Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Essential genes in a region of chromosome I in Caenorhabditis elegans Howell, Ann Marie 1989

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

Item Metadata

Download

Media
831-UBC_1989_A1 H68.pdf [ 5.5MB ]
Metadata
JSON: 831-1.0098283.json
JSON-LD: 831-1.0098283-ld.json
RDF/XML (Pretty): 831-1.0098283-rdf.xml
RDF/JSON: 831-1.0098283-rdf.json
Turtle: 831-1.0098283-turtle.txt
N-Triples: 831-1.0098283-rdf-ntriples.txt
Original Record: 831-1.0098283-source.json
Full Text
831-1.0098283-fulltext.txt
Citation
831-1.0098283.ris

Full Text

ESSENTIAL GENES IN A REGION OF CHROMOSOME IN CAENORHABDITIS  ELECANS  By ANN MARIE HOWELL B.Sc, Simon Fraser University, 1983  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY  in THE FACULTY OF GRADUATE STUDIES GENETICS PROGRAMME  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA May 1989 © Ann Marie Howell, 1989  In  presenting  degree freely  at  this  the  available  copying  of  department publication  of  in  partial  fulfilment  of  the  University  of  British  Columbia,  I  agree  for  this or  thesis  reference  thesis by  this  for  his thesis  and  scholarly  or for  her  of  The University of British Columbia Vancouver, Canada  DE-6  (2/88)  I further  purposes  gain  shall  that  agree  may  representatives.  financial  permission.  Department  study.  requirements  be  It not  that  the  be  Library  an  advanced  shall  permission for  granted  is  for  by  understood allowed  the  make  extensive  head  that  without  it  of  copying my  my or  written  ABSTRACT This thesis describes the identification and characterization of essential genes in a small region of the genome of the nematode Caenorhabditis  elegans. The region analyzed was defined  by a 1.2 map unit deficiency of chromosome l, hD/6. In order to do this, a system for recovering and analyzing a large number of autosomal recessive lethal mutations was developed. This system used a free duplication of the left third of chromosome I, sDp2. Lethal mutations were maintained with two mutant alleles on the normal chromosomes and a wild-type allele on the duplication. More than 31,000 chromosomes mutagenized with ethyl methane sulfonate were screened using the sDp2 system and 495 lethal mutations were recovered. Two translocations involving chromosome I (,hT1(l;V) and szTl(l;X)) were also used to recover lethal mutations. It was discovered that heterozygotes for one of these translocations, szTl(l;X),  exhibited an  increased frequency of recombination adjacent to the chromosome I breakpoint. This had not been observed previously with other translocations. The increase in recombination frequency may be the result of a disruption of the normal regulation of recombination in the region. Recombination mapping was used to position 59 EMS-induced sDp2-recovered lethal mutations. Deficiency mapping, duplication mapping, and inter se complementation analyses were used to define 36 essential genes to the left of the dpy-5 gene on chromosome I. Nineteen of these genes, defined by 54 lethal mutations, were uncovered by the deficiency hDf6. A small gamma radiation-induced deficiency, hDf7, was identified and found to uncover six of the essential genes in hDf6. Two duplications which have breakpoints in the hD/6 region, hDp3 and hDp25, were also used to position essential genes. The genes in hD/6 were thus positioned into five regions based on the breakpoints of these three other chromosomal rearrangements. The stage of lethal arrest was determined for the lethal mutations in hD/6. Mutations in genes in the left portion of hD/6 arrest development at an earlier stage than do mutations in genes in the right portion. As a whole, the hD/6 region seems to have a lower proportion of early arresting mutations than other comparably sized regions of the genome.  Ten of the 19 genes in hDf6 were represented by more than one lethal allele. One gene, let-354, was found to be an extremely mutable target since it was represented by seventeen alleles. A truncated Poisson analysis of the allele distribution indicated that the 19 genes identified represented a maximum of 75% of the essential genes predicted to be in hD/6. Extrapolating from this region to the entire genome, an estimate of 4,000 essential genes was obtained. The average forward mutation rate was determined to be 5 x 10"^ mutations per gene. It was estimated that 60,000 chromosomes mutagenized with 0.015 M EMS must be screened to identify mutationally most of the genes in any region of the genome. Thus, it would require the screening of another 30,000 mutagenized chromosomes to identify genetically almost every gene in the hDf6 region of C. elegans.  iv  TABLE OF CONTENTS Abstract  ii  List of Tables  vii  List of Figures  viii  Quotation  ix  Acknowledgements  x  I. INTRODUCTION  1  A. Studies of Genome Organization  3  B. Lethal Analysis in Drosophila  5  melanogaster  C. Lethal Analysis in Caenorhabditis  elegans  6  D. Lethal Recovery and Balancing Systems  7  E. Analysis of the hDf6  8  Region  II. MATERIALS AND METHODS A. C. elegans  Nomenclature  10  B. Nematode Culturing and Strains  11  C. Chromosomal Rearrangements Used  17  D. Lethal Screening Using sDp2  18  E. Recombination Mapping Using sDp2  Lethals  20  F. Complementation Testing Using sDp2 Lethals 1. Deficiencies  21  2. Duplications  21  3. Non-lethal mutations  22  4. Lethal mutations  22  C. Recombination Mapping Using sz7"7  23  H. Lethal Screening Using sz7"7  23  I. Lethal Screening Using hT1  24  J. Mapping h7"7-recovered Lethals to Chromosome I and Balancing Over szTl  24  III. RESULTS A. Recovery of Lethal Mutations  26  1. sDp2  26  2. szT1(l;X)  28  3. hT1(l;V)  30  B. Characterization of sz77  32  1. Recombination suppression  32  2. Recombination enhancement  32  3. Lack of any interchromosomal effect  35  C. Recombination Mapping 1. Non-lethal mutations  37 37  2. Lethal mutations a. EMS-induced lethal mutations  37  b. Gamma-induced lethal mutations  45  D. Complementation Analysis of Mutations Left of dpy-5  47  1. Deficiencies  47  2. Lethal mutations  49  3. Non-lethal mutations  51  a. unc-77  58  b. h/'m-7  58  4. Duplications  60  5. Stage of arrest of lethal mutations in hD/6  64  a. let-354  6. sDp2 has a lethal mutation in the hDf6 region  68  70  IV. DISCUSSION A. Conclusions of Research  71  B. Estimate of the Number of Genes in hDf6  72  VI  C. Forward Mutation Rates in C. elegans  77  D. Stage of Developmental Arrest  79  E. Comparison of Balancing Systems  85  F. Comparison of Mapping Systems  89  1. Recombination mapping  90  2. Deficiency mapping  90  3. Duplication mapping  91  C. Summary of Findings/Contributions  92  H. Proposals for Future Experiments  93  I.  94  General Discussion  Bibliography  97  Appendix I. Complementation tables  105  Appendix II. Construction of strains for DNA polymorphism mapping  109  LIST OF TABLES Table 1.  Strains used (not including lethal strains generated)  12  Table 2.  Recovery of EMS-induced lethal mutations using sDp2  29  Table 3.  Recombination suppression and enhancement in szTl(l;X)  Table 4.  Recombination on other chromosomes in szT7 heterozygotes  36  Table 5.  Two-factor mapping data for non-lethal mutations  38  Table 6.  Three-factor positioning data for non-lethal mutations  39  Table 7.  Two-factor mapping data for EMS-induced lethals  41  Table 8.  Two-factor mapping data for gamma-induced lethals  46  Table 9.  Deficiency mapping of EMS-induced sDp2-rescued lethals  48  Table 10.  Duplication and deficiency mapping of essential genes in hDf6  52  Table 11.  Distribution of alleles compared to truncated Poisson distribution  55  Table 12.  Duplication and deficiency mapping of hT7-recovered lethal mutations  56  Table 13.  X-chromosome nondisjunction in him-1 strains  59  Table 14.  Duplication mapping of sDp2-rescued lethal mutations  62  Table 15.  Stage of arrest of lethal mutations in hD/6  66  heterozygotes  33  viii LIST OF FIGURES  1. Map of the C. elegans genome showing the positions of the markers used  15  2. Segregation from the sDp2 screening strain  19  3. Map of genome showing extent of balanced regions  27  4. Distribution of sDp2-recovered essential genes  44  5. Genetic map of the unc-11 - dpy-5 region on chromosome I  63  6. Growth curve for Unc-13s and Dpy-5 Unc-13s  65  7. Distribution of mutations in hDf~6  74  8. Distribution of lethal arrest stages for genes in the hDf6 region  82  9. Distribution of lethal arrest stages for genes in three regions of the C. elegans genome  83  "The ease of handling of the nematode coupled with its small genome size suggests it is feasible to look for mutants in all of the genes to try to discover how they participate in the development and functioning of a simple multicellular organism." (Brenner, 1974).  X  ACKNOWLEDGEMENTS  I would like to express my sincere gratitude to my research supervisor, Dr. Ann Rose, for her support, advice and encouragement. I would like to thank my supervisory committee for helpful discussions; Dr. D. L. Baillie, Dr. D. Holm and Dr. R. McMaster. I wish to give lots of thanks to all of my colleagues in the Rose lab, past and present, especially Terry Starr (we made it!), Kim McKim, Joe Babity and Ken Peters, and anyone who ever had to help with a lethal screen Thanks also to the worm people at SFU, especially Raja Rosenbluth. I gratefully acknowledge the technical assistance of Ayesha Hassan, John Kam, Rohinish Kisun, Pearly Lee, and Kelly McNeil. I would like to thank my family, especially my parents, for their encouragement and faith in my abilities. 1 will always be grateful for the love and support of my friends who have stuck with me through it all; Rohinish, Kingsley, Lorri, Diane, and Rene. This research was financially supported by a University Graduate Fellowship and a Medical Research Council of Canada Studentship.  1  I. INTRODUCTION  One of the major problems in biology is the determination of how genes are organized in eukaryotic genomes. There are several possible explanations for the arrangements of genes. The linkage of some sets of genes may have a selective advantage to the organism, or the linkage may simply be an evolutionary artefact. The organization of genes may be involved in the regulation of expression required to allow development of a higher eukaryote. Little is understood concerning the coordinated regulation of large networks of genes involved in producing differentiated tissue types. The arrangement of genes along a chromosome may also be related to the ways chromatin is packaged during meiosis. An understanding of how these or other mechanisms are involved in organizing eukaryotic genomes will require the description of many different genomic regions. A complete description of the organization of the genome of an organism would facilitate investigations of the relevance of the mechanisms mentioned above. A complete description would allow for an extensive study of networks of gene regulation. The genetic description would include the identification of all genes, their regulatory regions and sites affecting chromosome behaviour. One approach by which the genome could be described would be to identify mutationally all of the genes essential for development and fertility. In the genetically characterized regions of the Drosophila  melanogaster  genome, most of the genes identified were  essential genes (for example, Hilliker et al., 1980; Hochman, 1971; judd, Shen and Kaufman, 1972; Lasko and Pardue, 1988; Leicht and Bonner, 1988; Marchant and Holm, 1988; Woodruff and Ashburner, 1979; Wright, 1987). The small free-living soil nematode Caenorhabditis  elegans is an excellent model system  for the study of eukaryotic genome organization (Brenner, 1974). (A brief summary of C. elegans nomenclature can be found in Materials and Methods). A unique opportunity exists with C. elegans to describe genetically large portions of the genome. The worms are easy to grow and maintain in the laboratory and strains can be kept frozen in liquid nitrogen for many years. In  2 Drosophila  melanogaster,  the maintenance of large numbers of strains is difficult because they can  not be recovered after freezing. The generation time of C. elegans is three and one half days at 20 ° C . A wild-type hermaphrodite will produce an average of 350 progeny by self-fertilization. Hermaphrodites have five pairs of autosomes and two X chromosomes. Males have one X chromosome and are produced spontaneously by X-chromosome nondisjunction (Hodgkin, Horvitz and Brenner, 1979) at a frequency of 0.1% at 20 ° C (Rose and Baillie, 1979). One half of the progeny resulting from mating a male to a hermaphrodite are male. These features make C. elegans extremely amenable to genetic analysis. The genetic system is well developed; there were approximately 800 genes on the most recently published genetic map (Edgley and Riddle, 1987). The organism is composed of highly differentiated tissue types; muscle, nervous system, intestine, cuticle, etc. There is a small number of cells in the adult (less than 1,000 in an adult hermaphrodite). Since the worms are transparent, cell divisions can be observed in living animals. The entire embryonic and post-embryonic cell lineages have been described (Sulston and Horvitz, 1977; Sulston et al., 1983). The patterns of cell divisions are practically invariant between wildtype individuals. This allows a precise determination of the effects of mutations on development of the organism. A physical map of the entire C. elegans genome is being constructed (Coulson et al., 1986, 1988) using cosmid contigs (sets of overlapping cosmid clones) and clones in yeast artificial chromosome vectors. This map is near completion and soon cloned DNA from any region of interest will be available. In many regions of the genome, the physical and genetic maps are being aligned (for example, Creenwald et al., 1987; Prasad, 1988; Starr ef al., 1989). If a complete genetic map of the C. elegans genome were also available, a completely colinear genomic map could be created. The purpose of the work described in this thesis was to identify essential genes in a small autosomal region in Caenorhabditis  elegans.  3 A. Studies of Genome Organization Because of their small size, bacteriophage genomes were the first to be completely described genetically (for example T4 reviewed by Mathews ef a/., (1983), and lambda reviewed by Hershey, (1971)). This small size also permitted rapid molecular analysis. For example, the bacteriophage lambda genome of almost 50,000 base pairs has been completely sequenced (Hendrix ef a/., 1983). It follows from these studies that an ideal organism for the study of genome organization requires a small genome and a powerful genetic system. E. coli is a good example. The entire genome has been cloned and restriction mapped (Kohara, Akiyama and Isono, 1987; Smith ef al., 1987) and approximately half of its genes have been identified (Bachmann, 1983). Saccharomyces  cerevisiae also has a small genome suitable for cloning and a  powerful genetic system, but progress towards identification of all of the genes is lagging behind E. coli. In E. coli, S. cerevisiae, D. melanogaster, and C. elegans, investigations of genome  organization have been carried out at several levels of resolution. Studies of individual genes have shown that there are some similarities and some differences between the structures and methods of regulation of gene expression of prokaryotic and eukaryotic genes. The lack of polycistronic messages in eukaryotes showed that genes were not grouped into operons as had been described by Jacob and Monod (1961) for £. coli. However, both prokaryotic and eukaryotic genes are transcriptionally regulated by c/s-linked regulatory regions and trans-acting factors. The lac operon system in E. coli is a classic example of transcriptionally regulated gene expression (reviewed by Miller and Reznikoff (1978)). A regulatory element was identified genetically by mutations at the 5' end of the rosy (xanthine dehydrogenase) gene in D. melanogaster  which result in over-  production or under-production of the rosy gene product (Chovnick ef al., 1976; McCarron ef al., 1979). Some genes in both eukaryotes and prokaryotes have more than one promoter to allow for more flexibility in regulation (reviewed by Schibler and Sierra, 1987). Many eukaryotic genes have been shown to be regulated by transcription factors which bind to specific DNA sequences known generally as enhancers (reviewed by Schaffner, Serfling and Jasin, 1985). Tissue-specific  4  regulation of some genes has been shown to be controlled by the binding of tissue-specific transcription factors; for example, a B-cell specific transcription factor binds to an enhancer in an immunoglobulin heavy chain gene (Banerji, Olson and Schaffner, 1983; Gillies eta/., 1983). Eukaryotic mRNAs are capped and polyadenylated. The choice of poly A addition site can be a step in regulation of gene expression (reviewed by Friedman, Imperiale and Adhya, 1987). Eukaryotic genes contain introns, as first described for adenovirus 2 mRNAs by Berget et al. (1977) and the rabbit beta-globin by Jeffreys and Flavell (1977). The presence of introns makes it possible for one gene to produce different protein products by alternative selection of splice junctions (reviewed by Breitbart, Andreadis and Nadal-Ginard, 1987). Alternative splicing can also be used to control the on/off regulation of some genes (reviewed by Bingham ef al., 1988). The organization of gene families has been examined in many different eukaryotic systems. Several patterns of organization have been observed. One of the simplest patterns is tandem repetition of identical duplicated genes or gene repeat units; for example, ribosomal RNA genes (Drosophila, Rissotta, Atwood and-Spiegelman, 1966; yeast, Bell ef al., 1977; C. elegans, Sulston and Brenner, 1974; Files and Hirsh, 1981; Nelson and Honda, 1985) or histone genes (sea star, Howell ef al. 1987; sea urchins and other organisms, Maxson, Cohn and Kedes, 1983). Gene families which arose from tandem gene duplications and subsequent divergence have been observed; for example, the human beta-globin gene cluster (Karlsson and Nienhuis, 1985), Drosophila cuticle protein genes (Snyder and Davidson, 1983), vitellogenin genes (Wahli, 1988), and the gene cluster around the dopa decarboxylase gene in D. melanogaster  (Wright, 1987).  Some gene families are dispersed apparently randomly throughout the genome although some small clusters are often seen; for example tRNA genes (Cortese ef al., 1978), C. elegans major . sperm protein genes (Ward ef al., 1988), and C. elegans collagen genes (Cox ef al., 1985). Most genes are present in a single copy in the genome. These are the type of genes expected to be identified in screens for recessive mutations. However, there are examples of genes which are present in more than one copy in every organism studied. For example, C. elegans has three genes encoding acetylcholinesterase; animals with mutations in any one of  5  these genes alone have no phenotype, but animals with mutations in all three genes are not viable (Johnson et al., 1988). Genes which are reiterated are not expected to be identified in screens for recessive mutations. They can be identified by rare dominant mutations, for example, the dominant mutations of the actin gene cluster in C. elegans (Landel et al., 1984; Waterston, Hirsh and Lane, 1984). If an individual member of a gene family has acquired a unique function, recessive mutations may have a phenotype. For example, the dpy-13 gene in C. elegans is a member of the collagen gene family which has a unique function and therefore could be identified by recessive mutations (von Mende ef al., 1988).  B. Lethal Analysis in Drosophila  Drosophila  melanogaster  melanogaster has traditionally been the metazoan organism of choice for  genetic analysis. Several regions of the D. melanogaster  genome have been analyzed genetically.  In many cases an attempt was made to define all of the essential genes in a small region. The first genetic saturation analysis was undertaken by Hochman (1971). He attempted to identify all of the essential genes on the smallest chromosome, chromosome 4, and identified 33 essential genes. He estimated that only three or four genes were not identified. Judd, Shen and Kaufman (1972) attempted to identify all of the essential genes in the zeste-white D. melanogaster  region of the  X chromosome. They probably identified all of the essential genes since they  were all represented by more than one allele. Woodruff and Ashburner (1979) analysed the region around the Adh gene. Hilliker ef al. (1980) analysed the region adjacent to the rosy gene. They predicted that the 20 complementation groups they had identified were nearly all of the essential genes in their region. Lasko and Pardue (1988) identified seventeen lethal complementation groups in the vestigial region. Leicht and Bonner (1988) undertook an analysis of a small region around a cluster of small heat shock polypeptide genes. They identified at least 80% of the essential genes in that region. Marchant and Holm (1988) predicted they identified all of the essential genes in the heterochrbmatin of chromosome 3.  6  In the Drosophila  melanogaster  experiments, most of the lethal mutations were recovered  in precomplementation screens using a deficiency. The mutations were then placed over a balancer chromosome which suppressed recombination (for example, Judd, Shen and Kaufman, 1972; Hilliker et al., 1980; Leicht and Bonner, 1988; Marchant and Holm, 1988). Some of the mutations recovered in deficiency screens were "haplo-specific lethal", meaning that they are homozygous viable but hemizygous inviable (over a deficiency). Nash and Janca (1983) found that one third of lethal mutations recovered in a screen with an X chromosome deficiency were of this type. They explained them as hypomorphic alleles with an activity below the threshold for survival . with one dose, but near enough to that threshold such that an individual with two mutant copies is viable. A possible problem with some deficiency screens is that mutations may be recovered from elsewhere in the genome which are synthetically lethal with the deficiency. In the screen reported by Leicht and Bonner (1988), 30% of the mutations recovered did not fall in the deficiency they were screening against. They proposed that the presence of a Minute locus in the deficiency was responsible for much of the synthetic lethality.  C. Lethal Analysis in Caenorhabditis  elegans  Several regions of the C. elegans genome have been studied with regard to the distribution of essential genes. Many groups have recovered recessive lethal mutations over small autosomal regions: the region around unc-15 (I) (Rose and Baillie, 1980); the region around unc-54 (I) (Anderson and Brenner, 1984); the region around unc-22 (IV) (Rogalski, Moerman and Baillie, 1982; Rogalski and Baillie, 1985; Clark et al., 1988); and the region around ama-7 (IV) (Rogalski and Riddle, 1988). Sigurdson, Spanier and Herman (1984) have used an unanalyzed cross-over suppressor on chromosome II, mnCI, to recover lethals over a large autosomal region. The well characterized reciprocal translocation e77 (Rosenbluth.and Baillie, 1981) has been used to balance lethal mutations over two large autosomal regions (Rosenbluth, Cuddeford and Baillie, 1983, 1985; Rosenbluth et al., 1988). Meneely and Herman (1979, 1981) used a duplication of X-  7  chromosome material linked to chromosome V to recover and analyse X-linked lethal mutations. There are still many regions which need to be analyzed in detail before the genome will be well characterized.  