Cell matrix plaques of initial bone formation

Abstract

An isolated and assentially purified cell matrix plaque of initial bone formation comprising of αvβ3 integrin and rapid assays using such cell matrix plaques to measure potentials of factors, regimens or tissues for simulation and/or inhibition of bone formation.

Description

BACKGROUND OF THE INVENTION

[0003] This invention relates to cell matrix plaques of initial bone formation and to rapid in vitro assays using such newly discovered cell matrix plaques to measure potentials of factors or tissues for stimulating and/or inhibiting bone formation. In particular, the invention relates to assays utilizing the process of αvβ3 expression, matrix protein secretion into the plaques and mineralization of such plaques developed using primary cultures from the human osteoblast lineage.

[0004] Bone cells are known to transduce mechanical signals from the environment into anabolic biochemical responses through the process of mechanotransduction. However, the precise mechanisms of mechanotransduction have been a mystery.

[0005] Osteoblasts are the skeletal cells that produce and regulate the deposition and mineralization of calcified bone matrix under the influence of the bone morphogenetic proteins (1-4).

[0006] Osteoblasts derive from osteoprogenitors, which in turn arise from multipotential mesenchymal cells (5,6). Osteoprogenitors exist in an uncommitted stage which may differentiate into one of several cell lineages, including osteoblasts, chondrocytes, fibroblasts, myocytes and adipocytes, depending on growth conditions (7-9).

[0007] The progression from early progenitors to fully functional osteoblasts that synthesize and mineralize matrix is gradual(10).

[0008] Differentiation leads through the preosteoblast stage, defined as cells expressing osteoblastic characteristics but not yet mineralizing the extracellular matrix.

[0009] The cells in the osteoblast program can be identified based on distinct morphological features and histochemical markers.

[0010] Mechanical strain is known to have potent anabolic effects on the skeleton and bone homeostasis in vivo (11), and well documented proliferative effects both in vivo and in vitro (12,13).

[0011] The tensegrity model of mechanotransduction (14) proposes that mechanical signals are integrated and converted to biochemical responses through changes in cellular architecture.

[0012] Transmembrane molecules (integrins) connecting extracellular and cytoskeletal structures may play key roles in tensegrity, however, the precise processes and molecules that transduce mechanical stimuli into cellular responses, including proliferation are not known.

[0013] Integrins are a diverse family of heterodimeric cell surface receptors which connect the cytoskeleton to the extracellular matrix and mediate a variety of signaling cascades.

[0014] The rationale for adapting the tensegrity hypothesis to the response of the cells in the osteoblast lineage to mechanical strain derived from the findings that repetitive strain of human osteoblast-like cells stimulated cell proliferation, transcription of bone matrix proteins and production of bone matrix (15-17).

[0015] The communication between osteoblasts and extracellular matrix proteins is believed to rely on integrins that are known to serve as receptors for matrix proteins.

[0016] Even though the precise role of integrins in human bone formation is just beginning to unravel, a number of cellular processes including proliferation are thought to result from the interactions between integrins present on bone cell surface and extracellular matrix proteins.

[0017] Human osteoblasts are known to express numerous α and β integrin subunits (18,19), each component having large extracellular domains responsible for ligand binding, a transmembrane domain and a short cytoplasmic domain responsible for interacting with the actin cytoskeleton (20,21).

[0018] Integrins are known to associate with signaling molecules in focal adhesions.

[0019] Focal adhesions are defined as cell—matrix adhesion structures which contain talin, vinculin and other proteins required for integrin attachment to actin filaments, and have been described in numerous cell types but not osteoblasts.

[0020] Focal adhesions play dual roles as cell signaling structures as well as architectural links between the cellular cytoskeleton and the extracellular matrix proteins (22).

[0021] One of the extracellular matrix proteins of bone, osteopontin, is a non-collagenous glycoprotein that binds to cell surface integrins via the adhesive arginine-glycine-aspartate (RGD) motif (23-25).

[0022] Osteoblasts express high levels of osteopontin during bone matrix maturation and mineralization, where it is known to modulate (either inhibit or stimulate) the mineralization of bone matrix (16).

[0023] In addition, osteopontin promotes cellular adherence of osteoblast-like cells (26-29) and was shown to specifically interact with the αvβ3 integrin (25,26,30).

[0024] A precise method for measuring new bone formation, particularly at the subcellular level, has long eluded those interested in bone disease, repair and regeneration. The absence of such a definitive method makes it difficult to evaluate target compositions or regimens for their ability to stimulate or inhibit bone formation or extracellular matrix mineralization. Prior art methods generally suffer from inaccuracy, require making gross determinations from whole cultures, and are typically both time and labor intensive.

SUMMARY OF THE INVENTION

[0025] Among the objects of the invention, therefore, may be noted the provision of isolated and essentially purified plaques of initial bone formation and methods of assaying utilizing such plaques. Such assays may be used to screen substances or regimens which stimulate or inhibit bone formation or matrix mineralization. Since the assays operate at the subcellular level, only minute amounts of tissue are required to conduct the assay. Further, the assays may be conducted repeatedly using a single bone donor. The assays can be conducted rapidly and are not labor intensive.

[0026] Briefly, therefore, the present invention is directed to a method of measuring new bone formation using a rapid in vitro assay.

[0027] The method involves:

[0028] 1. extracting bone marrow stromal cells from bone marrow cavities and growing them to semi-confluence;

[0029] 2. exposing the cells to a candidate medium or regimen for osteogenesis or inhibition of osteogenesis;

[0030] 3. determining the mineralization/development of bone plaque; and,

[0031] 4. optionally, comparing the mineralization/development of bone plaque generated using the candidate medium or regimen with the mineralization/development of bone plaque using a preselected control such as bone morphogenic factor-7 (BMP-7), also known as osteogenic factor-1 (OP-1).

[0032] Alternatively, the assay can be used as a diagnostic screen for healthy and/or deficient bone formation, making the variable the cells being observed for production of mineralization/bone plaque formation.

[0033] Other objects and features will be in part apparent and in part pointed out hereinafter.

BRIEF DESCRIPTION OF DRAWINGS AND ABBREVIATIONS

[0034] The following abbreviations used herein have the following meaings:

[0035] RGD-arginine-glycine-aspartate, BMP-bone morphogenetic protein, BSA-bovine serum albumin, Cy2-Cy2 cyanine, Cy3-indocarbocyanine, FAK-focal adhesion kinase, FCS-fetal calf serum, FITC-fluorescin isothiocyanate, OP-1-osteogenic protein-1

[0036] Figure legends:

[0037] FIG. 1 depicts localization of the αvβ3 integrin in human osteoprogenitor cells. The αvβ3 integrin was detected by immunostaining in the presence (images A and B) or in the absence (images C and D) of Triton TX-100 in the fixation protocol of cells mechanically strained for 48 hours (images B and D) and in control cells (images A and C). Cells were incubated with a mouse monoclonal antibody directed against anti-human αvβ3 integrin, followed by 7 an incubation with a secondary Cy-3 conjugated anti-mouse antibody, as described in Methods. The presented images were generated using Zeiss confocal microscope and are representative of greater than 15 experiments with similar results. The magnification of the optical system used was 630.

[0038] FIG. 2 depicts localization of the αvβ3 integrin in human preosteoblasts. The αvβ3 integrin was immunolocalized in osteoprogenitor cells incubated in OP-1 (BMP-7)-containing osteogenic medium for 48 hours in the presence (images A and B) or in the absence (images C and D) of Triton TX-100 cells mechanically strained for 48 hours (images B and D ) and in control cells (images A and C). Cells were incubated with a mouse monoclonal antibody directed against anti-human αvβ3 integrin, followed by an incubation with a secondary Cy-3 conjugated anti-mouse antibody, as described in Methods. The presented images are generated using a Zeiss confocal microscope and are representative of greater than 6 experiments with similar results. The magnification of the optical system used was 630.

