Overview
Three crucial biological functions occur within chromosomes: DNA replication, genetic recombination at meiosis (the specialized cell division that results in a single set of chromosomes in sex cells), and gene expression. The physical organization of chromosomes, determined by evolutionary selection, optimizes these processes. In studying the relationships, our laboratory has developed evidence that a large fraction of the mammalian genome is organized into functional domains. We find that extensive clustering of functionally related genes exists in the human, mouse and chicken genomes. In addition, we are developing and testing analytical methods for "walking" along chromosomes and testing contiguous sets of genes to determine if they have an increased likelihood to interact in the same biological network. Our ultimate goal is to understand the functional organization of chromosomes, both in terms of the evolutionary mechanisms involved and for its immediate relevance to mapping and identifying human disease susceptibility genes.
Research details
Functional Organization of Mammalian Genomes
Previous work from our group revealed a strong tendency for functionally related genes to be located in close proximity as a means of promoting the co-evolution and co-inheritance of favorable combinations of alleles at these loci (Petkov et al., 2005). We have now extended these studies using the complementary approaches of developing means to determine if groups of functionally related genes show genomic clustering, and means to test whether contiguous sets of genes are enriched in genes that interact in the same biological network. To compare clustering among sets of functionally related genes, we have developed an extension of r-scan statistics for application to mammalian systems. For any set of genes, it is now possible to calculate parameters that describe the extent to which the set shows genomic clustering, and that describe the anatomy of the set in terms of the probability that each gene in the set contributes to clustering.
We find that extensive functional clustering exists in the human, mouse, and chicken genomes. Clustering is not a consequence of recent gene duplications. Sequence comparisons show that nearly all gene duplications that gave rise to members of a set of functionally related genes arose, functionally diverged and relocated to distant sites, long before the mammalian radiation 75-80 million years ago, and have evolved very slowly since then. The genes encoding proteins that have retained any significant measure of sequence similarity are almost invariably on separate chromosomes; even adjacent genes encoding members of the same protein set, show little sequence identity. In contrast, orthologous proteins show little divergence between human and mouse; median identities are typically 88-89 percent over 100 percent of protein length. In addition, gene expression analyses show that functional clusters do not represent chromatin domains of shared regulation.
The primary evolutionary driver for functional gene clustering has been the selective advantage of co-inheriting co-adapted sets of alleles among functionally related genes, and to some extent, the tendency for certain classes of ancient gene duplications to remain in proximity, rather than disperse across the genome as others have done. Confirming the ancient evolutionary origin of these arrangements, there is considerable evidence of parallel functional clustering in the chicken genome.
In a complementary approach, we are developing and testing analytical methods for "walking" along chromosomes and testing contiguous sets of genes for participation in common functional pathways. Using the Ingenuity Pathways Analysis database, about 40 percent of testable genes occur in clusters that participate in common biological networks, with functionally unrelated genes interspersed among them. Using this approach, we have now tested the entire human and mouse genomes; our results suggest that nearly the entire mammalian genome is organized in this way.
What has emerged is a picture of a dynamic genome in which phenotypes are the outcome of interaction between alternate alleles at many genes rather than the sum of the individual components (in genetic terms, there is exceptionally extensive epistasis), with this organization rearranged at every generation by the independent assortment of chromosomes and the reshuffling of chromosomal regions by the process of recombination. This pattern of genomic organization influences both the nature of genetic variation within and between populations and the evolution of species, and has immediate practical consequences for efforts to map and identify genes important in human health and disease.
Because genetic recombination plays such a large role in producing new allelic combinations at each generation, we are carrying out parallel studies characterizing the location and properties of the chromosomal sites at which recombination occurs. In mammals, recombination occurs at special sites called hotspots, rather than being randomly located along chromosomes. Hotspots are typically about 1 Kb long, and are surrounded by long stretches, sometimes up to a megabase, where recombination is absent. Very little is known about hotspots beyond the fact that they exist and can vary several orders of magnitude in their ability to initiate recombination.
