Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Ubiquitin gene structure and expression in Caenorhabditis elegans Graham, Roger Walter 1990

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

Item Metadata


831-UBC_1990_A1 G72.pdf [ 12.52MB ]
JSON: 831-1.0098748.json
JSON-LD: 831-1.0098748-ld.json
RDF/XML (Pretty): 831-1.0098748-rdf.xml
RDF/JSON: 831-1.0098748-rdf.json
Turtle: 831-1.0098748-turtle.txt
N-Triples: 831-1.0098748-rdf-ntriples.txt
Original Record: 831-1.0098748-source.json
Full Text

Full Text

Ubiquitin Gene Structure and Expression i n Caenorhabditis elegans  By  Roger Walter Graham  B.Sc.(honours), The University of Winnipeg, 1984  A THESIS SUBMITTED IN P A R T I A L FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Department of Biochemistry  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA March 1990 © Roger Walter Graham  In  presenting  degree freely  at  this  the  University  available  copying  of  department publication  for  this or of  thesis  this  partial  of  British  reference  thesis by  in  for  his thesis  and  The  University  Vancouver,  Date  DE-6  of  or for  her  (2/88)  of  British  APRIL 3, 1990  Columbia,  I  study.  Columbia  I  further  purposes  gain  the  requirements  agree  be  It  is  shall  that  agree  may  representatives.  financial  BTOCHEMTSTKV  Canada  of  scholarly  permission.  Department  fulfilment  not  that  the  Library  an  by  understood allowed  advanced  shall  permission  granted  be  for  the that  without  for head  make  extensive of  copying my  it  my or  written  ii  ABSTRACT  Ubiquitin is a multifunctional 76 amino acid protein which plays critical roles in many aspects of cellular metabolism. Ubiquitin protein structure and gene structure are highly conserved among eukaryotes. In C. elegans the major source of ubiquitin R N A was shown to be the polyubiquitin locus, UbiA. UbiA was shown to be transcribed as a polycistronic mRNA which contained eleven tandem repeats of ubiquitin sequence and possessed a two amino acid carboxy terminal extension on the final repeat. Mature UbiA mRNA was demonstrated to acquire a 22 nucleotide leader sequence via a trans splicing reaction involving a 100 nucleotide splice leader R N A derived from a different chromosome. UbiA was also shown to be unique among known polyubiquitin genes in containing four cis spliced introns within its coding sequence. Thus UbiA was shown to be one of a small class of genes found in higher eukaryotes whose hnRNA undergoes both cis and trans splicing. The expression of the UbiA gene was studied under various heat shock conditions, and was monitored during larval molting and throughout the major stages of development.  These studies indicated that the expression of the UbiA gene was not  inducible by acute or chronic heat shock, and did not appear to be under nutritional or developmental regulation.  A second ubiquitin gene, UbiB, was cloned from C. elegans and a related nematode species C. briggsae. This gene was comprised of (at least) one ubiquitin unit followed by a basic 52 amino acid tail sequence.  iii TABLE O F CONTENTS Page Abstract  ii  Table of Contents  iii  List of Figures  viii  Abbreviations  x  Acknowledgements  I.  xiii  INTRODUCTION  1  1.1  Historical Perspective  2  1.2  Structure of Ubiquitin  3  1.3  Structure-Function Relations  5  1.4  1.3.1  Natural Variants  5  1.3.2  Site-directed Mutants of Ubiquitin  5  Mutations Affecting Ubiquitin Structure  6  Mutations Affecting Ubiquitin Conformation  7  Ubiquitin and Protein Degradation  8  1.4.1  Mechanism of Ubiquitin-Mediated Protein Degradation  9  1.4.2  Selectivity of Ubiquitin-Mediated Protein Degradation  11  1.413  Degradation of Proteins with a Free Amino Terminus  12 Degradation of Proteins with a Modified N Tenrdnus  14  Structure of Ubiquitin on Proteins Targeted for Proteolysis  15  1.5  Ubiquitin as a Protease  15  1.6  Ubiquitin and the Heat Shock Response  16  1.6.1  Heat Shock in Prokaryotes  17  1.6.2  Heat Shock in Eukaryotes  18  iv  1.7  Ubiquitin and Chromatin  21  1.7.1  Ubiquitinated Histones and the Cell Cycle  21  1.7.2  Ubiquitinated Histones and Transcription  22  1.8  Ubiquitin and the Cell Surface  23  1.9  Genetics of the Ubiquitin System  23  1.9.1 1.9.2  te85 Yeast Ubiquitin Genes  24  Yeast Polyubiquitin Gene  24  Yeast Ubiquitin Hybrid Genes  27  Ubiquitin-Protein Ligase System Genes  28  1.9.3  Ubiquitin Genes from Other Organisms  28  1.9.4  Ubiquitin-like Genes  29  H.  24  Ubiquitin Cross-Reactive Protein (UCRP)  29 GdX  31  Togavirus Ubiquitin  31  Baculovirus Ubiquitin  31  1.10  C. elegans as a Model System  32  1.11  Trans Splicing  33  1.12  The Present Study  34  EXPERIMENTAL PROCEDURES 2.1  Growth and Maintenance of Nematodes  35 35  2.1.1  Collection of Gravid Adult Nematodes  36  2.1.2  Preparation of Nematode Embryos  36  2.1.3  Viable Freezing of Nematodes  37  2.2  Nematode heat shock experiments  38  2.3  Preparation and Analysis of Nematode Genomic D N A  38  V  2.4  Preparation and Analysis of Nematode R N A  39  2.5  Screening of Recombinant D N A Libraries  40  2.5.1  Screening of Bacteriophage X Libraries  40  2.5.2  Isolation of Bacteriophage D N A  41  2.5.3  Screening of Cosmid and Y A C Panels  42  2.6  Transformations  42  2.7  Purification of plasmid D N A  42  2.8  Rapid Screening of Single Bacterial Colonies by PCR  43  2.9  D N A Sequencing  45  2.9.1  Single Stranded D N A Sequencing  45  2.9.2  Double Stranded D N A Sequencing  45  2.10  Preparation of Radioactive D N A Probes  46  2.10.1 Nick Translation  46  2.10.2 Primer Extension M13 probes  47  2.11  Nucleic Acid Hybridization  47  2.12  Dot hybridization Analysis  48  2.13  SI Nuclease Analysis  48  2.14  Primer Extension R N A Sequencing  49  2.15  PCR  50  2.15.1  Preparation of Total R N A for PCR  50  2.15.2  First Strand cDNA Synthesis  50  2.15.3  RACE cDNA Amplification  50  2.15.4  Analysis of PCR Products  51  2.15.5  Cloning PCR Products  51  2.15.6  Single Nematode PCR  52  2.15.7  PCR Assay of Transgenic Nematode R N A  53  vi  HI.  RESULTS 3.1  Isolation of Genomic Clones  54  3.2  Localization of the UbiA Gene  54  3.3  Analysis of the UbiA Gene Sequence  55  3.4  Genomic Southern Analysis  63  3.5  Northern Hybridization Analysis  66  3.6  Quantification of UbiA Expression  68  3.7  UbiA 3' Flanking Region  70  3.8  Intragenic Introns  73  3.9  UbiA Trans Spliced Intron  73  3.10  UbiA 5' Hanking Region  80  3.11  Expression of Ubiquitin  83  3.11.1  IV.  54  Developmental Analysis of Ubiquitin Expression  83  3.11.2 Ubiquitin Expression During Heat Shock  87  3.11.3 Heat Shock Inhibits Trans Splicing in C. elegans  91  3.12  Analysis of UbiA Promoter Function in Transgenic Nematodes  91  3.13  PCR Amplification of Ubiquitin cDNAs with Degenerate Primers  97  3.14  PCR Amplification of Trans Spliced cDNAs from C. elegans  105  DISCUSSION  110  4.1  Polyubiquitin Gene Structure  111  4.2  Nematode Ubiquitin Genes  112  4.3  Introns and Evolution  113  4.4  4.3.1  Evolution of the Cis Spliced Introns of U b i A  113  4.3.2  Trans Splicing of UbiA  115  Potential Regulatory Sequences in the 5' Flanking Region of U b i A  119  vii  4.5  Ubiquitin Gene Expression  121  4.5.1  Ubiquitin Expression during Development  121  4.5.2  Heat Shock and Ubiquitin  123  4.6  Heat Shock and Trans Splicing  126  4.7  Analysis of the UbiA Promoter in Transgenic Nematodes  126  4.8  Ubiquitin Hybrid Genes  129  4.9  PCR of Trans Spliced cDNAs from C. elegans  133  4.10  Potential Areas of Future Study  133  V.  REFERENCES  135  VI.  APPENDIX  145  A.  List of Oligonucleotides and their Sequences  145  B.  E. coli Strains and their Genotypes  146  C.  Calculation of UbiA mRNA Abundance  146  D. Sequence of Tail Regions of Nematode UbiB cDNAs  147  viii  LIST OF FIGURES Page 1.  Evolutionary variations in ubiquitin structure  4  2.  The ubiquitin-mediated system of non-lysosomal protein breakdown  10  3.  The ubiquitin genes of the yeast Saccharomyces cerevisiae  25  4  Comparison of cloned ubiquitin gene structures  29  5.  Single bacterial colony polymerase chain reaction  44  6.  Organization of the UbiA locus of C. elegans  56  7.  Structure of the UbiA gene of C. elegans  57  8.  D N A sequence of the C. elegans polyubiquitin gene, UbiA  58  9.  D N A sequence alignment of the eleven coding region repeats of UbiA  61  10.  Interspecies comparison of ubiquitin amino acid sequences  64  11.  Southern hybridization analysis of ubiquitin gene structure  65  12.  Northern hybridization analysis of C. elegans ubiquitin gene expression  67  13.  UbiB is not derived from the UbiA locus  69  14.  Quantitation of UbiA expression by R N A dot hybridization  71  15.  D N A sequence comparison of C. elegans and C. briggsae UbiA cDNAs  72  16.  D N A sequence comparison of the four UbiA cis spliced introns  74  17.  Northern hybridization analysis of the UbiA leader sequence  75  18.  SI nuclease protection analysis of the UbiA mRNA  77  19.  Primer extension sequencing of the UbiA and Actin3 mRNAs  79  20.  SI nuclease protection analysis of UbiA hnRNA  81  ix 21.  The life cycle of C. elegans  84  22.  Northern hybridization analysis of ubiquitin mRNA in development  85  23.  Northern hybridization analysis of ubiquitin mRNA in moulting  87  24.  R N A hybridization analysis of UbiA mRNA during chronic heat shock  89  25.  R N A hybridization analysis of UbiA mRNA at various temperatures  90  26.  Heat shock inhibits trans splicing of UbiA  92  27.  PCR strategies employed to assay transgenic C. elegans strains  93  28.  PCR amplification of promoter D N A from transgenic nematodes  94  29.  PCR amplification of reporter gene mRNA from transgenic nematodes  96  30.  Schematic illustration of the PCR strategy to amplify ubiquitin cDNAs  98  31.  Degenerate ubiquitin oligonucleotides used for PCR cDNA amplification  99  32.  PCR amplification of ubiquitin cDNAs from C. briggsae and yeast  100  33.  PCR amplification of ubiquitin cDNAs from other organisms  101  34.  D N A sequence comparison of UbiB cDNAs of C. elegans and C. briggsae  103  35.  Tail amino acid sequences from PCR-amplified yeast Ubi3-type cDNAs  104  36.  PCR strategy for the amplification of trans spliced cDNAs from C. elegans 106  37.  Southern hybridization analysis of PCR-amplified trans spliced cDNAs  107  38.  Partial sequence of a PCR-amplified C. elegans trans spliced cDNA  109  39.  A model for U b i A mRNA maturation in C. elegans  116  X  ABBREVIATIONS  AMP  adenosine-5'-monophosphate  ATP  adenosine-5'-triphosphate  bp  basepair(s)  BSA  bovine serum albumin  P-gal  (5-galactosidase  cDNA  complementary D N A  C. elegans  Caenorhabditis elegans  CEP  carboxy terminal extension protein  cpm  counts per minute  DNA  deoxyribonucleic acid  dATP  deoxyadenosine-5'-triphosphate  dCTP  deoxycytidine-5'-triphosphate  dGTP  deoxyguanosine-5'-triphosphate  dTTP  deoxythymidine-5'-triphosphate  ddATP  dideoxyadenosine-5'-triphosphate  ddCTP  dideoxycytidine-5'-triphosphate  ddGTP  dideoxyguanosine-5'-triphosphate  ddTTP  dideoxythymidine-5'-triphosphate  El  ubiquitin activating enzyme  E2  ubiquitin conjugating enzyme  E3  proteolytic substrate recognition factor  E. coli  Eschericia coli  EDTA  ethylenediamine tetraacetic acid  Gu-HCl  guanidinium hydrochloride  HEPES  N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid  xi  hsp  heat shock polypeptide  HSTF  heat shock transcription factor  HSE  heat shock element  I  inosine  kb  kilobase(s)  kd  kilodaltons  MAb  monoclonal antibody  mRNA  messenger R N A  MMLV  Moloney murine leukemia virus  N  adenine, cytosine, guanine, or thymine  NMR  nuclear magnetic resonance  nt  nucleotide(s)  PCR  polymerase chain reaction  PDGF  platelet-derived growth factor  PIPES  piperazine-N,N'-bis (2-ethansulfonic acid)  polyA  +  polyadenylated  PPi  pyrophosphate  RNA  ribonucleic acid  RNase  ribonuclease  rRNA  ribosomal R N A  S. cerevisiae  Saccharomyces cerevisiae  SDS  sodium dodecyl sulfate  SL  splice leader  SSPE  180 m M NaCl,l m M EDTA,10 m M N a H P 0 , p H 7.4  TAE  40 m M Tris-acetate p H 8.0, 2 m M EDTA  Taq  Thermus aquaticus  TBE  90 m M Tris-borate p H 8.3,1 m M EDTA  2  4  xii  TE  10 m M Tris-HCl p H 8.0,1 mM EDTA  tRNA  transfer R N A  Tris  tris (hydroxymethyl) aminomethane  ts  temperature sensitive  Ub  ubiquitin  UCRP  ubiquitin cross-reactive protein  xiii  ACKNOWLEDGEMENTS  The spirit of cooperation and collaboration that existed at UBC during the course of my research made every day in and out of the laboratory productive and playful. I would like to thank my supervisor, Peter Candidc, for the freedom and guidance that he readily provided. I would like to thank all those who encouraged me at every turn and those who shared the frenzied pace of packing what probably amounted to fifteen years into five. I would especially like to thank Roland Russnak, Rob Kay, and Dave Banfield for enlightened discussions and exaggerated beer consumption. Peter Candido, Don Jones, and Dennis Dixon turned morning coffee in the lab into a stimulating period of discussion, invention, and brainstorming. I would also like to thank all those who proferred advice, equipment, and enzymes in exchange for tall tales and improbable excuses. Finally, I would especially like to thank Kaos, a lovable group of easygoing, pleasure-seeking, intellectual gourmand volleyball players without whom the world would seem organized.  1 I.  INTRODUCTION  Ubiquitin is a 76 residue polypeptide that has been highly conserved in the evolution of eukaryotes. It has been implicated in vital roles in almost all aspects of cellular metabolism (for reviews see Rechsteiner, 1988; Schlesinger et al., 1988). In the nucleus ubiquitin is thought to play important roles in chromatin structure, thereby regulating gene activity (Levinger and Varshavsky, 1982), cell cycle events (Ciechanover et a l . , 1984a,b; Finley et al., 1984) and potentially D N A repair (Jentsch et al., 1987). In the cytoplasm ubiquitin is one of the most abundant proteins (Rechsteiner, 1987). It is the mediator of the major ATP-dependent non-lysosomal protein degradation pathway (for a review, see Hershko and Ciechanover, 1986 or Hershko, 1988a,b) and a normal component of the microtubule network (Murti et al., 1988). On the cell surface ubiquitin is believed to have a role in cell surface recognition and potentially in membrane protein turnover; antibodies to ubiquitin detect the lymphocyte homing receptor (Siegelman et al., 1986), the platelet-derived growth factor receptor (Yarden et al., 1986), growth hormone receptor (Leung et al., 1987), and several other unidentified external membrane proteins (Siegelman et al., 1986). Ubiquitin has also been associated with regulation of the heat shock response (Munro and Pelham, 1985) and with an autocrine growth factor (Okabe et al., 1986). In most of its roles, monomeric ubiquitin is conjugated to another protein (via its carboxy terminus, to epsilon amino groups) and may act as a recognition signal to guide distinct classes of specialized proteins and enzymes. Recently it has been shown that ubiquitin may itself possess proteolytic activity (Fried et al., 1987). The X-ray structure of ubiquitin has been determined to a resolution of 0.18 nm (VijayKumar et al., 1987a,b) and reveals a compact globular protein with a carboxy terminal tail, a hydrophobic core, and extensive hydrogen bonding.  2  1.1  Historical Perspective The amino acid sequence of ubiquitin was determined by Schlesinger in 1975,  shortly after Goldstein's discovery of ubiquitin (Goldstein et al., 1975). The sequence of the first complete ubiquitin gene, from S. cerevisiae. was published in 1984 (Ozkaynak et al., 1984). Ubiquitin was initially crystallized in 1985 by Vijay-Kumar et al., who has since refined the X-ray structure to 0.18 nm resolution (Vijay-Kumar et al., 1987). The proton N M R spectrum of ubiquitin has recently been completely assigned by Ecker et al.(1987). The history of the study of ubiquitin is one of accidental discoveries. Ubiquitin was initially discovered in 1975 by Goldstein et al. who purified it as a factor capable of inducing B-lymphocyte differentiation. Because antibodies to this factor crossreacted with proteins present in organisms as diverse as mammals, yeast, and plants, this protein was named ubiquitous immunopoietic polypeptide, or ubiquitin. Ubiquitin was "re-discovered" in 1977 by two groups who were studying A24, a rat liver chromosomal protein which was thought to be involved in gene regulation (Goldknopf and Busch, 1977; Hunt and Dayhoff, 1977). Protein A24 proved to be ubiquitin covalently linked to histone H2A via an isopeptide linkage. Ubiquitin resurfaced in 1980 as APF1, an ATP-dependent proteolysis factor found in rabbit reticulocyte lysates by Hershko et al.. The ubiquitin gene was initially cloned from chicken by "accident" when sequencing cDNA clones enriched for heat shock genes (Bond and Schlesinger, 1985). The surprising discovery that ubiquitin was present on the cell surface came from attempts to clone the lymphocyte homing receptor (Siegelman et al., 1986). Studies focussed specifically on characterizing ubiquitin have led to more surprising results, including the existence of polycistronic ubiquitin mRNAs in eukaryotes (e.g. Ozkaynak et al., 1984), the existence of ubiquitin fusion genes (e.g. Ozkaynak et al., 1987), and the intrinsic proteolytic activity of ubiquitin (Fried et al., 1987).  3 1.2  Structure of Ubiquitin All animal ubiquitins lack cysteine and tryptophan and have only one  methionine, one histidine, and one tyrosine. There are eleven acidic and eleven basic residues in addition to a single histidine. The only unusual feature of the primary structure is the presence of adjacent prolyl residues at positions 37 and 38. The few regions of variability in amino acid sequence between yeast, animal, and plant ubiquitin are clustered in one region of the tertiary structure opposite to the carboxy terminus (see Figure 1). There is every indication that the X-ray crystal structure and the proton NMR solution structure (Ecker et al., 1987) of ubiquitin are identical. From the ubiquitin crystal structure (Vijay-Kumar et al., 1987a, b; see also Figure 1) it can be seen that the ubiquitin protein is a single domain globular structure, made up of five strands of (3pleated sheet and three and one-half turns of alpha helix. About 90% of the polypeptide chain is involved in hydrogen bonded secondary structure. This extensive bonding and the hydrophobic core account for the pronounced thermal stability of this protein. In the folded ubiquitin molecule, the last four carboxy terminal amino acids protrude from the core of the protein in order to be accessible for formation of conjugates with the ubiquitin activating enzyme E l , ubiquitin conjugating enzyme E2, and with recipient protein substrates. Most of the lysyl and argininyl residues are buried or sterically blocked by flanlcing large or acidic side chains. On the surface of the molecule there is a partitioning of acidic residues to one face and basic residues to another. There is also a patch of hydrophobic residues on the surface of the protein near the carboxy terminus. These 'landmarks' may be recognized by the various enzymes that bind to and metabolize ubiquitin and its derivatives.  Figure 1. Evolutionary variations in ubiquitin structure. Positions of amino acids that are evolutionarily variant in cloned ubiquitin genes are represented on a ribbon approximation of the three dimensional structure of the human ubiquitin gene. The leucine-arginine-glycine-glycine carboxy terminus protrudes upward from the globular portion of ubiquitin in the diagrams, while the amino terminus is buried. Prominent parallel and antiparallel |3-sheets and an alpha helical stretch are evident. The molecule is viewed from the "back" orientation as per Vijay-Kumar et al. (1987). Amino acids (labelled, and represented by their side chains) variant between all published ubiquitin sequences are indicated. Amino acids variant between yeast, human, nematode, and barley ubiquitins are colored blue.  5  1.3  Structure-Function Relations Three main classes of enzyme interact with ubiquitin in proteolysis : E l  enzyme(s) which activate ubiquitin; E2 enzymes, which catalyze conjugation of activated ubiquitin to protein substrates; and E3 proteolytic substrate recognition enzymes (ubiquitin mediated proteolysis is discussed in detail in Section 1.4). In order to localize regions important for ubiquitin function, investigators have assayed naturally occurring and artificially created amino acid sequence variants and a variety of chemically modified ubiquitin derivatives in the presence of the enzymes of the protein degradation system. Specifically, these modified ubiquitins have been assayed for (1) their ability to be activated by E l (2) their ability to be conjugated to potential substrates of the proteolytic system and (3) their ability, once conjugated to substrate, to direct proteolysis of the substrate.  1.3.1  Natural Variants It has been demonstrated that natural oat or yeast ubiquitin have the same  activity in proteolytic assays as human ubiquitin or human ubiquitin with a single site-directed mutation, suggesting that the four positions of evolutionary variability between these ubiquitins are not crucial in the proteolytic function of ubiquitin (Wilkinson et al., 1986). This also suggests that the natural variants of ubiquitin are not the result of mutation followed by second site compensation; rather, the conservative substitutions in these amino acid positions are tolerated by the enzymes that utilize ubiquitin as a cofactor for proteolysis. It should be noted that the four amino acid residues which vary in the above ubiquitins may play a role in non-proteolytic extracellular functions of ubiquitin; there is evidence that human ubiquitin induces the differentiation of both T-precursor and B-precursor cells (Goldstein et al., 1975; Audhya and Goldstein, 1985).  6  1.3.2  Site-Directed Mutants of Ubiquitin  Mutations Affecting Ubiquitin Structure  The site-directed mutants of ubiquitin prepared and characterized functionally by Ecker et al. (1987) provide the most comprehensive data on ubiquitin structurefunction relationships. Ecker's group chemically synthesized an animal ubiquitin gene as a series of cassettes, making it easy to construct a series of single or double site-specific mutations to selectively perturb various regions of the molecule. Mutant ubiquitins were expressed in large quantity in E. coli (Ecker et al., 1987) or in yeast (Ecker et al., 1989b). Proton N M R was used to study the solution structure and folding of each mutant, and in vitro proteolysis assays were used to study function (Ecker etal., 1987). Mutations were constructed which focussed on the effects of changes on the protein surface, hydrophobic core, and tail region. Amino acid residues 72 to 76 of ubiquitin are highly mobile in solution; hence deletions and mutations in this region only alter the steric and chemical nature of the carboxy terminus, since it has no defined structure. However, a glycine-76 to alanine conservative change completely inactivated ubiquitin activity in proteolysis. Mutations in the ubiquitin tail destroy the biological activity of ubiquitin as a cofactor in proteolysis because the E l enzyme is unable to recognize the mutant ubiquitin tail. A tyrosine-59 to phenylalanine mutant was 70-100% as active as wild type ubiquitin in proteolysis assays, demonstrating that the hydroxyl group of tyrosine-59 is not necessary to stabilize the loop containing residues 50 to 60 as previously predicted from the crystal structure (Vijay-Kumar et al., 1987a,b). A histidine-68 to lysine mutant was only 30% as active as wild type ubiquitin in proteolysis assays, demonstrating the essential interaction of this histidine residue with the E l enzyme. When leucine-67 and leucine-69 were both converted to asparagine, the resulting mutant ubiquitin  7  was completely inactive in proteolysis. The two mutations are buried deep in the hydrophobic core of the protein, and may disrupt folding or, at the least, result in a substantially less stable hydrophobic core. The above results suggest that mutations adversely affecting points of contact between ubiquitin and the enzymes of the pathway diminish or abolish ubiquitin function in proteolysis. Activation of ubiquitin by E l appears to be essential for all ubiquitin-dependent proteolysis in in vitro assays, as no mutant was active m "in vitro proteolysis that was not active in the E l activation assay.  Mutations Affecting Ubiquitin Conformation  It has been proposed that conformational changes in ubiquitin, while conjugated to a target protein, determine whether that target protein is stable, deubiquitinated, or degraded. In order to study the requirement of conformation on the function of ubiquitin, Ecker et al. (1989a) inserted cysteine at strategic locations in the primary sequence of ubiquitin by site-directed mutagenesis, namely at two of phenylalanine-4, threonine-14, or threonine-66, allowing potential disulfide bridge formation across anti-parallel P-strands (4/14) or parallel pVstrands (4/66). The presence of disulfide bonds was confirmed by the absenced of titratable sulfhydryl groups in the purified mutant proteins. Disulfide bonds between pairs of these cysteine residues would severely limit potential ubiquitin conformations. Mutant 4/14 cross-links ubiquitin only at the first and second strands of the B-pleated sheet near the amino terminus; in contrast, 4/66 cross-links much more widely separated regions of the molecule, linking regions near the carboxy and amino terminus, thus more severely constraining the number of conformations available to ubiquitin. Substantial differences between the two disulfide mutants were seen in the proteolytic degradation assay; disulfide mutant 4/66 was consistently only 20-30% as active as wild type ubiquitin in degrading [ I]-BSA in vitro. The fact that disulfide 125  8  mutant 4/66 had low activity in the proteolytic assay relative to wild type ubiquitin suggests that disulfide cross-linked ubiquitin might be restricted from achieving a conformation that is active in signalling proteolysis. This result contradicts the hypothesis that ubiquitin is merely an inert flag for proteolytic enzymes. Rather, when conjugated to partially denatured proteins, the exposed hydrophobic region from the denatured protein could trigger a conformational change in ubiquitin; this altered conformation of ubiquitin might be recognized by the protease and signal substrate degradation (Wilkinson, 1987).  1.4  Ubiquitin and Protein Degradation Protein breakdown is responsible for essential cellular functions such as the  modulation of the levels of key enzymes and regulatory proteins and removal of abnormal proteins that arise through biosynthetic errors or postsynthetic damages. The process is highly selective : the half-lives of individual proteins vary from several minutes to many days (Goldberg and St. John, 1976). The major eukaryotic system for selective protein degradation in the cytoplasm is the ubiquitin pathway, in which proteins are committed to degradation by their ligation to ubiquitin. The central role of ubiquitin in cytosolic non-Iysosomal ATPdependent protein breakdown in reticulocytes was discovered by Hershko et al. in 1980. A variety of experiments have supported Hershko's initial model and have demonstrated the widespread occurrence of the ubiquitin degradation system in mammalian, yeast and plant cells : (1) in reticulocytes and in Ehrlich ascites cells the rapid degradation of amino acid analog-containing abnormal proteins is accompanied by a marked increase in the formation of ubiquitin-protein conjugates (Hershko et al., 1982). (2) a defect in the ubiquitin activating enzyme (El) renders mammalian cell line ts85 temperature sensitive - at the restrictive temperature there is a nearly  9  complete cessation of the degradation of abnormal and rapidly turning-over proteins (Finley et al., 1984). (3) in yeast cells, rapidly degraded derivatives of f3-galactosidase are converted to ubiquitin conjugates at a higher rate than slowly degraded derivatives (Bachmair etal., 1986). (4) in plant cells, exposure to a brief light pulse converts the photoreceptor phytochrome to another form that is degraded 100-fold faster than the original species - during this phase markedly increased levels of ubiquitin-phytochrome conjugates are observed (Shanklin et al., 1987; Jabben et al., 1989; Shanklin et al.,1989).  1.4.1  Mechanism of Ubiquitin-Mediated Protein Degradation The current model for the enzymatic steps in this pathway has been reviewed by  Hershko (1988a). Hershko's model is diagrammed in Figure 2 and is explained below. The initial reaction in the ligation process is the activation of the polypeptide by a specific ubiquitin-activating enzyme, designated E l (step 1 in Figure 2). E l catalyzes a two-step reaction sequence in which ubiquitin adenylate is first formed with the displacement of PPi from ATP, followed by the transfer of glycine-76activated ubiquitin to a thiol site of the enzyme, with the release of AMP. The energy-rich El-ubiquitin thiol ester has been shown to be the donor for conjugate formation with proteins. For this process, two further enzymes, designated E2 and E3 were found to be necessary. E2 has a ubiquitin-carrier function by virtue of an essential thiol site : it accepts activated ubiquitin from E l by transacylation (step 2) and then transfers it to amino groups of proteins. Pickart and Rose (1986) showed that there is a family of at least five E2 proteins, only one of which participates in E3-dependent ligation to substrates of the protein breakdown  10  ATP  AMP + PPi  Conjugates  (Type I — 111 binding sites)  Figure 2. The ubiquitin-mediated system of non-lysosomal protein breakdown. The cytoplasmic ubiquitin protein degradation model of Hershko (1988a) is presented. See the text for details. Ub : ubiquitin; CF1, CF2, CF3 : conjugatedegrading factors 1 to 3, respectively.  11 system. Some other species of E2 can transfer ubiquitin directly to histones in an E3independent process (step 3). E3 appears to have a central role in the selection of proteins suitable for degradation. Proteins with suitable structures are first bound to specific protein-binding sites of E3 (step 4) and only afterwards is ubiquitin transferred from E2-ubiquitin to the substrate (step 5). In this process, many units of ubiquitin may be ligated to the protein. Proteins conjugated to ubiquitin are degraded to small peptides by an ATPdependent multicatalytic protease complex or proteasome(Waxman et al., 1987). The mode of action of this protease and the role of ATP in this process are not known, although it has been speculated that this protease is a cysteine protease. When extracts from ATP-depleted reticulocytes are used for enzyme preparation, the high molecular weight protease is not present. Instead, three different factors have been resolved which are all required for conjugate breakdown (CF1-3). Upon addition of ATP, these three factors form a multienzyme complex which degrades ubiquitin conjugates. In addition to complex formation (step 6), ATP is probably involved in peptide bond cleavage as well (step 7), since ATP is continuously required for protease activity after the complex has been formed. The last reaction in the pathway (step 8) is presumably the release of free ubiquitin since the regeneration of reusable ubiquitin has been shown to occur in ubiquitin-dependent protein breakdown. Several of the enzymes involved in regenerating free ubiquitin, termed "ubiquitin protein hydrolases", have been identified.  