SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Objective

While the effects of biomechanical signals in the form of joint movement and exercise are known to be beneficial to inflamed joints, limited information is available regarding the intracellular mechanisms of their actions. This study was undertaken to examine the intracellular mechanisms by which biomechanical signals suppress proinflammatory gene induction by the interleukin-1-β (IL-1β)–induced NF-κB signaling cascade in articular chondrocytes.

Methods

Primary rat articular chondrocytes were exposed to biomechanical signals in the form of cyclic tensile strain, and the effects on the NF-κB signaling cascade were examined by Western blot analysis, real-time polymerase chain reaction, and immunofluorescence.

Results

Cyclic tensile strain rapidly inhibited the IL-1β–induced nuclear translocation of NF-κB, but not its IL-1β–induced phosphorylation at serine 276 and serine 536, which are necessary for its transactivation and transcriptional efficacy, respectively. Examination of upstream events revealed that cyclic tensile strain also inhibited the cytoplasmic protein degradation of IκBβ and IκBα, as well as repressed their gene transcription. Additionally, cyclic tensile strain induced a rapid nuclear translocation of IκBα to potentially prevent NF-κB binding to DNA. Furthermore, the inhibition of IL-1β–induced degradation of IκB by cyclic tensile strain was mediated by down-regulation of IκB kinase activity.

Conclusion

These results indicate that the signals generated by cyclic tensile strain act at multiple sites within the NF-κB signaling cascade to inhibit IL-1β–induced proinflammatory gene induction. Taken together, these findings provide insight into how biomechanical signals regulate and reduce inflammation, and underscore their potential in enhancing the ability of chondrocytes to curb inflammation in diseased joints.

Despite the known importance of cytokines in cartilage destruction and the established effectiveness of exercise or joint mobilization in the restoration of joint function, relatively little is understood about the molecular events underlying the beneficial effects of biomechanical signals on inflamed cartilage. Recent molecular studies have revealed that specific biomechanical stimuli and cell interactions generate intracellular signals that are powerful suppressors of cytokine-mediated proinflammatory gene transcription in chondrocytes, in vitro and in vivo (1, 2). Cytokine production by synovial cells as well as chondrocytes in joints has been linked to the pathogenesis of arthritic diseases (3–5).

Additionally, compelling evidence from studies of transgenic mice genetically engineered for overexpression of tumor necrosis factor α (TNFα) demonstrates that elevated levels of TNFα lead to early and severe induction of rheumatoid arthritis in the joints (6). Similarly, gene delivery of interleukin-1 receptor antagonist (IL-1Ra) or anti-TNFα/IL-1β antibodies to arthritic joints leads to suppression of cartilage destruction, further implicating proinflammatory cytokines in the etiopathology of arthritis (7). Given the importance of the role played by these cytokines in cartilage destruction and the potent effects of biomechanical signals in antagonizing proinflammatory gene transcription, it is critical to characterize the signaling cascade that may be essential in mediating the profound actions of exercise or joint mobilization in preventing joint destruction (8–10).

Several previous studies have identified NF-κB transcription factors as key regulators of TNFα-induced and IL-1β–induced gene activation in chondrocytes (11–15). Furthermore, biomechanical signals have been shown to regulate cytokine gene expression via regulation of the NF-κB signaling pathway. In chondrocytes, biomechanical signals have been found to promote or attenuate proinflammatory gene transcription in a magnitude-dependent manner (11). Biomechanical signals are transduced to cells by surface molecules such as β integrins and focal adhesion kinases (FAKs) (16). At high magnitudes these signals trigger activation of the NF-κB signaling cascade to induce proinflammatory gene transcription (11). Notably, at low magnitudes, biomechanical signals fail to activate NF-κB transcription factors, and act as potent inhibitors of IL-1β– and TNFα-dependent proinflammatory gene transcription (11, 17, 18). However, the molecular components of this signaling cascade regulated by biomechanical signals have not yet been described.

Members of the NF-κB family of transcription factors are sequestered in the cytoplasm of unstimulated cells as heterodimers or homodimers by binding to their inhibitory units, IκBα and IκBβ. Multiple cytokine-induced proinflammatory pathways converge at the signalsome comprising IKKα, IKKβ, and IKKγ/NF-κB–essential modulator, to activate downstream events in the NF-κB cascade. Upon phosphorylation by IKK, IκB proteins are ubiquitinated and marked for proteosomal degradation. The liberation of NF-κB from IκB complexes is followed by phosphorylation of NF-κB at multiple sites in a stimulant-dependent manner and eventual translocation to the nucleus.

The binding of NF-κB to its consensus sequences leads to transcription of a plethora of genes, including proinflammatory cytokines and mediators, as well as several of the molecules required for the activation of the NF-κB signaling cascade. Although this classic model of NF-κB activation by TNFα or IL-1β is well documented, its complexity evolves from its regulation at multiple intracellular levels, in a cell-dependent as well as stimulus-dependent manner.

In an effort to identify the key signaling and regulatory mechanisms that allow biomechanical signals to inhibit IL-1β–induced NF-κB activation, we examined the expression of NF-κB–controlled genes, such as tumor necrosis factor receptor–associated factor 1 (TRAF1) and TRAF2 and IL-1 receptor type I (IL-1RI) and IL-1RII, as transcription biomarkers. We focused on IL-1β–induced NF-κB nuclear translocation and prerequisite upstream events involved in its activation, in order to identify the key target molecule(s) that are regulated by biomechanical signals.

