Gill Bejerano - US grants

DESCRIPTION (provided by applicant): Characterization of regulatory elements leading to human limb malformations ABSTRACT Limb malformations are the second most common human congenital abnormality with a prevalence of 1 for every 500 births. Although several mutations in genes have been identified that explain syndromic forms (associated with other symptoms) of limb malformations, the characterization of mutations causing non- syndromic/isolated limb malformations has been less successful. A variety of molecular and clinical data suggests that mutations responsible for non-syndromic limb malformations may reside in distant noncoding regulatory sequences such as enhancers (sequences that regulate gene promoters). These data are based on position effects (chromosomal rearrangements that leave the gene intact but remove its regulatory elements) that lead to limb malformations, the observed modular nature of enhancers, and the recent example of a non- syndromic preaxial polydactyly in humans that has been linked to a long distance enhancer of the Sonic Hedgehog (SHH) gene. Long distance regulatory enhancers have traditionally been difficult to identify and very few of them have been characterized for limb regulatory expression thus far. In preliminary studies for this proposal, we have discovered 43 novel human limb enhancers using a mouse enhancer transgenic assay and verified several of them for pectoral fin expression in zebrafish. In order to discover additional human limb enhancers we are using advanced computational tools to dissect the unique sequence signatures in both the novel limb enhancers we discovered and previously reported ones. These signatures allow us to predict novel limb enhancers surrounding known limb-associated genes and throughout the human genome. These predicted limb enhancers will initially be tested in a high-throughput manner in zebrafish for fin expression. Positive fin enhancers will then be reverified in mice for limb expression. All characterized enhancers both in zebrafish and mouse will be available to the biomedical community through a web accessible browser. In addition, we have collected numerous DNA samples of patients with non-syndromic limb malformations and are in the process of collecting numerous more. We will conduct mutation analysis of these DNA samples within limb enhancers, and potential causative nucleotide changes will be tested for their effect on limb formation using the mouse as our model. The identification of causative sequences leading to non-syndromic limb malformations will result in improved patient counseling, the development of molecular testing including prenatal genetic testing, and an increased knowledge about the pathogenesis of human limb malformations and limb development. PUBLIC HEALTH RELEVANCE: Characterization of regulatory elements leading to human limb malformations Mutations in genes leading to limb malformations, the second most common human congenital abnormality with a prevalence of 1 for every 500 births, have been discovered on the majority in a syndromic form (associated with other symptoms). There is a variety of molecular and clinical data to suggest that non- syndromic (isolated) limb malformations can be caused by mutations in distant regulatory noncoding sequences (DNA switches that tell the genes when and where to turn on or off), but only a small number of limb regulatory noncoding sequences have been discovered thus far and only one of these sequences has been linked to human non-syndromic limb malformations. In this proposal we have discovered 43 novel human limb regulatory noncoding sequences and have collected numerous DNA samples from patients with non- syndromic limb malformations enabling us not only to screen these 43 DNA sequences for mutations in non- syndromic patients, but to also identify in a high-throughput manner using advanced computational tools and zebrafish and mouse assays, novel limb regulatory sequences throughout the human genome, which will make for additional limb malformation mutation candidates.

DESCRIPTION (provided by applicant): Zebrafish with its growing arsenal of tools that allow the generation of transgenics, gene knockdowns and knockouts, and mutant resources coupled with its high-throughput and cost efficiency is quickly becoming the major animal model for drug screens and gene related studies. However, as with other vertebrate genomes, the majority of the zebrafish genome (97%) is made up of non-genic sequences whose functional necessity remains largely unknown. One vital function that is clearly embedded in these regions is gene regulation, instructing genes when and where to turn on or off. However, unlike genes where we know their genomic location, their code, and the consequences of nucleotide changes within them, in gene regulatory sequences we don't have that knowledge. This knowledge is extremely vital, with a wide variety of clinical and molecular data supporting these sequences to be an important driver for development, evolution, diversity, and disease. In this proposal, we will combine advanced computational tools with high-throughput zebrafish functional studies to annotate this noncoding terrain. Using and refining multiple vertebrate genome alignments we have generated an unprecedented set of 166,693 zebrafish conserved noncoding elements (CNEs), with at least 8,805 regions having a direct ortholog in the human genome. Preliminary studies for a portion of these sequences using a zebrafish transgenic enhancer assay, find 41% of these sequences to function as enhancers at 24 to 48 hours post fertilization. Taking advantage of this transgenic assay we aim to screen 200 sequences a year for enhancer activity. These sequences will be selected from our large CNE set, sequences whose enhancer activity and tissue-timepoint specificity will be predicted using sophisticated computational tools, and community requested sequences. This characterization will not only allow the functional annotation of these sequences, but will also generate a novel and extremely important toolkit of gene regulatory elements that can drive expression of any gene of interest at precise locations and precise developmental time points. In addition, we will also use the annotated regulatory landscape to discover novel genes with potential important developmental function. This will be carried out by analyzing the expression patterns and functional consequences due to knockdown of less characterized genes that lie in rich regulatory regions, a common sign for the existence of important developmental gene regulators. Additional computational techniques will be used to discover genes under tight regulation in novel tissue contexts, as well as pathways which are currently not studied in the context we find them enriched in. All the data generated in this proposal, both computational and functional, will be made available to the community through a dedicated web browser (http://zebrafish.stanford.edu/) as well as integration into ZFIN, Ensemble, and the UCSC genome browser. Combined, our work will advance zebrafish as the major animal model for annotating and characterizing the noncoding portion of the vertebrate genome. PUBLIC HEALTH RELEVANCE: Computational &Functional Annotation of the Zebrafish Genome Regulatory Toolbox While genes make up less than 3% of our DNA, within the remaining 97% lie other numerous extremely important sequences such as gene regulatory elements, that instruct the genes when and where to turn on or off. Mutations in these gene regulatory elements can have a great impact on human disease, yet their location and code still remains on the majority unknown. In this proposal we will take advantage of the unique properties of the zebrafish model organism to couple advanced computational tools with rapid functional zebrafish assays to annotate these sequences and obtain a better understanding of the vertebrate gene regulatory code, which will be of extreme importance to our comprehension of the genetic cause for numerous human diseases.

