Composition, Structure, and Integrity of Casein Micelles" A Review 1

D O N A L D J. M c M A H O N 2 and R O D N E Y J. BROWN
Department of Nutrition and Food Sciences
Utah State University
Logan 84322

ABSTRACT proposed and reviewed (7, 21, 22, 28, 76, 84,
This review is an attempt to bring 89, 92, 96). A casein micelle consisting of
what is known about casein micelles numerous subunits is the most widely accepted
together into a coherent summary. Some model. Some calculated characteristics of casein
of the earlier models and theories no micelles are summarized in Table 2.
longer appear workable, and much is left Casein micelles exhibit a broad statistical
to be concluded on this subject. It is distribution of. sizes governed by physical
hoped with what we now know that chemical principles (5). Researchers using
those who already have contributed and inelastic light scattering have observed that 80%
others will continue to pursue the under- of micelles by weight have a diameter in the
standing of casein micelles. range 100 to 200 nm, with 95% between
80 and 440 nm (51), and probable diameter of
PHYSICAL PROPERTIES 160 nm. Laser light scattering has shown a
OF CASEIN MICELLES broader distribution of molecular weight
The biological function of bovine casein (MW) than has either electron microscopy
micelles is to provide efficient nutrition to the measurements or quasi-elastic light scattering
young calf. It does not require inherently (39) with a long tail extending up to 680
a high degree of ordered structure but, rather, nm diameter (53). Particles less than 20 nm
an effective mechanism for secretion of a highly diameter account for nearly 80% by number of
concentrated solution of protein, calcium, and all casein particles, but their volume comprises
phosphate. During the past 20 yr research has less than 3% of the total micellar volume (82).
been extensive to determine the composition If particles less than 20 nm diameter are ex-
and structure of casein micelles and to identify cluded, then significant differences in char-
forces that maintain their integrity. The ap- acteristics are obtained (Table 3).
proximate composition of bovine casein micelles
is in Table 1. In cows' milk, casein micelles T H E R M O D Y N A M I C FORCES
occur in colloidal dispersion. Structure and IN CASEIN MICELLES
properties of the casein group of proteins,
which comprise over 90% of the mass of casein Hydrophobic Interactions
micelles, have been reviewed (11, 13, 76, 89, Caseins are among the most hydrophobic of
90, 98). Casein micelles are highly hydrated and all proteins, and stabilization of casein micelles
spongelike colloidal particles containing about should involve hydrophobic interactions (23).
3.7 g H 2 0 / g protein (6, 47, 48). Relatively All casein associations are promoted by increase
little of this water (.5 g H 2 0 / g protein) is of temperature, indicating involvement of
bound to the protein. The remainder is occluded hydrophobic interactions (66). Proline reduces
within the micelle and moves with the micelle h y d r o p h o b i c i t y when it is bound to nonpolar
during hydrodynamic experiments. Numerous regions of protein, and it causes partial dis-
models describing casein micelles have been sociation of casein micelles (57). Urea is more
effective than oxalate in dissociating micelles,
again suggesting the importance of hydrophobic
interactions in stabilizing micelles (58).
Received July 14, 1983. Thermodynamic properties of a solution are
1Utah Agricultural Experiment Station Journal
Article No. 2849. Approved by the Director. the composite result of all components and are
2Kraft Foods Ltd., Salmon St., Port Melbourne, not soley due to the nature of the solute (5).
Victoria, 3207, Australia. Changes of water structure surrounding non-

1984 J Dairy Sci 67:499-512 499

500 McMAHON AND BROWN

TABLE 1. Approximate composition of bovine casein other rather than to remain surrounded by
micelles (77, 79). water molecules. A h y d r o p h o b i c interaction
occurs when two or m o r e n o n p o l a r groups
Component Content Reference c o m e into c o n t a c t (62). A p r o p e r t y u n i q u e to
h y d r o p h o b i c interactions is their d e p e n d e n c y
(g/lO0 g micelles) on solvent m e d i u m for existence.
as1 -casein 35.6 77 In an aqueous solution, proteins occur in a
C~s2-casein 9.9 77 solvent of which i n t e r m o l e c u l a r arrangements
~-casein 33.6 77
are influenced strongly by interactions witla
x-casein 11.9 77
Minor caseins 2.3 77 protein residues. Folding of a protein m o l e c u l e
Calcium 2.9 77 and stability of its various c o n f o r m a t i o n s are
Phosphate 2.9 77 the c o m b i n e d results of all its interactions. The
Magnesium .1 77 position o f equilibrium of i n t e r m o l e c u l a r forces
Sodium .1 77
Potassium .3 77 is governed by Gibbs free energy change, and at
Citrate .4 77 r o o m t e m p e r a t u r e unfavorable e n t h a l p y of
Sialic acid .3 79 f o r m a t i o n is m o r e than c o u n t e r b a l a n c e d by
Galactose .2 79 positive e n t r o p y changes. Increased e n t r o p y
Galactosamide .2 79
due to changes of water structure is the m o s t
i m p o r t a n t factor stabilizing h y d r o p h o b i c inter-
actions. F o r each square n a n o m e t e r protein
surface removed f r o m c o n t a c t with water,
polar solutes play an i m p o r t a n t role in the a p p r o x i m a t e l y 10 kJ of free energy is obtained
f o r m a t i o n o f h y d r o p h o b i c interactions and in (5). T h e e n d o t h e r m i c nature o f h y d r o p h o b i c
determining solution free energy (26, 62). interactions causes t h e m to be stronger with
When a n o n p o l a r solute dissolves in water, it increasing t e m p e r a t u r e (62). In n o n c o v a l e n t
modifies water structure towards greater associations m o d i f y c o n f o r m a t i o n a l f r e e d o m of
" c r y s t a l l i n i t y " , i.e., the water builds an "ice- flexible regions of a protein chain, an ap-
berg" around it. H y d r o g e n b o n d i n g of water in preciable c o n t r i b u t i o n to e n t r o p y results (5).
the i m m e d i a t e n e i g h b o r h o o d of the solute is For loosely packed proteins such as caseins, this
increased over its average in pure water. Inter- is a distinct possibility.
action of n o n p o l a r groups with water is un- H y d r o p h o b i c interactions, taken individually,
favorable, and there is a t h e r m o d y n a m i c are weaker than m o s t o t h e r side chain inter-
t e n d e n c y for n o n p o l a r groups to c o n t a c t each actions of proteins. T h e y b e c o m e i m p o r t a n t

TABLE 2. Average characteristics of casein micelles (6, 43, 60, 81).

Characteristic Value Reference

Diameter 130-160 nm 6, 81
Surface 8 × 10 -1° cm 2 43
Volume 2.1 × 10 - i s cm 3 43
Density (hydrated) 1.0632 g/cm 3 43
Mass 2.2 × 10 - I s g 43
Water content 63% 43
Hydration 3.7 g H 20/g protein 6
Voluminosity 4.4 cm 3/g 6, 60
Molecular weight (hydrated) 1.3 × 109 daltons 43, 60
Molecular weight (dehydrated) 5 × 108 daltons 6, 43, 60
Number of peptide chains
(MW: 30000) 104 6, 43
Number particles per ml milk 1014 - 1016 81
Whole surface of particle 5 × 104 cm 2/ml milk 81
Mean free distance 2 4 0 nm 81

Journal of Dairy Science Vol. 67, No. 3, 1984

 CASEIN MICELLE STRUCTURE 501

TABLE 3. Size distribution characteristics of casein particles (81).

