Clin Oral Invest DOI 10.1007/s00784-014-1219-0
ORIGINAL ARTICLE
In vitro effects of bisphosphonates on chemotaxis, phagocytosis, and oxidative burst of neutrophil granulocytes Nadine Hagelauer & Andreas Max Pabst & Thomas Ziebart & Holger Ulbrich & Christian Walter
Received: 10 May 2013 / Accepted: 25 February 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Objectives Bisphosphonate-associated osteonecrosis of the jaws is a serious side effect that mainly occurs in patients receiving highly potent, nitrogen-containing bisphosphonates. Usually the diagnosis is made due to exposed bone and a nonhealing wound. Neutrophil granulocytes are essential for sufficient wound healing; therefore, the influence of different bisphosphonates on neutrophil granulocytes was the focus of this study. Material and methods The effect of nitrogen-containing bisphosphonates (ibandronate, pamidronate, and zoledronate) and one non-nitrogen-containing bisphosphonate (clodronate) on chemotaxis, phagocytosis, and oxidative burst of neutrophil granulocytes in human whole blood was analyzed using standard cytometric flow assays. Results Chemotaxis of neutrophils was reduced by almost 50 % when cells were treated with ibandronate and zoledronate. All tested nitrogen-containing bisphosphonates moderately increased the percentage of phagocytizing neutrophils, whereas the percentage of oxidizing cells was extremely affected. Zoledronate increased the oxidative burst activity even at low concentrations. Treatment with ibandronate and pamidronate reached the same level, but only in at least 10 times the higher concentrations. The maximal burst activity of a single cell reached nearly 150 % compared to control. In this case, zoledronate also caused maximal effects even at low concentrations. Clodronate did not show any effects. N. Hagelauer : A. M. Pabst : T. Ziebart : C. Walter (*) Oral- and Maxillofacial Surgery, University Medical Center, Johannes Gutenberg-University, Augustusplatz 2, 55131 Mainz, Germany e-mail: [email protected] H. Ulbrich Institute of Pharmacy and Biochemistry, Johannes Gutenberg-University, Staudinger Weg 5, 55128 Mainz, Germany
Conclusion The results show a proinflammatory effect of the nitrogen-containing effect on neutrophil granulocytes which might contribute to the development of osteonecrosis. Clinical relevance The altered neutrophil defense might play a key role in the pathogenesis of bisphosphonate-associated osteonecrosis of the jaws, although the underlying causation between inflammatory reaction and the development of necrosis is yet unknown. Keywords Bisphosphonate . Bisphosphonate-associated osteonecrosis of the jaws . Neutrophil granulocyte . Immune defense
Introduction Several theories regarding the etiology of bisphosphonateassociated osteonecrosis of the jaws (BP-ONJ) are currently being discussed. The most commonly cited theory describes the influence of bisphosphonates on bone remodeling [1]. In addition to this, an antiangiogenic impact resulting in an avascular necrosis [2–4] and the influence of bone-covering soft tissues are described [3, 5–7]. Further risk factors might be a genetic predisposition [8, 9], smoking, obesity, further medication, and other diseases such as diabetes all of which are reasons that are a priori more often associated with the development of necrosis [10]. Nearly all BP-ONJ cases are described in patients receiving nitrogen-containing bisphosphonates. Among those, pamidronate and zoledronate are the most often involved bisphosphonates [11]. Interestingly, before most clinical diagnoses of BP-ONJ, a previous surgical procedure has preceded it with an accompanying wound that will not heal, resulting in BP-ONJ. Most of the procedures performed are tooth extractions [11, 12] due to periodontal disease in older patients [13]. Periodontal disease is closely related to microbiologically induced inflammation.
