IL-1β-Induced Mesenchymal Stem Cell Migration Involves MLCK Activation via PKC Signaling

Cell Culture

With the mothers’ consent, human umbilical cords were obtained from full-term births, and MSCs were isolated from umbilical cord Wharton's jelly as described previously (56). MSCs were maintained in low-serum medium consisting of 56% low-glucose Dulbecco's modified Eagle's medium (DMEM; Invitrogen, Carlsbad, CA, USA), 37% MCBD 201 (Sigma-Aldrich, St. Louis, MO, USA), 2% fetal bovine serum (Thermo Scientific, Waltham, MA, USA), 0.5 mg/ml of AlbuMAX® I (Invitrogen), 1× insulin–transferrin–selenium (Invitrogen), 1× antibiotic antimycotic solution (Thermo Scientific), 10 nM dexamethasone (Sigma-Aldrich), 50 nM L-ascorbic acid 2-phosphate (Sigma-Aldrich), 10 ng/ml of epidermal growth factor (PeproTech, Rocky Hill, NJ, USA), and 1 ng/ml of platelet-derived growth factor-BB (PeproTech) (57) and maintained at 37°C in 5% CO₂.

IL-1β and Inhibitor Treatment

All inhibitors were dissolved as a stock solution in DMSO, and DMSO was used as the control; the final DMSO concentration in the medium was 40 μM. Cells were starved in serum-free DMEM containing 0.1% bovine serum albumin (BSA) (starvation medium) for 16–24 h, then were treated for the indicated time with 100 ng/ml of recombinant human IL-1β (specific activity 1 × 10⁹ units/mg; Peprotech) in starvation medium. In inhibitor studies, calphostin C (Merck, Darmstadt, Germany) at the indicated concentrations was added to starvation medium and activated by light for 30 min; then 100 ng/ml of IL-1β was added. Go6976, BAPTA-AM, a specific PKCβ inhibitor, U0126, PD98059, and ML-7 (all from Merck), rottlerin (Enzo Life Sciences, Farmingdale, NY, USA), sanguinarine (Sigma), or IL-1R (Peprotech) was added to cells in starvation medium 2 h before IL-1β stimulation.

NOD Mice, Immunohistofluorescence, and IL-1β ELISA

Three-week-old female nonobese diabetic (NOD/ ShiLtJNarl) mice were purchased from the National Laboratory Animal Center, Taiwan. These mice show a spontaneous increase in blood sugar levels; those with levels of 540–660 mg/dl were used as the hyperglycemic group, while those with levels lower than 150 mg/dl were used as the normal group.

Human MSCs were incubated with 5 μg/ml of BrdU (Thermo Scientific) in low-serum medium for 2 days, then were transferred to Dulbecco's phosphate-buffered solution (Caisson Labs, North Logan, UT, USA), and 2 × 10⁶ cells were injected into the tail vein of six NOD mice with blood sugar levels less than 150 mg/dl (controls) and six with blood sugar levels of 540–660 mg/dl (hyper glycemic mice). Fourteen days later, the mice were anesthetized with Avertin (250 mg/kg; Sigma-Aldrich) and perfused intracardially with phosphate-buffered saline (PBS) (137 mM NaCl, 2.7 mM KCl, 10 mM Na₂HPO₄, 2 mM KH₂PO₄, pH 7.4; all chemicals used for PBS were purchased from Sigma-Aldrich) followed by 4% paraformaldehyde (Sigma-Aldrich) in PBS. Then the pancreas was embedded in OCT (Sakura Finetek, Torrance, CA, USA). Frozen sections (25 μm) were fixed in cold methanol for 30 min, immersed in 7 mM NaOH (Sigma-Aldrich) for 15 s, and incubated with DyLight-488-conjugated anti-BrdU antibody (Novus Biologicals, Littleton, CO, USA) for 1 h. Then nuclei were counterstained with Hoechst 33258 (Sigma-Aldrich), and images were acquired using a fluorescence microscope (DM-6000B; Leica, Wetzlar, Germany). BrdU⁺ cells in 20 sections from the head region of the pancreas for each mouse were counted using MetaMorph image analysis software (Molecular Devices, Sunnyvale, CA, USA). For ELISA analysis, six normal and six hyperglycemic NOD mice were sacrificed and pancreatic tissue lysed at 4°C in tissue protein extraction reagent (T-PER; Thermo Scientific) containing 10 ml/ml of Halt Protease Inhibitor Cocktail (Thermo Scientific). The lysate was centrifuged at 14,000 × g for 10 min at 4°C and the protein concentration in the supernatant measured using Coomassie Plus (Bradford) Protein Assay reagent (Thermo Scientific) and a Multimode microplate reader (Infinite 200; Tecan Group, Männedorf, Switzerland). IL-1β levels in lysates were measured using mouse IL-1β ELISA kits (Thermo Scientific) following the manufacturer's protocol.

