Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Autosomal products of meiosis arising from radiation-induced interchange in female Drosophila Melanogaster Gibson, William Glen 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

Download

Media
831-UBC_1977_A1 G52.pdf [ 7.57MB ]
Metadata
JSON: 831-1.0094205.json
JSON-LD: 831-1.0094205-ld.json
RDF/XML (Pretty): 831-1.0094205-rdf.xml
RDF/JSON: 831-1.0094205-rdf.json
Turtle: 831-1.0094205-turtle.txt
N-Triples: 831-1.0094205-rdf-ntriples.txt
Original Record: 831-1.0094205-source.json
Full Text
831-1.0094205-fulltext.txt
Citation
831-1.0094205.ris

Full Text

AUTOSOMAL PRODUCTS OF MEIOSIS ARISING FROM RADIATION-INDUCED INTERCHANGE IN FEMALE DROSOPHILA MELANOGASTER by / WILLIAM GLEN GIBSON B.A., University of Saskatchewan, 1966 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY THE FACULTY OF GRADUATE STUDIES In the Department of Zoology We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA July, 1977 © William Glen Gibson, 1977 In p r e s e n t i n g t h i s t h e s i s in p a r t i a l f u l f i l m e n t o f the r e q u i r e m e n t s f o r an advanced degree at the U n i v e r s i t y o f B r i t i s h Co lumb ia , I a g ree that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s tudy . I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y purposes may be g r a n t e d by the Head o f my Department o r by h i s r e p r e s e n t a t i v e s . It i s u n d e r s t o o d that c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i thou t my w r i t t e n p e r m i s s i o n . Depa rtment The U n i v e r s i t y o f B r i t i s h Co lumbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 i ABSTRACT The present study was initiated with the view of achieving two goals: 1) to establish a suitable genetic assay system for measuring the frequency of spontaneous and induced structural and numerical aberrations of autosomes during meiosis in females and 2) to provide a better understanding of the mechanisms responsible for the production of the aberrant classes recovered. By selective exclusion of a l l regular meiotic products this system enabled the recovery of large numbers of aberrant products. The multiplier system served as an internal dosimeter and provided an estimate of the population size from which the aberrancies arose which in turn provided a measure of the frequen-cies of each event. The four different classes of exceptional meiotic pro-ducts were named according to the source or the structural nature of the chromosomes: reductional nondisjunction as "matroclinous"; equational nondis-junction as "equationals"; loss of chromosome 2_ as "patroclinous"; and the attachment of homologous arms as "compounds". The results suggest that two main factors affect the recovery of induced aberrations: of most importance is isosequentiality and of lesser importance is genetic background. The three classes of simultaneously recovered progeny (excluding equational nondisjunctions) arise from a common mechanism of induc-tion; a mechanism which also accounts for free arm formation. The location of the breaks, the position of the chromatids and the method of reconstitution determine the type of aberration produced. The reconstitution of these breaks in aberrant ways are referred to as interchanges. Furthermore, i t would appear that the reconstitutions are restricted in that euchromatic breaks attach to euchromatic breaks and heterochromatic to heterochromatic. Inter-changes resulting from breaks on opposite sides of the centromeres of homolo-gues result in the formation of non-sister compound chromosomes and from breaks on opposite sides of the centromeres of sister chromatids result in the forma-tion of sister compound chromosomes. The interchange, i f between heterologues, could lead to the nondisjunction of a pair of chromosomes and be recovered, as in the present study, as matroclinous progeny. The reciprocal product of the interchange between heterologues would produce an equal number of nullo eggs observed as patroclinous progeny, but i f the dyad so formed is heteromorphic, i.e. chromatids of different length, i t would result in the greater recovery of patroclinous progeny because of the preferential inclusion of the shorter chromatid. The evidence for interchange mediated aberrations is provided by the recovery of free arms of chromosome 2_. Experimental support for these events is provided by the unequivocal identification of the centromeres in-volved, which, as in this study, is made possible through the use of metacentric autosomes. i i i TABLE OF CONTENTS PAGE ABSTRACT i TABLE OF CONTENTS i i i LIST OF TABLES v i LIST OF FIGURES v i i i ACKNOWLEDGEMENT x CHAPTER I. AN ASSAY SYSTEM TO RECOVER ABERRANT CHROMOSOME-2^  PRODUCTS OF MEIOSIS Introduction . 2 Materials And Methods . 8 Description Of Chromosomes 8 Radiation Treatment 8 Mating Procedure For Experimental And Dosimetry Test 9 Recovery Of Chromosomal Aberrations 10 Multiplier System And Internal Dosimeter 10 Results 16 Frequencies Of Chromosomal Aberrations 27 Internal Dosimeter 28 Discussion 30 Statistics 30 Internal Analysis 31 Internal Dosimeter 32 Chromosomal Aberrations 32 CHAPTER II. ANALYSIS OF RADIATION INDUCED COMPOUND CHROMOSOMES Introduction 34 Materials And Methods 43 Induction And Recovery 43 iv PAGE Analysis Of Compound Autosomes Generated In Females Carrying The Genetic Marker Analysis Of Compound Autosomes Generated In Females Carrying The Genetic Markers It pk cn 46 Results 47 Discussion . . . . . 57 CHAPTER III. ANALYSIS OF RADIATION INDUCED MATROCLINOUS AND PATROCLINOUS PROGENY Introduction 63 The Sex Chromosomes 63 The Autosomes 65 Equational Nondisjunction 67 Materials And Methods 69 Results 70 Spontaneous And Radiation Induced Matroclinous Progeny 70 Spontaneous And Radiation Induced Patroclinous Progeny . 74 Equational Matroclinous Progeny 76 Discussion 79 Spontaneous And Radiation Induced Nondisjunction 79 Quasibivalent Mediated Interchange Resulting In Autosomal Nondisjunction 85 Spontaneous And Radiation Induced Chromosome Loss 91 Proposed Mechanism For the Induction Of Chromosome-2 Loss 97 V PAGE Summary Of Nondisjunction And Chromosome Loss 98 Equational Nondisjunction 99 CHAPTER IV. AUTOSOMAL FREE-ARM FORMATION AS SUPPORT OF A UNIFYING THEORY FOR AUTOSOMAL PRODUCTS OF MEIOSIS ARISING FROM RADIATION-INDUCED INTER-CHANGE Introduction 102 Materials And Methods 112 Results 116 Discussion . 126 General Discussion Of Interchange 134 Intrachrbmosomal Breaks 135 Interhomologue Breaks 136 Interchromosomal Breaks 137 Interchanges 137 CHAPTER V. AN ASSAY FOR THE PRODUCTS OF CHEMICAL MUTAGENISIS Introduction 139 Materials And Methods 145 Results 146 Discussion . 153 BIBLIOGRAPHY 156 v i LIST OF TABLES TABLE PAGE I. Percent hatch of eggs recovered from C(2L)P,b;C(2R)P,px females mated to C(2L)VH2,lt;C(2R)P,px males. 18 II. Number of eggs laid by In(2L + 2R)Cy/ap W O W females when mated to C(2L)VH2,lt;C(2R)P,px or Basc/Y (standard second) males. 19 III. Multiplier values obtained for and corresponding estimated total number of progeny expected in each of the experiments involving standard females crossed to compound-2_ males. 20 IV. Number of progeny recovered as exceptional chromosome-^ products of meiosis and the number of X-linked recessive lethals obtained in corresponding multiplier tests. 24 V. Estimated percent recovery of the four classes of progeny that arose from aberrant chromosome-2_ products of meiosis. 25 VI. Comparisons, within and between treatment groups, of the percentage each class contributes to the total number of exceptional progeny recovered from each experiment. 26 VII. Estimated percent recovery of compound-^ chromosomes. 48 VIII. Comparison of the recovery of compound-2^  l e f t vs. compound-2^  right chromosomes. 50 IX. Comparison of sister vs. non-sister compound-chromosome formation for the right arm of chromosome 2. 51 X. Comparison of sister vs. non-sister compound-chromosome formation for the l e f t arm of chromosome 2. 53 v i i TABLE PAGE XI. Genetic analysis of newly generated compound-2_ chromosomes for the presence of duplications. 56 XII. Estimated percent recovery of matroclinous progeny. 71 XIII. Estimated percent recovery of patroclinous progeny. 75 XIV. Estimated percent recovery of equational (second division) nondisjunctions. 77 XV. Number of induced F(2L)'s and F(2R)'s recovered from In(2LR)SMl,Cy/b pr cn or In(2LR)SMl,Cy/Df(2R)M-S210 females mated to C(2L)VH2,1t;F(2R)bw/F(2R)bw males. 117 XVI. Genetic and cytological properties of induced free-2L chromosomes. 118 XVII. Genetic and cytological properties of induced free-2R chromosomes not carrying the Dp(2L)lt +. 124 XVIII. Genetic and cytological properties of induced free-2R chromosomes carrying Dp(2L)lt +. 125 XIX. Number of progeny recovered as exceptional chromosome-2^ products of meiosis and the number of X-linked recessive lethals obtained in corresponding multiplier tests. 147 XX. Estimated percent recovery of the four classes of progeny that arose from aberrant chromosome-^ products of meiosis. 148 XXI. Percent matroclinous progeny based on an estimated total number of progeny produced in, and corresponding mean multiplier value obtained for, each of the experiments. 149 XXII. Percent patroclinous progeny based on an estimated total number of progeny produced in, and corresponding mean multiplier value obtained for, each of the experiments. 151 LIST OF FIGURES FIGURE CHAPTER I 1. The experimental mating procedure that was followed to yield exceptional progeny and to provide an estimate, via the multi-plier system, of the population size from which the exceptions arose. 2. Males carrying compound-^ 2 chromosomes produce four classes of sperm in approximately equal frequencies. The only recoverable female gametes are either those resulting from irregular behaviour of the chromosomes during meiosis or those containing a newly generated compound-2^  autosome, thereby producing the four classes of progeny. CHAPTER II 3. Compound-chromosome formation i s thought to arise from simul-taneous breaks on either side of the centromere. The broken ends repair in such a way that the two l e f t or right arms are attached to a common centromere. CHAPTER III 4. An example of an interchange forming a quasibivalent between a second and a fourth chromosome. 5. An interchange between chromosome 2_ and chromosome 4_ produces a quasibivalent which causes the two heterologues to segregate at meiosis I. IX FIGURE PAGE CHAPTER IV 6. An interchange between chromosome 2_ and chromosome 4_ produces a quasibivalent which causes the two heterologues to segregate at meiosis I. 104 7. The segregation of free-2R chromosomes is shown for males. The two free-2R chromosomes segregate while the compound chromosome moves at random. 108 8. The F(2R) males produce two classes of sperm. Consequently, the only recoverable progeny, when mated to females with normal chromosomes, are those from new free arms generated through interchanges. 110 9. Males bearing the 2L-X chromosome produce two classes of sperm recoverable when mated to females carrying standard second chromosomes. 121 X ACKNOWLEDGEMENT I would like to thank Dr. D.G. Holm for his support, interest and advice. I would also like to thank Dr. A.J. H i l l i k e r for the many discussions we held and for his assistance in the cytology done as part of this thesis. Finally, I would like to thank the many fine people, who have been involved in Dr. Holm's laboratory, for their friendship and help. 1 CHAPTER I AN ASSAY SYSTEM TO RECOVER ABERRANT CHROMOSOME-2 PRODUCTS OF MEIOSIS 2 INTRODUCTION The present study was initiated with the view of achieving two goals: 1) to establish a suitable genetic assay system for measuring the frequency of spontaneous and induced structural and numerical aberrations of autosomes during meiosis in females, and 2) to provide a better understanding of the mechanisms responsible for the production of the aberrant classes recovered. The assay system developed in this study expanded beyond those previously used in that i t provides for the recovery and, in most cases, the unambiguous identification of four different classes of aberrant chromosome-2_ products of meiosis. By selective exclusion of a l l regular meiotic products, this system facili t a t e d the recovery of relatively large numbers of abnormal products. Furthermore, since the chromosome studied, chromosome 2_, is a metacentric chromosome in which the l e f t and right heterochromatic blocks have been characterized (Hilliker and Holm 1975; H i l l i k e r 1976), I was able to obtain a more detailed analysis of events leading to the production of some of the aberrant chromosomal classes. Parker (1968), Sobels (1971) and Abrahamson and Lewis (1971) have re-viewed the various methods previously used to determine genetic damage in Drosophila and the advantages of using this organism. As indicated by Parker (1968), studies may f a i l to detect induced genetic damage not because i t did not occur, but because an appropriate selective system, which avoided dominant lethality, was not available. This problem was more readily resolved with the sex chromosomes than with the autosomes and for this reason most of the assay systems thus far developed have focused on the X 3 and/or Y chromosome (Valencia 1970). Early studies with the X chromosome demonstrated i t s potential use in assay systems for detecting various types of irregular meiotic events. Bridges (1916) f i r s t demonstrated abnormal X chromosome segregation in females. These f l i e s produced female progeny that inherited both X chromo-somes from their mother, and male progeny that received the X chromosome from their father. These exceptional progeny were called matroclinous and patro-clinous respectively. Subsequent genetical and cytological analysis revealed that the mothers of these irregular progeny carried a free Y_ chromosome. Bridges (1916) believed that the Y chromosome caused the X's to nondisjoin. Thus, some eggs received both X chromosomes(disomic X) while others received none (nullo X). The former would give rise to matroclinous progeny when f e r t i l i z e d by a Y-bearing sperm, and the latter would give rise to patroclin-ous progeny when f e r t i l i z e d by an X-bearing sperm. Muller and Altenburg (1919), also using sex chromosomes bearing specific dominant genetic markers, were able to detect spontaneous recessive point mutations in females by the i n v i a b i l i t y of the hemizygous males that inherited the recessive lethals. Mavor (1924) and Muller (1927) subsequently showed that X-rays increased both the rate of nondisjunction and the frequency of recessive lethals. Although Mavor (1924) observed the increased rate of nondisjunction with X-rays, the results were not as he anticipated. Because formation of a nullo-X gamete was the reciprocal event to the formation of a disomic-X gamete, matroclinous and patroclinous progeny were expected in equal numbers. However, in his results the frequency of exceptional males was always greater than that of exceptional females. He concluded from this that patroclinous progeny were the sum of two independent events: 1) nondisjunction and 2) chromosome loss. Thus, the two phenomena have been studied concomitantly. Since the matro-4 clinous female progeny represented nondisjunctional events, the numerical difference between the patroclinous and matroclinous progeny was defined as an index of chromosome loss. The study of aneuploids, including nondisjunction and chromosome loss, has gained increased interest because surveys show 30 - 40 percent of a l l cases of spontaneous abortion in humans arise from nondisjunction and 0.4 percent of live-born children exhibit aneuploidy (Jacobs et a l . 1959; Jacobs 1971). At the same time studies were being conducted on nondisjunction, Morgan (1922) obtained results in sex linkage studies that could only be explained by the presence of two X chromosomes attached to a common centromere. Cyto-logical examination confirmed her genetic prediction. Shortly thereafter, Anderson (1925) generated an attaehed-X chromosome which carried arms hetero-zygous for a series of genetic markers. Thus, for the f i r s t time, half-tetrad analysis was available in Drosophila. In the years that followed, various arrangement of attached X's were recovered. Novitski (1954) described these various attachments and gave them the general name of compound-X chromosomes. Although much attention has been given to reductional nondisjunction, l i t t l e effort has been directed at equational, or second division, nondis-junction. The reasons for this are inherent in the nature of the X chromo-some. The X chromosome being acrocentric does not have euchromatin on both sides of the centromere. This makes genetic analysis of events in the heterochromatic regions ambiguous. Mavor (1924) and Anderson (1931) observed putative equational nondisjunctions of the X chromosome based on the recovery of progeny homozygous for a proximal marker from heterozygous female parents. They proposed that at meiosis II the two sister chromatids nondisjoined pro-5 ducing a disomic gamete. The same results could be explained by a crossover between the proximal marker and the centromere, followed by a reductional nondisjunction. The lack of a method for detecting a crossover in the proximal region of ah acrocentric chromosome makes i t impossible to differen-tiate between a crossover followed by a reductional nondisjunction and an equational nondisjunction. Similarly, the analysis of the centromere involved in a half transloca-tion, involving the X chromosome, is uncertain because of the ina b i l i t y to genetically define heterochromatic events in an acrocentric chromosome. This is of particular concern since most radiation induced events occur in the heterochromatic region (Abrahamson, Herskowitz and Muller 1954; Parker and McCrone 1958). The majority of studies in female Drosophila concerned with numerical and structural chromosomal aberrations, such as nondisjunction, both reduc-tional and equational, chromosome loss and translocations, have involved the X chromosome. In most cases i t has been d i f f i c u l t , i f not impossible, to study these events in the major autosomes (either the second or the third chromosomes). Events similar to those which occur in the X chromosome were thought to take place in the autosomes, but un t i l recently they had been un-detectable owing to the lack of a rescue system which compensated for the aneuploidy produced by the event. Although nondisjunction of the second or third chromosome might occur in females, the resulting zygote would be lethal without sperm that were genetically complementary. With the advent of compound autosomes, attempts have been made to study autosomal aberrancies. The f i r s t compound autosomes were constructed in the laboratory of E.B. Lewis (Rasmussen 1960; for a f u l l description of the pro-6 cedure followed in generating the original compound autosomes, see Holm 1976). Compound autosomes are formed by the attachment of two l e f t , or two right arms of the second (or third) chromosome to a common centromere. Thus in a compound-^ stock, the pair of standard homologous second chromosomes would be replaced by a pair of heterologues, namely a compound-2L and a compound-2R. The meiotic properties of compound chromosomes, which differ in males and females, have provided us with an important genetic tool for studying excep-tional meiotic behaviour of autosomes in Drosophila. Consistent with the theory of distributive pairing (Grell 1962, 1964) in the absence of heterologous rearrangements, the compound-left usually segre-gates from the compound-right autosome during meiosis in females (Holm, Deland and Chovnick 1967; Grell 1970; Holm and Chovnick 1975; Holm 1976). By con-trast, in males the complementary pair of compound autosomes assort independ-ently, thereby generating, in almost equal proportions, four classes of sperm (Scriba 1967, 1969; Holm, Deland and Chovnick 1967; Clark and Sobels 1973; Lutolf 1972; Holm and Chovnick 1975; Holm 1976). With specific reference to compound seconds the following four meiotic products would be obtained: sperm carrying C(2L) , sperm carrying C(2R) , disomic-2^ sperm and nullo-2_ sperm (Figure 2). The latter two classes of sperm are those which serve to rescue meiotic products from females arising from chromosome-2^ nondisjunction and loss. In addition, the f i r s t two classes of sperm provide the means of recovering newly generated C(2L) and C(2R) chromosomes. Bateman (1968) was the f i r s t to report a study on autosomal nondisjunc-tion in which the above approach was adopted. He mated C(2L);C(2R) males to irradiated females with standard chromosomes. The randomness of segregation of compound autosomes in males allowed both the recovery of nullosomic (chromosome loss) and the recovery of disomic eggs (nondisjunction). He also recovered viable progeny through the formation of new C(2L) or C(2R) chromo-somes. With one exception, these progeny derived one paternal compound and one newly generated compound from the female parent. There i s , however, one major disadvantage to this system of recovery; i t is not possible to obtain a direct measure of the frequencies of exceptional events as a l l regular eggs are inviable when f e r t i l i z e d by sperm from compound-2_ males. Consequently, previous studies have provided only relative comparisons of the recovery of exceptional classes. This f i r s t chapter contains the details of the multiplier system, a description of those experimental results that serve to support the adoption of such a system and the estimated frequencies of the four classes of aberrant products recovered from every experiment in which gamma radiation was used as the mutagenic agent and in which the exceptional events occurred spontaneously. The multiplier system served not only to estimate the size of the population from which the exceptional autosomal products were recovered but also as an internal dosimeter to provide an independent measure of the mutagenic effectiveness of the inducing agent. The internal dosimeter was obtained by carrying out an X-linked recessive lethal test on a proportion of the parental females in the multiplier system. While one might justifiably argue that such a precaution is not required in studies on radiation muta-genesis, i t s inclusion was viewed as absolutely essential to those studies in which the mutagenic properties of chemical agents are to be established. Therefore, a l l possible measurements were made on the effects of radiation to provide a wide scope of comparisons for studies on the effects of chemical mutagens. 8 MATERIALS AND METHODS Description Of Chromosomes: The study of radiation induced chromosomal aberrations was made on chromosome 1_ of Drosophila. Two chromosomes of stan-dard sequence were used: one carried the genetic m a r k e r a p W ° W (apterous-wow) and the other carried the markers, It pk cn (light, prickle and cinnibar). The above two chromosomes were examined in combination with homologues of two structurally different types: a structurally homozygous unmarked standard-second chromosome (designated +) and two homologues'that were structurally heterozygous owing to multiple inversions, In(2LR)SMl,Cy (Cy, curly) and In(2L + 2R)Cy. Both of these inverted chromosomes served as almost complete balancers for the second chromosome. The males used in a l l experiments were of two types: 1) those used in the multiplier system, which also served as the internal dosimeter, had stan-A A A j . . _ , V , • S1L 8R._ SI 8 a D SI dard seconds and the inverted X~chromosome In(l)sc sc +S,sc sc w B (sc , scute of Sinitakaya; sc^, scute-8; w&, white apricot; B, bar) which w i l l be abbreviated Base throughout the thesis, 2) those used to recover eggs that were aberrant for meiotic products of chromosome 2_ carried compound-2_ chromosomes taken from two different strains: the C(2L)VH2,lt;C(2R)P,px (It, light; px, plexus) strain or the C(2L)P,b;C(2R)P,px (b, black; px, plexus) strain. For a complete description of a l l genetic markers and rearranged stan-dard chromosomes consult Lindsley and Grell (1968). The terminology used in describing compound autosomes is discussed by Holm and Chovnick (1975) and Holm (1976). Radiation Treatment: Virgin females were collected over a 24 - 48 hour period. During the collection period, the virgin females were stored at 17°C on food without a live yeast supplement. Experimental females were etherized, placed in shell vials and treated with 2000 rads of gamma radiation from a 60 Co source in the Department of Chemistry at the University of British 60 Columbia. The dose rates of gamma radiation from the Co source were period-i c a l l y monitored by the Chemistry Department and were within an error of approximately + 3%. Control females were handled in the same manner exclud-ing treatment. Mating Procedure For Experimental And Dosimetry Test: In each experi-ment approximately 1,100 females were treated. For studies on induced chromosomal aberrations one thousand of the treated females were mated singly in shell vials to two males bearing compound-2_ chromosomes. Females with the a p W O W chromosome were mated to C(2L)VH2,It;C(2R)P,px males while those females carrying the It pk cn chromosome were mated to males carrying C(2L)P,b;C(2R)P,px. The two different strains of compound-2^  males were selected to f a c i l i t a t e the identification of newly generated compound auto-somes. The parents were l e f t in the vials for four days before being cleared. Progeny were scored daily for eleven days following the f i r s t eclosions. The remaining 100 females from each treatment were placed singly in shell vials each of which contained two Base males. A l l vials were cleared after four days. The total number of progeny recovered from each mating was" counted, and from each v i a l ten virgin females heterozygous for the genetic marker 15, were again single pair mated to two Base males. This resulted in 1,000 chromosomes being tested in each experiment. The criterion for clas s i -fication of a recessive lethal was the absence of wild type males in the F2 generation. Any series of ten vials which had 30% or more without wild type 10 males was classified as a premeiotic event and thus not included in the cal-culation of the recessive lethal frequency. An outline of the mating procedure both for the experimental crosses and for the dosimetry tests i s presented i n Figure 1. Recovery Of Chromosomal Aberrations: The C(2L);C(2R) males gave rise to four classes of sperm and each class was assumed to be produced in equal frequency: nullo-2^ disomic-2_, C(2L) bearing and C(2R) bearing sperm (Figure 2). Since these sperm were not complementary to any normal egg with regards to the second chromosome, a l l normal oocytes were lethal. The only recover-able oocytes were those containing certain aberrant meiotic products of chromosome 2_. These included: reductional or equational nondisjunction (disomic-2^ eggs) which were recovered by a nullo-2_ sperm; nullo-2_ eggs, re-covered by disomic-2^ sperm; and new compound-^ chromosomes, either l e f t or right, which were rescued by sperm carrying a complementary compound-2_ chromosome. In this study the four different classes of exceptional meiotic products rescued in the progeny were named according to the source or structural nature of the chromosomes in the following manner: reductional nondisjunction (disomic-2^ eggs) as "matroclinous", equational nondisjunction (disomic-2_ eggs homozygous for sister-chromatid markers) as "equationals"; loss of chromosome-2_ (nullo-2_ eggs) as "patroclinous"; and newly generated compounds either as C(2L) or as C(2R). This method of nomenclature avoided any inference as to the mechanism responsible for the production of irregular meiotic events. Multiplier System And Internal Dosimeter: As only exceptional meiotic products were recovered, i t was necessary to have an estimate of the size of the zygotic population from which the exceptions were obtained in order to 11 FIGURE 1 The experimental mating procedure that was followed to yield exceptional progeny and to provide an estimate, via the multiplier system, of the population size from which the exceptions arose. As an extension of the multiplier system an X-linked recessive lethal test was conducted as an internal dosimeter for the effectiveness of the mutagens. Base males carry the balancer X chromosome ,. , SZL 8R „ SZ 8 a^ IndJso so +S,sc so w B. EXPERIMENTAL MATING PROCEDURE 100 females (single pair matings) 10 virgin females from each of the above (single pair matings) t x 1,100 Treated Females I Base males h MULTIPLIER SYSTEM Base males 1,000 females (single pair matings) compound-2 males absence of a. wild type male indicates the presence of a recessive lethal INTERNAL DOSIMETER exceptional progeny: 1) new compounds 2) matroclinous 3) patroclinous 4) equational nondisjunction 13 FIGURE 2 Males carrying compoiind-2 chromosomes produce four classes of sperm in approximately equal frequencies. The only recoverable female gametes are either those resulting from irregular behaviour of the chromosomes during meiosis or those containing a newly gen-erated compound-2 autosome, thereby producing the four classes of progeny. 14 £ O E c i _ L_ n < n > o u 01 < New C(2R) > New C f 2 / j N u l l o - 2 D i s o m i c - 2 G a m e t e s f r o m Compound-.? M a l e s C(2L) C(2R) C(2LhC(2R) Nu l lo 2 >< new C(2R) >< new C(2L) >< p a t r o -c l i n ou s e — -— e m a t r o -c l i n o u s 15 determine the frequencies of a particular event. This was achieved by counting a l l progeny recovered from those matings in which (approximately 100) females were crossed to Base males for the X-Hnked recessive lethal (dosi-metry) test. The mean value of progeny from these vials gave an estimate of the average number of progeny expected from each experimental v i a l . To test the assumption that compound autosomes assort independently during meiosis, C(2L)P,b;C(2R)P,px females were single pair mated to C(2L)VH2,lt;C(2R)P,px males in shell vials containing charcoal colored medium. The parents were transferred through five 24 hour broods. Each v i a l was examined for the presence of hatched and unhatched eggs. To avoid including any unfertilized eggs, the count was initiated only when the previous v i a l had also contained hatched eggs. The use of a multiplier system implied that the number of eggs laid by the females were independent of the male used. Therefore, the number of eggs laid by In(2L + 2R)Cy/ap W O W females were determined when single pair mated to C(2L)VH2,lt;C(2R)P,px or to Basc/Y (standard second) males for four, one-day broods. 