SEARCH

SEARCH BY CITATION

Keywords:

  • Colony forming unit-osteoblast;
  • Expansion;
  • Osteoprogenitor;
  • Dexamethasone;
  • Self-renewal

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Osteogenic cells are an integral part of the dynamic tissue-remodeling process in bone and are potential tools for tissue engineering and cell-based therapies. We examined the role of glucocorticoids and cell density in the expansion of primary rat calvaria cell populations and osteoprogenitor subpopulations in adherent cell culture. Osteoprogenitor response to dexamethasone (dex, a synthetic glucocorticoid known to stimulate bone formation in vitro) supplementation and long-term osteoprogenitor cell proliferation and differentiation were quantified using functional (colony forming unit-osteoblast [CFU-O]) and phenotypic analyses. Although osteoprogenitor self-renewal occurred at both standard and high initiating cell densities, progenitor cell expansion (measured by changes in CFU-O number relative to input) was sustained and dramatically increased at high initiating cell densities (30-fold CFU-O expansion for standard-density cultures compared with a greater than 10,000-fold CFU-O expansion in high-density cultures). Cell density was also found to impact upon the potential of dex to recruit additional progenitors towards bone development. These multifaceted effects appeared to be independent of cell proliferation rates or population phenotypic expression. Together, our results emphasize a roll for cell-cell interactions and/or community effects in the control and maintenance of progenitor cells during in vitro culture.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Osteogenic cells are an integral part of the dynamic tissue-remodeling process in bone, and as such are potential tools for tissue engineering and cell-based therapies for skeletal pathologies. Osteoblast activities including bone matrix synthesis and mineralization are regulated by various local and systemic factors. Osteoblasts express alkaline phosphatase (ALP) and synthesize type I collagen and noncollagenous proteins such as osteopontin, osteonectin, osteocalcin, bone sialoprotein, and bone morphogenetic proteins [15]. In addition, they respond to osteotropic hormones, including parathyroid hormone (PTH), prostaglandin E2, 1,25 dihydroxyvitamin D3, and glucocorticoids such as corticosterone and dexamethasone (dex) [610].

Primary bone or bone marrow chicken-, human-, mouse-, and rat-derived cell cultures have been used for studies on the metabolism and hormonal responses of osteogenic cells in vitro [1015]. Like those of the fetal or newborn rat calvaria (RC) [1619], these osteogenic cells have the capacity to form discrete three-dimensional bone nodules with the histological, ultrastructural, and immunohistochemical characteristics of woven bone when grown in the presence of ascorbic acid and organic phosphate [11, 2022]. These nodules are thought to represent the end stage of the proliferation-differentiation sequence of individual osteoprogenitor cells present in the input cell population. Limiting dilution and clonal analysis have shown that bone nodules can arise from a single progenitor and thus comprise a colony forming unit-osteogenic (CFU-O) assay [23], with ∼0.3% of the RC cell population capable of nodule formation under standard culture conditions [11, 24]. Currently, however, the unambiguous identification and isolation of osteoprogenitor cells within mixed bone cell populations is a problem, as no unique morphological or biochemical criteria are known by which to define the cells. The CFU-O assay thus represents a powerful way to interrogate the effect of exogenous and endogenous parameters on in vitro bone development.

Nodule formation in vitro can be regulated by hormonal additives to the culture medium, such as the synthetic glucocorticoid dex, which increases nodule numbers in a dose-dependent manner [25]. Evidence suggests that the circulating levels of glucocorticoids in humans and animals are similar to the concentrations that have maximal stimulatory effects on in vitro osteoblast responses, i.e., 10−7–10−8 M [26]. In the RC system, the subpopulation of progenitor cells requiring dex to be recruited into or differentiate along the osteoblast lineage are considered more primitive than those that undergo differentiation in control medium [27]. Dex also appears to recruit or promote maturation of more primitive human osteoprogenitors [2830]. It has been suggested that dex may regulate osteoblast differentiation positively or negatively, depending on the differentiation stage of the cells [16, 26, 31].

