Evidence for Recent Invasion of the Medaka Fish Genome by the Tol2 Transposable Element

Corresponding author: Hiroshi Hori, Division of Biological Sciences, Graduate School of Science, Nagoya University, Nagoya 464-8602, Japan., hori{at}bio.nagoya-u.ac.jp (E-mail)

Communicating editor: M. J. SIMMONS

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Tol2 is a transposable element of the terminal-inverted-repeat class, residing in the genome of the medaka fish Oryzias latipes. The genus Oryzias contains more than 10 species for which phylogenetic relationships have previously been estimated. To infer the history of Tol2 in this genus we performed genomic Southern blots and PCR analyses of 10 of the species. It was revealed that Tol2 occurs in 2 of the 10 species (O. curvinotus and O. latipes) and that the length and the restriction map structure of Tol2 are identical in the two cases. Further, sequencing analysis revealed an extremely low level of divergence compared with that in a nuclear gene. These results suggest recent incorporation of Tol2 into one or both of the two species, implying horizontal transfer of Tol2 from one species to the other or into them both from a common source.

TRANSPOSABLE elements are repetitive sequences capable of moving from one chromosomal location to another. Several transposable elements, or transposable element families, are known to be distributed across species, genera, or even higher taxa. When the phylogeny of a transposable element is incongruent with the phylogeny of its host species, horizontal transfer of the element between species is often suggested as the explanation (for reviews, see KIDWELL 1992 ; CAPY et al. 1994 ). It is not known whether horizontal transfer is a frequent event under natural circumstances or just a rare accident. If the former is the case, it might play a significant role in the evolution of transposable elements (see MARUYAMA and HARTL 1991A ; KIDWELL 1992 ; LOHE et al. 1995 ), allowing them to survive the pressures of vertical inactivation and stochastic loss that would lead to their ultimate extinction from genomes.

Well-known examples of horizontal transfer are represented by the P element (DANIELS et al. 1990 ) and the mariner element (MARUYAMA and HARTL 1991A ) of Drosophila and the IS1 element of bacteria (LAWRENCE et al. 1992 ). With these, high sequence similarities were found among copies isolated from distantly related species. Similar observations have been reported for many elements in Drosophila, in related insects, and in bacteria. Examples from more diverse organisms should aid in inferring how frequent this phenomenon is.

We here report high nucleotide sequence similarity of a transposable element in two fish species, the most likely explanation for which is horizontal transfer. Tol2 is a terminal-inverted-repeat transposable element residing in the genome of the medaka fish Oryzias latipes (KOGA et al. 1996 ). It is 4.7 kb in length and contains four open reading frames (ORFs) having amino acid sequence similarity with members of the hAT family (CALVI et al. 1991 ; ATKINSON et al. 1993 ), a group of terminal-inverted-repeat transposable elements including hobo of Drosophila (MCGINNIS et al. 1983 ), Ac of maize (MCCLINTOCK 1948 ; FEDOROFF et al. 1983 ), and Tam3 of snapdragon (SOMMER et al. 1985 ). Features concerning the similarity, including amino acid sequence alignments, are described in KOGA et al. 1999 . The medaka fish genome contains 10 to 30 copies of Tol2 that are highly homogeneous in structure: no restriction map variation was found among more than 200 copies and no sequence variation among five randomly chosen clones (KOGA and HORI 1999 ). This situation contrasts with other hAT elements in which defective, shorter elements are common. To explain this anomaly, we have proposed three hypotheses: (1) lack of factors that are involved in the generation of defective copies; (2) less efficient amplification of defective copies; and (3) a short interval between the Tol2 invasion of the medaka fish and the appearance of evolutionary change (KOGA and HORI 1999 ). The fact that the homogeneity is seen even at the nucleotide sequence level appears to favor the last explanation. The genus Oryzias includes more than 10 species whose phylogenetic relationships have already been studied (SAKAIZUMI 1985 ; NARUSE et al. 1993 ; NARUSE 1996 ). If the third explanation is true, and, in addition, if the time span after the Tol2 invasion is short on the time scale of speciation in the genus, an inconsistency between the Tol2 phylogeny and the host phylogeny might be expected. To determine if this is the case, we first performed genomic Southern blotting and PCR to study the distribution and structure of Tol2-hybridizing sequences in 10 species of the genus Oryzias. Having found an inconsistency, we further conducted sequencing analysis of Tol2 copies and genes from their host species. The results suggested recent incorporation of Tol2 into at least one of the species.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Medaka fish:
This species (O. latipes) inhabits East Asia, including China, Korea, and Japan, and demonstrates geographical variation. According to data on isozyme frequencies, there are four regional populations: [1] Northern Japan, [2] Southern Japan, [3] Eastern Korea, and [4] China and Western Korea (SAKAIZUMI 1986 ; SAKAIZUMI et al. 1987 ). A total of 12 laboratory stocks, 3 from each population, were examined. Their original collection sites and designations were the following: [1a] Yokote, Japan; [1b] Kaga, Japan; [1c] Maizuru, Japan; [2a] Nagoya, Japan; [2b] Kasumi, Japan; [2c] Kobe, Japan; [3a] Yongchon, Korea; [3b] Sachon, Korea; [3c] Shinpyon, Korea; [4a] Shanghai, China; [4b] Maegkok, Korea; [4c] Pugang, Korea.

