WO2007104724A2 - Identification of osteoblast cells differentiation and bone tumor markers and uses thereof - Google Patents

Abstract

The present invention discloses a method for the diagnosis of a metabolic bone disease, and/or for assessing osteoporotic fracture risk and/or for the monitoring of therapeutic efficacy of a metabolic bone disease drug in a subject and a method for the diagnosis of bone tumor including the identification and quantification of bone cell or bone tumor specific proteins.

Description

Identification of osteoblast cells differentiation and bone tumor markers and uses thereof

Background of invention Bone contains two distinct cellular types, the bone-forming cells, namely osteoblasts, and the bone-resorbing cells, namely osteoclasts, whose action is balanced in a physiological process known as bone remodelling, which maintains skeletal homeostasis. Their function is intimately linked, specially because osteoblasts synthesize and secrete bioactive molecules acting on osteoclasts (1, 2). To obtain improved remedies for various bone metabolic diseases as osteoporosis, an endemic remodelling disease diffused in the western society, it has been widely emphasized the urgent need for a better comprehension of regulative processes controlling bone remodelling, therein osteoblasts and osteoclasts functionality. Research on bone cells requires selection of suitable in vitro model systems. As an example, bisphosphonates (BPs), the most important class of antiresorptive drugs currently used for treatment of metabolic bone diseases, exert its effects both on osteoclasts and osteoblasts (3, 4). To better investigate BPs activity on bone metabolism, cellular models in which osteodifferentiation and osteogenesis selectively occurs are necessary (5). Many culture systems have been developed using osteoblast-like cells derived from normal bone or osteosarcomas of animal or human origin. However, the use of animal or transformed cells raises the problem of extrapolation to human beings. Human osteosarcoma cells are often used for investigating osteoblast functions, although these cells have abnormal characteristics and particular responses to osteotropic agents (6). Moreover, because of the transformed nature of osteosarcoma cells, a deregulation of the tightly coupled relationships between proliferation and progressive expression of genes associated with bone cell differentiation has been suggested (7). Long-term maintained rat calvarial osteoblasts show maturation sequences and developmental stages comparable to those observed in neonatal bone (7), and represent an important model for in vitro studies of bone metabolism. Nevertheless, a direct and complete applicability of this model to human tissues has not been established yet (8). All these limitations suggest that well- differentiated cells taken from human adult bone and placed in culture are the most suitable model for specialized advanced studies. Primary cell cultures from human specimen should be preferable than cells passaged several times (9, 10), also because they allow taking into account the age and biological status of the patient from which they derive. Moreover, susceptibility to drugs of primary cell cultures may also be similar to what occurrs in vivo. In the last years, many important insights on osteoblasts function have come from molecular biology and biochemical studies on genetically defined human and mouse models (11, 12, 13, 14, 15). On the contrary, proteomic aspects of bone cells biology still remain unknown, although a preliminary investigation on human osteosarcoma cell line, SaOS-2, chaperone subproteome has been recently published (16).

Summary of invention The authors report the proteome analysis of primary-cultured human osteoblasts during cell differentiation. The authors analysed osteoblast cells both in proliferative and maturated status and identified proteins peculiar for this latter condition. Moreover, to evaluate significant phenotypic differences, a comparative proteome study was also carried out between normal human osteoblasts and SaOS-2 cells, which allowed characterizing proteins typically expressed in the latter ones to be considered as potential tumour markers. These results may be important for defining and standardizing osteoblast cultures type and differentiation stage with the ultimate goal to design experimental assays and evaluate at the molecular level the effects of different pharmacological treatments on osteoblast proliferation and differentiation. Indeed, in the present study the authors use serial primary cultures of human bone-derived osteoblasts as an in vitro model to study osteoblast differentiation. A proteomic study on human primary cultured normal specimen-derived osteoblast cells is carried out, defining a 2-DE reference map. Differently from previous studies on immortalized and/or animal cells, the present data on human primary cells can be directly extrapolated to human clinical cases.

The authors are able to profile protein expression during cell maturation; accordingly protein markers of osteoblast differentiation are identified. These molecular markers are important to characterize a human osteoblast cell model at a defined stage of maturation. Such markers can be utilized to standardize experimental data when assessing the biological effects of anabolic agents or pro-apoptotic drugs. Some of the differentiation markers of the present invention confirm results obtained at the gene level, while other have never been reported. Novel differentiation markers are relevant for bone research considering that the present study proved how classical histological bone differentiation protein markers failed to be sensitive enough at the proteome level. Molecular markers of bone turnover have gained increasing relevance in the evaluation of patients with metabolic bone diseases. Their clinical applications include assessment of risk for osteoporotic fracture, complementation of bone density measurements, diagnosis of certain metabolic osteopathies, therapeutic decision making and monitoring of therapeutic efficacy and patient compliance. In scientific setting, conventional and new markers of bone turnover may help elucidating formerly unknown mechanisms and pathways. Recently, the same osteoblast cell model described here has been successfully used to test bisphophonates activity (23) and the effects of osteogenic peptide on osteoblast (59). Comparative proteomic experiments, where osteoblasts and osteosarcoma SaOS-2 cells were evaluated, confirmed and validated the identification of differentiation markers in normal osteoblasts. The present study provides the characterization of the Human osteosarcoma proteome and highlights significant differences with respect to the proteome of normal osteoblasts. These findings suggest that osteosarcoma and immortalized cell lines should be use with caution for the study of age-related bone changes and/or of anti- osteoporotic drugs in humans.

Furthermore, the present invention characterizes specific proteins selectively over/under- expressed in SaOS-2 cells, thus identifies potential protein markers of tumor progression. Improving overall survival and reducing morbidity are major goals of childhood cancer research (60). Therefore new therapeutic strategies which optimize existing agents are needed, along with the development of new agents. With this view, it is extremely important to understand the role of individual genes in the regulation of cell growth and differentiation in order to design specific inhibitors. Gene therapy is another potential therapeutic strategy, however a prerequisite for a successful gene therapy approach to cancer is to identify the genes that confer cell differentiation.

