UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A genetic and molecular analysis of the bli-4 locus of Caenorhabditis elegans, an essential gene encoding… Peters, Kenneth W. 1992

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

Item Metadata

Download

Media
831-ubc_1993_spring_phd_peters_kenneth.pdf [ 6.35MB ]
Metadata
JSON: 831-1.0086382.json
JSON-LD: 831-1.0086382-ld.json
RDF/XML (Pretty): 831-1.0086382-rdf.xml
RDF/JSON: 831-1.0086382-rdf.json
Turtle: 831-1.0086382-turtle.txt
N-Triples: 831-1.0086382-rdf-ntriples.txt
Original Record: 831-1.0086382-source.json
Full Text
831-1.0086382-fulltext.txt
Citation
831-1.0086382.ris

Full Text

A GENETIC AND MOLECULAR ANALYSIS OF THE bli 4 LOCUS OF -  Caenorhabditis elegans, AN ESSENTIAL GENE ENCODING KEX2-LIKE SUBTILISIN-TYPE SERINE ENDOPROTEASES by KENNETH WILLIAM PETERS BSc. (Honours) Simon Fraser University, 1986  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Genetics Program Faculty of Medicine  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA JULY, 1992 © Kenneth William Peters  In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  (Signature)  Department of Medical Genetics The University of British Columbia Vancouver, Canada Date January 7, 1993  DE•6 (2/88)  ABSTRACT This thesis reports a characterization of genetic mutations and molecular structure of the bli-4 (I) gene of Caenorhabditis elegans. The bli-4 locus had been previously defined by a single recessive mutation, e937, which disrupts the structure of adult-stage cuticle causing the formation of fluid-filled separations of the cuticle layers, or blisters. A characterization of  e937 and eleven additional mutations is reported. The mutations were grouped into three classes based on phenotype and complementation analysis: Class I, represented by the single allele causing blistering, e937; Class II, nine lethal mutations that arrest development at the end of embryogenesis and that fail to complement all other mutations; and Class 111, two larval lethal mutations, s90 and h754, that complement e937. The complementation pattern provides evidence that all of these mutations are allelic, and that bli-4 is a complex locus with an essential function late in embryogenesis. A region of chromosome I that includes the bli-4 locus was identified by aligning the C. elegans genetic and physical maps. The bli-4 coding region was identified by using cosmids as hybridization probes to detect chromosomal alterations in the DNA of bli-4 mutant strains. Two mutations of bli-4 are small rearrangements of the gene; e937, is a 3.5 kilo base (kb) deletion, and h1010 is an insertion of the 1.6 kb transposable element, Tc1. Three protein products that differ at the carboxyl end, are predicted from alternately spliced bli-4 cDNA clones. The predicted proteins were designated blisterin A, blisterin B and blisterin C according to the order that the variantly spliced 3' ends occur on the chromosome. Of the three bli-4 cDNA clones characterized, only blisterin B includes an open reading frame beginning with an ATG start codon. The blisterin A and blisterin C open  ii  reading frames begin within blisterin B, and are likely to be incomplete. The predicted blisterin B gene product has a potential secretion signal peptide at its amino terminal. Blisterin C has a potential transmembrane domain near its carboxyl terminal while blisterin A and B lack this domain. The blisterins share significant sequence identity with kex2-like serine endoproteases, which are responsible for the cleavage of secreted proteins in yeast and mammals. This characterization of bli-4 provides the first evidence for an essential role of a KEX2-like gene in the development of a multi-cellular organism.  iii  TABLE OF CONTENTS Title page Abstract^  ii  Table of Contents^  iv  List of Tables^  viii  List of Figures^  ix  Acknowledgments^  xi  Dedication^  xii  Chapter One: introduction^  1  1.1 General^  1  1.2 Biology of C. elegans^  1  1.3 Cuticle structure^  4  1.4 Origin of bli 4 alleles^  6  1.5 Resources used to clone bli 4^  9  -  -  1.6 Review of dibasic endoproteases^  10  Chapter Two: Materials and Methods ^  15  2.1 Nomenclature^  15  2.2 Nematode culture conditions^  16  2.3 Determination of penetrance^  16  2.4 Determination of larval stage ^  17  2.5 Complementation testing^  18  2.6 Determination of developmental arrest stage ^  19  2.7 Allele mapping^  20  2.8 Screen for bli 4 alleles in a mutator strain^  20  2.9 Isolation of intact cuticle^  21  2.10 Cosmids and plasmids growth conditions ^  21  2.11 DNA preparation^  22  -  iv  2.12 Estimation of DNA concentration^  26  2.13 Restriction digests^  26  2.14 Gel electrophoresis^  27  2.15 Electroelution^  27  2.16 Subcloning^  28  2.17 Southern transfers ^  29  2.18 Northern transfers^  30  2.19 Preparation of hybridization probes^  31  2.20 Hybridization^  32  2.21 Autoradiography^  34  2.22 Construction of CB937 lambda-zap library^  34  2.23 cDNA isolation^  36  2.24 Lambda-zap phage library screening^  36  2.25 Isolation of plasmids from lambda-zap phage ^  37  2.26 Deletions^  38  2.27 DNA sequencing^  40  2.28 Polymerase chain reaction^  45  2.29 Sequence analysis^  46  Chapter Three: Results ^  47  3.1 Genetic analysis of the bli 4 locus^  47  3.1.1 Characterization of e937 ^  47  -  3.1.1.1 Blisters form between cuticle layers ^  47  3.1.1.2 Blistering is incompletely penetrant and is not temperature sensitive^  51  3.1.1.3 Blistering is adult specific^  51  3.1.1.4 Interaction of e937 with cuticle genes ^  56  3.1.2 Mapping of bli 4 alleles^ -  v  59  3.1.3 Screen for a mutator allele ^  60  3.1.4 Determination of developmental arrest stage^  63  3.1.5 Evidence that e937 is hypomorhic^  63  3.1.6 Complementation analysis of bli 4 alleles^  64  3.2 Molecular analysis of the bli 4 locus^  67  3.2.1 Identification of the bli 4 coding region^  67  -  -  -  3.2.1.1 Alignment of the genetic and physical maps^  67  3.2.1.2 Identification of rearrangements in the DNA of  bli 4 mutant strains^ -  80  3.2.1.3 Detection of RNA bands with KO4F10 fragments^82 3.2.2 Analysis of bli 4 cDNA clones^ -  89  3.2.2.1 Isolation and hybridization analysis ^  89  3.2.2.2 Sequence analysis of cDNA clones^  89  3.2.2.3 Northern analysis^  94  3.2.3 Analysis of bli 4 genomic DNA preliminary sequence data ^105 -  3.2.4 Analysis of the predicted bli 4 proteins^ -  111  3.2.4.1 Identification of similarity with kex2-like serine proteases ^111 3.2.4.2 Protein features^  123  3.2.5 Position of bli 4 mutations in predicted protein sequences^131 -  3.2.5.1 h1010^  131  3.2.5.2 e937^  131  Chapter Four: Discussion^  134  4.1 Evidence that bli-4 is a complex locus^  134  4.2 Evidence that bli 4 is the kex2-like coding element on KO4F10^137 -  4.3 b/i-4 mutations^  140  4.4 bli 4 gene products are related to kex2-like proteases^142 -  4.5 Concluding comments^  vi  144  Literature Cited^  146  Appendix A. Isolation and characterization of hT2^  160  Appendix B. Strains used in this study^  167  Appendix C. Bacterial strains^  168  Appendix D. Cosmid and plasmid clones^  169  Appendix E. vonHeinje identification of blisterin signal peptidase site ^172  vii  LIST OF TABLES Table 1. Alleles of bli-4^  7  Table 2. bli-4(e937) penetrance in Dpy homozygote and heterozygote backgrounds^  58  Table 3. bli-4 genetic map data^  61  Table 4. Inter se complementation data for bli-4 alleles^ 66 Table 5. hP5 three factor mapping data ^  73  Table 6. hDf8 dosage data^  78  Table 7. Summary of bli-4 cDNA sequence data^  91  Table 8. Summary of FASTA search of SWISSPROT protein database using blisterin B^  115  Table 9. Overall identity scores of blisterin B and kex2-like proteins^116 Table 10. Identity of blisterin domains with kex2-like enzymes^120 Table 11. Recombination in hT2 heterozygotes^  163  Table 12. Genetic screens using hT2^  165  Table 13. Strains used in this study^  167  Table 14. Bacterial strains^  168  Table 15. Cosmid clones^  169  Table 16. Plasmid clones^  170  viii  LIST OF FIGURES Figure 1. Developmental changes in cuticle structure Figure 2. Genetic map of the bli-4 region  ^  5  ^  Figure 3. Blistered adult CB937 bli-4(e937) worms  ^  Figure 4. Blistering in a whole worm and isolated cuticle Figure 5. Developmental stage of blistering  8 49  ^  ^  Figure 6. Screen for mutator-induced alleles of bli-4  Figure 8. Construction of N2/BO RFLP mapping strains Figure 9. Southern analysis of hP5 mapping strains  55  ^  Figure 7. Genetic and physical maps of the bli-4 region  50  62  ^  ^  69 70  ^  72  Figure 10. Dosage analysis of hDf8^  76  Figure 11. Detection of RFLDs by C44D11 in hDf8^  79  Figure 12. Southern analysis of KR1858 bli-4(h1010)^  84  Figure 13. Amplification of the h1010 Tc1 insertion site^ 85 Figure 14. Southern analysis of CB937 bli-4(e937)^  87  Figure 15. Northern analysis of bli-4 transcripts^  88  Figure 16. Alignment of cDNA clones with the bli-4 region^96 Figure 17. Sequencing strategies ^  98  Figure 18. Blisterin B Sequence data^  100  Figure 19. Blisterin C Sequence data ^  102  Figure 20. Blisterin A Sequence data^  104  Figure 21. bli-4 common genomic sequence^  106  Figure 22. Genomic sequence for the 3' end of blisterin A^108 Figure 23. Genomic sequence for the 3' end of blisterin C^109 Figure 24. Alignment of blisterin B and Bacillus subtilius amyolosaccharitis subtilisin^  117  ix  Figure 25. Alignment of blisterin B with kex2-like proteins ^118 Figure 26. Similarity of blisterin 50 amino acid segments with kex2-like proteins^  121  Figure 27. Blisterin and kex2-like protein structural features ^127 Figure 28. Hydropathy analysis of the blisterins^  129  Figure 29. Alignment of blisterin C and Furin Cysteine-rich regions^130 Figure 30. Sequence of e937 deletion breakpoints^  133  Figure 31. Recombination suppression in hT2 heterozygotes^164 Figure 32. Southern analysis of bli 4(11862)^  166  Figure 33. Genomic subclones of the bli 4 region^  171  -  -  x  ACKNOWLEGMENTS  The work presented in this thesis represents the result of six years of my life as a graduate student in the lab of Dr. Ann Rose. During that time I learned from and was supported by many individuals. In particular, I would like to acknowledge the instruction of Ann Marie Howell, who helped me with genetics, and Terry Starr and Shiv Prasad, who taught me molecular techniques. I would also like to thank Kim McKim, Monique Zetka and Joe Babity. My work was greatly facilitated by my collaboration with Jennifer McDowall, who identified many of the mutations of the bli-4 gene and who shared her germline transformation results with me. I am also grateful to Betty Leung for her assistance with DNA sequencing. The direction and support of my supervisory committee, Dr. David Baillie, Dr. Don Moerman, Dr. Tom Grigliatti and Dr. Fred Dill, is gratefully acknowledged. I would particularly like to thank my supervisor, Dr. Ann Rose, for her patience and support. Finally, I would like to thank my parents, Ron and Cynthia and my wife, Helen for their love and support. My work was made possible by Post Graduate Research Fellowships from the Medical Research Council of Canada, and the Natural Sciences and Engineering Research Council.  xi  This thesis is dedicated to Helen and Jeffrey.  xi i  CHAPTER ONE: INTRODUCTION 1.1 GENERAL Much of our present understanding of how organisms develop has come from analysis of heritable alterations in the genetic material. In many cases, genetic analysis has begun by the identification of a mutation with an unusual phenotype, and the curiosity of an investigator to know the basis of that phenotype. My thesis presents an analysis of the bli-4 locus. bli-4 is a member of a group of C. elegans genes that can be mutated to cause blisters. The blistered phenotype is fluid-filled separations of cuticle layers (Brenner, 1974). Blistering exhibits both variable penetrance and expressivity. When I began this work, six genes with mutations resulting in the blistered phenotype had been identified, named bli-1 through bli-6.  None of the blister genes had been cloned, and the physiological basis for blistering was unknown. The purpose of my thesis work was to determine the nature of the bli-4 gene product with the goal of elucidating the biological function of the gene. 1.2 BIOLOGY OF C. elegans  Caenorhabditis elegans is a free-living soil nematode that exists as self-fertilizing hermaphrodites and males. C. elegans is widely used as model organism for genetic studies, following the work of Brenner (1974). The biology of C. elegans has been studied extensively, leading to insights into the genetic control of its life cycle, genomic organisation, development, sexual differentiation, anatomy, muscle and nervous system, and cell lineage. This work is reviewed in the book, The nematode Caenorhabditis elegans (Wood, 1988). The hermaphrodite gonad produces approximately 300 hundred sperm during the last larval stage, then switches to the production of ova through its adult  1  1 Introduction life. Ova are fertilized as they pass through a spermatheca, where the sperm are stored. The fertilized egg, or zygote, begins to develop immediately, and is passed through the vulva into the surrounding environment 2 to 4 hours later. The embryo continues development until the embryo hatches out of its egg, about 14 hours post-fertilization at 25C. The worm passes through four larval stages lasting about 45 hours, termed L1, L2, L3 and L4. Each developmental stage is punctuated by a cuticle moult and the synthesis of a new cuticle (Singh and Sulston, 1978). Worms reach adulthood about 2.5 days after fertilization at 25C. An additional developmental stage, the dauer larvae, occurs as a substitute for the L2 stage under conditions of extreme stress such as starvation (reviewed by Riddle, 1988). At the time of hatching, hermaphrodites have 558 somatic cells, males have 560 (Sulston et al., 1983). Adult worms are about one mm in length, and have about 1000 somatic nuclei (Sulston and Horvitz, 1981). Sex is determined by X to autosome ratio; hermaphrodites have two X chromosomes (XX), males have one (XO) (Reviewed by Hodgkin, 1980). Male worms are produced periodically due to nondisjunction of the X chromosome (Hodgkin, Horvitz, and Brenner, 1979), at a frequency of about 0.1°/0 at 20C (Rose and Baillie, 1979). When males are crossed to hermaphrodites, the male sperm is used in preference to the hermaphrodite sperm, and about 50% of the cross-progeny are male. In the absence of males, C. elegans reproduces as a hermaphrodite, producing many progeny from a single individual. A single hermaphrodite worm produces about 300 progeny. This large number of progeny combined with the short life-cycle (3.5 days at 20C) and ease of manipulation makes the worm an ideal organism for genetic studies.  2  1 Introduction The C. elegans genome is small, comprised of 1 X 10 8 nucleotides, of which 17% is repetitive DNA (Sulston, J. and S. Brenner. 1974; J. Sulston, personal communication). C. elegans has six chromosomes, or linkage groups (LG); five autosomes, LGI, LGII, LGIII, LGIV and LGV, and a sexdetermining chromosome, LGX. Minimum estimates of the total number of essential genes based on lethal saturation screens over large regions of the genome range from 2,000 (Brenner, 1974) to 6,000 (Moerman and Baillie, 1979; Rogalski and Baillie, 1985; Clark et al., 1988; Howell, 1989; McDowall, 1990; Johnsen and Baillie, 1991). Based on the estimated 8.3 X 10 7 by of unique sequence in the genome, the genetic estimates of gene numbers predict a gene density of 10 kilo bases (kb) per gene to 40 kb per gene. Molecular estimates based on cross species hybridization (Heine and Blumenthal 1986; Prasad and Baillie, 1989) estimate one conserved region per 10 to 15 kb. Computer analysis of DNA sequence data for a 120 kb region of LGIII identified one potential coding region per 3 to 4 kb. Based on this estimate of gene density, the total number of genes could be as high as 15,000 (Sulston et al., 1992). Thus, C. elegans is estimated to have one gene for every 2,000 to 15,000 nucleotides. A variety of mutations altering the development, behavior and appearance of the worm have been identified. Morphological mutations include dumpy (dpy), roller (rol), squat (sqt), long (Ion), small (sma), and blister (bli) (Brenner, 1974; Higgins and Hirsh, 1977; Cox et al., 1980; Kusch and Edgar, 1986). Dpy worms are short and fat (Brenner, 1974); Rol worms roll to the right or left due to the helical twisting of their cuticle (Higgins and Hirsh, 1977); Sqt mutants are dominant Rol and recessive Dpy (Cox et al., 1980; Kusch and Edgar, 1986); Lon worms are longer than wild type (Brenner, 1974); Sma are smaller; and Bli have fluid-filled cuticular swellings  3  1 Introduction (Brenner, 1974). More than 900 C. elegans genes have been identified by mutational analysis (Edgley and Riddle, 1990). 1.3 CUTICLE STRUCTURE C. elegans has a complex, developmentally regulated extracellular cuticle that functions as a hydrostatic exoskeleton (Figure 1). For this reason, the cuticle of C. elegans has been proposed as a model system for the study of the assembly, architecture and function of extracellular matrices (Higgins and Hirsh, 1977; Cox et al., 1980). Biochemical and ultrastuctural analyses have revealed that the cuticle is arranged in two layers, a basal layer and a cortical layer, and is composed of collagenous proteins that are extensively cross-linked by disulfide bonds, as well as other proteins that are resistant to collagenase (Cox, Kusch and Edgar, 1981). The structures of the layers vary with developmental stage (Cox, Staprans and Edgar, 1981), illustrated in Figure 1. The adult cuticle has an additional layer consisting of a fluid-filled space spanned by columnar structures termed struts connecting the basal and cortical layers (Cox, Kusch and Edgar, 1981). The C. elegans collagen gene family consists of 50 to 150 members, encoding small collagens of 30 to 40 kilodaltons (kd) that are covalently cross-linked in the cuticle (Kramer Cox and Hirsh, 1982; Cox, Kramer and Hirsh, 1984). Collagen gene expression varies with developmental stage (Cox and Hirsh, 1985; Kramer, Cox and Hirsh, 1985). Recently three genes that affect cuticle morphology, dpy-13 (von Mende et al., 1988), sqt-1 (Kramer et a!., 1988), and rol-6 (Kramer et al., 1990) have been cloned and shown to be collagen genes.  4  1 Introduction  Figure 1. Developmental changes in cuticle structure  Diagrammatic representation of cuticle structure from adults, L4, dauer, and L1 worms. The basal and cortical layers in adult cuticle are separated by a fluid-filled layer and connected by columnar structures called struts. C = Cortical layer; B = Basal layer; St = Strut; Fl = Fluid layer; Fb = Fibrillar layer; SL = Striated layer. After Cox, Staprans and Edgar, 1981.  5  1 Introduction 1.4 ORIGIN OF bli 4 ALLELES -  bli 4(e937) was induced with 32 P (Babu and Brenner, unpublished -  results, cited in Brenner, 1974). e937 was the only allele of bli 4 identified -  prior to this study. bli 4 is located in the cluster of genes on LGI, and was -  positioned between dpy 5 and dpy 14 in a region that is covered by the free -  -  duplication sDp2 (Figure 2).  sDp2 has been used as a balancer in lethal screens by the Rose lab (Howell et al., 1987; Howell, 1989; McDowall, 1990). More than 500 lethal mutations in the sDp2 balanced region have been isolated following EMS mutagenesis. A number of the LGI lethal mutations that failed to complement the blistered phenotype of e937 were identified by complementation analysis (Peters, McDowall and Rose, 1991). The first such lethal mutation was h42 (Howell et al., 1987). Subsequently, seven other  sDp2 balanced lethal mutations that failed to complement the blistered phenotype of e937 were identified by complementation analysis. The complementation tests that identified these mutations were performed by the individuals listed in Table 1. All of the lethal mutations that failed to complement e937 also failed to complement the sDp2 balanced mutation  h754, and the mutation s90, which was identified by Rose and Baillie (1980), and assigned the gene name let 77. s90 and h754 complement e937. One -  goal of my thesis work was to explain this complementation pattern.  6  ^  1 Introduction  Table 1. Alleles of bli-4 ^Phenotype^Origin Allele Adult blisters^Brenner, 1974 e937 late embryonic arrest Howell et al, 1987 h42 .^D. Pilgrim and A.M. Rose h199 ^A.M. Howell and A.M. Rose h254 ^McDowall, 1990 h384 h427 h520 h699 I. h754 h791 Ll arrest^ s90a ^Rose and Baillie, 1980 II  II  II^  II  II^  II  11^  11  II^  11  II  as90 was assigned to let-77 in Rose and Baillie, 1980.  7  1 Introduction  1 ^1 ^ I 1 ^ 1 ^I  Diu  hD•19 hDp16 sDp2  -dpy-5  -unc-4 0  -bli -4  - unc-37 -dpy-14  - unc-13  T 0 . 1 mu  I I  .1_  Figure 2. Genetic map of the bli-4 region A partial genetic map of the region of LGI around bli-4 illustrating map positions for closely linked visible markers and duplications. sDp2 was mapped by Rose, Baillie and Curran, (1984). hDp16 and hDp19 were generated and mapped by McKim and Rose, 1990, and further mapped by McDowall (1990). Some map data is from Edgley and Riddle (1990).  8  1 Introduction 1.5 RESOURCES USED TO CLONE bli-4 An important resource exploited in this thesis was the C. elegans physical map. The physical map consists of contiguous sets (contigs) of overlapping cosmid and yeast artificial chromosome clones (YACs) covering most of the C. elegans genome (Coulson et al., 1986; Coulson et al., 1988). Cosmid contigs have been aligned with the genetic map through the cloning and mapping of genetic markers by numerous labs. Strain-specific fragment length difference (RFLD) mapping has been widely used in C. elegans to position genetic markers with respect to molecular markers (examples are found in Rose et al., 1982; Baillie et a!., 1985; Williams et al., 1992). A contig spanning an interval that includes bli-4 was aligned with the genetic map through the mapping of a strain specific RFLD, hP5 (RFLD) (Starr et al., 1989). Mapping of bli-4 on the physical map was accomplished by mapping hP5 and a breakpoint of the deletion hDf8. This mapping contributed to the cloning of bli-4 by restricting the range of cosmids that could include the locus. The technique of transposon tagging was used to identify the b/i-4 coding region. Insertion of a transposable element into a locus of interest creates a mutation that can be identified by Southern hybridization analysis. C. elegans genes that have been cloned by this method include lin-12 (Greenwald, 1985) and unc-22 (Moerman, Benian and Waterston, 1986). Tc1 is a 1610 by transposable element in C. elegans (Emmons et al., 1983; Rosenzweig, 1983). Tc1 is not normally mobile in C. elegans strain N2 (Eidie and Anderson, 1985a), but transposes at a high rate in C. elegans strain BO (Moerman and Waterston, 1984; Eidie and Anderson, 1985b). Loci responsible for the mutator activity have been crossed into an N2 background (Mori, Moerman and Waterston, 1988). In this study, I used the  9  1 Introduction mutator locus mut-6(st702) to isolate a spontaneous allele of bli-4, h1010. h1010 provided a detectable rearrangement that contributed to the identification of the cosmid containing the bli-4 locus. 1.6 REVIEW OF DIBASIC ENDOPROTEASES As I will describe in chapter three, bli-4 encodes a KEX2-like endoprotease. Therefore, I introduce this class of proteins here. The existence of prohormone processing enzymes was inferred when it was discovered that some proteins, such as pituitary hormones (Chretien and Li, 1967) and insulin (Steiner et al., 1967; Chance, Ellis and Brommer, 1968), are synthesized as inactive precursors. Many secreted proteins that are excised by cleavage at pairs of basic residues have been described since that initial discovery (reviewed in Docherty and Steiner, 1982; Thomas, Thorne and Hurby, 1988). The enzymes that catalyze the proteolytic processing of secreted proteins are subtilisin-type serine endoproteases related to the yeast KEX2 gene product. kex2: The Saccharomyces cerevisiae KEX2 gene product, kex2, was the first eukaryotic pro-protein processing enzyme to be isolated, and remains the best characterized. As such, the biochemical properties of kex2 form a paradigm for all eukaryotic serine endoprotease processing enzymes. KEX2 was first identified in screens for yeast cells unable to secrete the K1 killer toxin, an M1 dsRNA virus encoded toxin that kills cells lacking the killer plasmid (Liebowitz and Wickner, 1976; Wickner and Liebowitz, 1976). The name KEX derives from killer expression. kex2 mutants were also found to be defective in the ability to secrete the mature alpha-factor (Liebowitz and Wickner, 1976), a mating pheromone that acts on a mating-type cells causing them to arrest at the G1 phase of the cell division cycle (reviewed in Fuller, Sterne, and Thorner, 1988). Both the K1 killer toxin and the alpha-factor are  10  1 Introduction synthesized as inactive precursor proteins, and are activated by series of processing steps beginning with an endoproteolytic cleavage at pairs of basic residues (reviewed in Fuller, Sterne, and Thorner, 1988). In kex2 mutants, this cleavage does not take place. Instead, the prepro-alpha-factor is secreted in a heavily glycosolated form. Cleavage of the alpha-factor is restored by replacing the defective kex2 mutant with the normal gene on a plasmid (Julius et al., 1984). KEX2 was cloned by complementation of the mutant phenotypes with plasmid DNA containing genomic fragments (Julius et al., 1984). The KEX2 gene product, kex2, is structurally related to the bacterial subtilisin-like serine proteases (Fuller, Brake and Thorner, 1986; Mizuno et al., 1988). Studies assaying the protease activity of kex2 using short synthetic substrates have revealed that the enzyme is membrane bound and dependent on calcium, and cleaves its substrates on the carboxyl side of Arg-Arg and LysArg pairs of basic residues (Julius, Sheckman and Thorner, 1984; Mizuno et al., 1987; Fuller, Brake and Thorner, 1989a). Considerable evidence demonstrates that kex2 is localized in the Golgi body. First, a short hydrophobic domain with the hallmarks of a signal peptide is found at the amino-terminal of kex2 (Fuller, Brake and Thorner, 1988). Second, the effect of conditional mutations that affect the movement of proteins through the Golgi body on pro-alpha-factor processing indicate that kex2 cleaves the pro-alpha-factor in a late compartment of the Golgi body (Novick and Scheckman, 1979; Julius, Sheckman and Thorner, 1984; Franzuoff and Scheckman, 1989). Finally, kex2 has been localized by direct immunolocalization to the Golgi compartment (Franzuoff et al., 1991; Redding, Holcomb and Fuller, 1991).  11  1 Introduction Retention of kex2 in the Golgi body requires a hydrophobic domain, thought to be a transmembrane domain (Fuller, Sterne and Thorner, 1988), near the carboxyl-terminal of the protein. Deletion of the hydrophobic domain results in the secretion of a soluble form of the protein, confirming that the hydrophobic domain is required for retention of kex2 in the Golgi body (Fuller, Brake and Thorner, 1989b; Brenner and Fuller, 1992). Retention of kex2 in the Golgi body also requires clathrin, suggesting that kex2 may be cycled through secretory vesicles or the cell surface (Payne and Scheckman, 1989). kex2 can function in mammalian cells. The properties of kex2 resemble closely those predicted for mammalian prohormone processing enzymes. To determine if kex2 could function as a prohormone processing enzyme, Thomas et al. (1988) tested the ability of kex2 to correctly process the mammalian prohormone pro-opiomelanocortin (POMC). When POMC and kex2 were coexpressed in cells lacking an endogenous processing activity, POMC was processed to produce mature gamma-lipotropin (gamma-LPH), beta-endorphin (beta-End), and beta-lipotropin (beta-LPH). Similar experiments demonstrated that kex2 can convert pro-beta-nerve growth factor (pro-beta-NGF) to mature beta-NGF (Bresnahan et al., 1990), and the single-chain zymogen form of human protein C to the mature twochain active form (Foster et al., 1991). These experiments demonstrate that the functional characteristics of the mammalian prohormone processing enzymes are sufficiently conserved that the yeast kex2 can replace their function in mammalian cell culture lines. Furin: A human homologue of KEX2, the human fur gene (hfur) was first partially sequenced because of its proximity to the fes oncogene (fur stands for fes upstream region) (Roebroek et al., 1986) and was subsequently  12  1 Introduction identified as a homologue of KEX2 by sequence similarity (Fuller, Brake and Thorner, 1989b; van den Ouweland et al., 1990). The structure of the fur gene product, hfurin, is similar to that of kex2. Like kex2, hfurin includes an amino terminal signal peptide and a carboxyl-terminal transmembrane domain (Roebroek et al., 1986; van den Ouweland et al., 1990). hfurin was localized by immunoflorescence to the Golgi body (Bresnahan et al., 1990). Hfurin has been shown in transfection studies to correctly process pro vonWillebrand factor (van de Ven et al., 1990; Wise et al., 1990) and probeta-NGF (Bresnahan et al., 1990). The fur mRNA has been found in cell lines derived from a wide variety of tissues including non endocrine cell types (Bresnahan et al., 1990). This broad spectrum of expression has led to the speculation that hfurin is a component of the constitutive (tissue general) secretory pathway. Genes with more than 90% sequence identity to human furin have been cloned from mouse (Hatsuzawa et al., 1990) and rat (Misumi, Sohda and lkehara, 1990). PC1 and PC2: Recently, candidates for the mammalian prohormone converting (PC) enzymes regulated (tissue specific) secretory pathway have been cloned using the polymerase chain reaction. These include a partial clone, mPC1, and the complete mPC2, cloned from a mouse pituitary cell line (Seidah et al., 1990). In addition, hPC2, a homologue of mPC2, was cloned from human insulinoma total RNA (Smeekens and Steiner, 1990). A complete mPC1 cDNA was subsequently cloned from the mouse AtT20 pituitary cell line and the sequence published as mPC3 (Smeekens et al., 1991). These proteins are similar to kex2 and hfurin in that they have a secretion signal peptide and a subtilisin-type protease domain that has significant sequence identity with hfurin and kex2. However, in contrast to kex2 and hfurin, these proteins lack a carboxyl-terminal transmembrane  13  1 Introduction domain. It is not yet known if these proteins are retained in the Golgi body by some alternative mechanism, or if they are secreted. PC1(PC3) and PC2 are transcribed in endocrine and neuroendocrine tissues only (Seidah et al., 1990; Smeekens et al., 1991). The tissue distribution of expression of these genes has lead to the suggestion that they may be involved in providing the tissue specificity of prohormone processing. This suggestion is supported by the finding that tissue specific processing of mPOMC can be reconstituted by coexpression of PC1(PC3) and PC2 either alone or in combination in tissue culture cells (Thomas et al., 1991; Benjannet et al., 1991).  