Clin Oral Invest
Microbiological contamination is seen in almost every BPONJ sample [14, 15]; therefore, inflammation might play a key role in BP-ONJ or in the progression of BP-ONJ. This is particularly apparent since the clinical extent of BP-ONJ correlates with the dimension of inflammatory cell infiltration [15], although the sequel of necrosis and inflammation in the pathogenesis of BP-ONJ has not been elucidated yet. The immune system has an important part both in wound healing after surgical procedures and in preventing and controlling infections. Neutrophil granulocytes are important cells in wound healing, especially in the early phase of hemostasis and inflammation [16]. In the event of infection, particularly for periodontal disease, neutrophil granulocytes play a key role [17]. The epithelial section that is in contact with the tooth surface, called junctional epithelium, has fewer desmosomes compared with other epithelium and, occasionally, has gap junctions, making this part very fragile to bacterial infections. Many leucocytes are present at this site: lymphocytes, macrophages, and neutrophil granulocytes [18]. Interestingly, these cells derive from a common stem cell that can also differentiate into osteoclasts [19], which are the main target of bisphosphonates. Macrophages and granulocytes exhibit phagocytotic activity, just like osteoclasts. Therefore, these cells might be of utmost importance in the etiology of BPONJ. The aim of the present study was to evaluate the impact of bisphosphonates on neutrophil functions in human whole blood. The capacity for chemotaxis and phagocytosis of neutrophil granulocytes and their ability to produce reactive oxygen intermediates (ROIs) were determined in vitro. The influence of the nitrogen-containing bisphosphonates (N-BPs) ibandronate, pamidronate, and zolendronate was compared with the non-nitrogen-containing bisphosphonate clodronate.
without chemotactic peptide as a negative control. After an incubation of 30 min in a 37 °C warm water bath, the migration of neutrophils was stopped by removing the inserts from the wells. The complete volume of each well was transferred on ice. After adding counting beads and a vital DNA dye, the samples were analyzed by flow cytometry in order to determine the percentage of migrated neutrophils. Phagocytosis The phagocytic activity of neutrophils was measured by determining the percentage of phagocytizing neutrophils using a standard cytometric flow assay (Phagotest Kit, Glycotope Biotechnology GmbH, Heidelberg, Germany). Heparinized whole blood samples, except the control, were incubated with 50 μM bisphosphonates for 8, 16, and 24 h in an atmosphere of 5 % CO2 at 37 °C. In order to test for concentration dependency, whole blood samples were incubated in a second approach for 16 h with 2.5, 25, and 100 μM with the different bisphosphonates in the atmosphere mentioned above. After bisphosphonate incubation, the phagocytosis assay was started. Each of the bisphosphonate-containing whole blood samples and the control were mixed with opsonized fluorescein isothiocyanate (FITC)-labeled Escherichia coli bacteria at 0 °C. The mixtures were incubated for 10 min at 37 °C in a water bath. As a negative control, whole blood without BP and FITC-labeled E. coli were incubated at 0 °C. By placing the samples on ice, the phagocytosis was stopped. The fluorescence of the attached bacteria on the cell surface was quenched by using Coomassie brilliant blue. After two washing steps, lysing solution was added to remove erythrocytes. Just previously, flow cytometric analysis DNA staining solution was added to exclude aggregation artifacts of E. coli bacteria or cells.
