Mechanical strain and fluid movement both activate extracellular regulated kinase (ERK) in osteoblast-like cells but via different signaling pathways

Bone Vol. 31, No. 1 July 2002:186 –194

Mechanical Strain and Fluid Movement Both Activate Extracellular Regulated Kinase (ERK) in Osteoblast-like Cells but via Different Signaling Pathways H. L. JESSOP,1 S. C. F. RAWLINSON,1 A. A. PITSILLIDES,1 and L. E. LANYON1 1

Department of Veterinary Basic Sciences, The Royal Veterinary College, London, UK

Key Words: Mechanical strain; Osteoblast; Extracellular regulated kinase (ERK); Mechanotransduction; Cell signaling.

Extracellular regulated kinases (ERKs)-1 and -2 are members of the MAPK family of protein kinases involved in the proliferation, differentiation, and apoptosis of bone cells. We have shown previously that ROS 17/2.8 cells show increased activation of ERK-1 or -2, which is sustained for 24 h, when the strips onto which they are seeded are subjected to a 10 min period of cyclic four point bending that produces physiological levels of mechanical strain along with associated fluid movement of the medium. Movement of the strips through the medium without bending causes fluid movement without strain. This also increases ERK-1/2 activation, but in a biphasic manner over the same time period. Our present study investigates the role of components of signaling pathways in the activation of ERK-1/2 in ROS 17/2.8 cells in response to these stimuli. Using a range of inhibitors we show specific differences by which ERK-1 and ERK-2 are activated in response to fluid movement alone, compared with those induced in response to strain plus its associated fluid movement. ERK-1 activation induced by fluid movement was markedly reduced by nifedipine, and therefore appears to involve L-type calcium channels, but was unaffected by either L-NAME or indomethacin. This suggests independence from prostacyclin (PGI2) and nitric oxide (NO) production. In contrast, ERK-1 activation induced by application of strain (and its associated fluid disturbance) was abrogated by TMB-8 hydrochloride, LNAME, and indomethacin. This suggests that strain-induced ERK-1 activation is dependent upon calcium mobilization from intracellular stores and production of NO and PGI2. ERK-2 activation appears to be mediated by a separate mechanism in these cells. Its activation by fluid movement alone involved both PGI2 and NO production, but its activation by strain was not affected by any of the inhibitors used. The G protein inhibitor, pertussis toxin, did not cause a reduction in the activation of ERK-1 or -2 in response to either stimulus. These results are consistent with earlier observations of ERK activation in bone cells in response to both strain (with fluid movement) and fluid movement alone, and further demonstrate that these phenomena stimulate distinct signaling pathways. (Bone 31:186 –194; 2002) © 2002 by Elsevier Science Inc. All rights reserved.

Introduction Extracellular regulated kinases (ERKs)-1 and -23 are members of the MAPK family of protein kinases. This family is conserved among different species,3 and has been demonstrated to be involved in the regulation of cellular growth, differentiation, and apoptosis in a variety of cell types.9,11,19,30,49,53,55 ERK-1 and -2 can be activated by phosphorylation events carried out by MEK.10 Numerous signaling pathways induce the phosphorylation of ERK-1 and -2 via MEK (MAPKK). These pathways may be dependent upon Ras,25 G proteins,30,37 PKC,4 and calcium levels.12,22,23,25 In macrophages, the duration of ERK activation correlates with differences in proliferation or activation of the cell.53 Similarly, in the neuroblastoma cell line PC12, the duration of ERK activation is important in determining cell proliferation vs. cell differentiation.34 ERK-1 and -2 are also important in osteoblastic cell proliferation and differentiation.18,25,31,33,36,38,50,60 ERK has been shown to be involved in the proliferative response of osteoblasts to a variety of mitogens.33,36,50,60 In these circumstances, blockade of ERK activation has been shown to suppress osteoblast proliferation. A number of studies have also identified ERK as an important mediator of BMP-2-induced osteoblast differentiation. Independently, in three different cell lines (C2C12, C3H10T1/2, ROS 17/2.8), osteoblastic differentiation was shown to occur following bone morphogenetic protein-2 (BMP-2) treatment.18,33,38 On each occasion, inhibition of ERK resulted in the suppression of differentiation markers. More significantly, treatment of adult human mesenchymal stem cells with osteogenic supplements induces sustained ERK activation,25 whereas ERK inhibition in this system is associated with adipogenesis. This suggests that sustained ERK activation is a critical determinant in the promotion of osteoblastic differentiation. Studies involving the introduction of dominant negative mutants of ERK illustrate the array of osteoblastic functions in which it has a role. In addition to proliferation and differentiation these include cell adhesion and interaction with the extracellular matrix.31 Estrogen and parathyroid hormone (PTH), which have profound effects on bone remodeling and development, also activate ERK.17,25,27 Mechanical strain, another important modulator of bone remodeling,15,32 induces ERK-1/2 activation in both isolated osteoblasts and skeletal muscle cells.26,35 A number of intracellular pathways have been implicated in the response to mechanical strain. These include stretch-activated

Address for correspondence and reprints: Dr. L. E. Lanyon, Royal Veterinary College, Royal College Street, London NW1 0TU, UK. E-mail: [email protected] © 2002 by Elsevier Science Inc. All rights reserved.

