Journal of Microscopy, Vol. 167, P t 1,July 1992,pp. 105-121.

Received 1 1 October 199 1 ; revised and accepted 9 January 1992

The application of electron microscopy in dairy research

by D. G. SCHMIDT* and W. BUCHHEIMt, *Department of Biophysical Chemistry,

N I Z O (Netherlands Institute for Dairy Research), P O Box 20, 6710 B A Ede,
The Netherlands, and t Institut fur Chemie und Physik,
Bundesanstalt fur Milchforschung, Postfach 6069, W-2300 Kiel 14, Germany

KEY w 0 R D s. Shadow casting, negative staining, thin sectioning, freeze-fracturing,

cryomicroscopy, immuno-electron microscopy, dairy products, casein micelles, milk-
fat globules, butter, cream, cheese, evaporated milk, milk powder.

SUMMARY
A review is presented of the application of various electron microscopical techniques
in dairy research. Special attention is given to the use of cryofixation techniques, such as
freeze-fracturing, and of cryomicroscopy. These techniques are particularly well suited
for the investigation of dairy products, which are mostly characterized by a high water
content and by the presence of milk fat, having a low melting point. With these
techniques, fat globules in milk, cheese and cream, and casein micelles in milk and milk
products have been studied in an unperturbed state. By using immunogold-labelled
cryosections it could be demonstrated that in casein micelles in milk the kappa-casein is
located at the surface of the micelles, whereas the other caseins are distributed
throughout the whole micelle.

INTRODUCTION
Since the first application of electron microscopy in dairy research by Nitschmann
(1949), who investigated casein micelles in skimmed milk, the microstructure of milk
and dairy products has been the subject of many investigations (e.g. Brooker, 1979;
Kalab, 1979a, 1981; Buchheim, 1982; Schmidt, 1982a; Buchheim & Dejmek, 1990). A
rather complete bibliography of ultrastructure analysis of milk and dairy products has
recently been compiled by Holcomb (1991). Electron microscopy of most milk and
dairy products is hampered by several of their characteristic peculiarities, such as a high
water content, the absence of extended three-dimensional structures in fluid products
and the presence of fat with a low melting point (milk fat only solidifies completely at
temperatures below - 40°C). This milk fat is hardly fixed by the normal fixatives such
as glutaraldehyde and 0 ~ 0 4 Therefore,
. primarily non-fat products can be investigated
by techniques such as shadow casting, negative staining and conventional scanning
electron microscopy (SEM). In thin sections of plastic-embedded material, consider-
able extraction of fat may take place during the embedding procedure and deformation
of fat globules may occur during sectioning (Schmidt, 1982a).

(0 1992 T h e Royal Microscopical Society

105
106 D . G. Schmidt a n d W . Buchheim

Because of these difficulties it is not surprising that cryo-techniques, particularly

freeze-fracturing and freeze-etching, have found wide application in the electron
microscopy of milk and dairy products (Buchheim, 1982; Buchheim & Dejmek, 1990).
True cryomicroscopy, in which the specimen is observed at temperatures below
- 100°C,however, has only rarely been carried out, mostly for SEM of cream (Schmidt
& Van Hooydonk, 1980; Brooker et al., 1986)and cheese (Schmidt et al., 1979). Casein
micelles in frozen thin films of skimmed milk have been observed in transmission
electron microscopy (TEM) using a cryo-stage (van Bruggen et al., 1986).
Here we present a review of the ‘state of the art’ of the electron microscopy of milk
and dairy products.

T R A N S M I S S I O N ELECTRON MI C R O SC O PY
Shadow casting
Shadow casting of a dried specimen is a simple technique for the study of colloidal
protein particles such as casein micelles and their subunits (Nitschmann, 1949; Schmidt
& Buchheim, 1970). It has been demonstrated that simple air-drying of the specimen
yields collapsedmicelles even after proper fixation (Nitschmann, 1949; Schmidt, 1982a).
This distortion is due to the strong interfacial forces which occur when the receding
water surface passes over the particles. Better results have been obtained with freeze-
dried specimens, as shown in Fig. l(a) (Schmidt, 1982a). In addition, critical-point
drying in combination with a replica technique and rotary shadowing has produced
excellent results (Harwalkar et al., 1989). Shadow casting has also successfully been
applied to isolated membranes of milk-fat globules (Oortwijn et al., 1977).

Negative staining
Negative staining is another simple technique (Kalab, 1981) in which the particles are
more or less embedded in a thin layer of heavy metal salts such as potassium tungstate,
uranyl acetate or ammonium molybdate. Casein micelles thus retain their globular
structure without collapsing (Fig. 1b). The method has also been applied to the study of
heat denaturation of whey proteins (Heertje et al., 1985)and of the interaction between
carrageenan and kappa-casein (Snoeren et al., 1976). A combination of freeze-fracture
replication and negative staining (W. Buchheim, unpublished data) appeared particu-
larly suitable for studying air-serum interfaces in foams and interfacial layers of fat
droplets in milk (Fig. 2a).

