Sage Journals: Discover world-class research

Abstract

Data arising from the recent literature directed the researchers to study on the degree and extent of bisphosphonate toxicity on oral mucosa in further detail. The aim of this study is to determine the half maximal inhibitory concentration of pamidronate (PAM) and alendronate (ALN) on human gingival fibroblasts in vitro using 3-[4.5-thiazol-2-yl]-2.5-diphenyltetrazolium bromide (MTT) assay and to evaluate the effects of both agents on the proliferation and apoptotic indices. Cells used in the study were generated from human gingival specimens and divided into alendronate (n = 240), PAM (n = 240), and control groups (n = 60). Based on the MTT assay results, 10⁻⁴, 10⁻⁵, 10⁻⁶, and 10⁻⁷ M concentrations of both drugs were administered and the effects were evaluated for 6, 12, 24, 48, or 72 h periods. An indirect immunofluorescence technique was used to evaluate apoptotic (anti-caspase 3) and proliferation (anti-Ki67) indices. Toxicity of both PAM and ALN was found to be the most potent at 10⁻⁴–10⁻⁵ M range. The apoptotic index of PAM group was found to be significantly higher than ALN group for all concentrations especially at 24 h incubation time (p < 0.05). The decrease in the proliferation index was found similar in first 48 h for both drugs; however, after 72 h of incubation decrease in proliferation index in PAM group was found to be significantly higher (p < 0.05). Micromolar concentrations of not only PAM but also ALN rapidly affect cells generated from human oral gingival tissue by inducing apoptosis together with inhibition of proliferation. Cytotoxic effects of both ALN and PAM on primary human gingival fibroblasts, which cause significant changes in apoptotic and proliferative indices as shown in this in vitro study, suggests that the defective epithelialization of oral mucosa is possibly a major factor on the onset of bisphosphonate-related osteonecrosis of the jaw cases.

Keywords

Bisphosphonate-related osteonecrosis of the jaw alendronate oral mucosa fibroblast apoptosis/cell death toxicity testing

Introduction

Bisphosphonates (BPs) are potent inhibitors of bone resorption with long-lasting effects.^1,2 These drugs suppress the actions of osteoclasts and, thus, reduce bone resorption and increase bone density.^3,4 Today, BPs are widely prescribed for prevention and/or treatment of osteoporosis, multiple myeloma, malignant bone metastases, and Paget’s disease. Bisphosphonate-related osteonecrosis of the jaw (BRONJ) is accepted as the main serious side effect of long-term and high-dose BP usage.⁵ This condition is defined as an area of exposed necrotic bone in the maxillofacial region of a patient with no history of head or neck irradiation and no evidence of healing at lesion site for minimum 8 weeks.⁶ Defective or delayed epithelialization of oral mucosa has also been observed in almost all cases of BRONJ.⁷ It is generally agreed that the primary lesion for this condition is in the bone tissue; however, defective or delayed epithelialization in oral mucosa raised questions about the possible role of soft tissue in the pathogenesis of BRONJ.⁸

There are several reports regarding the toxic effects of BPs on oral mucosa. Investigators focused on the mechanism of the effects of BPs and determining cell proliferation rates and viability of fibroblasts and/or epithelial cells, the degree of the induction of apoptotic changes, migratory capacity of the cells, and the rate of collagen production using either cell lines or primary culture cells.^{9

–19}

As the information on the issue is still limited, this study is designed to outline the toxic effects of two common BP preparations in dose- and time-related manner in cultured human gingival fibroblasts. Among BP preparations, parenterally administered pamidronate (PAM) and orally administered alendronate (ALN) were selected for comparison. The previous reports are mainly focused on the effects of zolendronate (ZOL). In addition, there are several studies for PAM and a few for ALN. However, only one study has reported the comparative effects of both agents on collagen synthesis and proliferation rate so far.¹⁷ Thus, the aim of this study is to determine (i) half-maximal inhibitory concentrations (IC₅₀) of both ALN and PAM and (ii) the in vitro apoptotic and cytotoxic effects of these agents on primary human gingival fibroblasts at certain concentrations and time.

Methods

This study was approved by the Baskent University Institutional Review Board and Ethics Committee (Project No: D-KA 10/09). It was conducted in accordance with the Helsinki Declaration.

Primary fibroblast cell cultures

The gingival fibroblasts were derived from full-thickness gingival tissue specimens of six healthy patients (3 females and 3 males, aged 18–30). All included patients signed an inform consent. The specimens obtained from gingival tissue were released when impacted teeth were exposed during orthodontic tooth eruption procedures. Three groups were established for the study, namely, an ALN group, a PAM group, and an untreated control group.

