University at Buffalo - The State University of New York
Skip to Content
Gene Therapy - Anti-inflammatory effect of MAPK phosphatase-1 local gene transfer in inflammatory bone loss

Original Article

Gene Therapy (2011) 18, 344–353; doi:10.1038/gt.2010.139; published online 11 November 2010

Anti-inflammatory effect of MAPK phosphatase-1 local gene transfer in inflammatory bone loss

H Yu1,2, Q Li1, B Herbert1,4, R Zinna2, K Martin2, C R Junior1,2 and K L Kirkwood1,2,3

  1. 1Department of Craniofacial Biology, Medical University of South Carolina, Charleston, SC, USA
  2. 2Center for Oral Health Research, Medical University of South Carolina, Charleston, SC, USA
  3. 3Department of Microbiology and Immunology, Medical University of South Carolina, Charleston, SC, USA
  4. 4Department of Biology, College of Charleston, Charleston, SC, USA

Correspondence: Dr KL Kirkwood, Department of Craniofacial Biology, Medical University of South Carolina, DDS, PhD, 173 Ashley Avenue BSB 449, Charleston, SC 29425, USA. E-mail: klkirk@musc.edu

Received 21 May 2010; Accepted 7 September 2010; Published online 11 November 2010.

Top

Abstract

Alveolar bone loss associated with periodontal diseases is the result of osteoclastogenesis induced by bacterial pathogens. The mitogen-activated protein kinase (MAPK) phosphatase 1 (MKP-1) is a critical negative regulator of immune response as a key phosphatase capable of dephosphorylating activated MAPKs. In this study, rat macrophages transduced with recombinant adenovirus (Ad.)MKP-1 specifically dephosphorylated activated MAPKs induced by lipopolysaccharide (LPS) compared with control cells. Bone marrow macrophages from MKP-1 knockout (KO) mice exhibited higher interleukin (IL)-6, IL-10, tumor necrosis factor (TNF)-α, and select chemokine compared with wild-type (WT) mice when stimulated by LPS. In addition, bone marrow cultures from MKP-1 KO mice exhibited significantly more osteoclastogenesis induced by LPS than when compared with WT mice. Importantly, MKP-1 gene transfer in bone marrow cells of MKP-1 KO mice significantly decreased IL-6, IL-10, TNF-α and chemokine levels, and formed fewer osteoclasts induced by LPS than compared with control group of cells. Furthermore, MKP-1 gene transfer in an experimental periodontal disease model attenuated bone resorption induced by LPS. Histological analysis confirmed that periodontal tissues transduced with Ad. MKP-1 exhibited less infiltrated inflammatory cells, less osteoclasts and less IL-6 than compared with rats of control groups. These studies indicate that MKP-1 is a key therapeutic target to control of inflammation-induced bone loss.

Keywords:

MKP-1; cytokine; osteoclastogenesis; periodontal disease

Top

Introduction

Periodontitis is a chronic inflammatory disease that is characterized by loss of the periodontal ligament and alveolar bone, and is a major cause of tooth loss. The bone loss associated with periodontal diseases is the most prevalent disease of the bone in human, ranging from 10 to 15% of adult.1, 2 It has long been appreciated that periodontal bone loss results from inflammation-driven process associated with enhanced osteoclastogenesis.3 Osteoclasts, derived from hematopoietic monocyte/macrophage cell lineage,4 undergo cellular differentiation and maturation in response to many cytokines and hormones.5, 6, 7 Bacterial infection initiates innate immune responses that generate proinflammatory mediators, such as interleukin (IL)-1,8, 9 IL-6,10 tumor necrosis factor (TNF)-α,9 and prostaglandin E2,9 which are capable of stimulating osteoclastogenesis directly or indirectly leading to bone resorption.

The production of inflammatory cytokines is the consequence of activation key intracellular pathways, including mitogen-activated protein kinases (MAPKs) pathways11, 12 following activation of toll-like receptor complexes that recognize bacterial constituents. MAPKs consist of three major families, including extracellular signal-regulated kinases (ERKs), c-Jun N-terminal kinases (JNKs) and p38 MAPK. MAPK pathways relay, amplify and integrate signals, which modulate a series of physiological response, including cellular proliferation, differentiation, development, inflammatory responses and apoptosis.13

Although the activation of MAPK pathways is critical to initiate an innate immune response to help mount a protective immune response to eliminate invading pathogens, sustained production of proinflammatory cytokines can result in bone resorption. A number of studies have demonstrated that lipopolysaccharide (LPS) from Gram-negative bacteria were capable of inducing bone resorption in vivo.14, 15, 16, 17, 18 Therefore, modulating MAPK immune response to an appropriate level is essential to attenuate bone resorption associated with bacterial infection.

In mammalian cells, MAPKs are activated by phospohorylation on critical tyrosine, serine or threonine residues. Inactivation of activated MAPKs occurs by dephosphorylation by MAPK phosphatase (MKP), a group of dual-specificity protein phosphatases.19 To date, at least 11 MKP family members have been identified in mammalian cells, with MKP-1 being the founding member. Studies have shown that MKP-1 knockout (KO) mice are susceptible to endotoxic shock; exhibited a marked increase in production of proinflammamtory cytokine TNF-α, IL-6 and an anti-inflammatory cytokine IL-10 as compared with wild-type (WT) animals.20 MKP-1 KO mice also had a marked increase in both the incidence and severity of experimentally induced autoimmune arthritis,21, 22 and exhibited more bone loss after LPS challenge as compared with WT animals.23 Using an orally active p38 MAPK inhibitor, we have shown p38 signaling is critical to LPS-induced inflammatory response, osteoclastogenesis and alveolar bone loss.24 These results highlight the significance of the p38/MKP-1 axis of regulation on innate immune response to bacterial infections and maintaining bone homeostasis, suggesting that restriction of activated MAPKs is a potential therapeutic strategy for diseases associated with exaggerated MAPK responses.

In this study, we evaluated the therapeutic effects of local expression of MKP-1 in regulation of immune response and bone homeostasis following LPS challenge. We demonstrated that local adenoviral-delivered MKP-1 restrained the activation of MAPKs, reduced proinflammatory/anti-inflammatory cytokine responses, reduced osteoclastogenesis and in vivo bone loss associated with experimental periodontitis. Data obtained from these studies suggest MKP-1 is a potential therapeutic target for diseases associated with increased MAPK activation.

