Implant osseointegration relies on the interplay of implant surface and the surrounding bone cells. The purpose of this study was to investigate the in vitro response of the bone marrow stromal cells (BMSCs) to osteogenic drugs in patients undergoing primary total hip arthroplasty (THA). Bone marrow aspirates were obtained from the discarded femoral head in 13 THA patients. The number of STRO-1 and bone morphogenetic protein receptor (BMPR)-expressing cells were measured by flow cytometry. Real-time PCR was used to measure expression of osteogenic relevant genes of BMSCs. Osteogenic potential was evaluated by alkaline phosphatase (ALP) activity. The response of BMSCs to osteogenic drugs (BMP-2, dexamethasone and etidronate) was further evaluated. There was a strong association between the gene levels of BMPR1a, ALP, RunX-2 and MSX2 with the percent of STRO-1+ cells. BMP-2 (100 ng/ml) stimulated the BMPR-pSMAD1/5 signaling pathway. Both etidronate (250 ng/ml) and dexamethasone (10 nM) treatments raised ALP activity in 33% and 70% of BMSC samples, respectively. This was independent of the activation of pSMAD signaling. A better understanding of the individual variations of BMSCs in response to osteogenic drugs is important for the THA patient management and the strategic plan of implant surface drug delivery
More than 500,000 total hip arthroplasty (THA) operations are performed in the USA annually. Cementless THA is rapidly being accepted as the surgery for arthritic diseases of the hip joint . Implant stability and longevity require efficient osseointegration, which is a process requiring the recruitment of bone marrow mesenchymal stromal cells (BMSCs) to the prosthesis surface, resulting in the direct anchorage of an implant by bone formation at the bone-implant surface. The commitment of BMSCs to an osteoblastic differentiation pathway is under the control of both systemic and local growth factors, such as bone morphogenetic protein 2 (BMP-2). Current research focuses on the interplay between implant  and periprosthetic environment . A favorable bone marrow environment should have a sufficient number of osteoprogenitor cells able to differentiate into functional osteoblasts in response to growth factors . A better understanding of the genetics and responses of BMSCs of THA patients to osteogenic drugs is important but not well defined [5,6].
STRO-1 is a cell surface antigen expressed by BMSCs . The bone marrow STRO-1+ cells are capable of differentiating into multiple mesenchymal lineages including adipocytes, osteoblasts, and chondrocytes . The variation of STRO-1+ cells can be observed between human subjects . We have identified that there are no differences in Stro-1+ cells between male and female patients, though there are slight increases in the number of Stro-1+ cells in younger patients (data suggest that the age and gender of THA patients have little impact on the numbers of Stro-1+ BMSCs (4).
BMPs are a group of growth factors known for their ability to induce bone formation . BMPs activate target cells by binding to type-Ia, -Ib and -II BMP receptors (BMPRs) . BMPR1a is necessary for the extracellular matrix deposition by osteoblasts . Type II receptor (BMPR2) binds BMPs and then initiates intracellular signaling, including the recruitment and phosphorylation of cytoplasmic proteins, especially the ‘signaling mothers against decapentaplegic’ (SMAD) 1, 5 and 8 . Osyczka et al. found that human BMSCs exhibited a lower in vitro osteogenic response to BMP-2 compared to rodents. However, it is still unclear whether disparate BMSC responses to BMP also apply to other osteogenic molecules, such as bisphosphonate  and dexamethasone .
A detailed “bone marrow osteogenic potential” analysis of THA patients might be helpful for surgeons to predict the performance of implants after surgery. An inferior bone marrow environment (such as insufficient BMSCs, the lower response of BM-SCs to osteogenic molecules, etc.) may represent a risk factor for the dysfunctional osseointegration and earlier implant failure. The purpose of this study was to investigate the response of BMSCs collected during primary THA procedures to different osteogenic drugs.