Gingival tissue specimens were transferred to the laboratory in 3 ml transport medium Dulbecco’s modified Eagle’s medium (DMEM)-F12 (Biochrom AG, Germany) containing 1% streptomycin, neomycin, and penicillin mixture (Biological Industries, Israel). Initially, under sterile conditions, the specimen was cut into small pieces using a scalpel and enzymatically digested with type I collagenase (Biochrom AG, Germany) at a concentration of 5 mg collagenase/10 ml DMEM-F12 in a 37°C incubator for 60 min.⁹ The mixture was then centrifuged at 500g for 9 min. After the supernatant was removed, the cells were placed in cell culture medium comprised of DMEM-F12 solution containing 10% heat-inactivated fetal bovine serum (FBS; Biochrom AG, Germany), 2% l-glutamine (Biochrom AG, Germany), and 1% streptomycin, neomycin, and penicillin mixture. The plates were incubated at 37°C in a humidified atmosphere with 5% carbon dioxide (Heraeus, Hanau, Germany). The cell culture medium was refreshed every 2 days, and cells were subcultured when confluence was 80%. Fourth to sixth subculture of cells was used as experiment group. Viability of cells was evaluated using the trypan blue exclusion test.

Determination of cytotoxic drug concentration

Cytotoxic concentrations of ALN and PAM were determined using 3-[4.5-thiazol-2-yl]-2.5-diphenyltetrazolium bromide (MTT, Sigma, St Louis, Missouri, USA). Each drug was serially diluted from high to low concentrations on 96-well plates (maximum 10⁻⁴ M to minimum 5 × 10⁻⁹ M). A total of 1 × 10⁴ cells were then seeded to each drug containing well, and plates were incubated at 37°C for 48 h and 72 h. Twenty microliters of 5 mg/ml MTT (final volume: 0.59 mg/ml) was added to each well, and the plates were again incubated at 37°C for 4 h. As the final incubation step, 100 µl 10% of sodium dodecyl sulfate solution was added to each well, and the plates were incubated at 37°C overnight. The optical density of the chromogenic product was measured at 540 nm in an enzyme-linked immunosorbent assay (ELISA) reader (Biotech Instrument ELx800, Winooski, Vermont (VT), USA). Each drug’s IC₅₀ value was calculated from the logarithmic trend line equation of the ELISA reader outcome.

Treatments

Pure salt forms of ALN (alendronate sodium trihydrate, Sigma) and PAM (pamidronate disodium salt hydrate, Sigma) were applied to gingival fibroblasts in 10⁻⁴–10⁻⁷ M concentrations for 6, 12, 24, 48 and 72 h. Fibroblasts were grown in a six-well plate including in a lamella. After applications, the lamella was removed and dried at 4°C overnight.

Determination of proliferative index and apoptosis by immunohistochemical staining

Indirect immunofluorescence technique as described in Kaymaz et al.²⁰ was used in this study. Mouse anti-human caspase 3 (1123R908A, Thermo Scientific, Waltham, Massachusetts, USA), rabbit anti human Ki-67 (9106R1003A, Thermo Scientific) and mouse anti-human vimentin (M7020, Dako, Denmark) monoclonal antibodies were used as primary antibodies. Polyclonal rabbit anti-mouse immunoglobins fluorescein isothiocyanate (FITC; F0313, Dako) and polyclonal mouse anti-rabbit immunoglobins FITC (F0232, Dako) were used as secondary antibodies for appropriate primary antibodies.

First, each lamella was fixed with 96% acetone and a hydrophobic barrier pen was used to divide the lamella into the following four areas: (i) a positive control area in which anti-vimentin and polyclonal rabbit anti-mouse immunoglobulins FITC were applied; (ii) an area for evaluating apoptosis in which anti-caspase 3 and anti-mouse immunoglobulins FITC were applied; (iii) an area for evaluating cell proliferation in which anti-Ki-67 and polyclonal mouse anti-rabbit immunoglobulins FITC were applied; and (iv) a negative control area in which the primary antibody step was omitted and only polyclonal mouse anti-rabbit immunoglobulins FITC or anti-mouse immunoglobulins FITC were applied. For nuclear counterstaining, 10 µg/ml ethidium bromide (E8751, approximately 95% (high-performance liquid chromatography), Sigma Aldrich, 1G) solution was used in all slides.

