Biochem. J. (2014) 464, 355–364 (Printed in Great Britain)

355

doi:10.1042/BJ20140702

*Department of Biochemistry, Schulich School of Medicine & Dentistry, University of Western Ontario, London, Ontario, Canada, N6A 5C1 †Department of Molecular Biology and Genetics, Aarhus University, Gustav Wieds Vej 10, DK-8000 Aarhus, Denmark ‡School of Dentistry, Schulich School of Medicine & Dentistry, University of Western Ontario, London, Ontario, Canada, N6A 5C1 §Department of Physiology and Pharmacology, Schulich School of Medicine & Dentistry, University of Western Ontario, London, Ontario, Canada, N6A 5C1

Biomineralization is a complex process in the development of mineralized tissues such as bone and pathological calcifications such as atherosclerotic plaques, kidney stones and gout. Osteopontin (OPN), an anionic phosphoprotein, is expressed in mineralizing tissues and has previously been demonstrated to be a potent inhibitor of hydroxyapatite formation. The OPNdeficient (Opn − / − ) mouse displays a hypermineralized bone phenotype starting at 12 weeks postnatally. By isolating and culturing Opn − / − and wild-type (WT) osteoblasts, we sought to determine the role of OPN and two of its functional peptides in osteoblast development and mineralization. Opn − / − osteoblasts had significantly increased mineral deposition relative to their WT counterparts, with no physiologically relevant change in gene expression of osteogenic markers. Supplementation with bovine milk OPN (mOPN) led to a dramatic reduction in mineral

deposition by the Opn − / − osteoblasts. Treatment with OPNderived peptides corresponding to phosphorylated OPN-(220– 235) (P3) and non-phosphorylated OPN-(65–80) (OPAR) also rescued the hypermineralization phenotype of Opn − / − osteogenic cultures. Supplementation with mOPN or the OPN-derived peptides did not alter the expression of terminal osteogenic markers. These data suggest that OPN plays an important role in the regulation of biomineralization, but that OPN does not appear to affect osteoblast cell development in vitro.

INTRODUCTION

In bone, expression of OPN has been shown to be linked with the mineralization front of osteoid [7–9]. OPN expression has also been associated with pathological calcifications, such as atherosclerosis [1] and kidney stones [10]. The presence of OPN during de novo osteoid deposition and pathological calcification suggests that OPN may regulate mineral growth in both cases. The expression of OPN is also associated with varying degrees of post-translational modification in a tissue- and cell-specific manner, which may relate to its functional properties recently reviewed by Ganss and Bansal [3]. For example, boneextracted phosphorylated OPN is a more potent inhibitor of HA formation and growth in vitro than recombinant OPN lacking these modifications [3]. Mice with the Opn gene ablated (Opn − / − ) have been reported to have increased bone mineral density when assayed using Fourier-transform infrared (FTIR) imaging [11]. This phenotype develops progressively as the animals age [11], suggesting a dysregulation between mineral deposition and resorption by osteoblasts and osteoclasts, respectively. Rittling et al. [12] have demonstrated that Opn − / − mice have significantly reduced osteoclast activity, suggesting that the increased mineral density is due to decreased resorption. Currently, there do not appear to be any published studies examining a time course of mineralization and differentiation of osteoblasts derived from the Opn − / − mice. Previous work has presented conflicting evidence for OPN’s potential role in osteogenic cell differentiation [13,14].

Biomineralization of bone is an intricately controlled process that comprises osteogenic cell development as well as the nucleation and growth of hydroxyapatite (HA) crystals, imparting the tissue with strength and rigidity. Pathological dysregulation of mineralization results in a number of disorders such as atherosclerosis [1] and kidney stones [2]. A number of mechanisms have been proposed to account for the regulation of mineralization in vivo, including the involvement of proteins responsible for mediating osteogenic cell differentiation and the promotion or inhibition of mineral growth and/or nucleation. One protein implicated in the inhibition of mineral formation is osteopontin (OPN). OPN is an anionic phosphoprotein that is expressed in many tissues and cell types, and is also found in physiological fluids [3]. OPN is a member of the small integrin-binding ligand Nlinked glycoprotein (SIBLING) family of proteins that includes bone sialoprotein (BSP), dentin matrix phosphoprotein 1 (DMP1) and others [4]. These proteins are intrinsically disordered, with little or no secondary or tertiary structure [4–6]. They also contain RGD (Arg-Gly-Asp) integrin-binding sequences, highly acidic sequences that may include polyaspartate or polyglutamate sequences, and are highly post-translationally modified [3]. The expression patterns of OPN in different tissues are highly complex and dependent on a variety of regulatory elements.

Key words: differentiation, mineralization, osteoblast, osteopontin, small integrin-binding ligand N-linked glycoprotein (SIBLING).

Abbreviations: BSP, bone sialoprotein; G2P, glycerol 2-phosphate; HA, hydroxyapatite; mOPN, bovine milk OPN; MSC, mesenchymal stromal cell; OPAR, osteopontin polyaspartate region; OPN, osteopontin; qPCR, quantitative real-time PCR; TBST, TBS containing 0.1 % Tween 20; VSMC, vascular smooth muscle cell; WT, wild-type. 1 To whom correspondence should be addressed (email [email protected]).  c The Authors Journal compilation  c 2014 Biochemical Society

