within the protein's functional domains, for yeast Hsp82 and Hsc82 to function in place of mammalian Hsp90" and Hsp90! in Ahr signaling. It is noted however, that in these studies the level of Ahr or Arnt expression was never evaluated. While previous studies suggested that the yeast Hsp90 homolog ues, the HSP82 isoform in particular (Cox and Miller 2003), function in Ahr signaling in yeast, we asked if the expression level was sufficient to properly chaperone the Ahr and Arnt in our yeast model. If expression of yeast Hsps was inadequate in our st rains, the unfolded proteins could be targeted for rapid degradation or sequestered to inclusion bodies (Kamasawa 1999). Experiments in mammalian cell culture have shown that Ahr is rapidly degraded upon disruption of the AhrHsp90 interaction in the unli ganded cytoplasmic complex (Chen 1997, Song and Pollenz 2002). Treatment of mammalian cells with certain chemicals such as geldanamycin, inhibit and disrupt Ahr association with Hsp90 and
89 result in rapid turnover of Ah receptor protein in the absence of ligand treatment. This type of degradation is referred to as ligand independent degradation (Song and Pollenz 2002 and others) and the time course of degradation looks very similar to the results shown in Figures 3.6 3.8. We investigated whether Ahr loss could be accelerated in the presence of Hsp90 inhibitors treatment. This effect may be detectable in the yeast strains if the Ah receptor is indeed turning over because of an improper interaction with Hsp90. Cox and Miller (2003) reported that treatment of a yeast strain expressing a recombinant Ahr signaling pathway with Hsp90 inhibitors caused a reduction in Ahr signaling as measured by reporter activity. Based on the observed degradation of Ahr with Hsp90 inhibitor treatment in cell culture (Song and Pollenz 2002), one could hypothesize that reporter activity could be reduced in yeast strains with Hsp90 inhibitor treatment because the Ah receptor is degraded. But as stated previously, the mechanism to this reduction could not be correlated to reductio ns in Ahr levels because protein levels were never evaluated. The study by Cox and Miller (2003) referenced above showed that geldanamycin, a particular Hsp90 inhibitor often used to induce ligand dependent Ahr degradation, did not produce a statistically significant reduction in galactose reporter activity. However, they did show that several other compounds with similar function were effective at ablating Ahr signaling as measured by the galactosidase reporter. Specifically, novobiocin and radicico l treatment caused a dose dependent decrease in gal activity. In order to test the effects of Hsp90 disruption in the yeast strains, cultures of KHSY1547 were treated with DMSO, 33M novobiocin, and 5M radicicol. The effective doses for these dru gs were determined in the above referenced study. The cultures were pretreated for one hour prior to induction of Ahr and Arnt proteins with 2% galactose. Treatment with Hsp90 inhibitors or DMSO continued during the three hour
90 induction time and protein samples were harvested in triplicate using 20% TCA. Western blotting and antibody detection was carried out as previously described and even sample loading was confirmed with Ponceau S staining. A representative western blot is presented in Figure 3.14. FIGURE 3.14: Effect of Hsp90 inhibitors on Ahr and Arnt expressing yeast strain. The double knock in strain was evaluated for Hsp90 inhibitor mediated ligand independent degradation. A culture of KHSY1547 was split into three smaller cultures. Each was pretreated for one hour with DMSO (control), 33 M novobiocin, or 5 M radicicol. After one hour of pre treatment, the cells were spun down and subsequently brought up in medium containing 2% galactose for protein induction and the respective treatment. After three hours, protein samples were prepared in triplicate for each culture. In the western blots shown above, it appears that Ahr and Arnt protein levels are slightly reduced after treatment with novobiocin or radicicol as compared to the DMSO cont rol. The results show that treatment of cells with novobiocin and radicicol had minimal impact on the level of Ahr or Arnt protein levels at any of the time points evaluated. In mammalian cell culture, treatment with these compounds results in rapid and nearly complete degradation all Ahr protein within a three hour time course (Song and Pollenz 2002). It should be noted that treatment with these Hsp90 inhibitors was concurrent with activation of the GAL1 promoter, and thus expression was continuous thro ughout the stud y Impact of human H sp90 expression on Ahr stability in yeast While the effect of Hsp90 inhibitor treatment on Ahr levels was minimal, it was still possible that
91 chaperoning was involved in turnover of Ahr and Arnt proteins expressed in yeast. Increased turnover may have been due to the level of expression of the yeast Hsp90 homologues. Therefore, we obtained h uman Hsp90 expression vectors from Dr. Jill John son at the Universit y of Idaho and transformed these into the double knock in strain KHSY1547. We hypothesized that increased levels of mammalian Hsp90 might result in increased stability of Ahr. These plasmids allowed for e x pression of Xpress tagged human Hsp90 and Hsp90 isoforms, driven by the yeast constitutive GPD (glyceraldehydes 3 phosphate dehydrogenase) promoter. Maps of the Hsp vectors are shown in Figure 3.15A. FIGURE 3.15: Human Hsp90 vector maps and transformation of KHSY1547 with each plasmid. Human Hsp90 expression vectors were obtained from Dr. Jill Johnson at the University of Idaho. Expression of Xpress tagged Hsp90 and isoforms is driven by the constitutive GPD promoter. The AHR and ARNT expressing strain was transformed wi th empty ve ctor, Hsp90 ", and Hsp90! plasmids in three separate experiments. The transformed y east were selected on HIS dropout media and the resulting plates are shown above. A single clone of each was used in subsequent experiments to evaluate AHR protein stabilit y when Hsp90 is overexpressed. Three individual cultures of KHSY1547 were prepared and transformed with Hsp90 ", Hsp90! and an empty vector to prod uce three new strains called KSHY2228, KHSY2229, and KHSY2230 respectively. The transformed y east were sel ected on media lacking histidine and the resulting plates are shown in Figure 3.14B. One clone
92 from each transformation was used in subseq uent experiments to evaluate Ahr protein stability in cells expressing high levels of human Hsp90 FIGURE 3.16: Ah r turnover is evident in strains overexpressing human heat shock proteins. Two yeast strains, KHSY2228 and 2229 express human Hsp90" and isoforms from a constitutive reporter. The third yeast strain carries the empty vector. Ahr protein was induced wi th the addition of 2% galactose to the growth medium for three hours and was then turned off with the addition of glucose. Protein samples were prepared at 0, 1, and 2 hour time points after the promoter was turned off. Staining with Xpress antibody reve aled high levels of expression of the human heat shock proteins, with no staining detectable in the control strain. Staining with antibody specific to Ahr showed rapid turnover of the proteins over time, similar to the degradation observed in previous exp eriments. This result suggests that enhanced expression of the Hsp90 chaperone proteins does not stabilize Ahr in our yeast expression system. Upon generation of these Hsp90 expressing strains, we proceeded to test Ahr stability again. Ahr and Arnt proteins were induced with the addition of galactose for four hours in the various Hsp90 expressing yeast strains. After four hours, the inducible GAL 1 promoter was turned off to block transcription of Ahr and A rnt and to assess the
93 stability of thes e proteins. Samples were collected at 0, 60, and 120 minutes after the promoter was shut off and protein was prepared using 20% TCA as previously described. Results for Ahr are shown in Figure 3.16. Three identical SDS PAGE gels were prepared using equal amounts of each protein sample. Western blo tting and staining with antibodies specific to the Xpress tag (Invitrogen) revealed intense bands at approximately 90kDa in the KHSY2228 and KHSY2229 strains. Staining for the Xpress tag d id not show reactivity in the cells containing the empty vector It is noted that the level of expression for the Hsp90 protein is consistent at all time points and this was expected based on expression from a constitutive promoter. The same samples wer e also probed with Ahr specific antibodies and the resulting western blot shows a rapid loss of Ahr with a t 1/2 of ~1 hour. None of the strains showed a significant change in the loss of Ahr when compared to the strain with the empty vector. The loading of equal levels of protein across all samples is validated by the staining for tubulin. In summary, the resul ts from these experiments show that although Hsp90 can be expressed, there is no apparent change to the stability of the Ahr protein. Given the number of different experiments that were carried out to generate a more stable receptor in our inducible model we concluded that we should move forward and test the strain for ligand medi ated degradation of the Ahr while continuing to hypothesize about other solutions to the prob l em. Effect of Ahr Ligand Treatment in Yeast Strains. Extensive research in cell cul ture models has demonstrated that Ahr ligand exposure results in activation of the Ahr signaling pathway that terminates in degradation of the receptor ( Prokipcak and Okey 1991, Pollenz 1996, reviewed by Pollenz 2002) Degradation of Ahr is mediated by th e 26S proteasome (Ma and Baldwin 2000, Roberts and Whitelaw 1999, Song and
94 Pollenz 2002) and is a crucial step in the signaling pathway as it attenuates the dosage of Ahr and Arnt directed transactivation of regulated genes (Ma and Baldwin 2000 ; reviewed b y Pollenz 2002, 2010 ) While many studies clearly show that the ligand mediated Ahr turnover occurs via the 26S proteasome, the exact residues that are modified on the Ahr to target it for degradation remain unknown. As stated previously the Ahr and Ar nt expressing yeast strains were generated in order to screen for proteins involved in mediating Ahr turnover. The goal was to determine if Ahr degraded in yeast in a ligand dependent manner. Next, we would knock out genes from the yeast genome that coul d be responsible for targeting the receptor for proteasomal degradation. A candidate approach would be employed initially, focusing on E3 ligases as they are the suggested mediators between the protein targeted for degradation and the 26S proteasome. It could then be determine d if a particular E3 was responsible for targeting Ahr by assessing Ahr protein levels after ligand treatment. If the knocked out ligase was responsible for Ahr turnover, we would observe high levels of receptor after ligand treatme nt compared to the control strain that would still express the ligase in question. Unfortunately, the half life experiments shown in Figures 3.8 and 3.13 revealed degradation of the receptor in the absence of Ahr ligand and these results proved to com plicate this approach. To assess ligand mediated degradation it was essential to have a model that was amenable to observing a quantifiable reduction in Ah receptor levels after ligand treatment in order to test for blocked turnover in the a ppropriate E3 knock out. Since attempts at stabilizing the receptor expression were unsuccessful, it was still pertinent to determine if ligand treatment would result in an enhanced rate of degradation than was observed in the absence of ligand. Ligand mediated degrad ation of the Ahr was evaluated under several induction conditions. First, the KSHY1566 strain was treated with TCDD and induced with
95 galactose simultaneously. The nature of our inducible system allowed for such an approach and it was hypothesized that th e ability of the ligand to immediately bind a newly synthesized Ahr might result in immediate degradation. We would then observe a lack of accumulation of the synthesized Ahr compared to controls that did not receive ligand treatment. The KHSY1566 strain was used in the following experiments. This strain expresses both Ahr and Arnt under the GAL1 promoter but also carried the pLXRE5 Z reporter plasmid. We used this strain for the following experiments because we know that DNA binding is required for Ahr degradation to occur (Pollenz et al 2005). An overnight culture of KHSY1566 was split in half; one portion of the culture was treated with 20nM TCDD and 2% galactose and the other half of the culture was treated with DMSO and 2% galactose. Protein sampl es were prepared using 20% TCA at various time points after treatment/induction. All samples contained the same cell numbers. The resulting western blot, stained with anti Ahr antibodies, is shown in Figure 3.17. FIGURE 3.17: Induction of Ahr protein with galactose and simultaneous treatment with TCDD revealed equal levels of Ahr protein. KHSY1566 was treated with 20nM TCDD and 2% galactose or with DMSO and 2% galactose. Protein samples were prepared using 20% tri chloroacetic acid at 0, 60, 120, 150, 180, 210, and 240 minutes after treatment with TCDD and galactose induction. Western blotting and antibody detection revealed equivalent levels of Ahr protein induction after treatment with TCDD or DMSO.
96 The res ults revealed that Ahr protein expression levels increased over time in the yeast cells in the presence or absence of the known Ahr ligand, TCDD. This observation was in contrast to our anticipated result since it was hypothesized that reduced levels of A hr protein would be observed in the TCDD treated culture compared to the DMSO treated control. In spite of this result, we still wanted to explore the effect of ligand on Ahr protein under different induction conditions. In the next experiment, yeast cultures were treated with TCDD or DMSO after a three hour induction in galactose. These studies would result in a strain that expressed an Ahr protein pool prior to ligand treatment. The presence of Ahr protein in the yeast cells prior to treatment represents a more physiological scenario since endogenous levels of Ahr protein are present in mammalian cell culture prior to ligand treatment. In a culture of the KHSY1566 strain, the GAL1 promoter was activated with galactose for three hours to induce Ahr and Arnt protein expression. After three hours, the culture was split in half; one half was treated with DMSO and the other with 20nM TCDD. The promoter was left on during the three hour treatment time in order to prevent th e rapid ligand independent loss of Ahr reported throughout the previous sections. It was reasoned that we may observe some measurable difference in Ahr protein levels between treated and control cultures, even if the Ahr protein was not completely degrade d due to the continual induction by galactose. The results of this experiment are shown in Figure 3.18. Ahr and Arnt protein were detectable via western blotting after the 3 hour induction with galactose. However, subsequent treatment with TCDD for 3 hou rs did not result in reductions in the level of Ahr. This result is in contrast to experiments performed in mammalian cell culture where the Ah b 2 allele expressed in C 2 C 12 cells (mouse skeletal muscle myoblasts) was reduced by 90% after 2 3 hour exposu re to TCDD (Pollenz 1996).