D. Lethal Recovery and Balancing Systems  In C. elegans, translocations have been used successfully to recover and analyse large numbers of lethal mutations (Rosenbluth, Cuddeford and Baillie, 1983, 1985; Clark et al., 1988; Roseribluth et al., 1988). They are useful as balancers because they absolutely suppress recombination over large genetic intervals (Herman, 1978; Rosenbluth and Baillie, 1981; Herman, Kari and Hartman, 1982; Ferguson and Horvitz, 1985; Rosenbluth, Cuddeford and Baillie, 1985). A disadvantage to this approach is that the ends of the region under study are not precisely defined since lethal mutations which are linked to, but outside of, the boundaries of cross-over suppression are recovered. Screening for lethal mutations using a duplication or a deficiency will very precisely define the boundaries of the region under study. Deficiencies, however, generally can not be used to study large regions. A deficiency (sDf4) of approximately 2.5 map units on chromosome I has low viability as a heterozygote and could not practically be used in large-scale lethal screening experiments. Linked duplications have previously been used as balancers in Drosophila  melanogaster  (Judd, Shen and Kaufman, 1972) and in C. elegans (Meneely and Herman, 1979, 1981). A  .  drawback to the previous use of a linked duplication in C. elegans was that lethal mutations on the other chromosome were also recovered. In the study by Meneely and Herman (1979), only 21 of 176 putative lethals were in the region of interest. In order to carry out a genetic analysis of part of chromosome l in C. elegans, it was necessary to develop a system for the recovery and analysis of a large number of mutations in essential genes in an autosomal region. At the outset of this analysis, no translocations of chromosome I were available. A free duplication of the left third of chromosome l, sDp2, had  8 been described by Rose, Baillie and Curran (1984). Since sDp2 did not recombine with the normal homologues, Rose, Baillie and Curran (1984) suggested that it may be useful for balancing lethal mutations. An advantage to using a free duplication such as sDp2 is that all of the recovered lethal mutations are in a very precisely defined region on the linkage group of interest. The sDp2 system was also amenable to recombination mapping and complementation analyses. Subsequently, two chromosome I translocations became available, szT1(l;X) (Fodor and Deak, 1985) and hTl(l;V) (McKim, Howell and Rose, 1988). These translocations were used to recover and analyze lethal mutations in the same large region. In this thesis, I compare the relative merits of sDp2, szTl(l;X) and hTl(I;V) for rescuing lethals on chromosome l.  E. Analysis of the hDf6 region Dr. Rose's lab has been studying the genomic organization of the left half of chromosome I. I developed a system using sDp2 to isolate and characterize mutations in this region of the genome. For further analysis, I chose a smaller region which I could describe in greater detail. The region, to the left of dpy-5, was defined by the boundaries of the deficiency hDf6, and had not previously been examined with respect to the distribution of essential genes. This region and its environs were known to contain some interesting genes. Mutations in three genes to the left of the gene dpy-5, (unc-38, unc-63 and unc-74) confer resistance to the drug levamisole (Brenner, 1974; Lewis ef al., 1980) and have been proposed to code for subunits of the acetylcholine receptor. One of these three genes, unc-74, is uncovered by hD/6. Mutants of another gene in hD/6, unc-57, have a strong kinker uncoordinated phenotype (Brenner, 1974). They usually contract when touched on the head, but this is strongly inhibited in double mutants with unc-38, unc-63 and unc-74 (Lewis ef al., 1980), perhaps implying some functional interaction between these genes, unc-73 mutants are severely uncoordinated because the processes from the motor neurons are displaced (Hedgecock ef al., 1987). They are also defective in egg-laying because the hermaphrodite specific neurons do not migrate properly and form the proper connections with the vulval muscles (Desai ef al., 1988). Mutants in a gene involved in muscle  9  structure, unc-89, have disorganized thick filament assembly (Waterston, Thomson and Brenner, 1980). A gene which reduces X chromosome recombination and increases X chromosome nondisjunction, him-1, also maps to this region (Hodgkin, Horvitz and Brenner, 1979). Through a collective lab effort, 495 EMS-induced sDp2-recovered lethals have been analyzed. I have positioned those within the deficiency hD/6 by recombination mapping, deficiency mapping and duplication mapping. A second deficiency within hDf6, hDf7, was identified and mapped with respect to the essential genes. In all, nineteen essential genes were identified inside hDf6. The minimum number of essential genes in hD/6 was estimated to be twenty-five based on a calculation using the truncated Poisson distribution. Thus, approximately three quarters of the essential genes in this 1.25 map unit region of the C. elegans genome have been identified. The research described in this thesis was designed to examine several different questions relating to genes and genomic organization. The distribution of essential genes on chromosome I had not been described. It was not known whether the distribution of essential genes would follow the clustered distribution of visible genes on the chromosome (Brenner, 1974). It was not known how groups of genes were organized on chromosome I; for example, whether any apparent organization based on the developmental stage at which gene products are required could be discerned. The different mutabilities of genes had not previously been studied with such a large sample size. The analysis of essential genes in the hD/6 region resulted in the acquisition of new information regarding these questions.  10  MATERIALS AND METHODS  A. C. elegans Nomenclature  The nomenclature in this thesis follows the uniform system adopted for C. elegans (Horvitz et al., 1979), which is briefly summarized here. Gene names indicate the general phenotypic class; for example, dpy mutations cause the worms to be shorter and fatter than wildtype worms, unc mutations cause the worms to move in an uncoordinated fashion, and let mutations cause the worms to arrest development or to be sterile. Gene names are represented by three italicized lower case letters followed by a number. Mutant alleles are designated with one or two italicized lower case letters (signifying the lab of origin) followed by a number, and can be used alone or in parentheses after the gene name (for example, Iet-354(h79)). Phenotypes of mutant worms are indicated by using the appropriate gene name without italics and with-the first letter in upper case (e.g. Unc-13s). Nematode strains are identified with two upper case letters (signifying the lab of origin) followed by a number. The designations for A. Rose's lab are KR for strains and h for mutations. Chromosomal rearrangements are abbreviated as follows; Df for deficiency, Dp for duplication, and T for translocation. The five autosomes are indicated in upper case Roman numerals, and the X chromosome is indicated as "X". When discussing the reciprocal translocations, the nomenclature has been adapted from Horvitz ef al. (1979). When discussing the individual component chromosomes of a translocation, the adapted nomenclature describes the structure of the new chromosomes. For example, szTl (l;X) = I^X^szTI + and hTl(l;V) = l V hTl R  L  + l V hT1. L  R  The l X L  L  ftx^szTl  terminology implies that the new chromosome  carries the left portion of the X chromosome joined to the left portion of chromosome I. When describing genotypes of worms carrying duplications or translocations, the genotypes of translocation chromosomes (as in McKim, Howell and Rose, 1988) or duplication chromosomes (as in Rogalski, Bullerjahn and Riddle, 1988) are shown in brackets.  11 B. Nematode Culturing and Strains Caenorhabditis  elegans strains are usually maintained as self-fertilizing hermaphrodites  (5AA; XX). Males (5AA; XO) are found among the self progeny and result from X-chromosome nondisjunction (Hodgkin, Horvitz and Brenner, 1979). At 20 ° C they occur at a frequency of approximately 0.1% (Rose and Baillie, 1979). Wild-type and mutant strains were maintained and mated on petri plates containing nematode growth media (NCM), streaked with Escherichia coli OP50 (a uracil-requiring mutant) as described by Brenner (1974). The wild-type N2 strain and some mutant strains of C. elegans var. Bristol were obtained from D. Baillie, Simon Fraser University, Burnaby, Canada or the Caenorhabditis Genetics Center at the University of Missouri, Columbia. Strains used, with the exception of lethal-bearing strains, are listed in Table 1. Unless otherwise indicated, the e67 allele of dpy-5 was used. In addition to the new lethal mutations isolated in this study, the following mutant genes and alleles were used (listed by chromosome): I  bli-3(e579); bli-4(e937); dpy-5(e61); dpy-5(mn287); fog-1 (q253ts); him-l(e879); unc-11(e47); unc-63(e384);  dpy-14(e188);  fer-7(hc34);  Iet-354(ct42); Iin-6(e1466); spe-11(hc90); sup-11 (n403n682);  unc-13(e450); unc-15(e73); unc-29(e403); unc-38(e264); unc-67(e713); unc-73(e936); unc-74(x12); unc-75(e950);  II  dpy-10(e128);  unc-4(e120)  III  sma-2(e502);  unc-36(e251)  IV  unc-22(s7)  V  dpy-11(e224); sma-1(e30); unc-23(e25);  X  dpy-8(e130); lon-.2(e678); unc-3(e151)  unc-57(e406);. unc-89(e1460)  unc-42(e270)  Figure 1 a) is a partial C. elegans genetic map showing the map positions of the markers used. Figure 1 b) is an expansion of the unc-11 - unc-29 region of chromosome I.  12 Table 1. Strains used (not including lethal strains generated)  Strain  Genotype  Origin  AF1  +; dpy-8 + unc-3/szT1(l;X)[+;  BA34  fer-7(hc34)  BA662  spe-l1(hc90)  BW962  + Iet-354(ct42) dpy-5 +/unc-73 + + dpy-14  P. Mains  BC62  dpy-5 unc-75  A. Rose  BC207  dpy-5 unc-29  A. Rose  BC279  dpy-5 unc-13 unc-29  A. Rose  BC363  dpy-11 unc-23  R. Rosenbluth  BC417  dpy-5 unc-67  R. Rosenbluth  BC655  unc-36 sma-2  R. Rosenbluth  BC700  sDf4/bli-4  A. Rose  BC1142  unc-42 sma-1  D. Nelson  fog-1 (q253ts); him-5  K. Barton  KR182  unc-57 dpy-5  A. Rose  KR217  unc-73 dpy-5  A. Rose  KR235  dpy-5 unc-13 +1 dpy-5 + unc-15; sDp2  A. Rose  KR236  dpy-5 unc-13; sDp2  A. Rose  KR366  bli-3 dpy-5  A. Rose  KR900  dpy-5 unc-13; + unc-3/szT1[+  KR1012  dpy-5 unc-29 unc-75  A. M. Howell  KR1015  unc-11 unc-73  A. M. Howell  KR1069  sDf4; +/hT1[ + ; +]  K. McKim  KR1071  unc-11 dpy-5  A. M. Howell  JK?  .  + lon-2  +]  A. Fodor S. Ward D. Shakes  dpy-5; sDp2  dpy-14  +;lon-2  +]  A. M. Howell  Table. 1. (continued)  Strain  Genotype  Origin  KR1072  unc-63 dpy-5  A. M. Howell  KR1088  him-1 dpy-5  A. M. Howell  KR1106  unc-11 dpy-5; hDp2  K. McKim  KR1139  him-1 dpy-5; sDp2  A. M. Howell  KR1155  unc-38 dpy-5(mn287)  J. Babity  KR1238  unc-11 dpy-5; hDp3  K. McKim  KR1239  unc-11 dpy-5; hDp4  K. McKim  KR1241  unc-11 dpy-5; hDp6  K. McKim  KR1242  unc-11 dpy-5; hDp20  K. McKim  KR1250  unc-11 dpy-5; hDp5  K. McKim  KR1280  dpy-5 dpy-14; hDp13  K. McKim  KR1292  sup-11 dpy-5; +/szT1[+ +; lon-2]  A. M. Howell  KR1458  unc-11 dpy-14; +/szT1[+ +; lon-2]  K. McKim  KR1459  dpy-5 unc-13 +; +/hT1[+ + unc-29; +]  K. McKim  KR1469  dpy-5 dpy-14; hDp21  K. McKim  KR154.8  dpy-5 dpy-14; hDp22  K. McKim  KR1664  dpy-5 dpy-14; hDp23  K. McKim  KR1669  dpy-5 dpy-14; hDp25  K. McKim  KR1670  dpy-5 dpy-14; hDp24  K. McKim  KR1730  dpy-5 dpy-14; hDp34  K. McKim  KR1736  hDf6 dpy-5 unc-13; unc-3; l LX LszT1  K. McKim  KR1737  hDf6 dpy-5 unc-13; hDp31  A. M. Howell  KR1773  dpy-5 dpy-14; hDp31  K. McKim  KR1782  unc-89 dpy-5  K. McKim  Table 1. (continued)  Strain  Genotype  Origin  '  KR1783  dpy-5 dpy-14; hDp32  K. McKim  MT1403  sup-ll(n403n682)  J. Levin  MT1442  lin-6 dpy-5; +/szTl[+  SP74  dpy-10 unc-4  R. Herman  ZZ1012  unc-74 dpy-5  J. Lewis  + dpy-5/ + unc-11 + +; lon-2]  R. Horvitz  The strain names indicate the lab of origin. AF = A. Fodor's lab, Hungary; BA =' S. Ward's lab, Baltimore, MD; BC = D. Baillie's lab, Burnaby, B.C.; BW = B. Wood's lab, Boulder, CO; JK = J. Kimble's lab, Madison, Wl; KR = A. Rose's lab, Vancouver, B.C.; MT = R. Horvitz's lab, Boston, MA; SP - R. Herman's lab, St. Paul, MN; ZZ = J. Lewis' lab, Columbia, MO.  15  Figure 1 a). Partial genetic map of C. elegans showing the positions of the markers used. The region of chromosome I marked with asterisks is expanded in Figure 1 b).  sup-7 7 lin-6 fog-1 dpy-5 bli-3 unc-11 unc-29  I  I  1 I  I  1  unc-75  I  unc-54  1  I  **************  dpy-10  unc-4  III unc-36  I  sma-2  I  IV unc-22  V unc-23 sma-1 dpy-11 unc-42  I  I  I  I  X lon-2  dpy-8  I  I  unc-3  •I  16  Figure 1 b). Expansion of  unc-11  - unc-29  region of chromosome l showing extents of  deficiencies and duplications.  unc-89 unc-73 unc-74 unc-11 unc-57  I  I  I  I  him-1  I  dpy-5  I  bli-4  I  I  sDf4  hDf6  sDp2  hDp31,  32, 34  hDp13  hDp3  hDp25  I hDp20  hDp21  I hDp22  1 map unit  dpy-14  unc-15 unc-13  I I I  unc-29  [_  17  C. Chromosomal Rearrangements Used  The translocation szTl(l;X) was initially isolated by Fodor and Deak (1985) as a dominant X-chromosome crossover suppressor. It was induced on a lon-2(e678)(X) chromosome using 7000 R of X-radiation. Their analysis showed that szT7 consists of two abnormal chromosomes derived from the normal chromosome I and X chromosome and that the homozygotes arrest as embryos. They also showed that heterozygous hermaphrodites produce Lon-2 male self-progeny at a frequency of 0.08 - 0.12 and recombination was reduced in the dpy-7 - unc-3 (X) interval from 19.2 map units to 0.3 map units. hTl(l;V) was isolated by K. McKim in a screen for gamma-induced mutations causing pseudolinkage of markers on chromosomes I and V. It consists of the two translocation chromosomes l V hT1 L  R  and fiv^hTl. Recombination is completely suppressed along the left half  of chromosome! in translocation heterozygotes (from let-362 to let-88) (McKim, Howell and Rose, 1988). hT2(l;lll) was isolated by K. Peters in a screen for gamma-induced mutations causing pseudolinkage of markers on chromosomes ! and 111. Recombination is Completely suppressed along the left two thirds of chromosome I (at least as far as unc-75) in translocation heterozygotes (K. Peters and K. McKim, unpublished results). The deficiency, sDf4 (Figure 1 b), which deletes the unc-11 dpy-5 interval (Rose, 1980), was generated by treating wild-type male sperm with 0.07% formaldehyde as described by Rose and Baillie (1980). Strains heterozygous for sDf4 have no obvious visible phenotype, but have reduced progeny numbers. It has been maintained in strains heterozygous for bli-4 dpy-14 or the translocation hTl(l;V) by selecting phenotypically wild-type heterozygotes each generation. The deficiency, hDf6 (Figure 1 b), which was isolated as a gamma-induced lethal mutation (h545) balanced over szT7, was identified by K. McKim. K. McKim showed that h545 failed to complement unc-57 and unc-74 but not unc-11. ashD/6dpy-5 unc-13 (I); unc-3 (X); l X szT1, L  L  It was later named hDf6. hD/6 was maintained  or as hD/6 dpy-5 unc-13;  hDp31.  18 sDp2, a free duplication of the left third of chromosome I, carries the wild-type alleles of dpy-5 and dpy-14 but not of the unc-15 or the unc-13 loci (Rose, Baillie and Curran, 1984). Duplications with the h allele designation were generated and mapped with respect to visible markers by K. McKim (McKim and Rose, 1988; K. McKim, unpublished results). sDp2,  hDp13,  hDp31, hDp32 and hDp34 carry wild-type alleles for the farthest left known markers on chromosome I. The extents of all duplications used are shown in Figure 1 b.  D. Lethal Screening Using sDp2  In order to screen for lethal mutations of genes on the left third of chromosome I, the duplication sDp2 was used. A strain was constructed in which each chromosome I homologue was differentially marked. A single hermaphrodite of the genotype, dpy-5 + unc-131 dpy-5 unc-15 +; sDp2[+], was used to establish the strain KR235. Each of the segregating genotypes from KR235 was represented by a unique phenotype (Figure 2). Dpy-5 worms are approximately one-half the length of wild-type worms and are fatter. The two types of Uncs and Dpy Uncs present in these tests were phenotypically distinquishable from each other. Unc-13 animals are extremely uncoordinated and contract when touched, while Unc-15 worms have an extremely uncoordinated, limp phenotype. Diploid progeny were either Dpy (dpy-5 + unc-13 I dpy-5 unc-15 +) or Dpy Unc (dpy-5 unc-13 or dpy-5 unc-15). Duplication-carrying progeny were either Unc (dpy-5 unc-13/ dpy-5 unc-13; sDp2[+] unc-15/ dpy-5 unc-15; sDp2[+])  or "wild type" (dpy-5 unc-13 +1 dpy-5 + unc-15;  or dpy-5 sDp2[+])  ("wild type" = duplication-carrying non-Dpy, non-Unc, non-Dpy Unc). This strain was maintained by selecting the "wild-type" hermaphrodites and has been used for screening for chromosome I lethals. Lethal mutations in genes present in the sDp2 region were isolated and rescued by the wild-type allele on sDp2.  Individuals of the KR235 strain were treated either with  12, 15 or 17 mMethylmethane sulfonate (EMS) for four hours or with 1500 R (13 R/sec)  19  dpy-5  + unc-13  dpy-5  unc-15 +  "Wildtype"  Diploid individuals  sDp2-carrying individuals  + dpy-5  + unc-13  dpy-5  + unc-13  dpy-5  unc-15 +  dpy-5  unc-15 +  Dpy-5  dpy-5  "Wildtype"  unc-15 +  dpy-5  unc-15 +  dpy-5 • unc-15 +  dpy-5  unc-15 +  Dpy-5 Unc-15  Unc-15  + dpy-5  + unc-13  dpy-5  + unc-13  dpy-5  + unc-13  dpy-5  + unc-13  Dpy-5 Unc-13  Unc-13  Figure 2. Segregation from the sDp2 screening strain (KR235, dpy-5 + unc-13 I dpy-5 unc-15 +; sDp2). Lethal mutations induced on the dpy-5 unc-13 marked chromosome are identified by the absence of fertile Dpy-5 Unc-13 progeny and are maintained by recovery of their Unc-13 siblings. Individuals which receive two copies of sDp2 develop extremely slowly and are usually sterile.  20 of gamma radiation (Cobalt 60) as recommended by Rosenbluth, Cuddeford and Baillie (1983). Gravid "wild types" were individually placed on 10 x 60 mm culture plates after treatment. Five days later gravid "wild-type" F1s were placed individually on culture plates. Their progeny were screened for the absence of fertile Dpy-5 Unc-13 individuals. If fewer than 2 fertile Dpy-5 Unc-13s were observed, and if Unc-13s were present, a single Unc-13 was transferred to a fresh culture plate in order to confirm the existence of a lethal. A generation later the offspring of each Unc-13 strain were examined. Three types of strains were found: 1) both Uncs and fertile Dpy-5 Unc-13s were present (no lethal mutation); 2) no fertile Unc-13 could be recovered (a lethal not rescued by sDp2), and; 3) the desired cases, Unc-13s and developmentally arrested Dpy-5 Unc-13s were present. Thus, lethal mutations were rescued by the presence of a wild-type allele provided by sDp2.  Lethal mutations rescued in this way were maintained in strains with the  genotype (let-x dpy-5) unc-13/ (let-x dpy-5) unc-13; sDp2[+  +].  The parentheses indicate that  the let- gene could map either to the left or the right of dpy-5. These strains have an Unc-13 phenotype and segregate a single fertile phenotype (only Unc-13s).  E. Recombination Mapping Using sDp2 Lethals  Recombination mapping was done using procedures recommended by Rose and Baillie (1979). All recombination experiments were carried out at 20°C, and all of the progeny of a given hermaphrodite were scored. The total number of progeny was calculated as 4/3(wild types plus one recombinant class); the number of recombinants was calculated as two times one recombinant class. The recombination fraction, R, was calculated as (number of recombinants)/ (total progeny). The frequency of recombination, p, was calculated as 1 - (1-2R) ^ (Brenner, 1  1974). Confidence limits of 95% were calculated using Poisson statistics according to Crow and Gardner (1959). In order to generate the appropriate heterozygous individuals, N2 (wild-type) males were crossed to Unc-13 hermaphrodites from each of the lethal-bearing strains. Wild-type  21 heterozygotes which do not carry the duplication were required for mapping. Since the nonduplication carrying progeny were the first to reach adulthood, they were easily selected. Occasionally "wild types" carrying sDp2 were accidentally set up. These were easily identified because they segregated approximately 12% Unc-13 progeny whereas the heterozygotes lacking the duplication produced less than 1% Unc-13 progeny, the result of recombination events in the dpy-5 unc-13 interval. The self progeny of appropriate out-cross hermaphrodites (+  + +1 (let-x  dpy-5) unc-13) were scored. For those lethal mutations to the left of dpy-5, fertile Dpy-5 Unc-13 recombinants were recovered in a frequency proportional to the distance to the lethal. Unc-13 recombinants were recovered at the frequency expected for the dpy-5 unc-13 interval. For those lethal mutations to the right of dpy-5, Dpy-5 and Unc-13 recombinants were recovered. In this way, both right-left positioning relative to dpy-5 and two-factor recombination frequencies were obtained for each of the lethal mutations.  F. Complementation Testing Using sDp2 Lethals  1. Deficiencies Lethal mutations were tested for complementation to the two deficiencies, sD/4 and hD/6. N2 (wild type) males were crossed to Unc-13 hermaphrodites from the lethal-bearing strains, let-x dpy-5 unc-13/ + + + progeny males were mated to the deficiency strains. The presence of fertile Dpy-5s (for sDf4) or Dpy-5 Unc-13s (for hD/6) among the outcross progeny indicated complementation. A minimum of 30 wild-type outcross males were scored to ensure that the absence of the let-x/Df heterozygote was not due to reduced viability.  2. Duplications Lethal bearing strains (i.e. dpy-5 let-x unc-13; sDp2(l;f)) were crossed to unc-11 dpy-14; 0/szT1(l;X)[+;lon-2]  males. The resulting male progeny (let-x dpy-5 + unc-13/ + + dpy-14  +)  were then crossed to a) dpy-5 dpy-14; hDpz(l;f) or b) unc-11 dpy-5; hDpz(l;f) hermaphrodites.  22 Wild-type hermaphrodite progeny from cross a) were let-x dpy-5 + unc-13/ + dpy-5 dpy-14 + ; hDpz or + dpy-5 dpy-14 /unc-11 + dpy-14; hDpz if hDpz was dpy-14(+).  These two possibilites  were distinguished by examining their progeny. Upon examining the progeny of the former, Dpy-5 progeny were observed. Wild-type hermaphrodite progeny from cross b) were + let-x dpy-5 unc-13/ unc-11 + dpy-5 +; hDpz.  In both cases, fertile Unc-13 progeny were observed in  the next generation if the duplication carried let-x(+).  