Distribution
exclusive of
particles 20 nm Total
Characteristic diameter distribution

Concentration of micelles 1.2 X 1014/ml milk 6 × 1014/mI milk

Number mean diameter 65 nm 25 nm
Volume mean diameter 104 nm 86 nm
Volume median diameter 134 nm 129 nm

because of their high frequency in proteins with Electrostatic I nteractions

nonpolar side chains (62). They exhibit less Ionic interactions contribute little to stability
specificity than other interactions both with of a monomeric protein except when an ion
respect to steric requirements of side chain pair can be formed in a hydrophobic environ-
orientation and number and kind of side chains ment (23). They are critical for close packing
that can participate in their formation. Neither and proper formation of specific aggregates.
bond angle nor length is uniquely fixed, and When specific ion pairs such as phosphate-
relative orientation of nonpolar groups in calcium-phosphate bridges in casein are formed,
contact is unimportant; also the degree to they stabilize protein quaternary structure (5).
which groups overlap varies (62). Hydrophobic Such interactions play a greater role in casein
interactions would lead to incorrect associations structure than in most other proteins, and the
in the absence of complementary pairing high content of acidic groups increases calcium
surfaces (5). When two hydrophobic protein binding capacity and enhances crosslinking.
groups establish contact, the total number of Native casein micelles can approach each
water molecules in contact with them is de- other unhindered down to atomic distances and
creased. Removal from aqueous environment is experience considerable energy barrier in their
not complete as they still retain some water approach at interparticle distances of .1 nm
neighbors (62). Hydrophobic interactions of (67). Short interactions such as hydrophobic
various strengths are possible, and several side interactions, calcium bridging, and other charge
chains, belonging to different backbone portions interactions between positively charged para-K-
may associate to form a h y d r o p h o b i c regions casein and negatively charged regions on other
from which water is excluded completely (62). micelles then become predominant.
Distribution of protein side chains between The magnitude of London-van der Waals
interior and surface affects character and forces, which depends on differences in polar-
behavior of the macromolecule (44). Ionic izability of colloidal particles and surrounding
side chains generally are exposed fully to water. medium, is relatively low for casein micelles
A combination of high h y d r o p h o b i c i t y and because of their porosity. Low electrical
large molecular weight of casein prevents repulsion between micelles is suggested further
formation of a globular structure in which by electrokinetic measurements. However, that
nonpolar groups are buried completely in the casein particles can be flocculated at their
protein interior (80). Some casein side chains isoelectric point demonstrates charge is im-
still are exposed to water, and nonpolar areas portant for their stabilization (66).
on the protein surface are available for inter- The region surrounding a particle surface can
actions with other protein molecules (23, 62). be divided into two major parts. The first
Such residues find it favorable, in terms of free consists of ions adsorbed at the surface, which
energy, to cluster with similar residues from form an inner, compact double layer, and the
other molecules. This endows caseins with second is a diffuse Gouy layer (1). The inner
pronounced tendencies to associate and form layer can be subdivided into a layer o f de-
complexes, even in the absence of Ca 2+ ions solvated, chemically absorbed, and potential
(6). determining ions and a Stern layer, containing

Journal of Dairy Science Vol. 67, No. 3, 1984

502 McMAHON AND BROWN

counterions that partially are adsorbed electro- centrifugate ( - 1 0 . 2 mV cf. - 1 5 . 5 mV at 20°C,

statically and partially solvated. Thickness of pH 6.85).
the electrical double layer on micelle surfaces in MICELLE STABILITY
milk is approximately 1 nm (66). Ions and
surrounding medium in the Stern layer are held Effect of Micelle Charge
rigidly and immobile in the sense of resisting
Hydrophobic colloids generally are stabilized
shear, whereas the outer layer is populated by a
by surface charge, and repulsion between
diffuse distribution of anions and cations.
double layers prevents their aggregation (1).
Calcium, magnesium, phosphate, and citrate
Stability of casein micelles, however, is at-
adsorbed on casein micelles can be removed by
gel filtration (8) and show three distinct zones: tributed only in part to surface charge, because
adsorbed ions, bound minerals, and casein hydrophobic colloids with ~'-potentials less than
counterions. All of the citrate, 22% of the Ca, 20 to 30 mV, such as native casein micelles, are
50% of the Mg, and 47% of the phosphate of generally unstable (17). The exact contribution
whole micelles can be removed by gel filtration. of surface charge to casein micelle stability has
The potential at the shear region in the not been determined. It is an oversimplification
diffuse Gouy layer is known as the electrokinetic to view casein micelle coagulation as just a
(or ~'-) potential. The ~'-potentials of casein reversal of the balance of repulsive and attractive
micelles immersed in a medium of the same electrostatic forces. Other more specific inter-
ionic strength as milk depend only upon the actions are involved.
surface charge density of the micelles and not Cationic materials are absorbed rapidly and
on micelle size. Fixed charges within the extensively by micelles, but subsequent decrease
micelle are neutralized by ions in the serum and of micellar charge is insufficient to allow
coagulation (33). Lysozyme is unable to induce
do not contribute to electrokinetic potential.
clotting of casein micelles even though it carries
All micelles have similar electrokinetic mobilities
the same number of positive charges (Ca. 3 ×
and, therefore, have similar surface charge
10 x7 mg - 1 ) as negative charges on casein
densities and surface chemical composition
micelles (32). Also, micelle stability decreases
regardless of size (69).
with increases of temperature, yet surface
Calculated ~'-potentials of casein micelles
charge increases (Table 4), demonstrating the
vary with both temperature and pH as in Table
involvement of hydrophobic interactions in
4, but heat treatment itself has no signifi-
micelle aggregation (16). Cationic materials
cant effect. Milk can be heated to within a few
reduce rennet clotting time by increasing both
minutes of coagulation without exhibiting any
affinity of rennet for K-casein and aggregation
consistent change of ~'-potential (16). Cal-
rate.
culations of casein micelle ~'-potentials are
Specific interactions between positively
complicated because the micelle surface is
irregular, and its double layer has an unknown charged para-K-casein on one micelle and
negatively charged groups of as1- and K-casein
structure. Ionic composition of the serum phase
also affects electrophoretic mobility (17). on another micelle also may be involved in
Casein micelles have lower mobilities electro- aggregation. Calcium binding to casein is a
phoresed in milk ultrafiltrate than in milk function of Ca 2 + activity, and on cooling, Ca 2 +
activity increases because of dissociation of
colloidal calcium phosphate (CCP) from micelles
(73). Therefore, as temperature decreases,
amount of casein-bound calcium increases and
TABLE 4. Variation of electrokinetic potential of
casein micelles with temperature and pH (16). causes decrease of net negative charge of the
micelle. Micelle charge also is reduced as pH is
Potential (mY) lowered toward the isoelectric point of the
micelle. However, casein micelles do not
pH 20°C 30°C 45°C
coagulate at pH 4.6 below 5°C even though net
charge is negligible (17). This suggests electro-
5.7 --8 -I0 -14
6.8 --13 --17 --22 static forces contribute to but do not determine
overall micelle stability.

