Analysis of homeodomain specificities allows the family-wide prediction of preferred recognition sites

Novel sequence-based method for identifying transcription factor binding sites in prokaryotic genomes

MOTIVATION

Computational techniques for microbial genomic sequence analysis are becoming increasingly important. With next-generation sequencing technology and the human microbiome project underway, current sequencing capacity is significantly greater than the speed at which organisms of interest can be studied experimentally. Most related computational work has been focused on sequence assembly, gene annotation and metabolic network reconstruction. We have developed a method that will primarily use available sequence data in order to determine prokaryotic transcription factor (TF) binding specificities.

RESULTS

Specificity determining residues (critical residues) were identified from crystal structures of DNA-protein complexes and TFs with the same critical residues were grouped into specificity classes. The putative binding regions for each class were defined as the set of promoters for each TF itself (autoregulatory) and the immediately upstream and downstream operons. MEME was used to find putative motifs within each separate class. Tests on the LacI and TetR TF families, using RegulonDB annotated sites, showed the sensitivity of prediction 86% and 80%, respectively.

AVAILABILITY

http://ural.wustl.edu/∼gsahota/HTHmotif/

Quantitative analysis of the Drosophila segmentation regulatory network using pattern generating potentials

A new computational method uses gene expression databases and transcription factor binding specificities to describe regulatory elements in the Drosophila A/P patterning network in unprecedented detail.

Cis-regulatory modules that drive precise spatial-temporal patterns of gene expression are central to the process of metazoan development. We describe a new computational strategy to annotate genomic sequences based on their “pattern generating potential” and to produce quantitative descriptions of transcriptional regulatory networks at the level of individual protein-module interactions. We use this approach to convert the qualitative understanding of interactions that regulate Drosophila segmentation into a network model in which a confidence value is associated with each transcription factor-module interaction. Sequence information from multiple Drosophila species is integrated with transcription factor binding specificities to determine conserved binding site frequencies across the genome. These binding site profiles are combined with transcription factor expression information to create a model to predict module activity patterns. This model is used to scan genomic sequences for the potential to generate all or part of the expression pattern of a nearby gene, obtained from available gene expression databases. Interactions between individual transcription factors and modules are inferred by a statistical method to quantify a factor's contribution to the module's pattern generating potential. We use these pattern generating potentials to systematically describe the location and function of known and novel cis-regulatory modules in the segmentation network, identifying many examples of modules predicted to have overlapping expression activities. Surprisingly, conserved transcription factor binding site frequencies were as effective as experimental measurements of occupancy in predicting module expression patterns or factor-module interactions. Thus, unlike previous module prediction methods, this method predicts not only the location of modules but also their spatial activity pattern and the factors that directly determine this pattern. As databases of transcription factor specificities and in vivo gene expression patterns grow, analysis of pattern generating potentials provides a general method to decode transcriptional regulatory sequences and networks.

The developmental program specifying segmentation along the anterior-posterior axis of the Drosophila embryo is one of the best studied examples of transcriptional regulatory networks. Previous work has identified the location and function of dozens of DNA segments called cis-regulatory “modules” that regulate several genes in precise spatial patterns in the early embryo. In many cases, transcription factors that interact with such modules have also been identified. We present a novel computational framework that turns a qualitative and fragmented understanding of modules and factor-module interactions into a quantitative, systems-level view. The formalism utilizes experimentally characterized binding specificities of transcription factors and gene expression patterns to describe how multiple transcription factors (working as activators or repressors) act together in a module to determine its regulatory activity. This formalism can explain the expression patterns of known modules, infer factor-module interactions and quantify the potential of an arbitrary DNA segment to drive a gene's expression. We have also employed databases of gene expression patterns to find novel modules of the regulatory network. As databases of binding motifs and gene expression patterns grow, this new approach provides a general method to decode transcriptional regulatory sequences and networks.

Dissecting the functional specificities of two Hox proteins

Hox proteins frequently select and regulate their specific target genes with the help of cofactors like Extradenticle (Exd) and Homothorax (Hth). For the Drosophila Hox protein Sex combs reduced (Scr), Exd has been shown to position a normally unstructured portion of Scr so that two basic amino acid side chains can insert into the minor groove of an Scr-specific DNA-binding site. Here we provide evidence that another Drosophila Hox protein, Deformed (Dfd), uses a very similar mechanism to achieve specificity in vivo, thus generalizing this mechanism. Furthermore, we show that subtle differences in the way Dfd and Scr recognize their specific binding sites, in conjunction with non-DNA-binding domains, influence whether the target gene is transcriptionally activated or repressed. These results suggest that the interaction between these DNA-binding proteins and the DNA-binding site determines the architecture of the Hox-cofactor-DNA ternary complex, which in turn determines whether the complex recruits coactivators or corepressors.

Using protein design algorithms to understand the molecular basis of disease caused by protein-DNA interactions: the Pax6 example

Quite often a single or a combination of protein mutations is linked to specific diseases. However, distinguishing from sequence information which mutations have real effects in the protein’s function is not trivial. Protein design tools are commonly used to explain mutations that affect protein stability, or protein–protein interaction, but not for mutations that could affect protein–DNA binding. Here, we used the protein design algorithm FoldX to model all known missense mutations in the paired box domain of Pax6, a highly conserved transcription factor involved in eye development and in several diseases such as aniridia. The validity of FoldX to deal with protein–DNA interactions was demonstrated by showing that high levels of accuracy can be achieved for mutations affecting these interactions. Also we showed that protein-design algorithms can accurately reproduce experimental DNA-binding logos. We conclude that 88% of the Pax6 mutations can be linked to changes in intrinsic stability (77%) and/or to its capabilities to bind DNA (30%). Our study emphasizes the importance of structure-based analysis to understand the molecular basis of diseases and shows that protein–DNA interactions can be analyzed to the same level of accuracy as protein stability, or protein–protein interactions.

A novel sequence-based method of predicting protein DNA-binding residues, using a machine learning approach

Protein-DNA interactions play an essential role in transcriptional regulation, DNA repair, and many vital biological processes. The mechanism of protein-DNA binding, however, remains unclear. For the study of many diseases, researchers must improve their understanding of the amino acid motifs that recognize DNA. Because identifying these motifs experimentally is expensive and time-consuming, it is necessary to devise an approach for computational prediction. Some in silico methods have been developed, but there are still considerable limitations. In this study, we used a machine learning approach to develop a new sequence-based method of predicting protein-DNA binding residues. To make these predictions, we used the properties of the micro-environment of each amino acid from the AAIndex as well as conservation scores. Testing by the cross-validation method, we obtained an overall accuracy of 94.89%. Our method shows that the amino acid micro-environment is important for DNA binding, and that it is possible to identify the protein-DNA binding sites with it.

