Genome Evolution: Sex and the Transposable Element Stephen Wright* and David Finnegan†

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

R296 Dispatch

Genome evolution: Sex and the transposable element Stephen Wright* and David Finnegan†

Mating systems are thought to play an important role in retrotransposons, which move by reverse transcription of determining the fate of genomic parasites such as an RNA intermediate, and transposons, which move by a transposable elements. This is supported by recent somewhat more straightforward process of excision and studies which indicate that asexual genomes may be insertion (Figure 1). All transposable elements are repeated structured very differently to those of sexual species. within the genomes in which they are found, as transpositon usually leads to an increase in copy number, either Addresses: *Institute of Cell, Animal and Population Biology, University of Edinburgh, King’s Buildings, Edinburgh EH9 3JT, UK. †Institute of because its mechanism is intrinsically replicative, as in the Cell and Molecular Biology, University of Edinburgh, King’s Buildings, case of retrotransposons, or because of secondary events Edinburgh EH9 3JR, UK. such as the repair of double-stranded breaks generated E-mail: [email protected]; [email protected] during transpositon. Current Biology 2001, 11:R296–R299 Retrotransposons come in two flavours (Figure 1a), those 0960-9822/01/$ – see front matter with long terminal repeats (LTRs) equivalent to those of © 2001 Elsevier Science Ltd. All rights reserved. retroviruses, the LTR retrotransposons, and those with no such repeats, the non-LTR retrotransposons or LINE-like Transposable elements, which were once thought to be elements. Although both types of element transpose by genetic oddities, are now believed to be a major factor in reverse transcription, the details of the process differ genome evolution, making up a substantial proportion of markedly between the two and their reverse transcriptases the DNA in most species. Transposable elements are can be readily distinguished at the sequence level. The segments of DNA with the ability to move from one place similarity between LTR retrotransposons and retroviruses in the genome to another. They fall into two main classes: extends beyond the LTRs to the proteins they encode, and elements similar to the gypsy elements of Drosophila have genes equivalent to retroviral gag, pol and env genes. Figure 1 Transposons are more uniform in structure, having terminal inverted repeats flanking a gene coding for the trans- (a) Retrotransposons posase (Figure 1b). There are several distinct families of transposon, the most widespread containing elements LTR retrotransposons related to mariner elements of Drosophila and Tc1 elements LTR‘gag’ ‘pol’ LTR of Caenorhabditis elegans.

‘env’ At the genome level, one of the most striking differences LTR‘gag’ ‘pol’ LTR between eukaryotic species is their abundance of transposable elements. As an example, surveys of retrotransposons in flowering plants have demonstrated Non-LTR retrotransposon between-species differences in copy number over four NAB RT orders of magnitude [1]. Among animals, our own ‘obese’ An human genome comprises nearly 50% transposable elements [2], while the genome of the nematode C. elegans (b) Transposons has less than 5% [3]. These patterns have long puzzled biologists; why should some organisms harbour enormous Transposase numbers of actively self-replicating repetitive sequences Current Biology while others contain almost none? Several recent studies provide support for one long-standing hypothesis, which Transposable elements. (a) Retrotransposons. LTR retrotransposons may explain some of this variability; the dynamics and fate have long direct repeats, LTRs, at their termini and coding sequences of transposable elements are strongly influenced by the similar to the gag and pol genes of a retrovirus; some, like the gypsy element of Drosophila, have an env like gene as well. Non-LTR mode of reproduction. retrotransposons have no terminal repeats and usually have two coding sequences. The first, NAB, codes for a nucleic acid binding A dominant model for the control of element copy protein and the second, RT, a protein with reverse transcriptase number in natural populations is the parasitic DNA activity. (b) Transposons have terminal inverted repeats and encode a single protein, transposase, required for transposition. hypothesis [4]. It proposes that elements persist in populations as a function of their ability to self-replicate, Dispatch R297

despite having potentially deleterious effects on their host. Figure 2 Element abundance in populations can then be controlled by a balance between the forces of transposition increas- (a) ing element abundance, and the action of purifying natural selection at the host level removing individuals with high copy number [5]. Host regulatory mechanisms may also play an important role in reducing rates of transposition. This model does not preclude the possibility of rare beneficial insertions, but it does assume that the average effect of transposable element insertions is deleterious. There is substantial evidence from both natural populations and experimental studies that, on average, the effects of transposable elements are in fact deleterious [5,6].

If transposable elements are genomic parasites, then rates of transmission are expected to play a fundamental role in their evolutionary and population dynamics. If transmission rates are high between individuals, parasites easily spread to the entire population. In contrast, under low transmission, parasites are unlikely to spread if they have deleterious effects. Sexual reproduction and outcrossing provide transposable elements with a means of spreading to all individuals in a population. But in the absence of a (b) high frequency of horizontal transmission, rates of between- line infection are reduced to effectively zero in asexual populations, preventing the spread of transposable elements [7] (Figure 2).

