(The FASEB Journal. 1999;13:553-557.)
Laboratory of Cell Growth, Department of Medicine, Veterans Affairs Medical Center, San Francisco, California 94121, USA
In serum-deprived MC3T3-E1 osteoblasts, mechanical stimulation caused by mild (287 x g) centrifugation induced a 10-fold increase in mRNA levels of the proto-oncogene, c-fos. Induction of c-fos was abolished by the cAMP-dependent protein kinase inhibitor H-89, suggesting that the transient c-fos mRNA increase is mediated by cAMP. Down-regulation of protein kinase C (PKC) activity by chronic TPA treatment failed to significantly reduce c-fos induction, suggesting that TPA-sensitive isoforms of PKC are not responsible for c-fos up-regulation. In addition, 287 x g centrifugation increased intracellular prostaglandin E2 (PGE2) levels 2.8-fold (P<0.005). Since we have previously shown that prostaglandin E2 (PGE2) can induce c-fos expression via a cAMP-mediated mechanism, we asked whether the increase in c-fos mRNA was due to centrifugation-induced PGE2 release. Pretreatment with the cyclooxygenase inhibitors indomethacin and flurbiprofen did not hinder the early induction of c-fos by mechanical stimulation. We conclude that c-fos expression induced by mild mechanical loading is dependent primarily on cAMP, not PKC, and initial induction of c-fos is not necessarily dependent on the action of newly synthesized PGE2.—Fitzgerald, J., Hughes-Fulford, M. Mechanically induced c-fos expression is mediated by cAMP in MC3T3-E1 osteoblasts.
Key Words: gene expression • signal transduction • gravity • mechanical stimulation • PGE2 synthesis
THE EFFECT OF MECHANICAL STRESS in osteoblasts is complex and appears to be dependent on the magnitude of the strain applied. High strain increases cell proliferation, DNA synthesis, and prostaglandin production and decreases levels of the bone marker protein alkaline phosphatase (AP)3 (1 , 2 ). At more physiological strain levels, AP levels are increased and cell proliferation is decreased (3) .
However, the effect of mechanical stress on bone at the cellular level is relatively unknown. mRNA levels for type 1 collagen and osteopontin were up-regulated in chronic low frequency strain in OHS-4 osteosarcoma cell line (4) and ß1-integrin mRNA was increased in human osteosarcoma TE-85 cells exposed to low frequency mechanical stimulation (5) . Mechanical loading of embryonic chick tibiotarsi maintained in organ culture resulted in increased expression of type 1 collagen and was accompanied by an increase in AP activity (6) . Mechanical stimulation of rat caudal vertebra increased c-fos mRNA fourfold (7) . These studies clearly demonstrate that bone cells respond to mechanical strain at the level of gene expression.
Two studies have addressed the question of whether increased gravitational force can alter gene expression in cultured osteoblasts. First, when MC3T3-E1 osteoblasts were centrifuged up to several thousand times normal gravity, the oncogenes c-fos and c-jun were induced beginning at 50 x g and induction was shown to be dependent on a protein kinase C (PKC)-mediated signaling pathway (8) . Second, we showed in a previous report that changes in gene expression can occur with as little as 3 x g gravity. This small magnitude loading doubled c-fos mRNA levels and decreased by 50% osteocalcin mRNA levels in MC3T3-E1 osteoblasts (9) . Therefore, changes in the level of gene expression can be detected after both small and large magnitude stress. Understanding the mechanism behind changes in gene expression in osteoblasts is important since it directly relates to the growth and healing of bones after fracture and may indicate the existence of a mechanical load sensing mechanism.
The proto-oncogene c-fos is one of a family of transcription factors that includes c-fos, fosB, fra-1, and fra-2. Many signal pathways such as tyrosine kinases, p21 ras, MAP kinases, s6 kinases, and PKC can induce fos mRNA expression (10) . Fos protein is important in bone cells since recognition elements for the AP-1 complex are found in the promoter regions of several genes involved in the growth and mineralization of bone including osteocalcin, alkaline phosphatase, and collagen type 1. Experiments with transgenic mice indicate that regulation of c-fos gene expression is important for normal bone development 11-15) .
