The marine organism Oikopleura dioica is interesting for more than one reason. Firstly, the Oikopleurides family, of which this tiny animal is a member, is the second most abundant group of animals in the plankton of all the world’s seas. Furthermore, Oikopleura dioica has the smallest genome known to date in animals - only 72 Mb. This makes it an organism of choice for a sequencing project, and in addition, the animal multiplies easily and rapidly in captivity. Finally and above all, Oikopleura dioica belongs to the phylum Chordata, which also includes the Vertebrates. Its phylogenetic position makes it particularly interesting for the genomic study of the emergence of the Vertebrates and of their fundamental characteristics.
The Chordates, which appeared at least 600 million years ago, are animals which possess a notocord - a dorsal, rigid rod of turgescent cells which has the function of supporting the animal along its axis. The Chordates also have a dorsal neural tube. The phylum Chordata includes three sub-phylums: the Urochordates, or Tunicates, to which Oikopleura dioica belongs; the Cephalochordates, represented by the famous Amphioxus, a small marine organism which looks like a fish; and the Vertebrates, which constitute the sister group of the Cephalochordates, and in which the notocord disappears during development, to be replaced by the vertebral column. Within the Tunicates, Oikopleura belongs to the class Appendicularia. These animals conserve their notocord throughout their life. They differ in this respect from other tunicates, the ascidians, whose aquatic, tadpole-like larvae possess a notocord, but undergo a metamorphosis at the end of which this structure is lost; adult ascidians are immobile sack-like animals which do not look at all like their vertebrate cousins.
Interest in invertebrate chordates has increased tremendously recently, when it was confirmed that the molecular bases of numerous aspects of vertebrate development have been conserved in these animals. For example, the Brachyury gene plays a similar role in the development of the tail in both vertebrates and tunicates, and the homolog of the Pax-2 gene in tunicates intervenes in the establishment of the frontier between the mesencephalon and the rhombencephalon, as in vertebrates. Complex aspects of the development of vertebrates, such as the organization of the central nervous system, can therefore be studied profitably in these invertebrate chordates, which are simpler animals than the vertebrates.
The most frequently studied animals among the Cephalochordates and the Urochordates respectively are two species of amphioxus, Branchiostoma floridae and belcheri, and several species of ascidians including Ciona intestinalis, for which the 160 Mb genome was sequenced in 2001-2002. Amphioxus does not reproduce in captivity, which limits its usefulness as a model. Ciona intestinalis, on the other hand, has already been studied in numerous projects due to its simple growth and maintenance in an aquarium, its rapid embryogenesis and the ease of following cellular differentiation in the larval form. However, the adult form is a relatively large animal, sessile, and very different from the larva because the typical chordate body plan has been lost. Furthermore, the genetics of Ciona is still at a very early stage.
The choice of Oikopleura dioica as a tunicate model has been motivated by several characteristics which make the study of this animal quite complementary to that of Ciona. First, as in all the Appendicularia, the adult form, produced by a metamorphosis which is less severe than that of Ciona, conserves the typical chordate body plan, and in particular a mobile tail (even though the animal lives in a little gelatinous house secreted by the epidermis). Comparison with the vertebrates is therefore easier. The Oikopleura adult is very small (five millimeters long, with a trunk of only one millimeter), and hundreds of individuals may easily be produced in simple beakers of seawater supplied with microscopic algae. Another advantage is its extremely short generation time (four days at 20(C, compared with three months for Ciona intestinalis). Methods for maintaining and breeding these animals have been developed at the Sars centre at Bergen in Norway, and inbred lines have already been obtained by repeated brother-sister matings. This combination of characteristics confers to Oikopleura dioica (which is a main or secondary focus of research for at least eight groups in the world) a great potential for large-scale genetic studies. This tunicate model thus approaches the power of other invertebrate models such as Drosophila or Caenorhabditis elegans. Moreover, methods for transgenesis using retroviral vectors are being developed at the Sars centre. Finally, a major advantage of Oikopleura dioica is its genome of only 72 Mb, the smallest known in the animal world. This value was estimated by flow cytometry, and has been confirmed by different statistics on genome sequences (see below); it affirms the extreme compactness of this genome (very short intergenic sequences and introns), which is an advantage for a sequencing project, both at the level of assembly (fewer problems due to repeated sequences) and at the annotation level.
The discovery of the small size of the genome of Oikopleura dioica led the Sars international centre for marine biology (Bergen, Norway), as soon as 1999, to initiate studies based on data from large-scale sequencing. Two groups, those of Daniel Chourrout and EricThompson, are working on Oikopleura dioica at Sars, and have already obtained significant results regarding the organization of the genome and the chromatin, development, the cell cycle and gene expression. The Sars scientists have obtained the support of the Max Planck Institute for Molecular Genetics (MPIMG) in Berlin (Hans Lehrach’s department) for a first step of random sequencing. The assembly of these first shotgun reads in the beginning of 2002 produced 44 000 contigs covering 41 Mb of the Oikopleura genome. These random reads have confirmed, via their level of assembly and alignment of EST and BAC collections, the small size of the genome, with estimates which are even lower than 70 Mb. The MPIMG has also sequenced 25 BAC clones which contain genes of interest, such as Hox genes.
In 2003, the Sars centre decided to extend this initial sequencing effort, and proposed a project to be carried out at Genoscope : global random sequencing at 10X coverage, with the goal of obtaining an assembly covering the ensemble of the euchromatic portion of the genome. The assembly will profit from the use of an inbred line, which is for the moment a unique opportunity for the deuterostome invertebrates (this is indeed not the case for Ciona, amphioxus or the two sea urchins which are currently being sequenced).
What can we learn about chordate innovations from the analysis of the genome of Oikopleura dioica? The results obtained from the genome draft of Ciona intestinalis can give us an idea. The annotation of Ciona, as well as the preliminary analysis of sequence data from Oikopleura, indicate that both of these tunicates possess about 15,000 genes, about half the number of genes in vertebrates. This would reflect the importance of gene duplication events which occurred in the vertebrate lineage after its separation from the rest of the chordates. The genes of Ciona and Oikopleura thus have less redundancy than those of the vertebrates, which is an advantage for experiments in functional genomics.
Can we then consider that the genomes of these two tunicates correspond to the genome of the ancestral chordate, the last common ancestor of the ensemble of the chordates? An ascidian such as Ciona intestinalis is a highly specialized organism which has doubtlessly lost numerous genes of this ancestral chordate and has acquired others since the divergence of its lineage. The evolutionary distance between Ciona and Oikopleura, as well as their major differences in organization, will lead to a better model of the genetic identity of the urochordates and the results of their diverse adaptations, but also to an understanding of their differences from vertebrates and from non-chordate invertebrates. Using this information, we will be able to imagine the genome of the common ancestor of all the chordates with more precision. Comparison with the genomes of other invertebrates will reveal genes which have arisen specifically in the chordate lineage, while the comparison between the two fish and three mammalian genomes which have already been sequenced will reveal the genes which have appeared early in the vertebrate lineage (especially of interest are the neural genes and the primordial genes of the adaptive immune system).