Antibodies

The following antibodies were used: mouse monoclonal antibodies against human p-MEK1/MEK2 Ser218/222 (1:10,000), MLCK (1:2,000), p-PKCα Thr497 (1:1,000), p-IκB-α Ser32 (1:5,000), or NF-κB (1:1,000) (GeneTex, Irvine, CA, USA) and against histone H1 (1:1,000) (Upstate Biotechnology, Lake Placid, NY, USA), rabbit polyclonal antibodies against human p-ERK1/ERK2 Thr185/ 202 (1:10,000) or RAF1 (1:500) (GeneTex) and against human p-PKCδ Thr313 (1:1,000) (Cell Signaling Technology, Danvers, MA, USA), goat polyclonal antibodies against human p-MLCK Ser1760 (1:1,000) (Invitrogen), TRITC-labeled phalloidin (1:200) (Sigma-Aldrich), horseradish peroxidase-conjugated anti-mouse and anti-rabbit immunoglobulin G (IgG) antibodies (1:20,000) (KPL, Gaithersburg, MD, USA), Dylight-488-conjugated anti-rabbit IgG antibodies (1:400) (Jackson ImmunoResearch Laboratories, West Grove, PA, USA), and monoclonal anti-β-actin antibody (1:25,000) (Abcam, Cambridge, MA, USA).

Microarray Analysis

Total RNA isolated from untreated MSCs or MSCs previously incubated for 24 h with 100 ng/ml of IL-1β was hybridized onto a human genome U133 2.0 array (Affymetrix, Santa Clara, CA, USA) with two biological replicates for each treatment, and the cDNA array data were analyzed by the National Yang-Ming University VYM Genome Research Center. The results were expressed as the log value for the expression of the gene in the test sample divided by the expression of the gene in the control sample.

Quantitative Real-Time PCR

After incubation in starvation medium containing 100 ng/ ml of IL-1β for 24 h, MSCs were harvested and total RNA extracted using TriPure isolation reagent (Roche, Mannheim, Germany). cDNA was synthesized using SuperScript® III Reverse Transcriptase (Invitrogen) and Oligo (dT)20 primer (Invitrogen). PCR was carried out on a StepOnePlus” Real-Time PCR System (Applied Biosystems, Waltham, MA, USA) using a KAPA SYBR® FAST ABI Prism® 2× qPCR Master Mix (Kapa Biosystems, Wilmington, MA, USA) according to the manufacturer's protocol. The sequences of the primer sets were 5′-AGGGAGTCCCG CCACTTCCA-3′, MLCK forward primer; 5′-CTTGGCAT CGTCATCCCCGCA-3′, MLCK reverse primer; 5′-GGAG TCAACGGATTTGGTCGTA-3′, GAPDH forward primer; and 5′-GGCAATATCCACTTTACCAGAGT-3′, GAPDH reverse primer.

FRET-Based Kinase Assay

MLCK activity was measured using a modified FRET-based Z'LYTE kinase assay kit using Ser/Thr 13 peptide with a donor fluorophore (coumarin) on one end and an acceptor fluorophore (fluorescein) on the other end as the substrate to be phosphorylated by MLCK (Invitrogen) according to the manufacturer's protocol (25). This assay is based on the transfer of a phosphate group from ATP to the single Ser residue in the peptide that is only phosphorylated by MLCK. The nonphosphorylated peptides are cleaved by the development reagent in the kit. Kinase reaction buffer (50 mM HEPES, 10 mM MgCl₂, 1 mM EGTA, and 0.01% Brij-35), 50 μM ATP (Invitrogen, CA, USA), 2 μM Z'LYTE peptide substrate (Invitrogen), and the test sample were added to each well and the kinase reaction carried out at 37°C for 2 h. The development reagent provided by the manufacturer was then added and the mixture incubated for 1 h to cleave nonphosphorylated Z'LYTE peptide substrate. Then the reaction was stopped using stop reagent and the results read on a Multimode microplate reader (Infinite 200; Tecan, Switzerland) using an excitation wavelength of 400 nm and emission wavelengths of 445 and 520 nm. To establish the minimum and maximum emission ratio values for each assay, 0% phosphorylation controls (i.e., absence of ATP) and 100% phosphorylation controls (result using the standard MLCK sample in the kit) were used. Data were processed as the coumarin emission (445 nm)/fluorescein emission (520 nm) ratio for each well, which was used to calculate the percentage phosphorylation using the following equation:

( 1 − ( E m . R a t i o × F 100 % ) − C 100 % ( C 0 % − C 100 % ) + [ E m . R a t i o × ( F 100 % − F 0 % ) ] × 100

where Em. Ratio is the above emission ratio, C_100% and C_0% are the average coumarin emission for the 100% and 0% phosphorylated control, and F_100% and F_0% are the corresponding average fluorescein emission signals.