Other species:
Nine species of the genus Oryzias were obtained from the World Medaka Aquarium of the Nagoya City Higashiyama Zoological Garden. The nine species, in addition to O. latipes, are listed in Fig 1 with the present consensus on their phylogenetic relationships and their original collection sites. Detailed phylogenetic trees are available in NARUSE et al. 1993 and NARUSE 1996 . Swordtail fish (Xiphophorus helleri) and zebrafish (Danio rerio) were purchased from a pet shop in Nagoya. Commercially available genomic DNAs were used for assessment of the chicken and human cases.

View larger version (34K): In this window In a new window Download PPT slide   Figure 1. The 10 species of the genus Oryzias examined in the present study. (A) The original collection sites of the 10 species are shown by dots on the map. The species names are abbreviated with their first three letters. (B) NARUSE et al. 1993 and NARUSE 1996 constructed phylogenetic trees with nucleotide sequence data for the mitochondrial cytochrome b gene and the 12S ribosomal RNA gene, respectively. The scheme was redrawn from these references for the 10 species.

Analysis of genomic DNA:
Southern blotting and subsequent hybridization experiments were performed as described by KOGA et al. 1995 . Briefly, high molecular DNAs were digested with restriction enzymes, fractionated on 1.0% agarose gels, and transferred to nylon membranes. The membranes were then hybridized with ³²P-labeled probes at 65° for 16 hr. The hybridization solution contained 6x SSPE, 0.5% SDS, 50 µg/ml fragmented heterologous DNA, and 5 ng/ml probe DNA. Heterologous DNA was included as a blocking agent and salmon sperm DNA was used as long as only the two Tol2-containing species (O. curvinotus and O. latipes) were concerned. This is because similar analyses in our previous studies were conducted under this condition. In hybridization experiments including other species, yeast genomic DNA was used because salmon DNA might absorb the probe DNA if the salmon is phylogenetically closer to the origin of the probe DNA (medaka fish) than are the genome DNAs to be examined.

For the 10 species of the genus Oryzias, a 4-µg aliquot of genomic DNA was used for each gel slot. For the species outside the genus, 20 µg was used. The implications of this difference are described in the DISCUSSION.

Hybridization probes:
The Tol2 element was first identified as a DNA fragment inserted into the tyrosinase gene of an albino mutant fish (KOGA et al. 1996 ). This particular Tol2 copy is denoted Tol2-tyr (GenBank accession no. D84375). Parts of a Tol2-tyr clone generated by restriction enzyme digestion or PCR were subcloned into the pBluescript II vector. A fragment of the medaka fish tyrosinase gene was also cloned. These clones were used as probes for hybridization experiments after removal of their vector portions. The regions covered and the designations of the probes are shown in Fig 2.

View larger version (36K): In this window In a new window Download PPT slide   Figure 2. Probes used for Southern blot analysis. The tyrosinase gene region of the medaka fish is illustrated at the bottom. The open boxes indicate exons 1 to 5. The stippled box is Tol2-tyr, identified as a DNA fragment inserted in exon 5 as shown. The thick arrows in the box indicate the spans and directions of the four Tol2 ORFs. The thin arrows are internal inverted repeats of 302 bp and 303 bp. Because the direction of the Tol2 ORFs is opposite to that of transcription on the tyrosinase gene, the Tol2 portion is illustrated again by changing its direction. Parts of the tyrosinase gene and Tol2, indicated by double-headed arrows with their designations, were subcloned and used as hybridization probes. Restriction enzyme cutting sites (Bg, BglII; H, HindIII; Ps, PstI; Pv, PvuII) and the left and the right ends of the element (indicated by the letters L and R) are shown with their locations in kilobase pairs, 0 being assigned to the L site. Tn indicates the location of consecutive T's present in the first intron of Tol2. The double-headed arrow below the tyrosinase gene map is the region sequenced for construction of phylogenetic trees of fishes.

Other molecular techniques:
PCR, cloning, and sequencing were conducted as previously described (KOGA and HORI 1999 ).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Distribution of Tol2 among species:
Fig 3 illustrates autodiagrams of Southern blots to determine the presence of Tol2 in the 10 species of the genus Oryzias and other species outside the genus. Genomic DNAs were digested with BglII, which has a single cutting site in the Tol2-tyr sequence. In Fig 3A where probe Tyr-Bg (part of the medaka fish tyrosinase gene) was used, one or two hybridization signals were observed for all the fishes of the genus Oryzias, the swordtail fish, and zebrafish. In Fig 3B for which the hybridization conditions were the same as those for Fig 3A and probe L(0)R(4.7) (the entire Tol2 element) was used, strong hybridization signals, consisting of multiple bands, were observed for O. curvinotus and O. latipes. No signals were observed with the other 8 species of the genus Oryzias. For confirmation, an additional four fish samples for each of the 8 Oryzias species were examined by the same method as for Fig 3B and no bands due to Tol2 were observed (data not shown).