Deciphering the genome's regulatory code: the many languages of DNA

The generation of patterns and the diversity of cell types in a multicellular organism require differential gene regulation. At the heart of this process are enhancers or cis-regulatory modules (CRMs), genomic regions that are bound by transcription factors (TFs) that control spatio-temporal gene expression in developmental networks. To date, only a few CRMs have been studied in detail and the underlying cis-regulatory code is not well understood. Here, we review recent progress on the genome-wide identification of CRMs with chromatin immunoprecipitation of TF-DNA complexes followed by microarrays (ChIP-on-chip). We focus on two computational approaches that have succeeded in predicting the expression pattern driven by a CRM either based on TF binding site preferences and their expression levels, or quantitative analysis of CRM occupancy by key TFs. We also discuss the current limits of these methods and highlight some of the key problems that have to be solved to gain a more complete understanding of the structure and function of CRMs.

Transcriptional control of a plant stem cell niche

Despite the independent evolution of multicellularity in plants and animals, the basic organization of their stem cell niches is remarkably similar. Here, we report the genome-wide regulatory potential of WUSCHEL, the key transcription factor for stem cell maintenance in the shoot apical meristem of the reference plant Arabidopsis thaliana. WUSCHEL acts by directly binding to at least two distinct DNA motifs in more than 100 target promoters and preferentially affects the expression of genes with roles in hormone signaling, metabolism, and development. Striking examples are the direct transcriptional repression of CLAVATA1, which is part of a negative feedback regulation of WUSCHEL, and the immediate regulation of transcriptional repressors of the TOPLESS family, which are involved in auxin signaling. Our results shed light on the complex transcriptional programs required for the maintenance of a dynamic and essential stem cell niche.

Metamotifs--a generative model for building families of nucleotide position weight matrices

Background

Development of high-throughput methods for measuring DNA interactions of transcription factors together with computational advances in short motif inference algorithms is expanding our understanding of transcription factor binding site motifs. The consequential growth of sequence motif data sets makes it important to systematically group and categorise regulatory motifs. It has been shown that there are familial tendencies in DNA sequence motifs that are predictive of the family of factors that binds them. Further development of methods that detect and describe familial motif trends has the potential to help in measuring the similarity of novel computational motif predictions to previously known data and sensitively detecting regulatory motifs similar to previously known ones from novel sequence.

Results

We propose a probabilistic model for position weight matrix (PWM) sequence motif families. The model, which we call the 'metamotif' describes recurring familial patterns in a set of motifs. The metamotif framework models variation within a family of sequence motifs. It allows for simultaneous estimation of a series of independent metamotifs from input position weight matrix (PWM) motif data and does not assume that all input motif columns contribute to a familial pattern. We describe an algorithm for inferring metamotifs from weight matrix data. We then demonstrate the use of the model in two practical tasks: in the Bayesian NestedMICA model inference algorithm as a PWM prior to enhance motif inference sensitivity, and in a motif classification task where motifs are labelled according to their interacting DNA binding domain.

Conclusions

We show that metamotifs can be used as PWM priors in the NestedMICA motif inference algorithm to dramatically increase the sensitivity to infer motifs. Metamotifs were also successfully applied to a motif classification problem where sequence motif features were used to predict the family of protein DNA binding domains that would interact with it. The metamotif based classifier is shown to compare favourably to previous related methods. The metamotif has great potential for further use in machine learning tasks related to especially de novo computational sequence motif inference. The metamotif methods presented have been incorporated into the NestedMICA suite.

Origins of specificity in protein-DNA recognition

Specific interactions between proteins and DNA are fundamental to many biological processes. In this review, we provide a revised view of protein-DNA interactions that emphasizes the importance of the three-dimensional structures of both macromolecules. We divide protein-DNA interactions into two categories: those when the protein recognizes the unique chemical signatures of the DNA bases (base readout) and those when the protein recognizes a sequence-dependent DNA shape (shape readout). We further divide base readout into those interactions that occur in the major groove from those that occur in the minor groove. Analogously, the readout of the DNA shape is subdivided into global shape recognition (for example, when the DNA helix exhibits an overall bend) and local shape recognition (for example, when a base pair step is kinked or a region of the minor groove is narrow). Based on the >1500 structures of protein-DNA complexes now available in the Protein Data Bank, we argue that individual DNA-binding proteins combine multiple readout mechanisms to achieve DNA-binding specificity. Specificity that distinguishes between families frequently involves base readout in the major groove, whereas shape readout is often exploited for higher resolution specificity, to distinguish between members within the same DNA-binding protein family.

Genome-wide analysis of ETS-family DNA-binding in vitro and in vivo

Members of the large ETS family of transcription factors (TFs) have highly similar DNA-binding domains (DBDs)—yet they have diverse functions and activities in physiology and oncogenesis. Some differences in DNA-binding preferences within this family have been described, but they have not been analysed systematically, and their contributions to targeting remain largely uncharacterized. We report here the DNA-binding profiles for all human and mouse ETS factors, which we generated using two different methods: a high-throughput microwell-based TF DNA-binding specificity assay, and protein-binding microarrays (PBMs). Both approaches reveal that the ETS-binding profiles cluster into four distinct classes, and that all ETS factors linked to cancer, ERG, ETV1, ETV4 and FLI1, fall into just one of these classes. We identify amino-acid residues that are critical for the differences in specificity between all the classes, and confirm the specificities in vivo using chromatin immunoprecipitation followed by sequencing (ChIP-seq) for a member of each class. The results indicate that even relatively small differences in in vitro binding specificity of a TF contribute to site selectivity in vivo.

Conserved expression of mouse Six1 in the pre-placodal region (PPR) and identification of an enhancer for the rostral PPR

All cranial sensory organs and sensory neurons of vertebrates develop from cranial placodes. In chick, amphibians and zebrafish, all placodes originate from a common precursor domain, the pre-placodal region (PPR), marked by the expression of Six1/4 and Eya1/2. However, the PPR has never been described in mammals and the mechanism involved in the formation of PPR is poorly defined. Here, we report the expression of Six1 in the horseshoe-shaped mouse ectoderm surrounding the anterior neural plate in a pattern broadly similar to that of non-mammalian vertebrates. To elucidate the identity of Six1-positive mouse ectoderm, we searched for enhancers responsible for Six1 expression by in vivo enhancer assays. One conserved non-coding sequence, Six1-14, showed specific enhancer activity in the rostral PPR of chick and Xenopus and in the mouse ectoderm. These results strongly suggest the presence of PPR in mouse and that it is conserved in vertebrates. Moreover, we show the importance of the homeodomain protein-binding sites of Six1-14, the Six1 rostral PPR enhancer, for enhancer activity, and that Dlx5, Msx1 and Pax7 are candidate binding factors that regulate the level and area of Six1 expression, and thereby the location of the PPR. Our findings provide critical information and tools to elucidate the molecular mechanism of early sensory development and have implications for the development of sensory precursor/stem cells.

