Dr Findlay serves as a consultant for the Australian Orthopaedic Innovations Medical Advisory Board. All other authors have no conflict of interest
RANKL Expression Is Related to the Differentiation State of Human Osteoblasts
Article first published online: 1 JUN 2003
Copyright © 2003 ASBMR
Journal of Bone and Mineral Research
Volume 18, Issue 6, pages 1088–1098, June 2003
How to Cite
Atkins, G. J., Kostakis, P., Pan, B., Farrugia, A., Gronthos, S., Evdokiou, A., Harrison, K., Findlay, D. M. and Zannettino, A. C. (2003), RANKL Expression Is Related to the Differentiation State of Human Osteoblasts. J Bone Miner Res, 18: 1088–1098. doi: 10.1359/jbmr.2003.18.6.1088
- Issue published online: 2 DEC 2009
- Article first published online: 1 JUN 2003
- Manuscript Accepted: 31 DEC 2002
- Manuscript Revised: 13 NOV 2002
- Manuscript Received: 18 SEP 2002
- vitamin D3;
Human osteoblast phenotypes that support osteoclast differentiation and bone formation are not well characterized. Osteoblast differentiation markers were examined in relation to RANKL expression. RANKL expression was induced preferentially in immature cells. These results support an important link between diverse osteoblast functions.
Cells of the osteoblast lineage support two apparently distinct functions: bone formation and promotion of osteoclast formation. The aim of this study was to examine the relationship between these phenotypes in human osteoblasts (NHBC), in terms of the pre-osteoblast marker, STRO-1, and the mature osteoblast marker, alkaline phosphatase (AP), and the expression of genes involved in osteoclast formation, RANKL and OPG. The osteotropic stimuli, 1α,25(OH)2vitamin D3 (vitD3) and dexamethasone, were found to have profound proliferative and phenotypic effects on NHBCs. VitD3 inhibited NHBC proliferation and increased the percentage of cells expressing STRO-1 over an extended culture period, implying that vitD3 promotes and maintains an immature osteogenic phenotype. Concomitantly, RANKL mRNA expression was upregulated and maintained in NHBC in response to vitD3. Dexamethasone progressively promoted the proliferation of AP-expressing cells, resulting in the overall maturation of the cultures. Dexamethasone had little effect on RANKL mRNA expression and downregulated OPG mRNA expression in a donor-dependent manner. Regression analysis showed that RANKL mRNA expression was associated negatively with the percentage of cells expressing AP (p < 0.01) in vitD3- and dexamethasone-treated NHBCs. In contrast, RANKL mRNA expression was associated positively with the percentage of STRO-1+ cells (p < 0.01). In NHBCs sorted by FACS based on STRO-1 expression (STRO-1bright and STRO-1dim populations), it was found that vitD3 upregulated the expression of RANKL mRNA preferentially in STRO-1bright cells. The results suggest that immature osteoblasts respond to osteotropic factors in a potentially pro-osteoclastogenic manner. Additionally, the dual roles of osteoblasts, in supporting osteoclastogenesis or forming bone, may be performed by the same lineage of cells at different stages of their maturation.
CELLS OF THE osteoblastic lineage fulfill a range of functions that include new bone formation during bone modeling and remodeling,(1) organic matrix degradation at the initiation of remodeling,(2) and support of hematopoiesis,(3) notably osteoclastogenesis.(4) The way that these functional states of osteoblasts relate to one another, particularly for human osteoblasts, is not understood. A useful tool with which to investigate osteoblastic phenotypes is the monoclonal antibody (Mab) STRO-1.(5) STRO-1 binds to an, as yet, uncharacterized cell surface antigen expressed by a subpopulation of adult and fetal human bone marrow mononuclear cells (BMMC), including all assayable colony forming units-fibroblast (CFU-F).(5, 6) These multipotent mesenchymal cells identifiable by STRO-1 have been shown to form an array of stromal cell types, including smooth muscle cells, fibroblasts, adipocytes, chondrocytes, and functional osteoblasts, under the appropriate culture conditions.(5–11) Our previous study described the use of STRO-1 to identify several maturation stages of osteoblastic cells from heterogeneous populations of cultured normal human trabecular bone-derived cells (NHBCs).(7) Used in concert with the antibody B-478, which binds to the bone/liver/kidney isoform of alkaline phosphatase (AP), STRO-1 subdivides NHBCs into four subpopulations that represent different stages of osteogenic maturation. The least mature osteogenic cells in these populations are STRO-1+/AP−, which give rise to the progressively more mature STRO-1+/AP+, STRO-1−/AP+, and STRO-1−/AP− cells.(7, 8)
Reports of specific links between the osteogenic and osteoclast inductive properties of osteoblasts are sparse, despite the long-held view that the processes of bone resorption and bone formation are intimately coupled.(12) While the NHBC subpopulations described above have been extensively characterized with respect to their osteogenic activity,(7, 8) little is known regarding the ability of these cells to support osteoclast formation. In fact, the identification of the human osteoblast phenotype that promotes osteoclastogenesis has been elusive, and an in vitro model for the differentiation of human osteoclasts using human osteoblasts as a stromal layer has not yet been described. As recently reviewed,(4, 13) a number of osteoclastogenic agents act by inducing the surface expression of the tumor necrosis factor (TNF)-ligand family member, RANKL. The induced expression of RANKL by osteoblastic cells allows the maturation, differentiation, and activation of osteoclasts.(4, 13–15) RANKL promotes osteoclast development and survival by binding to the TNF receptor superfamily member, RANK, on osteoclast precursors, with the required presence of M-CSF.(16) Deletion of the RANKL gene in mice resulted in severe osteopetrosis and a complete lack of osteoclasts as a result of an inability of osteoblasts to support osteoclastogenesis.(17) The soluble TNF-receptor family member, osteoprotegerin (OPG), is a natural RANKL antagonist, which can inhibit osteoclast formation and bone resorption.(18)
The role of RANKL during human bone remodeling is implied by the above mechanisms, but has been described largely in rodent systems. The relevance of RANKL expression in human bone was highlighted by our recent study, which showed that histomorphometric indices of bone remodeling, erosion surface (% ES/BS), and osteoid surface (% OS/BS) associate tightly with expression of RANKL mRNA, relative to OPG mRNA, in normal human trabecular bone.(19) However, the cell types responsible for RANKL expression were not identified in that study.
