Transcriptional regulation by Brn-3 POU domain-containing transcription factors

Jonathan Hancock Dennis

A thesis submitted for the degree of Doctor of Philosophy

Medical Molecular Biology Unit Department of Molecular Pathology University College London London

January 2001 ProQuest Number: U644107

All rights reserved

INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.

In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. uest.

ProQuest U644107

Published by ProQuest LLC(2016). Copyright of the Dissertation is held by the Author.

All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code. Microform Edition © ProQuest LLC.

ProQuest LLC 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106-1346 ABSTRACT

Bm-3a and Bm-3b are closely related members of the POU domain-containing transcription factors. While Bm-3a is associated with neuronal differentiation and Bm-

3b with neuronal proliferation, their expression is not strictly limited to the nervous system. It has been shown that these Bm-3 are expressed in non-neuronal tissues including cervical epithelium, testis and breast. Moreover, these Bm-3 proteins functionally interact with the estrogen and in association with the have differing transactivation potentials.

The p i60 steroid receptor coactivators (Srcs) are also able to interact with the estrogen receptor in a ligand dependent manner and are able to enhance estrogen dependent transcription. This work describes a functional interaction between Bm-3 proteins and members of the steroid receptor coactivator family. In affinity chromatography and coimminoprecipitation experiments, Src-1 proteins were shown to physically interact with Bm-3a and Bm-3b. In addition, the transactivation potential of the Bm-3/Src complexes was tested on two promoters in three different cell lines. The steroid receptor coactivators potentiated the transcriptional effects of Bm-3 a and Bm-3b. The Src proteins, however, differed in their ability to potentiate the transcriptional activity of Bm-

3 proteins on different promoters in transiently transfected cells. In addition, maximal activation of the Bm-3/Src complex differed between cell lines, indicating that additional cell-type specific proteins are important for maximal activation by the complex. It has recently been shown that Bm-3b may play a role in regulating BRCA-1 in mammary tumors as its expression is enhanced in primary breast tumors with reduced

BRCA-1 expression. Moreover, this elevated Bm-3b expression is not seen in normal mammary cells, benign tumors, or malignant tumors which do not have reduced levels of

BRCA-1. This work also describes the effects of stable Bm-3b overexpression and reduction in the estrogen receptor positive breast adenocarcinoma cell line MCF7.

Overexpression of Bm-3b resulted in increased growth rate, saturation density, proliferation, and ability to form anchorage independent colonies when compared to mock transfected cells. MCF7 cells with reduced Bm-3b levels exhibited a significant decrease in growth rate, saturation density, proliferation, and ability to form anchorage independent colonies when compared to mock transfected controls.

Differential expression in these cell lines was examined using both specific western immunodetection as well as comprehensive DNA array technology. Immunodetection revealed increases in ER, HSP-27, and Bm-3 a in Bm-3b overexpressing cells as well as a decrease in these proteins in the Bm-3b reduced cell lines. The DNA array including

1200 cancer related revealed several genes differentially expressed when the Bm-

3b overexpressing and reduced cell line mRNAs were comj^ared. Finally, the results of three of the genes shown to be differentially regulated in the cDNA array differential display were corroborated by semi-quantitative RT-PCR. Dedicated to the memory of

Paul B. Sigler (1934-2000) ACKNOWLEDGMENTS

I would first like to thank Professor David S. Latchman for the opportunity to come to London and work in his laboratory. David’s broad base of knowledge, insight, practicality, efficiency, patience, optimism, and support have been invaluable over the past few years, and will continue to be characteristics I will work towards as I pursue my own career. I would also like to thank Dr. Vishwanie Budhram-Mahadeo for her guidance during this Ph.D. The work in this thesis stems from work she completed showing the relevance of Bm-3 proteins in estrogen responsiveness. In addition, I must extend my thanks to the members of the Bm-3 group, both past and present, who have always been generous with their time and expertise. In particular, I would like to thank Dr. Sally Dawson who advised and supervised me when I initially joined the laboratory. Several additional unmentioned individuals in the Windeyer Institute and The Institute of Child Health have helped with their assistance and friendship over the past four years.. .thank you.

I would like to thank my family for their constant support and encouragement throughout this Ph.D. and the circuitous path that led to it. Finally, I would like to thank Marie Charrel for her daily encouragement, support, patience, and love that helped me keep everything in perspective.

This work was supported by the Countess of Lisbume Medical Research Fellowship from the University College London Medical School, and an Overseas Research Studentship. LIST OF CONTENTS I General Introduction ...... 27 LI Transcription...... 27 1.1.1 Cellular communication results in the regulation of transcription ...... 27 1.1.2 Every cell contains the same set of genes ...... 28 1.1.3 There are three types of RNA polymerase ...... 28 1.1.4 Promoters, Enhancers, and UPEs control transcription...... 29 1.1.4.A Promoters ...... 29 1.1.4.B Upstream promoter elements ...... 29 1.1.4.C Enhancers...... 29 1.1.4.D The beta-globin gene ...... 30 1.1.5 Gene specific promoter elements ...... 30 1.1.6 The five stages of transcription...... 31 1.1.6. A The formation of the pre-initiation complex...... 31 1.1.6.A.a Stepwise assembly...... 31 l.l.ô.A.b The holo-RNA polymerase 11 complex...., ...... 32 1.1.6.B The initiation of transcription and promoter clearance ...... 32 1.1.6.C Transcript elongation and termination ...... 32 1.1.7 Gene specific transcription factors ...... 33 1.1.7.A Gene-specific transcription factors have a modular structure ...... 33 1.1.8 Transcription factors have several different types of DNA-binding domain.... 34 1.1.8.A.a Helix-tum-helix ...... 34 1.1.8.A.b Zinc-finger ...... 35 1.1.8.A.C and basic DNA-binding domain ...... 36 1.1.8.A.d Helix-loop-helix domain ...... 37 1.1.8.A.e Other DNA-binding motifs ...... 37 1.1.9 Activation and inhibition of gene transcription by transcription factors ...... 38 1.1.10 Chromatin remodeling by transcription factors ...... 38 1.1.11 Combinatorial regulation of transcription ...... 39 1.2 The superfamily...... 42 1.2.1 History...... 42 1.2.2 The general structural and functional domains of nuclear hormone receptors. 42 1.2.2.A Structural domains ...... 42 1.2.2.B Functional domains ...... 43 1.2.3 The members of the nuclear receptor superfamily...... 45 1.2.4 Class I, the steroid hormone receptors ...... 46 1.2.4.A Description...... 46 1.2.4.B The unliganded steroid nuclear receptor ...... 46 1.2.4.C The ligand-bound steroid nuclear receptor ...... 47 1.2.4.D The Estrogen receptor ...... 47 1.2.4.D.a History...... 47 1.2.4.D.b The role of estrogen receptors in cancer ...... 48 1.2.4.D.C Isoforms of the ER ...... 49 1.2.4.D.d Structural domains of the estrogen receptor ...... 50 1.2.4.D.d.i The A/B region ...... 50 1.2.4.D.d.ii The C region ...... 51 1.2.4.D.d.iii The E region ...... 51 1.2.4.D.d.iv The D region ...... 52 1.2.4.D.e ER interaction with response elements ...... 52 1.2.4.D.f General mechanism of ER activation ...... 53 1.2.4.D.g Different regions of the ER have different transactivation potential. 53 1.2.4.D.h The estrogen receptor must interact with other proteins ...... 54 1.2.4.D.h.i Dimerization interactions ...... 55 1.2.4.D.h.ii Estrogen receptors utilize transcriptional cross-talk ...... 55 1.2.4.D.h.iii Interactions of the ER with basal transcriptional machinery..... 56 1.2.4.D.h.iv Interactions of the ER with transcriptionalintermediary factors (TIPs)...... 56 • Coactivators ...... 56 • Corepressors ...... 57 • Integrators ...... 58 1.2.4.D.h.v Chromatin remodelling proteins ...... 58 1.3 The pi 60/SRC co-activators ...... 60 1.3.1 General structural and functional domains ...... 61 1.3.1.A bHLH/PAS...... 61 1.3.1.B Serine/threonine and glutamine rich regions ...... 61 1.3.1.C The activation domains ...... 61 1.3.1.D The steroid receptor binding domains ...... 62 1.3.1.D.a The LXXLL motif...... 62 1.3.1.D.b SRBl and SRB2...... 63 1.3.2 There are three SRC genes ...... 65 1.3.2.A SRC-1...... 65 1.3.2. A.a Isoforms of the SRC-1 proteins ...... 65 1.3.2.A.b Mechanism of action...... 68 1.3.2.A.C Interaction with ERp ...... 68 1.3.2.A.d Interactions with other proteins ...... 68 1.3.2.A.d.i Interaction with CBP ...... 69 1.3.2.A.d.ii Interaction with gene specific transcription factors ...... 69 1.3.2.A.d.iii Interactions with basal transcription factors ...... 69 1.3.2.A.d.iv Interactions with cell cycle proteins ...... 70 1.3.2.A.e SRC-1 knockout mice ...... 70 1.3.2.A.f Role in cancer ...... 70 1.3.2.B SRC-2...... 70 1.3.2.B.a Mechanism of action ...... 71 1.3.2.B.b Interactions with ER|3 ...... 71 1.3.2.B.C Interactions with other proteins ...... 72 1.3.2.C SRC-3...... 72 1.3.2.C.a Mechanism of action ...... 72 1.3.2.C.b Interaction with ERP ...... 73 1.3.2.C.C Interactions with other proteins ...... 73 1.3.2.C.d Role in cancers ...... 73 1.4 The POU family of transcription factors ...... 74 1.4.1 History and description...... 74 1.4.2 The POU domain ...... 74

8 1.4.2.A The POU-specifîc domain ...... 74 L4.2.B The POU-homeodomaia ...... 75 1.4.2.C The linker region ...... 75 1.4.3 Family members ...... 77 1.4.4 DNA response elements of the POU domains ...... 80 1.4.5 POU interactions with DNA response elements ...... 80 1.4.6 Arrangement on response elements ...... 83 1.4.7 Regulation of POU domain genes ...... 85 1.4.7.A Nuclear translocation ...... 85 1.4.7.B Phosphorylation ...... 85 1.4.8 Expression of POU domain proteins...... 86 1.4.9 /protein interactions ...... 86 1.4.9. A Homo- and hetero- dimerization of POU domains ...... 87 1.4.9.B Interactions of POU family members with other transcription factors 88 1.4.9.C Position 22 in the first alpha-helix of the POU homeodomain ...... 89 1.4.10 The Bm-3 family of POU domain containing proteins ...... 92 1.4.10.A History...... 92 1.4.10.B Three different Bm-3 genes: Bm-3 a, Bm-3b and Bm-3c ...... 93 1.4.10.C Altemative splicing of Bm-3 a and Bm-3b ...... 93 1.4.10.DThe POU IV box ...... 94 1.4.10.E Chromosomal localization of the Bm-3 proteins ...... 96 1.4.10.F Expression of the Bm-3 proteins ...... 96 1.4.10.F.a Expression of Bm-3 in neuronal tissue ...... 96 1.4.10.F.b Expression of Bm-3 in neuronal cancers ...... 97 I.4.10.F.b.i Expression in neuronal tissue...... 97 1.4.10.F.C Expression of Bm-3 in non-neuronal tissue and cancers ...... 97 1.4.10.G Bm-3 knockout mouse models ...... 98 1.4.10.G.a The Bm-3a knockout ...... 98 1.4.10.G.b The Bm-3b knockout ...... 98 1.4.10.G.C The Bm-3c knockout ...... 99 1.4.10.H DNA binding by Class IV Bm-3 Proteins ...... 99 I.4.10.I Promoters regulated by Bm-3 proteins ...... 99 I.4.10.J Bm-3 POU domain proteins play a role in estrogen dependent transcription ...... 102 I.5 Aims of the project ...... 104 II MATERIALS AND METHODS...... 105 II. 1 Materials...... 105 II. 1.1 Suppliers ...... 105 II. 1.2 Standard buffers and solutions ...... 107 II. 1.3 Description and sources of E. coli strains...... 108 II. 1.4 Description and sources of plasmids ...... 109 II. 1.5 Description and sources of cell lines ...... 113 II. 1.6 Description and sources of antibodies used ...... 114 II. 1.7 Description of oligonucleotide primers used ...... 117 II.2 Methods ...... 121 11.2.1 Bacterial Methods ...... 121 II.2.1.A Propagation of bacteria ...... 121 11.2.1.B Long term storage of bacteria ...... 121 11.2.1.C Transformation of bacteria ...... 121 11.2.2 DNA methods ...... 123 11.2.2.A Small scale plasmid DNA extraction from transformed bacteria ...... 123 11.2.2.B Large scale plasmid DNA extraction from transformed bacteria ...... 123 11.2.2.C Restriction enzyme digestion...... 124 11.2.2.D Phenol/chloroform extraction and precipitation of DNA ...... 124 11.2.2.E Agarose gel electrophoresis ...... 124 11.2.2.F Southem blot analysis ...... 125 11.2.2.F.a Resolution of DNA on an agarose gel ...... 125 11.2.2.F.b Transfer of DNA to nitrocellulose ...... 125 11.2.2.F.C Radiolabelling of DNA ...... 126 11.2.2.F.d Hybridisation...... 127 11.2.3 RNA Methods...... 128 11.2.3.A Isolation of total cellular RNA ...... 128

10 IL2.3.B Solubilization of RNA in formamide ...... 128 IL2.3.C cDNA synthesis ...... 129 IL2.3.D PCR...... 129 11.2.4 Protein Methods ...... 131 11.2.4.A SDS-polyacrylamide gel electrophoresis (SDS-PAGE) ...... 131 11.2.4.B Equalisation of protein loading ...... 131 IL2.4.B.a Coomassie Blue Staining...... 131 II.2.4.B.b Bradford Chemistry ...... 131 11.2.4.C Transfer of proteins to nitrocellulose membranes (western blotting).... 132 11.2.4.D Immunodetection of proteins on western blots ...... 132 11.2.4.E Expression of labeled proteins by in vitro translation ...... 133 11.2.4.F Preparation of GST fusion protein ...... 134 11.2.4.G Affinity purification of GST fusion proteins using Glutathione Sepharose 4B...... 134 11.2.4.H Glutathione-S-transferase affinity pull-down chromatography. 135 11.2.4.1 Coimmunoprecipitation of protein complexes from tissue culture samples ...... 135 11.2.5 Cell culture...... 137 11.2.5.A Culture conditions ...... 137 11.2.5.B Freezing and recovery of cell stocks ...... 137 11.2.5.C Routine cell passage ...... 137 11.2.5.D Transfection of plasmids for transient expression ...... 137 11.2.5.E CAT assay ...... 138 11.2.5.F Construction of stable cell lines ...... 139 11.2.5.G Cellular growth curves and saturation density limitation studies 140 11.2.5.H Saturation density limitation studies ...... 140 11.2.5.1 Tritiated thymidine incorporation proliferation studies ...... 140 11.2.5.J Anchorage independent growth studies ...... 141 11.2.5.K Standard protein extraction fi*om cultured cells ...... 141 III BRN-3 POU TRANSCRIPTION FACTORS AND pl60 STEROID RECEPTOR COACTIVATORS INTERACT AND COREGULATE TRANSCRIPTION FROM VARIOUS PROMOTERS IN DIFFERENT CELL TYPES...... 142

11 111.1 Introduction ...... 142 111.2 Results...... •...... 145 111.2.1 Src-1 specifically binds to Bm-3 proteins ...... 145 111.2.2 Bm-3a long and Bm-3 a short coimmunoprecipitate with Src-1 ...... 149 111.2.3 Bm-3/pl60 cotransfections elicit varied responses on different promoters in different cell lines...... 151 111.2.3.A Transient cotransfections of the Bm-3 proteins with the pl60 proteins ...... 151 111.2.3.A.a The promoters ...... 151 111.2.3.A.a.i The SNAP-25 promoter ...... 151 111.2.3.A.a.ii The Vitellogenin promoter ...... 151 111.2.3.A.b The cell lines...... 152 111.2.3.A.b.i ND7 cells ...... 152 111.2.3.A.b.ii BHK21 cells...... 152 IIL2.3.A.b.iii MCF7 cells ...... 152 111.2.3.B Experiments on the SNAP-25 promoter ...... 153 IIL2.3.B.a Experiments in ND7 cells ...... 153 111.2.3.B.b Bm-3 a long potentiates transactivation of the SNAP-25 promoter by Src-la in ND7 cells ...... 153 111.2.3.B.b.i Bm-3 a long potentiates transactivation of the SNAP-25 promoter by Src-le in ND7 cells ...... 153 111.2.3.B.C Experiments in BHK21 cells ...... 156 111.2.3.B.c.i Bm-3a long and Bm-3b short potentiate transactivation of the SNAP-25 promoter by Src-la in BHK21 cells...... 156 111.2.3.B.c.ii Bm-3 a long potentiates transactivation of the SNAP-25 promoter by Src-le in BHK21 cells ...... 156 IIL2.3.C Experiments on the Vit-tk promoter ...... 159 111.2.3.C.a Experiments with Src-la ...... 159 IIL2.3.C.a.i Bm-3 a long weakly activates while Bm-3b short weakly represses the activity of Src-la on the Vit-tk promoter in ND7 cells 159

12 111.2.3.C.a.ii Bm-3 a long and Bm-3b short reverse the weakrepression of the Vit-tk promoter by Src-la in BHK21 cells ...... 161 111.2.3.C.a.iii Neither Bm-3 a long, Bm-3b short, nor Src-la have an effect on the Vit-tk promoter in the absence of estrogen in MCF7 cells ...... 161 111.2.3.C.b Experiments with Src-le ...... 164 IIL2.3.C.b.i Bm-3a long responds to Src-le in a dose dependant manner, while Bm-3b short consistently represses the activity of Src-le on the Vit-tk promoter in ND7 cells ...... 164 111.2.3.C.b.ii Bm-3 a long and Bm-3b short consistently weakly potentiate repression of the Vit-tk promoter by Src-le in BHK21 cells ...... 164 111.2.3.C.b.iii Bm-3a long and Bm-3b short potentiate the repression of the Vit-tk promoter by Src-le in estrogen deprived MCF7 cells ...... 167 111.2.3.C.C Experiments with Src-3 ...... 169 IIL2.3.C.c.i Bm-3b short weakly represses the activity of AIBl on the Vit-tk promoter in ND7 cells ...... 169 111.2.3.C.C.Ü Bm-3 a long and Bm-3b short consistently weakly potentiate repression of the Vit-tk promoter by AIBl in BHK21 cells ...... 169 IIL2.3.C.c.iii Bm-3a long and Bm-3b short potentiate the repression of the Vit-tk promoter by AIBl in estrogen deprived MCF7 cells ...... 169 III.3 Discussion ...... 173 111.3.1 Bm-3 proteins physically interact with Src-1 proteins ...... 173 111.3.2 p i60 proteins coactivate transcription with Bm-3 transcription factors on a Bm-3 responsive promoter ...... 174 111.3.3 p i60 steroid receptor coactivators can activate steroid receptor independent transcription ...... 175 111.3.4 p i60 steroid receptor coactivator promoter regulation ...... 176 111.3.4.A The p i60 coactivators on the Vit-tk promoter construct ...... 177 111.3.4.A.a Src-la...... 177 111.3.4.A.b Src-le ...... 177 111.3.4.A.C Src-3...... 178 111.3.4.B The cell lines...... 178

13 III.3AB.aND7 cells ...... 179 111.3.4.B.b BHK21 cells...... 179 111.3.4.B.C MCF7 cells ...... 180 111.3.5 3.3.5 p i60 proteins may act as transcriptional integrators ...... 180 111.3.6 Further experiments to demonstrate the involvement of Src proteins as coregulatory proteins for the Bm-3 transcription factors ...... 181 111.3.7 Conclusion ...... 182 IV CELLULAR GROWTH AND ENDOCRINE CHARACTERISTICS OF MCF7 CELLS WITH ALTERED LEVELS OF BRN-3B...... 183 IV. 1 Introduction ...... 183 IV. 1.1 p i60 proteins enhance estrogen stimulated MCF7 proliferation ...... 183 IV. 1.2 The ER enhances estradiol stimulated MCF7 proliferation ...... 183 IV. 1.3 Bm-3b is associated with the proliferation of neuronal cells ...... 184 rV.1.4 Bm-3b may play a role in the regulation of BRCA I in breast cancer ...... 184 IV. 1.5 Altering specific protein levels in cell lines ...... 185 IV. 1.5.A Experimental strategy for increasing the levels of Bm-3b short in the MCF7 cell line ...... 186 IV.1.5.B Experimental strategy for decreasing the levels of Bm-3b in the MCF7 cell line...... 186 IV.2 Results...... 188 IV.2.1 Establishment of stable transformants ...... 188 IV.2.2 Confirmation of stable transformants by RT-PCR ...... 188 IV.2.3 Confirmation of stable transformants by Western immunoblotting ...... 191 IV.2.4 Effects on cell growth in monolayers ...... 194 IV.2.4.A Growth curve study...... 194 IV.2.4.B Saturation density limitation of growth study ...... 197 IV.2.4.C Tritiated thymidine incorporation study...... 199 IV.2.4.D Effects on cellular morphology ...... 199 IV.2.4.E Effects on growth in soft agar ...... 206 IV.2.5 Endocrine characteristics of the MCF stable clones ...... 212 IV.2.5. A Effects of estrogen and estrogen analogues on MCF7 cells ...... 212 IV.2.5.B Effects on growth ...... 212

14 IV.2.5.c Effects on cellular proliferation ...... 216 IV.2.5.D Effects on anchorage-independent growth...... 221 IV.3 Discussion ...... 225 V DIFFERENTIAL ANALYSIS OF MCF7 CELLS WITH ALTERED LEVELS OF BRN-3B...... 227 V. 1 Introduction ...... 227 V.1.1 Specific Western immunoblotting ...... 227 V.l.l.A Bm-3 proteins ...... 228 V.l.l.B Hormone receptors and hormones ...... 229 V.l.l.B.a The estrogen receptor ...... 229 V.l.l.B.b Human chorionic gonadotropin ...... 229 V.l.l.C Heat shock proteins ...... 230 V.l.l.C.aHsp90...... 230

V.l.l.C.b Hsp70...... 230 V.l.l.C.cHsp27 ...... 231 V.l.l.D Tumor suppressor gene ...... 231 V.l.l.D.a BRCA 1...... 231 V. 1.2 Differential display on a cDNA expression array ...... 232 V. 1.3 Corroboration of cDNA expression array results by RT-PCR ...... 232 V.2 Results...... 234 V.2.1 Specific Western immunoblotting ...... 234 V.2.1.A Bm-3 proteins ...... 234 V.2.1 .A.a Bm-3b long ...... 234 V.2.1.A.b Bm-3 a long ...... 234 V.2.1 .B Hormone receptors and hormones ...... 237 V.2.l.B.a Estrogen receptor ...... 237 C.2.1.B.0 3-hCG...... 237 V.2.l.C Heat shock proteins ...... 241 V.2.1.C.aHsp90...... 241 V.2.1.C.b Hsp/Hsc70 ...... 241 V.2.1.C.cHsp27 ...... 241 V.2.l.D Tumor suppressor gene ...... 245

15 V.2.1.D.a BRCA 1...... 245 V.2.2 Differential display of genes on the Atlas human cancer cDNA array...... 248 V.2.2.A Initial analysis of the array results reveals a broad spectrum of altered gene expression ...... 251 V.2.2.B Several genes differentially expressed in the clonal cell lines are regulated in ways that would allow the Bm-3b overexpressing cell line to maximize proliferation potential ...... 251 V.2.2.C Closer analysis of the genes differentially expressed in the array reveals some counterintuitive results ...... 252 V.2.2.D Several neuronal genes are regulated in the breast adenocarcinoma cell lines with altered levels of Bm-3b ...... 253 V.2.3 Confirmation of the cDNA array results with RT-PCR ...... 255 V.2.3.A Cyclophilin ...... 255 V.2.3.B Human RhoC ...... 255 V.2.3.C Human early growth response 1 (hEGRl) ...... 256 V.2.3.D Human junction plakoglobin ...... 256 V.2.3.E Summary of confirmation of the cDNA array results with RT-PCR.... 256 V.3 Discussion ...... 262 V.3.1 Western immunoblotting ...... 262 V.3.1.A Bm-3b short is not the only Bm-3 protein altered in the clonal cell lines. 262 V.3.l.B The levels of P-actin are altered in the cell lines with altered levels of Bm-3b...... 263 V.3.l.C Two types of differential expression of estrogen responsive genes were shown by Western immunoblotting ...... 263 V.3.2 cDNA array ...... 264 V.3.2.A Bm-3b and share molecular pathways ...... 265 V.3.2.B Bm-3 proteins play a role in auditory development ...... 266 V.3.2.C RhoC plays a role in both neuronal differentiation and tumor metastasis...... 267 V.3.2.D Early growth response protein 1 (hEGRl)...... 268

16 V.3.3 Conclusion...... 269 VI CONCLUSION...... 270 VI. 1 Discussion ...... 270 VI.2 Viewing the three results chapters as a whole ...... 272 VI.3 Future work ...... 273 VI.4 Conclusions ...... 275

17 LIST OF FIGURES AND TABLES Figure I.l: Comparison of the structural features of selected members of the nuclear receptor superfamily ...... 44 Table I.l: Classes of the nuclear receptor superfamily ...... 45 Table 1.2: Members of the pi 60 steroid receptor coactivator family ...... 60 Figure 1.2: Comparison of the structural features of selected members of the steroid receptor coactivator family ...... 64 Figure 1.3: Comparison of the SRC-la and SRC-le splice forms of SRC-1 ...... 67 Figure 1.4: The secondary structural elements of the POU domain ...... 76 Table 1.3: The six classes of POU transcription factors ...... 78 Table 1.4: Examples of DNA response elements for POU transcription factors ...... 82

Figure 1.5: Arrangement of the POUg and POUh on DNA response elements ...... 84 Table 1.5: Dimerization properties of some of the POU transcription factors ...... 87 Table 1.6: Interactions of some of the POU transcription factors with cofactors ...... 88 Figure 1.6: Comparison of the POU domains of Bm-3 a and Bm-3b ...... 91 Figure 1.7: Degenerate oligonucleotides used for the identification of POU domain containing proteins ...... 92 Figure 1.8: Sequence alignment comparing the common structural features of members of the Bm-3 family ...... 95 Table 1.7: Promoter regulation by Bm-3 family members ...... 101 Figure III.l: Affinity chromatography pull down of in vitro translated ^^S-metbionine SRC-la protein with GST/Bm-3 chimeras attached to glutathione sepharose 4B. 146 Figure III.2: Affinity chromatography pull down of in vitro translated ^^S-methionine SRC-le protein with GST/Bm-3 chimeras attached to glutathione sepharose 4B. 147 Figure III.3: Affinity chromatography pull down of in vitro translated ^^S-methionine luciferase protein with GST/Bm-3 chimeras attached to glutathione sepharose 4B...... 148 Figure III.4: Coimmunoprecipitation of Bm-3 a isoforms with SRC-1 fi*om the MCF7 breast epithelial cell line ...... 150 Figure III.5: Effects of Bm-3 a long and SRC-la cotransfection on SNAP 25 promoter transactivations in ND7 cells ...... 154

18 Figure III.6: Effects of Bm-3 a long and SRC-le cotransfection on SNAP 25 promoter transactivations in ND7 cells ...... 155 Figure III.7: Effects of Bm-3 a long or Bm-3b short and SRC-la cotransfection on SNAP 25 promoter transactivations in BHK21 cells ...... 157 Figure III.8: Effects of Bm-3 a long or Bm-3b short and SRC-le cotransfection on SNAP 25 promoter transactivations in BHK21 cells ...... 158 Figure IIL9: Effects of Bm-3a long or Bm-3b short and SRC-la cotransfection on VIT- TK promoter transactivations in ND7 cells ...... 160 Figure III. 10: Effects of Bm-3a long or Bm-3b short and SRC-la cotransfection on VIT- TK promoter transactivations in BHK21 cells ...... 162 Figure III. 11 : Effects of Bm-3a long or Bm-3b short and SRC-la cotransfection on VIT- TK promoter transactivations in MCF7 cells ...... 163 Figure III. 12: Effects of Bm-3a long or Bm-3b short and SRC-1 e cotransfection on VIT- TK promoter transactivations in ND7 cells ...... 165 Figure III. 13: Effects of Bm-3 a long or Bm-3b short and SRC-le cotransfection on VIT- TK promoter transactivations in BHK21 cells ...... 166 Figure III. 14: Effects of Bm-3a long or Bm-3b short and SRC-le cotransfection on VIT- TK promoter transactivations in MCF7 cells ...... 168 Figure III. 15: Effects of Bm-3 a long or Bm-3b short and SRC-3 cotransfection on VIT- TK promoter transactivations in ND7 cells ...... 170 Figure III. 16: Effects of Bm-3 a long or Bm-3b short and SRC-3 cotransfection on VIT- TK promoter transactivations in BHK21 cells ...... 171 Figure III. 17: Effects of Bm-3a long or Bm-3b short and SRC-3 cotransfection on VIT- TK promoter transactivations in MCF7 cells ...... 172 Figure IV. 1: RT-PCR analysis of antisense Bm-3b in clonal MCF7 cell lines transfected with Bm-3b antisense construct ...... 189 Figure IV.2: RT-PCR analysis of antisense Bm-3b in clonal MCF7 cell lines transfected with Bm-3b short construct ...... 190 Figure IV.3: Westem blot analysis of Bm-3b in clonal MCF7 cell lines with altered levels of Bm-3b ...... 192

19 Figure IV.4: Cellular growth curves of clonal MCF7 cell lines with altered levels of Bm- 3b...... 196 Figure IV.5: Saturation density of clonal MCF7 cell lines with altered levels of Bm-3b...... 198 Figure IV.6: Tritiated thymidine incorporation by clonal MCF7 cell lines with altered levels of Bm-3b ...... 200 Figure IV.7: Photomicrograph (lOx) MCF7 cells stably overexpressing pLTR Bm-3b short construct grown in monolayer ...... 201 Figure IV. 8: Photomicrograph (20x) MCF7 cells stably overexpressing pLTR Bm-3b short construct grown in monolayer ...... 202 Figure IV.9: Photomicrograph (40x) MCF7 cells stably overexpressing pLTR Bm-3b short construct grown in monolayer ...... 203 Figure IV. 10: Photomicrograph (20x) MCF7 cells stably overexpressing pLTR empty control construct grown in monolayer ...... 204 Figure IV.ll: Photomicrograph (20x) MCF7 cells stably overexpressing pJ4 Bm-3b antisense construct grown in monolayer ...... 205 Figure IV. 12: Anchorage independent colony formation by clonal MCF7 cell lines with altered levels of Bm-3b ...... 207 Figure IV. 13: Photomicrograph (20x) of anchorage independent colony formation by MCF7 cells stably overexpressing pLTR Bm-3b short construct ...... 208 Figure IV. 14: Photomicrograph (20x) of anchorage independent colony formation by MCF7 cells stably overexpressing empty pLTR control construct ...... 209 Figure IV. 15: Photomicrograph (20x) of anchorage independent colony formation by MCF7 cells stably overexpressing pJ4 Bm-3b antisense construct ...... 210 Table IV. 1: Cellular growth parameters of clonal MCF7 cell lines with altered levels of Bm-3b...... 211 Figure IV. 16: Effect of estradiol and tamoxifen, as determined by cell counts, on the growth of clonal MCF7 cells with altered levels of Bm-3b ...... 214 Figure IV. 17: Effect of estradiol and tamoxifen, as determined by cell counts, on the growth of clonal MCF7 cells with altered levels of Bm-3b ...... 215

20 Figure IV. 18: Effect of estradiol and tamoxifen, as determined by tritiated thymidine incorporation, on the proliferation of clonal MCF7 cells with altered levels of Bm- 3b...... 218 Figure IV.20: Effect of estradiol and tamoxifen, as determined by tritiated thymidine incorporation, on the proliferation of clonal MCF7 cells with altered levels of Bm- 3b...... 220 Figure IV.21: Effect of estradiol and tamoxifen on the anchorage independent growth of clonal MCF7 cells with altered levels of Bm-3b ...... 222 Figure IV.22: Effect of estradiol and tamoxifen on the anchorage independent growth of clonal MCF7 cells with altered levels of Bm-3b ...... 223 Table IV.2: Anchorage independent growth parameters of clonal MCF7 cell lines with altered levels of Bm-3b ...... 224 Figure V. 1. Westem blot analysis of Bm-3 a long in clonal MCF7 cell lines with altered levels of Bm-3b ...... 235 Figure V.2. Westem blot analysis of Bm-3a long in clonal MCF7 cell lines with altered levels of Bm-3b ...... 236 Figure V.3. Westem blot analysis of ER in clonal MCF7 cell lines with altered levels of Bm-3b...... 238 Figure V.4. Westem blot analysis of ER in clonal MCF7 cell lines with altered levels of Bm-3b...... 239 Figure V.5. Westem blot analysis of beta human chorionic gonadotropin long in clonal MCF7 cell lines with altered levels of Bm-3b ...... 240 Figure V.6. Westem blot analysis of Hsp90 in clonal MCF7 cell lines with altered levels of Bm-3b ...... 242 Figure V.7. Westem blot analysis of Hsp/Hsc70 in clonal MCF7 cell lines with altered levels of Bm-3b ...... 243 Figure V.8. Westem blot analysis of Hsp27 in clonal MCF7 cell lines with altered levels of Bm-3b ...... 244 Figure V.9. Westem blot analysis of BRCA I in clonal MCF7 cell lines with altered levels of Bm-3b ...... 246

21 Figure V.IO. Westem blot analysis of BRCA I in clonal MCF7 cell lines with altered levels of Bm-3b ...... 247 Table V.l. Results of the differential display of genes on the Atlas human cancer cDNA array ...... 249 Figure V .ll. RT-PCR analysis of cyclophilin in clonalMCF7 cell lines with altered levels of Bm-3b ...... 258 Figure V.12. RT-PCR analysis of RhoC in clonal MCF7 cell lines with altered levels of Bm-3b...... 259 Figure V.l 3. RT-PCR analysis of hEGRl in clonal MCF7 cell lines with altered levels of Bm-3b...... 260 Figure V.14. RT-PCR analysis of junction plakoglobin in clonal MCF7 cell lines with altered levels of Bm-3b ...... 261

22 ABBREVIATIONS SECTION:

4-OH T 4-hydroxy tamoxifen aa amino acid AD Alzheimer's disease ADI SRC activation domain 1 AD2 SRC activation domain 2 AF-1 activation function-1 domain of nuclear receptors AF-2 activation function-2 domain of nuclear receptors AIB amplified in breast cancer AMP adenosine monophosphate APP amyloid precursor protein APS ammonium persulphate AR ATP adenosine triphosphate bHLH basic helix-loop-helix bp CAT chloramphenicol acetyltransferase CBP CREB-binding protein CDK cyclin dependent kinase cDNA complementary DNA CREB cyclic AMP response element-binding protein DBD DNA-binding domain dCTP deoxycytosine triphosphate ddH20 double-distilled water DMSO dimethylsulphoxide DNA deoxyribonucleic acid DNA-bd DNA-binding domain DR direct repeat E2 17-(3 estradiol ECL enhanced chemiluminescence

23 EDTA ethylenediaminetetra-acetic acid ERa estrogen recepotor alpha ERP estrogen recepotor beta ERA? estrogen receptor related protein ERE estrogen response element ERR estrogen related receptor PCS fetal calf serum GR GST glutathione-S-transferase HAT histine acetyltransferase KBS HEPES-buffered saline HBSS Hank's balanced salt solution hCG human chorionic gonadotropin HEPES N-[hydroxyethyl] piperazine-N2'-[2-ethanesulphonic acid] HRE hormone response element HR? horseradish peroxidase Hsp heat shock protein HTH helix-tum-helix lAA isoamyl alcohol Inr initiator element IPTG isopropyl b-D-thiogalactopyranoside JR inverted repeat IVT in vitro translation kb kilobase kDa kilo Dalton LB Luria Bertani LBD ligand-binding domain LTR long terminal repeat LXD nuclear receptor interaction domain (LXXLL motif) MMLV Moloney murine leukemia vims mRNA messenger RNA

24 N.D. not determined NID nuclear receptor interaction domain (LXXLL motif) NLS nuclear localisation signal NR box nuclear receptor interaction domain (LXXLL motif) NRD nuclear receptor interaction domain (LXXLL motif) OLE oligo labelling buffer PAGE polyacrylamide gel electrophoresis PAS Per/Ant/Sim homology domain PBS phosphate-buffered saline PCR polymerase chain reaction PIC pre-initiation complex PMSF phenylmethylsulfonylfluoride Pol II RNA polymerase I POU Pit-Oct-unc family of transcription factors

POUh POU-homeodomain POUs POU-specific domain PR RAR retinoic acid receptor RE response element RIP receptor interacting protein RNA ribonucleic acid rRNA ribosomal RNA RT-PCR reverse transcription PCR RXR SDS sodium dodecyl sulphate SHR steroid SRA steroid receptor RNA activator SRBl SRC steroid receptor binding domain 1 SRB2 SRC steroid receptor binding domain 2 SRC steroid receptor coactivator SSC standard saline citrate

25 STAT signal transducer and activator of transcription TAE tris-acetate-EDTA Tam tamoxifen IBP TATA binding protein TBMED N,N,N’,N'-tetramethylethylene-diamine ThR tk thymidine kinase TNT transcription and translation tRNA transfer RNA UPE upstream promoter element UV ultraviolet VDR Vit vitellogenin

26 I G eneral Introduction

Organization in a mammalian cell begins with the nontrivial problem of packing 200cm or so of DNA into the 5|im diameter of the nucleus. Of course, the DNA backbone is highly charged, so basic (histone) proteins are required for charge neutralization to allow compact packing. The packaged DNA is then anything but inert. Transcription of the DNA template as well as its replication and repair all have to take place, and in an environment whose protein concentration exceeds that of most crystals. Once the transcript is made, it has to be processed. Every single transcript in eukaryotic cells, with the exception of the 5S rRNA in some species, is made as a precursor that requires nuclear processing to produce its mature form. These DNA- and RNA- based transactions are almost universally carried out by complex multimolecular machines whose relative distributions in the nucleus are just beginning to be understood (Berk & Mattaj, 1997).

LI Transcription

1.1.1 Cellular communication results in the regulation of transcription

A relatively limited number of fundamental pathways control all intra- and inter- cellular communication. One pathway utilises receptors anchored in the cytoplasmic membrane of the cell. Upon binding of ligand, these molecules undergo a conformational change which then induces an incompletely understood signal transduction cascade which ultimately results in the activation or repression of specific gene expression. Another pathway uses ion channels composed of pore forming proteins which, too, span the plasma membrane of the cell and open in response to specific stimuli. The ion channels may be regulated by: changes in electrical potential across the membrane (voltage gated), by ligand binding (ligand gated), or by mechanical forces (mechanically gated). Still another pathway involves receptors found elsewhere in the cell whose small and diffusible ligands may pass directly through the cellular and nuclear membranes. Once these ligands are bound by the nuclear receptor, the receptor undergoes a conformational change and may act directly as a ligand inducible , or may induce a signal transduction cascade which ultimately ends in the activation or repression of

27 specific genes. Each of these pathways results in the tight control of specific gene expression through the regulation of transcription.

1.1.2 Every cell contains the same set of genes

If as is generally implied in genetic work (although not often explicitly stated), all of the genes are active all of the time, and if the characters of individuals are determined by the genes, then why are not all the cells of the body alike?... Every cell comes to contain the same kind of genes. Why then is it that some cells become muscle, some nerve cells, and others remain reproductive cells? from Morgan‘s 1935 Nobel Prize speech (Sander, in 1985)

Precise and tight control of the production of proteins is the fimdamental basis for the generation of specific cell types from a single fertilized egg. Proteins are encoded by DNA genes which are selectively transcribed into RNA by RNA polymerase II, and the RNA may then be translated into protein. The region upstream of each gene is a regulatory region called a promoter. The promoter, as well as other DNA sequences called enhancers and upstream promoter elements (discussed in section 1.1.4), contain the necessary DNA sequence information that allows the gene product to be produced at the correct time and in the correct location.

1.1.3 There are three types of RNA polymerase

Three different types of RNA polymerase are responsible for the transcription of eukaryotic DNA templates into functional RNAs. The work herein focuses on the activities of RNA polymerase II which is responsible for the transcription of all of the genes which encode proteins (mRNAs), as well as genes which encode for small nuclear RNAs involved in post-transcriptional RNA splicing (Ul, U2, U3, etc.). RNA polymerase I is responsible for the transcription of ribosomal RNAs (28S, 18S, and 5.8S rRNAs). The transfer RNA (tRNA) genes, the 5S rRNA genes, and the U6 small nuclear RNAs are transcribed by RNA polymerase III.

28 1.1.4 Promoters, Enhancers, and UPEs control transcription

1.1.4.A Promoters

Prokaryotic and eukaryotic genes contain promoter regions 5' to the start of transcription. Two promoter elements, the TATA element and the Initiator (Inr) element, are capable of directing the specific transcription initiation of protein-encoding genes by RNA polymerase II. About 80% of eukaryotic polymerase II transcribed genes have an eight base pair TATA box centered at approximately -25 base pairs from the start site of transcription. The Initiator sequence found in the promoters of most housekeeping genes, as well as some tissue specific genes, is located over the start site of transcription between nucleotides -3 and +5 in relation to the start site of transcription. All RNA polymerase II promoters contain one or both of these elements (Weis & Reinberg, 1992).

1.1.4.B Upstream promoter elements

Other regulatory elements found located upstream of the promoter in a wide variety of genes with different patterns of expression have been termed upstream promoter elements (UPEs) (Goodwin, 1990). These sequence elements play a role in stimulating the constituitive basal activity of a promoter, and their elimination decreases or abolishes transcription from these promoters. The Spl box (a GC rich sequence) and the CCAAT box (sequence implied in name) are both UPEs, and promoters may contain one or both in single or multiple copies (Latchman, 1998b).

1.1.4.C Enhancers

Tight control of transcription is also aided by enhancers. Enhancers, like promoters contain specific DNA sequence elements, however they differ in that they may be located several thousand base pairs away and at either the 5' end, the 3' end, or even within the transcribed region of the gene that they control. Moreover, enhancer sequence elements

29 may be placed in any orientation relative to the promoter (Blackwood & Kadonaga, 1998).

I.1.4.D The beta-globin gene

The beta-globin gene promoter was one of the earliest and most thoroughly studied promoter regions. Studies in the late 1980s showed that the beta-globin gene and its flanking regions contain a promoter element and two enhancer elements which control erythroid specific regulation of transcription. Experiments have shown: 1. binding of ubiquitous and erythroid-specific transcription factors to these regions (deBoer et ah, 1988; Wall, deBoer & Grosveld, 1988); 2. deletions resulting in the silencing of the beta- globin gene (Van der Ploeg et al, 1980); 3. that the enhancer region is capable of increasing the expression of the beta-globin gene in a copy-dependent manner (Grosveld et al., 1987); and 4. that this control region is capable of activating both homologous and heterologous promoters in a tissue specific manner (Blom van Assendelft et al., 1989). Each of these experiments has led to our current understanding of the mechanisms of control which lie in the promoter and enhancer sequences.

1.1.5 Gene specific promoter elements

Different promoter response elements determine if a gene is constitutively expressed, expressed at a specific time, or expressed in a specific location. Some elements are recognized ubiquitously: such as the TATA box bound by the TATA binding protein (TBP) or those factors which bind to the Spl or CAAT box. Other promoter and enhancer elements are recognized in a tissue-specific or time-dependent manner. The rate of initiation of transcription is dependent upon transcription factors that recognize elements in enhancers and the promoter. Indeed, the upstream binding transcription factors influence the events at the TATA box by promoting the formation of a transcriptional pre-initiation complex (PIC).

30 Active forms of transcription factors which bind specific promoter or enhancer response elements are available for DNA-binding only under conditions when the gene is to be expressed. Absence of an activating form of a transcription factor or presence of an inhibitory form of a transcription factor results in the gene not being activated by that particular circuit. Activating and repressing forms of transcription factors will be discussed after the stages of transcription have been presented.

1.1.6 The five stages of transcription

The transcription of a gene may be divided into five separate stages: 1. formation of the pre-initiation complex (PIC), 2. initiation of transcription, 3. promoter clearance, 4. transcript elongation, and 5. termination of transcription.

1.1.6.A The formation of the pre-initiation complex

1.1.6.A.a Stepwise assembly

In 1989, Stephen Buratowski and colleagues published an elegant paper in which native gel electrophoresis, electromobility shift assay, DNAse I footprinting and gradient elution chromatography were used to gain the first clear insights into the stepwise assembly of the basal transcription factors and RNA polymerase II on a promoter template. Initially the TATA-binding protein (TBP) subunit of TFIID recognizes and binds the TATA box in the promoter. Simultaneously, or soon thereafter, TFIIA binds to the upstream side of TFIID stabilizing the TFIID/DNA complex. TFIIB binds the TFIIA/TFIID/DNA complex next. TFIIB is required for the formation of subsequent pre-initiation complexes. TFIIB most likely acts as a bridge between TFIIA/TFIID/DNA and RNA polymerase II. Finally, TFIIE, TFIIH, and TFIIJ bind the pre-initiation complex (Buratowski eta l, 1989).

31 I.1.6.A.b The holo-RNA polymerase II complex

The stepwise assembly of the functional class II PIC on the core promoter has been characterized and shown to occur in a particular ordered manner; however, it has also been shown that some preassembled RNA polymerase II may be found in solution already associated with a subset of the basal transcription factors (TFIIB, TFIIF, and TFIIH). It has been proposed that following the binding of TFIID and TFIIA to the core promoter element, that this preassembled holo-RNA polymerase II complex may bind, thereby reducing the number of steps to produce a functional PIC (Greenblatt, 1997).

By whichever path the assembly of the PIC occurs (stepwise or through the addition of the preassembled holo-RNA polymerase II complex), the complex of the seven basal transcription factors (TFIIA, B, D, E, F, H, and J) is sufficient for transcription to occur. Gene specific transcription factors may act to accelerate or inhibit the formation of this pre-initiation complex, thereby activating or repressing the transcription of a specific gene.

1.1.6.B The initiation of transcription and promoter clearance

The initiation of transcription is marked by the formation of the first phosphodiester bond linking the first two RNA residues. The less well-defined promoter clearance, however, occurs when the RNA polymerase II along with TFIIF leave the initiation complex to start transcript elongation. At this point TFIID and TFIIA remain bound to the promoter to allow for the formation of another PIC and subsequent cycles of transcription.

1.1.6.C Transcript elongation and termination

Five general elongation factors (P-TEFb, SII, TFIIF, Elongin (SIII), and ELL) are required to ensure that polymerase activity does not arrest or pause during mRNA synthesis (Reines, Conaway & Conaway, 1996).

32 Termination of transcription, to date, is poorly defined; the exact termination signals for RNA polymerase II within a “termination region” have not been elucidated. However, more important than the termination, is the cleavage of the primary mRNA transcript, which ultimately determines the 3’ end of the transcript. The precise cleavage point, which also determines the start point of the addition of a poly(A) tail, is marked by the consensus sequence 5’-AAUAA-3’ (Wahle & Keller, 1992).

1.1.7 Gene specific transcription factors

While the polymerase II general transcription factors are common to all polymerase II transcribed genes, gene specific transcription factors are responsible for the activation or repression of specific genes to ensure tightly controlled tissue- and time-specific gene expression. Other modes of gene specific regulation do occur post-transcriptionally (RNA splicing and post-translational modifications), however most regulation occurs at the level of deciding which genes will be transcribed into an mRNA transcript. Examination of promoter and enhancer regions reveals short DNA motifs, response elements (REs), common to genes with similar expression patterns and absent in genes with different expression patterns. For example, genes whose transcription is induced by the addition of the steroid hormone estrogen share a common response element, the estrogen response element (ERE), while those unaffected by the addition of estrogen generally do not contain an ERE. Moreover, when specific response elements are cloned into heterologous promoters, the heterologous promoter may be induced in response to the stimulatory factor. These DNA response elements act by binding gene-specific transcription factors. These transcription factors may control the constitutive expression of genes, the inducibe expression of genes, or the inactivation of gene expression at particular times in particular tissues to ensure tight, specific, and efficient control of gene expression.

1.1.7.A Gene-specific transcription factors have a modular structure

33 Gene-Specific transcription factors are composed of separate and functionally independent domains: a DNA-binding domain and a regulatory domain (Hope & Struhl, 1986; Ptashne, 1988). In general, the DNA-binding domain of transcription factors is used to classify a family of related transcription factors. It is not the DNA-binding domain, however, that accounts for the functional ability of a transcription factor as an activator or an inhibitor. Other domains in the transcription factor: activation domains or inhibitory domains, are the fimctional domains of the transcription factor. Interestingly, the DNA-binding domains and the functional domains of transcription factors are able to function independently. Thus, the DNA-binding domain of GAL4, a yeast transcriptional activator, may be fused to VP 16, a herpes virus activator unable to bind DNA, and the resulting chimera works as does the wild type GAL4 itself (Sadowski et al, 1988).

1.1.8 Transcription factors have several different types of DNA-binding domain

As mentioned earlier, transcription factors contain DNA-binding domains capable of recognizing and binding response elements under the proper cellular conditions. Most of the DNA binding domains of transcription factors may be grouped into a few families based on common structural motifs. The sections below will include these motifs as well as specific biochemical discussion of members of the groups containing these motifs.

I.1.8.A.a Helix-turn-helix

Helix-tum-helix (HTH) was the first structural DNA-binding motif recognized. In 1966, Walter Gilbert and Benno Muller-Hill isolated and characterized the E. coli lambda repressor, while in the next year Mark Ptashne showed that it was able to bind with high affinity to DNA containing its binding site, but not to DNA lacking this binding site (Muller-Hill, 1966; Ptashne, 1967). The lambda repressor contains a HTH DNA-binding motif: two alpha helices that cross at an angle of approximately 120° linked by a beta turn. One alpha helix is a DNA-recognition helix, which lies in the major groove of its DNA response element, while the other is responsible for stabilization of the motif

34 (Wharton, Brown & Ptashne, 1984). The DNA-recognition motif binds an imperfect palindrome in the major groove. This motif is conserved from bacteria to humans. Homeodomain-containing proteins are a family of highly conserved developmentally fimdamental transcription factors originally identified as playing a critical role in the development of Drosophila melanogaster (Dura & Ingham, 1988; Ingham, 1988). The conserved contains a 60 amino acid DNA-binding homeodomain. The homeodomain consists of three alpha helices. Helix 1 separated from helix 2 by a loop and helices 2 and 3 forming a helix-tum-helix motif. Indeed the HTH of the homeodomain and the prokaryotic HTH proteins have virtually superimposable structures (Qian et ah, 1989), even though the DNA response elements and the amino acid residues which contact the DNA phosphate backbone or bases differ (Kissinger et al., 1990; Otting et al., 1990; Pabo et al., 1990). More recently, another class of homeodomain-containing transcription factors has been identified, the POU family of transcription factors (Herr et al., 1988). The POU domain (discussed in section 1.4) is made up of two DNA-binding domains, the unique POU- specific (POUs) and the POU-homeodomain (POUh), which both contain HTH motifs (Herr & Cleary, 1995). The HTH subdomains are stmcturally independent domains that functionally cooperate as a single DNA-binding unit. The structural flexibility offered by the bipartite HTH DNA-binding domain leads to a unique functional versatility in transcriptional regulation (Herr & Cleary, 1995). It must be noted however, that while the HTH is a common stmctural motif, there is no singular pattern that relates response element sequences to a particular stmctural pattern of amino acid side chains. Thus, in the HTH/DNA interaction, several sets of side chains may interact with a given nucleic acid base pair, and likewise, more than one base pair may interact with a particular amino acid side chain (Harrison, 1991). Other DNA- binding motifs, however, do rely on specific amino acid stmctural patterns; one of these is the Zinc-finger motif.

I.1.8.A.b Zinc-fînger

In the Zinc-finger motif, a zinc molecule is coordinated by histidine and cysteine residues, thereby forming a stable finger shaped molecule, one side of which contains a

35 DNA-recognition alpha helix. Transcription factors may contain several zinc fingers, each possibly, but not necessarily making multiple contacts with several DNA response elements. The Zinc-finger proteins may be divided into three classes. The first class of Zinc-finger proteins conform to the structure:

O X C X.-4 C X3 O Xs O X. H X3 - 4 H X 2 .6 , where “O” is a hydrophobic amino acid, and “X” is any amino acid. In this structure, one zinc ion is liganded by two cysteines and two histidines (Brown, Sander & Argos, 1985; Miller, McLachlan & Klug, 1985), resulting in a folded structure with a twelve residue helix packed against a P-hairpin finger (Berg, 1988). Most proteins containing these motifs contain three or more of these motifs in direct succession, suggesting that they form a similar repeating structure when bound to their DNA target (Berg, 1988). The second class of Zinc-finger proteins includes members of the nuclear receptor superfamily (discussed in section 1.2), which generally bind as dimers to direct or inverted repeat DNA response elements. In these molecules, the Zinc-finger motif involves an approximately 70 amino acid domain in which two Zinc ions are each coordinated by four cysteines (Freedman et al, 1988a; Freedman et al., 1988b). The name double-loop Zn-helix (LZnH) has been proposed for this structure (Schwabe, Neuhaus & Rhodes, 1990), because it contains two loop helix structures in which two cysteines at the beginning of the loop and two cysteines at the end of the helix coordinate a single zinc ion. The resulting helices are packed orthogonal to one another (Hard et al., 1990). The class three Zinc-finger coordinates two zinc ions with six cysteines. All of the members of this class are found in yeast transcriptional activators, the most studied of which is the GAL4 DNA-binding domain (Johnston, 1987). GAL4 binds as a dimer to a symmetric 17 base pair site (Carey et al., 1989; Giniger, Vamum & Ptashne, 1985).

I.1.8.A.C Leucine zipper and basic DNA-binding domain

The Leucine zipper and basic DNA-binding domain does not refer specifically to a DNA- binding domain, but rather to the characteristic structure which allows two alpha helices to dimerize thereby facilitating the correct positioning of two basic regions N-terminal to the alpha helices to interact directly with acidic DNA (Gentz et al., 1989; Kouzarides &

36 Ziff, 1988; Landschulz, Johnson & McKnight, 1989; Turner & Tjian, 1989). In the helical portions of the leucine zipper proteins, every seventh amino acid is a leucine. As there are on average 3.6 amino acids per turn of an alpha helix, all of the leucines appear on the same side of the helix after every two helical turns, thereby forming a hydrophobic face. The leucines on each helix interdigitate forming an alpha helical parallel coiled coil (Oas et al,, 1990; O'Shea, Rutkowski & Kim, 1989). The DNA-binding portion of the motif is a highly cationically charged region containing several arginine and lysine residues. The DNA-binding region attaches to the major groove of the DNA in a “scissors-grip” manner (Vinson, Sigler & McKnight, 1989). The proto-oncogene products c-Jun and c-Fos which heterodimerize to form the AP-1 transcription complex, interact with one another through leucine zippers and with the AP- 1 response element through V-terminal basic residues. The leucine zipper and basic DNA-binding domain has been detected in several other transcription factors including the CCAAT-box binding protein C/EBP, and the yeast factor .

L1.8.A.dHelix-loop-helix domain

Like the leucine zipper proteins, the helix-loop-helix (HLH) proteins are characterized by a dimerization domain which allows a basic alpha helical region to make contacts with the DNA binding site. These proteins dimerize through an alpha helical coiled-coil, which, unlike the leucine zipper proteins, whose dimerization results in the creation of a two helix bundle, results in a four helix bundle (Ferre-D'Amare et al., 1993). Proteins containing this motif are capable of both homo- and hetero- dimerization. The classic HLH binding site is the E-box, which is defined by the sequence CANNTG (Ferre- D'Amare et al., 1993).

I.1.8.A.C Other DNA-binding motifs

It is important to note that several interesting and important DNA-binding domains exist that do not fall into the classes listed above. The DNA-binding domain of the human tumor suppressor gene product does not fall into any of the above three categories, but does bear some resemblance to the N-terminal section of the NFk B DNA-binding domain

37 (Muller et al., 1995). Similarly, the DNA-binding domains of the UBF ribosomal RNA transcription factor and the HMG proteins’ HMG-box share similarity with one another (Bazett-Jones et al, 1994; Grosschedl, Giese & Pagel, 1994; Hu et al, 1994). The Ets family of transcription factors were originally identified based on their with the v-ets oncogene product, but were later reclassified in the winged helix-tum-helix family based on structural data (Liang et al, 1994a; Liang et al, 1994b; Sharrocks et al., 1997). Similarly, a “winged-helix” defines the forkhead family of transcription factors which bind DNA as monomers in both the major and minor grooves and as a result of their binding, bend the DNA (Burley et al., 1993; Clark et al., 1993; Lai, Cleary & Herr, 1992). There are also transcription factors with DNA-binding domains which bear a resemblance to no other known DNA-binding motif; Ap-2 and CTF/NFI are examples (Latchman, 1998b). These transcription factors may represent the founding members of whole new classes of transcription factor DNA-binding motifs.

1.1.9 Activation and inhibition of gene transcription by transcription factors

Quite simply, a transcription factor may act as an activator by directly or indirectly aiding in the formation of the RNA polymerase II preinitiation complex either by facilitating the recmitment of transcriptional machinery to the PIC, or by modifying chromatin stmcture to make promoter and enhancer elements more accessible. Likewise, an inhibitory transcription factor interferes, either directly or indirectly, with the formation of the PIC by competing for activator response elements, interfering with the functional site of activators, interaction with the basal transcriptional machinery, or interaction with higher order chromatin structure.

1.1.10 Chromatin remodeling by transcription factors

Several transcription factors act directly on chromatin structure to regulate transcription. Chromosomal DNA is organized into a tight nucleoprotein package termed chromatin. Chromatin is composed of repeating subunits of nucleosomes. Each nucleosome is

38 composed of an octamer of two of each of the histone proteins (H2A, H2B, H3, and H4) with 147 base pairs of DNA wrapped around the octamer. A nucleofilament is then formed by linking nucleosome cores by short stretches of DNA bound by linker histones (HI and H5). Both the amino- and carboxy- terminal tails of histones are flexible and extend out from the nucleosome core. To date, acylation of the lysine residues on these tails has been the most extensively studied aspect of chromatin-based transcriptional regulation. The addition of an acyl group to the free amino group of these lysine residues results in a reduced net positive charge in the histone, resulting in a weaker ionic interaction and consequently less tightly packed chromatin (Glass & Rosenfeld, 2000). Chromatin may also be altered in an ATP-dependent manner. Two related remodeling complexes that rely on the hydrolysis of ATP have been most extensively studied in yeast (RSC and the SWI/SNF complex). In addition, homologues of the SWI/SNF complex have been identified in flies and humans. The discovery of complexes of proteins that acylate nucleosomes, deacetylate nucleosomes, and remodel nucleosomes in an ATP- dependent manner has aided the understanding of the important role that chromatin structure plays in regulating the transcription of eukaryotic genes (Krebs & Peterson, 2000; Wolffe & Guschin, 2000).

1.1.11 Combinatorial regulation of transcription

The presentation above may mislead one to the oversimplified assumption that single transcription factors on single response elements are responsible for the tight control of gene expression. In vivo, multiple different response elements for multiple different transcription factors are enabled at specific points in time in specific tissues. Activation or inhibition of gene expression is stimulated or counterbalanced more efficiently by the combination of multiple transcription factors bound at distinct response elements. This synergy and balance makes possible several physiologically distinct responses from a single gene. Unsurprisingly, this cooperation is not limited to different transcription factors acting on a single gene; multiple response elements for the same transcription factor may also result in a combinatorial synergy.