16 RESULTS The estimated frequency of the four exceptional meiotic products from females that were rescued by sperm from compound-2_ males was based on a multiplier system determined by the number of progeny. However, these deter-minations as well as comparisons among the experiments can be influenced by the following: 1) the frequency of egg hatch in the compound-2^  strain that provided the experimental male parents, 2) the influence mating with geno-typically (and chromosomally) different males had upon the rate of egg laying, and 3) the distribution and variance of progeny generated by the experimental females when crossed to standard males. Generally i t is assumed that compound autosomes assort randomly during meiosis in males, producing four classes of sperm in equal frequency. More-over, aneuploidy for compound autosomes i s known to result in embryonic lethality (Scriba 1967, 1969). It follows, therefore, that in crosses i n -volving females whose compound autosomes show regular segregation the expected frequency of hatch (viable progeny) is 25 percent. Indeed for compound-3_ strains, Holm and Chovnick (1975) found the percent hatch to differ insigni-ficantly from that expected. However, for some compound-2^  lines, slightly higher values have been obtained (Clark and Sobels 1973; Holm 1976). Since any estimate of the frequency with which aberrant meiotic products for chromosome 2_ recovered from standard females i s clearly dependent on the relative proportions of the four classes of sperm produced by compound-2^  males, I tested the frequency of egg hatch for C(2L)VH2,lt;C(2R)P,px males crossed to C(2L)P,b;C(2R)P,px females. In a previous study of egg hatch frequency, approximately 25 percent was found in compound-2 crosses involving 17 males from the C(2L)P,b;C(2R)P,px strain (Holm, unpublished data). Similarly, as shown in Table I, there was no significant difference between the hatch I observed and the theoretical expectation of 25 percent. Therefore, I have concluded that the segregation of compound-2 chromosomes was random in those male lines used throughout this study. The second consideration, namely the influence of different males on the rate of egg laying, was tested by crossing experimental females to males used in the multiplier system and males used to rescue aberrant meiotic products. It is evident, from the results recorded in Table II, that within this study the mean number of eggs laid was independent of the males used. The use of a multiplier system enabled me to estimate the population size from which exceptional events arose. In this study the mean number of f l i e s from each multiplier system varied greatly between experiments (Table III). In addition to differences in genetic background this variation was probably due to differences in such factors as the consistency of the medium (e.g. degree of hydration). As one precaution, therefore, a l l multiplier systems were run on the same batch of food as the corresponding experiment. A good-ness of f i t test confirmed that the number of progeny from the multiplier vials of each experiment were normally distributed. Although the mean number of progeny from each experiment differed, Bartlett's test for unequal sample size showed among the experiments there was homogeneity of variance. In view of the above findings, namely random segregation of compounds, normal d i s t r i -bution and homogeneity of variance, the values obtained from the multiplier system were taken as representing reasonable estimates of the mean number of progeny produced per experimental v i a l , and these values were used to calcu-late the frequencies of the various exceptional meiotic products. Moreover, TABLE I Percent hatch of eggs recovered from C(2L)P,b;C(2R)P3px females mated to C(2L)VH2ilt;C(2R)Pipx males. Number Total eggs Total Percent 95% females tested laid hatch hatch C.I. * 35 2527 660 25.31 23.02 - 27.65 * The unweighted mean percent hatch and the 95% confidence interval were determined by using the arcsin transformation values of individual results. TABLE II Number of eggs laid by In(2L + 2R)Cy/ccpWCW females when mated to C(2L)VH2ilt;C(2R)P3px or Baso/Y (stan-dard second) males. Brood _ „ , _ Total mean number „ , Total Number of _ , . , Male i o o / i • J r i °f e g g s l a i d 1 2 3 4 eggs lai d females _ °° . _ _ ° b per female + S.E. C(2L)VH23lt;C(2R)P3px 1329 485 405 487 2706 41 66.0+2.15 Basc/Y;+/+ 1350 397 454 379 2580 40 64.5 + 2.00 TABLE III Multiplier values obtained for and corresponding estimated total number of progeny expected in each of the experiments involving standard females crossed to compound-2 males. Female genotype Treatment in rads Number of multiplier v i a l s * Mean number of progeny per multiplier v i a l + S.E. Number of experimental v i a l s * * Estimated total number of progeny In(2L + 2R)Cy/apWOW 0 208 104.12 +2.81 1945 203,000 In(2L + 2R)Cy/lt pk on 0 105 63.85 + 1.57 874 55,800 In (2LR)SMl3 Cy/apWM 0 76 83.28 + 3.83 599 49,900 In(2L + 2R)Cy/apW<X> 2000 101 74.93 + 1.94 948 71,000 In(2L + 2R)Cy/lt pk on 2000 89 59.25 + 1.87 755 44,700 In(2LR)SMl1Cy/apWCW 2000 88 79.18+2.29 708 56,000 , , wow +/ap 0 103 92.21 + 2.07 765 70,500 +/lt pk on 0 88 71.18 + 2.01 880 62,600 , wow 2000 76 100.78 + 2.74 779 78,500 +/lt pk on 2000 119 64.13 + 2.11 885 56,800 * Females crossed to Base males. ** Females crossed to compound-2 males. 21 the random segregation of compound autosomes indicated that multiplying the measured value by four would provide a close estimate of the actual meiotic events. Before examining the four aberrant classes and the frequency with which they were generated, i t is important to br i e f l y consider the effect of gamma radiation at high doses, i.e. above 1000 rads. In agreement with the observa-tions made in this study on stage 14 oocytes, experiments show that stage 14 oocytes, the most mature stage, are extremely radio sensitive. It has been reported (Parker and Hammond 1957; Parker 1959) that at a dose of 2000 rads, approximately 97% dominant lethality can be expected at this stage. It should also be noted that Koch, Smith and King (1970), using the frequency of egg hatch as the measure of frequency of lethality, found at 2000 rads the rela-tive lethality of stage 14 oocytes was 75% when corrected for control values. In the present experiment the period of time over which the females were collected (24 - 48 hours) did not guarantee the exclusion of a l l stage 14 oocytes. However, at the radiation exposure (of 2000 rads) used in these experiments most of the stage 14 oocytes should have been eliminated. In this series of experiments I was primarily interested in stage 7 oocytes as Parker (1969), and Parker and Williamson (1970) have shown that interchanges which occur at this time direct the segregation of the chromosomes involved which can result i n nondisjunction. In contrast, segregation of stage 14 oocytes i s already determined and is independent of radiation induced interchange (Busby 1971; Williamson 1973). One could expect sister strand compounds at this dosage (and one was recovered in a preliminary test) but non-sister compounds would be rare as the repair, i t would appear, would be post-fertilization (Parker and Hammond 1958). 22 Although stage 7 oocytes were of prime consideration in this thesis, the magnitude of the number of virgins required (28,800 in 16 experimental crosses) in the recovery of the four classes of events made short collection times im-practical and consequently a 48 hour collection period was used. I did, how-ever, hold the virgins at 18°C on unyeasted medium prior to treatment. The literature dealing with dominant lethality of the two main stages, namely 7 and 14, is based on egg hatchability studies. I observed, from results which I w i l l discuss in detail later, that a large proportion of the progeny from crosses between F(2L)b pr/ln(2LR)SMl,Cy/F(2L)bw females and F(2L)nub b pr/  F(2L)nub b pr;C(2R)rl cn males, a l l of which were aneuploids, survived to pupation. Thus, at least for chromosome 2_, lethality should be based on the frequency of viable progeny not the frequency of egg hatch. In order to determine the contribution of the more sensitive stage 14 oocytes to the re-covered aberrant events I conducted the following experiments. Approximately 1000 In(2L + 2R)Cy/Pin females were collected, aged a minimum of three days to ensure the presence of stage 14 oocytes, and treated with 2000 rads of gamma radiation without etherization. These females were single pair mated to compound-2_ males in shell vials for 20 hours. Egg counts were done on approximately 100 of the vials. From these counts, and correct-ing for st e r i l e v i a l s , the total number of eggs laid by a l l females was e s t i -mated to be 16,500. The 19 recovered progeny from this experiment were as follows: 15 patroclinous (.091%), three matroclinous (.018%) and one sister-strand compound (.006%). (Percent recovery i s based on the number of progeny recovered per eggs laid.) The mean number of progeny recovered per v i a l was .024 (19/796), therefore, in the experiment, using In(2L + 2R)Cy/ap W O W treated with 2000 rads (Table III), the estimated maximum contribution of stage 14 to the total recovered progeny based on the above would be 22.56 progeny (.024 x 948 v i a l s ) . Correcting for this contribution the frequency of total aberrations v (i.e. the sum of the four classes) would have gone from a percent frequency of .740 (Table V) to a corrected value of .708, a rather small decrease. Furthermore, 100 of the originally treated females were mated to males with normal chromosomes marked by b pr It pk cn and brooded for the same 20 hours. Egg counts were performed on a l l vials and only those bearing f e r t i l i z e d eggs were included. Using eclosed progeny as the criterion,jthe survival rate of stage 14 oocytes treated with 2000 rads was calculated to be 5.96% (103 progeny/1,727 eggs). The mean number of eggs per v i a l was 18.37 with a standard deviation of 6.25. In addition, after 20 hours the parents were rebrooded for a period of 3.5 days, from which 5,906 progeny were re-covered giving a mean of 59.06 progeny/vial with a standard deviation of 17.2. It should be noted that six of the females which did not lay eggs in the f i r s t 20 hours, did give progeny in the subsequent brood. Finally, 150 structurally homozygous females collected and treated with 2000 rads within six hours of eclosion and aged two days following treatment, were mated to compound-2_ males with 15 pairs per bottle and l e f t for four days. A total of 37 progeny was recovered, of which 24 were patroclinous, five matroclinous and eight carried a newly induced C(2L) or C(2R). The proportion of the three classes to the total progeny recovered was 65% patroclinous, 13% matroclinous and 22% compounds which parallels the proportions recovered from structural homologues (+/ap W 0 W) 63%, 12% and 24% respectively, as recorded in Table VI. The above three findings using 2000 rads of gamma radiation, namely the low recovery of progeny from stage 14 oocytes (19 progeny/16,500 eggs), the low survival rate (5.96%) of stage 14 oocytes and the parallel proportions of TABLE IV Number of progeny recovered as exceptional chromosome-2 products of meiosis and the number of X-linked recessive lethals obtained in corresponding multiplier tests. Estimated Number of exceptional chromosome--2 products Recessive J-linked lethals* Female genotype Treatment in rads total number of progeny Compounds (2L or 2R) Matroelinous (Disomic-2 eggs) Patroclinous (Nullo-2 eggs) Equational Nondisjunctions In(2L + 2R)Cy/apWCW 0 203,000 0 137 220 6 13/1832 In(2L + 2R)Cy/lt pk on 0 55,800 0 23 38 1 2/941 In ( 2LR) SMI, Cy/apWCW 0 49,900 0 34 41 4 4/950 In(2L + 2R)Cy/apWCW 2000 71,000 38 159 328 4 27/857 In(2L + 2R)Cy/lt pk an 2000 44,700 21 89 138 1 14/821 In(2LR)SMl3Cy/apWOW 2000 56,000 10 92 169 7 18/715 , wow +/ap 0 70,500 0 10 9 1 1/981 +/tt pk cn 0 62,600 0 1 5 0 1/977 , , wow +/ap 2000 78,500 85 43 223 3 13/727 +/lt pk on 2000 56,800 33 30 78 2 13/958 * Number of Z-linked lethals per total number of tested chromosomes obtained from the multiplier test. See text for f u l l description. TABLE V Estimated percent recovery of the four classes or progeny that arose from aberrant chromosome-2 products of meiosis. Female genotype Treatment in rads Estimated frequencies of aberrant chromosome-2 products (in percent)* Compounds Matroclinous Patroclinous Equationals Percent recessive X- linked lethals** wow In(2L + 2R)Cy/ap In(2L + 2R)Cy/lt pk on In(2LR)SMliCy/apWOW 0 0 0 .000 ,000 .000 .067 .041 .068 .108 .068 .082 .003 .002 .008 0.71 0.21 0.42 wow In(2L + 2R)Cy/ap In(2L + 2R)Cy/lt pk on In(2LR)SMl3 Cy/apWOU) 2000 2000 2000 .054 .047 .018 .224 .199 .164 .462 ,309 .302 .006 ,002 .013 3.15 1.71 2.52 wow +/ap +/lt pk on 0 0 .000 ,000 .014 .002 .013 .008 ,001 .000 0.10 0.10 WOW +/ap +/lt pk on 2000 2000 ,108 .058 ,055 .053 ,284 .137 .004 .004 1.79 1.36 * Calculations were made using unmodified data. ** Based on results recorded in Table IV. TABLE VI Comparisons, within and between treatment groups, of the percentage each class contributes to the total number of exceptional progeny recovered from each experiment. Treatment Percent each class contributed to total aberrations Homogeneity test** remaie genotypes in rads Compounds Matroclinous Patroclinous Equationals In(2L + 2R)Cy/apWCW 0 0.0 37.7 60.6 1.6 In(2L + 2R)Cy/lt pk on 0 0.0 37.1 61.2 1.6 In(2L + 2R)Cy/apWCW 2000 * 32.4 66.8 0.8 In(2L + 2R)Cy/lt pk on 2000 39.0 60.5 0.4 Pooled Heterologues 0 0.0 37.6 60.7 1.6 Pooled Heterologues 2000 * 34.5 64.8 0.7 In(2L + 2R)Cy/apWM 2000 7.2 30.5 62.0 0.8 In(2L + 2R)Cy/lt pk on 2000 8.4 35.7 55.4 0.4 , , wow +/ccp 2000 24.0 12.2 62.9 0.9 +/lt pk on 2000 23.2 21.1 54.9 1.4 Pooled Heterologues 2000 7.6 31.9 59.9 0.6 Pooled Homologues 2000 24.0 14.6 60.3 1.0 0.01 3.27 3.20 6.99 n.s. n.s. 3.77 n.s. n.s, n.s. 93.09 <.05 * Since no spontaneous compounds were recovered this class has been excluded to provide a direct compari-son between treatment and controls for the percent distribution of the remaining three classes of exceptional meiotic products. Percent distributions including compounds are presented in lines 7 and 8. ** Based on actual numbers as recorded in Table IV. each aberration between definite stage 7 oocytes and those ranging in age up to 48 hours support the concept of high radiosensitivity of the later stage. Even though a 48 hour collection period was used in the main experiments i t would appear from these results that minimal contribution was made by any stage 14 oocytes which may have been present in the experimental females. Frequencies Of Chromosomal Aberrations: The system confirmed that four classes of aberrancies are recoverable. There is a major trend to the results: radiation definitely increases the frequency of the recovery of disomic-2_ eggs, nullosomic-2^ eggs, and newly generated compound autosomes but does not measurably increase the recovery of equational nondisjunctions. Table V contains the estimated frequencies of each event based on the actual numbers recorded in Table IV. The experiments are grouped according to treat-ment and structural arrangement of the chromosomes involved. While the frequency of recovering compound autosomes was higher for the structural homologues, there was a significant difference between the two lines of females studied (see Table VII). The +/lt pk cn heterozygote gave a wow lower frequency than +/ap , which may have been due to the lower v i a b i l i t y of sister-strand right arm attachments, as only one was recovered as compared to the recovery of eight compounds from the attachments of sister-strand l e f t arms. Only one spontaneous compound was recovered and i t was from a female heterozygous for ap W° W/In(2L + 2R)Cy. This compound i s not recorded in the table because the experiment from which i t was recovered had to be discontin-ued. The matroclinous progeny are not significantly different within groups (see Table XII). However, between group comparisons reveal that the highest 28 frequency of matroclinous progeny was produced by females structurally hetero-zygous for chromosome 2. This was not expected and presented a direct con-f l i c t with the results of Day and Grell (1966) upon which I w i l l comment later. One of the most important findings was the difference between the control and treated both for the structural heterologues and for the structural homologues, the increases being approximately 3X and 5X respectively. Patroclinous progeny, in general, followed a pattern of recovery similar to that of the matroclinous class. However, the differences within groups were usually significant (see Table XIII). It would appear (as described in a later section of this thesis) that the frequency of the two events parallel one another, which may imply a common mechanism. However, the results suggest additional factors may have greater influence upon the formation of patroclin-ous than upon matroclinous progeny. Moreover, the present study revealed for the f i r s t time the recovery of spontaneous equational nondisjunctions. In both induced and spontaneous the frequencies were estimated and while a comparison between spontaneous and i n -duced usually showed an increase for the latter, the differences were found not to differ significantly. Internal Dosimeter: Recessive lethal frequencies were determined for a l l experiments. As expected radiation treatment resulted in measurable increases in the recovery of recessive lethals. An unexpected observation, however, was the higher recovery of spontaneous X-linked recessive lethals from females structurally heterozygous for chromosome 2_ as compared to those that carried structural homologues (Table V). While an earlier study (Thompson 1960) re-vealed that inversions on chromosome 2_ and chromosome _3 would increase the frequency of spontaneous recessive lethals on the corresponding homologue, 29 both in males and in females, no increase in recessive lethals was shown for the X> However, studies on the X chromosome were conducted on males only. 30 DISCUSSION Statistics: One of the main objectives of this research was to recover and quantify four classes of autosomal aberrations, three of which, based on evidence from earlier studies, were presumed to arise as products of chromatid interchange (for a review see Parker and Williamson 1976). Since the frequency of exceptional events had to be determined from an independent multiplier system this presented a s t a t i s t i c a l problem in that the numerator was from a Poisson distribution and the denominator was from an external normal distribution (both of these distributions were s t a t i s t i c a l l y confirmed). Statistics could not be found to determine the confidence interval of fre-quencies determined in this manner. Nevertheless, as I have shown, the sampling of the population for the multiplier system was large (approximately 10% of the total females used), the variance between experiments for the multiplier vials was homogeneous, the numbers of progeny recovered from the multiplier vials were normally distributed, and the number of eggs lai d by the females was male independent. Therefore, the multiplier system appears to provide a good estimate of the population from which the aberrations were re-covered. This was supported further by the reproducibility of frequencies of recovery of matroclinous progeny. Within any group, (defined by treatment and structural arrangement of the chromosomes) there was no significant difference (see Table XII). However, between any two groups, the differences were sig-nificant . Assuming the validity of the multiplier system, there are few statistics which can cope with very low frequencies. For a l l s t a t i s t i c a l comparisons, among the recovered aberrant frequencies, I employed Stevens' Tables (Stevens 1942), using the correction factor for recoveries greater than 20. 31 Internal Analysis: Another approach to studying the frequency of recovery of chromosomal aberrations was to examine the results independently of the multiplier system. This was done by comparing between experiments the percentage each aberration contributed to the total aberrations recovered from that experiment (Table VI). This analysis indicates homogeneity for the controls of both the In(2L + 2R)Cy/ap W° W and the In(2L + 2R)Cy/lt pk cn genotypes. Similarly, the results from both these lines when treated were also homogeneous. Because there was homogeneity within both experiments, the results within the group could be pooled and these pooled data could be com-pared between groups. Thus I compared the results of the structural hetero-logues between the control and the treated. The compound chromosome class was removed from the treated group as there was no equivalent class in the untreated group. From this i t was shown the treated and untreated results were homogeneous. This would imply that radiation does not alter the types of aberrations but simply increases their frequency. With the exception of compound formation the mechanism for spontaneous or induced matroclinous or patroclinous progeny would appear to be similar. These results further indicate that the spontaneous aberrancies also arose primarily in stages prior to stage 14. It also should be noted that they arose randomly and never in clusters. Moreover, the frequencies of exceptional events from the treated structurally homologous strains, +/ap W O W and +/lt pk cn when compared, also proved to be homogeneous and thus could be pooled. However, when the pooled results of the treated structural heterologues are compared with the pooled results of the treated structural homologues i t i s evident that they are not homogeneous. Since the exceptional products recovered from the controls for the structural homologues were infrequent they did not lend themselves to 32 s t a t i s t i c a l analysis and therefore are excluded from Table VI. These results clearly indicate that radiation increases the frequency of a l l aberrant events, excluding equational nondisjunction. However, changing the structural arrangement of the homologues significantly alters the propor-tions of the various exceptional classes, i.e. when structural homologues were compared to structural heterologues, compound chromosome formation was i n -creased in the former and matroclinous progeny in the latter. Internal Dosimeter: The internal dosimeter (i.e. recessive lethal test) provided an independent demonstration of the effects of radiation (Table V). While indeed this internal dosimeter is viewed as important here, i t i s even more important, as described in a later chapter, when chemical mutagens are employed. The recessive lethal test also suggests additional information concerning the use of the It pk cn chromosome. Table V indicates that in every case where the It pk cn chromosome was used a slightly lower, although usually insignificant, value was obtained for every aberrant class when com-pared with the results from experiments using the a p W O W chromosome. The recessive lethal test also indicates a lower value for the It pk cn chromosome. This could imply that when irradiated the It pk cn bearing f l i e s did not receive as high a dosage of radiation. However, since i t is true for a l l four experiments where the It pk cn chromosome was irradiated this seems unlikely. One suggestion is the It pk cn stock may have a lower sensitivity to ionizing radiation resulting in the production of aberrant events (Spieler 1963,1 Sankaranarayanan and Sobels 1976). Chromosomal Aberrations: It can be seen from the results in Table V that structural heterozygosity leads to an increase in the recovery of matroclinous and patroclinous progeny while compound formation is decreased. However, no 32a measurable difference can be detected for the rather infrequent equational nondisjunction. The possible mechanisms leading to these exceptional meiotic products of chromosome 2_ are examined in depth in the following chapters and compared with a unifying hypothesis on the formation of these events (Parker 1969). More-over, a genetic testing system has been developed which for the f i r s t time provides direct evidence that the hypothesis applies to the major autosomes. Finally, the assay system was used to test two known chemical mutagens. From the classes of irregular progeny recovered, the possible modes of action of the mutagens were examined. CHAPTER II ANALYSIS OF RADIATION INDUCED COMPOUND CHROMOSOMES 34 INTRODUCTION The development of compound autosomes represented a major contribution which furthered genetic analysis in Drosophila. They have made i t possible to ini t i a t e a wide range of new inquiries as well as to extend to the autosomes experimental approaches that previously were restricted to the sex chromosomes: half tetrad analysis (Baldwin and Chovnick 1967), v i a b i l i t y of sperm nullo-somic for one or more of the arms of the major autosomes (McCloskey 1966; Lindsley and Grell 1969), spontaneous and induced nondisjunction and chromo-some loss in females (Bateman 1968; Gavin and Holm 1972; Clark and Sobels 1973), gene conversion (Chovnick et a l . 1970; Ballantyne and Chovnick 1971), genetic dissection and characterization of proximal heterochromatin (Hilliker and Holm 1975) and a genetic approach to the regulation of insect populations (Fitz-Earle, Holm and Suzuki 1973; Fitz-Earle 1976). Compound autosomes are chromosomes composed of two homologous arms attached to a common centromere. Although they have been widely reported, l i t t l e i s actually known about the nature of their formation. The original compound autosomes in Drosophila melanogaster were constructed in the labora-tory of E.B. Lewis (Rasmussen 1960). While at f i r s t they were referred to either as attached autosomes or as pseudo-isochromosomes (see Holm 1976) the term compound autosome was adopted following the conventional system of terminology proposed by Lindsley and Grell (1968). In contrast to the com-plex chromosome mechanics required for their i n i t i a l formation, now that compound-2_ and compound-^ chromosomes are available, additional compounds can be easily generated by treating females from a standard strain with ionizing radiation and mating them to compound-2_ (or compound-3) males (Holm 1976). Moreover, they can just as readily be generated in males (Leigh and Sobels 35 1970). In their experiments, Leigh and Sobels (1970) used males heterozygous for recessive markers on the second chromosome. The new compound autosomes they recovered were either heterozygous (called heteroisos) or homozygous (homoisos) for the paternal markers. The heteroisos were recovered from cells that were diploid at the time of treatment while homoisos were recovered both from primary and from secondary spermatocytes. Those from the latter obvious-ly had to be sister-chromatid attachments. Bateman (1968), upon mating radiation treated females bearing standard second chromosomes to males bearing compound-Z autosomes, recovered exception-a l progeny, 25% of which were newly generated C(2L)'s and C(2R)'s (he referred to these as isochromosomes). By using proximal markers Bateman concluded that approximately 25% of the newly induced "isochromosomes" were homozygous at the centromere, suggesting they were sister-chromatid attachments. Holm €it a l . (1969) indicated that newly generated compounds homozygous for proximal markers derived from heterozygous standards, could also occur by a crossover proximal to the marker followed by a non-sister chromatid attachment. However, Gavin (1971) recovered new compound-2_ chromosomes homozygous for proximal markers from females heterozygous for In(2LR)SMI,Cy, a balancer second chromosome that not only prevents crossing over but also by i t s physical structure cannot contribute to compound formation. This finding unquestion-ably demonstrated that true sister-strand attachments could be generated in females. Although the generation of both sister and of non-sister compound-^ chromosomes in both sexes was confirmed, two fundamental questions remained 36 unanswered: 1) what was the nature of their formation, and 2) what was the frequency of their formation both as spontaneous and as radiation induced products? For the mechanism of compound-autosome formation, Bateman (1968) pro-posed three models. His f i r s t model involved breaks on the same side of the centromere of two chromatids in which adjacent, rather than opposite, arms attached, resulting in a dicentric isochromosome and an acentric fragment. His second proposal involved "illegitimate" exchange formed by simultaneous breaks in two chromatids on opposite sides of the centromere with a subsequent rejoining of l e f t arm to l e f t arm or right arm to right arm in a manner similar to that proposed by Rasmussen (1960). His third proposal was similar to Darlington's (1939) in that i t implied a modified version of the centro-mere misdivision model. While genetic recognition of compound chromosomes originated with Morgan's (1922) discovery of attaehed-X chromosomes in Drosophila, concepts on the nature of their formation have been highly influenced by the centro-meric misdivision model originally proposed by Darlington (1939, 1940) to explain the occurrence of isochromosomes in plants. He observed, in cytolo-gical studies of F r i t e l l a r i a , chromosomes which he called isochromosomes because he thought they were the centromeric attachments of sister chromatids. Darlington proposed that isochromosomes arose through a transverse division of the centromere in an univalent followed by the fusion of the half-centro-meres of the sister chromatids. Consequently, as defined by Darlington, isochromosomes would be structurally isosequential from the point of attach-ment, that is they would neither bear duplications nor deficiencies, and would be isogenic. 