Previously, we examined the ability of dex to increase the proliferative capacity and differentiation potential of CFU-O capable of forming bone nodules through several passages [32]. We have now extended the analysis to longer times and compared bone progenitor expansion at low versus high initiating cell densities, the latter reported to maintain mineralized nodule formation for longer times in the rat bone marrow system [33]. We have found that CFU-O expansion is dex and cell density dependent, with a greater than 10,000-fold expansion observed under high-density conditions. Our results suggest that osteoprogenitor self-renewal, recruitment, and differentiation in response to dex is cell density dependent, with the hormone differentially influencing the proliferative capacity or onset of differentiation of a subpopulation of cells. The data also support the concept that progenitors at different stages of osteoblast differentiation exhibit selective responsiveness to hormone actions [34, 35]. Evidence of extensive bone progenitor self-renewal under multiple conditions emphasizes the role that cell-cell interactions or community effects play in revealing osteogenic potential and ultimately the ability to control bone development for therapeutic application.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Rat Calvaria Cell Isolation

Fetal RC cell populations were obtained by sequential enzymatic digestion of calvariae from 21-day-old fetuses of timed pregnant Wistar rats as previously described [36]. Cells from the last four digest steps (pop II-V) were plated for 24 hours in separate T-75 flasks containing α-minimum essential medium (α-MEM) from the Tissue Culture Media Prep (TCMP) core at the University of Toronto, 15% heat-inactivated fetal bovine serum (FBS) purchased from CanSera (Toronto, Ontario; http://www.cansera.ca), and the following antibiotics: 0.25 μg/ml fungizone and 50 μg/ml gentamycin sulfate (both from GIBCO/BRL; Burlington, Ontario; http://www.invitrogen.com), and 0.1 mg/ml penicillin G (Sigma Chemical Co.; St. Louis, MO; http://www.sigmaaldrich.com).

RC Cell Expansion

The general protocol used for expansion and assay of osteogenic capacity is outlined in Figure 1A.

thumbnail image

Figure Figure 1.. Schematic of the protocol used for subculturing and CFU-O assay.A) As described inMaterials and Methods, primary RC cells were plated at either 1,250 cells/cm2in 35-mm dishes (standard density) or at 12,500 cells/cm2in 60-mm dishes (high density). Cultures were subcultured at a 1:6-1:3 ratio every 3 days and CFU-O number (in either vehicle or dex) was assayed at each subculture. B) Representative dishes showing results at day 21 of a CFU-O assay in which cells were plated in dex at high density (day 0) and subcultured on day 3 into vehicle or dex assay conditions.

Download figure to PowerPoint

Standard Density

One day after isolation from calvariae, cells were trypsinized and viable cells were counted and plated in standard medium (α-MEM with 10% FBS and antibiotics [as above]) into six 35-mm dishes at 1.2 × 104 cells/dish (n = 3) for continuous subculture and expansion. After 24 hours, standard medium was replaced with differentiation medium (standard medium supplemented with 50 μg/ml ascorbic acid [Fisher Scientific; Hampton, NH; http://www.fisherscientific.com]) and 10 mM β-glycerophosphate (Sigma) and vehicle (0.11% anhydrous ethyl alcohol) or 10−8 M dex (both from Sigma). This is the maximum effective dose for dex in the RC cell system, established from complete dose-response curves [25].

Cells were grown in a 37°C humidified atmosphere containing 5% CO2 and were maintained in the exponential growth phase through continuous subculture as previously described [32]. Briefly, on day 3, nonadherent cells were removed and treated cells were detached with 0.2% trypsin (GIBCO/BRL) in citrate saline (TCMP) for 3–5 minutes at 37°C. Following centrifugation, the cells were plated into new 35-mm dishes at a 1:3 dilution with fresh medium containing supplements and vehicle or dex; this process was repeated every third day for at least 10 subcultures. Cell number was determined with a Beckman Coulter Z1 cell and particle counter (Mississauga, Ontario; http://www.beckmancoulter.com) at each subculture and medium was exchanged every 2 days.

