UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Dispersal of transposable elements Meister, Gerald Alan 1999

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

Item Metadata

Download

Media
831-ubc_1999-463893.pdf [ 16.82MB ]
Metadata
JSON: 831-1.0089311.json
JSON-LD: 831-1.0089311-ld.json
RDF/XML (Pretty): 831-1.0089311-rdf.xml
RDF/JSON: 831-1.0089311-rdf.json
Turtle: 831-1.0089311-turtle.txt
N-Triples: 831-1.0089311-rdf-ntriples.txt
Original Record: 831-1.0089311-source.json
Full Text
831-1.0089311-fulltext.txt
Citation
831-1.0089311.ris

Full Text

DISPERSAL OF TRANSPOSABLE ELEMENTS by GERALD ALAN MEISTER B.Sc, The University of British Columbia, 1986 M.Sc, The University of British Columbia, 1992 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Genetics Graduate Program) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA Sept 1999 © Gerald Alan Meister, 1999 Friday, October 8, 1999 UBC Special Collections - Thesis Authorisation Form Page: 1 In presenting this thesis in pa r t i a l fulfilment of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library shall make i t freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It i s understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. The University of B r i t i s h Columbia Vancouver, Canada http://www.library.ubc.ca/spcoll/thesauth.html 11 ABSTRACT This thesis describes four related sets of experiments: The first three experimental series investigated the dynamics of transposable element (TE) spread and accumulation following the introduction of low frequencies of Drosophila melanogaster containing specific TEs into populations that previously lacked the elements. In the first series (Chapter II), flies containing P elements with an inserted alcohol dehydrogenase gene were introduced. DNA hybridization assays showed that, despite an approximate three-fold increase in size over unmodified elements, the P element constructs were capable of rapid dispersal through the experimental populations. Moreover, assays for alcohol dehydrogenase activity revealed that many of the dispersed genes still encoded an active product. In the second series of experiments (Chapter III), hobo element containing flies were introduced. DNA hybridization assays showed that hobo elements were present within virtually all individuals within eight generations. The mean amount of hobo hybridizing DNA per element containing individual decreased in the first few generations, but then increased and stabilized at approximately 50% of that present within the element donating strain. The total hobo hybridizing DNA in the populations steadily increased until it also stabilized at about 50% of the amount in the donor strain. In the third series of experimental populations (Chapter IV), genomes containing both P and hobo elements were introduced. Within 8-10 generations both elements spread to all genomes within these populations and the P element DNA rapidly accumulated to the amount present in the element-donating strain. However, as with the populations in which only hobo I l l elements were introduced, the hobo DNA accumulated to much less than that present in the donating strain. In the final set of experiments (Chapter V), a differential DNA hybridization method for detecting moderately repetitive strain-specific TEs was developed. To demonstrate the effectiveness of this procedure, a lambda genomic library was prepared from DNA of a D. melanogaster strain that contained both P and hobo TEs. Duplicate plaque lifts of this library were then hybridized to DNA from this strain and to DNA from a strain which lacked the TEs. Differentially hybridizing plaques were isolated. Finally, reprobing with TE sequences demonstrated that plaques that hybridized differentially to the genomic probes did indeed contain TEs. This differential screen was then applied to several pest insect species in preliminary experiments. TABLE OF CONTENTS i v Abstract " List of Tables viii List of Figures ix Preface x Acknowledgments xi CHAPTER I: LITERATURE REVIEW 1 Transposable Elements 2 Classes of Transposable Elements 3 The P elements of Drosophila melanopaster 7 Hybrid Dysgensis and P elements 7 Structure ofP Elements and the Transposase Gene 8 Transposition of P Elements 9 Regulation ofP Element Transposition 11 Distribution ofP Elements in D. melanogaster 14 Distribution ofP elements in other species 17 Transposable Elements as Tools 20 Aims of this Research 21 CHAPTER II: RAPID SPREAD OF A P ELEMENT/ Adh CONSTRUCT THROUGH EXPERIMENTAL POPULATIONS OF D. MELANOGASTER 23 INTRODUCTION 24 MATERIALS AND METHODS 26 Drosophila Strains....... 26 Experimental Populations . 27 Single Flv Ovarv Blot Assays 27 Single Flv ADH Assays.... 30 RESULTS 32 Dispersal of P Element Sequences 32 Dispersal of ADH Activity 35 Phenotypes of Individuals 37 DISCUSSION..... 40 Dispersion of P Sequences 40 Dispersion of ADH Activity 42 Distribution of Phenotypes 44 Multiplication of P Elements 46 CHAPTER III: RAPID DISPERSAL AND ACCUMULATION OF HOBO ELEMENTS WITHIN MIXED EXPERIMENTAL POPULATIONS OF D. MELANOGASTER. 48 INTRODUCTION 49 MATERIALS AND METHODS 56 Experimental Populations 56 DNA preparation 56 Probes 57 Quantitative dot blots 58 Quantitative single flv Southern blots 63 Gonadal Dysgenesis Assay 65 RESULTS 66 Changes in total hobo element DNA within populations 66 Dispersal of hobo elements 71 v i Sizes of hobo elements present 74 Changes in amount of hobo DNA in single flies 80 Development of Regulatory Potential 83 DISCUSSION 85 Dispersal and accumulation of hobo elements 85 Regulation of hobo copy number 97 Summary 105 CHAPTER IV: CONCURRENT SPREAD OF P AND HOBO ELEMENTS WITHIN MIXED EXPERIMENTAL POPULATIONS OF D. MELANOGASTER 106 INTRODUCTION 107 MATERIALS AND METHODS 107 RESULTS 108 Presence of P and hobo DNA within individuals 108 Concurrent accumulation of P and hobo elements 110 DISCUSSION 114 CHAPTER V: A SIMPLE PLAQUE HYBRIDIZATION METHOD FOR THE DETECTION OF DIFFERENTIALLY REPRESENTED REPETITIVE DNA 117 INTRODUCTION 118 MATERIALS AND METHODS 122 Construction of the Drosophila Library 122 Screening of the Drosophila Library 122 Preparation of Recombinant Phage DNA 123 Insect Collections 125 v i i Application of Differential Screen to Pest Insects . 129 RESULTS 130 Drosophila as a Model System 130 Differential Screens of Pest Insects 132 DISCUSSION 135 Drosophila TEs as a Model System 135 Differential Hybridization Screens of Pest Insects 137 Future Perspectives 138 CHAPTER VI: SUMMARY OF THESIS RESEARCH 144 LITERATURE CITED 149 APPENDIX 1 169 Stepl 171 Step 2.. 171 Step 3 174 Step4 176 APPENDIX 2 178 APPENDIX 3 182 v m TIST OF TABLES Table I: The P Element Family of Transposons 19 Table II: Flies producing only 1.5 kb hobo homologous bands 76 Table III: Flies producing new sizes of hobo homologous bands 78 Table IV: Percent dysgenic ovaries from A* crosses 84 Table V: Predicted dispersal and accumulation of hobo elements in a 1% OR population based on segregation and recombination 88 Table VI: Predicted dispersal and accumulation of 3.0 kb hobo elements in a 1% OR population based on segregation and recombination 88 Table VII: Calculation of amount of hobo DNA in the 1% OR-A population.... 172 ix T 1ST OF FTGTTR.ES Figure 1: Structure of short inverted terminal repeat transposable elements.... 6 Figure 2: Setup of mixed experimental populations 28 Figure 3: Test of the sensitivity of the ovary blot assay 31 Figure 4: Ovary blots showing presence of P elements 33 Figure 5: Dispersal of P elements to individuals of tAP-8A populations 34 Figure 6: Demonstration of the ADH assay 36 Figure 7: Dispersal of ADH+ actuvity to flies of the tAP-8A populations 38 Figure 8: Percent of flies of each phenotype in the tAP-8A populations 39 Figure 9: Examples of normal and dysgenic ovaries in Drosophila 67 Figure 10: Dot blot showing increase in hobo DNA in populations 68 Figure 11: Changes in the amount of hobo element DNA within populations... 70 Figure 12: Dot blot showing an increase in frequency of flies containing hobo . 72 Figure 13: Southern blot showing an increase in flies containing hobo 73 Figure 14: Percent of flies from 1% OR populations which contain hobo 75 Figure 15: Changes in amount of hobo homologous DNA within individuals... 82 Figure 16: Presence of both P and hobo element DNA within individuals 109 Figure 17: Dot blot showing presence of P and hobo DNA within populations. Ill Figure 18: Changes in the amount of P and hobo DNA within populations.... 113 Figure 19: Simulium vittatum and Simulium arcticum 128 Figure 20: Hybridization of replicate plaque lifts of a n2 genomic library 131 Figure 21: Typical replicate Southern blots of Lymantria clones 134 Figure 22: Gonadal dysgenesis in P. Saucia? 142 Figure 23: Volume measurement using ellipses 170 Figure 24: Determination of micrograms DNA present in samples 173 Figure 25: Calculation of background hybridization to hobo element probe 175 Figure 26: Calculation of 100% OR strain hobo DNA 177 Figure 27: Volume measurement using the grid method 179 Figure 28: Quantitation using the line method 180 Figure 29: Comparison of 3 methods of selecting regions for quantitation 181 Figure 30: Comparison of two types of OR gradient. 183 PREFACE Much of the material presented in this thesis has been previously published. The material in Chapter II was published by MEISTER and GRIGLIATTI 1993. The section in Chapter V about using Drosophila as a model system for the differential screen has been published by MEISTER, LANSMAN and GRIGLIATTI 1995). In both cases, authors other than MEISTER provided a primarily supervisory role. xi ACKNOWLEDGMENTS I would like to first and foremost thank my supervisor, TOM GRIGLIATTI, for his continual encouragement, advice and patience throughout the duration of this work. I am also thankful for the additional guidance provided by my supervisory committee: A. ROSE, D. HOLM, T. BEATTY, and G. SCUDDER. For technical advice, I am indebted to R. LANSMAN, D. THEILMAN, H. BROCK, T. PFEIFER and B. DEVLIN. Many additional people have provided useful feedback, discussion, and support. Among these I especially thank R. MOTTUS, G. KHACHATOURIANS, and the anonymous reviewers of my publications. I would also like to thank the many scientists who kindly donated organisms for this research: W. R. ENGELS, W. H. GELBART, W. SOFER and J. POSAKONY supplied Drosophila strains; M. ISMAN and A. MOLDENKE kindly donated Peridroma isolates; T. PFEIFER contributed Lymantria that he originally obtained from L. HUMBLE and the USDA; and D. CURRTE, G. S. MlRANPURl and C. BROCKHOUSE all supplied Simulium. I also thank W.H. GELBART, K . O'HARE, and G.M. RUBIN for donating plasmids used as probes and D. THEILMAN for the use of the phospho imager. Finally, on a more personal note, I thank CLARE, MOM, DAD, GREG and BRENDA. Their unconditional support and encouragement made this possible. CHAPTER I: LITERATURE REVIEW: TRANSPOSABLE ELEMENTS IN DROSOPHILA MELANOGASTER 2 Transposable Elements: Transposable elements (TEs) are discrete segments of DNA that are able to mobilize and insert into new sites within the same genome. The existence of TEs was first deduced by Barbara McClintock in the 1940s using purely genetic analysis. She demonstrated that there were DNA sequences in maize that could move to new chromosomal locations and cause altered expression of nearby genes as well as chromosome breakage (MCCLINTOCK 1951). Further understanding of McClintock's "controlling elements" would await the passage of several decades and require the elucidation of the structure of DNA, the development of recombinant DNA technology, and the discovery of transposition in bacteria. Although originally considered to be rare components of genomes, TEs have now been found in a wide variety of organisms and are most likely ubiquitous (for reviews see BERG and HOWE 1989; SHERRATT 1995; SAEDLER and GlERL 1996). This remarkable presence of mobile DNA in both prokaryotes and eukaryotes has challenged the long accepted principle that genes and their arrangement within genomes are stable and transmitted with fidelity from parents to progeny. The genome of Drosophila melanogaster MEIGAN can be broadly separated into three categories of DNA based on reassociation kinetics: about 70% is single copy DNA, 12% is highly repetitive DNA (average reiteration frequency of about 24,000), and 12% is middle repetitive DNA (SCHACHAT and HOGNESS 1974; MANNING, et al. 1975). About 25% of this middle repetitive DNA is arranged in tandem repeats and contains genes for rRNA, 5S RNA and histones (SPRADLING and RUBIN 1981). The remainder of the middle repetitive DNA is comprised of 40-50 sequence families with a reiteration frequency of 35-100. The individual members of these families are found randomly interspersed with single copy sequences and they generally occupy different chromosomal locations in different 3 strains (MANNING, et al. 1975; YOUNG 1979). While some of the middle repetitive sequence families of D. melanogaster are found in sibling species, many are not. Moreover, closely related species of Drosophila often contain different families of repetitive elements (DOWSETT and YOUNG 1982; DOWSETT 1983). These observations suggest that these families of dispersed middle repetitive DNA, which together account for about 10% of the D. melanogaster genome, are relatively unstable components and are in fact TEs and/or remnants of TEs. Perhaps even more remarkably, it has been estimated that mobilization of these TEs is responsible for as much as 80% of all spontaneous mutations in D. melanogaster (GREEN 1988; ENGELS 1989; FlNNEGAN 1990). Genetic studies also reveal that TEs are the primary cause of chromosome rearrangements (LlM and SIMMONS 1994). Classes of Transposable Elements: For organizational purposes, TEs are separated into groups based on the type of transposition intermediate they form and on structural or sequence similarities. It is generally assumed that elements which are grouped together according to this classification scheme transpose via similar mechanisms. It is tempting to assume that elements that are grouped together are also derived from a common ancestral element and undoubtedly this is sometimes true. However, many TEs display a variety of unique features and this classification may represent an oversimplification. For example some, but not all elements that transpose directly from DNA to DNA, share a region of homology with elements that transpose via RNA intermediates. Thus, those DNA to DNA elements that contain this sequence, called the DDE signature (FAYET et al. 1990; KHAN et al. 1991; CAPY et al. 1996, 1997), may share a more recent common ancestor with elements that transpose via RNA intermediates than 4 they do with other DNA to DNA elements. In fact it is possible that DNA to DNA elements evolved independently several times. Yet, at the broadest level in this classification scheme we divide elements according to whether they have RNA or DNA intermediates. Elements that move via an RNA intermediate are called class I elements, whereas elements that move directly from DNA to DNA are termed class II elements (FINNEGAN 1989, 1990, 1992). Class I elements, are further divided into two main groups. In the D. melanogaster genome, the most frequent group is the retrotransposons which includes the copia, gypsy, mdg4 and blood TE families. Elements within each "family" share a high level of DNA sequence similarity and are named after the first element of each family discovered. All retrotransposons can be recognized structurally by long terminal direct repeats (LTRs). They also contain an open reading frame that encodes a protein related to viral reverse transcriptase. Retrotransposons are sometimes called retrovirus-like elements because they are believed to transpose by reverse transcription of an RNA intermediate using a mechanism related to the retroviral life cycle. The second group of class I elements is termed non-viral retroposons and in D. melanogaster this group includes the I, F, G, Doc and Jockey TE families. As with the retrotransposons, these elements encode a putative reverse transcriptase and are believed to transpose by a reverse transcription mechanism. However, instead of LTRs, the 3' end of one DNA strand of each element contains a poly-A sequence and the elements are often truncated at the 5' end of this same strand. It is believed that the two types of class I elements diverged long ago since their genes have little sequence similarity (XlONG and EICKBUSH 1990). Class II elements can also be subdivided into two main groups. The first of these groups is characterized by long inverted terminal repeats that vary in 5 length from a few hundred to a few thousand base pairs. In D. melanogaster there is only one known TE family of this type. This family is called fold-back based on the elements' tendency to "fold-back" on themselves when allowed to re-anneal at low concentration. The repeats of any one foldback element are not necessarily identical and may be found immediately adjacent to each other in the genome or may be separated by several kilobases of either related or unrelated DNA. The mechanism by which foldback elements mobilize is not known, however, RNA intermediates have not been detected. The second major group of class II TEs are those that have short inverted terminal repeats (ITRs) which are bounded by short (2-8 base pair) direct duplications of target DNA (Figure 1). The direct repeats are presumably produced by the filling in of a staggered cut created at the time of element insertion. ITR transposons from different families generally share a number of other features. Most contain a single gene which encodes transposase, a protein that is required for that specific element's mobilization. Surprisingly, transposase proteins from different ITR families have little if any sequence similarity and are not necessarily evolutionarily homologous. Each ITR element also usually exists in two forms: complete elements and their internally deleted derivatives. P, hobo and pogo elements are all ITR transposon families that were originally discovered in D. melanogaster. The other two ITR elements which have been isolated fromD. melanogaster, HB and Bari-1, are both members of the Tel transposon family that was originally identified in the nematode Caenorhabditis elegans MAUPAS (HARRIS et al. 1988; CAIZZI et al 1993). The thesis deals primarily with the dispersal of the P and hobo ITR transposon families within experimental populations of D. melanogaster. P elements are the most studied of all eukayotic transposable elements and our knowledge of P elements is often used as the basis of comparison for other ITR 6 7 transposons. I will therefore present substantially more background information on P elements below. Additional review of the literature related to hobo elements will be given in the INTRODUCTION of CHAPTER III. The P elements of Drosophila melanogaster: Hybrid Dysgensis and P elements: HlRAIZUMI (1971) crossed D. melanogaster males caught in the wild to females from a laboratory stock and observed recombination in the male progeny (a trait not normally seen in D. melanogaster). It was later shown that this male recombination could be correlated with several other abnormal traits including temperature sensitive gonadal sterility, high mutation rate, segregation distortion and chromosome aberration (KlDWELL et al. 1977; ENGELS 1979a). Since this syndrome occurred only in the germline of the hybrid progeny of certain crosses it was called hybrid dysgenesis. D. melanogaster strains can be phenotypically categorized according to their ability to induce and/or suppress the gonadal sterility phenotype in hybrid offspring (KlDWELL et al. 1983). In paternal (P) strains, males can induce gonadal sterility and females can suppress it. In maternal (M) strains, males cannot induce gonadal sterility and females are susceptible. Therefore, hybrid dysgenesis occurs only when P strain males are crossed to M strain females. It does not occur in the reciprocal cross (M males to P females) or in PxP or MxM crosses (for reviews, see BREGLIANO and KlDWELL 1983; ENGELS 1983). The first evidence that hybrid dysgenesis was due to transposable elements came from the observation that the P factors responsible mapped to many locations in P strain genomes (ENGELS 1979b). The instability of certain dysgenesis-induced mutations (GULUBOVSKIe* al. 1977; ENGELS 1979c; ENGELS 8 1981a) and the identification of chromosome breakage hotspots whose positions varied among chromosomes from unrelated P strains (ENGELS and PRESTON 1981) also supported the mobile element explanation for hybrid dysgenesis. The P factor hypothesis was confirmed when a transposable element insertion was found in a dysgenesis-induced allele of the white locus (RUBIN et al. 1982). Homologous elements were also found at chromosome breakage hot spots and other dysgenesis-induced mutations (BINGHAM et al. 1982). All of the traits associated with hybrid dysgenesis have since been shown to be in some way associated with an elevated level of P element activity in hybrid progeny. There is considerable variation in the P-M phenotype besides the P and M strain extremes. Strong P strains have 30-50 copies of P sequences, but as few as 30% of these may be complete (O'HARE and RUBIN 1983). There are two main subtypes of M strains. True M strains completely lack P elements by molecular analysis and have extremely high susceptibility to P strains (KlDWELL 1985); pseudo M (M') strains contain P elements, sometimes many of them, and their susceptibility to hybrid dysgenesis ranges from low to high (ENGELS 1984; BOUSSY and KlDWELL 1987). Neutral or Q strains are defined as strains that cause less than 10% gonadal sterility among hybrids with M strain females , but that are also resistant to the dysgenic activity of P strain males (KlDWELL and NOVY 1979; KlDWELL et al. 1983). Q strains may have only a subset of P sequences, or may have a reduced number of transposase producing P elements relative to P strains (BINGHAM et al. 1982). Structure of P Elements and the Transposase Gene: The complete P element is 2,907 base pairs long, is bounded by 31 base pair ITRs and creates an 8 base pair duplication of target DNA at each insertion site (O'HARE and RUBIN 1983). In germline tissues, the complete element expresses transposase, an 87 9 kilodalton (kDa) protein required for transposition and excision (BINGHAM et al. 1982; CHARE and RUBIN 1983; RIO et al. 1986). The P element has four open reading frames (ORF 0, 1, 2, and 3) and mutational analysis has revealed that all four are required in cis to encode a functional transposase protein (KARESS and RUBIN 1984; LASKI et al. 1986; RIO et al. 1986). Incomplete P elements are very heterogeneous and range in size from 0.5 to 2.9 kilobases (RUBIN et al. 1982). Their sequences can be derived from the complete element by internal deletions (O'HARE and RUBIN 1983). The incomplete elements do not encode a functional transposase, but can be mobilized in the presence of complete P factors (SPRADLING and RUBEN 1982; RUBIN and SPRADLING 1982). The complete and incomplete elements are therefore sometimes referred to as autonomous and nonautonomous elements respectively. Transposition of P Elements: The currently accepted model for P element transposition is conservative cut-and-paste transposition followed by gap repair of the excision site (ENGELS et al. 1990; KAUFMAN and RIO 1992). Modification of this model to include recent evidence produces the following scenario for repair: 1) excision occurs by staggered cuts at the element ends and leaves behind at least 33 nucleotides of single stranded sequence including the element's inverted repeats (O'BROCHTA et al. 1991; JOHNSON-SCHLITZ and ENGELS 1993; KEELER and GLOOR 1997); 2) each side of the break can then be acted on by exonucleases and is also free to invade an homologous template sequence which can be supplied by the sister chromatid, the homologous chromosome, an ectopic sequence, or even a plasmid (ENGELS et al. 1990, 1994; GLOOR etal. 1991; NASSIF et al. 1994; KEELER e£ al. 1996); 3) DNA synthesis proceeds independently from each end, displacing the newly synthesized 10 strands; 4) when a region of base pair complementarity is formed, the single strands anneal and use each other as a template to complete the repair process (FORMOSA and ALBERTS 1986; NASSIF et al. 1994). This mechanism is referred to as the synthesis-dependent strand annealing model (SDSA). It has long been known that proper termini are required for P element mobilization (RUBIN and SPRADLING 1983; KARESS and RUBIN 1984). MULLINS et al. (1989) used germline transformation with a variety of modified elements to determine the precise cis requirements for mobility. They found that about 150 base pairs of DNA from each end of the P element was essential. This region includes the 31 base pair terminal inverted repeats, unique sequences at the 3' and 5' ends, and internal 11 base pair inverted repeats that are located about 125 base pairs from the element ends. The P element transposase protein has been purified and characterized (KAUFMAN et al. 1989). It is a site-specific DNA binding protein that specifically interacts with a 10 base pair consensus sequence that is located just internal to the ITRs at both ends of the P elements. These sequences lie within the regions shown to be important for transposition. On the 5' end, the binding site also overlaps with sequences shown to be essential for P element transcription. This raises the possibility that transposase and/or the 66 kDa P element protein (described below) could regulate transcription from the P element promoter. Transposase protein does not directly interact with the inverted repeats. This may suggest that the transposition reaction and/or repair of the gap left after element excision requires the binding of Drosophila proteins to the P element termini. Requirements for host-encoded functions are common for prokaryotic transposition reactions (BERG and HOWE 1989). One host-encoded protein that interacts specifically with the outer 16 base pairs of both terminal repeats of the P element has been found (RIO and RUBIN 1988). It has been proposed that this 11 protein, called the inverted repeat binding protein (IRBP), may act to protect the P element ends left at the gap after element excision (STAVELEY et al. 1995; KEELER and GLOOR 1997). This proposal is supported by the observations that repair of chromosomal DNA breaks after P element excision is reduced and that any resulting excision products can contain large deletions in the absence of wild type IRBP function (BEALL and RlO 1996). IRBP may even assist in the search for homology, since IRBP is homologous with a protein involved in mammalian DNA repair (BEALL et al. 1994). Regulation ofP Element Transposition: The transposition of most if not all transposable elements is regulated, which limits deleterious affects on the host organism. There are at least three types of gene regulation associated with P element mobilization. First, P element transposition is restricted to germline cells. Second, the reciprocal cross differences that are characteristic of hybrid dysgenesis are due to a maternally inherited cellular condition, known as P cytotype. Finally, P element activity can be suppressed by chromosomal factors in zygotes. The current understanding of each of these three modes of P element regulation is summarized below. While P elements can transpose at high frequencies in the germline of certain hybrid flies, they are almost completely stable in somatic tissue. This tissue specificity, the first form of P element regulation, occurs at the level of pre-mRNA splicing of the intron between ORF 2 and ORF 3 of the P element (LASKI et al. 1986). The mature RNA in somatic tissue retains the third intron resulting in premature termination of translation and the production of a smaller 66 kDa protein instead of 87 kDa transposase. LASKI and colleagues (1986) put a modified element lacking the third intron into the genome and obtained high levels of somatic activity as indicated by somatic mosaicism for P 12 element induced mutations. There is now also direct evidence that the 66 kDa protein can act as a negative regulator of transposition (ROBERTSON and ENGELS 1989; MlSRA and RIO 1990), but this leaves unanswered the question of how the alternative splicing is controlled. SlEBEL and RIO (1990) demonstrated that the third intron can be spliced by a heterologous cell extract, but that this splicing was blocked by pre-incubation of the pre-mRNA with Drosophila somatic cell extract. Using UV cross-linking they went on to show that a 97 kDa Drosophila protein present in the somatic cell extract binds preferentially to sequences near the 5' splice site. This binding can be correlated with the inhibition of splicing in vitro. This evidence supports a model of tissue specific regulation in which specific proteins in somatic cells block the splicing of the third intron resulting in the absence of functional transposase and the production of a 66 kDa repressor of transposition. The second form of P element regulation gives rise to the non-reciprocal nature of the hybrid dysgenesis phenomena and is due to a regulatory ability of P strain females which is manifested in their eggs. Individuals that are resistant to the action of P elements have been described as having P cytotype, whereas those that are susceptible have been said to have M cytotype (ENGELS 1979b). Hence, hybrid dysgenesis occurs only when P elements are present in an embryo with M cytotype. Cytotype is bimodal; flies with mixed P and M ancestry usually display either an M or P cytotype rather than some intermediate state (ENGELS 1979b). The cytotype of a fly is determined by an unusual mode of inheritance that involves both the individual's genotype and the mother's cytotype. The frequency of P cytotype is higher in the progeny of mothers with P rather than M cytotype and this frequency increases with the number of P elements in the genome (ENGELS 1989). In the absence of P 13 elements the cytotype is always M. This suggests that repressor products produced by P elements in the P strain germline can be inherited maternally. The current model for cytotype regulation involves "Type F repressor proteins that are produced by P elements that retain at least the first three exons (GLOOR et al. 1993) and may sometimes be complete (LEMAITRE et al. 1993). Type I repressors act at the level of transcription (LEMAITRE and COEN 1991; LEMAITRE et al. 1993; ROCHE et al. 1995). LEMAITRE and his associates (1993) suggest that the 97 kDa protein, already shown to prevent splicing of the last intron in somatic cells, is also present in the germline, though in reduced quantities. When maternally inherited repressor is present in the germline (P cytotype), relatively low levels of P element mRNA are transcribed and the amount of the 97 kDa splice-blocker protein is sufficient to ensure that only the 66 kDa repressor is made. However in the absence of a substantial amount of maternally inherited repressor (M cytotype), more P element mRNAs are transcribed and these overwhelm the splice-blocker resulting in transposase production. Positive feedback of the Type I repressor stimulating more of its own production also explains the all or nothing bimodality of cytotype. The third type of P element regulation, suppression of hybrid dysgenesis exhibited by most Q and M' strains, differs in two ways from that exhibited by P strains (KlDWELL 1985; SIMMONS and BUCHOLZ 1985; BLACK etal. 1987). First, it is chromosomally inherited and therefore is fully transmissible through both the male and female lines. Second, instead of the bimodality associated with P and M cytotype, various combinations of M and M' chromosomes result in all levels of dysgenesis when crossed to P strain males. M' type suppression probably results from the presence of certain deleted elements, such as KP elements, that are present in high copy number in some M' strains (BLACK et al. 1987). KP elements are just one member of a group of elements called Type II 14 repressor elements. Type II elements are generally much smaller than Type I repressors and lack some of the sequences, including the 2-3 intron, shown to be necessary for Type I repression ( R A S M U S S O N et al. 1993; G L O O R et al 1993). At least three potential mechanisms of action have been proposed for Type II repressors. First, Type II repressors may "poison" heteromeric transposase complexes since their protein product retains the leucine zipper domain ( A N D R E W S and G L O O R 1995). Second, they may repress through protein-DNA interactions since, unlike P transposase, the protein product of KP elements binds to multiple sites at the P element termini including the 31 base pair ITRs ( L E E et al. 1996). Third, and perhaps most likely, it is possible that Type I repressors use the same mechanism as Type I repressors. This suggestion is supported by evidence that Type II repressors, like Type I repressors, probably act at the level of transcription ( L E M A I T R E et al. 1993; E N G E L S 1996). In addition, the ability of both Type I and Type II elements to act as repressors is extremely sensitive to genomic position ( R O B E R T S O N and E N G E L S 1989; R O N S E R R A Y et al. 1991; H I G U E T et al. 1992; G L O O R et al. 1993). The lack of maternal inheritance would be expected with this model since Type II elements lack the last P element exon and thus cannot be alternately spliced. They would therefore make a repressor regardless of the availability of splice-blocker (i.e. regardless of tissue type or the presence of maternally inherited repressor). The variable levels of repression exhibited by M strains can also be easily explained by this model since Type II repressors would lack the positive feedback proposed for Type I repressors. Distribution ofP Elements in D. melanogaster: Considerable variation in the P-M phenotype exists amongst worldwide populations of D. melanogaster. Surveys of wild populations from America, Japan, Europe, Asia, Africa and 15 Australia reveal these populations to be P, Q or M' (ANXOLABEHERE et al. 1982, 1984, 1985, 1988; BREGLIANO and KlDWELL 1983; KlDWELL et al. 1983; KlDWELL 1983; TAKADA et al. 1983; YAMAMOTO et al. 1984; KlDWELL and NOVY 1985; BOUSSY 1987; BOUSSY and KlDWELL 1987). Strains from a given geographical region are often predominantly of one type, although Q strains can be found in virtually all geographic regions (ENGELS 1989). Several broad patterns have been established. The frequency of Q strains gradually declines from west to east in Europe and central Asia (ANXOLABEHERE et al. 1985). Eastern Australian populations can be divided into three regions based on the predominant strain types which change over short distances from P to Q to M' in a north to south direction (BOUSSY 1987; BOUSSY and KlDWELL 1987; BOUSSY et al 1988). Such P-M variation suggests that P elements are not at equilibrium in wild populations of Drosophila. Nevertheless, all strains derived from natural populations since 1974 have been shown to contain at least some P elements (ANXOLABEHERE et al. 1988; BOUSSY et al 1988). In contrast to natural populations, long-established laboratory strains are usually of the true M type; that is, they contain no P elements at all (BINGHAM et al. 1982; KlDWELL 1983; KlDWELL et al 1983; BREGLIANO and KlDWELL 1983). Two hypotheses have been proposed to account for this difference between lab and natural populations. The stochastic loss hypothesis (ENGELS 1981b) suggests that P elements have always been present in substantial frequencies in natural populations, and that their absence from long-established laboratory populations is due to loss of P elements from these strains by genetic drift. ENGELS (1991b) reasoned that small lab populations would be more subject to genetic drift and that the only stable state would be when there were no P elements. In contrast, the recent invasion hypothesis (KlDWELL 1983) posits that P elements did not exist among natural populations prior to the 1950s and 16 that P sequences recently invaded natural populations of D. melanogaster, spreading rapidly by replicative transposition. A prerequisite to the recent invasion hypothesis is that P elements must be able to rapidly invade true M populations rapidly. Several experiments have monitored the phenotypes of gonadal sterility and P cytotype in mixed P-M populations. KlDWELL et al. (1981) showed that, in the absence of measurable gonadal dysgenesis (20°C), mixed populations changed unidirectionally toward the P type. KlYASU and KlDWELL (1984) found that most mixed populations also evolved to the P type under conditions of strong sterility (27°C). These observed increases in the frequency of P type flies provide strong evidence that P elements can spread once introduced into a true M population. However, the use of the dysgenesis phenotype alone as a marker to monitor the spread of P elements is complicated by the fact that there are multiple dispersed copies of these elements within each genome. P strains usually contain 30 to 50 elements per individual. It is possible that only a few elements may be required to induce and/or suppress hybrid sterility. Therefore, mixed population experiments which monitor only physiological phenotypes can only suggest inferences about the dispersal of elements to new genomes and the accumulation of elements within populations or within individuals. Experiments utilizing Southern blot or in situ hybridization analyses demonstrated that P elements accumulate within individual genomes of inbred lines over several generations (DANIELS et al. 1987; PRESTON and ENGELS 1989). These inbred lines were established from flies which had acquired one or more elements by transformation. Results obtained from such lines are not applicable to natural populations (see Introduction of Chapter III, page 54). Furthermore, these experiments cannot demonstrate the dispersal of elements to new genomes since all flies used to initiate the populations contain at least one element. 