Quantitative fine-tuning of photoreceptor cis-regulatory elements through affinity modulation of transcription factor binding sites

Given the remarkable recent progress in gene therapy-based treatments for retinal disease, there is an urgent need for the development of new approaches to quantitative design and analysis of photoreceptor-specific promoters. In this study, we determined the relative binding affinity of all single-nucleotide variants of the consensus binding site of the mammalian photoreceptor transcription factor, Crx. We then showed that it is possible to use these data to accurately predict the relative binding affinity of Crx for all possible 8 bp sequences. By rationally adjusting the binding affinity of three Crx sites, we were able to fine-tune the expression of the rod-specific Rhodopsin promoter over a 225-fold range in living retinas. In addition, we showed that it is possible to fine-tune the activity of the rod-specific Gnat1 promoter over ∼275-fold range by modulating the affinity of a single Crx-binding site. We found that the action of individual binding sites depends on the precise promoter context of the site and that increasing binding affinity does not always equate with increased promoter output. Despite these caveats, this tuning approach permits quantitative engineering of photoreceptor-specific cis-regulatory elements, which can be used as drivers in gene therapy vectors for the treatment of blindness.

Evidence for a myotomal Hox/Myf cascade governing nonautonomous control of rib specification within global vertebral domains

Hox genes are essential for the patterning of the axial skeleton. Hox group 10 has been shown to specify the lumbar domain by setting a rib-inhibiting program in the presomitic mesoderm (PSM). We have now produced mice with ribs in every vertebra by ectopically expressing Hox group 6 in the PSM, indicating that Hox genes are also able to specify the thoracic domain. We show that the information provided by Hox genes to specify rib-containing and rib-less areas is first interpreted in the myotome through the regional-specific control of Myf5 and Myf6 expression. This information is then transmitted to the sclerotome by a system that includes FGF and PDGF signaling to produce vertebrae with or without ribs at different axial levels. Our findings offer a new perspective of how Hox genes produce global patterns in the axial skeleton and support a redundant nonmyogenic role of Myf5 and Myf6 in rib formation.

Pbx1 represses osteoblastogenesis by blocking Hoxa10-mediated recruitment of chromatin remodeling factors

Abdominal-class homeodomain-containing (Hox) factors form multimeric complexes with TALE-class homeodomain proteins (Pbx, Meis) to regulate tissue morphogenesis and skeletal development. Here we have established that Pbx1 negatively regulates Hoxa10-mediated gene transcription in mesenchymal cells and identified components of a Pbx1 complex associated with genes in osteoblasts. Expression of Pbx1 impaired osteogenic commitment of C3H10T1/2 multipotent cells and differentiation of MC3T3-E1 preosteoblasts. Conversely, targeted depletion of Pbx1 by short hairpin RNA (shRNA) increased expression of osteoblast-related genes. Studies using wild-type and mutated osteocalcin and Bsp promoters revealed that Pbx1 acts through a Pbx-binding site that is required to attenuate gene activation by Hoxa10. Chromatin-associated Pbx1 and Hoxa10 were present at osteoblast-related gene promoters preceding gene expression, but only Hoxa10 was associated with these promoters during transcription. Our results show that Pbx1 is associated with histone deacetylases normally linked with chromatin inactivation. Loss of Pbx1 from osteoblast promoters in differentiated osteoblasts was associated with increased histone acetylation and CBP/p300 recruitment, as well as decreased H3K9 methylation. We propose that Pbx1 plays a central role in attenuating the ability of Hoxa10 to activate osteoblast-related genes in order to establish temporal regulation of gene expression during osteogenesis.

Mapping the complexity of transcription control in higher eukaryotes

Recent large analyses suggest the importance of combinatorial regulation by broadly expressed transcription factors rather than expression domains characterized by highly specific factors.

Recent genomic analyses suggest the importance of combinatorial regulation by broadly expressed transcription factors rather than expression domains characterized by highly specific factors.

A cis-regulatory signature for chordate anterior neuroectodermal genes

One of the striking findings of comparative developmental genetics was that expression patterns of core transcription factors are extraordinarily conserved in bilaterians. However, it remains unclear whether cis-regulatory elements of their target genes also exhibit common signatures associated with conserved embryonic fields. To address this question, we focused on genes that are active in the anterior neuroectoderm and non-neural ectoderm of the ascidian Ciona intestinalis. Following the dissection of a prototypic anterior placodal enhancer, we searched all genomic conserved non-coding elements for duplicated motifs around genes showing anterior neuroectodermal expression. Strikingly, we identified an over-represented pentamer motif corresponding to the binding site of the homeodomain protein OTX, which plays a pivotal role in the anterior development of all bilaterian species. Using an in vivo reporter gene assay, we observed that 10 of 23 candidate cis-regulatory elements containing duplicated OTX motifs are active in the anterior neuroectoderm, thus showing that this cis-regulatory signature is predictive of neuroectodermal enhancers. These results show that a common cis-regulatory signature corresponding to K50-Paired homeodomain transcription factors is found in non-coding sequences flanking anterior neuroectodermal genes in chordate embryos. Thus, field-specific selector genes impose architectural constraints in the form of combinations of short tags on their target enhancers. This could account for the strong evolutionary conservation of the regulatory elements controlling field-specific selector genes responsible for body plan formation.

Regional identity in embryos is defined by a few specific transcription factors that activate a large number of target genes through binding to common tags in regulatory sequences. In chordates it is unclear if such tags can be identified in the cis-regulatory regions of regionally expressed genes. To address this question we focused on the anterior nervous system where Otx codes for a transcription factor that triggers expression of many other head-specific genes. We analyzed an element that is active in the region bordering the anterior nervous system in the marine invertebrate Ciona intestinalis. We found that the crucial binding sites have to be duplicated and close enough. One of the pairs is bound by OTX. We showed that anterior nervous system genes are often flanked by duplicated OTX binding sites. We confirmed by transgenic assays that about half of these genomic sequences are active and drive expression anteriorly. This study unravels a simple regulatory logic in the anterior enhancers. It indicates that although there are major changes in the organization of the binding sites at short evolutionary range, conserved expression patterns are partly generated by a duplicated organization of conserved binding sites for region-specific transcription factors.

Comparing anterior and posterior Hox complex formation reveals guidelines for predicting cis-regulatory elements

Hox transcription factors specify numerous cell fates along the anterior-posterior axis by regulating the expression of downstream target genes. While expression analysis has uncovered large numbers of de-regulated genes in cells with altered Hox activity, determining which are direct versus indirect targets has remained a significant challenge. Here, we characterize the DNA binding activity of Hox transcription factor complexes on eight experimentally verified cis-regulatory elements. Hox factors regulate the activity of each element by forming protein complexes with two cofactor proteins, Extradenticle (Exd) and Homothorax (Hth). Using comparative DNA binding assays, we found that a number of flexible arrangements of Hox, Exd, and Hth binding sites mediate cooperative transcription factor complexes. Moreover, analysis of a Distal-less regulatory element (DMXR) that is repressed by abdominal Hox factors revealed that suboptimal binding sites can be combined to form high affinity transcription complexes. Lastly, we determined that the anterior Hox factors are more dependent upon Exd and Hth for complex formation than posterior Hox factors. Based upon these findings, we suggest a general set of guidelines to serve as a basis for designing bioinformatics algorithms aimed at identifying Hox regulatory elements using the wealth of recently sequenced genomes.

