• bone remodeling;
  • bone remodeling compartment;
  • histology;
  • lining cells;
  • bone histomorphometry


  1. Top of page
  2. Abstract
  7. Acknowledgements

We describe a sinus, referred to as a bone remodeling compartment (BRC), which is intimately associated with cancellous bone remodeling. The compartment is lined on its marrow side by flattened cells and on its osseous side by the remodeling bone surface, resembling a roof of flattened cells covering the bone surface. The flat marrow lining cells are in continuity with the bone lining cells at the margins of the BRC. We examined a large number of diagnostic bone biopsy specimens received during recent years in the department. Furthermore, 10 patients (8 women and 2 men, median age 56 [40–69] years) with the high turnover disease of primary hyperparathyroidism who were treated with parathyroidectomy and followed for 3 years were included in the histomorphometric study. Bone samples for the immuno-enzyme staining were obtained from an amputated extremity of child. The total cancellous bone surface covered by BRC decreases by 50% (p < 0.05) following normalization of turnover and is paralleled by a similar 50% decrease in remodeling surface (p < 0.05). The entire eroded surface and two-thirds of the osteoid surface are covered by a BRC. BRC-covered uncompleted walls are 30% (p < 0.05) thinner than those without a BRC. This indicates that the BRC is invariably associated with the early phases of bone remodeling, that is, bone resorption, whereas it closes during the late part of bone formation. Immuno-enzyme staining shows that the flat marrow lining cells are positive for alkaline phosphatase, osteocalcin, and osteonectin, suggesting that they are bone cells. The first step in cancellous bone remodeling is thought to be the lining cells digesting the unmineralized matrix membrane followed by their disappearance and the arrival of the bone multicellular unit (BMU). We suggest that the lining cell barrier persists during bone remodeling; that the old lining cells become the marrow lining cells, allowing bone resorption and bone formation to proceed under a common roof of lining cells; that, at the end of bone formation, new bone lining cells derived from the flattened osteoblasts replace the marrow lining cells thereby closing the BRC; and that the two layers of lining cells eventually becomes a single layer. The integrity of the osteocyte-lining cell system is reestablished by the new generation of lining cells. The BRC most likely serves multiple purposes, including efficient exchange of matrix constituents and minerals, routing, monitoring, or modulating bone cell recruitment, and possibly the anatomical basis for the coupling of bone remodeling.


  1. Top of page
  2. Abstract
  7. Acknowledgements

BONE AND bone marrow are two different tissues of intimate proximity. The bone tissue is morphologically(1) and physiologically(2) separated from the marrow by bone lining cells. The process of cancellous bone remodeling occurs on the surface of trabeculae, at the boundary between bone and marrow. Bone remodeling is brought about by a team of different bone cells (bone multicellular unit [BMU]), appearing in a predetermined sequence: osteoclasts, mononuclear cells, and osteoblasts.(3) Lining cells produce collagenase, which digests the layer of unmineralized matrix thereby exposing the bone surface to osteoclastic bone resorption.(4) Osteoblasts are recruited as part of the BMU and bone formation is initiated and fills up the resorbed Howship's lacuna.(5, 6) With the completion of bone formation, it is well established that some osteoblasts eventually become flat surface cells lining the quiescent surfaces, that is, bone lining cells.(7) Whereas bone lining cells are believed to differentiate from inactive osteoblasts that have ceased their matrix synthesis, the fate of bone lining cells at the onset of bone resorption is unknown.

It is described that the interface between bone forming surfaces and bone marrow are lined by vascular structures, paratrabecular sinusoidal capillaries. An osteoblastic layer makes up the osseous wall of the capillary, the opposite wall being formed by endothelial cells, and it is argued that the most likely explanation is a reduplication of the endosteal cells to form the paratrabecular sinusoid and the osteoblasts as part of its wall.(8) Bone remodeling occurring in a bone chamber is also related to the existence of and increased flow through microvessels that conform closely to the contour of the cancellous bone surface.(9) A looping blood vessel is a well-recognized component of cortical osteons during normal remodeling,(10) and the possibility of finding a similar vessel associated with cancellous osteons has linked the microvasculature to the regulation of coupling between bone resorption and bone formation.(6, 11)

We have identified a specialized structure, which we refer to as a bone remodeling compartment (BRC), associated with cancellous bone remodeling surfaces. The light microscopic morphology and cellular phenotype are described, and functional aspects are studied using bone morphometry. A hypothesis interlinking the existence and dynamics of the bone remodeling compartment to the dynamics of the bone remodeling process is proposed.