186

8756-3282/02/$22.00 PII S8756-3282(02)00797-4

Bone Vol. 31, No. 1 July 2002:186 –194

ion channels,15,16,20,45 release of Ca2⫹ from intracellular stores, increased cAMP and phosphoinositide 3-kinase levels, increased PKC activity,6 release of nitric oxide (NO) and prostaglandins,41,44 and upregulation of insulin-like growth factor-2 (IGF-2) mRNA.59 The extent to which these events contribute to mechanically related control of bone architecture through adaptive modeling and remodeling is uncertain. ERK activation has a downstream influence on the activity of the estrogen receptor (ER)␣,26,27 and there is increasing evidence to suggest that ER␣ activation is an important component of bone cells’ response to mechanical strain.13,14,58 In our previous studies,8,13,14,41,58 rat long-bone-derived osteoblasts and ROS 17/2.8 osteoblast-like cells were plated onto plastic strips, which were then subjected to a short period of dynamic four point bending. Such bending results in strain of the attached cells but also fluid movement around them, as the strip moves through the culture medium. Using this system, we have demonstrated previously that ERK-1/2 are activated in bone cells in response to both mechanical strain (with fluid movement) and to fluid movement alone (produced by moving the strips through the medium without bending them).26 However, the duration of ERK activation induced by these two stimuli differed. Strain (with its associated fluid movement) induced a sustained activation over 24 h, whereas the pattern of activation induced by fluid movement (without strain) was biphasic over the same time period. In the present study we aimed to identify the signaling pathway(s) responsible for ERK-1/2 activation in ROS 17/2.8 osteoblast-like cells after a single 10 min period of mechanical stimulation. We did this by using inhibitors to block the action of important components of known signaling pathways. Cells were exposed to strain (with fluid movement) or fluid movement alone in the presence of L-NAME (N-␻-nitro-2-arginine methylester), an inhibitor of nitric oxide synthase (NOS), to block NO production; indomethacin, an inhibitor of COX to block prostaglandin production; and pertussis toxin, to inhibit G protein signaling. In addition, we used two disrupters of calcium translocation, nifedipine to block L-type calcium channels and TMB-8 hydrochloride, an inhibitor of intracellular calcium mobilization. Materials and Methods Cell Culture The osteoblast-like cell line ROS 17/2.8 was maintained in Dulbecco’s minimum essential medium (DMEM; Life Technologies, Ltd., Paisley, UK) containing 10% heat-inactivated charcoal dextran-stripped fetal calf serum (FCS; Life Technologies) supplemented with 2 mmol/L L-glutamine (Life Technologies), 100 U/mL penicillin (Life Technologies), and 100 ␮g/mL streptomycin (Life Technologies), and incubated at 37°C in a humidified 5% CO2 incubator. Mechanical Straining of Cells ROS 17/2.8 cells were seeded at a density of 2 ⫻ 105 cells/strip onto cell culture-treated plastic strips (66 ⫻ 22 mm), custommade from four well dishes (Nunc, Dossel, Germany). Strips (n ⫽ 5 per condition) were incubated together in 150 cm2 plastic sterile dishes (Merck Eurolab, Poole, UK) with DMEM and 10% charcoal dextran-stripped FCS until the cells reached confluence. The strips onto which the cells were plated were subjected to a single 10 min period of four point bending or movement through the media without bending.8,13,26

H. L. Jessop et al. Strain activates ERK in bone cells

187

Each well contained 8 mL of fresh DMEM containing 10% charcoal dextran-stripped FCS and the strips were subjected to 600 1 Hz cycles of strain generating a peak strain of 3400 ␮⑀.8,13,26 Following the 10 min period of strain (with fluid movement) or fluid movement alone, strips were replaced in the plastic dish and incubated for a further 10 min with media transferred from the wells of the loading apparatus. Static (undisturbed) controls were removed from the incubator, provided with fresh media, and replaced in the incubator for 20 min (equal to 10 min strain plus 10 min incubation). All cells were washed with ice-cold PBS containing 20 ␮mol/L sodium orthovanadate (Sigma-Aldrich, Ltd., Poole, UK) and then lysed in buffer containing 63.5 mmol/L Tris-HCl (pH 6.8), 10% glycerol, 2% sodium dodecylsulfate (SDS), 1 mmol/L sodium orthovanadate, 1 mmol/L AEBSF (Sigma-Aldrich), and 50 ␮g/mL leupeptin (Sigma-Aldrich).54 Exposure to Inhibitors Cells were exposed to strain (with fluid movement) or fluid movement alone in the presence or absence of various inhibitors. With the exception of 10 ␮mol/L TMB-8 hydrochloride (Calbiochem-Novabiochem, Ltd.), which required a preincubation period of 1 h,48 all other inhibitors used (and vehicles) were added to cells in the loading apparatus. Cells were therefore exposed to the inhibitors during both the mechanical stimulus and the subsequent incubation period. The concentrations used were 100 ␮mol/L L-NAME (Sigma-Aldrich), 10⫺6 mol/L indomethacin (Sigma-Aldrich), 9 ⫻ 10⫺12 mol/L pertussis toxin (Sigma-Aldrich), and 10 ␮mol/L nifedipine (Sigma-Aldrich). Exposure to the inhibitors was terminated by rapid aspiration of the media and the cells were washed and harvested as described earlier. The conclusions drawn have been determined by the specificity of the compounds known at present. Because the specificity of some of these compounds is affected by their concentration we used only concentrations previously shown (or stated in the manufacturer’s literature) to have no additional affects. Nitrite Assay in Conditioned Media Because NO is labile, its production cannot be measured directly but rather is assessed from the concentration of nitrite in the medium. Media samples were stored at ⫺20°C prior to determination of nitrite concentration by chemiluminescence as described previously.39,41 Measurement of Prostacyclin Concentrations in Conditioned Medium Once the cells were harvested, conditioned medium was collected and stored (⫺20°C) prior to determination of 6-ketoPGF1␣ (the stable hydrolytic product of PGI2) levels by enzyme immunoassay (Assay Designs, Ltd.). None of the compounds added to the culture medium interfered with the immunoassays and the assay was carried out following the manufacturer’s instructions. Protein Assay The protein content of cell lysates, prior to 10% sodium dodecylsulfate-polyacrylamide gel electrophoresis (SDS-PAGE), was determined using BCA protein assay kits (Pierce & Warriner, Chester, Cheshire, UK). The assays were carried out using 40 ␮L aliquots of cell lysate following the manufacturer’s instructions and read on a Dynex MRX microplate reader (Dynex Technol-

