• osteoblasts;
  • green fluorescent proteins;
  • parathyroid hormone;
  • xylenol orange


  1. Top of page
  2. Abstract
  7. Acknowledgements

Primary calvarial osteoblast cultures derived from type I collagen promoter-GFP reporter transgenic mice were used to examine progression of the osteoblast lineage. This system was validated by assessing the effect of PTH on osteoblast growth in real time. The anabolic effect of PTH seemed to be the result of enhanced osteoblast differentiation rather than expansion of a progenitor population.

Introduction: Activation of green fluorescent protein (GFP) marker genes driven by Col1a1 promoter fragments has been associated with the level of osteoblast differentiation. GFP-marked cultures provide an approach to continuously monitor the level of osteoblast differentiation in real time without the termination of cultures.

Materials and Methods: Neonatal calvarial cells transgenic for pOBCol2.3GFP and pOBCol3.6GFP were used to establish calvarial osteoblast cultures. Parathyroid hormone (PTH) was added either continuous (days 1–21) or transient (days 1–7) to examine its diverse effect on osteoblast differentiation in cultures for 21 days. Three fluorescent markers were used: (1) pOBCol3.6GFP, which is activated in preosteoblastic cells; (2) pOBCol2.3GFP, which is restricted to differentiated osteoblasts; and (3) xylenol orange (XO), which stains the mineralized nodules. Progression of osteoblast differentiation indicated by fluorescent markers was documented throughout the entire period of culture. Recorded fluorescent images were analyzed in the patterns of expression and quantitated in the area of expression.

Results: Continuous PTH blocked osteoblast differentiation, which was evident by the attenuation of pOBCol3.6GFP and an absence of pOBCol2.3GFP. In contrast, transient PTH inhibited the initial osteoblast differentiation but ultimately resulted in a culture with more mineralized nodules and enhanced osteoblast differentiation expressing strong levels of pOBCol3.6GFP and pOBCol2.3GFP. Quantitative analysis showed that transient PTH first decreased then later increased areas of GFP expression and XO staining, which correlated with results of Northern blot and alkaline phosphatase activity. Transient PTH caused a decrease in DNA content during the treatment and after the removal of PTH.

Conclusion: GFP-marked cultures combined with fluorescent image analysis have the advantage to assess the effect of PTH on osteoblast differentiation in real time. Results suggest that the anabolic effect of transient PTH is caused by an enhancement in osteoblast differentiation rather than an increase in the population of progenitor cells.


  1. Top of page
  2. Abstract
  7. Acknowledgements

B one formation is accomplished by the recruitment and proliferation of osteoprogenitor cells and the subsequent differentiation into osteoblasts, which produce the bone matrix. Osteoblasts are derived from mesenchymal pluriprogenitor cells present in the bone marrow that have the potential to differentiate into other tissue-specific cells such as chondroblasts, myoblasts, fibroblasts, and adipocytes.(1,2) The progression of mesenchymal pluriprogenitors into cells of the osteoblast lineage can be divided into several differentiation stages based on morphological and biochemical criteria.(3) For instance, the multipotential osteoprogenitors are identified as proliferative fibroblastic cells with surface markers for STRO-1,(4,5) SB-10,(6) and HOP-26.(7) Pre-osteoblasts, which continue to proliferate to fibroblast-shaped cells, are the next level of differentiation based on the expression of type I collagen (Col1a1) and alkaline phosphatase (ALP).(8,9) Differentiated or mature osteoblasts are nonproliferative cuboidal cells located in multilayered nodules that express osteocalcin (OCN) and bone sialoprotein (BSP). The progression of osteoblast lineage can be examined conventionally through the primary osteoblast cultures isolated from neonatal calvaria or bone marrow stroma using a number of standard biochemical methods. However, because of the heterogeneous nature of cell populations, it is impossible to precisely assess the progression of different subpopulations when the culture is harvested as a whole. Furthermore, the termination of cultures to perform analysis has precluded the opportunity to continuously examine the progression of lineage in the same culture over time.

Green fluorescent protein (GFP) has drawn increasing attention for its application to visualize gene expression and protein interaction in living cells in real time and to isolate subpopulation of cells using fluorescence-activated cell sorting (FACS) technology.(10-13) Recently the strategy of using genetically engineered cells to express GFP under control of cell type-specific promoters has been successfully applied to identify subpopulations of cells at different stages of the osteoblast lineage in transgenic mice harboring the type I collagen promoter driven transgenes pOBCol3.6GFP and pOBCol2.3GFP.(14) Histological analysis of bone showed that pOBCol3.6GFP is active in periosteal fibroblasts and bone lining cells, indicating both pre-osteoblasts and differentiated osteoblasts are marked, whereas pOBCol2.3GFP is restricted to differentiated osteoblasts and osteocytes.(14) In primary osteoblast cultures derived from neonatal calvarial or bone marrow stromal fibroblasts, the pOBCol3.6GFP signal initiates at a low level concomitantly with ALP and Col1a1 mRNA expression, and the signal intensifies when the growing nodule shows signs of differentiation. In contrast, pOBCol2.3GFP expresses a uniform intensified level at the time when BSP and subsequent OCN are expressed and mineralization occurs.(14) Therefore, the initiation of weak levels of pOBCol3.6GFP marks the transition of multipotential osteoprogenitors to preosteoblasts, whereas strong levels of pOBCol3.6GFP and the activation of pOBCol2.3GFP expression are associated with the appearance of differentiated osteoblasts.