High Density

To examine effects of density, cells were also plated into a 60-mm dish at 3.5 × 105 cells/dish (n = 3) in standard medium. After 24 hours, medium was replaced with differentiation medium and vehicle or 10−8 M dex. As described above, cells were maintained in the exponential growth phase by subculturing into new dishes every 3 days, starting on the third day of culture. For high-density cultures, a 1:1 mixture of 0.4% trypsin and collagenase was required to generate a single cell suspension. After centrifugation, cells were replated at dilutions ranging from 1:6 to 1:3, into either one or two 60-mm plates for continued expansion under vehicle or dex treatment conditions.

In both standard and high-density conditions, cells were monitored for surface expression of ALP by flow cytometry during expansion, and subculture was continued until nodules no longer formed in the osteogenic assay (maximum of 20 subcultures were completed). Cell numbers were also determined at each subculture. The population doubling (PD) was calculated for each 3-day growth period (i.e., PD = log(N/No)/log2, where No is the seeded cell density and N is the cell density after 3 days). The cumulative population doubling level (CPDL) was determined by summing the population doublings between each subculture.

CFU-O Assay

The osteogenic potential of the expanded populations was assessed by taking aliquots of the cells recovered at each subculture of low- or high-density cultures and replating them into 35-mm dishes in standard medium at the same density used for expansion. Medium was changed 24 hours later to differentiation medium as above with vehicle or dex (standard density, n = 6; high density, n = 3 for each treatment) and then every 2 days until day 21.

Mineralized bone nodule formation was assessed at day 21 by double labeling for mineral (von Kossa stain) and ALP [37]. The number of bone nodules was determined by manual counting over a grid on a dissection microscope or by image analysis. For the latter, entire culture dishes were scanned in color using a flatbed scanner (AGFA) at a resolution of 800–1,200 ppi. To automate analysis, an image database was compiled with Image-Pro Plus software (Media Cybernetic, LP; Silver Spring, MD; http://www.mediacy.com) [38].

Flow Cytometry

Cell surface expression of ALP and parathyroid hormone/parathyroid hormone-related protein receptor (PTH1R) was monitored at alternating subculture time points during one of the high-density expansion experiments. Briefly, an aliquot of cells recovered on day 6 (and every second subculture thereafter) was resuspended in ice cold Hanks-buffered salt solution containing 2% FBS (HF) and incubated at ≤107 cells/ml for 45 minutes at 4°C, with the mouse monoclonal antibody recognizing rat bone/liver/kidney ALP (RMB211.13) at a 1:600 dilution in HF, described by Tursken and Aubin [27], and the rabbit polyclonal antibody against PTH1R (rat pep IV; generated from a sequence in the first extracellular loop region: CTLDESARLTEEELH) from Covance Research Products, Inc. (Princeton, NJ; http://www.covance.com) at 1:50 or 1:100 dilution, depending on the lot. Cells were then washed twice in HF and incubated for 45 minutes at 4°C with the secondary antibodies fluorescein isothiocyanate-conjugated donkey anti-mouse IgG at 1:200 (H+L) and R-phycoerythrin-conjugated AffiniPure F(ab′)2 fragment donkey anti-rabbit IgG (H+L) at 1:100 or 1:200 (Jackson ImmunoResearch Laboratories, Inc.; West Grove, PA; http://www.jacksonimmuno.com). Following two more HF washes, samples were resuspended in HF with 1 μg/ml 7-AAD (Molecular Probes, Inc.; Eugene OR; http://www.probes.com), and viable cells (7-AAD) were analyzed on EPICS XL flow cytometer (Beckman Coulter). Negative controls contained only the secondary antibody.

Statistical Analysis

The standard deviation of nodule numbers for all densities and conditions tested was determined (triplicate samples) and shown graphically as error bars. Trends were consistent between triplicate experiments; however, data were not pooled due to absolute number variation between the experiments. p values for comparison of mean nodule output or fold changes within an experiment were calculated using a t- or z-test of significance (p < 0.05; 2-tail).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Dexamethasone Supplementation Affects the Long-Term Growth of Osteoprogenitor Cells

