Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Purification and characterization of polyhomeotic associated proteins from Drosophila Kc1 cells Wang, Yong-Jun 2003

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2002-732568.pdf [ 15.43MB ]
Metadata
JSON: 831-1.0090643.json
JSON-LD: 831-1.0090643-ld.json
RDF/XML (Pretty): 831-1.0090643-rdf.xml
RDF/JSON: 831-1.0090643-rdf.json
Turtle: 831-1.0090643-turtle.txt
N-Triples: 831-1.0090643-rdf-ntriples.txt
Original Record: 831-1.0090643-source.json
Full Text
831-1.0090643-fulltext.txt
Citation
831-1.0090643.ris

Full Text

PURIFICATION A N D CHARACTERIZATION OF POLYHOMEOTIC ASSOCIATED PROTEINS F R O M DROSOPHILA K c l C E L L S  by  YONG-JUN W A N G B . Sc., Peking University, 1987 M . Sc., Peking Union Medical College, 1990  A THESIS S U B M I T T E D IN P A R T I A L F U L F I L M E N T OF T H E R E Q U I R E M E N T S FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in T H E F A C U L T Y OF G R A D U A T E STUDIES (Department of Zoology) W e accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH C O L U M B I A  July 2001 © Yong-Jun Wang, 2001  In  presenting this  degree at the  thesis in  partial  University of  fulfilment  of  the  requirements  British Columbia, I agree that the  for  an advanced  Library shall make it  freely available for reference and study. I further agree that permission for extensive copying  of this thesis for  department  or  by  his  scholarly purposes may be granted  or  her  representatives.  It  is  by the  understood  that  head of copying  my or  publication of this thesis for financial gain shall not be allowed without my written permission.  Department The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  fpQU, |  V  7-o°\  Abstract  The Polycomb group (PcG) genes encode repressors of homeotic and other genes, that are required to maintain silencing of target loci. Tethered P c G proteins repress reporter genes in cell lines and in Drosophila  embryos, showing that P c G proteins are  repressors. P c G genes are required for the maintenance but not the initiation of homeotic gene repression, because in P c G mutants, initiation of homeotic gene expression is normal, and then breaks down after a lag of several hours. Coimmunoprecipitation and cofractionation of P c G proteins, colocalization of P c G proteins on polytene chromosomes, and synergistic mutant phenotypes in double heterozygous mutants of P c G genes suggest that P c G proteins act through multimeric protein complexes. However, the mechanisms of PcG-mediated homeotic gene silencing are not known. One approach to this problem is to identify proteins that associate with P c G proteins in vivo in an effort to generate testable hypotheses about PcG-mediated silencing. Using epitope tagging followed by immunopurification, I purified Polyhomeotic (PH) proximal and its associated proteins from Drosophila  K c l cell nuclear extracts. I subsequently identified  the PH-associated proteins using mass spectrometry sequencing and western blotting analysis. I showed that molecular chaperones are associated with P H and that a mutation in chaperone Hsc70.4 enhances the extra sex combs phenotype of ph and Pc. These results suggest that chaperones may participate in the formation of PH-containing complexes, or may be required for silencing. I demonstrated that the histone deacetylase Rpd3 and histone binding protein p55 are associated with P H , and that the Rpd3 mutation enhances the extra sex combs phenotype of ph and Pc. Surprisingly, histone deacetylase activity was not detected in immunopurified P H . I showed using western blotting analysis that the T A T A - b i n d i n g protein (TBP) and its associating proteins T A F 4 2 and n  T A F 8 5 are also associated with P H . P H and P C were coimmunoprecipitated by antin  T B P antibody. In addition, Tbp mutants enhance the extra sex combs phenotype of ph but not Pc. Together, these findings suggest that P c G proteins use different means to silence gene expression including modifying histones and targeting the basal transcriptional machinery.  ii  Table of Contents Abstract  ii  Table of Contents  iii  List of Tables  viii  List of Figures  ix  Acknowledgements  xi  Chapter I INTRODUCTION  1  Gene Silencing  1  Homeotic Genes and Gene Silencing in Drosophila  3  The Polycomb group (PcG) Genes  4  Polycomb Group Response Elements (PREs)  5  The PcG and Silencing  8  The PcG Are More Than Homeotic Gene Repressors  10  PcG complexes  12  Cell Lines  16  Drosophila Cell Lines  17  The polyhomeotic (ph) Locus  18  Proteins Encoded by ph Proximal (PHP) and Distal (PHD)  21  Polycomb  23  Chapter II EXPRESSION OF EPITOPE TAGGED POLYHOMEOTIC PROXIMAL (PHP) IN DROSOPHILA KC1 CELLS INTRODUCTION  26 26  iii  Transfection of C e l l Lines  26  Expression System  29  Stably-transfected C e l l Lines versus Baculovirus-infected Expression Systems  30  Epitope Tagging  32  Advantages and Limitations of Epitope Tagging  32  Epitopes and Antibodies  33  Locating the Tag  34  Applications of Epitope Tagging  35  RESULTS A N D DISCUSSION  37  Expression Constructs for Stable Cell Transformation  37  Selection of Stable Cell Lines Expressing Tagged P H P ( F - P H P - H A )  41  Characterization of Tagged P H P Expression  45  Test of the Epitope for Immunoaffinity Purification  55  Selection of Stable C e l l Lines Expressing Tagged P C (F-PC)  58  Chapter III P U R I F I C A T I O N O F P H P A N D A S S O C I A T E D P R O T E I N S F R O M N U C L E A R EXTRACTS OF KC1 CELLS INTRODUCTION  62 62  Nuclear Extract Preparation from Cell Lines  62  General Considerations for Handling Proteins  63  Ion Exchange Chromatography  63  G e l Filtration Chromatography  65  Immunoaffinity Chromatography  65  iv  RESULTS A N D DISCUSSIOIN  66  Large Scale K c l Culture  66  Comparison of Native and Tagged P H P on G e l Filtration Chromatography  67  Development of Conventional Chromatographic Purification Scheme  68  Immunoaffinity Purification of F - P H P - H A Complexes  78  Double Immunoaffinity Purification of F - P H P - H A Complexes  84  Chapter IV F U R T H E R C H A R A C T E R I Z A T I O N O F PHP-ASSOCIATED PROTEINS  89  INTRODUCTION  89  RESULTS A N D DISCUSSION  91  Western Blotting Analysis of Immunoaffinity Purified P H P  91  P H P Cosediments with P C , R P D 3 and p55 on Glycerol Gradients  95  Mass Spectrometry Sequencing of PHP-associated Proteins  98  Hsc70.4 Enhances the Homeotic Phenotype of ph and Pc Mutations  103  Rpd3 Enhances the Homeotic Phenotype of ph and Pc Mutations  106  PHP-associated Proteins Have N o Significant Histone Deacetylase Activity in vitro  107  R N A Is Not Required for the Integrity of P H P and P C Complexes  109  Chapter V T A T A - B I N D I N G P R O T E I N (TBP) A N D T B P - A S S O C I A T E D F A C T O R S (TAF s) A S S O C I A T E W I T H PHP n  INTRODUCTION  114 114  The P c G and Gene Silencing  114  v  T B P and Transcriptional Repression  115  RESULTS A N D DISCUSSION  117  T B P Copurifies with P H P and P C  117  TBP-associated Factors (TAFs) Associate with P H P in vivo  120  Coimmunoprecipitation of T B P with P H P and P C  122  G S T - T B P Fusion Protein Pulls-down P H P and P C from Nuclear Extract 124 Size Fractionation of P H P , P C , T B P , T A F s  126  Tbp Mutations Enhance Homeotic Phenotype of ph but not Pc  131  u  Chapter VI C O N C L U S I O N S  138  P c G Complexes  138  Chaperones Associate with P H P  140  The Transcriptional Machinery and PcG-mediated Gene Silencing  141  Histone Deacetylation and PcG-mediated Gene Silencing  144  Chapter VII M A T E R I A L A N D M E T H O D S D N A Subcloning  146 146  pUASTPHPHA  146  pNTP14F-PHP-HA  146  pNTP14F-PC  147  pGEX4TlPHO  147  dTBP  148  Polymerase Chain Reaction ( P C R )  148  vi  Expression and Purification of G S T Fusion Proteins  148  Antibody Generation in Rabbits and Their Purification  149  Cell Transfection and Selection of Stable Transformed C e l l Lines  150  Nuclear Extract Preparation from Cell Lines  151  Protein Chromatography  153  Ion Exchange Chromatography  153  G e l Filtration Chromatography  154  Immunoaffinity Purification  154  Glycerol Gradient Ultracentrifugation  155  S D S - P A G E and Silver Staining  156  Sample Preparation for In G e l Proteolytic Digestion and Mass Spectrometry  157  Histone Deacetylase ( H D A C ) Assay  157  G S T Pull-down Assay  158  Co-Immunoprecipitation (Co-IP)  159  Western Blotting and Immunostaining  159  Fly Strain Maintenance and Genetic Crosses  160  Nomenclature  162  Bibliography  165  vii  List of Tables Table 1.1  P c G homologues in human and mouse  6  Table 1.2  P c G and interacting proteins  15  Table 3.1  Protein distribution on B i o - R e x column  72  Table 3.2  Protein distribution on S P Sepharose column  76  Table 3.3  Protein distribution on Q Sepharose column  80  Table 4.1  Proteins in the F - P H P - H A complex identified by U . L C / M S / M S  101  Hsc70.4 enhances the extra sex combs phenotype ofphandPc  104  Table 4.2  Table 4.3  Rpd3 enhances the extra sex combs phenotype ofphandPc  108  Table 5.1  Tbp enhances the extra sex combs phenotype of ph  133  Table 5.2  Tbp does not enhance the extra sex combs phenotype of Pc  135  viii  List of Figures Fig. 1.1  M a p of the ph locus  20  Fig. 1.2  Sequence motifs shared by P H P , P H D and mutant P H and P H 2  22  4  Fig. 2.1  M a p of p N T P 1 4 F - P H P - H A  38  Fig. 2.2  Internal initiation of p h translation  42  Fig. 2.3  M a p of p N T P 1 4 F - P C  43  Fig. 2.4  Diagram of cell growth under different  p  concentrations of Zeocin  46  Fig. 2.5  Western blotting analysis of selected clones resistant to Zeocin  48  Fig. 2.6  Western blotting analysis of the two stably transformed clones expressing tagged P H P  49  Fig. 2.7  Comparison of endogenous and tagged P H P  50  Fig. 2.8  Expression test of stably transfected tagged p h  Fig. 2.9  Comparison of K c l cell growth with and without copper ion  54  Fig. 2.10  Expression levels of endogenous and tagged p h  56  Fig. 2.11  Immunoaffinity test with anti-HA and a n t i - F L A G affinity beads  59  Fig. 2.12  Cloning and testing of p N T P 1 4 F - P C  61  Fig. 3.1  Western blotting analysis of gel filtration  53  p  p  fractions of nuclear extract  69  Fig. 3.2  Chromatogram of B i o R e x 70 chromatography  70  Fig. 3.3  Western blotting analysis of immunoaffinity-purified F - P H P - H A from B i o R e x column fractions  73  Fig. 3.4  Chromatogram of SP Sepharose chromatography ix  75  Fig. 3.5  Western blotting analysis of immunoaffinity-purified  F - P H P - H A from SP Sepharose column fractions  77  Fig. 3.6  Chromatogram of Q Sepharose chromatography  79  Fig. 3.7  Western blotting analysis of immunoaffinity-purified F - P H P - H A from Q Sepharose column fractions  81  Fig. 3.8  Silver staining analysis of immunoaffinity-purified F - P H P - H A and associated proteins  83  Fig. 3.9  Immunoblotting analysis of P S C across the column fractions  85  Fig. 3.10  Silver staining analysis of double immunoaffinity-purified F - P H P - H A and associated proteins  88  Immunoblotting analysis of immunoaffinity-purified PHP-associated proteins  94  Fig. 4.1  Fig. 4.2  Immunoblotting analysis of glycerol gradient fractions of F-PHP-HA-associated proteins  97  Fig. 4.3  Coomassie staining of S D S - P A G E gel  99  Fig. 4.4  Protein sequence alignment of the J domains of DnaJ chaperones  102  Fig. 4.5  Histone deacetylase assays of immuno-purified F - P H P - H A  110  Fig. 4.6  Silver staining analysis of F - P H P - H A and PC-associated proteins in the presence of RNase A  112  Fig. 5.1  Immunoblotting analysis of chromatography fractions and immuno-purified F-PHP-HA-associated proteins  119  Fig. 5.2  T A F 8 5 and T A F 4 2 associate with F - P H P - H A  121  Fig. 5.3  coimmunoprecipitation of F - P H P - H A and F - P C with T B P  123  Fig. 5.4  In vitro binding of P H and P C to T B P  125  Fig. 5.5  Cofractionation of F - P H P - H A , P C and T B P on gel filtration chromatography  128  Glycerol gradient analysis of immuno-purified PHP-associated proteins  130  Sketch map of plasmid p U A S T P H P H A  161  Fig. 5.6  Fig. 7.1  n  U  x  Acknowledgements I would like to thank my supervisory committee members, Drs. V . A u l d , H . Brock, C , Brown, T. Grigiliatti, and M . Roberge, who gave me many useful supports during the course of my study. Special thanks to D r . H Brock who did all the extra sex combs phenotype scoring work in this thesis. D r . Brock also helped me a lot on the thesis writing, especially he always gave me back very detail suggestions quickly. The freedom and encouragement he gave are highly appreciated.  I also want to thank the Brock L a b members, B o b . Argiropoulos, Jack. Chevalier, Ester O ' D o r , Cynthia. Fisher, Jacob Hodgson, who gave me a nice work environment and useful suggestions in those many lab meetings. Special thanks to Jacob Hodgson who helped me a lot on biochemical techniques, especially in the setting up of ion exchange columns and glycerol gradient analysis.  I am indebted to Paul Adler, Peter Becker, R i c k Jones, J i m Kadonaga, B o b Kingston, Yoshihiro Nakatani, Renato Paro, Jeff Simon, Robert Tjian, D a v i d Wasserman for providing me antibodies; and T o m Pfeiffer, and Dwyne Hedgedus and L o y V o l k m a n who supplied expression vectors.  I also would like to thank my family members, especially my mother and mother-in-law, who helped around my house when I worked in the nights and weekends; my daughter and my wife whose understanding for my absence from home so much.  xi  Chapter I Introduction Gene Silencing Gene silencing is a form of transcriptional repression that involves the assembly of specialized, heritable structures of chomatin confined to certain domains within chomosomes. Silencing has three recognizably separate phases: establishment, maintenance, and inheritance (Loo and Rine, 1995). Silencing systems include matingtype silencing of HML and HMR in yeast (reviewed in L o o and Rine, 1995); telomeric position effect (TPE) at the telomere of yeast (reviewed in Grunstein, 1998); heterochomatin position-effect variegation ( P E V ) in Drosophila (reviewed in Karpen, 1994; Henikoff, 1994); gametic imprinting in mammals (reviewed in Barlow, 1995); and X chomosome inactivation in female mammals (reviewed in Lee and Jaenisch, 1997). The features of silencing are: 1) it can act at long range; 2) it cannot distinguish among genes that use different promoters or that are transcribed by different R N A polymerases; 3) it is heritable for many cell generations or even for the whole life of the organism; 4) the silenced region of chomatin is generally not accessible to other proteins like restriction enzymes; 5) some types of silencing are metastable, for example, they can be overcome by overexpression of transcriptional activators. In Drosophila, the Polycomb group (PcG) of genes have become a model system for studying how gene expression is silenced. The lack of distinction between gene repression and silencing in the literature partly reflects the fact that we do not know very much about the mechanisms of gene repression, and is partly due to the fact that silencing is a special mode of repression and therefore the mechanisms overlap. T o support this, it has been found that S U M l p in  1  yeast can be either a gene specific repressor or silencer (Rusche and Rine, 2001). Therefore very often these two terms are interchangeable in the literature. To better understand silencing mechanisms, it is necessary to review briefly how gene repression is achieved. Important insights into the mechanism of gene repression derive from studies in phage and bacteria, in which the most frequently observed mechanisms involve competition between D N A binding proteins and general transcription factors or interference of transcription initiation (reviewed in Rojo, 1999; Rojo, 2001). It is now clear that a remarkable number of basic repression mechanisms are conserved between prokarotes and eukaryotes (Levine and Manley, 1989; Johnson, 1995; Maldonado et al., 1999; Struhl, 1999). Therefore it is not surprising to find out that similar mechanisms are used in eukaryotes for gene repression. In eukaryotes, these include: 1) masking a transcriptional activation domain; 2) blocking interaction of an activator with other components of the transcriptional machinery; 3) displacing an activator from the D N A (reviewed in Levine and Manley, 1989; Johnson, 1995; HannaRose and Hansen, 1996); and 4) inactivating the components of the core transcription machinery ( U m et al., 1995; reviewed in Maldonado et al., 1999). Furthermore, D N A response elements can exert allosteric effects on transcriptional regulators, such that regulators may activate transcription in the context of one gene, yet repress transcription in another (reviewed in Lefstin and Yamamoto, 1998). Are the general repression mechanisms described above responsible for gene silencing? The answer is not known yet but it would not be surprising to find out that they are partly responsible for at least some of the strategies that may be used for silencing. Nevertheless, the repression mechanisms described above are obviously too  2  simple for eukaryotic gene silencing as eukaryotic D N A is packaged into chomatin, and there exist fundamental differences in gene regulation between prokaryotes and eukaryotes (Struhl, 1999).  Homeotic Genes and Gene Silencing in Drosophila The Drosophila homeotic genes encode transcription factors that determine the identities of the parasegments of embryos and subsequently the segments of adults (Kaufman et al., 1980). There are 8 homeotic genes in Drosophila which are clustered together on chomosome 3 into two linkage groups called the Antennapedia-complex ( A N T - C ) and bithorax-complex ( B X - C ) (reviewed in Manak and Scott, 1994). Expression of homeotic genes is confined to unique domains along the body axis whose anterior and posterior boundaries are maintained thoughout development (reviewed in A k a m , 1987). Homeotic genes have the potential to be active outside their normal domains as they contain enhancers which mediate their expression thoughout the embryo. Ectopic expression is prevented by transcriptional repression operating in two consecutive steps: initiation and maintenance (Bienz, 1992). In the early embryos, expression of homeotic genes is delimited by a transient activity of the segmentation genes (Akam, 1987) which act at short ranges (repressors affect exclusively nearby activators) (Gray et al., 1994). After the initial signal fades away, the restricted expression of homeotic genes is ensured by a permanently acting mechanism of long range repression (i.e. repressors that affect activators bound to D N A anywhere within the gene) (Gray and Levine, 1996), which is also called transcriptional silencing (Brand et  3  al., 1985). This transcriptional silencing requires the activity of the P c G genes in Drosophila.  The Polycomb group (PcG) Genes The P c G genes in Drosophila were originally identified though their homeotic phenotypes. Sixteen P c G genes are well characterized genetically though mutant screens based on homeotic transformation (Lewis, 1978), cuticle defects (Nusslein-Volhard et al., 1984) , or enhancement of homeotic phenotypes (Kennison and Russell, 1987), and it is estimated that there are about 30-40 members in total (Landecker et al., 1994; Jurgens, 1985) . Thirteen of them have been cloned : Additional sex combs (Asx) (Sinclair et al., 1998), chomatin condensation factor (ccf) (Kodjabachian et al., 1998), cramped  (crm)  (Yamamoto et al., 1997), extra sex combs (esc) (Gutjah et al., 1995), Enhancer of Polycomb (E(Pc)) (Stankunas et al., 1998), Enhancer ofzeste (E(z)) (Jones and Gelbart, 1993), Polycomb (Pc) (Paro and Hogness, 1991), Pleiohomeotic 1998), Polycomb-like  (Pel) (Lonie et al., 1994), polyhomeotic  (Pho) (Brown et al.,  (ph) (DeCamillis et al.,  1992), Posterior sex combs (Psc) (Brunk et al., 1991a), Sex combs on midleg (Scm) (Bornemann et al., 1996), and Suppressor 2 ofzeste (Su(z)2) (Brunk et al., 1991b). In P c G mutants, early homeotic gene expression is normal but at later developmental stages the embryos show posterior transformations that resemble those exhibited by gain-of-function in homeotic genes (Struhl and A k a m , 1985; Jones and Gelbart, 1990; Soto et al., 1995). Therefore P c G genes act as repressors and are required at late stages to maintain the repression of homeotic genes outside of their normal expression domains.  4  In mammals, homologs of most Drosophila P c G genes have been described (Table 1.1) (reviewed in Schumacher and Magnuson, 1997; Gould, 1997). Knockout mice lacking P c G genes exhibit posterior transformations, showing that P c G genes are responsible for maintaining the repressed state of H o x genes. In mice, P c G genes also function in hematopoiesis (Schumacher and Magnuson, 1997). In addition, it has been shown that mouse M33 (Pc homolog) can rescue phenotypes of Pc in Drosophila  (Muller  et al., 1995). The conservation of structure and function from flies to mammals indicates that P c G genes are important gene regulators for development and differentiation in higher eukaryotes.  Polycomb Group Response Elements (PREs) A l l the P c G genes characterized so far are expressed ubiquitously (Gutjah et al., 1995; DeCamillis and Brock, 1994; Lonie et al., 1994; Martin and Adler, 1993; Paro and Zink, 1992). H o w do these ubiquitous proteins recognize specific targets only in domains where their repressive activity is required? H o w do they cooperate with each other to regulate different genes? Although none of the known P c G proteins binds to D N A except P H O (Brown et al., 1998) in vitro, in vivo P c G proteins associate with highly specific sites on polytene chomosomes. P R E s were originally defined by fusing a regulatory D N A element from the Ubx gene to a lacZ reporter gene which mimics the endogenous Ubx gene expression pattern. This pattern is sensitive to P c G mutations (Simon et al., 1993). P R E s have been identified from homeotic genes like proboscipedia Antennapedia  (pb), Sex combs reduced  (Antp), Ultrabithorax (Ubx), Abdominal A (AbdA),  5  (Scr),  Abdominal B  (AbdB),  Table 1.1  P c G homologues in human and mouse  Drosophila Proteins  Human Homologues  Mouse Homologues  Protein Motifs Shared  Additional sex combs  ASXL1, ASXL2  A s x l l , Asxl2  None known  Enhancer of Polycomb  EPC1,EPC2  E p c l , Epc2  None known  Enhancer of zeste  EZH1,EZH2  E n x l , Enx2  SET  Extra sex combs  EED  eed  W D - 4 0 repeat  Pleiohomeotic  YY1  -  Zinc finger D N A binding  Polycomb  HPC1,HPC2,HPC3  MPcl,MPc2  Chromodomain  Polycomblike  PHF1  -  P H D finger  Polyhomeotic  HPH1,HPH2  MPhl,MPh2  SAM, Zinc finger  Posterior sex combs  BMI1,MEL18  B m i l , mell8  R I N G finger  Sex combs on midleg  SCML1.SCML2, SCMH1  S c m l l , Scml2,  SAM,  Scmhl  Zinc finger  BMI1,MEL18  Bmil.Mell8  R I N G finger  Suppressor of zeste 2  6  and non-homeotic genes like engrailed (en) and polyhomeotic  (ph) (Kapoun and  Kaufman, 1995; Gindhart and Kaufman, 1995; Chan et al., 1994; Chiang et a l , 1995; Hagstrom et al., 1997; Kassis, 1994; Fauvarque and Dura, 1993). A l l P R E s have three properties in common: 1) P R E s exhibit pairing-sensitive repression of reporter genes that is sensitive to P c G genes mutations; 2) P R E s recruit P c G proteins on polytene chomosomes; 3) P R E containing transposons tend to insert at chomosomal sites containing P R E s , called the homing effect. Although P R E s have common properties, there is no consensus sequence among all the P R E s identified so far. Different P R E s respond to different P c G genes ( Kassis, 1994; Fauvarque and Dura, 1993; Kapoun and Kaufman, 1995; Gindhart and Kaufman, 1995). The Mcp P R E of Abd-B gene is not sensitive to Trl mutations (encoding G A G A Factor), but Trl mutations suppress the iab-7 P R E mediated silencing of miniwhite gene. The G A G A consensus sequence was found in the iab-7 P R E but not in that of Mcp P R E (Hagstrom et al., 1997). These data indicate that different protein complexes associate with different P R E s , perhaps accounting for the general lack of consensus sequence among the identified P R E s . If P R E s are the true D N A elements responsible for the specific targeting of P c G proteins to respective genes, then how do they do this? Sequence-specific D N A binding proteins are probably responsible for the specificity. Hagstrom et al. (Hagstrom et al., 1997) found that the silencing mediated by the iab-7 P R E appears to depend on a chomatin remodeling protein, the G A G A factor in Drosophila.  Consistent with this idea,  Horard et al. (Horard et al., 2000) showed that P c G binding to bxd P R E fragments is  7  dependent on consensus sequences for the G A G A factor. Hodgson et al. (Hodgson et al., 2001) showed that G A G A factor consensus sequence is required for the binding of a P H containing complex, and that d(GA)3 repeats, which bind G A G A factor, are important for bxd PRE-mediated silencing in flies. But G A G A sequences alone are not sufficient for silencing, bxd P R E fragments lacking G A G A binding sites can still bind P c G complexes in vitro (Horard et al., 2000), suggesting that other D N A sequences and respective binding proteins are also required. Interestingly, P H O , one of the P c G , has conserved zinc fingers of mammalian Y Y 1 , and can bind to D N A specifically (Brown et al., 1998). Consensus sites for P H O have been identified in many but not all P R E s (Mihaly et al., 1998). P H O sites are necessary but not sufficient for P R E activity in vivo (Brown et al., 1998; Fritsch et al., 1999; Shimell et al., 2000).  The PcG and Silencing It has been over a decade since it was demonstrated that P c G proteins are responsible for the maintenance of homeotic gene repression in Drosophila (Struhl and A k a m , 1985). Consistent with earlier genetic observations, later work tethering P c G proteins to reporter gene demonstrated that P c G proteins can repress reporter gene expression in cell lines (Bunker arid Kingston, 1994) and Drosophila embryos (Muller, 1995; Poux et al., 2001; Roseman et al., 2001). But the mechanisms of PcG-mediated homeotic gene silencing are not known. This is largely due to the fact that no P c G protein except P H O has D N A binding activity in vitro and that no chomatin modifying activity associated with P c G proteins had been identified until recently (van der V l a g and Otte, 1999; Tie et al., 2001).  8  Nevertheless many models have been proposed to explain PcG-mediated silencing. The similarity of P C and HP1 led to the speculation that P c G proteins might promote heterochomatinization of target loci to prevent access by transcription factors (Alberts and Sternglanz, 1990; Gaunt and Singh, 1990; Paro, 1990). Second, P c G proteins may act on R N A polymerase II or general transcription factors to repress transcription (Bienz, 1992; Laney and Biggin, 1992). Third, it has been suggested that P c G proteins maintain silencing by sequestering target genes to a particular nuclear compartment (Paro, 1993; Schlossherr et al., 1994). Fourth, P c G interactions between different sites along a chomosome may prevent enhancer-promoter interactions (Pirrotta and Rastelli, 1994; Pirrotta et al., 1995). Fifth, it has been suggested that P c G proteins antagonize chomatin remodelling complexes encoded by members of the trithorax group (trxG) (Shao et al., 1999). Lastly, P c G protein complexes deacetylate histone tails, leading to the compaction of chomatin and thus silencing (Lohuizen, 1999; van der V l a g and Otte, 1999; Tie etal.,2001). There may be so many models because some of the models described above are not mutually exclusive. For example, compartmentalization, transcriptional machinery interference and histone deacetylase models may be all used to ensure the requirements of long term and stable silencing. Silencing at the mating type loci in S. cerevisae, the most extensively studied gene silencing system, uses all thee strategies to achieve silencing (Maillet et al., 1996; Cockell and Gasser, 1999; Sekinger and Gross, 2001; Imai et al., 2000; Landry et al., 2000). Moreover, the T u p l - S S n 6 global repressor (a repressor that regulates many different classes of genes in the cell) in budding yeast represses its target genes by binding to histone tails and occluding the transcriptional machinery (Ducker and  9  Simpson, 2000), recruiting a histone deacetylase activity (Watson et al., 2000), and interacting with the components Srbl0/11 of R N A polymerase II holoenzyme (Zaman et al., 2001). The multiple mechanisms used together by one repression or silencing system can probably yield stable, long term repression that is required for developmental and differentiation-related genes.  The PcG Are More Than Homeotic Gene Repressors Several lines of evidence indicate that most P c G proteins are general transcriptional regulators rather than dedicated homeotic gene repressors. First, the ubiquitous cellular distributions of P c G proteins are consistent with a general role. Second, the pleiotropic phenotypes of many P c G genes mutations indicate that they are involved in numerous processes besides homeotic silencing. For example, P c G genes mutations have been described that affect oogenesis, dorsal-ventral patterning, neural development or cell proliferation (Phillips and Shearn, 1990; Paro, 1990; Adler et al., 1991). Third, all P c G proteins so far tested show immunolocalization on polytene chomosomes at 80 to 100 genomic sites in addition to the two homeotic loci (Franke et al., 1992; Martin and Adler, 1993; Rastelli et al., 1993; Lonie et al., 1994). Finally, mutations in P c G genes have been shown to affect expression of the segmentation genes engrailed, even-skipped, knirps and giant (Moazed and O'Farrell, 1992; Pelegri and Lehman, 1994; M c K e o n et al., 1994). Mutations in two P c G genes, E(z) and Asx exhibit both anterior and posterior transformations, consistent with a role in activation as well as repression of homeotic loci (Sinclair et al., 1992; LaJeunesse and Shearn, 1996; M i l n e et al., 1999). A recent study screened deletions of 70% of the Drosophila genome for their ability to either enhance or  10  suppress phenotypes for trxG genes mutations (Gildea et al., 2000). Surprisingly, five of the noncomplementing deletions uncover genes previously classified as members of the Polycomb group, suggesting that the products of these P c G genes are required to maintain chomatin in both transcriptionally inactive and active states. These five genes were renamed E T P (enhancers of trithorax and Polycomb) (reviewed in Brock and van Lohuizen, 2001). P c G proteins have been shown to associate with active genes by in vivo crosslinking (Strutt and Paro, 1997b). Polytene staining studies show that P C , P H , P S C and A S X bind to 2 D locus (Zink and Paro, 1989; DeCamillis et al., 1992; Rastelli et a l , 1993; Sinclair et al., 1998), where ph is located, indicating that ph expression may be regulated by these proteins. Therefore P c G binding does not necessarily mean repression, because the presence of the multimeric complex at the ph locus does not switch off the gene expression. In addition, in vivo crosslinking showed that P H and P S C are associated with expressed genes, such as gene VI, Abd-B, and empty spiracles (ems) in tissue culture cells (Strutt and Paro, 1997b). Intriguingly, some evidence indicates that P c G proteins can be activators. The first evidence is that amorphic allele Psc  Arpl  hypomorphic Psc  1445  and the  both cause less ph transcription, and the gain of function allele Psc  1  gives a higher transcription of ph , indicating a positive effect of P S C protein on ph expression, contrary to the negative effect of other P c G proteins (Fauvarque et al., 1995). Mutation in ph reduces gene VI transcription in tissue culture cells (Strutt and Paro, 1997b). Together, these results indicate that P c G proteins have multiple functions and can play roles in both activation and repression. To dissect the role of P c G proteins, it w i l l be  11  necessary to characterize the proteins that associate in vivo with P c G proteins, which is the principal aim of this thesis.  PcG Complexes Most, i f not all, proteins in the cell perform their diverse functions though interactions with different partners (Alberts and Miake-Lye, 1992; Alberts, 1998). Several lines of evidence indicate that P c G proteins form complexes. First, in genetic experiments, double- or triple-mutant combinations of the P c G genes show a synergistic enhancement of the homeotic phenotypes observed with single mutants (Jurgens, 1985; Cheng et al., 1994), consistent with their participation in a common structure or pathway. Second, antibody staining of polytene chomosomes reveals that P c G proteins bind to about 100 overlapping euchomatic sites (Zink and Paro, 1989; DeCamillis et al., 1992; Franke et al., 1992; Rastelli et al., 1993; Martin and Adler, 1993; Lonie et al., 1994). A large number of chomosomal target sites are common to different P c G proteins, though the cytological resolution does not exclude the possibility that they are bound to neighboring but independent sites. P H and P C are present in a soluble multimeric protein complex of 2-5 M D a that includes 10 to 15 other proteins (Franke et al., 1992). P c G homologues in other species also form complexes (Reijnen et al., 1995; Gunster et al., 1997; A l k e m a e t a l . , 1997). P c G proteins form different complexes on different target genes. Genetic evidence indicates that the efficient repression of the segmentation gap genes knirps and giant is dependent on E(z) and the Su(z)2 but not on other P c G genes (Pelegri and Lehman, 1994). Moreover, the P c G genes exhibit allele-specific repression of transformed P[en-  12  white] expression (Kassis, 1994). O f particular interest is the finding that telomeric P E V is affected by mutations in the Psc-Su(z)2 gene complex, not by Su(var) or Pc mutations (L Wallrath, S E l g i n , unpublished). Third, mutations in the P C chomodomain abolish the chomosomal association of P H (Messmer et al., 1992), and inactivation of the E(z) gene results in the loss of most, but not all, of the binding sites for P S C , SU(Z)2, P C and P H (Rastelli et al., 1993). P C , P H and P S C are differentially distributed on regulatory sequences of the engrailed-related  gene invected (Strutt and Paro, 1997b). These results  suggest that P c G proteins form complexes, and thus the complexes vary in composition depending on the chomosomal site. The domains of P c G proteins that are conserved in mammalian homologues, and that are found in other chomatin proteins, likely mediate protein-protein interactions (Table 1.1). Examples include the chomodomain in P C , Zinc finger and S P M domain in P H and S C M , R I N G finger in P S C and SU(Z)2, S E T domain in E(Z), W D 4 0 repeats in E S C , and P H D finger in P C L (Satijn and Otte, 1999). The S P M domain mediates direct interaction between S C M and P H , and homotypic interactions of S C M and P H (Peterson et al., 1997; K y b a and Brock, 1998a). P H coprecipitates with P S C and P C from nuclear extracts. P S C and P H interact though regions of sequence conservation with mammalian homologues, the homology domain I ( H I ) of P H (amino acids 1297-1418) and the helixturn-helix ( H T H ) domain of P S C (amino acids 250-335). P S C contacts P C primarily at the H T H motif and the R I N G finger (Kyba and Brock, 1998a). Table 1.2 lists some of the P c G and their interactors, more details are reviewed in Satijin and Otte (1999). Chomodomain swap experiments have demonstrated specific interactions between P c G proteins in vivo. H P 1 containing a P C chomodomain is recruited to P C binding sites on  13  polytene chomosomes (Platero et al., 1995), and can recruit P c G proteins to the heterochomatin (Platero et al., 1996). These results support the idea that, like the yeast SIR (silent information regulator) proteins, the P c G proteins form complexes held together by a chain of protein-protein interactions. Biochemically, a PHP-containing complex, P R C 1 , has been purified from transgenic embryos using an epitope-tagging strategy (Shao et al., 1999). P R C 1 contains the proximal isoform of P H ( P H P , see below), P S C , S C M , P C and six other unknown subunits. Preincubation of P R C 1 with nucleosomal arrays blocked the ability of these arrays to be remodeled by S W I / S N F . Addition of P R C 1 to arrays at the same time as S W I / S N F did not block remodeling. Thus the authors concluded that P R C 1 and S W I / S N F might compete with each other for the nucleosomal template. Furthermore, it has been shown that P R C 1 was active on nucleosomal arrays formed with tailless histones, suggesting that P R C 1 functions without interacting with histone tails, and that histone modifications are not required either. Genetic studies indicate that esc functions at a different time than Pc and ph (Simon et al., 1995), and biochemical studies support an interaction between E S C and E ( Z ) (Jones et al., 1998; Tie et al., 1998), suggesting that these proteins are part of a different complex. Consistent with this idea, a complex containing E S C and E ( Z ) was purified using F L A G tagged E S C from embryos (Tie et al., 2001). The E S C / E ( Z ) complex is distinguishable from P C on a gel filtration column. The E S C / E ( Z ) fractionates as a 600 K D a complex containing histone deacetylase R P D 3 and the histone binding protein P55, which is also a component of the chomatin remodeling complex N U R F and the chomatin assembly complex C A F - 1 (Martinez-Balbas et al., 1998; Tyler et al., 1996).  