Objective 1. Regulation of Anterior-Posterior (AP) digit pattern: How does Sonic hedgehog (Shh) signaling regulate the numbers and types of digits made?: Shh acts as a mitogen and cell survival factor in many adult processes, but acts as a morphogen in several developmental contexts. In the limb, Shh regulates both digit number and identity of different digits (A-to-P, thumb to pinky). Shh is thought to act as a morphogen forming a gradient along the limb AP axis, with higher concentrations specifying more posterior digit types. We determined the time-requirements for Shh function in limb (using a tamoxifen-regulated Cre to remove Shh at different times in mice). We find an invariant order of digit loss that is not consistent with predictions from current morphogen gradient models. The order of digit loss correlates inversely with the order in which digit primordia first form, and the phenotype severity at different times correlates with degree of apoptosis and decrease in mitotic index. Our results suggest that Shh regulates digit pattern only very early and transiently and later is needed mainly to ensure cell survival and/or proliferation. If the limb bud has fewer cells, fewer digit primordia can form but those that do form are normal. We are testing this provocative new model by restoring cell survival and/or proliferation to rescue the Shh mutant (collaboration with Gerard Evan, UCSF). What is the role of Gli3-Hoxd interaction in digit pattern?: Hoxd genes are TFs that cooperate in an additive fashion to regulate digit pattern and are thought to be key targets of Shh signals. We previously found that Hoxd-Gli3 interactions serve to modify the function of Gli3 as a nuclear mediator of Shh by converting Gli3-repressor into an activator of its target promoters. We are extending this analysis to determine: 1) target promoters regulated by Gli3-Hoxd interaction and 2) physiologic role of Gli3-Hoxd interaction during limb development. While Hoxd genes are no longer expressed in the adult, other related Hox genes are expressed, have highly conserved in Gli3-binding domains and may modify Hh-Gli3 targets in other contexts, such as skin and gut, during normal renewal of these epithelia or during neoplastic proliferation. We have determined requirements for Gli3-HoxD protein interaction and are testing the functional effects of a dominant interfering form of Gli3 (peptide) in transfections and in chick embryos. Dependent on the outcome of such experiments, long-range plans to introduce Hox-interaction domain mutations in Gli3 into mice for analysis will be undertaken. Objective 2. Late regulation of digit morphogenesis. What are the time requirements for Hoxd gene function?: Digit identity remains plastic even after the formation of the digit primordial chondrogenic condensations and is regulated by interdigit zones, which are also late sites of Hoxd and Gli3 expression. Mutant analysis has shown that several Hoxd genes (Hoxd11,12,13) together regulate digit morphogenesis. Collaborating with Denis Duboule (Univ. Geneva), we are analyzing the time dependence of Hoxd function in the limb using a conditional Hoxd13-d11 knock-out and tamoxifen-dependent Cre. We find a late role for Hoxd genes; late loss of function at interdigit stages results in a phenotype very similar to the early Hoxd gene removal, with all biphalangeal digits (thumb-like), as occurs in human brachydactyly syndromes. We are analyzing the status of possible candidate signals/factors that may operate downstream of Hoxd genes at these late stages. In a parallel study collaborating with Alex Joyner (MSKCC, NY), temporal requirements for Gli3 function in limb are being similarly examined. What role do Hoxd genes play in cartilage differentiation?: In addition to interdigit mesenchyme, Hoxd genes continue to be expressed very late at the periphery of the cartilage models for future digit bones. Hoxd expression normally shuts off as cartilage differentiation proceeds, while remaining on at the periphery. To assess whether this shut-off is important to allow chondrogenic differentiation to proceed, we developed an inducible transgenic model to sustain expression of Hoxd gene expressing in forming cartilage. Our preliminary results indicate that shut off of Hoxd gene expression is necessary for early steps in cartilage differentiation to occur. We find that active Hoxd genes in cartilage precursors repress Sox9 expression (a master-regulator of cartilage fate). This repression may play a key role in the normal segmentation that leads to digit joint formation, which occurs by local reversal of the cartilage differentiation program. ChIP assays for Hoxd binding (see below) will help to clarify whether the Sox9 promoter is a direct Hoxd target and identify other genes that regulate cartilage differentiation and are Hoxd targets. What signals regulate final digit morphogenesis?: Phalange length and number (extent of digit segmentation) are regulated at late stages by interdigit signals. We are analyzing individual interdigit samples in reverse phase protein arrays for levels of activated signaling phospho-intermediates in pathways implicated in cartilage growth and joint segmentation (Bmp, Wnt, Fgf, Hh), in collaboration with Lance Liotta (George Mason Univ.), who pioneered this proteomic approach. We are evaluating interdigits in species with evolutionary adaptations of digit morphology, to correlate morphogenetic changes with changes in signaling activity, comparing three vertebrates: chick, mouse, and bat (collaboration with John Rasweiler, SUNY). Both bats and birds have evolved striking digit adaptations for flight and also have highly adapted hindlimbs. The proteomic analysis will be complemented with global analysis of gene expression using DNA microarrays, to screen for differences at the RNA expression level. Comparing both gene expression and signaling phosphorylation status in the interdigits of different organisms will provide new insights on how digit identity is regulated and evolutionary adaptation occurs. In the mouse, well-characterized mutants are available that have digit alterations such as brachydactyly (short, biphalangeal digits), as seen in Hoxd13-11 mutants. Global genomic and proteomic analyses will also be applied to these mutants to gain further insight into critical signaling pathways regulating digit morphology. Objective 3. Identification of target promoters regulated during limb morphogenesis. Limb Initiation: How does Tbx5 regulate limb outgrowth?: We are developing ChIP assays in embryos for genome-wide direct identification of target promoters in vivo for transcriptional regulators of limb development and unravel the regulatory networks operating during pattern formation and morphogenesis. Tbx5 is an excellent prototype for this analysis: 1) It is highly expressed in early forelimb bud (also heart); Tbx5 mutations cause cardiac and limb abnormalities in humans (Holt-Oram disease) and similar defects in mice. 2) A direct Tbx5 target, Fgf10, has been identified, which will aid in validation and troubleshooting; nevertheless most targets are unknown. We have generated and validated high affinity polyclonal antibodies against Tbx5 that efficiently and specifically enrich the Fgf10 promoter by ChIP, optimized the approach, and are analyzing pilot genome-wide promoter microarrays. Digit morphogenesis: What is the relation between Hoxd and Gli3 targets?: We plan to extend ChIP analysis to several other developmental regulators important in digit specification and patterning. We have generated polyclonal antibodies for Hoxd12 and Gli3 to identify direct binding targets for Hoxd and Gli3 proteins and elucidate the role of Hoxd-Gli3 interaction in gene regulation