Top

Results

MKP-1 specifically dephosphorylates MAPKs after LPS stimulation

MKP-1 dephosphorylates critical tyrosine and serine/threonine residues of activated MAPKs, including p38 MAPK, ERK and JNK.25 First, to test the efficacy of MKP-1 expression by adenovirus (Ad.) MKP-1 transduction, rat macrophages (NR8383 cells) were transduced with Ad. MKP-1 or control vector, Ad. LacZ, at multiplicity of infection (m.o.i.) from 100 to 900 for 24–72h. Optimal MKP-1 protein expression was achieved at m.o.i. 300 for 48h (data not shown). The LacZ gene expression was confirmed by β-galactosidase staining in cells (data not shown). Following these preliminary experiments, rat macrophages were transduced either with Ad.MKP-1 or Ad.LacZ (m.o.i. 300), or treated with equal volume of HEPES-buffered saline. At 48h after the Ad. transduction, the cells were either unstimulated or stimulated with 1μgml−1 of LPS for various time periods (10, 30, 60, 120, 180 and 300min). The levels of MKP-1 protein expression in transduced cells were evaluated by western blot protein assay. As shown in Figure 1a, rat macrophages transduced with Ad.MKP-1 exhibited only mild level of MKP-1 protein expression (arrow shown, 39kDa) before LPS stimulation. However, the level of MKP-1 protein expression significantly increased at 60–300min following LPS stimulation peaking from 60 to 120min. In contrast, cells transduced with Ad.LacZ or treated with HEPES-buffered saline had undetectable level of MKP-1 protein expression at basal level. When incubated with a higher concentration MKP-1 antibody, cells transduced with Ad. LacZ or treated with HEPES-buffered saline only displayed a slight increase of MKP-1 expression at 60–300min (arrow shown, peaking from 60 to 120min) following LPS stimulation. The protein bands under the 39kDa MKP-1 protein are nonspecific protein bands due to incubation with a higher concentration of a polyclonal antibody.

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

MKP-1 gene transfer specifically dephosphorylated MAPKs in rat macrophages after LPS stimulation. Rat macrophages (NR8383 cells) were transduced with Ad.MKP-1 (m.o.i. 300), or Ad.LacZ (m.o.i. 300) or treated with equal volume of HEPES-buffered saline for 48h. Cells were either unstimulated or stimulated with 1μgml−1 of LPS (from A. actinomycetemcomitans) for various time periods (10, 30, 60, 120, 180 and 300min). The MKP-1, p-JNK (p54, p46), p-p38 MAPK, p-ERK (p44, p42), p-38 MAPK and p-NF-κB (p65, p105) protein expression were examined by western blot assays. The p-38 MAPK served as a loading control. (a) MKP-1 protein expression in rat macrophages. (b) Phosphorylated MAPKs protein expression in rat macrophages. (c) Phosphorylated NF-κB p65 and p105 protein expression in rat macrophages. The data is a representative from three separate experiments.

Full figure and legend (214K)

To investigate the effects of overexpression of MKP-1 on activated MAPKs, and nuclear factor (NF)-κB activity, phosphorylated (p)-MAPKs and p-NF-κB p65/p105 were analyzed by western blot protein assay. As shown in Figure 1b, there was significant induction of p-JNK (p54, p46), p-p38 and p-ERK (p44, p42) proteins at 60–300min after LPS stimulation in cells either treated with HEPES-buffered saline, or transduced with Ad.LacZ. Whereas in macrophages transduced with Ad.MKP-1, p-JNK (p54, p46) and p-p38 protein expression remained close to the basal levels following LPS stimulation. Although there was an increase in level of p-ERK (p44, p42) expression in cells transduced with Ad.MKP-1 at 30min after LPS stimulation, p-ERK (p44, p42) expression maintained at relatively lower level at 60–300min after LPS stimulation. This decrease of p-ERK (p44, p42) 60–300min after LPS stimulation correlated with the increase in levels of MKP-1 protein in cells transduced with Ad.MKP-1 at 60min following LPS stimulation. In contrast to p-MAPKs, overexpression of MKP-1 by Ad.MKP-1 failed to dephosphorylate NF-κB p65 and p105 protein after LPS stimulation as demonstrated in Figure 1c. These data demonstrated that overexpression of MKP-1 specifically inhibited LPS-stimulated MAPKs activities and this effect last at least for 5h following LPS stimulation.

MKP-1 gene transfer rescued the innate immune response in MKP-1 KO mice

Previous study has demonstrated that MKP-1 KO mice had compromised immune response, which exhibited enhanced production of IL-6, TNF-α, IL-10 after LPS challenge compared with MKP-1 WT mice.21 Studies in our lab also revealed that MKP-1 KO mice exhibited more alveolar bone loss after LPS challenge compared with MKP-1 WT mice.23 Therefore, a rescue experiment was conducted to determine the therapeutic effects of gene transfer of MKP-1 in modulating immune response and potentially restoring bone homeostasis after LPS challenge in MKP-1 deficient mice.

As bone marrow macrophages (BMMφ) are the primary cells that both generate inflammatory cytokines in an autocrine manner, differentiate into osteoclasts, we first analyzed the cytokine profile to verify whether MKP-1 gene transfer in MKP-1 KO mice could correct their compromised immune response induced by LPS. Primary BMMφ cells from MKP-1 KO mice and WT mice were transduced by either Ad.MKP-1 or Ad.LacZ (m.o.i. 300) for 48h. The cells were stimulated with LPS (100ngml−1) for 24h, and cell culture supernatants were analyzed for IL-6, IL-10, TNF-α and chemokine (C-X-C motif) ligand 1 (CXCL1) levels by enzyme-linked immunosorbent assay. As shown in Figure 2, BMMφ from MKP-1 KO mice displayed about twofold higher level of IL-6, threefold higher level of TNF-α and IL-10 and fourfold higher level of CXCL1 following LPS stimulation compared with the cytokine/chemokine levels in WT mice. Importantly, BMMφ cells from MKP-1 KO mice treated with Ad.MKP-1 exhibited a twofold decrease in IL-6, 2.7-fold decrease in TNF-α, 1.6-fold decrease in IL-10 and 2.4-fold decrease in chemokine levels compared with the cytokine levels in cells treated with Ad.LacZ (n=4, *P<0.05, **P<0.01, ***P<0.001). In addition, the cytokine levels in BMMφ from MKP-1 KO mice treated with Ad.MKP-1 reduced to the similar cytokine levels as seen in cells from WT mice treated with Ad.LacZ.