The results of the present study showed that, although IKK is the central target of biomechanical signals, these signals intercept multiple sites along the NF-κB signal transduction pathway to block IL-1β–induced proinflammatory gene expression. Recently, there has been a substantial effort to develop drugs and gene therapy approaches targeted at inhibiting NF-κB or IKK in order to prevent cartilage and bone destruction in arthritis (14, 19–21). Notably, appropriate biomechanical signals can inhibit both NF-κB and IKKβ, suggesting that these signals can be used as an effective therapeutic intervention, without side effects, to suppress joint inflammation in arthritic diseases. Furthermore, these signals, while functioning as antiinflammatory cues, also act as powerful inducers of genes that contribute to cartilage repair and matrix synthesis (19, 20). Hence, understanding the mechanisms of the intracellular actions by which biomechanical signals attenuate NF-κB–induced proinflammatory gene transcription to limit cartilage destruction is of critical importance.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Primary chondrocyte isolation and cell culture.

Articular cartilage was harvested from the knees of 10–12-week-old female Sprague-Dawley rats (Harlan, Indianapolis, IN), following approval by the Institutional Laboratory Animal Care and Use Committee at The Ohio State University. Primary chondrocytes were isolated and cultured as previously described (11). Primary chondrocytes were used during the first 3 passages, in which phenotypic markers, aggrecan, SOX9, and type II collagen were expressed, and type I collagen was not expressed, as assessed by real time polymerase chain reaction (PCR).

Application of cyclic tensile strain.

Chondrocytes (7 × 104/well) were grown for 4–5 days on type I collagen–coated BioFlex 6-well culture plates (Flexcell International, Hillsborough, NC) to 80% confluence in humidified 5% CO2 at 37°C. The medium was replaced with reduced serum medium (1% fetal bovine serum) 24 hours before experiments were initiated. Plates were placed in an FX-4000T bioreactor (Flexcell International) and subjected to cyclic tensile strain at a magnitude of 3% and frequency of 25 MHz. This magnitude of tensile forces translates into ∼10.8 kPa, assuming that the shear modulus of rat chondrocytes is 0.36 kPa (22). The following 4 different treatment groups were devised: untreated controls, chondrocytes treated with 1 ng/ml recombinant human IL-1β (Calbiochem, La Jolla, CA), chondrocytes subjected to cyclic tensile strain, and chondrocytes subjected to cyclic tensile strain and treated with recombinant human IL-1β. In these experiments the chondrocytes from the first 3 passages all responded to cyclic tensile strain in a similar manner.

Analysis of messenger RNA (mRNA) expression by real-time PCR.

Total RNA was isolated using the RNeasy Mini kit, according to the recommendations of the manufacturer (Qiagen, Chatsworth, CA). RNA (1 μg) was reverse-transcribed into complementary DNA, and real-time PCR was performed with the TaqMan assay in a volume of 25 μl at 65°C for 5 minutes. Real-time PCR was performed in a volume of 25 μl using a Bio-Rad iCycler iQ (Bio-Rad, Richmond, CA), as previously described (23). Primers and probes used are shown in Table 1.

Table 1. Primers and probes used in real-time polymerase chain reactions*
  • *

    NOS2 = type 2 nitric oxide synthase; IL-1RI = interleukin-1 receptor type I; TRAF1 = tumor necrosis factor receptor–associated factor 1.

NOS25′-TTCTGTGCTAATGCGGAAGGT-3′ (sense)
 5′-GCTTCCGACTTTCCTGTCTCA-3′ (antisense)
 5′-CCGCGTCAGAGCCACAGTCCT-3′ (probe)
IκBα5′-GGTATACTTAGCACCACAGCACACA-3′ (sense)
 5′-CCCCAAATTTCACAAGAACAACA-3′ (antisense)
 5′-CCTAGCCCCGAGCATTCTATTGTGGTGAT-3′ (probe)
IκBβ5′-CCATGTAGCTGTCATCCACAAAG-3′ (sense)
 5′-ACGTAGGCTCCGGTTTATGAG-3′ (antisense)
 5′-AGAGATGGTCCAACTGCTCAGGGATGCT-3′ (probe)
IL-1RI5′-AGCCCTGTGCCGTAATGTG-3′ (sense)
 5′-TCATATTCTCCAGGATGGAGGATT-3′ (antisense)
 5′-TTCATGCTGCCGAGGCTTGTGACA-3′ (probe)
IL-1RII5′-ATCGTGTGGTGGATGGCTAAC-3′ (sense)
 5′-CTGGTGGTGTAGCCCCTCAGT-3′ (antisense)
 5′-TCAGTGGCCTACCCAAGAGGCC-3′ (probe)
TRAF15′-TGGCTGTGGCTAATGTGAGACT-3′ (sense)
 5′-TGGTGTTGTGGTCTGTGTGAAG-3′ (antisense)
 5′-AGAGACAGTGGAGGAGAAGACAGAAGTGCT-3′ (probe)
TRAF25′-CTGTCCCAATGATGGATGCA-3′ (sense)
 5′-CAGCAGGAATGGGCACAGT-3′ (antisense)
 5′-ACCTTGAAAGAATACGAGAGCTGCCACGA-3′ (probe)

Analysis of NF-κB and IκB protein expression.

Regulation of NF-κB and IκB interaction was examined by Western blot analysis. Total cellular protein extracts were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) on a 10% gel and transferred onto nitrocellulose membranes (Bio-Rad). The membranes were probed with antibodies to anti-IκBα, anti-IκBβ (Santa Cruz Biotechnology, Santa Cruz, CA), anti–phospho-IκBα serine 32–36, anti–phospho-NF-κB p65 serine 536, and anti–phospho-NF-κB p65 serine 276 (Cell Signaling Technology, Beverly, MA). The binding of primary antibodies was detected with IR-dye 800CW– or IR-dye 680–conjugated secondary antibodies (Li-Cor, Lincoln, NE). The concentration curve for IR-dye was established by assessing the fluorescence of various concentrations of β-actin in Western blots. The total input of experimental proteins was within the linear range of the concentration curve (24). Membranes were imaged with the Odyssey Infrared Imaging System and software (Li-Cor) at 700 nm and 800 nm, using a resolution of 169 μm.

