Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Genetic and developmental studies of proximal segments of chromosome 3 of Drosophila melanogaster Sinclair, Donald A. R. 1977

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


831-UBC_1977_A1 S55.pdf [ 12.97MB ]
JSON: 831-1.0094310.json
JSON-LD: 831-1.0094310-ld.json
RDF/XML (Pretty): 831-1.0094310-rdf.xml
RDF/JSON: 831-1.0094310-rdf.json
Turtle: 831-1.0094310-turtle.txt
N-Triples: 831-1.0094310-rdf-ntriples.txt
Original Record: 831-1.0094310-source.json
Full Text

Full Text

C - \ GENETIC AND DEVELOPMENTAL STUDIES OF PROXIMAL SEGMENTS OF CHROMOSOME 3 OF DROSOPHILA MELANOGASTER by DONALD A.R. SINCLAIR B.Sc.(Hons), University of Manitoba, 1969 M.Sc, University of Manitoba, 1972 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in the Department of Zoology We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA March, 1977 © Donald A.R. Sinclair, 1977 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of Brit ish Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of Brit ish Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date ABSTRACT The present work deals with several approaches to the study of regions near the centromere of chromosome 3 of Drosophila melanogaster. The goals of this research were: (i) to examine spontaneous crossing over near the centromere in detail; ( i i ) to locate the Deformed locus genetically and to determine whether this lesion is a recessive lethal; ( i i i ) to test the efficacy of radiation-induced crossing over as a method of producing proximal aberrations; and (iv) to genetically and developmentally characterize a temperature-sensitive (ts) a l l e l e of a Minute locus, located near the centromere. CHAPTER 2 describes a study of recombination, which deals with short genetic regions near the centromere of chromosome 3, using the intervals, st-in-ri-eg^-Ki-p^ and Gl-p^-Sb-H. The following generali-zations have emerged: (i) an excess of multiple crossover chromosomes was recovered, and the intervals which immediately span the centromere showed the highest negative interference; ( i i ) a positive correlation of simultaneous exchange within closely-linked intervals, was noted for many of the multiple crossovers; and ( i i i ) several classes of reciprocal crossover products were not recovered equally. Three possible explanations for these results are: pre-meiotic exchange, chromatid interference and gene conversion. The results of one experi-ment also indicated that the interchromosomal effects of C(1)M3 are most pronounced within the st-in and Ki-p^ intervals. CHAPTER 3 describes a genetic study of the Deformed locus. The mapping results confirmed that Dfd is closely linked to Ki. Genetic analysis of the crossover chromosomes suggested that Dfd is homozygous viable and this was confirmed by the synthesis of homozygous Dfd stocks. This indicates that the Dfd locus is not located within section 84F in proximal 3R. CHAPTER 4 deals with experiments involving the use of radiation to produce crossovers near the centromere of chromosome 3, in males. Crossovers originating from exchange nearest the centromere, were associated with clusters more frequently than those originating from exchange within other proximally-adjacent segments. Induced exchange was frequently accompanied by mutation and/or chromosome damage, at or near the site of exchange. This was particularly true for cross-overs resulting from exchange in wholly euchromatic segments. It is suggested that many of the radiation-induced crossovers arise through asymmetrical exchange, and that this approach w i l l permit the isolation of proximal aberrations. CHAPTER 5 describes the genetic and developmental analysis of a ts Minute. As a ts all e l e of a proximally-located Minute locus, Q-III exhibits the classical dominant M trai t s , recessive lethality, and a highly pleiotropic phenotype, at 29°C. This phenotype was analysed in detail through the use of various temperature shift experiments. Q-III possesses a polyphasic temperature-sensitive period (TSP) for lethality extending from the f i r s t larval instar to late pupation. Shorter heat pulses defined discrete larval, larval/pupal, and pupal TSPs for lethality. In addition, homozygous Q-III females exhibit ts s t e r i l i t y and maternal effects, indicating that the Q-III gene product is essential throughout development. iv Heat-pulse experiments revealed a number of adult developmental abnormalities, involving derivatives of eye-antennal, leg, wing and genital imaginal discs. Many defects, for example, those involving the eye or antenna (eye-antennal disc), male genitalia (genital disc), and scutellum (wing disc), have larval TSPs; whereas others, such as bristle or sex comb traits, have pupal TSPs. It is suggested that the former defects may be related to c e l l death in the larval anlagen; while the latter are more likely due to blockages in differentiation during pupation. Q-III also interacts in ts fashion with several non-allelic muta-tions . Thus, at 29°C, Q-III is lethal when combined with DI, Ly_ and Dfd; suppresses the sex comb phenes of Msc and Pc; and produces wing nicking effects when combined with vg or Sex. TSPs were defined for the vg, Dl and Sex interactions. It is suggested that many of these interactions arec:metabolic rather than specific. The fact that Q-III phenotypically resembles bobbed and ts suppressor of forked—, strengthens the notion that Minute gene products are active in translation. It is concluded that translational defects can fully account for the pleiotropy of Q-III. V ACKNOWLEDGEMENTS I would like to express my gratitude to David Suzuki for his guidance, encouragement and patience throughout the course of this work. Above a l l I am thankful for his friendship. I would also like to thank Thomas Kaufman for his help, encour-agement and friendship. His creative approach has helped me immeasurably. As long as I have been in the lab, Shirley Macaulay has been the centre of strength to nearly everyone, a truly beautiful friend to whom I am indebted for so much. I thank David Holm for his helpful and c r i t i c a l l y knowledgeable discussions over the past few years. I am very grateful to Deborah Koo who did a beautiful job of the Scanning E. M. work and the photography. I would like to express my thanks to my patient and accomplished typists Mrs. Thorpe and Helen Curro. Finally, I would like to thank a l l of the friends who have affected me both s c i e n t i f i c a l l y and personally throughout the years in the fly lab. I w i l l treasure the memories. I am grateful for the financial support of the National Research Council and MacMillan Fund during the course of this work. Parts of CHAPTER 2 are published in Genetical Research. v i TABLE OF CONTENTS ABSTRACT i i ACKNOWLEDGEMENTS v LIST OF TABLES ix, x LIST OF FIGURES x i , x i i CHAPTER 1 GENERAL INTRODUCTION I. Background 1. Heterochromatin 1 2. The Genetic Importance of Proximal Regions The search for heterochromatic l o c i 2 The genetic study of proximally-adjacent segments of chromosome 3 6 3. Other Genetic Properties of Proximal Regions Crossing over 8 II. The Present Work 12 CHAPTER 2 CROSSING OVER BETWEEN CLOSELY LINKED MARKERS SPANNING THE CENTROMERE OF CHROMOSOME 3 I. Introduction 14 II. Materials and Methods 15 III. Results 22 IV. Discussion 38 CHAPTER 3 A GENETIC STUDY OF THE DEFORMED LOCUS I. Introduction 44 II. Materials and Methods 47 III. Results 48 IV. Discussion 61 v i i CHAPTER 4 A STUDY OF INDUCED CROSSING OVER NEAR THE CENTROMERE OF CHROMOSOME 3 I. Introduction 64 II. Materials and Methods 68 III. Results 1. Radiation-Induced Crossing Over in Males Numbers of progeny and frequency of crossing over 70 Numbers of crossover events 72 Recovery of mutants amongst crossovers 73 Regional comparison of crossovers and mutants 75 2. Analysis of Lethal Stocks Inter se complementation, pseudodominance and additional tests 77 Cytological analysis 81 Other complementation tests 83 IV. Discussion 85 CHAPTER 5 A GENETIC AND DEVELOPMENTAL STUDY OF Q-III, A - ' TEMPERATURE-SENSITIVE MINUTE MUTATION I. Introduction 88 II. Materials and Methods 92 1. Genetic Analysis Genetic mapping 93 Complementation of Q-III with other proximal mutations on chromosome 3 94 2. Developmental Analysis Tests for s t e r i l i t y and maternal effects 94 Regular temperature shift studies 95 Pulse shift studies 98 Scanning electron microscopy 99 Q-III interactions 99 Tests of other Minutes with homeotics 100 III. Results 1. Genetic Analysis Via b i l i t y Mapping 101 105 v i i i Tests of Q-III in triploids 108 Complementation tests 108 Developmental Analysis Duration of developmental periods 112 Tests for s t e r i l i t y and maternal effects 112 Stage distribution of lethality of Q-III homozygotes and heterozygotes 114 Temperature-sensitive periods for lethality of Q-III 117 Temperature-sensitive periods for dominant rough eye and bristle traits 125 Phenotypes revealed by shift experiments 128 Interactions displayed by Q-III and other Minutes 160 Scx-ts67 interactions 175 Summary of TSPs involving Q-III 178 IV. Discussion 1. The Nature of Minutes and Q-III Is Q-III a temperature-sensitive Minute? 182 The function of Minute genes 183 Is Q-III a single site lesion? 187 2. Developmental Characteristics of Q-III 189 Pattern defects and c e l l death 190 Defects resulting from heat treatment of Q-III during pupation 195 The interactions of Q-III 197 Additional uses for Q-III in studies of development 199 CHAPTER 6 OVERVIEW 201 BIBLIOGRAPHY 207 APPENDICES 1. Crossover Data and Estimates of Crossing Over 218 2. Summary of Results of 24-hour Exposure of Q-III Homozygotes to 29°C at Specific Times During Development 221 3. Summary of Results of 48-hour Exposures of Q-III Heterozygotes (TMl/Q-III) to 29°C at Specific Times During Development 223 ix LIST OF TABLES 1. Summary of AH Third Chromosome Mutations Used 16 2. Summary of Special'Mutant Chromosomes Used 18 3. Crossover Frequencies in the _st to JDP Interval in Chromosome 3 23 4. Types and Numbers of Recombinant Chromosomes Recovered 24 5. Coefficients of Coincidence Computed From A l l Multiples Recovered 25 6. Results of Mapping Experiments Using The Markers Gl Sb H/p_P 27 7. Interval-Specific Examination of Data of Females Producing Double Crossovers 29 8. Crossover Values in Progeny of Females Producing Multiple Crossovers Compared to Those of Females Producing None 31 9. Analysis of Crossing Over in Those Females Which Produced Triple Crossover Chromosomes 32 10. Tetrad Analysis of The Crossover Data 33 11. Comparison of Observed With Expected Numbers of Double Exchanges 34 12. Types and Numbers of Recombinant Crossovers Recovered 35 13. Results of The Cross of Dfd/Ki p P Females to Ki p P/Ki p P Males 49 14. Results of Inter Se and Deficiency Complementation Tests With Recombinant Lethals 53 15. Distribution of Crossovers in Proximal and Non-Proximal Intervals for sjz to _p_P Exchanges 71 16. Regional Summary of Lethals and Steriles Present on Crossover Chromosomes 74 17. Relative Occurrence of Crossovers and Lethal Events for Proximal and Non-Proximal Intervals 76 X 18. Results of Inter Se and Deficiency Complementation Crosses Involving Crossover Lethals 78 19. Summary of Pseudodominance and Complementation Tests Between Selected Crossover Mutants and Known Proximal Mutations 80 20. Phenotypic Description and Cytological Analysis of Crossover Mutants 83 21. Relative V i a b i l i t i e s of Q-III Homozygotes and Hetero-zygotes at Different Temperatures 102 22. Crossover Data From Crosses Designed to Localize Q-III 106 23. Relative V i a b i l i t y of Q-III in Combination With Various Mutations at 22° or 29°C 109 24. The Lengths of the Developmental Periods From Egg Deposition to Eclosion in Different Classes at Different Temperatures 111 25. Lethality of Control and Q-III-Bearing Progeny at Different Developmental Stages at Various Temperatures 115 26. Lethal and Visible Phenotypes of Heterozygotes for Q-III and Dl, vg_, Ly_, Sex, Msc and P_c at Different Temperatures 161 27. Interactions of Known Minutes With Different Homeotic Mutations Affecting Sex Combs 164 28. Effect of ts67 on Sex Expression 177 x i LIST OF FIGURES 1. Schematic representation of proximal regions of the third chromosome 19 2. Schematic representation of possible relative arrange-ments of Ki and Dfd 50 3. Complementation pattern emerging from inter se and deficiency complementation crosses (Dfd) 54 4. Proposed arrangements of lethal sites in proximal 3R in the original Dfd/Ki pP female 56 5. Scanning electron micrographs showing the eye development of Dfd adults 60 6. Results of the shift study to delineate a TSP for lethality of Q-III homozygotes 118 7. Results of 48-hour pulse shifts to delineate TSPs for lethality of Q-III heterozygotes and homozygotes - 120 8. Results of 24-hour pulse shifts to delineate TSPs for lethality of Q-III homozygotes 123 9. Results of the shift study to delineate TSPs for rough eyes and reduced bristles of Q-III heterozygotes 126 10. Scanning electron micrographs showing the effects of Q-III on macrochaete development 129, 131 11. Scanning electron micrographs showing the effects of Q-III on eye development 134, 136 12. Scanning electron micrographs showing the eye-antennal pattern defects of a Q-III homozygote 139 13. Scanning electron micrographs showing the effects of Q-III on the development of the scutellum at 17°C 142, 144 146, 148 14. A scanning electron micrograph showing a wing-like duplication in a Q-III/TM3 fly 151 15. Scanning electron micrographs, showing the effects of Q-III on sex comb development 154 16. Scanning electron micrographs showing foreleg fusion of a Q-III/TM3 fly 157 x i i 17. Results of the shift study to delineate a TSP for the vg-Q-III interaction 168 18. Results of the shift study to delineate a TSP for the Dl-Q-III interaction 170 19. Results of the shift study to delineate a TSP for the Scx-Q-III interaction 174 20. TSPs for lethal, sterile and adult morphological effects of Q-III 179 21. A schematic diagram of a mature eye-antennal imaginal disc 191 1 CHAPTER 1 GENERAL INTRODUCTION I. Background For many years, geneticists have studied segments that l i e near the centromeres of chromosomes. Attempts to characterize these proximal segments have been marked by considerable speculation (see Cooper, 1959), as well as by definitive experimentation. 1. Heterochromatin Long ago, Heitz (1933) reported that specific chromosome segments possess special cytological properties, in that they remain permanently condensed. Such segments are particularly prominent in regions near the centromeres of chromosomes of eukaryotes, and more recently they have been called constitutive heterochromatin (Brown, 1966) . When cytological length is considered, constitutive heterochromatin includes 20 to 25 percent of the two autosomes, 35 to 50 percent of the X chromosome and the entire Y chromosome of Drosophila melanogaster. The previous discovery of highly redundant sat e l l i t e DNA sequences in Drosophila and their preliminary localization within heterochromatin through _in situ hybridization (Gall e_t a_l., 1971) has recently led to the important study of Peacock e_t a_l. (1974). These workers detected 7 species of sa t e l l i t e DNA and determined that they were highly repetitive. In situ analysis of labelled RNA complementary to these DNA sequences revealed that the latter are located mainly in the chromocentre in salivary gland chromosome preparations. Nucleotide analysis of 3 of the satellites showed that the basic repeating units are relatively short and simple, thereby supporting the idea that most of this repetitive DNA is incapable of coding for a gene product of average complexity. 2. The Genetic Importance of Proximal Regions Directly or indirectly, several workers have given valuable in-formation concerning the nature of proximally-located genes. This information provides ample justification for continued study of this region of the chromosome. The search for heterochromatic l o c i Few genes capable of being mutated have been localized within proximal heterochromatin in any of the chromosomes of Drosophila. This observation led to the hypothesis that these chromosome regions are genetically inert (Muller and Painter, 1932). To test this, sev-eral approaches have been adopted to attempt to genetically dissect the various heterochromatic regions in the genome of this organism. Since the only phenotype which accompanies the lack of the entire Y chromosome is male s t e r i l i t y (Bridges, 1916), i t seemed that no v i t a l genes resided on this chromosome. Such a sterile phenotype was success fully exploited by Brosseau (1960), who provided a minimum estimate of 7, Y-linked male f e r t i l i t y factors. Muller jst al_. (1937) used the method of selecting for reciprocal products of meiotic exchange within overlapping inversions to recover heterochromatic deletions and duplications on the X-chromosome. They found no i n v i a b i l i t y associated with large duplications or deficiencies and concluded that aside from bobbed (bb), no essential genes are 3 present in the major blocks of heterochroma tin near the centromere. This has recently been corroborated by Schalet and Lefevre (1973). However, the latter workers mentioned that the suppressor of forked locus may be located within heterochromatin. The bobbed (bb) phenotype (short, fine bristles and delayed eclo-sion) has been mapped to a proximal location. Quantitative correla-tions between the dosage of bb and br i s t l e length (Stern, 1929), as well as the existence of a lethal a l l e l e , pointed to a probable hypo-morphic basis for bb alleles . High frequency mutation and reversion of bb had been noted in several studies. The cytological position of the Nucleolus Organizing (NO) region was located close to the bb locus (Cooper, 1959). Consequently, a combined genetic-biochemical approach was initiated to see i f the bb locus and the NO region were identical. First, Ritossa and Spiegelman (1965) used a nucleic acid hybridi-zation technique to measure the proportion of total DNA which codes for the 2 species of ribosomal (r)RNA (18S and 28S) and they found that each X chromosome carried about 130 copies of both types. Furthermore, they showed that f l i e s bearing different numbers of NO regions possessed different numbers of rRNA genes, thereby indicating that the NO region and the site(s) of these genes are either identical or closely linked. Finally, Ritossa et a l . (1966a) compared phenotypically different stocks of hb and found that the numbers of rDNA copies varied inversely with the phenotypic intensity. The instability of bb is attributed to the propensity of such a highly redundant locus to cause asymmetrical pairing of homologues, because via crossing over this would generate deficiency and duplication products. The apparent lack of l o c i within proximal heterochromatin of the 4 autosomes has stimulated a great deal of interest. Since highly repe-t i t i v e s a t e l l i t e DNA had been localized to constitutive heterochromatin and because of the existing evidence which equated bb with the tandemly redundant rRNA genes, Suzuki (1970; 1974a) suggested that genes con-trolling v i t a l functions such as c e l l division might be highly redun-dant and reside in heterochromatin. Thus, the loss of one or even a few copies of such genes through recessive mutation could be normalized by the presence of many wild-type duplicates. Two alternative methods of testing the hypothesis were formulated by Suzuki and his co-workers. First they used Ethyl methanesulphonate (EMS) to screen for Dominant temperature-sensitive (DTS) lethal muta-tions, following the rationale that any unconditional dominant mutation in such essential genes would of necessity be lethal or s t e r i l e . Sub-sequent screens yielded several such mutants, both on the second (Suzuki and Procunier, 1969) and the third (Holden and Suzuki, 1973) chromosomes. However, none was definitely localized to proximal heterochromatin, although DTS-6 mapped to a proximal position in chromosome 3. The second approach involved attempts to synthesize extensive deletions within heterochromatin through the method of attaching and detaching compound third chromosomes (Baldwin and Suzuki, 1971). The rationale was based on the idea that formation of compounds is fre-quently accompanied by asymmetrical exchange within the proximal seg-ments of the chromosome. This method enabled these workers to isolate a large number of recessive lethals whose complementation pattern suggested that many of these lesions were deficiencies. Many also displayed the dominant Minute phenotype (Minutes are recessive lethal 5 mutations having the dominant visible traits, thin bristles and delayed development). However, as with the PTS lethals, none of the above mutations could be unequivocally assigned to heterochromatin (although in this instance many were mapped to proximal positions). The lack of known heterochromatic mutations, as well as the rela-tive dearth of lesions in proximally-adjacent euchromatic segments, has made work with this region of chromosome 3 extremely d i f f i c u l t . In contrast, rolled (rl) and light (lt) have been genetically localized near or within the proximal heterochromatin of chromosome 2 and Pf(2R)M(2)S210 lacks a l l of the heterochromatin in 2R (see Lindsley and Grell, 1968). Hi l l i k e r and Holm (1975) used a method involving detachment of compound second chromosomes to isolate a large number of recessive lethals in both 2L and 2R heterochromatin. Several of these were shown to be deletions through pseudodominance tests with rJL and l t and complementation tests with Pf(2R)M(2)S210. H i l l i k e r (1976) next isolated EMS-induced alleles of some of these deficiencies and he was able to resolve a total of 13 heterochromatic l o c i (including It and r l ) in this chromosome. He argues that the relatively high muta-b i l i t y of these genes is inconsistent with the idea that they are redundant. The above genetic evidence, coupled with previously mentioned biochemical evidence indicating that the basic repetitive units of PNA in heterochromatin are simple, would suggest that heterochromatic genes are li k e l y structurally and functionally distinct from the surrounding simple sequence PNA. However, i t would be interesting to genetically probe the constitutive heterochromatin of chromosome 3 i n a similar fashion and to extensively study such uniquely-placed l o c i . The genetic study of proximally-adjacent segments of chromosome 3 Several approaches have proved successful in helping to genetically characterize regions near the constitutive heterochromatin of chromo-some 3. Three studies involved the direct selection for chromosome aberrations near the centromere, while an additional study provided relevant information even though i t was not i n i t i a l l y designed to investigate proximal l o c i . Lindsley _et al_. (1972) exploited the segregation properties of a large number of Y-autosome translocations to survey the v i a b i l i t y and phenotypic effects of aneuploidy for specific segments of the auto-somes. They verified the existence of two previously identified, proximal Minutes (M) on chromosome 3, M(3)S34 in 3L and M(3)S39 in 3R; and they discovered a new M site, M(3)LS4 in 3L and an additional haplo-insufficient locus, Splayed (Spl) which is located between the 3R heterochromatin and M(3)S39. The existence of a Triplo-lethal (Tpl) locus in a proximal part of 3R did not allow the characterization of an appreciable portion of this segment, since this locus is inviable when present in either haploid or triploid doses. A major drawback of this method is the fact that the duplications and deficiencies syn-thesized do not exist as stable stocks, but must be re-synthesized each time they are to be used. However, this characterization of the dosage-sensitivity of the genome is extremely valuable. In particular, the findings concerning the proximal Minutes in chromosome 3 are important since many workers are presently attempting to study these l o c i with a view to determining their primary functions. A group of so-called homeotic l o c i has been localized within proximal segments of the right arm of chromosome 3. Since mutations in these l o c i cause switches in developmental fates of imaginal discs resulting in the production of structures normally derived from other discs, such l o c i have generated considerable interest (for a review see Postlethwait and Schneiderman, 1973). One of the homeotic genes, Nasobemia (Ns) was suspected to be neo morphic (Muller, 1932). It had been previously shown that the neo-morph, K i l l e r of prune (K-pn) could be phenotypically reverted through the use of radiation and the genetic evidence suggested that such revertants were often deficient for the K-pn locus (Lifschytz and Falk 1969). Denell (1972, 1973) extended this method to the study of Ns, and recovered several revertants including a putative deletion. In a subsequent study, Duncan and Kaufman (1975) used this approach to iso-late a number of deletions in proximal 3R, 2 involving Ns revertants and -3 involving doublesex (dsx). They corroborated Denell's findings which indicated that Antennapedia (Antp), Ns and Extra sex comb (Sex) are mutations in the same locus (possibly including Multiple sex comb) Their experiments have also provided a number of stable deficiencies in proximal 3R which have been very useful for mapping other mutants, especially since the restrictive nature of proximal crossing over makes the map positions of proximal l o c i less meaningful. In a study that was designed to select for temperature-sensitive lethals along the entirety of chromosome 3, Tasaka and Suzuki (1973) obtained interesting results. Ninety percent of the ts mutations were mapped to the proximal region between scarlet (st) and Stubble (Sb). Genetically the sj: to Sb interval is small, although cytologically i t includes nearly 40 percent of the chromosome. One of the mutants was lethal at 29°C but viable at 25°C and at the latter temperature, 8 heterozygotes displayed a phenotype similar to that of Spl, which is located near the heterochromatin in the right arm. Some of the aforementioned Ns_ and dsx deficiencies have been used to map at least five of these ts lethals (T. C. Kaufman, personal communication), thereby supporting the results of the recombination mapping. The basis for the apparent preferential selection of ts mutations in proximal l o c i of chromosome 3 is not known. However, the eventual cytological mapping of a l l of the ts lethals may provide information concerning the genetic organization of these l o c i . The well-documented u t i l i t y of ts mutations for investigating developmental properties of a given gene (see Suzuki, 1970; Hartwell, 1974), should provide the impe-tus to determine how these proximal l o c i function during development. 3. Other Genetic Properties of Proximal Regions For decades, geneticists have observed numerous properties of chromosome segments residing near the centromere, particularly in Prosophila. The following category is the most relevant to my work. Crossing over The area of investigation involving the centromeric regions which has produced the most striking results, is that of crossing over. Several approaches to the study of crossing over near the centromere have been adopted, and in most cases, the results are descriptive in nature. Spontaneous meiotic crossing over in females is the type of ex-change most frequently studied. The work of several people has provided a number of interesting observations. For example, i t was found that while the most proximal segment makes up about 20 to 25 percent of the 9 mitotic length of chromosome 3 cytologically, i t constitutes only 1 percent of the total genetic length (Dobzhansky, 1930; Painter, 1935). Thus, the obvious suggestion was that for meiotic crossing over in females, exchange occurs only within euchromatin. The idea that no crossing over occurs within heterochromatin has been supported by the findings of Baker (1958) and Roberts (1965), and of particular import-ance in this regard is the work of Hi l l i k e r (1975). In addition to this, however, i t has long been recognized that crossing over between markers spanning proximal regions is severely restricted relative to comparable regions in more distal locations. Beadle (1932) used appropriate chromosome aberrations to displace markers from distal to more proximal positions and was able to show that crossing over between them was reduced, thereby suggesting that some sort of inhibitory effect of the centromere on crossing over exists. More recently, Thompson (1963a,b) proposed that i f exchange pairing of centromeric intervals was rapidly followed by localized centromeric repulsion just prior to exchange, this may explain the observed decrease in crossing over between proximal l o c i . Another interesting observation concerning spontaneous crossing over is that while positive interference usually governs double ex-change in adjacent intervals of the same chromosome arm (Morgan et a l . , 1925), studies of Drosophila have shown that simultaneous exchange within closely linked intervals which span the centromere is independent (Graubard, 1934; Stevens, 1936). In fact, Morgan et a l . (1925) found coincidence values of 1.3 for crossing over near the centromere of chromosome 3. 10 The above information poses some interesting questions with respect to spontaneous crossing over near the centromere. Does negative inter-ference definitely appear in multiple exchange between proximal loci? If so, is i t restricted to one arm of the chromosome? Besides causing mapping d i f f i c u l t i e s in work with proximally-located mutant l o c i , could centromeric inhibition obscure the nature of a given gene by maintaining close linkage of accumulated lethals to that gene? Induced crossing over has been demonstrated in both females (reviewed by Schultz and Redfield, 1951; and see Whittinghill, 1955) and males (Friesen, 1933; 1937a,b; Patterson and Suche, 1934; Whitting-h i l l , 1937). It has long been known that in females, recombinagenic agents increase crossing over preferentially near the centromere of the chromosome (see Schultz and Redfield, 1951; and Lucchesi and Suzuki, 1968). Similarly, i t has been reported by several people that induced exchange in males occurs mainly within proximal regions (Friesen, 1937b; Whittinghill, 1937; Puro, 1966; Hannah-Alava, 1968). However, no published study has conclusively established that induced crossing over in proximal regions occurs principally within hetero-chromatin. Friesen (1933) was the f i r s t to report that radiation could induce crossing over in males of Drosophila. From his data he concluded that radiation-induced exchange in males closely resembles spontaneous meiotic exchange in females. Contemporaries in this f i e l d (Patterson and Suche, 1934; Whittinghill, 1937) agreed with this conclusion and f e l t that induced exchange was relatively precise, since crossover chromosomes were not usually associated with lethality. However, Muller (1954, 1958) has argued that induced crossing over in both 11 sexes occurs by a mechanism similar to that which produces transloca-tions, and thus, crossover chromosomes formed in such an asymmetrical manner could be associated with aberrations. Recently, studies have shown that radiation-induced crossover chromosomes are frequently asso-ciated with lethality or s t e r i l i t y and in some cases chromosome aberra-tions (Hannah-Alava, 1968; Mglinets, 1972). If induced crossing over does involve asymmetrical exchange, and i f such crossing over occurs preferentially within or near heterochro-matin, the selection of induced crossovers might prove to be an important way of enriching for stable proximal deficiencies. 