Thin sectioning of plastic-embedded samples

One of the most widespread techniques of specimen preparation for electron
microscopy is thin sectioning of plastic-embedded samples. This technique comprises a
fixation step, for instance with glutaraldehyde and/or 0 ~ 0 4 dehydration
, with a graded
series of acetone or ethanol and finally impregnation in some suitable plastic monomer
such as Araldite or Epon. After hardening, thin sections are cut which can eventually be
post-stained with lead citrate. In this way extended three-dimensional structures may
be studied, for instance those in cheese. Dispersions such as casein micelles and/or fat
globules in milk or evaporated milk may be studied by this technique, using Salyaev’s
procedure to confine small aliquots of the sample in a tiny agar capsule, which can
subsequently be handled like a piece of tissue (Salyaev, 1968; Henstra & Schmidt,
1970a).
In dairy research, the thin-sectioning technique has been applied to study milk
(Henstra & Schmidt, 1970a, b), homogenized milk (Henstra & Schmidt, 1970b),
concentrated milk (Harwalkar & Vreeman, 1978), yoghurt (Kalab et al., 1983), curd
formation (Henstra & Schmidt, 1970c; Green et al., 1978), cheese (Kimber e t al., 1974;
Knoop & Buchheim, 1980) and foams (Brooker, 1985; Brooker et al., 1986). Examples
 Electron microscopy in dairy research 107

Fig. 1. Transmission electron micrographsof casein micelles by: (a) shadow casting, (b) negative staining, (c)
cryomicroscopy, (d) freeze-fracturing and unidirectional shadowing, (e) freeze-fracturing and rotary
shadowing, (f) freeze-etching and rotary shadowing. Micrographs (a)-(c) are from cows’ milk, (d)-(f) from
mares’ milk; scale bars = 100 nm.

of this technique are given by Fig. 3(a, b), showing fat-protein complexes in
concentrated milk, and by Fig. 4(a), showing fat globules and casein in Gouda cheese.
The method of thin sectioning of plastic-embedded specimens is very popular in
spite of the risk of the formation of artefacts, particularly during the dehydration and
embedding steps, as has been pointed out elsewhere (Schmidt, 1982a). Dehydration at
low temperatures followed by embedding at low temperatures using a Lowicryl-based
plastic might be less harmful, but has only occasionally been applied in dairy research
(Horisberger & Vautey, 1984; Horisberger & Rouvet-Vautey, 1984).

Freeze-fracturing and freeze-etching

Soon after the introduction of the freeze-fracturelfreeze-etch technique as a novel
108 D . G. Schmidt and W . Buchheim

Fig.2. View of the original milk-fat globule membrane: (a-c) cows’ milk, (d) mares’ milk. (a) Combination of
freeze-fracturing and negative staining; (b) freeze-etching (after fixation by glutaraldehyde and washing in
distilled water)-the arrowheads in (a) and (b) point to the boundary of the proteinaceous coat domain; (c)
inner view of one domain (obtained when a globule is cleft between the membrane and the fat core) showing a
pronounced paracrystalline order (asterisks) which is most probably due to the presence of xanthine oxidase;
(d) freeze-etching of washed globules-mucin-like glycoproteins (M) are well documented on the surface of
milk-fat globules of certain species (compare with b); scale bars = 1 pm. (Reprinted from Welsch et al. (1988),
courtesy of Springer-Verlag.)

TEM methpd for biological tissue and related systems, it was applied to the study of
protein and lipid structures in milk and in a wide variety of dairy products (Buchheim &
Dejmek, 1990). Depending on the water content, which can vary in dairy systems from
close to zero in dried products up to more than 997,, as in dilute milk protein solutions,
various preparatory pre-treatments prior to cryofixation had to be developed (Buch-
heim, 1981, 1982). Freeze-fracture replicas of fresh raw milk yield structural
information mainly on colloidal casein micelles (Fig. Id, e) and on fat droplets and their
surface layer, the milk-fat globule membrane (Fig. 2 ) , which originates from the apical
plasma membrane of the mammary epithelial cells (Buchheim, 1982; Buchheim &
Dejmek, 1990). The fine structure of the fat phase depends on whether the fat has been
 Electron microscopy in dairy research 109

Fig. 3. Micrographs of normally sterilized (15 min, 120°C) evaporated milk: (a) thin section, (c) freeze-
fracture; ultra high-temperaturesterilized (15s, 135°C)evaporatedmilk:(b)thin section, (d) freeze-fracture.
Note the difference in size of the fat-casein aggregates between the two types of evaporated milk; F: fat, C:
casein; scale bars = 1 pm.
liquid or (partly) crystallized prior to cryofixation (Buchheim, 1970a, b). Thus,
important structural information is obtained about the distribution of liquid and
crystallized fat, whether in the emulsified state or in a continuous phase as in butter
(Precht & Buchheim, 1979, 1980a, b).
Whereas caseins, having a loose randomly coiled structure, are easily visible on
freeze-fracture replicas, even in non-micellar solutions (Schmidt & Buchheim, 1976),
110 D . G. Schmidt and W . Buchheim