97 FIGURE 3.18: Effect of ligand treatment on Ahr protein in induced cultures. Ahr and Arnt proteins were induced with the addition of 2% galactose to the culture medium for three hours prior to treatment with 20nM TCDD or DMSO. TCDD or DMSO treatment was carried out for an additional three hours in the presence of galactose. Yeast protein samples were prepared using 20% TCA. Western blotting and antibody staining revealed expression of Ahr and Arnt proteins after three hours of induction and the expression levels increased after three additional hours in the presence or absence of TCDD. Undaunted by the previous results, the KHSY1566 strain was tested for the effect of TCDD on Ahr protein levels from whole cell extracts. A s discussed for Figure 3.13, it was a concern that the TCA precipitation method for protein extraction may have been bringing down functional Ahr proteins along with misfolded Ahr as a result of overexpression from the GAL1 promoter. If this was the case, the TCA sample preparation method may have impaired the ability to visualize ligand mediated degradation of the receptor in the previous experiments. For this experiment, a culture of KHSY1566 was grown in the presence of galactose for 3 hours and was the n split in half. Half of the culture was treated with 20nM TCDD and the other was treated with DMSO as a control still in the presence of galactose. Protein samples were prepared from these cultures by both TCA precipitation and whole cell extraction as previously described. F igure 3.19 show s the western blotting results of the TCA precipitated and soluble fractions of yeast protein after treatment with TCDD or DMSO for 0, 2, and 4 hours.
98 FIGURE 3.19 : Ahr protein levels after TCDD treatment in TCA precipitated and soluble protein fraction samples. Induction of Ahr and Arnt proteins was carried out for 3 hours with the addition of 2% galactose to the growth medium. After 3 hours, the culture was split in half and treated with either 20nM TCDD or DMSO. The cultures were quantified and 1.0 ODs of cells were prepared for protein using the TCA precipitation method. Additionally, 15.0 ODs of cells from the identical cultures were used to prepare soluble pr otein fractions Samples were prepared before exposure to ligand and 2 and 4 hours after ligand treatment. Western blotting of the samples revealed Ahr protein levels remained consistent over time with TCDD treatment in both sample preparation methods. The level of Ahr protein detected in the TCA precipitated samples after the initial three hour induction in galactose remained essentially constant throughout the experiment. An identical affect was observed in the whole cell extracts even though the ove rall level of Ahr protein loaded from the two different preparations was different. The inability to observe a ligand mediated degradation event is in contrast to published reports (Pollenz 1996) and suggests that the sample preparation method is not a va riable in our inability to detect Ahr protein degradation. As a final attempt to observe ligand mediated Ahr degradation, the issue of promoter activation was revisited. As stated before, the strength of the promoter directly impacts the level of protein that is accumulated in a cell (in conjunction with the protein half life) and in this system it could cause rapid production of Ahr protein that masks the effect of the ligand treatment. In this way, a decrease in Ahr protein levels by ligand
99 treatment is overcome by a rapid expression and synthesis. Even in stable mammalian cell culture models, the expression of the Ahr from the strong CMV promoter results in reduced magnitude of Ahr degradation of the recombinant proteins because the expression is much stronger than from the endogenous Ahr promoter (Song and Pollenz 2002 Pollenz and Dougherty 2005 ). In order to prevent synthesis of new Ahr protein once the ligand was added, the Ahr and Arnt proteins were first induced with galactose for several hours a nd then expression was turned off with the addition of glucose to the medium. TCDD or DMSO was added directly to the media at the time the promoter was turned off. It was hypothesized that a more rapid t 1/2 for the degradation of the Ahr in the presence of ligand would be observed compared to cultures treated with DMSO. It was also of interest to assess levels of Arnt protein for this experiment. Since Ahr, but not Arnt, undergoes ligand dependent degradation (Pollenz 1996), the rate of Ahr turnover cou ld be compared to Arnt and should therefore turnover more quickly if the degradation is specific to a ligand mediated event. Figure 3.20 shows a representative experiment. The results show that the Ahr and Arnt protein are reduced at 2 and 4 hours. Imp ortantly, treatment with TCDD had no impact on the level of either Ahr or Arnt proteins. Although these results were disappointing, they were not unexpected based on the previous results. For the ligand to have an impact on the level of Ahr in this syste m, the t 1/2 would have needed to be reduced to less than 30 minutes to distinguish the result from the level of reduction that was observed under control conditions.
100 FIGURE 3.20: Effect of TCDD on Ahr and Arnt protein after the promoter is turned off. A culture of KHSY1566 was grown in the presence of 2% galactose for 3 hours. That culture was then split in half, and each resulting culture was spun down to remove the galactose from the growth media. The cell pellets were resuspended in growth me dium containing 2% glucose to block to the promoter from producing more Ahr and Arnt. The cultures were immediately treated with 20nM TCDD or DMSO and placed in a shaking incubator. Yeast protein samples were prepared from each culture at 0, 2, and 4 hou rs of treatment after the promoter was turned off. Mechan ism of Ahr Degradation in Yeast All of the studies presented thus far clearly suggest that the Ahr is rapidly lost from the yeast cells when expression is turned off. The results also suggest that the ability to evaluate ligand mediated degradation is currently not supported by this model. However, the study of Ahr degradation has multiple components that include 1) normal homeostatic turnover, 2) ligand mediated turnover and 3) ligand independent turnover. Since the majority of studies that have evaluated both the ligand dependent and independent degradation of the Ahr have shown that the degradation is mediated by the 26S proteasome, the next series of experiments were carried out to determine w hether the Ahr degradation in our recombinant yeast model was also mediated by the 26S proteasome pathway. This series of experiments was critical, because it would
101 demonstrate that the loss of Ahr that is observed when the promoter was turned off was in fact related to degradation and not dilution. In addition, the ability to show that the Ahr was being degraded via the 26S proteasome in this model would allow future genetic analysis of the proteins involved in the process. Such findings might identify E3 ligases that are also functioning in mammalian cells. There are two distinct protein degradation pathways in yeast. Proteins can be degraded by way of the vacuolar system which is similar in function to the mammalian lysosome (Baba et al 1997) or by t he substrate specific ubiquitin proteasome pathway (Eisele and Wolf 2008). Studies in yeast have shown that the 26S proteasome pathway mediates degradation of unfolded cytoplasmic proteins (Lee and Goldberg 1997, Eisele and Wolf 2008). Therefore, the fol lowing experiments test ed the effect of proteasome inhibition on Ahr and Arnt stability in the recombinant yeast str a in. Generation of a Permeable Yeast Strain. Saccharomyces cerevisiae is resistant to proteasome inhibitor treatment due to the inability of these chemicals to permeate the yeast cell wall. Previous groups have generated mutant yeast strains in which a nonessential gene encoding the cell wall protein, ERG6 was disrupted ( Lee and Goldberg 1996, Lee and Goldberg 1998). Strains with the $ erg6 mutation demonstrated enhanced permeability to certain compounds. Therefore, we needed to disrupt the ERG6 gene in the Ahr and Arnt expressor strain so they could be treated with the a ppropriate inhibitors. The double knock in strain, KHSY1547, was used to generate a permeable yeast strain by knocking out ERG6 using homologous recombination. The ERG6 gene was knocked out of the KHSY1547 strain using homologous recombination as descri bed in chapter two. Primers were designed to amplify the HIS3 auxotrophic marker cassette from the pRS303 plasmid ( Sikorski and Hieter 1989 ) These primers amplified HIS3 with overhanging sequence homology to the yeast ERG6
102 gene in order to direct re combination of the HIS3 marker with the ERG6 gene The double knock in strain KHSY1547 was transformed with the resulting PCR product using lithium cations as described in chapter two. The resulting clones were selected for the $ erg6 mutation by platin g the transformed cells on medium lacking histidine. Figure 3.21A is a photo of the transformed clones that grew on the selective medium after two days in a 30¡C incubator. Several clones were selected for analysis and p roper integration of the HIS3 cass ette was confirmed via PCR. The growth rate of the ERG6 knock out was compared to the parental strain as an additional test for the transformed clones. Previous reports have shown that t he growth rate of the $ erg6 strain is about 50% lower than the grow th rate of wild type strains ( Lee and Goldberg 1998 ). Therefore, several clones were chosen and streaked on YPD medium to look for slowed growth in comparison to the parental strain FIGURE 3.21: Selection of erg6 Clones for use in Proteasome Inhibitor Studies. The HIS3 auxotrophic marker cassette was PCR amplified using primers with sequence homology to the yeast ERG6 gene. PCR product was combined with cells of the KHSY1547 strain for homologous recombination of the HIS3 cassette to the ERG6 locus. (A) Transformed cells were spread on medium lacking histidine to select for HIS3::ERG6 clones. (B) Several clones were chosen and streaked on YPD medium to test for growth compared to the parental strain. Clone #10, in the upper left quadrant, demonstrated growth comparable to the parental strain on the lower half of the plate. Clone #9 appears to have the $ erg6 mutation as it shows less growth on the plate compared to the parental strain.
103 Figure 3.21B shows the parental strain streaked on the bottom half of the plate. Two clones that grew on the plate shown in Figure 3.21A were streaked on quadrants of the top of the plate and it was placed in a 30¡C incubator for two days. The photo shows denser grow th of the parental strain that is comparable to the growth observed for clone #10. In contrast, significantly reduced growth was observed for clone #9 as compared to the parental strain. The obvious reduction in growth suggests that clone #9 may carry th e $ erg6 mutation To further confirm the erg6 mutation in clone #9, the doubling time was calculated in the presence or absence of MG132, a compound that inhibits the activity of the 26S proteasome. A study by Lee and Goldberg (1998) demonstrated that a strain with the $ erg6 mutation had a doubling time that was longer than wild type and importantly, Fujimuro (1998) showed that a similar yeast strain had an even longer doubling time with proteasome inhibitor treatment. These articles suggested that the mutation in the yeast cell wall impaired the ability of the yeast to grow at a normal rate. They also suggested that proteasome inhibitors were capable of entering the otherwise impermeable cells and showed evidence that the chemicals affected the turnov er of key proteins involved in cell cycle control. The observed reduction in growth rate due to stalled cell division also translates to an increased doubling time for the yeast. In the following experiment, the doubling time for the parental strain (KHS Y1547) was compared to the !erg6 clone that exhibited slowed growth in Figure 3.21. The double knock in strain served as the parental strain for the !erg6 mutant strain generation. The doubling time for this strain was tested in the presence and absence of 50M MG132. This dose was chosen based on a dose response experiment reported by Lee and Goldberg (1998) with 50M MG132 reported as the effective dos e The doubling time of the permeable yeast strain was also tested in the presence or absence of 50M MG132. Triplicate cultures of 1547 and the !erg6 mutant
104 were prepared at a starting optical density 600nm of 0.1. One milliliter of each culture was rem oved and quantified every hour for 8 hours. Data points were plotted in Excel for the samples after the yeast strains reached exponential growth. Using the equation of each exponential line, the time required for the cells to double was calculated. The average was taken for each replicate and the standard deviation was calculated. The results are provided in a graph shown in Figure 3.22. The results of Figure 3.22 show that KHSY1547 had a doubling time of 88.89 minutes with a standard deviation of 1.99 minutes. This strain demonstrated a doubling time comparable to wild type Saccharomyces cerevisiae that has a doubling time of approximately 90 minutes in rich medium (Sherman 2002). Treatment of the 1547 strain with MG132 did not affect the growth rate of the cells. The doubling time for KHSY1547 with MG132 was 88.12 minutes with a standard deviation of 1.28 minutes. This result is consistent with previous reports that MG132 is unable to penetrate the yeast cell wall. FIGURE 3.22: Effect of MG132 on the doubling time of the parental and permeable strains. The doubling time of KHSY1547 was tested in the presence or absence of a 26S Proteasome inhibitor, MG132. The !erg6 strain was also evaluated under the same conditions. A dose of 50M MG132 wa s added to the treatment cultures at the beginning of the experiment. Cultures were prepared in triplicate with a starting optical density (600nm) of 0.1. The density of the culture was measure every hour for 8 hours. The data points after the exponenti al growth phase was reached were plotted in Excel to calculate the average doubling time and standard deviation.