IKK activation.

IKK activation in response to IL-1β and cyclic tensile strain was investigated essentially as described previously (25). Briefly, chondrocytes were suspended in lysis buffer (20 mM Tris [pH 8.0], 500 mM NaCl, 0.25% Triton X-100, 1 mM EDTA, 1 mM EGTA, and 10 mM β-glycerophosphate) for 20 minutes at 4°C. IKK complexes were immunoprecipitated with anti-IKKγ antibody and A/G agarose beads (Santa Cruz Biotechnology), washed with lysis buffer, and then incubated with 2 μg of glutathione S-transferase (GST)–IκBα–purified protein and 0.5 μmoles of ATP at 30°C for 2 hours, in a final volume of 100 μl. The reaction was stopped by adding loading buffer, and the proteins were separated by size by SDS–10% PAGE, transferred to nitrocellulose membranes, and probed with either anti–phospho-IκBα serine 32 and 36 or anti-IκBα antibodies. IR-dye 800CW goat anti-rabbit antibody or IR-dye 680 goat anti-mouse antibodies were used as secondary antibodies, and membranes were imaged with the Odyssey Infrared Imaging System, as described above.

Immunofluorescence (IF).

Localization of NF-κB, IκBα, and IκBβ was assessed by IF using rabbit anti–NF-κB p65 IgG, anti-IκBα, and anti-IκBβ as primary antibodies and Cy3- or Cy2-conjugated goat anti-rabbit IgG (The Jackson Laboratory, Bar Harbor, ME) as secondary antibodies. Fluorescein isothiocyanate–conjugated phalloidin (Santa Cruz Biotechnology) was used to visualize filamentous actin. Stained chondrocytes on Bioflex membranes were mounted on slides and visualized under an epifluorescence microscope (Zeiss Axioimage; Carl Zeiss Instruments, Oberkochen, Germany), and the intensity of the fluorescence in cells was analyzed using Zeiss AxioVision software (Carl Zeiss Instruments). A nuclear mask was created using 4′,6-diamidino-2-phenylindole staining, and NF-κB fluorescence intensity was measured in the nuclear contour. Subsequently, NF-κB cytoplasmic fluorescence intensity was calculated by subtracting nuclear fluorescence from total fluorescence. Five different microscopic fields of 2.5 × 105 μm2 in each segment of a 6-well Bioflex plate were randomly selected and analyzed, and the mean and SEM were calculated.

Electrophoretic mobility shift assay (EMSA).

IR-dye–labeled NF-κB consensus duplex oligonucleotides (50 nmoles) (sense 5′-AGTTGAGGGGACTTTCCCAGGC-3′; Li-Cor) were added to the EMSA binding buffer (Lightshift EMSA kit; Pierce, Rockford, IL), containing 20 μg of nuclear proteins from chondrocytes exposed to various treatment regimens. DNA protein complexes were allowed to incubate at room temperature for 20 minutes. The complexes were resolved and separated from unbound labeled DNA by electrophoresis through a 5% nondenaturing polyacrylamide gel at 100V for 1 hour in 0.5× Tris–borate–EDTA buffer. Subsequently, the gel was imaged directly using an Odyssey Infrared Imaging System (26).

Statistical analysis.

Data were analyzed using SPSS 13.0 software (SPSS, Chicago, IL) and expressed as the mean ± SEM. At least 3 independent experiments were performed. One-way analysis of variance and Dunnett's post hoc multiple comparison test were used to compare stretched IL-1β–treated chondrocytes and unstretched IL-1β–treated chondrocytes. P values less than 0.05 were considered significant.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Suppression of IL-1β–dependent NF-κB nuclear translocation and DNA binding by cyclic tensile strain.

The inhibitory effects of cyclic tensile strain on IL-1β–dependent proinflammatory gene transcription have been documented previously (1). To further investigate the intracellular mechanisms of the antiinflammatory actions of cyclic tensile strain, we examined the possible involvement of the NF-κB signaling pathway as a target of cyclic tensile strain. In these experiments, chondrocytes were exposed to preoptimized magnitudes of cyclic tensile strain (3% equibiaxial strain) in the presence or absence of predetermined concentrations of IL-1β (1 ng/ml) for 10, 30, 60, or 90 minutes. The binding of NF-κB to its consensus sequences in the nucleus was determined by EMSA.

In unstimulated chondrocytes and chondrocytes subjected to cyclic tensile strain alone, binding of NF-κB to its consensus sequences was not observed between 10 and 90 minutes. Activation of chondrocytes with IL-1β resulted in a rapid increase in DNA binding of NF-κB by 10 minutes, which was sustained through 90 minutes (Figure 1A). Notably, concomitant exposure of chondrocytes to cyclic tensile strain and IL-1β resulted in near-total inhibition of IL-1β–induced NF-κB binding to its consensus sequences during the first 10 minutes, and >90% inhibition over the ensuing 80 minutes (Figure 1A).