In this study we show that the gene c-fos is induced by mechanical loading caused by 287 x g centrifugation. The early increase in c-fos mRNA by centrifugation is dependent on PKA-mediated signal transduction pathways and does not necessarily require newly synthesized PGE2.
|MATERIALS AND METHODS|
Minimum essential media alpha (aMEM) was purchased from Fisher Scientific (Pittsburgh, Pa.). 12-O-Tetradecanoylphorbol 13-acetate (TPA) and H-89 were obtained from LC Laboratories (Woburn, Mass.). Indomethacin and flurbiprofen were obtained from Sigma (St. Louis, Mo.). Fetal calf serum (FCS) was from Hyclone Labs Inc. (Logan, Utah). Moloney murine leukemia virus (MMLV), Taq DNA polymerase was from GIBCO-BRL (Grand Island, N.Y.). RNase inhibitor was from Boehringer-Mannheim (Indianapolis, Ind.). Oligonucleotides were ordered from Operon Technologies Inc. (Alameda, Calif.).
MC3T3-E1 cell line is clonally derived from embryonic mouse calvaria (16) . Cells were plated and grown to confluency in aMEM containing 10% FCS, antibiotic (100 U penicillin G/ml, 0.01 mg streptomycin/ml, 0.25 mg amphotericin B/ml in 0.85% saline), 20 mM L-glutamine, and 25 mM HEPES buffer. Cells were serum deprived for 16–18 h before the start of each experiment by incubation in aMEM containing 1% FCS. Confluent cultures of osteoblasts were treated for 60 min with various agents (see figure legends), centrifuged for 5 min in tissue culture plates at 1600 RPM, which corresponds to a centrifugal force of 287 x g, and returned to the incubator for 30 min. The centrifuge has trunnions especially designed to hold cell culture plates. The strain experienced by the cells is primarily through the apical–basal axis. Cell counts were performed in a ZBI Coulter counter.
RNA isolation, reverse transcription, and PCR
RNA from cultured MC3T3-E1 osteoblasts was isolated using a modified guanidinium thiocyanate method based on the protocol previously described by Chomczynski and Sacchi (17; unpublished results). RNA was quantitated and 1.5 µg was added to a reverse transcriptase (RT) reaction in 30 µl containing 50 mM Tris-HCl (pH 8.3), 75 mM KCl, 3 mM MgCl2, 1 mM dNTPs, 1.67 µM oligo d(T) primer, 6.67 units of MMLV, and 0.67 units of RNase inhibitor. The RT reaction was incubated at room temperature for 10 min, at 42°C for 45 min, and then 72°C for 10 min. The reaction was stopped by incubation at 99°C for 5 min. Five microliters of cDNA from the RT reaction was added directly to a 50 µl polymerase chain reaction (PCR) containing 20 mM Tris-HCl (pH 8.4), 50 mM KCl, MgCl2, 200 µM dNTPs, 25–50 pmol oligonucleotide primers, and 0.05 units of Taq DNA polymerase. The amplification conditions were as follows: 94°C/100 s, 63°C/70 s, 72°C/100 s. c-fos was amplified for 30 cycles and CPH1 for 22 cycles. Oligonucleotide primers were designed to span at least one intron in order to detect any contaminating genomic DNA carried over from the RNA isolation step. All primers span an intron and PCRs are amplified only in the linear range. c-fos primer sequences have been described previously (9) and CPH1 primer sequences were designed from Genbank sequences by M.H.F. as follows: cyclophilin, CPH1-F primer 5'-CGT CTC CTT TGA GCT GTT TGC AGA C-3' and CPH1-R primer 5'-CAT AAT CAT AAA CTT AAC TCT GCA ATC CAG C-3'. Conditions were established so that PCR was stopped in the linear range so the reaction products could be accurately quantified and compared. To correct for small variations between experiments, each c-fos PCR product was compared to CPH1 PCR products derived from the same RT reaction. Hence, the assay determines the relative levels of levels of c-fos mRNA. PCR products were electrophoresed on 2.5% agarose gels and photographed on Polaroid 667 film using a Polaroid DS-34 camera. Photographs were then scanned and digitized using the Lacie Silverscan II at 360 dpi into Adobe Photoshop v3.0. Density analysis was performed using the public domain NIH Image 1.58 program.