In some cases where the duplication did  not actually carry let-x(+), a few fertile Unc-13s were observed. These were recombinants which had lost the lethal mutation on one homoiogue. They were easily distinguished because they segregated fertile Dpy-5 Unc-13s.  3. Non-lethal mutations Lethal mutations were tested for complementation to non-lethal (visible) mutations. Visible mutations (m) linked to dpy-5 were crossed to N2 (wild-type) males, m dpy-5/ + + progeny males were mated to Unc-13 hermaphrodites from different lethal-bearing strains. If non-M Dpy-5 progeny were observed, the visible mutation complemented the lethal mutation. If M Dpy-5 progeny were observed, the visible mutation failed to complement the lethal mutation. If no Dpy-5 progeny were observed after a successful mating, the visible mutation failed to complement the lethal mutation.  4. Lethal mutations Heterozygous males generated by crossing sDp2 lethal strains to N2 (wild type) males were used for complementation testing. These males were mated to Unc-13 hermaphrodites from different lethal-bearing strains. The sDp2-carrying males are slower at developing and less effective at mating than males lacking the duplication (Rose, Baillie and Curran, 1984). The out-cross progeny were scored. The presence of Dpy-5 Unc-13 males and fertile Dpy-5 Unc-13 hermaphrodites indicated complementation. For lethals tightly linked to dpy-5, the presence of fertile Dpy-5 Unc-13 progeny was diagnostic of complementation. For lethals which mapped ten  23 or more map units to the left of dpy-5, out-cross progeny were scored to ensure that Dpy-5 Unc-13 offspring were present in excess of the number expected from recombination between the lethal and dpy-5.  G. Recombination Mapping in szTl Heterozygotes  Recombination frequencies between pairs of markers were determined by scoring the progeny of c/s-heterozygous hermaphrodites under the conditions described by Rose and Baillie (1979). When the interval being examined spanned the translocation breakpoint, one of the markers crossed onto the translocation chromosome. When the four types of gametes from a dpy-5 unc-zl szTl(l;X)  heterozygote occurred with equal frequencies, the recovery of the Unc-z  recombinant was expected at a frequency twice that of the reciprocal Dpy-5 recombinant when p was small (described in McKim, Howell and Rose, 1988). In these situations recombination frequency was calculated as; p = [(4A + 2W) - ((4A + 2W) - 16A(4A + W)) ]/2(4A + W) 2  1/2  where A is the number of recombinant Dpy-5 progeny and W is the number of wild types.  H. Lethal Screening Using sz77  Spontaneous Lon-2 males from the strain AF1, +; dpy-8 + unc-3/szTl(l;X)[+;  + lon-2 +],  were crossed to dpy-5 unc-13; unc-3 hermaphrodites to construct the strain KR900, dpy-5 unc-13; + unc-3/ szTl(l;X)[+  +; lon-2 +]. Hermaphrodites from this strain were treated with 1500 rads  of gamma radiation (Cobalt 60) as recommended by Rosenbluth, Cuddeford and Baillie (1985) and the progeny of individual F1 wild types were examined for the absence of Dpy-5 Unc-13 Unc-3 offspring. In this way strains carrying lethal mutations induced in the cross-over suppressed regions of either chromosome I or the X chromosome were isolated. Strains which failed to produce Unc-3 male progeny after crossing to dpy-5 unc-13/ + + males were assumed to carry  24 an X-linked lethal mutation and were not analyzed further. The lethals were mapped with respect to the dpy-5 and unc-13 markers (after replacing sz77 with a normal chromosome) and tested for complementation with visible and lethal mutations in the region.  1. Lethal Screening Using hTl  Young adult dpy-5 unc-13 +; +1 hT1(l;V)[ + + unc-29;+] hermaphrodites were treated with EMS in accordance with Brenner (1974), except that the dose used was only 0.017 M as recommended by Rosenbluth, Cuddeford and Baillie (1983). They were plated individually and allowed to self. The F1 progeny were checked to ensure segregation of fertile Dpy-5 Unc-13s (no lethal in the previous generation). Their wild type siblings were plated individually and allowed to self: The progeny of 3,000 F1s were screened for the absence of fertile Dpy-5 Unc-13s by A. Rose, K. McKim and myself. Strains were discarded if even one fertile Dpy-5 Unc-13 individual was seen since these would most likely carry lethal mutations just outside the boundaries of crossover suppression. If no fertile Dpy-5 Unc-13s were seen, the strain was assumed to be carrying a new lethal mutation on the balanced regions of chromosome I or V and was maintained for further analysis.  J. Mapping h77-recovered Lethals to Chromosome I and Balancing Over szT1  The lethal bearing strains described above could have had a lethal mutation on chromosome I or V. In order to assign each lethal mutation to the correct linkage group, the hT1 chromosomes were replaced with a normal chromosome V and the translocated I and X chromosomes from szTl(l;X).  This eliminated the pseudolinkage of chromosome I and V in the  lethal bearing strains. Lethal strains (let-x dpy-5 unc-13 +; +/ hTl(l;V)[+ dpy-5 unc-13 +; let-x/ hTl(l;V)[+ hT2(l;lll)  + + unc-29; +] or  + unc-29; +]) were mated to Lon-2 males of the genotype  [dpy-5 +; +; 0]/ szTl(l;X)[+  unc-29; +; lon-2]. Wild-type outcross hermaphrodites were  25 allowed to self. The segregation of Lon-2 males indicated the presence of szT1. The segregation of many fertile Dpy-5 Unc-13s indicated that the lethal mutation was on chromosome V. The presence of one or a few fertile Dpy-5 Unc-13s indicated that the lethal was on chromosome I but just outside of the boundary of cross-over suppression. If no fertile Dpy-5 Unc-13 individuals were observed, the lethal mutation was assumed to be in the cross-over suppressed region on chromosome I. The szH-balanced strains were used for any further analyses.  26 III. RESULTS  A. Recovery of Lethal Mutations  The major goal of this research was to identify essential genes in the region of the deficiency hD/6 on chromosome I. In order to do so, it was necessary to develop a usable system for recovering and analyzing a large number of lethal mutations in an autosomal region of Caenorhabditis  elegans. Three different methods were adopted for recovering mutations in  essential genes (described in detail in Materials and Methods). The free duplication, sDp2, and two translocations, szT1(l;X) and hT1(l;V), were used to rescue lethal mutations. Figure 3 shows the regions of the genome from which lethal mutations would be recovered using the three screening methods. Note that only with sDp2, all lethal mutations recovered were on chromosome I.  1. sDp2  In order to recover lethal mutations in the sDp2 region, a strain carrying the duplication and two differentially marked chromosome I's was used. This strain, KR235, had the genotype dpy-5 + unc-13/ dpy-5 unc-15 +; sDp2 [+].  After EMS mutagenesis, F1 progeny carrying a lethal  mutation were identified by the absence of Dpy-5 Unc-13 F2 progeny. An Unc-13 individual was isolated from each of these F1's in order to maintain mutations induced on the dpy-5 unc-13 chromosome within the sDp2 region. These lethal strains had the genotype, (let-x dpy-5) unc-13/ (let-x dpy-5) unc-13; sDp2 [+ +]. Since the duplication does not carry the wild-type allele of unc-13, these strains had an Unc-13 phenotype and segregated Unc-13 and developmentally arrested Dpy-5 Unc-13 animals. Because a lethal mutation is present on both normal homologues, only those mutations having a wild-type allele on the duplication were recovered in this way.  27  dpy-5  I  unc-29  I  xxxxxxxxxxxxxxxxxxxxxxxxxx x x  dpy-10  111 unc-36  I IV unc-22  I unc-60  dpy-11  I  unc-42  II  xxxxxxxxxxxxxxxxxxxxxxxxxxxxx X X dpy-3  I  lon-2  I  unc-3  I  Figure 3. Map of the C. elegans genome showing extents of balanced regions. Regions of the genome from which lethal mutations will be recovered using the three screens described are indicated as follows: ******  Region of the genome screened using sDp2(l;f)  xxxxx  Region of the genome screened using hTl(l;V)  —~ ~——  Region of the genome screened using szTl(l;X).  28 Several screens have been done in this laboratory (Table 2). A total of 552 lethal mutations have been recovered from 31,606 mutagenized chromosomes. The cumulative induction frequency was 1.7%. Fourteen strains did not grow upon thawing. In this study, 495 EMS-induced lethal bearing strains were analysed.  2.  szTl(l;X)  Spontaneous Lon males from the strain AF1, +; dpy-8 + unc-3/szTl (l;X)[+; + lon-2  +],  were crossed to dpy-5 unc-13; unc-3 hermaphrodites to construct the strain KR900, dpy-5 unc-13; + unc-3/ szT1(l;X)[+  +; lon-2 +]. The segregation from this strain was examined as a control  prior to mutagenesis. KR900 hermaphrodites segregated wild-type hermaphrodites (1009), Lon males (71), Dpy-5 Unc-13 Unc-3s (222 hermaphrodites and 19 males) and Unc-3s (50 hermaphrodites and 17 males). The absence of Dpy-5s (non-Unc-3s) and the low frequency of Unc-3s confirmed Fbdor and Deak's (1985) findings that the progeny of szT7 heterozygotes show a high degree of pseudolinkage between dpy-5 and unc-3. Lethal mutations would be recovered on the cross-over suppressed regions of both chromosome I and the X chromosome. The normal X chromosome was marked with unc-3 to make it easier to identify X-linked lethal mutations. The strain was designed so that the lethal mutations recovered from chromosome I would be linked to dpy-5 and unc-13, as in the sDp2 strains. In this way, recombination mapping would give both the right-left orientation of the lethal with respect to the markers and the relative distance. The right-left positioning would be determined in the same manner as for the sDp2-recovered lethals except that lethals could be recovered to the right of unc-13. These would give fertile Dpy-5 and Dpy-5 Unc-13 recombinants, but not Unc-13s. The screening strain (KR900) was treated with 1500 rads of gamma radiation and strains that segregated no fertile Dpy-5 Unc-13 Unc-3s were recovered. These strains could have carried lethal mutations in the cross-over suppressed regions of chromosome I or the X chromosome.  29 Table 2. Recovery of EMS-induced lethal mutations using  sDp2  EMS Dose  Screened  Lethals  Induction  01/1984  12 mM  1650  26  1.6  GG, AR  02/1985  12 mM  1933  32  1.7  AMH, JSK, AR  06/1985  17 mM  9356  134  1.4  DLB, LH, JSK,  Date  Participants  KMcN, AR, BR 01/1986  12 mM  2130  29  1.4  AMH, LH, AR, TS  01/1986  15 mM  8347  160  1.9  JB, AMH, LH, NM, AR, TS  01/1988  15 mM  3057  73  2.6  JB, JMcD, AMH, KMcK, AR, MZ  01/1988  15 mM  2663  51  2.1  AMH, KMcK, KP, AR, TS  01/1988  15 mM  TOTAL  2470  56  2.3  31606  552  1.7  AMH, AR, MZ  JB = Joseph Babity, DLB = David Baillie, JMcD = Jennifer McDowall, GG = Grant Gilmour, AMH = Ann Marie Howell, LH = Linda Harris, JSK = Jong Sun Kim, KMcK = Kim McKim, KMcN = Kelly McNeil, NM = Nasrin Mawji, KP. = Ken Peters, AR = Ann Rose, BR = Bruce Rattray, TS = Terry Starr, MZ = Monique Zetka  30 They were mated to dpy-5 unc-13/ + + males. Strains which gave no Unc-3 or Dpy-5 Unc-13 Unc-3 males but did give fertile Dpy-5 Unc-13 hermaphrodites after successful mating were assumed to carry an X-linked lethal mutation, and were discarded. Some strains gave wild-type and Dpy-5 Unc-13 males but no Dpy-5 Unc-13 hermaphrodites or Lon-2 or Unc-3 males after crossing. These did not behave like chromosome l or X-chromosome lethal strains and were not further characterized. Strains which gave Unc-3 and Dpy-5 Unc-13 Unc-3 males and fertile Dpy-5 Unc-13 hermaphrodites after crossing were assumed to carry a lethal mutation linked to dpy-5 unc-13.  Four of these (h546, h556, h563, h564) mapped to the right of dpy-5 and four (h549,  h550, h559, h565) mapped to the left of dpy-5.  3. hT1(l;V)  Phenotypically wild-type hermaphrodites of the genotype dpy-5 unc-13 +; +1 hTl(l;V)[ + + unc-29; +] were treated with EMS. Approximately 450 putative lethal mutations were recovered, from 3,000 screened chromosomes, of which 111 were analysed. In twelve strains the lethal mutation was lost due to recombination (i.e. they were outside of the boundaries of crossover suppression of hTl). Ninety-nine strains were crossed to Lon-2 males of the genotype hT2(l;lll)[dpy-5  +; +; 0]/szTl(l;X)[+  unc-29; +; lon-2].  In this way it was possible to determine  whether the lethal mutation in each strain was on chromosome 1 or chromosome V. If the lethal were on chromosome V, the resulting sz7~7 strain would have the genotype dpy-5 unc-13; +/ szT1[+  let-x/+;  +; lon-2] and would segregate many fertile Dpy-5 Unc-13s. The l^V^hTI  chromosome in the screening strain was marked with unc-29, as was the I^X^szTI chromosome in the strain used to rebalance the lethals. This was done so that the hTl/szTl  heterozygotes would  be Unc-29 and therefore easily distinguishable from the desired wild-type let-x dpy-5  unc-13/szTl  heterozygotes. The strain used to rebalance the lethals over sz7~7 also carried the translocation hT2(l;lll) marked with an induced dpy-5 allele. This was intended to keep the unc-29 on the 1^-X^szTI chromosome from crossing off. This was not completely effective, however, since two strains were created which contained an unmarked szTI chromosome in trans to h7"7.  31 Thirty-eight strains carrying lethal mutations on chromosome V were discarded. In thirteen strains the lethal mutation was on chromosome I, but outside of the cross-over suppressed region of szT7 on chromosome I. Two strains appeared to be chromosome I lethals when balanced over sz77, but were found to be heterozygous for hTl and sz7"7 which had lost the unc-29 marker. Forty-six strains were recovered which carried chromosome I lethal mutations in the cross-over suppressed region of szTl. The 111 strains analysed represented approximately one quarter of those recovered, and since they were chosen randomly they should be representative of one quarter of the chromosomes screened (750). This estimate is subject to any unknown bias that may have been placed on the lethals chosen for analysis. Forty-six lethal mutations in the szT7-balanced region of chromosome I were recovered giving an induction frequency of 6.1% (46/750).  32 B. Characterization of szT1  The reciprocal translocation szTl(l;X) was isolated and originally characterized by Fodor and Deak (1985). It was recovered as an X-ray induced, dominant cross-over suppressor on the X chromosome. It caused pseudolinkage of markers on chromosome I (unc-11, unc-38, dpy-5 andunc-15) and the X chromosome (unc-3 and unc-58).  In order to determine whether sz77 would  be a useful balancer for lethal mutations on chromosome I, its recombinational properties were further characterized. szT7 does suppress recombination on the left half of chromosome I. An interesting and novel observation was an increase in recombination adjacent to the chromosome 1 breakpoint in sz7"7 heterozygotes.  1. Recombination suppression Fodor and Deak (1985) observed reduced recombination in szTl(l;X) heterozygotes on the X chromosome, but did not investigate recombination suppression on chromosome l. Recombination frequencies were measured between chromosome I markers in controls and szT1 heterozygotes in order to determine the extent and severity of crossover suppression. No recombination was observed from let-362 (near the left end of chromosome I) to dpy-5 (Table 3). No recombination events occurred between dpy-5 and unc-13 in three different strains tested (Table 3).  2. Recombination enhancement Recombination frequencies were examined in regions of chromosome I to the right of the szT1(l;X) breakpoint. Recombination between unc-13 and unc-29 in sz77 heterozygotes was increased from 1.2 map units to 2.5 map units (Table 3). Since K. McKim observed no recombination in the unc-13 - let-88 interval, all of the increase must have occurred between let-88 and unc-29.  Normal recombination in this interval is less than 0.8 map units. By the same  33  Table 3. Recombination suppression and enhancement in szT1(l;X) heterozygotes  Genotype  let-362 dpy-5 unc-13/+  lei-362 dpy-5  +  +  unc-13;+/szTl  unc-11 dpy-5/ +  +  a  Wts  Recombinants  1406  149 Dpy-5 Unc-13  817 2463  b  pX100 (95% C.I.) 15.4 (13.0 - 17.9)  20 Unc-13  1.9(1.2-2.9)  0  0.0 (0 - 0.2)  33 Dpy-5  2.1 (1.8 - 3.3)  37 Unc-11 dpy-5;+/szT1  722  dpy-5 unc-13 unc-29/ + + +  2678  unc-11  0  0.0 (0 - 0.6)  34 Dpy-5  1.9 (1.3 - 2.6)  C  21 Unc-29 dpy-5  unc-13;+lszTl  dpy-5 unc-13 dpy-5  unc-29;+lszT1  unc-29;-hlszTl  1.2 (0.8 - 1.8)  c/  929  0  680  17 U n c ^ '  2.5 (1.5 - 3.8)  707  9 Dpy-5  2.5 (1.1 - 4.7)  0.0 (0 - 0.5) 0  14 Unc-29 dpy-5 unc-29 unc-75/+  dpy-5 unc-29  + +  unc-75;+/szT1  1767  1769  38 Unc-29 Unc-75  e  3.1 (2.1 - 4.1)  70 Unc-75^  5.8 (4.5 - 7.2)  26 Dpy-5  2.6(1.7-3.7)  e  66 Unc-29 Unc-75 234 Unc-75  f  12.9 (11.2 - 14.6)  100 Dpy-5 Unc-29 dpy-5  unc-75;+/szT1  581  45 Dpy-5 91 Unc-75  14.6 (10.5 - 19.0)  Table 3. (continued)  Genotype  Wts  Recombinants  pXlOO (95% C.I.)  2000 rads gamma radiation dpy-5 unc-13 unc-29/+  + +  2336  45 Dpy-5  3.0 (2.4 - 3.7)  C  51 Unc-13 Unc-29 28 Unc-29  c  a  From Table 7  b  From Table 5  The intervals assayed were c  dpy-5 - unc-13  d  unc-13 - unc-29  e  dpy-5 - unc-29  ^ unc-29 - unc-75  1.8 (1.2 - 2.5)  35 reasoning, all recombination between dpy-5 and unc-29 in the szTl heterozygotes (Table 3) must have taken place between let-88 and unc-29.  In the three different strains tested, the same level  of about 2.5 map units was measured. K. McKim mapped the sz77 breakpoint on chromosome I to this region. Thus this interval, immediately adjacent to the szT7 breakpoint on chromosome 1, increased in size three-fold in sz77 heterozygotes (from <0.8 map units to 2.5 map units). Recombination was also enhanced in the next interval to the right. The unc-29 - unc-75 interval increased in size two-fold from 5.8 map units to 12.9 map units (Table 3). In order to determine if the sz77-enhanced region is one that can be expanded by radiation, dpy-5 unc-13 unc-29/ + + + hermaphrodites were treated with 2000 rads of gamma radiation and the resulting recombination frequency was determined. In these experiments, the dpy-5 - unc-13 interval increased from 1.9 map units to 3.0 map units, an increase slightly less than that reported by Kim and Rose (1987) (1.6 map units to 3.9 map units). The unc-13 - unc-29 interval, which spans the position of the szT1 breakpoint, increased by 50% from 1.2 map units to 1.8 map units (Table 3??). The increase appears to be significant in comparison to the control. The increase indicates that the unc-13 - unc-29 interval is expandable by gamma radiation.  3. Lack of any interchromosomal effect Fodor and Deak (1985) reported increases in recombination on other chromosomes in their sz 77 strains. They tested an interval on each of chromosomes II, III and IV where they found increased recombination in sz 77 heterozygotes as compared to their controls. I measured recombination in four intervals on three autosomes and found no significant increases in recombination frequency (Table 4). Fodor and Deak (1985) reported that the dpy-10 - unc-4 interval on chromosome II increased from 2.5 map units to 5 map units in sz77 heterozygotes. I found no significant increase in that interval. It appears that sz77 does not affect recombination on autosomes other than chromosome I. It is possible that there was another mutation in the strain which Fodor and Deak were using which caused the apparent increases in recombination frequencies.  36 Table 4. Recombination on other chromosomes in szTI heterozygotes  Genotype  Wts  Recombinants  pX100(C.l.)  dpy-10 unc-4l+  + (II)  2484  47 Unc-4  2.8 (2.1 - 3.7)  dpy-10 unc-4l+  +;  1163  25 Unc-4  3.2 (2.2 - 4.5)  1693  4 Unc-36  0.4 (0.09 - 0.9)  731  1 Unc-36  0.2 (0.01 - 1.0)  1070  8 Unc-42  1.1 (0.4 - 2.1)  695  4 Unc-42  0.9 (0.2 - 2.2)  1186  19 Dpy-11  2.4 (1.4 - 3.6)  659  14 Dpy-11  3.2 (1.8 - 5.2)  +lszT1  unc-36 sma-2/ + + (III) unc-36 sma-2l+  +;  +/szTl  unc-42 sma-ll+  + (V)  unc-42 sma-V+  +;  +lszTl  dpy-11 unc-23l+  + (V)  dpy-11 unc-23/+  +;  +lszT1  37 C. Recombination Mapping  A genetic characterization of the hDf6 region should include determination of its size and its position on the chromosome with respect to other genes in the vicinity. Visible and lethal mutations inside of hDf6 and to both sides of it were positioned by recombination mapping. As well as positioning hDf6, the distribution of lethal mutations recovered by sDp2 could be derived from the mapping data. It was also important to test some of the initially recovered EMS-induced sDp2-recovered lethal mutations to ensure that they behaved predictably as single point mutations before undertaking large scale lethal screens.  1. Non-lethal mutations Previous to this study, several non-lethal mutations had been placed on the genetic map to the left of dpy-5 on chromosome I (Edgley and Riddle, 1987). However, some of these were not precisely positioned. In order to confirm the published positions of some of these genes, they were mapped by determining their recombination distance from dpy-5 (or unc-11). The data for the distances between dpy-5 and unc-11, unc-38, unc-57, unc-63 and unc-74 are presented in  Table 5 along with the distance between unc-11 and unc-73. Three-factor positioning of three genes was also carried out (Table 6). The positioning of him-1 between unc-57 and dpy-5 was confirmed, unc-73 had not previously been positioned with respect to unc-11.  It is now positioned between unc-11 and dpy-5. fer-7 failed to separate  from dpy-5 in 28 recombinants tested. Therefore it is tightly linked to, but not identified as being to the right or left of dpy-5.  2. Lethal mutations a. EMS-induced lethal mutations In order to characterize the distribution of an initial set of lethals, 59 EMS-induced sDp2recovered lethal mutations were mapped. Two-factor and three-factor data are reported in Table  38 Table 5. Two-factor mapping data for non-lethal mutations  Genotype  Wildtypes  Double Mutants  Recombinants  p (95% Cl.)  unc-11 dpy-5  2463  755  33 Dpy-5  37 Unc-11  2.1 (1.8 - 3.3)  unc-38 dpy-5  1951  568  .3 Dpy-5  3 Unc-38  0.2 (0.1 - 0.5)  unc-57 dpy-5  3703  1098  21 Dpy-5  26 Unc-57  1.0 (0.7 - 1.2)  unc-63 dpy-5  3586  1157  7 Dpy-5  7 Unc-63  0.3 (0.2 - 0.5)  unc-74 dpy-5  2053  631  23 Dpy-5  31 Unc-74  2.0 (1.5 - 2.5)  unc-11 unc-73  1072  298  ND = Not determined  7 Unc-11  ND  1.0 (0.5 - 1.9)  Table 6. Three-factor positioning data for non-lethal mutations  Genotype of Heterozygote  unc-57  +  +  him-1  dpy-5 +  Recombinant Chromosome  unc-57 unc-57 + +  +  (#)  +  (6)  +  (4)  dpy-5  (7)  him-1 dpy-5  (6)  him-1 +  unc-11  +  dpy-5  unc-11  +  +  (7)  +  unc-73  +  unc-11  unc-73  +  (5)  unc-11  (fer-7 +)  unc-11 (dpy-5 + ) +  ( + fer-7)  +  +  dpy-5  (13) (15)  Result  unc-57 (11/23) him-1 (12/23) dpy-5  unc-77 (5/12) unc-73 (7/12) dpy-5  unc-11 (28/28) (dpy-5 fer-7)  1 7. The recombination frequency for the dpy-5 unc-13 interval in a non-lethal strain is given in the first line of the table. These data were obtained from the strain KR236 (dpy-5 unc-13; sDp2  [+]).  