The expression of RANKL has recently been linked to the differentiation state of osteoblastic cells. A conditionally immortalized human marrow stromal cell line, hMS, was shown by Gori et al.(20) to support murine osteoclastogenesis. When these cells were induced to differentiate, their ability to support osteoclastogenesis was reduced. Significantly, the RANKL to OPG mRNA ratio was shown to decrease on their differentiation.(20) In primary mouse osteoblasts, although basal levels of RANKL mRNA did not change during mineralization, the response of more mature cells to vitD3, in terms of RANKL expression, was decreased.(21) VitD3 is commonly used to drive osteoclastogenesis in in vitro coculture models of osteoclastogenesis.(22–25) VitD3 is capable of directly inducing RANKL transcription(14, 16, 26) by virtue of vitamin D response elements (VDRE) in the RANKL promoter.(27) Similarly, the glucocorticoid dexamethasone is also used to enhance osteoclast formation. Glucocorticoid response elements (GREs) are also present in the RANKL promoter,(27) and dexamethasone has been shown to upregulate RANKL expression as well as inhibit OPG expression.(28)
In addition to its known effects on gene expression, VitD3 has also been shown to have differentiation effects on immature, STRO-1-derived osteoblastic cells.(6, 9) Dexamethasone has also been shown to differentiate immature precursor cells in vitro into osteoblasts with a mature phenotype.(6–9, 29–31)
In this study, we sought to examine the relationship in NHBCs between the osteogenic phenotype of osteoblasts in terms of the accepted stage-selective markers STRO-1 and AP, and the osteoclastogenic phenotype, in terms of expression of RANKL and OPG. We hypothesized that treatment of NHBC with vitD3 and dexamethasone, which have known effects on both osteogenic and osteoclastogenic pathways, would provide insights into how these two pathways are linked in the human. Our results are consistent with the emerging hypothesis that immature osteoblastic cells are involved in osteoclast formation, while more mature osteoblasts acquire an osteogenic phenotype.
MATERIALS AND METHODS
NHBCs were cultured as described previously.(7) Trabecular bone fragments were obtained from normal patients (58–80 years old) during joint replacement surgery in the Department of Orthopaedic Surgery and Trauma at the Royal Adelaide Hospital. Alternatively, bone fragments were isolated from bone marrow (BM) aspirates obtained from iliac crest biopsy of normal volunteers (less than 35 years old) from the Division of Hematology, Institute of Medical and Veterinary Science (IMVS), Adelaide, South Australia. NHBCs were cultured in 75-cm2 tissue-culture flasks in α-minimal essential medium (α-MEM: Flow Laboratories, Irvine, Scotland), supplemented with 10% fetal calf serum (FCS; Thermotrace, Noble Park, Australia), 2 mM L-glutamine, and 100 μM L-ascorbate-2-phosphate (Novachem, Melbourne, Australia) (α-MEM-10). Single-cell suspensions were obtained from confluent primary NHBC cultures by enzymatic digestion with collagenase (3 mg/ml) (Collagenase type I; Worthington Biochemical, Freehold, NJ, USA) and dispase (4 mg/ml) (Neutral Protease grade II; Boehringer Mannheim, GMBH, Mannheim, Germany) for 30 minutes at 37°C, followed by trypsin digestion for 5 minutes at 37°C. Cell suspensions were then washed with growth medium before being passed through a cell strainer (Becton Dickinson Labware, Franklin Lakes, NJ, USA) to obtain a single-cell suspension.