188

H. L. Jessop et al. Strain activates ERK in bone cells

Bone Vol. 31, No. 1 July 2002:186 –194

ogies, Billingshurst, UK) using REVELATION 3.04 software (Microsoft Corp., Redmond, WA). Immunoblotting Aliquots of cell lysates (50 ␮g) were resolved by SDS-PAGE (10%) and transferred onto PVDF (Immobilon-PT) membranes (Millipore, Bedford, MA). Membranes were blocked with 0.2% (w/v) I-block (Applied Biosystems, Foster City, CA) in TBST (50 mmol/L Tris, 150 mmol/L NaCl, 0.02% [v/v] Tween-20 [Sigma], pH 7.4) for 3 h at room temperature. The membranes were probed overnight at room temperature with agitation. Antiactive ERK-1/ERK-2 antibody (E4, Santa Cruz Biotechnology, Inc., Santa Cruz, CA) was used at a dilution of 1:10,000 in 0.2% (w/v) I-block/TBST. Anti-ERK-1/2 antibody (K-23, Santa Cruz) was used at a dilution of 1:10,000 in 0.2% (w/v) I-block/TBST. The blots were washed using 0.2% (w/v) I-block (5 ⫻ 10 min) and incubated with horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG (Pierce and Warriner) diluted 1:5000 in 0.2% (w/v) I-block for 45 min at room temperature or goat anti-mouse HRP-conjugated immunoglobulin G (IgG; Santa Cruz) for ERK-1/2 and active ERK-1/2, respectively. After further washing with 0.2% (w/v) I-block (5 ⫻ 10 min), immunoreactive bands were visualized using luminol reagent (Santa Cruz) and Hyperfilm-ECL film (Amersham International, Amersham, Bucks., UK) according to the manufacturer’s instructions. Densitometry and Statistical Analysis Changes in band densitometry were quantified using MOLECULARANALYST software (version 1.4, BioRad Laboratories, Hercules, CA) and expressed as percentage change (⫾standard error of the mean [SEM]) relative to the static controls. Statistical significance was determined for each comparison using the paired (one-tailed) t-test (Microsoft EXCEL software, 1997), and p ⬍ 0.05 was statistically considered significant. The percentages quoted are average changes from at least three independent experiments and are summarized graphically in the Results. Results Control blots probed with anti-ERK are shown in the upper panel of Figures 1 and 2 to illustrate the total amount of ERK protein present in each lane. Duplicate blots probed with anti-activated ERK (pERK) are shown in the lower panel of each figure. The average percentage increase in pERK-1 and -2 band intensity (⫾SEM) compared with the respective static control are summarized graphically. The p values given for changes in band intensity with strain (with fluid movement) ⫹ inhibitors or fluid movement alone ⫹ inhibitors are compared, respectively, with strain (with fluid movement) and fluid movement alone. Consistent with previous findings, the application of strain (with fluid movement) and fluid movement alone both produced marked activation of both ERK-1 and ERK-2. Calcium Signaling Disrupter Inhibition of Strain-related ERK-1 and -2 Inhibition of intracellular calcium mobilization using TMB-8 hydrochloride produced statistically significant decreases in the activation of ERK-1, but not ERK-2, in response to strain (with fluid movement) (see Figure 1). TMB-8 hydrochloride was the most potent inhibitor of strain-induced ERK-1 activation used in this study. In contrast, activation of ERK-1 and ERK-2 by fluid movement alone was not affected by the addition of TMB-8

Figure 1. The effects of nifedipine and TMB-8 hydrochloride on the phosphorylation of ERK-1/2 in response to strain (with fluid movement). Confluent cells in DMEM ⫹ 10% charcoal dextran-stripped FCS were exposed to a single 10 min period of strain (with fluid movement) or fluid movement in the presence or absence of 10 ␮mol/L nifedipine or 10 ␮mol/L TMB-8 hydrochloride. The cells were harvested 10 min after the end of the period of stimulation and 50 ␮g of whole cell lysates were subjected to western blot analysis using antibodies specific to (A) ERK-1/2 (upper panel) or activated ERK-1/2 (lower panel). (B) The mean percentage change in pERK band densitometry (compared with static controls) was determined from four independent experiments and plotted. The SEM is represented by the error bars. pERK-1 band densitometry in response to strain (with fluid movement) was reduced significantly by the presence of TMB-8 hydrochloride. pERK-2 band intensity was not significantly reduced by either compound. *p ⬍ 0.05, **p ⬍ 0.01, where a is compared with static, b is compared with strain, and c is compared with fluid disturbance alone.

hydrochloride (Figure 1A). Application of strain (with fluid movement) in the presence of the L-type calcium channel blocker, nifedipine, also caused a decrease in strain-induced ERK-1 activation, but this was not statistically significant (Figure 1B). Nifedipine was the only inhibitor used that produced statistically significant decreases in ERK-1 activation resulting from fluid movement alone (Figure 1B). These results indicate the differential roles for intracellular calcium mobilization and the ingress of extracellular calcium into the cells via L-type