To examine the long-term growth and self-renewal capacity of osteoprogenitors within the heterogeneous RC cell population, cultures were maintained in exponential growth by subculturing the cells at a 1:6 to 1:3 ratio every 3 days. In addition, two cell densities were investigated to examine the impact of the microcellular environment on the long-term self-renewal capabilities of progenitor cells. Cells cultured with this protocol (Fig. 1A) at standard density (Fig. 2A, 2B) and high density (Fig. 2C, 2D) proliferated extensively. Calculation of the CPDL demonstrated that cells in both treatment conditions grew with a constant and similar doubling time during the exponential growth phase at standard (Fig. 2B) and high (Fig. 2D) density, as demonstrated by linear regression of the population doubling with subculture (Table 1). In some experiments, growth crisis was observed in vehicle-treated cultures after ∼10–20 or ∼30 population doublings at standard or high density, respectively; in dex, however, exponential growth continued until culture termination (Table 1). Cell morphology (polygonal) was similar in vehicle and dex treatment during exponential growth, but as cultures approached crisis, cells assumed a more fibroblastic and then typically senescent shape (data not shown).

Table Table 1.. Summary of growth and osteoblast proliferation and differentiation capacities of triplicate experiments completed at two seeding densities
  1. a

    The maximum CPDL of the vehicle- (veh) and dex-treated populations is listed at the day cell number plateaued (†) or the experiment was terminated. The average growth rate of the three experiments was calculated from the CPDL and is designated as the doublings/day (DL/day). Capacity for osteoblast differentiation (ability to form bone nodules) in each expanded population was assayed in vehicle (*) and dex (**), with the last subculture day at which nodules were detected for each condition tabulated. In addition, the long-term proliferative capacity of osteoprogenitors is indicated by the maximum fold nodule expansion (compared with nodule output at subculture day 3) and the day this maximum was achieved.

  2. b

    Population plateau; end of exponential growth.

  3. c

    Maximum fold expansion relative to day 3.

  4. d

    *Expanded culture assayed in vehicle.

  5. e

    **Expanded culture assayed in dex.

  
 Expansion mediaVehDexVehDexVehDex
 Experiment      
Standard seeding density (1,250 cells/cm2)118.0 (30)22.7 (33)12, 2118, 241.0 ±.2 (6), 4.9 ± 1.2 (9)2.3 ± 2.2 (18), 12.5 ± 3.3 (15)
 216.3 (40)24.0 (47)17, 1717, 250.4 ± 0.3 (17), 1.9 ± 0.3 (10)0.8 ± 0.3 (10), 15.2 ± 5.1 (17)
 39.8 (27)14.2 (30)6, 99, 150.2 ± 0.1 (3), 0.6 ± 0.4 (6)0.5 ± 0.3 (6), 0.5 ± 0.2 (9)
Average growth rate (DL/day) 0.58 ± 0.160.63 ± 0.11    
 Expansion mediaVehDexVehDexVehDex
 Experiment      
High seeding density (12,500 cells/cm2)130.8 (54)38.4 (60)15, 2118, 242.5 ± 0.1 (9), 8.9 ± 0.7 (15)960 ± 70 (18), 8,000 ± 400 (24)
 222.0 (39)19.9 (30)21, 2130, 302.5 ± 0.2 (9), 22.9 ± 3.8 (12)13,000 ± 4,000 (27), 57,000 ± 8,000 (30)
 319.9 (36)21.4 (36)18, 1830, 3020.8 ± 2.6 (9), 48.6 ± 18.3 (15)31,000 ± 4,000 (30), 26,000 ± 1,000 (30)
Average growth rate (DL/day) 0.58 ± 0.010.63 ± 0.02    
thumbnail image

Figure Figure 2.. Population growth characteristics of RC cells.Primary RC cells were seeded at (A) standard (∼1,250 cells/cm2in 35-mm dish) or (C) high (∼12,500 cells/cm260-mm dish) density, expanded in vehicle () or dex () and split at a 1:6-1:3 ratio every 3 days. The CPDL achieved is plotted for the cultures initiated at standard (B) or high (D) density, respectively, with the linear regression of the curves overlaid (line). A representative example for each initiating density of three independent experiments is shown.