Biochemical Journal

Erik Holm*, Jared S. Gleberzon*, Yinyin Liao*, Esben S. Sørensen†, Frank Beier§, Graeme K. Hunter*‡ and Harvey A. Goldberg*‡1

www.biochemj.org

Osteopontin mediates mineralization and not osteogenic cell development in vitro

356

E. Holm and others

Overexpression of OPN in MC3T3-E1 pre-osteoblastic cells resulted in a decrease in the expression of mature osteoblast markers in addition to a decrease in mineral deposition [14]. However, overexpression of OPN in rat bone marrow cells was shown to enhance the expression of osteoblast markers and nodule formation [13]. These contradictory results have yet to be resolved. In order to better understand the role of OPN during bone formation and homoeostasis, it is critical to elucidate its function during osteoblast differentiation and biomineralization. As indicated above, OPN has also been shown to be a potent inhibitor of HA nucleation and growth in vitro [15,16]. Motifs of OPN shown to inhibit mineralization include the highly acidic polyaspartate region [17] and a number of highly acidic phosphorylated sequences [18]. These crystal-binding regions of OPN competitively perturb mineral growth by interacting with the positively charged regions of the crystal surfaces [19]. Bovine milk OPN (mOPN) has been used extensively in studies of OPN inhibition as it is highly phosphorylated and is readily available in large quantities [20]. mOPN has been shown to directly inhibit HA nucleation by binding crystal faces of HA using its multiple crystal-binding motifs [19,21]. Owing to OPN’s localization within atherosclerotic plaques, Giachelli et al. [22] undertook studies using vascular smooth muscle cells (VSMCs) isolated from the Opn − / − mouse. These cells were grown under high phosphate conditions to induce calcification, and were shown to have increased mineral deposition relative to wild-type (WT) controls [23]. Supplementation with exogenous OPN from neonatal rat smooth muscle cell cultures resulted in a decrease in mineral deposition in the same system [24]. These studies confirm that OPN is a potent inhibitor of mineral formation in VSMC cultures. Of the aforementioned OPN motifs, our studies have shown that OPN-(65–80) (SHDHMDDDDDDDDDGD), called OPAR (osteopontin polyaspartate region), and OPN-(220– 235) (pSHEpSTEQSDAIDpSAEK), termed P3, containing three attached phosphates, are potent inhibitors of crystal growth in vitro [18]. However, using osteogenic cells derived from the Opn − / − mouse provides an opportunity to test these peptides, in addition to mOPN, under pseudophysiological conditions to examine cell differentiation and mineral formation without interference from endogenous OPN. Comparing the effects on differentiation and mineralization for both mOPN and OPN-derived peptides, the latter lacking any apparent cell signalling properties, will contribute to our understanding of the potential mechanism of OPN effects on biomineralization. We hypothesize that OPN and its derived peptides are potent inhibitors of osteoblast-mediated mineralization. In the present study we have characterized the phenotype of Opn − / − osteoblasts. We have also determined the inhibitory capacity of mOPN and two functional peptides of OPN, namely P3 and OPAR. We demonstrate that Opn − / − osteoblast cultures have an increased mineralization phenotype that is independent of osteoblast differentiation. We also show that, upon treatment with mOPN and OPN derived-peptides, there is a reduction of mineral deposition with no alteration of the expression characteristics of terminal osteoblast markers. These data suggest that the main role of OPN in osteoblasts is to control extracellular mineralization and not cellular differentiation. MATERIALS AND METHODS Animal protocol

Animal procedures were performed in accordance with guidelines of the Canadian Council on Animal Care (CCAC) and Animal  c The Authors Journal compilation  c 2014 Biochemical Society

Care and Veterinary Services (ACVS), University of Western Ontario protocol number 2008-092. Mice were obtained from the Jackson Laboratory (JAX), originally documented by Liaw et al. [25] (JAX Stock number 004936, Sacramento, USA). Preparation and genotyping of Opn − / − homozygous mice and WT controls were performed as described by the supplier (JAX). Mice were maintained on a C57BL/6 background and were fed on a standard pelleted mouse diet (2018 Tekland Global 18 % protein diet, Harlan Laboratories) and tap water ad libitum. The Opn − / − and WT mice breed normally and both can be maintained on a homozygous background. mOPN and peptide preparation

mOPN, comprising a mixture of both intact and OPNderived peptides, was purified as described previously by Sørensen and Petersen [26]. Synthetic peptides were prepared as described previously [18]. Briefly, peptides (OPAR, SHDHMDDDDDDDDDGD, and P3, pSHEpSTEQSDAIDpSAEK) were synthesized using solid-phase method by Fmoc chemistry with free N- and C-termini. These were then purified using reversephase HPLC. Purity and protein content were confirmed using ESI–MS, HPLC and amino acid analysis respectively. Cell culture and in vitro assays

Isolation and culture of mesenchymal stromal cells (MSCs) was performed as described previously with minor modifications [27]. Cells were isolated by flushing the femurs of 8–10-weekold mice and were allowed to adhere to T-80 flasks (Nunc) for 3 days in Iscove’s modified Dulbecco’s medium (IMDM) with 10 % (v/v) FBS, 100 units/ml penicillin and 100 μg/ml streptomycin (all reagents from Life Technologies). The cell culture medium was changed every 48 h. At 70 % confluence, cells were passaged into 24- and 96-multiwell dishes (Nunc) at a density of 104 cells/cm2 . The cells were allowed to proliferate for 3 days and were then grown in mineralizing medium containing alpha minimal essential medium (αMEM) with 10 % (v/v) FBS, 100 units/ml penicillin and 100 μg/ml streptomycin, supplemented with 50 μg/ml ascorbate (Sigma– Aldrich) and 2 mM glycerol 2-phosphate (G2P) (Sigma–Aldrich). To study the effect of OPN on cell development and mineralization, the medium was supplemented with mOPN at final concentrations of 5, 10 or 20 μg/ml, or with the OPNderived peptides, OPAR and P3, at 20 μg/ml. Supplementation of the culture medium with OPN or its peptides was continued throughout the culture period. Metabolic activity was measured using alamarBlue® (Life Technologies) reagent, whose main constituent is resazurin which produces a fluorescently active metabolite in linear proportion to cell number [28,29]. Fluorescence was measured with excitation and emission wavelengths of 570 nm and 585 nm respectively on a Tecan Safire plate reader. At least three independent experiments were performed with each carried out in triplicate. Mineralization was assayed using Alizarin Red S dye (Ricca Chemical Company) and calcium quantification (QuantiChromTM Calcium Assay Kit, DICA-500, Bioassay Systems). Cells were fixed in anhydrous ethanol for 10 min at − 20 ◦ C. Calcium determination was performed according to the manufacturer’s protocol. Other cultures were then stained with 500 μl of 1 % (w/v) Alizarin Red S dye and incubated for 10 min. Unbound stain was removed by washing with distilled water, and then plates were allowed to dry overnight. Images of wells were taken using a LED2500 dissection microscope with integrated EC3 digital

