UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A structure-function analysis of the complex gene, bli-4, caenorhabditis elegans Srayko, Martin Anthony 1995

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

Item Metadata


831-ubc_1995-0413.pdf [ 5.88MB ]
JSON: 831-1.0086938.json
JSON-LD: 831-1.0086938-ld.json
RDF/XML (Pretty): 831-1.0086938-rdf.xml
RDF/JSON: 831-1.0086938-rdf.json
Turtle: 831-1.0086938-turtle.txt
N-Triples: 831-1.0086938-rdf-ntriples.txt
Original Record: 831-1.0086938-source.json
Full Text

Full Text

A S T R U C T U R E - F U N C T I O N A N A L Y S I S O F T H E C O M P L E X G E N E , BLI-4, I N CAENORHABDITIS ELEGANS by MARTIN ANTHONY SRAYKO B.Sc,  University of Alberta 1990  A THESIS S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in T H E F A C U L T Y O F G R A D U A T E STUDIES (Genetics Programme)  we accept this thesis as conforming to the required standard  T H E U N I V E R S I T Y O F BRITISH C O L U M B I A A p r i l , 1995 © M a r t i n A n t h o n y Srayko, 1995  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.  Department of  Mfzb/^AC  The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  / % ? y <P  /99S~  (h  £ a J £ T £  Abstract The purpose of this study was to elucidate the relationship between the molecular structure of bli-4 i n Caenorhabditis elegans and the functional role of the predicted proteins produced by this complex gene.  E n route to this goal,  the molecular structure of bli-4 has been redefined at the 5' and 3' ends of the gene; bli-4 is trans-spliced to the leader sequence SL1 and encodes 21 exons. Two approaches were adopted to study the relationship between the predicted BLI-4 products and bli-4 mutations.  The first employs a systematic  search through a portion of the gene i n five bli-4 mutant strains using P C R based heteroduplex analysis.  The second approach utilized the technique of  germline transformation rescue w i t h injected plasmid D N A .  Minigenes, or  clones of D N A that contain information necessary to encode a subset of the predicted isoforms of the gene, were constructed for this latter approach.  The  minigene rescue experiments provide a direct test for the capacity of a given isoform to rescue the different phenotypes of bli-4 (i.e., blistering a n d / o r lethality). One allele, hl99, was detected i n the 5' end of the gene as a polymorphism using the PCR-based heteroduplex technique.  D N A sequence  from homozygous arrested larvae indicated that hl99 is the result of a missense mutation, changing a histidine residue to leucine.  This amino acid  substitution is i n the amino terminus, proximal to the protease domain, and i n a region that is not particularly w e l l conserved among kex2/subtilisin-like family members. Genetic analysis suggests that the BLI-4 gene products provide at least two distinct functions: one, w h i c h when removed, gives rise to blisters, and the other, w h i c h when removed, results i n death.  Data from transgenic  minigene experiments, however, suggest that the structures of the isoforms are sufficiently similar to be functionally redundant.  In light of this new  data, it is likely that functional distinction between the bli-4 isoforms is due to pre post-translational localization and that either or both of these mechanisms are overridden by exogenous copies of minigenes that encode a subset of the total products of bli-4.  Table of Contents  Abstract  ii  List of Tables  vi  List of Figures  vii  Acknowledgements  Introduction The genetics of bli-4 bli-4 gene structure The kex2/subtilisin-like proprotein convertases  Materials and Methods Nematode culture conditions and strains Restriction enzyme digestion Agarose gel electrophoresis Polymerase chain reaction (PCR) PCR-based heteroduplex technique Generating heterozygous worms for PCR-based heteroduplex analysis Preparation of D N A for germline transformation G e r m l i n e transformation Subcloning of plasmid D N A Reverse transcriptase polymerase chain reaction (RT-PCR) C l o n i n g P C R products Testing for rrans-splicing of bli-4 R N A to leader sequences SL1 and SL2 D N A sequencing D N A sequence analysis Subcloning and the construction of bli-4 minigenes Rescue of bli-4 lethal alleles w i t h transgenic arrays Scoring the blistered phenotype  ix  1 1 3 7  11 11 13 13 13 14 17 17 18 19 22 22 23 23 24 24 25 35  V  Results  38  Section I. The 5' region of the bli-4 gene  38  A . Determination of the 5'-most sequence of bli-4  38  B. M a p p i n g mutations i n the 5' region of bli-4 B . l . hl99 is a missense mutation B.2. hl99 is a weak class II allele  41 41 47  Section II. The 3' region of the bli-4 gene  50  A . Sequence determination of pCeh207  50  B. Searching for the s90 lesion i n the 3'-specific exons of bli-4  50  Section III. The structure of the bli-4 gene Section IV. Transformation rescue" experiments w i t h subsets of bli-4 coding information A . Subclones encoding blisterase A rescued blistering and lethality B. A subclone encoding blisterase B partially rescued blistering, but not lethality C. A subclone encoding blisterases B, C , and D rescued blistering and lethality D. Transmission frequency of rescuing subclones verses proportion of rescued animals Discussion bli-4 is frans-spliced to SL1 M a p p i n g mutations i n the 5'region of bli-4 hl99 hl99/e937 hermaphrodites rarely blister Searching for the s90 lesion i n the 3'-specific exons of bli-4 The structure of the bli-4 gene Transformation rescue experiments w i t h subsets of bli-4 coding information blisterase B partially rescued blistering, but not lethality  56  56 59 62 62 62 70 70 71 73 74 77 79 80 82  Conclusions  85  References  86  Appendix 1 Appendix 2  91 92  List of Tables Table 1. Intragenic complementation of bli-4 alleles  4  Table 2. Abbreviations used i n this thesis  12  Table 3. Summary of PCR-based mismatch detection  43  Table 4. Progeny from 5 hl99/e937 hermaphrodites scored.  49  Table 5. Summary of rescue results for subclones of bli-4  60  Table 6. Strains constructed for transgenic rescue  61  Table 7. Partial rescue of blistering by pCeh238  63  Table 8. Progeny scored from transgenic lethal rescue experiments  65  List of Figures Figure  1.  Figure  2. The complex molecular nature of bli-4  Figure  3. Heteroduplex method of mismatch detection  15  Figure  4. Microinjection of plasmid D N A for germline transformation  20  5. Formation of hybrid extrachromosomal arrays is driven by homologous recombination  21  Figure  6. Construction of pCeh226  26  Figure  7. Construction of pCeh229  27  Figure  8. Construction of pCeh230  28  Figure  9. Construction of pCeh236  29  Figure 10. Construction of pCeh238  31  Figure 11. Construction of pCeh252  33  Figure 12. Strategy for determining transgenic rescue of bli-4 alleles balanced by sDp2  36  Figure 13. A test for rnms-spliced leaders SL1 or SL2 i n the bli-4 transcript  39  Figure 14. bli-4 is rrans-spliced to SL1 (spliced leader 1)  40  Figure 15. Location and sequence of primers designed to amplify the 5' end of bli-4  42  Figure 16. A polymorphism detected by heteroduplex analysis i n hl99/e937 a n i m a l s  44  Figure 17. Sequence of K R p 4 4 / K R p 4 5 P C R product from arrested hl99 homozygotes  45  Figure 18. Establishing the genomic arrangement of 3' exons i n Ceh207  51  Figure  Mutant phenotypes of bli-4  P  2 6  v i i i  Figure 19. Location of primers designed to amplify the blisterase B, C , and D 3' exons  52  Figure 20. Sequencing of 3' exons i n s90 homozygotes  53  Figure 21. The molecular structure of bli-4  57  Figure 22. Summary of results for transgenic rescue using subclones of bli-4  66  Figure 23. The relative rescuing ability of different subclones of bli-4  68  Figure 24. Subclones of the bli-4 region.  91  Acknowledgements  I w o u l d like to thank m y research supervisor, A n n Rose, for guidance and support throughout the course of this work; C o l i n Thacker for advice, encouragement, and many stimulating discussions; my supervisory committee members, D o n M o e r m a n and George Spiegelman for their constructive criticism and advice; past and present members of the Rose lab that have contributed to this work or my understanding of it; Carolyn Brown for comments on the thesis; D a v i d Pilgrim for giving me "worm fever"; and my father, A d a m , for constant support and inspiration.  Most of all, I w o u l d  like to thank Karalynn E l l for sharing this experience w i t h me.  Introduction  The genetics o f bli-4  The bli-4 gene of Caenorhabditis elegans was originally identified as a recessive mutation, e937, that results i n fluid-filled separations, or blisters, of the adult nematode cuticle (Brenner, 1974; Peters et al, 1991).  Characterization  of this mutation revealed incomplete penetrance of the phenotype; approximately 85% of e937 homozygotes exhibit the blistered phenotype.  This  reduced penetrance is a heritable feature of the mutation since a wild-type w o r m from an isogenic bli-4 strain produces the same number of blistered progeny as a blistered w o r m from the same population.  The physiological  basis for the blistered phenotype is not understood. Since the isolation of bli-4(e937), thirteen additional recessive alleles of this gene have been identified (Rose and Baillie, 1980; H o w e l l et al, 1987; Peters et al, 1991; Thacker, Srayko and Rose, unpublished data).  However,  these mutations all result i n late embryonic or early larval lethality (Figure 1).  Moreover, the host of bli-4 alleles exhibit a complex pattern of intragenic  complementation.  The thirteen lethal alleles of bli-4 have been grouped  into classes II or III, based on the phenotype observed when these lethals are i n heteroallelic combination w i t h the only k n o w n visible allele of bli-4, e937 (which belongs to class I; Peters et al,  1991).  Twelve lethal alleles, when i n  2  Figure 1. Mutant phenotypes of bli-4. Nomarski photomicrographs of (A) class I blistered phenotype of e937 homozygote, (B) class II q508 homozygote arresting development i n late embryogenesis. (C) class III s90 homozygote arresting development as an L I larva. 70% of s90 homozygotes appear as class II homozygotes (as i n B).  heteroallelic combination w i t h e937, produce blistered worms that survive to adulthood and are fertile (see Table 1).  These non-complementing alleles  have been termed class II lethal alleles.  A single lethal allele, s90,  complements e937 (heteroallelic animals appear completely wild-type) but does not complement any of the class II lethal alleles; s90 is termed a class III allele. Certain class II alleles, when placed i n heteroallelic combination w i t h e937, result i n 100% penetrance of the blistered phenotype (Thacker, et al, 1995; also see Discussion).  Because C. elegans populations are isogenic, this  result suggests that e937 is hypomorphic and that the class II alleles implicated i n this phenomenon are good candidates for n u l l alleles. Furthermore, the phenotype of animals homozygous for class II mutations represents the most severe phenotype of all alleles.  Muller's criterion (1937)  for designation of the n u l l state requires placement of the allele i n question over a deletion that removes the gene.  Unfortunately, no such deletion is  currently available for unequivocal categorization of bli-4 alleles at this time.  bli-4 gene structure  The cloning and molecular characterization of bli-4 was initiated by K e n Peters (1992).  A t that time, the complex nature of the gene's structure became  apparent (Figure 2).  Three different species of c D N A s were k n o w n and  matched to genomic regions, p r o v i d i n g evidence that the bli-4 gene produced at least three isoforms, arising v i a differential splicing. consistent w i t h the nature of a complex genetic locus. mutations of bli-4 (Figure 2).  The structure of bli-4 is Peters also mapped two  The blistering phenotype i n e937 homozygotes  Table 1. Intragenic complementation of bli-4 alleles  ClassI ClassI  Blistered (85%)  a  Class II  b  Class III  C  Class II  a  Class III  C  d  Blistered (100%) Wild-type  b  d  Arrest 3-fold Arrest 3-fold  Arrest 3 - f o l d / L I larvae  The complementation pattern for 12 lethal alleles and the visible allele, e937, is shown. A l l lethal mutants arrest development i n late embryogenesis. 30% of s90/s90 homozygotes survive until the L 1 / L 2 larval stage. (Peter's, 1992; Thacker, et al, 1995). a  C l a s s I mutant is represented by e937.  b  C l a s s II mutants are represented by hlOW and q508.  c  T h e class III mutant is represented by s90.  ^Percentage of blistered animals. Approximately 85% of e937 homozygotes exhibit the blistered phenotype. The blistered cuticle phenotype is fully penetrant i n the e937/class II animals.  is due to a deletion of 3.5 kb that removes an exon specific to one of the isoforms.  hlOW is a class II allele, recovered from a strain w i t h transposon-  based mutator activity that was mapped to the first 12 exons of bli-4. One of the goals of this thesis was to determine the relationship between the mutations associated w i t h bli-4 and the molecular structure of this gene.  Previous mutational analysis has provided important clues about  the function of the predicted products of the gene.  Based on Peter's data, the  blistering phenotype could be attributed to the loss of an exon that is specific to one of the isoforms produced by bli-4.  Indeed, it was later shown that the  expression of this isoform's transcript is absent i n e937 homozygotes (Thacker, et al, 1995).  Therefore, the transcript to w h i c h this exon belongs is  likely not essential for development.  Also, the non-complementing class II  allele, hlOW, is a transposon insertion into exon 9, an exon that all predicted transcripts share (Thacker, et al, 1995).  This mutation suggests that one or  more of the remaining products of bli-4 has an essential function. The region common to all transcripts (the first 12 exons shown i n Figure 2) shows sequence similarity to serine endoproteases and overall structural similarity to the kex2/subtilisin-like family of proprotein convertases (Thacker, et al, 1995).  This family of enzymes is implicated i n  the proteolytic activation of many biologically important precursor proteins. It is expected that bli-4 is involved in enzymatic processing of at least one protein that is essential for the development of the organism.  The variable  phenotypes associated w i t h this single gene may be due to mutations that alter functionally distinct proteases that arise v i a alternative splicing.  hl010::Tcl  1 kb e937  X  ^  ^  ^  ^  F ^ a l Sal  E SaKXh  E  E  y^ffiffij^  frTrU^J^  A  b  F  w  m  x  MA  ^  —  '-  ( A ) n  -(A)n  Alignment based on: |  Sequence data and hybridization Preliminary sequence data and hybridization Hybridization data only  Figure 2. The complex molecular nature of bli-4. The structure of bli-4 and the alignment of k n o w n c D N A clones by Peters (1992) is shown. A t this time, three isoforms were predicted to be encoded by bli-4, and the relative postions of unique 3' regions were positioned b y sequence a n d / o r hybridization analysis. The transposon insertion mutant hlOW was mapped to the common region but the exact insertion site was not known. e937 was found to be a deletion that removes 3.5 kb of D N A , including exon 13.  The kex2/subtilisin-like proprotein convertases  Regulating the activity of gene products is an important issue for any biological system.  