bli-4. This study presents evidence that the bli-4 locus of C. elegans is a kex2-like protease. With the exception of the yeast protein kex2, no mutations have been identified in genes encoding the kex2-like proteins. kex2 loss of function mutants result in the inability to process the alphamating factor and the K1 killer toxin. Neither of these functions is essential. In contrast, the most severe lethal phenotype of bli-4 is developmental arrest as late embryos. This observation indicates an essential role for the  bli-4 gene product prior to or at the end of embryogenesis. This is the first identification of mutations in a KEX2-like gene in a multicellular organism, and the first direct evidence that a KEX2-like gene is essential to development.  14  CHAPTER TWO: MATERIALS AND METHODS 2.1 NOMENCLATURE Nomenclature used conforms to the uniform system for  Caenorhabditis elegans (Horvitz et al., 1979). Nomenclature for translocations conforms to that used by McKim, Howell and Rose (1988). Genes are represented by a three letter name followed by a number indicating the order in which they were discovered. For example, bli-4 is the fourth locus defined by blister mutations. Allele names are given as one or two lower case letters followed by a number. The letters are a laboratory designation, while the numbers indicate the particular mutation. For example, for the allele h42, the designation h indicates that the mutation was isolated in the Rose laboratory. The number 42 indicates that the mutation is the 42nd mutation isolated in the Rose laboratory. Strain names are given as one or two upper case letters followed by a number. As is the case for alleles, the letters are a laboratory designation, while the numbers indicate the particular strain. The Rose laboratory strain. designation is KR. To indicate rearrangements, the laboratory designation letter is followed by a uppercase and lower case letters indicating the nature of the rearrangement; T indicates a translocation; Dp indicates a duplication; Df indicates a deficiency (deletion); P indicates a restriction fragment length difference (polymorphism). For example, hDf8 is the eighth deficiency generated in the Rose laboratory. Phenotypes for a given mutation are presented as the gene name with no italics and starting with a capital letter. For example, bli-  4 is a gene name, while Bli-4 is a phenotype. Names that refer to DNA, such as gene names, allele names, translocation names, polymorphism names, and deletion names, are italicized. Names that refer to whole  15  2 Methods animals, such as strain names and phenotypes, are not italicized. In this study, plasmids are pCeh followed by a number, where p stands for plasmid, Ce stands for C. elegans, and h stands for the Rose laboratory. For example, pCeh205 is Rose laboratory C. elegans DNA plasmid 205. A list of all strains and mutations with laboratory designations used in this study is presented in Appendix B. 2.2 NEMATODE CULTURE CONDITIONS Worms were maintained and mated on 10 X 35 mm petri plates containing strain OP50 E. coil streaked on Nematode Growth Medium (NGM) (Brenner, 1974) at 20C except where noted otherwise. In experiments where counting or examining progeny was necessary, hermaphrodites were transferred to fresh plates every twelve to sixteen hours to prevent overcrowding. Nematode Growth Media is constituted as follows:  NGM medium: 300g NaCI  After autoclaving (20 minutes)  17g Agar  1 ml Cholesterol (5mg/m1 in Ethanol)  5g Bactotryptone  1 ml 1M CaCl2  Distilled water to 1 L  1 ml 1M MgSO4 25 ml 1M KH2PO4 (pH 6.0)  2.3 DETERMINATION OF PENETRANCE Penetrance was determined by scoring adult hermaphrodites for blisters. In populations of worms resulting from hermaphrodite selffertilization, penetrance was:  16  2 Methods Penetrance =^100% X [9* Blistered worms] [total # of worms] In populations of worms resulting from cross-fertilization of bii 4(e937) -  homozygous hermaphrodites with bli 4(e937)/ + heterozygous males, half -  the progeny was expected to be heterozygous for bli 4(e937). Because the -  males were bli 4 heterozygotes, a maximum of 50% of the worms were -  expected to blister if penetrance was 100%. Therefore penetrance in the cross experiment was defined as the percentage of blistered worms out of one half the total number of progeny:  Penetrance = 100% X 2[# Blistered worms] [total # of worms]  2.4 DETERMINATION OF LARVAL STAGE To follow larval moults, the method of Cassada and Russell (1975) was used. Synchronous populations of worms were established as follows. A population of hermaphrodites was permitted to lay eggs on a 10 cm petridish for a period of one to two days. All of the worms were then washed from the plate with M9 buffer, leaving unhatched eggs adhering to the agar surface. Newly hatched L1 larva were harvested at one-hour intervals and placed on a fresh plate. Each population of worms established this way was synchronous with respect to development to within one hour. C. elegans undergoes a period of reduced activity prior to each moult, termed the lethargus period. During this period, movement is reduced and pharyngeal pumping ceases. Lethargus periods were monitored at 25° by plotting the percentage of worms that were pumping in synchronous populations with  17  2 Methods respect to time. Moults were indicated by rapid and marked reductions in the percentage of worms exhibiting pumping. 2.5 COMPLEMENTATION TESTING  sDp2 lethal alleles. Lethal alleles rescued by sDp2 were complementation tested inter se as described by Howell et al. (1987). Heterozygous males of the genotype dpy 5 let X unc 13/ + + + were -  -  -  mated to hermaphrodites of the genotype  sDp2/ dpy 5 let Y unc 13/ dpy 5 let Y unc 13. sDp2 complements dpy 5 but -  -  -  -  -  -  -  not unc 13 (Figure 2). The absence of fertile Dpy-5 Unc-13 in the cross -  progeny indicated failure to complement.  sDp2 lethal alleles and e937. Complementation tests were done in both of the following ways: a) Heterozygous males of the genotype  dpy 5 let X unc 13/ + + + were mated to hermaphrodites of the genotype -  -  -  bli 4(e937) unc 13/ bii 4(e937) unc 13. The presence of Bli-4 Unc-13 males -  -  -  -  indicated failure to complement. b) Heterozygous males of the genotype  dpy 5 let X unc 131 + + + were mated to hermaphrodites of the genotype -  -  -  bli 4(e937)/ bli 4(e937). The presence of Bli-4 males in the cross progeny -  -  indicated failure to complement. In reciprocal crosses, both bli 4 unc 13/ + -  -  + males and b/i 4/ + males were mated to sDp2/ dpy 5 let X unc 13/ dpy 5 -  -  -  -  -  let X unc 13 hermaphrodites. The presence of Bli-4 Unc-13 or Bli-4 -  -  hermaphrodites and males indicated failure to complement.  let 77(s90) and bli 4(h1010). let 77(s90) and bli 4(h1010) were not -  -  -  -  linked to dpy-5, making it necessary to use a different complementation testing protocol from that used for the sDp2 balanced lethal alleles. s90 and  h1010 were each balanced by the translocation szT/(I,A), and complementation tests performed as follows: a) dpy 5 let X unc 13/ + + + -  18  -  -  2 Methods males were crossed to let-77 unc-13; +I szT1(1;X)[lon-2] or unc-63 bli-  4(h1010) unc-I 3; +1 szTI(I;X)[Ion-2.1 hermaphrodites. The absence of Unc 13 -  progeny indicated failure to complement. Successful mating was indicated by the presence of wild type males. b) Spontaneous Lon-2 males of the genotype let-77 unc-I 3; 01 szTI(I;X)[lon-2] or unc 63 bli-4(h1010) unc-I 3; -  0/ szTI(I;X)[Ion-2.1 were crossed to bli-4(e937) unc-13 hermaphrodites. The presence of non bli-4 Unc-13 male progeny and the absence of Bli-4 Unc-13 male progeny indicated complementation (Bli Unc hermaphrodites in this experiment could have resulted from self-fertilization). In reciprocal crosses, the presence of non bli-4 Unc-13 and the absence of Bli-4 Unc-13 hermaphrodite and male progeny indicated complementation.  hDf8. The deletion hDf8 was isolated using formaldehyde in a screen for mutations that failed to compelement dpy- 14. (McKim, Starr and Rose, 1992). hDf8 is associated with a suppressor of chromosome I recombination, and does not carry any flanking markers, making complementation tests difficult. To test hDf8 for complementation of the  bli - 4(e937) allele, the translocation szT1 was used (Fodor and Deak, 1985). Males of the genotype hDf8 (I); 01 szT1(1;X)[lon - 2] were mated to hermaphrodites of the genotype dpy - 5 bli - 4(e937). In this cross, the only males produced have the genotype dpy- 5 bli - 4(e937)/ hDf8. If hDf8 deleted  bli - 4, the non Lon-2 males would be blistered; if hDf8 did not delete bli - 4, the male progeny would be wild type. Only wild type male progeny were observed, indicating that hDf8 does not delete b/i-4. 2.6 DETERMINATION OF DEVELOPMENTAL ARREST STAGE The stage at which lethal homozygotes arrested development was determined. Several heterozygous hermaphrodites of the genotype  19  2 Methods  dpy 5 let X unc 13/ + + + were permitted to lay eggs on an NGM plate for -  -  -  a short period (not more than two hours) and then the homozygous lethal progeny were examined by Normarski differential interference microscopy for terminal phenotype. 2.7 ALLELE MAPPING  h42, h199, h254 and e937 were three-factor mapped by scoring segregation from strains bearing cis-linked flanking markers dpy 5(e61) and -  unc 13(e450) in trans to an unmarked chromosome. In the case of e937, -  recombinants were picked as Dpy-5 or Unc-13 worms and their progeny screened for the presence of the Bli-4 phenotype. Because the blistered phenotype is not expressed in Dpy-5 worms, the Dpy recombinants were tested for the presence of e937 by complementation testing. s90, which was not induced on a dpy 5 chromosome, was two-factor mapped with respect -  to unc 13(e51), consistent with a position between dpy 5 unc 13. -  -  -  Recombination was determined under the conditions recommended by Rose and Baillie, 1979a. Hermaphrodites were kept at 20C and were transferred to fresh plates every 12 to 16 hours to prevent overcrowding. Recombination frequency was calculated using the mapping function p=1-(1-2R) 1 /2 where R is the fraction of recombinant progeny over total progeny, and total progeny is calculated as 4/3(the number of wild type plus one recombinant class) (Brenner, 1974). The alleles h384, h427, h520, h699, h754 and h791 were mapped to an interval around bli 4 defined by hDp16 and hDp19 (Figure 2) by McDowall -  (1990).  20  2 Methods 2.8 SCREEN FOR bli 4 ALLELES IN A MUTATOR STRAIN -  mut 6 causes high levels of transposition of the transposable genetic -  element Tc1 (Mori, Moerman and Waterston, 1988). A mutator strain (KR1822) of the genotype unc 63(e384) unc 13(e450); mut 6(st702) was -  -  -  constructed (mut 6 was from RW7097, a strain obtained from D. G. -  Moerman and R. H. Waterston). Mutator activity in KR1822 was confirmed by screening in 1% nicotine for twitcher worms resulting from the insertion of Tc1 into the unc 22 gene as described by Mori, Moerman and Waterston -  (1988). KR1822 segregates spontaneous twitchers at a rate of 3 X 10 -4 . KR1822 was screened for spontaneous bli 4 alleles by mating KR1822 -  hermaphrodites to dpy 5(e61) bli 4(e937)1 + + heterozygous males, and -  -  screening the progeny for blisters. Three Bli worms, two hermaphrodites and one male, were identified after screening 82,300 chromosomes, an induction frequency of 3.6 X 10 -5 . Of the three spontaneous blistered animals recovered, one survived. The surviving hermaphrodite carried a  bli 4 lethal allele designated h1010 and was maintained using the -  translocation szT1(I;X) in the strain KR1858. 2.9 ISOLATION OF INTACT CUTICLE (Cox, Kusch and Edgar, 1981) Intact cutcles were isolated by sonicating worms for three minutes in 10 mM Tris (pH 7.4), 1mM EDTA, 1mM phenylmethanesulfonyl floride and incubating at 100C for 2 minutes in 1% SDS, 0.125 M Tris (pH 6.8). 2.10 COSMIDS AND PLASMIDS GROWTH CONDITIONS Plasmids were constructed using the Stratagene Bluescript SK vector, which carries an ampicillin resistance gene and a multiple enzyme recognition site polylinker for subcloning. Cosmids used in this study were obtained from A. Coulson and ). Sulston at the MRC, Cambridge, England.  21  2 Methods Cosmid names beginning with F, T, or K used kanamycin-resistant lorist 2 vectors (Cross and Little, 1986; Gibson et al., 1987). Cosmid names beginning with C, B, or Z used ampicillin resistant p)138 vectors (IshHorowicz and Burke, 1981). 2.11 DNA PREPARATION 2.11.1 COSMIDS AND PLASMIDS  Mini-preps (Maniatis, Fritsch and Sambrook, 1982). 5 mL of L-broth containing either 50 ug/ul ampicillin or 25 ug/ul kanamycin was inoculated with a single bacterial colony picked from an L-broth plate. The culture was incubated in a shaking incubator at 300 rpm at 37C for 16 to 18 hours. Cells were transferred to a 1.5 ml microfuge tube and harvested by pelleting for 20 to 30 seconds in an Eppendorf 5415 microcentrifuge. The pellet was resuspended in 300 uL of solution I and incubated for five minutes at room temperature. 400 uL of solution II was then added, mixed by vigorous inversion, and incubated on ice for five minutes. To this mixture 300 uL of ice-cold solution III was added, mixed by vigorous inversion, and incubated on ice for five minutes. Cellular debris were removed by pelleting at 12,000 rpm in an Eppendorf 5415 microfuge for five minutes. The aqueous layer was transferred to a fresh microfuge tube and extracted once or twice with 850 uL water-saturated phenol and once with Sevag's reagent (24:1 Chloroform:isoamyl alcohol). The DNA was precipitated by adding 850 uL isopropanol and pelleted at 12,000 rpm for 10 minutes. The DNA pellet was washed with 70% ethanol, dried in a vacuum chamber, and resuspended in 100 uL 1X TE containing 20 ug RNase A (BRL). Samples were stored at -20C.  22  2 Methods Solution I:^  Solution II  50 mM sucrose^  0.2 M NaOH  10 mM EDTA^  1% S DS  25 mM Tris (pH 8.0) 5 mg/mL lysozyme (Boehringer-Mannheim) freshly added before each use  Solution 111^  TE  3 M potassium acetate^  10 mM Tris base (pH 7.6)  11.6% glacial acetic acid^  1 mM EDTA  Sevag's reagent^24:1 Chloroform:isoamyl alcohol.  L-broth  ^  5 g NaCI  L-broth Plates  ^  5 g Difco Yeast Extract 10 g Bacto-tryptone  5 g NaCI  ^  ^  Distilled water to 1 L  ^  5 g Difco Yeast Extract 10 g Bacto-tryptone Distilled water to 1 L  Large scale plasmid and cosmid DNA isolation (Davis, Botstein and Roth, 1980). 10 mL of L-broth containing either 50 ug/ul ampicillin or 25 ug/ul kanamycin was inoculated with a single colony picked from an L-broth plate. The culture was incubated in a shaking incubator at 300 rpm at 37C for 16 to 18 hours, then transferred to 500 mL of fresh L-broth containing 50 ug/mL ampicillin or 25 ug/mI kanamycin and incubated overnight at 37C with shaking. Cells were harvested in 250 mL Beckman tubes by centrifugation at 5,000 rpm for 5 minutes in a Beckman )21 Centrifuge using a JA-14 rotor.  23  2 Methods The supernatant was discarded and the pellet resuspended in 5 mL of lysis buffer. The cell suspension was transferred to a 45 mL Beckman tube and incubated at room temperature for five minutes. 20 mL of solution II was added and mixed by vigorous inversion, followed by incubation on ice 5 minutes. 10 mL of ice-cold solution III was then added, mixed by vigorous inversion and incubated on ice for a further 5 to 10 minutes. The cellular debris were removed by centrifugation at 16,000 rpm for 20 minutes at 4C using a JA-20 rotor. 36 mL of the supernatant was transferred to a 50 ml falcon tube. 24 mL of isopropanol was added an mixed by gentle inversion and the mixture stored at -20C for 30 minutes. The DNA was collected by centrifugation at 2,000 rpm for 10 minutes using an 1EC HN-SII Desk top Centrifuge. The precipitate was rinsed twice with 70 0/0 ethanol and dissolved in 5 mL of 1X TE; 7.5 mL CsCI saturated solution; 0.65 mL ethidium bromide (10 mg/mL). The mixture was transferred to a 12 mL Beckman heat sealable tube, and centrifuged at 60,000 rpm for 18 hours at 15C in a Beckman ultracentrifuge. The DNA band was removed from the tube using a needle and transferred to a disposable plastic culture tube. Ethidium bromide was removed by repeated extraction with water-saturated butanol until the aqueous layer was colourless. DNA was precipitated from the aqueous layer by adding two volumes of water and six volumes of 95% ethanol, followed by incubation at -20C for one hour. The DNA was collected by centrifugation at 2,000 rpm for 10 minutes, and the pellet washed with 70% ethanol and air-dried. The DNA was resuspended in 1 mL of 1X TE, and stored at 4C.  24  2 Methods  Lysis Buffer^  Solution I  25 mM Tris (pH 8.0)^  3 M potassium acetate  10 mM EDTA (pH 8.0)^  11.6 % glacial acetic acid  50 mM sucrose 5 mg/mL lysozyme (Boehringer-Mannheim) added just prior to use. 2.11.2 C. ELEGANS GENOMIC DNA C. elegans genomic DNA was prepared using the method of Emmons, Klass and Hirsh (1979) as modified by J. Curran and D.L. Baillie (personal communication). Worm cultures were grown at 20C on a lawn of wild type  E. coil in 100 mm petri plates containing NGM (Section 2.2) made with 0.7% agarose instead of agar. Cultures were incubated for five to ten days until the plates were crowded with adult worms, but not starved. Worms were harvested from the plates by washing with 0.5% NaCI and transferred to 50 mL Falcon tubes. The worms were collected by centrifugation at 2,000 rpm for 10 minutes and the supernatant discarded. To eliminate bacteria, worms were rinsed one or more times with 10 mL of 0.5% NaCI until the supernatant was clear, and recentrifuged. Worms were resuspended in 5 mL of proteinase K buffer and incubated with 5 mg proteinase K at 65C for 10 minutes or until the worms were dissolved. A further 10 mL of proteinase K buffer was added to increase the volume to 15 ml. Proteins and cellular debris were removed by extraction of the aqueous solution with water-saturated phenol. The aqueous and organic phases were separated by centrifugation at 2,000 rpm for 10 minutes. The extraction was repeated until the interface was clear of cellular debris. The aqueous layer was then extracted twice with 15 ml of Sevag's reagent and collected by centrifugation at 2,000 rpm for 5 minutes. The solution was transferred to a fresh 50 ml  25  2 Methods falcon tube, and the DNA was precipitated by adding 1/10 volume of 4M NH4OAc and one volume of isopropanol and mixed by gentle inversion. Precipitated DNA was harvested from the solution with a sealed 1 mL Pasture pipette and washed with 70% ethanol. DNA was resuspended in 1 mL of TE containing 20 ug RNase A by gentle inversion overnight at room temperature and stored at 4C.  Proteinase K buffer 0.1 M Tris (pH 8.0) 0.05 M EDTA (pH 8.0) 0.2 M NaCI 1% SDS  2.12 ESTIMATION OF DNA CONCENTRATION The concentration of DNA in solution was determined by measuring the absorbance of light at 260 nM (A260) of 6 ul of the DNA solution diluted to 600 ul with distilled water. Absorption readings were taken in 1 ml cuvettes using a Perkin-Elmer Lambda-3 UV spectrophotomer, and the results read as optical density units (OD). For double stranded DNA, 1 OD = 50 ug/ml. For single stranded oligonucleotides, 1 OD = 20 ug/ml. 2.13 RESTRICTION DIGESTS Restriction endonucleases were obtained from Pharmacia, Bio-Rad, and Boehringer-Mannhiem. Digestion of DNA with restriction enzymes was carried out under the conditions recommended by the manufacturer.  26  2 Methods 2.14 GEL ELECTROPHORESIS Agarose gels were prepared by melting 0.7% agarose in TBE buffer. 1 ug/mI Ethidium Bromide was added to permit visualization of DNA. Gels were cast in horizontal trays of either 15 cm X 20 cm (LKB 20212 Maxiphor Electrophoresis unit) or 6 cm X 10 cm (Bio-Rad Mini-sub DNA Cell). DNA samples were mixed with 1/10th volume of loading buffer containing 30% glycerol and 0.25% bromophenol blue in water. Electrophoresis was conducted with the gel submerged in TBE buffer at 30 volts for 16 to 18 hours for large gels or 95 volts for 1 to 2 hours for small gels. DNA bands were visualized by illumination with a 300 nm UV transilluminator.  TBE Buffer 0.89 M Tris 0.89 M boric acid 1 mM EDTA (pH 8.0) 2.15 ELECTROELUTION (Maniatis, Fritsch and Sambrook, 1982) Electroelution was used to recover DNA fragments from agarose gels following electrophoresis. The DNA band to be recovered was visualized using the 300 nm UV transilluminator. The band was excised from the gel using a razor blade and placed into 10 mm diameter Spectra/por standard dialysis tube cut to a length of about 5 cm. The tube was sealed at one end with a plastic clip and filled with approximately 400 ul of 0.5 X TBE (Section 2.14) All air bubbles were removed and the tube was sealed at the other end. The dialysis tube containing the gel slice was then placed in a minigel apparatus submerged in 1 X TBE and 90 volts was applied for about one  27  2 Methods hour to elute the DNA from the gel slice. The eluted DNA was recovered from the dialysis bag using a micropipette. Preparation of dialysis tubing: Prior to use, dialysis tubing was boiled for 10 minutes in 2% sodium bicarbonate solution, rinsed thoroughly in distilled water, and then placed in distilled water and autoclaved for 10 minutes. Prepared tubing was stored in water at 4C. 2.16 SUBCLONING The vector Bluescript by Stratagene was used for all subcloning. The vector was prepared by restriction enzyme digestion followed by either heat inactivation of the enzyme (incubation at 65C for 15 minutes) or extraction with TE-saturated phenol followed by extraction with Sevag's reagent. Fragments to be subcloned (insert fragment) were digested with the appropriate restriction enzyme and purified by agarose gel electrophoresis through Bio-Rad low melting temperature agarose. Insert fragments were recovered from the gel by electroelution (Section 2.14). 1 ug of insert fragment was mixed with 100 ng of restriction enzyme digested vector and precipitated by adding 1/10 volume of 8 M NH4OAc and 2 volumes of ethanol. The mixed vector and insert fragments were then resuspended in 40 ul BRL ligation mix and 1 unit of BRL ligase. The ligation mixture was incubated overnight at 16C. 10 ul of the ligation mixture was used to transform BRL competent DH5-alpha E. coli cells. The ligation mixture was added to 100 ul of competent cells, incubated on ice for 30 minutes and then heated to 42C for two minutes. The transformed cells were plated on L-broth plates containing 50 ug/ml ampicillin, 40 ug/ml 5-bromo-4-chloro-3indolyl-beta-D-galactoside (Xgal) in dimethyl formamide and 160 ug/ml Isopropopyl-beta-D-thiogalactoside (IPTG), and incubated overnight at 37C.  28  2 Methods Colonies containing inserts were white; colonies without inserts were blue. Single white colonies were picked and screened for the correct insert by restriction analysis of miniprep DNA. Note: Bluescript does not express beta-galactosidase as well as some other cloning vectors, making blue colonies somewhat difficult to see. The blue colour is enhanced by further incubation at 4C for several hours.  1 X BRL ligation mix 50 mM Tris 10 mM MgCl2 0.1 mg/ml BSA 1 mM ATP 10 mM dithiothreitol 5% polyethylene glycol 2.17 SOUTHERN TRANSFERS (Southern, 1975) Agarose gels containing DNA fragments to be transferred to a membrane were prepared as described (Section 2.14). 100 ng to 500 ng of restriction digested plasmid or cosmid DNA, or three to five ug of restriction digested C. elegans DNA was loaded in each lane. The gel was treated by soaking in 0.25 M HCI at room temperature for 10 minutes, 1.5 M NaCI; 0.5 M NaOH for 30 minutes, and 4 M NH4OAc for 15 minutes. The gel was rinsed in distilled water between each treatment. The gel was then placed on a Whatman 3MM filter paper wick soaked in 10 X SSC and supported by a glass plate. A Schleicher and Scheull Nytran membrane cut to the size of the gel was soaked in 4 M NH4OAc laid onto the gel. All air bubbles between the gel and the Nytran were removed. Two pieces of Whatman  29  2 Methods 3MM filter paper cut to the size of the gel were soaked in 4 M NH4OAc and laid onto the membrane, and a 6 to 10 cm stack of paper towels were laid onto the filter paper. The DNA was transferred to the membrane by capillary action for about 2 hours. To prevent the 10 X SSC from bypassing the gel, the wick was prevented from touching the paper towels by placing strips of plastic around the gel. Following the transfer, the DNA was fixed to the membrane by baking at 80 to 90C for 1 to 2 hours.  10 X SSC 350 g NaCI 176 g sodium citrate Distilled water to 4 L 2.18 NORTHERN TRANSFERS Total RNA used was a gift from S. Prasad and D.L. Baillie, Simon Fraser University, Burnaby, BC. 15 ug of C. elegans RNA was denatured in 2.2 M formaldehyde; 50% deionized formamide; 1X MOPS for 15 minutes at 60C. The RNA fractionated by electrophoresis through a 1.1% agarose gel (15 cm X 20 cm) containing 1 X MOPS; 2.2 M formaldehyde (pH 7.0) at 80 volts for two to three hours. 1 X MOPS buffer was used as an electrophoresis buffer. After electrophoresis the gel was soaked twice for 15 minutes in distilled water. The RNA was then transferred to a GeneScreen membrane following the GeneScreen protocol. Two pieces of Whatman 3MM paper were wetted with phosphate buffer and placed over an elevated glass plate so that its ends formed wicks in phosphate buffer. The gel was placed on the filter paper and plastic strips placed on each side of the gel. A GeneScreen membrane was cut to the exact size of the gel and soaked for 20 minutes in phosphate buffer. The membrane was then placed gently onto the gel and  30  2 Methods all air bubbles removed. Five pieces of Whatman 3MM filter paper cut to the size of the gel were placed on top of the membrane, and covered with a 10 cm stack of paper towels. After 12 hours, membrane was removed and baked at 90C for two hours to fix the RNA to the membrane.  1X MOPS solution 0.2 M morpholinopropanesulfonic acid (pH 7.0) 50 mM sodium acetate 1 mM EDTA (pH 8.0)  Phosphate Buffer 0.025M Na2HPO4/NaH2PO4 mixed to pH 6.5 2.19 PREPARATION OF HYBRIDIZATION PROBES All hybridization probes were separated from their cloning vectors. This was accomplished by restriction enzyme digestion followed by purification by agarose electrophoresis and electroelution to recover the probe from the gel. Probes were made radioactive (labelled) with 32 P using the oligo labelling technique (Feinberg and Vogelstein, 1984). 30 ul of water containing 1 ng/ul DNA was boiled in a sealed microfuge tube for 10 minutes, then placed immediately on ice. To the DNA was added 10 ul of OLB-A, 5 ul of 100 mg/ml BSA, one unit of DNA Klenow (polymerase II large subunit) and 5 ul 32 P-ATP (3,000 Ci/mmol; 10.0 mCi/mL; New England Nuclear). The labelling mixture was incubated at 16C overnight. The mixture was then diluted to 100 ul and passed through a spun column prepared from Pharmacia G-25 fine Sephadex to remove unincorporated  31  2 Methods  32 P-ATP. The probe was boiled in a sealed microfuge tube for 10 minutes and then immediately placed on ice for five minutes prior to use.  OLB-A  Solution A  solution A:B:C (1:2.5:1.5)  1.0 mL 1.25 M Tris (pH 8.0)  18 uL 2-mercaptoethanol  0.125 M MgCl2  5 uL 0.1 M dTTP (Pharmacia) 5 uL 0.1 M dCTP (Pharmacia)  Solution B  5 uL 0.1 M dGTP (Pharmacia)  2 M Hepes (pH 6.6)  solution C Random hexanucleotides (Pharmacia) suspended in 1X TE at 90 OD/mL.  2.20 HYBRIDIZATION 2.20.1 Southern blots. Southern blot filters were placed in heat sealable bags with 12 to 15 ml of hybridization solution (5 X SSPE; 0.3% SDS) and the hybridization probe (Section 2.19). Air bubbles were removed and the bag sealed. Hybridization was at 68C for 18 hours with constant agitation. The filter washed in 200 ml of 0.2 X SSC; 0.2 `)/0 SDS twice at minutes at room temperature for five minutes per wash, and twice at 65C for 30 minutes per wash. The wash solution was changed after each wash. The filter was then air-dried prior to autoradiography.  32  2 Methods  20 X SSPE 174 g NaCI 27.6 g Na2HPO4 7.4 g EDTA Distilled water to 1 L  2.20.2 Northern blots. Northern blots were hybridized using the GeneScreen protocol. A radioactive probe was prepared as described (Section 2.19). The hybridization membrane was prehybridized by incubation with hybridization buffer in a heat sealed bag for six hours at 42C with constant agitation. The probe was denatured by boiling in 1.5 ml of the hybridization buffer and then added to the hybridization bag. The bag was resealed, and incubated overnight at 42C with constant agitation. The hybridization buffer was then removed, and the membrane washed twice for 5 minutes in 2 X SSC (Section 2.17) at room temperature, twice in 2 X SSC and 1% SDS for 30 minutes at 65C, and twice in 0.1 X SSC for 30 minutes at room temperature. The membrane was then air-dried and autoradiographed as described (Section 2.21).  