39 Direct protein-protein interactions between transcription factors in the regulatory region of a single gene can also influence expression by several molecular mechanisms. A DNA-binding protein with marginal transcriptional activity may be converted to a strong activator by interacting with a non-DNA-binding protein with strong transcriptional activity, as is the case with Oct-1 and VP 16 (Stem, Tanaka & Herr, 1989). Or, as is the case with steroid hormone receptors, two proteins may interact to form a homodimer and then cooperatively bind to a response element under conditions in which the monomer could not bind (Tsai, Tsai & O'Malley, 1989). Similarly, heteromeric proteins, like the AP-1 transcription factors Fos and Jun, may complex to facilitate a single DNA-binding event (Umayahara et al, 1994). Finally, interactions between two DNA-binding transcription factors can augment the individual transcriptional regulation of each as is seen with the AP-1 complex and estrogen receptor (Gaub et al., 1990). A single activator bound at a single DNA site in a promoter generally results in low levels of transcriptional activation. Transcription is stimulated by factors of 5 - 1000, however, by the combination of multiple transcriptional activators bound at distinct promoter sites (Struhl, 1991). It is these interactions that give rise to diverse patterns of gene expression.

40 In the sections that follow, three groups of transcriptionally active molecules will be introduced. In the next section, 1.2, the first of these, the nuclear receptor superfamily of transcription factors, and specifically the estrogen receptor, will be introduced. This major family of transcription factors, however, does not regulate transcription without interacting with other transcriptional regulators and transcription factors. In section 1.3, the p i60 steroid receptor coactivators, a family of proteins which fimctionally interact with the steroid receptors will be introduced. Finally in section 1.4, the Bm-3 POU domain-containing proteins, which are also capable of functionally interacting with the estrogen receptor, will be discussed. With this background, the context for the work of this thesis will be set. The interplay between these factors in several experimental systems will be discussed in each of the three results chapters.

41 1.2 The nuclear receptor superfamilv

1.2.1 History

The lipophilic hormone ligands of the nuclear receptor superfamily (including steroids, retinoids, and thyroid hormone) have been known as powerful regulators of gene transcription since ecdysone (an insect hormone) was shown to induce chromosomal puffing in the of insects (Ashbumer, 1980). In the early 1970s radiolabelled steroid and thyroid hormones were used to identify their respective receptors (Jensen et al, 1972). These experiments led the way to the discovery that binding of a hormone by its receptor causes a change in the receptor resulting in its binding to specific sites in DNA (hormone response elements), leading to the activation or repression of hormone-responsive genes (Ivarie & O'Farrell, 1978). The first of these receptors to be cloned were estrogen receptor and glucocorticoid receptor (Mangelsdorf et al., 1995). Soon after, several receptors were identified, purified and cloned. The cloning of these receptors was crucial to support the model that chemically distinct hormone ligands activate biologically distinct cognate hormone receptors. Receptors for all the known nuclear hormones have been found. In addition, several orphan nuclear receptors have been identified for which the ligand is not yet known (Mangelsdorf et al., 1995).

1.2.2 The general structural and functional domains of nuclear hormone receptors

I.2.2.A Structural domains

The structure of nuclear receptors is defined by a central DNA-binding domain, which targets the ligand-bound receptor to specific DNA sequences known as hormone response

42 elements. The DNA-binding domain (Figure I.l) is the most highly conserved domain across the family and is composed of two conserved class II zinc fingers (discussed in section I.l.S.A.b) that differentiate nuclear receptors from other DNA-binding proteins. The zinc-fingers function as described earlier in the introduction and allow for the specific recognition of direct or inverted DNA sequence repeats (Zilliacus et al., 1995). The C-terminal portion of the receptor encompasses the ligand-binding domain (LED) (Figure I.l) which is responsible for hormone recognition and ensures both the specificity and selectivity of the physiological response. The LED acts as a switch that, upon binding ligand, shifts the receptor to its transcriptionally active state. Not suprisingly, the cooperation between an independent ligand-binding domain with an independent DNA- binding domain makes the nuclear receptors powerful transcriptional regulators. Consequently, the success of the family is reflected by the observation that evolution has selected nuclear receptors as the largest group of eukaryotic transcription factors (Weatherman, Fletterick & Scanlan, 1999).

I.2.2.B Functional domains

Most of the members of the nuclear receptor superfamily contain two activation domains, an activation function 1 domain (AF-1) in the N-terminus of the protein, and an activation function 2 domain (AF-2) in the ligand-binding domain (Danielian et al., 1992; Lees, Fawell & Parker, 1989; Tora et al, 1989).

43 A/BCD E F

1 185 250 311 551595 estrogen receptor a (ERa)

1 148 214 304 500 530 estrogen receptor P (ER^)

567 633 680 934

progesterone receptor (PR)

421 486 528 777 glucocorticoid receptor (GR)

124 89 192 427 vitamin D receptor (VDR)

1 88 153 198 462 retinoic acid receptor (RAR)

Figure 1.1: Comparison of the structural features of selected members of the nuclear receptor superfamily. The sequences have been aligned with reference to the DNA-binding domain, C. The figures above the protein representations indicate the number of amino acids with number 1 being the most A-terminal. The approximate location of each of the six domains (A-F) is indicated on the topmost line, and each of these domains is represented by a different colour in the figure (where: A/B is red, C is yellow, D is green, E is blue and F is purple).

44 1.2.3 The members of the nuclear receptor superfamily

Presently, over 300 different members of the nuclear receptor superfamily exist, including examples from worms to humans (Whitfield et al, 1999). This superfamily can be broadly divided into four classes based on dimerization and DNA-binding capabilities. The work herein focuses on the class I steroid hormone receptors.

THE CLASSES OF THE NUCLEAR RECEPTOR SUPERFAMILY

CLASS CHARACTERISTICS EXAMPLES DIMERIZATION DNA HALF SITE

steroid hormone receptors estrogen receptor (ER) homodimer IR 5' (G/A)GGTCA glucocorticoid receptor (GR) homodimer IR 5' PuGGTCA progesterone receptor (PR) homodimer IR 5‘ PuGGTCA

non-steroid hormone receptors retinoic acid receptor (RAR) heterodimerize with RXR DR 5' PuGGTCA vitamin D receptor (VDR) heterodimerize with RXR DR 5' PuGGTCA thyroid hormone receptor (ThR) heterodimerize with RXR DR 5' AGGTCA

orphan nuciear receptors Rev-ErbAa homodimer DR 5' PuGGTCA RLD-1 heterodimerize with RXR DR 5' PuGGTCA COUP-TF homodimer various DR and IR

IV orphan nuclear receptors NGFIB monomer 5' AAAGGTCA Rev-ErbAa monomer 5' (A/T)A(A/T)NTPuGGTCA

Table 1.1: Classes of the nuclear receptor superfamily. This table represents one proposed broad grouping of the nuclear receptors based on ligand characteristics and dimerization properties. The consensus half site is shown, however promoter response elements often contain imperfect versions of this consensus. Response elements found as inverted repeats are denoted “IR”; and, response elements found as direct repeats are denoted “DR.”

45 1.2.4 Class I, the steroid hormone receptors

1.2.4.A Description

The steroid hormone receptors (SHRs) are different in several respects from all other nuclear receptors. In general, the SHRs exert their effects in and adult homeostasis as hormone activated transcriptional regulators. Like the other nuclear hormone receptors, the class I SHRs have a modular structure, including the aforementioned DBDs and LBDs (Figure I.l), as well as the transcriptional activation domains, AF-1 and AF-2. In general, the steroid hormone receptors are different form the other nuclear receptors in that they are characterized by their unique ability to bind palindromic hormone response elements (HREs) as homodimers, with the receptors for glucocorticoids, mineralocorticoids, progesterone and androgens binding the DNA half­ site 5 ' AGAA.CA 3 ', and the estrogen receptor recognising 5 ' AGGTCA 3 '. Biophysical studies have shown that the half sites are distinguished by several amino acids in the different receptors’ DNA-binding domain called the P box. The amino acids of the P-box are part of a recognition helix that is coordinated by a zinc-binding motif and makes base-specific contacts with the major groove. The D-box contains several amino acids responsible for the specific homodimerisation of the steroid nuclear receptors. A second zinc atom coordinates an alpha helix which is orientated alongside the DNA with the D-box responsible for specific homodimerization (Glass, 1994). After binding to the DNA, the receptor is thought to interact with other regulatory factors including members of the basal transcriptional machinery. Several such interactions have been described (Katzenellenbogen, 1996; Katzenellenbogen & Katzenellenbogen, 1996); however, the exact and specific mode of action of the steroid nuclear receptors on different promoters remains elusive.

1.2.4.B The unliganded steroid nuclear receptor

46 In contrast to the other nuclear receptors, the unliganded steroid nuclear receptors are associated with a large multiprotein complex of chaperones (including Hsp90, Hsp70 and Hsp56) which maintains the receptors in an inactive but ligand-inducible state (Pratt, 1993). Bohen and Yamamoto in 1993 showed that yeast strains expressing mutant forms of Hsp90 did not lead to constituitively activated glucocorticoid receptor, rather the mutant chaperones led to significant impairment of hormone induction (Bohen & Yamamoto, 1993). In addition to the Hsps, other chaperones are necessary to maintain the hormone inducibility of the steroid hormone receptors (Caplan et a/., 1995; Kimura, Yahara & Lindquist, 1995).

1.2.4.C The ligand-bound steroid nuclear receptor

Upon binding of hormone, the steroid receptor undergoes a conformational change resulting in its dissociation from its multiprotein chaperone complex (Pratt & Toft, 1997). The free steroid receptor is then able to dimerize, then regulate transcription in the nucleus (Mangelsdorf etal., 1995).

1.2.4.D The Estrogen receptor

1.2.4.D.a History

Estrogen was the first nuclear receptor ligand shown to function through a nuclear receptor, which then bound to a specific response element, and activated transcription in a ligand-dependent manner (Jensen et al, 1973a; Jensen & DeSombre, 1973; Jensen et al, 1973b). In 1986, the groups of Green and Chambon cloned and sequenced the DNA for from MCF7 human breast cancer cells (Green et al., 1986a; Green et al, 1986b; Greene et al, 1986). Research following the identification of the estrogen receptor has revealed the fundamental role that this molecule plays in the normal development and maintenance of specific tissue types, and implicated deregulation of the estrogen receptor in a variety of disease states.

47 I.2.4.D.b The role of estrogen receptors in cancer

Cancers may result from the accumulation of random genetic errors, and the opportunity for these errors to accumulate is greater in proliferating cells (Bishop, 1987; Klein & Klein, 1985; Weinberg, 1989). Endogenous and exogenous hormones are associated with cellular proliferation, thus, this proliferation-accumulated-error mechanism of carcinogenesis is particularly appropriate to the hormone-related cancers: breast, endometrium, ovary, prostate, testis, thyroid, and osteosarcoma (Henderson & Feigelson, 2000). Estrogen receptor plays an important role in breast carcinomas. The functional role of ER under normal and malignant conditions has been extremely difficult to elucidate (Bissell & Werb, 1995). Estrogen receptor is measurable in about 7% of normal breast epithelial tissue (Ricketts et al., 1991), but is expressed in about 60% of breast carcinomas (Bissell & Werb, 1995). Quite surprisingly, however, the evidence from histopathological and clinical trials indicate that ER expression correlates with morphologically differentiated tumours and better survival (Tavassoli, 1995). The question remains: if ER positive cells comprise only 7% of the total cellular pool of the breast, why are up to 60% of breast tumors ER positive. One idea is that 60% of breast epithelial cells could activate the production of ER by some unknown mechanism that may act in the early stages of malignancy. Ricketts and colleagues proposed that the average value of 7% ER positive cells for all breast epithelium is inadequate for analysis, and that the total cellular pool should be divided into non-, low-, and high- expressing categories. Thus, although the high expressors represent a minority of the total cellular population, these cells are at a higher risk for developing into cancers and account for the 60% prevalence of ER in tumors (Ricketts et al, 1991). The ER positive status of a tumour changes with the progression of the disease state. Once the ER positive tumour is developed, ER expression is gradually lost with progression, possibly as a result of selection of less-differentiated ER-negative cells (Robertson, 1996). In short term cell culture, estrogen promotes the growth of cell lines expressing functional estrogen receptors through direct effects on the cell cycle machinery (Planas-

48 Silva & Weinberg, 1997; Prall et aL, 1997), and induction of specific growth factors and their receptors (Lippman & Dickson, 1989). To study the role of ER in normal cell conditions, groups have stably transfected ER into established normal breast epithelial cell lines. Paradoxically, the response to estrogen is the opposite of what is observed in tumours and in short term culture: the cells are growth inhibited (Lundholt et aL, 1996; Zajchowski, Sager & Webster, 1993). Finally, experiments with breast cancer cell lines xenografted into nude mice have been used to define the role of ER in tumour progression. In these cases, estrogen acts as a tumour promoter, as ER positive cells will not develop into tumours without the addition of endogenous or exogenous estrogens (Shafie & Grantham, 1981; Soule & McGrath, 1980). ER does not appear to evoke the same responses in primary tumour samples, as in cell culture models, as in xenograft tumour models. Much more must be learned about the regulation and fimction of ER during tumour development. The tissue specific control of gene expression (both ER-mediating and ER-mediated) that leads to tumour generation and progression must be investigated. To complicate this situation even further, two forms of the estrogen receptor with distinct patterns of expression and functional properties have been discovered.

I.2.4.D.C Isoforms of the ER

For several years after the cloning of the first estrogen receptor it was thought that this was the only estrogen receptor that existed. Initial attempts to clone a second estrogen receptor subtype failed and instead identified two orphan receptors: estrogen related receptors 1 and 2 (ERRl and ERR2, respectively) (Giguere et aL, 1988), neither of which use estrogen as their ligand. While examples of isoforms of non-steroidal nuclear receptors existed: thyroid receptor a and p, retinoic acid receptor a and P, and retinoid X receptor a , p, and y, no known examples of steroidal nuclear receptor subtypes existed until 1996 when Kupier and colleagues cloned ERP firom rat prostate and ovary. Soon after, ERP was isolated fi*om human prostate (Mosselman, Polman & Dijkema, 1996), mouse testis (Tremblay et aL, 1997), and mouse ovary (Pettersson et aL, 1997). ERa and P are encoded by separate genes located on different chromosomes, share strong sequence homology only in the DNA-binding domain, have different expression patterns, and exhibit different transactivation properties. Both ERa and p have approximately the same affinity for estradiol, and bind DNA in a similar manner (Cowley et al., 1997; Pettersson et al., 1997). In addition, ERa and p may form functional heterodimers on EREs, indicating a previously unrecognized pathway of estrogen signaling (Cowley et al., 1997; Pettersson et al., 1997). Finally, varying ratios of ERa and P may represent a new mode of control of estrogen responsiveness (Barkhem et al., 1998). It is now clear that any study involving estrogen receptor response must take into account both subtypes of estrogen receptor.

1.2.4.D.d Structural domains of the estrogen receptor

Like all members of the steroid nuclear receptor family, estrogen receptors have six common structural/functional domains named A through F from amino- to carboxy- terminus (Figure I.l) (Kumar et al., 1987).

1.2.4.D.d.i The A/B region The variable amino terminal region A/B which contains the activation function 1 (AF-1) domain. Here the two ER subtypes share only 15% sequence homology, indicating that regions in this area may be responsible for contacting different coregulatory proteins. This region possesses the capacity for transactivation, including ligand-independent transactivation. The exact features which constitute AF-1 have not been completely elucidated. There is evidence to suggest the AF-1 function may depend upon phosphorylation (Ali et al., 1993; Joel, Traish & Lannigan, 1998; Rogatsky, Trowbridge & Garabedian, 1999). Cyclin A2-CDK2 targets serine 104 and serine 106 in the AF-1 (Rogatsky et ah, 1999). Interestingly, the cyclin A2-mediated stimulation of ERa transactivation occurs whether the cells are treated with the agonist (a compound which activates the receptor) 17-p estrodiol, or the antagonist (a compound which inhibits the activity of the estrogen receptor) 4-hydroxy tamoxifen (4-OH tamoxifen). These results indicate that interactions

50 in this area are ligand-binding independent (“ligand-binding” discussed in section 1.2.4.D.d.iii), and support the previous observation that the agonist activity of 4-OH tamoxifen is mediated by AF-1 in ERa (Berry, Metzger & Chambon, 1990; Mclnemey & Katzenellenbogen, 1996).

1.2.4.D.d.ii The C region The estrogen receptor activates gene transcription directly by binding as a dimer to its cognate palindromic DNA sequences called estrogen response elements (EREs) (Beato & Sanchez-Pacheco, 1996), or indirectly by cooperative interactions with other transcription factors such as AP-1 (Gaub et al., 1990; Philips, Chalbos & Rochefort, 1993; Umayahara et al., 1994). ERa and ERp differ by only three amino acids in their DNA-binding domains, the C domain, which contains two functionally distinct Zinc-finger motifs which interact with high affinity to the classic inverted palindromic EREs: 5 ' -

GGTCAnnnTGACC-3 '. This domain also contains a nuclear localization motif and a constitutive dimerization domain.

1.2.4.D.d.iii The E region The third major functional domain, domain E, the ligand-binding domain (LBD), contains the activation function 2 (AF-2) domain. The general structure of the ligand- binding domains of ERa and ERj3 is similar to those seen in the LBD structures of thyroid hormone receptor (Wagner et al, 1995), retinoic acid receptor y (Renaud et al, 1995) and retinoid X receptor (Bourguet et al, 1995). The AF-2 domain functions as a ligand-dependent activation domain: the conformational change which results from ligand binding structurally activates the AF-2 domain within this ligand-binding domain (Brzozowski et al., 1997; Shiau et al., 1998; Tanenbaum et al., 1998). This ligand- induced structural change alters the conformation of the AF-2 domain such that it is able to recruit coactivators (Paech et al., 1997; Tanenbaum et al., 1998). This domain is conserved throughout the entire nuclear receptor superfamily. Residues in helices 3, 5, and 12 are critical for the activation function of this domain, as they provide the area for protein/protein interactions with cofactors which potentiate AF-2 ligand-dependent activation (Feng et al., 1998; Parker & White, 1996; Wurtz et al., 1996).

51 Helix twelve in the AF-2 of the LBD is essential for interaction with cofactors. Loss or mutation of this helix results in a LBD that is unresponsive to ligand binding (Danielian et al, 1992). This area has been shown to play a significant role in differential transactivation activities of ERa and ERp (Paech et al, 1997). Binding of ligand in this pocket created by the twelve alpha helices of the AF-2 domain results in a conformational change in the positioning of helix twelve (Tanenbaum et al., 1998). Different ligands induce different helix twelve distortions resulting in differing transcriptional activation potentials. These different distortions have been used to explain the molecular mechanisms of agonism (activation of the receptor by a compound which binds the receptor) and antagonism (inhibition of the receptor by a compound which binds the receptor) in the estrogen receptor. Brzozowski and colleagues demonstrated the helix twelve deformation in the AF-2 domain is fundamentally different for the agonist, 17-p estradiol than it is for the estrogen antagonist, Raloxifene. In the case of the 17-P estradiol, helix twelve is part of a competent AF-2 that is capable of interacting with other cofactors, while in the Raloxifene bound estrogen receptor, helix twelve is rotated 130° resulting in a non-functional AF-2 domain. Interestingly, the highly conserved lysine362 which is required for efficient ligand-dependent recruitment of coactivators is partially buried in the Raloxifene reoriented helix (Brzozowski et al, 1997). In addition, domain E is in the area in which Hsp90 is bound when the receptor is in its unliganded state (Baulieu, 1987).

1.2.4.D.d.iv The D region The D region acts as a flexible linker/hinge which seperates the DNA-binding and ligand- binding functions of the protein.

1.2.4.D.e ER interaction with response elements

As mentioned before, the estrogen receptor binds to an inverted palindromic repeat 5 ' -

GGTCAnnnTGACC-3 '. It has also been shown that the sequences immediately surrounding the ERE play a role in the affinity of the ER for the ERE (Klinge et al., 1997; Klinge et al, 1992). There are several examples of genes which contain EREs including Xenopus vitellogenin gene A1 (Walker et al., 1984), and B1 (Wahli et al.,

52 1989), chicken apo VLDII (Hache et al., 1983), human pS2 (Berry, Nunez & Chambon, 1989), human oxytocin (Richard & Zingg, 1990), human c-fos (Weisz & Rosales, 1990), human c- (Weisz & Bresciani, 1988), TGF alpha (El-Ashry et al, 1996), lactoferrin (Teng et al., 1992), prolactin (Berwaer, Martial & Davis, 1994), progesterone receptor (Kraus, Montano & Katzenellenbogen, 1994), and cathepsin D (Augereau et al., 1994). These EREs are often imperfect palindromes and are dependent upon sequence, spacing and flanking regions for optimal ER binding (Anolik et al, 1993; Anolik et al, 1996).

1.2.4.D.f General mechanism of ER activation

Similar to the other members of the class I nuclear receptors, ER activation requires the binding of speciflc ligand. Upon the binding of ligand, the 90kDa heat shock protein dissociates fi-om the E region allowing ER to dimerize and making the AF-2 region available for cofactor interaction (described in section I.2.4.D.d.iii) (Horwitz et al., 1996). The homodimer is then able to interact with estrogen response elements to activate the transcription of estrogen-responsive target genes. The ER in its unliganded chaperoned resting state is phosphorylated, and the ER bound to DNA is further phosphorylated at several serine and possibly tyrosine residues (Arnold et al, 1995b; Arnold et al., 1995c; Washburn et al., 1991). The physiological role of the phosphorylation is unknown, but it has been proposed that altered phosphoprotein status may regulate ligand, DNA, or protein binding to the ER (Arnold & Notides, 1995; Arnold et al, 1995a; Arnold, Vorojeikina & Notides, 1995d; Kato et al., 1995).

1.2.4.D.g Different regions of the ER have different transactivation potential Like the other members of the steroid receptor superfamily, the estrogen receptor has two activation functions, AF-1 in the A/B domain and AF-2 in the E/F domain, which flank the DNA-binding domain of the molecule (Figure I.l). Deletion studies in ERa have shown that AF-1 exhibits constituitive activity, while the activity of AF-2 is ligand- dependent. The transactivation functions of both the AF-1 and AF-2 are cell-type and promoter-context dependent. The transactivation properties of these domains are, however distinct. Thus, a truncated ER lacking the E/F ligand-binding domain, containing the AF-2, is only 5% as active as wild-type ER on the ovalbumin gene

53 promoter in the absence of ligand (Kumar et al, 1987). An ER construct lacking the A/B domain, containing the AF-2, can activate gene expression of a reporter construct containing a vitellogenin estrogen response element (ERE); whereas this deletion severely impairs activation of the estrogen responsive human pS2 gene promoter (Kumar e tal., 1987; Tora era/., 1989). While these domains have independent and distinct activation functions, ligand-binding is considered to induce a functional synergism between AF-1 and AF-2 activities. The precise molecular mechanism is however, to date, largely unknown (Mclnemey & Katzenellenbogen, 1996; Tasset et al., 1990). In 1995, Kraus and colleagues produced the A-terminal and C-terminal domains of ERa as separate polypeptides in mammalian cells and looked for transcriptionally productive interactions in the presence of estradiol, and tamoxifen. In these experiments, it was shown thatl7-p estradiol playes a role in the association of A-terminal and C-terminal regions of the proteins which led to a functional transcriptional synergism between AF-1 and AF-2 (Kraus, Mclnemey & Katzenellenbogen, 1995). More recently it has been proposed that p300 plays a role in the transcriptional synergism between AF-1 and AF-2 through direct contacts with the A/B domain of both ERa and ERP (Kobayashi et ai, 2000). It is important, particularly in the context of this thesis, to realise that the estrogen receptor, as well as other transcription factors, function in cooperative multiprotein assemblies. The above information highlights that particular areas of the estrogen receptor are involved in different promoter and cell-type specific effects, with some regions important in some promoter and cellular contexts, and others important in different promoter and cellular contexts. This allows one to hypothesise multiple specific mechanisms whereby estrogen responsiveness may be controlled.

I.2.4.D.h The estrogen receptor must interact with other proteins

In 1989 Meyer and colleagues (Meyer et al, 1989) showed in several different pairings of nuclear receptors, that overexpression of one receptor results in the inhibition of other receptors (ie: cotransfection of ER inhibited the stimulation of transcription of reporter genes by the PR in a dose and estradiol dependent manner), suggesting that other factors

54 might modulate nuclear receptor transcriptional regulation due to the titration of a limiting amount of cellular co-regulators. In 1990, Mukheijee and Chambon published a paper in which they showed that purified human estrogen receptor does not form a detectable complex with an estrogen response element (ERE) under conditions where hER-ERE complexes are readily formed with crude extracts from HeLa or yeast cells expressing the hER. This, too, indicates that other factors are necessary for ER-ERE complex formation (Mukheijee & Chambon, 1990).

1.2.4.D.h.i Dimerization interactions The classical mechanism for estrogen induced transcriptional activation (discussed in section I.2.D.f) supports ER homodimerization as the primary mode of activation of the estrogen receptor. The recent identification of ERP has complicated the view that ER functions solely as a homodimer. ERa and ERp can form functional heterodimers on EREs (Cowley et al., 1997; Pettersson et al., 1997). In addition, recent research has shown that ERa binds directly to a subset of nuclear receptors (including hepatocyte nuclear factor 4, ThR, RXR, RAR, and ER|3) through potential interactions between ligand-binding domains (Lee et al., 1998a). If these ER interactions are physiologically relevant, ER signaling may use novel signaling pathways that have remained uninvestigated until now.

1.2.4.D.h.ii Estrogen receptors utilize transcriptional cross-talk Estrogen receptor is not confined to members of the nuclear receptor family as interacting partners. Estrogen receptor can interact with the AP-1 transcription factor (composed of the proto-oncogene products c-Fos and c-Jun, discussed in section I.l.S.A.c) to regulate transcription fi*om an API DNA response element. The initial discovery came when the GR ligand-dependent repression of API on an API site was shown; in addition, API was shown to repress GR activity (Jonat et al., 1990; Schule et al., 1990; Yang-Yen et al., 1990). Later, the parallel situation was demonstrated for ER (Doucas, Spyrou & Yaniv, 1991). Interestingly, ERa and ERP function differently in a ligand-dependent manner on

55 the AP-1 site indicating that ERa and ER|3 have different roles in gene regulation via the AP-1 complex (Paech et aL, 1997).

1.2.4.D,h.iii Interactions of the ER with basal transcriptional machinery It is generally assumed that transcriptional activators and repressors act by either facilitating or interfering with the formation of the transcriptional pre-initiation complex. Supporting this idea, it has been shown that several steroid receptors directly interact with members of the basal transcriptional machinery. The rate-limiting step in the assembly of these basal factors into the pre-initiation complex is the binding of the TATA binding protein (TBP) to the initiation start site (the TATA box). It logically follows that proteins interacting with the components of the pre-initiation complex could strongly affect the activation or repression of transcription. The first interaction of nuclear receptors with members of the PIC was demonstrated in 1987 by Tsai and colleagues, showing that COUP-TF an orphan receptor interacts with TFIIB (Tsai et al, 1987). Since that time, PR, ER, TR, and RAR have all been shown to interact with TFIIB (Ing et a/., 1992; Lazennec et al, 1997; Sabbah et al, 1998). The ligand-independent interaction of the ER LBD with TBP was demonstrated in 1995 (Sadovsky et aL, 1995). Finally, Ing and colleagues demonstrated an interaction between ER and the TFIID associated factor TAFiiSO (Ing et aL, 1992), and interestingly, this interaction occurs in an region in ER that has a transactivation function in yeast cells (Pierrat et aL, 1992).

1.2.4.D.h.iv Interactions of the ER with transcriptional intermediary factors (TIFs)

• Coactivators

In 1989, Meyer and colleagues showed that the transcription of reporter genes by the progesterone receptor was inhibited by the coexpression of the estrogen receptor in both an ER and estrogen dose-dependent manner. Moreover, this inhibition was antagonised by the addition of antiestrogens (Meyer et aL, 1989). The authors suggested that this ligand-dependent squelching was the result of the two nuclear receptors competing for a limited amount of transcriptional intermediary factor. Recently there has been an intense

56 search to identify coregulatory proteins capable of enhancing estrogen-dependent transcription. In 1994 Halamichi and colleagues used a purified ligand-bound ER LBD to search for interacting proteins in MCF7 cells labelled with (Halachmi et aL, 1994). Two proteins were identified using this novel technique, a 140 kDa estrogen receptor associated protein (ERAP-140), and a 160 kDa estrogen receptor associated protein (ERAP-160). These experiments did not demonstrate transcriptional activation by these receptor-associated proteins, but several lines of evidence suggested that they played a role in ER function. The interaction between ER and these associated proteins was ligand-dependent and the interaction was disrupted by the estrogen antagonist 4-hydroxy tamoxifen and the pure antiestrogen ICI 164384. Finally, transcriptionally defective mutants of the estrogen receptor failed to interact with these proteins (Halachmi et aL, 1994). ERAP-160 was later cloned as SRC-1 by Onate and colleagues (Onate et aL, 1995). Onate and colleagues in 1995 identified a protein they named steroid receptor coactovator-1 (SRC-1) using the yeast two-hybrid system, with the PR as bait (Onate et aL, 1995). Importantly, this group showed that overexpression of SRC-1 reversed the ability of the estrogen receptor to squelch activation by hPR. Subsequently, SRC-1 has been shown to interact with and enhance the transcriptional activity of several nuclear receptors including ER. SRC-1 is a member of a large family of 160 kDa coactivator proteins which will be thoroughly discussed in the next section of this introduction.

• Corepressors

NCoR and SMRT were first identified as transcriptional silencers which constitutively bound to DNA-bound class II nuclear receptors (RAR, RXR, ThR), and prevented transcription until appropriate ligand was bound by the receptor (Chen & Evans, 1995; Horlein et aL, 1995; Kurokawa et aL, 1995). It was later shown that the corepressors could interact with the estrogen receptor (Jackson et aL, 1997). It has been proposed that these repressors may work via the recruitment of histone deacetylases which silence gene transcription by maintaining chromatin in a more condensed and less accessible state (Alland et aL, 1997; Heinzel et aL, 1997; Nagy et aL, 1997).

57 • Integrators

The cyclic AMP response element-binding protein (CREB) has an associated protein named CREB-binding protein (CBP) (Kwok et aL, 1994). CBP is essential for the activation of transcription by several transcription factors including the nuclear receptors. CBP is a coordinator of large multiprotein transcriptional complexes and is critical for the integration of signal transduction pathways. In addition to binding elements of the basal transcriptional machinery, RNA polymerase II (Kee, Arias & Montminy, 1996) and TFIIB (Kwok etal., 1994), CBP can also specifically bind the estrogen receptor (as well as other steroid receptor proteins) and in some cases enhance nuclear receptor transcriptional activity (Kamei et aL, 1996). CBP has been shown to interact with steroid receptors directly through the AF-2 domain in a ligand-dependent manner. Interestingly, CBP has also been shown to interact with the 160 kDa steroid receptor coactivators, as well (Smith et aL, 1996).

I.2.4.D.h.v Chromatin remodelling proteins

The members of the SWl/SNF complex were originally identified as positive regulators of transcription of the HO and SUC2 genes involved in yeast mating and yeast sucrose fermentation respectively (Neigebom & Carlson, 1984). Homologues of these proteins have since been identified in metazoans: the Drosophila Brahma complex and the human hBRM and BRG-1 complexes (Muchardt & Yaniv, 1999). Yamamoto and colleagues demonstrated a role for the SWl/SNF complex in steroid receptor mediated action in yeast with the glucocorticoid receptor. This work showed that mutations in the SWl/SNF complex disrupted GR fimction, and that a part of the SWl/SNF complex could be coimmunoprecipitated with the DNA-binding domain of GR (Yoshinaga et aL, 1992). Similar work followed in which the Drosophila Brahma complex and the human hBRG-1 proteins were shown to play a role in the transcriptional activity of GR, ER, and RAR (Chiba et aL, 1994; Muchardt & Yaniv, 1993). Several lines of investigation implicate these proteins in the remodelling of chromatin structure.

58 Mutations in histone genes in yeast alleviate the need for functional SWl/SNF genes in yeast (Hirschhom et aL, 1992; Winston & Carlson, 1992). In addition, SWl/SNF has intrinsic ATPase activity (Laurent, Treich & Carlson, 1993). Finally, a purified SWl/SNF alters nucleosomal structure in vitro in an ATP-dependent manner (Cote et aL, 1994; Imbalzano et aL, 1994; Kwon et aL, 1994). This evidence has led researchers to suggest that the dynamic remodelling of chromatin structure by the SWl/SNF to facilitate binding of transcription factors.

The next section of this introduction reviews a family of proteins which functionally interact with the estrogen receptor, which may also act as integrators for signal transduction pathways, and have chromatin remodelling histone acetyltransferase (HAT) activity. This is the p i60 steroid receptor coactivator family.

59 1.3 The P160/SRC co-activators

When nuclear receptor constructs were used as a bait in a series of biochemical and yeast hybrid experiments screening for proteins interacting with the nuclear receptor ligand binding domain, a family of related proteins collectively termed the pi 60 steroid receptor coactivators (the SRC family) were discovered. The three members of this family are encoded by three distinct genes and include: 1. SRC-l/NCoA, 2. SRC- 2/TIF2/GRIPl/NCoA2, and 3. SRC-3/pCIP/ACTR/AIBl. As well as sequence homology, the members of this family share the ability to stimulate ligand-dependant transactivation of several nuclear receptors.

GROUP NAME (REFERENCE) CLONED FROM INTERACTS WITH

SRC-1 mSRC-1 (Onate ef a/., 1995) mouse ER, GR, PR, TR, and NCoA-1 (Kamei etal., 1996) mouse RXR ERAP 160 (Halachmietal., 1994) mouse

SRC-2 GRIP1 (Hong et aL, 1997) mouse AR, ER, GR, and PR NCoA-2 (Torchia etal., 1997) mouse TIF2 (Voegel etal., 1996) human

SRC-3 SRC-3. (McKenna, Lanz & O'Malley, 1999a) mouse ER, PR, TR, p/CIP (Torchia etal., 1997) mouse RXR, and RAR RAGS (Li, Gomes & Chen, 1997) human ACTR (Chen etal., 1997) human AIB1 (Anzick eta/., 1997) human TRAM-1 (Takeshita eta/., 1997) human SRC-3 (Suen etal., 1998) human

Table 1.2: Members of the p160 steroid receptor coactivator family.

60 1.3.1 General structural and functional domains

1.3.1.AbHLH/PAS

The three SRC genes share the greatest degree of similarity in a region that contains a basic helix loop helix (bHLH) and a Per/Ant/Sim homology (PAS) domain (Figure 1.2). In general, bHLH/PAS domains mediate homo- and hetero- dimeric interactions between proteins that contain this domain (Hankinson, 1995; Swanson, Chan & Bradfield, 1995). The existence and conservation of this domain in the SRC family suggests that nuclear receptors may function through crosstalk between nuclear receptor mediated pathways and other bHLH/PAS factors (Kamei et aL, 1996).

1.3.1.B Serine/threonine and glutamine rich regions

All three SRC gene products share a serine/threonine rich area. In addition, a glutamine rich area is conserved at the C-terminal end of all three proteins (Figure 1.2, red box).

1.3.1.C The activation domains

In 1998, Onate and colleagues defined the activation domains of SRC-1. In their work, two regions capable of stimulating transcription in mammalian cells were identified: activation domain 1 (ADI), amino acids 1-93, and activation domain 2 (AD2), amino acids 840-948. Deletion of either ADI or AD2 resulted in a 50% reduction in transactivation ability, indicating that both domains are required for optimum coactivation. ADI includes the basic helix loop helix motif which is highly conserved among all SRC family members, while AD2 does not share homology with any known amino acid motif. Interestingly, these experiments demonstrated the intrinsic ability of SRC-1 fused to the DNA-binding domain of Gal4 to independently transactivate a GAL4-TATA reporter construct; these experiments were not carried out on steroid response elements. These results show that SRC-1 is a modular coactivator that possesses intrinsic activation ability when tethered to DNA. Moreover, the two activation

61 domains are both required for maximum coactivation function with steroid receptors (Onate et aL, 1998).

1.3.l.D The steroid receptor binding domains

I.3.1.D.a The LXXLL motif

The SRC family members interact with nuclear receptors in a ligand-dependent manner by means of an amphipathic helix which contains an LXXLL motif (where “L” is leucine and “X” is any amino acid) (Heery et aL, 1997). These LXXLL motifs are conserved in both sequence and spatial positioning in all three family members (Heery et aL, 1997; Torchia etal., 1997) (Figure 1.2). Interestingly, an additional LXXLL motif is present at the extreme C-terminus of the SRC-la splice form of SRC-1 (Figure 1.3, and discussed in section 1.3.2.A.a) (Kalkhoven et aL, 1998). The ability of SRC-1 to bind estrogen receptor and enhance transcriptional activity in a ligand-dependent manner depends upon the LXXLL motifs’ ability to interact with specific hydrophobic residues of helix 12 in the AF-2 in the LBD of the estrogen receptor (Heeiy et aL, 1997). Crystal structure studies have shown that LXXLL motifs form a two-tum amphipathic helix which binds in a hydrophobic pocket composed of helices 3, 4, 5, and 12 of the AF-2 of the nuclear receptor. Additionally, a conserved lysine in helix 3 and a glutamate in helix 12 form a “charged-clamp” which stabilizes the LXXLL motif by interactions at each end (Darimont et aL, 1998; Nolte et aL, 1998; Shiau et aL, 1998; Westin et aL, 1998). Recently, Needham and colleagues demonstrated that different LXXLL motifs have preferential affinity for particular nuclear receptors. They showed in a series of yeast two hybrid experiments the second LXXLL motif of SRC-1 preferentially interacts with ERa and ERp, while the fourth LXXLL motif (interestingly only found in the SRC-la isoform) preferentially interacts with peroxisome proliferator activated receptors alpha and gamma (PPARa and y), AR, and GR. These interactions depend not only on the LXXLL motifs, but also on residues 77-terminal to the motif (Needham et aL, 2000).

62 I.3.1.D.b SRBl and SRB2

Two steroid receptor-binding domains (SRBl and SRB2) were identified in a series of affinity chromatography pull-down, mammalian two-hybrid, and transient transfection experiments (Figure 1.2). SRBl, encoded by amino acids 361-1139, interacts with the steroid receptor AF-1 in the A/B region in a ligand-independent manner. In vitro^ amino acids 361 - 633 (the area containing the three LXXLL motifs), but not amino acids 782 - 1139, bind this N-terminal region of the steroid receptor. Although the region encoded by amino acids 782 - 1139 fails to interact in vitro, this region is active in mammalian two-hybrid systems, indicating that this region of SRC-1 requires additional cellular components for interaction with the A/B region of the steroid receptor. SRBl also interacted with the AF-2 motif in the D/E region of the steroid receptor but with only 20% of the efficiency of that seen in the A/B region (Onate et aL, 1998). SRB2 is largely responsible for binding to the steroid receptor in the D/E region, while no interaction was seen in the A/B region. This SRB2 region, therefore is proposed to provide specificity for the ligand-bound AF-2.

63 NLS HLH PAS NID SRBl SRB2

1441 III I SRC-1

1464 I ■ I m n SRC-2

1417 I MBIii IDD [ZH SRC-3

Figure 1.2: Comparison of the structural features of selected members of the steroid receptor coactivator family. The sequences have been aligned with reference to the A^-terminus. Gaps, represented by have been included where necessary to align more C-terminal common functional elements including the fourth LXXLL motif, the red box indicating a glutamine-rich (Q- rich) conserved region, and the relative endpoint of each protein. The figures above the protein representations indicate the number of amino acids with number 1 being the most jV-terminal. The approximate location of common functional elements is noted on the topmost line (NLS = nuclear localization signal, HLH = helix loop helix domain, PAS = Per/Ant/Sim domain, NID = nuclear receptor interaction domain, SRBl = steroid receptor binding region 1, SRB2 = steroid receptor binding region 2 ,1 = LXXLL motif).

64 1.3.2 There are three SRC genes

L3.2.A SRC-1

In 1995, Onate and colleagues, using a yeast two-hybrid screen with the ligand-binding domain of progesterone receptor as bait, identified steroid receptor coactivator-1 (SRC-1) as a coregulator that enhances human PR inducible transcriptional activity without affecting basal promoter activity. The C-terminal region of SRC-1 interacted with the LBD of PR in the presence of progesterone, but not in the presence of the antagonist RU- 486 (Onate et ah, 1995). In further studies, it was shown that SRC-1 enhances the transcriptional activity not only of the PR, but also of ER, GR, ThR, RXR, and RAR. Furthermore, it was shown that SRC-1 was not capable of transactivating other transcription factors including , E47 and CREB, indicating that it was not a general coactivator for all classes of transactivator (Onate et aL, 1995). As mentioned earlier, steroid receptors have the ability to squelch the activity of one another if one is overexpressed. Titrated overexpression of SRC-1 reverses this effect in a dose-dependent manner, suggesting that SRC-1 is a limiting coregulator (Onate et aL, 1995).

I.3.2.A.a Isoforms of the SRC-1 proteins

Kamei and colleagues in 1996 isolated two clones of SRC-1 identical up to amino acid 1385 and differing at their carboxy termini, SRC-la and SRC-le (Figure 1.3). The splice variants are generated via alternative intron/exon junctions. SRC-la is a 1441 amino acid protein containing 56 unique residues including an extra LXXLL motif at amino acids 1435 - 1439, and lacking the 14 most C-terminal residues present in SRC-le, a 1399 amino acid protein (Kamei et aL, 1996). The differences in these proteins exist in the area defined by Onate and colleagues as SRB2, responsible for the ligand-dependent interaction of SRC-1 with the steroid receptor AF-2 motif. Unsurprisingly, the two isoforms differ in their ability to interact with the AF-2 motif. The presence of the extra

65 LXXLL motif in SRC-la appears to stabilize the interaction of SRB2 with the AF-2 of the ERa, as increasing salt concentrations in GST pull-down experiments had no effect on SRC-la-GST/BRaAF-2 interactions; whereas, SRC-1 e-GST/ERaAF2 interactions were abolished at salt concentrations greater than 500mM (Kalkhoven et a/., 1998). Moreover, mutation of the fourth LXXLL motif, found only in SRC-la, to LXXAA resulted in a SRC-la with the salt-dependent AF-2 binding characteristics of SRC-le (Kalkhoven et ah, 1998). The two isoforms differ in their ability to potentiate ER mediated transcription. On promoters containing a single ERE, both SRC-la and SRC-le activated the promoter with SRC-le being the more potent activator. In experiments containing three EREs or the rat oxytocin promoter, which contains a natural ERE, SRC-la had a slightly inhibitory effect, while SRC-le continued to activate. Similar results were obtained in both Cos-1 cells and HeLa cells, with the maximal enhancement of ER activity by SRC- le being slightly less in the HeLa cells. Interestingly, mutation of the SRC-la-specific LXXLL motif to LXXAA did not affect the transcriptional properties of SRC-la. Moreover, the C-terminal 56 amino acids unique to SRC-la are probably responsible for the suppression of activity by SRC-la as a construct with these amino acids deleted stimulated ER mediated transcription in a manner similar to that of SRC-le (Kalkhoven etal., 1998).

66 NLS HLH PAS NID SRBl SRB2

1385 1441

SRC-la

m 1385 1399

SRC-le

Figure 1.3: Comparison of the SRC-1 a and SRC-le splice forms of SRC-1. The sequences have been aligned with reference to the A^-terminus. The figures above the protein representations indicate the number of amino acids with number 1 being the most TV-terminal. The approximate location of common functional elements is noted on the topmost line (NLS = nuclear localization signal, HLH = helix loop helix domain, PAS = Per/Ant/Sim domain, NID = nuclear receptor interaction domain, SRBl = steroid receptor binding region 1, SRB2 = steroid receptor binding region 2, I = LXXLL motif, red box = glutamine-rich (Q-rich) conserved region). The divergent areas of the proteins have been enlarged. The SRC-la form contains 56 unique amino acids (1385-1441) including an extra LXXLL motif (1435-1439). The SRC-le form contains 14 amino acids (1385-1399) not present at the C-terminus of SRC-la.

67 1.3.2.A.b Mechanism of action

After ligand-binding by the estrogen receptor, SRC-1 promotes the integration of N- and C-terminal steroid receptor AF-1 and AF-2 functions, allowing for full ERa activation (Mclnemey & Katzenellenbogen, 1996). In HepG2 and HeLa cells in which 4-OH tamoxifen can act as an estrogen agonist, SRC-1 enhanced both E2 and 4-OH tamoxifen stimulated transactivations in a dose-dependent manner (Smith, Nawaz & O'Malley, 1997a). Unsurprisingly, as mentioned earlier, SRC-1 has been shown to interact directly with the ER ligand-independent AF-1 domain which is the area responsible for mediating the agonist activity of 4-OH tamoxifen. (Webb et aL, 1998). The SRC-1/ERa interaction is dependent upon the LXXLL motifs in SRC-1 and critical residues in helix twelve of the ERa LBD (Heery et aL, 1997). Mutation of lysine366 in helix twelve of the ERa LBD does not affect ligand or DNA binding, but results in a marked decrease in the AF-2 activity and ability to bind SRC-1. This mutation, however had no effect on the ability of the steroid receptor to bind the RIP 140 coactivator (Henttu, Kalkhoven & Parker, 1997). These results indicate that lysinc366 plays an essential role in ERa/SRC-1 interactions, and that different nuclear receptor coactivators interact with different AF-2 amino acids.

1.3.2.A.C Interaction with ERP

ERp interacts with SRC-1 in a manner similar to that of ERa. In addition, SRC-1 is able to enhance the E2-dependent transactivations of an ERa/ERP heterodimer in transiently transfected Cos-1 cells (Tremblay et aL, 1997).

1.3.2.A.d Interactions with other proteins

68 1.3.2.A.d.i Interaction with CBP SRC-1 interacts with CBP and p300, which both interact with nuclear receptors in an AF- 2 dependent manner. SRC-1 has been shown to interact with a conserved C-terminal region of CBP and p300 (Eckner et ah, 1996; Yao et aL, 1996). Thus, SRC-1 allows the nuclear receptor two independent ways to interact with CBP/p300: directly, or indirectly via contacts with SRC-1. Likewise, CBP/p300 can interact with the nuclear receptor through the same direct or indirect mechanisms. If SRC-1 and CBP/p300 can interact simultaneously with the nuclear receptor, then the ability to form multiple independent contacts could result in the cooperative formation of a SRC-1, CBP/p300 coactivator complex. Consistent with this idea, cotransfection of CBP and SRC-1 with PR and ER results in a synergistic rather than an additive transcriptional response (Smith et aL, 1996).

1.3.2.A.d.ii Interaction with gene specific transcription factors SRC-1 can also interact with other transcription factors including: AP-1 (Lee et aL, 1998b), SRF (Kim, Kim & Lee, 1998a), NFkB (Na et aL, 1998), CREB and STATs (Torchia et aL, 1997), and p53 (Lee et aL, 1999). Interestingly, it is becoming more and more clear that SRC-1 does not require nuclear receptors for these interactions. Moreover, in several cases, transactivations involving these transcription factors can occur on the cognate transcription factor DNA-binding site rather than via nuclear receptor response elements.

1.3.2.A.d.iii Interactions with basal transcription factors SRC-1 can also interact with basal transcription factors. Interactions with both TBP and TFIIB have been described (Ikeda et aL, 1999). The interaction of SRC-1 with both gene specific and basal transcription factors supports the idea that SRC-1 may act as a bridge between ligand-activated, response element-bound nuclear receptors in gene-specific transcription factor complexes and the RNA polymerase II initiation complex.

69 L3.2.A.d.iv Interactions with cell cycle proteins An interaction between SRC-1 and cyclin-Dl, a key regulator of the cell cycle, has also been shown (Zwijsen et aL, 1998). Cyclin-Dl also interacts with the LBD of ERa and enhances ligand-independent ERa-mediated transcription (Zwijsen et aL, 1997). Cyclin- Dl was shown to increase the affinity of ERa for GST-SRC-1 in the absense of ligand (Zwijsen et aL, 1997). The authors suggest that cyclin-Dl recruits SRC-1 to the unliganded ERa establishing a SRC-1 mediated transcriptional role for cyclin-Dl (Zwijsen et aL, 1998).

1.3.2.A.C SRC-1 knockout mice

SRC-1 knockout mice have been created and homozygous mutants were both viable and fertile (Xu et aL, 1998). These mice showed decreased growth and development in estrogen responsive uterine and mammary tissues when compared to normal mice (Xu et aL, 1998). Finally, the homozygous null mutants showed an increase in the nuclear receptor coactivator TIFl, possibly as a compensatory mechanism (Xu et aL, 1998).

1.3.2.A.f Role in cancer

Jackson and colleagues proposed that alterations in the expression of coactivator and corepressor proteins play a role in tamoxifen resistance in breast cancer (Jackson et aL, 1997). Recently, semi-quantitative RT-PCR was used to screen for SRC-1 levels in primary breast tumor samples, and in cell lines (Bems et aL, 1998). Normal breast tissue contained the highest levels of SRC-1, lower levels were seen in tumor samples fi-om patients who responded to tamoxifen therapy, still lower levels were seen in tumor samples fi-om patients who did not respond to tamoxifen therapy, and the lowest levels of SRC-1 were detected in breast tumor cell lines. SRC-1 levels, however, did not correlate with ER status (Bems et aL, 1998).

1.3.2.B SRC-2

70 SRC-2 is equivalent to the proteins originally identified in mice as GRIPl (Hong et aL, 1997), and in HeLa extracts as TIF2 (Voegel et aL, 1996).

1.3.2.B.a Mechanism of action

SRC-2 interacts with ERa in the presence of E2, but not in the presence of 4-OH tamoxifen in GST pull-down, yeast two-hybrid, and transient transfection experiments in HeLa cells using a vitellogenin ERE reporter construct (Voegel et aL, 1996). In mammalian two-hybrid assays, SRC-2 was shown to interact with the AF-1 of ERa (Webb et aL, 1998; Webb et aL, 1999). In addition, like SRC-1, the AF-1 activity of tamoxifen-liganded ERa was stimulated by SRC-2 (Webb et aL, 1998; Webb et aL, 1999). Crystal structures of the ERa LBD and the second LXXLL motif of SRC-2 show that, like SRC-1, SRC-2 interacts with the hydrophobic pocket formed by helices 3, 4, 5, and 12 in the nuclear receptor LBD. In addition, this group also reviewed the structure of the 4-OH tamoxifen-bound ERa LBD showing that in this situation, helix 12 is bound to the hydrophobic coactivator groove where the SRC-2 LXXLL motif sat in the E2-bound structure (Shiau et aL, 1998).

1.3.2.B.b Interactions with ERP

In GST pull-down experiments, SRC-2 bound the AF-2 regions of ERa and ERP in a strictly ligand-dependent manner (Cowley & Parker, 1999). Unlike ERa, the A/B domain of ERp, containing the AF-1 region, does not interact with SRC-2 (Cowley & Parker, 1999). Transcriptionally SRC-2 has different effects on ERa and ERp. Full length ERP is only weakly transcriptionally activated by SRC-2, and ERP AF-1 activity is unaffected by the addition of SRC-2 (Webb et aL, 1998). The orphan nuclear receptor SHP can inhibit the transcriptional activity of ERa and ERP by binding the AF-2 region (Johansson et aL, 1999). In GST pull-down assays and transiently transfected cells, SHP competes with

71 SRC-2 to bind ERa AF-2 regions (Johansson et aL, 1999). These results suggest that SRC-1 interacts with ERa through both AF-1 and AF-2 regions, while SRC-2 interacts with ERJ3 only through contacts with the AF-2 region.

1.3.2.B.C Interactions with other proteins

It has been shown that the ERa LBD and full length ThR compete for binding both SRC- 1 and SRC-2 in vitro resulting in a limiting-coactivator-induced squelching of transcription. The overexpression of SRC-1 or SRC-2, however, fails to relieve the thyroid hormone- and E2- dependent squelching of TR transcriptional activity by the ERa LBD. Quite surprisingly, the transcriptional reporter activity was even further inhibited by the overexpression of either SRC-1 or SRC-2 (Lopez et aL, 1999). Because of this potentiated squelching by the SRC proteins, the authors suggested that an additional cofactor which recognized the SRC or the SRC-nuclear receptor complex, but not the nuclear receptor alone, was available only in limited concentrations. In the context of this particular experiment, the ERa LBD-SRC complex was proposed to sequester this second factor away from the DNA-bound TR-SRC complex. The additional overexpression of CBP, p300, pCAF, TFIIB, or TBP all of which interact with both SRC-1 and SRC-2 (and thus may have been the limiting reagent) each failed to relieve the inhibition of transcription. (Lopez et aL, 1999). The limiting cofactor remains unidentified.

1.3.2.C SRC-3

SRC-3 is equivalent to the proteins first identified as AIBl (Anzick et aL, 1997), ACTR {Chen etal., 1997), RAC3 (Li & Chen, 1998; Li et aL, 1997), TRAMl (Takeshita et aL, 1997), and p/CIP (Kim et aL, 1998b; Torchia et aL, 1997).

1.3.2.C.a Mechanism of action

72 SRC-3 functions similarly to SRC-1 and SRC-2 and has been shown to stimulate reporter gene expression from many nuclear receptors including ERa.

1.3.2.C.b Interaction with ERP

SRC-3 selectively enhances the transcriptional activity of ERa, but does not stimulate ERP-mediated transcription (Suen et al., 1998). This research represents the first indication that ERa and ERP utilize different coactivators. This differential effect could reflect the 60% difference in homology between the LBDs of the ER isoforms (McKenna etal., 1999b).

1.3.2.C.C Interactions with other proteins

SRC-3 has been identified as a substrate for the acetyltransferase activities of CBP and p300 both in vitro and in MCF7 cells (Chen et al., 1999a). In the experiments in MCF7 cells, acylation of three lysine residues immediately upstream of an LXXLL motif was stimulated by E2 treatment. Acylation of SRC-3 in vitro resulted in decreased affinity for ERa (Chenetal., 1999a).

1.3.2.C.d Role in cancers

The SRC-3 identified by Anzick and colleagues was overexpresssed in four of five ER positive breast cancer cell lines. In addition, higher expression of SRC-3 was demonstrated in 64% of primary breast tumors analyzed (Anzick et al, 1997). A more recent follow-up study including 1157 breast and 122 ovarian tumors showed amplification of SRC-3 in 4.8% of the breast cancers and 7.4% of the ovarian cancers. Moreover, in the breast tumors, SRC-3 amplification was correlated with ERa positivity, PR positivity, and tumor size (Bautista et al., 1998). Finally, a similar study reported SRC-3 to be amplified in six of nine pancreatic cancer cell lines studied (Ghadimi et al., 1999).

73 1.4 The POU family of transcription factors

1.4.1 History and description

In 1988, the simultaneous identification of three mammalian homeodomain-containing genes: Pit-1, Oct-1, and Oct-2, and the nematode gene unc-S6, and the subsequent comparison of their nucleotide and amino acid sequences identified a new family of transcription factors that shared a unique bipartite DNA-binding domain (DNA-bd). The DNA-bd of this family, the POU domain family (named after the first identified members: Pit-1, Oct-1, Oct-2 and unc-S6) contains a unique four-helix POU specific domain (POUs) of 70 - 80 amino acids, linked to another DNA-bd of 60 amino acids that shares striking similarity with the DNA-binding homeodomain of homeotic genes (POUh). Both the POUs and POUh have the ability to bind DNA response elements with the POUs conferring high-affinity, site-specific DNA-binding, and the POUh portion displaying weaker binding to A/T-rich sequences.

1.4.2 The POU domain

Biochemical and structural data indicate that the POUs and the POUh are structurally independent, however, both subdomains are always found together and are indispensable for sequence-specific DNA binding (Klemm et aL, 1994). Thus, the POU domain can be considered as a single functional unit (Botfield, Jancso & Weiss, 1992; Wegner, Drolet & Rosenfeld, 1993). (Figure 1.4)

I.4.2.A The POU-specific domain

The POUs domain consists of four alpha helices surrounding a hydrophobic core similar to the HTH DNA binding domains of bacteriophage X and 434 repressors (Assa Munt et aL, 1993; Dekker et aL, 1993; Klemm et aL, 1994). Helix 3 is the DNA-recognition helix for the HTH motif, and the crystal structure of Oct-1 on its response element has

74 shown that this helix makes contacts with the 5’ half of the octamer site (5 ' ATGC 3 " ) (Klemm et al., 1994). (Figure 1.4)

I.4.2.B The POU-homeodomain

The POUh consists of three alpha helices and its structure is virtually superimposable on those of classic homeodomains. Helices 2 and 3 form a HTH motif which, in the Oct- 1/DNA structure, make contact with the 3’ half of the octamer site (5 ' AAAT 3 ') on the opposite side of the DNA double helix. In this structure, helix 3 makes several contacts in the major groove of the DNA (Klemm et ah, 1994). (Figure 1.4)

Ï.4.2.C The linker region

Even though there are no protein/protein interactions between the POUs and the POUh, both subdomains may bind to the DNA in a cooperative manner, probably due to overlapping DNA contacts near the center of the response element (Klemm et al, 1994). The linker region (Figure 1.4), then, must serve as a flexible tether which facilitates complex formation by maintaining a high local concentration of the subdomains (the binding of the two independent domains in a unimolecular rather than a bimolecular reaction). Moreover, the covalent connection of the two subdomains increases the DNA sequence specificity, as binding of the composite domain is most efficient when both subdomains bind the DNA.

75 a helix 1 a helix 2 a helix 3 a helix 4 a helix 1 a helix 2 a helix 3 D—it— ^ FA K R G TQ G GSQ I RFE L L L W RT L F P L V RVWFCNRQ

POU-specific domain ______POU-homeodomain 75-82 amino acids 60 amino acids high affinity DNA-binding low affinity DNA-binding site specific binding relaxed specificity

Figure 1.4: The secondary structural elements of the POU domain. The secondary structural elements of the POU domain are shown with conserved amino acids directly below. The POU-specific domain consists of four a-helices (cylinders) two of which include a region of basic amino acids (indicated by BHI ). The POU- homeodomain consists of three a-helices. Helix 1 is preceded by a cluster of basic amino acids; and helix three is believed to contact DNA in the major groove. The POU-specific and POU-homeodomain are separated by a linker that varies in size (14-25 amino acids) among members of the family. (Adapted from Wegner 1993)

76 1.4.3 Family members

Comparison of the original four POU domain-containing proteins with those subsequently identified has led to the subdivision of this family into six distinct classes: I - VI (Table 1.3). The original subdivision of this group was based on similarities within the POUs and linker region, however, additional features are conserved within these classes. Classes I and III are expressed during the development of the neuroendocrine system, and class IV during the development of the central nervous and sensorineuronal system. The class III POU domain proteins share a core region of amino acid identity in which various homo- and poly- meric stretches of residues have been inserted as a consequence of evolutionary divergence. Similarly, the class IV genes share a related core of identity V-terminal to the POU domain with various homo- and poly- meric stretches of residues inserted (Veenstra, van der Vliet & Destree, 1997). Pit-1, the only class I member identified to date, contains several coding exons and several splicing variants have been identified. The class IV members contain an intron at a conserved position in the POU IV box (a region with identity to the original nematode unc-S6 gene and with the mammalian gene n-myc). The class III genes, however are intronless and at least one member, Tst-1, contains a poly-adenosine stretch in its 3’ untranslated region, lending support to the idea that this class arose through retroviral mediated insertion of a copy of an ancestral POU domain gene (Veenstra et al., 1997). Class I, III, and IV are structurally similar in that they contain an N-terminal transactivation domain linked to the POU domain.