Osteosarcoma is a malignant bone tumor evolved from cells characterized by an extended replication and altered proliferation. Consequently, osteosarcoma cells become different from normal osteoblasts at various cellular levels, like cell proliferation and matrix production. Thus, the identification of the proteins involved in the changes from normal osteoblasts function to abnormal function provides a better approach for the diagnosis of bone tumor. In particular, three SaOS-2 markers reported in the present invention, namely alkaline phosphatase, creatine kinase B and ORP 150, have features potentially related to the osteosarcoma metastatic potential and also with the so far not well characterized alterations of the osteosarcoma matrix.

It is therefore an object of the invention a method for the diagnosis of a metabolic bone disease, and/or for assessing osteoporotic fracture risk and/or for the monitoring of therapeutic efficacy of a metabolic bone disease drug in a subject, including the following steps: a) culturing subject's bone-derived cells; b) extracting and separating proteins from such cultured cells; c) identifying and quantifying at least one bone cell specific protein; d) comparing the amount of at least one bone cell specific protein with that of a control subject, or of a panel of control subjects.

Preferably, the bone cell specific protein is comprised in the group of: actin, tubulin, heat shock protein 27, cathepsin D, ubiquitin C-terminal hydrolase Ll, phosphoglycerate mutase 1, pyruvate kinase Ml, α-enolase, ATP synthase, glutathione S-transferase P, superoxide dismutase, glutamate dehydrogenase 1 or enoyl-CoA hydratase-like protein.

More preferably, the actin is beta or gamma actin and the tubulin is beta tubulin. Still preferably, the metabolic bone disease is a metabolic osteopathy. Preferably, the control subject is an healthy subject or a subject in course of a therapeutic treatment. In the present invention, the method can consist in the identification and quantification of a panel of 2, 3, 4 or more, or of all of the bone cell specific proteins.

It is a further object of the invention a method for the diagnosis of bone tumor in a subject comprising the steps of: a) culturing subject's bone-derived cells; b) extracting and separating proteins from cultured cells; c) identifying and quantifying at least one bone tumor specific protein; d) comparing the amount of at least one bone tumor specific protein with that of a control subject, or of a panel of control subjects.

Preferably, the bone tumor specific protein is comprised in the group of: pyruvate kinase Ml, L-lactate dehydrogenase B chain, triosephosphate isomerase 1, creatine kinase B chain, heat shock protein 90, 150KDa oxygen-regulated protein (ORP 150), Retinoblastoma-binding protein 4, alkaline phosphatase, galectin-1, annexin, tropomyosin 3 or UMP-CMP kinase. More preferably, the bone tumor specific protein is comprised in the group of: osteocalcin, osteonectin, Cathepsin D, heat shock protein 27, superoxide dismutase, glutamate dehydrogenase 1 or enoyl-CoA hydratase-like protein. Still preferably, the annexin is annexin I or II and the bone tumor is an osteosarcoma. In the present invention, the method can consist in the identification and quantification of a panel of 2, 3, 4 or more, or of all of the bone tumor specific proteins. The invention will be now illustrated by means of non limiting examples in reference of the following figures:

Figure 1: Scheme of human osteoblasts production

Figure 2: Cell proliferation assay. Graphs show the time course of human osteoblasts' growth in DMEM supplemented with 10% (v/v) FCS or 2% (w/v) UG, analyzed at days 1, 3, 6, 10, 16 and 20, respectively. Data represent the average of the mean values of four independent experiments. Error bars, SDs from the mean values.

Figure 3: Assessment of mature osteoblast phenotype by TEM. Sections of human osteoblasts cultured in vitro for 15 days in DMEM supplemented with 10% (v/v) FCS (panel A) or 2% (w/v) UG (panel B). In comparison with osteoblasts cultured with FCS, osteoblasts grown in UG-supplemented medium display a cytoplasm more abundant in endoplasmic rough reticulum (RER, indicated by the arrows) and Golgi complex. Nucleus (N); Original magnification 25.00Ox.

Figure 4: Alkaline phosphatase activity. Alkaline phosphatase activity in human osteoblasts cultured in DMEM supplemented with 10% (v/v) FCS or 2% (w/v) UG. Enzymatic activity was determined in 12-well plates and normalized to the relative number of viable cells. Graphs show the time course analyzed at days 4, 7, 10, 15 and 20, respectively. Data represent the average of the mean values obtained for four different experiments. Error bars, SDs from the mean values. Figure 5: Histochemical staining for alkaline phosphatase activity. The images show the difference between human osteoblasts cultured in DMEM supplemented with 10% (v/v) FCS (panel A) or 2% (w/v) UG (panel B). Original magnification 2Ox. Figure 6: Bone nodule formation and mineralization. Detection of mineralized nodules by alizarin Red S staining in human osteoblast cultures grown in DMEM supplemented with 10% (v/v) FCS (panel A) or 2% (w/v) UG (panel B). Original magnification 10x. Figure 7: Quantification of three-dimensional nodular structures in El and ElPl cell populations. FCS-cultured and UG-cultured ElPl cells data were reported at days 4, 12 and 20, respectively. Data represent the average of the mean values obtained for four different experiments. Error bars, SDs from the mean values. Figure 8: Proteome maps of human osteoblastic cells. Human pre-osteoblasts (El) (panel A) and differentiated osteoblasts (ElPl) (panel B). Proteins differently expressed are indicated. Identified proteins are reported in Table 1. Human osteosarcoma SaOS-2 cells (panel C). Proteins over-expressed (circles) or underexpressed (squares) with respect to normal human osteoblasts (ElPl) are indicated. Identified proteins are reported in Tables 2 and 3, respectively.