Carboxyfluorescein diacetate succinimidyl ester labeling of cells
Carboxyfluorescein diacetate succinimidyl ester (CFSE)(32) was used to track cell division and determine division-related phenotypic and functional change during differentiation of NHBCs. Suspensions of NHBCs were washed once and resuspended in PBS/0.1% wt/vol bovine serum albumin (BSA; Sigma Chemical Co., St Louis, MO, USA). CFSE (Molecular Probes, Eugene, OR, USA) was added to the cells to give a final concentration of 10 μM and incubated at 37°C for 10 minutes. The staining was quenched by the addition of 5 volumes of ice-cold α-MEM-10, followed by incubation on ice for 5 minutes. The cells were washed three times in α-MEM-10, and cultures were established at 3 × 105 cells per 75-cm2 culture flask in α-MEM-10, or α-MEM-10 supplemented with vitD3 (2 × 10−8 M), dexamethasone (1 × 10−8 M), or both vitD3 (2 × 10−8 M) and dexamethasone (1 × 10−8 M). Some cultures were treated with colchicine (100 ng/ml) to inhibit cell division and thus provide an input labeling index for the “parental” population. After 6 days, the cells were detached by trypsin-EDTA and were stained with STRO-1 and B-478 mAbs coupled to phycoerythrin (PE), as described below, and their corresponding isotype-matched negative controls. Cell proliferation was analyzed using dual-color flow cytometric analysis and ModFit LT software (Verity Software House Inc., Topsham, ME, USA).
Flow cytometric analysis
After enzymatic digestion, NHBCs were resuspended in blocking buffer Hank's balanced salt solution [HBSS] + 20 mM HEPES, 1% normal human AB serum, 1% BSA (BSA: fraction V; Sigma), and 5% FCS for 20 minutes on ice. Approximately 1–3 × 107 cells were resuspended in 200 μl of saturating concentrations of B-478 (anti-human bone/liver/kidney AP; Developmental Studies Hybridoma Bank, Iowa University, Iowa City, IA, USA) and STRO-1(5) for 45 minutes on ice. Isotype-matched negative control antibodies (provided by Prof LK Ashman, University of Newcastle, NSW, Australia) were used under identical conditions. The cells were then washed in HBSS with 5% FCS and incubated with a goat anti-mouse IgG (γ-chain specific) fluorescein isothiocyanate (FITC; 1/50) and a goat anti-mouse IgM (μ-chain specific) PE (1/50; Southern Biotechnology Associates, Birmingham, AL, USA) for 45 minutes on ice. For cell sorting, cells were washed twice and resuspended in αMEM-10 before sorting on a FACStarPLUS (Becton Dickinson, Sunnyvale, CA, USA) flow cytometer. For analytical flow cytometry, cells were washed after incubation with the second-stage reagents, as above, and then resuspended and fixed in PBS containing 1% wt/vol paraformaldehyde. Analysis was performed on an Epics XL (Beckman Coulter, Fullerton, CA, USA) flow cytometer.
VitD3 and dexamethasone treatment of STRO-1-sorted NHBC subpopulations
NHBCs were cultured and stained for STRO-1 or isotype-matched control antibody as described above. In some experiments, the top and bottom 10–15 percentiles of positivity for STRO-1 were collected immediately for mRNA analysis by reverse transcriptase-polymerase chain reaction (RT-PCR). Alternatively, sorted cells were plated in triplicate in 24-well plates at a density of 1 × 105 cells/well in α-MEM-10 in the presence or absence of vitD3 (2 × 10−8 M), dexamethasone (1 × 10−8M), or a combination of both factors. After a further 18 and 40 h of culture, total RNA was prepared and RT-PCR was performed, as described below.
Preparation of total RNA and RT-PCR
Total RNA was prepared from NHBCs and RT-PCR was performed, as we have recently reported.(7, 33) Briefly, cells were dissolved in Trizol reagent (Life Technologies, Gaithersburg, MD, USA). Total RNA was then prepared per manufacturer's instructions. RNA was reverse transcribed from total RNA from each sample, using a cDNA synthesis kit per the manufacturer's instructions (Promega Corp., Madison, WI, USA). cDNA was then amplified by PCR to generate products corresponding to mRNA encoding human RANKL, OPG, osteocalcin, bone sialoprotein-1, core binding factor alpha (CBFA)-1, vitamin D3 receptor (VDR), and GAPDH, using AmpliTaq Gold DNA polymerase (Perkin Elmer, Norwalk, CT, USA), under nonsaturating conditions, and PCR products were detected and quantitated by fluorimagery, as previously described.(7, 33)
Parametric data sets were compared using Student's t-test. Linear and polynomial regression analysis of data sets was performed using Excel software (Microsoft Corp., Redmond, WA, USA).
Growth effects of vitD3 and dexamethasone
Growth and differentiation of osteoblasts in rodents have been shown to be intrinsically linked, with the expression of specific gene sets at defined stages of differentiation.(34) We hypothesized that within NHBCs there would be a subpopulation more likely to display pro-osteoclastogenic properties (i.e., expression of RANKL and OPG). First, to test the effects of vitD3 and dexamethasone on cell growth and differentiation, we adapted a system that enabled the simultaneous tracking of cell division and changes in the expression of phenotypic markers, such as STRO-1 and AP. The cell permeant fluorescent dye CFSE allows the number of divisions of a labeled cell population to be tracked by flow cytometry, because daughter cells receive one-half of the CFSE of the parent cell.(32) A significant advantage of using this technique is that cell growth in relation to changes in phenotype can be monitored in specific subpopulations, whereas more conventional techniques measure only the average number of cell divisions. NHBCs were labeled with CFSE and cultured in the presence or absence of vitD3 and dexamethasone. Cells were harvested and labeled with either STRO-1, anti-AP, or negative control antibodies, which were detected with PE-conjugated second antibodies.