77 Table 1.3: The six classes of POU transcription factors. Facing page (Taken from Wegner et al 1993)

78 POU-domain proteins and their characteristics

Class lammaiiai1 Genomic Alternative Transactivation Expression Non-mammalian POU organization splice domain embryo adult POU

I Pit-1 Mouse 16 Pit-l/la Amino-terminai Neural tube, Pituitary pit-1 (GHF-1) 7 exons (GHF-1, (Ser-/Thr-rich) pituitary (salmon) GHF-2)

n Oct-1 Mouse chromosome 1 Oct-IA- Amino-terminai Ubiquitous Ubiquitous dOct-1 (OTF-1) (human chromosome 1) Oct-lC (Glu-rich) (pdm-1) (NF-Al) exon/intron (dPOU-19) (NFIII) structure dOct-2 (OBPlOO) (pdm-2) (dPOU-28)

Oct-2 Mouse Oct-2.1- Amino-terminal Neural tube, Lymphoid cells, (Drosophila ) (OTF-2) human chromosome 19 Oct-2.6 (Glu-rich) entire brain nervous system Oct-1 (N F-A2) 14 exons mini-Oct-2 Zarboxy-terminc except intestine, (chicken) (Ser-/Thr-rich) telencephalon testis, (Xenopus ) kidney

Skn-la/i Mouse chromosome 9 Skn-la Developing Epidermis exon/intron Skn-li epidermis (suprabasal cells, structure hair foiiicies)

III Brn-1 Mouse chromosome 1 Amino-terminal Developing central nervous zf-POUl intronless (Glu-rich) nervous system system, kidney (zebrafish) XPLOU-1 XPLOU-2 (Xenopus )

Brn-2 Mouse chromosome 4 Amino-terminai Developing CNS, Cfla N-0ct3 intronless (Giu-rich) nervous system glioblastoma, (Drosophila ) N-0ct5 neuroblastoma Ceh-6 (C. elegans )

Brn-4 Mouse chromosome X Amino-terminai Neural tube CNS (RHS2) intronless (forebrain) (N-0ct4)

Tstl Mouse chromosome 4 Amino-terminal Blastocyst Nervous system (SCIP) intronless (Giy-Aia-rich) ES ceils, (neurons, (Oct-6) EC ceils, brain myelinating glia), testis

Brn-3.0 Mouse chromosome 14 Developing Nervous system unc-86 (Brn3a) human chromosome 13) nervous system (retina, sensory (C. elegans ) (RDC-1) 3 exons (brainstem, spinal ganglia), spleen, l-POU cord, sensory leuro-epitheliomas tl-POU ganglia retina) Ewing's sarcomas (Drosophila )

Brn-3.1 Mouse chromosome 18 Developing 3 exons nervous system

Brn-3.2 Developing (Brn3b) nervous system

Oct-3/4 Mouse chromosome 17 Oct-3A Amino-terminai ES and EC ceils. Oocytes Oct-60 (Gets) (human chromosome 6) Oct-3B Pro-rich mbryonic ectoderm Oct-25 (NF-A3) 5 exons primordial germ ceils Oct-91 testis and ovary (Xenopus )

Emb Mouse chromosome 15 Developing Brain, kidney, pou[c] (Brn-5) 6 exons nervous system skeletal muscle, (zebrafish) testis

ES, Embryonic stem cell; EC, embryonic carcinoma ceil. 79 1.4.4 DNA response elements of the POU domains

The subdivision of the POU domain proteins into the six classes is useful not only for revealing similarities in genetic structure and expression, but also unsurprisingly, for revealing similarities in the cz^-acting DNA response elements recognized by the different classes. Class II proteins bind with high affinity to the canonical octamer site: ATCGAAAT. Other non-canonical sites may be recognised by the class II members, but with the expected commensurate decrease in binding affinity (Verrijzer et al., 1992; Verrijzer et al., 1991). Interestingly, recognition and binding of the octamer motif is not strictly limited to the POU domain-containing proteins; non-POU domain-containing proteins have been identified that also bind this response element (Bendall et al., 1993; van Leeuwen, Rensen & van der Vliet, 1997; Verrijzer et al., 1992; Verrijzer et al., 1991). The class I, III, and IV proteins bind sites distinctly different from the octamer site. The class I protein Pit-1 binds with maximum affinity to the consensus sequence ATGNATAWW ( where N = any base and W = A/T) (Ingraham et al., 1988; Nelson et al, 1988). Class III and Class IV bind with greatest affinity to sites with the consensus sequence CAT(N)nWAAT ( where N = any base, W = A/T, Class III n = 0, 2, or 3, Class IVn = 3) (U e ta l, 1993).

1.4.5 POU interactions with DNA response elements

As mentioned earlier, the two sub-domains of the POU domain, the POU specific (POUs) and the POU homeodomain (POUh) are separate structurally independent domains (Botfield et al., 1992; Botfield, Jancso & Weiss, 1994a; Botfield, Jancso & Weiss, 1994b; Botfield & Weiss, 1994). Both the POUs and the POUh domains contain helix-tum-helix DNA binding motifs which make sequence specific DNA contacts. This modular structure allows POU family members to recognise a diverse set of DNA response elements (Aurora & Herr, 1992; Cleary & Herr, 1995; Kemler et al., 1989). The precise structure of POU family transcription factors on their DNA response elements has been provided by x-ray crystallographic and nuclear magnetic resonance techniques. Several

80 structures have been solved: Pit-1 bound to DNA as a dimer (Jacobson et al, 1997), Pit-1 bound to DNA as a monomer (Jacobson et al., 1996), Oct-2 as a monomer (Sivaraja et al, 1994), Oct-1 bound to DNA as a monomer (Klemm et al., 1994), and Oct-1 POU- specific domain (Assa Munt et al., 1993; Dekker et al, 1993). The crystal structure of POU domain proteins, particularly the Pit-1 dimer on its response element (Jacobson et al., 1997) and Oct-1 on its DNA element (Klemm et al., 1994) have revealed in atomic detail the great flexibility allowed by the bipartite DNA binding domain (Table 1.4, Figure 1.4). Each member of each class of POU domain proteins show preferential binding to specific optimal sequences, however a sequence that is optimal for one POU domain protein does not exclude the possibility that the same sequence can be bound by a different POU domain protein. For example, the presence of a 5’ C residue in the canonical octamer sequence creates the preferred high-affinity binding site for both class I and Class III proteins:

5'- ATGCAAAT-3' Class I site

5'-C AT N N A A A T-3 ' Class II site

Ambiguous or promiscuous cfj-acting elements may have important implications in redundant and divergent functions resulting from coexpression of POU domain genes.

81 EXAMPLES OF DNA BINDING SITES FOR SOME POU PROTEINS Oct-1

H2B CTTATGCAAATGAG

Ad2 AATATGATAATGAG

OCTA+ ATGC TAATGATATTC

OCTA- GCGGTAATGATATAC Pit-1

rGH1 GC CATGAATAAATGATAGG

p3D ATGGTGAATAATAATGAAG

rGH2 CTGATGGATAATTTAGAAG

p1P TT CATGAATATATATATAA

p1D AAAATGCATTTTTTAATGC

Consensus ATGNATAWWT Unc-86

Consensus ATAATTAATNAG Brn-2

CRH-II TGCATAAATAATAGGC

CRH-IV GCGAT TAAT AAGC

CRH-V CACTTGGATAATAAGC Brn-3

Consensus ATAA(T/A)TAAT

Table 1.4: Examples of DNA response elements for POU transcription factors. The POU binding site is underlined. N = any nucleic acid, W = A or T (Adapted from Herr and Cleary, 1995)

82 1.4.6 Arrangement on response elements

It is clear now that the arrangement of the POU specific and the POU homeodomain on various cognate sites may differ in orientation, spacing and positioning of the subdomains on the response elements. Some of the DNA binding sites of POU family proteins are shown below. Arrangement of the POUs and the POUh , on the different sites may differ in orientation, spacing and positioning of the subdomains on the response element. (Figure 1.5)

83 ORITENTATION SPACING POSITIONING

TATGCAAATAA GC GCATAATTAAT

ATNmrTAATGCATNmrrAATGC

Oct-1 vs.Brn-2 Brn-2 Brn-3

Figure 1.5: Arrangement of the POUs andPOU h on DNA response elements. These domains may vary in orientation spacing and positioning on the DNA response element. The modular structure of the POU protein subdomains allows for this flexibility in binding and makes possible more programs of transcriptional control. (Adapted from Herr and Cleary 1995, Gruber et al 1997, and Ryan and Rosenfeld 1997)

84 1.4.7 Regulation of POU domain genes

Insights into the expression and regulation of POU domain genes elucidate the mechanisms by which cell-type specific tissues arise and are maintained. Ubiquitous and cell-type specific POU domain proteins exist, and it is becoming more and more clear that transcriptional regulation by these factors is not simply due to the expression (presence or absence) of these regulatory factors, but rather to the relative concentration, location, and modified states of these factors with respect to one another at specific times in specific tissues.

L4.7.A Nuclear translocation

Analysis of Pit-1 and Oct -6 (Sock et al., 1996; Theill et al., 1989) has shown that the nuclear translocation signal, the signal which allows these proteins to cross the nuclear membrane and gain access to DNA, is in the POUh. Veenstra and colleagues in 1996 showed the regulation of Oct-1 cellular localisation in Xenopus shifts to the nucleus around the mid-blastula transition, correlating well with the activation of the embryonic genome. Moreover, this group showed that the POU domain is responsible for this localization (Veenstra et al., 1997).

I.4.7.B Phosphorylation

Phosphorylation is a method of regulation common to several biochemical pathways. Phosphorylation provides a further control of POU domain's DNA-binding activity (Tanaka & Herr, 1990). The phosphorylation of Oct-1 during the cell cycle has been thoroughly studied (Roberts, Segil & Heintz, 1991; Segil, Roberts & Heintz, 1991). On the basis of charge differences, five different species of Oct-1 have been identified. In addition, proteolysis and phosphopeptide mapping indicates that Oct-1 may be phosphorylated at least ten different times throughout the cell cycle (Roberts et al., 1991).

85 During mitosis the Oct -1 POUh is phosphorylated at a serine residue and is thereby prevented from binding the H2B response element (Segil et al, 1991). Phosphorylation

of a threonine residue (homologous in position to the Oct -1 POUh serine residue mentioned earlier) in Pit-1 creates a drastic reduction in response element affinity (KapiloffgraA, 1991).

1.4.8 Expression of POU domain proteins

The class IV proteins show widespread specific expression in the developing sensorineuronal system, and likewise the class I and III proteins show widespread specific expression in the developing neuroendocrine system, and as the organism matures these proteins exhibit a pattern of progressive spatial restriction. By contrast, however, the expression of one member of the class II proteins, Oct-1, initially appears to show no spatial restriction. The ubiquitous expression and the presence of an octamer motif in the promoter of several housekeeping genes: histone 2B (Fletcher, Heintz & Boeder, 1987; Hinkley & Perry, 1992) and snRNA genes (Danzeiser, Urso & Kunkel, 1993; Mattaj et aL, 1985) suggested that Oct-1 is a constitutive housekeeping transcription factor. Analysis of the relative levels of Oct-1 expression in cell lines and tissues has shown that regulation of and by Oct-1 is not so simple. In an embryonic cell line Oct-1 expression may be upregulated 3-10 fold during treatment with IL -6 or retinoic acid to induce differentiation. Moreover this regulation occurs with a proportional downregulation of Oct-3 (Hsu & Chen-Kiang, 1993). In experiments studying the Xenopus embryo, Oct-1 levels may vary by 6 fold in different parts of the embryo (Veenstra et al., 1997). The relative concentrations of these factors whether designated ubiquitous or cell-type specific, play a major role in their ability to efficiently regulate transcription.

1.4.9 Protein/protein interactions

86 POU domain proteins bind several transcription factors including other POU domain proteins. These interactions may occur through the POU domain or other transactivating domains located elsewhere in the protein.

I.4.9.A Homo- and hetero- dimerization of POU domains

POU proteins have been shown to bind DNA as monomers and as homo- and hetro- dimers. Cooperative binding in the homo- and hetero- dimers act to increase transcriptional activity. Here again, the ability to form different functional units allows an even finer control for the regulation of transcription. (Table 1.5)

DIMERIZATION PROPERTIES OF POU DOMAIN PROTEINS

POU FORMPROMOTERREFERENCE Pit-1 homodimer (Ingraham etal., 1990)

Pit-1 monomer prolactin promoter (pi P) (Voss, Wilson & Rosenfeld, 1991) Oct-1 monomer prolactin promoter (p i P) (Voss ef a/., 1991) Pit-1/Oct-1 heterodimer prolactin promoter (pi P) (Voss ef a/., 1991)

Oct-1 homodimer Ig heavy chain (IgH) (LeBowitz et al., 1989) Oct-2 homodimer Ig heavy chain (IgH) (LeBowitz et a/., 1989) Oct-1/Oct-2 heterodimer Ig heavy chain (IgH) (Poellinger, Yoza & Roeder, 1989)

Table 1.5: Dimerization properties of some of the POU transcription factors.

87 I.4.9.B Interactions of POU family members with other transcription factors

In addition to their homo- and hetero- dimérisation properties, the POU proteins also interact with other proteins. POU domain interactions with other proteins may account for the fact that in some cases the POU domain alone is sufficient for the regulation of transcription, while in other cases the function of the protein depends on additional domains found in other areas of the protein (Katan et al., 1990). (Table 1.6)

INTERACTIONS OF POU DOMAINS WITH COFACTORS

POU COFACTOR DESCRIPTION INTERACTION REFERENCE

Oct-1 Spl transcription factor N.D. (Junker etal., 1990) Oct-1 PR/GR steroid receptor N.D. (Bruggemeier ef a/., 1991)

Unc-86 MEC3 activator POU h (Xue, Tu & Chalfie, 1993) Oct-1 VP-16 activator POU (Huang & Herr, 1996) Oct-3/4 El A coregulator N.D. (Scholer, Ciesiolka & Gruss, 1991) Oct-1 OCA-B coregulator POU (Luoetal., 1992) Oct-2 OCA-B coregulator POU (Luoetal., 1992) Pit-1 ER steroid receptor N.D. (Simmons etal., 1990) Brn-3 ER steroid receptor POU (Budhram Mahadeo, Parker & Latchman, 1998) Brn-3 p53 tumor supressor POU (Budhram Mahadeoetal., 1999a) Pit-1 TR steroid receptor N.D. (Schaufele, West & Baxter, 1992) Oct-1 PTF basal factor POU (Coenjaerts, van Oosterhout & van der Vliet, 1994) Oct-2 TBP basal factor N.D. (Arnosti et al., 1993)

Table 1.6: Interactions of some of the POU transcription factors with cofactors. Coregulatory proteins which interact with POU domain proteins are shown. The location of the interaction is indicated as well. N.D. indicates that the site of interaction on the POU domain has not yet been determined. (Adapted fi"om Herr and Cleary, 1995)

88 I.4.9.C Position 22 in the first alpha-helix of the POU homeodomain

The ability of the herpesvirus transactivator VP 16 to differentiate the seven amino acid difference between Oct-1 and Oct-2 allows for the specific heterodimerization with Oct-1 only (Cleary etal., 1993; Stem & Herr, 1991; Stem et al, 1989). Two (each in the first a-helix of the POUh) of the seven amino acid differences are recognised by VP 16 (Stem et al, 1989). The major determinant for the transactivator’s ability to differentiate between Oct-1 and Oct-2 is the glutamic acid at position 22 in the first a-helix of the Oct- 1 homeodomain. Replacement of the alanine found at position 22 in Oct-2 with the glutamic acid seen in Oct-1 is sufficient to promote Oct-2AT16 heterodimerization (Lai etal., 1992). A similar situation exists for the class IV POU domain proteins, Bm-3a and Bm-3b (discussed in section 1.4.10). The POU domain of Bm-3a generally acts as an activator of promoter constmcts bearing a Bm-3 site, while the POU domain of Bm-3b represses these same promoter constmcts. The interaction between OCT-1/2 and VP-16, and its dependence on the glutamic acid/alanine found at position 22 in the first a-helix of the homeodomain, is interesting in light of the fact that the amino acid in the first alpha helix in the POU domain proteins Bm-3a and Bm-3b is responsible for the activator and repressor activities of these proteins, respectively. A single amino acid change at position 22 in the POUh of Bm-3a’s valine to the corresponding isoleucine found in Bm- 3b changes the activity of Bm-3 a from an activator to a repressor on promoters in which the POU domain is sufficient for transcriptional activation. Moreover, the reciprocal change in Bm-3b changes Bm-3b to an activator (Figure 1.6) (Dawson, Morris & Latchman, 1996b; Dawson et al., 1998). Single amino acid changes have been well documented to effect loss of function changes in the transcriptional activity of regulatory proteins. However, this subtle change (both isoleucine and valine are hydrophobic amino acids with a steric difference of a single methyl group) and its drastic effects are unparalleled.

89 Neither the Oct-1/2 nor the Bm-3a/b mutations affect DNA binding activity. The crystal structures of other POU domains on their DNA targets reveal that the first a-helix of the

POUh is not involved in DNA contacts. In addition, gel shift experiments have shown that the isoleucine and valine mutants are capable of retarding the electromobility of DNA ft-agments containing a Bm-3 site. A deeper understanding of the way in which this activator-repressor/valine-isoleucine relationship functions is warranted.

90 POU-specifîc domain a helix 1 D a helix 2 D a helix 3 1 a helix 4 3D B rn -3a LEAFAERFKQRRIKLGVTQADVGSALANLKIPGVGSLSQSTICRFESLTLSHNNMIALKPILQAWLEEAE B r n - 3 b ------

linker

B rn -3a GAQREKMNKPELFNGG Brn-3b KSH LT ------A

POU-homeodomain

a helix 1 ahdix2 a helix 3

B rn-3 a EKKRKRTSIAAPEKRSLEAYFAVQPRPSSEKIAAIAEKLDLKKNWRVWFCNQRQKQK B r n - 3 b ------I ------

Figure 1.6: Comparison of the POU domains of Brn-3a and Brn-3b. The alpha-helical secondary structural elements of the POU domain are represented with cylinders. Identical amino acids are indicated with a The isoleucine at position 22 in the first a-helix of the Bm-3b POUh is in boldface type. (Adapted from Latchman 1999)

91 1.4.10 The Brn-3 family of POU domain containing proteins

1.4.10.A History

Degenerate oligonucleotides representing codons for amino acids conserved among POU domain proteins have been used to identify mRNAs encoding POU proteins in different tissue types (He et al., 1989; Latchman et at., 1992). The Bm-3 family of POU domain proteins was first identified in 1989 (He et al., 1989). Members of the Bm-3 family of transcription factors are expressed in neurons in both the central and peripheral nervous systems where they regulate neuronal differentiation (Gerrero et al., 1993; Lakin et al., 1995). Three different forms of Bm-3 have been found: Bm-3a (Bm-3.0), Bm-3b (Bm- 3.2), and Bm-3c (Bm-3.1) (Lillycrop et al., 1992; Tumer, Jenne & Rosenfeld, 1994; Xiang et al., 1993). (Figure 1.7)

a helix 1 a helix 2 a helix 3 a helix 4 a helix I a helix 2 a helix 3 Tw/-r3-r a FA K R G TQ G GSQ I RFE L L L W RT L F P L V RVWFCN RQ

5' TTCTAAAGTNAGNAGNATTAAATTNGG 3' 5

CA C C C GC G G GGT G C

forward primer reverse pnmer

Figure 1.7: Degenerate oligonucleotides used for the identification of POU domain containing proteins. Amplification of POU domain mRNAs using degenerate oligonucleotides containing all possible combinations encoding two highly conserved regions of the POU domain. (Adapted from Latchman 1996)

92 1.4.10.B Three different Brn-3 genes: Brn-3 a, Brn-3b and Brn-3c

Bm-3 a and Bm-3c, and Bm-3b appear to have antagonistic fimctions (discussed in section I.4.10.H) (Budhram Mahadeo et al., 1996; Milton et al., 1995). Generally, Bm- 3a and Bm-3c are able to activate promoters containing an octamer motif while Bm-3b inhibits the basal activity of these promoters as well as the activation by Bm-3 a (Morris et al., 1994). While Bm-3 a, Bm-3b, and Bm-3c show significant spatial overlap, they appear to be expressed at different points in the life of the cell, indicating fimctional roles at different developmental stages. (Figure 1.8)

1.4.10.C Alternative splicing of Brn-3a and Brn-3b

Bm-3 a and Bm-3b are encoded by two distinct genes located on separate chromosomes. Both of these genes are subject to altemative splicing. When the Bm-3 a and Bm-3b genes are transcribed firom a downstream promoter, they produce an intronless short form of the transcription factor. Likewise, when either of the genes is transcribed from an upstream promoter, a long RNA including a single intron is produced. Once this intron is removed, this longer RNA results in the translation of the long forms of Bm-3 a and Bm- 3b which contain additional A-terminal sequences (Theil et al., 1993). In the case of Bm- 3a, this W-terminal sequence is responsible for specific transcriptional activity (discussed in section I.4.10.I). Activation of the a-intemexin promoter for example, requires the TV- terminal domain of Bm-3a (Budhram Mahadeo et al., 1995b), while this domain is not necessary for the activation of test promoters containing Bm-3 sites (Morris et al., 1994). Thus the altemative splice forms of the Bm-3 proteins encode functionally distinct transcription factors (Liu, Dawson & Latchman, 1996c). (Figure 1.8)

93 I.4.10.D The POU IV box

The POU IV proteins share a region of significant homology, the POU IV box, N- terminal to the POU domain (Gerrero et al., 1993; Theil et al, 1993). The POU IV box contains about 40 amino acids and is well conserved in all members of the POU IV subfamily, including a 10 amino acid sequence entirely conserved between Bm-3 a and Bm-3c. The POU IV box bears DNA sequence homology to a conserved domain in the V-terminus of the cMyc family members (Collum et al., 1992). Thiel and colleagues suggested that this POU IV box, present in the long but not the short forms of Bm-3a, accounted for the ability of Bm-3a long but not Bm-3a short or Bm-3b short to tumourigenically transform primary fibroblasts cotransfected with Ha-ras (Theil et al., 1993).

94 20 30 40 bo 90 Brn-3a MMSMNSKQPHFAMHPTLPEHKYPSLHSSSEAIRRaclptpplqsnlfasidetllaraealaavdiavsqgK------shpfkpdatyht Brn-3b ------mcafylqlqsnifggIdasliaraealaavdivsqskSHHHHPPHhspfkpdatyht Brn-3c -MMAMNAKHRFGMHPVLQEPKFSSLHSGSEAMRRvclpapqlqgnifgsfdesllaraealaavdivshgk------nhpfkpdatyht 8888555333333311111111111111111111558 111111111111

100 110 120 130 140 150 160 170 180 190 Brn-3a mnsvpctststvplaH------hhhhhhhhqalEpgdlldhisspslalMAGAGGAGAAGGGGGAHdgpgggggpGGGGGPGGGGPGG B Brn-3b mntipctsaassssvPISHPSALAGTHHHHHHHhhhhhqphqal -egellehlspglalgAM------agpdgtvvsTPA------Brn-3c mssvpctstsptvpiSHPAALT------shphhavhqgl -egdllehispt 1 svsGL------gapehsvmpAQI------111333333333588 33333333333 3 3 3 3 3 3 3 3 3 3 5 5 8 8 8 888888888

200 210 220 230 240 250 260 270 280 290 Brn-3a GGGGGGPGGGGgapgggllgGSAHPHP hmhglghlshpAAAAAMNMPSGLPHPGLVAAAAHHGAAAAAAAAAAGQVAAASAAAAWGAAGLASICDSD Brn-3b ------haphmatmnPMHQAALSMAhahglpshmgcMSDVD------Brn-3c ------hphhlgamgHLHQAMGMS-hphavaphsamPACLSDVE------888888888 55555555588

Brn- 3a tdp releafaerfkqrrikl JVt jadvgsalanl tipgvgsl sqsticrfes Ltls hnnmialkpilqawleea s Brn-3b a dp rdleafaerfkqrrikl JVt jadvgsalanl tipgvgsl sqsticrfes Ltls hnnmialkpilqawleea s Brn- 3c sdp releafaerfkqrrikl JVt jadvgaalanl tipgvgsl sqsticrfes Ltls hnnmialkpvlqawleea i 1111111111111111111111111111111111111111111111111111111111111111111111111111

370 380 Brn-3a aqrekmnkpelfng Brn- 3b shrekltkpelfng Brn- 3c ayreknskpelfng 13111111111111

390 410 450 Brn-3a gekkrkrtsia apekrsleayfavq Drps sekiaaiaek Ldlk knvvrvwf cnqrqkqkr nkfsaty F Brn-3b aekkrkrtsia apekrsleayfaiq Drps sekiaaiaek Ldlk knvvrvwfcnqrqkqkk /kysagi Brn-3c serkrkrtsia apekrsleayfaia Drps sekiaaiaek Ldlk knvvrvwfcnqrqkqkr nkysavd 1111111111111111111111111111111111111111111111111111111111111133355

Figure 1.8: Sequence alignment comparing the common structural features of members of the Brn-3 family. The amino acid sequences of the murine Bm-3 a, Bm-3b, and Bm-3e sequences have been aligned from amino- to carboxy-termini using Match-box version 1.3. Amino acids are represented with their one letter code, and hyphens have been put in place as gaps. The numbering starts with the start methionine of Bm-3 a long. Aligned residues are printed in lowercase, while uppercase residues have not been aligned. Numbers below each aligned amino acid represent a score of 1 to 9, with lower scores indicating a high reliability of alignment. Underlined methionines are known start sites for the long and short forms of the proteins. Line A represents the A-terminal domains of the proteins (note: the entire A-terminus of Bm-3b has not been recorded in GenBank), including the POU IV box, covered by the dotted line. Line B marks the start of the short form of each protein; and, lines B and C together represent the central portions of the proteins. Line D represents the POU-specific domain, with each of the four alpha helices boxed. The linker region is shown in line E. Line F represents the POU-homeodomain, with each of the three alpha helices boxed.

95 L4.10.E Chromosomal localization of the Brn-3 proteins

Chromosomal assignments were made for murine Bm-3 members and have shown that each of the Bm-3 proteins are encoded by separate genes, that none of the Bm-3 family members are linked, and that these genes do not exist in gene clusters. These results indicate that each Bm-3 member is under its own independent regulation (Theil et al, 1994; Xia et al., 1993). Bm-3a, Bm-3b and Bm-3c are found on mouse chromosomes 14 (Xia etal., 1993), X (Theil etal., 1994), and 18 (Xia etal., 1993), respectively.

1.4.10.F Expression of the Brn-3 proteins

The expression pattem of Bm-3 in the developing and adult mouse has been studied using RNAse protection, Northem blot analysis, in situ hybridization, and RT-PCR, and the results suggest roles for Bm-3 in the developing sensorineuronal and neuroendocrine systems. (Table 1.3)

1.4.10.F.a Expression of Brn-3 in neuronal tissue

The Bm-3 proteins are highly similar to another class IV protein, unc-S6, whose functional deletion results in and ‘^/«coordinated” phenotype due to improper neuronal development in the nematode C. elegans. It is therefore unsurprising that that the Bm-3 proteins play a fundamental role in the growth and differentiation of cells in the nervous system (Gerrero et al., 1993). In the developing central nervous system, Bm-3 a, Bm-3b, and Bm-3c are expressed in the hindbrain and midbrain, but not in the forebrain (Tumer et al., 1994). In the adult brain, however, only Bm-3 a shows strong expression, and this expression is limited to the midbrain, hindbrain, and brainstem (Tumer et al., 1994; Xiang et al., 1995; Xiang et al., 1993). Bm-3 a and Bm-3b have been detected in the developing retina (Tumer et al, 1994), while the expression of all three Bm-3 proteins has been detected in the adult retina (Xiang et al, 1995). All three Bm-3 proteins have been detected in the embryonic and adult dorsal hom of the spinal cord.

96 1.4.10.F.b Expression of Brn-3 in neuronal cancers

1.4.10.F.b.i Expression in neuronal tissue Bm-3 a has been reported to be expressed in neuroepitheliomas and Ewings sarcomas (Collum et al., 1992). In addition Bm-3a has been shown to be overexpressed in aggressive neuroendocrine tumors (Leblond-Francillard etal, 1997). Bm-3 a and Bm-3b were shown to be reciprocally expressed in at least three human neuroblastoma cell lines, and are associated with the balance between proliferation and differentiation of the ND7 and other neuronal cell lines (Latchman, 1999). Specifically, while Bm-3a activates promoters associated with neuronal differentiation, Bm-3b not only represses these same promoters, but also is able to negate the stimulatory effects of Bm-3a on these promoters (Smith, Dawson & Latchman, 1997c). In addition, proliferating cell lines of neuroblastoma origin have a high ratio of Bm-3b to Bm-3 a. Likewise, when these cell lines are induced to differentiate, the levels of Bm-3b fall and the levels of Bm-3 a rise (Smith & Latchman, 1996). Finally, artificial over-expression of Bm-3a induces neurite outgrowth in the ND7 cell line, and artificial over-expression of Bm-3b inhibits neuronal process outgrowth and neuronal specific gene activation thus preventing the differentiation of these cells (Smith et al., 1997c). These results suggest that a role for Bm-3b may be to maintain these cells in a proliferative state.

1.4.10.F.c Expression of Brn-3 in non-neuronal tissue and cancers

Bm-3 a and Bm-3b have been detected in both normal and cancerous cervical epithelial cells (Ndisang, Budhram Mahadeo & Latchman, 1999). Bm-3a and Bm-3b have also been detected in the testis (Budhram Mahadeo, submitted). Bm-3 a and Bm-3b have been detected in the MCF7 human breast adenocarcinoma cell line (Budhram Mahadeo et al., 1998). Moreover, both Bm-3a and Bm-3b are expressed in primary breast tissue including: normal mammary cells, benign tumours, and malignant tumours. Bm-3b levels were previously studied in primary breast sample tumors by RT-PCR and Westem blot immunodetection. These experiments showed that the levels of Bm-3b, but not the

97 levels of Bm-3a, were enhanced in tumours compared to the levels in normal mammary cells. In addition, the levels of Bm-3b were elevated only in primary tumour samples with decreased levels of the breast cancer tumour suppressor gene BRCAI, indicating a role for Bm-3b in breast cancer.

1.4.10.G Brn-3 knockout mouse models Mice with a null mutation in both the maternal and paternal copies of individual Bm-3 have been prepared. The phenotypes correlate well with other Bm-3 expression data and the results reveal interesting information on the transcriptional regulation of and by Bm-3 proteins.

1.4.10.G.a The Brn-3a knockout The homozygous null Bm-3 a mutants are not viable. These animals die at postnatal day 1 (PI) as a result of starvation due to ««coordinated limb movement and impaired suckling ability. On the cellular level, these animals show a loss of neurons in the trigeminal ganglia, the medial habenula, the red nucleus, and the cervical but not the lumar dorsal root ganglia. The neurons of the retina however, appear normal (McEvilly et al., 1996). Interestingly, Bm-3b expression was reduced in the trigeminal and sensory ganglia, but not in the retinal ganglia, indicating that the expression of Bm-3 a has an effect, either direct or indirect, on the transcriptional regulation of Bm-3b.

I.4.10.G.b The Brn-3b knockout Unlike the Bm-3a knockout, the Bm-3b knockout is both viable and fertile. The null Bm-3b mutants also differ from their Bm-3 a counterparts, who had normal retinal development, in that these animals are blind. This blindness is the result of the loss of 60 - 70% of their retinal ganglia cells between postnatal days 17 and 20. The Bm-3b null animals have a different morphology in the cells of the superior colliculus, as well. Finally, there are reduced levels of Bm-3a in the retinal ganglia cells of the Bm-3b null mice (Erkman et al., 1996).

98 I.4.10.G.C The Brn-3c knockout The Bm-3c null mice are similar to the Bm-3b null mice in that they are both viable and fertile, and similar to the Bm-3 a null mice in that they display normal retinal development. These animals, however, are deaf and display a phenotype (wobbling and spinning) consistent with ataxic or vestibular defects. Unsurprisingly, these animals show a loss of cochlear and vestibular hair cells (Erkman et al., 1996).

I.4.10.H DNA binding by Class IV Brn-3 Proteins

Bm-3 a, Bm-3b and Bm-3c can bind to and regulate transcription from a subset of class III response elements (Li etal., 1993; Xiang et al., 1995). Xiang and colleagues in 1995 reported a consensus sequence for Class IV DNA response elements (Xiang et al., 1995). Dawson and colleagues in the next year showed that the class IV proteins rely not only upon the sequence but also upon the spacing and context of the specific response elements (Dawson et al., 1996a).

I.4.10.I Promoters regulated by Brn-3 proteins

Bm-3 proteins are capable of regulating transcription from a diverse set of promoters in different tissue types. Some of the transcriptional regulatory potential of these different class IV POU domain proteins can be attributed to the POU domain itself, while the amino terminus of the protein is also capable of transcriptional regulation. For example, in the case of Bm-3a, the short form of the protein, which lacks the V-terminal activation domain, is unable to activate the promoters of the genes encoding a-intemexin (Budhram Mahadeo etal., 1995b) and Bcl-2 (Smith et al, 1998a), while the long form of Bm-3 a is capable of activating these promoters. Several promoter studies have focused on the ability of Bm-3a and Bm-3b to regulate neuronal growth and differentiation. In general the POU domain of Bm-3 a activates promoters associated with neuronal differentiation such as the neurofilaments (Smith et

99 al, 1997d) and SNAP-25 (Lakin et al, 1995). Interestingly, not only is Bm-3b capable of repressing these promoters, but it is additionally able to negate the activation of Bm-3a on these promoters. Position 22 in the first a-helix of the POUh (discussed in section I.4.9.C) is of critical importance in this transcriptional antagonism. The Bm-3 response element has been mapped in several of the promoters studied, while in others, no Bm-3 site has been identified, indicating that the Bm-3 proteins may fimction indirectly through interactions with other transcription factors. Finally, it is now clear that the role of the Bm-3 protein as activator or repressor is dependent upon both the promoter and cell type. The activities of Bm-3 a and Bm-3b on several promoters is summarised in Table 1.7.

100 PROMOTER Bm-3a Brn-3b DOMAIN CITATION

a-internexin + A/-term (Budhram Mahadeo etal., 1995b)

SNAP-25 + - * POU (Lakin etal., 1995)

Pro-opiomelanocortin + . * (Gerrero etal., 1993)

HPV type 16 + - POU (Morris etal., 1994)

HPV type 18 + - POU (Morris etal., 1994)

Vitellogenein + POU (Budhram Mahadeo etal., 1998)

Neurofilaments + . * POU (Smith etal., 1997d)

Synapsin + + POU (Morris etal., 1996)

Bcl-2 + 0 A/-term (Smith et al., 1998b)

Bcl-x + 0 POU

BRCAI - - - N.D. (Budhram Mahadeo etal., 1999b)

Egr-1 + 0 POU (Smith et al., 1999)

Table 1.7: Promoter regulation by Brn-3 family members. This table shows some of the known promoters that are regulated by members of the Bm- 3 family. Activation of a promoter is indicated with a repression of a promoter is indicated with a no effect on the promoter is indicated with a “0.” An indicates that Bm-3b not only represses the promoter, but also negates activation by Bm-3a. “DOMAIN” indicates the Bm-3 domain responsible for the transcriptional effect.

101 I.4.10.J Brn-3 POU domain proteins play a role in estrogen dependent transcription

A number of POU domain transcription factors have been detected in the tissues of the reproductive tract (Jin et al, 1999). Tst-1, Sperm-1, Bm-5, and Oct-6 have been found in the testis, while Oct 3/4 has been detected on tissues of the ovary (Sock et al., 1996). Bm-3 a and Bm-3b have been detected in the tissues of the reproductive tract, as well (discussed in section I.4.10.F.c). Bm-3 a and Bm-3b have been detected in both normal and cancerous cervical epithelial cells (Ndisang et al., 1999). Bm-3 a and Bm-3b have also been detected in the testis (Budhram Mahadeo, submitted). As mentioned earlier, Bm-3a and Bm-3b have been detected in the MCF7 human breast adenocarcinoma cell line (Budhram Mahadeo et al., 1998). Moreover, both Bm-3 a and Bm-3b are expressed in primary breast tissues including normal mammary cells, benign tumours and malignant tumours. The levels of Bm-3b, but not the levels of Bm-3a, however, were enhanced in the tumours compared to the normal mammary cells (Budhram Mahadeo et al., 1999b).

Interestingly, Bm-3a and Bm-3b directly interact with the estrogen receptor proteins receptors (discussed in section I.2.4.D) which are highly expressed in these same tissues (Budhram Mahadeo et al., 1998). Indeed, it was demonstrated by Mukheijee and Chambon that the ER binds its DNA response element only in the presence of accessory proteins. The POU domains of Bm-3 a and Bm-3b were shown to interact with the DNA binding domain of the estrogen receptor in affinity chromatography pull-down assays (Budhram Mahadeo et al, 1998). This interaction was confirmed as physiologically relevant in coimmunoprecipitation experiments in homogenate from rat brain and rat ovary (Budhram Mahadeo et al., 1998). In addition, electromobility shift assays have shown that both Bm-3 a and Bm-3b increase the affinity of the estrogen receptor binding to the ERE (Budhram Mahadeo et al., 1998). Finally, Bm-3a and Bm-3b were shown to modulate the activity of promoters containing an estrogen response element: the vitillogenenin gene promoter and an artificial ERE containing promoter. In the presence of 17-P estradiol, Bm-3 a acts as a weak repressor and Bm-3b acts as a strong activator (BudhramMahadeo etal., 1998).

102 The ER has long been known to regulate the growth and differentiation of female reproductive organs (Lubahn et al, 1993; Muramatsu & Inoue, 2000), play a role in the progression of human breast cancers (Dickson et al., 1987; Dickson & Lippman, 1987; Elledge & Osborne, 1997; Fuqua, 1992; Fuqua et al, 1992; Lippman et al,, 1987), and consistently increase the estrogen-dependent proliferation of breast cancer cell lines which express it (Murphy & Sutherland, 1983; Reddel, Murphy & Sutherland, 1983; Reddel & Sutherland, 1983; Sutherland et al., 1983a; Sutherland, Hall & Taylor, 1983b; Taylor et al., 1983). The coexpression of the Bm-3 proteins and the estrogen receptor in these tissues and their transcriptionally functional relationship make them ideal candidates for the study of regulated gene expression.

103 1.5 Aims of the project

The Bm-3 proteins are potent transcriptional regulators which are able to functionally interact with and regulate transcription by the estrogen receptor. The p i60 steroid receptor coactivators, too, functionally interact with and potentiate the transcriptional activities of the estrogen receptor. Thus, the first set of aims for this project were to:

(1) determine if Bm-3 proteins physically interact with the p i60 steroid receptor coactivators.

(2) examine the transactivation properties of the Bm-3/SRC factors on both Bm-3 and estrogen responsive promoters in different cellular contexts.

Following the previous studies in this laboratory that focused on the role of Bm-3b in neuronal proliferation, the Bm-3/ER interaction, and the expression of Bm-3b in primary breast tumor samples; the observations from ( 1) and (2); and reports from the literature indicating that estrogen responsiveness plays an important role in the pathology of breast carcinomas, the next set of aims for this project were to:

(3) generate MCF7 estrogen responsive breast adenocarcinoma clones stably overexpressing a Bm-3b short constmct as well as clones stably expressing a Bm-3b antisense construct.

(4) examine the growth and endocrine characteristics of the Bm-3b altered clonal cell lines.

(5) examine the differential gene expression of the Bm-3b altered clonal cell lines.

104 II MATERIALS AND METHODS

II. 1 Materials

II. 1.1 Suppliers

Analytical grade general laboratory chemicals were obtained from Merck Ltd., (Poole, Dorset, U.K.), Boehringer Mannheim (Lewes, East Sussex, U.K.) or Sigma Chemical Company Ltd. (Poole, Dorset, U.K.). General disposable plasticware was supplied by Greiner (Stonehouse, Gloucester, U.K.) or Sterilin (Stone, Staffordshire, U.K.). Unless otherwise stated, additional laboratory materials were obtained from the following suppliers:

Amersham International pic. (Little Chalfont, Bucks., U.K.) a -“ P-dCTP (3000Ci/mM), ^^S-Methionine (>1000Ci/mM@10mCi/ml), ^H-Thymidine (400 Ci/mM), Rainbow'^’^ coloured protein molecular weight markers, Hybond^^-C and Hybond^'^-N^ membranes.

Qiagen (Crawley, West Sussex, U.K.) Qiagen “midi-prep” plasmid DNA extraction kits

Bio-Rad (Hemel Hempstead, Herts., U.K.) Ammonium persulphate, N,N'-methylene-bis-acrylamide, N,N,NTSf'-tetramethylethylene - diamine (TEMED)

Dako Ltd. (High Wycombe, Bucks., U.K.) All peroxidase conjugated secondary antibodies.

Difco Laboratories (Basingstoke, Hants., U.K.) Bacto-agar, Bacto-tryptone, Yeast extract.

105 Gelman Life Sciences (Ann Arbor, Michigan, U S A.) Disposable 0.45 and 0.2|Lim filters

Genosys Biotechnologies Inc. (Pampisford, Cambs., U.K.) RNA Isolator

Gibco-BRL Life Technologies Ltd. (Renfrewshire, Scotland, U.K.) 1 kb DNA ladder, all tissue culture solutions, media and supplements.

Nunc (Roskilde, Denmark) All tissue culture plasticware

Pharmacia Biotechnology Ltd. (St. Albans, U.K.) Hexanucleotides [pd(N) 6 ] for random prime labelling, dNTPs

Promega Corporation (Madison, Wisconsin, USA). All restriction and modifying enzymes and buffers.

Sigma Chemical Company Ltd. (Poole, Dorset, U.K.) Kodak X-OMAT imaging photographic film, Kodak Professional 64T colour film.

Whatman International Ltd. (Maidstone, Kent, U.K.) 3 MM chromatography paper

106 II. 1.2 Standard buffers and solutions

The following standard buffers and solutions were used throughout (all concentrations expressed at Ix):

TE: lOmM Tris-HCl pH 8.0 ImMEDTApHS.G

TAE: 400mM Tris base 200mM sodium acetate 20mM EDTA, pH 8.3

PBS: 137mMNaCl 2.7mM KCl

4.3mMNa2HP0^.7H20

1.4mM KH2PO4 SSC: ISOmMNaCl 15mM sodium citrate

Luria Bertani (LB) media: 1% (w/v) Bacto®-tryptone 1% (w/v) NaCl 0.5% Bacto®-yeast extract LB media was autoclaved at 120°C for 20 minutes at lOlb/square inch (psi)

Chloroform/isoamylalcohol: 96% (v/v) chloroform 4% (v/v) isoamyl alcohol

Tris equilibrated phenol: liquefied phenol equilibrated twice with excess O.IM Tris-HCl pH 8.0

107 II. 1.3 Description and sources oiE. coli strains

XLl-blue Genotype: 7 endAl gyrA96 thi-1 hsdR17supE44 relAl lac [F’proAB Lacf Z/M15, Tn 10 (Tet/] Source : Stratagene Ltd., Cambridge U.K.

BL21 (DE3) pLysS Genotype: F' ompT hsdSi^roQ) dcm^ Tet^ gai A.(DE3) endA [pLysS Cam'] Source : Stratagene Ltd., Cambridge U.K.

108 II. 1.4 Description and sources of plasmids

pGex 2TK Description: Prokaryotic expression vector under the control of the IPTG inducible tac promoter. Expresses GST protein for affinity chromatography purification. Resistance: Ampicillin Source/Reference: Amersham Pharmacia Biotech, U.K.

pGex 2TK Brn-3a long Description: Prokaryotic expression vector under the control of the IPTG inducible tac promoter. Expresses the long fi-om of Bm-3 a with a C- terminal GST protein tag for affinity chromatography purification. 2.3kb of the Bm-3a long cDNA is cloned in fi*ame into the BamHI site of the pGex2TK vector. Resistance: Ampicillin Source/Reference : Dr. T. Moroy, Phillips University, Marburg, Germany.

pGex 2TK Brn-3b short Description: Prokaryotic expression vector under the control of the IPTG inducible tac promoter. Expresses the short fi*om of Bm-3b with a C- terminal GST protein tag for affinity chromatography purification. Resistance: Ampicillin Source/Reference: Dr. T. Moroy, Phillips University, Marburg, Germany.

pGex 2TK Brn-3a POU Description: Prokaryotic expression vector under the control of the IPTG inducible tac promoter. Expresses the POU domain of Bm-3 a with a C- terminal GST protein tag for affinity chromatography purification. The

109 POU domain of Bm-3 a is cloned in frame into the BamHI/EcoRI sites of pGex2TK. Resistance: Ampicillin Source/Reference: Dr. S. Dawson, University College London, U.K.

pSG5 Srcla Description: Eukaryotic expression vector under the control of the early SV40 promoter for in vivo expression, and the T7 bacteriophage promoter for in vitro expression. Expresses the cDNA of Src-la cloned into the Smal/Bglll sites ofpSGS. Resistance: Ampicillin Source/Reference: Dr. M. Parker, Imperial Cancer Research Fund, London, U.K.

pSG5 Srcle Description: Eukaryotic expression vector under the control of the early SV40 promoter for in vivo expression, and the T7 bacteriophage promoter for in vitro expression. Expresses the cDNA of Src-le cloned into the Smal/Bglll sites of pSG5. Resistance: Ampicillin Source/Reference: Dr. M. Parker, Imperial Cancer Research Fund, London, U.K. pCDNA3.1 AIBl Description: Eukaryotic expression vector under the control of the CMV enhancer-promoter for in vivo expression, and the T7 bacteriophage promoter for in vitro expression. Expresses the 4841 bp cDNA of AIBl cloned into the Notl/Xbal sites of pCDNAS.l. Resistance: Ampicillin, Neomycin Source/Reference: Dr. P Meltzer, National Institute of Health, Bethesda, MD, U.S.A.

110 pLTR poly Description: Eukaryotic expression vector under the control of the MoMuLV promoter. Resistance: Ampicillin Source/Reference: Dr. T. Moroy, Phillips University, Marburg, Germany.

pLTR Brn-3a long Description: Eukaryotic expression vector under the control of the MoMuLV promoter for expression of the long form of Bm-3a. 3.5kb of the Bm-3 a long cDNA is cloned into the Sall/Nml site of the pLTR poly vector. Resistance: Ampicillin Source/Reference: Dr. T. Moroy, Phillips University, Marburg, Germany. pLTR Brn-3b short Description: Eukaryotic expression vector under the control of the MoMuLV promoter for expression of the short form of Bm-3b. 1.6kb of the Bm-3b short cDNA is cloned into the Sall/Hindlll site of the pLTR poly vector. Resistance: Ampicillin Source/Reference: Dr. T. Moroy, Phillips University, Marburg, Germany. pJ4 Brn-3b antisense Description: Eukaryotic expression vector under the control of the MoMuLV promoter for expression of the first 300 antisense base pairs of Bm-3b. The first 300 antisense bases of the Bm-3b short cDNA is cloned in reverse orientation into the Sall/BamHI site of the pJ4 vector. Resistance: Ampicillin Source/Reference : Dr. M Smith, University College London, U.K.

I ll pCAT Basic SNAP 25 Description: Eukaryotic promoter/reporter expression vector under the control of a 2.18 kb PstI - PstI (-2068/+114) fragment of the SNAP-25 containing both transcriptional start sites of the SNAP gene in pCAT- Basic expressing the bacterial CAT gene. Resistance: Ampicillin Source/Reference: Dr. M. Wilson, The Scripps Research Institute, LaJolla, CA, U.S.A.

pCAT Basic Vit-tk Description: Eukaryotic promoter/reporter expression vector under the control of bases -331/-87 of the Vit A2 promoter followed by -105/+5 of the herpes simplex virus thymidine kinase (tk) promoter expressing the bacterial CAT gene. Resistance: Ampicillin Source/Reference: Dr. M. Parker, Imperial Cancer Research Fund, London, U.K. pCi Neo Description: Eukaryotic expression vector which expresses the Neomycin phosphotransferase gene for selection of stably transfected cells. Resistance: Ampicillin, Neomycin Source/Reference: Promega Corporation, Madison, Wisconsin, U.S.A.

112 II. 1.5 Description and sources of cell lines

ND7 Description: Cells derived from rat dorsal root ganglion neurons immortalized by fusion with C l300 mouse neuroblastoma cells Culture media: Ix L15 10%(v/v) fetal calf serum 0.3%(v/v) D-glucose 0.37%(v/v) sodium bircarbonate 0.2mM L-glutamine Source/Reference: (Wood et al., 1990)

BHK21 clone 13 Description: Golden Syrian hamster kidney cells, Cells grown in monolayer and have a fibroblast morphology. Culture media: Ix DMEM 10%(v/v) fetal calf serum Source/Reference: (Macpherson & Stoker, 1962)

MCF7 Description: Human Caucasian breast adenocarcinoma cells. Cells grown in monolayer and have an epithelial morphology. Culture media: Ix DMEM 10%(v/v) fetal calf serum l%(v/v) non-essential amino acids

-or-

Ix DMEM (without phenol red) 10%(v/v) charcoal stripped fetal calf serum l%(v/v) non-essential amino acids

Source/Reference: (Brooks, Locke & Soule, 1973)

113 II.1.6 Description and sources of antibodies used

Brn-3a Immunogen: amino acids 186-224 Specificity: polyclonal Concentration/Dilution: crude rabbit antiserum used at 1:2000 Secondary : 1:2000 HRP-conjugated anti-rabbit Ig Source : BAbCo, Berkeley, CA, U.S.A.

Brn-3b Immunogen: human amino acids 184-252 Specificity: polyclonal Concentration/Dilution: crude rabbit antiserum used at 1:2000 Secondary : 1:2000 HRP-conjugated anti-rabbit Ig Source : BAbCo, Berkeley, CA U.S.A.

Bad (610391) Immunogen: human Bad Specificity: monoclonal Concentration/Dilution: 5pg of antibody used in coimmunoprecipitation Source : BD Transduction Laboratories

p53 (p53 Ab-1) Immunogen: human amino acids 371-380 Specificity: monoclonal Concentration/Dilution: 5p,g of antibody used in coimmunoprecipitation Source : Oncogene Research Products (Nottingham, U.K.)

Estrogen receptor alpha (SRA 1010)

114 Immunogen; human ERa amino acids 582-595 conjugated to KLH Specificity: monoclonal Concentration/Dilution: 3.5 mg/ml used at 1:2000 Secondary : 1:2000 HRP-conjugated anti-mouse Ig Source : StressGen, York, U.K.

Estrogen receptor alpha (SC 8005) Immunogen: Human ERa amino acids 120-170 Specificity: monoclonal Concentration/Dilution: 200pg/ml used at 1:2000 Secondary : 1:2000 HRP-conjugated anti-mouse Ig Source : SantaCruz Biotechnology, Santa Cruz, CA, U.S.A.

Human chorionic gonadotropin beta subunit Immunogen: beta subunit of human chorionic gonadotropin Specificity: polyclonal Concentration/Dilution: crude mouse serum used at 1:2000 Secondary : 1:2000 HRP-conjugated anti-mouse Ig Source : Dr. T. Lund, University College London, U.K.

Hsp90a (N 17) Immunogen: amino terminus of human Hsp90a Specificity: monoclonal Concentration/Dilution: 200|Xg/ml used at 1:2000 Secondary : 1:2000 HRP-conjugated anti-mouse Ig Source : SantaCruz Biotechnology, Santa Cruz, CA, U.S.A.

Hsp/Hsc70 (W 27) Immunogen: Specificity: monoclonal Concentration/Dilution: 200pg/ml used at 1:2000

115 Secondary : 1:2000 HRP-conjugated anti-mouse Ig Source : SantaCruz Biotechnology, Santa Cruz, CA, U.S.A.

Hsp25 (SPA 801) Immunogen: recombinant Hsp25 Specificity: polyclonal Concentration/Dilution: antiserum used at 1:2000 Secondary : 1:2000 HRP-conjugated anti-rabbit Ig Source : StressGen, York, U.K.

BRCAI (Abl) Immunogen: Specificity: monoclonal Concentration/Dilution: used at 1:2000 Secondary : 1:2000 HRP-conjugated anti-mouse Ig Source : CalBioChem, SanDiego, CA, U.S.A.

Actin (119) Immunogen: human actin carboxy-terminus Specificity: polyclonal Concentration/Dilution: used at 1:2000 Secondary : 1:2000 HRP-conjugated anti-goat Ig Source : SantaCruz Biotechnology, Santa Cruz, CA, U.S.A.

SRC-1 (MAI 840) Immunogen: GST:SRC-1 amino acids 477-947 Specificity: monoclonal

Concentration/Dilution: 1 mg/ml; used 5|Li 1 for coimmunoprecipitation Source : Affinity BioReagents, Golden, CO, U.S.A.

116 II. 1.7 Description of oligonucleotide primers used

Target: Polylinker of pJ4/5 Omega plasmids Genbank ascession number Primer name : pJ5PCR3 Sequence: AGA TCT GOT ACC ATC GAT Location from start (ATG=1): n/a Melting Temperature: Paired with/Product size : B3BAS R

Target: Bm-3b antisense strand Genbank ascession number S68377 Primer name : B3BAS R Sequence: TGG TGA TGG GIG CCA GCC A Location from start (ATG=I): 311 Melting Temperature: 69.51 Paired with/Product size : pJ5PCR3/320

Target: m/h Bm-3b Genbank ascession number S68377 Primer name : B3B271 F Sequence: GGT CTG CAT CCA CGT CGC TC Melting Temperature: 54.49 Paired with/Product size : B3B271 R /271

Target: m/h Bm-3b Genbank ascession number S68377 Primer name : B3B271 R Sequence: GGA GGG CGA GCT GCT TGA GC Melting Temperature: 56.48

117 Paired with/Product size: B3B271 F /271

Target: human Rho C Genbank ascession number L25081 Primer name : hrhocpSL Sequence: GCC TGG AAA ACA TTC CTG AG Location from start (ATG=1): 272 Melting Temperature: 59.67 Paired with/Product size : hrhocp3R/200

Target: human Rho C Genbank ascession number L25081 Primer name : hrhocp3R Sequence: AGG GTA GCC AAA GGC ACT GA Location from start (ATG=1): 471 Melting Temperature: 59.88 Paired with/Product size : hrhocp3L/200

Target: human early growth response 1 Genbank ascession number X52541 Primer name : hegrlp3L Sequence: CCG CAG AGT CTT TTC CTG AC Location from start (ATG=1): 302 Melting Temperature : 59.99 Paired with/Product size : hegrlp3R/200

Target: human early growth response 1 Genbank ascession number X52541 Primer name : hegrlp3R Sequence: TGG GTT GGT CAT GCT CAC TA Location from start (ATG=1): 501

118 Melting Temperature : 60.11 Paired with/Product size : hegrlp3L/200

Target: human junction plakoglobin Genbank ascession number M23410 Primer name : hplakp3L Sequence: AGC AAT AAG CCT GCC ATT GT Location from start (ATG=1): 1024 Melting Temperature: 59.74 Paired with/Product size: hplakp3R/195

Target: human junction plakoglobin Genbank ascession number M23410 Primer name : hplakp3R Sequence: GAC GTT GAC GTC ATC CAC AC Location from start (ATG=1): 1218 Melting Temperature: 60.01 Paired with/Product size: hplakp3L/195

Target: human cyclophilin A Genbank ascession number XM004890 Primer name : hcycp3L Sequence: AGG TCC CAA AGA CAG CAG AA Location from start (ATG=1): 83 Melting Temperature: 59.84 Paired with/Product size : hcycp3R/406

Target: human cyclophilin A Genbank ascession number XM004890 Primer name : hcycp3R Sequence: TGT CCA CAG TCA GCA ATG GT

119 Location from start (ATG=1): 488 Melting Temperature: 60.16 Paired with/Product size : hcycp3L/406

Target: rat p-actin (cross-reactive with human) Genbank ascession number V01217 Primer name : b-actinF Sequence: TCA TGA AGT GTG AGG TTG ACA TCC GT Location from start (ATG=1): 845 Melting Temperature: 68.88 Paired witb/Product size : b-actinR/279

Target: rat P-actin (cross-reactive with human) Genbank ascession number. V01217 Primer name : b-actinR Sequence: CCT AGA AGC ATT TGC GGT GCA CGA TG Location from start (ATG=1): 1123 Melting Temperature: 68.88 Paired witb/Product size : b-actinF/279

120 II.2 Methods

II.2.1 Bacterial Methods

11.2.1.A Propagation of bacteria

Bacteria were grown either in liquid LB media or on plates prepared from LB media containing 2% Bacto®-agar. Liquid cultures were grown overnight at 37°C in an orbital shaker at 200rpm. Bacterial plates were incubated overnight at 37°C in a standard incubator. Both media contained antibiotic selection as appropriate. Ampicillin was used at a final concentration of lOOjLig/ml. Tetracycline was used at a final concentration of 12.5|Lig/ml in plates and lOjig/ml in liquid culture.

11.2.1.B Long term storage of bacteria

For long-term storage, Exoli cells were grown as mentioned above to early log phase, OD580 of approximately 0.5. The bacteria were then pelleted by centrifugation at 3000rpm for 10 minutes and the LB was removed. The bacteria were resuspended in fresh LB medium with the appropriate antibiotic containing 30% (v/v) sterile glycerol. The culture was then distributed in 500|xl aliquots into autoclaved 1.5ml microcentrifuge tubes, which were immediately flash frozen in liquid nitrogen. Glycerol cell stocks were stored at -70°C until use.

IÏ.2.1.C Transformation of bacteria

Competent XL 1-Blue cells were prepared using a standard calcium chloride technique (Sambrook, Fritsch & Maniatis, 1989). A single bacterial colony was grown overnight in lOmls of LB containing the appropriate antibiotic. 100|il of this starter culture was used to inoculate lOOmls of LB containing no antibiotic and the culture was grown to an OD 580

121 of approximately 0.5. The bacteria were then pelleted by centrifugation at 3000rpm for 10 minutes and the LB was removed. The bacteria were resuspended in lOmls of ice-cold lOOmM CaCt, pelleted as before and resuspended in 4mls of ice cold CaCh. The competent cells were then incubated on ice until required and used not more than 72 hours after preparation. Competent cells were transformed by addition of DNA and subsequent incubation on ice for 30 minutes. The cells were then heat shocked for 90 seconds at 42°C and returned to ice for a further 2 minutes. 800|xl of LB was added and the cells incubated in an orbital shaker for 1 hour at 37°C/200rpm. The cells were then pelleted, resuspended in approximately lOOjuil of LB and plated onto LB agar plates containing the appropriate antibiotic selection.

122 II.2.2 DNA methods

11.2.2.A Small scale plasmid DNA extraction from transformed bacteria

The “mini-prep” extraction method used was based on a standard alkaline lysis protocol (Bimboim & Doly, 1979). Individual bacterial colonies were used to inoculate 3mls of LB containing the appropriate antibiotic selection. Cultures were incubated overnight at 37°C/200rpm. The cells from 1.5ml of culture were pelleted by centrifugation in a bench top microcentrifuge at 13000 rpm for 2 minutes and resuspended in 100|il of resuspension buffer (50mM Tris-HCl pH 7.5, lOmM EDTA pH8, lOOpg/ml RNase-A. Bacteria were then lysed by addition of 200pl of lysis buffer (200mM NaOH, 1% (v/v) Triton X-100) and neutralised by the addition of 150|xl of neutralisation buffer (3M sodium acetate pH5.5). The cell lysate was then centrifuged for 3 minutes at 13000rpm and the pelleted precipitate was removed and discarded. 500|il of isopropanol was then added to the supernatant which was then vortexed and centrifuged for 5 minutes at 13000 rpm to pellet the DNA The supernatant was removed and the DNA was washed with 500|il of 70% ethanol, dried under vacuum and then resuspended in lOOjil of double­ distilled water (ddfhO) containing 20|ig/ml RNase A. Plasmid DNA was stored at -20°C.

11.2.2.B Large scale plasmid DNA extraction from transformed bacteria

A single colony from a bacterial plate or approximately 50|il of liquid bacterial culture was used to inoculate 400mls of LB containing the appropriate antibiotic selection. This was then incubated overnight at 37°C at 200rpm. lOOmIs of this overnight culture were then spun down at 3000rpm for 10 minutes. Plasmid DNA was then extracted using the Qiagen midi-prep kit following the manufacturer’s instructions. A typical yield of DNA using this method was 100|ig, which was resuspended in lOOp.1 of ddH20.

123 11.2.2.C Restriction enzyme digestion

Restriction digests were performed for DNA analysis. Analytical digests were usually performed in 20|il total volume approximately 1 p,g DNA. Not more than 0.1 volumes of restriction enzyme(s) were added and the buffer recommended by the manufacturer was used at Ix concentration. Digests were incubated for at least 2 hours at the appropriate temperature. The digested DNA was electrophoresed on an agarose gel of appropriate percentage and bands visualized on a UV transilluminator. In general, a no enzyme control “mock” digest was prepared in parallel to detect changes that may have occurred independent of the enzyme and as reference against which to compare the enzyme cut samples.

