Abstract
A strategy for measuring Neisseria gonorrhoeae attachment and invasion in the human Fallopian tube organ culture (FTOC) model via computerized image analysis (CIA) combined with "digital" confocal microscopy (DCM) was tested. DCM on serial image stacks of fluorescent latex beads reduced out-of-focus light propagation in the Z-axis (p < 0.005) and improved the shape factor of lateral three-dimensional reconstructions of the beads (p < 0.001). Sections of tissue infected for 44 hr with piliated, Opa+ gonococci were stained with fluorescein-labeled monoclonal anti-gonococcal antibodies, rhodamine-labeled phalloidin, and Hoechst 33342. Serial images collected at identical focal planes for each fluorochrome were subjected to DCM. Epithelial cytoplasmic regions of interest defined by rhodamine-stained actin were superimposed on the corresponding fluorescein-stained and Hoechst-stained images. Fluorescent objects defined by gray-scale threshold were measured by computerized image analysis using different border treatments to differentiate attached from intracellular gonococci or count cell nuclei. Compared with raw images, measurement of DCM images was less dependent on threshold choice (p < 0.05). DCM augments conventional microscopy in removing out-of-focus light from fluorescent images, in reconstructing three-dimensional images, and in quantitatively differentiating extracellular from intracellular gonococci in a natural target tissue.