Exploring the topography of physiological genomics
1999; American Physical Society; Volume: 1; Issue: 1 Linguagem: Inglês
10.1152/physiolgenomics.1999.1.1.21
ISSN1531-2267
Autores Tópico(s)Bioinformatics and Genomic Networks
ResumoNews and ReportsExploring the topography of physiological genomicsVICTORIA McGOVERNVICTORIA McGOVERNBurroughs Wellcome Fund, Durham, North Carolina 27702Published Online:15 Jul 1999https://doi.org/10.1152/physiolgenomics.1999.1.1.21MoreSectionsPDF (22 KB)Download PDF ToolsExport citationAdd to favoritesGet permissionsTrack citations ShareShare onFacebookTwitterLinkedInEmail ongoing genomic work in a number of model systems and the pending completion of the Human Genome Project provide remarkable new opportunities for physiology. In the 1997 Banbury Conference organized by the American Physiological Society, “Genomics to Physiology and Beyond: How Do We Get There?”, a vision for developing new infrastructure for studying complex diseases—from training scientists to developing crosscutting databases, to increasing physiologists' contact with genomics, to developing a journal to highlight this emerging field—was laid out. The journal on your computer screen grew from this vision.Researchers surveyed the domain of an emerging field in April 1999 at the annual meeting of the American Physiological Society (at Experimental Biology '99 in Washington, DC). Victor Dzau, Editor-in-Chief of Physiological Genomics and Chair of Medicine at Brigham and Women's Hospital, introduced the Journal and the field at the opening of the meeting's “Physiology InFocus” feature track, which highlighted the interplay between physiology and genomics. Providing a broad introduction to genomic approaches, the discussions focused on how new genetic and genomic tools are already improving understanding of physiology and human health. Genomics is maturing through the accumulation of sequence and the improvement of technologies for gathering, sorting, and understanding expression data. As methodologies accumulate for determining gene function on a large scale, opportunities for new physiological insight will continue to expand.The Human Genome Project, Dzau summarized, is expected to generate a preliminary map by 2001. Expressed sequence tag (EST) sequencing, which focuses on rapidly identifying genes that are expressed in specific tissues or under different environmental conditions, has already proven valuable in finding and mapping physiologically important genes. At the same time, random genomic sequencing has provided new insight into genome structure and has identified more new genes. At present, approximately 30,000 of the 75,000–100,000 genes predicted to be encoded by the human genome have been identified. Already, normal sequence variation and diversity are being catalogued and quantified.The journal Physiological Genomics will look at the whole spectrum from molecular genetics to integrated physiology. “There are enormous opportunities to take advantage of gene discovery, finding normal gene functions and linking genes to diseases,” said Dzau. “We've barely scratched the surface.”The techniques of physiological genomics bring together familiar approaches—biochemistry, molecular and cell biology, genetics, and classical physiology—and pair them with recent genetic technologies: yeast 1 hybrid and yeast 2 hybrid screening, the development of transgenic animal systems, and improved strategies for generating controlled gene knockouts. Improvements in computing and development of new methods for efficient sample handling have allowed the development of microarrays, which allow simultaneous analysis of many genes or many tissue samples. At the same time, development of proteomics, i.e., the systematic analysis of an organism's complete spectrum of proteins, and improvements in prediction of protein structures are enhancing and accelerating the correlation of newly identified genes with their products. These high-throughput technologies enhance the researcher's ability to ask questions, but the questions are fundamentally familiar: how can normal and disease states be linked to proteins and genes?Dzau highlighted one emerging technique, expression profiling, which allows one to look under varying physiological conditions at the transcripts made within tissues. By examining overlap between genes expressed in disease and normal conditions, it is possible to begin identifying genes responsible for disease. Extending this analysis to look at gene expression under cell culture conditions, in in vivo models, and in human tissue provides a robust set of tools for studying how gene function and regulation differs between health and disease.Richard Mulligan, a Senior Editor of the Journal, described work that demonstrates gene transfer's potential as a tool for understanding tissue function and developing new therapeutic approaches. The serendipitous observation that the DNA-binding dye Hoechst-33342 is pumped out of a very early hematopoietic stem cell class has allowed Mulligan's group to develop reliable techniques for isolating these cells. They have used these isolated cells to discover unexpected differentiation potential of the hematopoietic line. Their work has exploited this potential to make improvements in the muscle fibers of a muscular dystrophy model mouse that lacks the structural protein dystrophin. After transferring bone marrow cells from male dystrophin-positive animals to females of the disease model strain, Mulligan's