Osteopontin mediates mineralization and not osteogenic cell development in vitro

Figure 1

357

Biomineralization is increased in the Opn − / − osteoblasts

Quantification of Alizarin Red S dye at A 415 (A) and total calcium (B) extracted from Opn − / − (broken line) and WT (unbroken line) cell cultures at weekly time points. (C) Representative images of the Alizarin Red S-dyed experimental wells. (D) Metabolic activity assay of Opn − / − and WT cell cultures at weekly time points. *P < 0.05, n = 4, total calcium n = 5.

camera (Leica Microsystems). Dye was then extracted using 1 ml of 0.6 M HCl and 0.7 M SDS (Sigma–Aldrich) for 1 h while shaking. A 200 μl aliquot of dye was transferred to a 96-well plate and absorbance was measured at 415 nm on an iMark plate reader (Bio-Rad Laboratories). At least three independent experiments were performed with each carried out in triplicate.

Quantitative real-time PCR (qPCR)

Total RNA was isolated from cells using TRIzol® (Life Technologies) according to the manufacturer’s guidelines. PCRs were performed using One-Step Master Mix (4309169, Life Technologies) on a 7900 HT system (Life Technologies) according to the manufacturer’s guidelines. Gene expression assays of Alpl, Ank, Bglap, Col1a1, Dmp1, Enpp1, Ibsp, Opn, Runx2 and Sp7 were performed using Taqman probes (Life Technologies). B2m was used as a reference gene for normalization of expression and relative quantification. These genes and their protein names are shown in Table 1.

Table 1 Gene names associated with their protein names and primer codes used for qPCR Gene symbol

Protein name

qPCR probe ID

Alpl Ank Bglap

Alkaline phosphatase (ALP) Ankylosis, progressive homologue Bone γ -carboxyglutamate protein/osteocalcin (OCN) β 2 -Microglobulin Type I collagen α1 Dentin matrix acidic phosphoprotein 1 (DMP1) PC1 (ectonucelotide pyrophosphatase/phosphodiesterase 1) Integrin-binding sialoprotein/BSP Runt-related transcription factor 2 Osterix

Mm00475834_m1 Mm00445040_ml Mm03413826_mH

B2m Col1a1 Dmp1 Enpp1 Ibsp Runx2 Sp7

Mm01269327_m1 Mm00801666_g1 Mm01208363_m1 Mm00501097_m1 Mm00492555_m1 Mm00501580_m1 Mm00504574_m1

 c The Authors Journal compilation  c 2014 Biochemical Society

358

Figure 2

E. Holm and others

Analyses of osteogenic marker expression in Opn − / − osteoblasts show few changes in developmental expression characteristics using qPCR

qPCR examining markers of osteoblastic differentiation taken at weekly time points. The expression of all genes is normalized to that of B2m. *P < 0.05, n = 4.

Immunoblots

Total cell lysate for immunoblots was collected using RIPA buffer (R0278, Sigma–Aldrich) according to the manufacturer’s guidelines. The protein was then quantified using the Bradford method, and 40 μg of lysate was loaded and separated by SDS/PAGE (12 % gels). The protein was then transferred on to a PVDF membrane (GE Healthcare) for immunoblotting. Membranes were blocked using 2.5 % blocking agent (RPN2125, GE Healthcare) in TBS containing 0.1 % Tween 20 (TBST) for 1 h at room temperature and then probed with primary antibody  c The Authors Journal compilation  c 2014 Biochemical Society

(1:1000-diluted rabbit anti-BSP, courtesy of Dr Renny Franceschi, Ann Arbor, MI, U.S.A., or 1:1000-diluted rabbit anti-OCN, Fl-95, Santa Cruz Biotechnology) in 2.5 % blocking solution overnight at 4 ◦ C. The membranes were then washed three times for 15 min in TBST and probed with a secondary antibody [1:10000diluted horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG, PI-1000, Vector Laboratories] at room temperature for 1 h. The membranes were washed again three times for 15 min in TBST and developed using ECL prime Western blotting reagent (RPN2232, GE Healthcare) and exposed on a ChemiDoc MP (Bio-Rad Laboratories). Densitometry was then performed to

Osteopontin mediates mineralization and not osteogenic cell development in vitro

Figure 3

359

Supplementation of mOPN attenuates Opn − / − osteoblast biomineralization

Quantification of mineralization of cultures at 28 days post-induction treated with mOPN (x -axis demonstrates mOPN concentration in μg/ml). Mineralization was assayed by measuring (A) extracted Alizarin Red S at A 415 and normalizing to WT levels or (B) by extraction and quantification of total calcium. (C) Representative images of the osteoblast cultures stained with Alizarin Red S. (D) Metabolic activity of the osteoblast cultures at day 28. *P < 0.05, n = 4.

quantify bands using the Image Lab software (version 5.0, BioRad Laboratories).

Statistical analyses

Statistical analysis was performed using a two-way ANOVA for time-course studies, and one-way ANOVA for single time-point experiments. Both were analysed with post-hoc Tukey’s test using GraphPad Prism version 4.00.