14  Tablel.2 P c G and interacting proteins Interacting Proteins  PcG Proteins Pc Pc MPcl HPC1 HPC2  Drosophila mouse human human  Psc, R L N G l M P c l , B m i l , R i n g 1A and B RING1 RINGl,CtBP  Psc Psc Bmil BMI1  Drosophila mouse human  Pc, P h MPhl,RinglB,MPh2 RING1,BMI1  Drosophila mouse mouse human human  Ph, S c m M P h l , B m i l , Scm B m i l , Ring IB HPH1,HPH2 HPH1,HPH2  Scm Scm  Drosophila  Scm, M P h l  E(z) E(z) Enxl Enx2  Drosophila mouse mouse  Esc eed, E E D , V a v eed  Esc Esc eed EED  Drosophila mouse human  E(z) E n x l , Enx2 Enxl  Ph Ph MPhl MPh2 HPH1 HPH2  15  Cell Lines C e l l line, or established cell line is a terminology proposed by Hayflick and Moorehead (Hayflick and Moorhead, 1961), to refer to any line that can be cultured for such a long time that it has apparently developed the potential to be subcultured indefinitely in vitro (Paul, 1973). Therefore cell lines are also called immortal cell lines. A s a general rule, a line of mammalian cells must have been subcultured at least 70 times, at intervals of 3 days, before being considered as fully established. For invertebrate cell lines, a comparable timetable should be around 40 subcultures at 3 to 5 day intervals, or some 120 cell division cycles (Simcox et al., 1985). Cell lines have advantages as well as drawbacks. For example, cell lines can be continuously cultured in vitro, and thus provide unlimited resources for analysis. C e l l lines are clonal, so they are relatively homogeneous with respect to genotype and phenotype. C e l l lines exhibit stable properties such as virus susceptibility, and give reproducible results.  Cell lines are adapted to in vitro conditions, so they are easier to  maintain than primary cultures; in contrast, establishment of primary cultures can be tedious, labour-intensive and time consuming (reviewed in Crane, 1999). Although cell lines have been used widely and still play important role in biology research, they are not ideal for all purposes. Since cell lines are derived either from tumors or normal cells transformed using viruses (Yeager and Reddel, 1999), their karotypes and physiology may be changed. Therefore, cell lines are not good for some development and differentiation research (Pfeiffer et al., 1977; Echalier, 1997; Kienlen-Campard et al., 2000).  16  Drosophila Cell Lines In vitro permanent cell lines of Drosophila were first established in 1968 (Gvozdev and Kakpakov, 1968). Since then, many cell lines have been made (Echalier and Ohanessian, 1969; Kakpakov et al., 1969; Schneider, 1972). These include cell lines from wild-type embryos and from those carrying particular mutations (Sang, 1981) or other temperature-sensitive lethals (Simcox et al., 1985). Most available Drosophila cell lines are derived from primary cultures of mechanically dissociated young embryos (6 to 12 h) (Echalier, 1997) or early larvae; success at deriving cell lines from other developmental stages, such larval imaginal discs, has been limited (Ashburner, 1989). The tissue origins of those lines derived from minced embryos are usually not known. A very small number of Drosophila cell lines have been extensively used: Echalier and Ohanessian's K c l line (Echalier and Ohanessian, 1969), Schneider's SI, S2, and S3 lines (Schneider, 1972), Kakpakov and Gvozdev's 67J25 (Kakpakov et al., 1969) and to a lesser degree, Mosna and Dolfini's GM1 and GM3 (Mosna and Dolfini, 1972) and Dubendorfer's D l line (Schneider and Blumenthal, 1978). Both the K c l and S2 lines, available from the American Tissue Culture Collection (Rockville, MD) and the European Collection of Animal Cell Cultures (Salisbury, UK) (Echalier, 1997), are the most widely used. The K c l line was derived from 8 to 12 h embryos of Drosophila melanogaster. The doubling time is 18 h at 25°C. The karyotype is diploid, with a female chomosomal set (XX), but with only one single 4 punctiform chomosome (Echalier, 1997). th  The most widespread use of Drosophila cell lines is as a system to assess the activities of exogenous genes. For example, the functions of mammalian and Drosophila  17  transcription factors have been reconstructed by cotransfections of transacting factor genes and target promoter-reporter constructs in Drosophila  cells (Krasnow et al., 1989)  (Yoshinaga and Yamamoto, 1991). Examples of use of K c l cells are: to analyze Drosophila  promoters using transient expression (Cherbas et al., 1991), and to express  Locust anti-diuretic hormone (Wang et al., 2000). Biochemically, K c l cells have been used to purify transcription factors (Parker and Topol, 1984), and silencing factors (Hodgson and Brock, 2001). Although transgenic flies are rich source of protein for purification, several disadvantages made me to choose transfected cell lines as a resource for purification of protein complexes. First, embryos are heterogenous with respect to cell types, and thus likely contain different P H complexes that would make purification of homogenous complex more difficult. Second, embryos contain more proteases than tissue culture cells, and this poses a serious problem during protein purification. Third, handling flasks of cells is more convenient than maintaining cages of flies. It is always faster and easier to obtain the same amount of nuclear extract from cultured cells than from collected embryos. Fourth, choosing a simpler system to answer questions is preferable in most, i f not, all scientific fields. Importantly, homeotic genes Ubx and abdominal A (abd-A) are silenced in K c l cells (Hodgson, personal communication). Therefore,in this thesis I use the K c l cells as a source for identifying proteins that associate with P H P in vivo.  The polyhomeotic (ph) Locus The ph locus was first described by Dura et al. (Dura et al., 1985), is X - l i n k e d and  18  maps cytologically to bands 2D2-3. Hypomorphic mutations in ph cause posterior transformations in embryos and adults similar to those of known dominant gain of function mutants in the A N T - C and B X - C (Dura et al., 1985; Dura et al., 1987). T w o mutagenic events are required to make null mutations of ph, which suggests that the locus is complex (Dura et al., 1987). Amorphic mutants die in mid-embryogenesis and completely lack ventral thoracic and abdominal epidermal derivatives (Dura et al., 1987; Smouse et al., 1988), exhibit abnormal patterns of homeotic and segmentation gene expression (Smouse et al., 1988; Dura and Ingham, 1988), and exhibit nervous system defects (Smouse et al., 1988). Genetic studies indicate that ph interacts with many other P c G loci including Asx, esc, E(Z), Pc, Pel, Psc, Scm, etc. (Dura et al., 1985; Cheng et al., 1994). Biochemical studies have shown that P H P interacts with P C , P S C , and S C M (Franke et al., 1992; K y b a and Brock, 1998a; Shao et al., 1999). Genomic sequencing indicates that the ph locus comprises two transcription units, which are similar at the molecular level (Deatrick et al., 1991). These transcription units have been termed polyhomeotic  proximal (ph ) and distal (ph/) (Dura et al., 1987). ph p  p  and ph appear to be functionally redundant because alleles with mutations in either d  transcription unit alone exhibit weak hypomorphic viable phenotypes (Dura et al., 1987). N u l l alleles are recovered only when lesions are generated in both transcription units (Dura et al., 1987). Homozygous distal (ph ), homozygous proximal (ph ), and double 401  heterozygous p h ^ ' / p h  409  409  mutants have similar phenotypes, suggesting that the proximal  and distal transcription units have similar functions. The ph locus is shown in F i g . 1.1 (DeCamillis et al., 1992; Hodgson et al., 1997).  19  0  5  1.0  15  20  5 (kb) Genomic DNA  Transcripts  Proximal  Distal  Fig. 1.1 Map of the ph locus. The thick line on the top represents the genomic DNA, and the number above the line indicates distance in kb which correspond to the coordinates of the ph sequence in Deatrick et al. (Deatrick et al., 1991). Below are the two transcripts as described in (DeCamillis et al., 1992) and (Hodgson et al., 1997). Exons are indicated in thicker lines and introns are indicated as finer lines. The arrows on the transcripts are the translation initiations.  20  Proteins Encoded by ph Proximal and Distal The two full-length P H proteins are very similar. The protein encoded by phf (PHP-170) has a 192 amino acid domain at its amino terminus that is not present in the protein encoded by ph (PHD), and a small region near the carboxy terminus (aa 1436d  1496) that exhibits only 42% identity compared to P H D . Otherwise, the proteins are 92% identical overall (Fig. 1.2) (Hodgson et al., 1997). T w o hypomorphic ph mutants, ph and 2  ph , deletions result from a P / M hybrid dysgenesis-induced rearrangement between 4  ph and ph to create a single new chimeric ph gene (Fig. 1.2) (Saget, 1994). p  d  ph is better characterized than ph (DeCamillis et al., 1992; Fauvarque and Dura, p  d  1993; Fauvarque et al., 1995; K y b a and Brock, 1998a; Shao et al., 1999). Hodgson et al. (Hodgson et al., 1997) found that in vitro translated ph? gave two products of 170 and 140 K D a (PHP-170 and PHP-140), detected on western blots with a common antibody that reacts with P H P and P H D .  Only PHP-170 was detected using an antibody directed  against the unique amino terminus of P H P .  The authors predicted that the P H P 140 band  detected after in vitro translation was due to internal translation initiation from M e t ^ (see Fig. 2.2 C ) . However, they detected only the 170 K D a product from a K c cell nuclear extract using the same common antibody used to detect the in vitro translated products. Therefore they concluded that K c cells use only the first methionine. In the same paper, they observed both PHP-170 and PHP-140 from embryonic nuclear extract, consistent with the results from Shao et al (Shao et al., 1999) who also detected these two bands using antibody against to a region common to both distal and proximal. ph encodes one protein of 135 K D a as observed after in vitro translation, i  consistent with the predicted molecular weight of conceptually translated protein.  21  PH-P  PH-D  PH'  proximal S A M domain  G ' rich n  |  | Ser/Thr rich  § Mammalian Homology | Zinc Finger  Fig. 1.2 Sequence motifs shared by P H P , P H D and mutant P H and P H (modified from Saget et al. (Saget, 1994)). Proximal specific domains are marked by diagonal stripes. Poly-glutamine stretches are marked by diagonal grid. Mamalian homolog regions are marked by horizontal bars. The S A M motifs (self-association motif) are marked by a checkerboard pattern. Serine and theonine rich regions are marked by stippled boxes. The black boxes indicate the location of zinc fingers. 2  22  4  Several smaller bands are observed from embryo extracts (Hodgson et al., 1997). The smaller proteins remain to be characterized, and could represent cleaved proteins, or products of alternatively spliced transcripts.  Polycomb Polycomb (Pc) is the founding member of Polycomb group gene and the best characterized among the P c G . Its c D N A encodes a 390 amino acid polypeptide that shows a high content of charged amino acids (20% basic and 15% acidic amino acids) (Paro and Hogness, 1991). P C migrates in S D S - P A G E as a band of apparently 64 K D a , although its calculated M is about 44 K D a . The chomodomain was identified as a r  protein motif homologous between P C and H P 1 , a heterochomatin-associated protein encoded by the suppressor of position effect variegation gene Su(var)205 (Paro and Hogness, 1991). Mutated P C proteins expressed as PC-beta-galactosidase fusion proteins showed that carboxy-terminal truncations of the Pc protein do not affect chomosomal binding of the fusion protein. However, mutations affecting only the chomo domain including in vitro generated deletions, as well as point mutations, abolish chomosomal binding (Messmer et al., 1992). Rapidly dividing nuclei were found to display a rather homogeneous P C - G F P (Green Fluorescent Protein) in flies. However, with increasing differentiation a pronounced subnuclear pattern was observed. In all investigated diploid somatic tissues the bulk of P C - G F P fusion protein is depleted from the chomosomes during mitosis: however, a detectable fraction remains associated (Dietzel et al., 1999).  23  P C interacts with nucleosomal core particles in vitro, and the main nucleosomebinding domain coincides with a region in the C-terminal part of P C previously identified as the repression domain (Breiling et al., 1999). P C has also been shown in a large protein complex (2-5 x 10 Da) with P H and other unknown proteins. During 6  embryogenesis the P C and P H show the same spatial distribution. In addition, by doubleimmunofluorescence labeling it was demonstrated that P C and P H have exactly the same binding patterns on polytene chomosomes of larval salivary glands (Franke et al., 1992). Therefore it might be expected that the proteins that associate with P H and P C in vivo should be similar. M u c h can be learned about the function of proteins by determining the identities of proteins that associate with the protein of interest in vivo. This is especially important in the case of P c G proteins, because we do not know the mechanism of PcG-mediated silencing. Using epitope tagging followed by immunopurification, I purified P H P and its associated proteins from Drosophila K c l cell nuclear extracts. I subsequently identified the PHP-associated proteins using mass spectrometry sequencing and western blotting analysis. I showed that molecular chaperones are associated with P H and that a mutation in chaperone heat shock 70 cognate 4 (Hsc70.4) enhances the extra sex combs phenotype of polyhomeotic  (ph) and Polycomb (Pc). I demonstrated that the histone deacetylase  Rpd3 and histone binding protein p55 are associated with P H , and that the Rpd3 mutation enhances the extra sex combs phenotype of ph and Pc. Surprisingly, histone deacetylase activity was not detected in immunopurified P H . I showed using western blotting analysis that T B P and its associated factors T A F 4 2 and T A F 8 5 are also associated with n  24  n  P H P . P H P and P C were coimmunoprecipitated by anti-TBP antibody. In addition, Tbp mutants enhance the extra sex combs phenotype of ph but not Pc.  25  Chapter II Expression of Epitope-tagged Polyhomeotic Proximal (PHP) in Drosophila Kcl Cells INTRODUCTION Transfection of Cell Lines The introduction of exogenous D N A into eukaryotic cell lines (or transfection) and the subsequent analysis of its expression and the expressed product have proved a very efficient method for studying gene regulation and the function of product m R N A s and proteins.  In transient transfection, D N A is introduced into cells without using drug  selection and samples are analysed within several days. In this way, the transformed D N A remains extrachomosomal. This method is fast, and is a convenient method for the rapid production of small quantities of protein for initial characterization. In addition, the method lends itself to the rapid testing of vector functionality as well as optimization of different combinations of promoters and other elements in expression vectors (reviewed in Makrides, 1999). Thus, results from transient expression assays allow one to proceed with confidence to the more time-consuming task of preparing permanent cell lines. One drawback of transient transformation is that the transfected D N A only exists for a couple of days before it is degraded, necessitating performing the analysis within short period of time. The other drawback is that the variation in expression levels between experiments makes the interpretation of results more difficult, requiring analysis of multiple samples. In contrast to transient gene expression, stable cell lines that permanently express the gene of interest depend on the stable integration of plasmid into the host cell chomosome. It is also possible, however, to generate stable cell lines that harbor vectors extrachomosomally (Yates et al., 1985). This method is particularly useful when large  26  quantities of protein are required for biochemical analysis (Burke and Kadonaga, 1996; Sif et al., 2001) and therapeutic use (Fusseneger et al., 1999; Pfeifer, 1998). Another advantage of stable cell lines is that the expression levels of transgenes are reproducible. The generation of stable cell lines necessitates the screening of large numbers of transfected cells. This is mainly due to the wide variation in the level of expression of the transfected gene, depending on the chomosomal site of plasmid integration (Wahl et al., 1984; Dobie et al., 1996; Lambert and Nordeen, 1998), and the copy number of integrated gene (Pikaart et al., 1998). Therefore it is cumbersome and time-consuming to generate a stable cell line. In this chapter, I describe the production of stable cell lines producing epitope-tagged PHP and PC. There are many methods to transfer expression vectors into cells, most of them initially devised for mammalian cells. The efficiency of transfection for naked D N A is quite low, and this problem can be overcome by using transfection vehicles. These vehicles can be classified as chemical (phosphate precipitation, liposomes, D E A E dextran, polybrene) and physical (electroporation) (Cherbas et al., 1994; Colosimo et al., 2000). C a P 0 coprecipitation, electroporation , and liposomes are the most common 4  choices. Calcium phosphate-DNA coprecipitation, adapted with minor modification from original mammalian cell transformation procedure (Wigler et al., 1979), has been used for a wide variety of Drosophila cell lines (Cherbas et al., 1994). A key factor affecting the efficiency of transformation appears to be the pH of the phosphate buffer (reviewed in Colosimo et al., 2000; Chen and Okayama, 1987; Wigler et al., 1979; Cherbas et al., 1994). It is simple and inexpensive, requires no special equipment, and works for most  27  cell lines. The major drawbacks of this procedure are that it requires high concentration of D N A (20 P-g/ml), is cytotoxic, and sublines differ enormously in their susceptibility to transfection (Cherbas et al., 1994). Electroporation is based on exposing cells to a pulsed electrical field of sufficient strength to cause a reversible permeabiliztion of the cell membrane in localized areas (Andreason and Evans, 1988; Shigekawa and Dower, 1988). The conditions for mammalian cells involve either high field strength coupled to a low capacitance or low field strength coupled to a high capacitance (reviewed in Colosimo et al., 2000). Electroporation is probably useful for all Drosophila lines, but parameters for its use have been optimized only for Drosophila Kcl67 cells (Cherbas et al., 1994). Liposome-mediated transfection (Feigner et al., 1987) is the most common chemical method and is based on the interaction of negatively charged D N A with positively charged artificial membrane vesicles. The vesicles either fuse with the cell membrane and deliver their contents into the cell (Matsui et al., 1997), or enter the cell though an endocytic pathway (Zabner et al., 1995). Liposome-mediated gene transfer offers several advantages over other chemical transfection systems. Among these are a relatively high efficiency (sometimes 100-fold more than the classical C a P 0 method) in 4  a variety of cell types, including Drosophila cell lines (Kc, S2) (Malone et al., 1989), the ability to transfect various cell types resistant to C a P 0 , successful delivery of a wide of 4  range of D N A (Lee and Huang, 1997), effective delivery of both R N A (Malone et al., 1989) and protein (Debs et al., 1990), and requirement for less D N A . Disadvantages of this method include the cytotoxicity of the liposome formulation, and the need to optimize the DNA-to-liposome charge ratio, the amount of D N A , cell density and the  28  transfection period (Gao and Huang, 1995) for each type of liposome. C e l l F E C T I N ™ , a liposome formulation of the cationic lipids ( T M - T P S and D O P E ) developed for transfection of Sf9 cells by G i b c o B R L , has been used successfully on different insect cell lines (Pfeifer et al., 1999). K c l cells were transfected successfully using C e l l F E C T I N ™ to express biologically active peptide (Wang et al., 2000).  Expression Systems The choice of an expression system for production of recombinant proteins depends on many factors, including cell growth characteristics, expression levels, intracellular and extracellular expression, posttranslational modifications and biological activity of the protein of interest (Marino, 1989; Goeddel, 1990; W u r m and Bernard, 1999). The essential elements of eukaryotic expression vectors include: (1) a constitutive or inducible promoter; (2) optimized m R N A processing and translational signals that include a K o z a k sequence, translational termination codon, and polyadenylation signal; (3) a transcription terminator; and (4) selection markers for the preparation of stable cell lines and vector propagation in bacteria (reviewed in Makrides, 1999). Depending on the promoter used in the expression vector, the expression can be strong or weak; and can be constitutive or inducible. Most current applications make use of cell lines with stably integrated genes under the control of a strong cellular or viral promoter (Kaufman, 1990; Pfeifer, 1998; Colosimo et al., 2000). Strong promoters are preferred because they produce more protein which is the goal for most applications. For example, the cytomegalovirus promoter, the most frequently used mammalian promoter, has been used in different cell lines for high-level protein expression (Makrides, 1999).  29  The other commonly used strong constitutive promoters are S V 4 0 and Rous Sarcoma virus promoters, and the murine 3-phosphoglycerate kinase promoter (reviewed in Makrides, 1999). Although strong constitutive promoters have the advantage of highlevel expression, high level expression may cause problems. For example, toxic proteins kill cells, and it is impossible to express growth inhibitory proteins, like components of the apoptotic cascade (Blau and Rossi, 1999). The problems of constitutive strong promoters can be overcome by inducible promoters. Commonly used inducible promoters are: the metallothionein promoter, the glucocorticoid response promoter, and tetracycline repressor (tetraR) controlled system (reviewed in Makrides, 1999). Inducible promoters are desirable for the production of . proteins that may be toxic to the host cell, such as diptheria toxin (Angrand et al., 1998), for the study of gene regulation during development in transgenic animals (Saez et al., 1997), and for experimental and therapeutic applications of gene transfer (Dhawan et al., 1995). The drawbacks of the inducible promoters are that the inducer might be cytotoxic or the promoter might be leaky (Cherbas et al., 1994).  Stably-transfected Cell Lines versus Baculovirus-infected Expression Systems The advantages and ease of expressing foreign genes at high levels in insect cell when recombinant baculovirus vectors are used has led to their wide use as a eukaryotic expression system (O'Reilly et al., 1992;.King and Possee, 1992). In baculovirus expression vectors ( B E V s ) , foreign proteins are expressed from promoters derived from the Autographa  californica  nuclear polyhedrosis virus ( A c M N P V ) which are strongly  activated during the very late phase of infection. The biggest advantage of the B E V s is  30  that proteins are usually expressed to very high levels (up to 200 mg/L of infected cells) (McCarroll and King, 1997). However, virus infection severely compromises the insect cell secretory pathway and this often results in relatively lower yields of proteins destined for the plasma membrane or for secretion (King and Possee, 1992). Infection also damages the insect cell's protein modification machinery and thus leads to reduction in proper posttranslational modifications of recombinant proteins (Lu et al., 1997). Second, during the late stages of infection, the baculovirus-infected cells are generally unable to process complex, intron-containing transcripts and produce recombinant proteins from genomic sequences. Third, it is relatively difficult to purify recombinant proteins as a result of cell lysis at the end of the infection process, and because release of proteases causes degradation of the overexpressed products (Lu et al., 1997). Last, the host range for the baculovirus is narrow leading to its lack of versatility with different cell lines (Hegedus et al., 1999). Attempts to overcome some of these problems have led to the development of a stable, plasmid based expression system (reviewed in McCarroll and King, 1997; Pfeifer, 1998). One obvious advantage of this system is that one can choose different cell lines according to different requirements. For example, proteins usually perform their functions with other proteins in protein complexes. To investigate protein function in its native protein complex, it is better to use the cell lines from the organism of interest. Many insect cell lines have been exploited for expression of heterologous genes. These include Spodoptera cell lines Sf9 (Joyce et al., 1993; Henderson et al., 1995; Jarvis et al., 1996; Pfeifer et al., 1997; Hegedus et al., 1998; Hegedus et al., 1999) and Sf21 (Lu et al.,  31  1997; McLachlin and Miller, 1997), Drosophila cell lines SL2 and Kc (Johansen et al., 1989; Buckingham et al., 1996; Segal et al., 1996; Pfeifer et al., 1997; Hegedus et al., 1998), and mosquito cells (Miller et al., 1987; Monroe et al., 1992; Lycett and Crampton, 1993).  Epitope Tagging Epitope tagging, first described by Munro and Pelham in 1984 (Munro and Pelham, 1984), is the process of fusing a set of amino acid residues that are recognized as an antigenic determinant to a protein of interest. An epitope-tagged protein is a special kind of fusion protein in which the added amino acids are few in number (typically 6 to 30) and do not add any new biological activity to the protein. These features make it particularly likely that a protein carrying the tag will retain, to a large degree, normal structure and function (reviewed in Jarvik and Telmer, 1998; Hickman et al., 2000).  Advantages and Limitations of EpitopeTagging The epitope tagging approach offers significant advantages over the use of antibodies generated directly against the protein of interest. First, the most obvious advantage of epitope tagging is that the time and expense associated with generating and characterizing antibodies against multiple proteins are obviated. Second, because the tag would be missing from extracts of cells that are not expressing a tagged protein, negative controls are unequivocal. Third, epitope tagging allows the protein to be purified in the absence of a functional assay. Fourth, the tagged protein can be monitored with wellcharacterized monoclonal antibody whose immunoprecipitation capabilities and spectrum  32  of cross-reactivity with non-tagged proteins are already known. Fifth, different but identically tagged proteins can be immunoprecipitated and affinity purified under the same conditions. Finally, because the experimenter has a choice for the tag insertion site in a protein, a site can be selected that is not likely to result in antibody interference with functional sites in the molecule, for example, sites that might be involved in proteinprotein interactions. There are limitations of the epitope tagging. A cloned and characterized gene or c D N A must be available; the epitope tag may interfere with the tagged protein structure and/or function (Brault et al., 1999); epitope-tagged genes are expressed at abnormal levels due to the use of heterologous promoters; and the epitope-tagged gene must be introduced into cell, tissue, or organism of interest.  Epitopes and Antibodies Most epitopes that have been popular for epitope tagging are highly charged. For example, the 11-amino acid H A tag contains two aspartic acids; the 10-amino acid c - M y c tag contains thee glutamic acids, one aspartic acid, and one lysine; and the 8-amino acid F L A G tag contains five aspartic acids and two lysines. Since one generally aims to place the tag on the external portion of the target protein, it is advantageous that the tag be charged rather than hydrophobic. Many validated epitope-antibody combinations are commercially available for use (BAbco, Santa Cruz Biotech., Sigma). The most commonly used epitopes include F L A G (Prickett et al., 1989), H A (Field et al., 1988), c - M y c (Munro and Pelham, 1987), and 6His. N o epitope tag is ideal for all applications. For example, only one antibody (9E10)  33  is available for c - M y c epitope. The advantage of c - M y c tag is that the hybridoma line expressing the 9E10 monoclonal obtainable from the A T C C for use in large scale projects. The disadvantages of c - M y c are that the 9E10 monoclonal may not immunoprecipitate reliably, and the endogenous c - M y c may interfere (reviewed in Fritze and Anderson, 2000). Three highly specific antibodies for H A , including monoclonal and polyclonal, are available. Most of the available antibodies recognize the epitope whether it is internal to the protein or at the termini, but in some cases recognition is dependent upon location. The advantages of the F L A G tag are that the epitope can easily cleaved off the tagged protein after purification (Prickett et al., 1989), and three good antibodies ( M l , M 2 , M 5 ) are available. The advantage of 6-His tag is that the tagged protein can be purified on N i  2 +  affinity matrix. The disadvantage of His tag is that there  is no good antibody available for immunodetection.  Locating the Tag In the majority of cases described in the literature, the tag has been placed at the extreme N or C terminus of the target protein. There are a number of reasons behind this choice of location.  One reason is historical. The first proteins to be tagged were tagged  at the termini, and when new proteins were tagged it seemed wise to do what had worked in the past.  A second reason is that it is simpler and more practical to tag using  expression vectors that automatically put the tag at a terminus because it is straightforward to engineer the correct reading frame for the fusion. Termini are frequently chosen in the belief that proteins w i l l tolerate additions more readily at these locations than at other sites, because termini are rarely included in active sites. The  34  termini also appear favorable because they are likely to be on the outside of the folded polypeptide, and therefore are more likely to be accessible to the antibody. It is impossible to give a quantitative estimate of the likelihood that a teminal tag w i l l be successful, because successful tagging events are more likely to be published than unsuccessful ones. However, a survey of the literature did allow Jarvik and Telmer (Jarvik and Telmer, 1998) to estimate the success rate for internal tagging events. The data gathered showed that the tag was well tolerated in about 63% of all insertions in terms of activity assay.  Applications of Epitope Tagging Since its first use by Munro and Pelham (Munro and Pelham, 1984), the epitope tagging strategy has been used to address a wide variety of questions. A s a testimony to that fact, epitope tagging was employed in some 30% (20 of 64) of the articles in a recent volume of the journal Cell (Volume 98, 1999) (calculated by Fritze and Anderson, 2000). Numerous studies have taken advantage of epitope tagging to determine, or confirm, the subcellular location of a gene product. Immunofluorescence microscopy is the most prevalent means for doing this (Goodson et al., 1996; X u et al., 1998). Visualization of the tag can reveal which parts of the integral membrane proteins, including receptors and channels, are intracellular or extracellular (Anand et al., 1993; Canfield et al., 1996). Epitope tagging has been used to study the movements of proteins though cellular compartments. G o l g i trafficking of G D 3 synthase (Martina et al., 1998) and P K C epsilon (Lehel et al., 1996) have been monitored. In a number of studies, coimmunoprecipitation of a protein complex with anti-tag antibodies has been used to  35  identify and analyze protein-protein interactions (Dear et al., 1997; Keys et al., 1996; Mende et al., 1995). For functional analysis , the epitope-tagged gene is introduced into cells or organisms with null or temperature-sensitive mutations in the gene of interest, and has been effectively applied in many studies (Chen et al., 1993; Handley-Gearhart et al., 1994; Jarvik et al., 1996). A n important feature of epitope tagging is that it provides a specific means to purify the tagged protein. Epitope tagging has been used successfully to purify many different protein complexes, including the Polycomb group P R C 1 complex from Drosophila et al., 1999), T B P containing T F D complex from Drosophila n  (Shao  (Burke and Kadonaga,  1996), S W I / S N F chomatin remodeling complex from Saccharomyces cerevisiae (Peterson et al., 1994), S W I / S N F from Hela cells (Sif et al., 1998), mSin3 containing B R G 1 complexes from Hela cells (Sif et al., 2001), and Drosophila B R M complex (Papoulas et al., 1998). A common problem in heterologous gene expression is proteolytic degradation. Using an affinity fusion strategy, eventual degradation products of the target protein are copurified with the full-length fusion protein. However, it has been observed that dual affinity approaches can have a stabilizing effect on several proteplytically sensitive proteins compared to single fusions. This dual affinity was achieved by fusing two different tags at each end of the target protein, allowing for two successive affinity purification to obtain full length tagged protein (Murby et al., 1991). Another new application of double tagging is to tag two different tags separated by a protease cleavage site to one end of target protein (Rigaut et al., 1999). The target protein can be purified by two consecutive immunoaffinity chomatography steps. The adavantage of this  36  strategy is that it employs two steps of high specific and efficient affinity, and avoids the use of low efficiency conventional chomatography, to achieve good quality of purification from low or natural level of expression of the tagged protein. In this chapter, I describe my successful attempts to express epitope-tagged P H P and P C in Drosophila  K c l cells, to provide the key reagents for attempts to identify  proteins associated with these P c G proteins in vivo.  RESULTS AND DISCUSSION Expression Constructs for Stable Cell Transformation A s noted in Chapter 1, ph is better characterized than ph , and the structures of ph p  d  and ph are highly conserved (Deatrick et al., 1991; Dura et al., 1987; Hodgson et al., d  1997; Saget, 1994). I chose to tag ph? as the first step toward identifying proteins that associate with P H in vivo. Because there is no assay for P H and its associated proteins, and to exploit the high specificity and purification of immunoaffinity chomatography, I tagged the ph? c D N A gene with one F L A G at its N-terminus and a double H A at its C-terminus (Fig 2.1). I reasoned that having two tags would increase the likelihood that at least one tag would be functional. M y concern was that i f P H P interacts with multiple proteins, then the epitope tag might be sterically hindered by one of the other subunits. Moreover, double tagging might allow me to purify P H P and its associated proteins using a double immunoaffinity approach, which has been demonstrated to give high purity and good yields in other systems (Hammarberg et al., 1989; Rigaut et al., 1999). A potential  37  p  :pnI/HincII 0.60  HincII/Xbal 6.00  FLAG: ATG GAC TAC AAA GAC GAT GAC GAT AAA Met Asp Tyr Lys Asp Asp Asp Asp Lys HA: TAC CCA TAC GAT GTT CCG GAT TAC GCT AGC CTC Tyr Pro Tyr Asp Val Pro Asp Tyr Ala Ser Leu  Fig. 2.1 M a p of p N T P 1 4 F - P H P - H A . (A) construct of epitope tagged phf for cell transfection. The direction of transcription is shown by the arrows. M t n , the metallothionein promoter. M t n poly A , the poly A signal from metallothionein gene of Drosophila. phf c D N A is indicated as ph. The H A H A stands for the tandem tag of H A , and the Flag stands for the F L A G epitope. (B) shows the nucleotide and amino acid sequences of epitopes of F L A G and H A .  38  drawback of this approach is that because there is always a possibility that the epitope tag w i l l interfere with the normal function of the tagged protein, having two tags increases this likelihood. The reasoning for using a tandem H A tag is discussed below. A s reviewed above, F L A G and H A are the most commonly used tags because their antibodies are good and available at reasonable cost. In preliminary experiments, I used the constitutive expression vector developed by Hegedus et al. (Hegedus et al., 1998). I found that the expression level of tagged P H drops with the passage of the cell line, and reached undetectable levels after about a month as assayed by western blotting. This might be caused by the toxicity of overexpressed P H or the transgene could be silenced. T o avoid these problems, the inducible metallothionein  (mtn) promoter (Bunch et al., 1988; Johansen et al., 1989;  Maroni et al., 1986; Otto et al., 1987) was chosen to drive the expression of tagged ph. Another advantage of the metallothionein promoter is that it is possible to control the expression level by using different levels of metal ions (Johansen et al., 1989; Hegedus et al., 1998). To generate Mtn driven double tagged ph?, I obtained the Drosophila  mtn promoter  and polyadenylation signal containing vector called p N T P 1 4 from L o y V o l k m a n ( U C Berkeley). A s described in the Materials and Methods, a c D N A encoding PHP-170 was tagged with double H A at its C-terminus with the addition of X b a l after the stop codon and with F L A G tag at its N-terminus with the addition of an A T G codon, K o z a k sequence, X h o l site. This double tagged P H P c D N A was then inserted into the H i n d i site of p N T P 1 4 vector downstream of the mtn promoter to generate the construct shown  39  in Fig. 2.1, which I named p N T P 1 4 F - P H P - H A . The engineered c D N A was sequenced and both tags are in frame. Before beginning the arduous and time-consuming task of selecting stable cell lines, I carried out preliminary experiments in transiently transfected K c l cells to ensure that the expression vector produced protein of the expected size, and that the tags were in frame. A s reported in Hodgson et al. (Hodgson et al., 1997), ph? uses two methionines (metl and met 244) as translation initiation codons in embryos and in an in vitro translation system. If this is also true in K c l cells, the F L A G antibody should only recognize PHP-170; and the H A antibody should detect PHP-170 and PHP-140. K c l cells were transiently transfected with the construct illustrated in Fig. 2.1. Expression was induced with increasing concentrations of copper ions for two days. After two days, a whole cell extract was prepared, and a sample was separated by electrophoresis using S D S - P A G E , and transferred to nitrocellulose as described in the Materials and Methods. Fig. 2.2 shows a western blot from whole cell K c l extract that has been probed with an anti-HA antibody (panel A ) , then stripped and reprobed with a n t i - F L A G antibody (panel B).  