Western Blotting and Immunoprecipitation

Cells were lysed for 1 h at 4°C in mammalian protein extraction reagent (M-PER; Thermo Scientific) containing 10 ml/ml of Halt Protease and Phosphatase Inhibitor Cocktail (Thermo Scientific) and the lysates centrifuged at 14,000 × g for 10 min at 4°C. The protein concentration in the supernatant was measured using Coomassie Plus (Bradford) protein assay reagent (Thermo Scientific) and a multimode microplate reader (Infinite 200). Samples (20 μg of protein) were then separated on a 6-10% SDS-PAGE gel and transferred onto polyvinylidene fluoride membranes. All subsequent steps were performed at room temperature. The membranes were incubated for 1 h in blocking buffer [136 mM NaCl, 2.7 mM KCl, 25 mM Tris-base, and 0.05% Tween-20 (TBST) containing 10% fish gelatin blocking buffer (AMRESCO, Solon, OH, USA)]. Then membranes were incubated for 1 h with the primary antibody diluted in TBST, washed with TBST, and incubated for 90 min with horseradish peroxidase-conjugated secondary antibody diluted in TBST. Bound antibody was detected using enhanced chemiluminescence substrate and a luminescence imaging system (LAS-4000; GE, Fairfield, CT, USA).

Immunoprecipitation was performed using a Pierce® Crosslink IP kit (Thermo Scientific) following the manufacturer's protocol. Protein A/G agarose beads (20 μl) were incubated with 2 μg of antibody for 2 h at room temperature. Then the mixture was incubated with lysates overnight at 4°C. After washing with wash buffer (0.025 M Tris, 0.15 M NaCl, 0.001 M EDTA, 1% NP-40, and 5% glycerol), protein bound to the beads was eluted using the elution buffer provided by the manufacturer. The samples were separated by SDS-PAGE and analyzed by Western blotting.

Nuclear Protein Extraction

Nuclear proteins were extracted from cells using a nuclear extraction kit (Merck Millipore, Darmstadt, Germany) following the manufacturer's protocol and examined by Western blotting.

In Vitro Migration Assay

The migration assay was performed in an 8.0-μm pore size chemotaxis chamber (Becton Dickinson Biosciences, Franklin Lakes, NJ, USA) as described previously (3).

Cells were incubated in starvation medium for 16-24 h, then for 2 h with and without inhibitors. The cells were then detached and 25,000 cells loaded into the upper chemotaxis chamber and 100 ng/ml of IL-1β in starvation medium added to the lower chamber. After incubation for 18 h, nonmigrated cells in the upper chamber were removed with a cotton swab, and the cells that had migrated to the lower chamber were fixed and stained with Hoechst 33258 (Sigma-Aldrich). Fluorescence microscopy was used to count the number of migrated cells in five randomly selected fields.

siRNA Transfection

For siRNA-mediated knockdown of MLCK and PKCδ, we used the ON-TARGET plus SMART pool (GE Dharmacon, Fairfield, CT, USA) containing a mixture of four designed MLCK-siRNA oligos (5′-CU AAGACCAUUCGCGAUUU-3′, 5′-GCAAUGAUCUC AGGGCUCA-3′, 5′-GGGAACUGCUCUUUAAUUA-3′, and 5′-GGGAUGACGAUGCCAAGUA-3′) or four designed PKCδ-siRNA oligos (5′-CCAUGUAUCCUGAG UGGAA-3′, 5′-CCAAGGUGUUGAUGUCUGU-3′, 5′-A AAGAACGCUUCAACAUCG-3′, and 5′-CCGCACCG CUUCAAGGUUC-3′). MSCs were transfected with siRNA using DharmaFECT transfection reagent (Dharmacon) following the manufacturer's protocol, and nontargeting siRNA-transfected cells were used as controls.

Wound Healing Assays

Wound healing assays were performed using an ECIS system (model 1600R; Applied BioPhysics, Troy, NY, USA) as described previously (6) or a cell culture insert system (Ibidi, Planegg, Germany).

For the ECIS assay, confluent si-control- or MLCK-siRNA-transfected MSC monolayers on ECIS 8W1E plates were incubated for 24 h in starvation medium with or without IL-1β, then wound formation was induced by subjecting the monolayers to 2,000-μA voltage pulses delivered at 40 kHz for 15 s, leading to the death and detachment of cells. Cells were then assessed by continuous resistance measurements for 12 h to measure wound healing.

Using the cell culture insert system, the insert was placed on the culture surface, generating two cell culture reservoirs separated by a 500-μm-thick wall. MSCs were then cultured in the reservoirs and incubated with or without inhibitors before removing the insert; then the cells were stimulated with IL-1β for ~18 h, and pictures were taken to visualize cell repair and migration at each time point.

Chromatin Immunoprecipitation

MSCs were incubated in starvation medium for 16-24 h, then were left untreated or were incubated for 2 h with 100 ng/ml of IL-1β in starvation medium, then intact chromatin was isolated using EZ-Magna ChIP A/G kits (Merck Millipore) as recommended by the manufacturer. A possible NF-κB binding site on the MLCK promoter was identified using Searching Transcription Factor Binding Sites software (ver. 1.3) and analyzed by PCR using the forward primer 5′-TT CCACCCTCTCTCTCTCCTTTGT-3′ and reverse primer 5′-TGCTGGTGTCTTGATCTTGGACTT-3′.

