The Physical Nature of the Lipid Globules in the Living Neurones of Helix aspersa as indicated by Measurements of Refractive Index

It has recently been shown by Chou (1957) that there are three different kinds of lipid globules in the neurones of the snail, Helix aspersa, and that they are clearly distinguishable from each other by histochemical tests. The kinds are as follows:

Rather irregularly shaped yellow globules (y in fig. 1, A), 1-2 p in diameter, found throughout the cytoplasm of the cell-body but particularly concentrated in the regions of the axon hillock. These are chemically complex bodies containing mixed lipids, carbohydrate, and protein.

Spherical colourless globules (c in fig. 1, A) about o-8 to 1 -4 p in diameter, found throughout the cytoplasm of the cell-body, but particularly abundant in the proximal part of the axon where the other two types of globule are usually absent. These probably contain only triglycerides, because they react negatively to all histochemical tests except the Sudans, which suggests that other substances are unlikely to be present. There is unfortunately no positive histochemical test for triglycerides.

(3) Nearly spherical ‘blue’ globules (b in fig. 1, A) about 1 to 2 p. in diameter, uniformly distributed throughout the cytoplasm of the cell-body. These are actually colourless in the living cell but are distinguishable from the colourless globules described above by the fact that they are stained blue in life by methylene blue, brilliant cresyl blue, and Nile blue. They are found to contain phospholipid.

Under the phase-contrast microscope, the yellow and colourless globules in the living cells immersed in saline presented an appearance typical of rather refractile objects, such as might be expected if they consisted of concentrated proteins and lipids, or pure lipids. The ‘blue’ globules, on the contrary, seemed strikingly less refractile, and this suggested that the phospholipid they contained might be associated with water.

It was thought desirable to make actual measurements of the refractive indices of each of the kinds of globules to see if this were true.

The method used for measuring the refractive indices of the globules was the same as that developed by Ross (1954) for measuring the refractive indices of the cytoplasmic inclusions in living cells. An interference microscope was used for measuring the retardation in phase of the light passing through the centre of each globule compared with the light passing through an adjacent region. The diameter of each globule was also measured and its refractive index was derived from these two measurements. It was, however, first necessary to immerse the cell itself in a mounting medium that had the same refractive index as the cytoplasm in order to ensure that the phase-change measurements made on the globules were unaffected by phase changes in the surrounding cytoplasm.

Suitable mounting media were made from a 20% solution of Armour’s bovine plasma albumin, fraction V in 0-5% saline, which has a tonicity approximately equal to that of a 0-7% saline solution, which is isotonic with snails’ blood (Ross, 1952). A series of dilutions of this protein solution were then made by adding 0-7% saline, and their refractive indices were measured with a Bellingham and Stanley pocket refractometer. This instrument has a built-in filter giving maximum transmission at 589 mp. It is sufficiently accurate to give figures of refractive index that are reliable to the third place of decimals. This third place is unlikely to have been affected by variation in roomtemperature during the course of the observations recorded in this paper.

Living neurones were obtained from the cerebral ganglion and ventral ganglion-mass, removed from the freshly decapitated snail. Pieces of this tissue were teased in each of the protein solutions and the cell suspensions were examined under a Smith interference microscope (manufactured by Messrs. Charles Baker of Holborn). It was found that in the protein solutions with refractive indices between 1’357 and 1’364, some neurones could be seen in which the cytoplasm appeared to be of exactly the same interference colour as the background, while, if nearly monochromatic light were used, the cytoplasm matched the background intensity at all settings of the analyser (fig. 1, c, D, and E). This meant that in these neurones, the cytoplasm had exactly the same refractive index as the mounting medium (or had a refractive index within o-ooi of this value). Phase-change measurements could therefore be made on the inclusions in those cells without it being necessary to take into account the refractive index and thickness of the surrounding cytoplasm.

Phase-change measurements (in the direction of the optical axis of the microscope) were made with the interference microscope by the extinction point method in green light with a measured mean wavelength of 542 m/q which was obtained by using a tungsten ‘pointolite’ lamp and an Ilford 807 (mercury green) gelatine filter. The analyser of the microscope was first turned until the background and cytoplasm appeared maximally dark (fig. 1, D) and then turned again until the centre of the globule appeared maximally dark (fig. i, E). With this material, phase-change measurements could thus be made accurately to the nearest 6° (a sixtieth of a wavelength).

