Wake Up of Transposable Elements Following Drosophila simulans Worldwide Colonization Cristina Vieira, David Lepetit, Sandy Dumont, and Christian Bie´mont Laboratoire de Biome´trie, Ge´ne´tique, Biologie des populations, UMR Centre National de la Recherche Scientifique, Universite´ Lyon 1, Villeurbanne, France Transposable elements (TEs) make up around 10%–15% of the Drosophila melanogaster genome, but its sibling species Drosophila simulans carries only one third as many such repeat sequences. We do not, however, have an overall view of copy numbers of the various classes of TEs (long terminal repeat [LTR] retrotransposons, non-LTR retrotransposons, and transposons) in genomes of natural populations of both species. We analyzed 34 elements in individuals from various natural populations of these species. We show that D. melanogaster has higher average chromosomal insertion site numbers per genome than D. simulans for all TEs except five. The LTR retrotransposons gypsy, ZAM, and 1731 and the transposon bari-1 present similar low copy numbers in both species. The transposon hobo has a large number of insertion sites, with significantly more sites in D. simulans. High variation between populations in number of insertion sites of some elements of D. simulans suggests that these elements can invade the genome of the entire species starting from a local population. We propose that TEs in the D. simulans genome are being awakened and amplified as they had been a long time ago in D. melanogaster.
Introduction Because of their capacity to promote mutations, transposable elements (TEs) are a serious threat to the genome. The way their copy numbers are regulated and contained is thus of major importance for genome integrity. One way of approaching this problem is a comparison between species differing in their overall amounts of repetitive sequences, as do the two sister species Drosophila melanogaster and Drosophila simulans. Indeed, D. melanogaster carries approximately three times as much middle repetitive DNA as D. simulans (Dowsett and Young 1982; Vieira and Bie´mont 1996a). It is thus suggested that TEs in D. simulans have a lower transposition rate, that the genome of D. simulans is less permissive than the genome of D. melanogaster to TE amplification, that a stronger selection acts against TE multiplication, and that D. simulans has a higher effective species population size than D. melanogaster (Aquadro, Lado, and Noon 1988; Aquadro 1992). The absence of evidence for a low rate of transposition in D. simulans (Eeken, De Jong, and Green 1987; Inoue and Yamamoto 1987; Vieira and Bie´mont 1997) favors the idea of particular cellular or physiological capacities of D. simulans in restraining copy number. If effective population size is involved in the smaller amount of TEs in D. simulans, then we expect some TEs to be affected in a similar way, at least those which are old components of the Drosophila genome and which present significant insertion site numbers on chromosomal arms. We do not, however, have a global view of copy numbers of many TEs in natural populations of both D. melanogaster and D. simulans, espeAbbreviation: TE, transposable element. Key words: transposable elements, copy number, Drosophila melanogaster, Drosophila simulans. Address for correspondence and reprints: Christian Bie´mont, Laboratoire de Biome´trie, Ge´ne´tique, Biologie des Populations, UMR Centre National de la Recherche Scientifique 5558, Universite´ Lyon 1, 69622 Villeurbanne cedex, France. E-mail:
[email protected]. Mol. Biol. Evol. 16(9):1251–1255. 1999 q 1999 by the Society for Molecular Biology and Evolution. ISSN: 0737-4038
cially when the various classes of TEs (long terminal repeat [LTR] retrotransposons, non-LTR retrotransposons, and transposons) are compared. We report here the insertion site numbers of 34 transposable elements (24 LTR retrotransposons, 6 non-LTR retrotransposons or retroposons, 4 transposons). We show that D. melanogaster has higher insertion site numbers for all elements except bari-1 (a transposon), gypsy and ZAM (two retroviruses), and 1731 (an LTR-retrotransposon), which are equally abundant in both species, and hobo, which shows more insertion sites in D. simulans. A high variation in TE insertion site numbers between populations of D. simulans, with some values similar to those for D. melanogaster, suggests that TEs are accumulating in the D. simulans species. Materials and Methods Natural Populations We worked on fly samples collected from several geographically distinct, natural populations. The D. melanogaster populations considered were from Arabia, Argentina (Virasoro), Bolivia, China (Canton), Congo (Brazzaville), France (St. Cyprien), Portugal (Chicharo), Re´ union Island, Senegal, and the U.S.A. (Seattle, Wash.), and the populations of D. simulans were from Australia (Canberra, Cann River, Eden), France (Valence), Kenya (Makindu), Madagascar, New Caledonia (Amieu), Polynesia (Noumea, Papeete), Portugal (Madeira), Re´union Island, Russia (Moscow), and Zimbabwe. These populations were maintained in the laboratory as isofemale lines or small mass cultures with around 50 pairs every generation. Transposable Elements We used probes from the following elements: LTR retrotransposons: 1731, 17.6, 297, 412, BEL, blood, burdock, copia, coral, flea, gypsy, HMS beagle, idefix, mdg1, mdg3, opus, osvaldo, prygun, roo/B104, springer, stalker, telemac, tirant, and ZAM; non-LTR retrotransposons: bilbo, doc, F, helena, I, and jockey; transposons: bari-1, gandalf, hobo, and pogo. In table 1, we 1251
1252
Vieira et al.