Cell Immunofluorescence

Cells plated on four-chamber polystyrene glass slides were incubated with or without inhibitor in starvation medium for 16–24 h, then with or without 100 ng/ml of IL-1β for 2 h. They were then fixed for 10 min in HistoChoice® MB Tissue Fixative (AMRESCO), permeabilized for 10 min with 0.2% Triton X-100 in PBS, then blocked for 1 h at room temperature with fish gelatin blocking buffer (AMRESCO). They were then incubated overnight at 4°C with primary antibody in blocking buffer, followed by overnight incubation at 4°C with TFITC-conjugated phalloidin to stain F-actin. After PBS washes, the cells were stained using Hoechst 33258 (Sigma-Aldrich) and mounted with fluorescence mounting medium (Dako, Glostrup, Denmark), and images were acquired using a laser confocal microscope (FV1000; Olympus, Tokyo, Japan).

Statistical Analysis

The quantitative real-time PCR, ELISA, Western blot, and migration assay data are expressed as the mean ± the standard deviation (SD) for at least three experiments. Statistical significance was analyzed by one-way ANOVA followed by post hoc least significant difference analysis on SPSS software versions 13 (IBM, Armonk, NY, USA). A value of p < 0.05 was considered significant.

IL-1β Stimulates MSC Migration via Activation of MLCK

Previous studies have shown that IL-1β, which is produced by macrophages, induces recruitment of lymphocytes and leukocytes to sites of injury (47) and that the pancreas of hyperglycemic NOD mice expresses high amounts of IL-1β (58,42,43). To determine whether IL-1β is involved in MSC homing, we performed an in vivo migration assay in NOD mice, which spontaneously developed autoimmune type I diabetes as a result of IL-1β- mediated pancreatic β-cell destruction (35,46). When mice began to exhibit high blood sugar levels, we measured pancreatic IL-1β levels using an enzyme-linked immunosorbent assay (ELISA) and found that hyperglycemic NOD mice had significantly higher IL-1β levels than normal NOD mice (Fig. 1A). To investigate MSC homing, we incubated human MSCs from Wharton's jelly with 5-bromo-2′-deoxyuridine (BrdU) for 2 days and injected them into the tail vein of the normal or hyperglycemic NOD mice, then sacrificed the mice 14 days later. Examination of frozen pancreatic sections using Dylight488-conjugated anti-BrdU antibodies showed that the pancreas of hyperglycemic mice contained significantly more BrdU⁺ cells than that of normal mice (Fig. 1B, C). In addition, blood sugar levels were significantly lower in the MSC-transplanted hyperglycemic NOD mice than in the untreated hyperglycemic NOD mice (326 ± 36 mg/dl vs. 584 ± 55 mg/dl, p < 0.001). Next, we examined the ability of IL-1β to induce human MSC migration in vitro in a Boyden chemotaxis chamber system and found that treatment for 18 h with 100 ng/ml of IL-1β (the concentration used in all in vitro experiments) significantly enhanced stem cell migration and that this effect was blocked by 2 h pretreatment with the IL-1β receptor antagonist IL-1RA (Fig. 1D). We then investigated the expression of molecules involved in the migration of IL-1β- treated MSCs using cDNA microarray analysis. Compared to MSCs incubated in serum-free medium, MSCs incubated for 24 h with recombinant human IL-1β showed increased levels of mRNAs for the cytoskeleton rearrangement-related proteins MLCK (log value for the IL-1β-treated cells/control cell ratio = 2.5, p < 0.005), NEDD9 (log value = 1.4, p < 0.005), and CDC42 (log value = 0.5, p < 0.005), with MLCK showing the greatest upregulation. Furthermore, quantitative realtime PCR showed that IL-1β treatment increased MLCK mRNA levels in a time-dependent manner (Fig. 1E) and that inhibition of MLCK by 2-h pretreatment of the cells with the MLCK inhibitor ML-7 (5 mM) blocked IL-1β-induced in vitro stem cell migration (Fig. 1F). These results suggest that MSC migration to the pancreas in hyperglycemic NOD mice is promoted by IL-1β.

OPEN IN VIEWER

Figure 1.

IL-1β stimulates MSC migration via MLCK activation. (A) IL-1β levels in the pancreas of hyperglycemic or normal NOD mice measured by ELISA (*p < 0.05). (B) BrdU-labeled MSCs in pancreas sections of normal or hyperglycemic mice were immunostained with Dylight488-conjugated anti-BrdU antibody (green) and Hoechst 33258 (blue) for cell nuclei. Scale bar: 100 μm. (C) Mean number of BrdU⁺ cells in the pancreas of hyperglycemic and normal NOD mice (*p < 0.05). (D) Migration of untreated MSCs or MSCs incubated with or without IL-1RA, then with IL-1β (*p < 0.05 vs. control cells; #p < 0.05 vs. IL-1β-treated cells). (E) MLCK mRNA levels in control MSCs or MSCs incubated with IL-1β expressed as a fold value compared to the untreated cells (*p < 0.05 vs. 0 h). (F) Migration of MSCs incubated with or without ML-7, then for 18 h with or without addition of IL-1β (*p < 0.05 vs. control cells, #p < 0.05 vs. IL-1β-treated cells). (G) MSCs were incubated with or without IL-1RA or ML-7, then for 6 h with or without addition of IL-1β, then MLCK kinase activity was measured using a FRET-based kinase assay (*p < 0.05 vs. control cells, p < 0.05 vs. IL-1β-treated cells). (H) p-MLCK levels in lysates of control MSCs or MSCs incubated with IL-1β (top panel) or control MSCs or MSCs incubated with or without IL-1RA, then with or without addition of IL-1β. The results shown are representative of those obtained in three experiments.