View larger version (73K): In this window In a new window Download PPT slide   Figure 3. Southern blot hybridization to examine the distribution of Tol2 among species. Genomic DNAs were digested with BglII. Two sets of samples were fractionated on 1.0% agarose gels and transferred to nylon membranes. One was hybridized with probe Tyr-Bg (part of the tyrosinase gene) (A) and the other with probe L(0)R(4.7) (the entire Tol2 element) (B). The fish [2c] (see MATERIALS AND METHODS) was used as a sample of O. latipes. The sizes and mobilities of the size marker DNA fragments are indicated along the left margin.

Genomic DNAs from swordtail fish and humans also lacked bands due to the Tol2 element. However, in the lanes for zebrafish and chicken, multiple weak signals were observed.

Distribution and variation of Tol2 in O. curvinotus and O. latipes:
Fig 4 shows an autodiagram of a Southern blot to determine the presence of Tol2 in an O. curvinotus sample and 12 samples of O. latipes. All 13 samples exhibited multiple bands hybridizing to probe L(0)Pv(0.5). Although this probe contains the left terminal inverted repeat (17 bp), it does not yield a signal for the right terminal inverted repeat (19 bp), as confirmed in a preliminary experiment using a Tol2 clone as the target (data not shown). The numbers of bands observed are shown in Fig 4. In the O. latipes samples, they are distributed in the range between 10 and 25, with an average of 16. This range is not very different from that observed for a single local population (11 to 30, with an average of 19; KOGA and HORI 1999 ).

View larger version (98K): In this window In a new window Download PPT slide   Figure 4. Southern blot hybridization to examine the presence of Tol2 in the samples of O. curvinotus and O. latipes. Genomic DNAs were digested with PvuII. The hybridization probe was L(0)Pv(0.5). The designation "cur" stands for the O. curvinotus sample. [1a], [1b], etc., are the O. latipes samples (see MATERIALS AND METHODS). The numbers shown in parentheses are the numbers of bands obtained by scanning the hybridization membrane along the lanes with a radioactivity scanner and by counting radioactivity peaks. The sizes and mobilities of the size marker DNA fragments are indicated along the left margin.

It has previously been shown, using various parts of Tol2 as probes, that Tol2 copies are highly homogeneous in their restriction map structures in fish samples from a single local population (KOGA and HORI 1999 ). In the present study, the same method was applied to O. curvinotus and the 12 O. latipes fish. The results (Fig 5) were the same as those for a single natural population obtained in the previous study: single bands were observed for the three probes of Tol2 internal regions and these bands were identical in size to those expected for Tol2-tyr. Results of PCR using the Tol2 terminal regions (28 bp from the ends) as primers also indicated no shorter copies (Fig 6). Thus, with the same logic described in KOGA and HORI 1999 , most, or possibly all, Tol2 copies carried by these 13 samples have lengths and restriction map structures identical to one another and to Tol2-tyr.

View larger version (49K): In this window In a new window Download PPT slide   Figure 5. Southern blot hybridization for examination of restriction patterns of Tol2 copies. Restriction enzymes and probes are indicated under A, B, and C.

View larger version (66K): In this window In a new window Download PPT slide   Figure 6. PCR products from the Tol2-tyr clone and genomic DNAs. The template DNAs, indicated above the lanes, were 10 pg of the Tol2-tyr clone, 100 ng of fish genomic DNAs, and no template DNA. The PCR primers were 28 nucleotides from the ends of Tol2-tyr. The conditions were [2 min at 94°] and 25 cycles of [20 sec at 94°, 20 sec at 60°, 6 min at 72°]. Reaction mixtures were fractionated on a 1.0% agarose gel and stained with ethidium bromide.

Sequencing analysis of Tol2 copies:
We have previously sequenced five Tol2 copies randomly chosen from genomic libraries of O. latipes and found that they are identical over the entire 4.7-kb element (KOGA and HORI 1999 ). The fish used were samples from the Southern Japan population. In the present study, using the same protocols, we cloned and sequenced one Tol2 copy from a fish of the Northern Japan population and one from O. curvinotus. Differences were found only in the number of consecutive T residues in intron 1 (nucleotides 677 to 705 of the Tol2-tyr sequence; see Fig 2): 29 T's in five copies from the O. latipes Southern Japan population; 32 T's for the O. latipes Northern Japan population; and 33 T's for the O. curvinotus sample. Sequences outside the Tol2 elements, including their target site duplications, differed from one another.

Sequencing analysis of the tyrosinase gene region:
The tyrosinase gene is a single-copy gene located on one of the chromosomes (INAGAKI et al. 1994 ). Its nucleotide sequence is available in GenBank under accession no. AB010101. This sequence was obtained from a fish of the Southern Japan population. To provide a frame of reference for considering the history of Tol2, we sequenced part of the tyrosinase gene of a fish from each of the Northern Japan population, O. curvinotus, O. luzonensis, and O. melastigma, the last one being used as an outgroup. The phylogenetic trees of the fishes based on the sequences obtained (Fig 7) were consistent with the phylogenetic tree that had previously been obtained (Fig 1).