1.4.2  Selectivity of Ubiquitin-Mediated Protein Degradation  Protein breakdown is elaborate and energy-consuming and hence must be a highly selective process. In order to recognize a protein as a potential substrate for proteolysis, the ubiquitin ligase system should be able to recognize two types of  12  structural feature : one is for the recognition of native protein structures and is required to distinguish between normal proteins of different half-lives, whereas the second would serve to recognize and remove damaged proteins.  Degradation of Proteins with a Free Amino Terminus  The vast majority of assays of ubiquitin-mediated proteolysis have been done using reticulocyte lysate Fraction n , a partially fractionated and reconstituted ubiquitin proteolytic system, which appeared to accurately reflect the in vivo system (Ciechanover et al., 1978). Studies with this system indicated that a free amino terminal a-amino acid is necessary for the ligation of protein substrates to ubiquitin and their degradation by the ubiquitin system (Hershko, 1984). Subsequently it was found that the nature of the amino terminal residue has a strong influence. Ferber and Ciechanover (1987) have observed that tRNA-dependent transfer of arginine to the amino terminus of protein substrates with an acidic amino terminal position converted them to good substrates for the ubiquitin system. Bachmair et al. (1986) systematically changed the amino terminal residue of Bgalactosidase by site-directed mutagenesis. Due to the design of the expression vector used in the experiment, all derivatives contained a forty amino acid residue extra carboxy terminal segment derived from an internal part of the lac repressor. Following the expression of the various derivatives in yeast cells, dramatic differences in their half-lives (ranging from 2 minutes to 30 minutes to over 12 hours) were observed. These workers proposed that an "N terminal rule" governs protein degradation by the ubiquitin system. In subsequent work, Bachmair and Varshavsky (1989) observed that variations in the amino terminal residue of dihydrofolate reductase did not much influence the stability of this protein expressed in yeast unless the extra carboxy terminal fragment of the B-galactosidase construct was fused to it. Because this segment presumably lacks structure, it was  13 suggested that an unstructured, mobile 'random coil' portion of a protein may be essential to bring a ubiquitin acceptor lysine residue to sufficient proximity to the ubiquitin ligase. More recent experiments have suggested that the amino terminal amino acid residue is not the only, and possibly not the predominant, structural determinant of a proteolytic substrate for the ubiquitin proteolytic system from reticulocytes. Reiss et al. (1988) showed that many good substrates have amino terminal residues that would be classified in the stabilizing category by the N terminal rule. Hershko et al. (1986) found that RNase A (which has a 'destabilizing' amino terminal lysine) was not a good substrate for the ubiquitin ligase system from reticulocytes, but it could be converted to an excellent substrate following the oxidation of its methionine residues to sulfoxide derivatives. A general correlation has been observed between the binding of different proteins to E3 and their susceptibilities to ligation to ubiquitin (Reiss et al., 1988). Thus, proteins with free amino termini (e.g. yeast cytochrome c or enolase) bound more strongly to purified E3 than do their N -acetylated counterparts, and proteins a  with oxidized methionyl residues bound more strongly to E3 than do the unmodified proteins (Hershko et al., 1986). Apparently there are two sites (I and n) on E3 with specificity for basic or bulky hydrophobic amino terminal amino acids. The major interaction of Type HI substrates is apparently not at the amino terminus, but at some other region of the protein molecule. It is interesting to note that protein unfolding and denaturation do not seem to have a major influence on susceptibility to ubiquitin ligation. Drastic oxidation of ribonuclease with performic acid (which cleaves disulfide bonds and converts the protein to a random coil configuration) has much less effect on increasing susceptibility to ubiquitin ligation than does milder oxidation which affects methionyl residues only (Hershko et al., 1986). Similarly, if one completely reduces  14  the disulfide bonds of BSA or completely denatures BSA or B-lactoglobin (by reduction and alkylation), one does not increase their rate of degradation by reticulocyte lysates (Evans and Wilkinson, 1985). These results are in contrast to the marked effects of protein unfolding on their susceptibility to the action of proteases. It may be assumed that drastic unfolding and denaturation of proteins occur only rarely in cells, arguing against the 'random coil' portion of the N terminal rule.  Degradation of Proteins with a Modified N Terminus  Most studies of the ubiquitin system and analysis of its components have been performed with the in vitro reticulocyte fractionated extract system. Naturally occurring proteins with acetylated amino termini were not degraded by this system. Because the vast majority of cellular proteins are blocked in the amino terminal position (e.g 80% of the soluble proteins in Ehrlich Ascites tumor cells have blocked amino termini : Brown and Roberts, 1976), a potential role for the ubiquitin system in their turnover in vivo is physiologically important. Mayer et al. (1989) compared the degradation of proteins with blocked termini in whole lysate versus Fraction n extract; N-acetylated proteins were only degraded in whole reticulocyte lysate. Thus an essential component that is required for degrading N  a  blocked proteins is removed or inactivated during the partial  separation of the reticulocyte lysate. Since whole lysate but not Fraction II can conjugate ubiquitin to Fraction IIlabelled (N-acetylated) histone H2A, it is likely that fractionation of the extract causes loss of a ubiquitin conjugating factor. Evidence arguing against the factor simply being a deacetylase includes the observation that actin, which has four successive acidic groups after the acetyl group, is RNaseA-insensitive for degradation in whole reticulocyte lysate (Mayer et al., 1989). If the factor deacetylates the substrate and exposes an acidic amino acid, then tRNA would be required for the  15  tRNA-dependent ubiquitin degradation pathway; thus degradation of acidic exposed termini should be sensitive to RNaseA, which degrades t R N A . These results suggest that recognition capability does not seem to require an exposed a amino terminal residue and may be directed to a structural feature (or features) of the substrate that is distinct and possibly downstream from the amino terminal residue. The factor lost during fractionation may be a novel E3 (or may affect the function of the already described E3) which interacts with substrates that have free a amino termini, however, at regions distinct from the amino terminus.  1.4.3  Structure of Ubiquitin on Proteins Targeted for Proteolysis The structure of ubiquitin-protein conjugates has not been characterized  sufficiently, but it appears that many ubiquitin molecules are linked (via their carboxy terminal glycine residue) to epsilon amino groups of one or two specific lysine residues of a proteolytic substrate by isopeptide linkages. In addition, the formation of polyubiquitin chains has been shown to occur, in which one molecule of ubiquitin is linked to lysine-48 of another molecule of ubiquitin, which in turn is linked to the protein substrate. Experiments with ubiquitin mutated at lysine-48 to arginine (arginine cannot be ubiquitinated) indicate that the multi-ubiquitin chain in a targeted protein is essential for the degradation of abnormal proteins containing amino acid analogs in vivo.  1.5  Ubiquitin as a Protease The existence of specific ubiquitin-protein conjugates that are not rapidly  degraded suggests that ubiquitin may be playing other roles than as an inert flag for non-lysosomal protein degradation. H u m a n erythrocyte ubiquitin purified by Fried et al. (1987) was shown to have intrinsic proteolytic activity comparable to that of other well-characterized proteases (e.g. Lysobacter (Lys-C) protease, mouse  16  submaxillaris (Arg-C) protease, and Staphylococcus V8 (Glu-C) protease). Ubiquitin does not appear to be a typical serine protease and may function by a novel catalytic mechanism. Ubiquitin cleaves predominantly onjhe carboxy terminal side of leucine residues, although it cleaves at other residues (including glycine, threonine, and glutamine) after long-term incubation. Structural features in addition to specific amino acid sequence may play a role in defining ubiquitin cleavage sites. It is possible that ubiquitin functions only as a free protease. As a free protease, the activity of ubiquitin may be highly selective and regulated by calcium or other cofactors. Fried et al. speculate that ubiquitin may also be a functional protease when conjugated to other proteins and thus be regulated, in part, by this specific localization (i.e. the ubiquitin catalytic domain may be coupled to a substrate recognition domain). As a protease conjugated to other proteins ubiquitin could function in two modes (1) in cis, ubiquitin could process the conjugate to degrade the ligated protein or to generate a new functional component e.g. ubiquitin bound to platelet-derived growth factor receptor may degrade the receptor in cis and downregulate the receptor, or (2) in trans, ubiquitin could cleave a separate component that interacts with the conjugate e.g. histone uH2A or uH2B could act to inactivate repressor (s) bound to adjacent regulatory elements and hence control gene expression in eukaryotes.  1.6  Ubiquitin and the Heat Shock Response A primary function of protein breakdown in eukaryotic and prokaryotic cells is  to eliminate nonfunctional or denatured polypeptides whose accumulation in vivo could be toxic and deleterious to cell survival. Proteins with highly abnormal structure may result from nonsense mutations, insertions or deletions, missense mutations, biosynthetic errors, incorporation of amino acid analogs into proteins, or postsynthetic damage to normal cell constituents.  17  The ability of cells to selectively hydrolyze abnormal proteins increases in a variety of stressful environmental conditions, for example exposure to elevated temperatures. Under these conditions, abnormal proteins may be generated in large amounts. During heat stress, the pattern of gene expression changes, and cells overproduce a characteristic set of proteins, the heat shock proteins (reviewed in Schlesinger et al., 1982; Craig, 1985; Goff et al-, 1988; Pardue et al., 1988). One important consequence of producing the heat shock proteins is to increase the cell's capacity to eliminate abnormal polypeptides and thus to help protect against the environmental conditions that generate them.  1.6.1  Heat Shock in Prokaryotes  The best characterized of the E. coli ATP-dependent endoproteases is protease La, the product of the Ion gene (Chung and Goldberg, 1981). A variety of observations suggest that the level of protease La is a critical factor determining the rate of breakdown of abnormal proteins : (1) mutations in the Ion gene reduce degradation of amino acid analogcontaining polypeptides by up to four-fold (Chung and Goldberg, 1981), (2) mutations in Ion also leads to decreased breakdown of certain short-lived regulatory proteins, and as a result, these mutants show a variety of unusual phenotypic alterations (Mizusawa and Gottesman, 1983) (3) when the cellular content of protease La is increased experimentally (Goff and Goldberg, 1987), the capacity to degrade abnormal proteins is enhanced. The exact role of the E. coli heat shock proteins in the heat shock response is not well understood but recent evidence suggests a role for a number of these proteins in the pathway of protein degradation. The product of the rpoH (htpR) gene has been cloned and shown to function in vitro and in vivo as a positive regulator of the heat shock genes (Neidhart et al., 1983; Tobe et al., 1984, Bloom et al., 1986). The  18  htpR gene has been shown to function as an alternative sigma factor (o 32) that redirects R N A polymerase to heat shock genes under conditions of thermal stress (Grossman et al., 1984). Increasing the amount of htpR in E. coli causes induction of the heat shock proteins. Protease La is itself a heat shock protein (Goff and Goldberg, 1987) and as such is likely a key player in degrading abnormal proteins which accumulate during the stress response. It should be noted that increased amounts of abnormal intracellular protein caused by amino acid analogs or puromycin have also been shown to induce Ion transcription (Goff and Goldberg, 1985). Goff et al. (1988) have proposed a model explaining the role of the cellular proteolytic system in the heat shock response. According to this model, the generation of abnormal proteins by various conditions that elicit a heat shock response results in saturation of the proteolytic enzymes of the cell and thereby protects a 32 from proteolysis. The buildup of this positive regulator enhances transcription of the heat shock genes. Since protein degradation is enhanced following increased synthesis of protease La and perhaps other protease (s) under control of the rpoH locus, the system should return to its original level or a near basal level once the abnormal proteins are degraded. Under these conditions a 32 would regain its original short half-life, resulting in a low rate of transcription of the heat shock genes. Accordingly, when cells become adapted to growth at high temperatures, the half-life of o 32 returns to that seen at 30°C.  1.6.2  Heat Shock in Eukaryotes  Heat shock transcription factors function only in heat shocked cells, but they must also be present in unstressed cells since inhibitors of protein synthesis do not block the increase in transcription of the heat shock genes following a temperature shift (Craig, 1985). Thus the heat shock transcription factor (HSTF) appears to  19  undergo some kind of activation prior to the increase in transcription of heat shock genes. A wide variety of treatments can induce the heat shock response in eukaryotes (reviewed in Burdon, 1986; Pardue et al., 1988). The only obvious common feature of these various inducers is that they can promote either the denaturation of preexisting cellular proteins or the synthesis of denatured proteins. Several observations indicate that the regulation of eukaryotic heat shock gene expression is somehow linked to the accumulation of abnormal proteins : (1) A mouse cell line, ts85, is temperature sensitive in the conjugation of ubiquitin to proteins (Finley et al., 1984) and has a defect in the degradation of short-lived proteins (Ciechanover et al., 1984a,b). Because of this defect in proteolysis, the cells are likely to accumulate denatured proteins at the nonpermissive temperature. These mutants also synthesize elevated levels of heat shock proteins under such conditions (Finley et al., 1984). (2) Deletion of the heat-inducible polyubiquitin gene from yeast renders these cells temperature sensitive for growth and viability. Yeast can still grow at low temperatures if they contain a normal content of ubiquitin derived from other ubiquitin genes. Increased production of ubiquitin is essential under stressful conditions when the concentration of free ubiquitin falls (Finley et al., 1987). (3) In the flight muscles of mutant Drosophila strains that synthesize an abnormal form of actin UI, heat shock proteins are produced constitutively (Hiromi etal., 1986; Ballet al., 1987). (4) Ananthan et al. (1986) microinjected proteins in their native or denatured forms into Xenopus oocytes carrying a B-galactosidase reporter plasmid with a hsp70 gene promoter. The denatured proteins caused a ten-fold increase in Bgalactosidase activity, establishing a causal link between the accumulation of denatured proteins and induction of the heat shock proteins.  20  Exactly how the presence of denatured proteins causes an activation of heat shock genes in eukaryotes is a matter of speculation (see Burdon 1986; Ananthan et al., 1986). By analogy to the bacterial heat shock regulatory system the large amount of abnormal protein accumulating under stressful conditions may exceed the degradative capacity of the cell's proteolytic machinery. These abnormal proteins should compete with HSTF, which is otherwise rapidly degraded via ubiquitin conjugation or direct proteolytic attack. Consequently, HSTF is more stable and accumulates. By this model most of HSTF is not functional in the unstressed cells because either it is rapidly degraded or another protein required for its activation or production is rapidly degraded. During heat shock active HSTF builds up and promotes heat shock gene transcription either because it is stabilized or because the activating enzyme is more stable. It is also possible that (1) HSTF assumes an active conformation upon interaction with abnormal proteins or (2) a separate system senses the accumulation of abnormal proteins and activates HSTF (Munro and Pelham, 1985). Heat shock is known to cause a wave of ubiquitination of cellular proteins, which can deplete the pool of free ubiquitin and decrease the levels of uH2A (Parag et al., 1987; Carlson et al., 1987). It should be noted, however, that there is no evidence of ubiquitinated forms of heat shock transcription factor in yeast or HeLa cells; rather it appears that this factor may be regulated by phosphorylation (Sorger et al., 1987 and 1988) and/or multimerization (Perisic et al., 1989; Sorger and Nelson, 1989). The heat shock response in Caenorhabditis elegans was initially characterized by Snutch and Baillie (1983) and Russnak et al. (1983). There has been extensive characterization of both the hsp70 genes [reviewed in Heschl and Baillie (1989, 1990)] and hsp!6 genes (reviewed in Candido et al., 1989). A putative heat shock transcription factor has not yet been isolated from C. elegans. Both major heat shock  21  gene classes are represented as gene families : the hsp70 family contains both inducible and constitutively expressed genes, while the hspl6 genes appear to be purely inducible. The powerful genetics available in C. elegans should allow extensive genetic characterization of the metazoan heat shock response.  1.7  Ubiquitin and Chromatin Even though substantial information exists concerning the presence of  ubiquitin in chromatin, little insight into its functional role has yet been achieved. Presumably, ubiquitination of histones affects chromatin structure, but it is uncertain how this modification influences transcriptional or mitotic processes. Ubiquitin has been found conjugated to H2A by an isopeptide bond through its carboxy terminal glycine-76 to the epsilon amino group of lysine-119 of H2A (Hunt and Dayhoff, 1977). Ubiquitin has subsequently been found on H2B, attached to lysine-120 (Thorne et al., 1987), but no ubiquitin derivatives of histones H I , H3, or H4 have been detected. Approximately 11% of the total amount of H2A and 1% of H2B is ubiquitinated (Bonner et al., 1988). uH2A has been detected in most organisms which have been assayed; however, in contrast to other eukaryotes, yeast growing under optimal conditions seem to lack uH2A completely (<0.1%). When yeast are grown at elevated temperature, a significant fraction of the H2A becomes ubiquitinated (Finley et al., 1988). This novel finding may lead to new insights regarding the role of histone ubiquitination.  1.7.1  Ubiquitinated Histones and the Cell Cycle The most definitive experiments on the behaviour of ubiquitinated histone  during the cell cycle were done by Mueller et al. (1985), who studied the uH2A and uH2B in Physarum polycephalum, a macroplasmodial mold in which the nuclear division cycle is very synchronous. Ubiquitinated histones were present during  22  prophase, anaphase and telophase as well as during S, G | , and G phases of the cell 2  cycle, but disappeared precisely at metaphase. This work confirmed earlier work by Matsui et al. (1979) that de-ubiquitination and ubiquitination occur during chromosome condensation and decondensation, respectively. De-ubiquitination is not necessarily the trigger for chromosome condensation however; evidence from ts85 cells argues against this model. At the non-permissive temperature ts85 chromatin remains dispersed despite the loss of all uH2A and uH2B (Ciechanover et al., 1984a,b).  1.7.2  Ubiquitinated Histones and Transcription  The association of ubiquitinated histones with transcriptionally active D N A sequences remains a completely open question despite numerous attempts to demonstrate a relationship. This is mainly because the experiments designed to show an association generate indirect correlational data. Levinger and Varshavsky (1982) found that a high percentage of the Drosophila hsp70 and copia gene D N A sequences co-migrated with mono- and di-ubiquitinated mononucleosomes.  In  heat shocked cells there was complete loss of D N A sequences hybridizing to an hsp70 gene probe, suggesting it was devoid of mononucleosomes (let alone ubiquitinated ones) under conditions of high transcription rate. However, treatment of mononucleosomes from mouse plasmacytoma cells with ubiquitin isopeptidase (which should remove ubiquitin from uH2A) and probing for active immunoglobulin kappa chain gene D N A demonstrated that ubiquitin may not be the important protein enriched in 'active nucleosomes' (Huang et al., 1986). To add to the confusion, untranscribed satellite D N A sequences may be associated with ubiquitinated histones (Levinger, 1985).  23  1.8  Ubiquitin and the Cell Surface The presence of ubiquitin on cell surface proteins was first discovered during  the isolation and cloning of the lymphocyte homing receptor (Siegelman et al., 1986). Monoclonal antibodies were raised against clonal lymphoid cell lines bearing the homing receptor; one such monoclonal, MEL-14, detected a cell surface determinant which proved to be the extracellular ubiquitin moiety of the lymphocyte homing receptor (St. John et al., 1986). Ubiquitin has also been found conjugated to the cytoplasmic domain of the platelet-derived growth factor receptor (Yarden et al., 1986). Examples of other known ubiquitinated cell surface proteins include the growth hormone receptor (Leung et al., 1987) and the Leu8 protein found on T-lymphocytes (Camerini et al., 1989). Of the several roles postulated for ubiquitin at the cell surface, perhaps the most interesting is that advanced by Butt et al. (1989). They have speculated that ubiquitin conjugation may facilitate the folding of conjugated cell surface receptors; ubiquitination may keep these hydrophobic proteins soluble and facilitate their translocation to the cell membrane.  1.9  Genetics of the Ubiquitin System The ubiquitin system was first approached experimentally using purely  biochemical methods. Genetic techniques provide not only an alternative way to address previously posed questions in this field, but also a strategy for the identification and dissection of the physiological functions of ubiquitin. Genetic studies have focussed primarily on yeast, whose ubiquitin system closely resembles that of mammals. While ubiquitin genes have now been cloned from a number of organisms, mutational analysis of the ubiquitin system has been restricted to yeast and the mammalian cell line ts85.  24  1.9.1  ts85 A detailed understanding of a biochemical pathway ultimately requires means  to perturb the pathway specifically in vivO. For example, the physiological substrates of the ubiquitin-dependent proteolytic pathway can be inferred by determining which proteins are stabilized when ubiquitination is selectively inhibited in vivo. The conditional ubiquitination phenotype of the murine cell line ts85, which results from a thermolabile ubiquitin activating enzyme (Finley et al., 1984), has been used to show that the bulk of short-lived intracellular proteins is degraded through a ubiquitin-dependent pathway (Ciechanover et al., 1984a,b). Amino acid analog-containing proteins, puromycyl peptides, and heat damaged proteins are similarly degraded by an ubiquitin-dependent pathway in ts85 cells, unlike most long-lived proteins (Ciechanover et al., 1984).  1.9.2  Yeast Ubiquitin Genes Analysis of ubiquitin in yeast takes advantage of the powerful molecular  genetics of this organism. In yeast, ubiquitin is encoded by a complex multigene family (see Figure 3; Ozkaynak et al., 1984; Ozkaynak et al., 1987). One striking feature of these genes is that while all of them contain ubiquitin coding sequences, none of them encode mature ubiquitin. Instead, the four yeast genes all encode hybrid proteins in which ubiquitin is fused at its carboxy terminal sequence either to itself, as in polyubiquitin, or to unrelated ("tail") amino acid sequences. Thus, in yeast ubiquitin is invariably a product of post-translational processing of precursor proteins.  Yeast Polyubiquitin Gene  Genetic analysis of the polyubiquitin gene UBI4 in yeast has been carried out by Finley et al. (1987). Deletion of the yeast UBI4 gene results in mutants which appear  25  Class  Translation Product  u u u uu  Final Product  Yeast Gene  u  UBI4 UB11 -UBI2  ll  ui  u +1  ll  u w<  u  22  UBI5  Figure 3. The ubiquitin genes of the yeast Saccharomyces cerevisiae.  The two classes of ubiquitin hybrid genes and their gene products are indicated (after Warner, 1989). U : ubiquitin monomer.  26  indistinguishable from wild type cells in terms of normal vegetative growth rate and the steady state levels of free ubiquitin in ubi4 mutant cells. Thus UBI4 appears to play a limited role in the normal physiology of exponentially growing yeast cells. However, the major transcript of the UBI4 locus is induced either by heat stress or in stationary phase cultures, suggesting that UBI4 expression may be under the control of the stress regulatory network. The UBI4 promoter region also contains a sequence with strong homology to the heat shock regulatory element sequence required in cis for stress inducibility of other eukaryotic genes. The ubi4 deletion mutants are hypersensitive to either carbon or nitrogen starvation. When yeast cells are starved for essential nutrients, they respond by repressing the initiation of new rounds of cell division and entering into stationary (G ) phase of the cell cycle. Starved ubi4 mutants appear to be capable of entering Q  stationary phase, thus the UBI4 gene does not appear to be required for cells to enter stationary phase but rather may be required for their survival once in stationary phase. Stress proteins can be induced not only by heat and starvation but also by a variety of toxic compounds (reviewed in Pardue et al., 1989). The plating efficiency of ubi4 deletion mutants on canavanine-containing plates is 1000-fold lower than that of wild-type strains (Finley et al., 1988). This hypersensitivity to amino acidanalogs strongly suggests that degradation of abnormal proteins is ubiquitindependent in yeast, as in murine ts85 cells. Another striking phenotype of ubi4 deletion mutants is that, although ubi4/UBI4 diploids can sporulate normally, UBI4 is subsequently required to maintain the viability (ability to germinate) of haploid spores. The ubiquitin system may be active even in mature yeast spores; alternatively in wild-type spores, but not in ubi4 spores, increased amounts of ubiquitin would either be stored within the  27  spore or produced upon germination via the UBI4 gene, thus accounting for the time-dependent loss of viability of ubi4 mutant spores.  Yeast Ubiquitin Hybrid Genes  Genetic analysis by Finley et al. (1989) has demonstrated that the small ubiquitin genes of yeast have a role distinct from that of the polyubiquitin gene. Deletions of these ubiquitin carboxy terminal extension (UbCEP) genes, unlike the deletion of UBI4, each confer a slow-growth phenotype. The strongest phenotype is seen in the ubi3 mutant, which has a doubling time of 6.8 hr as compared to 1.6 hr for wild-type cells under the same conditions. The slow growth phenotype of ubi3 mutants is largely due to a deficiency in ubi3 tail function : the level of free ubiquitin is only slightly reduced in ubi3 cells as compared to their wild-type counterparts and a plasmid that constitutively expresses free ubiquitin does not complement the ubi3 deletion. Single deletions of ubil or ubi2 confer milder but significant growth inhibition, while the ubil ubi2 double mutant is non-viable, suggesting that ubil and ubi2 together constitute an essential subfamily of the ubiquitin genes. The 52 amino acid tail protein encoded by the UBI2 gene is identical to that of UBIl. The 76 amino acid tail protein encoded by the UBI3 gene bears little similarity to that of UBIl and UBI2. Nonetheless, the three tails share a number of general structural features. The ubiquitin tail proteins are all basic and contain putative nuclear localization signals, although in the ubi3 protein this sequence is located at the beginning of the tail rather than at its end as in the ubil and ubi2 proteins (Ozkaynak et al., 1987). A TFHIa-like, putative nucleic-acid binding motif comprised of four cysteine residues is also present in the ubi3 tail (Ozkaynak et al., 1987). The yeast tail proteins have recently been shown to be ribosomal proteins (Finley et al., 1989). The 52 amino acid ubiquitin tail protein is found in the 60S ribosomal subunit in yeast. The 76 amino acid ubiquitin tail has been identified as  28  the 40S subunit protein S37 in yeast (Finley et al., 1989), whose mammalian homolog is ribosomal protein S27a (Redman and Rechsteiner, 1989). Ribosomal proteins are known to turn over rapidly unless they are assembled in the ribosome (Maicas et al., 1988). Ubiquitin might protect the ribosomal proteins from degradation. Butt et al. (1988) showed that ubiquitin fusion to an unstable protein such as metallothionein in E. coli led to its stabilization. UbCEP, also a metalloprotein, is perhaps relatively unstable in the absence of metal.  Yeast Ubiquitin-Protein Ligase System Genes  Enzymatic components of the S. cerevisiae ubiquitin-protein ligase system have been purified by Varshavsky's group (Goebl et al., 1988; Jentsch et al., 1987) using the ubiquitin affinity chromatography technique developed for purification of ubiquitin —  ligase system components from mammalian erythrocytes (Pickart and Rose,1985). There are at least five distinct E2 enzymes in the yeast S. cerevisiae (Jentsch et al., 1987). The gene encoding the E2-30K enzyme has been cloned and designated UBC1 (for UBiquitin-Conjugating enzyme). E2-20K (UBC2) was also cloned and found to be encoded by a previously cloned yeast gene, RAD6 (Jentsch et al., 1987). In addition the yeast cell cycle gene CDC34 has been identified as a UBC gene (Goebl etal., 1988). In yeast, the RAD6 gene plays a central role in one of the three epistasis groups of genes involved in D N A repair and mutagenesis (Lawrence, 1982; Prakash, 1989). RAD6-mediated ubiquitination of target proteins could induce structural alterations in chromatin, allowing access of enzymes to the underlying D N A lesions.  