Hybridization Buffer^  P Buffer  Made fresh just prior to use^1% BSA 5 ml deionized formamide^1% ficoll (M.W. 400,000) 2 ml P buffer^  250 mM Tris-HCL (pH 7.5)  2 ml 50% dextran sulfate^0.5% sodium pyrophosphate 0.58 g NaCI *^5%  SDS  1 ml 1 mg/ml denatured salmon sperm DNA  The solution is heated to 42C 10 minutes prior to the addition of NaCI.  33  2 Methods  2.21 AUTORADIOGRAPHY Probes bound to hybridization membranes were visualized using Kodak XAR5 or XRP X-ray film in cassettes using Dupont Cronex Lightning Plus or Dupont Par Speed enhancement screens. Autoradiographs of Southern blots of C. elegans genomic DNA and northern blots were exposed for one to two days at -70C. Autoradiographs of Southern blots of plasmid and cosmid DNA were exposed for 6 to 18 hours at room temperature. 2.22 CONSTRUCTION OF CB937 LAMBDA-ZAP LIBRARY A library of EcoRI digested CB937 genomic DNA in the Stratagene lambda-zap vector using a kit and protocol supplied by Stratagene. The vector was prepared as follows. First, the lamda cohesive ends (COS sites) were ligated. 10 ug lambda-Zap DNA was ligated in a 20 ul ligation mix overnight at 16C. The ligase reaction was stopped by incubating the mixture for 15 minutes at 68C. Next, the concatamerized vector was digested with EcoRl. 2.4 ul of EcoRI digest buffer and 2.0 ul of EcoRl (20 units) was added and the mixture incubated at 37C for one hour. The vector was then dephosphorylated with calf intestinal phosphatase (CIP). 5 ul CIP buffer and 1 ul CIP (0.00625 units) was added and the volume brought to 50 ul with distilled water. The dephosphorylation mixture was incubated for 30 minutes at 37C, and a second ul of CIP was added followed by a further 30 minute incubation. After the CIP treatment, 40 ul of water, 10 ul 10 X STE and 5 ul 10 % SDS was added. 105 ul phenol/Sevag's reagent was added, mixed by inversion, and separated by centrifugation. 90 ul of the aqueous phase was removed and the organic phase back extracted with 60 ul of TE.  34  2 Methods The resulting aqueous phases were pooled bringing the total aqueous volume to 150 ul. The aqueous solution was then re-extracted with 150 ul of phenol/Sevag's reagent and 135 ul of the aqueous phase removed. The organic phase was back extracted with 65 ul of phenol Sevag's reagent and the two aqueous phases combined to bring the volume to 200 ul. The aqueous phase was then extracted with 200 ul of Sevag's reagent and 185 ul of the aqueous phase removed. The vector was then precipitated by the addition of 18.5 ul of 8 M NH4OAc and 400 ul 95% ethanol, and harvested by centrifugation for 30 minutes at 4C. The vector DNA was resuspended in 5 ul 2 X ligation buffer. 1 ug of the insert DNA (KR1858 genomic DNA) was prepared by digestion with EcoRI in a volume of 20 ul followed by heat inactivation of the EcoRl. 2.5 ul of the insert mixture was mixed with 2.5 ul of the vector to bring the volume to 5 ul in 1 X ligation buffer. The ligation mixture was then incubated overnight at 16C. The library was then packaged using a Gigapack Gold lambda packaging extract. The library was amplified and titered prior to use.  Ligation Mix^  10 X Ligase Buffer  2 ul BSA (1 mg/ml)^  500 mM Tris-HCI (pH 7.5)  2 ul 10 mM ATP^  70 mM MgCl2  2 ul 100 mM dithiothreitol 2 ul 50% polyethylene glycol ^10 X CIP Buffer 2 ul ligase buffer^  500 mM Tris-HCI (pH 8.5)  1 ul T4 DNA ligase^  1 mM EDTA  Distilled water to 20 ul.  35  2 Methods  10 X STE^  Phenol/Sevag's reagent  1.0 M NaCI^  phenol:Sevag's reagent  100 mM Tris (pH 8.0)^  1:1  10 mM EDTA  2.23 cDNA ISOLATION cDNA clones were isolated from a C. elegans cDNA library constructed by Barstead and Waterston (1989) in the Stratagene lambda-zap vector. 40,000 phage from the cDNA library were screened using the h1010 probe (1.3 EcoRl fragment of pCeh181) as described (section 2.24). 2.24 LAMBDA-ZAP PHAGE LIBRARY SCREENING Host bacteria E. coil strain BB4 was grown in a 4 ml culture in TB broth at 37C overnight in a shaking incubator. The host bacteria were harvested by centrifugation at 2000 rpm for 10 minutes and prepared for use by resuspension in 2 ml 10 mM MgCl2. For each phage plate to be prepared, 5 X 10 4 phage diluted in SM buffer were incubated with 200 ul of BB4 cells for 15 minutes at 37C. The phage were then mixed with 3 ml of molten top agarose at 50C and poured onto 100 mm petri plates containing L-broth agar prewarmed to 37C. The phage plates were incubated at 37C of 16 hours. The plates were then cooled to 4C for two hours. The phage were transferred to 85 mm S&S NC nitrocellulose filters by laying the dry filters onto the bacterial lawn containing the phage plaques. The filters were then removed and soaked in 0.5 M NaOH; 1.5 M NaCI for 5 minutes, then 1 M NH4Ac for 5 minutes. The treated filters were air-dried and then baked for one hour at 80C. Filters were hybridized and washed as described (Section 2.20.1), and positive plaques identified by autoradiography (Section  36  2 Methods 2.21). Each hybridizing (positive) plaque was isolated from the phage plate by picking with a sterile toothpick and placed in SM buffer. The screen was then repeated twice more for each positive phage as described above, except that only 200 phage were plated on each plate. After three screens, all of the phage in a given stock were positive. One positive phage from each stock was then used to isolate the phage insert (section 2.25)  Top Agarose  SM Buffer  5 g NaCI  0.1 M NaCI  5 g yeast extract  0.01 M MgCl2  10 g bacto-tyrptone  0.05 M Tris (pH 7.5)  7.5 g Agarose  0.01% gelatin  Distilled water to 1 L 2.25 ISOLATION OF PLASMIDS FROM LAMDA-ZAP PHAGE Plasmids were isolated from lambda-zap phage stocks using the StrataGene protocol. E. coli strain BB4 cells were prepared as described (Section 2.24). 1 X 10 5 phage containing the desired insert were mixed with 1 X 10 7 R408 helper phage and added to 200 ul of BB4 cells. The mixture was incubated at 37C for 15 minutes and then 5 ml of 2 X YT was added followed by a further incubation of four hours at 37C in a shaking incubator. The cells were then killed by incubation at 70C for 20 minutes and removed from the solution by centrifugation at 2000 rpm for five minutes. The phage insert fragment was then contained in the supernatant in the Bluescript vector packaged in F1 phage particles. To recover the plasmid from the phage particles, 200 ul of the supernatant solution was mixed with 200 ul of BB4 cells and incubated at 37C for 15 minutes. 10 ul of cells were then  37  2 Methods plated on L-broth plates containing 50 ug/ml ampicillin and incubated at 37C overnight. The resulting bacterial colonies contained the insert in the Bluescript vector.  2 X YT 5 g NaCI 5 g yeast extract 8 g bacto-tryptone  2.26 DELETIONS Nested sets of plasmid insert deletions were constructed using the method of Henikoff (1984). First, the plasmid was digested with two restriction enzymes, BstXI and Xbal. The five base 3' overhang generated by  BstXl digestion is not a substrate for Exonuclease III (Exo III), and therefore protects the vector from deletion. The four base 5' overhang generated by Xbal digestion is a substrate for ExoIII digestion, and therefore permits Exo III to delete the insert. Both enzymes digest in Boehringer-Mannheim H buffer. However, BstXI requires 45C while Xbal requires 37C. 5 ug of plasmid DNA was added to 5 ul of 10 X H buffer, 5 ul of 1 mg/ml BSA, and 5 units of BstXI, brought to 50 ul with distilled water, and incubated at 45C for one hour. 5 units of Xbal was then added and the mixture incubated at 37C for one hour. The digestion mixture was then extracted with 50 ul of phenol/Sevag's reagent (1:1), the aqueous phase removed to a fresh microfuge tube and extracted with Sevag's reagent only. The aqueous phase was removed to a fresh microfuge tube and the DNA precipitated by the addition of 1/10th volume of 2 M NaCI and two volumes of 95% ethanol  38  2 Methods followed by incubation at -20C for 20 minutes. The DNA was pelleted in an Eppendorf 1514 microfuge at 12,000 rpm for 10 minutes, washed with 70% ethanol and dried under vacuum. The DNA pellet was resuspended in 60 ul of Exo Ill buffer and placed in a 37C heating block. 300 units of Exo Ill was added. At 30 second intervals, 2.5 ul aliquots of the Exo Ill digest mixture were removed and placed in individual tubes containing 7.5 ul of S1 mix on ice. After all the aliquots had been removed, the samples were incubated at room temperature for 30 minutes. S1 stop buffer was then added and the tubes heated to 70C for 10 minutes to inactivate the S1 nuclease. The tubes were then transferred to a 37C heating block, 1 ul of Klenow mix was added to each tube and the samples incubated for three minutes. 1 ul of dNTP mix was added and the samples incubated for a further five minutes at 37C. The samples were removed from the heating block, 40 ul ligation mix was added and the tubes incubated at 16C overnight. Deletion samples pooled in groups representing four adjacent time points and used to transform BRL DH5-alpha cells as described (Section 2.16). DNA prepared from bacterial cultures derived from individual bacterial colonies was extracted by miniprep (Section 2.11.1) and screened for deletion size by restriction digestion and agarose gel electrophoresis (Section 2.14).  10 X Exo III Buffer  7.4 X S1 Buffer  60 mM Tris (pH 8.0)  0.3 M potassium acetate (pH 4.6)  6.6 mM MgCl2  2.5 M NaCI 10 mM ZnSO4 50% glycerol  39  2 Methods  S1 Mix  S1 Stop Buffer  172 ul deionized water  0.3 M Tris base  27 ul 7.4 X S1 buffer  0.05 M EDTA  60 units S1 nuclease  1 X Klenow Buffer  Klenow Mix  20 mM Tris-HC1 (pH 8.0)  30 ul 1 X Klenow buffer  100 mM MgCl2  5 units Klenow  10 X Ligase Buffer  Ligase Mix  500 mM Tris-HCI (pH 7.6)  790 ul deionized water  100 mM MgCl2  100 ul 50% PEG  10 mM ATP  10 ul 100 mM dithiothreitol 5 units T4 DNA ligase  2.27 DNA SEQUENCING All sequence reactions were performed using sequencing kits and protocols from Applied Biosystems Inc (ABI) and an ABI- model 373A sequencing machine. The sequence methods used are based on the linear amplification technique of Craxton (1991) and the dideoxy-terminator method (Sanger and Coulson, 1975; Sanger, Nicklen and Coulson, 1977). ABI kits use florescent dyes covalently linked to either sequencing primers or dideoxynucleotide terminators. Reactions terminating in ddT, ddA, ddG and ddC were loaded into a single lane on a poly-acrylamide gel in the 373A sequencer, and the base order determined by fluorescence emitted from the dye as the band passed a scanning laser.  40  2 Methods 2.27.1 Template preparation. DNA templates were prepared using a QIAGEN miniprep kit. A bacterial culture was prepared as described (Section 2.11.1). 1.5 ml of the bacterial culture was harvested by centrifugation, resuspended in 300 ul of buffer P1, and incubated at room temperature for five minutes. 300 ul of buffer P2 was added and mixed by gentle inversion, and incubated at room temperature for a further five minutes. 300 ul of buffer P3 was then added, mixed by gentle inversion and the cellular debris removed by centrifugation in an Eppendorf 1514 microfuge for 15 minutes at 12,000 rpm at 4C. The supernatant was applied to a QIAGEN column, and the eluate precipitated with 0.7 volumes of isopropanol. The precipitate was washed with 70% ethanol, dried briefly under vacuum, and resupended in 20 ul TE (Section 2.11.1). The concentration of the DNA was estimated as described (Section 2.12). QIAGEN columns are anion exchange columns that were used as follows. The column was washed with 1 ml buffer QBT. The sample was then applied and allowed to enter the resin by gravity flow. The bound sample was washed twice with buffer QC. Finally, the DNA was eluted with 0.8 ml of buffer QF. Buffer P1  Buffer P2  Buffer P3  50 mM Tris-HCI  200 mM NaOH  2.55 M KOAC  10 mM EDTA  1% SDS  100 ug/ml RNase A pH 8.0  41  2 Methods  Buffer QBT  Buffer CF  Buffer QF  750 mM NaCI  1.0 M NaCI  1.5 M NaCI  50 mM MOPS  50 mM MOPS  50 mM MOPS  15% Ethanol  15% Ethanol  15% Ethanol  0.15% Triton X-100  pH 7.0  pH 8.2  pH 7.0 2.27.2 Dye-labelled primer sequencing. 1 - 1,5 ug of double stranded template DNA in 6.0 ul of TE was used for each set of reactions. Four 0.5 ml microfuge tubes were labelled A, C, G and T. The fluorescent dyes used with the A and C mixes were more fluorescent than those used with the G and T mixes. To compensate, the G and T reactions were doubled. One ul of each of the following reagents was added to A and C tubes, and two ul to G and T tubes: d/ddNTP mix; dye primer (0.4 pM/u1); 5 X cycle sequence buffer; DNA template; diluted Taq polymerase. The reactions were overlaid with 20 ul of mineral oil to prevent evaporation, and the tubes placed in a Cetus-Perkin-Elmer thermocyler preheated to 95C. Cycling was as follows: 15 cycles with 95C for 30 seconds, 55C for 30 seconds, and 70C for 60 seconds. This was followed by 15 cycles with 95C for 30 seconds and 70C for one minute. The samples were then removed from the tubes. Mineral oil was removed by rolling the samples on parafilm. The samples were then pooled in a microfuge tube containing 100 ul of 95% ethanol and 2 ul of 3 M sodium acetate and incubated for 15 minutes at room temperature. The reaction products were pelleted by centrifugation for 30 minutes at 12,000 rpm in an eppendorf 1514 microfuge at 4C, washed with 70% ethanol and dried briefly under vacuum. One hour prior to loading, the samples were resuspended in 4 ul  42  2 Methods of deionized formamide/ 50 mM EDTA 1:1 (v/v). Immediately before loading, the samples were heated to 90C for two minutes and quick-chilled on ice.  d/ddNTP Mixes d/ddA Mix  d/ddC Mix  1.5 mM ddATP  0.75 mM ddCTP  62.5 uM dATP  250 uM dATP  250 uM dCTP  62.5 uM dCTP  375 uM c 7 dGTP  375 uM c 7 dGTP  250 uM dTTP  250 uM dTTP  d/ddG Mix  d/ddT Mix  0.125 mM ddGTP  1.25 mM ddTTP  250 uM dATP  250 uM dATP  250 uM dCTP  250 uM dCTP  94 uM c 7 dGTP  375 uM c 7 dGTP  250 uM dTTP  62.5 uM dTTP  Dye-labelled primers Forward primer: 5' GTAAAACGACGGCCAGT 3' Reverse primer: 5' AACAGCTATGACCATG 3'  ^ 5 X Cycle Sequencing Buffer Diluted Taq Polymerase ^ 400 mM Tris-HCI (pH 8.9) 0.5 ul AmpliTaqR (8 units/u1) 100 mM (NH4)2 SO4 25 mM MgCl2  ^  ^  43  1.0 ul 5 X Cycle sequencing buffer  5.5 ul distilled Water  2 Methods 2.27.3 Dye-labelled terminator sequencing One ug of double-stranded DNA template was mixed with 7.25 ul reaction premix and 3.2 pmol of the sequencing primer in a 0.5 ml microfuge tube and brought to a volume of 20 ul with distilled water. The reaction mixture was overlaid with 40 ul of mineral oil and placed in a Cetus-Perkin-Elmer thermocycler preheated to 90C. The incubation temperatures were cycled as follows: 96C for 30 seconds, 50C for 15 seconds, and 60C for 4 minutes for 25 cycles. The samples were then removed from the tubes. Mineral oil was removed by rolling the reaction mix on parafilm. The sample was then loaded onto a Select D-50 spun column (5 Prime -> 3 Prime Inc) and eluted by centrifugation at 1000 X g for 2 minutes (2400 rpm, IEC HN-SII Desk top Centrifuge). The eluted sample was then precipitated by the addition of 40 ul of 95% ethanol and 0.3 M sodium acetate followed by incubation for 15 minutes at room temperature. The reaction products were pelleted by centrifugation for 30 minutes at 12,000 rpm in an eppendorf 1514 microfuge at 4C, washed with 70% ethanol and dried briefly under vacuum. One hour prior to loading, the samples were resuspended in four ul of deionized formamide/ 50 mM EDTA 1:1 (v/v). Immediately before loading, the samples were heated to 90C for two minutes and quick-chilled on ice. Reaction Premix (4 Reactions) ^  5 X TACS Buffer  16 ul 5 X TACS Buffer^  400 mM Tris-HCI  4 ul dNTP mix^  10 mM MgCl 2  2 ul DyeDeoxyTM A Terminator^100 mM (NH4)2SO4 2 ul DyeDeoxyTM C Terminator^pH 9.0 2 ul DyeDeoxyTM G Terminator 2 ul DyeDeoxyTM T Terminator 1 ul AmpliTaqR DNA Polymerase  44  2 Methods  dNTP Mix^  Abbreviations  750 uM dITP^  dNTP Deoxynucleoside triphosphate  150 uM dATP^ ddNTP Dideoxynucleoside triphosphate 150 uM dTTP^ c7dGTP 7-deazaguanosine triphosphate 150 uM dGTP^ ATP Adenosine triphosphate CTP Cytidine triphosphate TTP Thymidine triphosphate ITP Inosine triphospate  Custom Sequencing Primers Numbers in brackets indicate the priming position in the blisterin B cDNA (Section 3.2.2.2). Sense indicates that the primer is derived from the sense strand sequence; antisense indicates that the primer is derived from the antisense strand. KRp5  5' TGCTGGTTGACGGAAATC 3' (1349, antisense)  KRp6  5' CTACTCGGCTACTCCTGC 3' (2103, sense)  KRp7  5' TCCTTTTCCACCTCTGCC 3' (752, antisense)  KRp8  5' GAAGGCAACACCGACACC 3' (950, antisense)  KRp9  5' TCCACCAACTGCTCCACC 3' (692, antisense)  KRp10  5' ACTCTCTTCTTCGGTCGC 3' (499, antisense)  KRp11  5' GTGTCCTTGTTGTTTCCG 3' (414, antisense)  KRp13  5' TGGTGGAGCAGTTGGTGG 3'  (708, sense)  2.28 POLYMERASE CHAIN REACTION (Saki et al., 1988) PCR was performed using Perkin-Elmer-Cetus thermocyler and reagent kit. Cycling was performed as follows: 92C 60 sec, 58C 1 60 sec, 72C 60 sec, and 25 cycles. The reagents were mixed as listed below, and overlayed with 40 ul of mineral oil. Because KR1858 was heterozygous for the amplification target DNA, the actual quantity of template was 50 ng. The primers used were KRp13 (Section 2.27) and P618, a Tc1 specific primer  45  2 Methods (Williams et al., 1992). P618 was a gift from D. Moerman, University of British Columbia, Vancouver, BC. The sequence of P618 is: P618 5'- GAA CAC TGT GGT GAA GTT TC - 3' (Tc1 specific).  PCR Reaction mix (100 ul)  10 X Reaction buffer  100 ng Template DNA  100 mM Tris HCI (pH 8.3)  10 umoles P618  500 mM KCI  10 umoles KRp13  15 mM MgCl2  4 uM dATP  0.1% (W/V) Gelatin  4 uM dCTP 4 uM dTTP 4 uM dGTP 1 X Reaction Buffer 1 unit AmpliTaqR DNA Polymerase 2.29 SEQUENCE ANALYSIS Sequences were assembled using the Delany Sequence program to identify overlaps and the Eye-ball Sequence Editor ESEE by Eric Cabbot to edit and assemble sequences. Kyte-Doolittle hydropathy analysis (Kyte and Doolittle, 1982) was done using the GREASE program (Pearson and Lipman, 1988). Potential glycosolation sites were identified using the prosite program of PCGene. Searches of the Swissprot protein database was done using the FASTA algorithm with a ktup value of 2 (Pearson and Lipman, 1988).  46  CHAPTER THREE: RESULTS The results section of this thesis is divided into two major sections: the first section will report the results of a genetic analysis of bli-4; the second section will report results of a molecular analysis of bli-4.  3.1 GENETIC ANALYSIS OF THE bli-4 LOCUS This section begins with a characterization of the blistered phenotype resulting from the e937 allele of bli-4. This characterization includes a determination of where blisters form, determination of penetrance, when blisters form, and how blisters are affected by other mutations that affect cuticle morphology. Next, the complementation patterns of bli-4 alleles is examined, the arrest point of b/i-4 lethal alleles is determined, the map position of bli-4 alleles is determined, and finally, the isolation of a mutatorinduced allele is described. 3.1.1 Characterization of e937 3.1.1.1 Blisters form between cuticle layers  bli-4(e937) is a recessive mutation that results in fluid-filled blisters in the cuticle of adult-stage worms (Figure 3). The cuticle of larval-stage worms is not affected in a way that is observable under the dissecting microscope or by Normarski optics. No attempt at ultrastructural or biochemical characterization of larval-stage cuticles of bli-4 homozygote worms was made in this study. Therefore, the possibility of subtle alterations in larval cuticle by bli-4(e937) is not excluded. Under Normarski optics, blisters appeared to be a separation between the adult cuticle layers, rather than between the hypodermal tissue and the cuticle. The hypothesis that blisters result from a separation of cuticle layers predicts that blisters would remain intact in isolated cuticles. To test this  47  3 Results prediction the cuticle of blistered adult hermaphrodites was isolated as described in Section 2.9. Figure 4 shows the appearance of a blister in an adult hermaphrodite (A) compared to one in isolated cuticle (B). Visible in the isolated cuticle are some internal cuticle structures, including the pharynx and the gut, eggs, which are resistant to the treatment used to isolate the cuticle, and an intact blister. This experiment supports the conclusion that the blisters occur between the cuticle layers.  48  3 Results  A  Figure 3. Blistered adult CB937 bli 4(e937) worms -  A. Blister on adult male tail. B. Blister on adult hermaphrodite.  49  3 Results  B  Figure 4. Blistering in a whole worm and isolated cuticle A. Intact blistered CB937 adult hermaphrodite. B. Isolated blistered cuticle from a different CB937 hermaphrodite. Eggs and collagenous pharynx and gut lining remain intact in this preparation.  50  3 Results 3.1.1.2 Blistering is incompletely penetrant and is not temperature sensitive Blistering in CB937 worms is incompletely penetrant, and expressivity is variable. Penetrance is defined as the portion of mutant homozygotes in a population that are blistered out of the total number of mutant homozygotes in the population. Expressivity is defined as the extent to which an affected individual expresses the phenotype. To determine the penetrance of blistering in CB937 worms, adult hermaphrodites were scored for blisters. Of these, 1043 of 1092 (95.5%) were blistered at 20C. To determine if blistering is temperature sensitive, this experiment was repeated at 15C. 1706 out of 1811 (94.3%) adult hermaphrodites were blistered at 15C, indicating that blistering is not temperature sensitive within the 15 - 20C range. No experiments were done to quantify the expressivity of the blistered phenotype. However, anecdotally, in every population of blistered worms that were examined, expressivity varied from very small localized blisters to very severe blisters covering most of the animal. In general, in genetic backgrounds that enhance penetrance, blister expressivity in e937 worms is increased. In genetic backgrounds that reduce penetrance, blister expressivity is decreased. 3.1.1.3 Blistering is adult specific To determine exactly when during development e937 worms first blister, the method of Cassada and Russell (1975) was used to follow moulting as an indicator of developmental stage. Wild type worms go through four larval stages prior to maturing to adulthood and express the adult cuticle after the fourth moult. The larval to adult moult occurs at 35.5 hours after hatching at 25C (Wood et al., 1980). C. elegans undergoes a period of reduced activity prior to each moult, termed the lethargus period.  51  3 Results During this period, movement is reduced and pharyngeal pumping ceases. Lethargus periods in bli 4(e937) hermaphrodites were monitored at 25C by -  plotting the percentage of worms that were pumping in synchronous populations with respect to time (Figure 5A). The blistered phenotype is adult specific: larval stage worms did not blister. e937 hermaphrodites first expressed blisters about 2 hours after the adult moult at 25C. e937 worms reached the adult moult at 46 hours after hatching, 11 hours later than wild type worms. Thus, although e937 lacks visible effects on larval-stage worms, it slows growth by about 30%. Experiments were performed to determine if the adult specificity of blistering in e937 worms requires expression of adult cuticle or other adult-specific structures. For example, would blistering be expressed early if the adult cuticle is expressed early? Would blistering not be expressed if adult-stage worms expressed a larval cuticle or failed to express an adult cuticle? To address these questions, mutants of two heterochronic genes,  lin 29 (III) and lin 14 (X), that affect the timing of the adult cuticle were used. -  -  lin 29(n1440) causes reiteration of larval-stage cuticle in adults (that is, adult -  worms do not express the adult cuticle). lin 14(n179ts) causes precocious -  expression of the adult cuticle (ie adult-stage cuticle is expressed at larval stage four). The interaction of e937 with mutations of the heterochronic genes lin 29(n1440) and lin 14(n179ts) was determined. -  -  lin 29 loss-of-function alleles fail to make the L4 to adult cuticle -  switch, and reiterate the L4 stage cuticle, causing the animals to undergo extra moults (Ambros and Horvitz, 1984). This is the only known effect of  lin 29 mutations. I predicted that mutations in lin 29 would suppress -  -  blistering if expression of the e937 phenotype requires an adult cuticle. A  bli 4(e937); Iin 29(n1440) double mutant was constructed and screened for -  -  52  3 Results the expression of blisters. bli 4(e937); lin 29(n1440) worms did not express -  -  blisters: 0/1160 Fl progeny of bii 4; lin 29 hermaphrodites expressed blisters -  -  at any age. While the possibility that lin 29 has effects other than the simple -  reiteration of L4 cuticle cannot be ruled out, it is most likely that blisters could not form in Lin-29 hermaphrodites because they lacked an adult cuticle. This observation supports the conclusion that blisters cannot form in adult worms not expressing the adult cuticle. The lin 14 loss-of-function allele n179ts results in the precocious -  expression of the adult cuticle after the third moult at the restrictive temperature of 25C (Ambros and Horvitz, 1984; Ambros and Horvitz, 1987). Hermaphrodites undergo a fourth moult producing a second adult cuticle. If the expression of the blistered phenotype requires an adult cuticle, then the precocious expression of the adult cuticle in n179 worms was predicted to permit blistering one moult earlier than in wild type. A  bli 4(e937); lin 14(n179ts) double mutant was constructed. Lethargus -  -  periods in bli 4(e937); lin 14(n179ts) hermaphrodites at 25C were -  -  determined by plotting the percentage of worms that were pumping in synchronous populations with respect to time (Figure 5B). The interaction of  bli 4 and lin 14 was unexpectedly complex. At the restrictive temperature of -  -  25C, most bli 4; lin 14 animals were sterile. 20% of these worms arrested -  -  development prior to reaching the fourth moult. In addition, the rate of growth of the double mutant was variable. Consequently, synchronous populations quickly ceased to be synchronous. This may be seen by comparing the graph in Figure 5A with that in Figure 5B. Sterility, variable growth rates and variable larval arrests are not characteristics of either  lin 14(n179) or bli 4(e937) alone. Therefore, bli 4(e937) and lin 14(n179) -  -  -  -  produce an incompletely penetrant synthetic lethality in the double mutant.  53  3 Results From the data presented in Figure 5B, it appears that blistering in the  bli 4; lin 14 double mutant did not occur until after the fourth moult. All of -  -  the blistered worms appeared to be adults based on size. Moreover, no blistered worms were observed undergoing a lethargus period or moulting. Thus, lin 14(n179) does not seem to alter the expression of blistering with -  respect to the number of moults. The lack of blistering in 1in 14(n179ts) bli-  4(e937) L4 worms may indicate that the presence of the adult cuticle in L4 e937 animals is not sufficient for the expression of blisters. However, it is also possible that the cuticle expressed by lin 14(n179ts) worms is not a wild -  type adult cuticle.  54  •  3 Results A. bli-4(e937) 100  P 80  C E  N  T  80  40  M P •  20  G O  6 12 18 24 30 36 42 48 54 60 66  AGE (HOURS) • % Pumping  ^  % Blistered  B. bli-4(e937);lin-14(n179ts) 100  •  80  C  N  60  40  M P •  20  G O  6 12 18 24 30 36 42 48 54 60 66  AGE (HOURS) • % Pumping  ^  % Blistered  Figure 5. Developmental stage of blistering  Lethargus periods and expression of blistering in A. Bli-4, and B. Bli-4; Lin14 hermaphrodites. Synchronous populations of several hundred animals were hatched at time zero (+1- 1 hr) and grown on NGM plates at 25C using the method of Cassada and Russell (1975). At frequent intervals, 50 animals were observed for 5 seconds and the percentage with pharyngeal pumping recorded. 55  3 Results 3.1.1.4 Interaction of e937 with cuticle genes Roller, squat and some dumpy phenotypes are generally epistatic to blistering in BIi-1 and Bli-2 mutants (Higgins and Hirsh, 1977; Cox et al., 1980). It was observed that Dpy-5 also has a dominant effect on blistering. To determine the effect of Dpy mutations, double mutants of Bli-4 and 11 Dpy mutations were constructed. The penetrance of blistering in e937 homozygotes was determined in dpy X homozygous and heterozygous -  backgrounds (Table 2). The average body length for each Dpy mutant is presented as a measure of Dpy severity. Non-Dpy Bli-4 length is about 1.2 MM.  In non-Dpy Bli-4 animals, blister penetrance is 95%. In Dpy homozygotes, blistering was completely or almost completely suppressed by 9 of the 11 Dpy mutations. The exceptions were Dpy-14 and Dpy-17, in which blister penetrance was reduced to 41% and 54% respectively.  