Materials and methods
Oxidative burst
Functional assays
The percentage of oxidizing neutrophils as well as the oxidative burst activity of each cell was analyzed using the cytometric flow assay Phagoburst (Phagoburst Kit, Glycotope Biotechnology, Heidelberg, Germany). Heparinized whole blood was incubated with different BP concentrations (0, 2.5, 5, 25, and 50 μM) for 16 h in an atmosphere of 5 % CO2 and 95 % air at 37 °C. Mixtures of heparinized whole blood and the chemotactic peptide fMLP as a low physiological stimulus and a negative control sample without any stimulus were incubated in a 37 °C warm water bath for 10 min. Following this, the incubation was continued for an additional 10 min with the substrate, dihydrorhodamine (DHR). After removing erythrocytes by using lysing solution, DNA staining solution was added to exclude cell aggregation. Next to the percentage of neutrophils converting nonfluorescent dihydrorhodamine to fluorescent rhodamine 123, the mean
Chemotaxis The chemotactic function of neutrophils was measured using a standard cytometric flow assay (Migratest, Glycotope Biotechnology GmbH, Heidelberg, Germany), a modified Boyden chamber assay. Before starting the assay, heparinized whole blood was incubated with 50 μM of different bisphosphonates for 16 h in an atmosphere of 5 % CO2 and 95 % air at 37 °C. After this period, leucocytes were isolated using leucocyte separation medium. Leucocyte-rich plasma of each sample was placed in two cell culture inserts with a pore size of 3.0 μm. One insert was transferred to a well containing the chemoattractant N-formyl-methionyl-leucyl-phenylalanine (fMLP). The other one was put in a buffer solution
Clin Oral Invest
fluorescence intensity (amount of rhodamine 123 per cell) was analyzed by a flow cytometer. Flow cytometric analysis of functional assays Phagotest, Phagoburst, and Migratest were analyzed with flow cytometry using the blue-green excitation light (488 nm argon ion laser, FACSCalibur, CELLQuest software: Becton Dickinson, Heidelberg, Germany). For Phagotest and Phagoburst, 10,000 leukocytes were collected from each sample; for Migratest 2000, counting beats were acquired. The granulocyte populations were gated by using their forward and side scatter dot plots. For Phagotest, the percentage of granulocytes containing FITC-labeled E. coli was analyzed. For Phagoburst, the percentage of cells having produced reactive oxygen metabolites and the enzymatic activity of each granulocyte (the amount of cleaved substrate per cell detected by mean channel number) were analyzed. For testing
BP influence on neutrophil chemotaxis, the percentage of spontaneously migrated granulocytes of each approach was subtracted from the percentage of granulocytes that were attracted by fMLP (an example of flow cytometric analysis of migration assay is shown in Fig. 1).
Statistical analysis Continuous variables are expressed as mean±SD (standard deviation). Comparisons between groups were performed via analysis of variance (ANOVA, post hoc test: Tukey) for experiments with more than two subgroups. The software SPSS 17.0 for Windows was used for calculations. A p value
ORIGINAL ARTICLE
In vitro effects of bisphosphonates on chemotaxis, phagocytosis, and oxidative burst of neutrophil granulocytes Nadine Hagelauer & Andreas Max Pabst & Thomas Ziebart & Holger Ulbrich & Christian Walter
Received: 10 May 2013 / Accepted: 25 February 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Objectives Bisphosphonate-associated osteonecrosis of the jaws is a serious side effect that mainly occurs in patients receiving highly potent, nitrogen-containing bisphosphonates. Usually the diagnosis is made due to exposed bone and a nonhealing wound. Neutrophil granulocytes are essential for sufficient wound healing; therefore, the influence of different bisphosphonates on neutrophil granulocytes was the focus of this study. Material and methods The effect of nitrogen-containing bisphosphonates (ibandronate, pamidronate, and zoledronate) and one non-nitrogen-containing bisphosphonate (clodronate) on chemotaxis, phagocytosis, and oxidative burst of neutrophil granulocytes in human whole blood was analyzed using standard cytometric flow assays. Results Chemotaxis of neutrophils was reduced by almost 50 % when cells were treated with ibandronate and zoledronate. All tested nitrogen-containing bisphosphonates moderately increased the percentage of phagocytizing neutrophils, whereas