Glutaraldehyde and urea are polymerizable, even in the presence of substantial amounts of water, to produce an “aminoplastic” that can serve as a tissue embedment for ultrathin sectioning. The highly polar nature of the resin mixture permits the retention of lipid structures in situ. Also, since dehydration is not completed until after polymerization, the use of this embedding medium permits superior preservation of highly hydrated carbohydrated gels which otherwise are apt to collapse. An insoluble glutaraldehyde—urea polymer is produced only by acid catalysis (pH ≊ 4.0), easily achieved with oxalic acid. Two successful protocols are offered. One demands pH monitoring to initiate polymerization, the other involves the blending of two independently stable mixtures. Experiences with other related potential monomers are described. Polymerization is allowed to occur with the specimens in small, individual drops of the resin mixture. The final plastic is hard, and so brittle that rough trimming almost has to be done by grinding. It is quite hydrophilic, necessitating a relatively high cutting speed and a carefully controlled trough meniscus. Glass knives are much to be preferred. Representative tissues, stained in various ways, illustrate some of the potentialities, as well as the limitations, of the method. Sections exposed to osmium tetroxide vapors demonstrate completely stained cytomembranes. This is regarded as indicative of the retention of lipids in the embedment. The contents of lipid droplets are similarly stainable. Phosphotungstic and silicotungstic acids at a pH ≅ 4.5 demonstrate carbohydrate moieties, particularly the glycocalyx, and some of its specializations including the material of synaptic and desmosomal clefts (the stainable material drops out in tight junctions). Mucous droplets and layers are stained as well, as is frequently at least a part of the Golgi complex in secretory cells. Glycogen is preserved and stained. Spaces occupied by membrane lipids are then seen in negative contrast. After uranyl acetate and lead citrate staining, the hydrophobic layers of cytomembranes are not stained either, but are outlined by neighboring structures. These substances do stain proteins generally, but usually without good differentiation or contrast which is a decided limitation. The preservation of presumably normal spatial relations in lipid systems, including cytomembranes, is a special virtue of the technique. Also, normal relationships appear to be preserved in unparalleled fashion in systems dependent in life upon highly hydrated carbohydrate gels. This particularly is the case where specific gaps are maintained in contact relations between cells, in specializations of the extracellular compartment in such diverse places as kidney epithelium, neuropil and the intraperiod “line” of myelin.
Read full abstract