Promoter-dependent and -independent activation of insulin-like growth factor binding protein-5 gene expression by prostaglandin E2 in primary rat osteoblasts.

Insulin-like growth factor (IGF) action is mediated by high affinity cell surface IGF receptors and modulated by a family of secreted IGF binding proteins (IGFBPs). IGFBP-5, the most conserved of six IGFBPs characterized to date, uniquely potentiates the anabolic actions of IGF-I for skeletal cells. In osteoblasts, IGFBP-5 production is stimulated by prostaglandin E2 (PGE2), a local factor that mediates certain effects induced by parathyroid hormone, cytokines such as interleukin-1 and transforming growth factor-beta, and mechanical strain. In this study, we show that transcriptional and post-transcriptional events initiated by PGE2 collaborate to enhance IGFBP-5 gene expression in primary fetal rat osteoblast cultures. PGE2 treatment stimulated up to a 7-fold rise in steady-state levels of IGFBP-5 mRNA throughout 32 h of incubation. Analysis of nascent IGFBP-5 mRNA suggested that PGE2 had only a modest stimulatory effect on IGFBP-5 gene transcription, and transient transfection studies with IGFBP-5 promoter-reporter genes confirmed that PGE2 enhanced promoter activity by approximately 2-fold. Similar stimulatory effects were seen with forskolin. A DNA fragment with only 51 base pairs of the 5'-flanking sequence retained hormonal responsiveness, which may be mediated by a binding site for transcription factor AP-2 located at positions -44 to -36 in the proximal IGFBP-5 promoter. Incubation of osteoblasts with the mRNA transcriptional inhibitor 5,6-dichloro-1-beta-D-ribofuranosylbenzimidazole demonstrated that PGE2 enhanced IGFBP-5 mRNA stability by 2-fold, increasing the t1/2 from 9 to 18 h. The effects of PGE2 on steady-state IGFBP-5 transcripts were abrogated by preincubating cells with cycloheximide, indicating that the effects of PGE2 on both gene transcription and mRNA stability required ongoing protein synthesis. Therefore, both promoter-dependent and -independent pathways converge to enhance IGFBP-5 gene expression in response to PGE2 in osteoblasts.

Insulin-like growth factors-I (IGF-I) 1 and -II are abundant locally produced growth regulators in skeletal tissue (1). While the synthesis of IGFs is hormonally controlled, their actions are ultimately determined by way of signal-transducing receptors. IGF binding proteins (IGFBPs) are a family of secreted proteins that also avidly bind IGFs and modify their actions by altering their access to cell surface receptors (2,3). Of the six known IGFBPs, fetal rat osteoblasts synthesize five, IGFBP-2, -3, -4, -5, and -6 (4). However, IGFBP-5 is the only IGFBP with a demonstrated ability to potentiate IGF actions in bone cells (5). Therefore, agents that stimulate IGFBP-5 synthesis may have an important anabolic role in skeletal growth and the maintenance of skeletal integrity.
The initial identification of IGFBP-5 was based upon its ability to augment IGF-I activity in bone cell cultures, as well as its structural and sequence similarities to other IGFBPs (6 -9). The mechanism of potentiation of IGF action by IGFBP-5 has been attributed to its association within pericellular compartments (cell membrane and matrix components), resulting in a high local concentration of IGFs in close proximity to cell surface IGF receptors (5,10,11). Recent studies suggest the ability of IGFBP-5 to bind ligand decreases after its association with matrix or select polysaccharides and the actions of IGFBP-5-selective proteases (12)(13)(14)(15). The subsequent release of highly concentrated IGF in the pericellular environment thus enhances IGF receptor binding and its biological effects. IGFBP-5 has the additional attribute of high affinity binding to the calcium phosphate component of bone, which may serve to concentrate IGFs in inorganic bone matrix for storage and subsequent activation during periods of localized bone resorption (8).