Cooperative DNA-binding and sequence-recognition mechanism of aristaless and clawless

To achieve accurate gene regulation, some homeodomain proteins bind cooperatively to DNA to increase those site specificities. We report a ternary complex structure containing two homeodomain proteins, aristaless (Al) and clawless (Cll), bound to DNA. Our results show that the extended conserved sequences of the Cll homeodomain are indispensable to cooperative DNA binding. In the Al-Cll-DNA complex structure, the residues in the extended regions are used not only for the intermolecular contacts between the two homeodomain proteins but also for the sequence-recognition mechanism of DNA by direct interactions. The residues in the extended N-terminal arm lie within the minor groove of DNA to form direct interactions with bases, whereas the extended conserved region of the C-terminus of the homeodomain interacts with Al to stabilize and localize the third alpha helix of the Cll homeodomain. This structure suggests a novel mode for the cooperativity of homeodomain proteins.

Autoregulatory binding sites in the zebrafish six3a promoter region define a new recognition sequence for Six3 proteins

The homeodomain (HD) transcription factor Six3, which is a member of the Six/Sine oculis family, is essential for development of the eyes and forebrain in vertebrates. It has recently been claimed that the HDs of Six3 and other members of the Six family have a common recognition sequence, TGATAC. However, a different recognition sequence including the typical TAAT core motif, which has not yet been fully defined, has also been proposed for the Six3 HD in mice. Our study of the zebrafish orthologue six3a, which has an identical HD, shows that it binds in vitro to multiple TAAT-containing sites within its promoter region. Comparison of the different binding affinities for these sequences identifies three high-affinity sites with a common TAATGTC motif. Notably, this new recognition sequence, which is supported by our analysis of the influence of single-nucleotide substitutions on the DNA-binding affinity, is distinct from all of the DNA-binding specificities previously described in surveys of HDs. In addition, our comparison of Six3a HD binding to the novel TAATGTC motif and the common recognition sequence of Six family HDs (TGATAC) shows very similar affinities, suggesting two distinct DNA-binding modes. Transient reporter assays of the six3a promoter in zebrafish embryos also indicate that the three high-affinity sites are involved in autoregulation. In support of this, chromatin immunoprecipitation experiments show enrichment of Six3a binding to a six3a promoter fragment containing two clustered high-affinity sites. These findings provide strong evidence that the TAATGTC motif is an important target sequence for vertebrate Six3 proteins in vivo.

The FoxF/FoxC factor LET-381 directly regulates both cell fate specification and cell differentiation in C. elegans mesoderm development

Forkhead transcription factors play crucial and diverse roles in mesoderm development. In particular, FoxF and FoxC genes are, respectively, involved in the development of visceral/splanchnic mesoderm and non-visceral mesoderm in coelomate animals. Here, we show at single-cell resolution that, in the pseudocoelomate nematode C. elegans, the single FoxF/FoxC transcription factor LET-381 functions in a feed-forward mechanism in the specification and differentiation of the non-muscle mesodermal cells, the coelomocytes (CCs). LET-381/FoxF directly activates the CC specification factor, the Six2 homeodomain protein CEH-34, and functions cooperatively with CEH-34/Six2 to directly activate genes required for CC differentiation. Our results unify a diverse set of studies on the functions of FoxF/FoxC factors and provide a model for how FoxF/FoxC factors function during mesoderm development.

Structural rules and complex regulatory circuitry constrain expression of a Notch- and EGFR-regulated eye enhancer

Enhancers integrate spatiotemporal information to generate precise patterns of gene expression. How complex is the regulatory logic of a typical developmental enhancer, and how important is its internal organization? Here, we examine in detail the structure and function of sparkling, a Notch- and EGFR/MAPK-regulated, cone cell-specific enhancer of the Drosophila Pax2 gene, in vivo. In addition to its 12 previously identified protein-binding sites, sparkling is densely populated with previously unmapped regulatory sequences, which interact in complex ways to control gene expression. One segment is essential for activation at a distance, yet dispensable for other activation functions and for cell type patterning. Unexpectedly, rearranging sparkling's regulatory sites converts it into a robust photoreceptor-specific enhancer. Our results show that a single combination of regulatory inputs can encode multiple outputs, and suggest that the enhancer's organization determines the correct expression pattern by facilitating certain short-range regulatory interactions at the expense of others.

Integrating multiple evidence sources to predict transcription factor binding in the human genome

Information about the binding preferences of many transcription factors is known and characterized by a sequence binding motif. However, determining regions of the genome in which a transcription factor binds based on its motif is a challenging problem, particularly in species with large genomes, since there are often many sequences containing matches to the motif but are not bound. Several rules based on sequence conservation or location, relative to a transcription start site, have been proposed to help differentiate true binding sites from random ones. Other evidence sources may also be informative for this task. We developed a method for integrating multiple evidence sources using logistic regression classifiers. Our method works in two steps. First, we infer a score quantifying the general binding preferences of transcription factor binding at all locations based on a large set of evidence features, without using any motif specific information. Then, we combined this general binding preference score with motif information for specific transcription factors to improve prediction of regions bound by the factor. Using cross-validation and new experimental data we show that, surprisingly, the general binding preference can be highly predictive of true locations of transcription factor binding even when no binding motif is used. When combined with motif information our method outperforms previous methods for predicting locations of true binding.

Functional characterization of transcription factor motifs using cross-species comparison across large evolutionary distances

We address the problem of finding statistically significant associations between cis-regulatory motifs and functional gene sets, in order to understand the biological roles of transcription factors. We develop a computational framework for this task, whose features include a new statistical score for motif scanning, the use of different scores for predicting targets of different motifs, and new ways to deal with redundancies among significant motif–function associations. This framework is applied to the recently sequenced genome of the jewel wasp, Nasonia vitripennis, making use of the existing knowledge of motifs and gene annotations in another insect genome, that of the fruitfly. The framework uses cross-species comparison to improve the specificity of its predictions, and does so without relying upon non-coding sequence alignment. It is therefore well suited for comparative genomics across large evolutionary divergences, where existing alignment-based methods are not applicable. We also apply the framework to find motifs associated with socially regulated gene sets in the honeybee, Apis mellifera, using comparisons with Nasonia, a solitary species, to identify honeybee-specific associations.