Fig. 4. Micrographs of Gouda cheese: (a) thin section of a resin-embedded sample, scale bar = 1 pm;
(b) freeze-fractured sample, scale bar = 1 pm; (c) cry0 thin section, scale bar= 1 pm; (d) cry0 scanning
micrograph, scale bar = 2 pm. F: fat, C: casein, M: remnants of the milk-fat globule membrane.

the major whey proteins a-lactalbumin and P-lactoglobulin, which are compact
globular proteins, are difficult to identify in the background of the platinum-carbon
evaporation layer. After thermal denaturation of whey proteins, which often leads to
aggregation, their appearance on freeze-fracture replicas resembles that of aggregated
caseins (Jelen & Buchheim, 1976). a-Lactalbumin and P-lactoglobulin in their native
 Electron microscopy in dairy research 1 11

Fig. 5. Coffee cream (12% fat), homogenized and ultra high-temperature treated. (a) Freeze-fracture of the
original product (F: fat globule, the arrowheads point to adsorbed and free casein); (b) freeze-etch
micrograph of large fat-protein aggregates formed during flocculation (feathering) of the cream in a hot coffee
solution; scale bars = 1 pm.

state, however, have been visualized .n spray-frozen preparations of protein solutions in

very dilute buffers after freeze-etching (Schmidt & Buchheim, 1980). Many studies on
protein-lipid interactions in processed milks, e.g. the effects of homogenization,
thermal treatments, pH changes, etc., and in a variety of typical liquid milk products
like evaporated milks (Fig. 3c, d), homogenized creams (Fig. 5), ice cream emulsions
etc., have been performed by applying freeze-fracture conditions (Schmidt et al., 1971;
Heertje et al., 1985; Buchheim & Dejmek, 1990).

Fig. 6. Surface views of fat droplets as revealed by freeze-etching. (a) Fat homogenized in skimmed milk;
(b) fat homogenized in sodium caseinate solution; scale bars = 1 pm.
112 D . G. Schmidt and W . Buchheim

Fig. 7. Freeze-fracture preparations of a fresh cream cheese (a) and a processed cheese (b), both 60% fat in
dry matter. In the cream cheese fat globules (F) and milk protein, i.e. casein (arrowheads), form large
aggregates separated by an aqueous phase, i s . whey (W). In processed cheese the fat globules (F) are evenly
distributed in a rather uniform protein matrix (M); scale bars = 1 pm. (Reprinted from Buchheim & Dejmek
(1990), courtesy of Marcel Dekker Inc.)

T o obtain surface views of particles such as casein micelles (Fig. l f ) and fat droplets
(Figs. 2b, d, 5b and 6), freeze-etching must be carried out to a depth of up to several
micrometres. Since any deposition of foreign matter onto these surfaces must be
avoided, it is necessary to wash the dispersions in distilled water. Using this approach,
large areas of interfacial layers of fat droplets in milk and other dairy emulsions (Fig. 6)
have been made visible (Buchheim, 1986; Buchheim et al., 1988; Buchheim & Dejmek,
1990). One major finding has been the detection of peculiar mucin-like filamentous
glycoproteins on the surface of freshly secreted fat globules in milks of some species
(Fig. 2d; Buchheim, 1986; Buchheim et al., 1988).
The freeze-fracture/freeze-etch methods have also been successfully applied to
various types of natural as well as processed cheeses, providing information on the
distribution of fat and protein and their fine structure (Figs. 4b and 7; Knoop &
Buchheim, 1980; Klostermeyer & Buchheim, 1988). Freeze-fracture studies on various
types of dairy foams, such as whipped cream and ice cream (frozen or melted), have
revealed detailed information on the stabilization mechanisms at the air-serum
interface and also on effects of low-molecular-weight emulsifiers like monoglycerides
on lipid-protein interactions (Berger & White, 1971; Buchheim e t al., 1985; Brooker
1990; B,uchheim & Dejmek, 1990). Freeze-fracturing is also the only T E M method
providing detailed structural information on the fat phase and the aqueous phase in
dairy spreads like butter (Fig. 8) or low-fat dairy spreads which represent emulsions of
the water-in-oil type (Precht & Buchheim, 1979,1980aYb; Buchheim & Dejmek, 1990).
Water-soluble powders such as spray-dried whole milk powder can be prepared by
freeze-fracturing after suspending the powder in a non-aqueous medium such as
paraffin oil or polyethylene glycol (Fig. 9a; Buchheim, 1982).
 Electron microscopy in dairy research 1 13

Fig. 8. Freeze-fracture preparation of butter which represents a water-in-oil type emulsion. Water droplets
(W) are dispersed in a partly crystallized fat phase which still contains a varying amount of more or less
undestroyed fat globules (gF)which survived the churning process of the cream; scale bar = 1 pm. (Reprinted
from Buchheim & Dejmek (1990), courtesy of Marcel Dekker Inc.)