A total of 13 patients scheduled to undergo cementless THA were consecutively recruited for the study (8 men and 5 women, 50-84 years old, body mass index (BMI) ranged from 21.5 to 40) in the Department of Orthopaedic Surgery at Providence Hospital. The Clinical Investigation Plan and the Informed Consent Form were approved by the Ethics Committee of the hospital. Patients were entered into the study after reading and signing an informed consent form. Criteria for exclusion were rheumatoid arthritis, autoimmune disease, or symptoms or signs of infection or inflammation. Also excluded were patients taking nonsteroidal anti-inflammatory drugs within one month before the examination.
Bone marrow aspirates were obtained from the metaphyseal region of the discarded femoral head during THA. After filtration of the bone marrow aspirates through a 45 m-pore size cell strainer (Falcon, BD Biosciences, Bedford, MA), the bone marrow cells were collected and diluted 1:3 with Hanks’ Balanced Salt Solution and centrifuged at 200xg for 10 minutes. Supernatants were stored at -80ºC and cell pellets containing both red and white blood cells were plated in 75 mm flasks with alpha-modified Eagle’s medium (MEM) containing 20% glutamine, penicillin (100 U/ml), streptomycin sulfate (100 ugs/ml) and incubated at 37ºC in a humidified atmosphere of 5% CO2 . Aspirates were generally plated at a higher density (up to 20 X 107 /ml) because of the dilution of marrow with peripheral blood. After 2 days of culture, red blood cells were washed off the attached cells and media was replaced with media containing 10% FBS. Once the cultures reached 50-70% confluence they were passaged. When these passage 1 (P1) cells reached 70% confluence, they were harvested using TrpLE Express (Gibco, Grand Island, NY) and frozen or used to determine the percent of STRO-1+ cells in the culture.
Detection of cell surface markers (STRO-1, BMPR1a, and BMPR2) was performed using an Accuri C6 flow cytometer (Accuri, Ann Arbor, MI) equipped with CFlow software. The BMSC suspension was incubated with primary antibodies for 1 hour at 4°C followed by thirty-minute incubation with secondary antibodies. Mouse IgG anti-human STRO-1 (1 ug #MAB1038, R&D, Minneapolis, MN) was used with allophycocyanin-conjugated rat anti-mouse (#550676, BD Biosciences). For goat anti-human BMPR1A (1 ug #AF3460, R&D) and for goat anti-human BMPRII (1ug #AF811, R&D) the secondary antibody was phycoerythrin-conjugated donkey anti-goat IgG (#F010, R&D). At least 10,000 cells were analyzed per sample. Normal species-specific IgG1 was used as a negative control. After incubation, cells were washed and resuspended in 500 ?l of wash buffer and measured by flow cytometry. For the measurement of intracellular expression of pSMAD 1/5/8, cells were collected after 30 minutes of drug treatment and fixed with 2% formaldehyde for 10 minutes at 37°C. The fixed cells were then permeabilized with ice-cold 100% methanol. The permeabilized BMSCs were incubated with primary anti-pSMAD 1/5/8 antibody (14 ng, Cell Signaling #9516S, rabbit anti-human IgG) or isotype control normal rabbit IgG for 1 hour at 4o C. Unbound antibodies were removed by washing using wash buffer. The secondary antibody Alexa Fluor 488 conjugated-goat anti-rabbit (Cell Signaling #4412) were applied for 30 minutes. After incubation, cells were washed and resuspended in 500 ul of wash buffer and measured by flow cytometry. All data were analyzed using CFlow software (Accuri). The average fluorescence intensity (AFI) of the cell population was calculated for each antigen after subtraction of the AFI of the negative control samples from the AFI of the test samples (cells treated with the specific antibody).