Actin–myosin interactions are necessary for cell migration (55), and cell migration is regulated by activation of MLCK (14). Using an in vitro FRET-based kinase assay, we found that MLCK activity in cells treated for 2 h with IL-1β was more than threefold higher than that in control MSCs and that this effect was significantly inhibited by 2-h pretreatment with 500 ng/ml of IL-1RA or 5 mM ML-7 (Fig. 1G). The IL-1β-induced increase in MLCK kinase activity was not totally inhibited by ML-7 or IL-1RA, and it is possible that the concentrations used were not high enough to completely block MLCK activity. To assess MLCK activation in IL-1β-treated MSCs, cells were incubated with IL-1β for 1, 3, 6, 9, or 24 h prior to Western blotting of cell lysates for phosphorylated MLCK. As shown in Figure 1H, IL-1β treatment increased levels of MLCK phosphorylation in a time-dependent manner, a slight increase above basal levels being seen after 1 h and a much more pronounced increase after 6 to 24 h (top panel). Moreover, 2-h pretreatment of MSCs with IL-1RA completely blocked the effect of 6 h of IL-1β treatment on MLCK phosphorylation (bottom panel). These results show that IL-1β-induced MSC migration depends on MLCK activation.

IL-1β Activates MLCK via PKCα Signaling

PKCs consist of three subfamilies, classical PKCs (α, β, and γ isoforms), novel PKCs (δ, θ, η, and ε isoforms), and atypical PKCs (z and i isoforms). Classical PKCs activate MLCK and are regulated by diacylglycerol and calcium (29). To examine the role of PKC signaling in MLCK activation, MSCs were preincubated for 30 min with the photoactivated PKC inhibitor calphostin C, which inhibits classical and novel PKCs, or for 2 h with Go6976 (inhibits PKCa and PKCb), the calcium chelator BAPTA-AM (inhibits PKCa and PKCb), a specific PKCb inhibitor, or rottlerin (specifically inhibits PKCd at the low concentration used (16), then for 6 h with IL-1β; cell lysates were examined by Western blotting for phosphorylation of MLCK. The results showed that the IL-1β-induced increase in p-MLCK levels was blocked by BAPTA-AM, Go6976, or calphostin C (Fig. 2A, left panel, and B), but was not significantly inhibited by rottlerin or the PKCb inhibitor (Fig. 2A, right panel, and C). Similar results were obtained when MSCs were pretreated with PKC inhibitors, then examined in an in vitro kinase assay (Fig. 2D), demonstrating that PKCa activation is essential for IL-1β-induced MLCK activation. Inhibition of PKCa signaling by pretreatment for 30 min with calphostin C or for 2 h with Go6976 or BAPTA-AM also inhibited MSC migration induced by incubation for 18 h with IL-1β, whereas the specific PKCb inhibitor had no effect (Fig. 2E). When p-PKCa levels in IL-1β-treated cells were measured by Western blotting, we found a time-dependent increase in p-PKCα levels (Fig. 2F, left panel, and G) that was blocked by IL-1RA (Fig. 2F, right panel, and H). Since cell migration involves F-actin, we examined F-actin distribution in IL-1β-stimulated MSCs by immunofluorescence staining with TRITC-labeled phalloidin and found that the localized formation of actin filaments in lamellipodia protrusions seen in Il-1β-treated cells was significantly inhibited by pretreatment with ML-7 (data not shown), showing that inhibition of MLCK activity affected F-actin organization. Together, these findings show that PKCα is involved in IL-1β-stimulated MLCK activation and stem cell migration.

OPEN IN VIEWER

Figure 2.

IL-1β activates MLCK via PKCa. (A–C) MSCs were preincubated with calphostin C (30 nM) or with DMSO, BAPTA-AM (50 μM), Go6976 (15 μM), PKCb inhibitor (100 nM), or rottlerin (15 μM), then with IL-1β, and then p-MLCK levels in cell lysates were measured by Western blotting. The data shown in (A) are from a representative experiment, while those in (B) and (C) are the relative density of the p-MLCK band. (D) MSCs were treated as in (A); then kinase activity in the cell lysates was measured in the FRET-based kinase assay. (E) MSCs were treated as in (A) in a chemotaxis chamber, then in vitro migration was assessed. (F) MSCs were treated with IL-1β for the indicated time (left panel) or were incubated with or without IL-1RA for 2 h followed by incubation with or without IL-1β for 1 min (right panel), then p-PKCa levels in lysates were measured by Western blotting. The data shown in (F) are from a representative experiment, while those in (G) and (H) are the relative density of the p-PKCa band. *p < 0.05 versus control cells; #p < 0.05 versus IL-1β-treated cells.