Table 1 Average Euchromatic Insertion Site Number per Genome for 36 Transposable Elements in Populations of Drosophila melanogaster and Drosophila simulans SPECIES
ELEMENTS
D. melanogaster
D. simulans
Average Site Standard Number Error
Average Site Standard Number Error
LTR retrotransposons 1731 . . . . . . . . . 1.55 17.6 . . . . . . . . . 2.50 297 . . . . . . . . . . 23.40 412 . . . . . . . . . . 28.45 BEL . . . . . . . . . 5.25 blood . . . . . . . . 17.45 burdock . . . . . . 10.35 copia . . . . . . . . 24.05 coral. . . . . . . . . 15.85 flea . . . . . . . . . . 16.60 gypsy . . . . . . . . 1.70 HMS beagle . . 9.50 ide´fix . . . . . . . . 5.70 mdg1 . . . . . . . . 20.75 mdg3 . . . . . . . . 14.10 opus . . . . . . . . . 20.90 osvaldo . . . . . . 0.00 prygun . . . . . . . 11.35 roo/B104 . . . . . 67.60 springer . . . . . . 2.35 stalker . . . . . . . 6.50 telemac . . . . . . 0.00 tirant . . . . . . . . 11.45 ZAM . . . . . . . . . 0.35 Non-LTR retrotransposons bilbo. . . . . . . . . 0.00 dor . . . . . . . . . . 26.20 F. . . . . . . . . . . . 31.40 helena . . . . . . . 0.25 I . . . . . . . . . . . . 25.15 jockey. . . . . . . . 31.60 Transposons bari-1 . . . . . . . . 4.37 gandalf. . . . . . . 0.00 hobo . . . . . . . . . 49.90 pogo . . . . . . . . . 13.25 Pa . . . . . . . . . . . 14–33 marinera . . . . . . 0.00
1.10 2.70 6.16 4.76 2.97 3.41 2.39 3.85 4.18 4.33 1.34 3.15 2.52 4.09 4.69 4.73
3.14 0.49
1.00 0.00 1.00 13.88 0.58 2.50 5.27 3.88 1.88 3.42 1.54 2.77 1.00 0.19 3.35 4.81 0.00 0.81 38.46 0.00 0.38 0.00 1.62 0.23
4.74 10.73 0.55 5.47 7.51
0.00 13.81 1.77 10.23 12.58 3.27
3.99 13.87 1.53 4.08
2.50 11.38 4.71
4.88 0.00 66.23 0.00 0.00 0–10
0.96 0.84 15.00 0.71 2.72 3.29 1.12 1.70 3.53 1.29 3.07 0.00 0.41 1.80 2.43 1.84 7.05
NS *** *** *** *** *** *** *** *** *** NS *** *** *** *** ***
0.65
*** *** *** ***
1.15 0.58
*** NS
5.31 3.85 2.17 3.71 1.75
*** *** *** *** ***
2.73
NS
14.01
*** ***
NOTE.—The insertion site numbers for the two species were compared by a Mann-Whitney U-test done on distributions represented in figure 1. ***, P , 0.001; NS, P . 0.05. a Range value data from the literature.