37 Later Giles (1943) observed in a single plant of Gasteria, one chromo-some that regularly gave rise to isochromosomes. In these studies iso-chromosomes appeared to arise from normally paired and oriented bivalents rather than from univalents. He suggested a possible precocious division of the centromeres followed by the fusion of two sister chromatids resulted in isochromosome formation. Therefore, while Giles supported Darlington's original concept he found that isochromosomes could arise from bivalents as well as univalents. (It is noteworthy that Bateman's third model corresponds closely to that of Giles.) Brandam (1970) provided an alternative explanation to that of Giles (1943). He suggested that a single crossover within a pericentric inversion would result in apparent isochromosome formation. Brandom argues that they would not be true isochromosomes as they would only be isogenic distal to the point of crossing over. However, I wish to point out that this explanation would only account for non-sister chromatid attachments (pseudo-isochromosomes). When she originally generated compound autosomes in Drosophila, Rasmussen (1960) favored independent breaks on either side of the centromere (called illegitimate exchange by Bateman). Leigh , and Sobels (1970), study-ing compound-autosome formation in male Drosophila, also favored the model of two breaks on opposite sides of the centromeres resulting in newly formed compounds with small duplications and deficiencies (Figure 3). Observations leading to their support of this model were the following: 1) a high propor-tion of s t e r i l i t y was found among f l i e s carrying a newly formed compound, 2) not a l l l e f t and right compound combinations produced viable f l i e s , and 3) many of the new compounds had very low v i a b i l i t y . Earlier evidence also supported a translocation-type event (or i l l e g i t i -38 FIGURE 3 Compound chromosome formation i s thought to arise from simultan-eous breaks on either side of the centromere. The broken ends repair in such a way that the two l e f t or right arms are attached to a common centromere. 39 2 1 2 R C ( 2 L ) 40 mate exchange) giving rise to compound chromosome formation. Herskowitz, Schalet and del val Reuter (1962) studied induced crossing over in various regions of the X chromosome including the centromeric region as the females used were heterozygous for a euchromatic duplication (carrying the marker y +) on the right arm of one of the X chromosomes. From such females they re-covered attached X's. Since compound-X formation is extremely rare in the absence of such a duplication, they suggested attachments were the result of pseudocrossovers with one break to the right of the centromere in a duplica-tion-bearing chromatid, and the other to the l e f t of the centromere in one of the other three chromatids. Such a mechanism would allow the formation both of sister and of non-sister attachments. Similarly, Neuhaus (1936) observed, in males, that attached X's could be generated only i f the X chromosome carried a segment of the Y_ chromosome attached to the short arm. Subsequent studies supported a similar type of mechanism for the auto-somes. Kowalishyn (1971) generated compound chromosomes from a strain heterozygous for the SD-72 chromosome which carries a small pericentric inversion. When progeny carrying new right compound-2^  chromosomes hetero-zygous for the SD-72 inversion were mated to compound stocks marked by the r t a l l e l e , the lt_ phenotype was suppressed. This demonstrated that duplications were tolerated as the new right arm carried the locus for the r t a l l e l e . This evidence supported the idea of the locus for l_t being within the peri-centric inversion and was later confirmed cytologically (Hilliker, personal communication). In a subsequent investigation, Yeomans (1972) found in compound-2_ strains generated from standard chromosomes, one right compound chromosome that, when in combination with a l e f t compound homozygous for J-t, suppressed the r t 41 phenotype. In view of this finding, a l l compounds, le f t and right, were tested for duplications of proximal l o c i on the opposite arms. Although no other rights were found to carry a duplication for the lt_ locus, approximately one-half of the l e f t compounds carried duplications for the right proximal marker r_l. These results indeed suggested that compound chromosome formation most likely was caused by breaks occurring on opposite sides of the centromere even as far out as the most proximal markers. Moreover, these findings supported a translocation-like event in which the broken ends rejoined attach-ing l e f t or right arms to a common centromere i n such a way as to generate compound chromosomes heterozygous for varying lengths of proximal deficiencies and proximal duplications of the opposite arms. In addition to radiation induced compound autosomes, one of the spontaneous C(2L) chromosomes recovered by Yeomans carried a duplication for the rj. locus. This evidence suggested, therefore, that spontaneous, as well as induced, compound autosomes were the result of translocation-like events. The objectives of the present research were two fold. The f i r s t was to confirm the translocation nature of compound autosome formation, in part, by determining whether new compounds carried duplications and, additionally by determining i f the duplications included only those genes known to be located in heterochromatin (Hilliker and Holm 1975) or i f they also included genes within the proximal euchromatin. The second purpose was to devise a relative-ly accurate system to determine the frequency of compound autosome formation in females. The system for rescuing newly generated compound autosomes eliminated the recovery of a l l normal progeny. Thus, direct measures of frequencies could not be made. Leigh and Sobels (1970) attempted to quantify the fre-42 quency of compound formation in males by the use of a multiplier system. The system involved mating males to females with standard chromosomes to provide an estimate of the population from which they arose. A similar approach was taken in this study and the arguments to support i t have been given in Chapter I. 43 MATERIALS AND METHODS Induction And Recovery: The genetic markers, mating protocol, treatment and recovery of compound autosomes were described in Chapter I. The multiplier system, as previously described, provided the means of calculating the approximate frequency of compound-autosome formation. The frequencies were determined both for the structurally heterozygous and for the structurally homozygous combinations of second chromosomes. Analysis Of Compound Autosomes Generated In Females Carrying The Genetic  Marker a p W O W : The use of the inverted chromosome, In(2L + 2R)Cy, which acts as an effective balancer for chromosome 2^, in combination with a normal second proximally marked by a p w o w on the right arm, allowed for distinguishing be-tween sister versus non-sister chromatid attachments. A l l newly generated compound-2^  l e f t chromosomes were either phenotypically +, indicating a sister-strand attachment, or phenotypically Cy_, representing a non-sister attachment, i.e. a compound-left autosome heterozygous for In(2L)Cy. This inversion carries in addition to the genetic marker Cy_, recessive lethals which make i t impossible to recover sister-strand compounds for the In(2L)Cy chromosome. Similarly, the right compounds could be classified as sister-strand attachments by the expression of the a p W O W phenotype and the non-sisters by a + phenotype. Again the presence of recessive lethals associated with In(2R)Cy precluded the recovery of sister chromatid attachments of this chromosome. The genetic mark-er a p W O W was used as i t is viable when homozygous and in close proximity to the centric heterochromatin and thereby greatly reduced the probability of ex-change occurring between i t and the centromere, and event that would have led to the misclassification of compound-2R non-sister attachments. 44 In(2L + 2R)Cy/ap wow (treated and control) females were mated to C(2L)VH2, It; C(2R)P,px males. New compound-2L chromosomes were phenotypically identified as C(2L)+;C(2R)P,px (sister-chromatid attachments) or C(2L)Cy;C(2R)P,px (non-sister-chromatid attachments). Similarly, new compound-2R chromosomes were phenotypically C(2L)VH2,It;C(2R)apW O W (sister chromatid) or C(2L)VH2,lt;C(2R)+ (non-sister chromatids). The marker, Lt, on the compound-2L chromosome from the males was selected specifically to provide immediate identification of newly induced compound-2R chromosomes that carried a duplication of proximal 2L including the l t + a l l e l e . These f l i e s were phenotypically wild type or phenotypically a p W O W . Furthermore, a l l progeny bearing newly generated l e f t compounds were mated to C(,2L)P,b;C(2R)VH5,r 1 (b_, black; r_l_, rolled) to identify new compound lefts carrying a duplication for the right arm proximal marker r l . Those that carried a r l + duplication were crossed to C(2L)+;C(2R)stw (stw, straw) to determine whether the duplications extended dis t a l l y as far as the stw locus, which is a proximal marker known to be located in euchromatin. [See Lindsley and Grell (1968) for the positioning of the above markers and see H i l l i k e r and Holm (1975) for the relative posi-tion of It and r l . l One phenotypic class of progeny from this experiment, ap"""", could have had three possible modes of formation. Flies that were phenotypically a p W O W could have been the result of the simultaneous recovery of l e f t and right sister-strand attachments, thereby forming a disomic-2_ egg rescued by a nullo-2^ sperm. One such event was recovered previously (Gavin 1971). This phenotype could have also represented an equational or second division non-disjunction which would have been rescued by the same class of sperm. Additionally, as indicated above, this class could have represented a newly induced C(2R) sister-chromatid attachment carrying a duplication of 2L_ that wow 45 included the l t + a l l e l e . Hence, to distinguish among a l l the three possibili-ties, progeny of this phenotype were mated in turn to f l i e s of the following genotypes: C(2L)VH2,It;C(2R)P,px and In(2L + 2R)Cy/bwVl. The structurally heterozygous experiments were repeated using the multiple break inversion In(2LR)SMI,Cy (Lindsley and Grell 1968). This balancer chromosome allowed for the recovery only of sister-strand l e f t and wow right compounds from the chromosome carrying the ap genetic marker. Since In(2LR)SMl,Cy carries a large pericentric inversion, compounds formed from this chromosome whether they were sister strand or non-sister strand attach-ments would produce inviable segmental aneuploids. New sister compourid-2L chromosomes were phenotypically identified as C(2L)+;C(2R)P,px and sister compound-2R chromosomes were phenotypically identified as C(2L)lt;C(2R)apW°W. As described above, a l l phenotypically a p W O W progeny were further tested to determine i f they were equational nondisjunctions, simultaneous generation of le f t and right compounds or newly generated right compounds carrying a dupli-cation of 2R including the l t + a l l e l e . In those experiments in which the second chromosomes were isosequential for the standard arrangement, but heterozygous for a p W ° W (+/ap W° W), only the compound-2R's homozygous for a p W ° W could be classified as sister-strand attachments. The +/ap W O W females were mated to C(2L)VH2,lt;C(2R)P,px males. New compound-ZL chromosomes were phenotypically identified as C(2L)+;C(2R)P,px (both sister and non-sister attachments). New compound-2R chromosomes were phenotypically identified as C(2L)VH2,It;C(2R)apW°W (sister chromatids) or C(2R)VH2,lt;C(2R)+ (non-sister chromatids or sister chromatid attachments of the unmarked arm). Additionally, compounds recovered from +/ap W O W females were tested for duplications as described previously. 46 Analysis Of Compound Autosomes Generated In Females Carrying The Genetic  Markers It pk cn: The above series of experiments were repeated excluding the combination with In(2LR)SMl,Cy, replacing the chromosome carrying the genetic marker a p W O W by a chromosome carrying the genetic markers It pk cn. Females (treated and control) were mated to C(2L)P,b;C(2R)P,px males. From the structurally heterozygous females the new compound-ZL sister strands were phenotypically r t and the compound-2L non-sister strands were phenotypically Cy (Cy sister strands are not recoverable for the reasons presented earlier). The new right compounds were identified as sisters by means of the pk cn pheno-type and non-sisters were phenotypically + (wild type). Compound autosomes arising from sister-chromatid attachments in females structurally homozygous for chromosome 2_ could be identified only for the l e f t arm marked by lt^ or the right arm carrying the markers pk cn. Non-sister-chromatid attachments as well as sister-chromatid attachments for the unmarked chromosome were indistinguishable, i.e. they were phenotypically +. Compound-2^  chromosomes generated from experiments involving It pk cn were not tested for duplications as the chromosome carried the lt_ marker. 47 RESULTS Based on values estimated by the multiplier system, from a l l five control (0 rad) crosses recorded in Table IV, plus one additional cross involving In(2L + 2R)Cy/ap W O W females by C(2L)VH2,lt;C(2R)P,px males, which was not recorded in the table, only one spontaneous compound, a C(2L), was recovered from an estimated 500,000 progeny. While this exceptionally infrequent recovery of spontaneous events agrees with previous findings from this labora-tory (Gavin 1971; Yeomans 1972), Chadov (1973) reported the recovery of 29 spontaneously generated compound-2_ chromosomes. However, he failed to provide information pertaining to the magnitude of the experiments from which the chromosomes were recovered, and therefore no comparisons can be made. In view of the rare recovery of spontaneous events i t is evident that the vast majority of newly generated compound autosomes were radiation induced (Table VII). It is important to realize that, as a function of the meiotic behaviour of compound autosomes in males, only 25 percent of the newly generated chromosomes would be recovered. Therefore, the actual frequency of formation i s approximately four times that measured, which in fact i s strik-ingly close to the frequency of detachments of compounds at 2000 rads reported by H i l l i k e r and Holm (1975). In considering the mechanism of compound-chromosome formation one important parameter to measure is the frequency of recovering the various classes of expected attachment products. If, for example, compound autosomes arise as products of interarm interchanges (an interbrachial reciprocal trans-location) C(2L) and C(2R) chromosomes would be expected in a 1:1 ratio. More-over, since such events are viewed as interchromatid interchanges (Parker 1969, TABLE VII Estimated percent recovery of compound-2 chromosomes. Female genotype Treatment in rads Estimated total number of progeny Number of compounds (2L or 2R) Frequency in percent 95% Confidence intervals* wow In(2L + 2R)Cy/ap In(2L + 2R)Cy/lt pk on 0 0 203,000 55,800 0 0 0.000 0.000 0.000 - 0.002 0.000 - 0.007 WOW In(2L + 2R)Cy/ap In(2L + 2R)Cy/lt pk on 2000 2000 71,000 44,700 38 21 0.054 0.047 0.038 - 0.073 0.030 - 0.072 wow +/ap +/lt pk on 0 0 70,500 62,600 0 0 0.000 0.000 0.000 - 0.005 0.000 - 0.006 wow +/ap +/lt pk on 2000 2000 78,500 56,800 85 33 0.108 0.058 0.083 - 0.133 0.040 - 0.081 * Based on the corrected fiducial limits of expectation from Stevens' Table (Stevens 1942), 49 Williamson 1969), sister and non-sister chromatid attachments both would be expected. The results for a l l four experiments in which females were treated with 2000 rads are recorded i n Table VIII. Although the recovery of G(2L) and C(2R) are homogeneous in comparing a l l four experiments, the results do not f i t the 1:1 ratio expected but rather a 4:3 ratio of compound-2L to compound-2R respectively. Even though each experiment is consistent in moving in the direction of a 4:3 ratio of C(2L) to C(2R), and the pooled results differ significantly from a 1:1 ratio, individual results do not depart significantly from the expected random recovery of the reciprocal products. These dispro-portions w i l l be considered further in connection with the comparison between sister and non-sister chromatid attachments. Regarding the involvement of the various combinations of chromatids, let us f i r s t consider the right-arm attachments of chromosome 2. In a l l experi-ments the right arm of chromosome 2_ was heterozygous for a recessive marker, either a p W O W or pk cn. If, among the four chromatids, there was a random attachment of any two, there would be six possible combinations of any two chromatids involved in the formation of a new compound-right chromosome. Sister strand compound-2^  formation would result from attaching two sister chromatids to the same centromere; non-sister formation would involve two chromatids, one from each homologue. In the present experiments the formation of a sister-strand compound could be detected by the recovery of C(2R) chromo-somes homozygous either for the recessive marker a p W O W or for the markers nk cn. The recovery of such phenotypically marked f l i e s , therefore, should be 1/6 of the total right compound-2^  chromosomes produced from structural homologues. The results recorded in Table IX, for the structurally homozygous combinations, support the notion of random involvement of any two of the four chromatids, that i s , 1/6 of newly induced right compound chromosomes were TABLE VIII Comparison of the recovery of compound-2 l e f t vs. compound-2 right chromosomes. Female genotype Treatment in rads Number of compound-left chromosomes Number of compound-right chromosomes In(2L + 2R)Cy/apWOW In(2L + 2R)Cy/lt pk on , , wow +/ap +/tt pk on 2000 2000 2000 2000 24 14 12 9 47 38 19 14 Totals 102 75 * * The recovery of C(2L) to C(2R) was homogeneous (P = .9) for the nonrandom distribution of 4:3. TABLE IX Comparison of sister vs. non-sister compound-chromosome formation for the right arm of chromosome 2. Number of sister- Number of non-sister-Female genotype chromatid compound-2R chromatid compound-2R In(2L + 2R)Cy/apWOW 3 11 In(2L + 2R)Cy/lt pk on 1 8 +/apWOW 6 32 * +/lt pk on 1 13 * Totals 11 64 * This group includes sister-chromatid attachments of the unmarked (+) homologue. 52 homozygous for the recessive markers. For the structurally heterozygous second chromosomes, theoretically 1/5 of the progeny should be homozygous for the recessive allele as the inversion on the right arm carries a recessive lethal which presents the recovery of sister-strand attachments for this chromosome and, in theory therefore, eliminates 1/6 of the potential compound-2R chromosomes. Again, Table IX shows that for structural heterozygotes the theoretical expectation was supported by the experimental results. One of the d i f f i c u l t i e s in determining randomness of chromatids involved arises from the possibility of an apparent sister-strand attachment actually occurring as a result of compound formation from non-sister chromatids pre-ceeded by an exchange between the homologues. To reduce the possibility of exchange, proximal markers were used. The number of sister-strand attachments from structural homologues, where exchange could take place, is not s i g n i f i -cantly different from the number of sister-strand attachments from structural heterologues where exchanges are eliminated. In addition, when the chromosome bearing the genetic marker a p W O W was used in conjunction with the balancer, In(2LR)SMl,Cy, a chromosome bearing multiple inversions, crossing over is eliminated. The multiple rearrangements result in a l l non-sister and sisters from this chromosomes being inviable with the complementary compound from the male. The only recoverable compound chromosomes are sister-strand attachments from the a p W O W chromosome, of which four rights were recovered. Randomness for the l e f t chromatids was also investigated where possible. For the chromosomes In(2L + 2R)Cy/lt pk cn randomness was observed (Table X). TABLE X Comparison of sister vs. non-sister compound chromosome formation for the l e f t arm of chromosome 2. Female genotype Number of sister-chromatid compound 2L Number of non-sister-chromatid compound 2R In(2L + 2R)Cy/ccpWOW In(2L + 2R)Cy/lt pk on , , wow . +/ap * +/lt pk on 10 ** 1 8 * * 14 11 11 *** * Sister-chromatid compound-lefts could not be identified. ** The recovery of sister-chromatid attachments greatly exceeded the expected ratios of 1:5 for cross 1 and 1:6 for cross 4. The recovery of sister-2L compounds f i t s a Poisson distribution. *** This class includes both sister-strand attachments from the unmarked chromosome and the non-sister-strand attachments..• 54 However, for the other two detectable cases, +/lt pk cn and In(2L + 2R)Cy/apWQ1 a higher proportion of sister-strand attachments than the model of randomness would predict, were recovered. These results are d i f f i c u l t to reconcile as in the latter case the increase cannot be attributed to crossing over. It is interesting to note that the recovery of l e f t and right sister chromatids for In(2LR)SMI,Cy/apWOW are approximately equal (six l e f t s , four rights), indicat-ing that the excess recovery of sister lefts cannot be accounted for by the a p W O W chromosome, (data not included in Tables IX and X). The anomalous increase in recovery of sister lefts accounts for the greater total recovery of compound-2L's than compound-2R's. The preferential recovery or formation of compound-2L i s also found in the data of Bateman (1968) who recovered 430 lefts versus 392 rights and Chadov (1973) who recovered 28 lefts versus one right. These results indicate nonrandomness in the formation of compound-2^  chromosomes, although no explana-tion can be given at present for this deviation from the expected. From earlier work i t has been demonstrated that i t is extremely d i f f i c u l t to recover reciprocal translocations, especially in stage 7 oocytes (Traut 1967a, Parker 1969, Busby 1971). These results suggest that the formation of compound chromosomes in stage 7 oocytes would lead to the segregation of the reciprocal products of the translocation. This was supported by the finding that none of the putative equational nondisjunctions with the a p W ° W chromo-some proved to be reciprocal compounds. While the above findings are consis-tent with the concept that compound autosomes are products of reciprocal translocations of which only one-half the translocation i s recovered owing to the interchange event in females, these results alone provide insufficient evidence to refute the model of centromeric misdivision. Evidence that the 55 event i s indeed that of a translocation comes from the genetic analysis of newly generated compounds for duplications of genes located in the proximal regions of the opposite arm to that of the attachment. Kowalishyn (1971) demonstrated through the use of a small pericentric inversion on the SD-72 chromosome that duplications and heterozygous deficien-cies of proximal 2L and 2R heterochromatin could be tolerated. Moreover, Yeomans (1972) found that proximal duplications were sometimes produced as a result of compound formation. I tested a l l new compound chromosomes recovered from In(2L + 2R)Cy/ap W O W or +/ap W O W females for duplications of the complemen-tary arm. The l e f t compound chromosomes were tested for the proximal locus r l . As shown in Table XI, over 60% of the new l e f t compounds, sister and non-sister strand attachments both, carried duplications for the rl_ locus. A l l those compound-2L chromosomes with the duplication for r_l were tested further to determine i f the duplication covered the more distal'marker stw. However, none carried stw+. Similarly, newly generated compound right chromosomes were tested for a duplication of the l e f t arm bearing the J_t_ locus (Table XI). In one experiment (+/ap W° W) 13% carried the l t + duplication. In the other experiment (In(2L + 2R)Cy/ap W O W)none were recovered. However, fewer compound chromosomes were generated in this experiment and the absence of duplications for l_t was not s t a t i s t i c a l l y unexpected. In the cases for which C(2R) chromo-somes carrying duplication l t + were recovered, i t was somewhat surprising that none were sister-strand attachments. Interestingly the single compound i n -duced in stage 14 (see Chapter I) which was a sister-strand attachment for the right arm did carry a duplication for the lt_ locus. TABLE XI Genetic analysis of newly generated compound-2 chromosomes for the presence of duplications. Compound-.2 lefts carrying duplications of proximal 2R Female genotype vl~ rt+ total percent with r l + duplication* In(2L + 2R)Cy/apWCW 6 10 16 62.5 , , wow +/ap 14 25 39 64.1 Compound-2 rights carrying duplications of proximal 2L Female genotype It lt+ total percent with lt+ duplication* In(2L + 2R)Cy/apWM 13 0 13 0.0 , , wow +/ap 33 5 38 13.2 The results were homogeneous using G u modified by applying Yates correction for continuity (Sokal and Rohlf 1969). DISCUSSION The present research provides a new approach for estimating the frequency of induced compound-autosome formation in females. The mechanism of compound-2^  chromosome formation strongly favors the concept, originally proposed by Rasmussen (1960), of the joining of independent breaks on opposite sides of the centromere, a concept which Bateman (1968) referred to an illegitimate ex-change. The evidence from this study strongly favors this proposal, i.e. radiation induces independent breaks on opposite sides of the centromere either of homologous or sister chromatids. To be recovered these breaks must rejoin in such a way that an acentric arm joins with an homologous (or sister) centric arm (Figure 3). The breaks occur randomly and rejoining can occur between any two of the four chromatids i n the tetrad. Compound chromosomes generated in this manner would carry small duplications for the most proximal region of the opposite arm and would be heterozygous for proximal deficiencies. For example, a newly generated C(2L), as a consequence of joining an acentric 2L to a centric 2L chromatid, would carry a proximal segment of 2R and would be heterozygous deficient for a proximal segment of 2L. Although never applied to the present situation, Parker (1969) used the term interchange to describe induced exchange between nonhomologues. Williamson (1969) extended the term to include interbrachial interchanges in the Y chromosome. In the present study chromatid breaks resulting in a new compound chromosome would be considered an interchange. Furthermore, Parker (1969) found that when interchanges took place between nonhomologues they tended to segregate at meiosis. In preliminary studies matroclinous progeny were tested to see i f they were double compounds, but never was this the case. In the present study i t was possible that a l e f t and right sister-strand 58 attachment could occur simultaneously resulting in a f l y phenotypically iden-t i c a l to an equational or second division nondisjunction of which several were recorded. A l l such f l i e s were tested, but again none were found to be double compounds. Gavin (1971) observed one simultaneous recovery of a l e f t and a right compound, both of which were sister-strand attachments. This product was recovered in a late brood and therefore must have occurred in an early stage of meiosis. The strongest evidence for a translocation type event arose from a follow up of the original observations made by Kowalishyn (1971) and Yeomans (1972), that some newly generated compound autosomes carried duplications for proximal l o c i on the opposite arm. Upon testing a l l newly generated compounds re-covered from In(2L + 2R)Cy/ap W O W or +/ap W O W females, I found 63% of the induced l e f t arms carried a duplication that covered the marker r l located in the proximal heterochromatin of the right arm. Similarly, 9% of a l l right arms tested carried duplications for proximal regions of the l e f t arm, which included the r t locus. H i l l i k e r and Holm (1975) have demonstrated, through the detachments of compound seconds, that rl_ i s located in the heterochromatin of the right arm of chromosome 2_. Thus breaks resulting in the r l + duplica-tion occurred in the heterochromatic block between the r l locus and up to, or possibly including, proximal euchromatin. To test for breaks in euchromation the compound l e f t chromosomes were also tested for duplications involving breaks in the right arm distal to stw. None were found. The failure to recover euchromatic duplications may be explained in three ways: 1) there is a low frequency of recovery of breaks dis t a l to the euchromatic-heterochromatic junction, 2) f l i e s are unable to tolerate auto-somal duplications of this magnitude, and 3) the geometry of the chromosomal 59 arrangements at this stage of meiosis imposes a restrictive distribution on the interchange events such that heterochromatic breaks only rejoin with other heterochromatic breaks. It is interesting to note, in regard to the last point, that H i l l i k e r and Holm (1975) found that, with the exception of one unusual three break product, a l l products of compound-2_ detachments were res-tricted to heterochromatic-heterochromatic interchange events. The recovery of over 60% of the breaks i n the right arm distal to r l strongly implies that the £l locus f a l l s approximately in the middle of the right heterochromatic block, supporting the conclusion derived from detach-ment studies (Hilliker and Holm 1975). Fewer l t + duplications were recovered on new right compounds, possibly because i t is more distal than r l . Again, this i s supported by the detachment studies (Hilliker and Holm 1975). Many of the sister l e f t compounds carried the r l + duplication, indicating a common mechanism of formation both for sisters and for non-sisters. However, of the few rights carrying l t + duplications none were sister-strand attachments. As noted in the introduction to this chapter, results of other investi-gations also support a translocation like event. Neuhaus (1936) found that he could only recover spontaneous attached-X chromosomes in males i f they carried an attached X.Y chromosome. The frequency was much higher when the S L attached X.Y involved Y rather than Y_. He suggested, therefore, that a crossover between the Y fragment and the proximal region of the sister-chromatid X would result in the formation of an attached-X chromosome. Com-parably, Morgan (1938) thought the origin of attached-X chromosomes was a two step crossover event. The f i r s t exchange between an X and a Y made the attached X.Y chromosome, and in a subsequent generation a second exchange between the attached X.Y and a normal X led to the formation of an attached-X 60 chromosome bearing a Y_ centromere. In both of these studies attached X.Y chromosomes were viewed as spontaneous events occurring in males. Bateman (1968), although offering three models, appeared to favor the model of centromeric misdivision because the reversions he recovered from the detachment of compound-2^  chromosomes were a l l homozygous viable. In contrast, Baldwin and Suzuki (1971), upon analyzing detachment products of compound-3_ chromosomes, found that slightly less than half (66/162) were homozygous lethal. Similar results were found for the second chromosome (Hilliker and Holm 1975; Gibson, unpublished data). Although i t has been realized that the inab i l i t y to recover homozygous reversions i s , in some instances, due to the presence of deficiencies in compound autosomes as a function of their forma-tion (Baldwin and Suzuki 1971; H i l l i k e r and Holm 1975), compound-2 chromosomes diploid for a l l genetic l o c i have been identified (Hilliker and Holm 1975). While other models of compound chromosome formation cannot be excluded, they are not supported by the results obtained in this study. Darlington (1940) and Giles (1943) originally suggested centromeric misdivision as a method of compound chromosome, (or in their terminology isochromosome), forma-tion in plants. Darlington's model, specifically, could only give rise to sister-strand chromatids and my results definitely show that compounds can be produced from both sister and non-sister chromatids. Bateman's (1968) model of centromeric misdivision incorporated the formation of both, but as pre-viously mentioned, reversion studies indicated that deficiencies are produced in the formation of many compounds. Similarly, the formation of compound chromosomes leads to the recovery of duplications as supported by the findings in the present study. 