IL-1β-Induced MLCK Activation and Migration Depend on the PKCα/MEK/ERK Signaling Pathway

Because MEK/ERK signaling is involved in IL-1β-stimulated cell migration (18,33), we examined whether it was also involved in MSC migration by measuring MEK/ERK activation using Western blotting and antibodies specific for p-MEK and p-ERK. Strong activation of MEK or ERK was observed after, respectively, 5 or 10 min of IL-1β treatment, (Fig. 3A, left panel), and these effects were completely inhibited by pretreatment with IL-1RA (right panel), showing that MEK/ERK signaling was activated by IL-1β. As MEK/ERK signaling is also activated by PKC (4,31), we next examined whether PKC acts upstream of MEK/ERK signaling by pretreating the cells for 30 min with calphostin C or for 2 h with Go6976, BAPTA-AM, the specific PKCb inhibitor, or rottlerin before treatment for 5 min (p-MEK generation) or 10 min (p-ERK generation) with IL-1β. As shown in Figure 3B, calphostin C and Go6976 (top panel) and BAPTA/AM (center panel) completely inhibited IL-1β-induced MEK/ ERK activation, whereas the PKCb inhibitor or rottlerin had no effect (bottom panel), showing that IL-1β induces MEK/ERK activation via PKCα.

OPEN IN VIEWER

Figure 3.

MEK and ERK are involved in the IL-1β-induced activation of MLCK. (A) MSCs were treated with IL-1β (left panel) or were incubated with or without IL-1RA for 2 h, then IL-1β was added for 5 min (p-MEK) or 10 min (p-ERK), and p-MEK and p-ERK levels in MSCs lysates were examined by Western blotting. (B) MSCs were pretreated with DMSO or PKC inhibitors as in Figure 2A, then IL-1β was added, and p-MEK (left panels) and p-ERK (right panels) levels were examined by Western blotting. (C) MSCs were incubated with or without calphostin C, then with IL-1β. Cell lysates were then immunoprecipitated with anti-p-PKCα antibody to examine whether Raf-1 was present in the immunoprecipitate. (D) MSCs were pretreated for 2 h with DMSO, U0126, or PD98059, then with or without IL-1β for 6 h; then p-MLCK levels in cell lysates were examined by Western blotting. The Western blot results shown in (A–D) are representative of those obtained in three experiments. (E) MSCs were left untreated or were pretreated with U0126 or PD98059; then all were incubated with IL-1β for 6 h; then kinase activity in the cell lysates was measured in the FRET-based kinase assay. (F) MSCs were treated as in (E) in a Transwell assay, then the number of migrated cells was counted. *p < 0.05 versus control cells; #p < 0.05 versus IL-1β-treated cells.

In the MEK/ERK signaling pathway, Raf phosphorylates and activates MEK, which then phosphorylates and activates ERK (41). To determine whether p-PKCα was bound to Raf in our system, lysates of treated cells were immunoprecipitated with an antibody against p-PKCa; then the presence of Raf in the immunoprecipitate was assessed by Western blotting. As shown in Figure 3C, no binding of PKCα to Raf was seen in control cells but was clearly seen in cells incubated with IL-1β for 10 min, and this effect was blocked by calphostin C pretreatment.

We then examined whether MLCK was activated via the PKCa/MEK/ERK signaling pathway and found that IL-1β-induced MLCK phosphorylation (Fig. 3D) and kinase activity (Fig. 3E) were significantly decreased in cells pretreated for 2 h with the MEK inhibitor U0126 (10 mM) or the ERK inhibitor PD98059 (40 mM). We also found that pretreatment with U0126 or PD98059 blocked IL-1β-induced cell migration (Fig. 3F). These findings show that IL-1β activates the PKCa/MEK/ERK signaling pathway, leading to phosphorylation of MLCK and MSC migration.

IL-1β-Induced NF-κB Activation Involves the PKCδ Signaling Pathway

NF-κB signaling is required for inflammation-induced cancer cell migration and invasion (61), and IL-1β induces migration of human alveolar epithelial cells and alveolar carcinoma cells via NF-κB signaling (7,27). In the NF-κB signaling pathway, active IκB kinase phosphorylates IκB at Ser-32 and Ser-36, triggering ubiquitination and pro-teasomal degradation of IκB and translocation of NF-κB to the nucleus. To determine whether NF-κB signaling was involved in IL-1β-induced signaling in MSCs, we performed Western blotting on lysates of treated MSCs using antibodies specific for p-IκBα and, as shown in Figure 4A, found that IκBα was highly phosphorylated after 60 min of IL-1β treatment (left panel), and that this effect was reduced by 2-h pretreatment with IL-1RA (right panel). Next, we assessed the nuclear translocation of the NF-κB subunit p65 by Western blotting of nuclear proteins using anti-p65 antibody and observed translocation of p65 to the nucleus after 2 or 4 h of IL-1β treatment (Fig. 4B), a finding substantiated by immunofluorescent staining and confocal microscopy (Fig. 4C).