added the data for P and mariner from the literature. Most elements are from D. melanogaster, except gandalf (from Drosophila koepferae), osvaldo (from Drosophila buzzatii), bilbo (from Drosophila subobscura), and helena and telemac (from D. virilis). Information concerning all of these elements can be obtained through Flybase. In Situ Hybridization Polytene chromosomes from salivary glands of third-instar female larvae were prepared and treated with nick-translated, biotinylated DNA probes as previously described (Bie´mont 1994). Insertion sites were visual-
ized as brown bands resulting from a dye-coupled reaction with peroxidase substrate and diaminobenzidine. The insertion site numbers of the TEs were determined on all long chromosome arms (X, 2L, 2R, 3L, 3R) and were summed to give the total number of labeled sites per diploid genome. We did not take into account the insertions localized in pericentromeric regions 20, 40, 41, 80, and 81, because TE site number estimations in these regions are difficult and not reliable for all chromosomes or all squashes. For each element, the slides from the two species D. melanogaster and D. simulans were hybridized simultaneously so as to eliminate any problem associated with labeling intensity. To take into account TE insertion polymorphism, one female larva from each of two isofemale lines per population or two female larvae from each mass-mated stock were analyzed for TE copy number. Since the in situ technique does not allow us to distinguish homozygous from heterozygous sites, any increase in homozygosity level in the populations sampled due to inbreeding during laboratory culture may lead to an underestimation of total insertion sites in the diploid genomes. It has previously been shown, however, that such a protocol gives reliable data (Vieira and Bie´mont 1996a, 1996b). The bias in TE insertion site number estimates should not affect the comparison between D. melanogaster and D. simulans, because the populations of these two species were maintained in the laboratory under the same conditions. Results Higher TE Insertion Site Number in D. melanogaster than in D. simulans All TEs tested existed in both D. melanogaster and D. simulans, suggesting that they are old components of these species and were probably present before radiation. Table 1 reports average insertion site numbers on polytene chromosomes of the 34 TEs in 10 natural populations of D. melanogaster and 13 populations of D. simulans. Drosophila melanogaster appears to have higher insertion site numbers (mean insertion site number for TEs present in at least one species: 16.7) than D. simulans (mean insertion site number: 6.7) for all elements, excepting the two retroviruses (gypsy and ZAM), the LTR retrotransposon 1731, and the tranposons hobo and bari-1. Gypsy, ZAM, 1731, and bari-1 are equally represented in both species, while hobo has more insertion sites in D. simulans than in D. melanogaster. Among the LTR and non-LTR retrotransposons, the retrotransposon roo/B104 has the highest average site number in both species, but the transposon hobo has the highest site number of all TEs in D. simulans. Elements such as springer, P, and pogo are not represented on the chromosome arms in the D. simulans genome, and mariner is absent from D. melanogaster. The elements copia and idefix in D. simulans present three and one fixed sites, respectively; 17.6 is present at one pericentromeric insertion site at the base of the 2L chromosome arm; springer is absent from the chromosome arms but is detected in the chromocenter. Helena, an element from D. virilis, is peculiar because it has been
TE Copy Number in D. melanogaster and D. simulans
detected in the chromosome arms in only 2 populations out of 10 of D. melanogaster but is present at 10.32 sites on average in D. simulans. There is a discrepancy between our data for the 17.6 element and data from the literature, which show higher copy numbers (Dowsett and Young 1982; Kugimiya, Ikenaga, and Saigo 1983; Charlesworth, Jarne, and Assimacopoulos 1994); it may be that some 17.6 probes revealed 297 insertions, since these two elements are closely related (Kugimiya, Ikenaga, and Saigo 1983). The overall ratio of numbers for the melanogaster/simulans comparison is 499.82/201.34 5 2.48, a value not far from the crude estimate that D. melanogaster has about three times as many repetitive sequences as D. simulans (Dowsett and Young 1982). No insertion sites were detected with the elements osvaldo, gandalf, telemac, and bilbo. The Distribution of TE Insertion Site Numbers in Natural Populations As seen in figure 1, the TE insertion site numbers in D. melanogaster are globally normally distributed for many TEs, while they present more heterogeneous distributions in D. simulans because some populations have a large number of sites. For example, the elements HMS beagle, blood, flea, tirant, coral, prygun, jockey, and F have low