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

MKP-1 gene transfer corrected the defective immune responses seen in macrophages from MKP-1 KO mice after LPS stimulation. Bone marrow macrophages from MKP-1 KO mice or WT mice were transduced with Ad.MKP-1 or control Ad.LacZ (m.o.i. 300) for 48h, cells were stimulated with 100ngml−1 of LPS (from A. actinomycetemcomitans) for 24h. Cytokine expression in supernatant was evaluated by ELISA assay and was normalized by the protein concentration in cell lysates. The data represent the average of three separate experiments. (a) IL-6 expression in mouse bone marrow macrophages (n=4, **P<0.01, ***P<0.001). (b) TNF-α expression in mouse bone marrow macrophages (n=4, ***P<0.001). (c) IL-10 expression in mouse bone marrow macrophages (n=4, **P<0.01, ***P<0.001). (d) CXCL1 expression in mouse bone marrow macrophages (n=4, *P<0.05, ***P<0.001).

Full figure and legend (93K)

To further explore the potential therapeutic effects of gene transfer of MKP-1 in bone homeostasis, the bone marrow cells from MKP-1 KO mice and MKP-1 WT mice were first transduced with either Ad.MKP-1 or control Ad.LacZ (m.o.i. 25) for 48h, then the cells were subsequently incubated with cell culture media containing macrophage-colony stimulating factor (50ngml−1) ± LPS (1mgml−1) for additional 8 days. Osteoclastogenesis was measured in vitro as multinucleated (greater than or equal to3 nuclei) osteoclasts that stained positive for tartrate-resistant acid phosphatase (TRAP). Figure 3a showed that unstimulated cell cultures from both MKP-1 KO and WT mice failed to produce osteoclasts. In contrast, with LPS stimulation, there were significantly more multinucleated osteoclasts (arrows shown) in bone marrow cells from MKP-1 KO mice compared with cells from MKP-1 WT mice. Importantly, total bone marrow cultures treated with Ad.MKP-1 exhibited significant smaller and fewer numbers of TRAP+ osteoclasts in LPS containing cultures compared with control cells transduced with Ad.LacZ. The number of osteoclasts was quantified in each well. As shown in Figure 3b, there was a significant decrease in number of mature osteoclasts in bone marrow cells from MKP-1 KO mice treated with Ad.MKP-1 compared with cells treated with Ad.LacZ (Figure 3b, n=4, **P<0.01, ***P<0.001). We also noticed a decreased number of osteoclasts in bone marrow cells from MKP-1 KO mice treated with Ad.LacZ. However, this effect did not reach significant differences (P>0.05). Overall, these data demonstrate gene transfer of MKP-1 can modulate immune response and inhibit osteoclastogenesis following LPS challenge.

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

MKP-1 gene transfer inhibited osteoclasts formation in bone marrow (BM) cells from MKP-1 KO mice after LPS stimulation. BM cells were extracted from either 8-week-old MKP-1 KO mice or WT mice. Cells were transduced with Ad.MKP-1 or control Ad.LacZ (m.o.i. 25) for 48h and then incubated with MEM-alpha medium containing 10% FBS, 2mM glutamine, 50ngml−1 recombinant mouse M-CSF (R&D systems) with or without LPS (1μgml−1 from A. actinomycetemcomitans) for 8 days. Osteoclasts were stained by tartrate-resistant acid phosphatase (TRAP) staining. (a) Multinucleated osteoclasts (arrows shown) in bone marrow cells from MKP-1 KO mice and WT mice. (b) Number of osteoclasts (more than three nuclei) per well (96-well plate) in bone marrow cells from MKP-1 KO mice and WT mice (n=4, **P<0.01, ***P<0.001).

Full figure and legend (387K)

Local MKP-1 gene transfer alleviated inflammatory bone resorption

To test the therapeutic effect of MKP-1 gene transfer in normal animals under LPS stimulation, we initially needed to determine whether gene expression adenoviral gene expression persisted during the experimental protocol. To optimize the dose of adenoviral delivery, 8-week-old male Sprague–Dawley rats were injected with 4μl of either 1 × 109 plague-forming unit (p.f.u.) or 5 × 108p.f.u. of Ad.LacZ at palatal gingival tissues (21 rats per group). LacZ transgene expression from a single microinjection in gingival tissues was evaluated by β-galactosidase staining in frozen tissue sections 3, 21 and 28 days after adenoviral gene delivery (n=7 per group). Figure 4 illustrates that there was prominent amount of LacZ transgene expression at day 3 following Ad.LacZ injection. The LacZ gene expression was still detectable at days 21 and 28 following adenoviral delivery. These data are consistent with our previous reports26 and suggest that gene expression by local intraoral adenoviral delivery can persist over a long period.

Figure 4.
Figure 4 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

LacZ gene expression in rat gingival tissues after Ad.LacZ injection. Ad.LacZ (1 × 109, or 5 × 108p.f.u. in 4μl) was injected in the gingival tissues of 8-week-old male Sprague–Dawley rats (21 rats per group). Gingival tissues were collected at 3, 21, and 28 days after Ad.LacZ injection (seven rats per group). LacZ gene expression was evaluated by β-galactosidase staining in frozen tissue sections. The scale bars represent 200μm. Pictures represent average LacZ gene expression in each group.

Full figure and legend (229K)

To evaluate the effects of local MKP-1 gene transfer in an experimental model of periodontitis, 8-week-old male Sprague–Dawley rats were either injected with 4μl of either 1 × 109p.f.u. of Ad.MKP-1 (n=17) or control Ad.LacZ (n=17) or HEPES-buffered saline (n=17) at interproximal palatal regions between the maxilla first and second molars. At 48h after the local adenoviral gene delivery, rats were injected with either 2μl of phosphate-buffered saline (PBS) (n=7) or 2μl20μg−1 of LPS (n=10) three times a week for 4 weeks as previously established by our lab group.24, 26, 27 The rats were killed at the end of 4 weeks following LPS injection and the bone loss in the maxilla was evaluated by microcomputed tomography. As shown in Figure 5, rats transduced with Ad.MKP-1 exhibited significantly reduced bone resorption after LPS stimulation compared with rats treated with Ad.LacZ or HEPES control (n=10, *P<0.05). Although there was a trend showing that rats injected with Ad.LacZ reduced bone loss compared with rats injected with HEPES-buffered saline, there was no statistical significant difference (P>0.05) between those two groups.