Materials and Methods

Human bone-derived cell cultures Bone samples were obtained from one single male patient (35 years old) who underwent total hip replacement surgery. Trabecular bone fragments (10) were extensively washed in phosphate buffered saline (PBS) to remove blood and bone marrow, and then explanted into culture containing Dulbecco's modified Eagle's Medium (DMEM) (Gibco, Carlsbad, CA, USA) supplemented with 10% (v/v) fetal calf serum (FCS) (Sigma- Aldrich, St. Louis, MO, USA), 2 mM L-glutamine (Gibco, Carlsbad, CA, USA), penicillin (100 U/ml) and streptomycin (100 μg/ml). Cultures were incubated at 37°C in a humidified atmosphere of 7% CO₂/ 93% air (17). Cells were maintained by removing the conditioned medium and replacing it with a fresh one, every 7 days. After 3-6 weeks in culture, a cellular confluent monolayer had grown out from the bone fragments. The primary cell layers (El culture) were washed in PBS and then treated for 5 min with a solution containing 0.05% (w/v) trypsin and 0.02% (w/v) EDTA (Gibco, Carlsbad, CA, USA). Detached cells were passaged at a density of 60,000 cell/well into a 12-well multiplate (Corning Costar, Corning, NY, USA) and cultured in DMEM supplemented with 10% (v/v) FCS until confluence (ElPl culture). Bone fragments from El plates were also removed and placed in a different plate to obtain E2 and E2P1 cultures, following the procedure mentioned above. Cultures were carried out in parallel using DMEM supplemented with 10% (v/v) FCS or 2% (w/v) Ultroser G (UG, Gibco, Carlsbad, CA, USA), as culture medium (18). SaOS -2 cell culture The human osteosarcoma cell line SaOS-2 was obtained from ATCC (HTB-85) and cultured in DMEM supplemented with 10% (v/v) FCS, at 37°C, in a humidified atmosphere of 7% CC>2/93% air. Cells were collected when at confluence. Assessment of osteoblast viability and proliferation Cells were plated in 12 well plates at 3x10⁴ cells/well, and incubated in DMEM supplemented with 10% FCS or 2% Ultroser G, UG (Gibco, Carlsbad, CA) . At day 1, 3, 6, 10, 16 and 20 during cell culture, medium was removed and cells were washed with PBS. After a treatment with a solution containing 0.05% (w/v) trypsin and 0.02% (w/v) EDTA (Gibco, Carlsbad, CA, USA) for 5 min, they were counted using a Neubauer chamber. Cell viability was assessed by the Trypan blue dye exclusion method . Assessment of mature osteoblast phenotype by Transmission Electron Microscopy Osteoblast cells independently cultured with 10% FCS or 2% UG were processed for morphological analysis on the 15th day of culture. All samples were fixed for 2h at 4°C in cold Karnovsky fixative, rinsed overnight in 0.1M pH7.2 cacodylate buffer and post-fixed for Ih at 4°C in 1% OsO₄, dehydrated in a graded series of ethanol and embedded in Epon- Araldite. Ultra thin sections cut with an LKB III ultramicrotome were collected in copper grids, stained with uranyl acetate and lead citrate and then photographed with a Philips CMlO electron microscope. The authors observed at least 50 cells from each group for evaluation.

Biochemical alkaline phosphatase assay

Alkaline phosphatase (AP) activity was measured directly on the monolayer of the cultures. After medium removal, cells were washed three times with PBS and shaken for 30 min at 37°C in 1 ml of saline buffer containing 10 mM p-nitrophenylphosphate (PNP; Sigma-Aldrich, St. Louis, MO, USA). PNP solution was removed and cells were added with 1 ml of IN NaOH. Optical density was measured at 405 nm (9). Alkaline phosphatase activity values were normalized to the relative number of viable cells, as determined using the proliferation assay mentioned above. Histochemical alkaline phosphatase determination Cell cultures were washed three times with cold PBS, fixed in 90% (v/v) ethanol for 10 min, and incubated in 5-bromine-4-chlorine-3-indolyl phosphate/nitro blue tetrazolium (BCIP/NBT tablets Alkaline Phosphatase Substrate pH 9.5; Sigma-Aldrich, St. Louis, MO, USA) (19). Cells were observed under the light microscope without counterstaining. Nodules formation and mineralization Mineralized nodules formation and degree of mineralization were determined for osteoblast cultures grown in 6-well plates using Alizarin Red S staining. Briefly, medium was aspirated from the wells, and cells were rinsed twice with PBS. Cells were fixed with ice-cold 70% (v/v) ethanol for 1 h. Ethanol was removed, and the cells were rinsed twice with deionised water. Cells were then stained with 40 mM Alizarin Red S in deionised water (adjusted to pH 4.2), for 10 min, at room temperature. Alizarin Red S solution was removed by aspiration and the cells were rinsed five times with deionised water. Water was removed by aspiration, and the cells were incubated in PBS for 15 min at room temperature on an orbital rotator. PBS was removed, and the cells were rinsed once with fresh PBS. The counting of mineralized Alizarin Red S-positive nodules present in each well was then performed using light microscopy. Statistical analysis Data were expressed as mean ± SD of all cultures analyzed. Differences between the values were tested for statistical analysis of variance (ANOVA) using two-tailed Student's t test. The levels of significance were corrected by the Bonferroni test for multiple comparisons. The values of P<0.05 were considered to be statistically different to control. Protein extract preparation Cellular suspensions were centrifuged at 3000 x g for 5 min in a Beckman model J2-21 centrifuge equipped with a JAlO rotor. The supernatant was discarded and the pellet was washed three times with PBS. The cell pellets were immediately suspended and denatured in the buffer for 2-DE analysis consisting of a solution containing 8 M urea, 4% (w/v) CHAPS, 40 mM Tris, 65 mM dithioerythritol and a trace of bromophenol blue. Protein concentration determination Total protein content was determined in cell cultures lysates using the BioRad protein assay. The total protein values were normalized to the relative number of viable cells. Two-dimensional electrophoresis