site numbers in most D. simulans populations, excepting one or two populations in which they reach significantly larger insertion site numbers. We have tested whether some specific populations harbor more transposable elements than others by the nonparametric Friedman test. The statistic is nonsignificant for D. simulans but highly significant (P , 0.0001) for D. melanogaster populations. Populations with the lowest total copy numbers are from the African continent (Congo: 788, Senegal: 942, Re´union Island: 944), and total copy numbers increase in the other populations (Arabia: 966, Argentina: 1,009, Portugal: 1,028, U.S.A.: 1,067, China: 1,078, Bolivia: 1,081, France: 1,092). Discussion A D. melanogaster Bias or Specific Characteristics of the D. simulans Genome The higher TE copy numbers observed in D. melanogaster could result from a ‘‘melanogaster bias’’ resulting from the fact that high copy number and activity were the factors that permitted the detection of many TEs in D. melanogaster. There is indeed no a priori reason for these TEs to show the same high values in D. simulans. However, the observations that (1) even TEs with low copy numbers in D. melanogaster have still lower copy numbers in D. simulans; (2) gypsy, ZAM, and bari-1 have similar insertion site numbers in both species; and (3) hobo has very high values, with a higher value in D. simulans than in D. melanogaster, makes it unlikely that the bias hypothesis is the only explanation for the difference between the two species. The overall low TE site numbers in D. simulans suggest some specific evolutionary features of the D. simulans genome that have the effect of restraining the number of LTR and non-LTR retrotransposons (Aquadro, Lado,
1253
and Noon 1988; Aquadro 1992; Kimura and Kidwell 1994; Vieira and Bie´mont 1996a). However, this does not explain the systematic tendency to even lower site numbers in D. simulans for the TEs with low insertion site numbers in D. melanogaster. The observation that the 412 and roo/B104 elements have the highest average insertion site numbers in both species (28.45 and 13.88 for 412 and 67.60 and 38.46 for roo/B104 in D. melanogaster and D. simulans, respectively) suggests that specific characteristics of the TEs themselves must be involved, interacting with genomic properties that regulate TE copy number and that may differ between the two species. We cannot completely eliminate the hypothesis that some TEs in D. simulans are in the process of being lost. We would expect, under this hypothesis, to find many fixed sites for those TEs which present polymorphic insertion sites, which are not observed. Geographical Heterogeneity and TE Invasion Hypothesis The fact that in D. melanogaster the populations from Africa (the cradle of this species) harbor an overall smaller amount of TEs than do the populations from the other continents suggests that TE acquisition and accumulation are following the species colonization processes. There is no such significant association in D. simulans, but the existence of populations with very large insertion site numbers for a given TE in comparison with the mean species value, as reported here (see table 1 and fig. 1), suggests that some populations may overcome the regulatory constraints in this species. Moreover, in these D. simulans populations, TE copy numbers are similar to those observed in D. melanogaster. The geographical heterogeneity thus observed for these TEs in D. simulans suggests that they were mobilized in these specific populations and that the conditions constraining them were no longer in effect. This observation agrees with recent work on minisatellites that shows that D. simulans populations are perhaps much more variable than we thought (Irvin et al. 1998). We propose that the worldwide colonization process of D. simulans (Hyytia et al. 1985; Singh, Choudhary, and David 1987; Singh 1989; Capy, Pla, and David 1993) allows this species to encounter new environmental conditions capable of awakening some TEs. This fascinating possibility is illustrated by the correlation reported between the 412 copy number and minimal temperature in natural populations of D. simulans (Vieira et al. 1998). This suggests that TE insertion site number is a characteristic of a population more than of the species as a whole. Invasion of the entire species by TEs from these populations by step-by-step migration is thus a real possibility, as already inferred for the I, P, and hobo elements in D. melanogaster (Bucheton et al. 1984; Anxolabe´he`re, Kidwell, and Periquet 1988; Pe´riquet et al. 1994). If this is the case, we expect D. simulans populations to increase their insertion site numbers for many TEs, leading to distributions similar to those observed for D. melanogaster. The high TE insertion site numbers observed in D. melanogaster may thus only reflect the ancient mobilization of TEs when this species invaded
1254
Vieira et al.