In bone cell cultures, both IGFBP-5 and IGF-I synthesis are regulated by prostaglandin E 2 (PGE 2 ), parathyroid hormone (PTH), other agents that stimulate cAMP synthesis, or by cAMP itself (4,16). PGE 2 is produced by osteoblasts in response to PTH, to cytokines such as interleukin-1 and transforming growth factor-␤, and to mechanical strain, and PGE 2 has been shown to mediate various biological actions on osteoblasts initiated by these stimuli (17)(18)(19)(20)(21)(22). PGE 2 may thus serve as a local analog of PTH. Unlike PTH, however, its influence may be more highly focused due to its synthesis within the skeleton. Furthermore, PGE 2 has demonstrated anabolic actions in bone that depend on the cellular state of differentiation and on dose and duration of treatment (23)(24)(25).
While the molecular mechanisms by which PGE 2 regulates IGF-I synthesis in osteoblasts are only currently being eluci-dated (26,27), even less is known about IGFBP-5 synthesis in skeletal tissue. Previous observations reveal that PGE 2 enhances steady-state levels of IGFBP-5 mRNA (4). We now demonstrate that transcriptional and post-transcriptional pathways are both activated by PGE 2 to stimulate IGFBP-5 gene expression. These steps appear to require ongoing protein synthesis, indicating that promoter-dependent and -independent processes each regulate IGFBP-5 production in response to PGE 2 in primary fetal rat osteoblast-enriched cultures.

EXPERIMENTAL PROCEDURES
Cell Cultures-Primary osteoblast-enriched cell cultures were prepared from parietal bones obtained from 22-day-old Sprague-Dawley rat fetuses (Charles River Laboratories, Raleigh, NC). Rats were housed and euthanized by methods approved by the Yale University Animal Care and Use Committee. Cranial sutures were removed by dissection, and the bones were digested with collagenase for five sequential 20-min intervals. The cells released during the last three digestions exhibit biochemical charactertistics associated with differentiated osteoblasts, including high levels of alkaline phosphatase, PTH receptors, type I collagen synthesis, and a rise in osteocalcin expression in response to dihydroxyvitamin D 3 (28,29). Histochemical staining demonstrates approximately 80% of the cells express alkaline phosphatase, 2 although this itself cannot be considered entirely specific for osteoblasts. However, using these criteria, as well as differential sensitivity to transforming growth factor-␤, bone morphogenetic protein-2, various prostaglandins, and the ability to form mineralized nodules in vitro (30,31), these cells are well distinguished from the less differentiated cells released during earlier collagenase digestions. Cells obtained from the last three digestions were then plated at 9,400/cm 2 in Dulbecco's modified Eagle's medium containing 20 mM HEPES (pH 7.2), 0.1 mg/ml ascorbic acid, penicillin, and streptomycin (all from Life Technologies, Inc.) and 10% fetal bovine serum (Sigma). To examine IGFBP-5 transcript stability, 5,6-dichloro-1-␤-D-ribofuranosylbenzimidazole (DRB, Sigma) was added to the cultures at a final concentration of 75 M. Cycloheximide (Sigma) was used at a final concentration of 2 M, and its use preceded other treatments by 15 min to assure its effectiveness prior to vehicle or PGE 2 treatments.
Plasmids-Rat IGFBP-5 cDNA was kindly provided by Drs. S. Shimasaki and N. Ling. Murine IGFBP-5 promoter constructs have been described previously (32). Plasmids were propagated in Escherichia coli strain DH5␣ with ampicillin selection and were prepared by a modification of the alkaline extraction method (33,34).
