Isochores: dream or reality?

Abstract

In their Nature paper [1xInitial sequencing and analysis of the human genome. The International Human Genome Sequencing Consortium. Nature. 2001; 409: 860–921Crossref | PubMed | Scopus (12902)See all References[1], the International Human Genome Sequence Consortium (IHGSC) studied ‘the draft genome sequence to see whether strict isochores could be identified’ and failed to find any. They concluded that their results ‘rule out a strict notion of isochores as compositionally homogeneous’ and that ‘isochores do not appear to deserve the prefix “iso”.’The terminology ‘strict isochores’ was used by the authors to denote sequences that cannot be distinguished from random (uncorrelated) sequences in which every nucleotide is free to change. Therefore, the authors’ failure to identify random sequences masquerading as ‘strict isochores’ in the human genome could be predicted on three accounts. First, since the work of Rolfe and Meselson [2xThe relative homogeneity of microbial DNA. Rolfe, R and Meselson, M. Proc. Natl. Acad. Sci. U.S.A. 1959; 45: 1039–1043Crossref | PubMedSee all References[2] over 40 years ago, random sequences are known to be much more homogeneous than the least heterogeneous genomic DNAs, namely bacterial DNAs (which are, in turn, much less heterogeneous than mammalian DNAs). Second, it has been known for some time that the standard deviations of CsCl profiles of major DNA components from mammalian genomes (namely the compositional families of 50–100-kb DNA molecules derived from isochore families) are comparable to those of bacterial DNAs of the same size and composition [3.xAn analysis of the bovine genome by Cs2SO4-Ag+ density gradient centrifugation. Filipski, J et al. J. Mol. Biol. 1973; 80: 177–197Crossref | PubMed | Scopus (143)See all References, 4.xThe major components of the mouse and human genomes: preparation, basic properties and compositional heterogeneity. Cuny, G et al. Eur. J. Biochem. 1981; 115: 227–233Crossref | PubMedSee all References]. Third, ‘strict isochores’ cannot exist in any natural DNA because noncoding sequences are compositionally correlated with the coding sequences that they embed [5.xThe mosaic genome of warm-blooded vertebrates. Bernardi, G et al. Science. 1985; 228: 953–958Crossref | PubMed | Scopus (427)See all References, 6.xHuman coding and non-coding DNA: compositional correlations. Clay, O et al. Mol. Phylogenet. Evol. 1996; 5: 2–12Crossref | PubMed | Scopus (89)See all References], and coding sequences are made up of codons, in which the compositions of the three positions are correlated with each other [7xA universal compositional correlation among codon positions. D'Onofrio, G and Bernardi, G. Gene. 1992; 110: 81–88Crossref | PubMed | Scopus (58)See all References[7]. More detailed discussions of this problem are presented elsewhere [8.xMisunderstandings about isochores. Bernardi, G. Gene. 2001; 276: 3–13Crossref | PubMed | Scopus (91)See all References, 9.xThe isochores in human chromosomes 21 and 22. Clay, O and Bernardi, G. Biochem. Biophys. Res. Commun. 2001; 285: 855–856Crossref | PubMed | Scopus (10)See all References, 10.xCompositional heterogeneity within and among isochores in mammalian genomes. II. Some general comments. Clay, O and Bernardi, G. Gene. 2001; 276: 25–31Crossref | PubMed | Scopus (14)See all References, 11.xStandard deviations and correlations of GC levels in DNA sequences. Clay, O. Gene. 2001; 276: 33–38Crossref | PubMed | Scopus (21)See all References, 12.xA compact view of isochores in the draft human genome sequence. Pavlicek, A et al. FEBS Lett. 2002; 511: 165–169Abstract | Full Text | Full Text PDF | PubMed | Scopus (48)See all References].The conclusion of the IHGSC authors that isochores are more heterogeneous than random sequences (or than ‘strict isochores’, to use the misleading terminology of the authors) is correct, but it is something we have known for at least two decades. The definition of isochores as ‘fairly homogeneous sequences’ [4xThe major components of the mouse and human genomes: preparation, basic properties and compositional heterogeneity. Cuny, G et al. Eur. J. Biochem. 1981; 115: 227–233Crossref | PubMedSee all References[4] is still valid.Unfortunately, denying the existence of isochores not only represents a mistake in itself, but it also means denying the existence of compositional discontinuities in the human genome and going back to a genome organization characterized by a continuous compositional variation, a view shown to be wrong in the early 1970s [3xAn analysis of the bovine genome by Cs2SO4-Ag+ density gradient centrifugation. Filipski, J et al. J. Mol. Biol. 1973; 80: 177–197Crossref | PubMed | Scopus (143)See all References[3]. To deny the existence of isochores means to deny the existence of what has been recently summarized as ‘a fundamental level of genome organization’ [13xThe evolution of isochores. Eyre-Walker, A and Hurst, W. Nat. Rev. Genet. 2001; 2: 549–555Crossref | PubMed | Scopus (258)See all References[13].