61 Brandham (1970) thought that the Isochromosomes (I.e. compound chromo-somes) observed by Giles (1943) could have been the result of crossing over within a pericentric inversion. This model is plausible but f a i l s to explain how sister-strand attachments are produced or how compounds are formed in male Drosophila where crossing over does not occur. The suggestion of Brandham is not supported by the findings of Kowalishyn (1971). He generated compound chromosomes from an SD-72/cn bw stock, where the SD-72 chromosome carries a pericentric inversion. The compounds induced were 19.8% of the total recover-ed progeny, whereas in the present study, with the +/ap stock, the induced compounds represented 23.8% of the total progeny. This would indicate that a pericentric inversion does not increase the formation of compound chromosomes and certainly i s not a main contributing factor in f r u i t f l i e s . Moreover, Gavin (1971) was able to generate compound chromosomes in the presence of the crossover suppressor, c(3)G, indicating that the normal mechanisms required for crossing over were not necessary for the generation of compound chromo-somes. Previously, the spontaneous recovery of compound chromosomes was reported by Gavin (1971), Yeomans (1972) and Chadov (1973). The frequency of such events I found to be extremely low, in that I recovered only one out of an expected 5 x 10^ progeny. Nevertheless, this compound was of considerable interest in that i t was formed through the attachment of sister chromatids of the l e f t arm, and moreover, i t carried a duplication for the r_l locus. This suggests that the mechanism for the spontaneous formation, either as sister or as. non-sister attachments, is the same as that brought about by radiation. Finally, I conclude from the results of this study that compound autosomes arise as products of translocation. CHAPTER III ANALYSIS OF RADIATION INDUCED MATROCLINOUS AND PATROCLINOUS PROGENY 63 INTRODUCTION The Sex Chromosomes: Early studies on radiation induced X-chromosome nondisjunction led to the recovery of matroclinous progeny as well as the reciprocal class, patroclinous progeny. However, the latter class was always more frequent (Mavor 1924). The increase in the frequency of patroclinous progeny above that of matroclinous progeny was defined as chromosome loss, an event thought to be separate and independent from nondisjunction, but recover-ed simultaneously. Day and Grell (1966) examined the rate of X-ray induced primary nondis-junction with respect to structural heterozygosity. The sex chromosomes were appropriately marked so that the nondisjunctional female progeny (XXY) could be differentiated from the regular female progeny (XX), and the exceptional male progeny (XO) from the regular male progeny (XY). They used only the recovery of exceptional female progeny to determine the frequency of nondis-junctional events. Consequently this value had to be multiplied by a factor of four. Their study failed to show any significant difference in the frequency of induced nondisjunction of the X chromosome comparing structural homozygosity with structural heterozygosity. It is important to note that the homozygote used was homozygous for an inverted-X chromosome. The type of chromosomal aberrations induced is dependent upon the maturity of the oocytes involved. Oocyte staging was carried out by King, Rubinson and Smith (1956) and Koch, Smith and King (1970). The two main stages which can be readily identified are mature oocytes, which are consid-ered to be in the latest stages of prophase I or early metaphase I, and classified as stage 14, and immature oocytes, which represent earlier stages 64 of prophase I and classified as stage 7. A particular stage can be treated by altering the time between eclosion of the f l i e s and treatment. Those females treated immediately upon eclosion are considered to have their most mature oocytes at stage 7, while those f l i e s that are aged for one day or more before treatment w i l l have their most mature oocytes at stage 14. Traut (1970) re-ported the lack of stage sensitivity to radiation induced nondisjunction. This is in contrast to most other types of chromosomal aberrations for which mature oocytes show greater sensitivity (Parker and Hammond 1958; Parker 1959; Traut and Schmidt 1968; Traut 1967a, 1967b, 1968; Browning 1970; Traut and Scheid 1971). As previously mentioned, Mavor (1924) concluded that the exceptional male class included products of nondisjunction as well as products of chromosome loss. To determine the frequency of chromosome loss, Day and Grell (1966) subtracted the number of matroclinous progeny from the total number of patro-clinous progeny and multiplied by two, as only half of the patroclinous pro-geny could be recovered, i.e. the nullo gametes would only be recovered by X-bearing sperm. The exceptional males were genotypically X/0 and resulted either from the reciprocal event of nondisjunction producing a disomic egg or through some radiation induced event leading to the loss of normally.segre-gating X chromosomes. The excess of exceptional males was thought to be due to the latter event. Traut and Scheid (1971) have shown cytologically that chromosome loss may be partial or total in immature oocytes. This supports the earlier work of Grell, Munoz and Kirschbaum (1966) who found a correla-tion between the length of the chromosome and the induced loss. Considerable evidence supports the idea of chromosome loss and nondis-junction being separate events although recovered in the same manner. Traut 65 (1971) demonstrated that dose protraction but not dose fractionation reduced the frequency of nondisjunction, while both reduce chromosome loss. Traut proposes that some chromosome loss arises from the fusion of two independent breaks in sister chromatids thus leading to dicentric bridge formation. This would agree with the reduced effect of protraction and fractionation as the f i r s t break would be repaired before a second break i s produced. Chromosome loss as independent from nondisjunction is also supported by Muller (1940), Pontecorvo (1941) and Grell et a l . (1966). Further evidence differentiating the two events comes from the study of Day and Grell (1966), who did find a stage sensitivity for chromosome loss, with the greater recovery in stage 14, but did not find stage sensitivity for nondisjunction. Hence i t appears that there i s general agreement that nondisjunction and chromosome loss represent two distinct and separate events. The Autosomes: Since certain types of aneuploids for the sex chromosomes can be tolerated by Drosophila, namely XXY in females and XO in males, most of the studies on chromosome loss and nondisjunction have been limited to these chromosomes. However, Grell ert a l . (1966) introduced a new approach to studying these two phenomena by using a free-X duplication, which did not affect v i a b i l i t y , together with a free fourth chromosome. The second fourth chromosome was present as a translocation with one of the major autosomes. The free fourth and the X duplication thus would enter a non-competative dis-tributive pairing pool and would regularly segregate, i.e. the X duplication would segregate from the free fourth (Grell 1962, 1964). When females of this genotype were treated with radiation the frequency of induced nondis-junction of the free fourth and the X duplication did not differ significantly from that observed for normal X chromosomes. 66 The f i r s t attempt at studying nondisjunction in the major autosomes was performed by Bateman (.1968). He mated radiation treated females carrying standard chromosomes to males bearing C(2L);C(2R). This enabled the recovery of nondisjunctional products and loss of chromosome _2 as previously described in Chapter I. Ten percent of the total progeny recovered were matroclinous while sixty-six percent were patroclinous, the remainder carried newly formed compound autosomes. However, this approach enabled only relative values to be determined. In a study similar to Bateman's, Evans (1971) made an attempt to quantify the events by examining eggs for hatchability frequencies. This type of experiment is severely limited in size and furthermore, demands a clear understanding of the developmental stages at which aneuploidy for chromosome-2^  i s lethal. Wurgler, Ruch and Graf (1971) have also used the compound-^ system for studying induced chromosome loss in secondary oocytes. While they did reveal some chromosome loss at this stage, the test required the treat-ment of eggs 3 - 4 minutes after egg laying, and thereby severely limited the size of the experiment. Gavin (1971) examined the effect of radiation on nondisjunction of the second chromosome in normal females mated to males carrying compound-^ auto-somes. In her study no distinction was made between nondisjunction and chromosome loss. Gavin quantified this recovery of nondisjunction and newly formed compounds by crossing, in bottles, a number of treated females to normal males. This provided an estimate of the number of progeny that might be expected but was limited by the number of bottles used for the multiplier system and the effects of crowding from mass matings. It did demonstrate, however, a method of quantifying the exceptional events being studied. Gavin and Holm (1972), again using multiplier bottles as an estimate of total pro-geny, measured the frequencies of nondisjunction and chromosome loss of the second chromosomes of irradiated females. Clark and Sobels (1973) studied autosomal nondisjunction using irradiated females that were carrying compound seconds differentially marked from those carried by the males to which the females were mated. As mentioned previously, compound autosomes segregate at random in males, while in females compounds enter the distributive pairing pool (R.F. Grell 1962, E.H. Grell 1970, Holm and Chovnick 1975, Holm 1976). Provided the compound seconds are the only two chromosomes which enter the pool, they w i l l normally segregate. Thus any induced nondisjunction resulting in disomic oocytes w i l l be recovered by nullo sperm. The disomic sperm, as in previous studies, w i l l rescue oocytes repre-senting the reciprocal of the disomic egg class, which is chromosome loss. The normal segregational product in the female carrying either a compound-2L or a compound-2R w i l l be viable with the complementary compound chromosome from the male. This allows the ratio of exceptional to normal events to be calculated in the same manner as in the X~chromosome studies. It should be noted that the attached arms in females pair for exchange in prophase I, and that the pairing does not influence the distributive pairing properties which are apparently responsible for their regular segregation. Therefore, caution must be exercised in attempting to compare the results from compound chromo-somes to those of standard chromosomes. Equational Nondisjunction: Although considerable effort has gone into the study of nondisjunction, attention has been focused primarily upon reduc-tional nondisjunction, or nondisjunction of the f i r s t meiotic division. Nevertheless, from the work of Mavor (1924) we find that, although with con-68 siderably lower frequencies, putative equational, or second division, nondis-junction also results from irradiation treatment of the X's. Gavin (1971), using heterozygous markers on the second chromosome, also reported autosomal equational nondisjunction both from males and from females. However, frequen-cies of these events were not determined. In the present study an attempt was made to identify and to determine the frequency of second-division nondisjunc-tion of chromosome 2. 69 MATERIALS AND METHODS The procedure and chromosomes used were as outlined in Chapter I. The matroclinous progeny for chromosome 2_ were either phenotypically Cy (curly) or phenotypically + (wild type), depending on the females used. Patroclinous progeny were either It px or b px depending on the compound-2_ chromosome carried by the male. Equational nondisjunctions were identified by the recovery of progeny homozygous for markers that were heterozygous in the female parent. In the WOW ' case of ap , a putative equational nondisjunction could have been the result of a crossover between a p W ° W and the centromere followed by reductional non-disjunction. In such cases the progeny would carry the hypostatic marker Cy. A l l progeny that were putatively identified as arising from equational nondis-junctional events, from In(2L + 2R)Cy/ap W° W females, were tested for the presence of the Cy_ marker by mating them to + f l i e s . The absence of the Cy phenotype in progeny from this cross was taken as confirmation of an equation-al nondisjunction. With the It pk cn chromosome the recovery of progeny homozygous for markers l_t and £k, which flank the centromere, was used to identify equational nondisjunctions. It is important to note that, in a l l experiments, products of equational nondisjunction could be identified only in those progeny bearing recessive genetic markers; products of equational nondisjunction of the unmarked chromosome could not be identified. 70 RESULTS There are many ways of defining nondisjunction and chromosome loss. To reduce confusion and to avoid definitions implying possible mechanisms of formation, I w i l l simply refer to the exceptional products either as matro-clinous progeny, which clearly reflect reductional nondisjunction, or as patroclinous progeny, which may reflect nondisjunction or chromosome loss. The observed values of both the matroclinous and the patroclinous progeny w i l l be described without any corrections as only relative frequencies are import-ant. Estimated total frequencies of the actual meiotic occurrence can be ob-tained by multiplying the recorded values by a factor of four. Spontaneous And Radiation Induced Matroclinous Progeny: The spontaneous recovery of progeny, matroclinous for the X chromosome, have been studied previously under three structural arrangements: 1) structurally normal homologues (Uchida 1962, Traut 1970), 2) structurally heterozygous homologues (Day and Grell 1966), and 3) inversion homozygotes (Day and Grell 1966). In the present study, with the second chromosome, I considered only the f i r s t two configurations. The data in Table XII indicates that the structural organization of the chromosomes was more important in determining the spon-taneous frequency of matroclinous progeny (both second chromosomes coming from the mother) than the specific genetic markers used on standard chromosomes. The spontaneous recovery of matroclinous progeny was not significantly different for the two inversions, In(2LR)SMI,Cy and In(2L + 2R)Cy, in combin-ation with a normal chromosome. Similarly, the spontaneous frequency of non-disjunction did not differ significantly when comparing the structural homo-TABLE XII Estimated percent recovery of matroclinous progeny. Female genotype Treatment in rads Estimated total number of progeny Matroclinous progeny (Disomic-2 eggs) Frequency in percent 95% Confidence intervals* wow In(2L + 2R)Cy/ap In(2L + 2R)Cy/lt pk on In (2LR)SMl3 Cy/apWM 0 0 0 203,000 55,800 49,900 137 23 34 0.067 0.041 0.068 0.057 0.026 0.047 0.080 0.062 0.095 WOW In(2L + 2R)Cy/ap In(2L + 2R)Cy/lt pk on In(2LR)SMl3Cy/apWCW 2000 2000 2000 71,000 44,700 56,000 159 89 92 0.224 0.199 0.164 0.191 0.156 0.133 0.262 0.245 0.202 wow +/ap +/lt pk on 0 0 70,500 62,600 10 1 0.014 0.002 0.007 - 0.026 0.000 - 0.009 wow +/ap +/lt pk cn 2000 2000 78,500 56,800 43 30 0.055 0.053 0.040 - 0.074 0.036 - 0.075 * Based on the table or corrected fiducial limits of expectations from Stevens (1942). 72 / wow zygotes, +/ap and +/lt pk cn. However, there is a very significant difference in frequencies between the structural homologues and heterologues with the heterologues giving an increase of between five and twenty fold. Contrary to my results, Day and Grell (1966) reported a higher recovery of matroclinous progeny from structural homozygotes, which were homozygous for an inversion, than from structural heterozygotes. They considered structur-ally inverted homologues to be analogous to structurally normal homologues. These values were presented as modified spontaneous values without reference to run size or confidence limits. I w i l l consider this difference later. The recovery of matroclinous progeny following radiation treatment (of 2000 rads) was three times that of the control with the In(2L + 2R)Cy/ap W° W chromosomes (Table XII). I repeated the experiment with a second balancer chromosome, In(2LR)SMI,Cy, and again the radiation produced a greater than two fold increase in the recovery of matroclinous progeny. I questioned the high frequency of induced abnormal meiotic behaviour of structural heterologues. In(2L + 2R)Cy involves a large inversion in each arm which greatly reduces crossing over (Ward 1923). The reduced crossing over between homologues would have increased the probability of their inclusion in the distributive pairing pool (Grell 1962). The spontaneous results suggested nondisjunction was related to the frequency of the entry of the chromosomes to the distributive pairing pool. This was in direct conflict with the results of Day and Grell (1966). For the structurally heterozygous situation Day and Grell employed a Ml normal X and an X with two inversions; In(l)dl-49 and In(l)Bar These two inversions reduce crossing over to twenty-four percent of that normally 73 observed (Cooper 1945). In the structurally homozygous condition both chromosomes carried the inversions. In this case crossing over was estimated to be ninety percent of the normal rate. The ratio of crossing over between structural heterologues and structural homologues i s 24:90 giving a compara-tive crossover rate of approximately 1:4. However, they concluded the struc-tural rearrangements did not lead to any significant difference in the fre-quencies of induced nondisjunction. The two structural combinations, as set up in my experiment, give a much higher ratio of crossing over. In the case of the +/ap W O W (structural homozygotes) there is an exchange between homolo-gues about ninety-nine percent of the time and with the balancer, In(2L + 2R)Cy (structural heterozygotes), crossing over is reduced to less than one percent (Ward 1923). Therefore, the ratio of crossing over between the two combinations i s 1:00 rather than 1:4. If I assume that those chromo-somes which f a i l to undergo exchange pairing enter the distributive pairing pool, then the frequency of chromosomes entering this pool is inversely re-lated to the crossover ratio. The results recorded in Table XII suggest that chromosomes which enter the distributive pairing pool are more sensitive to induced nondisjunction than are exchange chromosomes. Therefore, I would suggest that in order to detect the effect of structural heterozygosity on nondisjunction i t i s important to maximize the number of non-exchange chromo-somes. My results show a four-fold increase in induced nondisjunction when the two chromosomes involved are structurally heterozygous rather than structurally homozygous. It is my impression that this increase reflects the greater availability of chromosomes to the distributive pairing pool. This series of experiments was repeated with one modification, a standard chromosome marked by It pk cn was substituted for the chromosome marked by wow ap . As shown in Table XII this substitution did not lead to any s i g n i f i -74 cant difference using the s t a t i s t i c a l method of Steven (1942). This lent support to the assertion that the events were real, and that the genetic back-ground exerted minimal effect. Although there was no significant difference when the a p W O W chromosome was replaced by the It pk cn chromosome, there was a generally consistent pattern of reduced values for the latter chromosome, implying there may be an underlying genetic influence in both the spontaneous and the induced experiments. Spontaneous And Radiation Induced Patroclinous Progeny: The other major class of recovered progeny were patroclinous: progeny that received both compound-^ chromosomes from their father. In studies on the sex chromosomes, chromosome loss was usually defined as the difference between the number of matroclinous and patroclinous progeny. The patroclinous class was always re-covered in a higher frequency and this was generally attributed to two events leading to their formation, namely nondisjunction and chromosome loss. In the present study I define chromosome loss as the number of recovered patroclinous progeny, the reasons for this w i l l be presented later. The fre-quencies of patroclinous progeny follow the same general pattern as nondis-junction (Table XIII). As with matroclinous progeny, both radiation and structural heterozygos-it y had a marked effect on the recovery of the patroclinous class. Altering the structural homology increased the spontaneous recovery eight fold. Rad-iation increased the patroclinous class by a factor of four for the structural heterologues, which i s similar to the observed increase in matroclinous pro-geny. The structural homologues showed approximately a 20 fold increase with radiation treatment over the controls. TABLE XIII Estimated percent recovery of patroclinous progeny. Female genotype Treatment in rads Estimated total number of progeny Patroclinous progeny (Nullo-2 eggs) Frequency in percent 95% Confidence intervals* wow In(2L + 2R)Cy/ap In(2L + 2R)Cy/lt pk on In(2LR)SMl,Cy/apWOW 0 0 0 203,000 55,800 49,900 220 38 41 0.108 0.068 0.082 0.095 0.048 0.059 0.124 0.094 0.112 wow In(2L + 2R)Cy/ap In(2L + 2R)Cy/lt pk on In (2LR)SMl3 Cy/apWOU> 2000 2000 2000 71,000 44,700 56,000 328 138 169 0.462 0.309 0.302 0.413 0.260 0.258 0.515 0.365 0.351 wow +/ap +/lt pk on 0 0 70,500 62,600 0.013 0.008 0.006 - 0.024 0.003 - 0.019 wow +/ap +/lt pk on 2000 2000 78,500 56,800 223 78 0.284 0.137 0.248 - 0.324 0.109 - 0.172 * Based on the table or corrected fiducial limits of expectation from Stevens (1942). 76 The s t a t i s t i c a l analysis of the patroclinous progeny within groups showed greater variation than was shown for the matroclinous progeny. In the control for the heterologues, the ninety-five percent confidence limits over-lap, but the means do not always f a l l within the limits. It would appear that genetic background plays a greater role in the recovery of patroclinous progeny than the recovery of matroclinous progeny. This confirms the earlier finding of Spieler (1963) who demonstrated higher spontaneous recovery of patroclinous progeny over matroclinous progeny and found both values varied with the strains used. The In(2L + 2R)Cy/ap females gave a higher spon-taneous patroclinous level than either In(2L + 2R)Cy/lt pk cn or In(2LR)SMI,Cy/apWOW. However, this was not the case for the matroclinous progeny. As indicated in Table XIII, significant differences for the recovery of patroclinous progeny exist among the different treated structural heter-ologues. Again, this was not shown for the recovery of matroclinous progeny. This difference, although significant, could be explained by the higher spon-taneous rate for the ln(2L + 2R)Cy/ap W° W females. However, statistics on data could not be corrected for spontaneous values as the determination of ninety-five percent confidence limits required the analysis be done on raw data, not on modified frequencies. Equational Matroclinous Progeny: The third class of chromosomal aber-ration recovered arose from equational (or second division) nondisjunction. These progeny were matroclinous and homozygous either for the recessive marker a p W ° W or for the markers It pk cn. The frequencies for equational nondisjunc-tion are presented in Table XIV. wow The recovery, of equational nondisjunctional products, from ap heterozygous females, could have been the result of a proximal exchange TABLE XIV Estimated percent recovery of equational (second division) nondisjunctions. Treatment Estimated total Number of Frequency 9 5 % Female genotype ± n r a d g number of progeny equationals n ^ Confidence intervals* 0 percent In(2L + 2R)Cy/apJUUJ 0 In(2L + 2R)Cy/lt pk on 0 In(2LR)SMl} Cy/apWCW 0 In(2L + 2R)Cy/apWCW 2000 In(2L + 2R)Cy/lt pk on 2000 In ( 2LR) SMI, Cy/apWM 2000 , / WOW -+/ap 0 +/lt pk on 0 +/apWCW 2000 +/lt pk on 2000 203,000 6 55,800 1 49,900 4 71,000 4 44,700 1 56,000 7 70,500 1 62,600 0 78,500 3 56,800 2 0.003 0.001 - 0.006 0.002 0.000 - 0.009 0.008 0.002 - 0.021 0.006 0.002 - 0.014 0.002 0.000 - 0.012 0.013 0.005 - 0.026 0.001 0.000 - 0.008 0.000 0.000 - 0.006 0.004 0.001 - 0.011 0.004 0.000 - 0.013 * Based on the fiducial limits of expectation (Stevens 1942). 78 between the marker and the centromere followed by reductional nondisjunction. The two events would have produced phenotypically identical progeny as the wow crossovers would carry the Cy_ marker, which is hypostatic to ap To reduce the possibility of exchange, the a p W O W mutant was used because of i t s proximal position to the centromere. A l l putative equational nondisjunctions, from In(2L + 2R)Cy/ap W° W or from In(2LR)SMI,Cy/apWOW, were mated to wild type f l i e s to determine whether any carried the Cy_ gene. None, however, expressed this phenotype. Thus, a l l were classified as equational nondisjunctions. From the irradiated, In(2L + 2R)Cy/ap W O W females, four progeny were phenotypically wild type, indicating a crossover had preceded nondisjunction. In this case, however, the exchange most probably occurred on the l e f t arm between In(2L)Cy and the centromeres. None were found in the controls. This problem did not exist for the It pk cn heterozygotes as the markers flank the centromere, therefore, the recovery of progeny homozygous for a l l three markers identified an equational nondisjunction. Table XIV shows that equational nondisjunction-als were recovered in a l l but one experiment. The frequencies were low and independent of the treatment or chromosomes used according to the ninety-five •percent confidence intervals derived from Stevens' Tables. 