Intra- and Interspecies Variation Among Bari-1 Elements of the Melanogaster Species Group

Corresponding author: Ruggiero Caizzi, Istituto di Genetica, Università di Bari, Via Amendola, 165/A, 70126 Bari, Italy., r.caizzi{at}biologia.uniba.it (E-mail).

Communicating editor: M. J. SIMMONS

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We have investigated the distribution of sequences homologous to Bari-1, a Tc1-like transposable element first identified in Drosophila melanogaster, in 87 species of the Drosophila genus. We have also isolated and sequenced Bari-1 homologues from D. simulans, D. mauritiana, and D. sechellia, the species constituting with D. melanogaster the melanogaster complex, and from D. diplacantha and D. erecta, two phylogenetically more distant species of the melanogaster group. Within the melanogaster complex the Bari-1 elements are extremely similar to each other, showing nucleotide identity values of at least 99.3%. In contrast, Bari-1-like elements from D. diplacantha and D. erecta are on average only 70% similar to D. melanogaster Bari-1 and are usually defective due to nucleotide deletions and/or insertions in the ORFs encoding their transposases. In D. erecta the defective copies are all located in the chromocenter and on chromosome 4. Surprisingly, while D. melanogaster Bari-1 elements possess 26-bp inverted terminal repeats, their D. diplacantha and D. erecta homologues possess long inverted terminal repeats similar to the terminal structures observed in the S elements of D. melanogaster and in several other Tc1-like elements of different organisms. This finding, together with the nucleotide and amino acid identity level between D. diplacantha and D. erecta elements and Bari-1 of D. melanogaster, suggests a common evolutionary origin and a rapid diversification of the termini of these Drosophila Tc1-like elements.

TRANSPOSABLE elements constitute a significant fraction of the repeated DNA present in the genome of virtually all organisms so far studied. Many families of transposons have been isolated and characterized and, depending on structure and mechanism of transposition, they are grouped into two main classes. Class I transposons move via an RNA intermediate while class II elements transpose directly from DNA into a new DNA target site (FINNEGAN 1989 ). The wide distribution of transposons raises interesting questions about their origin and about the roles they play in the evolution of host genomes and in regulating host structural genes (for reviews, see BERG and HOWE 1989 ; KIDWELL 1992A , KIDWELL 1992B ; MCDONALD 1993 ).

Some families of transposons are present in very distantly related taxa. A well-known example is represented by the Tc1-mariner superfamily comprising mariner-like and Tc1-like elements (for reviews, see ROBERTSON and LAMPE 1995 ; PLASTERK 1996 ). Detailed analysis of copies isolated from different organisms suggests that the Tc1-mariner elements have an ancient origin (see, for example, KIDWELL 1993 ; RADICE et al. 1994 ) and that horizontal transmission events play a substantial role in their diffusion, probably because their transposition mechanism is independent of host factors (LAMPE et al. 1996 ; VOS et al. 1996 ).

Three Tc1-like elements have been isolated in Drosophila melanogaster: HB1, S, and Bari-1. HB1 was the first element identified as a Tc1-like element of D. melanogaster (HENIKOFF and PLASTERK 1988 ). It possesses 28-bp inverted terminal repeats and on the basis of DNA sequencing of the analyzed copies it appears to be a defective element (BRIERLY and POTTER 1985 ). The S element was originally discovered during the characterization of a spontaneous suppressor of sable mutation (MERRIMAN et al. 1995 ). In D. melanogaster, where both defective and complete copies exist, the complete S element (pS3) possesses 234-bp-long inverted terminal repeats. S-like elements, partially complementary to the melanogaster probe, are present in D. simulans and D. mauritiana, the two species most closely related to D. melanogaster. The third Tc1-like element of D. melanogaster, Bari-1, was originally discovered in a molecular analysis of the heterochromatic h39 band (CAIZZI et al. 1993 ). It is present as an array of tandem repeats in the h39 region and as single copies in a few euchromatic polytene bands. By DNA sequencing, ORFs of heterochromatic and euchromatic copies of the element proved to be identical (CAIZZI et al. 1993 ). The apparent homogeneity of Bari-1 in the genome of D. melanogaster resembles that of Tc1 in Caenorhabditis elegans and of mariner in D. mauritiana rather than the variability in length and sequence of the D. melanogaster HB1 and S elements. Bari-1 is present in all strains of D. melanogaster and D. simulans so far analyzed (CAGGESE et al. 1995 ). This suggests a much wider diffusion in nature than that of the S elements.

To further study the evolution of the Bari-1 element and its phylogenetic relationship with other Tc1-like elements we carried out hybridization experiments to detect the presence of Bari-1 homologues in species representative of the Sophophora and Drosophila subgenera. DNA sequence analysis of elements isolated from all species of the melanogaster complex and from D. erecta and D. diplacantha suggests an evolutionary origin of these elements by diversification of a common ancestor element in different lineages.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Fly stocks:
Lines of D. sechellia, D. erecta, D. orena, D. teissieri, and D. simulans (Bordeaux strain) were obtained from the collection of R. COSTA, University of Padova, Italy. All other strains were obtained from the National Drosophila Species Resource Center, Bowling Green State University, Bowling Green, Ohio. The species examined for the presence of sequences homologous to Bari-1 are listed in Figure 2.