In some instances, proteins must be produced i n a cell  (where all transcriptional and translational machinery exists) but not allowed to function until they are safely sequestered i n some compartment of the cell, or are exported out of the cell.  One of the ways to prevent inappropriate  protein activity is to first make a nonfunctional form of the protein that can be activated at some later time.  Evidence for this mechanism of post-  translational control was inferred from the discovery that pituitary hormones (Chretien and L i , 1967) and insulin (Steiner et al, 1967; Chance, Ellis and Brommer, 1968) are synthesized as inactive precursors.  These and  other inactive precursor proteins are subjected to limited endoproteolytic cleavage u p o n their secretion.  This cleavage, w h i c h occurs most frequently  after pairs of basic amino acids such as L y s - A r g or A r g - A r g (reviewed by Docherty and Steiner 1982; Thomas et al, 1988) results i n activation of the molecule.  One mechanism of endoproteolytic cleavage found i n all  eukaryotes examined occurs via the processing activity of the kex2/subtilisinlike family of proprotein convertases (reviewed by Barr, 1991; Seidah and Chretien, 1992).  Kex2p, the prototype member of the convertases, is a  membrane-bound Ca^ -dependent serine endoprotease that cleaves the yeast +  pro-a-mating factor at dibasic residues (Mizuno et al, 1988; Fuller et al, 1989a; reviewed i n Fuller et al, 1988).  M a m m a l i a n members of the  convertase family include P C 1 / P C 3 (Seidah et al, 1991; Smeekens et al, 1991), P C 2 (Seidah et al, 1990; Smeekens and Steiner, 1990), P C 4 (Nakayama et al, 1992; Seidah et al, 1992), P C 5 / P C 6 (Lusson et al, 1993; Nakagawa et al, 1993), P A C E 4 (Keifer et al, 1991), and furin (Roebroek et al, 1986; Fuller et al, 1989b;  van den O u w e l a n d et al, 1990; Wise et al, 1990).  T w o genes isolated from  Drosophila, called Dfur-1 (Roebroek et al, 1992, 1994; also called dKLIP-1; Hayflick et al, 1992), and Dfur-2 (Roebroek et al, 1992) encode convertases w i t h sequence similarity to furin. For scientists studying the kex2/ subtilisin-like convertases, the relationship between the convertase enzymes and the precursor proteins on w h i c h they act is elusive.  Establishing the enzyme-substrate connection has  been attempted by coexpression studies; one can determine if proprotein processing occurs by exogenous expression of specific convertases w i t h candidate substrates.  However, i n this artificial context, many different  endoproteases are able to process the same substrates.  Whether such  functional redundancy exists amongst the family members in vivo remains a mystery.  One approach to determine possible convertase-substrate  interactions involves restricting the endogenous expression pattern and localization of the i n d i v i d u a l convertases a n d / o r potential substrates.  Such  analyses have shown that substrate specificity can be influenced by both restricting expression to particular tissues, and compartmentalization of the i n d i v i d u a l enzymes to specific intracellular locations (Seidah and Chretien, 1992). Some members of the family exhibit restricted expression patterns in vivo and participate i n the regulated secretory pathway.  For instance,  P C 1 / P C 3 and P C 2 is restricted to endocrine and neuroendocrine tissues (Seidah et al, 1990, 1991; Smeekens and Steiner, 1990; Smeekens et al, 1991), and P C 4 is restricted to the testis (Nakayama et al, 1992; Seidah et al, 1992). Therefore, some functional distinction can be inferred from those convertases whose expression pattern does not overlap.  A potential  substrate that is never localized i n the testis w o u l d not be predicted to be  processed by PC4, for instance.  In contrast, both furin and P A C E 4 are  expressed i n a broad range of tissues and participate i n the constitutive secretory pathway (Roebroek et al, 1986; van den Ouweland et al, 1990; V a n de V e n et al, 1990; Bresnahan et al, 1990; Kiefer et al, 1991).  This  constitutive expression complicates the task of determining convertase substrate interactions. tissue distribution.  P C 5 / P C 6 , like furin and P A C E 4 , exhibits a widespread  However, levels of expression are highest i n intestinal  tissue (Lusson et al, 1993; Nakagawa et al, 1993), indicating that P C 5 / P C 6 may participate i n both the constitutive and regulated secretory pathways. Intracellular compartmentalization may also influence proprotein convertase activity a n d / o r substrate specificity. Localization of the i n d i v i d u a l processing enzymes is likely intrinsic to the particular convertase examined.  For instance, furin is concentrated i n the rrans-Golgi network  (TGN) (Molloy et al, 1994).  A l l members of the kex2/subtilisin-like family  are first synthesized as an inactive zymogen (Leduc et al, 1992; reviewed i n Seidah and Chretien, 1992).  Furin becomes active at the same time the  enzyme is released from the endoplasmic reticulum and transported to the TGN.  The enzyme is also observed to cycle back and forth from the cell  surface v i a clathrin coated vesicles, much like E G F receptor molecules. Trafficking signals implicated i n this cycling are encoded w i t h i n the cytoplasmic tail of furin (Molloy et al, 1994). Defining the function of the proprotein convertases likely requires a system whereby the in vivo context of activity is not perturbed.  Therefore,  mutational analysis is a powerful method of identifying components of this complicated process of post-translational control.  The identification of  mutations w i t h i n a gene that encodes a proprotein convertase is a luxury few researchers i n this field possess; only yeast kex2 and C. elegans bli-4 mutants  10 have been reported.  In addition, the bli-4 gene offers an opportunity to  examine the function of structurally distinct isoforms that resemble convertase members of both the regulated secretory pathway and constitutive secretory pathway (Thacker, et al, 1995).  This gene represents a scientifically  challenging target of study due to its genetic and molecular complexity. By taking advantage of the amenable genetics and molecular biology of the C . elegans model system, bli-4 may reveal important insights into enzymatic processing i n a w i d e range of organisms/including humans.  Materials and M e t h o d s  Nematode culture conditions and strains  A l l C. elegans strains used i n this study were maintained on petri plates containing nematode growth media ( N G M ) streaked w i t h Escherichia coli OP50 at 20°C unless otherwise indicated (Brenner, 1974). /  The canonical wild-type nematode strain (+/+) is C. elegans, var. Bristol, strain N 2 . Mutations i n C. elegans are assigned two descriptors, a genetic locus, and a corresponding allele that represents the mutational event giving rise to the associated phenotype.  For example, the mutation e937  results i n a blistered phenotype (worms w i t h blistered cuticles), and is an allele of the bli-4 gene; this mutation is written as bli-4(e937). genotypes are italicized and phenotypes are not.  In general,  Upper case letters are used  when referring to gene products; for example, bli-4 encodes four BLI-4 protein isoforms.  Nomenclature guidelines for C. elegans have been  published by Horvitz, et al. (1979).  Nomenclature of strains and materials  used i n this thesis is presented i n Table 2.  Table 2.  Abbreviations used in this thesis.  Abbreviation  Description  Mi  Mutations in these genes give rise to a blistered cuticle  Bli  the Blistered phenotype; fluid filled separations of cuticular tissue  dpy  Mutations in these genes result in a dumpy phenotype  Dpy  the Dumpy phenotype: short, fat body morphology  unc  Mutations in these genes result in uncoordinated movement.  Unc  the Uncoordinated phenotype: an impairment or abolishment of locomotion or sensation  rol  Mutations to these loci result in a rolling phenotype  Rol  the Roller phenotype: a helical twisting of the body around the longitudinal axis, resulting in rolling motion as the worm moves forward  h  The Rose laboratory allele designation. A l l alleles, extrachromosomal arrays, chromosomal rearrangements and D N A constructs (i.e., genotypic alterations) designed in this laboratory are issued an h number.  KR  The Rose laboratory strain designation. A l l C. elegans strains isolated in the Rose laboratory are issued a KR number.  KRp  All oligonucleotides, or primers, designed in the Rose laboratory for the purpose of Polymerase Chain Reaction are issued a KRp number.  pCeh  D N A constructs are identified as pjasmids subcloned from Caenorhabditis elegans in the Rose (h) laboratory.  hEx  The exogenous D N A construct present within a nematode strain transformed via microinjection in the Rose (h) laboratory.  Restriction enzyme digestion  D N A was digested w i t h restriction enzyme(s) supplied by either Pharmacia or N e w England Biolabs.  Digests were performed w i t h a three-  fold excess of enzyme (i.e., 3 units/p.g of D N A ) i n the appropriate buffer (supplied by the manufacturer) for at least one hour at the recommended temperature.  Bovine serum a l b u m i n (BSA) was also included i n restriction  digests at a concentration of 100|ig/ml w h e n recommended by the restriction enzyme manufacturer.  Agarose gel electrophoresis  D N A was size-separated i n 0.7-1.5% w:v agarose gels i n 0.5X TBE buffer [IX TBE is 0.89 M Tris, 0.89 M boric acid, 1 m M E D T A ( p H 8.0)] w i t h approximately 0.1u.g/ml ethidium bromide.  Gels were electrophoresed i n  0.5X TBE running buffer at approximately 5V7cm.  The size standard  included i n all gels was 1 kb ladder (Bethesda Research Laboratories). D N A bands were detected and photographed under illumination w i t h 300nm U V .  Polymerase Chain Reaction (PCR)  Template D N A from i n d i v i d u a l worms or arrested embryos homozygous for bli-4 mutant alleles was extracted as described (Barstead, et al, 1991).  Most polymerase chain reactions were carried out i n a Perkin-  E l m e r / C e t u s P C R machine for 30 cycles of denaturation (94°C, 45 seconds),  annealing (54°C-60°C, depending on primers, 30 seconds), and extension (72°C, 1 minute) followed by extension at 72°C for 7 minutes.  Reactions were  performed w i t h Taq Polymerase and accompanying buffer system from Promega or w i t h Pfu polymerase and buffer system from Stratagene.  Some  reactions were carried out i n 25(il capillary tubes w i t h a hot-air thermocycler (Idaho) using the Idaho buffer system.  PCR-based Heteroduplex Technique  The principle behind the heteroduplex technique for identifying mutations is based on a well-studied property of D N A , the tendency of singlestranded molecules to associate, or hybridize, to their complementary forms. A s the title implies, heteroduplex technique involves detecting associated D N A molecules (duplexes) w h i c h are not completely complementary.  In  general, one selects a fragment of D N A to study and mixes this i n a 1:1 ratio w i t h the same fragment isolated from a mutant strain.  The D N A is  denatured into single-stranded form and allowed to slowly reanneal.  If a  mutation is present i n one half of the starting material, three forms of duplex molecules w i l l form: wt:wt homoduplex, m u t m u t homoduplex, and wt:mut heteroduplex.  Due to structural alteration of the duplex i n the  wt:mut form, this d s D N A w i l l migrate aberrantly through an acrylamide matrix.  Generally, mutations are detected as two bands on a gel, representing  both species of homoduplexes migrating at normal position compared w i t h a wt:wt control and a slightly fainter band migrating slightly slower, representing the heteroduplex D N A (see Figure 3) (Keeri, et al, 1991).  15  dpy-5  J  I  dpy-5  J  bli-4 (class II)  unc-13  bli-4 (e937)  L  bli-4 (class II)  I  unc-13  L  _ l  X  bli-4 (e937)  _ l  +  L +  L  c T  -I sDp2  dpy-5  bli-4 (class II)  unc-13  Figure 3. Heteroduplex method of mismatch detection. The procedure employed for the detection of class II alleles h791, h384, h254, hl99, and h670 is shown. Individual worms that are heteroallelic for each of the five mutations are used as template for P C R (30 cycles), using a set of primers that amplifies a segment of genomic D N A . The resulting P C R products are denatured and slowly reannealed to form stable duplexes. If a mutation is present, three types of duplexes w i l l form: wt:wt, wt:mut, and mut:mut. This D N A is run in a non-denaturing acrylamide matrix. Typically, heteroduplex migration is retarded when compared to homoduplexes and is slightly fainter due to 1/3 representation of the total pool of P C R products (Keen, et a l , 1991).  Polymerase C h a i n Reaction (PCR) was performed on heterozygous i n d i v i d u a l blistered adult worms of the genotype dpy-5 bli-4(class II allele) unc-13/bli-4(e937).  The products were checked for size, purity, and  approximate concentration by loading 2-3 ul (one tenth) of the total yield i n a 1-1.5 % agarose gel along w i t h a size standard (Gibco B R L 1 kb ladder) and a D N A source of k n o w n concentration (typically, pRF4 (rol-6) plasmid at 50, 100, and 200ng). Samples were incubated at 95°C for three minutes and slowly (at least 30 minutes) cooled to 37°C.  The samples were then removed from 37°C and  kept at room temperature prior to loading.  MDE™ (mutation detection  enhancement) matrix was prepared as suggested (AT Biochemicals). Approximately 200ng of D N A (in less than 14 ul volume) was loaded per well (with 1/10 loading buffer, provided by manufacturer) for each sample. A negative control (N2) and a positive control (unc-52/+)  were included i n  each gel. The positive control, CB1012/+ worms, were subjected to P C R using the antisense primer plO ( 5 - C T G G T G G G C T A T T C T C T G G-3') and the sense primer peg8 (5'-GAC A T C C A A G T G T T C A G C -3'), w h i c h amplify a -590 bp product of exon 16 and 17 from the unc-52 gene (primers were a gift from D o n Moerman).  W h e n using CB1012/ + worms as template for P C R ,  the mutation el012 (a G : C to A : T transition mutation i n exon 17 of unc-52) is amplified from one half of the starting material. 2 0 V / c m for a distance of ~30 cm.  The products were run at  The acrylamide matrix was stained w i t h  ethidium bromide ( l j i g / m l ) and photographed by illumination w i t h 300nm UV.  Generating heterozygous worms for PCR-based heteroduplex analysis  To generate heterozygosity at the bli-4 locus for heteroduplex analysis, the following cross was performed.  Balanced strains carrying lethal class II  alleles of the genotype dpy-5(e61) bli-4(class II lethal) unc-13(e450); sDp2 were crossed to homozygous males of the genotype bli-4(e937).  Due to intragenic  complementation (Table 1) the resulting F l heteroallelic progeny are blistered.  These blistered worms were used individually for P C R  amplification.  One exceptional allele, hl99, d i d not result in blistering very  often when i n heteroallelic combination w i t h e937 (see Results and Discussion); however, only blistered F l animals (hl99/e937) were used for heteroduplex analysis to ensure the correct heteroallelic genotype. h670 was maintained i n strain KR2486 as a heterozygote of the genotype:  + bli-4(h670) unc-13(e450)/dpy-5(e61) + unc-13(e450).  Unc  animals that gave approximately 1/4 dead progeny were used as template for PCR-based heteroduplex analysis.  Preparation of D N A for germline transformation  Plasmid D N A was purified from bacterial cultures by one of three methods.  