2-DE was carried out according to procedures detailed elsewhere (20). 45 μg of protein sample was applied to an Immobiline strip (IPG, Immobilized pH Gradient, Amersham Bioscience) consisting of a non-linear gradient, pH range 3.5-10, previously rehydrated. Isoelectric focusing was carried out on Multiphor II (Amersham Bioscience). The voltage was linearly increased from 300 to 3500 V during the first 3 h and then stabilized at 5000 V for 22 h (total 110 kV x h). The IPG strip was then equilibrated in 6 M urea, 30% (w/v) glycerol, 2% (w/v) SDS, 0.05 M Tris-HCl pH 6.8, 2% (w/v) dithioerythritol and then with 2.5% (w/v) iodoacetamide. Electrophoresis in the second dimension was carried out on a 9-16% polyacrylamide non-linear gradient gel (18 x 20 cm x 1.5 mm), at a constant current of 40 mA. Gels were stained with silver nitrate as previously described (20). Analysis was performed in triplicate. The digitalized images were obtained by scanning of the gels (Image Scanner, Amersham Bioscience) and then analysed qualitatively and quantitatively by the Melanie II 2D-P AGE and PDQuest softwares (Bio-Rad, Hercules, CA, USA). Spot intensities were obtained in pixel units and normalized to the total absorbance of the gel. The increasing/decreasing index (fold change) was calculated as the ratio of spot intensity (relative volumes) between the different gel maps. Electrotransfer and immunoblotting

Following 2-DE run, the electropherograms were electrotransferred onto a nitrocellulose membrane (Bio-Rad, 20 x 20 cm, 0.5 μm pore size) using a semidry blotting apparatus (Novablot II, Pharmacia-LKB) (21). Immunoblotting was performed with anti-alkaline phosphatase, osteonectin, osteopontin, osteocalcin and bone sialoprotein antibodies (Alexis, Lausen, Switzerland), followed by incubation with horseradish peroxidase- conjugated anti-rabbit Ig (Sigma-Aldrich, St. Louis, MO, USA). Final detection was obtained by enhanced chemiluminescence measurement (Amersham Bioscience). Mass spectrometry Spots from 2-DE were excised from the gel, triturated, alkylated and digested with trypsin as previously reported (22). Gel particles were extracted with 25 mM NH₄HC 03/acetonitrile (1:1 v/v) by sonication and peptide mixtures were concentrated. Samples were desalted using μZipTipC18 pipette tips (Millipore, Billerica, MA, USA) before MALDI-TOF-MS analysis. Peptide mixtures from 2-DE spots were loaded on the MALDI target together with CHCA as matrix, using the dried droplet technique. Samples were analysed with a Voyager- DE PRO spectrometer (Applera, Norwalk, CT, USA). Peptide mass spectra for PMF experiments were acquired in reflectron mode; internal mass calibration was performed with peptides derived from trypsin autoproteolysis. PSD fragment ion spectra were eventually acquired after isolation of the appropriate precursor, as previously reported (22). In both cases, data were elaborated using the DataExplorer 5.1 software (Applera, Norwalk, CT, USA). ProFound software was used to identify spots from NCBI non-redundant database by PMF experiments. Candidates with ProFound's Est'd Z scores > 2 were further evaluated by the comparison with Mr and pi experimental values obtained from 2-DE. Protein Prospector software was used to confirm spot identification using fragment ions obtained by PSD experiments.

Results and Discussion Primary human osteoblast cultures, biochemical and histochemical alkaline phosphatase determination, nodules formation and mineralization

Human primary osteoblast cultures deriving from a single patient were obtained as summarized in Figure 1. The human bone cell isolation procedure used yielded osteoblasts at different stages of differentiation up to mature osteoblasts. The osteoblastic phenotype was assessed by reverse transcriptase polymerase chain reaction using oligonucleotides suitable for Cbfal, alkaline phosphatase and osteocalcin detection (23). In general, parallel cultures of osteoblasts in the presence of FCS and UG in the culture medium showed significant differences in cell proliferation and morphology, AP activity and mineralized nodules formation. It has been reported that the composition of the osteoblast culture medium strictly influences cell differentiation (9). The authors chose to use the serum substitute Ultroser G containing steroids and glucocorticoids to obtain fully differentiated normal human osteoblast cells to be used as an in vitro model (19, 24, 25, 26, 27). Accordingly, cells grown in UG-rich medium displayed a lower proliferation rate than those obtained in FCS, generally acquiring the characteristic cubical-like shape associated to mature osteoblasts even in the first days of culture (Fig. 2 and 3). The achievement of a mature phenotype in which the cellular biosynthetic mechanisms are actively engaged in the production and secretion of the extracellular matrix is also evidenced by means of Transmission Electron Microscopy analysis (data not shown). Cells grown in UG stained strongly for alkaline phosphatase and had a stronger AP activity than cells cultured in the presence of FCS suggesting their occurrence at a pre-osteoblast stage towards an osteoblast stage of differentiation (Fig. 4 and 5). Congruently, mineralized bone nodule formation in the first 20 days of culture was extremely evident and earlier in UG-grown cells with respect to FCS (Fig. 6). In this view, the authors first analyzed cell cultures obtained 20 days after the migration of osteoblast progenitors (El and E2) from bone fragments placed in culture. Even if the osteoblast phenotype was confirmed by PCR assays, these cultures were mainly composed by spindle shaped cells, with low AP activity, and a low capacity to form mineralized nodules (Fig. 6 and 7). The authors then analyzed osteoblast cultures considered to be at an advanced maturation stage. These cultures (ElPl and E2P1), derived from El and E2 cells replated and carried out in culture for further 20 days, displayed a high number of mineralized nodules and were prevalently constituted by cells with the typical cuboidal shape of mature osteoblasts (ElPl, Fig. 6 and 7). Comparative proteomic analysis

