Genetica 86: 37-46, 1992. © 1992 Kluwer Academic Publishers. Printed in the Netherlands.

Evolution of the transposable element mariner in the Drosophila melanogaster species group E Capy 1,2, J. R. David I & D. L. Hartl 2. 1 Laboratoire de Biologie et Gdn~tique Evolutives, Centre National de la Recherche Scientifique, 91198 Gif/Yvette Cedex, France 2 Department of Genetics, Washington University School of Medicine, Box 8232, St Louis, MO 63110, USA * author for correspondence Received and accepted 13 March 1992

Key words: Mariner, Drosophila, molecular evolution, transposable element Abstract

The population biology and molecular evolution of the transposable element mariner has been studied in the eight species of the melanogaster subgroup of the Drosophila subgenus Sophophora. The element occurs in D. simulans, D. mauritiana, D. sechellia, D. teissieri, and D. yakuba, but is not found in D. melanogaster, D. erecta, or D. orena. Sequence comparisons suggest that the mariner element was present in the ancestor of the species subgroup and was lost in some of the lineages. Most species contain both active and inactive mariner elements. A deletion of most of the 3' end characterizes many elements in D. teissieri, but in other species the inactive elements differ from active ones only by simple nucleotide substitutions or small additions/deletions. Active mariner elements from all species are quite similar in nucleotide sequence, although there are some species-specific differences. Many, but not all, of the inactive elements are also quite closely related. The genome of D. mauritiana contains 20-30 copies of mariner, that of D. simulans O-10, and that of D. sechellia only two copies (at fixed positions in the genome). The mariner situation in D. secheUia may reflect a reduced effective population size owing to the restricted geographical range of this species and its ecological specialization to the fruit of Morinda citrifolia. Introduction

Determining the phylogenetic distribution of a family of transposable elements often sheds light on whether the pattern of occurrence among related species results primarily from vertical inheritance or horizontal transmission. In Drosophila, such studies are typically carried out by means of Southern blots, dot blots, or in situ hybridization with polytene chromosomes (Brookfield et al., 1984; Bucheton et aL, 1986; Anxolabebere & P6riquet, 1987; Simonelig et al., 1988; Silber et al., 1989). A number of novel transposable elements have been described in Drosophila species other than D. melanogaster, including the Tom element in D. ananassae (Shrimpton et aL, 1986; Tanda et al., 1988), Ulysses in D. virilis (Lozovskaya et aL,

1990; Scheinker et al., 1990), microcopia and minos in D. hydei (Huijser et al., 1988; Lankenau et al., 1988) and mariner in D. mauritiana (Jacobson et al., 1986; Haymer & Marsh 1986; Hartl 1989). However, most of the approximately 40 families of transposable elements described in the genus Drosophila were initially detected in Drosophila melanogaster (reviews in Ashburner, 1989; Berg & Howe, 1989). Consequently, most of the studies of transposable elements in Drosophila have focused on the melanogaster species subgroup. In this paper we focus on the evolution of the transposable element mariner in the melanogaster species subgroup. The occurence of mariner among the eight species of the melanogaster subgroup is summarized in Figure 1, along with data for P, /, hobo, copia, and FB. All of the species contain/,

38 hobo and FB, suggesting that these elements were probably present in the ancestor of the subgroup. Various patterns are observed for the remaining elements, suggesting either horizontal transmission or vertical inheritance with stochastic loss (or both). With regard to the P element, it is now generally accepted that the reason P elements can be detected only in D. melanogaster is that horizontal transfer probably occurred between D. willistoni and D. melanogaster (Kidwell, 1989; Daniels et al., 1990), and a specific mechanism of horizontal transfer has been suggested (Houck et al., 1991; see also Hartl et al. in this volume). The mariner element also has a discontinuous distribution among the species. However, in contrast to the P element, the presence of mariner in this species subgroup appears to result from vertical inheritance, and the absence of mariner in certain species could result from stochastic loss (Capy et al., 1990, 1991a; Maruyama & Hartl, 1991b). The peculiarities in mariner distribution are discussed below in light of the ecological status of the various species.