Cryomicroscopy
With the recent introduction of suitable cryo-stages for both T E M and SEM, cryo-
electron microscopy has become a practical tool for the examination of biological
materials close to their native state (Adrian et al., 1984; Roos & Morgan, 1990). A thin
film of a dispersion of small particles can be rapidly quench-frozen so that no ice
crystallization occurs. I t is then examined in the electron microscope at - 1 5 0 T in the
frozen state. In this way artefacts due to drying, staining, shadowing, fracturing or
sectioning can be avoided. Thus, casein micelles in their natural environment can easily
be studied, as is demonstrated in Fig. l(c) (van Bruggen et al., 1986). Due to the extreme
thinness of the frozen layer, the method is only applicable to small particles such as
casein micelles and viruses. Even then, artefacts will form due to boundary forces when
the particles adsorb at the liquid-air interface.
In the case of larger structures, which occur for instance in concentrated milk,
yoghurt, cheese, etc., it is necessary to use thin sections from cryofixed specimens which
have to be examined at - 150°C. T h e problem here is the proper cryofixation of the
sample, during which formation of ice crystals must be prevented. Except for the case of
114 D. G. Schmidt and W . Buchheim

Fig. 9. Micrographs of spray-dried whole milk powder. (a) Freeze-fracture micrograph of the powder
suspended in anhydrous polyethylene glycol 400 (PEG). This peripheral region of a cleft powder particle
reveals a thin layer of surface fat (sF) and details of the distributionof globular fat (gF) in the powder matrix
(PM) which represents the non-fat constituents of milk, i.e. milk proteins (arrowhead points to casein
particles), lactose, minerals, etc.; scale bar = 1 pm. (b) Scanning electron micrograph; note the vacuole (V) in
the cross-fractured particle and the relatively smooth surface of the particles; scale bar = 10 pm.

high-pressure cryofixation (Studer et al., 1989), only specimens smaller than 10 pm can
be frozen completely without damage due to the formation of ice crystals. Only when
cryoprotectants such as glycerol or sucrose are added is it possible to vitrify larger
samples, but this introduces the possibility of artefacts due to the cryoprotectant.
Without cryoprotectants only the outer 5-7 pm of a bulk sample will be vitrified with
immersion cryofixation. In most cases this will be sufficient for the study of the
structure in dairy products such as concentrated milk, yoghurt, custards, etc.
Particularly for the study of the interaction between proteins and substances like fat
and polysaccharides, which can hardly be fixed or stained in another way, cryo-electron
microscopy is the obvious technique for the future. Until now, cryo-electron
microscopy has rarely been applied in dairy research, which might be partly ascribed to
the scarcity of reliable cryo-electron microscopes and partly to the high water content of
most dairy products which would easily give rise to freezing artefacts.
In products with a low water content, however, there will be little crystal formation
during freezing. A piece of cheese can therefore be frozen without difficulties, sectioned
and studied in TEM at low temperature (Fig. 4c).

Immuno-electron microscopy
Immuno-electron microscopy has occasionally been applied in dairy research to
elucidate the location of kappa-casein in casein micelles, but the results are confusing
and mostly not in agreement with the current model of casein micelles (Schmidt,
1982b). Carroll & Farrell (1983), using ferritin-labelled antibodies against kappa-
casein, found that the kappa-casein was located at the micellar surface in large micelles
but distributed throughout the whole micelle in smaller ones. Horisberger &
Vonlanthen (1980) and Horisberger & Rouvet-Vautey (1984) used gold-labelled lectin
 Electron microscopy in dairy research 1 15

and Horisberger & Vautey (1984) used gold-labelled protein A with anti-kappa-casein
to locate the kappa-casein. Although their results were partly conflicting, they favoured
an even distribution of the kappa-casein in the micelles. In these examples the
immunolabelling was carried out using thin sections of plastic-embedded casein
micelles which had first been fixed by glutaraldehyde. Although good results may be
obtained in this way, artefacts will be present, particularly as a result of the possible
effects of fixation and embedding in resin on the reactivity of the antigen with the
antibodies.
Much better results would be expected by labelling thin sections cut from frozen
samples according to Tokuyasu’s method (Tokuyasu, 1980; Griffiths et al., 1984). This
technique has been used to localize kappa-casein in casein micelles, using gold-labelled
protein A and antibodies against whole casein and against kappa-casein (D. G.
Schmidt, unpublished data).
Casein micelles in fresh skimmed milk were fixed with 2% formaldehyde for 30 min,
after which they were isolated by centrifugation at 100,000 g for 1 h at 10°C. The
sedimented micelles were repeatedly redispersed and centrifuged in simulated milk
ultrafiltrate (Jenness & KOOPS,1962) and finally embedded in 2% gelatin. Small cubes
( < 1 mm’) of the solidified dispersion in gelatin were infiltrated with a 2.3 M solution of
sucrose for 24 h, subsequently mounted on copper stubs and frozen in liquid nitrogen.
Cryo-sections 70-100 nm thick were cut at - 100°C using a Reichert FC4 cryo-
ultramicrotome. The sections were transferred to carbon-coated pioloform films
mounted on nickel grids, and after thawing were then incubated with antiserum. The
excess antiserum was washed away with phosphate-buffered saline and the sections
were treated with protein A-gold complex. The excess protein A-gold complex was
then removed by washing and the sections were embedded and contrasted with 1%
methylcellulose and 0.1 %, uranyl acetate and finally observed by TEM.
Since the size of the micelles is of the same order of magnitude as the thickness of the
sections, stereo images are necessary for an accurate interpretation. Figure 10 shows
that kappa-casein is located at the micellar surface whereas the other caseins are
distributed throughout the whole micelle. However, these preliminary results must be
checked using antibodies specific to a,- and p-casein as well.