To fill this void, we have now mapped a 10 Mb region on mouse Chromosome 1 at very high resolution and found that hotspots can be clustered into "torrid zones" flanked by long stretches with no recombination. We are now mapping recombination hotspots along the entirety of mouse Chromosomes 1 and 11 and four other selected regions of the mouse genome using 6,000 offspring each from crosses between strains C57BL/6J (B6) and CAST/EiJ (CAST), and between WSB/EiJ (a pure M. m. domesticus strain) and PWD/PhJ (a pure M. m. musculus strain). We have learned that the average spacing between hotspots is approximately 100Kb, although this is highly variable; that hotspots of different activity classes are equally frequent, making most recombination occur at the hottest hotspots; and that recombination in both sexes occurs at the same hotspots, but often at different rates. The sex ratio of recombination is regionally determined (female higher v. male), but within regions, hotspots can vary in their ratio.
Using this data, we have shown that the difference in recombination rates between the sexes is a consequence of sex differences in the interference distance (the distance over which double-crossovers are forbidden), and that, in turn, this is a result of sex differences in degrees of chromatid compaction at the pachytene stage of meiosis I.
Extending these analyses, we have compared the recombination maps for distal Chromosome 1 obtained in crosses between the inbred strains B6 and CAST, and when only the distal half of Chromosome 1 is heterozygous B6/CAST, the remainder of genome being homozygous B6. A number of hotspots disappeared from the latter cross, while entirely new ones appeared. We conclude that trans-acting genes provide hotspot specific factors that control the activity of individual hotspots.
Our final project is an effort to bypass the interferon response in mammalian cells and use RNAi libraries as a means of developing a new mammalian somatic cell genetic system for identifying the genes underlying important cellular phenotypes. Using RNAi provides a means of overcoming the historical problems with somatic genetics - the difficulty of obtaining recessive mutations (RNAi suppression is dominant) and the difficulty of identifying the mutated gene (active RNAi molecules can be sequenced to determine their target gene).
Lab staff
Principal Investigator: Kenneth Paigen, Ph.D.
Co-Principal Investigator: Petko M. Petkov, Ph.D.
Postdoctoral Associates: Siemon Ng, Ph.D., Emil Parvanov, Ph.D., Lorin Petros, Ph.D.
Research Assistant IV: Evelyn Sargent, B.S.
Research Assistant I: Rose Madeira, B.S.
Laboratory Assistant III: Anita Hawkins
Bioinformatics Specialist: Michael B. Walker, B.S.
Research Administrative Assistant: Patricia Cherry
Publication listings
Perkov PM, Graber JH, Churchill GA, DiPetrillo K, King BL, Paigen K. 2007. Evidence of large-scale functional organization of mammalian chromosomes. PLoS Biol 5(5):2127.
Petkov PM, Broman KW, Szatkiewicz JP, Paigen K. 2007. Crossover interference underlies sex differences in recombination rates. Trend Genet 23(11): doi:10.1016/j.tig.2007.08.015.
Graber JH, Churchill GA, DiPetrillo KJ, King BL, Petkov PM, Paigen K. 2006. Patterns and mechanisms of genome organization in the mouse. J Exp Zool 305A:683-688.
Petkov P, Graber JH, Churchill GA, DiPetrillo K, King BL, Paigen. 2005. Evidence of a large scale functional organization of mammalian chromosomes. PLoS Genet 1(3):312-322.
Kelmenson PM, Petkov P, Wang X, Higgins DC, Paigen BJ, Paigen K. 2005. A torrid zone on mouse chromosome 1 containing a cluster of recombinational hotspots. Genetics 169:833-841.
Paigen K. 2004. Understanding the human condition: experimental strategies in mammalian genetics. ILAR J 43: 123-135.
Paigen K. 2003. One hundred years of mouse genetics: An intellectual history. II. The molecular revolution (1981-2002). Genetics 163:1227-1235.
Paigen K. 2003. One hundred years of mouse genetics: An intellectual history. I. The classical period (1902-1980). Genetics 163:1-7.