Figure 5.
Figure 5 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

MKP-1 gene transfer alleviated bone resorption in rats after LPS challenge. Male Sprague–Dawley rats (8-week old) (17 rats per group) were injected either Ad.MKP-1, or Ad. LacZ (1 × 109p.f.u. in 4μl) or HEPES-buffered saline (4μl). At 48h after the Ad. injection, the rats were injected with 2μl of either 20μg of LPS (from A. actinomycetemcomitans) or PBS three times a week for 4 weeks. (a) Representative microcomputed tomography images of rat maxillae from indicated treatment groups. (b) Volumic analysis of bone loss levels (n=7 for PBS groups, n=10 for LPS groups, *P<0.05).

Full figure and legend (132K)

Histological analysis revealed that there were no significant inflammatory cells and few multinucleated osteoclasts on the alveolar bone surface in the periodontal tissues injected with PBS (Figure 6a). In contrast, there were significantly more inflammatory cells (neutrophils, lymphocytes and macrophages), more fibroblasts and more multinucleated osteoclasts (arrows shown) presented in the periodontal tissues injected with LPS (Figure 6b). Importantly, Ad.MKP-1 transduced rats showed significantly reduced amount of inflammatory cells and multinucleated osteoclasts near the alveolar bone surface compared with rats injected with HEPES-buffered saline or Ad.LacZ after LPS stimulation. The multinucleated osteoclasts were confirmed by TRAP staining (Figure 7a), in which there were significantly fewer TRAP-positive osteoclasts on alveolar bone surfaces in periodontal tissues from rats injected with Ad.MKP-1 compared with the number of osteoclasts seen from rats injected with HEPES-buffered saline or Ad.LacZ after LPS stimulation. Quantification of TRAP-positive osteoclasts confirmed that there was a significant reduction of the number of osteoclasts after LPS stimulation in Ad.MKP-1 group compared with control groups (Figure 7b, P<0.001). Immunohistological staining revealed that MKP-1 was present in the periodontal tissues of rats injected with Ad.MKP-1, whereas MKP-1 was undetectable in control groups of rats (Figure 8a). Furthermore, IL-6 expression was significantly decreased in periodontal tissues from rats injected with Ad.MKP-1 compared with IL-6 levels in rats injected with HEPES buffered saline or Ad.LacZ (Figure 8a). Rats treated with Ad.MKP-1 exhibited lower score of IL-6 expression (Figure 8b) compared with control groups of rats.

Figure 6.
Figure 6 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

MKP-1 gene transfer reduced inflammatory infiltration and osteoclastogenesis in rats after LPS challenge. Rat tissue sections (7μm) were stained by hematoxylin and eosin. Histological appearance of lower magnification ( × 40) and higher magnification ( × 400) of periodontal tissues from rat maxillae injected with PBS or LPS. Arrows show multinucleated osteoclasts near the surface of alveolar bone. Scale bars represent 1500μm for lower magnification ( × 40), and 90μm for higher magnification ( × 400).

Full figure and legend (253K)

Figure 7.
Figure 7 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

MKP-1 gene transfer attenuated osteoclastogenesis induced by LPS. Rat tissue sections (7μm) were stained for tartate-resistant acid phosphatase (TRAP). (a) TRAP staining of representative tissue sections of rat maxillae from indicated treatment groups. TRAP-positive (TRAP+) osteoclasts were stained red. The scale bars represent 200μm. (b) Number of TRAP+ cells in each tissue section. Horizontal bar indicates mean TRAP+ cell counts (n=7 for PBS groups, n=10 for LPS groups, ***P<0.001).

Full figure and legend (252K)

Figure 8.
Figure 8 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

MKP-1 gene transfer decreased IL-6 expression in rat tissues after LPS challenge. Rat tissue sections (7μm) were immunohistochemically stained for MKP-1 and IL-6. (a) MKP-1 and IL-6 expression of representative tissue sections of rat maxillae from indicated treatment groups. The scale bars represent 200μm. (b) Histological scores of IL-6 expression in tissue sections (n=7 for PBS groups, n=10 for LPS groups).

Full figure and legend (425K)

Top

Discussion

Bone homeostasis is maintained through a process of coupled bone turnover in which bone formation by osteoblasts is balanced with bone resorptive osteoclasts.28, 29, 30 Osteoclasts are the primary bone resorbing cells, which have a pivotal role in creating bone marrow activity for hematopoiesis, and maintaining bone homeostasis.29, 31 Systemic hormones and cytokines provide the molecular cues that control osteoclastogenesis, and thus maintain homeostasis.30 Within the bone microenvironment, T and B lymphocytes, bone marrow stromal cells, macrophages and osteoblasts all can produce cytokines, which impact osteoclastogenesisc.30 As activation of MAPKs is critical for production of proinflammatory cytokines and osteoclastogenesis, dephosphorylating activated MAPKs by MKP-1 is a critical target to control inflammatory diseases associated with exuberant cytokine production and bone resorption. In this proof-of-concept study, we demonstrated that local MKP-1 gene transfer by adenoviral gene delivery reduced inflammatory mediators and inflammatory bone loss in vivo. Ad.MKP-1 inhibited exaggerated MAPK responses in which cells transduced with Ad.MKP-1 specifically dephosphorylated activated JNK, ERK and p38 MAPK after LPS stimulation, but not activated NF-κB signaling pathway. Ad.MKP-1 attenuated proinflammatory cytokine IL-6, TNF-α, CXCL1, as well as an anti-inflammatory cytokine IL-10 response after LPS stimulation. MKP-1 gene delivery corrected the compromised immune responses seen in MKP-1 KO mice compared with MKP-1 WT mice. In addition, Ad.MKP-1 reduced osteoclastogenesis in bone marrow cells of MKP-1 KO mice compared with controls. Furthermore, Ad.MKP-1 attenuated bone resorption in rats after LPS stimulation compared with bone loss in rats of control groups.