Bone Vol. 31, No. 1 July 2002:186 –194

H. L. Jessop et al. Strain activates ERK in bone cells

189

significant increases in ERK-2 activation induced by the application of both strain (with fluid movement) and fluid movement alone (Figure 2B). Indomethacin Reduction of Strain-related ERK-1 and Flow-related ERK-2 Indomethacin (100 ␮mol/L) was a potent inhibitor of strain (with fluid movement)-induced ERK-1 activation (Figure 2A, lanes 5 and 9) and one of only two inhibitors used in this study to produce statistically significant decreases in ERK-1 activation in response to strain (with fluid movement). Cells exposed to fluid movement in the presence of indomethacin did not show a reduction in ERK-1 activation, but the activation of ERK-2 was significantly reduced (Figure 2B). Strain-related ERK Activation Unaffected by Pertussis Toxin Increases in pERK-1 and -2 band intensities produced in response to the application of strain (with fluid movement) or fluid movement alone were unaffected by the exogenous addition of the G protein inhibitor pertussis toxin (Figure 3). This indicates that (pertussis toxin-sensitive) G proteins are unlikely to be involved in the regulation of ERK activation by either the strain (with fluid movement) or fluid movement alone produced in this system. Strain-related Nitrite Production Reduced by Indomethacin and L-NAME

Figure 2. The effects of L-NAME and indomethacin on pERK densitometry in response to strain (with fluid movement) and fluid movement alone. Confluent cells in DMEM ⫹ 10% charcoal dextran-stripped FCS were exposed to a single 10 min period of strain (with fluid movement) or fluid movement in the presence or absence of 1 ⫻ 10⫺6 mol/L indomethacin or 100 ␮mol/L L-NAME. The cells were harvested 10 min after the end of the stimulus and 50 ␮g of whole cell lysates were subjected to western blot analysis using antibodies specific to (A) ERK-1/2 (upper panel) or activated ERK-1/2 (lower panel). (B) The mean percentage change in pERK densitometry (compared with static controls) was determined from five independent experiments. The SEM is represented by the error bars. *p ⬍ 0.05, **p ⬍ 0.01, where a is compared with static, b compared with strain, and c compared with fluid movement. pERK-1 and pERK-2 band densitometry increased significantly in response to both stimuli. Strain (with fluid movement)-induced pERK-1 densitometry was reduced by L-NAME, but significantly reduced by indomethacin. pERK-2 densitometry in response to fluid movement alone was significantly reduced by the presence of indomethacin.

channels in ERK-1 activation induced by strain (with fluid movement) and fluid movement alone, respectively. L-NAME Reduction of Strain-related ERK-1 Activation The presence of 100 ␮mol/L L-NAME produced a small reduction in strain-related ERK-1 activation (Figure 2A, lanes 5 and 7). These changes in ERK-1 activation induced by strain (with fluid movement) were not statistically significant (p ⫽ 0.06), but represented a strong trend. In contrast, ERK-1 activation induced by fluid movement alone was increased, but not significantly, in the presence of L-NAME. Similarly, L-NAME produced non-

Media samples from these experiments were assayed for nitrite concentration to indicate NO production. The results are shown in Figure 4. Mechanical strain (with fluid movement) and fluid movement alone both stimulated the accumulation of nitrite in the media from ROS 17/2.8 cells. Indomethacin and L-NAME both independently reduced the accumulation of nitrite in the media, which was evident in response to the application of fluid movement alone and strain (with fluid movement). Strain-related Prostacyclin Production Reduced by Indomethacin and L-NAME Media samples from these experiments were also assayed for 6-keto-PGF1␣ concentration as an indicator of prostacyclin production. The application of both strain and fluid movement alone resulted in increased 6-keto-PGF1␣ concentration in the media. The concentration of 6-keto-PGF1␣ in the media of cells subjected to fluid movement alone was significantly reduced in the presence of either indomethacin or L-NAME (see Figure 5). The concentration of 6-keto-PGF1␣ in the media of cells subjected to strain (with fluid movement) was also reduced in the presence of either indomethacin or L-NAME, but not significantly (Figure 5). Discussion In the present study we investigated signaling pathways involved in the early activation of ERK-1 and ERK-2 in ROS 17/2.8 osteosarcoma cells in response to mechanical stimulation. Statistically significant decreases in strain (with fluid movement)induced ERK-1 activation occurred with the application of TMB-8 hydrochloride, an inhibitor of intracellular calcium mobilization, and indomethacin, which inhibits NO and PGI2 production. Fluid movement-related ERK-1 activation was significantly decreased only in the presence of nifedipine, a blocker of L-type calcium channels.

190

H. L. Jessop et al. Strain activates ERK in bone cells

Figure 3. The effects of pertussis toxin on pERK densitometry in response to strain (with fluid movement) and fluid movement alone. Confluent cells in DMEM ⫹ 10% charcoal dextran-stripped FCS were exposed to a single 10 min period of strain (with fluid movement) or fluid movement in the presence or absence of 9 ⫻ 10⫺12 mol/L pertussis toxin. The cells were harvested 10 min after the end of the stimulus and 50 ␮g of whole cell lysates were subjected to western blot analysis using antibodies specific to (A) ERK-1/2 (upper panel) or activated ERK-1/2 (lower panel). (B) The mean percentage change in pERK densitometry (compared with static controls) was determined from three independent experiments. The SEM is represented by the error bars. The presence of pertussis toxin had no statistically significant effect on the densitometry of pERK induced by either strain (with fluid movement) or fluid movement alone.

ERK-2 activation in response to strain (with fluid movement) was not significantly reduced by the application of any of the inhibitors used in this study. However, ERK-2 activation in response to fluid movement in the absence of strain was reduced significantly in the presence of indomethacin (as summarized in Figure 6). These data reveal two important observations: First, in ROS 17/2.8 cells, the signaling pathways important for ERK activa-

