Bisphosphonates inhibit bone resorption by reducing osteoclastic cell number and activity. Alendronate is a nitrogen-containing bisphosphonate analog used in the treatment of postmenopausal osteoporosis. The effects of alendronate in osteoclasts are well documented; however, there is limited information on the actions of alendronate in osteoblasts (Ob's). In this study, we investigated the effects of alendronate at concentrations of 1-100 μM on the synthesis of collagenase 3 or matrix metalloproteinase 13 (MMP-13) and tissue inhibitors of MMPs (TIMPs) 1, 2, and 3 in primary Ob-enriched cells from 22-day-old fetal rat calvariae. Alendronate at concentrations higher than 10 μM markedly stimulated the synthesis of collagenase messenger RNA (mRNA) and immunoreactive protein in Ob's. Alendronate did not stimulate the transcriptional rate of the collagenase 3 gene. However, in transcriptionally arrested cells, alendronate prolonged the half-life of collagenase transcripts. Alendronate did not alter the expression of TIMP 1 and 2, but modestly stimulated the expression of TIMP 3. The actions of alendronate in Ob's suggest potential additional effects in bone remodeling.
BISPHOSPHONATES ARE a class of drugs that are analogs of inorganic pyrophosphates with a high affinity for bone mineral. (1–3) Currently, several bisphosphonate analogs are used for the treatment of different metabolic bone disorders. Alendronate is one of the most potent nitrogen-containing bisphosphonates and it is widely used for the treatment of postmenopausal osteoporosis. Bisphosphonates, similar to inorganic pyrophosphates, can interfere with the formation, aggregation, and solubility of calcium phosphate crystals.(4) Thus, these agents can reduce mineralization, probably by impairing the calcification process.(5) Bisphosphonates inhibit bone resorption in vitro and in vivo by inhibiting the activity of osteoclastic cells. (6–9) After the internalization of bisphosphonates, osteoclasts undergo disorganization of their cytoskeletal structures, such as the actin ring, and lose the characteristic ruffled borders.(3,10) In addition to impairing osteoclastic activity, the uptake of bisphosphonates eventually leads to osteoclastic cell death by apoptosis.(11,12) Bisphosphonates also may decrease osteoclast formation by interfering with cell-cell fusion of osteoclastic precursors.(13,14) In contrast to the effects on bone resorption, the effects of bisphosphonates on osteoblasts (Ob's) and the aspects of bone formation are not well established.
Regulators of bone remodeling may affect the function of Ob cells, and agents that inhibit bone resorption may affect parameters of bone formation.(15) The major organic component of the bone matrix is type I collagen, and its synthesis and degradation are regulated by hormones, growth factors, and cytokines acting on the Ob. Collagenases are matrix metalloproteinases (MMPs) that collectively can degrade all the components of the extracellular matrix.(16) Collagenase 1 or MMP-1 is synthesized by cells of mesenchymal origin, including stimulated human Ob-like cells. Collagenase 2, or MMP-8, is synthesized primarily by neutrophils and collagenase 3, or MMP-13, is expressed by chondrocytes, synovial fibroblasts, and human and rodent Ob's. (17–19) Collagenase has been implicated in bone and cartilage remodeling and in the process of bone resorption.(18,20) Bone cells also synthesize other MMPs and tissue inhibitors of MMPs (TIMPs).
In this study, we investigated the effects of alendronate in primary cultures of rat Ob's from 22-day-old fetal rat calvariae (Ob cells) on the expression of collagenase 3 and TIMPs 1, 2, and 3.
MATERIALS AND METHODS
Twenty-two-day-old rat fetuses were removed from pregnant mothers and killed by blunt trauma to the nuchal area according to a protocol approved by the Animal Care and Use Committee at Saint Francis Hospital and Medical Center. Ob cells were isolated from parietal bone from 22-day-old fetal rats as described.(21) Cells were plated at a density of 10,000 cells/cm2 in Dulbecco's Modified Eagle's medium (DMEM) supplemented with nonessential amino acids, 100 μg/ml L-ascorbic acid, 20 mM HEPES (all from Life Technologies, Grand Island, NY, U.S.A.), and 10% fetal bovine serum (HyClone Laboratories, Logan, UT, U.S.A.) and cultured at 37°C in a CO2 incubator. At confluence, medium was replaced with serum-free DMEM for 16-24 h. Cells were incubated further with serum-free medium in the presence and absence of alendronate (kindly provided by Merck, West Point, PA, U.S.A.), which was dissolved in DMEM. 5,6-Dichlorobenzimidazole riboside (DRB; Sigma Chemical Co., Saint Louis, MO, U.S.A.) was dissolved in ethanol and diluted 1:200 in DMEM; control and treated cultures contained equal amounts of alcohol. At the completion of the culture period, the medium was collected for Western blot analysis, and Ob cells were harvested to isolate RNA for Northern blot analysis and nuclei for nuclear runoff assay.
