PATHOLOGY
In Vitro Effect of Bisphosphonates on Oral Keratinocytes and Fibroblasts Niall M. H. McLeod, FRCS(OMFS), FDS, MRCS,* Karwan A. Moutasim, BDS, MSc, PhD,y Peter A. Brennan, MD, FRCS(OMFS), FRCS, FDS,z Gareth Thomas, MScD, PhD,x and Veronika Jenei, MSc, PhDk Purpose:
Osteonecrosis of the jaws is a potential complication of bisphosphonate (BP) therapy. The underlying mechanisms remain unclear. Although most research has concentrated on the effects of BPs on osteoclast and osteoblast functions, the disease is diagnosed and classified based on of mucosal breakdown, suggesting that oral soft tissues may be involved in its pathogenesis. The aim of this study was to determine the effect of 3 different BP drugs (alendronate, zoledronate, and clodronate) on the function of oral keratinocytes and fibroblasts.
Materials and Methods:
Human oral keratinocytes (OKF6) and fetal foreskin fibroblasts (HFFF2) were exposed to each drug at several concentrations and the effect on cell proliferation was assessed by counting the viable cells after different lengths of treatment. The effect on cell migration was examined using Transwell migration assays. An organotypic coculture model using keratinocytes and fibroblasts, which recapitulated the morphology of the oral mucosa, was used to assess the effect of the drugs on epithelial stratification and differentiation.
Results:
The 3 BPs affected the viability and proliferation of OKF6 and HFFF2 cells at concentrations in keeping with their known relative in vitro potencies. There was no effect on cell migration or tissue architecture in organotypic cultures at subtoxic concentrations.
Conclusion:
The lack of effect of these drugs on cell migration below concentrations known to affect cell viability suggests that BP-related osteonecrosis is not caused through suppression of keratinocyte or fibroblast motility. Ó 2014 American Association of Oral and Maxillofacial Surgeons J Oral Maxillofac Surg 72:503-509, 2014
Bisphosphonate (BP) compounds have been synthesized since the 19th century, originally for use in industry. Although etidronate was first used clinically in the early 20th century, the applications of BPs became clarified and popularized only in the past 40 years.1 The principal action of BPs clinically is to inhibit bone resorption, although they have a secondary action of inhibiting mineralization.2 Several mechanisms of action have been proposed, including the inhibition
of osteoclast recruitment, adhesion, and activity and shortening of the osteoclast lifespan.3 The mechanism of action depends principally on the presence or absence of a nitrogen side chain in the BP structure, but different side chain structures have been associated with additional molecular actions.2,3 Osteonecrosis of the jaws (ONJ) is a complication of BP therapy first described almost 10 years ago.4 The pathophysiology of BP-related ONJ (BRONJ) is still
*Specialist Registrar, Department of Oral and Maxillofacial
This study was supported by a grant from the British Association
Surgery, Queen Alexandra Hospital, Portsmouth, United Kingdom.
of Oral and Maxillofacial Surgeons.
yResearch Fellow, Cancer Sciences Division, University of Southampton;
Clinical
Lecturer,
Cancer
Sciences
Address correspondence and reprint requests to Mr McLeod:
Division,
Department of Oral and Maxillofacial Surgery, The John Radcliffe,
University of Southampton, Southampton, United Kingdom.
Headley Way, Oxford OX39DU, UK; e-mail: [email protected]
zProfessor, Department of Oral and Maxillofacial Surgery, Queen
Received February 7 2013
Alexandra Hospital, Portsmouth, United Kingdom. xProfessor, Cancer Sciences Unit, University of Southampton,
Ó 2014 American Association of Oral and Maxillofacial Surgeons
Southampton, United Kingdom.
0278-2391/13/01033-1$36.00/0
Accepted August 1 2013
kSenior Research Fellow, Cancer Sciences Unit, University of
http://dx.doi.org/10.1016/j.joms.2013.08.007
Southampton, Southampton, United Kingdom.
503
504 poorly understood, and the relative importance of the effect of BPs on bone and soft tissues is unclear.5,6 Although it is known that BPs suppress bone turnover, how this translates into ulceration of the overlying mucosa is unclear. Although BPs are known to have an effect on mucosal tissues at high concentration, whether this effect is clinically relevant is unknown.7,8 This study compared the effect of 3 different BPs on human oral keratinocyte and fibroblast proliferation and migration.
Materials and Methods CELL CULTURE
Immortalized human fetal foreskin fibroblasts (HFFF2) were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) with 10% fetal bovine serum (FBS) and L-glutamine. Immortalized human oral keratinocytes (OKF6) were cultured in aMEM supplemented with 10% FBS, adenine, insulin, hydrocortisone, epidermal growth factor, and L-glutamine. Cells were cultured in a humidified atmosphere at 37 C and 10% CO2. BISPHOSPHONATE DRUGS
The chosen BP drugs reflected a spectrum of potencies in vivo and different modes of action.9 Zoledronate is a nitrogen-containing BP that is administered parenterally and is the most potent BP in vivo. Alendronate also contains nitrogen, but is principally administered orally and is the most potent oral BP. Clodronate is a non–nitrogen-containing BP that is principally administered orally and has relatively weak in vivo potency.9 PROLIFERATION ASSAY
The effect of different concentrations of BP on growth and viability of the chosen cell lines was assessed in the following way: 5 104 cells were seeded in 24-well plates with 500 mL of medium in the presence of H2O (control) or alendronate 104 to 107 mol/L, clodronate 102 to 105 mol/L, or zoledronate 0.2 106, 1 106, or 5 106 mol/L. Concentrations of each BP were selected based on previous in vitro cell studies.7-9 Each condition was tested in triplicate. At 1 day, 2 days, and 5 days, cells were trypsinized and counted using a Casy Counter (Roche Diagnostics, Penzberg, Germany).