One method involves standard mini-prep procedure w i t h  phenol/chloroform extraction and ethanol precipitation followed by resuspension of the D N A pellet w i t h T.E. + R N A s e (lOfig/ml).  The second  method utilizes a commercially available kit (Wizard M i n i - P r e p , Promega) to isolate plasmid D N A from an overnight culture of bacteria.  This procedure,  however, d i d not give any transformants unless the D N A was further  precipitated (70% ethanol, 0.3M N a O A c ) and washed w i t h 70% ethanol.  The  third method is C s C l preparation of plasmid D N A , as discussed i n Sambrook, et al, (1989).  The most consistent results were achieved w i t h the  C s C l preparation. The concentration of plasmid D N A after preparation was estimated by loading a fraction of the total volume i n an agarose gel along w i t h a control plasmid of k n o w n concentration, or, i n the case of C s C l D N A , estimated by measuring the optical density at 260 n m wavelength w i t h a spectrophotometer.  D N A constructs were coinjected w i t h the plasmid pRF4  (a gift from C . Mello), carrying the dominant marker rol-6(sul006) (Kramer, et al, 1990) at a total concentration of approximately lOOng/ul.  pRF4 confers  a dominant roller phenotype (Rol) to C. elegans strains that carry this plasmid. (Kramer et al, 1990).  Germline transformation  D N A was injected into the distal arm of one or both gonad arms i n adult hermaphrodites (shown i n Figure 4) after the method of M e l l o et al, (1991).  Recombination between separate coinjected D N A molecules has been  observed (Mello et al, 1991).  This recombination occurs i n regions of  sequence identity to create large concatemeric hybrid molecules w h i c h exist as extrachromosomal forms (see Figure 5). Approximately ten percent of F l s that carry an extrachomosomal array w i l l transmit that array i n a stable, non-Mendelian manner to the F2 generation.  Provided the array is large enough, it w i l l be faithfully  transmitted to progeny cells i n the developing egg after fertilization w i t h  sperm.  The inheritance of an array can be detected by the dominant R o l  phenotype;  rolling F l animals from an injected hermaphrodite are likely to  carry both the pRF4 plasmid and any other coinjected plasmid that bears sequence identity to pRF4.  For instance, pBluescript-based clones and pRF4  share sequence identity i n at least one location of their plasmid backbone, since both vectors carry the ampicillin resistance gene, amp . v  Stable  transgenic lines were used i n rescue analysis of bli-4 lethal alleles.  Subcloning of plasmid D N A  Target D N A and vector were digested w i t h appropriate restriction enzyme(s) supplied by either Pharmacia or N e w England Biolabs as described above.  Enzymes were inactivated as per manufacturer's suggestion w h e n  the products were used directly, otherwise, the resulting products were separated from enzymes and extraneous D N A v i a agarose gel electrophoresis.  Extraction of D N A from agarose gels was achieved w i t h the  Qiagen purification system.  Vector (pBluescript I (SK-) or (KS+)) (Stratagene)  and target D N A were ligated at 16°C overnight using T4 D N A ligase (New England Biolabs, as described by Sambrook, et al, 1989). Ligated D N A was used to transform competent E. coli strain D H 5 a (Bethesda Research Laboratories).  Transformed cells were identified as being resistant to  ampicillin, due to the production of ^-lactamase provided b y the amp gene r  present i n pBluescript.  The a-complementation system (Ullmann et al,  1967) was used to distinguish between parent and recombinant plasmids; cells were g r o w n i n the presence of a chromogenic substrate, X-gal (5-bromo4-chloro-3-indolyl-(3-D-galactoside) and IPTG (isopropyl thiogalactoside).  Figure 4.  Microinjection of plasmid D N A for germline transformation.  D N A prepared for germline transformation was microinjected into the distal gonad arm (arrow) of young adult hermaphrodites, as described i n Mello, et al., (1991). Photograph courtesy of J. M c D o w a l l .  stable hybrid array  Figure 5. Formation of h y b r i d extrachromosomal arrays is driven b y homologous recombination. Plasmid constructs coinjected into the distal arm of hermaphroditic gonads undergo recombination at sites of homology (for instance, the ampidllin resistance gene in pBluescript and pRF4(ro/-6) plasmids). Arrays consisting of many copies of each plasmid type are maintained by non-Mendelian inheritance. Strains exhibiting g o o d transmission (20% or greater) are used for subsequent crosses.  Parental plasmid-bearing clones appear blue and recombinants appear white when grown i n the presence of I P T G / X - g a l (Horwitz et al, 1964). Plasmid D N A was recovered from appropriate clones by conventional alkaline lysis miniprep procedure (Sambrook, et al, 1989).  Reverse transcriptase polymerase chain reaction (RT-PCR)  First strand c D N A was synthesized from 5ug of total R N A using an oligo-dT primer and the Superscript Preamplification system (Gibco BRL). Conditions for synthesis were as described by the manufacture. performed on poly A  +  R T - P C R was  R N A isolated from wild-type (N2) worms.  conditions for P C R were as described above.  Reaction  R N A was kindly provided by  C o l i n Thacker.  Cloning PCR products  C l o n i n g of P C R products involved size-separation of the appropriate band(s) v i a agarose gel electrophoresis, followed by purification w i t h either W i z a r d P C R Prep (Promega) or Qiagen systems.  Blunt-end ligation was  performed to insert the P C R products into the EcoRV site of the pBluescript(SK) polylinker. W h e n Taq polymerase was used i n the P C R reaction, a further step was required to repair the ends prior to blunt-end ligation.  This was accomplished by adding 1 unit of T4 D N A polymerase  (NEB) in a final volume of 50(4,1 of I X buffer (supplied) and a final concentration of 200mM dNTPs.  P C R reactions using Pfu polymerase  (Stratagene) were cloned directly into the EcoRV site without prior modification.  Testing for trans-splicing of bli-4 R N A to leader sequences SL1 and SL2  R T - P C R was performed on N 2 R N A w i t h an antisense primer K R p l O ( 5 - A C T C T C T T C T T C G G T CGC-3') situated i n exon III of bli-4, i n combination w i t h one of two sense primers, S L l ( p l u s Notl adapter) (5'-ATA A G A A T G C G G C C G C G G TTT A A T T A C C C A G T TG-3') or SL2 (plus Notl adapter) ( 5 - A T A A G A A T G C G G C C G C G G TTT T A A C C C A G T T A C T C A 3').  T w o bands were observed after amplification w i t h the SL1 primer, no  product of expected size was visible w i t h the SL2 primer.  The major product  was a band of smaller than expected size (~300bp) and was cloned directly after gel-purification; sequencing revealed this D N A was not bli-4 derived. The faint band of expected size (~500bp) amplified w i t h the SL1 and K R p l O primer was gel-purified and this D N A was used as template for a further round of P C R , using SL1 primer and a nested primer, K R p l l .  A resultant  single band of expected size (~450bp) was gel-purified. This fragment was cloned and sequenced (shown i n Figure 15, Results).  D N A sequencing  Sequences of pBluescript plasmid inserts were obtained w i t h Sequenase Version 2.0 (United States Biochemical (USB)), w i t h encorporation for detection of D N A by autoradiography.  3 5  S-dCTP  Preparation of  d s D N A template was accomplished by purifying D N A from 3-4.5 m l overnight bacterial cultures w i t h W i z a r d Purification preparations (Promega).  Subsequent steps performed to prepare template were as  suggested by USB.  Primers used for sequencing were  17,13,  universal,  reverse, S K , and K S , obtained from Stratagene.  D N A sequence analysis  D N A sequences were translated and restriction mapped w i t h the D N A Strider program for A p p l e Macintosh.  A m i n o acid sequences generated  from Strider were used to search SWISSPRO and G E N B A N K databases using the B L A S T network service (email address: blast@ncbi.nlm.nih.gov). Computations for searches were performed at the National Centre for Biotechnology Information (NCBI).  Subcloning and the construction of bli-4 minigenes  Most minigenes were constructed from four existing plasmid clones of bli-4, p C e h l 8 0 , pCeh.181, pCeh220, and pCeh221 (courtesy of K e n Peters and C o l i n Thacker). pCehl80.  Most 3' exons were subcloned from the parent clone  The putative 5' promoter region and 5' exons included i n all  minigenes were derived from pCeh220, pCeh221, p C e h l 8 1 and subclones thereof.  A p p e n d i x 1 shows all relevant subclones used i n this thesis.  cloning steps for constructing minigenes are outlined i n Figures 6 to 11.  The  Rescue of bli-4 lethal alleles with transgenic arrays  In order to determine whether a transgenic array rescued bli-4 lethality, the following crosses were performed.  Each.of the lethals, maintained w i t h  the balancer sDp2, were crossed to N 2 males.  q508 was maintained i n the  strain KR2572 w h i c h has the genotype: dpy-5(e61) bli-4(q508); sDp2. hl99 and s90 were maintained i n strains KR513 and KR2728 respectively and have the corresponding genotypes: dpy-5(e61) bli-4(hl99) unc-13(e450); sDp2, and dpy5(e61) bli-4 (s90) unc-13(e450); sDp2.  After allowing mating to occur (typically  16 hours) hermaphrodites were placed on i n d i v i d u a l plates.  F r o m these  plates, F l males (heterozygous for the lethal mutation and possibly carrying sDp2) were crossed to Rollers from the strain carrying the extrachromosomal array.  Males w h i c h carry sDp2 are k n o w n to mate very poorly so their  presence i n the F l s was not problematic (Rose, et al, 1984).  Rollers were  removed from the mating plate after 16 hours and allowed to lay eggs on i n d i v i d u a l plates.  O n l y those R o l hermaphrodites w h i c h gave male progeny  were used for further analysis.  From plates w i t h male progeny,  hermaphrodite R o l F2s were picked to individual plates.  O n l y plates of F2s  w i t h some arrested embryos (evidence for the lethal bli-4 allele) were scored for rescue; even if rescue is possible, only a fraction of the lethal homozygotes w i l l receive an unstable extrachromosomal array, thus arrested embryos should be present regardless of the rescuing capability of the array. A l l F3 progeny were scored from these F2 parents.  For alleles hl99 and s90,  rescue was evidenced by the presence of D p y U n c animals i n the F3 generation. array.  For q508, D p y animals were present if rescued b y the transgenic  Both Dpy-5 and Unc-13 are epistatic to Rol-6, therefore, the R o l  A Xb  lkb Xh  _L  Sac E  ' •  E P  E Sal Sal  E Sal XhPEXbEEK Sac K  E  [  E P Sal  Li I  pCeh220 E = EcoRI K = Kpnl P = Pstl Sac = Sad Sal = Sal\ Xb = Xbal Xh = Xhol  pCehl81  E Sal Sal I -I I  B  E Sal Xh 1 >T  P / K 5 k b fragment  Sal Sal  E Sal: K  pCeh220 (P/K)  pCeh226 Figure 6. Construction of pCeh226. A ) The substrate clones used i n the construction of pCeh226 are shown. B) A schematic representation of the cloning steps involved i n the construction of pCeh226.  27  A Xb  I  lkb  E  J  Xh  SacE  E  EP  1 nunuiL P  E Sal Sal  _a] I  1 wl  E Sal Xh PEXbEEK Sac K  1  )  E E  ''ill hffaf f^HB^innm^  P Sal  ^ ^  Ceh226 E = EcoRI K = Kvnl P = PstI Sac = Sacl Sal = Sail Xb = Xbal Xh = Xhol  B Xb  E  Xh  Sac E  E .EP  E Sal Sal  a  1  1  E Sal Xh  ' ~~1  pCeh226  digest with Sal religate  Xb  pCeh229  Figure 7. Construction of pCeh229. A ) The substrate clone pCeh226 used i n the construction of pCeh229 is shown. B) A schematic representation of the cloning steps involved i n the construction of pCeh229.  ^  A Xb  ikb E  Xh  Sac E  E  E  pCeh220  B  E Sal Sal  -UL-J  E Sal Xh PEXbEEK Sac K  1 M )\(&LManL-ll  E E  P Sal  H pCeh205  E = EcoRl K = Kpnl P = Pstl Sac = Sad Sal = Sail Xb = Xbal Xh = Xhol  p E Sal  K  pCeh205  P/K  p  E Sal  K  ligate to pCeh220 (P/K)  pCeh230 Figure 8. Construction of pCeh230. A) The substrate clones used i n the construction of pCeh230 are shown. B) A schematic representation of the cloning steps involved i n the construction of pCeh230.  1  Figure 9. Construction of pCeh236. A ) The substrate clones used i n the construction of pCeh236 are shown. B) A schematic representation of the cloning steps involved i n the construction of pCeh236. A r r o w s indicate direction of bli-4 reading frame.  A  lkb  Xb  Xh  P  Sac E  E  E Sal Sal  p C e h 180  E P Sal ^I 1  pCehl80  Ceh220  Xh \ E X b E E K SacK  toXh  E = Eco RI K = Kpnl P = PstI Sac = SscI Sal = Ss/I Xb = Xbal Xh = Xhol Sp = Spel Not = Notl  Sp  B  E [  E Sal XhPEXbEEK Sac K  E  E  Sal  Xh  ^ C l o n e P/Sp 3.4 fragment  Clone P/Sp 1.7 fragment  Not^—mmf \pCeh233 pCeh232  X  Cut with P Fill in P(blunt) Cut with Sp >Xbf y Clone into Xb(blunt)-Sp BSSK  Sp/Not  Sp  Blunt  \mm—•j-Not \pCeh234^/  V  pCeh235  >/  pCeh220 . P/Not Not  •^•pCeh236  Figure 10. Construction of pCeh238. A ) The substrate clones used i n the construction of pCeh238 are shown. B) A schematic representation of the cloning steps involved i n the construction of pCeh238.  32  A  ikb  Xb  Xh  P  Sac E  E Sal Sal  _UI  I  \ > f -Mi  '  '  Xh L  Sac E  -c:' '  P Sal  J J  ^  pCeh232  Ceh220 pCeh!81  B  E  E Sal X h P E ^ b E E K | Sac K  E P  -J_L  E Sal Sal  -Ul  E = EcoRl K = Kpnl P = Pstl Sac = Sacl Sal = Sail Xb = Xbal Xh = Xhol B = BamHl Sp = Spel  E Sal Xh  I L_^/  B  pCehl81 Clone P / E 1.7 fragment  P  K  Ceh232 digest with B fill in Bflblunt) digest with K  [A  Blunt  K  pCeh231 digest with E fill in E(blunt) digest with K  pCeh220  pCeh238  I  Figure 11. Construction of pCeh252. A ) The substrate clones used i n the construction of pCeh252 are shown. B) A schematic representation of the cloning steps involved i n the construction of pCeh250. pCeh244 is a PCR-derived product from single hl99 homozygous arrested embryos, amplified w i t h Pfu polymerase using primers KRp29 and KRp45 (Figure 15). The amplified product was gel-purified and digested w i t h CZal and Sad, and cloned into BSSK {Clal/Sacl). A parallel series of experiments using a P C R product from wild-type animals were also performed. pCeh250 was subsequently used to construct pCeh252, a subclone analogous i n structure to pCeh229 (Figure 7). The procedure was as follows: pCeh229 was digested w i t h PstI and Kpnl (Kpnl is present i n the poly linker, 3' of the Sail site), and the ~2 kb fragment containing exon 13 was ligated into pCeh250 (digested w i t h Pstl and Kpnl) to produce pCeh252. Wild-type sublcones analogous to hl99 subclones are: pCeh244(M99) and pCeh245(N2); pCeh246(W99) and pCeh247(N2); pCeh250(M99) and Ceh251(N2); pCeh252(/z299) and pCeh253(N2). P  34  A Xb  I  ikb  E  Sac Xh C C Sac E  1_L  E  EP  E Sal Sal  I I MbjUu&mmi^ H1 L_  E Sal Xh PEXbEEK Sac K  HpCeh244  E = EcoRI K = Kpnl P = Pstl Sac = Sad Sal = Sail Xb = Xbal Xh = Xhol C = CM  pCeh220 & pCeh221  I  pCehl81 pCeh249  B P  Ceh221  E E P Sal  pCeh224  Xh C (partial) 1.1 kb  C/Sac P  Xh C C  C  Ceh249  *, Sac Sac/K 3 kb  two factor ligation into Xh/Sac BSSK XhCC  Sac E  tsac  pCeh246  Xb  E  Sac Xh  K  pCeh220 (Xh/K).  Xb * =hl99  E  Xh Sac  I  E  E  E P  alteration  A-*-T substrate for  pCeh252  phenotype is masked i n the event of rescue by the transgene.  Stable D p y or  D p y U n c lines were established for each rescue and given a strain designation. Furthermore, a few animals from each rescued line were crossed to N 2 males to ensure that R o l progeny resulted, proof that the extrachromosomal array was present and faithfully propagated i n rescued animals.  