During animal development, the formation of the anteriorposterior (A-P) axis is of fundamental importance, because it sets the stage for subsequent tissue differentiation and organogenesis. Patterning along the A-P axis requires the expression of Hox genes to specify positional information. 1 In mammals, these genes are organized into 4 chromosomal (HOXA-D) clusters containing 39 members. Equivalent genes within each cluster are divided into 13 paralogous groups (HOX1-13). During A-P patterning, Hox genes are expressed in a temporal sequence that matches their 3' to 5' arrangement on the chromosome (temporal collinearity).
2This precisely controlled timing in Hox gene expression potentially provides a timing mechanism for the formation of A-P structures, that are laid down from head to tail progressively during early development.3 Actually, Hox temporal collinearity is prerequisite for establishing the correct spatial pattern of Hox gene expression, 4 which also reflects their 3' to 5' order on the chromosome (spatial collinearity). Therefore, not only is temporal collinearity an interesting feature of Hox genes, it is also likely to play a crucial role in A-P patterning. Nonetheless, the nature of Hox temporal collinearity has yet to be fully elucidated.In an attempt to understand what drives the Hox timer (temporal collinearity), we did hoxc6 loss of function (LOF) in the Xenopus by injection of a hoxc6 morpholino (MO) (5'-ATTCATATCTTCTCCTTTACCTGCC-3') at 2-cell or 4-cell stage. The effectiveness and specificity of this MO have been demonstrated previously.5 Remarkably, injection of 60ng hoxc6 MO cut off the posterior part of the axis completely, whereas injection of the same amount of control MO did not affect axis formation (Fig. 1A). Consistent with the phenotype, we observed that the expression domains of krox-20, a rhombomere marker, and otx-2, a midbrain marker, expanded posteriorly (Fig. 1B).We then asked: why did hoxc6 LOF give such a severe developmental consequence? To answer this question, we examined the expression of all the 9 genes in the HOXC cluster (hoxc4, 5, 6, 8, 9, 10, 11, 12 and13) and all the 3 genes in the hox6 paralogous group (hoxa6, b6 and c6) at st.26 (Fig. 1C). In hoxc6 LOF embryos, hoxc6 was downregulated by 50%, confirming the morpholino worked. The downregulation of hoxc6 was accompanied by upregulation of hoxc4 and c5 and by downregulation of most Hox genes that are posterior to hoxc6 in the HOXC cluster. These results indicate 2 previously identified types of Hox interactions within the HOXC cluster: posterior prevalence (the repression or antagonism of anterior Hox genes by posterior ones, e.g. the repression of hoxc4 and c5 by hoxc6), and posterior induction (the induction of posterior Hox genes by anterior ones, e.g., the induction of hoxc8,9, and 10-13 by hoxc6) (reviewed in Ref.4). Interestingly, hoxa6 and b6 were also downregulated by hoxc6 LOF, suggesting interactions also exist among Hox clusters. This explains why Hox genes from other clusters (e.g., hoxd3, a5, and d...