- Email: [email protected]
Morphological Changes Induced by Prostaglandin E in Cultured Rat Osteoblasts
Bone Vol. 22, No. 6 June 1998:629 – 636
Morphological Changes Induced by Prostaglandin E in Cultured Rat Osteoblasts R.-S. YANG,1 W.-M. FU,2 S.-M. WANG,3 K.-S. LU,3 T.-K. LIU,1 and S.-Y. LIN-SHIAU2 Departments of 1 Orthopaedics, 2 Pharmacology, and 3 Anatomy, College of Medicine, National Taiwan University, Taipei, Taiwan
metabolic activities, but also induce the shape change of osteoclasts and/or osteoblasts.7–10,31 Salmon calcitonin, PGs, and dibutyryl cyclic adenosine monophosphate (DBcAMP) decreased the cytoplasmic spreading of osteoclasts, but increased the cytoplasmic spreading in the presence of osteoblasts.7,9 On the other hand, PTH increased the spreading of the osteoclasts in the presence of osteoblasts.7 Rat endocranial osteoblasts in situ or isolated osteoblasts grown on plastic became elongated and aligned parallel to each other when treated with PTH.4,18 –21,30,39 The major morphological change was cytoplasmic spreading, which was supposed to be related to the functional status of the cells.7 Prostaglandins E (PGEs) are important regulators of bone metabolism. They may act on a separate group of bone cells to stimulate bone formation as well as bone resorption. PGEs, especially PGE2, may mediate the effects caused by PTH and other agents.37,43,44 PGEs have also been shown to be involved in the regulation of cell shape changes in other cells, including human rheumatoid synovial cells,15 cultured rat glomerular epithelial cells,23 rabbit corneal endothelial cells,32 human orbital fibroblasts,40,48 rat ovarian granulosa cells,25 etc. On the other hand, PGEs have been reported to be involved in various clinical disorders. The local tissue levels of PGE2 in periodonitis approach a concentration of 5 mmol/L or higher, suggesting an even higher concentration at the local interface. In addition, the biological response of PGEs differs in the presence of other factors.33–35,42 Therefore, PGEs are intimately related to the cellular physiology and play an important role in the pathophysiology of various disorders. In the current study, we investigated the possible mechanisms and significance of the PGE-induced morphological changes of osteoblasts. Two models were used, including: explant cultures of rat calvaria freed of connective tissue and periosteum but retaining the surface osteoblasts in situ; and the isolated osteoblasts from long bone fragments of neonatal rats. We investigated the PGE-induced slender cell transformation of rat osteoblasts, with special reference to the cytoskeletal organization. The possible involvement of protein kinase, cAMP, and Ca21 in the action of PGEs will also be assessed.
Prostaglandin E (PGE)-induced morphological changes of osteoblasts and its possible mechanisms were investigated in cultured calvaria and isolated osteoblasts from long bone fragments of neonatal rats. The control osteoblasts, either on the calvaria or isolated from the long bone fragments, were flat, polygonal in shape, and arranged in a monolayer under scanning electron microscopy (SEM) or phase contrast microscopy. Treatment with 1 mmol/L of prostaglandin E2 (PGE2, 2 h) caused these bone cells to contract a soma, whereas 10 and 100 mmol/L PGE2 (2 h) caused 18%–30% of the bone cells to elongate and expose the undersurface. Incubation of the cultured osteoblasts with PGE2 at different time periods showed a bell-shaped pattern with the optimal response at 2 h of incubation. A similar reaction can be induced by treatment with prostaglandin E1 (PGE1) or dibutyryl cyclic adenosine monophosphate (DBcAMP) in combination with 3-isobutyl-1-methylxanthine (IBMX). Furthermore, we assessed the percentage of responsive isolated bone cells to investigate interactions with other agents. The morphological changes induced by PGEs were inhibited by H-8, a protein kinase inhibitor. On the other hand, elevated intracellular calcium enhanced the PGE-induced morphological changes. Fluorescence labeling showed that PGEs caused the breakdown of the actin microfilaments, but spared the microtubules and vimentin filaments in the isolated osteoblast-like cells. These results suggest that the morphological changes of osteoblasts induced by PGEs may be related to the intracellular cAMP and calcium levels. (Bone 22:629 – 636; 1998) © 1998 by Elsevier Science Inc. All rights reserved. Key Words: Prostaglandin E; Rat osteoblast; Protein kinase; Cytoskeleton. Introduction There exists an intimate relationship between the morphology and function of bone cells that are regulated by many local and systemic factors.2,16,29 For example, the change in the osteoblast shape may regulate the access of the osteoclast to the bone surface during the bone remodeling process.38 Many hormones, such as calcitonin, prostaglandins (PGs), and parathyroid hormone (PTH), have been demonstrated not only to regulate the
Materials and Methods Calvaria Explant Culture Neonatal Wistar rats, weighing 5–7 g, were dissected using an aseptic technique under general anesthesia (pentobarbital 7.5 mg/mL, intraperitoneal injection 0.015 mL/g BW). The calvaria were removed immediately into Dulbecco’s modified Ca21-free Eagle’s medium (DMEM) (Gibco, Grand Island, NY) supple-
Address for correspondence and reprints: Rong-Sen Yang, M.D., Ph.D., Department of Orthopaedics, National Taiwan University Hospital, No. 7, Chung-Shan South Road, Taipei, Taiwan. E-mail: [email protected]. edu.tw © 1998 by Elsevier Science Inc. All rights reserved.
629
8756-3282/98/$19.00 PII S8756-3282(98)00055-6
630
R.-S. Yang et al. Prostaglandin-induced morphological changes in osteoblasts
mented with 10% fetal calf serum (Gibco), glutamine (58.5 mg/mL, Gibco), MgSO4 (200 mg/mL), and antibiotics (100 U penicillin and 100 U-streptomycin sulfate per milliliter, pH 7.6). They were kept in a humidified atmosphere containing 95% air and 5% CO2 at 37°C. All connective tissues, periosteum, and dura mater were carefully removed in phosphate-buffered saline (PBS) under a dissection microscope. The endocranial aspect of the parietal bone was carefully protected during dissection to keep the intact osteoblasts in situ for further investigation. A total of more than 50 pieces parietal calvaria were included for study. In order to study the maximal number of endocranial osteoblasts in situ, primary cultures of calvaria were immediately used for scanning electron microscopy (SEM) study to assess the PGEinduced morphological changes at various concentrations. All experiments were repeated at least six times. Cultures of Isolated Rat Osteoblasts Isolated osteoblasts from long bone fragments of neonatal rats were prepared as previously described.46,47 Briefly, the long bones of Wistar neonatal rats, including the femur, tibia, humerus, radius, and ulna, were dissected with an aseptic technique under anesthesia (pentobarbital) as previously described. After defleshing, the bones were diced into bare bone fragments and washed in PBS to remove the marrow cells. The primarily separated rat bone cells obtained from these explants were grown in the same culture medium and atmosphere as the calvaria culture. The viability of the cells was monitored by the trypanblue exclusion test. Explant pieces were removed when a confluence of the resulting monolayer was approached. Cell passage was carried out by incubating the monolayer for 5–10 min in Ca21- and Mg21-free Tyrode’s solution containing 0.25% trypsin (Sigma Chemical, St. Louis, MO) and replating the cells in fresh medium at about one third of their confluent density (10,000 –20,000 cells). Over 80% of the cells reattached within 24 h. The culture medium was changed twice or thrice weekly. The morphology of the cultured bone cells was observed under a phase contrast microscope. The characteristics of the cultured rat bone cells were identified by morphology identification, cytochemical staining, and biochemical assay of alkaline phosphatase (ALP) (more than 78% of cells were stained positive), SEM, transmission electron microscope (TEM), and monoclonal antibodies of osteocalcin (more than 95% positive).47 Cultured bone cells at confluence were used for study. It took 7–10 days of culture before the experiment. All experiments were performed on the cells with less than three passages. The isolated bone cells were used for study of the PGE-induced morphological changes and their possible mechanisms. SEM For SEM studies, the endocranial osteoblasts on the calvaria or on the glass slip of the primary culture were fixed in 2.5% glutaraldehyde in 0.1 mol/L cacodylate for 20 min. At completion of the fixation, the cells were rinsed in a phosphate buffer for 5 min, postfixed in 1% osmium tetroxide, dehydrated in a graded series of alcohol, and subsequently critical-point dried with liquid carbon dioxide. The specimens were ion sputter coated with a thin layer (about 10 nm) of gold and examined in a JEOL T330A scanning electron microscope at 15 or 30 kV. The morphological changes after treatment with PGE1 (1, 10, or 100 mmol/L, 2 h), PGE2 (1, 10, or 100 mmol/L, 2 h) and, DBcAMP (1 mmol/L) plus 3-isobutyl-1-methylxanthine (IBMX) (10 mmol/L) (2 h) were compared to those of control osteoblasts treated with a vehicle. PGE was dissolved in ethanol to make a 10 mmol/L stock solution, and this was added to the culture
Bone Vol. 22, No. 6 June 1998:629 – 636