A general scheme  is presented i n Figure 12.  Scoring the blistered phenotype  Blistering was scored as either positive or negative, w i t h no attention paid to severity of individual blisters (i.e. size or volume) or to the number of blisters per animal. In cases where a transgenic array was tested for ability to rescue, the dominant marker rol-6(sul006) was used to delineate worms carrying the transgene from those that do not (see germline transformation, above).  Adequate mobility by the w o r m is required to observe the R o l  phenotype conferred by the rol-6 marker  However, the blistered phenotype,  w h e n severe, can greatly impede movement, making it difficult to score the Rol phenotype.  Therefore, rollers were picked off onto a new plate as L4 or  earlier larvae (prior to the adult stage when blistering occurs) and then scored for presence or absence of blistering.  Figure 12. Strategy for determining transgenic rescue of bli-4 lethal alleles balanced by sDp2. Lethal alleles balanced b y sDp2 (e.g., hl99, q508, and s90) were crossed into R o l strains carrying the extrachromosomal array. R o l hermaphrodites, were picked from the F2 and set up individually for scoring. R o l F2s that d i d not produce any dead progeny were presumed to not carry the lethal allele and were not scored.  Unc dpy-5  J  bli-4(classll) bli-4(classll)  5  J  unc-13  _L  _J  Py-  d  WT  I  unc-13  I  H sDp2  dpy-5  N2  X  _L  bli-4(classll)  unc-13  J  1  +  + J_  +  pick wild-type f j s  -L  (may also carry sDp2)  WT dpy-5  bli-4(classll)  Rol unc-13  +  J  _L  c / X  + J  bli-4(e937)  +  I  L  bli-4(e937) I  + _  L  j-- --! /IEX ( i n c l u d e s ro/-6) 1  Bli dpy-5  J +  bli-4(classll)  WT unc-13  _l bli-4(e937)  J  I  +  + _L  L  Rol  Rol dpy-5  J + J  bli-4(classll)  + _l_  unc-13  _L  _J  1 bli-4(e937)  +  bli-4(e937)  J  I  |— -| hEx ( i n c l u d e s rol-6) L  pick R o l  bli-4(e937) I /JEX ( i n c l u d e s  for scoring  rol-6)  Results  SECTION I. The 5' region of the bli-4 gene.  A. Determination of the 5'-most sequence of bli-4  The 5' end of bli-4 was determined by performing R T - P C R on poly A R N A isolated from N 2 worms (see Materials and Methods).  +  The relative  positions of primers used for this task is presented i n Figure 13.  A single  band of expected size (~450bp), amplified by the sense primer derived from SL1 sequence and the bli-4 antisense primer K R p l l was gel-purified, cloned and sequenced. Figure 14.  The sequence of both ends of this product are presented i n  This experiment indicated that bli-4 poly A + R N A is trans-spliced  to the leader sequence SL1 and that the 5' portion of the gene begins w i t h the adenine residue immediately following the SL1 sequence.  This adenine is  not present at the 5'-most position of the c D N A containing the longest 5' end, p C e h l 9 7 .  A parallel experiment performed w i t h an SL2 trans-splice  leader primer failed to amplify bli-4 sequences, as determined by sequencing cloned D N A of slightly smaller than expected size.  Ikb  SLl  SL2  KRplO  SLl  5'-ATAAGAATGCGGCCGCGGTTTAATTACCCATGTTTG-3'  SL2  5'-ATAAGAATGCGGCCGCGGTTTTAACCCAGTTACTCA-3'  KRplO  5'-ACTCTCTTCTTCGGTCGC-3'  KRpll  5-GTGTCCT/rGTTGTTTCCG-3'  Figure 13. A test for trans-spliced leaders S L l or SL2 i n the bli-4 transcript. The primers shown were utilized i n reverse transcriptase-PCR to determine if bli-4 R N A was trans-spliced to leader sequences S L l or SL2. Arrows above (directed right, or 3') indicate sense primers and those below (directed left, or 5') indicate antisense primers. The predicted 5'UTR of bli-4 is shown i n white; the protease domain is i n grey; only those exons (112) common to all bli-4 transcripts are shown.  S L l crcrt t taa t tacccaacrt t tcracraccrttccrct tccraatcga tgtcag CGTTCGCTTCGGGTCGATGTCAG GCAAGCGAAGCCCAGCTACAGTC  aa  taaacgaaaaagaataacccgttgcacccagcgaatcgtcgaacattt  AATAAACGAAAAAGAATAACCCGTTGCACCCAGCGAATCGTCGAACATTT TTATTTGCTTTTTCTTATTGGGCAACGTGGGTCGCTTAGCAGCTTGTAAA tcaa tactcacccatatcagtcacccaaaacggatttattattattatta TCAATACTCACCCATATCAGTCACCCAAAACGGATTTATTATTATTATTA AGTTATGAGTGGGTATAGTCAGTGGGTTTTGCCTAAATAATAATAATAAT  gagcatatcatccaca11 t a  tgctcaaacgcgtgatgcgtatatcgatag  GAGCATATCATCCACATTTATTCTCAAACGCGTGATGCGTATATCGATAG CTCGTATAGTAGGTGTAAATAAGAGTTTGCGCACTACGCATATAGCTATC  gccggataccatggcaaa GCCGGATAGCATGGCAAATTCTGGCAGTTTTAATCGCAGTTGCATTCACT CGGCCTATCGTACCGTTTAAGACCGTCAAAATTAGCGTCAACGTAAGTGA ATTGAACATGATTCCATTTGCGATGAAAGTATAGGTGCCTGTGGGGAACC TAACTTGTACTAAGGTAAACGCTACTTTCATATCCACGGACACCCCTTGG AATACATACCGTAATACGTTTAGCAAAAAGAGATGATGAGCTTGCACGGC TTATGTATGGCATTATGCAAATCGTTTTTCTCTACTACTCGAACGTGCCG  ggcattatgcaaatc????ttctctactactcgaacgt???? GAATAGCTGCTGATCATGACATGCATGTAAAAGGTGATCCGTTTTTGGAT CTTATCGACGACTAGTACTGTACGTACATTTTCCACTAGGCAAAAACCTA  cttatcgacgactagtactgtacgtacattttcgactaggcaaaaaccta  ACTCACTACTTCCTTTATCACTCGGAAACAACAAGGACAC TGAGTGATGAAGGAAATAGTGAGCCTTTGTTGTTCCTGTG  KRp11  tgagtgatgaaggaaatagtgagcctttgttgtt  Figure 14. bli-4 is frans-spliced to S L l (spliced leader 1). The sequence of the -450 bp R T - P C R product amplified from N 2 poly A R N A w i t h primers S L l and K R p l l is shown. Italics indicates sequence obtained from the cloned P C R product and the relative position of the sequence is matched w i t h the c D N A that is most complete at the 5' end, pCehl97. S L l and K R p l l sequences are underlined. The sequencing gap between each end of the cloned P C R product is shown; question marks indicate unreadable sequence. +  B. Mapping mutations in the 5' region of bli-4  The technique of PCR-based heteroduplex analysis was employed to search for polymorphisms i n a subset of class II mutant strains.  Since the  genetics of bli-4 suggests that class II alleles affect the production a n d / o r function of all transcripts encoded by the gene, it was believed that the region common to a l l transcripts w o u l d be a good place to start searching for these mutations.  The following class II lethal alleles were chosen for the analysis:  h670, KISS, hi SI, h384 and h254.  Primers for P C R were constructed to  amplify overlapping fragments of the common region of bli-4 (Figure 15). The P C R products made from heteroallelic worms for the above stated lethal mutations were tested for polymorphisms (see Materials and Methods, Figure 3).  The 3' end of the common region was not tested due to the lack of  primers to amplify a fragment small enough to be useful for mutation detection by this technique.  A summary of the results for PCR-based  heteroduplex detection of mutations is shown i n Table 3.  B.l. hl99 is a missense mutation  A single p o l y m o r p h i s m was detected i n hl99/e937 heterozygotes from the P C R product of primers KRp44 and KRp45 (Figure 16). existence of a base pair alteration, hl99/hl99 template for P C R .  To confirm the  dead embryos were used as  Products from three independent P C R reactions were  cloned and sequenced i n both directions.  A s shown i n Figure 17, all three  clones show an A : T to T : A transversion i n hl99 but not i n e937 or N 2 controls.  Ikb KRp44  KRp29  KRp52  KRp54  KRp36  KRp34  KRp46  KRp6  •Xh  E  KRpll KRplO  KRp45  KRp35 KRp51  KRp37  KRp47  KRp53  A) KRp6  5-CTACTCGGCTACTCCTGC-3'  KRplO  5-ACTCTCTTCTTCGGTCGC-3'  KRpll  5'-GTGTCCTTGTTGTTTCCG-3'  KRp29  5'-TACTCACCCATATCAGTCAC-3'  KRp34  5'-AAACCACTGGATACATTTAC-3'  KRp35  5'-GCAGGAATATCAATTTTCAT-3'  KRp36  5'-CTATAAACCCTTAATTTGTC-3'  KRp37  5'-TGATAAGATTTGTAGAGAAC-3'  KRp44  5'-TGTTGAACGATTGGATTCAC-3'  KRp51  5'-ACCTGGTCGTTTTCAGAAAG-3'  KRp45  5'-TATAGTATAGGTGGTTGACG-3'  KRp52  5'-CATATTGATATATACTCAGC-3'  KRP46  5'-ATGCTACTGGTCAGTTTTCA-3'  KRp53  5'-CTAATAAGAGTTCACCAAAG-3'  KRp47  5'-TAAGATAGGCTTAGGAGGCT-3'  KRp54  5'-CGTTCAAAAGACCTTAATTA-3'  B)  Figure 15. Location and sequence of primers designed to amplify the 5' end of bli-4. Overlapping fragments of the bli-4 common region were amplified by using combinations of sense primers (drawn above the gene structure) w i t h antisense primers (drawn below the gene structure). The predicted 5'UTR of bli-4 is shown i n white; the protease domain is i n grey. A ) The host of primers available for heteroduplex analysis, determination of S L 1 / S L 2 splicing and 3' R A C E experiments. B) Four additional primers were constructed to further divide the protease domain (shown as grey boxes).  Table 3. Summary of PCR-based mismatch detection  allele primers  exons  h670  hl99  hi 91  h384  h254  KRp29 & KRplO  1-3  -  -  -  -  -  KRp44 & KRp45  3-5  -  +  -  -  -  KRp34 & KRp35  5-7  -  -  -  -  -  KRp36 & KRp37  8-10  -  -  -  -  -  Individual worms heteroallelic for each of the alleles presented were tested for polymorphisms v i a the heteroduplex method of mismatch detection (see Materials and Methods). + : polymorphism detected by PCR-based heteroduplex analysis - : P C R fragment appeared as homoduplex control  Ikb  KRp44  "Sac E  Xh  U  I  1  U  E  I  M  |  E  1  L  iBinmni KRp45  A)  B)  KRp44  5-TGTTGAACGATTGGATTCAC-3'  KRp45  5-TATAGTATAGGTGGTTGACG-3'  h791y h384^ ^e937 ^37  e937  y  M99,, h670^ unc-52, ^e937 + ^+ x  Figure 16. A polymorphism detected by heteroduplex analysis in hl99/e937 animals. A polymorphism associated with h!99 was observed upon heteroduplex analysis of the P C R product of KRp44 and KRp45. Four other alleles of bli-4 d i d not exhibit a polymorphism for this same region. The positive control primer set (peg8 and plO) on CB1012/+ (unc-52 heterozygote) was included; N 2 control for KRp44 and KRp45 was included i n this experiment and did not reveal a polymorphism (not shown).  Figure 17. Sequence of K R p 4 4 / K R p 4 5 P C R products from arrested hl99 homozygotes. A ) A summary of sequenced clones from three independent P C R products of arrested hl99 homozygotes. The three clones are listed i n (B). KRp44 and KRp45 sequences are underlined, as is the location of the transversion mutation. B)  i) T w o hl99 clones from the fragment amplified w i t h KRp44 and KRp45 primers using Taq polymerase were cloned and sequenced. Sequence differences found i n these clones, when compared to the e937 and N 2 clones, are bold-faced, ii) One hl99 and one N 2 clone from the fragment amplified w i t h KRp29 and KRp45 primers using Pfu polymerase were cloned and sequenced (see Materials and Methods). The hl99 clone was utilized for construction of a minigene that contains the hl99 A - T transversion, pCeh252 (Figure 11).  KRp44  GTTGAACGATTGGATTCACATCCAGCCGTCGAATGGGTTGAAGA CAACTTGCTAACCTAAGTGTAGGTCGGCAGCTTACCCAACTTCT ACAGCGACCGAAGAAGAGAGTCAAAAGAGATTATATTCTCCTGG TGTCGCTGGCTTCTTCTCTCAGTTTTCTCTAATATAAGAGGACC ATAATgttagttttttagaaatattcaacctacatgatatttat TATTAcaatcaaaaaatctttataagttggatgtactataaata ccagaatcattaccgtttttctttttatcttagctttagagttt ggtcttagtaatggcaaaaagaaaaatagaatcgaaatctcaaa ttcagGATGTTCATCTTCTAACCCCTTCCGCCGTTCGGTTTTGA aagtcCTACAAGTAGAAGATTGGGGAAGGCGGCAAGCCAAAACT ACCGTGATGGTACTCGTAGAGCTCAACGACAGCAGCCACAGTCT TGGCACTACCATGAGCATCTCGAGTTGCTGTCGTCGGTGTCAGA CCAGGAGAAATTCCATCACTTCCATTTCCTGATCCACTTTATAA GGTCCTCTTTAAGGTAGTGAAGGTAAAGGACTAGGTGAAATATT AGACCAGTGGTATTTGgtgagtttcaaattcaaattttctttaa TCTGGTCACCATAAACcactcaaagtttaagtttaaaagaaatt agagaaaaaaaaaccaactgatacatttacagCATGGTGGAGCA tctctttttttttggttgactatgtaaatgtcGTACCACCTCGT GTTGGTGGATATGATAT CAACCACCTATACTATA  KRp45  pCeh240 (hl99)  . . . t t t c a g G A T GTT C A C C T T T C T A A C C C C T T C CGC  pCeh241 (hl99) pCeh242 (e937) pCeh243 (WT)  . . . t t t c a g G A T GTT C A T C T T T C T A A C C C C T T C CGC . . . t t t c a g G A T GTT C A T C A T T C T A A C C C C T T C CGC . . . t t t c a g G A T GTT C A T C A T T C T A A C C C C T T C CGC  pCeh244 (hl99)  . . . t t t c a g G A T GTT C A T C T T T C T A A C C C C T T C CGC  pCeh245 (WT)  . . . t t t c a g G A T GTT C A T C A T T C T A A C C C C T T C CGC  However, one of the hl99 clones also revealed a second alteration i n the same region.  It was suspected that Taq polymerase introduced an error at  this site during the amplification process.  Taq polymerase is k n o w n to have  a significant rate of misencorporation of nucleotides during amplification. Therefore, one reaction was performed on hl99 and N 2 template using Pfu polymerase for P C R ; Pfu has approximately 12 fold higher fidelity than Taq polymerase (Lundberg, et al, 1991). A s shown i n figure 18B, the A : T to T : A change alone is present i n the Pfu amplified hl99, but not the N 2 control.  B.2. hl99 is a weak class II allele  Previous w o r k on intracomplementation between alleles of the bli-4 locus revealed that penetrance of blistering increases w h e n e937 is placed i n heteroallelic combination w i t h class II alleles (Thacker et al, 1995). However, i n performing crosses to acquire heterozygous hermaphrodites for PCR-based heteroduplex analysis, it was observed that hermaphrodite worms rarely blister.  hl99/e937  For instance, when an U n c  hermaphrodite of the genotype dpy-5 bli-4(hl99) unc-13; sDp2 was crossed to e937 homozygous males, the following non-Unc outcross progeny resulted: 58 blistered males, 3 blistered hermaphrodites, 4 non-blistered males, 62 non-blistered hermaphrodites.  These outcross progeny should have been  heteroallelic for hl99 and e937 at the bli-4 locus, unless they carried the sDp2 balancer.  Since the sDp2 balancer w o u l d have been transmitted to some of  the progeny, incomplete penetrance was expected.  However, this does not  explain w h y so few of the hermaphrodites blistered, since sDpl should have been transferred to both gametic lines (Rose, et al, 1984). In order to  determine whether this phenomenon was due to the hl99 allele itself, and not to preferential transmission of sDp2 to hermaphrodite outcross progeny, 3 blistered and 4 non-blistered hermaphrodites from the above cross were set up on i n d i v i d u a l plates and their progeny scored.  A l l seven worms gave  some dead progeny, suggesting that each hermaphrodite set up contained the hl99 lethal allele.  T w o of the non-blistered animals gave many U n c  progeny, w h i c h was evidence for the presence of sDp2; this balancer covers both dpy-5 and bli-4 loci, but not unc-13 (Rose, et al, 1984).  The progeny  from these two animals were not included i n the summary shown i n Table 4. If the hl99/e937 animals i n this experiment always blistered, we w o u l d have expected approximately 1/2 of 791 (=396) plus 85% of the e937 homozygotes (1/4 of 791, or 0.85 X 198) to be blistered (i.e., 564 animals). If the e937/hl99 animals never blistered, we w o u l d have expected only the e937 homozygotes, or approximately 85% of (1/4 of 791) = 169 animals to blister. The data i n Table 4 suggests that hl99 contributes very little to the blistered phenotype i n heteroallelic animals. Males of the genotype hl99/e937 appeared to blister w i t h complete, or near complete penetrance (58/62 males blistered, although some of these non-Bli males may have carried sDp2).  Similarly, e937 homozygous males  were always observed to blister i n an isogenic mating strain even though only about 85% of the hermaphrodites blistered i n the same population.  Table 4. Progeny from 5 hl99/e937 hermaphrodites scored.  Blistered 192  W i l d Type Dead(egg) Dead(larva) 428  150  17  Dpy  Unc  Total  4  0  791  dpy-5 bli-4(hl99) unc-13; sDp2 hermaphrodites were crossed to e937/e937 males, resulting i n non-Unc outcross progeny (Fl) of the genotype: dpy-5 bli-4(hl99) unc-13/+ e937 +, or, dpy-5 bli-4(hl99) unc-13/+ e937 +; sDpl 7 non-Unc F l s (two of w h i c h were also blistered) were plated o n i n d i v i d u a l plates to score the F2 progeny. 