[0039] FIG. 3 depicts colocalization of the αvβ3 integrin with Focal Adhesion Kinase in human preosteoblasts. The αvβ3 integrin shown in image B was colocalized (as presented in image C) with Focal Adhesion Kinase (image A) in the absence of Triton TX-100 in preosteoblasts incubated in OP-1 (BMP-7)-containing osteogenic medium and mechanically strained for 48 hours. Arrows indicated a few of the sites where the expression of FAK was confined to the boundaries of the αvβ3 integrin plaques. Cells are incubated with a combination of a mouse monoclonal antibody directed against anti-human αvβ3 integrin and a rabbit polyclonal antibody against FAK, followed by an incubation with a mix of two secondary antibodies: Cy-2 conjugated anti-mouse and Cy-3 conjugated anti-rabbit antibody, as described in Methods. The presented images were generated using a Zeiss confocal microscope and are representative of greater than 10 experiments with similar results. Colocalization of αvβ3 integrin with FAK in the plaques following strain was observed in osteoprogenitors similar to the preosteoblasts shown in this figure. The magnification of the optical system used was 630.

[0040] FIG. 4 is a graphic representation of flow cytometric analysis of the expression of the αvβ3 integrin on the cell surface of human osteoprogenitors and preosteoblasts. The imunofluorescence intensity due to the cell surface expression of the αvβ3 integrin was examined in single cell suspensions of cells mechanically strained for 48 hours as well as in their non-strained controls. Osteoprogenitors (left-side panels) and preosteoblasts (right-side panels) were incubated with a mouse monoclonal antibody directed against anti-human αv⊖3 integrin, followed by an incubation with a secondary FITC-conjugated anti-mouse antibody, as described in Methods. The presented histograms showing the fluorescene intensity (X axes) plotted against the relative cell number (Y axes) reflect the results of at least 3 experiments done in each cell culture type. The A, A′, C and C′ plots represent the fluorescence detected with a negative control mouse IgG. The B, B′, D and D′ plots who the fluorescence level observed in cells incubated with the anti-αvβ3 integrin antibody. A, B, A′ and B′ show the results of flow cytometric analysis in non-strained control cells, while C, D, C′ and D′ show the fluorescence level in strained cells. Each histogram is based on 10,000 events. The fluorescence levels shown in plots D (strained progenitors) and D′ (strained preosteoblasts) were significantly greater than those in B (non-strained progenitors and B′ (non-strained preosteoblasts) were significantly greater than those in B (non-strained progenitors) and B′ (non-strained preosteoblasts), respectively at p≦0.01.

[0041] FIG. 5 depicts colocalization of the αvβ3 integrin with osteopontin in non-permeabilized human osteoprogenitors. The αvβ3 integrin (shown in green; image B) was colocalized (image C) with osteopontin (shown in red; image A) in the absence of Triton TX-100 in cells mechanically strained for 48 hours. Cells were incubated with a combination of a mouse monoclonal antibody directed against anti-human αvβ3 integrin and a rabbit polyclonal antibody against human osteopontin, followed by an incubation with a mix of two secondary antibodies: Cy-2 conjugated anti-mouse and Cy-3 conjugated anti-rabbit antibody, as described in Methods. The presented images were generated using a Zeiss fluorescent microscope and are representative of at least 15 experiments. The magnification was 630.

[0042] FIG. 6 depicts analysis of the effects of mechanical strain on osteopontin secretion in human progenitors and preosteoblasts. Osteopontin levels were analyzed in the lysates (FIG. 6b) as well as in the concentrated cell culture medium (FIG. 6a) from the progenitor and preosteoblast cultures following mechanical strain for 48 hours (Str.), as well as in non-strained control cultures (Cont.). Following gel electrophoresis of equal amounts of cell culture medium protein in each well, and protein transfer onto Immobilon-P membrane, the protein blot was incubated with a rabbit polyclonal antibody against human osteopontin, followed by an incubation with a protein A—conjugated HRP, as described in Methods. Mechanical strain increased secretion of the immunoreactive protein band of highest intensity detected at approximately 168 kDa into cell culture medium. The densitometry rations of the 168 kDa band intensity in strained versus control cultures were 2.2±0.3 in progenitors and 0.92±0.06 in preosteoblasts (FIG. 6a). The strain over control stimulation index determined by scanning densitometry of the 78 kDa bands was approximately 1.0 for cell lysates (FIG. 6b). The presented data and densitometry ration values (mean±SE) are representative of 5 experiments to analyze osteopontin secretion into culture medium (FIG. 6a) and 3 experiments to examine osteopontin levels in cell lysates (FIG. 6b).

[0043] FIG. 7 depicts colocalization of tissue transglutaminase with the αvβ3 integrin in human osteoprogenitors. We examined the distribution of tissue transglutaminase in Triton TX-100-permeabilized cells (image A) and as well as in non-permeabilized cells (image B). Our attention was focused on the relatively large plaques of tissue transglutaminase that were visualized on the cell surface under non-permeabilizing conditions (image B). The co-immunostraining under such conditions of cell surface tissue transflutaminase (shown in image C) and the αvβ3 integrin (shown in image D) demonstrated the colocalization of the two molecules (image E) in cells mechanically strained for 48 hours. Images C-E were obtained from bone cells that were incubated with a combination of a rabbit antibody directed against tissue transglutarninase and a mouse monoclonal antibody against human αvβ3 integrin, followed by an incubation with a mix of two secondary antibodies: Cy-3 conjugated anti-rabbit and Cy-2 conjugated anti-mouse antibody, as described in Methods. The presented images were generated using a Zeiss fluorescent microscope and are representative of at least 5 experiments. The magnification was 630.

[0044] FIG. 8 depicts the effect of mechanical strain on the formation of mineralization nodules. Human bone marrow stromal cells were maintained in α-MEM supplemented with 10% FCS in culture until confluent. Cells were then exposed to osteogenic medium containing α-MEM supplemented with 10% FCS, 10 mM of β-glycerophosphate, 50 μg/ml of ascorbic acid and 40 ng/ml of osteogenic protein OP-1 for 18 days, mechanically strained for 48 hours (days 18-20) and then both strained cells (shown in FIGS. 8B, 8D and 8F) and their non-strained controls (shown in FIGS. 8A, 8C and 8E) were examined for the formation of mineralized nodules by alizarin red-S staining. Cells were fixed in 70% ice-cold ethanol for 1 hour and strained with 0.5% (w/v) alizarin red-S for 10 minutes, as described in Methods. Each pair of images (A-B, C-D and E-F) represent an independent experiment. The data is representative of at least 10 experiments. FIG. 8G shows colocalization (yellow) of mineralized nodules (as visualized by 1 μM calcein) and the αvβ3 integrin in cells mechanically strained for 48 hours. Cells are incubated with a combination of a mouse monoclonal antibody against human αvβ3 integrin and calcein, followed by an incubation with a secondary Cy-3 conjugated anti-mouse antibody, as described in Methods. The presented images were generated using a Zeiss fluorescent microscope and are representative of at least 5 experiments.

[0045] FIG. 9 depcits osteoblastic focal adhesions which do not contain the αvβ3 or the αvβ3 integrin. The αv integrin subunit was immunodetected in focal adhesions of mechanically strained osteoprogenitors (image A) using a mouse monoclonal antibody directed against human αv integrin subunit, followed by an incubation with a secondary Cy-3 conjugated anti-mouse antibody, as described in Methods. The presented αv integrin image was generated using a Zeiss confocal microscope. In order to determine the identity of the αv integrin in focal adhesions, the localization of the β5 and β6 integrin subunits, as shown in images B and D, respectively, was examined with primary rabbit polyclonal antibodies directed against these integrin subunits followed by a secondary Cy-3 conjugated anti-rabbit antibody. Image C shows the colocalization of the β5 integrin subunit (stained with a secondary Cy-3 conjugated antibody) with the αvβ3 integrin (stained with a secondary Cy-2 conjugated antibody), as described in Methods. Arrows indicate long cellular processes and arrowheads point to sites where cellular processes abut other cell bodies at their end. The presented images were generated using a Zeiss fluorescent microscope and are representative of at least 3 experiments. The magnification of the optical system was 630.