Journal of Dairy Science Vol. 67, No. 3, 1984

 CASEIN MICELLE STRUCTURE 503

Surface Protein Interactions Dimensions of Casein Submicelles

Casein micelle stabilization also involves Casein submicelles are not of fixed dimen-
sterie interactions of surface proteins (18, 35, sions but vary in size in an equilibrium system
61). When macromolecular layers interpenetrate, with a continual interchange operating between
the increased polymer segment concentration in serum casein monomers and submicelles (12).
the interaction zone between the two particles Restrictions on interchange by calcium-phos-
leads to a local osmotic pressure effect. Macro- phate bridges between submicelles tend to lock
molecules in the interaction zone also may casein molecules into position once a micelle is
compress without interpenetration, which may formed.
be more applicable to casein micelles because of Ion beam sputtering of casein micelles
the highly hydrated and concentrated protein enables details of the underlying structure of
surface. When compression occurs, protein submicelle spheres to be observed (38). At
molecules in the interaction region lose con- energy, submicelles are sputtered off, pre-
figurational entropy on collision and provide sumably by disintegration of calcium-phosphate
repulsive energy to prevent aggregation. bonds. Etching of submicelles reveals an in-
M I C E L L E STRUCTURE frastructure of curved rod-like components
about 2 nm diameter and 7 nm in length, which
A subunit model of casein micelle was corresponds to casein complex units produced
proposed as a result of internal structure of by treatment with oxalate (58).
casein micelles observed by electron microscopy
(EM) (83). An approximately spherical a,/3,K-
casein complex of molecular weight 3 × 10 s
Nature of Casein Submicelles
daltons and 10 nm diameter was described
as the building unit (submicelle). Such a model Not only K-casein but also 0/Sl-, &S2-, and
agrees with disintegration studies of micelles /3-caseins contain highly charged regions ex-
with three stages (58): pected to be located on the submicelle surface
(76). It has been postulated that submicelles
1. Dissolution of intercomplex CCP linkages have the nonpolar portion of each protein chain
producing calcium-casein complexes less oriented radially inward, whereas charged acidic
than 30 nm diameter. residues of the calcium-sensitive caseins and the
2. Removal of intracomplex calcium forming hydrophilic carbohydrate portion of K-casein
soluble casein complex units. are located on the submicelle surface. This
3. Disaggregation to monomers occurring by results in a hydrophobic core surrounded by a
disruption of hydrogen bonds, hydro- polar layer. The K-casein is located primarily in
phobic interactions, and salt linkages. one area of the submicelle (54), and, therefore,
there are phosphate-rich and phosphate-depleted
Early models implied that submicelles contain areas on the surface as well as hydrophilic and
different proportions of K-casein but that their hydrophobic areas (64). Segregation of highly
composition did not change rapidly (87). Those charged regions from hydrophobic regions
containing a high proportion of K-casein com- produces an amphipolar structure enabling
bine together to constitute a small micelle, formation of small aggregates.
whereas large numbers of subunits, each con- By size exclusion chromatography, to sepa-
taining less K-casein, constitute larger micelles. rate submicelles from micellar casein, sub-
From experiments of size exclusion added micelles consist of complexes of asl-casein with
K-casein elutes with micelles, suggesting casein ¢3-casein or K-casein and polymers of ~Sl- and/3-
components can rearrange randomly and that casein (63). Submicelles of as1- and /3-casein,
they exist in a dynamic system (12). The size, which are sensitive to Ca 2 + ions, would form the
composition, and molecular weight of sub- core region of the micelle. Some submicelles con-
micelles is determined by casein concentration, tain only asl-casein and presumably represent a
pH, and temperature. When K-casein is added, a type of submicelle situated in the mieellar core
new equilibrium position is reached in which region (63). Submicelles containing asa- and
added K-casein appears to be interchangeable K-casein eomplexed preferentially would be
with K-casein in the complex. located on the mieelle surface.

Journal of Dairy Science Vol. 67, No. 3, 1984

504 McMAHON AND BROWN

Submicellar Aggregation micelles with a low percentage of hydrophobic

A major difference between surface areas surface area (50%) are unable to aggregate
composed of as1- or ~-caseins and those of together and are on the micelle surface (87).
K-casein is that the phosphate groups are Addition and incorporation of K-casein into
potential calcium binding sites (100). Ionic submicelles would decrease surface hydrophobic
bonds between adjacent submicelles may be interactions and favor smaller micelles. In
produced by calcium binding, and the resultant contrast, addition of asl-casein would create
decrease of net negative charge would enhance more hydrophobic areas resulting in increased
hydrophobic interactions. average micelle size (87).
A possible first step of micelle formation Accumulated strain also has been proposed
involves tetrahedral arrangement of submicelles as a means of micelIe size regulation. Each
(84). In a system containing less than 15% submicelle experiences distortion to maxi-
K-casein, further association then would occur. mize its free energy of bonding to the growing
This would produce a minimum micelle of 14 micelle (5). Submicelles added late in the
submicelles (MW ca. 3.5 x 106 ) if enough aggregation process are distorted more than
K-casein is on the surface to hinder further those added early until finally the accumulated-
growth. Less K-casein on the surface would strain free energy is greater than bonding
allow further micelle growth and yield a wide energy and aggregation ceases.
distribution of sizes
Experimentally observed distribution of MICELLE POROSITY
micelle size has been produced mathematically The porous nature of casein micelles is
(85). Submicelles with maximum concen- consistent with high water content and shrinkage
tration when K-casein content is 15% contain 4 upon vacuum drying, partial dissociation
K-casein molecules and 26 a s- or ~3-caseins. Such of casein components without alteration of
submicelles are most likely to form the micelle hydrodynamic radii, and complete access of the
interior. C-terminal of all casein molecules to carbo-
An alternate scheme for building casein xypeptidase A (CPA) (28). The CPA has
micelles from subunits is to assign different molecular dimensions of 5.4 × 4.4 x 4.0 nm
functionalities to subunits of different com- and can penetrate channels or cavities in the
positions (14). Micelle size and size distribution micelle. Its diffusion time through a micelle of
then are governed by interactions of function- 100 nm diameter is reported as 10 ~4 to 10 - s s
a~ities rather than by geometry. (74). s o m e restrictions on availability of casein
molecules to proteolysis exist because pro-
Micelle Size Regulation teolysis by chymosin is dependent on the
Micelle growth is limited by concentration at physical state of the casein.
the micelle surface of submicelles rich in Component proteins of a casein system
K-casein. As submicelles poor in K-casein become progressively less susceptible to pro-
aggregate, average composition of remaining teolysis by chymosin as degree of aggregation
submicelles changes toward a larger fraction of increases (54). Considerable interaction occurs
K-casein and approaches the fraction of K-casein between t~Sl" and/3-caseins in a "soluble" casein
in milk serum, i.e., .29 (75, 85). Submicelles system. Susceptibility of as1- and /3-casein in
with low K-casein contents are buried in the such a system is reduced 10 to 15 times com-
interior of the micelle. Potential hydrophobic pared to isolated solutions. In colloidal phos-
interactions inward and sideways at the inter- phate free (CPF) milk, susceptibility of/~casein
acting surfaces tend to align entering submicelles to proteolysis is reduced further. In going from
so their hydrophilic areas extend radially CPF milk to native skim milk, a 50-fold increase
outward. As the radius of curvature of the of particle size occurs, but/3-casein susceptibility
micelle gets larger, steric hindrance causes less to proteolysis is decreased only slightly whereas
of the surface of a given micelle to be available asl-casein shows a significant decrease of
for further interaction (87). A surface rich in susceptibility.
K-casein is reached at a smaller radius as sub- There are three chymosin susceptible bonds
micelle ~-casein content is increased. Sub- in ~3-casein (all near the C-terminal), and a s l -

Journal of Dairy Science Vol. 67, No. 3, 1984

 CASEIN MICELLE STRUCTURE 505

casein has six of its seven such bonds in a Considerable heterogeniety exists for K-casein
similar area. Simple aggregation is unlikely to as a consequence of genetic variability of the
account for this reduced susceptibility, because polypeptide chain and varying amounts of
if the C-terminal is free for reaction with CPA, carbohydrate attached to the macropeptide (2).
then the bonds susceptible to chymosin would Variations of carbohydrate content do not
be expected to be free as well. Removal of CCP affect the ability of K-casein to stabilize ~sl-
renders micellar asl-casein more accessible to casein (19), because all K-casein fractions
chymosin, suggesting submicelles are linked by separated by ion-exchange chromatography on
calcium-phosphate bridges predominantly be- the basis of carbohydrate content are able to
tween asl-casein molecules, which would stabilize asl-casein against precipitation by
restrict micelle porosity. calcium (Table 5). Furthermore, S-carboxy-
methyl-K-casein possesses micelle stabilizing
ability and rennin sensitivity yet does not
MICELLE SURFACE STRUCTURE
contain any carbohydrate (56). Casein micelles
The initial effect of renneting casein micelle that have had 99% of the total sialic acid
solutions is a decrease of viscosity and volu- removed by neuraminidase also remain stable
minosity (95). If this were a result of reduction (58, 93). As long as the macropeptide portion is
of molecular weight only, these changes would attached, K-casein retains its stabilizing ability.
be less than observed. They, therefore, indicate This does not imply that the carbohydrate is
that the micelle surface is nonsperical. Sub- devoid of function, because there i s no in-
micellar structure of casein micelles results in a dication how stable such micelles are to heat or
surface that is neither smooth nor spherical (42, pH (37).
78) but has an unevenness equivalent to sub- That renneted micelles carry a net negative
micelle radius (i.e., 5 to 10 nm). Molecular charge whereas para-K-casein is positively
chains (or "hairs") protruding from the micelle charged shows that the exterior micelle surface
surface would help to explain the high volu- is not composed solely of K-casein. In whole
minosity of casein micelles (95). Reduction of casein the ratio of (asl-casein plus fl-casein) to
viscosity could be explained by their removal as K-casein has been calculated as 5:1 (36). By
part of the macropeptide portion of K-casein. In covalently immobilizing native micelles onto
heat-treated milk most micelles observed by EM glass beads and then releasing them after
exhibit small appendages, and it has been dissociation of the micelle by urea, it has been
proposed that heat treatment causes deposition observed that the ratio on the micelle surface is
of calcium and phosphate onto these "hairs" of approximately 1:1 (36). A surface position thus
K-casein/j3-1actoglobulin complex (27, 42). is favored for K-casein in the micelle, and in
small micelles the proportion of K-casein on the
surface is even higher.
Micelle Structure and K-Casein

The dichotomous function of K-casein to

intereact hydrophobically with asl-casein and
at the same time provide a hydrophilic surface
on the micelle is due to its primary structure
(37). The N-terminal two-thirds of K-casein is TABLE 5. Variations of carbohydrate and phosphate
hydrophobic, and the C-terminal third (the content of K-casein (19). Each fraction contains one
macropeptide) is hydrophilic. A physical model phosphoserine residue per mole.
of K-casein consists of a fairly rigid globular
structure, comprising two-thirds of the mol- moles/mole protein
ecule, to which a flexible, highly solvated tail is Fraction Sialic acid Hexose Hexosamine
attached (89). The hydrophobic para-K-casein
portion of the molecule could serve to anchor P2 0 0 0
the outwardly directed hydrophilic macro- P3 0 1 1
peptide portion to the casein micelle and P4 1 2 1
P5 2 3 2
endow it with a negatively charged, solvated P6 3 4 3
surface envelope (37).

Journal of Dairy Science Vol. 67, No. 3, 1984

506 McMAHON AND BROWN

Seventy-five percent of micellar sialic acid is CASEIN ASSOCIATIONS

on the micelle surface and accessible to im-
Caseins are strongly interacting proteins and
mobilized neuraminidase (58). Investigations
do not exist as monomers at physiological
using stains specific for glycosylated K-casein
conditions (90). Different mechanisms of
indicate the glycosylated K-casein is located
association between casein molecules occur.
mainly in the outer layers of large micelles,
For SH-K-casein either electrostatic or hydro-
whereas for small micelles it is distributed
phobic interaction can predominant under
evenly throughout (49). This implies that
various conditions (87). When as1 - and ~-casein
aggregation or interaction of K-casein with
are present, intermolecular hydrophobic inter-
other casein componennts may prevent addition
actions predominant.
of carbohydrate to all molecules (56). Oxidation of SH-K-casein in native micelles
Degree of aggregation does not affect access leads to covalently bound polymers of K-casein
of enzyme to the chymosin sensitive bond in (87, 88), suggesting K-caseins are adjacent to
K-casein. Skim milk and CPF milk (dissociated each other in micelles. Disulfide bonds may
micelles) show equivalent proteolysis by contribute to, but are not essential for, K-casein
chymosin (9). Investigations using pepsin- self-association (94) or C~sl-casein stabilization
dextran complexes (average Stokes radii 5, 8.5, (79, 91, 99). Electrostatic attraction between
and 11.5 nm) show K-casein is more accessible K-casein molecules results in discrete hydrophilic
for proteolysis in casein micelles than when surface areas on submicelles. The remainder of
micelles are dissociated or solubilized (10). It is the surface consists mainly of C~sl- and/3-caseins
perhaps in a more readily accessible conforma- that would be available for hydrophobic
tion, and also relatively little asl - or/3-casein is interactions.
available for binding to the enzyme. In com- Complex formation between as1- and
parison, as1- and /3-caseins are almost inac- K-casein is important in micelle formation.
cessible to proteolysis in native micelles but are Growth of miceUes from as1 , K-casein com-
rendered accessible upon removal of CCP (24). plexes in the presence of Ca 2+ depends on
interaction between Ca 2+ ions and phos-
Micelle Structure and ~Casein phoserine groups of asl-casein (71). Hydro-
dynamic radii for casein polymers and their
Relatively high proteolytic susceptibility of
hydrophobic surface area have been calculated
some of the micellar /3-casein is due to its
(Table 6). Upon complex formation the exposed
dissociability from the micelle. Average hydro-
hydrophobic surfaces of asl-casein polymers
phobicity of ~3-casein is greatest of all the
are buried in the complex interior (20). This
caseins, so it shows the largest effect of temper-
shows that as1, K-casein complexes are formed
ature on solubility (3). Part of the /3-casein is
as a result of both hydrophobic and electrostatic
associated only loosely with the micelle (by
interactions.
hydrophobic interactions) and reversibly dis-
sociates from it at low temperatures (7, 70).
Stabilization of Ors1-and ~-Caseins
The remainder (50%) is relatively firmly "fixed"
within the micelle and not free to dissociate. Gelatin is only 5% as effective as K-casein for
An intermediate role in binding of/3-casein stabilizing •sx- and ~-casein, and more gelatin
to the micelle is played by CCP (68). The than casein is required, indicating stabilization
amount of colloidal calcium removed from occurs by a different mechanism compared
the micelle by cooling affects release of/3-casein with K-casein. Stabilization by K-casein is a
(70). Solubilization of CCP may result in cooperative phenomenon, and protection of a
calcium-phosphate bridges between /3-casein proportion of as1- or fi-easein facilitates pro-
and the micelle being broken. This increases the tection of the rest (31). It involves specific
amount of/3-casein bound solely by hydrophobic interactions between K-casein and as1- or
interactions and shifts equilibrium towards 13-casein, primarily at hydrophobic regions, and
dissociation. When mineral migration is pre- probably involves definite sites in each molecule
vented, the /3-casein released during cooling is (30).
that bound to the micelle by hydrophobic Complex formation between K-casein and
interactions only (70). as1- or /3-casein does not occur in the absence

Journal of Dairy Science Vol. 67, No. 3, 1984

 CASEIN MICELLE STRUCTURE 507

TABLE 6. Molecular weight and size of casein polymers (20).