Parathyroid hormone (PTH) has diverse effects on the bone tissue depending on the mode of administration.(15,16) The catabolic consequences of PTH reported in hyperparathyroidism and in continuous treatment with PTH include rapid bone turnover, increased osteoclast activity, and decreased cortical bone mass.(17-22) On the other hand, the anabolic effects described in intermittent exposure to PTH show increased bone formation.(17, 21-25) At the cellular level, PTH regulates the osteoblast differentiation positively or negatively depending on the duration of PTH treatment and the stage of osteoblast differentiation when PTH is administered.(26-28) Whereas osteoblast differentiation was strongly inhibited by the continuous exposure to PTH, it was stimulated by the intermittent exposure to PTH. It was suggested that PTH may preferentially stimulate osteoblast differentiation in immature cells but inhibit it in mature osteoblasts.(15)

In this study, we applied the well-known inhibitory and stimulatory effects of PTH to validate GFP-marked cultures and tested the hypothesis that the status of osteoblast differentiation under the influence of PTH can be evaluated qualitatively and quantitatively through the expression of stage-dependent fluorescent markers. Neonatal calvarial osteoblast cultures derived from pOBCol2.3GFP and pOBCol3.6GFP transgenic mice were examined using three fluorescent markers: (1) pOBCol3.6GFP, which is expressed in preosteoblasts and differentiated osteoblasts; (2) pOBCol2.3GFP, which is restricted to differentiated osteoblasts; and (3) xylenol orange (XO), which stains the mineralized nodules. Specifically, progression of the osteoblast lineage from osteoprogenitors to differentiated osteoblasts was assessed through the expression of fluorescent markers in terms of the time, strength, and area.


  1. Top of page
  2. Abstract
  7. Acknowledgements

Mouse calvarial primary osteoblast cultures

Calvarial osteoblast cultures derived from mice transgenic for either pOBCol2.3GFP or pOBCol3.6GFP(14) were established to examine the osteoblast differentiation. Neonatal calvarial cells were isolated from 6- to 8-day-old mice using a modified sequential digestion described by Wong and Cohn.(29) Briefly, after removal of sutures and adherent mesenchymal tissues, calvariae were subjected to four sequential 15-minute enzyme digestions at 37°C in solution containing 0.05% trypsin-EDTA and 0.1% collagenase P (Roche Diagnostics). Cells released from the second to fourth digestions were pooled, centrifuged, resuspended, and plated at 1.5 × 104/cm2 (1.5 × 105/well) in 35-mm 6-well culture plates in DMEM (Invitrogen) containing 10% FCS, penicillin (100 units/ml), streptomycin (100 μg/ml), and nonessential amino acids (100 μM). In this study, the day of plating was counted as day 0. Plated cells became confluent around day 5-7; the culture medium was changed to differentiation medium, which was α-MEM (Invitrogen) containing 10% FCS, penicillin (100 units/ml), streptomycin (100 μg/ml), ascorbic acid (50 μg/ml), and β-glycerophosphate (4 mM). Medium was changed every other day for the entire duration of culture. The experiment protocol was approved by the University of Connecticut Health Center Animal Care Committee (Protocol 2001-120).

Administration of PTH

Human PTH [PTH (1-34); Bachem Bioscience, King of Prussia, PA, USA] powder was dissolved in 4 mM HCl with 0.1% bovine serum albumin (BSA) to make reconstituted stock solution that can be stored at −20°C. To exclude the possible effect of vehicle, cultures were treated with the same concentration of BSA. There was no difference between the control and vehicle-treated cultures. For continuous exposure, PTH was added at day 1 and thereafter with each change of medium for the entire culture period. For transient exposure, PTH was present from day 1 to day 7. The final concentration of PTH in culture medium was 25 nM (or 103 ng/ml). This concentration was chosen based on the results of a previous study on mouse calvarial osteoblast cultures.(27)

Photography, concatenation, and quantitation of fluorescent microscopic images

Procedures for acquisition of images are schematically presented in Fig. 1. Cell cultures were imaged with computerized Zeiss Axiovert 200 (Carl Zeiss, Thornwood, NY, USA) microscope using Openlab software (Improvision, Lexington, MA, USA). Fluorescent expression of pOBCol3.6GFP and pOBCol2.3GFP was examined using a Topaz (YFP) filter (Chroma Technology, Rockingham, VT, USA). The Zeiss Axiovert 200 microscope is equipped with the motorized X-Y-Z platform, motorized fluorescent cube, and AxioCam color digital camera that is controlled by the user-defined computation program. In this program, different parameters were set up regarding the date, position, magnification, filter, and exposure; thus, the images of cultures can be consistently and repeatedly recorded and analyzed under the same condition. The microscopic workstation allows the user to reproducibly record images of cultures at the same location at defined time points during the entire culture period. As current configuration, the microscopic workstation allows 63% of a 35-mm culture well to be repeatedly imaged and yields a series of 25 adjacent pictures that can be concatenated into a single image of the well. The development of fluorescent expression was recorded for digital analysis and scored for the size of expressed area. Concatenated images from different time-points were entered in a computation program developed from the Openlab software and adjusted to threshold to quantitate the area of fluorescent expression. The threshold was determined arbitrarily based on the extent of a particular requested level of fluorescence (such as area of weak or strong expression of pOBCol3.6GFP and pOBCol2.3GFP). For consistency and reproducibility, all fluorescent images from the same culture at different time-points were quantitated using the same threshold. The area of fluorescent expression was shown in pixels.

thumbnail image

Figure FIG. 1. Schematic representation of procedures in photography and concatenation of microscopic images. The image workstation (Zeiss Axiovert 200 operated with Improvision Openlab) can record the image of a well (35-mm diameter) from the left top corner to the right bottom corner in a series of 25 (5 × 5) small images that are labeled from y1×1 to y5×5. Each small image represents a 5.4 × 4.5-mm area. These 25 small images are concatenated seamlessly into a single large image (27 × 22.5 mm), which accounts for 63% of the area of a well. Patterns of GFP expression such as the location and intensity in culture can be observed and recorded reproducibly throughout the entire culture period.20

Download figure to PowerPoint

Northern blot, DNA quantitation, and ALP activity analysis

Northern blot:

Total RNA was prepared from cells using TRIzol Reagent (Invitrogen) according to the manufacturer's instructions. After the isopropanol precipitation, the RNA was separated on 1% agarose/1.1 M formaldehyde gel and transferred onto a nylon membrane. Northern blot was carried out using [32P]labeled DNA probes generated by random oligonucleotide labeling. Hybridization was performed in 50% formamide and 6× SSPE at 42°C. Hybridized membranes were processed for autoradiography and later digitized using a Storm 860 (Molecular Dynamics, Sunnyvale, CA, USA) phosphor-imager for signal quantitation. Hybridization to 18S RNA probe was used to normalize loading in different lanes.