RNA Analysis-Cultures of 9.6 cm 2 were solubilized in buffer consisting of 5 M guanidine monothiocyanate, 25 mM trisodium citrate, 0.5% Sarkosyl, and 0.1 M 2-mercaptoethanol, followed by extraction with phenol/chloroform/isoamyl alcohol (75:25:1) in the presence of 0.2 M sodium acetate (35). Total RNA was precipitated, ethanol washed, dried, and resuspended in diethylpyrocarbonate-treated water, and concentration and purity were determined by absorbance at 260 and 280 nm. Fifteen micrograms of RNA was denatured with 2.2 M formaldehyde, 12.5 M formamide at 65°C for 15 min and fractionated on a 1.5% agarose, 2.2 M formaldehyde gel. Co-electrophoresed RNA standards were excised and ethidium bromide stained, and the remaining gel was blotted onto charged modified nylon (GeneScreen Plus, DuPont NEN). A restriction fragment containing the rat IGFBP-5 cDNA clone was purified from an agarose gel and labeled with [␣-32 P]deoxycytidine triphosphate and [␣-32 P]thymidine triphosphate by random hexanucleotide primed second strand synthesis (36). Northern blots were hybridized with [ 32 P]IGFBP-5 cDNA, and the filters were washed under conditions of progressively increasing stringency. Final washes were with 0.2 ϫ SSC (20 ϫ SSC contains 3 M NaCl, 0.3 M trisodium citrate, pH 7.0) and 0.1% sodium dodecyl sulfate for 1 h at 55°C. The bound radioactive material was visualized by autoradiography using Amersham Hyperfilm ® and a DuPont Cronex intensifying screen. Filters were eluted of specifically bound 32 P-labeled cDNA by washing in deionized water for 5 min at 100°C before probing with 18 S antisense ribosomal RNA (Ambion, Austin, TX). RNase protection assays were performed with 5 g of total RNA and a ␣-32 P-labeled antisense RNA probe complementary to mouse IGFBP-5 exon 3 and the adjacent 3Ј intron, as described (37). Protected fragments were separated by electrophoresis on 6% polyacrylamide, 8.3 M urea gels and visualized by autoradiography, and the results were quantitated with a Betascope 603 betascanner (Betagen, Waltham, MA).
Transfection Studies-IGFBP-5 promoter-luciferase reporter plasmids (1.0 -1.5 g/9.6-cm 2 culture well) were co-transfected with a vector carrying the ␤-galactosidase gene under SV40 promoter control (1.0 g/culture well; pSV-␤-Galactosidase Control Vector, Promega Corp.) to normalize for transfection efficiency. Cultures at 75% confluent density were rinsed in serum-free medium and exposed to plasmids in the presence of Lipofectin (Life Technologies, Inc.) for 3 h. The solution was then replaced with growth medium containing 5% fetal bovine serum, and the cultures were grown to confluence (48 h). Confluent cultures were rinsed with serum-free medium and treated for 6 or 24 h with vehicle (ethanol diluted 1/1000 or greater), PGE 2 , or forskolin (both from Sigma). At the end of the treatment interval, the medium was aspirated, and cultures were rinsed with phosphate-buffered saline and then lysed in 100 l of 25 mM Tris phosphate (pH 7.8), 2 mM dithiothreitol, 2 mM 1,2-diaminocyclohexane-N,N,NЈ,NЈ-tetraacetic acid, 10% glycerol, 1% Triton X-100 (cell lysis buffer, Promega Corp.). Nuclei were pelleted at 12,000 ϫ g for 5 min, and the supernatants were stored at Ϫ75°C until assay. Commercial kits were used to measure luciferase (Promega Corp.) and ␤-galactosidase (Tropix, Bedford, MA). Protein was determined by the Bradford assay (38).
Statistical Analysis-When statistical analysis was conducted, data were assessed by one-way analysis of variance, using Kruskal-Wallis or Bonferonni methods for post hoc analysis. RESULTS PGE 2 increases the level of IGFBP-5 mRNA and protein in primary fetal rat osteoblast-enriched cultures (4). The magnitude of the rise in transcript abundance is time-dependent. In agreement with earlier studies (4), IGFBP-5 mRNA levels increased by 1.4 -7-fold following 4 -32 h of PGE 2 (1 M) treatment (Fig. 1). A subsequent analysis by RNase protection assay confirmed these observations and additionally demonstrated a time-dependent rise in abundance of nascent transcripts, first observed at 3 h after PGE 2 treatment, with a ϳ3-fold increase seen at 8 h (data not shown). These results suggest that PGE 2 may stimulate IGFBP-5 transcription, although this effect appears delayed when compared with another PGE 2 -stimulated gene, IGF-I, for which nascent transcripts increase within 30 min of PGE 2 treatment (26).