105 Next, the permeable yeast strain had a doubling time of 99.51 minutes with a standard deviation of 1.63 minutes. This result is significantl y different than the 1547 strain, but is not identical to the reports by other groups. The growth rate of a !erg6 mutant strain evaluated by Lee and Goldberg (1998) was approximately 50% lower than wild type. This would suggest a doubling time of approxi mately 135 minutes in the permeable strain. Although the observed growth rate was not identical to published reports, the doubling time for the !erg6 strain was markedly longer than the parental strains and was measured in the presence of MG132. The doub ling time was 149.71 minutes with a standard deviation of 4.89 minutes in the permeable strain treated with 50M MG132. This data is consistent with a report by Fujimuro (1998), where the doubling time for a similar permeable strain was increased in the p resence of MG132. The effect of MG132 on growth rate is notably only present in the permeable strain, as shown in Figure 3.21, suggesting that the compound is indeed able to enter the cells in this strain. Therefore, the !erg6 strain was used in the following s tudies to evaluate the mechanism of Ahr protein degradation in the recombinant expression mod e l. Ahr Protein Stability with Proteasome Inhibitor Treatment. The following experiments were carried out using the permeable strain to assess Ahr and Arnt turnover in the presence of proteasome inhibitors. Figure 3.6 3.8 demonstrated rapid degradation of the exogenously expressed proteins when the inducible promoter was turned off. Using a similar experimental strategy, we evaluated Ahr and Arnt levels when the promoter was turned off in the presence or absence of MG132. Since MG132 is known to block the activity of the 26S proteasome, the results of this experiment would determine whether ligand independent turnover of Ahr and Arnt in yeast is proteasome mediated. It was hypothesized that treatment with MG132 would block degradation of
106 Ahr and Arnt if the turnover was specifically mediated by the proteasome. Alternatively, if the MG132 has no effect on Ahr and Arnt prote in levels, a dilution effect caused by rapid division of yeast cells may be responsible for their degradation. The experimental paradigm for these studies was similar to that described for Figures 3.6 3.8. However, the !erg6 mutant strain was used due to its increased permeability to proteasome inhibitors. An overnight culture of this strain was used to prepare an induction culture with a starting OD 600 of 0.1 with 2% galactose. The culture was placed in a 30¡C shaking incubator to induce expression of A hr and Arnt proteins for three hours. After three hours, the culture was split in half. For both resulting cultures, the promoter was turned off with the removal of galactose and addition of 2% glucose to the medium, as previously described. A schematic for the induction and treatment time course is shown in Figure 3.23A. In this experiment, one culture was treated with 50M MG132 and the other was treated with DMSO. Again, the effective dose of MG132 treatment in yeast was determined by Lee and Goldbe rg (1998). At the zero time point, the inducible promoter was turned off together with MG132 or DMSO treatment. Duplicate 1.0 OD protein samples were prepared from each culture at the zero time point. The cultures were returned to the 30¡C shaking incub ator and protein samples were prepared again after 1, 2, 3, and 4 hours. The western blotting results shown in Figure 3.23B reveal equal levels of Ahr protein detected at the zero time point between both cultures. When the inducible promoter is turned off Ahr is rapidly lost in the DMSO treated culture with a t 1/2 of approximately 1 hour. This result is consistent similar with experiments carried out in KHSY1547, shown in Figures 3.6 3.8. However, treatment of the permeable strain with MG132 shifted the t 1/2 of Ahr to approximately 3.5 hours. While the level of Ahr protein in
107 FIGURE 3.23: Effect of MG132 on Ahr Degradation in Yeast. (A) A schematic for the induction and treatment of the permeable recombinant expression strain. A culture was prepared in 2% galactose for a 3 hour induction of Ahr and Arnt proteins. Then, t he galactose indu cible promoter was turned off and the culture was split into t wo equal samples, dosed with 50M MG132 or DMSO. Duplicate samples of 1.0 OD s were removed from each culture at 0, 1, 2, 3 and 4 hours following treatment and protein samples were prepared using 20% TCA. (B) Protein samples prepared from equal numbers of cells were resolved on 7% SDS PAGE gels. Western blotting was carried out us ing a semi dry blotting apparatus. The membrane was stained with 1g/ml anti Ahr polyclonal antibodies and ECL reagent was used for detection. (C) A hr protein expression was quantified using ImageJ software and was plotted such that Ahr expression at the zero time point is equivalent to 100%. Note the rapid Ahr protein turnover that was undetectable after three hours when the culture was treated with DMSO. However, Ahr protein levels were markedly higher than the control when treated with 50uM MG132. T his suggests that the observed turnover of Ahr and Arnt in Figures 3.6 3.8 was not a dilution effect caused by the yeast cell division, but is evidence of degradation mediated by the 26S proteas o me. the MG132 treated culture is still reduced over time, this loss of protein is consistent with the dilution of the protein from the expanding cultures. This finding is illustrated in Figure 3.24 and shows that the level of Ahr in the presence of MG132 trends toward the predicted loss based on dilution within the budding cells. It is important to note that this experiment was carried out several times and the same result was consistently observed with MG132 treatment.
108 FIGURE 3.24: Ahr protein levels shift towards predicted levels with MG132 treatment The histogram from Figure 3.22 is shown above with the addition of the green line that repre sents the predicted level of Ahr that would be present in the cells based on the c ulture densities Note that the Ah r protein level decreased over time and the trend is shifted to the right toward the predicted trend line with MG132 treatment. This observation suggests that the turnover of Ahr presented in this and previous figures was not a dilution effect caused by t he yeast cell division, but is evidence of degradation mediated by the 26S proteasome. The ability to detect higher levels of Ahr protein in the presence of MG132 demonstrates that Ahr degradation in yeast terminates through a proteaosme mediated mechanism. However, it is unclear whether this is specific to the Ahr. To determine the effect of MG132 treatment on Arnt protein degradation in yeast, the same protein samples that were evaluated in the Ahr studies (Figure 3.23 3.24) were used in western blotting with antibodies specific for the Arnt protein. The results of this experiment are sh own in Figure 3.25. Similar to the results observed with the Ahr, the Arnt protein levels were stabilized with proteasome inhibitor treatment. The western blots in Figure 3.25B reveal steady turnover of Arnt in the DMSO treated culture. In contrast, Ar nt protein is maintained at high levels when the promoter is turned off in the presence of MG132. Quantification of the western blots is shown in Figure 3.25C. Importantly, the trend for
109 Arnt protein loss in the MG132 treated cultures is consistent with the expected dilution of the protein from the yeast cultures shown in Figure 3.26. FIGURE 3.25: Effect of MG132 on Arnt Degradation in Yeast (A) A schematic for the induction and treatment of the permeable recombinant expression strain. A culture was prepared in 2% galactose for a 3 hour induction of Ahr and Arnt proteins. Then, t he galactose indu cible promoter was turned off and the culture was split into t wo equal samples, dosed with 50M MG132 or DMSO. Duplicate samples of 1.0 OD s were removed from each culture at 0, 1, 2, 3 and 4 hours following treatment and protein samples were prepared using 20% TCA. (B) Protein samples prepared from equal numbers of cells were resolved on 7% SDS PAGE gels. Western blotting was carried out using a semi d ry blotting apparatus. The membrane was stained with 0.5g/ml anti Arnt polyclonal antibodies and ECL reagent was used for detection. (C) A rnt protein expression was quantified using ImageJ software and was plotted such that Arnt expression at the zero t ime point is equivalent to 100%. Note the rapid Arnt protein turnover that was undetectable after three hours when the culture was treated with DMSO. However, Arnt protein levels were markedly higher than the control when treated with 50uM MG132. This s uggests that the observed turnover of Ahr and Arnt in Figures 3.6 3.8 was not a dilution effect caused by the yeast cell division, but is evidence of degradation mediated by the 26S proteasome.
110 FIGURE 3.26: Arnt protein levels shift towards predicted le vels with MG132 treatment The histogram from Figure 3.24 is shown above with the addition of the green line that repre sents the predicted level of Arnt that would be present in the cells based on the c ulture densities Note that the Arnt protein level decreased over time and the trend is shifted to the right toward the predicted trend line with MG132 treatment. This observation suggests that the turnover of Arnt shown in this and previous figures was not a dilution effe ct caused by the yeast cell division, but is evidence of degradation mediated by the 26S proteasome. Collectively, t he results presented in Figures 3.23 3.26 suggest that the reduction in Ahr and Arnt proteins over time is not due to dilution of the proteins from the yeast cultures. Instead, the results show that Ahr and Arnt proteins are being degra ded via the proteasome pathway when expressed exogenously in the yeast expression model. Based on these data, it appears that Ahr and Arnt proteins are degraded via the 26S proteasome when expressed in yea st strains. The implications of this result will be expanded and discussed further in chapte r 4. Summary of Experimental Results AHR and ARNT cDNAs were stably integrated into the genome of strains of S. cerevisiae. Expression of Ahr and Arnt proteins was induced upon activation of the GAL1 promoter.
111 The level of induction of proteins could be modulated with varying levels of galactose. Ahr and Arnt form functional heterodimers capable of binding XREs and driving transcription of a synthetic reporter. Several ligands at varying doses were shown to activate the recombinant Ahr signaling pathway. Ahr and Arnt protein levels were reduced when the inducible promoter was turned off. Modulation of induc tion conditions and sample preparation did not produce stable levels of Ahr and Arnt Degradation was not impacted with increased expression of mammalian chaperone proteins. L igand treatment does not illicit a n observable degradation event in our yeast model. The loss of Ahr and Arnt proteins is reversed when the 26S proteasome pathway is blocked.
112 CHAPTER 4: CONCLUSIONS, IMPLICATIONS, AND FUTURE DIRECTIONS Conclusions and Implications The data presented throughout this manuscript demonstrates an in depth investigation into the expression and stability of murine Ahr and Arnt proteins when expressed in Saccharomyces cerevisiae. The first aim of this research was to construct yeast strain s expressing Ahr and Arnt under control of the galactose inducible GAL1 promoter. Generation of the various strains is detailed in chapter 2. However, it is important to note that careful consideration went into the construction of these strains. The exp ressor strains were created so that a single copy of each cDNA was stably integrated downstream of an inducible promoter. The strains were produced in this manner to allow for modulation of the expression of the protein levels and to avoid overexpression that is likely occurring when proteins are expressed from a constitutive promoter. The ability to detect and control Ahr and Arnt protein expression was highlighted in chapter 3. Expression of Ahr and Arnt under the GAL1 promoter is Detectable in Yeast The results of these analyses showed for the first time that mammalian Ahr and Arnt protein could be detected in a recombinant yeast model using a western blotting approach. Importantly, the Ahr and Arnt proteins were detectable within minutes of
113 galac tose induction. These results demonstrated that extensive incubation with galactose is not required for GAL1 promoter driven expression to obtain detectable levels of protein. This was an important observation because these strains were constructed with the goal of generating a functional signaling pathway. It was crucial to have a regulated system that could be used to assess overexpression of Ahr and Arnt so that the integrity of the pathway is maintained. The use of the GAL1 promoter allowed the leve ls of expressed proteins to be carefully modulated and this is demonstrated in Figures 3.9 3.11. The results show that reducing the amount of galactose to the growth cultures coincided with induction of lower amounts of Ahr and Arnt proteins; however the modification of those induction conditions did not produce stable levels of Ahr and Arnt. Importantly, this could not have been evaluated if a constitutively active promoter had been used in these studies. Mammalian Ahr and Arnt Proteins are Functional w hen Expressed in Yeast. A series of reporter studies revealed that the Ahr signaling pathway could be activated in a ligand dependent manner in the yeast model. Reporter assays showed that Ahr and Arnt proteins can function as transcription factors when activated with TCDD and o ther ligands to drive transcription of an artificial reporter. Importantly, the lack of induction of reporter activity in the absence of ligand showed that only the ligand bound Ahr interacted with Arnt to form the DNA binding species when expressed in ye ast. This observation was critical as it confirmed that the basic components of the Ahr signaling pathway were intact in the yeast model. Additionally, these reporter assays suggested that the Ahr pathway was not falsely activated by any component of the yeast culture medium. It also showed that the level of protein induction in our yeast model was perhaps more "physiological" since other groups observed reporter activity in the absence of ligand due to overexpression of Ahr and Arnt.