ÏI.2.2.D Phenol/chloroform extraction and precipitation of DNA

To purify DNA the reaction mix was made up to 400|li1 with ddHiO and an equal volume of tris-equilibrated phenol was added. The mixture was vortexed, centrifuged at 13000 rpm for 2 minutes and the aqueous phase removed and re-extracted with one volume of chloroform/IAA. The aqueous phase was again removed and the DNA precipitated by addition of 0.1 volumes of 3M sodium acetate (pH5.5) and 2 volumes of -20°C 100% ethanol. Finally, the pellet was washed with 70% ethanol centrifuged at 13000 rpm for 2 minutes and the 70% ethanol removed. The pellet was then dried and resuspended in ddHiO.

11.2.2.E Agarose gel electrophoresis

Appropriate percentage agarose in Ix TAE gels were cast. Depending on the expected size of DNA fragments a 1.0% agarose gel was cast for fragments greater than Ikb, 1.5% agarose gel was cast for general purpose, and 2.0% agarose gels were cast to resolve DNA fragments less than Ikb. Ethidium bromide was added to a final concentration of 0.5|ig/ml. Approximately 0.1 volume of lOx loading buffer (Ix TAE, 50% v/v glycerol, 0.025% bromophenol blue) was added to DNA samples prior to loading. A DNA marker

124 appropriate to the size of the fragments being detected was used, usually a Gibco 1 kilobase (kb) ladder. DNA was electropboresed at approximately 100mA until the fragments were well separated (0.5 - 2 hours). Bands were visualised on a long wave UV transilluminator and photographed onto Polaroid film.

11.2.2.F Southern blot analysis

11.2.2.F.a Resolution of DNA on an agarose gel

Southern blots (Southern, 1975) were performed on RT-PCR amplified cDNA to confirm the expression of particular genes. 20|Li1 of a RT-PCR amplification reaction were directly loaded and electropboresed on a 2.0% agarose gel.

11.2.2.F.b Transfer of DNA to nitrocellulose

The DNA was visualised on a UV transilluminator and photographed. The gel was then placed in denaturing solution (1.5M NaCl, 0.5M NaOH) for 45minutes. The gel was then transferred to neutralising solution (2M NaCl, IM Tris pH5.5) for a further 45minutes. The gel was then placed upside down on a plastic support which was covered in a triple layer of 3MM Whatman paper which was used as a wick placed in a reservoir of 20x SSC. A piece of Hybond N nylon membrane cut to the same size as the gel was pre­ soaked in the neutralising solution and then carefully placed on the gel ensuring there were no air bubbles present. 10 pieces of 3MM Whatman paper, pre-soaked in 20x SSC, were placed on the nylon membrane and a stack of dry paper towels and a suitable weight were placed on top. Parafilm was used to isolate the wick from the paper towels. The DNA was then transferred by capillary action to the nylon membrane overnight. The membrane was removed, washed in 6x SSC and then air-dried for 30 minutes. The DNA was then cross-linked to the membrane using a UV Stratalinker 2400.

125 II.2.2.F.C Radiolabelling of DNA

Fragments of DNA were radiolabelled with a-[^^P]-dCTP for use as probes in Southern blot analysis. The method used was based on the random prime labelling reaction previously described (Feinberg & Vogelstein, 1983). Approximately Ipg of the DNA to be used as the probe was digested and run on a 1% low melting point agarose gel as described in sections 2.2.6 and 2.2.10 respectively. The required DNA fragment was then excised from the gel and 3 volumes of ddHzO were added. The gel slice was then heated to 100°C for Sminutes and than snap cooled by placing on ice for 2 minutes. 7 jli1 of the melted gel slice was then placed in a screw cap Eppendorf containing lOpf of oligo-labelling buffer, 5 units of DNA polymerase large fragment (Klenow) and SOpCi of a-[^^P]-dCTP in a total volume of 50pl. The labeling reaction was then incubated at 37°C for 1-2 hours. The reaction was then filtered through a Sephadex column (G25 or G50) to remove any unincorporated label. The radiolabelled DNA probe was then heated for 5 minutes at 100°C and snap cooled on ice for 2 minutes. The denatured probe was then added to the hybridization solution. Oligolabelling Buffer (OLB): Solution O: 1.25M Tris HCl pH 8.0 0.125M MgCfc Solution A: 1ml Solution O 18pl p-mercaptoethanol 5pl O.lMdATP 5pl O.lMdGTP 5pl O.lMdTTP Solution B: 2M HEPES pH6.6 Solution C: 90 units/ml random hexamers [pd(N)6] Dissolved in TE pH8.0

Oligolabelling buffer was prepared by mixing A:B:C in a ratio of 100:250:150.

126 II.2.2.F.d Hybridisation

The nylon membranes were pre-hybridised for at least 1 hour at 65°C with 3Omis of pre­ hybridisation solution (6x SSC, 5x Denhardt’s reagent, 0.5% w/v SDS in ddKbO containing lOOp-g/ml of denatured herring sperm DNA (100 x Denhardt’s reagent: 2% w/v bovine serum albumin, 2% Ficoll® (type 400), 2% w/v polyvinylpyrrolidone in ddH20)). The volume of the pre-hybridisation solution was reduced to approximately 5mls prior to adding the denatured probe. The probe was hybridised to the membrane overnight at 65°C. The membrane was then washed twice for 10 minutes in 2x SSC/0.1% w/v SDS, then twice for 5 minutes with in Ix SSC/0.1% w/v SDS, then twice for 5 minutes with in 0.1 x SSC/0.1% w/v SDS. The filters were then wrapped in cling film and exposed to X-ray film.

127 II.2.3 RNA Methods

11.2.3.A Isolation of total cellular RNA

5 X 10^ cells were washed twice in 1ml of Ix PBS and harvested in 1 ml of RNA Isolator (Genosys Biotechnologies), a RNA isolation solution based on the guanidinium thiocyanate isolation method. Total cellular RNA was then extracted according to the manufacturer’s instructions. Briefly, after the addition of the RNA isolator 200|Li1 of chloroform was added and the solution was vortexed for 15 minutes. The sample was then centrifuged for 10 minutes at 14,000 rpm at 4°C in a microcentrifuge. The aqueous phase was then transferred to a new 1.5 ml microcentrifuge tube and 500|il of isopropanol was added. The tube was inverted 10 times and allowed to incubate at room temperature for 5 minutes. The sample was then centrifuged for 10 minutes at 14,000 rpm at 4°C in a microcentrifuge. The supernatant was removed from the RNA pellet which was then washed with 75% ethanol. The tube was then centrifuged as above, the ethanol removed and the RNA pellet allowed to air dry for 5 minutes at room temperature.

11.2.3.B Solubilization of RNA in formamide

Extracted RNA pellets were resuspended in formamide. Resuspension of RNA in formamide has several benefits over conventional storage in water or ethanol. Formamide has the advantage that it will protect the RNA from RNases; consequently, the RNA is quite stable, allowing it to be stored overnight at 4°C or indefinitely at -20°C. In addition, samples may be stored at high concentrations, up to 4 mg/ml. Finally, formamide denatures the RNA, reducing the problems associated with secondary structure experienced in several RNA protocols.

After alcohol precipitation and drying, the RNA pellet was resuspended in 50jLil 100% deionized formamide. The RNA was quantitated on a UV spectrophometer, and the appropriate concentration of formamide was added to the blank to account for any background from the formamide.

128 II.2.3.C cDNA synthesis

The following components were assembled in a sterile 1.5 ml microcentrifuge tube: Template RNA 5 pg primer pd(N6) 1 pi (9 OD units/ml) (heated to 65°C then placed on ice) 5x AMV RT Buffer 10 pi RNase inhibitor 1 pi (40 units) dNTP mix 10 mM of each dNTP AMV reverse transcriptase 1 pi (200 units) DEPCed water to final volume of 50 pi

This solution was then removed from the ice and incubated at 37°C for 20 minutes followed by 42°C for a further 20 minutes. The cDNA was then used immediately in PCR reactions.

II.2.3.D PCR

The following components were assembled in a sterile 0.5 ml thin-walled microcentrifuge tube:

lOx Taq DNA polymerase Buffer 5 pi dNTP mix 10 mM of each dNTP 25 mM MgCb 3 pi downstream primer 50 pmol upstream primer 50 pmol cDNA template 2 ml (directly from 5pg cDNA reaction) Taq DNA polymerase 0.25 ml (5 units/pl) Nuclease-free water to final volume of 50 pi

The tubes were then placed in a controlled temperature beaat block and subjected to the following thermal cycling profile:

STEP TEMP TIME PROCESS

1. 94°C 30 seconds dénaturation

129 2. 94°C 30 seconds dénaturation 3. x^°C 30 seconds annealing 4. 72°C 30 seconds extension steps 2-4 cycled times

5. 72°C 60 seconds final extension

6. 4°C 30 seconds soak

^Annealing temperature: The annaealing temperature was 5°C less than the average melting temperature of the two primers involved in the PCR. See section II. 1.7 for a description of the oligonucleotide primers used.

^Number of cvcles: The number of cycles chosen for each set of primers was determined by taking samples at different numbers of cycles to locate the exponential phase of the amplification reaction. Cycle numbers that represented the exponential phase of the amplification reaction were individually chosen for each set of primers.

130 II.2.4 Protein Methods

11.2.4.A SDS-polyacrylamide gel electrophoresis (SDS-PAGE)

Protein samples were heated to 95°C for 5 minutes. Approximately 50-100|ig (total protein) of sample or 3pi of colored molecular weight protein standards (Rainbow Markers) were loaded per lane. Standard SDS-polyacrylamide gels were prepared and run in a vertical gel electrophoresis system as described by Sambrook and colleagues (Sambrook et al, 1989). The composition of the stacking and resolving gels was as described by Sambrook and colleagues (Sambrook et al., 1989).

Gels were run at constant current (30-40 mA/gel) with variable voltage in Ix running buffer (25mM Tris, 250mM glycine and 0.1% (w/v) SDS, pH 8.3). Gels were run until the protein of interest would be in approximately the middle of the gel (as determined by the migration of the colored molecular weight markers).

11.2.4.B Equalisation of protein loading 11.2.4.B.a Coomassie Blue Staining

Duplicate protein samples were separated by SDS-PAGE (section 2.5.3) and the gel placed in Coomassie stain solution (2% (w/v) Coomassie brilliant blue R250, 50% (w/v) methanol, 50% (v/v) glacial acetic acid) for 1 hour at room temperature with continual shaking. Any unbound stain was then removed by repeated replacements of destain solution (10% (w/v) glacial acetic acid, 30% (v/v) methanol). 11.2.4.B.b Bradford Chemistry

Protein concentration of duplicate protein samples were determined by Bradford chemistry using BCA Protein Assay Reagent From Pierce (Rockford, Illinois, U.S.A.) according to the manufacturers instructions. Briefly, BCA Reagent A and BCA Reagent B were combined and mixed at a ratio of 50 to 1. 300jxl of this mixture was added to the appropriate number of wells of a 96-well microtitre plate. 1, 5, or 10 pi of each protein

131 sample of unknown concentration was added to individual wells containing the BCA mixture on the plate. Finally, a titration series of known concentrations of bovine serum albumin were added to individual wells of the plate for the generation of a protein concentration standard curve. At least one well containing the BCA reagent mixture had

5|li1 of ddHzO added to be used as a blank against which to compare all other values. The plate was then incubated at 37°C for 30 minutes. After incubation, the plate was cooled to room temperature and the absorbance at 562nm was read on a microtitre plate reader. The value of the absorbance of the blank was subtracted ffon all other values, a standard curve was generated using the values obtained from the albumin titration, and protein concentrations of the experimental samples were determined by comparison to the albumin standard curve.

11.2.4.C Transfer of proteins to nitrocellulose membranes (western blotting)

Proteins separated on SDS-PAGE gels were transferred to Hybond C membranes using a wet-transfer method based on that of Sambrook and colleagues (Sambrook et a/., 1989). Briefly, the SDS-PAGE gel and the nitrocellulose membrane were pre-soaked in transfer buffer (50mM Tris, 380mM glycine, 0.1% (w/v) SDS, and 20% (v/v) methanol), sandwiched between sheets of pre-soaked Whatman 3MM paper, and a Trans-Blot™ Cell (BioRad) assembled according to manufacturer’s instructions. Transfer was carried out overnight at 200mA and 4°C. Following transfer, protein bands were viewed by staining the membrane with 0.5% (w/v) Ponceau S (in 1% (v/v) acetic acid) then washing the membrane with several changes of Ix PBS.

11.2.4.D Immunodetection of proteins on western blots

Following transfer, the membranes were blocked by incubation in 5% (w/v) skimmed milk powder in Ix PBS for 1 hour at room temperature with constant shaking. The membrane was then incubated with primary antibody diluted in 3% (w/v) skimmed milk powder in Ix PBS for 1-2 hours at room temperature with constant shaking. The

132 antibodies used in this thesis, their sources and the dilutions at which they were used are listed in section II. 1.6. Unbound antibody was removed by washing the membrane for 3 x 10 minutes in Ix PBS with 0.1% Tween-20 (PBST) at room temperature with constant shaking. The membrane was then incubated in an appropriate anti-IgG horseradish peroxidase (HRP) conjugated secondary antibody diluted in 3% (w/v) skimmed milk powder in Ix PBS for 1 hour at room temperature with constant shaking. Unbound antibody was removed by washing the membrane for 3 x 10 minutes in Ix PBST at room temperature with constant shaking. The bound HRP was then detected using an enhanced chemiluminescence system (ECL™) according to the manufacturer’s instructions. The membrane was then exposed to X-ray film to detect the resultant light emissions. Exposure times were between 1 second and 15 minutes, depending on the strength of the signal.

II.2.4.E Expression of labeled proteins by in vitro translation

^^S-labelled proteins were expressed by in vitro translation using the TNT-coupled transcription-translation system following conditions as described by the manufacturer (Promega Corp., Madison, WI, USA). Briefly, plasmid DNA was prepared specifically for use with the TNT system using standard alkaline lysis method described earlier. Great care was taken to ensure no ribonucleases were present in the final plasmid DNA suspension that was consistently resuspended at a concentration of SOOng per pi in DEPC-treated Milli-Q H2O. A standard TNT lysate reaction as given by the manufacturer is shown below:

Component ______Volume (in ill) TNT rabbit reticulocyte lysate 25 TNT reaction buffer 2 TNT RNA polymerase 1 Amino acid mixture (-Methionine) 1 S-Methionine (> 1 OOOCi/mM@l OmCi/ml) 2 RNasin ribonuclease inhibitor (400U/pI) 1 Plasmid DNA template (500ng/pl) 2 Nuclease fi*ee H2O 16

133 The reaction mixture (final volume 50jil) was assembled on ice in a 0.5ml microcentrifuge tube and incubated st 30°C for 30 minutes. The reaction mixture was then stored at -20°C until analysis with SDS/PAGE and autoradiography.

11.2.4.F Preparation of GST fusion protein

GST fusion protein expression and purification ware carried out as described by Pharmacia (Pharmacia 1996). A 20ml LB/100 |ig/ml ampicillin overnight (37°C, 200 rpm) culture innoculated with a single colony of BL21 (DE3) pLysS cells containing a recombinant pGEX 2TK plasmid was grown. 12-15 hours later, this culture was diluted 1:100 in fi-esh LB/100 jig/ml ampicillin and incubated 4 - 6 hours (37°C, 200 rpm) until the OD A600 reached 0.5. The cells were then induced with 400|iM IPTG and incubated (37°C, 200 rpm) for another 4 hours. The cells were then pelleted by centrifugation and resuspended in 1/20 growth volume of PBS containing 0.1% Triton X-100. The cell suspension was lysed by fi*eezing in the -80°C freezer for 1 hour, then allowed to slowly thaw overnight at 4°C. The cell debris was then pelleted by centrifugation at 10,000 rpm in a Sorvall SS-34 rotor for 10 minutes at 4^C. The supernatant was recovered and prepared for affinity purification using glutathione sepharose 4B.

11.2.4.G Affinity purification of GST fusion proteins using Glutathione Sepharose

Bacterial cell lysate supernatant was incubated with rotation with Ijitl of glutathione sepharose 4B per ml of original bacterial growth culture at room temperature for 1 hour or 4° C for 4 - 12 hours in Ix PBS containing 0.1% Triton X-100. After the incubation, the glutathione sepharose was pelleted and the supernatant removed. The glutathione sepharose was then washed four times with ice cold PBS containing 0.1% Triton X-100 at a volume that was equal to 1/20^^ of original bacterial growth culture. The glutathione sepharose containing the purified fusion protein was then pelleted and frozen at -20° C for use in future experiments.

134 11.2.4.H Glutathione-S-transferase affinity pull-down chromatography

Bacterially overexpressed GST/Bm-3 protein bound to glutathione sepharose was run on a SDS/15% polyacrylamide gel to check for size, purity, and approximate concentration of overexpressed protein. Approximately SOjig of total purified protein was used for each individual pull-down experiment. 50|Lig of total purified protein was resuspended in a 1.5ml microcentrifuge tube with 1.0ml NENT buffer (lOOmM NaCl, ImM EDTA, 20mM Tris pH 7.4, 0.5% NP-40, 0.5% milk powder) and incubated at room temperature on a rotating wheel for 30 minutes. The glutathione sepharose was then pelleted by centrifugation for 30 seconds at 2000 ipm in a microcentrifuge. The supernatant was then removed and the pelleted glutathione sepharose was then resuspended for blocking in NENT buffer containing 20% milk powder. The sepharose was then incubated with rotation in this NENT blocking solution for 15 minutes at room temperature. The sepharose was then pelleted, the supernatant removed, and washed three times with 1 ml of NENT buffer. After the final wash the sepharose beads were resuspended in lOOpl of transcription buffer (20mM HEPES pH 7.9, 60mM NaCl, ImM DTT, 6mM MgCh, 8.2% glycine, and O.lmM EDTA). The sepharose beads were then incubated on a rotating wheel for 60 minutes at room temperature with 5|il of labelled proteins expressed by in vitro translation using the TNT-coupled transcription-translation system with conditions as described by the manufacturer (Promega Corp., Madison, WI, USA). The sepharose beads were then washed 5 times each with 1.0 ml of NENT buffer. After the final wash the sepharose beads were resuspended in 20pl of SDS gel loading buffer, heated to 95°C for 5 minutes and then stored at 4°C or -20°C until analysis with SDS/PAGE and autoradiography.

II.2.4.I Coimmunoprecipitation of protein complexes from tissue culture samples

Adherent cells were washed twice with ice cold PBS. After the PBS fi’om the second wash had been removed from the cells 1.0ml of modified RIPA buffer (50mM Tris pH 7.4, 1% NP-40, 0.25% Na-deoxycholate, 150mM NaCl, ImM EDTA, ImM PMSF, ImM Na3V04, ImM NaF) per 160cm? flask, or 2x10^ cells. Cells were then scraped fi-om the

135 flask and pipetted into a 1.5ml microcentrifuge tube. Cells were then incubated on a rotating wheel for 15 minutes at 4°C. The cell lysate was then spun at maximum speed (14,000 rpm) for 10 minutes at 4^C in a refrigerated microcentrifuge. The supernatant was then immediately removed to a fresh microcentrifuge tube, and then pre-treated with lOOjLil of a 50% slurry of protein A/G sepharose on a rotating wheel for 15 minutes at 4°C to reduce the non-specific binding of proteins to the sepharose when used later in the experiment. The sepharose was then pelleted by centrifugation at 10000 rpm for 5 minutes at 4°C in a refrigerated microcentrifuge. The supernatant was removed to a fresh microcentrifuge tube and its concentration was determined using Bradford chemistry. The cell lysate was then diluted to a concentration of 5jig per jil. 5|ig of immunoprecipitating antibody was added to lOOjil (500p.g) of cell lysate and the resulting antibody/lysate mixture was then incubated on a rotating wheel for 90 minutes at room temperature. Immunocomplexes were captured by adding lOOjil of a 50% slurry of protein A/G sepharose and then incubating on a rotating wheel for 60 minutes at room temperature. The protein A/G sepharose beads were then collected by centrifugation (10 seconds at 14000 rpm). The supernatant was discarded and the sepharose beads were washed 5 times with 1ml of ice cold RIPA buffer. After the final wash the sepharose beads were resuspended in 20 |li1 of SDS gel loading buffer, heated to 95°C for 5 minutes and then stored at 4°C or -20°C until analysis with SDS/PAGE and western immunodetection.

136 II.2.5 Cell culture

11.2.5.A Culture conditions

All cell lines were stored long term in liquid nitrogen and during culture were maintained at 37°C in either a 5% CO^ incubator in a humidified. All manipulations of cells were carried out under sterile conditions using standard aseptic techniques.

IL2.5.B Freezing and recovery of cell stocks

2 Cell stocks for long term storage were prepared by suspending the cells fi*om one 175cm flask in 1.8 ml of 8% (v/v) dimethylsulphoxide (DMSO), 30% PCS, 68% appropriate medium. The freezing vials were slowly cooled to -70°C, then immersed in liquid nitrogen. Cells were recovered by rapidly thawing the contents of one vial and 2 transferring the cells to a 25cm flask of pre-warmed medium containing the appropriate selection if required. The medium was changed or the cells passaged the following day.

11.2.5.C Routine cell passage

Cells were grown in the appropriate growth media. When they were 80-90% confluent the cells were passaged as follows. For BHK21, MCF7, and MCF7-based cell lines, the monolayers were rinsed with Hank’s Balanced Salt Solution (HBSS) without calcium or magnesium, detached with trypsin/versene (1:10) and disaggregated. Fresh flasks were seeded at a ratio of 1:10. ND7 cells were rinsed with Hank’s Balanced Salt Solution (HBSS) without calcium of magnesium, detached by physically banging the cells off the flask, and fresh flasks were seeded at a ratio of 1:10.

11.2.5.D Transfection of plasmids for transient expression

137 The classic calcium-phosphate method of transfection was used for the introduction of plasmid DNA into tissue culture cell lines. Cells were plated at the appropriate density (usually 0.5 x 10^ in 6-well plates) one day before transfection. On the day of the transfection the medium was removed and replaced with 1.0ml pre-warmed fresh medium (in the case of transfections in ND7 cells, the L15-based medium was replaced DMEM + 10% PCS). The calcium phosphate was prepared as follows. The stated amount of each plasmid DNA (see specific transfection result) was combined with empty plasmid reporter construct to a final concentration of 20 jig of DNA per transfection in a sterile microcentrifuge tube. 6.2| li1 of freshly prepared 2M CaCh was added to each plasmid DNA solution, and the final volume was brought up to 50|il with sterile Milli-Q water. This solution was added to 50|il freshly prepared 2x HBS (1.64% (w/v) NaCl, 1.19% (w/v) HEPES, 0.04% (w/v) Na2? 0 4 , adjusted to pH 7.12 exactly), and vortexed immediately. The precipitate was then incubated at room temperature for 30 minutes and then vortexed once again. The entire lOOpl of the precipitate was then added to the medium on the cells. The cells were returned to the incubator and transfected for 5 hours. After 5 hours, the medium containing the precipitate was removed, the cells were washed once with pre-warmed sterile PBS which was removed and replaced with fresh pre- warmed growth medium.

H.2.5.E CAT assay

The promoter activity of transfected promoter constructs was measured by determining the activity of the chloramphenicol acetyltransferase (CAT) enzyme gene product which catalyses the transfer of an acetyl group from acetyl co-enzyme A to ^^C labeled chloramphenicol. This activity was assayed using thin layer chromatography as described by Sambrook and colleagues (Sambrook et al., 1989).

After three days transfected cells were washed once with PBS and harvested in 50|li1 of 0.25M Tris pH 7.8. The cells were then lysed by subjecting them to two freeze thaw cycles. Generally, 20)li1 of harvested cellular extract was made up to 90 pi with 0.25M Tris pH 7.8. This solution was then added to a fresh microcentrifuge tube containing 20pl 4mM acetyl co-enzyme A, 4pl labeled chloramphenicol, and 35pi sterile water.

138 The resulting solution was then incubated at 37°C for 60 minutes. After incubation the labeled chloramphenicol was extracted from the aqueous solution by vortexing with 1ml of ethyl acetate. The layers were separated by centrifugation and the upper organic layer was removed to a fresh tube and the ethyl acetate was evaporated in a vacuum dryer. The remaining labeled chloramphenicol was resuspended in 15|xl of fresh ethyl acetate and applied to silica gel thin layer chromatography plates. The unconverted, mono-acetylated and di-acetylated forms of the labeled chloramphenicol were resolved by ascending chromatography in a 95:5;:chloroform;methanol solvent. After air-drying, the plates were used to expose X-ray film. The resulting autoradiographs were analyzed by densitometry. The ability of the expression vector containing the factor of interest to activate the reporter construct was determined as a percentage of the activation observed with the empty expression vector.

II.2.5.F Construction of stable cell lines

MCF7 cells were plated out in six well dishes and transfected using the calcium phosphate method described above. The cells were cotransfected with a 5:1 molar ratio of plasmid containing the Bm-3b constructs to that containing a neomycin selectable marker. Stable cell lines expressing recombinant Bm-3b short, Bm-3b anti-sense, as well as control cell lines containing empty pLTR (all under the control of the Moloney Murine Lukemia Virus promoter) were prepared, selected, and isolated as follows. MCF7 cells (5x10^) in each well of a six well dish were transfected using pLTR Bm-3b short, pLTR empty, or pJ4 Bm-3b anti-sense (Spg each DNA) using the calcium phosphate method. In each of the transfections empty pCi Neo (l|ig DNA) was cotransfected for antibiotic selection of successfully transfected cells. The cells were allowed to grow for five days post-transfection (approximately two doublings) under non-selective conditions. After five days had passed full growth medium containing 800|ig/ml of G418 sulphate was used to select antibiotic resistant clones. After 14 days, (two days after 100% of the mock transfected cells had died under the G-418 selection) single colonies were evacuated in 150 |il using a Gilson P-1000 pipetteman and removed to an isolated well in a 24-well plate containing 1.0 ml of full growth medium. Individual clones were

139 expanded up to two 80cm? flasks. One flask of each 80cm? flask of each clone was then prepared for storage as a DMSO stock while the other was used to propagate cells to characterize each of the clones.

IL2.5.G Cellular growth curves and saturation density limitation studies

Two MCF7 clonal cell lines representing each of the stably integrated constructs were grown to 80% confluence, trypsinised, counted and plated into 24-well plates using 1 ml of cells at 1 X 10"^ cells per ml. The medium was changed every three days for 20 days. Triplicate wells were counted for each time point.

II.2.5.H Saturation density limitation studies

Two MCF7 clonal cell lines representing each of the stably integrated constructs were grown to 80% confluence, trypsinised, counted and plated into 24-well plates containing 13 mm diameter coverslips in each well using 1 ml of cells at 1 x 10^ cells per ml. After 48 hours, the coverslips were transferred to 6- well dishes containing 5 mis of medium. The medium was changed every three days for 16 days. Ten days post-transfer to the 6- well plates, triplicate wells were counted for each time point every 48 hours.

II.2.5.I Tritiated thymidine incorporation proliferation studies

Two MCF7 clonal cell lines representing each of the stably integrated constructs were grown to 80% confluence, trypsinised, counted and plated into 96-well plates using 200 |il of cells at 1 X 10^ cells per ml. Cells were grown for the indicated amount of time. At the appropriate time point cells were pulsed with 10 pi of 50 pCi/ml [^H] thymidine for one hour. Cells were then trypsinised and then harvested into glass filters, and washed using a Tomtec cell harvester. The glass filters were then treated with scintillant and counted on a Nuclear Enterprises NE 10600 scintillation counter.

140 11.2.5.J Anchorage independent growth studies

Anchorage independent growth studies were carried out in soft agar in 6-well plates. A base layer of 2 mis of growth medium containing 0.7% agarose was covered with 1 ml of growth medium containing 0.35% agarose and 1x10^ cells per ml. Finally, 2 mis of growth medium containing 0.7% agarose was layered atop the cells. Colonies were allowed to grow for 21 days. After 21 days, colonies comprised of at least 32 cells were counted on an inverted microscope.

11.2.5.K Standard protein extraction from cultured cells

1 X 10^ cells were washed twice in 1ml of Ix PBS and harvested in 100|il of standard protein sample buffer (5% (3-mercaptoethanol, SOmM Tris-HCl pH 8.0, 6% (v/v) glycerol, 2% (w/v) SDS, and 0.005% (w/v) bromophenol blue). The samples were immediately placed on ice and then heated to 95°C for 5 minutes. The samples were either loaded immediately onto a SDS-polyacrylamide gel or stored at -20°C.

141 m BRN-3 POU TRANSCRIPTION FACTORS AND p160 STEROID RECEPTOR COACTIVATORS INTERACT AND COREGULATE TRANSCRIPTION FROM VARIOUS PROMOTERS IN DIFFERENT CELL TYPES

III.l Introduction

The POU family of transcription factors represent a diverse and powerful group of transcription factors with roles in the cell fate specification of the developing embryo (Veenstra et al, 1997) as well as the maintenance of mature cell phenotypes (Latchman, 1999; McEvilly et al., 1996; McEvilly & Rosenfeld, 1999). The Bm-3 proteins originally identified in neuronal tissue, have also been found in the tissues of both the nervous system and the reproductive tract (Budhram-Mahadeo-Heads, 1995). The Bm-3 proteins are able to activate and repress a wide variety of promoters and may exert their effects independently or through interactions with cofactors.

Bm-3 a and Bm-3b have been detected in the tissues of the reproductive tract including cervical epithelium, testis, and breast (Budhram Mahadeo et al., 1999b; Budhram- Mahadeo-Heads, 1995; Lillycrop et al, 1992; Ndisang et al., 1999). In 1998, Budhraham-Mahadeo and colleagues showed an interaction between Bm-3 proteins and the estrogen receptor (ER) (Budhram Mahadeo et al, 1998). The estrogen receptor is a classic important for the estrogen inducible expression of a wide variety of cellular genes. Upon binding its ligand, estrogen, the receptor homodimerizes and becomes a powerful transcription factor. The estrogen receptor, however, is necessary, but not sufficient to act alone as a transcriptional activator (Mukherjee & Chambon, 1990). The estrogen receptor must interact with other transcriptional cofactors to mediate its effect on estrogen responsive genes. Bm-3 a and Bm-3b are examples of two of these coregulatory molecules. In a series of affinity chromatography pull down experiments, Bm-3 a and Bm-3b were shown to interact with in vitro translated estrogen receptor. Moreover, the ER/Bm-3 complex could be

142 immunoprecipitated from brain and ovary tissue. Both Bm-3 a and Bm-3b were shown to increase the affinity of the ER for the ERE in electromobility shift assays. Finally, transient transfection experiments in MCF7 cells showed that Bm-3 a and Bm-3b can modulate the activity of a promoter containing an ERE, with Bm-3 a repressing the activity of the promoter and Bm-3b strongly activating the promoter (Budhram Mahadeo et al, 1998). The POU domain containing proteins are potent regulators of transcription themselves, and it would be useful to elucidate the mode of action of the combination of these two sets of factors.

Another family of proteins, the p i60 steroid receptor coactivators, has recently been identified by its ability to interact with steroid nuclear receptors and coregulate transcription from steroid receptor response elements (Shibata et al., 1997). When nuclear receptors were used as bait in a series of biochemical and yeast hybrid experiments screening for proteins interacting with the nuclear receptor ligand binding domain, a family of related proteins collectively termed the p i60 steroid receptor coactivators were discovered (Onate et al., 1995). As well as sequence homology, the members of this family share the ability to stimulate ligand-dependant transactivation with several nuclear receptors.

A common feature to all the members of the p i60 family is the presence of several LXXLL motifs, where L is leucine and X is any amino acid (Heery et al., 1997; Torchia, Glass & Rosenfeld, 1998) (also named LXDs, NR boxes, or NIDs) which are responsible for the direct interaction with the nuclear receptor ligand binding domain. The crystal structures of the ligand binding domain of nuclear receptors show that upon ligand binding, the alpha helix containing the AF2 domain is severely distorted forming a “charged clamp” that accommodates pi 60 co-activators within a hydrophobic cleft of the ligand binding domain (Glass, Rose & Rosenfeld, 1997).

Experiments to date support the following model for steroid receptor coactivator stimulated transcription. Upon binding of ligand, the steroid receptor undergoes a conformational change in its ligand-binding domain. This conformational change allows

143 the steroid receptor coactivator to bind the steroid receptor in the hydrophobic cleft through direct contacts with the coactivators NR box (LXXLL motifs) found in the middle of the protein in the nuclear receptor domain (NRD) (Feng et al., 1998). Recent mutational mapping has shown that the second NR box in the NRD has the strongest interaction with the ER. Interestingly, when estrogen antagonists are bound to the ER ligand-binding domain, the distortion in the AF2 is such that alpha helix twelve (normally responsible for the creation of the hydrophobic cleft) occupies the area of the would-be cleft making binding by the coactivator impossible (Moras & Gronemeyer, 1998). One possible mode of action of the estrogen antagonists may therefore be to preclude any interaction with the pi 60 family of coactivators.

Therefore: > p i60 coactivators are capable of forming complexes with the estrogen receptor which in turn coactivate genes containing an ERE; > Bm-3 proteins are, too, capable of forming a complex with the estrogen receptor that in turn coactivates genes containing an ERE.

For these reasons, a functional interaction between the Bm-3 transcription factors and the p i60 coactivators was investigated.

144 III.2 Results

III.2.1 Src-1 specifically binds to Brn-3 proteins Because both Bm-3 proteins and the p i60 steroid receptor coactivators are known to interact with the estrogen receptor, it was tested if Bm-3 proteins physically interact with p i60 steroid receptor coactivators using glutathione-S-transferase affinity pull-down chromatography. GST alone and GST fused to Bm-3 constmcts (Bm-3a long, Bm-3b short and Bm-3 a POU domain) were expressed, purified and tested for interaction with in vitro translated ^^S-labelled Src-1 a and Src-le. As shown in Figures III.l and III.2, the radiolabelled Src-1 a and Src-le readily interacted with the GST-Bm-3 constmcts. Moreover, the Src-le constmct did not interact with the GST alone. The radiolabelled luciferase control did not bind any of the GST proteins, as was expected (Figure III.3). It is important to note here that each of the experiments shown in this section was completed at the same time and under identical conditions. Thus, Bm-3a long, Bm-3b short, and the Bm-3 a POU domain are able to specifically physically interact with Src-1 a and Src-le. Other cofactors relevant to steroid dependent gene transcription were also tested for interaction with the Bm-3 proteins. These included: rat ERp, human ERP, the nuclear receptor corepressor, SMRT, and the p300/CBP Associated Factor, p/CAF. Radiolabelled human ER{3, SMRT, and p/CAF interacted with the POU domains of Bm- 3 a and Bm-3b, while the rat ERp did not interact with either of the POU domains (data not shown).

145 I* I o ^ (/) Q. ( 0 . 0 ( 0 CO CO CO E £ £ m CO m 'S' P I— I— sS V) V) tn o o o o CL + + + (0 (0 (0 (Ü T T T V u o u u L_ L_ L_ k_ CO CO CO CO

> Ë ^ Ë

Src-1 a

%retention of load 100 33 51 28

Figure III.1: Affinity chromatography pull down ofin vitro translated ^^S- methionine SRC-1 a protein with GST/Brn-3 chimeras attached to glutathione sepharose 4B. In vitro translated SRC-la was incubated with the indicated GST proteins attached to glutathione sepharose resin. The resin with associated complexes was then washed and resolved on an SDS/12% polyacrylamide gel. Radiolabelled SRC-la was detected after the gel was dried and exposed to film overnight. The position of the SRC-la is indicated on the gel. The SRC-la interacts with each of the GST/Bm-3 constmcts. Percent of in vitro translate retained on the Bm-3 glutathione sepharose was determined using scanning densitometry.

146 Ils (0 ^ CO CO M M £ £ £ CO 0 0 CD sç (/)(/)(/) (0 o ü O O o SL + + + + Q) o o <1) o o o o o O l_ 1_ L_ 1- 1— CO CO CO C/3 CO ^ ^ ^ ^ ^

Src-1 e ------

%retention of load 100

Figure III.2: Affinity chromatography pull down ofin vitro translated ^^8- methionine SRC-1 e protein with GST/Brn-3 chimeras attached to glutathione sepharose 48. In vitro translated SRC-le was incubated with the indicated GST proteins attached to glutathione sepharose resin. The resin with associated complexes was then washed and resolved on an SDS/12% polyacrylamide gel. Radiolabelled SRC-le was detected after the gel was dried and exposed to film overnight. The position of the SRC-le is indicated on the gel. The SRC-le interacts with each of the GST/Bm-3 constructs. Percent of in vitro translate retained on the Bm-3 glutathione sepharose was determined using scanning densitometry.

147 1 1 8 o (A Q_ (Q .a (Q CO CO CO E £ E m CO m ^ P P P sS (0 (0 (O o o o o CL + + + Q) 0 )0)0 (/) (/)(/)(/) 2 2 2 2 S S S ^ Ü Ü Ü Ü 3 3 3 3

luciferase

Figure III.3: Affinity chromatography pull down ofin vitro translated ^®S- methionine luciferase protein with GST/Brn-3 chimeras attached to glutathione sepharose 48. In vitro translated luciferase was incubated with the indicated GST proteins attached to glutathione sepharose resin. The resin with associated complexes was then washed and resolved on an SDS/12% polyacrylamide gel. Radiolabelled luciferase was detected after the gel was dried and exposed to film overnight. The position of the luciferase is indicated on the gel. The luciferase is not retained by the Bm-3 glutathione sepharose.

148 III.2.2 Brn-3a long and Brn-3a short coimmunoprecipitate with Src-1

To independently test if the Bm-3/pl60 complexes exist in intact cells, coimmunoprecipitation experiments were used (Figure 111.4). In these experiments, aliquots of MCF-7 breast adenocarcinoma cell line extract were each individually incubated with anti-p53 antibody (positive control with a known interaction with Bm-3 a (Budhram Mahadeo et al., 1999a)), anti-pBad antibody (negative control), or anti-Src-1 antibody (experiment). The cell extracts were then incubated with protein A/G sepharose to purify the antibodies with the associated complexes from the supernatant. The protein A/G sepharose was then pelleted, washed, and run on a SDS/15% polyacrylamide gel. Proteins resolved in the gel were then blotted onto a nitrocellulose membrane and the membrane was then probed for Bm-3 a protein using a monoclonal anti-Bm-3a antibody. The Bm-3 a long form was detected in the p53 positive control, as well as in the Src-1 experimental samples in the MCF7 cell line extract. Small amounts of Bm-3 a were detected in the pBad negative control, while none was detected in the protein A/G sepharose alone control. Interestingly, the short form of Bm-3a was detected only in the Src-1 lanes from the extract, indicating that the interaction of the pi 60 protein and Bm-3 a occurs independently of the N-terminal region of Bm-3 a long.

149 MCF7 cell line

I c 2 o (/) o

Brn-3a long

Brn-3a short Protein A 0

Figure III.4: Coimmunoprecipitation of Brn-3a isoforms with SRC-1 from the MCF7 breast epithelial cell line. Protein extracts from MCF7 cells were incubated with the indicated antibodies. Immunocomplexes were collected on protein A sepharose resin, washed and resolved on a SDS/15% polyacrylamide gel and detected after Western blotting and probing with a monoclonal anti-Bm-3a antibody. The positions of the Bm-3a long and short forms as well as protein A are indicated.

150 III.2.3 Brn-3/pl60 cotransfections elicit varied responses on different promoters in different cell lines

111.2.3.A Transient cotransfections of the Brn-3 proteins with the pl60 proteins

Transient cotransfections of the Bm-3 proteins with the p i60 proteins on different promoter/reporter constructs, and in different cell lines were used to elucidate the functional interaction between the proteins.

111.2.3.A.a The promoters

111.2.3.A.a.i The SNAP-25 promoter Snap 25 is a presynaptic protein which is a key component in assembling a protein complex responsible for synaptic vesicle docking and fusion at nerve terminals. Bm-3a activates the SNAP-25 promoter in ND7 cells (Lakin et al., 1995) and Bm-3b represses this promoter in ND7 cells (Morris et al, 1996; Morris et al, 1994). The opposite and antagonistic effects of Bm-3 proteins on this promoter provide a good model for studying transcriptional activation and repression. In the experiments that follow, a 2.18 kb PstI - PstI (-2068/+114) fragment containing both transcriptional start sites of the SNAP gene in pCAT-Basic was used (Ryabinin et al., 1995).

111.2.3.A.a.ii The Vitellogenin promoter In experiments to investigate Bm-3/Src transcriptional regulation on an estrogen responsive promoter, rather than a Bm-3 response element, the vitellogenin A2 gene promoter was used. The Xenopus lavis vitellogenin (Vit) A2 gene encodes an egg yolk precursor protein whose production is stimulated by estrogen. The constmct used in the experiments which follow contained bases -331/-87 of the Vit A2 promoter followed by

151 -105/+5 of the Herpes simplex virus thymidine kinase (tk) promoter linked to the bacterial CAT gene (Klein-Hitpass et al, 1986).

111.2.3.A.b The cell lines

111.2.3.A.b.iND7 cells Because the promoter activities of Bm-3 a and Bm-3b have been most extensively characterized in neuronal cells, initial experiments were carried out in ND7 cells which were derived from rat dorsal root ganglion neurons immortalized by fusion with C l300 mouse neuroblastoma cells (Wood et al., 1990). Bm-3a is expressed at a minimal level in undifferentiated ND7 cells, but these cells do express relatively high levels of Bm-3b (Lillycrop et al, 1992). The levels of the endogenous Bm-3 proteins must be kept in mind when analyzing the results of the transfections.

111.2.3.A.b.ii BHK21 cells Because ND7 cells contain a basal level of Bm-3a and higher levels of Bm-3b, it is important to repeat Bm-3 transfections in a cell line which does not produce Bm-3a or Bm-3b. For these experiments, BHK21 cells which were derived from golden Syrian hamster kidney fibroblasts were used.

111.2.3.A.b.m MCF7 cells Because it is necessary to investigate the dependence of the Bm-3/pl60 interaction in a well-studied estrogen-responsive cell line, MCF7 cells were used in a set of experiments. MCF7 cells are derived from a human Caucasian breast adenocarcinoma, and are used routinely as an estrogen responsive cell line. One must keep in mind, however, that MCF7 cells contain endogenous Bm-3b and p i60 coactivator proteins.

152 111.2.3.B Experiments on the SNAP-25 promoter

111.2.3.B.a Experiments in ND7 cells

IIL2.3.B.b Bm-3 a long potentiates transactivation of the SNAP-25 promoter by Src-la in

ND7 cells.

In this experiment, increasing concentrations of Src-la were tested for their ability to regulate the SNAP-25 promoter both on its own and in the presence of Bm-3a. On its own, Src-la was able to activate the SNAP-25 promoter when the Src-la was transfected at high concentrations. The addition of Bm-3 a long potentiated the promoter activation of Src-la and allowed for the activation by the p i60 protein to occur at lower concentrations(Figure III.5).

111.2.3.B.b.iBm-3a long potentiates transactivation of the SNAP-25 promoter by Src-le in ND7 cells.

Similar to the results obtained with Src-la, Src-le was able to transactivate the SNAP-25 promoter in ND7 cells by itself in a dose dependent manner. Again, the addition of Bm- 3a long potentiated the promoter activity by Src-le and allowed transactivation by Src-le to occur at lower concentrations than on its own (Figure III.6).

153 [ ] pLTR empty

0 pLTR Brn-3a long 5 g

>

03 0) > CO 0) a:

SNAP CAT promoter (|ig) 2.5 2.5 2.5 2.5 2.5 2.5

pSG5 SRC-1 a (ng) 0.0 0.6 1.2 2.5 5.0 10.0

Figure 111.5: Effects of Brn-3a long and SRC-1 a cotransfection on SNAP 25 promoter transactivations in ND7 cells. ND7 cells were transfected with a SNAP promoter/CAT reporter construct and increasing amounts of pSG5 SRC-la in the presence of 2.5 pg of either empty pLTR or pLTR containing Bm-3a long. CAT reporter activity is reported from three experiments carried out in triplicate.

154 [D pLTR empty

@ pLTR Brn-3a long

n

SNAP CAT promoter (fig) 2.5 2.5 2.5 2.5 2.5 2.5

pSG5 SRC-1 e (|ig) 0.0 0.6 1.2 2.5 5.0 10.0

Figure III.6: Effects of Brn-3a long and SRC-1 e cotransfection on SNAP 25 promoter transactivations in ND7 cells. ND7 cells were transfected with a SNAP promoter/CAT reporter construct and increasing amounts of pSG5 SRC-le in the presence of 2.5 pg of either empty pLTR or pLTR containing Bm-3a long. CAT reporter activity is reported from three experiments carried out in triplicate.

155 111.2.3.B.C Experiments in BHK21 cells

111.2.3.B.c.i Bm-3a long and Bm-3b short potentiate transactivation of the SNAP-25 promoter by Src-la in BHK21 cells.

Src-la on its own is able to weakly activate the SNAP-25 promoter in BHK21 cells. However, the addition of Bm-3a long potentiated the activity of Src-la up to 9 fold. Concentrations of Src-la above a 1:1 ratio with Bm-3a reduced the transactivation in a dose dependent manner. While the basal activity of Bm-3b on the SNAP-25 promoter was uncharacteristically high, a similar pattern to the pattern of potentiation seen in the Bm-3a transfections was seen in the transfections that contained Bm-3b (Figure III.7).

111.2.3.B.c.ii Bm-3a long potentiates transactivation of the SNAP-25 promoter by Src-le inBHK21 cells.

Similar to the results in BHK cells with Src-la, Src-le was only able to activate the SNAP-25 promoter at high concentrations. Up to 8 fold transactivation was induced by the addition of Bm-3a long. It is noteworthy in this experiment that maximal potentiation of Src-le by Bm-3 a occurred when the ratio of Bm-3 a to Src-le was 4:1 or 2:1 with transactivation still occurring but decreasing at higher concentrations of Src-le. The result with Bm-3b was slightly more difficult to interpret due to the uncharacteristically high basal activity of Bm-3b on the SNAP-25 promoter (Figure III.8).

156 □ pLTR empty

H pLTR Brn-3a long

D pLTR Brn-3b short 5 o

SNAP CAT promoter (|ig) 2.5 2.5 2.5 2.5 2.5 2.5

pSG5 SRC-1 a (|ig) 0.0 0.6 1.2 2.5 5.0 10.0

Figure III.7: Effects of Brn-3a long or Brn-3b short and SRC-1 a cotransfection on SNAP 25 promoter transactivations in BHK21 cells. BHK21 cells were transfected with a SNAP promoter/CAT reporter construct and increasing amounts of pSG5 SRC-la in the presence of 2.5 pg of either empty pLTR, pLTR containing Bm-3a long, or pLTR containing Bm-3b short. CAT reporter activity is reported from three experiments carried out in triplicate.

157 Q pLTR empty

M pLTR Brn-3a long

O pLTR Brn-3b short S o

SNAP CAT promoter (jig) 2.5 2.5 2.5 2.5 2.5 2.5

pSG5 SRC-1 e (pg) 0.0 0.6 1.2 2.5 5.0 10.0

Figure III.8: Effects of Brn-3a long or Brn-3b short and SRC-1 e cotransfection on SNAP 25 promoter transactivations in BHK21 cells. BHK21 cells were transfected with a SNAP promoter/CAT reporter construct and increasing amounts of pSG5 SRC-le in the presence of 2.5 |ig of either empty pLTR, pLTR containing Bm-3 a long, or pLTR containing Bm-3b short. CAT reporter activity is reported from three experiments carried out in triplicate.

158 111.2.3.C Experiments on the Vit-tk promoter It was interesting to note that the p i60 steroid receptor coactivator proteins were able to coactivate transcription of Bm-3 proteins on a Bm-3 site. This experimental situation occurs out of the usual context of p i60 proteins acting as coactivators for steroid receptors on promoters containing steroid response elements. It was next investigated if Bm-3 proteins could regulate an estrogen responsive promoter when cotransfected with the p i60 coactivators. Bm-3b had previously been shown to produce a weak activation of the Vit-tk promoter in the absense of estrogens. This system was again explored to look at Bm-3/pl60 transactivation potential. The MCF7 cells in these experiments were grown in medium without the estrogenic phenolsulphonthalein (“phenol-red”) pH indicator (Berthois, Katzenellenbogen & Katzenellenbogen, 1986; Bindal et al., 1988; Bindal & Katzenellenbogen, 1988), containing semm pretreated with dextran coated charcoal to remove endogenous estrogens.

111.2.3.C.a Experiments with Src-la

111.2.3.C.a.iBm-3a long weakly activates while Bm-3b short weakly represses the activity of Src-la on the Vit-tk promoter in ND7 cells.

Both on its own and with the addition of Bm-3a or Bm-3b, Src-la had a minimal effect on the Vit-tk promoter constmct in ND7 cells (Figure 111.9).

159 □ pLTR empty

H pLTR Brn-3a long

CH pLTR Brn-3b short

VIT-TK promoter (jig) 2.5 2.5 2.5

pSG5 SRC-1 a (|ig) 2 . 5 5 . 0 10.0

Figure III.9: Effects of Brn-3a long or Brn-3b short and SRC-1 a cotransfection on VIT-TK promoter transactivations in ND7 cells. ND7 cells were transfected with a VIT-TK promoter/CAT reporter construct and increasing amounts of pSG5 SRC-la in the presence of 2.5 pg of either empty pLTR, pLTR containing Bm-3 a long, or pLTR containing Bm-3b short. CAT reporter activity is reported from two experiments carried out in triplicate.

1 6 0 111.2.3.C.a.ii Bm-3a long and Bm-3b short reverse the weak repression of the Vit-tk promoter by Src-la in BHK21 cells.

The results with Src-la on the Vit-tk promoter construct in BHK21 cells were similar to those of the same experiment in the ND7 cells: both on its own and with the addition of Bm-3a or Bm-3b, Src-la had a minimal effect on the Vit-tk promoter construct in ND7 cells (Figure III. 10).

111.2.3.C.a.iii Neither Bm-3 a long, Bm-3b short, nor Src-la have an effect on the Vit-tk promoter in the absence of estrogen in MCF7 cells.

No effect on the vit-tk basal promoter activity was seen by transfecting any concentration of Src-la into the estrogen deprived MCF7 cells. While on its own, Bm-3 a long had a slight repressive effect, no other effect was seen when Bm-3 a long was added to the Src- la transfections. Likewise, no effect was seen in any case when Bm-3b short was added to the transfection (Figure III.l 1).

161 H pL TR empty

■ pLTR Brn-3a long

O pLTR Brn-3b short

> 00o > 00 0) Ùl

VIT-TK promoter (ng) 2.5 2.5 2.5

pSG5 SRC-1a (|ig) 0.0 0.6 1.2 0 10

Figure 111.10: Effects of Brn-3a long or Brn-3b short and SRC-la cotransfection on VIT-TK promoter transactivations in BHK21 cells. BHK21 cells were transfected with a VIT-TK promoter/CAT reporter construct and increasing amounts of pSG5 SRC-la in the presence of 2.5 pg of either empty pLTR, pLTR containing Bm-3 a long, or pLTR containing Bm-3b short. CAT reporter activity is reported from two experiments carried out in triplicate.

162 □ pLTR empty

0 pLTR Brn-3a long

O pLTR Brn-3b short o

CD 0.4

VIT-TK promoter (|ig) pSG5 SRC-1 a (|ig) 5 . 0 10.0

Figure 111.11: Effects of Brn-3a long or Brn-3b short and SRC-1 a cotransfection on VIT-TK promoter transactivations in MCF7 cells. MCF7 cells were transfected with a VIT-TK promoter/CAT reporter construct and increasing amounts of pSG5 SRC-la in the presence of 2.5 jUg of either empty pLTR, pLTR containing Bm-3 a long, or pLTR containing Bm-3b short. CAT reporter activity is reported from two experiments carried out in triplicate.

163 111.2.3.C.b Experiments with Src-le

111.2.3.C.b.iBm-3a long responds to Src-le in a dose dependant manner, while Bm-3b short consistently represses the activity of Src-le on the Vit-tk promoter in ND7 cells.

On its own Src-le had no effect on the Vit-tk promoter in ND7 cells. The addition of Bm-3a long resulted in a dose-dependent bell shaped curve of initial and ultimate weak repression and slight activation when Bm-3a and Src-le were transfected in a 1:1 ratio. The addition of Bm-3b short consistently lowered the activity of the promoter in all cases irrespective of Src-le concentration (Figure III. 12).

111.2.3.C.b.ii Bm-3 a long and Bm-3b short consistently weakly potentiate repression of the Vit-tk promoter by Src-le in BHK21 cells.

The results with Src-le alone on the Vit-tk promoter construct in BHK21 cells were similar to those of the same experiment in ND7 cells: Src-le appeared to have no consistent effect on the Vit-tk promoter in BHK21 cells. The addition of Bm-3 a long, however, potentiated a Src-le dose-dependent repression of Vit-tk promoter activity. The same results held tme when Bm-3b short was added instead of Bm-3a long. The repression potentiated by Bm-3b short was stronger than that of Bm-3 a long (Figure III. 13).

164 [g pLTR empty

H pLTR Brn-3a long

n pLTR Brn-3b short

VIT-TK promoter (|ig) 2.5 2.5 2.5 2.5 2.5 2.5 pSG5 SRC-1e (|ig) 0.0 0.6 1.2 2.5 5.0 10.0

Figure 111.12: Effects of Brn-3a long or Brn-3b short and SRC-le cotransfection on VIT-TK promoter transactivations in ND7 cells. ND7 cells were transfected with a VIT-TK promoter/CAT reporter construct and increasing amounts of pSG5 SRC-le in the presence of 2.5 pg of either empty pLTR, pLTR containing Bm-3 a long, or pLTR containing Bm-3b short. CAT reporter activity is reported from two experiments carried out in triplicate.

165 □ pLTR empty

0 pLTR Brn-3a long

O pLTR Brn-3b short

VIT-TK promoter (pg) 2.5 2.5 2.5 2.5 2.5 pSG5 SRC-1 e (fig) 0.6 1.2 2.5 5.0 10.0

Figure 111.13: Effects of Brn-3a long or Brn-3b short and SRC-1 e cotransfection on VIT-TK promoter transactivations in BHK21 cells. BHK21 cells were transfected with a VIT-TK promoter/CAT reporter construct and increasing amounts of pSG5 SRC-le in the presence of 2.5 jig of either empty pLTR, pLTR containing Bm-3 a long, or pLTR containing Bm-3b short. CAT reporter activity is reported from two experiments carried out in triplicate.

1 6 6 III.2.3.C.b.iii Bm-3a long and Bm-3b short potentiate the repression of the Vit-tk promoter by Src-le in estrogen deprived MCF7 cells.

Src-le was on its own capable of repressing the activity of the Vit-tk promoter in a dose dependent manner. Repression with the addition of Bm-3 a long was consistently slightly stronger than with the Src-le alone and this repression consistently paralleled the repression seen by Src-le alone. The addition of Bm-3b short to the Src-le transfection resulted in a consistent strong repression of the basal promoter activity (Figure III. 14).

167 □ pLTR empty

H pLTR Bm-3a long

D pLTR Brn-3b short o OS

CD 0 4

VIT-TK promoter (fig) pSG5 SRC-1 e (fig) 1.2 2.5

Figure 111.14; Effects of Brn-3a long or Brn-3b short and SRC-1 e cotransfection on VIT-TK promoter transactivations in MCF7 cells. MCF7 cells were transfected with a VIT-TK promoter/CAT reporter construct and increasing amounts of pSG5 SRC-le in the presence of 2.5 pg of either empty pLTR, pLTR containing Bm-3 a long, or pLTR containing Bm-3b short. CAT reporter activity is reported from two experiments carried out in triplicate.

1 6 8 111.2.3.C.C Experiments with Src-3

111.2.3.C.c.iBm-3b short weakly represses the activity of AIBl on the Vit-tk promoter in ND7 cells.

AIB1 is capable of activating the Vit-tk promoter construct in a dose dependent manner. The addition of Bm-3 a long had no strong or consistent effect on the activity of AIBl on this promoter. However, similar to both the cases with Src-la and Src-le on the Vit-tk promoter in these cells, Bm-3b consistently repressed the activity of the promoter irrespective of AIBl concentration (Figure 111.15).

111.2.3.C.c.ii Bm-3a long and Bm-3b short consistently weakly potentiate repression of the Vit-tk promoter by AIBl in BHK21 cells.

Transfection of AIBl was did not alter the promoter activity of the Vit-tk promoter at any concentration. While Bm-3a long alone had no effect on the basal activity of the promoter constmct, it was able to potentiate a dose-dependent repression of the Vit-tk promoter by AIBl. And again, as was the case with Src-le for this promoter and cell line, Bm-3b short consistently potentiated a repression of the promoter activity by AIBl (Figure 111.16).

111.2.3.C.c.iii Bm-3a long and Bm-3b short potentiate the repression of the Vit-tk promoter by AIB 1 in estrogen deprived MCF7 cells.

Similar to the case seen above with the Src-le, AIBl on its own was capable of downregulating the basal activity of the Vit-tk promoter. The addition of Bm-3 a long had no strong consistent effect on the repression by AIB 1. The addition of Bm-3b short to the AIBl transfections, however, potentiated a consistent repressive effect on the Vit- tk promoter (Figure 111.17).

169 E] pLTR empty

9 pLTR Bm-3a long

D pLTR Brn-3b short 2 ê g- >

03 0) > 03 0) a:

VIT-TK promoter (jig) 2.5 2.5 2.5 2.5 2.5 2.5 pCDNA3.1 SRC-3 (pg) 0.0 0.6 1.2 2.5 5.0 10.0

Figure 111.15: Effects of Brn-3a long or Brn-3b short and SRC-3 cotransfection on VIT-TK promoter transactivations in ND7 cells. ND7 cells were transfected with a VIT-TK promoter/CAT reporter construct and increasing amounts of pCDNAS.l SRC-3 in the presence of 2.5 pg of either empty pLTR, pLTR containing Bm-3 a long, or pLTR containing Bm-3b short. CAT reporter activity is reported from two experiments carried out in triplicate.

170 □ pLTR empty

@ pLTR Brn-3a long

O pLTR Brn-3b short

VIT-TK promoter (|ig) 2.5 2.5 2.5 2.5 2.5

PCDNA3.1 SRC-3 (|ig) 0.0 0.6 1.2 2.5 5.0 1 0 . 0

Figure 111.16: Effects of Brn-3a long or Brn-3b short and SRC-3 cotransfection on VIT-TK promoter transactivations in BHK21 cells. BHK21 cells were transfected with a VIT-TK promoter/CAT reporter construct and increasing amounts of pCDNAS.l SRC-3 in the presence of 2.5 jig of either empty pLTR, pLTR containing Bm-3 a long, or pLTR containing Bm-3b short. CAT reporter activity is reported from two experiments carried out in triplicate.

171 m pLTR empty

0 pLTR Brn-3a long

D pLTR Brn-3b short

O 0.8

r o 0.4

VIT-TK promoter (|ig) 2.5 2.5 2.5 2.5 2.5 pCDNA3.1 SRC-3 (|ig) . 0 0.6 1.2 2.5 5.0 10.0

Figure 111.17: Effects of Brn-3a long or Brn-3b short and SRC-3 cotransfection on VIT-TK promoter transactivations in MCF7 cells. MCF7 cells were transfected with a VIT-TK promoter/CAT reporter construct and increasing amounts of pCDNAS.l SRC-3 in the presence of 2.5 |Lig of either empty pLTR, pLTR containing Bm-3a long, or pLTR containing Bm-3b short. CAT reporter activity is reported from two experiments carried out in triplicate.

172 111.3 Discussion

Direct physical and functional interactions exist between Bm-3 transcription factors and the p i60 coactivators. Moreover this interaction may work through both steroid dependent as well as steroid independent pathways. The molecular mechanism by which Bm-3 proteins potentiate the transcriptional activity of the p i60 proteins is unknown. The work described herein opens the possibility for several avenues of research to be pursued.