Over a 6-day period, untreated NHBC divided between 0–4 times (Fig. 1; Table 1). Consistent with its known effects in osteoblasts,(35, 36) cells treated with vitD3 were growth retarded, as evidenced by their ability to undergo a maximum of two cell divisions, compared with the untreated control cells, which had undergone up to four cell divisions. STRO-1 seemed to identify those cells that were least growth inhibited in response to vitD3 (Fig. 1B; Table 1). Dexamethasone treatment of NHBCs, on the other hand, caused little change in cell proliferation overall, but the cells that had divided most in the presence of dexamethasone were STRO-1− and AP+ (Fig. 1C; Tables 1 and 2). Dexamethasone treatment was also able to partially reverse the growth inhibition observed with vitD3 treatment when these agents were used in combination (Fig. 1D; Tables 1 and 2).
Phenotypic changes in response to osteotropic agents
Having assessed the effects of vitD3 and dexamethasone on the proliferation of NHBCs, in terms of STRO-1 and AP as single markers, we next wished to investigate the effect of these agents on the subpopulation distribution of NHBCs, as defined by these markers used in concert.(7, 8) Two-color flow cytometry was performed examining STRO-1 and AP expression by untreated cells or cells treated for 1, 3, 6, and 18 days with vitD3, dexamethasone, or a combination of both agents. In these cultures, we consistently observed that both the percentage of cells positive for STRO-1 and the mean fluorescence intensity (MFI) for STRO-1 were biphasic and peaked at 3 days after passage, declining thereafter (data not shown). In a reciprocal fashion, the percentage of cells positive for AP tended to decrease at day 3 and increase thereafter (data not shown). In response to vitD3, marked changes were observed, with an overall increase in the percentages of STRO-1+ cells, and in particular, the STRO-1+/AP+ population. As for the untreated cells, STRO-1 expression was often biphasic, peaking at day 3 in response to vitD3. Thereafter, the increase in the percentage of STRO-1+ cells, which correspond to an osteoprogenitor pool, was maintained with continuous exposure to vitD3, over a period of 18 days (Fig. 2).
Dexamethasone treatment consistently increased the percentage of STRO-1-expressing cells at day 1, with increases seen in either the STRO-1+/AP− or the STRO-1+/AP+ population in a donor-dependent fashion. However, the longer-term effect of dexamethasone was a progressive increase in the level of expression of AP and the percentage of cells expressing this marker. This result suggests that dexamethasone treatment led to an increase in the maturation of the culture as a whole and a corresponding depletion over the culture period of the STRO-1+ fraction, compared with untreated cells, and consistent with previous reports.(6–9, 29, 31) The combination of vitD3 and dexamethasone resulted in an increase in both STRO-1 and AP, giving rise to an increase in the proportion of STRO-1+/AP+ cells at days 3 and 6. By day 18 of culture, many of these cells had matured into STRO-1−/AP+ cells, and like dexamethasone treatment alone, the most immature population, STRO-1+/AP−, was depleted. However, it was found that more of the STRO-1+/AP+ population persisted with combination treatment compared with cells treated with dexamethasone alone (Fig. 2).
Both STRO-1 and AP expression associate with relative RANKL expression in NHBCs treated with vitD3 and dexamethasone
Given the effects of vitD3 and dexamethasone on cell phenotype (Fig. 2), we hypothesized that the expression of RANKL mRNA would be associated with a particular subset of cells within the mixed population, as defined by STRO-1 and AP. To test this, total RNA was isolated from cultures treated as described above and assayed by RT-PCR. NHBCs were found to express RANKL mRNA basally to varying degrees, but when treated with vitD3, these cells reproducibly increased and maintained their expression of RANKL mRNA (Fig. 3). VitD3 also reproducibly stimulated the expression of the osteoblast marker, osteocalcin, consistent with previous reports, but had little effect on OPG mRNA expression (Fig. 3). Dexamethasone did not consistently affect RANKL or osteocalcin mRNA expression at the time-points tested and affected OPG mRNA expression in a donor-dependent fashion. A representative example of these responses is shown in Fig. 3.
Potential relationships between gene expression and STRO-1/AP phenotype were then examined by plotting the RANKL:GAPDH mRNA ratio versus the percentage of cells positive for STRO-1 or AP (see data in Fig. 2), from a number of identically treated cultures. The RANKL:GAPDH mRNA ratio associated positively with the percentage of STRO-1+ cells, by regression analysis, independently of the AP status of the cells. Because, as previously stated, the underlying expression of STRO-1 was biphasic in nature, as indeed was the expression of RANKL mRNA (see Fig. 3), the plotted data resembled a quadratic curve. Therefore, a two-phase polynomial regression curve was fitted to the data, reflecting the temporal aspects of STRO-1 and RANKL expression (Fig. 4). Interestingly, the STRO-1 subpopulation with the greatest influence on RANKL mRNA expression varied between the STRO-1+/AP− and the STRO-1+AP+ population in a donor-dependent fashion (data not shown), suggesting that additional (unidentified) phenotypic markers would be necessary to further subdivide osteoblasts through this stage of their differentiation. In addition, the relative expression of RANKL mRNA associated negatively with the percentage of AP+ cells, independently of STRO-1. Again, because of the observed biphasic nature of AP expression, a two-phase polynomial regression curve was also fitted to these data (Fig. 5). The expression of OPG, osteocalcin, CBFA-1, and bone sialoprotein-1 mRNA did not consistently associate with the STRO-1 or AP status of the cells examined (data not shown). Thus, of the genes studied, only RANKL mRNA expression was found to be strongly associated with either STRO-1 or AP expression.