  1. Top of page
  2. Abstract
  7. Acknowledgements

Bone histology and histomorphometry

We examined a large number of diagnostic bone biopsy specimens received during recent years in the department. The biopsy specimens presented in Fig. 1 were selected because of their informative presentation of the structures studied. The histomorphometric analysis was based on 30 patients (25 women and 5 men), average age 54 (range, 27-73) years, with primary hyperparathyroidism who were treated with parathyroidectomy and followed up for three years.(12) Individuals who had liver or kidney diseases; who suffered from malabsorption, diabetes, or metabolic bone diseases other than primary hyperparathyroidism; who were taking drugs known to influence bone metabolism; or who were known to consume excessive amounts of alcohol were excluded from the study. Cases were also excluded if the mineral apposition rate was below the detection limit,(13) if one of the biopsy specimens in the pair was missing, or if the technical quality of the biopsy did not allow a complete evaluation of structures at the marrow-bone interface. This left 10 biopsy specimens (8 women and 2 men), average age 56 (40-69) years, for evaluation. The local ethics committee gave permission to the study, and participants gave their informed consent before being included.

thumbnail image

Figure FIG. 1. (A) Cancellous bone remodeling site showing bone resorption and bone formation associated with a common bone remodeling compartment (BRC) in a patient with primary hyperparathyroidism, explaining the slight fibrosis at the BRC roof. (C) A layer of flat cells coherent with the marrow and covering a bone formative surface, which is lined by osteoblasts. (E) A layer of flat cells covering a bone resorptive surface with an osteoclast situated in a small erosion lacuna just below the marrow lining cells. Bone resorption advances beneath the layer of flat marrow lining cells that are in continuum with the bone lining cells. (G) Bone surface with almost completed bone formation showing flattened osteoblasts paving a thin rim of osteoid just below another layer of flat cells coherent with the marrow; the BRC is almost closed. (I) Bone formation surface without osteoblasts and without bone remodeling compartment. A bone remodeling compartment stained with (B) Goldner trichrome, (D) alkaline phosphatase, (F) osteocalcin, (H) osteonectin, and (J) CD34.

Download figure to PowerPoint

Following a 2-8-2-day regimen (2 days of label, 8 days without label, 2 days of label) of tetracycline double-labeling, transiliac bone biopsy specimens were obtained at baseline and 3 years after surgery from the standard location 2 cm below the summit of the iliac crest and 2 cm posterior to the anterior superior iliac spine, using a trephine with an internal diameter of 8 mm. The biopsy specimens were fixated in 70% ethanol, rotated around their vertical axis,(14) and embedded undecalcified in methyl methacrylate at 10°C. Sections were cut at 7 μm using a heavy-duty microtome (Jung, Heidelberg, Germany) at a level one-third of the way through the biopsy. Sections were stained with Goldner trichrome for light microscopy or mounted unstained for epifluorescent microscopy. Between six and eight sections were examined from each biopsy. Resorptive surfaces were defined as eroded, that is, presenting eroded lamellae, regardless of the presence or absence of osteoclasts on the surface. Formative surfaces were defined as osteoid surfaces regardless of the presence or absence of tetracycline labels. In the histomorphometric study, measurements of eroded surface, osteoid surface, and mineralizing surface were in accordance with the guidelines provided by the American Society for Bone and Mineral Research.(15) The percentages of total cancellous bone surface (BRC/BS), erosion surface (BRC/ES), osteoid surface (BRC/OS), and quiescent surface covered by a BRC (BRC/QS) were measured. Uncompleted wall thickness with (BRC.W.Th) or without (W.Th) a covering bone remodeling compartment was uniformly sampled at random by surface(16) and measured by orthogonal intercepts according to Steiniche et al.,(17) using a digitizer and the Jandel SigmaScan software package (SPSS Science, Chicago, IL, USA). The eyepiece was equipped with an integrated line grid, which was rotated randomly rather than sine-weighted.(14) There was a risk of observer bias in the measurements of wall thickness because it could not be blinded as to whether the walls were associated with BRCs or not.

Data were normally distributed and tabulated as mean ± SEM. The paired t-test was used for statistical testing, and p < 0.05 was considered statistically significant.

Immunohistochemistry and enzyme histochemistry

Bone samples were obtained from an amputated upper extremity of an 8-year-old boy. Amputation was performed because of aggressive fibromatosis. Bone samples were fixated in 70% alcohol, dehydrated in increasing concentrations of ethanol, followed by 2-propanol and xylen, before a three-step infiltration with methyl methacrylate. Dehydration and infiltration were performed at 4°C in a vacuum desiccator. Polymerization at −20°C was accelerated with N,N-dimethyl-p-toluidine.(18) The specimen was cut into 7-μm serial sections using a heavy-duty microtome (Jung), mounted on SuperFrost Plus slides, and dried overnight at 37°C. Before staining, the sections were deplasticized in dimetoxyethylacetate, rehydrated, and washed with Tris-buffered saline (TBS, pH 7.4). In each series of sections, the staining panel including Goldner trichrome, alkaline phosphatase, CD34, osteocalcin, osteonectin, and negative controls was applied.