IN VITRO EFFECT OF BISPHOSPHONATES
chosen were concentrations for the migration assays that did not produce a significant decrease in cell proliferation or affect cell viability in the previous assay. HFFF2 or OKF6 cells (5 104) were seeded in the upper chamber of Transwell inserts (8-mm pore; Corning Inc) with 100 mL of serum-free migration buffer (DMEM with 0.1% heat-inactivated bovine serum albumin [BSA] and L-glutamine for HFFF2 or aMEM with adenine, L-glutamine, and 0.1% heat-inactivated BSA for OKF6) in the presence of H2O or alendronate 105 or 106 mol/L, clodronate 106 or 107 mol/L, or zoledronate 0.2 106 or 1 106 mol/L and allowed to migrate toward serum-free migration buffer and H2O (negative control) or the growth medium of the respective cell line with a positive control without BP. After 24 hours, cells migrating to the lower chamber were trypsinized and counted on a Casy Counter. ORGANOTYPIC CULTURES
Although single-cell migration of HFFF2 and OKF6 cells was not affected by subtoxic concentrations of BPs, the effect of BPs on 3-dimensional cell invasion was investigated using an organotypic coculture model that replicates the oral mucosal architecture, which shows whether BP-modulated epithelial stratification, differentiation, or invasion occurs. Organotypic cultures composed of OKF6 keratinocytes and HFFF2 fibroblasts were generated. OKF6 cells were treated with the highest concentration of BPs, which did not significantly decrease cell viability in the proliferation assays. Organotypic cultures were prepared as described previously.10 OKF6 cells (1 106) were seeded on gels comprising a 50:50 mixture of Matrigel and type 1 collagen containing 5 105 HFFF2 cells per gel. The day after seeding, the gels were raised at the air-tissue interface. The medium under the gels was supplemented with H2O or alendronate 1 105 mol/L, clodronate 1 105 mol/L, or zoledronate 1 106 mol/L and was changed every second day. After 7 days, the gels were bisected, fixed in formal saline, and processed to paraffin. Four-micrometer sections were stained with hematoxylin and eosin. The assay was replicated twice. STATISTICAL ANALYSIS
Data were collected and analyzed using the Student unpaired t test and Excel (Microsoft, Redmond, WA). Each BP was compared with the control for each concentration and time point. A P value lower than .05 was accepted as statistically significant.
MIGRATION ASSAY
Transwell (Corning Inc, Tewksbury, MA) migration assays were performed to assess the effect of the different BPs on cell motility for each cell line. Proliferation assays had shown that high concentrations of all 3 BPs resulted in loss of cell viability; therefore, concentrations of BP
Results PROLIFERATION ASSAY
Results of the proliferation assay are shown in Figure 1.
505
MCLEOD ET AL
A
B
C
FIGURE 1. Proliferation assays of HFFF2 and OKF6 cell lines with different bisphosphonate drugs. A, Alendronate. B, Zoledronate. C, Clodronate. The growth ratio of cells cultured to cells counted at 1, 2, and 5 days is depicted. All 3 bisphosphonates showed a significant decrease in cell proliferation and viability in the 2 cell lines at the highest concentrations used. McLeod et al. In Vitro Effect of Bisphosphonates. J Oral Maxillofac Surg 2014.
At a concentration of 104 mol/L, alendronate significantly suppressed proliferation of HFFF2 compared with the control from 1 day (P < .05). For OKF6 cells, although a similar effect was shown for alendronate at 104 mol/L, there also was a significant decrease in proliferation shown at a concentration of 105 mol/L at 5 days (P = .002). Zoledronate significantly decreased the proliferation of HFFF2 after only 2 days at its highest concentration of 5 106 mol/L (P = .017) and at 5 days for OKF6 (P = .021) at the same concentration.
Clodronate significantly affected proliferation of HFFF2 and OKF6 after 1 day at a concentration of 102 mol/L (P = .014) and at a concentration of 103 mol/L significantly decreased the proliferation of HFFF2 at only 5 days (P = .049). Results indicated that clodronate also decreased the proliferation of OKF6 at 5 days at a concentration of 104 mol/L (P = .014), although not at a concentration of 103 mol/L (P = .34). The reason for this variant result is not clear and is thought unlikely to truly represent cellular effects.
506 MIGRATION ASSAY
The results of the migration assay are shown in Figure 2. At these subtoxic BP concentrations, there was no difference in HFFF2 or OKF6 migration between BPtreated and non–BP-treated cells (positive control, P > .05) and significantly more compared with cells without migration buffer (negative control, P > .05), indicating that BP had no effect on cell migration. ORGANOTYPIC CULTURES
Figure 3 shows the organotypic assays achieved for the 3 BPs. Th OKF6 cells showed squamous morphology and stratification, with no differences noted between the control and the cultures with any of the BPs. There was some evidence of invasion of the basement layer on the photomicrographs, which probably reflects the abnormal behavior of this immortalized cell line.
