IL-1β Inhibits Human Osteoblast Migration
- 6 Downloads
Bone has a high capacity for self-renewal and repair. Prolonged local secretion of interleukin 1β (IL-1β), however, is known to be associated with severe bone loss and delayed fracture healing. Since induction of bone resorption by IL-1β may not sufficiently explain these pathologic processes, we investigated, in vitro, if and how IL-1β affects migration of multipotent mesenchymal stromal cells (MSC) or osteoblasts. We found that homogenous exposure to IL-1 β significantly diminished both nondirectional migration and site-directed migration toward the chemotactic factors platelet-derived growth factor (PDGF)-BB and insulinlike growth factor 1 (IGF-1) in osteoblasts. Exposure to a concentration gradient of IL-1β induced an even stronger inhibition of migration and completely abolished the migratory response of osteoblasts toward PDGF-BB, IGF-1, vascular endothelial growth factor A (VEGF-A) and the complement factor C5a. IL-1β induced extracellular signal-regulated kinases 1 and 2 (ERK1/2) and c-Jun N-terminal kinases (JNK) activation and inhibition of these signaling pathways suggested an involvement in the IL-1β effects on osteoblast migration. In contrast, basal migration of MSC and their migratory activity toward PDGF-BB was found to be unaffected by IL-1β. These results indicate that the presence of IL-1β leads to impaired recruitment of osteoblasts which might influence early stages of fracture healing and could have pathological relevance for bone remodeling in inflammatory bone disease.
Bone is a highly metabolic tissue with a high rate of self-renewal by bone remodeling and the potency for repair without forming scar tissue (1). Formation of bone during embryonic development or after fracture can occur via intramembranous or endochondral ossification. In case of fracture healing, bone formation processes are initiated in the fracture hematoma during the early inflammatory phase in which inflammatory cytokines and various growth factors such as growth factor β superfamily members and angiogenic factors are released (2,3). The concerted action of these factors regulates the recruitment and proliferation of multipotent mesenchymal stromal cells (MSC) and osteoblasts, the differentiation of MSC into osteoblasts and chondrocytes as well as angiogenesis to reestablish a sufficient blood supply (2,3). After the formation of primary bone, remodeling leads to the formation of secondary bone, which depicts the original anatomical and functional properties before the fracture (2, 3, 4).
Since recruitment of MSC and osteoblasts is indispensable for fracture healing and life-long bone remodeling, several studies have focused on growth factors that are present in the fracture hematoma or deposited in bone and released upon bone resorption through osteoclasts as being chemotactic agents for osteoblasts and MSC. The platelet-derived growth factor β (PDGF-BB) was found to be a very potent chemotactic factor for MSC, primary osteoblasts and osteoblastic cell lines (5, 6, 7, 8, 9, 10). Further examples are the insulinlike growth factors 1 (IGF-1) and 2 (IGF-2), which stimulate the site-directed migration of MSC and osteoblasts (5,11,12). In addition to its known angiogenic effects, vascular endothelial growth factor A (VEGF-A) was found to have a dose-dependent chemotactic effect on MSC and primary human osteoblasts, in vitro, that is mediated through VEGF receptor 1 activation (13,14).
In recent years, it has become clear that, in addition to growth factors, cells of the immune system, as well as cytokines and the complement system, have a profound impact on bone formation, bone remodeling and skeletal disorders (15,16). Mainly, excessive bone loss through osteoclasts observed in inflammatory and autoimmune diseases as well as the discovery of T cells and cytokines inducing osteoclastogenesis led to a rapidly emerging field called osteoimmunology (17). The complement factor C5a, locally released at sites of tissue damage and inflammation, has been reported to have a chemotactic effect on MSC (18). In contrast, MSC were recently found to express rather low levels of C5a receptor (C5aR) mRNA and to migrate toward C5a significantly less than osteoblasts, which show a higher C5aR mRNA expression (19).
