Retinoids and regulation of anterior Hox genes: the evolutionary history of breaking collinearity – RAnteriorHox

In bilaterian animals, Hox genes play important roles in conveying anteroposterior patterning information to the embryo. On the chromosome, Hox genes are usually linked in clusters and their order in these clusters correlates with their expression during embryogenesis. This so-called collinear expression is conserved in many animals, including most invertebrate chordates and vertebrates. Although in the vertebrate central nervous system (CNS) spatial collinearity of Hox genes is generally respected, there are some important exceptions to this rule. In jawed vertebrates, such as mice, Hox1 and Hox2 genes break collinearity. Whereas the anterior limit of Hox1 genes is limited to the border between rhombomeres 3 and 4, the Hox2 genes extend anteriorly to the limit of rhombomere 2. Importantly, there are even differences among the anterior limits of distinct Hox2 paralogs in jawed vertebrates. Whereas Hoxa2 has an expression limit at the border between rhombomeres 1 and 2 in all vertebrates analyzed including lampreys, Hoxb2 expression does not extend into rhombomere 2, but stops at the boundary between rhombomeres 2 and 3. In contrast, in the amphioxus CNS Hox1 and Hox2 obey collinearity, with the anterior border of Hox1 expression being more rostral than that of Hox2. In chordates, it has been shown that retinoic acid (RA), an endogenous vitamin A-derived morphogen, is directly involved in regulating collinear Hox expression along the anteroposterior body axis of the embryo. For example, in the vertebrate CNS RA treatments lead to an upregulation and anterior shift of Hox gene expression. These data suggest a link between the observed differences in collinear Hox expression and possible differences in the RA signaling pathway of invertebrate chordates and vertebrates. Thus, the aim of the proposed research is to understand the origins of the collinearity break between Hox1 and Hox2 and to elucidate the exact roles of RA signaling in this process. Using a comparative approach involving different animal systems that mark key positions of chordate phylogeny we will address the following questions: • Are there differences in RA signaling activity in the CNS of invertebrate chordates, jawless vertebrates and jawed vertebrates? • What roles does RA signaling have in establishing the anterior limits of Hox1 and Hox2 in invertebrate chordates, jawless vertebrates and jawed vertebrates? • Why is the anterior limit of Hox1 posterior to that of Hox2 in jawed vertebrates, but not in invertebrate chordates, and what is the expression of Hox1 (relative to Hox2) in jawless vertebrates? • Why do the different Hox2 genes in jawed vertebrates have distinct anterior expression limits in the CNS? The proposed work can be subdivided into three different sections: (a) RA signaling activity in the chordate CNS. In order to elucidate, in which regions of the chordate CNS RA signaling is active, we are going to study the functions of the nuclear receptors mediating the RA signal (RAR and RXR) as well as the genes implicated in synthesis (Raldh) and degradation (Cyp26) of endogenous RA. Moreover, we are going to assess the activity of a RA-sensitive construct in the developing chordate CNS. Since these topics have already extensively been studied in jawed vertebrates, we are going to focus our studies mainly on the invertebrate chordate amphioxus as well as on lampreys, which are jawless vertebrates. (b) Expression of anterior Hox genes in the chordate CNS. Here, we are going to study the expression of anterior Hox genes (Hox1 and Hox2) during development and in response to activation or inhibition of RA signaling. Again, we are going to focus mainly on amphioxus and lampreys. Together with the data already published in jawed vertebrates, our work on invertebrate chordates and jawless vertebrates will allow us to assemble a comprehensive gene expression data set, which will represent an important basis for in-depth comparative analyses. (c) Regulation of anterior Hox genes by RA in chordates. Finally, we will investigate the potential molecular link between the direct regulation of anterior Hox genes by RA signaling and the break of collinearity of Hox1 and Hox2 in the chordate CNS. Hence, we will be testing the activity of reporter constructs carrying regulatory sequences of anterior Hox genes in different chordate models (amphioxus, lampreys, chicken, mice). For selection of the regulatory regions, specific attention is going to be given to the location of RA-responsive elements (so-called RAREs). Ultimately, by addressing the four questions raised in this proposal with our comparative experimental studies, we aim to resolve why this alteration of collinear expression of Hox1 and Hox2 genes took place during chordate evolution. This work is thus going to allow a reconstruction of the evolutionary events leading to the elaboration of the anterior CNS early during vertebrate diversification.