View larger version (16K): In this window In a new window Download PPT slide   Figure 7. Maximum parsimony trees of tyrosinase gene sequences. Nucleotide sequences were determined for the region indicated by the arrow in Fig 2. This region includes exon 1 (828 bp; nucleotides 3505 to 4332 of the tyrosinase gene sequence), intron 1 (439 bp; nucleotides 4333 to 4771), exon 2 (217 bp; nucleotides 4772 to 4988), and part of intron 2 (567 bp; nucleotides 4989 to 5555). The trees were constructed using PAUP (SWOFFORD 1991 ). O. melastigma was used as an outgroup. The species are abbreviated with their first three letters. N and S stand for the Northern Japan and the Southern Japan populations, respectively. The numbers of mutational steps between the nodes are indicated above the branches. The percentages below each branch represent bootstrap values of 100 times. The sequences have been deposited in GenBank with accession nos. AB032694 (latN), AB032695 (cur), AB032696 (luz), and AB032697 (mel). (A) Sequences for exons 1 and 2 were used. (B) The whole sequences determined were used. The mel sequence proved to contain a 0.9-kb insertion in intron 2. This portion was excluded from the calculation.

Distribution of Tol2 copies in the genome:
O. latipes, O. curvinotus, and O. luzonensis have the same chromosome number, 2n = 48, and their karyotypes are similar to one another (UWA 1991 ). Of the three species, O. curvinotus and O. luzonensis can be crossed with each other and their hybrid progeny are fertile (SAKAIZUMI et al. 1992 ). Tol2 is present in O. curvinotus but not in O. luzonensis. Using these two species, we made a linkage analysis of Tol2 copies to ascertain if the copies are dispersed in the genome of O. curvinotus. Fig 8 illustrates the mating scheme we employed for this purpose. The first generation of the cross (G₀ to G₁) was conducted to make the copies "heterozygous." The following generation (G₁ to G₂) was to reduce the copy number for better identification of bands in Southern blots because the copy number at generation 1 was not sufficiently low, probably by chance and possibly by transposition. A single male of generation 2 (G₂ ) was then crossed to a female of O. luzonensis (G₂ ), and 10 fish of their progeny (G₃-1 to -10), together with the parents, were analyzed in a Southern blot for transmission patterns of Tol2 copies between generations 2 and 3. Fig 9 shows the autodiagram obtained. The male parent (G₂ ) contains 15 copies, which were transmitted to the progeny without novel insertion events, because new bands were not present in the progeny. The copy numbers in the progeny (G₃) were between 5 and 10, with an average of 6.7. The 15 bands carried by the male parent were numbered from the top to the bottom as shown in the photograph. Fisher's exact probability tests were conducted for all possible combinations of 2 bands out of the 15, and a significant association (P < 0.01) was shown only for the combination of bands no. 1 and no. 9. Therefore, the copies represented by these 2 bands appeared to be closely linked to each other on the same chromosome, and the other copies appeared to be transmitted independently of one another. Thus, the Tol2 copies are dispersed in the genome.

View larger version (16K): In this window In a new window Download PPT slide   Figure 8. Mating scheme for linkage analysis of Tol2 copies. cur and luz in parentheses stand for O. curvinotus and O. luzonensis, respectively. Fishes of these two species were crossed as shown in the scheme.

View larger version (69K): In this window In a new window Download PPT slide   Figure 9. Southern blot for examination of transmission patterns of Tol2 copies. Genomic DNAs were prepared from the fishes of generations 2 and 3 (see Fig 8). The numbers shown in parentheses are the numbers of bands in each lane. The male parent (G2 ) exhibits 15 bands, numbered from the top to the bottom along the right margin.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Distribution of Tol2:
Tol2 was first found in the medaka fish O. latipes (KOGA et al. 1996 ). Its copies are highly homogeneous even at the nucleotide sequence level (KOGA and HORI 1999 ). One possible explanation of this observation is that Tol2 is a recently arrived element in the medaka fish genome. With this possibility in mind, we conducted the present Southern blot and PCR analyses of genomic DNAs to study the distribution and structure of Tol2 in the genus Oryzias. The results show that Tol2 is present in 2 (O. curvinotus and O. latipes) of the 10 species (Fig 3B) with identical restriction map structures in both of them (Fig 5). By contrast, restriction map length polymorphisms were observed in the tyrosinase gene region (Fig 3A). The discontinuous distribution of Tol2 within the branch of the fish phylogeny consisting of O. latipes, O. curvinotus, and O. luzonensis (see Fig 1) is also revealing. These results are in accord with the hypothesis that Tol2 has recently entered the medaka fish genome.