RESULTS Characterization of Opn − / − osteoblasts

In order to study the effect of OPN on extracellular matrix mineralization, we first assayed the ability of Opn − / − osteoblasts to mineralize in tissue culture. Staining with Alizarin Red S and total calcium content determination showed a linear increase in mineralization in the WT osteogenic cultures throughout the time course. The Opn − / − cells demonstrated a similar linear increase in mineralization over time; however, there is a greater mineral content in cultures relative to WT cells (Figures 1A–1C). To address whether the observed difference is due to increased cell numbers, we performed a mitochondrial metabolic activity assay. There was no difference found between the Opn − / − and WT osteogenic cell cultures, suggesting that the differences

in mineralization are not due to differences in cell number (Figure 1D). To determine whether there are differences in the expression of genes related to osteogenic cells, qPCR was performed. No deviation was observed from the WT expression patterns for a number of osteogenic genes from the Opn − / − cells, including Ank, Alp1, Bglap and Dmp1 (Figure 2). Transcription factors Runx2 and Sp7 were also unchanged (results not shown). Other osteogenic genes showed some differences. Col1a1 expression was reduced by half at day 7, and by 40 % at day 14, in the Opn − / − osteoblasts relative to the WT cells. Expression of Ibsp at day 14 was increased by 100 % in the Opn − / − cells compared with the WT cultures. The expression of Enpp1 was also increased by 60 % in the Opn − / − osteogenic cultures at day 28 compared with the WT cells. Other than these exceptions, the expression characteristics are indistinguishable between the WT and Opn − / − cells. We also noted no difference in BSP or osteocalcin protein expression between the two genotypes at day 28 (see Figures 4C–4E). The overall similarity in expression of multiple genes throughout the time course suggests that OPN does not play a critical role in the maturation of osteoblasts. Treatment with mOPN

The enhanced mineralization observed in the Opn − / − osteoblasts, in the absence of any large changes in cell metabolic activity or expression of differentiation markers, suggests  c The Authors Journal compilation  c 2014 Biochemical Society

360

Figure 4

E. Holm and others

Treatment with mOPN does not change expression of terminal osteogenic markers of differentiation of Opn − / − osteoblasts

qPCR was used to measure expression of (A) Ibsp and (B) Bglap in the untreated and the mOPN-treated osteoblast cultures at 28 days post-induction. (C) Immunoblots for OCN (osteocalcin) and BSP protein expression at week 4 with β-actin shown as the loading control. Quantification of the (D) BSP and (E) OCN bands relative to the β-actin loading control. The x -axis represents the amount of mOPN supplemented in μg/ml. qPCR n = 4, immunoblot n = 3.

that OPN functions primarily to inhibit mineral deposition and/or growth. We therefore sought to determine whether supplementing the osteoblast cultures with mOPN could attenuate the increased mineralization phenotype. Upon treatment with mOPN, the Opn − / − cultures displayed a significant inhibition of mineralization (Figures 3A–3C). On the basis of both Alizarin Red S stain quantification and calcium assays, the amount of mineral detected in the mOPN-treated Opn − / − osteoblast cultures appears to decrease with increasing (5–20 μg/ml) mOPN concentration. We observed no change in metabolic activity (Figure 3D), neither were there any changes in terminal gene or protein expression (Figure 4) upon treatment with mOPN. This suggests that mOPN treatment inhibits mineralization independent of mediating cellular responses with the osteoblasts in our system.

Treatment with OPN-derived peptides

OPN has multiple sequences within the molecule that have been shown to be potent inhibitors of mineral formation by  c The Authors Journal compilation  c 2014 Biochemical Society

in vitro and in silico studies [18]. To determine whether some of these peptides affect cell-mediated mineralization, osteoblastic cells from the Opn − / − mouse were cultured in the presence of 20 μg/ml OPAR and P3. As before, there was increased mineral deposition in the Opn − / − osteogenic cell cultures compared with the WT cell cultures. The addition of OPAR and P3 reduced the mineralization relative to the untreated Opn − / − cells (Figures 5A–5C). These peptides did not alter the metabolic activity characteristics (Figure 5D), neither did they change Ibsp or Bglap gene expression (Figures 6A and 6B). DISCUSSION

The Opn − / − mice were originally characterized as having no significant phenotype [12]. As techniques became more sensitive, and as the mice were challenged, phenotypic differences became apparent. As adults, the Opn − / − mice show increased mineralization of their trabecular bone [11]. This increase in sitespecific mineral density was suggested to be due to decreased

Osteopontin mediates mineralization and not osteogenic cell development in vitro

Figure 5

361

Opn − / − osteoblast biomineralization is attenuated upon supplementation with 20 μg/ml P3 and OPAR

Quantification of mineralization of cultures at day 28. Mineralization was assayed by measuring extracted (A) Alizarin Red S at A 415 and normalizing to WT levels or (B) by extraction and quantification of total calcium. (C) Representative images of Alizarin Red S stain of cells treated with OPN-derived peptides. (D) Metabolic activity of the cell cultures at day 28. *P < 0.05, n = 3, calcium content n = 5.

osteoclast activity [12]. However, other studies suggest that OPN plays a role in promoting osteogenic development. For example, it was recently proposed that OPN interaction with cell-surface integrins promotes osteogenesis of MSCs by virtue of inhibiting adipogenesis [30]. It also has been shown that loss of OPN results in aberrations of signalling pathways in osteogenic cells mediated by, for example, parathyroid hormone (PTH) [31] and alkaline phosphatase 2 (AKP2) [32]. These findings suggest that OPN is a regulator of bone homoeostasis. In order to better understand the role of OPN in osteoblast-mediated biomineralization, we characterized a time course of gene expression and mineralization in Opn − / − osteoblasts and confirmed our findings by the addition of exogenous OPN and OPN-derived peptides. Our studies demonstrate that Opn − / − osteoblasts mediate significantly more mineral deposition in vitro, whereas the apparent cell number, based on assessment of metabolic activity, and expression of markers for differentiated osteoblasts remained essentially unchanged with a couple of exceptions, as discussed below. This suggests that the increase in mineralization is likely to be due to growth of the mineral component and not an increase in cell numbers or enhanced differentiation of osteogenic cells.

