A fractal model of chromosomes and chromosomal DNA replication.

With the aim of clarifying topological problems involved in the process of chromosomal DNA replication, a fractal model of chromosomes was built based on the assumption that a part of a chromosome, e.g. a radial loop, is similar in shape to a whole chromosome and each radial loop represents structures in the lower-order organization (an assumption of self-similarity). Several other assumptions used include (i) one continuous DNA fiber makes a whole chromosome (a unineme hypothesis), (ii) in situ DNA exists in the form of a double duplex or a tetraplex which is made of two duplex DNAs, although a duplex DNA may appear transiently in S-phase (multi-strandedness hypothesis) and (iii) torsional stress on a DNA fiber causes the fiber to supercoil and thus stabilizes chromosome structure (torque-based stabilization). This model allowed to calculate of a fractal dimension of a representative metaphase chromosome (e.g. d = 2.34), to predict the mode of replication of double duplex and to furnish a topological basis for the decondensation unit hypothesis. It must also be admitted that all the arguments in this report except for the possible existence of split telomeres hold true without assuming a tetraplex organization of chromosomes. Implications of this model was discussed and the importance of the fractal dimension as a measure of chromatin condensation stressed.