Discussion Osteonecrosis is an uncommon, but potentially serious, complication of BP therapy, which to date has been reported as affecting only the jawbones. The pathophysiology of the condition and the reason for its predilection for the jaws are still poorly understood. Multiple theories have been postulated, but it is increasingly recognized that the effect of BP on the oral soft tissues may play a role.5,6 Wound healing in the oral cavity involves cell proliferation and motility and a complex interplay of multiple cell types, including osteogenic cells, oral keratinocytes, fibroblasts, and endothelial cells. BRONJ most commonly presents after injury to the oral tissues (eg, after dental extraction) and its pathophysiology may arise from an effect on at least 1 of these individual cell types or an effect on their interactions. The effect of BPs on osteogenic cells has been studied extensively because these are the cells responsible for their principal therapeutic activity. The main focus of BP activity is the osteoclast. High BP concentrations result in osteoclast apoptosis, whereas lower concentrations inhibit bone resorption.3,9 BPs also are known to have an effect on osteocytes, where they have a protective effect against glucocorticoid-induced apoptosis.2 The effect of BPs on different soft tissues has been investigated for some time, because of the known side effects of BPs on the gastrointestinal tract, including esophagitis and gastric irritation.11,12 Oral soft tissue cells were initially investigated because of interest in the therapeutic use of BPs in dental trauma.7,8,11 An antitumor effect has been found with some BPs and is thought to be due to the induction of tumor cell apoptosis and the inhibition of tumor cell adhesion
IN VITRO EFFECT OF BISPHOSPHONATES
and invasion, and there has been interest in whether similar effects occur with oral soft tissue cells.13 With regard to the BPs studied, zoledronate has been shown to inhibit human endothelial cell proliferation and to modulate endothelial cell adhesion and migration.14 Walter et al15 found a decrease in cell viability of human umbilical vein endothelial cells in culture with the amino BPs and nonamino BPs. Moreira at al8 reported that alendronate mixed in a paste was rapidly toxic to endothelial cells at a concentration of 0.6 105 mol/L. In addition, when this paste was inserted subcutaneously in a rat, there was local tissue inflammation and necrosis. Correia et al7 found that use of alendronate at a concentration of 1 106 mol/L resulted in the loss of viability of human periodontal ligament cells and altered cell morphology at 1 105 mol/L. Scheper et al,16 using human gingival fibroblasts and human keratinocyte cell lines, also reported that zoledronate at a concentration of at least 2.5 107 mol/L resulted in cell apoptosis and at concentrations of 1 106 mol/L significantly decreased cell proliferation. These studies are similar to the present findings that high concentrations of BPs inhibit cell proliferation and result in a loss of cell viability. Amino BPs exert at least part of their effect by inhibiting prenylation of small guanosine triphosphatases (GTPases), which has an effect on osteoclast cell morphology, signaling, and apoptosis. Given the importance of small GTPases, such as Rac1 and cdc42, in cell motility, these BPs may suppress cell movement, and this may play a role in the pathophysiology of BRONJ. Various in vitro studies using a scratch woundhealing model have found that culturing cells with BPs can significantly delay scratch closure.12,17 However, in these studies, the concentrations of BP shown to suppress scratch closure also inhibited cell proliferation and decreased cell viability. The present study used Transwell migration assays to examine the motility of individual cells. The BP concentrations used for these assays were found not to have a significant effect on cell proliferation or viability, and at such concentrations no effect on cell migration was observed. Organotypic assays recapitulate the morphology of the oral mucosa and were used to assess the effect of the drugs on epithelial stratification and differentiation. The authors did not find any effect on epithelial stratification of BP at the chosen concentrations, which were concentrations shown not to have a significant effect on cell proliferation. Much has been made of the effect of different BPs on various fibroblast and keratinocyte cell lines, most of which are from nonhuman8,12 or nonoral cavity8,13
507
MCLEOD ET AL
FIGURE 2. Migration assays of HFFF2 and OKF6 cell lines with 3 different bisphosphonate drugs at subtoxic concentration. A, Alendronate. B, Zoledronate. C, Clodronate. Migration toward SF and H2O (column 1, negative control) was compared with migration toward FBS (column 2, positive control) and with 2 different subtoxic concentrations of each bisphosphonate (columns 3 and 4). No decrease in cell migration was found for the 3 bisphosphonates with either cell line. FBS, fetal bovine serum; SF, serum-free migration buffer. McLeod et al. In Vitro Effect of Bisphosphonates. J Oral Maxillofac Surg 2014.
cell lines. In line with the demonstration of mandibular and iliac crest bone marrow stem cells having different responses to BP, the choice of soft tissue cell line in experiments may prove to be more important than previously recognized. Malignant cell
lines also may behave differently, and contamination or transformation of immortalized nonmalignant cell lines make interpretation of results difficult to compare. Therefore, in vitro findings may not be directly comparable to the clinical setting.17
508
IN VITRO EFFECT OF BISPHOSPHONATES
FIGURE 3. Photomicrographs of organotypic culture of HFFF2 and OKF6 cells cultured with bisphosphonate drugs. A, Control. B, Alendronate at a concentration of 1 105 mol/L. C, Zoledronate at a concentration of 1 106 mol/L. D, Clodronate at a concentration of 1 105 mol/L. McLeod et al. In Vitro Effect of Bisphosphonates. J Oral Maxillofac Surg 2014.