SCANNING ELECTRON MICROSCOPY

SEM of dehydrated specimens
Soon after SEM had been introduced as a novel type of electron microscopy,
especially for viewing the surface morphology of solid samples, it was also applied in
dairy product research. In 1971 the first papers were published about SEM studies on
dried milks and related products (Buma & Henstra, 1971a, b). Further studies on dry
products like spray-, drum- and freeze-dried milks or whey, instant powders and on
lactose crystallization then followed (Kalab, 1979a, 1981; Saltmarch & Labuza, 1980;
Saito, 1985, 1988). SEM studies of dried milks yield information not only on external
particle surfaces-as for instance on the growth of lactose crystals as related to storage
conditions-but also on internal structures (of fragmented particles) like vacuoles,
cracks and capillaries which often are a result of composition and processing conditions
and may be closely related to certain product properties (Fig. 9b).
Investigation of water-containing dairy products by SEM requires adequate
reinforcement of fragile structures and also careful selection of drying procedures.
During the first years of such SEM applications low-fat systems, such as various types
of milk-protein gels and products like yoghurt and cottage cheese, were largely
investigated (Kalab, 1979a, b, 1981; Kalab et al., 1983). Structural stabilization was
achieved by fixation with glutaraldehyde and dehydration was performed mainly by
116 D. G. Schmidt and W . Buchheim

Fig. 10. Stereo electron micrographs of immunogold-labelled casein rnicelles. Top: casein rnicelle section
incubated with anti-whole casein antibodies and subsequently labelled with protein A-gold; both cutting
faces are labelled. Bottom: casein rnicelle section incubated with anti-kappa-casein antibodies and
subsequently labelled with protein A-gold; only the outer surfaces reacted with the gold label, while the
cutting face of the sectioned rnicelle did not react with the label and thus did not contain kappa-casein. This
indicates that kappa-casein is located exclusively on the rnicellar surface.
 Electron microscopy in dairy research 117

Fig. 11. Scanning electron micrograph of milk fat crystals; scale bar = 50 pm.

freeze-drying or critical-point drying (Kalab & Harwalkar, 1973). For S E M studies of

products containing substantial amounts of fat, such as most cheeses, a fat extraction
step was involved prior to the coating with gold (Ruegg et al., 1974). Allan-Wojtas &
Kalab (1984) and Gavaric et al. (1989), however, were able to retain the fat globules in
the matrix of a casein gel quite well by fixation in a 2",, solution of OsO4 in imidazole
buffer.
Further applications concern the fat distribution in various processed cheeses and the
analysis of crystalline material (e.g. citrates and phosphates) which may develop in
natural as well as in processed cheeses (Caric et al., 1985; Brooker, 1987). Several
studies by S E M on cultured milk products and natural cheeses were focused on the
distribution and appearance of micro-organisms associated with such products (Kalab
et al., 1983; Rousseau, 1984). Not only low-fat or defatted dairy systems have been
successfully analysed by conventional SEM, but attempts were also made to study
high-fat products or even pure butter fat. Schaap et al. (1975) demonstrated the
morphology of butter-fat crystals by applying a special replica technique (Fig. 11).

Cryo-SEM of hydrated specimens

With the development of cryo-stages it has become possible to analyse fully hydrated
specimens by S E M (Sargent, 1988). I t has already been mentioned that bulk samples
with a low water content could be frozen without the formation of ice crystals. Even
with rather primitive equipment, Schmidt et al. (1979) were able to obtain good results
with cheese. T h e sample was frozen in liquid nitrogen and fractured. T h e fracture plane
was coated with carbon and finally observed in a scanning electron microscope at
- 100°C. Figure 4(d) shows the fracture plane in cheese, in which fat globules are seen
to be embedded in a continuous matrix of casein.
118 D . G. Schmidt and W . Buchheim

Fig. 12. Cry0 scanning micrograph of whipped cream. A: inner surface of air cell; scale bar = 10 pm.

Cryo-SEM is particularly well suited to the study of the structure in whipped cream
and similar products. Since this structure is largely due to the fat, neither plastic
embedding with T E M nor SEM at room temperature are expected to yield insight into
these structures. With cryo-SEM, good micrographs of whipped cream have been
obtained by Schmidt & Van Hooydonk (1980) and by Brooker et al. (1986). Figure 12 is
a micrograph of whipped cream obtained by this technique, which suggests that the air
bubbles are surrounded by a layer of coalesced fat. For the study of crystalline fat in
butter and other spreads, Heertje et al. (1987) used a special de-oiling technique before
analysing the sample by cryo-SEM.