Northern blot analysis
Total RNA was isolated from Ob cells by using the RNeasy columns (Qiagen, Santa Clarita, CA, U.S.A.) according to the manufacturer's instructions. Total RNA (10-15 μg/lane) was fractionated on a 1% agarose-formaldehyde gel (Life Technologies) containing 100 μg/ml ethidium bromide as described.(22) Subsequent to electrophoresis, RNA was transferred onto a nylon membrane by capillary action. The integrity and equal gel loading of RNA and the efficiency of transfer were assessed by visualizing the 28S and 18S ribosomal RNA (rRNA) bands under UV light. The RNA was cross-linked to the nylon membrane using CL-1000 UV cross-linker (UVP, San Gabriel, CA, U.S.A.) and hybridized with radiolabeled complementary DNA (cDNA). The cDNA fragments were isolated by restriction endonuclease (New England Biolabs, Beverley, MA, U.S.A.) digestion of plasmid clones containing rat interstitial collagenase cDNA (kindly provided by Dr. Cheryl Quinn, St. Louis University School of Medicine, St. Louis, MO, U.S.A.); murine TIMP 1, TIMP 2, and TIMP 3 cDNAs (all kindly provided by Dr. Dylan Edwards, University of Calgary Health Sciences Center, Calgary, Canada); rat glyceraldehyde-3-phosphate dehydrogenase (GAPD) cDNA (kindly provided by Dr. Ray Wu, Cornell University, Ithaca, NY, U.S.A.); and mouse 18S rRNA cDNA (American Type Culture Collection, Rockville, MD, U.S.A.). (23–28) The cDNAs were radiolabeled by random hexanucleotide primed second-strand synthesis method using [α-32P]deoxyadenosine triphosphate (ATP), [α-32P]deoxycytosine triphosphate (CTP; 3000 Ci/mmol; DuPont, Wilmington, DE, U.S.A.), Klenow fragment (New England Biolabs), and hexanucleotide primers (Boehringer Mannheim Biochemicals, Indianapolis, IN, U.S.A.). The final low-stringency wash was performed at 55°C in 0.15 M sodium chloride, 0.015 M sodium citrate, pH 7 (1× SSC)/0.1% sodium dodecyl sulfate (SDS) for collagenase and TIMP probes, and at 65°C in 0.1× SSC/0.1% SDS for 18S probe. Autoradiography was performed by exposing the membranes to Kodak XAR film (Eastman Kodak, Rochester, NY, U.S.A.) in the presence of intensifying screens. The intensities of RNA bands were quantitated by densitometric scanning of the autoradiographs.
Nuclear runoff assay
Nuclei were isolated from Ob cells after homogenization using a Dounce homogenizer in Tris buffer containing 0.5% NP-40.(29) Nascent transcripts were radiolabeled by incubation of nuclei at room temperature for 30 minutes in a reaction buffer containing 250 μCi (800 μCi/mmol) of [α-32P]uridine triphosphate; 500 μM ATP, CTP, and guanosine triphosphate (GTP); and 150 U RNasin (Promega Corp., Madison, WI, U.S.A.). [32P]RNA was isolated by treatment with deoxyribonuclease I and proteinase K, followed by phenol-chloroform extraction and ethanol precipitation using ammonium acetate. Linearized plasmid DNA containing 1 μg of cDNA was immobilized onto a nylon membrane using a slot blot apparatus (Schleicher & Schuell, Keene, NH, U.S.A.). Equal counts per minute of [32P]RNA from each sample were hybridized to immobilized cDNAs at 42°C for 72 h and washed in 1× SSC/0.1% SDS at 55°C. Hybridization of nascent transcripts to different cDNAs was visualized by autoradiography and quantified by densitometry.