A previous study in Raw 264.7 cells has shown that LPS induced a moderate level of MKP-1 expression at 60–120min following LPS stimulation, which dephosphorylated p-JNK and p-p38, effectively.32 In this study, endogenous induction of MKP-1 by LPS in rat macrophages was very weak as shown in Figure 1a. The level of MKP-1 peaked around at 60–120min following LPS stimulation when incubated with a higher concentration of MKP-1 antibody. Apparently, weak induction of MKP-1 was not potent to dephosphorylate p-JNK, p-p38 and p-ERK as shown in Figure 1b. In contrast, macrophages transduced with Ad.MKP-1 had significantly higher level of MKP-1 expression at 60min following LPS stimulation. The overexpression of MKP-1 significantly inhibited the p-JNK and p-p38 expression, and to a less extent p-ERK expression at 60min following LPS stimulation. This effect lasted at least for 5h following LPS stimulation. These data demonstrate the potential therapeutic effect of overexpression of MKP-1 in regulation of activated MAPKs, both at magnitude and duration levels. The discrepancy between these studies might account for the different cell lines, different strains of LPS and the different dose of LPS used in those studies.

Many inflammatory cytokines have a synergic role in osteoclastogenesis including IL-1,8, 9 IL-6,10 TNF-α,9 and prostaglandin E2,9 which all have been reported to stimulate osteoclastic differentiation and bone resorption. Chemokines are chemotactic signals for monocytes that can facilitate the fusion of monocytes into multinucleated osteoclast.33 Therefore, elevated levels of proinflammatory cytokine and chemokines are osteolytic factors associated with inflammatory bone loss. In this study, bone marrow cells from MKP-1 KO mice displayed significant higher levels of IL-6, TNF-α and chemokine compared with those levels in MKP-1 WT mice (Figure 2). This observation was correlated with more osteoclasts seen in bone marrow cells of MKP-1 KO mice compared with osteoclasts from WT mice in response to LPS stimuli (Figure 3). In addition, this increase osteoclastogenesis in bone marrow cells from MKP-1 KO mice after LPS stimulation is in line with our previous finding that MKP-1 KO mice displayed more bone loss after LPS challenge compared with WT mice.23

In this study, bone marrow cells from MKP-1 KO mice transduced with Ad.MKP-1 significantly decreased of IL-6, TNF-α, IL-10 and CXCL1 levels compared with cells transduced with Ad.LacZ. Our observation demonstrated that MKP-1 has a key role in negatively regulation of IL-6, TNF-α, IL-10 and CXCL1 expression. This observation was in line with a previous study by Zhao et al.20 showing that MKP-1 KO mice produced significantly higher levels of IL-6, TNF-α and IL-10 than WT mice after LPS challenge. By downregulation of osteolytic proinflammatory cytokine and chemokine expression, delivery of Ad.MKP-1 significantly reduced the number of osteoclasts induced by LPS stimulation. Our study emphasized the important role of MKP-1 in maintaining bone homeostasis, suggesting the potential therapeutic effects of MKP-1 gene transfer in diseases associated with inflammatory bone loss. A recent study showed that MKP-1 KO mice had equivalent bone loss, but fewer osteoclasts compared with WT mice after ovariectomy.34 Instead, they showed that osteoclasts from MKP-1 KO mice had higher resorptive activity than osteoclasts from WT mice. Their observation is contradictory to our result that MKP-1 KO mice have more osteoclasts compared with WT mice. The discrepancy among those studies might account for the ovariectomy, which is not a strong stimulator of immune response. A previous study conducted by Li et al.35 revealed that p38 MAPK signals are required for osteoclast differentiation, but not for osteoclastic function. They demonstrated that p38 MAPK inhibitor SB203580 only inhibited osteoclast differentiation, but not affect the survival of osteoclasts or dentine-resorption activity induced by RANKL.

Our lab previously showed that inhibiting p38 MAPK suppressed oral bone loss after LPS challenge in rats.24 Other investigators36 also confirmed the therapeutic role of p38 MAPK inhibitor in inflammatory arthritis. They showed that systemic administration of two different p38 MAPK inhibitors significantly alleviated the TNF-α-mediated inflammatory bone destruction in arthritis. As MKP-1 not only regulates p38 MAPK, but also JNK and ERK activities, overexpression of MKP-1 has more advantages in restraining an overwhelming immune response and osteoclastogenesis in response to stimuli compared with p38 MAPK inhibitors. By local gene transfer of MKP-1, this study demonstrated that MKP-1 is a key factor in maintaining immune and bone homeostasis, MKP-1 gene transfer is a potential therapy in preventing inflammatory bone loss.

It is noteworthy that MKP-1 gene transfer in unstimulated cells only displayed a low level of MKP-1 protein expression. The MKP-1 protein expression was significantly increased at 60min after LPS stimulation. MKP-1 gene transfer did not raise the basal level of proinflammatory cytokine (IL-6, TNF-α), or anti-inflammatory cytokine IL-10 or chemokine levels. MKP-1 gene transfer only restrained an exaggerated immune response after LPS stimulation. Moreover, MKP-1 gene transfer not only restrained proinflamamtory cytokines (IL-6, TNF-α) and chemokine, but also inhibited an anti-inflammatory cytokine IL-10 after stress stimulation. These data specify the important role of MKP-1 in maintaining immune homeostasis. MKP-1 gene is only activated on stress stimuli, which has a critical role in restraining an exaggerated immune response and maintaining immune homeostasis after stress stimuli.

Recombinant Ad. may offer an advantage for local intraoral usage, as the target gene can be expressed over a relatively long period of time compared with chemical drugs, or protein-based drugs, which can be degraded quickly. In a previous study in oral delivery of an adenoviral vector carrying a luciferase-tagged gene,26 we demonstrated that luciferase activity persisted over 21 days after the adenoviral vector delivery. In this study, we demonstrated that the LacZ gene expression last 28 days following Ad.LacZ single injection. In this study, we noticed a trend of less number of osteoclasts and lesser bone loss in cells and animals treated with Ad.LacZ after stimulated by LPS, although it is not statistically significant different. This observation might be attributed to the property of Ad. in inhibiting cell growth, as we noticed less cell number/and less protein concentration with increasing amount of Ad. in cells.