17 To overcome the earlier shortcomings of using only physiological methods in the mixed populations, and also the limitations of studying transformed lines, we have previously investigated the dispersal and accumulation of P elements within mixed populations using molecular methods ( G O O D et al. 1989; M E I S T E R 1992). The invasion of P elements in natural populations of D. melanogaster was modelled by establishing laboratory populations with 0.5% and 5% P genomes. Two replicate populations at each frequency were monitored over twenty generations. The percentage of genomes that contained P elements was followed in each population by single fly ovary blots (see MATERIALS AND METHODS of CHAPTER II for a description of this technique). The distribution of P elements among individual flies from the two populations initiated with 5% P genomes was monitored by single fly Southern blots. Our results showed that the frequency of flies containing P elements increased each generation. The number of P elements within individual genomes decreased initially as expected, but then increased to equal or surpass the number of elements in the parental P strain. Finally, the distribution of P elements within the genomes of individuals from later generations varied considerably, and this pattern differed from the original P strain. These results suggest that an interaction between the assortment and recombination of chromosomal segments and some form of multiplicative transposition could result in the rapid spread of P elements in natural populations. Distribution of P elements in other species: Several groups have used hybridization under conditions of varying stringencies to search for P element homologous sequences within other species of Drosophilidae. An initial study detected no P homology within D. melanogaster sibling species which diverged some 2 million years ago ( B R O O K F I E L D et al. 1984). However, L A N S M A N et al. 18 (1985) later showed that P homologs were present within some species of every Drosophila species group in the subgenus Sophophora. The distribution within a species group can be discontinuous. For example, P sequences are apparently completely absent in Drosophila simulans S T U R T E V A N T and D. mauritiana which are the two most closely related sibling species of D. melanogaster. Interestingly, both the number of species in a species group that contained P elements and the similarity of the elements with the D. melanogaster element, were inversely correlated with the evolutionary relationship of the species groups. The first observation can be explained by either recent invasions or stochastic losses in several species. However, if species share elements due to vertical transmission from a common ancestor, then more closely related species should contain elements with higher sequence similarity. The second observation therefore argues that D. melanogaster acquired P elements through horizontal transfer from another species, followed by rapid dispersal through all natural populations. Several P elements from diverse Drosophilidae species have now been cloned and sequenced. These P element homologs are summarized in Table I. Examination of these sequences reveals that certain regions of the elements are more highly conserved, even within those elements which do not appear to be mobile. These regions include: (1) the inverted terminal repeats, transposase binding sites, and other regions required in cis for mobilization; (2) splice-junctions; and (3) the helix-turn-helix and leucine zipper motifs of the putative proteins. This conservation suggests that the elements may have only recently become immobile, or alternatively, that the elements containing these features are selectively maintained because they provide a selective advantage such as suppression of transposition. A comparison of the sequences from other Drosophila species confirmed the hybridization studies showing that more 19 CA c o CA o CA C « H CM O a to e cu s s ft* cu >fl H H WD PS cu H - CO cu o c tl cu 0-cu u c wo] u cu B O s u e CA fa P i o cu cj B cu M ,<u M >> M £ u a 6 CA g o 1/2 cu H Z 2; CD o a-co of co oo CA t—i o C ON ON Ru *J =3 b Q ID a CO OJ CO T 3 O cj c o a .-H CO ro o CJ c CO r o c o c 9 1 PC CO S If* IE CO CJ s Si CO «3 B O loo o 6 0 ro o cj a o c 5 2fr* oo CO Cu 13 c •c o c s •5b £ i s s CD •a a CD CO o 8 13 > c ca •a B O X! d> CN B s 2 ft< B O o t3 « 2 B a 13 CJ s 3 cu B o o s CJ B ~ <D O .8 a Q O f-( c^  </3 CD £ .S3 <D CS CD i H J 20 distantly related species often had P elements which were more similar to the D. melanogaster element. In fact, the P elements from Drosophila willistoni S T U R T E V A N T and D. melanogaster differ by a single base pair despite the two Drosophila species last sharing a common ancestor some 60 million years ago (DANIELS et al. 1990). This confirmed the recent invasion hypothesis and provided the "final nail in the coffin" to invalidate the stochastic loss hypothesis, at least with respect to the presence of P elements in D. melanogaster Hybridization studies have indicated that P element homologous sequences are rare outside of the Drosophilidae. Some sequences with low homology has been detected, but only within a minority of diptera and never in non-dipteran organisms ( A N X O L A B E H E R E and P E R I Q U E T 1987). The only two non-Drosophila P element homologs that have been sequenced were isolated from a genomic library of the Australian sheep blowfly Lucilia cuprina ( W I E D E M A N N ) (PERKINS and H O W E L L 1992). Pairwise comparison of nucleic acid sequences revealed that the two Lucilia elements each differ from the D. melanogaster P element (prc25.1) by approximately 42%; however, surprisingly, the two Lucilia elements also differ from each other by about 33%. Evidence from structural analysis and from genomic Southern blots suggested that the Lucilia elements are immobile; nevertheless, alignment of these elements with prc25.1 revealed regions of higher conservation similar to those noted above for the P element homologs of Drosophilidae species (PERKINS and H O W E L L 1992). Transposable Elements as Tools Transposons have proven to be valuable as tools for molecular geneticists. In particular, P elements have led to a revolution in Drosophila genetics. Besides the crucially important technique of germ line transformation (SPRADLING and R U B I N 1982; R U B I N and S P R A D L I N G 1983), the various uses of P 2 1 elements also include insertional mutagenesis or transposon tagging (BINGHAM et al. 1981; SEARLES et al. 1982), enhancer trapping (O'KANE and GEHRING 1987; BELLEN et al. 1989), targeted gene replacement (GLOOR et al. 1991), and generation of duplications and deletions in nearby genes (PRESTON et al. 1996). Many of these techniques have also been utilized successfully in several other well studied organisms for which convenient ITR transposons were available (see recent review by KAISER et al. 1995). The desire to apply these methods to the investigation of other less well characterized species, including insects of economic and/or medical importance, has led to efforts to use P elements as universal transformation vectors. The results of these efforts will be summarized in the INTRODUCTION to CHAPTER III. Aims of this Research: It has often been proposed that germline transformation could be utilized to introduce genetically engineered constructs into target insect populations. Such target populations might include insects which act as disease vectors or insects which are pests of agriculturally important plants and animals. For example, it has been suggested that effective long-term management of some insects could be achieved by the transformation and dispersal of either disease refractoriness (MILLER et al. 1997; KlDWELL and RIBEIRO 1992) or a conditionally regulated insecticidal system (PFEIFER and GRIGLIATTI 1996). Dispersal of engineered DNA might similarly be utilized to enhance useful insect populations, for example, by bestowing pesticide resistance to honey bees and other pollinators. These and similar proposals require not only transformation of economically important insects, but also that the engineered DNA be dispersed through the target insect populations. However, few methods for dispersal of engineered DNA have been suggested. 22 Transposable elements, especially ITR transposons such as the P elements of D. melanogaster, may provide ideal vehicles for dispersal of genetically engineered constructs. As described above, P elements have already proven to be valuable as vectors for transformation of D. melanogaster (SPRADLING and RUBIN 1982; RUBIN and SPRADLING 1983). In addition, we and others have demonstrated that P elements are capable of spreading very rapidly through experimental populations of D. melanogaster that were established with very low frequencies of P element containing individuals (MEISTER 1992; GOOD et al. 1989; KlDWELL et al. 1981; KlYASU and KlDWELL 1984). It is therefore plausible that mobile elements may be used as vectors not only for transformation, but also for dispersal of engineered DNA. In addition to the potential for rapid rates of dispersal of engineered DNA throughout a target population, TEs have the advantage of being generally species specific. Spread is usually limited to vertical inheritance within a single interbreeding population. Several fundamental questions must be answered before transposable elements can be seriously considered as dispersal vectors for engineered genes. Each series of experiments reported in this thesis addresses some of these questions. CHAPTER II asks whether or not a P element with a passenger gene is able to disperse and retain it's ability to encode a functional enzyme. CHAPTER III investigates if the ability to disperse is a property which is peculiar to P elements, or alternatively, whether hobo elements can also disperse. CHAPTER rV investigates the dynamics of the concurrent spread of P and hobo transposons and asks if the spread of one has a significant impact on the spread of the other. Finally, CHAPTER V develops a method of isolating strain-specific middle repetitive DNA from target insect species. Such DNA is likely to represent recently invading transposons which are still active. CHAPTER II: RAPID SPREAD OF A P ELEMENT/ Adh CONSTRUCT THROUGH EXPERIMENTAL POPULATIONS OF DROSOPHILA MELANOGASTER INTRODUCTION 24 One fundamental question that must be answered before transposable elements can be seriously considered as dispersal vectors for engineered genes is whether or not transposable elements, such as P elements, that have been modified by the insertion of foreign DNA are still capable of rapid dispersal through insect populations. The cis requirements for P element transposition have been well established and P element based transformation vectors are readily available (KARESS and RUBIN 1984; MULLINS et al. 1989). However, there are three lines of evidence indicating that transposition rates may decrease as element size increases. First, it has been found that transformation frequency is much higher for smaller elements (SPRADLING 1986). Second, P elements carrying the white eye color gene are less mobile than smaller elements containing the rosy eye color gene (ROBERTSON et al. 1988). Finally, analysis of six long-established lines that had been transformed with a helper P element and a P element carrying a rosy gene insert, revealed that of the many P elements in each line only one to three of the elements per individual contained rosy DNA (DANIELS et al. 1987). These results suggest that elements containing inserted passenger genes might be replicated and dispersed through insect populations at a lower rate than unmodified elements and/or that there is selection for internal deletions within the introduced P elements. Another question which must be answered is whether or not the inserted genes are replicated with good fidelity during the transposition process. There are two main structural classes of P elements; complete elements and internally deleted derivatives (O'HARE and RUBIN 1983). It is now generally accepted that the deleted elements arise by incomplete copying of a template element following excision of the original element (ENGELS et al. 1990; GLOOR et al. 1991). The 25 deletions often occur at the site of short duplications within the element (O'HARE and RUBIN 1983; ENGELS 1989), presumably because the duplications provide a region of homology at which the two invading strands can anneal during the SDSA repair process (see Transposition ofP elements in CHAPTER I). This model would explain why the resulting "deleted" element is missing one copy of the repeat plus the intervening sequence. When deletions occurred in a P element that contained an inserted rosy gene, it was demonstrated that one or both ends of the deletion may occur within the rosy gene (DANIELS et al. 1985). Obviously even very small deletions, or higher than average single base pair mutation rates, could result in loss of expression of inserted genes during transposition. In this chapter, I examine the rate at which a P element containing a 4.8 kb DNA fragment with a functional alcohol dehydrogenase (Adh) gene is dispersed through experimental D. melanogaster populations. I also ask whether or not the inserted gene retains its ability to code for a protein with enzymatic activity. My results show that the P element carrying a passenger gene spreads rapidly through the populations. The rate of this dispersal is comparable to the rate of dispersal exhibited by unmodified elements. Moreover, many of the Adh inserts still encode active alcohol dehydrogenase (ADH) proteins. 26 MATERIALS AND METHODS Drosophila Strains (1) b Adhn2 pr cn: Adh,n2 is an ethyl methanesulfonate induced allele of Adh-S derived from Canton-S (GRELL et al. 1968). Adhn2 has no detectable Adh activity (VlGUE and SOFER 1974). This strain contains the second chromosome markers b = black body (map position 48.5); pr = purple eyes (54.5); and cn = cinnabar eyes (57.5). This multiply marked strain was kindly supplied by W. SOFER (see O'DONNELLef al. 1975) and will henceforth be called simply Adhn2. This strain is a true M strain, that is, it contains no P elements (MEISTER, unpublished). ( 2 ) tAP-8A: A transformed strain which was derived from Adhf^^pr cn (O'DONNELL et al. 1975; BENYAJATI et al. 1983). This Adh null, true M strain was transformed with a P e\ement/Adh+ gene construct in which the functional 4.8 kb EcoRl fragment of the Adh gene (GOLDBERG 1980) was inserted into the EcoRl site of the P element vector pPL-1 (see ASHBURNER 1989, p. 1028). Transformation was carried out with the aid of p7t25.1 as a helper plasmid by J . POSAKONY. The tAP-8A strain was chosen for these experiments because it is unstable. That is, it contains integrated genomic copies of both the transposase-providing complete P elements, and the P element /Adh+ gene constructs. If large size is disadvantageous to the ability of P elements to spread, then the Adh+ providing constructs may not spread through the populations or may spread at a greatly reduced rate relative to the p7i25.1 elements. 27 Experimental Populations Experimental populations were maintained at 23°C on yeast-sucrose-cornmeal agar medium to which Tegosept (methyl-p-hydroxybenzoate) was added as a mold inhibitor. Populations were initiated with 800 Adhn% females. Either 1% or 10% of these females were mated to tAP-8A males; the remainder were mated to Adhn^ males (see Figure 2). This resulted in the introduction of a small proportion (0.5% or 5%) of genomes containing the P e\ement/Adh+ gene construct while minimizing the effects of fertility and genetic drift in the first generation of the experiment. Two populations were established at each of the initiation frequencies. Each population was maintained in 25 half-pint bottles. All the flies of each population were collected, pooled and mixed every generation. About 60 flies were then randomly distributed to each of 25 new bottles to establish the next discreet generation. Each population was monitored for 20 generations. Single Fly Ovary Blot Assays To test for the presence of P elements within individuals from each of the populations, ovary blot assays were done by a technique similar to the one used to study the distribution of P sequences in natural populations ( A N X O L A B E H E R E et al. 1985). Nitrocellulose filters were equilibrated on 2 thicknesses of Whatman 3MM filter paper that had been soaked in 5% SDS, 0.05 M EDTA. Ovaries were dissected from females onto the nitrocellulose. After 30 min, the nitrocellulose filters were floated on a 0.5 ml "puddle" of 0.5 M NaOH for 7 min to denature the DNA. The filters were then neutralized by soaking them twice in 1 M Tris-HCl (pH 7.4) for 2 min each. Finally, the filters were treated for 4 min with 1.5 M NaCl, 0.5 M Tris-HCl (pH 7.4), air dried and baked for 2 h at 80°C. The filters were wrapped in foil and stored so that several generations 28 P- /ADH- female X P+/ADH+ male X P-/ADH- male 10% , / 90% 1% 99% 2 replicate populations at each initiation frequency ^ allow random mating F2 F20 Figure 2: Setup of mixed experimental populations. The P-/ADH- flies are the Adh11^ strain that have no P elements and are null for ADH activity. The P+/ADH+ strain is tAP-8A which is derived from an ADH null strain that contained no P elements. tAP-8A has been transformed with the complete P element and a P element/ Adh gene construct which supplies ADH activity. 29 could be probed at once. This minimized variation due to the amount and intensity of the probe. Pre-hybridization and hybridization were carried out under conditions modified from MANIATIS et al. 1982. About 3 ml per filter of prewarmed prehybridization solution (50% formamide, 5X SSC, 50 mM sodium phosphate, 2X Denhardt's, 100 mg/ml sheared denatured salmon sperm DNA, 0.1% SDS) was added to filters in bags which were then heat-sealed and incubated for 3 h at 42°C. The 0.84 kb Hindlll fragment of prc25.1 (SPRADLING and RUBIN 1982; KARESS and RUBIN 1984) was nick-translated, filtered through a Sephadex G-50 spin column to remove unincorporated nucleotides (MANIATIS et al. 1982), and denatured. About 10 ng of probe DNA per ml of prehybridization solution was added to the bags. This internal P element fragment encompasses nucleotide positions 39 to 877 of the complete element and should hybridize to the majority of P elements: complete, internally deleted, and the P/Adh+ constructs. After hybridization at 42°C for about 14 h, the filters were washed 4 times for 10 min each in 3 mM Tris -HC1, 0.1% SDS and autoradiographed. Ovaries from two Adhn2 and two tAP-8A females were included on each filter as negative and positive controls respectively. In addition, each filter contained the ovaries of 25 females from the experimental populations. The amount of probe bound to ovary DNA from experimental flies was scored visually as positive or negative with respect to these controls. Any questionable dots were scored as negative. No attempt was made to quantify the intensity of the signal. However, to test the sensitivity of this assay, ovaries from several strains with known numbers of P elements were examined. Among the strains tested were tAP-1, tAP-3, tAP-5, and tAP-14 (GOLDBERG et al. 1983). Each of these strains is homozygous for a single, stable P element with an Adh insert. The ovaries from females of all three of these strains, as well as any strains with 30 more elements, always tested positive when compared to the Canton-S controls (see examples in Figure 3). Thus, it appears that this method can routinely detect an individual with only two P elements. Presumably this method can detect equally well those flies heterozygous for two elements at distinct loci and those flies homozygous for an element at one locus. Single Flv ADH Assays The histochemical assay for ADH activity described here is a modification of the one utilized by GRELL et al. (1968). After the ovaries of experimental flies were removed for the ovary blot assays, the remainder of each fly was squashed onto 3MM filter paper with the tip of an eppendorf tube. The filters were then incubated at 23-25°C for 5 minutes in a solution containing 125ml of 0.2M Tris (pH 9.0), 5ml of 95% ethanol, 3mg of phenazine methosulphate, 50mg of 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), and 200mg of B-Nicotinamide Adenine Dinucleotide. Squashes which contain ADH activity reduce MTT to the colored formazan and thus stain purple. Squashes which have no ADH activity turn a light pink due to background reduction of MTT. This simple colorimetric assay is semi-quantitative and quite sensitive (see results); however, experimental flies were scored only as positive or negative with respect to tAP-8A (+) and AdbpZ (-) controls. 31 1 # • • • Figure 3: Test of the sensitivity of the ovary blot assay. The right hand column, labelled C (for controls), contains ovaries from two Canton-S (true M strain) and two TC2 strain (P strain) females. The rest of the filter is divided into five rows, each containing five ovaries from a different test strain. For rows 1-5 respectively these strains are tAP-1, tAP-3, tAP-5, tAP-14, and Harwich. Each of these test strains is homozygous for a single P element/ Adh gene construct except for Harwich which is a strong P strain. This figure is reproduced from MEISTER 1992. RESULTS 32 Dispersal of P Element Sequences Four independent populations were established with 800 mated Adhn2 females. Two of these populations were initiated with females of which 10% had been mated to tAP-8A males. The other two populations were initiated with females of which 1% had been mated to tAP-8A males. In all cases the remainder of the females were mated to Adhn^ males. Each population was maintained as discreet generations. At generations 1 through 10, as well as generations 13, 15, 17 and 20, individual females from all four populations were tested for the presence of P element homologous DNA. This was accomplished by randomly selecting 200 females from each population and performing single fly ovary blots. Figure 4, panel A shows typical ovary blot results from generations 1,3,5, 8, and 10 of a population initiated with 10% tAP-8A mated females. Panel B shows similar results for the same generations of one of the populations initiated with 1% tAP-8A mated females. The right column of each filter contains the ovary DNA of two positive (bottom) and two negative (top) controls. The remainder of each filter contains the ovary DNA of 25 experimental females. With increasing generation there was a rapid increase in the number of experimental ovaries that contain P element homologous DNA. This was true regardless of whether the populations were initiated with 10% or 1% tAP-8A mated females. The percentage of flies containing P element sequences at each generation, from generations 1 through 10, for the two replicate populations initiated with 10% tAP-8A mated females is shown in Figure 5, panels A and B. In both populations, the number of individual flies that contain P element 33 lm* • • • • • • • B 8 . • • • • 10 Figure 4. Ovary blots showing presence of P elements: Representative ovary blots demonstrating the presence of P elements within individual flies from generations 1, 3, 5, 8, and 10 of one of the populations initiated with 10% tAP-8A mated females (panel A) and one of the populations initiated with 1% tAP-8A mated females (panel B). The 0.84-kb Hind III fragment of p7i25.1 was used as an internal P element probe. The top of the right hand column of each blot contains ovaries of two negative (AdhP^) controls. The bottom of the same column contains ovaries of two positive (tAP-8A) controls. In addition each blot has a 5X5 grid containing ovary DNA of 25 experimental females of the given generations. 34 lOO-604 AO 4 H H P H • 2 0 o 2 o H H Z H H z O 8 0 u H 6 0 2 P H o -1—1—1 1 1 1—1 1 1 r O l 2 3 4 6 6 7 8 9 1011 - i — i 1 1 1 1 1 1 1 r O l 2 3 4 6 6 7 8 9 IOII 1 OO-80-B 604 4 0 4 2 0 4 -1 1 1 1 1 1 1 1 1 r O l 2 3 4 5 6 7 8 9 1 O 11 IOO 804 604 40 4 2 0 4 D -T—1 1 1 1 1 1 1 1 r O l 2 3 4 6 6 7 8 9 IOII GENERATION Figure 5. Dispersal of P elements to individuals of tAP-8A populations. Panels A and B summarize the ovary blot data for generations 1 through 10 of the replicate populations initiated with 10% tAP-8A mated females; panels C and D summarize the same data for the two 1% populations. 35 homologous DNA increased dramatically until about generation 6. At this point approximately 90% of flies in each population contained detectable numbers of P sequences. From this point on the level ofP element containing flies plateaued. Figure 5, panels C and D, show very similar results for the percentage of flies containing P element DNA at each generation from 1 through 10 for the two 1% tAP-8A populations. In the initial generations of the 1% populations, especially the one represented in panel C, the dispersal of P elements appeared to lag somewhat as compared to the 10% populations. However, once the spread started, the increase in P element containing flies was dramatic. By generation 8, greater than 80% of the flies in both of the 1% populations had a detectable number of P sequences. Thus, the rate of increase in the percentage of individuals that contained P elements was similar in all four populations. It simply took several generations longer before the P elements in the 1% populations reached this rate of dispersal. The percentage of P element containing flies in all four populations remained greater than 90% after generation 10 (data not shown). Dispersal of ADH Activity The same 200 flies from each sampled generation that were tested for the presence of P element DNA were also assayed for ADH activity. ADH activity was determined by a colorimetric assay system (see MATERIALS and METHODS). Figure 6 demonstrates this assay for flies from strains that are positive and negative for ADH activity. I have also assayed flies with varying amounts of ADH activity (not shown). I can consistently score as positive any flies with 20% or more of normal ADH activity (data not shown). Figure 6. Demonstration of the ADH assay: This blot demonstrates the histochemical A D H assay for two negative Adhn2 control flies (left) and two positive tAP-8A control flies (right). 37 The percentage of flies that tested positive for ADH activity at each generation is shown in Figure 7. The populations are the same as those in Figure 5. The percentage of flies that were positive for ADH activity increased rapidly in all four populations and for each population this increase closely resembled the increase of flies with P element DNA (compare corresponding panels in Figures 5 and 7). Generally, for any given generation of all populations, the percentage of flies that contained P element DNA was slightly higher than the percentage of flies with detectable ADH activity. Phenotypes of Individuals With respect to presence of P elements and ADH activity there are four possible phenotypes within the populations. Where P+/" represents presence or absence of P element homologous DNA and ADH+/" represents positive or negative for ADH activity, these phenotypes are: P + and ADH+; P+ and ADH-; P" and ADH+; and finally, P" and ADH". Both the ovary blot assay for P elements and the histochemical assay for ADH activity were performed on each individual tested; therefore, all 200 flies tested at each sampled generation could be categorized according to these phenotypes. Figure 8 shows the distribution of the four phenotypes at representative generations for each of the populations. These generations were selected because they span the generations during which the major changes occurred within each population. The frequency of flies that were both P + and ADH + increased rapidly in all populations and then remained high. The frequency of flies that were both P" and ADH" underwent a corresponding rapid decrease and remained very low. The other two phenotypic categories of flies appeared to be present in low frequencies, but their frequencies were variable and were not consistently increasing nor decreasing. 38 IOO 804 H 4 0 4 o OH Q 2 0 4 -1—-1 1 1—1 1 1—1 1 r O l 2 3 4 B 6 7 8 9 IOII IOO 804 H y w OH 4 0 4 2 0 4 - 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1—r O l 2 3 4 6 6 7 8 9 IOII IOO 804 60 40 4 2 0 4 B - 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 — 1 — r O l 2 3 4 6 6 7 8 9 IOII IOO 804 60 40 4 204 D -1 1 1 1 1 1 1 1 1 r O l 2 3 4 6 6 7 8 9 1011 GENERATION Figure 7. Dispersal of ADH+ activity to flies of the tAP-8A populations. Panels A to D summarize the dispersal of ADH activity to individual flies from generations 1 to 10 of the same four tAP-8A populations represented in Figure 5. GENERATION • P+ \ADH+ E2 P+ \ A D H - B PAADH+ ^ P-VADH-Figure 8. Percent of flies of each phenotype in the tAP-8A populations. Panels A-D show the percent of individual flies that exhibit each of the four possible phenotypes at representative generations. The populations are the same as those shown in Figure 5 and Figure 7. DISCUSSION 40 The experiments described in this chapter model the introduction of low frequencies of flies which contain P elements, both with and without an active Adh+ gene insert, into randomly mating populations which originally contained no P elements and were Adh null. I was interested in whether or not the Adh+ containing constructs would be able to disperse through the populations, and whether or not the much larger constructs would be at a significant disadvantage relative to the standard p7i25.1 elements. Finally, if the modified elements were able to spread, I was interested in whether or not the Adh inserts would still code for an enzymatically active product. Dispersion of P Sequences The percentage of flies which contained P element homologous DNA, as measured by single fly ovary blots, increased rapidly with increasing generation in all populations (Figure 5). These ovary blots detect elements both with and without inserts, therefore, they should not to be taken as evidence for dispersal of the engineered construct. These data demonstrate that the P e\ement/Adh+ gene construct had no negative impact on the dispersal of P elements and confirm earlier results demonstrating that P elements can spread very rapidly in mixed populations of flies (KlDWELL et al. 1981; KlYASU and KlDWELL 1984; GOOD et al. 1989; MEISTER 1992). This rapid dispersal of P elements that are present in multiple copies per genome is likely to involve a combination of mechanisms including chromosome assortment and recombination, as well as some form of transposition that results in an overall increase in the number of elements (HlCKEY 1982; GlNSBURG et al. 1984; KAPLAN et al 1985; GOOD etui. 1989; see also more detailed explanation in DISCUSSION of CHAPTER III). 41 The probable mechanism for the multiplication of P elements will be discussed below. We have previously established mixed populations of flies using exactly the same method as was utilized in the experiments described here (GOOD et al. 1989; MEISTER 1992). Populations were initiated with Canton-S (true M) females; 10% of which were mated to TC2 males, while the remainder were mated to Canton-S males. Flies from the iz2 strain contain about 30 P elements per genome (BINGHAM et al. 1982). The frequency of flies that containeded elements was monitored through discreet generations using ovary blot assays. The percentage of flies that had detectable amounts of P sequences increased rapidly and reached a plateau of about 90% after 8 to 10 generations. The earlier results are thus very similar to those reported here, except that the earlier populations took several generations longer to reach the plateau. Slight differences in rate of dispersal of P elements are not surprising. If chromosome assortment and recombination play a role in the dispersal of the elements, then it is likely that the original number of elements per genome would have an effect on the rate of that dispersal. Furthermore, if transposition of the elements is important in the dispersal of the elements to new genomes, then any variable that has an effect on transposition rates would also affect the rate of spread. Such variables are likely to include temperature, element size, genomic position effects, and levels of both P transposase and any repressor molecules (see discussion by ENGELS 1989). Indeed, populations with large numbers of P elements are likely to contain a subset of elements that encode repressor activity and thus reduce the rate of element spread to naive members of the populations. It is therefore likely that the exact rate of transposable element dispersal will vary with each system tested. 42 Dispersion of ADH Activity The percentage of flies that exhibited ADH activity increased rapidly and immediately with increasing generation in the two 10% populations (Figure 7, panels A and B). A similarly rapid increase of ADH positive flies occurred in the 1% populations after an initial delay of three to four generations (Figure 7, panels C and D). These data demonstrate that P elements with a large insert are capable of rapid dispersal through D. melanogaster populations when introduced along with intact P elements. Apparently, any advantage in transposition that the intact elements might have relative to the approximately three-fold larger P element/Arf/i+ gene constructs is not sufficient to prevent or substantially retard the spread of the latter. These data also show that a significant proportion of the dispersed Adh+ inserts still encode a functional protein. In all four populations the spread of ADH positive flies appeared to lag slightly behind the dispersal of P element sequences (compare respective panels in Figures 5 and 7). This result is to be expected. The ovary blots detect P elements with and/or without inserts; whereas the ADH assay can detect only those elements which have inserts and encode active ADH enzymes. Since P elements both with and without the Adh+ gene insert are dispersing through the experimental populations, chromosome assortment and random segregation could give rise to individuals that contain only elements that lack inserts. Individuals containing such a subset of elements would test positive for P element sequences, but negative for ADH activity. On the other hand, if individuals arose that contained only elements with inserts, they should have both ADH activity and P sequences. However, due to differences in sensitivity of the two assay methods, these individuals may or may not test positive in both assays. 43 The ovary blot method for detection of P element homologous DNA can routinely detect individuals with two elements, and probably often a single element. The histochemical assay for ADH activity detects individuals with 20% or more of wild type diploid ADH activity. This means that a single Adh insert would have to provide 40% of the activity of a wild type Adh allele in order to be detected. Genomic position effects often result in lowered activity of genes transformed into D. melanogaster even though the gene is functional (SPRADLING and RUBIN 1983; GOLDBERG et al. 1983). Therefore, I would expect to have occasionally scored a fly as ADH" even though it contained a functional PI Adh construct. It should be noted that some selection for ADH positive flies, or for background genetic differences between the two founding Drosophila strains could occur. If Adhn2 eggs are placed on food containing 4% ethanol, none of them develops into adults (VlGUE and SOFER 1976). Tegosept is first dissolved in a small amount of ethanol before being added to our fly food. Less than 2% v/v of ethanol is added, and moreover, most of the ethanol evaporates when it is added to the hot (90°C) liquid food mixture. Nevertheless, some ethanol may remain. However, we routinely raise our Adh null Drosophila stocks on this food. In addition, viability and fertility tests of the Adhn2 and tAP-8A strains raised on this food have revealed no significant differences (data not shown). While I cannot rule out at least some positive selection for either ADH positive flies or for background genetic differences, it seems unlikely that selection could account for the large increase in ADH+ flies within all four experimental populations. In addition, the pattern of dispersal of the ADH activity in these populations is very similar to the pattern of dispersal of P elements in Adh+ M strains (GOOD et al. 1989; MEISTER 1992). This suggests that the dispersal of 44 ADH activity is a consequence of multiplicative transposition (see below) and dispersal of the P/Adh+ construct. Distribution of Phenotypes Since the ovary blot assays and the ADH assays were performed on the same individual females, the flies can be grouped according to four distinct phenotypes: P+ and ADH+; P+ and ADH'; P" and ADH+; P" and ADH". The distributions of the four phenotypes from selected generations of each population are shown in Figure 8. The frequency of flies that are both P + and ADH+ increased rapidly in all populations and then remained high. The frequency of flies that are both P" and ADH" underwent a corresponding rapid decrease and then remained very low. These two trends would be expected to occur as long as the majority of flies from later generations acquired at least one of the P element/Ac^ 4" gene constructs. The percentage of flies that are P+ and ADH" exhibited no obvious trend. If any generalization can be made, it is that the frequency of flies exhibiting this phenotype increased slightly in the middle generations, but then decreased. As discussed above, P elements both with and without Adh+ inserts are dispersing through the populations. The P + and ADH" class of flies may therefore include individuals which have, through random segregation, received only elements that lack Adh+ inserts. In addition, any individuals which received the P/Adh+ construct, but express the Adh+ gene at low levels due to genomic position effects, would be incorrectly included in this category. The intensity of the hybridization in the ovary blots is likely a rough reflection of the number of P element homologous sequences within the individual flies. Figure 4 therefore suggests that, in later generations, not only do more flies contain P elements, but also that individual flies contain more 45 elements (for each generation compare the intensity of hybridization to ovaries of experimental flies and to positive controls on the same filter). This is consistent with observations that the number of P sequences within individual flies increased with increasing generation in mixed P and M populations (GOOD et al. 1989; MEISTER 1992; CHAPTER IV this thesis). It seems likely that in the later generations the number of elements per fly increased and that most of the flies contained at least one element that supplies detectable ADH activity. The frequency of the P + and ADH" phenotype therefore decreased. The frequency of P" and ADH+ phenotype fluctuated with no apparent pattern; the percentage of flies that exhibited this phenotype only twice exceeded 10% and never exceeded 20%. This category should be very small, since most flies which obtain ADH activity should obtain the rest of the construct which includes P element DNA detectable by the P element probe. A few individuals of this phenotype could arise by targeted gene replacement of the in situ Adhn2 allele by a wild type Adh allele from one of the P/Adh+ constructs. Gene replacement could arise by homologous recombination, though this would be expected to be an extremely rare event. On the other hand, gene replacement could arise by excision of a resident PIAdh+ construct close to the Adh locus, followed by gap repair of the excision site by copying the wild type Adh gene from the P/Adh+ construct into the Adh locus by a gene conversion-like process (ENGELS et al. 1990; GLOOR et al. 1991). This process would require a source of transposase and a P element carrying an Adh gene inserted near the Adh locus. If both these criteria were met, then only about 1% of all progeny from such flies would copy any information into the Adh locus (GLOOR et al. 1991). Finally, by recombination and segregation, the progeny would have to lose both the original transposase source and the construct that provided the wild type copy of Adh. Clearly, even this relatively efficient process could account for very few 46 individuals of the P" and ADH + phenotype. I therefore believe that the majority of individuals within this phenotypic category are the result of differences in the resolving abilities of my scoring techniques. As discussed earlier, a single P/Adh+ construct would be detected in the genome by the ADH assay as long as it provided 40% of the ADH activity of a wild type allele; however, such a single construct might not be detected by the ovary blot assay for P elements which has a resolution of two copies. In addition, I was very conservative in interpreting both the ovary blot and ADH assays; any questionable results were scored as negative. In any case, the data do not indicate that the frequency of this phenotype has increased within the populations over time. Multiplication of P Elements As mentioned above, the intensity of the ovary blots imply that after the first few generations there is an overall increase in the number of P elements per fly. This accumulation of P elements within individuals occurs consistently in mixed populations of P and M type flies (GOOD et al. 1989; MEISTER 1992; CHAPTER TV). At the time of our earlier work, we suggested that the primary driving force for this increase in copy number was some form of transposition which results in a multiplication of elements. Despite substantial subsequent evidence that transposition of P elements is itself a conservative cut-and-paste event (see Transposition ofP Elements in CHAPTER I), there is now also a good model for the probable cause of this multiplication of P elements. Recall that, according to the SDSA model, when P elements transpose they leave behind a staggered cut at the element ends and the resulting single stranded ends invade an homologous template. DNA synthesis then proceeds independently from each end and, when a region of base complementarity is formed, the single strands anneal and use each other as a template to complete 47 the repair process. This strand invasion repair mechanism would result in multiplication of elements if a template containing an element were used preferentially. A recently invading element is not likely to be homozygous, but the repair could preferentially use the sister chromatid as a template (ENGELS et al. 1990), or even an element located elsewhere in the genome (GLOOR et al. 1991; ENGELS et al. 1994). Retention of the transposon ends at the excision site, as has been demonstrated for P elements (STAVELEYe* al. 1995; KEELER and GLOOR 1997), could create this preference for repair from aP element template since these ends would be involved in the search for homology. Moreover, since the PlAdh+ construct maintains the integrity of the P element ends, this same mechanism could also efficiently amplify the construct and thus assist in spreading the engineered DNA through the D. melanogaster populations. 