DNA quantitation:

DNA-binding fluorochrome 4′,6-diamimidino-2-phenylindole (DAPI; Sigma) was used to quantitate the amount of DNA in cultures. Scraped or trypsin-released cells were lysed in 0.5 M NaOH and neutralized with 0.5 M acetic acid. After the addition of DAPI reagent, the amount of DNA was determined by the binding of DAPI to DNA (excitation, 344 nm; emission, 466 nm) using Fluoro Count (Packard Instrument, Meriden, CT, USA).

ALP activity:

Cultured cells were scraped and homogenized in 10 mM Tris (pH 7.5) containing 0.1% Triton X-100. Analysis was performed by the hydrolysis of colorless p-nitrophenyl phosphate to a yellowish p-nitrophenol in the presence of 1 mM MgCl2. ALP activity was measured by spectrophotometry at 410 nm using μQuant spectrophotometer (Bio-Tek, Winooski, VT, USA). ALP activity (nmol/min/mg) was normalized with cellular protein measuring by BCA Protein Assay Reagents (Pierce, Rockford, IL, USA) according to the manufacturer's directions.

XO and von Kossa stainings

XO staining:

XO, a fluorochrome chelating reagent, has been administered systemically to label the newly calcified tissues in vivo.(30,31) We have applied XO to living cultures in vitro to mark the mineralized nodules where XO expresses red color under the microscope using a DsRed filter (Chroma Technology). XO powder (Sigma) was dissolved in distilled water and filtered to make the concentrated 20 mM stock and stored at 4°C. XO was added to culture medium at a concentration of 20 μM for 4 h to overnight. Before photography, the XO-containing medium was exchanged with fresh medium to avoid the nonspecific fluorescent background.

von Kossa staining:

Mineralization of nodules was assessed using modified von Kossa's silver nitrate staining method. Briefly, cultures were fixed in cold methanol for 15-20 minutes. After rinsing, the fixed plates were incubated with 5% silver nitrate solution under UV light using two cycles of auto-cross-link (1200 μJ × 100) in a UV Stratalinker (Strategene, La Jolla, CA, USA). Mineralized nodules were seen as dark brown to black spots.

Statistical analysis

Statistical analysis was carried out using SPSS-11.0 software (SPSS, Chicago, IL, USA). In comparisons among the three groups, data were analyzed with one-way ANOVA followed by Scheffe-posthoc test. In comparison between the two groups, data were analyzed with Student's t-test. All values were expressed as the mean ± SE, and p < 0.05 was considered statistically significant.


  1. Top of page
  2. Abstract
  7. Acknowledgements

Patterns of GFP expression and mineralization in calvarial cultures

In pOBCol3.6GFP cultures, GFP showed weak expression beginning on days 4-6 (data not shown), which intensified around day 10 (Fig. 2A), when osteoblast nodules started to develop. Acquisition of pOBCol3.6GFP expression was a dynamic process with continuous development of GFP expression from weak to strong levels. These two levels of expression could be better appreciated under a higher magnification (Fig. 2A, enlarged panel) and were characterized as diffuse weak levels and bright strong levels. At day 14 and thereafter, most areas previously having weak levels of pOBCol3.6GFP had enlarged and acquired strong GFP expression (Fig. 2A). In pOBCol2.3GFP cultures, the expression of GFP was not detected until day 10 (Fig. 2B). The intensity of expression was uniform and comparable with the strong levels of pOBCol3.6GFP (Fig. 2B, enlarged panel). Areas of GFP expression in both pOBCol3.6GFP and pOBCol2.3GFP cultures expanded as cells continued to differentiate and peaked in the third week.

thumbnail image

Figure FIG. 2. (A and B) Patterns of GFP expression and XO staining. Except the enlarged, each panel is an individual small image as described in Fig.1. (A) Expression of pOBCol3.6GFP started from weak levels and expanded and intensified to strong levels as cells continue to proliferate and differentiate. At day 10, two different levels of expression (diffuse weak levels and bright strong levels) were observed with higher magnification. No XO staining was observed at day 10. After day 14, GFP expression intensified and peaked during the third week. XO staining at day 14 and day 18 coincided with portions of areas that showed strong levels of pOBCol3.6GFP. (B) Expression of pOBCol2.3GFP was not detected until day 10, and the intensity was comparable to the strong levels of pOBCol3.6GFP. No diffuse weak level of expression was observed under higher magnification. GFP expression increased as cells continued to differentiate and congregate to form nodules at days 14 and 18. No XO staining was detected at day 10. By days 14 and 18, XO staining overlapped areas expressing pOBCol2.3GFP. (C) XO staining at different concentrations. Each panel is a concatenated large image as described in Fig.1. XO was added to day 20 calvarial osteoblast cultures at different concentrations (100 nM ∼ 1 mM). After overnight incubation with XO, cultures were photographed for XO fluorescence and terminated and stained for mineralized nodules with von Kossa. Low concentrations (100 nM and 1 μM) did not label the mineralized nodules, and the highest concentration of XO (1 mM) overstained the culture with nonspecific background fluorescence. The concentrations of 10 μM and 100 μM gave rise to specific red fluorescence that was co-localized exactly with von Kossa stain. (A and B) Scale bar = 0.5 mm, which applies to the enlarged and different time-point panels. (C) Scale bar = 10 mm, which applies to all panels.20

Download figure to PowerPoint

XO produces a strong red fluorescent signal in areas that are undergoing mineralization at the time of administration. In this study, XO was used to examine the mineralization in living cultures in real time without termination of cultures. XO was added to medium at different concentrations ranging from 100 nM to 1 mM. XO in the range of 10-100 μM produced the best fluorescent stains that were co-localized exactly with the silver nitrate stains using von Kossa's method (Fig. 2C). XO at 20 μM was used in this study and showed no significant effect on ALP activity and DNA content (data not shown), indicating the 20 μM XO does not change the growth or differentiation of the cultures. XO staining allowed us to monitor the pace of mineralization in the same culture over time.