12 II. The Present Work These background studies have established an important foundation for subsequent investigations of the proximal regions of the chromo-some. Since many of the features described above are in some ways related, I decided to adopt a multifaceted approach in my work with these regions. This thesis represents a report of the results of such an approach. CHAPTER 2 describes a series of experiments designed to explore proximal recombination in chromosome 3 of females with a view to ex-amining: (a) the degree and nature of negative interference in different genetic intervals near the centromere and (b) the extent of interchromosomal effects within these intervals. CHAPTER 3 represents a genetic study of the Deformed locus, in-cluding both an experiment designed to map this gene relative to other proximal markers in 3R and the subsequent analysis of resulting lethal crossover chromosomes. Since evidence exists which suggests that the lethality formerly ascribed to this locus is due to a closely-linked but separate mutation, the main aim of this study was to determine i f the Dfd lesion is i t s e l f a recessive lethal. The potential use of Dfd in future crossover studies, as well as the developmental interest of this mutant stemming from its temperature-sensitivity and i t s effects on the derivatives of the eye-antennal imaginal disc, provided the impetus for this study. CHAPTER 4 describes the use of radiation-induced, male crossing over to recover proximally-located aberrations and lethals. The ra-tionale for this method arises from recent evidence which argues that many induced crossovers may be the result of asymmetrical exchange events. Finally, CHAPTER 5 is a report concerned with the genetic and developmental investigation of a ts all e l e of a Minute locus, located proximally in chromosome 3. By studying the mutant I hoped to provide more information about (a) the basis of the Minute phenotype and (b) the effects of such a lesion on the development of the organism. 14 CHAPTER 2 CROSSING OVER BETWEEN CLOSELY LINKED MARKERS SPANNING THE CENTROMERE OF CHROMOSOME 3 I. Introduction Recently, the correct location of the centromere of chromosome 3, relative to the position of l o c i known to be tightly linked, has been ascertained. Thus, radius incompletus and inturned have been assigned to the l e f t arm (Arajarvi and Hannah-Alava, 1969) along with Polycomb (Puro and Nygren, 1975) and eagle (Holm, et aj.., 1969), while Kinked (Merriam and Garcia-Bellido, 1969) and Deformed (Holm et a l . , 1969) have been positioned in the right arm. The unequivocal l e f t and right localization of these genes makes i t possible to more accurately interpret crossover data from this region and therefore an intensive study of crossing over in the intervals adjacent to the centromere of chromosome 3 was initiated. 15 I I . Materials and Methods Tables 1 and 2 represent summaries of a l l third chromosome muta-tions and special chromosomes used in these experiments. A l l third chromosome balancers that have been used are described fully in Lindsley and Grell (1968). The balancer referred to as TM3 should henceforth be considered as equivalent to TM3, Stubble Serrate unless otherwise indicated. A l l experiments were performed at 22 i 0.5°C unless otherwise specified. Recombination was measured in the proximal region of chromosome 3 using the following markers (for a complete description, see Table 1 and consult Lindsley and Grell, 1968): s_t - scarlet (44.0), in -2 inturned (47), _ri - radius incompletus (47.0) , eg - eagle-2 (47.3), Ki - Kinked (47.6), and £^ - pink peach (48.0). Figure 1 is a sche-matic representation of the map positions of the markers along the chromosome (Lindsley and Grell, 1968). The centric blocks of hetero-chromatin are believed to be immediately flanked by eagle (Holm et. a l . , 1969) and Kinked (Merriam and Garcia-Bellido, 1969) on the l e f t and right respectively. The s_t - in interval was designated as 1, jLn -2 2 p r i as' 2, r i - eg-' as 3, eg - Ki as 4 and Ki - pj_ as 5. Note that the centromere lies in interval 4. Three different major studies of crossing over were performed: Experiment I. 100 st in r i eg^ Ki p P/ + + + + + + females were test-crossed for 5 consecutive 3-day broods and their progeny were scored. 2 Experiment II. Crossing over was similarly measured in 213 st in r i eg / + + + + Ki p P females; 108 of these were studied for five 3-day broods (Expt. Ila) and 105 for one 3-day brood (Expt. l i b ) . Experiment I I I . 16 Table 1 Summary of A l l Third Chromosome Mutations Used Mutant Symbol Genetic Position Phenotype Lyra Glued scarlet transformer inturned radius-incompletus Polycomb . 2 eagle Deformed Deformed-recessive Kinked roughened-eye ,G1 st  tra in r i Pc', 2 eg. Dfd Dfd1 Ki roe proboscipedia pb 40.5 41.4 44.0 45 47 47.0 47.7 47.3 47.5 47.5 47.6 47.6 47.7 Extra sex-comb Sex 47 Excised wing margins (rec. lethal) Rough small eyes Bright red eyes Transformation of females into sterile males Thoracic hairs and bristles directed towards midline Interruptions in L2 A l l legs of male possess sex combs (rec. lethal) Wings spread Reduced eyes (rec. lethal) Recessive a l l e l e of Dfd Short and twisted bristles and hairs Eyes rough Transformation of oral lobes into tarsus or arista A l l legs of male possess sex combs (rec. lethal) 17 Table 1 (continued) Mutant Symbol Genetic Position Phenotype Antenna-pedia Multiple-sex comb Antp 48 Nasobemia double sex pink peach  Stubble Delta Hairless ebony-sooty ,Msc Ant£ dsx £ P Sb DI H Ns 48.0 48.0 48 48.0 58.2 66.2 69.5 Transformation of antenna into leg structures (rec. lethal) A l l legs of male possess sex combs, associated with ::in(3R)84B;85F (rec. lethal) Transformation of antenna into leg structures Males and females intersexual Dull ruby eyes Short and thin bristles (rec. lethal) Termini of wing veins thick and broad (rec. lethal) Missing postvertical and abdominal bristles (rec. lethal) 70.7 Black body 18 Table 2 Summary of Special Mutant Chromosomes Used Chromosome Abbreviation Cytology Reference Df(3R)Antp +R2 Df (3R)Antp-Ns+R21 Df (3R)dsx-EH-R2 D+-R5 Df (3R)dsx- T(3;Y)P92 Dfd kar ry Antp-+R2 Ns' +R21 dsx-D+-R2 DH-R5 dsx1 DP-P92 Dfd-rk Df(3R)84B3;84Dl-2 Duncan and Kaufman, 1975 Df(3R)84A-B;84D-E " Df(3R)84D9-12;84F16 Df(3R)84F2-3;84Fl6 11 Insertion of 3R(84D10-ll;85Al-3) into Y " — Chovnick et a_l., 1971 FIGURE 1 Schematic representation of proximal regions of the third chromosome showing the genetic markers used. Published map positions are given below the symbols with numerically designated crossover intervals indicated above the line. / 2 3 4 5 — I H ir t^mmmxBnmm i 1 — 4 4 . 0 4 7 47.0 47.3 47.6 48.0 0 21 Crossing over was measured in 25 C(1)M3/Y; st in r i eg + + / + + + + Ki p P females for five 3-day broods. Expt. I l l females were tested in order to determine any interchromosomal effects of the inversions con-tained in each arm of the compound X (Lucchesi and Suzuki, 1968). A l l females tested were mated individually (within 40 hours of eclosion at 22 i 0.5°C) with 2 or 3 males homozygous for 2 o s st in r i eg Ki p e . The third chromosomes of a l l females were isogenized prior to use in order to minimize the presence of lethals in the stocks. However, the other chromosomes were not made co-isogenic and the Ki _p_^  chromosomes of Expt. l i b females were of different origins than those of Expt. Ha females. Progeny in each v i a l were scored daily until the eighteenth day after the parents had been introduced. III. Results A summary of the numbers of progeny examined and the crossover values for the region studied (including published map distances, Lindsley and Grell, 1968) are given in Table 3. Data for females carrying normal X chromosomes (columns 2 and 3, Table 3) reveal that recombination was consistently higher in Expt. I than in Expt. II 2 (X = 108.9, P = 0.05). These differences probably reflect random differences in genetic backgrounds in the two series. The insertion of C(1)M3 into test females (columns 4 and 5, Table 3) noticeably augmented recombination, thereby reconfirming i t s interchromosomal effects on crossing over near proximal heterochromatin. These effects were more prominent for the distalmost intervals; for example, re-combination in regions 1 and 5 increased 3- and 4-fold, respectively. Table 4 summarizes the number of different crossover chromosomes recovered. A total of 3,603 single, 85 double and 20 triple crossover chromosomes was scored. The most frequent class of doubles occurring in Expts. I and II involved regions 1 and 5 (nearly a third of the total) and 3 and 4 (more than a third of the total). Other doubles frequently recovered were 3, 5 and 1, 4. Double crossovers involving exchange in intervals known to be on the same side of the centromere were never recovered. However, 1, 2, 4 and 1, 3, 4 triple crossovers, which included exchange in two of these intervals, were recovered. Since double crossover chromosomes involving exchange in the two distalmost intervals (1 and 5) were frequently encountered, these data 23. Table 3 Crossover Frequencies in The s_t to _p_£ Interval in Chromosome 3 Experiment Number  Genetic Reference I Ila III Ratio Region Values rod X C(1)M3 III/IIa 1 3.0 3.99 1.77 5.39 3.05 2 0.06 0.25 0.20 0.30 1.50 3 0.30 0.21 0.08 0.17 2.13 4 0.30 0.43 0.22 0.64 2.90 5 0.40 0.65 0.25 1.11 4.44 Number of Fertile Females - 92 108 25 Number of Progeny - 36,948 33,139 4,063 Table 4 Types and Numbers of Recombinant Chromosomes Recovered Region Experiment Number I H a l i b III SINGLES 1 1393 574 217 547 2 71 63 12 13 3 27 17 4 30 4 101 50 24 58 5 183 64 41 114 Totals 1775 768 298 762 DOUBLES 1,2 0 0 0 0 1,3 0 0 0 0 1,4 3 2 0 1 1,5 , 14 8 1 2 2,3 0 0 0 0 2,4 2 0 0 1 2,5 4 0 0 1 3,4 22 6 1 4 3,5 5 2 2 1 4,5 0 3 0 0 Totals 50 21 4 10 TRIPLES • 1,2,3 0 0 0 0 1,2,4 3 0 0 0 1,2,5 0 0 0 0 1,3,4 2 2 0 0 1,3,5 0 0 0 0 1,4,5 4 4 2 2 2,3,4 0 0 0 0 2,3,5 0 0 0 0 2,4,5 0 0 0 0 3,4,5 1 0 0 0 Totals 10 6 2 2 25 Table 5 Coefficients of Coincidence Computed From A l l Multiples Recovered Experiment Number  Intervals I Ha III 1.2 0.81 1.3 0.64 4.25 1.4 1.90 6.21 1.43 1.5 1.90 8.17 1.23 2.3 -2.4 12.50 2.5 6.75 3.4 75.20 137.16 22.36 3.5 11.90 30.00 49.20 4,5 4.83 38.40 6.93 26 support the contention of other work in Drosophila (Graubard, 1934; Stevens, 1936), in Neurospora (Bole-Gowda et a l . , 1962) and yeast (Hawthorne and Mortimer, 1960), that positive interference does not extend to regions in different arms of the chromosome. This is fur-ther emphasized by the fact that most of the triple crossovers (12 of 20) involved exchanges in intervals 1, 4 and 5. However, i t should also be noted that a l l of the triple crossovers involved the most 2 proximal interval (eg to Ki). Coefficients of coincidence were calculated for doubles in a l l three experiments (excluding Expt. l i b ) . In a l l cases but 1, 2 and 1, 3 doubles of Expt. I, these values exceeded unity (Table 5). Ex-tremely high values for 3, 4 and 4, 5 exchanges (and 3, 5 exchanges for Expt. I l l ) indicate a very high negative interference in these intervals. Therefore, in spite of very tight linkage between s_t and p P, the recovery of multiple crossover chromosomes greatly exceeds conventional expectations. It is noteworthy that while single ex-changes were increased in a l l intervals by C(1)M3, a concomitant increase in the occurrence of multiple exchanges (except for 3, 5 doubles) did not occur, as shown by coincidence values. In order to further test different proximal intervals for inter-ference, two other mapping experiments were carried out with the markers: Glued (Gl), Stubble (Sb), Hairless (H) and (see Table 1 and consult Lindsley and Grell, 1968 for descriptions of these l o c i ) . Heterozygous Gl + Sb H/+ _p_P + + females were crossed to j>P/p_P males (20 males and 20 females per quarter pint bottle) and crossing over was measured for the three intervals: (a) Gl to p_P; (b) _p_P to Sb; and (c) Sb to H. Note that the (a) interval spans the centromere. 27 Table 6 Results of Mapping Experiments Using The Markers Gl Sb H/pP Experiment Number of Single Crossovers Multiple Crossovers Number Progeny a b c a,b a,c b,c a,b,c IV 1148 68 98 123 10 9 5 1 Map Distances (percent) 7.67 9.93 12.02 Coefficients of Coincidence 1.26 0.95 0.44 V 909 57 73 84 9 9 4 1 Map Distances 8.36 9.57 10.78 Coefficients of Coincidence 1.37 1.20 0.53 28 Table 6 summarizes the data from Experiments IV and V. The map dis-tances for the three intervals show good agreement with book values (see Lindsley and Grell, 1968) although (a) is slightly larger i n both experiments. Coincidence values were calculated for the different interval combinations: a, b (1.3 to 1.4); a, c (0.95 to 1.2); and b, c (0.44 to 0.53). It is apparent that more multiple crossovers than expected occurred for the combination involving the centromeric and the immediately adjacent intervals (a, b), while that involving the most distal and centromeric intervals (a, c) showed about the expected number of multiple crossovers. However, interference was positive when the most distal intervals are considered. The level of negative interference for this series of experiments was also much less marked than that of the major series which had involved more proximal markers. Thus, these data further support the idea that multiple exchanges are more common in adjacent intervals more closely associated with the centromere, and that although positive interference does not extend to both sides of the centromere, negative interference does. Previous workers have suggested that some rare multiple exchange chromosomes could, in fact, result from successive single crossover events. Thus, a mitotic crossover in a gonial c e l l could be followed by a meiotic exchange to produce an apparent double crossover chromosome (Whittinghill, 1955; Suzuki et a l . , 1966). Such a Two-Step model pre-dicts that the gonial exchange could be amplified through mitotic d i v i -sions, thereby generating doubles amidst a cluster of single crossovers (Suzuki et a l . , 1966). The progeny of individual females yielding 29 Table 7 Interval-Specific Examination of Data of Females Producing Double Crossovers (Experiment I) Female Type of Double Number of Doubles Singles Occurring in Either Interval Total Number of Progeny J 1,5 1 23 (16.8)* 391 (360) K 3,4 2 2 ( 1.3) 388 (360) L 2,4 1 3 ( 1.6) 465 (360) M 3,5 2 2 ( 2.6) 466 (360) N 1,4 1 13 (15.6) 429 (360) *Numbers in parentheses represent mean values of comparable data for 52 non-multiple females 30 double exchanges (but not triples) were examined for evidence of clustering of single crossovers in the regions where the doubles had occurred. A sample of 5 such females (Expt. I) is given in Table 7 (along with mean values for females yielding no multiple exchange progeny). No noticeable clusters of singles appeared to accompany doubles for the regions in question. If multiple recombinant chromosomes are generated by a Two-Step mechanism, then crossover values for multiple-producing females would be expected to be higher than the values from females producing no multiples. These subsets of data were significantly different (Table 8) and in both experiments, crossover values for regions 3 and 4 were higher in those females producing multiple crossovers. However, when the crossover data of the 9 females of Expt. I that had produced triple recombinant progeny (within the sjt to j ) P interval) were examined, in each case the distribution of crossover types followed a Poisson distribution (Table 9). A tetrad analysis as inferred from single strand recovery (Weinstein, 1936), was initiated with a view to distinguishing between a meiotic and a gonial origin of the triple exchanges (Table 10). The Two-Step production of rare multiple exchange chromosomes was inferred from an insufficient number of double exchange chromosomes predicted from a tetrad analysis of the multiple exchange chromosomes (Suzuki et a l , , 1966). In every case examined in present tests (Expts. I, Ha and l i b ) , the number of triple exchange tetrads was equivalent to or exceeded that of the double exchange tetrads. This supports a .31 Table 8 Crossover Values in Progeny of Females Producing Multiple Crossovers Compared to Those of Females Producing None Female Type Multiples Expt I Expt H a No multiples Expt I Expt H a Number of Females Number of Progeny Crossover Values > U cu +J ci M 2 3 4 5 42 19,284 4.09 0.28 0.26 0.53 0.56 24 7,111 1.65 0.22 0.21 0.36 0.40 52 15,378 3.71 0.17 0.07 0.24 0.59 85 26,029 1.76 0.18 0.05 0.15 0.20 Significant difference for X was indicated for a subset comparison of both experiments at P = 0.01 32 Table 9 Analysis of Crossing Over in Those Females Which Produced Triple Cross-over Chromosomes (Experiment I) Female Types 0 of exchange 1 (st to p^) 2 3 *Chi-Square Values A 431 28 0 1 0.002 B 438 35 1 1 0.329 C 460 27 0 1 0.060 D 482 25 0 1 0.081 E 423 15 1 1 0.046 F 163 10 0 2 2.200 G 467 23 0 1 0.427 H 401 29 2 1 2.280 I 295 21 1 1 1.333 0 = no exchange 1 = 1 exchange 2 = 2 exchanges 3 = 3 exchanges *In each case recombination -was found to approximate a Poisson distribution at P = 0.05 33 Table 10 Tetrad Analysis of The Crossover Data (Inferred From Recovery of Single Strands) Tetrad Distribution Experiment Number I H a l i b Triple Exchange 80 48 16 Double Exchange 80 12 16 Single Exchange 3,410 1,536 734 No Exchanges 33,378 31,543 10,088 Total Tetrad Sample 36,948 33,139 10,854 34 Table 11 Comparison of Observed With Expected (From Meiotic Triple Exchange Tetrads) Numbers of Double Exchanges Experiment Number  Classes of I H a l i b Doubles Expected Observed Expected Observed Expected Observed 1,2 3 0 0 0 0 0 1,3 2 0 2 0 0 0 1,4 9 3 6 2 2 0 1,5 4 14 4 8 2 1 2,3 0 0 0 0 0 0 2,4 3 2 0 0 0 0 2,5 0 4 0 0 0 0 3,4 3 22 2 6 0 1 3,5 1 5 0 2 0 2 4,5 5 0 4 3 2 0 35 Table 12 Types and Numbers of Recombinant Chromosomes Recovered Experiment I Experiment H a Region Genotype Number ic Recovered R Genotype Number Recovered R SINGLES st 726 2 D in r i eg Ki p 667 st in 2 „. P r i eg Ki p st in r i 2 v . P st in r i eg ,P Ki pf 56 15 16 11 58 44 st in r i eg Ki 75 108 1:1 3.5:1 1.5:1 1.5:1 1.5:1 st Ki p F 2 in r i eg st in Ki p* 2 r i eg st in r i Ki p 2 302 272 39 24 5 12 st in r i eg^ Ki p P 31 + + + + + + + 19 19 Ki 45 2 P st in r i eg p 1:1 1.5:1 2.5:1 1.5:1 2.5:1 DOUBLES 1,4 1,5 2,4 2,5 st Ki p F 2 in r i eg  st p P 2 „. in r i eg Ki st in Ki p P 2 n eg st in p 2 „. r i eg Ki 3 0 7 7 1 1 2 2 3:0 1:1 1:1 1:1 st 2 „. P in r i eg Ki p st Ki 2 p in r i eg p st in 2 r r . P r i eg Ki p F st in Ki 2 p r L eg P F 2 0 0 8 0 0 0 0 2:0 8:0 36 Table 12 (continued) Experiment I Experiment H a Region Genotype Number Recovered R Genotype Number Recovered R DOUBLES st in r i Ki p P 19 st in r i 6 3,4 2 eg_ st in r i 6 3 P P 1 :1 eg 2 Ki P p st in r i Ki 6:0 0 2 3,5 2 V eg Ki 4 4 :1 2 p eg p r 2:0 0 st in r i 2 P A eg p r 0 st in r i eg Ki 3 4,5 Ki 0 2! 3:0 0 * R = Ratio of reciprocal classes ( Two-Step explanation for the origin of the multiples. Comparison of the minimum expected numbers of doubles (i.e., doubles generated by triple exchange tetrads) with the actual numbers recovered (generated by both double and triple exchange tetrads) reveals (Table 11) that in a few cases (1, 2; 1, 3 and 4, 5) the observed numbers were appre-ciably less than expected, while in most of the remaining classes the observed numbers exceeded or approximated the expected. Table 12 shows a summary of the reciprocal crossover classes re-covered from Expts. I and 11a females, along with the numbers of each class obtained. In several cases (particularly for the more proximal intervals), these classes do not appear to be equally represented, despite the apparent lack of any obvious selective advantage for non-mutant al l e l e s . In order to rule out high revertability of the markers studied as a contributive factor to some of the multiple exchange chromosomes, homozygous stocks were screened for revertants and none was found among 1.1 X 10^ st in r i eg 2 and 2.0 X 10^ st in r i eg 2 Ki p P e S chromosomes. 38 IV. Discussion Although genetically small, the region studied in these experiments represents a large portion of the physical length of chromosome 3. Unexpectedly, these experiments have revealed evidence for the exist-ence of non-classical, non-reciprocal recombination events in this region of the chromosome. Thus, interference (expressed as coefficients of coincidence) for these intervals was high and negative. The present work is concerned solely with intergenic recombination near the centromere of one of the autosomes of Drosophila. Previously, i t had appeared that exchange within genetically short regions in this organism was generally accompanied by high positive interference (Morgan et a l . , 1925). Exceptions to this in Drosophila usually involved intragenic crossing over (e.g. Hexter, 1958; Green, 1959, 1960), although Sturtevant (1951) reported that negative interference could be detected i n the study of intergenic crossing over in the fourth chromosome of tr i p l o i d females. Similarly, negative interfer-ence was seen in intergenic crossing over in Aspergillus (Calef, 1957; Pritchard, 1960) and barley (S^gaard, 1974). One explanation offered to account for such negative interference is equivalent to the idea of effective pairing in bacteriophage (see Chase and Doermann, 1958), which essentially assumes that short localized regions of pair-ing exist, and within these regions recombination is highly probable. Both Calef (1957) and Pritchard (1960) used this hypothesis to explain coincidence values exceeding 100 for exchange between tightly linked l o c i . However, S(6gaard (1974) discounted any explanation invoking 39 localized pairing for his work with the eciferum l o c i of barley, since in this case he was dealing with relatively large interlocus intervals. Recently, workers have promoted another possibility to account for the recovery of multiple crossovers at unexpected frequencies. Thus, the occurrence of successive gonial and meiotic exchange to produce rare multiple crossovers (i.e. the Two-Step model), has been postulated (Whittinghill, 1955; Suzuki et a l . , 1966). In the present study some evidence supports this idea, v i z . the relative lack of double com-pared to triple exchange tetrads and the higher levels of recombina-tion i n those females producing multiple exchange progeny. However, examination of the data for individual females failed to show the clustering phenomenon that would be predicted for the females gener-ating multiples. Furthermore, the types of double crossover chromo-somes encountered were not dissimilar to those that would be predicted to arise from the different types of meiotic triple exchange tetrads. In most cases, the numbers of the recovered double crossovers exceeded those of the expected (see Table 11). However, the Two-Step model should be considered as a possible contributive factor to the results of these experiments, particularly since C. Sharpe (personal communi-cation) finds high coincidence values for proximal recombination near the centromere of chromosome 2 and claims that gonial exchange is involved. As previously mentioned, work in several organisms has indicated that exchange across the centromere is marked by a lack of positive chromosome interference and, in the case of chromosome 3 of Drosophila, negative interference has been observed (Morgan e_t a_l., 1925). In the 40 present study, the occurrence of negative chromatid interference for crossing over near the centromere could provide an explanation for the high negative chromosome interference that was detected, since an excess of two-strand doubles would generate more double relative to single crossover chromatids thereby inflating coincidence values. In this regard, i t is noteworthy that Strickland (1961) and Bole-Gowda et a l . (1962) found evidence of chromatid interference in Neurospora, particularly with respect to centromeric crossing over. Hawthorne and Mortimer (1960) have mentioned a similar situation in yeast. Howe (1956) and Stadler (1956) had repudiated earlier claims that this phenomenon occurs in Neurospora. Welshons (1955) reported that negative chromatid interference could be seen in the study of crossing over in short genetic intervals in attached-X chromosomes of Drosophila. However, Baldwin and Chovnick (1967) found no chromatid interference in exchange in compound third chromosomes. Davis (1974) also failed to detect chromatid interference when using the meiotic mutant mei-s332 for the recovery of half-tetrads. It should be mentioned that neither of the latter 2 studies examined crossing over in proximal regions and therefore chromatid interference cannot be eliminated as a characteristic of exchange in these segments of the chromosomes. The demonstration of conversion in Drosophila has previously been invoked as an explanation for the occurrence of exceptional events in intragenic exchange. For example, in his study of the white locus, Green (1959, 1960) could account for some exceptional chromosomes by assuming that rare true double exchange was involved. However, in 41 crossover studies of different white-apricot pseudoalleles, four ex-ceptions appeared which could be explained by gene conversion, while no single crossovers were recovered. Recently, i t has been argued that recombination and conversion may be manifestations of the same homologous exchange event, particularly in light of the findings in work with maroon-like (Smith e_t al_., 1970) and in yeast (Hurst et: a l . , 1972), which revealed that half of the convertants were associated with exchange of flanking markers. In this study, conversion must be considered as a possible mechan-ism for the frequent production of multiple recombinant chromosomes. 2 For example, simple conversion of eg to i t s wild-type a l l e l e and vice versa, would result in apparent 3, 4 double crossovers. Triples i n -volving 3,4 exchange might be explained by the conversion of eagle accompanied by exchange of either of the most distal markers. Extend-ing this logic, 1, 4, 5 triples could result from conversions of Ki or Ki*~ with a crossover in region 1, while 1, 2, 4 triples could be gener-ated by conversion of ri_ or ri"^" and an exchange in region 4. It must be emphasized that previous failure to detect intergenic exchange events which resemble conversion in Drosophila is l i k e l y related to the effects of high positive interference. The absence of interference across the centromere might permit the appearance of such a phenomenon. Indeed, evidence from other organisms suggests that when tightly-linked markers are studied, crossing over produces multiple exchange chromo-somes at inordinately high frequencies (see Calef, 1957; and S^gaard, 1974). Conversion has been mentioned as a possible contributor. Crossover frequencies in this present work indicate that the region from in to Ki is particularly small genetically. 42 The use of the inverted attached-X chromosome (C(1)M3) actually resulted in lower coincidence values (in most cases), suggesting that fewer multiple relative to single crossovers occurred. Previous demonstration of i n t r i n s i c a l l y (Schultz and Redfield, 1951) and extrinsically (Suzuki and Parry, 1964) mediated recombinagenesis in Drosophila were marked by decreased positive interference. Therefore, the present data are consistent with the suggestion that conversion may be contributing to the appearance of multiple crossover chromosomes, since one would expect recombinagenic agents to effect similar increases in the occurrence of true multiple crossovers as well as of singles. Green (1975) has reported similar results from his work with this region of the chromosome. Puro and Nygren (1975) also observed a double crossover involving radius incompletus when they were mapping Polycomb. These workers raise the possibility that conversion is involved. Other support for the conversion-based explanation for the multi-ple crossovers is provided by the inequalities of reciprocal crossover classes (Table 12),- even though spontaneous reversion of these l o c i was not observed. The different possibilities discussed above might be distinguished by using females carrying third chromosome pericentric inversions which include a l l of the l o c i used in this study. Recombinants recovered from such heterozygotes would almost certainly arise from even-numbered crossovers, since odd-numbered crossovers would produce inviable pro-geny bearing extensive duplications or deficiencies. Comparisons of the control frequencies of multiple crossovers with the number of pro-geny produced by these females should provide information about the origin of multiple crossover chromosomes. 43 It might also be possible to use meiotic mutants which effect in-creased levels of non-disjunction but do not alter recombination and in this way capture half-tetrads to test for reciprocality and chromatid interference for crossing over in proximal regions of this chromosome. However, given the low rates of recombination in these regions, this would be a formidable project. Finally, i t is noteworthy that the interchromosomal effects of C(1)M3 were more marked for the distal crossover intervals that were examined. This raises the possibility that most of the proximal i n -creases noted previously (see Lucchesi and Suzuki, 1968) may have occurred near, but not in heterochromatin. In fact, since several lines of evidence suggest that no crossing over occurs within hetero-chromatin (Baker, 1958; Roberts, 1965; H i l l i k e r , 1975), i t is possible that this sort of recombinagenesis is wholly euchromatic. It would be interesting to study this further. For example, one could compare these effects to the a r t i f i c i a l induction of crossing over, which also occurs preferentially near the centromere in both males and females (see Schultz and Redfield, 1951). Since i t appears that both l_t and xl_ l i e within heterochromatin (Hilliker and Holm, 1975), chromosome 2 would probably be more useful for this purpose. This type of approach w i l l t e l l us a great deal about the properties of heterochromatin, particularly with respect to crossing over. 44 CHAPTER 3 A GENETIC STUDY OF THE DEFORMED LOCUS I. Introduction Deformed (Dfd) is a mutation mapping in the proximal part of chromosome 3 at 47.5. Dfd mutants express a dominant phenotype re-sulting in ventral and lateral reduction of ommatidial tissue in the eyes of the adult. Concomitantly, tufted vibrissae are often observed. The penetrance of the eye phenotype is variable and sometimes Dfd/+ flie s are indistinguishable from wild-type. Recessive alleles (e.g. r Dfd—) have been described, with allelism based on the observation that TO * r r Dfd/Dfd— fl i e s are phenotypically more extreme than Dfd—/Dfd—. Pre-viously, i t had been thought that the Dfd lesion was also lethal when homozygous, although the recovery of aberrant homozygotes at a very low frequency has been reported (Lindsley and Grell, 1968). This locus is of particular interest from a developmental stand-rL point. A temperature-sensitive a l l e l e (Dfd-—) has permitted the de-lineation of a TSP for the Dfd gene in the f i r s t to second larval instars (Vogt, 1947). The eye phenotype could be due to the promo-tion of localized c e l l death within the eye-antennal disc (Fristrom, 1969). This may explain the frequent occurrence of mirror-image dupli-cations of antennae of Dfd f l i e s . This is supported by the observation that at 29°C, Dfd interacts synergistically when heterozygous with a temperature-sensitive Minute, causing lethality by preventing the .forma-tion of eye-antennal structures (see CHAPTER 5). The occurrence of extensive proliferation of vibrissae, which is also diagnostic of the 45 Dfd phenotype, may be due to a repatterning of surviving cells within the disc following c e l l death (see Postlethwait and Schneiderman, 1973). Furthermore, Dfd interacts with ophthalmontera to produce homeotic wing tissue i n the eye (Ouweneel, 1969). Thus, further study of this locus with respect to patterns of c e l l death i n the eye disc during develop-ment, should prove interesting. The cytological location of Dfd has not been determined. Its genetic location does not define the chromosome arm in which this locus resides. However, Holm et a l . (1969) found that the synthesis of com-pound chromosomes heterozygous for Dfd (Dfd/+) from normal homologues, resulted i n the high frequency association of this gene with the 3R elements, while no 3L element containing Dfd was ever recovered, thereby proving that Dfd is in the right arm of this chromosome. Duncan and Kaufman (1975) synthesized an array of proximal chromo-somal aberrations in 3R by selecting for radiation-induced revertants of the homeotic mutant Nasobemia (Ns) as well as of the dominant a l l e l e of double sex (dsx^). Three revertants of the latter are cytologically-observable deficiencies, with the smallest (dsx^" R^) lacking the 3R material in the 84B-F interval. A l l three are lethal in combination with either of two different chromosomes bearing Dfd. However, none r r of these deficiencies exposes the phenotype of Dfd—, that i s , Dfd—/ deficiencies are wild-type. Moreover, they were able to show that the leth a l i t y associated with both original Dfd stocks is covered by Dp-P92, which includes the. region of 3R between 84D and 85A (see Tables 1 and 2). They offered two possible explanations for these results: (i) The Dfd locus is actually located in 84B-F, but Dfd— is not an a l l e l e of 46 Dfd (i.e. the recessive lethality jLs associated with the Dfd locus), ( i i ) The lethal mutation is distinct from the Dfd locus, and therefore Dfd is located elsewhere in proximal 3R and is viable when homozygous. A minimal genetic characterization of a locus is a prerequisite to any analysis of its developmental properties. Therefore, the pre-sent study was initiated to accomplish two things: (a) to map Dfd with respect to Kinked in the hope that their correct relative positions might aid future genetic studies in this region of the chromosome; and (b) to genetically analyse any recombinant chromosomes derived from this mapping study, in order to determine which of the aforementioned possibilities concerning the nature of the locus is correct. 