The mariner element

The mariner element was first discovered in connection with an exceptional male having peachcolored eyes found in the Cambridge wildtype strain of D. mauritiana (Jacobson & Hartl, 1985; Haymer & Marsh, 1986). The mutation was designated white-peach (wPCh),and further analysis demonstrated that it was associated with the insertion of a mariner element in 5' region of the first exon of the white gene (Jacobson et al., 1986; Haymer & Marsh, 1986). The white-peach strain of D. mauritiana contains 20-30 copies of mariner (Jacobson et al., 1986), and the particular copy inserted in the white gene is defective and unable to excise or transpose in the absence of active elements elsewhere in the genome (Garza et aL, 1991; Medhora et aL, 1991). Several copies of mariner in this strain are potentially active, depending on their position in the genome (Maruyama et al., 1991), but the w pch allele in the strain is relatively stable as judged from the rate of germline reversion, which occurs at a frequency of 2-4 x 10 -3 in males and approximately twofold lower in females (Jacobson & Hartl, 1985; Haymer & Marsh, 1986). Somatic reversion, de-

tected phenotypically as flies with mosaic eye color, occurs at a frequency of approximately 4 X 10-3 in both sexes (Haymer & Marsh, 1986; Jacobson et aL, 1986). The particular copy of mariner present in the wPCh allele is designated the peach element. While inactive, its DNA sequence exhibits the same overall length and organization as functional mariner elements. In particular, the peach element is 1,286 base pairs (bp) in length and contains terminal inverted repeats of 28 bp having four mismatches. The element contains a single uninterrupted open reading frame of 1,038 bp, capable of coding for a protein of 346 amino acids, which is preceded by promoter-like and ribosome-binding sequences (Jacobson et al., 1986). Strains of D. simulans generally have many fewer copies of mariner than strains of D. mauritiana, and some laboratory strains of D. simulans lack mariner completely. The wPch allele was introduced into such a strain of D. simulans by repeated backcrossing and selection, eventually resulting in a strain called GB1, in which the only mariner element is the peach element present in the wPch allele. This useful strain was used in testcrosses to detect the presence or absence of active elements in natural populations of D. simulans (Capy et aL, 1990, 1991b). Distribution of mariner in the genus Drosophila The mariner element occurs in five of the eight species of the melanogaster species subgroup and in several other species of the subgenus Sophophora, particularly in species in the ananassae, montium, and suzukii subgroups (Maruyama & Hartl, 1991b; Capy, unpublished data). Outside Sophophora, presence of the mariner element has been found in several species of Zaprionus (Z. tuberculatus, Z. indianus, and Z. ghesquieri) (Maruyama & Hartl, 1991a; Capy, unpublished data), and one copy from Z. tuberculatus has been sequenced (Maruyama & Hartl, 1991a). More recently, sequences with significant DNA sequence identity with mariner have been found in some species very distantly related to Drosophila, namely, in the moth Hyalophora cecropia (Lidholm et al., 1991) and in the nematode Caenorhabditis elegans (Philip Morgan, personal communication).

39

The melanogaster subgroup

P

I

hobo

copia

D.orena

-

+

?

?

+

D.erecta

-

4-

+

?

÷

-

D.yakuba

--

+

÷

--

4-

4-

÷

÷

+

4-

÷

--

_~

FB m a r i n e r

D. t e i s s i e r i

D.melanogaster i

D.simulans

i

D.mauritiana D.sechellia

--

÷

÷

+

+

4-

÷

4-

4"

4-

4-

+

+

?

+

+

Fig. 1. Consensus phylogeny of the melanogaster species subgroup and occurrence of various transposable elements among the species. Transposon data from Martin et al. (1983), Brookfield et al. (1984), Stacey et al. (1986), Silber et aL (1989), Daniels et al. (1990) and

Maruyama and Hartl (1991b).

Within the genus Drosophila, most of the species containing mariner are endemic to southeast Asia or tropical Africa. Thus far, no Drosophila species endemic to the American or European continent has been found to contain mariner. This finding suggests that mariner entered the genome of an ancestor of the melanogaster species group through a species endemic to Asia or Africa. Furthermore, the similarity of the mariner elements in D. mauritiana and Z. tuberculatus strongly suggests horizontal transfer between the ancestors of these species (Maruyama & Hartl, 1991a), which are found in sympatry on Mauritius Island (David et aL, 1989).

