The organization of the osteocyte network mirrors the extracellular matrix orientation in bone
Introduction
Bone is a hierarchically structured material undergoing continuous remodeling (Buckwalter et al., 1995, Currey, 2003, Fratzl and Weinkamer, 2007). This process is traditionally attributed to activity of both osteoclasts, the bone resorbing cells and osteoblasts, the bone forming cells. During the formation process of new bone, a portion of the osteoblasts gets embedded in the mineralizing matrix and become osteocytes. Thus, the distribution of osteocytes buried in the bone matrix (Franz-Odendaal et al., 2006) can be considered as a fingerprint of the bone formation process. By establishing connections with neighboring cells during osteoid secretion, these cells do not only establish cell-to-cell communication but also produce a system of extracellular spaces through which bone tissue can be perfused – the osteocyte network (Lanyon et al., 1993). This three-dimensional network of lacunae and canaliculi contains non-mineralized, organic matrix, rich in osteopontin (Kaartinen et al., 2002), and the osteocyte cells (Burger and Klein-Nulend, 1999). Several functions have been proposed for osteocytes (Bonewald, 2007, Bonewald and Johnson, 2008), such as being ion-sensors or regulators for osteoid matrix maturation and mineralization (Kamioka et al., 2001). A general consensus exists that osteocyte function includes translating mechanical strain into biochemical signals (Klein-Nulend et al., 1995), forming the basis of the mechanotransduction pathway in bone tissue. The osteocyte network has a surface estimated to be 400-fold greater than that of the entire Haversian system and more than 100-fold greater than the trabecular bone surface (Aarden et al., 1994, Teti and Zallone, 2009). This enormous surface has the potential to substantially contribute to bone mineral homeostasis via a direct interaction of the osteocytes with their surrounding matrix, independently or in presence of bone remodeling and formation.
During bone formation osteoblasts produce type I collagen molecules which form the bulk of extracellular matrix (ECM). These triple-helical molecules organize into fibrils and further into more or less regular ordered 3D structures (Fratzl et al., 2004). It is as yet unclear whether the 3D organization of collagen fibrils in ECM is the result of a cell-directed construction (Jones et al., 1975, Palumbo et al., 2004) or of self-assembly processes (Giraud-Guille et al., 2008). Nevertheless, bone structures in their final state are well described in the literature (Buckwalter et al., 1995, Currey, 2003, Stover et al., 1992, Weiner and Wagner, 1998). During the formation of bone in mammalian embryos, initially a highly disordered tissue is deposited first and later either replaced and/or augmented by other bone types such as parallel fibered or lamellar bone showing a long range collagen matrix orientation (Weiner and Wagner, 1998). The initially deposited, highly disordered bone tissue can be characterized by the absence of long range order of the collagen matrix orientation, often being referred to as woven bone (Weiner and Wagner, 1998).
In endochondral ossification a specific lamellar pattern is formed as a response to stresses applied to bone (Chakkalakal et al., 1999, Frost and Jee, 1994, Liu et al., 2010, Vatsa et al., 2008) at later stages of development. Different animals grow bone differently in terms of speed of growth and bone structure. The mature femur of mouse bone consists of relatively high amounts of less organized bone, surrounded by later grown, more ordered parallel fibered or lamellar bone (Hörner et al., 1997). However, in femora of larger animals – such as sheep or horse – so called fibrolamellar (also called plexiform) bone is the main building block (Currey, 2002). This type of bone consists of alternating layers with relatively small amounts of woven bone and larger amounts of parallel fibered and lamellar bone tissue. An initial layer of woven bone is laid down quickly, covered by parallel fibered matrix and finally filled towards the blood vessels more leisurely with lamellar bone. Remodeling leads then to a bone structure with a fibrolamellar matrix in which secondary osteons are embedded (Currey, 2003). During the remodeling process the preexisting matrix gets resorbed by osteoclasts, including the osteocytic network, followed by a so-called reversal phase, where the eroded surface is smoothened and covered with a non-collagenous protein-rich substance (cement line) (Burr et al., 1988, Skedros et al., 2005) also containing osteopontin (McKee and Nanci, 1996). Osteopontin in cement lines may act as an interfacial adhesion promoter, therein maintaining the overall integrity of bone during the remodeling sequence and “bonding” dissimilar tissues. On such prepared surfaces the new bone matrix is then deposited by the osteoblasts.
