The objective of our research is to understand why meiosis and sexual reproduction occur in most eukaryotic species, and why the loss of these processes usually leads to early extinction.

Although asexual lineages occasionally arise from sexual populations, and initially may even thrive, they are almost invariably short-lived on an evolutionary time scale. In consequence, very few species of animals and plants reproduce only asexually and those that do rarely comprise an entire genus, let alone a taxon of higher rank. Despite this and other evidence that most asexual species do not last long enough to evolve higher taxa, there is no agreed explanation of what causes their early extinction, or of why meiosis and sexual reproduction prevent the early extinction suffered by most asexual lineages. Indeed, why meiosis and sex exist is a fundamental unsolved problem of biology.

Our approach to the problem is to ask if, contrary to the fate of most asexual lineages, there is a high-ranking taxon that has evolved without meiosis and sexual reproduction and, if so, to learn what has allowed it to escape extinction. The Class Bdelloidea of the Phylum Rotifera appears to be such an exception. Bdelloid rotifers are small fresh-water invertebrates comprising four families, 19 genera, and some 360 described species. Despite much study of field and laboratory populations since the first observation of bdelloids by Leeuwenhoek in 1677, the entire Class appears to be without males, hermaphrodites or meiosis. Eggs are produced from oocytes by two mitotic divisions, with no chromosome pairing and no reduction in chromosome number. Consistent with a lack of meiosis, bdelloid karyotypes contain chromosomes without morphological homologs.

Although remarkable, these observations do not preclude rare or unrecognized forms of sexual reproduction. We have therefore sought more rigorous evidence that bdelloid rotifers have evolved asexually. We have obtained three types of molecular and cytological evidence that strongly support the indication from the observations summarized above that the Bdelloidea evolved without meiosis and sex.

Our first approach made use of the fact that the genome of a sexually-reproducing diploid contains two sets of chromosomes that are kept closely homologous by segregation of haplotypes and genetic drift. In contrast, if reproduction is only mitotic and without segregation of any kind, sequences that were allelic will remain together, each gradually accumulating mutations, so that after many millions of years individual genomes will contain highly divergent copies of former alleles. In accord with this expectation, within individual genomes of diverse bdelloid species we found highly divergent copies of each of four genes examined (hsp82, RNApolIII, tbp , and tpi ) and no closely similar copies. Moreover, each copy of a given gene belongs to one or the other of two ancient lineages , which differ by approximately 50% at synonymous sites and which separated before the bdelloid radiation, when meiosis and sexual reproduction appear to have been lost. In contrast, as expected for sexually-reproducing diploids, we found only pairs of closely homologous copies in the genomes of individual rotifers of two other classes within the phylum, the facultatively sexual Class Monogononta and the obligately sexual Class Seisonida.

While the above results are in striking agreement with our expectation that former alleles would have become highly divergent in an anciently asexual lineage, they do not rule out an alternative possibility: Bdelloids could be sexually reproducing tetraploids, descended from a common tetraploid ancestor, with extremely low divergence between alleles, which therefore could have gone undetected in our sequencing studies. This possibility is ruled out, however, by our finding by fluorescent in situ hybridization (FISH) that bdelloid nuclei do not contain closely homologous chromosome pairs. For each of several bdelloid probes we tested, FISH to embryo mitotic nuclei labels only one chromosome, not pairs of chromosomes.

A third and altogether different line of evidence for asexual evolution of the Bdelloidea is our finding that, unlike all sexually-reproducing species tested, bdelloids lack both major families of transposable retroelements — LTR and non-LTR retrotransposons. The lack of these elements is consistent with the expectation that active deleterious vertically-transmitted elements will not persist in ancient asexual lineages. Eventually, either the host lineage will be driven to extinction by its increasing insertional load or, if it is to persist, such elements must be eliminated, attenuated, or inactivated. In sexually reproducing populations, however, deleterious mobile elements can persist indefinitely if recombination is sufficiently effective in constituting less-loaded genomes from more-loaded ones, or if some other process dependent on meiosis limits the deleterious load.

