Role of BMP and FGF signaling during limb development

发育的一个基本问题是确定细胞如何决定保持多能性或分化。这个问题的核心在于模式是如何在胚胎中产生的，反过来，当这些过程在肿瘤形成中被破坏时，模式是如何被破坏的。发育中的肢体是解决这些问题的理想系统，因为遗传学和实验胚胎学的研究提供了丰富的关于肢体发育的数据。之前，我们通过检测缺乏编码FGF配体基因的小鼠，发现了成纤维细胞生长因子（FGF）信号在肢体发育过程中的重要作用（Lewandoski et al. 2000 Nature Genetics 28:7 7; Sun et al. 2000 Nature Genetics 25:6）。然而，Fgf基因冗余导致的复杂性使我们考虑通过检查Fgf受体突变来解决这个问题，从而得出Fgf信号控制肢体大小的见解（Verheyden et al, 2005 . development 132:4235）。骨形态发生蛋白（Bone Morphogenetic Proteins, BMPs）调控的信号通路被认为在肢体生长的各个方面发挥作用：所有三个轴的早期模式、程序性细胞死亡和骨形成。摆在我们面前的任务是了解BMP和FGF信号通路在肢体发育过程中如何相互作用。BMP被认为在一个过程中发挥直接作用，即作为发生在间充质指间细胞的正常程序性细胞死亡的效应器，从而在出生时没有蹼肢的动物中去除它们并塑造最终的指骨模式。然而，我们在最近的出版物中对这种范式提出了挑战（Pajni-Underwood S. et al . 2007 Development 134: 2359）。通过同时灭活肢体芽外胚层中Bmp受体基因Bmpr1a以及Fgf8和Fgf4，我们已经为一种新的模型提供了遗传证据，在这种模型中，表面外胚层必须接收Bmp信号，导致Fgfs下调，进而诱导潜在间质凋亡（Pajni-Underwood S. et al 2007 Development 134: 2359）。因此，我们证明bmp通过调节FGF信号间接控制程序性细胞死亡。我们通过在肢体发育的特定区域表达Cre的小鼠品系研究BMP和FGF信号在肢体发育的各个方面的作用来扩展这些研究。例如，检验bmp作为细胞死亡的直接效应器这一假设的唯一方法是仅在经历细胞死亡的谱系中灭活bmp受体，而不影响附近细胞中FGF的表达。我们使用新的Cre系实现了这一点，这些Cre系允许Cre介导的基因失活。这些细胞系都在问：bmp是正常程序性细胞死亡的直接效应器吗？如果不是，程序性细胞死亡是如何控制的？如果是这样，bmp是如何实现这一目标的？在另一项研究中，我们发现了FGFs和BMP之间一个重要的信号传导节点，这对肢体骨骼的正常发育至关重要。我们之前的研究表明，从肢体芽的远端结构分泌的特定FGFs调节肢体的正常生长和模式。在目前的工作中，我们正在产生遗传证据，表明骨形成蛋白的信号传递给骨骼元件的祖先群体，通过控制FGF拮抗剂的表达来调节FGF信号。这两种信号通路的联系不仅是对肢体如何形成模式的独特见解，而且可能为两种通路在其他发育环境或癌症期间如何相互作用提供模型。

One fundamental problem in development is to determine how cells decide to remain multipotent or to differentiate. This problem lies at the heart of how pattern is generated in the embryo and, conversely, how pattern is disrupted when these processes are damaged as in tumor formation. The developing limb is an ideal system to address these problems because a wealth of data exists concerning limb development due to studies in genetics and experimental embryology. Previously, we have shown an essential role for Fibroblast Growth Factor (FGF) signaling during limb development by examining mice lacking genes that encode FGF ligands (Lewandoski et al. 2000 Nature Genetics 28:167, Sun et al 2000 Nature Genetics 25: 6 ). However, the complexity caused by Fgf gene redundancy has led us to consider approaching the problem by examining FGF receptor mutants leading to the insight that FGF signaling controls limb size (Verheyden et al 2005Development 132:4235). Signaling through the pathway governed by Bone Morphogenetic Proteins (BMPs) is thought to play a role in all aspects of limb outgrowth: early patterning in all three axes, programmed cell death and bone formation. The task before us is to understand how BMP and FGF signaling pathways interact during limb development. One process that BMP are thought to play a direct role in is as effectors of normal programmed cell death that occurs in mesenchymal interdigit cells, thus removing them and sculpting the final digit pattern in animals that are born without webbed limbs. However, we have challenged this paradigm in our recent publication (Pajni-Underwood S. et al 2007 Development 134: 2359). By simultaneously inactivating the Bmp receptor gene, Bmpr1a as well as Fgf8 and Fgf4 specifically in the limb bud ectoderm, we have produced genetic evidence for a novel model in which the surface ectoderm must receive a BMP signal, resulting in down regulation of Fgfs which in turn induces apoptosis of the underlying mesenchyme (Pajni-Underwood S. et al 2007 Development 134: 2359). Thus we demonstrated that BMPs control programmed cell death indirectly, by regulating FGF signaling. We have extended these studies by studying the role of BMP and FGF signaling in various aspects of limb development using mouse lines that express Cre in specific region of the developing limb. For example the only way to test the hypothesis that BMPs act as direct effectors of cell death is to inactivate BMPs receptors only in the lineage that undergoes cells death, without affecting FGF expression in nearby cells. We have achieved this using new Cre lines that allow Cre-mediated gene inactivation in these lineages. With these lines are asking: are BMPs are direct effectors of normal programmed cell death? If not, how is programmed cell death controlled? If so, how do BMPs achieve this endpont? In another study, we have uncovered an important node of signaling between FGFs and BMP that is essential for normal development of the limb skeleton. Our previous work, cited above, demonstrates that specific FGFs, secreted from a distal structure in the limb bud, regulate the normal outgrowth and patterning of the limb. In current work, we are generating genetic evidence that BMP signaling to the progenitor population of the skeletal elements regulates this FGF signal by controlling the expression of an FGF antagonist. This linking of the two signaling pathways is not only a unique insight into how the limb is patterned but may provide a model for how the two pathways interact in other developmental contexts or during cancer.