In summary, exuberant MAPK response is associated with increasing production of inflammatory cytokines, which leads to inflammatory bone loss. MKP-1 gene transfer by recombinant Ad. is a potential strategy for inflammatory bone loss and for those diseases associated with exuberant MAPK responses.

Top

Materials and methods

Cells and lipopolysaccharide

Rat macrophage NR8383 cell line (ATCC CRL-2192) was obtained from American Type Cell Collection (ATCC, Manassas, VA, USA). NR8383 cells were cultured with Ham's F12K medium (ATCC), supplemented with 15% heat inactivated fetal bovine serum, 100IUml−1 penicillin and100μgml−1 streptomycin in a 37°C incubator with 5% CO2. The human embryonic kidney 293A cell line was purchased from Invitrogen (Carlsbad, CA, USA). The 293A cells were cultured in Dulbecco's modified Eagle medium with high glucose, supplemented with 10% cosmic calf serum, 100IUml−1 penicillin and 100μgml−1 streptomycin in a 37°C incubator with 5% CO2. LPS from Aggregatibacter actinomycetemcomitans (strain Y4, serotype B) was extracted by the hot phenol-water method as described37 and diluted in PBS.

Adenovirus

Ad5-CMV-MKP-1 phosphatase, which express the full-length of MKP-1 gene under the cytomegalovirus promoter, and control Ad5-CMV-LacZ were obtained from Seven Hills Bioreagents (Cincinnati, OH, USA). Adenovirus were propagated in human embryonic kidney 293A cells, purified by cesium chloride density gradient ultracentrifugation method and desalted by PD-10 column (GE Healthcare, Pittsburgh, PA, USA) in HEPES-buffered saline.

Western blot analysis

Protein was extracted from NR8383 cells 48h after Ad. transduction with/or without LPS stimulation by cell lysis buffer (Cell Signaling Technology, Danvers, MA, USA). Total protein (30μg) was loaded on 10% Tris-HCl Ready Gel (Bio-Rad Laboratories, Hercules, CA, USA) and electrotransferred to nitrocellulose membranes, blocked and then incubated overnight at 4°C with primary antibodies. The primary antibody for MKP-1 (M-18) was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). The primary antibody of p-p38 MAPK; p-SAPK/JNK; p-p44/42 MAPK; p38 MAPK; and p-NF-κB p65 were purchased from Cell Signaling Technology. The presence of the primary antibodies was detected on radiographic film by using Horseradish peroxidase-conjugated secondary antibodies (Cell Signaling Technology) and SuperSignal West Pico Chemiluminescent Substrate (Thermo Fisher Scientific, Rockford, IL, USA). Digitalized images of the radiographic films were obtained in a gel documentation system (Bio-Rad Laboratories).

ELISA

Mouse IL-6, IL-10, TNF-α, and chemokine (C-X-C motif) ligand 1 (CXCL1) were measured in culture medium by enzyme-linked immunosorbent assay kits purchased from R&D systems (Minneapolis, MN, USA). The protein concentration in cell lysates was determined by DC protein Assay Kit (Bio-Rad Laboratories). The concentration of cytokines was normalized by protein concentrations in cell lysates.

Animals

MKP-1 KO mice and WT mice were provided by Bristol-Myers Squibb Pharmaceutical Research Institute (Lawrenceville, NJ, USA)and bred at Medical University of South Carolina. These mice are maintained on a mixed C57/129 background. Male Sprague–Dawley rats (8-week old) were purchased from Charles River Laboratory (Wilmington, MA, USA) with food and tap water ad libitum. All animal-related work was performed in accordance with the National Institute of Health guidelines. The institutional animal care and use committee at the Medical University of South Carolina approved all experimental protocols. For adenoviral gene expression experiment, 8-week-old male Sprague–Dawley rats were injected with 4μl of either 1 × 109 p.f.u. dose (21 rats per group) or 5 × 108p.f.u. dose (21 rats per group) of Ad.LacZ at palatal gingival tissues. The rats were killed at 3, 21 or 28 days following Ad. delivery. For bone resorption experiment, 8-week-old male Sprague–Dawley rats were injected with 4μl of either 1 × 109p.f.u. of Ad.MKP-1 (n=17) or control Ad.LacZ (n=17) or HEPES-buffered saline (n=17) at interproximal palatal regions between the maxilla first and second molars. At 48h after delivery of Ad., the rats were injected with 2μl of PBS (n=7) or 2μl20μg−1 of LPS (n=10) three times a week for 4 weeks. The rats were killed at the end of 4 weeks following LPS injection.

Bone marrow extraction and macrophage separation

Bone marrow cells were collected from 8-week-old MKP-1 KO/WT mice. The femurs and tibias were flushed with minimal essential alpha medium (Invitrogen) The bone marrow cells were grown in minimal essential α medium containing 10% fetal bovine serum, 2mM glutamine, 50ngml−1 recombinant murine granulocyte/macrophage colony stimulating factor (R&D systems), 100IUml−1 penicillin and 100μgml−1 of streptomycin. Cells were incubated at 37°C in 5% CO2 overnight. Nonadherent cells were collected. CD11b positive cells were magnetically selected using CD11b MicroBeads by an autoMACS Separator (Miltenyi Biotec, Auburn, CA, USA). Subsequently, the CD11b positive cells were grown in Dulbecco's modified Eagle medium containing 10% fetal bovine serum, 2mM glutamine, 50ngml−1 recombinant murine macrophage-colony stimulating factor (R&D systems) 100Uml−1 penicillin and 100μgml−1 of streptomycin for additional 7 days.

β-galactosidase staining in tissue sections

Tissue sections (30μm) were fixed with 0.5% glutaraldehyde in PBS for 15min at room temperature. After washing with PBS twice, the tissue sections were stained in staining solution at 37°C overnight. The staining solution contains 1mgml−1 X-gal (Sigma Aldrich, St Louis, MO, USA), 2mM MgCl2, 5mM potassium hexacyanoferrate (Sigma Aldrich) and 5mM potassium hexacyanoferrate trihydrate (Sigma Aldrich). After washing with PBS, the sections were counterstained with nuclear fast red solution (Sigma Aldrich) at room temperature for 3min. Finally the slides were washed in water and mounted with aqueous mounting medium.