The melanogaster subgroup For background we describe briefly the phylogenetic relationships of the species in the melanogaster species subgroup and discuss their geographical distributions and ecological status. The melanogaster subgroup (Fig. 1) is generally divided in two main 'complexes': the melanogaster complex includes D. melanogaster, D. simulans, D. mauri-

tiana and D. sechellia; and the yakuba complex includes D. yakuba, D. teissieri, D. orena and D. erecta. The principal ambiguity in the phylogeny concerns the relationship between D. simulans, D. mauritiana and D. sechellia. Depending on the trait examined, any of the theoretically possible relationships between three species can be observed. For practical purposes this means that the branching pattern is too close to discriminate, and so we represent it as a trifurcation. The geographical distribution and the breeding sites of these species are very different (reviewed in Lemeunier et al., 1986; Lachaise et al., 1988). D. melanogaster and D. simulans are cosmopolitan species with similar, and wide, ecological niches. D. mauritiana and D. sechellia are endemic to Mauritius Island and Seychelles Archipelago, respectively. The former is a generalist species (David et al., 1989) while the latter is specialized on the fruit of Morinda citrifolia, which is highly toxic for its sibling species D. mauritiana and D. simulans (R'Kha et al., 1991). The geographical distributions of D. teissieri and D. yakuba extend from northwest to southeast Africa, while D. erecta

40 is restricted to Western Africa (see species distribution map in Lemeunier et al., 1986). For D. orena, the species is known from a single female caught on Mount Lefo in The Cameroon (Tsacas & David, 1978). The breeding sites of the species are not well-defined, but for D. erecta it is primarily the fruit of Pandanus candelabrum (Lachaise & Tsacas, 1974; Rio et aL, 1983).

Occurrence of mariner in the melanogaster subgroup The melanogaster complex

Copies of mariner occur in the genome of three out of the four species in this complex. Only D. melanogaster seems free of this element. On the other hand, relatively few recently isolated strains of D. melanogaster have been investigated, and so it cannot be ruled out that some natural populations might contain the element. Each of the three remaining species exhibit a very different picture regarding mariner. The simplest situation is in D. sechellia. Only two mariner elements have been detected, and they are at fixed sites in the genome. One of the elements contains three deletions of 1,128 and 602 bp, the last of which includes all of the 3' half of the element. The other element in the genome is full-length, and two sequenced copies (PA2 and 228C, isolated from strains collected from Praslin and Cousin Island, respectively) are very similar in sequence to the full-length element described in D. mauritiana as well as similar to each other. However, PA2 and 228C are not completely identical, showing that polymorphisms between mariner elements exist. As noted, the two elements in the genome of D. sechellia seem to be immobile, since they are at identical positions in independent strains from Mahr, Praslin, Cousin and Frigate Islands (Capy et al., 1991a). The full-length element is localized in chromosome 3L in salivary chromosome bands 64A 10-11, and the deleted copy is in chromosome 2R at position 51A1-2. Several observations suggest that the full-length element of D. sechellia is at least potentially active. First, when males of D. sechellia are crossed with wPch females of D. simulans (the GB1 strain), 1% of the F 1 males have mosaic eyes (Capy et al., 1991a). Coyne (1989) also observed mosaic males in the

offspring of a cross between males of D. sechellia and wPch females of D. mauritiana; however, the proportion of mosaic males in the progeny was about 25%. Second, DNA corresponding to the complete element from D. sechellia (produced by the polymerase chain reaction) was injected into embryos of the GB1 strain of D. simulans; among 50 survivors, one female with three pigmented patches in one eye was obtained (Capy, unpublished observations). Third, the full-length elements from D. sechellia are more similar in sequence to the active mariner elements of D. mauritiana and D. simulans than they are to the inactive elements (Capy et al., 1991b). Finally, Maruyama et al. (1991) have shown that many inactive mariner elements have a T ---> A transversion at nucleotide position 1203, whereas the full-length mariner elements from D. sechellia have a T at this position. In D. simulans, all natural populations contain active mariner elements (Capy et al., 1990). Active elements present in males of a population are detected as mosaic male progeny in testcrosses with GB 1 females. Both the level of expression (number of pigmented spots) and the penetrance (proportion of males showing a mosaic phenotype) are highly variable among isofemale lines from the same population (Capy et al., 1990). Usually, most isofemale lines contain elements with low levels of expression and penetrance. However, high levels of expression and penetrance can be obtained by artificial selection (Capy and David, unpublished). The most likely interpretation is that the selection acts upon differences due to position effects (Garza et al., 1991; Maruyama etal., 1991). Southern blots of isofemale lines from various populations indicate that the copy number per genome varies from 0-10, and the average number of copies in a population may change over time (Chakrani and Capy, unpublished results). Although no recently established isofemale lines have been shown to lack mariner, several old laboratory lines lacking mariner have been found. Sequence analysis of mariner elements isolated from different natural populations and laboratory strains indicate that the active elements are very similar to the highly active Mosl element of D. mauritiana and that the inactive elements are very similar to the inactive peach element inserted in the wPch allele of D. mauritiana (Capy et al., 1991b).