the percentage of oxidizing cells was extremely affected. Zoledronate increased the oxidative burst activity even at low concentrations. Treatment with ibandronate and pamidronate reached the same level, but only in at least 10 times the higher concentrations. The maximal burst activity of a single cell reached nearly 150 % compared to control. In this case, zoledronate also caused maximal effects even at low concentrations. Clodronate did not show any effects. N. Hagelauer : A. M. Pabst : T. Ziebart : C. Walter (*) Oral- and Maxillofacial Surgery, University Medical Center, Johannes Gutenberg-University, Augustusplatz 2, 55131 Mainz, Germany e-mail: [email protected] H. Ulbrich Institute of Pharmacy and Biochemistry, Johannes Gutenberg-University, Staudinger Weg 5, 55128 Mainz, Germany
Conclusion The results show a proinflammatory effect of the nitrogen-containing effect on neutrophil granulocytes which might contribute to the development of osteonecrosis. Clinical relevance The altered neutrophil defense might play a key role in the pathogenesis of bisphosphonate-associated osteonecrosis of the jaws, although the underlying causation between inflammatory reaction and the development of necrosis is yet unknown. Keywords Bisphosphonate . Bisphosphonate-associated osteonecrosis of the jaws . Neutrophil granulocyte . Immune defense
Introduction Several theories regarding the etiology of bisphosphonateassociated osteonecrosis of the jaws (BP-ONJ) are currently being discussed. The most commonly cited theory describes the influence of bisphosphonates on bone remodeling [1]. In addition to this, an antiangiogenic impact resulting in an avascular necrosis [2–4] and the influence of bone-covering soft tissues are described [3, 5–7]. Further risk factors might be a genetic predisposition [8, 9], smoking, obesity, further medication, and other diseases such as diabetes all of which are reasons that are a priori more often associated with the development of necrosis [10]. Nearly all BP-ONJ cases are described in patients receiving nitrogen-containing bisphosphonates. Among those, pamidronate and zoledronate are the most often involved bisphosphonates [11]. Interestingly, before most clinical diagnoses of BP-ONJ, a previous surgical procedure has preceded it with an accompanying wound that will not heal, resulting in BP-ONJ. Most of the procedures performed are tooth extractions [11, 12] due to periodontal disease in older patients [13]. Periodontal disease is closely related to microbiologically induced inflammation.
Clin Oral Invest
Microbiological contamination is seen in almost every BPONJ sample [14, 15]; therefore, inflammation might play a key role in BP-ONJ or in the progression of BP-ONJ. This is particularly apparent since the clinical extent of BP-ONJ correlates with the dimension of inflammatory cell infiltration [15], although the sequel of necrosis and inflammation in the pathogenesis of BP-ONJ has not been elucidated yet. The immune system has an important part both in wound healing after surgical procedures and in preventing and controlling infections. Neutrophil granulocytes are important cells in wound healing, especially in the early phase of hemostasis and inflammation [16]. In the event of infection, particularly for periodontal disease, neutrophil granulocytes play a key role [17]. The epithelial section that is in contact with the tooth surface, called junctional epithelium, has fewer desmosomes compared with other epithelium and, occasionally, has gap junctions, making this part very fragile to bacterial infections. Many leucocytes are present at this site: lymphocytes, macrophages, and neutrophil granulocytes [18]. Interestingly, these cells derive from a common stem cell that can also differentiate into osteoclasts [19], which are the main target of bisphosphonates. Macrophages and granulocytes exhibit phagocytotic activity, just like osteoclasts. Therefore, these cells might be of utmost importance in the etiology of BPONJ. The aim of the present study was to evaluate the impact of bisphosphonates on neutrophil functions in human whole blood. The capacity for chemotaxis and phagocytosis of neutrophil granulocytes and their ability to produce reactive oxygen intermediates (ROIs) were determined in vitro. The influence of the nitrogen-containing bisphosphonates (N-BPs) ibandronate, pamidronate, and zolendronate was compared with the non-nitrogen-containing bisphosphonate clodronate.
