Finding general laws on the organization principle of living organisms is a particularly difficult task in biology but certainly a central one in systems biology. Part of the difficulty in this endeavor is probably linked to the fact that “by its very nature, life is both contingent and particular, each organism the product of eons of tinkering, of building on what had accumulated over the course of a particular evolutionary trajectory” (Keller, 2007, see also our post). Such laws are thus particularly significant when they emerge from evolutionary constraints alone. In a recent paper published in PNAS, Matthew Wright and colleagues may well provide such an example by looking at the “”http://dx.doi.org/10.1073/pnas.0610776104">chromosomal periodicity of evolutionarily conserved gene pairs" (Wright et al, 2007).
Using a comparative genomic approach, Wright and colleagues selected pairs of genes based on two simple criteria: 1) the genes of a pair have to have a tendency to be close together; 2) one gene of the pair should tend to be present only if the other gene is also present. Searching more than 100 bacterial genomes, 22’500 statistically conserved gene pairs could be identified. Looking at the distribution of distances between genes in a pair and at the density of conserved pair along E. coli chromosome, a strikingly regular pattern emerged: conserved pairs appear to be localized as clusters that are regularly spaced over the entire chromosome, with a regular inter-cluster interval of 117kb. In addition, this regular positional pattern correlates with the pattern of log-phase transcriptional activity along the chromosome: both positional and transcriptional grids are almost perfectly aligned, with the same 117kb periodicity (see figure below, from Figure 3b in Wright et al, Copyright 2007 The National Academy of Sciences of the USA).
The interpretation offered to explain these findings is that the regular spacing of conserved gene pairs may reveal underlying regularities of the structural spatial organization of E. coli chromosome. Specifically, a solenoid-like model with regular 117kb loops would imply that conserved pairs are preferentially located on one face of the chromosome. Correlation between the positional grid and the longitudinal profile of transcriptional activity suggests that this arrangement is coupled to functionally important characteristics (eg diffusion properties of the RNA polymerase or existence of transcription factories)
A patterned structure with a similar periodicity has been suggested by previous studies based the analysis of sequence features or on the profile transcriptional activity along the bacterial chromosome (Jeong et al, 2004, Carpentier et al, 2005, Allen et al, 2006). What is remarkable in the study by Wright and colleagues, is that the 117kb periodicity emerges so clearly by using solely evolutionary conservation criteria: chromosomal proximity and phylogenetic co-occurrence. The evolutionary forces that operate on a wide variety of genomes are thus able to reveal constraints on the overall structural organization of an entire bacterial chromosome. In turn, this finding implies that strong evolutionary selective pressure operate to shape the long-range organization of chromosomes. How general is this 117kb-periodicity law? Wright and colleagues were able to find a similar arrangement in C. crescentus and it will be interesting to see if a similar organization is observed in other genomes. Direct investigation of chromosomal conformation in vivo may also shed more light on the physical and functional mechanisms that explain the deep link between evolutionary conservation of local properties and a global architectural principle of a bacterial genome.