1.9.3  Ubiquitin Genes from Other Organisms  The gene structure of ubiquitin has also been remarkably conserved in evolution. Ubiquitin genes are usually arranged in two basic forms in most  29  organisms (typified by the yeast ubiquitin genes : see Figure 3). Class I is a polyubiquitin gene which encodes a polyprotein of up to 50 uninterrupted, tandemly repeated ubiquitins. Class II is a fusion between a single ubiquitin and one of two other sequences, of either 52 or approximately 80 predominantly basic amino acids, again conserved through evolution. Polyubiquitin genes have been isolated from a number of organisms including humans (Wiborg et al., 1985), chicken (Bond and Schlesinger, 1986), yeast (Ozkaynak et al., 1984), Drosophila (Arribas et al., 1986), Dictvostelium (Giorda and Ennis, 1987) and Arabidopsis (Burke et al., 1988); cDNAs corresponding to polyubiquitin mRNAs have been sequenced from Xenopus (Dworkin-Rastl et al., 1984) and barley (Gausing and Barkardottir, 1986). All display a characteristic spacerless, tandem array structure comprised of 228 bp repeats encoding 76 amino acids. This arrangement has presumably evolved by unequal crossing over and concerted evolution. Despite the large variation in gene structure and expression, the final product of monomeric ubiquitin is conserved. Ubiquitin carboxy terminal extension (UbCEP) genes have been identified in a number of species (e.g. yeast: Ozkaynak et al., 1987; Dictvostelium : Westphal et al., 1986; chicken : Mezquita et al., 1988; human : Lund et al., 1985). UbCEP genes are usually encoded as monomers, although they may occur as tandem repeats in trypanosomes (Swindle et al., 1988).  1.9.4  Ubiquitin-like Genes  Ubiquitin Cross-Reactive Protein(UCRP)  Treatment of the human lung carcinoma line A-549 or the mouse fibroblast line L929 with interferon induces the de novo synthesis of a 15 kDa protein immunologically related to ubiquitin (Haas et al., 1987). UCRP is composed of two domains, each of which bears striking homology to ubiquitin. The UCRP amino  30  Species  Number of loci  Human  3  Chicken  2  Xenopus  Val  1  1  tail  2  4  Tyr  3  >12  None  4  2  -15  ?  5  >4  >3  Lys  6  4  5  Asn  7  Yeast  Nematode  9  >2  Drosophila Barley  Number of repeats Extra Codon Reference  2  1  tail (52 aa)  8  1  tail (52 aa)  8  1  tail (76 aa)  8  Asp lie  9  tail (52 aa)  9  11 1  Figure 4.  Comparison of some cloned ubiquitin gene structures.  The repeat structure of the coding region and nature of the carboxy terminus of representative cloned ubiquitin genes is presented. References are as follows : 1. Wiborg et al., 1985; 2. Lund et al., 1985; 3. Bond and Schlesinger, 1986; 4. Dworkin-Rastl et aL 1984; 5. Arribas et al., 1986; 6. Gausing and Barkodottir, 1986; 7. Ozkaynak et al., 1984; 8. Ozkaynak et aL, 1987; 9. this thesis.  31 terminal domain of 80 residues is 29% similar to human ubiquitin; the carboxy terminal domain of 64 amino acid residues is 31% similar (if one allows for an error in D N A sequencing by the investigators which terminated translation prematurely, the carboxy terminal domain has all 76 amino acids of ubiquitin). GdX GdX was cloned as a single copy gene downstream from the glucose-6-phosphate dehydrogenase gene of the human X chromosome (Toniolo et al., 1988). The amino terminal 74 amino acids of GdX show 43% identity to human ubiquitin, however the characteristic Gly-Gly terminus of ubiquitin is lacking. The 80 amino acid carboxy terminus of GdX has no similarity to the tail of the yeast UBI3 class of fusion protein; however, a sequence similar to the carboxy terminal hormonogenic site of thyroglobulin is present.  Togavirus Ubiquitin  Bovine viral diarrhoea virus (BVDV) is a positive-stranded R N A togavirus. In the pathogenesis of mucosal disease, a non-cytopathic BVDV apparently mutates to a cytopathic biotype by taking up cellular sequences during a recombination event. Osloss strain BVDV has an insertion in a non-structural virus-encoded protein pl20. The sequence inserted into pl20 is 228 nucleotides and encodes a ubiquitin which differs in only 2 of 76 amino acids from the ubiquitin sequence conserved in all animals (Meyers et al., 1989).  Baculovirus Ubiquitin  Guarino (1990) has sequenced a portion of the baculovirus Autographa californica genome and identified a gene (v-ubi) which has 76% identity with ubiquitin. A polyubiquitin gene isolated and sequenced from Spodoptera  32  frugiperda, the host cell line for this baculovirus, is identical in amino acid sequence to animal ubiquitin. Most of the amino acid substitutions in v-ubi are conservative with respect to side chain, and many of the residues known to be important for ubiquitin function have been conserved.  1.10  C. elegans as a Model System Caenorhabditis elegans is a small free living soil nematode found commonly in  many parts of the world. It feeds primarily on bacteria and reproduces with a life cycle of about 3 days under optimal conditions. C. elegans is a simple organism, both anatomically and genetically. The adult hermaphrodite has only 959 somatic nuclei, and the adult male has only 1031. C. elegans has six haploid chromosomes. The haploid genome size is 8 x 10 bp or approximately eight times that of the yeast 7  S. cerevisiae. About 80% of the C. elegans genome is composed of single-copy sequences. Brenner (1974) estimated the total number of essential genes in C. elegans to be approximately 2000 based on the frequency of X linked lethal mutations. Clark et al. (1988) have revised this number to approximately 3500 essential genes in the C. elegans genome. Of special relevance to this thesis, the heat shock response of C. elegans has been extensively characterized [see Snutch and Baillie (1983) and Russnak et al. (1983)] and two major classes of heat shock gene have been studied in detail (reviewed in Heschl and Baillie (1989, 1990) and Candido etal. (1989). C. elegans is easily maintained in the laboratory; individual animals are easily manipulated and large numbers can be grown routinely in mass culture. The animals are transparent throughout the life cycle, which has allowed the complete cell lineage to be elucidated (Sulston et al., 1983). In addition, its small size has allowed a complete anatomical description of the animal and of the wiring of its nervous system at the electron microscope level (White et al., 1986).  33  C. elegans is well suited to genetic and molecular biological studies. Mutants are readily obtained following chemical mutagenesis or exposure to ionizing radiation. C. elegans has a well characterized transposon, Tel, which is useful for transposon tagging mutagenesis and cloning (e.g. Moerman et al., 1986). Homologous transformation by microinjection has recently been developed, allowing study of C. elegans gene expression in an appropriate context (Fire, 1986). The C. elegans genome physical map has been nearly completed, speeding acquisition of relevant D N A for gene analysis and mapping; indeed, researchers can now obtain specific segments of the C. elegans genome D N A by telephone!  1.11  Trans Splicing The splicing of exons from two distinct RNAs to form a single mRNA has  recently been shown to occur both in vivo (Murphy et al., 1986) and in vitro (Solnick, 1985). In these reactions, termed trans splicing, a 5' intron donor site in one R N A interacts with the 3' intron acceptor site in another R N A molecule (reviewed in Laird, 1989). In trypanosomes (e.g. T. brucei, L. enriettii, and C. fasciculata), trans splicing occurs with all mRNAs (Bruzik et al., 1988). The 5' ends of all trypanosome mRNAs consist of a capped, 39 nucleotide sequence termed the mini-exon or spliced leader which is derived from a 137 nucleotide R N A by trans splicing (Sutton and Boothroyd, 1986; Van der Ploeg, 1986). It has recently been shown that three of the four actin gene transcripts in the nematode C. elegans undergo trans splicing (Krause and Hirsh, 1987). C. elegans is the first example of a multicellular eukaryote which routinely carries out both trans and cis splicing of mRNA precursors in the nucleus. Hybrid arrest of translation experiments demonstrate that 10-15% of C. elegans mRNAs undergo trans splicing (Bektesh et al., 1988). trans splicing is apparently common in nematodes : the splice  leader sequence is found in Panagrellis redivivus, Haemonchus contortus, Anisakis spp., B. malayi, and Ascaris (Bruzik et al., 1988;Bektesh et al., 1988). The 22 nucleotide C. elegans splice leader SL1 is contained in a methylguanine7  capped, non-polyadenylated 100 nucleotide R N A (SL RNA) which is transcribed by R N A polymerase II from a region adjacent to the 5S ribosomal R N A gene (Nelson and Honda, 1989). There is also evidence of other SL RNAs in C. elegans besides SL1; Huang and Hirsh (1989) report the existence of a second splice leader (SL2) on the glyceraldehyde phosphate dehydrogenase gene. Although there is little primary sequence identity between the SL RNAs in nematodes or in other organisms such as trypanosomes, they can assume identical stem-loop secondary structures. The SL1 R N A of C. elegans also has a consensus binding site for the small nuclear ribonucleoprotein Sm(Van Doren and Hirsh, 1988; Thomas et al., 1988). The above properties suggest that the SL R N A might have a dual function in the trans-splicing process in which the 5' donated exon is covalently linked to an snRNA-like sequence (Blumenthal and Thomas, 1988).  1.12  The Present Study At the Heat Shock meeting at Cold Spring Harbor in 1985 there had been much  excitement upon presentation of data on the isolation of a yeast ubiquitin gene. The presence of a sequence with homology to the heat shock consensus element in the 5' flanking region of this gene, in conjunction with the experiments on the role of the proteolysis system in the bacterial heat shock response, led to speculation that ubiquitin might be the key controlling element in the eukaryotic heat shock response. Upon returning from the meeting, I set out to determine the role of ubiquitin in the nematode heat shock response. What transpired was a study in polyubiquitin gene expression, trans splicing, and the polymerase chain reaction.  H.  2.1  EXPERIMENTAL PROCEDURES  Growth and Maintenance of Nematodes Nematodes were maintained on N G M (nutrient growth medium : 0.3% NaCl,  0.25% tryptone, 5 Hg/ml cholesterol, 1 m M CaCl ,1 mM M g S 0 , and 25 m M K H P 0 2  4  2  p H 6.0) plates which were covered with a feeder lawn of E. coli strain OP50 as described by Brenner (1974). C. elegans Bristol (N2) strain and C. briggsae nematodes were grown at room temperature (22°C), while C. elegans Bergerac (BO) strain was grown at 17°C. When large quantities of nematodes were required for biochemical analysis sterile liquid culture was employed (adapted from Sulston and Brenner, 1974). Nematode embryos (1 to 1.5 grams/liter) were inoculated into Basal S medium (0.1 M NaCl, 50 m M K H P 0 p H 6.0) supplemented with 0.01 mg/ml cholesterol, 2 m M 2  4  potassium citrate, p H 6.0, 0.3 m M C a C l , 0.3 mM MgS0 ,1.3 u M FeSO^ 2.5 u M 2  4  EDTA, 0.5 u M Z n S 0 , 0.5 uM M n C l , and 0.05 uM CuS0 ) to which was added 4  2  4  approximately 48 grams per liter of E. coli strain B (obtained frozen from Grain Processing Corporation, Muscatine, Iowa). Heavy aeration was essential for normal nematode growth.  Frothing due to lysed bacteria and larval cuticle accumulation  was suppressed by addition of antifoam A emulsion (Sigma) as required. Cultures were maintained for only one nematode generation (approximately 3 to 4 days at 22°C) at which time gravid adult hermaphrodites were collected. Each new culture was inoculated with freshly prepared embryos in order to reestablish age synchrony, avoid crowding, and limit potential contamination. Embryos in liquid culture were either (a) fed immediately or (b) allowed to hatch and arrest in L l prior to feeding. The latter technique provided a greater degree of synchrony however overall yield was decreased due to death of a variable proportion of embryos during starvation.  4  36  Liquid culture also facilitated collection of synchronized populations of individual life cycle stages and dauerlarva stage nematodes. Dauerlarvae were collected by starvation of cultures in L2 stage; the presence of dauerlarvae was confirmed by their altered morphology and activity level and by testing for survival longer than 30 minutes in 0.1% SDS (Cassada and Russell, 1975).  2.1.1  Collection of Gravid Adult Nematodes  Gravid adults were collected from liquid cultures by centrifugation (400 x g for 5 minutes at 4°C). Large cultures (more than one liter) were bubbled on ice for 30 minutes then placed at 4°C until the nematodes had settled, then the supernatant was aspirated. The latter technique removed much of the residual bacteria (which remained in suspension) and reduced the volume of culture to be processed. Nematodes were purified away from pelleted bacteria by flotation in 30% sucrose : up to 15 grams of nematodes could be purified by centrifugation at 2000 x g for 30 seconds at room temperature (Sulston and Brenner, 1974).  2.1.2  Preparation of Nematode Embryos  Embryos were prepared by hypochlorite treatment (Emmons et al., 1979). Five grams or less of floated gravid adults were rinsed in Basal S medium and pelleted in 50 ml polycarbonate tubes (2000 x g, 10 seconds, room temperature). The nematode pellet was resuspended in 40 ml of freshly prepared alkaline hypochlorite solution (6% sodium hypochlorite (BDH), 0.5 M NaOH) and shaken for a total of 8 minutes at room temperature with one change of the bleaching solution. The exact treatment time was determined for each batch of bleaching solution by monitoring embryo release by phase contrast microscopy.  Following bleaching embryos were  immediately pelleted away from the bleach (2000 x g, 10 seconds, room temperature)  37  and rinsed twice with 40 ml of cold 0.14 M NaCl. Viable embryos were separated from a pellet of dead embryos and adult debris by flotation on sucrose as above. Embryo viability was checked by spreading approximately 100 embryos onto N G M plates and counting the number of unhatched embryos after overnight incubation. Alternatively, and more routinely, viability was checked by exclusion of the vital dye methylene blue - darkly staining embryos generally did not hatch. Hatching frequency after hypochlorite treatment was normally greater than 95%. Typical yields from a one liter culture inoculated with one gram of embryos were 2530 grams of gravid adults and 6 to 8 grams of viable embryos.  2.1.3  Viable Freezing of Nematodes  Nematodes can be viably stored in suspended animation at -70°C. Briefly, nematodes (preferably a mixed culture and unfortunately not embryos) are purified by flotation in 30% sucrose (2000 x g, 30 seconds, room temperature), washed twice with 40 ml sterile Basal S medium, then mixed (at more than one million nematodes/ml) with an equal volume of sterile Basal S containing 30% glycerol and frozen in 1 ml aliquots at - 7 0 ° C Once frozen, vials were transferred to liquid nitrogen for long-term storage. To revive frozen nematodes, 1 ml aliquots were rapidly thawed in a 37°C incubator. When just thawed, nematodes are diluted 40fold with Basal S medium, rapidly centrifuged (2000 x g, 10 seconds, room temperature), rinsed again with 40 ml sterile Basal S medium, then resuspended in 20-100 ul of Basal S medium. Droplets of nematodes were spotted on NGM-OP50 plates and nematodes were allowed to recover at room temperature. Typically more than 80% of nematodes recovered.  The most important steps for maximum  viability appeared to be slow freezing in styrofoam blocks, rapid thawing, and rapid, complete removal of glycerol-containing freezing medium on revival.  38  2.2  Nematode Heat Shock Experiments Heat shock experiments were performed by bubbling nematodes in 20 ml of  Basal S medium without food or additional salts. Nematodes were always purified away from bacterial food by sucrose flotation (as described previously) in order to facilitate collection of samples during a heat shock experiment and to avoid any contamination of nucleic acid preparations from bacterial sources.  Up to 5 grams of  nematodes were suspended in 30 ml of room temperature ('slow-rise') or pre-heated ('rapid rise') Basal S medium and left at 33°C for at least 15 minutes to induce the heat shock response. Temperature readings were determined by a thermometer in the sample rather than in the water bath. Aliquots of nematodes (equivalent to at least 0.2 grams packed nematodes) were harvested by immediate centrifugation (2000 x g, 10 seconds, room temperature), resuspended in approximately two packed nematode volumes of buffer A (250 m M sucrose; 10 m M Tris-HCl, p H 8.0; 10 m M M g C l ; 1 m M EGTA) or buffer A* (300 m M sucrose; 60 m M KC1; 15 m M NaCl; 0.15 2  m M spermine-HCl; 0.5 m M spermidine-HCl; 14 m M P-mercaptoethanol; 0.5 m M EGTA; 2 m M EDTA; 15 mM HEPES, p H 7.5 : Schibler et al., 1983), then dripped from a 1 ml pipet into liquid nitrogen. Frozen pellets were then stored at -70°C.  2.3  Preparation and Analysis of Nematode Genomic D N A High molecular weight genomic D N A was prepared from nematodes by the  method of Emmons et al. (1979) as follows. Frozen nematodes were allowed to thaw in proteinase K buffer (100 mM Tris-HCl p H 8.5, 50 m M EDTA, 200 m M NaCl, 1% SDS) at room temperature, then proteinase K (Boehringer Mannheim) was added to a final concentration of 0.05 mg/ml and the nematodes were placed at 55 to 65°C for one hour. The sample was then gently extracted with TE-saturated phenol three times in a separatory funnel, then with chloroform three times in polypropylene tubes. Separation of the organic and aqueous phases for the  39  chloroform extractions was by centrifugation (2000 x g for 3 minutes).  The aqueous  D N A supernatant was then chilled to 4°C and covered with two volumes of ice cold 95% ethanol. The D N A was gently wound out of solution by rotating the tube at an angle to disturb the interface. Ethanol was changed up to four times during winding. In this way much of the co-purifying RNA in the aqueous phase was lost to the ethanol washes. The wound D N A usually resuspended immediately upon addition of TE. Further purification of the nematode genomic D N A by cesium chloride gradient centrifugation was found to be unnecessary for almost every application. Typical yields from this procedure were 3 mg D N A per gram of frozen nematodes. For C. elegans, 1 \ig of genomic D N A represents approximately 10 million genome equivalents.  2.4  Preparation and Analysis of Nematode RNA RNA was isolated by the method of Ullrich et al. (1977) as described by  Antonucci (1985). This method removes contaminating D N A , allowing direct quantification of the RNA by spectrophotometry and ethidium fluorescence, and hence permitting direct comparisons between samples. Frozen nematodes (up to 5 grams) were powdered in a chilled mortar and pestle. As the powder thawed 3 ml of homogenization buffer (7.5 M guanidinium hydrochloride; 25 m M sodium citrate, p H 7.0; 0.1 M P-mercaptoethanol) was added. The homogenate was then passed five times through a 21 gauge needle to reduce sample viscosity and shear genomic DNA. The homogenate was gently layered over a 1 ml cushion of cesium chloride solution (5.7 M CsCl, 25 m M sodium citrate, p H 5.0 which had been treated with 0.07% diethylpyrocarbonate (DEPC) for 15 minutes and autoclaved) and was centrifuged (220,000 x g) for 16 hours at 2 2 ° C The RNA pellet was suspended in 300 ml of sterile, DEPC-treated water. The aqueous RNA was extracted once with phenol (saturated with DEPC-treated d H 0 ) then precipitated with 0.5 volumes of 2  40  7.5 M ammonium acetate and 2.5 volumes of 95% ethanol. Treatment of samples at 50°C for 10 minutes had little effect on RNA degradation and greatly facilitated resuspension of R N A pellets in DEPC-treated d H 0 . The quality of R N A was 2  routinely checked by agarose gel electrophoresis and by Northern hybridization analysis. Typical yields were 2 mg of total cellular RNA per gram of starting material. P o l y A R N A was prepared from total nematode R N A by batch adsorption to +  oligo-dT cellulose (Collaborative Research Inc.). Yields were higher than those obtained using an oligo-dT cellulose column because R N A could be eluted at a higher temperature. Typically, 80 ng polyA RNA was recovered per mg of total +  cellular R N A .  2.5  Screening of Recombinant D N A Libraries  2.5.1  Screening of Bacteriophage A, Libraries  Two C. elegans Bristol (N2) genomic D N A libraries were found-to contain recombinant ubiquitin clones. One was a partial EcoRI digest in the X Charon4 vector and was kindly provided by T. Snutch of Simon Fraser University. This phage was propagated in a E. coli LE392 host strain (see Appendix B for genotype) which was grown in N Z Y C M medium (1.0% NZ-amine; 0.5% yeast extract; 0.2% casamino acids; 10 m M M g C l ; 0.2% maltose; p H 7.0). The other library was a partial 2  EcoRI digest cloned into the X g t l l vector and was provided by R. Barsted of Washington University. This library was propagated in E. coli strain Y1088 (see Appendix B for genotype) in YT medium (0.8% tryptone, 0.5% yeast extract, 0.5% NaCl p H 7.0).  41  2.5.2  Isolation of Bacteriophage D N A  D N A was routinely prepared from 20 ml liquid cultures. 200 ul of an overnight culture of bacteria was pre-incubated with 200 ul of lambda dilution buffer containing 10 to 10 phage for 15 minutes at 37°C then was added to 20 ml of the 5  6  appropriate growth medium and shaken vigorously. Lysis usually occurred within 5 to 8 hours, at which time 3 ml of chloroform was added to the culture, and shaking was continued for an additional 10 minutes. The culture could then be stored at 4°C overnight. The phage supernatant mixture was decanted away from the chloroform then centrifuged twice (12,000 x g for 10 minutes) to remove all bacterial debris. To the cleared supernatant was added 3 ml of 5 M NaCl (final: 1 M) and 3 grams of polyethylene glycol powder (average M W 20,000) (final: 10%). The contents were thoroughly mixed and left at 4°C for two or more hours. The precipitated phage particles were collected by centrifugation (12,000 x g for 10 minutes at 4°C) then resuspended in 500 ul of DNase I buffer (50 m M HEPES p H 7.5, 5 m M M g C l , and 0.5 m M CaCl ). RNase A (100 ug) and DNase I (5 ug) were added 2  2  and the phage were incubated at 37°C for 60 minutes. In some cases the PEG precipitation and nuclease steps were performed twice to ensure removal of all bacterial remains. Following nuclease digestion, .50 ul of 10 x SET (IX = 10 m M TrisHCl p H 7.5, 20 m M EDTA, 0.5% SDS) and 150 ug of proteinase K were added, then the sample was incubated for 60 minutes at 65°C. This mixture was extracted once with an equal volume of phenol/chloroform and once with the same volume of chloroform. The phases were separated by centrifugation at 12,000 g for 3 minutes at room temperature. The bacteriophage D N A was precipitated from the aqueous phase with two volumes of 95% ethanol at room temperature for two minutes, and collected by centrifugation (12,000 x g for 4 minutes). The pellet was quickly resuspended in 50 ul of TE buffer containing 40 ug/ml RNase A . Typically 3 to 5 ug of X phage D N A was obtained with very little contamination by E. coli D N A .  2.5.3  Screening of Cosmid and Y A C Panels  X Charon4 bacteriophage D N A from a ubiquitin-encoding clone was sent to the MRC laboratory in Cambridge, England where it was compared to an available cosmid clone bank. In this way several cosmids which extended the available genomic D N A in the UbiA region were identified. These cosmids were obtained in E. coli hosts, from which cosmid D N A was prepared. Nylon filter replicas of YAC-containing yeast colonies (i.e. from yeast transformed with Yeast Artificial Chromosomes) were obtained from R. Waterston (Washington University). These filters were hybridized with a C. elegans ubiquitin clone by conventional means and a single positive colony reported to the MRC laboratory for alignment with the Y A C and cosmid bank.  2.6  Transformations Bacterial cells were made competent for transformation using the methods of  Hanahan (1983). Competent bacteria were prepared in bulk and frozen in single transformation aliquots at -70°C. Typically, competent bacteria had a transforming efficiency of 10 transformants per microgram of input plasmid. 7  2.7  Purification of plasmid D N A For the large scale isolation of plasmid or cosmid D N A 200 ml cultures were  incubated in TB medium (1.2% tryptone, 2.4% yeast extract, 0.4% glycerol, 17 m M K H P 0 , 72 m M K H P 0 p H 7.35). For the large scale isolation of M13 replicative 2  4  2  4  form (RF) DNA, 5 ml of an overnight E. coli JM109 culture and 150 ul of M13 infectious phage (approximately 10 phage particles) supernatant were added to 500 9  ml of YT medium and incubation was for 5 to 7 hours at 37°C. Typical yields were 1.5 mg D N A per 500 ml bacterial culture for plasmid preparations and 800 ug D N A per 500 ml bacterial culture for M13 RF preparations  43  when YT medium was employed. Frequently plasmid D N A was precipitated directly without purification by cesium chloride gradient centrifugation. The latter procedure increased plasmid D N A yields as much as ten-fold. Rapid small scale alkaline lysis plasmid and M13 RF isolations were carried out on 1.5 ml of overnight cultures as described in Maniatis et al. (1982) and typically yielded 10 ug of plasmid D N A or 2 ug of M l 3 RF DNA.  2.8  Rapid Screening of Single Bacterial Colonies by PCR Screening of recombinant clones following transformation of competent R coli  with plasmids involves the growth of liquid cultures from single colonies followed by the isolation of plasmid D N A , restriction analysis, and preparation of sequence quality template.  Analysing multiple colonies or multiple transformations  simultaneously is laborious and time consuming. I have developed a simple and rapid method based on the polymerase chain reaction (PCR: Saiki et al., 1988) which allowed the analysis of 36 or more recombinant colonies in less than 5 hours. Single bacterial colonies were dispersed in 15-20 ul of 1 x PCR buffer (0.5 m M MgCl ,10 m M 2  Tris-HCl p H 8.4, 0.05% Tween 20, 0.05% Nonidet P-40) (Innis et al., 1988). Bacterial suspensions were boiled for 90 seconds, then debris was pelleted by centrifugation (12,000 x g, 20 seconds, room temperature). A n aliquot of the supernatant was mixed with 20 pmol each of forward and reverse M13 oligonucleotide primers RG05 and RG02 (for oligonucleotide sequences see Appendix A) and was subjected to PCR amplification in 1 X PCR buffer containing 50 uM dNTP's and 1 unit T. aquaticus (Taq) D N A polymerase (Promega or Cetus). Amplifications were performed in a Perkin Elmer-Cetus D N A Thermal Cycler®. Following 30 cycles of PCR (94°C for 10 sec, 64°C for 40 sec, 72°C for up to 3.5 minutes depending on size of insert), 10% of each reaction was analyzed by agarose gel electrophoresis. Using this procedure (see Figure 5), I have successfully amplified fragments from 0.5 kb up to 4 kb from M13,  44  4.0 kb 2.4 kb  0.9 kb 0.7 kb 0.5 kb  Figure 5. Single bacterial colony polymerase chain reaction. Single bacterial colonies were boiled in polymerase chain reaction (PCR) buffer and subjected to 35 cycles of PCR as described in Experimental Procedures. Following PCR, 10% of each sample was analyzed on a 0.7% agarose gel containing 1.5 u g / m l ethidium bromide. PCR samples were loaded above as follows : M : A. D N A digested with Avail, 1-5 : 0.5 kb, 0.9 kb, 0.7 kb, 2.4 kb, and 4 kb fragments of amplified cloned C. elegans D N A . D N A fragments were cloned in the plasmid vectors M13mpl8, pUC19, pGEM4Z, pUC13, and pBluescript KSH (+), respectively, and were transformed into E. coli XL-1 Blue.  45  pUC, pGEM, and pBluescript vectors. In summary, this method allows the production of vector-free insert for routine insert size determination and restriction mapping.  2.9  D N A Sequencing  2.9.1  Single Stranded D N A Sequencing  Restriction fragments were subcloned into the sequencing vectors M13mpl8 and M13mpl9 (Messing, 1983) and were sequenced by the dideoxy chain termination method of Sanger et al. (1977). A detailed description of the methods used, including the subdoning of D N A fragments into M l 3 vectors and the preparation of single stranded phage D N A and dideoxy nudeotide reactions using the Klenow fragment of E. coli D N A polymerase is given by Messing (1983). In most cases M13 clones were first screened using only ddTTP reactions in order to avoid sequencing redundant clones.  2.9.2  Double Stranded D N A Sequendng  All double stranded D N A sequencing was performed on plasmids (e.g. pBluescript KS II (Stratagene)) isolated from bacterial hosts deficient in recombination (recAl) and nuclease (endA) activities (e.g. XL-1 Blue - see Appendix B for strains and genotypes). Plasmid subclones were pre-screened for the presence and size of insert by the PCR-based single colony screening procedure above. D N A sequendng was carried out using a modification of the procedure of Gaterman et al. (1988). Barterial pellets from overnight cultures were resuspended in 150 ul of lysis buffer (8% sucrose, 5% Triton X-100, 50 mM EDTA, 0.5 mg/ml lysozyme) and were boiled for two minutes. After a two minute centrifugation (12,000 x g at 4°C), pelleted debris was removed with a toothpick. D N A was predpitated from the supernatant by addition of 150 ul of 2-propanol. .Precipitated  46  D N A was resuspended in 35 ul of d H 0 and stored at - 2 0 ° C Plasmid D N A 2  (approximately 2 \ig) was mixed with 1 pmol of the appropriate oligonucleotide sequencing primer and freshly prepared N a O H was added to a final concentration of 0.2 N NaOH. This mixture was boiled for 2 minutes, then chilled and neutralized with the addition of ammonium acetate (pH 5.2) to a final concentration of 0.5 M . Denatured plasmid D N A was precipitated with 99% ethanol, then was redissolved at room temperature in 10 ul of Sequenase buffer (40 m M Tris-HCl, 20 m M M g C l , 2  50 m M NaCl) containing an additional 1 pmol of oligonucleotide sequencing primer. Following a 15 minute annealing period at room temperature, DTT, [ct- P] 32  dATP, and labelling nucleotide mix (dGTP, dCTP, and dATP) were added to 7 mM, 10 uCi, and 0.1 m M each nucleotide, respectively. Three units of T7 D N A polymerase (Sequenase , US Biochemical) were added and the labelling reaction 11  was allowed to proceed for 3 minutes at room temperature. Following the labelling step, aliquots were placed into wells of a microtiter plate containing G,A,T, or C termination mixes (each termination mix contained 80 u M dGTP, dATP, d l I P , dCTP and 50 m M NaCl - in addition, the *G' mix contained 8 u M ddGTP, the 'A' mix contained 8 u M ddATP, the T mix contained 8 u M ddTTP, and the ' C mix contained 8 uM ddCTP) pre-warmed to 49°C. Primer extension was then allowed to occur for 5 minutes at 49°C, then formamide dye mix (which contained 0.05% bromophenol blue and xylene cyanol dyes) was added to a final concentration of 20% (v/v). Samples were boiled for 3-5 minutes, then loaded onto 4%, 6%, or 8% polyacrylamide slab or wedge gels.  2.10  Preparation of Radioactive D N A probes  2.10.1 Nick Translation Preparation of radiolabelled D N A probes by nick translation was by standard methods (Maniatis et al., 1982).  