Computational & Functional Annotation of the Zebrafish Genome Regulatory Toolbox Zebrafish with its growing arsenal of tools that allow the generation of transgenics, gene knockdowns and knockouts, and mutant resources coupled with its high-throughput and cost efficiency is quickly becoming the major animal model for drug screens and gene related studies. However, as with other vertebrate genomes, the majority of the zebrafish genome (97%) is made up of non-genic sequences whose functional necessity remains largely unknown. One vital function that is clearly embedded in these regions is gene regulation, instructing genes when and where to turn on or off. However, unlike genes where we know their genomic location, their code, and the consequences of nucleotide changes within them, in gene regulatory sequences we don't have that knowledge. This knowledge is extremely vital, with a wide variety of clinical and molecular data supporting these sequences to be an important driver for development, evolution, diversity, and disease. In this proposal, we will combine advanced computational tools with high-throughput zebrafish functional studies to annotate this noncoding terrain. Using and refining multiple vertebrate genome alignments we have generated an unprecedented set of 166,693 zebrafish conserved noncoding elements (CNEs), with at least 8,805 regions having a direct ortholog in the human genome. Preliminary studies for a portion of these sequences using a zebrafish transgenic enhancer assay, find 41% of these sequences to function as enhancers at 24 to 48 hours post fertilization. Taking advantage of this transgenic assay we aim to screen 200 sequences a year for enhancer activity. These sequences will be selected from our large CNE set, sequences whose enhancer activity and tissue-timepoint specificity will be predicted using sophisticated computational tools, and community requested sequences. This characterization will not only allow the functional annotation of these sequences, but will also generate a novel and extremely important toolkit of gene regulatory elements that can drive expression of any gene of interest at precise locations and precise developmental time points. In addition, we will also use the annotated regulatory landscape to discover novel genes with potential important developmental function. This will be carried out by analyzing the expression patterns and functional consequences due to knockdown of less characterized genes that lie in rich regulatory regions, a common sign for the existence of important developmental gene regulators. Additional computational techniques will be used to discover genes under tight regulation in novel tissue contexts, as well as pathways which are currently not studied in the context we find them enriched in. All the data generated in this proposal, both computational and functional, will be made available to the community through a dedicated web browser (http://zebrafish.stanford.edu/) as well as integration into ZFIN, Ensembl, and the UCSC genome browser. Combined, our work will advance zebrafish as the major animal model for annotating and characterizing the noncoding portion of the vertebrate genome.

? DESCRIPTION (provided by applicant): Recent genetic advances have driven significant progress in scientists' abilities to classify and map neuronal cell types within the brains of mode organisms like laboratory mice. To better delineate neuronal cell types in the human brain, however, it is critical to have a deeper understanding of the way that neuronal cell types evolve across mammals. As a first step toward achieving this understanding, corresponding neuronal cell types will be directly compared in two closely related mammalian species: mice and rats. By closely examining differences in the properties of these cells, including the genes they express, we hope to identify genomic elements that control the properties of neuronal cell types, and to infer properties of the corresponding cell types in the human brain. Improving the precision with which we can classify human neuronal cell types could have a transformative impact on our ability to understand pathological changes in neuropsychiatric disease.