Bone Vol. 31, No. 1 July 2002:186 –194

tion in response to strain (with fluid movement) differ from those involved in ERK activation in response to fluid movement alone. This difference in the relative contribution of signaling pathways in response to strain (with fluid movement) or fluid movement alone may suggest that bone cells in situ are able to distinguish these stimuli, possibly leading to differences in cell behavior. The second observation relates to the effects of indomethacin and L-NAME on the NO and prostacyclin production by bone cells in response to strain (with fluid movement). The COX inhibitor, indomethacin, was associated with a reduction in prostacyclin production but also abrogated the increased NO production that is elicited in response to strain (with fluid movement) or fluid movement alone. The selective NOS inhibitor, L-NAME, was associated with its expected reduction in the production of NO, but also with reduced prostacyclin production. Klein-Nulend and coworkers29 observed a similar loss of prostaglandin production in response to pulsating fluid flow, in the presence of a different NOS inhibitor (NG-monomethyl-L-arginine). This concomitant loss of activity of NOS and COX in response to strain or fluid movement suggests that the activity of these two enzymes is associated in some manner under mechanical perturbation.42,43,45 The mechanism by which the activities of NOS and COX are associated is beyond the scope of the present study. Because we are unable to inhibit production of prostacyclin without also inhibiting NO production, we cannot distinguish their roles regarding relative importance for the activation of ERK in response to strain (with fluid movement) or fluid movement alone. The situation is complicated further by the observation that, although L-NAME reduced both NO and prostacyclin production (similar to indomethacin) induced by mechanical perturbation, strain (with fluid movement)-induced ERK-1 activation and fluid movement-induced ERK-2 activation were, nonetheless, not significantly reduced by the presence of LNAME (but were significantly reduced by indomethacin). This suggests that indomethacin may have additional effects, not shared by L-NAME, which result in the loss of ERK activation. The role of these signal transduction pathways in the osteogenic response to mechanical loading remains unclear. The separate administration of NOS or prostaglandin synthase inhibitors has been shown to reduce loading-related bone formation in vivo.28,40,47,51,52 This suggests that NO and PG production, which are both required for strain-induced ERK activation, are also necessary for loading-related bone formation in vivo. Furthermore, a number of studies have demonstrated a requirement for increased intracellular Ca2⫹ levels for PGE2 production,1,46 ERK activation,56,57 and osteoblast proliferation.5,24 Taken together, these data suggest that bone cells’ earliest responses to mechanical events involve Ca2⫹ signaling followed by NO and prostaglandin production, which in turn trigger signaling events that result in ERK activation. Once activated, ERK has numerous downstream targets, including transcription factors and other protein kinases, culminating in changes in gene expression and cell behavior. Thus, ERK may represent a point of convergence for a number of signaling pathways activated by mechanical stimuli. Whether ERK activation is sustained or transient may be a critical determinant of osteoblast behavior in terms of survival, proliferation, or differentiation. If this is the case, then the differences in the pattern of ERK activation observed in response to fluid movement alone as compared with strain with fluid movement,26 and the differing signaling requirements described in the present study, suggest that these two stimuli may have different effects on longer term changes in cell behavior. It is clear that the quantitative nature of our analysis by western blot is likely to be more stringent than interpretation

Bone Vol. 31, No. 1 July 2002:186 –194

H. L. Jessop et al. Strain activates ERK in bone cells

191

Figure 4. The concentration of nitrite in media from cells strained in the presence of indomethacin or L-NAME. Confluent cells in DMEM ⫹ 10% charcoal dextranstripped FCS were exposed to a single 10 min period of strain (with fluid movement) or fluid movement in the presence or absence of 1 ⫻ 10⫺6 mol/L indomethacin or 100 ␮mol/L L-NAME. The media was collected after the cells were harvested and the nitrite concentration determined as described in Materials and Methods. The mean concentration from three independent experiments was plotted and the SEM is represented by error bars. The nitrite concentration is indicative of the amount of nitric oxide released by the cells. *p ⬍ 0.05, **p ⬍ 0.01, where a is compared with static, and c is compared with fluid movement. The concentration of nitrite was increased significantly in media derived from ROS 17/2.8 cells subjected to either stimulus. The presence of either indomethacin or LNAME reduced the nitrite concentration detected.

based on a solely qualitative evaluation of the effects of the various inhibitors. However, although the procedures used in the present study were standardized, the qualitative nature of the western blot protocol and biological variability in the expression of ERK and responsiveness of the cells to strain may explain the large SEM values. Our interpretation is thus likely to underestimate the effects of these inhibitors, because only marked and quantitatively reproducible effects produce statistically significant differences in the response. In spite of the aforementioned data, it was clearly shown that the activation of ERK-1 in response to fluid movement was reduced significantly in the presence of nifedipine, but in none of the other inhibitors investigated. In contrast, ERK-2 activation in Figure 5. The concentration of 6-ketoPGF1␣ in media from cells subjected to strain (with fluid movement) or fluid movement in the presence of indomethacin or L-NAME. Confluent cells in DMEM ⫹ 10% charcoal dextran-stripped FCS were exposed to a single 10 min period of strain (with fluid movement) or fluid movement in the presence or absence of 1 ⫻ 10⫺6 mol/L indomethacin or 100 ␮mol/L L-NAME. The media was saved after the cells were harvested and the 6-keto-PGF1␣ concentration determined as described in Materials and Methods. The mean concentration from three independent experiments was plotted and the SEM is represented by error bars. The 6-keto-PGF1␣ concentration is indicative of the amount of prostacyclin produced by the cells. *p ⬍ 0.05, where a is compared with static, and c is compared with fluid movement alone. The concentration of 6-keto-PGF1␣was increased in the media of cells subjected to both stimuli. These increases were abrogated by the presence of either compound.

response to fluid movement was significantly reduced only in the presence of indomethacin. These observations highlight important differences in the relative contribution that particular signaling pathways may have in the activation of ERK-1 and -2 in response to fluid movement alone. They also indicate differences in response from fluid movement when experienced coincidentally with mechanical strain. The data suggest that, although ERK-1 activation in response to fluid movement alone involves calcium passage through L-type calcium channels, ERK-2 activation is dependent on a competent prostaglandin and/or NO production pathway. Similarly, ERK-1 activation in response to strain (with fluid movement) is also dependent upon prostaglandin and/or NO production and, in addition, requires the mobili-