Molecular Spherical Nonspherical Hydrophobic

Complex weight radii radii surface area

(nm) (nm 2)
asl-polymers 10 X 104 3.0 ... 198
K-polymers 78 X 104 6.1 6.6 357
as1 :K-complex 45 X 104 5.1 5.6 299

of Ca 2+ under the same conditions as in the condensation polymerization (13, 14). The first
presence of Ca 2+ (30). Interactions can occur two Ca 2+ ions bind to phosphoserine groups
directly by Ca-links, and Ca 2+ binding can and subsequent Ca 2+ ions bind to carboxylate
neutralize casein negative charges, enabling groups (13). The number average functionality
nonionic interactions to occur. Ionic inter- of the particles is close to two, corresponding
actions are important in maintaining maximum to between 4 to 10 Ca 2+ ions bound per
stabilization but are not essential for stabiliza- asl-casein molecule (14).
tion (30). In the absence of Ca 2+, hydrophobic
interactions are perhaps not strong enough to ROLE OF CASEIN PHOSPHATE ESTERS
prevent solvation of the casein molecules, b u t There are eight phosphoserine residues in
when Ca 2+ is added, caseins associate. Inter- asl-casein. The consequences of dephos-
action of nonpolar side groups and strong phorylation are.. 1) Ca 2+ induced precipitation
hydrogen bonds are major factors influencing occurs via nonphosphate Ca 2+ binding sites
complex formation, but growth of micelles (22). 2) Ability to be stabilized by K-casein in
from as1 , K-casein complexes in the presence of the presence of Ca 2+ is impaired but not
Ca 2 + depends on interaction between Ca 2 + and eliminated (22, 71). 3) Incorporation into
esterified phosphate residues of asl-casein micelles is decreased (97). 4) Micelle formation
molecules (71). and stability are reduced (4). 5 ) a s 1 , K-Casein
micelles are disrupted, indicating a minimum
Calcium-Induced Association of 0tsl-Casein number of phosphate groups is necessary for
Self association of asl-casein is mainly micelles to remain dispersed in their medium
electrostatic (15, 54). Binding of Ca 2+ to casein (97). 6) asl-Casein is shifted from the micelle
is important in micelle synthesis because phase to an insoluble phases, i.e., it precipitates
it renders 0~sx- and H-caseins liable to pre- (97).
cipitation (15). Critical Ca ~+ ion binding exists Chemical phosphorylation of K-casein impairs
at which level precipitation begins (13). A t its stabilizing ability (71). Low phosphate
Ca 2+ concentrations lower than the critical content for K-casein is thus a requisite for
binding, asl-casein is fully soluble whereas formation of stable micelles and explains why
above it solubility decreases sharply (14). specific phosphorylation of casein components
In the absence of phosphate, asl-casein is in the mammary gland is required. Native
precipitated between 6 and 7mM Ca 2+. A t low /3-casein has five phosphoserine groups and like
phosphate, aggregation is induced at 2raM K-casein is amphipolar, but it does not stabilize
Ca 2+, and at intermediate phosphate the asl-casein. Conversely, dephosphorylated ~3-
system can be stabilized as a colloid (40, 41). casein is not only soluble in the presence of
Precipitation occurs at higher phosphate be- Ca 2+ but also stabilizes asl-casein. Thus, three
cause there is insufficient asl-casein to maintain properties of a protein required to stabilize
stability of the colloidal coprecipitate. asl-casein are (100): 1) a hydrophobic region
Binding of Ca 2+ ions to casein creates units as site for interaction with asl-casein, 2) a
analogous to monomers in a polyfunctional hydrophilic region as site for interaction with

Journal of Dairy Science Vol. 67, No. 3, 1984

508 McMAHON AND BROWN

water, and 3) a low phosphate content. De- (PO4)2.CaHcitrate- (72). Numerous types of
phosphorylation of/3-casein reduces the rate at linkages are possible between CCP and casein
which 13-casein self associates but does not and within the CCP network (Table 7).
affect the final association (101). Serum casein content can be increased by
addition of phosphate or acid or by removal of
C O L L O I D A L CALCIUM PHOSPHATE CCP (75). Removal of calcium from micelles up
Calcium is in milk in a number of forms (6). to critical Ca 2 + causes no reduction of hydro-
Of the 32 mM calcium in skim milk 22 mM is in dynamic radius (7). As Ca 2+ activity is reduced
the colloidal state, and 10 mM is soluble, of below critical, molecular weight of the micelles
which only 3 mM is free ionic Ca 2+ (7). In the decreases as a result of dissociation of caseins,
serum phase it can exist as free hydrated ion or predominantly /3- and K-caseins (7). Further
complexed with citrate, phosphate, or serum removal of Ca 2+ causes dissociation of the
proteins. In the colloidal state it can be com- micelle. Addition of Ca 2+ initially causes
plexed with phosphate ester and carboxyl transfer of soluble casein to the micelle without
groups of micellar casein or it can be com- changing hydrodynamic radius. Further addition
plexed with phosphate (and perhaps citrate) causes formation of increasingly larger casein
associated with the casein micelle (55, 65, 80). micelles (7).
Skim milk contains 30 mM phosphate anions Two distinct forms of ions are associated
dispersed as: 19 mM colloidal, 5 mM free, and 6 with casein micelles, an outer system readily
mA4 bound to calcium. Of the 8.4 mM citrate .4 removed and an inner system resistant to
mM is incorporated into the casein micelle (7). removal (22). Removal of adsorbed ions does
Calcium is removed only slowly by dialysis not destroy integrity of the micelle. Forty
of skim milk (25) suggesting that nearly all the percent of micellar calcium is "hard-to-ex-
calcium, including "ionized" calcium, is at least change" and is part of the CCP (65). Sub-
loosely associated with non dialyzable con- stitution of calcium with Ba, Pb, or Ur followed
stituents. Integrity and structure of micelles by EM observations has shown that Ca is
remain essentially constant after dialysis for 24 not distributed homogenously within the casein
h with little change of CCP. This suggests that micelle but is located in a network system (45).
its rate of solution is slow and not in true Dye penetration studies also have shown that
equilibrium with the serum environment (25). the micelle is constructed rather loosely, and
Calorimetric studies of casein micelle systems this calcium network corresponds to connecting
suggest that CCP more closely resembles interstites between submicelles (45). At the pH
OH-apatite than amorphous calcium phos- of milk, CCP is insoluble, and casein prevents it
phate or any other crystalline form (52). It can from being precipitated. Early stages of micelle
be represented by an empirical formula of 3Ca3 formation may be promoted by CCP spatially
arranging submicelles (29) or by acting as
a nucleating agent (78).
TABLE 7. Types of linkages possible in colloidal Rapid mineral exchange occurs in micelles
calcium phosphate (59). XX = carboxylate or phos- formed without phosphate and leads to micelle
phoserine group. instability (13). Mineral exchanges are controlled
by salt solubility (70). Calcium phosphates are
/ Ca \ in saturated solution in the soluble phase of
Casein - x x - (Ca-PO4 p o , ) n -- Ca -- XX - Casein milk, and phosphate solubility in normal milk
/
\Ca at 37°C is limited by the saturation equilibrium
--H (70). An important function of calcium phos-)
-- CaOH
phate may be to slow exchange of calcium
-- CA -- PO 4 = Ca
between micellar phase and serum. Calcium
Casein - XX - Ca - PO~ - Ca - XX - Casein phosphates are slow to equilibrate, and both
Ca - Citrate = Ca Mg 2+ and pyrophosphates inhibit the crys-
Casein - XX - Ca - PrO4 - Ca - XX - Casein tallization of OH-apatite, indicating milk may
/
CaOH not be in true equilibrium with the solid phase
(52).