CONCLUSIONS
Milk and dairy products have been studied by electron microscopy more intensively
and systematicallythan other human foods. The main reason may be that these systems,
being homogeneous on a large scale, exhibit particular fine structures in the
microscopic and submicroscopic size ranges. During various technological treatments,
the microparticulate constituents of milk, i.e. the fat globules, the colloidal casein
micelles and the molecular dispersion of whey proteins, undergo significant physical
changes and mutual interactions. This gives rise to the characteristic macroscopic
structure and physical properties of the products. Therefore, electron microscopy of
dairy products is extremely useful in elucidating the relationship between macroscopic
properties of the product and its submicroscopic structure as altered by technological
treatments. In view of the present trend to incorporate separate milk constituents,
generally after physical, enzymatic or chemical modification, into other food, electron
microscopy will continue to play a significant role in the elucidation of the properties of
such new products. In this respect, the newly developed cryotechniques, with which
 Electron microscopy in dairy research 119

the products can be studied in their natural hydrated state, should prove to be quite
'fruitful.

ACKNOWLEDGMENTS
Micrographs l(a), 3, 4(a, b, d), 9(b), 11 and 12 were taken on microscopes of the
Technical and Physical Engineering Research Service (TFDL), Wageningen, micro-
graphs l(c) and 10 on a Philips EM 400 T microscope at the Biochemical Laboratory,
University of Groningen, in co-operation with D r F. P. Booy and Mr J. F. L. van
Breemen and micrograph 4(c) was taken on a Philips EM 400 microscope at the EMBL
in Heidelberg during a course on cryomicroscopy organized by Dr J. Dubochet in 1983.