View larger version (41K): In this window In a new window Download PPT slide   Figure 1. Genomic analysis of Bari-1 elements in species of the melanogaster complex. (A) DNAs from D. melanogaster (Canton-S, lanes 1, 5, 9), D. simulans (Bordeaux, lanes 2, 6, 10), D. mauritiana (lanes 3, 7, 11), and D. sechellia (lanes 4, 8, 12) were digested with HindIII (lanes 1–4), HindIII plus SmaI (lanes 5–8) or HindIII plus BglII (lanes 9–11). The amount of D. melanogaster DNA loaded was 10 times less than that of DNAs from other species to better visualize the position of the 1.7-kb HindIII band deriving from the heterochromatic cluster and the identical size of the hybridizing fragments from double digestions. (B) Restriction map of the D. melanogaster Bari-1 element showing the extension of the fragment used as a probe in A.

View larger version (52K): In this window In a new window Download PPT slide   Figure 2. Species examined for the presence of sequences that hybridize with Bari-1. The number of + signs indicates the relative strength of the hybridization signal. F, one or more bands were faintly discernible on the blot after long exposure; -, a complete absence of signal.

DNA blotting experiments:
Genomic DNA for Southern blot hybridization experiments was prepared from 50 to 100 flies immediately after the stock was received. DNA extraction, restriction digestion, gel electrophoresis, and transfer to nylon membranes were performed according to MANIATIS, FRITSCH and SAMBROOK 1982 . The 1.3-kb HindIII-BglII internal fragment of Bari-1 was labeled to high specific activity with [³²P]dATP by the method of FEINBERG and VOGELSTEIN 1983 . The hybridization mixture was 0.8 M NaCl, 0.05 M Na₂PO₄, 0.8% SDS, 0.005 M EDTA, and 0.1% sodium pyrophosphate. Two filter hybridization and washing conditions were used: (1) high stringency conditions (hybridization at 67° and final wash in 0.1x SSC, 0.1% SDS at 65°) for DNAs from species of the melanogaster complex; (2) reduced stringency conditions (hybridization at 60° and final wash in 0.5x SSC, 0.1% SDS at 56°) for DNAs from other species. Filters hybridized under reduced stringency conditions were exposed to X-ray films for 16–60 hr.

Genomic DNA libraries and cloning of PCR fragments:
Lambda genomic libraries of D. simulans, D. erecta, and D. diplacantha were constructed using the Lambda GEM-12 XhoI half-site arms cloning system (Promega, Madison, WI) following the protocol supplied by the manufacturer. In each case, roughly 30,000 recombinant phages were screened with the same probes and conditions used in the Southern blot experiments. Positive plaques were subjected to three rounds of purification. Appropriate restriction fragments recognized by the Bari-1 probe were gel eluted and ligated to the pUC19 vector for further analysis. Bari-1 elements from D. mauritiana and D. sechellia were isolated by PCR amplification using 50-ng genomic DNA as template, the two 26-bp terminal inverted repeats as primers (the inverted repeats have four mismatches), and 1 unit of Taq DNA polymerase (Promega) in the following conditions: 1', 95°; 1', 60°; 1', 72° for 35 cycles and 5', 72° as final extension. With both DNA templates a single 1.7-kb band was obtained which was gel-purified and ligated to the pGEM-T vector (Promega) for cloning.

DNA sequencing and computer analysis:
Plasmid DNAs harboring a complete Bari-1 element from each species of the melanogaster complex, or subclones obtained on the basis of the restriction maps shown in Figure 6, were sequenced using the appropriate vector primers. Synthetic primers (Genset, France) constructed on the basis of the published Bari-1 sequence (AC X67681) were also used. Sequencing of D. erecta and D. diplacantha plasmids was mainly performed by subcloning fragments generated on the basis of their restriction maps and using pUC vector primers. To complete the sequence of Er-1 and Di-3 clones three specific synthetic primers were also constructed. DNA sequencing was carried out with the dideoxy chain termination method (SANGER et al. 1977 ) using either Sequenase or Thermosequenase (Amersham, Arlington Heights, IL) cycle sequencing. Comparisons of homologues were done using the 1991 GCG Sequence Analysis Software Wisconsin Package V7. The sequence flanking the Bari-1 homologue in the Di-7 clone was identified as a copia-like element (70% similarity) in a Blast search of the EMBL databank. The DNA sequence of Di-3 and Er-1 clones have been archived under accession numbers Y13852 and Y13853, respectively.

View larger version (62K): In this window In a new window Download PPT slide   Figure 3. Genomic analysis of the species containing Bari-1 homologues. (A) Species from melanogaster group: 1, D. teissieri; 2, D. erecta; 3, D. orena; 4, D. lutescens; 5, D. takahashii; 6, D. eugracilis; 7, D. mimetica; 8, D. auraria; 9, D. birchii; 10, D. diplacantha; 11, D. serrata; 12, D. ananassae; 13, D. malerkotliana; 14, D. elegans. D. melanogaster, D.m., and D. virilis, D.v. DNAs were used to establish the strength of the hybridization signal. Exposure was overnight. (B) Comparison of hybridization intensity (species of the Sophophora subgenus, lanes 1–6 and species of the Drosophila subgenus lanes 7–17). 1, D. bicornuta; 2, D. lacteicornis; 3, D. triauraria; 4, D. vulcana; 5, D. equinoxialis; 6, D. tropicalis; 7, D. robusta; 8, D. melanica; 9, D. gibberosa; 10, D. arizonensis; 11, D. meridiana rioensis; 12, D. mojavensis baja; 13, D. mulleri; 14, D. peninsularis; 15, D. mercatorum; 16, D. repleta; 17, D. eohydei. Exposure was 24 hr. All DNAs were HindIII digested. The probe was the same as in Figure 1. Note that the 2.3 kb-band in lanes 10–17 of B is shared by all species of the repleta group.