47  2.10.2  Primer Extension M13 probes  The M13 templates (200 ng) were heated to 65°C for 10 minutes in Klenow Buffer (10 m M Tris-HCl p H 7.5, 7 m M M g C l , 7 m M fJ-mercaptoethanol) containing 2  0.8 pmol of universal forward M13 primer (P-L Biochemicals).  After cooling to  room temperature for 15 minutes, 2 ul each of 400 uM dGTP, dCTP, and dTTP were added along with 15 uCi of [ct- P] dATP (Amersham, 300 Ci/mmol) in a total 32  volume of 16 pi. Templates were extended at room temperature or 49°C for 5 minutes using 2 units of the Klenow fragment of E. coli D N A polymerase I (Pharmacia). The higher temperature of incubation helped to prevent termination of the extension at secondary structures. A chase reaction was initiated by the addition of dATP to 40 uM. Reactions were terminated by heating at 70°C for 10 minutes.  The primer extended D N A products were digested for 20 minutes with  appropriate buffer and restriction enzymes.  The reaction was terminated by the  addition of E D T A to a final concentration of 10 mM. The labelled D N A product was separated from unincorporated nucleotides by spin dialysis. The sample was centrifuged through a 1 ml mini-column packed with Sephadex G-50 superfine (Pharmacia). D N A thus prepared usually had a specific activity of 10 cpm/ug input 8  template. Higher activity probes could be obtained by using multiple labelled nucleotides during the pulse reaction. Low activity pulse-end-labelled probes could be prepared for SI nuclease reactions by labelling with 1 pi of 40 uM dATP present and extending for 30 seconds before initiating the chase reaction.  2.11  Nucleic Acid Hybridizations Southern hybridization analysis and Northern hybridization analysis were  performed by standard methods (Maniatis et al., 1982). All hybridizations were carried out in Seal-a-Meal® bags or covered glass Petri dishes using standard conditions (Maniatis et al.,1982). Pre-hybridization of filters was found to be  48  unnecessary and BSA was purposely omitted from any steps. Background nonspecific hybridization was eliminated by the inclusion of 5 |ig/ml heparin (sodium salt, Sigma) in hybridization cocktails.  2.12  Dot Hybridization Analysis For dot hybridization analyses, samples of RNA were heated to 55°C for 15  minutes in 200'ul of 1 M ammonium acetate p H 7.0 (DEPC-treated) then applied to a nitrocellulose membrane using a BIO-RAD manifold. The membranes were washed once with 0.5 ml of 1 M ammonium acetate p H 7.0, then were baked in vacuo at 80°C for 90 minutes. M l 3 single stranded D N A standards were boiled for 5 minutes in 1 M ammonium acetate p H 7.0 immediately prior to loading. Following hybridization and fluorography, signal intensities were quantified by densitometry.  2.13  SI Nuclease Analysis SI nuclease protection experiments were performed essentially as described by  Berk and Sharp (1977). Labelled single stranded D N A probes were prepared by primer extension of M13 templates in the presence of [ct- P] dATP (described above) 32  followed by gel purification on denaturing acrylamide preparative gels. Labelled fragments were cut from the gel and electroeluted in 0.5 X TBE buffer containing 0.05% SDS. Hybridization of nematode R N A and SI probe D N A was as follows. Approximately 200,000 cpm of labelled probe were mixed with 1 to 30 ug of total nematode R N A in 40 ul of hybridization buffer (50% formamide, 0.4 M NaCl, 40 m M PIPES p H 6.4,1 m M EDTA, 0.6 mg/ml E. coli tRNA). This solution was heated to 85°C for 10 minutes, then immediately transferred to 42°C and incubated overnight. Hybridization was terminated by the addition of 200 ul of ice-cold SI nuclease mixture [0.28 M NaCl, 50 m M sodium acetate p H 4.6, 4.5 m M Z n S 0 , and 4  60 to 100 units SI nuclease (Pharmacia)]. SI nuclease digestions progressed for one hour at 37°C and were terminated by the addition of 40 ul of 4 M ammonium acetate - 0.1 M EDTA solution. SI nuclease protected D N A fragments were precipitated with one volume of 2-propanol using 15 ug of E. coli tRNA as carrier. Samples were dissolved directly in formamide (which contained 0.05% xylene cyanol and bromophenol blue dyes) and were analyzed on thin 4% or 6% acrylamide gels containing 8 M urea.  2.14  Primer Extension RNA Sequencing Primer extension sequencing was by K. Van Doren (Synergen) using the  following protocol. A n oligonucleotide (S'-AGATTTGCATGATTG-S') was labelled with [y - P] dATP at its 5' end using polynucleotide kinase and was used as a primer 32  for sequencing the 5' end of the UbiA mRNA. One ug of oligonucleotide (approximately 200 pmol) was labelled using 7 units of T4 polynucleotide kinase in 20 ul of kinase buffer (50 m M Tris-HCl p H 7.4,10 m M M g C l , 5 m M DTT) containing 2  100 uCi of [y - P] dATP. The labelling reaction was incubated at 37°C for 30 minutes, 32  then was terminated by heating to 65°C for 5 minutes. Ten pmol of labelled primer was added to 56 ug of nematode total cellular RNA in 12 ul of annealing buffer (250 m M KC1,10 m M Tris-HCl p H 8.3), the mixture was heated at 80°C for 3 minutes, and then it was slowly cooled at 35°C for 45 minutes. Sequencing reactions contained 2 \il of annealed template/primer mixture and 1 ul of the appropriate 1 mM ddNTP stock (or no ddNTP) plus 3.3 ul reverse transcriptase buffer (24 m M Tris-HCl p H 8.3,16 m M M g C l , 8 m M DTT, 0.4 m M dATP, 0.4 m M dCTP, 0.8 m M 2  dGTP, 0.4 m M dTTP containing 3 units of Moloney murine leukemia virus (MMLV) reverse transcriptase). Reactions were incubated at 50°C for 45 minutes then stopped by addition of 2 ul of formamide dye mix. Samples were boiled 3 minutes and analyzed on a 10% acrylamide - 8M urea gel.  50  2.15 PCR 2.15.1  Preparation of Total R N A for PCR  Nematode total cellular RNA was prepared as described above. Plant R N A was prepared similarly, except that the guanidinium slurry of ground leaves was mixed well then pre-centrifuged (20,000 x g for 10 minutes) to pellet undissolved debris.  2.15.2  First Strand cDNA Synthesis  cDNA synthesis was performed by a combination of the procedures described in Maniatis et al. (1982) and Frohman et al. (1988). For cDNA synthesis, 1 to 10 ug of total RNA in 7.5 ul DEPC-treated d H 0 was mixed with 7.5 ul of 40 m M 2  methylmercuric hydroxide (freshly diluted from a 1 M stock) and incubated at room temperature for 15 minutes to allow denaturation. This mixture was then frozen in a dry ice/ethanol bath. A freshly prepared 30 ul cocktail consisting of 33 u M DTT, 1.66 X M M L V reverse transcriptase buffer (IX = 50 mM Tris-HCl p H 8.3, 75 m M KC1, 3 m M MgCl ,10 m M DTT), 1 unit/ul RNasin (Promega), 833 u M dNTP mix (i.e. 2  dATP, dGTP, dTTP, and dCTP), and 166 pmol/ul anchor oligo-dT oligonucleotide primer (for oligonucleotide sequence see Appendix A) was centrifuged into contact with the frozen pellet of denatured RNA. Immediately upon thawing of the pellet, 4.5 ul (900 units) M M L V reverse transcriptase (BRL) was added. The reaction was allowed to proceed at 37°C for 45 minutes, then was terminated by freezing the mixture in a dry ice/ethanol bath. It should be noted that chemical denaturation with methylmercuric hydroxide proved more efficacious than heat denaturation and that M M L V reverse transcriptase is required when using total R N A as substrate.  2.15.3  RACE cDNA Amplification  Rapid cDNA amplification was performed by a modification of the procedure of Frohman et al. (1988). A n aliquot (typically 2%) of the first strand cDNA reaction  51 containing RNA:cDNA hybrids was mixed with 50 pmol of the appropriate degenerate ubiquitin oligonucleotide (ubl or ub2; for oligonucleotide sequence see Appendix) and 50 pmol of the anchor oligo-dT in a volume of 5 ul. To this was added 45 pi of PCR cocktail consisting of 50 p M dNTPs (i.e. 50 pM each dNTP), 1 x PCR buffer (10 m M Tris-HCl p H 8.4, 0.05% Tween 20, 0.05% Nonidet P-40, 0.5 m M M g C l ), and 1 unit of T. aquaticus (Taq) D N A polymerase (Promega or Cetus Corp.). 2  This mixture was overlaid with 60 pi of light mineral oil (LifeBrand®), then inserted into the pre-heated (94°C) thermocycler block. A typical amplification program consisted of 35 cycles of : 94°C denaturation for 10 seconds, 53°C annealing for 30 seconds, then 72°C primer extension for 30 to 60 seconds. Annealing temperatures vary for different degenerate primers, but for the two ubiquitin primers (ubl and ub2) adequate results were obtained at the above temperature. A n additional 5 minute incubation at 72°C followed the final cycle.  2.15.4  Analysis of PCR Products  Routinely, 10% of a PCR sample was separated by electrophoresis on a TBEagarose gel to analyze the amplified D N A products; alternatively the amplified D N A was separated by electrophoresis on a TAE-low melting point (LMP) agarose gel. Individual D N A fragments could be readily re-amplified from TAE-LMP gels by extracting a plug of agarose from the band of interest, boiling this plug in 500 pi of d H 0 for 90 seconds, then re-amplifying using 2 pi of this D N A fragment solution 2  (Kelly Thomas, personal communication).  2.15.5  Cloning PCR Products  A protocol is included here because it was found that the products of PCR reactions do not appear to clone as efficiently as predicted for their mass abundance. One reason for this is that most PCR products, although in theory blunt ended, are  not, due to the tendency of T. aquaticus (Taq) D N A polymerase to add one or more extra dATP residues onto available 3' ends (R. Saiki, personal communication). This problem may be circumvented by the incorporation of restriction sites on the 5' end of PCR oligonucleotides; however, one must be certain that comparable internal restriction sites are not present in the D N A fragment to be amplified. To facilitate blunt end cloning of PCR amplified D N A fragments, amplified D N A was extracted with phenol:chloroform(l:l) then precipitated with 99% ethanol. The amplified D N A fragments were then redissolved in 18 ul d H 0 and 2  placed at 37°C. Klenow mix (2 ul of 20 m M Tris-HCl p H 8.0,100 m M M g C l , 0.03 2  units Klenow fragment of E. coli D N A polymerase/ul; Henikoff, 1984) was added and the D N A was incubated at 37°C to remove terminal overhanging single stranded DNA. After two minutes 2 ul of dNTP mix (125 u M each dNTP) was added and incubation was continued a further 3 to 4 minutes. The reaction was stopped by the addition of 12 5 ul of 7.5 M ammonium acetate, 30 ul of phenokchloroform, and 100 ng of appropriately digested vector D N A (e.g. EcoRV - digested pBluescript KS n plasmid for blunt end cloning). After phenolrchloroform extraction, the D N A was precipitated from the aqueous phase with two volumes of 99% ethanol. D N A was collected by centrifugation (12,000 x g for 15 minutes at 4°C), then and dissolved in an appropriate volume of d H 0 for ligation and subsequent transformation. 2  2.15.6  Single Nematode PCR  For PCR analysis of nematodes without pre-preparation of purified D N A , 1 to 10 nematodes were picked into 5 ul of low SDS solution (100 m M KC1, 20 m M Tris-HCl 8.4, 40 m M DTT, 0.0001 % SDS) (Li et al., 1988). The nematodes were frozen in this solution, then heated at 95°C for 5 minutes. Immediately, the heated nematodes were vortexed, briefly spun to the bottom of the tube, then frozen again. This freeze-thaw procedure was then repeated.  PCR cocktail (20 pmol of each oligonucleotide primer, 70 \iM dNTP (prepared in equimolar MgCl ), 1 X PCR buffer (see above), 1 unit of T. aquaticus (Taq) D N A 2  polymerase) was added to the thawed nematode mixture, which was then overlaid with 60 pi of mineral oil. D N A was amplified with the following temperature regime: cycle 1: 94°C x 3 minutes; cycles 2 to 35 : 94°C x 15 sec, 62°C x 1 minute, 72°C x 1 minute; cycle 51 = 72°C x 5 minutes. As with all PCR experiments, appropriate controls were done in parallel with samples. Suggested controls when initially assessing transgenic nematodes are : (a) no added substrate D N A (b) the microinjection plasmid (100 pg)  (c) genomic D N A of transformed and wild type  nematodes (approximately 100 ng)  (d) replicates of 1, 5, and 10 wild type and  transformed nematodes and (e) an extra sample for microscopy.  2.15.7  PCR Assay of Transgenic Nematode RNA  Total R N A and first strand cDNA were prepared from transgenic nematodes using the previously described protocol for amplification of ubiquitin genes. For detection of reporter gene transcription, PCR (40 cycles) was carried out as described except that P-gal01 primer oligonucleotide (for oligonucleotide sequence see Appendix A) was used in place of the degenerate ubiquitin oligonucleotide with anchor oligo-dT and the annealing temperature was 62°C. For detection of reporter gene trans splicing, first strand synthesis could be primed with oligonucleotide RG05 rather than anchor oligo-dT. PCR amplification would be performed with RG05 and SL oligonucleotides (for oligonucleotide sequences see Appendix A), annealing at 59°C.  Suggested controls are : (1) No cDNA (2) wild type cDNA (3) genomic D N A  from transformed and wild type nematodes and (4) if available, plasmid D N A which includes PCR oligonucleotide sequences.  54  m.  RESULTS  3.1  Isolation of Genomic Clones Preliminary Southern hybridization analysis confirmed the presence of  sequences homologous to yeast ubiquitin in restriction digests of C. elegans genomic D N A . The yeast polyubiquitin cDNA was provided by Dr. A. Varshavsky, MIT. Several hybridizing restriction fragments were detected and hybridization was found to retain specificity and increase in intensity under conditions of low stringency (5 x SSPE, 30% formamide, 42°C). These conditions were employed in screening X phage libraries containing C. elegans genomic D N A . A screen of a ACharon4 library representing ten genome equivalents revealed a single class of positive phage which hybridized to the yeast ubiquitin probe. This phage was termed XUbl. Subclones of XUbl which hybridized to yeast ubiquitin sequences were subjected to partial D N A sequencing and it was found that the phage represented an incomplete polyubiquitin gene, lacking the 5' sequences. Another clone (AUb2) was obtained in a subsequent screen of a Xgtll phage library. This clone overlapped XUbl by 220 bp and included the rest of the ubiquitin coding region as well as 5' noncoding sequences. One possible reason for the difficulty encountered in attempting to isolate a complete genomic clone of the polyubiquitin gene may be its tandemly repeated structure, which may be subject to recombinational deletion during the construction of phage libraries.  3.2  Localization of the UbiA Gene Employing the XUbl clone, John Sulston and Alan Coulson (MRC Laboratory,  Cambridge, England) were able to isolate a series of overlapping cosmids containing the UbiA gene (named ubq-1 for mapping purposes) as part of their C. elegans genome mapping project (Coulson et al., 1986). This cosmid 'contig* extends  approximately 100 kb downstream (i.e. 3') of the polyubiquitin gene. One of these cosmids, C16A7, has been nick translated and used as a probe for in situ hybridization to C. elegans chromosomes (D. Albertson, personal communication). Using these procedures, the UbiA gene has been mapped to the left half of chromosome i n , between the genetic markers ced-4 and mlc-3. A representation of the physical map of the UbiA gene is presented in Figure 6.  3.3  Analysis of the UbiA Gene Sequence A restriction map of an 8.6 kb region containing the C. elegans polyubiquitin  gene UbiA is presented in Figure 7. The overlapping parent Xphage clones XUbl and XUb2 are indicated. Coding regions were initially assigned by Southern blotting of restriction digests of XUbl and XUb2 D N A and were confirmed by sequencing. Introns and flanking noncoding regions of UbiA have also been indicated. Approximately 4500 bp of D N A within and around the UbiA gene was sequenced; the sequence of the coding region and immediately flanking D N A is presented in Figure 8. The coding region consists of 11 tandemly repeated copies of D N A encoding monomeric ubiquitin sequence with no spacer D N A between repeats; this polycistronic organization is unique to ubiquitin genes within eukaryotes. Assuming that the UbiA mRNA is translated into a polyprotein, its primary translation product would have a mass of approximately 96 kDa; this polyprotein would then be post-translationally cleaved into monomeric ubiquitin with a molecular weight of 8600. Several features of UbiA are worth noting and will be discussed below. Repeat units 1,4, 7, and 10 of UbiA are interrupted by short introns at glycine-47. As well, a 3' consensus splice site is found five nucleotides upstream of the translation initiation site; this site is used for trans splicing. It should also be noted that the  56  Chromosome III ubq-1 unc-93  tra-1  T01D10ZK331C41E11-  •F25B5  C16A7 ubq-1  Figure 6. Orgariization of the UbiA locus of C. elegans. C. elegans chromosome HI is represented as a rectangle made up of 100 units of physical D N A sequence. UbiA (here denoted ubq-1) is located between the genetic markers ced-4 and mlc-3. Two additional genetic markers are shown for scale. The lower portion of this figure represents a detailed cosmid map of the region surrounding UbiA (ubq-1). Lengths of the cosmids are not proportional to D N A size, rather they are proportional to the number of HindlH restriction sites present in each cosmid (as per Coulson et al., 1988).  SX  cc  H  s  •K2SSSS-  Ubi  Ub2  Figure 7. Structure of the UbiA gene of C. elegans. Thick arrows indicate tandemly repeated regions encoding monomeric ubiquitin. Open boxes within arrows represent cis spliced introns which are present in the primary transcript of UbiA. The shaded box represents, the trans spliced leader intron of UbiA. The hatched box represents the final two amino acids and the 3' untranslated region of UbiA. Lines below the gene represent the two overlapping bacteriophage X. clones, Ubil and Ubi2, which contain UbiA. Selected restriction sites are indicated as follows : S : Sail; X : Xbal; C : Clal; H : HindHI; E : EcoRI.  58 -847 -627 -807 -787 -7«7 -7*7 CCCCTTGTTCCTCATCATTTC7TTCCCT*CACAGC*CT(CTAbA*rQTTCTTqrTaTQC*Q***Q*OTOCCQTTTO*OTCAQCOACCCCCCCCCCCCCCCTCC7TTC7CTTGCTCTTCCTA  -727 -707 -687 -««7 -947 -«27 CTgGTTCTCCTttT*CCCC*CTTCTTgC7»Ae*a«AA0TaACC*T«aC*«C«TTTTTT«CTTT0TGGCCTTCA«T*«T*COTOCQTC0TTTH|TTi>G*«TSTTTeAClrAA*aTTe»«CCT -«07 -387 -567 -347 -S27 -507 ATTCACGTTTTGGGCGCTCTTTAATTTATTACTGTCAA8AATCAaTTTACCAAACGGTGAGTTTCTTTTTTTTTT0TCTAATTGTAAGATTTAGCGGGGTAAAACCAAC  GT A G A T T C A A T  -487 -4S7 -447 -427 -407 -387 TGTCATGCTTTTTTGAATAATCTCAATCAQTTGTTATATGAATTATTTTCCCATTTTAaCAATAC7GCTTaaTAQTATTTCG0TCA8AGAAACGAGGACATCAGCTGAACATCTG  AGAA A  -3S7 -347 -327 -307 -287 -267 CGTCTCTAACAACACTCGGGAAGCGAGTCAOTOTaCaCaTaCaTTGGGGTTTTATCCQATCOTTGAGCaQOCATACAOCAOTCATACACCCCATTCGACCAGACTCCGCTCCCGTGCCAC  -247 -327 ' -207 -187 -187 -147 TTCTCATTTCACTTGTCTCTACTC0GACATTACTCCTCATCGATAGCTCTTTACTACCATTTTACTTTTTATGCCTTTCTTTTTC8TTTGACTTGCCTATACGAGTGGGG  CTTGTCTCCA  -127 -107 -87 -«7 -47 ACAAGTTTGCTTTGTTAGTCTTAGCTAOTGTATCGATTTTTTGGaTAATAT77CaCAACTTTCTAaGACTTTCTTTCATAATCACCTCTTCTCTCGCCTCCTCATTCCA6TTTTATTCGC  -27  -7 • 13 33 33 73 93 M 0 I F V K T L T G K T I T L E V C A S O T l C N V K A K I 0 AC7CATTTTCTATTTTTTCAaCAATCATOCAAATCTTCQTCAAAACaT7QACTO0AAAAACTATCACCCTaOAQQTaaAaaCTTCCaATACCATCOAAAATaTCAAAGCCAAGATCCAAG  A 113  133  1S3  173  183  213  O K E a t P P O O O R L I F A ACAAGGAAGGAATTCCACCAGATCAGCAGAGACTTATTTTTGCTGGTACGTTGGCAAAATATCTAATATTTGACCTAAAATTTATTATATATTTTCAaaAAAGCAACTCGAOGArGGCCO 233  T  L  S  O  Y  N  253  r  O  K  E  S  T  L  273  H  L  V  L  B  L  R  T  S  a  M  293  O  I  F  V  313  K  T  L  T  S  K  T  333  I  T  L  E  V  e  TACCCTTTCGGATTACAATATCCAGAAGGAATCAACCCTCCATTTGGTCCTCCGCCTAAGAGGAGGAATGCAGATCTTCGTCAAGACTTTGACCGGAAAGACTATTACACTTGAGGTTGA  333  373  393  413  433  453  A S O T I E N V K A K I O O K E G I P P O O O P L I F A G K O L E O a s T L S O AGCTTCTGACACTATCGAGAATGTGAAGGCCAAGATCCAAGACAAGGAAGGTATCCCTCCGGATCAACAGCGTTTGATCTTTGCCGGAAAGCAACTCGAGGATGGCCGTACTCTCTCCGA  A73 Y  N  I  O  K  E  S  493 T  L  H  L  V  L  R  313 L  R  G  G ' M  533. O  I  F  V  K  T  L  533 T  G  K  T  I  T  L  573 E  V  E  A  S  O  T  TTACAACATCCAAAAGGAGTCTACTCTTCATCTGGTTCTGCGTCTCCGAGGAGGAATGCAAATCTTCGTCAAGACTCTTACTGaAAAGACCATCACCCTCGAAGTCGAAGCCTCCGATAC 593 G13 633 653 673 693 l E N V K A K I O O K E G I P P O O O R L I F A G K O l E O G R T L S O Y N I O CATCGAGAACGTaAAGGCCAAGATTCAGGACAAGGAAGGAATTCCACCAGATCAGCAGCGTCTCATCTTCGCCGGAAAGCAGCTCGAGGACGGCCGCACCCTTTCTGACTACAACATCCA 713 733 y 733 773 793 813 K E S T L H L V L R L R G G M O I F V K T L T G K T I T L E V E A S O T I E N V GAAGGAATCTACTCTTCACTTGGTTCTTCGTTTGAGAGGAGGAATGCAGATCTTTGTCAAGACTTTGACTGGAAAGACCATCACACTTGAAG7TGAAGCTTCCGACACGATCGAGAACGT 833 833 873 893 913 933 K A K I O O K E G I P P O O O R L I F A G K O L CAAGGCCAAGATTCAAGACAAGGAGGaAATCCCGCCAGATCAGCAGCGTCTTATCTTTGCTGGTATaTTACATATAACAAATTTTGTTCATGAGAGACTAATTTTTTCAGGAAAGCAATT 933 973 993 1013 1033 1053 E O a R T l S D Y N I O K E S T L H l V L R L R a G ' H O I F V K T L T G K T t T GOAAGATGGACGCACACTCTCTGATTACAATATTCAGAAAGAGTCTACTCTCCACTTGGTGCTCCGTCTCAGAGGAGGTATGCAGATCTTCGTCAAGACATTGACTGGAAAGACCATCAC 1073 1093 1113 1133 1153 1173 L E V E A S O T I E N V K A K I O O K E G I P P D O O R L I F A G K O L E Q G R ACTTGAAeTCGAAGCTTCCGACACGATCGAAAATGTCAAGGCTAAGATTCAAGATAAAGAAGGAATCCCACCAGATCAGCAAAGACTTATCTTCGCCGGAAAGCAGCTCGAGGACGGCCG '193 1213 1233 . 1253 1273 1293 T L S O Y N I O K E S T L H L V L P L R G G M 0 1 F V K T L T G K T I T L E V E CACCCTTTCGGACTACAACATCCAGAAGGAATCAACTCTTCATTTGGTTCTCCGTTTGAGAGGAGGTATGCAGATCTTCG7CAAGACAT7GACCGGAAAGACCATCACCC7CGAAGTCGA  1313 1333 1333 1373 1393 1413 A S D T I E N V K A K I O 0 K E G I P P 0 Q O R l . I F A a K O L E D G R T l . S O AGCCTCCGACACCATCGAAAATaTCAAGGCCAAaATCCAAGACAAGGAAGGAATTCCTCCAGATCAGCAACOTCTCATCTTCGCTGGAAAGCAGCTCGAAGACGGCCGCACCCTTTCGGA 1433 1483 f 1473 1493 1313 1933 Y N I O K E S T L H L V L f f L R G O M Q l F V K T L T Q I C T I T L E V C A S O T CTACAACATCCAaAAGaAATCAACTCTTCATTTOOTTCTCCOTTTOAQAGGAOGTArGCAAATCTTCOTOAAaACTTTGACTOOAAAOACTATCACCCTCGAAGTCOAAOCTTCTQArAC  1553 1573 1593 1813 1633 1633 I E N V K A K I Q O K E G I P P O O Q R L I F A CATCGAAAATGTaAAaGCCAAGATCCAGGACAAGGAAGGAATCCCACCAGATCAGCAGCGTCTTATCTTTGCCGGTAGCTTATATAGATATACATAACTCAAATCAACTATTATTATTTC 1673 1693 1713 1733 , 1753 1773 G K O L E O G R T L S O Y N t O K E S T L H L V L R L R . a G M O I F V K T L T G AGGAAAGCAATTGGAAGATGGGCGCACGCTCTCTaATTACAACATCCAGAAGGAATCTACTCTTCACTTGGTTCTCCGTCTCCGAGGAaaAATGCAGATCTTCGTCAAGACATTGACTGG  1793  1813  1833  1833  1873  <S93  K T I T L E V E A S O T I E N V K A K t O D K E G I P P O O O R L I F A G K O L AAAGACCATCACACTTGAAGTC8AAGCCTCTGATACCATCGAGAATGTGAAGGCCAAGATTCAAGACAAGGAAGGAATCCCACCAGATCAGCAGAGACTCATCTTCGCCGGAAAACAACT  "913 1933 I9S3 1973 1993 2013 E O a R T L S D Y N l O K E S T L H L V L R L R a o ' w O I F V K T L T G K T I T CeAAGACOaTCOTACCCTCTCCOACTACAACATCCAA**GO»OTCT*CTCTTC*TTTaOTTCTCCQTCTO*OAOGAGOT»TOCAOATCTTCOTCA»Q*CTCTT»CTGO*»*OACC»TCAC  2033 2033 2073 2093 2113 3133 L E V E A S O T I E N V K A K I O O K E O I P P O O O R L I F A G K O L E D G R ACTTGAAGTCGAAGCCTCTGATACCATCGAGAATGTOAASGCCAAGATTCAAGACAAGGAAGGAATCCCACCAGATCAGCAGCGCTTGATCTTCGCCGGAAAACAACTTGAAGACGGTCQ !"1 2173 2193 f 3313 2233 3253 T L S O Y N t O K 6 S T L H l . V L I J 1 . R S a M O I F V K T l . T a i e T I T t . E V E T*CCCTTTCCGACTACAACATTCAAAAGGAGTCTACTCTTCATTTGGTTCTCCGTCTGAGAGGAGGTATGCAGATCTTCOTCAAGACATTGACCGGAAAGACCATCACCCTCGAAGTCGA 2273 3293 2313 3333 33S3 3373 A S O T I E N V K A K I O D K E G I P P D O O R L t F A AGCCTCCGACACCATCGAAAATGTCAAGGCCAAGATCCAAaACAAOGAAGGAATTCCACCAOATCAGCAaAaACTTATTTTCGCTaGTGAaTTCATATTGTTTTAaAATTAAAACTAATT 3393 2413 3433 3493 2473 . 3493 ^ K O L E O a n T L S O V M I O K E S T L H L V L R L R a a * I O I F V TTTATTGTTTTTCAGGAAAGCAACTCGAGGATGGCCGTACCCTTTCGGACTACAATATCCAGAAGGAGTCTACTCTTCATTTGGTGCTCCOTCTCAGAGGAGGTATOCAGATCTTCaTCA 2S13 3333 3S53 3973 3993 2613 K T L T O K T I T L E V E A S O T I E N V K A K I O O K E O I P P O O O R L t F AGACTTTaACTGGAAAAACCATCACTCTCaAGGTCGAAGCTTCGGACACCATTGAGAATGTCAAAGCCAAAATCCAGGATAAGGAGGGAATCCCACCAGATCAGCAACOTTTOATCTTTa 2633  3693  2673  2693  .  3713  2733  A G K O L E D a R T L S D Y M I O K O S T L H L V L R L » a G < ! > ( I ) •  CTGGAAAGCAGCTCGAGGATGGACGCACTCTATCCGATTACAACATCCAAAAGCAGTCGACACTTCATCTCaTTCTTCGTCTTCGCGGAGGAGACATTTAAATCGAACCCATCAATTCAC 2793 27T3 2793 2813 2833 28S3 TCQTTATTCCTCCTCGAGATCTCCGTTCAAGTAACAATTATTTATTCTTTATTCTTCGGGAATTTCTGTATTTTAATQAACOAGCTCTaAATAAATTCATTTTCQTQTACTCAAACGATT  • 2863 2873 2883 2893 2903 3913 2923 TATCTTTATCTTTAACAATAACAAACAACAAAGATAACTACTCTATGAAOTGTAAGGTTCAACTATATTTATAGATC  Figure 8. D N A sequence of the C elegans polyubiquitin gene, UbiA. The A of the mefJiionine initiator codon of the first ubiquitin repeat unit has been numbered +1; thus, the start site of transcription is at -455 ( ). The translation of UbiA into amino acids is given above its nucleotide sequence. Individual repeat units are separated by small arrowheads. The 3' splice signal sequence used for trans splicing is underlined darkly, and the adjacent nucleotide ( A ) represents the site of ligation of the 5'-spliced leader mini-exon. The presumptive T A T A element, G A A T A A , is underlined at -487. Various potential regulatory sequences are also indicated; two elements closely matching the H S E consensus are boxed, a sequence resembling the binding site for the mammalian steroid hormone receptor is indicated by two arrows, a block of cytosine nucleotides is underlined with dots, and 8-base-pair direct repeats are indicated by broken underlines. The two non-ubiquitin amino acids at the end of polyubiquitin are circled, and the ochre stop codon is indicated by an asterisk. Three sequences potentially involved in efficient m R N A 3'-end formation axe underlined, as is the polyadenylation signal ( • ) . A  60  final repeat of UbiA has a two amino acid carboxy terminal extension which differentiates it from the other ten repeat units. The 76 amino acid ubiquitin sequence is identical for each of the eleven repeats; this is surprising, given that one might expect variation to be tolerated in such a redundant organization. There is degeneracy in the ubiquitin nucleotide sequence repeats in the wobble position of some codons (see Figure 9). Overall the codon usage of the UbiA gene is consistent with the highly asymmetric C. elegans codon bias (Emmons, 1988; Fields, 1989). Most codons in C. elegans are biased against wobble position purines. The most dramatic examples of the C . elegans codon asymmetry are found in the proline and glycine codon families : for all of the genes that have been sequenced from C. elegans, 88% of prolines are encoded by C C A and 82% of glycine codons are G G A , even though there are four members in each codon family. In UbiA, three of the glycine codons are G G A in all eleven ubiquitin repeats and overall 76% of the glycines in UbiA are G G A . The glycine 53 codon is the most skewed UbiA glycine codon, being G G A in only two of the eleven repeats. C C A is the most frequent proline codon for UbiA (82% of the 22 possible prolines). As well, UbiA has seven lysines per ubiquitin repeat, but A A A is found in only 9 of the 77 codons. C. elegans UbiA uses all types of leucine and arginine codons (which differ in both first and third position nucleotides). It is interesting to note the evolutionary selection pressure on the ubiquitin gene sequence when one considers the ubiquitin D N A sequence from a ciliate, T. pyriformis, which possesses an alternate nuclear genetic code. In the universal nuclear genetic code, C A A or C A G encode glutamine. In ciliates glutamine may also be encoded by T A A and T A G , normally eukaryotic stop codons (Hanyu et al., 1986). All four of the glutamine residues in T. pyriformis ubiquitin are encoded by T A A , rather than the typical C A A or C A G codons (Neves et al-, 1986). This  61  M  Q  I  F  V  K  T  L  T  G  K  T  I  T  L  E  V  ATG CAA ATC TTC GTC AAA ACG TTG ACT GGA AAA ACT ATC ACC CTG G Q —x C G T —A — T G — T C-T G —C C G G —T G —C A —T G G --A G —C A —T ~G G —A C G —C C G G G  E  A  G —A G — T C-T G —A  S  D  T  I  E  N  G —C G —C G —C  C  V  K  A  K  GAG GTG T —A —C —A — T —A —C —A —C  A — T —A —C A — T —A —C C —A —C  I  Q  D  K  E  GAG GCT TCC GAT ACC ATC GAA AAT GTC AAA GCC AAG ATC CAA GAC AAG GAA  G  I  P  P  D  Q  Q  R  L  I  F  A  G  K  Q  L  E  GGA ATT CCA CCA GAT CAG CAG AGA CTT ATT TTT GCT GGA AAG CAA CTC GAG — T —C — T —G A C-T —C —C —C —C G T  A C-T  C C —C —C —C —C T-G — C — C — C c  A A  T-G — A A T —A  62  I  Q  K  E  S  T  L  H  X  1  I  ooonc  I I I1  GAT GGC CGT ACC CTT TCG GAT TACAAT ATC CAG AAG GAA TCA ACC CTC —A —C —G — T — T — T —C —T —T —T A —C —A —C — T —T —A —G — T — T  i  —A — T —A  1  I  I I  I I  G —C —G —C —T —C — T C —C —C A —C — T —A —C  L  V  L  R  L  R  —C  G  G  D  I  —A  —T —G — T —G — T a x —G —G  —T —T —T —T —T —A  ——T  I p  N  Y  H  D  1  S  I  L  I  T  O O  R  1  G  I  D  1  —T —T — c —T —T —T —T  OCH  Figure 9.  1  c-c  1  I l  H H H  l l  o a Of  I I  1  —T —T —T —T  H  1  —T —T —G C-C — T — T  I I l I  TTG GTC CTC CGC CTA AGA GGA GGA c — — T — G — T — c C-A — T — T — T T-G —G —T — c —T —T —T —T --T — T T-G —T T-G  D N A sequence alignment of eleven coding region repeats of UbiA.  The D N A sequence from each of the eleven ubiquitin repeats of UbiA mRNA is presented. Triplet codons from successive repeats are aligned vertically. (-) indicates the occurence of an identical nucleotide. Amino acids are indicated above the codons by the standard one letter code. O C H : ochre stop codon.  63  alteration maintains the evolutionarily conserved amino acid at those positions in T. pyriformis ubiquitin. The predicted amino acid sequence of C. elegans ubiquitin is identical to that of human, plant, and yeast ubiquitin at a minimum of 72 out of 76 positions (Figure 10).  It most closely resembles human ubiquitin, differing only at amino acid 19.  There are only four sites of variation in the amino acid sequences for ubiquitin in the above organisms : these substitutions are at amino acids 19, 24, 28, and 57, and they are not necessarily conservative. Four additional sites of amino acid variation have recently become evident from further cloning and sequencing of ubiquitin genes (see Figure 10). The glutamic acid at amino acid 16 is replaced by an aspartic acid in the ciliated protozoan Tetrahymena (Neves et al., 1986) and the conserved threonine at position 22 of previously published ubiquitin sequences is replaced by asparagine in the slime mould Dictyostelium (Giorda and Ennis, 1987). Two additional sites of variation are present in trypanosome ubiquitin, namely at amino acids 9 and 52 (T. cruzi: Swindle et al., 1988 and T. brucei: Wong and Campbell, 1989). In general, Trypanosomes seem to favor the substitution of alanine at evolutionarily variant locations in ubiquitin.  