The diameter of the globules (in the direction at right angles to the optical axis of the microscope) was estimated to the nearest 0-2 p by means of an eyepiece micrometer scale. With the 2 mm (‘double-focus’) objective used, however, such measurements are only reliable to about the nearest 0-4 p, and this was the limiting factor in determining the accuracy of the refractive index measurements. A rather more accurate estimate was possible in the case of the colourless globules because those in the axon were often found to be arranged in straight rows of up to a dozen or so globules, apparently of identical size, so close together that they appeared to touch each other (fig. 1, B). In these cases, the maximum diameter of an individual globule can be accurately determined by measuring the overall length of a row (to the same degree of accuracy as it is possible to measure an individual globule) and dividing by the number of globules. It was, of course, possible that the globules did not quite touch each other and were therefore rather less in diameter than they appeared, but their upper size-limit could be clearly defined.

where ϕ = the measured retardation in phase of the light passing through the centre of the globule (expressed as an angle), d = the diameter of the globule estimated by eyepiece micrometer to the nearest 0 ·2 μ., λ = the mean wavelength of the light used (0 ·542 μ), and m = the refractive index of the protein mounting medium (and of the cytoplasm of the neurone in which the globules were measured). Since the two sets of measurements from which the refractive indices were derived were made in directions at right angles to each other, it was assumed that the globules were perfectly spherical and this was nearly true even in the case of somewhat irregularly shaped yellow globules.

Before they were measured, the globules were individually identified in white light with the polarizer of the interference microscope in the ‘out’ position, and the microscope stage was moved until they were in an easily recognizable position in relation to the eyepiece micrometer scale for subsequent measurements. The yellow globules and the colourless globules in the axon were quite easy to recognize, but for the certain identification of the ‘blue’ globules it was necessary to use cells which had been supravitally stained with Nile blue before being mounted in the protein solution. With the green filter used, it was extremely unlikely that the blue-green coloration of these globules affected the phase-change measurements because they were invisible when viewed with a green filter and no polarizer.

A total of 10 yellow globules were measured in two different cells with matched cytoplasm mounted in a protein medium with a refractive index of 1-358; and 5 colourless globules were measured in the axon region in a single matched cell in a similar mounting medium. Five ‘blue’ globules were measured in two other matched cells in another preparation where the refractive index of the mounting medium was 1-361.

Table 1 shows the measurements of the phase-change and diameter made on each globule and the refractive indices calculated from these. The final column of the table shows the upper and lower limits of the refractive index of each globule, on the assumption of an error of ± 0 · 2 in the diameter measurements. It is very unlikely that the refractive indices of any of the globules lay outside these limits, and indeed, these extreme errors themselves are most unlikely. It is probable that the refractive indices of the colourless globules were not as low as the lower limit shown because those that were measured were all either arranged in rows in the manner described above or.

were indistinguishable in diameter from those that were in rows. These values are therefore placed in brackets.

From table 1 it will be seen that the colourless globules have a mean refractive index of 1*491. This value is closely comparable with the refractive indices, at room temperature, of many fats and oils composed mainly of triglycerides, e.g. cotton seed oil (1-475), or olive oil (1-466). Indeed, at room temperature, nearly all fats and oils of vegetable or animal origin have refractive indices between 1-46 and 1-48 (Hodgman, 1945), so that the values are, if anything, rather higher than one might expect.

The refractive indices of the yellow globules are rather more variable, as might be expected in view of their rather complex chemical composition, but these values also are those that might reasonably be expected if these globules consist mainly of lipid material or rather concentrated protein. Their mean refractive index was 1-479, which was only a very little lower than that of the colourless globules.

The ‘blue’ globules on the contrary had very much lower refractive indices with a range which, even allowing for the maximum error in measurement, does not overlap those of the other two kinds of globule. Their mean refractive index was 1-416. This is lower than that of any pure lipid including phospholipids, although histochemical tests show the presence of no substance other than phospholipids. The latter, however, differ in one important respect from triglycerides in that they have a side chain in their molecule (the phosphoric acid / choline radicle) that is hydrophil. This means that, unlike triglycerides, they will associate with water molecules, and the low refractive index measurements suggest that in addition to phospholipid, the ‘blue’ globules probably contain quite appreciable amounts of water in an association of this kind.

Schmidt in 1939 used polarized light to investigate living cell structure and concluded that phospholipid and similar molecules tended to form bimolecular layers with the hydrophil chain orientated towards an aqueous phase (a in fig. 2). In spherical droplets containing phospholipid, he found a distinct ‘polarization cross’ which was absent in similar droplets composed of triglyceride. This indicated that while the triglyceride molecules were probably disorientated except at the surface of the droplet (b in fig. 2), the phospholipid molecules appeared to be radially orientated throughout the droplet from its surface to centre (c in fig. 2). This led him to suggest that such droplets probably had a series of concentric shells of water in between bimolecular layers of phospholipid molecules (c in fig. 2).

Our present results are in agreement with Schmidt’s hypothesis in that they strongly suggest that the globules containing phospholipid in the neurones of H. aspera must also contain a considerable amount of water. They also indicate that there may be other useful possibilities in applying these refractometric techniques in conjunction with histochemical investigations for determining the physico-chemical nature of cell constituents.

We are very much indebted to Dr. J. R. Baker for suggesting this investigation, which was carried out in his laboratory at Oxford, and for his invaluable advice and encouragement. The interference microscope for the work was provided out of a grant to one of us (K. F. A. R.) from the London University Central Research Fund; J. T. Y. C. was on an Inter-University Council Fellowship through the Carnegie Corporation of New York and on leave from the Department of Zoology, University of Hong Kong.