Roles of collagen cross-links and osteon collagen/lamellar morphotypes in equine third metacarpals in tension and compression tests

Abstract

Many bones experience bending, placing one side in net compression and the other in net tension. Because bone mechanical properties are relatively reduced in tension compared with compression, adaptations are needed to reduce fracture risk. Several toughening mechanisms exist in bone, yet little is known of the influences of secondary osteon collagen/lamellar 'morphotypes' and potential interplay with intermolecular collagen cross-links (CCLs) in prevalent/predominant tension- and compression-loaded regions. Paired third metacarpals (MC3s) from 10 adult horses were prepared for mechanical testing. From one MC3/pair, 5 mm cubes were tested in compression at several mid-shaft locations. From contralateral bones, dumbbell-shaped specimens were tested in tension. Hence, habitual/natural tension- and compression-loaded regions were tested in both modes. Data included: elastic modulus, yield and ultimate strength, and energy absorption (toughness). Fragments of tested specimens were examined for predominant collagen fiber orientation (CFO; representing osteonal and non-osteonal bone), osteon morphotype score (MTS, representing osteonal CFO), mineralization, porosity and other histological characteristics. As a consequence of insufficient material from tension-tested specimens, CCLs were only examined in compression-tested specimens (HP, hydroxylysylpyridinoline; LP, lysylpyridinoline; PE, pentosidine). Among CCLs, only LP and HP/LP correlated significantly with mechanical parameters: LP with energy absorption, HP/LP with elastic modulus (both r=0.4). HP/LP showed a trend with energy absorption (r=-0.3, P=0.08). HP/LP more strongly correlated with osteon density and mineralization than CFO or MTS. Predominant CFO more strongly correlated with energy absorption than MTS in both testing modes. In general, CFO was found to be relatively prominent in affecting regional toughness in these equine MC3s in compression and tension.

Keywords: Bone adaptation; Bone mechanical properties; Collagen fiber orientation; Cortical bone; Energy absorption; Haversian systems; Secondary osteons; Strain-mode-specific testing.

Fig. 1.

Six-point osteon morphotype scoring system. The six-point scoring scheme with examples of each osteon birefringence pattern (osteon collagen/lamellar morphotype) in circularly polarized light. The numerical values of the six morphotypes are used to calculate the osteon morphotype score (MTS) of entire microscopic images that contain many osteons (Skedros et al., 2009, 2011b). Four of the numerical scores shown include consideration of completeness and birefringence strength (brightness) of the peripheral ring ‘O’ or hoop (which is shown most clearly in osteon 4 of example 1 and has relevance in the debonding process that occurs when bone fails; Skedros et al., 2013a): 0=category N, a dark osteon with no birefringent lamellae; 1=category OWI, a combination of OI and OW; 2=category OW, similar to O but the birefringent ring is weak (W); 3=category OI, similar to O but the birefringent peripheral ring is incomplete (I); 4=category O osteon with dark interior and strongly birefringent peripheral lamellae; 5=‘distributed’ osteon group. This group includes ‘bright’ and ‘alternating’ osteons. Note there can be osteons that are considered to be ‘hybrid’ (h) morphotypes because they have variable gray-level patterns within the osteon wall (see the middle four osteon images in example 3 column). These osteons are also scored with the six-number scheme. Images in example 1 column are reproduced from Martin et al. (1996) with permission of Elsevier Science, Inc.

Fig. 2.

Transverse sections of a third metacarpal (MC3) and test specimens. (A) Drawings of transverse sections showing the locations of an MC3 where compression (left) and tension (right) test specimens were obtained: D, dorsal; L, lateral; M, medial; and P, palmar. NA, neutral axis. Black regions show where tension is prevalent/predominant; white regions indicate areas of habitual compression. The curved arrows in the top left drawing show rotation of the neural axis that occurs at faster gait speeds, which causes the lateral cortex (hatched region) to change from its ambient compression strain mode to tension (i.e. a strain-mode reversal occurs). (B) The dimensions of the tension (dumbbell-shaped) and compression (cubic) specimens. The hatched areas show regions of the specimens where histocompositional analyses were performed. BSE, backscattered electron imaging. Figure reproduced with permission from Skedros et al. (2006).

Fig. 3.

Forelimb and mid-diaphyseal cross-section of the MC3 of a Thoroughbred. (A) Top: equine forelimb and mid-diaphyseal cross-section of the MC3 showing regional differences in strain magnitude and mode (i.e. tension and compression) and the neutral axis (NA) at mid-stance in accordance with the strain distribution of the finite element model of Gross et al. (1992) as shown above. D, dorsal; L, lateral (for all four cross-sections shown). Bottom: toward the end of stance phase and especially at higher gait speeds the NA rotates clockwise (NA¹ to NA²), which places the lateral cortex in more prevalent/predominant tension (T) (C, net compression) (Nunamaker, 2001) (figure reproduced with permission from Skedros et al., 2007). (B) Mid-diaphyseal cross-section of the MC3 showing regional differences in total energy absorption data from Skedros et al. (2006). The asterisks indicate two regions that are significantly different from each other (P≤0.05). (C) Mid-diaphyseal cross-section of the MC3 demonstrating our hypotheses concerning regional differences in collagen cross-links (CCLs) and osteon MTSs (see Table 1).

Fig. 4.

Regional differences in predominant collagen fiber orientation (CFO), osteon MTS and CCL data. (A) Predominant CFO (expressed as weighted mean gray levels, WMGLs) and osteon MTS – results of paired comparisons between regions. Data (means±s.d.) were obtained exclusively from compression-tested specimens to allow for easier comparison with regional differences in CCLs. Letters in bars (corresponding to the cortical regions indicated) show significant differences (P<0.05) in paired comparisons of each characteristic (‡ indicates a trend). (B) CCL data – results of paired comparisons between compression-tested regions. HP, hydroxylysylpyridinoline; LP, lysylpyridinoline; PE, pentosidine (note the change in units for PE). The sample sizes for the two-sample t-tests in A and B are: (1) dorsal–lateral: 16–20 for each parameter, (2) lateral: 7–10 for each, (3) palmar–medial: 9 for CFO and 16–18 each for MTS and CCLs, and (4) dorsal–medial: 8 for MTS, 10 for CCLs, and 18 for CFO. ^‡0.06≤P≤0.09; *P≤0.05.

Fig. 5.

Extrinsic and intrinsic toughening mechanisms, loading modes and ASTM E399 fracture toughness standard. (A) The left side of this graphic primarily shows extrinsic toughening mechanisms in notched mechanical test specimens; these mechanisms include osteons and non-osteonal CFO. On the right are intrinsic toughening mechanisms, which include CCLs and, to some extent, also likely non-osteonal CFO. Figure reproduced from Launey et al. (2010) with permission of Annual Reviews, Inc. (B) Different modes of loading of notched specimens: mode I (opening mode), mode II (shear loading) and mode III (tearing and or anti-shear loading). (C) Orientations of notched specimens in accordance with ASTM E399 fracture toughness standard (1997, see Chittibabu et al., 2016). C, circumferential; L, longitudinal; R, radial. Figure reproduced from Chittibabu et al. (2016) with permission of Advances in Science and Technology Research Journal; Creative Commons Attribution 4.0 International (CC BY 4.0).