Epitopes were retrieved using 1% acetic acid for 10 minutes.(19) The three-step labeled streptavidin biotin method was used to probe for endothelial and osteoblastic characteristics of the cells lining the BRC-marrow interface using horse radish peroxidase (HRP) for visualization of the antibody complex. For the primary antibodies we used monoclonal mouse anti-human IgG (M7165; DAKO, Glostrup, Denmark) recognizing CD34, polyclonal rabbit anti-human IgG (593; BTI, Stoughton, MA, USA) recognizing osteocalcin, and polyclonal rabbit anti-human IgG (LF 37)(20) recognizing osteonectin.

Endogenous peroxidase activity was blocked with 3% hydrogen peroxide and 10% methanol in TBS (pH 7.4, 20 minutes), followed by washes in water and TBS. To avoid nonspecific tissue binding of antibodies, sections were incubated with TBS (pH 7.4) containing 10% normal goat serum (NGS) (DAKO X0970) for 10 minutes. Sections were incubated overnight with the primary antibody, washed in TBS, and incubated with the biotin-conjugated secondary antibodies for 1 h; biotinylated monoclonal goat anti-mouse IgG (DAKO E0433) was used for CD34 and biotinylated polyclonal goat anti-rabbit IgG (DAKO E0432) for osteocalcin and osteonectin. The slides were rinsed, incubated with the tertiary antibody peroxidase-conjugated streptavidin (DAKO P0397) for 1 h, developed with aminoethylcarbazol as a substrate for HRP, and counter-stained with Mayer's hematoxylin. Mouse IgG antibody (DAKO X0931) was used as the negative isotype control for the monoclonal antibody recognizing CD34. NGS (1%) was used as the negative control for polyclonal sera recognizing osteocalcin and osteonectin.


  1. Top of page
  2. Abstract
  7. Acknowledgements

Bone histology

We describe a sinus, referred to as a BRC, which is intimately associated with cancellous bone remodeling. On two-dimensional histological sections, the narrow sinus runs close to the bone surface resembling a cleavage between the remodeling trabecular surface and the bone marrow (Fig. 1A). The sinus is lined on its marrow side by flattened cells and on its osseous side by the remodeling bone surface, making it resemble a canopy of flattened cells over the remodeling bone surface. The space itself may appear empty or may contain erythrocytes or a few nucleated cells.

In light microscopy, flattened cells with thin, flat nuclear profiles border the marrow side. The cells are intimately associated with the bone marrow and are never seen as a separate layer detached from the marrow (Fig. 1C). Flattened cells do not line the remodeling osseous side of the BRC; the BRC opens directly to the remodeling bone surface (Figs. 1C and 1E). At the resorptive margins of the BRC, that is, at the initiation of bone resorption, marrow lining cells are in direct continuity with the bone lining cells on the bone surface and osteoclasts are observed immediately under the marrow lining cells lying in small resorption lacunae only a few lamellae deep (Fig. 1E). Using light microscopy, the marrow lining cells cannot be morphologically distinguished from bone lining cells. At surfaces where bone formation is close to completion, the marrow lining cells are proximate to the flat surface osteoblasts (Fig. 1G). Some bone forming surfaces are not associated with BRCs, and characteristically they are not paved by osteoblasts (Fig. 1I). Frequently, resorptive surfaces are in direct continuity with formative surfaces letting bone resorption and bone formation proceed in a common compartment (Fig. 1A).

Bone histomorphometry

The extent of the BRC relative to BRC/BS, BRC/BS, BRC/BS, and BRC/QS as well as the fraction of the total BRC-covered surface associated with eroded surface (ES/BRC, %), osteoid surface (OS/BRC, %), and quiescent surface (QS/BRC, %) were measured in iliac crest biopsy specimens obtained in the high turnover state of primary hyperparathyroidism and 3 years after curative surgery (Table Table 1). Total bone surface covered by BRC decreases by 50% (p < 0.05) following surgery and is paralleled by a similar 50% decrease in remodeling bone surface (p < 0.05). Using mineralizing surface (MS/BS) as an index of bone turnover, Figure 2 shows the changes from preoperative to postoperative values in BRC/BS in relation to changes in MS/BS. The general trend is a fall in BRC-covered bone surface with a fall in bone turnover. Virtually all eroded surfaces, irrespective of the presence of resorbing cells, are covered by BRCs, and two-thirds of the osteoid surfaces are only a few percent of quiescent surfaces. There are no statistically significant differences between the high turnover and the normalized state concerning the fraction of various remodeling surfaces covered by a BRC. The distribution of BRC on eroded, osteoid, and quiescent surfaces differ significantly from preoperative to postoperative values on OS/BRC, which is reduced by one third (p < 0.05). The corresponding changes in ES/BRC and QS/BRC are not statistically significant.