A s expected, the F L A G antibody detected only PHP-170, whereas the H A tag  antibody detected both PHP-170 and PHP-140. This result is consistent with the embryo results reported previously. This result therefore indicates that K c l cells use two initiation codons, and confirms that PHP-140 is the product of internal translation initiation from the ph tanscript. The low level expression from the non-induced lane p  seen in Panel A is probably because Zeocin is provided in the form of C u  40  2 +  salt, which  will itself cause induction. These experiments confirm that the expression vector is functional, that the inducible promoter works well in K c l cells, and that both tags are in frame. A s noted above, P C colocalizes with P H on polytene chomosome, and both proteins are present in one large complex (Franke et al., 1992). In addition, P H P and P C were copurified in P R C 1 complex (Shao et al., 1999). T o see whether P C containing complex is different from P H P containing complex in K c l cells, I also tagged Pc c D N A with F L A G epitope at its N-terminus with the same modifications as that used for P H P described above. The tagged D N A was then inserted into H i n d i site of p N T P 1 4 to generate the construct shown in Fig. 2.3, which I named p N T P 1 4 F - P C . The engineered D N A was sequenced, and the F L A G is in frame, and lacks mutations.  Selection of Stable Cell Lines Expressing Tagged PHP (F-PHP-HA) I chose Drosophila K c l cells as a starting point to obtain a stable, transfected cell line to express tagged P H P .  These cells offer important advantages. First, they are  robust and grow easily at room temperature, so no incubators are required. Second, they have no serum requirement, and are therefore cheap to grow. Third, the cells can be grown to high density ( 1 . 2 x l 0 cells/ml), making it relatively convenient to obtain large 7  numbers o f cells as source o f nuclear extracts. Fourth, they have been extensively used in our lab, and have been well-characterized i n previous studies (Pfeifer et al., 1999; Hodgson and Brock, 2001; Wang et al., 2000). The expression vector lacks drug selection marker for eukaryotic cells, therefore pZop2F that contains Zeocin™ gene was cotransfected with the expression vector.  41  Cu (uM):  o o  2+  0  o o o  o  200-  o o >n  © o  PHP170 PHP140  116  97H  B Anti-HA  Anti-FLAG  PHP FLAG  Vlet244  HAHA  Fig. 2.2 Internal initiation of phf translation. (A) western blotting analysis of transiently transfected K c l whole cell extract with H A antibody. Thee 30 m m diameter plates each containing 3 x l 0 K c l cells were transiently transfected with plasmid p N T P 1 4 F - P H P - H A , without copper added (0) or with copper ion inductions (500pM or 1000p:M,respectively) for two days post-transfection. Then 10 cells of each well were lysed using 2 x S D S sample buffer, and run on 6% S D S - P A G E . The lower molecular weight bands may be degradation products. (B) the same blot as ( A ) was stripped and re-probed with antiF L A G M 2 (Sigma). (C) sketch diagram of phf tagged with F L A G and H A epitopes. The F L A G epitope is marked as a zig-zag filled box, H A epitopes are designated as dotted boxes. The two arrows indicate the two translation initiation sites. 6  6  42  ).60  B  FLAG: ATG GAC TAC AAA GAC GAT GAC GAT AAA Met Asp Tyr Lys Asp Asp Asp Asp Lys  Fig. 2.3 M a p of p N T P 1 4 F - P C . (A) the construct of epitope tagged Pc used for cell transfection. The direction of transcription is shown by the arrows. M t n , the metallothionein promoter. M t n poly A , the poly A signal from metallothionein gene of Drosophila. Pc c D N A is indicated as Pc. The Flag stands for the F L A G epitope. (B) the nucleotide and amino acid sequences of epitopes of F L A G .  43  Zeocin is a member of the phleomycin family of antibiotics isolated from Streptomyces verticillus.  The toxicity of the phleomycin/bleomycin antibiotics results from their ability  to bind D N A followed by the generation of lethal double-strand breaks (Pfeifer et al., 1997). Resistance to these antibiotics is conferred by a 13.6 k D a protein that binds the antibiotic in a stoichiometric manner thereby inhibiting the antibiotic from binding to the cellular D N A and thus preventing strand cleavage. Therefore one advantage of Zeocin™ compared to other commonly used eukaryotic selection markers is that Zeocin exhibits toxicity towards a broad range of prokaryotic and eukaryotic organisms. This feature means that there are fewer concerns about bacterial contamination during long term selection of cell lines. The other benefit is the lower concentration of drug required for selection resulting in reduced cost. Different cell lines are sensitive to different amount of drug and different media may also affect the sensitivity. T o estimate the minimal concentration of Zeocin™ required to inhibit K c l cell growth, I incubated same density of cells with different concentrations of Zeocin , and counted the cell density every 24 h. Results are shown in Fig. 2.4. Twenty five p,g/ml Zeocin achieved maximal killing within a week. Although 50 and 100 Hg/ml Zeocin k i l l the cells more quickly, to lower the cost, I used 25 p.g/ml Zeocin for selection of stable cell lines. To select single clones of cells, it is necessary to dilute the transformed cells and then culture them in the presence of selection. However, no medium can support the growth of Drosophila cells at very low cell densities (Cherbas et al., 1994). Two ways have been reported to to overcome this problem. One is to use conditioned medium (Ashburner, 1989), and the other is to use feeder cells. The latter gives more  44  reproducible results and is more convenient to perform (Cherbas et al., 1994). Conventionally, X-ray or y-irradiated cells were used as feeder cells (Ashburner, 1989), but I found that non-irradiated cells work well. A t the low concentration of drug used, the non-transformed cells survived long enough to condition the medium, and to allow the transformed cells to achieve sufficiently high densities to survive. I used 100 u.1 of non-transformed K c l cells at l - 2 x l 0 / m l per well of a 96 well plate as feeder cells, and 5  added transformed cells at 100 to 1000 cells per well of 96-well plates. After about two weeks of selection at 25 (Xg/ml Zeocin in 96-well plates sealed with Parafilm to avoid drying, sixty-four single colony wells were obtained from 5 plates. These colonies of cells were transferred into 24-well plates, grown to 100% confluence, and then transferred to 6-well plates and subsequently to T-25 flasks. After testing for taggedPHP expression as described in the next section, two clones were expanded and stored in medium containingl0% D M S O at 5 x l 0 to 10 cells/ml at 7  8  - 7 0 ° C . T o maintain the stock, the frozen cells need to be revived, expanded and refrozen every year.  Characterization of Tagged PHP Expression Because it was not known if all cells expressed P H , or i f expression levels of F P H P - H A varied among clones, or i f leaky expression of ph? from the Mtn promoter was toxic to cells, it was necessary to test each clone for expression. About 10 cells from 6  each clone were heated with 20 u.1 of 2 x S D S sample buffer. Each sample was separated by electrophoresis on S D S - P A G E , and analyzed for expression of F - P H P - H A using antiH A antibody on western blots. A n example of this analysis is shown in F i g . 2.5. It can  45  -  Fig. 2.4  T i m e (day)  D i a g r a m o f c e l l g r o w t h under different concentrations o f Z e o c i n . T h e X - a x i s  stands for the n u m b e r o f days w h e n c e l l density was c o u n t e d . T h e Y - a x i s stands f o r the c e l l density at different days. T h e d i a g r a m was d r a w n u s i n g p r o g r a m C A - C r i c k e t .  46  be seen that one of the 11 clones tested expressed tagged P H P . This clone is called F P H P - H A # 1 5 . Similar analyses identified one other clone expressing F - P H P - H A , called F-PHP-HA#30. Therefore, I recovered two stable cell lines out of 64 tested. It is likely that the remaining clones had stably integrated Zeocin™ resistance, but did not have stably integrated F-PHP-HA, or had lost F-PHP-HA. Because clone #15 gave a cleaved tagged product when probed with H A antibody on Western blot, all the subsequent work was based on clone #30 (compare #15 to #30 in Fig. 2.6). To check whether F - P H P - H A made in transfected cells is similarly sized compared to endogenous P H P , western blots were used to analyze non-transformed and transformed nuclear extract using antibodies to endogenous P H and H A antibodies respectively. A s shown in Fig.2.7, both transfected and non-transfected cells produce PHP-170 and P H P 140, detected with P H antibody (lanes 1 and 2). For comparison, a parallel blot of nuclear extract from cells expressing F - P H P - H A was probed with a n t i - H A antibody (lane 3). The P H P proteins in extracts from transfected cells migrate slightly more slowly than the endogenous proteins. This difference in mobility is most likely caused by the increase in molecular weight caused by the tags themselves. The molecular weight of the tags is about 3.4 kDa. Therefore, i f the molecular weight of P H P - 1 7 0 is 170 k D a , and the molecular weight of F - P H P - H A - 1 7 0 would be 173.4 kDa, and this difference should be detectable by S D S - P A G E . The other obvious difference between the native and tagged P H P proteins is the ratio of PHP-170 to PHP-140. The endogenous PHP-140 is more abundant than PHP-170, whereas F - P H P - H A - 1 7 0 is more abundant than F - P H P - H A - 1 4 0 . This result suggests that the presence of the amino-terminal F L A G epitope changes the  47  Clone #:  15  22  28  29  31  32  34  36  44  47  54  200—  116  97  -  Fig. 2.5 Western blotting analysis of selected clones resistant to Zeocin. The numbers on the top are the clone numbers. The molecular weight standards are indicated in K D a to the left. About 10 cells were lysed by 2xSDS buffer and run on SDS-PAGE. The western blot was probed with H A antibody. 6  48  #15  #M  -PHP170 -PHP140  116—  97— •^Cleaved  66-  Fig. 2.6 Western blotting analysis of the two stably transformed clones expressing tagged PHP. About 10 cells of clone #15 and #30 were induced with 100 uM copper ion for 2 days, heated with 2xSDS buffer, and loaded onto SDS-PAGE. The western blot was probed with H A antibody. The molecular marker is indicated to the left in KDa. PHP170 and PHP140 bands are marked to the right. The cleaved product in clone #15 is indicated with an arrowhead to the right. 6  49  Abs'  1  PH  i  2  '  "  NE:  Native  HA 3 M i l  * Transfected  PHP 170 PHP140  Fig. 2.7 Comparison of endogenous and tagged P H P . Nuclear extracts (20u.g) prepared as described in the Materials and Methods were analyzed on S D S - P A G E , and the samples loaded are indicated at the bottom, non-transfected K c l (lanel) and clone #30 (lanes 2 and 3). Antibodies used to probe the western blot are indicated on the top, antiP H (lanes 1 and 2) and anti-HA (lane 3). PHP170 and PHP140 bands are indicated to the right.  50  relative initiation frequency at the first A U G codon used for PHP-170 relative to the internal A U G used for PHP-140. I do not know the effect that this change w i l l have on association of P H P with other proteins. Although copper ion is less toxic than cadmium to cell lines (Bunch et al., 1988), to avoid unknown side effects of copper on cells, I titrated the minimum copper ion concentration required for full induction on F - P H P - H A # 3 0 . A s shown in lanes 1 to 6 of Fig. 2.8, a western blot on the whole cell lysate of four months old #30 clone selected with 100 p:g/ml Zeocin, 100 p,M copper ion is enough to give full induction (lane 3). The low level expression from the non-induced lane is because Zeocin is provided in the form of C u  2 +  salt. A t 100 p.g/ml concentration, the corresponding C u  2 +  concentration is 65 p,M.  A t this concentration of copper ion, there is no obvious effects on the doubling time as reported (Echalier, 1997), which is about 24 h at 25°C (Fig. 2.9). Therefore, 100 \iM Cu was used for all the subsequent induction of expression for nuclear extract 2+  preparation, as described in Material and Methods. To check the stability of expression of clone #30, and to determine i f continuous selection was required to maintain expression of the tagged gene, I tested expression of F P H P - H A in #30 at different time points with and without Zeocin in the medium. F i g . 2.8 shows a western blot of these samples probed with anti-HA antibody. Lane 1 to 6 are the whole cell lysates from cells after 4 months selection with Zeocin and then induced with copper ion for one day. It can be seen that the transgene is stable for at least four months under selection. T o test i f continuous selection was required to maintain expression, the cells from the 4-month selection were then maintained in SF-900 medium as described in the Material and Methods with a low level of Zeocin (5p:g/ml) (lane 7) or without Zeocin  51  for another year. The non-selected cells were then split into two flasks, one induced with 100 p M C u  2 +  for one day (lane 9), and one without induction (lane 8). It can be seen that  Zeocin is not required for maintaining expression (lane 9) for at least one year. T o avoid bacterial contamination, 10 rUg/ml Zeocin was used regularly to maintain the transformed cell line. A tagged ectopic expression strategy has been used successfully for the purification of many protein complexes, including the Drosophila Polycomb complex P R C 1 (Shao et al., 1999), basic transcription factor T F I I D (Burke and Kadonaga, 1996), the chomatin remodelling complex S W I / S N F (Sif et al., 1998), and the Drosophila B R M complex (Papoulas et al., 1998). On one hand, high level expression is preferable because more tagged product w i l l incorporate into endogenous complex, allowing higher recovery. O n the other hand, more product might lead to promiscous interactions and thus form non-native complex(es) that might be co-purified with the native complex(es). For example, tagged E S C in flies gives an extra complex that is smaller than the endogenous complex (< 600 K D a ) and tagged E S C monomers, without significantly affecting the amount and size of larger native E S C complex that is similar to the non-overexpressed E S C complex in Superose 6 gel filtration chomatography (Peter Harte, personal communication). Therefore I compared the expression level of F - P H P - H A to that of endogenous P H . T o control the copper ion used, 100 p:M C u  2 +  was incubated with both non-transformed and  F - P H P - H A # 3 0 cells for three days, nuclear extracts were made, and extracts were analyzed by western blot probed with P H antibody, as described in Materials and Methods.  Results are shown in F i g . 2.10. T o estimate the relative expression levels of  52  1 year  o  o  ©  >o  © © in  o Zeocin(ug/ml): ©  © ©  ©  ©  © ©  liiili  ^ |  Cu  2 +  (uM):  1000  4 months 0  0  © ©  5  0  6  7  8  a  1  2  3  4  5  100  • 0  9  Fig. 2.8 Expression test of stably transfected tagged ph?. The number #30 clone was maintained in Sf900 S F M medium with 100 |J,g/ml Zeocin for 4 months after selection, then different concentration of copper ions were used to induce the tagged ph? expression for 2 days, and the whole cell lysates were analyzed on S D S - P A G E (lanes 1 to 6). After maintaining 4 months under 100 Ug/ml Zeocin, #30 clone was maintained with 5 |Xg/ml Zeocin (lane 7) or without Zeocin (lanes 8 and 9) for additional 1 year, then the whole cell lysates were analyzed using S D S - P A G E without copper ion induction (lanes 7 and 8) or with 100 u M copper ion (lane 9). The western blot was then probed with anti-HA. The cell line continues to express F - P H P - H A in the absence of selection.  53  #30 clone doubling at 25 degree Celsius  Time (day)  Fig. 2.9 Comparison of K c l cell growth with and without copper ions. Clone #30 was cultured without or with copper ion and the cell densities were measured every day. The X-axis stands for the number of days when the cell density was counted, the Y-axis stands for the cell density. The diagam was drawn using C A - C r i c k e t .  54  endogenous and tagged P H P , the density of the bands after western blotting was measured using the N I H Image program. The relative densities of lanes 1 to 3 are 76.97, 31.61, and 48.85, respectively. The ratio of F - P H P - H A to endogenous P H P is 1.5. Therefore F - P H P - H A is about one and half-fold more abundant than the endogenous PHP.  Test of the Epitopes for Immunoaffinity Purification Initially I used a tandem H A tag at the C-terminus for two reasons.  First, a double  tag w i l l have higher affinity for its antibody, allowing harsher wash conditions for the immunoaffinity column, which should yield a more purified P H complex. Second, phf has an internal initiation site, so the N-terminus tagging may have affected the internal intiation product. T o test the usefulness of H A affinity matrix and the availability of H A tag for immunoaffinity purification, I checked the efficacy of the H A tag using immunoprecipitation (IP) experiments followed by analysis using western blots. K c l nuclear extract prepared from stably transfected cells as described above were incubated, with lOp.1 of anti-HA Sepharose beads (BabCo) for 10 h at 4°C as described in the Materials and Methods. Beads were recovered by centrifugation and washed 5 times in buffer H E G B 0.15. The supernatant after the IP step was prepared for analysis on S D S P A G E (Fig. 2.11, lane 1 of panel A ) to obtain an estimate of the efficiency of the IP step. One sample of beads was eluted using competitor H A peptide (lane 3) as described in the Materials and Methods. Following the wash step, the beads were boiled in 2 S D S - P A G E sample buffer, and analysed on an S D S - P A G E (lane 2). Following transfer to nitrocellulose, I did a western using anti-HA antibody. A s shown in F i g . 2.11, panel A ,  55  N E  .  non-transfected 40ug 20ug  transfected 20ug anti-PH  Fig. 2.10 Expression levels of endogenous and tagged ptf. Nuclear extracts of nontransformed K c l cells (lanes 1 and 2) and clone #30 (lane 3) were analyzed on S D S P A G E . The western blot was probed with anti-PH. Overall levels of expression are similar.  56  there is much more protein in lane 2 than in lane 1, showing that the efficiency of the IP step is high. However, there is no tagged P H in the eluate after competition with H A peptide (lane 3), showing that the affinity of the anti-HA antibody for the tagged P H P is very high. Nevertheless, these results show that the tandem H A tag is efficacious for the recovery of P H P 1 7 0 and PHP140. The low elution efficiency of F - P H P - H A from anti-HA beads might be caused by the tandem H A tags which should have higher affinity for the affinity matrix. In support of this idea, Hernan et al. (Hernan et al., 2000) reported that a triple F L A G epitope tag gives tenfold more sensitivity than a single F L A G tag. Sigma started selling triple F L A G peptide to elute the triple F L A G - f u s i o n protein in 2000. It should be possible to elute the H A tagged P H P using double H A peptide. However, the cost and the possible difficulty of regenerating the matrix meant that for most experiments reported in this thesis, F - P H P H A was removed from the affinity matrix by eluting in 2% S D S . Similar experiments were carried out to demonstrate the efficacy of the F L A G tag, and to test recovery after elution with competitor peptide. Extracts prepared as described above were reacted with a n t i - F L A G M 2 agarose beads (Sigma). Following recovery of the beads by centrifugation, the beads were washed, and then the P H P was eluted by adding competitor F L A G peptide. Each sample was subjected to western analysis using anti-HA on the F L A G immunoaffinity purified samples (Fig. 2.11 B ) . Lane 1 is nuclear extract, lane 2 is supernatant of M 2 beads, lane 3 is l u l M 2 beads before elution, lane 4 is 0.1 p,l beads before elution, lane 5 is  M 2 beads after elution, lane 6 is 0.1pi beads  after elution, and lane 7 is the eluate. About 18% of the protein present in the extract is recovered after IP with a n t i - F L A G beads (compare lanes 1 and 2), estimated by  57  measuring the band density of P H P 170 using N I H image program (relative density is 150.21 and 122.90, respectively). It also appears that most of the bound tagged P H P was eluted from the beads using F L A G peptide (compare lanes 3 and 5).  Selection of Stable Cell Lines Expressing Tagged PC (F-PC) A similar strategy to that described for p N T P 1 4 F - P H P - H A was employed to select stably transformed cell lines expressing F L A G tagged P C . About 10 cells of Zeocin6  resistant clones were heated with 2 x S D S buffer and analyzed using western blotting. One of the western blots probed with a n t i - F L A G M 2 is shown in panel A of F i g . 2.12. Altogether, 9 of 60 clones expressed F - P C .  This higher efficiency of stable cloning than  F - P H P - H A (15% versus 3%) might be due to the smaller size of P C . Clone #5 was selected for the work described below. To evaluate the size of tagged P C compared to that of endogenous P C , nuclear extracts of non-transformed and clone #5 were analyzed by western blotting and probed with either anti-PC or a n t i - F L A G . The results are shown in panel B of F i g . 2.12. A s shown, the F L A G tagged P C migrates slightly slower than endogenous P C due to the tag which adds about 2.2% to the mass, which should be detectable on S D S - P A G E . It can also be seen that there is a smaller band in the #5 nuclear extract probed by anti-PC (lane 3 of B ) but not by a n t i - F L A G (lane 4 of B ) . I suspect that this is a degraded product due to overexpression, perhaps because some of the overproduced product is not able to participate in a protein complex, and thus is more vulnerable to proteases. Due to the multiple products seen in the #5 nuclear extract, it is difficult to estimate the expression level compared to endogenous P C , but I estimate that the expression level of tagged  58  A  B  1  2  3  4  5  6  7  Fig. 2.11 Immunoaffinity test with anti-HA and a n t i - F L A G affinity beads. (A) Lane 1 is the input nuclear extract, lane 2 is the anti-HA affinity beads after H A peptide elution, lane 3 is the the eluate using H A peptide. The blot was probed with H A antibody. (B) Lane 1 is the input nuclear extract (3p:g), lane 2 is supernatant of M 2 beads equivalent to the input amount in lane 1, lane 3 is l p j M 2 beads before elution, lane 4 is 0.1 | i l beads before elution, lane 5 is l f i l M 2 beads after elution, lane 6 is 0.1 jtxl beads after elution, and lane 7 is eluate using 0.2 mg/ml F L A G peptide. The western blot was probed with anti-HA.  59  version is within at least several-fold of the endogenous P C . Together, these experiments demonstrate that both F L A G and H A tags can be used to recover P H P proteins from K c l cell extracts. While the yield after IP is not maximal, these results encouraged me to make stable cell lines expressing F - P H P - H A . In this chapter, I have shown that PHP-170 can be successfully tagged, and expressed in K c l cells. The F - P H P - H A proteins are expressed at levels very close to those of endogenous levels, although there is a change in the relative abundance of F P H P - H A - 1 7 0 and F - P H P - H A - 1 4 0 compared to the endogenous PHP-170 and PHP-140. Preliminary experiments show that both the F L A G and H A tags can be used to IP F - P H P H A from cell extracts. The results of these experiments allowed me to go on to purify the proteins that associate with P H P in vivo. These experiments are described in chapter III.  60  A  clone #: 1  2  3  4  5  6  7  8  9  10  11  12  13  14  Fig. 2.12 Cloning and testing of p N T P 1 4 F - P C . (A) Western blotting analysis of Zeocin-resistant clones transfected with p N T P 1 4 F - P C . The numbers of clones are indicated on the top. The membrane was probed with anti-Flag M 2 . (B) Comparison of endogenous and tagged P C . Nuclear extracts loaded are indicated at the bottom, lanes 1 and 2 are from non-transformed K c l cells, and lanes 3 and 4 are from clone #5. Antibodies used to probe the western blot are indicated on the top, anti-PC (lanes 1,2 and 3) and a n t i - F L A G (lane 4). Different protein bands are indicated to the right.  61  Chapter III Purification of PHP and Associated Proteins from Nuclear Extracts of Kcl Cells INTRODUCTION Nuclear Extract Preparation from Cell Lines Although genetic analysis has demonstrated that ph interacts with many other genes in Drosophila (Cheng et al., 1994), we still do not understand silencing at the molecular level. One approach to understanding the mechanism by which P H maintains silencing is to determine what other proteins it associates with. T o characterize the members of the PHP-containing complex(es). I obtained cell-free extracts and isolated P H P and associated proteins using epitope tagging strategy. Initially, nuclear extracts were made for use in the in vitro transcription assay (Dignam et al., 1983; Parker and Topol, 1984; Heiermann and Pongs, 1985; L u e and Kornberg, 1987). In general, nuclei are isolated at low ionic strength (0.01 M KC1), and nuclear extracts are prepared by salt extraction (0.42 M N a C l or 0.36 M ammonium sulfate). Salt extraction removes histones, other non-specific D N A binding proteins like H M G s , and most of the nucleoplasm^ proteins from the nucleus. Both 0.42 M N a C l and 0.36 M ( N H ) S 0 , have been used successfully to make active extracts for in vitro 4  4  transcription and purification of transcription factors (Dignam et al., 1983; Parker and Topol, 1984). The difference between these two salts is the ionic strength, 0.42 and 1.08, for N a C l and ( N H ) S 0 respectively. Thus the former extracts fewer chomatin proteins, 4  4  but the latter extracts most chomatin proteins including histones H 3 and H 4 (Wolffe, 1998).  62  Traditionally, ammonium sulfate is used to precipitate nuclear extract because it has following advantages. First, at saturation, it has high molarity and higher ionic strength. Second, it produces less heat of solution when mixed. Third, its density is low (1.235 gem" at 4.04 M at 20°C), so does not interfere with sedimentation of precipitated 3  proteins. Fourth, it prevents bacterial growth. Finally and importantly, it protects most proteins from denaturation (Englard and Seifter, 1990).  General Considerations for Handling Proteins Because proteins are less stable than D N A , all protein work was performed either on ice or in the cold room and as fast as possible. Secondly, I attempted to keep protein solutions at as high a concentration as possible to minimize denaturation and dissolution of protein complexes stabilized by low affinity interactions. D T T was added to protect protein thiols from oxidation. Chelating agents such as E D T A and E G T A are useful in protecting enzymes from inactivation by heavy metals and from proteolysis by metalloproteases. Glycerol and other polyols were added to stabilize protein-protein interactions. Protease inhibitors with widely differing mechanisms and specificity were added to inactivate different proteases. Most solutions contained a low concentration of detergent to reduce loss on the walls of containers.  Ion Exchange Chomatography Ion exchange chomatography (IEC) is designed to separate ionic or ionizable compounds. I E C has both stationary and mobile phases, and differs from other types of liquid chomatography in that the stationary phase carries ionizable functional groups.  63  The oppositely charged ion paired with the functional groups is called the counterion. This counterion is not fixed and can be displaced. I E C is named on the basis of the sign of these displaceable charges. Thus, in anion I E C the fixed charges are positive and in cation I E C the fixed charges are negative. The type of the functional group determines the strength of the ion exchanger; their total number and availability determines the capacity. A variety of groups have been chosen for use in ion exchanges. Some of the anion functional groups are diethylaminoethyl ( D E A E ) , quaternary aminoethyl ( Q A E ) , and quaternary ammonium (Q); the cation functional groups are carboxymethyl ( C M ) , sulphopropyl (SP), and methyl sulphonate (S) (Himmelhoch, 1971). Q A E , Q, and S are used to form strong ion exchangers; D E A E , C M , and SP are used to form weak ion exchangers. The terms strong and weak refer to the extent of variation of ionization with p H and not the strength of binding. Strong ion exchangers are completely ionized over a wide p H range whereas with weak ion exchangers, the degree of dissociation and thus the exchange capacity varies much more markedly with p H . In theory, one could elute ion exchange columns by changing either buffer p H or ionic strength. In practice, changing ionic strength is more frequently used than changing p H because many proteins show minimum solubility in the vicinity of their isoelectric points. Many proteins also are more sensitive to p H change than to ionic change in terms of biological activity. In addition, it is more difficult to make continuous p H gradients at constant ionic strength, because simultaneous changes in ionic strength, although small, also occur, and the buffering capacities of the systems produced are pH-dependent.  64  Gel Filtration Chomatography G e l filtration is unique in that fractionation is based on the relative size of protein molecules. G e l filtration is performed using porous beads as the chomatographic support. A column constructed from such beads w i l l have two measurable liquid volumes, the external volume, consisting of the liquid between the beads, and the internal volume, consisting of the liquid within the pores of the beads. Large molecules w i l l equilibrate only with the external volume while small molecules w i l l equilibrate with both the external and internal volumes. This conceptualization has led to the gradual renaming of gel filtration as size-exclusion chomatography. Its main function is to determine i f proteins associate as monomers, or as members of high molecular weight complexes.  Immunoaffinity Chomatography The highly specific nature of the interaction of an antigen with its antibody has long been exploited to identify, quantitate, or purify antigens. The ability to perform these functions is based on the high degree of specificity of the reaction between antibody and antigen. Immunoaffinity chomatography ( I A C ) can be used to purify large amounts of a particular antigen. Because of the specificity of the antibody-antigen interaction, this technique exceeds all other single-column methods in yield and purity (routinely, 1,000to 10,000-fold in a single step) (Harlow and Lane, 1988). Therefore, all the purifications described in this thesis incorporate I A C . One big issue for I A C is the choice of antibody used. Polyclonal antibodies contain a range of molecules directed at all the epitopes on the surface of the protein and consist of antibodies with a wide range of affinities for each of the epitopes. The avidity of the  65  interaction between the antigen and portions of the antibodies render elution of the antigen almost impossible under conditions that retain the biological activity of the antigen. It was the development of monoclonal antibodies ( M A b s ) by Kohler and Milstein (Kohler and Milstein, 1975) that led to the increase in the use of I A C as a tool for protein purification (Jack, 1994). The first step of I A C is to prepare antibody affinity column, but for this work I used commercially available resins. The second step is the binding of the antigen to the affinity matrix. Because the antibody is not in solution, the time required for the antigenantibody interaction w i l l have different kinetics than soluble interactions, and takes considerably longer for equilibrium to be reached than for solution assays such as immunoprecipitation (Harlow and Lane, 1988). The third step is to elute the bound antigen off the column. The methods that can be used for elution include treating with harsh conditions such as low p H or detergent; or adding saturated competitor such as the peptide used for generating the antibody. Competition elution gives better purity and harsh conditions give better yield. Both procedures were used in the work described here.  Results and Discussion Large Scale Kcl Culture K c l cells were grown at room temperature (23°C to 25°C), with air as the gas phase. They were maintained in tissue culture plastic flask (T25). Cells were transferred every week with 1:20 dilution. K c l cells adhere to the surface loosely and can be dislodged by shaking the flask gently or blowing a gentle stream of medium from a  66  pipette at the surface. Although it has been reported that most Drosophila cell lines can be grown in spinner flasks with little adaptation (Cherbas et al., 1994), I found that K c l cells were not healthy and that it was hard to obtain high density in a spinner flask. Perhaps the stir bars break the cells at higher speed or do not give enough aeration at lower speed. I switched to using conical glass flasks with capacity of 6 liters for routine large-scale culturing. The cells were first maintained in T25 flasks and then transferred into 1 liter conical glass flask and then into 6 liter flasks, rotated at 75 rotations per min. Using these conditions, the cells behaved no differently from those maintained in plastic flasks. Both culture conditions permit cell doubling roughly every 24 h at 25°C (data not shown) and achieve a density of 1.2xl0 cells/ml. Routinely, 4 liters of culture from 3 7  flasks, about 25 to 35 m l packed cell volume, were collected for nuclear extract preparation.  Comparison of Native and Tagged PHP on Gel Filtration Chomatography P H P contains a S A M domain that homodimerizes and hetrodimerizes (Kyba and Brock, 1998b). This property may lead to the formation of non-native complexes of the tagged P H P , even though it is expressed at relatively low levels as indicated in chapter II. To determine i f the tagged and endogenous P H P had similar elution profiles, I used gel filtration chomatography to fractionate F - P H P - H A transfected and native nuclear extracts. Fig. 3.1 shows a western blot from every other fraction obtained after Sephacryl S400 H gel filtration chomatography. The upper panel is from the F - P H P - H A extract and probed with H A antibody. The lower panel is from native extract and probed with native anti-PH antibody (obtained from R. Kingston). A s shown, the F - P H P - H A elutes in  67  approximately the same fractions as native P H P , and the estimated M W ranges from 1-2 M D a . There is no monomer P H P or dimer P H P seen from either extract. These results strongly suggest that F - P H P - H A associates with other proteins to form high molecular weight complexes, and that F - P H P - H A associates similarly to endogenous PHP.  Development of Conventional Chomatographic Purification Scheme Since there is no activity assay available for P H , western blotting analysis was used to analyse the column fractions. I first tried a cation exchange column using B i o - R e x 70 resin (BioRad) because it has been used to purify P R C 1 (Shao et al., 1999), and to purify a bxd P R E binding complex (Hodgson and Brock, 2001). I used step elution because the elution properties of P H on B i o - R e x 70 have been previously characterized in the laboratory (Hodgson, personal communication). Routinely, 400 mg of nuclear extract was loaded onto 40 m l column volume of B i o R e x 70 column as described in the Material and Methods. The column was then washed with the column equilibration buffer. The flowthough (proteins not bound to column under the loading conditions) and the eluate following addition of column equilibration buffer was combined and named fraction BR0.1 ( 0 . 1 M KC1). I subsequently eluted the column with 0.18M, 0 . 3 M , and 0 . 6 M KC1 in the equilibration buffer as described in the Material and Methods. These fractions are called B R 0 . 1 8 , B R 0 . 3 , and B R 0 . 6 , respectively. A typical elution profile obtained by plotting the fraction number versus respective absorbance value of protein at 280 n m is shown in Fig. 3.2.  68  Fraction No.  21  23  wm~.  $£jgp  25  27  gjS§j»  sp&is  29  31  33  35  37  39  41  43  45  47  49  HA  PH  Fig. 3.1 Western blotting analysis of gel filtration fractions of nuclear extracts. The upper panel is a western blot of every other fraction obtained from gel filtration analysis of epitope-tagged P H nuclear extract and probed with anti-HA antibody. The lower panel is a similar blot from non-transfected K c l nuclear extract probed with P H antibody. The numbers on the top are the fraction numbers. Calibration markers are thyroglobulin (669 K D a ) , ferritin (440 K D a ) , and albumin (67KDa). The void volume is estimated according to the instructions from the matrix supplier (Amersham Pharmacia Biotech.).  69  1—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I o  >  n  o  i  n  o  v  "  }  O  v  ~  >  o  i  n  o  v  "  )  O  V  "  )  0  >  n  o  u  ~  )  Fraction no.  Fig. 3.2 Chomatogram of B i o R e x 70 chomatography. The X-axis indicates the fraction numbers. The Y-axis indicates the relative value of O D measured as described in the Materials and Methods. The elution peaks are labeled as B R 0 . 1 , B R 0 . 1 8 , B R 0 . 3 , and B R 0 . 6 respectively. The plot was drawn using program C A - C r i c k e t . 