To investigate potential transcriptional mechanisms influenced by PGE 2  cultures, and analyzed for reporter gene expression 48 h later. As shown in Fig. 2, IGFBP-5 promoter constructs with 1406, 1004, and 156 bp (IGFBP5-Luc3, -Luc4, and -Luc5) of 5Ј-flanking DNA directed comparably high luciferase activity (100, 95.6, and 84.3%, relative to IGFBP5-Luc3). In contrast, fusion plasmids with 75 bp or less of 5Ј-flanking DNA (IGFBP5-Luc6, -Luc7, and -Luc8) had progressively diminished activity (22.9 to 5.9%, relative to IGFBP5-Luc3). Luciferase expression for a positive control viral promoter-driven construct, pGL2-Control, was included for comparison. These data indicate that regions between Ϫ156 and Ϫ51 bp contain cis-acting regulatory element(s) needed for high level basal promoter activity in osteoblast cultures, analogous to the areas that we defined in Hep G2 (hepatocyte) and C2I (myoblast) cell lines (32,37). Two recombinant plasmids with longer 5Ј-flanking segments of 4100 and 3000 bp (IGFBP5-Luc1 and -Luc2) directed lower luciferase activity, suggesting the presence of inhibitory elements in the 5Ј-flanking region of the promoter. Promoter activity was orientation-specific. When the 1004-bp promoter fragment was inserted into the luciferase vector in the reversed orientation (IGFBP5-Luc4 rev, Fig. 2), reporter expression was minimal, being comparable with the promoterless pGL2-Basic parental plasmid.
Recombinant IGFBP-5 promoter-luciferase fusion constructs were next used to identify promoter elements that participated in PGE 2 -stimulated IGFBP-5 expression. While basal luciferase activity for IGFBP5-Luc2 through -Luc7 varied up to 14fold (Fig. 2), 6 h of PGE 2 treatment enhanced their ability to drive luciferase expression to a similar extent, ranging from 2.3-to 1.6-fold (Fig. 3). The shortest construct responsive to PGE 2 , IGFBP5-Luc 7, contained only 51 bp of 5Ј-flanking DNA. However, deletion of the next 20 bp (IGFBP5-Luc8) eliminated the effect of PGE 2 on reporter gene expression. Using IGFBP5-Luc 4, a comparable increase was seen after 24 h of PGE 2 treatment, and these effects were duplicated by treatment with forskolin (10 M), a strong inducer of adenylate cyclase activity (Fig. 4). Plasmids with internal deletions spanning Ϫ69 to Ϫ51 bp and Ϫ52 to Ϫ32 bp in the background of the very active IGFBP5-Luc 4 construct had diminished basal activity as shown previously (32). Importantly, however, each of these two deletion constructs clearly responded to PGE 2 treatment, although at modestly reduced levels (1.4-and 1.7-fold, respectively; Fig. 5).
In aggregate, these results show that the IGFBP-5 promoter is very active in primary osteoblast cultures but demonstrate that PGE 2 stimulated only a 2-fold increase in IGFBP-5 gene transcription, even after a 24-h incubation. This contrasts sharply with Northern blot data showing up to a 6-fold rise in steady-state IGFBP-5 mRNA levels at 24 h (Fig. 1). Thus, the discrepancy between maximal promoter activity and steadystate transcripts encoding IGFBP-5 in response to PGE 2 indicates the participation of transcriptional and post-transcriptional mechanisms in regulating IGFBP-5 gene expression. Therefore, the RNA polymerase II selective inhibitor DRB was used to examine the influence of PGE 2 on IGFBP-5 mRNA stability. Cultures were first treated with vehicle alone or PGE 2 (1 M) for 24 h, followed by DRB (75 M) to arrest gene transcription. PGE 2 treatment caused a 2-fold rise in the halflife of IGFBP-5 mRNAs; vehicle-treated control cultures had a t1 ⁄2 ϭ 9 h, while cultures pretreated with PGE 2 had t1 ⁄2 ϭ 18 h (Fig. 6). Consequently, PGE 2 enhanced the stability of IGFBP-5 transcripts. To explore further the mechanisms involved in induction of IGFBP-5 mRNA after PGE 2 treatment, we examined the effect of the protein synthesis inhibitor, cycloheximide. At a dose of 2 M, cycloheximide blocked Ͼ90% of ongoing protein synthesis, as measured by incorporation of 3 [H]proline into trichloroacetic acid-precipitable material in PGE 2 -treated cultures (Ϫcycloheximide, 12.6 Ϯ 0.9 ϫ 10 3 cpm versus ϩcyclo-

FIG. 2. Identifying regions controlling basal IGFBP-5 promoter function in osteoblast-enriched cultures.
Various IGFBP-5 promoterluciferase reporter plasmids (depicted in the left panel) were co-transfected with pSV-␤-galactosidase control vector into osteoblast-enriched cultures (9.6 cm 2 ) using Lipofectin. Cultures were grown to confluence (48 h), the growth medium was aspirated, and the cultures were rinsed with serum-free Dulbecco's modified Eagle's medium. Cultures were exposed to control medium (containing ethanol vehicle) for 6 h. Cytoplasmic extracts were prepared, and luciferase activity was determined as described under "Experimental Procedures." Data were corrected for transfection efficiency (␤-galactosidase expression) and for protein content of cytoplasmic extracts. Transfections were performed in duplicate or triplicate, and results are pooled data for three or more separate experiments for a total of eight or more replicate cultures. The mean Ϯ S.E. for pooled experiments is shown. Luciferase activity was determined by single channel photon counting, and background levels were ϳ930 cpm/g of protein.
Control transfections included an SV40 promoter/enhancer-luciferase reporter plasmid (pGL2-Control) and promoterless pGL-2 Basic (both from Promega Corp.). Numbers on the far right in parentheses indicate the percent maximal luciferase expression determined from the mean value for each construct, as compared with the mean value for IGFBP5-Luc3, which has been set at 100%. heximide, 1.3 Ϯ 0.1 ϫ 10 3 cpm). Cycloheximide also completely inhibited the induction of IGFBP-5 mRNA after PGE 2 treatment (Fig. 7). These results contrast with the lack of an effect of cycloheximide on the rise in IGF-I transcript abundance following PGE 2 treatment, 3 and demonstrate that inducible protein(s) contribute to the transcriptional and post-transcriptional effects of PGE 2 on IGFBP-5 gene expression in osteoblast-enriched cultures. DISCUSSION IGFBP-5 expression is activated through cAMP-dependent pathways in osteoblasts and in other cell culture models (4,16,39). While IGFBP-5 transcripts accumulate in response to PGE 2 and other agents that elevate cAMP, little is known about the mechanisms of hormone-induced IGFBP-5 synthesis in osteoblasts. We now present data demonstrating stimulation of IGFBP-5 promoter activity by PGE 2 and show that PGE 2 also enhances IGFBP-5 mRNA stability. These results indicate that promoter-dependent and -independent mechanisms function together to regulate IGFBP-5 gene expression.
Unstimulated fetal rat osteoblasts synthesize IGFBP-5 mRNA and protein (4,40,41). Results from transient transfection experiments confirm our earlier studies that near-maximal basal promoter activity resides within the first 156 bp of 5Ј-flanking DNA and that over 20% of basal activity is controlled by the proximal 75 bp of the promoter (32). Similar to our earlier evidence, promoter function is attenuated by internal deletions that eliminate nucleotides Ϫ69 to Ϫ51 or Ϫ52 to Ϫ32, encoding segments that span a DNase I-footprinted re-

FIG. 3. Effect of PGE 2 on IGFBP-5 promoter activity following transient transfection into osteoblast-enriched cultures.
Various IGFBP-5 promoter-luciferase reporter plasmids (depicted in Fig. 2) were co-transfected with a pSV-␤-galactosidase control vector into osteoblast-enriched cultures (9.6 cm 2 ) using Lipofectin. Cultures were grown to confluence (48 h), the growth medium was aspirated, and the cultures were rinsed with serum-free Dulbecco's modified Eagle's medium. Cultures were exposed to control medium (containing vehicle) or PGE 2 (1 M) for 6 h. Cytoplasmic extracts were prepared and luciferase activity was determined. Data are corrected for transfection efficiency (␤-galactosidase expression) and for protein content of cytoplasmic extracts. Transfections were performed in duplicate or triplicate, and results are pooled data for three or more separate experiments and for a total of eight or more replicate cultures. The mean Ϯ S.E. for luciferase expression compared with vehicle-treated cultures (-fold stimulation) for pooled experiments is shown. PGE 2 caused a statistically significant elevation in luciferase expression (p Ͻ 0.05 versus pGL-2 Basic) for IGFBP-5 promoter reporter plasmids IGFBP5-Luc2, -Luc3, -Luc4, -Luc5, -Luc6, and -Luc7.

FIG. 4. Forskolin and PGE 2 induce comparable increases in IGFBP-5 promoter activity in osteoblast-enriched cultures.
The IGFBP-5 promoter-luciferase reporter plasmid, IGFBP5-Luc4, was cotransfected with a pSV-␤-galactosidase control vector into osteoblastenriched cultures using Lipofectin, as described in Fig. 3 and under "Experimental Procedures." Cultures were grown to confluence (48 h), the growth medium was aspirated, and the cultures were rinsed with serum-free Dulbecco's modified Eagle's medium. Cultures were exposed to control medium (containing vehicle), forskolin at 10 M, or PGE 2 at 1 M for 6 or 24 h, as indicated. Cytoplasmic extracts were prepared and luciferase activity was determined. Data are corrected for ␤-galactosidase expression and for protein content of cytoplasmic extracts. Transfections were performed in duplicate or triplicate, and results are representative data for three separate experiments and for a total of eight or more replicate cultures. The mean Ϯ S.E. for luciferase expression is shown.

FIG. 5. Effect of PGE 2 on the activity of mutant IGFBP-5 promoter following transient transfection into osteoblast-enriched cultures.
IGFBP-5 promoter-luciferase reporter plasmids having deletion mutations spanning Ϫ69 to Ϫ51 bp, or Ϫ52 to Ϫ32 bp, in the IGFBP5-Luc 4 background (IGFBP5-Luc4⌬a and IGFBP5-Luc4⌬b, respectively) (32) were co-transfected with an pSV-␤-galactosidase control vector into osteoblast-enriched cultures using Lipofectin, as described under "Experimental Procedures" and in the legend to Fig. 2. Cultures were exposed to control medium (containing vehicle) or PGE 2 (1 M) for 6 h. Cytoplasmic extracts were prepared and luciferase activity was determined. Data are corrected for transfection efficiency (␤-galactosidase expression) and for protein content of cytoplasmic extracts. Transfections were performed in duplicate or triplicate, and results are pooled data for three or more separate experiments and for a total of eight or more replicate cultures. The relative effect of PGE 2 compared to vehicletreated cultures (-fold stimulation) is shown for pooled experiments. gion identified with Hep G2 nuclear protein extracts (32).
In transient transfections of osteoblasts, PGE 2 treatment for 6 or 24 h increased luciferase activity driven by the IGFBP-5 promoter 1.6 -2.3-fold in constructs containing as little as 51 bp or as much as 3000 bp of 5Ј-flanking DNA. The shortest promoter fragments mediating PGE 2 -induced gene transcription do not contain a consensus cAMP response element. However, a potential binding site for transcription factor AP-2 is present between nucleotides Ϫ44 and Ϫ36, and at least six AP-2 sites are dispersed throughout the 3000 bp of the IGFBP-5 promoter. The apparent decline in response to PGE 2 treatment seen with recombinant plasmids having progressively shorter promoter segments may reflect loss of individual AP-2 binding sites or other potential cAMP-responsive cis-elements.