[0046] FIG. 10 depicts identification of the αv integrin present in focal adhesions. The αv integrin subunit (shown in image A) was colocalized (image C) with the P integrin subunit (shown in image B) in cells mechanically strained for 48 hours. Cells were incubated with a combination of a mouse monoclonal antibody directed against anti-human αv integrin subunit and a rabbit polyclonal antibody against human β1 integrin subunit, followed by an incubation with a mix of two secondary antibodies: Cy-2 conjugated anti-mouse and Cy-3 conjugated anti-rabbit antibody, as described in Methods. The presented images were generated using a Zeiss fluorescent microscope and are representative of at least 8 experiments. The magnification was 630.

DESCRIPTION OF THE PREFERRED EMBODIMENT

[0047] The contents of each of the references cited herein are incorporated by reference.

[0048] In accordance with the present invention it has been discovered that primary bone cells adhering to vitronectin and Type I collagen, demonstrate a unique form of cell—matrix interaction and a major upregulation of the αvβ3 integrin following mechanical strain. The αvβ3 integrin localized to plaque-like sites of cell/substratum interaction along with osteopontin which was expressed as a macromolecule, possibly cross-linked by tissue transglutaminase which was also localized to the plaques. Such bone plaque sites, not previously isolated or essentially purified, are herein shown to be key indicators of bone formation and mineralization at the sub-cellular level. Such an indicator can be used, as described in detail below, to evaluate a candidate medium, or regimen (for example, employing an environmental stimulus such as mechanical strain alone or in combination with other factors) for its physiological or therapeutic potential as a stimulator or inhibitor of osteogenesis.

[0049] In general, bone cells transduce responses through the process of mechanotransduction. However, the mechanisms of mechanotransduction have been essentially unknown. To address this issue, applicants performed studies in primary human bone cells at three defined stages of differentiation: the osteoblasts, preosteoblasts and osteoprogenitors grown on collagen/vitronectin coated supports. Applicants discovered that a mechanical strain regimen of 70,000 μE at 0.05 Hz stimulated the expression of the αvβ3 integrin in irregular plaque-like areas of integrin-extracellular matrix protein interaction. The plaque-like areas were distinct from another type of cell matrix interaction expressed in osteoblastic cells represented by fine and regularly spaced focal adhesions (FA) that expressed αv and β1 integrins and colocalized with vinculin and other FA markers. The unique sites of cell—matrix interaction containing αvβ3 colocalized with focal adhesion kinase but not with other FA markers. Mechanical strain increased the number and size of the plaques defined by surface expression of αvβ3 Strain also increased secretion of matrix proteins into the plaque-like areas colocalizing with the αvβ3 integrin there. Osteopontin was secreted as a cross-linked macromolecular complex, likely through the action of tissue transglutaminase which was also found in the plaques of the αvβ3 integrin-cell matrix interaction. In addition, mechanical strain increased mineralization of the extracellular matrix by osteoblasts. Since the plaque-like areas of cell—matrix organization exhibit macromolecular assembly and mineralization, they represent subcellular indicators of bone formation, and αvβ3 integrin activation represents one mechanism by which mechanical strain stimulates bone formation.

[0050] Utilizing this discovery, a method of measuring new bone formation using a rapid in vitro assay has been developed. The method involves extracting bone marrow stromal cells from bone marrow cavities and growing them to semi-confluence; exposing the cells to a candidate medium or regimen for osteogenesis or inhibition of osteogenesis; determining the mineralization/development of bone plaque; and, optionally, comparing the mineralization/development of bone plaque generated using the candidate medium or regimen with the mineralization/development of bone plaque using a preselected control such as bone morphogenic factor-7 (BMP-7), also known as osteogenic factor-1 (OP-1).

[0051] Alternatively, the assay can be used as a diagnostic screen for healthy and/or deficient bone formation, making the variable the cells being observed for production of mineralization/bone plaque formation.

[0052] Using the techniques detailed below, a rapid, relatively high througput screen for large-scale evaluation of bone formation has been devised. For example, using techniques known in the art, immunocytochemical/fluorescent-activated cell sorting can be employed, in which the area of plaque growth is examined under the microscope, photographed and digitized for precise measurements. Examples of such techniques for assaying for the presence of the bone plaques isolated by applicants are detailed below.

[0053] The following examples illustrate the invention, but are not to be taken as limiting the various aspects of the invention as illustrated.

EXAMPLES

[0054] Materials

[0055] The anti-αvβ3 integrin and anti-osteopontin antibodies were kindly provided by Dr. David Griggs (Monsanto). The anti-αvβ3 integrin complex monoclonal antibody (P112) behaves almost identically to the LM609 antibody (31), it appears to bind the same epitope and operate through the same mechanism of action (personal communication)1. The anti-αv integrin subunit (L230) and anti-β6 integrin subunit (A10) antibodies were generous gifts from Drs. Scott Blystone and Eric Brown (Washington University) and Drs. Dean Sheppard and John Chen (UCSF), respectively. The anti-tissue transglutaminase antibody was a kind gift from Dr. Daniel Aeschlimann (University of Wisconsin). The anti-β1, -β5 and -α2 integrin antibodies were obtained from Chemicon International Inc. The anti-FAK (pp125FAK) antibody, anti-Goat Cy3-conjugated IgG and fetal calf serum were purchased from Sigma. The Cy3- and Cy2-conjugated AffiniPure Goat Anti-Mouse and Anti-Rabbit IgG were from Jackson Immunochemicals. The protein A-Horseradish Peroxidase conjugate was obtained from Transduction Research. Complete™ protease inhibitor cocktail was purchased from Boehringer Mannheim. Flex I and Flex II cell culture dishes were obtained from Flexcell Corporation. The Omegacell™ 10 ml disposable membrane cells were from Pall Filtron. Immobilon Transfer Membrane for Western blotting applications was purchased from Millipore. Ponceau S solution was obtained from Sigma. All other chemicals utilized were reagent grade. 1 Personal communication with Dr. Jeffrey W. Smith at Program on Cell Adhesion and Extracellular Matrix. The Burnham Institute 10901 North Torrey Pines Road, La Jolla, Calif. 92037.

[0056] Cell Cultures.

[0057] The primary human bone marrow stromal cell cultures were prepared as described previously. Briefly, bone marrow stromal cells were collected from bone marrow cavities, treated with 1 μg/ml DNase and 10 units/ml heparin in DME/F-12 medium, then pelleted, resuspended in α-MEM supplemented with 10% heat-inactivated FCS and separated by Ficoll/Hypaque density gradient. The interphase obtained by centrifugation was collected and resuspended in α-MEM supplemented with 10% heat-inactivated FCS, then grown to confluency. At that time, bone marrow stromal cells were plated on collagen Type I-coated Flex I and Flex II cell culture dishes at a density of 40, 000 cells/well and maintained in culture for 7-14 days, changing medium every 3-4 days. These cell cultures are referred to as osteoprogenitors (3). These osteoprogenitor cell cultures are characterized by low, but detectable, alkaline phosphatase activity of 400 nmole pNPP/min./μg protein; osteopontin expression; absence of osteocalcin expression; and mineralization of the extracellular matrix after 20 days of culture in osteogenic media.

[0058] Bone marrow stromal cells grown to semi-confluence and then exposed to osteogenic medium containing 10 mM of β-glycerophosphate, 50 μg/ml ascorbic acid and 40 ng/ml of osteogenic protein OP-1(32) also referred to as bone morphogenetic protein-7 (BMP-7), for 48 hours were defined as preosteoblasts. These preosteoblast cultures are characterized by alkaline phosphatase activity of 80,000 nmole pNPP/min./μg protein, osteopontin expression, absence of osteocalcin expression, and mineralization after 14 days in osteogenic media. For the purposes of this paper, bone marrow stromal cells exposed to OP-1-containing osteogenic medium for 14 days were defined as “mineralizing preosteoblasts” or osteoblasts. The osteoblast cultures are characterized by alkaline phosphatase activity of 60,000 nmole pNPP/min./μg protein, osteopontin expression, osteocalcin expression and mineralization of the extracellular matrix.