111.3.1 Brn-3 proteins physically interact with Src-1 proteins

In the GST pull-down assays, Src-la and Src-le were shown to interact in vitro with Bm- 3 a long, Bm-3b short, and the Bm-3 a POU domain. The results indicate that areas in the C-terminus of the Bm-3 proteins, specifically in the POU domain, are responsible for this interaction. This result was further confirmed in vivo in the immunoprécipitation experiments. Not only was the long form of Bm-3a seen in the immunoprécipitation, but also the short form. The results again indicate that the physical interaction between Bm- 3a and the p i60 protein relies on sequences that are not in the ^-terminal portion of the Bm-3 a protein. Further experiments must be undertaken to precisely define the location of the interaction. In the GST pull-down assays, however, there were differences when comparing the interaction between the Bm-3 proteins and Src-la and Src-le. The Bm-3/GST constmcts retained a greater percentage of the loaded Src-la than Src-le. The Bm-3 construct used retained between 28 and 51 percent of the radiolabelled Src-la loaded onto the sepharose. Much less of the loaded Src-le was retained by the Bm-3 constmcts in the same experiment. These differences may be interpreted in several ways. One explanation would include the idea that the interaction between the Bm-3 proteins and Src-la and Src-le occurs with different affinities. These different affinities in the pull-down assays could indicate that this interaction may rely on sequences in the divergent C-termini of

173 the Src-1 isoforms, and/or the absence of a cofactor which in vivo would mediate the interaction between the two groups of proteins. Alternatively, because: 1. less of the Src- 1 protein was pulled down in the affinity chromatography assay, 2. there was some interaction between the Src-1 protein and the pBad negative control in the coimmunoprecipitation, and 3. the Src family of proteins interact with many other families of transcription factors, of one might conclude that Src-1 may be a rather sticky protein, and thus that the weaker interaction of the Bm-3 poteins with Src-le may not be real. The relevance of Bm-3 proteins in systems associated with the regulation of steroid receptor dependent transcription is further supported by the interaction of the Bm-3 a and Bm-3b POU domains with the nuclear receptor corepressor SMRT and the p300/CBP Associated Factor (p/CAF) (data not shown). The physical interaction of Bm-3 transcription factors with each of these coregulatory molecules, which interact with and coregulate steroid responsive transcription with one another, indicates a general role for Bm-3 proteins in steroid receptor regulated systems.

III.3.2 p i60 proteins coactivate transcription with Brn-3 transcription factors on a Brn-3 responsive promoter

It is clear from the experiments in ND7 cells on the SNAP-25 promoter that Src-la and Src-le can act as coactivators for Bm-3 a. It is interesting to note, however that in this titration experiment, there is initial repression at low concentrations of Src-1 followed by activation as the concentration of Src-1 is increased. One interpretation of these results may take into account the levels of Bm-3 a and Bm-3b in the ND7 cells. Under normal conditions, proliferating ND7 cells contain low levels of Bm-3a, while the expression levels of Bm-3b are relatively high (Smith, Dawson & Latchman, 1997b; Smith et al, 1997c). Interestingly, in transfections with the SNAP-25 promoter/reporter constmct in ND7 cells there was a consistent repression of the promoter when low concentrations of Src-1 was cotransfected. An attractive model to describe this repression is that Bm-3b, which normally represses the SNAP-25 promoter, has a higher affinity for the Src-1

174 protein, and may use Src-1 as a corepressor in these cells on this promoter. In previous experiments it has been shown that not only can Bm-3b repress the SNAP-25 promoter, but also is able to intefere with the activation of this promoter by Bm-3a. This explanation is further supported by the GST pull-down data in which GST/Bm-3b retained almost twice as much of the IVT Src-1 a than GST/Bm-3a. Further studies are necessary to elucidate the interaction between the Bm-3 transcription factors and the Src- 1 proteins in ND7 cells on the SNAP 25 promoter. The situation is different in BHK cells which express neither Bm-3a nor Bm-3b. Here, Src-1 a has little effect on the transcriptional activity by Bm-3b, while Src-le is able to reduce the activity of the Bm-3b construct on the SNAP-25 promoter. It is interesting to note here that the Bm-3b short construct in these experiments does not repress the SNAP25 promoter in BHK21 cells as previously reported (Morris et al., 1997). The results described herein indicate that the precise state of the cells (including parameters such as passage number, exact density, or slight variations in growth conditions), or transfection conditions (including parameters such as exact composition of precipitate, length of time precipitate is incubated with cells, and exact transfection efficiency), may effect the activity of this Bm-3b short construct on the SNAP25 promoter/reporter construct. Further experiments are necessary to elucidate this phenomenon and accurately describe the activity of the Bm-3b transcription factor on this promoter.

III.3.3 pl60 steroid receptor coactivators can activate steroid receptor independent transcription

Briefly, the general model for steroid receptor coactivator binding (fully discussed in the introduction) is as follows: upon binding its ligand the steroid receptor changes its conformation in the ligand binding domain, thus enabling the recruitment of coactivators or corepressors. Indeed, the p i60 coactivators were initially identified by their ability to coactivate the transcription of hormone responsive genes by binding the ligand-bound steroid receptor. It is interesting to note that in the experiments described in this chapter,

175 the steroid receptor coactivators tested coactivated expression from a promoter with a transcription factor that is not a steroid receptor. Earlier in the discussion it was noted that the p i60 protein specifically physically interacts with the Bm-3 transcription factors. This lends support to the idea that this family of coactivators is not strictly limited to steroid receptors as partners for transcriptional regulation. Several proteins have been shown to interact with the p i60 coactivators including AP-1 (Lee et al, 1998b), SRF (Kim et aL, 1998a), NFkB (Na et al, 1998), CREE and STATs (Torchia et al, 1997), and p53 (Lee et al., 1999). The p i60 proteins have been shown to coactivate the transcriptional activity of each of these transcription factors. Bm-3 proteins may now be added to this list. One other characteristic that makes the Bm-3/pl60 coactivation on the SNAP-25 promoter interesting is that these activations occur on a steroid receptor independent promoter. Autoregulatory circuits control the expression of several homeobox genes ((Awgulewitsch & Jacobs, 1992; Bergson & McGinnis, 1990; Malicki et al., 1992; Packer et al, 1998; Popperl & Featherstone, 1992). Data from the next chapter indicates that Bm-3b may be capable of regulating its own expression (see Section IV.2.3 and Figure IV.3). It must be considered, then, that it is possible that the transcriptional effects seen by the addition of the Src proteins may be due, in part, to their regulating the levels of of Bm-3 produced in the transfected cells, rather than solely to the the ability of Srcs to act as transcriptional cofactors for Bm-3 proteins.

III.3.4 pl60 steroid receptor coactivator promoter regulation

Src-1 can specifically bind the and can potentiate serum response element mediated transactivations (Kim et al., 1998a). In addition, Src-1 is capable of enhancing p53 mediated transactivations (activations on promoters which contain p53 response elements instead of steroid response elements including: an artificial constmct containing 17 p53 sites, the MDM2 promoter, and the p21 promoter) (Lee et al., 1999). It is interesting that the p i60 coactivators are able to work entirely out of the context of interacting with steroid receptors and on steroid response elements. The work described

176 here supports the notion that p i60 proteins may act as general coactivators which may influence the transcription of genes other than steroid responsive genes. Indeed, there must be some fundamental interaction that allows the p i60 proteins to act as powerful coactivators in such different contexts (steroid receptors, constitutively active immediate early genes, tumor suppressor gene, and developmental transcription factor). Korzus and colleagues recognized this promiscuity in a 1998 Science article in which they showed that different mammalian transcription factors require distinct components of the CBP/p300 coactivator complex (including p i60 proteins) to function as transcriptional regulators. Hence, compositionally related but distinct CBP/p300 coactivator complexes exist in each physiological context. Moreover, each distinct complex has a different affinity for possible transcription factors (Korzus et al, 1998). This complex-dependent transcriptional regulation model may be applied to the results presented on the transfections on the different promoter constructs.

111.3.4.A The pl60 coactivators on the Vit-tk promoter construct

111.3.4.A.a Src-la

The addition of increasing concentrations of Src-la to Bm-3a long results in the activation of the Vit-tk promoter construct in ND7 cells, while these same increasing concentrations of Src-la have no effect on the activity of Bm-3b on the Vit-tk promoter construct. Thus, Src-la is part of a complex in ND7 cells which is able to interact with Bm-3 a long, but not Bm-3b short. The situation in BHK21 and MCF7 cells is different. The increasing concentrations of Src-la have no strong or consistent effect on the basal transcription activity of Bm-3 a long or Bm-3b short on the Vit-tk promoter, indicating that the Src-la complex here is either missing a component necessary for, or contains a component which precludes, fimctional interaction with either of the Bm-3 proteins.

111.3.4.A.b Src-le

177 No strong or consistent effect is seen when Bm-3a long or Bm-3b short are cotransfected with Src-le in ND7 cells. In BHK21 cells, Src-le is part of a complex which potentiates a weak repression of the Vit-tk promoter by Bm-3a and Bm-3b. A stronger repression by Bm-3b is seen in the MCF7 cells, while the addition of Bm-3 a does not vary from the addition of the Src-le transfected alone.

III.3.4.A.C Src-3

The results for Src-3/AIB 1 are similar to the results with Src-le. In ND7 cells no strong or consistent effect is seen with the addition of Bm-3a long or Bm-3b short. In BHK21 cells, however, AIBl must be part of a complex which can mediate the effects of Bm-3 and Bm-3b, as the repressive activity of both is potentiated by the addition of AIBl. In MCF7 cells, as was the case with Src-le, Bm-3b is able to interact with the AIB complex and repress transcription from the Vit-tk promoter, while Bm-3a has no strong effect in this context.

Here it is interesting to note the strongest unifying characteristic between Src-le and AIBl is that neither contains the extra C-terminal LXXLL motif found in Src-la. This extra motif may be responsible for differential transcriptional potential between the constmcts. This C-terminal LXXLL motif may then be responsible for the differential binding that was observed in the GST pull-down experiments when comparing Src-la and Src-le, in which the Bm-3 proteins retained 30-50% of the available Src-la, but less than 10% of the available Src-le.

IIÏ.3.4.B The cell lines The Src proteins were first identified as coactivators of steroid receptor dependent transcription. It is important, then, to understand that endogenous nuclear receptor in the cell lines may affect the outcome of the transient transfection experiments. The nervous system is, indeed, a target for steroid hormones (Shughme, Lane & Merchenthaler, 1997) which play roles in the proliferation, differentiation and maintenance of neuronal cells. Nuclear receptors including mineralocorticoid receptor (Kawata et al, 1998),

178 glucocorticoid receptor (Hu et al, 1997a; Hu et al, 1997b), and estrogen receptor (Osterlund et al, 2000) have each been loaclised in the brain. Moreover, several groups have specifically investigated the role of estrogen receptor in rat dorsal root ganglia (Patrone et al, 1999; Scoville, Button & Liuzzi, 1997; Yang et al, 1998a). The ND7 cell line is the result of a fusion between the mouse C l300 neurobalstoma cell line and rat dorsal root ganglia. Thus, one must be aware that the probability that the ND7 cells express the estrogen receptor is high. Less has been reported on the expression of nuclear receptors in kidney fibroblasts (fi*om which the BHK21 cell line was derived), but, it is clear that nuclear receptors play a role kidney development. The MCF7 breast adenocarcinoma cell line was chosen specifically for its estrogen receptor positivity. Therefore, it is necessary to be aware that endogenous nuclear receptors my play a role in the regulation of the expression of the promoter/reporter constructs used in each of these experiments.

111.3.4.B.a ND7 cells

In ND7 cells, Src-la is able to interact with Bm-3 a long which potentiates the activation of the Vit-tk promoter construct. However, because no strong effect is seen when Bm-3a long is added to the transfections with Src-le or AIBl, one may suggest that these p i60 proteins may not be able to form complexes with, or form complexes which preclude the addition of Bm-3 a long. The activity of the Bm-3b short on the Vit-tk promoter is generally unaltered in all three experiments with the p i60 proteins. The p i60 proteins may be unable to complex or complex in a non-functional manner with Bm-3b short in this cellular context. It is interesting to note here, then, that the Src-la complex in ND7 cells can differentiate between Bm-3 a long and Bm-3b short. This is not the case in the BHK2I cells.

111.3.4.B.bBHK21 ceUs

BHK cells appear to be missing one or more critical components that would allow the p i60 complex to differentiate between the Bm-3 proteins, as the effect of Bm-3a long

179 and Bm-3b short is generally the same for each of the individual p i60 proteins. Src-la is involved in a complex which is able to repress the Vit-tk promoter construct in BHK21 cells. However, the incorporation of Bm-3a long or Bm-3b short allows for the derepression of the promoter. The situation is similar but opposite for the complexes involving Src-le and AIBl. While each p i60 protein complex without the Bm-3 proteins has no strong consistent effect on the promoter, the incorporation of the Bm-3 proteins potentiates a consistent promoter repression.

III.3.4.B.C MCF7 cells

In the MCF7 cells, Src-la in the absence of ligand bound estrogen receptor is unable to form a complex, or forms a complex which is unable to interact with Bm-3a long or Bm- 3b short. This may be be due in part to its additional C-terminal LXXLL motif. Src-la and AIBl, however, are able to form a transcriptionally active complex with Bm-3 in the absence of ligand bound estrogen receptor. These p i60 complexes are able to incorporate at least Bm-3b short which potentiates repression of the Vit-tk promoter constmct in this cellular context. Thus, no strong effects were seen on the Vit-tk promoter in MCF7 cells deprived of estrogens. It would be quite interesting next to investigate if and how these transcriptional effects are altered in the presence of 17 13- estradiol.

III.3.5 3.3.5 pl60 proteins may act as transcriptional integrators

While the term integrator is loosely defined, it describes a protein molecule which is necessary to coordinate diverse signal transduction pathways by acting as a bridge between inducible transcription factors and the basal transcription machinery. CBP/p300 is the prototypical integrator protein. Because of the broad transcriptional flexibility of the Src-1 proteins, one group has proposed that they be regrouped into a class of integrator proteins like CBP/p300 (Lee et al., 1999; Lee et al., 1998b). The work described in this chapter would support such a move by adding Bm-3 transcription factors, a group of powerful developmental regulators and their DNA response element

180 to the catalogue of transcription factors and promoters manipulated by the p i60 coactivators.

III.3.6 Further experiments to demonstrate the involvement of Src proteins as coregulatory proteins for the Brn-3 transcription factors

Several experiments could further explore, confirm, and define the nature of the Src Bm- 3 interaction. Physical mapping experiments would give clues regarding the exact nature of the physical interaction between the two proteins. Data from the pull-down (Section 111.2.1 and Figures III.1-III.3) experiments and the coimmunoprecipitations (Section 111.2.1 and Figure III.4) indicate that the interaction between the Bm-3 proteins and the Src proteins occurs through the POU domain of the Bm -3 proteins. Further mapping of the domains of the steroid receptor coactivator domains required for this interaction is necessary. Affinity chromatography pull-down experiments using overlapping fragments representing segments of the Src-1 protein would help define the areas of this protein necessary for the Bm-3 interaction. Mutational studies concentrating on the specific areas of the interaction could then identify the specific residues in each protein necessary for the interaction. Further fimctional studies would help to conclusively demonstrate the involvement of Src-1 as a coactivator for Bm-3 proteins. Transient transfection experiments described herein have shown that Src proteins may coactivate Bm-3 dependent transcription. More transfections with dominant negative constructs of the steroid receptor coactivators would further confirm the results of the transfections using the wild-type constmcts. Electromobility shift assays in which the migration of Bm-3 proetins bound to radiolabelled Bm-3 response elements through a polyacrylamide could be retarded by the addition of Src proteins could demonstrate that the Src proteins interact with the Bm-3 proteins when bound to a specific Bm-3 response element. Finally, a most definite confirmation of the involvement of Src as a coregulator for Bm-3 proteins could be attained through a chromatin immunoprécipitation. In this experiment, the two proteins would be cross linked in the cell, coimmunoprecipitated on a specific Bm-3 responsive promoter, and the promoter site confirmed by PCR. Similarly, each of these experiments could be further extended to investigate a possible temary complex between

181 Bm-3 transcription factors, p i60 Srcs, and the estrogen receptor both on and off appropriate DNA response elements, thus revealing a broadened more flexible catalogue of functional interactions for each of the proteins.

III.3.7 Conclusion

Bm-3 transcription factors can functionally interact with p i60 coactivators and this interaction allows for tighter control over transcriptional regulation by each of these families of factors. In addition, this work broadens the catalogue of transcriptional activity of both families of proteins by producing evidence that both families can function outside of their originally defined areas of transcriptional activity. Finally, the functional interaction between these proteins results in promoter and cell-type specific transcriptional regulation. Different splice forms within the Src-1 group and a single member from the Src-3 group of p i60 proteins show activation, repression, or no effect when transfected with Bm-3 proteins. This complete array of effects would act as a powerful investigative tool to define the precise molecular environment the Bm-3 proteins require to act on a particular promoter. In addition, investigations of this type could reveal the cellular conditions which lead to different integrator complexes and the transcriptional activity of these complexes.

182 IV CELLULAR______GROWTH______AND______ENDOCRINE CHARACTERISTICS OF MCF7 CELLS WITH ALTERED LEVELS OF BRN-3B

IV. 1 Introduction

IV.1.1 pl60 proteins enhance estrogen stimulated MCF7 proliferation

The previous chapter identified clear physical and fimctional interactions between Bm-3 proteins and members of the p i60 family of steroid receptor coactivators. MCF7 breast adenocarcinoma cells stably expressing the p i60 coactivator Src-1 were produced by Tai and colleagues to study the action of 17 P-estradiol (E2) on cell growth as well as expression of E2 responsive genes (Tai, Kubota & Kato, 2000). In these experiments, Src-1 enhanced E2 stimulated growth as well as potentiated the E2 responsiveness of the pS2 gene whose expression has previously been shown to be ligand-bound estrogen receptor dependent (Brown et al., 1984).

IV. 1.2 The ER enhances estradiol stimulated MCF7 proliferation

The estrogen receptor has long been known to regulate the growth and differentiation of female reproductive organs (Endoh et al., 1999; Kato et al., 1995), play a role in the progression of human breast cancers (Dickson et al., 1987; Dickson & Lippman, 1987; Elledge & Osborne, 1997; Fuqua, 1992; Fuqua et al., 1992; Lippman et al., 1987), and consistently increase the E2 dependent proliferation of breast cancer cell lines which express it (Murphy & Sutherland, 1983; Reddel et al., 1983; Reddel & Sutherland, 1983; Sutherland et al., 1983a; Sutherland et ah, 1983b; Taylor et ah, 1983). Bm-3b has been shown to fimctionally interact with and potentiate the transcriptional effects of both ER (Budhram Mahadeo et ah, 1998) and p i60 proteins (described in the previous chapter).

183 IV. 1.3 Brn-3b is associated with the proliferation of neuronal cells

Bm-3 proteins are associated with the development of specific neuronal cells (Artinger et al, 1998; Fedtsova & Turner, 1997; Fedtsova & Turner, 1995). Bm-3a and Bm-3b were shown to be reciprocally expressed in at least three human neuroblastoma cell lines, and are associated with the balance between proliferation and differentiation of the ND7 and other neuronal cell lines (Latchman, 1999). Specifically, while Bm-3a activates promoters associated with neuronal differentiation, Bm-3b not only represses these same promoters, but also is able to negate the stimulatory effects of Bm-3 a on these promoters (Smith et al, 1997c). In addition, proliferating cell lines of neuroblastoma origin have a high ratio of Bm-3b to Bm-3a. Likewise, when these cell lines are induced to differentiate, the levels of Bm-3b fall and the levels of Bm-3a rise (Smith & Latchman, 1996). Finally, artificial over-expression of Bm-3 a induces neurite outgrowth in the ND7 cell line, and artificial over-expression of Bm-3b inhibits neuronal process outgrowth and neuronal specific gene activation thus preventing the differentiation of these cells (Smith et al., 1997c). These results suggest that a role for Bm-3b may be to maintain these cells in a proliferative state. The Bm-3 proteins are also present in other cancers. Bm-3a has been reported to be expressed in neuroepitheliomas and Ewings sarcomas (Collum et al., 1992). In addition, Bm-3 a has been shown to be overexpressed in aggressive neuroendocrine tumors (Leblond-Francillard a/., 1997).

IV. 1.4 Brn-3b may play a role in the regulation of BRCA I in breast cancer

In a separate set of experiments in this lab, Bm-3 a and Bm-3b levels were studied in primary breast sample tumors by RT-PCR and Westem blot immunodetection. These experiments showed that the levels of Bm-3b, but not the levels of Bm-3a, were enhanced in tumors, and that this correlated with reduced expression of the tumor suppressor gene BRCA I. Bm-3b expression was not increased in normal mammaiy

184 cells, benign tumors, or in malignant tumors without reduced levels of BRCA I (Budhram Mahadeo et al., 1999b). These results suggest that Bm-3b may play a role in the regulation of BRCA I in mammary tumorigenesis. Finally, Bm-3b was shown in a series of transient transfection assays in MCF7 cells to repress transcription of the luciferase gene from a promoter construct containing 4 kb of the BRCA I promoter (Budhram Mahadeo et al, 1999b).

Therefore: > Bm-3b can regulate estrogen receptor mediated as well as p i60 coactivator mediated gene expression, both of which have been shown to increase the proliferation of breast cancer cell lines; > Bm-3b is associated with maintaining neuronal cell lines in a proliferative state; > Bm-3b is expressed at greater than normal levels in breast tumors with decreased BRCA I expression; > Bm-3b has been shown to repress a 4 kb BRCA I promoter construct.

For these reasons, the effect of Bm-3b on the growth of the MCF7 adenocarcinoma cell line was investigated.

IV. 1.5 Altering specific protein levels in cell lines

For these experiments MCF7 cell lines stably expressing constructs that would either increase or decrease Bm-3b short expression were generated. Differences in the growth and endocrine characteristics were then investigated. Modification of the expression of a particular protein is a powerful tool for understanding its role in growth, development, differentiation and maintenance of homeostasis. Cell lines stably expressing the nucleic acid sequences of particular genes of interest are frequently used to study the regulatory role of proteins in a particular cellular context. Cell lines with altered levels of particular gene products may be created via homologous (Capecchi, 1989; Friedmann, 1989) or non-homologous (Scangos & Ruddle, 1981) recombination. This report describes altered gene expression of Bm-3b via non-

185 homologous recombination of a construct expressing the short form of Bm-3b in one case, and an antisense Bm-3b construct in the other. rV.l.5.A Experimental strategy for increasing the levels of Brn-3b short in the MCF7 cell line

A construct expressing the short form of Bm-3b under the control of the Moloney murine leukemia virus (MoMuLV) promoter was cotransfected in five fold excess with an empty pCiNeo construct into MCF7 cells using the calcium phosphate method of transfection (Gorman et al., 1982). Selection was carried out with the addition of Geneticin-G418 sulphate. Antibiotic resistant clones were further grown and analysed for increased production of the Bm-3b short gene product.

IV.1.5.B Experimental strategy for decreasing the levels of Brn-3b in the MCF7 cell line

Several antisense strategies exist for decreasing the production of a particular gene product in cell lines. The functional principle of antisense methods is based on DNA or RNA molecules binding a complementary region within the mRNA of a gene of interest, thereby blocking protein production. Three main methods to block protein production using antisense nucleic acids are currently in use. The first involves the use of antisense DNA oligonucleotides. In this method, DNA oligonucleotides are introduced into the cell by microinjection or via cellular uptake directly firom the medium (Colman, 1990). The effect of oligodeoxynucleotides is, however, transient. Another antisense strategy involves the use of mRNA-cleaving ribozymes (Colman, 1990). This method relies on mutating a 5’ four-base nucleotide sequence to bind sequences complementary to the mRNA that is to be cleaved. Because the specificity for this reaction is provided by only four bases, mismatches are not uncommon (Colman, 1990). Moreover the catalytic activity of ribozymes varies on different target sequences (Cantor et al., 1996). For antisense effects that are less transient than oligodeoxynucleotide treatments and more specific than ribozyme treatments, expression vectors encoding antisense DNA which is

186 the template for the synthesis of antisense mRNA in vivo are used (Colman, 1990). The mechanisms by which antisense mRNA achieves reduced protein levels, however are still unclear (Denhardt, 1992a; Denhardt, 1992b). Denhardt reviewed several hypotheses for decreased protein expression by antisense RNA including the formation of RNA duplexes which: block translation, enhance mRNA degradation, prevent mRNA translocation from the nucleus to the cytoplasm, or block the processing of the primary RNA transcript (Denhardt, 1992a; Denhardt, 1992b).

The work described herein utilised a eukaryotic expression vector encoding antisense cDNA for the first 300 bases of Bm-3b short. The first 300 nucleotides of Bm-3b short will not cross-hybridize with Bm-3a or Bm-3c mRNA. A pJ4 construct expressing the first 300 antisense bases of the short form of Bm-3b under the control of the Moloney murine leukemia virus (MoMuLV) promoter was cotransfected in five fold excess with an empty pCiNeo construct into MCF7 cells using the calcium phosphate method of transfection (Gorman et al, 1982). Selection was carried out with the addition of Geneticin-G418 sulphate. Antibiotic resistant clones were further grown and analysed for increased production of the Bm-3b short gene product.

187 IV.2 Results

IV.2.1 Establishment of stable transformants

To establish stable cell transformants with enhanced and reduced levels of Bm-3b, the human breast adenocarcinoma cell line MCF-7 was selected. This cell line is well known to be estrogen responsive, and previous work in the lab has shown that MCF-7 cells express Bm-3b (Budhram-Mahadeo-Heads, 1995). The cells were transfected with the short form of Bm-3b in pLTR, empty pLTR, or pJ4 expressing the first 300 antisense bases of Bm-3b. All transfections included pCiNeo at a five fold reduced concentration, compared to the expression constructs, for selection of individual clones with antibiotic resistance Proliferating MCF-7 cells were transfected using the calcium phosphate method (Gorman et al., 1982). Twelve independent G-418-resistant clones were selected form each transfection and grown up for further analysis.

IV.2.2 Confirmation of stable transformants by RT-PCR

Eleven of the twelve selected antisense clones grew up. RT-PCR with a set of primers corresponding to the polylinker of pJ4 and a region within the antisense construct (and therefore specific for the exogenous stably incorporated pJ4 construct and not endogenous Bm-3b) were used to confirm the expression of Bm-3b antisense transcripts (Figure IV. 1). Two of the clones appeared to express the construct at high levels. These two independent clones were grown up for further analysis.

Ten of the twelve Bm-3b overexpressing clones were viable and each was tested by RT- PCR with a set of primers corresponding to the short mouse Bm-3b transcript (Figure IV.2). Because the primer pairs used in this analysis could amplify both the endogenous and ectopically expressed Bm-3b, cycle numbers were kept low (18 cycles) to identify clones which overexpressed the most Bm-3b transcript. Two of these clones expressed the transcript at high levels and were grown up for further analysis.

188 CN a < < < OQ CJ u u

Brn-3b antisense-

Figure IV. 1: RT-PCR analysis of antisense Brn-3b in clonal MCF7 cell lines transfected with Brn-3b antisense construct. RT-PCR to detect overexpression of Bm-3b antisense transcript in nine clones: two transfected with empty pLTR vector controls, and seven transfected with the pJ4 Bm-3b anti sense construct. RNA was extracted from each of the clones, equalised for concentration by spectrophotometry, and used as a template for preparing cDNA. Because the forward primer annealed only to the polylinker region of pJ4, only clones stably expressing the pJ4 antisense construct produced the proper PCR product. Two clones gave strong bands: Al and A2, and these clones were used in all subsequent analysis.

189 »— I fNJ < < I —) h»—3 hI—] h J hH-) J Ç ? o Q h a , O h a . D h O h O h O h (N ri Tf c\imTfir)xr-ooo\

Brn-3b

Figure IV.2: RT-PCR analysis of antisense Brn-3b in clonal MCF7 cell lines transfected with Brn-3b short construct. RT-PCR to detect overexpression of Bm-3b transcript in ten clones: two transfected with empty pLTR vector controls, six transfected with the pLTR Bm-3b construct, and two transfected with the pJ4 Bm-3b antisense construct. RNA was extracted from each of the clones, equalised for concentration by spectrophotometry, and used as a template for preparing cDNA. Because the primer pairs could amplify both endogenous and exogenous Bm-3b, PCR cycle numbers were kept low (18 cycles) to identify the clones which overexpressed the most Bm-3b transcript. Two clones gave the strongest product bands: Z and Y, and these clones were used in all subsequent analysis.

190 IV.2.3 Confirmation of stable transformants by Western immunoblotting Having shown that the cell lines expressed the appropriate RNA transcripts, it was next determined whether overexpression of Bm-3b short led to a rise in Bm-3b levels, and likewise, if the levels of expression of the antisense construct led to a decrease in endogenous Bm-3b. 80% confluent samples of each clonally selected cell line were harvested in cell lysis buffer. 60 micrograms of total cell protein was run on a SDS/15% polyacrylamide gel, blotted, and incubated with a Bm-3b specific antibody. The overexpressing clones showed not only an increase in the short form of Bm-3b (33 kDa) but surprisingly also a smaller increase in the endogenous long form of Bm-3b (46 kDa). The antisense clones showed a reduction in the endogenous levels of both long and short forms of Bm-3b. The results are summarized in Figure IV.3.

191 Figure IV.3: Western blot analysis of Brn-3b in clonal MCF7 cell lines with altered levels of Brn-3b. (Facing page) Immunoblots to detect the levels of both isoforms of Bm-3b were carried out using six clones: two overexpressing Bm-3b (Z and Y), two empty pLTR vector controls (B and C), and two overexpressing the pJ4 Bm-3b antisense construct (Al and A2). Total cellular protein (60|ig per lane) was loaded onto a SDS/15% polyacrylamide gel, transferred to nitrocellulose, and probed with an anti-Bm-3b antibody. The level of Bm- 3b long antigen in extract from the pLTR Bm-3b overexpressing clones was 2.6-3.3 times the average endogenous level seen in the vector controls, and the level of Bm-3b long antigen in extracts from the pJ4 Bm-3b antisense clones was 0.2-0.4 times the average endogenous level seen in the vector controls. The level of Bm-3b short antigen in the extract from the pLTR Bm-3b overexpressing clones was 10.6-11.9 times the average endogenous level seen in the vector controls, and the level of Bm-3b short antigen in extracts from the pJ4 Bm-3b antisense clones was not detectable.

192 dLTR Bm-3b overexoressina dLTR emotv dJ4 Bm-3b antisense

B Al A2

Brn-3b long

Brn-3b short

actin

CLONE DENSITOMETRY(arbitrary units) FOLD EXPRESSION Brn-3b long Brn-3b short Brn-3b long Brn-3b short pLTR Brn-3b short Z 31440 41149 2.6 10.6 pLTR Brn-3b short Y 39539 45958 3.3 11.9 pLTR B 8062 3670 1.0 1.0 pLTR C 15271 4072 1.0 1.0 pJ4 Brn-3b antisense A1 1763 0 0.2 0.0 pJ4 Brn-3b antisense A2 4919 0 0.4 0.0

193 IV.2.4 Effects on cell growth in monolayers

The molecular mechanisms which control cellular growth are not clearly understood. Growth can be altered by changing proliferation, apoptosis, or differentiation rates. In normal tissue, cell number remains constant because of tight control over these three factors. In abnormal tissue, this balance can be deregulated such that increased cell number can result from either blocked death and/or differentiation pathways, or by an increase in proliferation without a commensurate increase in apoptosis or differentiation. Both of these deregulated pathways play a role in carcinogenesis.

As mentioned in the introduction to this chapter, Bm-3b is associated with neuronal proliferation. Moreover, Bm-3b can interact with the ER to potentiate the transcription effects of the ER, and the ER enhances estradiol stimulated MCF7 adenocarcinoma cell proliferation. Thus, changes in the growth of the MCF7 adenocarcinoma cell lines with altered levels of Bm-3b were investigated.

IV.2.4.A Growth curve study

Alterations in the levels of Bm-3b in MCF-7 cell lines has a significant effect on both population doubling time as well as cellular density at plateau phase. The Bm-3b short overexpressing, vector control, and Bm-3b antisense clones were studied to determine effects on cell growth. The effects of altered levels of Bm-3b is shown in figure IV.4. The clone “Z” overexpressing Bm-3b reached a plateau at an average density of 7.66 +/- 1.66 xlO^ cells per cm? after 14 days. The clone “Y” overexpressing Bm-3b reached a plateau at an average density of 7.43 +/- 1.66 xlO^ cells per cm? after 14 days. The pLTR control clone “B” reached a plateau at an average density of 4.17 +/- 0.39 xlO^ cells per cm^ after 12 days. The pLTR control clone “C” reached a plateau at an average density of 4.25 +/- 0.22 xlO^ cells per cnf after 12 days. The clone “A l” overexpressing the antisense construct of Bm-3b reached a plateau at an average density of 2.92 +/- 0.36 xlO^ cells per cnf after 10 days. The clone “A2” overexpressing the antisense construct

194 of Bm-3b reached a plateau at an average density of 2,12 +/- 0.24 xlO^ cells per cnf after 10 days. Using the chi squared statistical test, the individual cell counts for the Bm-3b overexpressing and Bm-3b antisense cell lines were compared to the pLTR control cell line cell counts at each time point. All time points after the initial lag phase of growth were statistically significant fi*om the expected growth rate gauged by the pLTR controls. The Bm-3b overexpressing cell lines grew significantly faster, and the Bm-3b antisense significantly slower than the pLTR control cell lines (all p-values < 0.05) (Figure IV.5 and Table IV. 1)

195 Growth curves of MCF7 stable clones

10

1

“o pJ4 Brn-3b antisense Al pJ4 Bm-3b antisense A2

È pLTR control B pLTR control C C pLTR Brn-3b short Z pLTR Bm-3b short Y U

*

0 24 48 72 96 120 144 166 192 216 240 264 288 312 336 360 384 408 432 456 480 Time (hours) Figure IV.4: Cellular growth curves of clonal MCF7 cell lines with altered levels of Brn-3b. Cells were plated out in six-well plates and grown in full growth medium. The number of cells at each time point represents the mean of three independent experiments +/- the standard deviation of the mean. denotes p-value less than 0.0001 resulting from a chi-squared test comparing all values at the indicated time point from the experimental (Bm-3b short) with the expected (pLTR control). denotes p-value less than 0.05 resulting from a chi-squared test comparing all values at the indicated time point from the experimental (Bm-3b antisense) with the expected (pLTR control).

196 IV.2.4 .B Saturation density limitation of growth study

In addition to the significant differences in each of the representative cell lines’ growth, the differences in cell density at the plateau phase of the cellular growth curves are also similarly different. A saturation density limitation growth study was carried out as described in the Materials and Methods Section to exclude the possibility that the different saturation densities observed were due to slow growth or exhaustion of medium. For this study 1 xlO^ cells were plated into 24 well plates containing a 13mm coverslip in each well. After 24 hours, the confluent coverslips were transferred to six well dishes containing 4 mis of medium which was changed every 48 hours. Starting at day 10 cells were trypsinized and counted every two days for eight more days. Three coverslips of each clone were used for each timepoint. The saturation densities of each clone were similar to the plateau densities for each clone shown in the growth curves. Using the chi squared statistical test, the individual cell counts for the Bm-3b overexpressing and Bm- 3b antisense cell lines were compared to the pLTR control cell line cell counts at each time point. The Bm-3b overexpressing cell lines grew to significantly higher densities (p-values < 0.00001) at all of the time points, and the Bm-3b antisense to significantly lower than the pLTR control cell lines (p-values < 0.05) in three out of four of the time points. In summary, the Bm-3b overexpressing clones had saturation densities over twice that of the control and antisense cell lines. The control cell lines generally grew to higher densities than the antisense cell lines, although these differences were slight. (Figure 1V.6 and Table IV.l).

197 Saturation cell density of MCF7 stable clones

12 1

1 0 - 1

*0 ■pJ4 Bm-3b antisense A pJ4 Bm-3b antisense A: I pLTR control C I pLTR control B § pLTR Bm-3b short Z pLTR Bm-3b short Y U

*

0 24 48 72 96 120 144 168 192 216 240 264 288 312 336 360 384 408 432 Time (hours) Figure IV.5: Saturation density of clonal MCF7 cell lines with altered levels of Brn-3b. Cells were plated out in six-well plates and grown in full growth medium. The number of cells at each time point represents the mean of three independent experiments +/- the standard deviation of the mean. “★ “ denotes p-value less than 0.0001 resulting from a chi-squared test comparing all values at the indicated time point from the experimental (Bm-3b short) with the expected (pLTR control). denotes p-value less than 0.05 resulting from a chi-squared test comparing all values at the indicated time point from the experimental (Bm-3b antisense) with the expected (pLTR control).

198 IV.2.4.C Tritiated thymidine incorporation study

To investigate if the growth differences seen in the cell lines were the result of differences in the rate of proliferation in the cell lines, incorporation of tritiated thymidine in the replicating cells was used as a measure of proliferative activity. Cells were incubated with tritated thymidine for one hour, then harvested and counted on a scintillation counter (Figure IV.6). It is clear from these studies that these cell lines proliferate at a rate commensurate with the level of Bm-3b protein, with the Bm-3b overexpressing cell lines consistently proliferating more than, and the Bm-3b antisense cell lines consistently proliferating less than the pLTR control cell lines.

IV.2.4.D Effects on cellular morphology

Morphological examination of the clonal cell lines revealed changes in the shape of the cells. 5-10% of the Bm-3b overexpressing cells had a severely altered shape. These cells appeared to adhere to other MCF7 cells but were never seen adhering to tissue culture plastic. These cells were 1-5 times larger the size of normal cells. These large cells had a vacuolar-type appearance with the cellular machinery appearing as a crescent in the side of the cell. Photomicrographs of these cells at lOx, 20x, and 40x are shown in Figures IV.7-IV.9) MCF7 cells transfected with the pLTR empty control plasmid had an appearance which did not vary from wild-type MCF7 cells. (Figures IV. 10) The MCF7 cells stably expressing the pJ4 Bm-3b antisense construct showed no changes in cellular morphology. Some cells at the edge of colonies, however, appeared to be unhealthy. (Figure IV. 11)

199 Tritiated thymidine incorporation of MCF7 stable clones

4500

4000- *

3500 T

3000-

O h 2500-

*

§ 2 0 0 0 - o JÎ • • •• VT- ‘tC.t." 1500- ’ I* »'k. • *

1 0 0 0 -

. a # 500- . @ 0 ♦ «T V-' .5;

A1 A2 MCF7 stable clone Figure IV.6: Tritiated thymidine incorporation by clonal MCF7 cell lines with altered levels of Brn-3b. Cells were grown in full growth medium for 48 hours and subsequently treated with tritiated thymidine. After one hour stimulation, cells were trypsinized, harvested onto glass filters, and counts per minute from the glass filters were recorded by a scintillation counter. The counts per minute from each cell line represent the mean of three independent experiments counted in triplicate +/- the standard deviation of the mean. denotes p-value less than 0.05 resulting from a chi-squared test comparing all values at the indicated time point from the experimental (Bm-3b short or Bm-3b antisense) with the expected (pLTR control).

200 - #P * ' ÎSî-ïSK; f t* 2 ^fmm/fMè^ g

Figure IV.7: Photomicrograph (10x) MCF7 cells stably overexpressing pLTR Brn- 3b short construct grown in monolayer. Ceils were plated in full growth medium and allowed to grow for three days at which point the medium was replaced with fresh medium. This photomicrograph was taken on the fifth day post-plating.

201 #

i

Figure IV.8: Photomicrograph (20x) MCF7 cells stably overexpressing pLTR Bm- 3b short construct grown in monolayer. Cells were plated in full growth medium and allowed to grow for three days at which point the medium was replaced with fresh medium. This photomicrograph was taken on the fifth day post-plating. “| y indicates vacuolar-type cells.

202 Figure IV.9: Photomicrograph (40x) MCF7 cells stably overexpressing pLTR Bm- 3b short construct grown in monolayer. Cells were plated in full growth medium and allowed to grow for three days at which point the medium was replaced with fresh medium. This photomicrograph was taken on the fifth day post-plating. “i indicates vacuolar-type cell.

203 &

m

#

Figure IV. 10: Photomicrograph (20x) MCF7 cells stably overexpressing pLTR empty control construct grown in monolayer. Cells were plated in full growth medium and allowed to grow for three days at which point the medium was replaced with fresh medium. This photomicrograph was taken on the fifth day post-plating.

204 Figure IV. 11: Photomicrograph (20x) MCF7 cells stably overexpressing pJ4 Brn- 3b antisense construct grown in monolayer. Cells were plated in full growth medium and allowed to grow for three days at which point the medium was replaced with fresh medium. This photomicrograph was taken on the fifth day post-plating.

205 IV.2.4.E Effects on growth in soft agar

The ability of cells to exhibit anchorage-independent growth is a hallmark of the transformed cell phenotype. We further tested the properties of these clones by measuring the ability of the clones to form colonies in a soft agar assay. Differences were observed in this experiment in both the number and size of colonies formed (summarized in Table IV. 1). In full growth medium containing 0.3% agarose the Bm-3b overexpressing and pLTR empty control vector cell lines had 229 +/- 44 and 188 +/- 39 colonies respectively. In general, the Bm-3b overexpressing colonies were consistently larger than the pLTR control colonies(Figure IV. 13). The Bm-3b antisense cells produced 51 +/- 9 colonies that were consistently smaller in size than the overexpressing or control colonies (Figure IV. 15). These results are summarized in Figure IV. 12. The colonies of the Bm-3b overexpressing cell line continued to produce the large vacuolar cells initially observed in the cells grown in monolayer (Figure IV. 13).

206 Anchorage independent growrh of MCF7 stable clones

200 -

ID, Ü Qh 150 . I u

MCF7 stable clones

Figure IV. 12: Anchorage independent colony formation by clonal MCF7 cell lines with altered levels of Brn-3b. Cells were plated in full growth 0.3% agarose medium. After 21 days, colonies comprised of at least 32 cells were counted. The number of colonies from each cell line represent the mean of three independent experiments counted in triplicate +/- the standard deviation of the mean. denotes p-value less than 0.05 resulting from a chi-squared test comparing all values at the indicated time point from the experimental (Bm-3b short or Bm-3b antisense) with the expected (pLTR control).

207 SB

Figure IV. 13: Photomicrograph (20x) of anchorage independent colony formation by MCF7 cells stably overexpressing pLTR Brn-3b short construct. Cells were plated in full growth 0.3% agarose medium. After 21 days, colonies comprised of at least 32 cells were counted. This photomicrograph was taken on the 24*^ day post plating.

208 4#

Figure IV. 14: Photomicrograph (20x) of anchorage independent colony formation by MCF7 cells stably overexpressing empty pLTR control construct. Cells were plated in full growth 0.3% agarose medium. After 21 days, colonies comprised of at least 32 cells were counted. This photomicrograph was taken on the 24*^ day post plating.

209 -it i. 4 ^ tjV \ ^ H.-L % m ■^r»V f V J i ^

Figure IV. 15: Photomicrograph (20x) of anchorage independent colony formation by MCF7 cells stably overexpressing pJ4 Brn-3b antisense construct. Cells were plated in full growth 0.3% agarose medium. After 21 days, colonies comprised of at least 32 cells were counted. This photomicrograph was taken on the 24^*^ day post plating.

210 Growth parameters of MCF7 stable clones

Clone Plateau phase Saturation Density Anchorage-independent growth days to cell number cell number colonies colonies (xIO® per cm (xIO® per cm^) (number per plate) (size)

3b01 14 7.66 +/-1.66 8.63 +/- 0.89 229 +/- 44 M

3b02 14 7.43 +/-1.66 8.81 +/-1.08

V1 12 4.17+/-0.39 3.95 +/- 0.82 188 +/- 39 S-M

V2 12 4.25 +/- 0.22 4.31 +/- 0.520

3bAS1 10 2.92 +/- 0.36 3.83 +/- 0.81 51 +/- 9 S

3bAS2 10 2.12+/-0.24 2.64 +/- 0.84

Table IV. 1: Cellular growth parameters of clonal MCF7 cell lines with altered levels of Brn-3b. The growth parameters of the pLTR Bm-3b short, pLTR control, and pJ4 Bm-3b antisense stable clones are described above. The cell density at plateau was determined using the cell number at the day 14 time point (the first time point at which all cell lines were in plateau phase) of the growth curve. The saturation density was determined from the mean of all four time points. Colonies in soft agar were counted directly using an inverted microscope. Relative colony size indicated by S (small), and M (medium). All values are the mean of at least triplicate counts +/- the standard deviation of the mean.

211 IV.2.5 Endocrine characteristics of the MCE stable clones.

IV.2.5.A Effects of estrogen and estrogen analogues on MCF7 cells

Because MCF-7 cells are well-documented to be estrogen responsive, and it has been shown in this lab that Bm-3b functionally interacts with the estrogen receptor, it was necessary to see if the altered levels of Bm-3b mediated their effects through estrogen mediated pathways. In general, estrogen stimulation of MCF7 cell growth is associated with an increase in growth fraction, a shortening of cell cycle time, and a decrease in cell death (Sutherland et ah, 1983b). Tamoxifen treatment of MCF7 cells causes an accumulation of cells in GO/Gl phase in the cell cycle with a commensurate decrease of cells in S phase (Sutherland et a/., 1983a). The effect of the concentration of the tamoxifen is important: at sub-5 pM the growth inhibition by tamoxifen is reversible with the addition of estrogen, but at concentrations greater than 5 pM tamoxifen has a cytotoxic effect. In order to elucidate the pathways through which the altered levels of Bm-3b may produce their effect, the effect of treatment with 17 p-estradiol (E2), and tamoxifen was tested on each of the MCF7 clonal cell lines. Changes in growth in monolayers, proliferation, and anchorage independent growth were monitored.

IV.2.5.B Effects on growth

To investigate whether the alterations in growth rate of the MCF7 cells were in part a result of alterations in estrogen responsive pathways, cells were plated and grown in the presence of estradiol, vehicle control, or tamoxifen in full growth medium as well as in medium without phenol red containing charcoal stripped serum. The addition of E2 to full growth medium increased the growth of all the MCF7 clonal cell lines. The Bm-3b antisense cell lines appeared to be the most affected by the addition of the E2, with greater than 250% increase in cell number. The estradiol mediated effects on the Bm-3b overexpressing cell,lines were similar to those for the pLTR control cell lines: around a 150% increase in cell number. (Figure IV. 16)

212 The addition of tamoxifen to the full growth medium resulted in a consistent and severe decrease in cell number when compared to the same cells treated with ethanol vehicle alone. Interestingly, however, the Bm-3b antisense cell lines were the least affected by the addition of tamoxifen to the culture medium losing only 40% of the cell number. This was most similar to the Bm-3b overexpressing cell lines which lost approximately 50% of their cell number. The pLTR control cells were the most extremely affected by the addition of tamoxifen, retaining an average of only 24% of their cells. (Figure IV. 16)

This same experiment was repeated in medium without phenol red prepared with dextran charcoal stripped serum. In these experiments, the addition of E2 to the stripped medium had dramatic effects on the growth of both the Bm-3b overexpressing and the Bm-3 antisense cell lines with an approximate increase in cell number in both groups of over 300% when compared to the same cells treated with vehicle. The addition of E2 to the medium of the control cell lines resulted in 150% growth stimulation only half as high as that seen in the experimental cell lines. (Figure IV. 17) The addition of tamoxifen to the stripped serum medium completely killed the Bm-3b antisense cell lines, while the Bm-3b overexpressing cell lines were unaffected by this treatment. The control cell lines lost an average of 35% of their cells when treated with tamoxifen. (Figure IV. 17)

213 Endocrine characteristics of MCF7 stable clones

E ü i_ Q.(D m O

MFGM + E2 0) HFGM + V □ FGM +TAIV E 3 C 0 O

F C MCF7 stable clones

n d LTR control 177 171 417 147 173 264 100 100 100 100 100 100 34 18 46 30 16 44 vehicle control : 146 100 57 120 100 36 160 100 17 122 100 30 275 100 61 264 100 64

Figure IV. 16: Effect of estradiol and tamoxifen, as determined by cell counts, on the growth of clonal MCF7 cells with altered levels of Brn-3b. Cells were plated in full growth medium and allowed to grow for two days and subsequently treated with estradiol, ethanol vehicle, or tamoxifen at a concentration of 200ng/ml. Cells were trypsinized and counted on a haemocytometer at the end of five days stimulation. The number of cells at each time point represents the mean of three independent experiments +/- the standard deviation of the mean.

214 Endocrine characteristics of MCF7 stable clones

OJ E o 1— 0) CL ID o □ SS + E2 X ü S S + v 1— □ SS + TAIV CD E 3 C 05 O

■ÉÜK am F C MCF7 stable clones

an d LTR control 306 127 225 279 123 230 100 100 100 100 100 100 25 5 0 13 6 0 vehicle control 336 100 103 318 100 190 118 100 47 175 100 77 682 100 0 264 100 0

Figure IV. 17: Effect of estradiol and tamoxifen, as determined by cell counts, on the growth of clonal MCF7 cells with altered levels of Brn-3b. Cells were plated in full growth medium which was replaced after 24 hours with estradiol-ffee medium (containing dextran charcoal-stripped serum). The cells were allowed to grow for a further two days and subsequently treated with estradiol, ethanol vehicle, or tamoxifen at a concentration of 200ng/ml. Cells were trypsinized and counted on a haemocytometer at the end of five days stimulation. The number of cells at each time point represents the mean of three independent experiments +/- the standard deviation of the mean.

215 IV.2.5.C Effects on cellular proliferation

To investigate if the growth differences seen in the cell lines were the result of differences in the rate of proliferation in the cell lines, incorporation of tritiated thymidine in the replicating cells was used as a measure of proliferative activity. Cells were incubated with tritated thymidine for one hour, then harvested and counted on a scinitllation counter.

The proliferative effects of the addition of estradiol or tamoxifen to the full growth medium of the clonal cell lines generally paralleled the results of the growth studies reviewed in the previous section. In full growth medium, the addition of estradiol had the strongest effect on the proliferation of the pLTR control and pJ4 Bm-3b antisense cell lines, which increased their proliferative activity by approximately 150% when compared to the vehicle alone control. Little effect was observed with the addition of estradiol to the pLTR Bm-3b overexpressing clones. The addition of tamoxifen to the full growth medium resulted in a decrease in the proliferation of all the cell lines with the antisense cell lines most severely affected. (Figure IV. 18)

This experiment was duplicated in stripped medium and analysed after two days of treatment (Figure IV. 19) and again after five days of treatment (Figure IV.20). After two days of treatment, the proliferation results paralled the cell count results: the addition of E2 appeared to stimulate the proliferation of the overexpressing and antisense cell lines by 150%, with less of an effect on the control cell lines, when each was compared with the same cells treated with ethanol vehicle. The addition of tamoxifen to each of the cell lines resulted in a consistent strong decrease in proliferation when compared to cells treated with vehicle alone. The proliferation of the Bm-3b overexpressing cells was, however, the least affected by the addition of tamoxifen decreasing proliferative activity by only 50% rather than by 70% as in the the other cell lines. (Figure IV.20)

216 After five days in the stripped medium, changes in the proliferation resulting from the addition of E2 or tamoxifen were more profound. In the presence of E2, the cells overexpressing Bm-3b were almost four times more proliferative than the same cell treated with vehicle, as was the case for the Bm-3b antisense cells. Likewise, the addition of tamoxifen resulted in the severe decreases in proliferation in all cell lines when compared to vehicle alone treatments; and again, the Bm-3b overexpressing cell lines was the least affected by this treatment, maintaining 25% of their proliferative activity. (Figure IV.20)

217 Endocrine characteristics of MCF7 stable clones

4500

4000-

3500

1500-

1000 -

500

F C MCF7 stable clones

in p LTR control 138 198 214 126 164 187 100 100 100 100 100 100 74 72 58 48 49 31 vehicle control : 101 100 53 111 100 56 146 100 56 148 100 42 150 100 39 142 100 31

Figure IV. 18: Effect of estradiol and tamoxifen, as determined by tritiated thymidine incorporation, on the proliferation of clonal MCF7 cells with altered levels of Brn-3b. Cells were plated in full growth medium and allowed to grow for two days and subsequently treated with estradiol, ethanol vehicle, or tamoxifen at a concentration of 200ng/ml. Cells were pulsed with tritiated thymidine for one hour, trypsinized, harvested onto glass fiber membranes, and counted on a scintillation counter at the end of five days stimulation. The number of cells at each time point represents the mean of three independent experiments +/- the standard deviation of the mean.

218 Endocrine characteristics of MCF7 stable clones

6000-,

^S S + E2 O. 3000- BSS + V □ SS +TAIV

2000 -

1000 -

F C MCF7 stable clones

b ean pLTRcontrol 205 106 152 158 99 143 100 100 100 100 100 100 96 54 50 64 40 33

% vehicle control ^gQ -|qq 149 100 49 102 100 32 84 100 36 164 100 31 150 100 28

Figure IV. 19: Effect of estradiol and tamoxifen, as determined by tritiated thymidine incorporation, on the proliferation of clonal MCF7 cells with altered levels of Bm-3b. Cells were plated in full growth medium which was replaced after 24 hours with estradiol-free medium (containing dextran charcoal-stripped serum). The cells were allowed to grow for a further two days and subsequently treated with estradiol, ethanol vehicle, or tamoxifen at a concentration of 200ng/ml. Cells were pulsed with tritiated thymidine for one hour, trypsinized , harvested onto glass fiber membranes, and counted on a scintillation counter at the end of two days stimulation. The number of cells at each time point represents the mean of three independent experiments +/- the standard deviation of the mean.

219 Endocrine characteristics of MCF7 stable clones

4000-,

3500

3000-

2500-

raSS-t-E2 0 Q. 2000 - (/) □ SS-HAIV "c U o 1500- O

1000 -

500-

F C A1 A2 MCF7 stable clones

^eanpLTRcomrol 212 153 826 100 100 100 100 100 100 77 42 89 49 45 268 % vehicle control 398 100 29 377 100 21 252 100 6 297 100 0 504 100 8 300 100 24

Figure IV.20: Effect of estradiol and tamoxifen, as determined by tritiated thymidine incorporation, on the proliferation of clonal MCF7 cells with altered levels of Brn-3b. Cells were plated in fiill growth medium which was replaced after 24 hours with estradiol-free medium (containing dextran charcoal-stripped serum). The cells were allowed to grow for a further two days and subsequently treated with estradiol, ethanol vehicle, or tamoxifen at a concentration of 200ng/ml. Cells were pulsed with tritiated thymidine for one hour, trypsinized , harvested onto glass fiber membranes, and counted on a scintillation counter at the end of five days stimulation. The number of cells at each time point represents the mean of three independent experiments +/- the standard deviation of the mean.

220 IV.2.5.D Effects on anchorage-independent growth

The effects of E2, and tamoxifen were tested on each of the clonal cell lines to look for differences in the ability to form anchorage-independent colonies. Cells were grown in full growth medium as well as in stripped medium.

In full growth medium, treatment of the Bm-3b overexpressing cells with E2 had little effect on the number of anchorage-independent colonies formed. Estradiol, however, had the greatest effect on increasing the number of anchorage independent colonies formed by the pJ4 Bm-3b antisense cell line with an approximate 340% increase in number of colonies formed. The results with the addition of tamoxifen to the pLTR control cell line were unsurprising, a 50% decrease in the number of cells capable of forming anchorage independent colonies, this treatment had no effect on the ability of Bm-3b overexpressing cells to form anchorage-independent colonies. Quite surprisingly, the antisense cells treated with tamoxifen formed approximately two fold more colonies in the presence of tamoxifen when compared to cells grown in vehicle alone (Table 2 and Figure 19).

The results in stripped serum were similar. The addition of E2 to each of the cell lines consistently resulted in an approximate two-fold increase in the number of anchorage- independent colonies formed. Again, quite suprisingly, the Bm-3b overexpressing and antisense cell lines, however, both had more cells that could form anchorage-independent colonies under tamoxifen treatment when compared to the vehicle alone control. The pLTR control cells responded to the tamoxifen with a 40% decrease in the number of cells capable of forming anchorage independent colonies.

221 Endocrine characteristics of MCF7 stable

350

300

250 m m IQ. i_ 0 200 Q. SFG M + E2 0 B FGM + V "c o □ FGM + TAM o 150 O

B A1 MCF7 stable clones

% mean oLTR control 94 122 244 100 100 100 64 27 112 % vehicle control 1 0 0 1 0 1 145 100 51 343 100 208

Figure IV.21: Effect of estradiol and tamoxifen on the anchorage independent growth of clonal MCF7 cells with altered levels of Brn-3b. Cells were plated in full growth 0.3% agarose medium with the stated endocrine treatments. After 21 days colonies comprised of at least 32 cells were counted. The number of cells at each time point represents the mean of three independent experiments +/- the standard deviation of the mean.

222 Endocrine characteristics of MCF7 cells

350

300

250 - 0 Q.

0 - Q. 200 (/) □ SS + E2 0 □ SS + V "c o □ SS + TAM Ô O

100 -

50 -

MW

A1 MCF7 stable clone

% mean pLTR control 118 113 308 100 100 100 57 37 106 % vehicle control 184 100 166 176 100 61 268 100 174

Figure IV.22: Effect of estradiol and tamoxifen on the anchorage independent growth of clonal MCF7 cells with altered levels of Brn-3b. Cells were plated in phenol redless 0.3% agarose medium prepared with dextran charcoal stripped serum and the stated endocrine treatments. After 21 days colonies comprised of at least 32 cells were counted. The number of cells at each time point represents the mean of three independent experiments +/- the standard deviation of the mean.

223 Endocrine characteristics of MCF stable clone anchorage-independent growth

Treatment FGM + Vehicle FGM + E2 FGM + Tamoxifen C o lo n i e s Number Size Number Size Number Size C lo n e

Z 229 +/- 44 S-M 257 +/- 53 L 232 +/- 53 S

B 188 +/- 39 -M 273 +/- 51 M-L 95+/-21 S

A1 51 +/-9 S 175 +/- 39 S-M 106+/-18 S

Endocrine characteristics of MCF stable clone anchorage-independent growth

Treatment SS + Vehicle SS + E2 SS + Tamoxifen C o lo n i e s Number Size Number Size Number Size C lo n e

Z 143 +/- 30 S-M 263 +/- 45 L 237 +/- 53 S

B 127 +/- 24 S-M 223 +/- 54 M 77+/- 17 S

A1 47+/-8 S 126 +/- 28 M 82+/- 16 S

Table IV.2: Anchorage independent growth parameters of clonal MCF7 cell lines with altered levels of Brn-3b. The growth parameters of the pLTR Bm-3b short, pLTR control, and pJ4 Bm-3b antisense stable clones are described above. Colonies in soft agar comprised of at least 32 cells were counted directly using an inverted microscope. Relative colony size indicated by S (small), and M (medium). All values are the mean of at least triplicate counts +/- the standard deviation of the mean.

224 IV.3 Discussion

As mentioned in the introduction to this chapter, Bm-3b is associated with neuronal proliferation. Moreover, Bm-3b can interact with the ER to potentiate the transcriptional effects of the ER, and the ER enhances estradiol-stimulated MCF7 cell proliferation. Thus, changes in the growth of MCF7 adenocarcinoma cell lines with altered levels of Bm-3b were investigated.

This chapter has clearly shown that the altered levels of Bm-3b have drastic effects on the growth characteristics of the MCF7 adenocarcinoma cell line. The success of the construct used to overexpress exogenous Bm-3b short as well as the construct used to decrease the levels of endogenous Bm-3b long and Bm-3b short was shown both at the RNA and protein levels. These increased and decreased levels were then shown to have significant effects on the growth characteristics of the cells grown in monolayers as well as cells grown as anchorage independent colonies. The cells overxpressing Bm-3b short consistently grew faster and to higher densities when compared to the mock transfected controls and the clones with reduced levels of Bm-3b. In the experiments in which the cells were grown in monolayer, the clones with reduced levels of Bm-3b were more growth inhibited than the mock transfected controls. The response to estrogen and tamoxifen was also different in the Bm-3b altered MCF7 clones. In the absence of estrogens, the Bm-3b cells show no growth advantage over the control cells (Figure IV. 17). Upon addition of 17-P estradiol however, the Bm-3b cells react much more strongly than their control counterparts (average increase in growth 377% and 146.5% respectively) to the addition of the chemical. Similarly, upon the addition of tamoxifen the Bm-3b cells are compromised only to the same degree that they were without estrogen, while the control cell counterparts are more sensitive to and growth compromised by the tamoxifen. Conclusions are more difficult to draw fi*om the antisense cell lines because the cell numbers are so low. Similar results are seen in the proliferation experiments (Figures IV. 19 and IV.20). In the absence of estrogens the Bm-3b overexperssing cells show no growth advantage over the

225 control cells, however upon addition of estradiol these cells respond with almost two fold proliferative activity when compared with the control counterparts. The proliferation is decreased, but not quenched, as is the case with the control cells, upon the addition of tamoxifen (Figure IV.20). Thus, the growth advantage in the presence of tamoxifen may be due more to a decrease in cell death rather than maintenance of the proliferative state. The effects seen in the stripped serum/phenol-redless medium would indicate that the 17- P estradiol does indeed play a role in the growth and proliferation of the Bm-3b altered cell lines. Further experiments, however, must be undertaken to confirm if these effects are mediated directly the estrogen receptor. Suprisingly, however, in the anchorage independent growth experiments, tamoxifen acted as an estrogen agonist in both the clones overexpressing Bm-3b short and the clones with reduced levels of Bm-3b. The control cells responded to the tamoxifen in the manner expected, fewer anchorage independent colonies were formed. These results may suggest that the altered levels of Bm-3b intefere with ligand-dependent signal transduction through the AF-2 domain. It has been well documented that tamoxifen may act as an estrogen agonist when functioning through the AF-1 domain of the estrogen receptor. This may be the case in the anchorage independent cell growth experiments with these clones.