View larger version (21K): In this window In a new window Download PPT slide   Figure 4. Relationships between genera, subgenera, groups and subgroups of the Drosophila species utilized in this study. Only the melanogaster group is detailed, and it is drawn according to LACHAISE et al. 1988 . Asterisks indicate the Drosophila species from which Bari-1 elements were sequenced.

View larger version (34K): In this window In a new window Download PPT slide   Figure 5. Alignment of eleven Bari-1 sequences from the melanogaster species complex. Only differing sites are shown. The numbers at the top indicate the positions of variations with respect to the reference D.mel 47D sequence. Single nucleotide deletions and insertions are indicated by - and + signs respectively. Nucleotide and, when applicable, amino acids variations are boldfaced. # indicates a frameshift and *, a stop codon. indicates in the D.mel 91F sequence the presence of the 18-bp TTTATCATCTTATCTTAT insertion, in the D.mel 41AB sequence the substitution of the GTTGAGTG with AA, and in the D.sim S1 sequence that CTGTTC is deleted.

View larger version (22K): In this window In a new window Download PPT slide   Figure 6. Structure of Bari-1 elements of D. erecta and D. diplacantha. The physical maps of Bari-1 elements isolated from D. erecta (Er-1, Er-4, Er-5) and from D. diplacatha (Di-1, Di-3, Di-7) are compared with the restriction map of the D. melanogaster Bari-1 element (bottom line). The structure of the long inverted terminal repeat (large arrows) and the presence of defective transposases (gray boxes) in D. erecta and D. diplacantha clones were revealed by DNA sequencing.

In situ hybridization:
Salivary gland chromosomes from third instar larvae of D. erecta were prepared essentially as described in PARDUE 1986 . The plasmid pEr/PK0.3, carrying the 380-bp PstI-KpnI fragment of the Er-1 element, was labeled by nick-translation with the fluorescent Cy3-dCTP precursor (Amersham). Chromosomes were stained with DAPI. Digital images were obtained using a Leica DMRXA epifluorescence microscope equipped with a cooled CCD camera (Princeton Instruments, NJ). Gray-scale images obtained separately recording Cy3 and DAPI fluorescence by specific filters were computer colored and merged for the final image using the Adobe Photoshop software.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Genomic analysis of Bari-1 elements in species of the melanogaster complex:
In a previous paper we showed that Bari-1 elements were present in all stocks analyzed from the two cosmopolitan sibling species D. melanogaster and D. simulans (CAGGESE et al. 1995 ). A striking difference between the two sibling species is the presence of a cluster of roughly 80 copies of Bari-1 elements in the heterochromatic band h39 in every D. melanogaster strain analyzed. This cluster was not observed in any D. simulans strain. These findings are suggestive of the presence of Bari-1 elements before the evolutionary splitting of the two species and of an origin of the h39 cluster of D. melanogaster after the separation of the two species but before the world-wide diffusion of D. melanogaster. We have tested this suggestion in D. mauritiana and D. sechellia, the two remaining species of the melanogaster complex, which are endemic to the Mauritius Island and to the Seychelles Archipelago, respectively.

Figure 1 shows the results of a Southern blot analysis of the genomic DNAs from the four species of the melanogaster complex when probed with an internal Bari-1 fragment after single and double digestions. Digestion with HindIII, which recognizes a restriction site present only once in the Bari-1 sequence, produces in D. melanogaster (Canton-S strain) a strong 1.7-kb band deriving from the monomerization of the 80 heterochromatin clustered copies, plus eight bands coming from single copies of the element inserted into different regions of the genome. D. simulans, D. mauritiana, and D. sechellia only produce discrete hybridization bands of similar intensity. Analogous patterns of hybridization are observed after digestion with BglII, SmaI, or KpnI (data not shown). These results provide further evidence that a cluster of tandem repeats is present only in D. melanogaster while single elements inserted in different genomic regions are present in all the four species of the melanogaster complex. From the single digestion patterns it is also possible to estimate the number of elements present in each genome. D. simulans, Bordeaux strain, contains 15–18 Bari-1 copies, the analyzed D. mauritiana strain contains six copies, and the analyzed D. sechellia strain contains three copies.

HindIII-SmaI or HindIII-BglII double digestions of all analyzed DNAs produced a main hybridization band at the same molecular weight, suggesting that Bari-1 is conserved in structure in all species of the complex. Hybridization bands appearing at higher molecular weights are presumably due to polymorphisms in one of the restriction sites used. From the intensity of the hybridization bands and from the size of fragments produced by double digestions we conclude that most Bari-1 copies present in D. simulans, D. mauritiana, and D. sechellia are very similar to the copies present in D. melanogaster.