thumbnail image

Figure 1. Inhibition of interleukin-1β (IL-1β)–induced NF-κB DNA binding, nuclear translocation, and transcription by cyclic tensile strain (CTS). Rat chondrocytes cultured on Bioflex plates were either untreated (control) or exposed to IL-1β (1.0 ng/ml) alone, cyclic tensile strain (3% and 0.05 Hz) and IL-1β, or cyclic tensile strain alone for 10, 30, 60, or 90 minutes, as indicated. A, Results of electrophoretic mobility shift assay (EMSA) of nuclear extracts, using IR-dye–labeled NF-κB consensus sequences. Lane 1, Untreated control cells; lanes 2, 4, 6, and 8, cells treated with IL-1β; lanes 3, 5, 7, and 9, cells treated with CTS and IL-1β; lane 10, IL-1β–treated cells incubated with a 100-fold excess of unlabeled probe (ExL); lane 11, cells treated with CTS; lanes 12 and 13, results of super-shift EMSA, demonstrating NF-κB subunit p65. B, Localization of cytoplasmic NF-κB p65 (green arrows) and nuclear NF-κB p65 (white arrows) following treatment with IL-1β in the presence and absence of cyclic tensile strain, determined by immunofluorescence staining. NF-κB p65 was stained red, and β-actin was stained green. Bar = 20 μm. C and D, Semiquantitative analysis of nuclear NF-κB (C) and cytoplasmic NF-κB (D) in chondrocytes exposed to IL-1β alone or to IL-1β and cyclic tensile strain. Values are the mean and SEM from 1 of 3 separate experiments with similar results. E, Phosphorylation of NF-κB p65 (p-p65) at serine 276 (ser 276) and serine 536 in whole cell extracts, determined by Western blot analysis and densitometric analysis of the bands. F, Inducible nitric oxide synthase (iNOS) mRNA expression measured by real-time polymerase chain reaction, using GAPDH mRNA expression as an internal control. Values are the mean and SEM from 3 separate experiments. Color figure can be viewed in the online issue, which is available at http://www.arthritisrheum.org.

Download figure to PowerPoint

We further investigated whether the observed inhibition of NF-κB DNA binding was due to the inhibition of its nuclear translocation or of its cytoplasmic activation (27–29). Supporting the EMSA findings (Figure 1A), IF analysis demonstrated that IL-1β induced rapid translocation of NF-κB within 10 minutes, and nuclear import was sustained during the subsequent 80 minutes (Figure 1B). Semiquantitative fluorescence analysis revealed that IL-1β–induced nuclear translocation of NF-κB was accompanied by a decrease in cytoplasmic NF-κB during the initial 30 minutes; thereafter its levels increased in both nuclear and cytoplasmic compartments, resulting in a total increase in NF-κB by 90 minutes, compared with untreated control cells (Figures 1B–D). Cyclic tensile strain dramatically suppressed IL-1β–induced NF-κB nuclear translocation over the entire 90-minute period, although some NF-κB was detected in the nuclei during the initial 30 minutes of activation (Figure 1C). Furthermore, in cells subjected to cyclic tensile strain and treated with IL-1β, the majority of NF-κB was found in the perinuclear cytoplasm by 60 and 90 minutes, and its total levels were lower than in cells treated with IL-1β alone (Figures 1B and C).

During IL-1β–induced gene activation in chondrocytes, NF-κB p65 phosphorylation at serine 536 and serine 276 is essential for transactivation and DNA binding of NF-κB. Therefore, we investigated whether cyclic tensile strain inhibits phosphorylation of NF-κB p65 to counteract IL-1β–dependent activation. Western blot data were quantified by near-infrared fluorescence, and values for total NF-κB were normalized to β-actin (23). NF-κB phosphorylation at serine 536 and serine 276 residues was absent or minimally detectable in control chondrocytes, and increased significantly at both sites within 10 minutes after stimulation with IL-1β. Cyclic tensile strain failed to inhibit IL-1β–induced NF-κB phosphorylation at serine 536 and serine 276 in chondrocytes from 10 minutes to 90 minutes (Figure 1E).

Furthermore, the lack of inhibition of IL-1β–induced NF-κB phosphorylation by cyclic tensile strain contrasted with cyclic tensile strain–dependent inhibition of type 2 nitric oxide synthase A mRNA expression at 60 and 90 minutes following IL-1β exposure (Figure 1F). These findings demonstrated that cyclic tensile strain–dependent attenuation of NF-κB DNA binding was not mediated by direct inhibition of NF-κB phosphorylation, but likely via inhibition of an event upstream of its phosphorylation, that ultimately blocked NF-κB nuclear translocation.

Regulation of activation and transcription of the IκBβ gene by cyclic tensile strain.

Because cyclic tensile strain did not regulate NF-κB activation at the phosphorylation level, we next examined whether cyclic tensile strain inhibited IκBβ degradation, thus inhibiting the release of inactivated NF-κB from IκBβ–NF-κB complexes in the cytoplasm. The effects of cyclic tensile strain on IκBβ protein levels were analyzed by Western blot analysis. As expected, total IκBβ protein levels decreased in cells during 10–90 minutes of IL-1β treatment (Figures 2A–C). In contrast, IL-1β–induced loss of IκBβ was markedly inhibited in the presence of cyclic tensile strain at 10, 30, and 60 minutes (Figure 2A).

thumbnail image

Figure 2. Inhibition of interleukin-1β (IL-1β)–induced degradation and synthesis of IκBβ by cyclic tensile strain (CTS). Chondrocytes were treated as described in Materials and Methods. A, Inhibition of IL-1β–induced IκBβ degradation by cyclic tensile strain, determined by Western blot analysis. The cytoplasmic extracts of articular chondrocytes were probed with rabbit anti-IκBβ, and the membrane was stripped and reprobed for β-actin as an internal control. B, Results of immunofluorescence analysis, demonstrating inhibition of IκBβ degradation by cyclic tensile strain. Arrows show IκBβ. In A and B, results are representative of 1 of 3 separate experiments with similar results. C, Semiquantitative analysis of the inhibition of IκBβ degradation by cyclic tensile strain. Values are the mean and SEM from 1 of 3 separate experiments with similar results. D, Expression of IκBβ mRNA in chondrocytes over a period of 10–90 minutes, determined by real-time polymerase chain reaction. Values are the mean and SEM from 3 separate experiments. ∗ = P ≤ 0.05 versus cells treated with IL-1β alone. Color figure can be viewed in the online issue, which is available at http://www.arthritisrheum.org.