Weak hybridization signals with Tol2 were observed for zebrafish and chicken. We conducted further Southern blot analysis of zebrafish using probes L(0)Pv(0.5), Pv(0.5)Pv(3.1), and Ps(2.4)Ps(4.3) under the same conditions as for Fig 3B. Similar signals were observed for probe Pv(0.5)Pv(3.1) but not for the other two probes (data not shown). It has been reported (IZSVAK et al. 1999 ) that the internal inverted repeats of the Tol2-tyr sequence (see Fig 2) are copies of another, short transposable element called Angel and that the zebrafish genome contains 10³ to 10⁴ copies of this element. The weak hybridization signals might thus have been due to Angel copies and the chicken genome might contain similar repetitive sequences.

Hybridization signals were not observed for the swordtail fish and human samples. Tol2 thus seems to be absent in these species or else might be too divergent to be detected with Southern blotting. It should be noted that the detectability in the hybridization analysis decreases as the genome size of the target species increases. To overcome this problem, we used 20 µg of genomic DNA for species outside the genus Oryzias, five times as much as for those within the genus (4 µg). The sizes of haploid genomes are 0.68–0.85 x 10⁹ bp for the medaka fish (TANAKA 1995 ), 1.7 x 10⁹ bp for zebrafish (POSTLETHWAIT et al. 1994 ), and 3.0 x 10⁹ bp for human (GARDINER 1995 ). Thus, at least for zebrafish and humans, the genome size should not be a significant factor affecting our inference.

Phylogeny of fish species:
The relationship among O. latipes, O. curvinotus, and O. luzonensis has been studied from various aspects. SAKAIZUMI et al. 1992 examined electrophoretic patterns of 15 proteins and showed that the number of common alleles is larger between O. curvinotus and O. luzonensis than between O. curvinotus and O. latipes, and than between O. luzonensis and O. latipes. UWA 1991 constructed a dendrogram of these three species based on their karyotypes, which is consistent with the phylogenetic tree shown in Fig 1. UWA 1991 and SAKAIZUMI et al. 1992 examined fertility and meiotic segregation in interspecific hybrids and showed that reproductive isolation is complete between O. latipes and O. luzonensis and between O. latipes and O. curvinotus but not between O. curvinotus and O. luzonensis. All findings support the conclusion that O. luzonensis is more closely related to O. curvinotus than to O. latipes.

Molecular phylogenetic studies including the three species have also been completed. NARUSE 1996 constructed a phylogenetic tree based on the nucleotide sequence divergence of the mitochondrial 12S ribosomal RNA gene; O. curvinotus and O. luzonensis clustered with a high bootstrap value of 86%, supporting the clustering of the two species. Another tree using the mitochondrial cytochrome b gene gave the same clustering pattern and exhibited a much higher bootstrap value, 100%, for the clustering of the two species (J. S. ALBERT and K. NARUSE, personal communication). In the present study, we examined the phylogenetic relationship of the three species using a single-copy nuclear gene and obtained the phylogenetic tree shown in Fig 9. Clustering of O. curvinotus and O. luzonensis is again apparent.

Possibility of horizontal transfer:
No deletion copies of Tol2 were detected, while heterogeneity in size due to the presence of internally deleted, smaller copies is common in the hAT family members (FEDOROFF et al. 1983 ; STRECK et al. 1986 ; WARREN et al. 1994 ). High homogeneity was observed not only in a local population (KOGA and HORI 1999 ) but also across the entire species and even between different species (Fig 5 and Fig 6). Although >100 base substitutions have accumulated in the tyrosinase gene region (Fig 7), variation in the Tol2 samples was observed only in the number of consecutive T's in one of its introns. No other variation was found even in the third bases of codons and in introns. This is a strikingly high homogeneity in comparison not only with hAT elements but also with more diverse eukaryotic elements including mariner, which has been reported to be highly homogeneous (MARUYAMA and HARTL 1991B ; CAPY et al. 1992 ). The most likely explanation of virtually no sequence divergence between O. curvinotus and O. latipes is a recent Tol2 invasion of one or both of the two species. This explanation involves horizontal transfer of Tol2 from one species to the other or into the two species from a common source.

An alternative explanation of the sequence homogeneity of Tol2 is to postulate a strong selective constraint on the Tol2 sequence. However, this is difficult to reconcile with the fact that Tol2 contains introns with a total length of 1992 bp (KOGA et al. 1999 ) because strong selective pressure can be considered not to work on introns. Another possibility would be to invoke concerted evolution of Tol2 copies. However, although homogeneity within species may thereby be conceivable, that observed between species is difficult to explain.

A new transposition event was not observed in the last generation of crosses (Fig 7 and Fig 8). Similarly, Southern blot analysis of an inbred strain did not give evidence for transposition in the last few generations (KOGA and HORI 1999 ). Thus, Tol2 is not an element that moves as frequently as P and mariner of Drosophila, at least under usual laboratory conditions. However, in medaka fish natural populations, Tol2 appears to be sufficiently active to create a situation in which Tol2 copies are dispersed in the genome (Fig 9) and Tol2 is widespread in the whole distribution area of O. latipes (Fig 4). We have demonstrated the activity of Tol2 in zebrafish by introducing a clone into fertilized eggs and observing its excision by PCR (KAWAKAMI et al. 1998 ; KAWAKAMI and SHIMA 1999 ). The same result has also been obtained using O. luzonensis (A. KOGA and H. HORI, unpublished results). These results indicate the potential of Tol2 to persist in the genomes of these fishes once introduced into them.

Time and place of horizontal transfer:
One of the possibilities about the time of occurrence of the horizontal transfer is that it took place after the collection of the original samples because the materials used in the present study have been maintained as fish stocks for 3–10 years. In other words, it might be that there was transfer of Tol2 among the laboratory stocks. This possibility, however, is negligible because the two species have been maintained in different places: O. curvinotus has been in the Nagoya City Zoological Garden and the O. latipes stocks have been in our lab at the University of Tokyo. The two species have not undergone encounter with each other before DNA preparation. Thus, it is more likely that the horizontal transfer occurred in nature.

As to where in nature horizontal transfer occurred, overlapping distribution areas of the two species may be a likely supposition. O. latipes is distributed in a wide range in East Asia, from Japan to China. O. curvinotus inhabits Southern China and Vietnam. It was examined whether the habitats of these two species overlap, with samples collected at various locations in China, but none was observed (UWA and PARENTI 1988 ). However, there is no obvious interruptive natural factor, such as a desert, between the distribution areas of the two species and it is conceivable that there might be an overlap somewhere in Southern China.

Concerning mechanisms involved in the horizontal transfer and also the direction of the horizontal transfer, our results so far obtained provide no suggestions.

Generality of horizontal transfer:
Horizontal transfer may be considered general at least for mariner/Tc1 family elements because of (1) their presence in diverse organisms (GARCIA-FERNANDEZ et al. 1993 ), (2) independence of the transposition activity from species-specific host factors (GARZA et al. 1991 ; LAMPE et al. 1996 ; VOS et al. 1996 ; GUEIROS-FILHO and BEVERLY 1997 ; FADOOL et al. 1998 ), and (3) accumulation of convincing evidence (MARUYAMA and HARTL 1991A ; ROBERTSON and LAMPE 1995 ). The hAT family elements are distributed in a wide range of organisms, comparable to that of mariner/Tc1 family elements: they are found in plants (HARTINGS et al. 1991 ; TSAY et al. 1993 ), fungi (KEMPKEN and KUCK 1996 ), insects (WARREN et al. 1994 ; COATES et al. 1996 ), and other invertebrate and vertebrate animals (DEVAULT and NARANG 1994 ; BIGOT et al. 1996 ; KOGA et al. 1996 ). Independence of species-specific host factors, at least to a certain extent, is demonstrated by activity in genomes that are free of the elements, as reported for hobo (HANDLER and GOMEZ 1995 ; LOZOVSKAYA et al. 1996 ), Ac (BAKER et al. 1986 ), and Tol2 (KAWAKAMI et al. 1998 ). Evidence for horizontal transfer from sequence divergence data has been shown for hobo (SIMMONS 1992 ). The results in this article also suggest horizontal transfer of Tol2. Horizontal transfer of hAT elements might be frequent, as it appears to be for mariner/Tc1 elements.

	ACKNOWLEDGMENTS

We are grateful to H. Hashikawa, M. Sato, and S. Susaki for providing the fish samples. We also thank K. Naruse, T. Iwamatsu, and D. L. Hartl for helpful discussions. This work was partly supported by grant nos. 09680672 and 10216025 to A.K. and nos. 08640786 and 09554053 to H.H. from the Ministry of Education, Science, Sports and Culture of Japan and also by the Takeda Science Foundation to A.K.

Manuscript received October 4, 1999; Accepted for publication January 25, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ATKINSON, P. W., W. D. WARREN, and D. A. O'BROCHTA, 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[Abstract/Free Full Text].

BAKER, B., J. SCHELL, H. LÖRZ, and N. FEDOROFF, 1986  Transposition of the maize controlling element "Activator" in tobacco. Proc. Natl. Acad. Sci. USA 83:4844-4848[Abstract/Free Full Text].

BIGOT, Y., C. AUGÉ-GOUILLOU, and G. PERIQUET, 1996  Computer analyses reveal a hobo-like element in the nematode Caenorhabditis elegans, which presents a conserved transposase domain common with the Tc1-Mariner transposon family. Gene 174:265-271[Medline].

CALVI, B. R., T. J. HONG, S. D. FINDLEY, and W. M. GELBERT, 1991  Evidence for a common evolutionary origin of inverted repeat transposons in Drosophila and plants: hobo, Activator, and Tam3.. Cell 66:465-471[Medline].

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. ANXOLABÉHÈRE, and T. LANGIN, 1994  The strange phylogenies of transposable elements: Are horizontal transfers the only explantation? Trends Genet. 10:7-12[Medline].

COATES, C. J., K. N. JOHNSON, H. D. PERKINS, A. J. HOWELLS, and 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[Medline].