Vimentin reactivity was referenced for both the characterization of cultured cells and as the positive control for the immunostaining. The cytoplasm of all cells exhibited strong reactivity for vimentin with their ethidium bromide-stained nuclei (Figure 1(a) and (b)). Thus, cultured cells were characterized as fibroblasts with their elongated processes also reactive for vimentin.

Figure 1.

(a) Strong fibrous reactivitiy using anti-vimentin antibody (green) is seen in cytoplasm of fibroblast; nuclei are counterstained with ethidium bromide (red). (pamidronate group for 10⁻⁶ M concentration at 72 h administration, original magnification ×100; indirect immunoflourescence – FITC labeled.). (b) The reactivity of anti-vimentin antibody (green) is seen in specific shaped cytoplasm of fibroblast; nuclei are counterstained with ethidium bromide (red). (pamidronate group for 10⁻⁶ M concentration at 12 h drug administration, original magnification ×100; indirect immunoflourescence—FITC labeled.). FITC: fluorescein isothiocyanate.

Two blinded researchers (SSS and HF) used a camera-equipped fluorescence microscope (Nikon Eclipse E600, Japan) to assess for caspase 3 and Ki-67 staining. First, apoptotic index (AI) and cell proliferation index (PI) were calculated independently by each researcher (ratio of effected cells to a total of 100 cells counted). Next, mean values of AI and PI were determined. To demonstrate apoptosis (caspase 3; Figure 2(a) and (b)) and proliferating (Ki-67; Figure 3(a) and (b)) of fibroblasts, digital microphotographs of some samples were taken.

Figure 2.

(a) and (b) Fluorescent photomicrographs of immunostaining using anti-caspase 3 show anti-caspase 3 reactive fibroblasts characterized by their irregular shape and appearing greatly shrunken with karyorrhectic nuclei (I); nuclei are counterstained with ethidium bromide (red). (original magnification (a) ×40, (b) ×100; indirect immunoflourescence—FITC labeled.). FITC: fluorescein isothiocyanate.

Figure 3.

(a) An example of speckled staining with anti-Ki-67 is observed in two small nuclei of fibroblast in the alendronate group for 10⁻⁴ M concentration at 6 h drug administration (∀); nuclei are counterstained with ethidium bromide (red); original magnification ×100; indirect immunoflourescence—FITC labeled.). (b) Fluorescent photomicrographs of immunostaining using anti-Ki-67 show an example of the reaction to be spread also to the cytoplasm in the alendronate group for 10⁻⁶ M concentration at 48 h drug administration (∀); nuclei are counterstained with ethidium bromide (red; original magnification ×100; indirect immunoflourescence—FITC labeled.). FITC: fluorescein isothiocyanate.

Statistical analysis

Analyses were performed using the Statistical Package for the Social Sciences (SPSS v. 17.0, IBM, Chicago, Illinois, USA). Variances of means were compared analysis of variance (ANOVA) test. The assumption of sphericity precondition for repeated-measures ANOVA was assessed using Mauchly’s sphericity test. When this assumption was violated, the Greenhouse Geiser repeated-measures ANOVA with correction for degrees of freedom was used. Whether these adjustments were not appropriate, additional Bonferroni–Dunn Friedman test was performed for multiple comparisons. Finally, the values of p < 0.05 were considered statistically significant.

Results

Cell viability and detection of IC₅₀ values

IC₅₀ values were found to be 0.041 and 0.350 µM for PAM and 0.355 and 7.7 µM for ALN at 48 h and 72 h, respectively. Accordingly, cell viability was significantly reduced following 48 h of drug administration, and this reduction was more prominent at 72 h for both ALN and PAM. The cytotoxicity of PAM was higher than ALN for both 48 h and 72 h (almost 10-fold). However, besides the higher cytotoxity rate of PAM, the time-dependent increase of the toxicity was only 10-fold. Viability of the cells significantly diminished below 20% when 10 µM or higher drug concentrations were administered. Regarding all data, 0.1–100 µM (10⁻⁷–10⁻⁴ M) drug concentrations were included for the experimental group design of the study.

Analysis of dose- and time-dependent changes for all groups

Depending on predetermined drug concentrations and timing, statistically significant differences were present between the AI and PI in all control, ALN, and PAM groups. More detailed statistical analysis of dual comparison of the groups regarding apoptosis and cell proliferation are summarized below.

Apoptotic indices

The mean AI for ALN, PAM, and control groups and their statistical comparison are shown on Table 1.

Table 1.