We develop a computational pipeline for predicting the functions of transcription factor motifs, through DNA sequence analysis. The pipeline is applied to the newly sequenced genome of the jewel wasp, Nasonia vitripennis. It exploits the wealth of molecular data available in another insect species, the fruitfly Drosophila melanogaster, and uses cross-species comparison to its advantage. Our main contribution is to show how this can be done despite the large evolutionary divergence between the two species. The methodology presented here may be applied more generally to other scenarios (genomes) where comparative regulatory genomics must deal with large evolutionary divergences.

Conserved expression without conserved regulatory sequence: the more things change, the more they stay the same

Regulatory regions with similar transcriptional output often have little overt sequence similarity, both within and between genomes. Although cis- and trans-regulatory changes can contribute to sequence divergence without dramatically altering gene expression outputs, heterologous DNA often functions similarly in organisms that share little regulatory sequence similarities (e.g. human DNA in fish), indicating that trans-regulatory mechanisms tend to diverge more slowly and can accommodate a variety of cis-regulatory configurations. This capacity to 'tinker' with regulatory DNA probably relates to the complexity, robustness and evolvability of regulatory systems, but cause-and-effect relationships among evolutionary processes and properties of regulatory systems remain a topic of debate. The challenge of understanding the concrete mechanisms underlying cis-regulatory evolution - including the conservation of function without the conservation of sequence - relates to the challenge of understanding the function of regulatory systems in general. Currently, we are largely unable to recognize functionally similar regulatory DNA.

The homeodomain protein HOXB13 regulates the cellular response to androgens

HOXB13 is a member of the homeodomain family of sequence-specific transcription factors and, together with the androgen receptor (AR), plays a critical role in the normal development of the prostate gland. We demonstrate here that, in prostate cancer cells, HOXB13 is a key determinant of the response to androgens. Specifically, it was determined that HOXB13 interacts with the DNA-binding domain of AR and inhibits the transcription of genes that contain an androgen-response element (ARE). In contrast, the AR:HOXB13 complex confers androgen responsiveness to promoters that contain a specific HOXB13-response element. Further, HOXB13 and AR synergize to enhance the transcription of genes that contain a HOX element juxtaposed to an ARE. The profound effects of HOXB13 knockdown on androgen-regulated proliferation, migration, and lipogenesis in prostate cancer cells highlight the importance of the observed changes in gene expression.

Practical computational methods for regulatory genomics: a cisGRN-Lexicon and cisGRN-browser for gene regulatory networks

The CYRENE Project focuses on the study of cis-regulatory genomics and gene regulatory networks (GRN) and has three components: a cisGRN-Lexicon, a cisGRN-Browser, and the Virtual Sea Urchin software system. The project has been done in collaboration with Eric Davidson and is deeply inspired by his experimental work in genomic regulatory systems and gene regulatory networks. The current CYRENE cisGRN-Lexicon contains the regulatory architecture of 200 transcription factors encoding genes and 100 other regulatory genes in eight species: human, mouse, fruit fly, sea urchin, nematode, rat, chicken, and zebrafish, with higher priority on the first five species. The only regulatory genes included in the cisGRN-Lexicon (CYRENE genes) are those whose regulatory architecture is validated by what we call the Davidson Criterion: they contain functionally authenticated sites by site-specific mutagenesis, conducted in vivo, and followed by gene transfer and functional test. This is recognized as the most stringent experimental validation criterion to date for such a genomic regulatory architecture. The CYRENE cisGRN-Browser is a full genome browser tailored for cis-regulatory annotation and investigation. It began as a branch of the Celera Genome Browser (available as open source at http://sourceforge.net/projects/celeragb /) and has been transformed to a genome browser fully devoted to regulatory genomics. Its access paradigm for genomic data is zoom-to-the-DNA-base in real time. A more recent component of the CYRENE project is the Virtual Sea Urchin system (VSU), an interactive visualization tool that provides a four-dimensional (spatial and temporal) map of the gene regulatory networks of the sea urchin embryo.

Regulation of Hox activity: insights from protein motifs

Deciphering the molecular bases of animal body plan construction is a central question in developmental and evolutionary biology. Genome analyses of a number of metazoans indicate that widely conserved regulatory molecules underlie the amazing diversity of animal body plans, suggesting that these molecules are reiteratively used for multiple purposes. Hox proteins constitute a good example of such molecules and provide the framework to address the mechanisms underlying transcriptional specificity and diversity in development and evolution. Here we examine the current knowledge of the molecular bases of Hox-mediated transcriptional control, focusing on how this control is encoded within protein sequences and structures. The survey suggests that the homeodomain is part of an extended multifunctional unit coordinating DNA binding and activity regulation and highlights the need for further advances in our understanding of Hox protein activity.

Profiling the human protein-DNA interactome reveals ERK2 as a transcriptional repressor of interferon signaling

Protein-DNA interactions (PDIs) mediate a broad range of functions essential for cellular differentiation, function, and survival. However, it is still a daunting task to comprehensively identify and profile sequence-specific PDIs in complex genomes. Here, we have used a combined bioinformatics and protein microarray-based strategy to systematically characterize the human protein-DNA interactome. We identified 17,718 PDIs between 460 DNA motifs predicted to regulate transcription and 4,191 human proteins of various functional classes. Among them, we recovered many known PDIs for transcription factors (TFs). We identified a large number of unanticipated PDIs for known TFs, as well as for previously uncharacterized TFs. We also found that over three hundred unconventional DNA-binding proteins (uDBPs)--which include RNA-binding proteins, mitochondrial proteins, and protein kinases--showed sequence-specific PDIs. One such uDBP, ERK2, acts as a transcriptional repressor for interferon gamma-induced genes, suggesting important biological roles for such proteins.

JASPAR 2010: the greatly expanded open-access database of transcription factor binding profiles

JASPAR (http://jaspar.genereg.net) is the leading open-access database of matrix profiles describing the DNA-binding patterns of transcription factors (TFs) and other proteins interacting with DNA in a sequence-specific manner. Its fourth major release is the largest expansion of the core database to date: the database now holds 457 non-redundant, curated profiles. The new entries include the first batch of profiles derived from ChIP-seq and ChIP-chip whole-genome binding experiments, and 177 yeast TF binding profiles. The introduction of a yeast division brings the convenience of JASPAR to an active research community. As binding models are refined by newer data, the JASPAR database now uses versioning of matrices: in this release, 12% of the older models were updated to improved versions. Classification of TF families has been improved by adopting a new DNA-binding domain nomenclature. A curated catalog of mammalian TFs is provided, extending the use of the JASPAR profiles to additional TFs belonging to the same structural family. The changes in the database set the system ready for more rapid acquisition of new high-throughput data sources. Additionally, three new special collections provide matrix profile data produced by recent alternative high-throughput approaches.

hPDI: a database of experimental human protein-DNA interactions

The human protein DNA Interactome (hPDI) database holds experimental protein-DNA interaction data for humans identified by protein microarray assays. The unique characteristics of hPDI are that it contains consensus DNA-binding sequences not only for nearly 500 human transcription factors but also for >500 unconventional DNA-binding proteins, which are completely uncharacterized previously. Users can browse, search and download a subset or the entire data via a web interface. This database is freely accessible for any academic purposes.