REFERENCES
Adrian, M., Dubochet, J., Lepault, J. & McDowall, A.M. (1984) Cryo-electron microscopy of viruses.
Nature, 308, 32-36.
Allan-Wojtas, P. & Kalab, M. (1984) Milkgel structure XIV. Fixation of fat globules in whole milk yoghurt
for electron microscopy. Milchwissenschaft, 39, 323-327.
Berger, K.G. & White, G.W. (1971) An electron microscopical investigation of fat destabilization in ice
cream. 3. Food Technol. 6,285-294.
Brooker, B.E. (1979) Milk and its products. Food Microscopy (ed. by J. G. Vaughan), pp. 273-31 1. Academic
Press, New York.
Brooker, B.E. (1985) Observations on the air-serum interface of milk foams. Food Microstructure, 4,289-296.
Brooker, B.E. (1987) The crystallization of calcium phosphate at the surface of mould-ripened cheeses.
Food Microstructure, 6, 25-33.
Brooker, B.E. (1990) The adsorption of crystalline fat to the air-water interface of whipped cream.
Food Microstructure, 9, 223-229.
Brooker, B.E., Anderson, M. & Andrews, A.T. (1986) The development of structure in whipped cream. Food
Microstructure, 5, 277-285.
van Bruggen, E.F.J., Booy, F.P., van Breemen, J.F.L., Brink, J., Keegstra, W. & Schmidt, D.G. (1986)
Cryo-EM of various biomolecular assemblies. l'roc. X I t h Int. Congr. Electron Microscopy, Kyoto (ed. by
T. Imura, S. Maruse and T. Suzuki), pp. 2423-2424. Japanese Society of Electron Microscopy.
Buchheim, W. (1970a) The process of fat crystallization in the fat globules of milk. Electron microscopical
studies by means of the freeze etching method. Milchwissenschaft, 25, 65-70.
Buchheim, W. (1970b) The molecular order of bi-refringent fat globules. Milchwissenschaft, 25, 223-227.
Buchheim, W. (1981) A comparison of the microstructure of dried milk products by freeze-fracturing powder
suspensions in non-aqueous media. Scanning Electron Microsc. 1981/III, 493-502.
Buchheim, W. (1982) Aspects of sample preparation for freeze-fracture/freeze-etchstudies of proteins and
lipids in food systems. A review. Food Microstructure, 1, 189-208.
Buchheim, W. ( 1 986) Membranes of milk fat globules-Ultrastructural, biochemical and technological
aspects. Kieler Milchw. Forsch. Ber. 38, 227-246.
Buchheim, W., Barfod, N.M. & Krog, N. (1985) Relation between microstructure, destabilization
phenomena and rheological properties of whippable emulsions. Food Microstructure, 4,22 1-232.
Buchheim, W. & Dejmek, P. (1990) Milk and dairy-type emulsions. Food Emulsions, 2nd edn (ed. by K.
Larsson & S . E. Friberg), pp. 203-246. Marcel Dekker, New York.
Buchheim, W., Welsch, U. & Patton, S. (1988) Electron microscopy and carbohydrate histochemistry of the
human milk fat globule membrane. Biology of Human Milk, Nestle' Nutrition Workshop (ed. by L. A.
Hanson), Vol. 15, pp. 27-44. Raven Press, New York.
Buma, T.J. & Henstra, S. (1971a) Particle structure of spray-dried milk products as observed by a scanning
electron microscope. Neth. Milk Dairy 3. 25, 75-80.
Buma, T.J. & Henstra, S. (1971b) Particle structure of spray-dried caseinate and spray-dried lactose as
observed by a scanning electron microscope. Neth. Milk Dairy 3. 25,278-281.
Caric, M., Gantar, M. & Kalab, M. (1985) Effects of emulsifying agents on the microstructure and other
characteristics of processed cheese-a review. Food Microstructure, 4,297-3 12.
Carroll, R.J. & Farrell, H.M., Jr (1983) Immunological approach to location of kappa-casein in the casein
micelle by electron microscopy. 3. Dairy Sci. 66, 679-686.
Gavaric, D.D., Caric, M. & Kalab, M. (1989) Effects of protein concentration in ultrafiltration milk
retentates and the type of protease used for coagulation on the microstructure of resulting gels. Food
Microstructure, 8, 53-66.
Green, M. J., Hobbs, D.G. & Morant, S.V. (1978) Intermicellar relationships in rennet-treated separated
milk. I. Preparation of representative micrographs. 3. Dairy Res. 45,405-411.
Griffiths, G., McDowall, A.M., Back, R. & Dubochet, J. (1984) On the preparation of cryosections for
immunocytochemistry. 3. Ultrastructure Res. 89,65-78.
120 D . G. Schmidt and W . Buchheim
Harwalkar, V.R., Allan-Wojtas, P. & Kalab, M. (1989) Effect of heating to 200°Con casein micelles in milk: a
metal shadowing and negative staining electron microscopic study. Food Microstructure, 8, 217-224.
Harwalkar, V.R. & Vreeman, H.J. (1978) Effect of added phosphates and storage on changes in ultra-high
temperature short-time sterilized concentrated skim-milk. 2. Micelle structure. Neth. Milk Dairy J. 32,
204-216.
Heertje, I., Leunis, M., van Zeyl, W. J.M. & Berends, E. (1987) Product morphology of fatty products. Food
Microstructure, 6, 1-8.
Heertje, I., Visser, J. & Smits, P. (1985) Structure formation in acid milk gels. Food Microstructure, 4,267-
277.
Henstra, S. & Schmidt, D.G. (1970a) Ultra thin sections from milk using the microcapsule method.
Naturwissenschaften, 57, 247-248.
Henstra, S. & Schmidt, D.G. (1970b) On the structure of the fat-protein complex in homogenized cow’s
milk. Neth. Milk Dairy J . 24, 45-51.
Henstra, S. & Schmidt, D.G. (1970~)Study of the rennet coagulation of cow’s milk by the microcapsule
method. Proc. 7th Int. Congr. Electron Microscopy (ed. by P. Favard), pp. 389-390. Societe Franqaise de
Microscopie Electronique, Paris.
Holcomb, D.N. (1991) Structure and rheology of dairy products. A compilation of references with subject
and author indexes. Food Structure, 10,45-108.
Horisberger, M. & Rouvet-Vautey, M. (1984) Localization of glycosylated kappa-casein on thin sections of
casein micelles by lectin-labelled gold markers. Histochemistry, 80, 523-526.
Horisberger, M. & Vautey, M . (1984) Localization of kappa-casein on thin sections of casein micelles by the
gold method. Hisrochemistry, 80, 9-12.
Horisberger, M. & Vonlanthen, M . (1980) Localization of glycosylated kappa-casein in bovine casein
micelles by lectin-labelled gold granules. J. Dairy Res. 47, 185-191.
Jelen, P. & Buchheim, W. (1976) Microstructure of heat-coagulated whey protein curd. Milchwissenschaft,
31,589-592.
Jenness, R. & Koops, J. (1962) Preparation and properties of a salt solution which simulated milk ultrafiltrate.
Neth. Milk Dairy 3. 16, 153-164.
Kalab, M. (1979a) Scanning electron microscopy of dairy products: an overview. Scanning Electron Microsc.
1979/III, 261-272.
Kalab, M . (1979b) Microstructure of dairy foods. 1. Milk products based on protein.3. Dairy Sci. 62,1352-
1364.
Kalab, M . (1981) Scanning electron microscopy of dairy products: a review of techniques. Scanning Electron
Microsc. 1981/III, 453-472.
Kalab, M., Allan-Wojtas, P. & Phipps-Todd, B.E. (1983) Development of microstructure in set-style non-fat
yoghurt-a review. Food Microstructure, 2, 51-66.
Kalab, M. & Harwalkar, V.R. (1973) Milk gel structure. I. Application of scanning electron microscopy to
milk and other food gels. J. Dairy Sci. 56, 835-842.
Kimber, A.M., Brooker, B.E., Hobbs, D.G. & Prentice, J.H. (1974) Electron microscope studies of the
development of structure in cheddar cheese. J. Dairy Res. 41, 389-396.
Klostermeyer, H . & Buchheim, W. (1988) Microstructure of processed cheese products. Kieler Milchw.
Forsch. Ber. 40, 219-231.
Knoop, A.-M. & Buchheim, W. (1980) T h e different development of the structure in Harzer-, Tilsit-, and
Camembert-cheese during ripening. Milchwissenschaft, 35, 482-488.
Nitschmann, H . (1949) Electron microscopical size determination of calcium caseinate particles in cow’s
milk. Helv. Chim. Acta, 32, 1258-1264.
Oortwijn, H., Walstra, P. & Mulder, H . (1977) T h e membranes of recombined fat globules. I. Electron
microscopy. Neth. Milk Dairy 3. 31, 134-147.
Precht, D. & Buchheim, W. (1979) Electron microscopic studies on the physical structure of spreadable fats.
I. The microstructure of fat globules in butter. Milchwissenschuft, 34, 745-749.
Precht, D . & Buchheim, W. (1980a) Electron microscopic studies on the physical structure of spreadable fats.
11. T h e microstructure of the interglobular fat phase in butter. Milchwissenschuft, 35,393-398.
Precht, D . & Buchheim, W. (1980b) Electron microscopic studies on the physical structure of spreadable fats.
111. T h e aqueous phase in butter. Milchwissenschaft, 35, 684-690.
Roos, N . & Morgan, A. J. (1990) Cryopreservation of Thin Biological Specimens for Electron Microscopy:
Methods and Applications. Oxford Science Publishers, Oxford.
Rousseau, M. (1984) Study of the surface flora of traditional Camembert cheese by scanning electron
microscopy. Milchwissenschaft, 39, 129-135.
Riiegg, M., Sieber, R. & Blanc, B. (1974) T h e structure of cheese as observed by a scanning electron
microscope. Schweiz. Milchwirt. Forschung, 3, 1-5.
Saito, Z. (1985) Particle structure in spray-dried whole milk and instant skim milk powder as related to lactose
crystallization. Food Microstructure, 4, 333-340.
Saito, Z. (1988) Lactose crystallization in commercial whey powders and in spray-dried lactose. Food
Microstructure, 7, 75-81.
 Electron microscopy in dairy research 12 1