2 / 7 F l s were observed to give many U n c F2s. This indicated the presence of sDp2 i n those parent F l s . These two parents' progeny were not included i n the summary.  SECTION II. The 3' region of the bli-4 gene.  A. Sequence determination of pCeh207  pCeh207 is a plasmid clone containing the 3' exons of the bli-4 transcripts, blisterase B, C , and D (Peters, 1992).  This clone was restriction  mapped and subcloned into smaller fragments for d s D N A sequencing (Figure 18).  This sequence information allowed the construction of primers  w i t h i n intronic sequences and also established the genomic arrangement of these 3' exons.  A summary of the regions sequenced is shown i n Figure 18B.  M u c h of the sequencing was performed on one strand since the coding sequence for this region had been previously determined by c D N A analysis (Peters, 1992).  Furthermore, the construction of primer pairs (shown i n  Figure 20) for P C R was rapidly accomplished and employed for other purposes (see below) i n c l u d i n g the confirmation of sequence information for this region.  B. Searching for the s90 lesion in the 3'-specific exons of bli-4  Primer pairs constructed from sequence obtained w i t h pCeh207 subclones were applied to amplify genomic D N A from arrested s90 larvae. A summary of the sequenced regions as w e l l as the overlapping investigative sequencing by C o l i n Thacker is shown i n Figure 20.  The procedure for  cloning and sequencing P C R products for this purpose is described i n Materials and Methods.  To date, the molecular lesion  51  A) The initial status of 3'exons (Peters, 1992). c D N A sequence was k n o w n (boxes) but genomic gaps prevented placement of intron/exon boundaries. Genomic sequence is represented by horizontal lines through the exons, which are numbered below. B) Restriction map and sequencing strategy of pCeh207. The restriction map for pCeh207 was determined. This allowed the placement of genomic sequence gaps (approximate size of each given) and provided information to produce subclones of pCeh207 for d s D N A sequencing. Each arrow represents direction and length of sequence obtained from subclones (lighter arrow represents sequence obtained from C o l i n Thacker). The polylinker of the plasmid is shown as an angled line at each end. C) The complete genomic arrangement of 3' exons i n pCeh207. Exon 18 was placed as a result of the identification of a fourth blisterase transcript (Thacker et al, 1995).  Ikb  KRp42  KRp38  E E  KRp40  K  KRp73  K  t  KRp43  KRp39  KRp72  KRp42  5'-ATTTCGACGTGCAATATCAG-3'  KRp43  5'-ACCTGTTATAAAGGTCAACT-3'  KRp38  5'-GTGGTGTCTATACCAGTTAC-3'  KRp39  5'-ATTTCCGCGAGCCGAAAAAC-3'  KRp40  5'-TGAAATTCTTGGCCTCTAAC-3'  KRp73  5'-TGCTAGTTATCGTCTCCGTA-3'  KRp72  5'-CTTGTAGTTAGGTTTAGGTCT-3'  KRp41  Figure 19. Location of primers designed to amplify the blisterase B, C , and D 3' exons. P C R fragments of the blisterase B, C , and D specific 3' exons were amplified using combinations of sense primers (drawn above the gene structure) w i t h antisense primers (drawn below the gene structure). KRp39 anneals within exon 18 but was designed previous to the discovery of the transcript to which this exon belongs (Thacker, et al, 1995). The predicted 3'UTRs for blisterases B, C , and D are shown i n white.  Figure 20. Sequencing of 3' exons i n s90 homozygotes. The entire sequence of pCeh207 plus sequence extending 5' of pCeh207, to include exon 14 is shown. P C R amplified fragments from s90 arrested larvae that were cloned and sequenced are shown w i t h underlines. Single underlines represent Pfu polymerase-based P C R reactions, performed by C o l i n Thacker; double underlines represent Taq polymerase-based P C R reactions and repeated Pfu polymerase-based P C R reactions. N o sequence differences between s90 and N 2 have been detected i n the regions tested. Lower case indicates introns, upper case indicates exons. The exon positions are as follows: exon 14 (139-425), exon 15 (721-793), exon 16 (1646-1824), exon 17 (1875-1983), exon 18 (2076-2209), exon 19 (2608-2835), exon 20 (2889-3092), and exon 21 (3143-3476).  54  tttttagctaatttttttcttatgtttctataagttttactgaaaagatg  50  aaatttcaacatacaatatcaaaatccctcrcaaacttatttttatttaca  100  attaaaattatttcatatttcaattataaGGTGATGAAGTTGTTGAAAGA  150  ATTCGAAATCATTGGGAAGTGACATTAGAAGAGAGTTCACATTGGAATTG  200  GGAGCATGCTCGTGAACATAAATCATTACAAGAATTGAACTCTTCTTCTC  250  GTACCCATAGTTTTTTATACTCTTTCACCAAATTTCAACCGATTTTCTTG  300  ATTATTCTTGTCTGTATTTTTGATGCCATTCATCGCCAATTCGCGGTTTG  350  AGACTATATGAATTCATTTTGGGTActaaaaatacatatttcaattcrtat  400  tcccatcaaattttcaaaatctcatcacaatctaccataatcccaatatt  450  gtgcctatttatgcttttatggttgtacacgtcgtggtgtgtgtgtgtgt  500  ctatttgtacttattgtcaattatctcaaatatttagttttgtttcttgc  550  tttccqtcatcagaacaaatttaatccaatttcttattttaaacaatatt  600  taaaaaaaaataataaaaaataataaaatacacaaacacacaacacacca  650  cattatccatccaaaacttttttaaatttacgaaacacaataaatattac  700  aataaatcattattttacaaGTTGAAGAGTCTGCTCCGATCTCATTTCCA  750  GATTTGACGTCGGCTGGAAATTGTCATGATGAATGTAATGGAGataaaaa  800  attaattttcacaattactataaatatttaaaaaatattttcaatttttt  850  tcaaaaatttcaaaaacaaaaattaaacattacaaaaattacaaaatcaa  900  ctaaaaatattatccaaatacaaaattttacaactattactacaaaaata  950  cggtacccggtctcgacacggccgattttttcaatgcaaaagaatgcgcg  1000  cctttaaggaatactgtagtttccacattttccgcgttgctcgattttct  1050  ctgattttcaaagctttctcatgttattcaaaaatcattgaagaattgtg  1100  ctgaagcagtttatcttactttaaaatacataaaatgttaagaaaactaa  1150  ggaaaatctataaaaaattgagctacagtacttgttttgcagtgcgttca  1200  gtggtattttatataagttaaactaacagtacacttggtgactctacaag  1250  gatttattatagacaactagtaacactaatagacaagtaacgcaagaaca  1300  taaagcaagaactgattcgggaagtcctccaaggttggcaatatggtcct  1350  tgttctttgttagcactgactgttggttgttccttgagcgatggtgatag  1400  aaatcgtcggtgttgttagcgtccatgtagagtacataattatgcaataa  1450  tatagaataattattaagtcgtgaataatggagatcctgctccatgctat  1500  tcgcgacagtactcctcaatgtaaaactttttcgcgttaaaaaattggtg  1550  gtgtcgataccagttactgtatttttgaattaaaatcgaaaaaaaaatgc  1600  acatatgagacatatttctgttaataaatttttgaatttttacagGTTGC  1650  ACAGAATCGAGCTCAGCAACATCATGTTTCGCTTGCAAACATTTAACTCA  1700  AACTCTTCGAAATAAAGGAGGAAGTGGATTCAAGTGTGTTCAAAAATGCG  1750  ATGATACTTACTACTTGGATGGTGACAAGTGTAAAATGTGCTCATCCCAT  1800  55  TGTCATACTTGTACGAAAGCTGAGqtatttqcrtqacataattqtttgcca  1850  aaaaaaataattctataattttaqGTGTGTGAAACGTGTCCTGGAAGTCT  1900  TTTGCTCATTGATGTGGATAATATGCCACACTATGATCATGGAAAGTGTG  1950  TTGAGTCATGTCCTCCAGGATTGGTTGCTGATTqtaaqttttataaqaqt  2000  ttcaqaataaaaaqttaatctccaqACGCTGAAAATTTCGATTTTTGCGC  2050  CAAAAATAACGAATCTGGTCGCGACACGACAGTGTTTGTTAAATTCAAAA  2100  AGCCGAGCGCCTTTAAAGATTACTGTAATTTGAAACTTGTTGATTTAGAG  2150  TTTTTTATTTTTTCTCTCTTATTTTAGTTTTTTTTTTGCTAATAAATTTT  2200  TTTTTATCGaaaaactttatttcgatttaaatctatcgatttaatttatc  2250  gattTaaaaaaaatccgcatcgatggaagtttgaaattacagtactcttt  2300  aaatgctctcagcttttcgaatttgacaaaaattgtcgtaggccgggtac  2350  cgtatttcttttatgaaaattgccaaaattcgccgctctggaaatagtga  2400  tcttgaagaagaaaaaagttacagcagaaaaatcaaaggaagaaaaactt  2450  ggtattgtgtgtgaatgtttgagtgctttcttttttcaggctaaatccta  2500  cttctacactgaaactttcctcttagaatatttacatctgaaattcttgg  2550  cctctaaccattttcctcatgttacaagttttttgtttaaacaatgtgaa  2600  atttcaqATGAATCGAATCTTGTTCAAGCTAAATGTATCTGGAGAAAAGA  2650  TCTTTGTGGTGACGGATATTACATCAACGCTGTTGGAAAATGTGATCTTT  2700  GCGACTCCTCATGCGAAACTTGCACCGCTCCTGGTCCAATGAGCTGTGAG  2750  AAATGTTCAAAAGGTTATGGTAAAGGATCGATTGGATACTGTAGACCGTG  2800  TTGTCCTGAAGGATCAACAAAGAGTTGGCAATGTGqt a a q q a a q a ca c t c  2850  tcaaaatcttccagqtattctcacaattattttttcaqAGGACTGCTCCA  2900  AGCCGGATCCTACACTTTTGATTGATTCTAATAAATCATCTGGATTTGGA  3000  TTGATGTTCTGGATTGTAGTTAGTTTGATTGCGGCTTGTGGAATCTGTGC  3050  CTGTAAAAAGTGTGCAAGTGAGACGAAAAGCTCAAACGTAGAqtaaqcct  3100  tqctaqttatcqtctccgtaattcqaattttaatatttttaqATATGCGC  3150  CGCTTGCTCAATATAATGCCACAAATGGTGCTATCAATTTAGGAGCACAC  3200  ACTGACGATGAAGACGATGATGAGGATGAAGTATTTGTGAACCCTCAAAT  3250  TGTTTAAACCAAACCTTTCAAAATTCATGTTTTTAATTGTAATTTTTCTG  3300  CCAACTTCTTTGTGTATGCTTCTGAAATCCGGGTAATTGTTTCTTTTTCA  3350  AAATTTTAAAACCTGAAACGTTTTCAAGTGGCCTTTTATCATGTGATTGT  3400  ATTGTTTCTTTTGTCTTATACCGGGTTTATTTTATTATACCGTACTGTTT  3450  CCATGTTATAAATAGATTGTTGAATTaa11ac ac aaqacaaaac a t c qac  3500  atattaaqactatacaaatqaqttccqacaqtaqacacctaaqccqactc  3550  qattcacttctqtqqcttcaqctcqqaaqtttcqaacttctqcacqcqat  3600  ggtttaggcacttctggtgactattgataatatcatctagtgtaggaaag  3650  responsible for the s90 mutant phenotype has not been discovered. However, the search has not yet included the exons common to all isoforms.  SECTION III. The structure of the bli-4 gene.  The results obtained for the 5' and 3' regions of bli-4 allowed the construction of the molecular structure map of the gene shown i n Figure 21. This schematic representation also serves as an indicator of the progress made towards characterizing the bli-4 gene during the course of this study by comparison to the former structure status depicted i n Figure 2.  The high  resolution map for the bli-4 gene enabled the construction of subclones encoding subsets of blisterase isoforms, as discussed below.  SECTION IV. Transformation rescue experiments with subsets of bli-4 coding information.  In order to test for rescue by subclones of bli-4, it was necessary to first establish the 5' promoter region of the gene.  A l t h o u g h defining the realm of  the promoter is not required at this level of analysis, it is important to include a sufficient amount of 5' sequences to ensure expression of the exogenous gene.  This was accomplished by fusing the putative promoter  region of bli-4 and a small amount of bli-4 coding sequences w i t h the lacZ coding sequences included i n the expression vector pPD21.28 (Fire, et al,  ~. cr 3 or oo  2  K2  ^  Hi  H  3fD  Crq  3 i-i  i-i  w S £ o 3&| cr n>  re rt>  CD <n f H CO 1  0) w  pj  ^  fD  3^ o  1  3  fD n  w  co  6"  fD  CO C O  rf O  8fD HI  3  '"I  1-1 03  0> CO  3  n  n £u fD i-i CO fD  ^ EL fD ju co cr  3- o g < <! fD  § ? 03  fD  i-i  <-t-  03 03  1  Cu  I—1  fD 03 w  -?. n  3  h-i  m° 8 ^  CD  fD  r-h fD cn fD  O  cd  C/)  r-tfD i-i  »J  CD  fD  r  1  3^ oT  3  Hi  c o CO  Hi  CO  a. 3 3 n  XT  H|  O  fD  Y*  H  3  fD  O  H3  x o  rr> g co ^ ^ r  u  Hj  3  o 2 3 3 i-i ™ 3' O 1-1  o r-t3 H!  i-r »-! 03  3  CO  n  i-i  »  tr  03 <-tH-l 03 Hi CO* Hi  fD  fD  <  Hj ^ a H-. O fD  03'  i-i  fD  s'  P S wo CJI  3  3 Xi CO  ^ 3  J  3 9 •6' CO  fD  CO  CO 03  O fD >-* n CO rtfD fD tT 3 O 3  3 ^  O O fD  •  £• H § s fx n  5: fD ff S5.  fD  r/i  <-h fD t-j  03  CO CO  fD  cd  3 Cd P- fD 3 o o o n co fD n co fD fD  fD fD T3 r+ O  3" a S Sf  fD n  cd  fD fD OQ r-t- fD  3? co  r-t-  fD <3-  ft)  cd CD  r<  w 2 "3 2. 3 2  3  fD  nl  3  *  Crq fD  S - 3fD OQ  On 3; o .. r 1 fD < p n 3ft) 3 Ht r+ fD  3^  fD  3*  03 <  nT - fD 3- ^ cr fD 2 to fD 3 3 3 fD r-tr-t- ^T* 1-1 03 3 i-i  H  fD  I  cd  fD  "oT 3/ Sen  co  H-  T3  2 3t 3 5'  3  el-  e-  -\Ti ->-d  i-l  7T  fD  3  r-th-»•  C/3 -P)  1990).  Sequence extending from the Xbal site approximately 5 kb upstream of  bli-4 exon 1, to the CM site i n exon 2 was found to express (3-galactosidase ((3gal) i n hypodermal cells, the ventral nerve cords, and v u l v a l cells (Thacker, et al, 1995). Expression was first observed at the two-fold stage of embryonic development, the stage at which most class II lethal homozygotes arrest.  A  second lacZ fusion was made w i t h 5' putative promoter sequences starting at the Xhol site just 700 bp upstream of exon 1 to the Clal site i n exon 2. construct d i d not express P-gal i n any tissue.  This  Therefore, the Xbal (5 kb) 5'  upstream sequence was included i n all minigenes used i n this study. Because this is a fairly large amount of sequence, (certainly many C. elegans genes are 5 kb or smaller) it was necessary to ensure that another gene i n this region was not responsible for rescue of bli-4 phenotypes.  pCeh238 d i d not  rescue the lethal mutants q508, hl99, or s90, and pCeh239 d i d not rescue blistering (see below). Both of these constructs carry the identical upstream promoter sequences.  This argues against another gene existing i n this  upstream region that is responsible for the rescues (see below). Minigenes derived from coding sequences of bli-4 were tested for their capacity to rescue each class of mutant bli-4 phenotype.  The alleles e937  (class I), q508 and hl99 (class II) and s90 (class III) were tested against each minigene constructed, as summarized i n Table 5.  For simplicity, e937  homozygous hermaphrodites were injected w i t h the subclones (see Materials and Methods) and rescue of blistering was assessed from stable lines.  The  arrays were then crossed into lethal bearing strains mentioned above (Figure 12).  A l l minigenes except for pCeh238, and pCeh239 rescued all three allelic  classes of bli-4 mutations; this data is summarized i n Figure 22.  The  dominant marker, rol-6(sul006), d i d not rescue lethality (since this marker was present i n arrays that do not rescue) and d i d not significantly contribute  to reduction of blistering when present alone (Table 7).  Rescued  homozygotes for each lethal allele were maintained and given strain designations (Table 6). A l l stable rescued lines gave some dead progeny due to less than 100% transmission of the extrachromosomal array.  After many  generations, rescued lines flanked by markers epistatic to rol-6 (e.g. dpy-5) still gave R o l progeny w h e n outcrossed to N 2 males, suggesting selective maintenance of the extrachromosomal array i n these rescued lines, as expected.  pCeh221 and pCeh230 were observed to poorly rescue both hl99  and s90, as evidenced by very few progeny recovered in each generation. Death of most animals was approximately late L 2 - early L 3 larval stage for each of the strains.  Since this is later than the typical arrest stage for these  two alleles alone, it is likely that bli-4 activity is reduced, but not absent, i n these animals.  Due to their severely reduced viability, these strains were lost  before a population could be archived at -70°C.  A. Subclones encoding blisterase A rescued blistering and lethality  Subclones pCeh226, pCeh229, and pCeh230 were injected into CB937 (e937/e937) hermaphrodites.  Stable transgenics were obtained for each of the  constructs injected (Table 6).  R o l animals were never observed to blister in  any of the transgenic strains.  U p o n outcrossing to the lethal (q508, hl99, s90)  bearing strains, each construct was found to rescue both classes of bli-4 lethal alleles, however, as mentioned previously, pCeh230 transgenic lines had reduced viability and were lost for hl99 and s90 alleles. A summary of rescue experiments is provided i n Table 5 and Figure 22.  Table 5. Summary of rescue results for subclones of bli-4 .  ClassI  ClassII  ClassIII  e937  q508  h!99  pCeh226  +  +  +  +  pCeh229  +  +  +  +  pCeh230  +  +  +  +  pCeh238  +/-  -  -  -  pCeh236  +  +  +  +  pCeh221  +  +  pCeh239  -  n.d.  a  a  +  a  n.d.  n.d.  animals survived but exhibited very poor viability and fecundity; strains were lost as a result. a  +/- : frequency of blistering is reduced, but not eliminated, n.d. : not done  Table 6. Strains constructed for transgenic rescue.  