Western immunoblot analysis
Aliquots from culture medium were adjusted to a final concentration of 0.1% polyoxyethylene sorbitan monolaurate (Pierce Chemical Co., Rockford, IL, U.S.A.), fractionated by polyacrylamide gel electrophoresis at denaturing conditions, and transferred onto an Immobilon P membrane (Millipore Corp., Bedford, MA, U.S.A.). After blocking with 2% bovine serum albumin, the membrane was exposed to a 1:1000 dilution of rabbit antiserum raised against rat interstitial collagenase (kindly provided by Dr. John J. Jeffrey, Albany Medical College, Albany, NY), followed by the addition of goat anti-rabbit immunoglobulin G (IgG) conjugated to horseradish peroxidase.(30) The blots were washed and developed with a horseradish peroxidase chemiluminescence detection reagent (DuPont). The chemiluminescent bands were visualized after exposure to DuPont reflection film employing refection intensifying screen (DuPont).
Data on collagenase mRNA decay were analyzed by linear regression, and the slopes of the regression lines obtained for control and alendronate-treated cells were compared for significant differences as described by Sokal and Rohlf.(31)
Effect of alendronate on collagenase synthesis
Alendronate markedly enhanced the synthesis of collagenase 3 in Ob cells. The effect of alendronate was dose dependent, and alendronate at 10-100 μM for 6 h increased collagenase 3 messenger RNA (mRNA) in Ob cells (Fig. 1). Alendronate, at 100 μM, caused a 3- to 14-fold increase in collagenase mRNA after 2-24 h (Fig. 2). The secretion of procollagenase into the culture medium was examined by Western blot analysis, using a previously characterized rat collagenase antibody(30) (Fig. 3). In cultures exposed to alendronate at 100 μM, there was a 17-fold increase in the accumulation of procollagenase in the medium after 24 h. The identity of the procollagenase band was confirmed by comigration with purified procollagenase.
To investigate whether the collagenase 3 gene is regulated by alendronate at the transcriptional level, we examined the changes in the collagenase gene transcriptional rate by nuclear runoff assay after exposing Ob cells to alendronate at 100 μM for 1 h. Alendronate did not increase the rate of collagenase transcription (Fig. 4). To determine whether the increase in collagenase mRNA levels by alendronate was caused by changes in mRNA stability, we tested the rate of decay of collagenase mRNA after transcriptional arrest by adding the RNA polymerase II inhibitor DRB at 75 μM.(32) Ob cells were treated with control medium or 100 μM alendronate for 2 h before and up to 8 h after DRB exposure (Fig. 5). The levels of collagenase mRNA from control and alendronate-treated cultures were measured before and 2-8 h after transcriptional arrest to determine the rate of collagenase mRNA decay. The half-life of collagenase mRNA was ∼3 h in control cultures and it was estimated, by extrapolation, to be ∼10 h in alendronate-treated cultures, indicating that alendronate increases the stability of collagenase transcripts.
Effect of alendronate on synthesis of TIMPs
The regulation of TIMP 1, 2, and 3 transcripts was evaluated after exposure to alendronate at 100 μM for 2-24 h. Alendronate failed to modify the levels of TIMP 1 and 2 mRNAs (Fig. 6). Alendronate did not affect the levels of TIMP 3 mRNA at 2 h and 6 h (data not shown), but it modestly stimulated the levels of TIMP 3 mRNA after 24 h (Fig. 7).
In this study, we show that alendronate stimulates the expression of collagenase 3 in primary Ob-enriched cells from fetal rat calvaria, a widely used cell culture model of Ob's. Because bisphosphonates are potent therapeutic agents used to reduce bone loss, the observation that alendronate caused a dramatic increase in the synthesis of collagenase 3, a matrix degrading protease, was surprising. Although we could only detect the procollagenase form in the cell culture medium, it is likely that the pro-form is converted into active collagenase by activators of procollagenase 3, such as gelatinase A and membrane type MMP-1 that are present in the bone microenvironment.(33) Active collagenase 3 can modulate aspects of both bone resorption and bone formation. In previous studies, a number of stimulators of bone resorption such as parathyroid hormone (PTH), interleukin (IL)-1, and IL-6 were shown to increase collagenase 3 expression in Ob's.(23, 34, 35) In contrast, inhibitors of bone resorption and stimulators of bone formation such as bone morphogenetic proteins (BMPs) inhibited collagenase 3.(19) Holliday et al. indicated the involvement of collagenase 3 in bone resorption by showing that osteoclastic bone resorption can be inhibited by a neutralizing antibody and inhibitors that interfere with collagenase activity.(20) A recent study by Zhao et al. showed that the PTH-mediated bone resorption is reduced in transgenic animals expressing type I collagen that is resistant to cleavage by collagenase.(36) Although the precise role of collagenase in bone resorption is not established, it has been proposed that the collagen fragments generated by collagenase 3 activate or increase the viability of osteoclasts by binding to specific integrin receptors.(20) Thus, it is possible that alendronate-mediated stimulation of collagenase 3 can generate collagen fragments that can cause an increase in the number of osteoclasts at bone remodeling sites. Subsequently, these osteoclasts may become susceptible to the direct cytotoxic effects of alendronate sequestered in the skeletal tissue leading to an overall decrease in bone resorption.(3, 10–12) Collagenase 3 increase by alendronate also may promote bone formation by enhancing the stimulatory effects of insulin-like growth factors (IGFs) in bone, because collagenase 3 can fragment IGF-binding protein 5 and these fragments may stimulate bone cell growth.(37) Further studies are needed to determine precisely how collagenase 3 stimulation by alendronate in Ob's contributes to the collective action of this bisphosphonate in bone.