group observed the development of Y chromosome-containing muscle cells expressing dystrophin within the muscles of the animals. Recent experiments have shown that the early class of hematopoietic stem cells also has the potential for development into vascular epithelium, vascularizing transplanted hearts in a mouse model system.Jeffrey Trent from the National Human Genome Research Institute described the use of the new “tissue chip” microarrays, reagents that allow simultaneous sampling of tissues from hundreds to thousands of clinical specimens. Samples on the tissue chip are linked to clinical data sets, making possible broad comparisons of both molecular and clinical data.Using cDNA microarrays, reagents in which thousands of genes may be simultaneously assayed for expression, Trent's group has looked at the development of fetal myofibroblasts into the rapidly invasive cancer, alveolar rhabdomyosarcoma. A subset of this cancer has been identified in which a specific chromosome 2-to-chromosome 3 translocation appears to be the discrete tumor-inducing event. Coupling these microarray technologies, it is possible to gain a better understanding of the disease: first, cDNA microarrays are used to identify the genes affected by the translocation, and then tissue microarrays allow examination of the expression of those genes within the context of clinically important human tumors.The interplay of environmental, genetic, and physiological factors that goes into controlling complex biological processes makes dissecting systems a daunting task. Senior Editor Allen Cowley, Jr., discussed the role genomics has played in identifying genes involved in blood pressure regulation and hypertension. With rapidly accumulating genomic tools and a wealth of biochemical and physiological data available, the rat is a vigorous model system for studying hypertension. Controlled inbreeding of rats allows the population variations that complicate human studies to be put aside. Whereas physiology has traditionally moved from function to gene, the use of genomic data and an animal model allows Cowley's group to work in the opposite direction, from gene to function. Using salt-stress experiments, Cowley has identified “likely determinant traits” or quantitative trait loci (QTLs), genes whose expression correlates with blood pressure.The volume of data produced by these experiments is daunting, however. An experiment using more than 300 mice to look at expression of more than 200 genes generates more that 15,000,000 data points, highlighting the need for data-management tools far more powerful than those traditionally used by biologists.Mark Adams, Vice President of Genome Programs at Celera Genomics, elaborated on the need for bioinformatics and described the state-of-the-art strategies now being used to generate sequence data. Assembly of fragmentary sequence data into the complete genomic sequence of an organism is a significant intellectual and computational task, Adams explained. Management of information from the human genome's nearly 3.5 billion base pairs and 75,000–100,000 genes would be a difficult task on its own. But because one-third of the human genome is in repetitive elements, assembling sequence reads into useful genomic data requires bioinformatic methods that can reliably discriminate between extensively similar sequences.Bioinformatics goes beyond simple data management, however, and provides tools for analyzing the multidimensional information that comes from global analysis of gene and protein expression. With thousands of genes of unknown function being identified in the human and other model system genome projects, methods for predicting the structure of proteins and for discerning how they interact have taken on new importance.Although DNA-based prediction of the structure of whole proteins remains difficult, prediction of protein substructure has improved in recent years. Even this limited structural information can provide insight into how proteins interact to form complex networks in cells. “To understand life, you have to get the parts list and get all the connections,” Roger Brent, who has done much of the development of two-hybrid selection systems, explained. Two-hybrid strategies provide an experimental approach to finding these connections by identifying, one pair at a time, proteins that interact in the cell.Beyond gaining insight into how proteins interact within pathways, understanding the proteome, the complete complement of proteins encoded by the genome, will be an important milestone for human biology. Lee Anderson described strategies that approach defining an organism's proteome. With the use of quantitative protein level measurements from high-resolution two-dimensional gel electrophoresis and high-throughput mass spectrometry of proteins, it is possible to gather data on protein expression by cells growing normally and in disease.Anderson described his group's work using two-dimensional gel electrophoresis to identify proteins involved in cholesterol synthesis. With these proteins identified, Anderson's group has backtracked to gain insight into the functions and mechanisms of the pathway's steps.Steve Gullans, who is leading the Normal Human Tissue Gene Expression Index Project at Brigham and Women's Hospital, described this growing database. The Index will be a comprehensive resource that organizes information about the normal scope of human gene expression in a range of different tissues. It will provide a substantial