Office of Biological and Environmental Research

DOE Low Dose Radiation Program Workshop III

Abstract

Title: Frequencies of Radiation-Induced Chromosome Interchanges and Randomness of Chromosome Territory Locations Relative to One Another.

Authors: RK Sachs,§ MN Cornforth,‡ KM Greulich-Bode,* L Hlatky, and DJ Brenner||

Institutions: §Department of Mathematics, University of California,
‡University of Texas Medical Branch,
*Department of Skin Carcinogenesis, German Cancer Research Center
DFCI, Harvard Medical School,
||Center for Radiological Research, Columbia University.

Leukemogenesis, and perhaps carcinogenesis in general, often involves specific chromosome translocations. Radiation-induced chromosome translocation frequencies are strongly influenced by how close participating chromosomes are to one another in an interphase cell nucleus. We sought to determine whether chromosomes in human peripheral blood lymphocytes are randomly situated with respect to one another (and therefore undergo radiation-induced interchanges at frequencies determined by their DNA content), as opposed to particular chromosomes being preferentially close together (and therefore having an enhanced frequency of interchanges). The analysis of 24-color combinatorial whole-chromosome painting (mFISH) data from laboratories in Galveston and Munich allowed us to determine whether any specific color pairs are biased toward extra interchanges after gamma-rays induce spatially random damage in the genome. The approach considered all (24×23/2)=276 pairs, many more than previous in vitro investigations. It had good statistical power for the detection of small to medium levels of non-randomness, and we concluded that there are no strong spatial correlations between specific chromosomes. Interchange frequencies are mainly dependent on DNA content, without major biases by chromosome-to-chromosome spatial correlations. Weak spatial correlations, or strong spatial correlations involving only comparatively very small portions of chromosomes (such as specific gene sites), are not inconsistent with the data, but the dominant whole-chromosome pattern is clearly one of randomness in the locations of territories relative to one another.

We further investigated whether the interchange radiosensitivity was linearly proportional to DNA content. In agreement with a number of other recent investigations, we found that large chromosomes tended to participate less frequently in exchanges than would be expected if sensitivity were strictly linearly proportional to DNA content, while the opposite was true of small chromosomes. This second result supports models of interphase nuclear structure where chromosome territories have comparatively little overlap with one another, with chromosome interchanges involving mainly chromatin near the outside of each territory.

» Return to top