Mean apoptotic index values of alendronate, pamidronate, and control groups.^a

	Time (h)	Concentration (10^−M)	Alendronate group	Pamidronate group	Control group
	Time (h)	Concentration (10^−M)	Mean (±SD)
Caspase 3 (apoptotic index)	6	4	17.8^b (3.1)	22.4^b (2.5)	2 (0.5)
		5	13.6 (3.4)	14.4^b (2.7)
		6	8 (1.8)	8.8 (1.5)
		7	2.8 (0.9)	4.4 (1.6)
	12	4	14.8^b (2.9)	20^b (1.9)	2.6 (0.8)
		5	11.4^b (2.2)	16^b (3.9)
		6	7 (1)	11.2 (4.07)
		7	2.8 (0.9)	9.2 (3.89)
	24	4	24.2^b (1.6)	53^b,c (9.3)	2.2 (1.3)
		5	15.6^b (1.8)	42^b,c (11.8)
		6	8.6 (1.8)	18.8^a (3.8)
		7	6 (1.3)	9.8^a (1.6)
	48	4	31.8^a (7.5)	53.4^c (7.02)	3.4 (0.7)
		5	23.2 (3.9)	34.6^b (6.6)
		6	10.2 (1.1)	20.2^c (1.7)
		7	6.4 (1.6)	10.8^c (1.3)
	72	4	48.8^b (7.6)	61.8^b (5.5)	4.8 (1.1)
		5	29.4 (1.4)	53^a,b (7.02)
		6	13.4 (3.4)	34.6^a (6.6)
		7	8.6 (1.9)	20.2^a (1.74)

^aComparative statistical analysis of the data according to drug concentration and incubation time. SD: standard deviation.

^bDifferent from control group.

^cDifferent from alendronate group.

When comparing ALN with the control group, at minimum drug concentration (10⁻⁷ M) and drug incubation time (6 h), the AI of the ALN and control group was 2.8% and 2%, respectively (p > 0.05). At maximum drug concentration (10⁻⁴ M) and maximum drug incubation time (72 h), the AI of the ALN and control group was 48.8% and 4.8%, respectively (p < 0.05). Differences between the AI of ALN and control groups at 10⁻⁴ M concentration for all incubation times were statistically significant.

When comparing PAM with the control group, at minimum drug concentration (10⁻⁷ M) and drug incubation time (6 h), the AI of the PAM and control group was 4.4% and 2%, respectively (p > 0.05). At maximum drug concentration (10⁻⁴ M) and maximum drug incubation time (72 h), the AI of the PAM and control group was 61.8% and 4.8%, respectively (p < 0.05). There were statistically significant differences between the AI of PAM and control groups at 10⁻⁴–10⁻⁵ M drug concentrations for all incubation times.

When comparing ALN with the PAM group, at minimum drug concentration (10⁻⁷ M) and drug incubation time (6 h), the AI of the ALN and PAM groups was 2.8% and 4.4%, respectively (p > 0.05). At maximum drug concentration (10⁻⁴ M) and maximum drug incubation time (72 h), the AI of the ALN and PAM group was 48.8% and 61.8%, respectively (p < 0.05). The AI of PAM group was significantly higher than ALN group for all drug concentrations at 24 h incubation period.

Proliferation indices

The mean PI for ALN, PAM, and control groups and their statistical comparison are summarized in Table 2.

Table 2.

Mean proliferative index of alendronate, pamidronate, and control groups.^a

	Time (h)	Concentration (10^−M)	Alendronate group	Pamidronate group	Control group
	Time (h)	Concentration (10^−M)	Mean (±SD)
KI-67 (proliferative index)	6	4	30.6^b (5.1)	27^b (1.4)	73.6 (5.21)
		5	31.4^b (5.06)	28.4^b (3.8)
		6	41.2 (5.1)	51.2 (6.4)
		7	57.8 (3.8)	62 (3.01)
	12	4	35.4^b (5.4)	32.6^b (4.1)	68.6 (4.04)
		5	44.8 (7.2)	41.4 (4.2)
		6	49.6 (6.1)	52.8 (5.6)
		7	60.6 (4.8)	58.6 (5.1)
	24	4	27.6^b (3.3)	26.8^b (3.6)	63.8 (4.7)
		5	34.8 (3.4)	33 (4.9)
		6	49.8 (5.6)	44 (4.3)
		7	63.2 (6.1)	56.6 (6.01)
	48	4	19.4^b (3.3)	24^b (5.3)	65.2 (10.6)
		5	31.8 (3.4)	31.8^b (4.8)
		6	38.8 (4.7)	46 (9.5)
		7	50.8 (4.4)	56.6 (9.9)
	72	4	18.8^b (1.4)	11^b,c (1.4)	55 (4.7)
		5	27 (3)	19^b,c (3.2)
		6	32.6 (2.7)	28 (5.2)
		7	47.4 (4.5)	36.2 (3.5)

^aComparative statistical analysis of the data according to drug concentration and incubation time. SD: standard deviation.