Strain KR2859 KR2860 KR2861 BA5hl99 KR2862 KR2863 KR2864 SalRLhl99 KR2865 KR3028 KR2868 KR2869 KR2870 KR2871 KR2998 KR2999 KR3001 KR3000 KR3006 KR3014 KR3003 KR3002  Method injection cross cross cross injection cross cross cross injection cross injection injection injection injection injection cross cross cross injection cross injection injection  Genotype bli-4(e937);hEx54 dpy-5(e61) bli-4(q508);hEx54 dpy-5(e61) bli-4(s90) unc-13(e450);hEx54 dpy-5(e61) bli-4(hl99) unc-13(e450);hEx54 bli-4(e937);hEx55 dpy-5(e61) bli-4(q508);hEx55 dpy-5(e61) bli-4(s90) unc-13(e450);hEx55 dpy-5(e61) bli-4(hl99) unc-13(e450);hEx55 Mi-4(e937);hEx56 dpy-5(e61) bli-4(q508);hEx56 bli-4(e937);hEx46 bli-4(e937);hEx47 bli-4(e937);hEx48 Mi-4(e937);hEx49 bli-4(e937);hEx58 dpy-5(e61) bli-4(q508);hEx58 dpy-5(e61) bli-4(s90) unc-13(e450);hEx58 dpy-5(e61) bli-4(hl99) unc-13(e450);hEx58 bli-4(e937);hEx59 dpy-5(e61) bli-4(q508);hEx59 bli-4(e937);hEx62 N2;hEx61  Plasmid pCeh226 Ceh226 Ceh226 Ceh226 pCeh229 pCeh229 pCeh229 pCeh229 pCeh230 pCeh230 pCeh238 pCeh238 pCeh238 pCeh238 pCeh236 pCeh236 pCeh236 pCeh236 pCeh221 pCeh221 pCehl81 pCeh239 P  P  P  B. A subclone encoding blisterase B partially rescued blistering, but not lethality  The minigene pCeh238 was injected into CB937 (e937/e937) animals. Four stable R o l lines were established and each was assayed for penetrance of blistering.  pCeh238 was found to only partially rescue blistering as an  extrachromosomal array; i n each of the four lines, some R o l progeny d i d blister.  After outcrossing pCeh238 to lethals q508, hl99, and s90, no rescue  was evident, indicating that blisterase B was unable to supply sufficient function to rescue lethality but it d i d reduce the penetrance of blistering. Table 7 presents a summary of the results obtained for the partial rescue of blistering by pCeh238.  C. A subclone encoding blisterases B, C, and D rescued blistering and lethality  pCeh236 was injected into CB937 (e937/e937) animals and a stable line was recovered.  R o l animals from this strain were never observed to blister,  indicating complete rescue of blistering by pCeh236. found to rescue q508, hl99, and s90.  This construct was also  A summary of rescue results for  pCeh236 is shown i n Table 5 and Figure 22.  D. Transmission frequency of rescuing subclones verses proportion of rescued animals.  In order to better assess the rescuing capacity of a given isoform, data for every rescue experiment was compiled. It was suspected, for instance, that the reason for loss of some of the transgenic lines that carried pCeh230  Table 7. Partial rescue of blistering by pCeh238  DNA injected in CB937  Stable line  % Blistering of Rollers  BILRol / total Rol  rol-6  KR2872  79  110/139  pCeh238 + rol-6  KR2868  43  79/185  pCeh238 + rol-6  KR2870  17  17/101  pCeh238 + rol-6  KR2869  8  41/523  pCeh238 + rol-6  KR2871  3  7/206  pCeh226 + rol-6  KR2859  0  0 / >1000  Four independent lines were recovered from a single round of injection experiments. A few animals from each line were picked and their progeny scored for penetrance of blistering among R o l progeny (see Materials and Methods), rol-6 marker was introduced w i t h pRF4 (Mello, et al, 1991).  and pCeh221 was due to very weak rescuing capacity by these constructs.  The  frequency of R o l animals was calculated for each line, as was the percentage of rescued animals (evidenced by the flanking marker(s) phenotype) from the total number of homozygotes for the lethal allele.  A comparison between  the frequency of transmission of the array [ R o l / ( W T + Rol)] and frequency of rescue [Rescued/(Rescued + dead)] provides an indication of the rescuing ability of a given array.  Data from transgenic rescue experiments is  summarized i n Table 8 and a graphical representation of relative rescuing ability of each construct is presented i n Figure 23.  65  o> u O N  o  T-j O  00  tn H oi  O m  oo O oo xl— i -n- > •/>  O m  c i CN c i v o r N m H - ol  o- xt" ^ 0O •* m  * o nci mm \o m m  T>  ol  O  -tf  in  <T)  0)  H m rH O0 OO CO  in  C/3  o §  i K  3 w c g "C o> X  o > 01 u in  -4-1  00 w c cS  H  Ol  60 </>  « 5 )H Mo H  o  r»  O  g.  oo  M S 9 0 >§° o 2  < N^  ol m ^ « \o ^ u oo m in *3- o l C \ i n o o  r- o OO CI"3o O Ol  —Ln  o> "al -a a Ol -B  >n  ^ 2 °  in  oo  ol  ol Ol O O  T3 T 3  ^  >  g > be . a  - L to  oo C-  o -=t  ^ C H  r~ m  V CIO CO I ^Ol o  OO  oo^  Ci § 2  O  2  • n  r -  o o N oo c  J!, i n i n ^  rt  m o l o-tl  3  m  o\ If) m  H  ,—i rt  O)  + bO  ^  QJ  s ° Q J. S CC M-lo c0> 1H CU + CD > a, w 8 + o Q CD CO  C  CO  T3  • § - § > b0 >•  ^  O  CU "-i-i  S  Ci Ci  •v  \  C i ) co  No l _— — -1 f l "1 UN\  » "£C <« co C U S § -si .2 3 g5  u C  Ol  O  Ci  —  ol  2  O  <N  B-i C C  T3  01  o u  O  o  S  0 0  N  CO  vo *  0 0  O  O  ol  xf  in  (S  in  n  m  -t> —sPL,  Ol  b O O  QJ  o  m tt in s o i n r - m [r H  N  *  ^  ci <N  vo o\ vo o o\ ol m ol -  -  T 3 CC QJ  O  ol N * h> n -o\< to\ CNmh o S l  +  CO CM  s^.  bO^H  b0 bO  QJ 1-4  0 0  I COtPIs O  oo  Ol > no o o r- O s r - c i •n  o\ O SO fN fN tn fN <T> S©  fN fN fN fN fN J= J= J3 Ol 0) Ol Ol 0) U U U U U a  a  a  a  a  as  CN CN Ol Oo o o O  <N  NO o \  O  l—1  SO  a  a  a  oo  00 o  O o t- V oC N C N CI  SO ON o  l-H SO  fN fN m fN m c ^ fN fN fN in fN fN fN fN fN fN fN fN fN fN jiJ= J= J3 J= J3 J= J= J3 Ol Ol Ol Ol Ol Ol Ol Ol Ol Ol U U U U U C^U U U U U a  a  a  a  a  a  a  OJ cC  OH  C  ,  QJ  ,  +  ° >;  QJ co  >-l COQ 2 -x a, * > ^ co i> Q J CQ OJ — o 2 CO S 0  Figure 22. Summary of results for transgenic rescue using subclones of bli-4.  A ) The structure of bli-4 (see Figure 21 for description)  B) The bli-4 minigenes, pCeh226, pCeh229, pCeh230, pCeh238, pCeh236, pCeh221 and pCeh239 were tested for rescue of blistering (Bli) by injecting into e937 homozygous hermaphrodites. Each of the constructs was then crossed into hl99, q508, and s90 strains (see Materials and Methods) to test for rescue of lethality (Let). T h i n lines indicate genomic sequence gaps i n plasmid constructs, compared to endogenous sequence. Construction details are provided i n Figures 6 - 1 1 . /  strains for hl99 and s90 were not able to be maintained w i t h these constructs +/- frequency of blistering is reduced but not eliminated n d not done a  67  Figure 23. The relative rescuing ability of different subclones of bli-4.  The subclones that rescued the lethal alleles q508, hl99, and s90 were graphed according to % animals that R o l ( R o l / R o l + Wild-type) and % animals rescued ( e.g. D p y U n c / D p y U n c + dead animals) shown along left margin. Data derived from Table 8. The Blistered phenotype was not scored, any Bli progeny were scored as wild-type; no BliRol were observed. Error bars represent 95% confidence interval calculated by the formula: 100 X p + 1 . 9 6 { p ( l - p ) / n } where p = %Rol/100 or %Rescue/100 for each strain and n = total number of animals used to calculate % R o l or %Rescue respectively. 1/2  ;  Discussion  bli-4 is rrans-spliced to SLl  Previous examination of bli-4 c D N A s predicted that the 5' end of the transcript (exons 1 to 12) is common to all four isoforms (Peters, 1992; Thacker, et al, 1995).  The c D N A w i t h the most complete 5' sequence  identified (pCehl97), contains 176 nucleotides of untranslated sequence (5'UTR), suggesting that this c D N A represents a complete, or nearly complete message at the 5' end (Peters, 1992).  However, the 5'-most sequence of the  processed transcript cannot be confirmed on this information alone.  In  order to determine the 5' end of the bli-4 R N A , an experiment was conducted to determine if the R N A is rrans-spliced to the leader sequences S L l or SL2. Approximately seventy-percent of C. elegans transcripts are transspliced to leader sequences S L l or SL2 (Spieth, et al, 1993).  A l t h o u g h the  molecular function of such leader sequences is not k n o w n , a deletion that removes many copies of the S L l gene results i n embryonic lethality, suggesting an essential component of at least some genes' expression (Kimberly Ferguson, Paul H e i d , and Joel Rothman, pers. comm.).  A benefit  of the rrans-splicing phenomenon to C. elegans researchers is that it enables rapid determination of the 5' most sequence of processed transcripts for genes frans-spliced to either of these leader sequences.  The general procedure to  detect trans-splicing to either S L l or SL2 leaders utilizes R T - P C R on poly A + R N A w i t h primers that are situated i n the leader sequence and w i t h i n the 5' portion of the gene of interest.  This approach led to the discovery that bli-4  is trans-spliced to S L l and established the 5'-most sequence of the processed bli-4 transcripts, w h i c h was one nucleotide longer than p C e h l 9 7 (Figure 14). Although it is possible that not all transcripts of bli-4 receive the S L l leader, transcript-specific differential trans-splicing has never been reported for other genes whose products undergo alternative splicing.  R T - P C R experiments to  test this possibility w o u l d necessitate the use of antisense primers located i n the distant 3' specific exons for each of the isoforms.  This w o u l d involve  amplification of sequences approximately 3 kb i n length, or longer. Considering the relatively high level of background amplification often observed w i t h the S L l and SL2 primers, this task may not be particularly easy.  Mapping mutations in the 5' region of bli-4  In this thesis, the technique of PCR-based heteroduplex detection was applied to five class II alleles of bli-4 to search for the molecular lesions responsible for each mutation.  Based on intracomplementation analysis of  bli-4, class II lethal alleles are expected to interfere w i t h the expression a n d / o r function of all isoforms.  Therefore, the initial search for class II mutations  began i n exons 1 to 12, the coding information that all isoforms share. Heteroduplex technique was employed for this task.  By using MDE™, an  acrylamide-based matrix through w h i c h D N A fragments are electrophoresed, one can rapidly detect single base pair alterations w i t h relatively high resolution (Keen, et al, 1991).  This is an important consideration, since  most of the alleles of bli-4 were induced w i t h E M S , w h i c h is k n o w n to preferentially cause single base pair alterations, most of w h i c h are G : C to A : T transitions.  Also, bli-4 spans a relatively large genomic region  (approximately 15 kb); considerable time and expense w o u l d be required to sequence the entire region for each mutation.  According to the  manufacturers, the resolving power of the heteroduplex technique w i t h the MDE™ product is dependent on three main variables, the size of the D N A fragment (mismatch detection has been observed for a 900 bp fragment, but 400 bp is optimal), the location of the mismatch i n the D N A fragment (a mismatch i n the centre of the fragment is optimal), and the type of mismatch (a G : G mismatch is optimal).  Based on the size criterion, exons 11 and 12  were excluded from the analysis, since the only primers available for this region amplify a product of 1011 bp. P C R primer pairs (KRp34/KRp35 and K R p 3 6 / K R p 3 7 ) amplify the protease domain of bli-4 i n two overlapping products of sizes 816 bp and 755 bp (Figure 16).  Both products were found to smear slightly i n the gel matrix.  N o polymorphisms were apparent i n any of the class II mutants analyzed i n the protease domain.  However, smearing of the D N A bands, combined  w i t h the relatively large product size, reduces the confidence of interpreting these results as evidence for the absence of mutations.  Therefore, four  additional primers that further divide this region into smaller products were designed for future analysis (Figure 15B). These primers have been tested and found to w o r k w e l l w i t h the previous four, but have not been applied to search for mutations at this time.  hl99  The hl99 mutation was induced w i t h E M S and recovered i n a screen to identify lethal alleles balanced by the free duplication sDp2 (Howell, 1989). Complementation analysis revealed that hl99 fails to complement the blistered phenotype of bli-4(e937), indicating that it is an allele of the bli-4 gene.  Subsequent mapping data obtained by H o w e l l (1989) and Peters et al,  (1991) suggested that the original hl99 mutation (in strain KR513) was linked to a second lethal mutation towards the left end of L G I , near unc-40. However, recent mapping of hl99 i n KR513 indicated that the second mutation no longer exists (mapping data obtained from R. Johnsen presented in A p p e n d i x 2).  Furthermore, subclones of bli-4 rescued the hl99 mutation  (Table 8, and Figure 22), w h i c h w o u l d not be expected if a second, linked lethal mutation was present i n the strain.  A new strain designation  (KR2997) has been assigned to distinguish this strain from KR513. I have mapped the hl99 lesion by heteroduplex mismatch detection and have sequenced the region of D N A containing the molecular lesion from hl99 homozygotes.  A n A : T to T A transversion was found i n three  independent hl99 clones from P C R products of exons 3 to 5 (Figure 17). A l t h o u g h E M S most often induces G : C to A : T transitions (Griffiths, et al, 1993), A : T to T A transversions resulting from E M S exposure have been reported i n C. elegans (Perry, et al, 1994). The transversion mutation results i n an amino acid substitution of H i s l 2 7 (CAT) to L e u (CTT).  This alteration is i n the amino terminus of the  common region of predicted bli-4 isoforms, proximal to the protease domain. The function of this region of the predicted bli-4 proteins is u n k n o w n and not w e l l conserved among proprotein convertase family members; h!99 may  reveal a previously unrecognized functional component specific to bli-4 convertases.  However, one cannot rule out the possibility that a common  function exists for this region among all family members, i n w h i c h case this mutation may provide novel insights into the function of this region i n other proprotein convertases.  hl99/e937 hermaphrodites rarely blister  Throughout the course of this work, it has been observed that e937/q508 and e937/h670 worms blister w i t h 100% penetrance and heightened expressivity, w h e n compared to e937/e937 worms. non-Unc progeny from the parent unc-63 bli-4(H670)  For instance,  unc-13/let(h661)  bli-4(e937); hT2 were always blistered; no wild-type progeny were ever observed.  hT2 is a translocation between L G I and LGIII ( M c K i m , et ah, 1992)  that carries the e937 mutation and has been used to identify class II alleles by precomplementation screens (Peters, et ah, 1991; Thacker, Srayko, Rose, unpublished).  e937 translocation homozygotes do not survive i n this strain  due to the existence of a linked lethal mutation (h661).  Therefore, any  blistered animal i n the above strain is a result of e937/h670.  The absence of  wild-type animals indicates 100% penetrance of the blistered phenotype when e937 is i n heteroallelic combination w i t h class II alleles such as h.670. q508/e937, h670/e937, and hl010/e937  animals have been observed to blister  w i t h complete penetrance when obtained from crossing e937 males to hermaphrodites that carry the lethal class II alleles (C. Thacker, M . Srayko, unpublished observations).  A l s o noted w i t h these strains was that blistering  was very severe i n heteroallelic animals; for instance, a large blister usually  enveloped the head soon after the fourth larval molt, rapidly extending d o w n the length of the w o r m , resulting in reduced viability.  Although a  rigorous quantification of this phenomenon was not performed, h42, h791, h384, and h254, also appeared to behave as q508, h670, and hlOW w h e n i n heteroallelic combination w i t h e937. A s presented i n this thesis, hl99/e937 hermaphrodites rarely blistered (Table 4).  