dpy 14(e188) and dpy 17(e164) are two of the least severe Dpy mutations -  -  (Table 2). This raises the possibility that Bli-4 penetrance in Dpy worms is related to Dpy severity. However, two other mild dpy mutations tested,  dpy 18(e364) and the e24 allele of sqt 3, did not permit the expression of -  -  blisters. In contrast, the el allele of dpy 1, one of the most severe Dpy -  mutations tested, produced three Dpy Bli worms. Thus, Bli-4 penetrance does not correlate with severity of the Dpy phenotype. To determine the magnitude of blister suppression by Dpy heterozygotes, bli 4; dpy-X double mutants were crossed to bli 4/ + males. -  -  Because the males were heterozygous, only half of the cross progeny were homozygous for bli 4. non Bli-4 worms could either be bli 4 heterozygotes -  -  or bli 4 homozygotes that did not express blistering. To compensate for b1i-  4 heterozygotes, the penetrance was determined by multiplying the  56  3 Results percentage of blistered worms by two. Thus the penetrance is only an estimate in this experiment, and the penetrance can be greater than 100% if more that half the progeny are homozygous for bli 4, as was the case for -  dpy 14, which gave a penetrance of 118% -  Blistering was completely or almost completely suppressed by 3 of the 11 heterozygous Dpy mutations. Dpy-5 and Dpy-13 both dominantly suppressed blistering completely in hermaphrodites and nearly completely in males; Dpy-6, which is X-linked, nearly completely suppressed blistering in hermaphrodites (Table 2). dpy 13(e184) is semi-dominant, but dpy 5(e61) -  -  and dpy 6(e14) are recessive. Dpy-9 and Dpy-10 dominantly reduce Bli-4 -  penetrance to about 20%. dpy 9(e424) and dpy 10(e128) are both recessive. -  -  Dpy-3, Dpy-14, Dpy-17, Dpy-18 and Sqt-3 all had moderate or no dominant effects on Bli-4 penetrance. Of these, only sqt 3(e24) is semi-dominant. The -  dominant suppression of blistering by recessive Dpy mutations indicates that the cuticle of the Dpy heterozygotes is not fully wild type.  57  3 Results  Table 2. bli-4(e937) penetrance in dpy homozygote and heterozygote backgrounds  dpy-X  Average lengths (mm)  Penetrancec in bli-4; dpv-X bli-4 +  Penetranceb in bli-4; dpv-X bli-4 dpy-X hermaphrodites  hermaphrodites  males  dpy-1  0.50 (0.036)  <1% (n=761)  58% (n=214)  51% (n=203) d  dpy-3  0.54 (0.040)  0% (n=909)  40% (n=421)  dpy-5  0.49 (0.051)  0% (n=735)  0% (n=459)  6% (n=442)  dpy-6  0.50 (0.051)  0% (n=245)  2% (n=346)  0% (n=337)  dpy-9  0.55 (0.036)  0% (n=1435)  19% (n=448)  22% (n=461)  dpy-10  0.53 (0.050)  0% (n=885)  19% (n=446)  12% (n=427)  dpy-13  0.49 (0.046)  0% (n=1143)  0% (n=437)  <1% (n=427)  dpy-14  0.57 (0.055)  41% (n=343)  118% (n=159)  88% (n=164)  dpy-17  0.72 (0.065)  54% (n=1564)  68% (n=361)  81% (n=370)  dpy-18  0.68 (0.066)  0% (n=687)  66% (n=373)  98% (n=361)  sqt-3  0.81 (0.054)  0% (n=345)  58% (n=442)  51% (n=424)  dpy +  1.26  0% (n=405)  95.5% (n=1092)  aAverage length of 20 dpy-X unc-13 adult hermaphrodites. Standard errors are presented in brackets.  bPenetrance is defined as the percentage of blistered worms out of the total number observed.  cBecause the bli-4;dpy-X hermaphrodites were mated to bli-4 heterozygous male in this experiment (Section 2.3), a maximum of 50% of the worms are expected to blister if penetrance is 100%. Therefore, penetrance is defined as the percentage of blistered worms out of one half the total number of observed. ddpy-3 and dpy-6 are X-linked and all male cross progeny are Dpy. eFrom Wood, 1988.  58  3 Results 3.1.2 Mapping of bli 4 alleles -  h42, h199, h254 and e937 were three factor mapped by scoring segregation from strains bearing cis-linked flanking markers dpy - 5 and  unc - 13 in trans to an unmarked chromosome (Table 3). Each allele mapped near the center of the interval, with the exception of h199 (see below). Recombinants were picked as Dpy-5 or Unc-13 worms and progeny tested for the presence of bli - 4. s90, which was not induced on a dpy- 5 chromosome, was two factor mapped to 0.6 m.u. from unc - 13(e51), consistent with a position between dpy- 5 unc - 13.  h199 mapped 0.8 m.u. from unc - 13, but failed to recombine with dpy- 5. This apparent crossover suppression could indicate that h199 is a deficiency spanning bli - 4, or that h199 is linked to a second mutation in an essential gene closely linked or to the left of dpy - 5. h199 is unlikely to be a deficiency, because it complements alleles of unc - 40, which is between bli - 4 and dpy- 5. However, the chromosome carrying h199 fails to complement sDf4, a deficiency of dpy- 5 (Howell, 1989). h199 could be an inversion with breakpoints in bli - 4 and an essential gene in sDf4. More likely, however, KR513, the strain carrying h199, carries a second site mutation, in a gene to the left of unc - 40. This conclusion is supported by the observation that EcoRI restriction endonuclease digested genomic DNA prepared from h199 heterozygous worms showed wild type band sizes on Southern blots when probed with bli - 4 genomic DNA clones (Section 3.2.1.2.2). The putative second site mutation in KR513 was designated h1360. The position of the e937 allele of bli - 4 on LGI between the markers  dpy - 5 and unc - 13 was confirmed by three factor mapping. Recombinant F1 progeny of hermaphrodites of the genotype  dpy- 5 bli - 4 unc - 13(e450)/ + + + were picked. Eleven Bli Unc, six Dpy, ten 59  3 Results Unc, and seven Bli Dpy recombinants were recovered. This gives the map position dpy 5 (17/34) bli 4 (17/34) unc 13, placing bli 4 at the center of the -  -  -  -  dpy 5 unc 13 interval, which is 1.6 m.u. (Howell et al., 1987). Having an -  -  accurate map position for bli 4 was important to the cloning of this gene. -  3.1.3 Screen for a mutator allele To facilitate the isolation of the bli 4 coding region, a screen was -  performed to identify potential transposable element insertion mutations of  bli 4 (Figure 6). The strain KR1822 was constructed for use in this screen -  and assayed for mutator activity as described in Section 2.8. KR1822 has the genotype unc 63(e384) unc 13(e450); mut 6(s1702). mut 6 causes high levels -  -  -  -  of transposition of the transposable genetic element Tc1 (Mori, Moerman and Waterston, 1988). Spontaneous alleles of bli 4 would occur in KR1822 -  whenever a transposable element inserted into the bli 4 locus. In the -  screen, blistered worms in the Fl generation could only arise if a spontaneous bli 4 mutation occurred in the gonad of the hermaphrodite. -  Because the males were heterozygous for bli 4, one-half of the Fl progeny -  were not informative in this cross (ie, one-half of the spontaneous bli 4 -  alleles would not be detected because they would receive a wild type bli 4 -  allele from the male). Three Bli worms, two hermaphrodites and one male, were identified after screening 164,600 worms (82,300 informative chromosomes), an induction frequency of 3.6X10 -5 . Of the three spontaneous blistered animals recovered, one survived. The surviving hermaphrodite carried a class II bli 4 lethal allele designated h1010 and -  maintained using the translocation szT1 (I;X) in the strain KR1858.  60  3 Results  Table 3. bli-4 genetic map data Recombination frequency (m.u.) Allele  Maternal genotype  Wild type  Dpy  Unc  h42  dpy-5 h42 unc-13 +^+^+  1376  10  h199  dpy-5 h199 unc-13 +^+^+  1781  h254  dpy-5 h254 unc-13 +^+^+  s90  s90 unc-13 +^+  dpy-5 to bli-4  bli-4 to unc-13  8  0.5 m.u. (0.3-0.9)a  0.4 m.u. (0.2-0.8)  0  10  0.0m.u. (0.0-0.1)  0.4 m.u. (0.2-0.9)  1033  9  3  0.7 m.u. (0.3-1.2)  0.2 m.u. (0.1-0.5)  1896  N/A  14  N/A  0.6 m.u. (0.3-0.9)  a95% confidence intervals are given in parentheses. Confidence intervals were  calculated using the table of Crow and Gardener (1959).  61  3 Results  dpy-5 b11-4(e937)  unc-63 unc-13 mut-6  Po  d  unc-63 unc-13 mut-6  Fl^  unc-63 •^bii-4(mut) unc-13  unc-63 • b11-4(+) unc-13 • dpy-5 b11-4(6937) #  dpy-5 b11-4(e 937) •  Wild-type (50%)  Blistered (Rare)  unc-63 • bli-4(*) unc-13 •^  unc-63 • b/i-4(mut) unc-13  • Wild-type (50%)^  Wild-type (Rare)  Figure 6. Screen for mutator-induced alleles of bli 4 -  Screen for spontaneous bli 4 lethal alleles in a mut 6(st702) background. -  -  KR1822 hermaphrodites were crossed to dpy 5 bli 4(e937)/ + + males and -  -  the Fl progeny screened for blisters. Blisters would result when an ovum carrying a novel bli 4 mutation in the germline of the KR1822 hermaphrodite -  was fertilized by a sperm carrying bli 4 (e937). Because dpy 5(e61) -  -  suppresses blistering in e937 homozygotes, only severe bli 4 mutations, that -  is, lethal class II alleles, would be detected in this screen.  62  3 Results 3.1.4 Determination of developmental arrest stage The stage at which lethal homozygotes arrested development was determined. Several heterozygous hermaphrodites of the genotype  dpy 5 let X unc 13/ + + + were permitted to lay eggs on an NGM plate for -  -  -  a short period (not more than two hours) and then the homozygous lethal progeny were examined by Normarski differential interference microscopy for time of arrest. All of the class II lethal alleles of bli 4 (including h1010) -  arrest development at or just before hatching, which is the end point of embryogenesis. Thus, the essential role of the bli 4 affected by the class II -  alleles occurs at or before the end of embryogenesis. The class Ill complementing alleles (s90 and h754) arrest development during the L1 stage, later than the non-complementing alleles. 3.1.5 Evidence that e937 is hypomorhic Muller (1937) defined a hypomorphic mutation as one that resulted in a reduction, but not a total loss, of function. Genetically, hypomorphic mutations have a phenotype that is more severe in trans to a deletion, or a null (loss of function) allele. No deletions of the bli 4 locus were available. -  However, the blistered phenotype of e937 exhibited higher penetrance, and greater expressivity in trans to class II alleles in a genetic background that did not normally permit blistering. The blistered phenotype of e937 hermaphrodites is suppressed in dpy 5(e61) heterozygotes (Section 3.1.1.4). -  This suppression was reversed when e937 was heterozygous to a class II lethal allele. That is, the blistered phenotype of e937 is more severe in trans to class II alleles. When bli 4(e937)/ bli 4(e937) hermaphrodites were -  -  crossed to dpy 5 bli 4(h42) unc 13/ + + + males, 46% of hermaphrodites -  -  -  and 50% of males were blistered. The maximum percentage of blistered progeny expected in this cross was 50%. Similar results were obtained with  63  3 Results other lethal alleles (data not shown). This result was interpreted as follows: that the function required to prevent blistering in adult cuticle is reduced in  e937 hermaphrodites, and is more severely reduced or absent in the lethal alleles tested. This result is consistent with Muller's definition (Muller, 1937) of a hypomorphic mutation. 3.1.6 Complementation analysis of bli 4 alleles -  The results of inter se complementation tests of bii 4 alleles are -  presented in Table 4. Mutations fell into three complementation classes that correlated with the phenotypes. Class I was the viable allele e937. Class II was the lethal alleles, h42, h199, 11254, h384, 11427, 11520, h699, h791 and  h1010. Class II alleles failed to complement all other alleles. Class III was the lethal alleles s90 and h754. These alleles complemented e937. Two explanations for this complementation pattern are as follows. In the first hypothesis, e937 and the class III complementing alleles are mutations of one gene that affect complementary domains or alternatively spliced exons of the gene. The second hypothesis is that e937 and the class Ill alleles are mutations of two separate genes, and all of the class 11 lethal mutations are deletions affecting both genes. In this hypothesis, the nine class II mutations are all deletions that delete the gene affected by e937 and the gene affected by the class III mutations. The two gene hypothesis can be proven incorrect if just one of the class II alleles is shown not to be a deletion. At least one class II allele, h1010 is not a deletion, it is an insertion (Section 3.2.1.2.1). Furthermore, four other class II alleles, h42, h199, h384 and h487, do not contain chromosomal alterations that are detectable by Southern analysis (Section 3.2.1.2.2). On this basis, I infer that e937 and the class III mutations s90 and h754 are complementing alleles of the same gene.  64  3 Results In summary, the class I allele e937 is hypomorphic, and is the only allele that results in blisters. Blisters are adult specific and occur between cuticle layers. Class II alleles fail to complement e937 and have the most severe phenotype, developmental arrest at the point of hatching. That class II alleles are the most severe is consistent with the hypothesis that they are null alleles. Class III alleles complement the blistered phenotype of e937, and arrest development at the LI stage.  65  3 Results  Table 4. Inter se complementation dataa for bli-4 alleles Class III Class II Class I Allele e937^h42 h199 h254 h384 h427 h520 h699 h791 h1010 h754 s90  +  e937 h42 h199 h254 h384 h427 h520 h699 h791 h1010 h754 s90  aComplementation tests were conducted as decribed (Section 2.5). Failure to complement is indicated by "-". Allelic combinations resulting in a blistered phenotype are shaded grey; other complementation failures are lethal. complementation is indicated by "+".  66  3 Results 3.2 MOLECULAR ANALYSIS OF THE bli-4 LOCUS A physical map of overlapping cosmid and YAC clones has been generated for the C. elegans genome (Coulson et al., 1986; Coulson et al., 1988). The bli-4 locus was cloned by aligning bli-4 with the physical map. Cosmids spanning bli-4 were used to as hybridization experiments to probe Southern blots of restriction enzyme digested DNA from bli-4 mutant strains. Clones of genomic DNA that detected restriction fragment length differences (RFLD) in bli-4 mutant strains were then used as probes to isolate cDNA clones. The DNA sequence of the cDNA clones was determined, and the sequence used to search DNA and protein computer data bases. This approach revealed that bli-4 encodes protein products that structurally resemble the kex2-like proteinases, enzymes that process secreted proteins by proteolytic cleavage. 3.2.1 Identification of the bli-4 coding region 3.2.1.1 Alignment of the genetic and physical maps A partial physical map of cosmids from the bli-4 region is presented in Figure 7. bli-4 had been mapped between dpy-5 and dpy-14 (Rose and Baillie, 1980). Approximately 35 cosmid clones are needed to cover the dpy-  5 to dpy-14 interval. The average size of the cosmid clone inserts is 34 kb (Coulson et al., 1986). Therefore, the dpy-5 to dpy-14 interval includes about 1,200 Kb of DNA. To define more precisely the position of bli-4 within the physical map, I used two strain-specific RFLDs markers flanking the locus. The RFLD hP5 and an RFLD associated with the left breakpoint of deletion hDf8 were used to place bli-4 within a 200 Kb interval.  67  3 Results 3.2.1.1.1 hP5 defines the left-most position of bli-4 in the physical map A first step in placing bli-4 within the physical map was to position the gene with respect to the molecular marker hP5. hP5 is an N2 (Bristol) BO (Bergerac) strain restriction fragment length polymorphism (RFLP) hP5. The hP5 RFLD was detected by the hybridization probe pCeh51. pCeh51 detected a 2.4 kb band in EcoRI digested N2 DNA and a 4.0 kb band in EcoRI digested BO DNA. To three factor map bli-4 with respect to  hP5, I constructed N2 BO recombinant mapping strains as described in Figure 8. Southern blots of EcoRI digested genomic DNA prepared from recombinant strains segregated from hP5 mapping strains were hybridized to the pCeh51 probe. An example of such an experiment is presented in Figure 9. The strains used and the data obtained are presented in Table 5. 29 strains derived from independent recombinants were analyzed. The map order inferred from these experiments is dpy-5 (7/29) hP5 (3/29) bli-4 (18/29)  unc-13. This result places hP5 between dpy-5 and bli-4, an interval of 0.9 cM. 3/10 of 0.9 cM is about 0.3 cM. The DNA density in the bli-4 region is about 500 kb per cM (Starr et A, 1989), predicting that bli-4 would be located about 150 kb to the right of hP5. The hP5 mapping data lead to the conclusion that hP5 defines the left-most position of bli-4 in the physical map (Figure 7).  68  3 Results  b11 -4  dpy-5 unc-40  I^I  A^  hP5  unc-87 dpy-14^unc-13 II hDf8 0.1 au  C40A4^ 032012^  A  722C4  ^  014131 I  B04430  018E7^  104F10  A  hDf8  hP5^  Figure 7. Genetic and physical maps of the bli-4 region  bli-4 was mapped genetically between the N2/B0 strain RFLD hP5 (0.3 mu to the left of bli-4), and the left breakpoint of the deletion hDf8. This interval is spanned by 200 kb of contiguous cosmid clones, except for a small gap between K06E6 and C44D11, which is spanned by YAC clones (Coulson et  al., 1986; Coulson et al., 1988). The positions of hP5 and the left break point of hDf8 in the physical map are indicated by arrows. All of the cosmids shown were used to probe bli-4 mutations.  69  3 Results  Figure 8. Construction of N2/BO RFLP mapping strains  Wild type BO hermaphrodites were crossed to dpy-5(e61) bli-4(e937) unc-13(e420)/ + + + N2 males. L4 hermaphrodites were picked from the Fl progeny and were placed individually on fresh plates. F2 progeny recombinant in interval A were picked as Dpy-5 or Bli-4 Unc-13 hermaphrodites. F2 progeny recombinant in interval B were picked as Unc-13 or Dpy-5 Bli-4 hermaphrodites. Because dpy-5(e61) suppresses blistering in e937 homozygotes, all Dpy-5 progeny were crossed to a bli-4 lethal bearing strain to determine presence of the bli-4(e937) allele.  70  3 Results  Construction of N2/B0 mapping strains BO Hermaphrodites^  N2 Males dpy-5^A^bli-4^B^unc-13  dpy-5  ^  A^on-Ar^B^unc-13 N2 BO  ^Self^fertilize Pick recombinants  +  dpy-5  +  dpy-5^bli-4  ^I  ^ I^  aPY  -  a AY - 5  5  bli-4  I  i  unc-13^dpy-5  bli-4  unc-13  bli-4  unc-13^aPY  I  bli-4  + -: I  unc-13  unc-13  -  5  unc-13  Self fertilize Pick homozygotes  aPY -5^  b1I-4  dpy-5^bli-4  -4  unc-13  unc-13  b1/- 4  unc-13  unc-13  121/  Recombinants in Interval A ^Recombinants in Interval B  71  3 Results  1^2 4.0 kb -  2.4 kb  -  1111111 eft  ao  5^6^7  ,  1111^  aria  Figure 9 Southern analysis of hP5 mapping strains -  Genomic DNA digested with EcoRl and probed with pCeh51, which detects the RFLD hP5. Lane 1, N2; lane 2, BO; lane 3, KR1210; lane 4, KR1211; lane 5, KR1212; lane 6, KR1220; lane 7, KR1221. The genotypes of the recombinant strains are presented in Table 5.  72  ^  3 Results Table 5 . hP5 Three factor mapping data. Strain^Recombinant Genotype^Recombinant^hP5^hP5 Recombined Intervala^Patternb^with bli-4? KR1187 + (BO) bli-4 unc-13 (N2)^A KR1189 + (BO) bli-4 unc-13 (N2)^A KR1190 + (BO) bli-4 unc-13 (N2)^A^BO^Yes KR1191 + (BO) bli-4 unc-13 (N2)^A KR1192 + (BO) bli-4 unc-13 (N2)^A^N2^No KR1193 + (BO) bli-4 unc-13 (N2)^A KR1194 + (BO) bli-4 unc-13 (N2)^A KR1195 + (BO) bli-4 unc-13 (N2)^A^BO^Yes KR1210^dpy-5 (N2) + + (BO)^A^BO^No KR1211^dpy-5 (N2) + + (BO)^A^BO^No KR1212^dpy-5 (N2) + + (BO)^A^N2^Yes KR1213^dpy-5 (N2) + + (BO)^A KR1214^dpy-5 (N2) + + (BO)^A^BO^No KR1215^dpy-5 (N2) + + (BO)^A^BO^No KR1216^dpy-5 (N2) + + (BO)^A^BO^No KR1217^dpy-5 (N2) + + (BO)^A^N2^Yes KR1218^dpy-5 (N2) + + (BO)^A^BO^No KR1196^+ + (BO) unc-13 (N2)^B^BO^No KR1197 + + (BO) unc-13 (N2)^B^BO^No KR1198^+ + (BO) unc-13 (N2)^B^BO^No KR1199 + + (BO) unc-13 (N2)^B^BO^No KR1200 + + (BO) unc-13 (N2)^B^BO^No KR1201^+ + (BO) unc-13 (N2)^B^BO^No KR1202 + + (BO) unc-13 (N2)^B^BO^No KR1203 + + (BO) unc-13 (N2)^B^BO^No KR1204 + + (BO) unc-13 (N2)^B^BO^No KR1205 + + (BO) unc-13 (N2)^B^BO^No KR1206 + + (BO) unc-13 (N2)^B^BO^No KR 1207 + + (BO) unc-13 (N2)^B^BO^No KR1208 + + (BO) unc-13 (N2)^B^BO^No KR1209 + + (BO) unc-13 (N2)^B^BO^No KR1219 dpy-5 bli-4 (N2) + (BO)^B^N2^No KR1220 dpy-5 bli-4 (N2) + (BO)^B^N2^No KR 1221 dpy-5 bli-4 (N2) + (BO)^B^N2^No KR1222 dpy-5 bli-4 (N2) + (BO)^B^N2^No aRecombinant interval A indicates recombination between dpy-5 and bli-4; interval B indicates recombination between bli-4 and unc-13. bhP5 pattern N2 indicates that pCeh51 detects a 2.4 kb EcoRI fragment in the recombinant strain; pattern BO indicates that pCeh51 detects a 4.0 kb band. 73  3 Results  3.2.1.1.2 hDf8 defines the right-most position of bli 4 on the physical map -  A deletion of dpy 14, hDf8 (McKim, Starr and Rose, 1992), provided a -  convenient tool to determine the right-most position of bli 4 within the -  cosmid map. To determine the position of hDf8 relative to bli 4, the hDf8 -  chromosome was tested for complementation of bli 4(e937) allele. Because -  hDf8 did not carry any genetic markers, the following protocol was used. Males of the genotype hDf8 (I); 0/ szT1(1;X)[lon 2.1 were mated to -  hermaphrodites of the genotype dpy 5 bli 4(e937). In this cross, the only -  -  males produced have the genotype dpy 5 bli 4(e937)/ + + hDf8. If hDf8 -  -  deleted bIi 4, the non Lon-2 males would be blistered; if hDf8 did not delete -  bli 4, the male progeny would be wild type. Only wild type male progeny -  were observed, indicating that hDf8 does not delete bli 4 and demonstrating -  that the gene lies outside of the deletion, and to the left of the deletion breakpoint. To localize hDf8 on the cosmid map, I determined the dosage of restriction fragments of cosmids on either side of the deletion breakpoint. These fragments were used to probe Southern blots of restriction digested wild type and deletion heterozygote genomic DNA. Using an LKB scanning densitometer, the intensity of autoradiograph bands was compared to an internal control probe, pCes233, which hybridizes to the molecular marker sP4 on LGIV (Baillie, Beckenbach, and Rose, 1985). EcoRI fragments of cosmids K06E6 (K06E6-E6) and C44D11 (C44D11-E5) were used (Figure 10). K06E6-E6 was present at equal intensity in both N2 and KR1816 (hDf8) strains, while C44D11-E5 was present at 1/2 the intensity in the deletion heterozygote strain as in the wild type strain (Table 6). Thus, the left break point of hDf8 is located between K06E6 and C44D11. The hDf8 mapping  74  3 Results data indicates that the bli-4 coding region is located to the left of C44D11 on the physical map (Figure 7). This region includes approximately 200 kb of DNA, consistent with the hP5 mapping data. In Southern analysis of EcoRI digested hDf8 heterozygote DNA, C44D11 detected the loss of an 8.5 kb band and a novel band at 5.5 kb (Figure 11). The RFLD detected by C44D11 is a candidate for the hDf8 left breakpoint. The right breakpoint of hDf8 occurs between dpy-14 and unc-  13, and is detected by the cosmid C14Al2 in Southern analysis (McKim, Starr and Rose, 1992).  75  3 Results  Figure 10. Dosage analysis of hDf8  A.  N2 and KR1000 DNA digested with EcoRI and probed with K06E6-E6 (the  6th largest EcoRl fragment of K06E6) and pCes233, which detects the RFLD sP4. B. LKB laser densitometer scan of the N2 lane from panel (A). C.  LKB laser densitometer scan of the KR1000 lane from panel (A).  D.  N2 and KR1000 DNA digested with EcoRl and probed with C44D11-E5  (the 5th largest EcoRl fragment of C44D11) and pCes233, which detects the RFLD sP4. E. LKB laser densitometer scan of the N2 lane from panel (D). F. LKB laser densitometer scan of the KR1000 lane from panel (D)  76  ^  MO. 4'. 0^I 664 40411006^■ 6 Ns • 1110 0••  0 6o 1  1444.^I.... . •6•6. II.^I A 50, • 4 0 ■ .4, • I  1  N2^KR1000  4I  , 6  10410 - MI  KO6E6- ONO S■ E6 .15 1.14 I^10  ow JM  1  4111•11115 sP4-^ SORRA  1.w 1M^114^1441/4  N2  405440  KR1000  I ttttt 4.«414  I  61^10^  ...."."--''''' '--  ^.114^10^  .  • 11 • 101^1.1.1.1^I • 4 40.4 1 1 4004 141^1-^•^I ,,,,,^14140  C. K06E6/sP4 - KR1000 Lane Scan  B. K06E6/sP4 - N2 Lane Scan  A.^K06E6  .0*^....04144.4^.  0 1^  4^ ••^ 11•40 (NI^I^•^lllll V^ •^4.04 10 • III^1.04 • 1^1 • 44 0104 1  ^5 b4. • 4 • i •■•• • 1•4 1  10••11•^1 0. 0 4.4100^i • 01 • 1. 410 1•61. • 1 11 1  1 1.4 I a,  1111.  1^11 ^  .1. 4  104  MID^044‘11  C44D11  ES  1 c4011 . E5 I •  1.11  1.A II • 0•54.5111^-^01 0.W  IM  4 • 1.w  sP4  -  gee^ANS  4 J.  -....-., .^ M^  1..1.4^I^ 0441^  D. C44D11  1.34 04 IN^114^1M^114^la^111 ./WI • •^4-•14 •^KIII• 101^1^I^414444 •• • 1 •^  E. C441)11/1P4 - N2 Lane Scan  •  -411.....410411014......^Ill^  .-K .4 141^1-^•  .^ 114^114^101^Ka • 04 • 01^t 00 • 1^1 • 45 Ki•••• 1 , 0a0^10^I  F. C44D11/sP4 - KR1000 Lane Scan  3 Results  Table 6. hDf8 dosage data Cosmida^Genomicb^Relative Areae^Cosmid probed KR1816e Deleted fragment^DNA^Cosmid Probe sP4 probe^sP4 probe^N2^in hDfa? K06E6 - E6^N2^52.31^37.12^1.41^1.05^No KR1816^52.88^35.46^1.49 C44D11 - E5^N2^45.81^54.19^0.42^0.49^Yes KR1816^26.93^64.16^0.85 aFragment to be tested for dosage relative to the LGIV control probe sP4. Probes were isolated by electroelution. K06E6 - E6 is the sixth largest EcoRI fragment of the cosmid K06E6. C44D 11 - E5 is the fifth EcoRI fragment of the cosmid C44D 11. bSource of genomic DNA to be probed with cosmid and test fragments. CRelative area of hybridization autoradiograph bands determined by scanning densitometer (see Figure 10). dRatio of cosmid probe band relative area to control probe band relative area. eRatio of cosmid probe band area to control probe band area determined in the N2 lane to that of the KR1816 lane. A ratio of 1 indicates that the cosmid test fragment is not deleted by hDf8; a ratio of 0.5 indicates that the test band is deleted by hDf8.  78  3 Results  2  5.5-  Figure 11. Detection of RFLDs by C44D11 in hDf8 N2 (Lane 1) and KR1000 (Lane 2) DNA digested with EcoRI and probed with C44D11. An 8.5 kb band detected in N2 DNA is absent in KR1000 DNA, and a novel 5.5 kb band is detected by C44D11 in KR1000 DNA.  79  3 Results 3.2.1.2 Identification of rearrangements in the DNA of bli 4 mutant strains -  3.2.1.2.1 A Tc1 insertion mutation in KR1858(h1010) DNA To determine the exact position of bli 4 within the interval defined by -  hP5 and hDf8, I used each of the cosmids shown in Figure 7 as a probe to compare the genomic DNA restriction patterns of two strains, wild type (N2) and KR1858, which carried the putative Tc1 mutation h1010 (Section 3.1.3). Each cosmid was used to probe EcoRl and Sall restriction digested genomic DNA prepared from wild type and KR1858 DNA (Figure 12). I chose to use EcoRl and Sall to digest the genomic DNA because, with some exceptions (Eidie and Anderson, 1985b), EcoRI does not cut inside Tc1, and would produce a novel band 1.6 kb larger than in the parental strain. Sall cuts twice within Tc1, and would produce three novel bands smaller than that of the parental strain. One of these novel bands, the Sall fragment internal to Tc1, would be only 247 by in length, a size that is difficult to see on a Southern blot. Therefore, only two new bands would be observed. Because KR1858 is heterozygous for the h1010 mutation, the N2 bands would also be observed. Only one of the cosmids, KO4F10, detected band shifts. A novel 2.9 kb band was observed in EcoRI digested KR1858 DNA, and two novel bands of 16.6 kb and 6 kb were observed in Sall digested DNA (Figure 12a). If the 2.9 kb band in KR1858 DNA detected by KO4F10 was a result of a 1.6 kb Tc1 insertion, then the 1.3 kb EcoRl would contain the insertion site. To test this prediction, EcoRI and Sall digested KR1858 DNA was probed with using the 1.3 kb EcoRl fragment of KO4F10. The same novel bands were detected (Figure 12b). To confirm my interpretation of the h1010 restriction data, I used the polymerase chain reaction to amplify across the h1010 Tc1 insertion site  80  3 Results (Figure 13). The primers used were p618 and KRp13. p618 is specific to a Tc1 sequence starting 72 by from one end of the transposable element. KRp13 is specific to a sequence 80 by to the 5' side of the 1.3 kb EcoR1 fragment containing the insertion (Section 3.2.3 reports the genomic sequence used to design the KRp13 primer). An amplification band of approximately 770 by was obtained (Figure 13b, lane 2). The presence of an amplification band confirms that the insertion is a Tc1 element. The size of the amplification band is consistent with an insertion site approximately 620 by into the 1.3 kb EcoRI fragment. As an additional control, DNA from KR1822, the mutator strain that produced KR1858, was probed. KO4F10 detects wild type band patterns in EcoRI and Sail restriction digested KR1822 DNA, the mutator strain that produced h1010. The Tc1 insertion in KR1858 is a candidate for the h1010 mutation for two reasons. First, the occurrence of the Tc1 insertion was concurrent with the occurrence of the h1010 mutation. Second, the position of the insertion is consistent with the mapping data of bli 4 with respect to -  hP5 (Section 3.2.1.1.1; Table 5). 3.2.1.2.2 A deletion mutation in CB937 (e937) DNA To identify additional rearrangements of the bli 4 locus, I used a 11 kb -  Xhol subclone of KO4F10 in the plasmid pCeh181 to probe Southern blots of genomic DNA prepared from other bli 4 mutant strains. I probed EcoRI -  digested DNA from strains carrying bli 4 alleles e937, h42, h199, h384, and -  h427. A rearrangement was detected in CB937 DNA, homozygous for e937, but not in the other strains tested (Figure 14). The 11 kb Xhol fragment of KO4F10 detected the disappearance of two adjacent EcoRI fragments of 2.2 and 3.0 kb and the appearance of a novel 1.7 kb band in CB937 DNA. The simplest interpretation of this pattern is that the rearrangement is a 3.5 kb  81  3 Results deletion spanning one EcoRI site, and fusing the 2.2 and 3.0 kb bands into one smaller 1.7 kb band. To confirm this interpretation and to further characterize the deletion, I constructed a lambda-zap genomic library using DNA from CB937. Using a mixture of the 2.2 and 3.0 kb EcoRI fragments of pCeh181 as probes, I isolated a 1.7 kb EcoRl fusion fragment from the CB937 library in the plasmid pCeh206. The fusion fragment was used to probe CB937 genomic DNA (Figure 14). The pattern shown in Figure 14 is consistent with a deletion of 3.5 kb. My interpretation of the of restriction data was confirmed by sequencing through the deletion breakpoints (See Section 3.2.5.2). The 11 kb Xhol fragment of KO4F10 detects rearrangements in the DNA of strains carrying two independent bli 4 mutations; KR1858, carrying -  h1010, and CB937, carrying e937. Based on my positioning of bli 4 relative -  to hP5 and hDf8, and my detection of two chromosomal alterations affecting the same 11 kb region, I conclude that the bli 4 coding region is at least -  partially contained within KO4F10. This conclusion is supported by the finding that an artificial duplication constructed by the germline injection of KO4F10 and the wholly overlapping cosmid C29F10 rescues the lethal phenotypes of the class II allele h42, and the class Ill allele h754 (Jennifer McDowall, personal communication). The 1.3 kb EcoRl fragment of KO4F10 will be referred to as the 1.3 kb h1010 probe. The 1.7 kb fusion fragment in pCeh206 will be referred to as the 1.7 kb e937 probe.  