Casein Interactions: Does the Chemistry Really Matter? - Food Colloids: Self-Assembly and Material Science - page 163

average of the number of cluster motifs is thus 1.5 per protein molecule. Or,
alternatively expressed, there are 6 cluster motifs available for every 4 casein
molecules.
The anhydrous composition of the casein is usually taken as 93% protein and
7% mineral ash. 11 Arguments exist as to the amounts of citrate or magnesium
found in this mineral component, but here it is assumed to represent calcium
phosphate. Thus, for every 100 kDa of molecular weight in the casein micelle,
93 kDa represents protein and 7 kDa represents calcium phosphate. We note
that 93 kDa is approximately the combined molecular weight of 4 casein
molecules, which, as calculated above, will have 6 phosphoserine cluster motifs
available for association with the 7 kDa of calcium phosphate. Analysis of the
mineral content of the casein micelle shows that the ratio of calcium to
inorganic phosphate is 1.5, close to that of the hydroxyapatite salt. Including

the organic ester phosphate in the analysis, the ratio drops to

B

1.0, close to the
di-calcium salt figure. 13 This provides us with the necessary information to
delineate the average composition of the minimum sized nanocluster. So, for
every cluster motif of 4 ester phosphates, there are 8 inorganic phosphate
anions and 12 matching calcium ions associated with the motif. The inorganic
portion here has a molecular weight of 1.24 kDa. The available 7 kDa is thus 'in
balance' with 5 or 6 of the available phosphoserine cluster motifs, indicating
that these are probably satiated in their associations with the calcium phos-
phate nanocluster.
Though not considered to be adsorbed to the surface of the nanocluster, the
phosphoserine cluster motifs control the surface area of the nanocluster by
contributing an outer layer of phosphate groups. Considering these as a 2 2
rhombus [Figure 6(b)], we could then have the 8 inorganic phosphates required
by stoichiometry forming a further two layers inwards, where (unless we have a
hollow shell) they must meet the phosphates proceeding inwards from the
opposite face of the nanocluster. Extending such groupings sideways in two
dimensions would give platelets whose existence has never been reported. What
this stoichiometry is really limiting, however, is the size of the nanocluster. The
number of phosphoserine cluster motifs controls the surface area; the stoic-
hiometry dictates the volume; the two together dictate the minimum size of the
nanocluster, but only if all phosphoserine cluster motifs are involved and
associated with nanoclusters.
There has been some controversy over the physical state of the calcium
phosphate - whether it is crystalline or amorphous. While current opinion
seems to favour the latter, 18 one might reasonably ask what is actually meant
by 'amorphous calcium phosphate'? It seems to be generally considered as
small clumps of calcium and phosphate (up to possibly 9 pairings), the clumps
being arranged in the solid in a random haphazard fashion with no regular
long-range periodic structure and no fingerprint radial distribution function
with distinct spectral lines. This is thought also to be the situation for micellar
calcium phosphate, but is the case clear-cut? While having close-range ordered
structure, the small microcrystallite cannot have a long-range extended perio-
dicity, as is the case for macroscopic crystal, because any spatial structure it

B