RNA was isolated from the first passage cells and used for Real-Time PCR measurement of gene expression (StepOnePlus System, Applied Biosystems, and Forest City, CA). Primers (Invitrogen, Carlsbad, CA) for Alkaline Phosphatase (ALP): 5’-ACCATTCCCACGTCT TCACATTG-3’ (forward) and 5’-AGACATTCTCTCGTTCACCGCC-3’ (reverse), BMP-2: 5’-GAGTTGCG-GCTGCTCAGCATGTT-3’ (forward) and 5’-ACATGTCTCTT GGAGACACCT-3’ (reverse), BMPR1a: 5’-CATAACTAATGGACATTGCT-3’ (forward) and 5’-GCAGCTGGAGAAGA TGATCATAGC-3’ (reverse), Msh Homeobox-2 (MSX-2): 5’-GCCATTTTCAGCTTTTCCAG-3’ (forward) and 5’-CCCTGAGGAAACACAAGACC-3’ (reverse), Runt-related transcription factor 2 (Runx2): 5’-AGATGATGACACTGCCACCTCTG-3’ (forward) and 5’-GGGATGA- AATGCTTGGGAACTGC-3’ (reverse), Glyceraldehyde 3-phosphate dehydrogenase (GAPDH): 5’-GAGCCACATCGCTCAGACAC-3’ (forward) and 5’-CATGTAGTTGAGGTCAATGAA GG-3’ (reverse) used Sybr Green Master Mix. TaqMan assays were used for Sclerostin (SOST) (HS00228830m1), BMPR1b (HS00176144m1), and GAPDH (HS99999905m1) (Applied Biosystems).
To standardize the target gene level with respect to variations in RNA and cDNA, the housekeeping gene GADPH was used as an internal control. To determine the relative level of gene expression, the comparative threshold cycle (CT) method with arithmetic formulae were used. Subtracting the CT of the housekeeping gene from the CT of target gene yielded the ΔCT for each gene. Choosing one patient sample as the control norm, all other samples were compared to determine if they had a more, less or similar expression of the same gene. This calculation yielded a ΔΔCT which was then converted to 2(-ΔΔCT) for the exponential amplification of gene expression.
Analysis of the osteogenic potential of P1 BMSCs
Osteogenic potential was measured by cellular ALP activity (BioVision Kit #K412-500) after culturing BMSCs in osteogenic medium (α-MEM/10% FBS supplemented with 100 nM dexamethasone (DX, D2915, Sigma), 10 mM glycerol 2-phosphate and 284 μM L-ascorbic acid) for 14 days. ALP activity in the cell lysate was measured spectrometrically at 405 nm (SynergyHT by BIO-TEK, Winooski, VT) utilizing the conversion of a colorless p-nitrophenyl phosphate to a colored p-nitrophenol (Sigma, St. Louis, MO). ALP levels were normalized to protein concentration (DC Protein Assay, BIO-RAD) allowing ALP activity to be expressed as μM pNPP/mg protein/min.
For this study, frozen P2 BMSCs were culture-expanded in basal ?MEM/10% FBS medium. P3 cells, 50% confluent, were serum starved (1% FBS) overnight and then exposed to either 100 ng/ml BMP-2 (Cell Signaling, 4697), 250 ng/ml etidronate (P5248, Sigma) or 10 nM DX (Sigma) in media supplemented with 10 mM β-glycerolphosphate and 284 mM L-ascorbic acid. Cells collected after 30 minutes or 24 hours were used for flow cytometry. Cells collected after 21 days of culture were used for ALP activity and Alizarin Red staining (#ECM815, Osteogenesis Assay Kit, Millipore). BMSCs without drug treatment were included as controls. Culture media and drugs were replaced every three days until the predetermined time of harvest.
Data were expressed as mean ± SE. Group means values were compared using one-way analysis of variance with Newman– Keuls test if a normal model could be set up; if not, the Mann– Whitney U test was used. The Spearman’s test was used for correlation analysis, with P values less than 0.05 considered to be statistically significant.
ALP activity of BMSCs
We measured ALP activity following 21 days of treatment of cultured cells with 10 mM β-glycerolphosphate and 284 mM L-ascorbic acid. The mean activity (+ SE, n=12) was 54.8 ± 11.9 U/mg protein. There was a considerable difference in ALP activities among cultures of patient samples (Figure 1). There was no significant relationship between ALP activity and the percent of BMSCs positive for STRO-1 (R=0.528, P=0.077). In addition, no relationship between ALP activity and the age, sex as well as BMI of the subjects was observed (data not shown).