3.4  Genomic Southern Analysis Southern hybridization analysis was undertaken to confirm the arrangement of  the UbiA gene in Bristol strain C. elegans (N2) genomic D N A and to determine whether the polyubiquitin structure is conserved in other nematodes. Figure 11 represents acomparison of two strains of C. elegans (Bristol: N2 and Bergerac : BO), a related species (C. briggsae). and a plant parasitic nematode (Steineremia feltiae). Restriction enzymes that cut outside (Sail) or within (EcoRI) the polyubiquitin coding region were selected and digests were probed with an EcoRI-Bgin fragment  64  U MQIFVKTL GKTITL VE SD I NVK KIQDKEGIPPDQQRLIFAGKQLE GRTL DYNIQKESTLHLVLRLRGG C.e. H.s. S.C. H.v.  T T T T  T.p. D.d. T.b. T.C.  T T A .. A  E E E E  A T P T S T S .T  D A - E G E A E S  T N T T  E E D D  A A S S  D D D D  S S S S  E E E E  A A A T  D D E D  S S A A  Figure 10. Interspecies comparison of ubiquitin amino acid sequences. Invariant residues are shown as the consensus sequence (U). Positions of interspecies variation are indicated as gaps; the corresponding amino acid substitutions in nematode (C.e. : Caenorhabditis elegans). human (H.s. : Homo sapiens : Wiborg et al., 1985), yeast (S.c.: Saccharomyces cerevisiae : Ozkaynak et al., 1984), plant (H.v.: Hordeum vulgare : barley : Gausing and Barkardottir, 1986), ciliate (T.p. : Tetrahymena pyriformis : Neves et al., 1988), slime mold (D.d. : Dictvostelium discoideum : Giorda and Ennis, 1987), and trypanosome (T.b.: Trypanosoma brucei: Wong and Campbell, 1989; T.c.: Trypanosoma cruzi: Swindle et al., 1988) ubiquitins are indicated. Asterisks indicate the codons interrupted by introns in published ubiquitin sequences : isoleucine codon 3 is interrupted in two of the yeast ubiquitin fusion genes (Ubil and Ubi2), while glycine codon 47 is interrupted in four of the eleven ubiquitin repeat units of the nematode U b i A polyubiquitin gene. Amino acids are represented using the standard one letter code.  65  EcoRI  Sail  Sail  BO N2 M BO N2  I  - 2 3 . 1 kb  w ft -  3.  N2 BO Cb Sf 99 6:6 kb k  b  4.4 kb -  2.3 kb 2.0 kb  -  0.5 kb  mm t  23.1 9.9 6.6 4.4  kb kb kb kb  2.3 kb 2.0 kb  0.5 kb  Figure 11. Southern hybridization analysis of ubiquitin gene structure. In the left panel, genomic D N A (2 ug) from C. elegans strain Bristol (lanes N2) or strain Bergerac (lanes BO) was digested with Sail or EcoRI, separated by electrophoresis on a 0.7% agarose gel, and transferred to nitrocellulose. The nitrocellulose replicas were hybridized to a portion of the UbiA coding region (EBT, an EcoRI-BglH fragment of UbiA - see Figure 12) and fluorographed. In the right panel, genomic D N A (2 ug) from nematodes related to C. elegans was digested with Sail and treated as above. C. briggsae (Cb) is a free-living nematode closely related to C. elegans. while S. feltiae (Sf) is a plant parasitic nematode. Size markers were X D N A digested with HindlH.  66  (EBT, described in legend of Figure 12) of the coding region which would cross-react with any ubiquitin sequence. As seen in the left panel of Figure 11, the UbiA gene is located on a 3.4 kb Sail fragment in genomic D N A of both C. elegans strains. Digestion with EcoRI yields the expected hybridizing fragments of 4.0 kb, 0.9 kb, 0.6 kb, and 0.5 kb corresponding to the UbiA gene D N A sequence (the 1.8 kb fragment containing the first 109 nt of the first ubiquitin repeat is not visible in this exposure). Similar results have been obtained for Xbal, SstI, PstI, Xhol, B g i n , H i n d i n , and Sau3AI digest comparisons (data not shown), confirming the U b i A organization predicted from XUbl and XUb2 phage D N A digests. Sail restriction digests of D N A from Steinerema feltiae (a related nematode genus) show a similar UbiA gene organization (however one extra ubiquitin repeat could also account for this result); surprisingly, however, there is some difference in UbiA organization in the D N A of a sister nematode species, C. briggsae, manifested as a restriction fragment length polymorphism. Unassigned hybridizing restriction fragments on the Southern blot in the left panel of Figure 11 (e.g. the 2.3 kb EcoRI fragment) must arise from ubiquitin gene(s) distinct from the UbiA gene. Southern analysis of cosmid digests indicate that there are no sequences homologous to the UbiA coding region within 100 kb downstream (data not shown). There is evidence for a class of ubiquitin hybrid gene(s) in C. elegans based on Northern hybridizations and polymerase chain reaction (PCR) analysis of cDNA from C. elegans and C. briggsae (discussed below).  3.5  Northern Hybridization Analysis Northern analysis of total nematode R N A was performed using the EBT coding  region probe of UbiA, as described in Figure 12.  EBT hybridization reveals two  main ubiquitin-encoding mRNAs in C. elegans (see left panel of Figure 12). The more abundant mRNA is 2600 nucleotides, sufficient to encode the eleven repeat  67  N R  2 N  A  EBT DNA Probe  Figure 12. Northern hybridization analysis of C. elegans ubiquitin gene expression The left panel shows the two main ubiquitin m R N A species found in C. elegans , U b i A (approximately 2600 bp) and UbiB (approximately 700 bp). Total cellular R N A (10 ug) from C. elegans Bristol strain nematodes was separated on an agaroseformaldehyde gel, transferred to a nylon membrane, then fixed by ultraviolet light crosslinking. This membrane was hybridized with an EcoRI-Bgin fragment of the U b i A gene coding region (EBT) in order to detect all ubiquitin mRNAs present in C. elegans. Size markers are indicated. EBT is shown schematically in the right panel; positions of the restriction sites in the nucleotide sequence of U b i A (see Figure 8) are indicated.  68  UbiA polyprotein, while the other, less abundant, mRNA is 700 nucleotides, sufficient to encode two to three ubiquitin repeats or a ubiquitin sequence fused to an unrelated sequence. There is precedent for trimeric ubiquitin genes (human UbiB : Baker and Board, 1987a) and for ubiquitin hybrid genes (yeast UBI1-3 : Ozkaynak et al., 1987). In order to determine whether the smaller mRNA was a processed form of UbiA, Northern hybridization analysis was performed using a D N A probe derived from the non-ubiquitin sequence in the 3' untranslated region of UbiA(FS, described in Figure 13). The fact that this probe hybridized exclusively to the U b i A mRNA indicates that the smaller mRNA(UbiB) arises from a different ubiquitin locus and is not a processed form of UbiA (see Figure 13). Southern hybridization data confirm that at least two distinct ubiquitin loci are present in C. elegans D N A and indicate that the gene encoding the 700 nucleotide UbiB mRNA is likely wholly or partially contained on a 2.3 kb EcoRI fragment(see Figure 11). PCR analysis of cDNA from C. elegans and C. briggsae indicates that the UbiB mRNA is analogous to the Ubil and Ubi2 mRNAs of yeast (Ozkaynak et al., 1987; see below).  3.6  Quantification of UbiA Expression Often mRNA abundance is estimated by the frequency of gene-specific clones  present in a cDNA library; unfortunately, these estimates are biased due to differences in cDNA cloning efficiencies.  R N A dot blot hybridization represents a  much more accurate method for quantifying mRNA abundance for a given gene. In order to quantify the relative amount of Ubi mRNA present in the nematode cellular R N A pool, C. elegans RNA was prepared and subjected to dot blot hybridization. Figure 14 shows a dot hybridization result used to quantitate UbiA mRNA. In order to specifically quantitate UbiA (and not UbiB) the probe chosen was the UbiA FS fragment (see Figure 13).  69  N  FS  2  RNA  DNA Probe §  _  2  5  k  — 0.7  k  kb  UMIO  Ubil 1  FnuDII 2697  3UTR  Sstl 2815  Figure 13. UbiB is not derived from the UbiA locus. The left panel is the result of Northern hybridization of C. elegans R N A to a FnuDII-SstI fragment (FS) of the 3' untranslated region of the UbiA gene. Conditions were comparable to those in Figure 12. It was expected that the FS D N A fragment would be unique to C. elegans UbiA. being outside the repeated ubiquitin coding region. FS is shown schematically in the right panel; positions in the D N A sequence of UbiA (see Figure 8) are indicated.  70  As can be seen from Figure 14, the hybridization signal for 2 ug of total R N A with the FS probe most closely corresponds to that obtained for 10 pg of FS fragment DNA. RNA.  Thus, UbiA R N A represents approximately 2 pg per ug of total nematode If one assumes that polyA R N A is 5% of total R N A , U b i A represents +  0.003% of polyA mRNA(for calculations see Appendix C). If all of the estimated +  5000 C. elegans genes were expressed equally, the UbiA mRNA would account for 1 out of every approximately 5000 transcripts, or 0.02% of polyA mRNA. Thus the +  UbiA gene product would not be classed as a highly abundant transcript.  3.7  U b i A 3' Flanking Region The UbiA polyubiquitin precursor polypeptide has a distinct carboxy terminal  sequence : the final ubiquitin repeat has a two amino acid extension prior to the ochre stop codon. The 3' untranslated region is approximately 140 nucleotides and includes a typical polyadenylation signal (AATAAA) 110 bp downstream of the stop codon. This region is rich in secondary structures and includes many sequences which have previously been shown to be involved in efficient 3' end formation, including A G T G T A A G (McLauchlin et al., 1985), T T C A A C and T C A A G (Urano et al., 1986)(see Figure 8). The exact site of polyadenylation of the UbiA mRNA has been determined by sequencing of UbiA cDNA clones obtained by conventional cDNA library screens (data not shown) and by ubiquitin cDNA PCR (see Figure 15). Polyadenylation occurs at or near the A residue (position 2846 in Figure 8) 18 nucleotides downstream of the A A T A A A sequence. The 3' end sequence of UbiA from C. elegans and C. briggsae are compared in Figure 15. The ubiquitin coding regions from the two UbiA cDNAs are nearly identical and encode identical translation products. The 3' untranslated regions of the two UbiA cDNAs are also highly conserved. There is evidence that regions of D N A sequence conserved  7  1  • ••  FS  D N A  f  R N A  Figure 14. (^antification of U b i A expression by R N A dot hybridization. Varying amounts of total cellular R N A from C. elegans embryos and denatured single stranded M13 D N A standards (i.e. M13 D N A with the FS insert shown in Figure 13 in the appropriate orientation) were applied to nitrocellulose paper. These samples were hybridized to radiolabeled FS D N A fragment then were fluorographed. Samples were as follows : a,b,c,d,e = 250,100, 50, 25,10,5 pg FS single stranded template D N A ; f,g,h,i = 4, 2,1, 0.5, 0.25 ug total R N A .  72  C.e.:  ATG CAG ATC TTC GTC AAG ACT TTG ACT GGA AAA ACC ATC ACT CTC  C.e.:  GAG GTC GAA GCT TCG GAC ACC ATT GAG AAT GTC AAA GCC AAG ATC  C.e.:  CAG GAT AAG GAG GGA ATC CCA CCA GAT CAG CAA CGT TTG ATC TTT  C.e.:  GCT GGA AAG CAG CTC GAG GAT GGA CGC ACT CTA TCC GAT TAC AAC  C.e.:  A T C CAA AAG GAG TCG ACA CTT CAT CTC GTT CTT CGT CTT CGC GGA  C.e.: C.b.:  GGA GAC ATT TAA ATCGAACCCA TCAATTCACT CGTTATTCCT CCTCGAGATC T G A-TT T—T A T  C.e.:  TCCGTTCAAG TAACAATTAT TTATTCTTTA TTCTTCGGGA ATTTCTGTAT  C.b.:  AT  C.e.: C.b.:  TTTAATGAAC GAGCTCTG AATAAA TTCATTTTCG TGTACTC (A) * -AA—C T-A—CA (A) *  •  T  A-A  C  A—C  T  T—A—  Figure 15. DNA sequence comparison of C. elegans and C. briggsae UbiA cDNAs. Comparison of the DNA sequences from the last ubiquitin repeat of the UbiA genes of C. elegans and C. briggsae. The C. elegans sequence(C.e.) was determined from three PCR amplified subclones and from a cDNA clone obtained from a X.ZAP cDNA library (this library was a gift from R. Barsted, Washington University). The C. briggsae cDNA sequence(C.b.) was determined from three PCR amplified subclones. It should be noted that the sequence of the first 20 nucleotides of the C. briggsae UbiA (which corresponds to the oligonucleotide primer used in the amplification reaction) can not be determined with absolute certainty using this technique. The cDNA sequence shown is from the start codon (ATG) of the final ubiquitin coding region of UbiA to the adenine nucleotide which is polyadenylated (* = polyA tail). Conserved positions in the C. briggsae UbiA cDNA sequence are indicated by dashes (-). The polyadenylation signal of UbiA is underlined.  73  between these two nematode species are important to gene function and regulation (Prasad and Baillie, 1989).  3.8  Intragenic Introns A notable feature of the C. elegans polyubiquitin gene is the presence of introns  within the coding region: all other reported polyubiquitin genes (e.g. yeast: Ozkaynak et al., 1984; insect: Arribas et al., 1986; human : Wiborg et al., 1985; and plant: Vierstra et al., 1987) appear to lack introns or spacers of any kind within the coding region. Short introns interrupt the glycine-47 codon G / G A in 4 of the 11 UbiA repeats; the position of these introns is identical in each repeat. The introns obey the loose C. elegans splice consensus sequences and show a degree of sequence similarity (see Figure 16). Southern data indicate that at least one other genus of nematode (Steineremia feltiae) may harbor intragenic introns within its UbiA gene, based on conserved restriction fragment length in Southern analysis (Figure 11).  3.9  UbiA Trans Spliced Intron The first ubiquitin repeat of UbiA is preceded by a sequence which closely  resembles a 3' splice consensus sequence. This 3' splice site in the U b i A upstream noncoding region is used for splicing; however it undergoes a trans, rather than conventional cis, splicing reaction (shown below). A 22 nucleotide leader exon from a separate R N A is joined to the splice acceptor site of the UbiA precursor in order to generate mature UbiA mRNA. This 22 nucleotide sequence is identical to that of SL1, the trans spliced leader of the actin genes, which is derived from the 5' end of a 100 nucleotide hnRNA (Krause and Hirsh, 1987; Nelson and Honda, 1989). Figure 17 shows the nucleotide sequence of 477 base pairs of 5' non-coding region and the first 109 nucleotides of the first ubiquitin coding unit, of C. elegans UbiA with the A of the A T G initiation codon labelled as nucleotide "+1". At  Intron  I  Intron  IV  Intron  II  Intron I I I C.e.  GrTACCTTGO^  — G A T  — T - T - G - T - T - G - A — A - A A - C — T T  A--T-TA-CAA-T-TK3-TCATGAG-GA  G C —AT-T-GATATAC  T  TGT-TTTCAG  C - A - - T -  AC-C-AA-C--CT  GTAAGT  TTTCAG  TTTCAG  TTTCAG  Figure 16. D N A sequence comparison of the four UbiA cis spliced introns. The D N A sequence of each of the four cis spliced introns of the UbiA locus are aligned relative to the first intron. (-) indicate conserved nucleotides. Gaps have been introduced where necessary to maximize alignment. The C. elegans cis intron consensus sequences (C.e., Blumenthal and Thomas, 1988) are shown below for comparison.  75  <-"C.ClCMM.KMSTC»OI(!T CCCCIOCGTT 0  M C  c;??^TCCO«ICCITMKCO=C«r.C«=C«="  121  ' -13* """WTTOCCTMMOlfcTOMiMljmTW 1  6  5  fi ?  CE  EX  <«3  1  5  . BT  CLAI  "O.O o« Oft  -.  -44 «  3  *  ^' 'c'''c''»'c»ccTcricTcTcc«TCCTc»TTCc.OT^^ :  •  0.5  A  rGGG . GrGGAGGCT tCCG»»tCCT lCG»A»T .GC t»»»GCC«AGJTCc2G . ScG liGG tl G»«TC lC ECOB I  Figure 17. Northern hybridization analysis of UbiA leader sequence. The right panel is a Northern hybridization analysis of the UbiA 5'-flanking region with two UbiA fragments : one fragment including ubiquitin-encoding sequences (EX) and the other consisting only of ubiquitin flanking sequence (CE). EX : EcoRI (+108) to Xba (-824) probe, CE, Clal (-213) to EcoRI (-1650) probe. XHindm size markers (in kb) are indicated on the right. The left panel shows the nucleotide sequence of the 5'-end and first 36 codons of the UbiA gene, and includes labelling pertinent to this figure and Figures 18-20. The 'A' of the methionine initiator codon for the first ubiquitin repeat has been numbered "+1". The 3'-splice consensus sequence (underlined) and the sequence of the oligonucleotide primer used for primer extension R N A sequencing in Figure 19 (overlined) are indicated. Triangles indicate positions of SI nuclease protection (see Figures 18 and 20): (a) the large open triangle indicates the 5'-most nucleotide of the UbiA gene which is present in the mature transcript; (b) the large solid triangle indicates the additional nucleotides of the EX probe protected by base pairing with the trans spliced leader sequence; (c) the small solid triangle at -455 indicates the probable site of transcript initiation.  position -8 is a C A G trinucleotide, preceded by 6 T's, which corresponds to known 3' splice site recognition sequences. This suggested the presence of an intron in the 5' non-coding region of this gene. In order to locate the 5' splice site of the intron, as well as potential promoter elements, a 1.6 kilobase segment of D N A upstream of the translation initiation codon was sequenced. Northern and SI analyses of C. elegans R N A were also performed using probes derived from the upstream region. Most C. elegans introns are short and clustered in length at approximately 50 bp (Blumenthal and Thomas, 1988), so it was expected that a 5' splice donor sequence would lie immediately upstream of the splice acceptor in the UbiA leader. A visual search of the UbiA upstream region D N A sequence revealed no good candidate 5' splice donor sequence. The right panel of Figure 17 shows that while a probe derived from the coding region of the UbiA gene (EX) detected the expected transcripts of 2600 and 700 nucleotides on Northern blots, a 1.2 kb Clal-EcoRI fragment (CE) extending upstream from position -213 showed no hybridization to mRNA, suggesting that the UbiA mRNA leader exon was also not found immediately upstream of the UbiA translation start site. SI nuclease protection analysis of ubiquitin mRNA was then performed using a probe extending from the EcoRI site within the first coding repeat to an Xbal site 829 base pairs upstream of the UbiA translation start site. Unexpectedly, two major protected fragments were seen (Figure 18); these fragments were also present when the experiment was performed using the technique of RNase mapping (data not shown, RNase mapping is described in Zinn et al., 1983). The smaller protected fragment corresponds to the region extending from the 3' splice junction to the EcoRI site in the coding region. The larger band of 119 nucleotides was of uncertain origin and did not appear to correspond to a 5' non-coding exon. Its nature will be explained below.  7 7  92  I  Figure 18. SI nuclease protection analysis of the UbiA mRNA. Total cellular RNA (10 ug) prepared from a C. elegans mixed life cycle stage culture was hybridized to a single stranded radiolabelled D N A fragment and the single stranded regions were then digested with SI nuclease. The single-stranded EX D N A fragment (extending from the EcoRI site within the first ubiquitincoding repeat to an Xbal site 829 base pairs upstream of the UbiA translation start site) was prepared by primer extension of an M13 template followed by gel isolation. The two main protected EX fragments are indicated by triangles. Hpall digested pBR322 D N A size markers are indicated on the left.  78  Since it was possible that the UbiA leader exon might be very short and as a result might have gone undetected in both the Northern and SI analyses, and because of the unexpected SI nuclease protection result, primer extension R N A sequencing was undertaken. A 15 base oligonucleotide complementary to the downstream region flanking the 3' splice site was used for R N A sequencing. The sequence complementary to the primer used is shown in Figure 17. It should be noted that the choice of priming oligonucleotide was restricted; oligonucleotides of greater length or complementary to sequence further 3' of the splice site would prime at multiple sites within the gene due to the tandemly repeated nature of the UbiA sequence. The results of chain termination sequencing reactions performed by K. Van Doren (Synergen, Boulder, Colorado) using M M L V reverse transcriptase and the above primer on C. elegans R N A are shown in Figure 19. A n unambiguous sequence of 17 nucleotides was obtained, corresponding exactly to the last 17 nucleotides of the SL1 trans spliced leader sequence seen on the C. elegans actin3 gene (Figurel9; Krause and Hirsh, 1987). The sequence corresponding to the start site of UbiA transcription is indicated by a small filled triangle in Figure 17. It can be seen that the adjacent sequence bears no homology to the splice leader sequence, proof that the UbiA splice leader is acquired by a trans, rather than cis, splicing event. In fact, the SL1 sequence does not occur upstream of the UbiA gene, within the 1.6 kb sequenced. It is also of interest to note that reverse primed sequencing of the actin genes often leads to ambiguous sequence due to cross homologies between the four genes (K. Van Doren, personal communication); the unambiguous sequence from UbiA may imply UbiB may also acquire the splice leader sequence.  79  Actin3 EGAATC  UbiA EGATC  Figure 19. Primer extension sequencing of the UbiA and Actin3 mRNAs. An oligonucleotide complementary to the sequence immediately downstream of the 3'-splice site of the UbiA 5'-noncoding region intron was used as a primer for sequencing the 5'-end of the UbiA mRNA. The panel on the right shows the results obtained with the UbiA primer. The panel on the left shows the same experiment performed with a primer complementary to the 5' region of the C. elegans actin 3 mRNA. Sequences of the oligonucleotide primers(ubiA, act3) are listed in Appendix A. E : primer extension in the absence of dideoxy nucleotides; G, A , T, C : primer extension using the corresponding dideoxy nucleotide. Data was obtained from K. Van Doren, Synergen Corporation.  80  The 3'-most six nucleotides of the 22 nucleotide leader are identical to the 3' splice site of the UbiA gene, with one mismatch. This may explain the origin of the 119 nucleotide fragment seen in the SI nuclease protection analysis in Figurel8 : the 119 nucleotide product occurs due to protection of the 3' portion of the SL1 sequence annealed to the trans spliced leader and the UbiA 3' splice site border, and the 113 nucleotide product occurs due to protection at the correct 3' splice site. Alternatively, the 113 nucleotide protected fragment could result from protection of the EX probe by the UbiB transcript.  3.10  U b i A 5' Flanking Region To ensure that the 22 nucleotide leader sequence found on the 5' end of the  mature UbiA mRNA is acquired by trans splicing, and in order to define the 5' noncoding regulatory sequences of UbiA it was necessary to locate the transcription start site for the gene. D N A sequences involved in the regulation of transcription are most commonly found in a short promoter region upstream of the site of transcription initiation. As the first nucleotides of the UbiA hnRNA are discarded in the formation of the mature trans spliced UbiA mRNA, it was necessary to detect the rare trans un-spliced UbiA hnRNA in order to localize the start site of transcription. SI nuclease protection assays were successfully employed to detect the start site of transcription for the UbiA gene(Figure 20). Preliminary attempts to detect unspliced UbiA hnRNA in preparations of total R N A from C. elegans were unsuccessful. It proved possible to improve the sensitivity of SI analysis by protecting an end-labelled Clal-Xbal (CX) probe derived solely from 5' intron sequence (and hence specific for the hnRNA) and by using large quantities of total RNA. The end-labelled CX probe protects an R N A fragment 244 nucleotides in length.  81  P  RNA  || 529 - I  406 -  311-  244 _ 240 "  219203 192 182 -  162 -  149-  I Figure 20. SI nuclease protection analysis of U b i A h n R N A . A n end-labeled single-stranded D N A fragment extending from C l a l (-213) to Xbal (-824) was prepared by primer extension. Total cellular R N A (34 ug) from C . elegans embryos heated at 3 4 ° C for 1 hour was hybridized to this fragment, then SI nuclease was used to digest un-hybridized single stranded R N A and D N A . The protected fragment is indicated by an asterisk. Lane P represents a sample containing only the C l a l to Xbal fragment. pBR322 size markers are indicated on the left.  82  This SI nuclease protection experiment localized the major site of transcript initiation to a sequence 450 nucleotides upstream of the 3' splice site used for trans splicing. The major UbiA promoter elements are thus likely to be located immediately upstream of this region. Careful scrutiny of the sequence immediately upstream of the UbiA start site reveals no consensus T A T A motif in the proximity of the start site; however, a G A A T A A sequence present at -485 is the likely analog for UbiA (see Figure 8). Multiple 'CAAT' box homologies are found within 200 nucleotides of the start site, including one between the G A A T A A box and the transcription initiation site. Since the D N A of C. elegans is 67% A / T rich (Sulston and Brenner,1974), the sequence C T C C located at -782 may well have some 1 5  functional significance to UbiA gene regulation. At positions —653 and —828 of UbiA are sequences which resemble the heat shock element (HSE) C - G A A - T T C - G (Pelham, 1982). This element is found upstream of most eukaryotic heat shock genes and has been shown to confer heat inducibility. The essential role of this sequence in the positive regulation of transcription has been demonstrated by in vivo expression of mutated heat shock genes (Dudler and Travers, 1984) and by the use of synthetic HSEs to make other genes heat inducible (Pelham and Bienz, 1982). The sequence at -653 matches the HSE consensus sequence at 12 of 14 positions but lacks dyad symmetry and is hence likely nonfunctional. The sequence at -828 matches the HSE consensus at 13 of 14 positions but is quite distant from the UbiA transcription start site. The UbiA 5' flanking region contains an inverted repeat at -822, overlapping the distal HSE. A search of published sequences revealed homology between this inverted repeat and the consensus binding motif for mammalian steroid hormone receptors (Scheidereit et al., 1983). Whether the inverted repeat in the C. elegans UbiA 5' end functions to confer steroid sensitivity to UbiA transcription is not known. This element is a considerable distance from the transcription start site and  83  hence may be only weakly functional. Another sequence of potential interest in the UbiA 5' flanking region is a 8 nucleotide direct repeat at -576 and -477 (underlined in Figure 8). The significance of this element is unknown.  3.11  Expression of Ubiquitin  3.11.1  Developmental Analysis of Ubiquitin Expression  In order to further characterize UbiA gene expression and understand its regulation, experiments were undertaken to assess whether ubiquitin expression is regulated during the development of C. elegans. The C. elegans life cycle is shown diagrammatically in Figure 21. Larval development of C. elegans proceeds from embryos through four larval stages followed by the adult hermaphrodite or male stage. This process occurs in approximately three days under laboratory conditions. It is thus relatively easy to obtain reasonable amounts of synchronized individuals from each life cycle stage. Pure total cellular R N A was prepared from synchronized populations of all C . elegans life cycle stages by the Guanidinium-CsCl method (Chirgwin et al., 1979) and was used for Northern hybridization analysis. When the cross-reacting EBT fragment from the UbiA coding region (diagrammed in Figure 12) was used for Northern hybridization of C. elegans life cycle stage RNAs, the level of both UbiA and UbiB mRNAs remained relatively constant throughout development. Figure 22 represents a typical experiment; levels of UbiA mRNA fluctuated in amount between experiments and relative to UbiB mRNA levels, however never in a consistent fashion. In this experiment it appeared that UbiA mRNA was induced up to three-fold in nematode life cycle or dauerlarva stages, however this observation could not be reproduced consistently in other experiments.  84  Figure 21. The life cycle of C. elegans. The progression of the C. elegans life cycle is diagrammed (after Riddle et al., 1981). The solid arrows represent periods of moulting between larval stages. The entire life cycle requires approximately three days at 25°C.  85  C. elegans Development Eggl_1 L2 L3 L4 A D  Figure 22. Northern hybridization analysis of ubiquitin mRNA in development. For each C. elegans life cycle stage, 10 ug of total cellular RNA was separated on a denaturing agarose-formaldehyde gel, transferred to nitrocellulose, and hybridized to a radiolabelled fragment of the coding region of UbiA (EBT, see Figure 12). UbiA is the larger RNA, while UbiB is the lower molecular weight mRNA. Lanes are labelled as follows : Egg : embryo; L l - L4 : larval stages; A : adult; D : dauerlarva. Size markers are indicated at right.  86  Ubiquitin transcription was also assayed during the larval moulting period from L l to 12. The moult should represent a period of heavy loading of the intracellular proteolysis system and it was thought that ubiquitin mRNA levels might rise to meet the temporary need for an increased number of free ubiquitin monomers. Furthermore, the possibility existed that a moulting hormone, possibly acting via the sequence with homology to a steroid hormone receptor binding site upstream of the UbiA transcriptional start site, could conceivably regulate UbiA gene expression during moulting. Synchronized larvae were sampled at varying times during the L l to L2 moulting period and Northern hybridization analysis was performed. No consistent changes in ubiquitin R N A levels were observed before, during and after moulting (Figure 23), indicating that ubiquitin gene expression is not altered during this process. —  The observed fluctuation in ubiquitin mRNA levels between samples in these  Northern analyses is not a reflection of unequal loading. R N A concentrations were determined by spectrophotometry and ethidium fluorescence, and the levels of UbiA appeared to vary independently relative to that of the small ubiquitin mRNA. Thus there was some variation in the level of ubiquitin mRNA, but it did not appear to be consistently correlated with progression through development or moulting. It should also be noted that extensive analysis of stage specific R N A has been carried out by Prasad and Baillie (1989), who examined the expression of seven putative coding regions from C. elegans as well as the expression of actin, hsp-1, and P8 (rat sodium channel). They found that most of these mRNAs (including actin) were expressed most abundantly in L2 larvae and could range in expression up to fivefold over the course of the life cycle.  87  L1/L2 1  Figure 23.  2  3  Moult 4  5  6  -  2.5 kb  -  0.7 kb  Northern hybridization analysis of ubiquitin mRNA levels in moulting.  