Hypothesis for the influence of fixatives on the chromatin patterns of interphase nuclei, based on shrinkage and retraction of nuclear and perinuclear structures

Abstract

Nuclear chromatin patterns are used to distinguish normal and abnormal cells in histopathology and cytopathology. However, many chromatin pattern features are affected by aspects of tissue processing, especially fixation. Major effects of aldehyde and/or ethanol fixation on nuclei in the living state include shrinkage, chromatin aggregation and production of a 'chromatinic rim'. The mechanisms of these effects are poorly understood. In the past, possible mechanisms of fixation-induced morphological change have been considered only in terms of the theoretical model of the nucleus, which involves only a random tangle of partly unfolded chromosomes contained within the nuclear membrane. Such a model provides no basis for chromatin to be associated with the nuclear envelope, and hence no obvious clue to a mechanism for the formation of the 'chromatinic rim' in fixed nuclei. In recent years, two new models of nuclear structure have been described. The nuclear membrane-bound, chromosomal-domain model is based on the discoveries of chromatin-nuclear membrane attachments and of the localisation of the chromatin of each chromosome within discrete, exclusive parts of the nucleus (the 'domain' of each partly unfolded chromosome). The nuclear matrix/scaffold model is based on the discovery of relatively insoluble proteins in nuclei, which it suggests forms a 'matrix' and modulates gene expression by affecting transcription of DNA. Here, a hypothesis for fixation-associated chromatin pattern formation based mainly on the first model but partially relying on the second, is presented. The hypothesis offers explanations of the variations of appearance of nuclei according to fixation (especially air-drying versus wet-fixation with formaldehyde, glutaraldehyde or ethanol); the appearances of the nuclei of more metabolically active versus less metabolically active cells of the same type; the appearances of nuclei after fixation with osmium tetroxide; and of the marked central clearing ('egg-shell' or 'orphan Annie' appearance) of tumour nuclei of papillary carcinoma of the thyroid gland. A similar process may underlie the phenomenon of 'chromatin margination' seen in apoptosis. Various tests of the hypothesis, such as time-lapse confocal microscopy of living nuclei during fixation, are suggested. The significance of the theory is that it suggests that chromatin patterns could be investigated in terms of qualitative and quantitative aspects of nuclear components, and hence be related to the results of studies of the structure and function of nuclei in health and disease.