Joint analysis of differential gene expression in multiple studies using correlation motifs.

The standard methods for detecting differential gene expression are mostly designed for analyzing a single gene expression experiment. When data from multiple related gene expression studies are available, separately analyzing each study is not ideal as it may fail to detect important genes with consistent but relatively weak differential signals in multiple studies. Jointly modeling all data allows one to borrow information across studies to improve the analysis. However, a simple concordance model, in which each gene is assumed to be differential in either all studies or none of the studies, is incapable of handling genes with study-specific differential expression. In contrast, a model that naively enumerates and analyzes all possible differential patterns across studies can deal with study-specificity and allow information pooling, but the complexity of its parameter space grows exponentially as the number of studies increases. Here, we propose a correlation motif approach to address this dilemma. This approach searches for a small number of latent probability vectors called correlation motifs to capture the major correlation patterns among multiple studies. The motifs provide the basis for sharing information among studies and genes. The approach has flexibility to handle all possible study-specific differential patterns. It improves detection of differential expression and overcomes the barrier of exponential model complexity.

The cell surface membrane proteins Cdo and Boc are components and targets of the Hedgehog signaling pathway and feedback network in mice.

Cdo and Boc encode cell surface Ig/fibronectin superfamily members linked to muscle differentiation. Data here indicate they are also targets and signaling components of the Sonic hedgehog (Shh) pathway. Although Cdo and Boc are generally negatively regulated by Hedgehog (HH) signaling, in the neural tube Cdo is expressed within the Shh-dependent floor plate while Boc expression lies within the dorsal limit of Shh signaling. Loss of Cdo results in a Shh dosage-dependent reduction of the floor plate. In contrast, ectopic expression of Boc or Cdo results in a Shh-dependent, cell autonomous promotion of ventral cell fates and a non-cell-autonomous ventral expansion of dorsal cell identities consistent with Shh sequestration. Cdo and Boc bind Shh through a high-affinity interaction with a specific fibronectin repeat that is essential for activity. We propose a model where Cdo and Boc enhance Shh signaling within its target field.

Genome modification in human embryonic stem cells.

Induced pluripotent stem cell (iPSC) technology has emerged as the most promising method for generating patient-specific human embryonic stem (ES) cells and adult stem cells (Takahashi et al., 2007, Cell 131:861-872; Wernig et al., 2007, Nature 448:318-324; Park et al., 2008, Nature 451:141-146). So far, most studies of direct reprogramming have been done by using lentiviruses/retroviruses encoding the reprogramming factors. This represents a major limitation to therapeutic applications since viral integration in the host genome increases the risk of tumorigenicity, and low-level residual expression of reprogramming factors may alter the differentiation potential of the human iPSCs (hiPSCs). As a result, more attention has been paid to developing new techniques to manipulate the human genome, with the goal of making safer hiPSCs that have fewer or no lesions or alterations in the genome. Additionally, the efficiency of reprogramming and of homologous recombination in gene therapy must be improved, if iPSC technology is to be a viable tool in regenerative medicine. Here, we summarize the recent developments in human genome manipulation for generating hiPSCs and advances in homologous recombination for gene targeting.

Using mechanistic Bayesian networks to identify downstream targets of the sonic hedgehog pathway.

BACKGROUND: The topology of a biological pathway provides clues as to how a pathway operates, but rationally using this topology information with observed gene expression data remains a challenge. RESULTS: We introduce a new general-purpose analytic method called Mechanistic Bayesian Networks (MBNs) that allows for the integration of gene expression data and known constraints within a signal or regulatory pathway to predict new downstream pathway targets. The MBN framework is implemented in an open-source Bayesian network learning package, the Python Environment for Bayesian Learning (PEBL). We demonstrate how MBNs can be used by modeling the early steps of the sonic hedgehog pathway using gene expression data from different developmental stages and genetic backgrounds in mouse. Using the MBN approach we are able to automatically identify many of the known downstream targets of the hedgehog pathway such as Gas1 and Gli1, along with a short list of likely targets such as Mig12. CONCLUSIONS: The MBN approach shown here can easily be extended to other pathways and data types to yield a more mechanistic framework for learning genetic regulatory models.

The Hedgehog-binding proteins Gas1 and Cdo cooperate to positively regulate Shh signaling during mouse development.

Hedgehog (Hh) signaling is critical for patterning and growth during mammalian embryogenesis. Transcriptional profiling identified Growth-arrest-specific 1 (Gas1) as a general negative target of Shh signaling. Data presented here define Gas1 as a novel positive component of the Shh signaling cascade. Removal of Gas1 results in a Shh dose-dependent loss of cell identities in the ventral neural tube and facial and skeletal defects, also consistent with reduced Shh signaling. In contrast, ectopic Gas1 expression results in Shh-dependent cell-autonomous promotion of ventral cell identities. These properties mirror those of Cdo, an unrelated, cell surface Shh-binding protein. We show that Gas1 and Cdo cooperate to promote Shh signaling during neural tube patterning, craniofacial, and vertebral development. Overall, these data support a new paradigm in Shh signaling whereby positively acting ligand-binding components, which are initially expressed in responding tissues to promote signaling, are then down-regulated by active Hh signaling, thereby modulating responses to ligand input.