The present results combined with those of previous studies confirm that, at high concentrations, BP drugs are toxic to different soft tissue cell lines, and that at these concentrations there is inhibition of healing using a scratch assay model. However, at subtoxic concentrations, there are no specific effects on singlecell motility or epithelial stratification. It is questionable whether induction of cell toxicity is the principal pathophysiologic mechanism in BRONJ, because oral soft tissues may never be exposed to the high concentrations of BPs required to induce cell death in vitro. Circulating BPs are rapidly incorporated into mineralized bone, within which it is metabolically inactive, and the plasma concentration of BP rapidly decreases after drug administration to levels that have been shown to have no functional effect on cells.1,18,19 For example, zoledronate has been shown to have a peak plasma concentration lower than 400 ng/mL, and this decreases to below 1% within 24 hours.18 The use of BP concentrations in vitro, which are comparable to plasma concentrations immediately after administration, is therefore questionable, because cells are unlikely to be subject to this concentration for extended periods in vivo. BPs are released during bone resorption and taken up by the osteoclasts on which they exert their effect. Although BP concentrations may reach high levels within the resorption space and Sato et al20 estimated that alendronate may reach concentrations of 0.5 mol/L, there is no clear evidence that similar concentrations may be found around the oral mucosal cells for any period.5,20,21
The present study and other studies have shown that at low concentrations BPs appear to have little effect on the proliferation and migration of fibroblasts or keratinocytes, whereas at high concentrations the drugs are significantly toxic. Many questions remain about the role of soft tissues in the pathophysiology of BRONJ. One of the key questions is the true concentration exposure of the tissues to BPs, because that would clarify the relevance of the in vitro findings of cell function inhibition.
References 1. Fleisch H: Bisphosphonates: Mechanisms of action. Endocr Rev 19:80, 1998 2. Russell RGG, Watts NB, Ebetino FH, et al: Mechanism of action of bisphosphonates: Similarities and differences and their potential influence on clinical efficacy. Osteoporos Int 19:733, 2008 3. Rogers MJS, Gordon HL, Benford FP, et al: Cellular and molecular mechanism of action of bisphosphonates. Cancer 88(suppl 12): 2961, 2000 4. McLeod NMH, Brennan PA, Ruggiero SL: Bisphosphonate osteonecrosis of the jaw: A historical and contemporary review. Surgeon 10:36, 2012 5. Allen MR, Burr DB: The pathogenesis of bisphosphonate related osteonecrosis of the jaw: So many hypotheses, so few data. J Oral Maxillofac Surg 67(suppl 1):61, 2009 6. Reid IR, Bolland MJ, Grey AB: Is bisphosphonate-associated osteonecrosis of the jaw associated with soft tissue toxicity? Bone 41:318, 2007 7. Correia VFP, Caldeira CL, Marques MM: Cytotoxicity evaluation of sodium alendronate of cultured human periodontal ligament fibroblasts. Dent Traumatol 22:312, 2006 8. Moreira MS, Katayama E, Bombana AC, et al: Cytotoxicity analysis of alendronate on cultured endothelial cells and subcutaneous tissue. A pilot study. Dent Traumatol 21:329, 2005 9. Ringe JD: The interpretation of preclinical data in predicting bisphosphonate response in the treatment of osteoporosis. Clin Therap 20:648, 1998
MCLEOD ET AL 10. Nystrom ML, Thomas GJ, Stone M, et al: Development of a quantitative method to analyse tumour cell invasion in organotypic culture. J Pathol 205:468, 2005 11. Twiss IM, Pas O, Ramp-Koopmanschap W, et al: The effects of nitrogen-containing bisphosphonates on human epithelial (caco-2) cells, an in vitro model for intestinal epithelium. J Bone Miner Res 14:784, 1999 12. Landesberg R, Cozin M, Cremers S, et al: Inhibition of oral mucosal cell wound healing by bisphosphonates. J Oral Maxillofac Surg 66:839, 2008 13. Santini D, Vespasiani Gentilucci U, Vincenzi B, et al: The antineoplastic role of bisphosphonates: From basic research to clinical evidence. Ann Oncol 14:1468, 2003 14. Wood J, Bonjean K, Reutz S, et al: Novel antiangiogenic effects of the bisphosphonate compound zoledronic acid. J Pharmacol Exp Therap 302:1055, 2002 15. Walter C, Klein MO, Pabst A, et al: Influence of bisphosphonates on endothelial cells, fibroblasts and osteogenic cells. Clin Oral Invest 14:35, 2010
509 16. Scheper MA, Badros A, Chaisuparat R, et al: Effect of zoledronic acid on oral fibroblasts and epithelial cells: A potential mechanism of bisphosphonate-associated osteonecrosis. Br J Haematol 144:667, 2008 17. Stefanik D, Sarin J, Lam T, et al: Disparate osteogenic response to mandible and iliac crest bone marrow stromal cells to pamidronate. Oral Dis 14:465, 2008 18. Skerjanec A, Berenson J, Hsu A, et al: The pharmacokinetics and pharmacodynamics of zoledronic acid in cancer patients with varying degrees of renal function. Clin Pharmacol 43: 154, 2003 19. Lin JH: Bisphosphonates: A review of their pharmacokinetic properties. Bone 18:75, 1996 20. Sato M, Grasser W, Endo N, et al: Bisphosphonate action. Alendronate localistaion in rat bone and effects on osteoclast ultrastructure. J Clin Invest 88:2095, 1991 21. Scheper MA, Chaisuparat R, Cullen KJ, et al: A novel soft-tissue in vitro model for bisphosphonate-associated osteonecrosis. Fibrin Tiss Rep 3:6, 2010
In Vitro Effect of Bisphosphonates on Oral Keratinocytes and Fibroblasts Niall M. H. McLeod, FRCS(OMFS), FDS, MRCS,* Karwan A. Moutasim, BDS, MSc, PhD,y Peter A. Brennan, MD, FRCS(OMFS), FRCS, FDS,z Gareth Thomas, MScD, PhD,x and Veronika Jenei, MSc, PhDk Purpose:
Osteonecrosis of the jaws is a potential complication of bisphosphonate (BP) therapy. The underlying mechanisms remain unclear. Although most research has concentrated on the effects of BPs on osteoclast and osteoblast functions, the disease is diagnosed and classified based on of mucosal breakdown, suggesting that oral soft tissues may be involved in its pathogenesis. The aim of this study was to determine the effect of 3 different BP drugs (alendronate, zoledronate, and clodronate) on the function of oral keratinocytes and fibroblasts.
