zebrafish and allow further comparative functional analysis ofTrap230 and its role in the migration of this vertebrate-specificpopulation of neural crest cells.

doi:10.1016/j.ydbio.2008.05.366

Program/Abstract # 344Optimization of the Gal4-UAS system and generation of a versatilecollection of UAS-reporters for zebrafishCourtney M. Akitake a, Mary G. Goll a, Jerry Rhee b,Michael J. Parsons b, Steven D. Leach b, Marnie E. Halpern a

a Department of Embryology, Carnegie Institution, Baltimore, MD, USAb Department of Surgery, Johns Hopkins Medical Institutions, Baltimore,MD, USA

A Tol2 transposon-mediated gene/enhancer trap was previouslyused to generate transgenic zebrafish lines that express Gal4-VP16 in atissue-specific manner and activate reporter genes under control ofmulticopy upstream activating sequences UAS (Davison et al., 2007).Gal4 mutations that are temperature-sensitive (ts) in yeast and Dro-sophila also function in zebrafish. By assaying for optimal temperatureregulation, we aim to devise methods to modulate Gal4 activity invivo. We generated a variety of UAS regulated reporter/effectors thatin conjunction with the Gal4 driver lines, can label subcellulardomains or ablate cells selectively. We are also attempting to traceneuronal projections in the brains of transgenic zebrafish usingtranssynaptic tracers under UAS control. One persistent drawback tocurrent Gal4-mediated methods is silencing of transgene expressionthrough methylation of the commonly used multicopy UAS. We havedevised a dual reporter system to assess transcriptional activation ofUAS sequence and copy number variants in transgenic zebrafish,independent of chromosomal context. With an optimized UAS,transgene variegation will be mitigated, and high levels of Gal4regulated expression achieved.

doi:10.1016/j.ydbio.2008.05.367

Program/Abstract # 345Paternal allele-specific methylation at Rasgrf1 is present inmonoalleleic and biallelic tissuesTamara L. Davis , Lauren Dockery, Rachel Horton , Christina Harview,Nelly KhaselevBiology Department, Bryn Mawr College, Bryn Mawr, PA 19010, USA

Genomic imprinting is a mammalian-specific phenomenonwhereby one of the two parental alleles is expressed. Monoallelicexpression is frequently associated with parent of origin-specific DNAmethylation and histone modifications. In mouse, the imprinted geneRasgrf1 is expressed solely from the paternal allele in a some tissueswhile in other tissues it is expressed from both parental alleles.Paternal allele-specific DNA methylation of the RasDMR is inheritedfrom sperm. We have investigated the relationship between DNAmethylation patterns and expression patterns at Rasgrf1 duringdevelopment. Our results demonstrate that paternal allele-specificmethylation is retained in the blastocyst, indicating that differentialmethylation of the RasDMR can serve as the primary imprintingmark for Rasgrf1. In addition, we observed paternal allele-specificmethylation in the placenta and head of 8.5 dpc embryos, illustratingthat both embryonic and extraembryonic tissues maintain theimprinting mark. Furthermore, we found paternal allele-specific

methylation in all neonatal tissues examined, including tissues thathave been reported to have monoallelic, biallelic and no expressionof Rasgrf1. We conclude that while DNA methylation may play a rolein marking the transcriptionally active paternal Rasgrf1 allele, it isnot sufficient to induce monoallelic expression since tissues withbiallelic expression also display paternal allele-specific methylationof the RasDMR. We are currently investigating the role histonemodifications play in differentiating between the maternal andpaternal Rasgrf1 alleles.

doi:10.1016/j.ydbio.2008.05.368

Program/Abstract # 346Maternal effects of CTCF, a multifunctional epigenetic regulatorLe-Ben Wan a, Hua Pan b, Yong Cheng c, Jun Ma b, Andrew Fedoriw a,Victor Lobanenkov d, Keith E. Latham c, Richard M. Schultz a,Marisa S. Bartolomei aa Department of Cell and Developmental Biology, University ofPennsylvania School of Medicine, Philadelphia, PA 19104, USAb Department of Biology, University of Pennsylvania, Philadelphia, PA19104, USAc The Fels Institute for Cancer Research and Molecular Biology, andDepartment of Biochemistry, Temple University School of Medicine,Philadelphia, PA 19140, USAd Laboratory of Immunopathology, NIAID, NIH, Rockville, MD 20852,USA