^bDifferent from control group.

^cDifferent from alendronate group.

When comparing ALN with the control group, at minimum drug concentration (10⁻⁷ M) and drug incubation time (6 h), the PI of the ALN and control group was 57.8% and 73.6%, respectively (p > 0.05). At maximum drug concentration (10⁻⁴ M) and maximum drug incubation time (72 h), the PI of the ALN and control group was 18.8% and 55%, respectively (p < 0.05). There were statistically significant differences between the PI of ALN and control groups for 10⁻⁴ M drug concentration at all incubation times.

When comparing ALN with the control group, at minimum drug concentration (10⁻⁷ M) and minimum incubation time (6 h), PI of PAM and control group was 62% and 73.6%, respectively (p > 0.05). At maximum drug concentration (10⁻⁴ M) and maximum drug incubation time (72 h), the PI of the PAM and control group was 11% and 55%, respectively (p < 0.05). PI of PAM group was significantly lower than control group especially for 10⁻⁴ M concentration for each period of time and for 10⁻⁵ M concentration at 48–72 h incubation periods.

When comparing ALN with the PAM group, at minimum drug concentration (10⁻⁷ M) and drug incubation time (6 h), the PI of the ALN and PAM group was 57.8% and 62%, respectively (p > 0.05). At maximum drug concentration (10⁻⁴ M) and maximum drug incubation time (72 h), the PI of the ALN and PAM group was 18.8% and 11%, respectively (p < 0.05). Statistically significant difference was only observed between ALN and PAM groups for 10⁻⁴–10⁻⁵ M drug concentration at 72 h drug incubation.

Discussion

Clinical management of BRONJ cases is still one of the major problems faced in Oral and Maxillofacial Surgery Departments. Extensive research has been undertaken to elucidate the pathogenesis of BRONJ. While the majority of this work was focused on bone tissue, many investigators also examined the role of soft tissue responses in BRONJ since 2008. Although the effect of BPs’ on many cell types and also some tissue samples belonging to BRONJ patients were investigated,^21

–26 there is only limited data on cultured gingival fibroblasts.^{10,12
–14,16

–19} Therefore, this study was designed to investigate the possible in vitro cytotoxic effects of ALN and PAM on primary human gingival fibroblasts according to apoptotic and proliferative changes. Primary cultures of human gingival fibroblasts were preferred over established cell lines since they mimic in vivo conditions better.

According to Simon et al.¹⁶ both ZOL and PAM are toxic for gingival fibroblasts. Similarly, another group reported that ZOL impedes proliferation and migratory capacity of gingival fibroblasts and downregulate the transcription of type-I collagen in keratinocyte and fibroblastic cell lines.¹³ Açil et al.¹⁷ demonstrated that ALN is also toxic and decreases collagen expression, alkaline phosphatase activity, and osteocalcin synthesis in gingival fibroblasts in addition to ZOL and PAM. All these reports suggest that BPs may interfere with the onset and development of BRONJ since fibroblastic activity was shown to be severely affected by these agents as the chief element of wound healing.

Reports on the issue, including the above mentioned, are focused on the apoptotic changes induced by BPs. Apoptotic changes were triggered by a variety of factors discussed in detail so far.²⁷ There are several reports in which BPs are accepted as potential toxins and extrinsic inducers of apoptosis in oral mucosa cells.^11–12 It is still controversial whether BPs in alveolar bone, saliva, or gingival cervicular fluid might have apoptotic effects on oral mucosa. Allen and Burr²⁶ reported direct toxicity of BPs on oral epithelium. Induction of apoptosis in oral epithelium may lead to bone exposure and restrict soft tissue healing, which is a characteristic feature of BRONJ. It also seems to be crucial to determine the apoptotic effects of different BPs on oral mucosa to understand this complex and unresolved process.

Although the cytotoxic effects of ALN to gastrointestinal mucosa are previously well documented by clinical and in vitro studies,^28
–30 the effect of ALN on apoptosis of human gingival fibroblasts has not been previously reported. The results of the current in vitro study present the apoptotic effect of ALN (which reached 50% at toxic concentration) on gingival fibroblasts and in concordance suggest a potential role in BRONJ.