A proteomic study on human osteoblasts differentiation was also carried out. Monitoring of several parameters during osteoblast growth allowed the authors to perform gel-analyses dealing with osteoblast populations with a defined stage of differentiation (pre-osteoblast and differentiated cells, El and ElPl, respectively). To evaluate proteome changes associated with cell differentiation, the authors compared the 2-DE maps of different human osteoblast cells (El, E2, ElPl and E2P1) grown under different nutritional conditions (FCS standard culture medium or UG). Reproducibility was assessed by performing culture experiments three times and sets of 12 gels were produced for each cell culture; all the gels associated with the same collection were completely superimposable. The author's analysis allowed ascertaining that El and E2 proteomes were equivalent as well as ElPl and E2P1, thus confirming morphological observations. Determination of the p/ and M_r scales on the gels was performed by gel matching with the calibrated human cells reference gels contained in the Swiss-2DPAGE database (http://www.expasy.ch/ch2d/). About sixty proteins with various functions were identified in both classes of cells (data not shown).

The authors also performed a qualitative and quantitative comparison of El and ElPl cell proteomes (Fig. 8A and 8B, respectively). Out of a total of around 2400 spots, 3 spots or spot series were typically present only in the El gel, while 32 were peculiar of the ElPl proteome. Changes in the intensity of the spots were recognizable by simple visual inspection. In fact, for specific spots the authors verified the complete absence in the El map and, conversely, a total occurrence in the ElPl map, or vice versa. For some other spots, the authors observed an increased/decreased presence in the respective maps. In this case, a quantitative evaluation was carried out by laser densitometry and the relative abundance of individual polypeptides was calculated. Differences between the maps by a factor greater than 2.0 were considered to be significant.

Identification of differentially expressed proteins was achieved by gel matching, immunoblotting and mass spectrometry procedures. Gel matching was carried out by comparing the authors gels to the reference human cells gels present in the Swiss-2DPAGE database. In some cases, identification was confirmed by mass spectrometry analysis (data not shown). Immunoblotting allowed identifying the main known osteoblast differentiation markers (osteocalcin, osteopontin, osteonectin, alkaline phosphatase and bone sialoprotein). Mass spectrometric identification of the remaining protein species was obtained by mass fingerprint methodologies and confirmed by two independent PSD experiments on most intense peptide ions. The results are summarised in Table I.

Table I. Identified proteins in ElPl osteoblast 2D proteomic map whose abundance increased with respect to El osteoblast map. In the lower part of the table, conventional osteoblast differentiation markers are also indicated, in order to show that their levels did not change significantly. MS, mass spectrometry; IB, immunoblotting; GM, gel matching.

The proteomic profile of other osteoblast-like cells was analysed as well (Figure 8C). In this case, SaOS-2 human osteosarcoma cells were compared to ElPl cells to search for proliferation indicators and potential tumour markers. Quantitative differences were evaluated by laser densitometry, as reported below. This comparison allowed the identification in SaOS-2 cells of specific protein spots over-expressed (Table II) and other under-expressed (Table III), which were identified by gel matching, immunoblotting and mass spectrometry procedures.

Table II. Identified proteins in SaOS-2 2D proteomic map whose abundance increased with respect to ElPl osteoblast map. MS, mass spectrometry; IB, immunoblotting; GM, gel matching.

Table III. Identified proteins in SaOS-2 2D proteomic map whose abundance decreased with respect to ElPl osteoblast map. MS, mass spectrometry; IB, immunoblotting; GM, gel matching.

Proteins over-expressed in differentiated osteoblasts with respect to pre-osteoblasts Seventeen protein molecular species (28) were identified as typically up-regulated in the proteome of ElPl cells (Table I). According to a functional classification, they can be grouped into four classes: cellular structure/motility, protein degradation, energy production and anti-oxidative defence. In addition, enoyl-CoA hydratase-like protein was also found strongly over-expressed in ElPl cells (3.5 fold increase) and, being so far a hypothetical protein with unknown function, was not included in any functional classification. Most of these identified proteins have never been reported before as implicated in the control of osteoblasts differentiation.