2 8 0  70  0  The fractions with A  2 8 0  30% higher than the base level were pooled for each step  elution, and the protein concentration of each protein peak was determined by Bradford method (Bradford, 1976) using B S A as a standard. The typical distribution of protein amounts of the four fractions is shown in Table 3.1. To decide which fraction to use for the next column, a western blotting analysis was performed on proteins from pooled B i o - R e x column fractions that were immunoprecipitated with beads containing a n t i - F L A G antibody. About 250p;g of protein from each fraction was immunoprecipitated with 20 p i a n t i - F L A G M 2 beads and analysed by electrophoresis on S D S gels as described in the Material and Methods. T o ensure that the F L A G beads were not saturated, I tested the binding capacity of the affinity beads (data not shown). The amount of beads used here is under saturation, so the results should be quantitative. F i g . 3.3 is a western blot of immunoprecipitated fractions probed with H A antibody. A s shown, the BR0.1 fraction has the least F - P H P H A . The other fractions contain about the same amount of protein. F r o m Table 3.1, it is clear that the B R 0 . 1 8 fraction contains the most protein. During the course of this thesis, Shao et al. (Shao et al., 1999) characterized the proteins associating with P H P present in high salt (0.85M KC1) nuclear extracts from embryos, so I chose to analyse a fraction that might be different from that previously characterized. For the second column, I initially chose an anion exchanger D E 5 2 (Whatman). But this resin only retains about 5-10% proteins of BR0.18 fraction, and the retentate contains much less F - P H P - H A than the flowthough as determined by western blotting analysis (data not shown). Therefore I chose another cation exchanger, SP Sepharose as next  71  Table 3.1 Protein distribution on B i o - R e x column Fractions  Protein amount (mg)  % of total  BR0.1 BR0.18 BR0.3 BR0.6  238 60 21 23  69.5 17.5 6 7  72  00  •  •  •  CD  CP  CP  fie  fiC  QS  &2 ^^^M|  ^ • ^ ^  os ^m^m]  co pwiwm^  ^ ^ ^ ^  *  CD f*Wi  fifi pMM^  |MMM^  PHP170 i PHP140  Fig. 3.3 Western blotting analysis of immunoaffinity-purified F - P H P - H A from B i o R e x column fractions. About 250 ug of B R 0 . 1 , BR0.18, B R 0 . 3 , and B R 0 . 6 was purified with anti-Flag affinity agarose beads, and the purified proteins were loaded onto lanes 1, 2, 3, and 4, respectively. The western blot was probed with H A antibody. The location of P H P 170 and P H P 140 are indicated on the right.  73  column. Routinely, 100 mg B R 0 . 1 8 fraction was loaded onto a 10 m l column in buffer H E M G containing 0 . 1 8 M KC1 as described in the Material and Methods. Then the column was washed with equilibration buffer, combined with the flowthough, and named SP0.18. The elution steps were achieved with the same buffer containing 0 . 3 M and 0 . 5 M KC1, and named SP0.3 and SP0.5 respectively. One typical protein versus fraction number profile is shown in Fig. 3.4. Similar to that of B i o - R e x chomatography described above, the fractions containing over 30% of base level of protein were pooled, and the concentration of each step eluate was determined. The protein distribution of each elution peak is shown in Table 3.2. From Table 3.2, it can be seen that the highest amount of protein is in the SP0.3 fraction. T o determine which fraction contains most F - P H P - H A , immuno-affinity precipitation was performed on 250 lUg of protein from each of the thee fractions of SP Sepharose. The immuno-purified proteins eluted with F L A G peptide were analyzed by western blotting. The results are shown in F i g . 3.5. The SP0.5 fraction contains more F P H P - H A than the other two fractions. The SP0.5 fraction gives 2.1 fold more F - P H P - H A than SP0.3 fraction as measured by the N I H image program. Therefore the SP0.5 fraction was chosen as the material for next column. Because two cation exchange columns had been used, and I had tested one anion exchanger, D E 5 2 , that did not work well, I therefore chose another strong anion exchanger, Q Sepharose (Pharmacia) as next column matrix. The SP0.5 fraction was dialyzed to buffer T E M G 0 . 1 as described in the Material and Methods. Routinely, 60 mg protein was loaded onto 12 m l column volume of packed column. A g a i n the flowthough  74  Fig. 3.4 Chomatogram of SP Sepharose chomatography. The X-axis shows the fraction numbers. The Y-axis shows the relative value of O D measured as described in the Material and Methods. The elution peaks are labeled as SP0.18, SP0.3, and SP0.5 respectively. The plot was drawn using program C A - C r i c k e t . 2 8 0  75  Table 3.2 Protein distribution on S P Sepharose column Fractions  Protein amount (mg)  % of total  SP0.18 SP0.3 SP0.5  20 45 23  23 51 26  76  QO  0* Xii  m  IT) ON GC  CZ)  PHP170 PHP140  Fig. 3.5 Western blotting analysis of immunoaffinity-purified F - P H P - H A from SP Sepharose column fractions. About 250 p:g of SP0.18, SP0.3, and SP0.5 was purified with anti-Flag affinity agarose beads, and the purified proteins were loaded onto lanes 1, 2, 3, respectively. The western blot was probed with H A antibody. The location of P H P 170 and P H P 140 are indicated on the right.  77  and the starting buffer washes were combined and named QO. 1. Then the column was eluted with buffer T E M G containing 0 . 3 M and 0 . 5 M K C 1 , and the eluates are called Q0.3 and Q0.5, respectively. The typical elution profile of protein versus the fraction number is shown in F i g . 3.6. The fractions containing at least 30% over the base level of proteins were pooled and the protein concentration of each peak were measured as before. The protein distributions of the three step eluates are shown in Table 3.3. A s had been done on the previous columns, 250 [ig of protein from each of the step eluates were immuno-purified by M 2 affinity beads as described in the Materials and Methods. The washed beads were eluted by F L A G peptide and the eluates were analyzed by western blotting. The western blot was probed with H A antibody and the results are shown in Fig.3.7. Q0.3 contains the most F - P H P - H A protein, and contains almost half the total protein (Table 3.3). Therefore this fraction was chosen as the material for immuno-affinity purification which w i l l be described in the next section.  Immunoaffinity Purification of F-PHP-HA Complexes A s shown in chapter II, it is difficult to elute F - P H P - H A protein with H A peptide from anti-HA affinity beads because of the tandem H A tag. Therefore a n t i - F L A G M 2 agarose from Sigma was used to purify proteins associating with F - P H P - H A . T o control for the non-specific binding of proteins to the affinity beads, a parallel mock purification was performed. T o do this, the non-transfected nuclear extract was purified exactly as described for transformed nuclear extract as described in the previous sections.  78  Fraction N o .  Fig. 3.6 Chomatogram of Q Sepharose chomatography. The X-axis shows the fraction numbers. The Y-axis shows the relative value of O D measured as described in the Material and Methods. The elution peaks are labeled as Q0.1, Q0.3, Q0.5, respectively. The plot was drawn using program C A - C r i c k e t . 2 g 0  79  Table 3.3 Protein distribution on Q Sepharose column Fractions  Protein amount (mg)  % of total  Q0.1 Q0.3 Q0.5  26 23.45 2.7  49.86 44.96 5.18  80  • o> •H  i/)•  •  o  o —*• PHP 170 PHP140  Fig. 3.7 Western blotting analysis of immunoaffinity-purified F - P H P - H A from Q Sepharose column fractions. About 250 \ig of fractions Q0.1, Q0.3, and Q0.5 was purified with anti-Flag affinity agarose beads, and the purified proteins were loaded onto lanes 1, 2, and 3, respectively. The western blot was probed with H A antibody. The location of P H P 1 7 0 and P H P 1 4 0 are indicated on the right.  81  About 37 mg Q0.3 fraction was incubated with 1 m l of the affinity beads for 10 h, then the beads were washed and eluted as described in the Material and Methods. The purified proteins were analyzed on 8% S D S - P A G E , and the gel was silver stained as described in the Material and Methods. The results are shown in Fig.3.8. Lane 1 is protein standard, lane 2 is from the mock purification, and lane 3 is the proteins associated with F - P H P - H A . It can be seen that there are 5 protein bands being purified in addition to the P H P 170 and P H P 140 bands. These specific bands are named with a P followed by a number of estimated M in K D a . The bands are P135, P120, P70, P64, and r  P48. The band described as "P64" appears to be a mixture of proteins judging by the diffuseness of the band, and "PI20" also appears to be thee closely-spaced bands. It can also be seen that there are 7 protein bands that are present in both mock and the real purification. These "sticky" proteins either have an epitope binding to the antiF L A G M 2 antibody on the beads or have certain degree of affinity to the agarose beads. Alternatively, the affinity matrix can be viewed as a ion exchange matrix because the antibody molecules are charged molecules. Therefore, it is not surprising that some of the 'sticky' proteins from the nuclear extract are 'ion-exchanged' and thus bind to the affinity beads. This non-specific binding is a common phenomenon in many immunological assays like immunobloting and immunohistochemistry. Therefore it is always necessary to perform a parallel negative (mock) control experiment (Harlow and Lane, 1988). It appears that the purification presented here has relatively low yields of protein suitable for analysis by mass spectrometry, although as w i l l be demonstrated in subsequent chapters, it is very suitable for analysis by western blotting.  82  4 #  •  $r  200 PHP170 PHP140 P135 P120  Fig. 3.8 Silver staining analysis of immunoaffinity-purified F - P H P - H A and associated proteins. Lane 1, 312.5 ng of marker, and mass labeled in K D a to the left. Lane 2,eluate of the mock-purified reaction. Lane 3, eluate of the F - P H P - H A associated proteins from Q0.3 fraction. The specific protein bands are indicated with lines, and the non-specific bands are labeled with arrowheads to the right.  83  One question is whether the proteins that associate with F - P H P - H A are the same or different as those found in P R C 1 (Shao et al., 1999). Although I chose to analyse a B i o Rex fraction different from that containing P R C 1 to purify proteins associated with F P H P - H A , it remained possible that some P R C 1 remained in the 0.18M salt fraction. Therefore, I did a western blotting analysis using antibody to P S C on all the fractions used for F - P H P - H A complex purification because P S C is a subunit of P R C 1 . I used antiP S C (gift of Dr. Paul Adler) to probe the blot. The results are shown in F i g . 3.9. P S C is present in all the fractions of the thee columns described above, but it is not present in the immunoaffinity-purified F - P H P - H A complexes from BR0.18, or SP0.5 or Q0.3 (lanes 11, 12, 13). In addition, S C M , present in P R C 1 , is not present in the immunopurified F - P H P H A either (data not shown). Therefore P R C 1 is not present in the immunopurified F P H P - H A , indicating that I have identified a different subset of proteins that associate with P H P than those previously identified (Shao et al., 1999).  Double Immunoaffinity Purification of F-PHP-HA Complexes The purification procedure described above is tedious and has low yields although it has been used for mass spectrometry analysis (data not shown) and for western blotting analysis as described in chapters I V and V . I therefore took another approach to purify the F - P H P - H A complex for mass spectrometry sequencing analysis, based on a double tagging strategy. A s introduced in chapter II, a double tagging strategy has been used to purify protein complexes for characterization (Rigaut et al., 1999; Nilsson et al., 1997; Murby et al., 1991). Because I had tagged P H P with both F L A G and H A epitope, I  84  Column Fractions oc o  X  T-H  o  X  ©  *  35  *0  00  ©  ©  *  T-H  g  FLAG IP T—<  •  r-H  C}  IT)  T«P  •  •  •  ©  CL  ©  ©  ©  CQ  8 a a a  8  9  10  in © PH  CA  ©  O*  PSC  11  12  13  Fig. 3.9 Immunoblotting analysis of P S C across the column fractions. Twenty jxg of the column fractions indicated on the top of the Fig. (lanes 1 to 10) and the equivalent amounts of one silver gel (Fig 3.8) purified F - P H P - H A complex from different fractions (lanes 11,12, 13) were analyzed on S D S - P A G E , and the western blot was probed with anti-PSC. The P S C protein bands are marked to the right with a line.  85  employed a two tandem immunoaffinity procedure to purify F - P H P - H A complex for mass spectrometry analysis (chapter I V ) . Routinely, 600 mg of nuclear extract was incubated with 6 m l a n t i - F L A G agarose and washed as described in the Material and Methods. The bound proteins were then eluted with F L A G peptide. The eluates were incubated with anti-HA Sepharose beads and washed as described in the Material and Methods. The bound F - P H P - H A and any associated proteins were eluted with 2% S D S because competition with H A peptide gives extremely low yield (chapter II). The purified proteins were analyzed on 8% S D S P A G E , and the gel was silver-stained. The results are shown in F i g . 3.10. Bands corresponding to proteins that associate with F - P H P - H A are labelled as P followed by the estimated M and non-specific bands that associate with immunoaffinity beads in a mock r  purification are labelled with arrowheads. Interestingly, the protein composition of this double immunoaffinity purified complex is very similar to that obtained by the conventional chomatography plus an immunoaffinity strategy described above. In particular, P135, P120, P70, P64,and P48 are found in both preparations. It is also noticeable that the proteins found in the mock preparation (data not shown) also cofractionate using the double immunoprecipitation procedure, suggesting that these proteins associate with the immunoaffinity beads. Together, the results suggest that the purification schemes are likely to identify very similar proteins that associate with P H P in vivo. In turn, this suggests the procedures developed here to identify PHP-interacting proteins are robust. In this chapter I describe the development of two independent ways of purifying P H P containing complex(es) from K c l cell nuclear extract. Both methods give similar  86  results, as determined by analysis of the most abundant proteins on silver-stained gels. Nevertheless, because there are only 5 proteins in addition to PHP-170 and PHP-140 detectable, it is not possible to speculate on the overall similarity of the PHP-associating proteins obtained by these methods. There may be proteins that are not detected by silver-staining that also associate with F - P H P - H A that are not detected here. It also seems clear that the overall yields of PHP-associated proteins are very low, making analysis by mass-spectrometry possible only for the most abundant proteins. A s w i l l be seen in the subsequent chapters, western analysis suggests that the overall composition of the PHP-associating proteins is similar after fractionation using each method. However, the western analysis also makes it obvious that there are PHP-associating proteins that can be detected with antibodies that are not detectable on silver-stained gels, or that are masked by background proteins. A n analysis of PHP-associating proteins by western analysis and by sequencing by mass spectrometry w i l l be presented in the following chapters.  87  200  PHP170 PHP140 P135 P120  116 97 84 P70  66  P64  55  i 45  P48  36  Fig. 3.10 Silver-staining analysis of double immunoaffinity-purified F - P H P - H A and associated proteins. Lane 1 was loaded with 312.5 ng of marker, and the molecular weights are indicated in K D a to the left. Lane 2 was loaded with the double immunoaffinity-purified F - P H P - H A . The specific protein bands are indicated with lines, and the non-specific bands are labeled with arrowheads.  88  Chapter IV Further Characterization of PHP-Associated Proteins INTRODUCTION Much can be learned about the function of proteins by determining the identities of proteins that associate with the protein of interest in vivo. This is especially important in the case of PcG proteins, because we do not know the mechanism of PcG-mediated silencing. Previous chapters have described my attempts to immunoprecipitate proteins that associate with PHP. As shown in chapter m, PHP associates with unknown subunits. To identify PHP-associated proteins, I employed two approaches: immunoblotting of candidate proteins and sequencing proteins bands from gel slices by mass spectrometry. Western blotting is a cheap, fast, and specific way to identify proteins if antibodies are available, and reasonable guesses can be made about candidate proteins. Twodimensional gel electrophoresis followed by western blotting analysis detected with E C L (enhanced chemiluminescence) reagents has been used to identify hundreds of human proteins (Celis and Gromov, 2000). Several groups have probed two-dimensional gels of proteins from allergy-causing organisms using antibodies derived from allergic patients (Breitenbach et al., 1996; Sander et al., 1998). The limitations of this method include: 1) the number of antibodies available for testing is limited, and thus the number of proteins that can be identified is limited; and 2) the resolution of gel electrophoresis is limited, so if proteins co-migrate, identification by western blotting may underestimate the number of proteins present. In this chapter, I test known chomatin proteins that have not been  89  linked previously to PcG complexes, and histone deacetylase, as this is thought to play a role in silencing (see below) (reviewed in Brock and van Lohuizen, 2001). On the other hand, mass spectrometry (MS) can overcome the above problems inherent in western blot analysis because it is much more sensitive and can deal with protein mixtures. Mass spectrometry relies on digestion of gel-separated proteins into peptides by a sequence-specific protease such as trypsin (Pandey and Mann, 2000). There are two main approaches to mass spectrometric protein identification (Yates, 1998; Griffiths et al., 2001). In the "peptide-mass mapping" approach, initially suggested by Henzel and co-workers (Henzel et al., 1993), the mass spectrum of the eluted peptide mixture is acquired, which results in a "peptide-mass fingerprint" of the protein being studied. This mass spectrum is obtained by a relatively simple mass spectrometric method, matrix-assisted laser desorption/ionization (MALDI) coupled to time of flight (TOF) mass analyzers. This mass data can be used to compare with masses calculated for theoretically possible enzymatic cleavage products for every sequence in a protein/DNA sequence database (Zhang and Chait, 2000). In a two-step procedure for rapid and unambiguous protein identification, M A L D I fingerprinting is the first step. The second step for protein identification relies on fragmentation of individual peptides in the mixture to gain sequence information. This is called tandem M S (MS/MS) (reviewed in de Hoffmann, 1996). In this method, the peptides are ionized by electrospray ionization (ESI) directly from the liquid phase. The peptide ions are sprayed into a tandem mass spectrometer which has the ability to resolve peptides in a mixture, isolate one species at a time and dissociate it into amino- or carboxy-terminal containing fragments (Yates et al., 1995; Griffiths et al., 2001). The  90  tandem mass spectrometric method is technically more complex and less scalable than M A L D I fingerprinting. Its main advantage is that sequence information derived from several peptides is much more specific for the identification of a protein than a list of peptide masses (Pandey and Mann, 2000). The organization of the higher order chomatin structure has been linked to the posttranslational modifications of histone tails, including acetylation, phosphorylation, and methylation. These modifications have been dubbed as "histone code" hypothesis (Strahl and A l l i s , 2000; Turner, 2000). The histone code hypothesis predicts that distinct modifications of the histone N termini can regulate interaction affinities for chomatinassociated proteins. For example, the bromodomain in transcription factors such as P / C A F and T A F 2 5 0 recognize histone tails only when they are acetylated at lysine n  residues (Dhalluin et al., 1999; Jacobson et al., 2000). A s introduced in chapter I, histone deacetylation is correlated to gene repression. To test the possible relationship of PcG-mediated silencing and histone deacetylation, I also performed histone deacetylation assays on proteins that associate with P H P .  RESULTS and DISCUSSION Western Blotting Analysis of Immunoaffinity Purified PHP To identify what proteins may associate with P H P after a single immunopurification step from nuclear extracts, I performed western blotting analysis with available antibodies against chomatin proteins. Results from this preliminary screen are shown in Fig. 4.1. In each case, lane 1 is the input nuclear extract, lane 2 is mock purification on the non-transformed cell nuclear extract, and lane 3 is immunopurified  91  PHP, obtained as described in the Materials and Methods. A s shown in panel A lane 3, F - P H P - H A was immunoprecipitated by F L A G beads but not in the mock purification (lane 2 of panel A ) , indicating that there is no background binding to the beads. The M i - 2 protein, originally identified as an autoantigen of the human disease dermatomyositis (Seelig et al., 1995; Woodage et al., 1997), belongs to the highly conserved C H D (chomodomain, helicase, D N A binding) family of proteins. In addition to an ATPase/helicase domain of the SWI2/SNF2 class, these proteins also contain two P H D zinc-finger motifs, two chomo domains, and a truncated helix-turn-helix D N A binding motif (Woodage et al., 1997). Homologues of the M i - 2 protein have been identified in vertebrates, Drosophila  (dMi-2), yeast, plants, and C. elegans (Delmas et al.,  1993; Woodage et al., 1997; Kehle et al., 1998; Eshed et al., 1999; von Zelewsky et al., 2000). In Drosophila,  Mi-2 has been correlated to P c G gene-mediated silencing of  homeotic genes (Kehle et al., 1998). I therefore tested if dMi-2 associates with P H P in vivo by probing the same western blot with dMi-2 antibody. A s shown in F i g . 4.1, panel B , dMi-2 does not associate with F - P H P - H A . This may indicate that either the genetic interaction seen between dMi-2 and Pc is indirect (Kehle et al., 1998), that the d M i - 2 containing P c G protein complex is not stable enough for purification, or that d M i - 2 is not associated with F - P H P - H A . To control for the non-specific binding of proteins to the affinity beads, I probed the same membrane with Drosophila  SLN3 (dSLN3) and M O D U L O antibodies. The yeast  and mammalian SIN3 proteins are components of corepressor complexes that also contain histone deacetylases (Ahinger, 2000). In Drosophila,  SLN3-RPD3 binds to less  condensed, hypoacetylated euchomatic interbands and is absent from moderately  92  condensed, hyperacetylated euchomatic bands and highly condensed, differentially acetylated centric heterochomatin (Pile and Wassarman, 2000). As shown in panel C of Fig. 4.1, dSIN3 does not coimmunoprecipitate with PHP. Similarly, M O D U L O , a suppressor of Position Effect Variegation (PEV) (Garzino et al., 1992) does not coimmunoprecipitate with F-PHP-HA (panel D). These negative results strengthen my confidence that the positive results are not likely to be result from adventitious binding to the beads. To test the hypothesis that histone deacetylation is involved in PcG-mediated silencing, I probed the western blot with antibody to RPD3, a class I histone deacetylase (Gray and Ekstrom, 2001; Johnson et al., 1998). As shown in Fig. 4.1, panel E , low levels of RPD3 coimmunoprecipitate with PHP (lane 3) but not in the mock purification (lane 2), suggesting that RPD3 might participate in PHP function in vivo. As a further test of the hypothesis that PHP associates with HDACs, I probed the same membrane with an antibody to a histone deacetylase-associated protein, p55. p55 was originally identified as a component of Drosophila  chomatin assembly factor 1 (CAF-1) (Tyler et  al., 1996), and of nucleosome remodelling factor (NURF) (Martinez-Balbas et al., 1998). p55 is a W D repeat-containing protein, a homolog of mammalian RbAp46/48, and a component of multiple chomatin remodelling complexes (Tyler et al., 1996) and functions in histone binding (reviewed in Kingston and Narlikar, 1999). As shown in Fig. 4.1, panel F, p55 coimmunoprecipitates with PHP but is not present in the mock purification. These results are consistent with a role for histone deacetylation in PcGmediated silencing, as suggested previously (Cavalli and Paro, 1999).  93  HA B  mm  dMi-2  dSIN3 D  Modulo  E  RPD3  F  P55  Fig. 4.1. Immuno-blotting analysis of immunoaffinity-purified PHP-associated proteins. L a n e l was loaded with the input nuclear extract. Lane 2 was the eluate from the mock purification. Lane 3 was the eluate of immuno-purified F - P H P - H A . Antibodies used are indicated to the right of each panel.  94  PHP Cosediments with PC, RPD3 and P55 on Glycerol Gradients The experiments described in the previous sections identify candidate interacting proteins but do not definitively establish that these proteins do interact with P H P in vivo. It may be that wash conditions are insufficiently stringent to remove all traces o f abundant proteins from the immunoprecipitate. However, negative results make it highly unlikely that a protein like d M i - 2 associates with P H P or P C in vivo. To establish that R P D 3 and p55 are members of a high molecular weight complex(es) containing P H P , it w i l l be necessary to show that these proteins cosediment on glycerol gradients, in fractions containing high molecular weight complexes. T o this end, a glycerol gradient ultracentrifugation analysis was performed on the immunoaffinity purified F - P H P - H A complex from Q0.3 fraction as described in chapter III. Every other fraction from the glycerol gradient was separated by S D S - P A G E . Western blotting analysis results are shown in F i g . 4.2. A s shown, F - P H P - H A (panel A ) , R P D 3 (panel C ) and p55 (panel D ) cosedimented in fractions 25-29 that contain high molecular weight complexes. A s noted in chapter I, P C and P H cofractionated as a large complex, so I probed the same membrane with P C antibody (panel B ) . P C is also present in fractions containing high molecular weight subunits. It can also be seen from F i g . 4.2 that P C , R P D 3 and p55 distribute in fractions containing proteins smaller than 158 K D a , and that p55 distributes i n most o f the fractions. Due to the limited resolution of this technique, it is difficult to tell whether these thee proteins are i n smaller complexes, or are present as monomers, or as other mixtures of proteins. There are at least two possibilities for the presence of P C , R P D 3 , and p55 in fractions containing lower molecular weight fractions. First, in vivo, these  95  proteins m a y n o r m a l l y b e present as subunits o f a m u l t i m e r i c c o m p l e x , but that the c o m p l e x i s not stable after e l u t i o n f r o m the i m m u n o a f f i n i t y beads, perhaps because d i l u t i o n results i n b r e a k i n g o f l o w affinity interactions, l e a d i n g to d i s s o c i a t i o n o f the c o m p l e x . T h e alternative p o s s i b i l i t y i s that i n a d d i t i o n to b e i n g subunits o f a m u l t i m e r i c c o m p l e x , P C , RPD3,and p55 react i n d i v i d u a l l y w i t h F - P H P - H A . available for  RPD3 a n d p55,  T h e r e are n o data  but P H P does not b i n d P C in vivo ( K y b a a n d B r o c k ,  u n p u b l i s h e d ) , l e a d i n g m e to f a v o u r the f o r m e r e x p l a n a t i o n . T h e s e results strongly support the c o n c l u s i o n that P H P , P C ,  RPD3, a n d P55  proteins are m e m b e r s o f h i g h  m o l e c u l a r w e i g h t m u l t i m e r i c c o m p l e x e s , but leave o p e n the p o s s i b i l i t y that these proteins m a y also associate w i t h P H P as m e m b e r s o f l o w e r m o l e c u l a r w e i g h t c o m p l e x e s .  These  results d o not p e r m i t m e to say i f there is one o r m o r e h i g h m o l e c u l a r w e i g h t c o m p l e x e s c o n t a i n i n g these proteins. A n issue raised b y these results is w h y are the proteins i d e n t i f i e d b y western analysis not v i s i b l e o n the C o o m a s s i e - s t a i n e d g e l s h o w n i n Fig.4.3, o r i n the s i l v e r stained gels s h o w n i n the p r e v i o u s chapter. T h e proteins i d e n t i f i e d b y western analysis m a y s i m p l y b e less abundant i n n u c l e a r extracts than needed to b e v i s i b l e b y s t a i n i n g . T h e i m p l i c a t i o n i s that there are m u l t i p l e c o m p l e x e s c o n t a i n i n g F - P H P - H A , s o m e m o r e abundant than others. A l t e r n a t i v e l y , i f there i s a single c o m p l e x c o n t a i n i n g F - P H P - H A , then it must b e true that some subunits are either present at l o w e r s t o i c h i o m e t r y than others, o r have l o w e r affinity f o r the c o m p l e x , a n d thus are m o r e e a s i l y d i s s o c i a t e d . T h e techniques e m p l o y e d here c a n not d i s t i n g u i s h these p o s s i b i l i t i e s . T h e r e f o r e further experiments are necessary to resolve this issue.  96  158KDa Top Fraction*:  1 1  3  5  7  9  1 11  232KDa  669KDa  1 13  Bottom 15  17  19  21  23  *  25  »•  27  29  wm  HA  B  C  •  «*,,  «#«•..";'  RPD3  Fig. 4.2 Immunoblotting analysis of glycerol gradient fractions of F-PHP-HA-associated proteins. Each panel of western blot was probed with antibodies to H A (A), P C (B), RPD3 (C), and P55 (D). The fraction numbers from the gradient tube are indicated on the top of the diagram. Molecular markers are indicated with arrows downward. The top fraction is to the left and the bottom fraction is to the right.  97  Mass Spectrometry Sequencing of PHP-associated Proteins To identify some of the abundant subunits associated with P H P in vivo, about 600 mg of nuclear extract was purified using double immunoaffinity purification as described in chapter III. The purified P H P was separated electrophoretically on an 8% S D S - P A G E gel. The gel was stained with Coomassie Blue R-250 as described in the Material and Methods. The result is shown in F i g . 4.3. Interestingly, the specific protein bands shown in the silver-stained gel (Fig. 3.10) are the most abundant bands here (PHP170, PHP140, P121, P70, P64 and P48), with the exception of P135, suggesting these are the most abundant proteins associated with F - P H P - H A . The individual bands were excised with a razor blade as tightly as possible to the centers of the bands to maximize the ratio of protein to gel volume. The gel bands were stored in untreated 1.5 m l Eppendorf tubes. T o control for chemical noise and generalized, non-specific protein background, an equivalent non-staining area of the same gel was also excised. The gel slices were then washed twice with 50% H P L C grade acetonitrile in water. After discarding the supernatant from the second wash, the gel slices were left moist but not submerged in the wash solution. The washed gel slices were then sent to Harvard Microchemistry Facility for in gel digestion with trypsin and subsequent sequencing. The sequence analysis was performed by microcapillary reverse-phase H P L C nanoelectrospray tandem mass spectrometry ( p , L C / M S / M S ) on a Finnigan L C Q D E C A quadruple ion trap mass spectrometer (Harvard). This instrument configuration is capable of acquiring individual sequence ( M S / M S ) spectra on-line as high sensitivity ( « < 1 femtomole) for multiple peptides in the chomatographic run. These M S / M S  98  200  PHP170 PHP140 P121  116 97 84 66  -P70 -P64  55 P48 45 36  Fig. 4.3 Coommassie staining of S D S - P A G E gel. The molecular markers (left lane, 45)xg loaded) are indicated to the left in K D a . The protein bands of the right lane exercised for sequencing are indicated to the right. A non-specific band is indicated with an arrowhead to the right.  99  spectra are then correlated with known sequences using the algorithm Sequest developed by Yates et al. (Yates et al., 1995) and programs by Chittum et al. (Chittum et al., 1998). The peptide sequences obtained and the corresponding identity of the possible candidates are shown in Table 4.1. P121 is a gene product with unknown function, as shown by searches of Flybase and B L A S T searches of the public databases. P70 is heat shock 70 cognate 4 gene product (HSC70.4). Interestingly, the P70 band purified using the thee conventional chomatographies and one F L A G M 2 immunoaffinity as described in chapter III, was also identified as HSC70.4 using mass spectrometry mass mapping done by D r . A M a z o (data not shown), suggesting the validity of this identification. It further suggests that the protein components identified using the two ways of purification are similar. P64 is a mixture of Dnop5 gene product and P C . Dnop5 is a member of the Nop/Sik family of the conserved r R N A processing factors. Its functional relationship with P c G at present is unknown. The presence of P C is expected from immunoblotting experiments (Fig.5.1). P48 is another chaperone protein based on homology to the DnaJ heat shock protein in E. coli, and its homologs have been identified in Caenorhabditis  elegans, Homo sapiens, Mus musculus, Rattus norvegicus and Saccharomyces cerevisiae (Flybase). F i g . 4.4 shows a comparison of the J-domains of the DnaJ family members from the six species. It can be seen that this domain is highly conserved from E. coli to humans. The overall conservation of the family members among eukaryotes is about 40%. The J-domain is believed to recruit Hsp70 by binding to its ATPase domain and accelerate the ATP-hydrolysis step of chaperone cycle (Kelley, 1998). J-domain proteins have also been found to participate in processes such as cell cycle control by D N A tumor viruses, or regulation of protein kinases (reviewed in Kelley, 1998).  100  Table 4.1 Proteins in the F - P H P - H A complex identified by u L C / M S / M S Protein Bands  Number of Peptides Sequenced  P121  Sequences Obtained  Identity  IYDIIYELNR ELLIQEVPAQR FTNHEDYVQLR VLSQLSAK EFSLGSDKR LLSTDDAER  Unknown (CGI 7509)  P70  21  DNNLLGKFELSGIPPAPR WLDANQLADKEEYEH KFDDAAVQSDMK STNKENKITITNDKGR NQVAMNPTQTIFDAK TVTNAVITVPAYFNDSQR KTFFPEEISSMVLTK STAGDTHLGGEDFDNR SVIHDIVLVGGSTR ARFEELNADLFR MKETAEAYLGK YRNEDEEQKETIAAK HWPFEVVSADGKPK AMTKDNNLLGKFELSGIPPAPR IEVTYKDEKK LVTHFVQEFKR LSKEDIER ATLDEDNLK STMDPVEK MVNEAEK DLTTNKR  Chaperone (Hsc70-4)  P64  12  LLVDDVQSSLLVADAK FNDTTEALAAATAAVEGK KLEQVDNLYQEFETPEK DNMATSDLSDILPEDVEEK QKDEEVEAVEEAAAPEPEDQPTAK QQADSLLGGLPK IITDNIAFVK THLYDYLK MMAMAPNLTVLVGDTVGAR YGLIYHAQLVGQASQK EEPAEEEAEVK SEVFTYQPEADNTLNVK  Dnop5  DNATDPVDLVYAAEK  PC  FFGAGFGGS GGGR KVLEVHIEK AISQAYEVLSDADKR GGADSGDFR TLDDRDLIVSTQPGEVIR QVYDEGGEAAIK NPMDFFEK KLQLQK  Chaperone (CG8863)  P48  101  10  Ec Sc Cb Dm Rn Mm Hs  \i V k -- -M \ K M V K - -  20  | V Y h j l L o \ S K 1 A r. K 1 V l> I L <; v P \ 1 A" T I I C,\\" > 1) A S Y I)fT (; \ i. r ^ U I L (; v K K P N A T 111V D | T h. 1 \ D I L (, A S I* * DPT L N A T i c; \ K p V G V K P N A" t Y |v  1) -  K\Q  M A N VA M \ M V K - K - - I  1  1 rr 1iK  1'  D  |  40  Ec Sc 03 Dm Rn Mm Hs  p p p p p p p  7  TD51LiK±J D o D D D  K K K K K  N  <, \ rn  50  K F K i: i h i k E x s Ar [F K 0 I S Q - -  Jj E (. 1 k k  ii(T  l' 'li  1  k K k K K  K K K K  \  \ \ \  ^  1.  II K ^LU K k <• \ i k * II |< k M v i, K [ T K  K  p N F. (; j> K F K \ I S y - N< p N (.[jTJk F E I s N p N E F y I s Q - N p G E K Fk K y I s o - N  • --  1.  Y I.S Y I A Y : v I, S A L S A Y Y I A Y I: v L A : v L S Y \  11 11  \1  1.  1 >LL 1 ) I) p --LI 1 D E U A 1 |N P 1 AS D LL D | A 1 :' v  Y  K A Y K K L k Y K -\ Y K K I. II K A \ K K \ i K ^n K A Y R k L \ I k \ ti  60  (• D K [ K ] A E S[K  30  R T V o\ D N K L V D K I. E. N K I. Q E K I, Q E I  70  s Q k E K  R K R R U  K K D K E K S K K K K.  3  A  A Y I) -  D  Y n Y V Y I) L Y V L Y I) . Y  Q Q E R  I I  1) 1) 1) -  Fig. 4.4 Protein sequence alignment of the J domains of DnaJ chaperones. The N terminus 70 amino acids of DnaJ chaperones from E. coli(Ec, DnaJ), S. cerevisae(Sc, Y D J 1 ) , C. elegans(Ce, F39B2.10), D.melanogaster(Dm, CG8863), R. notvegicus(Rn, DnaJa2), M. musculus(Mm, Hsj2), H. sapiens(Hs, DnaJ-like 2) were obtained from Flybase and compared using program Macvector™7.0.  102  HSC70.4 Enhances the Homeotic Phenotype of ph and Pc Mutations If HSC70.4 is a true member of P H P complex and participates in homeotic gene silencing, it should be possible to see enhancement of ph and Pc homeotic phenotypes in flies doubly heterozygous for mutations in Hsc70.4 and P c G mutations. T o test this hypothesis, Pc /TM6 Sb, ph /FM7c, or ph /ph females were crossed to males carrying 4  409  2  2  the Hsc70.4 dominant negative mutant Hsc70.4' /TM6 B identified as an enhancer of the 95  Notch signalling pathway (Hing et al., 1999). Briefly, the number of legs with sex combs is determined in male flies that carry heterozygous PcG mutations, in the presence or absence of a mutation in the gene of interest. Wild-type males have 2 legs with sex combs, and a fly with sex combs on all legs would be scored as 6.0. This assay measures the penetrance of a homeotic transformation caused by ectopic expression of Scr (Glicksman and Brower, 1988). This assay was used to determine i f Hsc70.4 mutants enhance the phenotypes of ph and Pc and the results are shown in Table 4.2. Hsc70.4'  95  enhances all the alleles of ph and Pc tested. This is consistent with the recent observation that Hsc70.4 is an enhancer of Polycomb in a genetic screen (Mollaaghababa et al., 2001). However, Mollaaghabada et al. (2001) showed that Hsc70.4  195  extra sex combs phenotype of ph  410  suppresses the  and Pc . But it is not clear from Mollaaghabada et al. 3  (2001) how they scored the phenotype. It may be that i f they scored the number of bristles/sex comb instead of the number of legs with sex combs, that this could account for the difference. A l s o , it should be noted that in Mollaaghabada et al. (2001), Hsc70.4'  95  enhances the A 4 to A 5 transformation of Pc , even though it suppresses the T 3  and T to T, transformation. This result suggests that Hsc70.4'  95  3  can have a variable  effect on PcG-mediated silencing. It may also be explained as an allele-specific effect,  103  2  Table 4.2 Hsc70.4 enhances the extra sex combs phenotype of ph and Pc Cross  Genotype  Pc ITM6 Sb x Hsc70.4 ITM6B Tb  Pc ITM6 Tb Pc /Hsc70.4  ph /FM7c x Hsc70.4' /TM6B Tb ph /ph x Hsc70.4' /TM6BTb  4  195  409  95  2  2  95  a  b  Number of flies  Average no. of legs with sex combs  82 112  2.00 2.38  ph /Y; TM6B Tb/+ ph /Y; Hsc70.4 /+  80 101  2.65 4.83  ph /Y; TM6B Tb/+ ph /Y; Hsc70.4 /+  93 111  2.60 4.00  4  4  m  409  409  195  2  2  m  X test was used to determine the significance of the data, P<0.01 X test was used to determine the significance of the data, P<0.001 2  2  104  a  b  b  because Pc used in Mollaaghabada et al. (2001) is an antimorph but Pc used in this 3  4  thesis is hypomorph (Flybase). M o r e puzzling is the opposite effect seen with ph  409  and  ph , as both are ph? null alleles. 