This data suggests that the basis  for lethality i n hl99 may be somewhat different than the other class II alleles. Class II alleles are thought to affect the expression a n d / o r function of all products produced by the bli-4 gene; consistent w i t h this hypothesis, three class II alleles have been mapped to the region shared by all transcripts of bli4 (Figure 2).  Based on Northern and R T - P C R analysis, e937 abrogates  expression of blisterase A , but does not eliminate the expression of blisterases B, C , and D (Thacker, et al., 1995).  One possible explanation for complete  penetrance of blistering i n e937/class II worms is that there is one less copy of a gene capable of producing blisterase B, C , and D i n e937/class II worms than in e937/e937 worms.  If slight functional redundancy exists between some of  the isoforms of this gene, one might predict an overall reduction i n activity of the bli-4 gene i n e937/class II animals, when compared to e937/e937 animals.  B y similar reasoning, it appears that hl99 does not reduce the  activity of bli-4 as much as these aforementioned class II alleles because e937/hl99 worms blister less often than e937/e937 worms.  H o w e v e r , this  weak intracomplementing ability is not paralleled by a weakness i n lethality; hl99 homozygotes arrest w i t h the same phenotype as other class II alleles examined and the penetrance of lethality appears complete (data not shown). . It is evident from the intracomplementation data that hl99 does not behave like these other class II alleles. The fact that hl99 appears to be a somewhat weaker class II allele is consistent w i t h the nature of the lesion  responsible for this mutation.  A missense mutation could give rise to a  protein w i t h partial function.  Perhaps the hl99 mutation affects the  function of the predicted essential bli-4 product(s) more severely than the predicted non-essential bli-4 product(s) implicated i n the adult cuticular function.  If true, this w o u l d be a particularly informative mutation,  considering it is located i n a region common to all predicted BLI-4 isoforms. A further complication is that males of the genotype hl99/e937 do seem to blister w i t h complete, or near complete penetrance (see Results, p49). However, it is possible that the male morphology is more susceptible to blistering and therefore, responsible for the differing penetrance of blistering between the two sexes of hl99/e937 animals.  For instance, e937 homozygous  males are always observed to blister in an isogenic mating strain even though only about 85% of the hermaphrodites blister i n the same population (C. Thacker, M . Srayko; unpublished observations). This argues against an extragenic basis for the difference, such as an X-linked suppressor of blistering.  Another possible explanation is that bli-4 expression is different  between males and hermaphrodites;  a sex-specific Northern blot could be  performed to test for this. pCeh250 is a minigene that contains the l\199 mutation and encodes only the blisterase A isoform (similar to pCeh229).  This minigene may be  used i n the future to assay the rescuing capability of a blisterase A isoform that carries the missense mutation.  This experiment w i l l provide  information about the effect of the mutation on the function of the predicted proteins.  For instance, pCeh250 may rescue blistering completely, since hl99  seems to not affect this function as much as other class II alleles i n II animals.  e937/class  However, one w o u l d not expect pCeh250 to rescue lethality since  there is no evidence that hl99 is weak w i t h respect to that phenotype.  In  addition, an epitope tag could be added to the hl99 exogeneous construct and a transgenic animal producing this flagged protein could be used i n immunolocalization studies to determine if the mutation affects intracellular trafficking, such as exit from the endoplasmic reticulum.  It has  been observed, for example, that pro-furin containing a defective kex2/subtilisin-like cleavage motif does not exit the endoplasmic reticulum (Molloy et al, 1994).  Similarly, the hl99 proteins may not be properly  processed; this could be detected by standard Western blot analysis on protein isolated from transgenic worms.  The H i s to L e u change is 11 amino acids  carboxyl to one of three potential kex2/subtilisin-like cleavage motifs (R-V-KR) present i n the common region of predicted bli-4 isoforms.  Searching for the s90 lesion in the 3'-specific exons of bli-4  The sequence determination of pCeh207 not only established the molecular structure of the 3' end of bli-4, but also allowed the design of primers for mutation searching i n this region of the gene.  These primers  were employed to begin testing a hypothesis originally proposed by Peters (1992), suggesting that the class III mutation, s90, affects isoform(s) other than blisterase A (the transcript that is eliminated i n e937 homozygotes).  This  hypothesis was based on intracomplementation analysis, w h i c h indicated that s90 complemented the blistered phenotype of e937; e937/s90 worms were wild-type i n appearance.  This complementation result w o u l d  normally suggest that these two mutations affect different genes, however, at least three pieces of evidence support the hypothesis that s90 is an allele of bli-4: 1) s90 failed to complement all class II alleles (Peters, 1992; Thacker, et  al, 1995); 2) the allele mapped to the bli-4 region genetically (Peters, et al, 1991); and 3) as shown i n this thesis, s90 was rescued by many subclones of the bli-4 gene.  Therefore, given that s90 is a bli-4 allele, one might predict  that the s90 lesion w i l l be found w i t h i n a region that affects the expression a n d / o r function of blisterase B, C , D , or some combination thereof.  One  possible location for such a mutation w o u l d be the 3'-specific exons for these isoforms. Sequencing of exons i n the 3' region of bli-4 was initiated by subcloning P C R fragments obtained w i t h Taq polymerase.  After sequencing  through both mutant and wild-type clones for exons 16, 19 and 20, it was decided that Pfu polymerase should be used to generate the amplified D N A from s90 homozygotes, especially after discovering a second nucleotide alteration i n the hl99 clone pCeh240, w h i c h was likely attributable to Taq polymerase (Figure 17).  However, no mutations were detected i n any of the  regions i n the s90 clones amplified w i t h Taq polymerase.  Subsequent  cloning and sequencing of P/u-generated P C R fragments was performed by C o l i n Thacker.  A summary of some regions sequenced is given i n Figure 21.  To date, all 3'-specific exons for each of the four transcripts, except 94 nucleotides at the end of exon 18 have been sequenced from s90 homozygotes (C. Thacker, pers. comm.). The absence of the s90 mutation i n a 3'-specific exon w o u l d promote a revision of the current hypothesis explaining the intragenic complementation between s90 and e937. However, a mutation that specifically affects a subset of the products produced by bli-4 does not have to reside i n a coding region of bli-4.  For instance, a mutation i n a czs-acting  regulatory element that directs expression of the gene i n early development, could result i n the lethal phenotype associated w i t h s90.  If the same allele  allowed normal expression of the gene at the adult stage, the blistering phenotype of e937 could be complemented.  Another way to affect a subset of  bli-4 products without mapping i n the 3' exons is if s90 resides i n one of the unsequenced introns of the 3' region, such that the post-transcriptional processing of specific transcripts is affected.  One example of an alteration  implicated i n this phenomenon i n C. elegans is the intragenic revertant of unc-52(e669), su250 (Rogalski, et al, 1995). su250 is a single nucleotide alteration that resides in the centre of intron 16 of the unc-52 gene, and reverts the paralyzed phenotype of the e669 mutation (which is a point mutation i n exon 17). The s90 mutation may also reside i n the region common to all isoforms.  In fact, the data from hl99/e937 animals (Table 4) suggests that at  least some distinction between the adult cuticular function and the early developmental function can be separated by a mutation i n the common region.  Complementation of blistering i n s90/e937 may be a severe example  of the same mechanism responsible for the reduction i n blistering observed in hl99/e937  hermaphrodites.  The structure of the bli-4 gene  Our understanding of the function of the predicted kex2/subtilisin-like proprotein convertases encoded by the bli-4 gene relies on a thorough understanding of the molecular structure of the gene.  Most of the genomic  D N A for the bli-4 gene has n o w been sequenced and a h i g h resolution restriction map is available.  Future work on bli-4 w i l l likely include  completing the sequencing of the gene, i n particular, the ~6 kb intron separating exons 13 and 14.  Transformation rescue experiments with subsets of bli-4 coding information  Genetic evidence, combined w i t h mapping data for molecular lesions of bli-4 suggest that at least some of the predicted isoforms of this gene perform distinct functions.  In an attempt to further elucidate the  relationship between the structure and function of the gene, subclones of bli-4 were assessed for transgenic rescuing ability against four alleles, representing each class of bli-4 mutant phenotype.  The preliminary results  indicated that the isoforms of bli-4 are sufficiently similar to provide overlapping function; exogenous blisterase A rescued lethality as w e l l as blistering, and exogenous blisterases B, C , and D rescued blistering and lethality w h e n expressed from the same construct.  This latter result was not  expected, since e937 homozygotes do not produce the blisterase A transcript, but do make blisterase B, C , and D transcripts (Thacker, et al., 1995).  This  apparent contradiction i n the predicted functional roles of the blisterases is perhaps due to aberrant expression of the exogenous gene, rather than functional equality.  In support of this reasoning, pCeh221 rescued blistering  and lethality even though it only encodes the first 12 exons of the gene and therefore, cannot make a protein identical to any of the predicted endogenous products.  A l s o , incomplete penetrance of blistering i n e937  homozygotes suggests that endogenous levels of the non-blisterase A isoforms can rescue blistering i n a small percentage of animals.  Transgenic rescue is a widely used method of determining the smallest fragment of D N A that is capable of reverting a mutant phenotype.  This is a  convenient way to determine the physical location of a gene and, i n some instances, the regulatory elements necessary for its expression.  However,  gene expression and control of many exogenous copies present i n the extrachromosomal array is a poorly understood process.  Rescue data derived  from transgenic experiments involving a w e l l characterized gene is sometimes contradictory to expectations.  For instance, mutant rescue w i t h  exogenous constructs that do not encode the complete gene (like pCeh221) are not uncommon (e.g., Bargmann, et al., 1995; Babity, 1992).  For this  reason, the conclusions made from transgenic rescue experiments are often limited to "positive" or "negative", especially w h e n dealing w i t h lethal mutations.  In general, the presence of a homozygous marker that flanks the  lethal mutation i n the F l progeny from a heterozygote carrying the array is sufficient to conclude that the array rescues lethality. In this thesis, however, I have scored all of the progeny from lethal transgenic heterozygotes, i n an attempt to carefully measure the rescuing capacity of the subclones of bli-4.  It was suspected that, if slight functional  redundancy d i d exist between isoforms in vivo, exogenous expression of inappropriate isoforms may have resulted i n incomplete rescue of lethality. This incomplete rescue w o u l d have been observed as lethal homozygotes that survived and produced progeny, but their frequency i n the population w o u l d have been lower than expected, much the same w a y that incomplete penetrance of blistering i n e937 was evident i n a population of worms, not i n an i n d i v i d u a l blistered w o r m .  For instance, a comparison between the  transmission frequency of the array (Rol/Rol+WT) and the frequency of rescue (Dpy/Dpy+dead) provides the information necessary to determine the  penetrance of lethality i n transgenic worms.  In this manner, if frequency of  transmission of the array and frequency of rescue by the array are equivalent, the conclusion w o u l d be complete rescue.  In two cases, the frequency of  transmission of the array was consistently higher (for all three alleles) than frequency of rescue by the array (Figure 24). By this criterion, pCeh221, and one of the minigenes that encodes blisterase A , pCeh230, d i d not rescue lethality as w e l l as the blisterase B, C, D minigene (pCeh236) or other blisterase A constructs (pCeh226 and pCeh229). If the reason for overlapping function from exogenous constructs was due to over-expression or aberrant expression, perhaps the reduced rescuing ability of pCeh221 and pCeh230 was a result of the removal of a large portion of intron 12.  Since the blisterase B,  C, D minigene, pCeh236, also lacks this region, it may be that the predicted isoforms encoded by this construct are more closely associated w i t h the essential role predicted for the bli-4 gene, and that even reduced expression still allowed complete rescue.  This interpretation is consistent w i t h the  intracomplementation data for the bli-4 gene.  Blisterase B partially rescued blistering but not lethality  The minigene pCeh238 can code for the blisterase B isoform. Transgenic animals homozygous for the e937 mutation were observed to blister at reduced penetrance, indicating that blisterase B partially rescued this phenotype.  U p o n introduction of pCeh238 into the lethal strains, it was  discovered that blisterase B was unable to rescue any of the lethal alleles, q508, hl99, or s90.  In light of the apparent functional redundancy exhibited  by the other isoforms of bli-4, these results were somewhat surprising.  The  carboxyl-terminus of blisterase B is quite similar to the carboxyl-terminus of blisterase A ; both regions are approximately equal i n length and contain several hydrophobic residues.  Therefore, it was expected that blistering  w o u l d be completely rescued, such as w i t h pCeh236 (blisterase B, C , and D , together). The blisterase B specific exon (exon 14) has a 3' untranslated region (3'UTR) of just 23 nucleotides, w h i c h is extremely small w h e n compared to most C. elegans genes.  This, combined w i t h Northern blot analysis  indicating blisterase B is weakly expressed i n mixed stage R N A (Thacker, et al, 1995), suggests that l o w endogenous blisterase B expression may be a result of reduced stability of this transcript.  Therefore, it is possible that  blisterase B levels from the array may be lower than other isoforms even if pCeh238 is transcribed as efficiently as the rescuing constructs.  A n additional  feature of pCeh238 that may have contributed to its lack of rescuing ability is that it contains the first exon of blisterase C / D .  