CTCF is a multifunctional nuclear factor involved in epigeneticregulation. Despite recent advances that include the systematicdiscovery of CTCF binding sites throughout the mammalian genome,the in-vivo roles of CTCF in adult tissues and during embryonicdevelopment are largely unknown. Using transgenic RNA interference,we depleted maternal stores of CTCF from growing mouse oocytes,and identify over a thousand misregulated genes. Moreover, CTCFdepletion causes meiotic defects in the egg, and mitotic defects in theembryo. Embryonic defects are accompanied by defects in zygoticgene expression, and culminate in apoptosis. Maternal pronucleartransfer and Ctcf mRNAmicroinjection experiments indicate that CTCFis a mammalian maternal effect gene whose roles in meiotic ma-turation and early embryonic development are independent. This isthe first study detailing a global and essential role for CTCF in mouseoocytes and preimplantation embryos.

doi:10.1016/j.ydbio.2008.05.369

Program/Abstract # 347Regulation of Tcf3 repressor activity in mouse embryonic stemcellsLaura Pereira , Fei Yi , Jackson Hoffman, Bradely MerrillDepartment of Biochemistry and Molecular Genetics, University ofIllinois at Chicago, IL, USA

The dual function of embryonic stem cells (ESC) requires themnot only to form new stem cells through self-renewal but also toform lineage-committed cells through differentiation. In self-renew-ing cells, the transcription factors Oct4, Sox2 and Nanog function asa feedforward circuit to promote a gene expression programsupporting pluripotency. We have found that Tcf3, a member ofthe WNT-regulated TCF/LEF family of transcription factors, isnecessary to counteract the effects of the Oct4/Sox2/Nanog factors

571ABSTRACTS / Developmental Biology 319 (2008) 565–575

on gene expression. Interestingly, chromatin immunoprecipitationrevealed Oct4, Nanog and Tcf3 co-occupy promoters of commonlyregulated genes. The underlying mechanism by which Tcf3 is ableto limit expression of active genes is being examined. Mutationalanalysis of TCF3 showed that the context-dependent regulatorydomain (CRD) is required for the repression, suggesting thatprotein-protein interactions with the CRD are mediating TCF3molecular activity in ESCs. Transfection experiments using Grouchoproteins showed that TCF3 repression is Groucho-independent,since it is not affected by expression of a dominant negative GRG5protein. To identify novel binding partners to TCF3-CRD, we haveused a recombinant GST-TCF3-CRD protein as a bait to probe forinteracting proteins in ESC nuclear extracts. The identity andfunctional significance of interacting proteins will help elucidatethe molecular mechanisms involved in controlling the coreregulatory circuitry controlling pluripotency.

doi:10.1016/j.ydbio.2008.05.370

Program/Abstract # 348Wrestling with melanocyte developmentPei-Chih Lee , Kimberly M. Taylor, Carole LaBonneDepartment of Biochemistry, Molecular Biology, and Cellular Biology,Northwestern University, IL, USA

Melanocytes in skin, hair and the choroid layer of the eye arederived from neural crest cells, a population of stem cell-like cells thatare found only in vertebrate embryos. Microphthalmia-associatedtranscription factor (Mitf) and SoxE factors act as the key regulatorsduring melanocyte development. Mis-regulation of Mitf or Sox10 cancause congenital diseases such asWaardenburg syndrome, as well as adangerous skin cancer, melanoma. However, it is still unclear howthese factors are regulated to control downstream target genes duringmelanocyte development. By co-expression of Mitf and SoxE factor,we demonstrated that these two factors synergistically activate theexpression of melanogenic marker Dct, and the synergistic effect ismodulated by the SUMOylation of either one of these proteins. Thiseffect is achieved by SUMO-dependent alteration of transcriptionalco-regulatory complexes. We provide evidence that SoxE, whichpreviously had been described as a dedicated transcriptional activator,is actually a context-dependent transcriptional regulator that recruitsa co-repressor when SUMOylated. We show that neither SoxE norSUMO alone is sufficient to interact with this co-repressor, but ratherrecruitment requires bivalent interactions with both factors. Theseresults and further studies on the regulation and cooperation of Mitfand SoxE factors can provide us a better understanding toward bothnormal melanocyte development and the melanocyte-relateddiseases.