Download figure to PowerPoint

Verification of these findings by IF revealed that IL-1β treatment resulted in a significant reduction in IκBβ levels between 10 to 90 minutes (Figure 2B, parts e–h and Figure 2C). However, cyclic tensile strain abrogated IL-1β–induced loss of cytoplasmic IκBβ, as evidenced by the presence of IκBβ in the cytoplasm at all time points tested (Figure 2B, parts i–l and Figure 2C). Additionally, the semiquantitative analysis of total IF in chondrocytes revealed a significant reduction in the levels of IκBβ at 60 and 90 minutes in chondrocytes exposed to IL-1β in the presence and absence of cyclic tensile strain (Figures 2B and C). Subsequently, we determined IκBβ mRNA expression in response to IL-1β and cyclic tensile strain. IκBβ mRNA was expressed at low basal levels in control cells and in cells subjected to cyclic tensile strain (Figure 2D). Surprisingly, chondrocytes treated with IL-1β exhibited significant up-regulation of IκBβ mRNA expression, which was approximately 60-fold higher than that in untreated controls at 60 minutes (Figure 2D).

As compared with cells treated with IL-1β alone, simultaneous exposure of cells to IL-1β and cyclic tensile strain resulted in a 61% and 87% decrease in IκBβ mRNA expression at 60 and 90 minutes, respectively. However, it is important to note that, while cyclic tensile strain inhibited IL-1β–induced IκBβ mRNA levels, these levels were not lower than those found in untreated control cells. Thus, cells treated with IL-1β and subjected to cyclic tensile strain exhibited a decrease in IL-1β–induced IκBβ gene transcription, but showed an increase in IκBβ levels in the cytoplasm. Since it is well documented that IL-1β induces IκBβ phosphorylation and subsequent degradation, it is likely that a reduction in IκBβ levels reflects its synthesis as well as its degradation.

Prevention of IκBα degradation and gene transcription by cyclic tensile strain.

IκBα contains both nuclear localization and nuclear export signals, and thus plays a major role in the regulation of NF-κB nuclear translocation as well as its nuclear export (30, 31). We next examined whether cyclic tensile strain represses the degradation and synthesis of IκBα to inhibit NF-κB nuclear translocation. Protein extracts from chondrocytes were examined for the presence of IκBα, by Western blot analysis (Figure 3A). Treatment of chondrocytes with IL-1β resulted in a rapid loss of constitutive levels of IκBα, within 10–30 minutes, followed by reexpression of IκBα by 60 and 90 minutes. The exposure of cells to both cyclic tensile strain and IL-1β resulted in inhibition of the loss of IκBα at 10 and 30 minutes; however, IκBα levels decreased at the later time points of 60 and 90 minutes. In control chondrocytes and those exposed to cyclic tensile strain alone, the levels of IκBα in the cytoplasm or nucleus remained unchanged (Figure 3B).

thumbnail image

Figure 3. Down-regulation of interleukin-1β (IL-1β)–induced degradation and synthesis of IκBα by cyclic tensile strain (CTS). Articular chondrocytes were either untreated or treated with IL-1β in the presence or absence of cyclic tensile strain for the indicated time periods. A, Degradation and synthesis of IκBα, determined by Western blot analysis. The membrane was probed with anti-IκBα antibody, stripped, and reprobed for β-actin as an internal control. B, Localization of cytoplasmic IκBα (white arrows) and nuclear IκBα (green arrows) and IL-1β–induced degradation of IκBα, determined by immunofluorescence staining using rabbit anti-IκBα IgG. Bar = 20 μm. C and D, Semiquantitative analysis of cytoplasmic IκBα (C) and nuclear IκBα (D) in chondrocytes exposed to IL-1β alone or to IL-1β and cyclic tensile strain. In A–D, results are representative of 1 of 3 separate experiments with similar results. E, Expression of IκBα mRNA in chondrocytes over a period of 10–90 minutes, determined by real-time polymerase chain reaction. Values are the mean and SEM from 3 separate experiments. ∗ = P ≤ 0.05. Color figure can be viewed in the online issue, which is available at http://www.arthritisrheum.org.

Download figure to PowerPoint

In addition to sequestering NF-κB in the cytoplasm, IκBα acts as a carrier to shuttle NF-κB from the nucleus to the cytoplasm. Therefore, we investigated the dynamic distribution of IκBα in response to IL-1β and/or cyclic tensile strain, by IF followed by densitometric analysis of nuclear and cytoplasmic IκBα, using Zeiss AxioVision software. In these experiments, the relative presence of IκBα was compared with that in untreated control cells. As expected, in chondrocytes treated with IL-1β, >93% of cytoplasmic IκBα degraded within 10 minutes, but IκBα levels gradually recovered to nearly 70% of control levels by 90 minutes (Figures 3B and C). Estimation of nuclear IκBα levels revealed that IκBα levels in cells treated with IL-1β alone were lower than those in control cells within 10 minutes, and levels of IκBα in the nuclei of IL-1β–treated cells remained similar at all time points observed (Figures 3B and D).

However, in chondrocytes exposed to cyclic tensile strain, IL-1β–induced reduction in cytoplasmic IκBα levels was rapidly inhibited (Figures 3B and C). Semiquantitative analysis of IF images confirmed a >70% inhibition of IL-1β–induced cytoplasmic IκBα degradation by cyclic tensile strain, during the first 10–30 minutes. Additionally, cyclic tensile strain up-regulated nuclear transport of IκBα, as was evident by a >5-fold increase in intranuclear IκBα within 10 minutes, which remained high until 30 minutes. However, in cells exposed to cyclic tensile strain and IL-1β, a 65% and 85% reduction in total cytoplasmic IκBα was observed by 60 minutes and 90 minutes, respectively (Figures 3B and C). This was followed by a subsequent increase in nuclear IκBα at 90 minutes (Figures 3B–D).