Having established that the Bdelloidea have almost certainly evolved without meiosis and sex, we have begun to use bdelloid rotifers to test various hypotheses that seek to account for the maintenance of sex. A principal class of hypotheses holds that sexual reproduction, by either stochastic or deterministic processes, limits the accumulation of deleterious nucleotide substitutions. This raises the possibility that bdelloids may have an unusually low mutation rate. Using relative rate tests and other methods of comparison, we found that the rates of synonymous and nonsynonymous nucleotide substitutions at hsp82 in bdelloids are not significantly different from those in the facultatively sexual rotifers of the class Monogononta.

We have suggested that it may be the limitation of deleterious insertions, rather than deleterious nucleotide substitutions that accounts for the maintenance of meiosis and sex. This would be consistent with the lack of retrotransposons in bdelloids and suggests that bdelloids either lost such elements close to the time when sex was lost, either by chance or else by some active mechanism of suppression or removal. In order to investigate these possibilities we are examining bdelloid genomes for relict retrotransposons. Phylogenenetic analysis of such relict sequences may allow us to determine the stage in bdelloid evolution when they ceased to be active and may even provide information regarding the mechanism of such inactivation. Recently, we have found what may be such a relict and are isolating and sequencing it from diverse bdelloid species. In order to detect possible active mechanisms by which bdelloids might suppress retrotransposition, we are introducing an active LINE-like retrotransposon into bdelloid genomes by microparticle bombardment and by gonadal injection followed by assay for transposition.

We are developing methods to render bdelloid genomes susceptible to genetic investigation. In addition to the introduction of transgenes mentioned above, we have developed a protocol for using RNAi to inactivate specific bdelloid gene transcripts.

Using the high-throughput sequencing capability of the Bay Paul Center, we have begun a basic characterization of the genomes from the major bdelloid orders. We will characterize and compare, inter alia, gene density, intronic structure, intergenic regions, relict or pseudo genes and insertions, discontinuities of sequence divergence, and Hox gene structure and organization. Moreover, in so unusual a genetic system as that of bdelloid rotifers, such genomic analysis may reveal important novel or unanticipated features.

Further opportunities for utilizing bdelloid rotifers in the study of basic genetic problems include investigations of DNA repair, genetic silencing, heterosis, population genetic diversity, and response to selection. We welcome inquiries regarding joining or collaborating with us.

Matthew Meselson is a member of the Mobile Genetic Element Cluster

Recent publications:

Mark Welch D.B, , J.L. Mark Welch, J.L., and M. Melselson. (2008) Evidence for degenerate tetraploidy in bdelloid rotifers. Proc Nat Acad Sci USA 105: 5145-5149.

Gladyshev, E., Meselson, M., 2008. Extreme resistance of bdelloid rotifers to ionizing radiation Proceedings of the National Academy of Sciences 10.1073/pnas.0800966105

Mark Welch D.B., M.P.Cummings, D.M Hillis and M. Meselson (2004). Divergent gene copies in the asexual class Bdelloidea (Rotifera) separated before the bdelloid radiation or within bdelloid families. Proc Natl Acad Sci U S A. 101 : 1622-1625.

Mark Welch J.L., D.B. Mark Welch and M.Meselson (2004). Cytogenetic evidence for asexual evolution of bdelloid rotifers. Proc Natl Acad Sci U S A. 101 : 1618-1621.

Meselson M. (2003). Interview with Matthew Meselson. Bioessays 12 : 1236-46.

Mark Welch D.B. and M. Meselson (2003) Oocyte nuclear DNA content and GC proportion in rotifers of the anciently asexsual Class Bdelloidea. Biological Journal of the Linnean Society 79 : 85-91.

Arkhipova, I.R., K.I. Pyatkov, M. Meselson and M.B. Evgen'ev (2003). Retroelements containing introns in diverse invertebrate taxa. Nature Genetics 33: 123-124.