48 CHAPTER III: RAPID DISPERSAL AND ACCUMULATION OF HOBO ELEMENTS WITHIN MIXED EXPERIMENTAL POPULATIONS OF DROSOPHILA MELANOGASTER INTRODUCTION 49 The transformation of germline cells by P element mediated transformation has led to the development of a variety of genome manipulation techniques and revolutionized the study of gene expression and function in Drosophila (see Transposable Elements as Tools in Chapter I; KAISER et al. 1995). There is obviously a strong desire to apply these techniques to the study and manipulation of economically or medically important insects. P element mediated transformation was successful in several Drosophila species including D. simulans (DANIELS et al. 1985), and the much more distantly related D. hawaiiensis ( B R E N N A N et al. 1984). In contrast, discouraging results were obtained from attempts to utilize P element transformation in mammalian cells and in non-drosophilid insects including mosquitoes, tephritid fruitflies, grasshoppers and houseflies (CLOUGH et al. 1985, KHILLAN et al. 1985; MILLER et al. 1987; M c G R A N E et al. 1988; O'BROCHTA and HANDLER 1988; MORRIS et al. 1989; WALKER 1989; HANDLER et al. 1993; R. LANSMAN, H. BROCK and T. G R I G L I A T T T , unpublished results). In these experiments, transformed cells were only rarely obtained and were always the result of illegitimate recombination as opposed to transposition events. This inability of D. melanogaster P elements to efficiently transform species outside of the Drosophilidae, likely reflects the requirement for specific host encoded functions in the transposition process. The general acceptance that P elements may not act as efficient transformation vectors in non-drosophilids has led to the investigation of the utility of transformation vectors based on other transposons. The hobo element, originally isolated from D. melanogaster as an insertion just 5' to a Drosophila glue protein gene (MCGlNNIS et al. 1983), is one element that shows promise. 50 Despite having no sequence homology, hobo and P elements resemble each other both structurally and genetically (STRECK et al. 1986; BLACKMAN et al 1987). Elements from both families have characteristic short inverted terminal repeats, produce 8 base pair duplications of genomic DNA at their sites of insertion, transpose as DNA intermediates, and are normally present as complete elements as well as internally deleted derivatives. Mobilization of either family of elements causes gonadal dysgenesis which is a correlated group of symptoms that includes male recombination, gonadal sterility, and increased levels of mutation. Finally, like P elements, hobo elements are able to mediate germ-line transformation of D. melanogaster (BLACKMAN et al. 1989). Three lines of evidence suggest that hobo elements may act as more universal transformation vectors than P elements. First, besides being present in a variety of Drosophila species (DANIELS et al. 1990; HANDLER and GOMEZ 1995), elements with significant homology to hobo have been identified in diverse insect species. Among the many hobo-like elements in dipterans are Hermes from the housefly Musca domestica LINNAEUS (O'BROCHTA et al. 1996), hermit from the Australian sheep blowfly L. cuprina (COATES et al. 1996), and Hector from the Australian bushfly Musca vetustissima WALKER (WARREN et al. 1995) . There are also hobo-like elements present in at least some lepidoptera (DEVAULT and NARANG 1994) and in the nematode C. elegans (BIGOT et al. 1996) . Second, it has been shown that hobo is capable of interplasmid transposition when injected into embryos of a wide variety of diptera including M. domestica and the Queensland fruitfly Bactrocera tryoni (FROGATT) (O'BROCHTA et al. 1994). The closely related Hermes element can also transpose in diverse dipteran species including D. melanogaster, B. tryoni and four families of cyclorraphan flies which are significant agricultural pests (O'BROCHTA et al. 1996; SARKAR et al. 1997), as well as the yellow fever mosquito Aedes aegypti 51 (JASINSKIENE et al. 1998). Moreover, both hobo and Hermes can transpose in at least one lepidoptera, the corn earworm Helicoverpa armigera (HUBNER) (PlNKERTON et al. 1996). Third, hobo elements belong to a broad superfamily of elements whose members tend to be able to transform heterologous species. The elements of this superfamily, sometimes called the hAT elements, are grouped together based mainly on the types of donor sites left behind after element excision and on limited sequence similarities in their transposase coding regions (STRECKet al. 1986; CALVIe* al. 1991; ATKINSONS al. 1993). Besides the hobo-like elements described above, the other primary members of the hAT family are the Ac element of Zea mays LINNAEUS (corn) and the Tam3 element from Antirrhinum majus LINNAEUS (snapdragon). Ac and Tam3 transformation have been demonstrated in a wide variety of plants including both monocotyledonous and dicotyledonous species (BAKER et al. 1986; YODER et al. 1988; KNAPP et al. 1988; VAN SLUYSef al. 1987; HARINGe* al. 1989; CHUCK et al. 1993; PETERSON and YODER 1993; HEHL 1994). More recently, the ability of hobo elements and hoho-like elements to act as germline transformation vectors in species other than those from which they were originally derived has been investigated. For example, the D. melanogaster hobo element has been used as a transformation vector in Drosophila virilis STURTEVANT (LOZOVSKAYA et al. 1996) and in B. tryoni (O'BROCHTA and ATKINSON 1996); and the M. domestica Hermes element can act as a very efficient transformation vector in D. melanogaster (O'BROCHTA et al. 1996). I have demonstrated that a P element modified by a large insert containing an active alcohol dehydrogenase gene is still capable of spreading through experimental Drosophila populations and that most of the recipient flies gain alcohol dehydrogenase activity (MEISTER and GRIGLIATTI 1993; CHAPTER II this thesis). This supports the proposal that transposons may be used as 52 vectors not only for transformation, but also, for rapid dispersal of engineered DNA throughout a target insect population. However, due to the inability of P elements to transform non-drosophilids, other transposable elements will need to be considered as vectors for dispersal of engineered DNA. The likelihood that hobo elements may transpose within and act as transformation vectors in a broad range of species provides a compelling reason to investigate the ability of hobo elements to disperse. In addition, unlike P and most other transposons, there is evidence that hobo can undergo significant transpositional activity in at least some populations containing many elements (BLACKMAN and GELBART 1989; HATZOPOULOUS et al. 1987; LlM 1988; HARADA et al. 1990; PASCUAL and PERIQUET 1991), although this is clearly not always the case (DOMINGUEZ and ALBORNOZ 1995; NUZHDIN and MACKAY 1995). Frequent transposition of hobo elements within element containing individuals may allow hobo elements to be used as vectors for transformation and dispersal even in insect populations which already contain them. Little is known about the ability of hobo elements to disperse. However, it is believed that hobo elements, like P elements (see CHAPTER I), have recently invaded natural populations of D. melanogaster. There are at least four lines of evidence which support this hypothesis. First, the sequence similarities of the primary members of the hAT superfamily of elements are considerably more similar than would be expected based on vertical transmission and independent divergence from a common ancestral element (CALVI et al. 1991). Second, the very narrow species distribution of hobo within Drosophila suggests that hobo was not present in the common ancestor of drosophilids (STRECK et al. 1986; DANIELS et al. 1990; BOUSSY and DANIELS 1991). Third, there is a correlation between collection date of D. melanogaster strains and the presence of hobo elements in their genomes. All strains that lack hobo elements are from older 53 collections, while all recently collected strains contain hobo elements (PERIQUET et al. 1989a; 1989b; DANIELS et al. 1990; PASCUAL and PERIQUET 1991). Finally, there is a very low level of sequence divergence between the hobo elements present in D. melanogaster, D. simulans and D. mauritiania suggesting that horizontal transmission may have occurred between these species (SIMMONS 1992). The apparent recent invasion of natural populations of D. melanogaster by hobo elements provides circumstantial evidence that hobo elements are capable of rapid dispersal and accumulation from a few individuals to entire populations. However, there is little experimental evidence to confirm this ability. Indeed, experiments monitoring the invasion of hobo transposons within genomes of transformed lines of D. melanogaster (LADEVEZE et al. 1994, 1998; GALINDO et al. 1995) cast some doubt on this conclusion. These experiments examined the hobo distribution in lines of D. melanogaster which originally contained no hobo elements and which had been transformed with autonomous (complete) hobo elements. In situ hybridization to polytene chromosomes revealed that until generation 52 each of six transformed lines accumulated an average of only four to six heterozygous and/or homozygous insertion sites of hobo elements per individual. Even if all sites were homozygous, this implies a maximum copy number of 12. In contrast, individuals from natural hobo containing strains have 50-100 copies (STRECK et al. 1986; YANNOPOULOS et al. 1987). The results with hobo transformed lines are also in contrast with previous results of lines transformed with autonomous P elements. Southern blots indicated that P elements had accumulated to about 110 copies per individual within a similar time frame (DANIELS et al. 1987). Even after 100 to 115 generations the hobo transformed lines reached a maximum of six to nine insertion sites, and the copy number actually appeared to be decreasing in some lines. Since an increase in 54 copy number per genome is expected to contribute to the ability of transposons to spread through populations (HlCKEY 1992), these experiments on transformed lines suggest that hobo elements may be much more restricted in their ability to spread through populations than are P elements. In addition to the studies on transformed lines, the dispersal and accumulation of P elements has also been examined in mixed populations initiated by introducing a few element containing genomes into populations otherwise devoid of elements (GOOD et al. 1989; KlDWELL et al. 1981; KlYASU and KlDWELL 1984). The studies of transformed lines have the advantage of assuring that there is no contamination by other elements or by genomic factors. Nevertheless, it should be noted that the two types of experiments, transformed lines versus mixed populations, may produce quite different dynamics and may model different stages of the invasion of natural populations by transposable elements. The primary differences result from distribution of elements within the initial populations: the transformed lines starting out with one or a few elements in every individual, whereas mixed populations start out with many elements in only a small subset of the flies. During the invasion of natural populations it is likely that one or a few transposable elements originally enter a single individual of the species by horizontal transfer. The elements would then spread to new genomes and increase in copy number within individuals of an isolated population. A few migrants from this population would then allow the elements to spread to other nearby populations and eventually worldwide. If this scenario of element spread does occur, then the study of transformed lines would in some ways resemble the initial horizontal transfer of element(s) from another species, and the early increase in copy number of elements per fly within an isolated population. However, the fact that all flies within the transformed lines initially contain 55 elements somewhat detracts from this analogy. In contrast, the study of mixed populations more closely resembles the spread of elements from the initially contaminated flies to the rest of the population, and certainly mimics the spread to further populations by migration. The study of mixed populations, where individual element donating flies contain many element copies, may also be more analogous to cases where the initial invasion is by interspecific hybridization rather than by horizontal transfer. Since exceptional fertile hybrids have been reported between D. melanogaster and its hobo element-containing sibling species, interspecific hybridization has been postulated as a possible mechanism for hobo element invasion ( P E R I Q U E T et al 1994). In this chapter I present data from experimental populations which were set up to introduce small proportions of hobo containing genomes into D. melanogaster populations which previously contained no hobo elements. The presence and quantity of hobo element DNA within these mixed populations was monitored at selected generations over 20 successive generations by quantitative dot blots of DNA prepared from multiple flies. In addition, both the presence and amount of hobo element DNA within individual flies were examined both by dot blots and by Southern blots. As an indicator of the development of hobo regulatory ability in the populations, female flies were also tested for the ability to suppress gonadal dysgenisis. MATERIALS AND METHODS 56 Experimental Populations: All populations were initiated with 800 D. melanogaster females from the Canton-S strain (CS). This strain contains neither hobo nor P elements ( B L A C K M A N et al. 1987; Y A N N O P O U L O S et al. 1987; B I N G H A M et al. 1982). Either 2% or 20% of these females were mated to males of the Oregon R-S strain which contains hobo elements but no P elements ( B L A C K M A N et al. 1987). The remainder of the females were crossed to CS males. This resulted in the introduction of 1% or 10% of hobo element containing genomes into naive populations, while minimizing the effects of genetic drift and fertility differences in the early generations. Three populations were set up at each of the initiation frequencies. To distinguish between these six populations when referring to them individually, I will henceforth call these the 1% and 10% OR-A, OR-B, and OR-C populations. Each population was maintained for 20 discrete generations in the same way as those described in the MATERIALS AND METHODS of CHAPTER II. DNA preparation: The following method of preparing DNA from single flies is a modification of the method I had used previously ( G O O D et al. 1989) and was adopted because it gave more consistent digestion with Xhol. Individual female flies were homogenized in 100 ul of a solution made up of 0.125 M Tris-HCl (pH 8.5), 0.08 M NaCl, 0.05 M EDTA, 0.16 M sucrose, and 0.5% SDS. The homogenate was transferred to a 1.5 ml micro-centrifuge tube and 100 pi of freshly prepared proteinase K solution (0.125 % SDS, 0.3 M Tris-HCl pH 9.0, 0.1 M EDTA, 5% sucrose, and 250 ug/ml proteinase K ) was added. Digestion was allowed to 57 proceed for a miiiimum of 3 h at 50°C and the samples were then extracted once with phenol chloroform and once with chloroform. The DNA was precipitated by the addition of two volumes of ice-cold 95% ethanol and spinning for 20 min at maximum RPM in a micro-centrifuge. The resulting pellets were washed with 70% ethanol and dried in a vacuum dessicator for 5 minutes. If the samples were to be analyzed by Southern blot then the pellets were resuspended directly in restriction enzyme mix (buffer, Xhol and RNase A). If the samples were to be analyzed by dot blot then the DNA pellets were resuspended in TE with RNase A. The preparation of DNA from multiple flies was performed essentially as described above for single flies. About 25 mixed male and female flies were homogenized together in 750 ul of the homogenization buffer and then 750 pl of the proteinase K solution was added. After proteinase K digestion, two such 25 fly homogenates were pooled for each of the tested generations. Following extraction, precipitation, and washing of the samples, the pellets were resuspended in TE with RNase A. Digestion was allowed to proceed for at least 30 min at 37°C. To remove the ribonucleotides the samples were again extracted once with phenol chloroform and once with chloroform, ethanol precipitated, and washed with 70% ethanol. Finally, the DNA preparations were resuspended in TE and their concentrations were determined using spectrophotometry. Probes; All DNA fragments used as probes were gel purified using standard methods and were labeled with 32p using a random primed DNA labeling kit (Boeringer Mannheim, Mannheim, Germany). A mixture of three labeled internal fragments, the 0.7 kb .EcoRI/ifindlll fragment and the two end fragments extending from the two EcoRl sites to the cloning site in the plasmid 58 pRG2.6X (BLACKMAN et al. 1987), were used for the hobo element probe. These fragments comprise approximately 80% of the hobo element, and should hybridize to the majority of both complete and internally deleted elements. The amount of DNA in each sample on each Southern blot or dot blot was measured using probes for two single copy regions (present once per haploid genome). The labeled fragments used to probe for single copy DNA were: 1) the 2.6 kb and 2.1 kb BaraHI/EcoRI fragments of the alcohol dehydrogenase gene as contained in the sAC-1 plasmid (GOLDBERG 1980), and 2) the 2.1 kb BamHl-Xhol fragment from the suppressor of forked region contained in the plasmid pBl.L (MlTCHELSON et al. 1993). For the Southern blots, the fragments of either one of these single copy regions was utilized. However, mixtures of the fragments from both the single copy regions were used for the dot blots. This effectively increased the "length" of single copy sequence used as a probe on the dot blots and thus increased the intensity of the hybridization obtained. Quantitative dot blots: About 2.5 pg of each sample DNA was loaded for dot blots of DNA prepared from multiple flies. For dot blots of DNA samples prepared from single flies, the entire DNA prep was loaded without any attempt to pre-determine the amount of DNA present. All DNA samples to be dot blotted, whether prepared from single flies or from multiple flies, were made up to a volume of 58 ul in TE and were transferred to a 96 well tissue culture plate in the same position that they were to be blotted. TE (58 pi) was added to any empty wells. Following addition of 2 pi of 0.5 M EDTA and 40 pi of 1.0M NaOH to each well, the samples were left at 37°C for 30-60 min to ensure that the DNA was denatured. The samples were then pipetted to Hybond-N nylon membranes (Amersham, Arlington Heights, IL) using gravity filtration through a Bio-Dot microfiltration 59 apparatus according to the manufacturer's instructions (Bio-Rad Laboratories, Richmond, CA). After air drying, the membranes were irradiated for 4-5 min on a UV transilluminator. All hybridizations and washes were then carried out at high stringency using the Hybond-N manufacturer's protocols (Amersham). All dot blot membranes were first hybridized to a single copy probe (see above). Digital images were produced with phosphor screen autoradiography using the Storm system (Molecular Dynamics, CA). The single copy probe was stripped from the membranes by incubating in a shaking water bath at 50°C for two washes in 0.4 M NaOH at 1 h per wash, followed by 1 h in 0.1 X SSC, 0.1 % SDS and 0.2 M tris-HCl (pH 7.5). A digital image was produced to check for complete removal of the probe and further stripping was performed as required. The membranes were then hybridized to a hobo element probe and digital images were again produced. Quantitation of the dot blot membranes was performed to determine the amount of hobo element hybridizing DNA within the samples from the OR initiated populations. In all cases, the amount of hobo hybridizing DNA was expressed as a percentage of that present in the OR hobo donating strain. I very carefully adjusted for variation in quantity of DNA in samples which arises due to slight differences in loading, transfer and binding of DNA to the membranes. In addition, I have scrupulously corrected for both specific and/or non-specific background hybridization of each probe. The overall analysis for each dot blot membrane consisted of four primary steps, each of which will be further explained in the paragraphs below. The detailed quantitation of one dot blot by the application of these four steps is also demonstrated in Appendix 1. In summary, these four steps are: 1) to measure the volume and subtract the background for each dot on the digital image produced by the single copy probe; 2) to use these volumes to perform a linear regression analysis to determine the 60 amount of DNA in each dot; 3) to measure the volume and subtract the background for all dots produced by the hobo probe; and 4) to compare the hybridization of the hobo probe to experimental samples and to a gradient of DNA from the original hobo donating strain. The volumes (total linear intensities of all the pixels in the area) of all dots on each digital image produced with the single copy probe were determined utilizing ImageQuaNT software according to the supplied user's guide (Molecular Dynamics, CA). For background correction, the mean volume of all the dots that contained no DNA was calculated for each image and this value was subtracted from the volumes of all other dots on that same image. The amount of DNA present in each experimental dot was calculated by linear regression analysis (see below). This method of measuring the amount of DNA actually present on the membranes corrects for the inherent variability that is often encountered in the transfer and binding of DNA to nylon membranes by the dot blot method (see A N C H O R D O G U Y et al. 1996). Such quantitation was especially necessary for blots containing DNA prepared from single flies because, unlike the DNA prepared from multiple fly samples, the concentrations of DNA in the samples were not pre-determined using spectrophotometry. A gradient of known amounts of CS strain DNA was included on each membrane to generate a linear regression equation. Linear regression equations were generated using the data analysis capabilities of Excel 5.0 (Microsoft software). For each membrane, the volumes of the CS gradient dots on the image produced by the single copy probe were plotted against the micrograms of CS DNA loaded (as determined by spectrophotometry). The amounts of DNA loaded in these gradients were selected to ensure that the range of CS DNA samples completely encompassed the amounts of DNA in all experimental samples on the same membrane. For 61 typical gradients included on multiple fly and single fly dot blots see Figure 10 and Figure 12 respectively. For all membranes utilized in quantitative analysis, the linear regression equations were generated from the volumes of a minimum of six CS gradient dots. There was an excellent linear correlation between the volumes of the gradient dots and the micrograms of CS DNA loaded; the correlation coefficients varied from 0.98 to 0.99 (confidence level of 95%). The individual volumes of the dots produced by the single copy probe hybridized to the experimental samples were then used to calculate the amount of DNA in each dot using the linear regression equation generated by the CS gradient present on the same blot. Experimental dots whose volume was not spanned by the gradient dots were not included in the final data. The volumes of all dots on the digital images produced with hobo element probe were also measured using ImageQuaNT. Background correction for the images produced by the hobo element probes was complicated by the slight hybridization obtained to CS control strain DNA which contains no canonical hobo elements , but does contain hobo homologous DNA (for example see CS gradient in Figure 10B). This hobo homologous DNA is present in virtually all Drosophila, including the strains used in our experiments, and will be discussed further below. To account for this background hybridization, the pixel volumes of the CS gradient dots on the image produced by the hobo probe were plotted against the micrograms of CS DNA present in each dot. This generated a linear regression equation for each image which could then be used to calculate that proportion of the pixel volume of experimental dots which was due to background hybridization. This background value was calculated for each experimental dot individually and was subtracted from the total pixel volume of that dot. 62 Once the amount of DNA in each sample on the membranes had been determined and the background had been subtracted from the volumes of the dots on all images produced by the hobo element probe, then the amount of hobo hybridizing DNA of experimental samples could be calculated as a percentage of that present in the OR strain. In addition to the CS gradient, each dot blot membrane contained a gradient of OR strain DNA. The volumes of the dots produced on each image by the hybridization of the hobo element probe to the gradient of OR strain DNA were plotted against the micrograms of DNA in the dots. For each image, this generated a final linear regression equation from which the volume corresponding to 100% of element donating strain hobo hybridizing sequences could be calculated for any given amount of DNA. The amount of hobo hybridizing DNA in the experimental samples (as a percentage of that present in the OR element donating strain) was then calculated using the formula: volume of experimental dot with hobo probe . . . . . X 100% volume of OR strain hobo for the amount of DNA in experimental dot (from gradient) It is worth noting that in the final calibration of percent of OR strain hobo homologous DNA I could have used a gradient of increasing amounts of OR strain DNA mixed into CS strain DNA, instead of the simple dilution series of OR DNA. However, I have experimentally demonstrated that the type of gradient used made no difference in the final results as long as the corrections were made for the high molecular weight hobo homologous DNA as described above (see Appendix 3). In fact, this experiment also provided a control to 63 check if the subtraction of the hobo homologous DNA was working correctly. Moreover, by using a simple dilution series, the OR strain gradient could also be used instead of, or in addition to, the CS gradient when determining the amount of DNA in the dots. Quantitative single fly Southern blots: DNA preparations from single female flies of the OR initiated populations were also analyzed by Southern blot. DNA samples were digested with Xhol, separated on 1% agarose gels, transferred to nylon membranes, and probed with radioactively labeled hobo fragments. Several autoradiograms of varying exposures were made on Kodak X-OMAT film to ensure that most bands would fall within the linear range of the film on one of the exposures. To allow for normalization for the amount of DNA loaded per lane, the hobo probe was removed and the blots were reprobed with one of two single copy probes (see above). Note that digestion with Xhol cuts the hobo element twice, thus yielding one internal fragment for each hobo element. When probed with internal hobo fragments these blots will therefore reveal one band for each size class of hobo element present in the genome. For example, Xhol digestion of the OR-S strain yields two bands of about 2.6 kb and 1.5 kb (see rightmost lane of the Southern blot in Figure 13). These bands correspond to 4-5 copies per genome of the presumably complete hobo element and about 55-70 copies per genome of a 1.9 kb deleted derivative (BLACKMAN et al. 1987). Both the OR-S and CS strains also contain high molecular weight hobo homologous DNA which corresponds to a ftooo-related transposable element which is no longer actively mobile and which has about 80% identity to the canonical hobo element (BOUSSY and ITOH 1996). 64 All autoradiograms of Southern blots were converted to TIFF image files using a video capture board and were analyzed using NIH Image version 1.57 (available via anonymous ftp from alw.nih.gov). To estimate the amount of hobo hybridizing DNA within bands from experimental samples, as a percentage of that in the OR strain, I followed a procedure similar to that described above for the dot blots. There were two main differences. First, the values compared were mean densities (i.e. the average gray value of the pixels within the selected bands) instead of volumes (total linear intensities of all the pixels). This difference is simply due to the two different analysis software programs used. The second difference is that the Southern blots contained only positive (OR) and negative (CS) controls to use as standards, as opposed to the gradients available on the dot blots. This second difference is not trivial; it makes it impossible to quantitate the amount of DNA per lane using linear regression analysis. Nevertheless, the amount of DNA per lane was normalized and an estimate of the amount of element homologous DNA present in each band, as a percent of that present in the element donating strain, was generated using the formula: (mean density of experimental band with hobo probe) (mean density of experimental band with single copy probe) . . . , x 100 % (mean density of OR strain band with hobo probe) (mean density of OR strain band with single copy probe) This value was calculated for each of the two size class of hobo element within each sample. The formula assumes that the density of gray on the autoradiograms is both linear and proportional to the amount of DNA on the membranes. The availability of various exposures allowed selection of exposures ; 65 where the bands being normalized fell within the linear range of the film, so I believe these assumptions are reasonable. Some flies analyzed by Southern blot displayed new sizes of hobo elements not present in the OR-S strain. I determined the approximate size of these new bands with Gel Frag Sizer software (available by anonymous ftp from net.bio.net). Since not all gels contained conventional DNA size standards, I first determined the size of a high molecular weight hobo homologous band which was visible on all autoradiograms. I thus had three DNA fragments of known size to use as standards on all Southern blots: the hobo homologous band (4.1 kb), the 2.6 kb hobo band corresponding to the full sized element, and the 1.5 kb hobo band corresponding to the deleted derivative present in the OR-S strain. Gonadal Dysgenesis Assay: In the OR strain the transposition of hobo elements is repressed ( B L A C K M A N et al. 1987; HO et al. 1993). This repression is removed and gonadal dysgenesis is induced when OR males are mated to females from strains which lack these regulatory mechanisms. Mating experimental females to OR males and checking progeny for dysgenic ovaries is therefore an indicator of the development of hobo regulatory ability within the populations. Such matings are referred to as A* crosses ( K l D W E L L and NOVY 1979; P A S C U A L and P E R I Q U E T 1991). Ten to twelve virgin females were randomly selected from each of the 1% OR populations at generations 5, 8, 10, 15 and 20. After allowing them to age for 2-3 days, each female was placed in a vial with 3 OR males at 25°C. Several days later the parents were discarded. About 3 days after the onset of eclosion, female progeny were transferred to fresh medium and aged 2-3 more days at 66 25°C to allow their ovaries to mature. Approximately 10 female progeny from each of the original females were dissected and scored for dysgenic ovaries (see Figure 9). The percentage of dysgenic ovaries for each tested generation of each population was then calculated as: number of dysgenic ovaries x ioo% 2 X (number of females dissected) RESULTS Changes in total hobo element DNA within populations; The experimental populations initiated with 1% and 10% OR (hobo containing) genomes were first examined for changes in the total amount of hobo homologous DNA with increasing generation. DNA was prepared from multiple flies of all six populations from each of the first 15 generations as well as various generations from 16 to 20. A 2.5 pg aliquot of each sample DNA was dot blotted onto each of two nylon membranes. Figure 10 shows phosphor images of a sample dot blot membrane containing DNA from multiple generations of two of the experimental populations initiated with 1% OR genomes. The relatively constant intensity of the dots produced when the blot was hybridized to a single copy probe (panel A) demonstrates that a similar amount of each experimental DNA sample was fixed to the membrane. Panel B shows the image produced when this same membrane was hybridized to a hobo element probe. From a simple visual inspection there is a clear increase in hobo element homologous DNA within the first four to five generations of both populations. In contrast, ;ure 9: Examples of normal (bottom) and TE induced dysgenic ovaries (top) in Drosophila. 68 Panel A: single copy probe 1 % OR-C 1 % OR-A CS OR 3 4 5 6 10 11 12 13 14 15 18 19 8 10 11 12 13 14 15 16 18 20 0.0 0.2 0.4 0.6 0.8 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Panel B: hobo element probe 1 % OR-C 10 11 12 13 14 15 18 19 1 2 10 11 12 13 14 15 16 18 20 1 % O R - A c s 0.0 0.2 0.4 0.6 0.8 OR Figure 10: Sample dot blot showing increase in hobo homologous DNA within populations. (A) Autoradiogram of dot blot filter containing DNA from two of the populations initiated with 1% OR genomes hybridized to a single copy probe to allow normalization for the amount of DNA per dot. The top four rows contain approximately 2.5 |ig of DNA prepared from multiple flies of the generations given below each dot. The bottom two rows contain gradients with increasing amounts of CS and OR DNA (values between these two rows indicate micrograms loaded). (B) Autoradiogram of the same filter hybridized to a hobo element probe. 