Alendronate regulates the expression of collagenase 3 by posttranscriptional mechanisms. Unlike most regulators of collagenase 3 expression in Ob's such as hormones, growth factors, and cytokines, there are no known specific receptors for bisphosphonates in Ob's. Therefore, it is difficult to envision a classical receptor-mediated mechanism of regulation of collagenase 3 expression by alendronate in Ob's. It is likely that the effects of alendronate on collagenase and possibly other genes are caused by modification of important intracellular regulatory proteins. (38–40) In a recent study, alendronate has been shown to increase phosphorylation of extracellular signal-regulated kinases (ERKs), mediators of several signal transduction pathways, in Ob's.(40) We speculate that RNA regulatory proteins mediating collagenase mRNA stability are among the downstream targets of ERKs or other cellular regulatory proteins modified by alendronate.
In addition to regulating collagenase 3 expression, alendronate stimulates TIMP 3 expression modestly. In our previous studies, we observed that stimulators of bone formation such as BMP-2 and transforming growth factor β increase expression of TIMP 3.(19,41) The stimulation of TIMP 3 by alendronate may contribute to the role of alendronate in reducing bone loss, possibly by limiting the activity of MMPs in bone matrix degradation. Because the effect of alendronate on collagenase 3 stimulation is greater than that on TIMP 3, there could be residual collagenase activity leading to an increase in collagen breakdown. Alternatively, an increase in collagenase expression may be important for enhanced bone turnover and formation, as it was demonstrated for membrane type 1 MMP (MTI-MMP).(42) Animals deficient in MT1-MMP develop osteopenia because of inadequate collagen turnover. Similarly, increased collagenase expression by Ob's may be necessary for bone collagen turnover and remodeling. Thus, it is conceivable that the stimulatory effects of alendronate on bone mineral density are related to its effects on collagenase 3 expression.(43,44) Another possible mechanism for the increased bone mass may include additional actions of alendronate on the Ob. In experimental models of glucocorticoid-induced osteopenia, alendronate opposed the effects of glucocorticoids on apoptosis of Ob's and osteocytes preserving a bone-forming cell population.(40) It is important to note that the concentrations of alendronate found to increase collagenase 3 are about two orders of magnitude higher than those that can be achieved in serum after the oral administration of alendronate at doses used for the treatment of postmenopausal osteoporosis.(45,46) However, bisphosphonates have high affinity for the skeletal tissue and can selectively accumulate in bone after long-term treatment for skeletal disorders. Therefore, it is possible that the concentration of a bisphosphonate in the bone microenvironment may reach a level sufficient to stimulate collagenase 3 and TIMP 3, as shown in this study.
In summary, alendronate causes a marked increase in the synthesis of collagenase 3 and a modest increase in the synthesis of TIMP 3. These effects of alendronate in Ob's may contribute to its overall actions in skeletal tissue.
The authors thank Dr. Cheryl Quinn for rat collagenase 3 cDNA; Dr. Dylan Edwards for murine TIMP 1, 2, and 3 cDNAs; Dr. Ray Wu for rat GAPD cDNA; and Dr. John Jeffrey for rat procollagenase antibody. The authors also thank Susan Bankowski and Susan O'Lone for expert technical assistance. This work was supported by a medical school grant from Merck & Co., Inc.