114 Ahr and Arnt Pro tein Levels are reduced in Yeast Cells over Time. A major finding in this study was that Ahr and Arnt protein levels were reduced upon transcriptional repression of the GAL1 promoter. T he reduction of the Ahr protein was not a result of premature initiat ion of the Ahr pathway by endogenous ligands present in the yeast or yeast media, as evidenced by baseline levels of galactosidase induction in vehicle treated controls. It was anticipated that since yeast protein samples were prepared during a time cou rse when the promoter was turned off, that the recombinant proteins would be "diluted" from the culture over time as the yeast propagated in the absence of galactose. Importantly, the loss of Ahr and Arnt protein levels from the protein samples was more r apid than could be explained by dilution through yeast cell division (budding). This result implied that Ahr and Arnt proteins were being actively degraded and may be unstable when expressed in ye ast. This result was not completely unexpected, since Ahr and Arnt are foreign proteins and y east do not express bHLH/PAS orthologs, but the findings clearly complicated the goal of a model that could be evaluated for ligand mediated degradation. Overexpression of Chaperone Proteins does not Impact the Loss o f Ahr and Arnt It was possible that Ahr and Arnt turnover was the result of overexpression in our recombinant yeast strains. Thus, it was hypothesized that the chaperone proteins required for Ahr folding may not be functioning efficiently in yeast. Thi s was the most logical hypothesis because of the critical role Hsp90 plays in the stability of Ahr in higher eukaryotes. It has been shown that treatment of mammalian cells with Hsp90 inhibitors that disrupt the AhrHsp90 interaction, causes a rapid turnov er of Ah receptor protein in the
115 absence of ligand treatment. It was hypothesized th at the interaction between the recombinant Ahr and Hsp90 may be compromised in yeast due to h i gher levels of Ahr than the yeast Hsp90 homologues. This imbalance could the n lead to rapid Ahr degradation in the absence of ligand. Interestingly, the results showed that overexpression of mammalian Hsp90 did not increase the stability of the proteins. It has also been shown that overexpression of exogenous proteins correlat es with an increase in expression of chaperone proteins to facilitate protein folding. However, in cases of gross protein overexpression, the cells may be unable to produce enough chaperones to properly fold the newly synthesized proteins. Although we di d not feel that the Ahr and Arnt proteins were being overexpressed to that extreme, it was possible that the rate of Ahr and Arnt synthesis was exceeding correlate expression of the yeast chaperones. Unfolded proteins in the cytoplasm of Saccharomyces cer evisiae are then targeted for degradation via the 26S proteasome. Again, the results of this study suggest that overexpression of mammalian Hsp90 did not impact Ahr or Arnt stability in these yeast strains. It should also be noted that this theory does n ot align with the ability to detect reporter activity. Specific activation of the Ahr through ligand binding implies that some population of the expressed Ahr pool is folded properly since unfolded Ahr will not be activated by ligand. Ligand Mediated Deg radation of Ahr was not detected in Yeast. Ligand mediated degradation of the Ahr could not be detected under any of the conditions tested during the course of these studies. Ahr protein levels were unchanged with ligand treatment when the promoter was a ctive or repressed. Even when cells were treated with ligand at the same time as induction, no significant loss of Ahr was detected. These observations were in contrast to our anticipated result and complicated the use of this model for assessment of Ahr degradation. The original aim of the research project was
116 to produce a model to genetically evaluate Ahr degradation in yeast in a ligand dependent manner, but a ligand dependent effect could not be observed. Ahr and Arnt Protein Degradation is Reversed with Proteasome Inhibitor Treatment. As stated in C hapter 3, yeast protein degradation occurs via the substrate specific 26S proteasome pathway or by way of the non specific vacuolar degradation pathway. It was of interest to determine which mechanism was responsible for the degradation of A hr and Arnt. This information c ould indicate whether these proteins were being shuttled nonspecifically to the vacuole, or if the proteins were being specifically targeted t o the proteasome for degradation. The results of this study revealed steady turnover of Ahr and Arnt in the DMSO treated culture but these proteins were maintained at high levels when the promoter was turned off in the presence of MG132, a chemical known to block the activity of the 26S proteasome. The ability to detect higher levels of Ahr and Arnt protein in the presence of MG132 demonstrates that their degradation in yeast terminates through a proteasome mediated mechanism. This is a positive finding because it does suggest that the study of Ahr degradation via the proteasome is still possible in this model a nd may be able to yield results that are translatable to a mammalian system (see Future Studi e s). Hypothesis for Ahr and Arnt turnover in Yeast. These data provide evidence that the mouse Ahr and Arnt proteins are being degraded through the 26S proteasome in the yeast strains due to high levels of expression from an inducible promoter. The mechanism of degradation may occur through a degradation signal that is not recognized in mammals and is specific to yeast. Importantly, Arnt is not degraded in a ligand specific manner and has a much longer half life in mammals than in these studies. The observation that Arnt is also turning over rapidly in t he yeast via the 26S
117 proteasome, suggests that it may also have a putative sequence recognized by the yeast enzymes. Unfortunately, there is no database of putative degradation signals in yeast (or in mammals for that matter) that can be used to assess th is question through amino acid analysis. To address this question directly would require mutagenesis of the Ahr and Arnt (although the yeast is amenable to selection schemes for such purposes). Future Direct i ons Further Analysis of Ahr and Arnt in Yeast There are additional experiments that would be of interest to provide further confirmation of the conclusions presented in this text. In particular, with regards to ligand treatment, it would be of interest to observe the effect of other known Ahr agonists on Ahr protein in these strains. Several ligands were tested and produced varying degrees of galatosidase reporter activity as shown in Figure 3.4. It would be of interest to evaluate Ahr and Arnt protein levels via western blotting with vari ous ligand treatments at a range of doses. However, it is unlikely that a ligand dependent degradation event would be detected, regardless of the ligand or dose used. This hypothesis is based on the fact that the ligand and dose used for the western blot ting studies produced significant levels of galactosidase reporter activity in parallel experiments. The gal results suggest that the Ah receptor was activated under the conditions tested and degradation would have been detectable according to the mam malian model. Another experimental approach that was not explored in this work would address the rate of yeast cell division. It has been shown that the yeast cell cycle can be arrested in S phase with the addition of hydroxyurea to the growth medium ( Ke et al 2003). In this work, the galactose inducible promoter was repressed together with hydroxyurea treatment. This, in effect, allowed for the detection of exogenous protein turnover in the absence of cell division. With this method, we could evalua te the stability
118 of Ahr and Arnt proteins without having to consider the dilution of the exogenous proteins due to cell division. It should be noted however, that Ahr is a transcription factor that has been shown to play a role in cell cycle progression ( Puga et al 2002). Since Ahr plays a role in cell cycle progression arresting the cell cycle du ring the course of the experiments could influence the integrity of the signaling pathway. Analysis of Ahr and Arnt Stability in E3 Knock out Strains. W hile this yeast model does not appear to be amenable for analysis of ligand mediated degradation of Ahr, these strains may be useful for the analysis of other aspects of Ahr signaling. Since these strains exhibit ligand independent degradation of Ahr and Arnt, it would be of interest to further explore this occurrence. One way to identify proteins involved in mediating the turnover of Ahr and Arnt in yeast would be to knock out candidate genes. A marker cassette is used to disrupt the open reading frame of evolutionarily conserved genes in the ubiquitin proteasome pathway in the Ahr and Arnt expressing strains. The goal would be to create a strain of yeast that does not degrade Ahr and Arnt p rotein s indicating that the knocked out gene is involved in the degradation process. A list of sixty yeast genes involved in the ubiquitin proteasome pathway was generated using the Saccharomyces cerevisiae genome database and is provided in Appendix C One particular E3 ligase was knocked out of the Ahr and Arnt expressor strain. The details concerning the generation of the !ubr1 knockout strain is provided in Appendix C Additionally, western blots stain ed for Ahr protein in the !ubr1 knockout strain are included in Figure C.1. The preliminary results of this work do not suggest that UBR1 is a mediator of Ahr degradation in yeast However, this was an important E3 ligase to evaluate be cause of its known role in degradation of unfolded cytoplasmic proteins (Eisele and Wolf 2008 )
119 Overexpression of Mammalian E3 Ligases in the Recombinant Yeast Strain. Since ligand mediated degradation of Ahr was not observed under the conditions evaluated, it was hypothesized that the particular E3 ligase that targets Ahr for degradation is n ot conserved from mammals to yeast. Therefore, it would be interesting to observe Ahr protein levels in the recombinant strain with overexpression of mammalian E3s. Using a candidate approach, particular E3s implicated in degradation of Ahr and other bHL H proteins could be cloned into yeast expression vectors. Transformation of these expression vectors and subsequent overexpression of the E3s in the Ahr and Arnt yeast strains could reveal a potential target that degrades the receptor. With this method, one could evaluate the effect of individual candidate E3s in a systematic way and in a simpler cell ular context As discussed in chapter 1, s eve ral reports have implicated Chip ( Carboxyl t erminus of Hsp70 interacting p rotein) as the E3 liga se responsible for targeting Ahr for degradation via the ubiquitin proteasom e pathway (Lees et al 2003, Morales and Perdew 2007). Although several reports have refuted the role of Chip in Ahr degradation using cell culture models, it would be of interest to test the eff ect of Chip overexpression in the recombinant yeast strain since yeast do not express a homologue. It has also been suggested that cullin4B interacts with ligand bound Ahr (Wormke et al 2003, Ohtake et al 2007). These studies provided evidence of an inte raction between these proteins, but did not offer direct evidence that Ahr degradation is mediated through cullin 4B. Again, one could evaluate the effect of overexpression of each individual ca ndidate E3 in a systematic way in the absence of other compli cating fact o rs Analysis of Ahr and Arnt Isoforms using a Yeast Model. While the re are many ways to further analyze Ahr and Arnt protein degradation in these strains, it is important to note that these strains could be used to analyze other aspects of the Ahr signaling
120 pathway. For example, the construction of a strain expressing Ah b 1 allele (t hat is naturally more stable than the Ah b 2 allele used in the recombinant strain ) would allow for a direct comparison of the signaling of the two different proteins. These two alleles of the receptor have been compared in vitro and in vivo but have yet to be expressed comparatively in yeast. A similar comparison could be made between the A rnt 1 and Arnt 2 isoforms to determine how they directly influence Ahr signaling in this model In this cas e, the Ahr signaling pathw ay could be constructed using Arn t 2 in place of Arnt 1. This new strain could be used to evaluate the ability of Ahr and Arnt 2 to form heterodimers and bind the galactosidase reporter in the absence of Arnt 1. This would be a very useful experiment, as an Arnt 1 knockout is lethal in mammals (Kozak et al 1997) thus complicating the abilit y to assess AhrArnt 2 interactions.
121 CHAPTER FIVE: MATERIALS AND METHODS Mate r ials Buffers P hosphate buffer is 20mM NaPi pH 7.7, 300mM NaoAc, 10% glycerol, pr otease inhibitor cocktail LacZ buffer is 60mM Na 2 HPO 4 40mM NaH 2 PO 4 1mM MgCl 2 and 10mM KCl. SDS PAGE sample buffer is 125 mM Tris, pH 6.8, 4% SDS, 25% glycerol, 4 mM EDTA, 20 mM dithiothre i tol, 0.005% bromophenol blue. Tris buffered saline is 50 mM Tris and 150 mM Na Cl, pH 7.5. TTBS is 50 mM Tris, 0.2% Tween 20, 300 mM N aCl, pH 7.5. TTBS+ is 50 mM Tris, 0.5% Tween 20, 300 mM NaCl, pH 7.5. B locking milk is 5% dry milk in TT B S. Reagents and Chem icals Restriction enzymes and buffers, AflII and ClaI, were purchased from Promega (Madison, WI). Alkaline phosphatase was purchased from New England Biolabs (Ipswich, M A ). 2 ,3,7,8 tetracholorodibenzo p dioxin (TCDD) was obtained from Radian Corporation (Austin, TX) and was solubilized in dimethyl sulfoxide (Me 2 SO) obtained from Research Organics Benzo[a]pyrene (BAP), 3 methylcholanthrene (3 MC) and naphthoflavone (!NF) were purchased from Sigma Aldrich (St. Louis, MO) and solubilized in dimethyl sulfoxide (Me 2 SO). MG132, Radiciol, and Novobiocin were pu rchased in solution from CalBiochem. G418 was purchased from Axxora LLC ( San Diego, CA ).