69 the hybridization to the samples of DNA from higher generations remained more or less constant. The intensity of hybridization of the 2.5 pg experimental DNA samples from both populations appears to have stabilized at about the intensity obtained with 1.0 to 1.5 pg of OR strain DNA. In order to quantify the accumulation of hobo elements described above, the DNA within each dot on all blots was normalized using the single copy gene probe and the amount of hobo element DNA per sample was calculated as a percentage of that present in the hobo donating OR strain. The three 1% OR initiated populations were each tested in duplicate and the data are summarized in Figure 11, panels A - C . As suggested by visual inspection, with increasing generation there was an immediate and rapid increase in hobo hybridizing DNA within all three populations. The quantitative analysis reveals that this rapid increase continued until about generation eight, at which time each population had between 35% and 45% of OR strain hobo. The rate at which the hobo DNA increased at each subsequent generation then slowed substantially, though the curves do not appear to have plateaued completely. By the end of 20 generations each of the 1% OR initiated populations had 40-50% of OR strain hobo DNA. Figure 11 panels D-F show the accumulation of hobo element DNA in the three 10% OR initiated populations. As with the 1% OR populations, there was an immediate and rapid increase in hobo hybridizing DNA within the early generations. For the populations represented in panels D and E, this period of rapid increase in hobo DNA ended at about generations four and nine respectively. The amount of hobo DNA in these populations then reached a plateau of about 40-50% of that present in the OR strain. The final 10% OR initiated population also reaches about 50% of OR strain hobo by generation 15, though the actual dynamics of this accumulation are less clear (panel F). 60 50 •= 40 a o a o -a o •c DC O "o 30 20 10 o o Q 6 ? ° ° o ° 1% OR-A «0 SO c o S3 I 40 Q. O a. o •Q o •e BC O 30 20 10 D & o A A o o 0 A o ? 10% — , — . i . — i OR-A 10 generation 15 20 10 15 generation 20 o •Q O •c oc o "o 8* 60 B <o E 50 50 40 o c o •a a 40 o ° ° oo o o Q o o 3 a. o o CL o o 30 o ° U|0 30 o « o o 20 p « GC o 20 o OO o 10 »e 10 o 1% OR-B 10% OR-B o o 5 10 15 20 0 5 10 15 20 generation generation 60 50 c o 1 40 a. o a. o o CC 20 o o se 10 * * A 4 * 0 o ° 0 6 ° o o o o © 9 1% OR-C 10 15 generation 20 10 15 generation 20 Figure 11: Changes in the amount of hobo element hybridizing DNA within populations. (A-C) Summary of quantitative dot blot results of the three populations initiated with 1% OR genomes. The populations are the same as those labeled as 1% OR-A and -C respectively in Figure 10. The amount of hobo homologous DNA is expressed as a percent of that present in the element donating OR strain. The circles and triangles represent results from replicate blots. (D-F) Similar results from the 10% OR initiated populations. 71 Dispersal of hobo elements: The quantitative dot blots on DNA prepared from multiple flies described above clearly demonstrate that hobo element sequences accumulated in all six OR initiated populations. I next investigated whether this accumulation within populations was associated with an increase in frequency of genomes containing hobo elements, or merely reflected a large increase in copy numbers within a few genomes each generation. The frequency of genomes containing hobo sequences was first examined by performing dot blots on DNA prepared from single flies of the 1% OR populations. Five single female flies from each of generations 2, 4, 6, 10, 14 and 18 were tested from the 1% OR-B and 1% OR-C populations. The 1% OR-A population was randomly chosen to be analyzed more extensively. For this population, 20 individuals from each of generations 1-6, 8, 10, 12, 14, 16, 18 and 20 were tested. A sample dot blot containing DNA from 10 individual female flies from selected generations of the 1% OR-A population is shown in Figure 12. Similar dot blots clearly demonstrate an increase in frequency of hobo containing flies with increasing generation of all three 1% OR populations. The change in frequency of genomes containing hobo elements was also examined by performing Southern blots on DNA prepared from single flies of both the 1% and 10% OR populations (see legend of Figure 14 and Table II for generations and numbers of individuals tested). Figure 13 shows a Southern blot containing the DNA from five individuals from each of the first four generations of the 1% OR-A population. The Southern blots confirm that there is a very rapid increase in frequency of hobo element containing flies with increasing generation of the three 1% OR initiated populations, and demonstrate a similar rapid increase for the 10% OR populations. The Southern blots also demonstrate that, among those individuals that contain hobo elements, most OR (+) gradient Figure 12: Dot blot showing an increase in frequency of flies containing hobo DNA. The left column contains DNA from single OR (+) and CS (-) control flies. From left to right (excluding the control column), the bottom row contains a gradient of 0.0, 0.03, 0.06, 0.09, 0.15, 0.20, 0.30, 0.60, 0.90, and 1.20 ug of OR strain DNA The rest of the blot contains 10 single fly DNA preps from each of the indicated generations of the 1% OR-A population. 73 P-2.6 kb 1.5 kb Figure 13: Southern blot showing an increase in flies containing both complete and deleted hobo elements. The two rightmost lanes contain Xhol digested DNA from single CS (-) and OR (+) control flies. The rest of the blot contains Xhol digested DNA from five single flies from each of generations 1-4 of the 1% OR-A population. (A) The hobo element probe reveals elements of two primary size classes: 1) the 2.6 kb band corresponding to presumably complete 3 kb hobo elements, and 2) the 1.5 kb band corresponding to the 1.9 kb defective hobo elements present in the OR-S strain. Note that all flies including the negative control also contain high molecular weight hobo homologous Xhol bands which are not canonical hobo elements. (B) Autoradiogram showing the hybridization obtained to the genomic Xhol fragment containing the complete Adh gene and used to compare the amount of DNA loaded within individual lanes. 74 have both the size classes of hobo elements that were present in the original OR strain (see Figure 13 legend). The percentage of flies containing hobo elements in the three populations initiated with 1% OR genomes, as determined both by dot blots and by Southern blots, is summarized in Figure 14. For the purpose of the Southern blot data presented in this figure, flies were counted as positive if they contained elements of either of the two main size classes of hobo elements present in the OR strain. The frequency of hobo element containing flies increased extremely rapidly within the first six generations of all three populations. Between 90% and 100% of individuals in generation six contained detectable numbers of hobo transposons. Moreover, hobo elements were present within virtually all individuals sampled from generation 8 or higher. The three populations initiated with 10% OR genomes produced very similar results though hobo elements dispersed to 100% of progeny individuals a few generations earlier (data not shown). Sizes of hobo elements present: Further analysis of the single fly Southern blots was undertaken to determine the size classes of hobo elements present within individuals from the six OR initiated populations. As mentioned above, most of the flies that contained hobo sequences had elements belonging to both the two main size classes present in the original OR strain. However, note that one of the flies from the fourth generation in the Southern blot (Figure 13) contains only the 1.5 kb band which corresponds to the 1.9 kb deleted hobo elements. The numbers of similar flies producing only the 1.5 kb hobo band, but not the 2.6 kb band coiresponding to the complete 3 kb hobo elements, are listed in Table II. For each generation of each population these numbers are expressed as fractions where the numerator 100 o A A @ A A O A A @ "§ 8 0 •c \? 60 contalnl o A A 5? 20 'o A 1%OR-A 0 5 10 15 20 generat ion 100 B @ o ® ' O * O o § 80 •C 0° O) „ c 60 c o contal • 20 - A O A 1%OR-B 0 5 10 15 20 generat ion 100 C i O A O A o Ing hobo 8 S • o ° o c CO 40 O o 5? 20 - A O o A 1%OR-C 0 5 10 15 20 75 generat ion Figure 14: Percent of flies from 1% OR populations which contain hobo elements. The populations are the same as those in Figure 11A-11C. Data from single fly Southern blots are represented by open circles. About 25 flies were sampled from tested generations of each population, except for generations 6 and 8 of the population shown in panel B, for which only 5 flies were sampled. Flies were counted as positive if they had either size class of hobo elements present in the OR strain (see Figure 13). Data from single fly dot blots are represented by open triangles. For each of the sampled generations, 20 flies were tested for the population shown in panel A and 5 flies were tested for the other two populations. LU _ J CQ H (0 •a c (0 J3 CO 3 O O) o o E o o O • c A lO >. c o c o 3 TJ O CO © o I 01 V P O CQ o O <: o O o I o O c o 2 a) . c. d) CD o CD o op a: O i n o CM CN o T — e > CM CD m o o CO LO CN t— CN O CN CM o S CO CD LO o o CN CJ) LO LO LO o o CN CN CN CNJ CN CO o LO CN LO o o CD LO CD o CD o LO o T — O O o CD o LO CN CN CN CN CN O O . T - T - o CN CM LO CD o CM LO CM O CM O CM o CJ) LO LO £M CM • t— T — CO T — o o LO LO LO LO CM O LO o CN CM CM CM CO 00 CO CO O O CN o3 CO CO CO O CM CD CD CM CJ) CO LO CO LO (0 o 77 is the actual count of flies which contained detectable numbers of only the 1.9 kb elements. The denominator is the number of flies tested each generation for which I feel the amount of DNA loaded and the intensity of hybridization was sufficient that low numbers of the full sized element would have been detected if they were present. With increasing generations there are no apparent trends towards increasing or decreasing frequency of flies producing only the 1.5 kb band. The final row of Table II shows the percent of flies tested that produced only the 1.5 kb band for each population. There are no obvious differences in the frequency of this class of flies between the 1% and 10% OR populations. Flies that had detectable numbers of the 3.0 kb elements, but not the 1.9 kb elements, were observed only twice among all flies tested by Southern blot analysis. Both of these exceptional individuals were from the 10% OR-C population. One occurred in generation three and the other occurred in generation four. Some flies which exhibited bands of both 1.5 kb and 2.6 kb sizes also exhibited new size classes of hobo elements. Table III lists individual flies which produced new sizes of elements within each sampled generation of each population. For each entry in the table, the approximate size of each new band is preceded by an asterisk and the number of flies which produced that size of band. Note that the sizes listed in the table are for bands on the Southern blot and that the elements they represent are expected to be 0.4 kb larger than the bands. At least four observations can be made from Table III: 1) New bands were detected in no more than one of the flies tested in any generation of the 10% populations, or in any generation up to 15 in the 1% populations. Since 20 to 25 flies were tested in most of these generations (see denominators in Table II), this suggests that 5% or less of the flies in these generations had new sizes of elements. In generation 20, the last generation sampled, the three 1% OR (0 •o c n> n tn 3 O cn o o E o o O m 2 CO V N o c c o 3 TJ O (0 O O a: o o i a: o o o o o cp. O O c o 5 Q) C O CM oo JD CO c i * JD JD 00 c i JD 00 CM JD CO JD OO JD CO csi co JD CD JD JD CO JD JD co O O o T — in 5 o CM O 79 populations had two to five flies which exhibited new sizes of elements. This translates to as many as 10-17% of the flies from generation 20 of these populations. 2) The four flies with 1.1 kb hobo homologous bands in generation 20 of the 1% OR-A population represent the only case where multiple flies sampled from a single population had the same size of new elements. While the sizes are only approximate, it is possible that these flies represent a new size of element which is in the process of becoming fixed within the population. 3) The single fly in generation 20 of the 1% OR-C population which contains both 1.8 kb and 2.8 kb bands is the only example of a fly which contained two or more different new sizes of hobo elements. 4) While most of the new hobo homologous bands are smaller than the 2.6 kb class that represent the complete hobo element, two flies contain elements which are about 2.8 kb in size. The smaller bands presumably represent new deleted derivatives, while the larger bands must represent a more complex rearrangement or a change resulting in the loss of one of the Xhol cut sites. A similar band which was greater than 2.6 kb has been observed in a study of long established Drosophila lines which had been transformed with hobo elements (LADEVEZE et al. 1998). For the purpose of Table III, I was very conservative in my interpretation of what constituted a new element; any ambiguous or extremely faint bands were not recorded. Results from less restrictive scoring of bands did not significantly change the results described above, though two observations are worth noting. First, less conservative scoring would increase the number of flies containing new sizes of elements by no more than about 50%. Second, some of the tentative bands in earlier generations correlated well with bands seen in generation 20 and may therefore indicate that some modified elements were becoming established in the populations. For instance, possible bands of about 1.1 kb and 1.2 kb were observed within flies of generation 15 of population 1% 80 OR-A. Perhaps most notably, a single individual from generation five of the 1% OR-C population had tentative bands of approximately 1.8 and 2.8 kb, the same sizes as those present in the single individual previously mentioned in generation 20 of this population. Changes in amount of hobo D N A in single flies; The preceding data clearly demonstrate that the total amount of hobo homologous DNA in the OR initiated populations increases with increasing generation, and that there is an associated increase in the frequency of flies containing detectable numbers of hobo elements. This led me to wonder about the dynamics of hobo accumulation within individuals as opposed to within the populations as a whole. If the initial dynamics of hobo element spread relied heavily on chromosome segregation, then the number of elements per element containing individual might be expected to decrease initially. I was therefore particularly interested in the amount of hobo homologous DNA within individuals of early generations. A visual analysis of the single fly dot blots reveals that there were noticeable differences in the intensity of hobo hybridization obtained to DNA from individual flies of varying generation. For example, in Figure 12 it appears that individuals from the fifth generation tend to have less hobo homologous DNA than flies from either earlier or later generations. Furthermore, flies of generation 15 or higher appear to have a relatively consistent amount of hobo hybridizing DNA, although this amount is clearly less than that of the OR ("+") control flies. Similar generalized statements could be made about all six OR initiated populations. That is, in the first few generations there was an apparent decrease in the amount of hobo homologous DNA per Ziofco-positive fly, but in later generations this trend is reversed. Moreover, from a visual 81 inspection of the single fly Southern blots, it appeared that the intensity of hobo hybridization to each of the two size classes of hobo elements followed the same trend in all populations. An accurate visual comparison of the individual signals on both the single fly dot blots and single fly Southern blots is complicated by the fact that there is considerable amount of variation in the amount of DNA obtained from individual flies. I therefore decided to confirm and quantify changes in the amount of hobo DNA within single flies of the 1% OR-A population by performing quantitative analysis of single fly dot blots. The amounts of DNA in the dots from individual flies were determined and the amount of hobo hybridizing DNA of each experimental fly was then calculated as a percentage of that present in the OR strain (see MATERIALS AND METHODS). These values were added for all flies of each generation and these totals were then divided by: 1) the total number of flies tested for that generation; and 2) the number of tested flies that contain hobo in that generation. Note that if all the flies of a given generation contain hobo elements, then these two values will be the same. The open triangles in Figure 15 show the first set of values which provide an estimate of the total amount of hobo homologous DNA in the population at given generations, as compared to a population made up of the element donating OR strain. There was an immediate and rapid increase in the total amount of hobo homologous DNA until a plateau or near plateau is reached at about 45-50% of that found in the OR strain. As might be expected, these results are very similar to those obtained from the multiple fly dot blots described previously (compare open triangles in Figure 15 with panel A of Figure 11). The open circles in Figure 15 show the second set of averages which provides an estimate of the amount of hobo hybridizing DNA per individual that contained hobo DNA in that generation, as compared to an OR strain individual. The average amount 82 60 50 «S 40 o 30 O O A 9 20 10 1 % OR-A 10 15 generation 20 Figure 15: Changes in amount of hobo homologous DNA within individuals. Data is for the 1% OR-A population as indicated by quantitation of single fly dot blots. The percentage of OR hobo DNA in sampled generations, expressed as a mean of all individuals tested (including negatives), is represented by open triangles. This provides an estimate of the percent of OR hobo DNA in the population as a whole. The percentage of OR hobo DNA expressed as a mean of only those flies which contain elements is represented by open circles. 83 of hobo hybridizing DNA per hobo containing individual decreased each generation up to generation six. However, by generation eight this trend was dramatically reversed and the amount of hobo homologous DNA per hobo containing fly then stabilized at about 45-50% of that present within individuals of the OR strain. From a visual inspection of the single fly Southern blots, it appeared that the intensity of hobo hybridization to each of the two main size classes of hobo elements followed the same general trends described above for the total amount of hobo hybridizing DNA per fly. To test this I measured the amounts of 1% OR-A DNA loaded per lane on the single fly Southern blots using densitometry and calculated the relative hybridization to each band as a percentage of that obtained to the OR strain controls on each blot. While the method of quantitation of the amount of DNA per fly on the Southern blots may be somewhat less precise than the measurements made by dot blot analysis (see MATERIALS AND METHODS of this chapter), the two approaches gave very similar results. Both the main size classes of hobo elements initially decreased in frequency and later increased and accumulated to approximately 50% of that found in the control strain (data not shown). Development of Regulatory Potential: The molecular analysis above clearly demonstrates that hobo elements dispersed and accumulated within all progeny individuals in the populations. I was curious as to whether or not this accumulation was associated with an increase in the regulation of hobo transposition. The ability of females from the 1% OR populations was therefore tested for the ability to repress gonadal dysgenesis using the A* cross (see MATERIALS AND METHODS). The percent dysgenic ovaries produced for tested generations of each population is 84 shown in Table IV. In all three populations, more than 30% of the ovaries were dysgenic at the fifth and eighth generations. By generation 10, the gonadal dysgenesis dropped below 30% in all populations. This decrease continued such that, by generation 20, the 1% OR-B and 1% OR-C populations produced less than 5%, and the 1% OR-A population produced about 10% dysgenic ovaries. The mean dysgenesis from the A* crosses of females from all three 1% OR populations shows a decrease from 36.4% in generation eight to 5.3% in generation ten. These results indicate that over these generations the populations are gaining the ability to supress gonadal dysgenesis and suggest that the rate of transposition of hobo elements is probably decreasing with increasing generation. TABLE IV: Percent Dysgenic Ovaries from A* Crosses g e n e r a t i o n 1 % O R - A 1 % O R - B 1 % O R - C m e a n S E M 1 G 5 34 .4% (n=160) 38 .5% (n=200) 3 2 . 0 % (n=200) 35 .0% 1.9 G 8 N.A. 33 .2% (n=196) 3 9 . 5 % (n=210) 36 .4% 3.2 G 1 0 29 .4% (n=170) 19 .1% (n=152) 2 3 . 0 % (n=200) 23 .8% 3.0 G 1 5 4 . 5 % (n=246) 8 .2% (n=208) 8 .9% (n=300) 7 .2% 1.4 G 2 0 10 .0% (n=140) 1.6% (n=124) 4 . 3 % (n=184) 5 .3% 2.5 1 SEM = standard error of the mean DISCUSSION 85 The OR initiated populations were established to model the introduction of a few hobo containing individuals into populations which initially contained no hobo elements. In this series of experiments I wanted to determine whether or not the hobo elements could spread through the populations. I also wished to observe what happened to the amount of hobo homologous DNA both within the populations as a whole, and within individuals of the populations. Dispersal and accumulation of hobo elements: Both dot blots and Southern blots performed on DNA prepared from single flies indicate that, with increasing generation, hobo elements spread very quickly to the descendants of all six OR initiated populations (summarized in Figure 14). Much of this rapid dispersal to new genomes, at least in early generations, can be explained by vertical transmission and simple Mendelian genetics. The hobo element donating OR strain used to initiate these populations has 60-75 copies of a 1.9 kb deleted element and 4-5 copies of the complete 3.0 kb element per genome ( B L A C K M A N et al. 1987). These hobo elements are located on all the major chromosome arms. To initiate the populations, males from this strain were mated to females from the CS strain. Low percentages (2% or 20%) of these OR mated females were then placed with other CS females (98% or 80%) which had been previously mated to CS males. Within the progeny of females which were mated to OR males, ignoring possible transposition for now, one member of each homologous pair of chromosomes will contain hobo elements. In the first generation these flies with 50% hobo genomes should comprise only 2% 86 or 20% of the populations, therefore I have called these the 1% and 10% OR initiated populations. The rest of the flies in the first generation will be progeny from the CS females that were mated to CS males and will contain no hobo elements, hence most of hobo containing flies will mate with others containing no elements. Recombination and segregation within these particular matings will result in generation of progeny that all contain at least some hobo elements, but on average each fly will contain about one half of the hobo elements present in its hobo containing parent (i.e. each will have about 25% hobo genomes). This second generation would also be expected to have approximately twice the number of hobo containing flies as the first generation, since all matings in which either parent contains hobo will result in all hobo containing offspring. Of course a few hobo containing flies within the first generation will actually mate with each other so this is a bit of a oversimplification. Nevertheless, these trends of an increase in frequency of flies containing elements and corresponding decrease in the number of elements per fly would continue each generation until most flies contained hobo elements, or until the number of elements within individual flies decreased to the point that not all progeny of hobo containing flies would receive elements. Based on the simple dynamics described above, and assuming that the multiple copies are highly dispersed throughout the genome, the ability of any middle repetitive element to spread to all progeny individuals depends on the initial frequency of the element containing flies and the number of elements per fly. GOOD and HlCKEY (1987) calculated the expected frequency of element containing flies and the expected copy number per fly for multiple generations of several hypothetical mixed populations, assuming random mating and these Mendelian dynamics. I have performed similar calculations to predict the 87 proportions of hobo containing flies and number of elements per fly within my 1% OR populations, that is, based on a first generation in which 2% of the flies contain 40 hobo elements each. The results are presented in Table V. Clearly recombination and segregation are sufficient to explain most of the very rapid dispersal of hobo elements to progeny genomes that is shown in Figure 14. However, the Southern blots indicate that this spread to new genomes included both the 1.9 and 3.0 kb size classes of hobo elements that are present in the OR strain. Flies containing only the 3.0 kb elements were observed only twice and flies containing detectable numbers of the 1.9 kb class alone were observed only slightly more frequently (Table II). Recombination and segregation are again sufficient to explain the occurrence of such flies containing only one size class of elements, but based on these same simple mechanisms one would expect a much greater frequency of flies containing only the deleted elements. Table VI shows the predicted frequency of flies containing complete elements and the copy number per fly using a first generation in which 2% of the flies have three complete elements each. After the first few generations of dispersal, individual flies would be expected to contain either one or no copies of the 3 kb elements. After this, no further increase in frequency of flies containing 3 kb elements would be expected. The extremely rapid dispersal of the low copy number 3.0 kb elements to all progeny flies thus clearly cannot be explained by recombination and segregation alone. There are two more results from the OR initiated populations which cannot be accounted for by recombination and segregation alone. First, quantitation of the dot blots on DNA prepared from multiple flies shows that the amounts of hobo hybridizing DNA increases rapidly in the early generations of all six OR populations (Figure 11). If dispersal was based on recombination and segregation alone, then the amount of hobo DNA within the populations as a 88 Table V. Predicted dispersal and accumulation of hobo elements in a 1% OR population based on segregation and recombination. Generation % with hobo copy # per fly % OR hobo in population GI 2.0 40 G2 3.9 20.3 1 G3 7.7 10.4 1 G4 14.9 5.4 1 G5 27.5 2.9 1 G6 47.4 1.7 1 G7 72.4 1.1 1 G8 92.4 0.9 1 Table VI. Predicted dispersal and accumulation of 3.0 kb hobo elements in a 1% OR population based on segregation and recombination. Generation % with copy # % OR hobo hobo per fly in population GI 2.0 3 1 G2 3.9 1.5 1 G3 7.7 0.8 1 89 whole would be expected to remain constant. Second, quantitation of the dot blots prepared from single flies of the 1% OR-A population show that, while there is a significant decrease in the amount of hobo DNA per hobo containing fly in the first few generations (open circles in Figure 15), this decrease is not as extreme as expected based on recombination and segregation (column 3 of Table V). Since I detected few new sizes of bands in the Southern blots on DNA from single flies and since the ratio of 2.6 kb to 1.5 kb bands appeared to be relatively stable from generation to generation, the amount of hobo hybridizing DNA within individuals should roughly reflect hobo copy number. The initial decrease in amount of hobo DNA per /iooo-containing fly in the 1% OR-A population indicates that segregation is occurring and there is no question that chromosome assortment and recombination assist in the elements' dispersal to new genomes. Nevertheless, the inescapable conclusion from the increase in copy number within individuals from later generations is that there are forces other than recombination and segregation involved in the dispersal of hobo elements within these mixed populations. Besides segregation, there are two important factors which could contribute to the ability of transposable elements to spread and to accumulate within populations of sexually outbreeding species: 1) selection acting at the cell or organismal level on either the hobo elements themselves or on associated chromosomal segments; and 2) an ability of the elements to replicate more quickly than the host genome. In most cases, and the cases of P and hobo elements in particular, transposable elements are thought to have a negative affect on host fitness (YUKUHIRO et al. 1985; MAC KAY 1986, 1989; FlTZPATRICK and SVED 1986; EANES et al. 1988; BOLSHAKOVe* al. 1994). It is nevertheless possible that hobo elements and/or the chromosomes containing them may confer an increase in 90 fitness on their host, especially in an evolutionary time frame (for review see MCDONALD 1995). For instance it has been suggested that transposable elements may contribute to the ability of organisms to adapt under stress conditions such as pesticides (BREGLIANO and KlDWELL 1983; VERESHCHAGINA 1994). Alternatively, the clustering of hobo and many other transposable elements within heterochromatin could allow the elements to progressively acquire a functional role in the host genome (PlMPINELLI et al. 1995). In any case, even if hobo does increase long term host fitness via these or other mechanisms, it is unlikely that this increase in fitness could be sufficient to explain the very rapid increase in hobo elements within our populations. As stated above, selection acting on the chromosomal segments that contain hobo elements could also lead to an increase in element copy number. The CS and OR strains used to initiate these mixed populations are both highly inbred. When two such strains are crossed, one would expect heterosis to occur. That is, as recombination snips and shuffles the combinations of alleles in the original chromosomes, selection will tend to favour heterozygotes in many chromosomal regions. In addition, a certain number of loci and their accompanying elements would undergo directional selection. However, the number of loci that experience positive and negative selection should be approximately equal. Moreover, a massive introgression of OR chromosomes would be required to explain the approximately 20 fold increase in copy number in the 1% OR populations within eight generations (Figure 11). Thus, while transposons "hitchhiking" on chromosome segments that were selected by heterosis and/or by directional selection could increase hobo copy numbers within these populations, it seems unlikely that either is the primary force. I therefore believe the spread and accumulation of hobo in our populations is largely due to an ability of the hobo elements to multiply independently of host 91 genes. This is the most parsimonious explanation given the rapid spread and increase in copy number within ten generations In sexual species, this ability of the transposable elements to multiply within a single genome can lead to their dispersal to many new genomes (HICKEY 1982, 1992; NANJUNDIAH 1985). In essence, some form of multiplicative transposition leads to the maintenance of multiple copies of the element on all or most of the chromosomes, and meiosis and conjugation ensure that all the progeny get elements if either of their parents had elements. Such multiplicative transposition can be considered as selection acting at the molecular or transposable element level (HICKEY 1992; PLASTERK1993). When there is positive selection of the element at the molecular level which outweighs negative selection at the organismic level, then the element will be 'driven' into the genome and it can be considered to be a molecular parasite (ORGEL and CRICK 1980; HICKEY 1982). Numerous studies have produced results consistent with the model of transposable elements as molecular parasites in eukaryotes. Some of these studies investigated the potential balance between multiplicative transposition and negative selection on the host (LANGLEY et al. 1983; MONTGOMERY and LANGLEY 1983; CHARLESWORTH and CHARLESWORTH 1983; CHARLESWORTH and LANGLEY 1986). Other studies demonstrated that P elements, despite their negative impact on host fitness, can rapidly spread through natural populations (KlDWELL 1983), and are capable of rapid dispersal in experimental populations containing genomes with and without P elements (KlDWELLef al. 1981; KlYASU and KlDWELL 1984; GOOD et al. 1989). This dispersal occurred even in the presence of strong dysgenic sterility selection (KlYASU and KlDWELL 1984). In such mixed populations containing low frequencies of element-containing flies, the ability of the elements to multiply must be sufficient to counteract not only 92 any negative selection against element containing flies, but also the dilution effect of repeatedly crossing most individuals from each generation to flies which contain no elements. My present results show that hobo elements, like P elements, can rapidly disperse in mixed populations and suggest that they may also be considered as genomic parasites. While this does not rule out the possible contribution of other forces, it does suggest that some form of multiplicative transposition is the dominant force in the ability of hobo elements to spread and to accumulate. This importance of transposition in copy number accumulation is supported by the observation that the large decrease in the rate at which hobo elements accumulated in the three 1% OR populations after generation 8 (Figure 16 panels A-C) was associated with an increase in the flies' ability to repress gonadal dysgensis (Table IV). That is, as the flies gained the ability to repress the high levels of transposition assumed to be responsible for hybrid dysgenesis, the rate of accumulation of hobo in the populations was drastically reduced. This of course raises two questions: 1) what is the mechanism of transposition of the hobo element; and 2) how does this mechanism ensure an increase in copy number of the element as compared to genomic DNA. There is unfortunately little direct evidence relating to the mechanism of transposition of hobo. Based on some sequence similarity within the transposase genes (CALVI et al. 1991; FELDMAR and KUNZE 1991) and on the type of footprints left behind after excision of the elements (ATKINSON et al. 1993), it was suggested that hobo excises and transposes like the Ac element of Z. mays (corn) and the Tam3 element of A. majus (snapdragon). Essentially the excision footprints consist of short deletions within the target site which was duplicated upon insertion of the element as well as addition of extra nucleotides which appeared to more or less 93 correspond to inverted duplications of sequences originally flanking the hobo element. At the time of the discovery of the footprints left by hobo, there were two competing models proposed for transposition of the plant transposable elements (for a review see ROMMENS et al. 1993). The first step in both proposed mechanisms was conservative "cut-and-paste" transposition of the element to leave a staggered double stranded gap at the donor site. It is the proposed mechanism through which this gap is repaired that distinguishes between the two models. The first model (SAEDLER and NEVERS 1985) proposed that the gap was filled through a combination of exonuclease and polymerase activities. The single stranded DNA on both the chromosome and the excised element would provide the polymerase with two potential templates. Exonuclease activity could explain the observed deletions within the target site duplications and polymerase switching between the two templates would explain the inversion of one or more base pairs at the junction(s). The second model for Ac transposition (GOEN et al. 1989) proposed that the gap was repaired by ligation of the free ends at the excision breakpoints to form two hairpin structures at the chromosome ends. These hairpins could then be resolved by further nicking and ligation. Subsequent replication would result in reciprocal products with extra inverted repeat structures. Both of the transposition models described above for the plant transposable elements are conservative. However the timing of conservative transposition could result in a multiplication of the elements. Both genetic and molecular analyses indicate that Ac excises during or shortly after DNA replication and then inserts into either replicated or unreplicated target sites (GREENBLATT and BRINK 1962; GREENBLATT 1974; G R E E N B L A T T 1984; CHEN et al. 1987,1992). This would result in one of the daughter cells having an element 94 at the original donor site, and also one or both of the daughter cells having an element at the target site. Thus, some cells would maintain their copy number, while an increase in copy number would be assured in others. But how is Ac transposition coupled to DNA replication? In many organisms, including Z. mays, DNA sites which are methylated in non-dividing cells are temporarily hemimethylated after passage of the replication fork. Such hemimethylation of the Ac transposase binding motif has been shown to increase transposase binding in vitro (KUNZE and STARLINGER 1989) and some putative transposase binding sites within Ac are methylated (WANG et al. 1996). The Ac transposon, like the bacterial element TnlO (ROBERTS et al. 1985), may thus be preferentially donated from newly replicated DNA through recognition of the state of DNA methylation. The problem with applying this mechanism of multiplicative transposition to the hobo element is that DNA is not methylated in Drosophila. One would thus have to propose either an alternate method for hobo to preferentially mobilize from newly replicated DNA, or a completely different method for hobo to increase in copy number. An interesting alternative is that, at least in some instances, the repair of the gap left after hobo, Tam3 and Ac element excision may take place using the SDSA repair mechanism proposed for P elements (see Transposition of P elements in CHAPTER I). This was originally suggested for the Ac element based on genetic evidence (CHEN et al. 1992). As pointed out by ROMMENS et al. 1993, this possibility is supported by observations that excision of Ac and Tam3 at least occasionally leave behind sequences other than the normal transposon footprints. These sequences include precise excisions, deleted derivatives of the elements, and insertion of an element previously found at a different location in the genome. The first two of these excision products have also been found for hobo elements (BLACKMAN and GELBART 1989), and all three are more easily 95 explained by the SDSA gap repair model proposed for P elements than for the repair mechanisms proposed for the plant transposons. Indeed, one of the primary strengths of the P element model is that it explains many of the structures left behind after P element transposition (ENGELS et al. 1990; GLOOR et al. 1991; NASSIF et al. 1994; KEELER and GLOOR 1997). Precise excision of the element sequences would occur when exonuclease digests both the single stranded ends at the excision sight back into the host derived flanking sequences. The gap could then be repaired by invading a homolog which does not contain a transposon at this site. On the other hand, replacement of the element with an element from another location, or "internal conversion", would occur when the transposon ends are retained at the excision site and invade the ectopic copy of the element during repair. Finally, elements with internal deletions would result when repair from another element template is cut short before the entire template is copied. As observed for P elements (O'HARE and RUBIN 1983; ENGELS 1989) and also for Ac elements (RUBIN and LEVY 1997), the internal deletions often occur at the sites of short direct duplications, presumably because the duplications provide a region of homology at which the two invading strands can anneal during the SDSA repair process. The production of some of these post-excision structures for Ac, Tam3 and hobo thus strongly suggest that at least occasionally the gap is repaired using the P element strand invasion mechanism. The reverse may also be true. It has often been observed that P element excision sites sometimes retain extra sequences, which are similar to, but generally more complex than Ac and Tam3 footprints. This may suggest that P elements are also capable of transposition and excision repair using mechanisms similar to those described above for the plant transposons (TOKASU-ISHIKAWA et al. 1992; B E A L L and RIO 1996), although these excision products might also be the result of DNA replication and 96 template slippage during the process of SDSA repair (O'HARE and RUBIN 1983; STAVELY et al. 1995). The double strand gaps left after excision of Ac, Tam3, hobo, P and perhaps other inverted repeat type elements may thus all be capable of repair via several different mechanisms. As suggested by ROMMENSe* al. (1993), the preferred mechanism of repair may be determined by the host. However, within D. melanogaster there are apparent differences in preference of repair mechanisms shown for hobo and P excision sites. For instance, precise excisions appear more common for P elements and element footprints appear more common for hobo. In addition, while most element containing fly strains contain many copies of either P or hobo deleted elements, there is one significant difference. Flies frequently contain many different sizes of P element deleted derivatives (O'HARE and RUBIN 1983), whereas they contain accumulations of only a few different sizes of hobo deleted derivatives (STRECK et al. 1986; BLACKMAN et al. 1987). These observations all suggest that SDSA repair may be more common for repair of P element gaps, but that the plant mechanism(s) are more common for repair of hobo element gaps. These differences for two elements within one species suggest that the choice of repair mechanism may at least in part be dependent on the particular transposable element involved. Such a preference of repair based on the transposable element could perhaps be the result of the type of gap left after the element excises and/or the result of the involvement of different host proteins. A combination of a large staggered cut during excision followed by the conscription of host proteins such as IRBP to protect the single stranded element ends, as has been shown for P (STAVELEY et al. 1995; KEELER and GLOOR 1997), may ensure that some elements are primarily repaired via the strand invasion mechanism. On the other hand, gaps left by the excision of other elements which leave short single 97 stranded ends or blunt ends may be preferentially repaired using one of the mechanisms originally proposed for the plant transposons. Similar gaps could also be created by elements which originally leave larger overhangs which are not protected from nuclease digestion by the binding of host proteins to the elements' termini. It has recently been shown that, while the Ac transposase protein binds primarily to internal Ac element sequences, it also binds weakly to the ends of the inverted repeats (BECKER and KUNZE 1997). Since Ac and hobo exhibit homology within both the D N A sequences of their terminal inverted repeats (STRECKei al. 1986), and within the protein coding regions known to be essential for transposition and D N A binding activity (CALVI et al. 1991; WARREN et al. 1994; BECKER and KUNZE 1997), it seems likely that hobo transposase also binds to the hobo element termini. This suggests that transposase induced excision of these elements could leave ends which are unprotected by host proteins. Again, this may predispose the gaps to being repaired by one of the mechanisms proposed for plant transposons. In any case, if the gap left after excision of hobo is, at least occasionally, repaired using the S D S A repair model, then the mechanisms proposed for rapidly increasing P element copy number by preferentially copying from an element-containing template (see Multiplication of P elements in DISCUSSION of CHAPTER II), may apply to hobo as well. Regulation of hobo copy number: More work will obviously be required to determine the mechanism(s) of hobo transposition. However, my thorough molecular analysis of the change in amount of element-hybridizing D N A , both within the O R initiated populations as a whole and within individuals of the 1% O R population, indicate that it must be a highly multiplicative process. There is a second important implication of 98 these quantitative data for the 1% OR-A population, namely, they show that the amount of hobo homologous DNA per fly does not reach the amount present in the element donating flies strain nor the copy numbers typically seen in natural populations (Figure 14). This is undoubtedly also true for individuals from the rest of the OR initiated populations which, according to the multiple fly dot blots, all plateaued or nearly plateaued at about 45-50% of the amount of hobo DNA present in the hobo element donating strain (Figure 11). A visual analysis of the Southern blots containing DNA of single flies from all six populations also supports the conclusion that all populations behaved very similarly in terms of amount of hobo hybridizing DNA within individuals. The rapid dispersal of hobo elements to all progeny genomes and drastic increase in element copy number closely mirrors that previously seen for P elements (GOOD et al. 1989; MEISTER 1992). However, unlike hobo elements, within 20 generations P elements did reach the copy numbers present in the element donating strains and typically seen in natural populations. It is of course possible that hobo elements take longer to accumulate than P elements and that they will eventually reach the copy numbers typical of natural populations. This is supported by the observation that the amount of hobo homologous DNA in at least several of the OR populations did not appear to plateau completely (Figure 11). Indeed, there is evidence that hobo copy numbers may still be slowly increasing even within natural populations (BONNTVARD et al. 1997). In any case, the lower accumulation of hobo elements, at least within a similar number of generations, suggests that the mechanism of copy number regulation may differ between these two Drosophila transposons. Based on theoretical and empirical population genetic studies, three primary models have been proposed to account for the maintenance of transposon copy number (for a recent review see CHARLESWORTH et al. 1994). 99 All three models assume that copy number will be determined by a balance between those forces which increase copy number and those forces which reduce copy number. The primary force for increasing copy number is some form of multiplicative transposition. Two of the models propose that selection at the host level acts to reduce copy number. The first model suggests that this selection could be the result of deleterious insertional mutations (CHARLESWORTH and CHARLESWORTH 1983; MONTGOMERY et al. 1987; LANGLEY et al. 1988; CHARLESWORTH and LANGLEY 1989; HOOGLAND and BlEMONT 1996), while the second model suggests that selection could be due to deleterious chromosome rearrangements caused by ectopic recombination between two elements located at different chromosomal sites (MONTGOMERY et al. 1987; LANGLEY et al. 1988; CHARLESWORTH and LANGLEY 1989). The third model suggests that the balancing force is a reduction of the transposition rate due to some form of negative regulation which is somehow induced by an increase in copy number (CHARLESWORTH and CHARLESWORTH 1983; CHARLESWORTH and LANGLEY 1986). It is of course possible that the copy numbers of different transposon families are maintained by different mechanisms or by different combinations of mechanisms. Both of the selection models of copy number control make specific predictions regarding the distribution of elements within genomes. Since many insertional mutations would be expected to be recessive, detrimental insertions would cause more of a selective disadvantage to the host when present on the X chromosome where they would be hemizygous in males. If insertional mutations play a major role in the restriction of element copy number, then one would predict that a population at equilibrium would have fewer insertions on the X chromosome than on the autosomes (MONTGOMERY et al. 1987). If selection due to ectopic recombination plays a significant role in copy number regulation, then 1 0 0 elements would be less detrimental in areas of the genome where crossing over is reduced ( L A N G L E Y et al. 1988). The ectopic recombination model therefore predicts an increase in element insertions within areas with decreased recombination, such as the bases or tips of chromosomes. Several studies have investigated the distribution of hobo elements within genomes. In general there appears to be no decrease in hobo insertions on the X chromosome (BIEMONT et al. 1988; ZABALOU et al. 1994; LADEVEZE et al. 1994), although YANNOPOULOS et al. (1994a) did report a significantly lower frequency of insertions on the X chromosome for one of the five populations in their study. There also appears to be an decrease rather than a increase in hobo element site number at the bases and tips of the chromosomes (ZABALOU et al. 1994; YANNOPOULOS et al. 1994a, 1994b), especially if insertion site number is weighted by DNA content (HOOGLAND and BIEMONT 1996). Thus, while the limited current literature on the genomic distribution of hobo elements does not strongly support a major role for either of the two selection models of copy number control, either or both could still be involved. Any role for selection in hobo copy number control might also predict a higher copy number to be tolerated in my experimental populations versus natural populations, since laboratory populations might be expected to lack much of the selective pressure present in nature. My results showing a lower copy number in the experimental populations therefore might also argue against a major role for selection in copy number control. Regulation of transposition, and hence multiplication, occurs with most transposable elements and may limit deleterious affects on the host (KLECKNER 1990; SMITH and CORCES 1991). In order for regulation of transposition to play a significant role in control of copy number, increased repression of transposition must accompany an increase in copy number (CHARLESWORTH and 101 CHARLESWORTH 1983). Transposition may be regulated by host cellular functions and/or may be autoregulated by element-specific mechanisms. Unfortunately, while regulation of hobo transposition almost certainly plays at least some role in copy number control of hobo, little is known about hobo regulation. Just as with P element transposition (see Regulation ofP Element Transposition in CHAPTER I), there is evidence to suggest at least three types of regulation of hobo transposition: 1) germ line specificity; 2) zygotic repression; and 3) maternal repression. The best understood of these forms of regulation of the hobo element is the restriction of transposition to the germline which occurs by the mechanism of transcriptional control (CALVI and GELBART 1994). Since transposition in somatic cells would not result in an increase in copy number in subsequent generations, the inhibition of hobo transposition in somatic cells should in no way reduce the ability of the element to accumulate. In fact, somatic inhibition undoubtedly reduces the genetic load caused by the element accumulation. The zygotic repression of hobo is in some way mediated by the elements themselves (HO et al. 1993; YANNOPOULOS et al. 1994b). This form of regulation is therefore likely to increase with increasing copy number and could play a significant role in the maintenance of copy number. For instance, zygotic repression could increase with copy number if the eventual insertion of elements at specific locations in the genome was required to induce repression. It has been suggested that the hobo elements clustered within heterochromatin may be involved in regulating hobo activity (KARPEN and SPRADLING 1992; PIMPINELLI et al. 1995). Alternatively, it has often been suggested that as the copy number of the element increases there is an increased likelihood of creating certain deleted derivatives that could be involved in regulation of the elements (see 102 below). There is currently little evidence to support either of these models for zygotic regulation of hobo transposition. For the P element system, regulation of transposition by deleted elements is one of several mechanisms supported. Specifically, the KP deleted derivative present in high copy number within many populations of Drosophila seems to have a role in repression (BLACK et al. 1987; JACKSON et al. 1988; RASMUSSON et al. 1993). A similar role in repression has been suggested for the Thl, Th2 and Oh deleted hobo elements which are ubiquitous or nearly ubiquitous in Drosophila strains collected since the mid-1960s (PERIQUET et al. 1989a, 1989b, 1990, 1994; BOUSSY and DANIELS 1991; BAZIN et al. 1993). These three elements are 1.51 kb, 1.49 kb, and 1.87 kb respectively. In this respect it could be significant that the OR strain we utilized to introduce hobo elements into our populations contained many copies of a 1.9 kb deleted element. Since Southern blots indicated that this deleted element quickly spread throughout the populations, it is possible that the presence of this element resulted in maintenance of a low copy number. However, if the presence of this element alone is sufficient to inhibit hobo transposition, then it is hard to understand how the hobo elements could have multiplicatively transposed enough to allow their dispersal throughout the populations. The Southern blots performed on DNA from single flies from all six OR initiated populations show the production of only a few new bands with increased generation (see Table III). The three deleted hobo elements with putative regulatory abilities described above would be expected to form Xhol fragments of 1.11 kb, 1.09 kb, and 1.47 kb. Thus, some of the bands of approximately 1.1 kb seen in generation 20 could correspond to these elements. However, it is unlikely that these new deleted elements could be responsible for restriction of copy number accumulation since they were detectable in no more 103 than a few individuals in any of the populations and since the hybridization intensity indicated that copy number remained very low. I cannot rule out the possibility of the accumulation of a modified repressor element which co-migrates on the Southern blots with either the 1.5 kb or 2.6 kb bands. Maternal repression of the hobo system, while present within many strains that contain hobo elements, is also present within some strains that contain no hobo themselves (YANNOPOULOS et al. 1987; STAMATIS et al. 1989; HO et al. 1993; SHEEN et al. 1993). Maternal repression must therefore be established by genomic factors distinct from the elements themselves. It is thus possible that the low accumulation of hobo in my experiments was due to genomic factors present in the particular CS or OR strains used to initiate the populations. I have at least partially ruled out this possibility for the OR strain by studying two completely separate mixed populations in which TT2 was used as the hobo donating strain instead of OR (see CHAPTER TV). The possibility of a repression inherent within the CS strain is especially notable in light of the observation that there are differences in repression potential of various E strains (YANNOPOULOS et al. 1987, 1994b; STAMATIS et al. 1989; HO et al. 1993; GALINDO et al. 1995; BAZIN and HlGUET 1996; BONNIVARD et al. 1997), and therefore warrants further investigation. The A* test crosses, which detect both zygotic and maternal repression, indicate that the average ability of females from the OR initiated populations to repress gonadal dysgenesis increased substantially after generation 8 (Table IV). The data also indicate that dysgenic ovaries were produced by the daughters of fewer of the experimental females in later generations (not shown). This suggests that not only was the average rate of transposition lower, but also that more of the individual flies had acquired the ability to repress hobo transposition. The period over which this repression potential developed roughly 104 corresponds to the generations during which rate of increase in hobo homologous DNA slowed dramatically. I therefore surmise that the observed repression of hobo transposition is largely responsible for hobo copy number control. This suggestion is supported by the observation that the 1% OR-A population, which shows the highest level of dysgenic ovaries in the A* test at generation 20 (Table IV), is also the population in which the amount of hobo DNA most appears be still increasing during the later generations (Figure 11). It is generally accepted that regulation of P element transposition is the primary balancing force for maintaining P element copy number. There is fairly strong evidence that P elements are not distributed within the genome as would be predicted if either selection against element insertions or element mediated rearrangements was the predominant force (BlEMONT 1992; YANNOPOULOS et al. 1994; BlEMONT et al. 1994; ZABALOU et al. 1994; HOOGLAND and BlEMONT 1996). Moreover, there is considerable evidence that after P elements invade a population there is short lag time, but then they quickly produce the self-inhibiting P cytotype and repress their own transposition (see Regulation of P element transposition in Chapter I). This development of P cytotype occurs regardless of whether the elements are introduced by the establishment of mixed populations (GOOD et al. 1989; MEISTER 1992) or by injection of embryos to form transformed lines (DANIELS et al. 1987). This suggests that P elements regulate their own copy number by increasing repression of transposition after a certain copy number is reached or specific conditions are attained. If repression of transposition is the primary force that counteracts multiplicative transposition and results in the regulation of both hobo and P element copy numbers, then differences in the dynamics of hobo and P element repression may explain the observed differences in copy number accumulation of P and hobo elements within mixed populations. With P elements, there is a lag 105 time but then they quickly develop P cytotype which results in complete repression of P element transposition and a stabilization of copy number. On the other hand, with hobo elements, the repression potential develops over a much longer period and is seldom complete. Copy number for hobo may therefore continue to increase, albeit at a continually decreasing rate. Consistent with this explanation are the observed continued development of hobo repression potential in natural populations over many years (BONNIVARD et al. 1997) and the variability of hobo repression potentials among established laboratory strains (HATZOPOULOUS et al. 1987; YANNOPOULOS et al. 1987; LlM 1988; BLACKMAN and GELBART 1989; HARADA et al. 1990; PASCUAL and PERIQUET 1991). Summary: The OR initiated populations presented in this chapter clearly demonstrate that hobo elements can spread rapidly and accumulate within mixed experimental populations of D. melanogaster established with low frequencies of hobo containing flies. This supports the role of hobo as a molecular parasite which has recently invaded natural populations of D. melanogaster and rapidly spread worldwide driven by some form of multiplicative transposition. These results suggest that the hobo element may be a valuable vector for the spread of engineered DNA in populations of some of the ever increasing number of species for which hobo transposition has been demonstrated. Further studies will be required to determine how hobo transposition results in multiplication of hobo sequences and to confirm why the accumulation of hobo DNA in these experimental populations appears to plateau, or at least to slow substantially, at about 50% of the amount present in the element donating strain (see also CHAPTER TV). CHAPTER IV: CONCURRENT SPREAD OF P AND HOBO ELEMENTS WITHIN MIXED EXPERIMENTAL POPULATIONS OF DROSOPHILA MELANOGASTER INTRODUCTION 107 I have previously monitored the dispersal and accumulation of P elements in several mixed populations including two CS populations into which 5% %2 strain genomes had been introduced (GOOD et al. 1989; MEISTER 1992). Since the 7t2 strain contains both P and hobo elements, I was curious as to whether or not the spread of P elements would have had a significant affect on the ability of hobo elements to spread within these populations. This may be relevant to natural populations since both P and hobo elements are thought to have invaded D. melanogaster within roughly the same period. If the hobo elements did disperse within these earlier experimental populations, I was also curious as to how their accumulation would compare to that seen in the OR initiated populations described in the previous chapter. I have therefore analyzed flies from these populations by both multiple and single fly dot blots in order to more thoroughly examine P element accumulation, and to see if hobo elements had also accumulated. MATERIALS AND METHODS The two populations I refer to in this thesis as the n2 initiated populations have been described previously and were originally used to monitor the dispersal and accumulation of P elements (GOOD et al. 1989; MEISTER 1992). These populations were set up and maintained in essentially the same manner as the OR initiated populations described in CHAPTER III except that: 1) the initiation frequency was 5% element containing genomes, and 2) the transposon providing males used to introduce the elements into the CS females were from 108 the %2 strain which contains both P and hobo elements (BINGHAM et al. 1982; O'HARE and RUBIN 1983; BLACKMAN et al. 1987; YANNOPOULOS et al. 1987). All DNA preparations, dot blotting to membranes, and hybridizations were carried out using the methods and conditions specified in CHAPTER III. Membranes were hybridized to a single copy probe, then stripped and hybridized to a P element probe, and finally stripped a second time and hybridized to a hobo element probe. The DNA fragments used as single copy and hobo element probes are the same as those described in the preceding chapter. For the P element, the fragments used were the 0.84 kb .HmdIII and the 1.53 kb HindllVSall fragments from p7r25.1 (O'HARE and RUBIN 1983). These fragments comprise approximately 80% of the hobo element and 82% of the P element and should therefore hybridize to the majority of both complete and internally deleted elements. Quantitative analysis was performed as described in the preceding chapter, except the quantity of both hobo and P element DNA was calculated as a percentage of the amount present in the K2 element donating strain. RESULTS Presence of P and hobo DNA within individuals: In order to examine the presence of both P and hobo DNA within individuals, dot blots were performed on DNA prepared from single females from selected generations of each of the two %2 initiated populations. Figure 16 shows a sample single fly dot blot from one of the %2 initiated populations. Panels A and B show the hybridization obtained with a P element probe and with a hobo element probe respectively. At generation six, numerous flies lacked 109 Panel A: Pelement probe controls G6 # . G8 # ' G11 G15 • G19 Panel B: hobo element probe + + '"•''wmfivK'' wrnmmmmmmm gradient controls G6 G8 G11 G15 G19 gradient Figure 16: Dot blot showing presence of both P and hobo element DNA within individuals. The top row contains DNA from single n2 (+) and CS (-) control flies. In the bottom row the dots from left to right contain 0.03, 0.06, 0.09, 0.15, 0.23, 0.30, 0.60, and 0.90 pg of K2 strain DNA. The rest of the blot contains DNA from single flies of the indicated generations of the 5% 7i2-B population. 110 detectable amounts of P and/or hobo DNA. However, both P and hobo elements were present within virtually all individuals tested from generation eight or higher. This confirms that P elements had dispersed within these populations (as reported by GOOD et al. 1989), and demonstrates that hobo elements had also dispersed. From the intensity of the hybridization obtained to the DNA from individuals of the various generations, it appears that many of the individual flies of later generations contained approximately the same amount of P element hybridizing DNA as the K2 controls. On the other hand, the amount of hobo element hybridizing DNA in flies of later generations clearly did not approach the amount present in the 7t2 strain. Concurrent accumulation of P and hobo elements; The potential accumulation of both P and hobo elements in the n2 initiated populations was analyzed by performing dot blots on samples of DNA prepared from multiple flies of selected generations. Phosphor images of a representative dot blot are shown in Figure 17. Panel A shows the hybridization obtained to a single copy gene probe and demonstrates that about 2.5 pg of each sample DNA was fixed to the membrane. The hybridization of this dot blot to a P element probe (panel B) shows that there was an increase in P homologous DNA between generations 6 and 17 of both populations. The 2.5 pg samples of DNA prepared from flies of generation 17 and higher appear to have about the same amount of P element hybridizing DNA as 2.0-2.5 pg of DNA from the element donating n2 strain. In contrast, there appears to have been little if any increase in the amount of hobo homologous DNA between generations 6 and 17 in either of the TC2 populations (panel C). Furthermore, even in the late I l l Panel A: single copy probe Tt-2-B 6 8 1 1 13 15 17 19 20 7t2-A c s i i f * i t f f f 0.2 0.4 0.6 0.8 1.0 1J> 2^0 2.5 3JD 3 J Panel B: Pe lement probe 7t2-B 6 8 11 13 15 17 19 2 0 7C2-A CS 0.2 0.4 0.6 0.8 1.0 1.5 2^0 2 ^ 3^> 3^5 Panel C: hobo e lement probe 7i2-B O # 6 8 11 13 15 17 19 20 CS 0.2 0.4 0.6 0.8 1.0 1^5 2J5 3.0 Figure 17: Dot blot showing presence of P and hobo DNA within populations. (A) Autoradiogram of a dot blot filter hybridized to a single copy probe to allow normalization for the amount of DNA per dot. The top two rows contain approximately 2.5 ng of DNA prepared from multiple flies of the given generations of the two 5% %2 populations. The bottom two rows contain gradients with increasing amounts of CS and 7i2 DNA (values indicate Lig). (B-C) The same filter hybridized to P and hobo element probes respectively. 1 1 2 generations, the experimental DNA samples never appear to contain more hobo homologous DNA than about 0.8-1.0 pg of 7c2 strain DNA. The amount of P and hobo element DNA in the replicate dot blots prepared from DNA of multiple flies from the 5% n2 initiated populations were quantitated by volume analysis. The results, summarized in Figure 18, reinforce the visually observed trends in amount of P and hobo element DNA. Panels A and B show the amount of P element homologous DNA in the two populations as a percent of that present in the iz2 strain. At generation six, both populations contained about 40% of 7t2 strain P element DNA. The amount of P DNA then increased relatively constantly such that by generation 20 both populations had about the same amount as the original K2 strain. These data confirm the results reported by GOOD et al. (1989) for the accumulation of P elements in these populations. Panels C and D show the amount of hobo homologous DNA in the two populations as a percent of that present in the n2 strain. As with the P element DNA, the amount of hobo DNA in each of the populations at generation six was about 40% of that found in the n2 strain. However, in contrast with the amount of P DNA, there is no further increase in the amount of hobo hybridizing DNA in successive generations of either population. 113 120 120 100 < z Q 80 c I o a 60 Q. ? 40 o 20 B 1 0 1 5 g e n e r a t i o n 2 0 o O o o o 5% 7i2-B 1 0 1 5 g e n e r a t i o n 2 0 100 < o 1 8* 80 60 40 20 c 5% 7i2-A • A A ® « o ° 120 ioo h o 1 3« 1 0 1 5 g e n e r a t i o n 2 0 1 0 1 5 g e n e r a t i o n 2 0 Figure 18: Changes in the amount of P and hobo DNA within populations. (A-B) Summary of P element DNA accumulation in the 5% TC2 populations. The circles and triangles represent data from the quantitation of replicate dot blots. (C-D) The hobo element DNA accumulation in the same populations. In each case, the amount of element homologous DNA is expressed as a percent of that present in the element donating iz2 strain. DISCUSSION 114 Dot blots of DNA prepared from single flies showed that by generation eight P elements had spread to virtually all individuals of both populations in which 7t2 acted as the element donating strain (Figure 16A). After 19 to 20 generations, many of the individual flies appeared to have about the same amount of P element hybridizing DNA as individuals of the TT.2 strain. Moreover, quantitative analysis of the dot blots containing DNA extracted from multiple flies show that, within the populations as a whole, P element DNA accumulated to as much or more than that present within the %2 strain (Figure 18). These results confirmed and extended earlier results that demonstratied the dispersal and accumulation of P elements in these populations (GOOD et al. 1989; M E I S T E R 1992). Re-probing of the single fly dot blots with hobo element fragments revealed that some flies did not contain a detectable amount of hobo homologous DNA at generation six, however, virtually all flies from generation eight or later contained hobo elements (Figure 16B). The dispersal of hobo elements thus took a few generations longer to reach all progeny individuals within the 5% 7i2 initiated populations than it did in the 1% OR initiated populations described in Chapter III. This slight difference could be the result of experimental variability, but it could also be due to the simultaneous spread of P elements in these populations or to some other differences between the two strains. Nevertheless, the ability of hobo elements to disperse clearly is not overly restricted by the concurrent dispersal of P elements. With increasing generation, there appeared to be less variability in the amount of hobo homologous DNA within individuals of the 7i2 initiated populations (Figure 16B). It is also clear that the amount of hobo hybridization 115 to DNA from individual flies from late generations never reached the amount present in the element donating TT2 strain. Quantitation of the multiple fly dot blots confirmed that the amount of hobo element homologous DNA in the n2 initiated populations was fairly constant from generation six onward and that each population has attained about 40-50% of the amount hobo DNA present in the element donating strain (Figure 18). This low accumulation of hobo is in contrast to the high accumulation of P elements within these populations, and closely mirrors the hobo accumulation in the OR initiated populations. Though the final accumulation of hobo elements as a percentage of that present in the hobo donating strain may be slightly lower in the n2 initiated populations (40% versus 50%), the simultaneous accumulation of P elements appears to have had little or no affect on the accumulation of hobo elements. As summarized in the DISCUSSION of CHAPTER III, the regulation of TE copy number is probably due to a balance between multiplicative transposition and either some sort of selection or an increase in the repression of element transposition as copy number itself increases. Both P and hobo elements are expected to create a selective disadvantage to their host. However, the current literature on transposable element distribution within genomes does not support a major role for selection in copy number control of either hobo elements or P elements (see DISCUSSION of CHAPTER III). My results showing similar spread and accumulation of hobo with or without the concurrent dispersal of P elements also suggest that selection is not a primary force in copy number control or that the spread of two elements is no more disadvantageous than spread of either element alone. That is, the elements' ability to increase in copy number by some sort of multiplicative transposition mechanism(s) must vastly outweigh any selective disadvantage caused to individual flies. 116 As I stated in the previous chapter, I therefore believe that the dominant force in the regulation of both hobo and P element copy number is the regulation of transposition. In the case of P elements, the current model for this regulation is that the P element transcript can be alternately spliced to produce either transposase or repressor (see Regulation of P element transposition in CHAPTER I). This alternate splicing is regulated by a host protein, so both the host and the P elements contribute to the regulation of transposition. The regulation of hobo element transposition is less well characterized. I suggested in CHAPTER III that the copy number regulation of hobo elements in the O R / C S mixed populations was due to the expression of a regulatory potential present in the genomes of individuals from either the C S or the O R strains. The similar results from the TC2/CS populations argue that this repression potential must be present in the C S strain, or that whatever potential is present in the O R strain is also present in n2 strain. Moreover, since the hobo element accumulation was very similar with or without the simultaneous accumulation of P elements, the two elements must spread and regulate copy number by independent mechanisms or else any host proteins involved in both mechanisms must not be in limiting supply. For example, the host DNA repair systems must not be overly taxed by the element mobilizations or TE excision and repair occurs at different times in the cell cycle for the two different elements. The simultaneous spread and accumulation of both P and hobo elements in the JC2 initiated populations support the notion that these elements could have spread rapidly through natural populations of D. melanogaster during approximately the same time period. These results also suggest that if the hobo element were to be utilized as a vector for the dispersal of engineered DNA, then the ability of elements to disperse would not be overly affected if a second TE was also actively mobile within the population. 117 CHAPTER V: A SIMPLE PLAQUE HYBRIDIZATION METHOD FOR THE DETECTION OF DIFFERENTIALLY REPRESENTED REPETITIVE DNA INTRODUCTION 118 Some transposable element (TE) families such as P elements exhibit a very limited species distribution, presumably due to the requirement for specific host encoded proteins in the transposition process (reviewed in CHAPTER I). Other TE families such as mariner, Tel, and to some extent hobo, are present in diverse species and apparently require either highly conserved or no host encoded proteins. It is generally considered that species-specificity is a liability because it makes it difficult to use P elements as tools to perform transformation and basic molecular analysis in other insects. In contrast, such species-specificity might be an asset for practical applications such as pest control because it would prevent horizontal transfer and generally restrict element dispersal to a single interbreeding population. In some instances it might therefore be advantageous to isolate TEs from the target insect one wishes to manipulate. The utilization of a variety of transposons may be necessary in any case. Once TEs become fixed in a population they generally establish a regulatory state in order to limit the genetic load they place on the host organisms and thus ensure their own survival. This suggests that the most suitable elements to act as vectors in a given insect may not be transposons which are endogenous to that species, but rather, transposons isolated from closely related species. The isolation of specific TEs from diverse organisms is obviously also of academic interest. TEs have been found in virtually all organisms for which any significant genome analysis or sequencing has been performed (BERG and HOWE 1989). However, until recently, individual TEs have been isolated primarily by serendipity. They have been discovered, not because researchers were searching for TEs, but because the elements were associated with some kind of genetic 119 instability or were integrated within DNA regions that were under study for other reasons. Similarly, serial passage of baculoviruses through lepidopteran larvae or tissue culture cells has resulted in the fortuitous acquisition of active TEs by the baculoviruses (FRASER et al. 1983). The intentional use of baculoviruses as 'traps' for TEs would have the advantage of preferentially selecting elements that are currently active. However, such a method would be limited to a relatively narrow range of organisms, since it requires baculoviruses which are virulent in both the insect species of interest and a cell culture line. Several methods have been developed which might be helpful for actively isolating TEs from less well characterized genomes. Each of these methods has both advantages and limitations. It is likely that experience may prove some to be more suitable for use with certain species or for detection of certain types of TEs. These methods can be broadly separated into two categories, namely, those which rely on homology to known TE families, and those which rely on techniques of phylogenetic screening. Low-stringency hybridization utilizing previously identified elements as probes (LANSMAN et al. 1985; HARRIS et al. 1988), and PCR with primers to conserved domains of TEs (WlCHMAN and VAN DEN BUSCHE 1992; ROBERTSON 1993; ROBERTSON and MACCLEOD 1993; RING et al. 1996) comprise the first category. While these methods can likely be applied to any species of interest, they are limited to specific elements or classes of elements. In addition these two techniques do not preferentially select for active elements. There are numerous techniques of phylogenetic screening which can be used to identify species-specific repetitive DNA. One method screens randomly amplified clones from a genomic library for differential hybridization to DNA from closely related species (WlCHMAN et al. 1985). A second method utilizes random PCR primers to amplify species-specific repeated DNAs (SKINNER 1992). 