Materials and Methods:
Human oral keratinocytes (OKF6) and fetal foreskin fibroblasts (HFFF2) were exposed to each drug at several concentrations and the effect on cell proliferation was assessed by counting the viable cells after different lengths of treatment. The effect on cell migration was examined using Transwell migration assays. An organotypic coculture model using keratinocytes and fibroblasts, which recapitulated the morphology of the oral mucosa, was used to assess the effect of the drugs on epithelial stratification and differentiation.
Results:
The 3 BPs affected the viability and proliferation of OKF6 and HFFF2 cells at concentrations in keeping with their known relative in vitro potencies. There was no effect on cell migration or tissue architecture in organotypic cultures at subtoxic concentrations.
Conclusion:
The lack of effect of these drugs on cell migration below concentrations known to affect cell viability suggests that BP-related osteonecrosis is not caused through suppression of keratinocyte or fibroblast motility. Ó 2014 American Association of Oral and Maxillofacial Surgeons J Oral Maxillofac Surg 72:503-509, 2014
Bisphosphonate (BP) compounds have been synthesized since the 19th century, originally for use in industry. Although etidronate was first used clinically in the early 20th century, the applications of BPs became clarified and popularized only in the past 40 years.1 The principal action of BPs clinically is to inhibit bone resorption, although they have a secondary action of inhibiting mineralization.2 Several mechanisms of action have been proposed, including the inhibition
of osteoclast recruitment, adhesion, and activity and shortening of the osteoclast lifespan.3 The mechanism of action depends principally on the presence or absence of a nitrogen side chain in the BP structure, but different side chain structures have been associated with additional molecular actions.2,3 Osteonecrosis of the jaws (ONJ) is a complication of BP therapy first described almost 10 years ago.4 The pathophysiology of BP-related ONJ (BRONJ) is still
*Specialist Registrar, Department of Oral and Maxillofacial
This study was supported by a grant from the British Association
Surgery, Queen Alexandra Hospital, Portsmouth, United Kingdom.
of Oral and Maxillofacial Surgeons.
yResearch Fellow, Cancer Sciences Division, University of Southampton;
Clinical
Lecturer,
Cancer
Sciences
Address correspondence and reprint requests to Mr McLeod:
Division,
Department of Oral and Maxillofacial Surgery, The John Radcliffe,
University of Southampton, Southampton, United Kingdom.
Headley Way, Oxford OX39DU, UK; e-mail: [email protected]
zProfessor, Department of Oral and Maxillofacial Surgery, Queen
Received February 7 2013
Alexandra Hospital, Portsmouth, United Kingdom. xProfessor, Cancer Sciences Unit, University of Southampton,
Ó 2014 American Association of Oral and Maxillofacial Surgeons
Southampton, United Kingdom.
0278-2391/13/01033-1$36.00/0
Accepted August 1 2013
kSenior Research Fellow, Cancer Sciences Unit, University of
http://dx.doi.org/10.1016/j.joms.2013.08.007
Southampton, Southampton, United Kingdom.
503
504 poorly understood, and the relative importance of the effect of BPs on bone and soft tissues is unclear.5,6 Although it is known that BPs suppress bone turnover, how this translates into ulceration of the overlying mucosa is unclear. Although BPs are known to have an effect on mucosal tissues at high concentration, whether this effect is clinically relevant is unknown.7,8 This study compared the effect of 3 different BPs on human oral keratinocyte and fibroblast proliferation and migration.
Materials and Methods CELL CULTURE
Immortalized human fetal foreskin fibroblasts (HFFF2) were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) with 10% fetal bovine serum (FBS) and L-glutamine. Immortalized human oral keratinocytes (OKF6) were cultured in aMEM supplemented with 10% FBS, adenine, insulin, hydrocortisone, epidermal growth factor, and L-glutamine. Cells were cultured in a humidified atmosphere at 37 C and 10% CO2. BISPHOSPHONATE DRUGS
The chosen BP drugs reflected a spectrum of potencies in vivo and different modes of action.9 Zoledronate is a nitrogen-containing BP that is administered parenterally and is the most potent BP in vivo. Alendronate also contains nitrogen, but is principally administered orally and is the most potent oral BP. Clodronate is a non–nitrogen-containing BP that is principally administered orally and has relatively weak in vivo potency.9 PROLIFERATION ASSAY
The effect of different concentrations of BP on growth and viability of the chosen cell lines was assessed in the following way: 5 104 cells were seeded in 24-well plates with 500 mL of medium in the presence of H2O (control) or alendronate 104 to 107 mol/L, clodronate 102 to 105 mol/L, or zoledronate 0.2 106, 1 106, or 5 106 mol/L. Concentrations of each BP were selected based on previous in vitro cell studies.7-9 Each condition was tested in triplicate. At 1 day, 2 days, and 5 days, cells were trypsinized and counted using a Casy Counter (Roche Diagnostics, Penzberg, Germany).