Mineral stained with XO (20 μM) appeared first between days 12-14 (Figs. 2A and 2B). Whereas XO staining correlated closely with the expression of pOBCol2.3GFP (Fig. 2B), it partially overlapped the expression of pOBCol3.6GFP (Fig. 2A). In general, areas stained with XO were smaller in size but located in the same areas expressing strong levels of GFP. A time lag was observed between GFP expression and XO staining. For example, in pOBCol3.6GFP cultures, the XO staining at day 18 corresponded spatially to strong levels of GFP expression at day 10 (Fig. 2A). Similarly, in pOBCol2.3GFP cultures, the XO staining at day 14 and day 18 agreed, respectively, with the GFP expression at day 10 and 14 (Fig. 2B). Our observations indicated that cells expressing strong levels of pOBCol3.6GFP and pOBCol2.3GFP were restricted to the nodules that will be mineralized eventually.

Effect of continuous PTH on GFP expression and osteoblast differentiation

PTH(1-34) was present at a concentration of 25 nM for the entire 21-day culture period to examine its effect on GFP expression and osteoblast differentiation. In comparison with vehicle-treated controls (Figs. 3A and 3D), the continuous presence of PTH maintained the weak levels of pOBCol3.6GFP expression (Fig. 3B) but completely suppressed the expression of pOBCol2.3GFP (Fig. 3E), indicating a block of osteoblast differentiation. The lack of XO staining at day 21 in both cultures confirmed the inhibition of osteoblast differentiation (Figs. 3B and 3E). Northern blot analysis showed weak expression of Col1a1 mRNA and no expression of BSP and OCN mRNA at all time-points (Fig. 4A). Moreover, there was a low level of ALP activity at all time-points (Fig. 4B). Our observations were consistent with previous reports that osteoblast differentiation is negatively regulated by the continuous administration of PTH.(27,28) Based on these results, continuous PTH blocks the forward progression of osteoprogenitors and pre-osteoblasts to full osteoblasts differentiation.

thumbnail image

Figure FIG. 3. Effect of PTH on GFP expression. (A and D) Control cultures without the addition of PTH. (B and E) Cultures treated with continuous PTH (days 1-21). (C and F) Cultures treated with transient PTH (days 1-7). Each panel is a concatenated large image as described in Fig.1. (A) Expression of pOBCol3.6GFP was very weak at day 7 and intensified during the second week of culture. GFP expression peaked during the third week. XO staining confirmed the presence of mineralized nodules at day 21. (B) Continuous PTH maintained the weak levels, but inhibited the strong levels of pOBCol3.6GFP. No mineralization was detected as indicated by the absence of XO staining at the end of the culture period. (C) Transient PTH enhanced and expanded the expression of pOBCol3.6GFP and resulted in larger areas of mineralized nodules as shown by XO staining at day 21. (D) Expression of pOBCol2.3GFP appeared in discrete nodules starting at day 10. Areas of expression expanded as culture continued to differentiate and mineralize. Expression of pOBCol2.3GFP was coincided with areas of XO staining. (E) Continuous PTH completely inhibited the expression of pOBCol2.3GFP and no mineralized nodule formed by day 21. (F) Transient PTH delayed the expression of pOBCol2.3GFP till day 14. After recovery, the areas with pOBCol2.3GFP expression expanded and exceeded those of control cultures by day 21. These increased GFP expression areas were also accompanied by larger areas of XO staining. (Scale bar = 10 mm, which applies to all panels).20

Download figure to PowerPoint

thumbnail image

Figure FIG. 4. Effect of PTH on cell differentiation. Except for difference in the expression of GFP, both pOBCol3.6GFP and pOBCol2.3GFP cells have the same pattern of endogenous gene expression (Col1a1, BSP, and OCN) and ALP activity in the calvarial osteoblast cultures. (A) Northern blot analyses of RNA derived from the pOBCol3.6GFP controls and cultures treated with transient PTH (PTH days 1-7) or continuous PTH at different time-points. Continuous PTH suppressed the expression of Col1a1 and completely inhibited the expression of BSP and OCN. Transient PTH inhibited the expression of BSP at day 10, but recovery to control level was seen by day 14. Elevated levels of OCN mRNA were seen from days 14 to 21 in transient PTH-treated cultures. When normalized to 18S rRNA levels at day 21, the signal intensity ratio of transient PTH to control is 1.3 for Col1a1, 1.2 for BSP, and 1.5 for OCN. Each lane contains 10 μg of total RNA. (B) ALP activity derived from pOBCol3.6GFP and pOBCol2.3GFP cultures at different time-points. In control cultures, ALP activity increased with time and reached a plateau after day 17. In contrast, there was a low level of ALP activity that did not increase over time in continuous PTH cultures. In cultures treated with PTH from days 1 to 7, although initial ALP activity was greater than control at day 7, ALP activity was suppressed and did not fully recover until day 17. At day 21, ALP activity in cultures treated with transient PTH was significantly higher than those of control and continuous PTH treatment. (* and # indicate a significant difference compared with the control at the same time-point, p < 0.05).20

Download figure to PowerPoint

Effect of transient PTH on GFP expression and osteoblast differentiation

PTH(1-34) (25 nM) was present in culture medium for the first 7 days and removed for the duration of culture. At day 7 in pOBCol3.6GFP cultures, there was a slight suppression in weak levels of GFP expression that was appreciated only at higher magnification (data not shown). This slight suppression of GFP was reversed shortly after the discontinuation of PTH administration. The intensity of pOBCol3.6GFP in the transient PTH treatment (Fig. 3C) exceeded that of the control at the end of culture. XO staining at day 21 showed more mineralized nodules in the culture treated with PTH from days 1 to 7 (Fig. 3C).

In pOBCol2.3GFP cultures, transient PTH treatment caused a significant delay in the GFP expression (Fig. 3F). The expression of pOBCol2.3GFP started to recover at day 14 and was even more intense than that of the control by the third week of culture (Fig. 3F). Similarly, at the end of culture, there was greater XO staining than untreated control (Fig. 3F).