47 II. Materials and Methods One hundred Dfd/Ki p P females were collected within 40 hours of eclosion and mass mated to homozygous st in r i eg 2 Ki p P males (hence-forth I w i l l refer only to the markers scored, Dfd, Ki and p P) in six quarter-pint milk bottles (20 females and 20 males per bottle). Every three days, the parents were transferred to fresh bottles and after nine days these fl i e s were discarded. Progeny were scored until the fifteenth day after the introduction of the parents. An early experi-ment using Dfd/Ki p P females crossed to Dfd/TM3 males was abandoned because of poor penetrance of Deformed in Dfd/TM3 f l i e s . However, several Dfd p P and Ki as well as one recombinant chromosomes from the latter were saved for genetic analysis. A l l recombinant and one parental chromosomes were balanced over either TM3 or CxD and examined for recessive lethality. A l l of the lethal-bearing recombinant chromosomes were tested for complementation inter se as well as with the following stocks: the Dfd parental stock; Dfd-rk (kindly contributed by Dr. D. G. Holm); and j • *. i M + R 2 1 A + R 2 A D+-R2 A D+R5 four deficiency stocks, Ns Antp dsx dsx ( a l l kindly contributed by Dr. T. C. Kaufman, see Table 2 for cytological descrip-tions) . 48 III. Results The results of the Deformed mapping experiment are presented in Table 13. A total of 5822 progeny was scored and 28 confirmed cross-overs between Dfd and _p_P were recovered (17 Dfd p P and 11 Ki types) . I n i t i a l l y , 34 progeny were scored as wild-type (i.e. Dfd^/Ki p P) recombinants but upon subsequent testing, 32 of these proved to be carrying Dfd (i.e. were parentals) and the remaining 2 (males) were st e r i l e . If the latter 2 were genuine crossovers, this would argue that Dfd is proximal to Ki in 3R (Figure 2a). The computed map dis-tance (excluding the two unconfirmed wildtypes) from Ki (or Dfd) to p P is 0.48 percent, a value reasonably close to that predicted from standard book values (Lindsley and Grell, 1968). Fourteen Dfd p P and seven Ki recombinant chromosomes were balanced. Recessive lethals were present on a l l of the former while 6 of 7 of the latter carried recessive lethals. In addition, 8 recessive lethal stocks including six Dfd p P and one Ki recombinants along with a single p P type were found in the early crossover experiment. The existence of the JD£ recombinant argues that Ki is proximal to Dfd in 3R (Figure 2b). For complementation purposes, the lethal-bearing recombinant stocks were designated as follows: Ki, lethals 1 to 7; Dfd p P, lethals 8 to 27; and lethal 28. The results of the complementation tests are summarized in Table 14 and Figure 3. Since a l l of the lethal crossover chromosomes were inviable in combination with the original Dfd parental chromosome, a l l must share a lethal site(s) in common with the parental stock. A l l Table 13 Results of The Cross of Dfd/Ki p P Females to Ki p P/Ki p P Males Progeny  Number of Number of Genotype Parentals Recombinants Unknowns Totals 2789 3003 17 11 2 (sterile) 2 5822 Dfd/Ki p P  Ki p P/Ki p P  Dfd p P/Ki p P Ki/Ki p P Dfd +/Ki p P 2789 3003 17 11 Map distance Dfd to j> = 0.48 percent FIGURE 2 Schematic representation of possible relative arrangements of Ki and Dfd in the proximal portion of the right arm of chromosome 3. In a, Dfd is proximal, Ki dista l ; in b, Ki is proximal and Dfd di s t a l . o Dfd K i + pp+ Dfd' r Ki K r Dfd I L 52 inter se combinations of the Dfd p P(p P) stocks as well as those of the Ki stocks were inviable, indicating that at least one common lethal is present on a l l of the chromosomes of a given recombinant class. How-ever, when members of the different recombinant classes were tested together in turn, and also each with the Dfd-rk and deficiency stocks, a differential pattern of complementation emerged (Figure 3). A l -though the results are not unequivocal, i.e. the deficiency or Dfd-rk stocks may contain more lethal sites, the simplest explanation is that the original Dfd parental chromosome contained a minimum of 3 st b c lethal sites designated as m , m , m , in addition to the Dfd locus. One possible arrangement of these lethals on the original Dfd chromosome is shown in Figure 4. A single crossover in region 2 would generate a Dfd p recombinant carrying m (Group IV also including b c lethal 28) and reciprocally a Ki recombinant carrying both m and m (Group I ) . Similarly, a single exchange in region 3 would provide Dfd p P recombinants carrying both m3 and (Group III) and the Ki reciprocals bearing m (Group II). A double exchange involving regions 2 and 4 could produce Dfd p P chromosomes carrying m3 and mC, thereby accounting for Group V types. Finally, i f Ki were proximal to Dfd, a crossover between them could generate a j 3 _ recombinant bearing m (lethal 28, Group IV). Thus, i t is proposed that is the lethal site which is carried by the Dfd-rk chromosome and that this site is exposed Df"R by both dsx deficiencies. The above proposal satisfies the requirement that a l l Ki recombin-c p ants carry a common lethal (m in Groups I and II) as do Dfd p types (m in both Group III and IV). Furthermore, the members of Groups I and III are non-complementing by virtue of the m3 site. Table 14 Results of Inter Se and Deficiency Complementation Tests With Recombinant Lethals lethals 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Dfd-3 ----•++ + + + + + + + + + + - + + + - + + + + + 4 _ _ _ + + + + + + + + + + + + - + + + - + + + + + 5 _ _ _ + + - - . - - + - + + - + + - - + + - + 6 - - + + + . + + - + + - - + + - + 7 + + + + + + + + + + + + - + + + - + + + + + 8 . . . -9 . . . - . . - - - . - - - + 10 + 11 - - - -12 " " 13 - - - - - - - - - -14 - - - -15 16 + 17 - - " " " " 18 " " " " + 19 . - - -. + 20 + 21 . - . . - - . + 22 - - - - - - + 23 24 " " " " + 25 - - - + 26 - - + 27 28 + D4-R2 dsx-—— + + + +- -+ -+ + - - - - - + - + + + + + - + + + - + IH-R5 dsx + + + +- -+ -+ + + - + + + + +- + + +- + Antpr^- + + + + + + + ++ + + + + + + + + + + + + + + + + + + + + +R21 Ns + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Dfd-par - - - - - - - - - - - - - - - - - - - - - - - - - - - -*Dfd-rk = Dfd rv stock (see Table 2) Dfd par. = o r i g i n a l Dfd parental chromosome FIGURE 3 Complementation pattern emerging from inter se and deficiency complementation crosses. The numbers 1 to 7 represent Ki recombinant lethal stocks while 8 to 27 are the Dfd p P lethals and 28 is the lethal. a b , c m , m and m represent proposed lethal sites originally present in Dfd parental stock. ^ ^dsx , D+R2 , , D4-R5 Def = dsx- and dsx  Def-rk = Dfd ry stock I, II, III, IV and V are distinct complementation groups. Dfd Parental Dfd-rk I 5,6 1,2,3,4,7 III 8, 11,12,13,14,15,17,23, 27 IV 9,10,16,18,19, 21, 22, 25,26, 28 V 20  24 FIGURE 4 Proposed arrangement of lethal sites in proximal 3R in the original Dfd/Ki p P female in which re-combination was studied, m , m and m represent the lethal sites. Regions 1, 2, 3 and 4 are possible crossover intervals. 6 7 m _L_ Dfd i m' _i O a+ — r -Ki mb+ pP m c+ 58 As represented in Figure 3, a l l Group IV Dfd p P recombinants are viable in combination with Dfd-rk. However, a l l of the heterozygotes (except lethal 28/Dfd-rk) displayed a very extreme form of the Deformed phenotype, which suggested that Dfd is viable when homozygous. The verification of the explanation for lethality of Dfd chromosomes would be the synthesis of Dfd homozygotes and such a stock was isolated in the following manner: Dfd pP/+ + (the lethal-26 stock was used) females were crossed to Dfd pP/TM3 males (from the lethal-26 stock, presumably carrying only m ) in bottles and the progeny scored. Surviving Dfd pP/Dfd p P progeny should arise only from the f e r t i l i z a t i o n of a ma+ Dfd p P recombinant oocyte by a m3 Dfd p P sperm. Three such re-combinant fl i e s (one male and two females) and a single JD^ (non-Dfd) individual were scored in a total of 2772 progeny. Stocks of the three Dfd p P recombinant chromosomes were produced and individual lines were established for each. In a l l three cases f e r t i l e Dfd p P/Pfd p P homozy-gotes were produced. Figure 5 is a scanning electron microscope picture of Dfd pP/+ + and Dfd pP/Dfd p P adults. The latter characteristically possess extremely reduced eyes (sometimes antennal structures are also missing) with extensively tufted vibrissae. The eye phenotype appears to be even more extreme than that of Dfd/Dfd—. In addition, mirror image duplications of the antennae (or in some cases aristae only) are frequently observed. That the Dfd locus does not appear to be located within the D1"R cytological interval 84A,B to 85A (which is delineated by the dsx—-deficiencies) is suggested by the observation that combining the newly isolated Dfd p P chromosome with the deficiency chromosomes, does not expose the more extreme phenotype which is characteristic of the homo-zygote. v> FIGURE 5 Scanning electron micrographs showing the eye development of Dfd heterozygotes (a, Dfd pP/+) and homozygotes (b, Dfd p P/Dfd p P), (magnification about x400). 61 IV. Discussion This study has shown that the Dfd locus is genetically separable from a lethal site present in the 84F interval in the proximal part of 3R of Dfd stocks (Duncan and Kaufman, 1975). Kinked and Deformed are very tightly linked and their relative positions are not definite and await further c l a r i f i c a t i o n through crossover studies. This should be facilitated through the use of Dfd in the homozygous condition,thereby eliminating the problem of misclassification of heterozygotes because of incomplete penetrance. Thus recombination experiments, for example, using females of the constitution st in r i eg 2 Dfd p P/Ki,should allow unequivocal ordering of the two genes in question and shed further light on the nature of proximal recombination in general. The question of the exact functional defect of the Dfd mutation remains unresolved, as does that of the cytological location of the locus. Deformed may be an amorph (Muller, 1932) in that the i n i t i a l mutation may produce the Dfd phenotype and the homozygote would simply exhibit an extreme version of the phenotype. Alternatively, Dfd could be an hypomorph, that i s , haplo-abnormal or dosage-sensitive. This is clearly possible since in their study of segmental aneuploidy in Drosophila, Lindsley et a l . . (1972) were unable to synthesize f l i e s heterozygous for a deficiency spanning the region 82CD to 83EF, an interval which includes the Triplo-lethal segment. This unique locus is lethal in either a haploid or tr i p l o i d condition. In a hypomorphic situation, +/deficiency would presumably be equivalent to Dfd/+. The possibility that Dfd is an hypermorph is more remote, since no Dfd phenotype was produced in f l i e s bearing i n t e r s t i t i a l duplications for 62 most of the proximal regions of 3R,and the only region resistant to trisomy was the 83DE (Tpl) interval (Lindsley et a l . , 1972). Dfd could be interpreted as an antimorph, particularly since there is evidence that i t is dominant in triploids (Lindsley and Grell, 1968). This could be tested directly by determining i f Dfd/+/+ f l i e s are phenotypically less mutant than Dfd/+ individuals. Finally, Dfd could be a neomorph. If the latter is the case, i t should be possible to induce deficiencies of the locus as has been done previously (Lifschytz and Falk, 1969; Denell, 1973; Duncan and Kaufman, 1975). However, i t should be noted that such attempts to revert Ki have not been successful (Duncan and Kaufman, 1975). If radiation-induced revertants of either of these mutations are pheno-typically indistinguishable from their respective heterozygotes, i t may not be possible to use such an approach to acquire proximal chromo-somal aberrations. This present study emphasizes the need to be cautious when one is assessing the v i a b i l i t y of a given mutation. Dfd has been known and studied for more than f i f t y years and a l l reports have assumed that lethality is a property of the Dfd mutation i t s e l f . This cautionary note is particularly important in the case of mutants isolated through the use of chemical or radiation mutagenesis, since their potent muta-genicity enhances the probability that double lesions w i l l be induced in single chromosomes. The lack of completely correct knowledge about the v i a b i l i t y of an a l l e l e is apt to mislead workers' attempts to genetically dissect a locus. Furthermore, the use of lethal flanking markers in recombinant systems demands precise knowledge of whether or 63 not an a l l e l e is lethal. Finally, studies of the developmental effects of specific genes would be d i f f i c u l t unless their v i a b i l i t y character-i s t i c s are well-defined. 64 CHAPTER 4 A STUDY OF INDUCED CROSSING OVER NEAR THE CENTROMERE OF CHROMOSOME 3 I. Introduction The lack of duplications and deficiencies for specific intervals has been one of the principal factors limiting the complete genetic dissection of Drosophila. The dosage dependent expression of many segments of the genome (Lindsley et a l . , 1972), particularly those where the structural genes of enzymes have been localized (Grell, 1962; Stewart and Merriam, 1974; Hodgetts, 1975), has permitted the cyto-logical mapping of several functions where aberrations which include them exist. However, segmental aneuploidy, produced by combinations of different translocations involving the Y chromosome and the autosomes (Lindsley et a l . , 1972), suffers from limitations stemming from the variety of disjunctional p o s s i b i l i t i e s . Therefore, workers have attempted to isolate heritably stable aberrations. The use of small, stable aberrations has permitted important genetic analyses. For example, a study of the interval between zeste and white on the X chromosome led to the conclusion that one functional unit exists in each polytene chromosome band (Judd e_t al_., 1972) . A recently reported method for the production and recovery of l o -calized aberrations involved selection of crossover chromosomes, where the crossovers were induced by radiation (Mglinets, 1972, 1973). Since normal meiotic crossing over occurs in females, i t follows that cross-overs arising from irradiated females would include both induced and 65 meiotic types. However, since no meiotic crossing over occurs in Drosophila melanogaster males, a l l exchange chromosomes recovered must be of the induced variety. The question of the origin of induced crossovers has previously centered on two distinct proposals: (a) in both sexes, induced ex-changes result from intimate pairing of homologues coupled with precise crossing over (breakage and rejoining), a situation analogous to the production of meiotic crossovers in females; or (b) in both sexes, induced exchange occurs in a manner similar to that of the production of translocations and therefore may involve mispairing and different break-points on each homologue and the potential for asymmetrical exchange. According to (a), induced crossover chromosomes should not be pre-ferentially associated with lethals and/or chromosome aberrations at or near the sites of exchange. Evidence for this has been provided by several studies. Patterson and Suche (1934) found that of a total of 59 third chromosome crossover progeny of X-irradiated males, only 9 carried recessive lethals. Moreover, most of these lethals were mapped to sites other than where exchange had occurred. They suggested that radiation might promote crossing over in males by releasing the normal constraints on meiotic crossing over. Friesen (1937a) found a similar situation for crossovers involving both chromosomes 2 and 3 as did whittinghill (1937) for heat-induced crossovers from males. Ives and Fink (1962) found that while only a low frequency of crossover progeny produced by gamma-irradiated males carried recessive lethals, translocations involv-ing non-crossover chromosomes were relatively frequent. Finally, Raytnayake (1970) found that the majority of crossover chromosomes of progeny produced by formaldehyde-treated males could be homozygosed. 66 Muller (1954, 1958) favoured the second alternative (b) and argued that the production of radiation-induced crossover chromosomes resembles the formation of translocations. In support of this idea, Herskowitz and Abrahamson (1957) found that radiation-induced exchange in centromeric regions in X chromosomes of females, conformed to a two-hit kinetic situation. O l i v i e r i and O l i v i e r i (1964) confirmed this in males and also showed that dose fractionation or administration in a nitrogen atmosphere decreased crossing over, while delivery df radiation in oxygen enhanced exchange. Recently, Williamson et a l . (1970) found that induced exchange involving the fourth chromosomes of females, followed two-hit kinetics. After studying her own data and those of other workers, Hannah-Alava (1968) concluded that there is a correlation between radiation-induced crossing over and the occurrence of recessive lethals and dominant s t e r i l i t y , particularly for meiotic broods sampled from male parents. She also found that crossovers involving intervals outside the centromeric region (the centromeric region is loosely defined as that region spanned by the most proximal markers used), were preferentially recovered in intermediate broods, while proximal crossovers (i.e. those occurring within the centromeric interval) were detected in later broods and were primarily associated with large clusters. Apparently, this brood pattern is a reflection of the meiotic and pre-meiotic origins of so-called non-proximal and proximal crossovers, respectively. She claims that earlier studies (e.g. Patterson and Suche, 1934; Whittinghill, 1937) failed to compensate for the occurrence of crossovers in clusters when estimating lethal frequencies amongst crossover stocks. She concluded that induced crossovers frequently arise in a translocation-like fashion or less l i k e l y , that exchange chromosomes must somehow be predisposed towards the possession of lethal sites. 67 Mglinets (1972) provided definitive support for the asymmetrical-exchange proposal through his finding that about 20 percent of re-combinant third chromosomes from gamma-irradiated males, possessed chromosome aberrations (especially duplications and deficiencies). Furthermore, nearly a l l of the aberrations had at least one breakpoint at or near the point of exchange. The majority of the crossover re-arrangements were present in 'meiotic' broods and few occurred in 'gonial' broods, a finding which parallels the results of Hannah-Alava (1968), He also found a significant correlation between sites of chromosome damage and sites of exchange in recombinant third chromo-somes derived from irradiated females. Thus, his data have raised the possibility that aberrations involving particular regions of the chromo-some can be recovered through the selection of appropriate crossover progeny. i The aim of this present study was to use radiation to induce crossovers within proximal intervals in chromosome 3 and to analyse the resulting crossover chromosomes genetically. I hoped that i t would be possible to determine i f such a method could provide a source of useful proximal deficiencies. 68 II. Materials and Methods Three separate experiments were performed. Heterozygous 2 p s st in r i eg Ki p e /+ + + + 4- + males, collected within 30 hours of, eclosion, were irradiated in gelatin capsules (the source of radiation was a cobalt-60 Gammacell in the U.B.C. Chemistry Department) and then 2 p s mass mated to homozygous st in r i eg Ki p e virgin females (these virgins were maintained for at least six days prior to mating) in quarter-pint milk bottles (using 8-10 males with 10 to 15 females per bottle). At the end of various intervals (depending upon the experi-mental protocol), new virgin females along with the irradiated males were added to fresh bottles. Wicks of f i l t e r paper were added to each bottle in order to maximize freedom of movement and mating a b i l i t y . The mass mating technique, coupled with observed differences in via-b i l i t y of female parents, precluded the use of a standardized defini-tive brooding procedure for any of the experiments. Two-hundred forty and 80 males were irradiated with 1000 R (Expt. I) and 2000 R (Expt. II) respectively and test crossed to females for six successive intervals of 3, 4, 4, 5, 5 and 5 days, for a total of 26 days. Two-hundred f i f t y males were irradiated with a dose of 3000 R and mated to females for five successive intervals of 3, 3, 6, 5 and 5 days, for a total of 22 days. Because the female parents frequently became mired in the wet food at the beginning of the third brood interval of this experiment, a new interval was started on the seventh day (the 55 non-crossover progeny recovered in the short interval were not included in the total) . 69 As a control, 90 males of the heterozygous genotype were test crossed for five 3-day intervals. The crossover (between s_t and p P) and non-crossover progeny were scored until the twenty-fourth day after the parents had been intro-duced. Although crossing over was not s t r i c t l y monitored between g and e_, several crossovers for this region were noted. 2 p s A l l crossovers and 30 st in r i eg Ki p e /+ + + + + + + non-crossover males (the latter were selected at random from cultures at days 12-22 in Expt. I l l ) were balanced with one of TM1, TM3 or CxD and tested for recessive lethality and each of the crossovers was coded according to its genotype. Sixteen of the lethal- and semi-lethal-bearing chromosomes (hence-forth the mutants w i l l a l l be referred to as lethals unless otherwise specified), collected from these experiments, were tested for comple-mentation inter se and with 4 cytologically-identifiable deletions lacking specific proximal segments in the right arm of chromosome 3 (see Table 2). In addition, some of the mutants were tested for complementation with the following dominant mutations: J?c, Ki, Msc, Q Sex and Antp ; and for pseudodominant expression of the recessive r visible mutations: tra, pb, Dfd , roe and dsx. .All of these l o c i are known to be genetically located in the proximal regions of chromosome 3 (see Table 1). Salivary gland chromosomes of 11 of the 16 mutant stocks were inspected by Dr. T. C. Kaufman using a standard technique. 70 I I I . Results 1. Radiation-Induced Crossing Over i n Males The crossover i n t e r v a l s w i l l hereafter be r e f e r r e d to as follows: 2 2 st to i_n - region 1; i n to r i - region 2; rji to eg - region 3, eg to K i - region 4; K i to p_^_ - region 5; and for reference, to e^ -region 6. Note that the centromere and proximal heterochromatin l i e i n region 4 (see Figure 1). Hereafter, region 4 w i l l be known as the Proximal i n t e r v a l while regions 1, 2, 3 and 5 w i l l be c o l l e c t i v e l y known as the Non-Proximal i n t e r v a l . Since t h i s study was designed to P s recover crossovers near the centromere and because some p_ to e_ cross-overs may not have been scored, data involving region 6 w i l l be considered 2 s only b r i e f l y . Although phenotypes such as eg , s t or e_ could a r i s e from mutation at a low frequency, a l l were scored as crossovers. Numbers of progeny and frequency of crossing over The control and experimental crossover data for the progeny of treated and untreated males are presented i n APPENDIX 1. For a l l three experiments i t i s evident (APPENDIX la) that the s i n g l e most frequent type of crossover occurred i n the Proximal i n t e r v a l as has been pre-vi o u s l y noted (see Hannah-Alava, 1968). In Expt. I, the t o t a l number of Proximal crossovers was s l i g h t l y l e s s than that of Non-Proximal crossovers (19 versus 22), while for both Expts. II and I I I the numbers of Proximal types greatly exceeded those of the Non-Proximal types (Expt. I I , 13 versus 3; Expt. I l l , 63 versus 12). Double crossovers were detected and these have been reported i n an e a r l i e r study (Mglinets, 1972). Table 15 Distribution of Crossovers in Proximal and Non-Proximal* Intervals for st t o ^ Exchanges Crossovers (Crossover Events)** Treatment of Intervals  Male Parent 1 _3 4 5_ Expt. I 11(7) 2(2) 20(8) 9(6) 1000 R Expt. II 3(3) - 13(5) 2000 R Expt. I l l 4(4) - 61(19) 5(4) 2(2) 1(1) 3000 R 3,4 5,6 * Proximal = Exchange in region 4 Non-Proximal = Exchange in region 1, 3 or 5 ** Crossover Event = Each cluster counted as one event 72 APPENDIX lb is a summary of the total crossover progeny for the entire st^ to p^ _ and p_^_ to e^ regions as well as an estimate of crossing over between s_t and p^ _ in Expt. III. Since i t is thought that gametic samples of earlier broods are mainly post-meiotic (i.e. with respect to irradiation, see Hannah-Alava, 1968), the crossover frequency includes only progeny recovered from day 7 to 22 inclusive. Furthermore, only the progeny of fourteen randomly selected cultures which had pro-duced a minimum of 25 progeny per brood interval, were used for this estimate. Thus, the estimated frequency of induced crossing over in males treated with 3000 R is 0.59 percent, while the comparable level for untreated males is 0.078 percent. A similar analysis of Puro's (1966) data gives a crossover frequency of 0.65 percent. The fact that he used a single male technique probably accounts for the frequency difference. Numbers of crossover events Presumably, radiation-induced crossovers can arise either in germ cells which have stopped dividing (i.e., those that have reached late gonial or early meiotic stages), or in germ cells which are s t i l l dividing. In the former case, one would expect to recover single crossover progeny of a given type (or of each reciprocal type); while in the latter case, one could recover clusters of identical crossovers (and/or reciprocals) that arose from a single exchange event. The size of the cluster would depend upon the number of gonial divisions occurring after irradiation. In the present work, the crossover frequency is low (about 1 in 150 to 200 progeny). Therefore, the probability that 2 (or more) pheno-typically identical crossover progeny in the same culture had arisen from independent crossover events, is even lower. Consequently, when 73 2 2 or more of the same (e.g., st in r i eg ) or reciprocal (e.g., 2 p s st in r i eg and Ki p e ) types were seen in the same culture, they could be collectively scored as a single crossover event. Many clus-D S ters were observed, with the largest including 26 Ki p e progeny in two consecutive brood intervals (Expt. III). Table 15 gives the total numbers of crossover progeny, along with the numbers of crossover events (in parentheses) for the three experi-ments. These data underscore the fact that in Expt. I, Non-Proximal and Proximal crossovers occurred at the same frequency and in Expts. II and III, the Proximal types were scored much more frequently. How-ever, when crossover events are considered, the Non-Proximals are about twice as frequent as the Proximals in Expt. I, while the differential between the two types is markedly reduced in both Expts. II and III (Expt. II, 13 Proximal:3 Non-Proximal crossovers versus 5 Proximal:3 Non-Proximal crossover events; Expt. I l l , 63 proximal:12 Non-Proximal crossovers versus 19 Proximal:ll Non-Proximal crossover events). Thus, i t seems that clustering is more characteristic of crossovers occurring within the region which includes heterochromatin than of those occurring outside of this region. This may indicate that Proximal crossovers are primarily gonial in origin, while Non-Proximal crossovers are not and this would support the findings of Hannah-Alava (1968). Recovery of mutants amongst crossovers Table 16 summarizes the distribution of recessive lethals (includ-ing 3 semi-lethal visibles) and putative dominant steriles amongst the various types of crossover chromosomes. The appropriate numbers of lethal or sterile events (after correcting for clusters) are included in parentheses. A total of 16 independently-induced lethals (or semi-lethals) was recovered amongst 52 tested crossover chromosomes which Table 16 Regional Summary of Lethals* and Steriles Present on Crossover Chromosomes Experiment Total Total Number _1 _3 _4 _5 _6 3,4 Proximal Non-Proximal"' Lethals 1(1)*"* - 1(1) 5(3) 1(1) - 1(1) 6(4) I Steriles 2(2) 1(1) 1(1) - - - 1(1) 3 ( 3 ) II III Lethals 1(1) - 1(1) - - - 1(1) 1(1) Steriles 1(1) _ _ _ _ _ _ Lethals 3(3) - 3(3) 2(2) 2(1) 1(1) 3(3) 6(6) Steriles - - 2(2) 1(1) - 1(1) 2(2) 2(2) * Includes semi-lethals ** Not including interval 6 *** Number in parentheses = Number of lethal or sterile events (each cluster counted as one event) 75 arose from independent exchanges between sj: and pjj 5 in Expt. I, 2 in Expt. II and 9 in Expt. III. The totals show that overall, most of the lethals occurred in Non-Proximal crossover chromosomes (11 of 16). A total of 9 of the crossover progeny were sterile (3 Proximals and 6 Non-Proximals). Since the mass mating technique makes i t d i f f i c u l t to define these as genuine steriles, they w i l l not be mentioned further. Regional comparison of crossovers and mutants. Table 17 shows the relative occurrence (as percent of total cross-over events) of Proximal versus Non-Proximal crossover events for the three experiments as well as the frequencies (in percent) of lethals amongst the crossover events. The latter frequencies were based upon the percentage of independent crossover stocks which carried lethals (since steriles could not be tested, they were omitted). These data (Table 17) reveal that in Expt. I most of the independent crossovers occurred outside the Proximal interval, while this trend was reversed at the higher doses of radiation (although in Expt. II only 2 Non-Proximals were tested since 1 was s t e r i l e ) . While 14 percent of the tested independent Proximal crossovers were associated with lethals in Expt. I, these frequencies were 20 percent in both Expts. II and III. In contrast, the lethal frequencies for Non-Proximal crossover events were 33.3, 50 and 66.7 percent in Expts. I, II and III, respectively (in Expt. II only two stocks could be tested). For reference, i t was determined that 16.7 percent (5 of 30) of wild-type parental chromo-somes from day 7 to 22 in Expt. I l l , carried recessive lethals. The above results support the idea that there is a preferential association of lethality with proximal crossovers that are induced in Table 17 Relative Occurrence of Crossovers and Lethal Events for Proximal and Non-Proximal Intervals Experiment Number I I I III Crossover Events (Percent of Total) Proximal 8(35.0) 5(62.5) 19(63.3) Non-Proximal 15(65.0) 3(37.5) 11(36.7) Number of Lethal-Associated Crossovers (Percent of Region-Specific Total)* Proximal 1(14.3) 1(20.0) 3(20.0) Non-Proximal 4(33.3) 1(50.0) 6(66.7) *Percent Lethality =((number of lethal events)/(number of crossovers - number of steriles)) x 100 the euchromatic portions of the chromosome in the region between st and p^ _. It would also appear from these results that more induced crossing over occurs (or at least is detected) outside of the Proximal interval at 1000 R, while at higher doses of radiation (i.e. 2000 and 3000 R), this trend is reversed and Proximal crossover events are more frequently observed. The question of whether the latter difference (between Expts. I, and II and III) is real or is due to artifacts of the technique, must await more analyses involving precise brood and single male studies. 