AVAILABILITY

http://bioinfo.wilmer.jhu.edu/PDI/.

The non-conserved C-terminal segments of Sine Oculis Homeobox (SIX) proteins confer functional specificity

The Sine Oculis Homeobox (SIX) proteins play critical roles in organogenesis and are defined by the presence of two evolutionarily conserved functional motifs: a homeobox DNA binding domain and the SIX protein-protein interaction domain. Members of this transcription factor family can be divided into three subgroups: Six1/2, Six4/5, and Six3/6. This partitioning is based mainly on protein sequence similarity and genomic architecture, and not on specificities of DNA binding or binding partners. In fact, it is well demonstrated that members of the different subgroups can bind to and activate common transcriptional targets as well as form biochemical complexes with communal binding partners. Here we report that the C-terminal segment, which is not conserved across different SIX subfamilies, may serve to functionally distinguish individual SIX proteins. In particular, we have dissected the C-terminal region of Optix, the Drosophila ortholog of mammalian Six3/6, and identified three regions that distinguish Optix from Sine Oculis, the fly homolog of Six1/2. Two of these regions have been preserved in all Six3/6 family members while the third section is present only within Optix proteins in the Drosophilids. The activities of these regions are required, in unison, for Optix function. We suggest that biochemical/functional differences between members of large protein families as well as proteins encoded by duplicate genes can, in part, be attributed to the activities of nonconserved segments. Finally, we demonstrate that a subset of vertebrate SIX proteins has retained the ability to function during normal fly eye development but have lost the ability to induce the formation of ectopic eyes.

Finding distal regulatory elements in the human genome

Transcriptional regulation of human genes depends not only on promoters and nearby cis-regulatory elements, but also on distal regulatory elements such as enhancers, insulators, locus control regions, and silencing elements, which are often located far away from the genes they control. Our knowledge of human distal regulatory elements is very limited, but the last several years have seen rapid progress in the development of strategies to identify these long-range regulatory sequences throughout the human genome. Here, we review these advances, focusing on two important classes of distal regulatory sequences-enhancers and insulators.

Different gene regulation strategies revealed by analysis of binding motifs

Coordinated regulation of gene expression relies on transcription factors (TFs) binding to specific DNA sites. Our large-scale information-theoretical analysis of > 950 TF-binding motifs demonstrates that prokaryotes and eukaryotes use strikingly different strategies to target TFs to specific genome locations. Although bacterial TFs can recognize a specific DNA site in the genomic background, eukaryotic TFs exhibit widespread, nonfunctional binding and require clustering of sites to achieve specificity. We find support for this mechanism in a range of experimental studies and in our evolutionary analysis of DNA-binding domains. Our systematic characterization of binding motifs provides a quantitative assessment of the differences in transcription regulation in prokaryotes and eukaryotes.

The complex spatio-temporal regulation of the Drosophila myoblast attractant gene duf/kirre

A key early player in the regulation of myoblast fusion is the gene dumbfounded (duf, also known as kirre). Duf must be expressed, and function, in founder cells (FCs). A fixed number of FCs are chosen from a pool of equivalent myoblasts and serve to attract fusion-competent myoblasts (FCMs) to fuse with them to form a multinucleate muscle-fibre. The spatial and temporal regulation of duf expression and function are important and play a deciding role in choice of fibre number, location and perhaps size. We have used a combination of bioinformatics and functional enhancer deletion approaches to understand the regulation of duf. By transgenic enhancer-reporter deletion analysis of the duf regulatory region, we found that several distinct enhancer modules regulate duf expression in specific muscle founders of the embryo and the adult. In addition to existing bioinformatics tools, we used a new program for analysis of regulatory sequence, PhyloGibbs-MP, whose development was largely motivated by the requirements of this work. The results complement our deletion analysis by identifying transcription factors whose predicted binding regions match with our deletion constructs. Experimental evidence for the relevance of some of these TF binding sites comes from available ChIP-on-chip from the literature, and from our analysis of localization of myogenic transcription factors with duf enhancer reporter gene expression. Our results demonstrate the complex regulation in each founder cell of a gene that is expressed in all founder cells. They provide evidence for transcriptional control—both activation and repression—as an important player in the regulation of myoblast fusion. The set of enhancer constructs generated will be valuable in identifying novel trans-acting factor-binding sites and chromatin regulation during myoblast fusion in Drosophila. Our results and the bioinformatics tools developed provide a basis for the study of the transcriptional regulation of other complex genes.

A global survey identifies novel upstream components of the Ath5 neurogenic network

Regulators of vertebrate Ath5 expression were identified by high-throughput screening; extending the current gene regulatory model network controlling retinal neurogenesis.

Background

Investigating the architecture of gene regulatory networks (GRNs) is essential to decipher the logic of developmental programs during embryogenesis. In this study we present an upstream survey approach, termed trans-regulation screen, to comprehensively identify the regulatory input converging on endogenous regulatory sequences.

Results

Our dual luciferase-based screen queries transcriptome-scale collections of cDNAs. Using this approach we study the regulation of Ath5, the central node in the GRN controlling retinal ganglion cell (RGC) specification in vertebrates. The Ath5 promoter integrates the input of upstream regulators to enable the transient activation of the gene, which is an essential step for RGC differentiation. We efficiently identified potential Ath5 regulators that were further filtered for true positives by an in situ hybridization screen. Their regulatory activity was validated in vivo by functional assays in medakafish embryos.

Conclusions

Our analysis establishes functional groups of genes controlling different regulatory phases, including the onset of Ath5 expression at cell-cycle exit and its down-regulation prior to terminal RGC differentiation. These results extent the current model of the GRN controlling retinal neurogenesis in vertebrates.