DANIELS, S. B., K. R. PETERSON, L. D. STRAUSBAUGH, M. G. KIDWELL, and A. CHOVNICK, 1990  Evidence for horizontal transmission of the P transposable element between Drosophila species. Genetics 124:339-355[Abstract].

DEVAULT, 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[Medline].

FADOOL, J. M., D. L. HARTL, and J. E. DOWLING, 1998  Transposition of the mariner element from Drosophila mauritiana in zebrafish. Proc. Natl. Acad. Sci. USA 95:5182-5186[Abstract/Free Full Text].

FEDOROFF, N., S. WESSLER, and M. SHURE, 1983  Isolation of the transposable maize controlling elements Ac and Ds.. Cell 35:235-242[Medline].

GARCIA-FERNÀNDEZ, J., G. MARFANY, J. BAGUÑÀ, and E. SALÓ, 1993  Infiltration of mariner elements. Nature 364:109-110[Medline].

GARDINER, K., 1995  Human genome organization. Curr. Opin. Genet. Dev. 5:315-322[Medline].

GARZA, D., M. MEDHORA, A. KOGA, and D. L. HARTL, 1991  Introduction of the transposable element mariner into the germline of Drosophila melanogaster.. Genetics 128:303-310[Abstract].

GUEIROS-FILHO, F. and S. M. BEVERLY, 1997  Trans-kingdom transposition of the Drosophila element mariner within the protozoan Leishmania.. Science 276:1716-1719[Abstract/Free Full Text].

HANDLER, A. M. and S. P. GOMEZ, 1995  The hobo transposable element has transposase-dependent and -independent excision activity in drosophilid species. Mol. Gen. Genet. 247:399-408[Medline].

HARTINGS, H., C. SPILMONT, N. LAZZARONI, V. ROSSI, and F. SALAMINI et al., 1991  Molecular analysis of the Bg-rbg transposable element system of Zea mays L. Mol. Gen. Genet. 227:91-96[Medline].

INAGAKI, H., Y. BESSHO, A. KOGA, and H. HORI, 1994  Expression of the tyrosinase-encoding gene in a colorless melanophore mutant of the medaka fish, Oryzias latipes.. Gene 150:319-324[Medline].

IZSVÁK, Z., Z. IVICS, N. SHIMODA, D. MOHN, and H. OKAMOTO et al., 1999  Short inverted-repeat transposable elements in teleost fish and implications for a mechanism of their amplification. J. Mol. Evol. 48:13-21[Medline].

KAWAKAMI, K. and A. SHIMA, 1999  Identification of the Tol2 transposase of the medaka fish Oryzias latipes that catalyzes excision of a nonautonomous Tol2 element in zebrafish Danio rerio.. Gene 240:239-244[Medline].

KAWAKAMI, K., A. KOGA, H. HORI, and A. SHIMA, 1998  Excision of the Tol2 transposable element of the medaka fish, Oryzias latipes, in zebrafish, Danio rerio.. Gene 225:17-22[Medline].

KEMPKEN, F. and U. KÜCK, 1996  restless, an active Ac-like transposon from the fungus Tolypocladium inflatum: structure, expression, and alternative RNA splicing. Mol. Cell. Biol. 16:6563-6572[Abstract].

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

KOGA, A. and H. HORI, 1999  Homogeneity in the structure of the medaka fish transposable element Tol2.. Genet. Res. Camb. 73:7-14[Medline].

KOGA, A., H. INAGAKI, Y. BESSHO, and H. HORI, 1995  Insertion of a novel transposable element in the tyrosinase gene is responsible for an albino mutation in the medaka fish, Oryzias latipes.. Mol. Gen. Genet. 249:400-405[Medline].

KOGA, A., M. SUZUKI, H. INAGAKI, Y. BESSHO, and H. HORI, 1996  Transposable element in fish. Nature 383:30[Medline].

KOGA, A., M. SUZUKI, Y. MARUYAMA, M. TSUTSUMI, and H. HORI, 1999  Amino acid sequence of a putative transposase protein of the medaka fish transposable element Tol2 deduced from mRNA nucleotide sequences. FEBS Lett. 461:295-298[Medline].

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].

LAWRENCE, J. G., H. OCHMAN, and D. L. HARTL, 1992  The evolution of insertion sequences within enteric bacteria. Genetics 131:9-20[Abstract].

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

LOZOVSKAYA, E. R., D. I. NURMINSKY, D. L. HARTL, and D. T. SULLIVAN, 1996  Germline transformation of Drosophila virlis mediated by the transposable element hobo.. Genetics 142:173-177[Abstract].

MARUYAMA, K. and D. L. HARTL, 1991a  Evidence for interspecific transfer of the transposable element mariner between Drosophila and Zaprionus.. J. Mol. Evol. 33:514-524[Medline].

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

MCCLINTOCK, B., 1948  Mutable loci in maize. Carnegie Inst. Wash. Year Book 47:155-169.

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[Medline].

NARUSE, K., 1996  Classification and phylogeny of fishes of the genus Oryzias.. Fish Biol. J. Medaka 8:1-10.