2. Analysis of Lethal Stocks Inter se complementation, pseudodominance, and additional tests The results of the inter se and deficiency complementation involv-ing the 16 crossover mutants are shown in Table 18. The mutants have been coded according to their genotypes and where more than 1 member of a given class occurred as a lethal, each was assigned a different 2 number. The mutant stocks st-3, st-4 and st in r i eg -4 actually possessed recessive semi-lethal sites and in each case less than 10 percent of the expected number of homozygotes survived to adulthood and these were extremely small (about 1/3 to 1/2 of normal size). It is clear from the inter se complementation results that most mutant combinations were viable. The non-complementing exceptions are: (a) st-1, st-2; (b) st in r i eg^-2, st in r i eg^-3; (c) in r i eg^ Ki p P e S, 2 2 st in r i eg -2; and a group of five (d) st-3, st-4, st in r i eg -4, 2 2 st in r i eg Ki-1 and st in r i eg Ki-2. Note that in groups a and b, non-complementation occurred between different lethals associated with crossovers within the same region, whereas in group c, crossing over Table 18 (A) (B) (C) (A) (B) (0 (D) (0 (F) fo) (H) (I) (J) <-> (L) (M) (N) (0) (P) •HUI +R2 DfR2 "&4-R5 Remits of Inter Se and Deficiency Complementation Crosses InvolvlnR Crossover Lethals (D) ( E) (F) (G) (H) (J) (K) , (D 2 (M) (N) (0) (P2-St-4 In r l eg 2 Kl pP e 8 s t in r l eg 2-l st in r l es2-2 st in r l en2-3 st in r i eftz-4 Kl p p e s st ln r l en K l - l s t in r l eg Ki-2 PP e s - l pP es-2 pP es-3 + + + + + + + + + + + + + + + + + + + + + + •t- + + _ + + + - + - - + + + + + + + + - + - - + + + + + + + + + + + + + + + + + + + + + + + + + + + + + , - + + + + + + + + + - - + f + + + + + + + + - + + + + + + + + + + + + + + + + + + + + + - + + + + + + + + + + + - + + + + + + - + + + + + + + + + + + - + + + + + + 00 79 occurred in different regions for the 2 lethals ( i t is possible that 2 st in r i eg -2 carries two lethal sites). Group d presents a hetero-geneous situation i.e. non-complementation occurred between similar as well as different crossovers. It should be mentioned that in the 2 latter group, both st in r i eg Ki mutant stocks were lethal as homo-zygotes and in combination with each other, but in combination with any of the other 3 members of this group, produced the characteristic small f l i e s at low frequencies. Most of the combinations between the crossover mutants and the proximal deficiencies were viable (Table 18). Four exceptions to this T,. p s,, EH-R2 p s,, EH-R5 p s ... ,_ +R2 , p s w e r e Ki p e /dsx , Ki p e /dsx , p e -3/Antp and p e -3/ N s — — S i n c e a l l of these deficiencies lack chromosome material between Ki and _p_P_, these data reveal that both the Ki p P e S and p P e S chromosomes possess lethals near or at the sites of exchange. The ~f~R21 D4"R2 failure of the Ns and dsx deficiencies to complement with 2 st in r i eg -1 w i l l be dealt with later. Table 19 represents a summary of pseudodominance and additional complementation tests involving four of the crossover lethal stocks and known proximal mutations in chromosome 3. Most of the combinations were non-mutant. However, a l l tra/st-1 progeny were males, thereby indicating that the st-1 stock f a i l s to complement the tra mutation. D S In addition, p e -3 was lethal when combined with Msc, Sex and Antp. Ns There is good evidence that Sex and Antp— are alleles (see Denell, 1973; Duncan and Kaufman, 1975), while the relation of both of these with Msc is unclear. Finally, the Ki/p e -3 combination was viable but the fl i e s displayed an enhanced Kinked phenotype reminiscent of Ki/Ki f l i e s . 80 Table 19 Summary of Pseudodominance and Complementation Tests Between Selected Crossover Mutants and Known Proximal Mutations  Proximal r Mutants tra Pc pb Ki Dfd— Msc Sex Antp roe dsx Crossover Mutants st-1 -*+ + + + + + + + + 2 st in r i eg -1 + + + + + + + + + + not Ki p P e S + + + done + + + + + + p P eS-3 + + + e + -** -** -** + + * A l l tra/st heterozygotes were males ** Lethal e = Ki appeared to be enhanced 81 Cytological analysis The results of the cytological analysis of 11 of the 16 crossover mutants are given in Table 20 along with some phenotypic descriptions. Eight of the 11 stocks examined displayed no obvious cytological dis-ruptions. However, the remaining 3 were a l l abnormal and displayed the following aberrations: st-1 = Df(3L)72F-73A;74BC, in r i eg 2 Ki p P e S 2 = In(3L)70C;74A, st in r i eg -1 = Df(3L)79E5-6;80 heterochromatin. Note that a l l of these stocks show low v i a b i l i t y as heterozygotes and 2 in addition, the basal deficiency (st in r i eg -1) includes a Minute, probably M(3)LS4 (Lindsley et a l . , 1972). The extent of the hetero-chromatic deficiency of this stock is not known. The low v i a b i l i t y of 2 -HR21 D+R2 heterozygotes involving the st in r i eg -1 and the Ns_ or dsx  deficiencies, may be due to the combined effect of the deficiencies in each case. It is worthy of mention that since the recessive lethality of Pc is not exposed by this basal deficiency (see Table 19) and because this gene has recently been localized to the interval between 75A-B to 80 (Puro and Nygren, 1975), these present results further narrow the interval containing Pc_ to 75A-B to 79D or E in proximal 3L. The fact that st-1 is a deletion explains i t s lack of complementa-tion with both st-2 and tra. Thus, in the case of a l l 3 aberration-bearing crossover chromosomes, the site of damage for at least one of the breakpoints corresponded to the region where crossing over had occurred. P S u s p e -3- and Ki p^ e -bearing f l i e s were weakly viable and since genetic evidence supports the idea that the former is a deletion which Ns exposes the Antp -Scx-Msc homeotic region as well as Ki, i t is sur-prising that neither of these stocks possess chromosome abnormalities. Table 20 Phenotypic Description and Cytological Analysis of Crossover Mutants  Crossover Phenotype (other than markers) _ st--1 Low v i a b i l i t y (as heterozygote) Df(3L)72F-73A; 74BC st--2 - + * st-•3 Low frequency recovery of very small sterile homozygotes + in r i eg 2 Ki P s P e Low v i a b i l i t y (as heterozygote) and infrequent In(3L)70C;74A haltere enlargement st in r i 2 eg --2 Minute, low v i a b i l i t y (as heterozygote), low Df(3L)79E5,6; female f e r t i l i t y , thick aristae, rough eyes 80 heterochromatin Ki P P s e Low v i a b i l i t y , gaps in wing vein L2 + st in r i 2 eg Ki-1 Low frequency recovery of very small homozygotes + P P s e --1 - + P P s e --2 - + P P s e -•3 Low v i a b i l i t y , sex combs have fewer teeth (5-7) than wild type + eg' I Slight broadening in wing vein L2 near tips, rough eyes + 83 Other complementation tests 2 Since the mutant stock eg /TM3 displayed a Delta-like phenotype, i t 2 was tested with an a l l e l e of DI. No DI/eg heterozygotes were scored, 2 thereby indicating that the eg chromosome carries an a l l e l e of DI. Tasaka and Suzuki (1973) recovered an EMS-induced ts visible muta-tion on the third chromosome, which although normal at 22°, when grown at 29° C produced homozygotes that were undersized. When this mutant (1 (3)ET*1S'*"^) was retested and crossed to the st in r i eg 2 Ki-1 stock, normal heterozygotes were produced at both 22° and 29° C. It is possible that the small body mutation which is characteristic of the inter se complementation group (d) is due to contamination. However, 2 the recessive lethality carried by both st in r i eg Ki members may be due to a second site which could s t i l l correspond to the regions of exchange. Of the five recessive lethals detected amongst the 30 non-crossover chromosomes tested (from Expt. I l l ) , four were distinct (i.e. two did not complement). None of these four was exposed by any of the six deficiencies which were at my disposal (the four described in Table 2 and the two newly synthesized deletions from the present study). In summary: I detected a total of 16 associated mutations amongst 52 independently-occurring crossovers. Three of the 16 possessed structural abnormalities coinciding with the sites of ex-2 s 2 change (st-1, in r i eg Ki p P e S, and st in r i eg -1) and the lesions p s p s of three additional lethals (st-2, p e -3 and Ki p e ) were cyto-logically mapped to sites close to or at the regions of exchange. Thus, at least 6 of the 16 mutants show correlations between the position of 84 lethal sites and sites of crossing over. The failure of st in r i eg^-2 2 and st in r i eg -3 to complement each other may indicate a similar correlation for these 2 lethal crossover chromosomes, thereby further increasing this total to 8 of 16. 85 IV. Discussion It is clear from the present study that selection of induced cross-overs between proximal markers in irradiated males, can enrich for proximally-located lethals and more specifically, for aberrations. Furthermore, the more frequent types of aberrations found in this case were deficiencies. Since the cytological positions of both the aberra-tions and many of the non-aberrant crossover lethals were very similar to the regions of exchange, these data support the contention that at least some of the crossover chromosomes originated from asymmetrical exchange events. The results of this study reveal that crossovers occurring within proximally-adjacent euchromatic segments are more frequently associated with mutations, than are crossovers occurring within the more proximal segment which spans the heterochromatin. This observation is consistent with the findings of Hannah-Alava (1968) and Mglinets (1972). At least 3 factors which could contribute to the apparent lower susceptibility of proximal exchanges to the concomitant occurrence of mutations are: (i) relatively few sites capable of being mutated to lethality exist in heterochromatin as compared to euchromatin (Muller et a l . , 1937; H i l l i k e r and Holm, 1975; H i l l i k e r , 1976), ( i i ) a mechanism which exists in the testis could allow for the selective elimination of pre-meiotic stem cells possessing chromosome breaks (Puro, 1966), and ( i i i ) stage-specific differences in pairing of homologous chromosomes during spermatogenesis could allow for more complete pairing of homo-logues in gonia and therefore promote reciprocal crossing over (Hannah-Alava, 1968; Mglinets, 1972). 86 The relevance of (i) is self-evident. Puro (1966) has found, evi-dence for a regenerative stem c e l l mechanism in the testes of Drosophila. Therefore i t is possible that where breakage of unpaired homologous chromosomes does occur in stem cells, resulting in crossovers which involve chromosome damage, these cells would be lethal, thereby account-ing for the relative dearth of crossovers in broods derived from such cells (see Hannah-Alava, 1968). The results of this present study corroborate those of earlier workers in showing that proximal cross-overs are more frequently associated with clusters. Such a selective mechanism as described above would likely lead to a situation where fewer proximal crossover chromosomes contained lethals. Finally, both Hannah-Alava (1968) and Mglinets (1972) have referred to evidence which suggests that homologous pairing occurs to different degrees during gonial stages and meiosis. Thus, more complete somatic pairing prior to meiosis would permit more exact homologous interchange to take place following radiation-induced chromosome breakage, thereby producing a situation analogous to that of meiotic exchange in females. The stable deficiencies isolated in the course of this study are potentially very useful. Thus, st-1 (Df(3L)st) has permitted the cyto-2 logical localization of tra and st in r i eg -1 (Df(3L)M(3)LS4) has further c l a r i f i e d the position of Pc in the l e f t arm of the chromosome. It is worthwhile to emphasize that the latter deficiency probably lacks some of the heterochromatin in 3L, although the extent of this hetero-chromatic deletion is unknown. The existence of such a mutant w i l l be an important tool for use in the continuing search for genes within heterochromatin. 87 The induced crossover technique offers a unique approach for the study of proximal regions of chromosome 3. To increase the resolution of this method, i t should be possible to construct a genetic screen whereby induced exchanges would be selected for and non-crossovers eliminated. For example, males containing two ts mutations (spanning the region of interest) linked in trans, could be irradiated and crossed to females homozygous for both ts mutations at 22°C. If the resulting cultures were raised at 29°C, only wild-type crossover pro-geny would be expected to survive, thereby enriching the selection system for radiation-induced crossover chromosomes. CHAPTER 5 A GENETIC AND DEVELOPMENTAL STUDY OF Q-III, A TEMPERATURE-SENSITIVE MINUTE MUTATION I. Introduction The phenotypic similarities between bobbed, some alleles of suppressor of forked and Minutes have been used to argue that they a l l represent defects in protein synthesis, since protein synthesis is abnormal in both bb (Ritossa et a l . , 1966a) and 1 ( l ) s u ( f ) t s 6 7 8 (Dudick e_t a_l., 1974; Lambertsson, 1975b). No evidence has been presented which unequivocally identifies a common mode of action for the various Minute l o c i . In fact, i t has been proposed that although a l l of these mutants probably do inhibit protein synthesis, mutants of different l o c i could be acting at different levels in the overall process (White, 1974). Thus, while some Minute genes could code for different iso-accepting species of tRNA, others could code for processing enzymes for tRNA or other molecules, amino-acyl synthetases, ribosomal proteins or any of the multitude of structural and functional components which comprise this system. The hypothesis that Minute l o c i code for tRNA (Ritossa et a l . , 1966b; Atwood, 1968) has attracted considerable interest because i t provides a plausible explanation for the identical phenotype of many different l o c i . Since tRNA genes are redundant, this hypothesis also explains why mutations at M l o c i are often deletions. Furthermore, the number of l o c i corresponds roughly to the expected number of tRNA 89 species. Experimentally, the tRNA-Minute relationship can be explored by correlating biochemical studies (viz. i_n situ hybridization, chromatographic isolation and nucleotide sequencing etc.) with genetic tests such as segmental aneuploidy (Lindsley et a l . , 1972) and fine structure mapping. The importance of further genetic study of Minutes therefore, can-not be overemphasized. In this regard, the recovery of EMS-induced Minutes (Holden and Suzuki, 1973; J. Stone, unpublished results; Huang and Baker, 1975) is particularly noteworthy,as EMS is assumed to induce single base transitions. The assessment of developmental anomalies of Minutes which are associated with cytologically-observable deletions, is fraught with d i f f i c u l t i e s . However, temperature-sensitive EMS-induced Minutes which are presumed to be point mutations (Suzuki, 1970), should permit more definitive and accurate examination of the onto-genic basis of defects produced by such lesions, especially during embryogenesis in Minute homozygotes. Moreover, there is evidence of the existence of recessive lethal sites which have no dominant visible effects and do not complement standard Minute alleles (A. Datagupta, personal communication). This presents the possibility that Minute l o c i are complex and that the analysis of different alleles and their bio-chemical properties w i l l provide an insight into the function(s) of Minutes. A significant proportion of EMS-induced mutations is temperature-sensitive in Drosophila (Suzuki, 1970). The relevance of such mutants to genetic, biochemical and developmental analysis of this organism should considerably expand the scope of Minute investigations. For 90 example, ts alleles of Minutes could allow construction of recombina-tion selective systems, thereby increasing the resolving power in fine structure studies of such l o c i . Furthermore, i f definitive evidence concerning the primary gene products of Minutes is forthcoming, the existence of a conditional a l l e l e could allow the unequivocal identi-fication of the potentially thermolabile gene product and its _Ln vivo and i_n vitro biochemical characterization. Temperature shift experiments during development have provided a wealth of information about the interval(s) when the gene product of a given locus is utilized (i.e. temperature-sensitive period, see Suzuki, 1970, 1974b). Furthermore, heat pulse experiments involving ts mutants or tests of such stocks at middle temperature ranges, have resolved phenotypes that are usually masked by lethality in orthodox TSP shift studies (Poodry e_t ajL., 1973; Tasaka and Suzuki, 1973). Heat pulse studies of a ts allele of a Minute locus with respect to the attenuated bristle phenotype and TSPs for any other M effects, would help to delineate the extent of Minute function during development. It is noteworthy that in their search for DTS lethals on chromo-some 3, Holden and Suzuki (1973) isolated two mutants, DTS-1 and DTS-6, which displayed the dominant Minute phenotype at 22°C as well as domi-nant lethality at 29°C. Neither could be made homozygous at 22°C. Owing to semi-sterility of DTS-6 females at 22°C, an accurate genetic location could not be found. However, the ts dominant lethality seemed to map proximally. Crossover studies involving DTS-1 placed the M and dominant lethal phenotypes at an identical site in a distal segment of 91 the right arm of the chromosome. Furthermore, the lethal phase (LP) of DTS-1 homozygotes is embryonic and this is also true for other known Minutes (see Brehme, 1939; Farnsworth, 1957a,b). It appears therefore that temperature-sensitive Minutes can indeed be induced and recovered. Since the autonomous cell-lethal nature of Minutes has been demon-strated (Stern and Tokunaga, 1971) and because many important features of pattern formation in imaginal discs have been described through the study of X-linked, ts autonomous cell-lethals (Russell, 1974; Simpson and Schneiderman, 1975; Arking, 1975), i t is likely that a ts Minute could provide similar as well as unique information about the develop-ment of Drosophila. The existence of a ts mutation that interacts phenotypically with a distinct, non-ts mutant has been used to define the time of activity of the non-ts mutant (Dudick et a l . , 1974). Thus, the ts a l l e l e of su(f) was used to determine the time of activity of the forked gene. Since i t is known that Minutes interact phenotypically with several non-allelic genes (see Schultz, 1929; Lindsley and Grell, 1968), a ts Minute might be exploited in a similar fashion and the time of the interaction between the Minute locus and the product of another non-ts locus determined. This chapter is a preliminary report on such genetic and develop-mental studies of a ts Minute, located near the centromere of chromo-some 3, which I recovered. The study was initiated to determine the potential u t i l i t y of the mutation in exploring both Minute function during development and i t s interaction with other l o c i . 92 II. Materials and Methods The temperature-sensitive Minute, Q-III was recovered by chance in a screen for EMS-induced, ts alleles of a known third chromosome mutation. The retarded development of Q-III/4- fl i e s at 29°C i n i t i a l l y resulted in the misclassification of this mutant as a lethal a l l e l e of the test chromosome. Crosses of Q-III with Oregon-R or _ _ P / j _ P resulted in the production of heterozygotes at 29°C. This led to the prelimin-ary classification of Q-III as a ts Minute. Henceforth, I w i l l simply refer to the temperature-sensitive Minute as Q-III. It should be noted that a l l experiments involve Q-III linked to j>P. Chromosomes used to balance Q-III include: TMl, CxD, and TM3. A l l references to TM3 throughout the text signify TM3, Sb Ser unless otherwise indicated. For f u l l descriptions of these balancers, see Lindsley and Grell (1968). Unless specified otherwise, a l l mapping, complementation and other crosses for assessing the properties of Q-III were carried out in quarter-pint milk bottles with standard Drosophila medium. Ten pairs of parents were introduced into each bottle and these were usually sub-cultured at least once on fresh medium. Where tests at 29°C were made, the females were allowed to lay for 1 or 2 days at 22°C before they were removed and then the culture bottles were shifted up to 29°C. A standard method of egg collection on petri dishes was used for the developmental and some of the genetic studies (see Tarasoff and Suzuki, 1970). Generally, the f i r s t two 2-hour batches of eggs laid were discarded and the third was used either by counting the eggs and 93 shifting the plates directly or more usually, by picking required numbers of eggs along with some of the medium and placing them in pre-incubated vials or petri dishes for shifts or other analyses. To test for v i a b i l i t y , lethal phases and lengths of developmental periods of Q-III homozygotes and heterozygotes, 22° and 29°C batches of eggs were collected in two control crosses: I j3PAp_P x £P/£P'» II CxD/TM3 males x ,£P/p_P females; and three experimental crosses: (A) Q-III/TM3 females x £ P / £ P males, (B) Q-III/TM1 x Q-III/TM1, (C) Q-III/TM3 x Q-III/TM3. The eggs were counted and transferred to pre-incubated 50mm petri plates on fresh medium (50 to 100 eggs per plate) which were then placed at 17°, 22° or 29°C. No attempt was made to distinguish between f e r t i l i z e d and unfertilized eggs (Wright, 1973) during this procedure. Stage-specific distribution of lethality was estimated for the control and 2 of the experimental crosses by inspecting cultures intermittently and computing the proportions of expected progeny which successfully survived the egg, larval and pupal stages. In some cases, lengths of developmental periods were estimated by inspecting the cultures at various intervals and noting the time when at least half of the total surviving progeny had eclosed, thus providing the period of time (in hours) from oviposition to eclosion. 1. Genetic Analysis Genetic mapping Q-IIIpP/TM3 males were crossed to Gl Sb H/Payne (see Table 1) or 2 2 o st in r i eg /st in r i eg females at 22 C. Fl Gl Sb H/Q-III p P and st in r i eg^/Q-IIIpP females were then crossed separately to <£P/p_P males. A l l matings were at 22°C. In the former cross, six bottles of 94 3-day cultures (3 originals, then sub-cultured) were kept at 22°C and 6 were transferred to 29°C. In the latter cross, the 3 originals and 3 subcultures were shifted to 29°C, while two more broods were retained at 22°C. Progeny of a l l bottles were scored until the twentieth day after the parents had been introduced. As many as possible of the Q-III-bearing recombinants (or putative multiples) recovered at 29°C, and a l l recombinants from 22°C cultures, were tested to verify their genotypes with respect to Q-III or other recessive markers. In addi-tion, the 22° recombinants were crossed at 29° and 22°C so that their progeny could be scored for other phenotypic traits associated with the Q-III chromosome. Complementation of Q-III with other proximal mutations on chromosome 3 Df(3L)M(3)LS4 is a Minute mutation associated with a cytologically-observable deletion in a proximally-located segment of the l e f t arm of chromosome 3 (see CHAPTER 4). Thus, males from this stock were crossed to Q-III/TM3 females to test for complementation at 22° and 29°C. Q-III heterozygotes sometimes display reduced eyes at 29°C. Since Ns this is also a t r a i t characteristic of Dfd and Antp—, males from the stocks Dfd pP/Dfd p P, Ns/Ns and ru h Ki Antp G eS/TM3 were crossed to Q-III/TM3 females to test for complementation at 22° and 29°C. 2. Developmental Analysis Tests for s t e r i l i t y and maternal effects Fifteen homozygous Q-III/Q-III females (0 to 24 hours in age) were individually mated wi th 5 __P/p_P males in shell vials and transferred for four consecutive 2-day broods to fresh vials at 22°C. Subsequent-ly, fresh males were added and the cultures exposed to 28°C for 95 two additional 3-day broods. New males were added for the second 28°C brood. The vials were l e f t at their respective temperatures and later examined for the appearance of any developmental stages. A similar study involving 15 Q-III/Q-III males was initiated. To test for any maternal effects, Q-III/Q-III females were mass mated to homozygous _£P/_£P males at 22°C and 322 eggs were collected over a 2-hour period. Of these, 102 were shifted to 29°C while the remaining 220 were kept at 22°C. A l l eggs were periodically examined for signs of development. Regular temperature shift studies For detailed descriptions of the rationale and experimental pro-cedures for determining TSPs, see Tarasoff and Suzuki (1970) and Suzuki (1970). For the present study, the beginning of the TSP was usually defined as the f i r s t point when a culture that was shifted to the per-missive temperature produced significant numbers of mutant animals (or decreases in v i a b i l i t y ) , and the end of the TSP was defined as the f i r s t point when a culture that was shifted to the restrictive tempera-ture produced non-mutant animals (or significant levels of v i a b i l i t y ) . Developmental stages present in the cultures were determined either by inspecting cultures at 12-hour intervals (or in a few cases, at 6-hour intervals) for the duration of development at 22° and 29°C or by scoring one of the cultures at the time of each shift. Since a l l shift experiments involved crosses producing a minimum of two classes, at least 20 (usually 30 to 40) progeny were staged at a given interval. The standard method of scoring larval mouthparts was used to distinguish between the different larval stages (Bodenstein, 1950). 96 Cultures were shifted from permissive (22°C) to restrictive temp-eratures (29°C) and vice versa at 12-hour intervals. A l l cultures were inspected every 12 to 18 hours and progeny scored for up to 25 days after the eggs were collected. (a) TSP for recessive lethality of Q-III A total of 400 to 500 eggs (50 to 60 eggs per vial) was shifted at each 12-hour interval after oviposition. Developmental stages reached in the cultures at the time of shifts were assessed by scoring the progeny in one extra v i a l . Furthermore, a detailed assessment of stages reached in cultures kept at 22° or 29°C was provided by inspec-tion at 6-hour intervals during the larval stages. (b) TSP for dominant eye and bristle phenotypes of Q-III Since Q-III heterozygotes at 29°C possess rough and less frequently, reduced eyes as well as short and thin bristles, the TSPs for these phenotypic traits were studied. The reciprocal crosses Q-III/TM3 females x J_P/__P males and Q-III/TM3 males x _ _ P / j _ P females were used to provide eggs. Since results in the two lines were similar, the samples were pooled. A total of 400 to 500 eggs was shifted (50 to 60 eggs per vial) at each interval. Developmental stages present in the samples were determined at the time of sh i f t . The larval stages present were further assessed by 6-hour inspection of parallel cultures at 22° and 29°C. The adult f l i e s were scored for the occurrence of roughened eye surfaces and bristle disruptions. (c) TSP for vg-Q-III interaction The recessive mutation vestigial (vg), on chromosome 2 (at 67.0, see Lindsley and Grell, 1968) causes a marked reduction of wing size. Since i t is known that in the presence of Minutes, vg/-fr f l i e s exhibit 97 wing scalloping (Green and Oliver, 1940), a vg_ stock was crossed to Q-III and the resulting 22° and 29°C progeny examined (see the part on Q-III interactions). At 29°C a high frequency of the double heterozy-gotes displayed nicked wing margins (particularly d i s t a l l y ) , while at 22°C, essentially no interaction was apparent. It was therefore decided to study the TSP of this interaction. Only one set of experi-ments was performed. Eggs were collected from the cross +/+; Q-III/ TM3 females x vg/vg;+/+ males, and 100 to 200 (50 to 100 per petri plate) were shifted at 12-hour intervals. Parallel 22° and 29°C cul-tures were examined every 12 hours to determine the developmental stages present. The adult f l i e s were scored for the scalloped wing phenotype. (d) TSP for Dl-Q-III interaction Since Schultz (1929) reported that some Minutes exhibit lower v i a b i l i t y when combined with different alleles of Delta (see Table 1 and Lindsley and Grell, 1968), i t was thought that Q-III might 2 interact similarly with DI. Preliminary crosses of a JJ1 stock (eg DI see CHAPTER 4) with Q-III produced no Dl/Q-III heterozygotes at 29°C, while normal heterozygotes survived at 22°C (see Q-III interactions). It was therefore decided to study the TSP of this interaction. Eggs were collected from the cross Q-IH/TM3 females x D1/TM3 males, and 250 to 300 of these were shifted at 12-hour intervals. Developmental stages present at the time of shift were determined by inspecting the culture in an extra vial for each shift at both temperatures. Q-III/D1 flies could be distinguished by their extremely retarded development at 29°C. The survival of adult f l i e s was scored at both temperatures. 98 (e) TSP for Scx-Q-IU interactions In testing for Scx-Q-IH interactions (these were primarily de-signed to see i f the sex comb phenotype of Sex could be suppressed, see the part on Q-III interactions), i t was discovered that Sex/Q-III heterozygotes exhibit low v i a b i l i t y at 29°C as well as marked scallop-ing or nicking of the posterior wing margin, while exhibiting no such phenotypic interaction at 22°C. It was decided to study the TSP of this interaction. Eggs were collected from the cross, Q-III/TM3 females x Scx/TM3 (ru h st r i Sex p P eS/TM3) males, and 200 to 300 of these (50 to 60 per vial) were shifted at 12-hour intervals. Develop-mental stages present in cultures at the time of shift were determined as for (d). The adult f l i e s were scored for the presence of nicks in the wing margin. It should be mentioned that the distinction between Q-III and other heterozygous progeny classes was considerably facilitated by the retardation of growth of Q-III larvae at 29°C. This was also true of Q-III/Q-III homozygotes at 22°C. Furthermore, both the Q-III homozy-gotes at 22° and heterozygotes at 29°C frequently possess diagnostic internal melanization which provided an additional marker. Pulse shift studies Eggs of Q-III/TM1 x Q-III/TM1 matings at 22°C were collected at 2-hour intervals and at various times during development, shifted to 29°C for a period of 24 to 48 hours. The numbers of eggs tested varied for different intervals (see APPENDICES 2 and 3). Petri plate cultures were used for the shifts (50 to 100 eggs per plate) and owing to the differential rate of development of Q-III, no synchronization 99 was attempted. Several plates were allowed to develop continuously at 22° and 29°C and the developmental stages reached in these cultures were assessed every 12 hours. The progeny were scored daily until at least 20 to 25 days after the eggs had been collected. The numbers of survivors and the occurrence of various phenotypic traits were noted. In some cases, imagoes incapable of emerging from the pupal cases, were dissected and inspected. The TSP of a particular mutant defect was defined as the developmental interval when exposure to 29°C would e l i c i t the mutant phene in a significant proportion of the progeny. Scanning electron microscopy Selected fl i e s were anaesthetized with CO^ , mounted on chucks using silver conductant paint and examined alive in a scanning electron microscope (SEM, Cambridge Instruments, Cambridge, England). Q-III interactions It has been found recently that some Minutes suppress the expres-sion of the extra sex comb phenotypes of the various sex comb homeotic mutations (R. Denell, personal communication). In order to test whe-ther Q-III is capable of effecting similar suppression, groups of Q-III/ s TM3 females were crossed separately to ru h Msc e /TM3, ru h st r i Sex p P eS/TM3 and Pc3/TM3 males in bottles (this and a l l subsequent Q-III tests involved 5 to 10 bottles at 22°C, sub-cultured to 29°C), and the progeny grown at 22° or 29°C. The adults were scored for v i a b i l i t y and the occurrence of visible phenotypes, particularly the presence of sex combs on the second pair of legs of males. Prelim-inary tests revealed that none of the recessive markers present on the 100 test chromosomes (singly or in combination) interacts with Q-III. Henceforth the stocks tested w i l l simply be referred to as Msc, Sex and Pc. A lethal interaction has been reported for the combination of Lyra and M ( 3 ) h — ( L i n d s l e y and Grell, 1968). To test for a similar inter-action between Q-III and Ly_, Q-III/CxD females were crossed to Ly/CxD males at 22° and 29°C and the offspring were scored for survival. Tests of other Minutes with homeotics In order to assess interactions of other Minutes with the sex comb homeotics, Df(3L)M(3)LS4/TM3 males were crossed to females carrying Sex, Msc or Pc. Five bottles of each cross were established and after 3 days, subcultured for another 3 days. Similarly, M(2)173/SM5 males were crossed to Sex and Msc in two bottles which were then each sub-cultured once after 3 days. The adult male progeny of a l l crosses were scored for the presence of sex combs on the second pair of legs. Also, v i a b i l i t y effects and the occurrence of any other phenotypes were noted in the adults. 101 III. Results Only the numbers of eclosing adults were recorded in the studies of v i a b i l i t y . In other tests, cultures were examined at least every 24 to 48 hours and the numbers of animals reaching key stages ( f i r s t instar larva, pupa and adult) were noted. The amount of lethality for each stage was estimated on the basis of the expected frequency of each class, relative to the number of eggs for a given cross. Simi-larly, percent v i a b i l i t i e s were calculated using the expected frequencies from the following crosses: Controls, I a l l progeny £ P/£ P; II 0.5 CxD/pP, 0.5 TM3/pP. Experimentals (A) 0.5 Q-III/p P 0. 5 TM3/pP; (B) 0.5 Q-III/TM1, 0.25 Q-III/Q-III, 0.25 TM1/TM1 (egg lethal); (C) 0.5 Q-III/TM3, 0.25 Q-III/Q-III, 0.25 TM3/TM3 (egg lethal). Percent v i a b i l i t i e s were calculated as: (Observed number of live progeny/Expected number of live progeny) x 100, where Expected number = Expected proportion of eggs x Total number of eggs. 1. Genetic Analysis Viability The crosses to assess Q-III v i a b i l i t y (Table 21A,B and C) show that relative to the controls, Q-III is essentially f u l l y viable at 22°C when heterozygous with p_P, TM3 (also at 17°C) or TM1, while at 29°C the percent v i a b i l i t i e s for these classes were 49, 13 and 0 respectively (where applicable, homozygotes for TMl, TM3 and CxD have been considered as egg lethals). The exact reason for the low via-b i l i t y of Q-III in combination with the two balancers is not known. Two of the inversions present in TM3 have the following breakpoints 102 Table 21 Relative V i a b i l i t i e s of Q-III Homozygotes and Heterozygotes at Different Temperatures Crosses Temp- Number Progeny Number Percent erature of eggs Genotypes of Adults Vi a b i l i t y Control I 22°C 195 Z / Z 151 77.4 Z/Z x Z / Z T\ T"\ 29°C 200 Z / Z 149 74.5 Control I I 22°C 105 CxD/_p_£ 38 72.4 TM3/pP 42 80.0 CxD/TM3 x Z/Z n CxD/j^ 54 74.5 29°C 145 TM3/pP 55 75.9 (A) Q Q - m / Z 309 83.6 22 C 739 TM3/pP 291 81.5 Q-III/TM3 x p P/p P n Q - I H / Z 192 49.0 29°C 783 TM3/pP 329 84.0 (B) Q-III/TM1 256 76.3 22°C 642 Q - I I I / Q - I I I 44 27.4 Q-III/TM1 x Q-III/TM1 Q-III/TM1 0 0 29°C 900 Q - I I I / Q - I I I 0 0 103 Table 21 (continued) Crosses Temp- Number Progeny Number Percent erature of eggs Genotypes of Adults V i a b i l i t y (C) 17°C 800 Q-III/TM3 x Q-III/TM3 22°C 825 29°C 1034 Q-III/TM3 335 88.8 Q-III/Q-III 62* 31.0 Q-III/TM3 357 86.6 0-111/Q-III 51 24.8 Q-III/TM3 69 13.4 Q-III/Q-III 0 0 *22/62 showed scutellar disruptions 104 (see Lindsley and Grell, 1968): In(3LR)79E;100C and In(3LR)76C;93A. It is noteworthy that M(3)LS4 resides in the segment 79D to 80 or 81, while M(3)S34 is located within the segment 75D to 76C. Thus, both proximal Minute l o c i could be under the influence of some sort of posi-tion effect which potentiates the lethal effects of Q-III. Alterna-tively, the dominant markers Sb and Ser carried by TM3 could be inter-acting with Q-III. However, the latter idea is less likely, since tests of TM3 without these dominants resulted in similar low frequen-cies of Q-III/TM3 progeny at 29°C. TMl has no comparable inversion breakpoints. However, i t is possible that Moire, a recessive lethal marker carried by this balancer, interacts with Q-III to produce . synthetic lethality of TMl/Q-III progeny at 29°C. This possibility was not pursued. In a l l crosses the heterozygotes which survived continuous expo-sure to 29°C, displayed small, thin bristles (see Figure 10), a roughened eye surface (sometimes reduced or malformed eyes), slightly pale body colour and occasionally, leg deformities. In contrast to complete v i a b i l i t y of Q-III heterozygotes, only 25 to 31 percent of the expected number of Q-III homozygotes (crosses B and C) survived to adulthood at 22° and 17°C, while none survived con-tinuous exposure to 29°C throughout development. The homozygous adults surviving at 22° and 17°C had bristles with a thickness that appeared to be intermediate between those of Q-III/+ and +/+ at 29°C. In addi-tion, they exhibited slightly roughened eyes, pale body colour and frequently, internal melanotic masses (particularly within the abdomen). At 17°C, one third or more of the homozygotes also displayed a disrupted 105 thorax phenotype. In the least severe cases this trait consists of extra or misplaced bristle sockets on the scutellum and in the most severe cases, grossly distorted or malformed scutella. The prolonged -development characteristic of Minute heterozygotes was also seen for Q-III/+ f l i e s at 29°C and Q-III/Q-III f l i e s at 17° and 22°C. Clearly the mutation Q-III is very pleiotropic with a complex of different phenotypic effects. Mapping The results of the mapping experiments are presented in Table 22. They show that Q-III is located between Gl and _p_p (Table 22a), indeed i t is between s_t and j>P (b) . Owing to the v a r i a b i l i t y in survival of Q-III-bearing f l i e s , unequivocal localization was d i f f i c u l t but a tentative position based on the dominant semi-lethal effect was com-puted. Thus, Q-III maps to 45.4 relative to the recessive markers. The location of Q-III cannot be unambiguously assigned. Its most like l y position is between _st and in, but the in to jp_P interval cannot be completely ruled out. The ambiguity arises from the recovery of st Q-IIl"1" p P (i.e. st i n + r i + p P), st in r i Q-III p P (i.e. st in r i p P —M) and st in Q-III p P (i.e. st in r i + pP—M) recombinants at 29°C. If Q-III lies between s_t and in, the latter two classes could be generated only by triple crossovers, and triple crossovers are again required to explain the occurrence of st Q-III + p P recombinants, i f Q-III is between in and _p_P. In either case, the observation of putative multiple ex-change classes extends the report that multiple crossovers occur within short genetic intervals in proximal regions of chromosome 3 (Sinclair, Table 22 Crossover Data From Crosses Designed to Localize Q-III a) Gl Sb H/Q-III p P Females x Males Type Genotype/p? • Number of Progeny in Each Class 22°C 29°C Gl Sb H 861 562 Parentals 748 30 SCO 1 Gl p P Sb + H + 35 2 (IM)* SCO 2 G l + p** Sb H 51 25 (IM) SCO 3 Gl p^ Sb + H + 138 60 SCO 4 G l + p P Sb H 123 4 (M) SCO 5 Gl p** Sb H + 92 85 SCO 6 G l + p P Sb + H 138 1 DCO 1 Gl p 1^ Sb + H 2 1 DCO 2 G l + p P Sb H + 2 1 (M) DCO 3 G l + P P + r Sb..H+ 5 2 DCO 4 Gl p P Sb + H 2 0 DCO 5 Gl p P Sb H 9 0 DCO 6 G l + p ^ Sb + H + 11 3 TCO 1 G l + p 1^ Sb + H 1 0 Totals 2218 776 107 b) st in r i eg /Q-III p Females x st in r i p /st in r i p Males Type Genotype/st in r i p Number of Progeny in Each Class Parentals SCO 1 SCO 2 SCO 3 SCO 4 SCO 5 st in r i . + .+ p st in r i p -+ • • P+ st in r i p .+ p st in r i p st in r i p + . + .+ p+ st in r i p 22°C 1315 1345 10 (M) 6 0 1 4 (M) 29°C 1405 402 8 (5M) 23 1 (M, sterile) 9 (IM) 4 (M) Totals 2681 1852 Map position of Q-III: 45.4 map units M = Minute phenotype (delayed eclosion, thin, small bristles) 1 108 1975). If Q-III is located between .st and in, i t is surprising that no Q-III in r i crossovers were recovered at 22°C. Progeny tests of the recombinants recovered at 22°C (included in Table 22b) support the suggestion that Q-III lies closely linked to in between _st and i_n. It is important to mention that a Minute locus exists between s_t and i_n, while two are located between r i and j>P, one on either side of the centromere. In addition to the i n i t i a l progeny tests, several recombinant chromosomes were cloned at 29° and 22°C. Almost a l l of the pleiotropic phenotypes (to be referred to later) attributable to Q-III, segregated with the mutation. Two Q-III p P Sb H recombinants that had been generated at 22 C were used to make stocks of Q-III for use in the shift studies. Thus, the marker H was removed from the chromosome via crossing over and with i t a second site cold-sensitive mutation that rendered Q-III lethal at 17°C. Test of Q-III in triploids In a preliminary test of Q-III in triploids, Q-III/TM3 males were 2 a crossed to C(1)RM, y sc w ec/FM6;3A females in shell vials and C(1)RM, y 2 sc wa ec/X;Q-IH/+/+ females isolated at 29°C. Five such females were scored and none displayed the M phenotypes, thereby indi-cating that Q-III is recessive in triploids. Complementation tests The complementation data have been converted to a v i a b i l i t y index parameter. Thus, the v i a b i l i t y index, V.I. = Number (Q-IIl/mutant)/ 109 Table 23 Relative Viability of Q-III in Combination With Various Mutations at 22° or 29°C Mutant V.I.* 22°C Total Progeny V.I. 29°C Total Progeny Df(3L)M(3)LS4 1.95 428 0 231 Dfd 1.25 234 0 200 Ns 1.18 804 0.09 246 Antp0- 1.44 814 0.19 38 . TI _ ,„. , . - . _ -r , s Number of Q-Ill/mutant * V.I. (Viability Index) = - — ,. 2... , ^—-— J Number of TM3/mutant 110 Number (mutant/TM3). Remember that at 29°C, a l l Q-III heterozygotes are half as viable as wild-type and therefore some decrease in the vi a b i l i t y of heterozygotes is expected. However, the V.I. estimates should not be lower than 0.5, i f Q-III is viable with these mutations. The results of the complementation crosses of Q-III are shown in Table 23. In each case, Q-III was viable in combination with the mutations tested at 22°C, while i t was lethal when combined with either Dfd or Df(3L)M(3)LS4 at 29°C. Furthermore, Q-III was semi-Q lethal in combination with Ns (V.I. = 0.09), and Antp- (V.I. = 0.19) at 29°C (although in the latter case, very few progeny of any class resulted). Any uneclosed Q-III/Dfd pupae (at 29°C) which were dissected showed a complete lack of head (i.e. eye-antennal disc) structures, and sometimes uneverted pigmented ommatidia within their thoraces. The phenotype of these f l i e s resembled Arking's (1975) i l l u s t r a t i o n of 1 (1)ts480, a sex-linked ts autonomous c e l l - l e t h a l . No Df(3L)M(3)LS4/Q-III pupae were found at 29°C, thereby showing that an earlier lethal phase exists for these heterozygotes. Since the most lik e l y map position of Q-III is distal to M(3)LS4, their lethality in combination may reflect the propensity of Q-III to interact with different l o c i such as Dfd, rather than allelism. Fur-thermore, the relatively normal v i a b i l i t y of Df(3L)M(3)LS4/Q-III fl i e s at 22°C argues against allelism. However, i t should be noted that ts6 7g 1 (l)su(f) & (Dudick est al_., 1974) is viable when heterozygous with a deficiency for the su(f) locus at 18° and 25°C, but no heterozygotes of this type survive at 29°C. Table 24 The Lengths of the Developmental Periods From Egg Deposition to Eclosion i n Different Classes at Different Temperatures Data From Different Temperatures 17°C 22°C 29°C Genotypes Genotypes of Parents of Progeny Total Eggs Number T 1/2* Total Number T 1/2 Total Eclosing (hrs.) Eggs Eclosing (hrs.) Eggs Controls 195 151 300±6 200 Number Eclosing 149 T 1/2 (hrs.) 192±6 II. CxD/TM3 CxD/pP TM3/pP 105 38 42 312±6 312±6 145 54 55 192±6 192±6 Experimentals A. Q-III/TM3 Q-III/p* TM3/pP 200 75 87 312±6 312±6 276 72 125 240+6 192+6 Q-III/TM3 C. Q-III/TM3 x Q-III/TM3 Q-III/Q-III 800 335 62 600±6 672±6 500 180 53 324±6 372±6 500 10 300±6 *T 1/2 (hours) = Time in hours from oviposition to eclosion of half of live progeny. 112 These data underscore the problems inherent in deciding whether particular mutants are a l l e l i c , since the possibility of synthetic lethality should always be considered. 2. Developmental Analysis Duration of developmental periods Table 24 summarizes the lengths of developmental periods of con-trols and Q-III heterozygotes and homozygotes. This interval was de-fined as the time (in hours) from oviposition (+ 2 hours) to the time when half the progeny of a given class had eclosed (since cultures were observed every 12 hours, each time period had an error of about 6 hours). Cross A provides an additional control since TM3/pP progeny are generated along with Q-III/p P individuals. The results of cross C show that the Q-III homozygotes took con-siderably longer to eclose than did the controls, while the Q-IH/TM3 heterozygotes developed at the same rate as the controls. At 29°C only a few heterozygotes survived and the latter took a long time to eclose relative to the controls. Development of Q-III/p P progeny in cross A is normal at 22°C but greatly prolonged at 29°C. Thus, in addition to the conditional bristle and eye phenotypes, Q-III/+ heterozygotes also develop more slowly at 29°C than at 22°C. At 22°C the homozygous individuals take longer to eclose than do the controls. Tests for s t e r i l i t y and maternal effects Of 15 female homozygotes brooded at 22°C, 5 produced no progeny or eggs while the remainder produced a total of 139 progeny (an average 113 of 14 of f s p r i n g per f e r t i l e female). Upon s h i f t i n g to 28°C, two of the above 10 f e r t i l e females died during the f i r s t brood, while the other 8 produced no progeny. Examination of the 28°C v i a l s revealed a few white (and some brown) eggs. A f t e r the second 28°C brood, no eggs were detected. Seven females that had been mated to p P / p P males fo r f i v e days at 28°C were dissected i n Drosophila Ringer^s s o l u t i o n and t h e i r ovaries examined with a compound microscope. Sperm were abundant i n the seminal receptacles . The ovaries contained degenerate oocytes which appeared to be heterogeneous for early stages of oogenesis and i n few cases, polytenic n u c l e i were detected.. The l a t t e r were presumably undegenerated nurse c e l l s ( M i l l e r , 1950). The study of f e r t i l i t y of Q-III/Q-III males was abandoned when only 2 of 15 showed f e r t i l i t y at 22°C, and that f e r t i l i t y was very poor. When such homozygotes were mated to j 5 P / p _ P females at 28°C, eggs were l a i d but f a i l e d to develop, thereby suggesting that both homozygous males and females are completely s t e r i l e at 28°C. When homozygous males and females are crossed at 22°C, white eggs are deposited but no development occurs. Of the 102 eggs which were produced by homozygous Q-III females (mated to j^/pj' males) and transferred to 29°C, 70 remained white while 30 turned dark a f t e r one or two days but development ceased. In only 2 cases did any l a r v a l development ensue and i n both cases, the larvae turned black and died s h o r t l y a f t e r hatching. Of the 220 eggs kept at 22°C, 107 exhibited varying degrees of darkening, 42 re-mained white and 71 hatched as larvae. Of the 71 larvae, 60 eclosed as phenotypically normal adults. I t i s worthy of mention that eggs 114 which remain white could be either unfertilized eggs or embryos in which development was blocked at very early stages. On the other hand, eggs which turn dark after a few days are assumed to be embryonic lethals (Wright, 1973) . It therefore appears that Q-III can exert a ts maternal effect, since eggs produced by homozygous Q-III females mated to normal males are essentially incapable of supporting normal development at 29°C, while at least a significant proportion of such eggs incubated at 22°C, develop normally. Stage distribution of lethality of Q-III homozygotes and heterozygotes Even as early larvae the Q-III/Q-III progeny could be distinguished from the Q-III/TM1 or Q-III/TM3 heterozygotes due to their slower devel-opment and internal melanization at 17° and 22°C. At 29°C, since the homozygotes never progressed past the f i r s t larval instar, this dis-tinction was also possible. Most Q-III/pP progeny developed considerably more slowly than p_P/ TM3 types at 29°C (but not at 22°C) and therefore, Q-IH/p P larvae could be unequivocally classified on this basis (as well as on that of internal melanization). However, some overlap of classes did exist. For this reason, stage-specific lethality was determined for a l l pro-geny of the cross of Q-III/TM3 females to j_P/p_P males, without attempt-ing to separate the classes, on the assumption that most ts lethality would be due to the death of Q-III/p P individuals which would be re-flected in the results. Table 25 illustrates the results of the analysis of stage by stage distribution of lethality during the embryo, larval and pupal stages Table 25 Lethality of Control and Q-III - Bearing Progeny at Different Developmental Stages at Various Temperatures Percent Mortality 17°C 22°C 29°C Genotype Number Number Number of of of of Progeny Eggs E L P Eggs _E L P Eggs E L P Controls P/ P - 195 14.4 5.6 2.6 200 8.5 16.5 0.5 CxD/pP CxD/TM3 - 105 12.4 4.8 6.7 145 8.3 10.3 6.2 Experimentals Q-III/TM1 - 300 0.5 20.5 5.5 450 21.6 71.9 6.5 j£/TM3j Q-III/p P - 268 8.7 9.4 5.0 250 15.0 23.1 12.1 Q-III/Q-III 100 3.0 26.0 40.0 200 2.0 44.0 22.0 225 52 48(L1)* 0 *L1 = First larval instar E = Embryo L = Larva P = Pupa 116 of the controls and of the various progeny arising from the crosses: Q-III/TM3 x p_P/£P (22° and 29°C); Q-III/TMl x Q-III/TMl (22° and 29°C); and Q-III/TM3 x Q-III/TM3 at 17°C (see Table 21). A l l Q-III/TMl heterozygotes died when exposed continuously to 29°C. More than 70 percent died as larvae. Larval death was also most common for Q-III/p P progeny at 29°C. At 22°C, again a high proportion of Q-III/TMl types died as larvae, while mortality was more evenly d i s t r i -buted between the three stages for Q-III/p P progeny. In comparison, Q-III homozygotes died equally as frequently as embryos or f i r s t in-stars at 29°, while at 22°C the lethality shifted more towards the later larval and pupal stages with very l i t t l e egg lethality. Further-more, at 17°C lethality of Q-III homozygotes was chiefly pupal, although frequent larval lethality s t i l l occurred. ,Thus, at progressively lower temperatures, Q-III homozygotes survived to later stages. Most of the Q-III/Q-III pupae at 17° and 22°C contained pale imagoes with bent legs. This may indicate that lethality results from defective sclerotization of the cuticle which leads to desiccation and lack of the muscular control necessary for eclosion. At 29°C, homozygous Q-III/Q-III larvae died as f i r s t instars. A l -though this death was often not immediate (they sometimes survived for up to 4 or 5 days), no growth or molting took place. They eventually showed internal discolouration and died. Considerable variation in patterns of lethality in different stages was observed in the controls at a l l temperatures and generally, larval death increased upon exposure to higher temperatures. 117 Temperature-sensitive periods for lethality of Q-III Percent v i a b i l i t y was computed for a l l cultures on the following basis. Since only 25 to 30 percent of Q-III/Q-III progeny survived at 22°, the 22°C level of v i a b i l i t y (see cross B, Table 21) was normalized to 100 percent. Thus, for example, the expected number of Q-III/Q-III progeny at 22°C would be: 0.274 x (total number of eggs x 0.25). This kind of analysis was used whenever calculation of such percent v i a b i l i t i e s was necessary. The TSP of Q-III recessive lethality can be inferred in Figure 6. This TSP extends continuously from the latter part of embryogenesis up to the second half of pupation, and therefore the Q-III + gene product is clearly required throughout most of development. Since long exposures to the restrictive temperatures can mask successive, but separate TSPs (Poodry ejt ajL., 1973; Suzuki, Kaufman, Falk e_t ajL., 1976), 24 and 48-hour pulse shifts were performed. Although a significant number of Q-III/TM1 heterozygotes usually survived most 48-hour heat pulses, three discrete temperature-sensitive intervals were resolved (Figure 7). One spans the latter half of the f i r s t instar and the f i r s t half of the second instar, another covers most of the third instar, and the third TSP of lethality spans the middle of pupation. On the other hand, the Q-III homozygotes appear to have a single, continuous TSP. These results re-emphasize the importance of the Q-III + gene product. A lack of this substance for even relatively short periods of development produces death. Forty-eight-hour heat pulses of homozygotes from the f i r s t to mid second larval instar, often resulted in sluggish larvae, many of which did FIGURE 6 Results of the shift study to delineate a temperature-sensitive period (TSP) for lethality of Q-III homozygotes. The data are given as percent eclosion of Q - I I I / Q T T I I progeny that were shifted from 22 to 29 C (closed circles), or from 29° to 22°C (open circles) at various times during development. Temporal estimates of the different developmental stages at 22°C, are indicated below; E = eclosion. Time of Shift (in Hours) Developmental Stages at 22°C Egg 1 .1st I 2nd 1 57d I Pu~^a~ 1 a y instar instar instar H u p a E FIGURE 7 Result of 48-hour pulse shifts to delineate temperature-sensitive periods (TSPs) for lethality of Q-III heterozygotes and homozy-gotes. The data are given as percent eclosion of Q-III/TMl (open triangles) or Q-III/Q-III (closed triangles) progeny, that were heat-pulsed for 48 hours, at various times during development (pulses were shorter during embryogenesis). The horizontal bars indicate the duration of the heat pulses to 29°C. Temporal estimates of the different developmental stages of the heterozy-gotes (het.) and homozygotes (hom.) at 22°C, are indicated above and below, respectively; E = eclosion. Developmental Stages at 22°C het.  I Egg I 1st 1 2nd I 3rd I P u p a instar instar instar 125} Time of Shift (in Hours) Developmental Stages at 22°C horn.  I ggg I 1st I 2nd I 3rd I P u p a 1 instar instar instar 122 not grow. These larvae exhibited different degrees of internal dis-colouration. They sometimes survived for several days after the shift, but most eventually died. Later shifts (i.e. early to late third in-star) frequently produced sluggish larvae that never metamorphosed but some also formed incomplete pupae, while s t i l l others reached various stages of pupation. The later shifts (late third instar and through-out the f i r s t half of pupation) produced pupae which reached various stages including the pharate stage, but never eclosed. Q-III/TMl heterozygotes survived 24-hour heat pulses at different stages. However, different lethal TSPs of Q-III homozygotes could be distinguished (Figure 8). The pattern of TSPs is strikingly similar to that resulting from 48-hour heat pulses to Q-III/TMl heterozygotes (see Figure 7), although the actual position of each TSP is different. Again, lethality from earlier pulses was primarily larval. For example, a heat pulse from 84 to 108 hours of development (early second instar) produced many dead larvae displaying considerable internal discolouration (probably diagnostic of generalized disruptions). Some pupae were formed but in most cases successful metamorphosis did not occur, although a few managed to complete pupation. Later larval pulses produced progressively more advanced development at the time of lethality. In some cases (for example a pulse from 108 to 132 hours), after surviving homozygous adults had eclosed and were scored, other third instar larvae began to appear in the cultures and these eventually pupated (after several days), but never eclosed. This failure to eclose may have been due to poor leg differentiation, since the legs of dissected pharates were often bent in appearance. FIGURE 8 Results of 24-hour pulse shifts to delineate temperature-sensitive periods (TSPs) for lethality of Q-III homozygotes. The data are given as percent eclosion of Q-III/Q-III progeny that were heat-pulsed for 24 hours, at various times during development. The hori-zontal bars indicate the duration of the heat pulses to 29°C. Temporal estimates of the different developmental stages at 22°C are indicated below; E = eclosion. 125 The occurrence of this second wave of pupation raises the interest-ing possibility that larval growth can be reversibly blocked in homozy-gous Q-III individuals by a brief 29°C exposure during development. This could be further studied by exposing large numbers of synchron-ously developing larvae to 29°C for even shorter intervals, or alter-natively, by employing very short intervals coupled with higher tempera-tures (e.g. 30° or 31°C). Finally, much of the lethality induced by 24-hour heat pulses during pupation occurred at the pharate stage, with some of the un-eclosed f l i e s possessing poorly differentiated legs. A l l of these were light in colour and many seemed to have been rapidly desiccated (as indicated by collapsed abdomens). Since only about 30 percent of Q-III homozygotes usually survive at 22°C, conclusions about the patterns of lethality should be viewed with caution. Nevertheless, i t does appear that 24-hour heat pulses have resolved three discrete TSPs for in v i a b i l i t y : one early in the second instar, one in the late third instar and the early part of pupation, and another just after mid-pupation. Phenotypic descriptions of. pulse survivors w i l l be dealt with later. Temperature-sensitive periods for dominant rough eye and bristle traits Since heterozygous Q-III (Q-III/+, Q-III/p P, Q-III/TM3) fli e s which survive continuous exposure to 29°C have rough eyes and short, thin bristles, TSPs for these phenotypes could be defined (Figure 9). The TSP for the eye phene extends from about a third of the way into the second instar up to the end of this instar. This TSP was also FIGURE 9 Results of the shift study to delineate temperature-sensitive periods (TSPs) for rough eyes and reduced bristles of Q-III heterozy-gotes. The eye data are given as percent ex-pression of rough eyes in Q-III/p P progeny that were shifted from 22° to 29° (closed ci r c l e s ) , or from 29° to 22°C (open circles), at various times during development. The TSP for the bristle phene was also provided and is indicated by the horizontal bar. Temporal estimates of the different developmental stages at 22° and 29°C, are indicated above and below, respectively; E = eclosion. Developmental Stages at 29°C i 7Z r ' E g g ' 1st ' 2nd instar instar 3rd instar Pupa 100 75 A >> UJ |> 50 0 OC c 9) o $ 2 5 0. — c • • • • o \ • \ c p - O - O - O - O - O - O - O - O - O - O " o-o-o oV I 1 ' £ 1 ( 1 I 1 f \ Bristle \ TSP jo -•• O O - O - O - Q 1 — « f > t n f » • o • y 60 120 180 Time of Shift (in Hours) 240 300 Developmental Stages at 22°C 1 E £99 iTt r instar 2ndT instar 3rd instar Pupa 128 observed after 24 and 48-hour pulses (see APPENDICES 2 and 3) and i t overlaps the earlier TSP for i n v i a b i l i t y that was also resolved by pulse shifts. It is noteworthy that this TSP also coincides with the developmental interval during which intense mitotic activity has been reported for the eye imaginal discs (see Nothiger, 1972). The probable basis for this phenotype w i l l be dealt with later. The TSP for the bristle phene was i n i t i a l l y delineated to a 34-hour period during the f i r s t half of pupation. The 24-hour pulse studies of heterozygous Q-III/TMl progeny allowed further resolution of this TSP to a 16-hour interval prior to mid pupation (before the time when yellow pigment is deposited in the eyes). Moreover, i t was possible to verify this TSP in homozygotes surviving 24-hour heat pulses (although the TSP of the bristle trait in the homozygote differs somewhat from that in the heterozygote, see Figure 20). Scanning electron micrographs of the bristle phenotype can be seen in Figure 10. Note the reduction in or absence of thoracic macrochaetae in the homo-zygote (c), whereas the phenotype is less severe in heterozygotes (b). The abdominal bristles were also slightly reduced in size in the homo-zygote. It is noteworthy that the bristle TSP corresponds to the pupal interval during which the thoracic and abdominal bristles are formed (Bodenstein, 1950). Phenotypes revealed by shift experiments Since nearly a l l heterozygous phenotypes observed in the shift experiments (both pulse and regular TSP shift studies) were also seen (and were usually more extreme) in Q-III homozygotes, only the pheno-types of the latter w i l l be dealt with in detail. However, reference FIGURE 10 Scanning electron micrographs showing the effects of Q-III on macrochaete development; (a) a control (p P/p P) f l y , grown at 22°C (magnification about x400); / 3 0 FIGURE 10 (b) a Q-III/p P heterozygote, grown continuously (continued) at 29 C; (c) a Q-III homozygote, heat-pulsed at 252-276 hours post oviposition (magnification about x400) . 133 w i l l be made to similar phenotypes of heterozygotes, and any unique traits w i l l be referred to specifically. APPENDIX 2 summarizes the types and frequencies of the phenotypes observed in surviving homozygotes after 24-hour heat pulses. These wi l l be dealt with on the basis of the individual imaginal discs from which the affected structures are derived. (a) pattern defects of the eye-antennal disc In addition to the roughened eye phenotype, heat-pulsed heterozy-gous and homozygous (and in the case of heterozygotes from regular shift experiments, those surviving shifts to 29° or 22°C) Q-III fl i e s displayed moderate to severe loss of ommatidial tissue. Figure 11a, b, c and d are scanning electron micrographs of such individuals. Note that there is considerable ventral displacement of eye tissue in the heterozygote (a). High power magnification (b) reveals that although inter-ommatidial bristles are absent in the displaced portion, they are duplicated for many of the other ommatidia. These duplications could be responsible for the rough eye phenotype mentioned earlier (see Poodry et al_., 1973). Note also the presence of extra vibrissae which extend across the eye at three o'clock. In the homozygote, the l e f t eye has been severely reduced (c), while no visible reduction of the right eye has taken place. Thus, although this reduction frequently can be extreme in'the homozygote, i t does not necessarily always occur for both eyes. Note also that this fly completely lacks the third segment (including the arista) of the l e f t antenna. The highest frequency (83 percent, APPENDIX 2) of ommatidial deficiencies of homozygotes was induced by a heat pulse at FIGURE 11 Scanning electron micrographs showing the effects of Q-III on eye development; (a) a Q-III/p P heterozygote, shifted to 29°C at 24 hours post oviposition (magnification about x400); (b) (the same f l y ) , one arrow (i) points out a duplicated interommatidial bristle, while another ( i i ) points out the enlarged vibrissae (magnification about x800); Z35-FIGURE 11 (c) a Q-III homozygote, heat pulsed at 72-96 (continued) hours post oviposition (magnification about x400); (d) (the same homozygote), an arrow points out the second antennal segment on the fly's l e f t (magnification about x800). 138 96 to 120 hours (this phenotype was also observed in pulses on either side of this interval). Therefore, the TSP of this phene spans most of the second and the early part of the third instar:. . Less often, deficiencies and duplications of antennae as well as palps and o c e l l i occurred within the limits of this TSP. In two cases where the palps were absent, the ipsilateral antennae were also missing. Thick, fleshy aristae were also often induced by heat pulses during this TSP. Since a l l of the above structures are derived from the eye-antennal disc, the correspondence in TSPs for the different defects is not surprising, but the spectrum of phenotypes is noteworthy. Dissection of several pharate pupae resulting from heat pulses at 84 to 120 hours revealed that in some cases almost no eye-antennal structures had formed in such pupal lethals. In a few cases, severe deficiencies of the head region were accompanied by complete antennal duplication. A pulse experiment was initiated to further investigate this. Two hundred larvae from a Q-III/TM3 x Q-III/TM3 cross were syn-chronized in the early part of the second instar and heat-pulsed from about 100 to 140 hours (mid second to early third instars) post ovi-position. Only four homozygotes eclosed. Of these, one displayed extreme eye deficiencies on both sides, while the other three exhibited unilateral antennal duplications along with less severe eye deficiencies. Figure 12 is a scanning electron micrograph of one of the latter. The following were observed in this individual: (a) mirror image duplica-tion of the l e f t antenna along with tri p l i c a t i o n of the arista (com-pare with the normal right antenna and arista and note also the reduc-tion of the ventral surface of the l e f t eye) and (b) a jointed FIGURE 12 Scanning electron micrographs showing the eye-antennal pattern defects of a Q-III homozygote, heat pulsed at 100-140 hours post oviposition; (a) arrows point out the duplicated l e f t antenna with triplicated arista; note the ventral reduction of the eye; (b) (the same f l y ) , arrows point out the unidentified jointed structure (magnification about x400) . 141 structure of unknown origin extending from the proboscis and positioned adjacent to the l e f t palp. Undefined structures have also been fre-quently observed at the periphery of reduced eyes in other experiments involving Q-III as well as with Dfd (D. Sinclair, personal observations). Of twenty-one dissections of dead pupae present in the cultures of this heat-pulse experiment, 17 possessed few or no head structures (in many cases, only the proboscis was found) and sometimes in these individuals, eye pigment globules could be seen within the thorax. This phenotype was also observed for Dfd/Q-III heterozygotes grown con-tinuously at 29°C. The other four dead pupae had severely reduced eyes and, in one case, bilateral duplication of the second and third antennal segments (along with the aristae) occurred. Thus, deficiencies of eye-antennal structures (particularly reduc-tion of the eyes) occur frequently in Q-III/Q-III f l i e s heat-pulsed for brief intervals during the second instar. In i t s most extreme form, this phene resembles the phenotype of a sex-blinked autonomous c e l l -lethal, 1(l)ts480, which was illustrated and described by Arking (1975). In heterozygotes, ommatidial deficiencies resembling (but less severe than) those of Q-III homozygotes were found i n survivors of 48-hour pulse shifts (see APPENDIX 3) and survivors of shifts up or down. Again the TSP for this phene appears to occur within the second larval instar. (b) pattern defects of the dorsal mesothoracic (wing) disc At least thirty percent of homozygotes that were continuously ex-posed to 17°C, possessed pattern defects of the thorax, particularly of the scutellum. As previously mentioned, expressivity of this trait FIGURE 13 Scanning electron micrographs showing the effects of Q-III on the development of the scutellum at 17°C (the effects are similar to those produced by heat-pulsing); (a) a p P/p P f l y grown at 22°C (magnification about x400); )4S FIGURE 13 (b) a Q-III homozygote (magnification about (continued) x400); (c) (the same f l y ) , at a higher power with an arrow pointing out the hairs in the region of the scar (magnification about x800); 145 FIGURE 13 (d) a Q-III homozygote (magnification about (continued) x400); (e) (the same fly as in d), at a higher power (magnification about x800); FIGURE 13 (f) a Q-III homozygote with an arrow pointing (continued) out a duplicated bristle (magnification about x800). A l l Q-III homozygotes were grown continu-ously at 17°C, while (a) was grown at 22°C. 150 ranged from the appearance of extra bristle sockets, randomly placed on the scutellum, to duplicated bristles and even bifurcation of the scutum and prescutum (Figure 13 a,b,c,d,e, and f) . Note the normal scutellum and bristle arrangement on the control f l y (a). A Q-III/ Q-III f l y reared at 17°C (b and c) had marked indentation of the scu-tellum. Higher magnification (c) shows that disarrangement of the small scutellar hairs has occurred in the v i c i n i t y of the scar. The fl y shown in d and e exhibited an analogous disruption, but i t is un-certain whether this represents a duplication of the posterior or anterior scutellar bristles. Finally, the f l y shown in (f) possessed a similar scutellar indentation, as well as what appears to be a non mirror-image duplication of a posterior scutellar b r i s t l e . By heat pulsing Q-III/Q-III larvae, similar phenotypes could be induced. When the pulse was applied from 108 to 132 hours after ovi-position, a l l of the progeny displayed the scutellar phenotype, with a minority having only extra bristle sockets at unusual positions on the' scutellum (APPENDIX 2). Very rarely, large deficiencies of the scutellum were seen. Their rarity may reflect that any more severe pattern defect in this disc produces death. The TSP for this anomaly thus encompasses the eye TSP in early to late second instar, although the peak for the former was attained just after the middle of this instar. Heterozygotes (Q-III/TMl, Q-III/TM3 and Q-IH/p P) also displayed the thoracic pattern phenotype at low frequencies, either when heat-pulsed (Q-III/TMl) or when shifted up during the TSP (Q-III/TM3, Q-III/ _£ P). Occasionally, Q-III/TM3 heterozygotes surviving shifts down possessed wing-like or haltere duplications (Figure 14). FIGURE 14 A scanning electron micrograph showing a wing-like duplication (arrow) in a Q-III/TM3 f l y that was shifted from 29° to 22°C at 156 hours post oviposition (magnification about x800). 153 It should be mentioned that the above Q-III-bearing heterozygotes which survived shifts up to 29°C also displayed disruption of wing venation, particularly for vein L2. However, this phenotype was not observed in homozygotes pulsed to 29°C and no attempt was made to define its TSP. Cell death within imaginal discs followed by different degrees of pattern reconstruction, could provide an explanation for most of the above phenotypes. This idea w i l l be more specifically dealt with in the Discussion. ' (c) thoracic macrochaetae To reiterate, thoracic (particularly scutellar) macrochaetae were severely reduced in homozygotes by heat pulses administered during the f i r s t half of pupation. (d) defects involving the leg discs Twenty-four hour heat pulses of Q-III homozygotes during late second or early third instars induced a low frequency of either mis-sing legs (particularly mesothoracic) or legs with shortened tarsi (APPENDIX 2). This was also observed in heterozygotes that had devel-oped continuously at 29°C or in the shift cultures which had been used to define the eye, bristle and lethality TSPs. Male homozygotes heat-pulsed from 228 to 276 hours (during pupa-tion) exhibited gaps in their sex combs (Figure 15). Note that the sex combs appear to be incompletely rotated, in that neither the upper nor the lower halves of the comb are aligned with the axis of the leg. Although the TSP for this phenotype corresponds roughly to the time of leg disc eversion and elongation (Bodenstein, 1950), the relationship between these phenomena and sex comb differentiation is unknown. FIGURE 15 Scanning electron micrographs showing the effects of Q-III on sex comb development; (a) the basitarsus of a normal male (p P/p P), grown at 22°C; (b) the basitarsus of a Q-III homozygote, heat-pulsed at 228 to 252 hours post oviposition (magnification about x2000). 156 A fused foreleg phenotype was observed only in specific shifts down (or heat pulses at specific times during larval development) in-volving Q-III/TM3 heterozygotes. This phene was marked by a progres-sive but variable fusion of the forelegs as seen in Figure 16. At most, only about one third of a l l heterozygotes in a culture exhibited this phenotype. Although such fusion was seen in shifts down and heat pulses within the interval 84 to 144 hours, i t was never observed in flies shifted up during this interval (even in dissections of dead pharate pupae). Furthermore, although the phenotype was repeatedly seen in several independent heat pulse tests using the balancer TM3, SbSer, when a TM3 balancer lacking the _Sb and Ser markers was used, leg fusion was never observed. Thus, i t appears that this t r a i t may be the result of some ts interaction between Q-III and these or other mutant alleles on the balancer. It is noteworthy that of the three pairs of discs producing legs, only the pair giving rise to the forelegs remain closely juxtaposed throughout development (Bodenstein, 1950). The fusion may reflect non-autonomous disc overgrowth, resulting from rapid c e l l proliferation at the medial edges of both discs. Simpson and Schneiderman (1975) described a similar phenotype for l(l)ts540, an X-linked autonomous ce l l - l e t h a l . (e) defects involving the genital disc A variable number of heterozygous (Q-III/TM3 or Q-III/pP) male progeny displayed malformation of their external genitalia when shifted up or down during development. This phene involved either incomplete rotation of the terminalia (see Miller, 1950) or in extreme cases, FIGURE 16 Scanning electron micrographs showing the forelegs of: (a) a normal f l y (p P/p P), grown at 22°C; and (b) a Q-III/TM3 fly, shifted from 29° to 22°C at 120 hours post oviposition, an arrow points out the proximal fusion of the legs (magnification about x400). 159 terminalization or lack of the genitalia. The TSP of this phene occurs during the second and part of the third larval instars. When homozygotes are heat-pulsed, higher proportions (see APPENDIX 2) of males treated during the larval intervals (particularly from 108 to 132 hours, 62 percent) exhibited the genital phene. Thus, the TSP of this phenotype can be more accurately defined. It li e s predominantly within the second half of the second instar (although i t spans the entire second as well as extending into the early third instar) and i t corresponds to the TSPs of many of the other phenotypes discussed heretofore. Miller (1950) claims that cells forming the external male genitalia (i.e. the terminal abdominal segments) normally undergo rotation during development. Whether the lack of Q-III + gene pro-duct directly prevents this normal process is unclear. Since female terminalia are not known to rotate, i t is not surprising that a com-parable phenotype was not observed in female progeny, (f) defects involving the abdominal histoblasts Although no homozygotes survived 48-hour heat pulses at 204 to 252 hours (during pupation),many of the dead imagoes (dissected from their cases) exhibited consistent abdominal anomalies including tergite malformation (the segments were missing, uneven, or etched) and incomplete pigmentation. In a few cases, whole patches of tergites were missing from the abdomen. This phenotype was observed at a lower frequency in homozygous survivors of shorter heat pulses at 204 to 228 hours. A few of the dead pharate pupae resulting from a 24-hour pulse showed tergite deficiencies or etching (see APPENDIX 2). 160 Interactions displayed by Q-III and other Minutes Some Minutes interact with non-allelic mutations to reduce via-b i l i t y or to enhance the expression of certain phenotypic abnormali-ties. As previously mentioned, some of the more well known interacting l o c i include: Delta (Schultz, 1929), vestigial (Green and Oliver, 1940) and Lyra (Lindsley and Grell, 1968). More recently i t has been observed that Minutes suppress the sex comb effects on the second or third pairs of legs of some homeotic mutations (R. Denell, personal communication). Consequently, possible interactions between Q-III and some of these mutants were tested for temperature-sensitivity. Although these inter-actions have both genetic and developmental significance, I have chosen to deal with the results of these tests here. Heterogeneity of genetic backgrounds in different stocks prevented a meaningful detailed assessment of differences in expressivity of the sex comb phenotypes. Therefore, a l l male f l i e s were scored s t r i c t l y for presence or absence of the supernumerary combs (any number of teeth) on either member of the second pair of legs. Table 26 summarizes the phenotypes and v i a b i l i t y of heterozygotes carrying Q-III and DI, V£, Ly_, Sex, Msc, or Pc^ when raised at 22° or 29°C. While Q-III/Dl, Q-III/Ly, Q-III/Sex and Q-III/Msc f l i e s were full y viable at 22°C (crosses 1,3,4 and 5), they were either totally inviable (Q-III/Dl, Q-III/Ly) or weakly viable (Q-III/Scx, Q-III/Msc) at 29°C. In addition, a l l Q-III/Sex fl i e s displayed a scalloping of the posterior wing margin at 29°C but not at 22°C. vg/+;Q-III/+ f l i e s were ful l y viable at both temperatures (cross 2, when compared with Q-III/p P at 29°C) and more than 80 percent possessed nicked wing Table 26 Lethal and Visible Phenotypes of Heterozygotes for Q-III and Dl, vg_, Ly_, Sex, Msc and _Pc at Different Temperatures Parental Genotype 1. Q-III/TM3 and D1/TM3 2. +/+;Q-III/TM3 and s s vg/vg;e /e 3. Q-III/CxD and CxD/Ly Surviving Progeny Progeny Genotype Q-III/TM3 D1/TM3 Dl/Q-III vg/+;Q-III/eS vg/-£;TM3/eS  Q-III/CxD Ly/CxD Ly/Q-III Number 22°C 29°C 119 5 121 175 127 0 236 172 220 40 30 50 359 4 139 0 Visible Phenotypes (Percent Expression) 22°C normal (100) slight Delta (100) n II normal wings (100) ebony (100) Dichaete (100) Lyra; Dichaete (100) Lyra (100) 29°C M (100) slight Delta (100) M (100); nicked wings (81.4) ebony (100) severe M (100); Dichaete (100) Lyra; Dichaete (100) Table 26 (continued) Visible Phenotypes (Percent Expression) Number Parental Genotype 4. Q-III/TM3 and Scx/TM3 5. Q-III/TM3 and Msc/TM3 6. Q-III/TM3 and Pc/TM3 Progeny Genotype Q-III/TM3  Scx/TM3  Sex/Q-III Q-III/TM3  Msc/TM3  Msc/Q-III Q-III/TM3 Pc/TM3 Pc/Q-III 22°C 249 269 292 202 197 203 77 69 83 29°C 61 970 10 7 893 102 4 291 143 22°C normal (100) *extra sex combs (51.6) normal wings (100); extra sex combs (48.2) normal (100) extra sex combs (58.3) " (68.4) normal (100) extra sex combs (96.4) (80) 29°C M (100) extra sex combs (99.6) M (100); nicked wings (100); extra sex combs (50) M (100) extra sex combs (91.1) M (100): extra sex combs (10) M (100) extra sex combs (100) M (100); extra sex combs (4.4) * Presence of sex combs on second legs of males 163 margins (particularly in distal margins) at 29°C, but none displayed such a phenotype at 22°C. This ts interaction resembles the non-ts effects of Minutes on vg. reported by Green and Oliver (1940). N. Dower (unpublished) discovered that EMS-induced Minutes also enhance' the expression of vg in heterozygotes. Eight of the 10 Sex/Q-III survivors at 29°C were males and 4 of these males had extra sex combs. Owing to the small numbers of these survivors, no further mention w i l l be made of them with respect to the extra sex comb phene. Q-III/Pc f l i e s were relatively viable at both temperatures. Amongst the male progeny of the Q-III/Pc and Q-IH/Msc constitutions, there was a marked reduction in the frequency of individuals displaying the extra sex comb tra i t at 29°C compared with that at 22°C (Pc/Q-III: 4 percent, down from 80 percent; Msc/Q-III: 10 percent, down from 68 percent). In summary, Q-III has the following interactions: (1) ts lethal-i t y with 1)1 and Ly. (2) reduced v i a b i l i t y with Msc and Sex at 29°C (3) ts scalloping of the wing margin with vg. and Sex and (4) ts sup-pression of the sex comb phenotypes of JPc and Msc. The ts suppression of homeotics by Q-III prompted an appraisal of combinations involving other Minutes and these l o c i . To this end, Df(3L)M(3)LS4 (hereafter abbreviated as M(3)LS4) and M(2)173 were tested with Sex, Msc and, in the case of the former M, JPc. The re-sults of the crosses performed are shown in Table 27. It should be mentioned that M(3)LS4 fli e s are poorly viable (undoubtedly due to the deletion), yet M(3)LS4/Scx fl i e s were even less viable, thereby re-sembling the combination, Sex/Q-III at 29°C. In contrast, v i a b i l i t y of Table 27 Interactions of Known Minutes With Different Surviving Progeny Parental Progeny (Percent) Genotype Genotype Number (Viability) M(3)LS4/TM3 M(3)LS4/TM3 252 (34.5) and Scx/TM3 Scx/TM3 357 (48.9) M(3)LS4/Scx 121 (16.6) M(3)LS4/TM3 M(3)LS4/TM3 129 (15.9) and Msc/TM3 Msc/TM3 373 (46.1) M(3)LS4/Msc 307 (37.9) M(3)LS4/TM3 M(3)LS4/TM3 6 (6.5) o and Pc /TM3 Pc/TM3 50 (54.4) M(3)LS4/Pc 36 (39.1) M(2)173/ SM5;+/+ M(2)173/+;TM3/+ 63 (21.6) and.+/+; Scx/TM3 SM5/+;Scx/+ 82 (28.1) M(2)173/+;Scx/+ 70 (24.0) SM5/+;TM3/+ 77 (26.3) otic Mutations Affecting Sex Combs Visible Number Phenotypes Showing (Percent Expression) Male Sex Comb Wing s Extra Normal Scalloped Normal 0 136 (100) 0 252 (100) 80(44.4) 100 (55.6) 0 357 (100) 3 (3.4) 85 (96.6) 121 (100) 0 0 70 (100) - -140 (70) 60 (30) - -21 (12) 154 (88) - -0 4 (100) - -20 (83.3) 4 (16.7) - -2 (11.8) 15 (88.2) - -0 32 (100) 0 63 (100) 24 (61.5) 15 (38.5) 0 82 (100) 11 (29.7) 26 (70.3) 35 (50) 35 (50) 0 35 (100) 0 77 (100) Table 27 (continued) Number Showing Surviving Progeny Visible Phenotypes (Percent"Expression) Male Sex Comb Wings Parental Progeny (Percent) Genotype Genotype Number (Viability) Extra Normal Scalloped Normal M(2)173/ SM5;+/+ M(2)173/+;TM3/+ 82 (21.5) 0 39 (100) - -and +/+;Msc/TM3 ' SM5/+;Msc/+ 107 (28.1) 51 (76 .1) 16 (23.9) - -M(2)173/+;Msc/+ 109 (28.6) 20 (37 .7) 33 (62.3) - -SM5/+;TM3/+ 83 (21.8) 0 46 (100) - -166 M(2)173/+;Scx/+ f l i e s was not decreased. Also no striking v i a b i l i t y effects were seen for M(3)LS4/Msc, M(2)173/+;Msc/+ or M(3)LS4/Pc progeny (although the cross which produced the latter yielded few progeny from 4 cultures). Both M(3)LS4/Scx and M(2)173/+;Scx/+ fli e s displayed a wing mar-gin phenotype characteristic of Q-III/Sex survivors at 29°C, with penetrance levels of 100 and 50 percent respectively. Expressivity of this phenotype was also greater in progeny of the former class. It is clear from the data (Table 27, columns 5 and 6) that M(3)LS4 suppresses expression of the sex comb phenotypes of Sex and Msc. Thus, 44 percent of Scx/TM3 males had extra sex combs while only 3 percent of Scx/M(3)LS4 did. Similarly, while 70 percent of Msc/TM3 males had extra sex combs, only 12 percent of M(3)LS4/Msc types did. Penetrance of _Pc is also affected, with more than 80 percent of Pc/TM3 males having had the mutant phene, while only 12 percent of M(3)LS4/Pc displayed i t . These results indicate that the other third chromosome Minute acts just like Q r l l l . However, i t should be remembered that Q-III might be an a l l e l e of M(3)LS4. The second chromosome Minute, M(2)173 also suppressed the sex comb traits of Msc and Sex (Table 27, columns 5 and 6). Thus, while 62 per-cent of the SM5/4-; Scx/+ males had extra sex combs, only 30 percent of M(2)173/+;Scx/+ males did. Further inspection shows that 76 per-cent of the SM5/+;Msc/+ males showed the tr a i t , compared to 38 percent of M(2)173/+;Msc/+ males. While suppression of Ms_c and Sex by M(2)173 is less efficient than by M(3)LS4, i t is nevertheless apparent that Minutes in general have a similar effect. Furthermore, Minutes 167 appear to interact with Sex to produce a phenotype which is character-ized by scalloping of the posterior wing margin in both males and females. This further strengthens the classification of Q-III as a genuine Minute mutation. The existence of interactions between Q-III and vg_, DI and Sex suggested the possibility that specific TSPs could be determined for these interactions. Consequently, regular shift studies were initiated for this purpose. (a) vg-Q-III wing scalloping vg/+;Q-III/+ progeny arising from the cross +/+;Q-III/TM3 x vg/vg; +/+.were scored for wing scalloping after shifts (Figure 17). If the beginning of the TSP is taken at the point where a shift down f i r s t gives mutant f l i e s , then this is clearly about or just after mid-second instar. The end of the TSP should coincide with the point where a shift up f i r s t yields non-mutants. These data indicate a very short TSP that terminates about two-thirds of the way through the second instar. Harnly (1936) exploited the suppression of the vg_ phenotype by high temperature to determine that the "temperature-effective period" for this gene extends from late second (about the second molt) to the early third instar, a later TSP estimate than that provided by this present study. (b) Dl-Q-III v i a b i l i t y Adult progeny arising from the cross Q-III/TM3 x D1/TM3 were counted after shift studies (Figure 18). It can be seen that the re-ciprocal shifts are not symmetrical, in that in the shifts down, a gradual reduction in v i a b i l i t y of DI/Q-III fl i e s occurred in cultures FIGURE 17 Results of the shift study to delineate a temperature-sensitive period for the vg-Q-III interaction. The data are given as percent expression of wing nicking in vg/Q-III progeny that were shifted from 22° to 29°C (closed circles), or from 29° to 22°C (open circ l e s ) , at various times during development. Temporal estimates of the different developmental stages at 29° and 22°C, are indicated above and below, respectively; E = eclosion. Developmental Stages at 29°C Time of Shift (in Hours) Developmental Stages J___i____^ at 22°C  I rr„„ I 1st I 2nd I 3rd T FIGURE 18 Results of the shift study to delineate a temperature-sensitive period for the Dl-Q-III interaction. The data are given as percent eclosion of Dl/Q-III progeny that were shifted from 22° to 29°C (closed cir c l e s ) , or from 29° to 22°C (open cir c l e s ) , at various times during development. Temporal estimates of the different developmental stages at 29° and 22°C, are indi-cated above and below, respectively; E = eclosion. Developmental Stages at 29°C I 1st I 2nd I Egg instar instar 3rd instar Pupa 100-•O-Q 75 A c o o 50-u Ul c o o I 25-O O • c » e o t>. 60 /20 180 Time of Shift (in Hours) 240 300 Developmental Stages at 22°C Egg 1 1st I 2~nd T instar instar 3rd instar T Pupa 1 E 172 shifted progressively later, while in the shifts up, normal v i a b i l i t y was attained f a i r l y rapidly. In this case, i f the beginning of the TSP is a r b i t r a r i l y taken as the point where 75 percent of DI/Q-III progeny survived shifts down (since shifts down thereafter generally produced more lethality) and similarly, i f the end of the TSP is set at the point where a significant increase in v i a b i l i t y followed a shift up, then an estimate of the TSP would be from about mid second to mid or late third instar. One of the problems encountered when collecting data for the above experiments was that Dl/Q-III progeny which emerged in cultures that were shifted down at 12-hour intervals from 92 to 192 hours (post oviposition), did so in a bimodal fashion in that, i n i t i a l l y , for a given shift only a few of the f l i e s eclosed, but after 3 or 4 days the bulk of the survivors were scored. This observation, along with the fact that DI/Q-III larvae showed extremely slow development at 29°C, indicates that even though prolonged exposure of these hetero-zygotes to 29°C greatly retards their growth, i t does not necessarily k i l l them. Most of the lethality of Q-III/Dl in cultures kept continuously at 29°C was larval, although a few dead pharate individuals were seen amongst a significant number of heterogeneous (with respect to devel-opment) pupae. Schultz (1929) mentioned that some M-Dl combinations died mainly as larvae. Although the Dl wing phenotype was severe in some of the Q-III/Dl individuals surviving shifts down during the TSP, no consistent pattern emerged. Possibly a more precise TSP could be derived from studying Q-III-mediated enhancement of the wing phenotype, 173 using other DI alleles that are less susceptible to Q-III-induced lethality. (c) Scx-Q-III wing scalloping Wings of Scx/Q-III survivors from the cross Scx/TM3 x Q-III/TM3 were scored for the presence of nicks and the results are shown in Figure 19. It should be mentioned that the pattern of semi-lethality of Sex/Q-III individuals is slightly earlier than the TSP for wing nicking. However, this lethality did create a problem. For example, few Sex/Q-III f l i e s survived shifts up at 24, 60 and 72 hours and none survived shifts up at 12, 36, 48 and 84 hours. Nevertheless, the shifts up immediately preceding the end of the TSP (96, 108, and 120-hour shifts) produced significant numbers of Scx/Q-III f l i e s , nearly a l l of which displayed the scalloped phenotype. It can be seen that the TSP for this interaction is confined to a small developmental period in the early part of the third instar. Qualitative differences in expression of this phenotype were noted, especially for shifts down at various times during the TSP. For example, many of the Scx/Q-III survivors from earlier shifts down showed anterior and/or posterior scalloping, while those shifted down progressively later (including the 132-hour shift) exhibited distal to proximal incisions in the wing blade. Indeed, some progeny pheno-typically resembled f l i e s carrying a less extreme form of apterous— (see Lindsley and Grell, 1968). Scx-ts67 interactions Since wing scalloping was displayed by fli e s carrying Sex along with Q-III, M(3)LS4 or M(2)173, i t could be postulated that i t is a FIGURE 19 Results of the shift study to delineate a temperature-sensitive period for the Scx-Q-III interaction. The data are given as percent ex-pression of wing nicking in Sex/Q-III progeny that were shifted from 22° to 29°C (closed circles), or from 29° to 22°C (open circles), at various times during development. Temporal estimates of the different developmental stages at 29° and 22°C, are indicated above and below, respectively; E = eclosion. 176 reflection of some general metabolic or developmental effect of Minutes. In that case, other non-Minute mutations such as the ts su(f) ts67g a l l e l e , l ( l ) s u ( f ) & (henceforth known as ts67) (Dudick e_t a l . , 1974) with similar biological activity, might also interact with Sex. To test this, a stock of ts67 (the X chromosome also bore a suppress-g ible a l l e l e of forked (f ) ) was obtained from T. Wright, and males of the genotype v f su(f)/Y;Scx/+ were constructed. These were crossed to ts67/ts67;+/+ females in bottles and the progeny allowed to develop at either 28° or 22°C (3 bottles at each temperature). Since ts67/Y males and ts67/ts67 females die at 30°C (Dudick e_t al_., 1974), cultures were maintained at 28°C in the hope that the males might survive, but at the same time display the 'deficiency' traits (small bristles, rough eyes, see Dudick et a l . , 1974), phenotypes which are similar to those of Minutes. This would permit a test for interactions between Sex and ts67 to be measured. The results of this test are summarized in Table 28. At 28°C a l l (183) of the male progeny were non-forked and displayed the deficiency phenotype. Fifty-five of the 183 males possessed extra sex combs, diagnostic of Sex and 31 of these 55 exhibited slight incisions in the posterior wing margin. In contrast,,at 22°C a l l males (136) had forked bristles (i.e. normal with respect to the 'deficiency' bristle phene) and 52 of these had extra sex combs. However, none of the males reared at 22°C possessed scalloped wings. A l l of the females (207) reared at 28°C were non-forked, and 58 of these displayed the deficiency pheno-type, but only 5 of the 58 had scalloped wings. In contrast, a l l of the females (150) produced at 22°C had forked bristles. Finally, no Table 28 Effect of ts67 on Sex Expression in Progeny of the Cross ts67/ts67;+/+ x v f S su(f)/Y;Scx/+ at 22° and 28°C Sex of Progeny Visible Phenotypes of Surviving Progeny  (X Chromosome Genotype) Scalloped Non-Scalloped Extra Sex Combs Normal Sex Combs Def. Non-Def. Males 22°C 0 136 52 84 0 136 22°C 0 150 - 0 150 Females (ts67/v f S su<f) ) 2 g o c 5 202 - - 58 149 Def. = Deficiency phenotype i.e. small, thin bristles Non-Def. = Normal bristles 178 consistent reduction in the penetrance of the extra sex comb phene was noted in ts67/Y;Scx/+ males at 28°C. These data suggest that the wing scalloping interaction between Minutes and Sex is probably not specific to M l o c i , since a similar, albeit less marked interaction exists between Sex and su(f). Summary of TSPs involving Q-III Figure 20 is a summary of the TSPs that have been provided by the shift experiments of the present study. It is clear that most of the TSPs affecting structures derived from the eye-antennal or wing discs, occur about the second instar (or early third) and that these overlap with the early lethal-sensitive as well as Dl-mediated i n v i a b i l i t y TSPs. In contrast, the TSPs associated with the differentiation of bristle derivatives (i.e. macrochaetae and sex combs) occur during early pupation, at almost the same time as the second TSP for lethality. This differential undoubtedly reflects a different basis for the origin of these defects. The TSP for female s t e r i l i t y (actually male s t e r i l i t y may be added as well) has been included for reference. The actual ex-tent of this TSP could include pre-eclosion stages, since such females were not tested for f e r t i l i t y . Finally, an embryonic TSP was included and i t is lik e l y that the latter could be more accurately delineated by using shorter shifts coupled with higher temperature. FIGURE 20 Temperature-sensitive periods (TSPs) for lethal, sterile and adult morphological effects of Q-III. Stippled bars indicate the extents of the TSPs relative to developmental stages given below. TSPs for recessive traits are summarized above the dashed line and include those for: embryonic lethality (le); larval and pupal lethality (1); s t e r i l i t y (s); defects involving aristae, scutella and male genitalia (a, s and g); eye reduction (e); sex comb rotation (c); and macrochaete reduction (ma); TSPs for dominant traits are summarized below the dashed line and include those for: rough eyes (E); macrochaete reduction (Ma); the lethal inter-action with DI (L(D1)); and the wing nicking inter-actions with vg (W(vg)) and Sex (W(Scx)). le a, s and g W(v3) W(Scx) Ma L(QI) Developmental Stages I I TTt I 2nd~l 3rd I I 181 IV. Discussion A temperature-sensitive mutation, Q-III has been found to produce a variety of ts phenes when homozygous and heterozygous. Furthermore, when Q-III is heterozygous with Dfd, vg, Sex, Pc, Msc and DI, ts inter-actions specific for each of these mutations are detectable. Unique or overlapping TSPs were determined for several of these attributes. Of the ts mutations known to have multiple, complex patterns of TSPs (Poodry et a l . , 1973; G r i g l i a t t i and Suzuki, 1970; Foster, 1973; Shellenbarger and Mohler, 1975; Holden and Suzuki, 1973), none has been found to be as highly pleiotropic as Q-III. The pattern of recessive lethality and extreme sensitivity of Q-III homozygotes to 48-hour heat pulses argues that the Q-III locus is indispensible for most of the larval as well as the pupal stage. However, the fact that the organism is relatively refractory to 24-hour exposures to 29°C (except for specific intervals) suggests that irreversible death is not mandatory unless a more prolonged period of deprivation of Q-III + product is involved. Thus, we might expect that larvae for example, might have two types of temperature-sensitivity, one in which a short heat pulse leads directly to death and others where such a pulse stops development but in a semi-reversible fashion. This is supported by the waves of pupation after heat treatment at different times. The temperature-sensitivity (of lethality) of Q-III contrasts with that of shibere, since the latter is extremely susceptible to very short exposures to high temperatures (Poodry et a l . , 1973). Preliminary observations show that the presence of functional Q-III + gene product is v i t a l also for embryogenesis and for adult 182 f e r t i l i t y . Eggs lai d by Q-III/Q-III females do not survive at 29°C. Even when Q-III/Q-III females are crossed to normal males, the Q-III/+ embryos f a i l to survive at the higher temperature, but do survive at 22°C. In contrast, many homozygous Q-III embryos produced by hetero-zygous females are i n i t i a l l y resistant to the lethal effects of high temperature (eventually succumbing after continuous exposure). This difference could indicate that functional gene product is supplied by the mother to the oocyte during oogenesis. The relationship of oogenesis vis-a-vis embryogenesis is interest-ing here, with reference to the nature of the Q-III lesion. If the process of oogenesis usually demands a minimum supply of Q-III + gene product, then i t is not surprising that partial or complete s t e r i l i t y is manifested at 22° and 29°C. The following 2 alternatives could be offered to explain why Q-III/+ progeny of homozygous Q-III females failed to develop at 29°C, while at least an appreciable proportion developed normally at 22°C: (a) the homozygous females can package enough essential molecules into the egg to support development at 22°C, but not at 29°C and (b) the Q-III gene product is thermolabile. Obviously the latter is the more attractive possibility, particularly with respect to eventual biochemical analysis of the Q-III gene product. 