A prominent cytokine released in the bone under various circumstances is IL-1β. In the fracture hematoma, IL-1β is released by macrophages together with tumor necrosis factor-α (TNF-α) and IL-6 to promote the recruitment of other inflammatory cells and initiate the repair process (20). IL-1β as well as IL-6 and TNF-α expression levels in the fracture hematoma were found to peak within the first 24-h after fracture and to decline rapidly afterward (20).
While short periods of IL-1β secretion are critical for successful fracture healing (20), persistent release of IL-1β can be associated with severe bone loss. In idiopathic osteoporosis patients, IL-1β release by peripheral blood monocytes, precursors of osteoclasts, was shown to be increased significantly when compared with healthy individuals (21,22). The notion of IL-1-mediated bone loss also is supported by a study where inhibition of IL-1 with soluble IL-1R was shown to significantly reduce inflammation, loss of connective tissue and bone resorption in the periodontium of Macaca fascicularis primates infected with pathogens (23). A different study showed that IL1−/−hTNFtg mice were completely protected from systemic bone loss mediated through the overexpression of TNF-α (24). A possible pathogenetic role for IL-1β in bone loss is further supported by the fact that IL-1β enables peripheral blood mononuclear cells to differentiate into osteoclasts in the presence of osteoblastic SaOS-2 cells without further stimulation (25). In murine osteoblasts, IL-1β induced receptor activator of NF-κB lig-and (RANKL) secretion while reducing osteoprotegerin (OPG) production and activated osteoclastogenesis (26). A subsequent study even found IL-1β to be essential for osteoclast formation as IL-1 receptor antagonist (IL-1RA) treatment inhibited RANKL/TNF-α-induced osteoclastogenesis (27).
Taken together, IL-1β has been shown to influence fracture repair and bone remodeling. The best studied function of IL-1β in bone metabolism is the activation of osteoclasts. Little, however, is known about the effects of IL-1β on migration of human osteoblast and MSC that are important for bone remodeling and repair. Therefore, we investigated in a human in vitro model if IL-1β has a chemotactic effect on osteoblasts and MSC or influences the chemotactic effect of other soluble factors active for osteoblast and MSC recruitment.
Materials and Methods
Primary Osteoblast and MSC Isolation and Culture
Human osteoblasts were harvested from cancellous bone obtained during routine surgical procedures (total knee replacements) with informed consent of the patients and according to the requirements of the Ethics Committee at the University of Ulm. As described earlier (13), 5 g cancellous bone were broken up in small pieces, washed with phosphatebuffered saline (PBS) and incubated in Dulbecco’s modified Eagle medium (DMEM) with 0.05% collagenase (Sigma-Aldrich, Schnelldorf, Germany) for 2 h at 37°C. After removing the supernatant and washing twice with PBS, bone pieces were cultured in 6-well plates with Ham’s F-12 medium supplemented with 10% fetal calf serum (FCS), 100 U/100 µg/mL penicillin/streptomycin, 2.5 µg/mL amphotericin B and 2 mmol/L L-glutamine (all Biochrom, Berlin, Germany) at 37°C, 95% humidity and 5% CO2. Medium was changed twice a week. After one week, osteoblasts started to grow out, which were then transferred into culture flasks and passaged at maximum 3 to 4 times in DMEM supplemented with 10% FCS, 100 U/100 µg/mL penicillin/streptomycin and 2 mmol/L L-glutamine (all Biochrom) before application in chemotaxis, adhesion and MTT assays. The cell culture technique preserves the osteoblastic phenotype as described previously (28).
Human MSC were harvested from bone marrow obtained during routine surgical procedures (triple osteotomy) with informed consent of the patients and according to the requirements of the Ethics Committee at the University of Ulm. As described earlier (6), mononuclear cells were harvested from the interphase after a density gradient centrifugation (600g) with Biocoll (Biochrom), washed with PBS and resuspended in DMEM supplemented with 10% FCS, 100 U/100 µg/mL penicillin/streptomycin and 2 mmol/L L-glutamine (all Biochrom). Mononuclear cells were given into 75-cm2 cell culture flasks and incubated at 37°C, 95% humidity and 5% CO2. Nonadherent cells were washed off 24 h later and 10 ng/mL fibroblast growth factor-2 was added to the medium. Medium was changed twice a week. MSC were split at approximately 60% confluency and used in passage 3–6.