Unc-13 hermaphrodites from KR236 were crossed to N2 males and the recombination frequency in the resulting heterozygotes calculated from the Unc-13 recombinant class. In a similar manner, recombination frequencies were obtained for each of the lethal strains (identified in Table 7 by the allele carried). The criteria for assignment of gene names is described in section lll.D.2 of this thesis. There were no anomalies in map position of alleles, i.e. all alleles of a gene mapped close to each other. For the lethal mutations which mapped to the left of dpy-5 (Section B of Table 7), Dpy-5 Unc-13 and Unc-13 recombinants were expected. This was the case for 30 of the EMS-induced mutations. The distance the lethal maps from dpy-5 was calculated from the Dpy-5 Unc-13 recombinant class. The lethals have been ordered by their distance from dpy-5, with the farthest left at the top of Table 7. Dpy-5 recombinants were not expected for this group of lethals since the cross-over chromosome (let-x dpy-5  + ) would carry the lethal and would not survive if  fertilized by a let-x dpy-5 unc-13 chromosome, in a few cases one cross-over chromosome (let-x dpy-5 +) was fertilized by another cross-over chromosome ( + dpy-5 unc-13). Twenty-one Dpy-5 recombinants of this type were observed. A second possible type of Dpy-5 recombinant could be the result of a double cross-over event ( + dpy-5 +/let-x dpy-5 unc-13).  None of the Dpy-5  recombinants listed in Section B of Table 7 were of the latter type, presumably the result of positive chromosomal interference. Section C shows the data for lethals mapping to the right of dpy-5. Gene names were assigned based on complementation analysis done by A. Rose. In these cases the distance to dpy-5 was calculated from the Dpy-5 recombinant class. Lethals are ordered by their distance from dpy-5, with the closest to dpy-5 at the top of the table. The distance from the lethal to unc-13 was also calculated from the Unc-13 class (data not shown) and no anomalies were observed. Section D shows the data for the lethals that were inseparable from dpy-5.  41 Table 7. Two-factor mapping data for EMS-induced lethals Gene  Allele  a  N  b  Progeny  0  DpyUnc  Dpy  Unc  p  d  (95% C. I.)  A: Control 30  dpy-5 unc-131 + +  6435  1422  52  51  0.016 (0.012 - 0.021)  B: EMS-induced lethals to the left of dpy-5: h86  9  2099  148  1  20  0.154 (0.130 - 0.179)  h93  9  2427  164  0  22  0.146 (0.122 - 0.171)  h92  8  1551  72  3  5  0.102 (0.079 - 0.129)  h99  8  2031  106  2  9  0.113 (0.092 - 0.136)  h108  10  2663  56  2  20  0.044 (0.034 - 0.057)  h129  10  2273  61  0  15  0.055 (0.042 - 0.070)  N. A.  h197  8  1719  40  1  19  0.049 (0.034 - 0.066)  let-360  h96  10  1716  39  0  10  0.046 (0.033 - 0.063)  let-357  h89.  8  1681  24  1  14  0.030 (0.020 - 0.044)  hi 32  10  2065  39  0  22  0.038 (0.027 - 0.051)  N. A.  hi 92  9  1791  21  2  11  0.026 (0.017 - 0.039)  let-356  h83  9  1356  15  1  16  0.024 (0.014 - 0.038)  let-351  h43  7  1041  12  0  16  0.023 (0.013 - 0.040)  let-366  h!12  9  1647  15  1  20  0.020 (0.012 - 0.031)  let-368  h121  10  1899  19  0  12  0.020 (0.012 - 0.031)  let-372  h!26  16  3413  31  4  24  0.018 (0.014 - 0.028)  let-369  h125  8  868  7'  0  4  0.016 (0.008 - 0.032)  let-354  h79  12  2577  16  1  24  0.013 (0.007 - 0.021)  h90  10  2068  14  1.  15  0.015 (0.008 - 0.022)  8  2125  13  0  17  0.012 (0.006 - 0.020)  let-362  lin-6  let-365  h267  '-,  42  Table 7. (continued) Progeny  Unc  (95% C. I.)  Allele  let-353  h46  6  872  4  0  12  0.009 (0.003 - 0.044)  him-1  h134  9  1719  4  0  18  0.005 (0.002 - 0.011)  let-361  h97  10  2263  5  0  19  0.004 (0.002 - 0.010)  h113  8  1441  0  0  11  0 ( 0 - 0.004)  h116  6  1131  5  1  8  0.011 (0.005 - 0.023)  10  2337  3  0  26  0.003 (0.001 - 0.007)  c  3  DpyUnc  Dpy  Gene  p  d  let-364  hi04 •  let-363  h98  8  1893  0  0'  10  0 ( 0 - 0.003)  hill  10  1883  3  0  14  0.003 (0.001 - 0.009)  h!14  10  2367  3  0  17  0.002 (0.001 - 0.007)  h131  9  2343  1  0  26  0.001 (0.0001- 0.005)  let-371  h123  16  2367  2  0  16  0.002 (0.0003- 0.006)  N. A.  h198  8  1696  2  0  15  0.002 (0.0004- 0.008)  let-352  h45  6  1360  1  0  12  0.001 (0.0001- 0.005)  let-359  h94  16  2891  1  0  23  0.001 (0.0001- 0.004)  G: EMS-induced lethalsto the right of dpy-5: let-376  h130  34  5768  0  1  44  0.0003(0.0001-0.002)  let-377  hllO  10  2088  0  2  16  0.002 (0.003 - 0.006)  let-378  hi 24  10  2156  0  2  14  0.002 (0.003 - 0.006)  let-388  h88  8  2051  0  4  14  0.004 (0.001 - 0.009)  let-379  h127  9  2043  0  4  10  0.004 (0.001 - 0.009)  iet-380  h80  9  1380  0  5  4  0.007 (0.003 - 0.016)  let-384  h84  9  1447  0  5  8  0.007 (0.003 - 0.016)  let-387  h87  8  1937  0  8  10  0.008 (0.003 - 0.016)  43 Table 7. (continued) Gene  Progeny  Allele*  0  DpyUnc  Dpy  Unc  p  d  (95% C. I.)  bli-4  h42  19  4116  0  22  21  0.011 (0.007 - 0.016)  let-382  h82  11  2345  0  13  2  0.011 (0.006 - 0.018)  let-391  h91  10  2332  0  13  15  0.011 (0.006 - 0.018)  let-381.  h107  10  2429  0  14  14  0.012 (0.007 - 0.019)  let-383  h!15  10  1972  0  12  4  . 0.012 (0.007 - 0.021)  let-386  hl17  10  2176  0  13  9  0.012 (0.006 - 0.020)  let-392  h120  8  1525  0  11  8  0.014 (0.007 - 0.025)  h122  10  2093  0  12  3  .0.012 (0.006 - 0.020)  9  2349  0  14  1  0.012 (0.007 - 0.020)  hW9  10  2145  0  25  8  0.023 (0.016 - 0.034)  M35  8  1840  0  14  5  0.015 (0.009 - 0.025)  10  2096  0  14  4  0.013 (0.008 - 0.022)  5  887  0  7  2  0.016 (0.007 - 0.032)  let-385  h85  let-389  h106  let-390  n 4 4  D: EMS•induced lethals inseparable from dpy-5: let-355  h81  31  7532  0  0  52  0 ( 0 -  let-367  M19  26  5356  0  0  35  0 ( 0 - 0.001)  let-370  hi 28  26  5267  0  0  45  0 ( 0 -  N. A.  h!94  9  1353  0  0  15  0 ( 0 - 0.005)  Lethal strain outcrossed to N2 males k Number of heterozygotes Calculated from wild types and recombinant progeny Frequency of recombination with dpy-5 N. A. = Not assigned  a  c  d  0.001)  0.001)  44  Figure 4. Distribution of sD 2-r covered essential genes. This histogram is based on the data P  presented in Table 7.  e  45 Figure 4 shows the distribution of essential genes inseparable from or to the left of dpy-5. The histogram displays the relative essential gene density per map unit for the left third of chromosome I. Genes on C. elegans autosomes form one large cluster per chromosome, leaving most of the rest of chromosome with a low gene density per map unit (Brenner, 1974). This distribution of essential genes shown in Figure 4 is very similar to the distribution of non-essential genes previously identified (Edgley and Riddle, 1987). Similarities in distributions of essential and non-essential genes on the other autosomes have been reported by others (Clark ef al., 1988; Rosenbluth ef al., 1988; Sigurdson, Spanier and Herman, 1984). b. Gamma-induced lethal mutations Gamma-induced lethal mutations were mapped in the same manner as the EMS-induced lethals. The results for eighteen lethal mutations are given in Table 8. The data in Table 8 is arranged similarly to that in Table 7. Another strain, carrying the mutation h330, gave fertile Unc-11 and Dpy-5 Unc-11 recombinants (1313 wild types, 11 Dpy-5 Unc-11s, 11 Dpy-5 Unc-13s, 28 Unc-11s, 15 Unc-13s; p = 0.021 for the unc-ll(h330)  - dpy-5 distance). h330 failed to  complement unc-11 (e47). Worms homozygous for the unc-11 (h330) dpy-5 unc-13 chromosome  arrested development at a late larval stage. It thus appeared that this viable unc-11 allele produced a synthetic lethal in combination with these markers. Since the unc-11 (h330) dpy-5 recombinants were viable, the synthetic lethality was probably due to unc-13.  I attempted to  construct an unc-11(e47) unc-13 double mutant strain to see if the e47 allele would behave in the same way. Hermaphrodites with the genotype unc-7 1 (e47) + unc-13/unc-l 1 (e47) dpy-5  +  segregated Unc-11s, Dpy-5 Unc-11s and late larval arrested Unc-13s (Unc-13s are quite paralyzed and would mask the Unc-11 phenotype). Thus, both alleles of unc-11 resulted in synthetic lethality with unc-13.  46 Table 8. Two-factor mapping data for gamma-induced lethals  Gene  Allele  N  Progeny  DpyUnc  Dpy  Unc  p (95% C. I.)  A: Gamma-induced mutations to the left of dpy-5 11  2293  174  1  17  0.166 (0.141 - 0.195)  h331  9  1797  18  0  13  0.020 (0.012 - 0.031)  hDf7  h54  6  1079  8  0  9  0.015 (0.006 - 0.028)  let-373  h70  10  1729  13  0  17  0.015 (0.008 - 0.025)  let-354  h72  9  1551  6  0  25  0.008 (0.003 - 0.017)  h327  9  276  1  0  3  0.007 (0.0003- 0.024)  let-361  h323  8  1235  3  0  8  0.005 (0.001 - 0.011)  let-363  h60  6  1487  3  0  16  0.004 (0.001 - 0.011)  h57  9  2191  2  0  16  0.002 (0.0002- 0.006)  h71  10  1663  2  0  h55  9  2096  1  0  18  0.001 (0.0001- 0.005)  h324  11  2761  1  0  14  0.001 (0.0001- 0.002)  h75  him-1  5-  0.002 (0.0003- 0.008)  B: Gamma-induced mutations to the right of dpy-5 h76  4  1539  •0  8  0  0.010 (0.004 - 0.020)  h59  6  348  0  2  2  0.012 (0.002 - 0.039)  C: Gamma-induced mutations inseparable from dpy-5 h56  10  517  0  0  8  0 ( 0 -  0.012)  h58  17  3664  0  0  9  0 ( 0  - 0.002)  h!67  7  1693  0  0  0  0 ( 0  - 0.004)  10  2197  0  0  0  0 ( 0  - 0.003)  hi 71  47  D. Complementation Analysis of Mutations Left of dpy-5  Essential genes were defined on the basis of extensive complementation analysis of mutations to the left of or inseparable from dpy-5. Several methods were used to divide the lethal mutations into groups for inter se complementation analysis. Mutations which had been positioned by recombination mapping or duplication mapping to the right of dpy-5, or which complemented sDf4, were not analysed. The remaining lethal mutations were tested for complementation with hDf6. The duplication hDp3 and the deficiency hDf7 were used to divide the hDf6 lethals into three groups for inter se complementation analysis. The lethal mutations from Table 7 which mapped to the left of (or inseparable from) dpy-5 were also subjected to inter se complementation analysis. Genes were assigned on the basis of EMS-induced mutations. Genes were not assigned using gamma-induced mutations because these could be small deficiencies.  1. Deficiencies The set of EMS-induced sDp2-recovered lethal mutations fell into three major categories: 1) 158 positioned with respect to dpy-5 by recombination mapping ; 2) 195 positioned to the right of dpy-5 by complementation to hDp!3 (A. Rose and J. McDowall, unpublished results); and, 3) 142 that had no prior positioning. The 85 lethal mutations from category one which mapped within 5 map units to the left of dpy-5 were tested for complementation with sDf4. In order to obtain some positional information for the unmapped lethal mutations (category 3), 90 were tested for complementation with sDf4. Strains heterozygous for sDf4 grow very slowly (even after back-crossing) and show low viability. A control cross was scored in order to determine the proper criteria for assessing complementation with sDf4. Heterozygous + +/ dpy-5 unc-13 males were generated by crossing N2 males to Unc-13 hermaphrodites from the KR236 strain. These males were mated to sDf4l bli-4 dpy-14 hermaphrodites and the outcross progeny were scored. Sixteen Dpy-5 and 208  48  Table 9. Deficiency mapping of EMS-induced sDp2-rescued lethals  Position  Number  In hDf6  Out hDfS  93  24  69  124  21  103  52  9  43  in sDf4 in hDp13  a  not tested with sDf4 or hDp73  not in sDf"4 not in hDp73 positioned right of dpy-5,^ far left of dpy-5, or  a  82  not tested  71  not tested  73  not tested  495  TOTAL  allelic to non-hD/6 gene  a = Data from A. Rose and J. McDowall b = Data from JB, JMcD, AMH, LHJSK, RK, KMcK, KMcN, DP, KP, BR, and TS. Initials are as listed for Table 2 with the addition of RK = Rohinish Kisun, DP = David Pilgrim.  49  wild-type males were observed giving a control segregation ratio of 1:13 (predicted 1:3). Comparable numbers of Dpy-5 and wild-type hermaphrodites were also present. The Dpy-5 hermaphrodites grew slowly and gave few progeny. A similar experiment was carried out with  sDf4 balanced over hT1. In this case the control segregation ratio was 1:6. In complementation tests with the lethal-carrying strains, more than 50 wild-type males were scored in order to ensure that apparent allelism was not the result of sD/4-heterozygqte inviability. Of the 175 lethal mutations tested, 93 failed to complement sDf4 (Table 9). Appropriate sDp2-recovered lethals were tested for complementation with hD/6 (Table 9). Since both hD/6 breakpoints are inside of sDf4, it was not necessary to test any lethal mutations outside of sDf4 for complementation with hDf6. Since the right breakpoint of hD/6 is to the left of dpy-5, it was not necessary to test any lethal mutations outside of hDp!3 for complementation with hD/6. A total of 54 EMS-induced sDp2-recovered lethal mutations failed to complement hD/6. The h77-recovered lethal mutations in the sz77-balanced strains were also tested for complementation with hD/6. Lon males from two of the 46 strains (h984 and hi003) failed to mate after several attempts and were not positioned. The remainder were tested regardless of their position in order to determine if there were mutations which could be rescued in a translocation heterozygote that would not be rescued by sDp2. Four of 44 tested failed to complement hD/6. The gamma-induced mutations recovered using sDp2 which mapped to the left of dpy-5 were tested with hD/6. Four of these (h54, h70, h72 and h327) failed to complement hD/6. The gamma-induced lethal mutations recovered with sz77 which mapped to the left of  dpy-5 were tested with hD/6. One, h549, failed to complement hD/6.  2. Lethal mutations The lethal mutations in Sections B and D from Table 7 (except h792, h194, h197 and h798) and all other lethal mutations which failed to complement hD/6 were subjected to inter se  50 complementation analysis. The lethals were divided into three groups for complementation analysis. Croup 1 mapped to the region from the left end of chromosome I to the left breakpoint of sDf4. Croup 2 included lethal mutations in sDf4 but not in hD/6 (the three mutations in section D of Table 7 were in sDf4). Croup 2 was further divided into two groups: 2a) included lethal mutations to the left of hD/6; and 2b) included lethal mutations to the right of hD/6. Group 3 included all the mutations in hD/6 and was further divided into three subgroups: 3a) contained lethal mutations complemented by hDp3 (see Section III.D.4); 3b) examined lethal mutations in hD/7 (see below); and, 3c) included lethal mutations outside both hDp3 and hD/7. Complementation tables indicating the direction of each cross can be found in Appendix I. The nine lethal mutations in group 1 define five complementation groups, four of which are represented by two alleles. Studies on h92 and h99 were published previously as alleles of let-358 (Howell et al., 1987). It was later determined that let-358 is the same gene as lin-6, previously described by Sulston and Horvitz (1981) who found that lin-6 is required for DNA synthesis and that mutations of this gene cause the absence of most postembryonic cell divisions. Since it was published first under the name //'n-6, that name takes precedence and the let-358 designation will no longer be used. The 18 lethals in Croup 2 define 13 complementation groups; three to the left of hD/6 and ten to the right, let-361 is represented by three alleles, and let-363 is represented by four alleles. It was reported by Howell et al. (1987) that h234 was an allele of let-372, a gene to the left of hD/6. It was later shown that h234 fails to complement hD/6, indicating the previous result was incorrect. h234 now defines the essential gene, let-373. The 54 EMS-induced sDp2-rescued lethal mutations in hD/6 (Croup 3) define nineteen complementation groups. The mapping of these genes with respect to duplication and deficiency breakpoints are reported in Tables 10 and 14, and shown in Figure 5. Ten genes are represented by more than one allele (Table 10). Table 11 compares the observed distribution of genes with multiple alleles to that predicted by the truncated Poisson formula (Beyer, 1976). This will be described in the Discussion.  51 The four hT7-recovered lethal mutations in hD/6 were complementation tested with the sDp2-recovered lethals. All four of them were allelic to complementation groups defined by the sDp2-recovered lethals (Table 12). The gamma-induced szTl-recovered lethal mutation in hD/6, h549, is an allele of let-354. The four gamma-induced sDp2-recovered lethal mutations in hD/6 were tested for complementation with each other and the EMS-defined complementation groups in hDf6. Three failed to complement single complementation groups: h70 is an allele of let-373; h72 is an allele of let-354; and h327 is an allele of let-504. h54 failed to complement alleles of six of the complementation groups in hD/~6 (Table 10). It was renamed hD/7. hD/7 complements lethal mutations outside of hD/6. hD/7 complements all of the visible mutations in hD/6 (see below). One gene uncovered by hD/7, let-507, is complemented by hDp3 (see below). This positions hD/7 to the right of unc-89, since unc-89 is to the left of hDp3.  3. Non-lethal mutations Some visible mutations are alleles of essential genes; for example bli-4 (Peters and Rose, 1988), rol-3 (Rosenbluth ef al., 1988), and unc-70 (Park and Horvitz, 1986) are known to have both lethal and visible alleles. It was thus appropriate to determine whether any of the visible mutations positioned by others to the left of dpy-5 on the genetic map (Edgley and Riddle, 1987) were allelic with essential genes identified in this study. The visible mutations were first tested for complementation with deficiencies. They were then tested for complementation with lethal mutations in the same region. unc-38, unc-63 and unc-67 fail to complement sDf4 but are outside of hD/6. unc-73 and unc-89 fail to complement hD/6. No visible mutation failed to complement hD/7. sup-11 and fog-1 are to the left of the left breakpoint of sDf4. No lethal alleles of these genes were found, even though the null phenotype of sup-11 is known to be lethal (Creenwald and Horvitz, 1982). unc-57 and unc-73 complemented all of the essential complementation groups in hD/6. Mutations in spe-11 result in male sperm-rescuable self-sterile hermaphrodites  Table 10. Duplication and deficiency mapping of essential genes in hD/6  Gene  let-356  Allele  h83  hDp3  hDf7  OUT  h501 h679  OUT OUT  h871 let-366  hll2 h265  OUT OUT  h4H h422 h505  OUT  h852 let-373  h234  OUT  OUT OUT  h573 let-501  h714  OUT  OUT OUT  h498 let-509  h521  OUT  OUT  hS22  OUT  h867  OUT  OUT  let-510  h740  OUT  OUT  let-511  h755  OUT  let-353  h46  OUT  IN  let-503  h313 h418  OUT  IN  h448  OUT  IN  let-504  h844  /  OUT  Table 10. (continued)  Gene  Allele  hDp3  hDf7  let-505  h426  OUT  IN  /ef-506  h300  OUT  IN  let-507  h439  IN  IN  let-354  h79  . I N  OUT  h90 h201 h267 h370 h390 h441 h482 h504 h508  IN  h693 h803 h809 h819 h841  IN  h863 h866 let-374  h251  IN  OUT  Table 10. (continued)  Gene  let-502  Allele  hDp3  hDf7  h392  OUT  h509 h732 h783 h835  IN  OUT  let-508  h452  IN  OUT  let-351  h43  IN  OUT  let-375  h259  IN  OUT  h391  For hDp3, IN indicates the lethal mutation is covered by the duplication and OUT indicates the lethal mutation is not covered by the duplication. For hDf7, IN indicates the lethal mutation fails to complement the deficiency and OUT indicates the lethal mutation complements the deficiency.  Table 11. Distribution of alleles compared to truncated Poisson distribution  Number of alleles  Observed number  0  Expected number 6  1  9  9  2  5  6  3  1  3  4  1  1  5  1  0  6  1  0  17  1  0  56  Table 12. Duplication and deficiency mapping of hT1 -recovered lethal mutations  Gene  Allele  hDp!3  hDp3  hDf6  h880  IN  OUT  OUT  h935  IN  OUT  OUT  h988  IN  OUT  OUT  h989  IN  OUT  OUT  hW04  IN  OUT  OUT  let-504  h888  IN  OUT  IN  let-366 .  h890  IN  OUT  IN  let-354  h934  IN  IN  IN  let-508  h995  IN  IN  IN  h883  IN  IN  OUT  h887  IN  IN  OUT  h889  IN  IN  OUT  h983  IN  IN  OUT  h936  IN  IN  OUT  h937  IN .  IN  OUT  h985  IN  IN  OUT  h987  IN  IN  OUT  h996  IN  IN  OUT  hlOOl _  IN  IN  OUT  h!006  IN  IN  OUT  h881  IN  OUT  h998  IN  OUT  hlOOO  IN  OUT  OUT  OUT  a  h878  Table 12. (continued)  Gene  a  a  Allele  hDp!3  hDp3  hDf6  h879  OUT  OUT  h882  OUT  OUT  h884  OUT  OUT  h885  OUT  OUT  h886  OUT  OUT  h931  OUT  OUT  h933  OUT  OUT  h938  OUT  OUT  h940  OUT  OUT  h986  OUT  OUT  h990  OUT  OUT  h991  OUT  h993  OUT  OUT  h994  OUT  OUT  h997  OUT  h999  OUT  h1005  OUT  IN  OUT  h1007  OUT  IN  OUT  IN  IN  OUT  OUT OUT  h932  OUT  hi 002  OUT  - Gene names were assigned only to the four mutations which failed to complement hD/6.  58  (L'Hernault, Shakes and Ward, 1988). This gene maps close to the left of dpy-5 but outside hD/6 (L'Hernault, Shakes and Ward, 1988; K. McKim, unpublished results). It was tested with the appropriate lethal mutations from Table 7 and found to complement them all. Three genes (unc-38, unc-63 and unc-74) whose mutant alleles confer resistance to the drug levamisole map to this region (Lewis et al. 1980). Lethal mutations were not tested for complementation to these genes because known null alleles are not lethal (Lewis et al. 1980), and are not synthetic lethals with unc-13.  unc-89 was not tested for complementation with lethals  because null alleles for this locus are not lethal.  a.  unc-11  unc-11 fails to complement sD/4 (Rose, 1980) but complements hD/6. unc-11 (e47)/sDf4 worms are fertile Dpy-5 Unc-11s. One of the sDp2-rescued lethal strains, KR776, gave sterile Dpy-5 Uncs when crossed to sD/4 and sterile Dpy-5 Unc-13s when crossed to hD/6. The phenotype of the former Dpy Uncs resembled that of Dpy-5 Unc-11 (Unc-11s have a distinctive backward jerking phenotype). KR776 failed to complement unc-11 (e47) for the uncoordinated phenotype but complemented the sterile phenotype. The fertile unc-11 mutation (h1008) was separated from the sterile mutation in hD/6 (h>452) by recombination. Thus, what at first appeared to be a sterile unc-11 mutation was in fact the result of two mutational events.  b. h/m-7 h/'m-7 was originally described by Hodgkin, Horvitz and Brenner (1979). The single EMSinduced allele, e879, causes a high incidence of males (Him phenotype) in the self progeny of homozygous hermaphrodites, the result of increased X-chromosome nondisjunction. e879 homozygotes produce about 20% male self progeny (Hodgkin, Horvitz and Brenner, 1979; Table 13). e879 also causes a reduction in recombination frequency on the X chromosome (Hodgkin, Horvitz and Brenner, 1979). h/m-7 maps outside hD/6 but within sD/4. e879/sDf4 heterozygotes produce significantly more males than e879 homozygotes (Table 13) suggesting that e879 is a  59  Table 13. X-chromosome nondisjunction in him-1 strains  Genotype  Hermaphrodites  Males  Frequency  e879/e879  764  181  19.2%  (16.9 - 21.4)  e879lsDf4  514  200  28.0%  (25.1 - 30.7)  e879/h55  325  144  30.7%  (27.0 - 34.2)  e879/h!34  499  176  26.1%  (23.1 - 28.9)  e879l +  3236  9  0.3%  ( 0.1 - 0.5)  3  95% C.I.  e879/e879;  sDp2  2154  27  1.2%  ( 0.8 - 1.7)  h134lh!34;  sDp2  1039  12  1.1%  ( 0.6 - 1.9)  991  2  02%  (0.04 - 0.7)  2480  3  0.1%  (0.03 - 0.3)  4663  5  0.1%  (0.04 - 0.3)  e879l + ; sDp2  + 1 + ;sDp2  + 1 +  b  :  a - e879, h55, and hl34 are alleles of him-1. e879 is on a chromosome marked with dpy-5. h55  and h134 are on chromosomes marked with dpy-5 and unc-13. sDf4 deletes him-1 and dpy-5.  b - Data from Rose and Baillie (1979).  