Non-homeodomain regions of Hox proteins mediate activation versus repression of Six2 via a single enhancer site in vivo

Hox genes control many developmental events along the AP axis, but few target genes have been identified. Whether target genes are activated or repressed, what enhancer elements are required for regulation, and how different domains of the Hox proteins contribute to regulatory specificity are poorly understood. Six2 is genetically downstream of both the Hox11 paralogous genes in the developing mammalian kidney and Hoxa2 in branchial arch and facial mesenchyme. Loss-of-function of Hox11 leads to loss of Six2 expression and loss-of-function of Hoxa2 leads to expanded Six2 expression. Herein we demonstrate that a single enhancer site upstream of the Six2 coding sequence is responsible for both activation by Hox11 proteins in the kidney and repression by Hoxa2 in the branchial arch and facial mesenchyme in vivo. DNA-binding activity is required for both activation and repression, but differential activity is not controlled by differences in the homeodomains. Rather, protein domains N- and C-terminal to the homeodomain confer activation versus repression activity. These data support a model in which the DNA-binding specificity of Hox proteins in vivo may be similar, consistent with accumulated in vitro data, and that unique functions result mainly from differential interactions mediated by non-homeodomain regions of Hox proteins.

dbx mediates neuronal specification and differentiation through cross-repressive, lineage-specific interactions with eve and hb9

Individual neurons adopt and maintain defined morphological and physiological phenotypes as a result of the expression of specific combinations of transcription factors. In particular, homeodomain-containing transcription factors play key roles in determining neuronal subtype identity in flies and vertebrates. dbx belongs to the highly divergent H2.0 family of homeobox genes. In vertebrates, Dbx1 and Dbx2 promote the development of a subset of interneurons, some of which help mediate left-right coordination of locomotor activity. Here, we identify and show that the single Drosophila ortholog of Dbx1/2 contributes to the development of specific subsets of interneurons via cross-repressive, lineage-specific interactions with the motoneuron-promoting factors eve and hb9 (exex). dbx is expressed primarily in interneurons of the embryonic, larval and adult central nervous system, and these interneurons tend to extend short axons and be GABAergic. Interestingly, many Dbx(+) interneurons share a sibling relationship with Eve(+) or Hb9(+) motoneurons. The non-overlapping expression of dbx and eve, or dbx and hb9, within pairs of sibling neurons is initially established as a result of Notch/Numb-mediated asymmetric divisions. Cross-repressive interactions between dbx and eve, and dbx and hb9, then help maintain the distinct expression profiles of these genes in their respective pairs of sibling neurons. Strict maintenance of the mutually exclusive expression of dbx relative to that of eve and hb9 in sibling neurons is crucial for proper neuronal specification, as misexpression of dbx in motoneurons dramatically hinders motor axon outgrowth.

Experimental advances in the characterization of metazoan gene regulatory networks

Gene regulatory networks (GRNs) play a vital role in metazoan development and function, and deregulation of these networks is often implicated in disease. GRNs depict the dynamic interactions between genomic and regulatory state components. The genomic components comprise genes and their associated cis-regulatory elements. The regulatory state components consist primarily of transcriptional complexes that bind the latter elements. With the availability of complete genome sequences, several approaches have recently been developed which promise to significantly enhance our ability to identify either the genomic or regulatory state components, or the interactions between these two. In this review, I provide an in-depth overview of these approaches and detail how each contributes to a more comprehensive understanding of GRN composition and function.

Motif discovery and motif finding from genome-mapped DNase footprint data

MOTIVATION

Footprint data is an important source of information on transcription factor recognition motifs. However, a footprinting fragment can contain no sequences similar to known protein recognition sites. Inspection of genome fragments nearby can help to identify missing site positions.

RESULTS

Genome fragments containing footprints were supplied to a pipeline that constructed a position weight matrix (PWM) for different motif lengths and selected the optimal PWM. Fragments were aligned with the SeSiMCMC sampler and a new heuristic algorithm, Bigfoot. Footprints with missing hits were found for approximately 50% of factors. Adding only 2 bp on both sides of a footprinting fragment recovered most hits. We automatically constructed motifs for 41 Drosophila factors. New motifs can recognize footprints with a greater sensitivity at the same false positive rate than existing models. Also we discuss possible overfitting of constructed motifs.

AVAILABILITY

Software and the collection of regulatory motifs are freely available at http://line.imb.ac.ru/DMMPMM.

Classification of sequence signatures: a guide to Hox protein function

Hox proteins are part of the conserved superfamily of homeodomain-containing transcription factors and play fundamental roles in shaping animal body plans in development and evolution. However, molecular mechanisms underlying their diverse and specific biological functions remain largely enigmatic. Here, we have analyzed Hox sequences from the main evolutionary branches of the Bilateria group. We have found that four classes of Hox protein signatures exist, which together provide sufficient support to explain how different Hox proteins differ in their control and function. The homeodomain and its surrounding sequences accumulate nearly all signatures, constituting an extended module where most of the information distinguishing Hox proteins is concentrated. Only a small fraction of these signatures has been investigated at the functional level, but these show that approaches relying on Hox protein alterations still have a large potential for deciphering molecular mechanisms of Hox differential control.

Internal regulatory interactions determine DNA binding specificity by a Hox transcription factor

In developing bilaterans, the Hox transcription factor family regulates batteries of downstream genes to diversify serially repeated units. Given Hox homeodomains bind a wider array of DNA binding sites in vitro than are regulated by the full-length protein in vivo, regions outside the homeodomain must aid DNA site selection. Indeed, we find affinity for disparate DNA sequences varies less than 3-fold for the homeodomain isolated from the Drosophila Hox protein Ultrabithorax Ia (UbxHD), whereas for the full-length protein (UbxIa) affinity differs by more than 10-fold. The rank order of preferred DNA sequences also differs, further demonstrating distinct DNA binding preferences. The increased specificity of UbxIa can be partially attributed to the I1 region, which lies adjacent to the homeodomain and directly impacts binding energetics. Each of three segments within I1-the Extradenticle-binding YPWM motif, the six amino acids immediately N-terminal to this motif, and the eight amino acids abutting the YPWM C-terminus-uniquely contribute to DNA specificity. Combination of these regions synergistically modifies DNA binding to further enhance specificity. Intriguingly, the presence of the YPWM motif in UbxIa inhibits DNA binding only to Ubx-Extradenticle heterodimer binding sites, potentially functioning in vivo to prevent Ubx monomers from binding and misregulating heterodimer target genes. However, removal of the surrounding region allows the YPWM motif to also inhibit binding to Hox-only recognition sequences. Despite a modular domain design for Hox proteins, these results suggest that multiple Hox protein regions form a network of regulatory interactions that coordinate context- and gene-specific responses. Since most nonhomeodomain regions are not conserved between Hox family members, these regulatory interactions have the potential to diversify binding by the highly homologous Hox homeodomains.

Frontorhiny, a distinctive presentation of frontonasal dysplasia caused by recessive mutations in the ALX3 homeobox gene

We describe a recessively inherited frontonasal malformation characterized by a distinctive facial appearance, with hypertelorism, wide nasal bridge, short nasal ridge, bifid nasal tip, broad columella, widely separated slit-like nares, long philtrum with prominent bilateral swellings, and midline notch in the upper lip and alveolus. Additional recurrent features present in a minority of individuals have been upper eyelid ptosis and midline dermoid cysts of craniofacial structures. Assuming recessive inheritance, we mapped the locus in three families to chromosome 1 and identified mutations in ALX3, which is located at band 1p13.3 and encodes the aristaless-related ALX homeobox 3 transcription factor. In total, we identified seven different homozygous pathogenic mutations in seven families. These mutations comprise missense substitutions at critical positions within the conserved homeodomain as well as nonsense, frameshift, and splice-site mutations, all predicting severe or complete loss of function. Our findings contrast with previous studies of the orthologous murine gene, which showed no phenotype in Alx3(-/-) homozygotes, apparently as a result of functional redundancy with the paralogous Alx4 gene. We conclude that ALX3 is essential for normal facial development in humans and that deficiency causes a clinically recognizable phenotype, which we term frontorhiny.