OPEN IN VIEWER

Figure 4.

IL-1β induces NF-κB activation via PKCα. (A) MSCs were treated with IL-1β (left panel) or were incubated with or without IL-1RA, then with or without IL-1β for 1 h; then p-IκBα and total IκBα levels were examined by Western blotting. (B) NF-κB p65 subunit levels in the nuclear and cytosolic fractions of MSCs incubated with or without IL-1β. (C) Immunofluorescence staining of p65 (green) in control MSCs and MSCs treated for 2 h or 4 h with IL-1β; cell nuclei are in blue. Scale bar: 20 μm. (D–G) p-IκBα levels in MSCs incubated with calphostin C or with DMSO, BAPTA-AM, Go6976, PKCβ inhibitor, rottlerin, U0126, or PD98059, then with or without IL-1β. The data shown in (D) are from a representative experiment, while those in (E–G) are the relative density of the p-IκBα band (*p < 0.05 vs. control cells; #p < 0.05 vs. IL-1β-treated cells). (H) MSCs were incubated with calphostin C or for 2 h with DMSO or rottlerin, then with or without IL-1β; then p65 levels in the nuclear fraction were analyzed by Western blotting. (I) Chromatin from MSCs incubated with or without IL-1β was immunoprecipitated using anti-p65 antibody; then MLCK promoter DNA was analyzed by PCR, together with input DNA. (J) p-PKCδ levels were examined in MSCs incubated with IL-1β (top panel) or with or without IL-1RA, then with or without IL-1β (bottom panel). (K) p-PKCδ levels (upper panel) or p-IκBα levels (lower panel) were measured in control-siRNA- and PKCδ-siRNA-transfected MSCs incubated with or without IL-1β either for 1 min (upper panel) or for 1 h (lower panel).

As NF-κB activation is regulated by PKC or MEK/ ERK (7,20,32), we next determined whether PKC or MEK/ERK activated NF-κB signaling in MSCs by pretreating the cells for 30 min with calphostin C or for 2 h with Go6976, BAPTA/AM, the PKCβ inhibitor, rottlerin, U0126, or PD98059 before incubation with IL-1β in the continued presence of the inhibitor. We then measured IκBα phosphorylation in cell lysates by Western blotting and found that IL-1β-induced IκBα phosphorylation was significantly inhibited by calphostin C or rottlerin (Fig. 4D, E), but not by the other inhibitors (Fig. 4D, F, and G). This showed that PKCα, MEK, and ERK are not involved in the activation of NF-κB and suggested that PKCδ is. Calphostin C and rottlerin pretreatment also reduced nuclear translocation of p65 after IL-1β treatment (Fig. 4H), showing that IL-1β-induced p65 translocation to the nucleus involves the PKC signaling pathway. We then examined whether IL-1β-induced MLCK expression was modulated by NF-κB using chromatin immunoprecipitation (ChIP) analysis with an anti-p65 antibody and PCR with primers for the MLCK promoter region and found that IL-1β treatment for 2 h induced binding of p65 to the MLCK promoter region (Fig. 4I). As calphostin C inhibits the classical and novel PKCs (including PKCδ), and because rottlerin at the concentration used is a specific PKCδ inhibitor (16), this suggests that IL-1β-induced NF-κB signaling is activated by PKCδ. As shown in Figure 4J, Western blotting using an antibody specific for p-PKCδ demonstrated that PKCδ was activated after 1 min of IL-1β treatment (upper panel) and that this effect was blocked by 2-h pretreatment with IL-1RA (lower panel).

Although rottlerin can act as an inhibitor of other novel PKC isoforms at higher concentrations (80-100 mM), as mentioned above, at the concentration used (15 mM) in our system, it is thought to act as a specific PKCδ inhibitor (16). However, to confirm that its effect on IL-1β-induced IκBα phosphorylation was due to PKCδ inhibition, we performed a study targeting PKCδ activity via RNA interference. As shown in Figure 4K, when MSCs were transfected with control or PKCδ siRNAs, Western blots showed that PKCδ siRNA inhibited IL-1β-induced PKCδ phosphorylation (upper panel) and IkBα phosphorylation (lower panel). These results show that IL-1β induces PKCδ phosphorylation, thereby activating NF-κB signaling in MSCs.