60  hypomorphic allele of him-1 (Muller, 1932). Two of the sDp2-recovered lethal mutations were found to be alleles of him-1. him-1 (h55) was a gamma-ray induced mutation and him-1 (hi34) was an EMS-induced mutation. Both are recessive, late larval lethals. e879/h55 and e879lh134 heterozygotes produce frequencies of males similar to that of e879/sDf4 (Table 13). They enhance the e879 phenotype to the same degree as a deficiency, so they are likely amorphic alleles. Since these two alleles are recessive lethals and both are inviable over sDf4, analyses of effects on recombination and X-chromosome nondisjunction can not be done. Hodgkin, Horvitz and Brenner (1979) reported that him-1 (e879) had a dominant Him phenotype, producing 0.7 % males; this value was significantly different from the wild-type control frequency. I also observed a slight dominant Him phenotype for eS79 (Table 13). The dominant Him phenotype of him-1 was more pronounced when in a ratio of two mutant copies to one wild-type copy (i.e. in sDp2 strains). e879/e879; sDp2 and h134lh!34;  sDp2 gave approximately  1 % males (Table 13). The presence of sDp2 was not responsible for this effect since it did not increase X-chromosome nondisjunction (Table 13, lines 8 and 9). It is possible that the wild-type allele of him-1 on sDp2 does not produce as much product as a wild-type allele on a normal chromosome, or the locus is dosage sensitive.  4. Duplications A number of duplications of chromosome 1 were available which could be used to position essential genes (see Figure 1 in Materials and Methods). hDp3 is a duplication which breaks in the hD/6 region between unc-74 and unc-89.  hDp3 was used to position the genes in  hD/6" (Table 10). Approximately one third (7/19) of the essential genes in hD/6 are covered by hDp3.  Table 14 shows the results of complementation testing some of the essential genes to the left of dpy-5 from Table 7 with some duplications. The table is arranged with essential genes farthest from dpy-5 at the top, and duplications with breakpoints closest to dpy-5 at the left. The  61 first three genes in Table 14 are in sDf4 to the left of hD/6. The two-factor recombination mapping distances for these three genes (Table 7) indicated that they were to the left of hD/6, and the fact that they are not covered by hDp3 confirms this positioning. The next nine genes are uncovered by hD/6. Lethals in hD/6 which were only tested with hDp3 are listed in Table 10. The last seven genes in Table 14 are between dpy-5 and the right breakpoint of hD/6. hDp22 includes dpy-5 but has not been found to complement any genes to the left of dpy-5.  hDp25 separates into two groups those genes which are found both in hD/6 and in hDp3.  hDp25 includes let-351 but not let-354 or let-374 (Table 14). Figure 5 is a map of the sD/4 region which is drawn according to the results presented in Tables 10 and 14. The genetic distance from the left breakpoint of hD/6 to the left end of the chromosome is at least thirteen map units. The distance from the right breakpoint of hD/6 to dpy-5 is approximately 0.75 map units. Forty-six lethal mutations in hDp!3 (but not in hD/6) were tested with hDp3. Twenty-four mapped to the right of hD/6 and 22 were to the left of hD/6. The frequency of recovery of lethal mutations per map unit is fifteen-fold higher in the region close to dpy-5 than in the region to the left of hD/6. If the relative mutability per gene is roughly equal across the chromosome, this would reflect a fifteen-fold difference in gene density in these two regions. The hT7-recovered lethals balanced by sz 77 were tested with hDp13 (Table 12). Twentythree of the 42 tested mutations were complemented by hDp13, positioning them either to the left of dpy-5 or close to the right of it. Using hT7, the frequency of induction of lethals in hDp13 was 3.1% (23/750). Of the 19 mutations tested, twelve were covered by hDp3 and seven were not. One of the h77-recovered lethals which failed to complement hD/6 (h995) appeared not to be rescuable by hDp!3.  This was because the chromosome carried two new mutations. The  hDp13 strain was a sperm-rescuable sterile hermaphrodite which layed unfertilized oocytes. The Dpy-5 Unc-13 worms that were heterozygous for the lethal-bearing chromosome and hD/6 arrested at a late larval stage, but did not produce oocytes. During a strain construction to  Table 14. Duplication mapping of sDp2-rescued lethal mutations  Duplication (hDp) Gene  . 22  21  20  25  let-368  OUT  OUT  OUT  let-369  OUT  OUT  let-372  OUT  let-353  OUT  let-356  OUT  OUT  let-366 '  OUT  OUT  let-373  OUT  let-507  IN  let-354  OUT  OUT  OUT  let-374  OUT  OUT  OUT  IN  OUT  IN  IN  let-351 let-375  IN  IN IN  OUT  IN  IN  IN  him-1  OUT  OUT  let-352  OUT  IN  let-359  OUT  IN  IN  let-361  OUT  IN  IN  let-363  OUT  IN  IN  IN  let-364  OUT  IN IN  IN  let-371  OUT  IN  OUT  '  OUT  IN  IN  63  let-356 let-366 let-373 let-501 let-509 let-510 let-Si 1  let-368 let-369 let-372 unc-11  I  unc-73  I  let-353 let-503 let-504 let-505 let-506  unc-89  ( let-508 ) let-354 let-374 let-351 let-502 let-375 let-507  I  spe-11 let-352 let-359 let-361 let-363 let-364 let-371 unc-38 unc-63  I  unc-74  I  unc-57  I  him-1  I  dpy-5  I  I  sDf4  hDf6  hDf7  hDp2, 3, 5  hDp25  hDp20  hDp21  hDp22  0.5 mu  Figure 5. Genetic map of the unc-11 - dpy-5 region on chromosome I. Essential and visible genes are positioned based on deficiency (Sections II1.D.1, lll.D.2. and III.D.3) and duplication mapping (Section III.D.4 and K. McKim, unpublished results). The positions of the endpoints of sDf4 are based on data of Rose (1980) and J. McDowall (unpublished results), let-508 was not tested with hDp25.  64  rebalance the lethal-bearing chromosome under sDp2, the sperm-defective mutation was lost by recombination. This resulted in a chromosome with the h995 mutation (in hDf6) which was rescuable by hDp13.  5. Stage of arrest of lethal mutations in hDfS The stage of lethal arrest was examined for each of the mutations in hD/6 (Table 15). For strains where arrested Dpy-5 Unc-13 individuals were easily visible, some were picked and allowed to sit for several days to ensure they developed as far as possible. Their lengths were measured and compared to a growth curve of Dpy-5 Unc-13s from KR236 (dpy-5 unc-13; sDp2) prepared by C. Duncan (see Fig. 6). For strains where arrested Dpy-5 Unc-13s were not easily visible, heterozygous hermaphrodites (let-x dpy-5 unc-13/ + + +) were allowed to lay eggs for about eight hours and then removed. Plates were checked the next day to see if all the eggs hatched. If all had hatched the plates were checked later to see when the Dpy-5 Unc-13s arrested development. Since no record was taken of larval moults, the arresting stages are classified as early, mid-larval, late larval or sterile adult. Minor differences were seen between alleles of a given essential gene for which multiple alleles were recovered. For example, three alleles of let-502 arrest at a mid-larval stage but the fourth arrests at a later larval stage. All 17 alleles of let-354 appear to arrest development at the same mid-larval stage (probably L2). None of the EMS-induced lethal mutations result in embryonic lethality. Very few (3/54) of the mutants arrest development at an early larval stage. The largest proportion (34/54 if let-354 is included let-354, 17/34 if let-354 is not included) arrest at a mid-larval stage, even if let-354 alleles are not counted. Six of the sterile adult mutations are slightly leaky, meaning that some of the homozygous lethal Dpy-5 Unc-13 hermaphrodites produced a few progeny. Both alleles of let-375 and all three alleles of let-509 are leaky. The escaping progeny from let-509 never grow past hatching, indicating that these mutations also have a maternal affect. Dpy-5 Unc-13 hermaphrodites from three of these leaky mutant strains (Iet-375(h259), Iet-375(h391) and  Figure 6. Growth curve for Unc-13s and Dpy-5 Unc-13s. The genotype of the Unc-13s was dpy-5 unc-13; sDp2 (strain KR236) and the Dpy-5 Unc-13s were the progeny of KR236 Unc-1  * - Dpy-5 Unc-13 T-0 is at egg lay.  -+- Unc-13  66  Table 15. Stage of arrest of lethal mutations in hDf6  Gene  Allele  Stage of arrest  let-351  h43  mid-larval  let-353  h46  mid-larval  let-354  all (17)  mid-larval  let-356  all (4)  mid-larval  let-366  all (6)  mid-larval  let-373  both  early larval  let-374  h257  mid-larval  let-375  both  sterile adult (leaky)  let-501  h714  early larval  h498  mid-larval  h392  mid-larval  h509  sterile adult  h732  mid-larval  h783  late larval  h835  mid-larval  let-503  both  sterile adult  let-504  h448  sterile adult  h844  late larval  let-505  h426  late larval  let-506  h300  late larval  let-502  Table 15. (continued)  Gene  Allele  Stage of arrest  let-507  h439  sterile adult (leaky)  let-508  h452  late larval  let-509  h521  sterile adult (leaky)  h522  sterile adult (leaky)  h867  late larval (leaky)  let-510  h740  late larval  let-511  h755  early larval  68  Iet-507(h439)) also produced a few progeny after crossing to males. Since the number of outcross progeny was not much more than self progeny, they are probably not sperm-defective mutations. They also did not exhibit the usual phenotype of sperm-defective mutants, that is they did not lay many unfertilized eggs. These three leaky mutations were not leaky over hDftj (and hDf7 for let-507); i.e. over a deficiency their phenotype was more severe. This is in agreement with Muller's (1932) definition of a hypomorphic mutation, so they are probably hypomorphic mutations. None of the other sterile adult mutants gave any progeny after attempted mating to wild-type males. An egg count was done for hD/6 dpy-5 unc-13/ + + + hermaphrodites. One quarter of the eggs should have been homozygotes for hDf6. From a total of 70 eggs laid, 17 did not hatch. It thus appears that hD/6 is an embryonic lethal. A second smaller deficiency of this region, hDf7, is an early larval lethal. All eggs from hDf7 dpy-5 unc-13/ + + + hermaphrodites hatched but the Dpy-5 Unc-13s did not develop to the L2 stage. Thus, hDf7 homozygotes seem incapable of any post-embryonic growth. The earliest blocking mutant identified in hD/7 was mid-larval, most of the lethals blocked development at a late larval stage or produced sterile adults. The stage of arrest for homozygotes of either deficiency is earlier than that for homozygotes of any single mutation uncovered by them. It is possible that earlier blocking mutations are yet to be identified, or that the earlier block is the result of a cumulative affect of the loss of several gene products.  a. let-354  Seventeen alleles of let-354 were identified among the 495 EMS-induced lethal mutations recovered by sDp2.  One EMS-induced allele was recovered in the hT1 screen. Two gamma-  induced alleles were also recovered. h72 was isolated from a screen using sDp2, and h549 was isolated from a screen using szTI.  In total, approximately 1/3 (20/61) of all mutations uncovered  by hD/6 were found to be alleles of let-354.  69 1 used the recessive lethal mutation,  ct42,  recovered by P. Mains (personal  communication) for complementation tests with lethal mutations in hD/6. 1 found it to be an allele of  Because  let-354.  ct42  also had a temperature-sensitive, dominant, maternal-effect lethal  phenotype (P. Mains, personal communication), I tested ten sensitive sterility. If any of them were similar to  at 25 °C. Only one allele,  h482,  they would have been sterile at 25 ° C . Late  ct42,  larval Unc-13 hermaphrodites from lethal strains (i.e.  alleles for temperature-  let-354  let-354  dpy-5  unc-13;  sDp2)  were incubated  was sterile at 25 °C. Many eggs did not hatch and no larvae  grew past hatching. None of the other nine alleles tested were sterile at this temperature. hD/6 dpy-5  unc-13;  hDp31  was also tested and found to be fertile at the restrictive temperature. Since  hD/6 must represent the null phenotype of  the amorphic phenotype is not temperature  let-354,  sensitive. Furthermore, the null phenotype is recessive lethality, not a dominant, maternal-effect lethality. The Iet-354(ct42)  ct42  allele of  dpy-5;  let-354  0/szT1[+  can not be rescued by sDp2.  +;lon-2]  resulting wild-type males (Iet-354(ct42) unc-29;  sDp2  Lon males with the genotype  were mated to Dpy-14 Unc-29 hermaphrodites. The dpy-5  +  +1  +  + dpy-14  unc-29)  were mated to  dpy-5  (Unc-29) hermaphrodites. Except for recombinants, all wild-type hermaphrodite  progeny from that cross were Iet-354(ct42)  dpy-5  +  +1  +  + dpy-14  unc-29;  sDp2.  They were  allowed to self at 16 °C because of the dominant temperature-sensitive sterility caused by Iet-354(ct42).  Thirty of their wild-type hermaphrodite progeny were picked. They were expected  be of the parental genotype or Iet-354(ct42)  dpy-5/  Iet-354(ct42)  dpy-5;  sDp2  in a 2:1 ratio.  Twenty-one were found to be of the parental genotype by progeny testing. The other nine laid fertilized eggs which did not hatch. hD/6  dpy-5  unc-13;  hDp31  were fertile at 16 °C, indicating  this maternal-effect can not be the result of insufficient wild-type product but that the is as an antimorph (Muller, 1932).  ct42  allele  70  6. sDp2 has a lethal mutation in the hD/6 region hDf6 can not be maintained in a homozygous state with sDp2 in the same manner as the sDp2-rescued lethal strains. Two different strains were constructed in attempts to recover an hD/6 dpy-5 unc-13; sDp2 strain. Both hD/6 dpy-5 + unc-13/ + dpy-5 + unc-15; sDp2 and hD/6 dpy-5  + unc-13/ + dpy-5 dpy-14 +; sDp2 hermaphrodites segregated sterile adult Unc-13 hermaphrodites. They did not lay oocytes and were not male sperm rescuable. Any fertile Unc-13s recovered were shown to be recombinants by progeny testing. A possible explanation of this result was that hD/6 could not be maintained in a 2:1 ratio of the deficiency chromosome to wild type. This was shown not to be the case by constructing two HDf6/hDf6; I^X^szTI strains. In this strain one of the translocated chromosomes of sz7"7 (I^X^szTI) is being used as a duplication of the left half of chromosome 1. hD/6 dpy-5 unc-13; unc-3; l X szT1 L  L  was isolated as an aneuploid segregant from hD/6 dpy-5 unc-13; unc-3/ szTI  [+  + +; + lon-2]. hD/6 dpy-5 unc-13; I^X^szTI was constructed by crossing hD/6 dpy-5 + unc-13/  + + dpy-14 + males to dpy-5 dpy-14; I^X^szTI hermaphrodites and picking the appropriate segregant in the next generation. The viability of the two hDf6/hDf6;  I^X^szTI strains showed that  null alleles of all hD/6 genes can be recovered in a 2:1 ratio of the deficiency chromosome to wild type. I^X^szTI covers more of chromosome I than does sDp2, so it was possible that the difference in ability to rescue hD/6 was due to the difference in size. This was addressed by testing hD/6 for complementation by three other duplications, hDp31, hDp32 and hDp34 (see Figure 1b). These three duplications could also rescue hD/6 in Unc-13 strains (i.e. hD/6 dpy-5 unc-13; hDpx).  Since these three duplications are shorter than sDp2, the inability of sDp2 to  rescue hD/6 was not a function of its size. Thus, it is more than likely that sDp2 carries a lethal mutation in the hD/6 region. The mutation can not be in the hD/7 region since it is rescuable by sDp2.  No mutations were found in the hT7 or szT7 screens which failed to complement the  lethal mutation on sDp2. a small deletion.  It is not known whether the lesion on sDp2 affects only one gene, or is  71  DISCUSSION  A. Conclusions of Research  I have undertaken an analysis of essential genes in a region of the Caenorhabditis  elegans  genome. The region was defined by the deficiency, hDf6, on chromosome I. In order to do this analysis, it was necessary to develop a system for identifying, maintaining and characterizing a large number of lethal mutations. The usefulness of a free duplication as a balancer for recessive lethal mutations in a large autosomal region in C. elegans was examined. In this thesis I report the development of a system for lethal recovery and analysis using sDp2, a free duplication of the left third of chromosome l. Using this system, 495 EMS-induced recessive lethal mutations have been analysed. The advantages of this system are discussed below. However, as a result of my analysis, it was shown that sDp2 carries a lethal mutation in the hD/6 region. sDp2 could therefore not be used to recover alleles of every essential gene uncovered by hD/6. This analysis has demonstrated the feasibility of using a free duplication as a genetic balancer for autosomal lethal mutations. The most common lethal balancing systems in C. elegans made use of translocations that suppress recombination (Clark ef al., 1988; Rosenbluth et al., 1988). The first genetically well characterized translocation in C. elegans was eTl(lll;V)  (Rosenbluth and Baillie, 1981). It has been  developed for use as a mutagen test system (Rosenbluth, Cuddeford and Baillie, 1983), used to study the effects of gamma mutagenesis in C. elegans (Rosenbluth, Cuddeford and Baillie, 1983), and to characterize a large number of essential genes on chromosome V (Rosenbluth et al., 1988). ln order to compare the usefulness of the duplication to translocations, two translocations involving chromosome l were used to screen for and maintain lethal mutations. One translocation, szT1(l;X), has been extensively characterized by Fodor and Deak (1985) and McKim, Howell and Rose (1988). sz77 was used for recovering and maintaining gamma-induced lethal mutations on chromosome I. sz77 spontaneously generates X-chromosome fusion events at a  72  frequency of approximately 3%, resulting in strains which appear to be carrying a lethal mutation (McKim, Howell and Rose, 1988). Because many false lethal strains are recovered, szTl is not very good for lethal screening. A second translocation, hTT(l;V) (isolated by K. McKim) was shown to recover EMS-induced lethal mutations at a higher frequency than that of sDp2, suggesting that it may be a more efficient system for the recovery of lethal mutations. However, because the lethal mutations recovered using hT7 are more difficult to map and analyse than sDp2-recovered lethals, the sDp2 system was used for the bulk of the analysis. Lethal mutations inseparable from or to the left of dpy-5 were analysed and a saturation analysis of the small deficiency, hD/6, was begun. A combination of recombination mapping, deficiency mapping, duplication mapping and inter se complementation analysis led to the identification of 36 new essential genes (listed in Sections B and D of Table 7 and in Table 10) in the region from dpy-5 to the left end of the chromosome. Nineteen of these are in hD/6. A maximum of three quarters of the essential genes in this deficiency have now been identified.  B. Estimate of the Number of Genes in hD/6  In an attempt to identify all of the genes in the small deficiency hD/6, nineteen essential genes were identified. These mutations were recovered by screening 31,606 mutagenized chromosomes. Ten of these genes were represented by more than one allele. An estimate of the number of genes yet to be identified can be gained from a statistical analysis of the allele frequencies. The Poisson distribution can be applied only if the following two assumptions are met; the probability of mutating is the same for each gene in the region, and the ability to detect mutants is the same for all genes (Lefevre and Watkins, 1986). The latter assumption holds if one considers only essential genes. The mutability of genes is far from being equal (Meneely and Herman, 1979; Hilliker, Chovnick and Clark, 1981; Lefevre and Watkins, 1986). The mutability of genes in hD/6 is obviously not the same since let-354 has seventeen alleles while others have none. The high mutation frequency indicates either that the gene is large or is extremely sensitive  73  to mutation. Two other frequent mutational targets in C. elegans,  unc-22  and unc-54, have been  cloned and shown to be large genes (Moerman, Benian and Waterston, 1986 for unc-22; MacLeod, Karn and Brenner, 1981 for unc-54). Barrett (1980) analysed published data on lethal analysis in Drosophila  melanogaster  and showed that it did not fit well to a Poisson distribution,  except when the sample sizes were small. Figure 7 shows a comparison of the observed distribution of mutations in hDf6 to that predicted by the truncated Poisson calculation. It is obvious that the observed data does not fit the Poisson distribution exactly; there are more genes with multiple alleles than expected. The data may be interpreted as representing three classes of genes; the genes with one to three alleles are genes with average mutability, the genes with four to six alleles are a different class of genes with much higher mutability, and let-354 with 17 alleles is a rare extremely mutable gene. Meneely and Herman (1979) proposed the use of a truncated Poisson formula which decreases the emphasis on highly mutable genes [f=  (1-e~ -me~ )/(l-e~ ), m  m  m  where f is the fraction of genes with more than one allele and m is the  average number of alleles]. Of the nineteen genes in hDf6, ten were represented by more than one allele, giving a value of 0.53 for f. By calculation from the above formula, m = 1.35. By examination of a table of cumulative terms for the Poisson distribution (Beyer, 1976) the proportion of genes with no alleles is 0.25. Thus, a maximum of 75% of the essential genes in hDf6  have now been identified. From these calculations one might conclude that a minimum of  six genes are yet to be identified, giving a total minimum estimate of 25 essential genes in hD/6. This estimate is likely an underestimate because, as Muller (1929) pointed out, the zero allele class will always be underestimated if the mutability is not constant. The frequency of recovery of mutations should be about the same across the entire sDp2 region, since the screen places no bias on the position of the lethal mutations recovered as long as a wild-type allele is present on the duplication. If complete complementation analysis were carried out for any other interval in the sDp2 region, the degree of saturation attained should be the same as for the hD/6 interval. The set of EMS-induced sDp2-recovered mutations should therefore represent 75% of all essential genes in the entire sDp2 region. My results are similar  74  Figure 7. Distribution of lethal mutations in hD/6. The observed distribution of lethal mutations is compared to that predicted by the truncated Poisson formula.  