Taxonomic distribution of Bari-1 elements:
The presence of sequences homologous to Bari-1 in many species of the Sophophora and Drosophila subgenera and in Zaprionus tuberculatus was investigated by Southern blotting experiments under conditions of moderate stringency. The results obtained with respect to 87 different species are summarized in Figure 2 and representative examples of the gel hybridizations are shown in Figure 3A and Figure B. The phylogenetic relationship of the species analyzed are shown in Figure 4. According to the banding patterns observed, and taking into account the relative amounts of DNA loaded on the gel, each species was assigned in Figure 2 a number of plus signs (+, ++, or +++) proportional to the intensity of the signal when hybridization bands were clearly visible after short exposure, the letter "F" when faint bands appeared only after long exposure, or the minus (-) sign when no signal was observed under either condition. Within the melanogaster subgroup only D. yakuba apparently lacks Bari-1 elements. Except for D. ficusphila, all species of each group of the Sophophora subgenus produced hybridization signals, although of variable intensity. Among species of the Drosophila subgenus, hybridization was observed in species of the virilis, robusta, and melanica groups and in some species of the other groups. In general, cross hybridization between species was less intense in the Drosophila subgenus than in the Sophophora subgenus. Z. tuberculatus, belonging to a different genus, also produced faint hybridization bands. These results indicate that, unlike the S elements, Bari-1 is widespread in the Drosophila genus.

DNA analysis of Bari-1 elements from species of the melanogaster complex:
The sequence of a D. melanogaster Bari-1 element cloned from the insertion site at polytenic band 47D has been previously reported (CAIZZI et al. 1993 ). In order to define the level of homogeneity of Bari-1 elements present in different species, we have cloned and sequenced elements from each species of the melanogaster complex. Bari-1 elements of D. simulans were isolated by screening genomic libraries, whereas Bari-1 clones from D. mauritiana and D. sechellia were obtained by PCR amplification using two 26-bp primers representing the inverted terminal repeat of the D. melanogaster element. We present here the sequence of five additional D. melanogaster clones: two independent heterochromatic clones (clones h19/5 and h1/1, both coming from the cluster located at the h39 region), and three clones from euchromatic regions 41AB, 55F and 91F. Heterochromatic clones can be distinguished from the euchromatic ones since they are present in phages carrying at least five Bari-1 copies in tandem repeat. Figure 5 summarizes the nucleotide differences between the new sequences obtained and the Bari-1 element from D. melanogaster band 47D taken as a reference.

The Bari-1 copies at 55F and at 91F were singled out for cloning and sequencing because they appear not to be mobile (CAGGESE et al. 1995 ). Since both elements have normal terminal repeats, their lack of mobility is likely to be due to some effect of the flanking regions or to differences in internal sequences. In contrast, both the heterochromatic clones sequenced lack the first two nucleotides (CA) of the terminal inverted repeats. It should be noted that the sequences of the heterochromatic clones, although reported as coming from isolated elements, are really composed of parts of two adjacent elements since the 1.7-kb HindIII fragments, subcloned from two different phages into the pUC vector, represent the joining of the right terminus of one element (from the HindIII site to the right inverted repeat) to the left part of another one (from the left terminus to the adjacent HindIII site). Therefore, the sequences of heterochromatic clones given in Figure 5 do not represent the entire sequence of a single element and no nucleotide difference can be unambigously assigned to a specific heterochromatic copy. Nevertheless, the heterochromatic elements clearly are very similar in nucleotide sequence to the euchromatic ones.

The two recovered D. simulans Bari-1 clones both have the typical TA duplication at the ends of their inverted repeats. Since the flanking regions are different, they are from two different (but unidentified) chromosomal sites. Both clones have inverted repeats identical to the reference sequence and possess intact ORFs.

One of the two sequenced Bari-1 elements of D. mauritiana, D.mau 9b, has a frameshift in the coding region, whereas the D.mau 6 clone has eight nucleotide substitutions, five of which are in the coding region. Finally, the D.sec 0 element from D. sechellia has a stop codon in the ORF.

The comparison of the 11 Bari-1 elements sequenced does not permit construction of a clear phylogenetic relationship since very few informative sites were found. In general, the nucleotide variations appear randomly distributed over the length of the element and, within the ORFs, are equally distributed in the three codon positions.

The pairwise divergence values from a comparison of Bari-1 sequences from the four species are shown in Table 1. Within D. melanogaster elements the highest divergence is between D.mel 55F and D.mel 19/5, both non-autonomous elements. Comparison between elements from any two species of the complex yields low divergence values with a maximum of 1% between D. sechellia and D. mauritiana. This low divergence makes Bari-1 similar to the mariner elements (MARUYAMA and HARTL 1991 ; CAPY et al. 1992 ) rather than to the Tc1-like S elements (MERRIMAN et al. 1995 ).

Cloning and characterization of Bari-1 homologues from D. erecta and D. diplacantha:
In order to investigate whether the structure and the DNA sequence of the Bari-1 elements were also conserved in species relatively distant from the melanogaster complex, we isolated Bari-1 homologues from D. erecta, a species of the melanogaster subgroup, and from D. diplacantha, a species of the montium subgroup. These species were chosen for detailed analysis because of the high level of hybridization with the Bari-1 probe observed in the Southern blotting experiments. In fact, when lambda libraries were screened with an internal Bari-1 fragment several positive plaques were found, most containing only a limited portion of the element. From each species three random clones were characterized. Figure 6 shows a preliminary restriction analysis of these clones and summarizes the most relevant features revealed by sequencing. The D. erecta Er-1 clone is the only one that contains a complete element flanked on both sides by the usual TA dinucleotide. The D. diplacantha Di-3 clone appears almost complete, lacking only a few nucleotides at the left end. The remaining cloned elements from D. erecta and D. diplacantha appear either to have been truncated during the cloning procedure or to be flanked by unrelated sequences. In one case, the Di-7 clone, the flanking DNA is related (70% similarity) to copia like sequences present in the databases. Comparison of the physical maps of the cloned elements shows that all of them share some restriction sites and that they are related to each other.