79 DISCUSSION Spontaneous And Radiation Induced Nondisjunction: Although the effect of heterozygosity for structural rearrangements on spontaneous and induced non-disjunction would appear not to have been previously intensively investigated, many experiments concerning induced nondisjunction of structural homologues have been recorded. From these studies have evolved a number of theories to explain the phenomena of induced nondisjunction. Early concepts of the mechanisms of nondisjunction are reviewed by Lea (1955). Models have been proposed to explain the mechanisms of induced, but not spontaneous, nondisjunction. It is possible however, that in both cases the mechanisms may be similar. Some workers believed that radiation induced a physiological change in the chromosome which caused centromeres to adhere (Marquardt 1938; Carlson 1941; Sax 1941). These conclusions were derived from cytological observations of metaphase chromosomes since investigators thought they could see chromosomes adhering to one another. Darlington's (1942) explanation for chromosome stickiness suggested that the matrix of nucleic acid of a metaphase chromosome was in a polymerized non-sticky form. However, when irradiated, the matrix through a physiological change became more flu i d and depolymerized resulting in surface stickiness. Marquardt (1938) suggested the change was reversible i f the cells were in interphase at the time of irradiation. Grell e_t a l . (1966) argued that the centromeric region was the target because i t was constant in size, and did not vary with the length of the chromosome. They based their idea on the observation that there was no sig-nificant difference between the nondisjunction of chromosomes of different 80 sizes, for example comparing X duplications and a free fourth with normal X chromosome. Day and Grell (1966) suggested that i f chromosome stickiness were responsible for nondisjunction there would be stage sensitivity corresponding with the degree of condensation of the chromosome. Their results indicated that this was not the case because they found no stage sensitivity for non-disjunction during the f i r s t 12 days of egg laying. Similar findings have been reported by Traut (1970). Although Day and Grell's results suggested that a physiological change was not the cause of nondisjunction their data did not provide evidence for centromeric malfunction. Day and Grell (1966) used a second method to study the nature of induced nondisjunction. They examined the effect of exchange and isosequentiality, and concluded that neither the absence of exchange nor the lack of isosequential-i t y affected the frequencies of radiation induced nondisjunction. However, I find that structural heterologues do cause a significantly increased rate of both induced and spontaneous nondisjunction and suggest that this may reflect the large differences in the frequency of exchanges that could take place between the two different combinations of homologues used in this study. My results, therefore, indicate that the rate of induced nondisjunction is affected by the frequency with which homologues enter the distributive pairing pool. The distributive pairing pool, as defined by Grell (1962), includes a l l chromosomes that failed to undergo exchange pairing. Moore and Grell (1971) found that various factors influence the association of chromosomes which enter the pool; the main one is that distributive pairing is size dependent but homology independent. The fourth chromosomes in female Drosophila rarely undergo exchange and, therefore, are usually members of the distributive pool. In my experiments involving In(2L + 2R)Cy/ap W O W as well as In(2LR)SMI,Cy/apWOW 81 both pairs of homologues were frequent members of the distributive pool. According to the size rule (Grell 1964), the difference in length between the fourths and the seconds would impede the association that leads to the segre-gation of these nonhomologues. The possibility of nonexchange X 1s, however, cannot be excluded. One of the more recent models of nondisjunction has been proposed by Traut (1970). He found a threshold level for radiation induced nondisjunction of the X chromosome. The rate of nondisjunction was not influenced by frac-tionating the treatment. However, when the treatment was protracted, nondis-junction was reduced to one quarter. He suggested that the dependency of nondisjunction on both dose and dose rate made the target hypothesis of centromeric malfunction improbable; radiation may have disrupted one of the processes involved in spindle formation. This probably takes place in pro-phase (stage 7 oocytes). The effect of the irradiation would be a physiolo-gical change requiring a definite number of hits (multiple hit theory). This idea- was supported by Forer (1966) who was able to interrupt chromosome movement during c e l l division by using an ultra violet microbeam to damage the spindle fibers. In light of these models, my results concerning the recovery of matro-clinous progeny, as well as the corresponding number of patroclinous progeny, might be explained as the result of chromosome stickiness, centromeric mal-function or spindle fiber damage. However, none of these models can explain the five fold increase in nondisjunction which occurs when the two chromosomes involved are structurally heterozygous. To explain nondisjunction and chromosome loss, Parker (1969) proposed a 82 model that represents a distinct departure from those described above. His model stemmed from earlier work on the detachment of compound-X chromosomes in the presence of a free-Y_ chromosome (Parker 1954b). Detachments of the compound X involved translocations either with the Y chromosome or with the fourth in stage 7 oocytes. These detachments, he assumed, resulted from "induced-exchanges", which later he called interchanges (Parker 1969). Parker (1969) concluded that two chromosomes formed a quasibivalent (e.g. the compound X and a fourth), which, owing to interchange, separated at meiosis I, while the recovery of the uninvolved fourth suggested random assortment. This supported his earlier findings. If the fourths always segregated owing to homology, but were not affected by interchanges, the free fourth recovered with the capped detached X would be either the sister chroma-tid of the fourth involved in the interchange or a chromatid from the free fourth homologue, and both would be recovered in equal frequencies. His results indicated that the chromatid from the homologue not involved in the interchange was recovered eighty-five percent of the time, while the sister chromatid of the involved homologue only ten percent. Consequently, Parker suggested that interchange usually led to the segregation of the fourth and the X, and that the production of eggs disomic for chromosome _4 were the result of the random movement of the non-involved fourth. In other words, interchange was an event leading to nondisjunction. This was tested further by Parker and Williamson (1970). When compound X females lacking a free Y chromosome were irradiated, the two predicted exceptional classes were recovered, i.e. the detachment product of a compound X and triplo-4 progeny. The latter class was found predominantly in males (269/278) as would be ex-pected i f the X usually segregated from the interchange fourth. Additionally, 83 the frequency of t r i p l o - 4 males, as well as the recovery of detachments of compound X's, increased in a nonlinear manner with dose, indicating that both events occurred as the result of double hits. They argued that, i f chromosome stickiness or centromeric properties were responsible, nondisjunction of chromosome 4_ would be independent of the segregational behaviour of the X chromosome. It should be noted, however, the argument applies only to im-mature stage 7 oocytes (Parker and Hammond 1957; Parker and McCrone 1958; Parker 1969; Parker and Williamson 1970). The addition of a free Y_ to the female genome increased the recovery of triplo -4_ progeny (Parker 1970). Evidently, four members in the distributive pairing pool not only increased the recovery of t r i p l o - 4 progeny, but also altered the sex-ratio recovery of triplo - 4^ progeny (4 males:1 female) as com-pared to the situation of no free Y where'triplo -4 's were recovered predomin-antly i n males (30 males:i female). In the absence of a Y chromosome, induced nondisjunction of the fourth would be primarly the result of interchanges with the X chromosome as described above. The presence of a Y chromosome, on the other hand, provided an alternative chromosome with which the fourths could undergo exchange. Consequently, i t was proposed that such events led to the random movement of the X, which in turn sometimes produced eggs carrying an X chromosome as well as being disomic for chromosome _4. Thus, more disomic-4 female progeny were recovered from mothers with, than from mothers without, a free Y_ chromosome. Parker and Busby (1973) extended the study of interchange mediated non-disjunction by using females with normal X chromosomes rather than compound-X chromosomes. As in the previous study, four chromosomes were involved. How-ever, in this situation the two X chromosomes were independent, joined only by 84 chiasmata, and consequently not regular members of the distributive pairing pool. Parker and Busby (1973) suggest that in such a situation a quasitri-valent may form between the two X's and a fourth i f an X-4 interchange occurs. In their results they estimate that seventy-five percent of a l l diplo-X eggs are nullo 4_. Nondisjunction of the fourths occurred in oocytes in which the X chromosome had simultaneously been lost. Moreover, many of the triplo-4 exceptions showed structural alterations in one of the maternally derived fourths, in that a duplication for the tip of the X appended to the l e f t arm of the fourth had been lost. Interchange mediated nondisjunction of a pair of homologues, e.g. the X's, could involve reciprocal nondisjunction of another pair of homologues, e.g. the fourths. In such an event only one of the nondisjunctional products could be recovered. However, this was d i f f i c u l t to demonstrate with a structurally normal genome. Parker and Busby (1973) found that not a l l diplo-X gametes were nullo 4_. Moreover, owing to the nature of their experiment, nullo-4_ gametes could not be recovered. However, they did recover diplo-4_ gametes that were nullo X. This implied that the reciprocal products, although not recoverable, were generated. Parker, Williamson and Gavin (1974) developed a system for rescuing the diplo-X;nullo-4 eggs by using males carry-ing a compound-4. chromosome. Therefore, the basic premise of their model was confirmed in normal genomes as one-half of the diplo-X eggs produced from irradiated females were nullo k_, while one-fourth of a l l nullo-X eggs were diplo 4_. Furthermore, in those cases of nondisjunction, there was a higher incidence of structural alterations than in the disjunctional gametes. Thus the induced nondisjunction in normal genomes appeared to follow the same pattern as shown in previous studies involving modified genomes. 85 The assortment of interchange chromosomes in mature oocytes (stage 14) appeared to follow different segregational properties (Busby 1971; Williamson 1970a, 1974b). Busby found that C(l)-4 interchanges at this stage did not alter their normal segregational pattern. That i s , both chromosomes involved in the interchange could be recovered at the same pole. This predicted that the recovery of reciprocal translocations in a single gamete was possible. Although none were actually observed, the conclusion was reached by observing that the sister chromatids of the fourths involved i n interchanges were re-covered with the detachment products in 15 out of 23 recoveries. This study indicated that when mature oocytes were treated, interchange chromosomes regularly disjoined from their homologues and consequently interchange did not lead to nondisjunction. In the present research such events would be recovered infrequently as a radiation dose of 2000 rads eliminates most stage 14 oocytes i f the are present (see Chapter I). Quasibivalent Mediated Interchange Resulting In Autosomal Nondisjunction: This explanation, when applied to my own findings, might be viewed in the following way: i f the second chromosomes, owing to structural heterozygosity or the absence of exchange, are members of the distributive pairing pool, one of them, at the time of radiation, may have experienced an induced interchange with, for example, one of the fourths, thereby forming a quasibivalent (Figure 5). This interchange would direct the chromosomes forming the quasi-bivalent to opposite poles, while the remaining second and fourth chromosomes may (or may not) move at random. Parker (1969) reported that those chromosomes which did not go through interchange but were part of the distributive pool 86 FIGURE 4 An example of an interchange forming a quasibivalent between a second and fourth chromosome. The interchange alters the length of the arms involved (2 and 3). Upon separation, the "drag" of the two longer chromatids (1 and 2) of each chromosome orients the centromeres, directing the shorter arms (3 and 4) to the poles of the c e l l . One of the polar or terminal chromatids (3 or 4) w i l l be preferentially incorporated into the func-tional gamete. When mated to compound-2 males the nullo 2 (4) is rescued by a disomic-2 sperm whereas the free arm (3) is lost owing to aneuploidy. I INTERCHANGE ANAPHASE I MEIOTIC PRODUCTS 00 88 FIGURE 5 An interchange between chromosome 2 and chromosome 4 produces a quasibivalent which causes the two heterologues to segregate at meiosis I. The eight classes of gametes depicted represent those predic-ted from the formation of a quasibivalent. MEIOSIS I MEIOSIS II G A M E T E S 1 I i \ n r r — f i . — o — — o A 1 1 Capped 2 nd A1 — o e A12 1 A A 2 A 2 1 Captured 2nd A 2 2 1 i \ n o B 1 1 Capped 2nd U B1 © B 1 2 — u  a±: J J B t r* n B 2 1 Captured 2nd B2 O — 9 B 2 2 tend to segregate from one another. In this manner then, during meiosis I, the two second chromosomes could go to the same pole, and a disomic or nullo-somic egg would be produced. While the involvement of chromosome 4_ is assumed, i t is possible that any other non-exchange chromosome could be i n -volved in an interchange. The above model provides a plausible explanation for the autosomal non-disjunction observed. It could also provide an explanation as to why structural heterologues yield a higher rate of nondisjunction than structural homologues. In the former case, the structural rearrangements would impede crossing over between the two second chromosomes, thus they would more fre-quently be members of the distributive pairing pool. Interchanges between one of the seconds and another chromosome, possibly the fourth, could then result in nondisjunction of the seconds. In this way, structural heterolo-gues would enhance the recovery of nondisjunctional events. Based on the above theory, the following predictions on autosomal non-disjunction can be made and tested. If an interchange took place between a second chromosome and a fourth there would be four possible meiotic products (Figure 4): 1) a normal chromosome 2^  2) a fourth with an attached-2_ arm, 3) a chromosome 2_ with a translocated arm of the fourth, and 4) a normal fourth. In this manner, one of the chromatids of chromosome 2_ would be broken into two parts, one bearing the centromere of the second, referred to as a capped second, and the other bearing the centromere of the fourth, referred to as a captured second. Based on the findings of Parker (1969) I w i l l assume that the non-interchange second and fourth chromosomes sometimes paired and segre-gated. The correctness of that assumption w i l l not affect the basic premise of the model. In Figure 5-A the two second chromosomes could possibly segre-gate to the same pole in meiosis I, yielding the four classes of gametes as shown. When mated to compound-2^  males, the only recoverable gametes would be 2 the disomic two (Al ), which would be rescued by a nullosomic sperm and the 2 progeny would be phenotypically matroclinous, and the nullo-2_ (A2 ), which would be rescued by a disomic sperm and the progeny would be phenotypically patroclinous. The other two gametes would be lethal with a l l possible sperm from compound-2^  males. If the non-interchange second segregates to the opposite pole (Figure 5-B), the gametes would be as lis t e d and a l l would be inviable with sperm from compound-2_ males. If the model applies to autosomal nondisjunction i t predicts the forma-tion of gametes Al"^ (Figure 5) which represents the l e f t arm and centromere of chromosome 2^  capped with an arm of the fourth, and A2"*" which represents the centromere of chromosome 4^  with the right arm of chromosome-2. If the interchange takes place on the l e f t arm, Al"'" w i l l have a capped right arm and A2^ " a captured l e f t arm. Spontaneous and Induced Chromosome Loss: Generally i t has been assumed that nondisjunction of any pair of homologues in females should result in an equal recovery of the reciprocal class, namely nullo gametes which when res-cued by disomic sperm would be phenotypically patroclinous. Any excess in the latter class was thought to arise from causes other than nondisjunction and this excess generally has been referred to as chromosome loss. How can this excess of patroclinous progeny be explained? 92 The earlier work of Mayor (1924) originally showed that the induced re-covery of matroclinous progeny was always less than that of patroclinous progeny; his control values did not reflect this difference. His study on the X chromosome was analogous to my study involving structurally homologous autosomes where I also recovered, from the untreated controls, both classes in equal frequencies, that is the number of- matroclinous and patroclinous progeny were not significantly different (see Table IV). In contrast, the results from structural heterozygotes revealed not only a marked increase in the spon-taneous level of.nondisjunction but also a marked disproportionate recovery of the two classes with the patroclinous invariably the more frequent class. Similarly, the data of Sturtevant and Beadle (1936) with the X chromosome indicated that structural heterozygosity increased the recovery of patroclin-ous progeny relative to the recovery of matroclinous progeny. They suggested the higher recovery of patroclinous progeny could be explained by the occur-rence of four-strand double exchanges within the inversion loop, resulting in the formation of two dicentric bridges. Consequently, the polar nuclei would not receive an X chromosome. Ptashne (1960) suggested that in structurally heterozygous X's, three strand double exchanges, one within the inversion and one between the centromere and the inversion could lead to an anaphase II bridge, resulting in nullo-X eggs. These models not only explain the present results but, are supported by my observations on the balancer chromosomes used. In(2L + 2R)Cy which includes a large inversion both on the l e f t and right arms of chromosome 2^, could result in the formation of nullo-2_ eggs via exchanges in a variety of ways. A four strand double exchange, one with-in each inversion loop, would lead to two dicentric bridges which might result nullo-2 eggs. Similarly, a three strand double, one within the inversion and one proximal to the inversion, would form a dicentric at anaphase II which 93 might also result in nullo-2^ eggs. Table XIII shows that the recovery of spontaneous patroclinous progeny was significantly higher than the recovery of matroclinous progeny (Table XII) for the structural heterologue, In(2L + 2R)Cy. On the other hand, the multiple-inverted chromosome, In(2LR)SMl,Cy, provided a far more effective balancer, which reduced crossing over, and consequently lowered the frequency of nullo-2_ eggs. The data supported this as there was no measurable difference in the spontaneous recovery of matroclinous and patroclinous progeny from the experiment involving this chromosome. Several explanations have been presented to explain induced X~ehromosome loss. I w i l l discuss these as related to my results with chromosome 2. The previous explanation of crossing over, which could explain spontaneous chromosome loss, could also account for induced chromosome loss as radiation increases the frequency of crossing over as specifically shown for chromosome 1_ by Mayor and Svenson (1924) and Muller (1925). Induced crossing over is increased primarily in the heterochromatic region of chromosome 2_ (Yeomans 1972) and this could lead to anaphase II dicentrics as proposed by Ptashne (1960). Traut (1968) suggested that induced chromosome loss in stage 7 oocytes could result from the reunion of breaks in sister chromatids in such a way as to form acentric and dicentric chromosomes. An anaphase II bridge may result in nullo eggs. Previously, Traut (1964) observed that radiation induced chromosome loss did not follow linear kinetics. This implied that both one and two hit events were responsible for chromosome loss. However, this was later shown to be stage dependent. Traut (1968) examined chromosome loss both in stage 7 and stage 14 oocytes. He found that fractionation and protraction of the dose decreased the X-loss frequency with immature oocytes but had no 94 effect on mature oocytes. The linear kinetics obtained with mature oocytes suggested that single hits caused chromosome loss. In immature oocytes the effects of protraction and fractionation suggested a two hit event. Traut postulated that in stage 14 oocytes the chromatids were broken and the lack of a repair mechanism led to eggs with terminally deleted chromosomes. Conse-quently, during the f i r s t division of the zygote a bridge had been formed resulting in the loss of the corresponding chromosome. Day and Grell (1966) looked at the effect of isosequentiality on the frequency of induced chromosome loss. They found no significant difference, and this, as in the case of nondisjunction, may have been due to the small difference in the frequency of crossovers between the two structural config-urations they examined. In the present study there was a direct correlation between chromosome loss and the in a b i l i t y of the chromosomes to engage in exchange. The investigations of interchange mediated nondisjunction led to the conclusion that the nondisjunction of one pair of homologues was not indepen-dent of that of another pair of homologues; one chromatid from each of two heterologues was involved in the interchange (Parker 1970). Similarly, using cytological techniques, Traut and Scheid (1971) recorded that eighteen percent of triplo-4_ progeny were X0 (complete X loss), which is in general agreement with genetic analysis of Parker and Busby (1973) who found twenty-two percent of diplo-4_ gametes were nullo X. The data from Parker and Busby (1973) i n -dicated that the probability of recovering nondisjunction of the fourth chromosomes in a nullo-X gamete, rather than in a gamete with a single X, was two orders of magnitude greater. Their explanation suggests that a common event, an interchange between an X and a fourth, could lead to both 95 results, namely the loss of an X and the gain of a fourth. The interchange between heterologous chromatids could result in both non-disjunction and chromosome loss, one being the reciprocal of the other. How-ever, the equal recovery of matroclinous and patroclinous classes, as predic-ted for reciprocal events, was not realized. An explanation i s required for the excess of the latter case. The work of Novitski (1951, 1967) demonstrated that interchange, which results in asymmetrical dyads, does not result in randomness of recovery. His observations were based on exchanges between two X chromosome inversions, 4 8 8 In(l)sc sc and In(l)sc . The proximal breaks of these inversions are hetero-8 4 8 chromatic, In(l)sc , however, is longer than In(l)sc sc . An exchange between the two inversions results in a heteromorphic dyad of unequal lengths. The 4 8 shorter exchange product bearing the tip of In(l)sc sc i s recovered more g often than the longer chromatid bearing the tip of In(l)sc . In the case where the dyads involved are of different length, the segregation i s dependent upon the orientation imparted upon the centromere. The two longest chromatids w i l l be the last to separate at the metaphase plate. This "drag effect" of the long chromatids w i l l orient the centromeres of the long chromatids so they are innermost in relation to the shorter two chromatids (Figure 4). In Drosophila females, the four products:of meiosis are arranged linearly with one of the two outermost becoming the egg pronuclues (Sturtevant and Beadle 1936, Sonnenblick 1965, also see review by King 1970). Subsequent studies have supported the general premise that the shorter chromatid w i l l be recovered non-randomly. Parker and McCrone (1958), in de-taching the C(X) chromosome recovered 21 Y-chromosome attachments carrying the y + duplication, while they recovered only four of the complementary class of 96 L S XY Y . In those detachments involving only one arm of the Y chromosome, the short arm was recovered more frequently than the long arm (19:9). Parker and Williamson (1970) found that nullo-X;d±plo-r4_ progeny were more common than C(X) , nullo 4_. This would support the recovery, following an interchange event, of the shorter chromatid as the fourth chromosome is much shorter than C(X). Finally, Parker (1974) using an X'4R translocation was able to generate both compound-X and compound-4R chromosomes. Theoretically, they should have been produced in equal numbers, providing both types of compound chromosomes were recovered randomly. The results, however, supported the concept of pre-ferential recovery of the shortest arm of the interchange chromosome as Parker recovered 52 compound-4 chromosomes and only ten compound-X chromosomes. The compound fourths would have been formed from the shorter chromatid of the heteromorphic dyad produced by an interchange and thus would have had a greater probability of being incorporated into the polar nucleus. In summary, the formation of gametes nullosomic for a given chromosome has been interpreted in several ways: 1) they could be the reciprocal event of nondisjunction, and this event may be the result of an interchange between nonhomologues (Parker 1970; Traut and Scheid 1971; Parker and Busby 1973; Parker, Williamson and Gavin 1974); 2) they could arise from four-strand double crossovers which would result in the formation of double dicentric bridges at meiosis I (Sturtevant and Beadle 1936) or from three-strand doubles which would result in single dicentrics at meiosis II (Ptashne 1960), both of which are supported by my results; 3) f i n a l l y , they may be formed by the re-union of sister chromatid breaks forming a dicentric bridge at meiosis II (Traut 1968). The second and third interpretations must be considered with reservation as Novitski (1955) demonstrated that not a l l anaphase bridges give 97 rise to chromosome loss. Proposed Mechanism For The Induction Of Chromosome 2_ Loss: In the present study crossing over may have contributed to both spontaneous and induced recovery of patroclinous progeny. Induced crossing over from radia-tion treatment, specifically within inversion loops, cannot account for the four-fold increase in the frequency of recovery of patroclinous progeny as radiation does not appear to increase crossovers in euchromatic regions to this extent (Roberts 1962; Yeomans 1972). Therefore, interchange may be the largest contributing factor to the recovery of the patroclinous class. The interchange model applied to the present study suggests that the second chromosomes occasionally pair with a nonhomologue, possibly one of the fourth chromosomes. Should an' interchange occur, i t would direct the segregation of the second and the involved (fourth) chromosome to opposite poles. If, in addition, the "drag hypothesis", as suggested by Novitski, is true, the shorter chromatids at anaphase II would more frequently be polar and conse-quently would be included in the egg pronucleus (Figure 4). This would include the uninvolved sister chromatid of the fourth chromosome that engaged in the interchange and the second chromosome interchange chromatid that was capped by the fourth (i.e. bearing the centromere of chromosome 2). If I assume that the other second and fourth distributively pair and orient at random, with respect to the interchange pair, then in half of the events the normal second w i l l go to the same pole as the capped second (Figure 5 Al"'"). The capped second along with the normal homologue w i l l produce an invisible_ segmental aneuploid with any sperm from compound 2^  males. Consequently, the gamete is lost. However, as mentioned by Novitski (1967) the longer chroma-tids (in this study the normal seconds) w i l l not always be excluded from the 98 polar nucleus. When i t is included, and the uninvolved second goes to the same pole, a nondisjunctional event w i l l arise and be recovered as a matro-2 clinous progeny (Figure 5 AI ). As a consequence of interchange, the opposite pole, as shown in Figure 5, would most frequently receive the normal fourth and no second resulting in patroclinous progeny receiving both compound-2^  chromosomes from their father. From this i t may be suggested that nondisjunc-tion and chromosome loss of autosomes are interrelated events arising as a consequence of nonhomologous interchange. The former has a lower recovery rate than the latter owing to the drag effect. It i s also possible that the second and fourth chromosomes not involved in the interchange move at random with respect to one another (i.e. they do not distributively pair). If this were so, not a l l nondisjunctions of the second would be maternally nullo for chromosome 4_. Parker and Busby (1973) and Parker, Williamson and Gavin (1974) suggest that with two free X's in the female there may be quasitrivalent formation, assuming chiasmata formed between the X's with only one free member. However, the large number of variable combinations makes this d i f f i c u l t to examine. Summary Of Nondisjunction And Chromosome Loss: In summary, these two phenomena, namely induced nondisjunction and chromosome loss of the second chromosome, may both be mediated by an interchange event. The formation of a quasibivalent, leading to an interchange, would direct one of the second chromosomes and the other chromosome involved in the interchange to opposite poles. The uninvolved second may segregate at random in which case, some of the time, the result would be the inclusion of both second chromosomes into the egg nucleus. Such a nondisjunctional event would be recovered as a matroclinous progeny. 