medium to make a final concentration of 1, 10, and 100 mmol/L. Thus, the concentration of ethanol ranged from 0.01% to 1% vol/vol, and control experiments were given a matched aliquot of ethanol without PGEs. All experiments were repeated at least six times. Estimation of the Percentage of Slender Cells in Cultured Osteoblasts For studies of the interaction effects of various agents with PGEs, the isolated rat osteoblasts were plated in 3-cm collagen-coated tissue culture dishes and treated with PGEs or other agents at various concentrations for different time periods. The control osteoblasts were treated with an ethanol vehicle (for PGE experiments) of the same concentration as the experimental group. The morphological changes of bone cells induced by PGE were monitored with light microscopy (3200 or 3400 Nikon Diaphot). The responsive cells were defined as the cells with cytoplasmic processes longer than twofold of the control cell long axis. The number of responsive cells showing such morphological changes before and after treatment with PGE and various drugs were counted and expressed as the population ratio of responsive cell (%). We took pictures under phase contrast microscopy of at least 10 areas under 3200 magnification for each experiment and counted the percentage of responsive cells (at least 150 cells were counted for each area). The time course and dose response after treatment with PGE1 (1, 10, and 100 mmol/L), PGE2 (1, 10, and 100 mmol/L), and DBcAMP (1 mmol/L) plus IBMX (10 mmol/L) (all from Sigma) were investigated. The effects of H-8 (N[2-(methylamino-)ethyl]-5-isoquinoline sulfonamide) (10 mmol/L, 2 h) and A23187 (2 mmol/L, 18 h) (all from Sigma) on the PGE-induced morphological changes were studied. H-8 was dissolved in PBS to make a 10 mmol/L stock solution, and A23187 was dissolved in dimethyl sulfoxide (DMSO) to make a 2 mmol/L stock solution. These solutions were added to culture medium to make a final concentration of 10 mmol/L each. The same concentration of solvent was added to control experiments without agents. The cell viability after treatment with PGE or various agents was monitored by the trypan-blue exclusion test, and proved to be more than 93% viable. All experiments were repeated at least six times. Statistical analyses were evaluated by Student’s t-test (twotailed) and analysis of variance (ANOVA). A Bonferroni multiple comparison test was also used to analyze the results. Fluorescent Labeling of Microfilaments, Microtubules, and Vimentin Filaments Cytoskeleton studies were performed in the primary culture of the osteoblasts. For F-actin staining, the cells were washed three times with PBS at 5 min intervals and then fixed with 3.7% formaldehyde for 10 min. After fixation, cells were washed three times with PBS and then permeabilized with acetone (220°C) for 3–5 min. Air-dried cells were then stained with fluorescein isothiocyanate (FITC)-phallotoxin in PBS (Molecular Probes, Eugene, OR) for 30 min, washed again with PBS, mounted in glycerol-PBS containing 0.1 mol/L n-propyl gallate, and examined immediately for fluorescent staining. In control experiments, cells were incubated with unlabeled phallotoxin before exposure to fluorescent probes. There was little or no labeling observed. For the microtubule study, a mouse anti-a-tubulin primary antibody (Sigma) was used. Rat osteoblasts were washed three times with a PHEM solution [60 mmol/L piperazine-N,N9-bis [2-ethanesulfonic acid] (PIPES), 25 mmol/L N-2-hydroxyeth-
Bone Vol. 22, No. 6 June 1998:629 – 636
R.-S. Yang et al. Prostaglandin-induced morphological changes in osteoblasts
631
ylpiperazine-N-2-ethanesulfonic acid (HEPES), 2 mmol/L MgCl2, 10 mmol/L ethyleneglycol-bis-(b-aminoethylether)N,N,N9,N9-tetraacetic acid (EGTA), pH 6.9] (all from Sigma) at 5 min intervals and then fixed with 100% methanol (220°C) for 3–5 min, followed by washing with PBS. The osteoblasts were incubated with a 1:100 mouse primary anti-a-tubulin antibody at 37°C for 1 h. After washing with PBS, the cells were incubated with 1:50 goat-antimouse FITC-conjugated secondary antibody (Sigma) at 37°C for 1–2 h. The cells were washed again with PBS and prepared as above. In control experiments, cells were incubated only with a secondary antibody and showed little or no labeling. For vimentin staining, a mouse antivimentin primary antibody (Sigma) was used. Rat osteoblasts were washed three times with a PBS solution at 5 min intervals and then fixed with 0.15% glutaraldehyde for 3–5 min, followed by washing with PBS. The osteoblasts were incubated with a 1:80 mouse primary antivimentin antibody at 37°C for 1 h. After washing with PBS, the cells were incubated with a 1:50 goat-antimouse FITC-conjugated secondary antibody at 37°C for 1–2 h. The cells were then prepared as mentioned above. All above experiments were repeated at least six times. Results PGE-Induced Morphological Changes in Rat Calvaria Osteoblasts Under the SEM, the endocranoial osteoblasts on the control calvaria were flat and polygonal in shape, and the cell surface was coated with a few microvilli. They were arranged in a single layer side by side and covered the underlying collagen fibrils (Figure 1A). Incubation of the rat calvaria with PGE2 (1 mmol/L, 2 h) induced the contraction of the soma and a few of the cells were elongated (Figure 1B). Higher concentrations of PGE2 (10 or 100 mmol/L, 2 h) caused more endocranial osteoblasts (40%– 80%) to elongate and expose the undersurface of the calvaria (Figure 1C). Incubation of the calvaria with PGE1 (1, 10, or 100 mmol/L, 2 h) or DBcAMP (1 mmol/L) plus IBMX (10 mmol/L) (2 h) also induced similar shape changes. PGE-Induced Morphological Changes in Cultured Rat Osteoblasts Under SEM, control cultured rat osteoblasts were polygonal with their own long axis (Figure 2A). The soma of cultured osteoblasts contracted and a few of the osteoblasts were elongated (10%–15%) after treatment with a low concentration of PGE2 (1 mmol/L, 2 h) (Figure 2B). However, after treatment with higher concentrations of PGE2 (10 or 100 mmol/L, 2 h), more osteoblasts (18%–30%) were progressively elongated (Figure 2C). The changes in cell shape observed under phase contrast light microscopy were similar (Figure 3A and B). Treatment with higher concentrations of PGE1 (10 or 100 mmol/L, 2 h) or DBcAMP (1 mmol/L) plus IBMX (10 mmol/L) (2 h) also induced similar morphological changes. Estimation of the Percentage of Slender Cells in Cultured Osteoblasts Because incubation with a concentration of lower than 1 mmol/L did not induce the morphological change, we treated the cultured rat osteoblasts with various concentrations of PGE2 (1, 10, and 100 mmol/L) for different time periods. The results showed a bell-shaped pattern with an optimal response at 2 h of incubation (Figure 4A). The PGE-induced slender cell formation of isolated
Figure 1. Morphological changes of rat calvaria osteoblasts induced by PGE2 under SEM. (A) Control osteoblasts treated with ethanol vehicle show sparse short cytoplasmic processes. (B) Incubation with 1 mmol/L PGE2 (2 h) caused the contraction of soma. (C) Treatment with 100 mmol/L PGE2 (2 h) induced significant elongation of calvaria and developed numerous long cytoplasmic projections. All experiments were repeated at least six times.
rat osteoblasts was concentration dependent. The percentage of responsive cells after incubation with PGE2 (100 mmol/L) reached a peak level (;30%) at 2 h. The dose-response curve of the peak effect at 2 h showed a positive correlation (r 5 0.91, p , 0.0001). The morphological changes were reversible and some cells returned to their normal appearance later. The results obtained after treatment with PGE1 also showed a bell-shaped response curve (Figure 4B). Furthermore, we investigated the effect of H-8, a protein kinase inhibitor, on the PGE-induced morphological changes of cultured rat osteoblasts. Pretreatment with H-8 (10 mmol/L, 2 h) decreased the percentage of slender cells induced by 100 mmol/L PGE2 from 28.6% 6 2% to 14.1% 6 2.1% (p , 0.05) (Figure 5). A similar inhibitory effect of H-8 on the morphological
632
R.-S. Yang et al. Prostaglandin-induced morphological changes in osteoblasts
Bone Vol. 22, No. 6 June 1998:629 – 636
Figure 3. Morphological changes of primary cultures of rat osteoblasts induced by PGE2 under phase contrast microscope. (A) Control osteoblasts treated with an ethanol vehicle show sparse short cytoplasmic processes. (B) Treatment with 100 mmol/L PGE2 (2 h) induced osteoblasts to elongate. All experiments were repeated at least six times.
IBMX (10 mmol/L) (up to 96% at 24 h) (Figure 6B and C). Thus the morphological changes of the osteoblasts induced by PGEs may probably be related to the intracellular cAMP and Ca21 levels. Figure 2. Morphological changes of primary cultures of rat osteoblasts induced by PGE2 under SEM. (A) Control osteoblasts treated with ethanol vehicle show a sparse flat polygonal shape with short cytoplasmic processes. (B) Treatment with 1 mmol/L PGE2 (2 h) induced contraction of some osteoblasts. (C) Treatment with 100 mmol/L PGE2 (2 h) induced a more elongated cell formation and displayed numerous long and slender cytoplasmic projections. All experiments were repeated at least six times.
changes induced by PGE1 (from 32.1% 6 2.1% to 11.2% 6 1.5%, p , 0.05) or DBcAMP plus IBMX (26.7% 6 1.9% vs. 16.8% 6 1.8%, p , 0.05) was also observed. Treatment with Ca21 ionophore A23187 (2 mmol/L) alone up to 18 h caused 15%–20% of osteoblasts to elongate. However, treatment with DMSO of the same concentration did not affect the morphology of the cells. The proportion of slender cells induced by PGE2 (10 mmol/L) increased up to 40% at 24 h in the cultured osteoblasts pretreated with A23187 (2 mmol/L, 18 h) (p , 0.05) (Figure 6A). Pretreatment with A23187 (2 mmol/L, 18 h) also increased the shape change induced by either PGE1 (10 mmol/L) (up to 74% at 24 h) or DBcAMP (1 mmol/L) plus
Roles of Cytoskeletal Proteins in the Action of Prostaglandin The normal rat osteoblasts showed a flat polygonal shape with stress fibers running a straight course throughout the cytoplasm (Figure 7A). However, the responsive cells induced by the prostaglandin showed the breakdown of actin microfilaments and diffused F-actin staining within the cells (Figure 7B). On the other hand, the distribution pattern of the microtubule (Figure 7C and D) and vimentin filaments (Figure 7E and F) in osteoblasts were not significantly affected by the prostaglandin, even though accompanied with the marked morphological changes of the cells. Discussion In this study we have demonstrated that treatment of rat calvaria osteoblasts and isolated osteoblasts from long bone fragments with PGEs (1 mmol/L, 2 h) induced the contraction of soma and a few cells were elongated, whereas higher concentrations of PGEs (10 and 100 mmol/L, 2 h) caused 18%–30% of the osteoblasts to develop the rapid elongation of soma, resulting in a morphological transformation. The PGE-
Bone Vol. 22, No. 6 June 1998:629 – 636
Figure 4. Concentration-dependent induction of morphological changes by PGEs and DBcAMP plus IBMX in cultured rat osteoblasts. (A) Time course of the effect of PGE2 (filled circle: 100 mmol/L; filled square: 10 mmol/L; filled triangle: 1 mmol/L). (B) The time-course curves of PGE1-induced morphological changes were similar (filled circle: 100 mmol/L; filled square: 10 mmol/L; filled triangle: 1 mmol/L). Data presented as mean 6 standard deviation (SD) (n 5 6). *p , 0.05 vs. 1 mmol/L at each time point.
