AUGUST, 1950 Vol. 75, No. 893 THE ANALYST PROCEEDINGS OF THE SOCIETY OF PUBLIC ANALYSTS AND OTHER ANALYTICAL CHEMISTS Fluorescence Microscopy as an Aid to Food and Drug Analysis BY J. KING AND R. E. WESTON SYNOPSIS-A simple method has been devised for using fluorescence microscopy in food and drug analysis, the wavelength of 3650 A. being selected from the light given by a 250-watt compact-source high-pressure mercury-vapour lamp. The advantages over normal microscopy are discussed, and illustrations are given from the authors’ experiences of the application of the method, with special reference to the detection of minute quantities of aneurine added in the form of a solution in fine droplets sprayed on to wheat flour for the purpose of “enrichment.” THE principle of fluorescence microscopy has been widely used in biology, and an excellent summary of the subject up to 1940 has been given by E1linger.l A more recent review by White2 and a new edition of Pringsheim’s book on fluorescence3 give useful references up to -1949.Unfortunately the apparatus used is expensive, difficult to procure and requires a skilled technique. It has been found, however, that fluorescence is frequently generated when light in the near ultra-violet region impinges on materials of plant or animal origin. The recent introduction of compact-source high-pressure mercury-vapour lamps induced us to investigate their potentialities in fluorescence microscopy as applied to food and drug analysis. The radiant power output per unit area of source in this type of lamp is extemely high; this enables an intense beam of radiation to be focussed on to the object to be examined whether by incident or transmitted light.The wavelength of 3650 A. has been selected for the following reasons. Extremely intense emission occurs in the near ultra-violet at this wavelength and this radiation can readily be isolated by filtering through Wood’s glass which transmits about 65 per cent. of the light, or through a Wratten filter No. 18A. Light of this wavelength is transmitted freely by most optical glass and an ordinary microscope can be used without quartz condensers, slides, etc. It has the further advantage of being innocuous to the eyes, and so obviates the special precautions necessary with lower wavelengths. When examining the specimens by transmitted light it is an advantage, although not essential, to replace the normal silvered mirror by one made of “super-purity” aluminium, polished and anodised. Such a mirror reflects about 80 per cent.of the incident light of 3650 A. wavelength. When dark-ground illumination is required, an Abb6 condenser can be used and at high magnifications (using objectives of shorter working distance than 4 mm.) this is necessary, as the angle of incidence is of necessity so small under these con- ditions that severe distortion of the image occurs. It is an advantage when using objectives of the order of 4 mm. to have the highest numerical aperture possible. If it is desired to use the shorter wavelengths of the mercury-vapour lamp, the condenser system described by 397398 KING AND WESTON : FLUORESCENCE MICROSCOPY -4s [Vol.75 Smiles4 and used with success by Barnard and Welch5 is recommended; it is then necessary to use transmitted light. Recently, a type of glass that transmits light of shorter wavelength has become commercially available and from this lenses and prisms can be made. This should enable the optical system formerly made from quartz to be made much more cheaply, and greatly extend the use of fluorescence microscopy at the shorter wavelengths. The essential details of the system adopted by us are shown in Fig. 1, in which A is a choke made for use with this type of lamp, furnished with tappings to match accurately the voltage of the A.C. supply when under load (the lamp is unsuitable for D.C.). The lamp housing should be well ventilated, in view of the heat to be dissipated, and constructed so as to prevent stray beams of ultra-violet light of’ various wavelengths escaping and affecting the eyes of the operator.The lenses at B are of any type of optical glass that will readily transmit light of wavelength 3650 A., and are spaced to enable the light to be focussed readily on to the stage of the microscope. It is an advantage to have these mounted on an optical bench, as this enables them to be adjusted so that when transmitted light is used a parallel beam may