Journal of Dairy Science Vol. 67, No. 3, 1984

 CASEIN MICELLE STRUCTURE 509

MICELLE SYNTHESIS major caseins (O~Sl , ~, and to). Properties of

artificial micelles depend on the way con-
Micelle Bioassem bly stituents are allowed to react (78, 81). One
Casein polypeptide synthesis occurs in the method starts with a salt solution of calcium,
rough endoplasmic reticulum (ER) of epithelial phosphate, and citrate milk serum concen-
cells. Addition of prosthetic groups and bio- trations (46). Casein and additional calcium,
assembly takes place in the Golgi apparatus phosphate, and citrate then are added at 15-rain
(GA) in preparation for cell export (50). There intervals. All the casein forms micelles within
appears to be a specific chronological order in the range 30 to 400 nm diameter with typical
which micellar components enter the GA and in submicelle structure. Another method starts
which phosphorylation and aggregation occur with sodium caseinate (pH 6.7, 37°C), and
(41). Globular particles I 0 to 20 nm in diameter inorganic constituents are added simultaneously
initially appear in newly formed Golgi vesicles in three separate solutions: calcium and mag-
(GV). These globular particles eventually nesium chlorides, sodium and potassium
aggregate to form individual casein micelles, phosphates, and potassium citrate. Volume
and only one micelle appears to be synthesized distribution closest to native micelles is obtained
per GV. Size of casein micelles ultimately may with ratios of asl:/3:K-casein of 6:2:1 (78).
be limited by size of GV (57). The GV con- In the presence of inorganic phosphate,
taining milk constituents bud off from the GA, addition of Ca 2 + initially causes formation and
move toward the cell apical portion, and growth of casein polymer to submicelle size and
discharge their contents into the glandular then promotes micelle formation (86). Once
lumen. A pool of unphosphorylated casein in submicellar size is reached, additional Ca 2+
mammary cells suggests phosphorylation of disrupts intrasubmicellar salt bridges, and aided
serine and threonine side chains following by inter-submicellar hydrophobic interactions,
polypeptide synthesis but before micelle forms longer chains or bridges between sub-
assembly (50). micelles. This occurs at the point where cal-
Micelle formation could begin under con- cium-sensitive caseins precipitate and argues for
ditions in which a coUoidally stable casein- their interaction in micelle formation with
phosphate-calcium coprecipitate, forming in K-casein providing stabilization by limiting the
the GA as phosphorylation is completed (41). precipitation process (78).
Bioassembly of casein micelles can be sum-
ACKNOWLEDGMENTS
marized as (22, 50): 1) After protein syn-
thesis on ribosomes, casein polypeptides The authors thank C. A. Ernstrom for
interact to form subunit particles composed of critical reading of the manuscript.
several monomers. 2) Upon reaching the GA,
these subunits are phosphorylated and aggregate REFERENCES
to form particles of about 10 to 20 nm diameter.
3) Addition of Ca 2+ (which is incorporated 1 Adamson, A. W. 1967. Physical chemistry of
surfaces. Ch. 4 in Electrical aspects of surface
into milk in the GA) initiates polymerization of chemistry. A. W. Adamson, ed. Wiley-Interscience,
these particles into partially condensed micelles. New York, NY.
4) Deposition of CCP yields the mature casein 2 Armstrong, C. E., A. G. Mackinlay, R. J. Hill, and
miceUe. R. G. Wake. 1967. The action of rennin on
K-casein. The heterogeniety and origin of the
Some newly formed GV contain long
soluble product. Biochim. Biophys. Acta 140:109.
stringy, protein filaments (50), and some are 3 Bingham, E. W. 1971. Influence of temperature
filled completely by loosely arranged sub- and pH on the solubility of C~sl-,/3- and K-casein.
micelles (34). More mature vesicles contain J. Dairy Sci. 54:1077.
casein micelles similar to those in secreted milk 4 Bingham, E. W., H. M. Farrell, Jr., and R. J.
Carroll. 1972. Properties of dephosphorylated
in the lumen. asl-casein. Precipitation by-calcium ions and
micelle formation. Biochemistry 11:2450.
Artificial Casein Micelle Synthesis 5 Bloomfield, V. A. 1979. Association of proteins.
J. Dairy Res. 46:241.
Artificial synthesis of casein micelles requires 6 Bloomfield, V. A., and R. J. Mead, Jr. 1975.
calcium, phosphate, and citrate as well as the Structure and stability of casein micelles. J. Dairy