Table Table 1.. Presence of Bone Remodeling Compartments (BRC) Over the Different Remodeling Surfaces and the Extent of the Remodeling Surfaces
Thumbnail image of
thumbnail image

Figure FIG. 2. The association between bone surface covered by bone remodeling compartments (BRC/BS) and bone turnover as indicated by mineralizing surface (MS/BS). Individual changes (—) in BRC/BS from preoperative (•) to postoperative values (○) in relation to changes in MS/BS BRC/BS.

Download figure to PowerPoint

The thickness of uncompleted walls associated with BRCs and not associated with BRCs were compared (Table 2). Walls without BRCs are 30% thicker than those covered by a BRC (before surgery, p < 0.001; after surgery, p < 0.05).

Table Table 2.. Thickness of Walls Associated With BRC Compared With Walls Not Associated With BRCs
Thumbnail image of

Because high quality sections were needed, the many exclusions (20 women and 30 men) lowered the statistical power of the study. However, statistical significance was achieved on the important indices of total bone surface covered by BRCs and on the thickness of walls with and without BRCs.

Immunohistochemistry and enzyme histochemistry

To address whether the phenotype of bordering cells are of endothelial cell type or bone cell type, immunostaining for osteocalcin, osteonectin, and myeloid progenitor cell antigen CD34 Class II (CD34) were applied together with enzyme-staining for alkaline phosphatase (Fig. 1). Osteocalcin and osteonectin are almost consistently expressed by the marrow lining cells (Figs. 1F and 1H) and at some formative surfaces also by osteoblasts, whereas lining cells at quiescent surfaces only occasionally express osteocalcin and osteonectin. Alkaline phosphatase is constantly displayed by osteoblasts and marrow lining cells (Fig. 1D); lining cells are alkaline phosphatase-negative. CD34 is expressed by the endothelium of small and large marrow vessels as well as by the endothelium of vessels in periosteal and striated muscle tissue, but the marrow lining cells do not express CD34 (Fig. 1J). Furthermore, the endothelium of marrow sinusoids is alkaline phosphatase-negative.


  1. Top of page
  2. Abstract
  7. Acknowledgements

The cellular events of cancellous bone remodeling is thought to be initiated by the digestion of the unmineralized matrix membrane by the lining cells; their disappearance and the arrival of osteoclasts attracted by the exposure of the bare bone surface.(4) As bone formation gradually ceases, the osteoblasts differentiate into bone lining cells(7) covering the quiescent bone surface until the next remodeling cycle is initiated. We suggest (Fig. 3) that the lining cell barrier persists during bone remodeling but is released from the cancellous bone surface by a disruption of the junctions between lining cells and embedded osteocytes. The old lining cells become the marrow lining cells letting bone resorption and bone formation proceed under a common roof of lining cells. At the end of bone formation, new bone lining cells derived from the flattened surface-osteoblasts replace the marrow lining cells thereby closing the BRC. The integrity of the osteocyte-lining cell system is reestablished by the new generation of lining cells. The two layers of lining cells, the old and the new, eventually becomes a single layer of lining cells on the quiescent bone surface.

thumbnail image

Figure FIG. 3. A schematic, not-to-scale presentation of a cancellous bone remodeling compartment during a complete remodeling cycle. Osteoclastic bone resorption begins under the lining cells that persist during the entire remodeling cycle. The final resorption depth is reached and bone formation follows. Osteoblasts differentiate into lining cells and the two lining cell layers becomes a single cell layer. During the rebuilding of bone the network between osteocytes and bone lining cells is reestablished.

Download figure to PowerPoint

Bone morphometry showed that practically all resorptive surfaces, more than one-half of the formative surfaces, and a few percent of the quiescent surfaces are covered by BRCs; and, that walls without a BRC are thicker than those covered by one. This implies that the formation of the BRC coincides with the initiation of bone resorption, and that it closes with the completion of bone formation. The closure of the BRC probably follows from the completion of matrix synthesis and the differentiation of osteoblasts into bone lining cells. Bone mineralization may, however, continue after the closure of the BRC. Because the formative rate declines during the formative period, more nearly completed than newly initiated walls are present in a cross-sectional study, explaining the observed positively skewed distribution of formative surfaces. Therefore, the presence of osteoid surface classified as “walls without BRC” may rather be osteoid surfaces where the BRC is collapsed representing the very last part of bone formation. The observation that the thickness of walls associated with BRCs is lower than those without BRCs supports this interpretation. The presence of BRCs over quiescent surfaces could represent activated surface that eventually will pass into bone resorption and run through a complete remodeling cycle.