[0059] Near-confluent human preosteoblasts, osteoprogenitors, and osteoblasts were placed on Flexcell Strain Apparatus that vacuum-stretches the flexible silicone bottoms of Flex culture dishes. Applicants applied cyclic strain at three cycles/minute (10 seconds on/10 seconds off; frequency of 0.05 Hz) for 48 hours. One characteristic of the Flexcell Strain Apparatus is the nonuniformity of the strain applied across the surface of the cell culture well. The magnitude of strain delivered in these experiments ranged from approximately 120,000 microstrain (μE) at the edge of the plate (12% maximal displacement) to 0 μE at the center of each cell culture well. Strain applied to cells in analyzed areas (approximately 2-5 mm from the edge of the well) was estimated at 70,000 μE using the strain curve described by Gilbert and colleagues(35). The strain parameters were as follows: strain was applied in the form of a square wave with an amplitude of 70,000 μE with approximately 0.5 second rise time associated with each cycle. Since the observations regarding strained cells were made approximately 2-5 mm from the edge of each well, Applicants presume that the cells portrayed as “strained” in the data contained herein received about 7% stretch. Each strained 6-well Flex I plate was accompanied by a non-strained 6-well Flex II control.

[0060] Immunocytochemistry and Microscopy

[0061] Immunocytochemical analysis essentially was performed as previously described (34). Briefly, primary human progenitors and preosteoblasts were rinsed with phosphate-buffered saline (PBS), and fixed for 30 minutes at room temperature (RT) with 4% (v/v) parafor-maldehyde in PBS. Cells were rinsed twice with PBS, incubated in 50 mM NH4Cl solution in PBS for 15 minutes to quench the fixative, and then permeabilized with 0.2% Triton X-100 in PBS for 15 minutes at RT. Potential sites for nonspecific antibody binding were blocked by a 30 minute incubation with 2% bovine serum albumin (BSA), 0.2% Triton X-100 and 0.04% sodium azide in PBS at RT. Each immunological co-localization study involved two primary antibodies raised in different species against two distinct antigens. Primary antibodies were typically diluted 1:100 (or used at 1-10 μg/ml) in 2% BSA, 0.2% Triton X-100, ), 04% sodium azide in PBS and incubated with the sample for 1 hour at RT. This was followed by four 15 minute washes with 0.02% Triton X-100 in PBS and a 1 hour incubation with an appropriate combination of a secondary indocarbocyanine (Cy3)- and Cy2 cyanine (Cy2)-conjugated goat anti-mouse, goat anti-rabbit or rabbit anti-goat IgG (1:100 dilution in 2% BSA, 0.2% Triton X-100 and 0.04% sodium azide in PBS). The cells were then washed as before with PBS containing 0.2% Triton X-100. The silicone cell culture supports were then excised and mounted on glass microscope slides with Aqua Polymount. The results were analyzed either on Zeiss Fluorescent Microscope (Axioscope) using a 63× oil immersion lens or on a Zeiss laser confocal microscope using a 63× lens. The magnification of the optical system was 630.

[0062] Flow Cytometry

[0063] Following the enzymatic dissociation, cells were resuspended in serum-free medium and pelleted at 4° C. The pellet was then resuspended in serum-free medium and kept on ice. Equal numbers of strained and of non-strained control cells, ranging from 1-2×106 cells and resuspended in 1 ml of serum-free medium were pipetted into polypropylene test tubes. Cells tested for the cell surface integrin expression were incubated with either 10 μg/ml (anti-αvβ3 antibody) or with 1:100 antibody dilution (anti-β1 and anti-β5 integrin subunit antibodies) or with 1:10 dilution of anti-β6 integrin subunit antibody for 1 hour on ice. The parallel negative control tubes were incubated with an irrelevant mouse antibody under identical conditions. Following the incubation with a primary antibody, cells were washed extensively in serum-free medium and incubated with the appropriate fluorescein isothiocyanate (FITC)-conjugated secondary goat anti-mouse or goat anti-rabbit IgG at 20 μg/ml for 1 hour on ice. Cells were then washed twice in serum-free medium and finally resuspended in 1 ml of serum-free medium. The fluorescence intensity was analyzed using a Becton Dickinson FACS SCAN flow cytometer.

[0064] Western Blot Analysis

[0065] Primary human cell cultures were mechanically strained on the Flexcell Strain Apparatus for 48 hours immediately following a replacement of cell culture medium. After the strain, medium was collected and pooled separately from the strained and non-strained wells, and concentrated approximately 6 fold using Omegacell™ 10 ml disposable membrane cells. The protein content in the concentrated medium was determined using a standard Bio-Red method. Medium samples from strained and control cells with equal total protein content were loaded onto a 6% SDS-PAGE gel. Immediately following the electrophoresis, proteins were transferred onto an Immobilon transfer membrane. The efficiency of protein transfer was judged after a 5 minute membrane incubation in Ponceau S solution. Nonspecific antibody binding sites on Immobilon membranes were then blocked by immersing the membrane in a 5% milk and 0.1% Tween 20 in PBS solution overnight at 4° C. Osteopontin was detected by incubating the membranes for 2 hours with anti-osteopontin rabbit polyclonal antibody diluted 1:1000 in 0.1% Tween 20 in PBS at RT, followed by washing the membrane once for 15 minutes and three times for 5 minutes with 0.1% Tween 20 in PBS. Applicants then incubated the membrane with protein A-horseradish peroxidase conjugate diluted 1:2000 in 0.1% Tween 20 in PBS for 1 hour at RT. The final washes included one 15 minute wash and four 5 minute washes in 0.1% Tween 20 in PBS. The antibody binding was detected by exposure to enhanced chemiluminescence using 10 mls of the ECL reagents mix for 60 seconds and then visualized on autoradiography film (Hyperfilm-ECL).

[0066] Mineralization Assay

[0067] Human bone marrow stromal cells were plated in Flex I and Flex II Type I collagen-coated cell culture dishes at 40,000 cells/well and maintained in α-MEM supplemented with 10% FCS in culture until confluent. Cells were then exposed to osteogenic medium containing α-MEM supplemented with 10% FCS, 10 mM of β-glycerophosphate, 50 mg/ml of ascorbic acid and 40 ng/ml of osteogenic protein OP-1 for 20 days, replacing medium every 3-4 days. For the purposes of this paper, bone marrow stromal cells exposed to these conditions were defined as osteoblasts. Following the incubation in osteogenic medium, cells plated in Flex I dishes were mechanically strained on the Flexcell Strain Apparatus for 48 hours, and then both strained cells and their non-strained controls were examined for the formation of mineralized nodules. Applicants applied a slight modification of the alizarin red-S histochemical staining described previously by Stanford and colleagues(35). Briefly, cells were washed 3 times with PBS and then fixed in 70% ice-cold ethanol for 1 hour at 4° C. Following 3 water rinses, cells were stained with 0.5% (w/v) alizarin red-S for 10 minutes at room temperature, and then extensively washed with water and PBS.

[0068] Alternatively, matrix mineralization was examined via calcein staining. Osteoblasts differentiated in osteogenic medium were washed 3 times with PBS and then fixed in 4% paraformaldehyde for 30 minutes at room temperature. Following 3 PBS rinses and a 30 minute incubation with 2% BSAIPBS, cells were stained with 1 mM calcein in 2% BSA/PBS for 1 hour, and then extensively washed with PBS.

[0069] Results

[0070] The present studies were carried out using primary human bone marrow stromal cell cultures at distinct functional stages of osteoblast differentiation: progenitors, preosteoblasts and osteoblasts (mineralizing preosteoblasts). The human bone marrow stromal cells cultured under conditions described in Methods, were defined as the osteoprogenitors. In these heterogenous cell cultures, the cells expressing osteopontin, low alkaline phosphatase levels and responding to mechanical strain with an increase in the αvβ3 integrin level were considered to be the osteoprogenitors. Preosteoblasts were defined as cells expressing high alkaline phosphatase levels, low or absent osteocalcin and absence of matrix mineralization. Osteoblasts were defined as cells expressing high alkaline phosphatase and osteocalcin levels that were mineralizing their extracellular matrix. These cells were produced by extending the culture of the preosteoblasts.

