SUMMARYWnt/-catenin signaling controls multiple steps of neural crest development, ranging from neural crest induction, lineage decisions, to differentiation. In mice, conditional -catenin inactivation in premigratory neural crest cells abolishes both sensory neuron and melanocyte formation. Intriguingly, the generation of melanocytes is also prevented by activation of -catenin in the premigratory neural crest, which promotes sensory neurogenesis at the expense of other neural crest derivatives. This raises the question of how Wnt/-catenin signaling regulates the formation of distinct lineages from the neural crest. Using various Cre lines to conditionally activate -catenin in neural crest cells at different developmental stages, we show that neural crest cell fate decisions in vivo are subject to temporal control by Wnt/-catenin. Unlike in premigratory neural crest, -catenin activation in migratory neural crest cells promotes the formation of ectopic melanoblasts, while the production of most other lineages is suppressed. Ectopic melanoblasts emerge at sites of neural crest target structures and in many tissues usually devoid of neural crest-derived cells. -catenin activation at later stages in glial progenitors or in melanoblasts does not lead to surplus melanoblasts, indicating a narrow time window of Wnt/-catenin responsiveness during neural crest cell migration. Thus, neural crest cells appear to be multipotent in vivo both before and after emigration from the neural tube but adapt their response to extracellular signals in a temporally controlled manner.

The CD27/CD70-interaction has been shown to provide a costimulatory and survival signal for T cells in vitro and in vivo. Recently, CD70 expression by DC was found to be important for the priming of CD8+ T cells. We show here that blocking CD70 interactions has a significant impact on priming of CD8+ T cell responses by vaccinia virus (VV), Listeria monocytogenes and vesicular stomatitis virus (VSV) in mice. However, the priming of specific CD8+ T cells upon infection with lymphocytic choriomeningitis virus (LCMV) was only marginally reduced by CD70-blockade. Blocking of CD70 prevented CD8+ T cell priming in DIETER mice, a model in which presentation of LCMV-derived epitopes can be induced selectively in dendritic cells (DC). In contrast, CD70-CD27 interactions were not important for the priming of VSV-specific CD4+ T cells or class switch of neutralizing antibodies. As we show that priming of CD8+ T cells by the pathogens used here is dependent on antigen presentation by DC and that infection results in up-regulation of CD70 on DC, we conclude that CD70 expression on DC plays an important role in the priming of CD8+ T cells by pathogens. Moreover, the lack of CD70 cannot be completely compensated for by other costimulatory molecules.

During vertebrate development, neural crest cells are exposed to multiple extracellular cues that drive their differentiation into neural and non-neural cell lineages. Insights into the signals potentially involved in neural crest cell fate decisions in vivo have been gained by cell culture experiments that have allowed the identification of instructive growth factors promoting either proliferation of multipotent neural crest cells or acquisition of specific fates. For instance, members of the TGFbeta factor family induce neurogenesis and smooth muscle cell formation at the expense of other fates in culture. In vivo, conditional ablation of various TGFbeta signaling components resulted in malformations of non-neural derivatives of the neural crest, but it is unclear whether these phenotypes involved aberrant fate decisions. Moreover, it remains to be shown whether neuronal determination indeed requires TGFbeta factor activity in vivo. To address these issues, we conditionally deleted Smad4 in the neural crest, thus inactivating all canonical TGFbeta factor signaling. Surprisingly, neural crest cell fates were not affected in these mutants, with the exception of sensory neurogenesis in trigeminal ganglia. Rather, Smad4 regulates survival of smooth muscle and proliferation of autonomic and ENS neuronal progenitor cells. Thus, Smad signaling plays multiple, lineage-specific roles in vivo, many of which are elicited only after neural crest cell fate decision.

Auto-reactivity of T cells is largely prevented by central and peripheral tolerance. Nevertheless, immunization with certain self-antigens emulsified in CFA induces autoimmunity in rodents, suggesting that tolerance to some self-antigens is not robust. To investigate the fate of nervous system-specific CD8 1 T cells, which only recently came up as being important contributors for MS pathogenesis, we developed a mouse model that allows inducible expression of lymphocytic choriomeningitis virus-derived CD8 1 T-cell epitopes specifically in oligodendrocytes and Schwann cells, the myelinating glia of the nervous system. These transgenic CD8 1 T-cell epitopes induced robust tolerance of endogenous auto-reactive T cells, which proved thymus-independent and was mediated by crosspresenting bone-marrow-derived cells. Immunohistological staining of secondary lymphoid organs demonstrated the presence of glia-derived antigens in DC, suggesting that peripheral tolerance of CD8 1 T cells results from uptake and presentation by steady state DC.