IN VITRO EFFECT OF BISPHOSPHONATES
chosen were concentrations for the migration assays that did not produce a significant decrease in cell proliferation or affect cell viability in the previous assay. HFFF2 or OKF6 cells (5 104) were seeded in the upper chamber of Transwell inserts (8-mm pore; Corning Inc) with 100 mL of serum-free migration buffer (DMEM with 0.1% heat-inactivated bovine serum albumin [BSA] and L-glutamine for HFFF2 or aMEM with adenine, L-glutamine, and 0.1% heat-inactivated BSA for OKF6) in the presence of H2O or alendronate 105 or 106 mol/L, clodronate 106 or 107 mol/L, or zoledronate 0.2 106 or 1 106 mol/L and allowed to migrate toward serum-free migration buffer and H2O (negative control) or the growth medium of the respective cell line with a positive control without BP. After 24 hours, cells migrating to the lower chamber were trypsinized and counted on a Casy Counter. ORGANOTYPIC CULTURES
Although single-cell migration of HFFF2 and OKF6 cells was not affected by subtoxic concentrations of BPs, the effect of BPs on 3-dimensional cell invasion was investigated using an organotypic coculture model that replicates the oral mucosal architecture, which shows whether BP-modulated epithelial stratification, differentiation, or invasion occurs. Organotypic cultures composed of OKF6 keratinocytes and HFFF2 fibroblasts were generated. OKF6 cells were treated with the highest concentration of BPs, which did not significantly decrease cell viability in the proliferation assays. Organotypic cultures were prepared as described previously.10 OKF6 cells (1 106) were seeded on gels comprising a 50:50 mixture of Matrigel and type 1 collagen containing 5 105 HFFF2 cells per gel. The day after seeding, the gels were raised at the air-tissue interface. The medium under the gels was supplemented with H2O or alendronate 1 105 mol/L, clodronate 1 105 mol/L, or zoledronate 1 106 mol/L and was changed every second day. After 7 days, the gels were bisected, fixed in formal saline, and processed to paraffin. Four-micrometer sections were stained with hematoxylin and eosin. The assay was replicated twice. STATISTICAL ANALYSIS
Data were collected and analyzed using the Student unpaired t test and Excel (Microsoft, Redmond, WA). Each BP was compared with the control for each concentration and time point. A P value lower than .05 was accepted as statistically significant.
MIGRATION ASSAY
Transwell (Corning Inc, Tewksbury, MA) migration assays were performed to assess the effect of the different BPs on cell motility for each cell line. Proliferation assays had shown that high concentrations of all 3 BPs resulted in loss of cell viability; therefore, concentrations of BP
Results PROLIFERATION ASSAY
Results of the proliferation assay are shown in Figure 1.
505
MCLEOD ET AL
A
B
C
FIGURE 1. Proliferation assays of HFFF2 and OKF6 cell lines with different bisphosphonate drugs. A, Alendronate. B, Zoledronate. C, Clodronate. The growth ratio of cells cultured to cells counted at 1, 2, and 5 days is depicted. All 3 bisphosphonates showed a significant decrease in cell proliferation and viability in the 2 cell lines at the highest concentrations used. McLeod et al. In Vitro Effect of Bisphosphonates. J Oral Maxillofac Surg 2014.
At a concentration of 104 mol/L, alendronate significantly suppressed proliferation of HFFF2 compared with the control from 1 day (P < .05). For OKF6 cells, although a similar effect was shown for alendronate at 104 mol/L, there also was a significant decrease in proliferation shown at a concentration of 105 mol/L at 5 days (P = .002). Zoledronate significantly decreased the proliferation of HFFF2 after only 2 days at its highest concentration of 5 106 mol/L (P = .017) and at 5 days for OKF6 (P = .021) at the same concentration.
Clodronate significantly affected proliferation of HFFF2 and OKF6 after 1 day at a concentration of 102 mol/L (P = .014) and at a concentration of 103 mol/L significantly decreased the proliferation of HFFF2 at only 5 days (P = .049). Results indicated that clodronate also decreased the proliferation of OKF6 at 5 days at a concentration of 104 mol/L (P = .014), although not at a concentration of 103 mol/L (P = .34). The reason for this variant result is not clear and is thought unlikely to truly represent cellular effects.
506 MIGRATION ASSAY
The results of the migration assay are shown in Figure 2. At these subtoxic BP concentrations, there was no difference in HFFF2 or OKF6 migration between BPtreated and non–BP-treated cells (positive control, P > .05) and significantly more compared with cells without migration buffer (negative control, P > .05), indicating that BP had no effect on cell migration. ORGANOTYPIC CULTURES
Figure 3 shows the organotypic assays achieved for the 3 BPs. Th OKF6 cells showed squamous morphology and stratification, with no differences noted between the control and the cultures with any of the BPs. There was some evidence of invasion of the basement layer on the photomicrographs, which probably reflects the abnormal behavior of this immortalized cell line.