NARUSE, K., A. SHIMA, M. MATSUDA, M. SAKAIZUMI, and T. IWAMATSU et al., 1993  Distribution and phylogeny of rice fish and their relatives belonging to the suborder Arianichthyoidei in Sulawesi, Indonesia. Fish Biol. J. Medaka 5:11-15.

POSTLETHWAIT, J. H., S. L. JOHNSON, C. N. MIDSON, W. S. TALBOT, and M. GATES et al., 1994  A genetic linkage map of the zebrafish. Science 264:699-703[Abstract/Free Full Text].

ROBERTSON, H. M. and D. J. LAMPE, 1995  Recent horizontal transfer of a mariner transposable element among and between Diptera and Neuroptera. Mol. Biol. Evol. 12:850-862[Abstract].

SAKAIZUMI, M., 1985  Species-specific expression of parvalbumins in the genus Oryzias and its related species. Comp. Biochem. Physiol. 80B:499-505.

SAKAIZUMI, M., 1986  Genetic divergence in wild populations of Medaka, Oryzias latipes (Pisces: Oryziatidae) from Japan and China. Genetica 69:119-125.

SAKAIZUMI, M., H. UWA, and S.-R. JEON, 1987  Genetic diversity of the East Asian populations of the freshwater fish, Oryzias.. Zool. Sci. 4:1003.

SAKAIZUMI, M., Y. SHIMIZU, and S. HAMAGUCHI, 1992  Electrophoretic studies of meiotic segregation in inter- and intraspecific hybrids among East Asian species of the genus Oryzias (Pisces: Oryziatidae). J. Exp. Zool. 264:85-92.

SIMMONS, G. M., 1992  Horizontal transfer of hobo transposable elements within the Drosophila melanogaster species complex: evidence from DNA sequencing. Mol. Biol. Evol. 9:1050-1060[Abstract].

SOMMER, H., R. CARPENTER, R. J. HARRISON, and H. SAEDLER, 1985  The transposable element Tam3 of Antirrhinum majus generates a novel type of sequence alteration upon excision. Mol. Gen. Genet. 199:225-231.

STRECK, R. D., J. E. MACGAFFEY, and S. K. BECKENDORF, 1986  The structure of hobo transposable elements and their insertion sites. EMBO J. 5:3615-3623[Medline].

SWOFFORD, D. L., 1991 PAUP: phylogenetic analysis using parsimony, version 3.0. Illinois Natural History Survey, Champaign, IL.

TANAKA, M., 1995  Characteristics of medaka genes and their promoter regions. Fish Biol. J. Medaka 7:11-14.

TSAY, Y.-F., M. J. FRANK, T. PAGE, C. DEAN, and N. M. CRAWFORD, 1993  Identification of a mobile endogenous transposon in Arabidopsis thaliana.. Science 260:342-344[Abstract/Free Full Text].

UWA, H., 1991  Cytosystematic study of the Hainan medaka, Oryzias curvinotus, from Hong Kong (Teleostei: Oryziidae). Ichthyol. Explo. Freshwaters 1:361-367.

UWA, H. and L. R. PARENTI, 1988  Morphometric and meristic variation in ricefishes, genus Oryzias: a comparison with cytogenetic data. Jpn. J. Ichthyol. 35:159-166.

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].

WARREN, W. D., P. W. ATKINSON, and D. A. O'BROCHTA, 1994  The Hermes transposable element from the house fly, Musca domestica, is a short inverted repeat-type element of the hobo, Ac, and Tam3 (hAT) element family. Genet. Res. Camb. 64:87-97[Medline].

This article has been cited by other articles:

				A. Hikosaka, T. Kobayashi, Y. Saito, and A. Kawahara Evolution of the Xenopus piggyBac Transposon Family TxpB: Domesticated and Untamed Strategies of Transposon Subfamilies Mol. Biol. Evol., December 1, 2007; 24(12): 2648 - 2656. [Abstract] [Full Text] [PDF]

				A. Koga, A. Iida, H. Hori, A. Shimada, and A. Shima Vertebrate DNA Transposon as a Natural Mutator: The Medaka Fish Tol2 Element Contributes to Genetic Variation without Recognizable Traces Mol. Biol. Evol., July 1, 2006; 23(7): 1414 - 1419. [Abstract] [Full Text] [PDF]

				B. Wickstead, K. Ersfeld, and K. Gull Repetitive Elements in Genomes of Parasitic Protozoa Microbiol. Mol. Biol. Rev., September 1, 2003; 67(3): 360 - 375. [Abstract] [Full Text] [PDF]

				E. Rubin, G. Lithwick, and A. A. Levy Structure and Evolution of the hAT Transposon Superfamily Genetics, July 1, 2001; 158(3): 949 - 957. [Abstract] [Full Text] [PDF]

				A. Koga and H. Hori Detection of de Novo Insertion of the Medaka Fish Transposable Element Tol2 Genetics, November 1, 2000; 156(3): 1243 - 1247. [Abstract] [Full Text]