Why so many noncoding nucleotides? The eukaryote genome as an epigenetic machine

Abstract

It is recalled that dispensability of sequences and neutral substitution rate must not be construed to be markers of nonfunctionality. Different aspects of functionality relate to differently-sized nucleotide communities. At the time cells became nucleated, a boom of epigenetic processes led to uses of DNA that required many more nucleotides operating collectively than do functions definable in terms of classical genetics. Each order of magnitude of nucleotide plurality was colonized by functions germane to that order. The eukaryote genome became a great epigenetic machine. Sequences of different levels of nucleotide plurality are briefly discussed from the point of view of their functional relevance. By their activities as both transcribed genes and cis-acting repeats, SINEs and LINEs are the principal link between genetic and epigenetic processes. SINEs can act as local repeats to produce position effect variegation (PEV) in a nearby gene. PEV may thus represent a general method of overall transcriptional regulation at the level of cell collectivities. When tracking the scale dependence of nucleotide function, one finds the 100 kb order of nucleotide plurality to provide epigenetically the basis at once for PEV, imprinting, and cell determination, with sectorial repressibility a trait common to the three. In sectorial repressibility, introns may play a structural role favoring the stability of higher-order chromatin structures. At that level of nucleotide involvement, nonconserved nonhomologous nonprotein-coding sequences may often play the same structural roles. In addition, genomic distance per se--and, therefore, the mass of intervening nucleotides--can have functional effects. Distances between enhancers and promoters need to be probed in this respect. At the 1,000 kb level of nucleotide function, attention is focused on the formation of centromeres. It is one of the levels of nucleotide plurality per function where specificity in the generation of DNA/protein complexes seems to depend more upon the structural fit among factors than upon the DNA sequence. This circumstance may explain in part the prevailing difficulty in recognizing the functional nature of sequences among non-protein-coding nucleotide arrays and the propensity among investigators to tag the majority of DNA sequences in higher organisms as functionally meaningless. Noncoding DNA often may not be 'selected' as an appropriate niche for a certain function, but be 'elected' in that capacity by a group of factors, as a preexisting sequence that is only now called upon to serve. Much of the non-protein-coding DNA may thus be only conditionally functional and in fact may never be elected to functions at a high level of nucleotide plurality. Eukaryotes are composites, at different levels of this plurality, of the functional and the nonfunctional, as well as of the conditionally functional and the outright functional. Thus, a sequence that is nonfunctional at one level of nucleotide plurality may participate in a functional sequence at a more inclusive level. In the end, every nucleotide is at least infinitesimally functional if, for metabolic and developmental reasons, the chromatin mass as such becomes a selectable entity. Given the scale dependence of nucleotide function, large amounts of 'junk DNA', contrary to common belief, must be assumed to contribute to the complexity of gene interaction systems and of organisms.