In patients who receive BPs, the known plasma level of ALN and PAM is 0.116 µM and 0.137 µM, respectively.³¹ However, the concentration of ALN or PAM at gingival tissue following systemic administration has not been determined to date. Thus, another primary goal of this study was to determine in vitro toxic drug concentrations of both agents for different administration periods. For this, we performed MTT analysis with a wide concentration range of PAM and ALN from 10⁻⁴–10⁻⁹ µM. The determined toxic concentration range was 0.1–100 µM for each time period. Our finding of the lowest toxic drug concentration (0.1 µM) is approximately equal to the pre-reported plasma level of these agents in BP-administered patients.³¹ However, no statistically significant change regarding the AI or PI was found between the control and experimental groups at the lowest toxic concentration, although a mild change was present. Similar findings were also evident for different time intervals for 1 µM ALN and PAM concentrations. The concentrations above 10 µM were clearly toxic as reflected by the statistically significant changes in AI and PI for all time intervals.

Different BP concentrations in a range of 0.1 µM to 0.3–0.5 mM were tested in previous studies. Some of the cytotoxicity studies were carried out on keratinocyte cell lines. The study by Landesberg et al.⁹ was the first to investigate the cytotoxic effects of BPs on murine oral epithelial cell cultures. Their results confirmed that 0.03–0.3 mM PAM inhibited proliferation and migration of oral keratinocytes but did not induce apoptosis of these cells. In contrast to Landesberg et al.’s findings, Kobayashi et al.¹¹ showed that 1–10 µM ZOL induced apoptosis and inhibited migration of keratinocytes of murine oral mucosa cell cultures. The latter studies were designed to investigate and solve these controversies, followed by the human gingival cell lines studies. Pabst et al.¹⁵ investigated the cytotoxic effects of PAM, ZOL, and some other non-nitrogen-containing BPs on gingival keratinocytes cell lines. They found that both PAM and ZOL induced apoptosis at 50 µM and decreased cell viability of gingival keratinocytes at 5 µM. In another study, Kim et al.¹⁹ evaluated the cytotoxic effects of PAM on three-dimensional synthetic fibroblast and keratinocyte cell cultures. They reported that 10 µM PAM induced keratinocyte senescence and had no cytotoxic effect on fibroblasts. They also reported that 50 µM PAM significantly increased apoptosis of fibroblasts and had no apoptotic effect on keratinocytes. Walter et al.^12,14 reported similar results and showed that 50 µM PAM significantly induced apoptosis and reduced the viability of fibroblast cultures. However, our results indicated that the lowest toxic PAM concentration of 10 µM induced significant apoptotic changes and decreased proliferation of primary gingival fibroblasts.

The results of this study indicate that both concentration and incubation time greatly influences the AI and PI in all ALN and PAM treatment groups. AI in PAM group was higher than that of ALN group for all incubation periods. The highest increase of the AI was observed at 24 h in PAM group. AI reached 62% in PAM group and 49% in ALN group at 100 µM and at 72 h. The decrease of PI in ALN group was found similar to PAM group in first 48 h, and then the decrease of PI was found higher in PAM group at the end of 72 h incubation period. PI decreased to 11% in PAM group and 18.8% in ALN group at 100 µM concentration and at 72 h. Agis et al.²³ and Açil et al.¹⁷ implied that the cytotoxic potency of ZOL was the highest among BPs tested, followed by ALN and PAM thereafter. Interestingly our findings showed that PAM had stronger apoptotic and antiproliferative effects than ALN. However, the cytotoxicity of ALN could not be underestimated.

Scheper et al.⁵ showed the release of low levels of the ZOL (0.25–3 µM) from bone, which may induce mucosal cell apoptosis and proliferation inhibition. They concluded that this mechanism may constitute an initiating mechanism for BRONJ. Whereas BP release from infected or disrupted bone may explain the delayed healing of oral mucosa that afflicts patients with BRONJ, spontaneous occurrence of this condition in the absence of infection or dentoalveolar surgical procedures has not yet been explained.³² The results of our in vitro study also support the presence cytotoxic effects of ALN and PAM shown by significant changes in PI and AI of human gingival fibroblasts.