Figure 1. ALP activity of BMSCs from bone marrow aspirates cultured in osteogenic media for 21 days (n=12)
The yield of STRO-1+ Cells from BMSCs
Cells cultured from thirteen different patient samples of bone marrow aspirates were analyzed for STRO-1+ cells by flow cytometry after P1. The mean percent yield (± SE) of STRO-1+ cells was 9.4% ± 16.5%, indicating a great deal of variability among patients (Figure 2). There were no significant difference in the percent of STRO-1+ cells between men (11.9% ± 7.2%, n=8) and women (5.4% ± 3.5%, n=5). Similarly, there was no significant correlation between the % of STRO-1+ cells and a patient’s age or BMI.
Figure 2. The percent of passage one cultured BMSC stained with antibody to STRO-1 and measured in Accuri C6 flow cytometer shows a high degree of variability between these thirteen total hip replacement patients.
Association of osteogenesis-relevant gene expression with ALP activity and STRO-1+ cells
To investigate the osteogenic potential of BMSCs, we measured a panel of genes, including Runx2, MSX2, BMPR-1a, BMPR-1b, BMP-2, ALP and SOST by real-time PCR using mRNA isolated from P1 BMSCs. Consistent with the data for ALP activity and STRO-1+ expression, the BMSC samples exhibited a wide range of gene expression. However, a significant positive correlation was found between the expression of four genes (ALP, RunX2, BMPR1a, and MSX-2) and the percent of STRO-1+ cells detected by flow cytometry (Figure 3). In addition, an association was observed between gene expression of ALP and ALP enzyme activity. There was no association between ALP enzyme activity and other osteogenesis-relevant genes measured.
Figure 3. Correlation analysis between osteogenic gene expression (2 (–ΔΔCT) for ALP, Runx2, BMPR1A, and MSX2) and the percent of STRO-1+ cells in BMSC cultures from 12 THA patients The correlation coefficient (R) analysis was done by SPSS bivariate correlation (Pearson) analysis.
Activation of pSMAD signaling in response to osteogenic drugs
SMADs regulate transcription of BMP-responsive genes. Phosphorylation of SMAD1/5/8 is a major, early activation step in signal transduction initiated by binding of BMP to BMPR. In order to determine whether the pSMAD signaling pathway was activated in response to osteogenic drug treatment, we harvested cells for flow cytometry 30 minutes after administration of BMP-2, etidronate or DX. Not surprisingly, BMP-2 gave the strongest response with measurement of intracellular pSMAD up to 4-fold higher than control treatments of ascorbic acid and β-glycerol phosphate (Figure 4a). The average stimulation was 275%. Etidronate and DX did not significantly stimulate pSMAD intracellular expression. The percent of different patient cell cultures responding to drugs is given in Table 1.
Figure 4. Treatment with the osteogenic drug, BMP2 increases the pSMAD signaling relative to control and to other treatments in BMSC cultured for 30 minutes (a). BMP2 as well as the other drugs, etidronate, and DX, have a more variable stimulating effect on the cell surface receptor, BMPR1A after a 2-day exposure to the drugs (b). Dexamethasone (DX) treatment for 21 days stimulates alkaline phosphatase activity as well as the subsequent calcium phosphate deposits revealed by Alizarin Red staining (c,d).
BMPR response to osteogenic drugs
Since binding of BMP to its cell surface receptor is required for initiation of the pSMAD signaling cascade, we investigated changes in this receptor following two days of osteogenic drug treatment. Detection of the BMPR2 cells surface antigen proved difficult and the data did not provide any conclusive results indicative of a change in receptor level following two days of osteogenic drug stimulation (data not shown). An increase in BMPR1a expression on the cell surface after two days of treatment could be seen in only a small percentage of patients in response to BMP-2, etidronate or DX (Table 1). The average magnitude of this change was 135% of control and it was extremely variable from culture to culture leading to very large standard error OR DEVIATION? (Figure 4b).
Table 1. The response of BMSC cultures to osteogenic drugs. A positive response was defined as a value greater than 125% of control-treated cells. Of the twelve samples tested 77% responded to BMP2 with an increase in pSMAD signaling (275 + 39 percent increase over control, mean + SE, n=12) whereas 70% responded to DX with an increased ALP activity (190 + 33 percent increase over control, mean + SE, n=10).