410  Together, the genetic results strongly support the biochemical results, and suggest that HSC70.4 associates with P H P in vivo, and that this association is necessary for silencing of Scr. Nevertheless, as with any genetic interaction, it is not possible to conclude from genetic data alone that two proteins interact directly. For example, HSC70.4 could be required to activate a P c G gene, so that a dominant negative mutation results in lowered amounts of P c G protein, causing the genetic interaction reported here. The molecular relationship between P c G silencing and H S C 7 0 . 4 is not clear even though the genetic interaction and copurification with P H P reported here suggests an important link. The evolutionary conserved members of the Hsp70 family play essential roles in preventing misfolding and aggregation of newly synthesized or unfolded proteins (Haiti, 1996; Bukau and Horwich, 1998; Bukau et al., 2000). More recently, Hsc70.4 was identified in a genetic screen as enhancer of Notch signaling (Hing et al., 1999), and as an enhancer of Polycomb (Mollaaghababa et al., 2001), suggesting that H S C 7 0 . 4 has new functions in gene expression. Consistent with this idea, in Drosophila, Hsc70.4 interacts with the ecdysone receptor genetically, and the H S C 7 0 . 4 containing chaperone complex is required and sufficient for the activation of ecdysone receptor D N A binding in vitro (Arbeitman and Hogness, 2000). In mammals, an H S C 7 0 . 4 containing complex has been shown to downregulate glucocorticoid receptor activity (Schneikert et al., 2000). Moreover, H S C 7 0 . 4 is required for the disruption of R b - E 2 F complexes by S V 4 0 T antigen in an ATP-dependent manner (Sullivan et al., 2000; Sullivan et al., 2001).  105  HSC70.4 belongs to the actin superfamily of ATPases that includes cytoskeletal actins, sugar kinases, glycerol kinase, and several prokaryotic cell cycle proteins (Holmes et al., 1993; Boyer and Peterson, 2000). A l l these proteins have the overall "actin fold" motif that is flexible and characterized by its ability to convert between conformational states in an ATP-dependent manner (Holmes et al., 1993; Kabsch and Holmes, 1995; Hurley, 1996). Interestingly, Actin-related proteins (Arps), a group of protein families that exhibit moderate sequence similarity among each other and to conventional actin (i.e., muscle actin), have been found in several chomatin remodelling protein compelxes (Cairns et al., 1998; Peterson et al., 1998; Zhao et al., 1998; Papoulas et al., 1998; Allard et al., 1999). In addition, Arp4p has been shown to bind to all four core histones (Harata et al., 1999). Consistent with this, mutations in an actin-related protein, act3, causes epigenetic effects on gene expression in yeast (Jiang and Stillman, 1996). What are the possible mechanisms of HSC70.4 action in PcG-mediated silencing? A s mentioned above, the available evidence suggests that H S C 7 0 . 4 and its related chaperones use the energy of hydrolyzed A T P to fold proteins to different forms. Thus one possibility is that HSC70.4 uses A T P binding and hydrolysis as molecular switches to regulate the activity of P c G complexes by assembling them with different subunits or changing their interaction partners. DnaJ chaperone (CG8863 product) may enhance the ATPase activity and thus increase the efficiency of switching.  Rpd3  Enhances the Homeotic Phenotype of ph and Pc Mutations In Drosophila, it has been shown that R P D 3 and P55 are subunits of the E S C - E ( Z )  complex, that Rpd3 is required for silencing mediated by a Polycomb response element  106  (PRE) in vivo, and that E(Z) and R P D 3 are bound to the Ubx P R E in vivo (Tie et al., 2001). These findings suggest that PcG-mediated repression might in part be achieved though histone modification. However, E S C and E(Z) are chromatographically distinct from the P R C 1 complex (Shao et al., 1999; Tie et al., 2001). A s shown above, R P D 3 coimmunoprecipitates with P H P and P C . T o my knowledge, this is the first demonstration of an association of P H or P C with an H D A C . T o test i f Rpd3 interacts with ph and Pc in vivo, Rpd3 /TM3 m  ph /FM7c, 409  Sb Ser males were crossed to Pc /TM3 Sb, 4  and ph /ph females, respectively. F I males were scored for sex combs on 2  2  all three pairs of legs. The results are shown in Table 4.3. It can be seen that Rpd  303  mutations weakly enhance the extra sex comb phenotype of ph and Pc alleles tested. These results are consistent with the coimmunoprecipitation data described above, and suggest that the association of R P D 3 with P H P is functionally significant. A s already noted, the genetic interaction reported here does not demonstrate a direct link between P H P and R P D 3 , as the same result could be obtained indirectly.  PHP-associated Proteins Have No Significant Histone Deacetylase Activity in vitro To determine whether the R P D 3 associated with P H P is enzymatically active, I assayed immuno-purified P H P from the Q0.3 fraction. The synthetic peptide corresponding to the N-terminal 20 amino acids of histone H 4 labelled with H-acetate 3  was used as substrate. Hela cell nuclear extract containing a rich source of histone deacetylase was used as a positive control. T o control for the specificity of histone deacetylase activity, butyrate, a specific inhibitor for histone deacetylase was used in parallel reactions. A s shown in Fig. 4.5, 5 p:g of Hela cell nuclear extract gives dramatic  107  Table 4.3 Rpd3 enhances the extra sex comb phenotype of ph and Pc Cross Pc /TM6 Sb x Rpd3 /TM3 Sb Ser 4  303  Genotype  Number of flies  Average no. of legs with sex combs  Pc /+ Pc /Rpd3  126 113  2.03 2.33  ph /Y; TM3/+ ph /Y; Rpd3 /+  44 66  2.80 4.10  ph /Y; TM6/+ ph /Y; Rpd3 /+  133 105  3.00 5.00  4  4  ph /FM7c x Rpd3 /TM3 Sb Ser 409  303  303  409  409  ph /ph x Rpd3 /TM3 Sb Ser 2  2  303  303  2  2  303  X test was used to determine the significance of the data, P<0.001 2  108  a  a  a  histone deacetylase activity and 100 m M butyrate inhibits about 90% of this activity. In contrast, 10 (ig of Q0.3 fraction gives low level histone deacetylase activity. This low level activity is specific because it can be inhibited to background level in the presence of butyrate. Surprisingly, the purified PHP-associated complex only gives the background level of activity. In yeast and mammals, the SIR2 protein has N A D - and ATP-dependent histone deacetylase activity. This activity is important for silencing (Imai et al., 2000; Landry et al., 2000; Smith et al., 2000). T o test whether an analagous N A D - and ATP-dependent activity associates with P H P , I performed a parallel reaction with immuno-purified P H P in the presence of N A D and A T P . A s shown in F i g , 4.5, P H P and associated proteins has no significant N A D - and ATP-dependent histone deacetylase activity. One obvious explanation for this negative result is that the histone deacetylase activity is inhibited in the complex. Interestingly, purified P R C 1 from  Drosophila  embryos contains R P D 3 but has no detectable histone deacetylase activity (R. Kingston, personal communication), suggesting that this might be general situation for P c G complexes in Drosophila.  It may be that the R P D 3 associated with P H P in vivo is at such  low levels that histone deacetylase activity cannot be detected, or that the amount of complex used was insufficient to allow detection.  RNA Is Not Required for The Integrity of PHP and PC Complexes Non-coding R N A is involved in several forms of gene silencing and activation. Xist R N A in required for X chromosome inactivation in female mammals (Mlynarczyk and Panning, 2000), and dosage compensation (gene activation) in male Drosophila  109  is  2000  (3 No Butyrate 1500 H  •  lOOmM Butyrate  g  ImM NAD  mi 4mM ATP Q.  IOOOH  o 500  •f-  Fig. 4.5 Histone deacetylase assays of immuno-purified F - P H P - H A . The Y-axis shows C P M of released free acetate. The X-axis shows the different samples assayed. Control is the reaction buffer without any protein added. Hela N . E . contains 5 u.g Hela cell nuclear extract (Upstate Biotech). Q0.3 contains 10 p.g of the Q Sepharose column fraction used for immuno-affinity purification of F - P H P - H A complex(es) as described in chapter III. L P . contains immuno-purified F-PHP-HA-associated proteins equivalent to the amount shown in F i g . 3.8.  110  achieved through the male-specific lethal ( M S L ) complex containing R N A s on X (roXl and roX2). In addition, M O F , one member of the M S L complex, specifically binds through its chromodomain to roX2 R N A in vivo (Akhtar et al., 2000). Non-coding R N A has also been shown to act as a coactivator, termed steroid receptor R N A activator ( S R A ) in SRC-1 complex (Lanz et al., 1999), and as trans- activator of transcription in viral gene (Sit et al., 1998). Therefore non-coding R N A s may be more commonly involved in organizing regulatory protein complexes than has been appreciated to date (Lanz et al., 1999; Eddy, 1999; Erdmann et al., 2001). Non-coding R N A s have also been reported from Ubx and abd-A of the bithorax complex (Lipshitz et al., 1987; Cumberledge et al., 1990). It was proposed that the R N A s may provide transregulatory functions (Lipshitz et al., 1987). T o test the hypothesis that R N A is required to maintain association of proteins with P H and P C in vivo, I immunopurified P H P and P C from nuclear extract in the absence or presence of RNase A . If R N A is important for the organization of P c G complexes, then one would expect that different proteins would associate with P H P or P C in the presence of RNase A . The results are shown in Fig.4.6. Lanes 1 and 2 are immuno-purified F - P C complex from stably- transfected cell nuclear extract, in the absence and presence of RNase A , respectively. Lanes 3 and 4 are purified F - P H - H A complex from stably transfected cell nuclear extract. It can be seen that RNase A does not change the composition of the bands associated with P H P and P C . Alternatively, it is possible that R N A is in the interior of the complexes and that is not accessible to the RNase A in the reaction. In this chapter, I described the identification of prominent protein bands associated with F - P H P - H A purified from previous chapter using tandem mass spectrometry and of  111  F-PC  F-PHP-HA  Fig. 4.6 Silver staining analysis of F-PHP-HA and PC-associated proteins in the presence of RNase A. Lanes 1 and 2 are the immuno-purified F-PC-associated proteins in the absence and presence of RNase A , respectively. Lanes 3 and 4 are the immunopurified F-PHP-HA-associated proteins in the absence and presence of RNase A . The molecular markers are indicated to the left in KDa. Two isoforms of PHP (PHP 170 and PHP 140), F L A G tagged PC (F-PC) and PC bands are indicated to the right with lines.  112  non-prominent associated proteins using immunoblotting analysis. Genetic tests of the avalilable mutants indicate that Hsc70.4 and Rpd3 are enhancers of the homeotic phenotype of ph and Pc, suggesting the significance of biochemical co-purification with P H P . R P D 3 and p55 are assoctiated with P H P but the immuno-purified proteins has no significant histone deacetylase activity. Using RNase A as a reagent in the immunopurification reaction, I also demonstrated that R N A seems not required for the integrity of both F - P H P - H A and PC-associated protein complexes. The mechanisms underlying PcG-mediated silencing are unknown. A s noted in chapter I, it has been suggested to involve histone deacetylation to create altered chromatin domains that are inaccessible to transcription factors. Recently it has been shown that the P c G proteins E S C and E ( Z ) are present in a complex containing the histone binding protein p55 and the histone deacetylase R P D 3 , and that R P D 3 is required for silencing mediated by a P R E in vivo (Tie et al., 2001). However, R P D 3 and p55 are not reported in the P R C 1 complex (Shao et al., 1999). These results also suggest that part of the mechanism mediated by P c G is due to the modification of chromatin. In Drosophila, p55 is an abundant protein and also present in at least two other chromatin remodeling complexes, C A F - 1 (Tyler et al., 1996) and N U R F (MartinezBalbas et al., 1998). The highly conserved mammalian p55 homologs, R b A p 4 8 and RbAp46, have been shown to bind to histone H 4 and H 2 A (Vermaak et al., 1999; Verreault et al., 1996). The role of p55 in the P H P complex described here is not known at present. It has been speculated p55 may target the histone deacetylase to its substrate (Tie et al., 2001). Alternatively, it may play role as an adapter between P c G and R P D 3 in the complex.  113  Chapter V TATA-binding Protein (TBP) and TBP-associated Factors (TAF„s) Associate with PHP INTRODUCTION The PcG and Gene Silencing A s reviewed in chapter I, many models have been proposed for the mechanisms of PcG-mediated silencing. These models can be categorized into two major types: chromatin packaging models and a basal transcription interference model (Pirrotta, 1998; Bienz and Muller, 1995; Paro, 1993; Brock and van Lohuizen, 2001). There exists evidence for both models. Based on the fact that P C and H P 1 , a heterochromatin protein (Paro and Hogness, 1991) share the chromodomain, it has suggested that P c G proteins may organize their target genes into higher order heterochromatin-like structures that are inaccessible to R N A polymerase II and transcriptional activators (Paro, 1990; Gaunt and Singh, 1990; Alberts and Sternglanz, 1990). Subsequently Orlando and Paro (Orlando and Paro, 1993) seemed to confirm the above model by using in vivo crosslinking and chromatin immunoprecipitation (ChIP). Later evidence suggested that P C is not distributed homogeneously on the regulatory regions of the repressed Ultrabithorax and abdominalA genes, but instead is highly enriched at discrete sequence elements (Strutt and Paro, 1997a; Orlando et al., 1998). Direct tests of chromatin accessibility yielded conflicting results. Schlossherr et al. (1994) found that P c G silencing had little effect on restriction enzyme cleavage at target loci. In addition, M c C a l l and Bender (1996) found that a reporter gene inserted within the Ubx transcription unit was efficiently silenced in parallel with Ubx itself while phage T7 R N A polymerase could still recognize its promoter  114  inserted in the same place. But B o i v i n and Dura (Boivin and Dura, 1998) showed that a P c G silenced gene was less accessible to E. coli D N A methyltransferase.  Together this  evidence suggests that the silenced chromatin is packaged differently than unsilenced chromatin to a certain extent. However, the silenced chromatin is not completely inaccessible to enzymes, or the accessibility is molecule specific. Other evidence supporting the chromatin packaging model is that histone acetylation/deacetylation is correlated with gene transcription/silencing at the Fab7 P R E (Cavalli and Paro, 1999). Alternatively, but not necessarily mutually exclusively, it has been suggested that P c G proteins maintain silencing by interfering with the basal transcriptional machinery (Bienz, 1992; Bienz and Muller, 1995; Pirrotta, 1998). Several observations support this model. P C associates with Ubx promoter in embryos only after the initiation of gene repression has occurred, suggesting that P c G presence at the promoter is correlated with silencing (Orlando et al., 1998). The bxd P R E works more efficiently when assayed together with Ubx promoter in embryos (Chan et al., 1994; Muller, 1995), and in tissue culture cells (Chang et al., 1995). Intriguingly, Laney and Biggin (1992) showed that the ability of the abx P R E to maintain embryonic silencing was abrogated i f deletions were made in the Ubx promoter.  TBP and Transcriptional Repression T B P is a general transcription factor for all three R N A polymerases (Hernandez, 1993). The S L 1 ( T B P / T A F , ) , T F D ( T B P / T A F ) , and T F B (TBP/TAF,,,) complexes n  n  n i  function at pol I, pol II,and pol Ill-transcribed promoters, respectively. T B P plays a critical role in the mechanism of transcriptional activation. For example, acidic activators  115  enhance the kinetics of T B P recruitment (Klein and Struhl, 1994), and tethering of T B P to a promoter bypasses the need for a transcriptional activator (Chatterjee and Struhl, 1995; Klages and Strubin, 1995). T F D is composed of T B P and approximately 14 distinct TBP-associated factors n  (TAF„s). T A F s were originally identified as coactivators for R N A pol II transcription n  (Goodrich and Tjian, 1994; Verrijzer and Tjian, 1996; Burley and Roeder, 1996). L i k e T B P , T A F s are present in different complexes, including T F D , yeast S A G A , human n  n  P C A F , human T F T C (reviewed in Bjorklund et al., 1999; Albright and Tjian, 2000; Pugh, 2000). Besides acting as coactivators, T A F s are also required to recognize sequence n  elements at promoters that lack T A T A boxes ( T A T A - l e s s promoters). T w o elements have been identified that are important for T A T A - l e s s promoters are the initiator (Inr) and downstream promoter element ( D P E ) (reviewed in Lee and Young, 1998). T B P is also a subunit of different repressing complexes that are functionally distinct from complexes required for activation (reviewed in Pugh, 2000). The repressive complexes include M o t l (Auble and Hahn, 1993; Chicca et al., 1998), N C 2 (Inostroza et al., 1992; Goppelt et al., 1996), N O T (Collart and Struhl, 1994; L i u et al., 1998), S A G A (yeast) (Sterner et al., 1999) or P C A F (human) (Belotserkovskaya et al., 2000). There are multiple mechanisms for repressing T B P transcriptional function, and the specific mechanism is repressor-dependent. For example, T B P self-dimerization prevents it from binding to the promoters, and thus minimizes promiscuous interaction with noninduced promoters (Coleman et al., 1995; Taggart and Pugh, 1996; Coleman and Pugh, 1997; Jackson-Fisher et al., 1999). M o t l , a member of the S N F 2 family of D N A dependent ATPases, uses the energy of A T P hydrolysis to dissociate T B P from D N A  116  (Darst et al., 2001). N C 2 , composed of two subunits each of D r l and D r a p l , binds T B P / T A T A complexes in vitro, does not appear to dissociate T B P / T A T A complexes, but instead inhibits the incorporation of T F A and T F B into the assembling transcription n  U  complex ( X i e et al., 2000). Interaction of transcriptional repressors with the R N A polymerase II holoenzyme plays crucial roles in repression. For example, the Drosophila Kruppel protein interacts in vitro with the small subunit of T F E (Sauer et al., 1995). The yeast transcriptional n  repressor T u p l interacts with the R N A polymerase II holoenzyme subunit SrblO (Zaman et al., 2001). Several repressors bind to T B P . These include the unliganded thyroid hormone receptor (Fondell et al., 1996), two homeodomain proteins, the Drosophila Even-skipped protein (Eve) ( U m et al., 1995) and the mouse M s x l protein (Zhang et al., 1996). Human M D M 2 protein, which is recruited to promoters by the P53 protein, can repress basal transcription in vitro and has been shown to interact with both T B P and T F E (Thut et al., n  1997). T o date, T B P has never been associated with a complex required for maintenance of silencing in any organism. Because of potential role for P c G proteins at the promoter, I tested the possibility that T B P or T A F s might be subunits of P c G complexes.  RESULTS and DISCUSSION TBP Copurifies with PHP and PC To test the possibility that T B P is present in the P H P complex purified in the previous chapter, I first probed a western blot, as shown in F i g . 5.1, of nuclear extract (lane 1), BR0.18(lane 2), SP0.5 (lane3), Q0.3 (lane 4), and F L A G immunoaffinity  117  purified P H complex from Q 0.3 fraction (lane 5) with anti-HA (panel A ) , and then with anti-dTBP provided by D r . James Kadonaga at U C S D (panel D ) . Interestingly, T B P is present in the F L A G purified P H complex (lane 5). To control for the non-specific binding of proteins to the immunoaffinity beads, a non-transfected K c l cell nuclear extract was mock immunopurified using the same conditions as that of the extract from F - P H P - H A transfected cells. The eluates after immunoaffinity purification were loaded onto the same S D S - P A G E gel for analysis by western blot. The results are shown in F i g . 5.1, panel D , T B P is not in the mock purified reaction (lane 6), demonstrating that T B P does not bind non-specifically to the immunoaffinity beads. To demonstrate that the purification procedure is specific for PHP-associated proteins, I probed the same membrane used above with antibody against Polycomblike provided by R i c k Jones (Fig.5.1, panel B ) . Surprisingly, P C L does not coimmunoprecipitate with P H P (lane 5), and is not present in the mock purification (lane 6), even though it is clearly detectable in previous fractions. This result argues that the association of T B P with P H P is specific, and does not result from a general affinity of transcription factors or chromatin proteins for P H P . P C and P H have previously been shown to colocalize on polytene chromosomes, to immunoprecipitate with each other (Franke et al., 1992), and to copurify from embryo extracts (Shao et al., 1999). Moreover, P C is present at the Ubx promoter in vivo (Orlando and Paro, 1998). Therefore I tested i f P C also associates with P H in vivo. I probed the same membrane with anti-PC from Renato Paro. A s shown in F i g . 5.1 panel C , P C coimmunoprecipitates with P H , but is not present in the mock purified reaction.  118  1 2  3  4  mm, PMM f " " 2? a5Sf?  F-PHP-HA  Fig. 5.1 Immunoblotting analysis of chromatography fractions and immuno-purified F PHP-HA-associated proteins. Lane 1 is the input nuclear extract. Lane 2 is BR0.18 fraction of B i o R e x column. Lane 3 is SP0.5 fraction of SP Sepharose column. Lane 4 is Q0.3 fraction of Q Sepharose column. Lane 5 is the eluate of immuno-purified F - P H P H A from Q0.3 fraction. Lane 6 is the eluate of a mock purification. The identity of proteins detected by corresponding antibodies is indicated to the right of each panel.  119  This result is consistent with previous results (see chapter I V and Franke et al., 1992), and demonstrates that P C is associated with P H P in vivo. It also can be seen from panel D of F i g . 5.1 that most T B P is not purified with P H , consistent with the previous evidence of T B P participating in many other protein complexes (see Introduction of this chapter). The relative amount of T B P associating with the affinity-purified F - P H P - H A appears to be less than the amount of P C . This difference could reflect differences in relative affinity of P C and T B P for F - P H P - H A or proteins interacting with it, heterogeneity of complexes containing F - P H P - H A , differences in subunit stoichiometry within one complex, or differences in accessibility of epitopes to the antibody, or some combination of these factors.  TBP-associated Factors (TAFs) Associate with PHP in vivo If T B P associates with P H in vivo, it is possible that some T A F s might also be associated with P H . Therefore I did a western blot on the immunopurified P H probed with different T A F antibodies. Results are shown in F i g . 5.2. Samples loaded from lane n  1 to lane 3 are nuclear extract, mock purification, and F L A G M 2 immuno-purified F P H P - H A from Q0.3 fraction, respectively. Panels A to panel E were probed with H A , D O M I N O B , T A F 8 5 , T A F 6 0 , and T A F 4 2 antibodies, respectively. Interestingly, two n  n  ir  of the three T A F s tested, T A F 8 5 and T A F 4 2 (panel C and E , respectively) associate n  n  n  with F - P H P - H A immunopurified from the Q0.3 fraction (lane 3), but not in the mock purification(lane 2). But T A F 6 0 (panel D ) , does not associate with F - P H P - H A , and is n  absent in the mock purification, suggesting that T A F 4 2 and T A F 8 5 associate n  120  n  1 2  3 ft?  B  Jm  F-PHP-HA  DOMINO B  TAF.,85 D  — »  TAFn60  5.2 T A F 8 5 and T A F n 4 2 associate with F - P H P - H A . Lane 1 is the input nuclear extract. Lane 2 is the eluate of mock purification from Q 0 . 3 fraction. Lane 3 is the eluate of immunopurified F - P H P - H A from Q 0 . 3 . The identity of protein of each panel detected by corresponding antibodies is shown to the right. Fig.  n  121  specifically with P H P . T o control for the specificity of purification, I probed the same membrane with D O M I N O B , a chromatin remodeling protein reported to enhance ph homeotic gene silencing (Ruhf et al., 2001), antibody (panel B ) . It can be seen that D O M I N O B is not present in either P H P complex or the mock purification. These data demonstrate that the association of T A F 4 2 and 85 with P H P is specific, and does not ir  result from a general affinity of transcription factors for P H P .  Coimmunoprecipitation of TBP with PHP and PC To confirm that T B P associates with P H P and P C  in vivo,  I used an anti-TBP  antibody to immunoprecipitate T B P from nuclear extracts prepared from cells expressing F - P H - H A or F - P C , detected with anti-HA and a n t i - F L A G antibodies respectively. Western blotting was used to analyze the immunoprecipitates to confirm that T B P is immunoprecipitated, and to determine if P H P and P C coimmunoprecipitate. The results are shown in F i g . 5.3. The precipitating antibody is shown above each panel, and the antibody used for the western is shown beside each panel. Input lanes in A and B were loaded with 10% of the nuclear extracts used for immunoprecipitation. T o control for non-specific binding to protein A beads, a mock IP with oc-SMRT was performed under the same conditions as that of a - T B P . It can be seen that a - T B P can coimmunoprecipitate F - P H - H A (middle lane of A ) but that oc-SMRT does not (right lane of A ) . Similarly, immunoprecipitate from F - P C nuclear extract is shown in panel B of F i g . 5.3. It also can be seen that a - T B P immunoprecipitated F - P C (middle lane of B ) but a - S M R T does not (right lane of B ) . Therefore, the reciprocal coimmunoprecipitation of P H P and P C with antibodies to endogenous T B P confirms the conclusion  122  Fig.5.3 Coimmunoprecipitation of F - P H P - H A and F - P C with T B P . (A) Samples loaded are indicated on the top. The Input lane is 10% of the nuclear extract containing F - P H P H A used for CoIP. Lane anti-TBP is the immunoprecipitate using anti-TBP. The antiS M R T lane is the immunoprecipitate using anti-SMRT. The antibodies used to probe the western blot are indicated to the right. (B) The samples loaded are indicated to the top. Input, 10% of the nuclear extract containing F - P C used for immunoprecipitation; antiT B P , the immunoprecipitate of anti-TBP; anti-SMRT, the immunoprecipitate of antiS M R T . The antibodies used to probe the western blot are indicated to the right.  123  that T B P associates with P H P and P C derived from coimmunoprecipitation experiments with epitope-tagged P H and P C . Ideally, the coimmunoprecipitation experiments would be repeated with endogenous proteins that are not over-expressed to confirm these results.  GST-TBP Fusion Protein Pulls-down PHP and PC from Nuclear Extract If T B P does participate in a complex containing P H or P C , it should be possible to pull down either P H P or P C by incubating bacterially expressed G S T - T B P fusion protein with nuclear extract. T o this end, I made a G S T - T B P fusion vector as described in the Materials and Methods. The G S T - T B P fusion protein was purified using Glutathione Sepharose beads as described in the Material and Methods. T w o p.g of G S T - T B P attached to the beads was incubated with 600 p.g of nuclear extracts of stably transformed F - P H P - H A or F - P C , respectively for 4 h at 4°C. T o control for the non-specific interaction of proteins to G S T beads, the same amount of nuclear extract was incubated with approximately the same amount of G S T beads. After centrifugation to recover proteins associated with G S T - T B P or G S T alone, the beads were washed extensively as described in the Materials and Methods, and the proteins associated were detected using western blotting. The results are shown in Fig.5.4. Panel A shows a Coommassie Blue stained gel of the input G S T and G S T fusion proteins, demonstrating that both proteins are intact, of the expected size, and largely free of contaminating bacterial proteins. Panel B shows a western blot of pull-downs probed with H A antibody to detect F - P H P H A . G S T - T B P can pull down tagged P H P from the nuclear extract, but G S T alone cannot. Panel C shows a western blot of pulls-down probed with F L A G antibody  124  B 3,  i 200-  -PHP170 -PHP140  116 C B B  97-  66 tx-HA  — -F-PC a-FLAG  Fig. 5.4 In vitro binding of P H and P C to T B P . Glutathione Sepharose beads containing bacterially expressed G S T or G S T - T B P (panel A , Coommasie-stained gel) were incubated with F - P H P - H A (panel B ) or F - P C (panel C ) containing nuclear extracts as described in the Material and Methods. After washing, western blotting was performed with these beads to analyze the proteins that had bound to G S T ( G S T lanes in B - and C panel) or G S T - T B P ( G S T - T B P lanes in B - and C-panel).  125  to detect F - P C . A s expected, G S T - T B P can also pull down tagged P C , but G S T alone cannot. These experiments support the hypothesis that T B P interacts with P H P and P C in vivo, but does not show that these proteins are found in one complex. G S T - T B P may exchange with the endogenous T B P present in complexes, allowing pull-down of P C and P H P . Alternatively, G S T - T B P may interact directly with tagged P H and P C present in the nuclear extract. If so, it should be possible to demonstrate that T B P interacts directly with P H P or P C using G S T pull-downs with bacterially synthesized T B P and P c G proteins translated in vitro. These experiments are in progress, but w i l l not be included in this thesis. These experiments provide indirect support for the hypotheses that T B P is a member of a complex containing P H P and/or P C , or that T B P interacts directly with P H P and P C .  Size Fractionation of PHP, PC, TBP, TAF s n  To see i f T B P and T A F s are present in large complex and cofractionated with n  P H P , in a preliminary experiment, I fractionated the 0.18 M fraction from the B i o R e x column reported in the previous chapter on Sephacryl S-400, a gel exclusion matrix capable of resolving high molecular weight complexes. The column was calibrated with size standards of known molecular weight as described in the Materials and Methods. The elution profiles of P H P , P C , and T B P were determined using western blotting of column fractions. A s shown in F i g . 5.5, these three proteins show overlapping elution profiles, consistent with these proteins being members of high molecular weight  126  complexes. However, the elution profiles are not identical. Therefore, the results do not demonstrate that T B P , P H P , and P C are members of one complex in vivo, as each protein could associate individually with a different multimeric complexes. Nevertheless, these results are consistent with the hypothesis that these proteins could associate in vivo, as demonstrated above for the coimmunoprecipitation experiments. To evaluate the possibilities that T B P and T A F s associate with monomeric F - P H P n  H A but are not present in a high molecular multimeric complex, it was important to show that T B P is a member of the high molecular weight complex that contains P H P . Therefore, I examined i f F - P H P - H A , P C , T B P , T A F 8 5 , and T A F 4 2 cosediment on U  n  glycerol gradients. T o this end, F - P H P - H A was purified from the transformed cell nuclear extract through three ion exchange columns, followed by F L A G immunoaffinity purification, and recovered by competitive elution with F L A G peptide as described in chapter III. The eluate was concentrated and loaded onto 15-40% glycerol gradient and the fractions were analyzed by western blotting with H A , P C , T B P , T A F 8 5 , and TAF„42 n  antibodies as described in the Materials and Methods, and the results are shown in F i g . 5.6. F - P H P - H A , P C , T B P , T A F 4 2 , and T A F 8 5 have similar sedimentation profiles, ir  n  with protein starting to be detected in fractions 19-21, and with the most protein found in fraction 29 at the bottom of the gradient. If P H P , P C , T B P , T A F 4 2 , and T A F 8 5 are n  n  members of one complex, then one would expect to see identical cosedimentation profiles. However, this is not the case. B y contrast with the other proteins P C is most abundant in fraction 27, and less abundant in fraction 29. P H P , T B P , and T A F 4 2 and n  127  Fig. 5.5 Cofractionation of F - P H P - H A , P C and T B P on gel filtration chromatography. Crude nuclear extract was fractionated on Sephacryl S-400 column, and every second fraction was analyzed by S D S - P A G E . The western blot was cut into three pieces and probed with anti-HA (panel A ) , anti-PC (panel B ) , and anti-TBP (panel C ) . The fraction numbers are indicated on the top of panel A . The molecular markers in K D a are indicated on the top with arrows. The void volume was estimated according to the instructions from the company (Amersham Pahrmacia Biotech.). The antibodies used are indicated to the right.  128  T A F 8 5 have very similar cosedimentation profiles, allowing for differences in staining n  intensity (exposure time) of the western blots. These results are consistent with the possibility that F - P H P - H A , P C , T B P , T A F 4 2 , and T A F 8 5 are members of multimeric n  n  complexes. B y inspection, it appears likely that this glycerol gradient has very low resolving power near the bottom of the gradient, as there are no fractions at the bottom of the gradient that do not contain P H P , T B P , T A F 4 2 , and T A F 8 5 . The probable n  n  explanation is that multimeric complexes containing these proteins have pelleted to the bottom of the tube because they have similar size. Under these conditions, it is not possible to conclude how many complexes containing these proteins exist, due to the low resolution. Formally, I cannot exclude the possibility that there has been aggregation of proteins that do not normally associate, and that the aggregates have pelleted. If so, then P C appears not to aggregate to the same extent. So while these results are consistent with F - P H P - H A , T B P , T A F 4 2 , and T A F 8 5 being members of multimeric complexes, they ir  n  do not unequivocally establish this fact. It would be better to use the Sephacryl S-400 column described in the previous section, which has been demonstrated to have better resolution to answer these questions. However, I did not have time to do these experiments for this thesis. Inspection of the western blots of the glycerol gradient also shows fractions that contain lower molecular weight proteins that contain each of P C , T B P , T A F 4 2 , and n  T A F 8 5 . These fractions may contain monomers or dimers of these proteins, U  subcomplexes arising from dissociation of a multimeric complex, or smaller complexes that cofractionate.  129  158kDa  Top Fraction*:  1  3  5  7  9  11 13  232kDa | 15  669kDa  Bottom  17 19 21 23 25  27  29 S I  I-^A  HA  PC  B  TBP D  TAF.,85  E  TAF.42  Fig. 5.6 Glycerol gradient analysis of immunopurified PHP-associated proteins from Q0.3 fraction. The eluate from M 2 affinity beads was concentrated, loaded onto 15-40% glycerol gradients and run as in the Material and Methods. Every other fraction was analyzed by S D S - P A G E . The western blot was cut into pieces according to the molecular weight of proteins examined. The antibodies used to probe their corresponding proteins are indicated to the right of each panel. Molecular markers are indicated towards to the fraction numbers with an arrow pointing downwards. The top fraction is to the left and the bottom fraction is to the right.  130  Tbp mutations enhance homoetic phenotypes of ph but not Pc The results so far indicate that T B P is physically associated with P H P and P C in nuclear extracts. It is important to demonstrate that this association is functionally important in vivo. The optimal experiment would be to demonstrate that PcG-mediated embryonic silencing is impaired in homozygous Tbp mutants. However, as would be expected for a gene encoding a protein required for all transcription, mutations in Tbp are cell-lethal. I examined the effect of heterozygous Tbp mutations on embryonic silencing mediated by the bxd P R E , but detected no effect (results not shown). The most likely explanation is there is a maternal contribution of T B P to the embryo that masks any effect. A s introduced in chapter I, homeotic phenotypes are enhanced between P c G gene mutations. The homeotic phenotypes of flies doubly heterozygous for two P c G mutations are much stronger that the homeotic phenotypes of either single mutant. This result suggests that P c G proteins function in the same process, because reducing the dosage of two P c G proteins has greater effects than a reduction in either gene alone. This has been interpreted as supporting evidence for the idea that P c G proteins are subunits of multimeric complexes. The homeotic phenotypes of P c G mutations arise as a result of ectopic expression of homeotic genes. For this reason, mutations in genes encoding proteins required for activation of homeotic loci suppress homeotic phenotypes of P c G mutations, because the homeotic gene is expressed less strongly, both in the normal and abnormal regions of the embryo. This is the reason that mutations of the trxG suppress homeotic phenotypes of P c G mutations.  