The flat cells that directly appose quiescent cancellous bone surfaces, the bone lining cells, extend their cytoplasm over the bone surface enveloping the bone(1); they are joined by junctions(21) and regulate calcium transport in and out of bone.(2) They are arranged in a single cell layer between the bone and the hematopoietic compartment,(22-25) yet there are areas without lining cells(24) and areas with more than one cell layer,(24) but this may be explained by artifacts or species differences.(26) Menton et al.(27) described a marrow sac comprising flat cells investing the bone marrow. They used a fracturing method whereby the cells in the marrow sac were physically separated with the marrow from the pavement of cells lying directly on the bone. The bone cells were described as osteoblasts, but it is not clear whether they are truly osteoblasts or bone lining cells because the two cell types are not easily distinguished by scanning electron microscopy (SEM). Supposing the cells on the bone surface were lining cells, then instead of one cell layer,(22-25) two cell layers, that is, one layer of bone lining cells and another of marrow sac cells would separate bone from the marrow compartment. Alternatively, if the cells on the bone surface were osteoblasts, then one layer of marrow sac cells would separate the osteoblasts from the marrow. It is proposed that the lining cells could persist as a canopy over the BMU during bone remodeling.(6, 28) Our observations are in line with this, and we suggest that the cells covering cancellous bone remodeling surfaces, referred to as marrow sac cells,(26, 27) are bone lining cells.

Ultrastructural studies of the intact bone-marrow interface show that flattened fibroblastlike cells cover the surface of osteoblasts(21, 22, 25) and possibly also osteoclasts.(22) Although it is anticipated that these cells are preosteoblasts, they may be bone lining cells with an osteogenic potential or they may even be the osteoprogenitor pool from which osteoblasts are derived. In rats, lining cells are activated to form bone de novo following treatment with intermittent parathyroid hormone(29) or exposure to mechanical loading.(30) Fibroblastlike mesenchymal cell populations exist that can differentiate into osteoblasts,(31) and some of these cell populations reside at the bone-marrow interface.(32) A so-called paratrabecular mesenchymal reaction with formation of loose connective tissue coinciding with intensive osteoblast recruitment is noted in the beginning of bone formation.(8) Accordingly, transmission electron microscopy (TEM) investigations show that multiple layers of putative preosteoblasts are present at some bone forming surfaces.(21, 22) We show that the flat cells covering formative surfaces stain positive for the osteoblast markers alkaline phosphatase, osteocalcin, and osteonectin, suggesting that they have the bone cell phenotype, and we suggest that they are lining cells. Lining cells covering quiescent bone surfaces only occasionally express osteocalcin and osteonectin, indicating that the lining cells change their metabolic state when they are activated at the beginning of bone remodeling. At the very end of bone formation we observed two layers of cells covering the bone surface, which could be one layer of flattened osteoblasts differentiating into bone lining cells and another layer of marrow lining cells (prior bone lining cells) that bordered the bone surface during the entire remodeling cycle. This is supported by TEM investigations showing that some bone surfaces are covered by two layers of flat cells.(24)

It is reported that some cortical BMUs stop their resorbing activity and permanently interrupt the normal bone remodeling sequence,(33, 34) and that a similar event occurs in cancellous bone.(35) The observations leading to this theory are that many more shallow resorption lacunae existing than are expected from the changes in resorption rate with time. There are alternative explanations to the permanently interrupted BMU activity, which are a temporarily interrupted bone resorption being comparable with the on-off phenomenon of bone formation or premature reversal of bone resorption into bone formation. Recent evidence suggests that cortical remodeling occurs in spatial clusters of synchronously activated osteons, super-osteons, that eventually coalesce.(36) Stereologically, this means that an individual osteonal profile may belong to a connected, branching system of cortical osteons. Similarly in cancellous bone, two individual BMU profiles may very well belong to the same three-dimensional BMU.(37, 38) Thus an individual, two-dimensional osteonal profile does not unambiguously identify a three-dimensional BMU,(39) which may have significant implications for the theory of abortive resorptions. Furthermore, the surfaces with interrupted resorption are lined with flat cells indistinguishable from those lining quiescent surfaces,(34) which is taken as an indication of the permanently interrupted BMU activity. It is however, in line with the BRC theory suggesting that flat cells border the interface between bone and marrow throughout the remodeling sequence; thus, arrested bone resorption is not in conflict with the presence of a bone remodeling compartment.