While this paper was in preparation, Duan and Clemmons (42) reported the involvement of AP-2 in basal and cAMPmediated IGFBP-5 transcription in human dermal fibroblast cell lines. In their study, forskolin stimulated IGFBP-5 promoter activity ϳ2.8-fold, a result similar in magnitude to our observation in osteoblasts. They identified an AP-2 site within nucleotides Ϫ55 to Ϫ36 in the human IGFBP-5 promoter as the key hormone response element (42). An identical AP-2 site is present in a comparable location in the murine promoter, as noted above. While deletion of this site in construct IGFBP5-Luc4⌬b reduced basal promoter activity, it caused only a modest decline in the effect of PGE 2 . Therefore, additional AP-2 sites or alternative cAMP response elements may be functional in osteoblasts.
The modest ϳ2-fold effect of PGE 2 on IGFBP-5 transcription does not account for the 6-fold increase in steady-state IGFBP mRNA seen following a 24-h incubation. As demonstrated here, PGE 2 also caused a doubling of IGFBP-5 transcript half-life, from ϳ9 to ϳ18 h. Of note, the t1 ⁄2 for IGFBP-5 mRNA in osteoblast-enriched cultures under basal conditions, ϳ9 h, is similar to the 11-12-h transcript half-life measured by us in C2I myoblasts (37) but differs somewhat from reported values of ϳ14 and ϳ20 h obtained under basal conditions in a similar osteoblast culture model (40,41), which may be accounted for by small differences in experimental design. Thus, both transcriptional and post-transcriptional effects of PGE 2 contribute to the induction of IGFBP-5 mRNA following hormone treatment. These dual actions on IGFBP-5 gene expression can be distinguished from transcriptional effects of PGE 2 on the IGF-I gene, which appear to be mediated through an element found in the proximal part of promoter 1, the major IGF-I gene promoter (27). In addition, the actions of PGE 2 to enhance IGFBP-5 gene expression require ongoing protein synthesis, since they were obliterated by preincubation with cycloheximide, while PGE 2 -stimulated IGF-I gene transcription occurs even in the absence of new protein synthesis. 4 Levels of IGFBP-5 in extracellular compartments are modulated not only by rates of gene expression and protein biosynthesis but also by post-translational mechanisms. The existence of IGFBP-5-selective proteases has been documented in a variety of cultured cells, including normal human osteoblasts (12)(13)(14)43), and IGF-mediated stabilization of IGFBP-5 abundance has been described in culture models derived from bone and other cell types (44). Since IGFBP-5 enhances the anabolic actions of IGF-I in bone cells (5), analysis of the multiple mechanisms involved in modifying IGFBP-5 availability within the skeleton should have direct impact in understanding how growth factors regulate skeletal cell metabolism. Confluent, serum-deprived cultures of osteoblasts were treated with control (C, vehicle) or PGE 2 (P, 1 M) for 6 h, in the absence or presence of 2 M cycloheximide. RNA blots were prepared as described under "Experimental Procedures," hybridized with 32 P-labeled rat IGFBP-5 cDNA, washed, and visualized by autoradiography. A representative Northern blot probing is shown in the upper panel. The hybridized probe was eluted, and the blot was hybridized with 32 P-labeled antisense 18 S ribosomal RNA riboprobe (Ambion, Houston, TX; shown below). RNA standards (Life Technologies, Inc.) were used to determine the length (in kilobases (kb)) of IGFBP-5 transcripts, shown on the left. Parallel cultures were treated with the same cycloheximide solution and cultures pulsed with [ 3 H]proline to assess effectiveness of the protein synthesis inhibitor; 90 -95% inhibition of protein synthesis was observed. The Northern blot is representative of three independent experiments.