120 These two phylogenetic screening methods may amplify recently invasive TEs that are likely to be active; however, they are best suited for the detection of those elements present in very high copy number. A third phylogenetic method, termed 'representational difference analysis' (RDA) amplifies small differences between genomes through the use of subtractive hybridization in combination with PCR (LISITSYN et al. 1993). While extremely sensitive, RDA requires complex constructions and numerous rounds of enrichment and amplification. It is thus both time consuming and costly. In this chapter I describe a method which utilizes differential hybridization of genomic DNA to plaque lifts of a genomic library in order to screen for species-specific or strain-specific DNA. By analogy to the P and hobo elements of D. melanogaster, it is assumed that such differentially represented DNA may often represent recently invading TEs. This chapter consists of two major sets of experiments. First, two D. melanogaster strains were utilized as a model system to demonstrate the effectiveness of this differential screen and to optimize the procedure prior to its application to pest insects. High molecular weight genomic DNA was prepared from: (1) the it2 strain which contains both P and hobo transposons, and ( 2 ) the Canton-S strain which contains neither of these two families of transposons. A lambda genomic library was constructed from 7i2 strain DNA and plated out on bacterial lawns. Plaques were transferred to nylon membranes and replicate sets of filters were probed with labeled K2 and Canton-S DNA. Plaques which hybridize more strongly to the 7t2 genomic DNA were noted. The filters were then stripped and reprobed with labeled P and hobo element DNA fragments. As anticipated, many of the plaques which hybridized differentially to the two genomic DNAs corresponded to plaques containing the two known transposons. 121 The second set of experiments described in this chapter involved collecting pest insects and taking preliminary steps towards the isolation of TEs from them. The application of differential hybridization to the detection of species-specific or strain-specific DNA requires isolates of specimens from diverse geographical sources and/or from different time periods. Through personal fieldwork and generous donations I have collected samples from four insect species or species complexes for which such diversity is available. The first species is the variegated cutworm Peridroma saucia HUBNER which is a garden nuisance locally and is a pest of mint production in the United States. The second insect, the Gypsy moth Lymantria dispar (LINNAEUS), is a persistent forest pest in much of the eastern part of North America. The latter two insects are blackflies of the Simulium vittatum ZETTERSTEDT species and from the Simulium arcticum MALLOCH species complex. Blackflies are biting insects whose host range includes birds and mammals (including cattle and humans). They are a major factor in limiting opportunities in northern Canada by affecting resource development and the tourism industry. In addition, certain species of blackflies cause allergic reactions in livestock resulting in serious loss of production in the beef, milk and egg industry. Most seriously, the Simulium damnosum THEOBALD sibling species complex acts as a vector for onchocerciasis (river blindness), a debilitating disease affecting 20-30 million people in Africa and Latin America. Genomic libraries were made and preliminary differential screens were applied to the two lepidopteran insects. However, it became evident that modification of this procedure will be required before its further application to these species and to similar insect species that contain large amounts of highly repetitive and dispersed DNA. MATERIALS AND METHODS 122 Construction of the Drosophila Library: High molecular weight genomic DNA was prepared from Canton-S and rc2 strains of Drosophila melanogaster as described in CHAPTER III). Aliquots of DNA from the %2 strain were digested for various lengths of time with the restriction enzyme Sau3A-l and small samples of each digest were separated on an agarose gel to determine the duration which gave primarily 9-23 kb DNA fragments. The appropriate partially digested DNA was then ligated into Lambda replacement vector LambdaGem-11 according to the manufacturer's directions (PROMEGA 1988). Finally, the recombinant lambda phage were packaged using a Gigapack II Plus Packaging System, once again following the manufacturer's instruction manual (STRATAGENE 1990). The resulting iz2 genomic library contains more than 85,000 individual clones with insert sizes greater than 9,000 base pairs Screening of the Drosophila Library: Part of the unamplified ix2 library was plated out on a lawn of KW251 host bacterial cells at approximately 300 plaques per plate. Duplicate plaque lifts were prepared onto Hybond-N 0.45 micron nylon membranes (Amersham, Arlington Heights, IL). The replicate sets of filters were then pre-hybridized for 3 h at 65°C in a solution containing 6 x SSC (20 x SSC: 3M NaCl and 0.3M sodium citrate), 10% SDS, 5 x Denhardt's solution (100X Denhardt's: 2% w/v bovine serum albumin, 2% w/v Ficoll, and 2% w/v polyvinyl pyrollidone) and 20 Lig/ml of denatured salmon sperm DNA. About 250 ng of Canton-S and %2 genomic DNA was digested to completion with Sau3A-l, and labeled using a random primed DNA labeling kit (Boehringer Mannheim Biochemica, Germany). 123 These probes were added directly to the pre-hybridization solution and hybridization was allowed to proceed for 18 h at 65°C. Each set of filters was washed in 500ml 2 x SSC for 15 min.; 500ml 2 x SSC and 0.1% SDS for 20 min.; and 0.1 x SSC and 0.1% SDS for 10 min. All washes were carried out at 65°C in a shaking water bath. Autoradiograms were made and plaques which hybridized with more intensity to the n2 DNA probe were noted. The genomic probes were then completely removed by washing the niters twice for 1 h in 500ml of 5mM Tris-HCl (pH 8.0), 0.2mM EDTA (pH 8.0), 0.02% sodium pyrophosphate, and 0.1 x Denhardt's solution. The same fragments of P element and hobo element DNA described in CHAPTER TV were labeled by random prime labeling. Since these fragments encompass approximately 82% of the P element and 80% of the hobo element, they should hybridize to the majority of plaques which contain DNA homologous to these two TEs. Each probe was hybridized to one replicate set of plaque lifts, and the niters were washed as described above. Finally, autoradiograms were made and plaques that hybridized to the mobile element DNAs were compared to those which hybridized differentially to the genomic DNAs. Preparation of Recombinant Phage DNA: In several cases it was desirable to prepare specific lambda clones from the various genomic libraries for DNA isolation and further analysis (see RESULTS). The following methods of lambda preparation and DNA extraction, modified from Promega Technical Bulletin number 57 ( P R O M E G A 1988), were adopted after trying several other protocols. This method provided DNA that was free of bacterial DNA contamination and digested consistently with a variety of restriction enzymes. 124 Well isolated plaques containing the recombinant lambda clones of interest were picked from agar plates by removing a plug of agar with a cut off P1000 micro-pipet tip. The plugs were placed into individual 1.5 ml microcentrifuge tubes containing 100 ul of phage buffer made up of 20 mM Tris-HCl (pH7.4), lOOmM NaCl, and 10 mM MgS04- A single colony of KW251 bacterial cells was inoculated into 10 ml of LB media supplemented with 0.2% maltose and lOmM MgSC»4. The cells were allowed to grow overnight at 37°C with agitation. A 300 pi aliquot of this overnight bacterial culture was added to each microcentrifuge tube containing an agar plug and the tubes were incubated at 37°C for 20 minutes. The entire contents of each tube were added to a prewarmed (37°C) 250 ml Erlenmeyer flask containing 50 ml of LB supplemented with 10 mM MgSC»4. The flasks were shaken at 37°C until lysis occurred. After lysis, or after 6 hours if no lysis occurred, 0.5 ml of chloroform were added to each flask and the flasks were shaken for an additional 15 minutes. Cellular debris was pelleted by centrifugation at 10000 X g for 10 minutes and the supernatants were transferred to sterile tubes. About 0.5 ml of chloroform was added to each of these cleared lysates and they were stored at 4°C. For each DNA extraction, 20 ml of cleared lysate was transferred to an Oakridge tube and DNase I and RNase A were added to a final concentration of 1 pg/ml and 2 pg/ml respectively. After incubation for 30 minutes at 37°C, an equal volume of solution containing 20% polyethylene glycol (PEG) and 2 M NaCl was added. The contents of the tubes were mixed and the tubes were left in an ice bath for 1 hour to allow the phage to precipitate. The phages were into pelleted by centrifugation at 10000 X g for 20 minutes and the supernatants carefully drained. Each pellet was resuspended in 0.5 ml of phage buffer, transferred to a 1.5 ml microcentrifuge tube, and centrifuged for 4 minutes at 125 12000 RPM to remove debris. The supernatant was transferred to a fresh tube and extracted with chloroform which does not break open the phage capsids. DNase I and RNase A were added (same concentrations as above) and digestion was allowed to proceed for 30 minutes at 37°C. This chloroform extraction followed by a second digestion of nucleic acids is the primary modification of other PEG precipitation methods and it greatly reduced the contamination of samples by bacterial nucleic acids. Samples were then extracted once with phenol, once with phenol/chloroform and once with chloroform. Finally, phage DNA was precipitated with isopropanol, washed with 70% ethanol, and gently resuspended in 100 pi of TE (10 mM Tris, 1 mM EDTA, pH 8.0). Insect Collections: The availability of diverse geographical and/or temporal isolates of a given species is required in order to apply the plaque hybridization screen to identify differentially represented repetitive DNA. Samples were obtained from four different insects that met these criteria (see below). The differential screen was tentatively applied to the lepidopteran insects, however, it became obvious that in its present form the screen would not work with these insects due to their form of genome organization (see DISCUSSION). Since the genome organization of the Simulium species is not known, I decided not to pursue the differential screen further at this time. Nevertheless, I have decided to include a description of the other two insect collections here for completeness and for future reference. The first two insects are lepidopterans. The first is the variegated cutworm P. saucia. I have obtained one isolate of this lepidopteran species from M. I S M A N in the Department of Plant Sciences at UBC. Dr. Isman maintains an active colony which was initiated from insects captured in the Winnipeg area about 10 years ago. I obtained a second isolate of P. saucia from A. M O L D E N K E 126 at Oregon State University. She also maintains an active colony, but the founding insects were isolated recently from coastal Oregon. Henceforth I will refer to these as the Winnipeg and Oregon Peridroma isolates. These populations of Peridroma clearly meet the critereon of geographical isolation. The second lepidopteran insect is the Gypsy moth L. dispar. Diverse isolates of both the Asian Gypsy moth (AGM) and North American Gypsy moth (NAGM) have been obtained and catalogued by T. PFEIFER, a postdoctoral fellow in T. GRIGLIATTTS research group. The NAGM variety is from a containment facility operated by the USDA Northern Center for Forest Research in Hamden, Connecticut. In contrast, the AGM variety is a derivative of recent invasions of North America and was obtained from the Canadian Gypsy moth containment facility maintained by LEE HUMBLE (Natural Resources Canada, Canadian Forest Service, Victoria, B.C.). These isolates of NAGM and AGM varieties of Gypsy moths meet both the geographic and temporal isolation criteria. The last two pest insect systems both involve blackflies of the family Simulium. The first blackfly system includes several diverse isolates of the widely available S. vittatum species. The second involves flies from several sibling species within the S. arcticum species complex. In both of blackfly systems, my collection now contains some contributions from my own field work and some donated specimens. I will therefore start with a brief description of my methodology for collection and identification of local blackfly species. I have collected blackfly larvae and pupae extensively over a two year period. Blackfly larvae are aquatic. In flowing water they naturally attach to trailing vegetation or rocks and filter feed though fan-like appendages called cephallic fans. I have placed pieces of plastic, which larvae consider a great substitute for vegetation, in the waters of the Cultus Lake, the Chilliwack River, and a small stream near Cultus Lake. Larvae anchor themselves to the plastic 127 and often pupate on it. Some of these larvae were placed in Carnoy's solution and can be used to confirm the various species cytogenetically. The remaining larvae were brought back to UBC and were placed in buckets of swirling water obtained from the three original locations. The larvae were maintained in an incubator at about 10°C until they pupated. Individual pupae were placed in 1.5 ml microcentrifuge tubes along with pieces of moist paper towels. As soon as flies eclosed from the pupae, the entire tubes were stored at -70°C for later identification and DNA preparation. Each of the three collection sites contains a variety of blackfly species. With the generous help of DOUG CURRIE, I was able to distinguish the various species complexes at both the late larval stages and the pupal stage, with the latter being perhaps the easiest and most reliable (see CURRIE 1986). It is much more difficult to distinguish species at the adult stage. Unfortunately, I found that it was the adults that consistently yielded good quality DNA using my extraction procedure. However, freshly eclosed adult flies were frozen individually in microcentrifuge tubes together with their pupal cocoons. The species of the adults can therefore be identified from these remnants of the pupal stage (Figure 19). In addition to having insects from diverse geographical and/or temporal isolates it is important to have pure species in order to apply the differential screen. The S. arcticum species complex is a group of very closely related blackfly species which cannot be distinguished morphologically and often live together, but which are reproductively isolated. The blackflies from my field collections should therefore be checked cytologically, however, earlier collections from these locations did contain pure species. I now have blackfly isolates of both S. vittatum and S. arcticum that should meet all the requirements for the differential screen. For S. vittatum I have three isolates: 1) my collection from Figure 19. Simulium vittatum (left) and Simulium arcticum (right). Panels A and B show the adult flies which are very difficult to consistently distinguish morphologically. Panel C shows an S. vittatum pupa case that has a distinctive slipper shape and clearly differs from the boot shape of the S. arcticum pupa shown in panel D. Panels E and F show close-ups of the pupal gill filaments. The gill filaments of most blackfly species are very distinctive and easy to identify therefore they are extremely useful taxonomic characters. S. vittatum (Panel E) has 14-16 filaments, but the isolate from the Cultus Lake outflow consistently had 14. S. arcticum normally has 12 very thin filaments. 129 the Cultus Lake outflow (Figure 19 panels A, C and E); 2) a recent collection from Saskatchewan donated by G.S. MlRANPURI; and 3) a collection of the prototypical species from Iceland made about 40 years ago (provided by D. CURRIE). For S. arcticum I also have three isolates: 1) my collection from the Chilliwack River (Figure 19 panels B, D and F); 2) my collection from the stream near the Cultus Lake outflow; and 3) a collection from the Athabascan River region of Alberta (from D. CURRIE). Application of Differential Screen to Pest Insects: I prepared DNA from late instar larvae of both isolates of P. saucia in much the same way as described for single fly DNA preparation in Chapter III. The outer cuticle of larvae were removed prior to homogenization since they contained a black pigment which otherwise remained in the DNA preparations. Single larvae were homogenized in 250-300 ul of homogenization buffer, and the volumes of all other solutions were adjusted accordingly. The only other deviation from the earlier single fly procedure was that the concentration of EDTA in the homogenization buffer was doubled to overcome degradation that otherwise occurred when the DNA preparations were stored for extended periods. DNA was also prepared from larvae of both the North American Gypsy moth (NAGM) and Asian Gypsy moth (AGM) by several other members of the lab using a variety of methods. All the experimental procedures for making and screening pest insect libraries were exactly as described above for the Drosophila library. I constructed a genomic library of DNA from the Oregon isolate of P. saucia. Similar libraries were also constructed for both Gypsy moth varieties by B. LEE and T. PFEIFER in this lab. I plated the Oregon Peridroma library, made duplicate plaque lifts, and probed these with DNA from the two Peridroma 130 isolates. Differentially hybridizing plaques were purified from the original plates. I performed a similar differential screen of the AGM library using AGM and NAGM genomic DNAs as probes, and B. LEE performed the "reciprocal" differential screen of the NAGM library with the same two gypsy moth DNAs. RESULTS: Drosophila as a Model System : The genomic library of the ix2 strain of D. melanogaster, which contains both P and hobo elements, was plated out on 12 plates at approximately 300 plaques per plate and replicate plaque lifts were made. Duplicate sets of filters were hybridized to Sau3A-I digested genomic DNA from either of two D. melanogaster strains: Canton-S, which contains neither P nor hobo elements, and ix2. I found that pre-digestion of the genomic DNAs greatly enhanced the hybridization differential obtained to the two probes (data not shown). This suggests that the size of the genomic DNA fragments used as a probe is important. Figure 20, panels A and B, show replicate filters of the n2 strain library probed with digested Canton-S and iz2 genomic DNA respectively. The arrows point to three of the seven plaques on this plate that hybridized more intensely to the K2 DNA. These plaques contained DNA present in the 7t2 strain, but absent or present in reduced copy number in the Canton-S strain. A total of 32 such differentially hybridizing plaques were noted in a quick comparison of all the duplicate autoradiograms. The genomic probes were removed and the replicate sets of filters were hybridized to either P element or hobo element probes. A total of 58 plaques hybridized to the hobo element probe, and 18 plaques hybridized to the P Figure 20. Hybridization of replicate plaque lifts of a p2 genomic library to: A) CS genomic DNA; B) p2 genomic DNA; and C) hobo element DNA. The three arrows indicate plaques which can be easily distinguished as differentially represented on panels A and B. On the original autoradiograms of these replicate lifts it was possible to detect seven differentially hybridizing plaques. All seven corresponded to plaques which contained hobo element DNA (panel C). 132 element probe. Figure 20, panel C, shows the filter from panel A reprobed with hobo element DNA. Note that all three of the previously noted plaques (arrows) contained hobo element sequences. Of the 32 differentially hybridizing plaques that were originally selected, 26 were hobo homologous, 3 were P homologous, and 3 were homologous to neither transposable element. The three plaques which appeared to hybridize differentially to the two genomic DNAs, but which did not correspond to the two known transposable elements, were examined further. The plaque-purified phages were plated out and transferred to nylon membranes. Genomic probes were unable to detect any differential hybridization to these membranes. For two of the clones, close examination of the original autoradiograms revealed that the differential hybridization initially observed could be explained by poor lifts around the edges of one of the duplicate membranes. The original differential hybridization to the third clone could not be readily explained. I therefore prepared a small amount of DNA from this clone, digested the DNA with several restriction enzymes, and transferred the DNA to nylon membranes. When these Southern blots were hybridized to the two genomic probes, no differential bands were observed (not shown). I conclude that all three of these plaques represented false positives, that is, they actually contained no differentially represented DNA. Differential Screens of Pest Insects : From the Peridroma screen of approximately 5000 plaques, I purified 25 putative differentially hybridizing clones. I also isolated 9 putative differentially hybridizing clones from the similar sized screen of the AGM, and B. LEE isolated about 20 putatives from the "reciprocal" differential screen of the NAGM library with the two gypsy moth DNAs. However, while each of these "putatives" appeared to hybridize more intensely to one of the two probes, in almost all cases 133 they were not clear +/- situations as had been observed in the Drosophila screen. These putatives were re-plated during the plaque purification process. With the Peridroma samples I made plaque lifts and checked them for differential hybridization of the purified clones. Despite attempts to equalize the size of the fragments of genomic DNA used for the two probes, and the amount of radioactivity incorporated, there was no clear evidence of differential hybridization. In an attempt to clarify these results, small amounts of lambda DNA were prepared of all putative differentially hybridizing clones from all three of the pest insect screens (Peridroma, AGM, NAGM). DNA was isolated from each preparation and digested with several different restriction enzymes. Each digested sample was then split in half and these duplicates were separated on parallel agarose gels and transferred to nylon filters. These filters were hybridized to digested genomic DNA from the two appropriate Peridroma or Lymantria isolates, and autoradiographed. A typical autoradiograph from my secondary screening of some putative clones from the AGM screen is shown in Figure 2 1 . The majority of bands on all autoradiographs hybridized with about the same intensity to the two different genomic DNAs. Within a few samples there were one or two bands that appeared to hybridize more strongly to the DNA from which the library was made. However, none of these differentially hybridizing bands proved to be reproducible so I must classify them as false positives. 134 Figure 21. Typical replicate Southern blots of Lymantria clones. Lanes 1-9 contain DNA from recombinant clones from the AGM library and the lane 10 contains a clone from the NAGM library. All DNA samples were digested with the restriction enzyme Sau3A-I. All lanes, including lane 9, contain roughly equal amounts of DNA. Differences in hybridization intensity are due to the number of times the DNA is repeated in the genome. These clones were isolated from plaques which appeared to hybridize differentially to genomic probes from the two Lymantria varieties. Panel A shows hybridization to an AGM genomic probe and panel B shows hybridization of a replicate Southern blot hybridized to a NAGM genomic probe. There are no clear differentially hybridizing bands. DISCUSSION 135 Drosonhila TEs as a Model System : The complete hobo andP TEs are very similar in size (3.0 kb versus 2.9 kb respectively), and the TE fragments that I used as probes encompass similar amounts of the complete elements (80% and 82%). Flies of the n2 strain contain 30-50 copies of P elements (BINGHAM et al. 1982; O'HARE and RUBIN 1983; GOOD et al. 1989) and roughly twice that number of hobo elements (BLACKMAN et al. 1987; YANNOPOULOS et al. 1987; MEISTER unpublished). One would therefore predict that roughly twice as many clones would contain hobo elements than would contain P elements. However, 58 plaques hybridized to the hobo element probe and only 18 plaques hybridized to the P element probe. The TE probes therefore detected a greater proportion of hobo homologous clones than expected. It should be remembered, however, that the element copy numbers given above are only estimates. Moreover, the precise number of each element will vary from fly to fly since both families of TEs are actively mobile. Utilizing the differential hybridization method, I initially detected about 45% of all hobo-containing plaques, but only about 17% of all P element-containing plaques. The differences in efficiency of detection of plaques containing the two families of TEs is probably largely due to their respective copy numbers in the n2 strain. Since there are about twice as many hobo elements per fly, one would expect the concentration of hobo element DNA to be roughly twice the concentration of P element DNA in the TC2 genomic probe. This higher concentration of hobo DNA would result in more efficient detection of /io&o-containing plaques. This suggests that the amount of genomic DNA utilized as a probe will need to be adjusted according to the copy number of the elements one wishes to detect. Of course, in two populations where the 136 differences in copy number of any DNA segment is very low, it would be difficult to detect these differences. My original selection of differentially hybridizing plaques was fairly conservative. Closer scrutiny of the autoradiograms made from genomic probes allowed detection of many more of the plaques which contained mobile elements without significantly increasing the number of false positives. In fact, we were able to detect up to 39 (67%) of the plaques containing hobo homologous DNA and 9 (50%) of the plaques containing P homologous DNA. I found that one way to greatly enhance the detection of hybridization intensity differences is to overlay clear acetates containing colorized laser photocopies of the autoradiograms. I routinely colorize the autoradiogram produced with one genomic probe blue and the autoradiogram produced with the second genomic probe red. When these are overlaid, plaques which hybridized equally to both probes appear purple. Differentially hybridizing plaques remain either blue or red. While colorization seldom results in detection of positives that would otherwise have been overlooked, it makes the comparison of the autoradiograms far less tedious and time consuming. The model system utilizing Drosophila has demonstrated that the differential hybridization screen is capable of identifying strain-specific low to middle repetitive DNA elements in Drosophila. Judicial modification of the hybridization conditions would allow for the selection of elements present in different copy numbers. This simple differential screen is more efficient than screening randomly amplified clones. For situations where repetitive sequences are expected, this technique is also adequately sensitive and much simpler and more cost effective than RDA. 137 Differential Hybridization Screens of Pest Insects : Once the differential hybridization technique was optimized in the Drosophila model system, the objective was to apply it to several pest insects. Four pest insect systems were originally selected based on the availability of pure species from diverse geographical and/or temporal isolates. The differential screen was applied to two of these systems: the variegated cutworm P. saucia and the gypsy moth L. dispar. The screen of the P. saucia Oregon genomic library resulted in the identification of 25 plaques which appeared to hybridize more intensely to genomic DNA from the Oregon isolate than from the Winnipeg isolate. Since the Oregon isolate was originally collected about 10 years later than the Winnipeg isolate, it was hoped that these plaques might contain TEs that had recently spread through the wild populations. The clones containing putative differentially represented DNA were purified and re-plated. Unfortunately there was no clear evidence of differential hybridization to the plaques formed by the purified clones. To confirm this result, small scale lambda DNA preps were prepared from the recombinant clones. These DNAs were digested with several restriction enzymes and the resulting fragments were electrophoresed on agarose gels, Southern blotted, and probed with the two genomic DNAs. If a clone comprised only DNA which was equally represented in the two Peridroma isolates then all analogous bands between duplicate samples would be of roughly equal intensity. However, if there were both differentially and non-differentially represented DNAs in these clones, then there should be a mixture of bands of both unequal and equal intensities between duplicate samples. There were no bands which clearly and reproducibly hybridized more intensely to either of the genomic probes. I therefore concluded that the putative differentially represented DNAs 138 were false positives, perhaps due to unequal transfer of plaques to the original duplicate niters. The original differential screening of both the AGM and NAGM Gyspy moth libraries also produced some plaques which appeared to contain DNA present in one variety, but absent or present in reduced copy number in the other variety. However, once again, closer scrutiny revealed these to be false positives. This does not mean that the differential screen 'failed" on Peridroma and Lymantria. It simply means that either there was no differentially represented repetitive DNA in the two isolates of P. saucia, or that detection of any such DNA was beyond the sensitivity of the plaque screening method under the conditions utilized. Future Perspectives: One potential limitation of the differential hybridization screen described above is that, at least using the present conditions, it may only be applicable to species which contain what has been termed the 'long period interspersion' pattern of organization of repetitive DNA within the genome (for review see DAVIDSON et al. 1975). This pattern is found in several economically and/or scientifically interesting species including D. melanogaster (MANNING et al. 1975), the honeybee Apis mellifera LINNAEUS (CRAIN et al. 1976), the red flour beetle Tribolium castaneum (HERBST) (BROWN et al. 1990), the midge Chironomous tentans FABRICIUS (WELLS et al. 1976), and the mosquito Anopheles quadrimaculatus SAY (BLACK and RAI 1988). Nevertheless, the majority of eukaryotes contain the alternate pattern of genome organization called 'short period interspersion1. Since organisms with such patterns contain large amounts of highly repetitive and highly dispersed DNA, virtually all lambda clones within a genomic library hybridize to genomic DNA. Such 139 hybridization to both genomic probes, regardless of their origin, would make differential screening very difficult. Highly repetitive dispersed DNA would cause similar complications to most other differential or subtractive hybridization based methods. It should be noted that the amount and pattern of interspersed repetitive DNA can vary greatly among even closely related insect species. For example in mosquitoes, species from the Anopheles genus exhibit the classic 'long pattern of interspersion', while species of the Aedes genus exhibit the alternative 'short pattern interspersion', and species of the Culex genus exhibit some sort of intermediate pattern (BLACK and RAI 1988, COCKBURN and MITCHELL 1989). The two classical patterns of genome organizations may thus represent two extremes of a continuum. It would obviously be valuable to know something about the repetitive DNA content of an insect species prior to the application of the differential screen. Since such information is available for only a limited number of species, this information would be of academic interest in any case. It has been established that hybridization with total genomic DNA can be used to identify recombinant clones containing repetitive sequences (DOWSETT and YOUNG 1982, HALL 1986). COCKBURN and MITCHELL (1989) have used this observation to develop a simple and rapid library hybridization technique to examine genome organization. A genomic (lambda phage) library is plated out and then individual well-separated plaques are inoculated onto a lawn of bacteria in a grid pattern. Plaque lifts are then made and probed with total genomic DNA under carefully controlled conditions. The intensity of hybridization to a given plaque depends on two factors: 1) what proportion of the cloned insert is repetitive DNA, and 2) what fraction of the probe is homologous to the insert. For moderately repeated genomic sequences the hybridization intensity should depend primarily on the latter, since the insert 140 DNA bound to the filter should be in excess. The fraction of the probe that is homologous to the insert will in turn be dependent upon the length of the repeat, its repeat frequency, and the size of the genome. Thus, the number of plaques with positive signals gives an indication of the interspersal of repetitive DNA, and the intensity of the signal gives an indication of its repeat frequency. C O C K B U R N and M I T C H E L L (1989) applied this technique to several species and demonstrated that the results were consistent with those obtained previously by Cot analysis. On the downside they point out that clones can be misrepresented in libraries for a variety of reasons, that clones without inserts could give false negatives, and that long moderately repeated sequences could give the same intensity of signals as shorter highly repeated sequences. While Cot analysis gives a more accurate picture of genome organization and genome size, library hybridization is faster than Cot analysis. Library hybridization should also provide enough information to determine which species are likely to be amenable to the differential screen I have proposed. In fact, while I have not purposefully analyzed my results to determine genome organization, by performing the differential screen I have essentially performed the library hybridization screen of C O C K B U R N and M I T C H E L L on both Peridroma and Lymantria. Since virtually all of the plaques hybridized to genomic DNA, it can be inferred that both these insect species do contain large amounts of highly dispersed and repetitive DNA sequences. Perhaps the most important advantage of the library hybridization screen over C o t analysis is that the library is then immediately available for use in the differential screen and/or for isolating full-length copies of any PCR amplified homologs of known TEs. Indeed, the PCR method has been used to obtain a full-length retrotransposon f r o m my Peridroma library ( R I N G et al. 1996). 141 I have suggested that species with highly dispersed and repetitive DNA may not be amenable to the differential screen under the conditions that I have utilized. However, there may be several ways to overcome this species limitation. For instance, one could make libraries with smaller inserts such that not all clones would contain the highly repetitive DNA. Alternatively, one could include excess unlabelled highly repetitive fractions of the genomic DNA as a blocking agent with the pre-hybridization mixture. Further experimentation is required to determine how widely applicable this differential hybridization method could be if such modifications were incorporated. The genome organizations of the various blackfly species that I have collected are unknown and their determination would require considerable effort and time. Moreover, substantially more experimentation would be required to modify the differential screen to work on genomes with highly repetitive and dispersed DNA. Even if these efforts were undertaken, there is still no guarantee that any differentially represented repetitive DNA is present within the Lepidopterans or blackflies that I have collected. Indeed the only indication that there are mobile TEs within any of these species is the observation of abnormal ovaries in the progeny of crosses between the Oregon and Winnipeg isolates of Peridroma (Figure 22). These ovaries are obviously reminiscent of those produced by P and hobo element induced hybrid dysgenesis in Drosophila (Figure 9). While I feel that further experimental application of the differential screen is worth pursuing, I must regrettably leave this for a later time or for others to pursue. 