10 N  u  m b e r  o f G e  n e s  2  3  MJUL 4  5  6  i 7  8  9  10  Mutations/Gene Observed  i  i  i  11 12 13 14 15 16 17  Expected  75 to those obtained by Clark ef al. (1988). They predicted they had identified 65% of the essential genes in the sD/2 region after screening 20,000 mutagenized chromosomes (D. V. Clark, personal communication). Meneely and Herman (1981) estimated they had identified one half of the genes in a region of the X chromosome after screening 11,500 mutagenized chromosomes. Their recovery rate was higher because they used a higher dose of EMS. Examination of the EMS dose response curve prepared by Rosenbluth, Cuddeford and Baillie (1983) indicates that double hit events are much more likely to be recovered at higher doses. These complicate the analyses by underestimating the total number of genes. To estimate what fraction of the genome is contained in hD/6, the size of hD/6 must be determined (see Figure 5). The size of hD/6 can be estimated from recombination mapping data of genes known to be in or out of it. The right boundary is between h/m-7 (0.5 map units from dpy-5, Table 7) and unc-57 (1.0 map units from dpy-5, Table 5). Duplication mapping confirmed that let-368, let-369 and let-372 are to the left of the left breakpoint of hD/6. The left boundary must therefore be somewhere between the position of these genes and let-356 and let-366, which are in hD/6. According to the recombination mapping data in Table 7, this is approximately 2 map units from dpy-5. Thus, hD/6 extends from 0.75 map units left of dpy-5 to 2.0 map units left of dpy-5. It is therefore about 1.25 map units across. This is 1/240 of the size of the total genome (300 map units). It is possible to derive an estimate of the number of genes in hD/6 by molecular studies if an estimate of its physical size is made. Its position in the genome must be taken into account for this estimate because the genetic map of C. elegans exhibits clustering of genes on the autosomes, first recognized for nonessential genes by Brenner (1974). It has since been shown to be the case for essential genes also (Section lll.C.2.a; Rose and Baillie, 1980; Sigurdson, Spanier and Herman, 1984; Clark et al., 1988; Rosenbluth ef a/., 1988). Greenwald ef al. (1987), Prasad (1988) and Starr ef al. (1989) have shown that this clustering is due to suppression of recombination in the apparently gene dense regions. Kim and Rose (1987) showed that succeptibility to gamma radiation induced map expansion changes across the chromosome I gene  76  cluster. It is believed that regions of the genome which exhibit gamma radiation induced map expansion have a suppressed frequency of recombination per base pair relative to the genomic average. The differential sensitivity of regions of the gene cluster to map expansion indicates that the amount of DNA per map unit changes across the gene cluster. Estimates of the DNA/map unit value based on cloning in the  dpy-5  - unc-13  region (Starr  the data of Kim and Rose (1987). For example, in the  dpy-14  et al.,  1989) correlated well with  - unc-13  interval, Starr  et al.  (1989)  found the DNAVmap unit value to be 4.5 times the genomic average; Kim and Rose (1987) found that recombination in the same region increased four-fold after radiation treatment. hD/6 is on the left edge of the gene cluster on chromosome I. Kim and Rose (1987) showed that the frequency of recombination in the unc-11 - dpy-5 interval increased 1.5 fold after radiation treatment. If the correlation between the amount of DNA/map unit and the increase in recombination frequency after radiation treatment extends to the left of  dpy-5,  one would predict  that the hD/6 region would contain 1.5 times the genomic average amount of DNA/map unit. Since the size of the genome has recently been re-estimated to be 100 Mb (J. Sulston, personal communication) the average size of a map unit would be 333 Kb. This would make hD/6 approximately 625 Kb. Both Prasad (1988) and Starr ef al. (1989) estimated there is one coding region for every 15 Kb in  C. elegans.  Heine and Blumenthal (1986) estimated 20 kb per gene.  This would mean there would be room for 30 to 40 genes in hD/6. Five non-essential genes and nineteen essential genes have been mapped to the hD/6 region. Including the six essential genes predicted to be present by the Poisson analysis, this would give a total of 30 genes in hD/6. Thus, the estimates of gene number in hD/6 based on genetic or molecular analyses are in reasonable agreement. The largest fraction of genes in the hD/6 region is most likely the essential genes. Of the identified genes in hD/6, approximately four-fifths are essential. There may be, however, several genes which would not be identified in screens for lethal mutations. Genes with subtle null phenotypes or redundant members of gene families would not be identified. The true fraction of  77  essential genes could be determined when the number of coding regions identified by molecular analysis can be compared to the number of genes identified by mutational analysis. Several different methods have been used to estimate the total gene number in C. elegans.  Brenner's (1974) original estimate of 2,000 essential genes was based on the  induction of X-linked recessive lethal mutations. A similar method was used by Baillie (D. L. Baillie, personal communication as cited in Moerman and Baillie, 1979) to arrive at an estimate of 4,000 essential genes. Rogalski (1983) obtained an estimate of 5,700 essential genes based on a saturation analysis of the small deficiency of chromosome IV, sDf2. Most recently, Clark er al. (1988) arrived at a minimum estimate of 3,500 essential genes by comparison of induction rates in a well characterized small region to a large region. However, they caution that their estimate may be an overestimate. If the essential gene density in the hD/6 region were representative of the entire genome, an estimate of approximately 6,000 (25 x 240) total essential genes would be obtained. However, the hD/6 region is probably not representative of the genome. hD/6 is on the left edge of the gene cluster on chromosome I, so the gene/map unit density is expected to be higher than the genomic average (see Fig. 4). If the density is 1.5 times the genome average, this would give an estimate of approximately 4,000 essential genes. This is in good agreement with the estimates of Baillie (Moerman and Baillie, 1979) and Clark et al. (1988).  C. Forward Mutation Rates in C. elegans C. elegans researchers often cite Brenner (1974) for his estimate of the average forward mutation rate per gene, that is, the rate of inducing a mutation in any given gene causing a mutant phenotype. Brenner had recovered 26 visible mutations from a screen of 736 mutagenized chromosomes. He estimated his target size as 77 genes, which was the total number of visible genes identified in screens of unmeasured sample sizes. Using these values, Brenner arrived at an estimate of 5 x 10"^ mutations per gene using 0.05 M ethyl methane sulfonate (EMS). The sample size of mutagenized chromosomes on which this estimate was based was quite small. The estimate of the target size was also inaccurate; many more than 77 genes in  78  C. elegans are now known to be mutable to a visible phenotype (Edgley and Riddle, 1987). The identification of visible mutations can be subjective because some mutant pheriotypes are quite subtle. The majority of genes in G elegans are essential genes, so an estimate based on mutability of visible genes may not represent the average gene. For these reasons, there is room for considerable error in Brenner's estimate. The data presented in this thesis can also be used to derive an estimate of the average forward mutation rate per gene in C. elegans.  The data for let-354 is omitted from this calculation  because its mutation rate is at least ten times higher than that for an average gene. From approximately 31,000 mutagenized chromosomes, 37 lethal mutations were recovered in hDf6. The target was the 24 essential genes predicted to be present in this region. Using these values, the average forward mutation rate per gene is 5 x 10"^. This is ten-fold lower than Brenner's estimate. According to the dose response curve for EMS mutagenesis published by Rosenbluth, Cuddeford and Baillie (1983), the difference in EMS dose used would result in a two-fold lower induction frequency. This would lower Brenner's estimate to 2.5 x 10" mutations per gene, 4  which is still five-fold higher than the estimate predicted by my data. The sample size of screened chromosomes on which this estimate is based is 40 times larger than Brenner's. The large sample allowed the recovery of mutations in less mutable targets. The identification of lethal mutations is also less subjective than identification of visible mutations. The induction frequency for lethal mutations based on the hDf6 analysis is in good agreement with previously published studies. If a two-fold correction is made for the EMS dose, the data on induction of X-linked lethal mutations from Meneely and Herman (1981) can be used to calculate an estimate of 5 x 10"-' mutations per gene. This estimate is identical to the estimate based on the results of this study. Using the data from Clark ef al. (1988) based on an estimated 20,000 screened chromosomes (D. Clark, personal communication), the induction frequency is calculated to be 6 x 10"^ mutations per gene. The forward mutation rate can be used to estimate the number of chromosomes which need to be screened when attempting to identify all of the genes in a region, or all of the genes  79  of a given type in the genome. In order to have 95% confidence that any gene has been identified, the number of chromosomes screened must be three times that required to recover one mutant allele. For the average gene in C. elegans, this would require the screening of approximately 60,000 chromosomes. If Brenner's estimate had been correct, only 12,000 mutagenized chromosomes (0.015 M EMS) would need to be screened to identify an average gene in the genome. If this were true, the three large studies (Meneely and Herman, 1981; Clark ef al., 1988; this thesis) would have identified all of the genes in the regions studied. Kemphues, Kusch and Wolf (1988) relied on Brenner's estimate of average forward mutation frequency to formulate the important conclusions of their study on maternal-effect lethal mutations in C. elegans. They identified 17 strict maternal-effect genes on chromosome II, 13 of which had one or two alleles and four of which had four or more alleles. They argued that if the data was subjected to a Poisson analysis, a minimum estimate of 37 strict maternal-effect genes would be obtained. They then calculated that this would give an average forward mutation rate of 8 x 10"^ mutations per gene. Since this was so much lower than Brenner's value, they did not believe that it could be correct. They then used only the genes with four or more alleles (with an< induction frequency of 5 x 10~ ) to make their conclusions. They assumed that since all of these 4  genes had more than three alleles, they had identified all of them. This led to an extremely low estimate of the predicted number of strict maternal-effect genes in the C. elegans genome. If they had known that the average forward mutation rate is actually only 5x10"^ per gene, they would have predicted the existence of many more strict maternal-effect genes.  D. Stage of Developmental Arrest  The stage of lethal arrest was determined for the nineteen essential genes represented by 54 EMS-induced sDp2-rescued mutations in hDf~6 (Table 15). The stage of lethal arrest was determined in strains which also carried mutations in dpy-5 and unc-13.  The presence of these  mutations may affect the phenotypes of the lethal mutations. The dpy-5 and unc-13 mutations  80  may alter the viability or the stage of lethal arrest of the lethal mutations. For example, viable mutant alleles of unc-11 are synthetically lethal in combination with unc-13 mutations (Section IIl.C.2.b). Some of the other apparent lethal mutations may be synthetically lethal with the marker mutations, but this was not tested. Most alleles of a gene, for which more than one allele has been identified, arrest development at approximately the same developmental stage. This finding has also been described by others (Shannon ef a/., 1972; Rose and Baillie, 1980; Meneely and Herman, 1981; Rogalski, Moerman and Baillie, 1982; Rosenbluth ef al., 1988). Shannon ef al. (1972) proposed that most variability between alleles is due to leakiness of later arresting mutations. Clark ef al. (1988) and Leicht and Bonner (1988) report, more variability between alleles than the other studies mentioned above. In some cases Clark ef al. (1988) showed that the later arresting mutations blocked development at an earlier stage when heterozygous with a deficiency, indicating that they are indeed hypomorphs (Muller, 1932). Leicht and Bonner (1988) proposed that some of the apparent variability between alleles of the essential genes they analysed was due to linked second site mutations. Since the lethal mutations described in this thesis were generated using a low dose of EMS, variability between alleles is not likely due to linked second site mutations. The most variability in stage of arrest was found for alleles of let-502, which ranged from mid-larval to sterile adult. The dissimilarities in phenotype could represent varying severities of mutant alleles, or differing mutated functions which are required at different times in development. There is considerable variability in the stage of arrest between genes in hDf6.  If the  earliest blocking allele is taken for genes with more than one allele, three genes are required for development past the first larval stage, seven are required for development past the mid-larval stage, six are required to mature to adulthood from the late larval stage, and three are required for fertility of adult hermaphrodites. No mutants which blocked development during embryogenesis were found. This is not surprising since C. elegans embryonic development is strongly maternally influenced. Maternal-effect lethal mutations were not studied in this thesis research, although it is possible to recover them using the sDp2 lethal screening system.  81 Figure 8 shows the distribution of arrest stages for the essential genes in hDf6. They have been divided into three regions; a) is the region left of hDf7, b) is the hDf7 region, and c) is the region right of hDf7. The results show that within a small region many of the lethals arrest at the same stage, and that this stage differs from region to region. For example, the three genes which are required for development past the first larval stage are clustered in the leftmost region of hDf6.  Individuals with mutations in five of the six essential genes in hDf7 arrest development as  late larvae or sterile adults. Individuals with mutations in four of the six essential genes in the rightmost region of hD/6 arrest development at a mid-larval stage. However, because of the small number of genes analysed, these differences are not statistically significant. The clustering of genes which are required at the same developmental stages may represent some type of clustering of genes with related functions. Analyses of stages of arrest for lethal mutations have been carried out in studies of different regions of the C. elegans genome. Figure 9 shows the distribution of arrest stages for three small regions; a) the hD/6 region, b) the unc-15 region (Rose and Baillie, 1980), and c) the unc-22 region (Clark ef al., 1988). The frequencies of arrest stages vary dramatically depending on the region of the C. elegans genome under analysis. Only three of the nineteen genes in hD/6 appear to be required for development past the first larval stage. Rose and Baillie (1980) reported that worms carrying mutations in 13 of the 1.6 essential genes in the region around unc-15 (I) arrested as early larva. Mutations in one third of the essential genes listed in Clark et al. (1988) from the unc-22 (IV) region arrest as early larva. The hD/6 region has the smallest proportion (0.16) of early larval lethals. The proportion of genes with mid-larval arrest phenotypes is approximately one third for both the hD/6 and unc-22 regions. The proportion of genes with late larval arrest phenotypes is much larger in the hD/6 region [0.32, (6/19)] than in the unc-22 region [0.17, (5/30)]. The proportion of genes with sterile adult mutations is somewhat larger in the hD/6 region [0.16, (3/19)] than the unc-22 region [0.10, (3/30)] (Clark ef al., 1988). The hD/6 region is similar to some other genomic regions for the proportions of mid-larval and adult sterile mutations recovered, but differs in the scarcity of early larval mutants recovered. Meneely and Herman  82  Figure 8. Distribution of lethal arrest stages for genes in the hD/6 region.  a) hOI6 - left  Early  Mid  Late  Adult  Stage of Arrest  b) hDI7  e '  2  o Q e n  1  0  Early  Mid  Late  Adult  Stage of Arrest  c) hDI6 - right  s I  Early  Mid  Late  Stage of Arrest  Adult  83 Figure 9. Distribution of lethal arrest stages for genes in three regions of the G elegans genome. The data for b) was from Rose and Baillie, 1980, and for c) was from Clark ef al., 1988.  a) note 8  IB  7  N u  m  6  b  Q  -  3  • Early  •  Mid  1  11 n Late  Adult  Stage of Arrest  b) unc-15 region  14  Early  Mid  Late  Adult  Stage of Arrest  c) unc-22 region  12  Early  Mid  Late  Stage of Arrest  Adult  84 (1981), Sigurdson, Spanier and Herman (1984) and Rosenbluth ef a/. (1988), studied regions of the genome which were much larger. If the data from their large regions is divided into small intervals, regional differences in stages of lethal arrest become evident. None of the sterile adult mutations in hD/6 appear to be due to defective sperm production. L'Hernault, Shakes and Ward (1988) carried out an extensive screen for spermdefective mutations on chromosome I. They identified eleven genes, none of which were in hD/6. From a Poisson analysis of the number of alleles recovered, they estimated there may be as few as three more spe- (sperm-defective) genes on chromosome I. There could be more than that since they did not screen enough chromosomes to saturate the region. Alleles causing temperature sensitive dominant or normal recessive lethal phenotypes have been recovered for one of the genes in hD/6, let-354.  Paul Mains isolated mutations in a screen  for temperature sensitive dominant maternal-effect embryonic lethal mutations (P. Mains, personal communication). Three of these failed to complement hD/6. They also had a non-conditional recessive mid-larval lethal phenotype. All three failed to complement each other for the recessive phenotype. One of these, cf42, is an allele of let-354. let-354 is a very large mutational target for both the recessive and dominant phenotypes. One third of all recessive lethal mutations in hD/6 described in this thesis are alleles of let-354. Three of eight dominant temperature-sensitive maternal-effect embryonic lethal mutations recovered in a screen of the whole genome were alleles of let-354 (P. Mains, personal communication). hD/6 represents the null phenotype for let-354 since the entire gene must be within hD/6 because genes on both sides of it are included in hD/6. hD/6 does not have a dominant temperature-sensitive maternal-effect lethal phenotype, so this could not be the null phenotype of any gene hD/6 deletes. Furthermore, most . temperature-sensitive mutations are the result of alterations in protein structure, not loss of function. The recessive mid-larval lethal phenotype is the most likely the null phenotype of . let-354. One wild-type allele on a duplication can rescue two null alleles (hD/6), but can not rescue cf42. Thus, cf42 is an antimorphic mutation because it interferes with the function of the wild-type gene product. The mutant cf42 gene product does not retain the function necessary for  85 larval development at any temperature, but at the restrictive temperature somehow causes dominant maternal-effect embryonic lethality. One of the sDp2-recovered EMS-induced alleles of let-354, h482, showed a dominant temperature-sensitive maternal-effect embryonic lethal phenotype. This phenotype was not as extreme as ct42; some of the eggs hatched at 25 ° C but none of the larvae grew after that. With cf42, most eggs do not hatch at 20 ° C and none hatch at 25 ° C (P. Mains/personal communication). h482 was rescuable by the duplication and the strain was quite fertile at 20 °C. The mutation is similar to cf42 but its antimorphic behaviour is not as strong. Perrimon ef al: (1986) showed that in Drosophila melanogaster many maternaleffect lethal mutations are rare hypomorphic or antimorphic alleles of ordinary zygotic essential genes. Recently, the same has been shown to be true in C. elegans (Kemphues, Kusch and Wolf, 1988). The alleles of let-354 which show maternal effects are probably of this type; the let-354 gene product may not normally play a role in embryogenesis.  E. Comparison of Balancing Systems  This thesis describes the effectiveness of the free duplication, sDp2, as a lethal rescue system. Lethals are induced in a strain designed to allow detection of new mutations in the F2 generation. The screening strain can be easily maintained owing to the tight linkage between the markers used, unc-15 and unc-13 (Waterston, Fishpool and Brenner, 1977; Rose and Baillie, 1980) and between the markers and the sDp2 breakpoint (Rose, Baillie and Curran, 1984). The progeny of over 31,000 individuals have now been screened using this strain with no indication of alteration in its genetic composition. Lethals could be isolated on the chromosome marked with unc-13 or unc-15. In this thesis lethals on the unc-13 chromosome were analysed because Unc-13s grow and mate better than Unc-15s. It is extremely important for a large scale lethal analysis to have all of the mutations linked to common markers for complementation analysis. The lethal mutations are maintained in a strain with an Unc-13 phenotype that segregates only Unc-13 progeny, all of which have an identical genotype. Thus these strains are self-  86 maintaining over any number of generations and transfer of the lethal-bearing individuals does not require prior familiarity with the phenotype of the strain. Furthermore, C. elegans strains can be frozen, making the maintenance of thousands of lethal-bearing strains feasible. Only lethal strains with the wild-type allele supplied by sDp2 are fertile since the lethal strains are maintained with two normal homologues carrying the same lethal mutation. This strictly defines the region from which lethal mutations will be recovered to that duplicated by sDp2. It is a straightforward procedure to produce males for complementation analysis when using this system since all the males resulting from an outcross have the lethal-bearing chromosome. Those that carry the duplication do not interfere with the analysis (Rose, Baillie and Curran, 1984). This simplifies and speeds up complementation testing by eliminating the need to maintain male strains that are heterozygous for the lethal or to identify the lethal-bearing males. A result of this research was the discovery of a lethal mutation on sDp2 in the hD/6 region. It could have occurred spontaneously during stock maintenance. Rosenbluth, Cuddeford and Baillie (1983) measured the spontaneous lethal mutation rate over a 40 map unit region of the C. elegans genome and determined that 0.06% of hermaphrodites per generation would carry a new spontaneous lethal mutation somewhere in this region. Since sDp2 is approximately one half of that size, 0.03% of hermaphrodites per generation woud be expected to carry a lethal mutation on the duplication. sDp2 was generated at a dose of 7000 R of gamma radiation (Rose, Baillie and Curran, 1984). This high dose often causes multiple hits (Rosenbluth, Cuddeford and Baillie, 1983) so the lethal event was likely generated at the same time as the duplication. Only one deficiency, hD/7, was recovered in a screen for gamma-induced lethal mutations. Since only mutations which are rescued by wild-type alleles on sDp2 would be recovered, any deletions with one lethal breakpoint outside sDp2 would be eliminated. sDp2 carries a lethal mutation in the hD/6 region, so deficiencies which span that site would not be recovered. For these reasons, sDp2 is not a good balancer to use for recovery of gamma-induced deficiencies.  87 Translocations in C. elegans suppress crossing-over from their breakpoint to one end of the chromosome (Rosenbluth and Baillie, 1981; Herman, Kari and Hartman, 1982; Ferguson and Horvitz, 1985; Rosenbluth, Cuddeford and Baillie, 1985; McKim, Howell and Rose, 1988). They are useful as balancers of recessive lethal mutations in their cross-over suppressed regions (Rosenbluth, Cuddeford and Baillie, 1983, 1985; Clark et a/., 1988; McKim, Howell and Rose, 1988; Rosenbluth et al., 1988). Three translocations of chromosome I that have been analysed (sz77, hTl and h7"2) suppress crossing over from the left end of the chromosome up to the breakpoint (McKim, Howell and Rose, 1988). Two of these translocations were used as balancers in this study; szTl(l;X) (Fodor and Deak, 1985) and hTl(l;V).  The boundary of suppression of  crossing over is between unc-13 and unc-29 for both of these translocations, so the region they balance on chromosome I extends approximately one map unit farther to the right than the region balanced by sDp2.  szTl was used in a screen for gamma-induced lethals and hT7 was used in a  screen for EMS-induced lethals. One of the problems with using a translocation as a balancer in screens is that mutations are recovered which are linked to, but outside of, the boundaries of cross-over suppression. This is less of a problem with sz7"7 screens because of the increased rate of recombination adjacent to the breakpoint on chromosome I. Lethal mutations outside of the szT7 breakpoint will be lost by recombination three times faster than with a translocation like h7"7 which does not have this increase in recombination frequency. Another difficulty with using the translocations as balancers is that the males can not be used directly for complementation analysis. The efficiency of recovering lethals that map under hDp!3  (Figure 1 b) is 1.4% using sDp2  and 3% using hTl. The apparent two-fold difference in induction frequency may be due to three factors. Firstly, it may be accounted for in part by the lethal mutation on sDp2.  This mutation  could be a small deficiency such that mutations in several genes may not be recovered. Secondly, some mutations may not be recovered in a 2:1 ratio to wild type. This was proposed by Howell et al. (1987) for sDp2 and shown to be true by Rosenbluth et al. (1988) for another region. Some lethal mutations may have a dominant influence on viability which would be enhanced in the  88  duplication system because the mutant allele is present in two doses. Since hD/6 can be rescued by a duplication which does not carry a lethal mutation in that region (e.g. hDp31), null alleles of all genes in hD/6 should be rescuable in a 2:1 ratio. Lastly, since recombination is not reduced in the presence of sDp2 (A. M. Howell, unpublished results; K. McKim, unpublished results), some of the lethals would be lost due to recombination in sDp2 screens. When the Unc-13s are set up to create the lethal-bearing strain, the probability that the lethal mutation would have crossed off one of the dpy-5 unc-13 chromosomes is directly proportional to the distance from the lethal mutation to dpy-5. Since often more than one Unc-13 is transferred onto one plate, the probability of finding fertile Dpy-5 Unc-13s in the next generation is even higher. This is not a serious problem for lethals in the hD/6 region because they are tightly linked to dpy-5. Approximately 2% of hD/6 lethals would be lost by recombination (i. e., only one or two lethals in hD/6 would have been lost). McKim, Howell and Rose (1988) showed that no recombination occurs in hT7 heterozygotes from the left end of chromosome I to the breakpoint. No lethal mutations would be missed in hTl screens due to recombination. The gamma-induced mutations recovered by sz77 included the deficiency hD/6. This deficiency would not have been recovered by sDp2 because of the lethal mutation carried by sDp2. This was the only deficiency found among 24 analysed lethal mutations on chromosome I recovered from the sz77 screen (K. McKim, unpublished results). This frequency is much lower than that obtained by Rosenbluth, Cuddeford and Baillie (1985) who found that six of 35 analyzed gamma-induced lethal mutations recovered using e7"7 were deficiencies of chromosome V. It seems to be difficult to isolate deficiencies of chromosome I, but there is no obvious explanation for this difference in rate in this part of the genome. The existence of sD/4 proves that there are no haplo-insufficient regions within the two map units left of dpy-5. in summary, the induction frequency of lethal mutations may be higher with the translocation screens than with the duplication screen. However, lethal mutations are recovered in two genomic locations in translocation screens, whereas with a free duplication all mutations recovered are in a strictly defined region. Complementation analysis is much simpler with lethals  89 from the duplication screen. For a large scale project, the duplication system is preferable because the resultant lethal-bearing strains are much easier to work with.  F. Comparison of Mapping Systems  The lethal mutations recovered in the duplication and translocation screens spanned the left third of chromosome I. This region includes at least 17 map units. It was important to develop ways of positioning large numbers of mutations efficiently, especially to divide the large region into smaller regions for complementation analysis. Three different methods were used to map mutations in this study; recombination mapping, deficiency mapping and duplication mapping.  1. Recombination mapping Recombination mapping using the sDp2 system is relatively simple. If the Unc-13s from the sDp2-rescued lethal strains are mated to wild-type males, all of the outcross wild-type hermaphrodites carry the lethal-bearing chromosome. Many of them also carry sDp2 and are not of use in recombination mapping. These can be selected against because they develop more slowly, are thinner and paler in appearance, and have a mildly uncoordinated phenotype. The lethal-bearing chromosomes are marked with two visible mutations. This allows for the right-left positioning of lethal mutations with respect to dpy-5 to be determined. The two marker phenotypes (Dpy-5 and Unc-13) are easily visible. This increases the accuracy of recombination mapping since even rare recombinants are not likely missed. However, Unc-13s are less viable than wild types, so their frequency may sometimes be under-estimated. This would result in an underestimation of the distance from unc-13 to dpy-5 or the lethal mutation. It is not as straightforward a procedure to obtain the correct heterozygous hermaphrodites for mapping experiments with the szTI system as with the sDp2 system. The lethal-bearing strains are wild-type in phenotype and so progeny resulting from crossing can not be visibly distinguished  90  from self progeny. The outcross progeny will be of two types; let-x dpy-5 unc-13/ + + + and + + +lszT1.  The types can only be distinguished in the next generation by observing the  progeny (the desired heterozygotes segregate no Lon-2 males). Many mutations were positioned by recombination mapping to one of four regions; right of dpy-5, close to the left of dpy-5, far to the left of dpy-5, or inseparable from dpy-5. Recombination mapping is not an efficient method for dealing with large numbers of mutations because brooding the heterozygotes and scoring the progeny are labor intensive tasks. If the lethal mutation is close to dpy-5, large numbers of progeny may need to be scored in order to find an informative recombinant. Recombination mapping gives a right-left position with respect to a marker, but the distance from the marker is not precisely determined unless a large number of recombinants are scored. 2. Deficiency mapping Deficiency mapping has been used very effectively in C. elegans to position large numbers of lethal mutations (Sigurdson, Spanier and Herman, 1984; Rosenbluth ef al., 1988). A series of overlapping deficiencies for chromosome I was not available. Deficiency mapping is easier than recombination mapping because it is not necessary to score large numbers of progeny to obtain results. The lethal mutation is positioned with respect to the two endpoints of the deficiency. For some deficiencies (e.g. hDf6) the heterozygous males will mate successfully and so many lethal stocks can be tested at one time, as opposed to recombination or duplication mapping where males must be obtained from each lethal strain. However, some deficiency strains are quite sick and difficult to work with. This may depend on the genotype of the strain. For example, the BC700 strain (sD/4 + +1 + bli-4 dpy-14) is very difficult to maintain, whereas the KR1069 strain (sDf4/hT1) grows well. The reduced viability of deficiency heterozygotes must be taken into account when determining the criteria for interpreting deficiency mapping results. Even in the hTT strain, the number of sDf4 heterozygotes is reduced from what would be expected after crossing.  91  sDf4 is the largest deficiency of the sDp2 region presently available. If a lethal mutation fails to complement sDf4, its position is well established. If it complements sDf4, however, it could be either to the left or the right of sDf4.  3. Duplication mapping The duplications of chromosome I isolated by K. McKim (McKim and Rose, 1988) have proven to be very useful for positioning lethal mutations. Since the duplications form a linear array with one fixed end, they give left-right positioning of lethal and visible mutations with respect to each other. One minor problem with duplication mapping is that some of the duplications are extremely unstable in the oocyte germ line. Duplications of chromosome I seem to be easier to recover than deficiencies, and are probably more useful for mapping genes. Meneely and Nordstrom (1988) discussed a possible problem associated with duplication mapping. They found that X-chromosome duplications increase gene expression of genes not under the duplication. If a mutation is a hypomorph, then increasing its expression may rescue the phenotype. It would then appear to map under the duplication even though it really did not. Their results may be specific for the sex chromosome and due to dosage compensation. If autosomal duplications behave in a similar manner, then special consideration should be given to mapping of leaky mutations. There is no evidence that the chromosome I duplications affect expression of genes they do not duplicate. In summary, duplication mapping is an efficient way to map large numbers of lethal mutations. The positioning of the lethals can be much more precise than that determined by recombination mapping.  92  G. Summary of Findings/Contributions  The following is a list of the findings of this thesis. 1.  sDp2 was shown to be an effective balancer for the region of chromosome I which it duplicates.  2.  In a collaborative laboratory effort, 31,606 chromosomes were screened for EMS-induced sDp2-rescued lethal mutations and 495 mutations were rescued.  3.  sDp2 was found to carry a lethal mutation in the hD/6 region.  4.  szT1(l;X) was used to recover gamma-induced lethal mutations. It was shown that sz77 is an effective balancer for the left half of chromosome 1, although it is not good for isolation of lethal mutations.  5.  It was discovered that szT7 heterozygotes have an increased frequency of recombination adjacent to the chromosome I breakpoint. This has not been found for any other translocation in C. elegans.  6.  hT1(l;V) was shown to be useful for screening and balancing chromosome I lethal mutations, but was difficult to use for large scale complementation analysis.  7.  h/m-7 was shown to be an essential gene.  8.  unc-11 was shown to have a synthetic lethal interaction with unc-13.  9.  let-354 was shown to be a large mutational target. Approximately one third of all lethal mutations recovered in the hD/6 region were alleles of let-354.  10.  A new chromosome I deficiency, hD/7, was identified and characterized.  11.  The average forward mutation rate per gene in C. elegans using 0.015 M EMS was estimated to be 5 x 10"^ mutations/gene. This value is five-fold lower than that estimated by Brenner (1974) even after a correction for EMS dosage is applied.  12.  Nineteen of the estimated minimum of twenty-five essential genes uncovered by the deficiency, hD/6, were identified. This led to the prediction of the existence of 4,000 essential genes in the C. elegans genome.  93  13.  The genes in hD/6 were positioned into five regions based on the breakpoints of hDf6, hD/7, hDp3 and hDp25.  14.  The stages of lethal arrest were determined for the mutations in hD/6. They appear to show some clustering of genes with similar stages of arrest. There were fewer early arresting lethal mutations and more adult sterile mutations in hD/6 than in other studied genomic regions.  H. Proposals for future experiments  The following is a list of experiments which could be done to further this work. I.  Lethal mutations could be tested with the amber suppressor tRNA mutants sup-5 (Waterston and Brenner, 1978) or sup-7 (Waterston, 1981). If suppressible alleles were found they would indicate that the mutation was in a protein coding region, and that the mutant probably represented the null phenotype. These alleles could then be tested with the tissue-specific amber suppressors (Hodgkin, 1985; Kondo, Hodgkin and Waterston, 1988) to assay for tissue-specific expression of the let- gene.  2.  The tissues in which the essential genes must be expressed could be determined by genetic mosaic analysis using a modification of the method described by Herman (1984).  3.  The phenotypes of the arrested Dpy-5 Unc-13s (or the arrested Unc-13s which result from complementation testing with a duplication which does not cover the let- gene) could be visually inspected by Nomarski optics to check for gross anatomical differences. Their muscle structure could also be examined by polarized light microscopy or antibody staining techniques.  4.  All possible combinations of let-354 alleles could be tested by inter se complementation analysis in order to look for intragenic complementation.  94 5.  A fine structure analysis of let-354 could be carried out. An approach similar to that described by Bullerjahn and Riddle (1988) for fine structure mapping alleles of an essential gene in the presence of a free duplication of the region could be employed.  6.  The saturation analysis of hDf6 could be continued. More lethal mutations could be recovered either in hTl screens, or in screens similar to sDp2 screens but which use a duplication known to be able to rescue hDf6. New lethal mutations in hDf6 could be used to identify the site of the lethal lesion on sDp2.  7.  The genes in hDtS could be positioned with respect to the breakpoints of more duplications. Right-left positioning of lethals with respect to visible markers in the same zone could be undertaken.  8.  A screen for transposon-induced mutations in visible or essential genes in hDf6 could be carried out. These mutations could be used to clone specific genes and obtain overlapping cosmids clones around them. They could also be used to align the position of a gene with respect to available cosmid clones.  9.  A set of overlapping cosmids clones is known to include the unc-38 gene. If the cloned DNA extends into the hDf6 region, cosmids to the right of the breakpoint could be used in transformation experiments to assign genes to cloned coding regions.  I. General Discussion  One current approach to studying eukaryotic genes and genome organization is to obtain the complete DNA sequence of a genome. It has been proposed that sequencing the human genome would be a worthwhile project. If this project is undertaken, it will result in the accummulation of incredible amounts of sequence data. The assignment of functions to the DNA sequence will be an extremely challenging task, however, since little genetic information is available for aligning the human physical and genetic maps. Coding regions can be identified  95 using the DNA sequence, but unless there is some identity with known sequences in the computer data bases, it will be difficult to determine the functions of the predicted proteins. The complete sequencing of the genome of a model organism with a good system for genetic analysis may be a more reasonable project. Genomic descriptions will be most valuable if they are complete at both the genetic and molecular levels. Individual genes and genomic organization are being studied in many metazoan systems; especially in nematodes, Drosophila, and mice. The short generation time for C. elegans makes it an especially favourable organism for genetic analysis, and the small genome size makes it amenable to molecular analysis. The work presented in this thesis provides another region of the C. elegans genome which is well characterized genetically. Twenty-four genes have been mapped to five zones defined by breakpoints of chromosomal rearrangements. The complete alignment of the physical and genetic maps of this region could be accomplished using cosmid clones and genetic transformation techniques. All of the genes in hDf6 should be contained in a set of 20 cosmid clones with an average insert size of 30 kilobases. The rearrangement breakpoints would divide the cosmids into five small sets which could be used to microinject and rescue appropriate lethal mutations. All of the coding regions could be sequenced and compared to known sequences to look for important functional regions. The major benefit of this model system is that genetic analysis can be used to analyze the functions of the coding regions. For example, it is possible to determine in which tissue type or developmental stage the gene product is required. The tissue type requirement can be determined genetically using tissue-specific amber suppressor tRNA mutations (Hodgkin, 1985; Kondo, Hodgkin and Waterston, 1988) or genetic mosaic analysis (Herman, 1984). The tissue type requirement can be determined biochemically by in situ hybridization of labelled antibodies to proteins or labelled probes for mRNA. Temperature shift experiments using temperaturesensitive mutations can determine the developmental time during which the gene product is required. Northern blots containing stage-specific RNA preparations could also be used to determine at which developmental stage a gene is expressed. Genetic analysis can also be used  96 to determine how gene products interact physically or in developmental pathways. Analyses of epistatic interactions between different mutations have been used to predict models for several developmental programs in C.  elegans;  including dauer larvae formation (Riddle, Swanson and  Albert, 1981), sex determination (Hodgkin, Doniach and Shen, 1986), specification of the vulval cell lineages (Ferguson, Sternberg and Horvitz, 1987), and development of the hermaphroditespecific neurons (Desai ef al., 1988). C. elegans  is presently one of the best metazoan organisms available for analysis of  genomic organization. The molecular map of the  C. elegans  genome is almost complete and the  genetic map is quite extensive. A fraction of genes are known to be evolutionarily conserved between nematodes and mammals (one example is the unc-86 gene product which shares identity with the human ocf-7 and ocf-2 transcription factors (Finney, Ruvkun and Horvitz, 1988)), so information gained by studying genes and genome organization in C. information which is applicable to other systems (including humans).  elegans  will provide  97  BIBLIOGRAPHY ANDERSON, P. and S. BRENNER, 1984 A selection for myosin heavy chain mutants in the nematode Caenorhabditis elegans. Proc. Natl. Acad. Sci. USA 81: 4470-4474. BACHMANN, B. J., 1983 Linkage map of Escherichia coli K-12, edition 7. Microbiol. Rev. 47: 180230. BAILLIE, D. L, K. A. BECKENBACH, and A. M. ROSE, 1985 Cloning within the unc-43 unc-31 IV interval of the Caenorhabditis elegans genome. Can. J. Genet. Cytol. 27: 457-466. BANERJI, J., L. OLSON and W. SCHAFFNER, 1983 A lymphocyte-specific cellular enhancer is located downstream of the joining region in immunoglobulin heavy chain genes. Cell 33: 729-740. BELL, G. I., L. ]. DeGENNARO, D. H. GELFAND, R. J. BISHOP, P. VALENZUELA and W. J. RUTTER, 1977 Ribosomal RNA genes of Saccharomyces cerevisiae: Physical map of the repeating unit and location of the region coding for the 5S, 5.8S, 18S and 25S ribosomal RNAs. J. Biol. Chem. 252: 8118-8125. BERGET, S. M., A. J. BERK, T. HARRISON, and P. SHARP, 1977 Spliced segments at the 5' termini of Adenovirus-2 late mRNA: A role for heterogeneous nuclear RNA in mammalian cells. Cold Spring Harbor Symp. Quant. Biol. 42: 523-529. BEYER, W. H. (Editor), 1976  CRC Handbook  of Tables for Probability and Statistics. CRC Press,  Cleveland, Ohio. BINGHAM, P. M., T. B. CHOU, I. MIMS and Z. ZACHAR, 1988 On/off regulation at the level of splicing. Trends Genet. 4: 134-138. BREITBART, R. E., A. ANDREADIS and B. NADAL-GINARD, 1987 Alternative splicing: A ubiquitous mechanism for the generation of multiple protein isoforms from single genes. Annu. Rev. Biochem. 56: 467-495. BRENNER, S., 1974 The genetics of Caenorhabditis  elegans. Genetics 77: 71-94.  BULLERJAHN, A. M. E. and D. L. RIDDLE, 1988 Fine-structure genetics of ama-1, an essential gene encoding the amanitin-binding subunit of RNA polymerase II in Caenorhabditis elegans. Genetics 120: 423-434. CHALFIE, M. and M. AU, 1989 Genetic control of differentiation of the Caenorhabditis touch receptor neurons. Science 243: 1027-1033.  elegans  CHOVNICK, A., W. GELBART, M. McCARRON, B. OSMOND, P. CANDIDO and D. BAILLIE, 1976 Organization of the rosy locus in Drosophila melanogaster. Evidence for a control element adjacent to the xanthine dehydrogenase structural element. Genetics 84: 233-255. CLARK, D. V., T. M. ROGALSKI, L. M. DONATI and D. L. BAILLIE, 1988 The unc-22(IV) region of Caenorhabditis elegans: Genetic analysis of lethal mutations. Genetics 129: 345-353. CORTESE, R., D. MELTON, T. TRANQUILLA and J. D. SMITH, 1978 Cloning of nematode tRNA genes and their expression in the frog oocyte. Nucleic Acids Res. 5: 4593-4611.  98  COULSON, A., J. SULSTON, S. BRENNER, and J. KARN, 1986 Towards a physical map of the genome of the nematode Caenorhabditis elegans. Proc. Natl. Acad. Sci. 83: 7821-7825. COULSON, A., R. WATERSTON, J. KIFF, J. SULSTON, and Y. KOHARA, 1988 Genome linking with yeast artifical chromosomes. Nature 335: 184-186. COX, C. N., S. CARR, ]. M. KRAMER and D. HIRSH, 1985 Genetic mapping of Caenorhabditis elegans collagen genes using DNA polymorphisms as phenotypic markers. Genetics 109: 513-528. CROW, E. L. and R. S. GARDNER, 1959 Confidence intervals for the expectation of a poisson variable. Biometrika 46: 441-453 DESAI, C , G. GARRIGA, S. L. MclNTIRE and H. R. HORVITZ, 1988 A genetic pathway for the development of the Caenorhabditis elegans HSN motor neurons. Nature 336: 638-646. EDGLEY, M. L. and D. L. RIDDLE, 1987 Caenorhabditis elegans. ln Genetic Maps 1987, Vol. 4, Edited by S. J. O'BRIEN. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. EMMONS, S. W., L. YESNER, K. RUAN, and D. KATZENBERG, 1983 Evidence for a transposon in C. elegans. Cell 32: 55-65. FERGUSON, E. L. and H. R. HORVITZ, 1985 Identification and characterization of 22 genes that affect the vulval cell lineages of the nematode Caenorhabditis elegans. Genetics 110: 17-72. FERGUSON, E. L, P. W. STERNBERG and H. R. HORVITZ, 1985 A genetic pathway for the specification of the vulval cell lineages of Caenorhabditis elegans. Nature 326: 259-267. FILES, J. G. and D. HIRSH, 1981 Ribosomal DNA of Caenorhabditis elegans. J. Mol. Biol. 149: 223-240. FINNEY, M., G. RUVKUN and H. R. HORVITZ, 1988 The C. elegans cell lineage and differentiation gene unc-86 encodes a protein with a homeodomain and extended similarity to transcription factors. Cell 55: 757-769. FODOR, A. and P. DEAK, 1985 The isolation and genetic analysis of a C. elegans translocation (szTI) strain bearing an X-chromosome balancer. J. of Genet. 64: 143-157. FRIEDMAN, D. I., M. J. IMPERIALE and S. ADHYA, 1987 RNA 3' end formation in the control of gene expression. Annu. Rev. Genet. 21: 453-488. GILLIES, S. D., S. L. MORRISON, V. T. Ol and S. TONEGAWA, 1983 A tissue-specific enhancer element is located in the major intron of a rearranged immunoglobulin heavy chain gene. Cell 33: 717-728. GREENWALD, I., A. COULSON, j. SULSTON and ]. PRIESS, 1987 Correlation of the physical and genetic maps in the lin-12 region of Caenorhabditis elegans. Nucleic Acids Res. 15: 22952307. GREENWALD, I. and H. R. HORVITZ, 1982 Dominant suppressors of a muscle mutant define an essential gene of Caenorhabditis elegans. Genetics 101: 211-225. HEDGECOCK, E., ). CULOTTI, D. HALL and B. STERN, 1987 Genetics of cell and axon migration in Caenorhabditis elegans. Development 100: 365-385.  99  HEINE, U. and T. BLUMENTHAL, 1986 Characterization of regions of the Caenorhabditis X chromosome containing vitellogenin genes. J. Mol. Biol. 188: 301-312.  elegans  HENDRIX, R. W., J. W: ROBERTS, F. W. STAHL and R. A. WEISBERC, 1983 Lambda II. (Cold Spring Harbor, New York: Cold Spring Harbor Laboratory). HERMAN, R. K., 1978 Crossover suppressors and balanced recessive lethals in elegans. Genetics 88: 49-65.  Caenorhabditis  HERMAN, R. K., 1984 Analysis of genetic mosaics of the nematode Caenorhabditis Genetics 108: 165-180.  elegans.  HERMAN, R. K., C. K. KARI and P. S. HARTMAN, 1982 Dominant X-chromosome non-disjunction mutants of Caenorhabditis elegans. Genetics 102: 379-400. HERSHEY, A. D. (Editor), 1971 The Bacteriophage Spring Harbor Laboratory).  Lambda. (Cold Spring Harbor, New York: Cold  HILLIKER, A. J., A. CHOVNICK and S. H. CLARK, 1981 The relative mutabilities of vital genes in D. melanogaster. Drosophila Inform. Serv. 56: 64-65. HILLIKER, A. J., S. H. CLARK, A. CHOVNICK and W. M. GELBART, 1980 Cytogenetic analysis of the chromosomal region immediately adjacent to the rosy locus in Drosophila melanogaster. Genetics 95: 95-110. HOCHMAN, B., 1971 Analysis of chromosome 4 in Drosophila melanogaster. methanesulfonate induced lethals. Genetics 67: 235-252.  II. Ethyl  HODGKIN, J. A., 1985 Novel nematode amber suppressors. Genetics 111: 287-310. HODGKIN, J. A., T. DONIACH and M. SHEN, 1986 The sex determination pathway in the nematode Caenorhabditis elegans: Variations on a theme. Cold Spring Harbor Symp. Quant. Biol. 50: 585-593. HODGKIN, J: A., H. R. HORVITZ and S. BRENNER, 1979 Nondisjunction mutants of the nematode C. elegans. Genetics 91: 67-94. HORVITZ, H. R., S. BRENNER, j. HODGKIN and R. HERMAN, 1979 A uniform genetic nomenclature for the nematode Caenorhabditis elegans. Mol. Gen. Genet. 175: 129-133. HOWELL, A. M., D. COOL, J. HEWITT, B. YDENBERG, M. J. SMITH and B. M. HONDA, 1987 Organization and unusual expression of histone genes in the sea star Pisaster ochraceus. Journal of Molecular Evolution 25: 29-36. HOWELL, A. M., S. G. GILMOUR, 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. JACOB, F. and J. MONOD, 1961 Genetic regulatory mechanisms in the synthesis of proteins. J. Mol. Biol. 3: 318-356. JEFFREYS, A. J. and R. A. FLAVELL, 1977 The rabbit beta-globin gene contains a large insert in the coding sequence. Cell 12: 1097-1108.  100  JOHNSON, C. D., J. B. RAND, R. K. HERMAN, B. D. STERN and R. L. RUSSELL, 1988 The acetylcholinesterase genes of C. elegans: Identification of a third gene (ace-3) and mosaic analysis of a synthetic lethal phenotype. Neuron 1: 165-173. JUDD, B. H., M. W. SHEN and T. C. KAUFMAN, 1972 The anatomy and function of a segment of the X chromosome of Drosophila melanogaster. Genetics 71: 139-156. KARLSSON, S., and NEINHUIS, A. 1985. Developmental regulation of human globin genes. Ann. Rev. Biochem. 54: 1071-1108. KEMPHUES, K. J., M. KUSCH and N. WOLF, 1988 Maternal-effect lethal mutations on linkage group II of Caenorhabditis elegans. Genetics 120: 977-986. KIM, J. S. and A. M. ROSE, 1987 The effect of gamma radiation on recombination frequency in Caenorhabditis elegans. Genome 29: 457-462. KOHARA, Y., K. AKIYAMA and K. ISONO, 1987 The physical map of the whole £ coli chromosome: Application of a new strategy for rapid analysis and sorting of a large genomic library. Cell 50: 495-508. KONDO, K., J. HODGKIN and R. H. WATERSTON, 1988 Differential expression of five tRNAij^^trp) amber suppressors in Caenorhabditis elegans. Mol. Cell: Biol. 8: 3627-3635. LANDEL, C. P., M. KRAUSE, R. H. WATERSTON and D. HIRSH, 1984 DNA rearrangements of the actin gene cluster accompany reversion of three muscle mutants. J. Mol. Biol. 180: 497-513. LASKO, P. F. and M. L. PARDUE, 1988 Studies of the vestigial microregion of melanogaster. Genetics 120: 495-502. LEFEVRE, G. and W. WATKINS, 1986 The question of the total gene number in melanogaster. Genetics 113: 869-895.  Drosophila Drosophila  LEICHT, B. G. and j. j. BONNER, 1988 Genetic analysis of chromosomal region 67A-D of Drosophila melanogaster. Genetics 119: 579-593. LEWIS, J. A., WU, C. H., BERG, H. and LEVINE, J. H. 1980. The genetics of levamisole resistance in the nematode C. elegans. Genetics, 95: 905-928. L'HERNAULT, S: W., D. C. SHAKES and S. WARD, 1988 Developmental genetics of chromosome I spermatogenesis-defective mutants in the nematode Caenorhabditis elegans. Genetics 120: 435-452. MacLEOD, A. R., J. KARN and S. BRENNER, 1981 Molecular analysis of the unc-54 myosin heavychain of Caenorhabditis elegans. Nature 291: 386-390. MARCHANT, G. E. and D. G. HOLM, 1988 Genetic analysis of the heterochromatin of chromosome 3 in Drosophila melanogaster. II. Vital loci identified through EMS mutagenesis. Genetics 120: 519-532. MATHEWS, C. K., E. M. KUTTER, G. MOSIG and P. B. BERCET (Editors), 1983 Bacteriophage (American Society of Microbiology, Washington, D.C).  14.  101 MAXSON, R. R. COHN and L. KEDES, 1983 Expression and organization of histone genes. Ann. Rev. Genetics 17: 239-277. ;  McCARRON, M., J. O'DONNELL, A. CHOVNICK, B. S. BHULLAR, J. HEWITT and E. P. M. CANDIDO, 1979 Organization of the rosy locus in Drosophila melanogaster. Further evidence in support of a c/s-acting control element adjacent to the xanthine dehydrogenase structural element. Genetics 91: 275-293. McKIM, K. S., A. M. HOWELL and A. M. ROSE, 1988 The effects of translocations on recombination frequency in Caenorhabditis elegans. Genetics 120: 987-1001. McKIM, K. S. and A. M. ROSE, 1988 Construction of an array of duplications to facilitate rapid mapping of LG I genes in Caenorhabditis elegans. Genome 30 (Suppl. 1): 298. MENEELY, P. M. and R. K. HERMAN, 1979 Lethals, steriles, and deficiencies in a region of the X chromosome of Caenorhabditis elegans. Genetics 92: 99-115. MENEELY, P. M. and R. K. HERMAN, 1981 Suppression and function of X-linked lethal and sterile mutations in Caenorhabditis elegans. Genetics 97: 65-84. MENEELY, P. M. and K. D. NORDSTROM, 1988 X chromosome duplications affect a region of the chromosome they do not duplicate in Caenorhabditis elegans. Genetics 119: 365-375. MILLER, ). H. and W. S. REZNIKOFF (Editors), 1978 The Operon. (Cold Spring Harbor, New York: Cold Spring Harbor Laboratory). MOERMAN, D. G. and D. L. BAILLIE, 1979 Genetic organization in Caenorhabditis structure analysis of the unc-22 gene. Genetics 91: 95-103.  elegans: Fine-  MOERMAN, D. G., G. M. BENIAN and R. H. WATERSTON, 1986 Molecular cloning of the muscle gene unc-22 in Caenorhabditis elegans. Proc. Natl. Acad. Sci. 83: 2579-2583. MULLER, H. J., 1929 The gene as the basis of life. Proc. Fourth Int. Congr. Plant Sci. [Ithaca, 1926] 1: 897-921. MULLER, H. J., 1932 Further studies on the nature and causes of gene mutations, pp 213-255. In: Proceedings  of the Sixth International  Congress of Genetics, Edited by D. JONES. Brooklyn  Botanic Gardens, Menasha, Wisconsin. NASH, D. and F. C. JANCA, 1983 Hypomorphic lethal mutations and their implications for the interpretation of lethal complementation studies in Drosophila. Genetics 105: 957-968. NELSON, D. W. and B. M. HONDA, 1985 Genes coding for 5S ribosomal RNA of the nematode Caenorhabditis elegans. Gene 38: 245-251. PARK, E.-C. and H. R. HORVITZ, 1986 C. elegans unc-105 mutations affect muscle and are suppressed by other mutations that affect muscle. Genetics 113: 853-867. PERRIMON, N., D. MOHLER. L. ENGSTROM and A. P. MAHOWALD, 1986 X-linked female-sterile loci in Drosophila melanogaster. Genetics 113: 695-712. PETERS, K. and A. M. ROSE, 1988 Complementing alleles of the bli-4 locus of elegans. Genome 30: s294.  Caenorhabditis  102 PRASAD, S., 1988 Molecular analysis within the dpy-20 to unc-22 region of chromosome IV in Caenorhabditis elegans. Ph.D thesis, Simon Fraser University, Burnaby, B.C. RIDDLE, D. L., M. M. SWANSON and P. ALBERT, 1981 Interacting genes in dauer larva formation. Nature 290: 668-671. RITOSSA, F. M., K. C. ATWOOD and S. SPIECELMAN, 1966 A molecular explanation of the bobbed mutants of Drosophila as partial deficiencies of "ribosomal" DNA. Genetics 54: 819-834. ROGALSKI, T. M., 1983 A genetic characterization of region around the unc-22 gene in Caenorhabditis elegans. PRASAD, S., 1988 Molecular analysis within the dpy-20 to unc-22 region of chromosome IV in Caenorhabditis elegans. Ph.D thesis, Simon Fraser University, Burnaby, B.C. ROGALSKI, T. M., D. G. MOERMAN and D. L. BAILLIE, 1982 Essential genes and deficiencies in the unc-22 region of Caenorhabditis elegans. Genetics 102: 725-736. ROGALSKI, T. R. and D. L. BAILLIE, 1985 Genetic organization of the unc-22 IV region of Caenorhabditis  elegans. Mol. Gen. Gen. 201: 409-414.  ROGALSKI, T. M., A. M. E. BULLERJAHN and D. L. RIDDLE, 1988 Lethal and amanitin-resistance mutations in the Caenorhabditis elegans ama-1 and ama-2 genes. Genetics 120: 409-422. ROGALSKI, T. M. and D. L. RIDDLE, 1988 A Caenorhabditis elegans RNA polymerase II gene, ama-1 IV, and nearby essential genes. Genetics 118: 61-74. ROSE, A. M., 1980 Genetic studies on the gene coding for paramyosin in Caenorhabditis elegans: unc-15 and the adjacent region. Ph.D. thesis, Simon Fraser University, Burnaby, B.C., Canada. ROSE, A. M. and D. L. BAILLIE, 1979 Effect of temperature and parental age on recombination and nondisjunction in Caenorhabditis elegans. Genetics 92: 409-418. ROSE, A. M. and D. L. BAILLIE, 1980 Genetic organization of the region around unc-15 (I), a gene affecting paramyosin in Caenorhabditis elegans. Genetics 96: 639-648. ROSE, A. M., D. L. BAILLIE, E. P. M. CANDIDO, K. A. BECKENBACH and D. NELSON, 1982 The linkage mapping of cloned restriction fragment length differences in Caenorhabditis elegans. Mol. Gen. Genet. 188: 286-291. ROSE, A. M., D. L. BAILLIE and J. CURRAN, 1984 Meiotic pairing behavior of two free duplications of linkage group I in Caenorhabditis elegans. Mol. Gen. Genet. 195: 52-56. ROSENBLUTH, R. E. and D. L. BAILLIE, 1981 Analysis of a reciprocal translocation, eTl(lll;V), Caenorhabditis elegans. Genetics 99: 415-428.  in  ROSENBLUTH, R. E., C. CUDDEFORD and D. L. BAILLIE, 1983 Mutagenesis in Caenorhabditis elegans. I. A rapid eukaryotic mutagen test system using the reciprocal translocation, eTl(lll;V). Mutation Research 110: 39-48. ROSENBLUTH, R. E., C. CUDDEFORD and D. L. BAILLIE, 1985 Mutagenesis in Caenorhabditis elegans. II. A spectrum of mutational events induced with 1500R of gamma-radiation. Genetics 109: 493-511. . •'  103  ROSENBLUTH, R. E., T. M. ROGALSKI, R. C. JOHNSEN, L. M. ADDISON and D. L. BAILLIE, 1988 Genomic organization in Caenorhabditis elegans: deficiency mapping on linkage group V(left). Genet. Res. 52: 105-118. SCHAFFNER, W., E. SERFLING and M. JASIN, 1985 Enhancers and eukaryotic gene expression. Trends Genet. 1: 224-230. SCHIBLER, U. and F. SIERRA, 1987 Alternative promoters in developmental gene expression. Annu. Rev. Genet. 21: 237-258. SHANNON, M. P., T. C. KAUFMAN, M. W. SHEN and B. H. JUDD, 1972 Lethality patterns and morphology of selected lethal and semi-lethal mutations in the zeste-white region of Drosophila melanogaster. Genetics 72: 615-638. SIGURDSON, D. C., G. j. SPANIER and R. K. HERMAN, 1984 C. elegans deficiency mapping. Genetics 108: 331-345. SMITH, C. L, J. C. ECONOME, A. SCHUTT, S. KLCO and C. R. CANTOR, 1987 A physical map of the Escherichia coli K12 genome. Science 236: 1448-1453. SNYDER, M., and N. DAVIDSON, 1983 Two gene families clustered in a small region of the Drosophila genome. J. Mol. Biol. 166: 101-118. STARR, T., A. M. HOWELL, J. McDOWALL, K. PETERS and A. M. ROSE, 1989 Isolation and mapping of DNA probes within the linkage group I gene cluster of Caenorhabditis elegans. Genome, in press. SULSTON, J.-E. and S. BRENNER, 1974 The DNA of Caenorhabditis  elegans. Genetics 77: 95-104.  SULSTON, J. E. and H. R. HORVITZ, 1977 Post-embryonic cell lineages of the nematode Caenorhabditis elegans. Dev. Biol. 56: 110-156. SULSTON, J. E. and H. R. HORVITZ, 1981 Abnormal cell lineages in mutants of the nematode Caenorhabditis elegans. Dev. Biol. 82: 41-55. SULSTON, J. E., E. SCHEIRENBERG, J. G. WHITE and J. N. THOMSON, 1983 The embryonic cell lineage of the nematode Caenorhabditis elegans. Dev. Biol. 100: 64-119. von MENDE, N., D. McK. BIRD, P. S. ALBERT and D. L. RIDDLE, 1988 dpy-13: A nematode collagen gene that affects body shape. Cell 55: 567-576. WAHLI, ]. 1988. Evolution and expression of vitellogenin genes. Trends Genet. 4: 227-232. WARD, S., D. j. BURKE, J. E. SULSTON, A. R. COULSON, D. G. ALBERTSON, D. AMMONS, M. KLASS and E. HOGAN, 1988 The genomic organization of transcribed major sperm protein genes and pseudogenes in the nematode Caenorhabditis elegans. J. Mol. Biol. 199:1-13. WATERSTON, R. H., 1981 A second informational suppressor, sup-7 (X), in elegans. Genetics 97: 307-325.  Caenorhabditis  WATERSTON, R. H. and S. BRENNER, 1978 A suppressor mutation in the nematode Caenorhabditis elegans acting on specific alleles of many genes. Nature 275: 715-719.  WATERSTON, R. K., R. M. FISHPOOL and S. BRENNER, 1977 Mutants affecting paramyosin in Caenorhabditis elegans. J. Mol. Biol. 117: 679-698. WATERSTON, R. H., D. HIRSH and T. R. LANE, 1984 Dominant mutations affecting muscle structure in Caenorhabditis elegans. J. Mol. Biol. 180: 473-496. WATERSTON, R. H., J. N. THOMSON and S. BRENNER, 1980 Mutants with altered muscle structure in Caenorhabditis elegans. Dev. Biol. 77: 271-302. WRIGHT, T. R.' F., 1987 The genetic and molecular organization of the dense cluster of functionally related, vital genes in the DOPA decarboxylase region of the Drosophila melanogaster genome, pp. 95-120. In: Structure and Function of Eukaryotic Chromosomes. Edited by W. Hennig. Springer-Verlag, Berlin.  105 APPENDIX l Below are complementation tables summarizing the data used to assign gene names to the new essential genes described in this thesis. In each table, a " + " indicates complementation and a "-" indicates failure to complement. Heterozygous lethal-bearing males are listed across the top of the tables, and sDp2-rescued lethal-bearing hermaphrodites are indicated down the side. The tables are ordered with the regions of the chromosome going from the left end to dpy-5. 1) Region left of sDf4 lin-6 h92 lin-6 lin-6 let-357 let-357 let-360 let-362 let-362 let-365 let-365  h92 h99 h89 h132 h96 h86 h93 hW8 h!29  lin-6 h99  let-357 h89  +  -  + + +  + +  +  let-360 h96  let-362 h86  let-362 h93  let-365 hi 08  let-365 h!29  +  +  +  +  +  -  +  +  +  + + + +  + +  +  +  let-357 hl32  +  + +  +  -  -  + +  +  -  2) Region in sDf4 to the left of hDf6 let-368 h!21 let-368 let-369 let-372  h121 h125 hi 26  let-369 h!25 +  + +  let-372 h126 + +  3a) Region between the left breakpoints of hDf6 and hDf7 let-356 h501 let-356 let-356 let-356 let-366 let-366 let-366 let-366 let-373 let-501 let-501 let-509 let-509 let-509  h83 h679 h871 h112 h265 h422 h852 h234 h498 h714 h521 h522 h867  let-356 h679  -  +  let-356 h871  let-366 h112  +  -  +  +  let-366 h265  let-366 h411  +  +  -  -  +  +  let-366 h505  +  let-366 h852  +  -  let-366 h890  + -  -  +  + +  +  +  -  •  +  +  + +  +  +  + + + +  + +  106 3b) Region between the left breakpoints of hD/6 and hD/7 let-373 h234 let-356 let-356 let-366 let-366 let-366 let-373 let-501 let-501 let-509 let-509 let-509 let-511  h83 h679 h112 h422 h505 h234 h498 h714 h521 h522 h867 h755  let-373 h573  let-501 h498  +  +  + ' + ' " + +  +  +  +  let-509 h522  +  + -  +  •+  +  +  +  let-509 h867  +  +  +  +  let-501 let-509 h>714 h521  +  let-511 h755  +  +  +  +  +  +  +  +  +  let-510 h740  +  +  +  +  +  +  +  +  4) hDf7 region let-353 h46 let-353 let-503 let-504 let-505 let-507  h>46 h418 h448 h426 h439  + +  let-503 h313  let-503 h418  +  -  +  +  let-504 h844  + +  let-504 h888  + + + +  let-505 h426  + + + +  let-506 h300  let-507 h439  let-354 h482  let-354 h508  + + + + +  + + +  5a) Region in hD/6 and hDp3 let-351 h43 let-351 let-354 let-354 let-354 let-354 let-354 let-354 let-354 let-354 let-374 let-375 let-375 let-502 let-502 let-502 let-508  h43 h79 h390 h441 h504 h803 h809 h841 h863 h251 h259 h391 h392 h732 h783 h452  let-354 h79  let-354 h90  +  -  +  let-354 h201  +  let-354 h267  +  let-354 h370  +  +  let-354 h693  +  _  -  + +  + + + + +  +  + + + + +  + +  + + + +  + +  107 5b) Region in hD/6 and hDp3 let-354 h819 let-351 let-354 let-354 let-354 let-354 let-354 let-354 let-374 let-375 let-375 let-502 let-502 let-502 let-508  h43 h79 h390 h803 h809 h841 h863 h251 h259 h391 h392 h783 h835 h452  let-354 h866  +  let-374 h251  + + +  -  +  + + +  +  let-375 h259  let-502 h392  + + +  -  let-502 h509  +  +  + +  + + + + + +  +  hm  h114 h131 hi 04 h119 h128 h123  +  let-508 h995  +  + + _  +  him-1 h134 h134 h45 h81 h94 h97 hi 13 h98  let-508 h452  _  +  -  +  +  +  +  6a) Region between dpy-5 and the right breakpoint of hD/6  him-1 let-352 let-355 let-359 let-361 let-361 let-363 let-363 let-363 let-363 let-364 let-367 let-370 let-371  let-502 h732  let-352 h45  +  + + + + + + +  + +  + +  let-355 h81  +  let-359 h94  let-361 h97  let-361 h113  +  + + +  -  +  + +  + +  +  +  + + +  +  +  + -  108 6b) Region between dpy-5 and the right breakpoint of hDf6 let-363 h98 him-1 let-352 let-355 let-359 let-361 let-361 let-363 let-363 let-363 let-363 let-364 let-367 let-371  h134 h45 h81 h94 h97 h113 h98 Mil h!14 hl31 h!04 h!19 h!23  let-363 hill  let-363 hl!4  + +  + +  + +  let-363 h!31  , let-364 h104  + + + +  let-367 hl19  let-370 h!28  + + + +  •+ + +  +  +  let-371 h!23  + +  +• +  + + +  + +  +  + + +  + +  APPENDIX II Construction of Strains for DNA Polymorphism Mapping  The availability of both a genetic and physical map of the chromosome I gene cluster will allow the detailed -study of genes and genomic organization in this large autosomal segment of the C. elegans genome. Since the physical map of the genome is nearly completed (Coulson ef al. 1986; Coulson ef al. 1988), the availability of genetically positioned cloned DNA probes from this region would allow for the alignment of the physical and genetic maps. One method for the isolation of DNA probes within a region of interest is Tc1-linkage selection (Baillie et al. 1985). Tc1-linkage selection takes advantage of the fact that the BO strain contains ten times the number of Tc1 elements found in N2 worms (Emmons ef al. 1983). Thus BO derived DNA can be distinguished from N2 DNA by determining which fragments contain new Tc1 elements. I have constructed an interstrain hybrid for Tc1-linkage selection. Two linked mutations  (dpy-5(h!4),  unc-29(h2)) were induced in the Bergerac (BO) strain by L. Harris and A. M. Rose. By replacing the dpy-5 unc-29 segment of the Bristol (N2) genome with this marked BO DNA, new Tc1 elements derived from the BO strain were crossed into an N2 background. The interval between these loci covers most of the chromosome I gene cluster. A hybrid, KR408, which was essentially free of BO DNA except for the BO-derived dpy-5 unc-29 interval was constructed. Tc1hybridizing fragments in the DNA of this strain have provided DNA markers for the dpy-5 unc-29 interval (Starr ef al., 1989).  Construction of the hybrid strain, KR408 The BO dpy-5 unc-29 double mutant strain was crossed to wild-type N2 males. F1 hermaphrodites were picked and allowed to self fertilize. One of their Dpy-5 Unc-29 offspring was isolated to generate the first backcross strain. This procedure was repeated until the sixth backcross strain was generated. This strain, KR408, contained BO DNA between the dpy-5 (h14) and unc-29 (h2) markers and approximately 97% N2 DNA for the rest of the genome.  110  Construction of chromosome I mapping strains Two strains which could be used to map polymorphic probes to chromosome I were constructed using methods modified from Rose ef al., (1982). Outcross progeny of the genotype dpy-5 unc-29 (BO) I + + (N2); unc-22 (N2) I + (N2) were allowed to self fertilize. Dpy-5 Unc-29 and Unc-22 progeny were isolated. DNA was isolated from the progeny of 50 Dpy-5 Unc-29 hermaphrodites (by T. Starr). This DNA preparation contained BO DNA in the dpy-5 unc-29 (I) region and N2 for the remainder of the genome. DNA was also isolated from the progeny of 50 Unc-22 hermaphrodites (by T. Starr). This DNA preparation contained N2 DNA from chromosome IV and a mixture of N2 and BO DNA for chromosome I. These DNA preparations were could then be used to verify that probes detecting strain polymorphisms isolated from KR408 mapped to chromosome l.  Construction of three-factor mapping strains Construction of recombinant chromosome I's containing both N2 and BO DNA allowed for three-factor analysis of polymorphisms isolated from KR408. Hermaphrodites of the genotype dpy-5 unc-29 (BO) I + + (N2) were allowed to self. Recombinant Dpy-5 and Unc-29 individuals were isolated (dpy-5 unc-29 (BO) I dpy-5 (BO) + (N2) and dpy-5 unc-29 (BO) I + (N2) unc-29  (BO)). From their progeny, individuals which were homozygous for the recombinant chromosome were isolated to generate recombinant strains. Thirteen Dpy-5 strains and nine Unc-29 strains were thus made available for mapping polymorphisms with respect to these two markers.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0098283/manifest

Comment

Related Items