C. elegans was cultured at 22°C and samples were taken at the following times relative to the appearance of mature L2 larvae : lane 1 : 6 hours prior to L2 stage (i.e. during L l stage); lane 2 : 45 minutes pre-lethargus; lane 3 : mid-lethargus; lane 4 : shedding of old cuticle; lane 5 : appearance of mature L2; lane 6 : 2 hours after appearance of mature L2. Hybridization was as described for Figure 22. Size markers are indicated at right.  88  3.11.2  Ubiquitin Expression During Heat Shock  In order to investigate the role of the HSE-like sequences in the UbiA promoter R N A was prepared" from embryos or synchronized populations of larvae subjected to various heat shock regimes. Nematodes normally grow at temperatures between 14 and 25°C. Typically, heating nematodes in liquid at 33°C for more than 15 minutes induces heat shock mRNAs (Jones et al., 1989). R N A dot hybridization experiments were conducted with samples of total R N A from C. elegans embryos which had experienced various times or temperatures of heat shock. In these experiments the levels of UbiA transcription were measured relative to those of a known heat shock gene (C. elegans hsp!6) and a gene known not to be induced by heat shock (C. elegans actin). Our initial assays of ubiquitin mRNA at control and heat shock temperatures yielded inconsistent results. It proved necessary to examine multiple samples for a given heat shock experiment; comparisons of single time points for temperature-stressed versus unstressed nematodes were unsatisfactory. As can be seen in Figure 24, a prolonged exposure to 33°C following a slow rise to this temperature does not appear to result in an increase in UbiA mRNA levels in C. elegans embryos. There is some fluctuation in R N A levels between samples but there is no linear trend. A similar result was obtained if embryos were rapidly raised to heat shock temperature or if L2 larvae were analyzed (data not shown). The expression of ubiquitin mRNA in embryos stressed for 45 minutes at different temperatures is presented in Figure 25. UbiA mRNA levels fluctuated slightly and nonlinearly as the temperature of incubation was increased from 31°C to 34°C. This result is in contrast to the behavior of a typical inducible mRNA such as hsp!6. the concentration of which increases linearly over this temperature range (Figure 25 and Jones et al., 1989). It should be noted that temperatures above 35°C result in extensive alterations in cell physiology, e.g. faulty splicing (Kay et al., 1987)  89  Figure 24. R N A hybridization analysis of UbiA mRNA during chronic heat shock. C. elegans embryos were shifted from 22°C to 33°C, and samples were removed at the following times after temperature elevation: 0 : immediately, 1 : 15 minutes, 2 : 30 minutes, 3 : 45 minutes, and 4 : 60 minutes. C represents control temperature embryos (22°C). Total RNA (3 ug) was analyzed by hybridization to one of the three [ct- P] labeled D N A probes as indicated. The probe for UbiA was the gene-specific FS fragment from the 3'-end of the UbiA locus (see Figure 13). Due to the large difference in mRNA abundance, UbiA and actin dots were exposed for 3 days, while the hsp!6 dots were exposed for only 1 day. Hybridization intensity for each dot was determined by densitometry and is expressed graphically as a proportion of the intensity of the sample with maximum hybridization for that probe (a : actin; u : UbiA; h : hsp!6). Also shown is a photograph of the ethidium fluorescence of the 28S and 18S rRNAs for each R N A sample, indicating equal loading. 32  90  C  Figure 25.  31  32  33  34  R N A hybridization analysis of UbiA mRNA at various temperatures.  R N A samples were prepared and analysed as in Figure 24, except that embryos were elevated to the indicated temperatures (in degrees Celsius; C represents control temperature of 22°C), and samples were taken after 45 minutes.  91  and lethality (Snutch and Baillie, 1983; D. Jones, personal communication), and as such do not accurately reflect heat induction.  3.11.3  Heat Shock Inhibits Trans Splicing in C. elegans  During the course of heat shock experiments an increase in the relative abundance of unspliced hnRNA from the UbiA locus was observed under heat shock conditions. The amount of unspliced hnRNA increases with the temperature of the heat shock (Figure 26); trans splicing reactions must become increasingly affected with higher temperature. Thus trans splicing is likely inhibited by heat.  3.12  Analysis of UbiA Promoter Function in Transgenic Nematodes In order to study the in vivo regulation of UbiA transcription and the role of  upstream D N A sequence elements experiments using transgenic nematodes have been undertaken in collaboration with D. Jones and E. Stringham (University of British Columbia). To assay the UbiA promoter, it was cloned upstream of the E. coli p-galactosidase gene in a microinjection plasmid vector developed by Andrew Fire (A. Fire, personal communication). I have designed PCR oligonucleotides which amplify various promoter regions cloned into this vector (see Figure 27). This type of experiment may be used to determine whether lack of expression of transformed genes is due to promoter damage/ rearrangement. Transgenic nematodes bearing the PC3 plasmid (kindly donated by Dennis Dixon, Johns Hopkins University) were employed in order to develop PCR assays for study of UbiA constructs. This plasmid bears a heat inducible promoter element derived from one of the nematode hsp!6 heat shock genes, the coding region of E. coli (3-galactosidase, and the 3' untranslated region from the nematode myosin heavy chain gene (unc-54). In Figure 28 hsp!6 promoter D N A was amplified from nematodes transformed with the PC3 plasmid using oligonucleotides flanking the  92  Heat Shock  Unspliced hnRN*A  Figure 26. Heat shock inhibits trans splicing of UbiA. C. elegans embryos were slowly elevated to the indicated heat shock temperatures (in degrees Celsius) and were maintained for 45 minutes. Total cellular R N A (35 ug) prepared from these embryos was subjected to SI nuclease protection analysis using the single stranded end-labelled probe CX (see Figure 19). The UbiA hnRNA, which has not undergone trans splicing, is indicated by the arrow. pBR322 size markers are indicated at the left.  93  Promoter  | p - G a l cDNA  PCR —  F i g u r e 27. P C R strategies e m p l o y e d to a s s a y t r a n s g e n i c C. e l e g a n s strains. PCR oligonucleotides were designed which w o u l d amplify various promoter r e g i o n s c l o n e d i n t o the m i c r o i n j e c t i o n p l a s m i d v e c t o r d e s i g n e d b y A n d r e w F i r e (Johns H o p k i n s U n i v e r s i t y ) . T h e s e o l i g o n u c l e o t i d e s ( R G 0 2 a n d R G 0 5 , f o r s e q u e n c e see A p p e n d i x A ) f l a n k the p r o m o t e r c l o n i n g site a n d h e n c e a l l o w a m p l i f i c a t i o n o f i n s e r t e d p r o m o t e r D N A f r o m n e m a t o d e s t r a n s f o r m e d w i t h t h i s vector. O l i g o n u c l e o t i d e s w e r e a l s o d e s i g n e d to a l l o w the d e t e c t i o n o f f J - g a l a c t o s i d a s e r e p o r t e r g e n e m R N A . F i r s t s t r a n d c D N A s y n t h e s i s f r o m total c e l l u l a r R N A is p r i m e d w i t h a m o d i f i e d o l i g o - d T p r i m e r ( a n c h o r o l i g o - d T , f o r s e q u e n c e see A p p e n d i x A ) . A p o r t i o n o f this first s t r a n d c D N A is t h e n a m p l i f i e d b y P C R u s i n g the a n c h o r o l i g o - d T p r i m e r i n c o n c e r t w i t h a n o l i g o n u c l e o t i d e d e s i g n e d to a 3'p o r t i o n o f the (5-galactosidase c D N A c o d i n g r e g i o n (P-galOl, f o r s e q u e n c e see A p p e n d i x A).  94  1  234567M89  Figure 28. PCR amplification of promoter D N A from transgenic nematodes. Nematodes transformed with the PC3 plasmid (see text for description) were assayed for the presence of the plasmid and the integrity of its promoter. The figure shows amplification of the hsp!6 promoter from as few as one transgenic nematode. Thirty-five cycles of PCR were performed as described in Figure 27, using the following substrates : 1,7: 5 pg of PC3 plasmid DNA; 2,3 : 200 ng and 20 ng of genomic D N A prepared from PC3 transgenic nematodes; 4,5,6 : five, three, and one adult PC3 transgenic nematode (s); 8 : 200 ng wild type C. elegans genomic DNA; 9 : five adult wild type C. elegans nematodes. 40% of the amplified D N A was then separated by electrophoresis on a 2,0% agarose gel. Amplified promoter D N A is indicated by an arrow. M : X  Hindm marker DNA.  95  promoter sequence. A n appropriate fragment was amplified when either PC3 plasmid D N A or genomic D N A isolated from PC3 nematodes (200 ng) was used as substrate, but not when wild type (N2) untransformed nematode genomic D N A was used. I have extended the utility of this method to the analysis of small numbers of transgenic animals. In lanes 4 to 6 of Figure 28 the D N A fragment represents amplification from 5, 3, and 1 adult transgenic nematode(s), respectively, which were boiled and then used directly as substrate for amplification. These amplified PC3 promoter D N A fragments were cloned and sequenced and corresponded to the expected promoter sequence from the microinjected plasmid. Oligonucleotides were also designed to allow the detection of the B-galactosidase reporter gene mRNA. Total R N A (1 ug) prepared from pD56 transgenic nematodes (the pD56 strain was a gift from Dennis Dixon, Johns Hopkins University; the pD56 plasmid bears the myosin light chain gene promoter, the E. coli B-galactosidase reporter gene, and the myosin heavy chain gene 3' untranslated region) was converted into first strand cDNA using M M L V reverse transcriptase and the anchor oligo-dT primer (for the sequence of anchor oligo-dT see Appendix A). A portion of this first strand cDNA was then amplified by PCR using the anchor oligo-dT primer in concert with B-galOl, an oligonucleotide derived from the 3' portion of the Bgalactosidase mRNA coding region (for the sequence of B-galOl see Appendix A). Amplification of cDNA prepared from "wild type" Bristol strain (N2) and pD56 transgenic nematodes is compared in Figure 29. The B-galactosidase-derived fragment is denoted by an arrow. The identity of this fragment was confirmed by D N A sequencing. In Figure 29 a second, non-transgene-derived fragment is observed due to cross-priming of an endogenous C. elegans gene which bears homology to the B-galOl oligonucleotide. The D N A sequence of this gene bears no apparent similarity to any sequence in the GeneBank Sequence Repository (data not shown).  96  1  2  h 2.0 kb  B-gal  cDNA  h 0.5 kb 0.1 kb  Figure 29. PCR amplification of reporter gene mRNA from transgenic nematodes.  Nematodes transformed with the PD56 plasmid (see text for description) were assayed for the expression of B-galactosidase reporter R N A under the control of the myosin light chain promoter. First strand cDNA prepared from wild type and transgenic total R N A was subjected to 35 cycles of PCR amplification as described in Figure 27. 40% of the amplified D N A was separated by electrophoresis on a 2.0% agarose gel. Lane 2 shows amplification of cDNA from the B-galactosidase reporter mRNA expressed in transgenic PC3 nematodes. Two amplified cDNA fragments are visible; the one derived from B-galactosidase is indicated by an arrow. The other amplified fragment is present in the control amplification from wild type C. elegans cDNA (lane 1). XFfindlTI size markers are indicated at the right.  97  3.13  PCR Amplification of Ubiquitin cDNAs with Degenerate Primers Due to the difficulty in obtaining non-UbiA ubiquitin clones from C. elegans  libraries, I decided to devise a PCR strategy for ubiquitin cloning, hopefully one which might provide a universal tool. PCR relies on the presence of unique oligonucleotide sequences in the gene of interest, or at least the presence of oligonucleotides which will define members of a set of related genes. Conceptually this is difficult for ubiquitin : ubiquitin is not only extremely highly conserved in D N A sequence but it tends to be found as part of a tandemly repeated gene structure. Two oligonucleotide primers which are specific for ubiquitin could prime in any pairwise combination of repeat units, ultimately generating a PCR product which is the lowest common denominator, i.e. a mixed sequence representing all of the ubiquitin repeats. In order to overcome this problem, only one ubiquitin-specific degenerate oligonucleotide is employed in combination with anchor oligo-dT primed cDNA as illustrated in Figure 30. The use of only one ubiquitin oligonucleotide primer gives polarity to the cDNA amplification. The amplified product represents the final ubiquitin repeat and the untranslated region of the represented locus, rather than a mixed sequence representing all repeats at any given locus. In practice this approach has been quite successful. Analysis of the ubiquitin sequence using a "Least Degenerate Oligonucleotide Sequence" program (PC GENE®, Intelligenetics Corp.) reveals two potential low degeneracy oligonucleotides : one at the immediate amino terminus of ubiquitin (MQIFVKT, ubl) and the other at amino acids 58 to 62 (DYNIQKE, ub2) (see Figure 31). Both regions are well conserved in evolution, and hence oligonucleotides from these regions prove useful for a wide range of species. Figure 32 shows that both ubl and ub2 oligonucleotides amplify truncated ubiquitins. Use of ubl allows recovery of the maximal amount of ubiquitin sequence. Amplifications using ubl or ub2 are presented in Figures 32 and 33.  98  Ub mRNfl  5'UT  Ub Primers  1 Mt>  I  I m+  I  3"UT  HHHHHHH <»TTTTI I . Anchor-dT  PCR Truncated Ub ds cDNA  Figure 30. Schematic illustration of the PCR strategy to amplify ubiquitin cDNAs. One ubiquitin-specific degenerate oligonucleotide is used in combination with anchor oligo-dT primed first strand cDNA as illustrated. The use of only one ubiquitin oligonucleotide primer gives polarity to the cDNA amplification and hence the final product represents a truncated "accurate" copy of the final ubiquitin repeat and the 3' untranslated region (3'-UT) of the represented locus, rather than a mixed sequence representing the coding sequence of all repeats at any given locus.  99  u b l ub2  u b l : 5' flTG Cfi_flT_TT_ GT_ flfl_ fiC 3' ub2 : 5' Gfl_ Tfl_ flfl_ HT_ Cfl_ flfl_ GR 3'  Figure 31. Degenerate ubiquitin oligonucleotides used for PCR cDNA amplification. Oligonucleotides ubl and ub2 are shown relative to their sites of priming in the last ubiquitin repeat (large filled arrow) of a ubiquitin cDNA. Also shown is the D N A sequence of the two oligonucleotides; degenerate positions in the triplet codons are underlined and were replaced by deoxyinosine in the oligonucleotide DNA.  100  S.c.  Cb.  ubZubl  ub2 ubl 2.0 kb 0.5 kb 0.1 kb  Figure 32. PCR amplification of ubiquitin cDNAs from C. briggsae and yeast. First strand cDNA prepared from C. briggsae or S. cerevisiae total cellular RNA using anchor oligo-dT was subjected to 30 cycles of PCR using anchor oligo-dT in combination with ubl or ub2 oligonucleotides (see Figure 31). Duplicate loadings of approximately 20% of the PCR amplified D N A was separated by electrophoresis on a 2.0% agarose gel (S.c.: Saccharomyces cerevisiae: C b . : Caenorhabditis briggsae). Amplified truncated ubiquitin cDNAs are indicated; their identity was confirmed by D N A sequencing (data not shown). Multiple amplified ubiquitin cDNAs are evident. As predicted, ubiquitin cDNAs amplified with ub2 migrated more slowly than those amplified with ubl due to the relative priming sites of these oligonucleotides in the ubiquitin coding repeat.  101  Ub cDNA — £  Figure 33. PCR amplification of ubiquitin cDNAs from other organisms. Ubiquitin cDNA amplification was performed as described in Figure 32, except that the ubl oligonucleotide was used exclusively. Duplicate amplifications from 1 : Ginko biloba RNA; 2 : Rhododendron RNA; 3 : Salmo gairdnerii RNA; and 4 : Drosophila melanogaster were analyzed on a 2.0% agarose gel. Amplified truncated ubiquitin cDNAs are indicated.  102  I have been able to re-isolate predicted truncated C. elegans UbiA cDNA clones as well as clones for a highly conserved UbiA homologue of C. briggsae. Alignment of the two UbiA cDNA sequences was shown in Figure 15. A class of non-UbiA cDNAs was also amplified from C. elegans and C. briggsae. These clones were similar to the yeast U b i l and Ubi2 genes (Ozkaynak et al., 1987); they contained an amino terminal ubiquitin sequence (i.e. at least one) and a basic carboxy terminal region of 52 amino acids. C. elegans and C. briggsae UbiB carboxy terminal moieties are compared to the published yeast UBI1 sequence in Figure 34. The nematode UbiB gene tail sequences differ in only two amino acids near the extreme carboxy terminus. The nematode tail sequences are also conserved relative to the yeast protein, differing in only 10 or 11 out of 52 amino acids. All four cysteines which may form a zinc finger in the yeast U b i l / U b i 2 tail protein are conserved, as is the stretch of lysines at amino acids 48 to 52 which are thought to be a nuclear localization signal (Ozkaynak et al., 1987). The 3' untranslated regions of the nematode UbiB genes are also highly similar and contain a polyadenylation signal eleven nucleotides upstream of the polyA site. It is interesting to note that the residue of UbiB which is polyadenylated differs for the two nematode species. This approach proved so successful with nematodes that I attempted to use it to isolate truncated ubiquitin cDNAs from members of the animal and plant kingdom for which cDNA libraries would not necessarily be readily available. These oligonucleotides have allowed the amplification of truncated ubiquitin cDNAs from Drosophila, Ginko, Rhododendron, Equisetum, trout, and newt (representative amplifications are shown in Figures 33). Amino acid translations from partial cDNA sequences of these clones are aligned with yeast Ubi3 and human UbCEP80 in Figure 35. The yeast Ubi3 gene tail  103  Tail Amino Acid Sequence Comparison C.e. Cb. S.c.  RGGI IEPSLiP<2LAQK^  3' Untranslated DNA Sequence Comparison C.e. : TAA TCTTCACTCGTTCCAGTCCA^ Cb. : GT CCTTC—T C—C  •  •  (A) * AAC (A) *  Figure 34. Amino Acid and D N A sequence of the UbiB cDNAs of C. elegans and C. briggsae. Three independent subclones of PCR amplified ubiquitin cDNA were sequenced for each nematode. Upper Panel : Comparison of the 52 amino acid tail sequences from the UbiB genes of the nematodes C. elegansCC.e.) and C. briggsae(C.b-) and the ubil gene(S.c) of the yeast S. cerevisiae (Ozkaynak et al., 1987). The tail sequences begin with the last three amino acids of the ubiquitin moiety (RGG). Standard one letter code is used to represent amino acids. Conserved positions are indicated by dashes (-). Lower Panel: Comparison of the 3' untranslated regions from C. elegans and C. briggsae UbiB cDNAs. The D N A sequence shown is from ochre stop codon (TAA) to the nucleotide which is polyadenylated (* = polyA tail; the first A residue may be encoded by the ubiquitin gene or may be part of the polyA tail). Conserved positions are indicated by dashes (-). The putative polyadenylation signal is underlined. The nudeotide sequence of the UbiB tail regions is presented in Appendix D.  104  S.c. H.s.  : :  RLRGG GKKRKK KVYTT PKKIK HKHKK VKLAV LSYYK VDAEG KVTKL K EN- -ISRA SN - —R K K K  EN- -IHREN IHREN- -IHR..  A A A A A A  S.C. : H.s. :  RRECS NPTCG AGVFL ANHKD RLYCG KCHSV YKVNA * P SDE M -S-F- -H CCLT YCF-K PGDK  D.m. : Ginko:  N-S S S-A T—K S S  —R N - —R N - —R  D.m. : Ginko: Rhodo: Equi . Trout: Newt :  P GEN P SEE-  INT- —R N - —K-  M -A-E- -H  NLT FVFSK PEEK  Figure 35. Tail amino add sequences from PCR-amplified yeast ubi3-type cDNAs. Two independent subclones of PCR amplified ubiquitin cDNA were sequenced for each organism. If the D N A sequences were not identical, a third cDNA was sequenced and the translation of the two identical sequences is reported. The cDNA tail sequences presented begin with the last five amino acids of the ubiquitin moiety (RLRGG) and are aligned vertically. Abbreviations for organisms are as follows : S.c. : Saccharomyces cerevisiae (the ubi3 gene from Ozkaynak et al., 1987); H.s.: Homo sapiens (the UbCEP80 gene from Lund et al., 1985); D.m.: Drosophila melanogaster; Ginko : Ginko biloba: Rhodo : Rhododendron sp.: Equi: Equisetum sp.; Trout: Salmo gairdnerii; Newt: Cynops pyrogaster. Standard one letter code is used to represent amino acids. Conserved amino adds are indicated by dashes (-); * indicates that the yeast sequence open reading frame stops at amino add 76; dotted lines indicate the end of partial cDNA sequence translations. R N A for PCR from trout and newt were kindly provided by David Banfield (University of British Columbia).  105  encodes 76 amino acids, while human and Drosophila encode 80 amino acids. All four cysteines which have been suggested to form a zinc finger in yeast Ubi3 are conserved, as is the stretch of lysines at amino acids 2 to 7 which are thought to be a nuclear localization signal (Ozkaynak et al., 1987). The amplified cDNAs appear to be more closely related to the human gene than the yeast gene; a comparison with the Drosophila 80 amino acid tail sequence shows 62 of the first 76 tail amino acids are identical to the human gene, while only 49 of 76 are identical to the yeast gene. This technique thus represents an efficient means of obtaining ubiquitin cDNA sequence from almost any eukaryote.  3.14  Amplification of Trans Spliced cDNAs from C. elegans I have attempted to develop a rapid method for the characterization of trans  spliced cDNAs from C. elegans and have devised a PCR-based strategy for the simultaneous amplification of all C. elegans trans-spliced cDNAs. Preliminary analyses of these cDNAs were undertaken and the results are presented below. Figure 36 shows a schematic illustration of the strategy used to amplify trans spliced C. elegans cDNAs. The substrate for amplification is first strand cDNA which has been synthesized using the anchor oligo-dT. Utilizing an oligonucleotide designed to the C. elegans SL1 splice leader sequence in place of a gene-specific oligonucleotide like ubl (as described previously), I have been able to amplify full length cDNAs which represent trans spliced mRNAs (see left panel of Figure 37). Amplified trans spliced cDNA has been analyzed in two ways : (1) The amplified cDNA was transferred to nitrocellulose and then hybridized with a coding region sequence from C. elegans actin, a known trans spliced gene. From the right panel of Figure 37, one can see a hybridization signal at approximately 1.5 kb, the size of the mRNA(s) for the trans spliced actin genes predicted from their published sequence  106  B  1 •c  Reverse Transcribe with TTTTT [  PCR with  TTTTT mnnil oligo dT primer and Splice Leader primer  Electrophoresis  Southern Hybridization  II in 1111A  Hnnnn B IIIIIIII  C  1  Subclone or directly sequence individual trans-spliced cDNAs Figure 36. A PCR strategy for the amplification of trans spliced cDNAs from C. elegans. First strand cDNA synthesis is primed with a anchor oligo-dT. The anchor sequence on the oligo-dT increases the hybridization temperature of this oligonucleotide primer for subsequent PCR amplifications. PCR is performed using the anchor oligo-dT and an oligonucleotide with the sequence of the C. elegans splice leader SL1. A n aliquot of the PCR product is separated by electrophoresis and may be blotted to confirm the presence of known trans-spliced genes. Individual or all trans spliced cDNAs may then be subcloned or directly sequenced.  107  M  12  Figure 37. Southern hybridization analysis of PCR-amplified trans spliced cDNAs. Trans spliced cDNAs were amplified from C. elegans using the method described in Figure 36. 40% of the amplified cDNA was analyzed on a 1.2% agarose gel, then transferred to a nylon membrane for hybridization with a radiolabelled C. elegans actin fragment (kindly provided by M . Krause, Fred Hutchinson Cancer Research Centre). The left panel shows the ethidium bromide stained agarose gel of the PCR amplified products; M : X HindUI markers; 1 : control amplification with SL oligonucleotide alone; 2 : amplification with SL oligonucleotide and anchor oligodT. The right panel shows Southern hybridization of the amplified splice leader cDNA with a radiolabelled C. elegans actin fragment. Note that the size of the hybridizing band (indicated by an arrow) is that predicted for the actin gene (Krause and Hirsh, 1987).  108  (Krause and Hirsh, 1987)  (2) Size-selected, re-amplified D N A from one of the  prominent bands was cloned into the pBluescript KSII plasmid and subjected to partial sequence analysis. The partial cDNA sequence and protein translation from the sequenced region is presented in Figure 38. The splice leader sequence used for amplification is underlined. No similarity was found between the open reading frame encoded by this cDNA and any of the known proteins in the SwissProt protein database or between the D N A sequence of this clone and any D N A sequence present in the Genebank database.  109  SLl M L D I I N Y GGTTTAATTACQZAAGTTTGAG C&GATTCAATC ATG CTT GAT ATC ATT AAC TAG  L L P F V V N Y S R C Q Y L N H CTT TTA CGC TTT GTT GTT AAT TAG AGC CGG TGC CAG TAT CTG AAC CAC  Q H Q N P R K I S G K N T A F K CAG CAC CAA AAC CCA CGT AAG ATC AGC GGC AAG AAT ACC GCC TTT AAA  N P Y A T T S N T N I A R T L K AAC CCA TAC GCC ACC ACC AGT AAC ACC AAC ATC GCC AGA ACT TTG AAA  L R Q I R I S T T R P F A G V C CTT CGC CAA ATC GGA ATT TCC ACC ACC CGG CCC TTC GCA GGT GTA TGC  A  C  R  GCC TGT GGA  Figure 38. Partial sequence of a PCR-amplified trans spliced cDNA. Size-selected, re-amplified D N A from one of the prominent splice leader cDNA fragments shown in Figure 37 was cloned and subjected to partial D N A sequencing. The partial D N A sequence is shown, with the splice leader oligonucleotide sequence (SLl) underlined. The amino acid translation of this sequence is presented above the D N A sequence.  110  IV.  DISCUSSION  Ubiquitin is the most conserved eukaryotic protein known, surpassing even the conservation of histone H4 (Sharp and Li, 1987a,b). Out of 76 amino acid residues, 68 are invariant in all sequenced ubiquitins. Yeast, plant and animal ubiquitins, representing over 3 billion years of evolution, differ in only 4 of 76 amino acid residues. This high degree of sequence conservation suggests that there is selection pressure from some source on virtually every amino acid of ubiquitin to maintain either the proper folding, stability, or function of ubiquitin. The ubiquitindependent proteolytic pathway involves at least five major classes of enzymes which all may have a binding site for ubiquitin. That ubiquitin interacts with so many other proteins may be one source of conservation pressure; a substitution that is harmless to structural interactions of ubiquitin with one enzyme may disrupt the interactions between ubiquitin and a different enzyme. A substitution which does not affect the function of ubiquitin as a cofactor in selective proteolysis may be detrimental to functions in the nucleosome or on cell surface receptors. Alternatively, the conservation pressure may be associated with some physical property of ubiquitin required for its diverse functions e.g. the ability to undergo conformational changes. The ubiquitin encoded by the C. elegans ubiquitin genes UbiA and UbiB most closely resembles that of human ubiquitin, differing in only one amino acid residue, namely human proline-19 is alanine in C. elegans ubiquitin. C . elegans has three amino acid substitutions relative to yeast at positions 19, 24, and 28. As with other organisms , there is no intragenic or intergenic variation in amino acid sequence encoded within the nematode polyubiquitin gene, although there is some divergence at the level of codon usage. It is interesting to note that not all of the amino acid sequence substitutions in C. elegans ubiquitin are conservative ones.  Ill The four sites of arruho acid sequence variation in the yeast, plant, human, and nematode forms of ubiquitin are clustered in a region of ubiquitin's surface almost directly opposite the carboxy terminal conjugation site. These variations are likely tolerated rather than being functionally important to the structure or folding of the protein; a recent comparison of yeast and human ubiquitins revealed no major differences in biochemical and physical properties (Wilkinson et al., 1986). However, it has been speculated by Ecker et al. (1987) that these four amino acid positions, which are spatially clustered (see Figure 1), may play a role in speciesspecific non-proteolytic functions.  4.1 Polyubiquitin Gene Structure The number of repeats per polyubiquitin locus varies considerably among and also apparently within species (see Figure 3), suggesting that these loci frequently engage in unequal crossover events (e.g. Arribas et al., 1986; Baker and Board, 1989). Such variation has also been found within the ubiquitin hybrid genes : although typically the ubiquitin genes that contain tail sequences have only a single ubiquitin coding element, a gene recently isolated from T cruzi contains a tail sequence similar to that of C. elegans UbiB, joined to a stretch of approximately 50 contiguous ubiquitin coding elements (Swindle et al., 1988). Studies in yeast demonstrate that even a monomeric unit placed within the context of polyubiquitin flanking sequences can adequately fulfill the role of the polyubiquitin gene (Finley et al., 1987). The polyubiquitin organization must confer some advantage to the cell:  (a) the multicopy arrangement may prevent the  indiscriminant attachment of ubiquitin to cellular proteins following its synthesis, and thus allow for regulation of ubiquitin conjugation or (b) alternatively, but not exclusively, the polyubiquitin arrangement may have evolved in order to minimize  112  the metabolic cost associated with producing large quantities of free ubiquitin monomer. Most polyubiquitin variants in higher eukaryotes also contain a single extra amino acid residue at the end of their last ubiquitin repeat, the extra residue being different in polyubiquitin variants from different species (see Figure 3). By blocking the carboxy terminus of the preceding Gly residue, the extra carboxy terminal residue could serve to prevent participation of unprocessed polyubiquitin in ubiquitin-protein conjugation. It should be noted that Xenopus polyubiquitin has no extra carboxy terminal amino acid and may require a different mechanism to prevent polyubiquitin conjugation.  Intraspecies diversity of carboxy terminal  extensions e.g. human UbC vs UbB (Wiborg et al., 1985; Baker and Board, 1987) might allow selective activation of these polyprotein precursors based on differential cleavage of the carboxy terminal blocking residue. The nematode polyubiquitin gene (UbiA) consists of eleven tandem copies of the ubiquitin-encoding repeat; this structure allows a large yield of free ubiquitin per transcription event. C. elegans and C. briggsae UbiA genes possess not one, but two additional amino acids at their carboxy terminus. Our Northern hybridization and Genomic Southern hybridization studies have indicated that developmental or recombinational polymorphism is not pronounced in C . elegans. despite the length of the gene. As well, restriction analysis indicates that the polyubiquitin gene structure is conserved in another nematode, Steineremia feltiae. This is in contrast to Drosophila (Arribas et al., 1986) and humans (Baker and Board, 1989), where recombination frequently generates new ubiquitin phenotypes.  