[0071] Mechanical Strain Enhances the Expression of the αvβ3 Integrin on the Cell Surface of Osteoprogenitors and Preosteoblasts

[0072] In order to examine whether mechanical strain has a qualitative and/or quantitative effect on the localization and expression level of integrins, near-confluent human preosteoblasts and osteoprogenitors were strained on Flexcell Strain Apparatus, as described in Methods. The applied mechanical strain was the nonuniform strain Applicants have previously described (15) using this apparatus. Using the immunocytochemical approach to immunostain bone cells with an anti-human αvβ3 monoclonal antibody, Applicants determined that the cells in human osteoblast lineage utilize αvβ3 as a cell surface integrin in our experimental conditions, and that osteoprogenitors (FIGS. 1A and 1B), preosteoblasts (FIGS. 2A and 2B) and osteoblasts (not shown) express the αvβ3 integrin. In the presence of Triton X-100, which leads to cell permeabilization and thus, allows examination of the distribution of intracellular as well as cell surface molecules, the αvβ3 integrin, as detected by an anti-human αvβ3 antibody, was localized along actin fibers and in small, granular structures adjacent to actin fibers (FIGS. 1A, 1B, 2A and 2B). Mechanical strain increased the expression of the αvβ3 integrin in progenitor cells (FIG. 1B), as compared to non-strained controls (FIG. 1A). Specifically, the number of the αvβ3-containing granular structures and the intensity of fluorescence were noticeably increased after 48 hours of cyclic strain. The effect of mechanical strain on the αvβ3 integrin expression was particularly well appreciated when cells were stained with the anti-αvβ3 integrin antibody in the absence of Triton X-100 (FIGS. 1C and 1D). Such a protocol dramatically diminished cell permeabilization defined by fluorescent 70,000 MW dextran staining (data not shown), and it produced improved focus for the examination of the integrin localization concentrated on the plasma membrane at sites of matrix interaction. Applicants demonstrated that the αvβ3 integrin on the cell surface was expressed in large, plaque-like sites of cell—matrix interaction, and it was not expressed in focal adhesions as discussed below. Upon exposure of progenitor and preosteoblast cells to 48 hours of mechanical strain, the number and size of the plaque-like sites increased significantly, and these complexes assumed an organized, directional pattern (FIG. 1D and FIG. 2D). In preosteoblasts, the effect of mechanical strain was in addition to a stimulation by OP-1 of the number of the plaque-like cell/matrix areas of the avb3 expression (compare FIG. 1C with FIG. 2C).

[0073] To better characterize the large, plaque-like sites of matrix organization, Applicants examined colocalization of an anti-αvβ3 integrin antibody with antibodies directed against Focal Adhesion Kinase (FAK), talin and vinculin in non-permeabilized cells. As shown in FIG. 3, FAK (FIG. 3A) and αvβ3 integrin (FIG. 3B) localized to the blotchy plaques on the cell surface of human bone cells. Each plaque containing the αvβ3 integrin also contained FAK (FIG. 3C), but the precise pattern of expression of each protein in a single plaque was different. Microscopic analysis of the cell surface plaques revealed abundant αvβ3 integrin present throughout the surface of the plaques (FIG. 3B), whereas the expression of FAK was confined to the boundaries of each blotchy plaque (see arrows in FIG. 3C). Vinculin and talin were not localized to the plaque-like areas of cell surface αvβ3 integrin expression (data not shown), but they were found in sites of cell—matrix interaction which were defined as osteoblastic focal adhesion complexes (see FIG. 9A).

[0074] Flow Cytometry Confirmed that Mechanical Strain Increased the Cell Surface Expression of the αvβ3 Integrin in Progenitors, Preosteoblasts, and Osteoblasts

[0075] The enhancement of the αvβ3 integrin expression on the cell surface of osteoprogenitors, preosteoblasts and osteoblasts, as detected by immunocytochemical techniques, was confirmed by flow cytometric analysis. FIG. 4 shows two representative plots of flow cytometric analysis of the αvβ3 integrin expression. In the right panel, Applicants demonstrated a shift in the fluorescence from non-strained (B′) to mechanically strained preosteoblasts (D′) shown in the overlay panel, which indicates an increase in the preosteoblast cell surface expression of the αvβ3 with strain. As shown in the left panel of FIG. 4, a similar shift in fluorescence was detected after osteoprogenitors were mechanically strained, hence showing that strain also increased the cell surface expression of the αvβ3 in these cells and thus corroborating our immunocytochemical results.

[0076] As demonstrated by flow cytometry in three to five experiments/cell type using an anti-human αvβ3 integrin antibody to react with single cell suspensions of osteoblast lineage cells, the mean fluorescence of the integrin on the cell surface of osteoprogenitors was increased significantly from 49.3±2.7 in non-strained cells to 78.3±6.7 in mechanically strained cells (Table 1). Similarly, in preosteoblasts, Applicants detected a remarkable mean fluorescence change from 45.3±8.5 in control cells to 90.0±6.6 in strained cells. Further differentiation of preosteoblasts in OP-1-containing osteogenic medium to cells expressing osteocalcin and mineralizing the extracellular matrix (osteoblasts) lead to an increased fluorescence intensity due to increased αvβ3 integrin expression in strained cells (59.8±10.6) as compared to non-strained cells (43.2±7.0).

TABLE 1
             MEAN
CELL TYPE    FLUORESCENCE ± SE
Progenitors
Not strained 49.3 ± 2.7
Strained     78.3 ± 6.7*
Preosteoblasts
Not strained 45.3 ± 8.5
Strained     90.0 ± 6.6*
Mineralized Preosteoblasts
Not strained 43.16 ± 7.0
Strained     59.8 ± 10.6*

[0077] Table 1. Changes in mean fluorescence as a measure of the αvβ3 integrin stimulation by mechanical strain. The flow cytometric analysis was performed in single cell suspensions of cells mechanically strained for 48 hours as well as in their non-strained controls. Osteoprogenitors, preosteoblasts and mineralizing preosteoblasts were incubated with a mouse monoclonal antibody directed against anti-human αvβ3 integrin, followed by an incubation with a secondary FITC-conjugated anti-mouse antibody, as described in Methods. Each data point is a mean of at least 3 experiments. *, p≦0.01 compared to cells not strained.

TABLE 2
αv	Blotchy
integrin	cell-matrix	Focal	Cell	Expressions
type	interactions	adhesions	projections	in cytosal
αvβ1	−	+	−	+
αvβ3	+	−	+	+
αvβ5	−	−	+	+
αvβ6	−	−	−	+

[0078] Table 2. Summary of specific integrin localization sites in bone cells. In addition to their expression in the cytosol, the αvβ1 integrin is apparently present in focal adhesions, the αvβ3 integrin is evident in blotchy cell—matrix interactions, and the αvβ5 integrin is found in cellular projections.

[0079] Mechanical Strain Increases Osteopontin Secretion Into Osteoprogenitor Medium

[0080] Using immunocytochemical techniques, Applicants demonstrated that human bone cells express osteopontin, a ligand for the αvβ3 integrin (FIG. 5A). In addition, Applicants observed a high degree of colocalization (FIG. 5C) between osteopontin (FIG. 5A) and the αvβ3 integrin (FIG. 5B) in non-permeabilized bone cells. Osteopontin co-localized with the αvβ3 integrin in the large plaque-like areas of cell matrix interaction. This finding coupled with the elevated level of αvβ3 integrin found in mechanically loaded cells suggests that in response to mechanical strain, osteopontin may serve as an integrin ligand.