The potential for OPN to alter the cell-maturation characteristics of osteoblasts is controversial. In a variety of cell types, including cancer cells, OPN’s interaction with cell-surface integrins or CD44 mediates cell signalling leading to specific responses [3,33]. As such, OPN is predicted to play a role in promoting osteoblast signalling [34,35], but, as discussed above, OPN has been shown to both promote [13] and inhibit [14] osteoblast differentiation. The contradictory results of these studies could be due to a multitude of factors, such as different cell types (primary rat osteoblasts compared with MC3T3-E1 pre-osteoblastic cells) or method of Opn gene delivery (adenovirus compared with transfection). The method of OPN delivery is important as OPN may be capable of initiating cell signalling through an intracellular mechanism [36,37]. In the present study, the time course of osteoblast marker expression showed no major overall changes in expression patterns other than Col1a1. A decrease in Col1a1 expression may be interpreted as fewer sites for early mineral formation and thus lower mineral deposition. Despite this, we still observed increased mineralization in the Opn − / − cells in the present study. It was of interest that the expression of Ank and Enpp1 in the Opn − / − cells,  c The Authors Journal compilation  c 2014 Biochemical Society

362

Figure 6

E. Holm and others

Supplementation of cultures with 20 μg/ml OPN-derived peptides does not change maturation characteristics of Opn − / − osteoblastic cells

qPCR of (A) Ibsp and (B) Bglap expression of the cultures treated with 20 μg/ml OPAR and P3 at day 28. (C) Immunoblots for OCN (osteocalcin) and BSP protein expression at week 4 with β-actin shown as the loading control. Quantification of the (D) BSP and (E) OCN bands relative to the β-actin loading control. qPCR n = 3, immunoblot n = 3.

except Enpp1 at day 28, did not show a difference when compared with WT cells. These genes encode transmembrane proteins that are thought to regulate biomineralization: Ank is responsible for the transport of pyrophosphate, a potent inhibitor of mineral formation, out of the cell [38,39]; PC1 (gene name Enpp1) is an enzyme that degrades NTPs to produce pyrophosphate [40– 42]. The observed increase in Enpp1 at day 28 may have been a response to the increased mineralization in these cultures, similar to the expression of Enpp1 by mineralizing cementoblasts in vitro [43]. In contrast with our studies, previous work has demonstrated that the osteoblasts derived from the Opn − / − mouse have enhanced expression of both Ank and Enpp1 [32], proposed to compensate for the loss of OPN’s mineral inhibitory activity. The discrepancies between expression of Ank and Enpp1 in the present study and that of Harmey et al. [32] may be due to differences in experimental procedures. Our system used stromalderived in contrast with calvarial-derived osteogenic cells used by Harmey et al. [32], thus differing in both cell lineage and stage  c The Authors Journal compilation  c 2014 Biochemical Society

of development when isolated. The mice from which the cells were isolated came from two independently produced Opn − / − mouse strains; ours from Liaw et al. [25], and that of Harmey et al. from Rittling et al. [12]. We also used 2 mM G2P, a more physiologically relevant concentration [44,45], compared with 5 mM G2P used by Harmey et al. [32]. It is of relevance that higher G2P concentration in the culture medium has been shown to increase inorganic phosphate, which can result in non-specific calcification [45,46] and promotion of osteoblast differentiation [47–49]. In the present study, the loss of OPN does not appear to alter the expression of the major osteogenic markers. To determine whether the hypermineralization phenotype of the Opn − / − cells could be rescued, we treated the cells with mOPN. The addition of mOPN resulted in levels of mineralization well below that of untreated Opn − / − osteogenic cell cultures. The unchanged levels of Ibsp and Bglap gene and protein expression suggest that mOPN treatment resulted in no change in osteoblast development. Furthermore, we observed no change in metabolic

Osteopontin mediates mineralization and not osteogenic cell development in vitro

activity upon treatment with mOPN, suggesting that changes in mineralization were not a reflection of changes in cell number. Previous work delineating the role of OPN in VSMCs has demonstrated a similar attenuation of mineralization upon addition of OPN [24]. Calcification in the VSMCs was induced using mineralizing medium (including 10 mM G2P), which was supplemented with 0.5–5 μg/ml OPN [24]. This treatment resulted in a decrease in mineralization similar to that shown in studies on MC-3T3-E1 cells [50] and our study on OPN − / − osteogenic cells. This confirms that in mineralizing cell culture systems, OPN is a potent inhibitor of mineralization. To further our understanding of OPN’s mode of mineral inhibition, we investigated OPN-derived peptides that were shown previously to inhibit the growth of HA [18] and calcium oxalate monohydrate crystals [51]. Upon supplementation with OPAR and the phosphorylated P3 peptide, mineral deposition in Opn − / − osteoblast cultures was reduced relative to the untreated Opn − / − cells, to a level comparable with that of the WT osteoblast cultures. Furthermore, the expression of mature osteoblast markers was unchanged by the treatments. These findings are in agreement with the effects of the OPN-ASARM peptide [human OPN-(115– 132)] on MC-3T3-E1 cell differentiation and mineralization [52]. Thus the two OPN-derived peptides, P3 and OPAR, were capable of rescuing the hypermineralization phenotype of the Opn − / − osteoblast cultures, with no change in terminal differentiation of the Opn − / − cells. In the present study, we have demonstrated that Opn − / − osteoblasts show greater biomineralization than WT cells in vitro and that the hypermineralization is rescued by the addition of mOPN and OPN-derived peptides. Since there was little change in osteoblast marker expression, we can conclude that OPN mediates its inhibitory effects in the extracellular matrix by affecting the nucleation and/or growth of mineral crystals. Lastly, the rescue of the hypermineralization phenotype by the inhibitory OPN-derived peptides, without any apparent effect on osteogenic cell differentiation, provides insight into a possible avenue for the development of clinical targets to prevent pathological calcifications.

