Characterization of topotecan-mediated redistribution of DNA topoisomerase I by digital imaging microscopy

Topographical image measures have been used to characterize the subnuclear distribution of DNA topoisomerase I in human tumor cell lines. This topographical analysis allowed a mathematical description of staining patterns to be produced that did not depend on subjective grading. The redistribution of topoisomerase I in response to increasing concentrations of topotecan was then monitored by this method. The cell lines were stained for topoisomerase I by indirect immunofluorescence methods. Digital imaging microscopy and image analysis were used to extract the nucleus from each cell, and nine parameters describing the topography of the distribution of topoisomerase I within the nucleus were computed for each. Use of multivariate analysis of variance enabled this nine-parameter set to be reduced to a single canonical variable, representing 60-90% of the observed internuclear variance. Plotting the canonical variable vs drug concentration resulted in dose-response curves that could be fitted well by a simple Emax model. From these curve fits, EC50 and Emax values for drug-induced redistribution of topoisomerase I were determined. Our results indicate that neither the maximum extent of topoisomerase I redistribution (Emax) nor the EC50 for drug-induced redistribution correlated well with the growth inhibition produced by continuous exposure to topotecan in these cell lines. However, the EC50 determined for the 1-h high-concentration exposure did reflect the growth inhibition produced in cells exposed to the drug for 1 h. The methodology described may also be generally applied to any antigen of interest.