Since the experiments described above showed that IκBα levels were significantly reduced in cells exposed to cyclic tensile strain and IL-1β, we next investigated whether cyclic tensile strain down-regulates IκBα transcription. As shown in Figure 3E, Iβ induced a 17-fold, 59-fold, and 31-fold increase in IκBα mRNA expression at 30, 60, and 90 minutes, respectively. Simultaneous exposure of cells to cyclic tensile strain significantly blocked IL-1β–induced IκBα mRNA expression by 82%, 62%, and 66% at 30, 60, and 90 minutes, respectively (P ≤ 0.05). These findings provide evidence that IκBα is not degraded at 60 minutes or at 90 minutes, but rather the observed decrease in the level of IκBα is due to a reduction in its synthesis (Figures 3B and D). Control cells and cells subjected to cyclic tensile strain alone expressed minimal levels of IκBα mRNA (Figure 3E).

Down-regulation of IL-1β–induced IKK activity by cyclic tensile strain.

IκB phosphorylation, ubiquitination, and eventual proteosomal degradation are essential for NF-κB nuclear translocation (32). Since cyclic tensile strain inhibited IL-1β–induced cytoplasmic IκB degradation, we next tested whether cyclic tensile strain prevents IL-1β–induced degradation of IκB via inhibition of IKK activity. The extent of phosphorylation of GST-IκBα substrate was examined by incubation with cytosolic IKK complexes from chondrocytes exposed to both cyclic tensile strain and IL-1β or to IL-1β alone. IKK complexes immunoprecipitated with anti-IKKγ IgG were incubated with equal amounts of GST-IκBα fusion protein in the presence of ATP.

As shown in Figure 4A, control chondrocytes exhibited minimal kinase activity. Stimulation of chondrocytes with IL-1β resulted in a significant increase in GST-IκBα phosphorylation within 10 minutes, which was sustained until 90 minutes. More importantly, when cells were exposed to both cyclic tensile strain and IL-1β, cyclic tensile strain inhibited IKK-dependent IκBα phosphorylation by 76%, 66%, 60%, and 93% at 10, 30, 60, and 90 minutes, respectively (Figure 4A). Taken together, these data indicate that IKK may be the intracellular target of cyclic tensile strain, and suppression of its kinase activity causes suppression of IκBα phosphorylation and thereby NF-κB nuclear translocation.

thumbnail image

Figure 4. Suppression of interleukin-1β (IL-1β)–induced IKK activity by cyclic tensile strain (CTS). Chondrocytes were treated with IL-1β in the presence or absence of cyclic tensile strain for the indicated time periods. A, Phosphorylation of IκB substrate, determined by Western blot analysis. IKK complexes immunoprecipitated from cell lysates with anti–IKKγ/NF-κB–essential modulator antibodies were incubated with glutathione S-transferase (GST)–IκB and ATP, as described in Materials and Methods. Phosphorylation of IκB substrate was then analyzed using p-IκBα mouse monoclonal antibody and N-terminal anti-IκB–GST peptide IgG as IκB loading control. The total amount of IKK immunoprecipitated from chondrocytes was assessed by probing the blots with rabbit anti-IKKγ IgG. Secondary antibodies labeled with IR-dye 680 or 800 were used to obtain densitometric measurements of blots with the Odyssey Infrared Imaging System. Results are representative of 1 of 4 separate experiments with similar results. The y-axis shows the fold increase over untreated control cells. B, Inhibition of expression of mRNA for NF-κB–regulated genes in the presence of IL-1β. Cells treated with IL-1β in the presence or absence of cyclic tensile strain were examined for expression of mRNA for tumor necrosis factor receptor–associated factor 1 (TRAF1) and TRAF2, by real-time polymerase chain reaction. Values are the mean and SEM from 2 separate experiments performed in triplicate. ∗ = P < 0.05. Color figure can be viewed in the online issue, which is available at http://www.arthritisrheum.org.

Download figure to PowerPoint

We next confirmed the effect of cyclic tensile strain on the expression of mRNA for NF-κB–regulated genes involved in inflammatory responses, such as IL-1RI, IL-1RII, TRAF1, and TRAF2. Consistent with the inhibition of IKK-dependent IκBα activation, we observed that cyclic tensile strain significantly inhibited transcriptional activation of TRAF1 and TRAF2 (Figure 4B) as well as the rest of these genes. Cyclic tensile strain alone did not influence transcriptional activation of these genes, confirming that it acts in an IL-1β–dependent manner (18).

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

In this study, we demonstrated that biomechanical signals at appropriate magnitudes attenuate IKK activity that could suppress the cytokine-mediated NF-κB signaling cascade. Previous studies have demonstrated inhibition of IL-1β and TNFα-induced proinflammatory gene induction through exposure of chondrocytes to low physiologic levels of tensile or compressive forces (1, 17, 33). In the present study, we examined the mechanisms of the antiinflammatory effects of tensile forces, because under physiologic loading cartilage undergoes both tension and compression, due to its intrinsic viscoelasticity (22, 34).