without chemotactic peptide as a negative control. After an incubation of 30 min in a 37 °C warm water bath, the migration of neutrophils was stopped by removing the inserts from the wells. The complete volume of each well was transferred on ice. After adding counting beads and a vital DNA dye, the samples were analyzed by flow cytometry in order to determine the percentage of migrated neutrophils. Phagocytosis The phagocytic activity of neutrophils was measured by determining the percentage of phagocytizing neutrophils using a standard cytometric flow assay (Phagotest Kit, Glycotope Biotechnology GmbH, Heidelberg, Germany). Heparinized whole blood samples, except the control, were incubated with 50 μM bisphosphonates for 8, 16, and 24 h in an atmosphere of 5 % CO2 at 37 °C. In order to test for concentration dependency, whole blood samples were incubated in a second approach for 16 h with 2.5, 25, and 100 μM with the different bisphosphonates in the atmosphere mentioned above. After bisphosphonate incubation, the phagocytosis assay was started. Each of the bisphosphonate-containing whole blood samples and the control were mixed with opsonized fluorescein isothiocyanate (FITC)-labeled Escherichia coli bacteria at 0 °C. The mixtures were incubated for 10 min at 37 °C in a water bath. As a negative control, whole blood without BP and FITC-labeled E. coli were incubated at 0 °C. By placing the samples on ice, the phagocytosis was stopped. The fluorescence of the attached bacteria on the cell surface was quenched by using Coomassie brilliant blue. After two washing steps, lysing solution was added to remove erythrocytes. Just previously, flow cytometric analysis DNA staining solution was added to exclude aggregation artifacts of E. coli bacteria or cells.
Materials and methods
Oxidative burst
Functional assays
The percentage of oxidizing neutrophils as well as the oxidative burst activity of each cell was analyzed using the cytometric flow assay Phagoburst (Phagoburst Kit, Glycotope Biotechnology, Heidelberg, Germany). Heparinized whole blood was incubated with different BP concentrations (0, 2.5, 5, 25, and 50 μM) for 16 h in an atmosphere of 5 % CO2 and 95 % air at 37 °C. Mixtures of heparinized whole blood and the chemotactic peptide fMLP as a low physiological stimulus and a negative control sample without any stimulus were incubated in a 37 °C warm water bath for 10 min. Following this, the incubation was continued for an additional 10 min with the substrate, dihydrorhodamine (DHR). After removing erythrocytes by using lysing solution, DNA staining solution was added to exclude cell aggregation. Next to the percentage of neutrophils converting nonfluorescent dihydrorhodamine to fluorescent rhodamine 123, the mean
Chemotaxis The chemotactic function of neutrophils was measured using a standard cytometric flow assay (Migratest, Glycotope Biotechnology GmbH, Heidelberg, Germany), a modified Boyden chamber assay. Before starting the assay, heparinized whole blood was incubated with 50 μM of different bisphosphonates for 16 h in an atmosphere of 5 % CO2 and 95 % air at 37 °C. After this period, leucocytes were isolated using leucocyte separation medium. Leucocyte-rich plasma of each sample was placed in two cell culture inserts with a pore size of 3.0 μm. One insert was transferred to a well containing the chemoattractant N-formyl-methionyl-leucyl-phenylalanine (fMLP). The other one was put in a buffer solution
Clin Oral Invest
fluorescence intensity (amount of rhodamine 123 per cell) was analyzed by a flow cytometer. Flow cytometric analysis of functional assays Phagotest, Phagoburst, and Migratest were analyzed with flow cytometry using the blue-green excitation light (488 nm argon ion laser, FACSCalibur, CELLQuest software: Becton Dickinson, Heidelberg, Germany). For Phagotest and Phagoburst, 10,000 leukocytes were collected from each sample; for Migratest 2000, counting beats were acquired. The granulocyte populations were gated by using their forward and side scatter dot plots. For Phagotest, the percentage of granulocytes containing FITC-labeled E. coli was analyzed. For Phagoburst, the percentage of cells having produced reactive oxygen metabolites and the enzymatic activity of each granulocyte (the amount of cleaved substrate per cell detected by mean channel number) were analyzed. For testing
BP influence on neutrophil chemotaxis, the percentage of spontaneously migrated granulocytes of each approach was subtracted from the percentage of granulocytes that were attracted by fMLP (an example of flow cytometric analysis of migration assay is shown in Fig. 1).
Statistical analysis Continuous variables are expressed as mean±SD (standard deviation). Comparisons between groups were performed via analysis of variance (ANOVA, post hoc test: Tukey) for experiments with more than two subgroups. The software SPSS 17.0 for Windows was used for calculations. A p value
