Content

In human patients, the sequence of KOA subchondral bone changes is less well understood. To gain insight into the osteochondral unit in KOA, we have used a multimodal approach to characterise the subchondral bone across the tibial plateau, using tibial plateaus taken from patients with KOA at total knee arthroplasty, as well as controls without KOA. We have been particularly interested in zones of subchondral bone represented by BMLs, which seem to correspond to the most severe OA changes. BMLs have acquired considerable clinical interest, since they appear to inform on clinically important changes in the subchondral bone, and thus might be useful as imaging biomarkers for both disease progression and response to treatment of KOA. Subchondral bone in BML zones was characterized by a number of important differences from healthy bone: vascular changes, increased bone matrix microdamage, increased resorptive sites, increased osteoid, and a focal sclerotic appearance, in both the subchondral plate and the underlying trabeculae. These changes correlated with the degree of degradation of the adjacent cartilage (1).

Micro-CT of the whole tibial plateaus showed key subregion-specific differences between non-OA controls, KOA without BMLs and KOA with BMLs. Limiting analyses to tibial plateaus containing a BML in the anterior medial (AM) compartment (the most frequent site of BMLs), between-group comparisons showed that the AM region of the OA-BML group had significantly higher histological cartilage degeneration (OARSI grade) (P<.0001, P<.05), thicker subchondral plate (P<.05, P<.05), trabeculae that are more anisotropic (P<.0001, P<.05), well connected (P<.05), and more plate-like (P<0.05, P<0.05), compared to controls and OA-no BML at this site. OA-no BML had significantly higher OARSI grade (P<.0001), and lower trabecular number (P<.05) compared to controls.
The subchondral trabecular bone micro-CT data were subjected to ITS analysis for plate-and-rod-based microstructural analysis. The subchondral bone of OA tibial plateaus containing BMLs was characterized by increased plate bone volume fraction (pBV/TV), (p=0.003), plate trabecular number (pTb.N), (p=0.04), and both rod and plate trabecular thickness (rTb.Th and pTb.Th) (p<0.0001 for both). Comparison between the anterior medial region (representing BML bone) and posterior medial region (representing no-BML) in OA subjects indicated that the anterior medial subregion representing BML bone is characterised by a greater number of plate and rod like trabeculae (p<0.0001 for all parameters) compared with posterior medial (no BML) bone. These differences also associated positively with increased OARSI grade in the BML region.

To investigate the nature of the bone matrix in the sclerotic BML regions of subchondral bone, Raman microspectroscopy analysis and quantitative backscattered electron imaging (qBEI) were performed. qBEI was used to determine bone mineralization density distribution and osteocyte and mineralized lacunar density. Raman spectra provided a distinct chemical fingerprint of each patient’s tibial bone sample. In particular, phosphate-to-amide I, phosphate-to-proline and phosphate-to-amide III were all reduced in the BML zones compared to control (p<0.002, p<0.02, p<0.03, respectively), suggesting lower mineralized bone matrix in BML bone. Consistent with this, the qBEI bone mineralization density distribution for OA-BML was shifted to low mineralization density compared to control (p<0.002), combined with an increased peak width compared to both OA-no BML (p<0.05) and control (p<0.002). The size and density of osteocyte lacunae did not differ between groups. However, the density of mineralized lacunae in bone from BML zones was lower compared to control (p<0.005). Thus, tibial BMLs in knee OA patients are characterized by low bone mineralization, in relation to the organic phase, together with greater mineral heterogeneity and a reduced number of mineralized osteocytes.

Taken together, our data support the concept of the sclerotic bone in BMLs representing localized areas of active bone remodeling in response to chronic bone injury in OA.

References

1. D. Muratovic, F. Cicuttini, A. Wluka, D. Findlay, Y. Wang, S. Otto, D. Taylor, J. Humphries, Y. Lee, A. Labrinidis, R. Williams, J. Kuliwaba, Bone marrow lesions detected by specific combination of MRI sequences are associated with severity of osteochondral degeneration, Arthritis Res Ther 18 (2016) 54.