99 On the other hand, the "drag hypothesis" predicts the preferential recovery of the shorter chromatids from the interchange chromosomes. Assuming chromosome 4_ is the other interchange chromosome, i t would be more frequently included in the egg nucleus. Thus, what is classically referred to as chromosome loss may be a reflection of the more frequent recovery of specific chromatids from an interchange event. The above model explains not only both events but also the high frequency of loss versus nondisjunction. Moreover, the lower recovery of matroclinous progeny may result, in part, from diplo-2_ gametes being nullo 4_. Haplo-_4 progeny from such gametes would only have an eighty percent v i a b i l i t y (Lindsley and Grell 1968), which could make a significant contribution to the decrease in the recovery of the matroclinous class. Such an explanation to account for nondisjunction and chromosome loss, apart from i t s in t r i n s i c appeal, provides a unifying explanation of the events leading to the formation of disomic and nullosomic eggs. More importantly, the model can be tested by developing an appropriate selective system to re-cover the interchange products, namely the capped or captured single arms of an autosome. Equational Nondisjunction: In a l l but one of the experiments, equational nondisjunctions were recovered, although in very low frequencies (Table V). Clearly such events have been shown for the X chromosome in males (Neuhaus 1936, Morgan 1938, Zimmering 1962). However, other than those studies design-ed to examine primary nondisjunction of the X's, no conclusive evidence has been documented for equational nondisjunction in females (see Zimmering 1976). The results in this study clearly demonstrates the occurrence of 100 equational nondisjunction of autosomes. This i s substantiated by two methods: the use of the balancer, In(2LR)SMl,Cy, which produces lethal aneuploids through any exchange and the use of markers that flank the centromere. It is also interesting to find that equational nondisjunctions do not appear to be affected by radiation, structural homology or the chromosome used. CHAPTER IV AUTOSOMAL FREE-ARM FORMATION AS SUPPORT OF A UNIFYING THEORY FOR AUTOSOMAL PRODUCTS OF MEIOSIS ARISING FROM RADIATION INDUCED INTERCHANGES 102 INTRODUCTION In the previous chapter I suggested that nondisjunction of chromosome 2_ could be the result of an interchange with another chromosome in the genome, thereby forming a quasibivalent. The interchange would direct chromosome 2_ and the other chromosome involved in the quasibivalent to opposite poles at meiosis I (see review by Parker and Williamson 1976). If the uninvolved homologue assorted at random, i t may, in a certain proportion of the events, also move to the same pole as the involved chromosome 2_. If, during meiosis II, a chromatid from the uninvolved second chromosome went to the same pole as the non-interchange chromatid from the dyad involved in the interchange 2 (Figure 6, Al ) the resulting egg would be disomic for chromosome 1_ and re-covered as a matroclinous progeny. The reciprocal class, the nullo-2 eggs, would be expected in equal numbers. However, in a l l experiments involving radiation treatment, the recovery of nullo-2^ eggs greatly exceeded the recovery of disomic-2_ eggs. A large contribution to the excess recovery of nullosomic-2^ eggs may have derived from quasibivalents because of the shorter chromatids being oriented in such a way that they were directed into the polar egg pronucleus. Such a segregation would'have resulted in the polar 1 2 1 2 nuclei being as shown in Al and A2 or Bl and B2 of Figure 6. However, as indicated in the discussion of Chapter III other factors may have contributed to the excess recovery of the nullosomic class. The quasibivalent model not only accounts for the greater recovery of patroclinous progeny over matroclinous progeny, but also i t provides an explanation for the dramatic increase of both classes when the two second chromosomes are structurally heterozygous rather than structurally homozygous. 103 FIGURE 6 An interchange between chromosome 2- and chromosome 4 produces a quasibivalent which causes the two heterologues to segregate at meiosis I. The eight classes of gametes depicted represent those predic-ted from the formation of a quasibivalent. 104 MEIOSIS I MEIOSIS II G A M E T E S I ! n — rr o — o A 1 1 Capped 2 nd A1 e — - e A12 1 A A 2 A 2 1 Captured 2nd A 2 2 I o B 1 1 Capped 2nd B1 e B 1 2 J I B B 2 1 Captured 2nd B2 B 2 2 105 Inversions interfere with normal exchange and therefore increase the probabil-it y of chromosomes segregating as a function of the distributive pairing phase of meiosis (Grell 1962). In the distributive phase (which i s also referred to as the distributive pairing pool) an interchange could occur either between homologues or any other member in the pool. Induced exchange between heterologues would direct their segregation which, in turn, could result in the formation of disomic or nullosomic gametes. I hesitate to suggest that only nonexchange chromosomes are involved in interchanges leading to nondisjunction for there is no experimental evidence to support this. Nevertheless, i t i s of interest to find that the greater frequency of induced exceptional events occur in females structurally heterozygous for chromosome 2^. Based on this model several predictions can be made. When the treated 2 2 females are mated to compound males only gametes Al and A2 (Figure 6) can be recovered. The A l ^ and A2^ gametes are lost owing to the in v i a b i l i t y of segmental aneuploids. The f i r s t prediction of the model is the formation of free le f t arms and free right arms of chromosome 2_ as shown in Figure 6. The second prediction i s based on Novitski's (1951, 1967) "drag hypothesis" which predicts the more frequent recovery of the shorter chromatid, a prediction supported by the work of Parker and McCrone 1958; Parker and Williamson 1970; and Parker 1974. The shorter chromatid recovered from ah interchange complex can be identified by determining the origin of i t s centromere. Based on the above hypothesis, and assuming the involvement of the fourths, of the free arms recovered, more should be capped (have the centromere of the second, Al"*" Figure 6) than captured (have the centromere of the other chromosome involved in the quasibivalent, A2"*" Figure 6) (Parker 1969; Parker and Busby 1972). This is not to suggest that a l l l e f t arms would be capped nor conversely that 106 a l l right arms would be captured. If the interchange were to take place on the l e f t arm rather than the right arm of chromosome as shown in Figure 6, the reverse results would occur (i.e. l e f t arms would be captured and right arms would be capped). Assuming the uninvolved second moves at random, the model predicts that some of the free arms should be recovered with their homologue, thereby suggesting that nondisjunction takes place as a result of interchange (Figure 6 AI"'"). Moreover, i f the uninvolved second moves at random, the model pre-dicts the equal recovery of l e f t and right free arms. The recovery of the individual l e f t and right free arms would give support to the theory of interchange as being a mode of generating progeny that are matroclinous or patroclinous for a major autosome. The generation of C(2L);F(2R) stocks by Holm (unpublished data) following the system developed by Grell (1970) enabled the above prediction to be tested. Free-2R stocks consist of a compound-2^  l e f t chromosome, as previously described and two free right arms (Figure 7). In males the homologous free arms normally segregate from one another and the compound-2L evidently moves at random. Gametes of two classes are produced: those bearing the compound-2L and a free-2R and those with only a free-2R (Figure 7). These males allow for the recovery of free arms generated through interchange as predicted by the above model. Interchange product AI"'' (Figure 6) would be rescued by a sperm carrying a right free arm and interchange product A2^ " (Figure 6) would be rescued by a sperm carrying a compound-2L plus a free-2R as demonstrated in Figure 8. Such a system for the recovery of interchange-generated free arms of chromosome _2 made i t possible to test the various predictions presented above. FIGURE 7 The segregation of free-2i? chromosomes is shown for males, two free-2jF? chromosomes segregate while the compound chromo-moves at random. S P E R M A T O C Y T E C ( 2 L ) F ( 2 R ) G A M E T E S 109 1 FIGURE 8 The F(2R) males produce two classes of sperm. Consequently, the only recoverable progeny, when mated to females with normal chromosomes, are those from new free arms generated through inter-changes. The source of the aberrant chromosome products Al"*" and A2^~ is shown in Figure 6. 110 Recoverable Aberrant Female Gametes A 1 1 e — ©-new F2L new F(2R) Gametes from C(2L): F(2R)/F(2R) Males C(2L); F(2R) F(2R) 0 — o Q I l l It also provided the means of identifying the other member of the genome involved in the interchange with the second chromosome. 112 MATERIALS AND METHODS The symbols used to designate free arms were in keeping with Grell (1970) which is somewhat inconsistent with the recommended terminology (Lindsley and Grell 1968). The l e f t and right centric free arms have been designated F(2L) and F(2R) respectively, followed by the appropriate symbols for any genetic markers they carry. The compound 2L; free-2R stock would be described as C(2L);F(2R)/F(2R) as suggested by Grell (1970). One combination of chromosomes in the present study has not been pre-viously described. It includes a free-2L chromosome, a free-2R chromosome and an intact chromosome-2^ homologue. Such genotypes w i l l be symbolized as F(2L)/In(2LR)SMI,Cy/F(2R) (as they a l l have been carried over the SMI balancer chromosome). If the strain i s homozygous for both l e f t and right free arms, i t w i l l be designated by F(2L)/F(2L);F(2R)/F(2R). Furthermore, the hetero-logous portion attached to the free arm w i l l be designated by i t s chromosome number and the arm involved; for example a free-2L derived from an inter-change with the right arm of the fourth would be written F(2L;4R). In the study of free arm recovery, C(2L)lt;F(2R)bw/F(2R)bw males, generated as described by Grell (1970), were mated to In(2LR)SMl,Cy/b pr cn females in bottles. The females were phenotypically Curly (Cy). Each experiment involved approximately 50 bottles with 20 females per bottle. The 60 females were treated with gamma radiation from a Co source for that time necessary to yield a total dose of 2000 rads. The radiation source was as described in Chapter I. The few progeny recovered were of three general phenotypes. Progeny 113 carrying putative new free 2L's had curly wings (Cy) and were of the follow-ing genotype F(2L)b pr/In(2LR)SMI,Cy/F(2R)bw. Progeny with new free 2R's expressed two possible phenotypes: lt^, which were genotypically C(2L)It;F(2R)bw/F(2R)cn or +, which were genotypically C(2L)lt;F(2R)bw/  F(2R)Dp(2L)lt ,cn. The last class was the result of an interchange between the l e f t arm of chromosome 2, dist a l to the It locus, and some other chromo-i some. Thus F(2R)'s that carried a duplication for the l e f t arm were immediate-ly identifiable as being capped rather than captured (compare product AI"'" to A2^ in Figure 6). Many of the + f l i e s were s t e r i l e , some of which expressed the phenotypes of either metamales (X/Y;2/2/2;3/3/3;4/4/4) or intersexes (X/X;2/2/2;3/3/3;4/4/4). The phenotypically classified free arms were analysed in the following manner. Progeny with a putative F(2L) were mated to f l i e s carrying In(2LR)SMI,Cy/ln(2LR)bw^\ a double balancer for chromosome 2. From each line crosses were made between F^ Cy_ males and females to test the v i a b i l i t y of the homozygous F(2L). In an attempt to determine whether F(2L)'s were capped, approximately 2000 virgin females from each F(2L)b pr/In(2LR)SMI,Cy/  F(2L)bw line were mass mated to F(2L)nub b pr/F(2L)nub b pr;C(2R)rl cn males. Rare nondisjunctional oocytes carrying only newly generated F(2L) chromosomes were rescued by F(2L)nub b pr;C(2R)rl cn sperm. If the F(2L) carried a dup-licat i o n of the right arm which covered r l , the progeny would be phenotypi-cally b pr.cn. Such F(2L)'s were then classified as being capped. Progeny that were b pr r l cn could not be immediately classified as a break could have occurred between r l and the centromere in which case the F(2L) would be capped but not identifiable by this test. 114 In one experiment the treated females as well as having the marked seconds, b p r cn/ln(2LR)SMl,Cy, also had marked fourths, one with cubitus interuptus dominant (ci^) and the other with eyeless dominant (ey^); each of these are recessive lethals. This was an attempt to determine the involvement of the fourth in quasibivalent formation leading to the generation of F(2L)'s and F(2R)'s. The recovery of free arms was so low and the v i a b i l i t y of those recovered so poor, that I did not continue the use of these marked fourth chromosomes. As an alternative, the new F(2L)'s were mated to a stock that was hetero-zygous for a chromosome-2^ inversion and homozygous for the recessive marker sparkling-poliert (spaP°"'") on the fourth chromosome. The F^ progeny were then carried through a test cross. The absence of the expression of spa^°^ in any of the progeny carrying the newly generated free arm was taken as evidence for the involvement of the fourth chromosome in the formation of the free arm. Upon analysis, I later learned that the absence of this expression could lead to a false interpretation. The F(2R)'s were mated to C(2L)dp;F(2R)bw/bw. The phenotypically d£ + f l i e s from this cross were mated to determine i f the F(2R)'s were homozygous viable in which case one would recover phenotypically dp cn f l i e s . The use of the C(2L)lt in the male parents, for the recovery of newly generated F(2R), provided the means of immediately identifying any new F(2R)'s that had a break point di s t a l to the It"*" locus on the interchange chromosome and thus were identified as capped. Those with more proximal break points could not be classified. A f i n a l attempt was made to determine the involvement of the fourths in 115 the formation of free arms of chromosome 2. The fourths in the free arm R n strain were made homozygous for the recessive markers c i ey sv (ci - cubitus interruptus, ey^ - eyeless-Russian, sv 1 1 - shaven-naked). The absence of expression of any of the recessive markers was interpreted as an interchange involving chromosome A l l viable lines confirmed to carry free arms were examined cytologically in an attempt to determine, from polytene chromosomes, the break points on chromosome 2_ as well as on the other chromosome involved in the interchange. 116 RESULTS Consistent with the prediction of the quasibivalent model, both F(2L) and F(2R) chromosomes were recovered (Table XV). The theory of equal recovery obviously was not realized in any of the three experiments and the accumulated results show an approximate five-fold greater recovery of F(2R)'s relative to F(2L)'s. The new F(2L)'s were analysed in various ways in an attempt to determine the origin of the centromere and thereby to classify as capped or captured. It was thought that the a b i l i t y to render the free arm homozygous might be a preliminary means to determine whether the F(2L) was capped. Table XVI shows that seven of the ten F(2L)'s recovered were homozygous viable. However, this i s not a particularly reliable test as F(2L)V17 was not homozygous viable and yet by another means was shown to be capped. F(2L)'s were then tested for a duplication of the right arm by constructing a stock with a F(2L) of known properties (F(2L)nub b pr, obtained from Oak-Ridge Laboratories) and a compound-2R homozygous for r_l and cn. The _rl locus has been shown to occupy a position within right heterochromatin of chromosome 2_ (Hilliker and Holm 1975). The absence of a _rl phenotype revealed that the break point, resulting in the formation of F(2L), was dist a l to the r l + locus in the right arm and thus confirmed that the F(2L) was capped. Of the six F(2L)'s that could be tested in this manner, that i s they were successfully incorporated into a C(2R) line, only one carried a r l + duplication (Table XVI). A third approach to analysing captured vs. capped F(2L)'s was to deter-mine i f any of the interchanges involved the X chromosome. The X chromosome being acrocentric implied that, i f "it were involved in an interchange, the TABLE XV Number of induced F(2L)'s and F(2R)'s recovered from In(2LR)SMl,Cy/b pr on or In(2LR)SMl3Cy/Df(2R)M-S2 females mated to C(2L)VE2}lt;F(2R)bw/F(2R)bw males. Number of Number of genotype p ( 2 L ) p ( 2 R ) In(2LR)SMl3Cyfi> pr on 7 20 In(2LR)SMl,Cy/df(2R)M-S210 1 8 In(2LR)SMl,Cy/b pr on ' 2 19 TABLE XVI Genetic and cytological properties of induced free-2L chromosomes. Code Dp(2E)rl+ Homozygous viable Interchange . Chromosome Break points *** Capped vs. captured V2 2 0 * + X hetero;hetero capped ** V9 1 - + X hetero;hetero capped ** V17 - - X hetero;hetero capped ** V46 + + 4 hetero;hetero capped ** V7 0 + X hetero;hetero capped ** V9 2 - + X hetero;hetero capped ** V31 - + 4 hetero;hetero capped ** W18 0 - unknown hetero;hetero unknown **' Y32 - + 4 hetero;hetero unknown Y42 0 - 2L 44C5-6;23D3-6 capped ** * An "0" indicates the line : was not tested. ** Basis for the classification is given in the text. *** Heterochromatic break points were assumed tin in polytene chromosomes. based on the cytological observation of unaltered euchroma-**** This free-2L chromosome, an example of a captured generated from an interchange of the Df(2R)M-S2 chromosome, rather than a capped free arm. appears to be 119 product would most lik e l y carry the centromere of the second. Products of interchange between chromosome _2 were detected in two ways. The discovery of F(2L)'s that could be made homozygous only in females provided a direct genetic method of determining the involvement of the X chromosome. A second method was employed for those free arms which could riot be made homozygous. Males of the genotype F(2L)b pr/ln(2LR)SMl,Cy/F(2R)bw, were mated to In(2LR)bwV1/In(2LR)SMI,Cy females. This cross should have given rise to progeny of three phenotypes: Cy bw V\ Cy and bwV~*" in both the male and the female progeny. The presence of only Cy bw^ males and only Cy_ or bw^ females was taken as confirmation of an interchange between the second and the X chromosome (Figure 9). As shown in Table XVI, five of the ten le f t free arms recovered were involved with the X chromosome, and therefore, were probably capped. Four of these interchange products were homozygous viable; one was not. Interestingly, a l l the X-2L interchange chromosomes were recovered from the same experiment. Further classification of free l e f t arms was made through cytological examinations of polytene chromosomes. Nine of the ten appeared perfectly normal, that is no euchromatic rearrangements in the polytenes could be identified, thereby indicating that interchanges took place in the hetero-chromatic regions. This supports the earlier work of Parker and MeCrone (1958) who thought most interchanges were heterochromatic in origin. The one exception in this study had a break in the right arm of chromosome 2_ at 44C5-6 of Bridges' map (see Lindsley and Grell 1968) which was capped by the tip of ZL broken at position 23D3-6. Since the homologue, In(2LR)SMl,Cy, over which the free arm was generated carried an inversion with a break point dist a l to 23D3-6, the interchange was recognized as having taken place between 120 FIGURE 9 Males bearing the 2L-X chromosome produce two classes of sperm recoverable when mated to females carrying standard second chromosomes. Since the X is attached to one arm of the second, i.e. 2L;X translocation, the free arms of chromosome 2 w i l l be recovered from this cross only in females. The male progeny w i l l inherit the standard second carrying the dominant marker Cy. Note that the Cy/Cy class in males are not recovered owing to a reces-sive lethal linked to this inverted chromosome. 121 Male Gametes X F(2L)^ FUR) Cy Female Gametes bw VI Cy Females Male Lethals bwvl Females Cy bwvl Males 122 the arms of sister chromatids. From the ten recovered l e f t fragments, seven were confirmed to be capped rather than captured by one of the means described above, excluding the test for homozygous v i a b i l i t y . Of the remaining three, one was classified as being capped from the test to determine the involvement of the fourth chromo-some in the formation of free arms of chromosome 2_. The fourths in the free arm strain were made homozygous for the recessive R n markers c i ey sv . The absence of expression of any of these markers in the F(2L) strain indicated the involvement of the fourth in the interchange. Such was the case with chromosomes F(2L)V31, F(2L)V46 and F(2L)Y32, although several anomalies were noted in the progeny from the stock carrying F(2L)V31. From this stock twenty-six percent (73/277) of the progeny did not express the n R more dist a l marker sv but did express the more proximal markers c i ey . A subsequent cross between males and females carrying F(2L)V31 and heterozygous R n for the c i ey sv also yielded anomalous results. In this case twenty-nine percent (23/79) did not f a l l into the expected classes: 11/23 expressed only c i ey^, 4/23 only expressed ey^ sv11, 7/23 only c l and 1/23 expressed ey^. Such results might be explained by position effect as has previously been shown for cl (Parker 1965, 1967), although such events have not been reported R n previously for ey and sv . Assuming these unusual progeny are the result of position effect would imply that the euchromatic component of chromosome 4_ had been transposed to a new heterochromatic site, i.e. the 2R heterochroma-ti n , and thus the F(2L) was probably capped. The F(2R)'s were grouped into those not carrying the l t + duplication from the right arm and those with the duplication. Of the 15 fragment rights 123 without the duplication (Table XVII), 13 were f e r t i l e and therefore analysed. Of these 13, seven were homozygous viable. The only method of determining whether they were capped or captured was by cytological means. A l l had normal polytene chromosomes and thus the other interchange member could not be identified. Again the break point of the right free arms as well as that of the interchange heterologue must have been heterochromatic. Because of the lack of a genetic system and the absence of cytological alterations in the polytene chromosomes of these fragments they could not be classified as capped or captured. By virtue of the fact that the remaining progeny (32 in total) carried the l t + marker, located in the 2L_ heterochromatin (Hilliker and Holm 1975), they were putatively classified as carrying newly generated F(2R)'s. This was confirmed upon subsequent matings with f l i e s of the genotype C(2L)dp;F(2R)be/F(2R)bw. Unfortunately, many of this class were sterile and therefore a distinction between free arm formation and some aneuploid state could not be differentiated; a l l such f l i e s were not included in Table XVIII. In general, the v i a b i l i t y of the l t + F(2R) stocks was very low. Many stocks died before they could be tested beyond confirming that they carried a new F(2R) (13 out of 32). Of those tested, a l l 19 were homozygous lethal. Where possible these free arms were also examined cytologically. Of the 12 examined, five appeared normal cytologically, indicating heterochromatic breaks on both members of the quasibivalent. Where the breaks leading to F(2R) formation were in the euchromatin of the l e f t arm of chromosome 1_, and therefore recognizable, the break i n the other chromosome involved in the interchange was invariably euchromatic. The captured euchromatic component, as indicated in Table XVIII, included 2L, 2R, 3L, 3R and 4R. Interestingly, no such euchromatic interchanges involving the X chromosome were identified. TABLE XVII Genetic and cytological properties of induced free-2i? chromosomes not carrying the Dp(2L)lt . Code Homozygous viable Interchange chromosome Break points** Capped vs. captured V8 + unknown hetero;hetero unknown V14 + unknown hetero;hetero unknown V18 - unknown hetero;hetero unknown V30 - unknown hetero;hetero unknown 2 V36 - unknown hetero ;hetero unknown V43 + unknown hetero ;hetero unknown W16 - unknown hetero;hetero unknown Y7 + unknown hetero;hetero unknown Y37 + unknown hetero;hetero unknown 2 Y45 - unknown hetero;hetero unknown 3 Y45 - unknown hetero;hetero unknown Y51 0 0 0 0 3 Y58 + unknown ; hetero;hetero unknown Y60 + unknown hetero;hetero unknown Y683 0 0 0 0 * An "0" indicates the line was not tested. ** Heterochromatic break points were assumed based on the cytological observation of unaltered euchromatin in polytene chromosomes. 125 TABLE XVIII Genetic and cytological properties of induced free-Si? chromosomes carry-ing Dp(2L)lt Code Homozygous viable Interchange chromosome Break points*** Capped vs. captured VI 0* 0 0 capped V2 0 0 0 capped V4 _ 0 0 capped V53 0 0 0 capped V6 - 0 0 capped V7 2 0 0 0 capped V 1 1 A 0 0 0 capped v i i 4 0 0 0 capped V13 - 0 0 capped V32 - 0 0 capped V36 - unknown hetero;hetero capped V37 - unknown hetero;hetero capped V39 0 0 0 capped V44 0 0 0 capped Wl 0 0 0 capped Wll - unknown hetero;hetero capped W14 - 0 0 capped W20 - 0 0 capped W23 - 0 0 capped W32^  0 0 0 capped W41 0 0 0 capped Y 22 - telomere 34D;telomere capped Y8* 0 0 0 capped Y8 Z - 3L 36F;61D capped Y16 - 2R 38E;32C** capped Y28 - 3R 37;99 capped Y34 - 2L 40A4-5;22B8-9 capped Y41 - 4R 35A;102E3-4 capped Y49 0 unknown heterojhetero capped Y55 - 0 0 capped Y58 - 2L 39C;26D capped Y68 unknown hetero;hetero capped * An "0" indicates the line was not tested ** Capped by the arm of In(2LR)Cy,SMl that carries the telomere of 2R. *** Heterochromatic break points were assumed based on the cytological ob-servation of unaltered euchromatin in polytene chromosomes. 126 DISCUSSION The attempt to recover l e f t and right free arms of chromosome 2_was con-ducted to lend support to the theory that quasibivalent interchanges give rise to nondisjunction and chromosome loss. As predicted, free arms were generated and recovered by using the appropriate selective system. The free-2L's, ten in total, f i r s t were analysed to determine whether they were capped or captured. Through the genetic identification of the pre-sence of a duplication for the r l locus, which i s located to the right of the centromere, cytological examinations of salivary gland polytene chromosomes, the genetic recognition of viable interchange products involving the X chro-mosome and position effect variegation, eight of the ten fragments were classified as being capped. The origin of the other two was indefinite. The analysis of the F(2L) chromosomes further included an attempt to determine i f the fourth chromosome was the other member of the quasibivalent. The fourth chromosome recessive marker, spa^°\ was made homozygous in each of the free-2L lines. The absence of expression of this marker in a F(2L) line was expected to genetically identify the fourth as the chromosome involved. Surprisingly, this method indicated that nine of the ten l e f t arm fragments arose from interchange with the fourth, i.e. F(2L;4R). However, five of the nine had previously been confirmed, by genetic means, to have been F(2L;X) interchanges. The fourth, therefore, definitely could not be involved unless each of the fragment formation events had been very complicated multichromosome interchange products. Although such events have been reported (Williamson and Parker 1968) they cannot conceivably explain this inordinately high rate of recovery. It would appear that, as a more li k e l y explanation, spa^°\ like spa, i s suppressed by an increase i n the amount of heterochromatin (Morgan 127 1947). Furthermore, Morgan showed that deficiencies for the heterochromatin of the right arm of chromosome _2 enhanced the expression of spa. As the heterochromatic effects on spa*30"*" appear to be the same as on spa the present findings indicate that the F(2L) chromosomes were capped, as captured seconds would be partially deficient for the heterochromatin of chromosome-2L and carry no heterochromatin of 2R- It seems most probable that a l l the F(2L)'s, involving the X chromosome, were capped by the X. Consequently, these inter-change chromosomes would carry duplications for the 2R heterochromatin but at the same time be deficient for varying amounts of the X heterochromatin. Results from the present study lend further support to this concept as F(2L)Y42 was generated from an euchromatic interchange with i t s sister chroma-ti d , thus i t definitely carried a duplication for heterochromatic-ZR. This free arm suppressed s p a P o 1 . Conversely, F(2L)W18 generated from a Df(2R)M-S210 chromosome, which i s deficient for most i f not a l l of the 2R heterochromatin (Morgan et a l . 