122 Yeast Media YPD (yeast extract peptone dextrose) consists of 10 g/l yeast extract (Fisher Scientific), 20 g/l Bacto peptone (BD Diagnostic Systems), and 2% glucose (Fisher Scienti fic). YPD plates consisted of the above components with the addition of bacto agar (BD Diagnostic Systems) at a concentration of 20 g/l. YPG consists of 10 g/l yeast extract (Fisher Scientific), 20 g/l Bacto peptone (BD Diagnostic S ystems), 20g/L agar, and 3 % (v/v) glycerol ( Sigma Aldrich). Synthetic medium consists of 6.7 g/l bacto yeast nitrogen base without amino acids (Fisher Scientific), 10g/l dropout amino acid mix minus tryptophan, uracil, o r leucine (US Biologicals), and 2% glucose (Fisher Scienti fic) or 2% galactose (Acros Organic s ). Plasmids. Various constructs were obtained from the sources listed below. TABLE 5.1: So u rces For Plasmids Plasmid Name Use in Experiments Source pRS303 HIS cassette Sikorski and Hieter 1989 pRS304 TRP cassette Sikorski and Hieter 1989 pRS306 URA cassette Sikorski and Hieter 1989 pFA6a PGAL1KanMx GAL promoter Longtine et al 1998 pLXRE 5Z Reporter construct Cox and Miller 2003 pLNcx 2 AHR b 2 Ahr cDNA Pollenz and Dougherty 2005 pCDNA ARNT Arnt cDNA Dougherty and Pollenz 20 0 8 Antibodies. Specific antibodies against A hr and A rnt were identical to those described previously (Holmes and Pollenz, 1997; Pollenz et al., 1994). Antibody to the Xpress tag was purchased from Invitrogen and tubulin antibody was purchased from Sigma Aldrich (St. Louis, MO) Peroxidase conjugated goat anti mouse antibody and peroxidase conjugated goat anti rabbit antibody was purchased from Jackso n
123 Immunoresearch. Antibodies were used at various concentrations that are included in the appropriate figure legends. Methods Yeast Transformatio ns. Yeast transformations were carried out according the the method of Gietz and Woods (2006) using lithium cations, single stranded carrier DNA, and polyethylene glycol. Parental yeast strains were streaked on pre warmed YPD agar plates and placed in a 30¡ C incubator for two days. A single colony of the parental strain was used to inoculate 2 ml of YPD liquid media that was placed in a 30¡C shaking incubator overnight. This culture was used to seed a 25 ml culture with an OD 600 = 0.2. When the OD 600 of t he culture reached 0.8 (+/ 0.04), the culture was transferred to a conical tube and centrifuged at 2000 rpm for 2 minutes. The supernatant was removed, rinsed with water, and centrifuged again. The cell pellet was resuspended with 1 ml of sterile 100mM LiAc (pH 7.4) and transferred to a sterile 1.5 ml Eppendorf tube. After centrifugation at 14000 rpm for 30 seconds, the supernatant was removed and the pellet was resuspended in the equivalent of 1% of the final culture volume of 100mM LiAc (pH7.4). The cell suspension was divided equally between four Eppendorf tubes and each was centrifuged at 14000 rpm for 1 minute. The supernatant was removed and each pellet was combined with the following transformation reagents without disturbing the cell pellet; 24 0l of 50% (w/v) PEG 3350, 36l of 1M LiAc (pH7.4), 75l PCR product or plasmid DNA in water, and 5l of boiled/snap cooled salmon sperm DNA (2mg/ml). Each tube was vortexed for one minute to resuspend the cell pellet in the transformation reagents. The tubes were placed in a 30¡C incubator for 30 minutes without shaking and were heat shocked in a 45¡C water bath for 15 minutes. The tubes were centrifuged at 7000 rpm for 1 minute and the supernatant was removed. The cell
124 pellet was combined with 100l o f sterile water and was spread on selective agar media. After two days, tran sformants were streaked again on selective media to obtain single colonies for further analys i s. Random Sporul ation M eiotic progeny was generated via random sporulation (Rockmill et al 1991) Diploids heterozygous for the desired cDNA integrations w ere grown overnight at 30¡C in YPD, washed, transferred to 0.1% potassium acetate (Fisher Scientific), and incubated for 5 days in a 30 ¡C shaking incubator Asci were i ncub ated in the presence of 500 g/ml zymolase (MP Biomedicals) in 1 M sorbitol (Fisher Scientific) for 20 min at 30¡C and enriched for haploid spores Spores were plated on YPD, incubated at 30 ¡C, and genotyped by spotting on synthetic drop out media (US Biological) to detect the presence of TRP1 and URA3 marker cassettes linked to the AHR and ARNT cDNAs The presence of the GAL1 promoter linked to the kanMX6 cassette was detect ed by the ability of haploids to grow on YPD supplem e nted with 200 g/ml G418 (Axxora LLC, San Diego, CA). Activation and Repression of the Galactose inducible Promoter. Yeast strains with integrated GAL1 promoter sequences were streaked on agar media and placed in a 30¡C incubator for two days. A single colony was used to inocul ate liquid YPD medium and was placed in a 30¡C shaking incubator overnight. The following day, the culture was centrifuged at 2000 rpm for 2 minutes. The YPD medium was removed, the pellet was rinsed with water, and the sample was centrifuged again. The supernatant was discarded and the cell pellet was resuspended in growth medium supplemented with 2% galactose (Acros Organics). The culture was returned to the 30¡C shaking incubator for various time courses to allow for protein induction. Repression of the inducible promoter occurred upon addition of 2% glucose to the culture med i um.
125 Preparation of Soluble Protein Frac tions Soluble protein fractions were prepared using a cold phosphate buffer extraction method. A liquid culture of the desired strain was propagated under conditions required for the given experiment. After the induction/treatment period, a sample of each culture was used to determine the cell density at OD 600 To obtain samples of even cell number, 15 ODs were removed from each cultur e. The optical density was converted to OD units by dividing 15 ODs by the optical density to determine the volume required. Each culture was centrifuged at 2000 rpm for 2 minutes and the pellet was suspended in 1ml of cold phosphate buffer (20mM NaPi pH 7.7, 300mM NaoAc, 10% glycerol, protease inhibitor cocktail). Each sample was centrifuged at 5000 rpm in a refrigerated centrifuge at 4¡C for five minutes. The pellet was resuspended in 100l of cold phosphate buffer and an equal volume of glass beads, and the tubes were placed in a mini bead beater for 4 times for one minute each, and placed on ice for two minutes in between each. The resulting lysate was spun again at 14,000 rpm for 30 minutes at 4¡C. The supernatant was then combined with an equal v olume of 20% trichloroacetic acid (TCA) followed by centrifugation at 14 ,000 rpm for one minute. The resulting protein pellet was resuspended in 100l of SDS PAGE sample loading buffe r and the pH of the sample buffer was adjusted with the addition of 2 M Tris. Each sample was centrifuged at 14000 rpm for one minute to remove cell debris and the supernatant was used in western blotting. P reparation of TCA Precipitated Protein Samples. Total protein was prepared from the yeast cultures according to the p rocedure of Wright et al ( 1989) using 20% TCA to precipitate proteins. The yeast culture density was quantified at OD 600 To obtain samples of even cell number, 1.5 ODs were removed from each culture. The optical density was converted to OD units by div iding 1.5 ODs by the optical density to
126 determine the volume required. E ach culture was centrifuged at 2000 rpm for 2 minutes and the medium was aspirated The cell pellets, containing 1.5 ODs of cells, were combined with 100l of 20% TCA and an equivale nt volume of glass beads and were placed in a mini bead beater for 4 minutes. Each sample was transferred to a fresh tube which was then centrifuged at 14 ,000 rpm for one minute. The resulting protein pellet was resuspended in 100l of SDS PAGE sample loading buffe r and the pH of the sample buffer was adjusted with the addition of 2 M Tris. Each sample was centrifuged at 1400 0 rpm for one minute to remove cell debris and the supernatant was used in western blotting. Western blotting Western bl otting and antibody detection were carried out as previously described (Pollenz, 1996) Protein samples we re resolved by denaturing electrophoresis on 7% SDS PAGE gels and were electrophoretically transferred to nitrocell ulose. Protein detection was carried out with varying concentrations of primary antibody in blocking milk for 1 hour The nitrocellulose membranes were washed with three changes of TTBS or TTBS+ solution for a total of 45 min. The blot was then incubated in blocking milk containing a 1:10,0 00 dilution of goat anti rabbit or goat anti mouse HRP secondary antibodies for 1 h our and again washed in TTBS or TTBS+ as above. Prior to detection, the blots were washed with PBS for 5 min utes Protein b ands were visualized with the enhanced chemiluminescence (ECL) kit (Amersham Biosciences, Piscataway, NJ). The relative concentration of target proteins w as determined by computer analysis of the autoradiographs. Blots were quantified first by scanning images using a Cannon Canoscan 8800F scan ner. Raw densitometric values were quantified using ImageJ software and were divided by Tubulin levels in o rder to normalize the target protein values in each sam p le.
127 Reporter Analysis. galactosidase activity was tested according to Miller (1997). One m i l liliter of yeast liqu id culture was removed from larger test cultures. Each sample was centrifuged for 1 minute at 14 ,000 rpm, the medium was aspirated, and the cell pellet was combined with 500l of lacZ buffer (60mM Na 2 HPO 4 40mM NaH 2 PO 4 1mM MgC l 2 10mM KCL, and ONPG ( 4mg/ml in lac Z buffer )). The pellet was resuspended in the solution and placed in a 37¡ C water bath and upon observation of a color change ; the reactions were stopped with the addition of 1.5M sodium carbonate. The samples were centrifuged at 14 ,000 rpm for one minute to pellet the cells and the optical density at 420nm was measured for the supernatant. The observed optical density was normalized using the following equation: (Absorption @ 420nm x 1000) / (Absorbtion @ 595nm) X (ml of cell suspension added) X (minutes of reaction time).
128 LITERATURE CITED Adachi, J., Y. Mori, S. Matsui, H. Kitagawa, C.A. Miller, III, T. Kato, K. Saeki, and T. Matsud a. 2001. Indirubin and Indigo are Potent Aryl Hydrocarbon Receptor Ligands Present in Human U r ine. The Journal of Biological Chemistry 276:31475 31478. Amati, B., S. Dalton, M.W. Brooks, T.D. Littlewood, G.I. Evan, and H. Land 1992. Transcriptional activation by the human c MYC oncoprotein in yeast requires interaction with Max. Nature 359:423 426. Baba, M., M. Osumi, S.V. Scott D.J. Klionsky, and Y. O hsumi 1997. Two Distinct Pathways for Targeting Proteins from the Cytoplasm to the Vacuole/Lysosome. The Journal of Cell Biology 139:1687 1695. Beischlag, T.V., S. Wang, D.W. Rose, J. Torchia, S. Reisz Porszasz, K. Muhammad, W.E. Nelson, M.R. Probst, M.G. Rosenfeld, and O. Hankinson. 2002. Recruitment of the NCoA/SRC 1/p160 Family of Transcriptional Coactivators by the Aryl Hydrocarbon Receptor/Ar yl Hydrocarbon Receptor Nuclear Translocator Complex. Molecular and Cellular B i ology :4319 4333. Berben, G., M. Legrain, V. Gilliquet, and F. Hilger. 1990. The Yeast Regulator Gene PHO4 Encodes and Helix Loop Helix Motif. Yeast 6:451 454. Bernassola, F., M. Karin, A. Ciechanover, and G. M elino 2008. The HECT Family of E3 Ubiquitin Ligases: Multiple Players in Cancer Development. C a ncer Cell 14:10 21. Betting, J., and W. Seufert. 1996. A Yeast Ubc9 M utant Protein with Temperature sensitive in Vivo Funct ion is Subject to Conditional Proteolys i s by a Ubiquitin and Proteasome dependent Pathway. The Journal of Biological Chemistry 271:25790 25796. Borkovich, K.A., F.W. Farrellu, D.B. Finkelstein, J. Taulien, and S. Lindquist. 1989. Hsp82 Is an Essential Pr otein That Is Required in Higher Concentrations for
129 Growth of Cells at Higher Temperat u res. Molecular and Cellular Biology 9:3919 3930. Bradfield, C.A., and A. Poland. 1988. A Competitive Binding Assay for 184.108.40.206 tetrachlorodibenzo p dioxin and Related Ligands of the Ah Rece p tor. Molecular Pharmacology 34:682 688. Braliou, G.G., E. Venieris, A. Kalousi, and G. Simos. 2006. Reconstitution of Human Hypoxia Factor HIF 1 in Yeast : A Simple In Vivo System to Identify and Characterize HIF 1" Effectors. Bio chemical and Biophysical Research Communica tions 346:1289 1296. Burbach, K.M., A. Poland, and C.A. Bradfield. 1992. Cloning of the Ah Receptor cDNA Reveals a Distinctive Ligand Activated Transcription Factor. Biochemistry 89:8185 8189. Carver, L.A., an d C.A. Brad field 1997. Ligand dependent Interaction of the Aryl Hydrocarbon Receptor with a Novel Immunophilin Homolog In Vivo The Journal of Biological Chemistry 272:11452 11456. Carver, L.A., V. Jackiw, and C.A. Bradfield. 1994. The 90 kDa Heat Shock Protein Is Essential for Ah Receptor Signaling in a Yeast Expression Sy s tem. The Journal of Biological Chemistry 269:30109 30112. Carver, L.A., J.J. LaPres, S. Jain, E.E. Dunham, and C.A. Bradfield. 1998. Characterization of the Ah Receptor associated P rotein, A RA9. The Journal of Biological Chemistry 273:33580 33587. Chen, B., W.H. Pie l L. Gui, E. Bruford, and A. Monteiro. 2005. The HSP90 Family of Genes in the Human Genome: Insights into their Divergence and Evolution. Genomics 86:627 637. Chen, H. S., S.S. Singh, and G.H. P erdew 1997. The Ah Receptor is a Sensitive Target of Geldanamycin Induced Protein Turnover. Archives of Biochemistry and Biophysics 348:190 198. Chen, Y. H., and R.H. Tukey. 1996. Protein Kinase C Modulates Regulation of the CYP1A1 Gene by the Aryl Hydrocarbon Rece p tor. The Journal of Biological Chemistry 271:26261 26266. Ciechanover, A., and A.L. Schwartz. 1994. The Ubiquitin Mediated Proteolytic Pathway: Mechanisms of Recognition of the Proteolytic Substrate and Involveme nt in the Degradation of Native Cellular Prot e ins. FASEB 8:182 191.