41

I D.mauritiana vx

~ ~

D. t e i s s i e r i

D. mauriti~ma

III

D.simulans iv II

vI

~ D.sechellia

_ _ ~ D . mauritiana I

~

D.simulans High activity ~

activit~

~~D.

mauritiana

naotiv°) [Onkno-- activityI

Fig. 2. Phylogenetic tree based on sequences of mariner elements present in species of the melanogaster species subgroup. Activity of the elements is symbolized in the geometrical shapes around the names. The species of origin of each of the mariner elements is indicated. The sequence data were analyzed by Phylogenetic Analysis Using Parsimony (PAUP of Swofford, 1989).

42 In D. mauritiana, the average number of mariner copies per genome is 20-30 (Hartl, 1989). Many of the elements are functional, but in most strains their level of activity is relatively low owing to position effects (Maruyama et al., 1991). Occasionally an element transposes to a location where a high level of expression can occur, and these events account for the spontaneous occurrence of high-level, heritable mosaicism in the wpch strain (Bryan et al., 1987; Maruyama et aL, 1991). Position effects have also been demonstrated when mariner elements from D. mauritiana are introduced into the genome of D. melanogaster by germline transformation (Garza et al., 1991; Maruyama et al., 1991; Medhora et al., 1991). The yakuba complex

Only two species in this complex contain mariner. D. yakuba has about four copies per genome, and D. teissieri has about 10. In D. yakuba, all of the mariner elements appear to be full-length, judging from hybridization with various segments of the D. mauritiana element (Maruyama & Hartl, 1991b). Two D. yakuba elements that have been sequenced are both full-length and show an average identity of 98% with active elements from D. mauritiana (Maruyama et al., 1991). It is therefore possible that the D. yakuba elements are functional. In D. teissieri, mariner elements with deletions of most of the 3' end have been detected by hybridization with probes corresponding to different segments of the D. mauritiana element. Detailed analysis of a genomic library suggests that about 20% of the mariner elements in D. teissieri are fulllength while the rest have the large deletion (Maruyama et aL, 1991). The sequence of the deleted element shows a deletion of 716 base pairs spanning positions 544-1260, but the inverted repeats of the element remain intact (Maruyama et aL, 1991), The occurence of deleted elements is also supported by the results of amplification of mariner from the D. teissieri genome using the polymerase chain reaction, using primer oligonucleotides corresponding to the inverted repeats (Capy, unpublished data). The amplification regularly yields two products, one of which is comparable in size to full-length mariner element, while the other is about 700 bp shorter.

Discussion and conclusions Phylogeny of mariner elements

A phylogeny of all sequenced mariner elements from species of the melanogaster subgroup is given in Figure 2. The tree is more complete than that in Capy et al. (1991b), since it includes the D. yakuba and D. teissieri elements. The latter elements are clearly separate from the others, and the differences are shown in Figure 3. A total of 13 nucleotide differences (at positions 47, 73, 136, 146, etc., listed in Figure 3) clearly distinguish mariner elements in the sibling species D. mauritiana, D. simulans, and D. sechellia from those in D. yakuba and D. teissieri. Furthermore, the five nucleotide differences between the mariner elements in D. yakuba and D. teissieri (asterisks in Fig. 3) are all in the direction of making the D. yakuba element more similar, at these positions, to the active elements from species in the melanogaster complex. On the other hand, the mariner elements in D. yakuba are themselves polymorphic, and there are seven nucleotide differences, most of them in the open reading frame, between the elements Yak2 and Yak3. Considering only the active elements in the three sibling species of the melanogaster complex, it appears that the mariner elements in D. simulans and D. mauritiana are more closely related to each other than they are to D. secheIfia. Between D. mauritiana and D. simulans elements there is only one nucleotide difference (at position 375, see Fig. 3), while there are four nucleotide differences between elements from D. simulans and D. secheltia, and six between elements from D. mauritiana and D. sechellia. These differences support a model in which D. mauritiana and D. sechellia derived independently from a D. simulans ancestor, as already argued from data on mitochondrial DNA (Solignac et al., 1986). Nevertheless, other traits support different phylogenetic relationships between the species (Lemeunier & Ashburner, 1976; Solignac et aL, 1986; Coyne & Kreitman, 1986; Cariou, 1987; Satta & Takahata, 1990).