Experimental studies in yeast have provided support for the prediction that transposable elements are less able to increase in frequency in asexual populations [8]. Researchers introduced active elements into sexual and asexual lines, and found a lower probability of spread, and a reduced abundance of genomic parasites in asexual populations relative to their sexual controls. However, the relevance of these data for understanding transposable element distributions in natural eukaryotic populations has remained unclear. Obligate asexual reproduction has been extremely difficult to demonstrate, as presumed cases of ancient asexual taxa often fail to hold up under close scrutiny. Furthermore, asexuality is a derived state in higher eukaryotes, so asexual species are likely to have Current Biology inherited transposons from their sexual ancestors alleviat- ing the need for initial spread. Effects of mode of reproduction on transposable elements. Circles represent individuals in a population, over four generations of (a) asexual reproduction and (b) sexual reproduction. Arrows What, then, is the expected fate of transposable elements represent parental contributions to progeny. Red circles are individuals present in a host which has secondarily evolved asexual infected with at least one transposable element, represented by bars. reproduction? Here again, predictions from models of Transposable elements are unable to spread to other lineages in an host–parasite evolution [9] may be relevant. If transmis- asexual (a) population and are therefore lost by genetic drift and/or purifying selection. In a sexual population (b), outcrossing and sion rates are high between individuals, transposable ele- recombination allow elements to spread throughout the population. ments are expected to be virulent; transposition in the host increases their fitness, as they can easily spread to new individuals. In contrast, if transmission rates are neg- present in the genome, the evolution of obligate asexual ligible, it is difficult for parasitic behaviour to persist, as reproduction leads to a reduction in the selective advan- elements are unable to escape their current genomic tage for replication, and may cause stochastic accumula- background. For transposable elements that are already tion of mutations in the genes involved in transposition, R298 Current Biology Vol 11 No 8

and their gradual extinction. Indeed, selection acting on albicans is a potential candidate, as it has been thought to the host should favour nonfunctional elements and, in the have no sexual cycle and the Candida genome project absence of any counterselection at the element level, allows direct screening of the genome for retrotransposons transposable elements are eventually expected to degen- [15]. This has uncovered sequences from several families erate over evolutionary time. of LTR retrotransposons, although most copies are incom- plete and it is not clear whether any functional elements The most thorough test of these ideas was reported are present, perhaps indicating a relatively recent adoption recently by Arkhipova and Meselson [10], who have of asexuality by Candida albicans. Indeed, this yeast species looked for transposable elements in 46 species represent- has recently been persuaded to mate after the manipula- ing 26 phyla. These included ten species of rotifers, five tion of mating type genes detected in its genome [16,17]. bdelloid species believed to have been asexual for several million years, and five species from classes known to Among prokaryotes, the complete genome sequencing undergo sexual reproduction. These DNAs were screened of the ancient endosymbiont Buchnera failed to identify for the sequences corresponding to elements from three any mobile elements [18], as is often the case with mito- types of transposable element: gypsy-like LTR retrotrans- chondrial genomes [6]. This is not surprising, as recent posons, non-LTR retrotransposons and mariner/Tc1-like population genetic analysis suggests a complete lack of transposons. This was possible because the enzymes recombination in this group [19], in contrast with its close required for transposition, reverse transcriptases and trans- relative Escherichia coli, which appears to have appreciable posases, are conserved between families of related ele- levels of both genetic exchange and transposable elements, allowing them to be detected using polymerase ments [20]. However, Buchnera also lacks many genes chain reactions (PCR) and primers for their most highly essential for the survival of free-living bacteria, and the conserved sequences. lack of repetitive DNA may also reflect strong selection for a compact genome coinciding with the evolution of a Primers for retrotransposons amplified fragments of the symbiotic life style. expected size from DNAs of each of the species tested, except for those of the five bdelloid rotifers. When the If evidence accumulates that asexual genomes have amplified fragments were examined by DNA sequencing, substantially reduced numbers of active transposons, this they were found to contain reverse transcriptase sequences will not only have implications for the evolution of genome of the expected type. In contrast, there was no successful structure, but may also relate to our understanding of the amplification of either gypsy-like or non-LTR elements in puzzling persistence of asexuality in lineages such as the the bdelloids. While the PCR-based approach does not bdelloids. In organisms with large numbers of active ele- allow for the detection of ancient remnants of transposable ments, transposable elements contribute significantly to elements which may remain in the genome, it seems clear the rate at which deleterious mutations are introduced in that bdelloids lack functional retrotransposons, and that populations. Theoretical models predict that high rates of this is the derived state. This is the result expected if deleterious mutation favour the maintenance of sexual bdelloid rotifers have indeed been asexual for a long time reproduction [21]. If the genomes of these ancient asexual and retrotransposons are simply genomic parasites. organisms have lost active transposable elements, the rate of deleterious mutation may be substantially reduced in Primers for transposase genes of mariner-like elements, on these populations [22], which will decrease the selective the other hand, amplified fragments from 24 of the 34 pressure to acquire a sexual cycle again, allowing them to species tested, including all five of the bdelloid rotifers. remain celibate over evolutionary time. The presence of transposons in the bdelloids is not neces- sarily at odds with the distribution of retrotransposons, as Acknowledgements extensive surveys of the distribution of mariner-like ele- We thank Doris Bachtrog and Brian Charlesworth for helpful discussions of ments suggest that they undergo horizontal transfer rela- the manuscript. tively frequently on an evolutionary time scale [11,12]. References Quite why this should be true of transposons but not retro- 1. Kumar A, Pearce SR, McLean K, Harrison G, Heslop-Harrison JS, transposons is not clear, particularly as LTR retrotrans- Waugh R, Flavell AJ: The Ty-1 Copia group of retrotransposons in posons like gypsy have an ‘env’ gene, and gypsy itself has plants: genomic organisation, evolution, and use as genetic markers. Genetica 1997, 100:205-217. been shown to form particles which are infective in 2. The Human Genome Sequencing Consortium: Initial sequencing Drosophila melanogaster [13,14]. It is also unclear whether and analysis of the human genome. Nature 2001, 409:860-921. horizontal transfer could be frequent enough to allow trans- 3. Kidwell MG, Lisch DR: Transposable elements and host genome evolution. Trends Ecol Evol 2000, 15:95-99. posable elements to become fixed in an asexual population. 4. Doolitle, WF, Sapienza, C: Selfish genes, the phenotypic paradigm and genome evolution. Nature 1980, 284:601-603. 5. Charlesworth B, Langley CH: The population genetics of It is not easy to identify other suitable eukaryotic asexual Drosophila transposable elements. Annu Rev Genet 1989, species to which this analysis could be extended. Candida 23:251-287. Dispatch R299

