Evaluation of a preparative method for X-ray microanalysis of soft tissues

We have adapted the cryopreparative methods designed for the radioautography of diffusible substances (Stumpf and Roth, 1966, 1967) to produce freeze-dried sections of soft tissues for electron probe microanalysis. This report concerns the evaluation of these methods for use with X-ray microanalysis with particular reference to: (1) the preservation of cellular morphology, (2) the introduction of structural artefact by ice crystal formation, (3) the preservation of natural elemental distributions under experimental conditions where known quantities of diffusible elements are present, (4) the effects of ice crystal formation upon possible artefactual elemental redistributions, (5) the effect of section thickness on elemental quantitation, and (6) the effect of tissue excision on element translocation. Freeze-dried sections were prepared from mouse pancreas and from 20% (w/v) solutions of bovine serum albumin (BSA) and gelatin containing known amounts of inorganic salts and were analysed in a scanning electron microscope fitted with energy dispersive X-ray detecting equipment. With the preparative methods used, the morphology of pancreatic acinar cells was well preserved. Acinar cell boundaries, nuclear boundaries, chromatin, nucleoli, ergastoplasm and zymogen granules were readily discernable. Ice crystals were present within sections of BSA, but 70% had a cut surface area < 1μm² and over 90% had a cut surface area of < 2μm². Characteristic peak-to-continuum ratios of elements in the BSA sections remained constant over an order of magnitude change in magnification Elemental redistributions were not detected until the magnification was such that the analysed area fell totally within the confines of a single ice crystal. No differences in elemental peak-to-continuum values were obtained between 2 and 4 μm sections of the gelatin-salt solution. Tissue excision did not cause element translocations when compared to tissues frozen in situ. We conclude that this method is valid for preparing tissues for microanalysis under our conditions (analysis of nuclear, cytoplasmic, and secretory compartments) and is limiting only when analyses are conducted at very high magnifications.