Inactive versus active elements

In the three species of the melanogaster complex the active elements show only the differences sum-

43 L 8/L 14/MB4 Mos6a/Mos6b/Mos2 MadB/Sey2/Pr 1

D. simulans Node I V

/ G--~A in position 375

G--~T in position 206 AIIeG in position 1017 T-I~A in position 1214

p.

in position 1220

D. mauritiana

/

D. seche 11ia

Mosl/MosJ/MosS/Ma351 Node III

Node V

PA/228C

G->T in position 127 GII>T in position 206 GII~A in position 375 AI~G in position 1017 T-~A in position 1214 C-I~T in position 1220

G~C in position 47

T-~A in position 246

A-->T in position 73

G-cA in position 678

T-->C in position 136

T-I~C in position 780

AI>T in position 146

A--)C in position 806

T-I~C in position 149

C-I~T in position 1056

AI)G in position 213

AI~G in position 1106 G I l A in position 1249

Node Vl

D. yakuba Yak2/Yak3

* A - ~ T in position 31 * AII>G in position 34

m.

D. t e i s s i e r i Te is32

* A-I~G in position 382 * C-I~T in position 493 * A-I~G in position 712

Fig. 3. Summary of fixed nucleotide sequence differences in mariner from different species in the melanogaster subgroup. Only the active elements of D. simulans and D. mauritiana are included.

marized in Figure 3, leading to the conclusion that active elements were probably present in the common ancestor of the species (Capy et al., 1991b). This conclusion is also supported by the comparisons of Maruyama et al. (1991). Furthermore, the species-specific characteristics in D. simulans must have been acquired prior to the worldwide expansion of D. simulans, since the differences are found in natural populations as distant as Europe, California, the West Indies, Tropical Africa, North Africa, Madagascar, and the Seychelles. The similarity between the inactive elements of D. simulans and D. mauritiana is more difficult to explain. Two possibilities have been proposed (Capy et aL, 1991b). First, the similarity between

inactive elements could result from selective constraints on inactive elements, perhaps because they are involved in regulation in a way analogous to the role of the deleted K P elements in the P-M system (Black et al., 1987; Rio, 1991). Secondly, the similarity could result from recent introgression of D. simulans into D. mauritiana. Introgression between the species has already been suspected from the analysis of mitochondrial DNA (Solignac & Monnerot, 1986; Aubert & Solignac, 1990). Sequence comparisons between highly active, weakly active, and inactive elements show that several nucleotide substitutions in the coding region can affect activity (Capy et al., 1991b). Indeed, three quite cohesive groups of elements can be

44 discerned in the tree in Figure 2. One group (defined by nodes I and II in Fig. 2) contains the inactive elements peach, PrA, BordA, Ma310 and Ma311. The main inconsistency among the inactive elements in the tree is in the position of the element MA331 of D. mauritiana. This element shows no activity when transformed into D. melanogaster (Maruyama et al., 1991), but its sequence is very similar to the highly active elements of D. mauritiana. It is very possible that Ma331 represents an independent class of inactive elements recently derived from an active element. A second cohesive group in Figure 2 consists of the weakly active elements Ma341 and MB1, and a third cohesive group (defined by node III and IV) contains the highly active elements Mosl, Mos5, MB4, L8, L14, MadB, Mos6a, Mos6b, Sey2, Mos2 and Prl. Between nodes I and II on the tree there are three consistent nucleotide differences, at positions 64, 154 and 305, and between nodes II and III there is a difference at position 1203, which was previously identified from the analysis of a smaller number of inactive elements as being important in the loss of activity (Maruyama et aL, 1991). In the tree in Figure 2, the mariner elements from D. sechellia and D. yakuba are grouped with the highly active elements, suggesting that the D. sechellia and D. yakuba elements could very well be active.

Stochastic loss or horizontal transfer The mariner element occurs within the drosophilids (Maruyama & Hartl, 1991) as well as in more distantly related taxa, such as Lepidoptera (Lidholm et aL, 1991) and nematodes (P. Morgan, personal communication). These observations suggest that mariner will be found in many other species as well. Among the eight species of the melanogaster subgroup, mariner is absent in three (D. melanogaster, D. erecta, and D. orena). Two scenarios, not mutually exclusive, can account for this distribution: mariner may have been present in the common ancestor of the species subgroup, or it invaded some of the species after their separation. If inheritance from a common ancestor is discounted as a possibility, then at least two interspecific horizontal transfers must be invoked to account for the distri-