The signals generated by cyclic tensile strain are perceived by cells via cell surface molecules such as β integrins and FAKs (35–37). However, cyclic tensile strain does not down-regulate IL-1β receptors during the initial hour of its activity (18). Rather, cyclic tensile strain acts by modulating those signal transduction pathways that ultimately result in the induction of proinflammatory gene transcription. Many of the downstream target genes of NF-κB function in the initiation and subsequent amplification of the NF-κB signaling cascade (38). Thus, the NF-κB family of transcription factors forms a particularly important target for biomechanical signals to mediate its antiinflammatory actions (Figure 5).

thumbnail image

Figure 5. Schematic representation of the mechanisms of intracellular action of cyclic tensile strain (CTS). Cyclic tensile strain suppresses interleukin-1β (IL-1β)–induced proinflammatory gene induction by intercepting salient steps in the NF-κB signaling cascade to inhibit transcription activity. Cyclic tensile strain suppresses IL-1β–induced IKK activation, and thus phosphorylation and proteosomal degradation of IκBα and IκBβ. This leads to the inhibition of nuclear translocation of NF-κB. During the initial stages of IL-1β–mediated activation of cells, cyclic tensile strain up-regulates IκBα nuclear translocation to prevent NF-κB binding to DNA and facilitate export of nuclear NF-κB, which may enter the nucleus. Cyclic tensile strain represses IL-1β–induced IκBα and IκBβ mRNA expression. Collectively, these actions of cyclic tensile strain inhibit proinflammatory gene induction as well as expression of multiple molecules involved in the regulation of the NF-κB signaling cascade to suppress IL-1β–induced inflammation. IL-1R = IL-1 receptor; TRAF1 = tumor necrosis factor receptor–associated factor 1; iNOS = inducible nitric oxide synthase; COX-2 = cyclooxygenase 2; MMPs = matrix metalloproteinases; TNF = tumor necrosis factor. Color figure can be viewed in the online issue, which is available at http://www.arthritisrheum.org.

Download figure to PowerPoint

Because of its dynamic control of many critical genes, the transcription activity of NF-κB is regulated at multiple levels, ranging from subcellular localization to posttranslational modifications, including phosphorylation, acetylation, and ubiquitination. In the present study, EMSA demonstrated that biomechanical signals generated by cyclic tensile strain inhibited IL-1β–induced binding of NF-κB to its consensus sequences. IF analysis confirmed that cyclic tensile strain abrogates IL-1β–induced nuclear translocation of NF-κB. Following activation of cells by IL-1β, NF-κB rapidly translocates to the nucleus, is retransported to the cytoplasm, and is resynthesized, resulting in a net increase in the total level of NF-κB in cells (11).

Cyclic tensile strain inhibited IL-1β–induced increases in cellular NF-κB levels. Since the NF-κB promoter region does not have NF-κB binding sites, it is not clear how inhibition of NF-κB nuclear translocation by cyclic tensile strain results in the down-regulation of NF-κB. The 2 major regulatory IL-1β–inducible phosphorylation sites of NF-κB p65 are at the serine 276 and serine 536 residues. Phosphorylation of serine 276 greatly increases NF-κB transactivation and promotes DNA binding and interactions with CBP/p300, whereas phosphorylation of serine 536 increases the transcriptional efficacy of NF-κB (27, 39, 40). Our experimental results suggest that, although cyclic tensile strain strongly modulates NF-κB signaling, it does not regulate the IL-1β–induced phosphorylation of NF-κB p65 at either serine 276 or serine 536.

At the intracellular localization level, both IκBα and IκBβ act as natural inhibitors of NF-κB activation by sequestering NF-κB in the cytoplasm. After exposure to IL-1β, both IκBα and IκBβ are phosphorylated, ubiquitinated, and targeted for proteosomal degradation, consequently allowing import of free cytoplasmic NF-κB into the nucleus. Characterization of the actions of cyclic tensile strain showed a consistent inhibition of IL-1β–induced degradation of both IκBα and IκBβ. Further experiments revealed the important finding that cyclic tensile strain abrogates the massive IL-1β–induced increase in IκBα and Iκββ gene expression to control levels.

While IκBα expression is controlled by NF-κB transcription activity, IκBβ has not been found to be under the transcriptional control of NF-κB (41). Since the pathways that culminate in IκBβ resynthesis after its degradation are incompletely characterized, the main conclusion that can be drawn from these findings is that cyclic tensile strain abrogates IL-1β–dependent IκBα and IκBβ degradation and synthesis, which inhibits NF-κB transcription activity. Moreover, these findings suggest that the signals generated by cyclic tensile strain likely interact with proteins other than those controlled by NF-κB to regulate proinflammatory gene induction.

We have observed that IL-1β induces a significant increase in IκBα and IκBβ mRNA expression; however, measurement of total proteins did not reveal a parallel increase in levels of IκBα and IκBβ proteins. This may be due to the simultaneous synthesis and degradation of both of these proteins in the presence of IL-1β (42). Furthermore, while cyclic tensile strain inhibited IL-1β–induced IκBα and IκBβ degradation during the first 30 minutes of exposure, the total levels of IκBα and IκBβ were found to have decreased at 60 and 90 minutes. This reduction in the levels of IκBα and IκBβ proteins may be due to cyclic tensile strain–mediated abrogation of IL-1β–induced IκBα and IκBβ mRNA expression, and to the constitutive loss of IκBα and IκBβ due to a relatively short half-life of 30–40 minutes in cells (42, 43). Thus, by inhibiting IκBα mRNA expression, cyclic tensile strain may reduce the synthesis of IκBα, while the reduction of the level of IκBα due to its short half-life in cells may lead to a total reduction of IκBα levels in cells treated with IL-1β and cyclic tensile strain.

More importantly, in the present study the levels of IκBα or IκBβ mRNA were not completely suppressed by cyclic tensile strain, as evidenced by the presence of IκBα and IκBβ in the cytoplasm at all time points tested (Figures 2B and 3B). In cells treated with IL-1β and subjected to cyclic tensile strain, the presence of NF-κB in the cytoplasm and its absence in the nuclei (Figure 1B) further confirmed that IκBα and IκBβ played an active role in sequestering NF-κB in cells exposed to cyclic tensile strain. Thus, suppression of IL-β–induced IκBα and IκBβ mRNA expression, as well as inhibition of nuclear translocation of NF-κB, may be the key regulatory actions involved in the suppression of proinflammatory gene induction by cyclic tensile strain.