6. Zeyl C, Bell G: Symbiotic DNA in eukaryotic genomes. Trends Ecol Evol 1996, 11:10-15. 7. Hickey D: Selfish DNA: a sexually transmitted nuclear parasite. Genetics 1982, 101:519-531. 8. Zeyl C, Bell G, Green DM: Sex and the spread of retrotransposon Ty3 in experimental populations of Saccharomyces cerevisiae. Genetics 1996, 143:1567-1577. 9. Ewald PW: The evolution of virulence: a unifying link between parasitology and ecology. J Parasitol 1995, 81:659-669. 10. Arkhipova I, Meselson M: Transposable elements in sexual and ancient asexual taxa. Proc Natl Acad Sci USA 2000, 97:14473-14477. 11. Robertson HM: The Tc1-mariner superfamily of transposons in animals. J Insect Physiol 1995, 41:99-105. 12. Robertson HM, Lampe DJ: Recent horizontal transfer of a mariner transposable element among and between Diptera and Neuroptera. Mol Biol Evol 1995, 12:850-862. 13. Song SU, Kurkulos M, Gerasimova T, Boeke JD, Corces VG: An env- like protein encoded by a Drosophila retroelement: evidence that gypsy is an infectious retrovirus. Genes Dev 1994, 8:2046-2057. 14. Kim AI, Terzian C, Santamaria P, Pelisson A, Prid’homme N, Bucheton A: Retroviruses in invertebrates: the gypsy retrotransposon is apparently an infectious retrovirus of Drosophila melanogaster. Proc Natl Acad Sci USA 1994, 91:1285-1289. 15. Goodwin TJ, Poulter RT: Multiple LTR-retrotransposon families in the asexual yeast Candida albicans. Genome Res 2000, 10:174-191. 16. Magee BB, Magee PT: Induction of mating in Candida albicans by construction of MTLa and MTLalpha strains. Science 2000, 289:310-313. 17. Hull CM, Raisner RM, Johnson AD: Evidence for mating of the ‘asexual’ yeast Candida albicans in a mammalian host. Science 2000, 289:307-310. 18. Shigenobu S, Watanabe H, Hattori M, Sakaki Y, Ishikawa H: Genome sequence of the endocellular bacterial symbiont of aphids Buchnera sp. APS. Nature 2000, 407:81-86. 19. Funk DJ, Wernegreen JJ, Moran N: Intraspecific variation in symbiont genomes. Bottlenecks and the aphid-buchnera association. Genetics 2001, 157:477-489. 20. Hartl DL: Population genetics of microbial organisms. Curr Opin Genet Dev 1992, 2:937-942. 21 Kondrashov AS: Classification of hypotheses on the advantage of amphimixis. J Hered 1993, 84:372-387. 22 Keightley PD, Eyre-Walker A: Deleterious mutations and the evolution of sex. Science 2000, 290:331-333.