3.2.1.3 Detection of RNA bands with KO4F10 DNA fragments The 1.3 kb h1010 probe, and the 1.0 and 1.9 kb EcoRI fragments flanking the 1.3 kb h1010 probe each detected four bands on a northern blot of C. elegans mixed stage total RNA (Figure 15). These bands were 3.5,  82  3 Results 3.1, 2.6. and 1.6 kb in size. The detection of RNA bands by the DNA fragments flanking the Tc1 insertion site in h1010 indicates that this region includes a gene.  83  3 Results  EcoRI^Sall ^EcoRI^Sail ^0  .9.  Z r  ^0 01  I?:  csi 2 cg 2 z -c z .e  Z 4  16k b-  VW < NUM  am  Ivor  5.8kb-  2.9k bE< air 41111. 11110  1.3kb-  41111 rr  B  A  Figure 12. Southern analysis of KR1858 b/i-4(h7070)  A. Genomic DNA from N2 (wild type) and KR1858, carrying the h1010 allele  in trans to a wild type allele, was digested with EcoRI or Sall, probed with whole cosmid K04F10. Bands that are altered in size in KR1858 are marked by arrows. B. Genomic DNA from the same strains digested by the same enzymes as in (A) probed with the 1.3 kb EcoRI fragment of KO4F10, designated the 7.3 kb  h7010 probe.  84  3 Results  Figure 13. Amplification of the h1010 Tc1 insertion site  Polymerase chain reaction amplification of the Tc1 insertion site in h1010. A. Restriction map of the genomic region cloned in the cosmid KO4F10 spanning the insertion site, and the primers used to amplify the insertion point. B. Agarose gel analysis of amplification products. Lane (1) lamda  Hinc1111/EcoRI marker; lane (2) Amplification using primers KRp13 and p618 with 100 ng of KR1858 DNA as template; lanes (3) to (6), Controls, same as lane (2) except as follows: lane (3) No template; lane (4) No primers. lane (5) Primer p618 only. lane (6) Primer KRp13 only.  85  3 Results  X^1.9^  1.3^ 1.0  102p13  29  Tcl  p618  PCR  A  1^2^3^4^5^6  770 by -  B  86  3 Results  EcoRI  EcoRI  N2 e937  N2 e937 >15kb4.5kb-  4.0kb-  3.0kb-  3.0kb- woo 2.4kb-  2.2kb1.9kb-  1.7kb-^MID  1.7kb-  •  1.3kb- Wt. MS  1.0kb-  A  •  1.0kb-  B  Figure 14. Southern analysis of CB937 bli 4(e937) -  A. Genomic DNA from N2 (wild type) and CB937, carrying the e937 allele, was digested with EcoRl and probed with pCeh181. B. Genomic DNA from the same strain as in (A) digested with EcoRl and probed with a 1.7 kb fusion band in the plasmid pCeh206 isolated from a CB937 genomic library.  87  3 Results  A  ^ ^ ^ ^  C  D  E  F  3.5-2.6-  1.6-  Figure 15. Northern analysis of bli 4 transcripts -  Mixed stage total RNA was probed with: A. The 1.9 kb EcoRI of pCeh181. B. The 1.3 kb EcoRl fragment of pCeh181 flanking the Tcl insertion site of  h1010. C. The 1.0 kb EcoRI of pCeh181. D. The 0.8 kb EcoRI/Sall fragment of pCeh181 containing genomic DNA encoding the unique 3' end of blisterin A. E. 211 by EcoRl fragment from the unique 3' end of blisterin B. F. 204 by BamH11BgIll fragment from the unique 3' end of blisterin C.  88  3 Results 3.2.2 Analysis of bli-4 cDNA clones 3.2.2.1 Isolation and hybridization analysis Six bli-4 cDNA clones were isolated from the cDNA library of Barstead and Waterston (1989) using the 1.3 kb MOW probe (Section 3.2.1.2.1) to screen approximately 40,000 phage plaques. To determine the approximate extent of the bli-4 locus, the cDNA clones were used as probes in a Southern analysis of the genomic region around bli-4 contained in the cosmid KO4F10 (Figure 16a). cDNA clones hybridized to restriction fragments of KO4F10 corresponding to about 15 kb of genomic DNA. As shown in Figure 16a the hybridization pattern was discontinuous. Three of the cDNA clones were chosen for further characterization. These cDNA clones and their predicted products will be referred to as blisterin A (pCeh200), blisterin B (pCeh197) and blisterin C (pCeh196). 3.2.2.2 Sequence analysis of cDNA clones The following sections present sequence data derived from the cDNA clones blisterin A, blisterin B, and blisterin C. The three cDNA clones share overlapping sequence at the 5' end, but differ at the 3' end. The overlapping sequence will be referred to as the "common" sequence. The different 3' ends of the cDNA clones will be referred to as "unique" sequences. The point at which the 3' sequence of the three cDNA cones diverge from one another will be referred to as the "divergence point" (Figure 16b). Blisterin B is the only one of the three cDNA clones that appears to include a complete open reading frame (section 3.2.2.2.1). No evidence exists to support the conclusion that any of the cDNA clones are complete cDNAs. Blisterin C and blisterin A begin abruptly within the blisterin B open reading frame at points indicated in Figure 18. Blisterin A and blisterin C do  89  3 Results not contain open reading frames begining with an ATG start codon. Instead, they begin abruptly within the blisterin B open reading frame at points indicated in Figure 18. Therefore, a working hypothesis is that the blisterin A and blisterin C open reading frames are not complete at the 5' end. Blisterin A and blisterin C are identical to blisterin B from their repective 5' ends up to the point where the three cDNA clones diverge from each other. All three cDNA clones terminate in a poly-A track, and are therefore likely to be complete cDNA clones at the 3' end. Because blisterin A and blisterin C begin within the blisterin B open reading frame, these cDNA clones will be described in relationship to blisterin B. For this reason, I will discuss blisterin B first. Sequencing strategies are presented in Figure 17. A summary of the sequence data is presented in Table 7.  90  3 Result  Table 7. Summary of bli-4 cDNA sequence data cDNA  Plasmid  Total Length by  ORFa  Start Point in Blisterin Bb  Stop codon  Protein length  Blisterin A  pCeh200  1459  1281  905  TGA (1284)  427  Blisterin B  pCehl97  2421  2193  0  TGA (2370)  731  Blsiterin C  pCehl96  1956  1713  1295  TAA (1755)  571  cDNA  Divergence Point  Unique Sequence Length  Unique ORF Length  Unique Protein length  5' UTRc  3' UTR  Blisterin A  1244  195  39  13  n/a  150  Blisterin B  2149  245  219  73  175  26  Blisterin C  855  1081  858  13  n/a  223  aORF: Qpen Reading Frame. bBlisterin A and blisterin C contain incomplete open reading frames that begin within the blisterin B ORF. eUTR: Entranslated Region  91  3 Results 3.2.2.2.1 Blisterin B The sequence of the blisterin B cDNA is presented in Figure 18. blisterin B contains an EcoRI insert of 2,448 nucleotides. This number includes 27 adenosine residues at the 3' end of the sequence, indicating that the cDNA is complete at the 3' end. Blisterin B has two potential ATG methionine initiation codons at the 5' end. The first ATG, at nucleotide 36, is followed by an in-frame stop codon, TAA, at the fourth codon position, and eight additional stop codons within the next 300 nucleotides. In addition, the first ATG is preceded by a pyrimidine at the -3 position, in violation of the Kozak initiation sequence (Kozak, 1986). The second, ATG, at nucleotide 176, has a purine at the -3 position, consistent with the Kozak sequence. The second ATG begins an open reading frame of 2,193 nucleotides encoding a potential protein of 731 amino acid residues. This reading frame terminates with the nonsense codon TGA beginning at position 2,370. Two in-frame stop codons are present upstream from the second ATG. The second ATG, therefore, is more likely to be a bona fide translation initiation site. The open reading frame begining with the second ATG is preceded by a 5' untranslated region of 175 nucleotides, and is followed by a 3' untranslated region of 26 nucleotides, excluding the poly-A track. On the basis of this long open reading frame and presence of a polyA track, I conclude that blisterin B contains a complete open reading frame. 3.2.2.2.2 Blisterin C The sequence of the blisterin C cDNA is presented in Figure 19. blisterin C contains an EcoRI insert of 1,946 nucleotides. This number includes 10 adenosine residues at the 3' end. The 5' end of the blisterin C insert begins at nucleotide position 1,295 of blisterin B. The first 835 nucleotides of blisterin C is identical to blisterin B from nucleotide 1,295 to  92  3 Results nucleotide 2,149, the point at which the three cDNA clones diverge from each other. The 1091 nucleotides (excluding the poly-A track) following the divergence point are different from the 3' end of blisterin B. The open reading frame of blisterin C begins at the third nucleotide at its 5' end, and is 1,713 nucleotides in length encoding a potential protein fragment of 571 amino acid residues (Figure 19). The blisterin C open reading frame terminates with a TAA stop codon at nucleotide 1,715. The blisterin C open reading frame is followed by an untranslated region of 223 nucleotides, excluding the poly-A track. 3.2.2.2.3 Blisterin A The sequence of the blisterin A cDNA is presented in Figure 20. The  3' end of blisterin A terminates in 20 adenosine residues. The 5' end of the blisterin A insert begins at nucleotide position 905 of blisterin B, and is identical to blisterin B as far as it was sequenced, to blisterin B nucleotide 1,140. The 3' end sequence is identical to blisterin B starting at blisterin B nucleotide 2,118, and then diverges from the blisterin B sequence at blisterin B nucleotide 2,149. The remaining 215 nucleotides are unique to blisterin A. I will assume that the internal sequence of blisterin A is identical to the portion of blisterin B that it overlaps. This assumption is supported by the estimated size of the blisterin A insert and restriction fragment pattern as determined by agarose gel electrophoresis. The sequence of blisterin A is identical to blisterin B from the blisterin A 5' end to nucleotide 1244. blisterin A diverges from the blisterin B sequence at blisterin B nucleotide 2,149. The 195 nucleotides (excluding the poly-A track) following the divergence point are different from the 3' end of both blisterin B and blisterin C. The open reading frame of blisterin A begins at the fourth nucleotide at its 5' end, and is 1,1281 nucleotides in length, encoding a  93  3 Results potential protein fragment of 427 amino acid residues (Figure 20). The blisterin A open reading frame terminates with a TGA stop codon at nucleotide 1,284. The blisterin A open reading frame is followed by a 3' untranslated region of 150 nucleotides, excluding the poly-A track. I conclude from the cDNA sequence data that bIi-4 encodes at least three different gene products generated by alternative splicing. This conclusion is supported by two observations. First, the sequences of the three cDNA clones differ starting at the same point. Second, the point at which the three cDNA sequences diverge is identical in all three clones and corresponds to an Exon/Intron junction (Section 3.2.3). 3.2.2.3 Northern Analysis The 1.9 kb, 1.3 kb, and 1.0 kb EcoRl fragments of the genomic subclone pCeh181 (Figure 16), encoding the 5' common sequence of the blisterin cDNA clones, each detected four bands on a northern blot of C.  elegans mixed stage total RNA (Section 3.2.1.3, Figure 15). These bands were 3.5, 3.1, 2.6. and 1.6 kb in size. To correlate these bands with the cDNA clones, northern blots of N2 RNA were probed with DNA fragments predicted to be unique to each of blisterin A, blisterin B and blisterin C (Figure 15). The probes used were as follows. To obtain a probe to detect blisterin A, the 0.8 kb EcoRI genomic DNA fragment encoding the unique 3' end of blisterin A (Section 3.2.3) was subcloned into plasmid pCeh205. To obtain a probe to detect blisterin B, a 211 by EcoRl fragment from the 3' unique end of blisterin B was isolated by electroelution from an EcoRl digest of the blisterin B plasmid. To obtain a probe to detect blisterin C, a 204 by EcoRI fragment from the 3' unique end of blisterin C was isolated by electroelution from an EcoRl digest of the blisterin C plasmid.  94  3 Results Of the three probes used, only the blisterin A probe detected a unique band on the northern blot, at 3.1 kb. The blisterin A cDNA clone insert is 1.4 kb in length. The fact that the RNA band detected by the blisterin A probe indicates that blisterin A is not a complete cDNA. The blisterin B and blisterin C probes failed to detect unique bands. Both probes detected bands at 3.5, 3.1 and 1.6 kb. The blisterin B probe detected the 3.5 most intensely, while the blisterin C probe detected the 3.1 band most intensely. The failure of these probes to detect unique RNA bands may indicate that they are contaminated by other fragments due to the electroelution procedure used to isolate the probes. Alternatively, the prediction that the blisterin B and blisterin C probes would detect unique RNA bands could be incorrect. None of the three cDNA probes detected the major band of 2.6 kb detected by the genomic fragments encoding the 5' common sequences. The 2.6 kb band may, therefore, represent a major transcript class not included in the cDNA clones that have been characterized.  95  3 Results  Figure 16. Alignment of cDNA clones with the bli-4 region A restriction map of the portion of KO4F10 containing the bli-4 locus is shown at the top of each panel. bli-4 cDNA clones are shown in the bottom portion of each panel. A. An alignment of bli-4 cDNA clones based on hybridization data is presented. Stippling indicates that the cDNA clone hybridized to the corresponding genomic restriction fragment in the restriction map immediately above. No conclusions about the positions of the coding regions within the restriction fragments is implied. B. An alignment of bli-4 cDNA clones based on both sequence (Section  3.2.3) and hybridization data. Exons are represented by boxes in the cDNA clones; introns are represented by lines. Exons and introns are indicated in the cDNA clones only in the regions of the genomic DNA that have been sequenced completely. The number and size of exons encoding the unique  3' ends of cDNA clones encoded by cDNA regions that have not been completely sequenced (blisterin B and blisterin C) is not implied. The genomic restriction map was constructed as follows. First, a Xhol Sall restriction map of the region was made directly from the cosmid KO4F10. Second, the 11 and 11.6 kb Xhol fragments were subcloned and further restriction mapped. The 11 kb Xhol fragment in pCeh181 was further subcloned as two Xhol Pstl fragments, and these fragments were independently restriction mapped. Ambiguities in the map were resolved by Southern analysis. X=Xhol; P= Pstl; S = Sall; E=EcoRl.  96  3 Results  X^E^E^E P 1.9^1.3^1.0^.3 I^1111  E S^S^E S/X E E I 1.2^1 1.3^I . 7 1 0.8 1  1.9^1 .9^1.0  3.3  pCehl 95 pCehl 98 pCeh200 pCehl 97 pCehl99 pCeh 1 96  cDNA Divergence ^ Dense dots indicate that the cDNA clone hybridized Point to the genomic restriction tragement in the map above.  A. Alignment of cDNA clones based on hybridization data.  X^E^E^E P  ^  E S S^E SIX E E  ^  E  1.9^1.3^1.0^.3^1.9^.8^1,0^1 . 2^1.3^.7^0.8^3.3  Blisterin A  A'n  pCeh200 Blisterin B  -(A)n  pCehl97 Blisterin C  (A)n  pCehl96 Alignment based on: °DNA^ Divergence^ Point^  IliSequence data and hybridization \\\\\\Preliminary sequence data and hybridization =Hybridization data only  B. Alignment of cDNA clones based on sequence data.  97  3 Results  Figure 17. Sequencing strategies Arrows indicate individual sequence reactions. # indicates that the sequencing reaction used a custom primer with dye-labeled di-deoxy terminators. All other reactions used dye-labeled forward or reverse M13 primers on templates generated by Exonuclease III deletions (Henikoff, 1984). A. Genomic DNA clones. The upper portion is a restriction map of the bli-4 region. Heavy black lines indicated coding regions determined by hybridization and sequencing. The DNA encoding the 5' ends of the blisterins is labeled. A, B, and C indicate the DNA encoding the unique 3' ends of blisterin A, blisterin B and blisterin C. B. cDNA clones. The cDNA divergence point is indicated.  98  Common 6 ' Sequence  A^  B^C  E^EEP^ES^S^E^SXPEE  X  ,s9 e^e40,146.11,116V  ^r  E^E  049.59e4P,o9e 9 4s f9.e9,-,49439,49eliee if ,  A cDNA divagenco point  e^,iss)  rfP^B BlIstedn  ..•■•■■■■■■►  Blisterin A i^  B$Ist ran cP  49 ti9^cs9 ■-  B  •■■■•■•••  3 Results Figure 18. Blisterin B. Sequence of blisterin B and predicted protein product. Points at which blisterin A and blisterin C begin and the cDNA divergence points are indicated. The predicted secretory signal peptide is underlined. The protease domain is double underlined. The catalytic amino acids, Asp, His and Ser are indicated by an up arrow (t). Potential autocatalytic sites are indicated by an upward arrowhead (•). gcttgccagtaaaaaagtacgttcgcttcgggtcgatgtcagaataaacggaaaagaataacccgttgcaccagcgaatcgtcgaacattttcaatactc 100 acccatatcagtcacccaaaacggatttattattattattagagcatatcatccacatttattctcaaacgcgtgATGCGTATATCGATAGGCCGGATAG 200 MRISIGRI8 CATGGCAAATTCTGGCAGTTTTAATCGCAGTTGCATTCACTATTGAACATGATTCCATTTGCGATGAAAGTATAGGTGCCTGTGGGGAACCAATACATAC 300 A W O I L A V L I A V A F T I E H D S I C D E S I G A C G E P I H T 42 CGTAATACGTTTAGCAAAAAGAGATGATGAGCTTGCACGGCGAATAGCTGCTGATCATGACATGCATGTAAAAGGTGATCCGTTTTTGGATACTCACTAC 400 VIRLAKRDDELARRIAADHDMHVKGDPFLDTHY 75 • • TTCCTTTATCACTCGGAAACAACAAGGACACGGCGACATAAAAGAGCGATTGTTGAACGATTGGATTCACATCCAGCCGTCGAATGGGTTGAAGAACAGC 500 FLYHSETTRIRRHKRAIVERLDSHPAVEWVEE0 108 • GACCGAAGAAGAGAGTCAAAAGAGATTATATTCTCCTGGATAATGATGTTCATCATTCTAACCCUTCCGCCGTTCGGTTTTGAACCGTGATGGTACTCG 600 RPKKRVKRDYILLDNDVHHSNPFRRSVLNRDGTR142  • •^•  •  TAGAGCTCAACGACAGCAGCCACAGTCTCCAGCAGAAATTCCATCACTTCCATTTCCTGATCCACTTTATAAAGACCAGTGGTATTTGCATGGIGGAGCA 700 RAORQOPOSPAEIPSLPFPDPLYKDOWYLHGGA175 GTTGGTGGATATGATATGAATGTTCGTCAAGCGTGGCTTCAAGGATATGCAGGCAGAAATGTTTCAGTTTCGATTCTTGACGATGGAATTCAAAGAGATC 800 VGGYDMNVROAWLOGYAGRNVSVSILDDGICIRD 208  t  ATCCTGATTTGGCAGCGAACTACGATCCACTCGCGTCAACAGATATCAATGATCACGATGATGATCCAACACCACAAAACAATGGAGATAACAAACATGG 900 HPDLAANYDPLASTDINDHDDDPTPONNGDNKHG242  r . Blisterin A 5' end  TACAAGATGTGCTGGAGAAGTAGCTGCACTCGCCGGAAACAATCAATOGGTGTCGGIGTTGCCITCAAAGCAAAAATAGGAGGAGTTCGTATGCTTGAT 1000 TRCAGEVAALAGNNOCGVGVAFKAKIGGVRMLD275 GGAGCTGTTTCGGATTCTGTCGAAGCTGCGTCATTGTCTTTGAATCAAGATCATATTGATATATACTCAGCATCATGGGGACCTGAAGACGACGGAAAGA 1100 GAVSDSVEAASLSLNODHIDIYSASWGPEDDGK 308 CTTTTGATGGTCCAGGTCCACTTGCCCGAGAAGCTTTTTATCGTGGAATCAAGAATGGCAGAGGTGGAAAAGGAAACATTTTTGTATGGGCCAGTGGAAA 1200 TFDGPGPLAREAFY RGIKNGRGGKGNIFVWASGN342  r. Blisterin C CGGTGGATCAAGCCAAGACTCATGTTCAGCTGATGGTTACACAACTTCAGTCTACACGCTTTCCATATCTTCGGCTACTTATGATAATCACAGACCATGG 1300 GGSSODSCSADGYTTSVYTISISSATYDNHRPW375  fi' end TATTTGGAAGAATGCCCATCATCAATTGCAACAACATACAGTTCTGCGGATTTCCGTCAACCAGCAATTGTGACCGTTGATGTTCCTGGAGGATGTACTG 1400 YLEECPSSIATTYSSADFROPAIVIVDVPGGCT 408  3 Results Figure 18. Blisterin B. Continued ATAAGCATACAGGAACTTCTGCGTCTGCTCCATTGGCAGCTGGAATTATTGCTCTTGCTITAGAGGCTAATCCTGAATTAACATGGCGTGATATGCAACA 1500 DKHIGTSASAPLAAGIIALALEANPELTWROM011442  t TCTTGTTCTTCGAACCGCCAACTGGAAACCACTTGAGAATAATCCTGGATGGTCAAGAAATGGTGTTGGACGTATGGTTAGCAATAAATTCGGTTATGGT 1600 LVIRTANWKPLENNPGWSRNGVGRMVSNKFGY0475 CTAATCGATGOGGTGCACTIGTGCAATATGGCAAAAACATGGGAAGACAGTCCCAGAGCAGCACATTTGCACATATGAGATACAGACTTGCTAATCCAA 1700 LIDGGAIVOYCKNMGROSOSSTFAHNRYRIANP 508 ATCCTCGGCCAATTUGGGICGTTTCCAACTTAATTTCACTTTAGATGTGAATGGATGCGAGTCAGGAACCCCTGTITTATATTIGGAACACGTTCAAGT 1800 NPRPIVGRFOLNFTLOVNGCESGTPVLYLEHVOV542 GCATGCTACTGTTCGATATCTGAAACGTGGCGATCTGAAGCTTACGCTCTTCTCACCATCCGGCACCCGATCGGTTCTTCTTCCACCACGACCACAAGAT 1900 HATVRYLKRGDIKITIFSPSGIRSVLIPPRPaD575 TTCAATGCTAACGGATTCCACAAGTGGCCTTTCCTTTCAGTCCAACAATGGGGAGAAGATCCACGTGGAACATGGCTTCTCATGGTGGAATCAGTCACCA 2000 FNANGFHKWPFISVC/OWGEDPRGTWILMVESVT 608 CTAATCCAGCTGCTACTGGTACGTTCCACGATTGGACTCTTCTTCTTTATGGAACTGCTGATCCAGCTCAATCAGGTGATCCTGTCTACTCGGCTACTCC 2100 INPAATGIFHDWILILYGTADPACISGDPVYSATP642  cDNA Divergence point TGCAACATCCCAAGGAGTICTCTCACGCGTTCATCAACTCACTICTCAGGGTGATGAAGTTGTTGAAAGAATTCGAAATCATTGGGAAGTGACATTAGAA 2200 ATSOGVISRVHOLTS0GDEVVERIRNHWEVTLE675 GAGAGTTCACATTGGAATTGGGAGCATGCTCGTGAACATAAATCATTACAAGAATTGAACTCTTCTTCTCGTACCCATAGTTUTTATACTCTTTCACCA 2300 E S S H W N W E H A R E H K S L O E L N S S S R T H S F L Y S F T 708 AATTTCAACCGATTTICTTGATTATTCTTGTCTGTATTTTTGATGCCATTCATCGCCAATTCGCOTTTGAgactatatgaattcattttgggtaaaaaa 2400 KFOPIFIIIIVCIFDAIHROFAV*^ 731 aaaaaaaaaaaaaaaaaaaaa^  2421  101  3 Results Figure 19. Blisterin C. Sequence of blisterin C and predicted protein product. The cDNA divergence point is indicated. The predicted transmembrane domain is underlined in bold. The protease domain is double underlined. The catalytic amino acid, Ser, is indicated by an up arrow (t). CCATGGTATTTGGAAGAATGCCCATCATCAATTGCAACAACATACAGITCTGCGGATTTCCGTCAACCAGCAATTGTGACCGTTGATUTCCTGGAGGAT 100 P W Y L E E C P S S I ATI Y S S AD F R OP A I VIVDVPGG 33 GTACTGATAAGCATACAGGAACTICTGCGTCTGCTCCATTGGCAGCTGGAATTATTGCTCTTGCTITAGAGGCTAATCCTGAATTAACATGGCGTGATAT 200  CID): NIG I S A S A PL A AG  I I AL AL E A N P E L TWRDM67  t  GCAACATCTTGTTCTTCGAACCGCCAACTGGAAACCACTTGAGAATAATCCTGGATGGICAAGAAATGGTGTTGGACGTATGGTTAGCAATAAATTCGGT 300 O HL V I R I A N W K P L E N N P G U S R N G V G R M V S N K F G 100 TATGGTCTAATCGATGUGGTGCACTIGTGCAATATGGCAAAAACATGGGAAGACAGTCCCAGAGCAGCACATTTGCACATATGAGATACAGACTTGCTA 400 Y GL ID G G A L V Q Y GKNMGR OSOSS I F A H M R Y R LA 133 ATCCAAATCCTCGGCCAATTGTGGGICGTTTCCAACTTAATTICACTTTAGATGTGAATGGATGCGAGTCAGGAACCCCTGTTTTATATTTGGAACACGT 500 N P N P R P I V G R F CIL NFILD V N G C E S G T P V L Y L E H V 167 TCAAGTGCATGCTACTGTTCGATATCTGAAACGTGGCGATCTGAAGCTTACGCTCTTCTCACCATCCGGCACCCGATCGGITCTTCTICCACCACGACCA 600 O VH A I V R YLKR G D L K L I L F SPSGIR SVLLPPRP 200 CAAGATTTCAATGCTAACGGATTCCACAAGTGGCCTTTCCTTTCAGTCCAACAATGGGGAGAAGATCCACGTGGAACATGGCTTCTCATGGTGGAATCAG 700 O DFNANGF H K VP F L SVGONGEDPR GTWILMVES 233 TCACCACTAATCCAGCTGCTACTGGTACGTTCCACGATTGGACTCTICTTCTITATGGAACTGCTGATCCAGCTCAATCAGGTGATCCTGTCTACTCGGC 800 ^ T I N P AA T GIP MILLI Y G T ADP AGSGDP V YSA267  cDNA Dvergence point TACTCCTGCAACATCCCAAGGAGTTCTCTCACGCGTTCATCAACTCACTICTCAGGTTGAAGAGTCTGCTCGATCTTCATTTCCAGATTTGACGTCCGGC 900 T P A T S Q G V L SR V H 0 1 TS Q V E E S AR SS F P D L T S G 300 TGGAAATTGTCATGTGATGAATGTAATGGAGGTTGCACAGAATCGAGCTCAGCAACATCATGTTTCGCTTGCAAACATTTAACTCAAACTCTTCGAAATA 1000 W K L S C D E C N G G C T E S S S A I S C F A C K HL T OT LRN 333 AAGGAGGAAGTGGATTCAAGCGTGITCAAAAATGCGATGATACTTACTACTIGGATGGTGACAAGTGTAAAATGTGCTCATCCCATTGICATACTTGTAC 1100 K G G S G F KR VOK C D D T Y Y L D G D K CKMCSSHCHICT367 GAAAGCTGAGGCGTGTGAAACGTGTCCTGGAAGTCTTTTGCTCATTGATGTGGATAATATGCCACACTATGATCATGGAAAGTUGTTGAGTCATGTCCT 1200 K AEA CE T C P G S L L L ID V D N M P H Y D H G K C V E S C P 400 CCAGGATTGGTTGCTGATTATGAATCGAATCTTGTTCAAGCTAAATGTATCTGGAGAAAAGATCTTIGTGGTGACGGATATTACATCAACGCTGTTGGAA 1300 P G L V A D YESNI %/OAK C I WRKOL C G D C Y Y I N A V G 433 AATGTGATCTTTGCGACTCCTCATGCGAAACTTGCACCGCTCCTGGTCCAATGAGCTGTGAGAAATGTTCAAAAGGTTATGGTAAAGGATCGATTGGATA 1400 K CD L C D S S C E IC I AP G P M S C E K C S K C Y GK GS I GY467  102  3 Results Figure 19 Blisterin C continued.  CIGTAGACCGTGTTGTCCTGAAGGATCAACAAAGAGTTGGCAATGTGAGGACTGCTCCAAGCCGGATCCTACACTITTGATTGATTCTAATAAATCATCT 1500 CRPCCPEGSTKSWOCEDCSKPOPTLLIDSNKSS 500 GGATTIGGATTGATGITCTGGATTGTAGTTAGITTGATTGCGGCTIGTGGAATCTGTGCCTGTAAAAAGTGTGCAAGTGAGACGAAAAGCTCAAACGTAG 1600 G F G L M F U I V V S L I A A C G I C A C K K C A S E T K S S N V 533 AATATGCGCCGCTTGCTCAATATAATGCCACAAATGGIGCTATCAATTTAGGAGCACACACTGACGATGAAGACGATGATGAGGATGAAGTATTTGTGAA 1700 EYAPLAQYNATNGAINLGAHIDDED 0 0 E 0 EVFVN567 CCCICAAATTGITTAAaccaaacctttcaaaattcatgtttttaattgtaatttttctgccaacttctttgtgtatgcttctgaaatccgggtaattgtt 1800 PQ I V*^  571  tctttttcaaaattttaaaacctgaaacgttttcaagtggccttttatcatgtgattgtattgtttcttttgtcttataccgggtttatttttattatac 1900 cgtactgtttccatgttataaatagattgttgaattaaaaaaaaaa^  103  1946  3 Results  Figure 20. Blisterin A. Sequence of blisterin A and predicted protein product. The cDNA divergence point is indicated. The protease domain is double underlined. The catalytic amino acid Ser is indicated by an up arrow (1' ). GATGTGCTGGAGAAGTAGCTGCACTCGCCGGAAACAATCAATUGGTGTCGGIGTTGCCTICAAAGCAAAAATAGGAGGAGTTCGTATGCTTGATGGAGC 100 CAGEVAALAGNNOCGVGVAFKAKIGGVRMIDGA 33 TGTTTCGGATTCTGTCGAAGCTGCGTCATTGTCTTTGAATCAAGATCATATTGATATATACTCAGCATCATGGGGACCTGAAGACGACGGAAAGACTTTT 200 VSDSVEAASLSLNODHID1YSASWGPEDDGKTF 66 GATGGTCCAGGTCCACTTGCCCGAGAAGCTTITTAtcgtggaatcaagaatggcagaggtggaaaaggaaacatttttgtatgggccagtggaaacggtg 300 DGPGPLAREAFYRGIKNGRGGKGN1FVWASGNG  99  gatcaagccaagactcatgttcagctgatggttacacaacttcagtctacacgetttccatatcttcggetacttatgataatcacagaccatggtattt 400 GSSODSCSADGYTTSVYTISISSATYDNHRPWYL133 ggaagaatgcccatcatcaattgcaacaacatacagttctgcggatttccgtcaaccagcaattgtgaccgttgatgttcctggaggatgtactgataag 500 EECPSSIATTYSSADFROPAIVIVDVPGGCTDK166 catacaggaacttctgcgtctgctccattggcagctggaattattgctcttgctttagaggctaatcctgaattaacatggcgtgatatgcaacatcttg 600 HIGTSASAPLAAGIIALALEANPELTWRDMOHL 199  t ttcttcgaaccgccaactggaaaccacttgagaataatcctggatggtcaagaaatggtgttggacgtatggttagcaataaattcggttatggtctaat 700 VLRTANWKPLENNPGWSRNGVGRMVSNKFGYGL1233 cgatggtggtgcacttgtgcaatatggcaaaaacatgggaagacagtcccagagcagcacatttgcacatatgagatacagacttgctaatccaaatcct 800 DGGALVOYGKNMGROSOSSTFAHMRYRLANPNP266 cggccaattgtgggtcgtttccaacttaatttcactttagatgtgaatggatgcgagtcaggaacccctgttttatatttggaacacgttcaagtgcatg 900 RP1VGRFOLNFTLDVNGCESGTPVLYLEHVOVH 299 ctactgttcgatatctgaaacgtggcgatctgaagcttacgctcttctcaccatccggcacccgatcggttcttcttccaccacgaccacaagatttcaa 1000 ATVRYLKRGDLKLTLFSPSGTRSVLLPPRPODFN333 tgctaacggattccacaagtggcctttcctttcagtccaacaatggggagaagatccacgtggaacatggcttctcatggtggaatcagtcaccactaat 1100 ANGFHKWPFLSVOOWGEDPRGTWILMVESVTIN366 ccagctgctactggtacgttccacgattggactcttcttctttatggaactgctgatccagctcaatcaggtgatcctgtctactcggctactcctgcaa 1200 PAATGTFHOWILLLYGTADPAOSGDPVYSATPA 399  r  cDNA Divergence point  catcccaaggagTTCTCTCACGCGTTCATCAACTCACTTCTCAGATACTAATCACCATCGCTATCCACCTCGTCGTCAACGCATGAattattttctgttt 1300 TSOGVLSRVHOLTS0 ILITIA1HLVVNA*^427 cactgctcacagettcatagagcttaaaattagatttattgtccctttatccttggttacgatgtgttctgaaacttgtctccccattttctgtgtcatt 1400 ttcccttttgacctcactatgatgaatactttttacgagaaaaaaaaaaaaaaaaaaaa^  104  1459  3 Results 3.2.3 Analysis of bli 4 Genomic DNA preliminary sequence data -  Preliminary sequence data for the genomic DNA including exons common to all three cDNA clones was derived from the plasmid pCeh202, and is presented in Figure 21. This sequence includes exons encoding all of the cDNA sequences 5' to the cDNA divergence point. The relationship between the genomic sequence and the cDNA clones is presented in Figure 16b. The ATG start codon of the blisterin B open reading frame occurs at nucleotide position 374. The cDNA divergence point occurs at the 3' end of Exon XII and coincides with an exon/intron boundary. This observation supports the hypothesis that the unique 3' sequences of the cDNA clones arose through alternative splicing. Genomic DNA sequence corresponding to the unique 3' end of blisterin A was derived from pCeh205, and is presented in Figure 22. All of the blisterin A unique 3' sequence occurs in a single exon. This exon is deleted by the e937 deletion (Section 3.2.5.2). Genomic DNA sequence corresponding to the unique 3' end of blisterin C was derived from pCeh207, and is presented in Figure 23. pCeh207 includes genomic DNA encoding all of the unique 3' sequence of blisterin C. The blisterin C sequence is encoded by at least five exons. The sequence labeled "Exon XV" in Figure 22 falls into a gap in the pCeh207 sequence data, and therefore may be encoded by more than one exon. In combination with the hybridization of cDNA clones to genomic DNA presented in section 3.2.2.1 and summarized in Figure 16, the genomic sequence data presented here supports the conclusion that the genomic DNA encoding the three cDNA clones is entirely contained within the cosmid KO4F10.  