The effect of transient PTH treatment on GFP expression (i.e., inhibition followed by stimulation) was correlated with results of Northern blot showing inhibition of BSP expression at day 10 and recovery at day 14, and stronger OCN expression from day 14 to day 21 (Fig. 4A). Compared with the control, the transient PTH treatment resulted in an initial higher ALP activity at day 7 (Fig. 4B). Whereas it was increasing with time in the control cultures, the ALP activity in transient PTH-treated cultures was inhibited and did not recover until day 17 (Fig. 4B). At day 21, there was higher ALP activity in the transient PTH-treated cultures (Fig. Fig. 4B).

Quantitation of GFP expression and XO staining

The close correlation between patterns of pOBCol2.3 GFP expression and osteoblast differentiation implies that osteoblast differentiation may be assessed quantitatively by the magnitude of GFP expression. Using the Improvision OpenLab software, we developed a computation program to calculate the total area of pOBCol2.3GFP expression. Compared with controls, pOBCol2.3GFP cultures treated with PTH days 1-7 showed no GFP expression at day 10 and a decrease in the total area of pOBCol2.3GFP at day 14 (Fig. Fig. 5A). This reduction in the area of expression was followed by recovery at day 17 and augmentation at day 21 (Fig. 5A). The pattern (i.e., inhibition [RIGHTWARDS ARROW] recovery [RIGHTWARDS ARROW] enhancement) in GFP expression levels was consistent with the results of the Northern blot and ALP activity analyses, suggesting that the tempo of osteoblast differentiation can be assessed by the real time expression of GFP.

thumbnail image

Figure FIG. 5. Quantitative analysis of GFP expression, XO staining and DNA content in cultures. (A) Quantitation of GFP expression in pOBCol2.3GFP cultures treated with transient PTH (PTH days 1-7) showed that GFP expression was decreased at both day 10 and day 14 and recovered and exceeded control levels thereafter. Quantitation of XO staining (hatched area) showed that transient PTH treatment resulted in more mineralized areas at day 21. When cultures were treated with continuous PTH, the expression of GFP was completely inhibited and no XO staining was detected (data not shown). (B) Continuous PTH prevented an increase in DNA content at all time-points. In cultures treated with transient PTH, there was a significant decrease in DNA content at both day 7 and day 10. (* and # indicate a significant difference compared with the control at the same time-point, p < 0.05).20

Download figure to PowerPoint

The total area stained with XO was also quantitated with the same computation program modified for red fluorescence. At day 17, the total area stained with XO in cultures treated with PTH days 1-7 was similar to that in control cultures (hatched area in Fig. 5A). However, at day 21, there was a significant increase in XO-stained area in the transient PTH-treated cultures (hatched area in Fig. 5A). At day 21, the proportion of GFP expression area that also stained with XO was 70% in control cultures and 85% in the transient PTH-treated cultures. This result suggests that, in the transient PTH-treated cultures, a greater proportion of osteoblasts within the nodules are capable of depositing mineral.

Effect of PTH on cell accumulation

Given that PTH first caused a delay in cell differentiation (i.e., maintenance in the progenitor stage) and the notion that there is more proliferation in less differentiated cells, we assumed that, at the cellular level, the increased mineralized nodule formation in cultures treated with transient PTH was because of an expansion of total precursor cell number. However, this did not prove to be the case in the DNA content measurement. We chose to estimate cell number within the cultures by total DNA content rather than direct cell counting to overcome potential inconsistency in the recovery of cells by proteolytic enzyme digestion. As the cultures mature and form a complex extracellular matrix, it is difficult to release entrapped cells leading to an underestimation of the total cell number. Measuring total DNA content overcomes the problem of cells that are not released or are ruptured by the harvesting technique.

DNA content measurement showed that, compared with the controls, there was a significant decrease in DNA content at all time-points in cultures treated with continuous PTH (Fig. 5B). In cultures treated with transient PTH, there was an initial significant decrease of DNA content at both days 7 and 10 (Fig. 5B). Although DNA content recovered gradually to the level of controls after day 15, it did not exceed the control level at the end of cultures (Fig. 5B). This result, combined with the data of the Northern blot, ALP activity, and GFP expression, suggests that the greater formation of mineralized nodules by transient PTH may arise from enhanced osteoblast differentiation but not an increase in expansion of the progenitor population.


  1. Top of page
  2. Abstract
  7. Acknowledgements

Primary cells isolated from neonatal calvaria or marrow stroma and immortalized cells derived from tumor or transformed by virus are the two major cell sources that have been used in cultures to assess the progression of osteoprogenitor cells to full osteoblast differentiation.(32) To assess the status of osteoblast differentiation in cultures, duplicate plates are terminated and subjected to a variety of examinations at different time-points. Consequently, a complete examination of culture is assembled from a series of observations and analyses from parallel replicate plates. This prevents the continuous examination of a culture exactly in the same plate from the beginning to the end. Because of this limitation, the effect of an agent or mutation on the tempo of the differentiation process is often overlooked, which probably is a crucial determinant of bone health. In addition, the heterogeneous nature of primary cell cultures has been amply shown by histological stains for ALP, fat, sulfated proteoglycan, and mineralization, and routines have been developed to image and quantitate these features of lineage progression. However, these procedures are rather insensitive to early markers of lineage progression and cannot be examined in a prospective manner.