Microcomputed tomography

Nondemineralized rat maxillae (n=10 per group) were scanned by a cone beam microcomputed tomography system (GE Healthcare BioSciences, Chalfont St Giles, UK) as previously described.24 Each scan was reconstructed at a mesh size of 18 × 18 × 18μm, and three-dimensional digitized images were generated for each specimen. Using GEHC MicroView software (GE Healthcare BioSciences, Chalfont St Giles, UK) (version viz.+2.0build 0029), the images were rotated into a standard orientation and threshold to distinguish between mineralized and nonmineralized tissue. For each specimen, a grayscale voxel value histogram was generated to determine an optimal threshold value. Volumic alveolar bone losses expressed as bone volume fraction were measured as previously described.24

Hematoxylin and eosin staining and immunohistological staining in tissue sections

Formalin-fixed specimens were decalcified in a 10% EDTA solution for 2 weeks at 4°C. The EDTA solution was changed three times per week. The maxillas were paraffin-embedded, and 7μm sagittal sections were prepared. Some sections were stained with hematoxylin and eosin for descriptive histology.

For immunohistological staining, tissue sections were deparaffinized in xylene and rehydrated in graded ethanol. After antigen retrieval in citric buffer at 100°C for 10min, the slides were incubated with 3% H2O2 for 10min to quench endogenous peroxidase activity. After washing and incubation with block serum for 30min at room temperature, the sections were incubated with anti-MKP-1 (M-18 1:200, Santa Cruz Biotechnology), or with anti-rat IL-6 (1:800, R&D systems) at 4°C overnight. IL-6 expression was detected using a VECTASTAIN Universal ABC kit (Vector Laboratories Inc., Burlingame, CA, USA) following manufacturer's instruction. Finally, the sections were counterstained with hematoxylin, dehydrated in graded ethanol and covered with coverslips in mounting media. The control sections were incubated with preimmune serum to assess background staining. Histological slides were evaluated and graded by an experienced pathologist.

Tartate-resistant acid phosphatase staining

TRAP staining was performed in cultured bone marrow cells and tissue sections using a leukocyte acid phosphatase kit (Sigma Aldrich). The tissue sections were counterstained with hematoxylin after TRAP staining. Active osteoclasts were defined as multinucleated TRAP-positive cells in contact with bone surface.

Statistical analysis

Data were analyzed by Student's t test with Welch's correction for unequal variances. All statistical tests were performed using a GraphPad Prism software (GraphPad Software Inc., La Jolla, CA, USA). P values of 0.05 or less were considered significant.

Top

Conflict of interest

The authors declare no conflict of interest.