4.2 Nematode Ubiquitin Genes C. elegans possesses two ubiquitin genes, UbiA and UbiB. UbiA is not highly expressed, being only 0.003% of mRNAs in C. elegans. C. elegans and C. briggsae  113 UbiA and UbiB cDNAs obtained by PCR have similar sequence and structure. The UbiB mRNA is derived from a ubiquitin locus distinct from U b i A : sequencing of UbiB cDNA shows that it is a fusion gene similar to the yeast Ubil,2 genes. There is no evidence for monomeric ubiquitin gene in C. elegans; this indicates that all free ubiquitin monomer in C. elegans is probably derived from the cleavage of the ubiquitin fusion proteins produced by UbiA and UbiB.  4.3 Introns and Evolution 4.3.1 Coding Region cis Introns The lack of spacer sequences between tandem repeats is a common feature of polyubiquitin genes. To date, however, no reported sequence of a polyubiquitin gene has contained coding region introns. Intragenic introns are not completely absent in the ubiquitin gene family : an intron occurs in two of the yeast monomeric ubiquitin fusion genes, interrupting the ubiquitin moiety within the isoleucine codon at position 3 : A T / C or T (Ozkaynak et al., 1987) and an intron is found within the 5' noncoding sequence of the chicken polyubiquitin (Bond and Schlesinger, 1986) and human tri-ubiquitin genes (Baker and Board, 1987) (and potentially the human nine repeat ubiquitin gene : Wiborg et al., 1985). In each of these examples the introns are in analogous positions but are not part of the tandemly repeated sequence at that locus and lack sequence similarity. The lack of coding region introns for polyubiquitin genes might be explained by postulating (a) recombinational deletion of introns by a mechanism which compares tandem repeat units and (b) selection pressure- genes crucial to metabolic survival under conditions of environmental stress (e.g. heat shock genes) typically lack introns in order to bypass stress-sensitive splicing reactions and to minimize the metabolic cost of producing stress proteins.  114  The nematode UbiA gene thus represents a unique and important exception to other published ubiquitin genes. In addition to containing an unusual leader intron which is acquired by a trans splicing mechanism (see below), UbiA also contains 4 coding region cis introns. These coding region introns were identified by (a) the presence of flanking splice site consensus sequences  (b) interruption of the  tandemly repeated open reading frame and (c) partial D N A sequencing of UbiA cDNA. This gene structure appears to be conserved and recombinationally stable in C. elegans strains based on the lack of polymorphisms for a number of restriction enzymes that cut within and around the polyubiquitin locus (see Figure 10). Thus, in contrast to other polyubiquitin genes, the synthesis of ubiquitin via the UbiA gene is intimately associated with proper splicing, both cis and trans. The coding region introns of UbiA are found in identical locations within ubiquitin repeats l,4,7,and 10, interrupting the glycine-47 codon G / G A . The introns are all approximately 50 bp in length, typical of the short introns which predominate in C. elegans genes. Most C. elegans 50 bp introns display only splice site similarity (there is no strong lariat consensus sequence apparent for nematode introns (Blumenthal and Thomas, 1988). When the UbiA cis intron sequences are aligned, a degree of sequence similarity can be seen; however, the nature of intron sequence is to diverge much more rapidly than protein coding sequence. The conservation of the position of the introns within the tandemly repeated coding sequences strongly suggests that the C. elegans polyubiquitin gene evolved by a series of duplication and crossing over events. The repeating +— structure implies that the duplicating cassette may have been a +- dimer or a +— trimer. However, it should be noted that codon structure in the intron-containing versus intron-free ubiquitin repeats is not substantially different, arguing against this model. It can be assumed that the UbiA eleven repeat structure arose from an original monomeric ubiquitin gene which possessed an intron or acquired an intron after  115 duplication. If the former, this implies that all other polyubiquitin genes have lost their introns; if the latter, this implies that C. elegans acquired an intron in the ubiquitin gene early in its evolution and that it maintained it (or was unable to eliminate it). Recent evidence suggests that introns existed in primitive genes and have since been deleted in many cases, notably in prokaryotes (Doolittle, 1978; Stone et al., 1985; Gilbert et al., 1986; Lambowitz, 1989). In this interpretation, C. elegans has been able to shorten its ubiquitin introns but has been unable to completely eliminate them. This situation has analogy to other nematode genes, which possess introns while their evolutionary homologs in other species typically do not. Examples of this include : (1) the hsp!6 genes of C. elegans possess an approximately 50 bp cis spliced intron (Russnak and Candido, 1985) and (2) the hsp-1 hsp70 gene of C. elegans possess both cis and trans spliced introns (Snutch et al., 1988; Bektesh et al., 1988).  4.3.2 Trans Splicing of UbiA I have demonstrated that hnRNA from the nematode polyubiquitin gene is one of the few documented RNAs from a higher eukaryote to undergo a trans splicing reaction similar to that found in trypanosomes. A 22 nucleotide mini-exon encoded by a 5S ribosomal repeat on chromosome V is trans spliced onto the UbiA hnRNA from chromosome 111 via a 3' splice site in the UbiA 5' noncoding leader sequence. This same splice leader sequence is trans spliced onto three of the four C. elegans actin genes (Krause and Hirsh, 1987). Figure 39 presents a model for UbiA mRNA maturation. The start site of transcription for the UbiA gene was mapped by SI nuclease protection to a position 450 nucleotides upstream of the A T G of the first ubiquitin repeat. This indicates that the first 450 nucleotides of the UbiA hnRNA are not included in the mature mRNA. The 450 base 5' intron of the UbiA gene is quite  116  Promoter RNA  II I Start C  I  v  —I  Site E  3Splica  ATG  Stop  PolyA  I  •'—r  U b i A Gene  Transcription  S p l i c e Leader (SL) hnRNA  Start S i t * Probe  UbIA  Figure 39. A model for UbiA polyubiquitin R N A maturation  mRNA  in C. elegans.  The primary transcript of the UbiA gene on chromosome i n includes a leader intron which is replaced by a 22 nucleotide mini-exon derived from the splice leader R N A gene S L l on chromosome V. The single-stranded D N A probe used for determination of the UbiA transcript initiation site is indicated, as is the location of the oligonucleotide used for primer extension sequencing of the UbiA mRNA. Major features of the UbiA gene are indicated. Restriction sites flanking SI nuclease probes are indicated by single letters: X : Xbal; C : Clal; E : EcoRI.  117  long for a nematode intron; over 80% of the known nematode cis introns are between 48 and 52 bases in length (Blumenthal and Thomas, 1988). It is tempting to speculate that there may be important sequence information to guide the trans splicing apparatus within the 5' intron. One may speculate that the UbiA transcript leader intron may at one time have possessed a 5' splice site and hence have been cis spliced; loss of the 5' splice site would then have resulted in the use of the trans splicing apparatus. Particularly intriguing in the case of the UbiA transcript is the finding that other species have a ubiquitin gene containing a conventional, cis spliced intron at its 5' end, in a position analogous to that of the trans spliced intron of the C elegans gene. It is interesting to note that the C. elegans UbiA trans-spliced intron is located at a position conserved in polyubiquitin genes from other species. Both the chicken Ubil (715 bp : Bond and Schlesinger, 1986) and human UbiB (654 bp : Baker and Board, 1987) polyubiquitin genes have cis spliced introns in the same position of the 5' untranslated sequence. The human UbC polyubiquitin gene has a 3' splice site in an analogous position and hence may also have an intron in its leader. It is difficult to speculate on a reason for the existence of a conserved leader intron in the polyubiquitin gene family since coding region introns are rare. It is possible that leader introns are remnants of the original gene structure; yeast Ubi4 may have lost its leader intron due to the high recombination rate found in yeast (reviewed in Fink, 1987). C. elegans may be in the process of losing this intron since it only has one half of the necessary information for proper cis splicing (i.e. the 3' splice site). The existence of trans splicing in C elegans immediately raises a number of questions concerning the evolutionary origins and cellular function(s) of this process. Besides the three actin genes and the UbiA gene, other transcripts in C. elegans are known to contain the spliced leader sequence, as shown by Northern blot analysis using the 22 nucleotide leader sequence as a probe (Bektesh et al., 1988).  118 The UbiA gene has recently been mapped to chromosome ITJ in C elegans, whereas the actin genes acquiring the trans spliced leader are on chromosome V (Krause and Hirsh, 1986), as is the splice leader R N A coding sequence (Nelson and Honda 1985, 1986) . Thus it is likely that the leader can be spliced to transcripts derived from unlinked regions throughout the Q elegans genome. There is no similarity between the C. elegans SLl D N A sequence and the trypanosome SL sequence. Also, there is no apparent similarity between the transspliced leader of the C. elegans UbiA transcript and the cis spliced leader of the chicken Ubil mRNA. Thus perhaps it is not the sequence of the leader itself, but rather the splicing event or the mere possession of a leader (whether acquired by a cis or a trans mechanism) that serves some physiological function for ubiquitin metabolism within cells. It is not known why only some C. elegans transcripts undergo trans splicing, or what signals direct this process. However, a comparison between the 100 nucleotide splice leader R N A and the UbiA 5' sequence yields no strong similarities except between the 5' mini exon of the leader R N A and the 3' splice consensus at the UbiA 5' end. It remains to be determined whether this similarity is relevant to the mechanism of trans splicing. In trypanosomes, cis splicing has not been detected, and it has been suggested that the co-existence of cis and trans splicing mechanisms competing for the same splice acceptor sites might be intolerable within the same cell (Braun, 1986; Sharp, 1987) . Clearly the two processes can co-exist not only within the same cell, but within the same transcript in C. elegans. The C. elegans UbiA gene has a trans spliced leader at the 5' end of its primary transcript and four coding region ubiquitin repeats which contain a single short intron which appears to undergo conventional cis splicing. Thus, the UbiA transcript must undergo both trans and cis splicing in order to yield a mature mRNA. This raises the question of how the cellular splicing machinery distinguishes between these two processes.  119  It is believed that trypanosomes are very ancient eukaryotic cells which diverged from other lineages before the separation of plants, animals and fungi. The existence of only trans splicing in these cells suggests that this process may pre-date, and may indeed have been the precursor of, cis splicing. C. elegans is the first known example of a eukaryote which possesses both processes in the nucleus. The finding of both trans and cis splicing within the same organism provides an opportunity to investigate the mechanism and functional role of these processes within cells.  4.4 Potential Regulatory Sequences in the 5' flanking region of U b i A The site of transcript initiation for the UbiA gene was mapped to a region 450 bp upstream of the A T G of the first ubiquitin repeat by SI nuclease protection experiments. This is an important step as the transcriptional start site delimits the promoter region for most eukaryotic RNA polymerase II transcribed genes. The major UbiA promoter elements are thus likely to be located immediately upstream of this region. In all previous examples of trans spliced mRNAs from trypanosomes and C. elegans transcription start sites have been arbitrarily assigned based on similarity to known start sequence contexts or by assuming transcription initiates 3035 nucleotide downstream from the nearest consensus T A T A sequence. No consensus T A T A motif is evident near the UbiA transcription start site; the T A T A box influences the accuracy of transcript initiation in eukaryotes and is almost always found 20 to 35 nucleotides upstream of the mRNA cap site (Goldberg, 1979).  A derivative of the T A T A motif, G A A T A A , is found approximately 30  nucleotides upstream of the UbiA hnRNA start site. This inexact T A T A sequence may account for the inexactness of UbiA transcription initiation: minor alternative sites of hnRNA initiation can be detected by SI analysis (data not shown). It should be noted that not all genes rely on a T A T A sequence e.g. human transforming  120 growth factor alpha and human CD3-epsilon gene lack T A T A sequences (Jakobovits et al., 1988; Clevers et al., 1988) and human chorionic somatomammotropin-1 uses C A T A A 10% of time instead of T A T A (Tanaka et al., 1988). It has recently been demonstrated that nuclear factor ETF specifically stimulates expression from the EGF receptor gene which doesn't have T A T A and will recognize polydC, polydG, and various GC-rich sequences (Kageyama et al., 1989). The sequence C T C C at -280 from the transcription start site of UbiA likely 1 5  represents a functionally important regulatory sequence; overall, C. elegans D N A is 67% A / T rich so the probability of random occurrence of such a sequence is low. A similar sequence exists upstream of the mammalian B-globin gene and is DNasel hypersensitive. The globin sequence may be a site for binding of an SPl-like transcription factor termed BGP1 (Lewis et al., 1988). Transcription factor SP1 has been postulated to regulate transcription from constitutively expressed (housekeeping) genes and to have a binding preference for GC-rich D N A (Letovsky and Dynan, 1989). At positions -190 and -360 relative to the hnRNA start site are sequences which resemble the heat shock element (HSE) C - G A A - T T C - G defined by Pelham (1982). This element is found upstream of most eukaryotic heat shock genes and has been shown to confer heat inducibility both in vitro and in vivo. The sequence 190 nucleotides upstream of the start site matches the HSE consensus sequence at 12 of 14 positions but lacks dyad symmetry and is hence likely nonfunctional. The sequence 360 nucleotides upstream of the transcription start site matches the HSE consensus at 13 of 14 positions but is quite distant from the UbiA transcription start site. Functional HSE sequences have been noted for other polyubiquitin genes e.g. yeast Ubi4 has an HSE element approximately 250 nucleotides upstream of its assumed transcriptional start site (Ozkaynak et al., 1987) and the chicken UbI gene has two overlapping HSE sequences approximately 360 nucleotides upstream from  121  the polyubiquitin start site (Bond and Schlesinger, 1986). Polyubiquitin R N A levels are increased up to five-fold by heat shock in these organisms.  4.5 Ubiquitin Expression 4.5.1 Ubiquitin Expression during Development The expression of polyubiquitin and ubiquitin hybrid genes is under developmental and metabolic control in yeast. The yeast genes U b i l , Ubi2, and Ubi3 are all expressed in exponentially growing cells, while the U b i l and Ubi2 genes are turned off when cells are heat stressed or starved. Ubi3, unlike most yeast genes, is apparently expressed under all of these conditions (Finley et al., 1989). A third pattern of regulation is shown by the yeast Ubi4 gene, which is expressed at low levels in exponentially growing cells but is strongly induced by heat shock, starvation, and other stresses (Finley et al., 1987). The importance of ubiquitin in development has been elegantly demonstrated by a genetic analysis of the yeast Ubi4 polyubiquitin gene. Deletion studies have revealed a requirement for a functional polyubiquitin locus to ensure proper spore development and viability (Finley et al., 1987). Possible roles for ubiquitin during development include the proteolysis of transient developmental structures and gene activation via ubiquitin conjugation to histones. Ubiquitin-dependent proteolysis has been postulated to play important roles in cytoplasmic reorganization during red blood cell maturation (Rapoport, 1981) and eye lens development (Jahngen et al., 1986). Developmental induction of one or more ubiquitin mRNA (s) has been documented in many organisms : (1) in barley leaf tips, the level of small ubiquitin mRNAs is highest in the least differentiated tissues (Gausing and Barkardottir, 1986)  (2) Dictvostelium penta-ubiquitin gene expression  is low throughout the life cycle but rises dramatically in spores and germinating spores (Giorda and Ennis, 1987) (3) Xenopus ubiquitin mRNA levels are highest in  122  unfertilized eggs and early embryos and decrease subsequently; certain ubiquitin mRNAs are also tissue-specific (Dworkin-Rastl et ah, 1984) (4) Drosophila polyubiquitin is present in a chromosomal region (63F) that displays salivary chromosome ecdysone-induced puffs in early development (Arribas et al., 1986) (5) chicken Ubil! , initially thought to be functional but inactive, may be expressed during spermiogenesis (Mezquita et al., 1987). I investigated whether a transient rise in UbiA mRNA occurred with the development of C. elegans larvae. Larval development of C. elegans proceeds from embryos through four larval stages and into the adult hermaphrodite or male stages. With adequate food throughout the life cycle and at 20°C one nematode generation takes approximately four days. Both hermaphrodites and males live for approximately seventeen days after reaching adulthood. R N A was prepared from each of the C. elegans life cycle stages and ubiquitin expression was assayed by Northern hybridization. It was found that the level of both ubiquitin mRNAs remained relatively constant throughout development, suggesting that nematode ubiquitin is constitutively expressed, unlike the ubiquitin genes in other organisms. It has been demonstrated that yeast Ubi4 mRNA levels respond to the nutritional status of the cell, increasing five-fold in stationary phase (starved) cells relative to exponentially growing vegetative cells (Finley et al., 1987). When the food supply is limited early in development, C. elegans can take an alternative developmental pathway at the L2/L3 moult to produce the dauerlarva, a specialized L3 stage that does not feed, is resistant to desiccation, and can survive for up to three months without further development (Cassada and Russell, 1975). If food becomes available during this period, the dauerlarva moults to become an L4, which resumes normal development. Dauerlarvae do not undergo a true dormancy; rather than becoming quiescent, this stage swims rapidly and searches for food. Gene expression is streamlined to maximize survival. UbiA expression was investigated in  123  dauerlarvae, on the hypothesis that levels of UbiA mRNA might be induced by the stressful environment that leads to dauerlarva formation and the likelihood that abnormal proteins might accumulate in the absence of nutrition. There was no apparent change in UbiA mRNA level, even upon heat shock of dauerlarvae. Thus dauerlarvae must have an adequate pool of free ubiquitin to fulfill their metabolic requirements. The four C. elegans larval stages are punctuated by moults. New cuticle is synthesized under the old, and pharyngeal pumping ceases during a brief period called lethargus, while the old cuticle is shed. Synchronized L l larvae were sampled prior to pre-moult lethargus, during lethargus, through shedding of the cuticle, and into L2. Ubiquitin transcription was assayed during the larval moulting period. It was reasoned that moulting might represent a period of increased turnover of intracellular structures and regulatory proteins, which in turn might lead to an increased demand for ubiquitin-mediated intracellular proteolysis. Furthermore, the sequence with similarity to a steroid hormone receptor binding site upstream of the U b i A transcriptional start could conceivably play a role in moulting. We could see only slight fluctuations in UbiA mRNA levels with no correlation to the onset of moulting; the level of the small ubiquitin mRNA remained unchanged as well. Several interpretations of this result are possible. It is possible that newly induced R N A synthesis might be masked by the relatively large pre-existing pool of ubiquitin mRNA. Alternatively, a transient requirement for more ubiquitin might be met by changes at the translational or post-translational levels. The hormone receptor binding sequence in the UbiA promoter may or may not be functional; it does appear, however, to be insensitive to changes associated with moulting.  124  4.5.2 Heat Shock and Ubiquitin There are two HSE sequences upstream of the C. elegans UbiA gene at -190 and -360 relative to the transcriptional start site. A kinetic analysis was undertaken to look for changes in ubiquitin mRNA synthesis upon temperature elevation. The level of UbiA mRNA, as measured by hybridization to a UbiA coding region probe, fluctuates somewhat during a two hour heat shock and subsequent recovery (Figure 23 and data not shown); however, no parallel increase or decrease is observed for U b i A in relation to temperature. UbiA mRNA expression is not induced if nematode embryos were rapidly raised to any temperature between 30°C and 34°C and R N A was prepared after 30 to 60 minutes at that temperature. The expression of UbiB mRNA is also unaffected by chronic or acute heat stress (data not shown). The levels of UbiA mRNA and of the smaller ubiquitin mRNA do not increase significantly during chronic or acute heat stress. R N A polymerase IT transcriptional run-off experiments with isolated C. elegans nuclei confirm that C. elegans UbiA is not induced by heat shock (Jones et al., 1989). The two HSE sequences in the UbiA 5' end thus appear to be weak or nonfunctional, possibly due to their distance from the transcript initiation site. It is possible that a more sensitive assay of de novo transcription might allow a better estimation of changes in the transcriptional status of the nematode UbiA gene. However, there must be only subtle differences in mRNA synthesis if they are masked by pre-existing UbiA mRNA levels. The fact that no change in transcription of the UbiA gene is detected by Northern and dot hybridization suggests (a) that UbiA mRNA is relatively stable and present at sufficient concentration to accommodate temporary changes in requirement for free ubiquitin and (b) that the concentration of free ubiquitin is regulated at another level, for example trans splicing or post-transcriptional processing. It may be equally plausible to suggest that it is not the level or activity of ubiquitin that is important to  125  regulation of the heat shock response, but rather the level or activity of a factor which interacts with ubiquitin, for instance ubiquitin activating enzyme. It is possible that the requirement for ubiquitin is regulated at the level of the free ubiquitin protein pool in a cell, and as such chicken cells with four ubiquitin repeats per polyubiquitin mRNA, and yeast with five, might require more mRNA to replenish depleted levels of free ubiquitin during heat shock than does C. elegans which contains eleven ubiquitin repeats per polyubiquitin mRNA. Nematode cells might be able to increase the level of free ubiquitin monomer sufficiently by translational or post-translational means (e.g. by regulating the level or activity of a ubiquitin-specific hydrolase). It should also be noted that the Drosophila polyubiquitin gene (15 repeat units : Lee et al., 1988; Arribas et al., 1986) and the Dictyostelium polyubiquitin gene (5 repeat units : Giorda and Ennis, 1987) both lack HSE sequences and thermal induction of transcription. Evidence contrary to this interpretation does exist; for example, the hypersensitivity to chronic heat stress displayed by yeast Ubi4 mutants is complemented by a ubiquitin minigene containing only one copy of ubiquitin carried on a single copy, centromerecontaining plasmid (Finley et al., 1987). Due to the involvement of the ubiquitin system in the normal turnover of short-lived (regulatory) proteins and the above evidence linking ubiquitin to the heat shock response, Munro and Pelham (1985) proposed that the the level of ubiquitin conjugation to the eukaryotic heat shock transcription factor (HSTF) may control HSTF activity via selective proteolysis. Recent experiments with purified yeast and Drosophila HSTF preparations indicate that HSTF may undergo phosphorylation, but there is no evidence for direct ubiquitination (Sorger and Pelham, 1988; Sorger and Nelson, 1989; Perisic et al., 1989). The interaction of ubiquitin with the heat shock pathway is thus more complex than originally assumed; alternative substrates within the heat shock regulatory pathway could be  126  regulated by ubiquitin attachment, for instance a HSTF kinase.  The lack of  universal heat induction for polyubiquitin genes could imply organismal variation in the regulation of the ubiquitin degradation pathway : one can envision thermal regulation of the synthesis or activity not only of ubiquitin, but potentially of E l , E2, or E3 enzymes. How the signal of increased protein denaturation is translated into increased proteolytic activity remains an unanswered question. The true regulator of ubiquitin metabolism in the cell may be the amount of denatured protein present which requires degradation by ubiquitin-mediated proteolysis; heat shock and the developmental regulation of transcription may simply give rise to temporary periods of increased proteolysis.  4.6 Heat Shock and Trans Splicing During the course of heat shock experiments an increase in the relative abundance of unspliced hnRNA from the UbiA locus was observed under heat shock conditions. These results imply that trans splicing, like cis splicing (in Drosophila : Yost and Lindquist, 1986; in mouse cultured cells : Kay et. al, 1987), is inhibited by heat shock in C. elegans. This conclusion is supported by a recent publication showing that trans splicing is inhibited by heat shock in Trypanosomes (Munich and Boothroyd, 1989). It is likely that trans splicing, if not a specialized form of cis splicing, is likely to be at least as complicated and to involve a number of temperature sensitive reactions.  4.7 Analysis of the UbiA Promoter in Transgenic Nematodes In order to study the in vivo regulation of UbiA transcription and the role of upstream D N A sequence elements Don Jones and Eve Stringham (University of British Columbia) have begun to carry out experiments using transgenic nematodes. This approach involves the introduction of a P-galactosidase reporter gene into C.  127  elegans under the control of all or portions of the cloned UbiA promoter region. The reporter construct is introduced into oocytes of living hermaphrodite nematodes by microinjection and those nematodes that transiently or stably maintain the introduced D N A are assayed for B-galactosidase gene function. In this way the transcription of the B-galactosidase gene closely simulates that occurring in C. elegans from the UbiA promoter. The sequences which define the U b i A promoter should be defined by manipulating the length and structure of the promoter region. As C. elegans lacks endogenous B-galactosidase activity, histochemical and biochemical assays of B-galactosidase activity in transgenic nematodes can be used as measures of relative promoter function for manipulated UbiA promoter constructions (Andrew Fire, personal communication). Histochemical assays also allow assay of tissue localization and developmental progression of UbiA transcription. I have designed PCR oligonucleotides which amplify various promoter regions cloned into the microinjection plasmid vector. These oligonucleotides allow (a) rapid size analysis and sequencing for verification of plasmid constructions (b) the assay of one or very few individual transgenic worms, without the need to prepare D N A from them and (c) rapid assay of transgenic progeny for (i) the presence/absence of microinjected plasmid D N A (ii) the integrity of promoter D N A , and potentially (iii) promoter copy number quantification by comparison to an internal standard.  A PCR-based assay saves much of the time previously  required for progeny growth, genomic D N A preparation, and Southern analysis. I was able to amplify transgene promoter D N A from nematodes transformed with the PC3 plasmid using oligonucleotides flanking the hsp!6 promoter sequence. PC3 plasmid D N A or genomic D N A isolated from PC3 nematodes was used as substrate. Encouraged by these results, I have extended the utility of this method to the analysis of small numbers of transgenic animals.  Amplification of PC3  128  promoter D N A has been achieved from as few as one adult transgenic nematode which was boiled and then used as substrate for amplification. PCR D N A fragments were cloned and sequenced, verifying that the amplified D N A corresponded to the promoter sequence from the transgenic plasmid (data not shown). The biggest advantage of this screening protocol is in its speed and capacity to test many different pools of progeny animals. As well, it can be used as a simple test to periodically verify transgenic stocks. I have also designed oligonucleotides which allow the detection of the Pgalactosidase reporter gene mRNA. The result of this type of experiment is seen in Figure 27 (lower panel), where p-galactosidase mRNA expression is detected in pD56 transgenic nematode R N A by the amplification of an approximately 300 bp fragment corresponding to the P-galactosidase mRNA 3' end. Implicit in the detection of functional p-galactosidase activity by histochemical staining is that a functional P-galactosidase mRNA has been synthesized. As the UbiA promoter directs synthesis of a "leaderless" transcript, correct trans splicing of the transgenic P-galactosidase mRNA must occur in order for translation and expression of P-galactosidase enzyme activity to occur. In order to study the trans splicing of the P-galactosidase mRNA one should be able to perform a PCR assay using an oligonucleotide internal to the 5' end of the P-galactosidase mRNA and an oligonucleotide corresponding to the SL1 sequence normally trans spliced onto U b i A mRNA. Transgenic promoter studies with the UbiA upstream region may thus also provide information on the signals required for, and the mechanism of, trans splicing in C. elegans. Preliminary histochemical staining of UbiA transformed nematodes has already revealed some unexpected results : contrary to expectation, UbiA-p-galactosidase was only expressed in a limited number of cells, not constitutively in all cells. The observed tissue distribution is identical under heat shock conditions (E. Stringham,  129  D. Jones personal communication). These results may reflect tissue-specific signals directing UbiA transcription or possibly some form of tissue specific UbiA R N A processing. In future it should also be possible to quantify B-galactosidase mRNA levels in various transgenic nematodes by PCR. This type of result may be integrated with histochemical staining (which is not quantitative) to obtain information on U b i A promoter strength and tissue-specificity.  