We have identified a novel zinc finger-containing transcription factor, called Osterix (Osx), that is specifically expressed in all developing bones. In Osx null mice, no bone formation occurs. In endochondral skeletal elements of Osx null mice, mesenchymal cells, together with osteoclasts and blood vessels, invade the mineralized cartilage matrix. However, the mesenchymal cells do not deposit bone matrix. Similarly, cells in the periosteum and in the condensed mesenchyme of membranous skeletal elements cannot differentiate into osteoblasts. These cells do, however, express Runx2/Cbfa1, another transcription factor required for bone formation. In contrast, Osx is not expressed in Runx2/Cbfa1 null mice. Thus, Osx acts downstream of Runx2/Cbfa1. Because Osx null preosteoblasts express typical chondrocyte marker genes, we propose that Runx2/Cbfa1-expressing preosteoblasts are still bipotential cells.

Calcification of the extracellular matrix (ECM) can be physiological or pathological. Physiological calcification occurs in bone when the soft ECM is converted into a rigid material capable of sustaining mechanical force; pathological calcification can occur in arteries and cartilage and other soft tissues. No molecular determinant regulating ECM calcification has yet been identified. A candidate molecule is matrix GLA protein (Mgp), a mineral-binding ECM protein synthesized by vascular smooth-muscle cells and chondrocytes, two cell types that produce an uncalcified ECM. Mice that lack Mgp develop to term but die within two months as a result of arterial calcification which leads to blood-vessel rupture. Chondrocytes that elaborate a typical cartilage matrix can be seen in the affected arteries. Mgp-deficient mice additionally exhibit inappropriate calcification of various cartilages, including the growth plate, which eventually leads to short stature, osteopenia and fractures. These results indicate that ECM calcification must be actively inhibited in soft tissues. To our knowledge, Mgp is the first inhibitor of calcification of arteries and cartilage to be characterized in vivo.

Potential repair by cell grafting or mobilizing endogenous cells holds particular attraction in heart disease, where the meager capacity for cardiomyocyte proliferation likely contributes to the irreversibility of heart failure. Whether cardiac progenitors exist in adult myocardium itself is unanswered, as is the question whether undifferentiated cardiac precursor cells merely fuse with preexisting myocytes. Here we report the existence of adult heart-derived cardiac progenitor cells expressing stem cell antigen-1. Initially, the cells express neither cardiac structural genes nor Nkx2.5 but differentiate in vitro in response to 5 -azacytidine, in part depending on Bmpr1a, a receptor for bone morphogenetic proteins. Given intravenously after ischemia͞reperfusion, cardiac stem cell antigen 1 cells home to injured myocardium. By using a Cre͞Lox donor͞ recipient pair (␣MHC-Cre͞R26R), differentiation was shown to occur roughly equally, with and without fusion to host cells. C ardiomyocytes can be formed, at least ex vivo, from diverse adult pluripotent cells (1-5). Apart from therapeutic implications and obviating ethical concerns aroused by embryonic stem cell lines, adult cardiac progenitor cells might provide an explanation distinct from cell cycle reentry, for the reported rare occurrence of cycling ventricular muscle cells (6). However, recent publications suggest the failure of certain stem cells' specification into neurons, skeletal muscle, and myocardium in vivo (7,8) and recommend greater conservatism in evaluating claims of adult stem cell plasticity, for cogent reasons (9-11).The rarity of cardiogenic conversion by endogenous hematopoietic cells (2, 12), requirements for intracardiac injection (3), or mobilization by cytokines (13), uncertain proof for myocytes of host origin in transplanted human hearts (14), and the confounding possibility of cell fusion after grafting in vivo (15, 16) highlight unsettled issues surrounding stem cell plasticity in heart disease. For donor cell types already in clinical studies, the predominant in vivo effect of bone marrow or endothelial progenitor cells may be neoangiogenesis, not cardiac specification (17, 18), and skeletal myoblasts, despite integration and survival, are confounded by arrhythmias, perhaps reflecting lack of transdifferentiation (19). These obstacles underscore the need to seek cardiac progenitor cells beyond the few known sources. Materials and MethodsFlow Cytometry and Magnetic Enrichment. A ''total'' cardiac population was isolated from 6-to 12-wk-old C57BL͞6 mice by coronary perfusion with 0.025% collagenase, as for viable adult mouse cardiomyocytes (20). More typically, a ''myocytedepleted'' population was prepared, incubating minced myocardium in 0.1% collagenase (30 min, 37°C), lethal to most adult mouse cardiomyocytes (20). Cells were then filtered through 70-m mesh. Bone marrow cells (21) were compared, with or without collagenase and filtration. Cells were labeled with stem cell antigen 1 (Sca-1)-phycoerythrin (PE), Sca-1-FITC, c-kit-PE; CD4-...