1940, H i l l i k e r and Holm 1975), was the only F(2L) which did not suppress spa^ °l In fact, the expression of spaP°^" was enhanced in the presence of this free l e f t arm. Because of the unreliability of spa^ 0^ owing to i t s suppression by excess 2R heterochromatin, four F(2L)'s were tested for their involvement with the fourth chromosome, detected by the in a b i l i t y to recover the expression of three different recessive markers on the fourth chromosome. Of the four lines tested, three were confirmed interchanges between a second and fourth chromo-some, that is the products were 2L-4R. However, in each case the results revealed anomalies. Two of the lines, V46 and Y32, gave the expected classes _ VI . R. n _ j *V1 , . _ . R n , Cy/bw ; c i ey sv , Cy and bw The absence of Cy;ci ey sv and bwV"*";ci ey R sv" served to identify the involvement of the fourth. Although the above classes were expected, one class was missing, namely Cy/bw^ f l i e s 128 that were wild type for markers on chromosome-4. This class should have been heterozygous for the recessive markers having received a wild type fourth from the free arm parent. This result implies that in these two lines there was only one free marked fourth, the wild type fourth being attached to the F-2L. Consequently, the wild type fourths would never be recovered in the Cy/bw^ progeny and thus a l l these progeny would be Cy/bw^ "*";ci ey R sv", receiving only the free fourth from the free arm parent. The question can then be asked: how would a stock be maintained with only a single free fourth? One possible explanation i s that during meiosis the 2L-4R chromosome regularly segregated from the free fourth, as well as from In(2LR)SMI,Cy thus the In(2LR)SMl,Cy chromosome and the free fourth segregated to the same pole. In this way, by mating F(2L)b pr/In(2LR)SMl,Cy/F(2R)bw males and females, the stock would maintain a single free fourth. The third fragment ;(V31) , also identified as an interchange with the fourth, differed from the two previously mentioned VI F(2L) lines in two aspects: 1) this line produced Cy/bw phenotypes i n d i -cating there were two free fourths in the F(2L) line, one wild type fourth and the other carrying the recessive markers c i ey^ sv 1 1, 2) of the F(2L) progeny recovered, twenty-six percent (73/277) were homozygous for only the R n two most proximal markers, cl and ey ; sv was not expressed. The loss of c i and ey could have been due to some type of position effect suppression of the two proximal genes as has previously been reported for c^ L (Parker 1965; Parker 1967; Spofford 1976) or could have been the expression of an inter-im change in which only the 4_R dis t a l to ey was translocated to F(2L). However, cytological studies did not support the second possibility. A further examination;.of this strain, by crossing F(2L)b pr/bw^/F(2R)bw;ci ey R sv" males and females, resulted in the recovery of the two expected chromosome-2_ VI classes, bw and b pr bw (the latter class i s the expression of both homo-129 zygous F(2L);F(2R) chromosomes). None of the regular progeny expressed a l l three recessive markers on the fourth as the F(2L;4R) carried the wild type alleles. However, from this cross seventeen percent (24/143) of the progeny expressed the c i ey^ phenotypes but not the sv11, as found in the previous case, and 12 progeny showed other variations of expression of the three genes: four had only the ey^ sv11 phenotypes, seven had only cjL phenotype and one expressed only the ey phenotype. These findings further point out the necessity of caution in the use of a marked fourth chromosome in the analysis of chromosomal interchanges. These results simply report a strange phenomenon, further experiments are required to better determine the cause of these results. The fourth line tested, W18, indicated the fourth chromosome was not i n -volved in the free arm formation as progeny with the free arm expressed a l l three recessive genes on the fourth. As this free arm was generated from a chromosome deficient for 2L heterochromatin (Df(2R)M-52"'"^) and as the polytene preparations appeared normal and as i t enhanced the expression of spa^ '0"'", this chromosome was possibly a captured F(2L). The model of quasibivalent formation followed by interchange resulting in nondisjunction is supported by the present data. The recovery of F(2L)'s, with the selection system used, required that the homologue of the second involved in the interchange also be included in the gamete with the free arm. In studies of nondisjunction i t is v i t a l that this class include only those cases where homologous centromeres go to the same pole. Thus in this study, as in previous studies using acrocentric chromosomes, nondisjunction is not necessarily synonomous with matroclinous progeny. An interchange can lead to the exchange of centromeres in which case matroclinous progeny would actually be the consequence of disjunction rather than nondisjunction (e.g. B2"^  Figure 130 6). In acrocentric chromosomes, i.e. 4ths and X's, a progeny matroclinous for the X chromosome could involve a normal X and an interchange X. The interchange X may carry the centromere of the 4th in which case the matro-clinous progeny would result from disjunction rather than nondisjunction. With metacentric chromosomes, nondisjunction per se is synonomous with matroclinous progeny. Furthermore, interchange mediated nondisjunction also can not be substantiated unless the chromosome in question (i.e. X or 4th) can be confirmed as being capped. However, from the present study, with the free arms generated from natural metacentric chromosomes, confirmation of the centromeres involved demonstrated that interchanges lead to the nondis-junction of homologous centromeres. In total, 47 confirmed F(2R)'s were recovered of which 32 were definite-ly capped as they carried the l t + marker from the 2L heterochromatin (Hilliker and Holm 1975). The remaining 15 were classified as unknown as they could not be tested for the absence or presence of the centromere of chromosome 2. Thus from the 57 induced fragments produced and tested (both l e f t and right arms), seventy percent (40/57) were confirmed as being capped, i.e. they carried the centromere of chromosome 2. In conjunction with the Novitski (1951, 1967) "drag hypothesis", the quasibivalent model of nondisjunction predicts a preferential recovery of the shorter chromatid of that chromosome 2_ involved in the interchange. It follows, therefore, that the excess recovery of patroclinous over matroclin-ous progeny, which was previously called chromosome loss, i s a function of the preferential recovery of the shorter chromatid as a direct result of quasibivalent interchange. This would also imply the preferential recovery of a capped rather than a captured interchange chromatid. The results, 131 obtained in the present study, are consistent with this model. I do not mean to suggest, however, that the "drag" effect is an exclusive mechanism for the production of excess nullo-2_ eggs, but rather that i t contributes to this class of exceptional products. In view of the quasibivalent model, my original prediction was that the number of l e f t and right free arms generated would be equal. This definitely was not the case as only ten lefts were recovered as compared to 47 rights. There are two possible explanations for this discrepancy. The f i r s t suggests either that the other chromosome involved i n quasibivalent formation prefer-entially pairs with the l e f t arm or that interchanges occur here more often. The second possible explanation is that the quasibivalent frequently results in the formation of a trivalent. This would include both seconds and the heterologous chromosome involved in the interchange. The latter explanation agrees with the results of Williamson (1974a) who reported preferential segregation of the X and Y chromosomes in a compound-X-bearing female, even though the X had been involved in an interchange with chromosome-4_. In the present study the two second chromosomes regularly would be members of the distributive pool as the balancer, ln(2LR)Cy,SMI, would prevent crossovers and according to the size rule would pair distributively. An interchange could result in the formation of both l e f t and right free arms, but as the homologues would usually disjoin, more gametes would contain only a free arm (either l e f t or right) than a free arm accompanied by a normal homologue. The present study supports previous observations that the interchange chromo-somes separate at meiosis I (Parker 1969; Busby 1971; Parker and Williamson 1970). However, i t is possible that the two second chromosomes pair prior to an interchange between one of the second chromosomes and some other heterolo-132 logue. The pairing of the homologues possibly guarantees their separation at meiosis I and, similarly, the interchange causes the two interchange products to segregate. If segregation were to occur, gamete Bl"^ would be recovered more frequently than gamete AI"*" (Figure 6). Thus when mated to C(2L);F(2R)/F(2R) more right free arms than l e f t free arms would be recovered, bearing in mind the interchanges leading to free right arms must take place to the l e f t of the centromere in chromosome 2_, not as depicted in Figure 6 where the breaks are shown on the right arm. The type of sperm available from the C(2R);F(2R)bw/F(2R)bw males could recover either induced l e f t or right free arms. Nevertheless, the rescuing of a l e f t free arm can only occur i f i t is recovered with a normal homologue as well. But i f the second chromosomes usually segregate, the higher incidence of recovery of right free arms over l e f t would be explained. In previous studies, in the absence of a free Y chromosome, detachment products of the compound X usually involved chromosome k_ (Parker and Hammond 1957; Parker 1969; Parker and Williamson 1970). As the 4ths are normally members of the distributive pairing pool they would be likely candidates for interchange with other chromosomes entering the pool. From a total of 57 fragments, the other chromosome involved in the interchange was identifiable by genetic or cytological means in 16 cases. Of these only four were con-firmed as being the fourth. The other identifiable interchanges included the l e f t arm of the X and both l e f t and right euchromatic portions of chromo-some 2_ and chromosome 3^. However, a very important observation was made concerning the break points. In the X fragments, in which the break i n chromosome 2_ was euchromatic, the break in the capping segment was also euchromatic. Similarly, in the 27 cases in which the chromosome-2^ free arm break point was evidently in heterochromatin, the capping segment also 133 appeared to have i t s break in heterochromatin as the polytene chromosomes appeared normal. Moreover, 14 of these 27 interchange products were homo-zygous viable. This finding suggests that possibly euchromatic breaks can rejoin only with euchromatic breaks and likewise, heterochromatic breaks can rejoin only with heterochromatic breaks. While no exceptions to this last point were witnessed in the present study, one such exception, a quasirecip-rocal translocation involving C(2L);C(2R) and a standard third was recovered by H i l l i k e r and Holm (1975). Further studies are required to provide more information about the temporal and structural organization of chromosomes during meiosis. Because of the previous observations, i t can be implied that in those free arms in which normal cytological preparations were observed the other chromosome involved was probably the fourth. Involvement with any other chromosome except an entire X (of which five were identified) would be lethal. Many of the euchromatic breaks provide evidence that very large dupli-cations of the autosomes are tolerated. I also observed, in the crosses between normal females and free arm stocks, almost as many pupae cases as in normal crosses, although only the rare nondisjunctional progeny eclosed. 134 General Discussion Of Interchange: One of the overall objectives of the present study was to recover and analyse aberrant products of chromosome _2 induced during meiosis. The total number of induced aberrations in chromo-some 1_> recovered when mated to compound-2^  males, was very high and included four classes of gametes: nullo 2_, diplo 2_, compound-2L and compound-2R. For example, In(;2L 4- 2R)Cy/ap W O W females treated with 2000 rads produced 529 recovered aberrant events. Those recovered represent approximately a total of one quarter of those induced as only one quarter of the sperm were capable of rescuing each of the four aberrant classes. Thus, the estimated number of actual aberrations generated was 2,116. This represented 2.98 percent of an estimated total number of 71,000 progeny. It must be noted, however, that this represents only the damage induced and recoverable for chromosome _2 by using the compound-2^ selective system. The total chromosomal damage to the entire genome based on the recoverable aberrant events that were recovered from chromosome 2^  would be considerable. The above estimates do not take into consideration translocations which are not recoverable except possibly with males bearing free arms. Nevertheless, even then many are lost due to segmental aneuploidy. The results suggest there are two main factors affecting the recovery of induced aberrations. The most important variable is the isosequentiality of the chromosomes involved. Alterations such as inversions greatly increase the recovery of some exceptional meiotic events. Although of less s i g n i f i -cance, but s t i l l a recognizable factor, i s the genetic background of the stocks used. The It pk cn stock, for example, was less sensitive to the effects of radiation in regards to the four classes of apparent products re-covered, as well as the rate of induced recessive lethals, supporting the results of Sobels (1971) on stock radiation resistance or sensitivity. Furthermore, I believe that three of the classes of simultaneously recovered progeny arise from a common mechanism of induction; a mechanism which also accounts for free arm formation. Compound chromosomes, matro-clinous progeny, and some patroclinous progeny as well as free arm formation, require two independent breaks occurring at the level of the chromatid. The location of the breaks, the position of the chromatids, and the method of reconstitution determine the recoverable events. It is assumed that many induced breaks w i l l be reconstituted to their original condition. The events recorded here are the result of abnormal associations of these breaks, i.e. interchanges. Intrachromosomal Breaks: Abnormal reconstitution could occur between breaks within a chromosome, resulting in a variety of rearrangements. The breaks could lead to sister compound chromosome formation, providing one break occurred in each chromatid, and the breaks were on opposite sides of the centromere. Two breaks followed by such reconstitution leading to the formation of a compound chromosome could be referred to as an interbrachial interchange as suggested by Williamson (1969) in the formation of isomarked Y chromosome fragments. Although not observed in these experiments, intrachromatid breaks resulted in ring chromosomes (Hilliker, personal communication) by the re-joining of breaks on opposite sides of the centromere of a single chromatid. The heterochromatic l e f t and right breaks joined forming a heterochromatic ring, cytologically identified in patroclinous progeny. If the two breaks occur on the same side of the centromere, two conse-136 quences are possible. The adjacent arms could rejoin forming an acentric and a dicentric isochromosome. This would involve a dicentric bridge at anaphase II, possibly resulting in a nullo egg recovered as a patroclinous progeny (Traut 1968). The other alternative i s the rejoining of opposite arms form-ing two normal chromosomes bearing either small duplications or deficiencies, i.e. an induced sister chromatid crossover event. Interhomologue Breaks: Simultaneous breaks also may occur in chromatids of homologues. The repair of these breaks can result in non-sister compound formation in a manner similar to that previously described for sister compound chromosome formation. If'the homologues are heterozygous for a paracentric inversion and i f the breaks result in an induced heterochromatic exchange accompanied by an exchange within the inversion, a dicentric bridge would occur at meiosis II which might lead to the production of a nullo 2_ egg (Novitski 1955; Ptashne 1960). Induced exchanges did occur as evidenced by the recovery of four wild type progeny from In(2L + 2R)Cy/ap W O W females. None were observed from the In(2LR)Cy,SMI/apWOW females presumably as induced crossing over would only lead to segmental aneuploidy owing to the peri-centric nature of the inversion. The results (Table V) indicate a s i g n i f i -cantly lower recovery of induced patroclinous progeny from the latter class, lending support to this notion. Although, Sturtevant and Beadle (1936) demonstrated double dicentric bridges of the X chromosome lead to nullo eggs recovered as patroclinous progeny, Novitski (1952) found this was true only for acrocentric chromo-somes. When the X chromosome had an arm attached (i.e. from the Y chromo-some) fewer patroclinous progeny were recovered. Novitski (1955) speculated that the formation of a submetacentric chromosome changed the strength of 137 the centromere such that the dicentric bridge would fracture resulting in fragments. This would lead to dominant lethals rather than nullo-X eggs. In view of Novitski's finding with a modified X (by the attachment of a Y_) breaks between homologues could result in dicentric bridges as well as non-sister compound chromosomes. The dicentric formed in this fashion would result in strong centromeres which would possibly break the bridge. When mated to males carrying compound-2^  chromosomes, the bridge products would be lost. Interchromosomal Breaks: A third type of event could occur between nonhomologous chromosomes and are referred to as interchanges by Parker (1969). As suggested by Parker, interchanges between two nonhomologues of unequal size produce heteromorphic dyads resulting in chromatids of different length attached to the same centromere. As a consequence of interchange, the heterologues segregate at meiosis I, which would result in nondisjunction providing the homologue of one of the interchange chromosomes goes to the same pole. The heteromorphic dyad supposedly orients chromatids of different lengths in such a way as to place the shorter chromatid at the pole as suggested by the "drag hypothesis" (Novitski 1951, 1967) and thus increases the probability of i t s incorporation into the pronucleus. Assuming, in the present study, that the fourth chromosome was most frequently involved with the second in the formation of a heteromorphic dyad, the results should most often result in the production of patroclinous progeny . Interchanges: If I extend the use of the term interchange to include breaks in any pair of chromatids, homologues or nonhomologues, i t could account for the various meiotic products recovered in immature oocytes, depending upon the manner in which the broken chromatids are rejoined. Such 138 interchanges could result in both the formation of sister and the formation of non-sister compound chromosomes. The interchange, i f between heterologues, could lead to the nondisjunction of a pair of chromosomes and be observed, in the present study, as matroclinous progeny. The reciprocal product of the interchange between heterologues would produce an equal number of nullo eggs, observed as patroclinous progeny, but i f the dyad so formed is heteromorphic, i.e. chromatids of different length, i t would result in the greater recovery of patroclinous progeny because of the preferential inclusion of the shorter chromatids. Interchange could also result in dicentric formation which may also contribute to this last class. Although the evidence is minimal, interchanges may also be responsible for induced crossing over in hetero-chromatin. Finally, interchanges can result in the recovery of a half-translocation (which now more suitably might be expressed as one-half the product of a reciprocal translocation) as observed in the free arm studies. The generation and analysis of interchange events point out four major features: 1) i f i t is assumed that nondisjunction and chromosome loss are interchange mediated then interchange is increased in the presence of struc-tural heterologues, 2) i t also follows from the results of this study that the genetic backgrounds affect the frequency of interchange, 3) probably any interchange can be induced providing an appropriate selective system is available to recover the event, and 4) interchanges appear to occur primarily between euchromatic-euchromatic or heterochromatic-heterochromatic breaks. CHAPTER V AN ASSAY FOR THE PRODUCTS OF CHEMICAL MUTAGENESIS 140 INTRODUCTION Chemically induced chromosomal aberrations are of interest to geneticist as well as environmentalists. The increasing numbers of new chemicals appear-ing in the environment increases the need for sensitive and reliable assay systems to detect chemically induced genetic damage. The identification of mutagens, the genetic manifestation of their action, and the frequency of i n -duced aberrancies are a l l specific aspects to be considered. Chemically induced genetic damage may range from toxic lethality, to much more subtle changes such as specific substitutions of the nucleotides in DNA. The former is a very gross analysis and eliminates a l l levels of damage below total lethality. Different chemical agents have been found to vary greatly in their toxicity as compared with their mutagenic activity. For example, ethyl methane sulfonate (EMS), although highly mutagenic, has been reported to have l i t t l e effect on v i a b i l i t y of bacteria (Loveless and Howarth, 1959). Such findings pose two main questions concerning assays for induced genetic damage: what organism should be used, and what level of damage should be used as the criterion for mutagenicity? Before these two questions can be answered, i t is necessary to look at characteristics of a good assay system. It should 1) be economical 2) be reliable 3) yield relatively fast results 4) depend upon objective interpre-tation 5) cover a wide range of damage, and 6) involve an in vivo system. Although various eukaryotic organisms could be used, a l l having certain advantages and disadvantages, the f r u i t f l y , Drosophila melanogaster, best f i t s the requirements for an i n i t i a l assay system for environmental mutagens. Flies can be reared more economically and in larger numbers than most other 141 multicellular organisms. Their short l i f e cycle, of approximately 12 days, yields rapid results, and the wealth of accumulated information on Drosophila greatly fa c i l i t a t e s interpretation of the results. Finally, the simplicity of the Drosophila genome, is also an asset as there are only four linkage groups. The second question, the level of damage used as a criterion, must be concerned with the genetic material. A good assay system must be able to detect a wide range of damage. Ideally, i t should reveal point mutations, chromosomal aberrations and abnormal meiotic behaviour of the chromosomes. Parker (1968), Sobels (1971) and Abrahamson and Lewis (1971) have re-viewed the various methods of determining genetic damage in Drosophila and the advantages of this organism. Parker (1968) suggests two distinct approaches towards the study of induced radiation damage during meiosis: either measure the frequency of the damage or gain insight into the mechanism of meiosis. Most of the assay systems thus far developed have focused on the X and/ or Y chromosome (Valencia 1970). As indicated by Parker (1968), studies may f a i l to detect induced genetic damage not because i t did not occur but because an appropriate sective system, which avoided dominant lethality, was not available. This problem was more readily resolved with the sex chromo-somes than with the autosomes, and for this reason they have received the greatest attention. With the use of the multiplier and appropriate selective systems, I was able to extend experimental approaches that previously had been restricted to the sex chromosomes. I was able to detect and quantify simultaneously four 142 classes of structural or numerical autosomal aberrations during meiosis in females as well as recessive lethals via means of the internal dosimeter test. The multiplier system served as an internal dosimeter by testing for recessive lethals. Having developed an assay system for autosomes, using gamma radiation as a mutagenic agent, I tested the system with known chemical mutagens to deter-mine whether or not they were radiomimetic. Two known mutagenic compounds were tested: ethyl methane sulfonate (EMS) and mitomycin C (MC). The former is a monofunctional alkylating agent and the latter may act either as a monofunctional or as a bifunctional alkylating agent. The molecular mechanisms of these chemicals are not completely known, but i t is thought that they react at the N-7 position of guanine, resulting in the excision of the nitrogenous base from the DNA molecule (for a review see Drake 1969; Freese 1963). It is not the purpose of this paper to examine the effect of alkylating agents at the molecular level, but rather to determine the general effect of these chemicals on chromosomes. The type of damage caused by the alkylating agents should be elucidated by the recovery of some or a l l of the four classes of aberrations discussed in previous chapters. One of the questions concerning alkylating agents is their a b i l i t y to induce chromosomal breakage. Fristom (1970) reported the recovery of EMS induced breaks. Additionally, Williamson (1970b), treating male Drosophila, reported the recovery of two deletions within the Y chromosome but Lim and Snyder (1968) recovered only one translocation out of 2,000 sperm tested. On the other hand, Cattanach and Williams (1971) did not recover in mice, EMS induced translocations or fragments when they treated spermatogonia. How-143 ever, Cattanach, Pollard and Isaacson (1968) did recover EMS induced breaks in postmeiotic cells of mice. More recently, H i l l i k e r (1976), using EMS, induced 113 mutations in the heterochromatic region of chromosome 2_ of Drosophila melanogaster. Analysis of these mutations indicated that none were deletions. This finding was in keeping with the results of Lim and Snyder (1974), who found no deletions out of 82 induced mutations in the zeste-white region of the X chromosome. Further evidence indicating the in a b i l i t y of EMS to produce chromosomal breaks was reported by Schewe ej^ a l . (1971a), who failed to recover X-Y or Y-Y interchanges when treating female Drosophila. In general, the results from mutagenic studies on EMS indicated that i t does not cause chromosomal break-age, although the evidence was not conclusive. MC, a bifunctional alkylating agent, appeared to cause chromosomal breaks. Natarajan and Schmid (1971) observed chromosomal aberrations induced by MC in Chinese Hamster c e l l s , mainly in constituative heterochromatin. Similar results obtained from studies in plants have been reported by Rao and Natarajan (1967). Furthermore, Schewe et a l . (1971b) recovered X-Y and Y-Y, MC-induced interchanges both in female and in male Drosophila. Several studies have been directed towards the effect of alkylating agents on nondisjunction. At the time, nondisjunction was usually considered to be independent of any interchange event, and the results like those from studies on chemical induced breakage, have been inconclusive. Hayashi and Suzuki (1970) reported increased nondisjunction of the X chromosomes in females of Drosophila when treated with MC, which is in contrast with the results of Schewe et^ a l . (1971b). Moreover, Schewe et. a l . (1971a) did not 144 find EMS to have any measurable effect on nondisjunction. Because of the interchange nature of the aberrant meiotic products re-covered from radiation treated females, the assay system provides a means of determining whether the alkylating agents are capable of causing similar effects. As well, the comparative recovery of the various classes should confirm whether or not the aberrant events are exclusively break induced. The assay system provides a means of examining more extensively some of the effects of known mutagens as well as testing new environmental chemicals. The present system, because of i t s many advantages, should be able to extend studies on mutagenic agents. For example, i t allows mutagenic studies of a broader nature to be conducted on autosomes, rather than being restricted to sex chromosomes. It also allows for the simultaneous recovery of four different classes or progeny. The selective exclusion of a l l regular meiotic products fa c i l i t a t e s the recovery of a relatively larger number of aberrant eggs without massive and laborious screening techniques. Finally, the multi-plier system enables the calculation of the frequency of each event and acts as an internal dosimeter to lend credence to negative results. r 145 MATERIALS AND METHODS The stocks used for this study and the mating protocol was the same as that presented in Chapter I. Virgin females, aged three days, were fed in bottles for 24 hours on a solution of 0.025M EMS in one percent sucrose (Lewis and Bacher 1968) or 125ug/ml MC in one percent sucrose, then allowed to recover for 24 hours before being mated to compound-^ males. A recessive lethal test on the X chromosome (as described in Chapter I) was conducted for each experiment. The exceptional progeny were not tested be-yond confirmation of their original classification. 146 RESULTS As recorded in Table XXI, the mean number of progeny per v i a l was signi-ficantly reduced by the administering of alkylating agents, the reduction being greater for MC than for EMS. Although the mean number of f l i e s per multiplier v i a l differed, the variances were homogeneous using Bartletts test of homogeneity of variance for different sample sizes. Table XX contains the frequencies of the four classes of aberrant chromosome-2_ products, as well as the frequency of recessive lethals. These frequencies were calculated using the data recorded in Table XIX. Consistant with a l l previous experiments, fewer aberrations were recovered from the It pk cn stock, suggesting that their resistance to radiation possibly also applies to chemical mutagens. This i s in agreement with Browning (1970) who found that various strains of Drosophila showed different sensitivies to EMS induced recessive lethals. The system definitely indicates differences between the monofunctional and bifunctional alkylating agents. While EMS failed to induce compound chromosomes, one was recovered from the In(2L + 2R)Cy/ap W O W strain and nine from the +/lt pk cn strain when they were treated with MC (Table XIX). More compounds were recovered from the structural homozygotes than from inversion heterozygotes; a finding that was in keeping with those made in the radiation studies. Interestingly, the nine compounds generated from +/lt pk cn were a l l sister-strand attachments of the It pk cn chromosome, one l e f t and eight rights. Each of these compounds represented independent events as each was derived from a different female. TABLE XIX Number of progeny recovered as exceptional chromosome-2 products of meiosis and the number of Z-linked recessive lethals obtained in corresponding multiplier tests. Number of exceptional chromosome-2 products Estimated c ; c  total Compounds Matroclinous Patroclinous Equational Recessive Female genotype Treatment , f number or o r 2 R . ( D i s o m i c - 2 (Nullo-2 nondisjunctions lethals * progeny . . v eggs) eggs) In(2L + 2R)Cy/apWCW 0 203,000 0 137 220 6 13/1832 In(2L + 2R)Cy/lt pk an 0 55,800 0 23 38 1 2/941 In(2L + 2R)Cy/apWOW EMS 66,600 0 87 114 7 20/837 In(2L + 2R)Cy/lt pk on EMS 36,300 0 26 19 0 23/815 In(2L + 2R)Cy/apW™ MC 24,700 1 34 29 7 17/717 In(2L + 2R)Cy/lt pk on MC 24,000 0 16 8 3 8/603 +/lt pk on 0 62,600 0 1 5 0 1/977 +/lt pk on EMS 42,700 0 13 5 5 20/456 +/lt pk on MC 19,400 9 22 5 0 22/615 * Number of Z-linked lethals per total number of tested chromosomes obtained from the multiplier test. See text for f u l l description. TABLE XX Estimated percent recovery of the four classes of progeny that arose from aberrant chromosome-2 pro-ducts of meiosis. Female genotype Treatment Estimated frequencies of aberrant chromosome-2 products (in percent)* , Recessive Compounds Matroclinous Patroclinous Equationals lethals** WOW In(2L + 2R)Cy/ap In(2L + 2R)Cy/lt pk on wow In(2L + 2R)Cy/apu In(2L + 2R)Cy/lt pk on In(2L + 2R)Cy/apWOW In(2L + 2R)Cy/lt pk on +/lt pk on +/lt pk on +/lt pk on 0 0 EMS EMS MC MC 0 EMS MC 0.000 0.000 0.000 0.000 0.004 0.000 0.000 0.000 0.046 0.067 0.041 0.131 0.072 0.138 0.067 0.002 0.030 0.113 0.108 0.068 0.171 0.052 0.117 0.033 0.008 0.012 0.026 0.003 0.002 0.011 0.000 0.028 0.013 0.000 0.012 0.000 0.710 0.212 2.389 2.822 2.371 1.327 0.102 4.386 3.577 * Calculations were made using unmodified data. ** Based on the results recorded in Table XIX. TABLE XXI Percent matroclinous progeny based on an estimated total number of progeny produced i n , and correspond-ing mean multiplier value obtained for, each of the experiments. Mean number Estimated Number of -n -| _ _ of progeny total matroclinous Frequency „ ,. ° Female genotype Treatment , ,- . ' Confidence per multiplier number of progeny m percent interval * v i a l + S.E. progeny (Disomic-2 eggs) wow In(2L + 2R)Cy/ap In(2L + 2R)Cy/lt pk on 0 104.12 + 2.81 203,000 0 63.85 + 1.57 55,800 137 23 0.067 0.057 - 0.080 0.041 0.026 - 0.062 wow In(2L + 2R)Cy/ai> EMS In(2L + 2R)Cy/lt pk on EMS 74.78 + 2.26 66,600 44.72 + 1.39 36,300 87 26 0.131 0.105 - 0.161 0.072 0.047 - 0.105 wow In(2L + 2R)Cy/ap In(2L + 2R)Cy/lt pk on MC 40.27 + 3.73 24,700 MC 33.75 + 2.40 24,000 34 16 0.138 0.095 - 0.193 0.067 0.038 - 0.108 +/lt pk on 71.18 + 2.01 62,600 0.002 0.000 - 0.009 +/lt pk on EMS 56.08 + 2.19 42,700 13 0.030 0.016 - 0.052 +/lt pk cn MC 29.52 + 2.06 19,400 22 0.113 0.071 - 0.172 * Based on the table or corrected fiducial limits of expectations from Stevens (1942). 