Download figure to PowerPoint

Dexamethasone Supplementation and Density Affect the Maintenance of the Differentiation Capacity of Osteoprogenitor Cells

At each passage, the presence and number of CFU-O were assayed by replating the expanded cell populations (vehicle- or dex-treated) into either vehicle or dex conditions to track progenitor self-renewal and the differentiation or recruitment effects of dex. CFU-O was defined as cells giving rise to colonies with cuboidal cells that were ALP positive and containing mineralized matrix (von Kossa positive), i.e., bone nodules (Fig. 1B). Treated cultures initiated at standard density and expanded in vehicle (Fig. 3A) or dex (Fig. 3B) generated detectable bone nodules for 5–8 subcultures in both vehicle and dex assay conditions. Based on the assumption that one osteoprogenitor cell gives rise to one nodule [24], a theoretical yield of bone nodules was calculated based on input progenitor numbers (as observed in CFU-O assay) and the dilution factor of each subculture. To emphasize graphically that progenitors self-renewed, the calculated yield was overlaid (lines) on the observed yield of nodules (Fig. 3A, 3B). The number of CFU-O did not exceed the theoretical yield in the expanded populations assayed in vehicle (Fig. 3A, 3B). In contrast, CFU-O number was significantly (p < 0.05; 2-tail z-test) higher than the theoretical yield when cells were assayed in dex on days 6 and 9 for populations expanded in vehicle and for all time points beyond 3 days (with the exception of day 12 and day 30 where the increase failed to reach statistical significance) for populations expanded in dex (Fig. 3A, 3B). Thus, self-renewal was observed for 4 (vehicle) or more than 12 (dex) population doublings for treated cells. Additionally, during the time that cells were capable of self-renewal, a population of CFU-O was recruited by dex in both culture treatments (p < 0.05; 2-tail t-test), yielding an ∼2–20- or ∼2–25-fold increase in CFU-O output in dex versus vehicle assay conditions (Fig. 3A, 3B).

thumbnail image

Figure Figure 3.. CFU-O number in sequentially passaged RC cells.The number of mineralized nodules after 21 days in vehicle (hatched) or dex (open) assay conditions from cell populations initiated at standard density and treated during exponential growth with vehicle (A) or dex (B). A representative experiment of three separate experiments is plotted and the mean number of nodules ± standard deviation of triplicate wells per assay condition is shown. Overlaid lines represent the expected osteoprogenitor dilution curve if progenitors did not proliferate in vehicle () or dex (). The significance (p < 0.05; t-test assuming unequal variances, 2-tail) of the separation between nodule formation in the dex assay following expansion in both conditions and the osteoprogenitor dilution curve is indicated (*). Nodule outputs of the vehicle and dex assay are significantly different (p < 0.05, t-test, 2-tail) until day 9 and until day 18 (with the exception of day 12) for vehicle- or dex-treated expansion cultures. Nd indicates that nodules were not detected.

Download figure to PowerPoint

To further quantify the osteoprogenitor pool with time, the fold change in CFU-O number was calculated at each subculture, relative to the number of CFU-O detected on the first subculture (day 3). Mineralized nodules were detectable for a limited number of passages under both cultivation conditions, but the numbers differed between conditions. In populations initiated at standard density that were passaged and assayed in vehicle (Fig. 4A, hatched bars), CFU-O steadily declined in number and could not be detected after 12 doublings (day 15). When the same population was assayed in dex (Fig. 4A, open bars), CFU-O number first increased then decreased to a plateau value prior to a loss of detection after 15 doublings (day 21). Strikingly, in populations initiated at standard density but passaged in dex (Fig. 4B), CFU-O numbers were higher and were maintained longer in both vehicle and dex assay conditions, with maximal numbers in dex assay conditions (Table 1). In cultures initiated at high density, nodules were detected for a greater number of subcultures for nearly all conditions (Fig. 4C, 4D; Table 1). Cells expanded in vehicle at high density demonstrated a higher level of self-renewal and progenitor maintenance over cells expanded at standard density, assayed in both vehicle (p < 0.05; 2-tail z-test; to day 9) and dex (to day 15, except day 9) conditions (Fig. 4A, 4C). Dramatically, high-density expansion in dex significantly (p < 0.05) amplified and sustained the CFU-O number measured in vehicle and dex assay conditions until just prior to or until culture termination (14–18 doublings) (Fig. 4D; Table 1). The CFU-O amplification was greater than 104-fold in the dex assay condition, significantly higher than levels previously reported.