1. The Nature of Minutes and Q-III Is Q-III a temperature-sensitive Minute? Certainly Q-III possesses the basic phenotypic characteristics of Minute mutations. At the restrictive temperature, the Q-III hetero-zygote has lowered v i a b i l i t y , lengthened developmental period, and dominant expression of thin, small bristles, rough eyes, and other 183 less penetrant dominant phenotypes. In addition, the Q-III lesion is recessive lethal at 29°C. These properties have been reported for other mutations called Minutes (Lindsley and Grell, 1968). However, at 22°C no dominant effects are seen. The interactions of Q-III with Dl, Ly and vg have been described for other Minutes (see Schultz, 1929 Green and Oliver, 1940). At 29°C, Q-III/+/+ triploids have wild-type bristles and normal developmental periods, a result which has been reported for other Minutes in triploids (Schultz, 1929). Probably the strongest evidence is that the Q-III interactions led to predictions of Minute interactions that were indeed f u l f i l l e d . Although Q-III is lethal in combination with Df(3L)M(3)LS4, genetic mapping positions Q-III in a region between s_t and in which does not include the deleted portion of this proximal deficiency. There is a possibility that a Minute point mutant or a cytologically-invisible deficiency of the Q-III locus is present on the deficiency chromosome. At any rate, Q-III amply f u l f i l l s the properties of a genuine temperature-sensitive Minute. The function of Minute genes The burning question to be answered i s , what is the primary function of Minute loci? This study has not provided any definitive answers concerning the specific nature of Minutes. The proposal that some or many of these l o c i code for tRNA (K.C. Atwood, 1968) is s t i l l tenable, although preliminary genetic evidence argues against the idea that they are redundant (Huang and Baker, 1975). Temperature-TYR sensitive mutations of the tRNA locus have been reported for E. c o l i (Smith et a l . , 1970: Nomura, 1973). In one case, in vivo 184 experiments suggested that conformational changes in the mutant tRNA species led to irreversible inactivation of the molecule due to degrada-tion at high temperatures (Nomura, 1973). In yeast, a ts nonsense suppressor has been identified and its genetic properties indicated TYR that a mutant tRNA was likely involved (Rasse-Messenguy and Fink, 1973) . The in vitro a b i l i t y of tRNA isolated from a super-suppressor mutant in yeast to translate nonsense codons, argues that at least for lower eukaryotes, nonsense suppression occurs by tRNA-mediated transla-tional defects (Capecchi et a l . , 1975). These studies caution us against assuming that thermolabile gene products must be proteins, thereby leaving the possibility open that Q-III is a ts mutation in a tRNA gene. The pleiotropic phenotype of Q-III is clearly compatible with the 'tRNA hypothesis', especially in view of the quantitative and qualita-tive changes that occur for different tRNA species during development in Drosophila (White et a l , , 1973). The disruptive effects of tRNA abnormalities on protein synthesis which might be regulated differently in various c e l l types and discs, could adequately account for the pleiotropy, although the additional contribution of translational (Ilan, 1968) and post-translational (Jacobsen, 1971) control, is also possible. In situ hybridization to salivary gland chromosomes permits cyto-logical mapping of loci specifying tRNA transcripts ( G r i g l i a t t i et a l . , 1974) . If i t can be shown that an iso-accepting species maps near the Minutes i n proximal 3L, the possession of such a mutant w i l l be invalu-able (especially since the Q-III a l l e l e can be made homozygous at 22°C) 185 for the isolation and biochemical study of this gene product. It may be possible for example, to study the a b i l i t y of the homozygote to i n -corporate labelled amino acids into protein at the restrictive tempera-ture (see Farnsworth, 1970), thereby directly testing for disruption of protein synthesis in Q-III. Another idea that is mentioned in the literature but seldom con-sidered in detail is that Minute l o c i might code for ribosomal proteins. Previous evidence suggesting that the basal region of the X chromosome in Drosophila contained a cluster of genes specifying ribosomal proteins (Steffensen, 1973; Finnerty et a l . , 1973) has been recently disputed (Lambertsson, 1975b; Vaslet and Berger, 1976). Therefore, at least some of the Minute mutations could represent lesions in structural genes for these proteins. Berger and Weber (1974) found almost no polymorphic electrophoretic variants in ribosomal proteins of several different mutants including su(f) and various Minutes, as well as wild-type strains of Drosophila  melanogaster. One protein from the small ribosomal sub-unit in f l i e s bearing a third chromosome Minute did show altered electrophoretic migration. However, such an observation could arise from a second site mutation in a non-Minute locus. The authors did not indicate whether quantitative differences in protein patterns could be measured. The Minutes in which no differences in ribosomal proteins were noted could have been deficiencies or hypomorphs, in which case no qualitative differences would be expected. Lambertsson (1975a) analysed ribosomal protein patterns electro-phoretically at various stages of development in Drosophila. He was 186 able to detect from 69 (in pupae and adults) to 74 (in larvae) di f f e r -ent proteins, numbers which far exceed the estimated number of Minute l o c i (Lindsley e_t ajL., 1972). In addition, he found qualitative and quantitative changes in some of these proteins during development, particularly in the third instar. He (Lambertsson, 1975b) was unable to find evidence for the existence of a mutant ribosomal protein in ts67g the l( l ) s u ( f ) 6 strain. However, he did find that pattern changes characteristic of the larval to pupal transition in the wild-type were delayed in ts67 at higher temperatures, and he concluded that the mutation probably causes a severely reduced larval a b i l i t y to synthe-size imaginal ribosomes. Sussman (1970) proposed a model involving quantitative and quali-tative control exerted by ribosomes on translation during development. DeWitt and Price (1974) found differences in ribosomal protein patterns which corresponded temporally with stage-specific appearance of immature erythrocytes in Rana catesbeiana, and they suggested that Sussman's model may have merit. Thus, absence or abnormality of a ribosomal pro-tein could directly (through aberrant translational control), or in-directly (through generalized disruptions in protein synthesis) preci-pitate a large array of developmental defects. Future considerations of the nature of Minutes should be concerned with the fundamental genetic organization and control of these genes as well as their biochemical properties. Indeed, i t w i l l likely be possible to genetically answer many of the outstanding questions, be-fore their biochemical dissection has been accomplished. 187 Is Q-III a sin g l e s i t e lesion? The multiple phenes displayed by Q-III could have a t r i v i a l basis such as several d i f f e r e n t mutations on the Q-III chromosome. Cytologi-c a l examination of the s a l i v a r y gland chromosomes of Q-III by T. C. Kaufman, revealed no c y t o l o g i c a l l y - v i s i b l e aberration i n chromosome 3. While a l l EMS-induced ts mutations usually map g e n e t i c a l l y at single s i t e s (Suzuki, 1970), the occurrence of multiple mutations must be considered as a p o s s i b i l i t y . Since a l l experiments of the present study were performed using a recombinant stock i n which the d i s t a l 3R region of the Q-III chromosome, from Hairless(69.5) to the t i p was re-placed, this l i k e l i h o o d i s diminished. However, this i s only a r e l a -t i v e l y small segment of the chromosome and second s i t e mutations could reside elsewhere. Three main findings support the notion that defects associated with Q-III represent expression of a sin g l e mutant s i t e . (1) In no case i n ei t h e r of the mapping experiments (see Table 22) were the rough eye, b r i s t l e or l a t e - e c l o s i n g phenes separated from each other by recombination. For example, a l l 7 l a t e - e c l o s i n g recom-binants between Gl and H possessed small b r i s t l e s and rough eyes as did a l l 11 l a t e - e c l o s i n g crossovers i n the s_t to p_P i n t e r v a l . Further-more, while many of the l a t e eclosing, e y e - b r i s t l e recombinants ex-h i b i t e d leg anomalies (e.g. shortened t a r s i , gnarled legs) and wing v e i n d i s r u p t i o n , as well as absence of p o s t - v e r t i c a l s , none of the recombinants with normal b r i s t l e s , eyes and developmental periods displayed any of these d e f e c t s . T h i s was also v e r i f i e d for recombinants between s_t and £ P recovered at 22°C and retested at 29°C. While a l l 188 late-eclosing recombinants which displayed rough eyes and bristle phenes produced some progeny with different combinations of eye, wing, leg, and genitalia defects, none of these phenotypes was ever seen in + recombinant progeny that were M . (2) Some of the ts phenotypes of Q-III such as ommatidial defi-ciencies and rough eyes have also been observed for ts67 (Dudick et a l . , 1974). Furthermore, the effects of the combination of Q-III and Sex on the wings at 29°C were also apparent for ts67 males (and a few females) carrying Sex at 28°C. Since Dudick e_t al_. claimed that most of the original X chromosome containing this mutation was replaced, ts67 is li k e l y due to a single lesion. Thus, a single site ts Minute such as Q-III, could mimic the known pleiotropic effects of ts67. (3) If other third chromosome mutant sites are responsible for some of the mutant phenes expressed by Q-III, either: (i) the second site(s) must be a ts a l l e l e of a separate locus or ( i i ) the other site(s) must interact with Q-III at 29°C but not at 22°C. The proba-b i l i t y of (i) would be low since the frequency of EMS-induced third chromosome Minutes is 1 in 3000 tested chromosomes (0.00033) and that of third chromosome recessive ts lesions is 0.055 (including both lethals and visibles) at the same dose of EMS (Tasaka and Suzuki, 1973). The probability of such a double mutant would be 0.00033 x 0.055 = 1.8 x 10 ^ or about 1 in more than 55,000 chromosomes. S t i l l , such a possibility cannot be dismissed entirely because Simpson and Schneiderman (1975) reported the recovery of a doubly mutant X chromo-some containing a ts a l l e l e of scalloped ,as well as a ts autonomous cell-l e t h a l mutation. 189 The second possibility ( i i ) is d i f f i c u l t to rule out. The best way to counter this is to separate the smallest segment of the original chromosome which s t i l l carries the ts Minute locus, or alternatively, to map this locus cytologically using deficiencies. Thus, while i t is possible that more than one mutant site is involved in Q-III, i t is extremely unlikely. 2. Developmental Characteristics of Q-III With a few exceptions, shorter (i.e. 24-hour) exposures of Q-III cultures to 29°C did not k i l l the animals. However, such treatment resulted in the production of many imaginal defects. The type of defects produced, depended on when during development the cultures were pulsed to 29°C. Therefore, the pulse experiments helped to de-fine several TSPs for the pleiotropic phenotype of Q-III. While most of the adult defects resulted from exposure of Q-III larvae (usually during the second or third ins tars) to 29°C, a few resulted from exposure during pupation. The former phenes w i l l here-after be referred to as pattern defects. These are particularly ex-emplified by deficiencies, duplications or other abnormalities which involve derivatives of the eye-antennal and dorsal mesothoracic discs. Such defects were observed in heat-treated progeny which bore Q-III alone, or in combination with Dfd (eye-antennal disc), vg or Sex (dorsal mesothoracic disc) . Pattern defects involving derivatives of other imaginal discs, such as the genital and leg discs were also seen, but at lower frequencies. Since nearly a l l of the mitotic activity of the imaginal discs of Drosophila is restricted to the larval stages (see Nothiger, 1972), i t i s not surprising that the pattern defects of Q-III are associated with heat treatment during these stages. 190 Pattern defects and c e l l death Cell death in different parts of an imaginal disc, followed by varying degrees of proliferation could explain most of the eye-antennal phenes of Q-III homozygotes and heterozygotes. Deficiencies, particularly of eye tissues, were the most frequently encountered pattern defects in derivatives of the eye-antennal discs of Q-III-bearing f l i e s . Figure 21 is a schematic representation of an eye-antennal disc. If Q-III-induced c e l l death occurred more frequently in region a, lack of adequate c e l l replacement would produce ommatidial deficiencies . Less frequently, extensive c e l l death could embrace several regions simultaneously, thereby accounting for the occasional concomitant absence of eye (a)antennal (b), and head (d) structures, etc. (assuming that the presumptive cells were not replaced). More restricted c e l l death involving specific regions, for example, b, c, and d, could lead to deficiencies of individual antennal structures, o c e l l i or palps, respectively. According to Bryant (1971), localized c e l l death in an imaginal disc could produce deficiencies, as well as regeneration of the missing structures, or mirror-image duplication of structures already present. The type of result depends on the repatterning of the new c e l l s . Bryant suggests that there is a "developmental" gradient in which cells at the 'high' end are totipotent, whereas cells at the 'low' end are more limited. Thus, i f c e l l death occurred and was mitotically compensated for in a region of the disc that is nearer to the top of the developmental gradient, the new cells could be re-patterned to produce structures normally derived from portions of the disc that are FIGURE 21 A schematic diagram of a mature eye-antennal imaginal disc showing the following structures: presumptive f i r s t (AI), second (All) and third (AIII) antennal segments; presumptive arista (Ar); presumptive eye (E); presumptive head (H); presump-tive palpus (P); and the proposed ocellar region (0). Regions a, b, c, d, e and f are hypothetical zones of c e l l death; (after Gehring, 1966). 193 further down the gradient and regeneration would result. However, ce l l death and subsequent replacement in a section of the disc that is low in the gradient would allow only for the 'regeneration' of lower structures and as a result, mirror-image duplication would occur. In the case of the schematic eye-antennal disc (Figure 21), i f ce l l death \was. induced in region b to obliterate the antennal portion of the disc, adequate replacement of cells at region e could lead to regeneration of the antenna (Gehring, 1966; Bryant, 1971). On the other hand, i f c e l l death in the area of the disc between e and f was followed by c e l l divisions near the antennal anlagen (i.e. at f ) , this could result in mirror-image duplication of the antenna. Finally, i f c e l l death in this same section was followed by simultaneous prolifera-tion at both e and f, concomitant regeneration and duplication could take place, thereby giving rise to a tri p l i c a t i o n of antennal struc-tures. This present study has demonstrated that a l l of these eye-antennal pattern defects occur in Q-III-bearing f l i e s exposed to 29°C during the larval period. Verification of ts induction of c e l l death by Q-III w i l l depend upon histochemical tests of discs. Such investigations could indi-cate whether c e l l death is random or localized. In this regard, the temperature-sensitivity of Q-III should prove amenable to jLn vitro analyses of c e l l death in isolated discs. If we can assume that Q-III is an autonomous cell-lethal like other Minutes (Stern and Tokunaga, 1971), then this is the f i r s t re-port of a ts autonomous cell-lethal mutation on an autosome. Since Minutes have been shown to effect cell-autonomous reduction in mitotic rate (Morata and Ripoll, 1975), this may explain the low frequency of 194 duplications and triplications found in this study. Thus, even brief exposure of the disc to high temperature could prevent c e l l replacement that might normally follow c e l l death. In light of the above, i t would not be surprising i f some sort of ce l l death in the wing disc is also responsible for the scutellar pattern defects observed in Q-III-bearing individuals. According to Bryant (1975), fragments isolated from the mesothoracic (wing) discs which include the anlagen of the scutellum, frequently give rise to structural duplications. However, since misplaced bristles or sockets are fre-quently scattered over the scutella of Q-III f l i e s , clear pattern dis-ruptions, as well as what appear to be duplications, are occurring. While wing scalloping was not seen in Q-III f l i e s , in adults bear-ing Q-III in concert with Sex or vg at 29°C, such wing margin effects are quite striking, particularly in Sex/Q-III f l i e s . Fristrom (1969) showed that the wing phenotypes of y j * , Beadex (Bx), cut (ct) and Xa apterous— are the result of c e l l death either during the third larval instar, or early pupation (in the case of ct). Santamaria and Garcia-Bellido (1975) used an elegant approach involving clonal analysis to show that the c e l l death in Bx occurs at about the middle of the third instar. In the present work, the TSP for the Q-III-Scx wing phene occurs during the f i r s t half of the third instar and therefore i t is likely that c e l l death occurs at that time or shortly thereafter. It is worthy of mention however, that the TSP for the Q-III-vg interaction seen in this study is considerably earlier than the time when c e l l death purportedly takes place. Even i f Harnly's (1936) TSP for vg_ at the end of the second instar is quite precise, Fristrom (1969) has 195 reported that c e l l death is not detectable until the late part of the third instar in wing discs of vg f l i e s . Cell death routinely occurs in wild-type discs of Drosophila. For example, Spreij (1971) observed degenerating cells in the wing pouch and other sections of the dorsal mesothoracic disc, at about mid third instar. In addition, Fristrom (1969) detected similar necrosis in the wild-type eye-antennal discs, particularly near the area between the eye and antennal portions, in both the second and third instars. Since c e l l death has been offered as a potential contributor to morphogenesis in normal discs (see Nothiger, 1972), one hypothetical Xa mechanism by which mutants such as vg, ap—, Bx and cJ: could produce wing defects, would involve an extension of the boundaries of normally-restricted c e l l death which is assumed to occur during the formation of the wing margin (Bryant, 1975). A similar explanation could apply to eye mutants such as Bar, eyeless, and Dfd with regard to extensive destruction of presumptive facet tissue promoted by these mutants. Presumably, Q-III alone, or in combination with mutants like Sex (wing disc) and Dfd (eye-antennal disc) might similarly extend the boundaries of c e l l death upon exposure to 29°C by making more cells susceptible to the genetically programmed, regionally-specific c e l l death. Further study of these Q-III-mediated interactions should pro-vide considerable information about c e l l death during development in imaginal discs, and i t s relationship to pattern phenomena such as duplication formation and regeneration. Defects resulting from heat treatment of Q-III during pupation Whereas most of the pattern defects of Q-Ill-bearing f l i e s (homo-zygotes) were produced by briefly exposing larvae to 29°C, other imaginal 196 defects were expressed by Q-III individuals that had been heat-pulsed during pupation. For example, the classical M phene of short bristles, as well as the comb gap phene, were observed in Q-III adults which emerged from cultures treated in this manner. This raises the possi-b i l i t y that these latter abnormalities are directly related to the disruptive effects of the Q-III lesion on protein synthesis at the time of differentiation. This situation contrasts with that of the pattern defects, which probably originate from imaginal c e l l death at much earlier stages than differentiation (i.e., during the larval stages). The finding that the TSP for the Minute bri s t l e phenotype occurs at the time of bristle formation during the f i r s t half of pupation (Bodenstein, 1950) is consistent with the idea that rapid accumulation of protein is a prerequisite to bristle synthesis (Howells, 1972) and that any disturbance in the translation process (i.e. bb, s u ( f ) t s ( ^ Q r Minutes) directly results in attenuated bristles. My findings are in agreement with those of Dudick e_t a l . (1974) who determined that ts67 suppression of forked occurs at the time of bristle formation. It should be possible to use this phenotype to select for dominant and recessive mutations that affect translation. While the TSP of the comb gap phenotype also occurs during the f i r s t half of pupation, the situation here with respect to the effects of Q-III on protein synthesis, is unclear. Cell lineage studies of the legs of Drosophila males (Tokunaga, 1962) have revealed that cells in the region of the basitarsus which contain the presumptive sex comb tissue (the combs are actually modified macrochaetae) undergo a char-acteristic shift and rotation of about 90 degrees, so that the formerly transverse row becomes longitudinally placed. Tokunaga makes no mention 197 of when this shift actually occurs in the presumptive sex comb tissue. If rotation occurs during the early part of pupation, then presumably the lack of Q-III~*~ gene product could prevent successful rotation. For example, the effects of Q-III on translation could k i l l key cells in-volved in the rotation. On the other hand, rotation might occur earlier and removal of the Q-III + product at, or just prior to leg eversion could de-stabilize the sex-comb alignment. Remember that rotation of the terminalia in males is also blocked by exposing Q-III progeny to 29°C. However, in this case, the TSP is larval. Further study of this ts Minute w i l l clearly increase our knowledge about such morphogenetic processes. The interactions of Q-III This study has shown that lesions such as Q-III can be useful for the developmental study of other genes for which ts alleles are un-available (see Dudick et a l . , 1974). Thus, TSPs have been defined for yg, Dl and Sex. The TSP for the vg-Q-III interaction occurs in the second half of the second instar, while that for the Scx-Q-III inter-action f a l l s in the f i r s t half of the third instar. The Dl-Q-III TSP for lethality extends from the mid second to mid third instars. It is interesting that the TSP for Sex is larval since this may also be true for its homeotic effect. However, i t should be emphasized that more than one TSP may exist for a given gene product ( G r i g l i a t t i and Suzuki, 1970; Mglinets, 1975). Therefore, the possibility of additional TSPs should be considered for the above genes. Similar use of Q-III could permit the delineation of the time of action of the gene products of Dfd and Ly_. The suppression of the sex 198 comb phenotypes of some of the homeotics by Minutes has been demonstrated. Q-III might also provide an estimate of the interval in development when the gene product of Polycomb (the most fully penetrant of the sex comb homeotics) is ut i l i z e d . It is noteworthy that of the mutants affected by Q-III at 29°C, both vg_ and _p_l are dosage-sensitive (Lindsley et a l . , 1972). The basis for the Dfd lesion is unknown and there is some evidence which suggests that at least Pc is an antimorph (Puro and Nygren, 1975). Thus, fl i e s heterozygous for deficiencies for y_g_ and Dl display the respective phenotypes (Lindsley et a l . , 1972). The interaction of Minutes with these mutations could therefore be due to the inhibitory effects of Minutes on protein synthesis. In other words, the Minutes could be producing a synthetic hypomorphic situation by decreasing the amount of vg_ or 1)1 product. If this is true then we might expect that the expression of heterozygous deficiencies for these l o c i would be en-hanced by Q-III at 29°C and this would effectively rule out the possi-b i l i t y that Q-III is interacting with mutant yj» or D_l gene products. Other dosage-sensitive l o c i such as Ultrabithorax, Intersex and Hairless on chromosome 3, and Star, black and Plexate on chromosome 2 (Lindsley e_t ajL., 1972) could be tested to see i f they are also enhanced by Q-III. The interaction of Q-III with the homeotics is clearly complex. While Q-III suppresses the sex comb phenes of JPc and Msc, i t decreases the v i a b i l i t y of Msc-, Sex-, Antp-, and possibly Ns-bearing f l i e s . Furthermore, i t interacts with Sex to produce a new mutant phene, wing scalloping. The latter phene is not specific to Minutes, suggesting 199 that this interaction is metabolic rather than specific. It might be argued that the retarding effects which Minutes have on development actually allow the accumulation of the antimorphic gene products of (some of) the sex comb homeotics, thereby leading to the production of new phenotypes or lethality. The suppression of the sex comb phenes is puzzling since Q-III does not inhibit sex comb formation on the forelegs. This phenomenon clearly merits further study. The fact that ts67 interacts with Sex in a fashion similar to Q-III, underscores the contention that many of the phenotypic inter-actions described for different l o c i in Drosophila (i.e. suppression or enhancement) may be metabolic rather than specific (Kaufman et a l . , 1973) . Therefore, i t is important to exercise caution when attempting to interpret interactions in specific terms as has been done with the proposed nature of the suppression of forked by ts67 (Dudick jet a l . , 1974) . It may be that this suppression is due to generalized decreases in protein synthesis rather than to ribosomally-mediated, informational suppression. Additional uses for Q-III in studies of development Several potential uses for Q-III in the study of development emerge from the fact that Minutes lower mitotic rates in a cell-autonomous fashion (Morata and Ripoll, 1975). For example, i t should be possible to produce clones of Q-III homozygous cells in a Q-III heterozygous background at 22°C by somatic crossing over. Since the background cells would be wild-type at 22°C, such an approach may allow more information to be gleaned from studies which exploit this phenomenon. In one + + instance, Garcia-Bellido _et _al. (1973) used the tendency for M /M 200 clones to overgrow their heterozygous background to investigate devel-opmental compartmentalization in the wing disc. The possession of Q-III should considerably expand the scope of similar investigations, particularly those involving the eye-antennal disc. The detrimental effects of Q-III on v i a b i l i t y and growth of larvae are amply demonstrated in this study. Some of the Q-III-induced larval lethality could be due to the inability of the larvae to metamorphose, which in turn could be related to lack of competence of imaginal disc cells (see Nothiger, 1972) because of death or slow mitosis. However, larval death is undoubtedly also due to the direct effects of Q-III-mediated disruptions in protein synthesis. It is known that larval cells grow by increases in c e l l size, while imaginal disc grow mitoti-cally (Bodenstein, 1950). It should therefore be possible to speci-f i c a l l y probe the growth of imaginal vis a vis that of larval tissue. 201 CHAPTER 6 OVERVIEW The research described in this thesis was designed to investigate regions near the centromere of chromosome 3 of Drosophila melanogaster, with the following aims: (a) to investigate proximal recombination in females with a view to examining interference and interchromosomal effects (b) to determine i f Deformed, a mutation which maps near the centromere, is recessive lethal and to map this locus relative to Kinked (c) to see i f selecting for radiation-induced, proximal cross-overs w i l l enrich for deletions in proximal segments and (d) to genetically and developmentally characterize Q-III, a ts a l l e l e of a proximally-located Minute. For the most part these objectives have been realized. The experiments described in CHAPTER 2 have helped to further characterize crossing over near the centromere of chromosome 3. First, some results suggested that much of the increase in proximal crossing over caused by the inverted attached-X chromosome C(1)M3, preferentially takes place within centrically-adjacent euchromatin. Since i t is likely that l i t t l e or no spontaneous crossing over occurs in heterochromatin, such recombinagenesis may be confined solely to proximal euchromatic segments. Second, this study has provided results suggesting that multiple (double and triple) crossovers are detectable at higher than expected frequencies in proximal intervals. The latter results are of particular interest since they mark a striking departure from two classical rules of intergenic recombination: (a) positive chromosome interference within a given region varies 202 inversely with the genetic size of that region and (b) crossing over within different arms of the same chromosome is independent, i.e. positive interference does not extend across the centromere. Three possible explanations were offered to account for these results: (1) the Two-Step model (mitotic followed by meiotic crossovers) (2) chromatid interference and (3) gene conversion. The following experimental evidence supports the f i r s t possibility: (i) females producing multiples also showed higher frequencies of crossing over than did those females producing no multiples and ( i i ) fewer double relative to triple exchange tetrads were observed in a tetrad analysis of the data. However, the predicted clustering of single crossovers which would likely accompany a Two-Step production of multiple crossovers, did not occur for females producing double crossovers. Two lines of evidence were presented in support of the idea that gene conversion may be involved in the production of putative multiple crossovers. First, equal numbers of reciprocal crossover classes were not recovered. Second, when crossing over was measured in females carrying C(1)M3, negative interference in most proximal intervals showed a relative decrease, in spite of the fact that crossing over was increased in a l l intervals, whatever the cause of this observed high negative interference, these findings have introduced a totally new dimension to the consideration of linked exchange in this organism. CHAPTER 3 described a genetic study of the Dfd locus. The results of the mapping experiment suggest that Ki and Dfd are very close, genetically. The results of the analysis of a large number of recombin-ant crossover chromosomes and successful synthesis of a homozygous Dfd 203 stock, confirm the notion that the Dfd lesion is by nature homozygous viable. Therefore, most Dfd stocks must carry at- least 1 lethal site in addition to the Dfd mutation. However, the Dfd stock examined in the present study probably carried a minimum of 3 extra lethals. This information w i l l allow a more complete assessment of devel-opmental and genetic studies of Dfd. It also emphasizes the fact that cytological mapping is preferable to crossover mapping, particularly when the mutation(s) in question lies near the centromere. Thus, the need for a wider inventory of stable proximal deficiencies and duplica-tions is obvious. The results of the study described in CHAPTER 4 agree with the idea that a large proportion of induced crossovers recovered from irradiated males, arise via asymmetrical exchange. Some of the results also support other workers in their claim that lethals are more common to crossover chromosomes, when the crossovers are produced by induced exchange within proximally-adjacent euchromatic segments. In future, similar investigations w i l l not only be sources of proximal aberrations, but w i l l also provide considerable information about induced crossing over. Of particular interest w i l l be the com-parison of exchange within heterochromatic and proximally-adjacent euchromatic segments of the chromosome. CHAPTER 5 was a detailed description of the genetic and develop-mental properties of Q-III, a ts a l l e l e of a Minute which is located near the centromere of chromosome 3. This represents the f i r s t report of a truly conditional Minute a l l e l e i.e. one which e l i c i t s the dominant M traits under restrictive conditions, but produces no such effects under permissive conditions. 