Nucleotide sequence variations in D. erecta and D. diplacantha elements:
The relationship between the isolated D. erecta and D. diplacantha elements was confirmed by sequencing. Figure 7 aligns the nucleotide sequences of Er-1 and Di-3, the two almost complete elements, with the reference D. melanogaster Bari-1 copy. Several deletions and insertions affecting both coding and noncoding regions are present in the two elements and in the sequences of the incomplete clones. Thus, it appears that most Bari-1 homologues present in the genome of D. erecta and D. diplacantha are defective elements. Table 2 shows pairwise similarities of the sequences compared. The three clones from D. erecta have on average 95% identity to each other and approximately 70% identity with Bari-1 of D. melanogaster. Among clones from D. diplacantha we can distinguish a subclass, represented by the Di-3 sequence, which is 93% similar to elements of D. erecta and 70% to Bari-1, from an apparently different one represented by clones Di-1 and Di-7, which are only ~70% similar to Di-3 and to the D. erecta elements. Interestingly, elements of the latter subclass show higher similarity, ~88%, to the D. melanogaster element. This evidence suggests that at least two different members of the Bari-1 family exist in D. diplacantha.

View larger version (75K): In this window In a new window Download PPT slide   Figure 7. Comparison of the Bari-1 D. melanogaster sequence with its homologues from D. diplacantha and D. erecta. The complete DNA sequences of Di-3 (AC Y13852) and Er-1 (AC Y13853) clones are aligned to the standard D.mel 47D sequence. The differing nucleotides are shown in lower case letters. (—) Deletion; (.) introduced to optimize the alignments. The start and stop codon of the Bari-1 standard transposase are boldfaced. (¶) A cloning site.

Optimal alignment of the D. diplacantha Di-3 ORF with the D. melanogaster ORF is obtained by introducing three single nucleotide gaps and deleting two nucleotides. The best alignment of the Er-1 ORF with the Bari-1 ORF is obtained by introducing three gaps of 1, 15 and 23 nucleotides respectively. The deduced proteins putatively encoded by Di-3 and Er-1 comprise 338 and 325 amino acids, respectively. As shown in Table 3, the deduced Di-3 protein is 86% identical to the Er-1 protein and 67% identical to the Bari-1 protein. Such high levels of identity clearly demonstrate that Er-1, Di-3 and Bari-1 belong to the same family of transposable elements.

The inverted repeats of Er-1 and Di-3 elements:
Careful analysis of the termini of the Er-1 element, the only cloned element that apparently maintains both ends (see Figure 6), revealed a different organization with respect to the ends of Bari-1. The Er-1 element has 254-bp terminal inverted repeats that are 96.4% identical to each other, with six gaps. In the Di-3 element a similar long inverted repeat could also be postulated since its left end, although presumably truncated during the cloning, still possesses a 198-bp sequence that is repeated in opposite orientation at its right end (97.9% identity without gaps). The 254 bp at the right end of the Er-1 and Di-3 elements show 96.4% identity without gaps. Within each long inverted terminal repeat two 18-bp direct repeats are present: the outer direct repeat starts 8 bp from the end of the element and the inner one terminates the inverted repeat. This structure, called IR-DR, is also found in other Tc1-like transposable elements of invertebrates (FRANZ and SAVAKIS 1991 ; MERRIMAN et al. 1995 ; PETROV et al. 1995 ) as well as in Tc1-like transposable elements from some fishes (RADICE et al. 1994 ; IZSVAK et al. 1995 ).

Table 3 lists all the invertebrate Tc1-like transposases which have been compared to the reconstructed transposase of D. diplacantha and D. erecta elements. While the overall homology of the ORFs strongly relates Er-1 and Di-3 to Bari-1, the structure of their terminal repeats is more related to other Tc1-like elements than to Bari-1. However, when the 26-bp inverted repeat of Bari-1 is compared with the Er-1 26-bp terminal sequence comprising the terminal 8 bp of the element plus the 18-bp outer direct repeat, a strong similarity is found. Moreover, the last eight nucleotides of the outer direct repeat are identical to the corresponding nucleotides of the S elements. The relationships among terminal sequences are shown in Figure 8.

View larger version (19K): In this window In a new window Download PPT slide   Figure 8. Comparison of terminal repeats of Bari-1 and its Er-1 D. erecta homologue. The upper diagram represents the long inverted repeats at both ends of the Er-1 element. The small arrows within each box represent the outer and inner direct repeats. The terminal 26 nucleotides of Er-1 and Bari-1 are compared to show the high level of similarity. Identical nucleotides are shown by asterisks. The 18-bp sequences of the outer direct repeats of Er-1 are in uppercase letters. The underlined nucleotides perfectly match the corresponding nucleotides in the terminal repeats of the S element.

Copy number and distribution of Bari-1 related elements in D. erecta and D. diplacantha:
We have estimated the copy number of Bari-1-like elements in D. erecta and D. diplacantha by Southern hybridization of genomic DNAs probed under stringent conditions with the 380-bp PstI-KpnI internal fragment of the Er-1 element. Within this region Er-1 and Di-3 share 96% nucleotide identity. From the number of hybridization bands we estimate that 12–15 copies of the element are present in D. diplacantha and 18–25 in D. erecta (Figure 9A). Furthermore, by in situ hybridization experiments in D. erecta, the elements appear to be located in the pericentromeric heterochromatin and/or the Y chromosome. Heavy labeling sites are also present on chromosome 4 (Figure 9B).