Journal of Dairy Science Vol. 67, No. 3, 1984

510 McMAHON AND BROWN

Sci. 50: 592. 25 Fox, P. F., and B. M. Nash. 1979. Physico-

7 Bloomfield, V. A., and C. V. Morr. 1973. Struc- chemical characteristics of casein micelles in
ture of casein micelles: physical methods. Nether- dilute aqueous media. J. Dairy Res. 46:357.
lands Milk Dairy J. 27:103. 26 Frank, H. S., and M. W. Evans. 1945. Free
8 Boulet, M., A. Yang, and R. R. Riel. 1970. volume and entropy in condensed systems. 3.
Examination of the mineral composition of the Entropy in binding liquid mixtures; partial molal
micelles of milk by gel filtration. Can. J. Biochem. entropy in dilute solutions; structure and thermo-
48:817. dynamics in aqueous electrolytes. J. Chem. Phys.
9 Chaplin, B., and M. L. Green. 1980. Determination 13:478.
of the proportion of K-casein hydrolysed by 27 Freeman, N. W., and M. E. Mangino. 1981.
rennet on coagulation of milk. J. Dairy Res. Effects of ultra-high temperature processing on
47:351. size and appearance of casein miceIles in bovine
10 Chaplin, B., and M. L. Green. 1981. The rate of milk. J. Dairy Sci. 64:1772.
hydrolysis of x-casein in various complexes by 28 Garnier, J. 1973. Models of casein micelle struc-
pepsin and soluble dextran-pepsin conjugates. ture. Neth. Milk Dairy J. 27:240.
Neth. Milk Dairy J. 35:377. 29 Garnier, J., and B. Riabadean Dumas. 1970.
11 Cheeseman, G. C., and W. Manson. 1982. Milk Structure of the casein micelle. A proposed
proteins: molecular, colloidal and functional model. J. Dairy Res. 37:493.
properties. J. Dairy Res. 49:679. 30 Green, M. L. 1971. The specificity for K-casein as
12 Creamer, L. K., and G. P. Berry. 1975. A study the stabilizer of as-Casein and 13-casein. 1. Re-
of the properties of dissociated bovine casein placement of K-casein by other proteins. J. Dairy
micelles. J. Dairy Res. 42:169. Res. 38:9.
13 Dalgleish, D. G. 1978. Recent advances in the 31 Green, M. L. 1971. The specificity for K-casein as
physical chemistry of milk proteins. Proc. 20th the stabilizer of as-casein and /3 casein. 2. Re-
Int. Dairy Congr. (Paris), 74ST. placement of K-casein by detergents and water
14 Dalgleish, D. G., and T. G. Parker. 1979. Quantita- soluble polymers. J. Dairy Res. 38: 25.
tion of asl-casein aggregation by the use of 32 Green, M. L. 1973. Studies on the mechanism of
polyfunctional models. J. Dairy Res. 46:259. clotting of milk. Neth. Milk Dairy J. 27:278.
15 Dalgleish, D. G., and T. G. Parker. 1980. Binding 33 Green, M. L., and R. J. Marshall. 1977. The
of calcium ions to bovine ast-casein and pre- acceleration by cationic materials of the co-
cipitability of the protein-calcium ion complexes. agulation of casein micelles. J. Dairy Res. 44: 521.
J. Dairy Res. 47:113. 34 Helminen, H. J., and J.L.E. Ericsson. Studies on
16 Darling, D. F., and J. Dickson. 1979. The deter- mammary gland involution. 1. On the ultra-
mination of the ~'-potential of casein micelles. J. structure of the lactating mammary gland. J.
Dairy Res. 46: 329. Ultrastruct. Res. 25:193.
17 Darling, D. F., and J. Dickson. 1979. Electro- 35 Hesselink, F.Th., A. Vrij, and J.Th.G. Overbeek.
phoretic mobility of casein micelles. J. Dairy Res. 1971. On the theory of the stabilization of
46:441. dispersions by adsorbed macromolecules. 2.
18 Darling, D. F., and A.C.M. Van Hooydonk. 1981. Interaction between two flat surfaces. J. Phys.
Derivation of a mathematical model for the Chem. 75:2094.
mechanism of casein micelle coagulation by 36 Heth, A. A., and H. E. Swaisgood. 1982. Ex-
rennet. J. Dairy Res. 48:189. amination of casein micelle structure by a method
19 Doi, H., F. Ibuki, and M. Kanamori. 1979. for reversible covalent immobilization. J. Dairy
Heterogeneity of reduced bovine K-casein. J. Sci. 65:2047.
Dairy Sci. 62:195. 37 Hill, R. J., and R. G. Wake. 1969. Amphiphile
20 Dosako, S., S. Kaminogawa, S. Taneya, and K. nature of K-casein as the basis for its micelle
Yananchi. 1980. Hydrophobic surface areas and stabilizing property. Nature 221:635.
net charge of as1 -, K-casein and as1 -casein: K-casein 38 Hojou, K., T. Oikawa, K. Kanaya, T. Kimura, and
complex. J. Dairy Res. 47:123. K. Adachi. 1977. Some applications of ion beam
21 Ernstrom, C. A., and N. P. Wong. 1974. Milk- sputtering to high resolution electron microscopy.
clotting enzymes and cheese chemistry. Page 703 Micron 8:151.
in Fundamentals of dairy chemistry. B. H. Webb, 39 Holt, C., D. G. Dalgleish, and T. G. Parker. 1973.
A. H. Johnson, and J. A. Alford, ed. 2nd ed. Avi Particle size distribution in skim milk. Biochim.
Publ. Co., Westport, CT. Biophys. Acta 328:428.
22 Farrell, H. M., Jr. 1973. Models for casein micelle 40 Home, D. S. 1979. The kinetics of the pre-
formation. J. Dairy Sci. 56:1195. cipitation of chemically modified as1-casein. J.
23 Farrell, H. M., Jr., and M. P. Thompson. 1974. Dairy Res. 46:265.
Physical equilbria: Proteins. Page 456 in Fun- 41 Horne, D. S. 1982. Calcium-induced precipitation
damentals of dairy chemistry. B. H. Webb, A. M. of as~-casein: effect of inclusion of citrate or
Johnson, and J. A. Atford, ed. 2nd ed. Avi Publ. phosphate. J. Dairy Res. 49:107.
Co., Westport, CT. 42 Kalab, M., B. E. Phipps-Todd, and P. Allan-Wojtas.
24 Fox, P. F. 1970. Influence of aggregation on the 1982. Milk gel structure XIII. Rotary shadowing
susceptibility of caseins to proteolysis. J. Dairy of casein micelles for electron microscopy.
Res. 37:173. Milchwissenschaft 37: 513.