130 Ciechanover, A., and A.L. Schwartz. 1998. The Ubiquitin Proteasome Pathway: The Complexity and Myriad Functions of Proteins Death. Proc. Natl. Acad. Sci. 95:2727 2 7 30. Cox, M.B., and C. A. Mill er III 2002. The p23 Co chaperone Facilitates Dioxin Receptor Signaling in a Yeast Model System. Toxicology Letters 129:13 21. Cox, M.B., and C.A. Miller, III. 2003. Pharmacological and Genetic Analysis of 90 kDa Heat Shock Isoprotein Aryl Hydro carbon Receptor Compl e xes. Molecular Pharmacology 64:1549 1556 Cox, M.B., and C.A. Miller. 2004. The Cooperation of heat shock protein 90 and p23 in Aryl Hydrocarbon Receptor Signaling. Cell Stress and Chaperones 9:4 20. Crews, S.T. 1998. Control of Ce ll Lineage Specific Development and Transcription by bHLH PAS Pro teins Genes and Development 12:607 620. Dale, Y.R., and S.E. Eltom. 2006. Calpain Mediated the Dioxin Induced Activation and Down Regulation of the Aryl Hydrocarbon Receptor. Molecular Pharmac ology 70:1481 1487. Davarinos, N.A., and R.S. Pollenz. 1999. Aryl Hydrocarbon Receptor Imported into the Nucleus following Ligand Binding is Rapidly Degraded via the Cytoplasmic Proteasome following Nuclear Export. The Journal of Biological Chemis try 274:28708 28715. Denison, M.S., J.M. Fisher, and J.P. Whitlock, Jr. 1988. The DNA Recognition Site for the Dioxin Ah Receptor Complex. The Journal of Biological Chemistry 263:17221 17224 Denison, M.S., and S.R. Nagy. 2003. Activation of the Aryl Hydrocarbon Receptor by Structurally Diverse Exogenous and Endogenous Chemicals. Annual Review of Pharmocological Toxicology 43:309 334. Dougherty, E.J., and R.S. Pollenz. 2008. Analysis of Ah Receptor ARNT and Ah Receptor A RNT2 Comp l exes In Vitro and in Cell Culture. Toxicological Sciences 103:191 206. Eisele, F., and D.H. Wolf. 2008. Degradation of Misfolded Protein in the Cytoplasm is Mediated by the Ubiquitin Ligase Ubr1. FEBS Letters 582:4143 4146.
131 Ema, M., N. Ohe, M Suzuki, J. Mimura, K. Sogawa, S. Ikawa, and Y. Fuji Kur iyama 1994. Dioxin Binding Activities of Polymorphic Forms of Mouse and Human Aryl hydrocarbon Recep t ors. The Journal of Biological Chemistry 269:27337 27343. Escamilla Powers, J.R., and R.C. Sear s 2007. A conserved pathway that controls c Myc protein stability through opposing phosphorylation events occurs in yeast. Journal of Biological Chemistry 282:5432 5442. Flaveny, C., R.K. Reen, A. Kusnadi, and G.H. Perdew. 2008. The Mouse and Human Ah R eceptor Differ in Recognition of LXXLL Mo t ifs. Archives of Biochemistry and Biophysics 471:215 233. Fujimuro, M., H. Takada, Y. Saeki, A. Toh e, K. Tanaka, and H. Yokosawa. 1998. Growth Dependent Change of the 26S Proteasome in Budding Yeast. Biochemical and Biophysical Research Communica tions 251:818 823. Fukunaga, B.N., M.R. Probst, S. Reisz Porszasz, and O. Hankinson. 1995. Identification of Functional Domains of the Aryl Hydrocarbon Receptor. The Journal of Biological Chemistry 270:29270 29278. Gi annone, J.V., W. Li, M. Probst, and A.B. Okey 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. Biochemical Pharmacology 55:489 497. Giannon e, J.V., A.B. Okey, and P.A. H arper 1995. Characterization of Polyclonal Antibodies to the Aromatic Hydrocarbon Receptor. Can J Physiol Pharmacol 73:7 17. Gietz, D., and R.A. Woods. 2006. Yeast Transformation by the LiAc/SS Carrier DNA/PEG Method. Methods in Molecular Biology 313:107 120. Goll, D.E., V.F. Thompson, H. Li, W. Wei, and J. Cong. 2 003. The Calpain System. Physiol Rev 83:73 1 801 Greenberg, M.L., and J.M. Lopes. 1996. Genetic Regulation of Phospholipid Biosynthesis in Saccharomyces cerevi s iae Microbiology Review 60:1 20. Gu, Y.Z., J.B. Hogenesch, and C.A. Bradfield. 2000. The PAS Superfamily: Sensors of Environmental and Developmental Signals. Annual Review of Pharmocological Toxicology 40:519 561.
132 Hanel, F., K. Peukert, and T. M under 1996. Control of Myc Max Mediated Transactivation in yeast by Natural Promoter Elements. Journal of Basic Microbiology 37:23 28. Hankinson, O. 2005. Role of Coactivators in Transcriptional Activation by the Aryl Hydrocarbon Receptor. Archives of Biochemistry and Bioph ysics 433:379 386. Hayes, S.A., and J.F. Dice. 1996. Roles of Molecular Chaperones in Protein Degradation. The Journal of Cell Biology 132:255 258. Heath Pagliuso, S., W.J. Rogers, K. Tullis, S.D. Seidel, P.H. Cenijn, A. Brouwer, and M.S. De nison 1998. Activation of the Ah Receptor by Tryptophan and Tryptophan Metabolites. Biochemistry 37:11508 11515. Heid, S.E., R.S. Pollenz, and H.I. Swanson. 1999. Role of Heat Shock Protein 90 Dissociation in Mediating Agonist Induced Activat ion of Aryl Hydrocarbon Rece p tor. Molecular Pharmacology 57:82 92. Hochstrasser, M. 1 996. Ubiquitin Dependent Protein Degradation. Annual Review of Genetics 30:405 439. Holmes, J.L., and R.S. Pollenz. 1997. Determination of Aryl Hydrocarbon Receptor Nu clear Translocator Protein Concentration and Subcellular Localization in Hepatic and Nonhepatic Cell Cultures Lines: Development of Quantitative Western Blotting Protocols for Calculation of Aryl Hydr ocarbon Receptor and Aryl Hydrocarbon Receptor Nuclear T ranslocator Protein in Total Cell Lys a tes. Molecular Pharmacology 52:202 211. Ikuta, T., H. Eguchi, T. Tachibana, Y. Yoneda, and K. Kawajiri. 1998. Nuclear Localization and Export Signals of the Human Aryl Hydrocarbon Receptor. The Journal of Biological Chem istry 273:2895 2904. Ito, H., Y. Fukuda, K. Murata, and A. Kimura. 1983. Transformation of Intact Yeast Cells Treated with Alkali Cations. Journal of Bacteriology 153:163 168. Jain, S., K.M. Dolwick, J.V. Schmidt, and C.A. Bradfield. 1994. Potent T ransactivation Domains of the Ah Receptor and the Ah Receptor Nuclear Translocator Map to their Carboxyl Ter m ini. The Journal of Biological Chemistry 269:31518 31524. Jentsch, S. 1 992. The Ubiquitin Conjugation System. Annual Review of Genetics 26:179 207.
13 3 Jiang, J., C.A. Ballinger, Y. Wu, Q. Dai, D.M. Cyr, J. Hohfeld, and C. Patterson. 2001. CHIP is a U box dependent E3 Ubiquitin Ligase. The Journal of Biological Chemistry 276:42938 42944. Kamata, R., F. Shiraishi, D. Nakajima, H. Takigami, and H. Shir aishi 2009. Mono hydroxylated Polychlorinated Biphenyls are Potent Aryl Hydrocarbon Receptor Ligands in Recombinant Yeast Assays. Toxicology in Vitro 23:736 743. Kawanishi, M., M. Sakamoto, A. Ito, K. Kishi, and T. Yagi. 2003. Construction of Reporter Yeasts for Mouse Aryl Hydrocarbon Receptor Ligand Acti v ity. Mutation Research 540:99 105. Kazlauskas, A., L. Poellinger, and I. Pongratz. 1999. Evidence that the Co chaperone p23 Regulates Ligand Responsiveness of the Dioxin (Aryl Hydrocarbon) Rece p tor. The Journal of Biological Chemistry 274:13519 13524. Ke, P. Y., C. C. Yang, I. C. Tsai, and Z. F. Chang. 2003. Degradation of Human Thymidine Kinase is Dependent on Serine 13 Phosphorylation: Involvement of the SCF Mediated Pathway. Biochemical So ciety 370:265 273. Kewley, R.J., M.L. Whitelaw, and A. Chapman Smith. 2004. The Mammalian Basic Helix Loop Helix/PAS Family of Transcriptional Regulators. The International Journal of Biochemistry & Cell Biology 36:189 204. King, R.W., M. Glotzer, a nd M.W. Kirs chner 1996. Mutagenic Analysis of the Destruction Signal of Mitotic Cyclins and Structural Characterization of Ubiquitinated Intermediates. Molecular Biology of the Cell 7:1343 1357. Kippert, F. 1995. A Rapid Permeabiliza tion Procedure for A ccurate Quantitative Determination of Galactosidase Activity in Yeast C e lls. Microbiology Letters 128:201 206. Ko, H.P., S.T. Okino, Q. Ma, and J.P. Whitlock, Jr. 1997. Transactivation Domains Facilitate Promoter Occupancy for the Dioxin Inducible CYP1 A1 Gene In V ivo. Molecular and Cellular Biology 17:3497 3507. Koegl, M., T. Hoppe, S. Schlenker, H.D. Ulrich, T.U. Mayer, and S. Jentsch. 1999. A Novel Ubiquitination Factor, E4, is Involved in Multiubiquitin Chain Assembly. Cell 96:635 644. Kohlmann, S., A. Schafer, and D.H. Wolf 2008. Ubiquitin Ligase Hul5 is Required for Fragment Specific Substrate Degradation in Endoplasmic Reticulum Associated Degradation. The Journal of Biological Chemistry 283:16374 16383.
134 Kornitzer, D., and A. Ciechanover. 20 00. Modes of Regulation of Ubiquitin Mediated Protein Degrada t ion. Journal of Cellular Physiology 182:1 11. Kozak, K.R., B. Abbott, and O. Hankinson. 1997. ARNT Deficient Mice and Placental Differentiation. Developmental Biology 191:297 305. Larsen, J.C. 1995. Levels of Pollutants and their Metabolites: Exposures to Organic Substa n ces. Toxicology 101:11 27. Lee, D.H., and A.L. Goldberg. 1996. Selective Inhibitors of the Proteasome dependent and Vacuolar Pathways of Protein Degradation in Sac charomyces cerevi s iae The Journal of Biological Chemistry 271:27280 27284. Lee, D.H., and A.L. Goldberg. 1998. Proteasome Inhibitors Cause Induction of Heat Shock Proteins and Trehalose, which together confer Thermotolerance in Saccharomyces cerevi s iae Molecular and Cellular Biology 18:30 38. Lees, M.J., D.J. Peet, and M.L. Whitelaw. 2003. Defining the Role for XAP2 in Stabilization of the Dioxin Receptor. The Journal of Biological Chemistry 278:35878 35888. Legraverend, C., R.H. Hannah, H.J. Eiden, I.S. Owens, D.W. Nebert, and O. Hank inson 1982. Regulatory Gene Product of the Ah Locus. The Journal of Biological Chemistry 257:6402 6407. Li, H., L. Dong, and J.P. Whitlock, Jr. 1994. Transcriptional Activation Function of the Mouse Ah Receptor Nuclear Translo cator The Journal of Biological Chemistry 269:28098 28105. Liu, G., Z. Niu, D.V. Nierkerk, J. Xue, and L. Zheng. 2008. Polycyclic Aromatic Hydrocarbons (PAHs) from Coal Combustion: Emission, Analysis, and Toxicology. Reviews of Environme ntal Contamination and Toxic ology 192:1 28. Liu, Y., P. Choudhury, C.M. Cabral, and R.N. Sifers. 1999. Oligosaccharide Modification in the Early Secretory Pathway Directs the Selection of a Misfolded Glycoprotein for Degradation by the Proteasome. The Jo urnal of Biological Chem istry 274:5861 5867. Lohr, D., P. Venkov, and J. Zlatanova. 1995. Transcriptional Regulation in the Yeast GAL Gene Family: a Complex Genetic Network. FASEB 9:777 787.