Induction of RANKL mRNA expression in response to osteotropic agents
Because STRO-1 expression positively associated with RANKL mRNA expression in NHBCs, we tested whether the expression of RANKL mRNA was different among the various subpopulations sorted by FACS. No consistent differences were observed between the four subpopulations in terms of basal RANKL mRNA expression (data not shown). Because the phenotypic constitution of NHBC cultures did not change markedly in untreated cultures but changed profoundly in response to vitD3 and dexamethasone (Fig. 2), we hypothesized that the extent of stimulated RANKL mRNA expression might vary as a function of STRO-1 expression. Technically, it was difficult to obtain sufficient viable cells from each of the four subpopulations for culture. NHBCs were therefore sorted into populations representing STRO-1bright or STRO-1dim populations, each containing between 10% and 20% of the total stained cells, as indicated in Fig. 6A. The STRO-1bright cells represent relatively immature cells, irrespective of their AP status.(7, 8) The STRO-1dim population excludes the immature fraction and consists of mature STRO-1−/AP+ and STRO-1−/AP− cells. The isolated cell populations were treated with vitD3, dexamethasone, or a combination of both. Because of the observed phenotypic drift of these populations with successive cell divisions,(7) we chose an early time-point (18 h) for examination, in which time a maximum of one cell division had occurred, as determined by CFSE labeling experiments (data not shown). Total cellular RNA was extracted, and the expression of RANKL, OPG, osteocalcin, and VDR was examined using semi-quantitative RT-PCR. In sorted cells from four different donors, vitD3 consistently elicited a greater increase in RANKL mRNA expression in the STRO-1bright population compared with the STRO-1dim population. In the example shown, vitD3 treatment significantly increased RANKL mRNA expression in the unfractionated (p < 0.01) and the STRO-1bright cells (p < 0.03; Fig. 6B). A small decrease in RANKL mRNA expression was seen in the STRO-1dim population, which although significant in this case (p < 0.02), was not reproducible between donors. In contrast, osteocalcin mRNA expression was induced by vitD3 in both populations of cells isolated from each of the four donors tested (data not shown), indicating that both populations of cells could respond to vitD3. Consistent with this, the levels of VDR mRNA were not different between the sorted cell populations (data not shown). At the time-point shown, dexamethasone did not affect RANKL mRNA expression in the unfractionated cells, although moderate induction of RANKL expression was observed in cells from some donors, at later times between 24 and 40 h (data not shown), consistent with previous reports.(28) However, this induction was consistently not observed in either of the sorted fractions treated for up to 40 h, and in the example shown, a reduction in RANKL expression can be seen in response to dexamethasone (Fig. 6B). Together, these results suggest that the type, or the kinetics, of the response to dexamethasone is heterogeneous among subpopulations of NHBCs. Dexamethasone, when added simultaneously with vitD3 to unfractionated NHBCs, seemed to alter the kinetics and the extent of the RANKL mRNA response seen with vitD3 alone (see Fig. 3), and this was observed in both unfractionated cells and the STRO-1bright cells in these experiments (Fig. 6B). Both vitD3 and dexamethasone suppressed OPG mRNA expression in a donor-dependent manner, and as seen in Fig. 6C, the effect was not STRO-1-dependent, consistent with the lack of correlation between OPG mRNA expression and the STRO-1/AP phenotype seen in unfractionated cultures, as discussed in the previous section.
Most studies regarding the nature and activity of human osteoblasts have focused on their osteogenic properties. Previous studies(7, 8) have demonstrated that the stromal cell precursor marker, STRO-1, and the mature osteoblast marker, membrane AP, can be used to subdivide cultured NHBCs into four distinct subpopulations with varying osteogenic and differentiative potential. In essence, STRO-1+ NHBCs represent an osteoprogenitor phenotype, whereas mature osteogenic NHBCs lose STRO-1 expression and variably express AP.(7, 8) However, relatively few studies have addressed the question of the pro-osteoclastogenic nature of human osteoblasts. A key phenotypic property of osteoblastic cells able to support osteoclast formation is the expression of RANKL, which is induced by various osteotropic factors, including vitD3 and glucocorticoids such as dexamethasone.(4, 14, 15, 26, 28) In the present study, we sought to examine the apparent dual functionality of human cells of the osteoblast lineage by examining the regulation of RANKL and its inhibitor, OPG, with respect to the STRO-1/AP status of NHBC.