AUTHOR CONTRIBUTION Erik Holm designed and performed the experiments and wrote and edited the paper before submission. Jared Gleberzon performed the peptide-treated experiments and edited the paper before submission. Yinyin Liao performed the calcium extraction and quantification experiments. Esben Sørensen provided the mOPN and provided a critical review of the paper before submission. Frank Beier provided experimental design critique and edited the paper before submission. Graeme Hunter provided the OPN-derived peptides P3 and OPAR and edited the paper before submission. Harvey Goldberg critiqued the experimental design and co-wrote and edited the paper before submission.

ACKNOWLEDGEMENTS We thank Dr Martha Somerman and Dr Brian Foster for their critical review of the paper before submission.

FUNDING This study was supported by the Canadian Institutes of Health Research [grant numbers MOP 93598 and MOP 68827].

REFERENCES 1 O’Brien, K. D., Kuusisto, J., Reichenbach, D. D., Ferguson, M., Giachelli, C., Alpers, C. E. and Otto, C. M. (1995) Osteopontin is expressed in human aortic valvular lesions. Circulation 92, 2163–2168 CrossRef PubMed 2 McKee, M. D., Nanci, A. and Khan, S. R. (1995) Ultrastructural immunodetection of osteopontin and osteocalcin as major matrix components of renal calculi. J. Bone Miner. Res. 10, 1913–1929 CrossRef PubMed

363

3 Ganss, B. and Bansal, A. K. (2012) Osteopontin: a plurifunctional molecule. In Phosphorylated Extracellular Matrix Proteins of Bone and Dentin (Goldberg, M., ed.), pp. 283–318, Bentham Science Publishers, Paris 4 Fisher, L. W., Torchia, D. A., Fohr, B., Young, M. F. and Fedarko, N. S. (2001) Flexible structures of SIBLING proteins, bone sialoprotein and osteopontin. Biochem. Biophys. Res. Commun. 280, 460–465 CrossRef PubMed 5 Wuttke, M., Muller, S., Nitsche, D. P., Paulsson, M., Hanisch, F. G. and Maurer, P. (2001) Structural characterization of human recombinant and bone-derived bone sialoprotein: functional implications for cell attachment and hydroxyapatite binding. J. Biol. Chem. 276, 36839–36848 CrossRef PubMed 6 Tye, C. E., Rattray, K. R., Warner, K. J., Gordon, J. A., Sodek, J., Hunter, G. K. and Goldberg, H. A. (2003) Delineation of the hydroxyapatite-nucleating domains of bone sialoprotein. J. Biol. Chem. 278, 7949–7955 CrossRef PubMed 7 Cowles, E. A., DeRome, M. E., Pastizzo, G., Brailey, L. L. and Gronowicz, G. A. (1998) Mineralization and the expression of matrix proteins during in vivo bone development. Calcif. Tissue Int. 62, 74–82 CrossRef PubMed 8 Chen, J., McKee, M. D., Nanci, A. and Sodek, J. (1994) Bone sialoprotein mRNA expression and ultrastructural localization in fetal porcine calvarial bone: comparisons with osteopontin. Histochem. J. 26, 67–78 CrossRef PubMed 9 Chen, J., Singh, K., Mukherjee, B. B. and Sodek, J. (1993) Developmental expression of osteopontin (OPN) mRNA in rat tissues: evidence for a role for OPN in bone formation and resorption. Matrix 13, 113–123 CrossRef PubMed 10 Bautista, D. S., Denstedt, J., Chambers, A. F. and Harris, J. F. (1996) Low-molecular-weight variants of osteopontin generated by serine proteinases in urine of patients with kidney stones. J. Cell. Biochem. 61, 402–409 CrossRef PubMed 11 Boskey, A. L., Spevak, L., Paschalis, E., Doty, S. B. and McKee, M. D. (2002) Osteopontin deficiency increases mineral content and mineral crystallinity in mouse bone. Calcif. Tissue Int. 71, 145–154 CrossRef PubMed 12 Rittling, S. R., Matsumoto, H. N., McKee, M. D., Nanci, A., An, X. R., Novick, K. E., Kowalski, A. J., Noda, M. and Denhardt, D. T. (1998) Mice lacking osteopontin show normal development and bone structure but display altered osteoclast formation in vitro . J. Bone Miner. Res. 13, 1101–1111 CrossRef PubMed 13 Kojima, H., Uede, T. and Uemura, T. (2004) In vitro and in vivo effects of the overexpression of osteopontin on osteoblast differentiation using a recombinant adenoviral vector. J. Biochem. 136, 377–386 CrossRef PubMed 14 Huang, W., Carlsen, B., Rudkin, G., Berry, M., Ishida, K., Yamaguchi, D. T. and Miller, T. A. (2004) Osteopontin is a negative regulator of proliferation and differentiation in MC3T3-E1 pre-osteoblastic cells. Bone 34, 799–808 CrossRef PubMed 15 Boskey, A. L., Maresca, M., Ullrich, W., Doty, S. B., Butler, W. T. and Prince, C. W. (1993) Osteopontin–hydroxyapatite interactions in vitro : inhibition of hydroxyapatite formation and growth in a gelatin-gel. Bone Miner. 22, 147–159 CrossRef PubMed 16 Hunter, G. K., Kyle, C. L. and Goldberg, H. A. (1994) Modulation of crystal formation by bone phosphoproteins: structural specificity of the osteopontin-mediated inhibition of hydroxyapatite formation. Biochem. J. 300, 723–728 PubMed 17 Goldberg, H. A. and Hunter, G. K. (1995) The inhibitory activity of osteopontin on hydroxyapatite formation in vitro . Ann. N.Y. Acad. Sci. 760, 305–308 CrossRef PubMed 18 Azzopardi, P. V., O’Young, J., Lajoie, G., Karttunen, M., Goldberg, H. A. and Hunter, G. K. (2010) Roles of electrostatics and conformation in protein–crystal interactions. PLoS ONE 5, e9330 CrossRef PubMed 19 Hunter, G. K., Grohe, B., Jeffrey, S., O’Young, J., Sørensen, E. S. and Goldberg, H. A. (2009) Role of phosphate groups in inhibition of calcium oxalate crystal growth by osteopontin. Cells Tissues Organs 189, 44–50 CrossRef PubMed 20 Sørensen, E. S., Højrup, P. and Petersen, T. E. (1995) Posttranslational modifications of bovine osteopontin: identification of twenty-eight phosphorylation and three O-glycosylation sites. Protein Sci. 4, 2040–2049 CrossRef PubMed 21 Kumura, H., Minato, N. and Shimazaki, K. (2006) Inhibitory activity of bovine milk osteopontin and its fragments on the formation of calcium phosphate precipitates. J. Dairy Res. 73, 449–453 CrossRef PubMed 22 Giachelli, C. M., Bae, N., Almeida, M., Denhardt, D. T., Alpers, C. E. and Schwartz, S. M. (1993) Osteopontin is elevated during neointima formation in rat arteries and is a novel component of human atherosclerotic plaques. J. Clin. Invest. 92, 1686–1696 CrossRef PubMed 23 Speer, M. Y., Chien, Y. C., Quan, M., Yang, H. Y., Vali, H., McKee, M. D. and Giachelli, C. M. (2005) Smooth muscle cells deficient in osteopontin have enhanced susceptibility to calcification in vitro . Cardiovasc. Res. 66, 324–333 CrossRef PubMed 24 Wada, T., McKee, M. D., Steitz, S. and Giachelli, C. M. (1999) Calcification of vascular smooth muscle cell cultures: inhibition by osteopontin. Circ. Res. 84, 166–178 CrossRef PubMed 25 Liaw, L., Birk, D. E., Ballas, C. B., Whitsitt, J. S., Davidson, J. M. and Hogan, B. L. M. (1998) Altered wound healing in mice lacking a functional osteopontin gene (spp1). J. Clin. Invest. 101, 1468–1478 CrossRef PubMed  c The Authors Journal compilation  c 2014 Biochemical Society