[0081] Applicants also addressed the issue whether osteopontin levels, both intracellular and secreted into cell culture media, were affected by mechanical strain. By performing Western blot analyses using anti-human osteopontin antibodies, Applicants detected several forms of osteopontin that were secreted into bone cell culture medium (FIG. 6A). The predominant osteopontin band had a molecular size of approximately 168 kDa and may represent either a cross-linked osteopontin-matrix protein macro-molecule or an osteopontin homomultimer (FIG. 6A). In cell lysates, the osteopontin bands included predominant 75 kDa and a 78 kDa bands (FIG. 6B). The 75 kDa band has been frequently seen in several cell types as a mature OPN(36,37). FIG. 6A is representative of five experiments demonstrating stimulation of osteopontin secretion into the medium of mechanically strained cells in the osteoblast differentiation program. The strain over control stimulation index determined by scanning densitometry of the 168 kDa bands was more than 2.0 in osteoprogenitors and approximately 1.0 for preosteoblasts. The level of osteopontin secretion in non-strained osteoprogenitors in the presented blots seems lower than that in non-strained preosteoblasts exclusively due to technical reasons (as not to overexpose the band representing strained progenitors). Thus, the effect of strain was apparent only in the osteoprogenitors. In contrast to the findings in cell culture mediums, Western blotting of cell lysates revealed predominant osteopontin bands of 75 and 78 kDa and no differences in osteopontin levels between strained and non-strained bone cells, regardless of the cell differentiation stage (FIG. 6B). These data suggest that the effect of strain includes stimulation of secretion of the 168 kDa molecule such that it does not accumulate in the cell.

[0082] αvβ3 Integrin Colocalizes with Tissue Transglutaminase

[0083] To further analyze the basis of the large osteopontin molecular sizes, Applicants analyzed whether mechanical strain affected osteoblastic expression of tissue transglutaminase, which cross-links proteins into homo and heteromultimers through the formation of γ-glutamyl-Σ-lysine cross-links between substrate proteins. Bone osteopontin and osteonectin are tissue transglutaminase substrates(38-40). Based on the immunocytochemical analysis, tissue transglutaminase in osteoprogenitors is expressed throughout the cytoplasm in punctate subcellular structures, as shown in detergent-permeabilized cells (FIG. 7A). Interestingly, in cells fixed in the absence of Triton X-100, tissue transglutaminase was also found in large, plaque-like sites (FIG. 7B and FIG. 7C) that closely resemble the cell surface αvβ3 integrin plaques (FIG. 7D). Moreover, the immunocytochemical approach demonstrated a colocalization of tissue transglutaminase and αvβ3 integrin in well defined plaques in osteoprogenitors (FIG. 7E).

[0084] The Effect of Mechanical Strain on Formation of Mineralized Nodules

[0085] Since osteoblasts are known to produce, deposit and mineralize the bone matrix, Applicants examined the influence of mechanical strain on the formation of mineralized nodules. Alizarin red S histochemical staining of osteoblasts derived from bone marrow stromal cells demonstrated that although mineralization was detected in both strained (FIGS. 8B, 8D, 8F) and non-strained cultures (FIGS. 8A, 8C, 8E), the intensity of nodules produced by mechanically strained cells was dramatically enhanced at least twofold, as compared with controls. These osteoblastic cells were maintained in osteogenic medium containing ascorbate, β-glycerophosphate and OP-1 for 18 days prior to 48 hours of mechanical strain (days 18-20). Furthermore, on the basis of colocalization of another mineralization marker, calcein with the anti-αvβ3 integrin antibody (FIG. 8G) in osteoblasts derived from bone marrow stromal cells, Applicants observed that the αvβ3 integrin plaques colocalized with calcein-labeled mineralization nodules. Moreover, in mineralizing cultures the αvβ3 plaques assumed a nodular shape (FIG. 8G) indicated by their expansion in the Z axis (vertically), unlike their level, flatter shape observed in non-mineralizing cells (compare with FIG. 1D and FIG. 2D).

[0086] Mechanical Strain Does not Affect Osteoblastic Focal Adhesions

[0087] Application of a monoclonal anti-human αv antibody in our immunocolocalization protocol followed by confocal analysis demonstrated that the αv subunit epitope detected by monoclonal antibody L230 was present in the plaque-like areas of cell—matrix interaction discussed above, but it was predominantly found in fine, regularly and densely spaced structures morphologically similar to focal adhesions described in other cell types (41-43). These focal adhesions were observed in each of the osteoblast differentiation stages examined (FIG. 9A). Besides their characteristic shape, the structures containing the αv subunit were defined as focal adhesions on the basis of their co-localization with known focal adhesion markers vinculin and talin. Interestingly, an identical pattern of αv localization was obtained in both permeabilized and non-permeabilized cells, thus suggesting that this αv subunit epitope is primarily exposed during cell surface/substratum interaction. Moreover, our results indicate a high level of focal adhesion expression in resting osteoblastic cell cultures obscuring the effects of mechanical strain on the level of αv integrin expression in the focal adhesion of human bone cells. Studies showing a cell surface colocalization of the αv subunit (L230 antibody) with osteopontin in plaque-like structures provided proof that this αv subunit was also found in the αvβ3 integrin plaques, since osteopontin colocalized closely with the αvβ3 in those plaques (FIG. 5C).

[0088] To identity the β subunit of the αvβ heterodimer detected in the focal adhesions of our bone cell cultures (FIG. 9A), Applicants immunologically colocalized the anti-αv antibody with each of the potential αvβ candidate heterodimers of the osteoblast. The anti-β5 and the anti-β6 integrin subunit antibodies revealed a cytoplasmic distribution of these integrins distinct from the αv subunit-containing fine focal adhesions, as shown in FIGS. 9B and 9D, respectively. Interestingly, the colocalization of the β5 integrin subunit and the αvβ3 integrin revealed that our human bone cell cultures were characterized by long cellular projections containing both the β5 subunit and the αvβ3 integrin (FIG. 9B and FIG. 9C). These cellular projections abut other cell bodies at their end, thus, suggesting the involvement of the αvβ5 integrin in cell-cell communication (see arrows in FIG. 9C). As demonstrated by flow cytometry, mechanical strain increases the expression of the β5 integrin subunit by approximately 25% on the surface of preosteoblasts, but not on the osteoprogenitor plasma membrane.

[0089] In contrast to the β5 subunit, the β6 integrin subunit expression level, as measured by flow cytometry, was not affected at all by mechanical strain. The immunolocalization studies demonstrated that the primarily cytosolic staining pattern of the β6 integrin subunit was different from that of fine focal adhesions containing the αv subunit (FIG. 9D). This is the first demonstration of the αvβ6 integrin in osteoblasts.

[0090] Finally, Applicants compared the distribution of the β1 integrin subunit with the staining pattern of fine focal adhesions (FIG. 10). As shown in FIG. 10B, the β1 integrin subunit was expressed in a rather diffuse staining pattern, as expected from the large number of osteoblastic b1 integrins. However, the immunochemical analysis indicated that a significant portion of the β1 integrin subunit colocalized closely with the αv integrin subunit, as shown in FIG. 10C. Careful microscopic analysis of a large number of samples demonstrated that the αv and β1 integrin subunits are present in the focal adhesions of osteoprogenitors, preosteoblasts and osteoblasts.

[0091] Discussion

[0092] Applicants have demonstrated strain-induced osteoblastic mineralization in vitro. It provides several important new clues as to the molecular basis of the cellular events that lead to mechanical strain induced bone formation. The general conclusion from this investigation is that in response to mechanical strain, the αvβ3 integrin of the osteoblast lineage is upregulated and that it may participate in the organization of the osteopontin-containing extracellular matrix in unique plaque-like areas of cell -matrix interaction.

[0093] Applicants have presented evidence that primary human cells of the osteoblast lineage express at least two types of cell—matrix interactions. One type represents fine and densely spaced cell surface structures that colocalize with vinculin and represent osteoblastic focal adhesions. A second type of relatively large, plaque-like areas of integrin/extracellular matrix protein interaction colocalizes with Focal Adhesion Kinase (FAK) but not with vinculin. Applicants have demonstrated that the αvβ3 integrin specifically colocalizes to the second type of cell—matrix interaction sites and that its expression was further stimulated by mechanical strain. Upon mechanical strain, both types of cell—matrix interactions assumed an organized, directional pattern (FIG. 1D and FIG. 9A). In agreement with earlier reports by Buckley and colleagues (49), this new orientation was perpendicular to the vectors of strain. FAK has been previously shown to colocalize with several components of cellular focal adhesions, such as tensin, vinculin and talin(45). Furthermore, these plaque-like areas are specific to the osteoblast lineage in that Applicants have failed to demonstrate them in other cell types (i.e. human melanoma cells), and cells differentiating from hematopoetic progenitors (i.e. macrophages and osteoclasts).