Treatment of NHBCs with vitD3 increased the percentage of cells expressing STRO-1 over an extended culture period. By using the cell division tracking dye, CFSE, we were able to show that the effects of vitD3 were associated with inhibition of growth and maturation, although the cells expressing STRO-1 proliferated to a greater extent than the STRO-1− cells. Dexamethasone had the effect of progressively increasing the percentage of cells expressing AP and decreasing the percentage of STRO-1+ cells, consistent with previous findings.(6–9, 29, 30) Analysis of gene expression of unsorted NHBCs in response to vitD3 and dexamethasone (alone or in combination) in time-course studies spanning 1 week revealed that RANKL mRNA expression changed as a function of the phenotypic constitution of the cultures, as defined by their STRO-1/AP status. Thus, the RANKL mRNA signal positively correlated with the incidence of immature (STRO-1+) cells and correlated negatively with the increasing maturation of the culture, as indicated by the incidence of AP+ cells. The level of induced RANKL mRNA derived largely from the STRO-1+ fraction of cells was supported by experiments using FACS-sorted NHBCs, in which STRO-1bright cells expressed RANKL mRNA in response to vitD3 to a greater extent than the corresponding STRO-1dim population.
The mechanism for the differential responsiveness of the osteoblastic cell populations is intriguing and was not caused simply by the relative levels of VDR mRNA expression. Both populations exhibited a similar response to vitD3 in terms of an increase in osteocalcin mRNA expression, consistent with the findings of Siggelkow et al.,(37, 38) confirming that human osteoblasts do not express osteocalcin in a stringent maturation stage-dependent fashion as has been reported for rodent osteoblasts.(34, 39) Transcriptional activation by vitD3 is complex, involving heterodimerization of retinoid X-receptors (RXR) and VDR, as well as multiple coactivator proteins.(40) This complexity likely accounts for some of its observed gene- and stage-specific effects. It has been observed that the RANKL promoter can be CpG methylated in late passage, compared with early passage, mouse ST-2 cells, which coincided with a lack of responsiveness of these cells to vitD3, in terms of induction of RANKL expression.(27) It is conceivable that inactivation of the RANKL promoter by a similar mechanism may abrogate the responsiveness of mature human osteoblasts to osteotropic factors in terms of RANKL expression. Further characterization of the STRO-1+ population of NHBCs treated with vitD3 is currently underway in our laboratory.
The association between inducible RANKL mRNA expression and the immature osteoblast phenotype seen in NHBCs seems to resemble the situation in the mouse: in mineralizing cultures of mouse calvarial osteoblasts, the induction of RANKL mRNA expression in response to vitD3 was decreased in late-stage, mature osteogenic cultures.(21) Our results imply that, regardless of the overall maturation stage of a population of human osteoblasts, it is the immature cells in that population that largely contribute to the induced RANKL signal. Furthermore, and consistent with previous findings,(26, 28, 41) vitD3 and dexamethasone, or a combination of these, tended to downregulate OPG mRNA expression and did so in a STRO-1-independent fashion. This implies that in unfractionated NHBC cultures, the overall decrease in OPG expression, in combination with an increase in expression of RANKL mRNA by the STRO-1bright cells in the presence of vitD3 and/or dexamethasone, would result in a change in the local RANKL:OPG ratio in favor of an osteoclastogenic phenotype.
Whether the changes in RANKL and OPG expression by human osteoblasts in response to vitD3 and dexamethasone, as reported here, could promote the de novo formation of osteoclasts is currently being pursued in our laboratory. However, in vivo, these agents might instead regulate the dynamics of existing bone remodeling units (BMUs). Erben et al.(42) demonstrated that short-term administration of vitD3 to rats resulted in a transient increase in osteoclast activity, an increase in the numbers of osteoblast precursors, and an increase in the osteogenic activity of existing remodeling sites. Thus, vitD3 may extend the resorptive phase by promoting the number and the duration of immature osteoblasts and the production by these of RANKL. Our results suggest that vitD3 may maintain STRO-1 expression by retarding cell growth, as it has been shown previously that loss of STRO-1 occurs through cell division.(6–8) We found that with continuous exposure to vitD3, the osteoprogenitor (STRO-1+/AP−) population of cells was increased and maintained for at least 18 days. Furthermore, after a peak in STRO-1 expression after 1 week, the numbers of osteoblasts expressing AP increased over the remaining culture period in the presence of vitD3, consistent with the known effects of vitD3 on osteoblast differentiation.(6, 9) This implies that in vivo, vitD3 may act to increase the osteoprogenitor pool and subsequently would lead to an increase in the number of osteogenic osteoblasts at a given site, leading in turn to a net increase in bone formation. Together with observations that vitD3 has pro-osteogenic effects on mature osteoblasts, the observed increase in the osteoprogenitor pool reported here is consistent with numerous reports that administration of vitD3 is anabolic both in vitro and in vivo.(6, 39, 42–45)
Glucocorticoids, such as dexamethasone, may shorten the resorptive phase of bone remodeling by promoting the differentiation of immature osteoprogenitor cells into osteogenic osteoblasts. Initially, this effect may be counteracted by an increase in the survival of existing osteoclasts,(46) possibly mediated by a transient increase in the RANKL:OPG ratio.(28) As reported here and elsewhere,(5, 31) continuous exposure to dexamethasone leads to a depletion of the osteoprogenitor pool and is known to disrupt the synthesis of type I collagen and osteocalcin.(47) These effects of dexamethasone could lead to a net decrease in bone formation over time, consistent with its observed effects in glucocorticoid-induced osteoporosis.(48) Clearly, the complex interplay of osteotropic factors in determining an osteogenic or an osteoclastogenic response requires further elucidation.