Figure 5. BMSC sample #5 cultured in osteogenic medium for 21 days. Control (a), etidronate (b), BMP2 (c), DX (d).
ALP activity and Alizarin Red staining of BMSCs in response to osteogenic drugs
To determine whether and how BMSCs samples are responsive in a functionally relevant manner to osteogenic drugs, P3 BMSCs were treated with BMP-2, etidronate or DX. Cells were collected after 21 days of culture. The ALP activity in BMSCs without drug treatment was used to normalize values for each patient culture. The positive drug response was considered to be an ALP activity greater than 125% of control ALP. As shown in Figure 4c, increased ALP activity was observed in some but not all P3 BMSCs samples with a response rate for BMP-2 (27%), etidronate (33%) and DX (70%), respectively (Table 1). Alizarin red S staining has been used for decades to evaluate calcium-rich deposits by cells in culture. Alizarin Red S analysis was used to assess the mineralization of drug-treated cells (Figure 5). As shown in Fig. 5d, adding DX significantly enhanced mineralized nodule formation. However, BMP-2 treatment only slightly increased mineralization as compared to controls (Figure. 5C), and etidronate treatment had no effect (Figure.5B).
There are conflicting results regarding an association of THA success rate with patient age, sex, or body weight [16, 17]. This may, in part, be due to the influences of ethnic status, general disease background, physical activity level,  and surgical variables. A better understanding of a patient’s bone marrow microenvironment profiles is important but has not been well studied. In this study, we demonstrated that there are highly diversified BMSC profiles characterized by the percentage of STRO-1+ cells, ALP activity and osteogenic-relevant gene profiles in this group of 13 THA patients.
STRO-1+ is a cell surface antigen expressed in human bone marrow cells. An STRO-1+ enriched subset of BMSCs is capable of differentiating into multiple mesenchymal lineages including adipocytes, osteoblasts and chondrocytes . The percent of BMSC STRO-1+ cells has been shown to vary with age, sex [8, 20], and osteoarthritis background . It is still unclear whether the variation of the BMSC STRO-1+ cells represents a biomarker for BMSC osteogenic potential.
Osteoblast differentiation is regulated by a panel of transcription factors such as Runx2 and MSX2 . Runx2 commences the expression of key osteoblast proteins such as ALP and osteocalcin and is expressed in bone cells throughout the osteogenic lineage . MSX2 plays critical roles in bone formation and osteoblast differentiation . We discovered a strong association between the number of STRO-1+ cells and the genes expressing BMPR1A, MSX2, Runx2, and ALP (Figure 3). In addition, we found a correlation between the number of STRO-1+ cells and ALP activity in early passage BMSCs but not in later passages. This leads us to suspect that the culture of human BMSCs can result in the predominance of different cell phenotypes. Since we did not measure STRO-1+ cells in P3 this cannot be confirmed.
This study was the first attempt to test the responses of BMSCs to a panel of osteogenic drugs. We found that the response of BMSCs to a panel of osteogenic drugs was highly diversified. Though BMPs might play a critical role in enhancing implant osseointegration , the response of human BMSCs to BMP-2 was quite varied. Current clinical studies have shown the variable success rate of recombinant BMP in the treatment of fracture repair and nonunion . As compared to results observed in animal models, BMPs are relatively inefficient in inducing human BMSCs to undergo osteogenesis . We found that BMP-2 stimulated the BMPR-pSMAD1/5 signaling pathway but this did not correlate with ALP activity in these cultures, suggesting there is a distinctive molecular mechanism or signaling pathway for its osteogenic action.