bution pattern, as indicated by the phylogenetic tree in Figure 1: one an invasion of the common ancestor of D. yakuba and D. teissieri, and the other an invasion of the common ancestor of the D. simulans complex (Maruyama & Hartl, 1991). The alternative to the multiple invasion scenario is the vertical transmission of mariner accompanied by stochastic loss. If we assume that mariner was present in the common ancestor of the subgroup, then two independent losses must be invoked to explain the present distribution: one in the common ancestor of D. erecta and D. orena, and the other in the lineage leading to D. melanogaster. Two arguments favor the model of stochastic loss. First, some laboratory strains of D. simulans lack mariner elements, although they are widespread in natural populations (Capy et al., 1990, 1991 a), which typically contain 1-10 mariner elements per diploid genome (Capy, unpublished results). Hence, the loss of the elements from certain laboratory strains probably occurred during laboratory maintenance and the accompanying inbreeding. The second argument is that the molecular phylogeny of the mariner elements in the species of the melanogaster subgroup is quite similar to the phylogeny of the species themselves (Fig. 2). Such a finding, which not conclusive, is more simply explained by vertical transmission than horizontal transfer (Maruyama & Hartl, 1991). At the species level, stochastic loss may be promoted by a reduced effective population size (Ne), and there is evidence that some of the species in the melanogaster subgroup have small effective population sizes. For example, D. orena is rare in nature and is known only from a single isofemale line; and its closest relative, D. erecta, is specialized on the fruits of Pandanus, and may undergo severe bottlenecks of population size every year. (On the other hand, especially considering the situation in some laboratory strains of D. simulans, it is possible that the absence of mariner from the genome of D. orena is a laboratory artifact). At the DNA level, D. melanogaster is less polymorphic than D. simulans, suggesting a much smaller effective population size (Aquadro et al., 1988). The lowest level of polymorphism among all species in the melanogaster subgroup, which probably reflects the smallest effective population size (Cariou et al., 1990), is found in D. sechellia; in this species two copies of mariner are present at fixed positions in the

45 genome, and one copy is clearly defective (Capy et aL, 1991a). An additional consideration that may bear on the long-term maintenance of mariner in a species is possible genetic variation in the regulation of transposition. The importance of this factor is difficult to evaluate at this time, since little is known at present about regulatory mechanisms governing mariner. In particular, it is not known whether the absence of mariner in D. melanogaster, and the large variation in copy number in D. simulans, D. mauritiana, and D. sechellia, are somehow determined by genetic factors that differ among the species.

Acknowledgements This work was supported by NIH grant GM33741 to DLH and by a NATO Fellowship to PC.

References Anxolabthtre, D. & G. Ptriquet, 1987. P-homologous sequences in Diptera are not restricted to the Drosophilidae family. Genet. Iber. 39:211-222. Aquadro, C. F., K. M. Lado & W. A. Noon, 1988. The rosy region of Drosophila melanogaster and Drosophila simulans. I. Contrasting level of naturally occurring DNA restriction map variation. Genetics 119: 875-888. Ashburner, M., 1989. Drosophila: A Laboratory Handbook. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. Aubert, J. & M. Solignac, 1990. Experimental evidence for mitochondrial DNA introgression between Drosophila species. Evolution 44: 1272-1282. Berg, D. E. & M. M. Howe (editors), 1989. Mobile DNA. American Society for Microbiology, Washington, D.C. Black, D. M., M. S. Jackson, M. G. Kidwell & G. A. Dover, 1987. KP elements repress P-induced hybrid dysgenesis in D. melanogaster. EMBO J. 6: 4125-4135. Brookfield, J. E Y., E. Montgomery & C. H. Langley, 1984. Apparent absence of transposable elements related to the P element of D. melanogaster in other species of Drosophila. Nature 310: 330-332. Bryan, G. J., J. W. Jacobson & D. L. Hartl, 1987. Heritable somatic excision of a Drosophila transposon. Science 235: 1636-1638. Bucheton, A., M. Simonelig, C. Vaury & M. Crozatier, 1986. Sequences similar to the I transposable element involved in I-R hybrid dysgenesis in D. melanogaster occur in other Drosophila species. Nature 322: 650-652. Capy, P., E Chakrani, E Lemeunier, D. L. Hartl & J. R. David, 1990. Active mariner transposable elements are widespread in natural populations of Drosophila simulans. Proc. Royal