Currently, main tool Ki-67 expression levels were significantly changed. More detailed molecular and histological studies are required to understand the overall mechanism of soft tissue damage in BRONJ. Additionally, new strategies should be developed to prevent or reverse the caspase used for the management of BRONJ to try stimulating soft tissue healing. Regarding the previous and present findings, approaches aiming to prevent apoptosis may also be beneficial as it seems to be a key limiting factor for the soft tissue damage observed in BRONJ. Since caspase 3 is one of the main regulators of cellular apoptosis, blocking of caspase 3 and 9 by geranylgeraniol or recombinant human platelet-derived growth factor may be effective as reported previously.^10,18,33 Saito et al.³⁴ demonstrated that ZOL significantly suppresses the Ki-67 labeling index, cause downregulation of TGF-β/Smad signaling in oral keratinocytes, and thus may be the main factor for the impairment observed in reepithelialization in BRONJ cases. Our results also support this suggestion as caspase 3 and mechanism of oral soft tissue apoptosis and inducing proliferation of oral mucosal cells in patients under BPs regimen.

In conclusion, here we report that (1) both ALN and PAM induce apoptosis and inhibit proliferation of human gingival fibroblasts in a dose- and time-dependent manner; (2) in vitro IC₅₀ concentrations of both ALN and for PAM is in 10⁻⁴–10⁵ M range; and (3) cytotoxic effects of PAM and ALN on soft tissue cells such as fibroblasts may have a significant role in the pathogenesis of BRONJ.

Footnotes

Conflict of interest

The authors declared no conflicts of interest.

Funding

This study was approved by Baskent University Institutional Review Board (Project no: D-KA 10/09) and supported by Baskent University Research Fund.

References

Woo

Hellstein

Kalmar

. Narrative (corrected) review: bisphosphonates and osteonecrosis of the jaws. Ann Intern Med 2006; 144: 753–761.

Gómez Font

Martínez García

Olmos Martínez

. Osteochemonecrosis of the jaws due to bisphosphonate treatments: update. Med Oral Patol Oral Cir Bucal 2008; 13: 318–324.

O’Ryan

Gordon

Predicting risk of osteonecrosis of the jaw with oral bisphosphonate exposure (PROBE)investigators. Prevalence of osteonecrosis of the jaw in patients with oral bisphosphonate exposure. J Oral Maxillofac Surg 2010; 68: 243–253.

Martin

Grin

. Bisphosphonates-mechanism of action. Aust Prescr 2000; 23: 130–132.

Scheper

Chaisuparat

Cullen

A novel soft-tissue in vitro model for bisphosphonate-associated osteonecrosis. Fibrogenesis Tissue Repair 2010; 1: 3–6.

Marx

. Pamidronate (Aredia) and zoledronate (Zometa) induced avascular necrosis of the jaws: a growing epidemic. J Oral Maxillofac Surg 2003; 61: 1115–1117.

Lorenzo

Trapassi

Corradino

Histology of the oral mucosa in patients with BRONJ at III stage: a microscophic study proves the unstability of local mucosa flaps. J Clin Med Res 2013; 5: 22–25.

Reid

. Osteonecrosis of the jaw — who gets it, and why? Bone 2009; 44: 4–10.

Landesberg

Cozin

Cremers

Inhibition of oral mucosal cell wound healing by bisphosphonates. J Oral Maxillofac Surg 2008; 66: 839–847.

10.

Sheper

Badros

Chaisuparat

Effect of zoledronic acid on oral fibroblasts and epithelial cells: a potential mechanism of bisphosphonate-associated osteonecrosis. Br J Haematol 2009; 144: 667–676.

11.

Kobayashi

Hiraga

Ueda

Zoledronic acid delays wound healing of tooth extraction socket, inhibits oral epithelial cell migration, and promotes proliferation and adhesion to hydroxyapatite of oral bacteria, without causing osteonecrosis of the jaw, in mice. J Bone Miner Metab 2010; 28: 165–175.

12.

Walter

Pabst

Ziebart

Bisphosphonates affect migration ability and cell viability of HUVEC, fibroblast and osteoblast in vitro. Oral Dis 2011; 17: 194–199.

13.

Ravosa

Ning

Liu

Bisphosphonate effects on the behaviour of oral epithelial cells and oral fibroblasts. Arch Oral Biol 2011; 56: 491–498.

14.

Walter

Klein

Pabst

Influence of bisphosphonates on endothelial cells, fibroblasts, and osteogenic cells. Clin Oral Investig 2009; 14: 35–41.

15.

Pabst

Ziebart

Koch

The influence of bisphosphonates on viability, migration, and apoptosis of human oral keratinocytes – in vitro study. Clin Oral Investig 2012; 16: 87–93.

16.