131  Because T B P is normally considered a transcriptional activator, mutations in Tbp should suppress P c G mutations. However, i f T B P is also required for silencing as well as activation of homeotic loci, then it is possible that Tbp mutations might enhance homeotic phenotypes of P c G mutations. T o test this hypothesis, I did genetic crosses between Tbp and ph mutants. Three Tbp mutants were obtained from D r . Spyros Artavanis-Tsakonas: Tbp , Tbp ; and Tbp . Males heterozygous mutant for each s0U  sEm  sC4  of the three Tbp alleles balanced over CyO were crossed to ph and ph /FM7c 2  409  virgins,  respectively. Wild-type males have sex combs only on the first thoracic leg, and would therefore have 2.0 legs with sex combs. Mutations in ph cause a low frequency of extra sex combs. The experiment is to compare F I males mutant for ph (detectable because the X chromosome is marked with white, whereas FM7c is marked with yellow), and heterozygous for Tbp or for CyO for the presence of extra sex combs on the second and third thoracic legs. If Tbp mutants suppress this homeotic phenotype of ph, then ph; Tbp double heterozygotes would show fewer legs with extra sex combs than the ph/Y; CyO/+ males. Conversely, i f Tbp mutants enhance this homeotic phenotype, then the ph/Y; Tbp/+ males w i l l show more extra sex combs than the ph/Y; CyO/+ males. The results are shown in Table 5.1. It can be seen that all three Tbp alleles enhance the ph and ph 2  409  extra sex comb phenotype. These results are surprising, and support the  hypothesis that P H might prevent activation at the Scr promoter by T B P . However, as with all genetic experiments, these results cannot show that the effect o f Tbp mutations is direct. For example, i f the Tbp mutations lower the amount of P c G proteins present in the imaginal discs, then this would enhance the homeotic phenotype of ph mutations, but  132  Table 5.1 Tbp enhances the extra sex combs phenotype of ph Cross  Genotype  ph /ph ;+/+ x +/Y; 2  2  Tbp /CyO s012  ph /Y; CyO/+ ph /Y; Tbp /+  Number of flies 110 171  Average No. of legs with sex combs 2.31 2.74  ph /Y; CyO/+ ph /Y; Tbp /+  110 119  2.60 3.39  ph /Y; CyO/+ ph /Y; Tbp /+  100 101  2.52 3.04  ph /Y; CyO/+ ph /Y; Tbp /+  110 113  3.10 3.70  ph /Y; CyO/+ ph /Y; Tbp /+  129 166  3.20 3.80  ph /Y; CyO/+ ph /Y; Tbp /+  99 120  3.45 5.00  2  2  ph /ph ; +/+ x +/Y; 2  2  Tbp /CyO sEU4  2  2  ph /ph ; +/+ x +/Y; Tbp /CyO 2  2  sC4  409  Tbp /CyO s0,2  409  +/+ x +/Y;  Tbp /CyO sEV4  409  +/+ x +/Y; Tbp /CyO sC4  s012  409  409  ph /FM7c;  sC4  409  409  ph /FM7c;  sEU4  2  2  ph /FM7c; +/+ x +/Y;  s012  sEU4  409  409  sC4  X test was used to determine the significance of the data, P<0.001 2  133  a  a  a  a  a  a  not necessarily show that P H acts by interfering with the function of T B P . Mutations in the genes encoding T A F 4 2 and T A F 8 5 are not available, so similar experiments could ir  n  not be undertaken for these genes. However, in contrast to ph alleles, Tbp does not enhance the extra sex combs phenotype of Pc (Table 5.2). This is surprising because both P C and T B P associate with 4  F - P H P - H A as shown above. O n one side, this result suggests that the genetic enhancement of Tbp to ph is not due to a general reduction of ph expression caused by less T B P , but that the interaction is direct. Another allele of Pc, Pc™  109  also shows no  enhancement of homeotic phenotypes by Tbp mutations (data not shown), suggesting that if T B P does have a role in silencing, this role is more sensitive to changes in dosage of P H than of P C . The data described in this chapter demonstrate that T B P , T A F 4 2 and T A F 8 5 n  n  coimmunoprecipitate with P H P and P C . In addition, mutations in Tbp enhance the homeotic phenotype of ph mutations. These data are consistent with the previous observation in which P C was found localized at the Ubx promoter (Orlando et al., 1998), and in agreement with the prediction of basic transcriptional machinery involved in P c G mediated gene silencing (Bienz and Muller, 1995; Pirrotta, 1998). Whether T B P in the P c G complex is inactivated and the relationship of the compelx to gene silencing are not clear at present. But several possibilities can be predicted from what we have learned from other systems of T B P involved in gene repression. First, one obvious way is that T B P is trapped in P c G complex, making it unable to organize a productive preinitiation complex (PIC) for transcription. This is similar to one of the TBP-containing repressive complexes, N C 2 , in which T B P is disabled to interact with  134  Table 5.2 Tbp does not enhance the extra sex combs phenotype of Pc Cross  Genotype  Pc /TM3 Sb x Tbp /CyO 4  s012  Number  CyO/+; T M 3 Sb/+ CyO/+; Pc /+ Tbp/+; TM3 Sb/+ Tbp/+; Pc /+  74 132 126 125  Average No. of legs with sex combs 2.0 2.5 2.0 2.2  CyO/+; TM3 Sb/+ CyO/+; Pc /+ Tbp/+; TM3 Sb/+ Tbp/+; Pc /+  53 108 117 147  2.0 2.1 2.0 2.1  CyO/+; TM3 Sb/+ CyO/+; Pc /+ Tbp/+; TM3 Sb/+ Tbp/+; Pc /+  81 141 102 173  2.0 2.1 2.0 2.1  4  4  Pc /TM3 Sb x Tbp /CyO 4  sEU4  4  4  Pc /TM3 Sb x Tbp /CyO 4  sC4  4  4  135  T F A , and thus prevent the loading of pol II holoenzyme and the formation of P I C ( X i e et U  al., 2000). Second, it was reported that T B P can inhibit its T A T A binding by selfdimerization (Coleman et al., 1995; Coleman and Pugh, 1997). P H P contains a S A M domain that can cause homodimerization of P H P or heterodimerization of P H P and S C M (Kyba and Brock, 1998b). Therefore it is formally possible that P H P drives the T B P dimer formation and prevents binding to the promoter. Third, in the S A G A complex in which T B P is bound to two subunits of S A G A , Spt3 and Spt8, T B P is not available to other members of preinitiation complex (PIC), thus inactivating its transcription function. But when spt3 and 8 are mutated, T B P can interact with other T A F s in the same complex and initiate transcription (Belotserkovskaya et al., 2000). So it is possible that T B P in the P c G complex could be either active or inactive depending on modulation of its binding partner. A s mentioned in the introduction, T A F s form different complexes and recognize u  different promoter elements. It is especially interesting is that T A F 6 0 and 42 recognize n  the D P E of some T A T A - l e s s promoter in Drosophila (Burke and Kadonaga, 1996; Burke and Kadonaga, 1997; Kutach and Kadonaga, 2000). Most of the homeotic gene promoters are T A T A - l e s s , thus it is possible that P c G can inactivate T B P or T A F s in the n  P c G containing T A F s and T B P complex. But T A F 6 0 does not associate with P H in n  n  vivo, so the significance of this observation remains unclear. Interestingly, five TAF„s are present in the yeast S A G A and human P C A F complexes and their function is not known yet. Therefore it may be some time before the role of T A F association with P H and P C is understood.  136  In the future, it is necessary to find out which proteins bind T B P , using far westerns or pull-down experiments. Another important issue is how T B P is inactivated. T o answer this question, it is necessary to test i f T B P associated with F - P H P - H A can still bind to T A T A box by doing footprinting in vitro. If this result is obtained, one obvious possibility to test is that T B P is a dimer in the compelex. Chemical crosslinking can be used to test this hypothesis. Taken together, these data strongly suggest that one of the mechanisms by which the P c G maintains homeotic gene silencing is to inactivate T B P or inhibit the formation of functional P I C .  137  Chapter VI Conclusions  Overall, the main conclusions of my thesis are as follows: 1) B y using epitope tagging, immunopurification and mass spectrometry, I showed that chaperones are associated with F - P H P - H A in vivo. Genetic analysis shows that one of the chaperones, Hsc70.4, enhances the homeotic phenotype of ph and Pc. 2) I demonstrated using immunoblotting analysis that the histone deacetylase R P D 3 ( H D A C 1 ) and histone binding protein p55 are associated with F - P H P - H A , and by genetic analysis that RpdS mutations enhance the homeotic phenotype of ph and Pc. 3) I showed that the core subunit of basal transcription factor T B P and its-associated factors, T A F 4 2 and T A F 8 5 U  n  associate with P H P , and genetic analysis indicates that T B P may be involved in the homeotic gene silencing.  PcG Complexes The phenotypes of individual P c G mutants, while generally similar, also differ in detail, suggesting that P c G proteins form different multimeric protein complexes ( M c K e o n and Brock, 1991; Simon et al., 1992; Soto et al., 1995). Consistent with this inference, two P c G protein complexes, E S C - E ( Z ) and P R C 1 , have been purified from transgenic embryos using epitope tagging and immunoaffinity chromatography (Tie et al., 2001; N g et al., 2000; Shao et al., 1999). Some other evidence suggests that there are more than two P c G protein complexes. For example, as noted in chapter I, ph mutants also show an epidermal defect not seen with other P c G mutants (Dura et al., 1987). Furthermore, P C , P H , and P S C are differentially distributed on regulatory sequences of  138  the engrailed-related gene invected (Strutt and Paro, 1997b). These suggest that there may be multiple different Polycomb group protein complexes that function at different target sites. Interestingly, the PHP-associated proteins purified here partially overlap with the subunits of P R C 1, but are also different. P R C 1 contains P S C , P H P , S C M , P C , and six other unknown members (p276, p202, p84, p47, p43, p32, the numbers are the apparent molecular weight in K D a ) . T w o components of P R C 1 , P H P and P C , are present in the F P H P - H A associated proteins identified in this thesis, but the P c G protein P S C is not present. The F - P H P - H A associated proteins T A F 4 2 and T A F 8 5 may correspond to p43 U  n  and p84 of P R C 1, and the chaperone p48 may correspond to the p47 of P R C 1. However immunoblotting or sequencing data was not presented in Shao et al. (Shao et al., 1999) to confirm these hypotheses. The abundant proteins associated with P H P identified here are two chaperones, at least one of which, H S C 4 , has a molecular weight of 70 K D a , and appears not to be present in P R C 1 . The other less abundant proteins associated with P H P are histone deacetylase R P D 3 (apparent molecular weight is around 68 K D a ) and p55, which are apparently not present in P R C 1 . That different proteins associate with P H P , and presumably, that there are multiple complexes containing P H P , is consistent with the genetic evidence described above. This evidence suggests that it is necessary to purify more P c G protein containing complexes and to identify their associated members in the future for a better understanding of the functions of P c G . For example, although it has been shown that P H and P C colocalize 100% on polytene chromosome (Franke et al., 1992), and copurify biochemically (Shao et al., 1999 and this study), ph and Pc behave differently as noted above. In fact, P H and  139  P C present in crude extracts show different distribution on a gel filtration column although they do overlap (Fig. 5.5), and preliminary data shows that P H and P C might associate with different proteins (Fig. 4.5). Therefore it w i l l be interesting to purify P C containing complexes and to identify their members, so that one could compare with P H P containing complexes. These studies should yield new insights into fundamental aspects of P c G complexes and suggest how PcG-mediated silencing can occur.  Chaperones associate with PHP One surprising discovery is that two chaperone proteins associate with P H , as shown by two methods of purification. Genetic analysis also demonstrates that Hsc70.4 indeed enhances both ph and Pc homeotic phenotype. Intriguingly, Hsc70.4 was identified as an enhancer of Pc during a genetic screen. It was suggested that HSC70.4 might be needed for the proper folding of a component of the Polycomb repression complex, or it may be a functional member of that complex (Mollaaghababa et al., 2001). One interesting and necessary experiment that should be done in the near future is to size fractionate the immunopurified P H P containing complex to determine i f HSC70.4 associates with a high molecular weight complex containing F - P H P - H A . Because P H P , HSC70.4, P C and P48 are the most prominent members of the F - P H P - H A containing complexes, it is important to dissect the interaction network of these members, and to determine which proteins interact directly. The in vitro assays like GST-fusion protein pulldowns and Far-westerns could be used to answer this question. T o determine i f HSC70.4 binds to P R E s , and is required directly for homeotic gene silencing in vivo, embryo transformation of homeotic PRE-reporters constructs can be employed. The  140  rationale behind this experiment is that i f H S C 7 0 . 4 is involved directly in PcG-mediated silencing, it should be possible to see that HSC70.4 associates with the ectopic P R E s using immunostaining of polytene chromosomes, and to show that Hsc70.4 mutations cause the derepression of the reporter in vivo. It has been demonstrated that molecular chaperones transform the ecdysone receptor heterodimer E c R - U S P (Ecdysone Receptor-Ultraspiracle) to the D N A binding state using a gel shift assay (Arbeitman and Hogness, 2000). Similarly, i f H S C 7 0 . 4 is required for the association of P c G to polytene chromosome, then it should be possible to see the dissociation of P c G from P R E in an Hsc70.4 mutant background using chromatin immunoprecipitation.  The Transcriptional Machinery and PcG-mediated Gene Silencing Traditionally, gene silencing was viewed as epigenetic and without promoter specificity (Loo and Rine, 1995). Therefore it is not surprising that no connection has been made between the silencing machinery and the basal transcriptional machinery. However, recently it has been found that repressive, SIR-generated heterochromatin is permissive to the constitutive binding of an activator, H S F , and two components of the preinitiation complex (PIC), T B P and P o l H . The heterochromatic H M R a l promoter is also occupied by T B P and P o l II, suggesting that SIR regulates gene expression not by restricting factor access to D N A but rather by blocking a step downstream of P I C recruitment (Sekinger and Gross, 2001). In Drosophila, no direct evidence is available to support the hypothesis that P c G mediated silencing acts through the basal transcriptional machinery. This possibility was  141  suggested initially by Bienz (Bienz, 1992). Nevertheless, recent evidence has demonstrated that P C is detected outside core P R E s and binds to fragments containing promoters (Orlando et al., 1998). Additionally, Laney and Biggin (Laney and Biggin, 1992) have shown that the Ubx promoter is required for the abx PRE-mediated silencing. These observations prompted me to look for basal transcriptional factors in the P H complexes I purified. Interestingly, the basal transcription factor T B P and two T A F s n  associate with P H in vivo. The non-association of T A F 6 0 with P H , makes a strong n  argument that the biochemical co-purification is specific, and that my preparations are not contaminated with T F I I D . It is especially interesting that TBP mutations enhance the homeotic phenotype of ph but not Pc. One would expect that TBP mutations should suppress the phenotype of ph because T B P is a basal transcription factor required for gene activation. The failure to detect enhancement of Pc by TBP mutations suggests that PcG-mediated silencing at promoters specifically requires P H , or alternatively, that T B P mediated silencing is especially sensitive to changes in P H dosage. The former possibility would be supported i f further analysis shows that a complex that contains P H P , T B P , T A F 4 2 , and T A F 8 5 exists, but this is distinguishable from complexes that n  n  contain P C . Such data could be obtained by careful analysis of fractions from size exclusion columns containing high molecular weight complexes. The latter possibility is harder to address, but in principle could be approached by examining how changing amounts of P H and P C using transgenic flies under the control of inducible promoters in TBP mutant backgrounds affects homeotic phenotypes. However, compared to the two chaperones, T B P , T A F 4 2 and T A F 8 5 are less n  n  abundant, and thus can only be detected by immunoblotting, indicating that these proteins  142  are either less stably associated with P H , are present in one complex, but at lower stoichiometry than the chaperones, or most likely that these proteins are part of a less abundant and different complex than PH-Chaperone complex. T o determine which of these hypotheses is correct, immunopurified F - P H P - H A can be analyzed on gel filtration columns or glycerol gradient ultracentrifugation, followed by western blotting analysis to see i f the subunits distribute similarly or differently. A s noted in the previous chapters, the function of the association of T B P with P H P is not known, and the mechanisms of PcG-mediated silencing remains elusive. M a n y questions need to be addressed in the future. For example, can T B P in the complex still bind to the promoter? H o w does the P c G repress transcription? Is T B P present in other P c G complexes? The first question can be addressed using in vitro footprinting or immunoprecipitation of D N A fragments containing T A T A boxes. The remaining questions w i l l require further biochemical analysis of complexes containing T B P . Theoretically, the P c G protein complexes could repress promoter activity through a number of distinct mechanisms. P c G complexes could act by sterically hindering access of activators to their cognate sites within the promoter. A second possibility is that the D N A binding capability of activators is preserved, yet the subsequent recruitment of P I C is impaired, possibly through the P c G complexes interfering with the function of activation domains or by imparing access of either the T F D or P o l II holoenzyme n  complexes. Finally, P c G repressed chromatin may be permissive to both activator binding and P I C assembly, yet exert its inhibitory effect at the level of promoter clearance or polymerase elongation. However, real understanding of the mechanism  143  awaits the development of in vitro PRE-mediated silencing assays that function with reconstituted activities.  The Histone Deacetylation and PcG-mediated Gene Silencing Activation and repression of gene expression correlate with the acetylation state of histones (Workman and Kingston, 1998). The evidence of histone deacetylation involved in PcG-mediated silencing is controversial. First, Strouboulis et al. (Strouboulis et al., 1999) showed that transcriptional repression by x P c l , a Polycomb homolog in Xenopus laevis, is independent of histone deacetylase. Then van der V l a g and Otte (van der V l a g and Otte, 1999) showed that transcriptional repression mediated by the human P c G protein E E D involves histone deacetylation, but they were not able to detect H D A C s in other P c G immunoprecipitates. Consistent with this observation H D A C s do not copurify with P R C 1 (Shao et al., 1999). Recently, in Drosophila, histone deacetylase R P D 3 and histone binding protein p55 are shown to be in the E S C - E ( Z ) complex (Tie et al., 2001). However, genetic evidence showed that esc is required only for early maintenance of homeotic gene silencing, and other P c G genes are required for late maintenance (Simon, 1995; reviewed in Simon et al., 1995). So i f P H or P C complexes do not contain H D A C , then how is repressed chromatin kept in hypoacetylated state? The preliminary evidence from this thesis suggests that P H does associate with R P D 3 and histone binding protein p55. This result is the first to show that a histone deacetylase associates with P H P . Importantly, an RPD3 mutation enhances the extra sex comb phenotypes of ph and Pc. Although the genetic result could arise indirectly, the  144  result is consistent with the possibility that R P D 3 is required for PcG-mediated silencing  in vivo. In Drosophila, p55 has also been found present in chromatin assembly factor-1 (dCAF-1) (Tyler et al., 1996) and chromatin remodeling factor N U R F (Martinez-Balbas et al., 1998). T o eliminate the possibility of copurification of these complexes with P H P containing complexes, it is necessary to check whether other subunits of d C A F - 1 or N U R F are present in the purified P H P complexes. To address the question of whether R P D 3 is involved directly in PH-dependent homeotic gene silencing, the obvious next experiment is to test whether R P D 3 and P H bind directly to ectopic transgenic P R E s in vivo, and i f Rpd3 mutations derepress the reporter. It w i l l also be interesting to undertake similar experiments with p55. The findings described in the thesis form a foundation for new investigations on the P c G and the mechanisms of silencing. It is clear that identifying the interacting members of P c G complexes is just the beginning, not the end of our understanding of P c G functions. The next big goals w i l l be to understand how many P c G complexes exist, how they recognize their targets, how different complexes function in silencing, and to understand the roles of individual subunits in PcG-mediated silencing.  145  Chapter VII Material and Methods DNA Subcloning A l l subcloning procedures follow Sambrook J. et. al. (Sambrook et al., 1989). Enzymes were purchased from Boehringer Manheim or N e w England Biolabs. The sequence of all PCR-engineered constructs was confirmed by D N A sequencing ( N A P S Unit, U B C ) . Details of key constructs are given below. Details of P C R reactions are given in section on P C R . pUASTPHPHA.  The forward primer PHP1 C A A G G T G G A G T C C A T T A A G  and the reverse primer P H P 2 C A G T C T A G A C T A G A G G C T A G C G T A A T C C G G AAC ATC GTA TGG GTA G A G GCT A G C GTA ATC CGG A A C ATC GTA TGG G T A C T G C G C T C C T G G A T C C T T were used in a P C R reaction to add two H A tags (underlined sequence) at the C-terminus of phf c D N A . The engineered P H P H A was inserted into the p U A S T vector (Phelps and Brand, 1998) at E c o R l and X b a l sites. The H A peptide sequence is Y P Y D V P D Y A S L . This construct was named p U A S T P H P H A , and is illustrated in F i g . 7.1. pNTP14F-PHP-HA.  The template primer A C T G C T C G A G C A A C A T G G A C  T A C A A A G A C G A T G A C G A T A A A G A A T T C G C C G , the forward primer P I A C T G C T C G A G C A A C A T G and the reverse primer P2 C G G C G A A T T C T T T A T C G T were used to generate a peptide ( D Y K D D D D K ) with a X h o l site at its N-terminus, a E c o R l site at its C-terminus. The D N A was inserted into the X h o l and E c o R l sites of p Z O P 2 F (Hegedus et al., 1998) to make the tagged vector p Z O P 2 F . The P H P - H A in p U A S T P H P H A was excised with E c o R l and X b a l and inserted into p Z O P 2 F . Then the F - P H P - H A in p Z O P 2 F - P H P - H A was excised with K p n l and X b a l , blunted with T4 D N A  146  polymerase, inserted into C I P treated H i n d i site of p N T P 1 4 (provided by D r . L o y Volkman at U C Berkeley ). The construct was named p N T P 1 4 F - P H P - H A , and is illustrated in F i g . 2.1. pNTP14F-PC. The forward primer P c P l ( G C G A A T T C A T G A C T G G T C G A G G C A A G ) and reverse primer PcP2 ( C T G G A T C C T C A A G C T A C T G G C G A C G A ) were used to amplify a Pc c D N A from the plasmid T P c containing P c c D N A provided 3  by D r . R . Paro. The P C R product was digested with E c o R I and B a m H I, and was then inserted into the pBS(+KS) vector for sequencing. Subsequently, the Pc c D N A was digested with E c o R I and B a m H I and inserted into p Z O P 2 F . The tagged Pc was subsequently excised with X h o I and X b a I, filled-in with the large fragment of E. coli D N A polymerase I (Klenow fragment) and inserted into Hinc II site o f p N T P 1 4 . The correct orientation was confirmed by E c o R I and B a m H I digestion. This construct was named p N T P 1 4 F - P C , and is illustrated in F i g . 2.3 PGEX4T1PHO. The reverse primer PhoP2 G C T A G G A T C C T C A G T C T G C A T A T A C C A C was used for reverse transcription of Pho m R N A . Total R N A isolated from 5 x l 0 K c l cells using T R I z o l reagent ( G B 3 C O / B R L ) was used for reverse 6  transcription. Reverse transcription was performed with Superscript ( G D 3 C O / B R L ) and followed the protocol provided by the manufacturer. The reverse transcribed Pho c D N A was amplified in the P C R reaction with forward primer P h o P l ( A T C G G A A T T C A T G G C A T A C G A A C G T T T T G ) and PhoP2 as a reverse primer. P C R was performed at 94°C/45 s,48°C/45 s,72°C/2 min for 30 cycles. The resulting c D N A was digested with E c o R I and B a m H I , inserted into pBS(+KS) vector (Stratagene) and sequenced. Then the Pho c D N A was exercised with E c o R I and N o t l and inserted into p G E X 4 T l  147  (AmershamPharmacia Biotech.) to express G S T fusion protein. T o avoid the conserved zinc fingers found in the C-terminus of P H O , c D N A encoding the N-terminus (1-276 aa) was digested with E c o R I, Hinc II and cloned into E c o R l and S m a l sites of p G E X 4 T l for antibody generation. This construct was named p G E X 4 T l P H O ( 1-276). dTBP. The same strategy described above to clone the Pho c D N A was used to obtain a c D N A of T B P from total R N A of K c l cells. The forward primer d T B P P l ( C G G A A T T C A T G G A C C A A A T G C T A A G C C C and reverse primer d T B P P 2 ( C G G G A T C C T T A T G A C T G C T T C T T G A A C T T C) was used in the P C R . The resulting P C R product was digested with E c o R I and B a m H I and inserted into pBS(+KS). The sequence was confirmed using T3 and T7 primers to sequence the insert ( N A P S Unit, U B C ) . For G S T fusion construction, the full-length c D N A was excised from pBS(+KS) with E c o R I and Not I and inserted into sites of E c o R I and Not I of p G E X 4 T l .  Polymerase Chain Reaction (PCR) A standard fifty p i reaction contained 10-20 ng of plasmid template or reverse transcribed c D N A from 2 x 1 0 K c l cells, 0.3 p,g forward and reverse primers, 200 p:M 6  dNTPs, and 2.5 units Taq D N A polymerase. The reaction was performed at 94°C/45s, 54°C/45s, 72°C/2min for 30 cycles on a D N A Thermal Cycler 480 (Perkin Elmer).  Expression and Purification of GST-fusion Proteins L o g phase E. coli D H 5 a containing GST-fusion constructs regulated by the tac promoter were induced with 0.5 to 1 m M I P T G (isopropylthio-p-D-galactoside) for 7 h at 37°C or overnight at room temperature. Bacteria were collected by centrifugation at 5000  148  rpm using a G S A rotor (Sorvall DuPont), and the pellet was washed once with ice cold P B S . The bacterial pellet was resuspended in 5 volumes of lysis buffer (PBS containing I m M P M S F , 0.1% Triton X-100, I m M D T T , 0.5 mg/ml lysozyme), and incubated on ice for 30 min. Bacteria were lysed by 4 cycles of freezing in liquid N and thawing in a 2  37°C water bath. D N a s e l was added to the viscous lysate at 1 fig/ml, and then the mixture was incubated on ice for about 15 min. Then the lysate was centrifuged at 13,000 rpm for 15 min on a Sorvall SS34 rotor to collect supernatant. The supernatant was incubated with glutathione sepharose beads (Amersham Pharmacia Biotech) with rotation for 1 h at 4°C. The beads were then washed 4 times with lysis buffer without lysozyme. GST-Pho( 1-276) fusion protein, which is not soluble, was heated in 2x S D S buffer and purified from S D S - P A G E for use as an immunogen. G S T - d T B P was kept on the beads and stored in l x P B S containing I m M D T T , 0.1% N P - 4 0 , 50% glycerol at -20°C.  Antibody Generation in Rabbits and Their Purification Insoluble GST-Pho( 1-276) protein was isolated from an S D S - P A G E gel by electroelution from the excised gel band. For immunization, 0.5 to 1.0 mg purified protein was mixed with M P L ® + T D M + C W S Adjuvant System (Sigma, #M6661) to make about 1.0 ml solution. Boost injections were applied subcutaneously every four weeks. T w o weeks after the fifth boost injection, 0.5 mg antigen without adjuvant was injected through lymph nodes. Ten days later, the rabbit was bled. The antibody was purified with G S T - P h o (1-276) coupled to CNBr-activated Sepharose 4B(Amersham Pharmacia). T o couple GST-Pho( 1-276) to Sepharose 4 B , 0.5  149  g Cyanogen bromide Sepharose 4 B was treated sequentially with 1.0 m M H C I at room temperature for 15 min, 50 m M ethanolamine-HCl, pH8.4 at room temperature for 15 min, 50 m M N a H C 0 , pH8.4 at room temperature for 1 min. Then the beads were 3  washed once with 1 M propionic acid(l:14 dilution with P B S ) , pH2.5, and twice with P B S . T w o mg of gel purified G S T fusion protein was dialyzed against 2 liters of 100 m M N a H C 0 , pH8.4 overnight, and then mixed with beads treated as described above at 3  room temperature for 2 h. The coupled beads were incubated with 50 m M ethanolamineH C l , pH8.4 at room temperature for 1 h, and washed once with 100 m M N a H C 0 , p H 3  8.4, 500 m M N a C l , and twice with P B S . The affinity beads were stored in P B S with 0.03% N a N . 3  For immunoaffinity purification, Sepharose C L - 4 B (Amersham Pharmacia) precleared serum was incubated with antigen-Sepharose beads at 4°C for 2 h. The mixture was packed into a column. The column was washed twice with 50 column volumes of PBS/1 M N a C l , and twice with 50 column volumes of P B S . The column was eluted with 4 column volume (cv) cold 1 M propionic acid (1:14 dilution with H 0 ) and 2  0.5 cv fractions were collected into equal volumes of 0.5 M T r i s - H C l , pH8.0 to neutralize. Peak fractions were determined by spectrophotometry at A  2 8 0  . Protein peak  fractions were combined and dialyzed against 2 liters of P B S containing 0.05% Tween 20 ( B D H ) for 12 h to overnight at 4°C.  Cell Transfection and Selection of Stable Transformed Cell Lines K c l cells were maintained in S F - 9 0 0 I I S F M ( G I B C O / B R L ) without serum at room temperature at a density of 5 x l 0 t o 10 cells/ml. For transformation, a 4:1 molar ratio of 5  7  150  pNTP14PhHA or pNTP14Pc to pZop2F, about 10 |ig and 1 |Xg respectively, were used to cotransfect 80% confluent K c l cells in 6-well plates using 40 u.1 CellFECTIN Reagent (GEBCO/BRL) according to the manufacturer's directions. Two days after transfection, 100 to 1000 transformed cells and 2 x l 0 feeder cells (non-transformed cells) were plated 4  into 96-well plates in 100 ul of medium containing 25 (ig/ml Zeocin (Invitrogen). After two weeks of selection, wells containing a single colony of viable cells were transferred into 24-well plates with 0.5 ml medium containing 25 |ig/ml Zeocin. After the cells grew to 100% confluence, they were transferred to a 25 cm tissue culture flask (Corning). To 2  check for expression of the appropriate fusion proteins, about 10 cells were collected and 6  boiled with 2 X SDS-PAGE buffer for 5 min, and proteins were examined on an SDS-gel, prepared as described. The positive cell lines were expanded and stored in liquid nitrogen in medium containing 10% DMSO.  Nuclear Extract Preparation from Cell Lines Cells were grown to 10 cells/ml in a 6 liter glass flask rotated at 75 rpm containing 7  1 to 1.2 litres per flask. For transformed cells, C u S 0 was added to 100-200 u M to 2  4  induce expression of the tagged protein three days before making nuclear extract. Nuclear extract preparation was according to Parker C. and Topol J.(1987). A l l steps were performed at 4°C. Briefly, cells were collected by centrifugation in a GS A rotor (Sorvall) at 5000 rpm (4,080g) for 5 min, and then washed once with ice cold PBS. The washed cell volume (packed cell volume, pcv.) was measured and cells were suspended in 5 pcv buffer BFI-A ( 15mM HEPES-K , pH7.6 at 4°C,10 m M KC1, 0.1 m M E D T A , +  0.1 m M E G T A , 1.0 m M DTT, 2 m M MgCl ) containing protease inhibitors ( I m M 2  151  benzamidine-HCl, 0.5 m M PMSF, 1.0 u.g/m.1 bestatin, 1.0 ug/ml leupeptin, and 2.0 U.g/m.1 aprotinin), serine/threonine phosphatase inhibitor K F (5 mM), and tyrosine phosphatase inhibitor sodium molybdate (0.2 mM). Cells were pelleted and resuspended in 5 ml/1 culture of buffer BFI-A containg protease and phosphatase inhibitors. The cells were homogenized in a 40 ml Dounce homogenizer with pestle A using 15 strokes. A 1/10 volume of buffer BFI-B( 50 m M HEPES-K , pH7.6, 1M KC1, 30 m M M g C l , 0.1 +  2  m M E D T A , 0.1 m M E G T A , 2 m M DTT) containing protease and phosphatase inhibitors was added to adjust the tonicity and the nuclei were pelleted at 10,000 rpm (12,061g) for 10 min in an SS34 rotor. The pelleted nuclei were washed once with buffer A B (15 m M HEPES-K , pH7.6 110 m M KC1, 0.1 m M E D T A , 0.1 m M E G T A , 2 m M DTT, 5 m M +  MgCl ) containing protease and phosphatase inhibitors, and then resuspended in buffer 2  A B by homogenization with pestle B. The homogenate was mixed with l/10(v/v) 4 M ( N H ) S 0 (pH7.0), and incubated for 20 min with rotation at 4 °C. Extracted nuclei were 4  2  4  pelleted at 142,000g for 60 min (45,000 rpm on 70.1 T i rotor (Beckman), and the supernatant was recovered. ( N H ) S 0 was added to 0.3 g/ml supernatant and mixed for 4  2  4  20 min at 4°C. Precipitated protein was pelleted at 15,000 rpm (27,138g) for 20 min on SS34 rotor at 4°C. The protein pellet was dissolved in 1/2 pcv of HEMG0.1 (25mM HEPES-K , pH7.6, 0.1 m M E D T A , 0.1 m M E G T A , 1 m M DTT, 12.5 m M M g C l , 100 +  2  m M KC1, 0.1% Brij36T, 10% glycerol, and protease and phosphatase inhibitors), and dialyzed against 50-100 volumes of HEMG0.1 containing protease and phosphatase inhibitors for 12 to 15 hours with 2 buffer changes. The dialyzed nuclear extract was centrifuged at 13,000 rpm for 20 min on SS34 rotor at 4 °C. Supernatant was collected and stored at -70 °C.  152  Protein Chromatography A l l procedures were performed at 4°C, and all reagents used were equilibrated and p H adjusted at the same temperature. Column matrices were equilibrated with starting buffer using a batch method before packing columns. Ion exchange chromatography. B i o - R e x 70 Resin (Bio-Rad, #142-5852, 200-400 mesh) was equilibrated with 5 bed volume of H E M G 0 . 1 for several hours, then changed to fresh buffer. Equilibration was repeated several times until p H and conductivity were the same as that of H E M G 0 . 1 , a day before performing chromatography. The column bed dimension was 2.45x8.5 cm, and volume was 40 m l . The packing rate was 8-10 cv/h, the sample loading rate was 2 cv/h, and elution rate was 4 cv/h. Elution volume for each salt concentration was 4 cv. Nuclear extract was applied atlO mg/ml bed volume. The fraction size was 18% of cv. After elution, 20 pj of every the other fraction was mixed with 80 u.1 T B S (10 m M T r i s - H C l , pH7.5, 150 m M N a C l ) , and the O D  2 8 0  was  measured. The protein peak was pooled and the protein concentration of each peak was determined using B i o - R a d Protein Assay (# 500-0006). SP Sepharose Fast F l o w (Pharmacia, # 17-0729-01) chromatography was carried out using similar conditions to those described for B i o - R e x 70. Some minor modifications were that the column bed dimension was 1.45x6 cm, and the elution rate was 3cv/h. Q Sepharose Fast F l o w (Pharmacia, #17-0510-01) was equilibrated with T E M G 0 . 1 (25 m M T r i s - H C l , pH7.6 at 4°C, 0.1 m M E D T A , 12.5 m M M g C l , 100 m M KC1, 10% glycerol, 0.1% Brij 36T, 1 m M 2  D T T , and protease and phosphatase inhibitors). A l l other operations were the same as  153  described for Bio-Rex chromatography, except that the sample loaded was 5 mg/ml bed volume.  Gel filtration chromatography. Sephacryl S400HR (Pharmacia) was equilibrated with T E M G 0 . 3 (25 m M T r i s - H C l , pH7.6 at 4 ° C , 0.2 m M E D T A , 12.5 m M M g C l , 0.3 M 2  KC1, 0.1% Brij 36T, 10% glycerol, 1 m M D T T , with protease and phosphatase inhibitors). The column dimension was 1.2x50 cm. The column was packed with a PI pump (Pharmacia) at rate of 32 ml/h for 5 h. The bed dimension was 1.01x45 cm, and the volume was 36 ml. The column was equilibrated with T E M G 0 . 3 at rate of 3.0 ml/h (3.9 ml/cm h) overnight. 2  The column was calibrated with Blue Dextran, thyroglobulin,  ferritin, adolase and ovalbumin (Amersham Pharmacia, #17-0441-01, #17-0442-01) as instructed by the manufacturer. The sample loaded did not exceed 2% of cv (0.72 ml, 2% cv), and contained 4-5 mg protein. The sample was loaded by gravity, then the column was eluted with T E M G 0 . 3 at rate of 3.0 ml/h. Fractions contained 0.74 ml were collected. The A  2 8 0  was determined for every the other fraction. Sixty p i of every other  fraction was heated in 5xSDS loading buffer without glycerol and loaded 8% S D S - P A G E for analysis by western blot.  Immunoaffinity Purification Fifty mg crude nuclear extract or 30 mg conventional chromatography fractions were incubated with 1 ml anti- M 2 agarose beads (Sigma, #A1205) equilibrated with buffer H E G B 0 . 1 5 ( 2 5 m M H E P E S - K , pH7.6, O . l m M E D T A , 0.1 m M E G T A , 150 m M +  KC1, 10% glycerol, 0.1% Brij36T, 0.2 m M D T T , L O m M bezamidine H C I , 1.0|ig/ml  154  bestatin, l.Oug/ml leupeptin, 2.0u.g/ml Aprotinin, 0.5 mM PMSF, 5 mM KF, 0.2 mM sodium molybdate) at 4°C for 10 h. The beads were then washed 5 times with ice cold HEGB0.15 for 10 minutes each. For elution, the beads were incubated with 1 ml peptide (DYKDDDDK, Sigma, #F-3290) at final concentration of 0.25 mg/ml in HEGB0.15 at 4°C for 1 h. The elution was repeated 4 times and eluates were pooled. For subsequent H A affinity purification, the column eluate was concentrated 5 to 10 times with molecular weight cut off 10 kDa Centricon (Milipore, #4241) filters. The concentrated R  eluate was incubated with 0.2 ml H A . l 1 Affinity Matrix (BAbCo, # AFC-101P-1000) at 4°C for 4 h. The bound H A beads were then washed 5 times with ice cold HEGB0.15 for 10 min each. The bound PhHA complex was eluted with 2% SDS for 3x5 min at room temperature.  