Soc., Lond. B 242: 57-60. Capy, R, K. Maruyama, J. R. David & D. L. Hartl, 1991a. Insertion sites of the transposable element mariner are fixed in the genome of Drosophila sechellia. J. Mol. Evol. 33: 450-456. Capy, E, A. Koga, J. R. David & D. L. Hartl, 1991b. Sequence variation among active mariner elements in natural populations of Drosophila simulans. Genetics 130: 499-506. Cariou, M. L., 1987. Biochemical phylogeny of the eight species in the Drosophila melanogaster species subgroup, including D. sechellia and D. orena. Genet. Res., Camb. 50: 181-185. Cariou, M. L., M. Solignac, M. Monnerot & J. R. David, 1990. Low allozyme and mtDNA variability in the island endemic species Drosophila sechellia (D. melanogaster complex). Experientia 46:101-104. Coyne, J.A., 1989. Mutation rates in hybrids between sibling species of Drosophila. Heredity 63: 155-162. Coyne, J. A. & M. Kreitman, 1986. Evolutionary genetics of two sibling species, Drosophila simulans and Drosophila sechellia. Evolution 40: 673-691. Daniels, S. B., K. R. Peterson, L. D. Strausbaugh, M. G. Kidwell & A. Chovnick, 1990. Evidence for horizontal transmission of the P transposable element between Drosophila species. Genetics 124: 339-355. David, J. R., S. E McEvey, M. Solignac & L. Tsacas, 1989. Drosophila communities on Mauritius and ecological niche of D. mauritiana (Diptera, Drosophilidae). Revue Zool. Afr.-J. Aft. Zool. 103: 107-116. Garza, D., M. Medhora, A. Koga & D. L. Hartl, 1991. Introduction of the transposable element mariner into the germline of Drosophila melanogaster. Genetics 128:303-310. Hartl, D. L., 1989. Transposable element mariner in Drosophila species, pp. 531-536 in Mobile DNA, edited by D. E. Berg and M. M. Howe. American Society for Microbiology, Washington, D.C. Haymer, S. & J. L. Marsh, 1986. Germ line and somatic instability of a white mutation in Drosophila mauritiana due to a transposable element. Dev. Genet. 6:281-291. Houck, M. A., J. B. Clark, K. R. Peterson & M. G. Kidwell, 1991. Possible horizontal transfer of Drosophila genes by the mite Proctolaelaps regalis. Science 253:1125-1129. Huijser, E, C. Kirchoff, D. H. Lankenau & W. Hennig, 1988. Retrotransposon-like sequences are expressed in the Y chromosomal lampbrush loops of Drosophila hydei. J. Mol. Biol. 203: 689-697. Jacobson, J. W., M. M. Medhora & D. L. Hartl, 1986. Molecular structure of a somatically unstable transposable element in Drosophila. Proc. Natl. Acad. Sci. USA 83: 8684-8688. Jacobson, J. W. & D. L. Hartl, 1985. Coupled instability of two X-linked genes in Drosophila mauritiana: Germinal and somatic mutability. Genetics 111: 57-65. Kidwell, M. G., 1989. Regulatory aspects of the expression of P-M hybrid dysgenesis in Drosophila, pp. 183-194 in Transposable Elements as Mutagenic Agents, edited by M. E. Lambert, J. E McDonald and I. B. Weinstein. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Lachaise, D., M. Cariou, J. R. David, E Lemeunier, L. Tsacas & M. Ashbumer, 1988. Historical biogeography of the Drosophila melanogaster species subgroup. Evol. Biol. 22: 159227.