105  3 Results  Figure 21. Blisterin common genomic sequence. DNA sequence encoding the 5' ends of blisterin cDNA clones. Introns are lower case; exons are upper case. tgcttattccccgaatgaaactttcccagcttagtagtgcctccgatagccaattttaaacaaatattctgaaattattaatttttcttgtttttcgcga 100 r170 Exon I aaaaactgtcattttctttattattctettttttaaccgctttccaaattaattcggctatttttaaagCGTTCGCTTCGGGICGATGTCAGAATAAACG 200 r298 Intron I GAAAAGAATAACCCGTTGCACCAGCGAATCGTCGAACATTTTCAATACTCACCCATATCAGTCACCCAAAACGGATTTATTATTATTATTAGAGCATgtg 300 r346 Exon II ^rStart of blisterin B 0RF agttcgttgaatttgtgaatttattctataaatctacaatttcagATCATCCACATTTATTCTCAAACGCGTGATGCGTATATCGATAGGCCGGATAGCA 400 ,  r450 Intron II^ X.95 95 Exon III TGGCAAATTCTGGCAUTTTAATCGCAGTTGCATTCACTATTGAACATGgttccgtaagttatttcaatcacntctaaaattentattatacttATTCCA 500 TTTGCGATGAAAGTATAGGTGCCTGTGGGGAACCAATACATACCGTAATACGTTTAGCAAAAAGAGATGATGAGCTTGCACGGCGAATAGCTGCTGATCA 600 TGACATGCATGTAAAAGGTGATCCGTITTTGGATACTCACTACTTCCTTTATCACTCGGAAACAACAAGGACACGGCGACATAAAAGAGCGATTGTTGAA 700 X788 Intron III CGATTGGATTCACATCCAGCCGTCGAATGGGTTGAAGAACAGCGACCGAAGAAGAGAGTCAAAAGAGATTATATTCTCCTGGATAATgttagttttttag 800 r898  aaatattcaacctacatggatatttatccagaatcattaccgtttttctttttatcttagctttagagttttttcctgacaattaatatttttccagGAT 900 Exon IV GTTCATCATTCTAACCCCTTCCGCCGTTCGGTTTTGAACCGTGATGGTACTCGTAGAGCTCAACGACAGCAGCCACAGTCTCCACGAGAAATTCCATCAC 1000 r1041 Intron IV TTCCATTTCCTGATCCACITTATAAAGACCAGTGGTATTTGgtgagtttcaaattcaaattttctttaaagagaaaaaaaaaccactggatacatttaca 1100 r1102 Exon V gCATGUGGAGCAGTTGGTGGATATGATATGAATUTCGTCAAGCGTGGCTTCAAGGATATGCAGGCAGAAATGTTTCAGTTTCGATTCTTGACGATGGA 1200 ATTCAAAGAGATCATCCTGATTTGGCAGCGAACTACGATCCACTCGCGTCAACAGATATCAATGATCACGATGATGATCCAACACCACAAAACAATGGAG 1300 r1308 Intron V^ 0390 Exon VI ATAACAAgtaaggatttttaaacacaagttcttcagaaatttttatggncattaantgttttttgggccaaggtacaagnntgtntccgACATGGTACAA 1400 GATGTGCTGGAGAAGTAGCTGCACTCGCCGGAAACAATCAATGTGGTGTCGGTGTTGCCTTCAAAGCAAAAATAGGAGGAGTTCGTATGCTTGATGGAGC 1500 TGTTTCGGATTCTGTCGAAGCTGCGTCATTGTCTTTGAATCAAGATCATATTGATATATACTCAGCATCATGGGGACCTGAAGACGACGGAAAGACTTTT 1600 GATGGTCCAGGTCCACTTGCCCGAGAAGCTTTTTATCGTGGAATCAAGAATGGCAGAGGTGGAAAAGGAAACATTTTTGTATGGGCCAGTGGAAACGGTG 1700 r1708 Intron VI^ r1754 Exon VII GATCAAGgtatttaggttttgaaaagagaattttgtgcaagaagaatgtttagCCAAGACTCATGTTCAGCTGATGGTTACACAACTTCAUCTACACGC 1800 r1835 Intron VII TTTCCATATCTTCGGCTACTTATGATAATCACAGgtatgattttatattactataaaccataatttgictgaatgaaaattgatattcctgcaagataa 1900 r1948 Exon VIII aaaatggatttccctactcatcaaattcttattctacttgtctacagACCATGGTATTTGGAAGAATGCCCATCATCAATTGCAACAACATACAGTICTG 2000 CGGATTTCCGTCAACCAGCAATTGTGACCGTTGATGTTCCTGGAGGATGTACTGATAAGCATACAGGAACTICTGCGTCTGCTCCATTGGCAGCTGGAAT 2100 r2124 Intron VIII^ r2176 Exon IX TATTGCTCTTGCTTTAGAGGCTAAgttagtttattaattccagaaaacttgctcactgtttttttttaatttcagTCCTGAATTAACATGGCGTGATATG 2200  106  3 Results  Figure 21. Blisterin common genomic sequence. Continued. CAACATCTTGTTCTTCGAACCGCCAACTGGAAACCACTTGAGAATAATCCTGGATGGTCAAGAAATGGTGTTGGACGTATGGTTAGCAATAAATTCGGTT 2300 ATGGTCTAATCGATGGTGGTGCACTTGTGCAATATGGCAAAAACATGGGAAGACAGTCCCAGAGCAGCACATTTGCACATATGAGATACAGACTTGCTAA 2400 r2454 Exon X r2407 Intron IX^ TCCAAAgtgaggttcaatcatagatttctctggaaaaaatacgatcattacagTCCTCGGCCAATTOGGGICUTTCCAACTTAATTTCACTITAGATG 2500 r2564 Intron X TGAATGGATGCGAGTCAGGAACCCCTUTTTATATTIGGAACACGTTCAAGTGCATGCTACTGgtcagttttcaattgatactgttctctacaaatctta 2600 rExon XI tcaaattgaattcagTTCGATATCTGAAACGTGGCGATCTGAAGCTTACGCTCTTCTCACCATCCGGCACCCGATCGGTTCTTCTTCCACCACGACCACA 2700 AGATTICAATGCTAACGGATTCCACAAGTGGCCTTTCCITTCAGTCCAACAATGGGGAGAAGATCCACGTGGAACATGGCTICTCATGGTGGAATCAGTC 2800 r2823 lntron XI ACCACTAATCCAGCTGCTACTGgtgagaatttatttaaatgt...1/...taaaatttgaacgaacggaaggattttcgtttttctttttttttttaatg 2892 aaaaatccatitttcccaaaaaaacgacagcnctgatttctatttaaaaaatatgttcgcctttgaagagtactgtaagttcaaacttttgttgctgetg 2992 aattttggttgattttttcatagtttttcaataagaccatatttatttatttaaaaaaaacgcaaatttttagcggttaacaaaaatttgtcgtgctgag 3092 accggtttcgatacttctggcgtaaaatttcgtgactaggtaatagtaaaaaattcaaaattgaagcgaaaaatttccctataaatatttganccntata 3192 r3213 Exon XII tggtgatttcagGTACGTTCCACGATTGGACTCTTCTTCTTTATGGAACTGCTGATCCAGCTCAATCAGGTGATCCTGTCTACTCGGCTACTCCTGCAAC 3292 r3344 cDNA divergence point ^Intron XII ATCCCAAGGAGTTCTCTCACGCGTTCATCAACTCACTTCTCAGgtacgcacatttttcggaggattcggatagaatcatattgtnctttgtgtgtgtgtt 3392 gtgttgnctcctttgnttttatttncatatttaggggcccttgaattcaaactgatttacccttcccctctnctcactttctatctaatttggtttcggt 3492 gatagttgtgttttaattttgaaaattcccaacccacagagtgtgtgtccttgcgagtgtgtgagcctggtgtgataatcaattttcaaattgaaaaatt 3592 gaaaaaatcaacaaaatgcctaattttctcataattattcatcgtecttacctctggcacaattttcagtntattcatgtttctattcataaattncttg 3692 tgtatgtatctgtatgtctctatgtaaatgtgtcgctgactttttaatgtttcttgtgaactcgtcttcgtctatgttcttgttgttattgtgatgtgaa 3792 tcaactttttaattaaaaagagaattgattaacagttatatttatttctttatttctttaatttctttgttcaactagacgtccacattgttttctgcag 3892 3' end of pCeh202^  3894  107  3 Results Figure 22. Genomic DNA encoding the unique carboxy terminal of Blisterin A. DNA sequence derived from pCeh205. A single exon encodes the unique 3' end of blisterin A. rEcoRI guItcgaaaagaaaaatgtaatataactgtaaattgattctctgagtgtcacgtcgtcgtctttttaaattggttctacggtgtttttaattgttattt 100 cacttgcatgtcgttgggtggcacaaaatattaagtgaaaagtaagattttcagagaagtaaaaaataagtgtataagcagaataaaaatatgatgagcg 200 acctcgtagtgaataaacttctacggtaaaaataaaataaaaaaaaaggaaaaaatactggangcccgtaagngtaaantaaaaaatngaaaaaaaatcc 300  r 3' unique sequence of blisterin A tttgggnaaaaatttggnccccctttttaagATACTAATCACCATCGCTATCCACCTCGTCGTCAACGCATGAATTATTTTCTGTTTCACTGCTCACAGC 400 TTCATAGAGCTTAAAATTAGATTTATTGTCCUTTATCCTIGGTTACGATGTGTTCTGAAACTTGICTCCCCATTTICTGTGTCATTTTCCCTITTGACC 500  r Poly A site TCACTATGATGAATACTTTTTACGAGaacattttgttggttgtccaaaaattaaacaaaaaaaacaaatctgcatgctcccagcgtttcatctattttcc 600 rSaTI tattcccccactacagattggaatgccgagtgttctcgttgtcattcaaatcatcgtcatatgtctcgtacttggcgttggcctttatqtcgac^694  108  3 Results  Figure 23. Genomic sequence encoding the unique 3' end of blisterin C. Genomic sequence derived from pCeh207 (top) compared to the unique 3' end of blisterin C (bottom). Gaps in the genomic sequence are indicated by dots (...); introns are indicated by dashes (---); matched positions are indicated by vertical bars (1 ). Genomic cacgacattgnccanceccaaacttttttgaatngncgaaacgcagtagatattaeggtagatcgttgttttacagGGTGAAGAGICTACTCGATCTTCA 1^1111111111^11111111111 GTTGAAGAGTCTGCTCGATCTTCA BLi C ^ Genomic TUCCAGATTTGACGT.CGGCTGGAAATTGTCA—TGATGAATGTAATGGAGGTgaggaattgattttcgcgattactataaatgtttaaaaagtatttt 1111111111111111^1111111111111111^1111111111111111111 Sli C TTTCCAGATTTGACGTCCGGCTGGAAATTGTCATGTGATGAATGTAATGGAGGT ^ Genomic cggtttttttcagaaatttcaaagncaaaaattaaacattgcagaaattncaaaatcaactggaaatattgtccagatgcgaaattttgcgactattgct Bli  C  Genomic ncaaaaatacggtacccggtctcgacacggccgattttttcaatccaaaagaatgcgcgcctttaaggaatactgtagtttccacattttccgcgttgct Bti C Genomic cgattttcgtggagctccaattcaccntatagtgagtcntatggcaattcaaa ^ Genomic sequencing.gap ^ Bli C ^ Genomic  TGCACAGAATCGAGCTCAGCAACATCATGTTTCGCTTGCAAACA  Genomic  sequencing  gap  Bli C TTTAACTCAAACTCTTCGAAATAAAGGAGGAAGTGGATTCAAGCGTUTCAAAAATGCGATGATACTTACTACTTGGATGGTGACAAGTGTAAAATGTGC Genomic  Genomic  sequencing  gap  Bli C TCATCCCATTGTCATACTTGTACGAAAGCTGAGGCGTGTGAAACGTGTCCTGGAAGTCTTTTGCTCATTGATGTGGATAATATGCCACACTATGATCATG  Genomic ^ Bli  aaaacnggggnattgnntgtgaatgttnagtncttccctttnnnaggctaaaccc  C GAAAGTGTGTTGAGTCATGTCCTCCAGGATTGGTTGCTGATT ^  Genomic tacntccnaccacctgaaacttccccccttagaattatttaccatcttgaaatnccttggcctcctaacccattnnccctcatgttacaagtttttttgt Bli C Genomic ttaaaccaatgtgaaatttcagATGAATCGAATCTTGTTCAAGCTAAATGTATCTGGAGAAAAGATCTTIGTGGTGACGGATATTACATCAACGCTGTTG 111111111111111111111111111111111111111111111111111111111111111111111111111111 Bli C ^ ATGAATCGAATCTTGTTCAAGCTAAATGTATCTGGAGAAAAGATCTTIGTGGTGACGGATATTACATCAACGCTGTTG Genomic GAAAATGTGATCTTTGCGACTCCTCATGCGAAACTTGCACCGCTCCTGGTCCAATGAGCTGTGAGAAATGTTCAAAAGGTTATGGTAAAGGATCGATTGG 11111111111111111,11111111111111111111111111111111111111111111111111111111,1111111111111111111111111 Bli C GAAAATGTGATCTITGCGACTCCTCATGCGAAACTTGCACCGCTCCTGGTCCAATGAGCTGTGAGAAATGTTCAAAAGGITATGGTAAAGGATCGATTGG  109  3 Results Figure 23. Genomic sequence encoding the unique 3' end of blisterin C. Continued.  Genomic ATACTGTAGACCGTGTTUCCTGAAGGATCAACAAAGAGTTGGCAATUggtaaggaagacactotcaaaatcttccaggtatctcacaatttattmt 1111111111111111111111111111111111111111111111111 Bli C ATACTGTAGACCGTGTTGTCCTGAAGGATCAACAAAGAGTTGGCAATGT ^ Genomic caaGAGGACTGUCCAAGCGGGATCCTACACTTTTGATTGATTCTAATAAACCATUGGATTTGGATTGATGNICTGGATTGTAGTTAGTTTGATTGCGG 1111111111111111^1111111111111111111111111111111^11111111111111111111^111111111111111111111111111 Bli C ---GAGGACTGCTCCAAGCCGGATCCTACACTTTTGATTGATTCTAATAAATCATUGGATTIGGATTGATGTTCTGGATTGTAGTTAGTTTGATTGCGG Genomic CTTGIGGAATCTGTGCCTGTAAAAAGTGTGCAAGTGAGACGAAAAGCTCAACCGTAGAgtaagccttgctagttatcgtotccgtaattcgaattttaat 111111111111111111111111111111111111111111111111111^111111 Bli C CTTGTGGAATCTGTGCCTGTAAAAAGTGTGCAAGTGAGACGAAAAGCTCAAACGTAGA ^ Genomic atttttaGATATGCGCCGCTTGCTCAATATAATGCCACAAATGGTGCTAACAATTNAGGAGCACACACTGACGATGAAGACGATGATGAGGATGAAGTAT 11111111111111111111111111111111111111111 ^11111^11111111111111111111111111111111111111111111 Eli C ^ GATATGCGCCGCTTGCTCAATATAATGCCACAAATGGTGCTATCAATTTAGGAGCACACACTGACGATGAAGACGATGATGAGGATGAAGTAT Genomic TTGTGAACCCTCAAATTUTTAAACCAAACCITTCAAAATTCATGTUTTAATTGTAATTTUCTGCCAACTTCTITUGTATGCTIGTGAAATCCGGGT 111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 ^111111111111 Bli C TTGTGAACCCTCAAATTUTTAAACCAAACCTTICAAAATTCATUTTTTAATTGTAATTTTTCTGCCAACTTCTTTGTGTATGCTICTGAAATCCGGGT Genomic AATTGTTTCTITTCCAAAATITTAAAACCTGAAACGTTITCAAGTGGCCITTTATCATGTGATTGTATTGTTTCTTTTGTCTTATACCGG ^  1111111111111^1111111111111111111111111111111111111111111111111111111111111111111111111111  Bli C AATTGTTTCTITTTCAAAATTTTAAAACCTGAAACGTTTTCAAGTGGCCTITTATCATGTGATTGTATTGTTTCTTTTGICTTATACCGGGTTTATTITT Genomic ^ Bli C ATTATACCGTACTUTTCCATGTTATAAATAGATTGTTGAATT 3' end of blisterin C.  3 Results  3.2.4 Analysis of the predicted bli-4 proteins 3.2.4.1 Identification of similarity with kex2-like serine proteases A search of the translated EMBL DNA data base, SWISSPROT using the FASTA search algorithm with a k-tup value of 2 was conducted using all three predicted proteins. All proteins with an optimized similarity score of 100 or greater are listed in Table 8. All of the proteins listed in Table 8 are subtilisin-type serine endoproteases. The proteins identified in the FASTA search of the SWISSPROT library fell into two distinct groups; those with higher scores, ranging from 846 to 1316, and those with lower scores, ranging from 100 to 159. All of the proteins in the higher scoring group were eukaryotic subtilisin-type serine endoproteases. The scores in the lower scoring group were prokaryotic subtilisins. The higher scoring group included mouse, rat and human furin, mouse and human prohormone convertase 2 (PC2), mouse prohormone convertase 1 (PC3), the Saccharomyces cervisiae KEX2 gene product kex2, and the Kluvermyces lactis KEX1 gene product Kex1. Because some of these proteins are more than 95% identical to each other, a representative of each type will be used in sequence comparisons to the blisterins. Human furin (hfurin) will be used to represent human, rat and mouse furin. Human PC2 (hPC2) will be used to represent human and mouse PC2. S. cerevisiae kex2 will be used to represent kex2 and its K. lactis homologue, kex1. The eukaryotic proteins as a group will be referred to as the kex2-like proteins, because kex2 was discovered first. The identity scores reported by FASTA for the proteins listed in Table 8 are for the region of highest similarity rather than the whole protein. To determine the overall identity scores of the blisterins and the kex2-like  111  3 Results proteins, the ALIGN program was used to compare the proteins listed Table 9. The overall identity scores between blisterin B and the eukaryotic kex2like enzymes ranged from 25.8% for kex2, to 36.4 `)/0 for hfurin. Subtilisin scored much lower, at 11.0% identical amino acid positions. The identity scores between the blisterins and the kex2-like proteins are in the same range as scores obtained when the kex2-like proteins are compared to each other (Table 9). Thus, blisterins are as related to the kex2-like proteins as the kex2-like proteins are to each other. An alignment of blisterin B and the mature subtilisin amylosaccharits revealed a domain of similarity between subtilisin and blisterin B, residues Tyr(191) to Gln(441) (Figure 24). Within this domain, 26% of amino acid residues are identical. This domain includes most of the 275 residue subtilisin protein, and will be referred to as the "Subtilisin-domain" of blisterin. Alignment of the eukaryotic kex2-like enzymes revealed regions of similarity extending beyond the subtilisin domain (Figure 25). These regions begin near the beginning of the blisterin B and extend to residue Ala(629). The domain of similarity to the amino side of the subtilisin domain is referred to as the "Pre-domain" of the blisterins. The Pre-domain includes residues Met(1) to Gly(190). The subtilisin domain extends from Tyr(191) to Gln(441). The domain of similarity among the kex2-like endoproteases to the carboxyl-side of the Subtilisin Domain has been noted previously (Fuller, Brake and Thorner, 1989b; Smeekins and Steiner, 1990) and is referred to as the "P-domain". The P-domain includes residues His(442) to Ala(629). All three domains, the Pre-domain, the Subtilisin-domain and the P-domain occur within the blisterin common sequence, which ends at Gln(658).  112  3 Results The identity scores obtained by the alignment of whole protein sequences provides a rough index of relative similarity of the sequences. To obtain a more sensitive ranking of protein identities, ALIGN was used to compare the individual blisterin domains to the corresponding segments of the kex2-like proteins (Table 10). In the pre-domain of blisterin B, the highest identities were obtained for mPC3 (23.1%), hfurin, (21.7%) and hPC2 (21.6%). The lowest identity was obtained for kex2 (15.1%). In the subtilisin-domain, the highest scores were for hfurin (67.8%), while the lowest score for the eukaryotic proteins was obtained for kex2 (43.3%). For the P-domain, the highest identity was obtained for hPC2 (38.9%), hfurin (38.3%), and mPC3 (37.2%). Again, the lowest score was obtained for kex2 (23.5%). The only clear pattern that emerged from these alignments was that the identity of blisterins with kex2 was consistently lower than that of blisterins with the proteins of higher eukaryotes. Based on these results, it is not possible to conclude that the blisterins are most like any one of the kex2-like proteins. To identify small regions of similarity between the blisterins and the kex2-like proteins, the blisterin B sequence was divided into segments of 50 amino acid residues and used individually to search a library of kex2-like sequences with the FASTA program (Figure 26). In the pre-domain, only segment four, residues 151-200, identified a similar region in all of the protein sequences. Thus, most of the identity between the Pre-domain and the kex2-like proteins in Table 10 occurs in the last 25% of the Pre-domain. Exceptions to this observation were similarities identified with segment two to hPC2 and mPC3, and segment three to mPC3. The more extended similarity of these proteins with the blisterin pre-domain is reflected by the high percentage of identical amino acid positions in Table 10. Segments five  113  3 Results to nine span the subtilisinldomain. These segments identified similar regions in all of the proteins, with the exception of segment seven, which did not identify a similarity in subtilisin. Segment seven also had the lowest scores among the subtilisin-domain segments for all of the other proteins, indicating that the region of residues 301 to 350 are less conserved that other parts of the subtilisin domain. Segments 10 to 13 span the P-domain. Segments 10 and 12 identified stronger similarities than segments 11 and 13, with the exception of kex2, which did not have a region similar to segment 10. In general, the P-domain segments had lower similarity scores than did the subtilisin domain segments. The blisterin sequences diverge from each other at blisterin B residue 658, in segment 14. Of the segments in the unique carboxyl-ends of the blisterins, only one identified a similar region in the kex2-like proteins. A segment of blisterin C identified a region of similarity in hfurin (See section 3.2.4.2.4).  114  Table 8. Summary of Fastaa search of SWISSPROT protein database using Blisterin C. Protein ID  Protein name  Optimized similarity score'  Identityc  Reference  )  FURI$MOUSE  Mouse furin (mfurin)  1316  52.8% (511)  Hatsuzawa et a/. , 1990  FURI$RAT  Rat furin (rfurin)  1322  50.0% (558)  Misurni, Sohda and Ikehara, 1990  FUR1$HUMAN  Human furin (hfurin)  1309  52.4% (510)  Roebroek et al., 1986; van den Ouweland et al„ 1989  FURH$MOUSE  Mouse prohormone convertase 1 (mPC1 also known as mPC3)  1301  43.7% (565)  Seidah et at., 1990 Smeekins et al., 1991  KEX2$YEAST  S. cerevisiae kex2 (kex2)  917  36.2% (527)  Mizuno et al., 1988  KEXl$KLULA  K. lactis kexl (klkexl)  846  35.4% (492)  Tanguy-Rougeau, Wesolowski and Fukuhara, 1988  KEX2$HUMAN  Human prohormone convertase 2 (hPC2)  1139  42.5% (570)  Smeekins and Steiner, 1990  KEX2$MOUSE  Mouse prohormone convertase 2 (mPC2)  1139  43.0% (570)  Seidah et al., 1990  SUBT$BACSA  B. subtilis amylosaccaritis mature subti I isin  100  34.8% (92)  Yoshimoto et al., 1988  aLippman and Pearson (1985). bCalculated for region of similarity using the PAM250 matrix. cPercentage of identical amino acids within the region of similarity. Numbers in brackets indicate the number of nucleotides in the region of similarity identified by FASTA.  3 Results  Table 9. Overall identity scoresa of blisterin B and kex2-like proteins. Protein blisterin B hfurin kex2 hPC2 mPC3 Subtilisin  Subtilisin 11.8% 9.8% 9.5% 11.6% 9.0% 100%  mPC3 35.8% 40.2% 28.3% 38.8% 100%  hPC2 35.0% 35.0% 62.0% 100%  kex2 24.8% 26.5% 100%  aIdentity scores were obtained using the ALIGN program.  116  hfurin 36.4% 100%  blisterin B 100%  3 Results  Figure 24. Alignment of blisterin B and B. subtilus subtilisin. Alignment of blisterin B (top) and Bacillus subtilus amylosaccharitis subtilisin. Within the blisterin B interval On o° to G1n441, 26% of the amino acid positions are identical. Active site amino acids are indicated by asterisks (*); gaps introduced into the alignment are indicated by hyphens (---); identical positions are indicated by vertical bars (:::); conservative positions are indicated by dots (...). The alignment was determined using ALIGN. MRISIGRIAWOILAVLIAVAFTIEHDSICDESIGACGEPIHTVIRLAKRDDELARRIAADHDMHVKGDPFLDTHYFLYHSETTRTRRHKRAIVERLDSHP AQ ^ SVPYGISQ ^  IKAPA ^  X190 Subtilisin AVEWVEEORPKKRVKRDYILLDNDVHHSNPFRRSVLNRDGTRRAOROOPOSPAEIPSLPFPDPLYKDOWYLHGGAVGGYDMNVROAWLOGYAGRNVSVSI •-• •^ • ^ LHS ^ QGYTGSNVKVAV  Domain LDDGIORDHPDLAANYDPLASTDINDHDDDPTPONNGDNKHGTRCAGEVAALAGNNOCGVGVAFKAKIGGVRMLDGAVSDSVEAASLSLNODHIDIYSAS •  •  ..  IDSGIDSSHPDLNVR----GGASFVPSETNP---YODGSSHGTHVAGTIAALN-NSIGVLGVAPSASLYAVKVLDSTGSG ^ OYSWIINGIE WGPEDDGKTFDGPGPLAREAFYRGIKNGRGGKGNIFVWASGNGGSSODSCSADGYTTSVYTLSISSATYDNHRPWYLEECPSSIATTYSSADFROPAI-V .^.^: : ...... : . : . : .^...^ •^•^•^• WAISNNMDVINMSLGGPSGSTALKTVVDKAVSSGIVVAAAAGNEGSSGSSSTVGYPAK-YPSTIAVGAVNSSNOR ^ASFSSAGSELDVMAPGVSI  r441 End of Subtilisin Domain TVDVPGGCTDKHIGTSASAPLAAGIIALALEANPELTWRDMOHLVLRTANWKPLENNPGWSRNGVGRMVSNKFGYGLIDGGALVOYGKNMGROSOSSTFA • •^• OSTLPGGTYGAYNGTSMATPHVAGAAALILSKHP--TWTNAO ^ HMRYRLANPNPRPIVGRFOLNFTLDVNGCESGTPVLYLEHVOVHATVRYLKRGDLKLTLFSPSGTRSVLLPPRPODFNANGFHKWPFLSVONGEDPRGTW -VRDRL  ESTATYL  WLLMVESVTTNPAATGTFHDWTLLLYGTADPAOSGDPVYSATPATSOGVLSRVHOLTSOGDEVVERIRNHWEVTLEESSHWNWEHAREHKSLOELNSSSR GDSFYYG  KGLINVQAAAQ  THSFLYSFTKFQPIFLIILVCIFDAIHROFAV*  117  3 Results Figure 25. Alignment of blisterin B and kex2-like proteins. Amino acid residues in the kex2-like proteins identical to blisterin B are indicated by hyphens (-). Similar amino acid residues are lower case. Dissimilar amino acids are upper case. B1iB: blisterin B; hfur: human furin; mPC2: mouse prohormone convertase 2; mPC3: mouse prohormone convertase 3; kex2: yeast kex2. ALIGN was used for protein alignments r.  Pre-domian  MRISIGRIAWQILAVLIAVAFTIEHDSICDESIGACGEPIHTVIRLAKR DDELARRIAADHDMHVKG DPFLDTHYFLYH SETTRTRRHKRAIVER hfur -Elr^p-L1Wv-AAtGt1v1 La^aDaq -qKvF-NtwAvrIP GgPAv-nsv-RK-gfLNL-QI FGDY-HFW-RGV-K-SLSPH-PRHSmPC2 -kGgCVS q-k^A-ag-Lf^-VMVfasAeR-vF-NHF-veLHKGgeDK--qv--E-gfG- RKL--AEGL-HF--NGLAKAK--RSLHHKOO mPC3 -EQr^g-TL^qCt--af FCVW-aL-sVK akrQFvNEwA-eIP Ggq-A-sA--eELgYd1L-OIGS-EN---GK-K-HPR-S--SALH-TKkex2 -kvrKY -tLCFWWAfstS-ivssqqi^PLKdliTsRqYFa-E^SneT-s-LEeMNpnWkYEHDVRG-PN--VFSKELLKLGK-SSLEELOGD 0' bliB LDSHPAVEWVEEORPKKRVKRDYILLDNDV HHSNPFRRSVLNRDGTRRAOROOPOSPREIPSLPFPDPLYKDOWYLIIG ^GAVGGYDMNVROAWLOG hfur -QRE-Q-Q-L-Q-VA-r-T---^VYQEPT^  --kFPq----S-^-tOR-1--ka--A--  mPC2 -ERD-R-KMALO-EGFD-K--G-^R-I NEIDINMN^  ---Ftd----InTGQAdgtP-L-L--Ae--EL-  mPC3 -SDDDR-T-A-Q-YE-E-S--S ^VQKD-A-DL^ -N--mUnq----qdTRMTa-1PKL-1h-IPv-Ek^ NN-H--SVH-LFPRNDL-K- -PVPAPPMDSSLL-VKEA- DK-SIN---FeR--h-VnPS ^FP-S-i--LdL-Ynn kex2 Subtilisin domain OHS YAGRNVSVSILDDGIQRDHPDLAANYDPLASTDINDHDDDPTPQNN GDNKHGTRCAGEVAALAGNNQCGVGVAFKAKIGGVRMLDGA VSDSVEAASL hfur -t-hgiV ^ ekn -^g- -G--F-v--q-P--0-rYtOMn- r ^ v-n-gV^ fn-r ^ e -t-a---R-mPC2 -t-kg-t1g-m- -dYL ^ s -nAE- Y-fssn-PY-F-rYtDDWF-s ^ s-A-n -I ^ fns-va-i----OpFmt-ii--s-i mPC3 It-kg-VItv----Lewn-t-iY ^ E- Y-f--n-h--F-rYdLTne ^ i-MQ-n-hk ^ fnsdN -i ^ I -t-ai--s-i kex2 It-Ag-VAa-v---LdYenE--Kd-fCAEG-W-f--ntnL-k-r^Ls-dY ^ i--KK- -F ^ gfn- -s-i-i-s-DIttEd ^ !DUB SLNODHIDIYSASWGPEDDGKTFDGPGPLAREAFYRGIKNGRGGKGNIFVWASGNGGSSODSCSADGYTTSVYTLSISSATYDNHRPWYLEECPSSIATT hfur g- pn -h ^ V- -ar--e- -f--vsq- -L s ^ reh -nC- -n-i ^ OFgnV---S-a-s-t1--mPC2 -HMpqL ^ t-n- -V- -RDvtLq-mAd-vnk ^ s-y ^ d---Yd- -nC---as-mw-i--n--in-grTaL-D-s-s-t1-smPC3 gf-pg-v ^ n^ Ve- -r--qk -EY-v-q--O- s- -N ^ rOG-n-dC- -d-i- i ^ kex2  ^ trG-n-nY----n-m-Sit-saIDhkdLh-P-S-g-saVm-v-  r•  P-domain  YSSADFRQPA IVTVDVPGGCTDKHIGTSASAPLAAGIIALALEANPELTWRDMOHLVLRTANWKPLENN PG WSRNGVGRMVSNKFGYGLIDGGALV hfur ---griOneKq ---t-lrOK--es ^ t-^Kn ^ vq-skPAh-na- D -aT ^ k--hsy----1-a--mmPC2 f-ngRK-n-eAGva-t-1Y-n--Lr-s----a--E- -vf ^ Lg ^ tvL-skrnq-hdeVhQ -r ^ ^ mPC3 -tsa-Lhnd--et ^ f ^ n ^ vw-seyd--as-^-kk--a-L--nsr--f--lnaK--kex2 ---gs^gEY-HsS-iN-R-Sns-g^-a ^ vYT-L ^ n ^ v-y-SiLs- VG--k-AD-D-rDsam-kkY-hry-f-K--aHK-I bliB OYCKNMGROSO SSTFAHMRYRLANPNPRPIVGRFOLNFTLDVNGCESGT PVLYLEHVOVHATVRYLKRGDLKLTLFSPSGTRSVLLPPRPODFNAN G hfur aLaq nwTtV apORKCi I DILte-kD-GK-IEvrK-vt a-LgepNhITR---a-arL-1s-NR----AiH-V--M----t--aa--h-ys-d mPC2 kMa- dwktV peR-HCvGG sVqd-EK-PsTGK-V1--ttda--gKeNF-R ^ aVI--nATR----ninmT--M--k-i--sr--r-DdakVmPC3 dLadPrtwrnV pekKeCv VkDn-Fe--alKanGEvIVEiPtra--gOeNaIKS ^ Fe-  ^  -  kex2 eMs- twenVNa-wFylpTLYvsgstnstEETLESVi-isEds LOdANFKRi---T-TvdiDTEI--TTtvd-i--a-iI-N-GVV--r-Vsse Blisterin divergence point FHKWPFLSVQQWGEDPRGTWLLMVESVTTNPAATGTFHDWTLLLYGTA DPAQSGDPVYSATPA TSOGVLSRVHOLTSOGDEVVERIRNHWEVTLEESS hfur -ND-A-MTTHS-D---S-E-V-EI-NTSE ANNY--LTKF--V ^ EGLPV-PE--gCK-L--S0aCV-CeeGF-LliqkSCVqHCPpGFap0VLdT mPC2 -D----MTTHT----A----T-EL GFVGSAPQK-VLKE---M-H--Q SAPYIDQV- ^RDYQ-k Lam-kKee-eeeL--a---sLkSILNkN mPC3 -KN-D-M--HT---N-V---T-KITDMSGRMONE-RIVN-K-I-H--S SOPEHMKOPRVY-sYN-V-nDRrG-ekmvnVVekRPtqKsLnGn1LvPknkex2 -KD-T-M--AH---NGV-D-KIK-K --ENGHRID--S-R-K-F-ESI-SSKTETF-F ^GnDKeevepaat-StVsQYsASstsisI-A  118  3 Results  Figure 25. Alignment of blisterin B and kex2-like proteins. Continued. bliB HWNWEHAREHKSOELNSSSRTHSFLYSFUNPIFLIILVCIFDAIHROFAV hfur -ysT-ndV-TIrASVCapchAsCaTCOgPaLTdCLSCPsHaslDPvEqTCSROSOSSRESPPQMIPPRLPPEVEAMIRLRAGLLPSHLPEVVAGLSCAFI mPC3 SSaVeg-Rde0v-gTP-kamLalina-s-NaLsKOOKKsPSAKISIpyeSFYEALEKLNKPSKLEGSEDSLYSDYVDVFYNTKPYKHRDDRUCIALM kex2 TStSsIsIgVetSaIPqtttAsTdPDSdPntPkICLSSPRQaMHyFLUFLigaTFLVLYFMFFMKSRRRIRRSRAETYEFDIIDTDSEYDSTLDNGTSGI hfur VLMTVFLVLOLRSUSFRGVKVYTMDRGLISYKGLPPEAWDEECPSDSEEDEGRGERTAFIKDOSAL mPC3 DILNEEN kex2 TEPEEVEDFDFDLSDEDNLASLSSSENGDAENTIDSVLTNENPFSDPIKUFPNDANAESASNKLUELCIPDVPPSSGRS  119  3 Results  Table 10. Identity of blisterin domains with kex2-like enzymes. Protein hfurin kex2 hPC2 mPC3 Subtilisinc  Pro 21.7%b 15.1% 21.6% 23.1%  Blisterin domaina Subtilisin-like 67.8% 43.3% 53.5% 59.6% 26%  P 38.3% 23.5% 38.9% 37.2%  aBlisterin domains: Pro = residues 1-190; Subtilisin-like = 191-440; P = 441-629. bIdentities were determined using the ALIGN program to align each blisterin domain with the corresponding segment of the kex2-like enzyme. cBacillus subtilus amylosaccharitis subtilisin.  120  3 Results  Figure 26. Similarity of blisterin 50 amino acid segments with kex2-like proteins Blisterin sequences were divided into segments of 50 amino acid residues which were used individually to search a library of kex2-like sequences with the FASTA program. The scores were normalized to the score obtained when a blisterin segment was aligned with itself. A score of 1 indicates completely identical sequence. A score of 0 indicates that the blisterin segment did not align with the corresponding region of the kex2-like protein.  121  3 Results BItsterin B Protease Darriciln  1  1  2  3^1 4^15^1 6^1 7  1  ^  P C>amcaln  1 8 1 9 I 10 I 11 I12I13114I 15 hfurin  1 z,c 0.8 •  X 0.6• E  0.4 •  *(7'  0.2 • 0 1^2^3^4^5^6^7^8^9^10 11^12 13 14 15 Furin 1  t 0.8  -  0.6 It, 0.4 <-71 0.2 0 1  1111 11,1111.1,11111  2^3^4^5^6^7^8^9^10 11^12 13 14 15 mPC2  ..11,11111.1  0.6 • ,2  0.4 0.2  1^2^3^4^5^6^7^8^9^10 11^12 13^14 15 mPC3  0.8 t 0.6 • ;6 E  0.4  in 0.2 0 1  2  3  1111 III I I 1111 III I I^I.. I II_  4^5^6^7^8^9^10 11^12 13 14 15 Kex2  1 -0  0.8 0.6 0.4  Z -15 0.2 0 1^2^3^4^5^6^7^8^9^10 11^12 13 14 15 Subtilisin  122  3 Results 3.2.4.2 Protein features The structural features of the Blisterins are presented in Figure 27. 3.2.4.2.1 Active site The region of the kex2-like proteins corresponding to the blisterin subtilisin domain includes the serine protease active site. The catalytic amino acids of the serine protease active site of the kex2-like proteins, aspartic acid, histidine and serine are found at amino acid residues 202, 241 and 415 respectively. The alignment of the blisterins and kex2-like proteins presented in Figure 25 shows that the position of these residues within the subtilisin domain site is conserved. The observation that the blisterin subtilisin-domain is the most conserved segment of the protein sequence (Table 10) is consistent with the fact that this region corresponds to the active site of the kex2-like proteins. 3.2.4.2.2 Prediction of potential post-translation modification sites kex2 is localized in a late compartment of the golgi body (Julius, Schekman and Thorner, 1984; Fuller, Brake and Thorner, 1989a; Franzuoff et  al., 1991; Redding et al., 1991; Wilcox and Fuller, 1991). Because it passes through the endoplasmic reticulum, the nascent kex2 protein chain includes a signal peptide that is removed by cleavage by signal peptidase, probably at residue 19 (Fuller, Brake and Thorner, 1989a; Wilcox and Fuller, 1991). The kex2 is both N-glycosylated and 0-glycosylated (Wilcox and Fuller, 1991). In addition, a pro-region is removed from the maturing protease by autocatalysis at residues Lys(108)Arg(109) (Brenner and Fuller, 1991). Because of its similarity to kex2, blisterin sequence was examined for potential signal peptidase, glycosylation, and autocatytic sites. The first 29 amino acid residues of the blisterin common sequence meet the criteria of von Heijne (1983) for a secretion signal peptide. The 24  123  3 Results amino-terminal residues have an overall hydrophobic character, as determined by hydropathy analysis (Figure 28)(Kyte and Doolittle, 1983). A 16 amino acid residue sequence composed of mostly hydrophobic amino acid residues flanked by charged amino acids is found between Arg(7) and Glu(24). A core of ten exclusively hydrophobic amino acid residues is found from between residues Ile(12) and Phe(21). A potential signal peptidase cleavage site was identified between residues Cys(29) and Asp(30) using the technique of von Heijne (1983) (Appendix E). Potential N-linked glycosylation sites with the motif Asn-X-Ser/Thr are found at positions 195 and 695 of the blisterin common sequence. Additional potential Nglycosylation sites are found at amino acid residue 695 of blisterin B, and residue 497 of blisterin C. Potential sites for autocatalytic cleavage of the Blisterins occur after Lys(48)Arg(49), Arg(55)Arg(56), Arg(87)Arg(88), Lys(90)Arg(91), Lys(111)Lys(112), Lys(112)Arg(113), Arg(131)Arg(132), Arg(142)Arg(143). 3.2.4.2.3 A potential transmembrane domain in blisterin C Kex2 and furin both have a potential hydrophobic transmembrane domain near the carboxyl-terminal. In the case of kex2, this domain is required to maintain the enzyme within the golgi body; deletion of the transmembrane domain results in secretion of the majority (70%) of the enzyme (Fuller, Brake and Thorner, 1989b). The transmembrane domain divides kex2 and hfurin into a luminal portion, which includes the proteolytic domain, and a negatively charged cytosolic tail of 115 amino acid residues in kex2, and 49 amino acid residues in furin. Blisterin C includes a hydrophobic domain of 24 amino acid residues, between residues 500 and 524 (Figure 28). This hydrophobic domain could potentially span a membrane. The hydrophobic domain is followed by a 48 amino acid  124  3 Results carboxyl-terminal region which includes 11 acidic (glutamic or aspartic acid) residues. In particular, nine of the last 17 residues are acidic. By comparison to hfurin and kex2, the hydrophobic domain of blisterin C is likely to be a transmembrane domain. In contrast to blisterin C, blisterin A unique and blisterin B do not include a transmembrane-like domain. Of the 13 amino acid residues of the blisterin A unique carboxyl-terminal, nine residues are non-polar residues (lieu, Leu, Val and Ala), two, Thr and Asn, are uncharged polar residues, and one, His, is basic. Thus, the dominant feature of the blisterin A unique sequence is its non-polar nature. The 73 amino acid residues of the blisterin B unique carboxyl-terminal also include a short hydrophobic region of 13 amino acid residues. However, 13 amino acids is not enough to form a membrane spanning domain. Therefore, neither blisterin B nor blisterin A are likely to be membrane spanning proteins; both hydrophobic sequences are too short, not strongly hydrophobic and located right at the carboxylterminus. 3.2.4.2.4 A cysteine-rich region in hfurin and blisterin C hfurin contains a cysteine-rich region to the amino side of its transmembrane domain. Optimal alignment of this segment of the proteins using ALIGN revealed that eight Cys residues in an interval of 90 amino acids in the blisterin C sequence align with Cys residues in the hfurin sequence (Figure 29). In blisterin C, this interval, Asp(391) to Ser(457), is located 43 residues to the amino side of the potential blisterin C transmembrane domain. In hfurin, the interval containing the conserved Cys residues, His(606) to Pro(663), is located 66 residues to the amino side of the furin transmembrane domain. The conserved Cys residues would, therefore, be located near the membrane in the lumen of the golgi body. 125  3 Results 3.2.4.2.5 A conserved Cell attachment site A search of the blisterins for protein motifs using the PCGENE PROSITE program revealed the presence of a cell attachment site, Arg(551)Gly(552)Asp(553). This sequence is characteristic of proteins such as fibronectin that interact with receptors on the cell surface (reviewed by Rouslahti and Pierschbacher, 1986). This site occurs in the P-domain of all of the kex2-like proteins except kex2, which has the sequence Arg-Gly-Thr at this position. The cell attachment site occurs 110 residues to the carboxylside of the blisterin subtilisin domain, and is found in similar positions in the kex2-like proteins (Figure 25).  126  3 Results  Figure 27. Blisterin and kex2-like protein structural features Pairs of basic residues, potential autocatalytic sites are shown as vertical bars on the amino side of the active site. Secretion signal peptides, SP, are shown as hatched boxes; transmembrane domains, TMD, are shown as cross hatched boxes; active sites are shown as dense dots. (?) indicates potential glycosylation sites. The positions of the catalytically important amino acids Asp, His, and Ser are indicated by D, H and S.  127  Protease Domain  SP  Blisterin B  r  D H  10^1  731  427  Blisterin A  571  Blisterin C Furin  Kex2  M^11 t^t  M^1 1  I  67:8 ..  I  43.3% ??  mPC2  53 5%  mPC3  59.6%  ^  794  814  638  753  3 Results  Blisterin A 140 120 100 7.= 80 60 0 40 20 0  r--^No") CO CZY) CV Lc) co^r-tr) CS) •1- CO CV CO^Cr) CS)^03 N N.^Lt-) •-•- CV CV rn^-cl- •cr U) Le) CO CO  Amino acid position Blisterin B 140 >. 120 -  c=5  •-G la  100 80  ° 60 ..= 2 40 ›... = 20  VII  0 ^  <- N. 0 c•") CO ca) CV 1.(7 00 ^<1- N. 0 rn CO 01 c:3-)^00 cNi CD^1-r)^oo CV N.^CC) 0) CV CV N) M rn <1- 9^(C) CO CO CO  CC)  Amino acid position Blisterin C 140 >. 