induced slender cell transformation of rat osteoblasts was related to the breakdown of actin microfilaments. In addition, pretreatment with H-8 decreased whereas A23187 increased the percentage of responsive cells. Thus, the elevated intra-
Figure 5. The inhibitory effect of H-8 on the actions of prostaglandins inducing morphological changes. The H-8 antagonized the effect of 100 mmol/L PGE and DBcAMP (1 mmol/L) plus IBMX (10 mmol/L) induced morphological changes of cultured osteoblasts. Data presented as mean 6 SD (n 5 6). *p , 0.05.
R.-S. Yang et al. Prostaglandin-induced morphological changes in osteoblasts
633
Figure 6. Enhancement by A23187 on the prostaglandin-induced morphological changes in cultured rat osteoblasts. (A) Pretreatment with A23187 (2 mmol/L, 18 h) showed that 15%–20% of the osteoblasts changed shape. Pretreatment with A23187 (2 mmol/L, 18 h) followed by addition of PGE2 (10 mmol/L) (filled square) increased a higher proportion of slender cells of the osteoblasts than the controls (filled circle). The effects of PGE1 (10 mmol/L) (B) and DBcAMP (1 mmol/L) 1 IBMX (10 mmol/L) (C) on the morphological changes were also enhanced by A23187 [filled circle: controls; filled square: pretreatment with A23187 (2 mmol/L, 18 h)]. Data presented as mean 6 SD (n 5 6). *p , 0.05 for comparison at each time point.
cellular cAMP and calcium levels may be related to these PGE-induced morphological changes. The rat calvaria model provides a good model for the study of the bone cell biology with special reference to cell– cell and cell–matrix interactions.6,18,19,28,30,39,45 The osteoblasts are arranged in a monolayer with their own long axis. The polygonal calvaria osteoblasts elongated after exposure to PTH.18 –21 However, the effect of PGE2 on the morphological changes of osteoblasts was controversial.1,4 Ali et al. demonstrated that 1 mmol/L PGE2 did not cause a shape change of the rat calvaria osteoblasts.1 However, a rapid change of the shape of cultured osteoblasts could be induced after treatment with PGEs.4 In the current study, we used a higher concentration of PGE and demonstrated a dose-dependent effect of PGE-induced morphological changes of either calvaria osteoblasts or cultured bone cells from long bone fragments of neonatal rats. These results correspond to other reports.30,39 However, the effects of PGEs on the isolated cultured osteoblasts in this study differed from those on the calvaria osteoblasts, which preserved intact cell–matrix interactions. Many agents have been proven to increase the intracellular cAMP of osteoblasts and osteosarcoma cells.3,12–14 The mode of actin of PGE probably involves the accumulation of cAMP or
634
R.-S. Yang et al. Prostaglandin-induced morphological changes in osteoblasts
Bone Vol. 22, No. 6 June 1998:629 – 636
Figure 7. Effects of PGE2 on the organization of cytoskeletons in cultured rat osteoblasts under fluorescence microscopy. (A) The control cells treated with an ethanol vehicle showed typical stress fibers of actin throughout the cytoplasm. (B) Two hours after treatment with 100 mmol/L PGE2, bundles of actin fibers disappear, accompanied with the cell transformation. The microtubules [(C) and (D)] and vimentin [(E) and (F)] were not significantly affected after treatment with PGE2. All experiments were repeated at least six times.
elevated intracellular calcium in the bone cells. In the present study, treatment with PGE2, PGE1, or DBcAMP plus IBMX induced a similar shape change in the osteoblasts. Furthermore, the inhibition of the PGE-induced slender cell formation by H-8, a preferential protein kinase A inhibitor, suggests that a cyclic AMP-dependent protein kinase may be involved. Thus the morphological change may be closely related to an increased level of intracellular cAMP. Previous studies have demonstrated that an increased intracellular cAMP level is closely related to the shape change in various cell types.22,23,25,30,36,48 The contracted cytoplasm could respread when the cAMP returned to normal levels.3,4,41 On the other hand, H-8 was not fully inhibitory, perhaps suggesting additional mechanisms. H-8 may also inhibit protein kinase C, which may be related to regulation of the cell morphology. In addition, intracellular calcium may also be involved in the PGE-induced changes in cell morphology. The possible interaction of the mechanism merits further investigation. PGE has been demonstrated to cause the dissolution of actin microfilament bundles and inhibit the induction of actin polymerization or the phosphorylation of myosin light chain kinase.4,24 Although the role of kinases in osteoblasts is not well understood, it has been shown that only thin filaments reaching the cortex of the cell are associated with the soma and serve as the anchoring mechanism of the cells. In this study, treatment with PGEs induced the breakdown of actin microfilaments without affecting microtubule and vimentin filaments. These results indicated that the effect of prostaglandin on osteoblasts may be
intimately related to the breakdown of the actin microfilament. Lomri et al. demonstrated that cAMP and calcium can modulate the synthesis and organization of the cytoskeleton of mouse osteoblasts.26,27 The marked elevation of intracellular cAMP in the osteoblasts may induce the dissolution of the microfilament. However, the complexity of the phosphorylation pathways makes it difficult to sort out the exact mechanisms of the action of the prostaglandin. Other studies showed that PGE2 may enhance the metabolism of phosphoinositol and the IP3 level increases in the mouse osteoblast and osteoclast.11,13 Elevated IP3 may mobilize the release of Ca21 from the nonmitochondrial pool. Therefore, the Ca21 level is increased after treatment with PGE2. The present study demonstrated that the elevation of intracellular Ca21 by A23187 is able to enhance the PGE-induced slender cell formation. Thus the PGE-induced morphological changes may be related to the elevated intracellular Ca21. However, the definite interaction of the mechanisms needs further investigation. On the other hand, in the current study, we demonstrated that normal microtubule and vimentin assembly was preserved after treatment with PGEs. This finding corresponds to other studies.5 In addition, Shen et al.39 found that colchicine, an inhibitor of microtubule polymerization, inhibited the shape changes of the osteoblast induced by PTH and/or PGE2. Therefore, the internection and interaction among the microtubule, vimentin, and actin may be closely related to achieve the final shape of the osteoblasts. These organizations among cytoskeletal proteins are prob-
Bone Vol. 22, No. 6 June 1998:629 – 636
ably involved in the cell morphology induced by hormones and nucleotides. In the normal bone remodeling process, changes of the osteoblast morphology may expose the underlying matrix and provide a convenient access to osteoclasts.38 Bone remodeling also plays an important role in the pathophysiology of various kinds of clinical disorders. The bone remodeling is regulated by many factors, including PGs. In fetal rat calvaria cultures, PGE2 and other PGs (PGE1, PGF1a, PGA1, PGA2, 0.1–100 mmol/L) inhibited the incorporation of the radioproline into newly formed collagen without affecting the synthesis of noncollagen protein.17 The effect was latent, not acute, and occurred only at the higher end of the dose range. On the other hand, the local tissue levels of PGE2 in cirvicular fluid of periodonitis approach a concentration of 5 mmol/L or higher,33–35 suggesting a higher local concentration of PGE2 at the interface of the resorbing bone surface. The local calcium levels released at the resorbing bone surface is probably high. From the current study, local high concentrations of PGEs and calcium may act synergistically and potentiate the morphological changes of bone cells to expose more bone matrix. In addition, the biological response of PGEs differs in the presence of other factors.35,42 Therefore, the definite clinical implication of the PGE-induced cell morphological changes in the current study merits further investigation. In conclusion, this study suggested that PGE-induced morphological changes of cultured osteoblasts were intimately related to the cytoskeletal microfilaments. The elevated intracellular cAMP or Ca21 may be closely related to the morphological changes.
R.-S. Yang et al. Prostaglandin-induced morphological changes in osteoblasts
14.
15.
16.
17.
18. 19.
20.
21.
22.
23.
Acknowledgments: This work was supported by grants from the National Science Council (NSC82-0115-B002-551 and NSC81-0412-B002-649).
24.
References
25.
1. Ali, N. N., Melhuish, P. B., Boyde, A., Bennett, A., and Jones, S. J. Parathyroid hormone, but not prostaglandin E2, changes the shape of osteoblasts maintained on bone in vitro. J Bone Miner Res 5:115–121; 1990. 2. Arnett, T. R. Update of bone cell biology. Eur J Orthodont 12:81–90; 1990. 3. Atkins, D. and Martin, T. J. Rat osteogenic sarcoma cells: Effects of some prostaglandins, their metabolites and analogues on cyclic AMP production. Prostaglandins 13:861– 871; 1977. 4. Aubin, J. E., Alders, E., and Heersche, J. N. M. A primary role for microfilaments, but not microtubules, in hormone induced cytoplasmic retraction. Exp Cell Res 143:439 – 450; 1983. 5. Beertsen, W., Heersche, J. N. M., and Aubin, J. E. Free and polymerized tubulin in cultured bone cells and Chinese hamster ovary cells: The influence of cold and hormones. J Cell Biol 95:387–393; 1982. 6. Boyde, A., Jones, S. J., Binderman, I., and Harrell, A. Scanning electron microscopy of bone cells in culture. Cell Tissue Res 166:65–70; 1976. 7. Chambers, T. J., Athanasou, N. A., and Fuller, K. Effect of parathyroid hormone and calcitonin on the cytoplasmic spreading of isolated osteoclasts. J Endocrinol 102:281–286; 1984. 8. Chambers, T. J. and Dunn, C. J. The effect of parathyroid hormone, 1,25dihydroxy-cholecalciferol and prostaglandins on the cytoplasmic activity of isolated osteoclasts. J Pathol 137:192–203; 1982. 9. Chambers, T. J., Fuller, K., and Athanasou, N. A. The effect of prostaglandin I2, E1, E2, and dibutyryl cyclic AMP on the cytoplasmic spreading of rat osteoclasts. Br J Exp Pathol 65:557–566; 1984. 10. Chambers, T. J. and Moore, A. The sensitivity of isolated osteoclasts to morphological transformation by calcitonin. J Clin Endocrinol Metab 57:819 – 824; 1983. 11. Dolce, C., Schneeberger, G., and Dziak, R. Effects of prostaglandin E2 on phosphoinositide metabolism in osteoblastic cells. J Dent Res 66:1577; 1987. 12. Dziak, R. M., Hurd, D., Miyasaki, K. T., Brown, M., Weinfeld, N., and Hausmann, E. Prostaglandin E2 binding and cyclic AMP production in isolated bone cells. Calcif Tissue Int 35:243–249; 1983. 13. Farndale, R. W., Sandy, J. R., Atkinson, S. J., Pennington, S. R., Meghji, S.,
26.
27.
28.
29. 30. 31.
32.
33.
34.
35.
36.
635
and Meikle, M. C. Parathyroid hormone and prostaglandin E2 stimulate both inositol phosphates and cyclic AMP accumulation in mouse osteoblast cultures. Biochem J 252:263–268; 1988. Farr, D., Pochal, W., Brown, M., Shapiro, E., Weinfeld, N., and Dziak, R. Effects of prostaglandins on bone cell calcium. Arch Oral Biol 29:885– 891; 1984. Gadher, S. J. and Woolley, D. E. Comparative studies of adherent rheumatoid synovial cells in primary cultures: Characterization of the dendritic (stellate) cell. Rheumatol Int 7:13–22; 1987. Hagel-Bradway, S., Tatakis, D. N., and Dziak, R. Prostaglandin-induced changes in calcium uptake and cAMP production in osteoblast-like cells: Role of protein kinase C. Calcif Tissue Int 48:272–277; 1989. Harvey, W. Source of prostaglandins and their influence on bone resorption and formation. In: Harvey, W. and Bennett, A., Eds. Prostaglandins and Bone Resorption. Boca Raton, FL: CRC; 27, 1988. Jones, S. J. and Boyde, A. Morphological changes of osteoblasts in vitro. Cell Tissue Res 166:101–107; 1976. Jones, S. J. and Boyde, A. Experimental study of changes in osteoblastic shape induced by calcitonin and parathyroid extract in an organ culture system. Cell Tissue Res 169:449 – 465; 1976. Jones, S. J., Boyde, A., and Shapiro, I. M. The response of osteoblasts to parathyroid hormone (PTH 1-34) in vitro. Metab Bone Dis Relat Res 2:335– 338; 1981. Jones, S. J. and Ness, A. R. A study of the arrangement of osteoblasts of rat calvarium cultured in medium with or without added parathyroid extract. J Cell Sci 25:247–263; 1977. Jumblatt, M. M. and Paterson, C. A. Prostaglandin E2 effects on corneal endothelial cyclic adenosine monophosphate synthesis and cell shape are mediated by a receptor of the EP2 subtype. Invest Ophthalmol Visual Sci 32:360 –365; 1991. Kreisberg, J. I., Patel, P. Y., Venkatachalam, M. A., and Taylor, G. Elevations of intracellular cAMP result in a change in cell shape that resemble dome formation in cultured rat glomerular epithelial cells. In Vitro Cell Develop Biol 22:392–396; 1986. Lamb, N. J. C., Fernandez, A., Conti, M. A., Adelstein, R., Glass, D. B., Welch, W. W., and Feramisco, J. R. Regulation of actin microfilament integrity in living nonmuscle cells by the CAM-dependent protein kinase and the myosin light chain kinase. J Cell Biol 106:1955–1971; 1988. Lawrence, T. S., Ginzberg, R. D., Gilula, N. B., and Beers, W. H. Hormonally induced cell shape changes in cultured rat ovarian granulosa cells. J Cell Biol 80:21–36; 1979. Lomri, A., Marie, P. J., Escurat, M., and Portier, M. Cytoskeletal protein synthesis and organization in cultured mouse osteoblastic cells. FEBS Lett 222:311–316; 1987. Lomri, A. and Marie, P. Distinct effects of calcium- and cyclic AMP-enhancing factors on cytoskeletal synthesis and assembly in mouse osteoblastic cells. Acta Biochim Biophys 1052:179 –186; 1990. MacDonald, B. R. F., Gallagher, J. A., Ahnfekt-Ronne, I., Beresford, J. N., Gowen, M., and Russell, G. R. Effects of bovine parathyroid hormones and 1,25-dihydroxyvitamin D3 on the production of prostaglandins by cells derived from human bone. FEBS Lett 169:49 –52; 1984. Martin, T. J., Ng, K. W., and Suda, T. Bone cell physiology. Endocrinol Metab Clin 18:833– 858; 1989. Miller, S. S., Wolf, A. M., and Arnaud, C. D. Bone cells in culture: Morphological transformation by hormones. Science 192:1340 –1343; 1976. Murrills, R. J., Shane, E., Lindsay, R., and Dempster, D. W. Bone resorption by isolated human osteoclasts in vitro: Effects of calcitonin. J Bone Miner Res 4:259 –268; 1989. Neufeld, A. H., Jumblatt, M. M., Matkin, E. D., and Raymond, G. M. Maintenance of corneal endothelial cell shape by prostaglandin E2: Effects of EGF and indomethacin. Invest Ophthalmol Visual Sci 27:1437–1442; 1986. Offenbacher, S., Odle, B. M., and Dyke, T. E. The use of cervicular fluid prostaglandin E2 levels as a predictor of preiodontal attachment loss. J Periodontol 21:101–112; 1986. Offenbacher, S., Williams, R. C., Jeffcoat, M. K., Howell, T. H., Odle, B. M., Smith, M. A., Hall, C. M., Johnson, H. G., and Goldhaber, P. Effects of NSAIDs on beagle cervicular cyclooxygenase metabolites and periodontal bone loss. J Periodontol Res 27:207–213; 1991. Offenbacher, S., Heasman, P. A., and Collins, J. G. Modulation of host PGE2 secretion as a determinant of periodontal disease expression. J Periodontol 64:432– 444; 1993. Porter, K. R., Puck, T. T., Hsie, A. W., and Kelly, D. An electron microscope
636
37.
38. 39.
40.
41.
42.
43.
R.-S. Yang et al. Prostaglandin-induced morphological changes in osteoblasts study of the effects of dibutyryl cAMP on Chinese hamster ovary cells. Cell 2:145–162; 1974. Powles, T. J., Easty, D. M., Easty, G. C., Bondy, P. K., and Munro-Neville, A. Aspirin inhibition of in vitro osteolysis stimulated by parathyroid hormone and PGE1. Nature (New Biol) 245:83– 84; 1973. Rodan, G. A. and Martin, T. J. The roles of osteoblasts in hormonal control of bone resorption—a hypothesis. Calcif Tissue Int 33:349 –351; 1981. Shen, V., Rifas, L., Kohler, G., and Peck, W. A. Prostaglandins change cell shape and increase intercellular gap junctions in osteoblasts cultured from rat fetal calvaria. J Bone Miner Res 1:243–249; 1986. Smith, T. J., Sempowski, G. D., Wang, H. S., Del Vecchio, P. J., Lippe, S. D., and Phipps, R. P. Evidence for cellular heterogenercity in primary cultures of human orbital fibroblasts. J Clin Endocrinol Metab 80:2620 –2625; 1995. Spruill, W. A., White, M. G., Steiner, A. L., Tres, L. L., and Kierszenbaum, L. Temporal sequence of cell shape changes in cultured rat sertoli cells after experimental elevation of intracellular cAMP. Exp Cell Res 131:131–148; 1981. Stashenko, P., Dewhirst, F. E., Peros, W. J., Kent, R. L., and Ago, J. M. Synergistic interactions between interleukin 1, tumor necrosis factor, and lymphotoxin in bone resorption. J Immunol 138:1464 –1468; 1987. Tashjian, Jr., A. H., Ivey, J. L., Delclos, B., and Levine, L. Stimulation of prostaglandin production in bone by phorbol diesters and melittin. Prostaglandins 16:221–232; 1978.
Bone Vol. 22, No. 6 June 1998:629 – 636 44. Tashjian, Jr., A. H. and Levine, L. Epidermal growth factor stimulates prostaglandin production and bone resorption in cultured mouse calvaria. Biochem Biophys Res Commun 85:966 –975; 1978. 45. Yang, C. Y., Gonnerman, W. A., Taylor, L., Nimberg, R. B., and Polgar, P. R. Synthetic human parathyroid hormone fragment stimulates prostaglandin E2 synthesis by chick calvaria. Endocrinology 120:63– 69; 1987. 46. Yang, R. S., Lu, K. S., Fu, W. M., Liu, T. K., and Lin-Shiau, S. Y. Staurosporine-induced morphological changes in the rat osteoblasts. Cell Biol Int 17:75– 82; 1992. 47. Yang, R. S., Liu, T. K., Tsai, K. S., Lin-Shiau, S. Y., and Lu, K. S. Morphological and immunocytochemical characterization of osteoblast cultures from long bones of neonatal rats. Arch Histol Cytol 55:415– 422; 1992. 48. Wang, H. S., Keese, C. R., Giaever, I., and Smith, T. J. Prostaglandin E2 alters human orbital fibroblast shape through a mechanism involving the generation of cyclic adenosine monophosphate. J Clin Endocrinol Metab 80:3553–3560; 1995.
Date Received: November 15, 1996 Date Revised: April 3, 1997 Date Accepted: February 19, 1998
Morphological Changes Induced by Prostaglandin E in Cultured Rat Osteoblasts R.-S. YANG,1 W.-M. FU,2 S.-M. WANG,3 K.-S. LU,3 T.-K. LIU,1 and S.-Y. LIN-SHIAU2 Departments of 1 Orthopaedics, 2 Pharmacology, and 3 Anatomy, College of Medicine, National Taiwan University, Taipei, Taiwan
metabolic activities, but also induce the shape change of osteoclasts and/or osteoblasts.7–10,31 Salmon calcitonin, PGs, and dibutyryl cyclic adenosine monophosphate (DBcAMP) decreased the cytoplasmic spreading of osteoclasts, but increased the cytoplasmic spreading in the presence of osteoblasts.7,9 On the other hand, PTH increased the spreading of the osteoclasts in the presence of osteoblasts.7 Rat endocranial osteoblasts in situ or isolated osteoblasts grown on plastic became elongated and aligned parallel to each other when treated with PTH.4,18 –21,30,39 The major morphological change was cytoplasmic spreading, which was supposed to be related to the functional status of the cells.7 Prostaglandins E (PGEs) are important regulators of bone metabolism. They may act on a separate group of bone cells to stimulate bone formation as well as bone resorption. PGEs, especially PGE2, may mediate the effects caused by PTH and other agents.37,43,44 PGEs have also been shown to be involved in the regulation of cell shape changes in other cells, including human rheumatoid synovial cells,15 cultured rat glomerular epithelial cells,23 rabbit corneal endothelial cells,32 human orbital fibroblasts,40,48 rat ovarian granulosa cells,25 etc. On the other hand, PGEs have been reported to be involved in various clinical disorders. The local tissue levels of PGE2 in periodonitis approach a concentration of 5 mmol/L or higher, suggesting an even higher concentration at the local interface. In addition, the biological response of PGEs differs in the presence of other factors.33–35,42 Therefore, PGEs are intimately related to the cellular physiology and play an important role in the pathophysiology of various disorders. In the current study, we investigated the possible mechanisms and significance of the PGE-induced morphological changes of osteoblasts. Two models were used, including: explant cultures of rat calvaria freed of connective tissue and periosteum but retaining the surface osteoblasts in situ; and the isolated osteoblasts from long bone fragments of neonatal rats. We investigated the PGE-induced slender cell transformation of rat osteoblasts, with special reference to the cytoskeletal organization. The possible involvement of protein kinase, cAMP, and Ca21 in the action of PGEs will also be assessed.
Prostaglandin E (PGE)-induced morphological changes of osteoblasts and its possible mechanisms were investigated in cultured calvaria and isolated osteoblasts from long bone fragments of neonatal rats. The control osteoblasts, either on the calvaria or isolated from the long bone fragments, were flat, polygonal in shape, and arranged in a monolayer under scanning electron microscopy (SEM) or phase contrast microscopy. Treatment with 1 mmol/L of prostaglandin E2 (PGE2, 2 h) caused these bone cells to contract a soma, whereas 10 and 100 mmol/L PGE2 (2 h) caused 18%–30% of the bone cells to elongate and expose the undersurface. Incubation of the cultured osteoblasts with PGE2 at different time periods showed a bell-shaped pattern with the optimal response at 2 h of incubation. A similar reaction can be induced by treatment with prostaglandin E1 (PGE1) or dibutyryl cyclic adenosine monophosphate (DBcAMP) in combination with 3-isobutyl-1-methylxanthine (IBMX). Furthermore, we assessed the percentage of responsive isolated bone cells to investigate interactions with other agents. The morphological changes induced by PGEs were inhibited by H-8, a protein kinase inhibitor. On the other hand, elevated intracellular calcium enhanced the PGE-induced morphological changes. Fluorescence labeling showed that PGEs caused the breakdown of the actin microfilaments, but spared the microtubules and vimentin filaments in the isolated osteoblast-like cells. These results suggest that the morphological changes of osteoblasts induced by PGEs may be related to the intracellular cAMP and calcium levels. (Bone 22:629 – 636; 1998) © 1998 by Elsevier Science Inc. All rights reserved. Key Words: Prostaglandin E; Rat osteoblast; Protein kinase; Cytoskeleton. Introduction There exists an intimate relationship between the morphology and function of bone cells that are regulated by many local and systemic factors.2,16,29 For example, the change in the osteoblast shape may regulate the access of the osteoclast to the bone surface during the bone remodeling process.38 Many hormones, such as calcitonin, prostaglandins (PGs), and parathyroid hormone (PTH), have been demonstrated not only to regulate the
Materials and Methods Calvaria Explant Culture Neonatal Wistar rats, weighing 5–7 g, were dissected using an aseptic technique under general anesthesia (pentobarbital 7.5 mg/mL, intraperitoneal injection 0.015 mL/g BW). The calvaria were removed immediately into Dulbecco’s modified Ca21-free Eagle’s medium (DMEM) (Gibco, Grand Island, NY) supple-
Address for correspondence and reprints: Rong-Sen Yang, M.D., Ph.D., Department of Orthopaedics, National Taiwan University Hospital, No. 7, Chung-Shan South Road, Taipei, Taiwan. E-mail: [email protected]. edu.tw © 1998 by Elsevier Science Inc. All rights reserved.
629
8756-3282/98/$19.00 PII S8756-3282(98)00055-6
630
R.-S. Yang et al. Prostaglandin-induced morphological changes in osteoblasts
mented with 10% fetal calf serum (Gibco), glutamine (58.5 mg/mL, Gibco), MgSO4 (200 mg/mL), and antibiotics (100 U penicillin and 100 U-streptomycin sulfate per milliliter, pH 7.6). They were kept in a humidified atmosphere containing 95% air and 5% CO2 at 37°C. All connective tissues, periosteum, and dura mater were carefully removed in phosphate-buffered saline (PBS) under a dissection microscope. The endocranial aspect of the parietal bone was carefully protected during dissection to keep the intact osteoblasts in situ for further investigation. A total of more than 50 pieces parietal calvaria were included for study. In order to study the maximal number of endocranial osteoblasts in situ, primary cultures of calvaria were immediately used for scanning electron microscopy (SEM) study to assess the PGEinduced morphological changes at various concentrations. All experiments were repeated at least six times. Cultures of Isolated Rat Osteoblasts Isolated osteoblasts from long bone fragments of neonatal rats were prepared as previously described.46,47 Briefly, the long bones of Wistar neonatal rats, including the femur, tibia, humerus, radius, and ulna, were dissected with an aseptic technique under anesthesia (pentobarbital) as previously described. After defleshing, the bones were diced into bare bone fragments and washed in PBS to remove the marrow cells. The primarily separated rat bone cells obtained from these explants were grown in the same culture medium and atmosphere as the calvaria culture. The viability of the cells was monitored by the trypanblue exclusion test. Explant pieces were removed when a confluence of the resulting monolayer was approached. Cell passage was carried out by incubating the monolayer for 5–10 min in Ca21- and Mg21-free Tyrode’s solution containing 0.25% trypsin (Sigma Chemical, St. Louis, MO) and replating the cells in fresh medium at about one third of their confluent density (10,000 –20,000 cells). Over 80% of the cells reattached within 24 h. The culture medium was changed twice or thrice weekly. The morphology of the cultured bone cells was observed under a phase contrast microscope. The characteristics of the cultured rat bone cells were identified by morphology identification, cytochemical staining, and biochemical assay of alkaline phosphatase (ALP) (more than 78% of cells were stained positive), SEM, transmission electron microscope (TEM), and monoclonal antibodies of osteocalcin (more than 95% positive).47 Cultured bone cells at confluence were used for study. It took 7–10 days of culture before the experiment. All experiments were performed on the cells with less than three passages. The isolated bone cells were used for study of the PGE-induced morphological changes and their possible mechanisms. SEM For SEM studies, the endocranial osteoblasts on the calvaria or on the glass slip of the primary culture were fixed in 2.5% glutaraldehyde in 0.1 mol/L cacodylate for 20 min. At completion of the fixation, the cells were rinsed in a phosphate buffer for 5 min, postfixed in 1% osmium tetroxide, dehydrated in a graded series of alcohol, and subsequently critical-point dried with liquid carbon dioxide. The specimens were ion sputter coated with a thin layer (about 10 nm) of gold and examined in a JEOL T330A scanning electron microscope at 15 or 30 kV. The morphological changes after treatment with PGE1 (1, 10, or 100 mmol/L, 2 h), PGE2 (1, 10, or 100 mmol/L, 2 h) and, DBcAMP (1 mmol/L) plus 3-isobutyl-1-methylxanthine (IBMX) (10 mmol/L) (2 h) were compared to those of control osteoblasts treated with a vehicle. PGE was dissolved in ethanol to make a 10 mmol/L stock solution, and this was added to the culture