The exogenous PGE2 levels were quantitated using the PGE2-monoclonal enzyme immunoassay kit (Cayman Chemical; Ann Arbor, Mich.) according to the manufacturer's instructions.
|RESULTS AND DISCUSSION|
We have previously shown that 3 x g centrifugation alters c-fos and osteocalcin gene expression in MC3T3-E1 osteoblasts, although the changes in mRNA levels detected were relatively small (9) . Since 3 x g centrifugation approximates the mechanical stress experienced during mild exercise, such as walking, we wanted to determine whether changes in gene expression could be detected in larger mechanical loading in osteoblasts under conditions that simulate vigorous exercise or near-fracture forces. Previous experiments using MC3T3-E1 osteoblasts have shown that two apparently different kinds of mechanical stress—mild centrifugation (9) and vibration (18) —result in very similar changes in gene expression, which suggests that centrifugation is a valid method for applying a mechanical load. We chose to study expression of the c-fos gene because it is known to be involved in bone growth and responds to different kinds of stress including chemical, UV, and mechanical stress; its mechanism of gene activation is relatively well studied (see reviews in 10, 19, 20 ).
MC3T3-E1 osteoblasts were subjected to a force of 287 x g in the presence of various agents. Total RNA was isolated and RT and PCR performed on the c-fos and cyclophillin genes; the PCR products were photographed and imaged as described in Materials and Methods. Representative gels are shown in Fig. 1 and Fig. 2 and a summary of three to four experiments is shown in Fig. 3 . The data in Fig. 3 represent mean c-fos mRNA after different treatments and are expressed in comparison to c-fos levels after centrifugation alone, which is set at 100.
After a 5 min centrifugation, c-fos mRNA levels were increased by 10-fold compared to noncentrifuged controls 30 min after the end of the centrifugation run (Fig. 1 , lanes 1 and 2). This result agrees with our earlier study in MC3T3-E1 osteoblasts, where a mild mechanical stress corresponding to 3 x g centrifugation induced a 1.7-fold increase in c-fos mRNA levels (9) . The difference in magnitude of mRNA up-regulation between these two studies may reflect the increased magnitude of mechanical strain applied. Previous experiments have shown that in vitro mechanical loading transiently stimulates c-fos expression in condylar tissue (21) , cardiac myocytes (22) , caudal vertebra (7) , and tibial periosteum (1) . It is noteworthy that c-fos mRNA levels did not return to control levels after 3 h (Fig. 3 , lane 5), which occurred in the 3 x g experiments, and suggests that although c-fos is up-regulated in both experimental situations, additional systems may be acting to maintain c-fos gene activation at higher magnitudes of force. Such subtle responses to different magnitude of force may be important for bone repair after breakage but are not required to be activated by very mild mechanical stress.
To identify the signal transduction pathways involved in the up-regulation of c-fos by mechanical loading, changes in the mRNA level for c-fos in osteoblasts treated with protein kinase activators and inhibitors were determined. In cells pretreated for 1 h with a specific cAMP-dependent kinase (PKA) inhibitor, H-89, up-regulation of c-fos mRNA levels induced by centrifugation was completely abolished (Fig. 1 , lane 6, and Fig. 3 ). These data suggest that the increase in c-fos mRNA induced by centrifugal mechanical stress is dependent on increased cAMP levels and subsequent activation of PKA.
Next, we asked whether the up-regulation of c-fos depended on Ca2+-dependent PKC. Short exposure to the potent PKC activator TPA stimulates PKC activity and chronic (16 h) treatment is known to down-regulate PKC activity (23) . Treatment with TPA alone increased c-fos mRNA levels after 30 min in serum-depleted osteoblasts (Fig. 2 , lane 3), in agreement with a previous observation (J. F. and M. H. F., unpublished observations). Sixteen hours of treatment with TPA alone or followed by reapplication of TPA for 30 min failed to induce c-fos, confirming that PKC activity is down-regulated by chronic TPA treatment (Fig. 2 , lanes 4 and 5, and Fig. 3 ). However, after a 16 h incubation with TPA, a 287 x g force was still able to cause an increase in c-fos mRNA levels, suggesting that the induction does not occur by a TPA-sensitive PKC-dependent mechanism (Fig. 2 , lane 6). These data suggest that c-fos induction by centrifugation is mediated by cAMP and PKA but not PKC in MC3T3-E1 cells under these conditions.