be reflected through the substage condenser. If light of shorter wavelength is required, these lenses must be of quartz or the special glass referred to above, as must also be the slip covering the preparation to be examined; C is a sheet of Wood’s glass approximately 3 mm.in thickness, or a Wratten filter No. 18A. For some purposes it is essential to utilise the maximum intensity of ultra-violet light, and this is made possible by careful focussing on to a small sheet of paper impregnated with a dilute solution of acidified quinine sulphate and mounted on a glass slide in the usual position on the microscope stage. Unfortunately, the image is in the form of an elongated ellipse, which makes even illumination of the field impossible at the highest light intensities when using low-power objectives. Even- ness of illumination is improved greatly by adopting an “out-of-focus” position, but at the expense of intensity. A compromise may be obtained by presenting the stage of the microscope at an appropriate angle to the incident light or by interposing a lens having a cylindrical face.The microscope slides and cover-slips rnay be of ordinary glass, but they should be examined for mechanical inclusions by observing them under ultra-violet light and rejecting any showing fluorescent patches or spots. These spots as well as any extraneous matter that may have settled on the top of the slide may be troublesome when working with objectives of greater power than 16 mm., and on all occasions when a hazy fluorescence is encountered it is essential to focus accurately on to the material responsible, so that its nature may be identified. The appearance of objects seen under ;t powerful beam of ultra-violet light differs markedly from that under normal lighting and the microscopist will need to build up his own experience with this medium.Objects such as leaves, powders, hairs, textiles, etc., should first be examined in the dry state by incident light under a low power objective, about 16 mm., with a x 10 eyepiece. Specimens should be reduced t o a suitable size and form for examination in a water or dilute alkali mount with superimposed cover-slip. The usual clearing agents may be employed, but this is seldom necessary except when using transmitted light. A brief study has been made of the appearance under ultra-violet light of leaves, whole and powdered, seeds, fibres, animal and vegetable hairs, powdered roots, cereals, flour and insects. Chase and Pratt6 have described the colour of the general fluorescence observed under comparatively low intensities of ultra-violet light of wavelength approximately 3650 A.of 151 powdered drugs, and of their alcoholic extracts ; but our investigation relates to the fluorescence, under intense ultra-violet light, exhibited by individual structures. Stanfill‘ has also given some account of the detection of contamination by rodents. Certain constituents of vegetable and ardmal matter exhibit marked fluorescence, e.g., aesculin, fisetin, lignin, xylin, riboflavine, lyochromes, porphyrins, vitamin A. Differential staining with fluorescent dyes such as fluorescein or acriflavine may be used when the sub- stance under examination does not exhibit a natural fluorescence, but this subject has not been pursued by us, our attention having been given entirely to naturally fluorescent substances.A fluorescent paste has also been describeda to render interstices in surfaces fluorescent. As a rule no special treatment of objects to be examined by ultra-violet light is necessary. They may be examined in the dry state or mounted in fluids such as water, dilute glycerol, 1 per cent. sodium hydroxide, or 2 N hydrocliloric acid, without clearing. The translucency of starch gelatinised by the alkaline ferricyanide mounting fluid referred to later enables ultra-violet light to penetrate readily to cellu1a.r structure in starchy powders ; if this fluoresces,August, 19501 AN AID TO FOOD AND DRUG ANALYSIS 399 its fine structure is seen much more clearly than when submitted to normal microscopy with transmitted light.The jelly holds the fragments of fibre in positions that enable the structure to be seen in perspective if a binocular microscope is used with the direct rays, III I and this greatly facilitates the examination of starchy powders such as wheat flour. The interior structure of leaves can only be seen by drying and powdering or by suitable