120 100 80  ° 60 2 40 = 20 0  O r.') c0 C7) CV LC) 00^<1- N. 0 rn CO Cr) CV Cr) 03 LC, 0-3 <7' 00 CV CO^CS) •I' CO CV N.^LC)^•/- 03 CV CV CV N) c''') r4") <1-^CC) L.r, CO CO CO n n 00 00  Amino acid position Figure 28. Hydropathy analysis of the blisterins. Hydropathy values for each amino acid position were calculated using GREASE with a window of 19. Blisterin A and blisterin C were aligned with coresponding positions in blisterin B for comparison.  129  cr)  3 Results  3 4^5 6^7 8 1^2^ DHGKCVESCPPGLVADYESNLVQAKCIWIRKDLCGDGYYINAVGKCDLCDSSCETCTAPGPMSCEKCS HQKSCVQHCPPG - - - -FAPQVLDTHYSTENDVET --- -IRA -SVCAPCHASCATCQGPALTDCLSCP *.*^*. *..**.** .*,^.* .*, .**. ****^  Figure 25. Alignment of blisterin C and Furin Cysteine-rich regions. Alignment of blisterin C (top sequence), Asp(391) to Ser(457), with hfurin (bottom sequence), His606 to Pro(663). 26.9% amino acids are identical in a 67 amino acid residue overlap. Identical amino acids are indicated by "*"; similar amino acids are indicated by ".". Cysteine residues that are present in both sequences are numbered 1 to 8.  130  3 Results  3.2.5 Position of bli 4 mutations in predicted protein sequences -  3.2.5.1 h1010 Restriction mapping and Southern analysis indicated that the h1010 Tc1 insertion is in a 1.3 Kb EcoRI fragment of pCeh202 (Section 3.2.1.2.1). Analysis of the sequence of pCeh202 revealed that the boundaries of this fragment are EcoRI sites at nucleotide 1,199, in Exon V, and at nucleotide 2,608 in Intron X (Figure 18). To confirm my interpretation of the h1010 restriction data, I used PCR to amplify across the h1010 Tc1 insertion site (Figure 29). The primers used were p618 and KRp13. p618 is specific to a Tc1 sequence starting 72 by from one end of the transposable element. KRp13 is specific to a sequence 80 by to the 5' side of the 1.3 kb EcoR1 fragment containing the insertion. A band of approximately 770 by was obtained following PCR (Figure 30, lane 2), indicating that the Tc1 element was inserted approximately 620 by into the 1.3 kb EcoRI fragment. The size of the amplification band indicated that the h1010::Tcl insertion occurred within the active site of the kex2-like protease.  3.2.5.2 e937 Restriction mapping and Southern analysis indicated that the e937 mutation is a 3.5 Kb deletion (Section 3.2.1.2.2). Based on the genomic sequence data, the left break point occurs in Intron XII, to the 3' side of Exon XII. This interpretation of the left breakpoint was confirmed by sequencing through the deletion breakpoint. This was accomplished by sequencing both ends of the e937 fusion fragment cloned in pCeh206. As shown in Figure 31, the left breakpoint of the e937 deletion occurs in a 0.4 kb EcoRI/Pstl restriction fragment of pCeh181, 170 by into the fragment. The right deletion breakpoint occurs in the 1.0 kb Sall fragment of pCeh181,  131  3 Results 674 by into the fragment. Thus, the left breakpoint of the e937 deletion does not affect the blisterin common sequence. Moreover, e937 does not delete genomic DNA encoding blisterin B or blisterin C. e937 does, however, delete Exon XIII, which encodes the unique 13 amino acid residue carboxyl-terminal of blisterin A. Thus, e937 may affect only one of the three  bli-4 gene products. This may explain why the deletion does not result in a lethal phenotype.  132  3 Results Figure 30. Sequence of e937 deletion breakpoints. Partial sequence of e937 fusion fragment in plasmid pCeh206 (Top) compared to genomic sequence (Bottom). Restriction enzyme recognition sites are underlined. The exon encoding the 3' end of blisterin A is in uppercase letters. Identical positions are indicated by vertical bars (1). &ORLI e937^qaattcaaactgatttacccttcccctcttctcactttctatctaatttggtttcggtgatagttgtgttttaattttgaaaattcccaacccacagagt 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 genomic gaattcaaactgatttacccttcccctcttctcactttctatctaatttggtttcggtgatagttgtgttttaattttgaaaattcccaacccacagagt e937^gtgtgtccttgcgagtgtgtgagcctggtgtgataatcaattttcaaattgaaaaattgaaaaaatcaacaaaat IIIIIIIIIII1111111111111111111111111111111111111111111111111111111111111111 genomic gtgtgtccttgcgagtgtgtgagcctggtgtgataatcaattttcaaattgaaaaattgaaaaaatcaacaaaatgcctaattttctcataattattcat genomic cgtccttacctctggcacaattttcagtntattcatgtttctattcataaattncttgtgtatgtatctgtatgtctctatgtaaatgtgtcgctgactt genomic tttaatgtttcttgtgaactcgtcttcgtctatgttcttgttgttattgtgatgtgaatcaactttttaattaaaaagagaattgattaacagttatatt  Pst1 genomic tatttctttatttctttaatttctttgttcaactagacgtccacattgttttctqcag... 2kb gap ...ccnaaaaaaaaattggnngagcgcccc genomic ccgnaggggnaaaaaancnccgngggnaaaaaanaaaantaaaaaaagngaaaaaatactggangcccgtaagngtaaantaaaaaatngaaaaaaaatc  r Unique 3' end of blisterin A .  genomic ctttgggnaaaaatttggnccccctttttaagATACTAATCACCATCGCTATCCACCTCGTCGTCAACGCATGAATTATTUCTGTTTCACTGCTCACAG genomic CTICATAGAGCTTAAAATTAGATTTATTGTCCCTTTATCCTIGGITACGATGTUTCTGAAACTTGICTCCCCATTITCTGTGTCATTTTCCUTTTGAC genomic CTCACTATGATGAATACTTTTTACGAGaacattttgttggttgtccaaaaattaaacaaaaaaaacaaatctgcatgctcccagcgtttcatctattttc  Sall genomic ctattcccccactacagattggaatgccgagtgttctcgttgtcattcaaatcatcgtcatatgtctcgtacttggcgttggcctttatgcatqtcqacg genomic acaacaagttcaagattctacagttggagctccgactccttcagctccagaagacgttgcaatgacttcactatcacctagggaatctggaaggatgagc genomic agcttcggaaggcaattgcagcgtcaattgagggaggaagctgaaccgaaaaatggattgaaagagcattgtcagacgctgaaaagctctagtttttttt genomic tcttctcacagttatattttagtttatttctagccgtttctcatttccagtttgaggtcattcatttacacggttttctcttgtttcgggattcaatccg genomic tttcttgctcaaatttgatgttttataatgtttaacttcatatattatattctccttgtcacgggtatttttatctgtgctaagaatgtttcttcatcct genomic ccgagtatgtaattccaattgttgatattcctaattcctgcataatctgaatgagtttgagcctggttttcaattgattaatttaattttaattgcatga e937 taatttccctgaatggaatttcaaccaaatttcagttttactttcccatccacttctccgctcgattaataacttgaattaacttgatttttcccctt genomic tttaatttccctgaatggaatttcaaccaaatttcagttttactttcccatccacttctccgctcgattaataacttgaattaacttgatttttcccctt  4 Discussion  CHAPTER FOUR: DISCUSSION In this thesis, I have presented evidence supporting the following conclusions. First, that the bli 4 gene of Caenorhabditis elegans is a complex -  locus with at least one essential role in development, as well as a nonessential role in the adult cuticle. Second, the bli 4 coding region is -  contained within the cosmid KO4F10 Third, the bli 4 transcripts are -  alternatively spliced to produce more than one protein product. Fourth, that the gene products produced by bli 4 are related to the kex2-like mammalian -  prohormone convertases. In this discussion, I will examine the basis of these conclusions and attempt to demonstrate their validity. 4.1 EVIDENCE THAT bli 4 IS A COMPLEX LOCUS -  I have divided the alleles of bli 4 into three classes based on -  phenotype and complementation pattern. Class I was e937, which was homozygous viable, and caused adult cuticle blisters. Class II included nine lethal mutations that caused developmental arrest at the end of embryogenesis and failed to complement the blistered phenotype of e937. Class III included two lethal mutations, that caused developmental arrest at an early larval stage and complemented the blistered phenotype of e937. The blistered phenotype of e937 was not complemented by class II alleles, but was complemented by class III alleles. Class II and class III alleles did not complement each other. The complementation data can be interpreted in two ways; first, that there is one gene with complementing alleles, and second, that there are two genes, and the class III alleles affect both. I favor the one gene hypothesis. My evidence is discussed in the following paragraphs. I will first consider the two-gene hypothesis in light of the genetic and molecular data. The two gene hypothesis is that two genes have been  134  4 Discussion identified, bli 4, defined by e937, and an essential gene, which I will call let-  77, defined by the class III alleles. In this hypothesis, the class II alleles fail to complement both bli 4 and the let 77 mutations because they affect both -  -  genes. Mechanisms for this affect could include either or both of the following. First, each of the class II alleles could be a two mutation event, with mutations in both bli 4 and let 77. Second, each of the class II alleles -  -  could be a deletion, affecting both bli 4 and let 77. -  -  The two-hit mechanism requires that all nine of the class II alleles simultaneously acquire mutations in bli 4 and let 77. Each such double hit -  -  event is very unlikely. A further problem with the two hit hypothesis is that it does not account for the fact that the class II have a different phenotype than the class III allele: class II alleles arrest development earlier than class III alleles. Thus, in addition to requiring a highly improbable series of events, all of the double hit mutations involving let 77 would be more severe than -  s90 and h754. While the double hit hypothesis cannot be formally ruled out on the basis of genetic evidence alone, it is not a reasonable explanation for the complementation data. The deletion mechanism for the two-gene hypothesis requires that all nine of the class II alleles be deletions of bli 4 and let 77. Deletions are -  -  rarely recovered following EMS mutagenesis. McDowall (1990) demonstrated that the class II alleles h384, h427, h520, h699 and h799 complement alleles of all other genes in the interval around bli 4 defined by -  the duplications hDp16 and hDp19 (Figure 1). These mutations, therefore, cannot be large deletions. If the class II alleles were all deletions, we would expect the 10 kb Xhol fragment of K04F10 that detects h1010::Tc1 and e937 to detect restriction fragment length differences in the genomic DNA of at least some of the other class II lethal alleles. No such differences were  135  4 Discussion detected in the alleles h42, h199, h384, and h427. It is possible that these alleles are deletions with breakpoints outside of the 10 kb fragment. However, these alleles complement alleles of all other genes around bli 4, -  and therefore cannot be large deletions. Finally, at least one class II lethal,  h1010, cannot be a deletion: none of the cosmids spanning 200 kb around bli 4 that were used to probe h1010 DNA detected deletions. I conclude, -  therefore, that the deletion mechanism for the two gene hypothesis is incorrect. In the one-gene hypothesis, e937, class II and class III mutations affect a single genetic locus with complementing alleles. In this hypothesis, which I will call the one-gene hypothesis, the bli 4 locus has more than one -  function. e937 affects one function, required in the adult cuticle, while s90 and h754 affect another function. The class II alleles affect all functions of the gene. Several lines of evidence lead me to conclude that the one gene hypothesis is correct. First, the induction frequency of class II alleles is very high. Nine alleles were identified in the sDp2 lethal set, which was produced by screening 30,000 mutagenized chromosomes. The induction frequency of bli 4 alleles in the sDp2 set was, therefore, approximately 1 in -  3000. A hypothesis accounting for the nature of the class II alleles must take this frequency into account. The event resulting in a class II allele must be common, not rare. Mutations involving double hits, EMS induced deletions or mutations in a regulatory region shared by two genes are all likely to be very rare. In contrast, point mutations in a large gene that result in the loss of function of that gene are likely to be common. Second, I have shown the that insertion of Tc1 into the active site of the kex2-like gene in KO4F10 resulted in a concurrent loss of the ability of the h1010 allele to complement the function disrupted by e937. The latter observation demonstrates that  136  4 Discussion the kex2-like protease active site is required to prevent blisters. Finally, the genomic DNA encoding at least one transcript that includes the kex2-like active site, blisterin A, is affected by e937. I conclude, that e937 and all of the bli 4 lethal alleles are alleles of one gene. -  4.2 EVIDENCE THAT bli 4 IS THE kex2-LIKE CODING ELEMENT ON KO4F10 -  As a first step in identifying the bli 4 coding region, I used a probe -  detecting a strain-specific restriction fragment length difference and a deletion to define the limits of bli 4 within the C. elegans physical map. In -  doing so, I restricted the possible range of the bli 4 locus to about 200 kb of -  genomic DNA. My next step was to use the cosmids in this interval to probe the mutation h1010, which was isolated in a screen specifically designed to generate alleles that would be detectable by southern analysis. Only one of the cosmids in the interval containing bli 4 detected rearrangements in -  h1010 DNA. The pattern of restriction fragments detected by KO4F10 was consistent with the insertion of a Tc1 element. Subsequent PCR analysis demonstrated that the band shifts detected by KO4F10 were indeed a result of the insertion of Tc1. The genomic fragment that the Tc1 element was inserted into detected bands on Northern blots, demonstrating that the fragment containing the Tc1 insertion site is transcribed into RNA. The genomic restriction fragments flanking the fragment containing the Tc1 insertion site detected the same bands. Based on these data, I concluded that the Id element detected by KO4F10 was inserted into a coding region. PCR amplification using a Tc1-specific primer, p618, and a primer that directed DNA synthesis into the 1.3 kb EcoRl fragment containing the Tc1 insertion in h1010 produced a band of approximately 770 bp. based on the  137  4 Discussion position of the primers, I estimated that the Tc1 insertion site was about 580 nucleotides into the 1.3 kb EcoRI fragment. Based on the sized of the amplification band, examination of preliminary sequence data revealed that the Tc1 insertion site was near the center of the genomic DNA encoding the predicted active site of the kex2-like protease sequence identified in KO4F10. Two questions are of critical importance to my contention that the kex2-like coding region in K04F10 is bli 4. First, did the lethal phenotype of -  h1010 result from the TO element insertion detected by KO4F10? Second, if the lethal phenotype of h1010 did result from the Tc1 element insertion detected by KO4F10, was it the disruption of the kex2-like coding element by the Tc1 insertion that was responsible for the h1010 phenotype, or was it some other effect of the TO insertion? For example, the Tc1 insertion could be inserted into an enhancer required for the expression of a coding element other than the kex2-like gene. KR1822, the parent strain of KR1858 did not contain a lethal mutation linked to flanking markers unc 63 and unc 13. Nor did the parental strain -  -  contain a Tc1 insertion at this position. Therefore, h1010 and the Tc1 insertion arose at the same time. The answer to the first question, does the  h1010 phenotype result from the Tc1 element insertion, must be consistent with this observation. The hypothesis that the h1010 phenotype does not result from the Tc1 insertion, therefore, requires two separate but simultaneous mutation events. Since the Tc1 insertion was the only genetic alteration detected in southern analysis using the cosmids of the bli 4 -  interval as probes, the mutation causing the h1010 phenotype must be either a very small rearrangement or point mutation. Since KO4F10 rescues  bli 4 lethal mutations, these events must have both occurred in the genomic -  DNA cloned in KO4F10.  138  4 Discussion The hypothesis that the Tc1 insertion is responsible for the h1010 phenotype requires only one mutation event. The position of the Tc1 insertion is consistent with the physical mapping and rescue data, and with the observation that the e937 deletion mutation affects the same kex2-like coding region as the Tc1 insertion affects. Definitive proof that the h1010 phenotype resulted from the Tc1 insertion could have been provided by the recovery of a spontaneous revertant of the h1010 mutation and demonstration that the Tc1 element had excised in the revertant strain. In the absence of such proof, I conclude that the hypothesis that the hi010 phenotype is results from the Tc1 insertion is the most probable. Given that the lethal phenotype of h1010 did result from the Tc1 element insertion detected by KO4F10, was it the disruption of the kex2-like coding element by the Tc1 insertion that was responsible for the h1010 phenotype, or was it some other effect of the Tc1 insertion? Two points support the hypothesis that it is the disruption of the kex2-like gene that is responsible for the h1010 phenotype. First, h1010 is inserted into the active site of the kex2-like gene. The Tc1 element is therefore very likely to affect the function of the kex2-like gene. Second, the h1010 mutation cannot complement the blistered phenotype of e937, which affects the same kex2like gene. These observations, combined with the fact that KO4F10 rescues  bli - 4 lethal mutations, lead us to conclude that the Tc1 insertion is responsible for the mutant phenotype of h1010. The h1010 data, however, does not prove that the kex2-like protease disrupted by the TO insertion is bli - 4. It is possible that the h1010 phenotype is a result of the Tc1 insertion in the kex2-like protease active site interfering in some way with the expression of the 'real' bli - 4, a gene other than the kex2-like gene. Given that KO4F10 rescues bli - 4 lethal mutations,  139  4 Discussion  bli 4 would also be on KO4F10 in this hypothesis. For example, the Tc1 -  could be inserted into an intron containing an enhancer required for the expression of the bli 4. Therefore, the other gene hypothesis cannot be -  eliminated on the basis of the h1010 data alone. The same 10 kb Xhol KO4F10 restriction fragment that detects  h1010::Tcl also detects a chromosomal rearrangement associated with the bli 4 allele, e937. The e937 rearrangement was shown to be a 3.5 kb -  deletion by restriction mapping and sequencing of the deletion breakpoints. e937 was induced using 32 P as a mutagen (Brenner, 1974). 32 P causes chromosomal breaks, consistent with the finding that e937 is a deletion.  e937 deletes genomic DNA encoding the last exon of blisterin A. Blisterin A includes at least part of the kex2-like gene active site. Therefore, like  h1010::Tcl, e937 disrupts the kex2-like protease. The fact that both h1010::Tcl and e937 affect the kex2-like protease leads us to conclude that the kex2-like protease is the bli 4 gene. -  In summary, KO4F10 rescues bli 4 mutants, and detects -  rearrangements in two bli 4 alleles. Both rearrangements affect the same -  coding region. Together, these observations provide compelling evidence for the conclusion that the affected coding region is bli 4. -  4.3 bli 4 MUTATIONS -  That e937 is a deletion was initially surprising given that e937 is the least severe of the bli 4 mutations. This can be explained by the observation -  that is an internal deletion affecting genomic DNA encoding only one of the three messages identified, blisterin A; e937 does not delete genomic DNA encoding blisterin B or blisterin C. The fact that the e937 deletion is homozygous viable indicates that the function of blisterin A is not essential, consistent with the hypothesis that disruption of blisterin A is reponsible for  140  4 Discussion the blistered phenotype. It is possible, however, that e937 could also delete other exons that have not yet been identified. Thus, blisterin A should be considered as a candidate only for the protein required in the adult cuticle. Given that e937 affects DNA encoding only one of the three transcripts identified, a mechanism for intragenic complementation of bli 4 -  alleles might be that the two complementing alleles, s90 and h754, affect only one of the other two transcript classes. In this hypothesis, the alleles complement because the e937 allele produces some bli 4 transcripts, -  including, blisterin B and C, but not others, such as blisterin A. s90 and  h754 would affect different transcripts from those affected by e937. Class II alleles would eliminate all bli 4 functions. This idea is supported by the -  observation that the complementing alleles arrest development slightly later than the class II alleles, which is evidence that the complementing alleles retain some function. This hypothesis could be tested by sequencing the blisterins from s90 and h754 homozygotes. Of the three bli 4 mutation classes, class II mutations occur most -  frequently, and have the most severe phenotype. It is therefore likely that the class II phenotype is the null phenotype. Class II mutants do not arrest development until the end of embryogenesis. This could indicate that bli 4 -  is not required until the end of embryogenesis. More likely, the blisterins are required earlier in embryogenesis, and a maternal endowment of the bli-  4 transcripts provides the null mutants with sufficient blisterin function to survive to this stage. Examples of this are null mutations of the ama 1 and -  ama 2 genes in C. elegans, which encode the large and small subunits of -  RNA polymerase II (Rogalski and Riddle, 1988; Rogalski, Bullerjahn and Riddle, 1988). In these mutants, worms survive to the end of embryogenesis with no zygotic RNA polymerase II-dependent transcription at all.  141  4 Discussion Of the three cDNA clones sequenced, only blisterin B contained a complete open reading frame. This conclusion is based on the presence of an ATG beginning at position 176 of the cDNA, and upstream in frame stop codons. It is possible that the blisterin B cDNA clone does not contain the entire 5' untranslated region. Particularly since the length of the blisterin B message detected in northern analysis is 3.5 kb, much longer that the 2.4 kb cDNA insert. The Blisterin A and Blisterin C cDNA clones are likely to contain incomplete open reading frames. This conclusion is based on the observations that both are shorter than the bands they detect in northern analysis, and that they start within the open reading frame of blisterin B. An attractive hypothesis is that blisterin A and Blisterin C are derived from mRNA transcripts with the same 5' end as blisterin B. The fact that three of the pCeh181 EcoRI genomic fragments encoding all of the 5' end sequence of blisterin A detected all of the messages identified supports this conclusion. However, it is possible that the 5' ends of these molecules differs from blisterin B through alternative splicing. The fact that a blisterin C unique probe detected a smaller band than did a blisterin B unique probe supports the latter possibility. Thus, no conclusions on the nature of the 5' ends of blisterin A and C can be drawn based on this data. 4.4 bli-4 GENE PRODUCTS ARE RELATED TO kex2-LIKE PROTEASES I have shown that the blisterin sequences have a high degree of identity and similarity with the kex2-like serine endoproteases. Sequence similarity can originate in three ways. Coincidence, convergence, and evolutionary conservation. The identity of the predicted blisterin sequences with the kex2-like proteins was as great as 60% over more than 300 amino acid residues. Evolutionary conservation is the only likely explanation for this degree of identity. On this basis, I conclude that bli-4 is a homologue  142  4 Discussion of the kex2-like genes: bli-4 and the kex2-like genes evolved from a common ancestral sequence. All of the members of the kex2-like proteins that have been tested have been shown to be serine endoproteases that cleave substrates after pairs. Blisterin B, the only bli-4 cDNA that appeared to contain a complete open reading frame, was found to be as similar to members of the kex2-like serine endoproteases as the kex2-like proteases were to each other. It is therefore likely that the sequence conservation of the blisterins and the kex2-like proteases is reflected in functional conservation: the blisterins are likely to be serine endoproteases. The similarity of the blisterin B primary amino acid sequence to other members of the kex2-like enzyme family is greatest in the 300 amino acid sequence including the protease domain. Outside of this region, other structural features are conserved, including a signal peptide, potential autocatalytic sites, and, in blisterin C, a Cys-rich region and potential transmembrane domain followed by an acidic carboxy tail. By analogy to the kex2-like proteases, therefore, the blisterins are likely to be proprotein processing enzymes that cleave substrates in the golgi body during secretion. Structurally, the kex2 protein family can be divided into two classes based on the presence or absence of a transmembrane domain (TMD) near the carboxy terminus. PC1(PC3), PC2, blisterin A and blisterin B lack a TMD. Kex2, hfurin, and blisterin C contain a TMD. bli-4 is the first example of both types of enzyme arising through alternative splicing. Although alternative splicing in other KEX2-like genes has not been reported, alternative splicing cannot be ruled out. PC1(PC3) and PC2 detect multiple bands on Northern blots (Seidah et al., 1990; Smeekens and Steiner, 1990).  143  4 Discussion It is possible that alternatively spliced transcripts from these genes exist but have not yet been identified. The TMD is required for protein retention within the golgi body: deletion of the TMD in kex2 results in the secretion of the protein (Fuller, Brake and Thorner, 1989b). It has been suggested that PC1(PC3) and PC2 might be membrane associated through an amphipathic helix structure (Smeekins et al., 1991). No direct evidence to support this hypothesis has been produced. It is possible that the different proteins encoded by the bli-  4 locus fulfill functions in the nematode that are fulfilled in mammals by different genes. Of the other members of the KEX2-like gene family, only KEX2 was identified genetically: null mutations in KEX2 result in the inability to process the alpha-factor mating pheromone and the K1 killer toxin (Wickner and Leibowitz, 1976; Leibowitz and Wickner, 1976). KEX2 is not essential to viability in yeast, indicating that kex2 does not play an essential intracellular role (that is, kex2 is not an essential metabolic gene). In contrast, null mutations in bli-4 result in developmental arrest near the end of embryogenesis prior to hatching. Such evidence demonstrates that bli-4 is essential to development. Why is bli-4 essential where KEX2 is not? Unlike yeast, C. elegans is multicellular, and has a complex development process. The observation that the blisterins are essential to development implies the existence of blisterin substrates required for embryogenesis. Such substrates could include both structural molecules and peptide signals. 4.5 CONCLUDING COMMENTS No essential cellular interaction involving a secreted protein derived from a proprotein has yet been described in C. elegans. The disruption of development by bli-4 mutations implies the existence of such interactions.  144  4 Discussion This work provides the first link between the molecular analysis of kex2-like enzymes and genetic analysis of mutations in a multicellular eukaryote.  145  References Ambros, V. and H.R. Horvitz. 1984. Heterochronic mutants of the nematode Caenorhabditis elegans. Science 226, 409-416.  Ambros, V. and H.R. Horvitz. 1987. The Lin-14 locus of Caenorhabditis elegans controls the time of expression of specific post-larval events. Genes Dev. 3, 399-414.  Baillie, D.L., K.A. Beckenbach, and A.M. Rose. 1985. Cloning within the unc43 to unc-31 interval (linkage group IV) of the Caenorhabditis elegans  genome using Tc1 linkage selection. Can. 1. Genet. Cytol. 27, 457-466. Barstead, R.J. and R.H. Waterston. 1989. The basal component of the nematode dense-body is vinculin. I. Biol. Chem. 264, 10177-10185. Benjannet, S., N. Rondeau, R. Day, M. Chretien, and N.G. Seidah. 1991. PC1 and PC2 are proprotein convertases capable of cleaving proopiomelanocortin at distinct pairs of basic residues. Proc. Natl. Acad. Sci. 88, 3564-3568.  Brenner, C. and R.S. Fuller. 1992. Structural and enzymatic characterization of a purified prohormone-processing enzyme: secreted, soluble Kex2 protease. Proc. Natl. Acad. Sci. 89, 922-926. Brenner, S. 1974. The genetics of Caenorhabditis elegans. Genetics 77, 71-94. Bresnahan, P.A., R. Leduc, L. Thomas, J. Thorner, H.L. Gibson, A.J. Brake, P.J. Barr, and G. Thomas. 1990. Human fur gene encodes a yeast KEX2-like endoprotease that cleaves pro-beta-NGF in vivo. J. Cell Biol. 111, 2851-9. Bullock, W.O., J.M. Fernadez, and J.M. Short. 1987. XL1-Blue: a high efficiency plasmid trasforming recA Escherichia coli strain with betagalactosidase selection. BioTechniques 5, 376.  146  Cassada, R.C., and R.L. Russell. 1975. The dauer larva, a post-embryonic developmental variant of the nematode Caenorhabditis elegans. Dev. Biol. 46, 326-342. Chance, R.E., R.M. Ellis, and W.W. Brommer. 1968. Porcine proinsulin: characterization and amino acid sequence. Science 161, 165-167. Chretien, M. and C.H. Li. 1967. Isolation, purification, and characterization of gamma-lipotropic hormone from sheep pituitary glands. Can. I. Biochem. 45, 1164-1174. Clark, D.V., Rogalski, T.M., L.M. Donati, and D.L. Baillie. 1988. The unc22(1V) region of Caenorhabditis elegans: genetic analysis of lethal mutations. Genetics 119, 345-353. Coulson, A., J. Sulston, S. Brenner, and J. Karn. 1986. Towards a physical map of the genome of the nematode Caenorhabditis elegans. Proc. Natl. Acad. Sci. U.S.A. 83, 7821 7825. -  Coulson, A., R. Waterston, J. Kiff, J. Sulston, and Y. Kohara. 1988. Genome linking with yeast artificial chromosomes. Nature (London) 355, 184186. Cox, G.N., J.S. Laufer, M. Kusch and R.S. Edgar. 1980. Genetic and phenotypic characterization of roller mutants of Caenorhabditis elegans. Genetics 95, 17 339. -  Cox, G.N., M. Kusch and R. S. Edgar. 1981. The cuticle of Caenorhabditis elegans: Its isolation and partial characterization. J. Cell Biol. 90, 7-17. Cox, G.N., S. Staprans and R.S. Edgar. 1981. The cuticle of Caenorhabditis elegans II. stage specific changes in ultrastructure and protein composition during post-larval development. Dev. Biol. 86, 456-470.  