Soluble Factors and Inhibitors Employed in Chemotaxis and Adhesion Assays
For chemotaxis and adhesion assays, the following factors were used. Recombinant human PDGF-BB, VEGF-A, IGF-1 and IL-1β were purchased from Peprotech, Hamburg, Germany. Recombinant human C5a was purchased from Sigma-Aldrich. For ERK1/2 and JNK inhibition, PD 98059 and JNK inhibitor II (both Callbiochem, Darmstadt, Germany) were used, respectively.
Chemotaxis and Adhesion Assay
In vitro adhesion and chemotaxis assays were performed in a modified Boyden chamber (NeuroProbe, Gaithersburg, MD, USA) using polycarbonate filters with 8-µm pores between the lower well containing the chemotactic factor and the upper well containing osteoblasts. To evaluate the influence of IL-1β on the adhesion and chemotaxis of osteoblasts, IL-1β was either added only into the lower chamber to establish an IL-1β gradient or to both upper and lower chamber to create a uniform IL-1β concentration. Growth factors and cytokines were diluted in serum-free DMEM at concentrations that were found effective in previous studies (6,11,13,19): 10, 100 or 1000 pg/mL IL-1β, 100 ng/mL C5a, 10ng/mL PDGF-BB, 100 ng/mL IGF-1 and 100 ng/mL VEGF-A and filled in quadruplicates into the lower wells of the Boyden chamber, which was then covered by the polycarbonate filter. For negative control, DMEM alone was added to the lower well. Osteoblasts of passage 3–4 were trypsinized, washed and resuspended in serum-free DMEM. For the adhesion assay, 5 × 102 osteoblasts in 50 µL serum-free DMEM were given into the upper well of the Boyden chamber and incubated for 30, 60 and 90 min. In case of uniform IL-1β concentrations in upper and lower well of the Boyden chamber, IL-1β also was added to the cell suspension at 100 pg/mL. Adherent cells on the upper side of the polycarbonate filter were fixed, stained with Giemsa (Merck, Darmstadt, Germany) and counted.
For chemotaxis assay, 104 osteoblasts in 50 µL serum-free DMEM were given into the upper wells of the Boyden chamber. In case of uniform IL-1β concentrations in upper and lower wells of the Boyden chamber, IL-1β was added to the cell suspension at the appropriate concentration. For inhibition of chemotaxis with mitogen-activated protein kinase (MAPK) inhibitors, 10 µmol/L PD 98059 (ERK1/2 inhibition) or JNK inhibitor II (JNK inhibition) were given in both upper and lower wells of the Boyden chamber. After 4 h of incubation, the filter was obtained and adherent, nonmigrated cells on the upper side were scraped off with PBS and a rubber scraper. The migrated cells on the lower side were stained with Giemsa staining after fixation in a 4% formaldehyde solution. All migrated cells were counted at 20× magnification.
Immunocytochemistry of MSC and Osteoblasts
To detect IL-1 receptor type I (IL-1R1) expression in cultured MSC and osteoblasts, cells were seeded at 104 cells per well on 4-well glass slides with polystyrene vessels (BD Falcon) in complete medium and left overnight to adhere. Serum-free medium was added 30 min before staining. For detection of IL-1R1 expression in migrating MSC and osteoblasts, chemotaxis toward PDGF-BB or without chemoattractant was performed as described above. After migration, nonmigrated cells on the upper side were scraped off with PBS and a rubber scraper. Both the culture slides and the chemotaxis filter were fixed in a 4% formaldehyde solution and cells were stained with 2 µg/mL anti-IL-1R1 antibody (AF269; R&D Systems, Wiesbaden, Germany) and the Dako LSAB + System-HRP (K0690; Dako, Hamburg, Germany) according to the manufacturer’s protocol. Negative controls were stained without exposure to the primary antibody (anti-IL-1R1).