It is observed that a vascular sinus closely conforming to the surface of trabeculae and referred to as a paratrabecular sinusoidal capillary is associated with bone forming surfaces,(8) and it is known that the vascular system is associated with osteogenesis during bone development(40-42) and bone remodeling in vitro(43, 44) and in vivo.(45-47) The anatomical structure described by Burkhardt et al.,(8) in which an osteoblastic layer makes up the osseous wall of the capillary and endothelial cells the opposite wall, is very similar to what is noted by Bi et al.(26) and to the BRC that we observe. In iliac crest biopsy specimens the BRC often contains blood cells but it may also have marrow cells or fat cells in its lumen, which could have been forced into the compartment during the biopsy procedure because of the hydrostatic pressure of the injured microcirculation. Using a methyl methacrylate corrosion casting technique in human vertebral bodies, it is shown that small tubular extensions reach from the surface of marrow sinusoids toward areas of bone resorption and formation.(48) These extensions could be direct connections between the marrow vessels and the BRC establishing a circulation through the compartment. Burkhardt et al.(8) suggested that the flat cells bordering the marrow were endothelial cells. In different tissues, endothelial cells show different characteristics, but CD34 is characteristic of essentially all endothelial cells.(49) CD34 is not expressed by the marrow lining cells of the BRC but is extensively expressed by the endothelium of bone marrow sinusoids. It has been recently shown that cancer cells may form blood supply channels that are not lined by endothelial cells(50) and that osteoblasts are polarized(51) and have mutual zonula-adherens-like junctions at their membrane facing the BRC.(21) Thus, the absence of endothelial markers in the flat marrow lining cells does not preclude the BRC from being of vascular nature. If the BRC is not a vascular compartment, then the finding that the marrow sac cells (i.e., the bone lining cells) apparently do not have junctions(27) could still ensure an effective exchange of amino acids and minerals between the bone compartment and the extracellular fluid being essential to both bone resorption and bone formation. As discussed, fibroblastlike populations of mesenchymal cells possess osteogenic potential, but so do mesenchymal cells associated with blood vessels,(52-54) so suggesting that the marrow lining cells are endothelial cells may still locate the site of osteoprogenitor cells to the wall opposing the marrow.

The BRC most likely serves multiple purposes. It could assure a rapid and efficient exchange of matrix constituents and minerals between the bone fluid compartment and the extracellular fluid compartment during the process of bone remodeling. Second, it could be the route, the monitor, the modulator, or even the source of bone cell recruitment. Mitoses are not seen in bone cells at the bone surface but they are observed in the adjacent marrow as part of the generalized mesenchymal reaction including mitotic activity of endothelial cells, fibroblasts, and preosteoblasts.(8) Third, it could be the anatomical basis for the coupling of bone remodeling. Previous attempts to integrate the microvascularity of bone and the process of bone remodeling have lead to conflicts with the coupling theory.(8) However, our observations may be interpreted in agreement with this theory. The BRC may provide a microenvironment that links bone formation to bone resorption through local signaling. The activating events leading to initiation of the remodeling process through osteoclastic resorption in certain areas on the bone surface are not fully understood. The osteocyte may hold a central position in determining the initiation, the extent, and the depth of bone resorption through propagating events between neighboring osteocytes.(55) Similarly, the thickness of the new bone structural unit (BSU) could be determined by nutritional demands of the osteocytes.(55) If the BRC is no obvious candidate for controlling the activation, propagation, termination or the balance of bone remodeling, there remains numerous possibly integrating roles for the BRC to play in bone remodeling and perhaps disorders of bone remodeling.


  1. Top of page
  2. Abstract
  7. Acknowledgements

The Danish Research Council, The Novo Nordic Center for Research in Growth and Regeneration, and the Center of Molecular Gerontology supported the study.

We thank laboratory technicians Rita Ullerup, Anette Baatrup, and Jette Barlach for preparation of bone samples. Drs. Peer Christiansen and Torben Steiniche are thanked for kindly making available the bone samples from patients with primary hyperparathyroidism.