In particular, actin, tubulin and Hsp27 are proteins playing a critical role in cell architecture/motility and are essential for various life cyclic processes. In fact, the transformation of osteoblasts to osteocytes necessarily requires a significant change in morphology, adhesion and motility (58). The observed increased expression of actin and tubulin during osteoblast differentiation process was congruent with data previously reported (14, 29). Moreover, Hsp27, a small cell- and tissue-specific chaperone that undergoes estrogen regulation, has a clear regulatory role in actin polymerization but an unknown control mechanism and has been also found associated with differentiation of MCF-7 human breast cancer cells (30). The authors observed an increased expression of two Hsp27 isoforms, presumably related to different phosphorylated species (30, 31). Thus, the increased levels observed for these architecture/motility proteins may contribute to generate actin-rich cell projections, typical of mature osteocytes (13). Two proteins involved in polypeptide degradation, namely cathepsin D and ubiquitin C- terminal hydrolase Ll, showed a strong expression increase associated with cell differentiation. Cathepsin D, here occurring as three different isoforms, is an aspartic protease previously detected in ossifying cartilage by immunohistochemistry approaches (32). It plays a role in extracellular processing of potent stimulators of osteoblastic mitogenesis, namely TGF-β and IGF-II. Arrayed cDNA hybridization techniques revealed its increased expression associated with maturation of osteoblast-like MC3T3-E1 cells (15). Moreover, cathepsin D is a marker of ganglion cell differentiation (33). On the other hand, UCH-Ll is a thiol-hydrolase involved in the proteolytic processing of polymeric ubiquitin and generation of monomeric ubiquitin, the active component of eukaryote ubiquitin-dependent proteolytic systems. In eukaryotes, the ubiquitin degradation pathway is essential for regulation of normal growth, proliferation and differentiation (34). UCH-Ll is known to have a controversial role in Parkinson's disease etiology and its gene has been associated to both opposite hydrolase and ligase activities (35). A previous work reported UCH-Ll as being implicated in differentiation of Reh leukemia lymphoblasts (36), suggesting a role as a transcriptional or post-transcriptional regulator. UCHL-I was also found being over-expressed during human neuronal differentiation (37). In this sense, UCHL-I seems integrated in the ubiquitin-proteasome pathway and required for the regulation of the nuclear transcription factor kB. The present work is the first report on UCHL-I over-expression during bone cell differentiation. An increased amount of phosphoglycerate mutase 1, pyruvate kinase Ml, α-enolase and ATP synthase was observed in ElPl cells compared to El ones (Table I). Enzymes involved in glycolysis and ATP biosynthesis have been generally reported as markers of the energy-demanding differentiation. So far, their up-regulation has never been reported in osteoblasts. Their concomitant over-expression has been associated with an accelerated glycolysis rate that consequently influences tricarboxylic acid and oxidative phosphorylation rates, increasing ATP levels. Alpha-enolase was found up-regulated in various proliferating human cells and down-regulated in differentiating cell types (38). In particular, the authors found over-expression of a single α-enolase isoform in differentiating osteoblasts, similarly to what observed during Caco-2 cells differentiation (39).

Finally, three protein species involved in anti-oxidative defence were found being up- regulated, namely glutathione S-transferase P, SOD and glutamate dehydrogenase 1 (Table I). It has been widely reported that oxidation reduces both proliferation and differentiation in human osteoblast-like cells and SOD prevents oxidative damage favouring differentiation (40, 41). SOD overexpression is related to an ameliorate malignant SaOS-2 phenotype (42). Glutathione S -transferases have been reported as up-regulated during Caco-2 differentiation (43). The authors also monitored expression of all the classical markers of osteoblast differentiation functionally involved in extracellular matrix/cell interaction (osteocalcin, osteopontin, bone sialoprotein, alkaline phosphatase and osteonectin) by specific immunoblotting assays. The quantitative proteomic analysis revealed that these markers were not differentially expressed, with the only exception of osteocalcin. This finding is not surprising since it has been demonstrated that there is no quantitative correlation between mRNA and corresponding protein levels (44). As far as alkaline phosphatase, the authors found only a slight increased expression in one of the two isoforms in ElPl proteome. Proteins over-expressed in SaOS-2 with respect to differentiated osteoblasts Osteosarcoma is the most frequent sarcoma in children. In this disease, abnormal proliferation and osteoblast cell function as well as production of an incompletely mineralized matrix are observed. Osteosarcoma physico-chemical and molecular features are poorly delineated so far. Gene expression information on osteoblast maturation leading to osteosarcoma tumorogenesis are limited. The present invention identifies proteins over- expressed in SaOS-2 cells with respect to osteoblasts (Table II). These proteins can be used as markers in investigations on malignant bone cells and to profile molecular-cellular changes associated with tumour progression. In particular, different glycolytic enzymes were found as over-expressed in SaOS-2 cells, namely pyruvate kinase, lactate dehydrogenase and triosephosphate isomerise (Table II). Lactate dehydrogenase and pyruvate kinase have been already reported as tumour markers (43), although they have never been reported in osteosarcoma so far. Similarly, the authors found over-expression of Hsp90, which has been recently classified as a tumour indicator in various human tumour cell lines, including SaOS-2 cells (16).

The expected reduced activity of alkaline phosphatase in osteosarcoma, normally considered as an osteoblast differentiation marker in cells where bone formation does not take place, did not found a relative protein expression counterpart in SaOS-2 proteomic map. In fact, the authors observed a specific 4-fold increase for one of the two alkaline phosphatase isoforms (Table II). These two molecular species, having a similar M_r but a different p/, should possibly present a different post-translational modification degree, thus hypothetically playing a different role within the cell. The nature of different proteomic profiles and alkaline phosphatase amounts observed in SaO S-2 and normal osteoblast cells highlighted the erroneous definition of osteosarcoma as osteoblast-like cells, underlining the necessity to test many therapeutics for bone metabolic diseases on the more appropriate and defined osteoblast cell type.

Creatine kinase B (BB-CK) was also found differently over-expressed in one of the two molecular species occurring in SaOS-2 proteomic map (Table II). BB-CK is an enzyme that regulates intracellular ATP/ ADP concentration and acts as a general marker for the action of hormones involved in energy-demanding processes such as cell growth. BB-CK is known to be synthesized in SaOS-2 and other cell lines whose response to gonadal steroids is gender-specific (45). Its presence is usually associated with contemporary expression of bone marker osteocalcin. The authors found an under-expression of osteocalcin in SaOS-2 (Table III), congruently with the active proliferation of osteosarcoma cells that do not differentiate and do not produce bone tissues. The present work is the first report on differential BB-CK synthesis among immortalized cell lines and normal human osteoblasts. Two additional proteins were found for the first time over-expressed in SaOS-2 cells, namely 15OkDa oxygen-regulated protein (ORP150) and retinoblastoma-binding protein 4, and are considered as potential tumour markers. The latter is encoded by a gene that is up- regulated in acute myelocytic leukemia and controls cell proliferation, differentiation and survival, preventing apoptosis during leukemo genesis (46). The mRNA of an analogous protein, namely RBPlLl, was found over-expressed in breast, lung, colon, pancreas and ovarian cancers, as well as over-transcripted in normal testis (47). Thus, the present findings originally confirm at a proteome level previous transcriptome reports on other types of tumour. On the other hand, ORP 150 is a so far uncharacterized protein that probably functions as a molecular chaperone. A significant correlation has been reported in bladder cancer between expression of ORP150 and matrix metalloproteinase 2 (MMP-2), suggesting a role of ORP150 in MMP-2 secretion for tumour invasion and metastasis (48). ORP 150 also plays an important role in successful adaptation to oxygen deprivation during osteonecrosis (49). Proteins under-expressed in SaOS-2 with respect to differentiated osteoblasts