147  Cox, G.N., J.M. Kramer and D. Hirsh. 1984. Number and organization of collagen genes in Caenorhabditis elegans. Mol. Cell. Biol. 4, 2389-2395. Cox, G.N. and D. Hirsh. 1985. Stage specific patterns of collagen gene expression during development of Caenorhabditis elegans. Mol. Cell. Biol. 5, 363-372.  Craxton, M. 1991. Methods: A companion to Methods in Enzymology 3, 2026. Cross, S.H., and P. Little. 1986. A Cosmid vector for systematic chromosome walking. Gene 49, 9-22. Crow, E.L. and R. S. Gardener. 1959. Confidence intervals for the expectation of a poisson variable. Biometrika 46, 441-453. Davis, R.W., D. Botsein and J.R. Roth. 1980. A manual for genetic engineering: Advanced bacterial genetics.. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. Docherty, K. and D.F Steiner. 1982. Post-translational proteolysis in polypeptide hormone biosynthesis. Ann. Rev. Physol. 44, 625-638. Edgley, M.L. and D.L. Riddle. 1990. The nematode Caenorhabditis elegans. In Genetic maps, vol 5. Eidie, D., and P. Anderson. 1985a. The gene structures of spontaneous mutations affecting a Caenorhabditis elegans myosin heavy chain gene. Genetics 109, 67-79. Eidie, D., and P. Anderson. 1985b. Transposition of Tc1 in the nematode Caenorhabditis elegans. Proc. Natl. Acad. Sci. 82, 1756 1760. -  Emmons, S.W., M.R. Klass, and D. Hirsh. 1979. Analysis of the constancy of DNA sequences during development and evolution of the nematode C. elegans. Proc. Natl. Acad. Sci. 76, 1333 1337. -  148  Emmons, S.W., L. Yesner, K.-S. Ruan, and D. Katzenberg. 1983. Evidence for a transposon in Caenorhabditis elegans. Cell 32:55-65. Feinberg, A.P. and B. Vogelstein. 1984. A technique for radio-labeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 137, 266-267.  Fodor, A. and P. Deak. 1985. The isolation and genetic analysis of a Caenorhabditis elegans translocation (szT1) strain bearing an  X-chromosome balancer. J. Genet. 64, 143-1673. Foster, D.C., R.D. Holly, C.A. Sprecher, K.M. Walker, and A.A. Kumar. 1991. Endoproteolytic processing of the human protein C precursor by the yeast Kex2 endopeptidase coexpressed in mammalian cells. Biochem. 30, 367-72.  Franzuoff, A. and R. Scheckman. 1989. Functional compartments of the yeast golgi apparatus are defined by the sec? mutation. EMBO 1 8, 26952702. Franzuoff, A., K. Redding, J. Crosby, R.S. Fuller, and R. Scheckman. 1991 Localization of components involved in protein transport and processing through the yeast Golgi apparatus. 1. Cell Biol. 112, 27-37. Fuller, R.S., A. Brake, and J. Thorner. 1986. The Saccharomyces cerevisiae KEX2 gene, required for processing prepro-alpha-factor, encodes a calcium dependent endopeptidase that cleaves after Lys-Arg sequences. In Microbiology, L. Leive, editor. American Society for Microbiology, Washington, DC. pp 273-278. Fuller, R.S., R.E Sterne, and J. Thorner. 1988. Enzymes required for yeast prohormone processing. Ann. Rev. Physiol. 50, 345-362.  149  Fuller, R.S., A. Brake, and J. Thorner. 1989a. Yeast prohormone processing enzyme (KEX2 gene product) is a Ca2 +-dependent serine protease. Proc. Natl. Acad. Sci. 86, 1434-1438. Fuller, R.S., A.J. Brake, and J. Thorner. 1989b. Intracellular targeting and structural conservation of a prohormone-processing endoprotease. Science 246, 482-486. Gibson, T.J., A.R. Coulson, J.E. Sulston, and P. Little. 1987. Lorist 2, a cosmid with transcriptional terminators insulating vector genes from interference by promoters within the insert: effect of DNA yield and cloned insert frequency. Gene 53, 275-281. Greenwald, I. 1985. lin-12, a nematode homeotic gene, is homologous to a set of mammalian proteins that includes epidermal growth factor. Cell 43, 583-590. Hanahan, D. 1983. Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 166, 557 Hatsuzawa, K., M. Hosaka, T. Nakagawa, M. Nagase, A. Shoda, K. Murakami, and K, Nakayama. 1990. Structure and expression of mouse furin, a yeast kex2-related protease. J. Biol. Chem. 265, 22075-22078. Heine, U., and T. Blumenthal. 1986. Characterization of regions of the C. elegans X Chromosome containing vitellogenin genes. J. Mol. Biol.  188, 301-312. Henikoff, S. 1984. Unidirectional digestion with exonuclease Ill in DNA sequence analysis. Gene 28, 351-259. Higgins, B.J. and D. Hirsh. 1977. Roller mutants of Caenorhabditis elegans. Mol. Gen. Genet. 150, 63-72. Hodgkin, J. 1980. Primary sex determination in the nematode C. elegans. Genetics 46, 649-664.  150  Hodgkin, J., H.R. Horvitz, and S. Brenner. 1979. Nondisjunction mutants of the nematode Caenorhabditis elegans. Genetics 91, 67 94. -  Horvitz, H.R., S. Brenner, J. Hodgkin, and R. Herman. 1979. A uniform genetic nomenclature for the nematode Caenorhabditis elegans. Mol. Gen. Genet. 175, 129-133. Howell, A.M. 1989 Essential genes in a region of chromosome / in Caenorhabditis elegans. Ph.D. Thesis, University of British Columbia,  Vancouver, B.C., Canada. Howell, A.M., S.G. Gilmore, R.A. Mancebo, and A.M. Rose. 1987. Genetic analysis of a large autosomal region in Caenorhabditis elegans by the use of a free duplication. Genet. Res. 49, 207-213. Ish-Horowicz, D. and J.F. Burke. 1981. Rapid and efficient cosmid cloning. Nucleic Acids Res. 9, 1323 1338. -  Johnsen, R.C. and D.L. Baillie. 1991. Genetic analysis of a major segment [LGV(left)] of the genome of Caenorhabditis elegans. Genetics 129, 735-752. Julius, D., R. Schekman, and J. Thorner. 1984. Glycosylation and processing of pre-pro-alpha-factor in the yeast secretory pathway. Cell 36, 309318. Julius, D., A. Brake, L. Blair, R. Kunisawa, and. J. Thorner. 1984. Isolation of the putative structural gene for the lysine-arginine-cleaving endopeptidase required for processing of yeast prepro-alpha-factor. Cell 37, 1075-1089. Kozak, M. 1986. Point mutations define a sequence flanking the AUG initiator codon that modultates translation by eukaryotic ribosomes. Cell 44, 283-292.  151  Kramer, J.M., G.N. Cox and D. Hirsh. 1982. Comparisons of the complete sequences of two collagen genes of Caenorhabditis elegans. Cell 30, 599-606. Kramer, J.M., G.N. Cox and D. Hirsh. 1985. Expression of the Caenorhabditis elegans collagen genes col-/ and col 2 is developmentally regulated. J. -  Biol. Chem. 260, 1945-1951.  Kramer, J.M., J.J. Johnson, R.S. Edgar, C. Basch and S. Roberts. 1988. The sqt 1 gene of C. elegans encodes a collagen critical for organismal -  morphogenesis. Cell 55, 555-565. Kramer, J.M., R.P. French, E.C. Park, and J. J. Johnson. 1990. The Caenorhabditis elegans rol 6 gene, which interacts with the sqt 1 -  -  collagen gene to determine organismal morphology, encodes a collagen. Mol. Cell. Biol. 10, 2081-9. Kusch, M. and R. S. Edgar. 1986. Genetic studies of unusual loci that affect body shape of the nematode Caenorhabditis elegans and may code for cuticle structural proteins. Genetics 113, 621-639. Kyte, J. and R.F. Doolittle. 1982. A simple method for displaying the hydropathic character of a protein. I. Mol. Biol. 157, 105-132. Liebowitz, M.J. and R.W. Wickner. 1976. A chromosomal gene required for killer plasmid expression, mating and sporulation in Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. 73, 2061 2065. -  Maniatis, T., E.F. Fritsch and J. Sambrook. 1982. Molecular cloning (a laboratory manual). Cold Spring Harbor Laboratory, Cold Spring Harbor, New York.  152  McDowall, J. 1990. Essential genes in the hDp16/hDp19 region of LG/ in Caenorhabditis elegans. M.Sc. Thesis, University of British Columbia,  Vancouver, B. C., Canada. McKim, K.S., A.M. Howell, and A.M. Rose. 1988. The effects of translocations on recombination frequency in Caenorhabditis elegans. Genetics 120: 987-1001. McKim, K.S. and A.M. Rose. 1990. Chromosome I duplications in Caenorhabditis elegans. Genetics 124, 115 132. -  McKim, K.S., T.V. Starr, and A.M. Rose. 1992. Genetic and molecular analysis of the dpy 14 region in Caenorhabditis elegans. Mol. Gen. Genet. 233, -  241-251. McKim, K.S., K. Peters, and A.M. Rose. 1992. Genetic analysis of homologue pairing in Caenorhabditis elegans. Genetics, in press. Misumi, Y., M. Sohda and Y. Ikehara. 1990. Sequence of the cDNA encoding rat furin, a possible propeptide-processing enzyme. Nucleic Acids Res. 18, 6719.  Mizuno, K., T. Nakamuta, K. Takada, S. Sakakibara, and J. Matsuo. 1987. A membrane-bound, calcium-dependent protease in yeast alpha-cell cleaving on the carboxyl side of paired basic residues. Bioch. Biophys. Res. Comm. 144, 807-814.  Mizuno, K., T. Nakamura, T. Ohshima, S. Tanaka, and H. Matsuo. 1988. Yeast KEX2 gene encodes an endopeptidase homologous to subtilisin-like serine proteases. Biochem. Biophys. Res. Comm. 156, 246-54. Moerman, D.G. and D.L. Baillie. 1979. Genetic organization in Caenorhabditis elegans: Fine-structure analysis of the unc 22 gene. -  Genetics 91, 95-103.  153  Moerman, D.G. and D.L. Baillie. 1981. Formaldehyde mutagenesis in the nematode Caenorhabditis elegans. Mutat. Res. 80, 273 279. -  Moerman, D.G., and R.H. Waterston. 1984. Spontaneous unstable unc 22 /V -  mutants in Caenorhabditis elegans var. Bergerac. Genetics 108, 859877. Moerman, D.G., G.M. Benian and R.H. Waterston. 1986. Molecular cloning of the muscle gene unc-22 in C. elegans by Tc1 transposon tagging. Proc. Natl. Acad. Sci. 83, 2579 2583. -  Mori, I., D.G. Moerman, and R.H. Waterston. 1988. Analysis of a mutator activity necessary for germline transposition and excision of Tc1 transposable elements in Caenorhabditis elegans. Genetics 120, 397407. Muller, H.J. 1937. Further studies on the nature and causes of gene mutations, pp 213 255 in Proceedings of the Sixth International -  Congress of Genetics, edited by D. Jones. Brooklyn Botanical  Gardens, Menasha, Wisconsin. Novick, P., and R. Scheckman. 1979. Secretion and cell-surface growth are blocked in a temperature-sensitive mutant of Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. 76, 1858 1862. -  Payne, G.S. and S. Schekman. 1989. Clathrin: a role in the intracellular retention of a Golgi membrane protein. Science 245, 1358-1365. Pearson, W.R. and D.J. Lipman. 1988. Improved tools for biological sequence comparison. Proc. Natl. Acad. Sci. 85, 2444-2448. Peters, K., J. McDowall, and A.M. Rose. 1991. Mutations in the bli 4 (I) locus -  of Caenorhabditis elegans disrupt both adult cuticle and early larval development. Genetics 129, 95-102.  154  Prasad, S.S., and D.L. Baillie. 1989. Evolutionarily conserved coding sequences in the dpy-20 to unc-22 region of Caenorhabditis elegans. Genomics 5, 185-198.  Redding, K., C. Holcomb, and R.S. Fuller. 1991. Immunolocalization of Kex2 protease identifies a putative late golgi compartment in the yeast Saccharomyces cerevisiae. J. Cell Biol. 113, 527-538.  Riddle, D. 1988. The Dauer Larva. In: The nematode Caenorhabditis elegans. W.B. Wood, Ed. Cold Spring Harbor Laboratories, Cold  Spring Harbor. Roebroek, A.J.M., J.A. Schalken, J.A.M. Leunissen, C. Onnekink, H.P.J. Bloemers, and W. Van de Ven. 1986. Evolutionary conserved close linkage of the c-fes/fps proto-oncogene and genetic sequences encoding a receptor-like protein. EMBO J. 5, 2197-2202. Rogalski, T.M. and D.L. Baillie, 1985. Genetic organization of the unc-22 region of Caenorhabditis elegans. Mol. Gen. Genet. 201, 409-414. Rogalski, T.M. and D.L. Riddle. 1988. A Caenorhabditis elegans RNA polymerase II gene, ama-1 /V, and nearby essential genes. Genetics 118, 61-74. Rogalski, T.M., A.M.E. Bullerjahn, and D.L. Riddle. 1988. Lethal and amanitin resistance mutations in the ama-1 and ama-2 genes of Caenorhabditis elegans. Genetics 120, 409-422.  Rose, A.M. and D.L. Baillie. 1979. The effects of temperature and parental age on recombination and nondisjuction in Caenorhabditis elegans. Genetics 92, 409-418.  Rose, A.M. and D.L. Baillie. 1980. Genetic organization of the region around unc-15 (I), a gene affecting paramyosin in Caenorhabditis elegans. Genetics 96, 639-648.  155  Rose, A.M., D.L. Baillie, E.P.M. Candido, K.A. Beckenbach, and D. Nelson. 1982. The linkage mapping of cloned restriction fragment length differences in Caenorhabditis elegans. Mol. Gen. Genet. 188, 286 291. -  Rose, A.M., D.L. Baillie, and J. Curran. 1984. Meiotic pairing behavior of two free duplications of linkage group I of Caenorhabditis elegans. Mol. Gen. Genet. 195, 52 56. -  Rosenzweig, B., L.W. Liaom and D. Hirsh. 1983. Sequence of the Caenorhabditis elegans transposable element Tc1. Nucleic Acids Res. 11: 4201-4209. Rouslahti, E. and M.D. Pierschbacher. 1986. Arg-Gly-Asp: a versatile cell recognition signal. Cell 44, 517-518. Saki, R.K., D.H. Gelfand, S. Stoffel, S.H. Scharf, R. Higuchi, G.T. Horn, K.B. Mullis and H.A. Erlich. 1988. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239, 487-491. Sanger, F., and A.R. Coulson. 1975. A rapid method for determining sequences in DNA by primed synthesis with DNA polymerase. J. Mol. Biol. 94, 444-448. Sanger, F., S. Nicklen and A.R. Coulson. 1977. DNA sequencing with chainterminating inhibitors. Proc. Natl. Acad. Sci. 74, 5463-5467. Seidah, N.G., L. Gaspar, P. Mion, M. Marcinkiewicz, M. Mbikay, and M. Chretien. 1990. cDNA sequence of two distinct pituitary proteins homologous to Kex2 and furin gene products: tissue-specific mRNAs encoding candidates for pro-hormone processing proteinases. DNA and Cell Biol. 9, 415-424. Singh, R.N. and J. Sulston. 1978. Some observations on moulting in Caenorhabditis elegans. Nematologica 24, 63 71. -  156  Smeekens, S.P., A.S. Avruch, J. LaMendola, S.J. Chan, and D.F Steiner. 1991. Identification of a cDNA encoding a second putative prohormone convertase related to PC2 in AtT20 cells and islets of Langerhans. Proc. Natl. Acad. Sci. 88, 340 344. -  Smeekens, S.P. and D.F. Steiner. 1990. Identification of a human insulinoma cDNA encoding a novel mammalian protein structurally related to the yeast dibasic processing protease Kex2. I. Biol. Chem. 265, 2997-3000. Southern, E. 1975. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98, 503-517. Starr, T., A.M. Howell, J. McDowall, K. Peters, and A.M. Rose. 1989. Identification of DNA probes within the linkage group I gene cluster of Caenorhabditis elegans. Genome 32, 365 372. -  Steiner, D.F., D. Cunningham, L. Spigelman, and B. Aten. 1967. Insulin biosynthesis: evidence for a precursor. Science 157, 697-700. Sulston, J. and S. Brenner. 1974. The DNA of Caenorhabditis elegans. Genetics 77, 95-104.  Sulston, J.E. and H.R. Horvitz. 1981. Post-embryonic cell lineages of the nematode, C. elegans. Dev. Biol. 56, 110-156. Sulston, J.E., E. Schiernberg, J.G. White, and J.N. Thomson. 1983. The embryonic cell lineage of the nematode C. elegans. Dev. Biol. 100, 64119. Sulston, H., Z. Du, K. Thomas, R. Wilson, L. Hiller, R. Staden, N. Hallorna, P. Freen, J. Thierry-Mieg, L. Qui, S. Dear, A. Coulson, M. Craxton, R. Durbin, M. Berks, M. Metzstein, T. Hawkins, R. Ainscough, and R. Waterston. 1992. The C. elegans genome sequencing project: a beginning. Nature, 356, 37-41.  157  Tanguy-Rougeau, C., M. Wesolowski-Louvel and H. Fukuhara. 1988. The Kluyveromyces lactis KEX1 gene encodes a subtilisin-type serine  proteinase. FEBS Letters 234, 464-470. Thomas, G., B.A. Thorne, and D.E. Hurby. 1988. Gene transfer techniques to study neuropeptide processing. Ann. Rev. Physiol. 50, 323-332. Thomas, G., B.A. Thorne, L. Thomas, R.G. Allen, D.E. Hruby, R. Fuller, and J. Thorner. 1988. Yeast KEX2 endopeptidase correctly cleaves a neuroendocrine prohormone in mammalian cells. Science 241, 226230. Thomas L., R. Leduc, B.A. Thorne, S.P. Smeekens, D.F. Steiner, and G. Thomas. 1991. Kex2-like endoproteases PC2 and PC3 accurately cleave a model prohormone in mammalian cells: evidence for a common core of neuroendocrine processing enzymes. Proc. Natl. Acad. Sci. 88, 5297-5301. van den Ouweland, A.M., H.L. van Duijnhoven, G.D. Keizer, L.C. Dorssers, and W.J.M. Van de Ven. 1990. Structural homology between the human fur gene product and the subtilisin-like protease encoded by yeast KEX2. Nucleic Acids Res. 18, 664. van de Ven, W.J.M., J. Voorberg, R. Fontijin, H. Pannekoek, A.M.W. van den Oweland, H.L.P. van Duijnhoven, A.J.M. Roebroek, and R.J. Siezen. 1990. Furin is a subtilisin-like proprotein processing enzyme in higher eukaryotes. Mo/. Biol. Rep. 14, 265-275. von Heijne, G. 1983. Patterns of amino acids near signal sequence cleavage sites. Eur. J. Biochem. 133, 17-21. von Mende, N., D. Bird, P.S. Albert and D.L. Riddle, 1988 dpy-13: a nematode collagen that affects body shape. Cell 55, 567-576.  158  Wickner, R.W. and M.). Liebowitz. 1976. Two chromosomal genes required for killing expression in killer strains of Saccharomyces cerevisiae. Genetics 82, 429-442.  Williams, B.D., B. Schrank, C. Huynh, R. Shownkeen, and R.H. Waterston. 1992. A genetic mapping system in Caenorhabditis elegans based on polymorphic sequence-tagged sites. Genetics 131, 609-624. Wise, R.J., P.J. Barr, P.A. Wong, M.C. Kiefer, A.J. Brake, and R.J. Kaufman. 1990. Expression of a human proprotein processing enzyme: correct cleavage of the von Willebrand factor precursor at a paired basic amino acid site. Proc. Natl. Acad. Sci. 87, 9378-82. Wilcox, C.A. and R.S. Fuller. 1991. Posttranslational processing of the prohormone-cleaving Kex2 protease in the Saccharomyces cerevisiael secretory pathway. J. Cell Biol. 115, 297-307. Wood, W.B., R. Hecht, S. Carr, R. Vanderslice, N. Wolf and D. Hirsh. 1980. Paternal effects and phenotypic characterization of mutations that affect early development in Caenorhabditis elegans. Dev. Biol. 74, 446-469. Wood, W.B.(Ed). 1988. The Nematode Caenorhabditis elegans. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. Yoshimoto, T., H. Oyama, T. Honda, H. Tone, T. Takeshita, T. Kamiyama, and D. Tsuru. 1988. Cloning and expression of subtilisin Amylosacchariticus gene. I. Biochem. 103, 1060-1065.  159  APPENDIX A APPENDIX A. ISOLATION AND CHARACTERIZATION OF hT2  Screen for a bli-4 balancer: Isolation of hT2. Chromosomal translocations in C. elegans suppress recombination in the translocated DNA and cause pseudolinkage between markers on the affected chromosomes (Rose and Baillie, 1981). To develop a tool for the isolation of new bli 4 mutations, I -  designed a screen to identify translocations of the left half of LGI and the right half of LGIII that would include the bli 4 allele e937 on the translocated -  portion of LGI. Such a translocation would permit rapid precomplemention screens for new bli 4 alleles. A screen to identify potential translocations of -  the left arm of LGI and the right arm of LGIII was performed as follows. Homozygous e937 males were treated with 1500R of gamma radiation using an Atomic Energy Commission of Canada Gamma Cell and crossed to hermaphrodites (KR169), which were homozygous for unc 13 (e51) I; dpy 18 -  -  (e364) ///. In the absence of a translocation, the Fl progeny from this cross would have the genotype bli 4 + / + unc 13 (I); + / dpy 18. Segregation -  -  -  from hermaphrodites of this genotype would be expected to produce progeny in the following ratios: 6 wild-type: 3 Bli-4: 3 unc-13: 2 Dpy-18: 1 BIi-4 Dpy-18 : 1 Dpy-18 Unc-13 (ignoring suppression of the blistered phenotype by Dpy-18). However, a hermaphrodite heterozygous for a translocation involving chromosomes I and III would segregate 4 wild-type: 1 Bli-4: 1 Dpy-18 Unc-13. This apparent linkage, or psuedo-linkage of Unc-13 and Dpy-18 was used to identify candidates for 1;111 translocations. The progeny of 1400 Fl hermaphrodites were screened for Dpy-18 Unc-13 pseudo-linkage. One strain, KR1235, was identified as having a candidate translocation designated hT2. Wild type segregants of KR1235 had the genotype unc 13/hT2 I; dpy 18I hT2[bli 4(e937)] ///. Homozygous individuals -  -  -  segregated from KR1235 were blistered, with the genotype hT2 I; hT2[bli-  160  APPENDIX A  4(e937)] ///, and were used to establish the strain KR1234. The strain names reflect the order in which the strains were frozen, not isolated. To eliminate adult translocation homozygotes among the progeny of KR1235 hermaphrodites, KR1235 worms were treated with 0.012 M EMS for 4 hours and the Fl progeny screened for hermaphrodites that did not segregate Bli-4. One such hermaphrodite was isolated and used to establish the strain KR1274. This hermaphrodite presumably had a lethal mutation, h661, on one of the translocation arms, either linked or pseudo-linked to  hr2[bli 4(e937)] /// with the following genotype: unc 13/ hT2 I; dpy 18 (111)1 -  -  -  hT2I let (h661) bli 4(e937)] W. h661 was not mapped, and could be on -  -  -  either hT2 1 or hT2 III. Characterization of crossover suppression by hT2. To determine the extent of the translocation, the boundries of recombination suppression were determined. Recombination was determined from hT2 heterozygotes of the genotypes listed in Table 11. Recombination was eliminated or greatly reduced in the intervals tested between let 386 and unc 101 on LGI, and -  -  unc 64 to unc 32 on LGIII (Figure 31). This work has been refined and -  -  expanded by K. McKim (McKim, Peters and Rose, 1992). Isolation of bli-4 alleles using hT2 To identify new bli 4 alleles, three screens were conducted, -  summarized in Table 12. First, KR1274 worms were treated with formaldehyde. Hermaphrodites were suspended in 0.1% formaldehyde for four hours (Moerman and Baillie, 1981). 5000 F1 progeny were screened for blisters. No Bli-4 F1 progeny were identified. Second, KR1274 was treated with 1500 rads of gamma radiation. 30,000 Fl progeny were screened, and four Bli-4 worms were recovered in the Fl progeny; two died without progeny, and two survived. The two that survived were designated h665  161  APPENDIX A and h666. Third, KR1274 was treated with EMS. 40,000 Fl progeny were screened, and four putative bli - 4 alleles isolated, h667, h668, h669 and h670. None of these putative b/i-4 alleles have been characterized or complementation tested with other bli - 4 alleles. However, based on preliminary hT2 screen results, the hT2 system provides a rapid method for the isolation of new bli - 4 mutations. One additional putative bli - 4 mutation arose spontaneously in KR1407, a strain of the genotype unc - 74 unc - 13 I/ hT2 I; + III / hT2[Iet - (h661) bli-  4(e937)] M. This strain segregated a homozygous Unc-74 Bli-4 Unc-13 hermaphrodite, designated strain KR1803. Complementation analysis indicated that this strain carried a viable allele, h862, of bli - 4. However, when EcoRl digested DNA prepared from this strain was probed using pCeh181, the band pattern observed for the e937 deletion was observed (Figure 32). h862, therefore, is likely to be a reisolate of e937.  162  APPENDIX A  Table 11. Recombination in h72 heterozygotes. Maternal Genotype  Wild-type  Bli-4  DpyUnc  Dpy  Unc  LG I h72/ dpy-5 unc-54  238  36  10  44  41  hT2 / dpy-5 unc-101  194  40  51  0  0  hT2 / let-362 dpy-5 unc-13  132  21  0  0  la  dpy-5 unc-54 / + +  389  53  89  62  let 362 dpy-5 unc-13 / + + +  511  35  0  5  LG III hT2/ dpy-18 unc-64  447  95  87  la  la  hT2/ unc-36 dpy-18  623  117  115  0  0  hT2 / dpy-1 unc-32  127  21  11  25  14  aRecombinants were not confirmed by progeny testing.  163  APPENDIX A  14.0^1.8^  14.0  694 0.8^7.5^15.0  Figure 31. Recombination suppression in hT2 heterozygotes. Partial genetic map of LGI (top) and LGIII (bottom). Regions that do not recombine are shown in black. Wild-type genetic map distances shown below each chromosome are from Edgley and Riddle, 1990.  164  APPENDIX A  Table 12. Genetic screens using h72. Maternal Strain  Mutagen  KR1274 KR1274 KR1274  Formaldehyde Gamma radiation EMS  KR1401  Spontaneous  Chromosomes screened 5,000 31,000 40,000 Spontaneous  165  Putative bli-4 alleles identified None h66S, h666 h667, 12668, h669, h670 h862  APPENDIX A  Figure 32. Southern analysis of bli-4(h862) EcoRl digested genomic DNA probed with pCeh181. Lane 1: N2; lane 2: CB937 bli-4(e937); lane 3: KR1803 bli-4(h862). pCeh181 detects bands at 3.0 kb and 2.2 kb in N2 DNA that are absent in both CB937 and KR1803. A novel 1.7 kb band is present in both CB937 and KR1803.  166  APPENDIX B APPENDIX B. STRAINS USED IN THIS STUDY Table 13. Strainsa used in this study (excluding hP5 mapping strains, presented in Table 5) Strain  ^  Strain^Genotype  Genotype  KR2301 bli-4(e937) I; lin-14 (n179ts) X  KR0513 sDp2/dpy-5(e61) bli-4(h199)  KR1922 bli-4(e937) I lin-29(n1440) II KR1000 unc-11(e47) dpy-5(e61) dpy-14 (e188)  KR0573 sDp2/10-77 dpy-14(e188) unc-13(e450) I KR0709 sDp2/dpy-5(e61) bli-4(h384)  unc-13(e450) I  unc-13(?)/ (hDR) BC0069 dpy-14(e188) unc-13(e51) I BC0189 unc-13 (e450) I BC0835 unc-13 (e51) I; dpy-18(e364) III CB0001 CB0014 CB0024 CB0027 CB0128  dpy-1(el) III dpy-6(e14) X sqt-3(e24) V dpy-3(e27) X dpy-10(e128) II  CB0130 dpy-8(e130) X CB0164 dpy-17(6164) III CB0184 dpy-13(e184) IV CB0364 dpy-18(e364) CB0424 dpy-9(e424) IV CB0933 unc-17(e933) IV CB0973 bli-4(e937) I CB1214 unc-15 (e1214) I KR0001 Bristol (N2) KR0003 Bergerac (BO) KR0081 unc-74(e883) dpy-5(e61) I KR0214 unc-74(e883) unc-13(e51) I KR0290 sDp2/dpy-5(e61)(e51) bli-4(h42)  unc-13(e450) I KR0752 sDp2/dpy-5(e61) bli-4(h427) unc-13(e450) I KR0841 sDp2/dpy-5(e61) bli-4(h520) unc-13(e450) I KR0870 let-77 dpy-14(e188) unc-13(e450) I/ dpy-5(e61) unc-13(e450) I KR0912 dpy-5(e61) bli-4(e937) unc-13(e450) I KR1025 dpy-5(e61) bli-4(e937) I KR1123 bli-4(e937) unc-13(e450) KR1234 hT2 I; hT2[bli-4(e937)] KR1235 unc-13(e51 )/h72 I; dpy-18(e364)/h721bli-4(e937)1Ill KR1274 unc-13 (e51 )/h72 I; dpy-18(e364)/ h72(let-X(h661) bli-4(e937)fill KR1407 unc-74(e883) unc-13(e51 )/h72 I; + / hT2[let-X(h661) KR1340 sDp2/dpy-5(e61) bli-4(h699) unc13 (e450) I KR1395 sDp2/dpy-5(e61) bli-4(h754) unc-13(e450) I KR1395 sDp2/dpy-5(e61) bli-4(h754) unc-13(e450) I KR1434 sDp2/dpy-5(e61) bli-4(h791) unc-13(e450) I KR1822 unc-63(e384) unc13(eM) (I); mut-6 KR1844 lin-14(n129ts) him-5 KR1845 lin-29 (n1440)/C1 KR1858 szTl /unc-63 (384) bli-4(h1010) unc-13(e51) I RW7096 mut-6(st702) unc-22(st192::Tc1) RW7097 mut-6(st702)  unc-13(e450)  aStrains and allele designations: KR h^A.M. Rose, University of British Columbia, Vancouver, BC. CB e^MRC-LMB, Cambridge, England BC s^D L Baillie, Simon Fraser University, Burnably, BC. RW st^R. Waterston, Washington University, St.Louis, MO  167  APPENDIX C APPENDIX C. BACTERIAL STRAINS.  Table 14. Bacterial Strains  E. coli Strain OP50 BB4 DH5a  Use Nematode food X-Zap host cells Plasmid host cells  Origin Brenner, 1974 Bullock, Fernandez and Short, 1987 Hanahan, 1983; Brethesda Research Laboratories  168  APPENDIX D  APPENDIX D. COSMID AND PLASMID CLONES  Table 15. Cosmid Clones Clonea  Vectorb  B0480 C27D2 C32G12 C40A4 C44D11 C44D11 C48E7 KO4F10 K06E6 T22C4 ZC308  pJB8 pJB8 pJB8 pJB8 pJB8 pJB8 pJB8 Lorist 2 Lorist 2 Lorist 2 pJB8  Antibiotic resistance Ampicillin Ampicillin Ampicillin Ampicillin Ampicillin Ampicillin Ampicillin Kanamycin Kanamycin Kanamycin Ampicillin  aAll cosmids were a gift from J. Sulston and A. Coulson of the Medical Research Council, Cambridge, England (Coulson et al., 1986; Coulson et al. 1988). bpJB8 is described in Ish-Horowicz and Burke (1981). Lorist 2 is described in Cross and Little (1986) and Gibson et al. (1987).  169  APPENDIX D  Table 16. Plasmid clones Plasmid clone pCehl80 pCehl81 pCehl82  Insert size 11.6 Kb 11.0 2.4  pCehl95 pCehl96 pCehl97 pCehl98 pCehl99 pCeh200 pCeh201 pCeh202  6.4 4.6  pCeh205  1.7  pCeh206  0.8  pCeh207 pCeh210  3.3 0.6  Insert site  Insert source  Vectora  XhoI  KO4F10  BS KS (M13+)  Xhol Xhol EcoRI EcoRI EcoRI EcoRI EcoRI EcoRI PstIIXhoI PstIlXhol EcoRI EcoRIISalI  KO4F10  BS KS (M13+)  KO4F10  BS KS (M13+)  cDNAb  BS SK (M13-)  cDNA  BS SK (M13-)  cDNA cDNA  BS SK (M13-) BS SK (M13-)  cDNA  BS SK (M13-)  cDNA pCehl81  BS SK (M13-) BS KS (M13+)  pCehl81  BS KS (M13+)  pCehl81  BS KS (M13+)  CB937 genomic libraryc pCehl80 pCehl80  BS KS (M13+)  EcoRI EcoRI  BS KS (M13+) BS KS (M13+)  aBluescript vectors are abrievated BS. SK and KS refer to the orientation of the cloning site. " + " and "-" indicates the orientation of the M13 intragenic region. bcDNA clones were isolated from the Barstead and Waterston (1988) lamda-Zap library using the 1.3 kb h1010 probe. cpCeh206 was isolated from a lamda-Zap library of EcoRI digested CB937 genomic DNA.  170  pCeh181  x  pCeh180  x  1 1.0  1 1.6  ^I  pCeh201 pCeh202^  4.6 ^  X  pCeh207  P^6.4^ ^ X E^3,3^E  i^  I^I  E.6 E i-I  --k^  pCeh206^pCeh210  1.14^E0.8S  I----1  XP X^E^E EP^ESS ESHEE^E E P^E S^X 1^1.9^i 1.3 1 1.01 1^2.1^10.81 1.01 1.2 j^1.3_1.6,1.61^3.3^1 1.11.41^1.5^1.41^3.3  1  Figure 33. Genomic subciones of the b11-4 region Genomic restriction map (bottom) of 21 kb of KO4F10. Subclones are shown above the map and are labeled with plasmid names labeled. Numbers above lines indicate restriction fragment sizes in kb. E=EcoRI; S=Sall; P=PstI; X=Xhol.  APPENDIX E APPENDIX E. VON HEUNE PREDICTION OF BLISTERIN SIGNAL PEPTIDASE SITE. 39 amino acids at the amino terminal of blisterin B are presented at the top of this figure. A probable site for cleavage by signal peptidase was detected using the method of VonHeijne (1983). In this method, the first four consecutive hydrophobic amino acid residues are identified starting at position i, and used to define a window, i +12 to i + 20. The probability of any the amino acids within this window occuring at positions +1 to -5 in the signal peptidase recognition site was obtained from von Heijne's table. The mutiple of the probabilities for each amino acid at each possible cleavage site within the window was determined, and is presented as a probability in the table below. Using this method, the signal peptidase was predicted to cleave after Cys29,  i+12^i+20 MRISIGRIAWQILAVLIAVAFTIEHDSICDESIGACGEP +1  -1  -2  -3  -4  -5  Probabiltiy  E24 1.0  I 0.0  T 0.6  F 1.0  A 1.0  V 0.7  0.0  H25E 1.0 0.0  I 1.0  T 3.0  F 1.0  A 1.0  0.0  D26 1.0  HE 0.0 1.0  I 1.0  T 1.0  F 0.7  0.0  S27 1.0  D 0.0  H 2.0  E 0.0  I 1.0  T 1.0  0.0  128 1.0  S 1.0  D 1.0  H 0.0  E 1.0  I 0.7  0.0  C29 1.0  I 0.0  S 0.6  D 0.0  H25 1.0  E 1.0  0.0  D30 1.0  C 1.0  I 1.0  S 3.0  D 1.0  H 1.0  3.0  E3IDC 1.0 0.0 0.6  I 1.0  SD 1.0 1.0  0.0  S32 1.0  C 3.0  I 1.0  0.0  E 0.0  D 1.0  S 1.0  172  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0086382/manifest

Comment

Related Items