Introduction

The tidemark is a structure that marks the boundary between articular cartilage and the calcified cartilage layer in synovial joints. It is an undulating structure around 3 µm-thick. A tidemark is also found at the tendon-bone interface, more specifically in the fibrocartilagenous enthesis that is found for example in the rotator cuff, patellar tendon, and Achilles tendon ¹. The tidemark is most easily seen in decalcified tissue sections stained by basophilic dyes that bind to polyanionic species: hematoxylin and DAPI ^2,3.

Tidemark formation accompanies the special mineralization process of the calcified cartilage layer, which could be viewed as a tide of mineralization factors thrown up against a cartilage tissue whose matrix is replete with factors that inhibit mineralization. However unlike waves at the seashore that cast debris on the sand in a unidirectional manner, the literature has shown that the tidemark interface between cartilage and calcified cartilage is the result of a 3-directional exchange of mobile factors that diffuse from synovial fluid and blood through the cartilage, calcified cartilage and bone matrix ⁴. Although blood and synovial fluid may be viewed as 2 distinct pools of body fluid, most solutes and small proteins readily exchange between the synovial fluid and blood plasma ⁵. Pro-mineralization and anti-mineralization factors and enzymes circulate in both the blood stream and the synovial fluid ⁶.

Content

Simkin previously proposed that the tidemark is composed of the “mortal remains” of dying calcified chondrocytes that collect at the cartilage-bone junction ³. However data from our laboratory and others are consistent with the calcified chondrocytes having a role in maintaining the tidemark. Hematin-stained lines resembling the tidemark are visible above and encircling calcified chondrocytes in selected rabbit and human osteochondral tissue sections. These hematin-stained lines appear to diffuse as a wave front from a group of calcified chondrocytes through mineral, while gaps in the tidemark are visible over empty calcified chondrocyte lacunae. The polyanionic hematin-stained lines have the appearance of being released from calcified chondrocytes in waves, cycles or bursts over time.

Modest tidemark duplication is a normal process of development and aging. In the normal aging process, the calcified cartilage layer can “creep” or “advance” into the hyaline cartilage deep zone. During aging, the calcified cartilage layer stays the same thickness because of slow endochondral ossification. It can be speculated that once created, tidemarks do not disappear – the new mineral is simply deposited beyond the previous tidemark, leading to “duplication” or multiple wavy lines that resemble tree rings generated during annual growth spurts. This notion is supported by the observation that heavy metals such as lead can accumulate in human specimens at the original tidemark at the cartilage-calcified cartilage junction ⁷ and also in deeper (older) duplicated tidemarks closer to the bone ⁸. In skeletally aged sheep, the calcified cartilage layer shows heavy tidemark duplication. Tidemark duplication and roughness was also associated with osteoarthritis (OA) progression ⁹. In a rat model, mechanical unloading of the hind limbs for 28 days by tail suspension was associated with thickening or advancement of the calcified cartilage layer ¹⁰. In some sheep condyles following cartilage repair surgery, the flanking cartilage can undergo extensive tidemark advancement in just several months relative to residual calcified cartilage inside the cartilage defect. All of these observations are consistent with the hypothesis proposed by Xiao et al ¹¹ that OA progresses due to tidemark/osteochondral advancement at the expense of the articular cartilage layer. We would have to suppose that if tidemark advancement were caused by diffusion of a factor/factors released from calcified chondrocytes and/or the subchondral bone, that there must be mechanisms to suppress mineral advancement into the articular cartilage itself.