142 Figure 22. Gonadal dysgenesis in P. Saucia? Panel A shows normal Peridroma ovaries. Panel B shows a set of ovaries in a progeny female from a cross of Oregon males to Winnepeg females. These are obviously reminiscent of P or hobo element induced gonadal dysgenesis in Drosophila. However, further studies should carefully monitor both the age and rearing temperature of the dissected moths. 143 All of the phylogenetic screening methods (see INTRODUCTION) should allow the isolation of strain-specific or species-specific repetitive DNAs from a variety of organisms. While I have focused my efforts on recovering TEs, it should be noted that not all such differentially represented DNA would represent TEs. For instance, it has been shown that satellite DNA fractions can vary greatly between closely related species ( B R U T L A G 1980). While this may be a complication in terms of isolating TEs, any species-specific DNAs would be useful as phylogenetic markers. Further characterization of these sequences would indicate which, if any, might be active TEs with potential as tools for the study and eventual manipulation of insect genomes. CHAPTER VI: SUMMARY OF THESIS RESEARCH 145 The combination of genetic engineering and germline transformation offers the potential for practical alterations of insect genomes. If these techniques are to be used to enhance beneficial insects or to control pest insects, then the engineered DNA must be dispersed through the target insect populations. TEs may provide ideal vehicles for dispersal of genetically engineered constructs. The experiments described in this thesis address some of the fundamental questions that must be answered before this proposal can be given serious consideration. In addition these experiments either investigate the ability of various Drosphila TEs to disperse, or develop methods for the isolation of endogenous TEs from species of interest. In Chapter II the ability of a TE with an inserted gene to spread through experimental populations was examined. In addition, I tested whether or not the passenger gene retained its ability to encode an active protein. Several D. melanogaster laboratory populations were initiated with female flies null for alcohol dehydrogenase activity that contained no P elements. Most of the females were mated to males of the same strain; however, 1% or 10% of the females were mated to males from a strain which had previously been transformed with a helper P element and a P element/alcohol dehydrogenase gene construct. The dispersal of P elements to new genomes was monitored at each generation by randomly selecting females and performing DNA hybridization assays on dissected ovarian tissue. Each of the same females was also tested for alcohol dehydrogenase activity using a simple histochemical assay. I found that, despite an approximate three-fold increase in size, the P element constructs containing a functioning gene are still capable of rapid dispersal through the experimental populations. I also showed that many of the inserted alcohol dehydrogenase genes still encode an active product. 146 There is some evidence that hobo elements, like P elements, have recently spread through natural populations of D. melanogaster. The experiments described in Chapter III tested whether or not hobo elements are capable of rapid dispersal and accumulation in mixed populations. The populations were initiated with non-hobo containing Canton-S (CS) females, 2% or 20% of which had been previously mated to males from the hobo containing strain Oregon R (OR). The remainder had been mated to CS males. Southern blots of DNA prepared from single flies show that hobo elements spread rapidly and were present within virtually all individuals of all populations within ten generations. Dot blots reveal that, within those flies that contain elements, the mean amount of hobo hybridizing DNA per individual: 1) decreased in the first few generations, 2) increased in subsequent generations until about generation ten, and 3) stabilized at approximately 50% of the amount of hobo DNA present within individuals of the element donating OR strain. The total hobo hybridizing DNA in the populations underwent a steady increase and also stabilized at about 50% of that present in the OR strain. Since it is hypothesized that P and hobo elements spread through natural Drosophila populations within the same time period, it is of interest to know whether or not both elements can disperse and accumulate simultaneously within experimental populations. Populations similar to those in Chapter III, but in which P and hobo element-containing genomes of the n2 strain were introduced into CS populations, are described in Chapter IV. I had previously shown that P elements quickly spread to all progeny genomes and that the amount of P element homologous DNA in these populations quickly approached the amount of P element DNA of the element donating TC2 strain. I have now used dot blots containing DNA prepared from both single and multiple flies to confirm this earlier result and to show that hobo elements also spread to all 147 progeny individuals. The amount of hobo homologous DNA also accumulated, but only to about 40% of that present in the it2 strain. Chapter V described a differential DNA hybridization method for detecting moderately repetitive strain-specific or species-specific DNA. Two D. melanogaster strains, one with and one without TEs, were utilized as a model system to demonstrate the effectiveness of this procedure. A genomic library was constructed from flies of the K2 strain. Duplicate plaque lifts of this library were probed with DNA from the same strain and with DNA from the Canton-S strain. Plaques which hybridized more strongly to the genomic DNA that contained elements were noted, and then the filters were stripped and reprobed with P and hobo element DNA. Many of the differentially hybridizing plaques were shown to contain DNA homologous to either the hobo or P elements known to be present in the n2 strain, but absent in the CS strain. This differential screen was then tentatively applied to two pest insect species, however, some modification of the procedure may be required before it will be widely applicable. The results of all of these experiments are supportive of the possible use of TEs as vectors for the dispersal of engineered DNA through target insect populations. The utility of TEs as dispersal vectors is predicated on the ability of TEs carrying a large DNA insert to spread rapidly through a population. In addition, the inserted DNA must be replicated with a high degree of fidelity during this dispersal. Chapter II demonstrated that P elements satisfy both of these requirements. However, one potential limitation of using P elements is that their mobility appears to be very species specific. One way around this difficulty is to utilize other TEs which are less host specific. Chapter III showed that hobo elements are also capable of rapid dispersal and accumulation. This ability, combined with their broader host range, may allow hobo elements or hobo-like elements to act as dispersal vectors in many insect species. 148 Furthermore, Chapter IV showed that the ability of hobo elements to spread was not greatly decreased by the concurrent dispersal of a second TE. This suggests that the dispersal of hobo elements containing engineered DNA may not be significantly impeded if one or more endogenous TEs are active within the target population. Another way to overcome the species specificity limitation would be to use endogenous TEs from species closely related to the target population. This may be necessary in any case since the mobility of most TEs becomes restricted once the element becomes established in a population. Chapter V described a differential hybridization screen that, with some modification, should allow the isolation and cloning of any repetitive DNA present in one species or isolate , but absent or present in reduced copy number in another species or isolate. By analogy to the recent invasion of D. melanogaster by P elements, such differentially represented DNA is likely to be recently invading TEs which are actively mobile and which may be adapted into dispersal vectors. TE mediated germline transformation has recently been demonstrated in several economically or medically important pest insect species. For example: (l)a Tei-like element called Minos from Drosophila hydei S T U R T E V A N T has been used to transform Ceratitis capita ( W I E D E M A N N ) , the Mediterranean fruit fly that causes immense destruction of coffee and fruit crops ( L O U K E R I S et al. 1995) ; and (2) the mariner element isolated from D. mauritiana has been used to transform Aedes aegypti ( L I N N A E U S ) , the mosquito that acts as vectors for yellow fever ( C O A T E S et al. 1998). The genetic control of these species through the TE mediated dispersal of engineered DNA may soon follow. 149 LITERATURE CITED A N C H O R D O G U Y , T . J., D . L . CR A W F O R D , I. HA R D E W I G and S. C. H A N D , 1996. Heterogeneity of DNA binding to membranes used in quantitative dot blots. Biotechniques 20: 754-756. A N D R E W S , J. D. and G. B. GLOOR, 1995. A role for the XPluecine zipper in regulating P element transposition in Drosophila melanogaster. Genetics 141: 587-594. A N X O L A B E H E R E , D., K. HU, D. N O U A U D , G. P E R I Q U E T , and S. R O N S S E R A Y , 1984. The geographical distribution of P-M hybrid dysgenesis in Drosophila melanogaster. Genetique, Selection et Evolution 16:15-26. A N X O L A B E H E R E , D., M. G. KlDWELL and G. P E R I Q U E T , 1988. Molecular characteristics of diverse populations are consistent with a recent invasion of Drosophila melanogaster by mobile P elements. Mol. Biol. Evol. 5: 252-269. A N X O L A B E H E R E , D., D. N O U A U D , and G. P E R I Q U E T , 1982. Cytotype polymorphism of the P-M system in two wild populations of Drosophila melanogaster. Proc. Natl. Acad. Sci. USA 79: 7801-7803. A N X O L A B E H E R E , D., D. N O U A U D , G. P E R I Q U E T and P. T C H E N , 1985. P element distribution in Eurasian populations of Drosophila melanogaster: a genetic and molecular analysis. Proc. Natl. Acad. Sci. USA 82: 5418-5422. A N X O L A B E H E R E , D. and G. P E R I Q U E T , 1987. P-homologous sequences in Diptera are not restricted to the Drosohilidae family. Genet. Iber. 39: 211-222. A S H B U R N E R , M. 1989. Drosophila. A Laboratory Handbook. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. A T K I N S O N , P. W., W. D. W A R R E N and D. A. O ' B R O C H T A , 1993. The hobo transposable element of Drosophila can be cross-mobilized in houseflies and excises like the Ac element of maize. Proc. Natl. Acad. Sci. USA 90: 9693-9697. B A K E R , B., J. S C H E L L , H. LORZ andN. V. F E D O R O F F , 1986. Transposition of the maize controlling element Activator in tobacco. Proc. Natl. Acad. Sci. USA 83: 4844-4848. BAZIN, C. and D. HlGUET, 1996. Lack of correlation between dysgenic traits in the hobo system of hybrid dysgenesis in Drosophila melanogaster. Genet. Res. 67: 219-226. BAZIN, C , J. WILLIAMS, J. BELL and J. SlLBER, 1993. A deleted hobo element is involved in the unstable thermosensitive vg(al) mutation at the vestigial locus in Drosophila melanogaster. Genet. Res. 61:171-176. 150 B E A L L , E. L . , A. A D M O N and D. C. R l O , 1994. A Drosophila protein homologous to the human p70 Ku autoimmune antigen interacts with the P transposable element inverted repeats. Proc. Natl. Acad. Sci. USA 91:12681-12685. B E A L L , E. L . and D. C. RlO, 1996. Drosophila IRBP/Ku p70 corresponds to the mutagen-sensitive mus309 gene and is involved in P-element excision in vivo. Genes and Dev. 10:921-933. B E C K E R , H. and R. K U N Z E , 1997. Maize Activator transposase has abipartate binding domain that recognizes subterminal sequences and the terminal inverted repeats. Mol. Gen. Genet. 254: 219-230. B E L L E N , H . J . , C. J . O ' K A N E , C. W I L S O N , U. G R O S S N I K L A U S , R. K. P E A R S O N et al., 1989. P-element-mediated enhancer detection: a versatile method to study development in Drosophila. Genes Dev. 3: 1288-12300. B E N Y A J A T I , C , A. R. P L A C E and W. S O F E R , 1983. Formaldehyde mutagenesis in Drosophila: molecular analysis of ADH-negative mutants. Mut. Res. 111:1-7. B E R G , D. E. and M. M. H O W E , 1989. Mobile DNA. Washington, D.C: American Society of Microbiology. B I E M O N T , C , C. G A U T I E R and A . H E I Z M A N N , 1988. Independent regulation of mobile element copy number in Drosophila melanogaster inbred lines. Chromosoma 96: 291-294. B I G O T , Y . , C. A U G E - G O U I L L O U and G. P E R I Q U E T , 1996. Computer analysis reveals a hobo-like element in the nematode Caenorhabditis elegans, which presents a conserved transposase domain common with the Tel-Mariner transposon family. Gene 174:265-271. B I N G H A M , P. M., M. G. K l D W E L L and G. M. R U B I N , 1982. The molecular basis of P-M hybrid dysgenesis: the role of the P element, a P-strain-specific transposon family. Cell 29: 995-1004. B r N G H A M , P.M., R. L E V I S and G. M. R U B I N , 1981. The cloning of the DNA sequences from the white locus of Drosophila melanogaster using a novel and general method. Cell 2 5 : 693-704. B L A C K , D. M., M. S. J A C K S O N , M. G. K l D W E L L and G. A. D O V E R , 1987. KP elements repress P-induced hybrid dysgenesis in Drosophila melanogaster. EMBO J . 6: 4125-4135. BLACK, W. C , IV and K. S. RAI, 1988. Genome evolution in mosquitoes: intraspecific and interspecific variation in repetitive DNA amounts and organization. Genet. Res. Camb. 51:185-196. 151 B L A C K M A N , R. K. and W. M. G E L B A R T , 1989. The transposable element hobo of Drosophila melanogaster, pp. 523-530 in Mobile DNA, edited by D. E. B E R G and M. M. H O W E . American Society for Microbiology, Washington D.C. B L A C K M A N , R. K., R. G R I M A I L A , M. M A C Y , D. K O E H L E R and W. M. G E L B A R T , 1987. Mobilization of hobo elements residing within the decapentaplegic gene complex: suggestion of a new hybrid dysgenesis system in Drosophila melanogaster. Cell 49: 497-505. B L A C K M A N , R. K., M. M. K O E H L E R , R. G R I M A I L A and W. M. G E L B A R T , 1989. Identification of a fully-functional hobo transposable element and its use for germ-line transformation of Drosophila. EMBO J. 8: 211-217. B O L S H A K O V , V. N . , A. P. G A L K I N , L. Z . K A I D A N O V , V. A. G V O Z D E V and C. L O U I S , 1994. Closely related Drosophila melanogaster strains with altered fitness also show changes in their hobo element properties. Genet. Sel. Evol. 26: 205-216. B O N N I V A R D , E., D. H I G U E T and C. B A Z L N , 1997. Characterization of natural populations of Drosophila melanogaster with regard to the hobo system; a new hypothesis on invasion. Genet. Res.69: 197-208. B O U S S Y , I.A., 1987. A latitudinal cline in P-M gonadal dysgenesis potential in Australian Drosophila melanogaster populations. Genet. Res. (Camb.) 49: 11-18. B O U S S Y , I. A. and S. B. D A N I E L S , 1991. hobo transposable elements in Drosophila melanogaster and D. simulans. Genet. Res. 58: 27-34. B O U S S Y , I. A., M. J. H E A L Y , J . G. O A K E S H O T T and M. G. KlDWELL, 1988. Molecular analysis of the P-M gonadal dysgenesis cline in eastern Australian Drosophila melanogaster. Genetics 119: 889-902. B O U S S Y , I. A. and M. I T O H , 1996. The hobo transposable element family invaded Drosophila melanogaster twice (Abstract), p. 27 in Horizontal gene transfer: Implications and consequences. Fallen Leaf Lake Conference, University of California, Davis. B O U S S Y , I. A. and M. G. KlDWELL, 1987. The P-M Hybrid dysgenesis cline in Eastern Australian Drosophila melanogaster: Discrete P, Q and M' regions are nearly contiguous. Genetics 115: 737-745. B R E G L I A N O , J . C. and M. G. KlDWELL, 1983. Hybrid dysgenesis determinants, pp. 402-404 in Mobile Genetic Elements, edited by Y. A. S H A P I R O . Academic Press, London. B R E N N A N , M. D . , R . G . R O W A N and w. J. D I C K I N S O N , 1984. introduction of a functional P element into the germ-line of Drosophila hawaiiensis. Cell 38: 147-151. 152 B R O O K F I E L D , J. F . Y., E. M O N T G O M E R Y and C. L A N G L E Y , 1984. Apparent absence of transposable elements related to P elements in Drosophila melanogaster in other species of Drosophila. Nature 310:330-332. B R O W N , S. J., J. K. H E N R Y , W. C. B L A C K IV and R. E. D E N E L L , 1990. Molecular genetic manipulation of the red flour beetle: genome organization and cloning of a ribosomal protein gene. Insect Biochem. 20: 185-193. B R U T L A G , D. L. 1980. Molecular arrangement and evolution of heterochromatic DNA. Ann. Rev. Genet. 14: 121-144. CAIZZI , R., C. C A G G E S E andS. P l M P I N E L L I , 1993. Bari-1, a new transposon-like family in Drosophila melanogaster with a unique heterochromatic organization. Genetics 133: 335-345. C A L V I , B. R. and W. M. G E L B A R T , 1994. The basis for germline specificity of the hobo transposable element in Drosophila melanogaster. EMBO J. 13:1636-1644. C A L V I , B. R., T. J. H O N G , S. D. F l N D L E Y and W. M. G E L B A R T , 1991. Evidence for a common evolutionary origin of inverted repeat transposons in Drosophila and plants: hobo, Activator, and Tam3. Cell 66: 465-471. C A P Y , P . , R. V l T A L I S , T. L A N G I N , D. H l G U E T and C. B A Z I N , 1996. Relationships between transposable elements based upon the integrase-transposase domains: is there a common ancestor? J. Mol. Evol. 42: 359-368. C A P Y , P . , T. L A N G I N , D. H l G U E T P . M A U R E R and C. B A Z I N , 1997. Do the integrases of LTR retrotransposons and class II element transposases have a common ancestor? Genetica 100:63-72. C H A R L E S W O R T H , B. and D. C H A R L E S W O R T H , 1983. The population dynamics of transposable elements. Genet. Res. 42: 1-27. C H A R L E S W O R T H , B. and C. H . L A N G L E Y , 1986. The evolution of self-regulated transposition of transposable elements. Genetics 112: 359-383. C H A R L E S W O R T H , B. and C. H . L A N G L E Y , 1989. The population genetics of Drosophila transposable elements. Annu. Rev. Genet. 23: 251-287. C H A R L E S W O R T H , B., P . S N I E G O W S K I and W. S T E P H A N , 1994. The evolutionary dynamics of repetitive DNA in eukaryotes. Nature 371:215-20. C H E N , J., I. M. G R E E N B L A T T and S. L. D E L L A P O R T A , 1987. Transposition of Ac from the P locus of maize into unreplicated chromosomal sites. Genetics 117: 109-16. C H E N , J., I. M. G R E E N B L A T T and S. L. D E L L A P O R T A , 1992. Molecular analysis of Ac transposition and DNA replication. Genetics 130:665-676. 153 C H U C K , G . , T . R O B B I N S , C . N I J J A R , E . R A L S T O N , N . C O U R T N E Y - G U T T E R S O N et al., 1993. Tagging and cloning of a Petunia flower color gene with the maize transposable element Activator. Plant Cell 5: 371-378. C L O U G H , D . W., H . M. L E P I N S K E , R . L . D A V I D S O N and R . V . S T O R T I , 1985. Drosophila P element-enhanced transfection in mammalian cells. Moi. Cell. Biol. 5: 898-901. COATES, C J., N . JASINSKIENE, N . MlYASHIRO, and A. A. JAMES, 1998. Mariner transposition and transformation of the yellow fever mosquito, Aedes aegypti. Proc. Natl. Acad. Sci. U S A 95: 3748-3751. COATES, C. J., K . N . JOHNSON, H . D. PERKINS, A. J. HOWELLS, D. A. O'BROCHTA et al., 1996. The hermit transposable element of the Australian sheep blowfly, Lucilia cuprina, belongs to the hAT family of transposable elements. Genetica 97: 23-31. COCKBURN, A. F. and S. E. MITCHELL, 1989. Repetitive DNA interspersion patterns in diptera. Arch, of Insect Bioch. and Phys. 10: 105-113. C O E N , E. S., T . P. R O B B I N S , J. A L M E I D A , A. H U D S O N and R . C A R P E N T E R , 1989. Consequences and mechanisms of transposition in Antirrhinum majus, pp. 413-436 in Mobile DNA, edited by D. E. BERG, and M. M. HOWE. American Society for Microbiology, Washington D.C. CRAIN, W. R . , E. H . DAVIDSON and R . J. BRITTEN, 1976. Contrasting patterns of DNA sequence arrangement in Apis mellifera (honeybee) and Musca domestica (housefly). Chromosoma (Berl.) 59:1-12. CURRIE, D. C . , 1986. An annotated list of and keys to the immature black flies of Alberta (Diptera: Simuliidae). Memoirs of the Entomological Society of Canada 134. DANIELS, S. B., A. CHOVNICKand I. A. BOUSSY, 1990. Distribution of hobo transposable elements in the genus Drosophila. Moi. Biol. Evol. 7: 589-606. D A N I E L S , S. B., S. H . C L A R K , M. G. K I D W E L L and A. C H O V N I C K , 1987. Genetic transformation of Drosophila melanogaster with an autonomous P element: phenotypic and molecular analyses of long-established transformed lines. Genetics 115: 711-723. DANIELS, S. B., M. MCCARRON, C. LOVE and A. CHOVNICK, 1985. Dysgenesis-induced instability of rosy locus transformation in Drosophila melanogaster: analysis of excision events and the selective recovery of control element deletions. Genetics 109:95-117. D A V I D S O N , E. H . , G. A. G A L A U , R . C. A N G E R E R and R . J. B R I T T E N , 1975. Comparative aspects of D N A organization in metazoa. Chromosoma (Berl.) 51: 253-259. 154 D E V A U L T , J . D. and S. K . NARANG, 1994. Transposable elements in lepidoptera: hobo-like transposons in Heliothis virescens and Helicoverpa zea. Biochem. Biophys. Res. Commun. 203: 169-175. DOMINGUEZ, A. and J . ALBORNOZ, 1996. Rates of movement of transposable elements in Drosophila melanogaster. Mol. Gen. Genet. 251:130-8. DOWSETT, A. P., 1983. Closely related species of Drosophila can contain different libraries of middle repetitive DNA sequences. Chromosoma 88: 104-108. DOWSETT, A. P. and M. W . YOUNG, 1982. Differing levels of dispersed repetitive DNA among closely related species of Drosophila. Proc. Natl. Acad, of Sci. (USA) 79: 4570-4574. E A N E S , W . J . , C. W E S L E Y , J . H E Y , D . H O U L E and J . W . A J I O K A , 1988. The fitness consequences of P element insertion in Drosophila melanogaster. Genet. Res. 52: 17-26. ENGELS, W . R., 1979a. Germ line aberrations associated with a case of hybrid dysgenesis in Drosophila melanogaster males. Genet. Res. 33: 137-146. ENGELS, W . R., 1979b. Hybrid dysgenesis in Drosophila melanogaster: Rules of inheritance of female sterility. Genet. Res. (Camb.) 33: 219-236. ENGELS, W . R., 1979C. Extrachromosomal control of mutability in Drosophila melanogaster. Proc. Natl. Acad. Sci. (USA) 76: 4011-4015. ENGELS, W . R., 1981a. Germline hypermutability in Drosophila and its relation to hybrid dysgenesis and cytotype. Genetics 98: 565-587. ENGELS, W . R., 1981b. Hybrid dysgenesis in Drosophila and the stochastic loss hypothesis. Cold Spring Harbor Symp. Quant. Biol. 45: 561-565. ENGELS, W . R., 1983. The P family of transposable elements in Drosophila. Ann. Rev. Genet. 17:315-344. ENGELS, W . R., 1984. A trans-acting product needed for P factor transposition in Drosophila. Science 226:1194-1196. ENGELS, W . R., 1989. P elements in Drosophila melanogaster. pp. 437-484. In: Mobile DNA, Edited by D. E . BERG and M. M. HOWE. American Society for Microbiology, Washington, D.C. ENGELS, W . R., 1996. P elements in Drosophila. pp. 103-123. In: Transposable Elements, Edited by H. SAEDLER and A. GlERL.. Springer-Verlag, Berlin. 155 ENGELS, W. R., D. M. JOHNSON-SCHLITZ, W. B. EGGLESTON and J. SVED, 1990. High-frequencyP element loss in Drosophila is homolog dependent. Cell 62: 515-525. ENGELS, W. R. and C. R. PRESTON, 1979. Hybrid dysgenesis in Drosophila melanogaster: The biology of female and male sterility. Genetics 92:161-174. ENGELS, W. R. and C. R. PRESTON, 1981. Identifying P factors in Drosophila by means of chromosome breakage hotspots. Cell 26: 421-428. ENGELS, W. R., C. R. PRESTON and D. M. JOHNSON-SCHLITZ, 1994. Long-range cis preference in DNA homology search over the length of a Drosophila chromosome. Science 263: 1623-1625. FAYET, O., P. RAMOND, P. POLARD, M. F. FRERE and M. CHANDLER, 1990. Functional similarities between retroviruses and the IS3 family of bacterial insertion sequences? Moi. Microbiol. 4:1771-1777. FELDMAR, S. and R. KUNZE, 1991. The ORFa protein, the putative transposase of maize transposable element Ac, has a basic DNA binding domain. EMBO J. 10:4003-4010. FlNNEGAN, D. J., 1989. Eukaryotic transposable elements and genome evolution. Trends Genet. 5: 103-107. FlNNEGAN, D. J., 1990. Transposable elements. Drosoph. Inform. Serv. 68: 371-382. FITZPATRICK, B. and J. A. SVED, 1986. High levels of fitness modifiers induced by hybrid dysgenesis in Drosophila melanogaster. Genet. Res. 48: 89-94. FORMOSA, T. AND D. ALBERTS, 1986. DNA synthesis dependent on DNA recombination: characterization of a reaction characterized by purified bacteriophage T4 enzymes. Cell 47: 793-806. FRASER, M. J., G. E. SMITH and M. D. SUMMERS, 1983. Acquisition of host cell DNA sequences by baculoviruses: relationship between host DNA insertions and FP mutants of Autographa californica and Galleria mellonella nuclear polyhedrosis viruses. J. Virol. 47: 287-300. GALINDO, M. I., V. LADEVEZE, F . L E M E U N I E R , R. K A L M E S , G. PERIQUETet al, 1995. Spread of the autonomous transposable element hobo in the genome of Drosophila melanogaster. Moi. Biol. Evol. 12:723-734. GlNSBURG, L . R., P. M. BINGHAM and S. YOO, 1984. On the theory of speciation induced by transposable elements. Genetics 107:331-341. 156 G L O O R , G. B . , N . A. N A S S I F , D . M . J O H N S O N - S C H L I T Z , C. R . P R E S T O N and W. R . ENGELS, 1991. Targeted gene replacement in Drosophila via P element-induced gap repair. Science 253: 1110-1117. G L O O R , G. B . , C. R . P R E S T O N , D. M. J O H N S O N - S C H L I T Z , N. A. N A S S I F , R . W . P H Y L L I S , W. K. B E N Z , H . M. R O B E R T S O N and W. R . E N G E L S , 1993. Type I repressors of P element mobility. Genetics 135:81-89. GOLDBERG, D. A., 1980. Isolation and partial characterization of the Drosophila alcohol dehydrogenase gene. Proc. Natl. Acad. Sci. USA 77: 5794-5798. GOLDBERG, D. A., J. W. POSAKONY and T . MANIATIS, 1983. Correct developmental expression of a cloned alcohol dehydrogenase gene transduced into the Drosophila germline. Cell 34: 59-73. GOMEZ, S. P. and A. M. HANDLER, 1997. A Drosophila melanogaster hobo-white(+) vector mediates low frequency gene transfer in D. virilis with full interspecific white(+) complementation. Insect Mol. Biol. 6: 165-171. GOOD, A. G. and D. A HICKEY, 1987. Hybrid dysgenesis in Drosophila melanogaster: the elimination of P elements through repeated backcrossing to an M-type strain. Genome 29: 195-200. G O O D , A. G., G. A. M E I S T E R , H . W . B R O C K , T . A . G R I G L I A T T I and D. A. H I C K E Y , 1989. Rapid spread of transposable P elements in experimental populations of Drosophila melanogaster. Genetics 122:387-396. GREEN, M. M., 1988. Mobile DNA elements and spontaneous DNA mutation, pp. 41-50. In: Eukaryotic Transposable Elements as Mutagenic Agents, Edited by M. E . LAMBERT, J. F . MACDONALD and I. B. WEINSTEIN. Cold Spring Harbor Labotratory Press, Cold Spring Harbor. GREENBLATT, I. M., 1974. Movement of Mudulator in maize: A test of an hypothesis. Genetics 77: 671-678. GREENBLATT, I. M., 1984. A chromosomal replication pattern deduced from pericarp phenotypes resulting from movements of the transposable element, Modulator, in maize. Genetics 108:471-485. GREENBLATT, I. M. and R . A. BRINK, 1962. Twin mutations in medium variegated pericarp maize. Genetics 47: 489-501. GRELL, E . H., K. B. JACOBSON and J. B. MURPHY, 1968. Alterations of the genetic material for analysis of alcohol dehydrogenase isozymes of Drosophila melanogaster. Ann. N . Y . Acad. Sci. 151:441-455. GULUBOVSKI, M. D., Y . N. IVANOVand M. M. GREEN, 1977. Genetic instability in Drosophila melanogaster: Putative multiple insertion mutants at the singed bristle locus. Proc. Natl. Acad. Sci. (USA) 74: 2973-2975. 157 HALL, H. G., 1986. DNA differences found between Africanized and European honeybees. Proc. Natl. Acad. Sci. USA 83: 4874. HANDLER, A. M. and S. P. GOMEZ, 1995. The hobo transposable element has transposase-dependent and - independent excision activity in drosophilid species. Moi. Gen. Genet. 247: 399-408. HANDLER, A. M., S. P. GOMEZ and D. A. O'BROCHTA, 1993. A functional analysis of the P-element gene-transfer vector in insects. Arch. Insect Biochem. Physiol. 22: 373-84. HARADA, K., K. YUKUHIRO and T. MUKAI, 1990. Transposition rates of movable genetic elements in Drosophila melanogaster. Proc. Natl. Acad. Sci. USA 87: 3248-3252. HARING, M. A , J. GAO, T. VOLBEDA, C. M. ROMMENS, H. J. NlJKAMPe* al., 1989. A comparative study of Tam3 and Ac transposition in transgenic tobacco and petunia plants. Plant Moi. Biol. 13:189-201. HARRIS, L. J., D. L. BAILLIE, and A. M. ROSE, 1988. Sequence identity between an inverted repeat family of transposable elements in Drosophila and Caenorhabditis. Nucleic Acids Res. 16:5991-5998. HATZOPOULOS, P., M. MONASTIRIOTI, G. YANNOPOULOS and C. LOUIS, 1987. The instability of the TE-like mutation Dp (2;2)GYL of Drosophila melanogaster is intimately associated with the hobo element. EMBO J. 6: 3091-3096. HEHL, R., 1994. Transposon tagging in heterologous host plants. Trends Genet. 10:385-386. HlCKEY, D. A., 1982. Selfish DNA: a sexually-transmitted nuclear parasite. Genetics 101: 519-531. HlCKEY, D. A , 1992. Evolutionary dynamics of transposable elements in prokaryotes and eukaryotes. Genetica 86: 269-274. HIGUET, D., D. ANXOLABEHERE and D. NOUAUD, 1992. A particular P-element insertion is correlated to the P-induced hybrid dysgenesis repression in Drosophila melanogaster. Genetical Research 60:15-24. HIRAIZUMI, Y., 1971. Spontaneous recombination in Drosophila melanogaster males. Proc. Natl. Acad. Sci. (USA) 68: 268-270. HO, Y. T., S. M. WEBER and J. K. LlM, 1993. Interacting hobo transposons in an inbred strain and interaction regulation in hybrids of Drosophila melanogaster. Genetics 134:895-908. HOOGLAND, C. and C. BlEMONT, 1996. Chromosomal distribution of transposable elements in Drosophila melanogaster: test of the ectopic 158 recombination model for maintenance of insertion site number. Genetics 144: 197-204. HOUCK, M. A., J. B. CLARK, H. R . PETERSON and M. G. KlDWELL, 1991. Possible horizontal transfer of Drosophila genes by the mite Proctolaelaps regalis. Science 253: 1125-1129. JACKSON, M. S., D. M. BLACK and G. A. DOVER, 1988. Amplification of KP elements associated with the repression of hybrid dysgenesis in Drosophila melanogaster [published erratum appears in Genetics 1989 Mar;121(3):629]. Genetics 120:1003-1013. J A S I N S K I E N E , N . , C. J . C O A T E S , M. Q. B E N E D I C T , A. J. C O R N E L , C. S. R A F F E R T Y , A. A. JAMES and F. H. COLLINS. 1998. Stable transformation of the yellow fever mosquito, Aedes aegypti, with the Hermes element from the housefly. Proc. Natl. Acad. Sci. USA 95: 3743-3747. JOHNSON-SCHLITZ, D. M. and W. R. ENGELS, 1993. P-element-induced interallelic gene conversion of insertions and deletions in Drosophila melanogaster. Mol. Cell. Biol. 13:7006-7018. JOHNSON-SCHLITZ, D. and J. K. LlM, 1987. Cytogenetics of Notch mutations arising in the unstable X chromosome Uc of Drosophila melanogaster. Genetics 115:701-709. KAISER, K., J. W. SENTRY and D. J. FINNEGAN, 1995. Eukaryotic transposable elements as tools to study gene structure and function, pp. 69-100. In: Mobile Genetic Elements, Edited by D. J. SHERRAT. IRL Press, Oxford. KAPLAN, N., T. DARDEN and C. H. LANGLEY, 1985. Evolution and extinction of transposable elements in Mendelian populations. Genetics 109:459-480. KARESS, R. E. and G. M. RUBIN, 1984. Analysis of P element functions in Drosophila. Cell 38: 135-146. KARPEN, G. H. and A. C. SPRADLING, 1992. Analysis of subtelomeric heterochromatin in the Drosophila minichromosome Dp 1187 by single P element insertional mutagenesis. Genetics 132:737-753. KAUFMAN, P. D., R. F. DOLL and D. C. RlO, 1989. Drosophila P element transposase recognizes internal P element sequences. Cell 59: 359-371. KAUFMAN, P. D. and D.C. RlO, 1992. P element transposition in vitro proceeds by a cut-and-paste mechanism and uses GTP as a cofactor. Cell 69:27-39. KEELER, K. J., T. DRAY, J. E. PENNY and G. B. GLOOR, 1996. Gene targeting of a plasmid-borne sequence to a double-strand DNA break in Drosophila melanogaster. Mol. Cell. Biol. 16:522-528. 159 KEELER, K. J. and G. B. GLOOR, 1997. Efficient gap repair inDrosophila melanogaster requires a maximum of 31 nucleotides of homologous sequence at the searching ends. Mol. Cell. Biol. 17:627-634. KHAN, E., J. P. G. MACK, R. A. KATF, J. KULKOSKYand A. M. SKALKA, 1991. Retroviral integrase domains: DNA binding and recognition of LTR sequences. Nucleic Acids Res. 19:851-860. KHILLAN, J. S., P. A. OVERBEEK and H. WESTPHAL, 1985. Drosophila P element integration in the mouse. Dev. Biol. 109:247-250. KlDWELL, M. G., 1983. Evolution of hybrid dysgenesis determinants in Drosophila melanogaster. Proc. Natl. Acad. Sci. USA 80: 1655-1659. KlDWELL, M. G., 1985. Hybrid dysgenesis in Drosophila melanogaster: Nature and inheritance of P element regulation. Genetics 111:337-350. KlDWELL, M. G., 1992. Horizontal transfer of P elements and other short inverted repeat transposons. Genetica 86: 275-286. KlDWELL, M. G , T. FRYDRYK and J. NOVY, 1983. The hybrid dysgenesis potential of Drosophila melanogaster strains of diverse temporal and geographic origin. Drosophila Inform. Serv. 59: 63-69. KlDWELL, M. G , J. F. KlDWELL and J. A. SVED, 1977. Hybrid dysgenesis in Drosophila melanogaster: A syndrome of aberrant traits including mutation, sterility and male recombination. Genetics 86: 813-833. KlDWELL, M. G. and J. B. NOVY, 1979. Hybrid dysgenesis in Drosophila melanogaster. sterility resulting from gonadal dysgenesis in the P-M system. Genetics 92: 1127-1140. KlDWELL, M. G. and J. B. NOVY, 1985. The distribution of hybrid dysgenesis determinants in North American populations of D. melanogaster. Drosophila Inform. Serv. 61:97-100. KlDWELL, M. G., J. B. NOVY and S. M. FEELEY, 1981. Rapid unidirectional change of hybrid dysgenesis potential in Drosophila. J. Hered. 72: 32-8. KlDWELL, M. G. and J. M. C. RIBEIRO, 1992. Can transposable elements be used to drive disease refractoriness genes into vector populations? Parasitol. Today 8: 325-329. KlYASU, P. K. and M. G. KlDWELL, 1984. Hybrid dysgenesis inDrosophila melanogaster: the evolution of mixed P and M populations maintained at high temperature. Genet. Res. 44: 251-259. KLECKNER, N., 1990. Regulation of transposition in bacteria. Annu. Rev. Cell Biol. 6:297-327. 160 K N A P P , S., G. C O U P L A N D , H . U H R I G , P. S T A R L I N G E R and F. S A L A M I N I , 1988. Transposition of the maize transposable element Ac in Solanum tuberosum. Moi. Gen. Genet. 213: 285-290. K U N Z E , R and P. S T A R L I N G E R , 1989. The putative transposase of transposable element Ac from Zea mays interacts with subterminal sequences of Ac. EMBO J . 8:3177-3185. L A D E V E Z E , V., I. G A L I N D O , N. C H A M I N A D E , L. P A S C U A L , G. P E R I Q U E T e t al., 1998. Transmission pattern of hobo transposable element in transgenic lines of Drosophila melanogaster. Genet. Res. 71:97-107. L A D E V E Z E , V., M. I. G A L I N D O , L. P A S C U A L , G. P E R I Q U E T and F. L E M E U N I E R , 1994. Invasion of the hobo transposable element studied by in situ hybridization on polytene chromosomes of Drosophila melanogaster. Genetica 93: 91-100. L A N G L E Y , C. H . , J . F. Y . B R O O K F I E L D and N. K A P L A N , 1983. Transposable elements in Mendelian populations. I. A theory. Genetics 104:457-471. L A N G L E Y , C. H . , E. M O N T G O M E R Y , R. H U D S O N , N. K A P L A N and B. C H A R L E S W O R T H , 1988. On the role of unequal exchange in the containment of transposable element copy number. Genet. Res. 52: 223-235. L A N S M A N , R. A., S. N. S T A C E Y , T. A. G R I G L I A T T I and H . W. B R O C K , 1985. Sequences homologous to the P mobile element of Drosophila melanogaster are widely distributed in the subgenus Sophophora. Nature 318: 561-563. L A S K I , F. A., D. C. R I O and G. M. R U B I N , 1986. Tissue specificity of Drosophila P element transposition is regulated at the level of mRNA splicing. Cell 44: 7-19. LEE, C. C , M. M. MUL and D. C. RlO, 1996. The Drosophila P-element KP repressor protein dimerises and interacts with multiple sites on P element DNA. Molecular and Cellular Biology 16:5616-5622. L E M A I T R E , B. and D. C O E N , 1991. P regulatory products repress in vitro the P promoter activity in P-lacZ fusion genes. Proc. Natl. Acad. Sci. USA 88: 4419-4423. L E M A I T R E , B., S. R O N S S E R A Y and D. C O E N , 1993. P cytotype repression of the P element promoter in the germ-line of Drosophila melanogaster: a model for the P cytotype. Genetics 135:149-160. LIM, J . K . , 1988. Intrachromosomal rearrangements mediated by hobo transposons in Drosophila melanogaster. Proc. Natl. Acad. Sci. USA 85: 9153-9157. LIM, J . K . and M. J . S I M M O N S , 1994. Gross chromosome rearrangements mediated by transposable elements. BioEssays 16: 169-275. 161 LISITSYN, N., N. LISITSYN and M. WlGLER, 1993. Cloning the differences between two complex genomes. Science 259: 946-951. LOUKERIS, T. G . , I. LIVADARAS, B. ARCA, S. ZABALOU and C. SAVAKIS, 1996. Gene transfer into the Medfly, Ceratitis capitata, with a Drosophila hydei transposable element. Science 270: 2002-2007. LOZOVSKAYA, E. R., D. I. NURMINSKY, D. L. HARTL and D. T. SULLIVAN, 1996. Germline transformation of Drosophila virilis mediated by the transposable element hobo. Genetics 142:173-177. MACKAY, T. F. C , 1986. Transposable element-induced fitness mutations in Drosophila melanogaster. Genet. Res. 48: 77-87. MACKAY, T. F., 1989. Transposable elements and fitness in Drosophila melanogaster. Genome 31:284-295. MANIATIS, T., E. F. FRITSCH and J . SAMBROOK, 1982. Molecular cloning: A laboratory manual. Cold Spring Harbor Laboratory. Cold Spring Harbor, N.Y. MANNING, J . E., C. W. SCHMID and N. DAVIDSON, 1975. Interspersion of repetitive and nonrepetitive DNA sequences in the Drosophila melanogaster genome. Cell 4: 141-155. MCCLINTOCK, B., 1951. Chromosome organization and genetic expression. Cold Spring Harbor Symp. Quant. Biol. 16:13-47. MCDONALD, J . F., 1995. Transposable elements: possible catalysts of organismic evolution. Trands Ecol. Evol. 10: 123-126. MCGINNIS, W., A. W. SHERMOEN and S. K. BECKENDORF, 1983. A transposable element inserted just 5' to a Drosophila glue protein gene alters gene expression and chromatin structure. Cell 34: 75-84. M C G R A N E , V . , J . O . C A R L S O N , B . R. M I L L E R and B. J . B E A T Y , 1988. Microinjection of DNA into Aedes triseriatus ova and detection of integration. Am. J . Trop. Med. Hyg. 39: 502-510. MEISTER, G. A , 1992. Spread of P elements in Drosophila melanogaster. M.Sc. Thesis, University of British Columbia, Vancouver. MEISTER, G. A. and T. A. GRIGLIATTI, 1993. Rapid spread of a P elementlAdh gene construct through experimental populations of Drosophila melanogaster. Genome 36: 1169-1175. MEISTER, G. A , R. A. LANSMAN and T. A. GRIGLIATTI, 1995. Simple plaque hybridization method for the detection of differentially represented repetitive DNA. Biotechniques 18:250-255. 