As the formation of bone is performed by osteoblasts in conjunction with the setup of the osteocyte network, the question arises how the arrangement of the ECM is linked to the architecture of the osteocyte network. As mentioned above, a debate still exists as to whether the 3D organization of collagen fibrils is cell-driven or a result of self-assembly, thus the influence of the arrangement of the cells on the final structure of bone is still an open question (Bouligand, 1972, Canty et al., 2004, Giraud-Guille, 1992, Giraud-Guille et al., 2008, Hynes and Destree, 1978). On the nanometer length scale up to about 20 μm (in the range of the cell size) it is indisputable that a self-assembly process dominates. On the other hand, approaching the question of organization from the higher length scales (millimeter range) a mechanically controlled feedback loop is active in bone, regulating bone mass and architecture (mechanostat) (Frost, 1987) and therefore indirectly also the 3D organization of collagen fibrils. The question remains, which mechanisms exist on the intermediate length scales (the range of many matrix-producing cells up to millimeter range) to organize the ECM in different ways as it is reflected in different bone types such as lamellar and woven bone.
In order to understand the origins of the 3D organization of the collagen matrix over the mentioned intermediate length scales in bone, we investigate the architecture of the osteocyte network with respect to the organization of the collagen fibrils within the ECM. We combined confocal laser scanning microscopy (CLSM) (Kamioka et al., 2001, Kamioka et al., 2009, Sugawara et al., 2005) with rhodamine-based staining (Smith et al., 2006) to investigate the structure of the osteocyte network non-destructively, hence without the need to demineralize the examined bone samples. In the present study we investigated the structure of three bone types from four different animals: horse, mouse and sheep respectively the cow. These types were chosen because they feature a high structural diversity in terms of degree of organization. Equine metacarpal bone (horse) shows the highest degree of organization, while murine bone (mouse) shows the lowest, and fibrolamellar bovine and ovine bone (cow and sheep) is situated between those two from a structural perspective. In addition to CLSM, back-scattered electron imaging (Roschger et al., 2008) and polarized light microscopy were used to study the microstructural organization of these different bone types. These investigations shed light on the correlation of the organization of the osteocyte network and the extracellular matrix orientation. Moreover, cross-correlation of osteocyte network geometry with extracellular matrix properties gives insights into the formation dynamics of the different examined bone types during physiological growth and remodeling.
Section snippets
Sample preparation
Osteonal equine bone from the 3rd metacarpal bone (wrist bone) from a 7 year old horse, fibrolamellar ovine bone from the mid-diaphysis of the femur of a 5 year old sheep (for the high vacuum back-scattered electron imaging a comparable bone sample from the femur from a 2–3 year-old cow was used) as well as murine bone from the mid-diaphysis of the femur from a 12 month-old mouse, without initial aldehyde fixation, was transversally sectioned with an initial thickness of 200 microns. Samples were
Results
Undemineralized bone sections of osteonal cortical bone from horse, plexiform (fibrolamellar) bone from sheep and mouse femur were stained with rhodamine. The treatment of the bone samples with the rhodamine solution resulted in a labeling of all internal bone surfaces and consequently revealing the blood vessels, the lacunae and the canaliculi. Thus, the whole osteocytic network is dyed with a fluorescent medium that can be visualized and analyzed in a 3D mode with confocal laser scanning
Discussion
The rhodamine CLSM-imaging technique combined with polarized microscopy and back-scattered electron imaging provides the possibility to cross-correlate structural features of bone material. We demonstrated that the organization of the osteocytic network mirrors the extracellular matrix orientation. This is evident from comparisons made of the organization of the osteocyte network and ECM of different bone types from different animals. It also shows that different length scales must be
Acknowledgments
We thank P. Leibner, S. Weichold, A.M. Martins and B. Schonert from the Max Planck Institute of Colloids and Interfaces for technical assistance and H. Schell from the Julius Wolff Institute for providing ovine bone samples collected in the SFB 760. M.K. is funded by Bundesministerium für Bildung und Forschung (BMBF, Grant-number: 01 EC 1006C, project 9133). Contributions were made possible by DFG funding through the Berlin-Brandenburg School for Regenerative Therapies GSC 203. The institute of
References (48)
- et al.
Osteocytes, mechanosensing and Wnt signaling
Bone
(2008) Twisted fibrous arrangements in biological-materials and cholesteric mesophases
Tissue & Cell
(1972)- et al.
Composition of the cement line and its possible mechanical role as a local interface inhuman compact-bone
Journal of Biomechanics
(1988) - et al.
Fatigue life of compact bone. 2. Effects of microstructure and density
Journal of Biomechanics
(1976) - et al.
Repair of segmental bone defects in the rat: an experimental model of human fracture healing
Bone
(1999) The many adaptations of bone
Journal of Biomechanics
(2003)- et al.
Nature’s hierarchical materials
Progress in Materials Science
(2007) Liquid crystallinity in condensed type-I collagen solutions – a clue to the packing of collagen in extracellular matrices
Journal of Molecular Biology
(1992)- et al.
Liquid crystalline properties of type I collagen: perspectives in tissue morphogenesis
Comptes Rendus Chimie
(2008) - et al.
Morphological features of bone in human osteopetrosis
Bone
(1991)