The Distribution Of Cholinesterase in Cholinergic Neurons Demonstrated With the Electron Microscope

0 Department of Anatomy, University of Cambridge SUMMARY The thiocholine technique for cholinesterase has been successfully adapted for the demonstration of enzyme distribution with the electron microscope. After fixation in buffered glutaraldehyde, thin slices oftissuewere taken through a histochemical procedure designed to minimize diffusion artifacts and cytological damage; appropriate areas were dissected out, fixed with osmium tetroxide and embedded in Araldite. This technique has been used to study the distribution of enzyme in and around known cholinergic neurons in the rat. In diaphragm muscle the intense staining of the motor end plates is confined to the actual synaptic clefts. The distribution of cytoplasmic staining was similar in all three types of cholinergic neurons studied (ventral horn cells from the cervical cord, cells from the dorsal motor nucleus of the vagus, and cells from the hypoglossal nucleus). There was intense staining of the space contained within the individual bilaminar sheets of the rough endoplasmic reticulum and occasionally in areas of the nuclear envelope. Mitochondria, lysosomes, the smooth endoplasmic reticulum and most of the plasma membrane were unstained. In cholinergic nerve fibres staining was particularly intense at the axonal membrane and absent from the myelin sheath. In several regions of the brain known to receive an afferent cholinergic innervation many of the identifiable synaptic areas were heavily stained. Staining here was usually present round most of the presynaptic process, spreading over part of the postsynaptic process and often penetrating into the actual synaptic space. The mitochondria, microvesicles and the general cytoplasm of the synaptic processes were all unstained. In some areas, such as the hippocampus, there was intense membrane staining offinecholinergic neuropil. The enzyme specificity of the technique is not in doubt and evidence is adduced for the view that diffusion artifacts are small. The close correlation between the electron-microscopic results and evidence from physiological and biochemical investigations is discussed. THE DISTRIBUTION OF CHOLINESTERASE Several attempts have been made to use histochemical techniques for cholinesterases at the electron-microscope level. The results obtained by Lehrer & Ornstein (1959), using a simultaneous coupling azo-dye technique, were not encouraging, largely because the reaction product was not sufficiently electron-dense. More satisfactory results have been obtained with the thiolacetic acid technique: the reaction product is very electron-dense and sufficiently well localized for at least some purposes disadvantage of any procedure based on thiolacetic acid, however, is the lack of enzyme We therefore attempted to adapt the highly specific thiocholine technique for electron microscopy. A brief outline of our method has been published as a preliminary communication (Lewis & Shute, 1964); and a method based on a somewhat different principle has been published by Karnovsky (1964). Our ultimate aim was to study cholinesterase distribution in many areas of the brain where cholinergic mechanisms are implicated, following on previous work at the light-microscope level (Shute & Lewis, 1963, 1965). But first it was essential to test the technique on structures known, unequivocally, to be concerned in cholinergic transmission. The procedure finally devised gives very good enzyme localization and adequate, though not perfect, preservation of fine cytological detail. The precise procedure used is given in some detail here because it should be directly applicable to many tissues besides brain and muscle. MATERIALS AND METHOD Osmium tetroxide is not suitable as a primary fixative because it is so toxic to enzymes. For much of our material glutaraldehyde has been satisfactory, but it was usually necessary to fix the whole animal by perfusion before attempting to dissect out individual pieces of tissue. In some regions of the brain (for example, hippocampus) glutaraldehyde alone did not give adequate preservation of cytological detail; such tissues were transferred after preliminary fixation in glutaraldehyde to an equivalent formalin solution for a few hours before proceeding to the next stage (Shute & Lewis, 1966). Aldehyde-fixed tissue sectioned on the freezing microtome did not provide material satisfactory for electron microscopy. Although many of the cytoplasmic organelles were well preserved there were numerous scattered lacunae which were obviously artifacts, presumably caused by ice-crystal formation. Such artifacts are quite unacceptable in brain, where intercellular relationships are vitally significant. It was therefore necessary to cut thin freehand sections from the face of a chilled, but not frozen, block of aldehyde-fixed brain. In practice it proved difficult to cut sections thinner than about 150 /i without damaging them mechanically; so, in general, sections about 200-250 ft thick were cut and incubated for electron-microscopic study. A simple device facilitates the cutting of these sections. Two coverslips are mounted with balsam on an ordinary 3 in. x i\ in. glass slide with a gap of, say, 1 cm between them. A clean-cut face of the block of tissue is then firmly held against the slide and a sharp razor blade with its ends held flat against the coverslips is drawn across the block. The thickness of the coverslips determines the thickness of the tissue slice obtained. If the normal histochemical procedure, as applied routinely to frozen sections, is used for electron microscopy the results are disappointing. The deposit of copper sulphide is so electron-dense that ordinary cytological detail is obscured and there is clear evidence of diffusion artifacts. In an attempt to improve the electron-microscopic picture various modifications of the standard procedure were tried, the choice of modifications being based on the results of an earlier study with the light microscope (Lewis, 1961). Some of the modifications were aimed at improving the efficiency of the capture reaction, that is, the reaction of enzymically released thiocholine with cupric ions to give the initial precipitate, since it is at this stage that most of the diffusion occurs. Other modifications were aimed at preserving the fine cytology: thus all solutions were buffered and made approximately isotonic, and calcium ions were incorporated into all except the incubation medium. The perfusion fluid was used at room temperature, but all the other solutions were used at about 4 C. Albino rats were killed with an overdose of ether and immediately perfused through the heart. The composition of the perfusion fluid, which was made up and filtered immediately before use, was as follows: 16 ml 25% glutaraldehyde, 50 ml 0-2 M sodium cacodylate, 2 ml 0-2 M calcium acetate, made up to 200 ml with distilled water. The final pH was adjusted if necessary to y-o-J'S The tissues required were (...truncated)