dis- section. Many oils, the dry endosperm of cereals, etc., fluoresce so brilliantly that the visible light emitted greatly impairs the definition of the fine structure of cellular matter. It is therefore an advantage to mount in a fluid, and sometimes to remove the oil by extraction400 KING AND WESTON: FLUOR.ESCENCE MICROSCOPY AS [Vol. 75 with a suitable solvent. Hairs of either plants or animals almost invariably fluoresce under ultra-violet light.The cell walls of most plants exhibit a green fluorescence in 1 per cent, caustic soda solution and blueviolet in an acid or neutral aqueous medium. In some cases, e.g., powdered rue, the aqueous mounting mediuim, particularly if alkaline, becomes fluorescent. On the whole, as stated by Helmhol~,~ the image is better differentiated, and the detail of structure is more readily discernable, when fluorescent substances are associated with the objects under examination, and it is our experience that for many substances differentiation and recognition is easier by the ultra-violet technique than by ordinary microscopy, and certainly with far less preliminary preparation. The following is a short account of the fluorescence microscopy of some well-known foods and drugs, with their characteristic features.Hairs-Most vegetable hairs fluoresce in the dry state, but it is better to examine them in an aqueous medium which may be neutral, acid or alkaline. Wheat hairs fluoresce with a blue colour when mounted in 1 per cent. sodium hydroxide, and the structure is well defined. Rat and mouse hairs also fluoresce with a blue colour in an aqueous medium, and the structure is thrown into sharp relief. This structure can readily be seen in contaminated flour in 1 per cent. sodium hydroxide solution, but unless contamination is gross, the method cannot replace that described for hairs and insect fragments by Kent- Jones, Amos, Elias, Bradrjhaw and Thackray.lo Vegetable strmture-Many naturally occurring oils fluoresce both in the oil cells of plants and as oils. Thus the germ oils of cereals fluoresce with a violet colour, and the oil-containing cells also appear to be violet, the cell walls being green in an alkaline mounting fluid.The bast fibres of roots, stems, etc., fluoresce with a blue or blue-green colour when mounted in water or in an alkaline medium. Imects-Mounted dry, the outer surface of insects exhibits a variety of brown and red colours, the hairs standing out in sharp relief. The compound eyes fluoresce with a blue colour. Insect fragments can be recognised in powdered cereals mounted in sodium hydroxide solution, but a slight contamination could only be established after a most exhaustive examination by this method.Detection of aneurine hydrochloride added to wheat flour in the form of a solution as a f i n e spray-In view of the possibility of flours being sprayed with a solution of aneurine hydrochlo- ride for enrichment purposes, a special study of the method as applied to the detection of such flours was undertaken. Experience has shown that, while the method of Hintonll for the detection of added powdered aneurine hydrochloride is adequately diagnostic, sufficient light to be visible to the unaided eye is not obtainable when the vitamin is added as a solution of aneurine hydrochloride. If, however, the specjimens are examined by fluorescence microscopy using our technique, the violet fluorescent spots due to thiochrome formed by the action of the oxidising medium are readily apparent, even when the addition has been made in the form of extremely fine drops produced by an atomiser.The technique is as follows: the mounting fluid consists of an aqueous solution containing 0.1 per cent. of K,Fe(CN), and 1 per cent. of sodium hydroxide, with the addition of a wetting agent such as “Teepol.” The latter enables the microscope slide to be wetted easil.y, and makes the preparation of a bubble-free slide much easier. The mounting fluid is spread with a glass rod on to a microscope slide 3 inches x 19 inches (as free as possible from flaws that fluoresce). The flour is sprinkled as evenly as possible over the wetted surface by means of a small sieve (mesh about 0.7 mm. square). Only sufficient flour is used to cover amply but to avoid thick patches, which are not completely wetted by the oxidising fluid and in the dry state fluoresce brilliantly. It is essential to avoid fluorescence due to this dry flour.A slide of