The fact that the identity of the cells that initiate metastasis in most human cancers is unknown hampers the development of antimetastatic therapies. Here we describe a subpopulation of CD44 cells in human oral carcinomas that do not overexpress mesenchymal genes, are slow-cycling, express high levels of the fatty acid receptor CD36 and lipid metabolism genes, and are unique in their ability to initiate metastasis. Palmitic acid or a high-fat diet specifically boosts the metastatic potential of CD36 metastasis-initiating cells in a CD36-dependent manner. The use of neutralizing antibodies to block CD36 causes almost complete inhibition of metastasis in immunodeficient or immunocompetent orthotopic mouse models of human oral cancer, with no side effects. Clinically, the presence of CD36 metastasis-initiating cells correlates with a poor prognosis for numerous types of carcinomas, and inhibition of CD36 also impairs metastasis, at least in human melanoma- and breast cancer-derived tumours. Together, our results indicate that metastasis-initiating cells particularly rely on dietary lipids to promote metastasis.

A longstanding unsettled question is whether pancreatic beta cells originate from exocrine duct cells. We have now used genetic labeling to fate map embryonic and adult pancreatic duct cells. We show that Hnf1beta+ cells of the trunk compartment of the early branching pancreas are precursors of acinar, duct, and endocrine lineages. Hnf1beta+ cells subsequent form the embryonic duct epithelium, which gives rise to both ductal and endocrine lineages, but not to acinar cells. By the end of gestation, the fate of Hnf1beta+ duct cells is further restrained. We provide compelling evidence that the ductal epithelium does not make a significant contribution to acinar or endocrine cells during neonatal growth, during a 6 month observation period, or during beta cell growth triggered by ligation of the pancreatic duct or by cell-specific ablation with alloxan followed by EGF/gastrin treatment. Thus, once the ductal epithelium differentiates it has a restricted plasticity, even under regenerative settings.

During embryogenesis, the pancreas arises from dorsal and ventral pancreatic protrusions from the primitive gut endoderm upon induction by different stimuli from neighboring mesodermal tissues. Recent studies have shown that Retinoic Acid (RA) signaling is essential for the development of the pancreas in non-mammalian vertebrates. To investigate whether RA regulates mouse pancreas development, we have studied the phenotype of mice with a targeted deletion in the retinaldehyde dehydrogenase 2 (Raldh2) gene, encoding the enzyme required to synthesize RA in the embryo. We show that Raldh2 is expressed in the dorsal pancreatic mesenchyme at the early stage of pancreas specification. RA-responding cells have been detected in pancreatic endodermal and mesenchymal cells. Raldh2-deficient mice do not develop a dorsal pancreatic bud. Mutant embryos lack Pdx 1 expression, an essential regulator of early pancreas development, in the dorsal but not the ventral endoderm. In contrast to Pdx 1-deficient mice, the early glucagon-expressing cells do not develop in Raldh2 knockout embryos. Shh expression is, as in the wild-type embryo, excluded from the dorsal endodermal region at the site where the dorsal bud is expected to form, indicating that the dorsal bud defect is not related to a mis-expression of Shh. Mesenchymal expression of the LIM homeodomain protein Isl 1, required for the formation of the dorsal mesenchyme, is altered in Raldh2--/-- embryos. The homeobox gene Hlxb9, which is essential for the initiation of the pancreatic program in the dorsal foregut endoderm, is still expressed in Raldh2--/-- dorsal epithelium but the number of HB9-expressing cells is severely reduced. Maternal supplementation of RA rescues early dorsal pancreas development and restores endodermal Pdx 1 and mesenchymal Isl 1 expression as well as endocrine cell differentiation. These findings suggest that RA signaling is important for the proper differentiation of the dorsal mesenchyme and development of the dorsal endoderm. We conclude that RA synthesized in the mesenchyme is specifically required for the normal development of the dorsal pancreatic endoderm at a stage preceding Pdx 1 function.

The basic helix-loop-helix transcription factor Neurogenin 3 (NGN3) controls endocrine cell fate specification in uncommitted pancreatic progenitor cells. Ngn3-deficient mice do not develop any islet cells and are diabetic. All the major islet cell types, including insulin-producing beta-cells, derive from Ngn3-positive endocrine progenitor cells. Therefore, the characterization of this population of immature cells is of particular interest for the development of novel strategies for cell replacement therapies in type 1 diabetes. To explore further the biology of islet progenitor cells we have generated a mouse in which Ngn3-expressing cells are labeled with the enhanced yellow fluorescent protein (EYFP) using a knock-add-on strategy. In this approach, the EYFP cDNA is introduced into the 3'-untranslated region of the proendocrine transcription factor, Neurogenin 3, without deleting any endogenous coding or regulatory sequences. In Ngn3(EYFP/+) and Ngn3(EYFP/EYFP) mice, the EYFP protein is targeted to Ngn3-expressing progenitors in the developing pancreas, and islets develop normally. Islet progenitors can be purified from whole embryonic pancreas by fluorescence-activated cell sorting from Ngn3(EYFP/+) mice and their development can be monitored in real time in pancreas explant cultures. These experiments showed that endocrine progenitors can form de novo and expand, in vitro, in the absence of signals from the surrounding mesenchyme, suggesting that endocrine commitment is a default pathway. The Ngn3(EYFP) mice represent a valuable tool to study islet cell development and neogenesis in normal and diabetic animals as well as for the determination of the conditions to generate beta-cells in vitro.