Bone Vol. 22, No. 6 June 1998:629 – 636
medium to make a final concentration of 1, 10, and 100 mmol/L. Thus, the concentration of ethanol ranged from 0.01% to 1% vol/vol, and control experiments were given a matched aliquot of ethanol without PGEs. All experiments were repeated at least six times. Estimation of the Percentage of Slender Cells in Cultured Osteoblasts For studies of the interaction effects of various agents with PGEs, the isolated rat osteoblasts were plated in 3-cm collagen-coated tissue culture dishes and treated with PGEs or other agents at various concentrations for different time periods. The control osteoblasts were treated with an ethanol vehicle (for PGE experiments) of the same concentration as the experimental group. The morphological changes of bone cells induced by PGE were monitored with light microscopy (3200 or 3400 Nikon Diaphot). The responsive cells were defined as the cells with cytoplasmic processes longer than twofold of the control cell long axis. The number of responsive cells showing such morphological changes before and after treatment with PGE and various drugs were counted and expressed as the population ratio of responsive cell (%). We took pictures under phase contrast microscopy of at least 10 areas under 3200 magnification for each experiment and counted the percentage of responsive cells (at least 150 cells were counted for each area). The time course and dose response after treatment with PGE1 (1, 10, and 100 mmol/L), PGE2 (1, 10, and 100 mmol/L), and DBcAMP (1 mmol/L) plus IBMX (10 mmol/L) (all from Sigma) were investigated. The effects of H-8 (N[2-(methylamino-)ethyl]-5-isoquinoline sulfonamide) (10 mmol/L, 2 h) and A23187 (2 mmol/L, 18 h) (all from Sigma) on the PGE-induced morphological changes were studied. H-8 was dissolved in PBS to make a 10 mmol/L stock solution, and A23187 was dissolved in dimethyl sulfoxide (DMSO) to make a 2 mmol/L stock solution. These solutions were added to culture medium to make a final concentration of 10 mmol/L each. The same concentration of solvent was added to control experiments without agents. The cell viability after treatment with PGE or various agents was monitored by the trypan-blue exclusion test, and proved to be more than 93% viable. All experiments were repeated at least six times. Statistical analyses were evaluated by Student’s t-test (twotailed) and analysis of variance (ANOVA). A Bonferroni multiple comparison test was also used to analyze the results. Fluorescent Labeling of Microfilaments, Microtubules, and Vimentin Filaments Cytoskeleton studies were performed in the primary culture of the osteoblasts. For F-actin staining, the cells were washed three times with PBS at 5 min intervals and then fixed with 3.7% formaldehyde for 10 min. After fixation, cells were washed three times with PBS and then permeabilized with acetone (220°C) for 3–5 min. Air-dried cells were then stained with fluorescein isothiocyanate (FITC)-phallotoxin in PBS (Molecular Probes, Eugene, OR) for 30 min, washed again with PBS, mounted in glycerol-PBS containing 0.1 mol/L n-propyl gallate, and examined immediately for fluorescent staining. In control experiments, cells were incubated with unlabeled phallotoxin before exposure to fluorescent probes. There was little or no labeling observed. For the microtubule study, a mouse anti-a-tubulin primary antibody (Sigma) was used. Rat osteoblasts were washed three times with a PHEM solution [60 mmol/L piperazine-N,N9-bis [2-ethanesulfonic acid] (PIPES), 25 mmol/L N-2-hydroxyeth-
Bone Vol. 22, No. 6 June 1998:629 – 636
R.-S. Yang et al. Prostaglandin-induced morphological changes in osteoblasts
631
ylpiperazine-N-2-ethanesulfonic acid (HEPES), 2 mmol/L MgCl2, 10 mmol/L ethyleneglycol-bis-(b-aminoethylether)N,N,N9,N9-tetraacetic acid (EGTA), pH 6.9] (all from Sigma) at 5 min intervals and then fixed with 100% methanol (220°C) for 3–5 min, followed by washing with PBS. The osteoblasts were incubated with a 1:100 mouse primary anti-a-tubulin antibody at 37°C for 1 h. After washing with PBS, the cells were incubated with 1:50 goat-antimouse FITC-conjugated secondary antibody (Sigma) at 37°C for 1–2 h. The cells were washed again with PBS and prepared as above. In control experiments, cells were incubated only with a secondary antibody and showed little or no labeling. For vimentin staining, a mouse antivimentin primary antibody (Sigma) was used. Rat osteoblasts were washed three times with a PBS solution at 5 min intervals and then fixed with 0.15% glutaraldehyde for 3–5 min, followed by washing with PBS. The osteoblasts were incubated with a 1:80 mouse primary antivimentin antibody at 37°C for 1 h. After washing with PBS, the cells were incubated with a 1:50 goat-antimouse FITC-conjugated secondary antibody at 37°C for 1–2 h. The cells were then prepared as mentioned above. All above experiments were repeated at least six times. Results PGE-Induced Morphological Changes in Rat Calvaria Osteoblasts Under the SEM, the endocranoial osteoblasts on the control calvaria were flat and polygonal in shape, and the cell surface was coated with a few microvilli. They were arranged in a single layer side by side and covered the underlying collagen fibrils (Figure 1A). Incubation of the rat calvaria with PGE2 (1 mmol/L, 2 h) induced the contraction of the soma and a few of the cells were elongated (Figure 1B). Higher concentrations of PGE2 (10 or 100 mmol/L, 2 h) caused more endocranial osteoblasts (40%– 80%) to elongate and expose the undersurface of the calvaria (Figure 1C). Incubation of the calvaria with PGE1 (1, 10, or 100 mmol/L, 2 h) or DBcAMP (1 mmol/L) plus IBMX (10 mmol/L) (2 h) also induced similar shape changes. PGE-Induced Morphological Changes in Cultured Rat Osteoblasts Under SEM, control cultured rat osteoblasts were polygonal with their own long axis (Figure 2A). The soma of cultured osteoblasts contracted and a few of the osteoblasts were elongated (10%–15%) after treatment with a low concentration of PGE2 (1 mmol/L, 2 h) (Figure 2B). However, after treatment with higher concentrations of PGE2 (10 or 100 mmol/L, 2 h), more osteoblasts (18%–30%) were progressively elongated (Figure 2C). The changes in cell shape observed under phase contrast light microscopy were similar (Figure 3A and B). Treatment with higher concentrations of PGE1 (10 or 100 mmol/L, 2 h) or DBcAMP (1 mmol/L) plus IBMX (10 mmol/L) (2 h) also induced similar morphological changes. Estimation of the Percentage of Slender Cells in Cultured Osteoblasts Because incubation with a concentration of lower than 1 mmol/L did not induce the morphological change, we treated the cultured rat osteoblasts with various concentrations of PGE2 (1, 10, and 100 mmol/L) for different time periods. The results showed a bell-shaped pattern with an optimal response at 2 h of incubation (Figure 4A). The PGE-induced slender cell formation of isolated
Figure 1. Morphological changes of rat calvaria osteoblasts induced by PGE2 under SEM. (A) Control osteoblasts treated with ethanol vehicle show sparse short cytoplasmic processes. (B) Incubation with 1 mmol/L PGE2 (2 h) caused the contraction of soma. (C) Treatment with 100 mmol/L PGE2 (2 h) induced significant elongation of calvaria and developed numerous long cytoplasmic projections. All experiments were repeated at least six times.
rat osteoblasts was concentration dependent. The percentage of responsive cells after incubation with PGE2 (100 mmol/L) reached a peak level (;30%) at 2 h. The dose-response curve of the peak effect at 2 h showed a positive correlation (r 5 0.91, p , 0.0001). The morphological changes were reversible and some cells returned to their normal appearance later. The results obtained after treatment with PGE1 also showed a bell-shaped response curve (Figure 4B). Furthermore, we investigated the effect of H-8, a protein kinase inhibitor, on the PGE-induced morphological changes of cultured rat osteoblasts. Pretreatment with H-8 (10 mmol/L, 2 h) decreased the percentage of slender cells induced by 100 mmol/L PGE2 from 28.6% 6 2% to 14.1% 6 2.1% (p , 0.05) (Figure 5). A similar inhibitory effect of H-8 on the morphological
632
R.-S. Yang et al. Prostaglandin-induced morphological changes in osteoblasts
Bone Vol. 22, No. 6 June 1998:629 – 636
Figure 3. Morphological changes of primary cultures of rat osteoblasts induced by PGE2 under phase contrast microscope. (A) Control osteoblasts treated with an ethanol vehicle show sparse short cytoplasmic processes. (B) Treatment with 100 mmol/L PGE2 (2 h) induced osteoblasts to elongate. All experiments were repeated at least six times.
IBMX (10 mmol/L) (up to 96% at 24 h) (Figure 6B and C). Thus the morphological changes of the osteoblasts induced by PGEs may probably be related to the intracellular cAMP and Ca21 levels. Figure 2. Morphological changes of primary cultures of rat osteoblasts induced by PGE2 under SEM. (A) Control osteoblasts treated with ethanol vehicle show a sparse flat polygonal shape with short cytoplasmic processes. (B) Treatment with 1 mmol/L PGE2 (2 h) induced contraction of some osteoblasts. (C) Treatment with 100 mmol/L PGE2 (2 h) induced a more elongated cell formation and displayed numerous long and slender cytoplasmic projections. All experiments were repeated at least six times.