Journal of Dairy Science Vol. 67, No. 3, 1984

 CASEIN MICELLE STRUCTURE 511

43 Kirchmeier, O. 1973. Arrangement of com- tionated bovine milk caseins. Micelles from
ponents, electrical charge and interaction en- analytical ultracentrifugation and diffusion mea-
ergies of casein micelles. Neth. Milk Dairy J. surements. J. Dairy Sci. 56:415.
27:191. 61 Napper, D. H. 1977. Steric stabilization. Science
44 Klotz, I. M. 1970. Comparison of molecular 58:390,
structures of proteins: helix content; distribution 62 Nemethy, G., and H. A. Scheraga. 1962. The
of apolar residues. Arch. Biochem. Biophys. structure of water and hydrophobic bonding in
138:704. proteins. 3. The thermodynamic properties of
45 Knoop, A. M., E. Knoop, and A. Wiechen. 1973. hydrophobic bonds in proteins. J. Phys. Chem.
Electron microscopical investigations on the 66:1773.
structure of casein micelles. Neth. Milk Dairy J. 63 Ono, T., S. Odagiri, and T. Takagi. 1983. Separa-
27:121. tion of the submicelles from micellar casein by
46 Knoop, A. M., E. Knoop, and A. Wiechem. 1979. high performance gel chromatograph on a TSK-
Sub-structure of synthetic casein miceltes. J. GEL 94000SW column. J. Dairy Res. 50:37.
Dairy Res. 46:357. 64 Parker, T. G., and D. G. Dalgleish. 1981. Binding
47 Korolczuk, J. 1981. Voluminosity and viscosity of calcium ions to bovine/3-casein. J. Dairy Res.
of casein solutions. 1. The correlation between 37:259.
voluminosity, protein concentration and viscosity. 65 Parry, R. M. 1974. Milk coagulation and protein
Milchwissenschaft 36:414. denaturation. Page 603 in Fundamentals of dairy
48 Korolczuk, J. 1981. Voluminosity and viscosity chemistry. B. H. Webb, A. H. Johnson and J. A.
of casein solutions. 2. The correlation between Alford, ed. 2nd ed. Avi Publ. Co., Westport, CT.
voluminosity and the empirical constants of 66 Payens, T.A.J. 1966. Association of caseins and
regression equations. Milchwissenschaft 36:467. their possible relation to structure of the casein
49 Kudo, S., S. Iwata, and M. Mada. 1979. An micelle. J. Dairy Sci. 49:1317.
electron microscopic study of the location of 67 Payens, T.A.J. 1977. On enzymatic clotting
~-dasein in casein micelles by periodic-silver processes. 2. The colloidal instability of chymosin-
methenamine staining. J. Dairy Sci. 62:916. treated casein micelles. Biophys. Chem. 6:263.
50 Larson, B. L. 1979. Biosynthesis and secretion of 68 Payens, T.A.J. 1979. Casein micelles: the colloid-
milk proteins: a review. J. Dairy Res. 46:161. chemical approach. J. Dairy Res. 46:291.
51 Lin, S.H.C., R. K. Dewan, V. A. Bloomfield, and 69 Pearce, K. N. 1976. Moving boundary electro-
C. V. Morr. 1971. Inelastic light-scattering study phoresis of native and rennet-treated casein
of the size distribution of bovine milk casein micelles. J. Dairy Res. 43:27.
micelles. Biochemistry I0:4788. 70 Peirre, A., and G. Brule. 1981. Mineral and
52 Lyster, R.L.J. 1979. The equilibrium of calcium protein equilibrium between the colloidal and
and phosphate ions with the micellar calcium soluble phases of milk at low temperature. J.
phosphate in cow's milk. J. Dairy Res. 46:343. Dairy Res. 48:417.
53 McGann, T.C.A., W. J. Donnelly, R. D. Kearney, 71 Pepper, L., and M. P. Thompson. 1963. De-
and W. Buchheim. 1980. Composition and size phosphorylation of as,- and K-casein and its
distribution of bovine casein micelles. Biochim. effect on micelle stability in the K-c~-casein
Biophys. Acta 630:261. system. J. Dairy Sci. 46:764.
72 Pyne, G. T., and T.C.A. McGann. 1960. The
54 McGann, T.C.A., and P. F. Fox. 1973. Casein
colloidal phosphate of milk. II. Influence of
micelle structure: susceptibility of various casein
systems to proteolysis. J. Dairy Res. 40:299. citrate. J. Dairy Res. 27:13.
73 Qvist, K. B. 1979. Reestablishment of the original
55 McGann, T.C.A., and P. F. Fox. 1974. Physico-
rennetability of milk after coding. 1. The effect
chemical properties of casein micelles reformed
of cooling and LTST pasteurization of milk and
from urea-treated milk. J. Dairy Res. 41:269.
renneting. Milchwissenschaft 34:467.
56 MacKinlay, A. G., and R. G. Wake. 1965. Frac- 74 Ribadeau Dumas, B., and J. Garnier. 1970.
tionation of S-carboxymethyl-~-casein and char- Structure of the casein micelle. J. Dairy Res.
acterization of the components. Biochim. Bio-
37:269.
phys. Acta 104:167. 75 Rose, D. 1968. Relation between micellar and
57 Marshall, R. J., and M. L. Green. 1980. The serum casein in bovine milk. J. Dairy Sci. 51 : 1897.
effect of chemical structure of additives on the 76 Schmidt, D. G. 1980. Colloidal aspects of casein.
coagulation of casein micelle suspensions by Neth. Milk Dairy J. 34:42.
rennet. J. Dairy Res. 47:359. 77 Schmidt, D. G. 1982. Association of caseins and
58 Mehaia, M. A., and M. Cheryan. 1983. Treatment casein micelle structure. Ch. 2. in Developments
of casein micelles with soluble and immobilized in dairy chemistry. 1. Proteins. P. F. Fox, ed.
neuraminidase: implications in structure of the Appl. Sci. Publ., London.
micelle. J. Dairy Sci. 66: 390. 78 Schmidt, D. G., J. Koops, and D. Westerbeek.
59 Morr, C. V., V. Josephson, R. Jenness, and P. B. 1977. Properties of artificial casein micelles. 1.
Manning. 1971. Composition and properties of Preparation, size distribution and composition.
submicellar casein complexes in colloidal phos- Neth. Milk Dairy J. 31:328.
phate-free milk. J. Dairy Sci. 54:1555. 79 Schmidt, D. G., and B. W. Van Markwijk. 1968.
60 Mort, C. V., S.H.C. Lin, R. K. Dewan, and V. A. Further studies on the associating subunit in
Bloomfield. 1973. Molecular weights of frac- sl-Casein. Biochim. Biophys. Acta 154:613.

Journal of Dairy Science Vol. 67, No. 3, 1984

512 McMAHON AND BROWN

80 Schmidt, D. G., andT.A.J. Payens. 1976. Mice/far Association of s-casein C in the alkaline pH range.
aspects of casein. Page 165 in Surface and colloid Arch. Biochem. Biophys. 125:344.
science. Vol 9. E. Matijevic, ed. Wiley-Interscience, 92 Thompson, M. P., and H. M. Farrell, Jr. 1973.
Amsterdam. The casein micelle - the forces contributing to
81 Schmidt, D. G., C. A. van der Spek, W. Bucheim, its integrity. Neth. Milk Dairy J. 27:220.
and A. Hinz. 1974. On the formation of artificial 93 Thompson, M. P., and L. Pepper. 1962. Effect of
casein micelles. Milchwissenschaft 29:455. neuraminidase on kappa-casein. J. Dairy Sci.
82 Schmidt, D. G., P. Walstra, and W. Buchheim. 45 : 794.
1973. The size distribution o f casein micelles in 94 Vreeman, H. J. 1979. The association of bovine
cows milk. Neth. Milk Dairy J. 27:128. SH-K-casein at pH 7.0. J. Dairy Res. 46:271.
83 Shimmin, P. D., and R. D. Hill. An electron 95 Walstra, P. The voluminosity of bovine casein
microscope study of the internal structure of micelles and some of its implications. J. Dairy
casein micelles. J. Dairy Res. 31:285. Res. 46:317.
84 Slattery, C. W. 1976. Review: casein micelle 96 Waugh, D. F., and B. Talbot. 1971. Equilibrium
structure, an examination of models. J. Dairy Sci. casein micelle systems. Biochemistry 10:4153.
59:1547. 97 Whikehart, D. R., and G. W. Rafter. 1970.
85 Slattery, C. W. 1977. Mode/calculations of casein Effects of varying protein to protein-phosphate
micelle size distributions. Biophys. Chem. 6:59. ratio of as-Casein on as-K-casein micelles. J. Dairy
86 Slattery, C. W. 1979. A phosphate-induced Sci. 53:1171.
sub-micelle-micelle equilibrium in reconstituted 98 Whitney, R. McL., J. R. Brunner, K. E. Ebner, H.
casein micelle systems. J. Dairy Res. 46:253. M. Farrell, Jr., R. V. Josphson, C. V. Morr, and
87 Slattery, C. W., and R. Evard. 1973. A model for H. E. Swaisgood. 1976. Nomenclature of the
the formation and structure of casein micelles protein of cow's milk: a review. J. Dairy Sei.
from subunits of variable composition. Biochim. 59:785.
Biophys. Acta 317:529. 99 Woychik, J. K. 1965. Preparation and properties
88 Swaisgood, H. E., J. R. Brunner, and H. A. of reduced K-casein. Arch. Biochem. Biophys.
Lillevik. 1964. Physical parameters of -casein 109:542.
from cow's milk. Biochemistry 3 : 1616. 100 Yoshikawa, M., R. Sasaki, and H. Chiba. 1981.
89 Swaisgood, H. E. 1973. The caseins. CRC Crit. Effects of chemical phosphorylation of bovine
Rev. Food Technol. 3 : 375. casein components on the properties related to
90 Swaisgood, H. E. 1982. Chemistry of milk casein micelle formation. Agric. Biol. Chem.
protein. Chap. 1 in Developments in dairy 45:909.
chemistry. I. Proteins. P. F. Fox, ed. Appl. 101 Vun, S., K. Ohmiya, and S. Shimizu. 1982. Role
Sci. Publ., London. of the phosphoryl group of B-casein in milk
91 Swaisgood, H. E., and S. N. Timasheff. 1967. curdling. Agric. Biol. Chem. 46:1505.

Journal of Dairy Science Vol. 67, No. 3, 1984