Search

DNA methylation is an essential epigenetic mark. Three classes of mammalian proteins recognize methylated DNA: MBD proteins, SRA proteins and the zinc-finger proteins Kaiso, ZBTB4 and ZBTB38. The last three proteins can bind either methylated DNA or unmethylated consensus sequences; how this is achieved is largely unclear. Here, we report that the human zinc-finger proteins Kaiso, ZBTB4 and ZBTB38 can bind methylated DNA in a sequence-specific manner, and that they may use a mode of binding common to other zinc-finger proteins. This suggests that many other sequence-specific methyl binding proteins may exist.

HIPK2 is a eukaryotic Serine-Threonine kinase that controls cellular proliferation and survival in response to exogenous signals. Here, we show that the human transcription factor ZBTB4 is a new target of HIPK2. The two proteins interact in vitro, colocalize and associate in vivo, and HIPK2 phosphorylates several conserved residues of ZBTB4. Overexpressing HIPK2 causes the degradation of ZBTB4, whereas overexpressing a kinase-deficient mutant of HIPK2 has no effect. The chemical activation of HIPK2 also decreases the amount of ZBTB4 in cells. Conversely, the inhibition of HIPK2 by drugs or by RNA interference causes a large increase in ZBTB4 levels. This negative regulation of ZBTB4 by HIPK2 occurs under normal conditions of cell growth. In addition, the degradation is increased by DNA damage. These findings have two consequences. First, we have recently shown that ZBTB4 inhibits the transcription of p21. Therefore, the activation of p21 by HIPK2 is two-pronged: stimulation of the activator p53, and simultaneous repression of the inhibitor ZBTB4. Second, ZBTB4 is also known to bind methylated DNA and repress methylated sequences. Consequently, our findings raise the possibility that HIPK2 might influence the epigenetic regulation of gene expression at loci that remain to be identified.

The transcriptional corepressor C-terminal binding protein (CtBP) is thought to be involved in development and oncogenesis, but the regulation of its corepressor activity is largely unknown. We show here that a novel BTB-zinc finger protein, CIBZ (CtBP-interacting BTB zinc finger protein; a mouse ortholog of rat ZENON that was recently identified as an e-box/dyad binding protein), redistributes CtBP to pericentromeric foci from a diffuse nuclear localization in interphase cells. CIBZ physically associates with CtBP via a conserved CtBP binding motif, PLDLR. When heterologously targeted to DNA, CIBZ represses transcription via two independent repression domains, an N-terminal BTB domain and a PLDLR motif-containing RD2 region, in a histone deacetylase-independent and -dependent manner, respectively. Mutation in the PLDLR motif abolishes the CIBZ-CtBP interaction and transcriptional repression activity of RD2, but does not affect the repression activity of the BTB domain. Furthermore, this PLDLR-mutated CIBZ cannot target CtBP to pericentromeric foci, although it is localized to the pericentromeric foci itself. These results suggest that at least one repression mechanism mediated by CIBZ is recruitment of the CtBP/HDAC complex to pericentromeric foci, and that CIBZ may regulate pericentromeric targeting of CtBP.

We report directed differentiaion of retinal precursors in vitro from mouse ES cells. Six3+ rostral brain progenitors are generated by culturing ES cells under serum-free suspension conditions (SFEB culture) in the presence of Wnt and Nodal antagonists (Dkk1 and LeftyA), and subsequently steered to differentiate into Rx+ cells (16%) by treatment with activin and serum. Consistent with the characteristics of early neural retinal precursors, the induced Rx+ cells coexpress Pax6 and the mitotic marker Ki67, but not Nestin. The ES cell-derived precursors efficiently generate cells with the photoreceptor phenotype (rhodopsin+, recoverin+) when cocultured with embryonic retinal cells. Furthermore, organotypic culture studies demonstrate the selective integration and survival of ES cell-derived cells with the photoreceptor phenotype (marker expression and morphology) in the outer nuclear layer of the retina. Taken together, ES cells treated with SFEB/Dkk1/LeftyA/serum/activin generate neural retinal precursors, which have the competence of photoreceptor differentiation.

Fox factors (winged-helix transcription factors) play important roles in early embryonic patterning. We show here that FoxD3 (Forkhead 6) regulates neural crest determination in Xenopus embryos. Expression of FoxD3 in the presumptive neural crest region starts at the late gastrula stage in a manner similar to that of Slug, and overlaps with that of Zic-r1. When overexpressed in the embryo and in ectodermal explants, FoxD3 induces expression of neural crest markers. Attenuation of FoxD3-related signaling by a dominant-negative FoxD3 construct (FoxD3delN) inhibits neural crest differentiation in vivo without suppressing the CNS marker Sox2. Interestingly, these loss-of-function phenotypes are reversed by coinjecting SLUG: In animal cap explants, neural crest differentiation induced by Slug and Wnt3a is also inhibited by FoxD3delN but not by a dominant-negative form of XBF2. Loss-of-function studies using dominant-negative forms of FoxD3 and Slug indicate that Slug induction by Zic factors requires FoxD3-related signaling, and that FoxD3 and Slug have different requirements in inducing downstream neural crest markers. These data demonstrate that FoxD3 (or its closely related factor) is an essential upstream regulator of neural crest determination.

A replication initiator protein (RepE54) complexed with iteron DNA at its binding site was crystallized by the hanging drop vapor diffusion method. The crystals belong to monoclinic space group C2 with unit cell dimensions of a = 108.4 A, b = 81.9 A, c = 73.9 A, and beta = 121.5 degrees, where one molecule of the protein-DNA complex exists per asymmetric unit. They diffract X-rays up to 2.6 A resolution with synchrotron radiation.