Why Are Multiple Regulators Required for Transcription of Each Gene?

Introduction

Reports of various regulatory networks for gene expression are now accumulating in mammals, demonstrating that branched complex networks are present. It has been suggested that mammals have a large number of DNA-binding proteins for gene expression. In contrast, one regulatory protein is generally involved in individual gene expression, and such expression is often mediated by only RNA polymerase without other DNA-binding proteins in bacteria. The basic question is why multiple DNA-binding proteins are required for the regulation of single gene expression in mammals. One explanation may be that mammals have more complex functions than bacteria. However, the functions operating in a single mammalian cell are less complex than those in a single bacterial cell, although the functions in the whole body are highly complex in total. It is indispensable for our improved understanding of gene-regulating networks to clarify why multiple regulators are required for each gene expression in mammalian cells.

The human genome has been almost completely analyzed (International Human Genome Sequencing Consortium, 2004), and numerous mechanisms for transcriptional regulation have been proposed (Abnizova & Gilks, 2006; Barrett, Fletcher, & Wilton, 2012). In the first step of transcription, various regulatory proteins are bound to their specific sites with a nucleotide sequence called a “consensus sequence”. Experimental data have demonstrated that a given regulatory protein recognizes its consensus sequence to regulate transcription. The bioinformatics of mammalian genomes suggest numerous sites for the binding of regulatory proteins which cooperatively regulate gene expression (Abnizova, Subhankulova, & Gilks, 2007; Barrett, Fletcher, & Wilton, 2012).

The human genome has approximately 3 billion base pairs, and many consensus sequences reported to date have less than 16 base pairs (Barrett, Fletcher, & Wilton, 2012). Theoretically, a sequence consisting of less than 16 base pairs is present randomly in the human genome if its sequence is close to random. In contrast, if the human genome is not random, consensus sequences will be conserved only at sites required for gene expression. It is therefore indispensable to clarify whether or not the human genome is close to random.

To clarify this, the following were theoretically assessed using a computer in the present study: (1) whether or not sequences consisting of 11 to 16 base pairs created randomly are present in the human genome; (2) whether or not consensus sequences reported to date can be found in an artificial genome created by random numbers using a computer. The results suggested that the human genome has areas whose sequences are close to random and the areas account for more than 2.7x10^8 (4^14) base pairs.

In molecular biology, theoretical examinations have been few to date. Theoretical verification of biological phenomena is indispensable for our improved understanding of the regulation of gene expression to complement arguments based on experimental studies. In this chapter, we first discuss why multiple regulators are required for the expression of a single gene, then, we focus on the benefits of having multiple regulators, and finally, on the question why mammalian cells have complex systems for signal pathways. The discussion is based on theoretical results obtained with a computer simulation.