Increased nondisjunction of chromosome 21 with age in human peripheral lymphocytes

Fluorescence in situ hybridization (FISH) on binucleated cells with chromosome-specific DNA probes provides a convenient way to visualize reciprocal segregation patterns in daughter nuclei, and overcomes most problems related to the artefactual loss or gain of chromosomes that flaw chromosome preparations. In this study, FISH was employed to evaluate age- and sex-effects on spontaneous malsegregation, nondisjunction and loss of chromosome 21 in human lymphocytes after the first division in culture. A total of 68 healthy nonsmokers and nondrinkers of alcohol (37 males and 31 females) were grouped by age as Group I (0-10 years), Group II (20-30 years), Group III (40-50 years) and Group IV (60-70 years), with at least seven subjects per group and sex. FISH with a pericentric chromosome 21 specific DNA probe was carried out on binucleated lymphocytes, cytokinesis-blocked by cytochalasin B (6 microg/ml for 26 h) at 44 h after initiation of cultures. Linear regression analyses demonstrated a significant age-related increase in the frequency of micronuclei without chromosome 21 (MN-21)(r=0.73, p<0.001 in females; r=0.69, p<0.001 in males) in all binucleated cells, with a steeper slope in females (0.1758) than in males (0. 1241). Analysis using the 2x2 chi-square (chi(2)) test on the frequencies of MN-21 showed significant age-related differences in both males and females, except males in Group III and Group IV (p>0. 05). A significant sex-related difference was found only in subjects over 60 years (p<0.05), with females having more MN-21 (12.57 per thousand vs. 8.43 per thousand) than males. Loss of chromosome 21, occurring at mean levels of 0.38 per thousand in all binucleated cells and 0.24 per thousand in binucleated cells containing four FISH signals, was shown not to be age- or sex-related. A positive age-related increase in nondisjunction of chromosome 21 was shown in males (r=0.50, p<0.01), females (r=0.61, p<0.001) and all subjects (r=0.55, p<0.001) by linear regression analysis. An age effect was found only between children and adults (p<0.01 for females, p<0.05