150 The results recorded in Table XXI indicate that the alkylating agents used significantly increases the recovery of matroclinous progeny from the structural heterozygotes when using the a p W O W chromosome. With the It pk cn structural heterozygotes the results show a moderate increase. This demon-strates the importance of the sensitivity of the strains used in any assay system. Additionally, when comparing the effects of EMS and MC on any given strain of structural heterozygotes, there is no difference. However, when comparing the effects of EMS and MC on the structurally homozygous strain, +/lt pk cn, a far greater increase is observed in the recovery of matroclin-ous progeny with the MC treatment than with the EMS treatment. As revealed by the results recorded in Table XXII, the only increase in recovery of patroclinous progeny was obtained from In(2L + 2R)Cy/ap W O W females upon treatment with EMS. Since increases in the recovery of matro-clinous progeny were taken to be nondisjunctional events, an equal increase was expected for the reciprocal product, namely, patroclinous progeny. However, this was not realized. Consequently, i t has not been possible to offer an interpretation to the events giving rise to these exceptional progeny. The results of the chemical mutagens are not as precise or consistent as those found for gamma radiation. Two factors may account for this v a r i a b i l -i t y : 1) i t is much more d i f f i c u l t to standardize a chemical treatment than i t is to standardize radiation exposure, and 2) alkylating agents greatly reduce v i a b i l i t y and increase s t e r i l i t y of Drosophila females indicating the biological potency of these chemicals. Thus, these agents affect the genetic material as well as the physiology of the f l y . Such variables make i f d i f f i c u l t to determine accurately subtle differences. However, this assay system, with a built in dosimeter (Tables XIX and XX), makes negative results TABLE XXII Percent patroclinous progeny based on an estimated total number of progeny produced in, and correspond-ing mean multiplier value obtained for, each of the experiments. Female genotype Treatment Mean number of progeny per multiplier v i a l + S.E. Estimated total number of progeny Number of patroclinous progeny (Nullo-2 eggs) Frequency in percent 95% Confidence intervals* WOW In(2L + 2R)Cy/ap In(2L + 2R)Cy/lt pk cn 0 104.12 + 2.81 203,000 0 63.85 + 1.57 55,800 220 38 0.108 0.095 - 0.124 0.068 0.048 - 0.094 wow In(2L + 2R)Cy/ap In(2L + 2R)Cy/lt pk cn EMS EMS 74.78 + 2.26 44.72 + 1.39 66,600 36,300 114 19 0.171 0.141 - 0.206 0.052 0.032 - 0.082 wow In(2L + 2R)Cy/ap In(2L + 2R)Cy/lt pk cn MC MC 40.27 + 3.73 33.75 + 2.40 24,700 24,000 29 8 0.117 0.079 - 0.169 0.033 0.014 - 0.065 +/lt pk cn 71.18 + 2.01 62,600 0.008 0.003 - 0.019 +/lt pk cn EMS 56.08 + 2.19 42,700 0.012 0.004 - 0.027 +/lt pk cn MC 29.52 + 2.06 19,400 0.026 0.008 - 0.059 * Based on the table or corrected fiducial limits of expectation from Stevens (1942). 152 more meaningful as the recessive lethal test provides confirmation that the mutagen penetrated the target cells. A note of caution must be made concerning comparisons. One assumption required for comparisons between any agents, whether radiation, chemical and/ or control, i s that oocytes of similar maturity are being treated and recover-ed. This makes comparisons between the chemical mutagens and radiation d i f f i c u l t . The radiation treatment of 2000 rads effectively destroys ninety percent or more of the most mature (stage 14) oocytes. Thus, radiation induced aberrations basically represent stage 7, or immature, oocytes. The chemical mutagens in these experiments presumably w i l l affect both stages. Thus the comparison of results between the two treatments may not be that meaningful. There may also be a residual effect of these chemicals as possibly indicated by the increased frequency of equational nondisjunctions recovered in four of the experiments as shown in Tables XIX and XX. 153 DISCUSSION The assay system used in this study, including the internal dosimeter, is an effective means of testing the consequences of exposing female Drosophila to chemical mutagens. The system, via the broad spectrum of recoverable events, provides for an analysis of the types of damage in f l i c t e d by a mutagen. Chapter II provided evidence that compound chromosome formation is a translocation event, or more specifically, an interchange event between sister or non-sister chromatids. The interchange event requires two independent breaks which rejoin in the appropriate way to form a new compound chromosome. The recovery of ten compound chromosomes from MC treated females and none from the EMS treated females supports the findings of Cattanach and Williams (1971), H i l l i k e r and Holm (1975), Lim and Synder (1974), Natarajan and Schmid (1971), Schewe, Suzuki and Erasmus (1971a,b) that EMS does not induce chromo-somal rearrangements. In the previous chapter I discussed three classes of interchange: intrachromosomal, interhomologous and interchromosomal. The fact that a l l ten compound chromosomes recovered from the MC treated females were sister-strand attachments implies that this chemical can cause breaks, but i t would appear that the interchanges are restricted to the intrachromo-somal type. Based on the above evidence that MC, but not EMS, induces breaks, one would expect a higher recovery of matroclinous progeny from the former, assuming nondisjunction is caused by interchanges. The results in Table XXI indicate that both alkylating agents caused a significant increase over the controls in the recovery of matroclinous progeny from structural heterologues. 154 A comparison of the frequencies of recovery of matroclinous progeny, between In(2L + 2R)Cy/ap treated with EMS and with MC revealed no significant difference. Moreover, there is no significant difference between In(2L + 2R)Cy/lt pk cn when treated with the two alkylating agents, although this strain shows an overall lower frequency. This failure of MC to increase the frequency of matroclinous progeny over EMS may suggest that although MC can cause interchanges, for unknown reasons, the interchanges are limited to the intrachromosomal type (as indicated by the results of compound chromosome formation) thus interchromosomal interchanges leading to nondisjunction do not occur. Furthermore, not a l l nondisjunction may be interchange mediated. The mechanism by which nondisjunction is induced by alkylating agents should be investigated futher. The frequency of recovery of patroclinous progeny as recorded in Table XX for structural heterologues was not significantly greater than the fre-quency of recovery of matroclinous progeny when treated with either alkylating agent with the exception of In(2L + 2R)Cy/ap W° W when treated with EMS. More-over, the frequency of patroclinous progeny, except for the one noted, was not greater than the frequency of the controls. If the control values had been ignored these results would have suggested that the patroclinous progeny were the reciprocal product of nondisjunction, i.e. that is no chromosome loss. However, in view of the control values, i t is d i f f i c u l t to accept this interpretation. It would appear, nevertheless, that chemically induced non-disjunction is not interchange mediated as, according to the "drag hypothesis", one would expect a preferential recovery of the nullo eggs, which would be observed as a higher recovery of patroclinous progeny. These studies on the effect of alkylating agents upon the formation of 155 aberrant chromosome-2 products of meiosis are only preliminary. However, they do demonstrate a method by which the developed assay system can not only detect the effect of mutagens, but also provide an approach to disclose the mechanism of their action. 156 BIBLIOGRAPHY Abrahamson, S., Herskowitz, I.H. and Muller, H.J. (1954). Genetic proof for half-translocations derived from irradiated oocytes of Drosophila  melanogaster. Genetics 39_, 955. (Abst). Abrahamson, S. and Lewis, E.B. (1971). The detection of mutations in Drosophila melanogaster. In: Chemical Mutagens (Hollaender, A. Ed). New York: Plenum Press 2, 461-484. Anderson, E.G. (1925). Crossing over in a case of attached-X chromosomes in Drosophila melanogaster. Genetics 10, 403-417. Anderson, E.G. (1931). The constitution of primary exceptions obtained after X-ray treatment of Drosophila. Genetics 16_, 386-396. Baldwin, M. and Chovnick, A. (1967). Autosomal half-tetrad analysis in Drosophila melanogaster. Genetics 5_5, 277-293. Baldwin, M.C. and Suzuki, D.T. (1971). A screening procedure for detection of putative deletions in proximal heterochromatin of Drosophila. Mutation Research 11, 203-213. Ballantyne, G.H. and Chovnick, A. (1971). Gene conversion in higher organ-isms; Non-reciprocal recombination events at the rosy cistron in Drosophila melanogaster. Genetical- Research" 1_7, 139-149. Bateman, A.J. (1968). Nondisjunction and isochromosomes from irradiation of chromosome 2_ in Drosophila. In: Effects of Radiation on Meiotic  Systems. Vienna: International Atomic Energy Agency pp. 63-70. Brandham, P.E. (1970). The consequence of crossing-over in pericentric inversions in acrocentric chromosomes. Heredity Z5, 125-129. Bridges, C.B. (1916). Nondisjunction as proof of the chromosome theory of heredity. Genetics Princeton 1_, 1-52, 107-163. Browning, L.S. (1970). Recessive lethals produced during oogenesis in Drosophila melanogaster by EMS. Drosophila Information Service 45_, 75. Busby, N. (1971). Segregation following interchange induced by irradiating mature oocytes of Drosophila melanogaster. Mutation Research 11, 391-396. Carlson, J.G. (1941). Effects of X-irradiation on grasshopper chromosomes. Cold Spring Harbor Symposium 9_, 104. Cattanach, B.M. , Pollard, CE. and Isaacson, J.H. (1968). Ethylmethanesulfonate-induced chromosome breakage in the mouse. Mutation Research 6, 297-307. 157 Cattanach, B.M. and Williams, E.G. (1971). A search for chromosomal aber-rations induced in mouse spermatogonia by chemical mutagens. Mutation Research 13, 371-375. Chadov, B.H. (1973). The spontaneous formation of the second isochromosome in female Drosophila melanogaster with normally and structurally altered genotypes. Genetika _6, 1267-1268. Chovnick, A., Ballantyne, G.H., B a i l l i e , D.L. and Holm, D.G. (1970). Gene conversion in higher organisms: half-tetrad analysis of recombination within the rosy cistron of Drosophila melanogaster. Genetics 66, 315-329. Clark, A.M. and Sobels, F.H. (1973). Studies on nondisjunction of the major autosomes in Drosophila melanogaster. I. Methodology and rate of induc-tion by X-rays for compound second chromosomes. Mutation Research 18, 47-61. Cooper, K.W. (1945). Normal segregation without chiasmata in female Drosophila melanogaster. Genetics _30_, 472-484. Darlington, CD. (1939). Misdivision and the genetics of the centromere. Journal of Genetics _37_, 341-364. Darlington, CD. (1940). The origin of isochromosomes. Journal of Genetics 39, 351-361. Darlington, CD. (1942). Chromosome chemistry and gene action. Nature, London 149, 66. Day, J.W. and Grell R.F. (1966). Radiation-induced nondisjunction and loss of chromosomes in Drosophila melanogaster females. II. Effects of exchange and structural heterozygosity. Mutation Research J3, 503-509. Drake, J.W. (1969). Mutagenic mechanisms. Annual Review of Genetics j$, 247-268. Evans, W.H. (1971). Preliminary studies on frequency of autosomal nondis-junction in females of Drosophila melanogaster. Drosophila Information Service 46, 123-124. Fitz-Earle, M. (1976). Insect population control using genetic engineering. Bulletin of the Entomogical Society of America 22, 11-14. Fitz-Earle, M., Holm, D.G. and Suzuki, D.T. (1973). Genetic control of insect populations. I. Cage studies of chromosome replacement by compound autosomes in Drosophila melanogaster. Genetics 7^ 4, 461-475. Forer, A. (1966). Characterization of the mitotic traction system and evidence that birefringent spindle fibers neither-produce nor transmit force for chromosome movement. Chromosoma 19, 44-48. 158 Freese, E. (1963). Molecular mechanism of mutations. In: Molecular Biology. New York: Academic Press pp. 207-269. Fristrom, J.W. (1970). The developmental biology of Drosophila. Annual Review of Genetics 4_, 325-345. Gavin, J.A. (1971). The effect of gamma radiation and alkylating agents on the recovery of nondisjunction and compound autosomes in Drosophila melanogaster. B.Sc. Thesis, University of British Columbia. Gavin, J.A. and Holm, D.G. (1972). Gamma ray induced nondisjunction and chromosome loss of chromosome 2_ in females. Drosophila Information Service 48, 143-144. Giles, N.H. (1943). The origin of isochromosomes at meiosis. Genetics 28, 512-524. Grell, E.H. (1970). Distributive pairing: mechanism for segregation of compound autosomal chromosomes in oocytes of Drosophila melanogaster. Genetics Jx5, 65-74. Grell, R.F. (1962). A new hypothesis on the nature and sequence of meiotic events in the female of Drosophila melanogaster. Proceedings of the National Academy of Science U.S.A. 4_8, 165-172. Grell R.F. (1964). Distributive pairing: the size-dependent mechanism for regular segregation of the fourth chromosomes in Drosophila melanogaster. Proceedings of the National Academy of Science U.S.A. j>2_, 226-232. Grell, R.F., Munoz, E.R. and Kirschbaum, W.F. (1966). Radiation-induced nondisjunction and loss of chromosomes in Drosophila melanogaster females. Mutation Research _3, 494-502. Hayashi, S. and Suzuki, D.T. (1970). A comparative study on induced increases in proximal recombination in Drosophila melanogaster females. Canadian Journal of Genetics and Cytology JL0, 276-282. Herskowitz, I.H., Schalet, A. and del val Reuter, Sister Mary, I.H.M. (1962). Induced changes in female germ cells of Drosophila. VII. Exchanges induced in different X chromosome regions after X-raying oocytes and oogonia. Genetics 47, 1663-1678. Hi l l i k e r , A.J. (1976). Genetic analysis of the centromeric heterochromatin of chromosome 2^  of Drosophila melanogaster: Deficiency mapping of EMS-induced lethal complementation groups. Genetics 83_, 765-782. Hi l l i k e r , A.J. and Holm, D.G. (1975). Genetic analysis of the proximal region of chromosome 1_ in Drosophila melanogaster. I. Detachment products of compound autosomes. Genetics 81_, 705-721. Holm, D.G. (1976). Compound autosomes. In: Genetics and Biology of Drosophila Volume lb (Novitski, E. and M. Ashburner, Eds.). New York: Academic Press pp. 529-562. 159 Holm, D.G.., Baldwin, M., Duck, P. and Chovniek, A. (1969). The use of compound autosomes to determine the relative centromeric position of chromosome 3^. Drosophila Information Service 4_4, 112. Holm, D.G. and Chovniek, A. (1975). Compound Autosomes in Drosophila melanogaster: The meiotic behaviour of compound thirds. Genetics 81, 293-311. Holm, D.G., Deland, M. and Chovniek, A. (1967). Meiotic segregation of C(3L) and C(3R) chromosomes in Drosophila melanogaster. Genetics 56, 565-566 (Abstract). Jacobs, P.A. (1971). Human population cytogenetics. Proceedings 4th International Congress of Human Genetics, Paris (September). Jacobs, P.A., Baikie, A.G., Court Brown, W.M. and Strong, J.A. (1959). The somatic chromosomes in mongolism. Lancet 1, 710. King, R.C. (1970). Ovarian Development in Drosophila melanogaster. New York: Academic Press. King, R.C, Rubinson, A.C, and Smith, R.F. (1956). Oogenesis in Adult Drosophila melanogaster. Growth 20, 121-157. Koch, E.A., Smith, P.A. and King, R.C. (1970). Variations in the radio sensitivity of oocyte chromosome during meiosis in Drosophila melano- gaster. Chromosoma J0_, 98-108. Kowalishyn, F.J. (1971). The meiotic behaviour of segregation distorter in compound-2_ chromosomes of Drosophila melanogaster. B.Sc. Thesis, University of British Columbia. Lea, D.E. (1955). Action of radiation on livi n g cells (2nd Ed.). London: Cambridge University Press pp. 189-281. Leigh, B. and Sobels, F.H. (1970). Induction by X-rays of isochromosomes in the germ cells of Drosophila melanogaster males: Evidence for nuclear selection in embryogenesis. Mutation Research 10, 475-487. Lewis, E.B. and Bacher, F. (1968). Method of feeding ethyl methanesulfonate (EMS) to Drosophila males. Drosophila Information Service 4_3, 193. Lim, J.K. and Snyder, L.A. (1974). Cytogenetic and complementation analysis of recessive lethal mutations induced in the X chromosome of Drosophila by three alkylating agents. Genetical Research 24_, 1-10. Lindsley, D.L. and Grell, E.H. (1968). Genetic variation of Drosophila  melanogaster. Carnegie Institute of Washington, Publ. No. 627. Lindsley, D.L. and Grell, E.H. (1969). Spermiogenesis without chromosomes in Drosophila melanogaster. Genetics (Suppl.) _61, 69-78. 160 Loveless, A. and Howarth, S. (.1959). Mutations in bacteria at high levels of survival by ethyl methane sulfonate. Nature, London 184, 1780-1782. Lutolf, H.V. (1972). Meiotic segregation of compound-3 chromosomes in Drosophila. Genetica 43, 431-442. McCloskey, J.D. (1966). The problem of gene activity in the sperm of Drosophila melanogaster. American Naturalist 100, 211-218. Marquardt, H. (1938). Roentgen-pathology of mitosis. Zeitschrift Fuer Botanik 32, 401. Mavor, J.W. (1924). The production of nondisjunction by X-rays. Journal of Experimental Zoology 39, 381-432. Mavor, J.W. and Svenson, H.K. (1924). An effect of X-rays on the linkage of Mendelian characters in the second chromosome of Drosophila melanogaster. Genetics 9_, 70-89. Moore, C.M. and Grell, R.F. (1971). Factors affecting recognition and non-disjunction of chromosomes at distributive pairing in female Drosophila  melanogaster. I. Total length vs. arm length. Genetics 7J), 567-581. Morgan, L.V. (1922). Non-criss-cross inheritance in Drosophila melanogaster. Biological Bulletin 42, 267-274. Morgan, L.V. (1938). Origin of attached-X chromosomes in Drosophila melano- gaster and the occurrence of nondisjunction of the X's in the male. American Naturalist 7_2, 434-446. Morgan, L.V. (1947). A variable phenotype associated with the fourth chromosome of Drosophila melanogaster and affected by heterochromatin. Genetics 30, 200-219. Morgan, T.H., Schultz, J. and Curry, V. (1940). Investigations on the constitution of the germinal material in relation to heredity. Carnegie Institute of Washington. Ybk 39, 251-255. Muller, H.J. (1925). The regionally differential effect of X-rays on cross-ing over in autosomes of Drosophila. Genetics 10, 470-507. Muller, H.J. (1927). A r t i f i c i a l transmutation of the gene. Science 66, 84-87. Muller, H.J. (1940). An analysis of the process of structural change in chromosomes of Drosophila. Journal of Genetics 40_, 1-66. Muller, H.J. and Altenburg, E. (1919). The rate of change of hereditary factors in Drosophila. Proceeding of the Society for Experimental Biological Medicine 17_, 10-14. 161 Natarajan, A.T. and Schmid, W. (1971). Differential response of constitutive and facultative heterochromatin in the manifestation of mitomycin induced chromosome aberrations:in Chinese hamster cells in vitro. Chromosoma 33, 48. Neuhaus, M.J. (.1936). Production of attached X chromosomes in Drosophila  melanogaster males. Nature 137, 996-997. Novitski, E. (1951). Non-random disjunction in Drosophila. Genetics 36, 267-280. Novitski, E. (1952). The genetic consequences of anaphase bridge formation in Drosophila. Genetics 37, 270-287. Novitski, E. (1954). Compound-X chromosomes in Drosophila. Genetics 39, 127-140. Novitski, E. (1955). Genetic measures of centromeric activity in Drosophila  melanogaster. Journal of Cellular and Comparative Physiology 45 supp. , 151-169. Novitski, E. (1967). Nonrandom disjunction in Drosophila. Annual Review of Genetics 1, 71-86. Parker, D.R. (1954a). Radiation induced exchanges in Drosophila females. Proceedings of the National Academy of Science, U.S.A. 40, 795-800. Parker, D.R. (1954b). Role of the Y chromosome in induced detachments of attached-X chromosomes in Drosophila melanogaster. Genetics 39^ , 986 (Abst.). Parker, D.R. (1959). Dominant lethal mutations in irradiated oocytes. Biological contributions, The University of Texas. 5914, 113-127. Parker, D.R. (1965). Chromosome pairing and induced exchange in Drosophila. Mutation Research 2, 523-529. Parker, D.R. (1967). Induced heterologous exchange at meiosis in Drosophila. I. Exchanges between the Y and fourth chromosome. Mutation Research 4_, 333-377. Parker, D.R. (1968). A survey of methods for the induction of aberrations in meiotic stages in Drosophila females and for observation of their disjunctional properties in the ensuing meiotic divisions. In: Effect  of Radiation on Meiotic Systems. Vienna: International Atomic Energy Agency. Parker, D.R. (1969). Heterologous interchange at meiosis in Drosophila. II. Some disjunctional consequences of interchange. Mutation Research 393-407. Parker, D.R. (1970). Coordinate nondisjunction of Y and fourth chromosomes in irradiated compound X female Drosophila. Mutation Research 9_, 307-322. 162 Parker, D.R. (1974). Radiation induced nondisjunction and Robertsonian translocation in Drosophila. Mutation Research _24, 149-162. Parker, D.R. and Busby, N. (1972). Chromosomal interchange in mature oocytes of Drosophila. Mutation Research 1(5, 49-58. Parker, D.R. and Busby, N. (1973). Observations concerning the effects of radiation on the segregation of chromosomes. Mutation Research 18, 33-46. Parker, D.R. and Hammond, A. (1958). The production of translocations in Drosophila oocytes. Genetics Princeton 43, 92-100. Parker, D.R. and McCrone, J. (1958). A genetic analysis of some rearrange-ments induced in oocytes of Drosophila. Genetics 4_3, 172-186. Parker, D.R. and Williamson, J.H. (1970). Heterologous interchange at meiosis in Drosophila. III. Interchange-mediated nondisjunction. Mutation Research 9^, 273-286. Parker, D.R. and Williamson, J.H. (1976). Aberration induction and segrega-tion in oocytes. In: Genetics of Biology of Drosophila Volume lc (Novitski, E. and M. Ashburner, Eds.). New York: Academic Press pp. 1252-1264. Parker, D.R., Williamson, J.H. and Gavin, J. (1974). The nature and time of occurrence of radiation-induced nondisjunction of the acrocentric X and fourth chromosomes in Drosophila melanogaster females. Mutation Research 24, 135-148. Pontecorvo, G. (1941). The induction of chromosome losses in Drosophila sperm and their linear dependence on dosages of irradiation. Journal of Genetics 41, 195-215. Ptashne, M. (1960). The behaviour of strong and weak centromeres at second anaphase of Drosophila melanogaster. Genetics 4_5, 499-506. Rao, R.N. and Natarajan, A.T. (1967). Somatic association i n relation to chemically induced chromosome aberrations in Vicia faba. Genetics 57, 821-835. Rasmussen, I.E. (1960). Reports on new mutants. Drosophila Information Service 3_4, 53. Roberts, P. (1962). Interchromosomal effects and the relation between crossing over and nondisjunction. Genetics 47, 1691-1709. Sankaranarayanan., K. and Sobels,F.- (1976) . Radiation genetics. In: Genetics and Biology of Drosophila Volume lc (Novitski, E. and M. Ashburner, Eds.). New York: Academic Press pp. 1090-1250. Sax, K. (1941). Types and frequencies of chromosomal aberrations induced by X-rays. Cold Spring Harbor Symposium of Quantitative Biology 9_, 93. 163 Schewe, M.J., Suzuki, D.T. and Erasmus, U. (1971a). The Genetic effects of mitomycin C in Drosophila melanogaster. II. Induced meiotic recombina-tion. Mutation Research 12, 269-279. Schewe, M.J., Suzuki, D.T. and Erasmus, U. (1971b). The Genetic effects of mitomycin C in Drosophila melanogaster. I. Induced mutations and X-Y chromosomal interchange. Mutation Research IZ, 255-267. Scriba, M.E.L. (1967). Embryonale entwiek lungsstorungen bei defizienz und tetraploidie des 2. Chromosoms von Drosophila melanogaster. Roux Archiv. Fuer Entw. 159, 314-345. Scriba, M.E.L. (1969). Embryonale entwiek lungsstorungen bei nullosomic und tetrasomie des 3. Chromosoms von Drosophila melanogaster. Devel. Biol. 19, 169-177. Smith, P. and King, R.C. (1968). Genetic control of synaptonemal complexes in Drosophila melanogaster. Genetics ^ 0, 335-351. Sobels, F.H. (1971). The role of Drosophila in the f i e l d of mutation research. Archiv Fuer Genetik 45, 101-125. Sokal, R.S. and Rohlf, F.J. (1969). Biometry. San Francisco: W.H. Freeman and Company. Sonnenblick, B.P. (1941). Germ c e l l movements and sex differentiation of the gonads in the Drosophila embryo. Proceedings of the National Academy of Science, U.S.A. 2T7, 1204-1206. Spieler, R.A. (1963). Genie control of chromosome loss and nondisjunction i n Drosophila melanogaster. Genetics 4_8, 73-90. Spofford, J.B. (1976). Position effect variegation in Drosophila. In: Genetics and Biology of Drosophila Volume l c (Novitski, E. and M. Ashburner, Eds.). New York: Academic Press pp. 955-1009. Stevens, W.L. (1942). Accuracy of mutation rates. Journal of Genetics 43, 301-306. Sturtevant, A.H. and Beadle, G.W. (1936). The relationship of inversions in the X chromosome of Drosophila melanogaster to crossing-over and dis-junction. Genetics _21, 554-604. Thompson, P.E. (1960). Effects of some autosomal inversions on lethal muta-tions in Drosophila melanogaster. Genetics 45, 1567-1580. Traut, H. (1964). The dose-dependence of X chromosome loss and nondisjunc-tion induced by X-rays in oocytes of Drosophila melanogaster. Mutation Research 1, 157-162. Traut, H. (1967a). X-ray induction of 2:3 translocations in mature and im-mature oocytes of Drosophila melanogaster. Genetics J56_, 265-272. 164 Traut, H. (1967b). X chromosome loss induced by low X-ray doses in mature and immature oocytes of Drosophila melanogaster. Mutation Research 4_, 510-513. Traut, H. (1968). Experiments on the mechanism of X-ray-induced chromosome loss. Mutation Research 109-115. Traut, H. (1970). Nondisjunction induced by X-rays in oocytes of Drosophila melanogaster. Mutation Research 10, 125-132. Traut, H. (1971). The influence of the temporal distribution of the X-ray dose of the induction of X-chromosomal nondisjunction and X-chromosome loss in oocytes of Drosophila melanogaster. Mutation Research 12, 321-327, Traut, H. and Scheid, W. (1970). Cytological analysis of partial and total X chromosome loss induced by X-rays in oocytes of Drosophila melanogaster. Mutation Research 10, 583-589. Traut, H. and Scheid, W. (1971). The production of monosomic and trisomic individuals in Drosophila melanogaster by X-irradiation of immature oocytes. Mutation Research 13, 429-432. Traut, H. and Schmidt, P. (1968). Repair of dominant lethal damage induced by X-rays in immature oocytes of Drosophila melanogaster. International Journal of Radiation Biology 13, 405-415. Uchida, I.A. (1962). The effect of maternal age and radiation on the rates of nondisjunction in Drosophila melanogaster. Canadian Journal of Genetics and Cytology 4., 402-408. Valencia, R.M. (1970). A cytogenic study of radiation damage in entire genome of Drosophila. Mutation Research J_0, 207-219. Ward, L. (1923). The genetics of curly wing in Drosophila. Another case of balanced lethal factors. Genetics 276-300. Williamson, J.H. (1969). On the nature of Y chromosome fragments induced in Drosophila melanogaster females. I. Immature oocytes. Mutation Research 8, 327-335. Williamson, J.H. (1970a). On the nature of Y chromosome fragments induced in Drosophila melanogaster females. II. Mature oocytes. Mutation Research 9, 85-90. Williamson, J.H. (1970b). Ethyl methanesulfonate-induced mutants in the Y chromosome of Drosophila melanogaster. Mutation Research 10, 597-605. Williamson, J.H. (1973). Heterologous interchange and nondisjunction of distributively paired chromosomes in Drosophila melanogaster oocytes. Mutation Research 18, 273-277. 165 Williamson, J.H. (1974a). Heterologous interchange and nondisjunction of distributively paired chromosomes in Drosophila melanogaster immature oocytes. Mutation Research 23, 189-196. Williamson, J.H. (1974b). Heterologous interchange and nondisjunction of distributively paired chromosomes in Drosophila melanogaster mature oocytes. Mutation Research _18, 273-277. Williamson, J.H. and Parker, D.R. (1968). Recovery of multiple break re-arrangements from irradiated oocytes. Drosophila Information Service 43, 176. Wurgler, F.E., Ruch, R. and Graf, U. (1972). X-ray-induced loss of the maternal second chromosome in inseminated eggs of Drosophila melanogaster. Mutation Research 15_, 31-40. Yeomans, T.C. (1972). Induced exchange and compound autosome formation in females of Drosophila melanogaster. B.Sc. Thesis, University of British Columbia. Zimmering, S. (1962). Genetic evidence of X-ray induced exchange occurring at a four-strand stage in Drosophila spermatocytes. Journal of Heredity 53_, 254-256. Zimmering, S. (1976). Genetics and cytogenetics aspects of altered segrega-tional phenomena in Drosophila. In: Genetics and Biology of Drosophila Volume lb (Novitski, E. and M. Ashburner, Eds.). New York: Academic Press pp. 569-665. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0094205/manifest

Comment

Related Items