Chondrogenesis results in the formation of cartilages, initial skeletal elements that can serve as templates for endochondral bone formation. Cartilage formation begins with the condensation of mesenchyme cells followed by their differentiation into chondrocytes. Although much is known about the terminal differentiation products that are expressed by chondrocytes, little is known about the factors that specify the chondrocyte lineage. SOX9 is a high-mobility-group (HMG) domain transcription factor that is expressed in chondrocytes and other tissues. In humans, SOX9 haploinsufficiency results in campomelic dysplasia, a lethal skeletal malformation syndrome, and XY sex reversal. During embryogenesis, Sox9 is expressed in all cartilage primordia and cartilages, coincident with the expression of the collagen alpha1(II) gene (Col2a1) . Sox9 is also expressed in other tissues, including the central nervous and urogenital systems. Sox9 binds to essential sequences in the Col2a1 and collagen alpha2(XI) gene (Col11a2) chondrocyte-specific enhancers and can activate these enhancers in non-chondrocytic cells. Here, Sox9 is identified as a regulator of the chondrocyte lineage. In mouse chimaeras, Sox9-/- cells are excluded from all cartilages but are present as a juxtaposed mesenchyme that does not express the chondrocyte-specific markers Col2a1, Col9a2, Col11a2 and Agc. This exclusion occurred cell autonomously at the condensing mesenchyme stage of chondrogenesis. Moreover, no cartilage developed in teratomas derived from Sox9-/- embryonic stem (ES) cells. Our results identify Sox9 as the first transcription factor that is essential for chondrocyte differentiation and cartilage formation.

The liver and exocrine pancreas share a common structure, with functioning units (hepatic plates and pancreatic acini) connected to the ductal tree. Here we show that Sox9 is expressed throughout the biliary and pancreatic ductal epithelia, which are connected to the intestinal stem-cell zone. Cre-based lineage tracing showed that adult intestinal cells, hepatocytes and pancreatic acinar cells are supplied physiologically from Sox9-expressing progenitors. Combination of lineage analysis and hepatic injury experiments showed involvement of Sox9-positive precursors in liver regeneration. Embryonic pancreatic Sox9-expressing cells differentiate into all types of mature cells, but their capacity for endocrine differentiation diminishes shortly after birth, when endocrine cells detach from the epithelial lining of the ducts and form the islets of Langerhans. We observed a developmental switch in the hepatic progenitor cell type from Sox9-negative to Sox9-positive progenitors as the biliary tree develops. These results suggest interdependence between the structure and homeostasis of endodermal organs, with Sox9 expression being linked to progenitor status.