Human Phospho-MAPK Array of IL-1β-Stimulated Osteoblasts
To determine the phosphorylation profile of MAPK in IL-1β-stimulated osteoblasts, we performed a Human Phospho-MAPK Array from R&D Systems. Osteoblasts were seeded at 1.5 × 106 cells per 10-cm cell culture dish and incubated in complete medium overnight to adhere. After serum-free medium was added, the control was left untreated while 100 pg/mL IL-1β was added to the other dish. After 30 min of incubation, cells were lysed and the protein content of both the control and the IL-1β treated sample was measured with the NanoDrop ND 1000 (Peqlab, Erlangen, Germany) at 280 nm to be of equal quantity. The array procedure was performed according to the manufacturer’s protocol and spots were detected with the LAS-4000 (GE Healthcare, Munich, Germany).
All data are presented as the mean of independent donors. Error bars represent standard error of the mean (SEM). Chemotaxis results were analyzed for their significance with two-way analysis of variance (ANOVA) followed by a Bonferroni post hoc test using GraphPad Prism 5 (GraphPad Software).
Effect of IL-1β on Cell Migration of Osteoblasts and MSC
Expression of IL-1R1 in MSC and Osteoblasts
Phosphorylation Profile of MAPK in IL-1β Treated Osteoblasts
Inhibition of Migration Specific Pathways
We could not further evaluate if one of the p38 MAPK actually plays a role in the IL-1β mediated inhibition of migration, since the p38 inhibitor SB203580 completely abolished any migration activity of osteoblasts (data not shown) which previously has been observed for several other cell types as well (29). Taking together the phosphorylation profile of MAPK upon IL-1β stimulation and our inhibitor experiments, we conclude that IL-1β reduces osteoblast migration in an ERK1/2- and JNK-dependent manner.
Controlled release of IL-1β is necessary for successful fracture repair (3,20). If IL-1β levels are increased permanently, however, severe bone loss can occur, since IL-1β has been shown to stimulate osteoclastogenesis and bone resorption by inducing RANKL secretion and suppressing OPG secretion (25, 26, 27). While activation of osteoclasts is a well-studied function of IL-1β, few studies are known concerning the inhibition of bone forming cells such as MSC and osteoblasts through IL-1β. We could show, for the first time, that IL-1β completely abolishes osteoblast migration toward several chemotactic factors if presented in a concentration gradient, while MSC migration was not inhibited through IL-1β. To the best of our knowledge, this is the only situation identified so far in which a concentration gradient of a soluble factor inhibits the directed migration of osteoblasts in vitro. We could exclude that the overall process of cell adhesion was compromised under homogenous or gradient exposure to IL-1β. Nevertheless, effects on site-dependent involvement of adhesion receptors, proteases and cytoskeletal dynamics clearly deserve further investigation.
Osteoblasts used in this study are able to migrate nondirectionally and in the direction of a chemotactic gradient as shown previously (6,11,13,19). When exposed to IL-1β, however, osteoblasts’ basal migration and migration toward several chemotactic factors was significantly decreased. A study by Gilardetti et al. indicated that preincubation of osteoblasts with IL-1β significantly reduces migration toward PDGF-AA by specifically downregulating the binding capacity of osteoblasts for PDGF-AA, but has few effects on the binding of PDGF-BB and migration toward PDGF-BB (30). In another study, IL-1β preincubation was even able to enhance the migratory response of murine osteoblast-like cells toward PDGF-AA (31). The present study shows that, without preincubation, an IL-1β gradient strongly inhibits basal migration of human osteoblasts and completely abolishes the chemotactic function of several growth factors such as PDGF-BB, IGF-1 and VEGF-A, as well as the inflammatory factor C5a in vitro. Therefore, we suggest that an IL-1β gradient inhibits more general mechanisms of cell migration in osteoblasts such as expression of migration-relevant adhesion proteins or matrixmetalloproteases. Focal adhesion formation or cytoskeletal rearrangements could also be affected. The signaling mechanism by which IL-1β inhibits migration in osteoblasts, we suggest, depends on ERK1/2 and JNK activation since the IL-1β-induced inhibition of migration could be rescued through ERK1/2 and partially also through JNK inhibitors. Both JNK and ERK1/2 are, among others, known to be central kinases of the IL-1R1 signaling transduction (32) and we found them to be highly phosphorylated upon IL-1β stimulation.