thumbnail image

Figure Figure 4.. Fold expansion of CFU-O with successive passaging.Capacity for nodule production calculated as the fold expansion of mineralized nodules relative to day 3 (or day 0) nodule output for treatment with vehicle (A, C) or dex (B, D) when cells were initiated at standard (A, B) or high (C, D) density. CFU-O number was assayed in vehicle (hatched) or dex (open) conditions. A representative experiment at each density from three independent experiments is shown with the fold expansion ± standard deviation of the calculation based on population growth and nodule frequencies. Nd indicates that nodules were not detected; * indicates that vehicle and dex assay results for each expansion condition were significantly different (p < 0.05; z-test, 2-tail). A comparison of the fold expansion of nodules during standard and high-density culture showed significant differences (p < 0.05; z-test, 2-tail) to day 9 or day 15 (with the exception of day 9) in vehicle-treated cultures assayed in vehicle and dex, respectively, and differences to day 18 and day 24 for dex-treated cultures assayed in vehicle or dex.

Download figure to PowerPoint

Populations initiated at high density and expanded in dex exhibited similar CFU-O numbers in dex versus vehicle assay conditions for 5–7 passages (data not shown). Further examination of the steadily increasing nodule number that occurred with dex treatment was done by calculating the cumulative nodule doubling level (CNDL) in the vehicle and dex CFU-O assay and comparing it with the growth of the CPDL (Fig. 5). Replicate experiments showed that production of nodules occurred at an indistinguishable rate from the overall population growth, and thus neither progenitor cells nor the bulk fibroblastic cells exhibited a clear growth advantage.

thumbnail image

Figure Figure 5.. Comparison of nodule and population expansion in dex cultures initiated at high density.A plot of the CNDL in the vehicle () or dex () CFU-O assay and the CPDL is shown for one representative experiment. A reference line for a 1:1 ratio of CNDL:CPDL is shown.

Download figure to PowerPoint

Separation of Expression of the Osteoblastic Phenotype of Expanded Cells and Capacity for Nodule Formation

To examine the proportion of the population exhibiting markers of the osteoblastic phenotype, the high-density culture was characterized in terms of surface expression of ALP and PTH1R by flow cytometry at alternating subcultures (Fig. 6). There was a clear separation between nodule formation, ALP and/or PTH1R expression, and population growth in both expansion conditions. Both vehicle- and dex-expanded cultures stopped producing nodules during a time of stable cell expansion. CFU-O capacity was lost without an associated loss in ALP/PTH1R expression or population growth, and changes in measured marker expression were not coupled to population growth changes. In vehicle, an approximate 60:40 split of ALP+:ALP cells was seen (Fig. 6A), while in dex, an inverse 40:60 distribution was seen (Fig. 6B). Cells expressing ALP remained fairly stable over 16–17 doublings (day 24) in both vehicle and dex conditions (Fig. 6A, 6B), although populations expanded in vehicle demonstrated a reduced capacity for nodule formation in comparison to their dex-treated counterpart. The ALP+PTH1R+ population appeared in a cyclic fashion over time; however, changes in expression did not correlate to nodule production. At the time that 80%–100% of the total nodules had formed under vehicle conditions, cells assayed in dex had produced only 1%–10% of their total nodule capacity (days 12–18). The largest changes in ALP expression occurred after nodule formation was no longer observed, while ALP+PTH1R+ expression varied over the duration of culture. In the control population, the number of cells expressing ALP steadily declined following 16 doublings (day 24), and the dex population exhibited a reduced number of ALP+ cells prior to a later decline.

thumbnail image

Figure Figure 6.. Surface expression of ALP and PTH1R was monitored by flow cytometry during expansion of high-density cultures.The percentage of ALP positive cells () or ALP and PTH1R double positive cells () is shown for vehicle (A) and dex (B) expanded cells (right axis) for one representative experiment. A cumulative representation of the total nodule formation from the CFU-O assays in vehicle () or dex () is also overlaid (left axis). NS indicates that there was no sample available.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