The effects of the alteration of the levels of Bm-3b are clear and obvious. Further investigation is, however necessary to determine the pathways and transcriptional mechanisms affected by the altered levels of Bm-3b. The next chapter investigates differential gene expression in the clonal cell lines.

226 V DIFFERENTIAL GENE EXPRESSION ANALYSIS OF MCF7 CELLS WITH ALTERED LEVELS OF BRN-3B

V.l Introduction

Differences in the cellular growth characteristics of each of the Bm-3b altered clonal cell lines discussed in the previous chapter are clear. The next goal was to gain an insight into the alterations in molecular pathways and gene programs that led to these differences. In order to accomplish this, the expression of genes both as final protein product and as mRNA transcript was investigated to discover if differences existed among the clonal cell lines. In the initial experiments, Western immunoblotting was used to examine the protein expression levels of genes of specific interest. In later experiments, the mRNA expression profile, in the Bm-3b overexpressing and Bm-3b reduced cell lines, of 1176 well-characterized cancer-related genes on a nylon cDNA hybridization array was investigated. Finally, the results of the DNA array were corroborated with quantitative RT-PCR of five genes fi*om the array.

V.1.1 Specific Western immunoblotting

Specific protein expression in each of the cell lines was analyzed using Western immunodetection. Briefly, each cell line was harvested in cell lysis buffer and soluble extract was equalized for protein content by Bradford assay. Equal total protein concentration from each cell line was loaded onto an SDS/polyacrylamide gel of appropriate percentage acrylamide and then electrophoresed. The resolved proteins were then transferred to nitrocellulose and the nitrocellulose was blotted with the appropriate antibody. The Bm-3b overexpressing and Bm-3b antisense cell lines were compared to the empty vector controls.

227 The genes analyzed by Western immimoblot may be grouped into four categories: 1. Bm- 3b proteins (including the long from of Bm-3b and the long form of Bm-3a), 2. steroid receptor and peptide hormone (including the estrogen receptor and the beta subunit of human chorionic gonadotropin), 3. heat shock proteins (including Hsp90, HspTO, and Hsp27), and 4. tumor suppressor genes (including BRCA I).

V.l.l.A Bm-3 proteins

Homeobox, and specifically POU domain containing proteins, have been observed to be expressed out of their normal physiological context in both cancer cell lines and malignant tissue samples. Bm-3b and Bm-3 a play important roles in the proliferation and subsequent differentiation of at least one mouse and three different human neuroblastoma cell lines (Latchman, 1996; Latchman, 1998a; Latchman, 1999; Latchman et al., 1992). Bm-3 a is also overexpressed in aggressive neuroendocrine tumors (Leblond-Francillard er a/., 1997). Bm-3 a also activates the expression of the human papillomavims types 16 and 18 E6 and E7 oncogenes (Morris et al., 1994). Recently it has been shown that Bm-3 a is expressed over 300 fold more in human cervical intraepithélial neoplasia grade 3 lesions than in normal human cervical biopsies. Moreover, elevated expression of Bm-3 a was detected in areas adjacent to CIN3 lesions, indicating that the increased expression levels are not a result of the oncogenic process, and suggesting that Bm-3a may activate HPV oncogene expression in the lesions (Ndisang et al, 1999). Expression of POU homeobox genes has been investigated in the MCF7 breast adenocarcinoma cell line, as well. Using degenerate oligonucleotide RT-PCR, Westem blotting, and electromobility shift assays, four POU products Oct-1, Oct-2, Oct-3, and Oct-11 were identified in the MCF7 cells. Oct-3 was further shown to be expressed in both human breast cancer cell lines and human breast cancer primary tumors, but not in normal human breast tissue, indicating that Oct-3 may play a role in mammary carcinogenesis (Jinera/., 1999). Finally, primary mammary tumors which show reduced expression of the tumor suppressor gene BRCA I show enhanced levels of Bm-3b. In addition, this enhanced

228 level of Bm-3b is not seen in normal mammary cells, benign tumors or in malignant tumors which do not have reduced levels of Bm-3b (Budhram Mahadeo et al, 1999b). Given the strength of POU domain proteins as versatile transcription factors, and the knowledge that these factors are aberrantly expressed in cancers, it is important to investigate the differential expression of these genes in the MCF-7 mammary carcinoma cell lines with altered levels of Bm-3b. Changes in the expression of Bm-3b long and Bm-3a long were investigated.

V.l.l.B Hormone receptors and hormones

V.l.l.B.a The estrogen receptor

As mentioned earlier, Bm-3b and the estrogen receptor functionally interact, and Bm-3b is able to activate transcription from estrogen responsive promoters. In addition, the MCF7 human adenocarcinoma cell line was chosen for these studies because it expresses the estrogen receptor. Finally, the addition of E2 or tamoxifen to the cell lines with altered levels of Bm-3b had significant effects on the growth, proliferation and formation of anchorage-independent clones (previous chapter). For these reasons, altered levels of estrogen receptor in these cells was investigated.

V.l.l.B.b Human chorionic gonadotropin

Human chorionic gonadotropin (hCG) is a glycoprotein hormone consisting of two non- covalently linked subunits homologous in structure to leutinizing hormone. hCG appears to convey a hormonal stimulus from the implanted fertilized ovum to the corpus leuteum which results in an augmented production of steroid hormones (Bishop, Nureddin & Ryan, 1976). While the beta subunit of hCG (p-hCG) has been detected in several tumors including those of ovarian, breast, and testicular origin, its role in the physiology of these tumors is not known. Recently, Remier and colleagues showed a strong correlation between P-hCG levels and progesterone receptor (PgR) levels in malignant breast tissue indicating that P-hCG may be involved in estrogen inducible pathways of

229 gene regulation (Reimer et al., 2000). Because P-hCG is expressed in breast cancers (Russo & Russo, 2000), is involved in steroidogenesis, and like Bm-3b is associated with estrogen inducible pathways of gene regulation, the level of p-hCG in these cells was investigated.

V.l.l.C Heat shock proteins

Heat shock proteins are a group of highly conserved proteins in both prokaryotes and eukaryotes which act as molecular chaperones and as a defense mechanism against environmental stress. As well as fulfilling several cellular roles to ensure the maintenance of homeostasis, heat shock proteins are associated with the proper function of steroid receptors (Graumann & Jungbauer, 2000), and are used as prognostic indicators in cancers. We chose to investigate the regulation of Hsp90, HspTO, and Hsp27.

V.l.l.C.aHsp90

In the general introduction, the association of the newly translated steroid receptor monomer with the 90 kDa heat shock protein was discussed (Rutherford & Zuker, 1994). Because in earlier experiments in this chapter it was noted that the levels of expression of the estrogen receptor increased in the cell lines overexpressing Bm-3b short and possibly decreased in the cell lines expressing Bm-3b antisense, we chose to investigate if the 90 kDa chaperone of the unliganded steroid receptor rose and fell commensurately.

V.l.l.C.b HspTO

Several laboratories have shown HspTO, like Hsp90, to be a part of the large protein hetero-complexes which keep steroid receptors in their inactive, unliganded state (Graumann & Jungbauer, 2000). In addition a specific interaction between HspTO and p53 may play a role in p53-associated cellular oncogenesis (Chen, Lin-Shiau & Lin, 1999b; Davidoff, Iglehart & Marks, 1992; Iwaya et al., 1995; Maehara et al., 2000; Park et al., 1999). While in breast cancers HspTO has been used as a prognostic indicator for

230 survival, reports differ as to whether high levels of HspTO predict short (Ciocca et al, 1993) or long (Elledge et al, 1994) survival. Finally, Landall and colleagues, in 1997, proposed that HspTO, through its ability to augment the estrogen receptor’s binding to its response element, contributes to the activation of genes which contain EREs (Landel et al, 199T). For these reasons the levels of HspTO were investigated in the clonal cell lines.

V.l.l.C.cHsp2T

Hsp2T is a member of the small heat shock protein family and is identical with the 24 kDa protein originally identified in MCFT cells and in human breast carcinomas (Adams et al, 1983), expressed in several hormone sensitive organs (Ciocca et al, 1983), and human tumors (Ciocca, Puy & Fasoli, 1989; Fuqua, Blum-Salingaros & McGuire, 1989). Several labs have shown that the expression of Hsp2T correlates well with the presence of ER and PgR in breast tumor biopsy samples (Ciocca et al, 1989). In addition, Hsp2T expression is estrogen inducible (Ciocca et al, 1983). Because Hsp2T plays a role in breast cancers, is correlated with estrogen receptor, and is estrogen inducible, its levels of expression in the Bm-3b altered cell lines were investigated.

V.l.l.D Tumor suppressor gene

V.l.l.D.aBRCAI

The tumor suppressor gene BRCA I which was cloned in 1994 (Miki et al, 1994) is mutated in 50 - 80% of all familial breast cancers. Overexpression of BRCA I results in a growth inhibition of breast and ovarian cell lines (Holt et al, 1996), while reduction of BRCA I via antisense methods results in the proliferation of mammary epithelial cells (Thompson et al, 1995). In addition, as mentioned earlier, mammary tumors with reduced levels of BRCA I have increased levels of Bm-3b, while normal cells, benign tumors, and malignant tumors with normal levels of BRCA I have normal levels of Bm- 3b (Budhram Mahadeo et al, 1999b). Finally, Bm-3b has been shown to repress

231 transcription from a BRCA I promoter/firefly luciferase reporter construct (Budhram Mahadeo étal., 1999b). For these reasons the expression of BRCA I was investigated in the Bm-3b altered cell lines.

V .l.2 Differential display on a cDNA expression array cDNA array technology is used to examine gene expression of hundreds to thousands of specific genes in an entire cDNA population on a single blot. Genome mapping projects have generated large scale sequence data for thousands of genes, however the biological function of these genes remain to be elucidated. Prior to understanding the functional significance of specific gene products, it is necessary to define differential gene expression profiles by comparing expression patterns in different tissue types, developmental stages, and disease states. Expression analysis techniques such as RT- PCR, Northern blot analysis, and Westem immunodetection have been widely used in the past, but these techniques are applicable only to a restricted number of genes. cDNA gene arrays provide a systematic approach for the simultaneous, sensitive and quantifiable detection of differential regulation of gene expression based on differential mRNA expression. The analytical principle of the cDNA expression array is based on a reverse Northern blot hybridization. Several groups have used similar techniques to demonstrate changes that occur after particular treatments (Iyer et al, 1999) or in cells from different tissue samples (Golub et ah, 1999; Soukas et al., 2000; Zhao et al., 2000). To further characterize the gene networks involved in the different growth characteristics of the MCF7 clonal cell lines, an array containing cDNA fragments representing 1176 human cancer-related genes (200-600 bp long, without poly(A) tails, repetitive elements, or highly homologous sequence) immobilized on a positively charged nylon membrane was used. The membrane was probed with radiolabelled cDNA form the Bm-3b overexpressing cell line Z and in a separate experiment another membrane was probed with radiolabelled cDNA form the Bm-3b antisense cell line A2.

V .l.3 Corroboration of cDNA expression array results by RT-PCR

232 The data from the cDNA expression must be corroborated by another method. Reverse transcription PCR is ideally suited to this task. Ultimately genes from the array that exhibit the same pattern of regulation as shown by RT-PCR would be tested in Northern blot analysis.

233 v .l Results

V.2.1 Specific Western immunoblotting

V.2.1.A Bm-3 proteins

V.2.1.A.aBm-3b long

The cell lines stably expressing the Bm-3b short construct show a 10.6 - 11.9 fold increase in the levels of Bm-3b short (refer to previous chapter, Figure IV.4). Similarly, the cell lines stably expressing the Bm-3b antisense show no detectable Bm-3b short and a 0.2 - 0.4 fold decrease in the levels of Bm-3b long. The experimental design, however, can not account for the 2.6 - 3.3 fold increase seen in the levels of Bm-3b long expression in the Bm-3b short overexpressing cell lines when compared to the vector controls. This increase indicates that Bm-3b short may be able to autoregulate its own promoter.

V.2.1. A.b Bm-3 a long

The levels of Bm-3 a long rise in the Bm-3b short overexpressing cell lines. In the first experiment the levels of Bm-3a in the Bm-3b short overexpressing cell lines showed a 1.5 - 2.8 fold increase when compared to the empty vector controls (Figure V.l) and in the second experiment these increases were 2.3 - 2.6 fold (Figure V.2). Only the second experiment included the Bm-3b antisense cell lines and in these experiments no decrease in Bm-3 a protein production was seen when compared to the empty vector control cell lines.

234 Figure V.1. Western blot analysis of Brn-3a long in clonal MCF7 cell lines with altered levels of Brn-3b. Immunoblots to detect the levels of Bm-3a long were carried out using six clones: two overexpressing pLTR Bm-3b (Z and Y) each in duplicate, and two empty pLTR vector controls (B and C). Total cellular protein (60 micrograms per lane) was fractionated on a SDS/15% polyaerylamide gel, transferred to nitrocellulose, and probed with an anti-Bm- 3a antibody. Expression levels were normalized to total protein densitometry of the coomassie satined gel after westem blotting. The level of Bm-3 a long antigen in extract from the pLTR Bm-3b overexpressing clones was 1.5 - 2.8 times the average endogenous level seen in the vector controls.

pLTR Bm-3b overexpressing pLTR empty B

Brn-3a (I)

CLONE DENSITOMETRY(arbitrary units) FOLD EXPRESSION Brn-3a Coomassie raw normalized •

pLTR Brn-3b short Z 524294 551569 5.63 1.46

pLTR Brn-3b short Z 548829 453077 5.89 1.86

pLTR Brn-3b short Y 639352 350176 6.86 2.80

pLTR Brn-3b short Y 702403 393675 7.54 2.74

pLTR B 69532 217177 0.75 0.49

pLTRC 116787 118621 1.25 1.51

235 Figure V.2. Western blot analysis of Brn-3a long in clonal MCF7 cell lines with altered levels of Brn-3b. Immunoblots to detect the levels of Bm-3a long were carried out using six clones: two overexpressing pLTR Bm-3b (Z and Y), two empty pLTR vector controls (B and C), and two overexpressing the pJ4 Bm-3b antisense construct (A1 and A2). Total cellular protein (60 micrograms per lane) was fractionated on a SDS/15% polsyacrylamide gel, transferred to nitrocellulose, and probed with an anti-Bm-3a antibody. The level of Bm- 3a long antigen in extract from the pLTR Bm-3b overexpressing clones was 2.3 - 2.6 times the average endogenous level seen in the vector controls, and the level of Bm-3a long antigen in extracts from the pJ4 Bm-3b antisense clones was 0.6 - 0.9 times the average endogenous level seen in the vector controls.

pLTR Bm-3b overexpressing pLTR empty pJ4 Bm-3b antisense B C A1 A2

B r n - 3 a (I)

a c t i n

CLONE DENSITOMETRY(arbitrary units) FOLD EXPRESSION Brn-3a Iona Actin raw normalized

pLTR Brn-3b short Z 333091 561578 2.04 2.61

pLTR Brn-3b short Y 326225 629120 2.00 2.28

pLTRB 223087 740746 1.37 1.33

pLTR C 102835 670602 0.63 0.67

pJ4 Brn-3b antisense A1 112428 819828 0.69 0.60

pJ4 Brn-3b antisense A2 182554 908405 1.12 0.88

236 v.2.1.B Hormone receptors and hormones

V.2.1 .B.a Estrogen receptor

In these studies, two different monoclonal antibodies were used to detect levels of estrogen receptor in the clonal cell lines. In the first experiment, a monoclonal antibody produced by Santa Cruz biotechnology was used and it revealed a 2.3 - 2.9 fold increase in estrogen receptor protein in the Bm-3b short overexpressing cell lines while no noteworthy alterations in expression were seen in the Bm-3b antisense cell lines (Figure V.3). In the second experiment, a monoclonal antibody from StressGen was used, and these results were similar to those obtained with the Santa Cruz antibody, namely a 1.6 - 2.0 fold increase in the levels of estrogen receptor in the Bm-3b short overexpressing cell lines. Interestingly, a decrease of 0.4 - 0.6 fold was seen in the Bm-3b antisense cell lines when compared to the empty vector controls (Figure V.4). Taken together, these results indicate that upregulation of Bm-3b short leads to an upregulation of the estrogen receptor.

V.2.1.B.b B-hCG

In these experiments, only the Bm-3b short cell lines were tested against the empty vector control cell lines. In duplicate lanes, the Bm-3b overexpressing cell lines showed clear and strong upregulation of p-hCG, with clone Z showing 3.6 - 4.1 fold increase and clone Y showing a 5.7 - 7.1 fold increase when compared to the empty vector controls. (Figure V.5)

237 Figure V.3. Western blot analysis of ER in clonal MCF7 cell lines with altered levels of Brn-3b. Immunoblots to detect the levels of ER were carried out using six clones: two overexpressing pLTR Bm-3b (Z and Y), two empty pLTR vector controls (B and C), and two overexpressing the pJ4 Bm-3b antisense construct (A1 and A2). Total cellular protein (60 micrograms per lane) was fractionated on a SDS/12% polyacrylamide gel, transferred to nitrocellulose, and probed with an anti-ER antibody. The level of ER antigen in extract from the pLTR Bm-3b overexpressing clones was 2.3 - 2.9 times the average endogenous level seen in the vector controls, and the level of ER antigen in extracts from the pJ4 Bm-3b anti sense clones was 0.8 - 1.2 times the average endogenous level seen in the vector controls.

pLTR Bm-3b overexpressing pLTR empty pJ4 Bm-3b antisense B A1 A2

& ER

actin

CLONE DENSITOMETRY(arbitrary units) FOLD EXPRESSION E R Actin raw normalized

pLTR Brn-3b short Z 250627 561578 2.29 2.89

pLTR Brn-3b short Y 224256 629120 2.05 2.31

pLTR B 126890 740746 1.16 1.11

pLTRC 92317 670602 0.84 0.89

pJ4 Brn-3b antisense A1 96137 819828 0.88 0.76

pJ4 Brn-3b antisense A2 166421 908405 1.52 1.19

238 Figure V.4. Western blot analysis of ER in clonal MCF7 cell lines with altered levels of Brn-3b. Immunoblots to detect the levels of ER were carried out using six clones: two overexpressing pLTR Bm-3b (Z and Y), two empty pLTR vector controls (B and C), and two overexpressing the pJ4 Bm-3b antisense construct (A1 and A2). Total cellular protein (60 micrograms per lane) was fractionated on a SDS/12% polyacrylamide gel, transferred to nitrocellulose, and probed with an anti-ER antibody. The level of ER antigen in extract from the pLTR Bm-3b overexpressing clones was 1.6 - 2.0 times the average endogenous level seen in the vector controls, and the level of ER antigen in extracts from the pJ4 Bm-3b antisense clones was 0.4 - 0.6 times the average endogenous level seen in the vector controls.

pLTR Bm-3b overexpressing pLTR empty pJ4 Bm-3b antisense B Â I Â 7

actin

CLONE DENSITOMETRY(arbitrary units) FOLD EXPRESSION ER Actin raw normalized

pLTR Brn-3b short Z 84578 561578 1.62 2.02

pLTR Brn-3b short Y 75803 629120 1.45 1.62

pLTR B 46624 740746 0.89 0.84

pLTR C 57792 670602 1.11 1.16

pJ4 Brn-3b antisense A1 36794 819828 0.70 0.60

pJ4 Brn-3b antisense A2 26498 908405 0.51 0.39

239 Figure V.5. Western blot analysis of beta human chorionic gonadotropin long in clonal MCF7 cell lines with altered levels of Brn-3b. Immunoblots to detect the levels of beta-hCG were carried out using six clones: two overexpressing pLTR Bm-3b (Z and Y) each in duplicate, and two empty pLTR vector controls (B and C). Total cellular protein (60 micrograms per lane) was fractionated on a SDS/15% polyacrylamide gel, transferred to nitrocellulose, and probed with an anti-beta- hCG antibody. Expression levels were normalized to total protein densitometry of the coomassie satined gel after westem blotting. The level of beta-hCG antigen in extract from the pLTR Bm-3b overexpressing clones was 3.6 - 7.1 times the average endogenous level seen in the vector controls.

pLTR Bm-3b overexpressing pLTR empty B

P-hCG

CLONE DENSITOMETRYtarbitrarv units) FOLD EXPRESSION Brn-3a Coomassie raw normalized

pLTR Brn-3b short Z 1114033 551569 11.55 3.58

pLTR Brn-3b short Z 1047134 453077 10.86 4.10

pLTR Brn-3b short Y 1393001 350176 14.45 7.06

pLTR Brn-3b short Y 1255212 393675 13.02 5.66

pLTRB 130236 217177 1.35 1.06

pLTRC 62588 118621 0.65 0.94

240 V.2.1.C Heat shock proteins

V.2.1.C.a Hsp90

The levels of the 90 kDa chaperone of the estrogen receptor remain strikingly constant with expression not varied by more than 0.4 fold in both the Bm-3b overexpressing cell lines as well as in the Bm-3b antisense cell lines. We may conclude from these experiments that the levels of Hsp90 are not altered in the cell lines with altered levels of Bm-3b. These results are valuable in confirming equal protein loading in this set of experiments, and thus provide a useful control against which other protein increases or decreases may be judged. (Figure V.6)

V.2.1.C.b Hsp/Hsc70

Like the levels of the 90 kDa heat shock protein the levels of Hsc70 (the constituitively expressed form of HspTO) remain convincingly unchanged with the fold variation in expression differing by even less than the Hsp90. We may conclude from these experiments that the levels of HspTO are not altered in the cell lines with altered levels of Bm-3b. Again, these results are valuable in confirming equal protein loading in this set of experiments, and thus provide a useful control against which other protein increases or decreases may be judged. (Figure V.T)

V.2.1.C.C Hsp2T

In these experiments, Hsp2T levels are increased in the Bm-3b overexpressing cell lines 1.2 fold, and reduced in the Bm-3b antisense cell lines 0.5 fold. While the differences in expression are not large, they are clear and consistent. The levels of Hsp2T are altered in the Bm-3b altered cell lines. (Figure V.8)

241 Figure V.6. Western blot analysis of Hsp90 In clonal MCF7 cell lines with altered levels of Brn-3b. Immunoblots to detect the levels of Hsp90 were carried out using six clones: two overexpressing pLTR Bm-3b (Z and Y), two empty pLTR vector controls (B and C), and two overexpressing the pJ4 Bm-3b anti sense construct (A1 and A2). Total cellular protein (60 micrograms per lane) was fractionated on a SDS/12% polyacrylamide gel, transferred to nitrocellulose, and probed with an anti-Bm-3a antibody. The level of Hsp90 antigen in extract from the pLTR Bm-3b overexpressing clones was 1.3 - 1.4 times the average endogenous level seen in the vector controls, and the level of Hsp90 antigen in extracts from the pJ4 Bm-3b anti sense clones was 0.9 - 1.0 times the average endogenous level seen in the vector controls.

pLTR Bm-3b overexpressing pLTR empty pJ4 Bm-3b antisense B ÂI ÂZ

Hsp90

. #

actin

CLONE DENSITOMETRY(arbitrary units) FOLD EXPRESSION Hsd90 Actin raw normalized

pLTR Brn-3b short Z 153216 561578 1.11 1.38

pLTR Brn-3b short Y 160372 629120 1.16 1.29

pLTRB 118088 740746 0.85 0.80

pLTRC 158769 670602 1.15 1.20

pJ4 Brn-3b antisense A1 149413 819828 1.08 0.92

pJ4 Brn-3b antisense A2 182395 908405 1.32 1.01

242 Figure V.7. Western blot analysis of Hsp/Hsc70 in clonal MCF7 cell lines with altered levels of Brn-3b. Immunoblots to detect the levels of HspTO were carried out using six clones: two overexpressing pLTR Bm-3b (Z and Y), two empty pLTR vector controls (B and C), and two overexpressing the pJ4 Bm-3b antisense construct (A1 and A2). Total cellular protein (60 micrograms per lane) was fractionated on a SDS/12% polyacrylamide gel, transferred to nitrocellulose, and probed with an anti-HspTO antibody. The level of HspTO antigen in extract from the pLTR Bm-3b overexpressing clones was 0.8 - 0.9 times the average endogenous level seen in the vector controls, and the level of HspTO antigen in extracts from the pJ4 Bm-3b antisense clones was 0.6 - O.T times the average endogenous level seen in the vector controls.

pLTR Bm-3b overexpressing pLTR empty pJ4 Bm-3b antisense B S ÂI ÂZ :

Hsp70

actin

CLONE DENSIT0METRY(arbitrai7 units) FOLD EXPRESSION Hsd70 Actin raw normalized

pLTR Brn-3b short Z 210391 658868 0.92 0.83

pLTR Brn-3b short Y 121142 600856 0.93 0.92

pLTR B 269305 592461 1.18 1.18

pLTRC 187355 593171 0.82 0.82

pJ4 Brn-3b antisense A1 165375 628846 0.72 0.68

pJ4 Brn-3b antisense A2 146693 644862 0.64 0.59

243 Figure V.8. Western blot analysis of Hsp27 in clonal MCF7 cell lines with altered levels of Brn-3b. Immunoblots to detect the levels of Hsp27 were carried out using six clones: two overexpressing pLTR Bm-3b (Z and Y), two empty pLTR vector controls (B and C), and two overexpressing the pJ4 Bm-3b antisense construct (A1 and A2). Total cellular protein (60 micrograms per lane) was fractionated on a SDS/15% polyacrylamide gel, transferred to nitrocellulose, and probed with an anti-Hsp27 antibody. The level of Hsp27 antigen in extract from the pLTR Bm-3b overexpressing clones was 1.2 times the average endogenous level seen in the vector controls, and the level of Hsp27 antigen in extracts from the pJ4 Bm-3b anti sense clones was 0.5 times the average endogenous level seen in the vector controls.

pLTR Brn-3b overexpressing pLTR empty pJ4 Bm-3b antisense

Hsp27

actin

CLONE DENSITOMETRY(arbitrary units) FOLD EXPRESSION Hsd27 Actin raw normalized

pLTR Brn-3b short Z 286750 658868 1.48 1.17

pLTR Brn-3b short Y 268631 600856 1.38 1.20

pLTRB 182309 592461 0.94 0.83

pLTR C 206468 593171 1.06 0.94

pJ4 Brn-3b antisense A1 126525 628846 0.65 0.54

pJ4 Brn-3b antisense A2 128712 644862 0.66 0.54

244 v . 2 . 1 . D Tumor suppressor gene

V.2.1.D.aBRCAI

Because BRCA I is a large protein (220 kDa) two experiments were undertaken. One experiment resolved the protein on a 2.0M urea/SDS/8% polyacrylamide gel to help keep the protein in solution and reduce precipitation. The other experiment used a conventional SDS/6% polyacrylamide gel to resolve the protein. The results from each experiment were consistent with one another: In the urea gel BRCA I was seen 1.6 - 1.9 fold overexpressed in the Bm-3b short overexpressing cell lines (Figure V.9), as compared to 1.8 - 2.1 for the conventional gel (Figure V.IO). The levels of BRCA I in the antisense cell lines did not seem to vary significantly from the empty vector controls in the urea experiment, but did appear to decrease in the conventional gel. Thus, counterintuitively, BRCA I appears to be upregulated in the Bm-3b short overexpressing cell lines.

245 Figure V.9. Western blot analysis of BRCA I in clonal MCF7 cell lines with altered levels of Brn-3b. Immunoblots to detect the levels of BRCA I were carried out using six elones: two overexpressing pLTR Bm-3b (Z and Y), two empty pLTR vector controls (B and C), and two overexpressing the pJ4 Bm-3b antisense construet (A1 and A2). Total cellular protein (100 micrograms per lane) was fractionated on a 2.0M urea/SDS/8% polyacrylamide gel, transferred to nitrocellulose, and probed with an anti-BRCA I antibody. The level of BRCA I antigen in extract from the pLTR Bm-3b overexpressing clones was 1.6 - 1.9 times the average endogenous level seen in the vector controls, and the level of BRCA I antigen in extracts from the pJ4 Bm-3b antisense clones was 1.2 times the average endogenous level seen in the vector controls.

pLTR Brn-3b overexpressing pLTR empty pJ4 Bm-3b antisense B A1 A2

BRCA I

a m

actin

CLONE DENSITOMETRY(arbitrary units) FOLD EXPRESSION BRCA 1 Actin raw normalized

pLTR Brn-3b short Z 934639 340577 2.27 1.64

pLTR Brn-3b short Y 965597 310292 2.34 1.86

pLTR B 182529 232622 0.44 0.47

pLTRC 641414 250351 1.56 1.53

pJ4 Brn-3b antisense A1 570996 294787 1.39 1.16

pJ4 Brn-3b antisense A2 714570 359358 1.74 1.19

246 Figure V.10. Western blot analysis of BRCA I in clonal MCF7 cell lines with altered levels of Brn-3b. Immunoblots to detect the levels of BRCA I were carried out using six clones: two overexpressing pLTR Bm-3b (Z and Y), two empty pLTR vector controls (B and C), and two overexpressing the pJ4 Bm-3b antisense construct (A1 and A2). Total cellular protein (60 micrograms per lane) was fractionated on a SDS/6% polyacrylamide gel, transferred to nitrocellulose, and probed with an anti-BRCA I antibody. The level of BRCA I antigen in extract from the pLTR Bm-3b overexpressing clones was 1.8 - 2.1 times the average endogenous level seen in the vector controls, and the level of BRCA 1 antigen in extracts from the pJ4 Bm-3b antisense clones was 0.7 - 0.9 times the average endogenous level seen in the vector controls.

pLTR Bm-3b overexpressing pLTR empty pJ4 Bm-3b antisense B Â I Â 7

BRCA I

actin

# - c l #

CLONE DENSITOMETRY(arbitrary units) FOLD EXPRESSION BRCA 1 Actin raw normalized

pLTR Brn-3b short Z 324945 658868 2.04 1.84

pLTR Brn-3b short Y 338821 600856 2.13 2.10

pLTR B 143247 592461 0.90 0.90

pLTR C 174640 593171 1.10 1.10

pJ4 Brn-3b antisense A1 154674 628846 0.97 0.92

pJ4 Brn-3b antisense A2 121940 644862 0.77 0.71

247 V.2.2 Differential display of genes on the Atlas human cancer cDNA array

Partial cDNAs 1176 cancer-related genes were tested for differential expression in the pLTR Bm-3b short overexpressing cell line Y clone and the pJ4 Bm-3b antisense A1 clone. The overexpressing and antisense clones were chosen for this analysis so as to maximise differences seen in the differential expression of genes. This strategy has inherent in it both advantages and disadvantages. By selecting cell lines that contain extreme differences in the levels of Bm-3b, one would hope to reveal the greatest possible number of genes that may be altered either directly or indirectly by altered levels of Bm-3b using the fewest financially restrictive array membranes. Using this strategy, however, one must realise that the absence of a particular factor does not necessarily result in the opposite effect of the presence of the factor. Indeed, presence of a particular factor may regulate gene programs entirely different from the alterations in gene programs resulting from its absence. Thus, interpretations of the results of the differential display may act only as a starting point for the analysis of gene expression. These results, as with any differential display results, must be confirmed and corroborated by alternative methods including but not limited to, RT-PCR, northern analysis, and promoter studies. The practical work for this experiment was undertaken by the Clontech Atlas Array Service. Briefly, mRNA was prepared from the Bm-3b short overexpressing and the Bm-3b antisense cell lines. This mRNA was reverse transcribed into cDNA using primers specific to the partial cDNAs found on the membrane, hybridized to the nylon array, and then stringently washed. The membrane was exposed to a phosphorimager and the resulting information was analyzed using Atlaslmage software. Fifty-one genes were selected on the basis of differences in spot intensity ratios of 1.69 (as calculated by the Atlaslmage software based on the results of the hybridisation) or higher (Table V.l).

248 Table V.1. Results of the differential display of genes on the Atlas human cancer cDNA array. Facing page

249 t control I http://atlas.cl VS. brn-3b short anti-sense | Array tot# Normalization n Thresholds Ü 9100044 method coefficient 1 ratio difference S Atlas Array global 1.30 1 1.69 15 § Cancer 1.2k Array

# coordinate anti-sense bm-3b stiort Ratio Difference UP DOWN Gene 1 A01c 66 28 0.42 -38 2.4 c-jun proto-oncogene; transcription factor AP-1 2 AOIh 3 25 8.33 22 8.3 interferon-inducible protein 9-27 3 A03c 71 36 0.51 -35 2.0 c-myc oncogene 4 A03g 256 533 2.08 277 2.1 c-myc binding protein MM-1 5 A04j 51 88 1.73 37 1.7 cell division protein kinase 4; cyclin-dependent kinase 4 (0DK4); PSK-J3 6 A09I 50 116 2.32 66 2.3 cyclin-dependent kinase inhibitor 1 (0DKN1A); melanoma differentiation-associated protein 6 (MDA6); ODK-interacting protein 1 (0IP1); WAF1 7 AlOk 84 142 1.69 58 1.7 cyclin-dependent kinase regulatory subunit 1 (0KS1 ) 8 A12j 44 24 0.55 -20 1.8 cdc2-related protein kinase PISSLRE 9 A12n 35 10 0.29 -25 3.5 G1 to S phase transition protein 1 homolog; GTP-binding protein GST1-HS 10 B02a 381 220 0.58 -161 1.7 ADP/ATP carrier protein 11 B02I 49 23 0.47 -26 2.1 protein phosphatase 20 gamma 12 B04j 68 129 1.90 61 1.9 rhoO (H9): small GTPase (rhoO) 13 B04n 170 288 1.69 118 1.7 B-cell receptor-associated protein (hBAP) 14 B05I 35 18 0.51 -17 1.9 ? calmodulin 1; delta phosphorylase kinase 15 B07m 41 3 0.07 -38 13.7 zyxin + zyxin-2

16 BIOc 30 51 1.70 21 1.7 c-jun N-terminal kinase 2 (JNK2); JNK55 17 B10m 167 37 0.22 -130 4.5 junction plakoglobin (JUP); desmoplakin 111 (DP3)

18 B13j 29 10 0.34 -19 2.9 ? guanine nucleotide-binding protein G(l)/G(S)/G(0) gamma-10 subunit 19 COIg 46 19 0.41 -27 2.4 DNA ligase 1; polydeoxyribonucleotide synthase (ATP) (DNL1) (LIG1) 20 C04b 127 237 1.87 110 1.9 tumor necrosis factor type 1 receptor associated protein (TRAP1 )

21 C04g 34 19 0.56 -15 1.8 ? DNA excision repair protein ER001 22 C06n 27 10 0.37 -17 2.7 ? interferon regulatory factor 3 (IRF3) 23 C08I 27 7 0.26 -20 3.9 ? retinoic acid receptor alpha 1 (RAR-alpha 1; RARA) PML-RAR protein

24 Cl 21 40 5 0.13 -35 8.0 TIS11B protein; EGF response factor 1 (ERF1 ) 25 012j 28 5 0.18 -23 5.6 early growth response protein 1 (hEGRI); transcription factor ETR103; KROX24, protein 225; AT225 26 014m 25 3 0.12 -22 8.3 fuse-binding protein 2 (FBP2) 27 014n 49 24 0.49 -25 2.0 transcription factor -1 ; AP2 aamma transcription factor 28 D06e 60 32 0.53 -28 1.9 integrin beta 4 (ITGB4); CD104 antigen

29 D07b 65 31 0.48 -34 2.1 high mobility group protein HMG2 30 D08f 37 11 0.30 -26 3.4 paxillin 31 D09d 44 19 0.43 -25 2.3 alphal catenin (CTNNA1); cadherin-associated protein; alpha E- catenin

32 D09m 55 109 1.98 54 2.0 glutathione-S-transferase (GST) homolog 33 D14I 19 3 0.16 -16 6.3 ? cysteine-rich fibroblast growth factor receptor ; Golgi membrane sialOQlycoorotein MG160 (GLG1 ) 34 E02n 55 10 0.18 •45 5.5 78-kDa glucose regulated protein precursor (GRP 78); immunoglobulin heavy chain binding protein (BIP) 35 E03j 780 384 0.49 -396 2.0 cathepsin D precursor (CTSD) 36 E07f 130 64 0.49 -66 2.0 interleukin-1 beta precursor (IL-1 ; IL1B); catabolin 37 E07h 239 440 1.84 201 1.8 macrophage migration inhibitory factor (MIF); glycosylation-inhibiting factor (GIF) 38 E09h 25 9 0.36 -16 2.8 ? iaaaed2 (JAG2) 39 F04k 61 29 0.48 -32 2.1 60S ribosomal protein L5 40 F05e 14 36 2.57 22 2.6 ornithine decarboxylase 41 F08j 30 15 0.50 -15 2.0 ? HSC70-interacting protein; progesterone receptor-associated P48 protein 42 F08k 28 9 0.32 -19 3.1 ? eukaryotic translation initiation factor 3 beta subunit (EIF-3 beta); EIF3 P116 43 F08m 75 19 0.25 -56 3.9 PM5 protein 44 FI Ob 24 9 0.38 -15 2.7 ? IMP dehydrogenase 1 45 FlOj 110 64 0.58 ^6 1.7 suppressor for yeast mutant 46 FI 2d 25 10 0.40 -15 2.5 ? uridine 5'-monophosphate synthase (UMP synthase) 47 F12f 141 77 0.55 -64 1.8 type II cytoskeletal 2 epidermal keratin (KRT2E); cytokeratin 2E (K2E; CK2E) 48 F13k 52 15 0.29 -37 3.5 glycyl tRNA synthetase 49 FI 4b 24 42 1.75 18 1.8 aminoacvlase 1 (ACY1) 50 G27 2180 5230 2.40 3050 2.4 liver glyceraldehyde 3-phosphate dehydrogenase (GAPDH) 51 G43 1004 397 0.40 -607 2.5 cytoplasmic beta-actin (ACTS)

N/C = not calculated due to manually-detennlnecl inconsistencies (signal bleeding, background, etc.) in one or both spots Results generated by Atlaslmage 1.5k (Final) To sort columns, click on column heading, then Data/Sort tittn//atlas.clontectr com/

250 V.2.2.A Initial analysis of the array results reveals a broad spectrum of altered gene expression.

Several classes of protein molecules are differentially regulated in the two cell lines. The classes of molecules encompass: transcription factors (including: c-jun, c-myc, c-myc binding protein, EGF response factor 1, early growth response protein 1, fuse binding protein, retinoic acid receptor, and AP2 gamma), signal transduction molecules (including: inteferon-inducible protein 27, zyxin, alpha E-catenin, and Jagged2), proteins involved in cell cycle regulation (including: cell division protein kinase 4 and cyclin dependent kinase inhibitor l/p21), heat shock proteins (including: tumor necrosis factor type 1 receptor associated protein and 78 kDa glucose regulated protein precursor), GTP binding proteins (including: G1 to S phase transition protein 1 homologue and rhoC), and cytoskeletal structural proteins (including: junction plakoglobin, P-actin and paxillin)

V.2.2.B Several genes differentially expressed in the clonal cell lines are regulated in wavs that would allow the Brn-3b overexpressing cell line to maximize proliferation potential.

• Cell division protein kinase 4 (CDK4), upregulated 1.7 fold, is a fundamental component of a holoenzyme assembly with D-type cyclins that facilitates exit from G1 phase of the cell cycle by phosphorylating key substrates such as (reviewed in (Sherr, 1995)). In addition, the stimulatory effect of estrogen on cell cycle progression in the G1 phase is mediated by the activation of CDK4 and CDK2. • RhoC, upregulated 1.9 fold, can play a role in cellular transformation (reviewed in (Ridley, 1997)), and is overexpressed in ductal adenocarcinoma of the pancreas (Suwa g/ a/., 1998). • Zyxin, downregulated 13.7 fold, has been implicated in pathways important for , and may serve as a molecular scaffold for the assembly of

251 multimeric protein complexes that function in the nucleus and at the sites of cellular adhesion (reviewed in (Beckerle, 1997)). • Junction plakoglobin, downregulated 4.5 fold, plays a key role in cell adhesion, and downregulation of this protein is a determinant in the metastatic capability of individual malignant cells (Buchner et a/., 1998). • Early growth response protein EGR, downregulated 5.6 fold, had significant transformation suppression activity (Liu, Adamson & Mercola, 1996a; Liu et al, 1996b; Liu et al., 1998). • Paxillin, downregulated 3.4 fold, is involved in connecting the actin cytoskeletin to the extracellular matrix (Salgia et al, 1995a; Salgia et al., 1995b; Salgia et al., 1995c). • Catenin, downregulated 2.3 fold, is involved in cell-cell adhesion, and is downregulation has been implicated as a possible contributor to cancer invasion and metastasis (Shimoyama et al., 1992). • Glucose regulated protein 78 (GRP78), downregulated 5.5 fold, has been shown to participate in the downregulation of the epidermal growth factor signaling pathway by forming a stable complex with the EGF receptor, thereby inhibiting the translocation of the receptor to the cell surface (Cai et al., 1998). • Jagged2, downregulated 2.8 fold, is associated with the differentiation of the sensory epithelium of the mammalian cochlea, and genetic deletion of Jagged2 results in a significant increase in these sensory hair cells (Lanford et al., 1999; Zhang et al., 2000; Zine, Van De Water & de Ribaupierre, 2000). • Retinoic acid receptor alpha 1, downregulated 3.9 fold, has been asscoiated with decreased cellular proliferation and increased cellular adhesion in ER responsive breast cancer cell lines (Zhu et al., 1999).

V.2.2.C Closer analysis of the genes differentially expressed in the array reveals some counterintuitive results.

• c-Jun, downregulated 2.4 fold, behaves as a positive regulator of cell growth and may cause cell transformation when overexpressed (Musti et al., 1996).

252 • Interferon-inducible protein 27, upregulated 8.3 fold, may be found as part of a multiprotein complex which mediates the antiproliferative effects of inteferon alpha (Deblandre et al, 1995). • c-Myc, downregulated 2.0 fold, is frequently amplified in many human tumors including breast tumors (Nass & Dickson, 1997). • The cyclin kinase inhibitor p21, upregulated 2.3 fold, plays a fundamental role on regulating the cell cycle, and can induce G1 arrest and block entry into S phase by interacting with cyclin dependent kinases or inhibiting proliferating cell nuclear antigen (Cartel, Serfas & Tyner, 1996). • G1 to S phase transition protein, downregulated 3.5 fold, is essential for G1 to S phase transition of the cell cycle (Hoshino et al., 1989).

V.2.2.D Several neuronal genes are regulated in the breast adenocarcinoma cell lines with altered levels of Brn-3b.

Bm-3b plays a significant role in the growth and development of cells derived from the neural crest, thus, it is unsurprising that its alteration in the MCF7 cell line should result in the differential expression of some ubiquitous genes expressed in neuronal tissue.

• c-Jun is one of the earliest and most consistent markers for neurons that respond to nerve fiber transection. Moreover, expression of c-jun can kill neonatal neurons, but may also be involved in neuroprotection and regeneration (Herdegen, Skene & Bahr, 1997). • RhoC and its family members are key players in neuritogenesis as well as neurite branching (Albertinazzi et al., 1998; Patrone et al., 2000). • Tumor necrosis factor type-1 associated protein (TRAPl) is expressed in the brain • Fuse binding protein is expressed in both neural and non-neuronal cell lines, but is expressed at higher levels in neuronal cells (Min et al, 1997).

253 Activating enhancer binding protein 2 gamma (AP2y) is required for the development of tissues of neuroectodermal origin, including those of the neural crest and skin (Williamson 1996). GRP78 interacts with the amyloid precursor protein (APP), a central protein in Alzheimer’s disease (AD) pathogenesis, in the endoplasmic reticulum, and this interaction modulates APP maturation and processing, and may facilitate its correct folding (Yang, Turner & Gaut, 1998b). Additionally, overexpression of GRP78 within neurons suggests that it may protect neurons from AD damage (Hamos et al., 1991).

254 V.2.3 Confirmation of the cDNA array results with RT-PCR

Four genes were selected for analysis by RT-PCR for the initial corroboration of the cDNA array results. These genes are: cyclophilin, RhoC, early growth response 1 (EGRl), and junction plakoglobin (Figures V.l 1 to V.14). These genes represent a gene that was upregulated in the cDNA array (RhoC upregulated 1.9 fold), as well as genes that were downregulated in the array (hEGRI downregulated 5.6 fold, and junction plakoglobin downregulated 4.5 fold). Cyclophilin was included as an invariant control against which to equalise the fold expression of the genes from the array.

V.2.3 .A Cyclophilin

Cyclophilin was used as an invariant control to ensure equal loading and equal amplification efficiency. In these experiments, the levels of cyclophilin are indeed invariant. The levels in the pLTR overexpressing cell lines are slightly higher than in the other cell lines, but these differences are minimal and differences seen in the following RT-PCR experiments are stronger than those seen here. (Figure V.l 1)

V.2.3.B Human RhoC

In these experiments, RhoC levels are clearly increased in the Bm-3b overexpressing cell lines cell lines, and decreased in the Bm-3b antisense when compared to the average mRNA levels seen in the Bm-3 control cell lines. These results confirm the results of the differential display in which the levels of RhoC were 1.9 fold higher in the Bm-3b overeexpressing cells when compared to the levels seen in the antisense cell lines. In this experiment, the levels of RhoC in the Bm-3b overexpressing cell lines were approximately 2 fold that of the control cells, while the antisense cell lines expressed approximately 60% of the RhoC mRNA seen in the control cell lines. (Figure V.l 2)

255 V.2.3.C Human early growth response 1 (hEGRI)

In these experiments there is a clear increase in the level of hEGRI transcript in the Bm- 3b antisense cell lines. These results confirm the results of the differential display in which hEGRI was downregulated 5.6 fold in the Bm-3b overexpressing cells when compared to the levels seen in the antisense cell lines. In this experiment, the levels of hEGRI in the Bm-3b overexpressing cell lines was only 60% that seen in the control cell lines. The levels of hEGRI transcript in the Bm-3b antisense cell lines was greater than 2.5 fold that seen in the control cell lines. (Figure V .l3)

V.2.3.D Human junction plakoelobin

In these experiments, junction plakoglobin levels are reduced in the Bm-3b overexpressing cell lines. The levels of transcript in the Bm-3b antisense cell line are close to those of the control cell lines. These results confirm the results of the differential display in which junction plakoglobin was downregulated 4.5 fold in the Bm-3b overeexpressing cells when compared to the levels seen in the antisense cell lines. In this experiment, the levels of junction plakoglobin in the Bm-3b antisense cell lines differed only minimally from the control cell lines. The levels of junction plakoglobin in the Bm- 3b overexpressing cell lines was only 17% that seen in the control cell lines. (Figure V.14)

V.2.3 E Summary of confirmation of the cDNA array results with RT-PCR

These results confirm the results of the Clontech cDNA array. In addition the results of the RT-PCR further reveal subtleties in the transcriptional regulation of these genes by Bm-3b. It is interesting to note that three types of regulation are seen in these experiments. In the RhoC experiment, it is clear that the amount of RhoC transcript present is commensurate with the amount of Bm-3b expression. In the case of hEGRI, however, only the absence of the Bm-3b transcription factor results in higher levels of hEGRI transcript production, as the levels of the hEGRI transcript remain constant in the

256 control and Bm-3b overexpressing cell lines. Finally, in the case of junction plakoglobin, increased levels of Bm-3b result in a decrease in transcript production without a commensurate increase when the Bm-3b levels are reduced.

257 pLTR Brn-3b short pLTR control pJ4 Brn-3b antisense B A1 A2

' *• . ' , ,, * ^‘v-- “K'^t-* , -/ ' ,î -\y -f ■ - . T ? '’ ',;*>■ * , * ,<“*'»* , '. ;x.\,t. -\ .'■**. ^ ' - ifc, -snififlEf!*!

Figure V.11. RT-PCR analysis of cyclophilin in clonal MCF7 cell lines with altered levels of Brn-3b. Southern hybridisation was used to detect a reverse transcription PCR product. cDNA prepared from total cellular RNA from each of the cell lines indicated. Primers specific for cyclophilin (see section II. 1.7) were used to amplify a 200 base pair fragment of the cyclophilin gene. PCR reactions were resolved on a 2% agarose gel, transferred to Hybond N+ nylon membrane and probed with the same PCR product prepared previously and labelled with ^^P. This experiment was undertaken to check for equal amplification of each of the cDNAs. The RNA used to prepare the cDNA for each sample is shown directly below the autoradiograph.

258 pLTR Brn-3b short pLTR control pJ4 Brn-3b antisense

Figure V.12. RT-PCR analysis of RhoC in clonal MCF7 cell lines with altered levels of Brn-3b. Southern hybridisation was used to detect a reverse transcription PCR product. cDNA was prepared from total cellular RNA from each of the cell lines indicated. Primers specific for RhoC (see section II. 1.7) were used to amplify a 200 base pair fragment of the RhoC. PCR reactions were resolved on a 2% agarose gel, transferred to Hybond N+ nylon membrane and probed with the same PCR product prepared previously and labelled with ^^P. In the chart below the autoradiograph, expression levels were normalised to the levels of cyclophilin expression (Figure V.l I).

Information from Clontech Atlas Cancer Array Name______Spot intensity Difference Ratio Antisense Overexp RhoC 68 129 61 up 1.9

259 pLTR Brn-3b short pLTR control pJ4 Brn-3b antisense A1 A2

Figure V.13. RT-PCR analysis of hEGRI in clonal MCF7 cell lines with altered levels of Brn-3b. Southern hybridisation was used to detect a reverse transcription PCR product. cDNA was prepared from total cellular RNA from each of the cell lines indicated. Primers specific for hEGRI (see section II. 1.7) were used to amplify a 200 base pair fragment of the hEGRI. PCR reactions were resolved on a 2% agarose gel, transferred to Hybond N+ nylon membrane and probed with the same PCR product prepared previously and labelled with ^^P. In the chart below the autoradiograph, expression levels were normalised to the levels of cyclophilin expression (Figure V.l 1).

Information from Clontech Atlas Cancer Array Name______Spot intensity Difference Ratio Antisense Overexp hEGRI 28 5 -23 down 5.6

260 pLTR Brn-3b short pLTR control pJ4 Brn-3b antisense Z Y B C A1 A2

£ oa :r-m u S' _

3 a ...«% a s Ï • 0c x ^ “l

21 o c

Figure V.14. RT-PCR analysis of junction plakoglobin in clonal MCF7 cell lines with altered levels of Brn-3b. Southern hybridisation was used to detect a reverse transcription PCR product. cDNA was prepared from total cellular RNA from each of the cell lines indicated. Primers specific for junction plakoglobin (see section II. 1.7) were used to amplify a 200 base pair fragment of the junction plakoglobin. PCR reactions were resolved on a 2% agarose gel, transferred to Hybond N+ nylon membrane and probed with the same PCR product prepared previously and labelled with ^^P. In the chart below the autoradiograph, expression levels were normalised to the levels of cyclophilin expression (Figure V.l 1).

Information from Clontech Atlas Cancer Array Name______Spot intensity Difference Ratio Antisense Overexp junction plakoglobin 167 37 -130 down 4.5

261 V.3 Discussion

V.3.1 Western immunoblotting

V.3.1.A Brn-3b short is not the only Brn-3 protein altered in the clonal cell lines.

The primary change observed in each of the experimental cell lines (overexpressing and antisense) is in the levels of Bm-3b short. The differential gene expression described herein, however, may not be solely due to alterations in the expression of Bm-3b short. As mentioned in the general introduction, Bm-3a and Bm-3b are encoded by two distinct genes, and the primary transcript of each of these genes is alternatively spliced to produce two distinct mRNAs encoding long and short isoforms that differ at the A-terminus of each protein. In neuronal cells and in primary neurons, the ratio of the isoforms may be modulated by specific stimuli (Liu et al., 1996c). The changes in the ratios of these isoforms indicates specific functional roles for the long and short forms of the protein. In neuronal cells, the short form of Bm-3b acts predominantly as a repressor of transcription of the genes associated with differentiation (Budhram Mahadeo, Lillycrop & Latchman, 1995a; Morris et al, 1994). In contrast, the long form has been reported to activate promoters with octamer binding sites (Turner et al., 1994). Bm-3b can also activate the neuronal nicotinic acetylcholine receptor a2 gene (Milton et al., 1995; Milton et al., 1996),and the snyapsin I gene (Morris et al., 1996). Thus it may be reasonably suggested that the short and long forms of Bm-3b can have opposite functions. The unanticipated upregulation of Bm-3b long must be considered when analyzing these results involving differential gene expression. In addition, immunoblots revealed an upregulation of Bm-3 a long. Bm-3 a and Bm-3b have long been known to act on the same promoters in an opposite and antagonistic manner. As mentioned above Bm-3b may have a stimulatory or repressive effects on transcription. Bm-3 a can activate the SNAP-25 (Morris et al., 1996), the a intemexin (Budhram Mahadeo et al, 1996; Budhram Mahadeo et al., 1995b), and the three neurofilament (Smith et al, 1997d) promoters, while Bm-3b not only has a repressive

262 effect, but also inhibits the stimulatory effect of Bm-3 a. The effect of increased Bm-3 a long must also be kept in mind when analyzing the differential protein expression results.

V.3.1.B The levels of B-actin are altered in the cell lines with altered levels of Brn- 3K

The values for the fold expression of proteins investigated by Westem immunoblot are expressed both as raw values and a value normalized for each corresponding actin concentration. The cDNA array showed a 2.5 fold dovmregulation in the levels of cytoplasmic p-actin in the Bm-3b overexpressing cell lines. In general, actin levels are highest in the Bm-3b antisense cell line clones in the Westem immunoblotting experiments. It is because of this disparity that actin should not be used for normalizing relative concentrations of total protein in this system. Because equal total protein concentration was loaded onto the gels in all cases, the raw values for fold expression may be of more value for data analysis. In addition, the levels of Hsp90 and Hsp/Hsc70 were unchanged in each of the cell lines and are thus valuable in confirming equal protein loading in this set of experiments, and thus provide a useful control against which other protein increases or decreases may be judged. In all cases in the text, the range of protein expression encompasses the most extreme values in both the raw and normalized fold expression data.

V.3.1.C Two types of differential expression of estrogen responsive genes were shown by Western immunoblotting.

Both the levels of the Bm-3b proteins and the levels of the estrogen receptor are altered in the same direction in the cell lines (an increase in Bm-3b results in an increase in the estrogen receptor). It was unsurprising to note that the levels of Hsp 27 are upregulated in the same manner, as well. One may not, however differentiate between this upregulation as simply the result of increased estrogen receptor, or a Bm-3b mediated potentiation of the Hsp27 activation by the estrogen receptor.

263 In another Westem immunoblot, the protein levels of a different estrogen inducible gene were noted. BRCA I protein levels increased approximately two fold in the cell lines which overexpress Bm-3b. This was unexpected because, as mentioned earlier, primary breast tumor tissue with decreased levels of BRCA I has decreased levels of Bm-3b, and Bm-3b short has been shown to repress the activity of a BRCA I promoter construct (Budhram Mahadeo et al, 1999b). At least three laboratories (Gudas et al., 1995; Lane et al., 1995; Marquis et al, 1995) have reported an increase in BRCA I expression following estrogen treatment. Gudas and colleagues showed in Northern and Westem blots low levels of BRCA I in estrogen deprived MCF7 cells, and a gradual increase in BRCA I 8 - 24 hours after cells were fed with fresh medium containing estradiol (Gudas et al., 1995). The kinetics of this upregulation more closely paralleled the estrogen induced upregulation of the S phase regulated gene Cyclin A, than the more quickly upregulated pS2 gene which is specifically directly induced by estrogens in MCF7 cells (KaMioven et al., 1994). The authors conclude that the effects of the estrogens on BRCA I expression may be due to overall changes in the proliferative status of the MCF7 cells as opposed to the effect of a ligand induced estrogen receptor on DNA sequences in the BRCA I gene itself. These findings may be applied to an explanation of the upregulation of BRCA I in the Bm-3b overexpressing cell lines. The Bm-3b overexpressing cell lines proliferate at a greater rate than the control or antisense cell lines (see previous chapter). The increased proliferative status of the Bm-3b overexpressing cells may account for the upregulation of BRCA I in much the same manner as in the Gudas paper. An interesting experiment, then, would be to directly compare the levels of BRCA I in the overexpressing and control cell lines both with and without estradiol treatment. Experiments of this nature would reveal more clearly, the effect the Bm-3b has on the BRCA I promoter, and hopefully separate the similar effects of increased proliferation and transcriptional upregulation.

V.3.2 cDNA array

264 The molecular mechanisms which control cellular growth are not clearly understood. Growth can be altered by changing proliferation, apoptosis or differentiation. In normal tissue, cell number remains constant because of tight control over these three factors. In abnormal tissue, this balance can be deregulated such that increased cell number can result from either blocked death and/or differentiation pathways or by an increase in proliferation without a commensurate increase in the apoptosis or differentiation pathway. Both of these deregulated pathways play a role in carcinogenesis. The Atlas human cancer 1.2 array allowed some insight into the particular genes which were altered in the different clonal cell lines. With this information it is possible to propose mechanisms and pathways which account for the altered cellular growth characteristics of the Bm-3b altered clones. A few examples of the genes shown to be altered in the cDNA array are discussed in the context of their cellular function, and how that function may be linked with the alterations in Bm-3 proteins.

V.3.2.A Brn-3b and retinoic acid receptor share molecular pathways.

Retinoids, natural and synthetic vitamin A derivatives, modulate growth differentiation and development (Guzey et al, 1998a; Guzey, Sattler & DeLuca, 1998b), and mediate their effects through two classes of nuclear receptors, reitinoic acid receptors (RARs), and retinoid X receptors (RXRs). In 1994, Turner and colleagues showed that Bm-3b is downregulated in F9 cells treated with retinoic acid (RA) (Turner et al., 1994). This effect was paralleled but to a lesser extent in neuroblastoma cells. In both cases the effect was reversed by replacing the RA treated medium with fresh untreated medium. The level of Bm-3a in the cells was unaffected by the treatment (Turner et al., 1994). It is quite interesting to note, then, that the cDNA array revealed a 3.9 fold downregulation in RAR in the Bm-3b overexpressing cell line. Possibly Bm-3b and the RAR are involved in a reciprocal feedback regulation loop to maintain tight control on metabolic pathways in which both are involved. An example of a pathway which appears to be regulated by both follows.

265 Both retinoic acid receptor and paxillin are downregulated in the Bm-3b overexpressing cell line. The addition of RA to the estrogen responsive MCF7 cell line results in inhibited cell growth and increased adhesion to fibronectin, while ER negative breast cancer cell lines fail to respond in this way (Zhu et aL, 1999). Phosphorylation experiments showed that the addition of RA resulted in the tyrosine phosphorylation of focal adhesion kinase (FAK) and paxillin (a specific tyrosine residue within paxillin is a primary target for phosphorylation by FAK). The Bm-3b overexpressing cell line showed enhanced cell growth and decreased anchorage dependent growth. Moreover, both RAR and paxillin were downregulated on the cDNA array. These results indicate that Bm-3b may play a role in the RA activated pathways that lead to FAK and paxillin phosphorylation, and ultimately to increased cell growth and decreased adhesion dependence. Interestingly, increased tyrosine phosphorylation of FAK and paxillin have been shown to be responsive to NGF (Melamed et al., 1995) and the neuropeptides bombesin, endothelin, and vasopressin (Sinnett-Smith et al., 1993; Zachary et al., 1993) which may make this system relevant in neuronal systems as well.