The approach of monitoring osteoprogenitor progression by visual GFP markers provides the advantage to continuously analyze developmental processes in real time in heterogeneous cell cultures. First, the analysis of GFP can be performed on living cells noninvasively. Two stage-specific GFPs were used in this study: (1) the pOBCol3.6GFP, which appears initially in pre-osteoblastic cells and intensifies in differentiated osteoblasts, and (2) the pOBCol2.3GFP, which is activated in differentiated osteoblasts. These two unique markers expressing distinguishable color of GFP provide the possibility to continuously examine osteoprogenitor progression in the same culture plate. Second, the analysis of GFP can be monitored at the level of the colony/nodule or even the single cell instead of the whole culture. Combined with the powerful FACS technique to separate cell populations based on GFP expression, it is now possible to obtain more homogeneous cell population and analyze more precisely the progression of osteoblast development.(33)

In pOBCol3.6GFP cultures, three different cell populations were observed based on the expression of GFP (i.e., cells without GFP expression, cells with weak levels of GFP expression, and cells with strong levels of GFP expression). The acquisition of pOBCol3.6GFP is a continuous dynamic process and a function of time. Our time-course observations on pOBCol3.6GFP indicate that the initial expression expands and intensifies with a pattern such that the previous weak levels of expression advance to a stronger level and become surrounded by the newly emerging areas with weak levels of expression. Areas with strong levels of GFP expression were mostly confined to the mineralized portions of a nodule that can be labeled with XO. A QuickTime movie of the process can be viewed at (click Image Center). In contrast to pOBCol3.6GFP with different level of expression, pOBCol2.3GFP was activated uniformly at a strong level of expression. Cells expressing pOBCol2.3GFP were restricted to the nodules that eventually become mineralized as indicated by XO staining. The close correlation between GFP expression and XO staining suggests that pOBCol2.3GFP is an excellent visual marker for monitoring osteoblast differentiation in real time.

Five levels of increasing differentiation can be appreciated in the present primary calvarial culture model in real time: pOBCol3.6GFP negative [RIGHTWARDS ARROW] pOBCol3.6GFP positive (weak level) [RIGHTWARDS ARROW] pOBCol3.6GFP positive (strong level) [RIGHTWARDS ARROW] pOBCol2.3GFP positive [RIGHTWARDS ARROW] pOBCol2.3GFP positive associated with mineralization (Fig. 6). The initiation in weak levels of pOBCol3.6GFP marks the commitment of multipotential osteoprogenitors into the osteoblast lineage at the preosteoblast stage. Microarray data indicates that the acquisition of weak levels of pOBCol3.6GFP is associated with a suppression of numerous muscle genes involved with myofibroblasts, suggesting that the differentiation potential of this lineage is being restrained.(34) Because strong levels of pOBCol3.6GFP do not correlate completely with the XO-labeled nodules, there is the possibility that cells expressing strong levels of pOBCol3.6GFP may not yet achieved the ability to deposit mineral, or they have entered a non-osteoblast lineage. In addition, our data suggest that the expression of pOBCol2.3GFP is activated initially in differentiated osteoblasts located in presumptive mineralized nodules, which is consistent with microarray data on isolated pOBCol2.3GFP+ cells showing dramatic upregulation of differentiation-associated genes.(34) The interesting time lag between strong levels of GFP expression and XO staining in pOBCol2.3GFP and pOBCol3.6GFP cultures (Figs. 2A and 2B) indicates cells with activated pOBCol2.3GFP are more advanced in osteoblast lineage and require less time to mineralize compared with the cells accumulating strong levels of pOBCol3.6GFP.

thumbnail image

Figure FIG. 6. Expression of fluorescent markers and proposed change in osteogenic cell population during progression of the osteoblast lineage. No GFP is activated in the multipotential osteoprogenitors (MOPs). Expression of weak levels of pOBCol3.6GFP is initiated in the pre-osteoblasts (PreOB), and the intensity is elevated to strong levels in differentiating osteoblasts. Expression of pOBCol2.3GFP starts in the mature osteoblast and persists over time. The end product of osteoblast differentiation in vitro is the mineralized bone nodules that can be labeled with XO. Progenitors and less differentiated cells continue to proliferate, but only a portion of the cells in each stage differentiates and advances to the next stage. We propose that transient PTH treatment commits a greater proportion of progenitor cells to the osteoblast lineage, such that more cells undergo full osteoblast differentiation and deposit mineralized nodules. The gray area indicates a portion of total population that has the potential or ability to secrete bone matrix at the end of culture. The hatched area indicates an increase in osteoblast differentiation as a result of transient PTH treatment.20

Download figure to PowerPoint

Techniques of digital analysis have been applied to images of terminated cultures stained for ALP activity or von Kossa to quantitate the number of ALP+ or mineralized nodules.(35-38) In this study, we have shown that the size of area with GFP expression is a function of osteoblast differentiation and the quantitation of GFP can be a real time indicator for osteoblast differentiation. In addition, our image analysis model showed the use of XO to quantitate the mineralization in cultures without termination of cultures. XO has been used as an in vivo fluorescent dye to mark the newly mineralized tissue in living animals at the time of administration.(30,31) However, it has never been used to label mineralized nodules in the primary osteoblast cultures. Currently, von Kossa staining remains the routine standard method for labeling mineralized nodules, with a drawback that the cultures need to be terminated for the subsequent staining. Our result showed the red fluorescent stains of XO correlated exactly with the dark silver stains of von Kossa without deleterious effect on cells, suggesting that XO can be used as a real time fluorescent dye to mark the mineralized nodule in cultures.

In this study, PTH was used to validate the advantages of GFP culture and image analysis in assessing osteoblast differentiation. Our results showed that the inhibitory and stimulatory effect of PTH can be assessed through the quantitative analysis of GFP expression. Although it suppressed the expression of GFP during the phase of direct exposure, transient PTH treatment did accelerate the acquisition of GFP after the removal of PTH and resulted in a larger GFP expression area. The accelerated and enhanced osteogenesis represented by GFP and mineralization (XO and von Kossa stainings) implicates an increase in differentiated osteoblasts by the previous PTH treatment. At the cellular level, this increase may have resulted from two fundamentally different mechanisms: (1) an increase in the total number of progenitor cells either by enhanced proliferation or diminished apoptosis, in either case, more progenitor cells will be available to differentiate into osteoblasts, or (2) no increase in the total number of progenitors but a greater proportion of progenitor cells that achieve full osteoblast differentiation. Our result showed that, at day 7, the total DNA content in the transient PTH-treated culture was less than that in control culture, suggesting that the anabolic effect of PTH observed at the end of culture may have resulted from an increasing proportion of multipotential progenitor cells committed to osteoblast lineage rather than increased cell proliferation.