In this study, we analyzed osteoprogenitor (CFU-O) self-renewal, recruitment, and differentiation in response to dex in RC cell populations. We found evidence of progenitor self-renewal in cultures initiated at both standard and high density. A fraction of progenitors was recruited by dex either to induce nodule formation or to complete the osteoblastic differentiation program, with the most robust effect in cultures initiated at standard density. In populations initiated at high density, the need for dex was reduced, suggesting that the dex-responsive fraction was either diminished compared with the fraction present at low density or was no longer dependent on dex to complete the differentiation process. Finally, populations subcultured at high density in dex exhibited an elevated capacity for self-renewal and overall CFU-O expansion. The osteoprogenitors produced were able to differentiate in vehicle or dex conditions equally well for five to seven passages, although dex stimulation was required for subsequent passages prior to a striking loss of nodule formation.

The self-renewal capacity of osteoprogenitors has been studied in rat calvaria [32] and bone marrow stromal [33] cell populations, as well as in human bone marrow stromal cell populations [3942]. Initiating density (103–5 × 104 cells/cm2) and subculturing regimes have varied, although a 1:3-1:2 dilution at subcultivation was typically completed every 3–5 days or once cultures were nearly confluent. In some experiments, ability to express multipotentiality (adipogenesis, chondrogenesis, osteogenesis) has been monitored, with a gradual loss in early progenitor properties reported and dependent on the conditions and duration of culture. However, optimized growth conditions for maximal expansion and maintenance of osteogenic potential have not been reported, and most studies lack quantitative assays to track osteoprogenitors in the population and the ability to characterize fold expansion. We were able to track osteoprogenitor expansion using RC cells coupled with functional assays.

Repeated passaging to cyclically reduce cell density and promote continued exponential growth demonstrated that primary RC cell populations are capable of extensive proliferation: more than 16 population doublings in vehicle and more than 20 doublings in dex. Consistent with the findings of an earlier study on the proliferation and differentiation of osteoprogenitor cells in the presence or absence of dex [32], CFU-O were not detectable beyond 18–20 population doublings. It is also of interest that in the hematopoietic system dex in combination with erythropoietin and stem cell factor was able to induce erythroid progenitors to undergo a similar number of cell divisions (15–22), corresponding to a 105–106-fold amplification before a loss in proliferation and differentiation capacity was measured [43]. Dex acts directly on human erythroid progenitors, maintaining their colony-forming capacity and delaying their differentiation, with the glucocorticoid receptor acting as a key regulator for the decision between self-renewal and differentiation [43, 44]. The loss experienced by both the osteogenic and hematopoietic systems may reflect an inherent limit on the capacity of these progenitors to expand long-term in vitro.

The progeny of osteoprogenitor cell divisions may not enter cycle or have sufficient proliferative capacity to elaborate a bone nodule with repeated subculture. A standard 21-day CFU-O assay in vehicle or dex was used to assess osteogenic capacity; however, it remains possible that a change in culture conditions or a longer culture time would allow nodule formation in the repeatedly passaged populations. For example, Owen et al. [45] reported that mature mineralized cultures of RC cells (after day 35) were capable of reinitiating expression of osteoblast phenotype markers and mineralizing under appropriate culture conditions following trypsinization and replating. These passaged cells exhibited a variable length of time to develop a mineralized extracellular matrix, i.e., fourth passage cells developed mineralized nodules 66 days after plating. However, in our study, it did not appear that cultures were stopped prematurely, as ALP expression and size of the nodules decreased with subculture (data not shown).

Generally, bone nodules were not detected following eight or nine subcultures, reflecting either a loss in the functional capacity of the proliferating cells, the dilution of a rare cell population past the detection limit of the assay, or the loss of stimulatory or gain of inhibitory accessory cells that influence CFU-O development cell nonautonomously [37]. Consistent with the possibility that loss of CFU-O reflects loss of functional capacity or a change in accessory cells is the observation that cultures initiated at high density demonstrated a striking loss in mineralized nodules from ∼2,500/35-mm culture dish to zero in one subculture. In addition, the rate of nodule production for high-density cultures matched the doubling rate of the mixed population, suggesting that at high density the loss of nodule formation did not result from the dilution of a rare cell type.