135 Long, W.P., M. Pray Grant, J.C. Tsai, and G.H. Perdew. 1998. Protein Kinase C Activi t y is Required for Aryl Hydrocarbon Receptor Pathway Mediated Signal Transduction. Molecular Pharmacology 53:691 700. Longtine, M.S., A. McKenzie, D.J. Demarini, N.G. Shah, A. Wach, A. Brachat, P. Phillipsen, and J.R. Pringle. 1998 Additional Modules for Versatile and Economical PCR Based Gene Deletion and Modification in Saccharomyces cerevi s iae Yeast 14:953 961. Lorenz, M.C., R.S. Muir, E. Lim, J. McElver, S.C. Weber, and J. Heitman. 1995. Gene Disruption with PCR Products in Saccharomyces cerevisiae Gene 158:113 1 17. Luscher, B., and L.G. Larsson. 1999. The Basic Region/ Helix Loop Helix/ Leucine Zipper Domain of Myc Proto oncoproteins: Function and Regulation. Oncogene 18:2955 2966. Ma, Q., and K.T. Baldwin. 2000. 2,3,7, 8 Tetrachlorodibenzo p dioxin induced Degradation of Aryl Hydrocarbon Receptor (AhR) by the Ubiquitin Proteasome Pathway. The Journal of Biological Chemistry 275:8432 8438. Ma, Q., L. Dong, and J.P. Whitlock. 1995. Transcriptional Activation by the Mous e Ah Rece p tor. The Journal of Biological Chemistry 270:12697 12703. Ma, Q., and J.P. Whitlock. 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 di o xin. The Journal of Biological Chemistry 272:8878 8884. Mahon, M.J., and T.A. Gasiewicz. 1995. Ah Receptor Phosphorylation: Localization of Phosphorylation Sites to the C Terminal Half of the Protein. Archives of Biochemist ry and Bioph ysics 318:166 174. Massari, M.E., and C. Murre. 2000. Helix Loop Helix Proteins: Regulators of Transcription and Eucaryotic Organisms. Molecular and Cellular Biology 20:429 440. Miller, C.A., III. 1997. Expression of the Human Aryl Hydrocarbon Receptor Complex in Y e ast. The Journal of Biological Chemistry 272:32824 32829. Miller, C.A., III. 1999. A Human Aryl Hydrocarbon Receptor Signaling Pathway Constructed in Yeast Displays A dditive Responses to Ligand Mixtures. Toxicology and Applied Pharmacology 160:297 303.
136 Miller, C.A. 2002. Tw o Tetratricopeptide Repeat Proteins Facilitate Human Aryl Hydrocarbon Receptor Signaling in Y e ast. Cellular Signaling 14:615 623. Miller, C.A., III, M.A. Martinat, and L.E. Hyman 1998. Assessment of Aryl Hydrocarbon Receptor Complex Interactions using pBEVY plasmids: Expression Vectors with Bi directional Promoters for use in Saccharomyces cerevi s iae Nucleic Acids Research 26:3577 3583. Morale s, J.L., and G.H. P erdew 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 v i tro Biochemistry 46:610 621. Nguyen, T.A., D. Hoivik, J. E. Lee, and S. Safe. 1999. Interactions of Nuclear Receptor Coactivator/Corepressor Proteins with the Aryl Hydrocarbon Receptor Complex. Archives of Biochemistry and Biophysics 367:250 257. Noguerol, T. N., S. Boronat, M. Casado, D. Raldua, D. Barcelo, and B. Pina 2006. Evaluating the Interactions of Vertebrate Receptors with Persistant Pollutants and Antifouling Pesticides using Recombinant Yeast Assays. Analytical and Bioanalytical Chemistry 385:1012 1019. Ohtake, F., A. Baba, I. Takada, M. Okada, K. Iwasaki, H. Miki, S. Takahashi, A. Kouzmenko, K. Nohara, T. Chib a, Y. Fuji Kuriyama, and S. Kato. 2007. Dioxin Receptor is a Ligand Dependent E3 Uniquitin Li g ase. Nature Letters 446:562 566. Ohura, T., M. Mori ta, M. Makino, T. Amagai, and K. S himoi 2007. Aryl Hydrocarbon Receptor Mediated Effects of Chlorinated Poylcyclic Aromatic Hydrocarbons. Chemical Research in Toxicology 20:1237 1241. Pahl, H.L., and P.A. Baeuerle. 1996. Control of Gene Expression by Pr oteol y sis. Current Opinion in Cell Biology 8:340 347. Pemberton, L.F., and B.M. Paschal. 2005. Mechanisms of Receptor Mediated Nuclear Import and Nuclear Export. Traffic 6:187 198. Perdew, G.H. 1 988. Associated of the Ah Receptor with the 90 kDa Heat S hock Pr otein The Journal of Biological Chemistry 263:13802 13805. Platt, A., and R.J. Reece. 1998. The Yeast Galactose Genetic Switch is Mediated by the Formation of a Gal4p Gal80p Gal3p Complex. EMBO Journal 17:4086 4091.
137 Poland, A., and J.C. Knutson 1982. 2,3,7,8 Tetrachlorodibenzo p dioxin and Related Halogenated Aromatic Hydrocarbons: Examination of the Mechanism of Toxicity. Annual Review of Pharmocological Toxicology 22:517 554. Pollenz, R.S. 1996. The Aryl Hydrocarbon Receptor, but not the Ar yl Hydrocarbon Receptor Nuclear Translocator Protein, is Rapidl y Depleted in Hepatic and Nonhepatic Culture Cells Exposed to 2,3,7,8 Tetrachlorodibenzo p di o xin. Molecular Pharmacology 49:391 398. Pollenz, R.S. 2002. The Mechanism of AH Receptor Protein Down Regulation (Degradation ) and its Impact on AH Receptor Mediated Gene Regulation. Chemico Biological Interactions 141:41 61. Pollenz, R.S. 2007. Specific Blockage of Ligand Induced Degradation of the Ah Receptor by Proteasome but not Calpain Inhibitors in Cell Culture Lines from Different Spe c ies. Biochemical Pharmacology 74:131 143. Pollenz, R.S., and E.R. Barbour. 2000. Anal ysis of the C omplex Relationship between Nuclear Export and Aryl Hydrocarbon Receptor Mediated Gene Regula t ion. Molecular and Cellular Biology 20:6095 6104. Pollenz, R.S., and E.J. Dougherty. 2005. Redefining the Role of the Endogenous XAP2 and C termin al Hsp70 interacting Protein on the Endogenous Ah Receptors Expressed in Mouse and Rat Cell L i nes. The Journal of Biological Chemistry 280:33346 33356. Pollenz, R.S., J. Popat, and E.J. Dougherty. 2005. Role of Carboxyl terminal Transactivation Domain and Active Transcription in the Ligand Induced and Ligand Independent Degradation of the Mouse Ah b 1 Rece p tor. Biochemical Pharmacology 70:1623 1633. Pollenz, R.S., C.A. Sattler, and A. Poland. 1993 The Aryl Hydrocarbon Receptor and Aryl Hydrocarbon Receptor Nuclear Translocator Pro tein Show Distinct Subcellular Localizations in Hepa 1c1c7 Cells by Immunofluorescence Micros c opy. Molecular Pharmacology 45:428 438. Pollenz, R.S., S.E. Wilson, and E. J. Doug herty 2006. Role of Endogenous XAP2 Protein in the Localization and Nucleocytoplasmic Shuttling of the Endogenous Mouse Ahb 1 Receptor in the Presence and Absence of Ligand. Molecular Pharmacology 70:1369 1379.
138 Pongratz, I., G.G.F. Mason, and L. Poell inger 1992. Dual Roles of the 90kDa Heat Shock Protein Hsp90 in Modulating Functional Activities of the Dioxin Receptor. The Journal of Biological Chemistry 267:13728 13734. Prokipcak, R.D., and A.B. Okey. 1991. Downregulation of the Ah Receptor in Mouse Hepatoma Cells Treated in Culture w ith 2 ,3,7,8 tetrachlorodibenxo p dioxin. Can J Physiol Pharmacol 69:1204 1 2 10. Puga, A., Y. Xia, and C. Elferink. 2002. Role of Aryl Hydrocarbon Receptor in Cell Cycle Regulation. Chemico Biological Interactions 141:117 130. Ramadoss, P., J.R. Petrulis, B.D. Hollingshead, A. Kusnadi, and G.H. Perdew. 2004. Divergent Roles of Hepatitis B Virus X Associated Protein 2 (XAP2) in Human versus Mouse Ah Receptor Complexes. Biochemistry 43:700 709. Rannug, A., U. Rann ug, H.S. Rosenkranz, L. Winqvist, R. Westerholm, E. Agurell, and A.K. Grafstrom. 1987. Certain photooxidized derivatives of tryptophan bind with very high affinity to the Ah receptor and are likely to be endogenous signal substa n ces. The Journal of Biologi cal Chem istry 262:15422 15427. Rechsteiner, M., and S.W. Rogers. 1996. PEST Sequences and Regulation of Proteolysis. Trends in Biochemical Science 21:267 271. Reisz Porszasz, S., M.R. Probst, B.N. Fukunaga, and O. Hankinson. 1994. Identification of Functional Domains of the Aryl Hydrocarbon Receptor Nuclear Translocator Protein (ARNT). Molecular and Cellular Biology 14:6075 6086. Reyes, H., S. Reisz Porszasz, and O. Hankinson. 1992. Identification of the Ah Receptor Nuclear Translocator Protein (Arnt) as a Component of the DNA Binding Form of the Ah Rece p tor. Science 256:1193 1195. Rockmill, B., E.J. Lambie, and G.S. Roeder. 1991. Spore Enrichment. Methods in Enzymology 194:146 149. Roberts, B.J., and M.L. Whitelaw. 1999. Degradation of the Basic Helix Loop Helix/Per ARNT Sim Homology Domain Dioxin Rec e ptor Via the Ubiquitin/Proteasome Pathway. The Journal of Biological Chemistry 274:36351 36356. Robinson, K.A., and J.M. Lopes. 2000. Saccharomyces cerevisiae Basic Helix Loop Helix Proteins Regulate Diverse Biological Proce s ses. Nucleic Acids Research 28:1499 1505.
139 Schmidt, K.H., and R.D. Kolodner. 2004. Requirement of Rrm3 Helicase for Repair of Spontaneous DNA Lesions in Cells Lacking Srs2 or Sgs1 Helicase. Molec ular and Cellular Bi ology 24:3213 3226. Sherman, F. 2 002. Getting Started with Yeast. Methods in Enzymology 350:3 41. Sikorski, R .S., and P. H ieter 1989. A System of Shuttle Vectors and Yeast Host Strains Designed for Efficient Manipulation of DNA in Saccharomyces cerevi s iae Genetics 122:19 27. Song, Z., and R.S. Pollenz. 2002. Ligand Dependent and I ndependent Modulation of Aryl Hydrocarbon Receptor Localization, Degradation, and Gene Regula t ion. Molecular Pharmacology 62:806 816. Song, Z., and R.S. Po llenz 2003. Functional Analysis of Murine Aryl Hydrocarbon (AH) Receptors Defective in Nuclear Import: Impact on AH Receptor Degradation and Gene Regulation. Molecular Pharmacology 63:597 606. Sorimachi, H., S. Ishiura, and K. Suzuki. 1997. Struc ture and Physiological Function of Calp a ins. Journal of Biochemistry 328:721 732. Soshilov, A., and M.S. Denison. 2008. Role of the PER/ARNT/SIM Domains in Ligand Dependent Transformation of the Aryl Hydrocarbon Receptor. The Journal of Biological Chem istry 283:32995 33005. Sreedhar, A.S., E. Kalmar, P. Csermely, and Y. F. Shen. 2004. Hsp90 Isoforms: Functions, Expression, and Clinical Importance. FEBS Letters 562:11 15. Sugihara, K., T. Okayama, S. Kitamura, K. Yamashita, M. Yasuda, S. Miyairi, Y. Minobe, an d S. Ohta. 2008. Comparative Study of Aryl Hydrocarbon Receptor Ligand Activities of Six Chem i cals In Vitro and In Vivo Toxicokinetics and Metabolism 82:5 11. Swanson, H.I., W.K. Chan, and C.A. Bradfield. 1995. DNA Binding Specificities a nd Pairing Rules of the Ah Receptor, ARNT, and SIM Prot e ins. The Journal of Biological Chemistry 270:26292 26302. Swanson, H.I., and G.H. Perdew. 1993. Half Life of Aryl Hydrocarbon Receptor in Hepa 1 Cells: Evidence for Ligand Dependent Alterations in C ytosolic Receptor Le v els. Archives of Biochemistry and Biophysics 302:167 174. Tanaka, K. 1998. Proteasomes: Structure and Bio l ogy. Journal of Biochemistry 123:195 204.