changes induced by PGE1 (from 32.1% 6 2.1% to 11.2% 6 1.5%, p , 0.05) or DBcAMP plus IBMX (26.7% 6 1.9% vs. 16.8% 6 1.8%, p , 0.05) was also observed. Treatment with Ca21 ionophore A23187 (2 mmol/L) alone up to 18 h caused 15%–20% of osteoblasts to elongate. However, treatment with DMSO of the same concentration did not affect the morphology of the cells. The proportion of slender cells induced by PGE2 (10 mmol/L) increased up to 40% at 24 h in the cultured osteoblasts pretreated with A23187 (2 mmol/L, 18 h) (p , 0.05) (Figure 6A). Pretreatment with A23187 (2 mmol/L, 18 h) also increased the shape change induced by either PGE1 (10 mmol/L) (up to 74% at 24 h) or DBcAMP (1 mmol/L) plus
Roles of Cytoskeletal Proteins in the Action of Prostaglandin The normal rat osteoblasts showed a flat polygonal shape with stress fibers running a straight course throughout the cytoplasm (Figure 7A). However, the responsive cells induced by the prostaglandin showed the breakdown of actin microfilaments and diffused F-actin staining within the cells (Figure 7B). On the other hand, the distribution pattern of the microtubule (Figure 7C and D) and vimentin filaments (Figure 7E and F) in osteoblasts were not significantly affected by the prostaglandin, even though accompanied with the marked morphological changes of the cells. Discussion In this study we have demonstrated that treatment of rat calvaria osteoblasts and isolated osteoblasts from long bone fragments with PGEs (1 mmol/L, 2 h) induced the contraction of soma and a few cells were elongated, whereas higher concentrations of PGEs (10 and 100 mmol/L, 2 h) caused 18%–30% of the osteoblasts to develop the rapid elongation of soma, resulting in a morphological transformation. The PGE-
Bone Vol. 22, No. 6 June 1998:629 – 636
Figure 4. Concentration-dependent induction of morphological changes by PGEs and DBcAMP plus IBMX in cultured rat osteoblasts. (A) Time course of the effect of PGE2 (filled circle: 100 mmol/L; filled square: 10 mmol/L; filled triangle: 1 mmol/L). (B) The time-course curves of PGE1-induced morphological changes were similar (filled circle: 100 mmol/L; filled square: 10 mmol/L; filled triangle: 1 mmol/L). Data presented as mean 6 standard deviation (SD) (n 5 6). *p , 0.05 vs. 1 mmol/L at each time point.
induced slender cell transformation of rat osteoblasts was related to the breakdown of actin microfilaments. In addition, pretreatment with H-8 decreased whereas A23187 increased the percentage of responsive cells. Thus, the elevated intra-
Figure 5. The inhibitory effect of H-8 on the actions of prostaglandins inducing morphological changes. The H-8 antagonized the effect of 100 mmol/L PGE and DBcAMP (1 mmol/L) plus IBMX (10 mmol/L) induced morphological changes of cultured osteoblasts. Data presented as mean 6 SD (n 5 6). *p , 0.05.
R.-S. Yang et al. Prostaglandin-induced morphological changes in osteoblasts
633
Figure 6. Enhancement by A23187 on the prostaglandin-induced morphological changes in cultured rat osteoblasts. (A) Pretreatment with A23187 (2 mmol/L, 18 h) showed that 15%–20% of the osteoblasts changed shape. Pretreatment with A23187 (2 mmol/L, 18 h) followed by addition of PGE2 (10 mmol/L) (filled square) increased a higher proportion of slender cells of the osteoblasts than the controls (filled circle). The effects of PGE1 (10 mmol/L) (B) and DBcAMP (1 mmol/L) 1 IBMX (10 mmol/L) (C) on the morphological changes were also enhanced by A23187 [filled circle: controls; filled square: pretreatment with A23187 (2 mmol/L, 18 h)]. Data presented as mean 6 SD (n 5 6). *p , 0.05 for comparison at each time point.
cellular cAMP and calcium levels may be related to these PGE-induced morphological changes. The rat calvaria model provides a good model for the study of the bone cell biology with special reference to cell– cell and cell–matrix interactions.6,18,19,28,30,39,45 The osteoblasts are arranged in a monolayer with their own long axis. The polygonal calvaria osteoblasts elongated after exposure to PTH.18 –21 However, the effect of PGE2 on the morphological changes of osteoblasts was controversial.1,4 Ali et al. demonstrated that 1 mmol/L PGE2 did not cause a shape change of the rat calvaria osteoblasts.1 However, a rapid change of the shape of cultured osteoblasts could be induced after treatment with PGEs.4 In the current study, we used a higher concentration of PGE and demonstrated a dose-dependent effect of PGE-induced morphological changes of either calvaria osteoblasts or cultured bone cells from long bone fragments of neonatal rats. These results correspond to other reports.30,39 However, the effects of PGEs on the isolated cultured osteoblasts in this study differed from those on the calvaria osteoblasts, which preserved intact cell–matrix interactions. Many agents have been proven to increase the intracellular cAMP of osteoblasts and osteosarcoma cells.3,12–14 The mode of actin of PGE probably involves the accumulation of cAMP or
634
R.-S. Yang et al. Prostaglandin-induced morphological changes in osteoblasts
Bone Vol. 22, No. 6 June 1998:629 – 636
Figure 7. Effects of PGE2 on the organization of cytoskeletons in cultured rat osteoblasts under fluorescence microscopy. (A) The control cells treated with an ethanol vehicle showed typical stress fibers of actin throughout the cytoplasm. (B) Two hours after treatment with 100 mmol/L PGE2, bundles of actin fibers disappear, accompanied with the cell transformation. The microtubules [(C) and (D)] and vimentin [(E) and (F)] were not significantly affected after treatment with PGE2. All experiments were repeated at least six times.
elevated intracellular calcium in the bone cells. In the present study, treatment with PGE2, PGE1, or DBcAMP plus IBMX induced a similar shape change in the osteoblasts. Furthermore, the inhibition of the PGE-induced slender cell formation by H-8, a preferential protein kinase A inhibitor, suggests that a cyclic AMP-dependent protein kinase may be involved. Thus the morphological change may be closely related to an increased level of intracellular cAMP. Previous studies have demonstrated that an increased intracellular cAMP level is closely related to the shape change in various cell types.22,23,25,30,36,48 The contracted cytoplasm could respread when the cAMP returned to normal levels.3,4,41 On the other hand, H-8 was not fully inhibitory, perhaps suggesting additional mechanisms. H-8 may also inhibit protein kinase C, which may be related to regulation of the cell morphology. In addition, intracellular calcium may also be involved in the PGE-induced changes in cell morphology. The possible interaction of the mechanism merits further investigation. PGE has been demonstrated to cause the dissolution of actin microfilament bundles and inhibit the induction of actin polymerization or the phosphorylation of myosin light chain kinase.4,24 Although the role of kinases in osteoblasts is not well understood, it has been shown that only thin filaments reaching the cortex of the cell are associated with the soma and serve as the anchoring mechanism of the cells. In this study, treatment with PGEs induced the breakdown of actin microfilaments without affecting microtubule and vimentin filaments. These results indicated that the effect of prostaglandin on osteoblasts may be
intimately related to the breakdown of the actin microfilament. Lomri et al. demonstrated that cAMP and calcium can modulate the synthesis and organization of the cytoskeleton of mouse osteoblasts.26,27 The marked elevation of intracellular cAMP in the osteoblasts may induce the dissolution of the microfilament. However, the complexity of the phosphorylation pathways makes it difficult to sort out the exact mechanisms of the action of the prostaglandin. Other studies showed that PGE2 may enhance the metabolism of phosphoinositol and the IP3 level increases in the mouse osteoblast and osteoclast.11,13 Elevated IP3 may mobilize the release of Ca21 from the nonmitochondrial pool. Therefore, the Ca21 level is increased after treatment with PGE2. The present study demonstrated that the elevation of intracellular Ca21 by A23187 is able to enhance the PGE-induced slender cell formation. Thus the PGE-induced morphological changes may be related to the elevated intracellular Ca21. However, the definite interaction of the mechanisms needs further investigation. On the other hand, in the current study, we demonstrated that normal microtubule and vimentin assembly was preserved after treatment with PGEs. This finding corresponds to other studies.5 In addition, Shen et al.39 found that colchicine, an inhibitor of microtubule polymerization, inhibited the shape changes of the osteoblast induced by PTH and/or PGE2. Therefore, the internection and interaction among the microtubule, vimentin, and actin may be closely related to achieve the final shape of the osteoblasts. These organizations among cytoskeletal proteins are prob-
Bone Vol. 22, No. 6 June 1998:629 – 636
ably involved in the cell morphology induced by hormones and nucleotides. In the normal bone remodeling process, changes of the osteoblast morphology may expose the underlying matrix and provide a convenient access to osteoclasts.38 Bone remodeling also plays an important role in the pathophysiology of various kinds of clinical disorders. The bone remodeling is regulated by many factors, including PGs. In fetal rat calvaria cultures, PGE2 and other PGs (PGE1, PGF1a, PGA1, PGA2, 0.1–100 mmol/L) inhibited the incorporation of the radioproline into newly formed collagen without affecting the synthesis of noncollagen protein.17 The effect was latent, not acute, and occurred only at the higher end of the dose range. On the other hand, the local tissue levels of PGE2 in cirvicular fluid of periodonitis approach a concentration of 5 mmol/L or higher,33–35 suggesting a higher local concentration of PGE2 at the interface of the resorbing bone surface. The local calcium levels released at the resorbing bone surface is probably high. From the current study, local high concentrations of PGEs and calcium may act synergistically and potentiate the morphological changes of bone cells to expose more bone matrix. In addition, the biological response of PGEs differs in the presence of other factors.35,42 Therefore, the definite clinical implication of the PGE-induced cell morphological changes in the current study merits further investigation. In conclusion, this study suggested that PGE-induced morphological changes of cultured osteoblasts were intimately related to the cytoskeletal microfilaments. The elevated intracellular cAMP or Ca21 may be closely related to the morphological changes.
R.-S. Yang et al. Prostaglandin-induced morphological changes in osteoblasts
14.
15.
16.
17.
18. 19.
20.
21.
22.
23.
Acknowledgments: This work was supported by grants from the National Science Council (NSC82-0115-B002-551 and NSC81-0412-B002-649).
24.
References
25.