The interplay between transcription factors and microRNAs in genome-scale regulatory networks

Metazoan genomes contain thousands of protein-coding and non-coding RNA genes, most of which are differentially expressed, i.e., at different locations, at different times during development, or in response to environmental signals. Differential gene expression is achieved through complex regulatory networks that are controlled in part by two types of trans-regulators: transcription factors (TFs) and microRNAs (miRNAs). TFs bind to cis-regulatory DNA elements that are often located in or near their target genes, while miRNAs hybridize to cis-regulatory RNA elements mostly located in the 3' untranslated region of their target mRNAs. Here, we describe how these trans-regulators interact with each other in the context of gene regulatory networks to coordinate gene expression at the genome-scale level, and discuss future challenges of integrating these networks with other types of functional networks.

Cis-regulatory mechanisms of left/right asymmetric neuron-subtype specification in C. elegans

Anatomically and functionally defined neuron types are sometimes further classified into individual subtypes based on unique functional or molecular properties. To better understand how developmental programs controlling neuron type specification are mechanistically linked to programs controlling neuronal subtype specification, we have analyzed a neuronal subtype specification program that occurs across the left/right axis in the nervous system of the nematode C. elegans. A terminal selector transcription factor, CHE-1, is required for the specification of the ASE neuron class, and a gene regulatory feedback loop of transcription factors and miRNAs is required to diversify the two ASE neurons into an asymmetric left and right subtype (ASEL and ASER). However, the link between the CHE-1-dependent ASE neuron class specification and the ensuing left-right subtype specification program is poorly understood. We show here that CHE-1 has genetically separable functions in controlling bilaterally symmetric ASE neuron class specification and the ensuing left-right subtype specification program. Both neuron class specification and asymmetric subclass specification depend on CHE-1-binding sites (;ASE motifs') in symmetrically and asymmetrically expressed target genes, but in the case of asymmetrically expressed target genes, the activity of the ASE motif is modulated through a diverse set of additional cis-regulatory elements. Depending on the target gene, these cis-regulatory elements either promote or inhibit the activity of CHE-1. The activity of these L/R asymmetric cis-regulatory elements is indirectly controlled by che-1 itself, revealing a feed-forward loop configuration in which che-1 restricts its own activity. Relative binding affinity of CHE-1 to ASE motifs also depends on whether a gene is expressed bilaterally or in a left/right asymmetric manner. Our analysis provides insights into the molecular mechanisms of neuronal subtype specification, demonstrating that the activity of a neuron type-specific selector gene is modulated by a variety of distinct means to diversify individual neuron classes into specific subclasses. It also suggests that feed-forward loop motifs may be a prominent feature of neuronal diversification events.

The sine oculis homeobox (SIX) family of transcription factors as regulators of development and disease

The sine oculis homeobox (SIX) protein family is a group of evolutionarily conserved transcription factors that are found in diverse organisms that range from flatworms to humans. These factors are expressed within, and play pivotal developmental roles in, cell populations that give rise to the head, retina, ear, nose, brain, kidney, muscle and gonads. Mutations within the fly and mammalian versions of these genes have adverse consequences on the development of these organs/tissues. Several SIX proteins have been shown to directly influence the cell cycle and are present at elevated levels during tumorigenesis and within several cancers. This review aims to highlight aspects of (1) the evolutionary history of the SIX family; (2) the structural differences and similarities amongst the different SIX proteins; (3) the role that these genes play in retinal development; and (4) the influence that these proteins have on cell proliferation and growth.

The six family of homeobox genes in development and cancer

The homeobox gene superfamily encodes transcription factors that act as master regulators of development through their ability to activate or repress a diverse range of downstream target genes. Numerous families exist within the homeobox gene superfamily, and are classified on the basis of conservation of their homeodomains as well as additional motifs that contribute to DNA binding and to interactions with other proteins. Members of one such family, the Six family, form a transcriptional complex with Eya and Dach proteins, and together these proteins make up part of the retinal determination network first identified in Drosophila. This network is highly conserved in both invertebrate and vertebrate species, where it influences the development of numerous organs in addition to the eye, primarily through regulation of cell proliferation, survival, migration, and invasion. Mutations in Six, Eya, and Dach genes have been identified in a variety of human genetic disorders, demonstrating their critical role in human development. In addition, aberrant expression of Six, Eya, and Dach occurs in numerous human tumors, and Six1, in particular, plays a causal role both in tumor initiation and in metastasis. Emerging evidence for the importance of Six family members and their cofactors in numerous human tumors suggests that targeting of this complex may be a novel and powerful means to inhibit both tumor growth and progression.

Hox specificity unique roles for cofactors and collaborators

Hox proteins are well known for executing highly specific functions in vivo, but our understanding of the molecular mechanisms underlying gene regulation by these fascinating proteins has lagged behind. The premise of this review is that an understanding of gene regulation-by any transcription factor-requires the dissection of the cis-regulatory elements that they act upon. With this goal in mind, we review the concepts and ideas regarding gene regulation by Hox proteins and apply them to a curated list of directly regulated Hox cis-regulatory elements that have been validated in the literature. Our analysis of the Hox-binding sites within these elements suggests several emerging generalizations. We distinguish between Hox cofactors, proteins that bind DNA cooperatively with Hox proteins and thereby help with DNA-binding site selection, and Hox collaborators, proteins that bind in parallel to Hox-targeted cis-regulatory elements and dictate the sign and strength of gene regulation. Finally, we summarize insights that come from examining five X-ray crystal structures of Hox-cofactor-DNA complexes. Together, these analyses reveal an enormous amount of flexibility into how Hox proteins function to regulate gene expression, perhaps providing an explanation for why these factors have been central players in the evolution of morphological diversity in the animal kingdom.

Toward a systems-level understanding of developmental regulatory networks

Developmental regulatory networks constitute all the interconnections among molecular components that guide embryonic development. Developmental transcriptional regulatory networks (TRNs) are circuits of transcription factors and cis-acting DNA elements that control expression of downstream regulatory and effector genes. Developmental networks comprise functional subnetworks that are deployed sequentially in requisite spatiotemporal patterns. Here, we discuss integrative genomics approaches for elucidating TRNs, with an emphasis on those involved in Drosophila mesoderm development and mammalian embryonic stem cell maintenance and differentiation. As examples of regulatory subnetworks, we consider the transcriptional and signaling regulation of genes that interact to control cell morphology and migration. Finally, we describe integrative experimental and computational strategies for defining the entirety of molecular interactions underlying developmental regulatory networks.