Physiological and pathological functions of E3 ubiquitin ligases Smurfs

Division Of Basic Sciences - Nci

Investigators

Linked publications & trials

Abstract

My group maintains a long-term interest in identifying physiological functions of Smurfs. Toward that end, we have generated Smurf1-, Smurf2-, and Smurf2fl mouse alleles through homologous recombination. We also obtained Smurf1-/-;Smurf2-/- double knockout (DKO) as well as Smurf1-/-;Smurf2fl/fl conditional KO mice through breeding. Characterization of these mice has led to a number of important findings that revealed both common and unique functions of each Smurf during embryogenesis and in maintaining adult physiological homeostasis. Our previous analyses of Smurf1-/- mice revealed a novel mechanism in regulating osteoblast functions and bone homeostasis, suggesting that targeting Smurf1 may prove to be an effective strategy for treating age-related bone losses in osteoporosis 59. Recently, we showed that Smurf1-/- mice constitute an animal model for non-alcoholic fatty liver disease (NAFLD). These mutant mice were over-weight, obese, and exhibited glucose intolerance as well as mild insulin resistance. Our data indicated that Smurf1 catalyses the K63-linked non-proteolytic ubiquitin modification of peroxisome proliferator-activated receptor gamma (PPARg), thereby directly regulating the transcriptional activity of PPARg and the expression of its target genes involved in lipid synthesis and fatty acid uptake. We further showed this causal relationship of Smurf1 deficiency and hepatosteatosis observed in mice was extended to an inverse correlation of low SMURF1 expression to high body mass index values and metabolic dysregulation in human patients, thus revealing the medical relevance of Smurf1 to NAFLD pathogenesis. Our previous work on characterizing Smurf2-/- mice clarified contradictory reports in the literature about Smurf regulation of TGF-beta signaling by showing that Smurf2 indeed has an inhibitory role, but it does so by attenuating Smad3 activity through mono-ubiquitination rather than promoting its degradation as previously reported. We and others have showed that although deleting either Smurf1 or Smurf2 in mice has no impact on gross development and fertility, removing both leads to embryonic lethality, indicating that Smurf1 and Smurf2 share essential overlapping functions. The majority of double knockout (Smurf1-/-;Smurf2-/-) embryos do not develop beyond the Theiler stage 13 and no such embryo can be recovered after E10. The mutant phenotypes fall into one of the two distinct categories: the type I mutants show gastrulation defects whereas the type II mutants survive through gastrulation, but with a pleiotropic range of phenotypes including a completely open neural tube and a broadened neural floor plate. The type II phenotypes have been attributed to the inability of double knockout embryos to establish planar cell polarity (PCP) as well as the inability to sustain Shh signaling that specify neural tube development along the dorsal-ventral axis. However, the underlying cause for the type I phenotype unmains uncharterize. To address the physiological functions of Smurfs in gastrulation, we have isolated embryonic stem cells (ESCs) from E3.5 blastocyts of Smurf2f/f or Smurf1-/-;Smurf2f/f embryos. This affords us a genetic control of shutting down Smurf2 or both Smurf1 and Smurf2 expression in ESCs by way of a cell-permeable CRE recombinase, TAT-Cre. Ablation of all Smurf alleles does not affect the pluripotency of ESCs, as they did not show any significant difference in morphology and expression of ESC specific markers. This indicates that Smurfs do not play a role in maintaining the undifferentiated state of ESCs, which is consistent with the fact that there was no detectable phenotype during early embryonic development until E7.0. We then induced the ESCs to form embryoid bodies (EB), followed by transcriptional profiling and qRT-qPCR quantification of marker genes to distinguish three germ layer lineages. We found that deleting both Smurfs caused a developmental delay or arrest at the mesendoderm stage. The TGF-beta/activin/Nodal signaling pathways are known to be required to maintain the self-renewal of epiblasts and promote mesendoderm differentiation. Attenuation of TGF-beta/Nodal signaling after mesendoderm is required for exit from EpiSCs and mesendoderm and move forward to mesoderm and neuroectoderm lineage. We indeed observed downregulation of Nodal expression at day 6 in WT EBs but not in Smurf-deficient DKO EBs. Consistent with the sustained Nodal expression in DKO EBs, we found the level of p-Smad2/3 stayed high in differentiated DKO EBs, which contrasted with the drastically decreased p-Smad2/3 in WT or Smurf single KO EBs. Giving that Nodal, Gsc and Eomes are direct transcriptional targets of Nodal-activated Smad2, these results suggested that transcriptional activity of Smad2 stayed high in the DKO EBs, which was further confirmed by examine other Smad2/3 transcriptional target gene expression. These results indicate that unsuppressed TGF-beta/Nodal-induced pSmad2 transcriptional activity is the underlying reason for limited differentiation of the mutant DKO cells after formation of mesendoderm.