The effects of glucocorticoids on osteoblast activities in vitro are biphasic and appear dependent on concentration, duration of exposure, and the model investigated [25]. Generally, physiological levels of glucocorticoids appear necessary for osteoblast differentiation, while high levels of glucocorticoids reduce proliferation of osteoblast precursors [30, 46, 47]. Dex may have multiple effects on the RC cell population; it may selectively stimulate osteoprogenitor cell recruitment and proliferation, and additionally provide an advantage in the maintenance of functional capacity. The recruitment capability of dex was demonstrated as cells passaged at standard plating density in both of the treatments (vehicle or dex) and assayed for CFU-O produced more detectable nodules in dex than in vehicle. The long-term detection of CFU-O with dex treatment at high density likely results from proliferation, with osteoprogenitors dividing at a comparable rate to the entire population (Fig. 5). Initiating the cells at high density greatly magnified the stimulatory effects of dex on nodule formation and CFU-O expansion. Redistribution studies demonstrated that nodule-forming progeny of CFU-O were produced only during exponential growth in cultures without dex [32], while in its presence, they were produced both during the exponential growth phase and after confluency. This observation may explain the large induction that results with dex at high density, as these cultures typically reached confluence and began multilayering prior to subculture.

Initiating cells at high density (∼12,500 c/cm2) also eliminated some of the requirement for dex to recruit or stimulate nodule formation. This phenomenon was apparent for cells from early passages of both treatments and suggests that at high cell density, community effects or cell-cell interactions/ secreted factors may regulate osteogenesis or progenitor maintenance. Glucocorticoid treatment also maintained osteogenic capacity longer than vehicle treatment, suggesting that dex may also act directly and/or indirectly through accessory cells that provide favorable factors or cell-cell interactions. For example, in vivo, basic fibroblast growth factor (bFGF) stimulates endosteal bone formation, and, in vitro, bFGF cooperates with glucocorticoids to stimulate the formation of a bone-like matrix [48]. We have not yet systematically examined specific growth/soluble factors. However, in preliminary experiments in which medium collected after 2 days of conditioning from day-6 or day-30 expansion cultures was used 1:1 with fresh medium on primary RC cells, a consistent effect on nodule number or size was not observed (i.e., certain conditions influence nodule numbers while others influence total mineralized area) when compared with fresh medium alone (data not shown). Additionally, populations in which the PTH1R signaling pathways were stimulated during expansion via treatment with PTH (10−9 M) expressed the same density-dependent effects as seen here in vehicle and dex (data not shown), but highly increased expansion of progenitors in high-density versus standard-density cultures was not observed. More extensive studies are required to elucidate whether the cell density-dependent growth we observed is attributable to cell-cell contact or secreted factor(s), or both.

Identifying a relationship between population growth, nodule formation, and expression of surface or other osteoblast markers is the object of much study [15, 16, 28, 4952]. To this end, separation of the input RC population into four purified fractions (ALPPTH1R, ALPPTH1R+, ALP+PTH1R, ALP+PTH1R+) showed that nodules are able to form within each fraction [53]. However, as we demonstrate here, there is no direct correlation between ALP and/or PTH1R expression and nodule formation in the passaged populations, and nodule formation is lost even in populations in which expression of these markers persists. The inhibitory effects of fibroblastic cells on osteoprogenitor differentiation or osteoblast function in different cell systems have been reported previously [5456]. As surface markers remain limited, ongoing studies examining gene expression patterns may shed further light on these population dynamics.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

K.A.P. gratefully acknowledges student fellowship support from Ontario Graduate Scholarships in Science and Technology. This work was supported by operating grants from Canadian Institutes of Health Research (J.E.A., MT-12390), Premier's Research Excellence Awards (P.W.Z.), the Stem Cell Network (J.E.A.; P.W.Z.), and Natural Sciences and Engineering Research Council of Canada (P.W.Z.).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References