  1. Top of page
  2. Abstract
  7. Acknowledgements
  • 1
    Baud CA 1968 Submicroscopic structure and functional aspects of the osteocyte. Clin Orthop 56:227236.
  • 2
    Talmage RV 1970 Morphological and physiological considerations in a new concept of calcium transport in bone. Am J Anat 129:467476.
  • 3
    Hattner R, Epker BN, Frost HM 1965 Suggested sequential mode of control of changes in cell behaviour in adult bone remodeling. Nature 206:489490.
  • 4
    Chambers TJ, Fuller K 1985 Bone cells predispose bone surfaces to resorption by exposure of mineral to osteoclastic contact. J Cell Sci 76:155165.
  • 5
    Jaworski ZFG 1971 Some morphologic and dynamic aspects of remodeling on the endosteal-cortical and trabecular surfaces. In: MenczelJ, HarellA (eds.) Calcified Tissue, Structural, Functional and Metabolic Aspects. Academic Press, New York, NY, USA, pp. 159160.
  • 6
    Parfitt AM 1994 Osteonal and hemi-osteonal remodeling: The spatial and temporal framework for signal traffic in adult human bone. J Cell Biochem 55:273286.
  • 7
    Marotti G 1996 The structure of bone tissues and the cellular control of their deposition. Ital J Anat Embryol 101:2579.
  • 8
    Burkhardt R Bartl R Frisch B Jäger K Mahl C Hill W Kettner G 1984 The structural relationship of bone forming and endothelial cells of the bone marrow. In: ArletJ, FicatRP, HungerfordDS (eds.) Bone Circulation. Williams and Wilkins, Baltimore, MD, USA, pp. 214.
  • 9
    McClugage SJ, McCuskey RS 1973 Relationship of bone blood flow to bone resorption and growth in situ. Microvscular Research 6:132134.
  • 10
    Marotti G, Zambonin Zallone A 1980 Changes in vascular network during the formation of Haversian systems. Acta Anat 106:84100.
  • 11
    Parfitt AM 2000 The mechanism of coupling: A role for the vasculature. Bone 26:319323.
  • 12
    Steiniche T, Christiansen P, Vesterby A, Ullerup R, Hessov I, Mosekilde LE, Melsen F 2000 Primary hyperparathyroidism: Bone structure, balance, and remodeling before and 3 years after surgical treatment. Bone 26:535543.
  • 13
    Hauge E, Mosekilde L, Melsen F 1999 Missing observations in bone histomorphometry on osteoporosis. Implications and suggestions for an approach. Bone 25:389395.
  • 14
    Baddeley AJ, Gundersen HJG, Cruz-Orive LM 1986 Estimation of surface area from vertical sections. J Microsc 142:259276.
  • 15
    Parfitt AM, Drezner MK, Glorieux FH, Kanis JA, Malluche H, Meunier PJ, Ott SM, Recker RR 1987 Bone histomorphometry: Standardization of nomenclature, symbols, and units. J Bone Miner Res 2:595610.
  • 16
    Kragstrup J, Gundersen HJG, Melsen F, Mosekilde L 1982 Estimation of the three-dimensional wall thickness of completed remodeling sites in iliac trabecular bone. Metab Bone Dis Relat Res 4:113119.
  • 17
    Steiniche T, Christiansen P, Vesterby A, Hasling C, Ullerup R, Melsen F 1992 Reconstruction of the formative site in trabecular bone by a new, quick, and easy method. Bone 13:147152.
  • 18
    Erben RG 1997 Embedding of bone samples in methylmethacrylate: An improved method suitable for bone histomorphometry, histochemistry, and immunohistochemistry. J Histochem Cytochem 45:307313.
  • 19
    Ingram RT, Clarke BL, Fisher LW, Fitzpatrick LA 1993 Distribution of noncollagenous proteins in the matrix of adult human bone: Evidence of anatomic and functional heterogeneity. J Bone Miner Res 8:10191029.
  • 20
    Riminucci M, Fisher LW, Shenker A, Spiegel AM, Bianco P, Gehron RP 1997 Fibrous dysplasia of bone in the McCune-Albright syndrome: Abnormalities in bone formation. Am J Pathol 151:15871600.
  • 21
    Yamazaki K, Eyden BP 1995 A study of intercellular relationships between trabecular bone and marrow stromal cells in the murine femoral metaphysis Anat Embryol (Berl) 192:920.
  • 22
    Luk SC, Nopajaroonsri C, Simon GT 1974 The ultrastructure of the endosteum: A topographic study in young adult rabbits. J Ultrastruct Res 46:165183.
  • 23
    Vander Wiel J, Grubb SA, Talmage RV 1978 The presence of lining cells on surfaces of human trabecular bone. Clin Orthop 134:350.
  • 24
    Miller SC, Bowman BM, Smith JM, Jee WS 1980 Characterization of endosteal bone-lining cells from fatty marrow bone sites in adult beagles. Anat Rec 198:163173.
  • 25
    Deldar A, Lewis H, Weiss L 1985 Bone lining cells and hematopoiesis: An electron microscopic study of canine bone marrow. Anat Rec 213:187201.
  • 26
    Bi LX, Simmons DJ, Hawkins HK, Cox RA, Mainous EG 2000 Comparative morphology of the marrow sac. Anat Rec 260:410415.
  • 27
    Menton DN, Simmons DJ, Orr BY, Plurad SB 1982 A cellular investment of bone marrow. Anat Rec 203:157164.
  • 28
    Rasmussen HB Bordier PHJ 1974 The Physiological and Cellular Basis of Metabolic Bone Disease, Williams and Wilkins, Baltimore, MD, USA
  • 29