Our results may explain the obligatory linkage that exists in bone remodeling between bone resorption and bone formation. A number of mechanisms for coupling have been proposed, including release of preosteoblast chemoattractants from the bone matrix by the osteoclasts, biomechanical signals caused by focal weakness of the bone due to resorption, and more recently, coupling through the vasculature.(12) In the latter model, a capillary-like blood vessel proximal to a site of remodeling would behave as a conduit for osteoclast precursors to access the site of resorption but also the endothelial cells function to attract preosteoblasts to the site. Pericytes surrounding these blood vessels and endothelial cells constituting the vessels have been proposed to be actual sources of osteoblast precursors, and both of these cell types display osteoblastic features.(12) It is noteworthy that bovine vascular pericytes have been shown to be STRO-1+ and display osteogenic capabilities.(49) Our results suggest that the migration of osteoprogenitor cells to a site targeted for remodeling, or proliferation of these cells at that site, might initially provide a stromal layer conducive to osteoclast generation and survival, by virtue of their ability to express RANKL in response to pro-osteoclastogenic stimuli. As the osteoprogenitors proliferate and differentiate into osteogenic osteoblasts and lose RANKL expression, they would no longer support osteoclastogenesis, hence limiting the degree of osteoclast formation and survival at a given site.
Our studies begin to address the complexities surrounding human osteoblast biology, and this is the first report, to our knowledge, that directly addresses the dual functionality of the osteoblast in human bone remodeling. Our results imply that the functions of osteogenesis and osteoclastogenesis in human bone, in response to osteotropic factors, are performed by the same lineage of osteoblasts at different stages of their maturation. The human model described in this report is of additional potential importance, given that bone remodeling is a process that, perhaps, cannot be adequately studied using rodent models.
This work was supported by grants from the National Health and Medical Research Council of Australia and Eli Lilly. The authors gratefully acknowledge the assistance of Mr. Andrew (Sandy) Macintyre of the Flow Cytometry Facility, Hanson Institute, for help in cell sorting.
- 11984 The cellular basis of bone remodeling: The quantum concept reexamined in light of recent advances in the cell biology of bone. Calcif Tissue Int 36:S37–S45.
- 22002 The bone lining cell: Its role in cleaning Howship's lacunae and initiating bone formation. J Bone Miner Res 17:77–90., , , , , ,
- 31996 Human osteoblasts support human hematopoietic progenitor cells in in vitro bone marrow cultures. Blood 87:518–524., ,
- 42000 The roles of osteoprotegerin and osteoprotegerin ligand in the paracrine regulation of bone resorption. J Bone Miner Res 15:2–12., , , , ,
- 51991 Identification of stromal cell precursors in human bone marrow by a novel monoclonal antibody STRO-1. Blood 78:55–62.,
- 61994 The STRO-1+ fraction of adult human bone marrow contains the osteogenic precursors. Blood 84:4164–4173., , ,
- 71999 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 14:47–56., , , , ,
- 81999 Further characterization of cells expressing STRO-1 in cultures of adult human bone marrow stromal cells. J Bone Miner Res 14:1345–1356., , , , , ,
- 91999 Isolation and characterization of human clonogenic osteoblast progenitors immunoselected from fetal bone marrow stroma using STRO-1 monoclonal antibody. J Bone Miner Res 14:351–361., , ,
- 101999 Immunomagnetic isolation of osteoprogenitors from human bone marrow stroma. Lab Invest 79:449–457., ,
- 112002 The STRO-1+ marrow cell population is multipotential. Cells Tissues Organs 170:73–82., , ,
- 122000 The mechanism of coupling: A role for the vasculature. Bone 26:319–323.
- 132002 RANK-L and RANK: T cells, bone loss, and mammalian evolution. Annu Rev Immunol 20:795–823., ,
- 141998 Osteoclast differentiation factor mediates an essential signal for bone resorption induced by 1 alpha, 25-dihydroxyvitamin D3, prostaglandin E2, or parathyroid hormone in the microenvironment of bone. Biochem Biophys Res Commun 246:337–341., , , , , , , , , ,
- 151998 Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 93:165–176., , , , , , , , , , , , , , , , , , , , , , ,
- 161998 Osteoclast differentiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc Natl Acad Sci USA 95:3597–3602., , , , , , , , , , , , , , ,
- 171999 OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis. Nature 397:315–323., , , , , , , , , , , , , , , ,
- 181997 Osteoprotegerin: A novel secreted protein involved in the regulation of bone density. Cell 89:309–319., , , , , , , , , , , , , , , , , , , , , , , , , , , , ,
- 192001 The ratio of messenger RNA levels of receptor activator of nuclear factor kappaB ligand to osteoprotegerin correlates with bone remodeling indices in normal human cancellous bone but not in osteoarthritis. J Bone Miner Res 16:1015–1027., , , ,