Chondrogenesis is a multistep process that is essential for endochondral bone formation. Previous results have indicated a role for ␤-catenin and Wnt signaling in this pathway. Here we show the existence of physical and functional interactions between ␤-catenin and Sox9, a transcription factor that is required in successive steps of chondrogenesis. In vivo, either overexpression of Sox9 or inactivation of ␤-catenin in chondrocytes of mouse embryos produces a similar phenotype of dwarfism with decreased chondrocyte proliferation, delayed hypertrophic chondrocyte differentiation, and endochondral bone formation. Furthermore, either inactivation of Sox9 or stabilization of ␤-catenin in chondrocytes also produces a similar phenotype of severe chondrodysplasia. Sox9 markedly inhibits activation of ␤-catenin-dependent promoters and stimulates degradation of ␤-catenin by the ubiquitination/proteasome pathway. Likewise, Sox9 inhibits ␤-catenin-mediated secondary axis induction in Xenopus embryos. ␤-Catenin physically interacts through its Armadillo repeats with the C-terminal transactivation domain of Sox9. We hypothesize that the inhibitory activity of Sox9 is caused by its ability to compete with Tcf/Lef for binding to ␤-catenin, followed by degradation of ␤-catenin. Our results strongly suggest that chondrogenesis is controlled by interactions between Sox9 and the Wnt/␤-catenin signaling pathway. Chondrogenesis, an obligatory process in endochondral bone formation, starts with the recruitment of chondrogenic mesenchymal cells into condensations. This is followed by the differentiation of these cells into chondrocytes, which produce cartilage-specific extracellular matrix (ECM) proteins including type II collagen and the proteoglycan aggrecan. Chondrocytes then undergo a unidirectional proliferation to form orderly parallel columns, exit the cell cycle, become prehypertrophic, and then hypertrophic. Sox9, a high-mobility-group (HMGbox) transcription factor, is required at sequential steps in this pathway (Bi et al. 1999(Bi et al. , 2001Akiyama et al. 2002).Both the human disease campomelic dysplasia, which is caused by heterozygous mutations in the Sox9 gene and is due to Sox9 haploinsufficiency, as well as Sox9 heterozygous mutant mice are characterized by a general hypoplasia of endochondral bones (Foster et al. 1994;Wagner et al. 1994). Inactivation of Sox9 in limb buds using the Cre recombinase/loxP recombination system before chondrogenic mesenchymal condensations results in the complete absence of mesenchymal condensations and of subsequent cartilage and bone formation, indicating that Sox9 is needed for an early step in chondrogenesis, that of mesenchymal condensations (Akiyama et al. 2002). A similar conclusion was also reached by analysis of mouse embryo chimeras derived from homozygous Sox9 mutant embryonic stem (ES) cells (Bi et al. 1999). That Sox9 is needed at sequential steps is shown by the severe generalized chondrodysplasia of mouse embryos in which Sox9 is deleted after chondrogenic mesenchymal conde...

Several studies have implicated Wnt signalling in primary axis formation during vertebrate embryogenesis, yet no Wnt protein has been shown to be essential for this process. In the mouse, primitive streak formation is the first overt morphological sign of the anterior-posterior axis. Here we show that Wnt3 is expressed before gastrulation in the proximal epiblast of the egg cylinder, then is restricted to the posterior proximal epiblast and its associated visceral endoderm and subsequently to the primitive streak and mesoderm. Wnt3-/- mice develop a normal egg cylinder but do not form a primitive streak, mesoderm or node. The epiblast continues to proliferate in an undifferentiated state that lacks anterior-posterior neural patterning, but anterior visceral endoderm markers are expressed and correctly positioned. Our results suggest that regional patterning of the visceral endoderm is independent of primitive streak formation, but the subsequent establishment of anterior-posterior neural pattern in the ectoderm is dependent on derivatives of the primitive streak. These studies provide genetic proof for the requirement of Wnt3 in primary axis formation in the mouse.

Bone morphogenetic proteins (BMPs) are secreted proteins that interact with cell-surface receptors and are believed to play a variety of important roles during vertebrate embryogenesis. Bmpr, also known as ALK-3 and Brk-1, encodes a type I transforming growth factor-~ (TGF-[3) family receptor for BMP-2 and BMP-4. Bmpr is expressed ubiquitously during early mouse embryogenesis and in most adult mouse tissues. To study the function of Bmpr during mammalian development, we generated Bmpr-mutant mice. After embryonic day 9.5 (E9.5), no homozygous mutants were recovered from heterozygote matings. Homozygous mutants with morphological defects were first detected at E7.0 and were smaller than normal. Morphological and molecular examination demonstrated that no mesoderm had formed in the mutant embryos. The growth characteristics of homozygous mutant blastocysts cultured in vitro were indistinguishable from those of controls; however, embryonic ectoderm (epiblast) cell proliferation was reduced in all homozygous mutants at E6.5 before morphological abnormalities had become prominent. Teratomas arising from E7.0 mutant embryos contained derivatives from all three germ layers but were smaller and gave rise to fewer mesodermal cell types, such as muscle and cartilage, than controls. These results suggest that signaling through this type I BMP-2/4 receptor is not necessary for preimplantation or for initial postimplantation development but may be essential for the inductive events that lead to the formation of mesoderm during gastrulation and later for the differentiation of a subset of mesodermal cell types.