[0094] Applicants also demonstrated that the αvβ5 integrin once considered a dominant osteoblast adhesion molecule is largely cytoplasmic in primary human bone cells, and that it is also expressed along extended cellular projections that contact distant osteoblast cell bodies. Applicants believe that these cellular projections participate in cell-cell communication.

[0095] Applicants have also identified at least one matrix protein, osteopontin as a ligand for the αvβ3 integrin contained in the plaque-like cell—matrix organization sites. Osteopontin in our studies in the osteoblast may act as an autocrine matrix protein, secreted in the focal adhesion in response to regulatory stimuli and important in the proliferation and differentiation of cells during the osteoblast program. Based on previous reports, osteopontin is present in human milk and melanoma cells, as well as in rat osteosarcoma cells as an approximately 75 kDa protein, and in osteoclasts as a protein of 55-65 kDa (36,37,47-48). Applicants have demonstrated the presence of several osteopontin bands of different molecular sizes in the cell culture medium of osteoprogenitors and osteoblasts. The osteopontin band of highest intensity was detected at approximately 168 kDa, thus representing a cross-linked macromolecular complex of either osteopontin and other matrix proteins or of osteopontin oligomers.

[0096] Applicants have discovered that in human cells of the osteoblast lineage, osteopontin colocalizes with the αvβ3 integrin in plaque-like sites of cell—matrix interaction. Moreover, the αvβ3 integrin was found in the mineralization nodules in osteoblasts derived from bone marrow stromal cells, as determined by anti-αvβ3 antibody/calcein staining. Applicants also demonstrated that mechanical strain leads to an increased formation of mineralized nodules (FIG. 9). Osteopontin and other extracellular proteins (including fibronectin and bone sialoprotein) were previously shown to initiate and/or direct the spatial deposition of mineral in the extracellular matrix (49-51). Therefore, Applicants' results indicate that mechanical strain-enhanced interaction between osteopontin and (αvβ3 integrin is indicated to lead to enhanced matrix mineralization.

[0097] Osteopontin, a known substrate of tissue transglutaminase, may be covalently cross-linked by this enzyme (52). Tissue transglutaminases are a widely distributed class of enzymes found intra- as well as extracellularly (53). Tissue transglutaminase participates in the assembly and organization of extracellular matrix via its activity as an externalized enzyme, and thus has an impact on cell spreading and adhesion (54). Tissue transglutaminase catalyses a calcium-dependent formation of a specific isopeptide bond, therefore promoting intra- and/or intermolecular cross-linking in proteins (55). Tissue transglutaminase is a highly selective enzyme, with only a few native proteins identified as its substrates (53,56). The high specificity of this enzyme renders the tissue transglutaminase—catalyzed cross-linking of proteins, including osteopontin, physiologically significant.

[0098] Our study indicates that the αvβ3 integrin colocalizes with tissue transglutaminase in bone cells (FIG. 7), and suggests that osteopontin colocalizes with tissue transglutaminase in the large, blotchy sites of matrix organization, thus suggesting that other matrix proteins may be cross-linked with osteopontin in these complexes.

[0099] Finally, Burger and Veldhujzen have reported that production of extracellular matrix decreases at high strain levels, while cellular proliferation increases (57). Applicants observed the enhancement of mineralization of extracellular matrix in osteoblasts grown in osteogenic medium for 20 days, resulting from an average mechanical strain level of approximately 70,000 μE, as determined on the basis of the strain curve described by Gilbert et al (33). Thus, Applicants suggest that the majority of the cells in our studies were exposed to a strain level at which no inhibition of bone matrix production occurred. Since Applicants were limited to a non-uniform strain distribution by our apparatus, Applicants were unable to determine the applied strain that lead to increased mineralization. Our data indicates that the sequential process of matrix production, deposition and mineralization was enhanced by the level of mechanical strain employed.

[0100] In view of the above, it will be seen that the several objects of the invention are achieved.

[0101] As various changes could be made in the above compositions and processes without departing from the scope of the invention, it is intended that all matter contained in the above description be interpreted as illustrative and not in a limiting sense.

[0102] References

[0103] 1. Wozney, J. J., Rosen, V., Celeste, A. J., Mitsock, L. M., Whiters, M. J., Kriz, R. W., Hewick, R. M., and Wang, E. A. (1988) Science 242, 1528-1534

[0104] 2. Cheng, S.-L., Yang, J. W., Rifas, L., Zhang, S.-F., and Avioli, L. V. (1994) Endocrinology 134, 277-286

[0105] 3. Lecanda, F., Avioli, L. V., and Cheng, S. L. (1997) J. Cell. Biochem. (In press)

[0106] 4. Hanada, K., Dennis, J. E., and Caplan, A. I. (1997) J. Bone Miner. Res. 12, 1606-1614

[0107] 5. Owen, M. (1988) J. Cell Sci. Suppl. 10, 63-76

[0108] 6. Caplan, A. I. (1991) J. Orthop. Res. 9, 641-650

[0109] 7. Yamaguchi, A., Katagiri, T., Ikeda, T., Wozney, J. M., Rosen, V., Wang, E. A., Kahn, A. J., Suda, T., Yoshiki, S. (1991) J. Cell Biol. 113, 681-687

[0110] 8. Katagiri, T., Yamaguchi, A., Komaki, M., Abe, E., Takahashi, N., Ikeda, T., Rosen, V., Wozney, J. M., Fujisawa-Sehara, A., and Suda, T. (1994) J. Cell. Biol. 127, 1755-1766

[0111] 9. Katagiri, T., Yamaguchi, A., Ikeda, T., Yoshiki, S., Wozeny, J. M., Rosen, V., Wang, E. A., Tanaka, H., Omura, S., and Suda, T. (1990) Biochem. Biophys. Res. Comm. 172, 295-299

[0112] 10. Aubin, J. E. and Kahn, A. (1996) in Primer on the metabolic bone diseases and disorders of mineral metabolism (Favus, M. J., ed) pp. 35-38, Lippincott-Raven, Philadelphia, New York

[0113] 11. Pead, M. J., Skerry, T. M., and Lanyon, L. E. (1988) J. Bone Miner. Res. 3, 647-656

[0114] 12. Brighton, C. T., Fisher, J. R. S., Levine, S. E., Corsetti, J. R., Reilly, T., Landsman, A.S., Williams, J. L., and Thibault, L. E. (1996) J. Bone and Joint Surg. 78, 1337-1347

[0115] 13. Einhorn, T. A. (1998) J. Bone Miner. Res. 13, 10-12

[0116] 14. Ingber, D. E. (1997) Ann. Rev. Physiol. 59, 575-599

[0117] 15. Harter, L. V., Hruska, K. A., and Duncan, R. L. (1995) Endocrinology 136, 528-535

[0118] 16. Toma, C. D., Ashkar, S., Gray, M. L., Schaffer, J. L., and Gerstenfeld, L. C. (1997) J. Bone Miner. Res. 12, 1626-1636

[0119] 17. Salter, D. M., Robb, J. E., and Wright, M. O. (1997) J. Bone Miner. REs. 12, 1133-1141

[0120] 18. LaFlamme, S. E. and Auer, K. L. (1996) Cancer Biology 7, 111-118

[0121] 19. Schwartz, M. A., Schaller, M. D., and Ginsberg, M. H. (1995) Annu. Rev. Cell Biol. 11, 549-600

[0122] 20. Sanchez-Mateos, P., Cabanas, C., and Sanchez-Madrit, F. (1996) Seminars in Cancer Biology 7, 99-109

[0123] 21. Sonnenberg, A. (1993) Current topics in microbiology and immunology 184, 7-35

[0124] 22. Craig, S. W. and Johnson, R. P. (1996) Curr. Opin. Cell. Biol. 8, 74-85

