“Know Thyself”

Sage advice, particularly if you are a hermaphrodite organism seeking to take advantage of sexual reproduction. Promotion of out-breeding requires that male gametes do not fertilize the eggs being released from the same organism, or only do so at very low efficiency. This problem impacts a variety of organisms from flowering plants (Rea and Nasrallah, 2008) to ascidians, the sea squirts. Indeed the basic process is equally fundamental to the vertebrate immune response (Burnet, 1971). Interestingly the receptor/ligand pairs that mediate gamete self/non-self discrimination (allorecognition) are not phylogenetically conserved, indicating that these systems are a striking example of convergent evolution. Ciona intestinalis (photo by Lixy Yamada) In the ascidian Ciona intestinalis it has long been known that the key allorecognition event occurs at the level of the egg's vitelline coat (Morgan, 1923). In the current issue, Akira Yamaguchi and colleagues from Hitoshi Sawada's laboratory move us a significant step closer to understanding how the gametes of C. intestinalis manage this task. Previous work from this group had identified two highly polymorphic, unlinked loci: themis A and B, which each encode two products, a sperm membrane receptor (s-themis) and an egg coat ligand (v-themis). Should the sperm s-themis receptors bind to v-themis ligand encoded by the same allele, “self” is signaled and fertilization is prevented (Harada et al., 2008). While this elegant system explains Morgan's observations on gamete compatibility, which gave rise to the haploid sperm hypothesis, it is not complete. After all it is the failure to recognize self, for example, no binding between s-themis on the sperm head and v-themis in the vitelline coat, that is the signal for fertilization. Clearly there must be additional receptor–ligand pairs involved in sperm–egg interactions. A clue comes from other studies by the same group using a second ascidian species, Halocynthia roretzi. This work implicated a second pair of proteins: HrVC70, an EGF repeat protein in the vitelline coat, and HrUrabin, a GPI linked CRISP protein on the sperm plasma membrane (Urayama et al., 2008). Blockade of the HrUrabin/HrVC70 interaction blocks fertilization in this species. Is a similar interaction occurring in C. intestinalis? In the current article, the authors clone CiUrabin, a protein that interacts with the C. intestinalis ortholog of HrVC70, CiVC57. Despite low levels of amino acid identity, both CiUrabin and HrUrabin are members of the CRISP family of proteins and have a similar domain structure, have key conserved cysteine residues, and appear to be attached to the sperm membrane by GPI linkages. This allows the proposal of an elegant two-step model to allorecognition: (1) initial binding of sperm of C. intestinalis to the vitelline coat via CiUrabin/CiVC57 is followed by (2) an allorecognition event mediated by the themis proteins. If there is no recognition of self then fertilization proceeds, but if s-themis A and B bind to their allelic v-themis then a signaling cascade triggers the shedding of the sperm from the vitelline coat and self-fertilization is prevented. This model raises a number of questions. One is the nature of the signaling cascade derived from s-themis receptors. This signaling event must be rapid enough to prevent sperm from penetrating the vitelline coat, but it requires that both s-themis proteins (A and B) be bound to prevent fertilization. How does the spermatozoon manage this task? While both s-themis proteins are members of the polycystin family, only s-themis-B is structurally complete with 11 transmembrane domains. The s-themis A protein is truncated, suggesting that two distinct signaling cascades are activated and must cooperate to block fertilization. A second question is the diversity in the molecules ascidians use to mediate gamete recognition, and how this relates to the allorecognition event. In H. roretzi HrVC70 is highly polymorphic and has been proposed to play an indirect role in gamete allorecognition as it shows a higher binding affinity towards non-self sperm (Sawada et al., 2004). It is unknown if a similar phenomenon occurs in C. intestinalis. Divergence of the gamete allorecognition system within ascidians may not be that surprising given that in the colonial form of Botryllus schlosseri, the allorecognition molecules that govern colony fusion versus rejection are distinct from both the themis and HrVC70/CiVC57/Urabin proteins (McKitrick et al., 2011). The lack of primary sequence conservation between the HrUrabin and CiUrabin proteins may also hint at selective pressures driving their divergence. Analysis of the allelic structures of CiVC57 and also determining the presence or absence of themis loci within H. roretzi will be essential. It may well be that there are many paths to self-knowledge.