IL-1β-Induced MLCK Expression Involves PKCδ/NF-κB Signaling

We also investigated whether IL-1β regulated MLCK expression. PCR showed that MLCK mRNA levels were increased after 24 h of IL-1β treatment, and this effect was inhibited by pretreatment with IL-1RA (Fig. 5A), while Western blots showed an increase in MLCK protein levels after 48 h of IL-1β treatment (Fig. 5B). We next examined whether PKC, MEK/ERK, or NF-κB signaling was involved in IL-1β-induced MLCK expression by pretreating the cells for 30 min with calphostin C or for 2 h with Go6976, BAPTA/AM, PKCβ inhibitor, rottlerin, U0126, PD98059, the NF-κB inhibitor sanguinarine, or IL-1RA before treatment with IL-1β and found that the IL-1β-induced increase in MLCK mRNA levels was blocked or markedly inhibited by calphostin C, rottlerin, sanguinarine, or IL-1RA (Fig. 5A). Similar results were obtained for MLCK protein levels (Fig. 5B), showing that IL-1β induces MLCK expression via NF-κB and the PKCδ signaling pathway. We also found that IL-1β-induced MLCK expression was knocked down by PKCδ-siRNA transfection (Fig. 5C), showing that IL-1β induces MLCK expression via PKCδ/NF-κB signaling. These findings demonstrate that IL-1β activates the PKCδ/NF-κB signaling pathway, leading to translocation of NF-κB subunit p65 to the nucleus, where it binds to the MLCK promoter to upregulate MLCK expression.

OPEN IN VIEWER

Figure 5.

IL-1β induces MLCK mRNA expression via the PKCd/NF-κBsignaling pathway. (A) MSCs were incubated with calphostin C (Cal) or with DMSO, BAPTA-AM (BA), Go6976 (Go), rottlerin (Ro), PKCb inhibitor (b in), U0126 (U0), PD98059 (PD), IL-RA (RA), or sanguinarine (Sang; 1 mM), then with IL-1β for 24 h. Total RNA was then extracted and reverse transcribed and MLCK mRNA levels quantified by real-time PCR (*p < 0.05 vs. control cells, #p < 0.05 vs. IL-1β-treated cells). (B) MSCs were pretreated as in (A), then incubated for 48 h with or without IL-1β in the continued presence of the inhibitor, when MLCK protein levels were measured by Western blotting. (C) Control- or PKCd-siRNA-transfected MSCs were incubated with or without IL-1β for 48 h; then MLCK levels were measured by Western blotting.

Knockdown of MLCK Expression Suppresses IL-1β-Induced Cell Migration and Wound Healing

After establishing that IL-1β-induced MLCK expression is mediated by the PKCδ/NF-κB signaling cascade, we examined whether MLCK expression was required for MSC migration. PCR and Western blots showed that MLCK-siRNA transfection of MSCs resulted in knockdown of MLCK mRNA levels (Fig. 6A) and protein levels (Fig. 6B) and suppressed IL1-β-induced cell migration (Fig. 6C).

OPEN IN VIEWER

Figure 6.

IL-1β-induced MLCK expression in MSCs is required for increased cell migration. (A) Control- and si-MLCK-transfected MSCs were incubated with or without IL-1β for 24 h, then total RNA was extracted and reverse transcribed and MLCK mRNA quantified by real-time PCR. Values are normalized to GADPH expression (*p < 0.05 compared to untreated si-control-transfected cells; #p < 0.05 compared to IL-1β-treated si-control). (B) Control- or MLCK-siRNA-transfected MSCs were incubated with or without IL-1β for 48 h, then MLCK in cell lysates was measured by Western blotting; the results shown are representative of those obtained in three experiments. (C) Control- and MLCK-siRNA-transfected MSCs were incubated with or without IL-1β for 18 h in a chemotaxis chamber, then cells that had migrated though the membrane filter to the lower chamber were counted in five randomly selected fields (*p < 0.05 vs. control cells; #p < 0.05 vs. IL-1β-treated cells). (D) Control- or MLCK-siRNA-transfected MSCs were seeded in ECIS electrode wells and grown to confluence, then were incubated with or without IL-1β for 24 h. They then received an elevated voltage pulse to induce a wound, and the resistance was recorded.

To assess wound healing, we used continuous electric cell-substrate impedance sensing (ECIS) (45). Control-siRNA- or MLCK-siRNA-transfected MSCs cultured on ECIS 8W1E plates were left untreated or were treated with IL-1β for 24 h before being subjected to a 2,000-μA voltage pulse delivered at 40,000 Hz for 15 s followed by measurement of the resistance for 12 h. The high electrical current completely killed cells attached to the microelectrode. Afterward, the surrounding viable cells migrate into the wounded electrodes, and cells act as insulators leading to an increase in resistance. The results showed that the resistance of wounded IL-1β-treated control-siRNA-transfected MSCs increased more rapidly than that of the same cells without IL-1β treatment (Fig. 6D), whereas the resistance of wounded MLCK-siRNA-transfected MSCs remained low regardless of whether IL-1β was present. Using the Ibidi wound healing culture-insert system, we also found that pretreatment of MSCs with 5 mM ML-7 for 2 h or with 30 nM calphostin C for 30 min decreased IL-1β-induced migration, whereas 2-h pretreatment with the specific PKCβ inhibitor (100 nM) had no effect (data not shown). Again this showed that IL-1β-induced MSC migration involves MLCK activity and the PKC pathway. These results show that a reduction in MLCK expression by siRNA or inhibition of MLCK activity blocks the migration of MSCs.