FIG. 1.—Insertion site number distributions of 30 TEs in populations of D. melanogaster (gray histograms) and D. simulans (black histograms). The elements bilbo, gandalf, osvaldo, and telemac, which have no chromosomal insertion sites on both species, are not represented.
the world, long before D. simulans (Hyytia et al. 1985; Singh, Choudhary and David 1987; Singh 1989; Capy, Pla, and David 1993). Such invasion is not likely to happen for retrovirus elements such as gypsy and ZAM, which seem to be strongly constrained in the genome. However, D. melanogaster itself could be under a dy-
namic process of invasion by some TEs, as observed for the I, P, and hobo elements, which have been mobilized during the last 70 years or so (Bucheton et al. 1984; Anxolabe´he`re, Kidwell, and Periquet 1988; Periquet et al. 1994). Our results widen the debate to TEs not involved in dysgenesis and to D. simulans.
TE Copy Number in D. melanogaster and D. simulans
Time of TE mobilization, transposition and excision rates, and selection coefficients against TE insertions are specific characteristics that can vary between TEs and can interfere with the host genome, accounting for the differing insertion site numbers now observed in D. simulans populations. Effective population size may be involved in such a process, as it may determine the impact of transposition and selection on the equilibrium frequency spectrum of TE sites (Aquadro, Lado, and Noon 1988; Charlesworth and Langley 1991; Aquadro 1992; Brookfield and Badge 1997). It is not clear, however, why effective size does not affect all TEs simultaneously, since we observe that TEs do not present high insertion site numbers in the same populations (data not shown). Although more knowledge on the ecology of Drosophila is necessary, the possibility of a genomic invasion by TEs during the worldwide spread of a species is a challenging hypothesis that warrants research. Acknowledgments We thank D. Blesa, I. Boussy, B. Charlesworth, J. Costas, M. A. Csink, A. Dominguez, A. Fontdevila, R. de Frutos, N. Junakovic, V. Ladeve`ze, H. D. Lipshitz, E. G. Pasyukova, D. Petrov, N. Tchurikov, C. Vaury, J. Vieira, and J. Wahlstrom for their gifts of various transposable element DNAs. We are grateful to N. Gautier and R. Grantham for comments. This work was supported by the Junta Nacional de Investigac¸ao Cientifica (Portugal), the Centre National de la Recherche Scientifique, the Fondation pour la Recherche Me´dicale, and the Bureau des Ressources Ge´ne´tiques. LITERATURE CITED
ANXOLABE´HE`RE, D., M. G. KIDWELL, and G. PERIQUET. 1988. Molecular characteristics of diverse populations are consistent with the hypothesis of a recent invasion of Drosophila melanogaster by mobile P elements. Mol. Biol. Evol. 5: 252–269. AQUADRO, C. F. 1992. Why is the genome variable? Insights from Drosophila. Trends Genet. 8:355–362. AQUADRO, C. F., K. M. LADO, and W. A. NOON. 1988. The rosy region of Drosophila melanogaster and Drosophila simulans. I. Contrasting levels of naturally occurring DNA restriction map variation and divergence. Genetics 119:875– 888. BIE´MONT, C. 1994. Dynamic equilibrium between insertion and excision of P elements in highly inbred lines from an M9 strain of Drosophila melanogaster. J. Mol. Evol. 39:466– 472. BROOKFIELD, J. F. Y., and R. M. BADGE. 1997. Population genetics models of transposable elements. Genetica 100:281– 294. BUCHETON, A., R. PARO, H. M. SANG, A. PE´LISSON, and D. J. FINNEGAN. 1984. The molecular basis of IR hybrid dysgenesis in D. melanogaster: identification, cloning and properties of the I factor. Cell 38:153–163. CAPY, P., E. PLA, and J. R. DAVID. 1993. Phenotypic and genetic variability of morphometrical traits in natural populations of Drosophila melanogaster and D. simulans. I. Geographical variations. Genet. Sel. Evol. 25:517–536.
