Evolution and expression of the highly variable cell adhesion molecule Dscam in the crustacean Daphnia and other arthropods

The Down syndrome cell adhesion molecule (Dscam) family, is within the cell adhesion molecules, a
family whose members are characterized by being composed of immunoglobulin (Ig) and fibronectin
domains and which are known to play an essential role in the development of the nervous system in both
vertebrates and invertebrates.
In insects, one member of the Dscam family diversified extensively due to internal exon duplications
and a sophisticated mechanism of mutually exclusive alternative splicing (AS). This enables a single
individual to generate somatically thousands of Dscam isoforms which differ in half of two Ig domains
and in another complete Ig domain. That creates a high diversity of adhesion properties which are used by
nervous cells and also by immune cells (hemocytes).
How this situation evolved is best understood my means of comparative studies. I have studied aspects
of the evolution and expression of this diversified member of the Dscam family mainly in the brachiopod
crustacean Daphnia magna and to lesser extent, in other representatives of the arthropod phyla. I have
shown that like in insects, a highly variable Dscam gene evolved in crustaceans, which also express
Dscam diversity in nervous and in immune cells. Additionally I could demonstrate that not only Dscam’s
ectodomains are diversified but that several cytoplasmic tails with different signal transduction capacities
can also be expressed. The comparison between Daphnia and insects revealed furthermore that there is
high amino acid conservation among distantly related species for most Dscam domains except for the Ig
regions that are coded by the multiple exons, suggesting that the latter evolved under different selective
constraints.
Dscam has been proposed as an exciting candidate molecule for mediating specific immune responses
in arthropods. Nevertheless, the involvement of Dscam in immunity remains largely elusive. I tested the
effect of parasite infection on the expression of total Dscam and on the diversity of some duplicated exons
at the RNA level and found no significant effect. Yet, hemocytes expressed reduced transcript diversity
relative to the brain, but each transcript was likely more abundant. This would be consistent with a
function in the immune system given that each Dscam isoform would be present in higher concentrations
which would increase their functional capacity.
Dscam isoforms engage in dimer formation with other identical isoforms, promoting cell-cell
recognition. It has been demonstrated that the variable parts of Dscam coded by the duplicated exons
mediate dimer formation. The genetic diversification caused by exon duplication and AS has thus direct
functional implications. I estimated signatures of selection on some of the regions involved in dimer
formation by comparing sequences from different Daphnia magna populations and from different species
of Daphnia and Drosophila. The results indicated that diversity created by duplication followed by
divergence is maintained by purifying selection against new mutations and against new gene conversion
events. That is consistent with the essential role of Dscam diversity in the nervous system. Contrastingly, I
found that some parts of the variable regions which are not involved in dimer formation and are oriented
towards the dimer’s external environment, may evolve under positive selection, which would be consistent
with an immune function.
To understand the evolutionary history of the molecule, I searched for Dscam related genes in
representatives of chelicerates (Ixodes scapularis) and myriapodes (Strigamia maritima), two other groups
of arthropods. In both myriapodes and chelicerates, Dscam diversified extensively by whole gene
duplications and by duplications of some internal exons coding for one Ig domain region, but not several,
like in insects and crustaceans. Similar duplications could have provided the raw material from which the
highly diverse Dscam evolved uniquely in the ancestors of crustaceans and insects. I propose a speculative
scenario under which the evolution of this remarkable gene might have occurred.