Simon

Niehoff

Kimmig

Expression profile and synthesis of different collagen types I, II, III, and V of human gingival fibroblasts, osteoblasts, and SaOS-2 cells after bisphosphonate treatment. Clin Oral Investig 2010; 14: 51–58.

17.

Açil

Möller

Niehoff

The cytotoxic effects of three different bisphosphonates in-vitro on human gingival fibroblasts, osteoblasts and osteogenic sarcoma cells. J Craniomaxillofac Surg 2012; 40: e229–E235.

18.

Cozin

Pinker

Solemani

Novel therapy to reverse the cellular effects of bisphosphonates on primary human oral fibroblasts. J Oral Maxillofac Surg 2011; 69: 2564–2578.

19.

Kim

Lee

Williams

Bisphosphonates induce senescence in normal human oral keratinocytes. J Dent Res 2011; 90: 810–816.

20.

Kaymaz

Dagdeviren

Asan

. Antigenic profile of human thymus in concurrence with “Clusters of Thymic Epithelial Staining” classification. Ann Anat-Anat Anz 2003; 185: 163–171.

21.

Hewitt

Lissina

Green

The bisphosphonate acute phase response: rapid and copious production of proinflammatory cytokines by peripheral blood gamma delta T cells in response to aminobisphosphonates is inhibited by statins. Clin Exp Immunol 2005; 139: 101–111.

22.

Correia

Caldeira

Marques

. Cytotoxicity evaluation of sodium alendronate on cultured human periodontal ligament fibroblasts. Dental Traumatol 2006; 22: 312–317.

23.

Agis

Blei

Watzek

Is zoledronate toxic to human periodontal fibroblasts?

J Dent Res 2010; 89: 40–45.

24.

Moreau

Guillet

Massin

Comparative effects of five bisphosphonates on apoptosis of macrophage cells in vitro. Biochem Pharmacol 2007; 73: 718–723.

25.

Mackie

Fisher

Zhou

Bisphosphonates regulate cell growth and gene expression in the UMR 106-01 clonal rat osteosarcoma cell line. Br J Cancer 2001; 84: 951–958.

26.

Allen

Burr

. The pathogenesis of bisphosphonate-related osteonecrosis of the jaw: so many hypotheses, so few data. J Oral Maxillofac Surg 2009; 67: 61–70.

27.

Popov

Villasmil

Bernardi

. Lethal toxin of Bacillus anthracis causes apoptosis of macrophages. Biochem Biophys Res Commun 2002; 293: 349–355.

28.

Suri

Mönkkönen

Taskinen

Nitrogen-containing bisphosphonates induce apoptosis of Caco-2 cells in vitro by inhibiting the mevalonate pathway: a model of bisphosphonate-induced gastrointestinal toxicity. Bone 2001; 29: 336–343.

29.

de Groen

Lubbe

Hirsch

Esophagitis associated with the use of alendronate. N Engl J Med 1996; 335: 1016–1021.

30.

Wallace

Dicay

MCKnight

N-bisphosphonates cause gastric epithelial injury independent of effects on the microcirculation. Aliment Pharmacol Ther 1999; 13: 1675–1682.

31.

Cremers

Pillai

Papapoulos

. Pharmacokinetics/pharmacodynamics of bisphosphonates: use for optimisation of intermittent therapy for osteoporosis. Clin Pharmacokinet 2005; 4: 551–570.

32.

Badros

Weikel

Salama

Osteonecrosis of the jaw in multiple myeloma patients: clinical features and risk factors. J Clin Oncol 2006; 24: 945–952.

33.

Ziebart

Koch

Klein

Geranylgeraniol—a new potential therapeutic approach to bisphosphonate associated osteonecrosis of the jaw. Oral Oncol 2011; 47: 195–201.

34.

Saito

Izumi

Shiomi

Zoledronic acid impairs re-epithelialization through down-regulation of integrin αvβ6 and transforming growth factor beta signalling in a three-dimensional in vitro wound healing model. Int J Oral Maxillofac Surg 2014; 43: 373–380.

Effects of alendronate and pamidronate on apoptosis and cell proliferation in cultured primary human gingival fibroblasts

Abstract

Keywords

Introduction

Methods

Primary fibroblast cell cultures

Determination of cytotoxic drug concentration

Treatments

Determination of proliferative index and apoptosis by immunohistochemical staining

Statistical analysis

Results

Cell viability and detection of IC50 values

Analysis of dose- and time-dependent changes for all groups

Apoptotic indices

Proliferation indices

Discussion

Footnotes

Conflict of interest

Funding

References

Cell viability and detection of IC₅₀ values