V.3.2.B Brn-3 proteins play a role in auditory development.

Disruption of the Bm-3c gene has been shown to result in defects in the auditory system with the failure of the cochlear and vestibular hair cells of the inner ear to differentiate (Brkman et al., 1996). It was interesting, then, that the Bm-3b overexpressing cell line exhibited a downregulation in the Notch receptor family ligand Jagged2. Notch signaling mediated by Jagged2 plays an essential role during limb, craniofacial, and thymic development in mice (Bell et al, 1998; Jiang et al, 1998). Moreover, Notch 1 and Jagged2 are expressed in alternating cell types in the developing sensory epithelium of the mammalian cochlea (the Organ of Corti). The development of the epithelial cells here exhibits a precise mosaic patterning and the determination of cell fates in the cochlear mosaic occurs via inhibitory interactions between adjacent cells; developing hair cells suppress differentiation in their immediate neighbors through lateral inhibition. Genetic deletion of Jagged2 results in a significant increase in the sensory hair cells, presumably resulting fi"om a decrease in Notch activation (Lanford et al., 1999). Bm-3b

266 is associated with cellular proliferation (Smith & Latchman, 1996) and Bm-3c is associated with cochlear hair cell differentiation (Xiang et al, 1998). These results would support the idea that the Jagged2 gene could be either directly or indirectly regulated by Bm-3 proteins, possibly with Bm-3b repressing and Bm-3c activating it to maintain a proliferative state, or induce differentiation respectively.

V.3.2.C RhoC plays a role in both neuronal differentiation and tumor metastasis.

The small GTP-binding protein, RhoC, is upregulated 1.9 fold in the Bm-3b overexpressing cell line. Calvet and colleagues showed that RhoA and RhoC along with other proteins play a critical role in the heparin signal transduction pathway which can control axonal and dendritic growth (Calvet, Doherty & Prochiantz, 1998). Recently, members of the small G-protein family were shown to play a role in the estradiol induced differentiation of SK-N-BE neuroblastoma cells. In cells transfected with ERa, differentiation proceeds with increased length and number of neurite processes, while in cells transfected with ERp differentiation proceeds with neurite elongation only. The effects of the different ERs was then tested in the presence of the wild type or dominant negative form of the small G-protein Raclb. ERa dependent, but not ER|3 dependent, differentiation events are observed only in the presence of the active form of the small G- protein (Patrone et al., 2000). Bm-3 a and Bm-3b are key players in the regulation of proliferation and differentiation in neuronal cells (Smith et al, 1997c; Smith & Latchman, 1996). In the ND7 cell line, which can be induced to differentiate from a proliferating cell type to a non-dividing neuronal-like cell bearing numerous processes, the level of Bm-3 a increases from a very low level in the undifferentiated cell, to a high level in the differentiated cell. Likewise, the high levels of Bm-3b in the proliferating cell drop to a much lower level in the differentiated cell (Budhram Mahadeo et al, 1994; Lillycrop et al, 1992). Unsurprisingly, Bm-3a is able to activate promoters of genes encoding SNAP-25 (Lakin et al, 1995), the intermediate filament protein a intemexin (Budhram Mahadeo et al, 1996; Budhram Mahadeo et al, 1995b), and the three neurofilaments (Smith et al, 1997d), and these same promoters are repressed by Bm-3b. The interesting question

267 becomes: could the Bm-3 proteins be responsible for the direct or indirect induction of the small GTP-binding proteins which are involved in actin polymerization which accompanies changes in the cytoarchitecture such as the generation and lengthening of neurite processes. Moreover, the action of ERa is dependent upon the small G-protein Raclb. The Bm-3 proteins interact with ERa and potentiate its transcriptional effects. Possibly the Bm-3 proteins, ERa, and the small GTP-binding proteins are together involved in the regulation and control of neuriteogenesis in neuronal cells.

In an unrelated set of experiments, RhoC was implicated as playing an essential role in tumor metastasis. In these experiments, the gene expression profiles of melanoma variants with high or low metastatic potential were compared on a cDNA array. RhoC was upregulated in pulmonary metastatic tumors of poorly metastatic cell lines when compared to the parental cell line non-metastatic subcutaneous tumor xenografts. RhoC cloned into a retrovims vector was then used to infect a poorly metastatic cell line. Cells expressing high levels of RhoC were isolated using FACS and then subjected to an experimental metastasis assay. The cells expressing RhoC had a markedly increased metastatic capability (Clark et al, 2000). The Bm-3b overexpressing cells have an increased ability to grow in an anchorage independent manner while the Bm-3b antisense cells consistently grow less well in soft agar studies. With the knowledge that the Bm-3 proteins act in related pathways and at related times in the neuronal as the small G-proteins, it is interesting to propose that the Bm-3 proteins may play a role in the pathways involved in gross changes in cytoarchitecture and may mediate their effects through pathways that utilize the small GTP-binding proteins.

V.3.2.D Early growth response protein 1 (hEGRI)

Discussion of the relationship between Bm-3 proteins and early growth response protein 1 (hEGRI) may be found in the Conclusion, Chapter 6.

268 V.3.3 Conclusion The identification of families of genes involved in different gene programs that control the different phenotypes of the Bm-3b altered cell lines opens a vast number of varied research opportunities. The role of the Bm-3 proteins in cellular proliferation, hormone responsiveness, cellular adhesion, cellular mosaic patterning, and changes in cytoarchitecture are only the most immediately interesting topics generated by the work of this chapter. The specific transcriptional mechanisms by which these changes occur have not been defined. Previous work describing a transcriptionally fimctional interaction between Bm-3 proteins and the ER, and work discussed herein describing a functional interaction between Bm-3 proteins and the pl60/SRCs must be included in any discussion.

269 VI CONCLUSION

VI. 1 Discussion

Cells respond to a diverse set of stimuli via alterations in gene expression. While gene expression may be regulated at any of the several steps of its production, transcription is accepted as one of the most fundamental points at which gene regulation occurs. Eukaryotic transcription is an intricate process involving several factors which function with spatio-temporal specificity to achieve particular patterns of gene expression. In the cell, several large multiprotein complexes including various components of the transcriptional machinery have been identified as combinatorial regulators of gene expression. The POU family of transcription factors is one family whose members play a role in function as as diverse as cellular homeostasis (i.e. Oct-1 regulation of the H2B promoter (Pierani et al., 1990)) as well as specific spatio-temporal developmental processes (i.e. Bm-3 control of the differentiation of cells derived from the neural crest (Smith et al., 1997b; Smith et al., 1997c)). Within the POU family of transcription factors, the Bm-3 subfamily represents a particularly potent group of flexible transcription factors. Originally identified in neuronal tissue, the Bm-3 proteins have since been identified in the tissues of the breast, ovary, utems and testes. Moreover, the Bm-3 members’ catalogue of protein partners has grown as well. The Bm-3 proteins have been shown to interact with the ER (Budhram Mahadeo et al., 1998) as well as other proteins. The interaction with the ER led this lab to investigate whether the Bm-3/ER interaction extended to the interaction of Bm-3 proteins with other factors involved in mediating estrogen-responsive transcription. In general, ligand-bound ER is able to recmit coactivator complexes and subsequently activate transcription. The p i60 family of transcription factors is one family of these coactivator molecules. This family consists of three members: Src-1, Src-2, and Src-3, which interact with the ligand-bound ligand binding domain of steroid receptors through a centrally located series of LXXLL motifs in the Src protein. Chapter III discusses the

270 investigation of physical and functional interactions between Bm-3 transcription factors and Src proteins. Physical and functional interactions occur between Bm-3 transcription factors and p i60 coactivators. Depending on the promoter or cellular context, these proteins may exist in complexes that activate, repress, or have no effect on transcription. Moreover, these factors are able to act in both estrogen dependent and estrogen independent pathways. The transcriptional versatility of these factors is clear. These results broaden the transcriptional flexibility of the Bm-3 transcription factors, and strengthen the notion that Bm-3 proteins may play a significant role in ER-mediated transcriptional regulation. It has recently been shown that the Bm-3b POU family transcription factor is overexpressed in several cases of sporadic breast cancer compared to its level in normal mammary epithelium. Additionally, the cases of breast cancer which show elevated levels of Bm-3b, also show reduced levels of the anti-oncogene BRCAI. These results suggest that Bm-3b may play a role in breast cancer. Moreover, the implication of ER in the development of normal and neoplastic breast tissue, combined with the discovery that Bm-3b can functionally interact with both the ER and Srcs, led to the investigation of the role of Bm-3b in estrogen receptor positive breast cancer. Chapter IV reports the cellular growth and endocrine characteristics of MCF7 ER positive breast epithelial cell lines with altered levels of Bm-3b. Alterations in the levels of Bm-3b have profound effects on the cellular growth characteristics of the MCF7 adenocarcinoma cell line. Upregulation of Bm-3b short (and upregulation of Bm-3b long) results in cells which have a net increase in growth rate. These cells have greater proliferation, saturation densities, and anchorage independent growth than their normal empty vector control counterparts. Moreover, these cells respond more strongly to estrogens and estrogen agonists, and are less debilitated by estrogen agonists than their normal counterparts. Reassuringly, MCF7 cells with decreased levels of Bm-3b long and short are not as proliferative as the control cells in any of the assays mentioned above. In addition, these are more responsive to estrogen antagonists than their empty pLTR control counterparts in all assays except anchorage- independent growth. In the anchorage-independent growth assays, the antisense clones

271 responded to tamoxifen with an increased ability to form anchorage-independent colonies. Because these several growth parameters showed such strong and clear differences in the cell lines with altered levels of Bm-3b, and because Bm-3b is such a potent and flexible transcription factor, alterations in gene expression in the Bm-3b altered cell lines were investigated, using Westem immunoblotting, differential display technology, and reverse transcription PCR. Upregulation of Bm-3b short results in an upregulation of the estrogen receptor as well as an estrogen responsive gene, Hsp27. Reassuringly, a comconitant decrease in each of these proteins was seen in the cell lines with reduced levels of Bm-3b. Several other genes were differentially regulated (including; transcription factors and immediate early genes, signal transduction proteins, proteins involved with cell cycle kinase programs, GTP-binding proteins, heat shock proteins, and cytoskeletal structural proteins). RT- PCR was used to successfully corroborate and confirm the regulation of three genes (one upregulated and two downregulated) chosen from the cDNA array.

VÏ.2 Viewing the three results chapters as a whole

While the results from each individual chapter may stand alone as interesting and significant, it is with the application of the cumulative information of the results from three related but distinct chapters that one may maximise their potential. An example follows:

Recent research has shown that the Src-3 protein AIBl is a phosphoprotein in vivo and can be phosphorylated in vitro by mitogen activating protein kinase (MAPK). The transcriptional activity of AIBl is enhanced by this MAPK phosphorylation. Finally, MAPK phosphorylation of AIBl stimulates its recmitment of the integrator p300 (Font de Mora & Brown, 2000). Thus, the authors conclude, the ability of growth factors to modulate estrogen action may be mediated through MAPK activation of AIBl.

272 Nerve growth factor (NGF) and its receptor signal by triggering a kinase cascade which results in the activation of MAPK and the ultimate induction of immediate early genes. One immediate early gene induced by NGF is NGFI-1 (or EGR-1, as it was referred to in the Clontech cancer array mentioned earlier). Bm-3 a has been shown to activate the EGR promoter in primary neuronal cultures as well as in neuronal cell lines. This activation is dependent upon a region in the promoter within its serum response element bridge region. Electromobility shift experiments, however, failed to identify a site bound directly by Bm-3 a, suggesting that the role of Bm-3 a may be to coordinate protein/protein interactions which regulate EGR-1 expression in neurons (Smith et al, 1999).

The p i60 protein Src-1 is able to interact with the serum response factor and regulate transcription fi*om the serum response element independent of a steroid receptor binding site (Kim et al, 1998a).

One possible hypothesis that may be drawn from these research papers and information found in each chapter of this thesis follows:

The mitogen, NGF, may induce a kinase cascade that results in the activation of MAPK. MAPK translocates to the nucleus where it phosphorylates Src-1. The phosphorylation of Src-1 stimulates the recmitment of p300 and this complex is able to integrate the regulatory transcription factors including Bm-3 a. Once a part of the multiprotein complex, Bm-3a is able to potentiate transcriptional activation of EGR-1, not fi*om the element that bears a homology to the Bm-3 binding site, but fi’om the serum response element fi*om which the p i60 protein has been shown to function.

VT.3 Future work

Several current papers support the notion that related but distinct multiprotein complexes are responsible for the different transcriptional activities of the p i60 coactivators, and it has been suggested that the p i60 activators should be viewed as integrator proteins that

273 coordinate the effects of several transcriptional regulatory proteins (Lee et ah, 1999; Lee et al., 1998b). The differential transcriptional effects noticed upon the addition of Bm-3 transcription factors to the p i60 transfections further supports the p 160-coactivators-as- multiprotein-integrators hypothesis. The transcriptional activity of Bm-3a and Bm-3b when added to the p i60 transfections was varied according to promoter and cellular context such that each could function as activator, repressor, or have no effects on promoter activity. The transfection systems described herein, then, provide an ideal opportunity to dissect the protein interactions and molecular mechanisms behind the differential effects of the transcription factors. Thus, experiments such as those involving competition, phosphorylation, complex formation kinetics, and complex composition should be undertaken. In addition, the effect of Bm-3/pl60 cotransfection should be investigated on each of the promoters which Bm-3 proteins have been shown to directly or indirectly regulate.

While the growth and proliferation of the cell lines with altered levels of Bm-3b has been investigated using several techniques, the cell death in each of these cell lines has not. Growth, alteration in the size of a cell mass is determined by five interrelated processes: proliferation, differentiation, cell death, cell contacts, and nutrient supply. To fully understand the growth kinetics of the Bm-3b altered cell lines, cell death must be accounted for. While there exists considerable disagreement in the scientific community around the most relevant indicators of cell death, several techniques exist and are easily available for this work.

The genes differentially regulated in the Bm-3b altered cell lines must be shown to be either directly or indirectly regulated by Bm-3 proteins. Further RT-PCR and Northem blot analysis of the regulation of these genes in different tissues and cell types must be undertaken. Promoter studies of these genes regulated by Bm-3 proteins must be undertaken to identify both the mechanism of regulation and the physiological context in which this regulation takes place.

274 VI.4 Conclusions Bm-3 transcription factors can functionally interact with steroid receptor coactivators and this interaction allows for greater flexibility in transcriptional regulation by each of these families of factors. This work broadens the catalogue of transcriptional activity of both families of proteins by producing evidence that both families can function outside their originally defined areas of transcriptional activity. Moreover, the functional interaction between these proteins results in promoter and cell-type specific transcriptional regulation. Thus, it is clear from the results that the already diverse transcriptional effects of the Bm-3 proteins have further grown to include regulation of transcription with the pi 60 family of steroid receptor coactivators. Some of the physiological implications of this increased transcriptional regulatory activity are revealed in the work with the MCF-7 breast cancer model. Bm-3b plays a key role in the growth and proliferation of these human breast cancer cells in vitro and enhances their ability to grow in an anchorage independent manner in vivo. This work provides a firm link to the previous studies that have shown that Bm-3b is overexpressed in primary breast cancer tissue samples. Further investigations have been initiated to determine the pathways and transcriptional mechanisms affected by the altered levels of Bm-3b in vivo. The further analysis of Bm- 3b should greatly enhance our understanding of the mechanisms of human breast cancer growth, proliferation, and endocrine response, and may potentially identify Bm-3 proteins a therapeutic target for breast cancer.

275 BIBLIOGRAPHY Adams, D. J., Hajj, H., Bitar, K. G., Edwards, D. P. & McGuire, W. L. (1983). Purification of an estrogen-regulated breast cancer protein by monoclonal antibody affinity chromatography. Endocrinology 113,415-7. Albertinazzi, C., Gilardelli, D., Paris, S., Longhi, R. & de Curtis, I. (1998). Overexpression of a neural-specific rho family GTPase, cRaclB, selectively induces enhanced neuritogenesis and neurite branching in primary neurons. J Cell 5/0/142,815-25. Ali, S., Metzger, D., Bomert, J. M. & Chambon, P. (1993). Modulation of transcriptional activation by ligand-dependent phosphorylation of the human oestrogen receptor A/B region. Embo J 12, 1153-60. Alland, L., Muhle, R., Hou, H., Jr., Potes, J., Chin, L., Schreiber-Agus, N. & DePinho, R. A. (1997). Role for N-CoR and histone deacetylase in Sin3-mediated transcriptional repression [see comments]. Nature 387,49-55. Anolik, J. H., Klinge, C. M., Bambara, R. A. & Hilf, R. (1993). Differential impact of flanking sequences on estradiol- vs 4-hydroxytamoxifen-liganded estrogen receptor binding to estrogen responsive element DNA. J Steroid Biochem Mol 5/0/46,713-30. Anolik, J. H., Klinge, C. M., Brolly, C. L., Bambara, R. A. & Hilf, R. (1996). Stability of the ligand-estrogen receptor interaction depends on estrogen response element flanking sequences and cellular factors. J Steroid Biochem Mol Biol 59,413-29. Anzick, S. L., Kononen, J., Walker, R. L., Azorsa, D. O., Tanner, M. M., Guan, X. Y., Sauter, G., Kallioniemi, O. P., Trent, J. M. & Meltzer, P. S. (1997). AIBl, a steroid receptor coactivator amplified in breast and ovarian cancer. Science 277, 965-8. Arnold, S. F. & Notides, A. C. (1995). An antiestrogen: a phosphotyrosyl peptide that blocks dimerization of the human estrogen receptor. Proc Natl Acad Sci U S A92^ 7475-9. Arnold, S. P., Oboum, J. D., Jaffe, H. & Notides, A. C. (1995a). Phosphorylation of the human estrogen receptor by mitogen-activated protein kinase and casein kinase II: consequence on DNA binding. J Steroid Biochem Mol Biol 55, 163-72. Arnold, S. P., Oboum, J. D., Jaffe, H. & Notides, A. C. (1995b). Phosphorylation of the human estrogen receptor on tyrosine 537 in vivo and by src family tyrosine kinases in vitro. Mol Endocrinol 9^ 24-33. Arnold, S. P., Oboum, J. D., Yudt, M. R., Carter, T. H. & Notides, A. C. (1995c). In vivo and in vitro phosphorylation of the human estrogen receptor. J Steroid Biochem Mol Biol SI, 159-71. Amold, S. P., Vorojeikina, D. P. & Notides, A. C. (1995d). Phosphorylation of tyrosine 537 on the human estrogen receptor is required for binding to an estrogen response element. J Biol Chem 270, 30205-12. Amosti, D. N., Merino, A., Reinberg, D. & Schaf&ier, W. (1993). Oct-2 facilitates functional preinitiation complex assembly and is continuously required at the promoter for multiple rounds of transcription. Embo J 12, 157-66. Artinger, K. B., Pedtsova, N., Rhee, J. M., Bronner-Praser, M. & Tumer, E. (1998). Placodal origin of Bm-3-expressing cranial sensory neurons. JNeurobiol 36, 572- 85.

276 Ashbumer, M. (1980). Chromosomal action of ecdysone. Nature 285,435-6. Assa Munt, N., Mortishire Smith, R. J., Aurora, R., Herr, W. & Wright, P. E. (1993). The solution structure of the Oct-1 POU-specific domain reveals a striking similarity to the bacteriophage lambda repressor DNA-binding domain. Cell 73, 193-205. Augereau, P., Miralles, P., Cavailles, V., Gaudelet, C., Parker, M. & Rochefort, H. (1994). Characterization of the proximal estrogen-responsive element of human cathepsin D gene. Mol Endocrinol 8, 693-703. Aurora, R. & Herr, W. (1992). Segments of the POU domain influence one another's DNA-binding specificity. Mol Cell Biol 12,455-67. Awgulewitsch, A. & Jacobs, D. (1992). Deformed autoregulatory element from Drosophila functions in a conserved manner in transgenic mice. Nature 358, 341- 4. Barkhem, T., Carlsson, B., Nilsson, Y., Enmark, E., Gustafsson, J. & Nilsson, S. (1998). Differential response of estrogen receptor alpha and to partial estrogen agonists/antagonists. Mol Pharmacol 54, 105-12. Baulieu, E. E. (1987). Steroid hormone antagonists at the receptor level: a role for the heat-shock protein MW 90,000 (hsp 90). J Cell Biochem 35, 161-74. Bautista, S., Valles, H., Walker, R. L., Anzick, S., Zeillinger, R., Meltzer, P. & Theillet, C. (1998). In breast cancer, amplification of the steroid receptor coactivator gene AIBl is correlated with estrogen and progesterone receptor positivity. Clin Cancer Res 4, 2925-9. Bazett-Jones, D. P., Leblanc, B., Herfort, M. & Moss, T. (1994). Short-range DNA looping by the Xenopus HMG-box transcription factor, xUBF [see comments]. Science 264, 1134-7. Beato, M. & Sanchez-Pacheco, A. (1996). Interaction of steroid hormone receptors with the transcription initiation complex. Endocr Rev 17, 587-609. Beckerle, M. C. (1997). Zyxin: zinc fingers at sites of cell adhesion. Bioessays 19, 949- 57. Bell, S. M., Lan, Y., Jiang, R. & Gridley, T. (1998). Exclusion of Jagged2 as a candidate for the legless gene. Mamm Genome 9, 778-9. Bendall, A. J., Sturm, R. A., Danoy, P. A. & Molloy, P. L. (1993). Broad binding-site specificity and affinity properties of octamer 1 and brain octamer-binding proteins. EurJBiochem 217, 799-811. Berg, J. M. (1988). Proposed structure for the zinc-binding domains from transcription factor IIIA and related proteins. Proc Natl Acad Sci U SA 85, 99-102. Bergson, C. & McGinnis, W. (1990). An autoregulatory enhancer element of the Drosophila homeotic gene Deformed. Emho J 9 ,4287-97. Berk, A. & Mattaj, I. W. (1997). Nucleus and gene expression [editorial]. Curr Opin Cell BW 9, 307-9. Bems, E. M., van Staveren, I. L., Klijn, J. G. & Foekens, J. A. (1998). Predictive value of SRC-1 for tamoxifen response of recurrent breast cancer. Breast Cancer Res Treat 48, 87-92. Berry, M., Metzger, D. & Chambon, P. (1990). Role of the two activating domains of the oestrogen receptor in the cell-type and promoter-context dependent agonistic activity of the anti-oestrogen 4-hydroxytamoxifen. Embo J 9,2811-8.

277 Berry, M., Nunez, A. M. & Chambon, P. (1989). Estrogen-responsive element of the human pS2 gene is an imperfectly palindromic sequence. Proc Natl Acad Sci U S ^ 86, 1218-22. Berthois, Y., Katzenellenbogen, J. A. & Katzenellenbogen, B. S. (1986). Phenol red in tissue culture media is a weak estrogen: implications concerning the study of estrogen-responsive cells in culture. Proc Natl Acad Sci U S AH3, 2496-500. Berwaer, M., Martial, J. A. & Davis, J. R. (1994). Characterization of an up-stream promoter directing extrapituitary expression of the human prolactin gene. Mol Endocrinol 8, 635-42. Bindal, R. D., Carlson, K. E., Katzenellenbogen, B. S. & Katzenellenbogen, J. A. (1988). Lipophilic impurities, not phenolsulfonphthalein, account for the estrogenic activity in commercial preparations of phenol red. J Steroid Biochem 3 1 ,287-93. Bindal, R. D. & Katzenellenbogen, J. A. (1988). Bis(4-hydroxyphenyl)[2- (phenoxysulfonyl)phenyl]methane: isolation and structure elucidation of a novel estrogen from commercial preparations of phenol red (phenolsulfonphthalein) [published erratum appears in J Med Chem 1989 Dec;32(12):2582]. J Med Chem 31, 1978-83. Bimboim, H. C. & Doly, J. (1979). A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res 7, 1513-23. Bishop, J. M. (1987). The molecular genetics of cancer. Science 235, 305-11. Bishop, W. H., Nureddin, A. & Ryan, R. J. (1976). Pituitary luteinising and follicle- stimulating hormones, pp. 273-98. In: Parsons JA, ed. Peptide hormones. Baltimore, Univ Park Press, . Bissell, M. J. & Werb, Z. (1995). Introduction: basic science aspects of breast cancer. Semin Cancer Biol 6, 117-8. Blackwood, E. M. & Kadonaga, J. T. (1998). Going the distance: a current view of enhancer action. Science 2 8 1 ,61-3. Blom van Assendelft, G., Hanscombe, O., Grosveld, F. & Greaves, D. R. (1989). The beta-globin dominant control region activates homologous and heterologous promoters in a tissue-specific manner. Cell 56, 969-77. Bohen, S. P. & Yamamoto, K. R. (1993). Isolation of Hsp90 mutants by screening for decreased steroid receptor fimction. Proc Natl Acad Sci USA 90, 11424-8. Botfield, M. C., Jancso, A. & Weiss, M. A. (1992). Biochemical characterization of the Oct-2 POU domain with implications for bipartite DNA recognition. Biochemistry 31,5841-8. Botfield, M. C., Jancso, A. & Weiss, M. A. (1994a). An invariant asparagine in the POU- specific homeodomain regulates the specificity of the Oct-2 POU motif. Biochemistry 33, 8113-21 . Botfield, M. C., Jancso, A. & Weiss, M. A. (1994b). Mapping critical residues in eukaryotic DNA-binding proteins: a plasmid-based genetic selection strategy with application to the Oct-2 POU motif. Biochemistry 33,6177-85. Botfield, M. C. & Weiss, M. A. (1994). Bipartite DNA recognition by the human Oct-2 POU domain: POUs-specific phosphate contacts are analogous to those of bacteriophage lambda repressor. Biochemistry 3 3 ,2349-55.

278 Bourguet, W., Ruff, M., Chambon, P., Gronemeyer, H. & Moras, D. (1995). Crystal structure of the ligand-binding domain of the human nuclear receptor RXR-alpha [see comments]. Nature 375, 377-82. Brooks, S. C., Locke, E. R. & Soule, H. D. (1973). Estrogen receptor in a human cell line (MCF-7) from breast carcinoma. Journal o f Biological Chemistry 248, 6251- 6253. Brown, A. M., Jeltsch, J. M., Roberts, M. & Chambon, P. (1984). Activation of pS2 gene transcription is a primary response to estrogen in the human breast cancer cell line MCF-7. Proc Natl Acad Sci U SA^\, 6344-8. Brown, R. S., Sander, C. & Argos, P. (1985). The primary structure of transcription factor TFIIIA has 12 consecutive repeats. FEES Lett 186,271-4. Bruggemeier, U., Kalff, M., Franke, S., Scheidereit, C. & Beato, M. (1991). Ubiquitous transcription factor OTF-1 mediates induction of the MMTV promoter through synergistic interaction with hormone receptors. Cell 64, 565-572. Brzozowski, A. M., Pike, A. C., Dauter, Z., Hubbard, R. E., Bonn, T., Engstrom, O., Ohman, L., Greene, G. L., Gustafsson, J. A. & Carlquist, M. (1997). Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 389, 753-8. Buchner, A., Obemeder, R., Riesenberg, R., Keiditsch, E. & Hofstetter, A. (1998). Expression of plakoglobin in renal cell carcinoma. Anticancer Res 18,4231-5. Budhram Mahadeo, V., Lillycrop, K. A. & Latchman, D. S. (1995a). The levels of the antagonistic POU family transcription factors Bm-3a and Bm-3b in neuronal cells are regulated in opposite directions by serum growth factors. Neurosci Lett 185, 48-51. Budhram Mahadeo, V., Morris, P. J., Lakin, N. D., Dawson, S. J. & Latchman, D. S. (1996). The different activities of the two activation domains of the Bm-3 a transcription factor are dependent on the context of the binding site. J Biol Chem 271,9108-13. Budhram Mahadeo, V., Morris, P. J., Lakin, N. D., Theil, T., Ching, G. Y., Lillycrop, K. A., Moroy, T., Liem, R. K. & Latchman, D. S. (1995b). Activation of the alpha- intemexin promoter by the Bm-3 a transcription factor is dependent on the N- terminal region of the protein. J Biol Chem 270,2853-8. Budhram Mahadeo, V., Morris, P. J., Smith, M. D., Midgley, C. A., Boxer, L. M. & Latchman, D. S. (1999a). p53 suppresses the activation of the Bcl-2 promoter by the Bm-3 a POU family transcription factor. J Biol Chem 274, 15237-44. Budhram Mahadeo, V., Ndisang, D., Ward, T., Weber, B. L. & Latchman, D. S. (1999b). The Bm-3b POU family transcription factor represses expression of the BRCA-1 anti-oncogene in breast cancer cells. Oncogene 18, 6684-91. Budhram Mahadeo, V., Parker, M. & Latchman, D. S. (1998). POU transcription factors Bm-3 a and Bm-3b interact with the estrogen receptor and differentially regulate transcriptional activity via an estrogen response element. Mol Cell Biol 18, 1029- 41. Budhram Mahadeo, V., Theil, T., Morris, P. J., Lillycrop, K. A., Moroy, T. & Latchman, D. S. (1994). The DNA target site for the Bm-3 POU family transcription factors can confer responsiveness to cyclic AMP and removal of serum in neuronal cells. Nucleic Acids Res 22, 3092-8.

279 Budhram-Mahadeo-Heads, V. (1995). Regulation and Function of POU domain transcription factors, Bm-3a and Bm-3b. Ph.D. thesis, University Colege London. Buratowski, S., Hahn, S., Guarente, L. & Sharp, P. A. (1989). Five intermediate complexes in transcription initiation by RNA polymerase II. Cell 56, 549-61. Burley, S. K., Clark, K. L., Ferre-D'Amare, A., Kim, J. L. & Nikolov, D. B. (1993). X- ray crystallographic studies of eukaryotic transcription factors. Cold Spring Harb Symp Quant Biol 58, 123-32. Cai, B., Tomida, A., Mikami, K., Nagata, K. & Tsuruo, T. (1998). Down-regulation of epidermal growth factor receptor-signaling pathway by binding of GRP78/BiP to the receptor under glucose-starved stress conditions. / Ce// Physiol 177,282-8. Calvet, S., Doherty, P. & Prochiantz, A. (1998). Identification of a signaling pathway activated specifically in the somatodendritic compartment by a heparan sulfate that regulates dendrite growth. J Neurosci 18, 9751-65. Cantor, G. H., Stone, D. M., McElwain, T. F. & Pahner, G. H. (1996). Comparison of the antiviral efficacy of ribozymes and antisense RNA directed against bovine leukemia virus rex/tax. Antisense Nucleic Acid Drug Dev 6, 301-4. Capecchi, M. R. (1989). Altering the genome by homologous recombination. Science 244, 1288-92. Caplan, A. J., Langley, E., Wilson, E. M. & Vidal, J. (1995). Hormone-dependent transactivation by the human androgen receptor is regulated by a dnaJ protein. J Biol Chem 270, 5251-7. Carey, M., Kakidani, H., Leatherwood, J., Mostashari, F. & Ptashne, M. (1989). An amino-terminal fi*agment of GAL4 binds DNA as a dimer. J Mol Biol 209, 423- 32. Chen, H., Lin, R. J., Schiltz, R. L., Chakravarti, D., Nash, A., Nagy, L., Privalsky, M. L., Nakatani, Y. & Evans, R. M. (1997). Nuclear receptor coactivator ACTR is a novel histone acetyltransferase and forms a multimeric activation complex with P/CAF and CBP/p300. Cell 90, 569-80. Chen, H., Lin, R. J., Xie, W., Wilpitz, D. & Evans, R. M. (1999a). Regulation of hormone-induced histone hyperacetylation and gene activation via acétylation of an acetylase. Cell 98,675-86. Chen, J. D. & Evans, R. M. (1995). A transcriptional co-repressor that interacts with nuclear hormone receptors [see comments]. Nature 377,454-7. Chen, Y. C., Lin-Shiau, S. Y. & Lin, J. K. (1999b). Involvement of heat-shock protein 70 and P53 proteins in attenuation of UVC-induced apoptosis by thermal stress in hepatocellular carcinoma cells. Photochem Photohiol 70, 78-86. Chiba, H., Muramatsu, M., Nomoto, A. & Kato, H. (1994). Two human homologues of Saccharomyces cerevisiae SWI2/SNF2 and Drosophila brahma are transcriptional coactivators cooperating with the estrogen receptor and the retinoic acid receptor. Nucleic Acids Res 22,1815-20. Ciocca, D. R., Adams, D. J., Edwards, D. P., Bjercke, R. J. & McGuire, W. L. (1983). Distribution of an estrogen-induced protein with a molecular weight of 24,000 in normal and malignant human tissues and cells. Cancer Res 43, 1204-10. Ciocca, D. R., Clark, G. M., Tandon, A. K., Fuqua, S. A., Welch, W. J. & McGuire, W. L. (1993). Heat shock protein hsp70 in patients with axillary lymph node-negative breast cancer: prognostic implications. JNatl Cancer Inst 85, 570-4.

280 Ciocca, D. R., Puy, L. A. & Fasoli, L. C. (1989). Study of estrogen receptor, progesterone receptor, and the estrogen-regulated Mr 24,000 protein in patients with carcinomas of the endometrium and cervix. Cancer Res 49,4298-304. Clark, E. A., Golub, T. R., Lander, E. S. & Hynes, R, O. (2000). Genomic analysis of metastasis reveals an essential role for RhoC [In Process Citation]. Nature 406, 532-5. Clark, K. L., Halay, E. D., Lai, E. & Burley, S. K. (1993). Co-crystal structure of the HNF-3/fork head DNA-recognition motif resembles histone H5. Nature 364, 412- 20. Cleary, M. A. & Herr, W. (1995). Mechanisms for flexibility in DNA sequence recognition and VP16-induced complex formation by the Oct-1 POU domain. Mol Cell Biol 15, 2090-100. Cleary, M. A., Stem, S., Tanaka, M. & Herr, W. (1993). Differential positive control by Oct-1 and Oct-2: activation of a transcriptionally silent motif through Oct-1 and VP 16 corecruitment. Genes Dev 7,72-83. Coenjaerts, F. E,, van Oosterhout, J. A. & van der Vliet, P. C. (1994). The Oct-1 POU domain stimulates adenovirus DNA replication by a direct interaction between the viral precursor terminal protein-DNA polymerase complex and the POU homeodomain. Embo J 13, 5401-9. Collum, R. G., Fisher, P. E., Datta, M., Mellis, S., Thiele, C., Huebner, K., Croce, C. M., Israel, M. A., Theil, T., Moroy, T. & et al. (1992). A novel POU homeodomain gene specifically expressed in cells of the developing mammalian nervous system. Nucleic Acids Res 20,4919-25. Colman, A. (1990). Antisense strategies in cell and developmental biology. J Cell Sci 97, 399-409. Cote, J., Quinn, J., Workman, J. L. & Peterson, C. L. (1994). Stimulation of GAL4 derivative binding to nucleosomal DNA by the yeast SWI/SNF complex. Science 265, 53-60. Cowley, S. M., Hoare, S., Mosselman, S. & Parker, M. G. (1997). Estrogen receptors alpha and beta form heterodimers on DNA. J Biol Chem 272, 19858-62. Cowley, S. M. & Parker, M. G. (1999). A comparison of transcriptional activation by ER alpha and ER beta. J Steroid Biochem Mol Biol 69, 165-75. Danielian, P. S., White, R., Lees, J. A. & Parker, M. G. (1992). Identification of a conserved region required for hormone dependent transcriptional activation by steroid hormone receptors [published erratum appears in EMBO J 1992 Jun;ll(6):2366].Em6oyil, 1025-33. Danzeiser, D. A., Urso, O. & Kunkel, G. R. (1993). Functional characterization of elements in a human U6 small nuclear RNA gene distal control region. Mol Cell Biol 13, 4670-8. Darimont, B. D., Wagner, R. L., Apriletti, J. W., Stallcup, M. R., Kushner, P. J., Baxter, J. D., Fletterick, R. J. & Yamamoto, K. R. (1998). Structure and specificity of nuclear receptor-coactivator interactions. Genes Dev 12, 3343-56. Davidoff, A. M., Iglehart, J. D. & Marks, J. R. (1992). Immune response to p53 is dependent upon p53/HSP70 complexes in breast cancers. Proc Natl Acad Sci U S A 89, 3439-42.

281 Dawson, S. J., Liu, Y. Z., Rodel, B., Moroy, T. & Latchman, D. S. (1996a). The ability of POU family transcription factors to activate or repress gene expression is dependent on the spacing and context of their specific response elements. Biochem /3 1 4 ,439-43. Dawson, S. J., Morris, P. J. & Latchman, D. S. (1996b). A single amino acid change converts an inhibitory transcription factor into an activator. J Biol Chem 271, 11631-3. Dawson, S. J., Palmer, R. D., Morris, P. J. & Latchman, D. S. (1998). Functional role of position 22 in the homeodomain of Bm-3 transcription factors. Neuroreport 9, 2305-9. Deblandre, G. A., Marinx, O. P., Evans, S. S., Majjaj, S., Leo, O., Caput, D., Huez, G. A. & Wathelet, M. G. (1995). Expression cloning of an interferon-inducible 17-kDa membrane protein implicated in the control of cell growth. J Biol Chem 270, 23860-6. deBoer, E., Antoniou, M., Mignotte, V., Wall, L. & Grosveld, F. (1988). The human beta-globin promoter; nuclear protein factors and erythroid specific induction of transcription. EMBO J l^ 4203-12. Dekker, N., Cox, M., Boelens, R., Verrijzer, C. P., van der Vliet, P. C. & Kaptein, R. (1993). Solution stmcture of the POU-specific DNA-binding domain of Oct-1. Nature 362, 852-5. Denhardt, D. T. (1992a). Antisense strategies come of age. Antisense strategies sponsored by the New York Academy of Sciences, Philadelphia, PA, USA, January 12-15, 1992. New Biol 4,473-81. Denhardt, D. T. (1992b). Mechanism of action of antisense RNA. Sometime inhibition of transcription, processing, transport, or translation. Ann N Y Acad Sci 660, 70-6. Dickson, R. B., Kasid, A., Huff, K. K., Bates, S. E., Knabbe, C., Bronzert, D., Gelmann, E. P. & Lippman, M. E. (1987). Activation of growth factor secretion in tumorigenic states of breast cancer induced by 17 beta-estradiol or v-Ha-ras oncogene. Proc Natl Acad Sci U SA 84, 837-41. Dickson, R. B. & Lippman, M. E. (1987). Estrogenic regulation of growth and polypeptide growth factor secretion in human breast carcinoma. Endocr Rev 8, 29-43. Doucas, V., Spyrou, G. & Yaniv, M. (1991). Unregulated expression of c-Jun or c-Fos proteins but not Jun D inhibits oestrogen receptor activity in human breast cancer derived cells. Embo J 10,2237-45. Dura, J. M. & Ingham, P. (1988). Tissue- and stage-specific control of homeotic and segmentation gene expression in Drosophila embryos by the polyhomeotic gene. Development 103,733-41. Eckner, R., Yao, T. P., Oldread, E. & Livingston, D. M. (1996). Interaction and functional collaboration of p300/CBP and bHLH proteins in muscle and B-cell differentiation. Genes Dev 10,2478-90. El-Ashry, D., Chrysogelos, S. A., Lippman, M. E. & Kem, F. G. (1996). Estrogen induction of TGF-alpha is mediated by an estrogen response element composed of two imperfect palindromes. J Steroid Biochem Mol Biol 59,261-9.

282 Elledge, R. M., Clark, G. M., Fuqua, S. A., Yu, Y. Y. & Allred, D. C. (1994). p53 protein accumulation detected by five different antibodies: relationship to prognosis and heat shock protein 70 in breast cancer. Cancer Res 54, 3752-7. Elledge, R. M. & Osborne, C. K. (1997). Oestrogen receptors and breast cancer [editorial]. Bmj 314, 1843-4. Endoh, H., Maruyama, K., Masuhiro, Y., Kobayashi, Y., Goto, M., Tai, H., Yanagisawa, J., Metzger, D., Hashimoto, S. & Kato, S. (1999). Purification and identification of p68 RNA helicase acting as a transcriptional coactivator specific for the activation fimction 1 of human estrogen receptor alpha. Mol Cell Biol 19, 5363- 72. Erkman, L., McEvilly, R. J., Luo, L., Ryan, A. K., Hooshmand, F., O'Connell, S. M., Keithley, E. M., Rapaport, D. H., Ryan, A. F. & Rosenfeld, M. G. (1996). Role of transcription factors Bm-3.1 and Bm-3.2 in auditory and visual system development. Nature 381, 603-6. Fedtsova, N. & Tumer, E. E. (1997). Inhibitory effects of ventral signals on the development of Bm-3.0-expressing neurons in the dorsal spinal cord. Dev Biol 190, 18-31. Fedtsova, N. G. & Tumer, E. E. (1995). Bm-3.0 expression identifies early post-mitotic CNS neurons and sensory neural precursors. Mech Dev 53,291-304. Feinberg, A. P. & Vogelstein, B. (1983). A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 132, 6-13. Feng, W., Ribeiro, R. C., Wagner, R. L., Nguyen, H., Apriletti, J. W., Fletterick, R. J., Baxter, J. D., Kushner, P. J. & West, B. L. (1998). Hormone-dependent coactivator binding to a hydrophobic cleft on nuclear receptors. Science 280, 1747-9. Ferre-D'Amare, A. R., Prendergast, G. C., Ziff, E. B. & Burley, S. K. (1993). Recognition by Max of its cognate DNA through a dimeric b/HLH/Z domain [see comments]. Nature 363, 38-45. Fletcher, C., Heintz, N. & Roeder, R. G. (1987). Purification and characterization of OTF-1, a transcription factor regulating cell cycle expression of a human histone H2b gene. Cell 31, 773-81. Font de Mora, J. & Brown, M. (2000). AIBl is a conduit for kinase-mediated growth factor signaling to the estrogen receptor. Mol Cell Biol 20, 5041-7. Freedman, L. P., Luisi, B. F., Korszun, Z. R., Basavappa, R., Sigler, P. B. & Yamamoto, K. R. (1988a). The function and structure of the metal coordination sites within the glucocorticoid receptor DNA binding domain. Nature 334, 543-6. Freedman, L. P., Yamamoto, K. R., Luisi, B. F. & Sigler, P. B. (1988b). More fingers in hand [letter]. Cell 54,444. Friedmann, T. (1989). Progress toward human gene therapy. Science 244, 1275-81. Fuqua, S. A. (1992). Where is the lesion in hormone-independent breast cancer? [editorial; comment]. JNatl Cancer Inst 84, 554-5. Fuqua, S. A., Blum-Salingaros, M. & McGuire, W. L. (1989). Induction of the estrogen- regulated "24K" protein by heat shock. Cancer Res 49,4126-9. Fuqua, S. A., Fitzgerald, S. D., Allred, D. C., Elledge, R. M., Nawaz, Z., McDonnell, D. P., O'Malley, B. W., Greene, G. L. & McGuire, W. L. (1992). Inhibition of

283 estrogen receptor action by a naturally occurring variant in human breast tumors. Cancer Res 52,483-6. Cartel, A. L., Serfas, M. S. & Tyner, A. L. (1996). p21-negative regulator of the cell cycle. Proc Sac Exp Biol Med 213, 138-49. Gaub, M. P., Bellard, M., Scheuer, I., Chambon, P. & Sassone-Corsi, P. (1990). Activation of the ovalbumin gene by the estrogen receptor involves the fos-jun complex. Ce//63, 1267-76. Gentz, R., Rauscher, F. J. d.. Abate, C. & Curran, T. (1989). Parallel association of Fos and Jun leucine zippers juxtaposes DNA binding domains. Science 243, 1695-9. Gerrero, M. R., McEvilly, R. J., Tumer, E., Lin, C. R., O'Connell, S., Jenne, K. J., Hobbs, M. V. & Rosenfeld, M. G. (1993). Bm-3.0: a POU-domain protein expressed in the sensory, immune, and endocrine systems that functions on elements distinct from known octamer motifs. Proc Natl Acad Sci U SA 90, 10841-5. Ghadimi, B. M., Schrock, E., Walker, R. L., Wangsa, D., Jauho, A., Meltzer, P. S. & Ried, T. (1999). Specific chromosomal aberrations and amplification of the AIBl nuclear receptor coactivator gene in pancreatic carcinomas. Am J Pathol 154, 525-36. Giguere, V., Yang, N., Segui, P. & Evans, R. M. (1988). Identification of a new class of steroid hormone receptors. Nature 331, 91-4. Giniger, E., Vamum, S. M. & Ptashne, M. (1985). Specific DNA binding of GAL4, a positive regulatory protein of yeast. Cell 40, 767-74. Glass, C. K. (1994). Differential recognition of target genes by nuclear receptor monomers, dimers, and heterodimers. Endocr Rev 15, 391-407. Glass, C. K., Rose, D. W. & Rosenfeld, M. G. (1997). Nuclear receptor coactivators. Curr Opin Cell Biol 9,222-32. Glass, C. K. & Rosenfeld, M. G. (2000). The coregulator exchange in transcriptional functions of nuclear receptors. Genes Dev 14, 121-41. Golub, T. R., Slonim, D. K., Tamayo, P., Huard, C., Gaasenbeek, M., Mesirov, J. P., Coller, H., Loh, M. L., Downing, J. R., Caligiuri, M. A., Bloomfield, C. D. & Lander, E. S. (1999). Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 286, 531-7. Goodwin, G. H. (1990). Sequence specific DNA-binding proteins involved in transcription. In Chromosomes: Eukaryotic, Prokaryotic, and Viral, vol. 1 (ed. K. W. Adolph), pp. 31-85. CRC Press, Boca Raton, FL. Gorman, C. M., Merlino, G. T., Willingham, M. C., Pastan, I. & Howard, B. H. (1982). The Rous sarcoma vims long terminal repeat is a strong promoter when introduced into a variety of eukaryotic cells by DNA-mediated transfection. Proc Natl Acad Sci US Al% 6777-81. Graumaim, K. & Jungbauer, A. (2000). Quantitative assessment of complex formation of nuclear-receptor accessory proteins. Biochem J 345 Pt 3,627-36. Green, S., Walter, P., Greene, G., Kmst, A., Goffin, C., Jensen, E., Scrace, G., Waterfield, M. & Chambon, P. (1986a). Cloning of the human oestrogen receptor cDNA. J Steroid Biochem 24, 77-83. Green, S., Walter, P., Kumar, V., Kmst, A., Bomert, J. M., Argos, P. & Chambon, P. (1986b). Human oestrogen receptor cDNA: sequence, expression and homology to v-erb-A. Nature 320, 134-9.

284 Greenblatt, J. (1997). RNA polymerase II holoenzyme and transcriptional regulation. Curr Opin Cell Biol 9, 310-9. Greene, G. L., Gilna, P., Waterfield, M., Baker, A., Hort, Y. & Shine, J. (1986). Sequence and expression of human estrogen receptor complementary DNA. Science 231, 1150-4. Grosschedl, R., Giese, K. & Pagel, J. (1994). HMG domain proteins: architectural elements in the assembly of nucleoprotein structures. Trends Genet 10, 94-100. Grosveld, F., Antoniou, M., van Assendelft, G. B., de Boer, E., Hurst, J., Kollias, G., MacFarlane, F. & Wrighton, N. (1987). The regulation of expression of human beta-globin genes. Prog Clin Biol Res 251, 133-44. Gudas, J. M., Nguyen, H., Li, T. & Cowan, K. H. (1995). Hormone-dependent regulation of BRCAl in human breast cancer cells. Cancer Res 55,4561-5. Guzey, M., Demirpence, E., Criss, W. & DeLuca, H. F. (1998a). Effects of retinoic acid (all-trans and 9-cis) on tumor progression in small-cell lung carcinoma. Biochem Biophys Res Commun 242, 369-75. Guzey, M., Sattler, C. & DeLuca, H. F. (1998b). Combinational effects of vitamin D3 and retinoic acid (all trans and 9 cis) on proliferation, differentiation, and programmed cell death in two small cell lung carcinoma cell lines. Biochem Biophys Res Commun 249, 735-44. Hache, R. J., Wiskocil, R., Vasa, M., Roy, R. N., Lau, P. C. & Deeley, R. G. (1983). The 5' noncoding and flanking regions of the avian very low density apolipoprotein II and serum albumin genes. Homologies with the egg white protein genes. J Biol Chem 258, 4556-64. Halachmi, S., Marden, E., Martin, G., MacKay, H., Abbondanza, C. & Brown, M. (1994). Estrogen receptor-associated proteins: possible mediators of hormone- induced transcription. Science 264, 1455-8. Hamos, J. E., Oblas, B., Pulaski-Salo, D., Welch, W. J., Bole, D. G. & Drachman, D. A. (1991). Expression of heat shock proteins in Alzheimer's disease. Neurology 41, 345-50. Hankinson, O. (1995). The aryl hydrocarbon receptor complex. Annu Rev Pharmacol Toxicol 35, 307-40. Hard, T., Kellenbach, E., Boelens, R., Maler, B. A., Dahlman, K., Freedman, L. P., Carlstedt-Duke, J., Yamamoto, K. R., Gustafsson, J. A. & Kaptein, R. (1990). Solution structure of the glucocorticoid receptor DNA-binding domain. Science 249, 157-60. Harrison, S. C. (1991). A structural taxonomy of DNA-binding domains. Nature 353, 715-9. He, X., Treacy, M. N., Simmons, D. M., Ingraham, H. A., Swanson, L. W. & Rosenfeld, M. G. (1989). Expression of a large family of POU-domain regulatory genes in mammalian brain development [published erratum appears in Nature 1989 Aug 24;340(6235):662]. Nature 340, 35-41. Heery, D. M., Kalkhoven, E., Hoare, S. & Parker, M. G. (1997). A signature motif in transcriptional co-activators mediates binding to nuclear receptors [see comments]. Nature 387, 733-6. Heinzel, T., Lavinsky, R. M., Mullen, T. M., Soderstrom, M., Laherty, C. D., Torchia, J., Yang, W. M., Brard, G., Ngo, S. D., Davie, J. R., Seto, E., Eisenman, R. N., Rose,

285 D. W., Glass, C. K. & Rosenfeld, M. G. (1997). A complex containing N-CoR, mSin3 and histone deacetylase mediates transcriptional repression [see comments]. Nature 387, 43-8. Henderson, B. E. & Feigelson, H. S. (2000). Hormonal carcinogenesis. Carcinogenesis 21,427-33. Henttu, P. M., Kalkhoven, E. & Parker, M. G. (1997). AF-2 activity and recruitment of steroid receptor coactivator 1 to the estrogen receptor depend on a lysine residue conserved in nuclear receptors. Mol Cell Biol 17, 1832-9. Herdegen, T., Skene, P. & Bahr, M. (1997). The c-Jun transcription factor—bipotential mediator of neuronal death, survival and regeneration. Trends Neurosci 20, 227- 31. Herr, W. & Cleary, M. A. (1995). The POU domain: versatility in transcriptional regulation by a flexible two-in-one DNA-binding domain. Genes Dev 9, 1679-93. Herr, W., Sturm, R. A., Clerc, R. G., Corcoran, L. M., Baltimore, D., Sharp, P. A., Ingraham, H. A., Rosenfeld, M. G., Finney, M., Ruvkun, G. & et al. (1988). The POU domain: a large conserved region in the mammalian pit-1, oct-1, oct-2, and Caenorhabditis elegans unc-86 gene products. Genes Dev 2, 1513-6. Hinkley, C. & Perry, M. (1992). Histone H2B gene transcription during Xenopus early development requires functional cooperation between proteins bound to the CCAAT and octamer motifs. Mol Cell Biol 12,4400-11. Hirschhom, J. N., Brown, S. A., Clark, C. D. & Winston, F. (1992). Evidence that SNF2/SWI2 and SNF5 activate transcription in yeast by altering chromatin structure. Genes Dev 6,2288-98. Holt, J. T., Thompson, M. E., Szabo, C., Robinson-Benion, C., Arteaga, C. L., King, M. C. & Jensen, R. A. (1996). Growth retardation and tumour inhibition by BRCAl [see comments] [published erratum appears in Nat Genet 1998 May; 19(1): 102]. Nat Genet 12,298-302. Hong, H., Kohli, K., Garabedian, M. J. & Stallcup, M. R. (1997). GRIPI, a transcriptional coactivator for the AF-2 transactivation domain of steroid, thyroid, retinoid, and vitamin D receptors. Mol Cell Biol 17, 2735-44. Hope, 1. A. & Struhl, K. (1986). Functional dissection of a eukaryotic transcriptional activator protein, GCN4 of yeast. Cell 46, 885-94. Horlein, A. J., Naar, A. M., Heinzel, T., Torchia, J., Gloss, B., Kurokawa, R., Ryan, A., Kamei, Y., Soderstrom, M., Glass, C. K. & et al. (1995). Ligand-independent repression by the thyroid hormone receptor mediated by a nuclear receptor co­ repressor [see comments]. Nature 377, 397-404. Horwitz, K. B., Jackson, T. A., Bain, D. L., Richer, J. K., Takimoto, G. S. & Tung, L. (1996). Nuclear receptor coactivators and corepressors. Mol Endocrinol 10, 1167- 77. Hoshino, S., Miyazawa, H., Enomoto, T., Hanaoka, F., Kikuchi, Y., Kikuchi, A. & Ui, M. (1989). A human homologue of the yeast GSTl gene codes for a GTP-binding protein and is expressed in a proliferation-dependent manner in mammalian cells. Em6oJ8, 3807-14. Hsu, W. & Chen-Kiang, S. (1993). Convergent regulation of NF-1L6 and Oct-1 synthesis by interleukin-6 and retinoic acid signaling in embryonal carcinoma cells. Mol Cell Biol 13,2515-23.

286 Hu, C. H., McStay, B., Jeong, S. W. & Reeder, R. H. (1994). xUBF, an RNA polymerase I transcription factor, binds crossover DNA with low sequence specificity. Mol Cell Biol 14,2871-82. Hu, Z., Yuri, K., Ozawa, H., Lu, H., Yang, Y., Ito, T. & Kawata, M. (1997a). Adrenalectomy-induced granule cell death is predicated on the disappearance of glucocorticoid receptor immunoreactivity in the rat hippocampal granule cell layer. Brain Res 11^, 293-301. Hu, Z., Yuri, K., Morimoto, M., Ozawa, H. & Kawata, M. (1997b). Chronic loss of glucocorticoids following adrenalectomy down-regulates the expression of glucocorticoid receptor mRNA in the rat forebrain. Eur J Neurosci 9, 1778-83. Huang, C. C. & Herr, W. (1996). Differential control of transcription by homologous homeodomain coregulators. Mol Cell Biol 16,2967-76. Ikeda, K., Steger, D. J., Eberharter, A. & Workman, J. L. (1999). Activation domain- specific and general transcription stimulation by native histone acetyltransferase complexes. Mol Cell Biol 19, 855-63. Imbalzano, A. N., Kwon, H., Green, M. R. & Kingston, R. E. (1994). Facilitated binding of TATA-binding protein to nucleosomal DNA [see comments]. Nature 370, 481- 5. Ing, N. H., Beekman, J. M., Tsai, S. Y., Tsai, M. J. & O'Malley, B. W. (1992). Members of the steroid hormone receptor superfamily interact with TFIIB (S300-II). J Biol Chem 267, 17617-23. Ingham, P. W. (1988). The molecular genetics of embryonic pattern formation in Drosophila [published erratum appears in Nature 1988 Oct 20;335(6192):744]. Nature 335, 25-34. Ingraham, H. A., Chen, R. P., Mangalam, H. J., Elsholtz, H. P., Flynn, S. E., Lin, C. R., Simmons, D. M., Swanson, L. & Rosenfeld, M. G. (1988). A tissue-specific transcription factor containing a homeodomain specifies a pituitary phenotype. Ce//55, 519-29. Ingraham, H. A., Flynn, S. E., Voss, J. W., Albert, V. R., Kapiloff, M. S., Wilson, L. & Rosenfeld, M. G. (1990). The POU-specific domain of Pit-1 is essential for sequence-specific, high affinity DNA binding and DNA-dependent Pit-1-Pit-1 interactions. Cell 61, 1021-33. Ivarie, R. D. & O'Farrell, P. H. (1978). The glucocorticoid domain: steroid-mediated changes in the rate of synthesis of rat hepatoma proteins. Cell 13,41-55. Iwaya, K., Tsuda, H., Fujita, S., Suzuki, M. & Hirohashi, S. (1995). Natural state of mutant p53 protein and heat shock protein 70 in breast cancer tissues. Lab Invest 72, 707-14. Iyer, V. R., Eisen, M. B., Ross, D. T., Schuler, G., Moore, T., Lee, J. C. F., Trent, J. M., Staudt, L. M., Hudson, J., Jr., Boguski, M. S., Lashkari, D., Shalon, D., Botstein, D. & Brown, P. O. (1999). The transcriptional program in the response of human fibroblasts to serum [see comments]. Science 283, 83-7. Jackson, T. A., Richer, J. K., Bain, D. L., Takimoto, G. S., Tung, L. & Horwitz, K. B. (1997). The partial agonist activity of antagonist-occupied steroid receptors is controlled by a novel hinge domain-binding coactivator L7/SPA and the corepressors N-CoR or SMRT. Mol Endocrinol 11, 693-705.

287 Jacobson, E. M., Li, P., Leon del Rio, A., Rosenfeld, M. G. & Aggarwal, A. K. (1997). Structure of Pit-1 POU domain bound to DNA as a dimer: unexpected arrangement and flexibility. Genes Dev 11,198-212. Jacobson, E. M., Li, P., Rosenfeld, M. G. & Aggarwal, A. K. (1996). Crystallization and preliminary X-ray analysis of Pit-1 POU domain complexed to a 28 base pair DNA element. Proteins 24,263-5. Jensen, E. V., Brecher, P. I., Numata, M., Mohla, S. & De Sombre, E. R. (1973a). Transformed estrogen receptor in the regulation of RNA synthesis in uterine nuclei. Adv Enzyme Regul 11, 1-16. Jensen, E. V. & DeSombre, E. R. (1973). Estrogen-receptor interaction. Science 182, 126-34. Jensen, E. V., Jacobson, H. I., Smith, S., Jungblut, P. W. & De Sombre, E. R. (1972). The use of estrogen antagonists in hormone receptor studies. Gynecol Invest 3, 108-23. Jensen, E. V., Mohla, S., Brecher, P. I. & DeSombre, E. R. (1973b). Estrogen receptor transformation and nuclear RNA synthesis. Adv Exp Med Biol 36, 60-79. Jiang, R., Lan, Y., Chapman, H. D., Shawber, C., Norton, C. R., Serreze, D. V., Weinmaster, G. & Gridley, T. (1998). Defects in limb, craniofacial, and thymic development in Jagged2 mutant mice. Genes Dev 12, 1046-57. Jin, T., Branch, D. R., Zhang, X., Qi, S., Youngson, B. & Goss, P. E. (1999). Examination of POU homeobox gene expression in human breast cancer cells. Int J Cancer 104-12. Joel, P. B., Traish, A. M. & Lannigan, D. A. (1998). Estradiol-induced phosphorylation of serine 118 in the estrogen receptor is independent of p42/p44 mitogen- activated protein kinase. Chem 273, 13317-23. Johansson, L., Thomsen, J. S., Damdimopoulos, A. E., Spyrou, G., Gustafsson, J. A. & Treuter, E. (1999). The orphan nuclear receptor SHP inhibits agonist-dependent transcriptional activity of estrogen receptors ERalpha and ERbeta. J Biol Chem 274, 345-53. Johnston, M. (1987). Genetic evidence that zinc is an essential co-factor in the DNA binding domain of GAL4 protein. Nature 328,353-5. Jonat, C., Rahmsdorf, H. J., Park, K. K., Cato, A. C., Gebel, S., Ponta, H. & Herrlich, P. (1990). Antitumor promotion and antiinflammation: down-modulation of AP-1 (Fos/Jun) activity by glucocorticoid hormone. Cell 62, 1189-204. Junker, S., Pedersen, S., Schreiber, E. & Matthias, P. (1990). Extinction of an immunoglobulin kappa promoter in cell hybrids is mediated by the octamer motif and correlates with suppression of Oct-2 expression. Cell 61,467-74. Kalkhoven, E., Kwakkenbos-Isbrucker, L., de Laat, S. W., van der Saag, P. T. & van der Burg, B. (1994). Synthetic progestins induce proliferation of breast tumor cell lines via the progesterone or estrogen receptor. Mol Cell Endocrinol 102,45-52. Kalkhoven, E., Valentine, J. E., Heery, D. M. & Parker, M. G. (1998). Isoforms of steroid receptor co-activator 1 differ in their ability to potentiate transcription by the oestrogen receptor. Embo J 11, 232-43. Kamei, Y., Xu, L., Heinzel, T., Torchia, J., Kurokawa, R., Gloss, B., Lin, S. C., Heyman, R. A., Rose, D. W., Glass, C. K. & Rosenfeld, M. G. (1996). A CBP integrator complex mediates transcriptional activation and AP-1 inhibition by nuclear receptors. Cell 85,403-14.