Bisphosphonates are widely used as therapeutic agents in bone disorders including osteoporosis due to their osteoclast inhibitory effect . Recent data show that bisphosphonates may also induce bone-building by stimulating osteoblast activity , including an increase in BMP-2 gene expression . In animal models, the local application of bisphosphonate improves periprosthetic bone quality and osseointegration . Wilkinson et al.  found that bisphosphonate therapy reduced femoral calcar bone loss and bone turnover in a group of patients over 2 years after THA. von Knoch et al.  evaluated the effects of bisphosphonates on the proliferation and osteogenic differentiation of human BMSCs isolated during THA. Cultured BMSCs were treated with or without a bisphosphonate (alendronate, risedronate, or zoledronate) and analyzed over 21 days. Bisphosphonate treatment enhanced BMSC proliferation and initiated osteoblastic differentiation. However, the drug responses were highly diversified among the different donors . The molecular mechanism for the variation of donor responses has not been fully investigated. Our data demonstrated that bisphosphonate (etidronate) treatment increased ALP activity in 3 of 11 patient cultures through a pSMAD-independent signaling pathway. Additionally, only one of eleven patient cultures responded to etidronate with an increase in BMPR1a expression.
It is well established that DX increases ALP activity in cultured BMSCs. Diefenderfer et al.  reported that osteogenesis can be induced by DX (100 nM). Our results confirmed this in 70% of BMSC samples collected from THA patients during surgery. In agreement with these findings, we demonstrated that the continuous presence of DX was an absolute requirement for in vitro osteogenesis in human BMSCs. All cultures treated with DX exhibited Alizarin Red staining for calcium phosphate production after 21 days of continuous treatment. Because DX is inexpensive and its clinical properties are well understood, it could provide a convenient, safe alternative to BMP as a means of inducing osseointegration in THA patients.
Taken together, we found that the response of BMSCs to osteogenic drugs was highly diversified in these 13 THA patients. The bone marrow microenvironment is a complex, three-dimensional structure composed of many cell types and abundant extracellular matrix. More efforts are needed to elucidate how the bone marrow composition is related to the osteogenic potential (drug response and ALP activity) and whether this would allow prediction of a clinical outcome. One critical avenue for further investigations is to define the number and quality of osteoblast precursor cells among BMSCs. It remains unclear whether the cell surface marker STRO-1 identifies a population of osteoblast precursor cells in the bone marrow. Additional markers, such as BMPRs, may be required to define the osteoblast precursor cells in the bone marrow.
Data generated from this study demonstrated individual variations in the response of BMSCs to a panel of osteogenic drugs. The lack of further clinical outcome measurements, such as computed tomography (CT) or dual-energy X-ray absorptiometry (DXA), to assess bone changes is a potential limitation.
Data generated from this study is important, because a better understanding of an individual patient’s bone marrow profile and response sensitivity to different osteogenic drugs may help clinicians to predict and prevent post-THA implant failure due to the disruption of osseointegration. A comprehensive “bone marrow osteogenic profile” analysis will help clinicians and biomedical engineers to provide patient-specific implant surface-based osteogenic drug delivery strategies [32-34].
The authors wish to thank Dianna Fan, MSs for data collection and literature reviews.
1. Berry DJ, Harmsen WS, Cabanela ME, Morrey BF. Twentyfive-Year Survivorship of Two Thousand Consecutive Primary Charnley Total Hip Replacements : Factors Affecting Survivorship of Acetabular and Femoral Components. J Bone Joint Surg Am. 2002, 84(2): 171-177.
3. Ponche A, Bigerelle M, Anselme K. Relative influence of surface topography and surface chemistry on cell response to bone implants materials Part 1: physico-chemical effects. Proc Inst Mech Eng H. 2010, 224(12): 1471-1486.
4. Shen Y, Wang W, Li X, Liu Z, Markel D et al. Impacts of age and gender on bone marrow profiles of BMP7, BMPRs and Stro-1 + cells in patients with total hip replacement. International Orthopaedics. 2012, 36(4): 879-886.
7. Zannettino AC, Paton S, Kortesidis A, Khor F, Itescu S et al. Human mulipotential mesenchymal/stromal stem cells are derived from a discrete subpopulation of STRO-1bright/CD34 / CD45(-)/glycophorin-A-bone marrow cells. Haematologica. 2007, 92(12): 1707-1708.
8. Byers RJ, Brown J, Brandwood C, Wood P, Staley W et al. 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. 1999, 187(3): 374-381.
10. Mishina Y, Starbuck MW, Gentile MA, Fukuda T, Kasparcova V et al. Bone morphogenetic protein type IA receptor signaling regulates postnatal osteoblast function and bone remodeling. J Biol Chem. 2004, 279(26): 27560-27566.