1255
CHARLESWORTH, B., P. JARNE, and S. ASSIMACOPOULOS. 1994. The distribution of transposable elements within and between chromosomes in a population of Drosophila melanogaster. III. Element abundances in heterochromatin. Genet. Res. 64:183–197. CHARLESWORTH, B., and C. H. LANGLEY. 1991. Population genetics of transposable elements in Drosophila. Pp. 150–176 in R. K. SELANDER, A. G. CLARK, and T. S. WHITTAM, eds. Evolution at the molecular level. Sinauer, Sunderland, Mass. DOWSETT, A. P., and M. W. YOUNG. 1982. Differing levels of dispersed repetitive DNA among closely related species of Drosophila. Proc. Natl. Acad. Sci. USA 79:4570–4574. EEKEN, J. C. J., A. W. M. DE JONG, and M. M. GREEN. 1987. The spontaneous mutation rate in Drosophila simulans. Mutat. Res. 192:259–262. HYYTIA, P., P. CAPY, J. R. DAVID, and R. S. SINGH. 1985. Enzymatic and quantitative variation in European and African populations of Drosophila simulans. Heredity 54: 209–217. INOUE, Y. H., and M. T. YAMAMOTO. 1987. Insertional DNA and spontaneous mutation at the white locus in Drosophila simulans. Mol. Gen. Genet. 209:94–100. IRVIN, S. D., K. A. WETTERSTRAND, C. M. HUTTER, and C. F. AQUADRO. 1998. Genetic variation and differentiation at microsatellite loci in Drosophila simulans: evidence for founder effects in new world populations. Genetics 150: 777–790. KIMURA, K., and M. G. KIDWELL. 1994. Differences in P element population dynamics between the sibling species Drosophila melanogaster and D. simulans. Genet. Res. 63: 27–38. KUGIMIYA, W., H. IKENAGA, and K. SAIGO. 1983. Close relationship between the long terminal repeats of avian leukosis sarcoma virus and copia-like movable genetic elements of Drosphila. Proc. Natl. Acad. Sci. USA 80:3193–3197. PERIQUET, G., F. LEMEUNIER, Y. BIGOT, M. H. HAMELIN, C. BAZIN, V. LADEVE`ZE, J. EEKEN, M. I. GALINDO, L. PASCUAL, and I. BOUSSY. 1994. The evolutionary genetics of the hobo transposable element in the Drosophila melanogaster complex. Genetica 93:79–90. SINGH, R. S. 1989. Population genetics and evolution of species related to Drosophila melanogaster. Annu. Rev. Genet. 23:425–453. SINGH, R. S., M. CHOUDHARY, and J. R. DAVID. 1987. Contrasting patterns of geographic variation in the cosmopolitan sibling species Drosophila melanogaster and Drosophila simulans. Biochem. Genet. 25:27–40. VIEIRA, C., P. AUBRY, D. LEPETIT, and C. BIE´MONT. 1998. A temperature cline in copy number for 412 but not roo/B104 retrotransposons in populations of Drosophila simulans. Proc. R. Soc. Lond. B Biol. Sci. 265:1–5. VIEIRA, C., and C. BIE´MONT. 1996a. Selection against transposable elements in D. simulans and D. melanogaster. Genet. Res. 68:9–15. . 1996b. Geographical variation in insertion site number of retrotransposon 412 in Drosophila simulans. J. Mol. Evol. 42:443–451. . 1997. Transposition rate of the 412 retrotransposable element is independent of copy number in natural populations of Drosophila simulans. Mol. Biol. Evol. 14:185–188.
PIERRE CAPY, reviewing editor Accepted May 25, 1999