204 The phenotypic similarities between Q-III, ts67 ( s u ( f ) — " ) and bobbed, favour the idea that Minute l o c i are involved at some level(s) in the process of translation. No conclusions about the exact nature of M gene products were reached in this study. In the case of Q-III, the primary product may be thermolabile. However, this would not eliminate the possibility that Minutes code for products as diverse as tRNA. species or ribosomal proteins. Indeed, the assortment of M gene products may not be homogeneous with respect to the different components of translation. The pleiotropy of Q-III is impressive. By exposing cultures to 29°C at various developmental intervals, i t was possible to k i l l Q-III homozygotes as well as to produce homozygous (and heterozygous) adults with a large spectrum of imaginal defects. While the TSP of lethality is polyphasic, TSPs of the different phenotypes are generally mono-phasic. The patterns of lethality and s t e r i l i t y show that the Q-IIl"*" product is essential for nearly a l l stages of development. However, i f the organism is deprived of this substance for shorter periods, v i a b i l i t y is more normal, but the phenotypic anomalies are seen. The observation that homozygous Q-III individuals (produced by heterozy-gous mothers) frequently reach the f i r s t larval instar before dying at 29°C, suggests that considerable Q-III + product is supplied to the developing oocyte. This is further supported by the observed ts maternal effects of Q-III. Sensitivity of Q-III larvae to heat-induced lethality i s marked by either a reversible blockage of growth (usually followed by death), or more frequently by f a i r l y rapid death. On the other hand, pupal 205 sensitivity appears to be related to blockages in differentiation. A parallel difference is apparent for some of the imaginal defects. Thus, most pattern defects,for example, those involving eye-antennal or wing disc derivatives, were produced by exposure of larvae to 29°C (usually during the second or third instar), whereas other defects such as attenuated bristles or abnormal sex combs, resulted from heat exposure during pupation. It is lik e l y that the former phenes are caused by c e l l death within the respective anlagen in larvae, while the TSPs of the latter defects imply that they are due to the direct disruption of differentiation in Q-III individuals. The Q-III-mediated translational d i f f i c u l t i e s can easily account for a l l of the phenotypic traits of this mutant. A l l developmental stages of Q-III homozygotes which require protein synthesis would inevitably succumb to heat-induced death. At 29°C, an embryo produced + by a Q-III female could survive only as long as its supply of Q-III product lasted. Cell death, which is lik e l y responsible for pattern defects exhibited by imagoes heat-treated as larvae, could be the result of the cell-autonomous failure of translation. On the other hand, imaginal defects s,uch as the small bristle and comb gap phenes, which are seen in individuals heat treated during pupation, could be produced by impaired differentiation due to translational collapse at c r i t i c a l intervals in Q-III pupae. Finally, a lack of protein syn-thesis would probably result in the s t e r i l i t y of Q-III (homozygous) adults at 29°C. Several intriguing ts interactions between Q-III and non-allelic genes were documented in this study. Some of these interactions (but 206 non ts) had been previously reported for Minutes, v i z . synthetic lethality with DI and Ly_, production of wing scalloping with vg_ and suppression of the sex comb phenes of some of the homeotic mutants. In addition, I was able to show that Q-III is lethal in combination with Dfd at 29°C and that this situation is due to the failure of eye-antennal disc derivatives to develop. Q-III/Scx and Q-III/Msc com-binations are less viable than controls at 29°C and the former indi-viduals possess variable nicking of the posterior wing margin. It was possible to determine TSPs of the gene products of vg, Dl and Sex. In each case the TSP is larval and in the case of Sex and vg, c e l l death i s probably involved in the production of these tr a i t s . The notion that these interactions are specific rather than meta-bolic in nature, is challenged by the observations that ts67 and other Minutes interact similarly with Sex at 29°C, and also that at least one additional Minute mutation suppresses the sex comb phene of Msc. These findings imply that the reduced translational capacity of Q-III-bearing individuals is sufficient to account for the observed interactions, without invoking the idea of gene-product interactions involving Q-III and the other l o c i . The wide spectrum of Q-III properties has thus provided additional scope for the analysis of Minutes in general, and this proximally-located Minute in particular. That Q-III possesses great potential for the study of developmental processes peculiar to i t s e l f or other non-a l l e l i c mutations, has been amply demonstrated by these experiments. \ 207 BIBLIOGRAPHY Arajarvi, P., and Hannah-Alava, A. (1969). Cytogenetic mapping of in and r i . Dros. Inf. Serv. 44: 73. Arking, R. (1975). Temperature-sensitive cell-lethal mutants of Drosophila: isolation and characterization. Genetics 80: 519-537. Atwood, K. C. (1968). In "Genetic Variations of Drosophila melano-gaster" (D. L. Lindsley and E. H. Grell, eds.) Carnegie Inst. Wash. Publ. 627. Baker, W. K. (1958). Crossing over in heterochromatin. Amer. Natur. 92: 59-60. Baldwin, M., and Chovnick, A. (1967). Autosomal half-tetrad analysis in Drosophila melanogaster. Genetics S5: 277-293. Baldwin, M. C, and Suzuki, D. T. (1971). A screening procedure for the detection of putative deletions in proximal heterochro-matin of Drosophila. Mutation Res. _11: 203-213. Beadle, G. W. (1932). A possible influence of the spindle fiber on crossing over in Drosophila. Proc. Natl. Acad. Sci. (U.S.A.) 18: 160-165. Berger, E. M., and Weber, L. (1974). The ribosomes of Drosophila. II. Studies on intraspecific variation. Genetics _78: 1173-1183. Bodenstein, D. (1950). The postembryonic development of Drosophila, In "Biology of Drosophila" ( (M. Demerec, ed.) pp. 175-367. John Wiley and Sons, New York. Bole-Gowda, B. N., Perkins, D. D., and Strickland, W. M. (1962). Crossing over and interference in the centromere region of linkage group I of Neurospora. Genetics 4_7: 1243-1252. Brehme, K. S. (1939). A study of the effect on development of Minute mutations in Drosophila melanogaster. Genetics 24: 131-161. Bridges, C. B. (1916). Non-disjunction as proof of the chromosomal theory of heredity. Genetics 1: 1-52. Brosseau, G. E., Jr. (1960). Genetic analysis of the male f e r t i l i t y factors on the Y chromosome of Drosophila melanogaster. Genetics 45: 257-274. Brown, S. W. (1966). Heterochromatin. Science (Washington) 151: 417-425. 208 Bryant, P. J. (1971). Regeneration and duplication following opera-tions in situ on the imaginal discs of Drosophila melanogaster. Devel. Biol. 26: 606-615. Bryant, P. J. (1975). Pattern formation in the imaginal wing disc of Drosophila melanogas ter: fate map, regeneration and duplica-tion. J. Exp. Zool. 193: 49-78. Calef, E. (1957). Effect on linkage maps of selection of crossovers between closely linked markers. Heredity 1_1: 265-279. Capecchi, M. R., Hughes, S. H., and Wahl, G. M. (1975). Yeast super-suppressors are altered tRNAs capable of translating a nonsense codon in vitro. Cell 6_: 269-277. Chase, M., and Doermann, A. H. (1958). High negative interference over short segments of the genetic structure of bacteriophage T 4 . Genetics 43: 332-353. Chovnick, A., Ballentyne, G. H., and Holm, D. G. (1971). Studies on gene conversion and i t s relationship to linked exchange in Drosophila melanogas ter. Genetics 6S>: 179-209. Cooper, K. W. (1959). Cytogenetic analysis of major heterochromatic elements (especially Xh and Y) in Drosophila melanogaster, and the theory of "heterochromatin". Chromosoma (Berlin) 10: 535-588. Davis, B. K. (1974). Chromatid interference in Drosophila melanogaster. Genetics 7_7: 16s. Denell, R. E. (1972). Reversion studies of Nasobemia. Dros. Inf. Serv. 48: 45. Denell, R. E. (1973). Homeosis in Drosophila. I. Complementation studies with revertants of Nasobemia. Genetics 7_5: 279-297. DeWitt, W., and Price, R. P. (1974). Structural changes in ribosomes during development. Biochem. Biophys. Res. Comrnun. 5_6: 593-598. Dobzhansky, T. (1930). Cytological map of the second chromosome of Drosophila melanogaster. Biolog. Zent. 50: 671-685. Dudick, M. E., Wright, T. R. F., and Brothers, L. L. (1974). The developmental genetics of the temperature-sensitive a l l e l e of the suppressor of forked, 1 ( l ) s u ( f ) — — ^ in Drosophila  melanogas ter. Genetics _76: 487-510. Duncan, I. W., and Kaufman, T. C. (1975). Cytogenetic analysis of chromosome 3 in Drosophila melanogaster: mapping of the proximal portion of the right arm. Genetics 80: 733-752. 209 Farnsworth, M. W. (1957a). Effects of the homozygous Minute - IV deficiencies on the development of Drosophila melanogaster. Genetics 42: 7-18. Farnsworth, M. W. (1957b). Effects of homozygous f i r s t , second and third chromosome Minutes on the development of Drosophila  melanogas ter. Genetics 42: 19-27. Farnsworth, M. W. (1970). Uptake and incorporation of amino acids in Minute mutants of Prosophila. J. Exp. Zool. 175: 375-382. Finnerty, V. G., Barton, L. M., Schalet, A., Elmer, W. A., and Smith, P. D. (1973). The suppressor of forked mutation: a putative protein synthesis mutant in Drosophila melanogaster. Genetics 74: 579. Foster, G. G. (1973). Temperature-sensitive mutations in Drosophila  melanogaster. XIII. Temperature-sensitive periods of the lethal and morphological phenotypes of selected combinations of Notch-locus mutations. Devel. Biol. 32: 282-296, Friesen, H. (1933). A r t i f i c i a l l y induced crossing-over in males of Drosophila melanogaster. Science (Washington) 78: 513-514. Friesen, H. (1937a). A r t i f i c i a l release of crossing-over in meiosis and mitosis. Nature 140: 362. Friesen, H. (1937b). Mechanism of crossing-over in males of Pros ophila  melanogaster. J. Genetics 35: 141-150. Fristrom, D. (1969). Cellular degeneration in the production of some mutant phenotypes in Drosophila melanogaster. Moi. Gen. Genet. 103: 363-379. Gall, J. G., Cohen, E. H., and Polan, M. L. (1971). Repetitive DNA sequences in Drosophila. Chromosoma (Berlin) j}3: 319-344. Garcia-Bellido, A., Ripoll, P., and Morata, G. (1973). Developmental compartmentalisation of the wing disk of Drosophila. Nature N.J.B. 245: 251-253. Gehring, W. J. (1966). Uebertragung und Aenderung der Determina-tionsqualita'ten in Antennen-Scheiben-Kulturen von Drosophila  melanogaster. J. Embryol. Exp. Morphol. _15: 77-111. Graubard, M. A. (1934). Temperature effect on interference and crossing over. Genetics 19: 83-94. Green, M. M. (1959). Double crossing over or gene conversion at the white locus in Drosophila melanogaster? Genetics 45: 15-18. 210 Green, M. M. (1960) . Apparent double crossing over in a short genetic interval in Drosophila melanogas ter. Nature 126: 990-991. Green, M. M. (1975). Conversion as a possible mechanism of high coincidence values in the centromere region of Drosophila. Moi. Gen. Genet. 139: 57-66. Green, M. M., and Oliver, C. P. (1940). The action of certain mutants upon the penetrance of heterozygous vestigial wing in Drosophila melanogaster. Genetics 2_5: 584-592. Grell, E. H. (1962). The dose effect of ma-l+ and r y + on xanthine dehydrogenase activity in Drosophila melanogaster. Genetics 47: 956. G r i g l i a t t i , T., and Suzuki, D. T. (1970). Temperature-sensitive mutations in Drosophila melanogaster. V. A mutation affecting concentrations of pteridines. Proc. Natl. Acad. Sci. (U.S.A.) 67: 1101-1108. G r i g l i a t t i , T. A., White, B. N., Tener, G. M., Kaufman, T. C, Holden, J. H., and Suzuki, D. T. (1974). Studies on the tRNA genes of Drosophila. Cold Spring Harbor Symp. Quant. Biol. 38: . 461-474. Hannah-Alava, A. (1968). Induced crossing-over in the presterile broods of Drosophila melanogaster males. Genetica 39_: 94-152. Harnly, M. H. (1936). The temperature-effective periods and the growth curves for length and area of the vestigial wings of Drosophila  melanogaster. Genetics 2JL: 84-103. Hartwell, L. (1974). Saccharomyces cerevisiae c e l l cycle. Bacteriol. Rev. 38: 164-198. Hawthorne, D. C, and Mortimer, R. K. (1960). Chromosome mapping in Saccharomyces: centromere linked genes. Genetics 45: 1085-1110. Heitz, E. (1933). Die somatische Heteropyknose bei Drosophila melano-gaster und ihre genetische Bedenteung. Z. Zellforsch 20: 237-287. Herskowitz, I. H., and Abrahamson, S. (1957). Induced changes in female germ cells of Drosophila. IV. Dependence of induced crossover-like exchanges in oocytes and oogonia upon X-ray intensity. Genetics 42: 444-453. Hexter, W. M. (1958). On the nature of the garnet locus in Drosophila  melanogaster. Proc. Natl. Acad. Sci. (U.S.A.) 44: 768-771. Hilli k e r , A. J. (1975). Genetic analysis of the proximal hetero-chromatin of chromosome 2 of Drosophila melanogaster. Ph.D. thesis, University of British Columbia. 211 Hil l i k e r , A. J. (1976). Genetic analysis of the centromeric hetero-chromatin of chromosome 2 of Drosophila melanogas ter: defi-ciency mapping of EMS-induced lethal complementation groups. Genetics &3: 765-782. Hil l i k e r , A. J., and Holm, D. G. (1975). Genetic analysis of the proximal region of chromosome 2 of Drosophila melanogas ter. I. Detachment products of compound autosomes. Genetics 81: 705-721. Hodgetts, R. B. (1975). The response of dopa decarboxylase activity to variations in gene dosage in Drosophila: a possible location of the structural gene. Genetics 7_9: 45-54. Holden, J. J., and Suzuki, D. T. (1973). Temperature-sensitive muta-tions in Drosophila melanogaster. XII. The genetic and developmental effects of dominant lethals on chromosome 3. Genetics 7_3: 445-458. Holm, D. G., Baldwin, M., Duck, P., and Chovnick, A. (1969). The use of compound autosomes to determine the relative centromeric position of chromosome 3. Dros. Inf. Serv. 44: 112. Howe, H. B. (1956). Crossing over and nuclear passing in Neurospora  crassa. Genetics 41: 610-622. Howells, A. J. (1972). Levels of RNA and DNA in Drosophila melano-gas ter at different stages of development: a comparison between one bobbed and two phenotypically non-bobbed stocks. Bioch.i Genetics 6: 217-230. Huang, S. L., and Baker, B. S. (1975). The mutability of the Minute lo c i of Drosophila melanogaster with Ethyl methanesulfonate. Mutation Res. 34: 407-414. Hurst, D. D., Fogel, S., and Mortimer, R. K. (1972). Conversion-associated recombination in yeast. Proc. Natl. Acad. Sci. (U.S.A.) 69: 101-105. Ilan, J. (1968). Amino acyl incorporation and aminoacyl transfer in an insect pupal system. J. Biol. Chem. 243: 5859-5866. Ives, P. T., and Fink, G. R. (1962). Comparison of translocation and crossover chromosomes produced by y irradiation of Drosophila males. Genetics 47: 963. Jacobsen, K. B. (1971). Role of an isoacceptor transfer ribonucleic acid as an enzyme inhibitor: effect on tryptophan pyrrolase of Drosophila. Nature N. B. 231: 17-19. Judd, B. H., Shen, M. W., and Kaufman, T. C. (1972). The anatomy and function of a segment of the X chromosome of Drosophila melano-gaster. Genetics 7_1: 139-156. 212 Kaufman, T. C, Tasaka, S. E., and Suzuki, D. T. (1973). The interac-tion of two complex l o c i , zeste and bithorax in Drosophila  melanogaster. Genetics 7_5: 299-321. Lambertsson, A. (1975a). The ribosomal proteins of Drosophila melano-gaster. IV. Characterization by two-dimensional gel electro-phoresis of the ribosomal proteins from nine post-embryonic developmental stages. Moi. Gen. Genet. 139: 133-144. Lambertsson, A. (1975b). The ribosomal proteins of Drosophila melano-gaster . V. Analysis by two-dimensional gel electrophoresis of the ribosomal proteins of the temperature-sensitive lethal a l l e l e of suppressor of forked, l ( l ) s u ( f ) t s ^ 7 g : a putative ribosomal protein mutant. Moi. Gen. Genet. 139: 145-156, Lifschytz, E., and Falk, R. (1969). A system for screening of rare events in genes of Drosophila melanogaster. Genetics 63: 353-352. Lindsley, D. L., and Grell, E. H. (1968). "Genetic Variations of Drosophila melanogaster." Carnegie Inst. Wash. Publ. 627. Lindsley, D. L., Sandler, L., Baker, B. S. B., Carpenter, A. T. C, Denell, R. E., Hall, J. C, Jacobs, P. A., Miklos, G. L. G., Davis, B. K., Gethmann, R. C , Hardy, R. W., Hessler, A., Miller, S. M., Nozawa, H., Parry, D. M., and Gould-Somero, M. (1972). Segmental aneuploidy and the genetic gross structure of the Drosophila genome. Genetics 7JL: 157-184. Lucchesi, J. C, and Suzuki, D. T. (1968). The interchromosomal control of recombination. Annu. Rev. Genet. 2_:. 53-86. Merriam, J. R., and Garcia-Bellido, A. (1969). Linkage data, D. melanogaster. Dros. Inf. Serv. 44: 51. Mglinets, V. A. (1972). Irradiation-induced crossing-over in Drosophila males. Cytological investigation of crossovers. Genetika 8: 193-201. Mglinets, V. A. (1973). Cytological analysis of crossovers induced by irradiation in females of Drosophila melanogaster. Genetika 2: 323-328. Mglinets, V. A. (1975). Time of gene activity in the ontogenesis of Drosophila. I. Temperature-sensitive period for the spineless effect in aristapedia mutants. Ontogenez j5: 416-421. Miller, A. (1950). The internal anatomy and histology of the imago of Drosophila melanogas ter, In "Biology of Drosophila" (M. Demerec, ed.), pp. 420-534. John Wiley and Sons, New York. Morata, G., and Ripoll, P. (1975). Minutes: mutants of Drosophila autonomously affecting c e l l division rate. Devel. Biol. 42: 211-221. 213 Morgan, T. H., Sturtevant, A. H., and Bridges, C. B. (1925). The genetics of Drosophila. Bibl. genet. 2: 1-262. Muller, H. J. (1932). Further studies on the nature and causes of gene mutations. Proc. 6th Int. Cong. Genet. 1: 213-255. Muller, H. J. (1954). The nature of the genetic effects produced by radiation, In "Radiation Biology" (A. Hollander, ed.) v 1, pp. 351-474. McGraw-Hill, New York. Muller, H. J. (1958). Pseudo-crossingover near centromere of the third chromosome induced in late oocytes by X-rays. Dros. Inf. Serv. 32: 140-141. Muller, H. J., and Painter, T. S. (1932). The differentiation of the sex chromosomes of Drosophila into genetically active and inert regions. Mu. indukt. Abstamm.-vererb. Lehre 62_: 316-365. Muller, H. J., Raffel, D., Gershenson, S. M., and Prokofyeva-Belgovskaya, A. A. (1937). A further analysis of l o c i in the so-called "inert region" of the X chromosome of Drosophila. Genetics 22.: 87-93. Nomura, Y. (1973). Irreversible inactivation at high temperature of temperature-sensitive mutant tRNA^^ In vivo. Nature N. B. 2-42: 12-14. Nothiger, R. (1972). The larval development of imaginal disks, In "Results and Problems in Cell Differentiation: The Biology of Imaginal Disks" (H. Ursprung and R. Nothiger, eds.) v 5, pp. 1-34. Springer-Verlag, New York, Heidelberg and Berlin. O l i v i e r i , G., and O l i v i e r i , A. (1964). Evidence for the two-hit nature of Xfray induced crossing over in the centromeric region of Drosophila males. Mutation Res. 1: 279-295. Ouweneel, W. J. (1969). Morphology and development of loboid-ophthalmoptera, a homeotic strain in Drosophila melanogaster. wllhelm Roux Arch. Entwicklungsmech. Organ. 178: 247-275. Painter, T. S. (1935). The morphology of the third chromosome in the salivary gland of Drosophila melanogaster and a new cytological map of this element. Genetics 2_0: 301-326. Patterson, J. T., and Suche, M. L. (1934). Crossing over induced by X-rays in Drosophila males. Genetics 1_9: 223-236. Peacock, W. J., Brutlag, D., Goldring, E., Appels, R., Hinton, C. S., and Lindsley, D. L. (1974). The organization of highly repetitive DNA sequences in Drosophila melanogas ter chromo-somes. Cold Spring Harbor Symp. Quant. Biol. _28: 405-416. 214 Poodry, C. A., Hall, L., and Suzuki, D. T. (1973). Developmental pro-perties of shiberp*-^: a pleiotropic mutation affecting larval and adult locomotion and development. Devel. Biol. 32: 373-386. Postlethwait, J. H., and Schneiderman, H. A. (1973). Developmental genetics of Drosophila imaginal discs. Annu. Rev. Genet. 7_: 381-433. Pritchard, R. H. (1960). Localized negative interference and i t s bearing on models of gene recombination. Genetical Res. 1: 1-24. Puro, J. (1966). Mutational response of the premeiotic germ-cell stages of adult Drosophila melanogaster males to X-irradiation. Ann. Zool. Fenn. 3: 99-126. Puro, J., and Nygren, T. (1975). Mode of action of a homeotic gene in Drosophila melanogaster: localization and dosage effects of Polycomb. Hereditas 81: 237-248. Rasse-Messenguy, R., and Fink, G. R. (1973). Temperature-sensitive nonsense suppressors in yeast. Genetics 7j>: 459-464. Raytnayake, W. E. (1970). Studies on the relationship between induced crossing-over and mutation in Drosophila melanogaster. Mutation Res. 9: 71-83. Ritossa, F. M., Atwood, K. C., and Spiegelman, S. (1966a). A molecular explanation of the bobbed mutants of Drosophila as partial deficiencies of "ribosomal" DNA. Genetics .54: 819-834. Ritossa, F. M., Atwood, K. C., and Spiegelman, S. (1966b). On the redundancy of DNA complementary to amino acid transfer RNA and its absence from the nucleolar organizer region of Drosophila melanogaster. Genetics .54: 663-676. Ritossa, F. M., and Spiegelman, S. (1965). Localization of DNA comple-mentary to ribosomal RNA in the nucleolus organizer, region of Drosophila melanogaster. Proc. Natl, Acad. Sci. (U.S.A.) 53: 737-745. Roberts, P. A. (1965). Difference in the behavior of eu- and hetero-chromatin: crossing over. Nature 205: 725-726. Russell, M. A. (1974). Pattern formation in the imaginal discs of a temperature cell - l e t h a l mutant of Drosophila melanogaster. Devel. Biol. 40: 24-39. Santamaria, P., and Garcia-Bellido, A. (1975). Developmental analysis of two wing scalloping mutants ct^ and Bx^ of Drosophila  melanogaster. Wilhelm Roux Arch. Entwicklungsmech. Organ. 178: 233-245. 215 Schalet, A., and Lefevre, G., Jr. (1973). The localization of "ordinary" sex-linked genes in section 20 of the polytene X chromosome of Drosophila melanogaster. Chromosoma (Berlin) 44: 183-202. Schultz, J. (1929). The Minute reaction in the development of Drosophila melanogaster. Genetics 14: 366-419. Schultz, J., and Redfield, H. (1951). Interchromosomal effect on crossing over in Drosophila. Cold Spring Harbor Symp. Quant. Biol. 16: 175-197. Shellenbarger, D. L., and Mohler, J. D. (1975). Temperature-sensitive mutations of the Notch locus in Drosophila melanogaster. Genetics _81: 143-162. Simpson, P., and Schneiderman, H. A. (1975). Isolation of temperature-sensitive mutations blocking clone development in Drosophila  melanogaster, and the effects of a temperature-sensitive c e l l lethal mutation on pattern formation in imaginal discs. Wilhelm Roux Arch. Entwicklungsmech. Organ. 178: 247-275. Sinclair, D. A. (1975). Crossing over between closely linked markers spanning the centromere of chromosome 3 in Drosophila melano-gaster . Genetical Res. 11: 173-185. Smith, J. D., Barnett, L., Brenner, S., and Russell, R. L. (1970). More mutant tyrosine transfer ribonucleic acids. J. Moi. Biol. 54: 1-14. Smith, P. D., Finnerty, V. G., and Chovnick, A. (1970). Gene conver-sion in Drosophila: non-reciprocal events at the maroon-like cistron. Nature 228: 442-444. S^gaard, B. (1974). The localization of eciferum l o c i in barley. III. Three point test of genes on chromosome I in barley. Hereditas 76: 41-47. Spreij, T. E. (1971). Cell death during the development of the imaginal disks of Calliphora erythrocephala. Neth. J. Zool. 2_1: 221-264. Stadler, D. R. (1956). Double crossing over in Neurospora. Genetics 41: 623-630. Steffensen, D. M. (1973). Mapping genes for the ribosomal proteins of Drosophila. Nature N. B..244: 231-234. Stern, C. (1929). Uber die additive Wirckung multipler A l l e l e . Biolog. Zent. 49: 261-290. Stern, C, and Tokunaga, C. (1971). On c e l l lethals in Drosophila. Proc. Natl. Acad. Sci. (U.S.A.) 68: 329-331. 216 Stevens, W. L. (1936). The analysis of interference. J. Genetics 32: 51-64. Stewart, B. R., and Merriam, J. R. (1974). Segmental aneuploidy and enzyme activity as a method for cytogenetic localization in Drosophila melanogaster. Genetics 76: 301-309. Strickland, W. N. (1961). Tetrad analysis of short chromosome regions of Neurospora crassa. Genetics 46: 1125-1141. Sturtevant, A. H. (1951). A map of the fourth chromosome in Drosophila  melanogas ter based on crossing over in triploid females. Proc. Natl. Acad. Sci. (U.S.A.)37: 405-407. Sussman, M. (1970). Model for quantitative and qualitative control of mRNA translation in eukaryotes. Nature 225: 1245-1246. Suzuki, D. T. (1970). Temperature-sensitive mutations i n Drosophila  melanogaster. Science (Washington) 170: 695-706. Suzuki, D. T. (1974a). Genetic analysis of crossing over and its rela-tion to chromosome structure and function in Drosophila  melanogaster. Oregon State University Genetics Lectures 3^: 7-32. Suzuki, D. T. (1974b). Temperature-sensitive Mutations in Drosophila  melanogaster, In "Handbook of Genetics" (R. C. King, ed.) p. 653-668. Plenum, New York. Suzuki, D. T., B a i l l i e , D., and Parry, D. (1966). The origin of multiple crossing over chromatids in short genetic intervals in Drosophila melanogaster. Genetics 5_4: 1359-1370. Suzuki, D. T., Kaufman, T. C , and Falk, D. R. (1976). Conditionally expressed mutations in Drosophila melanogaster, In "The Genetics and Biology of Drosophila" (M. Ashburner and E. Novitski, eds.) v a, pp. 207-263. Academic Press, London, New York, and San Francisco. Suzuki, D. T., and Parry, D. M. (1964). Crossing over near the centro-mere of chromosome 3 in Drosophila melanogaster females. Genetics 50: 1427-1432. Suzuki, D. T., and Procunier, D. (1969). Temperature-sensitive muta-tions in Drosophila melanogaster. III. Dominant lethals and semilethals on chromosome 2. Proc. Natl. Acad. Sci. (U.S.A.) 62: 369-376. Tarasoff, M., and Suzuki, D. T. (1970). Temperature-sensitive mutations in Drosophila melanogaster. VI. Temperature effects on devel-opment of sex-linked recessive lethals. Devel. Biol. 23: 492-509. 217 Tasaka, S. E., and Suzuki, D. T. (1973). Temperature-sensitive muta-tions in Drosophila melanogaster. XVII. The genetic proper-ties of heat- and cold-sensitive lethals on chromosome 3. Genetics 74: 509-520. Thompson, P. E. (1963a). Centric pairing and crossing over in Drosophila melanogaster. Genetics 48: 697-701. Thompson, P. E. (1963b). Evidence on the basis of the centromere effect in the large autosomes of Drosophila melanogaster. Genetics 49: 761-769. Tokunaga, C. (1962). Cell lineage and differentiation on the male fore-leg of Drosophila melanogaster. Devel. Biol. _+: 489-516. Vaslet, C. A., and Berger, E. M. (1976). Electrophoretic identity among r-proteins in sibling and mutant species of D. melanogaster. Genetics _83: 79s. Vogt, M. (1947). Zur labilen Determination der Imaginalscheiben von Drosophila. I II. Analyse der Manifestierungsbedingungen sowie der Wirkungsweise der zu Antennen und Palpusrerdoppel-^ ungen fuhrenden Genmutation Deformed-recessive-Luers (Dfd——) . Biolog. Zent. _66: 81-105. Weinstein, A. (1936). The theory of multiple-strand crossing over. Genetics 2_L: 155-199. Welshons, W. J. (1955). A comparative study of crossing over in attached-X chromosomes of Drosophila melanogaster. Genetics 40: 918-936. White, B. N. (1974). An analysis of tRNAs in five Minutes and two suppressors. Dros. Inf. Serv. 51_: 58-59. White, B. N., Tener, G. M., Holden, J., and Suzuki, D. T. (1973). Analysis of tRNA's during the development of Drosophila. Devel. Biol. 33: 185-195. Whittinghill, M. (1937). Induced crossing over in Drosophila males and its probable nature. Genetics 22: 114-129. Whittinghill, M. (1955). Crossover v a r i a b i l i t y and induced crossing over. J. of Ce l l . Comp. Physiol. 45: 189-220. Williamson, J. H., Parker, D. R., and Manchester, W. G. (1970). X-ray-induced recombination in the fourth chromosome of Drosophila melanogaster females. I. Kinetics and brood patterns. Mutation Res. 9_: 287-297. Wright, T. R. F. (1973). The recovery, penetrance, and pleiotropy of X-linked, cold sensitive mutants in Drosophila. Moi. Gen. Genet. 122: 101-118. APPENDIX 1 Crossover Data and Estimates of Crossing Over a) Crossover Data from Irradiated "Males and Controls Brcod Single Crossover Progeny Treatment Pa rental Genotypes Region 1 Region 3 of Male Parents Interval (davs) + in 2 ... p s r i eg Ki p e st 2 „. p s in r i eg Ki p e 2 _ _ . p s eg Ki p e Control 0-3 3-6 1,565 665 1,584 623 Mo Radiation 6-9 9-12 12-15 638 822 384 707 858 429 Totals 4,074 4,201 Expt I 0-3 3-7 1,518 552 1,585 544 1000R 7-11 11-16 16-21 918 3,699 5,391 920 3,710 8,293 1 3 4 1 2 1 1 Totals 21-26 1,678 13,756 1,716 13,768 ~~8 ~3 ~2 Expt II 0-3 3-7 140 97 141 109 2000R 7-11 11-16 16-21 589 309 1,693 549 303 1,703 2 1 -21-26 732 760 Totals 3,560 3,565 ~2 1 Expt III 0-3 3-6 1,142 628 1,11.3 680 I 3000R 6-12 12-17 17-22 805 3,614 3,881 848 3,586 3,981 1 1 1 Totals 10,070 10,208 ~~3 APPENDIX la (continued) Treatment of Male Brood Interval Single Crossover Progeny Doubles Region 4 Region 5 Region 6 3,4 5,6 o Parents (days) st in r i eg Ki D P e S st in r i 2 eg Ki P s P e 2 D st in r i eg Ki p s e ^ P eg_ p _ Control 0-3 3-6 1 No Radiation 6-9 9-12 12-15 3 Totals ~3 "T Expt I 0-3 3-7 1 1000R 7-11 11-16 16-21 6 2 1 1 1 2 1 1 1 21-26 6 5 2 1 Totals 6 13 4 5 ~T ~2 Expt II 0-3 3-7 2000R 7-11 11-16 16-21 21-26 1 7 1 1 3 Totals ~~8 5 Expt III 0-3 3-6 1 3000R 6-12 12-17 5 5 5 12 2 1 2 1 2 2 17-22 8 26 Totals 18 43 ~3 ~2 ~2 APPENDIX 1 (continued) b) Summary of Male Crossover Data and Estimation of Crossing Over Between st and p P. Treatment of Number of Number of Crossovers Experiment Male Parents Parentals p st to p P ^ s p to c Totals Control 8,275 3 1 8,279 Expt I 27,524 41 3 27,568 Expt II 7,125 16 - 7,141 Expt III 20,278 73 3 20,354 Experiment Totals 54,927 130 6 55,063 Control 3,835 3 -Expt III* Estimations of Crossing Over 10,796 Control Expt III = 0.08 = 0.59 64 4 Percent (3838 Progeny) Percent 10,864 * Progeny of fourteen cultures, days 7 to 22 APPENDIX 2 Summary of Results of 24-Hour Expi osures of Q-III Homozygotes to 29°C at Specific Times During Development Hours Number Corrected Percent Expression of Various Phenotypes of Surviving Progeny of of Percent Ommatidial Antennal Antennal Thick Palp Pulse Adults Vi a b i l i t y Deficiencies Deficiencies Duplications Aristae Deficiencies 0-24 14(300)* 58.3 0 0 0 0 0 12-36 12(242) 63.2 0 0 0 0 0 24-48 28(300) 116.0 0 0 0 0 0 36-60 22(285) 95.7 0 0 0 0 0 48-72 25(515) 60.6 0 0 0 4.0 0 60-84 34(600) 70.8 26.1 8.8 0 11.8 0 72-96 37(504) 93.8 53.3 0 0 13.5 0 84-108 12(500) 30.0 72.3 0 8.3 25.0 8.3 96-120 35(500) 87.5 83.1 0 " 0 22.9 2.9 108-132 34(500) 85.0 58.3 8.8 8.8 58.8 2.9 120-144 20(200) 125.0 20.0 0 0 50.0 5.0 132-156 17(225) 106.0 5.9 0 0 5.9 5.9 144-168 13(250) 65.0 0 0 0 0 0 156-180 42(300) 120.0 0 0 0 0 0 168-192 29(300) 120.8 0 0 0 0 0 180-204 27(200) 130.0 0 0 0 0 0 192-216 14(300) 58.3 0 0 0 0 0 204-228 20(300) 47.6 0 0 0 0 0 228-252 6(300) 25.0 0 0 0 0 0 252-276 16(360) 55.6 0 0 0 0 0 276-300 13(342) 47.5 7.6 0 0 0 0 300-324 8(360) 27.7 0 0 • 0 0 0 324-348 22(300) 91.7 0 0 0 0 0 * Number in parentheses = Number of Eggs Shifted APPENDIX 2 (continued) Percent Expression of Various Phenotypes of Surviving Progeny Hours Ocellar Small Sex Thoracic Wing Misrotated of Duplications Thoracic Comb Pattern Vein Male Leg Etched Pulse or Deficiencies Macrochaetae Gaps Disruptions Disruptions Terminalia Deformities Tergites 0-24 0 0 0 0 0 0 0 0 12-36 0 0 0 0 0 0 0 0 24-48 0 0 0 0 0 0 0 0 36-60 0 0 0 0 0 0 0 0 48-72 0 0 0 12.0 0 4.0 0 0 60-84 0 0 0 32.4 0 10.0 0 0 72-96 5.4 0 0 21.6 0 20.0 0 0 84-108 0 0 0 41.7 10.0 50.0 0 0 96-120 8.6 0 0 94.3- 39.1 33.3 0 0 108-132 14.7 0 0 100.0 5.0 61.5 8.8 0 120-144 0 0. 0 65.0 0 37.5 30.0 0 132-156 0 0 0 23.5 0 50.0 0 0 144-168 0 0 0 33.3 0 33.3 0 0 156-180 0 0 0 0 0 0 0 0 168-192 0 0 0 0 0 0 0 0 180-204 0 0 0 0 0 26.6 0 0 192-216 0 0 0 0 0 25.0 0 0 204-228 0 0 0 0 0 0 0 5.0 228-252 0 16.5 100.0 (males) 0 0 0 0 16.7 252-276 0 100.0 100.0 0 0 0 0 0 276-300 0 7.7 0 23.1 0 0 0 0 300-324 0 0 0 0 0 0 0 0 324-348 0 0 0 0 0 0 0 0 APPENDIX 3 Summary of Results of 48-Hour Exposures of Q-III Heterozygotes (TMI/QIII) to 29°C at Specific Times During Development Hours Number Corrected Percent Expression of Various Phenotypes of of Percent Roughened Ommatidial Antennal Pulse Adults Viab i l i t y Eyes Deficiencies Deficiencies 0-48 191 86.6 0 0 0 24-72 152 82.6 0 0 0 48-96 93 42.9 98.8 53.1 0 72-120 308 53.0 98.9 63.3 1.0 96-144 244 90.2 98.4 3.2 0 120-168 58 85.3 16.4 11.5 0 144-192 30 21.8 0 0 0 168-216 26 28.0 0 0 0 192-240 129 70.1 0 0 0 216-264 204 100.0 0 0 0 204-252 265 123.8 0 0 0 252-300 300 56.2 0 0 0 300-348 270 126.0 0 0 0 348-396 304 120.0 0 0 0 APPENDIX 3 (continued) Percent Expression of Various Phenotypes Hours Thoracic Small of Antennal Wing Pattern Thoracic Pulse Duplications Duplications Disruptions Macrochaetae 0-48 0 0 0 0 24-72 0 0.7 0.7 0 48-96 0 0 ' 0 0 72-120 0.7 0 0 0 96-144 0.4 0 0 0 120-168 0 0 0 0 144-192 0 0 0 20.0 168-216 0 0 0 100.0 192-240 0 0 0 98.4 216-264 0 0 0 62.0 204-252 0 0 0 25.0 252-300 0 0 0 0 300-348 0 0 0 0 348-396 0 0 0 0 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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


Related Items