Genome Structure and Modularity

March 20, 2008 | By Dan In Books, Journal Club |

A minireview recently in Genomics caught my eye with the title Coexpression, coregulation, and cofunctionality of neighboring genes in eukaryotic genomes that sounded just like a passage that I recalled from Richard Dawkins’ The Selfish Gene:

…the ‘environment’ of a gene consists largely of other genes, each of which is itself being selected for its ability to cooperate with its environment of other genes. (page 39) … Genes are selected, not as ‘good’ in isolation, but as good at working against the background of other genes in the gene pool. A good gene must be compatible with, and complementary to, the other genes with whom it has to share a long succession of bodies. (page 84)

This appears to in fact be the case. Gene linkage has been a cornerstone of molecular genetics since shortly after the rediscovery of Mendel’s laws of inheritance, occurs when particular genetic loci or alleles for genes are inherited jointly. Genetic loci on the same chromosome are physically connected and tend to segregate together during meiosis, and are thus genetically linked. Alleles for genes on different chromosomes are usually not linked, due to independent assortment of chromosomes during meiosis.

Crossing over can occur during meiosis, however, occasionally recombining the ways in which genes are grouped on a given chromosome. Because crossing over runs the risk of breaking up a ‘good team’ of gene alleles, it could be potentially more detrimental to the organism than beneficial. And indeed crossing over does not occur extremely often; moreover, when it does happen, it inevitably results in swapping of homozygous alleles for some loci, and the net change in gene groupings is restricted more to the groupings of polymorphisms. In this way, Dawkins’ analogy of a professional team sport, where one or more members may be traded from time to time, is very appropriate to changes in chromosomal arrangements.

The minireview that I started off this post has more to say however, by noting that multiple mechanisms exist that promote the co-expression of neighboring genes.

This local coexpression, typically measured as a correlation between the expression levels of genes positioned close to each other, can be explained by multiple biochemical, genetic, evolutionary, and technological factors. Such elements of genomic structure as overlapping genes, tandemly duplicated genes, homologous genes, and operons come first to mind as logical candidate determinants of coexpression. Although these elements seem to enrich coexpression clusters in some cases, they do not account for the remaining part of the coexpression pattern. Therefore, it has been hypothesized that coexpression of neighboring genes can be determined by chromatin domains, or multigene segments of DNA, which, in a given cell at a given moment, are consistently either euchromatin or heterochromatin. When chromatin opens during gene expression this may simultaneously facilitate expression of genes from neighborhoods of the open region. Genomes would thus be compartmentalized into chromatin domains, with their location possibly varying between cell types to deliver tissue-specific chromatin conformation and concerted transcriptional activity. Alternatively, coexpressed gene clusters are formed through local sharing regulatory elements such as transcription factors, promoters, and enhancers. In the human genome, for example, more than 10% of genes form head-to-head pairs that may be subject to bidirectional expression mediated by common promoter sequences.

This model explicitly focuses on cis-acting elements in the untranscribed regions surrounding gene loci, and the authors note that, while not discussed, trans-acting factors are indispensible. Moreover, modularity does appear to be an indispensible rule for managing complexity of any system exhibiting coexpression, coregulation, and cofunctionality of its basic components. Call modularity an emergent property if you will.