Glycerol Gradient Ultracentrifugation A 4ml linear glycerol gradient in HEMG0.1 plus protease inhibitors was prepared using a gradient maker connected to a pump (LKB Varioperperx II). Two ml of 15% glycerol in leading chamber and 2.1 ml of 40% glycerol in lagging chamber, with outflow from the 15% glycerol leading chamber was poured into 5 ml ultracentrifuge tubes at a rate of about 400 jxl/min (pump set at "2.5x10"). Four hundred ul of the anti- M  2  immunoaffinity purified complex was layered on top of a single gradient and the gradients were developed for 12 hours at 45,000 RPM (with acceleration set at "1" and deacceleration at "0") in a Beckman SW50.1 rotor. Catalase, thyroglobulin and aldolase (2 mg each) were run in parallel tubes for calibration. Four hundred ul was run as above. One hundred and sixty ul fractions were collected using P200 pipette from the top of the  155  tubes. To estimate the molecular weight of tested protein complexes, 15 pi of every fraction from the calibration tubes was mixed with 85 pi of HEMG0.1 for A  2 8 0  measurement. Ten pi of every the other fraction from the tested complex was used for Western blot analysis.  SDS-PAGE and Silver Staining SDS-PAGE was performed according to Sambrook et.al.(1989). Eight percent acrylamide gels were used throughout the project, with the exceptions indicated in the figure legends. The ratio of acrylamide to N,N -methylenebisacrylamide was 30:0.8. The gel size was 18.5x13x0.075 cm. Silver staining of gels was carried out according to Ansorge (Ansorge, 1985). Distilled water was used for all steps. Chemicals are either certified A.C.S. from Fisher or Anala'R' from BDH. Gels were stained in a 4.2 L microwave plastic container. All procedures were performed at room temperature, and the volume of solutions used was 500 ml. Molecular weight markers (Sigma, #M-3788) were loaded on each gel (312.5 ng). After gel electrophoresis, the gel was shaken in 12% TCA, 50% methanol for 30 min to overnight, followed by a wash in 10% ethanol, 5% acetic acid for 30 min. Then the gel was sensitized with precooled 0.06% KMn0 , 0.02% CuCl for 8 min. The 4  2  sensitized gel was washed 3 times with precooled 10% ethanol for 1 min, 8 min and 40 min, then one time with H 0 in a fresh tank for 30 min. The washed gel was shaken in 2  0.1% AgN0 for 45 min to 1 h, rinsed in H 0 for 30 seconds, then developed in 2% 3  2  K C0 , 0.05% HCHO (36.5-38%) until the weakest marker became visible (usually takes 2  3  about 7 min). The reaction was stopped with 1% HAc for 30 min. The gel was washed  156  in H 0 for 1 h with several changes, and dried on slab gel dryer (Savant, #SGD4050) at 2  80°C for 1 h.  Sample Preparation for In Gel Proteolytic Digestion and Mass Spectrometry The eluate from the double IP described in chapter III was run on a S D S - P A G E as described in the previous section except that the thickness of the gel, was 1.0 m m instead of 0.75 mm. The gel was stained using 0.1% Coomassie Brilliant Blue ( C B B ) R250 (Sigma, #B-0149) in 50% methanol and 10% acetic acid for 30 min. The gel was destained thoroughly using the staining solution without C B B for 3-4 h with 3 buffer changes. A razor blade was used to exercise protein bands as tightly as possible to increase the ratio of protein to the gel volume. The gel slices were stored in 1.5 m l plain Eppendorf tubes. T o control for chemical noise and generalized, non-specific protein background, an equivalent area of the same gel was exercised. The gel slices were washed in the tube 2 times for 2-3 min each with H P L C grade 50% acetonitrile in water. After discarding the supernatant from the second wash, the gel slices were left moist, but not submerged or swimming in any excess liquid. Then the gel slices were stored at - 7 0 ° C or sent to Harvard Microchemistry Facility for in gel digestion and subsequent mass spectrometry sequencing.  Histone Deacetylase (HDAC) Assay The histone H 4 N-terminus peptide (2-20), radiolabeling kit was purchased from Upstate Biotech. Labeling and purification of peptide was according to manufacturer's instructions. Briefly, 1.25 m C i of [ H] acetic acid ( N E N , 2.7 C i / m m o l , #NET-003H) was 3  157  added to 100 \ig of lyophilized histone H 4 peptide, and mixed thoroughly. Five u l of freshly prepared B O P solution [0.24 M Benzotriazol-l-yloxytris(dimethylamino) phosphonium hexafluorophosphate, 0.2 M triethylamine in acetonitrile] was added and the solution was rocked gently overnight at room temperature. The labeled peptide was purified using a supplied spin column. For the H D A C assay, a 50 u l reaction containing 25 m M T r i s - H C l , p H 8.0, 10 m M N a C l , 0.5 m M E D T A , 2 m M D T T 10% glycerol, 20,000 C P M [ H]-acetyl histone H 4 3  peptide, and a source of H D A C was incubated at 30°C for 2 h. The reaction was stopped by adding 18 u l quenching solution (1 M HC1, 0.12 M acetic acid). The released [ H]3  acetate was extracted with 500 |j,l ethyl acetate , and 400 u l ethyl acetate was added to scintillation vial containing 5 ml A C S I I (Amersham, #NACS204). Free acetate was measured as counts per minute in a scintillation counter. T o control for specificity, 150 m M sodium butyrate was added to a control reaction. T o detect dSIR2 H D A C activity, 1 m M N A D or 4 m M A T P was added.  GST Pull-down Assay Nuclear extract (0.6mg) from K c l cells stably transfected with p N T P 1 4 P C or p N T P 1 4 P H H A was incubated with G S T or G S T - d T B P sepharose beads in H E M G 0 . 1 for 4 h at 4°C. Beads were washed 4x5 min with N E B C buffer (20 m M T r i s - H C l , pH7.6 at 4°C, 150 m M N a C l , 0.5 m M E D T A , 10 m M N a F , 1 m M N a V 0 , 10 m M NaPPi) 3  4  containing protease inhibitors. Beads were then heated with 2x S D S loading buffer for 5 min at 95°C. Samples were separated in 8% S D S - P A G E . Western blots were probed with either anti-HA or a n t i - F L A G .  158  Co-Immunoprecipitation (Co-IP) One p i of polyclonal anti-TBP antibody was incubated with 5 p i protein A agarose ( P A A ) beads for 1 h at 4°C, then washed with N E B C buffer (20 m M T r i s - H C l , pH7.6 at 4°C, 150 m M N a C l , 0.5 m M E D T A , 10 m M N a F , 1 m M N a V 0 ) 3x5 min. 0.5 mg of 3  4  nuclear extract was added to the antibody bound P A A beads and incubated for 1-2 h at 4°C. Beads were washed 4x5 min with N E B C containing protease inhibitors. A n t i S M A R T bound P A A beads were used as control IP. In some cases, a n t i - F L A G or antiH A agarose was used instead, and mouse I g G agarose was used for mock IP. Washed beads were heated with 2 x S D S loading buffer and samples were electrophoresed on 8% gel for analysis by Western blot.  Western Blot and Immunostaining Routinely, proteins separated on 8% S D S - P A G E were transferred to nitrocellulose membrane (BioRad, #162-0112) using a semi-dry apparatus for 6-8 h at room temperature using Tris-glycine buffer (25 m M Tris, pH8.3, 149 m M glycine). The blot was then stained with 0.5% Ponseau S in 3% T C A to detect molecular weight markers, and blocked in 5% skim milk powder in T B S for 1 h to overnight. The filter was incubated with primary antibody in 5% powdered milk in T B S for 1 h at room temperature with shaking. The filter was washed 3x5 min with T B S T ( T B S , 0.1%Tween-20), and then secondary antibody in 5% milk in T B S T was added, and incubated for 1 h at room temperature. The membrane was washed 4x15 min at room temperature with T B S T . The N E N chemiluminescence kit ( N E N , #NEL104) was used to  159  detect bands. Kodak film (Kodak, #870-1302) was exposed to membrane for 30 seconds to 1 h to detect band recognized by the antibody. The primary antibodies used were P C (Renato Paro) at 1:5000, P H (Bob Kingston) at 1:750, P H (Jeff Simon) at 1:2000, P C L (Rick Jones) at 1:2000, P S C (Paul Adler) at 1:25, dSIN3(David Wasserman) at 1:2000, d R P D 3 (David Wasserman) at 1:2000, d M i - 2 (Peter Becker) at 1:10000, P55 (Jim Kadonaga) at 1:20000, d T B P (Jim Kadonaga) at 1:10000, dTAFII42 and dTAFII85 (Yoshihiro Nakatani) at 1:2000, dTAFII60 (Robert Tjian) at 1:25. Secondary H R P conjugated anti-rabbit and anti-mouse antibodies were from Pierce and used at 1:10000.  Fly Strains Maintenance and Genetic Crosses Flies were maintained at 25°C on standard medium containing Tegosept as a mold inhibitor. A l l alleles are described in www.flybase.org. Hsc70.4 /TM6 m  Tbp /CyO, and Tbp /CyO sEU4  B, Tbp /CyO, sC4  were obtained from D r . S. Artavanis-Tsakonas.  s0n  Rpd3 /TM3 Sb Ser was a gift of Randy Mottus. About 50 to 100 females o f the strains 303  ph /FM7c {phf null); ph /ph (viable mutant fusing phf and ph ), Pc f 409  2  2  d  4  e /TM6c (loss of s  function) were crossed to 50 to 100 Hsc70.4' /TM6 B, Rpd3 7TM3 Sb Ser, 95  Tbp /CyO,  303  sEU4  Tbp /CyO, and Tbp /CyO males, and their male progeny were scored for the presence sC4  s012  of extra sex combs. The significance of the data was determined using a X test. 2  r  160  A Hindlll BamHI Hindlll  Hindlll 8.90  , BamHI 4.60 Xbal 4.86 BamHI 5.80  B  HA: TAC CCA TAC GAT GTT CCG GAT TAC GCT AGC CTC Tyr Pro Tyr Asp Val Pro Asp Tyr Ala Ser Leu  Fig. 7.1 Sketch map of plasmid p U A S T P H P H A . ( A ) ph cDNA attached with a double H A tag at its C-terminus in frame was inserted into sites E c o R I and X b a I of vector p U A S T . The promoter is from the heat shock 70 gene (hsp70). U A S contains 5 yeast G A L 4 binding sites. The p o l y A signal is from S V 4 0 virus. The Drosophila white gene is a selection mark for transgenic flies. P 5 ' and P 3 ' are the P-element terminal repeats for chromosome integration. (B) H A peptide sequence and its corresponding nucleotide sequence used for tagging. p  161  Nomenclature AcMNP  Autographa californica mononuclear polyhedrosis virus  Arps  Actin-related proteins  ASX  Additional sex combs  BR  BioRex  bxd  bithoraxoid  CAF-1  Chromatin Assembly Factor 1  CHD  Chromo-Helicase-DNA binding  CM  Carboxymethyl  Co-IP  Co-immunoprecipitation  DEAE  Diethylaminoethyl  DPE  Downstream promoter element  ECL  Enhanced chemiluminescence  EDTA  Ethylenediaminetetraacetic  EGTA  Ethylene Glycol-bis (p-aminoethyl Ether)N,N,N'N'-tetraacetic A c i d  ESI  Electrospray ionization  ESC  Extra sex combs  ETP  Enhancer of trithorax and Polycomb  Eve  Even-skipped  E(Z)  Enhancer of zeste  GFP  Green fluorescent protein  GST  Glutathione-S-transferase  HA  Hemagglutinin  HDAC  Histone deacetylase  HEPES  N-[2-hydroxyethyl]-piperazine-N'-[2-ethanesulfonic  HML  Homothallic left  HMR  Homothallic right  HP1  Heterochromatin protein 1  HSC70.4  Heat shock 70 cognate 4  iab  infra-abdominal  Acid  162  acid]  IAC  Immunoaffinity chromatography  IEC  Ion exchange chromatography  Inr  initiator  MALDI  Matrix-assisted laser desorption ionization  Mcp  Miscadastral pigmentation  MS  Mass spectrometry  MSL  M a l e specific lethal  Mtn  Metallothionein promoter  NURF  Nucleosome remodeling factor  PC  Polycomb  PCAF  E300/CBP-associated factor  PcG  Polycomb Group  PCL  Polycomblike  pcv  Packed cell volume  PEV  Position effect variegation  PH  Polyhomeotic  ph"  ph distal gene  ph"  ph proximal gene  PHO  Pleiohomeotic  PHP  P H proximal protein  PIC  Preinitiation complex  PMSF  Phenylmethylsulfonyl fluoride  Pol II  R N A polymerase II  PRC1  Polycomb repressor complex 1  PREs  P c G response elements  PSC  Posterior sex combs  Q  Quaternary ammonium  QAE  Quaternary aminoethyl  RbAp46/48  Rb-associated protein 46/48  RPD3  Reduced potassium dependency 3  rpm  Rotation per minute  163  SAGA  Spt-Ada-Gcn4-acetyltransferase  SCM  Sex combs on midleg  Scr  Sex combs reduced  SET  Su(var)3-9-E(z)-Trx  SIN  Switch independent  SIR  Silent Information Regulator  SMRT  Silencing mediator for RAR and TR  SP  Sulphopropyl  SPM  Scm-Ph-Mbt  SRA  Steroid RNA activator  Srb  Suppressor of RNA polymerase B  SRC-1  Steroid receptor complex 1  Su(var)  Suppressor of variegation  SWI/SNF  Switch/Sucrose non-fermenting  TAF  TBP-associated factor  TBP  TATA binding protein  TFTC  TBP-free TAF containing complex  TOF  Time of flight  TPE  Telomere position effect  trxG  trithorax Group  Ubx  Ultrabithorax  YY1  Yin Yang 1  164  Bibliography Adler, P. N . , Martin, E . C , Charlton, J., and Jones, K . (1991). Phenotypic consequences and genetic interactions of a null mutation in the Drosophila Posterior Sex Combs gene. Dev Genet 12, 349-61. Ahringer, J. (2000). N u R D and SLN3 histone deacetylase complexes in development. Trends Genet 16, 351-6. A k a m , M . E . (1987). The molecular basis for metameric pattern in the Drosophila embryo. Development 101, 1-22. Akhtar, A . , Zink, D . , and Becker, P. B . (2000). Chromodomains are protein-RNA interaction modules. Nature 407, 405-9. Alberts, B . (1998). The cell as a collection of protein machines: preparing the next generation of molecular biologists. C e l l 92, 291-4. Alberts, B . , and M i a k e - L y e , R. (1992). Unscrambling the puzzle of biological machines: the importance of the details. C e l l 68, 415-20. Alberts, B . , and Sternglanz, R. (1990). Chromatin contract to silence. Nature 344, 193194. Albright, S. R., and Tjian, R. (2000). T A F s revisited: more data reveal new twists and confirm old ideas. Gene 242, 1-13. Alkema, M . J., Brock, M . , Verhoeven, E . , Otte, A . , Veer, L . J. v. t., Berns, A . , and Lohuizen, M . (1997). Identification of B mil-interacting proteins as constituents of a multimeric mammalian Polycomb complex. Genes Dev 11, 226-240. Allard, S., Utley, R. T., Savard, J., Clarke, A . , Grant, P., Brandl, C . J., Pillus, L . , Workman, J. L . , and Cote, J. (1999). N u A 4 , an essential transcription adaptor/hi stone H 4 acetyltransferase complex containing E s a l p and the ATM-related cofactor T r a l p . E M B O J 18, 5108-19. Anand, R., Bason, L . , Saedi, M . S., Gerzanich, V . , Peng, X . , and Lindstrom, J. (1993). Reporter epitopes: a novel approach to examine transmembrane topology of integral membrane proteins applied to the alpha 1 subunit of the nicotinic acetylcholine receptor. Biochemistry 32, 9975-84. Andreason, G . L . , and Evans, G . A . (1988). Introduction and expression of D N A molecules in eukaryotic cells by electroporation. BioTechniques 6, 650-660.  165  Angrand, P. O., Woodroofe, C . P., Buchholz, F., and Stewart, A . F . (1998). Inducible expression based on regulated recombination: a single vector strategy for stable expression in cultured cells. Nucleic Acids Res 26, 3263-9. Arbeitman, M . N . , and Hogness, D . S. (2000). Molecular chaperones activate the Drosophila ecdysone receptor, an R X R heterodimer. C e l l 101, 67-77. Ashburner, M . (1989). Drosophila: A Laboratory Handbook (Cold Spring Harbor, N Y , C o l d Spring Harbor Laboratory Press). Auble, D . T., and Hahn, S. (1993). A n ATP-dependent inhibitor of T B P binding to D N A . Genes Dev 7, 844-56. Barlow, D . (1995). Gametic imprinting in mammals. Science 270, 1610-1613. Belotserkovskaya, R., Sterner, D . E . , Deng, M . , Sayre, M . H . , Lieberman, P. M . , and Berger, S. L . (2000). Inhibition of T A T A - b i n d i n g protein function by S A G A subunits Spt3 and Spt8 at Gcn4-activated promoters. M o l C e l l B i o l 20, 634-47. Bienz, M . (1992). Molecular mechanisms of determination in Drosophila. Current Op Cell B i o l 4, 955-961. Bienz, M . , and Muller, J. (1995). Transcriptional silencing of homeotic genes in Drosophila. Bioessays 17, 775-784. Bjorklund, S., A l m o u z n i , G . , Davidson, I., Nightingale, K . P., and Weiss, K . (1999). Global transcription regulators of eukaryotes. C e l l 96, 759-67. Blau, H . M . , and Rossi, F . M . V . (1999). Tet B or not tet B : Advances in tetracyclineinducible gene expression. Proc Natl A c a d S c i U S A 96, 797-799. B o i v i n , A . , and Dura, J . - M . (1998). In vivo chromatin accessibility correlates with gene silencing in Drosophila. Genetics 150, 1539-1549. Bornemann, D . , M i l l e r , E . , and Simon, J. (1996). The Drosophila Polycomb group gene Sex comb on midleg (Scm) encodes a zinc finger protein with similarity to polyhomeotic protein. Development 122, 1621-1630. Boyer, L . A . , and Peterson, C . L . (2000). Actin-related proteins (Arps): conformational switches for chromatin-remodeling machines? BioEssays 22, 666-72. Bradford, M . M . (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. A n a l B i o c h e m 72, 248-54.  166  Brand, A . H . , Breeden, L . , Abraham, J., Sternglanz, R., and Nasmyth, K . (1985). Characterization of a "silencer" in yeast: a D N A sequence with properties opposite to those of a transcriptional enhancer. C e l l 41, 41-8. Brault, V . , Sauder, U . , Reedy, M . C , A e b i , U . , and Schoenenberger, C . A . (1999). Differential epitope tagging of actin in transformed Drosophila produces distinct effects on myofibril assembly and function of the indirect flight muscle. M o l B i o l C e l l 1999 Jan; 10(1): 135-49 70, 135-49. Breiling, A . , Bonte, E . , Ferrari, S., Becker, P. B . , and Paro, R. (1999). The Drosophila polycomb protein interacts with nucleosomal core particles: i n vitro v i a its repression domain. M o l C e l l B i o l 19, 8451-60. Breitenbach, M . , Ferreira, F., Jilek, A . , Swoboda, I., Ebner, C , HoffmannSommergruber, K . , Briza, P., Scheiner, O., and Kraft, D . (1996). Biological and immunological importance of Bet v 1 isoforms. A d v E x p M e d B i o l 409, 117-26. Brock, H . W . , and van Lohuizen, M . (2001). The Polycomb group—no longer an exclusive club? Curr Opin Genet Dev 11, 175-81. Brown, J. L . , M u c c i , D . , Whitely, M . , Dirksen, M . - L . , and Kassis, J. A . (1998). The Drosophila Polycomb Group gene pleiohomeotic encodes a D N A binding protein with homology to the transcription factor Y Y 1 . M o l C e l l 1, 1057-1064. Brunk, B . P., Martin, E . , and Adler, P. N . (1991a). Molecular genetics of the Posterior sex combs/Suppressor 2 ofzeste region of Drosophila: Aberrant expression of the Suppressor ofzeste gene results i n abnormal bristle development. Genetics 128, 119-132. Brunk, B . P., Martin, E . C , and Adler, P . N . (1991b). Drosophila genes Posterior sex combs and Suppressor two ofzeste encode proteins with homology to the murine bmi-1 oncogene. Nature 353, 351-353. Buckingham, S. D . , Matsuda, K . , Hosie, A . M . , Baylis, H . A . , Squire, M . D . , Lansdell, S. J., Millar, N . S., and Sattelle, B . (1996). Wild-type and insecticide-resistant homooligomeric G A B A receptors of Drosophila melanogaster stably expressed in a Drosophila cell line. Neuropharmacology 35, 1393-401. Bukau, B . , Deuerling, E . , Pfund, C , and Craig, E . A . (2000). Getting newly synthesized proteins into shape., C e l l 101, 119-22. Bukau, B . , and Horwich, A . L . (1998). The Hsp70 and Hsp60 chaperone machines. Cell 92, 351-66. Bunch, T. A . , Grinblat, Y . , and Goldstein, L . S. (1988). Characterization and use of the Drosophila metallothionein promoter in cultured Drosophila melanogaster cells. Nucleic Acids Res 16, 1043-61.  167  Bunker, C . A . , and Kingston, R. E . (1994). Transcriptional repression by Drosophila and mammalian Polycomb group proteins in transfected mammalian cells. M o l C e l l B i o l 14, 1721-1732. Burke, T. W . , and Kadonaga, J. T. (1996). Drosophila T F B D binds to a conserved downstream basal promoter element that is present in many T A T A - b o x -deficient promoters. Genes & Dev 10, 711-724. Burke, T. W . , and Kadonaga, J. T. (1997). The downstream core promoter element, D P E , is conserved from Drosophila to humans and is recognized by T A F I I 6 0 of Drosophila. Genes & D e v 11, 3020-3031. Burley, S. K . , and Roeder, R. G . (1996). Biochemistry and structural biology of transcription factor D D (TFIID). A n n u Rev Biochem 65, 769-99. Cairns, B . R., Erdjument-Bromage, H . , Tempst, P., Winston, F., and Kornberg, R. D . (1998). T w o actin-related proteins are shared functional components of the chromatin-remodeling complexes R S C and S W I / S N F . M o l C e l l 2, 639-51. Canfield, V . A . , Norbeck, L . , and Levenson, R. (1996). Localization of cytoplasmic and extracellular domains of N a , K - A T P a s e by epitope tag insertion. Biochemistry 35, 1416572. Cavalli, G . , and Paro, R. (1999). Epigenetic inheritance of active chromatin after removal of the main transactivator. Science 286, 955-8. Celis, J. E . , and Gromov, P. (2000). High-resolution two-dimensional gel electrophoresis and protein identification using western blotting and E C L detection. E X S 88, 55-67. Chan, C.-S., Rastelli, L . , and Pirotta, V . (1994). A Polycomb response element in the Ubx gene that determines an epigenetically inherited state of repression. E M B O J 13, 25532564. Chang, Y . - L . , K i n g , B . O., O'Connor, M . , Mazo, A . , and Huang, D . - H . (1995). Functional reconstruction of trans regulation of the Ultrabithorax promotor by the products of two antagonistic genes, trithorax and Polycomb. M o l C e l l B i o l 15, 66016612. Chatterjee, S., and Struhl, K . (1995). Connecting a promoter-bound protein to T B P bypasses the need for a transcriptional activation domain. Nature 374, 820-2. Chen, C , and Okayama, H . (1987). High-efficiency transformation of mammalian cells by plasmid D N A , M e l C e l l B i o l 7, 2745-52.  168  Chen, Y . T., Holcomb, C , and Moore, H . P. (1993). Expression and localization of two low molecular weight GTP-binding proteins, Rab8 and RablO, by epitope tag. Proc Natl Acad Sci U S A 90, 6508-12. Cheng, N . N . , Sinclair, D . A . R., Campbell, R. B . , and Brock, H . W . (1994). Interactions of polyhomeotic with Polycomb group genes of Drosophila melanogaster. Genetics 138, 1151-1162. Cherbas, L . , Lee, K . , and Cherbas, P. (1991). Identification of ecdysone response elements by analysis of the Drosophila Eip28/29 gene. Genes & Dev 5, 120-131. Cherbas, L . , Moss, R., and Cherbas, P. (1994). Transformation techniques for Drosophila cell lines. Methods in Cell Biology 44, 161-179. Chiang, A . , O'connor, M . B . , Paro, R., Simon, J., and Bender, W . (1995). Discrete Poly comb-binding sites in each parasegmental domain of the bithorax complex. Development 121, 1681-1689. Chicca, J. J. n., Auble, D . T., and Pugh, B . F. (1998). Cloning and biochemical characterization of T A F - 1 7 2 , a human homolog of yeast M o t l . M o l C e l l B i o l 18, 170110. Chittum, H . S., Lane, W . S., Carlson, B . A . , Roller, P. P., Lung, F . D . , Lee, B . J., and Hatfield, D . L . (1998). Rabbit beta-globin is extended beyond its U G A stop codon by multiple suppressions and translational reading gaps. Biochemistry 37, 10866-70. Cockell, M . , and Gasser, S. M . (1999). Nuclear compartments and gene regulation. Curr Opin Genet Dev 9, 199-205. Coleman, R. A . , and Pugh, B . F . (1997). Slow dimer dissociation of the T A T A binding protein dictates the kinetics of D N A binding. Proc Natl A c a d S c i U S A 94, 7221-6. Coleman, R. A . , Taggart, A . K . , Benjamin, L . R., and Pugh, B . F. (1995). Dimerization of the T A T A binding protein. J B i o l Chem 270, 13842-9. Collart, M . A . , and Struhl, K . (1994). N O T l ( C D C 3 9 ) , N O T 2 ( C D C 3 6 ) , N O T 3 , and N O T 4 encode a global-negative regulator of transcription that differentially affects T A T A element utilization. Genes Dev 8, 525-37. Colosimo, A . , Goncz, K . K . , Holmes, A . R., Kunzelmann, K . , N o v e l l i , G . , Malone, R. W . , Bennett, M . J., and Gruenert, D . (2000). Transfer and expression of foreign genes in mammalian cells. BioTechniques 29, 314-331. Crane, M . S. J. (1999). Mutagenesis and cell transformation in cell culture. Methods in Cell Science 21, 245-253.  169  Cumberledge, S., Zaratzian, A . , and Sakonju, S. (1990). Characterization of two R N A s transcribed from the cis-regulatory region of the abd-A domain within the Drosophila bithorax complex. Proc Natl A c a d Sci U S A 87, 3259-63. Darst, R. P., Wang, D . , and Auble, D . T. (2001). M O T 1-catalyzed T B P - D N A disruption: uncoupling D N A conformational change and role of upstream D N A . E M B O J 20, 202840. de Hoffmann, E . (1996). Tandem mass spectrometry: a primer. J Mass Spectrometry 31, 129-137. Dear, T. N . , Hainzl, T., Folio, M . , Nehls, M . , Wilmore, H . , Matena, K . , and Boehm, T. (1997). Identification of interaction partners for the basic-helix-loop-helix protein E47. Oncogene 14, 891-8. Deatrick, J., Daly, M . , Randsholt, N . B . , and Brock, H . W . (1991). The complex genetic locus polyhomeotic in Drosophila melanogaster potentially encodes two homologous zinc-finger proteins. Gene 105, 185-195. Debs, R., Brunette, E . , Fuchs, H . , L i n , E . , Shah, M . , Hargis, A . , and Montgomery, A . B . (1990). Biodistribution, tissue reaction, and lung retention of pentamidine aerosolized as three different salts. A m Rev Respir Dis 142, 1164-7. DeCamillis, M . , and Brock, H . W . (1994). Expression of the polyhomeotic locus in development of Drosophila melanogaster. Roux's A r c h Dev B i o l 203, 429-438. DeCamillis, M . A . , Cheng, N . , Pierre, D . , and Brock, H . W . (1992). The polyhomeotic gene of Drosophila encodes a chromatin protein that shares polytene chromosome binding sites with Polycomb. Genes Dev 6, 223-232. Delmas, V . , Stokes, D . G . , and Perry, R. P. (1993). A mammalian D N A - b i n d i n g protein that contains a chromodomain and an SNF2/SWI2-like helicase domain. Proc Natl A c a d Sci U S A 90, 2414-8. Dhalluin, C , Carlson, J . E . , Zeng, L . , He, C , Aggarwal, A . K . , and Zhou, M . M . (1999). Structure and ligand of a histone acetyltransferase bromodomain. Nature 399, 491-6. Dhawan, J., Rando, T. A . , Elson, S. L . , Bujard, H . , and Blau, H . M . (1995). Tetracyclineregulated gene expression following direct gene transfer into mouse skeletal muscle. Somat C e l l M o l Genet 21, 233-40. Dietzel, S., Niemann, H . , Bruckner, B . , Maurange, C , and Paro, R . (1999). The nuclear distribution of Polycomb during Drosophila melanogaster development shown with a G F P fusion protein. Chromosoma 108, 83-94.  170  Dignam, J. D . , Lebovitz, R . M . , and Roeder, R. G . (1983). Accurate transcription initiation by R N A polymerase II in a soluble extract from isolated mammalian nuclei. N u c l Acids Res 11, 1475-1489. Dobie, K . W . , Lee, M . , Fantes, J . A . , Graham, E . , Clark, A . J., Springbett, A . , Lathe, R., and McClenaghan, M . (1996). Variegated transgene expression in mouse mammary gland is determined by the transgene integration locus. Proc Natl A c a d S c i U S A 93, 6659-64. Ducker, C . E . , and Simpson, R. T. (2000). The organized chromatin domain of the repressed yeast a cell-specific gene S T E 6 contains two molecules of the corepressor T u p l p per nucleosome. E M B O J 19, 400-9. Dura, J . - M . , Brock, H . W . , and Santamaria, P. (1985). Polyhomeotic: a gene in Drosophila melanogaster required for correct expression of segment identity. M o l Gen Genet 198, 220-231. Dura, J . - M . , and Ingham, P. W . (1988). Tissue- and stage-specific control of homeotic and segmentation gene expression in Drosophila embryos by the polyhomeotic gene. Development 103, 733-741. Dura, J . - M . , Randsholt, N . B . , Deatrick, J., Erk, I., Santamaria, P., Freeman, D . J., Freeman, S. F., Weddell, D . , and Brock, H . W . (1987). A complex genetic locus, polyhomeotic, is required for segmental specification and epidermal development i n D . melanogaster. C e l l 57, 829-839. Echalier, G . (1997). Drosophila cells i n culture (San Diego, Academic Press). Echalier, G . , and Ohanessian, A . (1969). Isolement, en cultures in vitro, de lignees cellulaires diploides de Drosophila melanogaster. C R A c a d S c i Ser III S c i V i e 268D, 1771-1773. Eddy, S. (1999). Noncoding R N A genes. Curr Opin Genet Dev 9, 695-699. Englard, S., and Seifter, S. (1990). Precipitation techniques. Methods i n Enzymology 182, 285-306. Erdmann, V . A . , Barciszewska, M . Z . , Szymanski, M . , Hochberg, A . , de Groot, N . , and Barciszewski, J . (2001). The non-coding R N A s as riboregulators. Nucleic Acids Res 29, 189-93. Eshed, Y . , Baum, S. F . , and Bowman, J. L . (1999). Distinct mechanisms promote polarity establishment in carpels of Arabidopsis. C e l l 99, 199-209. Fauvarque, M . - F . , and Dura, J . - M . (1993). polyhomeotic regulatory sequences induce developmental regulator-dependent variegation and targeted P-element insertions i n Drosophila. Genes Dev 7, 1508-1520.  171  Fauvarque, M . - O . , Zuber, V . , and Dura, J . - M . (1995). Regulation of polyhomeotic transcription may involve local changes in chromatin activity in Drosophila. M e c h Dev 52, 343-355. Feigner, P. L . , Gadek, T. R., H o l m , M . , Roman, R., Chan, H . W . , Wenz, M . , Northrop, J. P., Ringold, G . M . , and Danielsen, M . (1987). Lipofection: a highly efficient, lipidmediated DNA-transfection procedure. Proc Natl A c a d S c i U S A 84, 7413-7. Field, J., Nikawa, J., Broek, D . , MacDonald, B . , Rodgers, L . , Wilson, I. A . , Lerner, R. A . , and Wigler, M . (1988). Purification of a RAS-responsive adenylyl cyclase complex from Saccharomyces cerevisiae by use of an epitope addition method. M o l C e l l B i o l 8, 2159-65. Fondell, J. D . , Brunei, F., Hisatake, K . , and Roeder, R. G . (1996). Unliganded thyroid hormone receptor alpha can target T A T A - b i n d i n g protein for transcriptional repression. M o l Cell B i o l 16, 281-7. Franke, A . , DeCamillis, M . A . , Zink, D . , Cheng, N . , Brock, H . W . , and Paro, R . (1992). Polycomb and polyhomeotic are constituents of a multimeric protein complex in chromatin of Drosophila melanogaster. E M B O J 11, 2941-2950. Fritsch, C , Brown, J. L . , Kassis, J . A . , and Muller, J. (1999). The D N A - b i n d i n g Polycomb group protein Pleiohomeotic mediates silencing of a Drosophila homeotic gene. Development 126, 3905-3913. Fritze, C . E . , and Anderson, T. R . (2000). Epitope tagging: General method for tracking recombinant proteins. Methods in Enzymology 327, 3-16. Fusseneger, M . , Bailey, J . E . , Hauser, H . , and Mueller, P. P. (1999). Genetic optimization of recombinant glycoprotein production by mammalian cells. Trends i n Biotech 17, 3542. Gao, X . , and Huang, L . (1995). Cationic liposome-mediated gene transfer. Gene Ther 2, 710-722. Garzino, V . , Pereira, A . , Laurenti, P., Graba, Y . , Levis, R. W . , L e Parco, Y . , and Pradel, J. (1992). Cell lineage-specific expression of modulo, a dose-dependent modifier of variegation in Drosophila. E M B O J i i , 4471-9. Gaunt, S. J., and Singh, P. B . (1990). Homeogene expression patterns and chromosomal imprinting. Trends Genet 6, 208-212. Gildea, J. J., Lopez, R., and Shearn, A . (2000). A screen for new trithorax group genes identified little imaginal discs, the Drosophila melanogaster homologue of human retinoblastoma binding protein 2. Genetics 156, 645-63.  172  Gindhart, J. G . , and Kaufman, T. C . (1995). Identification of Polycomb and trithorax group responsive elements in the regulatory region of the Drosophila homeotic gene Sex combs reduced. Genetics 139, 797-814. Glicksman, M . A . , and Brower, D . L . (1988). Misregulation of homeotic gene expression in Drosophila larvae resulting from mutations at the extra sex combs locus. Dev B i o l 126, 219-227. Goeddel, D . V . (1990). Systems for heterologous gene expression. Methods in Enzymology 185, 3-7. Goodrich, J. A . , and Tjian, R. (1994). T B P - T A F complexes: selectivity factors for eukaryotic transcription. Curr Opin C e l l B i o l 6, 403-9. Goodson, H . V . , Anderson, B . L . , Warrick, H . M . , Pon, L . A . , and Spudich, J . A . (1996). Synthetic lethality screen identifies a novel yeast myosin I gene ( M Y 0 5 ) : myosin I proteins are required for polarization of the actin cytoskeleton. J C e l l B i o l 133, 1277-91. Goppelt, A . , Stelzer, G . , Lottspeich, F., and Meisterernst, M . (1996). A mechanism for repression of class II gene transcription through specific binding of N C 2 to T B P promoter complexes v i a heterodimeric histone fold domains. E M B O J 75, 3105-16. Gould, A . (1997). Functions of mammalian Polycomb group and trithorax group related genes. Curr Opin Genet Dev 7, 488-494. Gray, S., and Levine, M . (1996). Transcriptional repression in development. Curr Opin Cell B i o l 8, 358-364. Gray, S., Szymanski, P., and Levine, M . (1994). Short range repression permits multiple enhancers to function autonomously within a complex promoter. Genes & D e v 8, 18291838. Gray, S. G . , and Ekstrom, T. J. (2001). The human histone deacetylase family. Experimental C e l l Research 262, 75-83. Griffiths, W . J., Jonsson, A . P., L i u , S., R a i , D . K . , and Wang, Y . (2001). Electrospray and tandem mass spectrometry in biochemistry. Biochem J 355, 545-61. Grunstein, M . (1998). Yeast heterochromatin: regulation of its assembly and inheritance by histones. C e l l 93, 325-328. Gunster, M . J., Satijn, D . P. E . , Hamer, K . M . , Blauwen, J. L . v. d., Bruijn, D . d., Alkema, M . J., Lohuizen, M . v., Driel, R . v., and Otte, A . P. (1997). Identification and characterization of interactions between the vertebrate Polycomb-Group protein B M I 1 and human homologs of Polyhomeotic. M o l Cell B i o l 77, 2326-2335.  173  Gutjahr, T., Frei, E . , Spicer, C , Baumgartner, S., White, R . A . H . , and N o l l , M . (1995). The Polycomb-group gene, extra sex combs encodes a nuclear member of the W D - 4 0 repeat family. E M B O J 14, 4296-4306. Gvozdev, V . A . , and Kakpakov, V . T. (1968). Culture of embryonic cells of Drosophila melanogaster in vitro. Genetika4, 129-142. Hagstrom, K . , Muller, M . , and Schedl, P. (1997). A Polycomb and G A G A dependent silencer adjoin the Fab-7 boundary in the Drosophila bithorax complex. Genetics 146, 1365-1380. Hammarberg, B . , Nygren, P. A . , Holmgren, E . , Elmblad, A . , Tally, M . , Hellman, U . , Moks, T., and Uhlen, M . (1989). Dual affinity fusion approach and its use to express recombinant human insulin-like growth factor II. Proc Natl A c a d S c i U S A 86, 4367-71. Handley-Gearhart, P. M . , Stephen, A . G . , Trausch-Azar, J. S., Ciechanover, A . , and Schwartz, A . L . (1994). Human ubiquitin-activating enzyme, E l . Indication of potential nuclear and cytoplasmic subpopulations using epitope-tagged c D N A constructs. J B i o l Chem 269, 33171-8. Hanna-Rose, W . , and Hansen, U . (1996). Active repression mechanisms of eukaryotic transcription repressors. Trends Genet 12, 229-234. Harata, M . , Oma, Y . , M i z u n o , S., Jiang, Y . W . , Stillman, D . J., and Wintersberger, U . (1999). The nuclear actin-related protein of Saccharomyces cerevisiae, Act3p/Arp4, interacts with core histones. M o l B i o l C e l l 10, 2595-605. Harlow, E . , and Lane, D . (1988). Antibodies: A laboratory manual. (Cold Spring Harbor, C o l d Spring Harbor Laboratory). Haiti, F . U . (1996). Molecular chaperones in cellular protein folding. Nature 381, 571-9. Hayflick, L . , and Moorhead, P. S. (1961). The serial cultivation of human diploid cell strains. E x p C e l l Res 25, 585-621. Hegedus, D . D . , Pfeifer, T. A . , Hendry, J., Theilmann, D . A . , and Grigliatti, T . A . (1998). A series of broad host range shuttle vectors for constitutive and inducible expression of heterologous proteins in insect cell lines. Gene 207, 241-249. Hegedus, D . D . , Pfeifer, T. A . , Theilmann, D . A . , Kennard, M . L . , Gabathuler, R . , Jefferies, W . A . , and Grigliatti, T. A . (1999). Differences in the expression and localization of human melanotransferrin in lepidopteran and dipteran insect cell lines. Protein Expr Purif 15, 296-307.  174  Heiermann, R., and Pongs, O. (1985). In vitro transcription with extracts of nuclei of Drosophila embryos. Nucleic Acids Res 13, 2709-30. Henderson, J., Atkinson, A . E . , Lazarus, C . M . , Hawes, C . R., Napier, R . M . , Macdonald, H . , and K i n g , L . A . (1995). Stable expression of maize auxin-binding protein in insect cell lines. F E B S Lett 371, 293-6. Henikoff, S. (1994). A reconsideration of the mechanism of position-effect. Genetics 138, 1-5. Henzel, W . J., B i l l e c i , T. M . , Stults, J. T., Wong, S. C . , Grimley, C . , and Watanabe, C . (1993). Identifying proteins from two-dimensional gels by molecular mass searching of peptide fragments in protein sequence databases. Proc Natl A c a d S c i U S A 90, 5011-5. Hernan, R., Heuermann, K . , and Brizzard, B . (2000). Multiple epitope tagging of expressed proteins for enhanced detection. Biotechniques 28, 789-93. Hernandez, N . (1993). T B P , a universal eukaryotic transcription factor? Genes Dev 1993 Jul;7(7B): 1291-308 7, 1291-308. Hickman, H . D . , Batson, C . L . , Prilliman, K . R., Crawford, D . L . , Jackson, K . L . , and Hildebrand, W . H . (2000). C-terminal epitope tagging facilitates comparative ligand mapping from M H C class I positive cells. Human Immunology 61, 1339-46. Himmelhoch, S. R . (1971). Chromatography of proteins on ion-exchange adsorbents. Methods in Enzymology 22, 273-286. Hing, H . K . , Bangalore, L . , Sun, X . , and Artavanis-Tsakonas, S. (1999). Mutations in the heatshock cognate 70 protein (hsc4) modulate Notch signaling. Eur J C e l l B i o l 78, 690-7. Hodgson, J. W . , Argiropoulos, B . , and Brock, H . W . (2001). Site-Specific Recognition of a 70-Base-Pair Element Containing d(GA)n Repeats Mediates bithoraxoid Polycomb Group Response Element-Dependent Silencing. M o l C e l l B i o l 21, 4528-4543. Hodgson, J . W . , Cheng, N . N . , Sinclair, D . A . R., K y b a , M . , Randsholt, N . B . , and Brock, H . W . (1997). The polyhomeotic locus of Drosophila melanogaster is transcriptionally and post-transcriptionally regulated during embryogenesis. M e c h Dev 66, 69-81. Holmes, K . C , Sander, C , and Valencia, A . (1993). A new A T P - b i n d i n g fold i n actin, hexokinase and Hsc70. Trends Cell B i o l 3, 53-59. Horard, B . , Tatout, C , Poux, S., and Pirrotta, V . (2000). Structure of a Polycomb Response Element and in vitro binding of Polycomb group complexes containing G A G A factor. M o l Cell B i o l 20, 3187-3197.  