Discussion Osteonecrosis is an uncommon, but potentially serious, complication of BP therapy, which to date has been reported as affecting only the jawbones. The pathophysiology of the condition and the reason for its predilection for the jaws are still poorly understood. Multiple theories have been postulated, but it is increasingly recognized that the effect of BP on the oral soft tissues may play a role.5,6 Wound healing in the oral cavity involves cell proliferation and motility and a complex interplay of multiple cell types, including osteogenic cells, oral keratinocytes, fibroblasts, and endothelial cells. BRONJ most commonly presents after injury to the oral tissues (eg, after dental extraction) and its pathophysiology may arise from an effect on at least 1 of these individual cell types or an effect on their interactions. The effect of BPs on osteogenic cells has been studied extensively because these are the cells responsible for their principal therapeutic activity. The main focus of BP activity is the osteoclast. High BP concentrations result in osteoclast apoptosis, whereas lower concentrations inhibit bone resorption.3,9 BPs also are known to have an effect on osteocytes, where they have a protective effect against glucocorticoid-induced apoptosis.2 The effect of BPs on different soft tissues has been investigated for some time, because of the known side effects of BPs on the gastrointestinal tract, including esophagitis and gastric irritation.11,12 Oral soft tissue cells were initially investigated because of interest in the therapeutic use of BPs in dental trauma.7,8,11 An antitumor effect has been found with some BPs and is thought to be due to the induction of tumor cell apoptosis and the inhibition of tumor cell adhesion
IN VITRO EFFECT OF BISPHOSPHONATES
and invasion, and there has been interest in whether similar effects occur with oral soft tissue cells.13 With regard to the BPs studied, zoledronate has been shown to inhibit human endothelial cell proliferation and to modulate endothelial cell adhesion and migration.14 Walter et al15 found a decrease in cell viability of human umbilical vein endothelial cells in culture with the amino BPs and nonamino BPs. Moreira at al8 reported that alendronate mixed in a paste was rapidly toxic to endothelial cells at a concentration of 0.6 105 mol/L. In addition, when this paste was inserted subcutaneously in a rat, there was local tissue inflammation and necrosis. Correia et al7 found that use of alendronate at a concentration of 1 106 mol/L resulted in the loss of viability of human periodontal ligament cells and altered cell morphology at 1 105 mol/L. Scheper et al,16 using human gingival fibroblasts and human keratinocyte cell lines, also reported that zoledronate at a concentration of at least 2.5 107 mol/L resulted in cell apoptosis and at concentrations of 1 106 mol/L significantly decreased cell proliferation. These studies are similar to the present findings that high concentrations of BPs inhibit cell proliferation and result in a loss of cell viability. Amino BPs exert at least part of their effect by inhibiting prenylation of small guanosine triphosphatases (GTPases), which has an effect on osteoclast cell morphology, signaling, and apoptosis. Given the importance of small GTPases, such as Rac1 and cdc42, in cell motility, these BPs may suppress cell movement, and this may play a role in the pathophysiology of BRONJ. Various in vitro studies using a scratch woundhealing model have found that culturing cells with BPs can significantly delay scratch closure.12,17 However, in these studies, the concentrations of BP shown to suppress scratch closure also inhibited cell proliferation and decreased cell viability. The present study used Transwell migration assays to examine the motility of individual cells. The BP concentrations used for these assays were found not to have a significant effect on cell proliferation or viability, and at such concentrations no effect on cell migration was observed. Organotypic assays recapitulate the morphology of the oral mucosa and were used to assess the effect of the drugs on epithelial stratification and differentiation. The authors did not find any effect on epithelial stratification of BP at the chosen concentrations, which were concentrations shown not to have a significant effect on cell proliferation. Much has been made of the effect of different BPs on various fibroblast and keratinocyte cell lines, most of which are from nonhuman8,12 or nonoral cavity8,13
507
MCLEOD ET AL
FIGURE 2. Migration assays of HFFF2 and OKF6 cell lines with 3 different bisphosphonate drugs at subtoxic concentration. A, Alendronate. B, Zoledronate. C, Clodronate. Migration toward SF and H2O (column 1, negative control) was compared with migration toward FBS (column 2, positive control) and with 2 different subtoxic concentrations of each bisphosphonate (columns 3 and 4). No decrease in cell migration was found for the 3 bisphosphonates with either cell line. FBS, fetal bovine serum; SF, serum-free migration buffer. McLeod et al. In Vitro Effect of Bisphosphonates. J Oral Maxillofac Surg 2014.
cell lines. In line with the demonstration of mandibular and iliac crest bone marrow stem cells having different responses to BP, the choice of soft tissue cell line in experiments may prove to be more important than previously recognized. Malignant cell
lines also may behave differently, and contamination or transformation of immortalized nonmalignant cell lines make interpretation of results difficult to compare. Therefore, in vitro findings may not be directly comparable to the clinical setting.17
508
IN VITRO EFFECT OF BISPHOSPHONATES
FIGURE 3. Photomicrographs of organotypic culture of HFFF2 and OKF6 cells cultured with bisphosphonate drugs. A, Control. B, Alendronate at a concentration of 1 105 mol/L. C, Zoledronate at a concentration of 1 106 mol/L. D, Clodronate at a concentration of 1 105 mol/L. McLeod et al. In Vitro Effect of Bisphosphonates. J Oral Maxillofac Surg 2014.