Specific mineralization of the calcified cartilage layer is largely controlled by developmentally-timed expression and anatomical deposition of factors in the extracellular matrix. Some gamma-carboxyglutamic acid (gla)-domain factors are matricellular molecules that form complexes with apatite mineral and also with cells. The gla domain is modified by a Vitamin K-dependent process to become hypercarboxylated, and this highly acidic domain can form complexes with calcium and apatite. Gla domain factors that promote mineralization (bone gla protein, osteocalcin, osteonectin/SPARC) are deposited in the mineral phase, and matricellular factors that inhibit mineralization (matrix gla protein/MGP, osteopontin) are abundant in cartilage matrix, especially in the deep zone ¹². As famously shown, mouse strains with null mutations in MGP develop abnormal calcification in arteries and growth plate cartilage ¹³. Warfarin inhibits Vitamin K and blocks gla-domain carboxylation; it is used clinically for its anticoagulant activity although fetal exposure was associated with an interference with MGP performance, leading to hyper-mineralized growth plates; Vitamin K-deficiency in adults was previously tied to joint space narrowing ^14,15. There is evidence that apatite-binding factors accumulate in the tidemark. In a study of the rabbit paw, osteopontin was detected in calcified chondrocytes and in the tidemark by immunostaining ¹⁶. In addition to matricellular factors, cartilage calcification is controlled through extracellular vesicles (phospholipid matrix vesicles), and phosphatases.

Alkaline phosphatase (ALP) is an enzyme known to promote mineralization, but in extreme cases, ALP and acid phosphatase can both demineralize tissue. Tidemark erosion could be an initiating factor in tidemark advancement into the articular cartilage deep zone. Tidemark erosion were previously observed in carrageenan-injected rabbit knees in an experimental model of inflammatory arthritis ¹⁷. Kidney pathology, diabetic states and hypophosphatemia (rickets) was previously found to induce heavy ALP activity at the tidemark and the Achilles enthesis leading to erosion of the tidemark and weakening of the junction ^18,19. In a rabbit model of bone marrow stimulation for cartilage repair, Tartrate-resistant Acid Phosphatase (TRAP) activity was specifically detected in calcified chondrocytes at 1 to 2 weeks post-surgery (unpublished observations). To elucidate structural relationships between osteocytes and the calcified cartilage layer, we analyzed osteocyte morphology in silver stained osteochondral sections from New Zealand White rabbit knees. One knee was intact and the contralateral knee had a chronic cartilage defect with one week of healing after bone marrow stimulation surgery. Compared to intact knees, osteocyte dendrite length in microdrilled knees was on average 1 μm less (p=0.04, 6.2 vs 7.3 µm) along with less variation in dendrite length, signifying that on average the osteocytes retracted their dendrites and had more uniform/regular dendrite lengths after microdrilling. Shorter dendrite lengths are expected to alter osteocyte cell communication and promote trabecular bone perilacunar/canalicular remodeling (PLR). Although osteocyte dendrites occasionally extended to the border of the bone/calcified cartilage interface, the canalicular network did not extend into the calcified cartilage layer. These results suggest that the cells in the calcified cartilage and bone communicate through paracrine factors, matrix vesicles, and mechanical cues and not through direct cell-cell communication.

To summarize, the tidemark is a dynamic structure that is formed at cartilage-calcified cartilage interface. The tidemark can be remodeled by enzymes released at the cartilage-bone junction in pathological states. Tidemark advancement is a sign of OA progression. Tidemark irregularity in early OA was proposed to be a reversible event ⁹, suggesting that more knowledge of mechanisms involved in tidemark remodeling could help preserve whole joint function following surgical procedures that temporarily disrupt the tidemark or fibrocartilage enthesis.