46 Lachaise, D., J. R. David, E Lemeunier, L. Tsacas & M. Ashburner, 1986. The reproductive relationships of Drosophila sechellia with D. mauritiana, D. simulans, and D. melanogaster from the afrotropical region. Evolution 40:262-271. Lachaise, D. & L. Tsacas, 1974. Les Drosophilidae des savanes prrforestibres de la region tropicale de Lamto (C6te d'Ivoire). Le peuplement des fruits de Pandanus candelabrum (Pandanac~es). Ann. Univ. Abidjan Ser. E. Ecol. 7: 153-192. Lankenau, D.-H., P. Huijser, E. Jansen, K. Miedema & W. Hennig, 1988. Microcopia: A retrotransposon of Drosophila combining structural features of DNA viruses, retroviruses and non-viral transposable elements. J. Mol. Biol. 204: 233246. Lemeunier, E & M. Ashburner, 1976. Relationships within the melanogaster species subgroup of the genus Drosophila (Sophophora). Chromosoma 89:343-351. Lemeunier, F., J. R. David, L. Tsacas & M. Ashbumer, 1986. The melanogaster species group, pp. 147-256 in The Genetics and Biology of Drosophila, Volume 3, edited by M. Ashburner and H. L. Carson, Academic Press, N.Y. Lidholm, D.-A., G. H. Gudmundsson & H. G. Boman, 1991. A highly repetitive, mariner-like element in the genome of Hyalophora cecropia. J. Biol. Chem. 266: 11518-11521. Lozovskaya, E. R., V. S. Scheinker & M. B. Evgen'ev, 1990. A hybrid dysgenesis syndrome in Drosophila virilis. Genetics 126: 619-623. Martin, G., D. Wiemasz & P. Schedl, 1983. Evolution of Drosophila repetitive-dispersed DNA. J. Mol. Evol. 19: 203-213. Maruyama, K. & D. L. Hartl, 1991a. Evidence for interspecific transfer of the transposable element mariner between Drosophila and Zaprionus. J. Mol. Evol. 33: 514-524. Maruyama, K. & D. L. Hartl, 199 lb. Evolution of the transposable element mariner in Drosophila species. Genetics 128: 319-329. Maruyama, K., K. D. Schoor & D. L. Hartl, 1991. Identification of nucleotide substitutions necessary for trans-activation of mariner transposable elements in Drosophila: Analysis of naturally occurring elements. Genetics 128: 777-784. Medhora, M., K. Maruyama & D. L. Hartl, 1991. Molecular and functional analysis of the mariner mutator element Mosl in Drosophila. Genetics 128:311-318. R°Kha, S., P. Capy & J. R. David, 1991. Host-plant specialization in the Drosophila melanogaster species complex: A physiological, behavioral and genetical analysis. Proc. Natl. Acad. Sci. USA 88: 1835-1839. Rio, B., G. Couturier, F. Lemeunier & D. Lachaise, 1983. Evolution d'une sprcialisation saisonni~re chez Drosophila erecta

(Dipt., Drosophilidae). Ann. Soc. Ent. Fr. (N.S.) 19: 235248. Rio, D. C., 1991. Regulation of Drosophila P element transposition. Trends Genet. 7: 282-287. Satta, Y. & N. Takahata, 1990. Evolution of Drosophila mitochondrial DNA and the history of the melanogaster subgroup. Proc. Natl. Acad. Sci. USA 87: 9558-9562. Scheinker, V., E. Lozovskaya, J. G. Bishop, V. G. Corces & M. B. Evgen'ev, 1990. A long terminal repeat-containing retrotransposon is mobilized during hybrid dysgenesis in D. virilis. Proc. Natl. Acad. Sci. USA 87: 9615-9619. Shrimpton, A. E., E. A. Montgomery & C. H. Langley, 1986. Om mutations in Drosophila ananassae are linked to insertions of a transposable element. Genetics 114: 125-135. Silber, J., C. Bazin, E Lemeunier, S. Aulard & M. Volovitch, 1989. Distribution and conservation of the Foldback transposable element in Drosophila. J. Mol. Evol. 28: 220-224. Simonelig, M., C. Bazin, A. Pelisson & A. Bucheton, 1988. Transposable and nontransposable elements similar to the I factor involved in inducer-reactive (IR) hybrid dysgenesis in Drosophila melanogaster coexist in various Drosophila species. Proc. Natl. Acad. Sci. USA 85: 1141-1145. Singh, R.S., 1989. Population genetics and evolution of species related to Drosophila melanogaster. Ann. Rev. Genet. 23: 425-453. Solignac, M. & M. Monnerot, 1986. Race formation, speciation and introgression within Drosophila simulans, D. mauritiana, and D. sechellia inferred from mitochondrial DNA analysis. Evolution 40:531-539. Solignac, M., M. Monnerot & J.-C. Mounolou, 1986. Mitochondrial DNA evolution in the melanogaster species subgroup of Drosophila. J. Mol. Evol. 23: 31-40. Stacey, S. N., R. A. Lansman, H. W. Brock & T. A. Grigliatti, 1986. Distribution and conservation of mobile elements in the genus Drosophila. Mol. Biol. Evol. 3: 522-534. Swofford, D. L., 1989. PAUP: Phylogenetic Analysis Using Parsimony, Version 3.0. Illinois Natural History Survey, Champaign, Illinois. Tanda, S., A. E. Shrimpton, C. Ling-Ling, H. Itayama, H. Matsubayashi, K. Saigo, Y. N. Tobari & C. H. Langley, 1988. Retrovirus-like features and site specific insertion of a transposable element, Tom, in Drosophila ananassae. Mol. Gen. Genet. 214: 405-411. Tsacas, L. & J. R. David, 1978. Une spetieme espece appartenant au sous-groupe Drosophila melanogaster Meigen: Drosophila orena spec. nov. du Cameroun (Diptera: Drosophilidae). Beitr. Entomol. 28: 179-182.