192

H. L. Jessop et al. Strain activates ERK in bone cells

Bone Vol. 31, No. 1 July 2002:186 –194

Figure 6. Diagrammatic summary of the probable pathways responsible for activation of ERK in osteoblast-like cells in response to mechanical stimuli. Mechanistic interpretation determined by the action of specific inhibitors (shown as ⵬).

zation of intracellular calcium. Paradoxically, however, ERK-2 activation in response to strain (with fluid movement) was not reduced significantly by the presence of any of the inhibitors used in this study. The implications of these differences may lie in the targets downstream from these kinases. Differences in cell behavior in response to strain (with fluid movement) and fluid movement alone may reflect differences in the profile of ERK activation. The work described herein expands on previous observations in ROS 17/2.8 cells,26 which showed that a single 10 min period of mechanical strain (with fluid movement) was sufficient to increase the phosphorylation of estrogen receptor (ER)␣. It was determined in that study that the observed increase in ER␣ phosphorylation, in response to estrogen or strain, was ERKdependent. It is of interest that the single most important component on the strain stimulus with regard to ERK activation appears to be mobilization of intracellular calcium. ImprotalBrears24 showed that estrogen-induced ERK activation is also inhibited by blocking of intracellular calcium mobilization. Exposure to estrogen has also been shown to lead to NO release.2,21 These findings suggest that there is a common pathway in the early responses to mechanical strain and estrogen in bone cells, which involves intracellular calcium mobilization, NO production, ERK activation, and ER␣ phosphorylation.7,23,42

Acknowledgments: The authors thank the Biotechnology & Biological Sciences Research Council (BBSRC) and the Wellcome Trust for funding.

References 1. Ajubi, N. E., Klein-Nulend, J., Alblas, M. J., Burger, E. H., and Nijweide, P. J. Signal transduction pathways involved in fluid flow-induced PGE2 production by cultured osteocytes. Am J Physiol 276:E171–E178; 1999. 2. Armour, K. E. and Ralston, S. H. Estrogen upregulates endothelial constitutive nitric oxide synthase expression in human osteoblast-like cells. Endocrinology 139:799 –802; 1998. 3. Boulton, T. G., Yancopoulos, G. D., Gregory, J. S., Slaughter, C., Moomaw, C., Hsu, J., and Cobb, M. H. An insulin-stimulated protein kinase similar to yeast kinases involved in cell cycle control. Science 249:64 –66; 1990. 4. Buchner, K. The role of protein kinase C in the regulation of cell growth and in signalling to the cell nucleus. J Cancer Res Clin Oncol 126:1–11; 2000. 5. Carpio, L. C., Stephan, E., Kamer, A., and Dziak, R. Sphingolipids stimulate cell growth via MAP kinase activation in osteoblastic cells. Prostagland Leukot Essent Fatty Acids 61:267–273; 1999. 6. Carvalho, R. S., Scott, J. E., Suga, D. M., and Yen, E. H. K. Stimulation of signal transduction pathways in osteoblasts by mechanical strain potentiated by parathyroid hormone. J Bone Miner Res 9:999 –1011; 1994. 7. Chang, L. and Karin, M. Mammalian MAP kinase signalling cascades. Nature 410:37–40; 2001. 8. Cheng, M. Z., Zaman, G., Rawlinson, S. C. F., Mohan, S., Baylink, D. J., and Lanyon, L. E. Mechanical strain stimulates ROS cells proliferation through IGF-II and mechanical estrogen through IGF-I. J Bone Miner Res 10:1742– 1750; 1999. 9. Cook, S. J., Aziz, N., and McMahon, M. The repertoire of Fos and Jun proteins expressed during the G1 phase of the cell cycle is determined by the duration of mitogen-activated protein kinase activation. Mol Cell Biol 19:330 –341; 1999. 10. Crews, C. M., Alessandrini, A., and Erikson, R. L. The primary structure of MEK, a protein kinase that phosphorylates the ERK gene product. Science 258:478 –480; 1992. 11. Cross, T. G., Scheel-Toellner, D., Henriquez, N. V., Deacon, E., Salmon, M., and Lord, J. M. Serine/threonine protein kinases and apoptosis. Exp Cell Res 256:34 –41; 2000.

