Chemistry Reference
In-Depth Information
different ways. Thus, while Hanson and Eyre (1996) found a predominant participation of
α
1(I)K(OHK)
9N
2(I)K(OHK)
5N
in the pyrrole cross-links, Brady and Robins (2001)
×α
1(I)K(OHK)
9N
was the predominant form.
In contrast to the pyrroles, the pyridinoline based cross-links are mainly derived
from the C-telopeptide regions (Yamauchi et al. 1981, 1988, 1989,1992; Katz et al. 1989,
1992; Otsubo et al. 1992). Virtually all of the available K and OHK in the C-telopeptides
were found to be quantitatively converted to the aldehyde form, and these aldehydes
stoichiometrically cross-linked to K87 on both
1(I)K(OHK)
9N
reported that
α
×α
2 chains of adjacent molecules.
Moreover, the telopeptide K in bone was completely hydroxylated, present as OHK.
However, the ratio of cross-linked
α
1 and
α
2-chains was 3 to 1 rather than 2 to 1,
suggesting that there was some specificity of structure with regard to the azimuthal
orientation in the packing of the bone collagen fibrils. It is important to recall here that
the C-telopeptide of
α
1-chains to
α
α
2(I) is devoid of K and hence cannot participate in cross-linking.
However, the
2(I) K87 can be present as OHK87 and participate in a cross linkage.
Otsubo et al. (1992) and Yamauchi et al. (1989, 1992) showed that the cross-linking in
dentin was related to the extent of mineralization. Collagen is secreted from the
odontoblasts into an adjacent nonmineralized layer called predentin. Predentin
subsequently mineralizes to become the dentin. No collagen is secreted directly into the
dentin. The predentin cross-linking was all in the form of DHLN, and at the level of
α
2
moles of cross-link per mole of collagen. All the cross-linkages were between C-telo-K
16C
and helix-K
87
, so that molecules must be linked to both nearest neighbors as allowed by the
quasi-hexagonal packing model formation of an extended system of the appropriate 1-5-1-
5- etc. cross-links throughout a fibril. The surprising feature of these data was that in the
mineralized dentin, the DHLN cross link content was less than half that of the predentin,
0.86 cross-links per mole of collagen. Moreover, 0.08 moles/mole collagen were HLN and
0.10 mole/mole collagen were in the form of pyridinoline (LP+HLP, PYR)). Further, the
predentin DLHN cross-links do not mature to PYR, and it is evident that about half of the
bonds are disrupted during mineralization. Perhaps the precursor form of the cross-
linkages,
∼
HLHN, may exist in the predentin and their maturation is overtaken by the
mineralization (15-20 h) before the Amadori reaction has had time to occur, so that
in situ
the bonds are still labile and reversible. In support of this idea, the free aldehyde content is
greater in dentin than in predentin. The mineralization of the dentin matrix may reorganize
the fibril network and provide enough energy to disrupt the C-telo to helix bonding
network. Any collagen differences between dentin and predentin must be the result of post-
secretory processing, tissue maturation and mineralization. The insertion of the mineral
phase can change both the fibril organization and chemistry.
More recently, in addition to the pyridinoline tri-functional cross-linking related to the
C-telopeptide to helix bonds described above, a set of N-telopeptide to helix pyridinolines
were found in bone. The main N-telo related pyridinoline involved
∆
1(I)K
9N
2(I)K
5N
α
−α
−
α
1(I)OHK(K)930 (Hanson and Eyre 1996). The chemical properties of this cross-linkage
are such that the N-telo pyridinoline fluorescence, the main procedure used to detect its
presence, was quenched on chromatography, leading to the underestimation of its presence.
Pyridinolines linking two
1(N) telopeptides to helix were a minor component. The cross-
link ratio of HLP to LP differed between N-telopeptide and C-telopeptide sites, and
between the individual interchain combinations. Cross-linked N-telopeptides accounted for
two-thirds of the total lysylpyridinoline in bone.
It seems paradoxical that the domains with the highest chain packing densities and
crystallinity (Hulmes et al. 1995) within bone and dentin collagen fibrils are so readily
modified by deposition of the mineral phase. As pointed out earlier, the gap or hole zone
spaces within which the mineral crystals may nucleate and grow are not uniform in
α