View larger version (72K): In this window In a new window Download PPT slide   Figure 9. Copy number and in situ hybridization of D. erecta elements. (A) Southern blot of genomic DNAs from D. diplacantha (D) and D. erecta (E) digested with PstI and probed with the O.35-kb PstI/KpnI internal fragment of Er-1 clone. Bars on the left represent the position of HindIII marker fragments. (B) Salivary glands chromosomes of D. erecta showing the hybridization of the same probe used in A over chromocenter (C) and on the fourth chromosome (4).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Bari-1-like elements in species of the melanogaster group:
Within the melanogaster species group Bari-1-like elements show several interesting features. First, the elements are widely distributed in the Sophopora subgenus (Figure 2), with only a few apparent cases of taxonomic discontinuity. Second, different copies of the element are extremely similar, at least within the species of the melanogaster complex. All Bari-1 copies so far isolated from D. melanogaster, D. simulans, D. mauritiana, and D. sechellia are similar in size and physical map. Third, sequences isolated from D. erecta and D. diplacantha, two species phylogenetically quite distant from the melanogaster complex, are clearly related to the melanogaster Bari-1 elements. However, their terminal repeats have a different organization.

The taxonomic distribution of many families of transposable elements can be explained by vertical transmission because the phylogeny of the transposon closely follows the phylogeny of the host species (see, for example, MARUYAMA and HARTL 1991 ; RADICE et al. 1994 ). However, both vertical and horizontal transmission appear to have contributed to the distribution of some families of mobile elements (see, for example, ABAD et al. 1991 ; ROBERTSON 1993 ; LOHE et al. 1995 ). The wide distribution of Bari-1-like elements in the Sophophora subgenus could simply indicate that they have an ancient origin and have been able to remain in the genome of this group of organisms for a long time.

Our data indicate that in Drosophila two subclasses of Bari-1-like elements exist, strongly related by sequence similarity but differing by the structure of their terminal repeats. The IR-DR structure is found in the subclass represented by the Er-1 and Di-3 elements and the simple IR structure is found in the subclass represented by D. melanogaster elements. Similar observations have been reported for fish Tc1-like elements (RADICE et al. 1994 ; IZSVAK et al. 1995 ). However, in several fish species both types of elements coexist and IR-DR elements from different species are more similar than elements with simple IR termini within a species. Subfamilies of the mariner element with independent evolutionary histories have been also reported (ROBERTSON and LAMPE 1993 ; LOHE et al. 1995 ). Peculiar to the Bari-1-like elements of Drosophila, however, is the higher level of nucleotide similarity between IR-DR and non IR-DR elements. Moreover, no Drosophila species has as yet been identified possessing elements of both Bari-1-like subclasses. Assuming that IR-DR Bari-1-like elements were present in an ancestor of the melanogaster subgroup, then they must have been lost in the lineage leading to the melanogaster complex, presumably by a stochastic process (LOHE et al. 1995 ), while being maintained in D. erecta and D. diplacantha.

The Bari-1 elements that lack IR-DR are present only in the melanogaster species complex. Thus, either they were not present in the ancestor of the melanogaster subgroup or they were lost in the erecta-orena and yakuba-teissieri lineages. Their existence in all strains of D. melanogaster, D. simulans, D. mauritiana, and D. sechellia suggests that they were present before these species split. However, DNA sequencing reveals a low rate of divergence among the elements in the melanogaster complex. Repeated invasion of Bari-1 in the common ancestor of the simulans/mauritiana/sechellia lineage and in the ancestor of the melanogaster lineage could explain the low observed level of divergence. Interestingly, the rate of divergence for Bari-1 elements falls in the range reported for mariner elements in the simulans complex (MARUYAMA and HARTL 1991 ; CAPY et al. 1992 ). Phylogenetic results suggest that the mauritiana subfamily of mariner was probably present before the melanogaster subgroup diverged and was then lost in some lineages. Amplification of a few ancestral elements is thought to be responsible for the sequence homogeneity. We cannot exclude that in the melanogaster complex Bari-1 had a rapid amplification similar to the mariner elements.

CAPY et al. 1994 have pointed out several alternatives that could explain the seemingly strange phylogeny of many transposable elements. However, our data are still limited to too few species to conclude which scenario best explains the evolution of Bari-1 homologues in Drosophila. The isolation and sequencing of elements from more species may elucidate to what extent vertical inheritance and horizontal transmission have contributed to the strikingly wide diffusion of the element in Drosophila.

Relationship between the Bari-1 inverted repeats and the inverted repeats of Er-1 and Di-3:
Although strongly related to Bari-1 by sequence similarity, the Er-1 and Di-3 elements are surprisingly different from this element in the structure of their terminal repeats. Bari-1 possesses short 26-bp inverted repeats while Er-1 and Di-3 possess a terminal IR-DR structure. This type of structure is present in many families of Tc1-like elements, but it is likely not to be an essential feature since it is not found in other members of the Tc1 superfamily (IVICS et al. 1996 ). As shown in Figure 8, the terminal 26 bp of Er-1, which include the 18-bp outer direct repeats of the IR-DR structure, are almost identical to the terminal 26 bp of Bari-1, suggesting that the 26 bp Bari-1 sequence originated from the Er-1 IR-DR structure. The loss of the long inverted repeat in Bari-1 cannot be explained simply by an unequal recombination event between the direct repeats within each long inverted repeat because the product of such an event would be about 450 bp shorter than the original element; however, Er-1 and Bari-1 are similar in length.