Since arachidonic acid and PGE2 is released in response to mechanical loading in bones in vitro and in vivo (24 , 25 ) and PGE2 is known to induce c-fos (26, 27; unpublished observations), we designed an experiment to minimize the production of PGE2 by serum depriving the cells and thereby reducing substrate levels for PGE2. PGE2 levels in the extracellular media were assayed 30 min after centrifugation by immunoassay (Fig. 4 ). Under these conditions, centrifugation induced a threefold increase in PGE2 compared to the noncentrifuged control. Since this laboratory and others have shown that the arachidonic acid metabolite, prostaglandin E2, (PGE2) induces c-fos in several cell types (26 , 28 ), including MC3T3-E1 osteoblasts (27; unpublished observations), we pretreated the cells for 60 min with indomethacin or flurbiprofen and measured c-fos activation by RT/PCR. Figure 1 (lanes 3 and 4) and Fig. 3 shows that pretreatment with 5 µM indomethacin for 60 min had little effect on c-fos induction. Since COX inhibitors used at concentrations known to inhibit PGE2 release do not alter centrifugation-induced c-fos mRNA levels, we suggest that new synthesis of PGE2 is not necessary for the induction of c-fos by mechanical forces. Earlier we (9) examined the effect of 3 x g centrifugation on c-fos expression in a study designed to examine gene activation under similar conditions (9) . We found that c-fos mRNA levels increased in the absence of newly released PGE2 synthesis, suggesting that PGE2 was not responsible for the observed changes in gene expression. In the experiments described here, approximately 100-fold more mechanical loading was applied, and we detected a centrifugation-induced PGE2 release. Two separate NSAIDs completely inhibited the centrifuged induction of PGE2 synthesis while not affecting c-fos expression. Although there was a slightly higher PGE2 level in both the noncentrifuged and centrifuged indomethacin treated cells, indomethacin completely blocked the centrifugation induced PGE2 synthesis without blocking c-fos induction. We think that centrifugation induction of c-fos is not dependent on PGE2 for three reasons. 1) We have previously shown that c-fos can be induced in the absence of new PGE2 production (9) ; 2) it takes approximately 20 min to synthesize the prostaglandins measured; after synthesis, prostaglandins require 20–25 min to maximally up-regulate c-fos; and 3) no increase in c-fos was seen in the noncentrifuged indomethacin cells, which had the same prostaglandin content as the treated centrifuged osteoblast. However, we believe that PGE2 plays a significant role in later osteoblast gene expression in untreated centrifuged cells and may be one factor in the extended period of c-fos induction seen 3 h after centrifugation.
Together these data suggest that centrifugation at 287 x g induces an increase in c-fos mRNA and a release of PGE2. The increase in c-fos expression is sensitive to a PKA inhibitor and insensitive to down-regulation of PKC activity, suggesting that induction is mediated by PKA and not TPA-sensitive PKC. Nevetheless, the involvement of additional signaling mechanisms cannot be ruled out. Although we have previously shown that increased intracellular cAMP levels can induce c-fos and that PGE2 can stimulate the cAMP signaling pathway, our data suggest that the centrifugation-induced new synthesis of PGE2 is not the only stress activated pathway that can induce early c-fos expression. These findings demonstrate that mechanical stress not only induces PGE2, which is known to independently induce c-fos, but also induces c-fos independent of new PGE2 synthesis, suggesting two independent pathways by which mechanical stress induces c-fos expression.
This work suggests that researchers should practice care when isolating any cells where centrifugation is used prior to RNA isolation, since the centrifugation itself can cause major changes in gene expression.
This work was supported by NASA grant NAG-2-1086 and by the Department of Veteran's Affairs, VAMC, San Francisco.
1 Correspondence: Laboratory of Cell Growth, Department of Medicine, Veterans Affairs Medical Center, Mail code 151F, 4150 Clement St., San Francisco, CA, 94121, USA. E-mail: firstname.lastname@example.org
2 Current address: Orthopaedic Molecular Biology Research Unit, Department of Paediatrics, University of Melbourne, Royal Children's Hospital, Parkville, Victoria, 3052, Australia.
3 Abbreviations: AP, alkaline phosphatase; FCS, fetal calf serum; aMEM, minimum essential medium alpha; MMLV, Moloney murine leukemia virus; NSAID, nonsteroidal anti-inflammatory drug; PCR, polymerase chain reaction; PG, prostaglandin; PKC, protein kinase C; RT, reverse transcriptase; TPA, 12-O-tetradecanoylphorbol 13-acetate.
Received for publication June 26, 1998. Revision received November 3, 1998.
|HOME||HELP||FEEDBACK||SUBSCRIPTIONS||ARCHIVE||SEARCH||TABLE OF CONTENTS|