Bone Vol. 31, No. 1 July 2002:186 –194 12. Cruzalegui, F. H. and Bading, H. Calcium-regulated protein kinase cascades and their transcription factor targets. Cell Mol Life Sci 57:402–410; 2000. 13. Damien, E., Price, J. S., and Lanyon, L. E. The estrogen receptor’s involvement in osteoblasts adaptive response to mechanical strain. J Bone Miner Res 13:1275–1282; 1998. 14. Damien, E., Price, J. S., and Lanyon, L. E. Mechanical strain stimulates osteoblast proliferation through the estrogen receptor in males as well as females. J Bone Miner Res 15:2169 –2177; 2000. 15. Duncan, R. L. and Hruska, K. A. Chronic, intermittent loading alters mechanosensitive channel characteristics in osteoblast-like cells. Am J Physiol 267:F909 –F916; 1994. 16. Duncan, R. L. and Turner, C. H. Mechanotransduction and the functional response of bone to mechanical strain. Calcif Tissue Int 57:344 –358; 1995. 17. Endoh, H., Sasaki, H., Maruyama, K., Takeyama, K.-I., Waga, I., Shimizu, T., Kato, S., and Kawashima, H. Rapid activation of MAP kinase by estrogen in a bone cell line. Biochem Biophys Res Commun 235:99 –102; 1997. 18. Gallea, S., Lallemand, F., Atfi, A., Rawadi, G., Ramez, V., Spinella-Jaegle, S., Kawai, S., Faucheu, C., Huet, L., Baron, R., and Roman-Roman, S. Activation of mitogen-activated protein kinase cascades is involved in regulation of bone morphogenetic protein-2-induced osteoblast differentiation in pluripotent C2C12 cells. Bone 28:491–498; 2001. 19. Gotoh, Y., Nishida, E., Matsuda, S., Shiina, N., Kosako, H., Shiokawa, K., Akiyama, T., Ohta, K., and Sakai, H. In vitro effects on microtubule dynamics of purified Xenopus M phase-activated MAP kinase. Nature 349:251–254; 1991. 20. Harter, L. V., Hruska, K. A., and Duncan, R. L. Human osteoblast-like cells respond to mechanical strain with increased bone matrix protein production independent of hormonal regulation. Endocrinology 136:528 –535; 1995. 21. Haynes, M. P., Sinha, D., Russell, K. S., Collinge, M., Fulton, D., MoralesRuiz, M., Sessa, W. C., and Bender, J. R. Membrane estrogen receptor engagement activates endothelial nitric oxide synthase via the PI3-kinase-Akt pathway in human endothelial cells. Circ Res 87:677–682; 2000. 22. Huang, Z., Cheng, S. L., and Slatopolsky, E. Sustained activation of the extracellular signal-regulated kinase pathway is required for extracellular calcium stimulation of human osteoblast proliferation. J Biol Chem 276: 21351–21358; 2001. 23. Huang, S., Maher, V. M., and McCormick, J. J. Involvement of intermediary metabolites in the pathway of extracellular Ca2⫹-induced mitogenactivated protein kinase activation in human fibroblasts. Cell Signal 11: 263–274; 1999. 24. Improtal-Brears, T., Whorton, A. R., Codazzi, F., York, J. D., Meyer, T., and McDonnell, D. P. Estrogen-induced activation of MAPK requires mobilization of intracellular calcium. Proc Natl Acad Sci (USA) 96:4686 – 4691; 1999. 25. Jaiswal, R. K., Jaiswal, N., Bruder, S. P., Mbalaviele, G., Marshak, D. R., and Pittenger, M. F. Adult human mesenchymal stem cell differentiation to the osteogenic or adipogenic lineage is regulated by mitogen-activated protein kinase. J Biol Chem 275:9645–9657; 2000. 26. Jessop, H. L., Sjo¨ berg, M., Cheng, M. Z., Zaman, G., Wheeler-Jones, C. P. D., and Lanyon, L. E. Mechanical strain as well as estrogen activates estrogen receptor ␣ in bone cells. J Bone Miner Res 16:1045–1055; 2001. 27. Kato, S., Endoh, H., Masuhiro, Y., Kitamoto, T., Uchiyama, S., Sasaki, H., Masushige, S., Gotoh, Y., Nishida, E., Kawashima, H., Metzger, D., and Chambon, P. Activation of the estrogen receptor through phosphorylation by mitogen-activated-protein kinase. Science 270:1491–1494; 1995. 28. Keller, J., Bayer-Kristensen, I., Bak, B., Bunger, C., and KjaersgaardAndersen, P. Indomethacin and bone remodelling. Effect on cortical bone after osteotomy in rabbits. Acta Orthopaed Scand 60:119 –121; 1989. 29. Klein-Nulend, J., Semeins, C. M., Ajubi, N. E., Nijweide, P. J., and Burger, E. H. Pulsating fluid flow increases nitric oxide (NO) synthesis by osteocytes but not periosteal fibroblasts — correlation with prostaglandin upregulation. Biochem Biophys Commun 217:640 –648; 1995. 30. Kyriakis, J. M. and Avruch, J. S6 kinases and MAP kinases: Sequential intermediates in insulin/mitogen-activated protein kinase cascades. Woodgett, J. R., ed. Protein Kinases. Oxford, UK: Oxford University; 85–148; 1996. 31. Lai, C.-F., Chaudhary, L., Fausto, A., Halstead, L. R., Ory, D. S., Avioli, L. V., and Cheng, S.-L. Erk is essential for growth, differentiation, intergrin expression, and cell function in human osteoblastic cells. J Biol Chem 276:14443– 14450; 2001. 32. Lanyon, L. E. Control of bone architecture by functional load bearing. J Bone Miner Res 7(Suppl.):S369 –S375; 1992.