At present it is not clear why two kinds of termini exist in the Tc1-like superfamily nor what mechanisms are responsible for maintaining the inverted repeats in the transposon (for review see PLASTERK 1996 ). The sequences of the direct repeats in IR-DR Tc1-like elements from very distant species are clearly related, while the surrounding sequences diverge substantially (IZSVAK et al. 1995 ; MERRIMAN et al. 1995 ). The DR motifs represent the binding site of the transposase (VOS et al. 1993 ; IVICS et al. 1997 ) and in Tc3, which has an internal repeat in a different position with respect to the IR-DR Tc1 subclass, the removal of the internal DR binding site does not reduce the rate of transposition (COLLOMS et al. 1994 ). This suggests that the internal repeats do not have a functional significance directly linked to the mobility of the element, but they could have regulating purposes (IVICS et al. 1997 ).

The relationship between the defective elements of D. erecta and the homogeneous elements of the melanogaster complex:
All elements as yet isolated from D. erecta and D. diplacantha are defective. Moreover, the 18–25 copies in the D. erecta genome are localized either in the chromocenter or on chromosome 4, which, by analogy with the D. melanogaster chromosome 4, can be considered heterochromatic (MIKLOS et al. 1988 ). Thus, it is reasonable to suppose that the D. erecta Bari-1 homologues are the evolutionary relics of an autonomous ancestor possessing IR-DR terminal structures. The lack of any relics of the element in euchromatic regions of the genome may be the result of loss by genetic drift or elimination by natural selection (reviewed in CHARLESWORTH et al. 1994 ). The defective heterochromatic elements are unlikely to have been present before the evolutionary splitting of the melanogaster subgroup, since they are not found in any species of the subgroup except D. erecta. According to the model proposed by HARTL et al. 1997 the dynamics of transposon mobility and mainteinance within a host genome lead inevitably to loss of the element, whose survival is only possible by invasion of a new host. We suggest that the transposition machinery of a primitive IR-DR mobile element was error-prone in D. erecta, and possibly also in D. diplacantha, due, for example, to the inability of the transposase to recognize the termini of the element correctly because of identical direct repeats present within each long inverted repeat. From an evolutionary standpoint, therefore, survival of the transposon required either modification of the transposase so as to acquire more specificity, or modification of the sequence responsible for the transposase's error-propensity. This second strategy may have been responsible for the generation of Bari-1-like elements that lack terminal IR-DR.

	ACKNOWLEDGMENTS

We thank RODOLFO COSTA or providing some of the species used in this work, MARIANO ROCCHI for helping us with the analysis and managing of digital images from fluorescence in situ hybridization, and M. SIMMONS and the anonymous reviewers for helpful comments. We are grateful to NICOLA DITURI for expert technical assistance. This work was supported by funds from Ministero dell'Università e della Ricerca Scientifica e Technologica (MURST; ex 40%, 1996) and from Consiglio Nazionale delle Richerche (N. 96.03264.CT04) to R.C.

Manuscript received December 11, 1997; Accepted for publication June 8, 1998.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ABAD, P., C. QUILES, S. TARES, C. PIOTTE, and P. CASTAGNONE-SERENO et al., 1991  Sequences homologous to the Tc(s) transposable elements of Caenorhabditis elegans are widely distributed in the phylum nematoda. J. Mol. Evol. 33:251-258[Medline].

BERG, D., and M. M. HOWE, 1989 Mobile DNA. American Society for Microbiology, Washington, DC.

BREZINSKY, L., G. V. L. WANG, T. HUMPHREYS, and J. HUNT, 1990  The transposable element Uhu from Hawaiian Drosophila—member of the widely dispersed class of Tc1-like transposon. Nucleic Acids Res. 18:2053-2059[Abstract/Free Full Text].

BRIERLY, H. L. and S. S. POTTER, 1985  Distinct characteristics of loop sequences of two Drosophila foldback transposable elements. Nucleic Acids Res. 13:485-500[Abstract/Free Full Text].

CAGGESE, C., S. PIMPINELLI, P. BARSANTI, and R. CAIZZI, 1995  The distribution of the transposable element Bari-1 in the Drosophila melanogaster and Drosophila simulans genomes. Genetica 96:269-283[Medline].

CAIZZI, R., C. CAGGESE, and S. PIMPINELLI, 1993  Bari-1, a new transposon-like family in Drosophila melanogaster with a unique heterochromatic organization. Genetics 133:335-345[Abstract].

CAPY, P., A. KOGA, J. R. DAVID, and D. L. HARTL, 1992  Sequence analysis of active mariner elements in natural populations of Drosophila simulans. Genetics 130:499-506[Abstract].

CAPY, P., D. AUXOLABEHERE, and D. LANGIN, 1994  The strange phylogenies of transposable elements: are the horizontal transfers the only explanation? Trends Genet. 10:7-12[Medline].

CHARLESWORTH, B., P. SNIEGOWWSKI, and W. STEPHAN, 1994  The evolutionary dynamics of repetitive DNA in eukaryotes. Nature 371:215-220[Medline].

COLLOMS, S. D., H. G. VAN LUENEN, and R. H. PLASTERK, 1994  DNA binding activities of the Caenorhabditis elegans Tc3 transposase. Nucleic Acids Res. 22:5548-5554[Abstract/Free Full Text].

FEINBERG, A. P. and B. VOGELSTEIN, 1983  A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132:6-13[Medline].

FINNEGAN, D. J., 1989  Eukaryotic transposable elements and genome evolution. Trends Genet. 5:103-107[Medline].