[0125] 23. Denhardt, D. T. and Guo, X. (1993) FASEB J. 7, 1475-1482

[0126] 24. Grzesik, W. J. and Robey, P. G. (1994) J. Bone Miner. Res. 9, 487-496

[0127] 25. Miyauchi, A., Alvarez, J., Greenfield, R. M., Teti, A., Grano, M., Colucci, S., Zambonin-Zallone, A., Ross, F. P., Teitelbaum, S. L., Cheresh, D., and Hruska, K. A. (1991) J. Biol. Chem. 266, 20369-20374

[0128] 26. Oldberg, a. Franzen, A., Heinegard, D., Pierschbacher, M., and Ruoslahti, E., (1988) J. Biol. Chem. 263, 19433-19436

[0129] 27. Sauk, J. J., Van Kampen, C. L., Foster, N. K., and Somerman, M. J. (1990) Experimental Cell Research 188, 105-110

[0130] 28. Sommerman, J. J., Prince, C. P., Sauk, J. J., Foster, R. A., and Moehring, J. M. (1989) Matrix 9, 49-54

[0131] 29. Sommerman, J. J., Prince, C. W., Sauk, J. J., Foster, R. A., and Butler, W. T. (1987) J. Bone Miner. Res. 2, 259-265

[0132] 30. Ross, F. P., Chappel, J., Alvarez, J. I., Sander, D., Butler, W. T., Farach-Carson, M. C., Mintz, K. A., Robey, P. G., Teitelbaum, S. L., and Cheresh, D. A. (1993) J. Biol Chem. 268, 9901-9907

[0133] 31. Cheresh, D. A. and Spiro, R. C. (1987) J. Biol. Chem. 262, 17703-17711

[0134] 32. Smpath, T. K., Maliakal, J. C., Hauschka, P. V., Jones, W. K., Sasak, H., Tucker, R. F., White, K. H., Coughlin, J. E., Tucker, M. M., Pag, R. H. L., Corbett, C., Ozkaynak, E., Opperrnann, H., and Rueger, D. C. (1992) J. Biol. Chem. 267, 20352-20362

[0135] 33. Gilbert, J. A., Weinhold, P. S., Banes, A. J., Link, G. W., and Jones, G. L. (1994) J. Biomechnaics 27, 1169-1177

[0136] 34. Wozniak, M. and Limbird, L. E. (1996) J. Biol. Chem. 271, 5017-5024

[0137] 35. Stanford, C. M., Jacobson, P. A., Eanes, E. D., Lembke, L. A., and Midura, R. J. (1995) J. Biol. Chem. 270, 9420-9428

[0138] 36. Kubota, T. M., Yamauchi, M., Onozaki, S. S., Suzuki, Y., and Sodek, J. (1993) Archives of Oral Biology 38, 23-30

[0139] 37. Senger, D. R., Peruzzi, C. A., Papadopoulos, A., and Tenen, D. G. (1989) Biochimica et Biophysica Acta 996, 43-48

[0140] 38. Prince, C. W., Dickie, D., and Krurndieck, C. L. (1991) Biochemical and Biophysical Research Communications 177, 1205-1210

[0141] 39. Aeschlimann, D., Wetterwald, A., Fleisch, H., and Paulsson, M. (1993) J. Cell Biol. 120, 1461-1470

[0142] 40. Sorensen, E. S., Rasmussen, L. K., Moller, L., Jensen, P. H., Hojrup, P., and Petersen, T. E. (1994) Biochemistry Journal 304, 13-16

[0143] 41. Hotchin, N. A. and Hall, A. (1995) J. Cell. Biol. 131, 1857-1865

[0144] 42. Larava, H., Pletonen, J., Akiyama, S. K., Yamada, S. S., Gralnick, H. R., Uitto, J., and Yamada, K. M. (1990) J. Cell. Biol. 110, 803-815

[0145] 43. Ridley, A. J. and Hall, A. (1992) Cell 70, 389-399

[0146] 44. Buckley, M. J., Banes, A. J., and Jordan, R. D. (1990) J. Oral Maxillofac Surg 48, 276-282

[0147] 45. Zachary, I. And Rozengurt, E. (1992) Cell 71, 891-894

[0148] 46. Hruska, K. A., Rolnick, F., Huskey, M., Alvarez, U., and Cheresh, D. (1995) Endocrinology 136, 2984-2992

[0149] 47. Chellaiah, M. and Hruska, K. A. (1996) Mol. Biol. Cell 7, 743-753

[0150] 48. Chellaiah, M., Fitzgerald, C., Filardo, E. J., Cheresh, D. A., and Hruska, K. A. (1996) Endocrinology 137, 2432-2440

[0151] 49. Chen, Y., Bal, B. S., and Gorski, J. P. (1992) J. Biol. Chem. 267, 24871-24878

[0152] 50. Gerstenfeld, L. C., Gotoh, Y., McKee, M. D., Nanci, A., Landis, W. J., and Glimcher, M. J. (1990) Anat Rec 228, 93-103

[0153] 51. Hunter, G. K. and Goldberg, H. A. (1993) PNAS 90, 8562-8565

[0154] 52. Beninati, S., Senger, D. R., Cordellα-Miele, E., Mukherjee, A. B., Chcckalaparampil, I., Shanmugarn, V., Singh, K., and Mukherjee, B. B. (1994) Journal of Biochemistry 115, 675-682

[0155] 53. Aeshlimann, D. and Paulsson, M. (1994) Thrombosis and hemostasis 71, 402-415

[0156] 54. Jones, R. A., Nicholas, B., Mian, S., Davies, P. J. A., and Griffin, M. (1997) Journal of Cell Science 110, 2461-2472

[0157] 55. Folk, J. E. (1980) Ann. Rev. Biochem. 49, 517-531

[0158] 56. Aeschlimann, D., Kaupp, O., and Paulsson, M. (1995) J. Cell Biol. 129, 881-892

[0159] 57. Burger, E. H. and Veldhuijzen, J. P. (1993) in Bone-Bone Growth (Hall, K., ed) pp.37-56, CRC Press, Melbourne

Claims

1. An isolated and essentially purified cell matrix plaque comprising αvβ3 integrin.

2. The plaque of claim 1 further comprising focal adhesion kinase, osteopontin and tissue transglutaminase.

3. A method of measuring new bone formation comprising extracting bone marrow stromal cells from bone marrow cavities and growing said cells to semi-confluence, exposing said cells to a candidate medium or regimen for osteogenesis or inhibition of osteogenesis, and determining the mineralization or the development of bone plaque.

4. The method of claim 3 further comprising the steps of comparing the mineralization or development of bone plaque generated using said candidate medium or regimen with the mineralization or development of bone plaque using a preselected control.

5. The method of claim 4 wherein said preselected control is a bone morphogenic factor.

6. The method of claim 5 wherein said bone morphogenic factor is BMP-7.

7. The method of claim 3 wherein said determining the mineralization or development of bone plaque comprises examining for presence of bone plaque sites containing αvβ3 integrin.

8. The method of claim 7 wherein said bone plaque sites further contain focal adhesion kinase.

9. The method of claim 7 wherein said examination further comprises measurement of the number and size of said plaque sites.

10. The method of claim 7 further comprising microscopically photographing said plaque sites and digitizing components thereof for precision measurements.

11. The method of claim 7 wherein said examination further comprises utilization of flow cytometry.

12. A method for screening the bone formation capacity of sample cells comprising exposing said cells to a medium or regimen known to generate osteogenesis, and determining the mineralization or the development of bone plaque.

13. The method of claim 12 wherein the medium or regimen comprises a bone morphogenic factor.

14. The method of claim 13 wherein the factor comprises BMP-7.

15. The method of claim 12 wherein said determining the mineralization or development of bone plaque comprises examining for presence of bone plaque sites containing αvβ3 integrin.

16. The method of claim 15 wherein said bone plaque sites further contain focal adhesion kinase.

17. The method of claim 15 wherein said examination further comprises measurement of the number and size of said plaque sites.

18. The method of claim 15 further comprising microscopically photographing said plaque sites and digitizing components thereof for precision measurements.

19. The method of claim 15 wherein said examination further comprises utilization of flow cytometry.