288 Kapiloff, M. S., Farkash, Y., Wegner, M. & Rosenfeld, M. G. (1991). Variable effects of phosphorylation of Pit-1 dictated by the DNA response elements. Science 253, 786-9. Katan, M., Haigh, A., Verrijzer, C. P., van der Vliet, P. C. & O'Hare, P. (1990). Characterization of a cellular factor which interacts functionally with Oct-1 in the assembly of a multicomponent transcription complex. Nucleic Acids Res 18, 6871-80. Kato, S., Endoh, H., Masuhiro, Y., Kitamoto, T., Uchiyama, S., Sasaki, H., Masushige, S., Gotoh, Y., Nishida, E., Kawashima, H. & et al. (1995). Activation of the estrogen receptor through phosphorylation by mitogen-activated protein kinase. Science 21^^ 1491-4. Katzenellenbogen, B. S. (1996). Estrogen receptors: bioactivities and interactions with cell signaling pathways. Biol Reprod 54,287-93. Katzenellenbogen, J. A. & Katzenellenbogen, B. S. (1996). Nuclear hormone receptors: ligand-activated regulators of transcription and diverse cell responses. Chem Biol 3, 529-36. Kawata, M., Yuri, K., Ozawa, H., Nishi, M., Ito, T., Hu, Z., Lu, H. & Yoshida, M. (1998). Steroid hormones and their receptors in the brain. J Steroid Biochem Mol Biol 65, 273-80. Kee, B. L., Arias, J. & Montminy, M. R. (1996). Adaptor-mediated recruitment of RNA polymerase II to a signal-dependent activator. J Biol Chem 271,2373-5. Kemler, I., Schreiber, E., Muller, M. M., Matthias, P. & Schaffiier, W. (1989). Octamer transcription factors bind to two different sequence motifs of the immunoglobulin heavy chain promoter. Embo 2001-8. Kim, H. J., Kim, J. H. & Lee, J. W. (1998a). Steroid receptor coactivator-1 interacts with serum response factor and coactivates serum response element-mediated transactivations. JBiol Chem 273, 28564-7. Kim, H. J., Lee, S. K., Na, S. Y., Choi, H. S. & Lee, J. W. (1998b). Molecular cloning of xSRC-3, a novel transcription coactivator from Xenopus, that is related to AIBl, p/CIP, and TIF2. Mol Endocrinol 12, 1038-47. Kimura, Y., Yahara, I. & Lindquist, S. (1995). Role of the protein chaperone YDJl in establishing Hsp90-mediated signal transduction pathways [see comments]. Science 26S, 1362-5. Kissinger, C. R., Liu, B. S., Martin-Bianco, E., Komberg, T. B. & Pabo, C. O. (1990). Crystal structure of an homeodomain-DNA complex at 2.8 A resolution: a framework for understanding homeodomain-DNA interactions. Cell 63, 579-90. Klein, G. & Klein, E. (1985). Evolution of tumours and the impact of molecular oncology. Nature 315, 190-5. Klein-Hitpass, L., Schorpp, M., Wagner, U. & Ryffel, G. U. (1986). An estrogen- responsive element derived from the 5' flanking region of the Xenopus vitellogenin A2 gene functions in transfected human cells. Cell 46, 1053-61. Klemm, J. D., Rould, M. A., Aurora, R., Herr, W. & Pabo, C. O. (1994). Crystal structure of the Oct-1 POU domain bound to an octamer site: DNA recognition with tethered DNA-binding modules. Cell 77,21-32.

289 Klinge, C. M., Bodenner, D. L., Desai, D., Niles, R. M. & Traish, A. M. (1997). Binding of type II nuclear receptors and estrogen receptor to full and half-site estrogen response elements in vitro. Nucleic Acids Res 25, 1903-12. Klinge, C. M., Peale, F. V., Jr., Hilf, R., Bambara, R. A. & Zain, S. (1992). Cooperative estrogen receptor interaction with consensus or variant estrogen responsive elements in vitro. Cancer Res 52, 1073-81. Kobayashi, Y., Kitamoto, T., Masuhiro, Y., Watanabe, M., Kase, T., Metzger, D., Yanagisawa, J. & Kato, S. (2000). p300 mediates functional synergism between AF-1 and AF-2 of estrogen receptor alpha and beta by interacting directly with the N-terminal A/B domains. J Biol Chem 275, 15645-51. Korzus, E., Torchia, J., Rose, D. W., Xu, L., Kurokawa, R., Mclnemey, E. M., Mullen, T. M., Glass, C. K. & Rosenfeld, M. G. (1998). Transcription factor-specific requirements for coactivators and their acetyltransferase functions. Science 279, 703-7. Kouzarides, T. & Ziff, E. (1988). The role of the leucine zipper in the fos-jun interaction. Nature 336, 646-51. Kraus, W. L., Mclnemey, E. M. & Katzenellenbogen, B. S. (1995). Ligand-dependent, transcriptionally productive association of the amino- and carboxyl-terminal regions of a steroid hormone nuclear receptor. Proc Natl Acad Sci U S A92^ 12314-8. Kraus, W. L., Montano, M. M. & Katzenellenbogen, B. S. (1994). Identification of multiple, widely spaced estrogen-responsive regions in the rat progesterone receptor gene. Mol Endocrinol 8,952-69. Krebs, J. E. & Peterson, C. L. (2000). Understanding "active" chromatin: a historical perspective of chromatin remodeling. Crit Rev Eukaryot Gene Expr 10, 1- 12. Kumar, V., Green, S., Stack, G., Berry, M., Jin, J. R. & Chambon, P. (1987). Functional domains of the human estrogen receptor. Cell 51, 941-51. Kurokawa, R., Soderstrom, M., Horlein, A., Halachmi, S., Brown, M., Rosenfeld, M. G. & Glass, C. K. (1995). Polarity-specific activities of retinoic acid receptors determined by a co-repressor [see comments]. Nature 377,451-4. Kwok, R. P., Lundblad, J. R., Chrivia, J. C., Richards, J. P., Bachinger, H. P., Brennan, R. G., Roberts, S. G., Green, M. R. & Goodman, R. H. (1994). Nuclear protein CBP is a coactivator for the transcription factor CREB [see comments]. Nature 370, 223-6. Kwon, H., Imbalzano, A. N., Khavari, P. A., Kingston, R. E. & Green, M. R. (1994). Nucleosome dismption and enhancement of activator binding by a human SWI/SNF complex [see comments]. Nature 370,477-81. Lai, J. S., Cleary, M. A. & Herr, W. (1992). A single amino acid exchange transfers VP16-induced positive control from the Oct-1 to the Oct-2 homeo domain [published erratum appears in Genes Dev 1992 Dec;6(12B):2663]. Genes Dev 6, 2058-65. Lakin, N. D., Morris, P. J., Theil, T., Sato, T. N., Moroy, T., Wilson, M. C. & Latchman, D. S. (1995). Regulation of neurite outgrowth and SNAP-25 gene expression by the Bm-3a transcription factor. J Biol Chem 270, 15858-63.

290 Landel, C. C., Potthoff, S. J., Nardulli, A. M., Kushner, P. J. & Greene, G. L. (1997). Estrogen receptor accessory proteins augment receptor-DNA interaction and DNA bending. J Steroid Biochem Mol Biol 63, 59-73. Landschulz, W. H., Johnson, P. F. & McKnight, S. L. (1989). The DNA binding domain of the rat liver nuclear protein C/EBP is bipartite. Science 243, 1681-8. Lane, T. P., Deng, C., Elson, A., Lyu, M. S., Kozak, C. A. & Leder, P. (1995). Expression of Brcal is associated with terminal differentiation of ectodermally and mesodermally derived tissues in mice [published erratum appears in Genes Dev 1996 Feb 1;10(3):365]. Genes Dev 9,2712-22. Lanford, P. J., Lan, Y., Jiang, R., Lindsell, C., Weinmaster, G., Gridley, T. & Kelley, M. W. (1999). Notch signalling pathway mediates hair cell development in mammalian cochlea [see comments]. Nat Genet 21,289-92. Latchman, D. S. (1996). Activation and repression of gene expression by POU family transcription factors. Philos Trans R Soc Lond B Biol Sci 351, 511-5. Latchman, D. S. (1998a). The Bm-3 a transcription factor. Int J Biochem Cell Biol 30, 1153-7. Latchman, D. S. (1998b). Eukaryotic Transcription Factors, 3 edition. Academic Press, London. Latchman, D. S. (1999). POU family transcription factors in the nervous system. J Cell Physiol 179, 126-33. Latchman, D. S., Dent, C. L., Lillycrop, K. A. & Wood, J. N. (1992). POU family transcription factors in sensory neurons. Biochem Soc Trans 20, 627-31. Laurent, B. C., Treich, I. & Carlson, M. (1993). The yeast SNF2/SWI2 protein has DNA- stimulated ATPase activity required for transcriptional activation. Genes Dev 7, 583-91. Lazennec, G., Ediger, T. R., Petz, L. N., Nardulli, A. M. & Katzenellenbogen, B. S. (1997). Mechanistic aspects of estrogen receptor activation probed with constitutively active estrogen receptors: correlations with DNA and coregulator interactions and receptor conformational changes. Mol Endocrinol 11, 1375-86. Leblond-Francillard, M., Picon, A., Bertagna, X. & de Keyzer, Y. (1997). High expression of the POU factor Bm3a in aggressive neuroendocrine tumors. J Clin Endocrinol Metab 82, 89-94. LeBowitz, J. H., Clerc, R. G., Brenowitz, M. & Sharp, P. A. (1989). The Oct-2 protein binds cooperatively to adjacent octamer sites. Genes Dev 3, 1625-38. Lee, S. K., Choi, H. S., Song, M. R., Lee, M. O. & Lee, J. W. (1998a). Estrogen receptor, a common interaction partner for a subset of nuclear receptors. Mol Endocrinol 12, 1184-92. Lee, S. K., Kim, H. J., Kim, J. W. & Lee, J. W. (1999). Steroid receptor coactivator-1 and its family members differentially regulate transactivation by the tumor suppressor protein p53. Mol Endocrinol 13,1924-33. Lee, S. K., Kim, H. J., Na, S. Y., Kim, T. S., Choi, H. S., Im, S. Y. & Lee, J. W. (1998b). Steroid receptor coactivator-1 coactivates activating protein-1-mediated transactivations through interaction with the c-Jun and c-Fos subunits. J Biol Chem 273, 16651-4. Lees, J. A., Fawell, S. E. & Parker, M. G. (1989). Identification of two transactivation domains in the mouse oestrogen receptor. Nucleic Acids Res 17, 5477-88.

291 Li, H. & Chen, J. D. (1998). The receptor-associated coactivator 3 activates transcription through CREB-binding protein recruitment and autoregulation. J Biol Chem 273, 5948-54. Li, H., Gomes, P. J. & Chen, J. D. (1997). RAC3, a steroid/nuclear receptor-associated coactivator that is related to SRC-1 and TIF2. Proc Natl Acad Sci U SA 94, 8479- 84. Li, P., He, X., Gerrero, M. R., Mok, M., Aggarwal, A. & Rosenfeld, M. G. (1993). Spacing and orientation of bipartite DNA-binding motifs as potential functional determinants for POU domain factors. Genes Dev 7,2483-96. Liang, H., Mao, X., Olejniczak, E. T., Nettesheim, D. G., Yu, L., Meadows, R. P., Thompson, C. B. & Fesik, S. W. (1994a). Solution structure of the ets domain of Fli-1 when bound to DNA. Nat Struct Biol 1, 871-5. Liang, H., Olejniczak, E. T., Mao, X., Nettesheim, D. G., Yu, L., Thompson, C. B. & Fesik, S. W. (1994b). The secondary structure of the ets domain of human Fli-1 resembles that of the helix-tum-helix DN A-binding motif of the Escherichia coli catabolite gene activator protein. Proc Natl Acad Sci U SA 9\,\ 1655-9. Lillycrop, K. A., Budrahan, V. S., Lakin, N. D,, Terrenghi, G., Wood, J. N., Polak, J. M. & Latchman, D. S. (1992). A novel POU family transcription factor is closely related to Bm-3 but has a distinct expression pattern in neuronal cells. Nucleic Acids Res 20, 5093-6. Lippman, M. E., Bates, S., Huff, K. K., Davidson, N. & Dickson, R. B. (1987). Estrogens regulate production of specific growth factors in hormone-dependent human breast cancer. JLab Clin Med 109,230-5. Lippman, M. E. & Dickson, R. B. (1989). Mechanisms of normal and malignant breast epithelial growth regulation. J Steroid Biochem 34, 107-21. Liu, C., Adamson, E. & Mercola, D. (1996a). Transcription factor EGR-1 suppresses the growth and transformation of human HT-1080 fibrosarcoma cells by induction of transforming growth factor beta 1. Proc Natl Acad Sci f/ *5 ^ 93, 11831 -6. Liu, C., Calogero, A., Ragona, G., Adamson, E. & Mercola, D. (1996b). EGR-1, the reluctant suppression factor: EGR-1 is known to function in the regulation of growth, differentiation, and also has significant tumor suppressor activity and a mechanism involving the induction of TGF-betal is postulated to account for this suppressor activity. Crit Rev Oncog 7, 101-25. Liu, C., Rangnekar, V. M., Adamson, E. & Mercola, D. (1998). Suppression of growth and transformation and induction of apoptosis by EGR-1. Cancer Gene Ther 5, 3- 28. Liu, Y. Z., Dawson, S. J. & Latchman, D. S. (1996c). Alternative splicing of the Bm-3 a and Bm-3b transcription factor RNAs is regulated in neuronal cells. J Mol Neurosci 7, 77-85. Lopez, G. N., Webb, P., Shinsako, J. H., Baxter, J. D., Greene, G. L. & Kushner, P. J. (1999). Titration by estrogen receptor activation fiinction-2 of targets that are downstream fi-om coactivators. Mol Endocrinol 13, 897-909. Lubahn, D. B., Moyer, J. S., Golding, T. S., Couse, J. F., Korach, K. S. & Smithies, O. (1993). Alteration of reproductive function but not prenatal sexual development after insertional dismption of the mouse estrogen receptor gene. Proc Natl Acad SciUSA9{i, 11162-6.

292 Limdholt, B. K., Madsen, M. W., Lykkesfeldt, A. E., Petersen, O. W. & Briand, P. (1996). Characterization of a nontumorigenic human breast epithelial cell line stably transfected with the human estrogen receptor (ER) cDNA. Mol Cell Endocrinol 119,47-59. Luo, Y., Fujii, H., Gerster, T. & Roeder, R. G. (1992). A novel B cell-derived coactivator potentiates the activation of immunoglobulin promoters by octamer-binding transcription factors. Cell 71,231-41. Macpherson, I. & Stoker, M. (1962). Virology 16,147-151. Maehara, Y., Oki, E., Abe, T., Tokunaga, E., Shibahara, K., Kakeji, Y. & Sugimachi, K. (2000). Overexpression of the heat shock protein HSP70 family and p53 protein and prognosis for patients with gastric cancer. Oncology 58, 144-51. Malicki, J., Cianetti, L. C., Peschle, C. & McGinnis, W. (1992). A human H0X4B regulatory element provides head-specific expression in Drosophila embryos. Nature 358, 345-7. Mangelsdorf, D. J., Thummel, C., Beato, M., Herrlich, P., Schütz, G., Umesono, K., Blumberg, B., Kastner, P., Mark, M., Chambon, P. & et al. (1995). The nuclear receptor superfamily: the second decade. Cell S3, 835-9. Marquis, S. T., Raj an, J. V., Wynshaw-Boris, A., Xu, J., Yin, G. Y., Abel, K. J., Weber, B. L. & Chodosh, L. A. (1995). The developmental pattern of Brcal expression implies a role in differentiation of the breast and other tissues. Nat Genet 11, 17- 26. Mattaj, I. W., Lienhard, S., Jiricny, J. & De Robertis, E. M. (1985). An enhancer-like sequence within the Xenopus U2 gene promoter facilitates the formation of stable transcription complexes. Nature 316, 163-7. McEvilly, R. J., Erkman, L., Luo, L., Sawchenko, P. E., Ryan, A. F. & Rosenfeld, M. G. (1996). Requirement for Bm-3.0 in differentiation and survival of sensory and motor neurons. Nature 384, 574-7. McEvilly, R. J. & Rosenfeld, M. G. (1999). The role of POU domain proteins in the regulation of mammalian pituitary and nervous system development. Prog Nucleic Acid Res Mol Biol 63,223-55. Mclnemey, E. M. & Katzenellenbogen, B. S. (1996). Different regions in activation function-1 of the human estrogen receptor required for antiestrogen- and estradiol-dependent transcription activation. J Biol Chem 271, 24172-8. McKenna, N. J., Lanz, R. B. & O'Malley, B. W. (1999a). Nuclear receptor coregulators: cellular and molecular biology. Endocr Rev 20, 321-44. McKenna, N. J., Xu, J., Nawaz, Z., Tsai, S. Y., Tsai, M. J. & O'Malley, B. W. (1999b). Nuclear receptor coactivators: multiple enzymes, multiple complexes, multiple functions. J Steroid Biochem Mol Biol 69, 3-12. Melamed, I., Tumer, C. E., Aktories, K., Kaplan, D. R. & Gelfand, E. W. (1995). Nerve growth factor triggers microfilament assembly and paxillin phosphorylation in human B lymphocytes. J Exp Med 181,1071-9. Meyer, M. E., Gronemeyer, H., Turcotte, B., Bocquel, M. T., Tasset, D. & Chambon, P. (1989). Steroid hormone receptors compete for factors that mediate their enhancer function. Cell 57,433-42.

293 Miki, Y., Swensen, J., Shattuck-Eidens, D., Futreal, P. A., Harshman, K., Tavtigian, S., Liu, Q., Cochran, C., Bennett, L. M., Ding, W. & et al. (1994). A strong candidate for the breast and ovarian cancer susceptibility gene BRCAl. Science 266, 66-71. Miller, J., McLachlan, A. D. & Klug, A. (1985). Repetitive zinc-binding domains in the protein transcription factor IIIA from Xenopus oocytes. Embo J4 , 1609-14. Milton, N. G., Bessis, A., Changeux, J. P. & Latchman, D. S. (1995). The neuronal nicotinic acetylcholine receptor alpha 2 subunit gene promoter is activated by the Bm-3b POU family transcription factor and not by Bm-3 a or Bm-3c. J Biol Chem 270, 15143-7. Milton, N. G., Bessis, A., Changeux, J. P. & Latchman, D. S. (1996). Differential regulation of neuronal nicotinic acetylcholine receptor subunit gene promoters by Bm-3 POU family transcription factors. Biochem J 317,419-23. Min, H., Turck, C. W., Nikolic, J. M. & Black, D. L. (1997). A new regulatory protein, KSRP, mediates exon inclusion through an intronic splicing enhancer. Genes Dev 11, 1023-36. Moras, D. & Gronemeyer, H. (1998). The nuclear receptor ligand-binding domain: stmcture and function. Curr Opin Cell Biol 10, 384-91. Morris, P. J., Dawson, S. J., Wilson, M. C. & Latchman, D. S. (1997). A single residue within the homeodomain of the Bm-3 POU family transcription factors determines whether they activate or repress the SNAP-25 promoter. Neuroreport 8, 2041-5. Morris, P. J., Lakin, N. D., Dawson, S. J., Ryabinin, A. E., Kilimann, M. W., Wilson, M. C. & Latchman, D. S. (1996). Differential regulation of genes encoding synaptic proteins by members of the Bm-3 subfamily of POU transcription factors. Brain Res Mol Brain Res 43, 279-85. Morris, P. J., Theil, T., Ring, C. J., Lillycrop, K. A., Moroy, T. & Latchman, D. S. (1994). The opposite and antagonistic effects of the closely related POU family transcription factors Bm-3 a and Bm-3b on the activity of a target promoter are dependent on differences in the POU domain. Mol Cell Biol 14, 6907-14. Mosselman, S., Polman, J. & Dijkema, R. (1996). ER beta: identification and characterization of a novel human estrogen receptor. FEBS Lett 392,49-53. Muchardt, C. & Yaniv, M. (1993). A human homologue of Saccharomyces cerevisiae SNF2/SWI2 and Drosophila brm genes potentiates transcriptional activation by the glucocorticoid receptor. Embo J 12,4279-90. Muchardt, C. & Yaniv, M. (1999). ATP-dependent chromatin remodelling: SWI/SNF and Co. are on the job. J Mol Biol 293, 187-98. Mukheijee, R. & Chambon, P. (1990). A single-stranded DNA-binding protein promotes the binding of the purified oestrogen receptor to its responsive element. Nucleic Acids Res 18, 5713-6. Muller, C. W., Rey, F. A., Sodeoka, M., Verdine, G. L. & Harrison, S. C. (1995). Stmcture of the NF-kappa B p50 homodimer bound to DNA [see comments]. Nature 373, 311-7. Muller-Hill, B. (1966). Suppressible regulator constitutive mutants of the lactose system in Escherichia coli. J Mol Biol 15, 374-6. Muramatsu, M. & Inoue, S. (2000). Estrogen receptors: how do they control reproductive and nonreproductive functions? Biochem Biophys Res Commun 270, 1-10.

294 Murphy, L. C, & Sutherland, R. L. (1983). Antitumor activity of clomiphene analogs in vitro: relationship to affinity for the estrogen receptor and another high affinity antiestrogen-binding site. J Clin Endocrinol Metab 57, 373-9. Musti, A. M., Treier, M., Peverali, F. A. & Bohmann, D. (1996). Differential regulation of c-Jun and JunD by ubiquitin-dependent protein degradation. Biol Chem 377, 619-24. Na, S. Y., Lee, S. K., Han, S. J., Choi, H. S., Im, S. Y. & Lee, J. W. (1998). Steroid receptor coactivator-1 interacts with the p50 subunit and coactivates nuclear factor kappaB-mediated transactivations. J Biol Chem 273, 10831-4. Nagy, L., Kao, H. Y., Chakravarti, D., Lin, R. J., Hassig, C. A., Ayer, D. E., Schreiber, S. L. & Evans, R. M. (1997). Nuclear receptor repression mediated by a complex containing SMRT, mSin3A, and histone deacetylase. Cell 89, 373-80. Nass, S. J. & Dickson, R. B. (1997). Defining a role for c-Myc in breast tumorigenesis. Breast Cancer Res Treat 44, 1-22. Ndisang, D., Budhram Mahadeo, V. & Latchman, D. S. (1999). The Bm-3a transcription factor plays a critical role in regulating human papilloma virus gene expression and determining the growth characteristics of cervical cancer cells. J Biol Chem 274,28521-7. Needham, M., Raines, S., McPheat, J., Stacey, C., Ellston, J., Hoare, S. & Parker, M. (2000). Differential interaction of steroid hormone receptors with LXXLL motifs in SRC-la depends on residues flanking the motif. J Steroid Biochem Mol Biol 72, 35-46. Neigebom, L. & Carlson, M. (1984). Genes affecting the regulation of SUC2 gene expression by glucose repression in Saccharomyces cerevisiae. Genetics 108, 845-58. Nelson, C., Albert, V. R., Elsholtz, H. P., Lu, L. I. & Rosenfeld, M. G. (1988). Activation of cell-specific expression of rat growth hormone and prolactin genes by a common transcription factor. Science 239,1400-5. Nolte, R. T., Wisely, G. B., Westin, S., Cobb, J. E., Lambert, M. H., Kurokawa, R., Rosenfeld, M. G., Willson, T. M., Glass, C. K. & Milbum, M. V. (1998). Ligand binding and co-activator assembly of the peroxisome proliferator-activated receptor-gamma. Nature 395, 137-43. Oas, T. G., McIntosh, L. P., O'Shea, E. K., Dahlquist, F. W. & Kim, P. S. (1990). Secondary structure of a leucine zipper determined by nuclear magnetic resonance spectroscopy. Biochemistry 29,2891-4. Onate, S. A., Boonyaratanakomkit, V., Spencer, T. E., Tsai, S. Y., Tsai, M. J., Edwards, D. P. & O'Malley, B. W. (1998). The steroid receptor coactivator-1 contains multiple receptor interacting and activation domains that cooperatively enhance the activation function 1 (AFl) and AF2 domains of steroid receptors. J Biol Chem 273, 12101-8. Onate, S. A., Tsai, S. Y., Tsai, M. J. & O'Malley, B. W. (1995). Sequence and characterization of a coactivator for the steroid hormone receptor superfamily. Science 21^^ 1354-7. O'Shea, E. K., Rutkowski, R. & Kim, P. S. (1989). Evidence that the leucine zipper is a coiled coil. Science 243, 538-42.

295 Osterlund, M. K., Grandien, K., Keller, E. & Hurd, Y. L. (2000). The human brain has distinct regional expression patterns of estrogen receptor alpha mRNA isoforms derived from alternative promoters. JNeurochem 75, 1390-7. Otting, G., Qian, Y. Q., Billeter, M., Muller, M., Affolter, M., Gehring, W. J. & Wuthrich, K. (1990). Protein—DNA contacts in the structure of a homeodomain- DNA complex determined by nuclear magnetic resonance spectroscopy in solution. Embo J9 , 3085-92. Pabo, C. O., Aggarwal, A. K., Jordan, S. R., Beamer, L. J., Obeysekare, U. R. & Harrison, S. C. (1990). Conserved residues make similar contacts in two repressor-operator complexes. Science 247, 1210-3. Packer, A. I., Grotty, D. A., Elwell, V. A. & Wolgemuth, D. J. (1998). Expression of the murine Hoxa4 gene requires both autoregulation and a conserved retinoic acid response element. Development 125, 1991-8. Paech, K., Webb, P., Kuiper, G. G., Nilsson, S., Gustafsson, J., Kushner, P. J. & Scanlan, T. S. (1997). Differential ligand activation of estrogen receptors ERalpha and ERbeta at API sites [see comments]. Science 277, 1508-10. Park, C. S., Joo, I. S., Song, S. Y., Kim, D. S., Bae, D. S. & Lee, J. H. (1999). An immunohistochemical analysis of heat shock protein 70, p53, and estrogen receptor status in carcinoma of the uterine cervix. Gynecol Oncol 74, 53-60. Parker, M. G. & White, R. (1996). Nuclear receptors spring into action [news; comment]. Nat Struct Biol 3, 113-5. Patrone, C., Andersson, S., Korhonen, L. & Lindholm, D. (1999). Estrogen receptor- dependent regulation of sensory neuron survival in developing dorsal root ganglion. Proc Natl Acad Sci USA96, 10905-10. Patrone, C., Pollio, G., Vegeto, E., Enmark, E., de Curtis, I., Gustafsson, J. A. & Maggi, A. (2000). Estradiol induces differential neuronal phenotypes by activating estrogen receptor alpha or beta. Endocrinology 141, 1839-45. Pettersson, K., Grandien, K., Kuiper, G. G. & Gustafsson, J. A. (1997). Mouse estrogen receptor beta forms estrogen response element-binding heterodimers with estrogen receptor alpha. Mol Endocrinol 11, 1486-96. Philips, A., Chalbos, D. & Rochefort, H. (1993). Estradiol increases and anti-estrogens antagonize the growth factor-induced activator protein-1 activity in MCF7 breast cancer cells without affecting c-fos and c-jun synthesis [published erratum appears in J Biol Chem 1993 Dec 5;268(34):26032]. J Biol Chem 268, 14103-8. Pierani, A., Heguy, A., Fujii, H. & Roeder, R. G. (1990). Activation of octamer- containing promoters by either octamer-binding transcription factor 1 (OTF-1) or OTF-2 and requirement of an additional B-cell-specific component for optimal transcription of immunoglobulin promoters. Mol Cell Biol 10, 6204-15. Pierrat, B., Heery, D. M., Lemoine, Y. & Losson, R. (1992). Functional analysis of the human estrogen receptor using a phenotypic transactivation assay in yeast. Gene 119, 237-45. Planas-Silva, M. D. & Weinberg, R. A. (1997). Estrogen-dependent cyclin E-cdk2 activation through p21 redistribution. Mol Cell Biol 17,4059-69. Poellinger, L., Yoza, B. K. & Roeder, R. G. (1989). Functional cooperativity between protein molecules bound at two distinct sequence elements of the immunoglobulin heavy-chain promoter. Nature 337, 573-576.

296 Popperl, H. & Featherstone, M. S. (1992). An autoregulatory element of the murine Hox- 4.2 gene. Embo J 11, 3673-80. Prall, O. W. J., Sarcevic, B., Musgrove, E. A., Watts, C. K. W. & Sutherland, R. L. (1997). Estrogen-induced activation of Cdk4 and Cdk2 during Gl-S phase progression is accompanied by increased cyclin D1 expression and decreased cyclin-dependent kinase inhibitor association with cyclin E-Cdk2. J Biol Chem 272, 10882-94. Pratt, W. B. (1993). The role of heat shock proteins in regulating the function, folding, and trafficking of the glucocorticoid receptor. JBiol Chem 268,21455-8. Pratt, W. B. & Toft, D. O. (1997). Steroid receptor interactions with heat shock protein and immunophilin chaperones. Endocr Rev 18, 306-60. Ptashne, M. (1967). Specific binding of the lambda phage repressor to lambda DNA. Nature 214,232-4. Ptashne, M. (1988). How eukaryotic transcriptional activators work. Nature 335, 683-9. Qian, Y. Q., Billeter, M., Otting, G., Muller, M., Gehring, W. J. & Wuthrich, K. (1989). The structure of the Antennapedia homeodomain determined by NMR spectroscopy in solution: comparison with prokaryotic repressors [published erratum appears in Cell 1990 May 4;61(3):548]. Cell 5% 573-80. Reddel, R. R., Murphy, L. C. & Sutherland, R. L. (1983). Effects of biologically active metabolites of tamoxifen on the proliferation kinetics of MCF-7 human breast cancer cells in vitro. Cancer Res 43,4618-24. Reddel, R. R. & Sutherland, R. L. (1983). N-desmethyltamoxifen inhibits growth of MCF 7 human mammary carcinoma cells in vitro. Eur J Cancer Clin Oncol 19, 1179- 81. Reimer, T., Koczan, D., Muller, H., Friese, K., Krause, A., Thiesen, H. J. & Gerber, B. (2000). Human chorionic gonadotrophin-beta transcripts correlate with progesterone receptor values in breast carcinomas. J Mol Endocrinol 24, 33-41. Reines, D., Conaway, J. W. & Conaway, R. C. (1996). The RNA polymerase II general elongation factors. Trends Biochem Sci 21, 351-5. Renaud, J. P., Rochel, N., Ruff, M., Vivat, V., Chambon, P., Gronemeyer, H. & Moras, D. (1995). Crystal structure of the RAR-gamma ligand-binding domain bound to all-trans retinoic acid. Nature 378, 681-9. Richard, S. & Zingg, H. H. (1990). The human oxytocin gene promoter is regulated by estrogens. JBiol Chem 265, 6098-103. Ricketts, D., Turnbull, L., Ryall, G., Bakhshi, R., Rawson, N. S., Gazet, J. C., Nolan, C. & Coombes, R. C. (1991). Estrogen and progesterone receptors in the normal female breast. Cancer Res 51, 1817-22. Ridley, A. J. (1997). The GTP-binding protein Rho. Int J Biochem Cell Biol 29^ 1225-9. Roberts, S. B., Segil, N. & Heintz, N. (1991). Differential phosphorylation of the transcription factor Octl during the cell cycle. Science 253, 1022-6. Robertson, J. F. (1996). Oestrogen receptor: a stable phenotype in breast cancer. Br J Cancer 73,5-12. Rogatsky, I., Trowbridge, J. M. & Garabedian, M. J. (1999). Potentiation of human estrogen receptor alpha transcriptional activation through phosphorylation of serines 104 and 106 by the cyclin A-CDK2 complex. J Biol Chem 274, 22296- 302.

297 Russo, I. H. & Russo, J. (2000). Hormonal approach to breast cancer prevention. J Cell Biochem 77, 1-6. Rutherford, S. L. & Zuker, C. S. (1994). Protein folding and the regulation of signaling pathways. Cell 79, 1129-32. Ryabinin, A. E., Sato, T. N., Morris, P. J., Latchman, D. S. & Wilson, M. C. (1995). Immediate upstream promoter regions required for neurospecific expression of SNAP-25. J Mol Neurosci 6,201-10. Sabbah, M., Kang, K. I., Tora, L. & Redeuilh, G. (1998). Oestrogen receptor facilitates the formation of preinitiation complex assembly: involvement of the general transcription factor TFIIB. Biochem J 336, 639-46. Sadovsky, Y., Webb, P., Lopez, G., Baxter, J. D., Fitzpatrick, P. M., Gizang-Ginsberg, E., Cavailles, V., Parker, M. G. & Kushner, P. J. (1995). Transcriptional activators differ in their responses to overexpression of TATA-box-binding protein. Mol Cell Biol 15, 1554-63. Sadowski, I., Ma, J., Triezenberg, S. & Ptashne, M. (1988). GAL4-VP16 is an unusually potent transcriptional activator. Nature 335, 563-4. Salgia, R., Brunkhorst, B., Pisick, E., Li, J. L., Lo, S. H., Chen, L. B. & Griffin, J. D. (1995a). Increased tyrosine phosphorylation of focal adhesion proteins in myeloid cell lines expressing p210BCR/ABL. Oncogene 11, 1149-55. Salgia, R., Li, J. L., Lo, S. H., Brunkhorst, B., Kansas, G. S., Sobhany, E. S., Sun, Y., Pisick, E., Hallek, M., Ernst, T. & et al. (1995b). Molecular cloning of human paxillin, a focal adhesion protein phosphorylated by P210BCR/ABL. J Biol Chem 270, 5039-47. Salgia, R., Uemura, N., Okuda, K., Li, J. L., Pisick, E., Sattler, M., de Jong, R., Druker, B., Heisterkamp, N., Chen, L. B. & et al. (1995c). CRKL links p210BCR/ABL with paxillin in chronic myelogenous leukemia cells. J Biol Chem 270, 29145-50. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning. Cold Spring Harbor Publications, Cold Spring Harbor. Sander, K. (1985). The role of genes in ontogenesis - evolving concepts firom 1883-1983. In A history o f embryology: the Eighth Symposium o f the British Society for Developmental Biology (ed. T. J. Horder, Wilkowski J., Wylie, C.C.), pp. 477. Cambridge University Press, Cambridge. Scangos, G. & Ruddle, F. H. (1981). Mechanisms and applications of DNA-mediated gene transfer in mammalian cells - a review. Gene 14,1-10. Schaufele, F., West, B. L. & Baxter, J. D. (1992). Synergistic activation of the rat growth hormone promoter by Pit-1 and the thyroid hormone receptor. Mol Endocrinol 6, 656-65. Scholer, H. R., Ciesiolka, T. & Gruss, P. (1991). A nexus between Oct-4 and El A: implications for gene regulation in embryonic stem cells. Cell 66,291-304. Schule, R., Rangarajan, P., Kliewer, S., Ransone, L. J., Bolado, J., Yang, N., Verma, I. M. & Evans, R. M. (1990). Functional antagonism between oncoprotein c-Jun and the glucocorticoid receptor. Cell 62, 1217-26. Schwabe, J. W., Neuhaus, D. & Rhodes, D. (1990). Solution structure of the DNA- binding domain of the oestrogen receptor. Nature 348,458-61.

298 Scoville, S. A., Bufton, S. M. & Liuzzi, F. J. (1997). Estrogen regulates neurofilament gene expression in adult female rat dorsal root ganglion neurons. Exp Neurol 146, 596-9. Segil, N., Roberts, S. B. & Heintz, N. (1991). Mitotic phosphorylation of the Oct-1 homeodomain and regulation of Oct-1 DNA binding activity. Science 254, 1814- 6 . Shafie, S. M. & Grantham, F. H. (1981). Role of hormones in the growth and regression of human breast cancer cells (MCF-7) transplanted into athymic nude mice. J Natl Cancer Inst 67, 51-6. Sharrocks, A. D., Brown, A. L., Ling, Y. & Yates, P. R. (1997). The ETS-domain transcription factor family. Int J Biochem Cell Biol 29, 1371-87. Sherr, C. J. (1995). D-type cyclins. Trends Biochem Sci 20, 187-90. Shiau, A. K., Barstad, D., Loria, P. M., Cheng, L., Kushner, P. J., Agard, D. A. & Greene, G. L. (1998). The structural basis of estrogen receptor/coactivator recognition and the antagonism of this interaction by tamoxifen. Cell 95, 927-37. Shibata, H., Spencer, T. E., Onate, S. A., Jenster, G., Tsai, S. Y., Tsai, M. J. & O'Malley, B. W. (1997). Role of co-activators and co-repressors in the mechanism of steroid/thyroid receptor action. Recent Prog Horm Res 52, 141-64. Shimoyama, Y., Nagafuchi, A., Fujita, S., Gotoh, M., Takeichi, M., Tsukita, S. & Hirohashi, S. (1992). Cadherin dysfimction in a human cancer cell line: possible involvement of loss of alpha-catenin expression in reduced cell-cell adhesiveness. Cancer Res 52, 5770-4. Shughrue, P. J., Lane, M. V. & Merchenthaler, I. (1997). Comparative distribution of estrogen receptor-alpha and -beta mRNA in the rat central nervous system. J Comp Neurol 388, 507-25. Simmons, D. M., Voss, J. W., Ingraham, H. A., Holloway, J. M., Broide, R. S., Rosenfeld, M. G. & Swanson, L. W. (1990). Pituitary cell phenotypes involve cell-specific Pit-1 mRNA translation and synergistic interactions with other classes of transcription factors. Genes Dev 4, 695-711. Sinnett-Smith, J., Zachary, I., Valverde, A. M. & Rozengurt, E. (1993). Bombesin stimulation of p i25 focal adhesion kinase tyrosine phosphorylation. Role of protein kinase C, Ca2+ mobilization, and the actin cytoskeleton. J Biol Chem 268, 14261-8. Sivaraja, M., Botfield, M. C., Mueller, M., Jancso, A. & Weiss, M. A. (1994). Solution structure of a POU-specific homeodomain: 3D-NMR studies of human B-cell transcription factor Oct-2. Biochemistry 33, 9845-55. Smith, C. L., Nawaz, Z. & O'Malley, B. W. (1997a). Coactivator and corepressor regulation of the agonist/antagonist activity of the mixed antiestrogen, 4- hydroxytamoxifen. Mol Endocrinol 11,657-66. Smith, C. L., Onate, S. A., Tsai, M. J. & O'Malley, B. W. (1996). CREB binding protein acts synergistically with steroid receptor coactivator-1 to enhance steroid receptor-dependent transcription. Proc Natl Acad Sci U SA 93, 8884-8. Smith, M. D., Dawson, S. J., Boxer, L. M. & Latchman, D. S. (1998a). The N-terminal domain unique to the long form of the Bm-3a transcription factor is essential to protect neuronal cells fi*om apoptosis and for the activation of Bcl-2 gene expression. Nucleic Acids Res 26,4100-7.

299 Smith, M. D., Dawson, S. J. & Latchman, D. S. (1997b). The Bm-3a transcription factor induces neuronal process outgrowth and the coordinate expression of genes encoding synaptic proteins. Mol Cell Biol 17, 345-54. Smith, M. D., Dawson, S. J. & Latchman, D. S. (1997c). Inhibition of neuronal process outgrowth and neuronal specific gene activation by the Bm-3b transcription factor .Chem 272, 1382-8. Smith, M. D., Ensor, E. A., Coffin, R. S., Boxer, L. M. & Latchman, D. S. (1998b). Bcl-2 transcription fi*om the proximal P2 promoter is activated in neuronal cells by the Bm-3a POU family transcription factor. J Biol Chem 273, 16715-22. Smith, M. D., Ensor, E. A., Stohl, L., Wagner, J. A. & Latchman, D. S. (1999). Regulation of NGFI-A (Egr-1) gene expression by the POU domain transcription factor Bm-3a. Brain Res Mol Brain Res 74, 117-25. Smith, M. D. & Latchman, D. S. (1996). The functionally antagonistic POU family transcription factors Bm-3a and Bm-3b show opposite changes in expression during the growth arrest and differentiation of human neuroblastoma cells. Int J Cancer 67, 653-60. Smith, M. D., Morris, P. J., Dawson, S. J., Schwartz, M. L., Schlaepfer, W. W. & Latchman, D. S. (1997d). Coordinate induction of the three neurofilament genes by the Bm-3a transcription factor. JBiol Chem 272,21325-33. Sock, E., Enderich, J., Rosenfeld, M. G. & Wegner, M. (1996). Identification of the nuclear localization signal of the POU domain protein Tst-l/Oct6. J Biol Chem 271, 17512-8. Soukas, A., Cohen, P., Socci, N. D. & Friedman, J. M. (2000). Leptin-specific patterns of gene expression in white adipose tissue. Genes Dev 14,963-80. Soule, H. D. & McGrath, C. M. (1980). Estrogen responsive proliferation of clonal human breast carcinoma cells in athymic mice. Cancer Lett 10, 177-89. Southern, E. M. (1975). Detection of specific sequences among DNA firagments separated by gel electrophoresis. J Mol Biol 98, 503-17. Stem, S. & Herr, W. (1991). The herpes simplex vims trans-activator VP16 recognizes the Oct-1 homeo domain: evidence for a homeo domain recognition subdomain. Genes Dev 5, 2555-66. Stem, S., Tanaka, M. & Herr, W. (1989). The Oct-1 homoeodomain directs formation of a multiprotein-DNA complex with the HSV transactivator VP 16. Nature 341, 624-30. Struhl, K. (1991). Mechanisms for diversity in gene expression pattems. Neuron 7, 177- 81. Suen, C. S., Berrodin, T. J., Mastroeni, R., Cheskis, B. J., Lyttle, C. R. & Frail, D. E. (1998). A transcriptional coactivator, steroid receptor coactivator-3, selectively augments steroid receptor transcriptional activity. J Biol Chem 273,27645-53. Sutherland, R. L., Green, M. D., Hall, R. E., Reddel, R. R. & Taylor, I. W. (1983a). Tamoxifen induces accumulation of MCF 7 human mammary carcinoma cells in the GO/Gl phase of the cell cycle. EurJ Cancer Clin Oncol 19, 615-21. Sutherland, R. L., Hall, R. E. & Taylor, I. W. (1983b). Cell proliferation kinetics of MCF-7 human mammary carcinoma cells in culture and effects of tamoxifen on exponentially growing and plateau-phase cells. Cancer Res 43, 3998-4006.

300 Suwa, H., Ohshio, G., Imamura, T., Watanabe, G., Arii, S., Imamura, M., Narumiya, S., Hiai, H. & Fukumoto, M. (1998). Overexpression of the rhoC gene correlates with progression of ductal adenocarcinoma of the pancreas. Br J Cancer 77, 147- 52. Swanson, H. I., Chan, W. K. & Bradfield, C. A. (1995). DNA binding specificities and pairing rules of the Ah receptor, ARNT, and SIM proteins. J Biol Chem 270, 26292-302. Tai, H., Kubota, N. & Kato, S. (2000). Involvement of nuclear receptor coactivator SRC- 1 in estrogen-dependent cell growth of MCF-7 cells. Biochem Biophys Res Commun 267, 311-6. Takeshita, A., Cardona, G. R., Koibuchi, N., Suen, C. S. & Chin, W. W. (1997). TRAM- 1, A novel 160-kDa thyroid hormone receptor activator molecule, exhibits distinct properties fi-om steroid receptor coactivator-1. Chem 272,27629-34. Tanaka, M. & Herr, W. (1990). Differential transcriptional activation by Oct-1 and Oct-2: interdependent activation domains induce Oct-2 phosphorylation. Cell 60, 375-86. Tanenbaum, D. M., Wang, Y., Williams, S. P. & Sigler, P. B. (1998). Crystallographic comparison of the estrogen and progesterone receptor's ligand binding domains. Proc Natl Acad Set U SA 95, 5998-6003. Tasset, D., Tora, L., Fromental, C., Scheer, E. & Chambon, P. (1990). Distinct classes of transcriptional activating domains function by different mechanisms. Cell 62, 1177-87. Tavassoli, F. A. (1995). Pathology of advanced primary breast carcinoma. Surg Oncol Clin NAm 4, 591-600. Taylor, I. W., Hodson, P. J., Green, M. D. & Sutherland, R. L. (1983). Effects of tamoxifen on cell cycle progression of synchronous MCF-7 human mammary carcinoma cells. Cancer Res 43,4007-10. Teng, C. T., Liu, Y., Yang, N., Walmer, D. & Panella, T. (1992). Differential molecular mechanism of the estrogen action that regulates lactoferrin gene in human and mouse. Mol Endocrinol 6, 1969-81. Theil, T., McLean Hunter, S., Zomig, M. & Moroy, T. (1993). Mouse Bm-3 family of POU transcription factors: a new aminoterminal domain is crucial for the oncogenic activity of Bm-3 a. Nucleic Acids Res 21, 5921-9. Theil, T., Zechner, U., Klett, C., Adolph, S. & Moroy, T. (1994). Chromosomal localization and sequences of the murine Bm-3 family of developmental control genes. Cytogenet Cell Genet 66,267-71. Theill, L. E., Castrillo, J. L., Wu, D. & Karin, M. (1989). Dissection of functional domains of the pituitary-specific transcription factor GHF-1. Nature 342, 945-8. Thompson, M. E., Jensen, R. A., Obermiller, P. S., Page, D. L. & Holt, J. T. (1995). Decreased expression of BRCAl accelerates growth and is often present during sporadic breast cancer progression. Nat Genet 9,444-50. Tora, L., White, J., Brou, C., Tasset, D., Webster, N., Scheer, E. & Chambon, P. (1989). The human estrogen receptor has two independent nonacidic transcriptional activation functions. Cell 59,477-87. Torchia, J., Glass, C. & Rosenfeld, M. G. (1998). Co-activators and co-repressors in the integration of transcriptional responses. Curr Opin Cell Biol 10, 373-83.

301 Torchia, J., Rose, D. W., Inostroza, J., Kamel, Y., Westin, S., Glass, C. K. & Rosenfeld, M. G. (1997). The transcriptional co-activator p/CIP binds GBP and mediates nuclear-receptor function [see comments]. Nature 387, 677-84. Tremblay, G, B., Tremblay, A., Copeland, N. G., Gilbert, D. J., Jenkins, N. A., Labrie, F. & Giguere, V. (1997). Cloning, chromosomal localization, and functional analysis of the murine estrogen receptor beta. Mol Endocrinol 11, 353-65. Tsai, S. Y., Sagami, I., Wang, H., Tsai, M. J. & O’Malley, B. W. (1987). Interactions between a DNA-binding transcription factor (COUP) and a non-DNA binding factor (S300-II). Ce//50, 701-9. Tsai, S. Y., Tsai, M. J. & O'Malley, B. W. (1989). Cooperative binding of steroid hormone receptors contributes to transcriptional synergism at target enhancer elements. Cell 57,443-8. Turner, E. E., Jenne, K. J. & Rosenfeld, M. G. (1994). Bm-3.2: a Bm-3-related transcription factor with distinctive central nervous system expression and regulation by retinoic acid. Neuron 12, 205-18. Turner, R. & Tjian, R. (1989). Leucine repeats and an adjacent DNA binding domain mediate the formation of functional cFos-cJun heterodimers. Science 243, 1689- 94. Umayahara, Y., Kawamori, R., Watada, H., Imano, E., Iwama, N., Morishima, T., Yamasaki, Y., Kajimoto, Y. & Kamada, T. (1994). Estrogen regulation of the insulin-like growth factor I gene transcription involves an AP-1 enhancer. J Biol Chem 269, 16433-42. Van der Ploeg, L. H., Konings, A., Oort, M., Roos, D., Bemini, L. & Flavell, R. A. (1980). gamma-beta-Thalassaemia studies showing that deletion of the gamma- and delta-genes influences beta-globin gene expression in man. Nature 283, 637- 42. van Leeuwen, H. C., Rensen, M. & van der Vliet, P. C. (1997). The Oct-1 POU homeodomain stabilizes the adenovims preinitiation complex via a direct interaction with the priming protein and is displaced when the replication fork passes. J Biol Chem 272, 3398-405. Veenstra, G. J., van der Vliet, P. C. & Destree, O. H. (1997). POU domain transcription factors in embryonic development. Mol Biol Rep 24, 139-55. Verrijzer, C. P., Alkema, M. J., van Weperen, W. W., Van Leeuwen, H. C., Strating, M. J. & van der Vliet, P. C. (1992). The DNA binding specificity of the bipartite POU domain and its subdomains. Embo J\\, 4993-5003. Verrijzer, C. P., van Oosterhout, J. A., van Weperen, W. W. & van der Vliet, P. C. (1991). POU proteins bend DNA via the POU-specific domain. EMBO /lO , 3007-14. Vinson, C. R., Sigler, P. B. & McKnight, S. L. (1989). Scissors-grip model for DNA recognition by a family of leucine zipper proteins. Science 246,911-6. Voegel, J. J., Heine, M. J., Zechel, C., Chambon, P. & Gronemeyer, H. (1996). TIF2, a 160 kDa transcriptional mediator for the ligand-dependent activation function AF- 2 of nuclear receptors. EMBO J 15, 3667-75. Voss, J. W., Wilson, L. & Rosenfeld, M. G. (1991). POU-domain proteins Pit-1 and Oct- 1 interact to form a heteromeric complex and can cooperate to induce expression of the prolactin promoter. Genes Dev 5, 1309-20.

302 Wagner, R. L., Apriletti, J. W., McGrath, M. E., West, B. L., Baxter, J. D. & Fletterick, R. J. (1995). A structural role for hormone in the thyroid hormone receptor. Nature 378, 690-7. Wahle, E. & Keller, W. (1992). The biochemistry of 3'-end cleavage and polyadenylation of messenger RNA precursors. Annu Rev Biochem 61,419-40. Wahli, W., Martinez, E., Corthesy, B. & Cardinaux, J. R. (1989). cis- and trans-acting elements of the estrogen-regulated vitellogenin gene B1 of Xenopus laevis. J Steroid Biochem 34,17-32. Walker, P., Germond, J. E., Brown-Luedi, M., Givel, F. & Wahli, W. (1984). Sequence homologies in the region preceding the transcription initiation site of the liver estrogen-responsive vitellogenin and apo-VLDLII genes. Nucleic Acids Res 12, 8611-26. Wall, L., deBoer, E. & Grosveld, F. (1988). The human beta-globin gene 3' enhancer contains multiple binding sites for an erythroid-specific protein. Genes Dev 2, 1089-100. Washburn, T., Hocutt, A., Brautigan, D. L. & Korach, K. S. (1991). Uterine estrogen receptor in vivo: phosphorylation of nuclear specific forms on serine residues. Mol Endocrinol 5, 235-42. Weatherman, R. V., Fletterick, R. J. & Scanlan, T. S. (1999). Nuclear-receptor ligands and ligand-binding domains. Annu Rev Biochem 68, 559-81. Webb, P., Nguyen, P., Shinsako, J., Anderson, C., Feng, W., Nguyen, M. P., Chen, D., Huang, S. M., Subramanian, S., McKinemey, E., Katzenellenbogen, B. S., Stallcup, M. R. & Kushner, P. J. (1998). Estrogen receptor activation function 1 works by binding p i60 coactivator proteins. Mol Endocrinol 12, 1605-18. Webb, P., Nguyen, P., Valentine, C., Lopez, G. N., Kwok, G. R., Mclnemey, E., Katzenellenbogen, B. S., Enmark, E., Gustafsson, J. A., Nilsson, S. & Kushner, P. J. (1999). The estrogen receptor enhances AP-1 activity by two distinct mechanisms with different requirements for receptor transactivation functions. Mol Endocrinol 1672-85. Wegner, M., Drolet, D. W. & Rosenfeld, M. G. (1993). POU-domain proteins: structure and function of developmental regulators. Curr Opin Cell Biol 5,488-98. Weinberg, R. A. (1989). Oncogenes, antioncogenes, and the molecular bases of multistep carcinogenesis. Cancer Res 49, 3713-21. Weis, L. & Reinberg, D. (1992). Transcription by RNA polymerase II: initiator-directed formation of transcription-competent complexes. FasebJS, 3300-9. Weisz, A. & Bresciani, F. (1988). Estrogen induces expression of c-fos and c-myc protooncogenes in rat uterus. Mol Endocrinol 2, 816-24. Weisz, A. & Rosales, R. (1990). Identification of an estrogen response element upstream of the human c-fos gene that binds the estrogen receptor and the AP-1 transcription factor. Nucleic Acids Res 18, 5097-106. Westin, S., Kurokawa, R., Nolte, R. T., Wisely, G. B., Mclnemey, E. M., Rose, D. W., Milbum, M. V., Rosenfeld, M. G. & Glass, C. K. (1998). Interactions controlling the assembly of nuclear-receptor heterodimers and co-activators. Nature 395, 199-202. Wharton, R. P., Brown, E. L. & Ptashne, M. (1984). Substituting an alpha-helix switches the sequence-specific DNA interactions of a repressor. Cell 38, 361-9.

303 Whitfield, G. K., Jurutka, P. W., Haussier, C. A. & Haussier, M. R. (1999). Steroid hormone receptors: evolution, ligands, and molecular basis of biologic function. J Cell Biochem Suppl, 110-22. Winston, F. & Carlson, M. (1992). Yeast SNF/SWI transcriptional activators and the SPT/SIN chromatin connection. Trends Genet 8, 387-91. Wolffe, A. P. & Guschin, D. (2000). Review: chroinatin structural features and targets that regulate transcription. J Struct Biol 129, 102-22. Wood, J. N., Bevan, S. J., Coote, P. R., Dunn, P. M., Harmar, A., Hogan, P., Latchman, D. S., Morrison, C., Rougon, G., Theveniau, M. & et al. (1990). Novel cell lines display properties of nociceptive sensory neurons. Proc R Soc Lond B Biol Sci 241, 187-94. Wurtz, J. M., Bourguet, W., Renaud, J. P., Vivat, V., Chambon, P., Moras, D. & Gronemeyer, H. (1996). A canonical structure for the ligand-binding domain of nuclear receptors [see comments] [published erratum appears in Nat Struct Biol 1996 Feb;3(2):206]. Nat Struct Biol 3, 87-94. Xia, Y. R., Andersen, B., Mehrabian, M., Diep, A. T., Warden, C. H., Mohandas, T., McEvilly, R. J., Rosenfeld, M. G. & Lusis, A. J. (1993). Chromosomal organization of mammalian POU domain factors. Genomics 18, 126-30. Xiang, M., Gao, W. Q., Hasson, T. & Shin, J. J. (1998). Requirement for Bm-3c in maturation and survival, but not in fate determination of inner ear hair cells. Development 125, 3935-46. Xiang, M., Zhou, L., Macke, J. P., Yoshioka, T., Hendry, S. H., Eddy, R. L., Shows, T. B. & Nathans, J. (1995). The Bm-3 family of POU-domain factors: primary stmcture, binding specificity, and expression in subsets of retinal ganglion cells and somatosensory neurons. JNeurosci 15,4762-85. Xiang, M., Zhou, L., Peng, Y. W., Eddy, R. L., Shows, T. B. & Nathans, J. (1993). Bm- 3b: a POU domain gene expressed in a subset of retinal ganglion cells. Neuron 11, 689-701. Xu, J., Qiu, Y., DeMayo, F. J., Tsai, S. Y., Tsai, M. J. & O'Malley, B. W. (1998). Partial hormone resistance in mice with disruption of the steroid receptor coactivator-1 (SRC-1) gene. Science 279, 1922-5. Xue, D., Tu, Y. & Chalfie, M. (1993). Cooperative interactions between the Caenorhabditis elegans homeoproteins UNC-86 and MEC-3. Science 261, 1324- 8. Yang, Y., Ozawa, H., Lu, H., Yuri, K., Hayashi, S., Nihonyanagi, K. & Kawata, M. (1998a). Immunocytochemical analysis of sex differences in calcitonin gene- related peptide in the rat dorsal root ganglion, with special reference to estrogen and its receptor. Brain Res 791, 35-42. Yang, Y., Turner, R. S. & Gaut, J. R. (1998b). The chaperone BiP/GRP78 binds to amyloid precursor protein and decreases Abeta40 and Abeta42 secretion. J Biol Chem 273,25552-5. Yang-Yen, H. F., Chambard, J. C., Sun, Y. L., Smeal, T., Schmidt, T. J., Drouin, J. & Karin, M. (1990). Transcriptional interference between c-Jun and the glucocorticoid receptor: mutual inhibition of DNA binding due to direct protein- protein interaction. Cell 62,1205-15.

304 Yao, T. P., Ku, G., Zhou, N., Scully, R. & Livingston, D. M. (1996). The nuclear hormone receptor coactivator SRC-1 is a specific target of p300. Proc Natl Acad Sci U SA 93, 10626-31. Yoshinaga, S. K., Peterson, C. L., Herskowitz, I. & Yamamoto, K. R. (1992). Roles of SWIl, SWI2, and SWI3 proteins for transcriptional enhancement by steroid receptors. Science 25^, 1598-604. Zachary, I., Sinnett-Smith, J., Turner, C. E. & Rozengurt, E. (1993). Bombesin, vasopressin, and endothelin rapidly stimulate tyrosine phosphorylation of the focal adhesion-associated protein paxillin in Swiss 3T3 cells. J Biol Chem 268, 22060-5. Zajchowski, D. A., Sager, R. & Webster, L. (1993). Estrogen inhibits the growth of estrogen receptor-negative, but not estrogen receptor-positive, human mammary epithelial cells expressing a recombinant estrogen receptor. Cancer Res 53, 5004- 11. Zhang, N., Martin, G. V., Kelley, M. W. & Gridley, T. (2000). A mutation in the Lunatic fiinge gene suppresses the effects of a Jagged2 mutation on inner hair cell development in the cochlea. Curr Biol 10, 659-62. Zhao, R., Gish, K., Murphy, M., Yin, Y., Notterman, D., Hoffinan, W. H., Tom, E., Mack, D. H. & Levine, A. J. (2000). Analysis of p53-regulated gene expression pattems using oligonucleotide arrays. Genes Dev 14, 981-93. Zhu, W. Y., Jones, C. S., Amin, S., Matsukuma, K., Haque, M., Vuligonda, V., Chandraratna, R. A. & De Luca, L. M. (1999). Retinoic acid increases tyrosine phosphorylation of focal adhesion kinase and paxillin in MCF-7 human breast cancer cells. Cancer Res 59, 85-90. Zilliacus, J., Wright, A. P., Carlstedt-Duke, J. & Gustafsson, J. A. (1995). Stmctural determinants of DNA-binding specificity by steroid receptors. Mol Endocrinol 9, 389-400. Zine, A., Van De Water, T. R. & de Ribaupierre, F. (2000). Notch signaling regulates the pattern of auditory hair cell differentiation in mammals [In Process Citation]. Development 127, 3373-83. Zwijsen, R. M., Buckle, R. S., Hijmans, E. M., Loomans, C. J. & Bernards, R. (1998). Ligand-independent recruitment of steroid receptor coactivators to estrogen receptor by cyclin Dl. Genes Dev 12, 3488-98. Zwijsen, R. M., Wientjens, E., Klompmaker, R., van der Sman, J., Bernards, R. & Michalides, R. J. (1997). CDK-independent activation of estrogen receptor by cyclin Dl. Cell 88,405-15.

305