A new method for preparing cells for microscopic examination is presented in which cell mixtures are fractionated by dielectrophoretic forces and simultaneously collected into characteristic zones on slides. The method traps cells directly from the suspending medium onto the slide, reducing cell loss. Furthermore, it exploits differences in the dielectric properties of the cells, which sensitively reflect their morphology. Because different cell types are trapped in characteristic zones on the slide, the technique represents an advance over existing methods for slide preparation, such as centrifugation and smears where cells are randomly distributed. In particular, the new method should aid in the detection of rare and anomalous cell subpopulations that might otherwise go unnoticed against a high background of normal cells. As well as being suitable for traditional microscopic examination and automated slide scanning approaches, it is compatible with histochemical and immunochemical techniques, as well as emerging molecular and proteomic methods. This paper describes the rationale and design of this so-called electrosmear instrumentation and shows experimental results that verify the theory and applicability of the method with model cell lines and normal peripheral blood subpopulations.
KeywordsSlide preparation; Dielectrophoresis; Cell separation; Cell discrimination; Microelectrodes Advances in cytological slide preparation techniques have facilitated the identification and characterization of different cell types and have improved the ease and accuracy of disease diagnosis. In particular, numerous staining procedures, including chemical dye-, 1 fluorescent-, 2 enzyme-linked-, 3 immunosensitive-, 4 and specific molecular-methods 5 have been developed. While these techniques allow target cells to be differentiated from other cell types, specimen screening efficiency depends on how well the pathologist discriminates target components such as bacteria, precancerous lesions or cancerous cells from background cells. In diseases characterized by the presence of extremely small numbers of marker cells against a background of large numbers of normal cells, identifying the randomly distributed target cells is tedious, time consuming, and prone to error. Recently, a number of instruments have been introduced that automatically scan slides and attempt to identify putative marker cells for later review by a pathologist. 6 These instruments are based on machine vision through a high quality microscope and they are expensive and relatively slow.