The present results combined with those of previous studies confirm that, at high concentrations, BP drugs are toxic to different soft tissue cell lines, and that at these concentrations there is inhibition of healing using a scratch assay model. However, at subtoxic concentrations, there are no specific effects on singlecell motility or epithelial stratification. It is questionable whether induction of cell toxicity is the principal pathophysiologic mechanism in BRONJ, because oral soft tissues may never be exposed to the high concentrations of BPs required to induce cell death in vitro. Circulating BPs are rapidly incorporated into mineralized bone, within which it is metabolically inactive, and the plasma concentration of BP rapidly decreases after drug administration to levels that have been shown to have no functional effect on cells.1,18,19 For example, zoledronate has been shown to have a peak plasma concentration lower than 400 ng/mL, and this decreases to below 1% within 24 hours.18 The use of BP concentrations in vitro, which are comparable to plasma concentrations immediately after administration, is therefore questionable, because cells are unlikely to be subject to this concentration for extended periods in vivo. BPs are released during bone resorption and taken up by the osteoclasts on which they exert their effect. Although BP concentrations may reach high levels within the resorption space and Sato et al20 estimated that alendronate may reach concentrations of 0.5 mol/L, there is no clear evidence that similar concentrations may be found around the oral mucosal cells for any period.5,20,21
The present study and other studies have shown that at low concentrations BPs appear to have little effect on the proliferation and migration of fibroblasts or keratinocytes, whereas at high concentrations the drugs are significantly toxic. Many questions remain about the role of soft tissues in the pathophysiology of BRONJ. One of the key questions is the true concentration exposure of the tissues to BPs, because that would clarify the relevance of the in vitro findings of cell function inhibition.
References 1. Fleisch H: Bisphosphonates: Mechanisms of action. Endocr Rev 19:80, 1998 2. Russell RGG, Watts NB, Ebetino FH, et al: Mechanism of action of bisphosphonates: Similarities and differences and their potential influence on clinical efficacy. Osteoporos Int 19:733, 2008 3. Rogers MJS, Gordon HL, Benford FP, et al: Cellular and molecular mechanism of action of bisphosphonates. Cancer 88(suppl 12): 2961, 2000 4. McLeod NMH, Brennan PA, Ruggiero SL: Bisphosphonate osteonecrosis of the jaw: A historical and contemporary review. Surgeon 10:36, 2012 5. Allen MR, Burr DB: The pathogenesis of bisphosphonate related osteonecrosis of the jaw: So many hypotheses, so few data. J Oral Maxillofac Surg 67(suppl 1):61, 2009 6. Reid IR, Bolland MJ, Grey AB: Is bisphosphonate-associated osteonecrosis of the jaw associated with soft tissue toxicity? Bone 41:318, 2007 7. Correia VFP, Caldeira CL, Marques MM: Cytotoxicity evaluation of sodium alendronate of cultured human periodontal ligament fibroblasts. Dent Traumatol 22:312, 2006 8. Moreira MS, Katayama E, Bombana AC, et al: Cytotoxicity analysis of alendronate on cultured endothelial cells and subcutaneous tissue. A pilot study. Dent Traumatol 21:329, 2005 9. Ringe JD: The interpretation of preclinical data in predicting bisphosphonate response in the treatment of osteoporosis. Clin Therap 20:648, 1998
MCLEOD ET AL 10. Nystrom ML, Thomas GJ, Stone M, et al: Development of a quantitative method to analyse tumour cell invasion in organotypic culture. J Pathol 205:468, 2005 11. Twiss IM, Pas O, Ramp-Koopmanschap W, et al: The effects of nitrogen-containing bisphosphonates on human epithelial (caco-2) cells, an in vitro model for intestinal epithelium. J Bone Miner Res 14:784, 1999 12. Landesberg R, Cozin M, Cremers S, et al: Inhibition of oral mucosal cell wound healing by bisphosphonates. J Oral Maxillofac Surg 66:839, 2008 13. Santini D, Vespasiani Gentilucci U, Vincenzi B, et al: The antineoplastic role of bisphosphonates: From basic research to clinical evidence. Ann Oncol 14:1468, 2003 14. Wood J, Bonjean K, Reutz S, et al: Novel antiangiogenic effects of the bisphosphonate compound zoledronic acid. J Pharmacol Exp Therap 302:1055, 2002 15. Walter C, Klein MO, Pabst A, et al: Influence of bisphosphonates on endothelial cells, fibroblasts and osteogenic cells. Clin Oral Invest 14:35, 2010
509 16. Scheper MA, Badros A, Chaisuparat R, et al: Effect of zoledronic acid on oral fibroblasts and epithelial cells: A potential mechanism of bisphosphonate-associated osteonecrosis. Br J Haematol 144:667, 2008 17. Stefanik D, Sarin J, Lam T, et al: Disparate osteogenic response to mandible and iliac crest bone marrow stromal cells to pamidronate. Oral Dis 14:465, 2008 18. Skerjanec A, Berenson J, Hsu A, et al: The pharmacokinetics and pharmacodynamics of zoledronic acid in cancer patients with varying degrees of renal function. Clin Pharmacol 43: 154, 2003 19. Lin JH: Bisphosphonates: A review of their pharmacokinetic properties. Bone 18:75, 1996 20. Sato M, Grasser W, Endo N, et al: Bisphosphonate action. Alendronate localistaion in rat bone and effects on osteoclast ultrastructure. J Clin Invest 88:2095, 1991 21. Scheper MA, Chaisuparat R, Cullen KJ, et al: A novel soft-tissue in vitro model for bisphosphonate-associated osteonecrosis. Fibrin Tiss Rep 3:6, 2010