While IL-1β inhibited osteoblast migration, both nondirectional and even more potent in the direction of a chemotactic gradient, we observed that migration of MSC stayed unaffected by IL-1β stimulation in vitro. This is supported by a study by Ponte et al. in which preincubation of MSC with a ten-fold higher IL-1β concentration did not impact basal MSC migration and migration toward FCS (33). Since MSC used in this study migrated normally in the absence and presence of a PDGF-BB gradient compared with previous studies (6), full migratory function of MSC under exposure to IL-1β can be presumed. Moreover, the positive staining for IL-1R1 indicates that the nonreactive behavior of MSC upon IL-1β exposure in terms of migration does not depend on a lack of IL-1R1 expression.
Inflammatory cytokines including IL-1β, TNF-α and IL-6 are expressed especially during the very early phase after the fracture by immune cells in the fracture hematoma (3,20). A preferential recruitment of MSC instead of osteoblasts in the presence of IL-1β being observed in our study might be important in this early fracture healing stage. In a sheep bone-healing model, for example, more flexible stabilization was associated with a higher inflammatory response, and delayed healing (34).
The importance of IL-1β for bone loss during inflammatory diseases has been shown in various studies as explained before (21, 22, 23). The most prominent function of IL-1β in bone loss is the capacity to activate osteoclastogenesis (25, 26, 27). We could show in this study that, in addition to promoting bone resorption, IL-1β also seems to influence bone formation processes as it inhibits osteoblast recruitment in the sense of a soluble repellent factor. The IL-1β-concentrations used were within the range of those found in synovial fluid of patients with rheumatoid arthritis and culture supernatant of explanted murine calvaria after implantation of polyethylene or titanium particles indicating physiological relevance (35). Since IL-1β levels in inflammatory bone diseases are elevated over longer periods of time, the balance of bone metabolism might be particularly disturbed by simultaneously elevated bone resorption through osteoclast activation and impaired osteoblast migration. The combination of both mechanisms may be the central pathogenetic factor for development of osteolytic lesions observed in locally persistent inflammatory situations such as osteomyelitis, subchondral bone lesions in rheumatoid arthritis or periodontal disease.
In this study we show, in vitro, that chemotactic activity of osteoblasts is reduced in the presence of IL-1β, while MSC stay unaffected. This mechanism may interfere with osteoblast recruitment through osteoclast-mediated release of bone-derived growth factors leading to disturbed coupling of bone formation and bone resorption. The IL-1β-induced inhibition of osteoblast migration was most pronounced in the presence of a concentration gradient indicating that this cytokine acts functionally as a repellent factor. Future studies will have to investigate if more cytokines act as soluble repellent factors on cells of the osteoblast lineage, in vitro and in vivo, to understand bone remodeling and fracture healing as well as their dysregulation by inflammatory processes in more detail.
The authors declare they have no competing interests as defined by Molecular Medicine, or other interests that might be perceived to influence the results and discussion reported in this paper.
The excellent technical assistance of Giovanni Ravalli is gratefully acknowledged. The study was funded in part by the German Research Council (Deutsche Forschungsgemeinschaft; KFO 200).
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, and provide a link to the Creative Commons license. You do not have permission under this license to share adapted material derived from this article or parts of it.
The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
To view a copy of this license, visit (https://doi.org/creativecommons.org/licenses/by-nc-nd/4.0/)