Depending on the culture systems and conditions used, in vitro studies have shown inconsistent results in the effect of PTH on cell proliferation. On the other hand, an increasing body of in vivo evidence indicates the anabolic effect of PTH may arise from the alteration of osteoblast differentiation rather than proliferation.(15,16) In vivo labeling studies have shown no evidence of increased osteoprogenitor proliferation stimulated by PTH.(39,40) At the cellular level, the anabolic effect of PTH at least can be partly achieved by maintaining the number of osteoblasts through the inhibition of apoptosis, which prolongs the longevity of osteoblasts, or increasing the number of osteoblasts through the activation of bone lining cells on quiescent bone surface.(40-42) Our results suggest in the presence of PTH, osteoblast differentiation is suppressed and maintained at the stages of osteoprogenitors and preosteoblasts. On the removal of PTH, osteoblast differentiation is recovered, accelerated, and enhanced. PTH treatment may not only reprogram the multipotential progenitors to promote the osteoblast commitment but also enhance the capacity of osteoblast differentiation once PTH is removed.

Altogether the approach of GFP-marked cultures combined with image analysis has shown the advantage of continuously monitoring osteoblast differentiation in living cultures in real time. Using this approach, the diverse effect of PTH on osteoblast differentiation was shown successfully. We suggest that the anabolic effect of PTH may arise from enhanced commitment of cells into the osteoblast lineage. Our ultimate goal is to develop a culture model with multiple differentiation stage-specific GFP transgenes to mark different developmental stages of osteoblast differentiation and use these markers to isolate subpopulations of cells by FACS for further cellular and molecular analysis.


  1. Top of page
  2. Abstract
  7. Acknowledgements

The authors thank Drs B Kream and M Kronenberg for helpful discussion on manuscript and Dr W Zhang for assistance in statistical analysis. This study was supported by NIH grants K12-HD01409 and R03-DE015224. Y-HW was selected as a “Building Interdisciplinary Research Careers in Women's Health” (BIRCWH) Scholar, under an award from the Office of Research on Women's Health and the National Institute of Child Health and Human Development.

The authors have no conflict of interest.