11. Nohe A, Hassel S, Ehrlich M, Neubauer F, Sebald W et al. The mode of bone morphogenetic protein (BMP) receptor oligomerization determines different BMP-2 signaling pathways. J Biol Chem. 2002, 277(7): 5330-5338.
14. Su Y, Su Q, Liu W, Lim M, Venugopal JR et al. Controlled release of bone morphogenetic protein 2 and dexamethasone loaded in core-shell PLLACL-collagen fibers for use in bone tissue engineering. Acta Biomater. 2011, 8(2): 763-771.
16. Robertsson O, Knutson K, Lewold S, Lidgren L. The Swedish Knee Arthroplasty Register 1975-1997: an update with special emphasis on 41,223 knees operated on in 1988-1997. Acta Orthop Scand. 2001, 72(5): 503-513.
17. Furnes O, Espehaug B, Lie SA, Vollset SE, Engesaeter LB et al. Early failures among 7,174 primary total knee replacements: a follow-up study from the Norwegian Arthroplasty Register 1994-2000. Acta Orthop Scand. 2002, 73(2): 117-129
18. Fortin PR, Penrod JR, Clarke AE, St Pierre Y, Joseph L, etc. Timing of total joint replacement affects clinical outcomes among patients with osteoarthritis of the hip or knee. Arthritis Rheum. 2002, 46(12): 3327-3330.
20. Brusnahan SK, McGuire TR, Jackson JD, Lane JT, Garvin KL et al. Human blood and marrow side population stem cell and Stro-1 positive bone marrow stromal cell numbers decline with age, with an increase in quality of surviving stem cells: correlation with cytokines. Mech Ageing Dev. 2010, 131(11- 12): 718-722.
21. Giurea A, Ruger BM, Hollemann D, Yanagida G, Kotz R et al. STRO-1+ mesenchymal precursor cells located in synovial surface projections of patients with osteoarthritis. Osteoarthritis Cartilage. 2006, 14(9): 938-943.
23. Shimizu T, Tanaka T, Iso T, Matsui H, Ooyama Y et al. Notch signaling pathway enhances bone morphogenetic protein 2 (BMP2) responsiveness of Msx2 gene to induce osteogenic differentiation and mineralization of vascular smooth muscle cells. J Biol Chem. 2011, 286(21): 19138-19148.
24. Yan MN, Dai KR, Tang TT, Zhu ZA, Lou JR. Reconstruction of peri-implant bone defects using impacted bone allograft and BMP-2 gene-modified bone marrow stromal cells. J Biomed Mater Res A. 2010, 93(1): 304-313.
27. Dobnig H, Hofbauer LC, Viereck V, Obermayer-Pietsch B, Fahrleitner-Pammer A. Changes in the RANK ligand/osteoprotegerin system are correlated to changes in bone mineral density in bisphosphonate-treated osteoporotic patients. Osteoporos Int. 2006, 17(5): 693-703.
28. Blazsek J, Dobo NC, Blazsek I, Varga R, Vecsei B et al. Aminobisphosphonate stimulates bone regeneration and enforces consolidation of titanium implant into a new rat caudal vertebrae model. Pathol Oncol Res. 2009, 15(4): 567-577.
29. von KF, Jaquiery C, Kowalsky M, Schaeren S, Alabre C et al. Effects of bisphosphonates on proliferation and osteoblast differentiation of human bone marrow stromal cells. Biomaterials. 2005, 26(34): 6941-6949.
30. Wilkinson JM, Eagleton AC, Stockley I, Peel NF, Hamer AJ et al. Effect of pamidronate on bone turnover and implant migration after total hip arthroplasty: a randomized trial. J Orthop Res. 2005, 23(1): 1-8.
34. Stadelmann VA, Gauthier O, Terrier A, Bouler JM, Pioletti DP. Implants delivering bisphosphonate locally increase periprosthetic bone density in an osteoporotic sheep model. A pilot study. Eur Cell Mater. 2008, 16: 10-16.