References

1. Benjamin, M., et al. The skeletal attachment of tendons - tendon 'entheses'. Comp. Biochem. Physiol. A-Mol. Integr. Physiol. 133, 931-945 (2002).
2. Hoemann, C.D., et al. International Cartilage Repair Society (ICRS) Recommended guidelines for histological endpoints for cartilage repair studies in animal models and clinical trials. Cartilage 2, 153-172 (2011).
3. Simkin, P.A. A biography of the chondrocyte. Annals of the Rheumatic Diseases 67, 1064-1068 (2008).
4. Hoemann, C.D., Lafantaisie-Favreau, C.-H., Lascau-Coman, V., Chen, G. & Guzmán-Morales, J. The Cartilage-Bone Interface. J Knee Surg 25, 085-098 (2012).
5. Burstein, D., et al. Protocol issues for delayed Gd(DTPA)(2-)-enhanced MRI: (dGEMRIC) for clinical evaluation of articular cartilage. Magnetic Resonance in Medicine 45, 36-41 (2001).
6. Lehman, M.A., Kream, J. & Brogna, D. Acid and alkaline phosphatase activity in the serum and synovial fluid of patients with arthritis. J. Bone Joint Surg.-Am. Vol. 46, 1732-1738 (1964).
7. Zoeger, N., et al. Determination of the elemental distribution in human joint bones by SR micro XRF. X-Ray Spectrometry 37, 3-11 (2008).
8. Roschger, A., et al. Differential accumulation of lead and zinc in double-tidemarks of articular cartilage. Osteoarthritis Cartilage 21, 1707-1715 (2013).
9. Deng, B., et al. Quantitative study on morphology of calcified cartilage zone in OARSI 0 similar to 4 cartilage from osteoarthritic knees. Current Research in Translational Medicine 64, 149-154 (2016).
10. O'Connor, K.M. Unweighting accelerates tidemark advancement in articular cartilage at the knee joint of rats. Journal of Bone and Mineral Research 12, 580-589 (1997).
11. Xiao, Z.F., Su, G.Y., Hou, Y., Chen, S.D. & Lin, D.K. Cartilage degradation in osteoarthritis: A process of osteochondral remodeling resembles the endochondral ossification in growth plate? Med. Hypotheses 121, 183-187 (2018).
12. Muller, C., et al. Quantitative proteomics at different depths in human articular cartilage reveals unique patterns of protein distribution. Matrix Biol. 40, 34-45 (2014).
13. Luo, G.B., et al. Spontaneous calcification of arteries and cartilage in mice lacking matrix GLA protein. Nature 386, 78-81 (1997).
14. Hale, J.E., Fraser, J.D. & Price, P.A. The identification of matrix Gla protein in cartilage. Journal of Biological Chemistry 263, 5820-5824 (1988).
15. Neogi, T., Felson, D.T., Sarno, R. & Booth, S.L. Vitamin K in hand osteoarthritis: results from a randomised clinical trial. Annals of the Rheumatic Diseases 67, 1570-1573 (2008).
16. King, K.B., Opel, C.F. & Rempel, D.M. Cyclical articular joint loading leads to cartilage thinning and osteopontin production in a novel in vivo rabbit model of repetitive finger flexion. Osteoarthritis Cartilage 13, 971-978 (2005).
17. Bogoch, E.R., Lee, T.C., Fornasier, V.L. & Berger, S.A. Articular damage is associated with intraosseous inflammation in the subchondral bone marrow of joints affected by experimental inflammatory arthritis and is modified by zoledronate treatment. Journal of Rheumatology 34, 1229-1240 (2007).
18. Liang, G., VanHouten, J. & Macica, C.M. An Atypical Degenerative Osteoarthropathy in Hyp Mice is Characterized by a Loss in the Mineralized Zone of Articular Cartilage. Calcif. Tissue Int. 89, 151-162 (2011).
19. Karaplis, A.C., Bai, X.Y., Falet, J.P. & Macica, C.M. Mineralizing Enthesopathy Is a Common Feature of Renal Phosphate-Wasting Disorders Attributed to FGF23 and Is Exacerbated by Standard Therapy in Hyp Mice. Endocrinology 153, 5906-5917 (2012).

Acknowledgments

Funding: George Mason University Start-up funds (CDH), CIHR (MOP 133729), NSERC (STPGP 36525), Ortho RTi/Prima Quebec Research contract.