foundation for studies correlating genotype with phenotype within the background of normal population variation. Comparison of gene expression within and across tissues provides an effective window for appreciating normal tissue function. Collating the data from expression studies across both tissues and across populations creates a potent tool for investigating both basic function and disease.Once genes have been identified and the normal state of their expression is understood, the ability to create specific mutations at will is invaluable for dissecting function. Brian Sauer, who has developed the Cre-lox recombination system into a powerful tool for manipulating gene expression, described the function of the Cre protein, a site-specific recombinase, and the lox site, the target of Cre. By exploiting this recombination system, it is possible to develop both externally controlled molecular switches for the study of tissue-specific gene expression and to insert assayable markers into naturally regulated genes of interest. These approaches allow the study of temporal or tissue-specific expression of genes in different contexts.Curtis Sigmund has used the Cre-lox system to further his studies of blood pressure homeostasis. Sigmund's lab uses the renin-angiotensin system to dissect mechanisms that separate endocrine from tissue-specific effects. He has made constructs that under normal conditions are unchanged in tissue-specific expression of angiotensinogen but, in the presence of exogenously added Cre, lose angiotensinogen gene function, allowing him to knock out the gene after the animal has developed.Susumu Tonegawa has used the Cre-lox system in a different way, to study the complex physiology of memory. By specifically knocking out expression of a group of receptors in a single layer of the mouse hippocampus before the brain develops, he has been able to experimentally manipulate a behavioral property, the ability of mice to carry out certain memory-dependent tasks.In the final segment of the discussion of genomics and physiology, several researchers using different tools to genetically dissect the function of the sarcomere described complementary work. John Solaro described how his group has been using transgenic animals to study thin filament regulation. They have systematically substituted altered proteins into myofilaments, allowing exploration of the interactions between proteins within this sarcomeric machine. Joe Metzger discussed experiments in which he used gene transfer by DNA viruses to gain insight into the same thin filament structure.Using the virus system, Metzger's group has generated isolated virus-infected cardiac muscle cell systems that express mutations linked to familial hypertrophic cardiomyopathy, essentially creating single cell transgenic models that are robust for several days before the isolated cells dedifferentiate. By generating tropomyosin mutants that mirror each of four missense mutations linked to human hypertrophic cardiomyopathy, Metzger has been able to show that force response to calcium varies in the mutants. Margaret Westfall discussed experiments using gene transfer into adult cardiac myocytes to look at the effects of troponin I isoforms on cardiac function. Westfall has also done work generating chimeric troponin I proteins to explore thin filament regulation.Eric Olson also described work focused on factors involved in hypertrophism of the heart. Looking for calcium signaling systems involved in heart development, Olson's group has identified a family of calcineurin-regulated transcriptional regulators sharing homology in a regulatory domain. Work with transgenic mice and experiments in tissue culture have allowed Olson's group to explore the activation of a pathway that can lead to cardiac hypertrophy or to more efficient working of the healthy heart. Jeff Robbins has also used transgenic mice, studying the effects of expression of different isoforms of the myosin light chain.Leslie Leinwand discussed gene-targeting strategies for understanding chemomechanical coupling in skeletal muscle cells. She has inactivated the two major heavy chain gene classes, which give rise to 80–85% of the myosin heavy chains in adult cells, and surprisingly sees very different phenotypes, depending on which of these similar proteins are removed.The 1997 Banbury question—genomics to physiology and beyond: how do we get there?—may answer itself over time, for who could stop this field from getting there? Genomics is bringing broader, more physiological approaches to understanding biology. Genetics, and especially the use of transgenesis, has already had a tremendous impact on physiological studies. Resources ranging from the proteome to the Normal Human Tissue Gene Expression Index to microarrays are tools that have only begun to reveal their potential. Physiology's questions can increasingly be approached with an ambition and scope unimaginable a generation ago.This article has no references to display. Previous Back to Top Next FiguresReferencesRelatedInformationCited ByDevelopment of physiological regulatory systems: altering the timing of crucial eventsZoology, Vol. 106, No. 2 More from this issue > Volume 1Issue 1July 1999Pages 21-23 Copyright & PermissionsCopyright © 1999 the American Physiological Societyhttps://doi.org/10.1152/physiolgenomics.1999.1.1.21PubMed11015557History Published online 15 July 1999 Published in print 15 July 1999 PDF download Metrics Downloaded 71 times
Referência(s)