doi:10.1016/j.ydbio.2008.05.371

Program/Abstract # 349Transcriptional regulation of the FoxO1 gene during mousedevelopmentBarbara Villarejo Balcells , Peter W.J. Rigby, Jaime J. CarvajalThe Institute of Cancer Research, Section of Gene Function andRegulation, Chester Beatty Laboratories, London, SW3 6JB, England, UK

FoxO1, a member of the Forkhead Box family of transcriptionfactors, regulates many processes including cell cycle progression,

differentiation and apoptosis. FoxO1 KO mice die early in gestationdue to vascular defects. Comparative sequence analysis of N250 kbspanning FoxO1 reveals many non-coding sequences highly con-served between species that are likely to contain transcriptionalcontrol elements. We have used three overlapping BACs spanning theFoxO1 locus to define regulatory regions responsible for differentsubsets of expression. Using recombineering we have inserted a LacZreporter gene at the translational start site in each BAC and generatedtransgenic mice.All BACs are able to drive LacZ expression in theumbilical cord, myotome and developing gut at various develop-mental stages suggesting sequences within the common interval(−38 kb to +104 kb) control expression in these regions. In addition,BAC38 and BAC61 show expression in the heart. Between 9.0 dpc and11.0 dpc, BAC38 drives vascular expression that rapidly disappears by11.5 dpc. In contrast, BAC61 only drives adult vascular expressionsuggesting two distinct vascular enhancers, an embryonic elementbetween +104 kb and +148 kb, and an adult element between −61 kband +104 kb. Finally, BAC116 is able to drive expression in the neuraltube and cartilage indicating the presence of specific elementsbetween −116 kb and −61 kb.In conclusion, FoxO1 expression isregulated by elements distributed over N264 kb acting at differenttimes and anatomical locations.

doi:10.1016/j.ydbio.2008.05.372

Program/Abstract # 350Foxn1 is a regulatory target of Hoxc13 in ectodermal developmentand dysplasiaChristopher S. Potter a, Nathanael D. Pruett a, Michael J. Kern b,Alan R. Godwin c, John P. Sundberg d, Alexander Awgulewitsch a

a Department of Medicine, Medical University of South Carolina,Charleston, SC, USAb Department of Cell Biology and Anatomy, Medical University of SouthCarolina, Charleston, SC, USAc Department of Molecular and Integrative Physiology, University ofKansas Medical Center, Kansas City, KS, USAd The Jackson Laboratory, Bar Harbor, ME, USA.

Hoxc13 null (Hoxc13tm1Mrc) mice share multiple phenotypiccharacteristics with the nude (Foxn1nu) mouse, including hair-lessness and a severe nail dystrophy. Previous DNA microarray andin situ hybridization data obtained with Hoxc13 overexpressingmice indicated downregulation of Foxn1 in the abnormally differ-entiated hair follicles of these mice. Considering the overlap ofFoxn1 and Hoxc13 expression domains in the precortical region ofanagen hair follicles and the presence of multiple bona fide Hoxc13binding sites in the Foxn1 promoter region, we hypothesize thatFoxn1 is a direct target of Hoxc13 regulation. Both genes are likelypart of a regulatory network essential for both normal hair and nailplate development. We present a comparative analysis of Hoxc13overexpressing mice, Hoxc13 null mice, and Foxn1 null mice. Dataobtained through immunohistochemistry, in-situ hybridization, real-time PCR, and Hoxc13-chromatin immuno-precipitation (ChIP)assays of the Foxn1 promoter region, suggest that Foxn1 is indeeda direct regulatory target of Hoxc13. Additionally, the evidenceprovided here suggests that Hoxc13 is a key regulator of both nor-mal hair and nail plate development and likely plays a central role inthe complex regulatory networks underlying various forms ofEctodermal Dysplasia.

doi:10.1016/j.ydbio.2008.05.373

572 ABSTRACTS / Developmental Biology 319 (2008) 565–575