364

E. Holm and others

26 Sørensen, E. S. and Petersen, T. E. (1993) Purification and characterization of three proteins isolated from the proteose peptone fraction of bovine milk. J. Dairy Res. 60, 189–197 CrossRef PubMed 27 Malaval, L., Wade-Gueye, N. M., Boudiffa, M., Fei, J., Zirngibl, R., Chen, F., Laroche, N., Roux, J. P., Burt-Pichat, B., Duboeuf, F. et al. (2008) Bone sialoprotein plays a functional role in bone formation and osteoclastogenesis. J. Exp. Med. 205, 1145–1153 CrossRef PubMed 28 Nakayama, G. R., Caton, M. C., Nova, M. P. and Parandoosh, Z. (1997) Assessment of the Alamar Blue assay for cellular growth and viability in vitro . J. Immunol. Methods 204, 205–208 CrossRef PubMed 29 Voytik-Harbin, S. L., Brightman, A. O., Waisner, B., Lamar, C. H. and Badylak, S. F. (1998) Application and evaluation of the alamarBlue assay for cell growth and survival of fibroblasts. In Vitro Cell. Dev. Biol. Anim. 34, 239–246 CrossRef 30 Chen, Q., Shou, P., Zhang, L., Xu, C., Zheng, C., Han, Y., Li, W., Huang, Y., Zhang, X., Shao, C. et al. (2014) An osteopontin–integrin interaction plays a critical role in directing adipogenesis and osteogenesis by mesenchymal stem cells. Stem Cells 32, 327–337 CrossRef PubMed 31 Ono, N., Nakashima, K., Rittling, S. R., Schipani, E., Hayata, T., Soma, K., Denhardt, D. T., Kronenberg, H. M., Ezura, Y. and Noda, M. (2008) Osteopontin negatively regulates parathyroid hormone receptor signaling in osteoblasts. J. Biol. Chem. 283, 19400–19409 CrossRef PubMed 32 Harmey, D., Johnson, K. A., Zelken, J., Camacho, N. P., Hoylaerts, M. F., Noda, M., Terkeltaub, R. and Millan, J. L. (2006) Elevated skeletal osteopontin levels contribute to the hypophosphatasia phenotype in Akp2 − / − mice. J. Bone Miner. Res. 21, 1377–1386 CrossRef PubMed 33 Bellahcene, A., Castronovo, V., Ogbureke, K. U., Fisher, L. W. and Fedarko, N. S. (2008) Small integrin-binding ligand N-linked glycoproteins (SIBLINGs): multifunctional proteins in cancer. Nat. Rev. Cancer 8, 212–226 CrossRef PubMed 34 Liu, Y. K., Uemura, T., Nemoto, A., Yabe, T., Fujii, N., Ushida, T. and Tateishi, T. (1997) Osteopontin involvement in integrin-mediated cell signaling and regulation of expression of alkaline phosphatase during early differentiation of UMR cells. FEBS Lett. 420, 112–116 CrossRef PubMed 35 Yabe, T., Nemoto, A. and Uemura, T. (1997) Recognition of osteopontin by rat bone marrow derived osteoblastic primary cells. Biosci. Biotechnol. Biochem. 61, 754–756 CrossRef PubMed 36 Zohar, R., Cheifetz, S., McCulloch, C. A. G. and Sodek, J. (1998) Analysis of intracellular osteopontin as a marker of osteoblastic cell differentiation and mesenchymal cell migration. Eur. J. Oral Sci. 106, 401–407 PubMed 37 Zohar, R., Suzuki, N., Suzuki, K., Arora, P., Glogauer, M., McCulloch, C. A. and Sodek, J. (2000) Intracellular osteopontin is an integral component of the CD44–ERM complex involved in cell migration. J. Cell. Physiol. 184, 118–130 CrossRef PubMed 38 Hakim, F. T., Cranley, R., Brown, K. S., Eanes, E. D., Harne, L. and Oppenheim, J. J. (1984) Hereditary joint disorder in progressive ankylosis (ank /ank ) mice. I. Association of calcium hydroxyapatite deposition with inflammatory arthropathy. Arthritis Rheum. 27, 1411–1420 CrossRef PubMed Received 3 June 2014/29 August 2014; accepted 13 October 2014 Published as BJ Immediate Publication 13 October 2014, doi:10.1042/BJ20140702