Sixteen protein species were remarkably under-expressed in SaOS-2 cells with respect to differentiated osteoblasts (Table III). These proteins may be important to understand biochemical mechanisms associated with osteosarcoma tumorogenesis. Particularly interesting were the proteins whose expression was reduced in pre-osteoblasts and SaOS-2 cells with respect to differentiated osteoblasts, namely Hsp27, cathepsin D, superoxide dismutase, glutamate dehydrogenase, osteocalcin and enoyl-CoA hydratase-like protein. The data reported here emphasize their use as specific markers of osteoblast differentiation. These data also confirm differences between cell lines and primary culture cells. In the present invention, quantitative findings on cathepsin D in SaOS-2 cells were congruent with previous observations in MG-63 osteoblast-like cells (50). Interestingly, the authors observed a contemporaneous decrease in both SOD and osteocalcin in SaOS-2 cells. This result is different from what is required for osteoprogenitors differentiation (52). This can be explained by the fact that in osteosarcoma proliferation takes place instead of differentiation and bone formation.

Furthermore, the authors observed a remarkable under-expression of galectin-1 and annexins (Table III). Galectin 1 is a protein that is tightly associated with nuclear matrix during osteoblast differentiation. As observed in rat calvarial osteoblasts, it is present in the cytoplasm and nucleus of both proliferating and differentiated cells (53); accordingly, the authors did not observed differences in galectin 1 content between El and ElPl proteome maps. The reduced expression of galectin 1 observed in SaOS-2 contrasts with previous reports on various cancer cells, where its levels were correlated with loss of differentiation- specific functions, although independently from tumorigenicity (54). On the other hand, annexin I is known to induce cell differentiation (55), promote apoptosis (56) and its diminished expression is congruent with SaOS-2 cells biological features. Annexin II is over-expressed in osteoclasts and osteoclastoma, since it promotes bone resorption. It is considered as a tumour marker for many cancer types. Nevertheless, it has been found under-expressed in some cancers (57) and it plays an important role in binding collagen in the extracellular matrix. Its under-expression in SaOS-2 cells could account for the reduced production of extracellular matrix observed in osteosarcoma.

Bibliography 1 Fujikawa, Y., Sabokbar, A., et al, Bone 2001, 28, 261-267.

2 Hofbauer, L. C, et al, J Musculo skelet Neuronal Interact. 2004, 4, 268-275.

3 Fleisch, H., Endocr. Rev. 1998, 19, 80-100.

4 Rodan, G. A., Annu. Rev. Pharmacol. Toxicol. 1998, 38, 375-388.

5 Reinholz, G. G., Getz, B., et al., Cancer. Res. 2000, 60, 6001-6007. 6 Im, G. L, Qureshi, S. A., et al., Biomaterials. 2004, 25, 4105-4115.

7 Stein, G. S., et al., FASEB J. 1990, 4, 3111-3123.

8 de Pollack, C, et al., Cell Biochem. 1997, 64, 128-139.

9 Rattner, A., et &\.,In Vitro Cell Dev. Biol. Animal 1997, 33, 757-762.

10 Gartland, A., et al., Methods MoI. Med. 2005, 107, 29-54. 11 Ducy, P., Schinke, T., Karsenty, G., Science 2000, 289, 1501-1504.

12 Beresford, L. F., et al., Metab. Bone Dis. Relat. Res. 1984, 5, 229-234.

13 Billiard, J., Moran, R. A., et al., J. Cell. Biochem. 2003, 89: 389-400.

14 Billiard, M., et al., Suppl. Clin. Neurophysiol. 2000, 53, 362-365.

15 Kitching, R., et al., J. Bone Miner. Metab. 2002, 20, 269-280. 16 Myung, J-K., et al., Proteome Science 2004, 2, 1-21.

17 Walsh, C. A., et al., J Bone Miner Res 1995, 10, 17-25.

18 Garcia-Moreno, C, et al., Bone 1998, 22, 233-239.

19 Durant, S., Duval, D., et al., J. Steroid Biochem. 1989, 33, 1103-1110.

20 Bernardini. G., Renzone, G., et al., Proteomics 2004, 4, 2893-2926. 21 Towbin, H.; ,Stahelin, T., Gordon, J., PNAS 1979, 76, 4350-4354.

22 Talamo, F., DAmbrosio, C, et al., Proteomics 2003, 3, 440-460.

23 Frediani, B., et al., Bone 2004, 35, 859-869.

24 Green, E., Todd, B., Heath, D., Eur. J. Biochem. 1990, 188, 147-153. 25 Wong, M. M., et al., /. Bone Miner. Res. 1990, 5, 803-813.