FRANZ, G. and C. SAVAKIS, 1991  Minos, a new transposable element from Drosophila hydei, is a member of the Tc1-like family of transposons. Nucleic Acids Res. 19:6646[Free Full Text].

HARTL, D. L., E. H. LOZOVSKAYA, D. I. NURMINSKY, and A. L. LOHE, 1997  What restricts the activity of mariner-like transposable elements? Trends Genet. 13:197-201[Medline].

HENIKOFF, S. and R. H. A. PLASTERK, 1988  Related transposons in C. elegans and D. melanogaster. Nucleic Acids Res. 16:6234[Free Full Text].

IVICS, Z., Z. IZSVÀK, A. MINTER, and P. B. HACKETT, 1996  Identification of functional domains and evolution of Tc1-like transposable elements. Proc. Natl. Acad. Sci. USA 93:5008-5013[Abstract/Free Full Text].

IVICS, Z., P. B. HACKETT, R. H. PLASTERK, and Z. IZSVÀK, 1997  Molecular reconstruction of Sleeping Beauty, a Tc1-like transposon from fish, and its transposition in human cells. Cell 91:501-510[Medline].

IZSVÀK, Z., Z. IVICS, and P. B. HACKETT, 1995  Characterization of a Tc1-like transposable element in zebrafish (Danio rerio). Mol. Gen. Genet. 247:312-322[Medline].

KIDWELL, M. G., 1992a  Horizontal transfer. Curr. Opin. Genet. Dev. 2:868-873[Medline].

KIDWELL, M.G., 1992b  Lateral transfer in natural populations of eukaryotes. Annu. Rev. Genet. 27:235-256[Medline].

KIDWELL, M. G., 1993  Evolutionary biology. Voyage of an ancient mariner. Nature 362:202[Medline].

LACHAISE, D., M. CARIOU, J. R. DAVID, F. LEMEUNIER, and L. TSACAS et al., 1988  Historical biogeography of the Drosophila melanogaster species subgroup. Evol. Biol. 22:159-227.

LAMPE, D. J., M. E. A. CHURCHILL, and H. M. ROBERTSON, 1996  A purified mariner transposase is sufficient to mediate transposition in vitro. EMBO J. 15:5470-5479[Medline].

LOHE, A. R., E. N. MORIYAMA, D. A. LIDHOLM, and D. L. HARTL, 1995  Horizontal transmission, vertical inactivation, and stochastic loss of Mariner-like elements. Mol. Biol. Evol. 12:62-72[Abstract].

MANIATIS, T., E. F. FRITSCH and J. SAMBROOK, 1982 Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

MARUYAMA, K. and D. L. HARTL, 1991  Evolution of the transposable element mariner in Drosophila species. Genetics 128:319-329[Abstract].

MCDONALD, J. F., 1993  Evolution and consequence of transposable elements. Curr. Opin. Genet. Dev. 3:855-864[Medline].

MERRIMAN, P. J., C. D. GRIMES, J. AMBROZIAK, D. A. HACKETT, and P. SKINNER et al., 1995  S elements family of Tc1-like transposons in the genome of Drosophila melanogaster. Genetics 141:1425-1438[Abstract].

MIKLOS, G. L. G., M.-T. YAMAMOTO, J. DAVIES, and V. PIRROTTA, 1988  Micro-cloning reveals a high frequency of repetitive sequences characteristic of chromosome 4 and the ß heterochromatin of Drosophila melanogaster. Proc. Natl. Acad. Sci. USA 85:2051-2055[Abstract/Free Full Text].

PARDUE, M. L., 1986 In situ hybridization to DNA of chromosomes and nuclei, pp. 111–137 in Drosophila, A Practical Approach, edited by D. B. ROBERTS. IRL Press, Oxford.

PETROV, D. A., J. L. SCHUTZMAN, D. L. HARTL, and E. R. LOZOVSKAYA, 1995  Diverse transposable elements are mobilized in hybrid dysgenesis in Drosophila virilis. Proc. Natl. Acad. Sci. USA 92:8050-8054[Abstract/Free Full Text].

PLASTERK, R. H., 1996  The Tc1/mariner transposon family. Curr. Top. Microbiol. Immunol. 204:125-143[Medline].

RADICE, A. D., B. BUGAJ, D. H. A. FITCH, and S. W. EMMONS, 1994  Widespread occurrence of the Tc1 transposon family: properties of Tc1-like transposons from teleost fish. Mol. Gen. Genet. 244:606-612[Medline].

ROBERTSON, H. M., 1993  The mariner transposable element is widespread in insects. Nature 362:241-245[Medline].

ROBERTSON, H. M. and E. G. LAMPE, 1993  Five major subfamilies of mariner transposable elements in insects, including the Mediterranean fruit fly, and related arthropods. Insect Mol. Biol. 2:125-139[Medline].

ROBERTSON, H. M. and D. J. LAMPE, 1995  Distribution of transposable elements in arthropods. Annu. Rev. Entomol. 40:333-357[Medline].

SANGER, F., S. NICKLEN, and A. R. COULSON, 1977  DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463-5467[Abstract/Free Full Text].

VOS, J. C., H. G. A. M. VAN LUENEN, and R. H. A. PLASTERK, 1993  Characterization of the Caenorhabiditis elegans Tc1 transposase in vivo and in vitro. Genes Dev. 7:1244-1253[Abstract/Free Full Text].

VOS, J. C., I. DE BAERE, and R. H. A. PLASTERK, 1996  Transposase is the only nematode protein required for in vitro transposition of Tc1. Genes Dev. 10:755-761[Abstract/Free Full Text].