 c The Authors Journal compilation  c 2014 Biochemical Society

39 Ho, A. M., Johnson, M. D. and Kingsley, D. M. (2000) Role of the mouse ank gene in control of tissue calcification and arthritis. Science 289, 265–270 CrossRef PubMed 40 Terkeltaub, R., Rosenbach, M., Fong, F. and Goding, J. (1994) Causal link between nucleotide pyrophosphohydrolase overactivity and increased intracellular inorganic pyrophosphate generation demonstrated by transfection of cultured fibroblasts and osteoblasts with plasma cell membrane glycoprotein-1: relevance to calcium pyrophosphate dihydrate deposition disease. Arthritis Rheum. 37, 934–941 CrossRef PubMed 41 Johnson, K., Vaingankar, S., Chen, Y., Moffa, A., Goldring, M. B., Sano, K., Jin-Hua, P., Sali, A., Goding, J. and Terkeltaub, R. (1999) Differential mechanisms of inorganic pyrophosphate production by plasma cell membrane glycoprotein-1 and B10 in chondrocytes. Arthritis Rheum. 42, 1986–1997 CrossRef PubMed 42 Johnson, K., Moffa, A., Chen, Y., Pritzker, K., Goding, J. and Terkeltaub, R. (1999) Matrix vesicle plasma cell membrane glycoprotein-1 regulates mineralization by murine osteoblastic MC3T3 cells. J. Bone Miner. Res. 14, 883–892 CrossRef PubMed 43 Foster, B. L., Nagatomo, K. J., Nociti, Jr, F. H., Fong, H., Dunn, D., Tran, A. B., Wang, W., Narisawa, S., Millan, J. L. and Somerman, M. J. (2012) Central role of pyrophosphate in acellular cementum formation. PLoS ONE 7, e38393 CrossRef PubMed 44 Eidelman, N., Chow, L. C. and Brown, W. E. (1987) Calcium phosphate saturation levels in ultrafiltered serum. Calcif. Tissue Int. 40, 71–78 CrossRef PubMed 45 Khouja, H. I., Bevington, A., Kemp, G. J. and Russell, R. G. (1990) Calcium and orthophosphate deposits in vitro do not imply osteoblast-mediated mineralization: mineralization by betaglycerophosphate in the absence of osteoblasts. Bone 11, 385–391 CrossRef PubMed 46 Bellows, C. G., Heersche, J. N. and Aubin, J. E. (1992) Inorganic phosphate added exogenously or released from β-glycerophosphate initiates mineralization of osteoid nodules in vitro . Bone Miner. 17, 15–29 CrossRef PubMed 47 Usui, Y., Uematsu, T., Uchihashi, T., Takahashi, M., Takahashi, M., Ishizuka, M., Doto, R., Tanaka, H., Komazaki, Y., Osawa, M. et al. (2010) Inorganic polyphosphate induces osteoblastic differentiation. J. Dent. Res. 89, 504–509 CrossRef PubMed 48 Beck, Jr, G. R. (2003) Inorganic phosphate as a signaling molecule in osteoblast differentiation. J. Cell. Biochem. 90, 234–243 CrossRef PubMed 49 Beck, Jr, G. R., Moran, E. and Knecht, N. (2003) Inorganic phosphate regulates multiple genes during osteoblast differentiation, including Nrf2. Exp. Cell Res. 288, 288–300 CrossRef PubMed 50 Addison, W. N., Azari, F., Sørensen, E. S., Kaartinen, M. T. and McKee, M. D. (2007) Pyrophosphate inhibits mineralization of osteoblast cultures by binding to mineral, up-regulating osteopontin, and inhibiting alkaline phosphatase activity. J. Biol. Chem. 282, 15872–15883 CrossRef PubMed 51 Grohe, B., O’Young, J., Ionescu, D. A., Lajoie, G., Rogers, K. A., Karttunen, M., Goldberg, H. A. and Hunter, G. K. (2007) Control of calcium oxalate crystal growth by face-specific adsorption of an osteopontin phosphopeptide. J. Am. Chem. Soc. 129, 14946–14951 CrossRef PubMed 52 Addison, W. N., Masica, D. L., Gray, J. J. and McKee, M.D. (2010) Phosphorylation-dependent inhibition of mineralization by osteopontin ASARM peptides is regulated by PHEX cleavage. J. Bone Miner. Res. 25, 695–705 CrossRef PubMed

Copyright of Biochemical Journal is the property of Portland Press Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.