The presence of exon 15,  w h i c h has a canonical splice acceptor sequence ( T T T A C A G ) may result i n the production of an aberrant transcript, further reducing the amount of blisterase B. Further evidence that l o w exogenous expression of blisterase B was responsible for its inability to completely rescue blistering is that one strain believed to contain an integrated pCeh238 array does rescue blistering completely (data not shown). The negative result for lethal rescue by pCeh238 and the negative rescue result for blistering by pCeh239 (C. Thacker, pers. comm.) indicates that the common region of bli-4 is necessary for rescue of the bli-4 mutant phenotypes.  However, pCeh221 rescue data suggests that this region is also  sufficient to rescue the mutant phenotypes (albeit at a reduced level for lethality) w h e n i n the context of this exogenous construct.  This apparent  contradiction implies a difference i n function or expression between the products of these constructs.  A l t h o u g h the function of the 3' ends of the  predicted bli-4 isoforms still remains a mystery, these structural features may account for the differing rescuing capacity of pCeh221, pCeh238 and pCeh239. Future analysis of the distribution of the predicted bli-4 isoforms i n the w o r m v i a immunolocalization may provide some evidence for their function. tissues.  For instance, individual isoforms may be targeted to different It w o u l d be informative to subject the transgenic worms to  immunolocalization studies to determine if inferred functional redundancy is due to aberrant localization of the exogenously produced proteins. bli-4 potentially produces at least four protein isoforms, each of w h i c h are identical at the amino-terminus (Figure 21) but differ i n their relatively small carboxyl-termini.  Therefore, the question of whether each isoform  performs a distinct in vivo function is difficult to assess w i t h transgenic rescue experiments.  Aberrant expression from an extrachromosomal array  could override endogenous bli-4 function, especially if the distinction between bli-4 isoforms in vivo is due to temporal a n d / o r spatial expression differences.  Conclusions  1. A refined structure of bli-4 was constructed.  The gene encodes 21 exons  and is trans-spliced to the leader sequence SL1.  2. The class II lethal mutation hl99 was mapped by PCR-based heteroduplex technique.  hl99 is an A : T to T : A transversion that results i n a H i s to L e u  substitution i n a region proximal to the protease domain, w h i c h all isoforms share.  hl99/e937 animals rarely blister, suggesting that hl99 retains some  activity, at least w i t h respect to the adult cuticular function associated w i t h this gene.  3. The s90 mutation has not been found, despite a search through most of the coding region from the 3' specific exons of the gene.  A revision of the  current hypothesis regarding the location of this complementing allele may be required.  4. Transgenic minigene experiments suggest that the isoforms produced by bli-4 are sufficiently similar to be functionally redundant, at least w h e n exogenously expressed.  The penetrance of lethality of rescued transgenic  lines was measured, providing evidence that not all minigenes are equivalent i n their ability to rescue lethality.  References  Babity, J. M . 1993. L o w level T e l transposition and Tcl-induced mutation in dpy-5. P h . D . Thesis, University of British Columbia, Vancouver, B.C., Canada. Barr, P. J. 1991. M a m m a l i a n subtilisins: The long-sought dibasic processing endoproteases. Cell 66: 1-3. Barstead, R.J., L . Kleiman, and R . H . Waterston. 1991. Cloning, sequencing, and mapping of an a-actinin gene from the nematode Caenorhabditis elegans. Cell Motil. Cytoskel. 20: 69-78. 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. H u m a n fur gene encodes a yeast KEX2-like endoprotease that cleaves pro-beta-NGF i n vivo. /. Cell Biol. I l l : 28512859. 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 . L i . 1967. Isolation, purification, and characterization of gamma-lipotropic hormone from sheep pituitary glands. Can J. Biochem. 45, 1164-1174. Docherty, K . and D.F Steiner. 1982. Post-translational proteolysis i n polypeptide hormone biosynthesis. Annu. Rev. Physol. 44: 625-638. Fuller, R.S., A . Brake, and J. Thorner. 1989a. Yeast prohormone processing enzyme (KEX2 gene product) is a Ca -dependent serine protease. Proc. Natl Acad. Sci USA. 86: 1434-1438. 2+  Fuller, R.S., A.J. Brake, and J. Thorner. 1989b. Intracellular targeting and structural conservation of a prohormone-processing endoprotease. Science 246: 482-486. Fuller, R.S., R.E Sterne, and J. Thorner. 1988. Enzymes required for yeast prohormone processing. Ann. Rev. Physiol 50: 345-362. Griffiths, A . , J. Millers, D . Suzuki, R. Lewontin and W . Gelbart, 1993. A n introduction to genetic analysis, E d . 5. W . H . Freeman & Co., N e w York.  Hayflick, J.S., W.J. Wolfgang, M . A . Forte, and G . Thomas. 1992. A unique Kex-2 like endoprotease from Drosophila melanogastor is expressed i n the central nervous system d u r i n g early embryogenesis. /. Neurosci. 12: 705-717. Horvitz, H.R., S. Brenner, J. H o d g k i n , and R. Herman. 1979. A uniform genetic nomenclature for the nematode Caenorhabditis elegans. Moi. Gen. Genet. 175: 129-133. Horwitz, J. P., J. Chua, R. J. Curby, A . J. Tomson, M . A . DaRooge, B. E. Fisher, J. Mauricio, and I. Klundt. 1964. Substrates for cytochemical demonstration of enzyme activity. I. Some substituted 3-indolyl-p-Dglycopyranosides. J. Med. Chem. 7: 574-583. H o w e l l , A . M . , 1989. Essential genes i n a region of chromosome I i n Caenorhabditis elegans. P h . D . Thesis, University of British C o l u m b i a , Vancouver, B.C., Canada. H o w e l l , 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. Keen, J., D . Lester, C . Inglehearn, A . Curtis, and S. Bhattacharya. 1989. Rapid detection of single-base mismatches as heteroduplexes on H y d r o L i n k gels. Trends in Genetics. 7: 5 Kiefer, M . C , J.E. Tucker, R. Joh, K . E . Landsberg, D. Saltman, and P.J. Barr. 1991. Identification of a second h u m a n subtilisin-like protease gene i n the fes/fps region of chromosome 15. DNA. Cell Biol. 10: 757-769. Kramer, J.M., R.P. French, E-C. Park, and J.J. Johnson. 1990. The Caenorhabditis elegans rol-6 gene, w h i c h interacts w i t h the sqt-1 collagen gene to determine organismal morphology, encodes a collagen. Moi. Cell Biol. 10: 2081-2089. Kuwabara, P. E., and J. Kimble. 1992. Molecular genetics of sex determination in C. elegans. Trends in Genetics. 8: 164-168. Lundberg, K . S., D . D . Shoemaker, M . W . A d a m s , J . M . Short, J. A . Sorge, and E. J. Mathur. 1991. High-fidelity amplification using a thermostable D N A polymerase isolated from Pyrococcus furiosus. Gene. 108: 1-6 Lusson, J., D . Vieau, J. Hamelin, R. Day, M . Chretien, and N . G . Seidah. 1993. c D N A structure of the mouse and rat subtilisin/kexin-like PC5: a candidate proprotein convertase expressed i n endocrine and nonendocrine cells. Proc. Natl. Acad. Sci. 90: 6691-6695.  M c K i m , K.S., A . M . H o w e l l , and A . M . Rose. 1988. The effects of translocations o n recombination frequency i n Caenorhabditis elegans. Genetics 120: 987-1001. M e l l o , C.C., J . M . Kramer, D . Stinchcomb, and V . Ambros. 1991. Efficient gene transfer i n C. elegans: extrachromosomal maintenance and integration of transforming sequences. EMBO f. 10: 3959-3970. M i z u n o , K . , T. N a k a m u r a , T. Ohshima, S. Tanaka, and H . Matsuo. 1988. Yeast K E X 2 gene encodes an endopeptidase homologous to subtilisin-like serine proteases. Biochem. Biophys. Res. Commun. 156: 246-254. M o l l o y , S.S., L . Thomas, J.K. VanSlyke, P.E. Stenberg, and G . Thomas. 1994. Intracellular trafficking and activation of the furin proprotein convertase: localization to the T G N and recycling from the cell surface. EMBO J. 13: 18-33. M o r i , I., D . G . Moerman, and R . H . Waterston. 1988. Analysis of a mutator activity necessary for germline transposition and excision of T e l transposable elements i n Caenorhabditis elegans. Genetics 120: 397407. M u l l e r , H.J. 1937. Further studies on the nature and causes of gene mutations, pp 213-255 i n Proceedings of the Sixth International Congress of Genetics, edited by D . Jones. Brooklyn Botanical Gardens, Menasha, Wisconsin. • Nakayama. 1993. Identification and functional expression of a new member of the mammalian Kex2-like processing endoprotease family: its striking structural similarity to P A C E 4 . J. Biochem. 113: 132-135. Nakayama, K., W.-S. K i m , S. Torii, M . Hosaka, T. Nakagawa, J. Ikemizu, T. Baba, and K.; M u r a k a m i . 1992. Identification of the fourth member of the mammalian endoprotease family homologous to the yeast Kex2 protease. /. Biol. Chem. 267: 5897-5900. Perry, M . D . , C . Trent, B. Robertson, C . Chamblin and W . Wood. 1994. Sequenced alleles of the Caenorhabditis elegans sex-determining gene her-1 include a novel class of conditional promoter mutations. Genetics 138: 312-327. Peters, K . W . 1992. A genetic and molecular characterization of the bli-4 locus of Caenorhabditis elegans, an essential gene encoding kex2-like subtilisin-type serine endoproteases. P h . D . Thesis, University of British Columbia, Vancouver, B.C., Canada.  Peters, K., J. M c D o w a l l , 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. Roebroek, A . J . M . , J.A. Schalken, J . A . M . Leunissen, C . Onnekink, H.P.J. Bloemers, and W . J . M . V a n de V e n . 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. Roebroek, A . J . M . , J . W . M . Creemers, I.G.L. Pauli, U . K u r z i k - D u m k e , M . Rentrop, E.A.F. Gateff, J . A . M . Leunissen, and W . J . M . V a n de V e n . 1992. C l o n i n g and functional expression of Dfurin2, a subtilisin-like proprotein processing enzyme of Drosophila melanogastor w i t h multiple repeats of a cysteine motif. /. Biol. Chem. 267: 17208-17215. Rogalski, T. M . , E. J. Gilchrist, G . P. M u l l e n , and D . G . Moerman. 1995. Mutations in the unc-52 gene responsible for body w a l l muscle defects in adult Caenorhabditis elegans are located i n alternatively spliced exons. Genetics. 195: 52-56. Rose, A . M . , D.L. Baillie, and J. Curran. 1984. Meiotic pairing behavior of two free duplications of linkage group I of Caenorhabditis elegans. Moi. Gen. Genet. 195: 52-56. Sambrook, J., E.F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual. C o l d Spring Harbor Laboratory, C o l d Spring Harbor, N Y . Seidah, N . G . , L . Gaspar, P. M i o n , M . Marcinkiewicz, M . M b i k a y , and M . Chretien. 1990. c D N A sequence of two distinct pituitary proteins homologous to Kex2 and furin gene products: tissue-specific m R N A s encoding candidates for pro-hormone processing proteinases. DNA and Cell Biol. 9: 415-424. Seidah, N . G . , M . Marcinkiewicz, S. Benjannet, L . Gaspar, G . Beaubien, M . G . Mattei, C . Lazure, M . M b i k a y , and M . Chretien. 1991. C l o n i n g and primary sequence of a mouse candidate prohormone convertase P C I homologous to P C 2 , furin and Kex2: distinct chromosomal localization and messenger R N A distribution i n the brain and pituitary compared to P C 2 . Moi. Endocrinol. 5: 111-122. Seidah, N . G . , and Chretien, M . (1992). Proprotein and prohormone convertases of the subtilisin family. Trends Endocrinol. Meta. 3: 133140.  Seidah, N . G . , R. Day, J. Hamelin, A . Gaspar, M . W . Collard, and M . Chretien. 1992. Testicular expression of PC4 i n the rat: molecular diversity of a novel germ cell-specific kex2/subtilisin-like proprotein convertase. Mol. Endocrinol. 6: 1559-1570. Smeekens, S.P. and D.F. Steiner, D.F. 1990. Identification of a human insulinoma c D N A encoding a novel mammalian protein structurally related to the yeast dibasic processing protease Kex2. /. Biol. Chem. 265: 2997-3000. Smeekens, S.P., A.S. A v r u c h , J. LaMendola, S.J. Chan, and D.F Steiner. 1991. Identification of a c D N A encoding a second putative prohormone convertase related to P C 2 i n AtT20 cells and islets of Langerhans. Proc. Natl. Acad. Sci. 88: 340-344. Spieth, J. Brooke, G . Kuersten, S. Lea, K . and T. Blumenthal. 1993. Operons i n C. elegans: polycistronic m R N A precursors are processed by transsplicing of SL2 to downstream coding regions. Cell 73: 521-32. Thacker, C , Peters, K., Srayko, M . , and A . M . Rose. 1995. The bli-4 locus of Caenorhabditis elegans encodes structurally distinct k e x 2 / s u b t i l i s i n like endoproteases essential for early development and adult morphology. Genes and Devel. 9: 956-971. Thomas, G . , B . A . Thorne, and D.E. H u r b y . 1988. Gene transfer techniques to study neuropeptide processing. Annu. Rev. Physiol. 50: 323-332. U l l m a n n , A . , F. Jacob, and J. M o n o d . 1967. Characterization by in vitro complementation of a peptide corresponding to an operator-proximal segment of the [3-galactosidase structural gene of Escherichia coli. J. Mol. Biol. 24: 339-343. V a n de V e n , 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 i n higher eukaryotes. Mol. Biol. Rep. 14: 265-275. van den Ouweland, A . M . , H . L . van Duijnhoven, G . D . Keizer, L . C Dorssers, and W . J . M . V a n de V e n . 1990. Structural homology between the human fur gene product and the subtilisin-like protease encoded by yeast K E X 2 . Nuc. Acids Res. 18: 664. Wise, R. J., Barr, P. J., Wong, P. A . , Kiefer, M . C , Brake, A . J., and Kaufman, R. J. 1990. Expression of a human proprotein processing enzyme: correct cleavage of the v o n Willebrand factor precursor at a paired basic amino acid site. Proc. Natl. Acad. Sci. 87: 9378-9382.  Appendix 1.  earn QJ  ca  w-i  CO CM  cu  X QJ  o  X,-  u  <N  a, in  ex  X  W-  CO QJ  CM CO CM  00  LO CO CM  X QJ  u  — II  1  §  OJ  >  O  QJ QJ  jgJ  5-,  C  £ £  LO CM  13C/5  QJ  u  03-  ]  CM X QJ  cj  O  o  m-  OH  00  QJ CO CM  QJ rcj  X  o  QJ  U  00 '  •rH SH  '5b x i 2 3 OJ  CO  u  cC  o  cu  g  wSO  '5b  CN  J-i  X  QJ  O  X co -<-' CO  2 12 co QJ  4->  O  3  QJ  oo  u  0  4) U D  CM CM QJ  £-1  u vO, CM CM  QJ  u  0 CM CM  X QJ  o CN CM  X CD U O  QJ  "ti  QJ  «  CCS  C/3 QJ  c jo X C/J  CM QJ *-<  g co  .ti QJ 55 o o OH «s  cu  <a 5H  CJ  QJ  X  ^ ii H .5  Appendix 2. Evidence that KR513 no longer contains a linked lethal allele w i t h i n sDf4.  The following cross was performed by R. Johnsen: dpy-5 let(?) bli-4(hl99) unc-13; sDp2 X unc-11 dpy-14; szTl  (Lon males)  W T progeny recovered: ± unc-11  dpy-5 let(?) Ui-4(hl99) + + + + dpy-14  unc-13 +  cross the above animals to sDf4/hT2: dpy-5 let(?) bli-4(h!99) sD/4  unc-13  D p y progeny if let(?) is outside sD/4  M a n y D p y progeny were recovered (four were set up and each gave only D p y progeny). Therefore, if there is a lethal mutation other than bli-4(hl99), it is outside of sD/4.  M a p p i n g data: KR513 X N 2 males, set up heterozygous F l s for scoring (discard any F l s that give many Unc-13 F2 progeny, w h i c h is evidence for sDp2 i n F l ) wild-type 195 246 270 218 205 103 1237  Dpy 3 0 0 1 0 1 5  Unc 1 0 1 0 1 0 3  Dpy,Unc 0 0 0 0 0 0 0  Map distance = 0.3 m.u. between dpy-5 and hl99, as compared to 0-0.1 (Peters, et al, 1991), indicating that the lethal once linked to M99 in KR513 is no longer in this strain. This strain was reassigned as KR2997. Recombination frequency was calculated by p = 1-(1-2R) where R is the fraction of recombinant progeny over total progeny (Brenner, 1974) and total progeny is 4/3 (wild-type + Dpy) (Rose and Baillie, 1979) 1/2  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items