175  Hurley, J. (1996). The sugar kinase/heat shock protein family: implications of conserved structure for mechanism. A n n u Rev Biophys B i o m o l Struct 25, 137-162. Imai, S., Armstrong, C . M . , Kaeberlein, M . , and Guarente, L . (2000). Transcriptional silencing and longevity protein Sir2 is an NAD-dependent histone deacetylase. Nature 403, 795-800. Inostroza, J. A . , Mermelstein, F . H . , H a , I., Lane, W . S., and Reinberg, D . (1992). D r l , a T A T A - b i n d i n g protein-associated phosphoprotein and inhibitor of class II gene transcription. C e l l 70, 477-89. Jack, G . W . (1994). Immunoaffinity Chromatography. Molecular Biotechnology 1, 5986. Jackson-Fisher, A . J., Chitikila, C , Mitra, M . , and Pugh, B . F. (1999). A role for T B P dimerization in preventing unregulated gene expression. M o l C e l l 3, 717-27. Jacobson, R. H . , Ladurner, A . G . , K i n g , D . S., and Tjian, R. (2000). Structure and function of a human TAFII250 double bromodomain module. Science 288, 1422-5. Jarvik, J. W . , Adler, S. A . , Telmer, C . A . , Subramaniam, V . , and Lopez, A . J. (1996). CD-tagging: a new approach to gene and protein discovery and analysis. Biotechniques 20, 896-904. Jarvik, J. W . , and Telmer, C . A . (1998). Epitope tagging. A n n u Rev Genet 32, 601-18. Jarvis, D . L . , Weinkauf, C , and Guarino, L . A . (1996). Immediate-early baculovirus vectors for foreign gene expression in transformed or infected insect cells. Protein Expr Purif 8, 191-203. Jiang, Y . W . , and Stillman, D . J. (1996). Epigenetic effects on yeast transcription caused by mutations in an actin-related protein present in the nucleus. Genes D e v 10, 604-19. Johansen, H . , van der Straten, A . , Sweet, R., Otto, E . , Maroni, G . , and Rosenberg, M . (1989). Regulated expression at high copy number allows production of a growthinhibitory oncogene product in Drosophila Schneider cells. Genes & D e v 3, 882-889. Johnson, A . D . (1995). The price of repression. C e l l 1995 Jun 2;81(5):655-8 81, 655-8. Johnson, C . A . , Barlow, A . L . , and Turner, B . M . (1998). Molecular cloning of Drosophila melanogaster c D N A s that encode a novel histone deacetylase d H D A C 3 . Gene 221, 127-134. Jones, C . A . , N g , J., Peterson, A . J., Morgan, K . , Simon, J., and Jones, R. S. (1998). The Drosophila esc and E(z) proteins are direct partners in Polycomb Group-mediated repression. M o l C e l l B i o l 18, 2825-2834.  176  Jones, R. S., and Gelbart, W . M . (1990). Genetic analysis of the Enhancer ofzeste locus and its role i n gene regulation in Drosophila melanogaster. Genetics 126, 185-199. Jones, R. S., and Gelbart, W . M . (1993). The Drosophila Polycomb-group gene Enhancer ofzeste shares a domain of sequence similarity with trithorax. M o l C e l l B i o l 13, 63576366. Joyce, K . A . , Atkinson, A . E . , Bermudez, I., Beadle, D . J., and K i n g , L . A . (1993). Synthesis of functional G A B A A receptors in stable insect cell lines. F E B S Lett 335, 614. Jurgens, G . (1985). A group of genes controlling the spatial expression of the bithorax complex in Drosophila. Nature 316, 153-155. Kabsch, W . , and Holmes, K . C . (1995). The actin fold. F A S E B J 9, 167-174. Kakpakov, V . T., Gvozdev, V . A . , Platova, T. P., and Polukarova, L . G . (1969). In vitro establishment of embryonic cell lines of Drosophila melanogaster. Genetika 5, 67-75. Kapoun, A . M . , and Kaufman, T. C . (1995). Regulatory regions of the homeotic gene proboscipedia are sensitive to chromosomal pairing. Genetics 140, 643-658. Karpen, G . H . (1994). Position-effect variegation and the new biology of heterochromatin. Curr Op Genet Dev 4, 281-291. Kassis, J . A . (1994). Unusual properties of regulatory D N A from the Drosophila engrailed gene: three "pairing-sensitive" sites within a 1.6kb region. Genetics 136, 10251038. Kaufman, R. J. (1990). Vectors used for expression in mammalian cells. Methods i n Enzymology 185, 487-512. Kaufman, T. C , Lewis, R., and Wakimoto, B . (1980). Cytogenetic analysis of chromosome 3 in Drosophila melanogaster: the homeotic gene complex in polytene interval 8 4 A - B . Genetics 94, 115-133. Kehle, J., Beuchle, D . , Treuheit, S., Christen, B . , Kennison, J. A . , Bienz, M . , and Muller, J. (1998). d M i - 2 , a Hunchback-interacting protein that functions in Polycomb repression. Science 282, 1897-1900. Kelley, W . L . (1998). The J-domain family and the recruitment of chaperone power. Trends Biochem S c i 23, 222-227. Kennison, J. A . , and Russell, M . A . (1987). Dosage-dependent modifiers of homeotic mutations in Drosophila melanogaster. Genetics 116, 75-86.  177  Keys, D . A . , Lee, B . S., Dodd, J. A . , Nguyen, T. T., V u , L . , Fantino, E . , Burson, L . M . , Nogi, Y . , and Nomura, M . (1996). Multiprotein transcription factor U A F interacts with the upstream element of the yeast R N A polymerase I promoter and forms a stable complex. Genes D e v 10, 887-903. Kienlen-Campard, P., Tasiaux, B . , and Octave, J. N . (2000). The processing and biological function of the human amyloid precursor protein ( A P P ) : lessons from different cellular models. E x p Gerontol 35, 843-50. K i n g , L . A . , and Possee, R. D . (1992). The baculovirus expression system: a laboratory manual. (London, Chapman and Hall). Kingston, R. E . , and Narlikar, G . J. (1999). ATP-dependent remodeling and acetylation as regulators of chromatin fluidity. Genes Dev 13, 2339-52. Klages, N . , and Strubin, M . (1995). Stimulation of R N A polymerase II transcription initiation by recruitment of T B P in vivo. Nature 374, 822-3. Klein, C , and Struhl, K . (1994). Increased recruitment of T A T A - b i n d i n g protein to the promoter by transcriptional activation domains in vivo. Science 266, 280-2. Kodjabachian, L . , Delaage, M . , Maurel, C , Miassod, R., Jacq, B . , and Rosset, R. (1998). Mutations in ccf, a novel Drosophila gene encoding a chromosomal factor, affect progression through mitosis and interact with P c - G mutations. E M B O J 17, 1063-75. Kohler, G . , and Milstein, C . (1975). Continuous cultures of fused cells secreting antibody of predifined specificity. Nature 256, 495-497. Krasnow, M . A . , Saffman, E . E . , Kornfeld, K . , and Hogness, D . (1989). Transcriptional activation and repression by Ultrabithorax proteins in cultured Drosophila cellls. Cell 57, 1031-1043. Kutach, A . K . , and Kadonaga, J. T. (2000). The downstream promoter element D P E appears to be as widely used as the T A T A box in Drosophila core promoters. M o l Cell  B i o l 20, A15A-A16A. K y b a , M . , and Brock, H . W . (1998a). The Drosophila Polycomb Group protein Psc contacts ph and Pc through specific conserved domains. M o l C e l l B i o l 18, 2712-2720. Kyba, M . , and Brock, H . W . (1998b). The S A M domain of polyhomeotic, rae28 and S c m mediates specific interactions through conserved residues. Dev Genet 22, 74-84. LaJeunesse, D . , and Shearn, A . (1996). E(z). a polycomb group gene or a trithorax group gene? Development 122, 2189-2197.  178  Lambert, J. R., and Nordeen, S. K . (1998). Steroid-selective initiation of chromatin remodeling and transcriptional activation of the mouse mammary tumor virus promoter is controlled by the site of promoter integration. J B i o l C h e m 273, 32708-14. Landecker, H . L . , Sinclair, D . A . R., and Brock, H . W . (1994). Sscreen for enhancers of Polycomb and Polycomblike in Drosophila melanogaster. Dev Genet 15, 325-434. Landry, J., Sutton, A . , Tafrov, S. T., Heller, R. C , Stebbins, J., Pillus, L . , and Stenglanz, R. (2000). The silencing protein SIR2 and its homologs are NAD-dependent protein deacetylases. Proc Natl A c a d S c i U S A 97, 5807-5811. Laney, J. D . , and Biggin, M . D . (1992). zeste, a non-essential gene, potently activates Ultrabithorax transcription in the Drosophila embryo. Genes Dev 6, 1531-1541. Lanz, R. B . , M c K e n n a , N . J., Onate, S. A . , Albrecht, U . , Wong, J., Tsai, S. Y . , Tsai, M . J., and O'Malley, B . W . (1999). A steroid receptor coactivator, S R A , functions as an R N A and is present in an S R C - 1 complex. C e l l 97, 17-27. Lee, J., and Jaenisch, R. (1997). The epigenetic control of mammalian X-chromosome inactivation. Curr Op Genet Dev 7, 274-280. Lee, R. J., and Huang, L . (1997). Lipidic vector systems for gene transfer. Crit Rev Ther Drug Carrier Syst 14, 173-206. Lee, T. I., and Young, R. A . (1998). Regulation of gene expression by TBP-associated proteins. Genes & Dev 12, 1398-1408. Lefstin, J. A . , and Yamamoto, K . R. (1998). Allosteric effects of D N A on transcriptional regulators. Nature 392, 885-8. Lehel, C , Olah, Z . , Petrovics, G . , Jakab, G . , and Anderson, W . B . (1996). Influence of various domains of protein kinase C epsilon on its P M A - i n d u c e d translocation from the Golgi to the plasma membrane. Biochem Biophys Res Commun 223, 98-102. Levine, M . , and Manley, J. L . (1989). Transcriptional repression of eukaryotic promoters. Cell 59, 405-408. Lewis, E . B . (1978). A gene complex controlling segmentation in Drosophila. Nature 276, 565-570. Lipshitz, H . D . , Peattie, D . A . , and Hogness, D . S. (1987). N o v e l transcripts from the Ultrabithorax domain of the bithorax complex. Genes Dev 1, 307-22. L i u , H . Y . , Badarinarayana, V . , Audino, D . C , Rappsilber, J., Mann, M . , and Denis, C . L . (1998). The N O T proteins are part of the C C R 4 transcriptional complex and affect  179  gene expression both positively and negatively. E M B O J 1998 Feb 16;17(4): 1096-106 17, 10960-106. Lohuizen, M . v. (1999). The trithorax-group and Polycomb-group chromatin modifiers: implications for disease. Curr Op Genet Dev 9, 355-361. Lonie, A . , D'Andrea, R., Paro, R., and Saint, R. (1994). Molecular characterisation of the Polycomblike gene of Drosophila melanogaster, a trans-acting negative regulator of homeotic gene expression. Development 120, 2629-2636. Loo, S., and Rine, D . (1995). Silencing and heritable domains of gene expression. A n n Rev Cell Dev B i o l 11, 519-548. L u , M . , Farrell, P. J., Johnson, R., and Iatrou, K . (1997). A baculovirus (Bombyx mori nuclear polyhedrosis virus) repeat element functions as a powerful constitutive enhancer in transfected insect cells. J B i o l C h e m 272, 30724-8. Lue, N . F., and Kornberg, R. D . (1987). Accurate initiation at R N A polymerase II promoters in extracts from Saccharomyces cerevisiae. Proc Natl A c a d S c i U S A 84, 883943. Lycett, G . J., and Crampton, J. M . (1993). Stable transformation of mosquito cell lines using a hsp70::neo fusion gene. Gene 136, 129-36. Maillet, L . , Boscheron, C , Gotta, M . , Marcand, S., Gilson, E . , and Gasser, S. M . (1996). Evidence for silencing compartments within the yeast nucleus: a role for telomere proximity and Sir protein concentration in silencer-mediated repression. Genes & Dev 10, 1796-811. Makrides, S. C . (1999). Components of vectors for gene transfer and expression in mammalian cells. Protein Expr Purif 17, 183-202. Maldonado, E . , Hampsey, M . , and Reinberg, D . (1999). Repression: targetting the heart of the matter. C e l l 99,455-458. Malone, R. W . , Feigner, P. L . , and Verma, I. M . (1989). Cationic liposome-mediated R N A transfection. Proc Natl A c a d S c i U S A 86, 6077-81.  Manak, J. R., and Scott, M . P. (1994). A class act: conservation of homeodomain protein functions. Development Supplement, 61-77. Marino, M . H . (1989). Expression systems for heterologous protein production. B i o P h a r m 2 , 18-33.  180  Maroni, G . , Otto, E . , and Lastowski-Perry, D . (1986). Molecular and cytogenetic characterization of a metallothionein gene of Drosophila. Genetics 112, 493-504. Martin, E . C . , and Adler, P. N . (1993). The Polycomb group gene Posterior sex combs encodes a chromosomal protein. Development 117, 641-655. Martina, J. A . , Daniotti, J. L . , and Maccioni, H . J. (1998). Influence o f N-glycosylation and N-glycan trimming on the activity and intracellular traffic of G D 3 synthase. J B i o l Chem 273, 3725-31. Martinez-Balbas, M . A . , Tsukiyama, T., Gdula, D . , and W u , C . (1998). Drosophila N U R F - 5 5 , a W D repeat protein involved in histone metabolism. Proc Natl A c a d S c i U S A 95, 132-7. Matsui, H . , Johnson, L . G . , Randell, S. H . , and Boucher, R. C . (1997). Loss of binding and entry of liposome-DNA complexes decreases transfection efficiency in differentiated airway epithelial cells. J B i o l Chem 272, 1117-26. M c C a l l , K . , and Bender, W . (1996). Probes for chromatin accessibility i n the Drosophila bithorax complex respond differently to Polycomb-mediated repression. E M B O J 15, 569-580. M c C a r r o l l , L . , and K i n g , L . A . (1997). Stable insect cell cultures for recombinant protein production. Curr Opin Biotechnol 8, 590-4. M c K e o n , J., and Brock, H . W . (1991). Interactions of the Polycomb group of genes with homeotic loci of Drosophila. Roux's A r c h Dev B i o l 199, 387-396. M c K e o n , J., Slade, E . , Sinclair, D . A . R., Cheng, N . , Couling, M . , and Brock, H . W . (1994). Mutations in some Polycomb group genes of Drosophila interfere with regulation of segmentation genes. M o l Gen Genet 244, 474-483. M c L a c h l i n , J. R., and M i l l e r , L . K . (1997). Stable transformation of insect cells to coexpress a rapidly selectable marker gene and an inhibitor of apoptosis. In Vitro C e l l Dev B i o l A n i m 33, 575-9. Mende, U . , Schmidt, C . J., Y i , F., Spring, D . J., and Neer, E . J. (1995). The G protein gamma subunit. Requirements for dimerization with beta subunits. J B i o l C h e m 270, 15892-8. Messmer, S., Franke, A . , and Paro, R. (1992). Analysis of the functional role of the Polycomb chromo domain in Drosophila melanogaster. Genes Dev 6, 1241-1254. Mihaly, J., Mishra, R. K . , and Karch, F. (1998). A conserved sequence motif in Polycomb-response elements. M o l C e l l 1, 1065-1066.  181  Miller, L . H . , Sakai, R. K . , Romans, P., Gwadz, R. W . , Kantoff, P., and Coon, H . G . (1987). Stable integration and expression of a bacterial gene in the mosquito Anopheles gambiae. Science 237, 779-81. Milne, T. A . , Sinclair, D . A . R., and Brock, H . W . (1999). The Additional sex combs gene of Drosophila is required fo activation and repression of homeotic loci, and interacts specifically with Polycomb and super sex combs. M o l Gen Genet 261, 753-761. Mlynarczyk, S. K . , and Panning, B . (2000). X inactivation: Tsix and X i s t as yin and yang. Curr B i o l 10, R899-903. Moazed, D . , and O'Farrell, P. H . (1992). Maintenance of the engrailed expression pattern by Polycomb group genes in Drosophila. Development 116, 805-810. Mollaaghababa, R., Sipos, L . , Tiong, S. Y . , Papoulas, O., Armstrong, J. A . , Tamkun, J. W . , and Bender, W . (2001). Mutations in Drosophila heat shock cognate 4 are enhancers of Polycomb. Proc Natl A c a d S c i U S A 98, 3958-63. Monroe, T. J., Muhlmann-Diaz, M . C , Kovach, M . J., Carlson, J. O., Bedford, J. S., and Beaty, B . J. (1992). Stable transformation of a mosquito cell line results in extraordinarily high copy numbers of the plasmid. Proc Natl A c a d Sci U S A 89, 5725-9. Mosna, G . , and Dolfini, S. (1972). Morphological and chromosomal characterization of three new continuous cell lines of Drosophila melanogaster. Chromosoma 38, 1-9. Muller, J. (1995). Transcriptional silencing by the Polycomb protein in Drosophila embryos. E M B O J 14, 1209-1220. Muller, J., Gaunt, S., and Lawrence, P. A . (1995). Function of the Polycomb protein is conserved in mice and flies. Development 121, 2847-2852. Munro, S., and Pelham, H . R. (1984). Use of peptide tagging to detect proteins expressed from cloned genes: deletion mapping functional domains of Drosophila hsp 70. E M B O J 3, 3087-93. Munro, S., and Pelham, H . R. (1987). A C-terminal signal prevents secretion of luminal E R proteins. C e l l 48, 899-907. Murby, M . , Cedergren, L . , Nilsson, J., Nygren, P. A . , Hammarberg, B . , Nilsson, B . , Enfors, S. O., and Uhlen, M . (1991). Stabilization of recombinant proteins from proteolytic degradation in Escherichia coli using a dual affinity fusion strategy. Biotechnol A p p l Biochem 14, 336-46. N g , J., Hart, C . M . , Morgan, K . , and Simon, J. A . (2000). A Drosophila E S C - E ( Z ) protein complex is distinct from other polycomb group complexes and contains covalently modified E S C , M o l C e l l B i o l 20, 3069-78.  182  Nilsson, J., Stahl, S., Lundeberg, J., Uhlen, M . , and Nygfen, P. A . (1997). Affinity fusion strategies for detection, purification, and immobilization of recombinant proteins. Protein E x p r P u r i f 11, 1-16. Nusslein-Volhard, C , Wieschaus, E . , and Kluding, H . (1984). Mutations affecting the pattern of the larval cuticle in Drosophila melanogaster. Roux's A r c h Dev B i o l 193, 267282. O'Reilly, D . R., M i l l e r , L . K . , and Luckow, V . A . (1992). Baculovirus expression vectors: a laboratory manual. (New York, W . H . Freeman and Co.). Orlando, V . , Jane, E . P., Chinwalla, V . , Harte, P. J., and Paro, R. (1998). The binding of Trithorax and Polycomb proteins to the bithorax complex: dynamic changes during early Drosophila embryogenesis. E M B O J 77, 5141-5150. Orlando, V . , and Paro, R. (1993). Mapping Polycomb-repressed domains in the bithorax complex using in vivo formaldehyde cross-linked chromatin. C e l l 75, 1187-1198. Otto, E . , A l l e n , J. M . , Young, J. E . , Palmiter, R. D . , and Maroni, G . (1987). A D N A segment controlling metal-regulated expression of the Drosophila melanogaster metallothionein gene M t n . M o l C e l l B i o l 7, 1710-5. Pandey, A . , and Mann, M . (2000). Proteomics to study genes and genomes. Nature 405, 837-846. Papoulas, O., Beek, S. J., Moseley, S. L . , M c C a l l u m , C . M . , Sarte, M . , Shearn, A . , and Tamkun, J. W . (1998). The Drosophila trithorax group proteins B R M , A S H I and A S H 2 are subunits of distinct protein complexes. Development 725, 3955-3966. Parker, C . S., and Topol, J. (1984). A Drosophila R N A polymerase II transcription factor contains a promoter-region-specific D N A - b i n d i n g activity. C e l l 36, 357-369. Paro, R. (1990). Imprinting a determined state into the chromatin of Drosophila. Trends Genet 6, 416-421. Paro, R. (1993). Mechanisms of heritable gene repression during development of Drosophila. Curr Op C e l l B i o l 5, 999-1005. Paro, R., and Hogness, D . S. (1991). The Polycomb protein shares a homologous domain with a heterochromatin-associated protein of Drosophila. Proc Natl A c a d S c i U S A 88, 263-7. Paro, R., and Zink, B . (1992). The Polycomb gene is differentially regulated during oogenesis and embryogensis of Drosophila melanogaster. M e c h Dev 40, 37-46.  183  Paul, J. (1973). C e l l and tissue culture, 4th ed. (London, Churchill Livingstone). Pelegri, F., and Lehman, R. (1994). A role of Polycomb group genes in the regulation of gap gene expression in Drosophila. Genetics 136, 1341-1353. Peterson, A . , K y b a , M . , Bornemann, D . , Morgan, K . , Brock, H . W . , and Simon, J. (1997). A domain shared by the Polycomb group proteins S c m and ph mediates heterotypic and homotypic interactions. M o l C e l l B i o l 17, 6683-6692. Peterson, C . L . , Dingwall, A . , and Scott, M . P. (1994). Five S W I / S N F gene products are components of a large multisubunit complex required for transcriptional enhancement. Proc Natl A c a d S c i U S A 91, 2905-2908. Peterson, C . L . , Zhao, Y . , and Chait, B . T. (1998). Subunits of the yeast S W I / S N F complex are members of the actin-related protein ( A R P ) family. J B i o l C h e m 273, 236414. Pfeifer, T. A . (1998). Expression of heterologous proteins in stable insect cell culture. Curr Opin Biotechnol 9, 518-521. Pfeifer, T. A . , Hegedus, D . , Wang, Y . J., Zhao, Y . , Meredith, J., Brock, H . W . , Phillips, J. E., Grigliatti, T. A . , and Theilmann, D . A . (1999). Analysis of an insect neuropeptide, Schistocerca gregaria ion transport peptide (ITP), expressed in insect cell systems. A r c h Insect Biochem Physiol 42, 245-52. Pfeifer, T. A . , Hegedus, D . D . , Grigliatti, T. A . , and Theilmann, D . A . (1997). Baculovirus intermediate early promoter-mediated expression of the Zeocin resistance gene for use as a dominant selectable marker in Dipteran and Lepidopteran insect cell lines. Gene 188, 183-190. Pfeiffer, S. E . , Betschart, B . , Cook, J., Mancini, P., and Morris, R. (1977). G l i a l cell lines. In C e l l , Tissue, and Organ Cultures in Neurobiology., S. Federoff, and L . Hertz, eds. (New York, Academic Press), pp. 287-346. Phillips, M . D . , and Shearn, A . (1990). Mutations in polycombeotic, a Drosophila Polycomb group gene, cause a wide range of maternal and zygotic phenotypes. Genetics 725,91-101. Pikaart, M . J., Recillas-Targa, F., and Felsenfeld, G . (1998). Loss of transcriptional activity of a transgene is accompanied by D N A methylation and histone deacetylation and is prevented by insulators. Genes Dev 12, 2852-62. Pile, L . A . , and Wassarman, D . A . (2000). Chromosomal localization links the S I N 3 R P D 3 complex to the regulation of chromatin condensation, histone acetylation and gene expression. E M B O J 19, 6131-40.  184  Pirrotta, V . (1998). Polycombing the genome: P c G , trxG and chromatin silencing. C e l l 93, 333-336. Pirrotta, V . , Chan, C . S., M c C a b e , D . , and Qian, S. (1995). Distinct parasegmental and imaginal enhancers and the establishment of the expression pattern of the Ubx gene. Genetics 141, 1439-1450. Pirrotta, V . , and Rastelli, L . (1994). white gene expression, repressive chromatin domains, and homeotic gene regulation in Drosophila. Bioessays 8, 549-556. Platero, J. S., Hartnett, T., and Eissenberg, J. C . (1995). Functional analysis of the chromodomain of H P 1 . E M B O J 14, 3977-3986. Platero, J. S., Sharp, E . J., Adler, P. N . , and Eissenberg, J. C . (1996). In vivo assay for protein-protein interactions using Drosophila chromosomes. Chromosoma 104, 393-404. Poux, S., M c C a b e , D . , and Pirrotta, V . (2001). Recruitment of components of Polycomb Group chromatin complexes in Drosophila. Development 128, 75-85. Prickett, K . S., Amberg, D . C , and Hopp, T. P. (1989). A calcium-dependent antibody for identification and purification of recombinant proteins. Biotechniques 7, 580-9. Pugh, F. B . (2000). Control of gene expression through regulation of the T A T A - b i n d i n g protein. Gene 255, 1-14. Rastelli, L . , Chan, C . S., and Pirotta, V . (1993). Related chromosome binding sites for zeste, suppressor of zeste and Polycomb group protein and their dependence on Enhancer ofzeste function. E M B O J 12, 1513-1522. Reijnen, M . J., Hamer, K . M . , Blaauwen, J. L . d., Lambrechts, C , Schoneveld, I., Driel, R. v., and Otte, A . P. (1995). Polycomb and bmi-1 homologus are expressed in overlapping patterns in Xenopus embryos and are able to interact with each other. M e c h Dev 53, 35-46. Rigaut, G . , Shevchenko, A . , Rutz, B . , W i l m , M . , Mann, M . , and Seraphin, B . (1999). A generic protein purification method for protein complex characterization and proteome exploration. Nat Biotechnol 17, 1030-2. Rojo, F. (1999). Repression of transcription initiation in bacteria. J Bacteriology 181, 2987-2991. Rojo, F . (2001). Mechanisms of transcriptional repression. Curr Opin M i c r o b i o l 4, 145151. Roseman, R. R., Morgan, K . , M a l l i n , D . R., Roberson, R., Parnell, T. J., Bornemann, D . J., Simon, J. A . , and Geyer, P. K . (2001). Long-range repression by multiple Polycomb  185  group (PcG) proteins targeted by fusion to a defined dna-binding domain in drosophila. Genetics 158, 291-307. Ruhf, M . L . , Braun, A . , Papoulas, O., Tamkun, J. W . , Randsholt, N . , and Meister, M . (2001). The domino gene of Drosophila encodes novel members of the S W I 2 / S N F 2 family of DNA-dependent ATPases, which contribute to the silencing of homeotic genes. Development 128, 1429-41. Rusche, L . N . , and Rine, J. (2001). Conversion of a gene-specific repressor to a regional silencer. Genes & Dev 15, 955-67. Saez, E . , N o , D . , West, A . , and Evans, R. M . (1997). Inducible gene expression in mammalian cells and transgenic mice. Curr Opin Biotechnol 1997 Oct;8(5):608-16 8, 608-16. Saget, O., Randsholt, N . B . (1994). Transposon-induced rearrangements in the duplicated locus ph of Drosophila melanogaster can create new chimeric genes functionally identical to wild-type. Gene 149, 227-235. Sambrook, J., Fritsch, E . F., and Maniatis, T. (1989). Molecular cloning: a laboratory manual. ( U S A , C o l d Spring Harbor Laboratory Press). Sander, I., Raulf-Heimsoth, M . , Siethoff, C , Lohaus, C , Meyer, H . E . , and Baur, X . (1998). Allergy to Aspergillus-derived enzymes in the baking industry: identification of beta-xylosidase from Aspergillus niger as a new allergen (Asp n 14). J Allergy C l i n Immunol 102, 256-64. Sang, J. H . (1981). Drosophila cells and cell lines. A d v Cell Culture / , 125-182. Satijn, D . P. E . , and Otte, A . P. (1999). Polycomb group protein complexes: do different complexes regulate distinct target genes? B i o c h i m Biophys A c t a 1447, 1-16. Sauer, F., Fondell, J. D . , Ohkuma, Y . , Roeder, R. G . , and Jackie, H . (1995). Control of transcription by Kruppel through interactions with TFID3 and TFIJE beta. Nature 375, 162-4. Schlossherr, J., Eggert, H . , Paro, R., Cremer, S., and Jack, R. S. (1994). Gene inactivation in Drosophila mediated by the Polycomb gene product or by position-effect variegation does not involve major changes in the accessibility of the chromatin fibre. M o l Gen Genet 243,453-462. Schneider, I. (1972). Cell lines derived from late embryonic stages of Drosophila melanogaster. J Embryol E x p Morphol 27, 353-365. Schneider, I., and Blumenthal, A . B . (1978). Drosophila cell and tissue culture. In The genetics and biology of Drosophila. V o l 2a (London and New York, Academic Press).  186  Schneikert, J., Hubner, S., Langer, G . , Petri, T., Jaattela, M . , Reed, J., and Cato, A . C . (2000). Hsp70-RAP46 interaction in downregulation of D N A binding by glucocorticoid receptor. E M B O J 19, 6508-16. Schumacher, A . , and Magnuson, T. (1997). Murine Polycomb- and Trithorax-group genes regulate homeotic pathways and beyond. Trends Genet 13, 167-170. Seelig, H . P., Moosbrugger, I., Ehrfeld, H . , Fink, T., Renz, M . , and Genth, E . (1995). The major dermatomyositis-specific M i - 2 autoantigen is a presumed helicase involved in transcriptional activation. Arthritis Rheum 38, 1389-99. Segal, D . , Cherbas, L . , and Cherbas, P. (1996). Genetic transformation of Drosophila cells in culture by P element-mediated transposition. Somat C e l l M o l Genet 22, 159-65. Sekinger, E . A . , and Gross, D . S. (2001). Silenced chromatin is permissive to activator binding and P I C recruitment. Cell 105, 403-14. Shao, Z . , Raible, F . , Mollaaghababa, R., Guyon, J. R., W u , C.-t., Bender, W . , and Kingston, R. E . (1999). Stabilization of chromatin structure by P R C 1 , a Polycomb complex. C e l l 98, 37-46. Shigekawa, K . , and Dower, W . J. (1988). Electroporation of eukaryotes and prokarryotes: a general approach to the introduction fo macromolecules into cells. BioTechniques 6, 742-751. Shimell, M . J., Peterson, A . J., Burr, J., Simon, J. A . , and O'Connor, M . B . (2000). Functional analysis of repressor binding sites in the iab-2 regulatory region of the abdominal-A homeotic gene. Dev B i o l 218, 38-52. Sif, S., Saurin, A . J., Imbalzano, A . N . , and Kingston, R. E . (2001). Purification and characterization of mSin3A-containing B r g l and h B r m chromatin remodeling complexes. Genes & Dev 15, 603-18. Sif, S., Stukenberg, P. T., Kirschner, M . W . , and Kingston, R. E . (1998). Mitotic inactivation of a human S W I / S N F chromatin remodeling complex. Genes & Dev 12, 2842-51. Simcox, A . A . , Sobeih, M . M . , and Shearn, A . (1985). Establishment and characterization of continuous cell lines derived from temperature-sensitive mutants of Drosophila melanogaster. Somat C e l l M o l Genet 11, 63-70. Simon, J. (1995). Locking in stable states of gene expression: transcriptional control during Drosophila development. Curr Opin Cell B i o l 7, 376-385.  187  Simon, J., Bornemann, D . , Lunde, K . , and Schwartz, C . (1995). The extra sex combs product contains W D 4 0 repeats and its time of action implies a role distinct from other Polycomb group products. M e c h Dev 53, 197-208. Simon, J., Chiang, A . , and Bender, W . (1992). Ten different Polycomb group genes are required for spatial control of the abd-A and Abd-B homeotic products. Development 114, 493-505. Simon, J., Chiang, A . C , Bender, W . , Shimell, M . J., and O'connor, M . (1993). Elements of the Drosophila bithorax complex that mediate repression by Polycomb group products. Dev B i o l 158, 131-144. Sinclair, D . A . R., Campbell, R. B . , Nicholls, F . , Slade, E . , and Brock, H . W . (1992). Genetic analysis of the Additional sex combs locus of Drosophila melanogaster. Genetics 750,817-825. Sinclair, D . A . R., M i l n e , T. A . , Hodgson, J. W . , Shellard, J., Salinas, C . A . , K y b a , M . , Randazzo, F., and Brock, H . W . (1998). The Additional sex combs gene of Drosophila encodes a chromatin protein that binds to shared and unique Polycomb group sites on polytene chromosomes. Development 725, 1207-1216. Sit, T. L . , Vaewhongs, A . A . , and L o m m e l , S. A . (1998). RNA-mediated trans-activation of transcription from a viral R N A . Science 281, 829-32. Smith, J. S., Brachmann, C . B . , Celic, I., Kenna, M . A . , Muhammad, S., Starai, V . J., Avalos, J. L . , Escalante-Semerena, J. C , Grubmeyer, C , Wolberger, C , and Boeke, J. D . (2000). A phylogenetically conserved NAD+-dependent protein deacetylase activity in the Sir2 protein family. Proc Natl A c a d Sci U S A 97, 6658-63. Smouse, D . , Goodman, C . S., Mahowald, A . , and Perrimon, N . (1988). Polyhomeotic: a gene required for the embryonic development of axon pathways in the central nervous system of Drosophila. Genes Dev 2, 830-842. Soto, M . C , Chou, T . - B . , and Bender, W . (1995). Comparison of germ-line mosaics of genes in the Polycomb group of Drosophila melanogaster. Genetics 140, 231-243. Stankunas, K . , Berger, J., Ruse, C , Sinclair, D . A . R., Randazzo, F., and Brock, H . W . (1998). The Enhancer of Polycomb gene of Drosophila encodes a chromatin protein conserved in yeast and mammals. Development 725, 4055-4066. Sterner, D . E . , Grant, P. A . , Roberts, S. M . , Duggan, L . J., Belotserkovskaya, R., Pacella, L. A . , Winston, F . , Workman, J. L . , and Berger, S. L . (1999). Functional organization of the yeast S A G A complex: distinct components involved in structural integrity, nucleosome acetylation, and T A T A - b i n d i n g protein interaction. M o l C e l l B i o l 79, 86-98.  188  Strahl, B . D . , and A l l i s , C . D . (2000). The language of covalent histone modifications. Nature 403, 41-45. Strouboulis, J., Damjanovski, S., Vermaak, D . , Meric, F., and Wolffe, A . P . (1999). Transcriptional repression by X P c l , a new Polycomb homolog in Xenopus laevis embryos, is independent of histone deacetylase. M o l Cell B i o l 19, 3958-3968. Struhl, G . , and A k a m , M . E . (1985). Altered distribution of Ultrabithorax transcripts in extra sex combs mutant embryos of Drosophila. E M B O J 4, 3259-3264. Struhl, K . (1999). Fundamentally difference logic of gene regulation in eukaryotes and prokaryotes. C e l l 98, 1-4. Strutt, H . , and Paro, R. (1997a). Co-localization of Polycomb protein and G A G A factor on regulatory elements responsible for the maintenance of homeotic gene expression. E M B O J 16, 3621-3632. Strutt, H . , and Paro, R. (1997b). The Polycomb group protein complex of Drosophila melanogaster has different compositions at different target genes. M o l C e l l B i o l 17, 6773-6783. Sullivan, C . S., Cantalupo, P., and Pipas, J. M . (2000). The molecular chaperone activity of simian virus 40 large T antigen is required to disrupt R b - E 2 F family complexes by an ATP-dependent mechanism. M o l C e l l B i o l 20, 6233-43. Sullivan, C . S., Gilbert, S. P., and Pipas, J. M . (2001). ATP-dependent simian virus 40 Tantigen-Hsc70 complex formation. J V i r o l 75, 1601-10. Taggart, A . K . , and Pugh, B . F . (1996). Dimerization of T F I I D when not bound to D N A . Science 272, 1331-3. Thut, C . J., Goodrich, J. A . , and Tjian, R. (1997). Repression of p53-mediated transcription by M D M 2 : a dual mechanism. Genes Dev 11, 1974-86. Tie, F., Furuyama, T., and Harte, P. J. (1998). The Drosophila Polycomb Group proteins E S C and E ( Z ) bind directly to each other and co-localize at multiple chromosomal sites. Development 725, 3483-96. Tie, F., Furuyama, T., Prasad-Sinha, J., Jane, E . , and Harte, P. J . (2001). The Drosophila Polycomb Group proteins E S C and E ( Z ) are present in a complex containing the histonebinding protein p55 and the histone deacetylase R P D 3 . Development 128, 275-86. Turner, B . M . (2000). Histone acetylation and an epigenetic code. BioEssays 22, 836845.  189  Tyler, J. K . , Bulger, M . , Kamakaka, R. T., Kobayashi, R., and Kadonaga, J. T. (1996). The p55 subunit of Drosophila chromatin assembly factor 1 is homologous to a histone deacetylase-associated protein. M o l C e l l B i o l 16, 6149-59. U m , M . , L i , C , and Manley, J. L . (1995). The transcriptional repressor even-skipped interacts directly with T A T A - b i n d i n g protein. M o l C e l l B i o l 15, 5007-16. van der V l a g , J., and Otte, A . P. (1999). Transcriptional repression mediated by the human polycomb-group protein E E D involves histone deacetylation. Nature Genetics 23, 474-8. Vermaak, D . , Wade, P. A . , Jones, P. L . , Shi, Y . B . , and Wolffe, A . P. (1999). Functional analysis of the SIN3-histone deacetylase RPD3-RbAp48-histone H 4 connection in the Xenopus oocyte. M o l C e l l B i o l 19, 5847-60. Verreault, A . , Kaufman, P. D . , Kobayashi, R., and Stillman, B . (1996). Nucleosome assembly by a complex of C A F - 1 and acetylated histones H 3 / H 4 . Cell 87, 95-104. Verrijzer, C . P., and Tjian, R. (1996). T A F s mediate transcriptional activation and promoter selectivity. Trends Biochem S c i 21, 338-42. von Zelewsky, T., Palladino, F., Brunschwig, K . , Tobler, H . , Hajnal, A . , and Muller, F . (2000). The C . elegans M i - 2 chromatin-remodelling proteins function in vulval cell fate determination. Development 127, 5277-84. Wahl, G . M . , de Saint Vincent, B . R., and DeRose, M . L . (1984). Effect of chromosomal position on amplification of transfected genes in animal cells. Nature 307, 516-20. Wang, Y . - J . , Zhao, Y . , Meredith, J., Phillips, J. E . , Theilmann, D . A . , and Brock, H . W . (2000). Mutational analysis of the C-terminus in ion transport peptide (ITP) expressed in Drosophila K c l cells. Archives of Insect Biochemistry and Physiology 45, 129-138. Watson, A . D . , Edmondson, D . G . , Bone, J. R., M u k a i , Y . , Y u , Y . , D u , W . , Stillman, D . J., and Roth, S. Y . (2000). Ssn6-Tupl interacts with class I histone deacetylases required for repression. Genes & Dev 14, 2737-44. Wigler, M . , Pellicer, A . , Silverstein, S., A x e l , R., Urlaub, G . , and Chastin, L . (1979). D N A mediated transfer of the adenine-phosphoribosyltransferase locus into mammalian cells. Proc Natl A c a d S c i U S A 76, 1373-1376. Wolffe, A . (1998). Chromatin: structure & function., 3rd edn (San Diego, Academic Press). Woodage, T., Basrai, M . A . , Baxevanis, A . D . , Hieter, P., and Collins, F . S. (1997). Characterization of the C H D family of proteins. Proc Natl A c a d Sci U S A 94, 11472-7.  190  Workman, J. L . , and Kingston, R. E . (1998). Alteration of nucleosome structure as a mechanism of transcriptional regulation. A n n u Rev Biochem 67, 545-579. Wurm, F., and Bernard, A . (1999). Large-scale transient expression in mammalian cells for recombinant protein production. Current Opinion in Biotech 10, 156-159. X i e , J., Collart, M . , Lemaire, M . , Stelzer, G . , and Meisterernst, M . (2000). A single point mutation in T F I I A suppresses N C 2 requirement in vivo. E M B O J 79, 672-82. X u , L . , Gonzalez-Agosti, C , Beauchamp, R., Pinney, D . , Sterner, C , and Ramesh, V . (1998). Analysis of molecular domains of epitope-tagged merlin isoforms in Cos-7 cells and primary rat Schwann cells. E x p C e l l Res 238, 231-40. Yamamoto, Y . , Girard, F., Bello, B . , Affolter, M . , and Gehring, W . J. (1997). The cramped gene of Drosophila is a member of the Polycomb-group, and interacts with mus209, the gene encoding Proliferating C e l l Nuclear Antigen. Development 124, 33853394. Yates, J. L . , Warren, N . , and Sugden, B . (1985). Stable replication of plasmids derived from Epstein-Barr virus in various mammalian cells. Nature 313, 812-5. Yates, J. R., 3rd, E n g , J. K . , McCormack, A . L . , and Schieltz, D . (1995). Method to correlate tandem mass spectra of modified peptides to amino acid sequences in the protein database. A n a l Chem 67, 1426-36. Yates, J. R., III. (1998). Mass spectrometry and the age of the proteome. J Mass Spectrometry 33, 1-19. Yeager, T. R., and Reddel, R. R. (1999). Constructing immortalized human cell lines. Current Opinion in Biotechnology 10, 465-469. Yoshinaga, S. K . , and Yamamoto, K . R. (1991). Signaling and regulation by a mammalian glucocorticoid receptor in Drosophila cells. M o l Endocrinol 5, 844-853. Zabner, J., Fasbender, A . J., Moninger, T., Poellinger, K . A . , and Welsh, M . J. (1995). Cellular and molecular barriers to gene transfer by a cationic lipid. J B i o l C h e m 270, 18997-9007. Zaman, Z . , Ansari, A . Z . , K o h , S. S., Young, R., and Ptashne, M . (2001). Interaction of a transcriptional repressor with the R N A polymerase II holoenzyme plays a crucial role in repression. Proc Natl A c a d S c i U S A 98, 2550-4. Zhang, H . , Catron, K . M . , and Abate-Shen, C . (1996). A role for the M s x - 1 homeodomain in transcriptional regulation: residues in the N-terminal arm mediate T A T A binding protein interaction and transcriptional repression. Proc Natl A c a d S c i U S A 93, 1764-9.  191  Zhang, W . , and Chait, B . T. (2000). ProFound: an expert system for protein identification using mass spectrometric peptide mapping information. A n a l C h e m 72, 2482-9. Zhao, K . , Wang, W . , Rando, O. J., X u e , Y . , Swiderek, K . , K u o , A . , and Crabtree, G . R. (1998). Rapid and phosphoinositol-dependent binding o f the S W I / S N F - l i k e B A F complex to chromatin after T lymphocyte receptor signaling. C e l l 95, 625-36. Zink, B . , and Paro, R. (1989). In vivo binding pattern of a trans-regulator of the homeotic genes in Drosophila melanogaster. Nature 337, 468-471.  192  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0090643/manifest

Comment

Related Items