4.8 Ubiquitin Hybrid Genes Based on Northern analysis with 3'-end specific probes it was determined that C. elegans must possess at least two classes of ubiquitin genes. The major ubiquitin gene, UbiA, was cloned by conventional library screening methods. Numerous attempts were made to isolate non-UbiA member(s) of the ubiquitin family. Unfortunately, exhaustive screening of available C. elegans cDNA and genomic libraries and attempts to directly clone size-fractionated genomic D N A led only to the re-isolation of U b i A clones. This led to the conclusion that an alternative approach was required. I decided to devise a PCR strategy for ubiquitin cloning, hopefully one which might provide a universal tool. In order to overcome the problem of the tandemly repeated organization of ubiquitin genes, a strategy was developed in which only one ubiquitin-specific degenerate oligonucleotide was used, in combination with anchor oligo-dT primed cDNA (see Results). Ubiquitin PCR cDNA amplification gives rise to a final product which represents a copy of the final ubiquitin repeat and the 3-UTR (untranslated region) of the represented locus. Sequencing of this product gives information regarding (1) the amino acid sequence of ubiquitin, (2) the nature of the carboxy terminus of the ubiquitin locus, (3) the site of polyA addition for an uncharacterized locus and (4) the nature of the untranslated sequences present at  130  the 3* end of that ubiquitin mRNA. The 3' "non-ubiquitin" sequence may also be used as a probe to isolate one particular locus or may provide D N A sequence for the design of locus-specific primer(s) for PCR amplification of that locus. It should be noted that the same strategy can be used to obtain the 5' flanking "non-ubiquitin" leader sequence of a ubiquitin locus if the first strand cDNA is tailed with a homopolymer and a conserved ubiquitin oligonucleotide is used in the opposite orientation. This approach to cloning ubiquitin cDNAs was quite successful. Using degenerate primers designed to amino acids 1 to 7 (MQIFVKT, ubl) and amino acids 58 to 63 (DYNIQKE, ub2), which are extremely conserved between animal, plant, and yeast ubiquitin amino acid sequences, both polyubiquitin loci and ubiquitin fusion protein loci cDNAs have been amplified from a variety of organisms. Using this system, predicted truncated C. elegans UbiA cDNA clones and clones for a highly conserved UbiA of C. briggsae have been isolated. Alignment of the two UbiA D N A sequences is shown in Figure 15. The ubiquitin amino acid sequences are identical (data not shown). The sequence between the stop codon and the polyA addition site is also highly conserved in the two UbiA genes. The availability of C. briggsae UbiA sequence paves the way for studies aimed at characterizing sequence elements important for ubiquitin gene regulation and R N A processing. Sequence comparisons between these two nematode species have been used successfully to define coding regions and regulatory elements for cloned C. elegans DNAs e.g. hsp70 (Heschl and Baillie, 1990). The versatility of the ubiquitin PCR primers will also allow rapid comparison with other nematode UbiA genes. It should be possible to use these two primers and the splice leader oligonucleotide to determine whether the polyubiquitin genes of other nematodes undergo cis and/or trans splicing. A class of non-UbiA cDNAs from C. elegans and C. briggsae was also isolated using this procedure. These clones, termed UbiB, are similar to the yeast U b i l and  131  Ubi2 genes (Ozkaynak et al., 1987). These genes have been shown, together with Ubi3. to be comprised of ubiquitin fused to either a 60S (Ubil and Ubi2) or 40S (Ubi3) ribosomal protein in yeast (Finley et al., 1989). The yeast and nematode genes both contain an amino terminal ubiquitin sequence (i.e. at least one) fused to a basic carboxy terminal extension region of 52 amino acids. If one assumes that there is only one ubiquitin repeat fused to this tail sequence, this locus could account for the small ubiquitin mRNA (approximately 700 nucleotides) seen on C. elegans Northern blots. Northern analysis using a UbiB-specific D N A fragment could be performed to confirm this hypothesis. The UbiB genes of C. elegans and C. briggsae are highly similar : they have nearly identical 3* untranslated regions and only two of 52 positions in the tail amino acid sequence are different versus 11 of 52 between C. elegans and yeast. Nematode UbiB retains the same basic nature and has the conserved nuclear localization sequence and zinc finger-like arrangement of four conserved cysteine residues. Likely these nematode tail proteins act as ribosomal proteins much like their yeast counterparts ubil and ubi2 (Ozkaynak et al., 1987; Finley et al., 1989). cDNAs that encode the ubil,2-type fusion proteins have also been isolated from the slime mold D. discoideum (Westphal et al. 1986) and the parasitic protozoan T. cruzi (Swindle et al., 1988). The tail amino acid sequences, like the ubiquitin coding sequence, are thus conserved to a high degree over great evolutionary distances. Cloning ubiquitin cDNAs by PCR proved so successful with nematodes that the technique was used to isolate truncated ubiquitin cDNAs from members of the animal and plant kingdom for which cDNA libraries would not necessarily be readily available. These oligonucleotides have allowed amplification of truncated ubiquitin cDNAs (with and/or without tail extensions) from Drosophila, Ginko, Rhododendron, Equisetum, trout, and newt (shown in Figure 32 aligned with yeast  132  Ubi3 and human UbCEP80). Most of the Ubi3-type cDNAs isolated by this method resemble human CEP80, both in length and sequence conservation. Studies by Ecker et al. (1989) suggest that ubiquitin fusion proteins are very short-lived and possibly never exist in eukaryotic cells - instead they are rapidly processed during or shortly after translation. If ubiquitin fusion proteins are immediately processed during or soon after translation, why did this gene structure evolve? It is possible that processing itself plays some important role in their function - e.g. processing may be required to localize ubiquitin to specific cellular sites where it exerts its function. It has also been suggested that these extensions act as nuclear signal sequences, which localize ubiquitin to specific sites in the nucleus (Ozkaynak et al., 1987). Alternatively, it may be that ubiquitin plays a "chaperone" role to the tail proteins. It has been demonstrated that the ubiquitin moiety of Ubi3 is helpful but not essential in yeast - if the Ubi3 tail protein is expressed without ubiquitin, many copies of the tail construct are required to fully replace one copy of the fusion gene (Finley et al., 1989). Ubiquitin may be used as a translational spacer for these very short proteins or as a 'wedge' to assist in assembling them into the ribosome (Warner, 1989). Since it has been observed that ribosomal proteins turn over rapidly unless they are assembled in the ribosome (Maicas et al., 1988), ubiquitin could protect the ribosomal proteins from degradation. Butt et al. (1989) show that ubiquitin fusion to an unstable protein such as metallothionein in E. coli leads to its stabilization. The tail protein of the ubiquitin hybrid protein is apparently a metalloprotein and may also be relatively unstable in the absence of metal. They propose that the presence of ubiquitin on the ribosomal tail proteins may stabilize the ribosomal tail protein during translation until a metal atom is coordinated and the ubiquitin is cleaved.  133  4.9 PCR of Trans Spliced cDNAs from C. elegans The polymerase chain reaction is a powerful tool for the rapid isolation of genes (especially those with rare transcripts) which avoids most of the complications associated with construction and screening of conventional libraries in bacteria. I have developed a PCR method for the rapid cloning of a family of trans spliced cDNAs from C. elegans. Most C. elegans trans spliced RNAs receive the same 5' leader sequence, donated by a mini-exon gene located adjacent to the 5S RNA gene on chromosome V. I have devised a strategy to rapidly isolate and characterize C. elegans RNAs which undergo trans splicing in order to attempt to understand the nature of genes which are trans spliced and hence the functional significance of trans splicing in C. elegans. Preliminary data demonstrate the utility of this approach for amplifying and analyzing trans spliced gene products (which represent approximately 10% of all C. elegans genes : Bektesh et al., 1988). It is worth noting that the amplified material contains several prominent bands; one of these appears to be C. elegans actin and another a previously uncharacterized protein. Further analysis of this material would involve attempts to identify the gene product for each of the remaining prominent bands, to generate a size-selected trans spliced cDNA library, and to characterize the larger members of this library (which would be selected against in construction of a typical cDNA library).  4.10 Potential Areas of Future Study 1) The ubiquitin PCR techniques developed here could be used to isolate ubiquitin cDNAs from a number of related species beyond C. elegans and C. briggsae, allowing further evolutionary comparisons within the phylum Nematoda. Strategies also exist to isolate 5' and 3' flanking regions of cloned genes providing some unique sequence is available (e.g by inverted PCR (Ochman et al., 1989). One  134  could design oligonucleotide PCR primers based on 3* untranslated region D N A sequence for cloning specific ubiquitin loci and flanking regions from genomic D N A . With this type of information one could study interspecies functionality of ubiquitin flanking regions e.g. by microinjection of chimaeric ubiquitin genes into C. elegans. 2) Experiments devised for the characterization of trans spliced mRNAs could be continued. This work could lead to attempts to isolate trans splicing mutants or to study the mechanism of trans splicing in in vitro extracts; the ability to inhibit trans splicing with heat shock could prove a useful tool in this type of experiment. 3) It would be of interest to isolate and characterize the components of the C. elegans ubiquitin-mediated proteolysis system e.g. ubiquitin activating and conjugating enzymes, ubiquitin hydrolases, isopeptidases etc. C. elegans homologs of already characterized E2 genes could be obtained rapidly by PCR. 4) Little is known about the regulation of transcription from constitutively expressed genes; ubiquitin promoters would be a good model system from which to characterize regulatory factors which interact with constitutively expressed genes.  135 V.  REFERENCES  Ananthan, J., A.L. Goldberg, and R. Voellmy. 1986. Science 232: 522-524. Antonucci, T.K. 1985. Recombinant D N A Techniques 6: 22-24. Univ. of Michigan. Arribas, C , J. Sampedro, and M . Izquierdo. 1986. Biochim. Biophys. Acta 868:119127. Audhya, T. and G. Goldstein. 1985. Methods Enzymol. 116: 279-291. Bachrnair, A., D. Finley, A . Varshavsky. 1986. Science 234: 179-185. Bachmair, A . and A. Varshavsky. 1989. Cell 56: 1019-1032. Baker, R.T. and P.G. Board. 1987. Nucleic Acids Res. 15: 443-463. Baker, R.T. and P.G. Board. 1989. Proc. Natl. Acad. Sci. USA 86: 7751-7755. Ball, E., C C Karlik,C.J. Beall, D.L. Saville, J.C Sparrow, B. Bullard, and E.A. Fyrberg. 1987. Cell 51: 221-228. Bektesh, S., K. Van Doren, and D. Hirsh. 1988. Genes Devel. 2: 1277-1283. Berk, A.J. and P.A. Sharp. 1977. Cell 12: 721-732. Bloom, M., S. Skelly, R. VanBogelen, F. Neidhardt, N . Brot, and H . Weissbach. 1986. J. Bacteriol. 166: 380-384. Blumenthal, T. and J. Thomas. 1988. Trends Genet. 4: 305-308. Bond, U . and M.J. Schlesinger. 1985. Mol. Cell. Biol. 5: 949-956. Bond, U . and M.J. Schlesinger. 1986. Mol. Cell. Biol. 6: 4602-4610. Bonner, W.M., C L . Hatch, and R.S. Wu. 1988. in Ubiquitin (Rechsteiner, M . , ed.) pp. 157-172, Plenum Press, New York. Braun,R. 1986. Bioessays 5: 223-227. Brenner, S. 1974. Genetics 77: 71-94. Brown, J.L. and W.K. Roberts. 1976. J. Biol. Chem. 251: 1009-1014. Bruzik, J.P., K Van Doren, D. Hirsh, and J.A. Steitz. 1988. Nature(London) 335: 559562. Burdon, R.H. 1986. Biochem. J. 240: 313-314.  136  Burke, T.J., J. Callis, and R.D. Vierstra. 1988. Mol. Gen. Genet. 213: 435-443. Butt, T.R., M.I. Khan, J. Marsh, D.J. Ecker, and S.T. Crooke. 1988. J. Biol. Chem. 263: 16364-16371. Butt, T.R., Jonnalagadda, S., B.P. Monia, E.J. Sternberg, J.A. Marsh, J.M. Stadel, D.J. Ecker, and S.T. Crooke. 1989. Proc. Natl. Acad. Sci. USA 86: 2540-2544. Camerini, D., S.P. James, I. Stamenkovic, and B. Seed. 1989. Nature (London) 342: 78-82. Candido, E.P.M., D. Jones, D.K. Dixon, R.W. Graham, R.H. Russnak, and R.J. Kay. 1989. Genome 31:690-697. Carlson, N . , S. Rogers, and M . Rechsteiner. 1987. J. Cell Biol. 104: 547-555. Cassada, R.C. and R.L. Russell. 1975. Dev. Biol. 46: 326-342. Chau, V., J.W. Tobias, A . Bachmair, D. Marriott, D.J. Ecker, D.K. Gonda, and A. Varshavsky. 1989. Science 243: 1576-1583. Chirgwin, J.M., A.E. Przbyla, R.J. McDonald, and W.F. Rutter. 1979. Biochemistry 18: 5294-5299. Chung, C.H. and A.L. Goldberg. 1981. Proc. Natl. Acad. Sci. USA 78: 4931-4935. Ciechanover, A., Y. Hod, and A . Hershko. 1978. Biochem. Biophys. Res. Commun. 81:1100-1104. Ciechanover, A., D. Finley, and A. Varshavsky. 1984a. Cell 37: 57-66. Ciechanover, A., D. Finley, and A . Varshavsky. 1984b. J. Cell. Biochem. 24: 27-53. Clark, D.V., T.M. Rogalski, L.M. Donati, and D.L. Baillie. 1988. Genetics 119:345-353. Clevers, H.C., S. Dunlap, T.E. Wileman, and C. Terhorst. 1988. Proc. Natl. Acad. Sci. USA 85: 8156-8160. Coulson, A., J. Sulston, S. Brenner, and J. Karn. 1986. Proc. Natl. Acad. Sci. USA 83: 7821-7825. Craig, E.A. 1985. Crit. Rev. Biochem. 18: 239-280. Doolittle, W.F. 1978. Nature (London) 272: 581-582. Dudler, R. and A . Travers. 1984. Cell 38: 391-398. Dworkin-Rastl, E., A. Shrutkowski, and M.B. Dworkin. 1984. Cell 39: 321-325.  137  Ecker, D.J., T.R. Butt, J. Marsh, E.J. Sternberg, N . Margolis, B.P. Monia, S. Jonnalagadda, M.I. Khan, P.L. Weber, L. Mueller, and S.T. Crooke. 1987. J. Biol. Chem. 262:14213-14221. Ecker, D.J., J.M. Stadel, T.R. Butt, J.A. Marsh, B.P. Monia, D.A. Powers, J.A. Gorman, P.E. Clark, F. Warren, A . Shatzman, and S.T. Crooke. 1989a. J. Biol. Chem. 264: 77157719. Ecker, D.J., T.R. Butt, J. Marsh, E.J. Sternberg, A . Shatzman, J.S. Dixon, P.L. Weber, and S.T. Crooke. 1989b. J. Biol. Chem. 264:1887-1893. Emmons, S.W., M.R. Klass, and D. Hirsh. 1979. Proc. Natl. Acad. Sci. USA 76:13331337. Emmons, S.W. 1988. in The Nematode Caenorhabditis elegans (W.B. Wood, W.B., ed.), pp. 47-81, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. Evans, A . C , Jr. and K.D. Wilkinson. 1985. Biochemistry 24: 2915-2923. Ferber, S. and A . Ciechanover. 1987. Nature (London) 326: 808-811. Fields, C. 1990. Worm Breeder's Gazette 11: 13. Fink,G.R. 1987. Cell 49: 5-6. Finley, D., A. Ciechanover, and A . Varshavsky. 1984. Cell 37: 43-55. Finley, D., E. Ozkaynak, and A. Varshavsky. 1987. Cell 48: 1035-1046. Finley, D., E. Ozkaynak, S. Jentsch, J.P. McGrath, B. Bartel, M . Pazin, R M . Snapka, and A . Varshavsky. 1988. in Ubiquitin (Rechsteiner, M . , ed.) pp. 39-75, Plenum Press, New York. Finley, D., B. Bartel, and A. Varshavsky. 1989. Nature (London) 338: 394-401. Fire, A . 1986. EMBO J. 5: 2673-2680. Fried, V.A., H.T. Smith, E. Hildebrandt, and K. Weiner. 1987. Proc. Natl. Acad. Sci. USA 84: 3685-3689. Frohman, M.A., M.K. Dush, and G.R. Martin. 1988. Proc. Natl. Acad. Sci. USA 85: 8998-9002. Gaterman, K.B., G.H. Rosenberg, and N.F. Kaufer. 1988. Biotechniques 6: 951-952. Gausing, K. and R. Barkardottir. 1986. Eur. J. Biochem. 158: 57-62. Gilbert, W., M . Marchionni, and G. McKnight. 1986. Cell 46: 151-154.  138  Giorda, R. and H.L. Ennis. 1987. Mol. Cell. Biol. 6: 2097-2103. Goebl, M.G., J. Yochem, S. Jentsch, J.P. Mcgrath, A. Varshavsky, and B. Byers. 1988. Science 241:1331-1335. Goff, S.A. and A.L. Goldberg. 1985. Cell 41: 587-55. Goff, S.A. and A.L. Goldberg. 1987. J. Biol. Chem. 262: 4508-4515. Goff, S.A., R. Voellmy, and A.L. Goldberg. 1988. in Ubiquitin (Rechsteiner, M . , ed.) pp. 207-238, Plenum Press, New York. Goldberg, A. and A . St. John. 1976. Annu. Rev. Biochem. 45: 747-803. Goldberg, M.L. 1979. Ph.D. thesis, Stanford University, Stanford, California. Goldknopf, I.L., and H . Busch. 1977. Proc. Natl. Acad. Sci. USA Z4: 864-868. Goldstein, G., M . Scheid, U . Hammerling, E.A. Boyse, D.H. Schlesinger, and H.D. Niall. 1975. Proc. Natl. Acad. Sci. USA 72:11-15. Grossman, A.D., J.W. Erickson, and C.A. Gross. 1984. Cell 38: 383-390. Guarino, L.A. 1990. Proc. Natl. Acad. Sci. USA 87: 409-413. Haas, A . L., P. Abrens, P. M . Bright, and H . Ankel. 1987. J. Biol. Chem. 262: 1131511323. Hanahan, D. 1983. J. Mol. Biol. 166: 557-580. Hanyu, N . , Y. Kuchino, and S. Nishimura. 1986. EMBO J. 5: 1307-1311. Henikoff,S. 1984. Gene 28: 351-359. Hershko, A., A. Ciechanover, H . Heller, A.L. Haas, and LA. Rose. 1980. Proc. Natl. Acad. Sci. USA 77:1783-1786. Heschl, M.F.P. and D.L. Baillie. 1989. D N A 8:233-243. Heschl, M.F.P. and D.L. Baillie. 1990. J. Mol. Evol. (in press). Heschl, M.F.P. and D.L. Baillie. 1990. Comp. Biochem. Phys. (in press). Hershko, A., E. Eytan, A . Ciechanover, and A.L. Haas. 1982. J. Biol. Chem. 257: 13964-13970. Hershko, A., H . Heller, E. Eytan, G. Kaklis, and LA. Rose. 1984. Proc. Natl. Acad. Sci. USA 81: 7021-7025.  139  Hershko, A. and A. Ciechanover.  1986. Prog. Nucl. Acid Res. Mol. Biol. 33: 19-56.  Hershko, A., H . Heller, E. Eytan, and Y. Reiss. 1986. J. Biol. Chem. 261: 11992-11999. Hershko, A. 1988a. J. Biol. Chem. 263: 15237-15240. Hershko, A. 1988b. in Ubiquitin (Rechsteiner, M . , ed.) pp. 325-332, Plenum Press, New York. Hiromi, Y., H . Okamoto, W.J. Gehring, and Y. Hotta. 1986. Cell 44: 293-301. Huang, S.-Y., M.B. Barnard, M . Xu, S.I. Matsui, S.M. Rose, and W.T. Garrard. 1986. Proc. Natl. Acad. Sci. USA 83: 3738-3742. Huang, X.-Y. and D. Hirsh. 1989. Proc. Natl. Acad. Sci. USA 86: 8640-8644. Hunt, L.T. and M.O. Dayhoff. 1977. Biochem. Biophys. Res. Commun. 74: 650-655. Innis, M.A., K.B. Myambo, D.H. Gelfand, and M.A.D. Brow. 1988. Proc. Natl. Acad. Sci. USA 85: 9436-9440. Jabben, M . , J. Shanklin, and R.D. Vierstra. 1989. J. Biol. Chem. 264: 4998-5005. Jahngen, J. H . , A. L. Haas, A. Ciechanover, J. Blondin, D. Eisenhauer, and A. Taylor. 1986. J. Biol. Chem. 261: 13760-13767. Jakobovits, E.B., U . Schlokat, J.L. Vannice, R. Derynck, and A.D. Levinson. 1988. Mol. Cell. Biol. 8: 5549-5554. Jentsch, S., J.P. Mcgrath and A. Varshavsky. 1987. Nature (London) 329: 131-134. Jones, D., D.K. Dixon, R W . Graham, and E.P.M. Candido. 1989. D N A 8: 481-490. Kageyama, R , G.T. Merlino, and I. Pastan. 1989. J, Biol. Chem. 264: 15508-15514. Kay, R.J., R.H. Russnak, D. Jones, C. Mathias, and E.P.M. Candido. 1987. Nucleic Acids Res. 15: 3723-3741. Krause, M . and D. Hirsh. 1986. in Cell and Molecular Biology of the Cytoskeleton, (J.W. Shay, ed.) pp. 151-178, Plenum Publishing Corporation, New York. Krause, M . and D. Hirsh. 1987. Cell 49: 753-761. Laird, P.W. 1989. Trends Genet. 5: 205-208. Lambowitz, A . M . 1989. Cell 56: 323-326. Lawrence, C W . 1982. Adv. Genet. 21: 173-254.  140  Lee, H . , J.A. Simon, and J.T. Lis. 1988. Mol. Cell. Biol. 8: 4727-4735. Letovsky, J. and W.S. Dynan. 1989. Nucleic Acids Res. 17: 2639-2653. Leung, D.W., S.A. Spencer, G. Cachianes, R.G. Hammonds, C. Collins, W.J. Henzel, R. Barnard, M.J. Waters & W.I. Wood. 1987. Nature (London) 330: 537-543. Levinger, L. and A . Varshavsky. 1982. Cell 28: 375-385. Levinger, L. 1985. J. Biol. Chem. 260: 11799-11804. Lewis, C D . , S.P. Clark, G. Felsenfeld, and H . Gould. 1988. Genes Dev. 2: 863-873. Li, H . , U.B. Gyllensten, X. Cui, R.K. Saiki, H.A. Erlich, and N . Arnheim. 1988. Nature (London) 335: 414-417. Lund, P. K., B. M . Moats-Staats, J.G. Simmons, E. Hoyt, A.J. D'Ercole, F. Martin, and J.J. Van Wyk. 1985. J. Biol. Chem. 260: 7609-7613. Maicas, E., F.G. Pluthero, and J.D. Friesen. 1988. Mol. Cell. Biol. 8: 169-175. Maniatis, T., E.F. Fritsch, and J. Sambrook. 1982. Molecular Cloning, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. Matsui, S.-L, B. Seon, and A . Sandberg. 1979. Proc. Natl. Acad. Sci. USA 76: 63866390. Mayer, R.J., M . Landon, F.J. Doherty, J.S. Lowe, G.P. Reynolds, and E.J. Byrne. 1989. Nature (London) 340:193. McLauchlin, J., D. Gaffney, J.L. Witton, and J.B. Clements. 1985. Nucleic Acids Res. 13:1347-1368. Messing, J. 1983. Methods Enzymol. 101: 20-78. Meyers, G., T. Rumenapf, and H.-J. Thiel. 1989. Nature(London) 341: 491. Mezquita, J., R. Oliva, and C. Mezquita. 1987. Nucleic Acids Res. 15: 9604. Mizusawa, S. and S. Gottesman. 1983. Proc. Natl. Acad. Sci. USA 80: 358-362. Moerman, D.G., G.M. Benian, and R.H. Waterston. 1986. Proc. Natl. Acad. Sci. USA 83: 2579-2583. Mueller, R. D., H . Yasuda, C L . Hatch, W.M. Bonner, and E.M. Bradbury. 1985. J. Biol. Chem. 260: 5147-5153. Muhich, M.L. and J.C. Boothroyd. 1989. J. Biol. Chem. 264: 7107-7110.  141  Munro, S. and H . Pelham. 1985. Nature (London) 317: 477-478. Murphy, W.J., K.P. Watkins, and N . Agabian. 1986. Cell 47: 517-525. Murti, K.G., H.T. Smith, and V.A. Fried. 1988. Proc. Natl. Acad. Sci. USA 85: 30193023. Neidhardt, F . C , R.A. VanBogelen, and E.T. Lau. 1983. J. Bacterid. 153: 597-603. Nelson, D.W. and B.M. Honda. 1985. GENE 38:245-251. Nelson, D.W. and B.M. Honda. 1986. Can J. Genet. Cytol. 28:545-553. Nelson, D.W. and B.M. Honda. 1989. Nucleic Acids Res. Yh 8657-8667. Neves, A . M . , I. Barahona, L. Galego and C. Rodrigues-Pousada. 1988. GENE 73: 8796. Ochman, H . , A.S. Gerber, and D.L. Hartl. 1988. Genetics 120: 621-623. Okabe, T., M . Fujisawa, A. Mihara, S. Sato, N . Fujiyoshi, and F. Takuku. 1986. J. Cell Biol. 103: 442A. Ozkaynak, E., D. Finley and A. Varshavsky. 1984. Nature (London) 312: 663-666. Ozkaynak, E., D. Finley, M . Solomon, and A. Varshavsky. 1987. EMBO J. 6:14291439. Parag, H.A., B. Raboy and R.G. Kulka. 1987. EMBO J. 6: 55-61. Pardue, M.L., J.R. Feramisco, and S. Lindquist. 1988. Stress-Induced Proteins. Alan R. Liss, Inc., New York. Pelham, H.R.B. 1982. Cell 30: 517-528. Pelham, H.R.B. and M . Bienz. 1982. EMBO J. 1: 1473-1477. Perisic, O., H . Xiao, and J.T. Us. 1989. Cell 59: 797-806. Pickart, C M . and I.A. Rose. 1985. J. Biol. Chem. 260:1573-1581. Pickart, C M . and LA. Rose. 1986. J. Biol. Chem. 261: 10210-10217. Prakash, L. 1989. Genome 31: 597-600. Prasad, S.S. and D.L. Baillie. 1989. Genomics 5:185-198.  142 Rapoport, S.M., W. Dubiel, and M . Muller. 1981. Acta Biol. Med. Germ. 40: 12771283. Rechsteiner, M . 1987. Annu. Rev. Cell Biol. 3: 1-30. Rechsteiner, M . 1988. Ubiquitin. Plenum Press, New York. Redman, K.L. and M . Rechsteiner. 1989. Nature (London) 338: 438-440. Reiss, Y., D. Kaim, and A. Hershko. 1988. J. Biol. Chem. 263: 2693-2698. Riddle, D.L., M . M . Swanson, and P.S. Albert. 1981. Nature (London) 290: 668-671. Russnak, R.H., D. Jones, and E. P.M. Candido. 1983. Nucleic Acids Res. 11:3187-3205. Russnak, R.H. and E.P.M. Candido. 1985. Mol. Cell. Biol. 5: 1268-1278. Saiki, R.K., S. Scharf, F. Faloona, K.B. Mullis, G.T. Horn, H.A. Erlich, and N . Arnheim. 1988. Science 239: 487-491. Sanger, F., S. Nicklen, and A.R. Coulson. 1977. Proc. Natl. Acad. Sci. USA 74: 54635467. Scheidereit, C , S. Geisse, H . M . Westphal, and M . Beato. 1983. Nature (London) 304: 749-752. Schibler, U , O. Hagenbuchle, P.K. Wellauer, and A.C. Pittet. 1983. Cell 33: 501-508. Schlesinger, D.H., G. Goldstein, and H.D. Niall. 1975. Biochemistry 14: 2214-2218. Schlesinger, M.J., M . Ashburner, and A . Tissieres. 1982. Heat Shock : From Bacteria to Man. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. 1982. Shanklin, J., M . Jabben, and R.D. Vierstra. 1987. Proc. Natl. Acad. Sci. USA 84: 359363. Shanklin, J., M . Jabben, and R.D. Vierstra. 1989. Biochemistry 28: 6028-6034. Sharp, P.A. 1987. Cell 50:147-148. Sharp, P.M. and W. Li. 1987. J. Mol. Evol. 25: 58-64. Sharp, P.M. and Wen-Hsiung Li. 1987. TREE 2: 328-332. Siegelman, M . , M.W. Bond, W.M. Gallatin, T. St. John, H.T. Smith, V.A. Fried, and LL. Weissman. 1986. Science 231: 823-829. Snutch, T.P. and D.L. Baillie. 1983. Can. J. Biochem. Cell Biol. 61:480-487.  143 Snutch, T.P., M.F.P. Heschl, and D.L. Baillie. 1988. Gene 64:241-255. Solnick,D. 1985. Cell 42:157-164. Sorger, P.K., M.J. Lewis, and H.R.B. Pelham. 1987. Nature (London) 329: 81-84. Sorger, P.K. and H.R.B. Pelham. 1988. Cell 54: 855-864. Sorger, P.K. and H . C . M . Nelson. 1989. Cell 59: 807-813. St. John, T., W.M. Gallatin, M . Siegelman, H.T. Smith, V.A. Fried,I.L. Weissman. 1986. Science 231: 845-849. Stone, E.M., K.N. Rothblum, and R.J. Schwartz. 1985. Nature (London) 313: 498-500. Sulston, J.E., E. Schierenberg, and J.G. White. 1973. Dev. Biol. 100: 64-119. Sulston, J.E. and S. Brenner. 1974. Genetics 77: 95-104. Sutton, R.E. and J.C. Boothroyd. 1986. Cell 47: 527-535. Swindle, J., J. Ajioka, H . Eisen, B. Sanwal, C. Jacque-Mot, Z. Browder, and G. Buck. 1988. EMBO J. 7:1121-1127. Tanaka, M . , N . Masuda, M . Watahiki, M . Yamakawa, K. Shimizu, J. Nagai, and K. Nakashima. 1988. Biochemistry International 16: 287-292. Thomas, J.D., R.C. Conrad, and T. Blumenthal. 1988. Cell 54: 533-539. Thome, A.W., P. Sautiere, G. Briand, and C. Crane-Robinson. 1987. EMBO J. 6: 10051010. Tobe, T., K. Ito, and T. Yura. 1984. Mol. Gen. Genet. 195:10-16. Toniolo, D., M . Persico, and M . Alcalay. 1988. Proc. Natl. Acad. Sci. USA 85: 851-855. Ullrich, A., J. Shine, J. Chirgwin, R. Pictet, E. Tischer, W.J. Rutter, and H . M . Goodman. 1977. Science 196: 1313-1319. Urano, Y., K. Watanabe, M . Sakai, and T. Tamaoki. 1986. J. Biol. Chem. 261: 32443251. Van der Ploeg, L.H.T. 1986. Cell 47: 479-480. Van Doren, K. and D. Hirsh. 1988. Nature(London) 335: 556-559. Vierstra, R.D., S.M. Langan, and G.E. Schaller. 1985. J. Biol. Chem. 260: 12015-12021. Vierstra, R.D., S.M. Langan, and G.E. Schaller. 1986. Biochemistry 25: 3105-3108.  144  Vierstra, R.D., T.J. Burke, J. Cortner, P.H. Hatfield, M . Jabben, and J. Shanklin. 1987. in Plant Gene Systems and Their Biology, (Key, J. and L. Mcintosh, eds.) pp. 251-262, Alan R. Liss, Inc., New York. Vierstra, R.D. and M.L. Sullivan. 1988. Biochemistry 27: 3290-3295. Vijay-Kumar, S., C.E. Bugg, K.D. Wilkinson, and W.J. Cook. 1985. Proc. Natl. Acad. Sci. USA 82: 3582-3585. Vijay-Kumar, S., C.E. Bugg, and W.J. Cook. 1987a. T. Mol. Biol. 194: 531-544. Vijay-Kumar, S., C.E. Bugg, K.D. Wilkinson, R.D. Vierstra, P.M. Hatfield, and W.J. Cook. 1987b. J. Biol. Chem. 262: 6396-6399. Warner, J.R. 1989. Nature (London) 338: 7. Waxman, L., J.M. Fagan, and A.L. Goldberg. 1987. J. Biol. Chem. 262: 2451-2457. Westphal, M . , A. Muller-Taubenberger, A. Noegel, and G. Gerisch. 1986. FEBS Lett. 209: 92-96. White, J.G., E. Southgate, J.N. Thomson, and S. Brenner. 1986. Philos. Trans. R. Soc. Lond. B Biol. Sci. 314: 1-340. Wiborg, O., M.S. Pedersen, A. Wind, L.E. Berglund, K.A. Marcker, and J. Vuust. 1985. EMBO J. 4: 755-759. Wilkinson, K.D., M.J. Cox, L.B. O'Connor, and R Shapira. 1986. Biochemistry 25: 4999-5004. Wilkinson, K.D. 1987. Anti Can. Drug 2: 211-229. Wong, S. and D.A. Campbell. 1989. Mol. Bioch. Parasit. 37:147-50. Wood, W.B. 1988. The Nematode Caenorhabditis elegans. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. Yarden, Y. A., J.A. Escobedo, W. Kuang, T.L. Yang-Feng, T.O. Daniel, P.M. Tremble, E.Y. Chen, M.E. Ando, R.N. Harkins, Y. Francke, V. A. Fried, A. Ullrich and K.T. Williams. 1986. Nature (London) 323: 226-232. Yoo, Y., K. Rote, and M . Rechsteiner. 1989. J. Biol. Chem. 264: 17078-17083.  145  Yost, H.J. and S. Lindquist. 1986. Cell 45:185-193. Zinn, K., D. Dimaio, and T. Maniatis. 1983. Cell 34. 865-879.  146 VL APPENDIX  A. List of Oligonucleotides and their Sequences  Primer Extension Sequencing Oligonucleotides : UbiA Act3  5' A G A T T T G C A T G A T T G 3* 5' G A T T A G T T T T T A A T G T A C 3'  PCR Oligonucleotides* : RG05 RG02 B-galOl primer Anchor oligo-dT SLl splice leader ubl degenerate** ub2 degenerate**  5' G C T G T C T A G A A G G G T T T T C C C A G T C A C G A C 3' 5' G A T C C T G C A G G T G T G G A A T T G T G A G C G G A T 3' 5' T C A T G G A T C C G A G C C C G T C A G T A T C G G C G G 3' 5' C G A G C A T G C G T C G A C A G G C A T 3* 5' T C A T G G A T C C G G T T T A A T T A C C C A A G T T T G A G 3' 5' C A T T G G A T C C G T ATGCAIATTTTIGTIAAIAC 3' 5' G T A C G G A T C C G GAITALAAIATICALAAIGA 3' 1 7  * Oligonucleotides were synthesized with 5' restriction enzyme recognition sequences in order to facilitate cloning of PCR products.  ** Ubiquitin oligonucleotides were synthesized with deoxyinosine in the third position of degenerate codons.  147  B. E. coli Strains and their Genotypes*  JM109  recAl endAl gyrA96 thi-1 hsdR17 supE44 relAl Ddac- proAB) mcrANal F'(traD36 lacA lacZDM15) r  XL-lBlue  recAl endAl gyrA96 thi-1 hsdR17 supE44 relAl X~ lac (F proAB* laciq lacZDM15 TnlO(tet )) R  DH5a  recAl endAl gyrA96 thi-1 hsdR17supE44 relAl AlacU169 (<|)80 lacZDM15)  Y1088  supE supF metB trpR hsdR' h s d M tonA21 strA AlacU169 mcrA proC::Tn5/pMC9  LE392  hsdR514 supE44 supF58 lacYl galK2 galT22 metBl trp55 mcrA  +  * JM109 was used to propagate M13 vectors; XL-1 Blue and DH5a were used for plasmid vectors; Y1088 was used for Xgtll bacteriophage; and LE392 was used for A,Charon4 bacteriophage.  C. Calculation of UbiA mRNA Abundance  With regards to the R N A dot hybridization in Figure 14 : 1.  The hybridization signal for 10 pg of FS D N A is approximately equivalent to  that for 2 ug of total cellular R N A (i.e. 5 pg = 1 ug) 2.  The FS insert represents 80 bp of the 7380 bp single stranded plasmid  3. The FS insert represents 80 bp of the 2600 bp UbiA mRNA 4.  5 pg x (80/7380) / (80/2600) =1.8 pg UbiA mRNA per ug total cellular R N A  5. If polyA mRNA is approximately 5% of total cellular R N A (i.e. 50 ng per n g ) , then UbiA represents 1.8 pg / 50,000 pg or 0.003% of polyA mRNA. +  +  148  D.  Sequence of Tail Regions of Nematode UbiB cDNAs*  R C.b.:  N  I  I  E  P  S  L  R  Q  L  A  Q  K  C  D  K  Q  I  C  R  K  C  Y  A  R  L  TAC AAC TGC GAC AAG CAG ACT TGC AGA AAG TGC TAC GCT CGT CTC  P C.b.:  G  CGT GGA GGA ATC ATC GAG CCA TCG CTC CGT CAG CTT GCC CAG AAG  Y C.b.:  G  P  R  A  S  N  C  R  K  K  K  C  G  H  S  CCA CCA CGT GCC AGC AAC TGC AGA AAG AAG AAG TGC GGA CAC TCC  1  2  L  R  I  K  K  K  L  K  Och  C.b.:  AAC GAT CTC CGC ATC AAG AAG AAG CTG AAG TAA  C.e.:  -GC ~ G  * Standard single letter code is used for amino acid sequences; 1,2 : N,D in C. briggsae: S,E in C. elegans; Och : Ochre stop codon.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items