  1. Top of page
  2. Abstract
  7. Acknowledgements
  • 1
    Liang JB, Stein GS 2001 Osteoblast biology. In: MarcusR, FeldmanD, KelseyJ (eds.) Osteoporosis, 2nd ed., vol. 1. Academic Press, San Diego, CA, USA, pp. 2171.
  • 2
    Aubin JE, Triffitt JT 2002 Mesenchymal stem cells and osteoblast differentiation. In: BilezikianJP, RaiszLG, RodanGA (eds.) Principles of Bone Biology, 2nd ed., vol. 1. Academic Press, San Diego, CA, USA, pp. 5981.
  • 3
    Aubin JE 1998 Bone stem cells. J Cell Biochem Suppl 30–31: 7382..
  • 4
    Simmons PJ, Torok-Storb B 1991 Identification of stromal cell precursors in human bone marrow by a novel monoclonal antibody, STRO-1. Blood 78: 5562.
  • 5
    Gronthos S, Graves SE, Ohta S, Simmons PJ 1994 The STRO-1+ fraction of adult human bone marrow contains the osteogenic precursors. Blood 84: 41644173.
  • 6
    Bruder SP, Ricalton NS, Boynton RE, Connolly TJ, Jaiswal N, Zaia J, Barry FP 1998 Mesenchymal stem cell surface antigen SB-10 corresponds to activated leukocyte cell adhesion molecule and is involved in osteogenic differentiation. J Bone Miner Res 13: 655663.
  • 7
    Joyner CJ, Bennett A, Triffitt JT 1997 Identification and enrichment of human osteoprogenitor cells by using differentiation stage-specific monoclonal antibodies. Bone 21: 16.
  • 8
    Aubin JE, Turksen K 1996 Monoclonal antibodies as tools for studying the osteoblast lineage. Microsc Res Tech 33: 128140.
  • 9
    Gronthos S, Zannettino AC, Graves SE, Ohta S, Hay SJ, Simmons PJ 1999 Differential cell surface expression of the STRO-1 and alkaline phosphatase antigens on discrete developmental stages in primary cultures of human bone cells. J Bone Miner Res 14: 4756.
  • 10
    Misteli T, Spector DL 1997 Applications of the green fluorescent protein in cell biology and biotechnology. Nat Biotechnol 15: 961964.
  • 11
    Hadjantonakis AK, Macmaster S, Nagy A 2002 Embryonic stem cells and mice expressing different GFP variants for multiple non-invasive reporter usage within a single animal. BMC Biotechnol 2: 11.
  • 12
    Hadjantonakis AK, Nagy A 2000 FACS for the isolation of individual cells from transgenic mice harboring a fluorescent protein reporter. Genesis 27: 9598.
  • 13
    van Roessel P, Brand AH 2002 Imaging into the future: Visualizing gene expression and protein interactions with fluorescent proteins. Nat Cell Biol 4: E15E20.
  • 14
    Kalajzic I, Kalajzic Z, Kaliterna M, Gronowicz G, Clark SH, Lichtler AC, Rowe D 2002 Use of type I collagen green fluorescent protein transgenes to identify subpopulations of cells at different stages of the osteoblast lineage. J Bone Miner Res 17: 1525.
  • 15
    Aubin JE, Heersche JNM 2001 Cellular actions of parathyroid hormone on osteoblast and osteoclast differentiation. In: BilezikianJP, MarcusR, LevineMA (eds.) The Parathyroids, 2nd ed. Academic Press, San Diego, CA, USA, pp. 199211.
  • 16
    Hock JM, Fitzpatrick LA, Bilezikian JP 2002 Actions of parathyroid hormone. In: BilezikianJP, RaiszLG, RodanGA (eds.) Principles of Bone Biology, 2nd ed., vol. 1. Academic Press, San Diego, CA, USA, pp. 463481.
  • 17
    Tam CS, Heersche JN, Murray TM, Parsons JA 1982 Parathyroid hormone stimulates the bone apposition rate independently of its resorptive action: Differential effects of intermittent and continuous administration. Endocrinology 110: 506512.
  • 18
    Delling G 1987 Bone morphology in primary hyperparathyroidism - a qualitative and quantitative study of 391 cases. Appl Pathol 5: 147159.
  • 19
    Silverberg SJ, Shane E, de la Cruz L, Dempster DW, Feldman F, Seldin D, Jacobs TP, Siris ES, Cafferty M, Parisien MV, Lindsay R, Clemens TL, Bilezikian JP 1989 Skeletal disease in primary hyperparathyroidism. J Bone Miner Res 4: 283291.
  • 20
    Parisien M, Silverberg SJ, Shane E, de la Cruz L, Lindsay R, Bilezikian JP, Dempster DW 1990 The histomorphometry of bone in primary hyperparathyroidism: Preservation of cancellous bone structure. J Clin Endocrinol Metab 70: 930938.
  • 21
    Uzawa T, Hori M, Ejiri S, Ozawa H 1995 Comparison of the effects of intermittent and continuous administration of human parathyroid hormone(1–34) on rat bone. Bone 16: 477484.
  • 22
    Hock JM, Gera I 1992 Effects of continuous and intermittent administration and inhibition of resorption on the anabolic response of bone to parathyroid hormone. J Bone Miner Res 7: 6572.
  • 23
    Wronski TJ, Yen CF, Qi H, Dann LM 1993 Parathyroid hormone is more effective than estrogen or bisphosphonates for restoration of lost bone mass in ovariectomized rats. Endocrinology 132: 823831.
  • 24
    Lane NE, Thompson JM, Strewler GJ, Kinney JH 1995 Intermittent treatment with human parathyroid hormone (hPTH[1–34]) increased trabecular bone volume but not connectivity in osteopenic rats. J Bone Miner Res 10: 14701477.
  • 25
    Lane NE, Kimmel DB, Nilsson MH, Cohen FE, Newton S, Nissenson RA, Strewler GJ 1996 Bone-selective analogs of human PTH(1–34) increase bone formation in an ovariectomized rat model. J Bone Miner Res 11: 614625.
  • 26
    Bellows CG, Ishida H, Aubin JE, Heersche JN 1990 Parathyroid hormone reversibly suppresses the differentiation of osteoprogenitor cells into functional osteoblasts. Endocrinology 127: 31113116.
  • 27
    Isogai Y, Akatsu T, Ishizuya T, Yamaguchi A, Hori M, Takahashi N, Suda T 1996 Parathyroid hormone regulates osteoblast differentiation positively or negatively depending on the differentiation stages. J Bone Miner Res 11: 13841393.
  • 28
    Ishizuya T, Yokose S, Hori M, Noda T, Suda T, Yoshiki S, Yamaguchi A 1997 Parathyroid hormone exerts disparate effects on osteoblast differentiation depending on exposure time in rat osteoblastic cells. J Clin Invest 99: 29612970.
  • 29
    Wong GL, Cohn DV 1975 Target cells in bone for parathormone and calcitonin are different: Enrichment for each cell type by sequential digestion of mouse calvaria and selective adhesion to polymeric surfaces. Proc Natl Acad Sci USA 72: 31673171.
  • 30
    Rahn BA, Perren SM 1971 Xylenol orange, a fluorochrome useful in polychrome sequential labeling of calcifying tissues. Stain Technol 46: 125129.
  • 31
    Horobin RW 2002 Hydroxy triarylmethanes. In: HorobinRW, KiernanJA (eds.) Conn's Biological Stains, 10th ed. BIOS Scientific Publishers, Oxford, UK, pp. 203217.
  • 32
    Majeska RJ, Gronowicz GA 2002 Current methodologic issues in cell and tissue culture. In: BilezikianJP, RaiszLG, RodanGA (eds.) Principles of Bone Biology, 2nd ed., vol. 2. Academic Press, San Diego, CA, USA, pp. 15291541.
  • 33
    Purpura KA, Zandstra PW, Aubin JE 2003 Fluorescence activated cell sorting reveals heterogeneous and cell non-autonomous osteoprogenitor differentiation in fetal rat calvaria cell populations. J Cell Biochem 90: 109120.
  • 34
    Kalajzic I, Stahl A, Yang W, Wu S, Johnson S, Feyen J, Rowe D 2003 Microarray analysis of subpopulations of cells within differentiating murine osteoblast cultures. J Bone Miner Res 18: S2; S139.
  • 35
    Nefussi JR, Ollivier A, Oboeuf M, Forest N 1997 Rapid nodule evaluation computer-aided image analysis procedure for bone nodule quantification. Bone 20: 516.
  • 36
    Benestad HB, Sorensen T, Iranpour KM, Liestol K, Yogesan K, Strom-Gundersen I, Wang X, Lovhaug D 1998 Digital image analysis of hematopoietic colonies in vitro. Exp Hematol 26: 936941.
  • 37
    Martin I, Mastrogiacomo M, De Leo G, Muraglia A, Beltrame F, Cancedda R, Quarto R 2002 Fluorescence microscopy imaging of bone for automated histomorphometry. Tissue Eng 8: 847852.
  • 38
    Purpura KA, Aubin JE, Zandstra PW 2003 Two-color image analysis discriminates between mineralized and unmineralized bone nodules in vitro. Biotechniques 34: 11881198.
  • 39
    Onyia JE, Bidwell J, Herring J, Hulman J, Hock JM 1995 In vivo, human parathyroid hormone fragment (hPTH 1–34) transiently stimulates immediate early response gene expression, but not proliferation, in trabecular bone cells of young rats. Bone 17: 479484.
  • 40
    Dobnig H, Turner RT 1995 Evidence that intermittent treatment with parathyroid hormone increases bone formation in adult rats by activation of bone lining cells. Endocrinology 136: 36323638.
  • 41
    Jilka RL, Weinstein RS, Bellido T, Roberson P, Parfitt AM, Manolagas SC 1999 Increased bone formation by prevention of osteoblast apoptosis with parathyroid hormone. J Clin Invest 104: 439446.
  • 42
    Hock JM 1999 Stemming bone loss by suppressing apoptosis. J Clin Invest 104: 371373.