Top

References

  1. Brown LJ, Loe H. Prevalence, extent, severity and progression of periodontal disease. Periodontol 2000 1993; 2: 57–71. | Article | PubMed | OpenURL | ChemPort |
  2. Pilot T, Miyazaki H. Periodontal conditions in Europe. J Clin Periodontol 1991; 18: 353–357. | Article | PubMed | ISI | OpenURL | ChemPort |
  3. Hall TJ, Chambers TJ. Molecular aspects of osteoclast function. Inflamm Res 1996; 45: 1–9. | Article | PubMed | OpenURL |
  4. Reddy SV. Regulatory mechanisms operative in osteoclasts. Crit Rev Eukaryot Gene Expr 2004; 14: 255–270. | Article | PubMed | OpenURL |
  5. Nair SP, Meghji S, Wilson M, Reddi K, White P, Henderson B. Bacterially induced bone destruction: mechanisms and misconceptions. Infect Immun 1996; 64: 2371–2380. | PubMed | ISI | ChemPort |
  6. Taubman MA, Valverde P, Han X, Kawai T. Immune response: the key to bone resorption in periodontal disease. J Periodontol 2005; 76: 2033–2041. | Article | PubMed | OpenURL |
  7. Mundy GR. Inflammatory mediators and the destruction of bone. J Periodontal Res 1991; 26: 213–217. | Article | PubMed | OpenURL |
  8. Hanazawa S, Amano S, Nakada K, Ohmori Y, Miyoshi T, Hirose K et al. Biological characterization of interleukin-1-like cytokine produced by cultured bone cells from newborn mouse calvaria. Calcif Tissue Int 1987; 41: 31–37. | Article | PubMed | OpenURL |
  9. Zou W, Bar-Shavit Z. Dual modulation of osteoclast differentiation by lipopolysaccharide. J Bone Miner Res 2002; 17: 1211–1218. | Article | PubMed | OpenURL |
  10. Ishimi Y, Miyaura C, Jin CH, Akatsu T, Abe E, Nakamura Y et al. IL-6 is produced by osteoblasts and induces bone resorption. J Immunol 1990; 145: 3297–3303. | PubMed | ISI | ChemPort |
  11. Yang SH, Sharrocks AD, Whitmarsh AJ. Transcriptional regulation by the MAP kinase signaling cascades. Gene 2003; 320: 3–21. | Article | PubMed | ISI | OpenURL | ChemPort |
  12. Whitmarsh AJ. Regulation of gene transcription by mitogen-activated protein kinase signaling pathways. Biochim Biophys Acta 2007; 1773: 1285–1298. | Article | PubMed | OpenURL |
  13. Keyse SM. Protein phosphatases and the regulation of mitogen-activated protein kinase signalling. Curr Opin Cell Biol 2000; 12: 186–192. | Article | PubMed | ISI | OpenURL | ChemPort |
  14. Orcel P, Feuga M, Bielakoff J, De Vernejoul MC. Local bone injections of LPS and M-CSF increase bone resorption by different pathways in vivo in rats. Am J Physiol 1993; 264: E391–E397. | PubMed |
  15. Nishida E, Hara Y, Kaneko T, Ikeda Y, Ukai T, Kato I. Bone resorption and local interleukin-1alpha and interleukin-1beta synthesis induced by Actinobacillusactinomycetemcomitans and Porphyromonasgingivalis lipopolysaccharide. J Periodontal Res 2001; 36: 1–8. | Article | PubMed | OpenURL | ChemPort |
  16. Jiang Y, Mehta CK, Hsu TY, Alsulaimani FF. Bacteria induce osteoclastogenesis via an osteoblast-independent pathway. Infect Immun 2002; 70: 3143–3148. | Article | PubMed | ISI | OpenURL | ChemPort |
  17. Zhuang L, Jung JY, Wang EW, Houlihan P, Ramos L, Pashia M et al. Pseudomonasaeruginosa lipopolysaccharide induces osteoclastogenesis through a toll-like receptor 4 mediated pathway in vitro and in vivo. Laryngoscope 2007; 117: 841–847. | Article | PubMed | OpenURL |
  18. Ozaki Y, Ukai T, Yamaguchi M, Yokoyama M, Haro ER, Kaneko T et al. Locally administered T cells from mice immunized with lipopolysaccharide (LPS) accelerate LPS-induced bone resorption. Bone 2009; 44: 1169–1176. | Article | PubMed | OpenURL |
  19. Owens DM, Keyse SM. Differential regulation of MAP kinase signalling by dual-specificity protein phosphatases. Oncogene 2007; 26: 3203–3213. | Article | PubMed | ISI | OpenURL | ChemPort |
  20. Zhao Q, Wang X, Nelin LD, Yao Y, Matta R, Manson ME et al. MAP kinase phosphatase 1 controls innate immune responses and suppresses endotoxic shock. J Exp Med 2006; 203: 131–140. | Article | PubMed | ISI | OpenURL | ChemPort |
  21. Chi H, Barry SP, Roth RJ, Wu JJ, Jones EA, Bennett AM et al. Dynamic regulation of pro- and anti-inflammatory cytokines by MAPK phosphatase 1 (MKP-1) in innate immune responses. Proc Natl Acad Sci USA 2006; 103: 2274–2279. | Article | PubMed | OpenURL | ChemPort |
  22. Salojin KV, Owusu IB, Millerchip KA, Potter M, Platt KA, Oravecz T. Essential role of MAPK phosphatase-1 in the negative control of innate immune responses. J Immunol 2006; 176: 1899–1907. | PubMed | ISI | ChemPort |
  23. Sartori R, Li F, Kirkwood KL. MAP kinase phosphatase-1 protects against inflammatory bone loss. J Dent Res 2009; 88: 1125–1130. | Article | PubMed | OpenURL |
  24. Rogers JE, Li F, Coatney DD, Otremba J, Kriegl JM, Protter TA et al. A p38 mitogen-activated protein kinase inhibitor arrests active alveolar bone loss in a rat periodontitis model. J Periodontol 2007; 78: 1992–1998. | Article | PubMed | OpenURL | ChemPort |
  25. Wang X, Liu Y. Regulation of innate immune response by MAP kinase phosphatase-1. Cell Signal 2007; 19: 1372–1382. | Article | PubMed | OpenURL |
  26. Patil CS, Liu M, Zhao W, Coatney DD, Li F, VanTubergen EA et al. Targeting mRNA stability arrests inflammatory bone loss. Mol Ther 2008; 16: 1657–1664. | Article | PubMed | OpenURL |
  27. Kirkwood KL, Li F, Rogers JE, Otremba J, Coatney DD, Kreider JM et al. A p38alpha selective mitogen-activated protein kinase inhibitor prevents periodontal bone loss. J Pharmacol Exp Ther 2007; 320: 56–63. | Article | PubMed | OpenURL | ChemPort |
  28. Marie PJ. Human endosteal osteoblastic cells: relationship with bone formation. Calcif Tissue Int 1995; 56 (Suppl 1): S13–S16. | PubMed |
  29. Miyamoto T, Suda T. Differentiation and function of osteoclasts. Keio J Med 2003; 52: 1–7. | PubMed |
  30. Yavropoulou MP, Yovos JG. Osteoclastogenesis—current knowledge and future perspectives. J Musculoskelet Neuronal Interact 2008; 8: 204–216. | PubMed |
  31. Teitelbaum SL. Osteoclasts: what do they do and how do they do it? Am J Pathol 2007; 170: 427–435. | Article | PubMed | OpenURL | ChemPort |
  32. Chen P, Li J, Barnes J, Kokkonen GC, Lee JC, Liu Y. Restraint of proinflammatory cytokine biosynthesis by mitogen-activated protein kinase phosphatase-1 in lipopolysaccharide-stimulated macrophages. J Immunol 2002; 169: 6408–6416. | PubMed | ISI | ChemPort |
  33. Kim MS, Magno CL, Day CJ, Morrison NA. Induction of chemokines and chemokine receptors CCR2b and CCR4 in authentic human osteoclasts differentiated with RANKL and osteoclast like cells differentiated by MCP-1 and RANTES. J Cell Biochem 2006; 97: 512–518. | Article | PubMed | ISI | OpenURL | ChemPort |
  34. Carlson J, Cui W, Zhang Q, Xu X, Mercan F, Bennett AM et al. Role of MKP-1 in osteoclasts and bone homeostasis. Am J Pathol 2009; 175: 1564–1573. | Article | PubMed | OpenURL |
  35. Li X, Udagawa N, Itoh K, Suda K, Murase Y, Nishihara T et al. p38 MAPK-mediated signals are required for inducing osteoclast differentiation but not for osteoclast function. Endocrinology 2002; 143: 3105–3113. | Article | PubMed | ISI | OpenURL | ChemPort |
  36. Zwerina J, Hayer S, Redlich K, Bobacz K, Kollias G, Smolen JS et al. Activation of p38 MAPK is a key step in tumor necrosis factor-mediated inflammatory bone destruction. Arthritis Rheum 2006; 54: 463–472. | Article | PubMed | ISI | OpenURL | ChemPort |
  37. Wilson ME, Hamilton RG. Immunoglobulin G subclass response of localized juvenile periodontitis patients to Actinobacillusactinomycetemcomitans Y4 lipopolysaccharide. Infect Immun 1992; 60: 1806–1812. | PubMed |
Top

Acknowledgements

We acknowledge the continued support of the University of Michigan Orthopedic Research Labs and the Muscloskeletal Core for the microcomputed tomography. This study is supported by NIH 1R01DE018290 and 2P20 RR017696. This study was conducted in a facility constructed with support from the NIH C06 RR015455 from the Extramural Research Facilities Program of the National Center for Research Resources.