- 202000 The expression of osteoprotegerin and RANK ligand and the support of osteoclast formation by stromal-osteoblast lineage cells is developmentally regulated. Endocrinology 141:4768–4776., , , , ,
- 212001 Changing RANKL/OPG mRNA expression in differentiating murine primary osteoblasts. J Endocrinol 170:451–460., , ,
- 221989 The bone marrow-derived stromal cell lines MC3T3-G2/PA6 and ST2 support osteoclast-like cell differentiation in cocultures with mouse spleen cells. Endocrinology 125:1805–1813., , , , , , ,
- 231996 The human osteoclast precursor circulates in the monocyte fraction. Endocrinology 137:4058–4060., , , ,
- 241999 Bidirectional signaling between stromal and hemopoietic cells regulates interleukin-1 expression during human osteoclast formation. Bone 25:269–278., , , , ,
- 252000 Coordinated cytokine expression by stromal and hematopoietic cells during human osteoclast formation. Bone 26:653–661., , , , ,
- 261998 Osteotropic agents regulate the expression of osteoclast differentiation factor and osteoprotegerin in osteoblastic stromal cells. Endocrinology 139:4743–4746., , ,
- 271999 Promoter structure of mouse RANKL/TRANCE/OPGL/ODF gene. Biochim Biophys Acta 1445:134–141., ,
- 281999 Stimulation of osteoprotegerin ligand and inhibition of osteoprotegerin production by glucocorticoids in human osteoblastic lineage cells: Potential paracrine mechanisms of glucocorticoid-induced osteoporosis. Endocrinology 140:4382–4389., , , , , ,
- 291999 Osteoblastic differentiation and mRNA analysis of STRO-1-positive human bone marrow stromal cells using primary in vitro culture and poly (A) PCR. J Pathol 187:374–381., , , , , , ,
- 302001 Differentiation of osteoblasts and in vitro bone formation from murine embryonic stem cells. Tissue Eng 7:89–99., , , , , , ,
- 312001 High concentrations of dexamethasone suppress the proliferation but not the differentiation or further maturation of human osteoblast precursors in vitro: Relevance to glucocorticoid-induced osteoporosis. Rheumatology 40:74–83., , , ,
- 321994 Determination of lymphocyte division by flow cytometry. J Immunol Methods 171:131–137.,
- 332000 Expression of osteoclast differentiation signals by stromal elements of giant cell tumors. J Bone Miner Res 15:640–649., , , , , ,
- 341990 Progressive development of the rat osteoblast phenotype in vitro: Reciprocal relationships in expression of genes associated with osteoblast proliferation and differentiation during formation of the bone extracellular matrix. J Cell Physiol 143:420–430., , , , , , , , ,
- 352001 Vitamin D control of osteoblast function and bone extracellular matrix mineralization. Crit Rev Eukaryot Gene Expr 11:199–226., , ,
- 362001 Regulation of osteoblast growth by interactions between transforming growth factor-beta and 1alpha, 25-dihydroxyvitamin D3. Crit Rev Eukaryot Gene Expr 11:299–317.,
- 371999 Development of the osteoblast phenotype in primary human osteoblasts in culture: Comparison with rat calvarial cells in osteoblast differentiation. J Cell Biochem 75:22–35., , , , , , ,
- 381999 1, 25 dihydroxyvitamin-D3 attenuates the confluence-dependent differences in the osteoblast characteristic proteins alkaline phosphatase, procollagen I peptide, and osteocalcin. Calcif Tissue Int 64:414–421., , , , ,
- 391991 Pleiotropic effects of vitamin D on osteoblast gene expression are related to the proliferative and differentiated state of the bone cell phenotype: Dependency upon basal levels of gene expression, duration of exposure, and bone matrix competency in normal rat osteoblast cultures. Endocrinology 128:1496–1504., , , , ,
- 402001 Vitamin D receptor and nuclear receptor coactivators: Crucial interactions in vitamin D-mediated transcription. Steroids 66:171–176., , , ,
- 411998 Osteoprotegerin mRNA is expressed in primary human osteoblast-like cells: Down-regulation by glucocorticoids. J Endocrinol 159:191–195., , ,
- 421997 Short-term treatment of rats with high dose 1, 25-dihydroxyvitamin D3 stimulates bone formation and increases the number of osteoblast precursor cells in bone marrow. Endocrinology 138:4629–4635., , , ,
- 432002 Prevention of bone loss in ovariectomized rats by combined treatment with risedronate and 1, 25-dihydroxyvitamin D3. J Bone Miner Res 17:1498–1511., , , , , , ,
- 442000 Increased formation and decreased resorption of bone in mice with elevated vitamin D receptor in mature cells of the osteoblastic lineage. FASEB J 14:1908–1916., , , , , , , , , ,
- 452001 Combination treatment with estrogen and calcitriol in the prevention of age-related bone loss. J Clin Endocrinol Metab 86:3618–3628., , ,
- 462002 Promotion of osteoclast survival and antagonism of bisphosphonate-induced osteoclast apoptosis by glucocorticoids. J Clin Invest 109:1041–1048., , , , , , , ,
- 472002 Prolonged culture of HOS 58 human osteosarcoma cells with 1, 25-(OH)2-D3 TGF-beta, and dexamethasone reveals physiological regulation of alkaline phosphatase, dissociated osteocalcin gene expression, and protein synthesis and lack of mineralization. J Cell Biochem 85:279–294., , , , , ,
- 482001 Glucocorticoid-induced osteoporosis: Summary of a workshop. J Clin Endocrinol Metab 86:5681–5685.,
- 491998 Vascular pericytes express osteogenic potential in vitro and in vivo. J Bone Miner Res 13:828–838., , , , ,