Genomic characterization of Gli-activator targets in sonic hedgehog-mediated neural patterning.

Sonic hedgehog (Shh) acts as a morphogen to mediate the specification of distinct cell identities in the ventral neural tube through a Gli-mediated (Gli1-3) transcriptional network. Identifying Gli targets in a systematic fashion is central to the understanding of the action of Shh. We examined this issue in differentiating neural progenitors in mouse. An epitope-tagged Gli-activator protein was used to directly isolate cis-regulatory sequences by chromatin immunoprecipitation (ChIP). ChIP products were then used to screen custom genomic tiling arrays of putative Hedgehog (Hh) targets predicted from transcriptional profiling studies, surveying 50-150 kb of non-transcribed sequence for each candidate. In addition to identifying expected Gli-target sites, the data predicted a number of unreported direct targets of Shh action. Transgenic analysis of binding regions in Nkx2.2, Nkx2.1 (Titf1) and Rab34 established these as direct Hh targets. These data also facilitated the generation of an algorithm that improved in silico predictions of Hh target genes. Together, these approaches provide significant new insights into both tissue-specific and general transcriptional targets in a crucial Shh-mediated patterning process.

Maternal inheritance of Cre activity in a Sox2Cre deleter strain.

The Sox2 gene is expressed in several undifferentiated cell types. In an earlier study we described a Sox2Cre transgene that mediates epiblast-specific Cre-mediated modification of gene activity in the embryo. Here we report that this transgene is active in the female germline. Consequently, all offspring that arise from female mice heterozygous for the Sox2Cre transgene have demonstrable Cre activity irrespective of whether they inherit the transgene itself. Maternal inheritance of Cre activity allows the efficient modification of gene activity for functional analysis.

Dose dependency of Disp1 and genetic interaction between Disp1 and other hedgehog signaling components in the mouse.

Genetic analyses in Drosophila have demonstrated that a transmembrane protein Dispatched (Disp) is required for the release of lipid-modified Hedgehog (Hh) protein from Hh secreting cells. Analysis of Disp1 null mutant embryos has demonstrated that Disp1 plays a key role in hedgehog signaling in the early mouse embryo. Here we have used a hypomorphic allele in Disp1(Disp1(Delta)(2)), to extend our knowledge of Disp1 function in Hh-mediated patterning of the mammalian embryo. Through genetic combinations with null alleles of patched 1 (Ptch1), sonic hedgehog (Shh) and Indian hedgehog (Ihh), we demonstrate that Disp1 genetically interacts with Hh signaling components. As Disp1 activity is decreased we see a progressive increase in the severity of hedgehog-dependent phenotypes, which is further enhanced by reducing hedgehog ligand levels. Analysis of neural tube patterning demonstrates a progressive loss of ventral cell identities that most likely reflects decreased Shh signaling as Disp1 levels are attenuated. Conversely, increasing available Shh ligand by decreasing Ptch1 dosage leads to the restoration of ventral cell types in Disp1(Delta2/Delta2) mutants. Together, these studies suggest that Disp1 actively regulates the levels of hedgehog ligand that are available to the hedgehog target field. Further, they provide additional support for the dose-dependent action of Shh signaling in patterning the embryo. Finally, in-vitro studies on Disp1 null mutant fibroblasts indicate that Disp1 is not essential for membrane targeting or release of lipid-modified Shh ligand.

Hedgehog signaling in the neural crest cells regulates the patterning and growth of facial primordia.

Facial abnormalities in human SHH mutants have implicated the Hedgehog (Hh) pathway in craniofacial development, but early defects in mouse Shh mutants have precluded the experimental analysis of this phenotype. Here, we removed Hh-responsiveness specifically in neural crest cells (NCCs), the multipotent cell type that gives rise to much of the skeleton and connective tissue of the head. In these mutants, many of the NCC-derived skeletal and nonskeletal components are missing, but the NCC-derived neuronal cell types are unaffected. Although the initial formation of branchial arches (BAs) is normal, expression of several Fox genes, specific targets of Hh signaling in cranial NCCs, is lost in the mutant. The spatially restricted expression of Fox genes suggests that they may play an important role in BA patterning. Removing Hh signaling in NCCs also leads to increased apoptosis and decreased cell proliferation in the BAs, which results in facial truncation that is evident by embryonic day 11.5 (E11.5). Together, our results demonstrate that Hh signaling in NCCs is essential for normal patterning and growth of the face. Further, our analysis of Shh-Fox gene regulatory interactions leads us to propose that Fox genes mediate the action of Shh in facial development.

Mouse brain organization revealed through direct genome-scale TF expression analysis.

In the developing brain, transcription factors (TFs) direct the formation of a diverse array of neurons and glia. We identifed 1445 putative TFs in the mouse genome. We used in situ hybridization to map the expression of over 1000 of these TFs and TF-coregulator genes in the brains of developing mice. We found that 349 of these genes showed restricted expression patterns that were adequate to describe the anatomical organization of the brain. We provide a comprehensive inventory of murine TFs and their expression patterns in a searchable brain atlas database.