similar dimensions is wetted with the same fluid and carefully placed wetted side down on to the prepared sample engaging one edge first and making complete contact slowly so as to avoid the formation of air bubbles. The slides are then inverted and examined systematically, on a travelling stage, with a 2-inch objective in combination with a x 10 eyepiece. The presence of diffused round spots of violet fluorescence which cannot be sharply focussed and are not associated with recognisable cell structure are indicative of a sprayed solution of the vitamin. Care should be taken not to mistake for thiochrome, fluorescence derived from other sources such as wheat germ oil, flaws in the glass slidle, or dry material in the preparation or on the surface of the slide.A little experience is necessary before such spots can be recognised with certainty and it may be necessary to examine a number of slides, depending on the quantity added and the size of the drops, as even with flours known to have been enriched A few hairs exhibit a pale pink colour. Mites exhibit a very pale violet fluorescence.August, 19501 AN AID TO FOOD AND DRUG ANALYSIS 401 by fluid aneurine sprays, a proportion of slides exhibiting no typical spots may be encountered. Comparison with flours of known authenticity is necessary in the event of doubt. In the sam- ples of flour examined by us as little as 0.1 pg.of aneurine per g. of flour, added in the form of finely atomised droplets, has been detected. We prefer to use a binocular microscope and mechanical stage for the examination of these specimens, the slides being made from old photographic quarter-plates from which the gelatin has been removed. For rapid scanning, the slides may be manipulated by hand, resort being had to the screw mechanism only when a systematic survey of the whole of the material on such large slides is necessary. A moderate contamination by insect fragments or rodent hairs may be detected by this arrangement. The various parts of the wheat grain appear as follows when mounted in water or alkaline solutions. Wheat hairs-In water mounting the fluorescence is blue-green ; the fluorescence is much stronger in 1 per cent.sodium hydroxide. Germ-In water the fluorescence is pale blue-violet; in 1 per cent. sodium hydroxide the contents of the cell fluoresce with a stronger violet colour and the cell walls brilliant green, After de-fatting, which removes the oil, the violet colour no longer appears. CeZZuZar tissue-In water the fluorescence is pale blue; in 1 per cent. sodium hydroxide the aleurone cells walls are a brillibt green. Parts of the surface of the outer bran may be a pale brown in colour and show little distinctive green fluorescence. WhoZemeaZ-Mounted in 1 per cent. sodium hydroxide the starchy matter is invisible, but the structures enumerated above can be seen quite clearly. Occasionally a single field contains specimens of hairs (blue), internal cellular structure (cell walls brilliant green) and germ (cell walls green with violet contents), the whole exhibiting a very striking effect and giving a much clearer delineation than is possible by using ordinary microscopy.Much time is saved as no previous preparation of the specimen is necessary, the structure being as a rule more clearly defined. The authors wish to express their thanks for gifts of samples of “enriched” flour prepared by Dr. D. FV. Kent-Jones, Dr. L. George, and Dr. T. Moran and his staff of the Research Association of British Flour Millers, and also to the Government Chemist for permission to publish this paper. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Ellinger, P., Biol. Rev., 1940, 15, 323. White, C. E., Anal. Chem., 1950, 22, 69. Pringsheim, P., “Fluorescence and Phosphorescence, ’’ Interscience Publishers, Inc. , New York, Smiles, J., J . R. Micro. Soc., 1933, 53, Ser. 111, 207. Barnard, J. E., and Welch, F. K., Ibid., 1936, 56, Ser. 111, 361; 1937, 57, Ser. 111, 256. Chase, C . R., and Pratt, R., J . Amer. Pharm. Ass., 1949, 28, 324. Stanfill, R. C., Quart. Bull. Assn. Food and Drug Ojicial U S . , 1946, 10, 141. Deribere, M., Rev. gen. Mecan, 1949, 33, 251. Helmholtz, H., Ann. Phys., LJ~z., Jubelband, p. 557. Kent-Jones, D. W., Amos, ,4. J., Elias, P. S., Bradshaw, R.C.A., and Thackray, G. B., Analyst, Hinton, J. J . C., Chem. and Ind., 1946, 65, 94. 1949. 1948, 73, 128. GOVERNMENT LABORATORY STRAND, LONDON, W.C.2 February, 1950