162 M I L L E R , L . F L , R. K . S A K A I , P . R O M A N S , R . W. G W A D Z , P. K A N T O F F and H .G. COON. 1987. Stable integration of a bacterial gene in the mosquito Anopheles gambiae. Science 237 : 779-781. MlSRA, S. and D. C . RlO, 1990. Cytotype control of Drosophila P element transposition: The 66 kd protein is a repressor of transposase activity. Cell 6 2 : 269-284. MITCHELSON, A., M . SlMONELIG, C . WILLIAMS and K . O'HARE, 1993. Homology with Saccharomyces cerevisiae RNA14 suggests that phenotypic suppression in Drosophila melanogaster by suppressor of forked occurs at the level of RNA stability. Genes Dev. 7: 241-249. MONTGOMERY, E . , B. CHARLESWORTH and C . H. LANGLEY, 1987. A test for the role of natural selection in the stabilization of transposable element copy number in a population of Drosophila melanogaster. Genet. Res. 49 : 31-41. MONTGOMERY, E . A , and C . H. LANGLEY, 1983. Transposable elements in Mendelian populations. II. Distribution of three copia-like elements in a natural population of Drosophila melanogaster. Genetics 104:473-483. MORRIS, A. C , P. EGGELSTON and J. M . C R A M P T O N , 1989. Genetic transformation of the mosquito Aedes aegypti by micro-injection of DNA. Med. Vet. Entomol. 3 . MULLINS, M . C , D. C. RIO and G. M . RUBIN, 1989. Cis-acting DNA sequence requirements for P element transposition. Genes Dev. 3 : 729-738. N A N J U N D I A H , V., 1985. Transposable element copy number and stable polymorphisms. J. Genet. 64 : 127-134. N A S S I F , N., J. PENNEY, S. PAL, W. R. E N G E L S and G. B. GLOOR, 1994. Efficient copying of nonhomologous sequences from ectopic sites via P- element-induced gap repair. Mol. Cell. Biol. 14:1613-1625. NUZHDIN, S. V. and T. F. MACKAY, 1995. The genomic rate of transposable element movement in Drosophila melanogaster [letter]. Mol. Biol. Evol. 12:180-181. O'BROCHTA, D. A. and P. W. ATKINSON, 1996. Transposable elements and gene transformation in non-drosophilid insects. Insect Biochem. Mol. Biol. 26 : 739-53. O'BROCHTA, D. A , S. P. GOMEZ and A. M . HANDLER, 1991. P element excision in Drosophila melanogaster and related drosophilids. Mol. Gen. Genet. 225 : 387-394. O'BROCHTA, D. A., and A. M . HANDLER, 1988. Mobility of P elements in drophilids and non-drosophiids. Proc. Natl. Acad. Sci. 8 5 : 6052-6056. 163 O'BROCHTA, D. A., W. D. WARREN, K. J. SAVILLE and P. W. ATKINSON, 1994. Interplasmid transposition of Drosophila hobo elements in non-drosophilid insects. Moi. Gen. Genet. 244 : 9-14. O'BROCHTA, D. A , W. D. WARREN, K. J. SAVILLE and P. W. ATKINSON, 1996. Hermes, a functional non-Drosophilid insect gene vector from Musca domestica. Genetics 142:907-914. O'DONNELL, J., L. GERACE, F. LEISTERand W. SOFER, 1975. Chemical selection of mutants that affect alcohol dehydrogenase in Drosophila. II. Use of 1-pentyne-3-ol. Genetics 79: 73-83. O'HARE, K. and G. M. RUBIN, 1983. Structures of P transposable elements and their sites of insertion and excision in the Drosophila melanogaster genome. Cell 34 : 25-35. O'KANE, C. J. and W. J. GEHRING, 1987. Detection in situ of genomic regulatory elements in Drosophila. Proc. Natl. Acad. Sci. USA 84: 9123-9127. ORGEL, L. E. and F. H. CRICK, 1980. Selfish DNA: the ultimate parasite. Nature 284 : 604-607. PASCUAL, L. and G. PERIQUET, 1991. Distribution of hobo transposable elements in natural populations of Drosophila melanogaster. Moi. Biol. Evol. 8: 282-296. PERIQUET, G., M. H. HAMELIN, Y. BIGOT and K. Hu, 1989a. Presence of the deleted hobo element Th in Eurasian populations of Drosophila melanogaster. Genet. Sel. Evol. 21:107-111. PERIQUET, G., M. H. HAMELIN, Y. BlGOTand A. LEPISSIER, 1989b. Geographical and historical patterns of distribution of hobo elements in Drosophila melanogaster populations. J. Evol. Biol. 2 : 223-229. PERIQUET, G., M. H. HAMELIN, R. KALMES and J. EEKEN, 1990. hobo elements and their deletion-derivative sequences in D. melanogaster and its sibling species D. simulans, D. mauritiana and D. sechellia. Genet. Sel. Evol. 2 2 : 393-402. PERIQUET, G., F. LEMEUNIER, Y. BIGOT, M. H. HAMELIN, C. BAZIN et al, 1994. The evolutionary genetics of the hobo transposable element in the Drosophila melanogaster complex. Genetica 9 3 : 79-90. PERKINS H. D. and A. J. HOWELL, 1992. Genomic sequences with homology to the P element of Drosophila melanogaster occur in the blowfly Lucilia cuprina. Proc. Natl. Acad. Sci. USA 89 : 1753-1757. PETERSON, P. W. and J. I. YODER, 1995. Amplification of Ac in tomato is correlated with high Ac transposition activity. Genome 38 : 265-276. 164 PFEIFER, T. A. and T. A. GRIGLIATTI, 1996. Future perspectives on insect pest management: engineering the pest. J. Invertebr. Pathol. 67 : 109-119. PIMPINELLI, S., M. BERLOCO, L. FANTI, P. DlMITRI, S. BONACCORSI et al., 1995. Transposable elements are stable structural components oi Drosophila melanogaster heterochromatin. Proc. Natl. Acad. Sci. USA 9 2 : 3804-3208. PINKERTON, A. C , D. A. O'BROCHTA and P. W. ATKINSON, 1996. Mobility of hAT transposable elements in the Old World bollworm, Helicoverpa armigera. Insect Mol. Biol. 5: 223-227. PLASTERK, R. H., 1993. Molecular mechanisms of transposition and its control. Cell 74: 781-786. PRESTON, C. R. and W. R. ENGELS, 1989. Spread of P elements in inbred lines of Drosophila melanogaster. Prog. Nucleic Acid Res. Mol. Biol. 3 6 : 71-85. PRESTON, C. R., J. A. SVED and W. R. ENGELS, 1996. Flanking duplications and deletions associated with P-induced male recombination in Drosophila. Genetics 144:1623-1638. PROMEGA. 1988. LambdaGEM™ll Xho I Half-Site Arms Cloning System. Promega, Inc. Brochure TB057, December 1988. Promega, Inc., 2800 Woods Hollow Road, Madison, WI 53711-5399. RASMUSSON, K. E., J. D. RAYMOND andM. J. SIMMONS, 1993. Repression of hybrid dysgenesis in Drosophila melanogaster by individual naturally-occurring P elements. Genetics 133:605-622. RING, M., T. A. PFEIFER and T. A. GRIGLIATTI, 1996. Isolation of a non-LTR-retrotransposon from the variegated cutworm, Peridroma saucia, using PCR. Insect Biochem. and Molec. Biol. 26 : 511-518. RIO, D. C , F. A. LASKI and G. M. RUBIN, 1986. Identification and immunochemical analysis of biologically active Drosophila P element transposase. Cell 44 : 21-32. RIO, D. C. and G. M. RUBIN, 1988. Identification and purification of a Drosophila protein that binds to the terminal 31-base-pair inverted repeats of the P transposable element. Proc. Natl. Acad. Sci. (USA) 8 5 : 8929-8933. ROBERTS, D., B. C. HOOPES, W. R. MCCLURE and N. KLECKNER, 1985. IS10 transposition is regulated by DNA adenine methylation. Cell 43 : 117-130. ROBERTSON, H. M., 1993. The mariner transposable element is widespread in insects. Nature (London) 3 6 2 : 241-245. ROBERTSON, H. M. and W. R. ENGELS, 1989. Modified P elements that mimic the P cytotype in Drosophila melanogaster. Genetics 123:815-824. 165 R O B E R T S O N , H. M. and E. G. MCLEOD, 1993. Five major subfamilies of mariner transposable elements in insects, including the Mediterranean fruit fly, and related arthropods. Insect Molecular Biology 2: 125-139. R O B E R T S O N , H. M., C. R. P R E S T O N , R. W. PHILLIS, D. J O H N S O N - S C H L I T Z , W. K. BENZ and W. R. ENGELS, 1988. A stable genomic source of P element transposase in Drosophila melanogaster. Genetics 118:461-470. ROCHE, S., M. SCHIFF and D. C. RlO, 1995. P element repressor autoregulation involves germline transcriptional repression and reduction of third intron splicing. Genes and Development 9: 1278-1288. ROMMENS, C. M., M. J . V A N HAAREN, H. J . NlJKAMP and J . HlLLE, 1993. Differential repair of excision gaps generated by transposable elements of the 'Ac family'. Bioessays 15:507-512. R O N S S E R A Y S . , M. LEHMANN and D. A N X O L A B E H E R E , 1991. The maternally inherited regulation of P elements in Drosophila melanogaster can be elicited by two P copies at cytological site 1A on the X chromosome. Genetics 129:501-512. RUBIN, G. M., M. G. KlDWELL and P. M. B I N G H A M , 1982. The molecular basis of P-M hybrid dysgenesis: The nature of induced mutations. Cell 29: 982-994. RUBIN, E. and A. A. LEVY, 1997. Abortive gap repair: underlying mechanism for Ds element formation. Mol. Cell. Biol. 17:6294-6302. RUBIN, G. M. and A. C. SPRADLING, 1982. Genetic transformation of Drosophila with transposable element vectors. Science 218:348-353. RUBIN, G. M. and A. C. S P R A D L I N G , 1983. Vectors for P element-mediated gene transfer in Drosophila. Nucleic Acids Res. 11: 6341-6351. SAEDLER, H. and A GlERL, 1989. Transposable Elements. Berlin: Springer-Verlag. SAEDLER, H. and P. NEVERS, 1985. Transposition in plants: a molecular model. EMBO J . 4: 585-590. S A R K A R , A., C. J . COATES, S. WHYARD, U. W I L L H O E F T , P. W. A T K I N S O N and D. A. O ' B R O C H T A , 1997. The Hermes element from Musca domestica can transpose in four families of cyclorrhaphan flies. Genetica 99: 15-29. SCHACHAT, F. H., and D. S. HOGNESS, 1974. Repetitive sequences in isolated Thomas circles from Drosophila melanogaster. Cold Spring Harbor Symp. of Quant. Biol. 38: 371-381. SEARLES, L. L., R. S. J O K E R S T , P. M. B I N G H A M , R. A. VOELKER and A. L. GREEN LEAF, 1982. Molecular cloning of sequences from a Drosophila RNA polymerize II locus by P element transposon tagging. Cell 31:585-592. 166 S H E E N , F., J. K. LIM and M. J. S I M M O N S , 1993. Genetic instability in Drosophila by hobo transposable elements. Genetics 133: 315-334. S H E R R A T T , D. J. 1995. Mobile Genetic Elements. Oxford: IRL Press. S I E B E L , C. W. and D. C. R I O , 1990. Regulated splicing of the Drosophila P transposable element third intron in vitro: Somatic repression. Science 248: 1200-1208. SlMONELIG, M., and D. A N X O L A B E H E R E . 1991. A P element of Scaptomyza pallida is active in Drosophila melanogaster. Proc. Natl. Acad. Sci. (USA) 88: 6102-6106. S I M M O N S , G. M., 1992. Horizontal transfer of hobo transposable elements within the Drosophila melanogaster species complex: evidence from DNA sequencing. Moi. Biol. Evol. 9: 1050-1060. S I M M O N S , M. J. and L. M. BUCHOLZ., 1985. Transposase titration in Drosophila melanogaster: A model of cytotype in the P-M system of hybrid dysgenesis. Proc. Natl. Acad. Sci. (USA) 82: 8119-8123. S K I N N E R , D.Z. 1992. A PCR-based method of identifying species-specific DNAs. BioTechniques 13: 210-213. S M I T H , D., J., B. R. C A L V I , I. F R A N K L I N and W. M. G E L B A R T , 1993. hobo enhancer trapping mutagenesis in Drosophila reveals an insertion specificity different from P elements. Genetics 135:1063-1076. S M I T H , P. A. and V. G. C O R C E S , 1991. Drosophila transposable elements: mechanisms of mutagenesis and interactions with the host genome. Adv. Genet. 29: 229-300. S P R A D L I N G , A.C., 1986. P element-mediated transformation, pp. 175-198. In: Drosophila: A practical approach, Edited by D. B. Roberts. IRL Press, Oxford. S P R A D L I N G , A. C. and G. M. R U B I N , 1981. Drosophila genome organization: Conserved and dynamic aspects. Ann. Rev. of Genet. 15:219-264. SPRADLING, A. C. and G. M. RUBIN, 1982. Transposition of cloned P elements into Drosophila germ line chromosomes. Science 218:341-347. S P R A D L I N G , A. C . and G. M. R U B I N , 1983. The effect of chromosomal position on the expression of the Drosophila Xanthine dehydrogenase gene. Cell 34: 47-57. S T A M A T I S , N., M. M O N A S T I R I O T I , G. Y A N N O P O U L O S and C. L O U I S , 1989. The P-M and 2.3MRF (hobo) systems of hybrid dysgenesis in Drosophila melanogaster are independent of each other. Genetics 123: 379-387. 167 STAVELEY, B. E., T. R. HESLIP, R. B. HODGETTS and J. B. BELL, 1995. Protected P-element termini suggest a role for inverted-repeat-binding protein in transposase-induced gap repair in Drosophila melanogaster. Genetics 139:1321-1329. STRATAGENE. 1990. Gigapack™II Plus Packaging Extract. Stratagene protocol, Catalogue # 200211, March 1990. Stratagene, 11099 North Torrey Pines Road, La Jolla, CA 92037. STRECK, R. D., J. E. MACGAFFEYand S. K. BECKENDORF, 1986. The structure of hobo transposable elements and their insertion sites. EMBO J. 5: 3615-3623. TAKADA, S., M. MURAI, E. TAKANASHI, K. OHISHI, K. FUKAMI, N. HAGIWARA, Y. SATTA and S. ISHIWA, 1983. On the P-M system in natural populations ofD. melanogaster in and around Japan. Jap. J. of Gen. 58: 686. TAKASU-ISHIKAWA, E., M. YOSHIHARA and Y. HOTTA, 1992. Extra sequences found at P element excision sites in Drosophila melanogaster. Moi. Gen. Genet. 232: 17-23. VAN SLUYS, M. A , J. TEMPE and N. FEDOROFF, 1987. Studies on the introduction and mobility of the maize Activator element in Arabidopsis thaliana and Daucus carota. EMBO J. 6: 3881-3889. VERESHCHAQINA, N. M., I. K. ILIADI and O. A. NlKITICH, 1994. Comparative studies of P- and /io6o-element activity, fitness components and recombination parameters in two natural populations of Drosophila melanogaster in Moldova. Hereditas 120:91-98. VIGUE, C. and W. SOFER, 1974. Adh115: A temperature-sensitive mutant at the Adh locus in Drosophila. Biochem. Genet. 11: 387- 396. VIGUE, C. and W. SOFER, 1976. Chemical selection of mutants that affect ADH activity in Drosophila. III. Effects of ethanol. Biochem. Genet. 14:127-135. WALKER, V. K., 1989. Gene transfer in insects. Adv. Cell Culture 7: 87-124. WANG, L., M. HEINLEIN and R. KUNZE, 1996. Methylation pattern of Activator transposase binding sites in maize endosperm. Plant Cell 8: 747-758. WANG, L. and R. KUNZE, 1998. Transposase binding site methylation in the epigenetically inactivated Ac derivative Ds-cy. Plant J. 13:577-582. WARREN, W. D., P. W. ATKINSON and D. A. O'BROCHTA, 1994. The Hermes transposable element from the housfly, Musca domestica, is a short inverted repeat-type element of the hobo, Ac and Tam3 hAT element family. Genetical Res. Camb. 64: 87-97. 168 WARREN, W. D., P. W. ATKINSON and D. A. O'BROCHTA, 1995. The Austrahan bushfly Musca vetustissima contains a sequence related to transposons of the hobo, Ac and Tam3 family. Gene 154:133-134. WELLS, R., M. D. ROYER and C. P. HOLLENBERG, 1976. Non Xenopus-like DNA sequence organization in the Chironomous tentans genome. Molec. Gen. Genet. 147: 45-51. WlCHMAN, H. A., S. S. POTTER and D. PINE, 1985. Mys, a family of mammalian transposable elements isolated by phylogenetic screening. Nature 317: 77-81. WlCHMAN, H.A. and R.A. VAN DEN BUSSCHE, 1992. In search of retrotransposons: exploring the potential of the PCR. BioTechniques 13: 258-264. XIONG, Y. and T. H. ElCKBUSH, 1990. Origin and evolution of retroelements based on their reverse transcriptase sequences. EMBO J. 9: 3353-3362. YAMAMOTO, A., F. HIHARA, and T. K. WATANABE, 1984. Hybrid dysgenesis in Drosophila melanogaster: Pre-dominance of Q factor in Japanese populations and its change in the laboratory. Genetica 63: 71-77. YANNOPOULOS, G., N. STAMATIS, M. MONASTIRIOTI, P. HATZOPOULOS and C. LOUIS, 1987. hobo is responsible for the induction of hybrid dysgenesis by strains of Drosophila melanogaster bearing the male recombination factor 23.5MRF. Cell 49: 487-495. YANNOPOULOS, G., S. ZABALOU and S. N. ALAHIOTIS, 1994a. Distribution of P and hobo mobile elements in environmentally manipulated long-term Drosophila melanogaster cage populations. Hereditas 121: 87-102. YANNOPOULOS, G., S. ZABALOU, N. STAMATIS and G. TSAMATHIS, 1994b. Differential regulation of P and hobo elements by two laboratory strains in Drosophila melanogaster. Genet. Res. 63: 129-137. YODER, J. I., J. PALYS, K. ALPERT andM. LASSNER, 1988. Ac transposition in transgenic tomato plants. Mol. Gen Genet. 213: 291-296. YOUNG, M. W. 1979. Middle repetitive DNA: A fluid component of the Drosophila genome. Proc. Natl. Acad, of Sci. (USA) 76: 6274-6278. YUKUHIRO, K., K. HARADA and T. MUKAI, 1985. Viability mutations induced by the P elements inDrosophila melanogaster. Jpn. J. Genet. 60: 531-537. ZABALOU, S., S. N. ALAHIOTIS and G. YANNOPOULOS, 1994. A three-season comparative analysis of the chromosomal distribution of P and hobo mobile elements in a natural population of Drosophila melanogaster. Hereditas 120: 127-140. 169 APPENDIX 1 The purpose of this appendix is to show in detail the steps involved in the quantitation of one dot blot, starting with the phosphor images as described in the METHODS of CHAPTER III. A previously stated, the overall analysis for each dot blot membrane consisted of four primary steps: 1) to measure the volume and subtract the background for each dot on the digital image produced by the single copy probe; 2) to use these volumes to perform a linear regression analysis to determine the amount of DNA in each dot; 3) to measure the volume and subtract the background for all dots produced by the hobo probe; and 4) to compare the hybridization of the hobo probe to experimental samples and to a gradient of DNA from the original hobo donating strain. The blot that I will use as an example is the one shown in Figure 10. For simplicity, I will show the calculation of the amount of OR strain hobo DNA present in the 1% OR-A population only. There are several ways that the dots can be measured. After experimenting with several of these (see Appendix 2), I determined that the most accurate and reproducible protocol was to draw an ellipse around each dot and then have ImageQuaNT measure the volume (total linear intensities of all the pixels in the area) of each dot. To keep the ellipse areas constant, and to speed up the repetitive process of drawing an 8 X 12 grid of ellipses, a template of 96 ellipses was made which could be overlain on each of the images (Figure 23A). This complete template was aligned as well as possible on each image and then, if necessary, individual ellipses were re-centered on the appropriate dot (Figure 23B). Once ImageQuaNT had measured the volumes of each dot, the data were imported into an Excel document (Microsoft Software) for further analysis. 170 o o K - 0 1 A - 0 1 o o K - 0 2 B - 0 1 o o K-03C - 0 1 o o K-04D-01 o o E-01E-02 o o F - 0 1 F - 0 2 o o K-14G -01 o o K-15H-01 O O O O A - 0 2 A - 0 3 A - 0 4 A - 0 5 O O O O B - 0 2 B - 0 3 B - 0 4 B - 0 5 o 0 0 0 C - 0 2 C - 0 3 C - 0 4 C - 0 5 O O O O D - 0 2 D - 0 3 D - 0 4 D - 0 5 o o o o E-03 E-04 E-05 E - 0 6 o o o o F - 0 3 F -04 F - 0 5 F - 0 6 o o o o G - 0 2 G - 0 3 G - 0 4 G - 0 5 O O O O H - 0 2 H - 0 3 H - 0 4 H - 0 5 O O O O A - 0 6 A - 0 7 A - 0 8 A - 0 9 O O O o B - 0 6 B - 0 7 B - 0 8 K - 0 7 O O O O C - 0 6 C - 0 7 C - 0 8 C - 0 9 O O O O D - 0 6 D - 0 7 D - 0 8 D - 0 9 o o o o E-07 E-08 E-09 E - 1 0 o o o o F - 0 7 F - 0 8 F - 0 9 F - 1 0 O O O O G - 0 6 G - 0 7 G - 0 8 G - 0 9 O O O O H - 0 6 H - 0 7 H - 0 8 H - 0 9 0 a K-05K-06 O O K-08K-09 O O K-10K-11 0 a K-12K-13 O O E-11 E-12 O O F-11 F-12 O O G - 1 0 K - 1 6 o o H-10K-17 B o ® ® © i i i i t l o o K-01 A-01 A - 0 2 A - d 3 A - 0 4 A - b 5 A - b 6 A - 0 7 A - 6 8 A - 0 9 K - 0 5 K - 0 6 o # # # • # $ # # o 0 0 K-02 B-61 B-02 Bit)3 B-04 B-05 B-06 B-07 B-08 K-07K-08K-09 0 0 B m © # © III O O K-03C-01 C - 0 2 C - 0 3 C - 0 4 C - 0 5 0-06 0*07 C-08C-09 K-10K-11 o #. m m m m m W f t n o a K-04D-01 D-02D-03 D-04 D-05 D-06 D-67 D-08 D-09 K-12K-13 o a o o o o © o m # m # E-01E-02 E-03 E-04 E-05 E-06 E-07 E-08 E-09 E-10 E-11 E-12 o o o o m m m m m mmm F-01 F-02 F-03 F-04 F-05F-06 F-07 F-08 F-09 F-^0 F-T1 F^f2 o o m o ® ® ® w ® & m o K-14G-01 G-02G-03 G - 0 4 G - 0 5 G - 0 6 G-07G-08 G-09G-10 K-16 o o o o o # o K-15H-  H- 2 H- H-04 H  H-0 H-07 H-08H-09 H-'l 0 7 Figure 23: Volume measurement using ellipses. Panel A shows a template of 96 ellipses that was used to direct ImageQuaNT to the correct areas upon which to perform volume quantitation. Panel B shows this template overlaid on the image of a dot blot probed with labeled hobo element fragments. See column 2 of Table VII (and also Figure 10) for a description of the DNA samples. 171 STEP 1; The first column of Table VII lists the 'names' of the ellipses on the template and the second column shows the sample DNA that was loaded into the corresponding well of the vacuum manifold. The third column shows the volume of each dot for the single copy probe. For this particular image, the mean volume of all dots which contained only TE was 1236 (see values in bold in column 3). Column 4 shows the volumes of the dots on the image produced with the single copy probe, once they had been corrected for background by subtracting 1236 from each of the values in column 3. This completes step 1. STEP 2: The next step was to calculate the micrograms of DNA present in each experimental dot by linear regression analysis using the data analysis capabilities of Excel. To accomplish this, the volumes of the CS gradient dots on the single copy image (column 4) were plotted against the micrograms of CS DNA loaded in each dot (see column 2). At first, all possible data points were plotted. However, in the case of this blot it was apparent that values below 0.8 pg were not linear and that both the 2 and 4 pg samples had either not transferred efficiently or had partially leaked from their wells. The volumes of the remaining six dots (bold values in column 4) were therefore used to plot a new graph and generate a linear regression equation for normalization of this particular image (Figure 24). There was an excellent linear correlation between the volumes of the gradient dots and the micrograms of CS DNA loaded; the correlation coefficient (r^ ) was 0.980 (confidence level of 95%). Where X = volume and Y = micrograms of DNA, the linear regression equation generated by this graph was X = -10272 + 70934Y. This equation was applied to the volumes in column 4 to calculate the amount of DNA present in each dot as shown in 172 Table VII: Calculation of the amount of hobo DNA in the 1% OR-A population as a percentage of that present in the element donating OR strain. 1 2 3 4 5 6 7 8 9 1 0 1 1 Name Sample SC vol SC-back ug hobo vol backgrd hobo-back 100% OR %OR comments ELPS-C-01 1%OR-AF1 169501 168265 2.5 33938 25967 7972 501535 1.6 ELPS-C-02 1%OR-AF2 164571 163335 2.4 37237 25035 12202 480491 2.5 ELPS-C-03 1%OR-AF3 223435 222199 3.3 120226 36156 84070 731762 11.5 ELPS-C-04 1%OR-AF4 204166 202930 3.0 139318 32516 106802 649510 16.4 ELPS-C-05 1%OR-AF5 182612 181376 2.7 143603 28444 115160 557502 20.7 ELPS-C-06 1%OR-AF6 130312 129076 2.0 78021 18563 59458 334251 17.8 ELPS-C-07 1%OR-AF7 110044 108808 1.7 92969 14734 78235 247731 31.6 ELPS-C-08 1%OR-AF8 114127 112891 1.7 120604 15505 105099 265161 39.6 ELPS-C-09 1%OR-AF9 151564 150328 2.3 156173 22578 133595 424969 31.4 ELPS-D-01 1%OR-A F10 224372 223136 3.3 300203.3 36333 263870 735760 35.9 ELPS-D-02 1%OR-AF11 200606 199370 3.0 257634 31843 225791 634313 35.6 ELPS-D-03 1%OR-A F12 162158 160922 2.4 185506 24579 160927 470191 34.2 ELPS-D-04 1%OR-A F13 112930 111694 1.7 116660 15279 101381 260049 39.0 ELPS-D-05 1%OR-A F14 126842 125606 1.9 174297.7 17907 156390 319436 49.0 ELPS-D-06 1%OR-A F15 84069 82833 1.3 66468 9827 56642 136852 41.4 ELPS-D-07 1%OR-A F16 128050 126814 1.9 173402 18136 155266 324594 47.8 ELPS-D-08 1%OR-A F18 167250 166014 2.5 242322.9 25541 216782 491926 44.1 ELPS-D-09 1%OR-A F20 134431 133195 2.0 196765.6 19341 177424 351832 50.4 ELPS-E-01 0.0 ug CS 2276 1040 0.2 1612 -5626 7238 -212296 -3.4 invalid ELPS-E-02 0.2 ug CS 17475 16239 0.4 2628 -2755 5382 -147418 -3.7 Invalid ELPS-E-03 0.4 ug CS 34618 33382 0.6 3648 484 3164 -74237 -4.3 Invalid ELPS-E-04 0.6 ug cs 41077 39841 0.7 5010 1704 3305 -46667 -7.1 invalid ELPS-E-05 0.8 ug CS 53812 52576 0.9 5586 4110 1476 7695 19.2 invalid ELPS-E-06 1.0 ug CS 60414 59178 1.0 6416 5358 1058 35878 2.9 invalid ELPS-E-07 1.5 ug CS 100490 99254 1.5 11977 12929 -951 206947 -0.5 ELPS-E-08 2.0 ug CS 91669 90433 1.4 11341 11262 78 169292 0.0 (blurry) ELPS-E-09 2.5 ug CS 146196 144960 2.2 18682 21564 -2881 402051 -0.7 ELPS-E-10 3.0 ug CS 209955 208719 3.1 31930 33609 -1679 674220 -0.2 ELPS-E-11 3.5 ug CS 247409 246173 3.6 42837 40685 2152 834100 0.3 ELPS-E-12 4.0 ug CS 249431 248195 3.6 41814 41067 747 842731 0.1 wk hybe ELPS-F-01 0.0 ug OR 1413 177 0.1 1648 -5789 7437 -215979 -3.4 invalid ELPS-F-02 0.2 ug OR 13757 12521 0.3 14464 -3457 17921 -163288 -11.0 Invalid ELPS-F-03 0.4 ug OR 22527 21291 0.4 28621 -1800 30421 -125853 -24.2 Invalid ELPS-F-04 0.6 ug OR 27488 26252 0.5 33463 -863 34325 -104677 -32.8 invalid ELPS-F-05 0.8 ug OR 41901 40665 0.7 59712 1860 57851 -43149 -134.1 invalid ELPS-F-06 1.0 ug OR 50021 48785 0.8 90583 3394 87189 -8490 -1026.9 invalid ELPS-F-07 1.5ug OR 93778 92542 1.45 197318 11661 185657 178295 104.1 ELPS-F-08 2.0 ug OR 113012 111776 1.7 279820 15295 264525 260403 101.6 ELPS-F-09 2.5 ug OR 129592 128356 2.0 341469 18427 323042 331174 97.5 ELPS-F-10 3.0 ug OR 210452 209216 3.1 694729 33703 661026 676343 97.7 ELPS-F-11 3.5 ug OR 219404 218168 3.2 737994 35394 702599 714556 98.3 ELPS-F-12 4.0 ug OR 240124 238888 3.5 866237 39309 826928 803001 103.0 ELPS-K-01 bckgrd 1 1175 -61 0.1 2509 -5834 8343 -216998 -3.8 ELPS-K-02 bckgrd 2 1204 -32 0.1 3347 -5828 9175 -216873 -4.2 ELPS-K-03 bckgrd 3 1475 239 0.1 2202 -5777 7979 -215715 -3.7 ELPS-K-04 bckgrd 4 1378 142 0.1 2377 -5796 8173 -216130 -3.8 ELPS-K-05 bckgrd 5 1125 -111 0.1 1651 -5843 7494 -217212 -3.5 ELPS-K-06 bckgrd 6 830 -406 0.1 1635 -5899 7534 -218468 -3.4 ELPS-K-07 bckgrd 7 1308 72 0.1 1803 -5809 7611 -216430 -3.5 ELPS-K-08 bckgrd 8 820 -416 0.1 1596 -5901 7497 -218510 -3.4 ELPS-K-09 bckgrd 9 1051 -185 0.1 1433 -5857 7291 -217526 -3.4 ELPS-K-10 bckgrd 10 1074 -162 0.1 1606 -5853 7459 -217430 -3.4 ELPS-K-11 bckgrd 11 820 -416 0.1 1575 -5901 7476 -218510 -3.4 ELPS-K-12 bckgrd 12 3018 1782 0.2 1892 -5486 7378 -209128 -3.5 ELPS-K-13 bckgrd 13 1257 21 0.1 1657 -5819 7476 -216648 -3.5 ELPS-K-14 bckgrd 14 1127 -109 0.1 1583 •5843 7426 -217201 -3.4 ELPS-K-15 bckgrd 15 1253 17 0.1 1769 -5819 7589 -216665 -3.5 ELPS-K-16 bckgrd 16 1206 -30 0.1 2292 •5828 8121 -216866 -3.7 ELPS-K-17 bckgrd 17 893 -343.12 0.1 1589 -5887 7477 -218201 -3.4 mean back 1236 0.0 0.0 173 0-1 : — i H 1 H 1 1 ' 0 0.5 1 1.5 2 2.5 3 3.5 micrograms CS strain D N A Figure 24: Determination of micrograms DNA present in samples. The X axis represents the micrograms of DNA loaded into the manifold wells for the gradient of CS strain DNA, as estimated by spectrophotometry. The Y axis is the volume of dots on a digital image produced by the hybridization of a single copy probe to this CS gradient, after background has been removed. Linear regression analysis of the data points (squares) yielded the line shown which has the formula X = -10272 + 70934Y. This linear regression equation can be used to determine the actual amount of micrograms of DNA present in any dot on this same blot, based on the volume of the dot on the digital image. 174 column 5. This completes step 2. Any experimental dots which contained amounts of DNA outside of the range used to generate the normalization equation on a particular blot were not used in the final data and were noted as invalid in column 11. However, in this case, all experimental dots containing DNA from the 1% OR-A population contained between 1.3 and 3.3 micrograms of DNA and were therefore spanned by the 0.8 to 3.5 microgram samples of CS DNA used to generate the regression equation. STEP 3 : The volumes of dots on the digital images produced with hobo element probe on this same dot blot, as measured by ImageQuaNT, are shown in column 6. Background correction for the images produced by the hobo element probes was complicated by the slight hybridization obtained to CS control strain DNA which contains high molecular weight hobo homologous DNA. To simultaneously account for this background hybridization and for non-specific background on the image, the volumes of the CS gradient dots on the image produced by the hobo probe (bold values in column 6) were plotted against the micrograms of CS DNA present in each dot (column 5) to yield the graph in Figure 25. Note that only those data points for dots which contained between 0.9 and 3.6 micrograms were used. Also, since it is now known how much DNA is present in the dots that did not contain as much DNA as expected (the "2" and "4" ug CS samples), these data are included. This graph generated the linear regression equation X = -7763 + 13401 Y (r2 = 0.988) which was then be used to calculate that proportion of the pixel volume of experimental dots which was due to background (X values). This background value was calculated for each experimental dot individually to yield the values in column 7. These values were subtracted from the total volume of each dot on the image produced by the hobo 175 45000 0-1—: 1 1 1 1 H : 1 1 1 0 0.5 1 1.5 2 2.5 3 3.5 4 micrograms CS strain D N A Figure 25: Calculation of background hybridization to hobo element probe. The X axis represents the micrograms of CS strain DNA present in the gradient, as calculated by linear regression. The Y axis is the volume of dots on a digital image produced by the hybridization of a hobo element probe to these CS gradient dots. Linear regression analysis of the data points (squares) yielded the line shown which has the formula X = -7763 + 13401Y. For any given volume of dot on the image produced by the hobo probe, this linear regression equation will determine the volume that is due to background. 176 element probe (column 6) to determine the volume that was due to probe hybridization to canonical hobo elements in the experimental samples (column 8). This completes step 3. STEP 4 : Finally, the amount of hobo hybridizing DNA of experimental samples could be calculated as a percentage of that present in the OR strain. The volumes of the dots produced on each image by the hybridization of the hobo element probe to the gradient of OR strain DNA (bold values in column 8) were plotted against the micrograms of DNA in the dots to produce the graph shown in Figure 26. This generated a final linear regression equation X = -260584 + 302795Y (r2 = 0.997) from which the volume (X) corresponding to 100% of OR element donating strain hobo hybridizing sequences could be calculated for any given amount of DNA (Y). These values are shown in column 9 for the micrograms of DNA present in each dot. The amount of hobo hybridizing DNA in each experimental samples, expressed as a percentage of that present in the OR element donating strain, was then calculated using the formula: (column 8 / column 9) x 100%. These final values are shown in bold in column 10 for the 1% OR-A population and are also shown as one of the two sets of data in Figure 11 A. Note that those dots which contained greater than 0.8 ugs of CS or OR strain DNA were calculated to have very close to 0% and 100% of OR hobo DNA respectively (also shown as bold values in column 10). These provide internal negative and positive controls for the quantitation of this image. 177 900000 9 a l—H o > © © 0.5 1.5 2.5 3.5 micrograms OR strain DNA Figure 26: Calculation of 100% OR strain hobo DNA. The X axis represents the micrograms of OR strain DNA present in the OR gradient, as calculated by linear regression. The Y axis is the volume of dots on a digital image produced by the hybridization of a hobo element probe to these OR gradient dots, after the removal of background. Linear regression analysis of the data points (squares) yielded the line shown which has the formula X = -260584 + 302795Y. For any given amount of DNA in a dot, this linear regression equation will determine the volume that would be expected on the hobo image, if the DNA sample in the dot was from the OR strain. That is, it will calculate the volume expected for 100% OR strain DNA. 178 APPENDIX 2 Several different methods can be utilized to perform volume quantitation with ImageQuaNT. Although the software can automatically identify dots, I quickly discovered that it was easier to identify the dots manually than it was to select the correct preferences for the automatic spot finder. I proceeded to test experimentally three different ways of manual identification. The first method is to draw ellipses around each of the dots (as described in Appendix 1 and shown in Figure 23). The second method is to draw a grid such that each of the experimental dots is contained within a box (Figure 27A). In both of these methods, ImageQuaNT creates a three dimensional plot of pixel locations on the image (Figure 27B) and then calculates the volume under each peak. The third method of volume quantitation is to draw a line through each row of dots (Figure 28A). In this case ImageQuaNT creates a two dimensional plot of pixel values (Figure 28B) and calculates the area under each peak. To compare these methods, the images of several gradients with known amounts of DNA (such as the one used to generate the plot in Figure 27B) were quantitated by all three methods. Typical results are shown in Figure 29. Obviously all three methods produced very linear results with high correlation coefficients (r2 > 0.95). In general, the r2 value for the ellipse method was very slightly higher than the others. It is likely that this method is less sensitive to any background spots which might be included in the larger areas of the squares in the grid method, and also less sensitive to variation in the intensity within the experimental dots than the line method. I therefore chose to use only the ellipse method for quantitation of my experimental images. Figure 27: Volume measurement using the grid method. Panel A shows an 8 X 12 grid overlaid on the digital image produced by a single copy probe hybridized to the same blot shown in Figure 23. Panel B shows a surface plot generated from the top six rows of this image. Note that the top four rows contain about the same amount of DNA from various generations of the OR experimental populations (Panel A) and they give similar three dimenional peaks (Panel B). On the other hand, the fifth and sixth rows contain gradients of CS and OR DNA respectively, and the peaks clearly reflect this. 180 Counts 1500-1 1400 J 1300 • 1200-1100 • 1000-Figure 28: Quantitation using the line method. Panel A shows the same digital image in Figure 27, overlaid with lines to direct ImageQuaNT to the dots. Panel B shows the two dimensional plot generated by the fifth line from the top which corresponds to a gradient of CS DNA. Notice that the seventh peak is smaller than expected. This dot represents the 2 ug DNA sample which I previously indicated was not included in Figure 24 because the DNA either had not transferred efficiently or had partially leaked from the well. 181 400000 U.O 0.2 0.4 0.6 1.0 micrograms CS D N A Figure 29: Comparison of three methods of selecting regions for quantitation. The data points show volumes (after background subtraction) on images produced by dots of a gradient of CS DNA probed with a single copy probe, as measured by the three different methods. The correlation coefficients of the lines generated by regression analysis are given above each line. APPEND LX 3 182 Two different types of gradients could have been used to perform the linear regression and calculate the volumes corresponding to 100% OR strain hobo for a given number of micrograms of DNA (see Step 4 in Appendix 1). One type of gradient is obtained by simply placing increasing amount of OR strain DNA into consecutive wells of the vacuum manifold. The second type of gradient is obtained by keeping the total amount of DNA in each well constant, but increasing the amount of OR strain DNA compared to CS strain DNA. It might be argued that the second type of gradient is more appropriate, especially since all Drosophila DNA has some hobo homologous DNA. This appendix describes an experiment performed to compare the results obtained from the two types of gradients. A simple gradient of increasing amounts of OR DNA up to 1 ug was placed in wells of a vacuum manifold. A second gradient containing the same increasing amounts of OR DNA, but supplemented with enough CS DNA so that each well contained a total of 1 ug of DNA, was loaded into adjacent wells. These samples were blotted to nylon membranes and hybridized to a hobo element probe. A phosphor image was produced and the resulting dots were subjected to volume analysis. The resulting volumes are plotted against the micrograms of OR DNA in the samples in Figure 30A. As expected, the volumes of the dots produced by the two types of gradient were very similar. The samples containing CS DNA in addition to the OR DNA were generally slightly higher, presumably due to hybridization of the probe to hobo homologous DNA with the CS strain. The hybridization to this high molecular weight hobo homologous DNA was then subtracted as described in Step 3 of Appendix 1. The results for the two types of gradient are even more alike after this adjustment for the hobo 183 2 0 0 0 0 J 10000 f-o > micrograms OR DNA 20000 I ,£j 10000 O > B simple O R gradient O R into CS gradient 0.8 1.0 micrograms OR DNA Figure 30: Comparison of two types of OR gradient. The two panels show the volumes of the dots produced by hybridization of a hobo element probe to the two types of OR gradient. Panel A shows the results before correction for background hybridization to the hobo homologous DNA present in all Drosophila strains. Panel B shows the results after this correction 184 homologous DNA (Figure 30B). This experiment was repeated several times and I conclude that the two types of gradient yield equivalent results. The simple gradient of increasing amount of OR DNA is obviously easier to produce. In addition, on images which have complications such as poor transfer of some DNA samples, it can be used as a substitute for the CS gradient when calculating the amount of DNA in sample dots (Step 2 of Appendix 1). I therefore used this simple type of gradient on all my dot blots. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0089311/manifest

Comment

Related Items