IL-1β–induced activation of NF-κB is balanced by the positive transcriptional regulation of the IκBα gene under the direct control of NF-κB proteins. Newly synthesized IκBα down-regulates NF-κB activity by interacting with transcriptionally active NF-κB in the nucleus and in the cytoplasmic compartments of cells (44). The results presented here suggest that, in addition to inhibiting IκBα degradation, signals generated by cyclic tensile strain regulate the trafficking and/or interactions of IκBα with NF-κB to down-regulate NF-κB binding to DNA. For example, we have observed a marked nuclear translocation of IκBα at 10 and 30 minutes following the application of cyclic tensile strain, and then again at 90 minutes.

This rapid entry of IκBα may be important in inhibiting the binding of any residual nuclear NF-κB to DNA, and thus terminating its transcriptional activation. Additionally, simultaneous migration of IκBα along with NF-κB to the nucleus and cytoplasm, demonstrated by IF analysis, suggests that nuclear IκBα may also be involved in facilitating the export of nuclear NF-κB–IκBα complexes to the cytoplasm (45). Furthermore, EMSA demonstrated that, even though some NF-κB translocates to the nucleus while costimulated with IL-1β and cyclic tensile strain (Figure 1B), it fails to productively bind to its consensus sequences (Figure 1A), suggesting that IκBα trafficking may be another mechanism by which cyclic tensile strain inhibits NF-κB activity.

Because the IκBα promoter is under the control of NF-κB, it is not surprising that inhibition of NF-κB nuclear translocation leads to suppression of IκBα mRNA expression and synthesis. Thus, inhibition of nuclear NF-κB binding to DNA by IκBα, and inhibition of IκBα degradation and its synthesis, reflect collective mechanisms by which cyclic tensile strain may attenuate IL-1β–induced NF-κB transcription activity.

IL-1β–induced regulation of IκB is a tightly controlled event that is regulated by phosphorylation by IKK (46). In the canonical pathway, IKKβ is essential for the phosphorylation of IκBα and IκBβ (47). Upon phosphorylation, IκBα and IκBβ are ubiquitinated and degraded to allow the nuclear translocation of NF-κB. In the in vitro assay system used in the present study, IL-1β induced a rapid and sustained phosphorylation of GST-IκBα proteins, reflecting the presence of activated IKK in chondrocytes. In contrast, cyclic tensile strain markedly abrogated IL-1β–dependent IKK activation, as reflected by the drastic reduction in the phosphorylation of exogenous IκBα in chondrocytes treated with IL-1β and subjected to cyclic tensile strain.

Recently, NF-κB regulation has been shown to be mediated by activation of IKK in a transforming growth factor β–activated kinase 1 (TAK-1)– or MEKK-3–dependent manner (48). TAK-1 activates IKKβ and phosphorylates IκBα to promote their degradation. The MEKK-3–dependent pathway phosphorylates IKKα, which results in NF-κB activation without degradation of IKKα (48). Since IL-1β–induced IκBα phosphorylation results in IκBα degradation, and cyclic tensile strain intercepts IL-1β–induced IκBα phosphorylation and degradation, it is likely that cyclic tensile strain actions involve IKK activation that is mediated by TAK-1.

NF-κB directly regulates several genes that are essential for the regulation of its own signaling cascade. By repressing IKK activity, cyclic tensile strain inhibits transcriptional up-regulation of such genes as TRAF1 and TRAF2, which are required for IKK activation. Thus, by suppressing the expression of these genes, cyclic tensile strain causes a multistep inhibition of IL-1β activation in chondrocytes, which may help in inhibiting proinflammatory responses in a sustained manner.

In conclusion, our findings represent the first demonstration that biomechanical signals generated by appropriate magnitudes of cyclic tensile strain attenuate IL-1β–dependent activation of IKK to inhibit NF-κB transcription activity. These signals act at multiple steps within the NF-κB signaling cascade to inhibit the transcription activity of NF-κB itself by preventing its nuclear import, as well as inhibiting the activation and gene expression of the NF-κB inhibitors IκBα and IκBβ (Figure 5). Cyclic tensile strain–enforced dynamic distribution of NF-κB–IκBα complexes may represent another level of regulation, which could inhibit the activity of NF-κB and subsequent downstream signaling events involved in proinflammatory gene induction.

The NF-κB signal transduction pathway has been exploited in many pharmacologic agents to reduce inflammation of the joints. For example, acetylsalicylic acid inhibits synthesis of proinflammatory mediators by inhibiting IKKβ (49). Additionally, several studies have shown that the application of inhibitors of IKK or NF-κB may be efficacious for the treatment of arthritic joints, suggesting that inhibition of this pathway is an attractive therapeutic target for joint inflammation (14, 21, 50, 51). However, these approaches may provide only temporary relief, and are limited due to the transient nature of their effects. Our findings demonstrate that the effects of appropriate biomechanical signals are more substantial and generate a sustained inhibition of proinflammatory gene induction (1).

Thus, the present findings provide insight into how biomechanical signals regulate and reduce inflammation. Further studies may assist in targeting specific treatment strategies by the use of appropriate exercise to enhance the ability of chondrocytes to curb inflammation in diseases affecting the joint.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Dr. Agarwal had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study design. Dossumbekova, Agarwal, He.

Acquisition of data. Dossumbekova, Anghelina, Madhavan, He.

Analysis and interpretation of data. Dossumbekova, Anghelina, He, Quan, Knobloch.

Manuscript preparation. Dossumbekova, Madhavan, He, Knobloch, Agarwal.

Statistical analysis. Dossumbekova, He.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES