Establishing an Immortalized Human Osteoprecursor Cell Line: OPC1

Authors


Abstract

The present studies evaluated the feasibility of establishing a conditionally immortalized osteoprecursor cell line derived from human fetal bone tissue. Primary cultures were transfected with a plasmid in which the Mx-1 promoter drives the expression of SV40 T-antigen when activated by human A/D interferon. Several neomycin (G418)-resistant colonies were characterized for cell growth and alkaline phosphatase (ALP) enzyme activity. The clone, designated OPC1 (osteoblastic precursor cell line 1), which exhibited the highest ALP enzyme activity at passage 10 (P10), was selected for additional osteogenic phenotypic characterization. Reverse transcription-polymerase chain reaction (RT-PCR) phenotyping revealed abundant mRNA for osteocalcin (OC), osteonectin (ON), osteopontin (OP), parathyroid hormone receptor (PTHr), ALP, and procollagen type I (ProI). In addition, the levels of quantitative RT-PCR product of ON, OP, PTHr, and ProI mRNAs exhibited a marked up-regulation when maintained in medium containing an osteogenic supplement (OS). The ability to stimulate osteogenic differentiation was characterized in postconfluent OPC1 cells maintained in tissue culture medium supplemented with recombinant human bone morphogenetic protein-2 (rhBMP-2) either with or without an OS. All treatment groups exhibited a striking up-regulation of ALP enzyme activity that coincided with ALP histochemical observations. Postconfluent cells also exhibited the ability to form mineralized nodules under all treatments (confirmed by von Kossa histochemical staining and calcium deposition). An enzyme immunosorbent assay (EIA) was utilized to measure intact human OC from the OPC1 line under the various treatments. Abundant OC was evident in the tissue culture medium indicating de novo sythesis and release from the OPC1 line under appropriate conditions. The clonal human-derived OPC1 line represents a homogeneous osteogenic cell line that not only has maintained a consistent bone phenotype from P10 to at least P30, but has also exhibited the capacity to generate programmed differentiation in the presence of low dose rhBMP-2 (10 ng/ml). Thus, the OPC1 line is a human-derived osteoprecursor that provides a sensitive in vitro cell culture system to evaluate bone development, cell/biomaterial interactions, and may be a useful screen for putative bone differentiating factors.

INTRODUCTION

Bone regenerative therapeutics under clinical development, as well as animal models such as the critical-sized defects, have utilized biomaterials as scaffolds to provide an anchor for the attachment and differentiation of endogenous migratory precursor cells (reviewed in Winn et al. and Hollinger et al.(1,2)). Recently, additional efforts have combined the scaffold philosophy with potent morphogenetic factors such as recombinant human bone morphogenetic protein-2 (rhBMP-2) to act as a chemoattractant and differentiating factor for endogenous precursor cells to take residence in the regenerative environment.(1–7) The expanding field of tissue engineering is becoming prominent in bone regeneration and utilizes combinations of a biomaterial scaffold with either an osteoblast or precursor cell source and generally BMPs prior to implantation.(8,9) Although many osteoblastic and precursor cell lines are available,(1–2,10–19) the majority have liabilities limiting biomaterial/human bone cell interactions, as well as being deficient for clinical applications. Furthermore, repeated isolations of primary human osteoblasts from multiple donors have been reported to result in cellular preparations with variable osteogenic potential.(20) Therefore, generating a human-derived clonal osteoprecursor cell line would provide a consistent and reproducible culture system for direct comparisons of bone-graft substitutes. In addition, generating a history of a cell line from the isolation through the genetic manipulation to the continued characterization of the stability and safety of a transgene are important considerations in the development of human-derived cell lines. Consequently, the reported study was driven by these issues.

The SV40 oncogenes, small t-antigen (tag) and large T-antigen (Tag), are nuclear phosphoproteins that transform a broad range of cell types.(21–23) In the present study, the pMx-1-SV40Tag-Neo-195 plasmid DNA was utilized in standard transfection protocols to generate a conditional immortalized cell line.(24,25) The Mx-1 promoter has been previously shown to be highly inducible by interferon to direct the expression of SV40 Tag.(25) In these studies, glial cell lines derived from Mx-1-SV40Tag transgenic mice exhibited reversible induction of oncogenic expression.(25) Cells with the pMx-1-SV40Tag-Neo-195 DNA will exhibit increased proliferation by driving the SV40 Tag in the presence of human A/D interferon and will diminish their mitotic activity when interferon is removed. Several laboratories have investigated the establishment of bone cell lines transfected with a gene constituitively expressing the SV40 Tag(15,23) while others have utilized a gene coding for a temperature-sensitive mutant, tsA58, of SV40 Tag which conditionally immortalizes the human fetal osteoblastic cell line under permissive conditions.(17)

A number of concerns have arisen regarding the safety and reliability of transformed cell lines expressing the SV40 Tag, such as phenotypic variabilities; nevertheless, it remains as one of the most effective methods of either constituitively or conditionally immortalizing cell lines.

To evaluate cell/material interactions in a culture system that would be pertinent to the clinical situation, developing and utilizing human cell lines would facilitate understanding of the cellular and molecular events that occur with bone substitutes. Furthermore, advances in molecular biology and gene therapy warrant investigation for engineering and developing human cell lines as a clinical tissue engineered product, especially if the technology can provide a universal donor cell. The present study describes the establishment of a human fetal osteoblastic precursor cell line (OPC1) that has been characterized for maintenance of a stable phenotype and continued transgene expression (i.e., SV40 Tag expression). In addition, the influence of exogenously applied rhBMP-2 on the expression of the OPC1 osteogenic phenotype is described.

MATERIALS AND METHODS

Materials

All reagents were purchased from Sigma Chemical Co. (St. Louis, MO, U.S.A.) or GIBCO BRL Inc. (Grand Island, NY, U.S.A.) unless otherwise noted. Falcon tissue culture plasticware was obtained from Becton Dickinson and Co. (Franklin Lakes, NJ, U.S.A.). A QuickPrep Micro mRNA Purification Kit was purchased from Pharmacia Biotech, Inc. (Piscataway, NJ, U.S.A.), and an Access reverse transcriptase-polymerase chain reaction (RT-PCR) system was purchased from Promega, Inc. (Madison, WI, U.S.A.). PCR plasticware was purchased from Perkin Elmer, Inc. (Norwalk, CT, U.S.A.). NuSieve 3:1 agarose was obtained from FMC BioProducts (Rockland, ME, U.S.A.). rhBMP-2 was kindly provided by Genetics Institute, Inc. (Andover, MA) and the methods of production and purification have been previously described.(26) The recombinant human A/D interferon is an alpha hybrid constructed from the recombinant human interferons Hu-IFN-αA and Hu-IFN-αD and was purchased from PBL (West Caldwell, NJ, U.S.A.).

Isolation and transfection of OPCs

Fetal tissue of gestational age of ∼12–13 weeks was obtained under institutionally approved protocols. The tissue was maintained in Earle's Balanced Salt Solution (EBSS) supplemented with 10 mM HEPES, pH 7.4, and transported to the laboratory for dissection and isolation. Osteoblasts were derived from fetal human periosteum and femur utilizing a repeated digestion technique involving 0.3% collagenase P (Boeringer-Mannheim, Indianapolis, IN, U.S.A.) and 0.25% trypsin.(27) Cells were plated at 2.5 × 105 in 75-cm2 tissue culture flasks in alpha modified essential medium (α-MEM) with 5% fetal bovine serum (FBS). The remaining tissue pieces were collected, washed with calcium and magnesium-free EBSS, and digested with 0.25% trypsin-EDTA for 30 minutes. These cells were also plated at 2.5 × 105 in 75-cm2 tissue culture flasks in α-MEM with 5% FBS. These cultures represented the initial isolation, P0, and once confluent, the cells were subcultured after enzymatic removal with 0.25% trypsin-EDTA to passage 1 (P1).

The early passage bone cells were maintained and expanded to P3, at which time they were transfected by a standard calcium phosphate-mediated methodology (Stratagene, La Jolla, CA, U.S.A.) to incoporate 10 μg of double-CsCl purified pMx-1-SV40T-antigen-Neo-195 plasmid DNA into the host cell genome (Fig. 1). The plasmid pMx-1-SV40T-Neo-195 was constructed as follows. A 2.3-kb mouse Mx-1 promoter was fused to a 2.1-kb SV40 large T fragment in the cloning vector pSP65. A 1.9-kb mouse beta globin 3′ untranslated region (3′UTR) was introduced into the plasmid at the BamHI and XbaI sites; the resulting plasmid was designated pMx-1-SV40T. A 1518-bp HincII–XmnI fragment containing the neomycin phosphotransferase gene driven by the SV40 promoter was isolated from pcDNA3 (InVitrogen, San Diego, CA, U.S.A.), subcloned into pMx-1-SV40T digested with EcoRI, and filled in by Klenow generating the pMx-1-SV40T antigen-Neo-195 plasmid DNA. The P3 bone cells were transfected overnight in a mitogenic serum-free defined medium UltraCULTURE (BioWhittaker, Inc., Walkersville, MD, U.S.A.) containing no antibiotics. Following the transfection protocol, plates were rinsed with medium and placed into fresh α-MEM/5% FBS overnight. The following day, the transfected cells were selected in α-MEM/5% FBS medium supplemented with 0.5 mg/ml G418-sulfate (neomycin analog). The G418 allows the selection of stable transfectants that have incorporated the gene conferring resistance to neomycin toxicity. After a selection period of 10–14 days, the medium was changed to the α-MEM/5% FBS supplemented with 750 U/ml human A/D interferon and 0.2 mg/ml G418. Clonal lines were obtained by a standard limiting dilution protocol of the polyclonal transfectants, and preference was determined by the clonal cell's morphology, growth rate of ∼3.5–4 doublings/week, and for their expression of alkaline phosphatase (ALP). Mock transfected cells served as a control for the selection protocol.

Figure FIG. 1.

The pMx-1-SV40T-Neo-195 plasmid DNA. A 2.3-kb mouse Mx-1 promoter was fused to a 2.1-kb SV40 large T fragment in the cloning vector pSP65. A 1.9-kb mouse beta globin 3′ untranslated region (3′UTR) was introduced into the plasmid at the BamHI and XbaI sites. A 1.52-kb HincII–XmnI fragment containing the neomycin phosphotransferase driven by the SV40 promoter was isolated from pcDNA3, subcloned into pMx-1-SV40T, digested with EcoRI, and filled with a Klenow sequence. SV40, simian virus 40.

Cell safety testing

Cryovials containing 5 × 106 cells of the OPC line maintained in antibiotic and antimycotic-free tissue culture medium for at least three passages were packaged and sent to ViroMED Laboratories (Minneapolis, MN, U.S.A.) to test for the presence of Cytomegalovirus, Hepatitis B, Hepatitis C, HIV-1, HIV-2, and HTLV I/II. Conditioned medium was also collected and shipped to Microbiological Associates, Inc. (Rockville, MD, U.S.A.) for assessing sterility. Last, mycoplasma detection was performed utilizing the CELLshipper kit as described by BIONIQUE Testing Laboratories, Inc. (Saranac Lake, NY, U.S.A.) at P4 and later at P20.

Cell proliferation assay

The OPC1 line was evaluated for growth kinetics while maintained in the presence of four types of tissue culture medium. The media consisted of: a serum-free defined medium, CM; CM containing 750 U/ml recombinant human A/D interferon (CM IFN); the base medium of α-MEM/5% FBS, BM; and BM supplemented with 750 U/ml recombinant human A/D interferon (BM IFN). OPCs were initially seeded at 2.5 × 104 cells/well in a 12-well plate (Falcon Labware), maintained in their respective media and a feeding schedule of every other day was followed. At 4, 7, and 10 days following the cell seeding, cell proliferation was measured in duplicate with a modified crystal violet dye-binding assay.(28) Cultures were rinsed with Tyrode's balanced salt solution and fixed for 15 minutes in 1% (v/v) buffered glutaraldehyde. The fixed cells were rinsed twice with distilled water and air dried. The dried cultures were stained for 30 minutes with 0.1% crystal violet (w/v) in distilled water. The crystal violet was extracted from the cells by a 4-h incubation at room temperature in 1% Triton X-100. Triton extracts were measured at 600 nm on a microplate reader (MRX; Dynatech Labs., Chantilly, VA, U.S.A.). Absorbance values were converted into cell numbers extrapolated from established standard curves.

Evaluating osteogenic phenotypes

Alkaline phosphatase:

Confirmation of an osteoblast/preosteoblast phenotype was performed at various passages (P10, P20, P30). The levels of ALP enzyme activity was quantitatively measured by the method of Lowry et al.(29) in cultures at days 4, 9, and 16 following an initial seeding of 2.5 × 104 cells/well in a 6-well plate (Falcon Labware). The wells with the OPCs contained the base medium of α-MEM/5% FBS overnight after their initial seeding. On the following day, the base medium provided a negative control (group 1) and the additional groups included: (group 2) base medium supplemented with 10 ng/ml of rhBMP-2; (group 3) base medium supplemented with 50 ng/ml of rhBMP-2; (group 4) base medium supplemented with 100 ng/ml of rhBMP-2; (group 5) base medium supplemented with the osteogenic supplement (OS) 10 mM β-glycerophosphate, 10−7 M Dexamethasone, 50 μg/ml of ascorbic acid phosphate (Wako Chemical, Osaka, Japan); and (group 6) base medium supplemented with OS + 50 ng/ml rhBMP-2. ALP activity was measured in triplicate cultures after rinsing the wells with calcium and magnesium-free EBSS, collecting the cells by scraping and incubating 5 × 104 cells/well in a 96-well plate with 5 mM p-nitrophenyl phosphate in 50 mM glycine, 1 mM MgCl2 at 37°C for 5–20 minutes. Enzyme activity was calculated after measuring the absorbance of p-nitrophenol product formed at 405 nm on a microplate reader (MRX; Dynatech Labs.) and compared with serially diluted standards. Enzyme activity is expressed as nanograms of p-nitrophenol/minute/50,000 cells. In addition, ALP histochemistry was performed on cultures at P30 according to standard protocols described in Sigma Kit 85. However, in this series of experiments, group 4 from above (base medium supplemented with 100 ng/ml of rhBMP-2), was replaced with base medium supplemented with 50 ng/ml of basic fibroblast growth factor (bFGF), to evaluate mitogenicity and/or the ability to influence programmed osteogenic differentiation.

Extracellular calcium:

The OPC1 line was evaluated for the ability to mineralize the extracellular matrix, which is produced 7–10 days following confluence in the base medium, but especially following maintenance in the base medium containing OS. The extracellular calcium content was measured quantitatively by first rinsing the cell layers twice with phosphate-buffered saline (PBS), removing the cell layers with a cell scraper, collecting in 1.5 ml eppendorf tubes, and exposing the cells to 0.1 N HCl. The calcium was extracted from the cells by shaking for 4 h at 4°C, collecting the cells by centrifugation, and the supernatant was used for calcium determination according to the manufacturer's protocol in Sigma Kit 587. Absorbance of samples was measured on the multiplate reader (MRX, Dynatech Labs) at 570 nm 5–10 minutes after the addition of reagents. Total calcium was calculated from standard solutions prepared in parallel and expressed as micrograms per well.

A histochemical analysis of mineralization was evaluated utilizing a staining procedure of von Kossa. In brief, postconfluent cells were fixed in 1% (w/v) paraformaldehyde in PBS (pH 7.4) for 1 h, rinsed with PBS and treated with 5% (w/v) silver nitrate in the dark for 15 minutes. Thereafter, the cells were rinsed thoroughly with distilled water, subjected to ultraviolet light for 5–7 minutes, and visualized with light microscopy.

Osteocalcin enzyme immunosorbent assay:

The osteocalcin (OC) level in conditioned medium was determined by an enzyme immunosorbent assay (EIA) kit for intact OC (Biomedical Technologies, Inc., Stoughton, MA, U.S.A.). Pre- and postconfluent OPC1 cells in 6-well multiwell plates (characterized in triplicate) were maintained in 1 ml of the serum-free medium UltraCULTURE with the various supplements included as previously described in the measurement of ALP. Samples were collected after 48 h, and a 50-μl sample was innoculated onto the microtiter plate and assayed according to the manufacturer's protocol. Data were expressed as nanograms per milliliter, and the limit of detection for the EIA was 0.1 ng/ml.

RT-PCR phenotyping:

Established RT-PCR qualitative analysis techniques were used to characterize (PCR phenotyping) the presence of OC, osteonectin (ON), osteopontin (OP), parathyroid hormone receptor (PTHr), ALP, and procollagen type I (ProI)(17,30) transcripts. The oligonucleotide RT-PCR primer sequences are listed in Table 1 and were purchased from GIBCO BRL. Briefly, the procedure involved 3–5 × 106 OPCs pelleted at 12,000 rpm in a microfuge (Eppendorf 5412; Brinkman Instruments, Inc., Westbury, NY, U.S.A.) and either used immediately or frozen at –80°C for storage. The mRNA was isolated using the QuickPrep Micro mRNA purification kit (Pharmacia Biotech, Inc.) according to the manufacturer's specifications resulting in a 200-μl final mRNA elution volume. The mRNA concentration was determined by spectrophotometric absorbance using A260 × 40 μg/μl, and mRNA concentrations of the elution volumes were determined to be in the range of 50–160 μg/μl. The cDNA was synthesized from 50 μg of the mRNA according to the Access RT-PCR System (Promega, Inc.). Aliquots of the total cDNA were amplified in each PCR with 2.5 U of Taq polymerase (Promega, Inc.), and amplifications were performed in a GeneAmp 2400 thermal cycler (Perkin-Elmer, Inc., Foster City, CA, U.S.A.) for 30 cycles after an initial 30 s denaturation at 94°C, annealed for 2 minutes at 55°C, and extended for 2 minutes at 72°C. The amplification reaction products were resolved on 2.5% NuSieve agarose/TBE gels (FMC BioProducts), electrophoresed at 85 mV for 90 minutes, and visualized by ethidium bromide. Base ladders of 50 bp and 100 bp (Boehringer Mannheim, Inc.) were included as standards.

Table Table 1. Oligonuclueotide Primer Sequences Utilized in the RT-PCR
original image

Quantitative RT-PCR was performed as follows. Heated total RNA (5 μg in 10.5 μl total volume in ultrapure H2O at 65°C for 5 minutes) was added to tubes containing 4 μl 5× MMLV-RT buffer (GIBCO), 2 μl dNTPs, 2 μl dT17 primer (10 pmol/ml), 0.5 μl RNAsin (40 U/ml). Samples were incubated at 37°C for 1 h, the enzyme inactivated at 95°C for 5 minutes with 80 μl ultrapure H2O added. Transcribed samples (5 μl) were subjected to PCR as described. Briefly, 50-μl sample volumes were added to tubes containing water and appropriate amounts of PCR buffer, 25 mM MgCl2, dNTPs, the forward and reverse primers for GAP (internal control), ON, OP, PTHr, and ProI, [32P]dCTP, and Amplitaq (Perkin-Elmer). Primers were standardized to run consistently at 22 cycles with the following conditions for denaturation, annealing, and extensions (GAP, 94°C, 30 s; 58°C, 30 s; 72°C, 20 s; ON, 94°C, 30 s; 64°C, 30 s; 72°C, 20 s; OP, 94°C, 30 s; 58°C, 30 s; 72°C, 20 s; PTHr, 94°C, 30 s; 64°C, 30 s; 72°C, 20 s; ProI, 94°C, 30 s; 53°C, 30 s; 72°C, 20 s).

For quantitation, RT-PCR products received 5 μl/tube of loading dye, were mixed, heated (65°C) for 10 minutes, centrifuged, and 10 μl each subjected to 12% polyacrylamide:bis gel electrophoresis (15 V/well; constant current) under standard conditions. Gels were incubated in gel preserving buffer (10% v/v glycerol, 7% v/v acetic acid, 40% v/v methanol, 43% deionized water) for 30 minutes, dried (80°C; BioRad GelDryer, BioRad Labs, Hercules, CA, U.S.A.) in vacuo for 1–2 h and developed in a Molecular Dynamics PhosphorImager cassette (Sunnyvale, CA, U.S.A.) for 6–24 h. Visualized bands were analyzed for SUM counts (above background) using the Molecular Dynamics analysis programs and plotted graphically with Microsoft Excel 97 Professional (Microsoft Corp., Seattle, WA, U.S.A.).

Determining T-antigen expression

Immunostaining for the presence of Tag expression was performed on the OPC1 line maintained in medium that drives the Mx-1 promoter for Tag, human A/D interferon, and in cultures without the stimulus. The Tag monoclonal antibody (PAB-419) was a generous gift from CytoTherapeutics, Inc. (Providence, RI, U.S.A.) and the staining protocol followed standard immunoperoxidase techniques utilizing a Vectastain Elite ABC Mouse Kit. The SCT-1/hNGF cell line was utilized as the positive control for the nuclear Tag.(31)

Statistical analysis

Data are presented in the text and in all relevant figures as means and 1 SD from the mean (mean ± SD) of samples characterized in triplicate. Data from the ALP enzyme activity, calcium content and OC production were subjected to an analysis of variance and post hoc multiple comparison tests with time and treatments as repeated measures. Statistical significance was established at p < 0.05.

RESULTS

Cell safety

The OPC1 line has been screened by ViroMED Laboratories to test for Cytomegalovirus, Hepatitis B, Hepatitis C, HIV-1, HIV-2, and HTLV I/II. These tests indicated the OPC1 line was negative for these viruses. Conditioned antibiotic-free medium was tested and the cell line was sterile and mycoplasma free.

Morphology and nontransfected cells

The osteoblast precursor cell preparations at the time of the initial plating (P0) established an adherent culture with cells exhibiting generally a polygonal, and infrequently a fusiform morphology, as well as expressing a faint birefringence surrounding the cells (Fig. 2A). Within 5–8 days of the P0 plating, the cells grew into a confluent monolayer with a morphology consistent with other osteogenic cell lines. There were no hyperconfluent, multilayered colonies of randomly oriented cells. Control, nontransfected cells were maintained and passaged in parallel with the transfected cells to determine the growth rates and limits to propagation. The growth rate of the normal human osteoblast-like cells was reduced as compared to the transfected clonal lines 1, 2, and 3, but did not differ from OPC clones 4–7 (Table 2). The nontransfected normal human osteoblast-like cells also exhibited a growth rate that significantly diminished after P20. Between P25 and P28, the nontransfected cells entered a crisis phase, became senescent and ceased propagation.

Table Table 2. Growth and Phenotypic Characteristics of Osteoprecursor Cell Types/Lines
original image
Figure FIG. 2.

The osteoblast precursor cell preparations at the time of the initial plating (P0) established an adherent culture with cells exhibiting generally a polygonal, and infrequently, a fusiform morphology, as well as expressing a faint birefringence surrounding the cells (A). The OPC1 clonal cells possessed an epithelial morphology when grown to a confluent monolayer (B). Positive immunostaining for nuclear large T-antigen (Tag) expression can be observed on the OPC1 line maintained in medium that apparently drives the promoter for Tag (C). Scale bar = 20 μm in (A), (C), and 40 μm in (B).

Transfection

The cells were subcultured until P3, after which they were transfected with the pMx-1-SV40T-Neo-195 DNA expression vector. Following transfection, selection in the 0.5 mg/ml G418-sulfate (neomycin analog) containing medium was carried out for 10–14 days in which 1–2% of the initial plated cells survived. In contrast, all of the mock transfected cells died within a 4–7-day period. The stable transfectants were maintained in α-MEM/5% FBS containing 750 U/ml human A/D interferon, and thereafter, the cultures were not maintained under continual selection pressure (i.e., G418-containing medium). Seven clonal lines designated OPC1–OPC7 were obtained by a standard limiting dilution protocol of the polyclonal transfectants in 96-well plates. Three of the clones, OPC4, OPC5, and OPC7 were eliminated from additional evaluation as they exhibited a fusiform, tapered-end morphology and growth characteristics that were equal to or reduced as compared with the nontransformed cells (Table 2). The remaining four clones were selected for additional expansion and characterization. Morphologically, these clonal cells generally exhibited a polygonal morphology with the extension of short dendritic processes at low density, which when grown to a confluent monolayer possessed an epithelial morphology (Fig. 2B). Cell proliferation kinetics for the four clones, OPC1, OPC2, OPC3, and OPC6, approximated 3.5–4.2 doublings per week (population doubling time of 40–49 h), and the clone designated OPC1 (Table 2) was selected as the lead candidate based on the ability of the OPC1 line to up-regulate significantly the ALP activity to low-dose rhBMP-2, 10 ng/ml, as assessed at day 9.

Immunostaining for the presence of positive large Tag expression on the OPC1 line maintained in medium that drives the promoter for Tag, in the presence of human A/D interferon, revealed positive nuclear staining (Fig. 2C). In the absence of the stimulating agent (IFN), some of the OPCs retained a positive nuclear immunostaining for the Tag antibody (5–8%), suggesting that the Tag may be expressed constitutively in some of the OPC1 line, rendering these cells immortal and incapable of reversal. The SCT-1/hNGF cell line exhibited immunopositive nuclear Tag as a positive control.(31)

Cell proliferation

Growth kinetics of the OPC1 line maintained in the presence of four types of tissue culture medium are presented in Fig. 3 at days 4, 7, and 10 following the initial seeding period. The media included the serum-free defined media CM, the base medium (BM) of α-MEM/5% FBS, and both types of medium supplemented with 750 U/ml recombinant human A/D interferon (IFN). At day 4, the OPC1 cell counts in the wells containing the BM ± IFN with 5% FBS were significantly greater than the wells containing the CM ± IFN. A marked increase in cell numbers was observed in the OPC1 line maintained in the BM ± IFN from day 4 to day 7 that subsided from day 7 to day 10. By day 7, the OPCs exhibited cellular confluence in the wells containing the BM ± IFN media. At the 7- and 10-day interval, a significant increase in cell numbers was observed within each type of medium in the wells containing the tissue culture medium supplemented with the human IFN.

Figure FIG. 3.

Cell proliferation kinetics of the OPC1 line maintained in four types of tissue culture medium. At day 4, the OPC1 cell counts in the wells containing the BM ± IFN with 5% FBS were significantly greater than the wells containing the serum-free CM ± IFN. A marked increase in cell numbers was observed in the OPC1 line maintained in the BM ± IFN from day 4 to day 7 that subsided from day 7 to day 10. By day 7, the OPCs exhibited cellular confluence in the wells containing the BM ± IFN media, contributing to a diminished rate of proliferation from day 7 to day 10. At the 7- and 10-day interval, a significant increase in cell numbers was observed within each type of medium in the wells containing the tissue culture medium supplemented with the human IFN.

Phenotypic characteristics

Alkaline phosphatase: The OPC1 line has been evaluated at P10, P20, and P30 for ALP activity in the presence of several supplemented media at days 4, 9, and 16 after an initial seeding density of 2.5 × 104 cells/well (Table 3). At P20, with the exception of the control, the OPCs exhibited a statistically significant increase in the ALP enzyme activity in all medium conditions at 4, 9, and 16 days after the initial seeding period. A striking up-regulation of ALP enzyme activity was observed at 4 days in the OS, as well as the OS + 50 ng/ml rhBMP-2 groups, which was also the time the wells became nearly confluent. These groups exhibited activities of 63.2 ± 10.2 and 262.3 ± 14.8 ng of p-nitrophenol/minute/50,000 cells, respectively, as compared with 9.6 ± 2.4 for the control. By day 9, peak ALP enzyme activities in nanograms of p-nitrophenol/minute/50,000 cells were observed: control = 16.6 ± 2.6; 10 ng of BMP-2 = 127.8 ± 19.5; 50 ng of BMP-2 = 225.5 ± 22.8; 100 ng BMP-2 = 292.2 ± 24.4; OS = 310.1 ± 19.2; OS + 50 ng BMP-2 = 407.8 ± 19.5. Similar observations were noted for the OPC1 line at P10 and P30, with the complete data presented in Table 3. ALP histochemistry revealed staining that was consistent with the enzyme activity data. By day 9 following the initial seeding, a marked increase in the ALP staining was observed in the BMP groups ± the OS (wells B, C, E, and F) compared either with the negative control well (A) or the group treated with 50 ng/ml bFGF (well D) (Fig. 4). Note the intense ALP expression in the wells with rhBMP-2 (Figs. 4B and 4C), and with the OS + 50 ng/ml BMP group (Fig. 4F) at high power (Fig. 4H) compared with the negative control (Fig. 4G).

Table Table 3. ALP Enzyme Activity of OPC1 Cells with Various Treatments
original image
Figure FIG. 4.

ALP histochemistry at day 9 following the initial cell seeding indicated a marked increase in the ALP staining observed in the BMP groups ± the OS (wells B, C, E, and F) as compared with either the negative control well (A) or the group treated with 50 ng/ml bFGF (well D). Note the especially intense ALP expression in the wells with rhBMP-2 (B and C) and with the OS + 50 ng/ml BMP group (F) at high power (H) compared with the negative control (G).

Mineralization:

The OPC1 line was characterized quantitatively (biochemically) for extracellular calcium deposition during the formation of the mineralized nodules at P20. The data for the various conditions at day 9 and 16 are presented in Fig. 5. All groups were negative for calcium content at day 4, while the groups with OS (± 50 ng/ml rhBMP-2) exhibited a significant increase in the quantity of extracellular calcium deposition at day 9. By day 16, all of the groups exhibited a significant increase in the deposition of extracellular calcium, while the groups maintained in the OS deposited nearly 20 μg of calcium in a 6-well plate.

Figure FIG. 5.

Calcium deposition of the OPC1 line at P20 as characterized quantitatively (biochemically) during the formation of the mineralized nodules. All conditions were negative for calcium content at day 4, while the groups with OS (±50 ng/ml rhBMP-2) exhibited a significant increase in the quantity of extracellular matrix deposition at day 9. By day 16, all of the groups exhibited a significant increase in the deposition of extracellular calcium, while the groups maintained in the OS deposited nearly 20 μg of calcium in a 6-well plate.

The OPC1 line forms mineralized nodules in tissue culture wells under all treatment conditions. To visualize the mineral deposition within the nodules, a von Kossa stain was used and cultures were examined by light microscopy. The von Kossa stained specimens were detected by day 4–5 postconfluency and were extensive by days 7–10 in treatments maintained in the OS + 50 ng rhBMP-2. Once the cells reached confluency, a small number of mineralized nodules became visible in the cells maintained in base medium of α-MEM/5% FBS at day 9 following the initial seeding (well A), exhibiting multilayered nodules. This was in contrast to the cells treated with 50 ng/ml bFGF (well D) in which no mineralized nodules were evident. However, in groups maintained with BMP-2 ± the OS (wells B, C, E, and F), and especially in the 50 ng rhBMP-2 + OS (well F), the number of mineralized nodules was markedly greater than in the control treatment group (well A) (Fig. 6). The formation of mineralized nodules in the cells maintained in base medium without either β-glycerophosphate or dexamethasone was consistent with a previous report for immortalized human fetal osteoblastic cells(17) but was otherwise unusual for osteoblastic cell lines.

Figure FIG. 6.

Histochemical staining of the OPC1 line at day 9 following the initial cell seeding by the von Kossa method. Positive staining for nodular aggregates (mineralized) can be observed in the negative control well (well A, with (G) representing the boxed region in A). Mineralization was increased in the groups maintained with 10 ng (B) and 50 ng (C) of BMP, as well as the OS group (E). The most intense staining was observed in the OS + 50 ng rhBMP-2 (well F), with a highlighted region from well F displayed in (H). Note the absense of von Kossa positive nodules in the group maintained in 50 ng/ml bFGF (D).

Osteocalcin EIA:

The data indicated that intact OC (ng/ml/24 h) was measured in the tissue culture medium under the following conditions: control = 0.9 ± 0.3; 10 ng BMP-2 = 1.2 ± 0.3; 50 ng BMP-2 =1.8 ± 0.4; 50 ng bFGF = 0.8 ± 0.2; OS =2.1 ± 0.5; OS + 50 ng BMP-2 =8.6 ± 2.7. A significant increase (p < 0.05) as compared with the control group was observed in the treatment groups except the 10 ng BMP-2 and bFGF conditions. The lysate values from ∼1 × 106 OPC1 cells maintained in the control medium was negligible (below the limit of detection), further indicating that the OC measured in the present study was intact and secreted de novo from the OPC1 line.

RT-PCR phenotyping:

RT-PCR phenotyping of the OPC1 line demonstrated that the OPC1 line expressed abundant mRNA for OC, ON, OP, PTHr, ALP, and ProI (Fig. 7A). The representative RT-PCR products (bands) were resolved by agarose gel electrophoresis and corresponded to the expected molecular sizes of 258, 369, 348, 571, 367, and 599 bp for OC, ON, OP, PTHr, ALP, and ProI, respectively (Table 1).

Figure FIG. 7.

RT-PCR analysis for the expression of osteoblast genes in the OPC1 line. Confluent cells had their mRNA extracted, reverse transcribed into cDNA, and amplified using specific oligonucleotide primers as described in the Materials and Methods. Separate PCRs were performed for OC, ON, OP, PTHr, ALP, and ProI. The reaction products were stained with ethidium bromide and resolved on agarose gels. Control ladders of 100 bp and 50 bp were run on the outside lanes and the base pair product lengths of 258, 369, 348, 571, 367, and 599 (A) were observed for OC, ON, OP, PTHr, ALP, and ProI, respectively (Table 1). (B) represents the quantitative RT-PCR products from the OPC1 line at ∼80–90% confluence (Pre Con) normalized to 1.0 and compared with postconfluent cells maintained for 4 days in the basal medium (D4–) or the OS + 50 ng BMP-2 medium condition (D4+). A significant increase in the levels of RT-PCR products of the selected mRNAs ON, OP, PTHr, and ProI were observed following the 4-day maintenance in the OS + 50 ng BMP-2 medium condition (+) as compared with the D4 (–) group or the preconfluent control group.

The levels of mRNA expression of the osteogenic markers for ON, OP, PTHr, and ProI were further characterized by quantitative RT-PCR. The RT-PCR products from the OPC1 line at ∼80–90% confluence (preconfluence) was normalized to 1.0 and compared with postconfluent cells maintained for 4 days in the basal medium (–) or the OS + 50 ng BMP-2 medium condition (+). A significant increase in the levels of RT-PCR products of the selected mRNAs ON, OP, PTHr, and ProI were observed following the 4-day maintenance in the OS + 50 ng BMP-2 medium condition (+) as compared with the D4 (–) group or the preconfluent control group (Fig. 7B). Under the (+) medium condition, the OP, PTHr, and ProI from the OPC1 line exhibited a 7.4-, 9.1-, and 10.3-fold increase in the level of mRNA expression. Human-derived glioblasts were used as a control cell type and were negative for OC, ON, OP, and PTHr as determined by RT-PCR.

DISCUSSION

The present report details the establishment and characterization of a series of new human fetal osteoprecursor cell lines (OPCs) which are immortalized with a gene coding for the SV40 Tag. One of these lines, OPC1, was chosen as the lead candidate for additional characterization based on the ability of this cell line to exhibit the capacity to generate programmed differentiation in the presence of low dose rhBMP-2 (10 ng/ml). The incorporation of the DNA plasmid into the primary cultures was intended to conditionally express SV40 Tag when the Mx-1 promoter was activated by human A/D interferon. The OPC1 line exhibited a significant increase in cell proliferation when maintained in either a serum-free defined medium or a nutrient-rich base medium containing 5% (v/v) FBS that was supplemented with human A/D interferon. However, abundant growth was observed in the nutrient-rich base medium containing 5% (v/v) FBS, suggesting that the Tag is expressed constitutively. This observation is consistent with the finding that the Mx-1 promoter can be activated in the presence of serum in the absence of human A/D interferon (S.C. Wong, personal communication). The OPC1 growth curve for cells maintained in the presence of the serum-free CM media without interferon also exhibits a slow but steady growth rate, with concurrent Tag immunopositive nuclear staining evident in ∼2–3% of the cells. Additionally, when the OPC1 line became confluent in standard two-dimensional tissue culture plastic, a concurrent down-regulation of immunopositive nuclear Tag staining was observed that coincided with a decrease in the cell proliferation rate from day 7 to day 10 in the BM ± IFN. To determine the impact of this observation in transplants in situ, we have evaluated cell-loaded polymer implants in athymic rats, which have not exhibited any tumorigenic activity. These cells have been repeatedly frozen and thawed (at least three times) and continue to maintain consistent levels of the osteogenic phenotypic markers as described and characterized in the present study.

Numerous phenotypic markers have been monitored to determine the stability of expression in the OPC1 line in association with osteoblastic differentiation. These include ALP expression, the ability to mineralize, measurement of intact OC, and mRNA expression of OC, ON, OP, PTHr, ALP, and ProI. The present study indicated that postconfluent cultures of the OPC1 line express high levels of these osteoblastic-associated markers at P10, P20, and P30. Quantitative RT-PCR has shown that a significant increase in the levels of RT-PCR products of mRNAs for ON, OP, PTHr, and ProI were observed following the 4-day maintenance in the OS + 50 ng BMP-2 medium condition (+) as compared with the D4 (–) group or the preconfluent control group. Additional studies are under way to determine if the PTHr's are functional by characterizing the induction of cAMP to various agonists.(32) However, it has been reported previously that the distribution of PTHr's in bone cells is variable (and restricted), indicating that not all osteoblasts, notably the mature osteoblasts, bind PTH.(33) Studies also are underway to determine if the OPC1 line contains the mRNA for an additional osteoblast marker, OSF-2, that was recently cloned from an MC3T3-E1 library and shown to be expressed in primary osteoblasts, the MC3T3-E1 cell line and in lung tissue.(34) Interestingly, the OSF-2 osteoblast marker was recently utilized to distinguish an osteoblast/osteocyte phenotype, which was shown to be negative in an established osteocyte-like cell line, MLO-Y4.(35)

Many of the bone morphogenetic factors, a subfamily within the transforming growth factor-beta superfamily, have been shown repeatedly to play critical roles in bone formation.(1–2,36,37) Interactions of rhBMP-2 with the nonosteogenic cell lines, C2C12 myoblasts,(38) as well as C3H10T1/2 fibroblasts,(11) have induced the differentiation of these cells into osteoblasts. Thus, rhBMP-2 appears to be a potent regulatory factor to control osteoblast differentiation. In the present studies, the ability of rhBMP-2 to elicit a stimulatory effect on ALP activity in the OPC1 line in vitro is consistent with previous reports of rhBMP-2 activity on marrow stromal precursors,(14,39) as well as rat-derived osteoblast-like cells, ROB-C26, described as potential osteoblast precursor cells.(13) In addition, reports document that rhBMP-2 has no stimulatory effect either on differentiated osteoblasts obtained from human iliac bone(38) or on a differentiated rat-derived osteoblast cell, ROB-C20.(13) Thus, based on the OPC1 line's capacity to generate programmed differentiation in the presence of low-dose rhBMP-2 (10 ng/ml), with a markedly large ALP stimulation index (10 ng BMP/control), the OPC1 line appears to represent a homogeneous osteoprecursor cell line. Furthermore, the ability to demonstrate programmed differentiation at a dose of 10 ng/ml rhBMP-2 is markedly lower than the dose described for the mouse-derived MC3T3-E1,(40) a rat-derived “potential” osteoblast precursor cell line,(13) and differentiated osteoblasts from human iliac bone.(38)

To provide a consistent and reproducible culture system for characterizing bone-graft substitutes, it is imperative to utilize a model system that is consistent, sensitive, and a biomimetic for endogenous human osteoprecursor cells. In addition, many characteristics of the biomaterial bone-graft substitute influence the ability of the osteoblast and precursor cell to initiate and achieve a timely sequence of programmed differentiation.(1–2,41,42) To characterize reproducibly the potential influence that many of the physicochemical characteristics of a material, as well as the surface modification on substrates may have on the programmed cascade of osteogenesis, a well-defined human-derived osteoprecursor cell line is needed that demonstrates programmed osteogenic differentiation. Although some cell lines, such as other osteoblast human-derived immortalized,(15,23) as well as conditional immortalized cell lines,(17) are available and have been utilized as a model system to develop bone-graft substitutes, the OPC1 line is a human-derived clonal cell that exhibits a consistent pattern of differentiation over at least 70 passages. In addition, generating a history for a cell line from the isolation through the genetic manipulation to the continued characterization of the stability and safety of a transgene are important safety and regulatory considerations in the development of human-derived cell lines that may be applied clinically. The OPC1 line fulfills these criteria and exhibits the capacity to generate programmed differentiation in the presence of a low rhBMP-2 (10 ng/ml) dose, a property defining osteoprecursor cells.(13–14,39)

In summary, the OPC1 line is a contaminant-free, human-derived immortalized osteoprecursor cell line that undergoes programmed osteogenic differentiation. This cell line exhibits a stable incorporation of the SV40 Tag transgene without continual selection pressure that does not appear negatively to impact the growth, maintenance, or differentiation genome of the host cell. Currently, the OPC1 line has been maintained for greater than 70 passages and has not exhibited the growth crisis and senescence observed in the nontransformed parent cell line. Thus, the continued proliferative capability of the OPC1 cell line appears to be dependent upon the expression of the Tag oncogene. Additional studies will evaluate other conditional immortalizing transgenes and describe the ability to stably incorporate a second transgene to produce and secrete a growth or differentiation factor. We have established and characterized an osteoprecursor cell line that can provide a sensitive in vitro cell culture system to evaluate bone development, cell/biomaterial interactions and may be useful for screening putative bone differentiating factors.

Acknowledgements

The authors gratefully acknowledge the skillful technical assistance of David Buck, John Schmitt, Rich Sipe, and Cynthia Bohan from the OHSU and Zhong-Yi Hu from CytoTherapeutics, Inc. We also acknowledge materials provided by CytoTherapeutics, Inc. and Genetics Institute, Inc. In addition, we thank our friends and colleagues from CytoTherapeutics and Genetics Institute for their feedback and continued support.

Ancillary