140 Varshavsky, A. 1997. The N end Rule Pathway of Protein Degradation. Genes to Cells 2:13 28. Whitelaw, M., I. Pongratz, A. Wilhelmsson, J. A. Gustafsson, and L. Poellinger. 1993. Ligand Dependent Rectruitment of the Arnt Coregulator Determines DNA Recognition by the Dioxin Receptor. Molecular and Cellular Biology 13:2504 2514. Whitela w, M.L., J. McGuire, D. Picard, J. A Gustafsson, and L. Poellinger. 1995. Heat Shock Protein Hsp90 Regulates Dioxin Receptor Function In Vivo Proc. Natl. Acad. Sci. 92:4437 4441. Whitlock, J.P., Jr. 1 999. Induction of Cytochrome P4501A1. Annual Review o f Pharmocological Toxic ology 39:103 125. Wormke, M., M. Stoner, B. Saville, K. Walker, M. Abdelrahim, R. Burghardt, and S. Safe. 2003. The Aryl Hydrocarbon Receptor Mediates Degradation of Estrogen Receptor alpha through Activation of Proteasomes. Molecu lar and Cellular Bi ology 23:1843 1855. Wright, A.P., M. Bruns, and B.S. Hertley. 1989. Extraction and rapid inactivation of proteins from Saccharomyces cerevisiae by trichloroacetic acid precipita t ion. Yeast 5:51 53.
142 Appendix A : Table o f Strains Generated TABLE A 1: List o f Strains Generated Strain Genotype Use in Experiments KHSY421 MATa, ura3 52, trp1 63, his3 200, leu2 1 Parental strain for AHR expression KHSY422 MAT ura3 52, trp1 63, his3 200, leu2 1 Parental strain for ARNT expression KHSY1538 MATa, ChrXV::HIS3MX6 PGAL1 AHR.TRP1, ura3 52, trp1 63, his3 200, leu2 1 AHR expressor strain KHSY1541 MAT ChrIV::HIS3MX6 PGAL1 V5 ARNT.URA3, ura3 52, trp1 63, his3 200, leu2 1 ARNT expressor strain KHSY1547 MATa, ChrXV::HIS3MX6 PGAL1 AHR.TRP1, ChrIV::HIS3MX6 PGAL1 V5 ARNT.URA3, ura3 52, trp1 63, his3 200, leu2 1 Double knock in strain expressing both AHR and ARNT KHSY1565 MATa, ChrXV::HIS3MX6 PGAL1 AHR.TRP1, ura3 52, trp1 63, his3 200, leu2 1, pLXRE5 Z AHR expressor strain with AHR activated reporter plasmid KHSY1566 MATa, ChrXV::HIS3MX6 PGAL1 AHR.TRP1 ChrIV::HIS3MX6 PGAL1 V5 ARNT.URA3, ura3 52, trp1 63, his3 200, leu2 1, pLXRE5 Z Double knock in strain with AHR activated reporter plasmid KHSY2228 MATa, ChrXV::HIS3MX6 PGAL1 AHR.TRP1, ChrIV::HIS3MX6 PGAL1 V5 ARNT.URA3, ura3 52, trp1 63, his3 200, leu2 1, pHGPD Double knock in strain with empty vector used in HSP90 studies KHSY2229 MATa, ChrXV::HIS3MX6 PGAL1 AHR.TRP1, ChrIV::HIS3MX6 PGAL1 V5 ARNT.URA3, ura3 52, trp1 63, his3 200, leu2 1, pHGPDHsp90alpha Double knock in strain over expressing hum an HSP90 alpha from a transformed plasmid KHSY2230 MATa, ChrXV::HIS3MX6 PGAL1 AHR.TRP1, ChrIV::HIS3MX6 PGAL1 V5 ARNT.URA3, ura3 52, trp1 63, his3 200, leu2 1, prHGPDHsp90beta Double knock in strain over expressing human HSP90 beta from a transformed pla smid KHSY2501 MATa, ChrXV::HIS3MX6 PGAL1 AHR.TRP1, ChrIV::HIS3MX6 PGAL1 V5 ARNT.URA3, ura3 52, trp1 63, his3 200, leu2 1, erg6 Permeable double knock in strain for protease studies KHSY2502 MATa, ChrXV::HIS3MX6 PGAL1 AHR.TRP1, ChrIV::HIS3MX6 PGAL1 V5 ARNT.URA3, ura3 52, trp1 63, his3 200, leu2 1, hul5 Double knock in strain with HUL5 E3 ligase knocked out KHSY2502 MATa, ChrXV::HIS3MX6 PGAL1 AHR.TRP1, ChrIV::HIS3MX6 PGAL1 V5 ARNT.URA3, ura3 52, trp1 63, his3 200, leu2 1, ubr1 Do uble knock in strain with UBR1 E3 ligase knocked out
143 Appendix B: Additional Figures FIGURE A .1 : Reduction of Ahr protein levels in yeast upon addition of glucose to the growth medium. The KHSY1538 strain was induced for Ahr protein expression with the addition of 2% galactose to the growth medium overnight (A), for 2 hours (B), or for 1 hour (C). Subsequently, the galactose containing media was removed and replaced with glucose containing media. The glucose acted to shut down the GAL promoter located upstream of the AHR cDNA. Upon addition of glucose to the culture medium, the Ah receptor is degraded over time regardless of the length of the initial induction of the protein expression. FIGURE A 2 : Levels of Arnt protein expressed in the KHSY1541 strain are reduced following addition of glucose to the growth medium. A two hour induction in 2% galactose followed by a change to growth media containing 2% glucose caused the GAL1 promoter to stop transcription of the ARNT mRNA. Protein samples prepared from samples of the culture taken over time show a rapid degradation of the Arnt protein that is comparabl e to the Ahr degradation pattern previously described.
144 Appendix C: E3 Knockout Preliminary Results FIGURE A 3 : Ahr protein is degraded in the !ubr1 strain. The Ahr and Arnt expressor strain was used to knock out the yeast E3 ligase, UBR1. This strain was tested for Ahr stability as previously described. Ahr and Arnt protein expression was induced in the double knock in strain and the knock out str ain with 2% galactose for two hours. The GAL1 promoter was turned off and protein samples were obtained 0, 1, 2, and 3 hours after the promoter was turned off. Note that Ahr is degraded over time in both the double knock in strain and the E3 mutant strai n over time. The rate of degradation in both strains appears comparable suggesting that UBR1 is not likely the intermediate E3 ligase that targets Ahr for proteasomal degradation. TABLE A 2: Yeast Genes Involved In Proteasome Mediated Degradation Gene ID Name & APC1 Largest subunit of the Anaphase Promoting Complex Cyclosome (APC/C), which is a ubiquitin protein ligase (E3) I APG7, CVT2, ATG7 Autophagy related protein and dual specificity member of the E1 family of ubiquitin activating enzymes V CDC16 metal binding nucleic acid binding protein, interacts with Cdc23p and Cdc27p to catalyze the conjugation of ubiquitin to cyclin B (putative) I CDC4 F box protein which acts as ubiquitin protein ligase (E3) I DSK2 Nuclear enriched ubiquitin like polyubiquitin binding protein V HSE1 Subunit of the endosomal Vps27p Hse1p complex required for sorting of ubiquitinated membrane proteins into intralumenal V
145 vesicles prior to vacuolar degradation HUB1 Ubiquitin like protein modifier V HRT1 HRT2 RBX1, ROC1 Skp1 Cullin F box ubiquitin protein ligase (SCF) subunit I HUL4 Protein with similarity to hect domain E3 ubiquitin protein ligases (E3) V HUL5 Protein with similarity to hect domain E3 ubiquitin protein ligases (E3) V LEO1 Component of the Paf1 complex V PAF1 RNA polymerase II associated protein V RAD23 Protein with ubiquitin like N terminus V RUB1 Ubiquitin like protein with similarity to mammalian NEDD8 V STP22 VPS23 Component of the ESCRT I complex, which is involved in ubiquitin dependent so rting of proteins into the endosome V TUL1 Golgi localized RING finger ubiquitin ligase (E3) V UBA1 Ubiquitin activating enzyme (E1), involved in ubiquitin mediated protein degradation I UBC1 Ubiquitin conjugating enzyme (E2) that mediates selective degradation of short lived and abnormal proteins I UBC2, RAD6 Ubiquitin conjugating enzyme (E2) V UBC3, CDC34 Ubiquitin conjugating enzyme or E2 (cell cycle) woks with SKP1, RBX1, CDC53 V UBC4 Ubiquitin conjugating enzyme (E2) that mediates degradation of short lived and abnormal proteins V UBC5 Ubiquitin conjugating enzyme (E2) that mediates selective degradation of short lived and abnormal proteins V UBC6, Ubiquitin conjugating enzyme (E2) involved in ER associated V
146 DOA2 protein degradation UBC7, QRI8 Ubiquitin conjugating enzyme (E2), involved in the ER associated protein degradation pathway V UBC8 Ubiquitin conjugating enzyme (E2) that negatively regulates gluconeogenesis V UBC9 SUMO conjugating enzyme involved in the Smt3p conjugation pathway I UBC10, PAS2, PEX4 Peroxisomal ubiquitin conjugating enzyme (E2) V UBC11 Ubiquitin conjugating enzyme (E2) most similar in sequence to Xenopus ubiquitin conjugating enzyme E2 C V UBC12 Enzyme that mediates the conjugation of Rub1p, a ubiquitin like protein, to other proteins; related to E2 ubiquitin conjugating enzymes V UBC13 Ubiquitin conjugating enzyme (E2) involved in the error free DNA postreplication repair pathway V UBI1, RPL40A Fusion protein, identical to Rpl40B p, that is cleaved to yield ubiquitin V UBI2, RPL40B Fusion protein, identical to Rpl40Ap, that is cleaved to yield ubiquitin V UBI3, RPS37 Fusion protein that is cleaved to yield a ribosomal protein of the small (40S) subunit and ubiquitin V UBI4 SCD2 Ubiquitin gene V UBP12 Ubiquitin specific protease present in the nucleus and cytoplasm that cleaves ubiquitin from ubiquitinated proteins V UBP1 Ubiquitin specific protease that removes ubiquitin from ubiquitinated proteins V UBP2 Ubiquitin specific protease that removes ubiquitin from ubiquitinated proteins V UBP3 Ubiquitin specific protease that int eracts with Bre5p V
147 UBP4, DOA4 DOS1, MUT4 NPI2, SSV7, Ubiquitin hydrolase, required for recycling ubiquitin from proteasome bound ubiquitinated intermediates V UBP5 Putative ubiquitin specific protease that does not associate with the proteasome V UBP6 Ubiquitin specific protease situated in the base subcomplex of the 26S proteasome V UBP7 Ubiquitin specific protease that cleaves ubiquitin protein fusions V UBP8 Ubiquitin specific protease that is a component of the SAGA (Spt Ada Gcn5 Acetyltransferase) acetylation complex V UBP9 Ubiquitin specific protease that cleaves ubiquitin protein fusio ns V UBP10, DOT4 ubiquitin specific protease that deubiquitinates ubiquitin protein moieties V UBP11 Ubiquitin specific protease that cleaves ubiquitin from ubiquitinated proteins V UBP12 Ubiquitin specific protease present in the nucleus and cytoplasm that cleaves ubiquitin from ubiquitinated proteins V UBP13 Putative ubiquitin specific protease V UBP14 Ubiquitin specific protease that specifically disassembles unanchored ubiquitin chains V UBP15 ubiquitin specific protease that may play a role in ubiquitin precursor processing V UBP16 Putative ubiquitin specific protease V UBR1, PTR1 Ubiquitin protein ligase (E3) that interacts with Rad6p/Ubc2p to ubiquitinate substrates of the N end rule pathway V UBR2, Cytoplasmic ubiquitin protein ligase (E3) V UFD1 Protein that interacts with Cdc48p and Npl4p, involved in recognition of polyubiquitinated proteins and their presentation to the 26S proteasome for degradation I
148 UFD2 Ubiquitin chain assembly factor (E4) also functions as an E3 V UFD3, DOA1, ZZZ4 WD repeat protein required for ubiquitin mediated protein degradation V UFD4 Ubiquitin protein ligase (E3) that interacts with Rpt4p and Rpt6p V UFD5, RPN4, SON1 Transcription factor that stimulates expression of proteasome genes; Rpn4p levels are in turn regulated by the 26S proteasome in a negative feedback control mechanism V URM1 Ubiquitin like protein with only weak sequence similarity to ubiquitin; depends on the E1 like activating enzyme Uba4p V YUH1 Ubiquitin C terminal hydrolase that cleaves ubiquitin protein fusions to generate monomeric ubiquitin V