H. L. Jessop et al. Strain activates ERK in bone cells

193

33. Lou, J., Tu, Y., Li, S., and Manske, P. R. Involvement of ERK in BMP-2 induced osteoblastic differentiation of mesechymal progenitor cell line C3H10T1/2. Biochem Biophys Res Commun 268:757–762; 2000. 34. Marshall, C. J. Specificity of receptor tyrosine kinase signaling: Transient versus sustained extracellular signal-regulated kinase activation. Cell 80:179 – 185; 1995. 35. Martineau, L. C. and Gardiner, P. F. Insight into skeletal muscle mechanotransduction: MAPK activation is quantitatively related to tension. J Appl Physiol 91:693–702; 2001. 36. Mathov, I., Plotkin, L. I., Sgarlata, C. L., Leoni, J., and Bellido, T. Extracellular signal-regulated kinases and calcium channels are involved in the proliferative effect of bisphosphonates on osteoblastic cells in vitro. J Bone Miner Res 11:2050 –2056; 2001. 37. Naor, Z., Benard, O., and Seger, R. Activation of MAPK cascades by Gprotein-coupled receptors: The case of gonadotrophin-releasing hormone receptor. Trends Endocrinol Metab 11:91–99; 2000. 38. Palcy, S. and Goltzman, D. Protein kinase signalling pathways involved in the up-regulation of the rat alpha1(I) collagen gene by transforming growth factor beta1 and bone morphogenetic protein 2 in osteoblastic cells. Biochem J 343:21–27; 1999. 39. Palmer, R. M., Ferrigo, A. G., and Moncada, S. Nitric oxide release accounts for biological activity of endothelium-derived releasing factor. Nature 325: 524 –525; 1987. 40. Pead, M. J. and Lanyon, L. E. Indomethacin modulation of load-related stimulation of new bone formation in vivo. Calcif Tissue Int 45:34 –40; 1989. 41. Pitsillides, A. A., Rawlinson, S. C., Suswillo, R. F., Bourrin, S., Zaman, G., and Lanyon, L. E. Mechanical strain-induced NO production by bone cells: A possible role in adaptive bone (re)modelling? FASEB J 9:1614 –1622; 1995. 42. Pitsillides, A. A., Rawlinson, S. C. F., Zaman, G., Suswillo, R. F. L., Mosley, J. R., Cheng, M. Z., and Lanyon, L. E. Regulation of nitric oxide synthase and cyclooxygenase activities in bone cells: A possible loading-related linkage. Calcif Tissue Int 64(Suppl. 1):S42; 1999. 43. Rawlinson, S. C. F. Early loading-related responses of resident cells in mammalian bone organ explants. University of London: PhD thesis; 1999. 44. Rawlinson, S. C. F., El Haj, A. J., Minter, S. L., Bennet, A., Tavares, I. A., and Lanyon, L. E. Load-related release of prostaglandins in cores of cancellous bone in culture — a role for prostacyclin in adaptive bone remodelling. J Bone Miner Res 6:1345–1351; 1991. 45. Rawlinson, S. C. F., Pitsillides, A. A., and Lanyon, L. E. Involvement of different ion channels in osteoblasts’ and osteocytes’ early responses to mechanical strain. Bone 19:609 –614; 1996. 46. Reich, K. M., McAllister, T. N., Gudi, S., and Frangos, J. A. Activation of G proteins mediates flow-induced prostaglandin E2 production in osteoblasts. Endocrinology 138:1014 –1018; 1997. 47. Samuels, A., Perry, M. J., and Tobias, J. H. High-dose estrogen-induced osteogenesis in the mouse is partially suppressed by indomethacin. Bone 25:675–680; 1999. 48. Schmidt, K. N., Traenckner, B.-M., Meier, B., and Baeuerle, P. A. Induction of oxidative stress by okadaic acid is required for activation of transcription factor NK-␬B. J Biol Chem 270:27136 –27142; 1995. 49. Shichiri, M., Yokokura, M., Marumo, F., and Hirata, Y. Endothelin-1 inhibits apoptosis of vascular smooth muscle cells induced by nitric oxide and serum deprivation via MAP kinase pathway. Arterioscler Thromb Vasc Biol 20:989 – 997; 2000. 50. Swarthout, J. T., Doggett, T. A., Lemker, J. L., and Partridge, N. C. Stimulation of extracellular signal-regulated kinases and proliferation in rat osteoblastic cells by parathyroid hormone is protein kinase C-dependent. J Biol Chem 276:7586 –7592; 2001. 51. Tornkvist, H., Lindholm, T. S., Netz, P., Stromberg, L., and Lindholm, T. C. Effect of ibuprofen and indomethacin on bone metabolism reflected in bone strength. Clin Orthopaed Rel Res 187:255–259; 1984. 52. Turner, C. H., Takano, Y., Owan, I., and Murrell, G. A. C. Nitric oxide inhibitor L-NAME suppresses mechanically induced bone formation in rats. Am J Physiol 270:E634 –E639; 1996. 53. Valledor, A. F., Comalada, M., Xaus, J., and Celada, A. The differential time course of extracellular-regulated kinase activity correlates with the macrophage response toward proliferation or activation. J Biol Chem 275:7403–7409; 2000. 54. Wheeler-Jones, C. P. D., May, M. J., Houliston, R. A., and Pearson, J. D. Inhibition of MAP kinase kinase (MEK) blocks endothelial PGI2 release but has no effect on von Willebrand factor secretion or E-selectin expression. Fed Eur Bone Soc Lett 388:180 –184; 1996.

194

H. L. Jessop et al. Strain activates ERK in bone cells

55. Xiao, G., Jiang, D., Thomas, P., Benson, M. D., Guan, K., Karsenty, G., and Franceschi, R. T. MAPK pathways activate and phosphorylate the osteoblastspecific transcription factor, Cbfa1. J Biol Chem 275:4453–4459; 2000. 56. Yamaguchi, T., Chattopadhyay, N., Kifor, O., Sanders, J. L., and Brown, E. M. Activation of p42/p44 and p38 mitogen-activated protein kinases by extracellular calcium-sensing receptor agonists induces mitogenic responses in the mouse osteoblastic MC3T3-E1 cell line. Biochem Biophys Res Commun 279:363–368; 2000. 57. You, J., Reilly, G. C., Zhen, X., Yellowley, C. E., Chen, Q., Donahue, H. J., and Jacobs, C. R. Osteopontin gene regulation by oscillatory fluid flow via intracellular calcium mobilization and activation of mitogen-activated protein kinase in MC3T3-E1 osteoblasts. J Biol Chem 276:13365–13371; 2001. 58. Zaman, G., Cheng, M. Z., Jessop, H. L., White, R., and Lanyon, L. E. Mechanical strain activates estrogen response elements in bone cells. Bone 27:233–239; 2000.

Bone Vol. 31, No. 1 July 2002:186 –194 59. Zaman, G., Suswillo, R. F. L., Cheng, M. Z., Tavares, I. A., and Lanyon, L. E. Early responses to dynamic strain change and prostaglandins in bone-derived cells in culture. J Bone Miner Res 12:769 –777; 1997. 60. Zhang, W., Lee, J. C., Kumar, S., and Gowen, M. ERK pathway mediates the activation of Cdk2 in IGF-1-induced proliferation of human osteosarcoma MG-63 cells. J Bone Miner Res 14:528 –535; 1997.

Date Received: November 22, 2001 Date Revised: February 7, 2002 Date Accepted: February 7, 2002