Saltmarch, M. & Labuza, T.P. (1980) SEM investigation of the effect of lactose crystallization on the storage
properties of spray-dried whey. Scanning Electron Microsc. 1980/III, 659-665.
Salyaev, R.K. (1968) A method of fixation and embedding of liquid and fragile materials in agar
microcapsulae. Proc. 4th Eur. Reg. Conf. Electron Microscopy (ed. by D . S. Bocciarelli), pp. 37-38.
Tipografia Poliglotta Vaticana, Rome.
Sargent, J.A. (1988) The application of cold stage scanning electron microscopy to food research. Food
Microstructure, 7, 123-135.
Schaap, J.E., Badings, H.T., Schmidt, D.G. & Frede, E. (1975) Differences in butterfat crystals, crystallized
from acetone and from the melt. Neth. Milk Dairy J . 29, 242-252.
Schmidt, D.G. (1982a) Electron microscopy of milk and milk products: problems and possibilities. Food
Microstructure, 1, 151-165.
Schmidt, D.G. (1982b) Association of casein and casein micelle structure. Developments in Dairy Chemistry
(ed. by P. F. Fox), Vol. 1, pp. 61-86. Applied Science Publishers, Barking.
Schmidt, D.G. & Buchheim, W. (1970) An electron microscopical investigation of the sub-structure of the
casein micelles in cow’s milk. Milchwissenschaft, 25, 596-600.
Schmidt, D.G. & Buchheim, W. (1980) The size of alpha-lactalbumin and beta-lactoglobulin molecules as
determined by electron microscopy using the spray-freeze-etching technique. Milchwissenschaft, 35,
209-2 11.
Schmidt, D.G. & Buchheim, W. (1976) Particle size distribution in casein solutions. Neth. Milk Dairy J . 30,
17-28.
Schmidt, D.G., Buchheim W. & Koops, J. (1971) An electron microscopical study of the fat-protein
complexes in evaporated milk, using the freeze-etching technique. Neth. Milk Dairy J . 25, 200-2 16.
Schmidt, D.G., Henstra, S. & Thiel, F. (1979) A simple low-temperature technique for scanning electron
microscopy of cheese. Mikroskopie ( W i e n ) , 35, 50-55.
Schmidt, D.G. & Van Hooydonk, A.C.M. (1980) A scanning electron microscopical investigation of the
whipping of cream. Scanning Electron Microsc. 1980/III, 653-658.
Snoeren, T.H.M., Both, P. & Schmidt, D.G. (1976) An electron microscopic study of carrageenan and its
interaction with kappa-casein. Neth. Milk Dairy J . 30, 132-141.
Studer, D., Michel, M. & Muller, M. (1989) High pressure freezing comes of age. Scanning Microsc. Suppl. 3,
253-269.
Tokuyasu, K.T. (1980) Immunocytochemistry of ultrathin frozen sections. Histochem. 3. 12, 381-403.
Welsch, U., Buchheim, W., Schumacher, U., Schinko, I . & Patton, S. (1988) Structural, histochemical and
biochemical observations on horse milk fat globule membranes and casein micelles. Hisrochemistry, 88,
357-365.