Histone deacetylases (HDACs) are major epigenetic regulators. We show that HDAC1 and HDAC2 functions are critical for myelination of the peripheral nervous system. Using mouse genetics, we have ablated Hdac1 and Hdac2 specifically in Schwann cells, resulting in massive Schwann cell loss and virtual absence of myelin in mutant sciatic nerves. Expression of Sox10 and Krox20, the main transcriptional regulators of Schwann cell myelination, was greatly reduced. We demonstrate that in Schwann cells, HDAC1 and HDAC2 exert specific primary functions: HDAC2 activates the transcriptional program of myelination in synergy with Sox10, whereas HDAC1 controls Schwann cell survival by regulating the levels of active β-catenin.

The neural crest (NC) is an embryonic stem/progenitor cell population that generates a diverse array of cell lineages, including peripheral neurons, myelinating Schwann cells, and melanocytes, among others. However, there is a long-standing controversy as to whether this broad developmental perspective reflects in vivo multipotency of individual NC cells or whether the NC is comprised of a heterogeneous mixture of lineage-restricted progenitors. Here, we resolve this controversy by performing in vivo fate mapping of single trunk NC cells both at premigratory and migratory stages using the R26R-Confetti mouse model. By combining quantitative clonal analyses with definitive markers of differentiation, we demonstrate that the vast majority of individual NC cells are multipotent, with only few clones contributing to single derivatives. Intriguingly, multipotency is maintained in migratory NC cells. Thus, our findings provide definitive evidence for the in vivo multipotency of both premigratory and migrating NC cells in the mouse.

The emergence of craniofacial skeletal elements, and of the jaw in particular, was a crucial step in the evolution of higher vertebrates. Most facial bones and cartilage are generated during embryonic development by cranial neural crest cells, while an osteochondrogenic fate is suppressed in more posterior neural crest cells. Key players in this process are Hox genes, which suppress osteochondrogenesis in posterior neural crest derivatives. How this specific pattern of osteochondrogenic competence is achieved remains to be elucidated. Here we demonstrate that Hox gene expression and osteochondrogenesis are controlled by epigenetic mechanisms. Ezh2, which is a component of polycomb repressive complex 2 (PRC2), catalyzes trimethylation of lysine 27 in histone 3 (H3K27me3), thereby functioning as transcriptional repressor of target genes. Conditional inactivation of Ezh2 does not interfere with localization of neural crest cells to their target structures, neural development, cell cycle progression or cell survival. However, loss of Ezh2 results in massive derepression of Hox genes in neural crest cells that are usually devoid of Hox gene expression. Accordingly, craniofacial bone and cartilage formation is fully prevented in Ezh2 conditional knockout mice. Our data indicate that craniofacial skeleton formation in higher vertebrates is crucially dependent on epigenetic regulation that keeps in check inhibitors of an osteochondrogenic differentiation program.

No abstract

To identify the cells at the origin of melanoma, we combined single-cell lineage-tracing and transcriptomics approaches with time-lapse imaging. A mouse model that recapitulates key histopathological features of human melanomagenesis was created by inducing a BRafV600E-driven melanomagenic program in tail interfollicular melanocytes. Most targeted mature, melanin-producing melanocytes expanded clonally within the epidermis before losing their differentiated features through transcriptional reprogramming and eventually invading the dermis. Tumors did not form within interscales, which contain both mature and dormant amelanotic melanocytes. The hair follicle bulge, which contains melanocyte stem cells, was also refractory to melanomagenesis. These studies identify varying tumor susceptibilities within the melanocytic lineage, highlighting pigment-producing cells as the melanoma cell of origin, and indicate that regional variation in tumor predisposition is dictated by microenvironmental cues rather than intrinsic differences in cellular origin. Critically, this work provides in vivo evidence that differentiated somatic cells can be reprogrammed into cancer initiating cells.