    Dobnig H 1995 Evidence that intermittent treatment with parathyroid hormone increases bone formation in adult rats by activation of bone lining cells. Endocrinology 136:36323639.
  • 30
    Chow JW, Wilson AJ, Chambers TJ, Fox SW 1998 Mechanical loading stimulates bone formation by reactivation of bone lining cells in 13-week-old rats. J Bone Miner Res 13:17601767.
  • 31
    Benayahu D, Kletter Y, Zipori D, Wientroub S 1989 Bone marrow-derived stromal cell line expressing osteoblastic phenotype in vitro and osteogenic capacity in vivo. J Cell Physiol 140:17.
  • 32
    Ashton BA, Eaglesom CC, Bab I, Owen ME 1984 Distribution of fibroblastic colony-forming cells in rabbit bone marrow and assay of their osteogenic potential by an in vivo diffusion chamber method. Calcif Tissue Int 36:8386.
  • 33
    Jaworski ZFG, Meunier PJ, Frost HM 1972 Observations on two types of resorption cavities in human lamellar cortical bone. Clin Orthop Rel Res 83:279285.(abstract)
  • 34
    Parfitt AM 1983 The physiologic and clinical significance of bone histomorphometric data. In: ReckerRR (ed.) Bone Histomorphometry: Techniques and Interpretation, 1st ed., CRC Press, Boca Raton, FL, USA, pp. 143221.
  • 35
    Parfitt AM 1993 Morphometry of bone resorption: Introduction and overview. Bone 14:435441.
  • 36
    Bell KL, Loveridge N, Reeve J, Thomas D, Feik S, Clement J 2000 Cortical remodelling clusters (super-osteons) in the human femoral shaft. J Bone Miner Res 15:S1; S237.(abstract)
  • 37
    Jones SJ, Boyde A 1993 Histomorphometry of Howship's lacunae formed in vivo and in vitro: Depths and volumes measured by scanning electron and confocal microscopy. Bone 14:455460.
  • 38
    Kragstrup J, Melsen F 1983 Three-dimensional morphology of trabecular bone osteons reconstructed from serial sections. Metab Bone Dis Relat Res 5:127130.
  • 39
    Hauge E, Mosekilde L, Melsen F 1994 Stereological considerations concerning the measurements of individual osteoid seams and resorption cavities. Bone Miner 26:8990.
  • 40
    Winet H, Bao JY, Moffat R 1990 A control model for tibial cortex neovascularization in the bone chamber. J Bone Miner Res 5:1930.
  • 41
    Hunter WL, Arsenault AL, Hodsman AB 1991 Rearrangement of the metaphyseal vasculature of the rat growth plate in rickets and rachitic reversal: A model of vascular arrest and angiogenesis renewed. Anat Rec 229:453461.
  • 42
    Pechak DG 1986 Morphology of bone development and bone remodeling in embryonic chick limbs. Bone 7:459472.
  • 43
    Macintyre IM, Zaidi M, Alam AS, Datta HK, Moonga BS, Lidbury PS, Hecker M, Vane JR 1991 Osteoclastic inhibition—an action of nitric oxide not mediated by cyclic GMP. Proc Natl Acad Sci USA 88:29362940.
  • 44
    Alam AS, Gallagher A, Shankar V, Ghatei MA, Datta HK, Huang CL, Moonga BS, Chambers TJ, Bloom SR, Zaidi M 1992 Endothelin inhibits osteoclastic bone resorption by a direct effect on cell motility: Implications for the vascular control of bone resorption. Endocrinology 130:36173624.
  • 45
    Semb H 1969 Experimental limb disuse and bone blood flow. Acta Orthop Scand 40:552562.
  • 46
    Wootton R, Tellez M, Green JR, Reeve J 1981 Skeletal blood flow in Paget's disease of bone. Metab Bone Dis Relat Res 3:263270.
  • 47
    Burkhardt R, Kettner G, Bohm W, Schmidmeier M, Schlag R, Frisch B, Mallmann B, Eisenmenger W, Gilg T 1987 Changes in trabecular bone, hematopoiesis and bone marrow vessels in aplastic anemia, primary osteoporosis, and old age: A comparative histomorphometric study. Bone 8:157164.
  • 48
    Møller JF 1999 Microvascular morphology in bone studied in experimental arthritis in the rabbit knee joint and in lumbar vertebral bodies from elderly humans, Ph.D. thesis, University of Aarhus, Aarhus, Denmark.
  • 49
    Garlanda C, Dejana E 1997 Heterogeneity of endothelial cells. Specific markers. Arterioscler Thromb Vasc Biol 17:11931202.
  • 50
    Maniotis AJ, Folberg R, Hess A, Seftor EA, Gardner LM, Pe'Er J, Trent JM, Meltzer PS, Hendrix MJ 1999 Vascular channel formation by human melanoma cells in vivo and in vitro: Vasculogenic mimicry. Am J Pathol 155:739752.
  • 51
    Ilvesaro J, Metsikko K, Vaananen K, Tuukkanen J 1999 Polarity of osteoblasts and osteoblast-like UMR-108 cells. J Bone Miner Res 14:13381344.
  • 52
    Brighton CT, Hunt RM 1991 Early histological and ultrastructural changes in medullary fracture callus. J Bone Joint Surg (Am) 73:832847.
  • 53
    Diaz-Flores L, Gutierrez R, Lopez-Alonso A, Gonzalez R, Varela H 1992 Pericytes as a supplementary source of osteoblasts in periosteal osteogenesis. Clin Orthop 275:280286.
  • 54
    Doherty MJ, Ashton BA, Walsh S, Beresford JN, Grant ME, Canfield AE 1998 Vascular pericytes express osteogenic potential in vitro and in vivo. J Bone Miner Res 13:828838.
  • 55
    Martin RB 2000 Toward a unifying theory of bone remodeling. Bone 26:16.