26 Ogata, Y., et al., J., Eur. J. Biochem. 1995, 230, 183-192.

27 Walsh, S., et al., Rheumatology 2001, 40, 74-83.

28 Junglblut, P. R., Microbes Infec. 2001, 831-840. 29 Lomri, A., Marie, P. J., Bone Miner. 1990, 10, 1-12.

30 Chen, S-T., Pan, T-L., Tsai, Y-C, Huang, C-M., Cancer Letters 2002, 181, 95-107.

31 Horman, S., Fokan, D., Galand, P., Cell Prolif. 2000, 33, 331-340.

32 Poole, A. R., Hembry, R. M., Dingle, J. T., Calcif Tissue Res 1973, 12, 313-321.

33 Magro, G., Ruggirei, M., et al., Virchows Arch 2000, 437, 406-412. 34 Ciechanover, A., Orian, A., Schwartz, A. L., / Cell Biochem 2000, Suppl 34, 40-51.

35 Liu, Y., Fallon, L., et al., Cell 2002, 111, 209-218.

36 Maki, A., Mohammad, R. M., et al., Differentiation 1996, 60, 59-66.

37 Satoh, J. L, Kuroda, Y., Neuropathol. Appl. Neurobiol. 2001, 27, 95-104.

38 Olsen, E., Rasmussen, H. H., Celis, J. E., Electrophoresis 1995, 16, 2241-2248. 39 Stierum, R., Gaspari, M., et al., Bioch. Biofis. Acta 2003, 1650, 73-91.

40 Hikiji, H., et al., Am. J. Physio. I Endocrinol. Metab. 2000, 278, E1031-1037.

41 Da Rocha, F. A. C, et al., J. Bone Miner. Res. 2002, 17, 434-442.

42 Komatsu, et al., Int. J. Oncol. 2005, 26, 853-862

43 Perrin, A., et al., Biochimie. 2002, 84, 1003-1011. 44 Gygi, S. P., et al., MoI. Cell. Biol. 1999, 19, 1720-1730.

45 Katzburg, A. M., et al., J. Endocrinol. Invest. 2001, 24, 166-172.

46 Casas, S., et al., Cancer Genet. Cytogen. 2003, 146, 89-101.

47 Cao, J., et al., J. Natl. Cancer Inst. 2001, 93, 1159-1165.

48 Koshida, K., et al., Himyokika Kiyo 2001, 47, 815-818. 49 Sato, M. et al., J. Bone Joint Surg. Br. 2001, 83, 751-759.

50 Sourla, A., et al., Anticancer Res. 1996, 16, 2773-2780.

51 Miller, A. C, et al., J. Inorg. Biochem. 2002, 25, 246-252.

52 Wang, F. S., et al., J. Biol. Chem. 2002, 277, 10931-10937.

53 Choi, J-Y., et al., J. Cell Science 1998, 111, 3035-3043. 54 Chiarotti, L., et al., Cell Growth Different. 1994, 5, 769-775.

55 Sato, F. E., et al., Biochim. Biophys. Acta 1995, 1266, 23-30.

56 Canaider. S., et al., Life Sci. 2000, 66, PL265-270.

57 Chiang, Y., et al., Biochim. Biophys. Acta 1996, 1313, 295-301. Luegmayr, E., et al., Bone 1996, 18, 591-599. Spreafico, A., et al., /. Cell Biochem. 2006, 98(4): 1007-20. Jones, L. K., Saha, V., Br. J. Haematol. 2005, 130, 489-500. Thorpe, W., Loeb, W., Hyatt, C, Rosenberg, S.A., Cancer Res. 1979, 39, 277-279. Thorpe, W.P., Reilly J., Rosenberg, S.A., Cancer 1979, 43, 2178-2181. Levine, A.M., Rosenberg, S. A., Cancer 1979, 44, 2291-2293.

Claims

Claims

1. A method for the diagnosis of a metabolic bone disease, and/or for assessing osteoporotic fracture risk and/or for the monitoring of therapeutic efficacy of a metabolic bone disease drug in a subject, including the following steps: a) culturing subject's bone-derived cells; b) extracting and separating proteins from such cultured cells; c) identifying and quantifying at least one bone cell specific protein; d) comparing the amount of at least one bone cell specific protein with that of a control subject, or of a panel of control subjects.

2. The method according to claim 1 wherein the bone cell specific protein is comprised in the group of: actin, tubulin, heat shock protein 27, cathepsin D, ubiquitin C-terminal hydrolase Ll, phosphoglycerate mutase 1, pyruvate kinase Ml, α-enolase, ATP synthase, glutathione S-transferase P, superoxide dismutase, glutamate dehydrogenase 1 or enoyl- CoA hydratase-like protein.

3. The method according to claim 2 wherein the actin is beta or gamma actin.

4. The method according to claim 2 wherein the tubulin is beta tubulin.

5. The method according to claim 1 wherein the metabolic bone disease is a metabolic osteopathy.

6. The method according to claim 1 wherein the control subject is an healthy subject or a subject in course of a therapeutic treatment.

7. A method for the diagnosis of bone tumor in a subject comprising the steps of: a) culturing subject's bone-derived cells; b) extracting and separating proteins from cultured cells; c) identifying and quantifying at least one bone tumor specific protein; d) comparing the amount of at least one bone tumor specific protein with that of a control subject, or of a panel of control subjects.

8. The method according to claim 7 wherein the bone tumor specific protein is comprised in the group of: pyruvate kinase Ml, L-lactate dehydrogenase B chain, triosephosphate isomerase 1, creatine kinase B chain, heat shock protein 90, 150KDa oxygen-regulated protein (ORP 150), Retinoblastoma-binding protein 4, alkaline phosphatase, galectin-1, annexin, tropomyosin 3 or UMP-CMP kinase.

9. The method according to claim 7 wherein the bone tumor specific protein is comprised in the group of: osteocalcin, osteonectin, Cathepsin D, heat shock protein 27, superoxide dismutase, glutamate dehydrogenase 1 or enoyl-CoA hydratase-like protein.

10. The method according to claim 8 wherein the annexin is annexin I or II.

11. The method according to claims 7-10 wherein the bone tumor is an osteosarcoma.