1. Ali, N. N., Melhuish, P. B., Boyde, A., Bennett, A., and Jones, S. J. Parathyroid hormone, but not prostaglandin E2, changes the shape of osteoblasts maintained on bone in vitro. J Bone Miner Res 5:115–121; 1990. 2. Arnett, T. R. Update of bone cell biology. Eur J Orthodont 12:81–90; 1990. 3. Atkins, D. and Martin, T. J. Rat osteogenic sarcoma cells: Effects of some prostaglandins, their metabolites and analogues on cyclic AMP production. Prostaglandins 13:861– 871; 1977. 4. Aubin, J. E., Alders, E., and Heersche, J. N. M. A primary role for microfilaments, but not microtubules, in hormone induced cytoplasmic retraction. Exp Cell Res 143:439 – 450; 1983. 5. Beertsen, W., Heersche, J. N. M., and Aubin, J. E. Free and polymerized tubulin in cultured bone cells and Chinese hamster ovary cells: The influence of cold and hormones. J Cell Biol 95:387–393; 1982. 6. Boyde, A., Jones, S. J., Binderman, I., and Harrell, A. Scanning electron microscopy of bone cells in culture. Cell Tissue Res 166:65–70; 1976. 7. Chambers, T. J., Athanasou, N. A., and Fuller, K. Effect of parathyroid hormone and calcitonin on the cytoplasmic spreading of isolated osteoclasts. J Endocrinol 102:281–286; 1984. 8. Chambers, T. J. and Dunn, C. J. The effect of parathyroid hormone, 1,25dihydroxy-cholecalciferol and prostaglandins on the cytoplasmic activity of isolated osteoclasts. J Pathol 137:192–203; 1982. 9. Chambers, T. J., Fuller, K., and Athanasou, N. A. The effect of prostaglandin I2, E1, E2, and dibutyryl cyclic AMP on the cytoplasmic spreading of rat osteoclasts. Br J Exp Pathol 65:557–566; 1984. 10. Chambers, T. J. and Moore, A. The sensitivity of isolated osteoclasts to morphological transformation by calcitonin. J Clin Endocrinol Metab 57:819 – 824; 1983. 11. Dolce, C., Schneeberger, G., and Dziak, R. Effects of prostaglandin E2 on phosphoinositide metabolism in osteoblastic cells. J Dent Res 66:1577; 1987. 12. Dziak, R. M., Hurd, D., Miyasaki, K. T., Brown, M., Weinfeld, N., and Hausmann, E. Prostaglandin E2 binding and cyclic AMP production in isolated bone cells. Calcif Tissue Int 35:243–249; 1983. 13. Farndale, R. W., Sandy, J. R., Atkinson, S. J., Pennington, S. R., Meghji, S.,
26.
27.
28.
29. 30. 31.
32.
33.
34.
35.
36.
635
and Meikle, M. C. Parathyroid hormone and prostaglandin E2 stimulate both inositol phosphates and cyclic AMP accumulation in mouse osteoblast cultures. Biochem J 252:263–268; 1988. Farr, D., Pochal, W., Brown, M., Shapiro, E., Weinfeld, N., and Dziak, R. Effects of prostaglandins on bone cell calcium. Arch Oral Biol 29:885– 891; 1984. Gadher, S. J. and Woolley, D. E. Comparative studies of adherent rheumatoid synovial cells in primary cultures: Characterization of the dendritic (stellate) cell. Rheumatol Int 7:13–22; 1987. Hagel-Bradway, S., Tatakis, D. N., and Dziak, R. Prostaglandin-induced changes in calcium uptake and cAMP production in osteoblast-like cells: Role of protein kinase C. Calcif Tissue Int 48:272–277; 1989. Harvey, W. Source of prostaglandins and their influence on bone resorption and formation. In: Harvey, W. and Bennett, A., Eds. Prostaglandins and Bone Resorption. Boca Raton, FL: CRC; 27, 1988. Jones, S. J. and Boyde, A. Morphological changes of osteoblasts in vitro. Cell Tissue Res 166:101–107; 1976. Jones, S. J. and Boyde, A. Experimental study of changes in osteoblastic shape induced by calcitonin and parathyroid extract in an organ culture system. Cell Tissue Res 169:449 – 465; 1976. Jones, S. J., Boyde, A., and Shapiro, I. M. The response of osteoblasts to parathyroid hormone (PTH 1-34) in vitro. Metab Bone Dis Relat Res 2:335– 338; 1981. Jones, S. J. and Ness, A. R. A study of the arrangement of osteoblasts of rat calvarium cultured in medium with or without added parathyroid extract. J Cell Sci 25:247–263; 1977. Jumblatt, M. M. and Paterson, C. A. Prostaglandin E2 effects on corneal endothelial cyclic adenosine monophosphate synthesis and cell shape are mediated by a receptor of the EP2 subtype. Invest Ophthalmol Visual Sci 32:360 –365; 1991. Kreisberg, J. I., Patel, P. Y., Venkatachalam, M. A., and Taylor, G. Elevations of intracellular cAMP result in a change in cell shape that resemble dome formation in cultured rat glomerular epithelial cells. In Vitro Cell Develop Biol 22:392–396; 1986. Lamb, N. J. C., Fernandez, A., Conti, M. A., Adelstein, R., Glass, D. B., Welch, W. W., and Feramisco, J. R. Regulation of actin microfilament integrity in living nonmuscle cells by the CAM-dependent protein kinase and the myosin light chain kinase. J Cell Biol 106:1955–1971; 1988. Lawrence, T. S., Ginzberg, R. D., Gilula, N. B., and Beers, W. H. Hormonally induced cell shape changes in cultured rat ovarian granulosa cells. J Cell Biol 80:21–36; 1979. Lomri, A., Marie, P. J., Escurat, M., and Portier, M. Cytoskeletal protein synthesis and organization in cultured mouse osteoblastic cells. FEBS Lett 222:311–316; 1987. Lomri, A. and Marie, P. Distinct effects of calcium- and cyclic AMP-enhancing factors on cytoskeletal synthesis and assembly in mouse osteoblastic cells. Acta Biochim Biophys 1052:179 –186; 1990. MacDonald, B. R. F., Gallagher, J. A., Ahnfekt-Ronne, I., Beresford, J. N., Gowen, M., and Russell, G. R. Effects of bovine parathyroid hormones and 1,25-dihydroxyvitamin D3 on the production of prostaglandins by cells derived from human bone. FEBS Lett 169:49 –52; 1984. Martin, T. J., Ng, K. W., and Suda, T. Bone cell physiology. Endocrinol Metab Clin 18:833– 858; 1989. Miller, S. S., Wolf, A. M., and Arnaud, C. D. Bone cells in culture: Morphological transformation by hormones. Science 192:1340 –1343; 1976. Murrills, R. J., Shane, E., Lindsay, R., and Dempster, D. W. Bone resorption by isolated human osteoclasts in vitro: Effects of calcitonin. J Bone Miner Res 4:259 –268; 1989. Neufeld, A. H., Jumblatt, M. M., Matkin, E. D., and Raymond, G. M. Maintenance of corneal endothelial cell shape by prostaglandin E2: Effects of EGF and indomethacin. Invest Ophthalmol Visual Sci 27:1437–1442; 1986. Offenbacher, S., Odle, B. M., and Dyke, T. E. The use of cervicular fluid prostaglandin E2 levels as a predictor of preiodontal attachment loss. J Periodontol 21:101–112; 1986. Offenbacher, S., Williams, R. C., Jeffcoat, M. K., Howell, T. H., Odle, B. M., Smith, M. A., Hall, C. M., Johnson, H. G., and Goldhaber, P. Effects of NSAIDs on beagle cervicular cyclooxygenase metabolites and periodontal bone loss. J Periodontol Res 27:207–213; 1991. Offenbacher, S., Heasman, P. A., and Collins, J. G. Modulation of host PGE2 secretion as a determinant of periodontal disease expression. J Periodontol 64:432– 444; 1993. Porter, K. R., Puck, T. T., Hsie, A. W., and Kelly, D. An electron microscope
636
37.
38. 39.
40.
41.
42.
43.
R.-S. Yang et al. Prostaglandin-induced morphological changes in osteoblasts study of the effects of dibutyryl cAMP on Chinese hamster ovary cells. Cell 2:145–162; 1974. Powles, T. J., Easty, D. M., Easty, G. C., Bondy, P. K., and Munro-Neville, A. Aspirin inhibition of in vitro osteolysis stimulated by parathyroid hormone and PGE1. Nature (New Biol) 245:83– 84; 1973. Rodan, G. A. and Martin, T. J. The roles of osteoblasts in hormonal control of bone resorption—a hypothesis. Calcif Tissue Int 33:349 –351; 1981. Shen, V., Rifas, L., Kohler, G., and Peck, W. A. Prostaglandins change cell shape and increase intercellular gap junctions in osteoblasts cultured from rat fetal calvaria. J Bone Miner Res 1:243–249; 1986. Smith, T. J., Sempowski, G. D., Wang, H. S., Del Vecchio, P. J., Lippe, S. D., and Phipps, R. P. Evidence for cellular heterogenercity in primary cultures of human orbital fibroblasts. J Clin Endocrinol Metab 80:2620 –2625; 1995. Spruill, W. A., White, M. G., Steiner, A. L., Tres, L. L., and Kierszenbaum, L. Temporal sequence of cell shape changes in cultured rat sertoli cells after experimental elevation of intracellular cAMP. Exp Cell Res 131:131–148; 1981. Stashenko, P., Dewhirst, F. E., Peros, W. J., Kent, R. L., and Ago, J. M. Synergistic interactions between interleukin 1, tumor necrosis factor, and lymphotoxin in bone resorption. J Immunol 138:1464 –1468; 1987. Tashjian, Jr., A. H., Ivey, J. L., Delclos, B., and Levine, L. Stimulation of prostaglandin production in bone by phorbol diesters and melittin. Prostaglandins 16:221–232; 1978.
Bone Vol. 22, No. 6 June 1998:629 – 636 44. Tashjian, Jr., A. H. and Levine, L. Epidermal growth factor stimulates prostaglandin production and bone resorption in cultured mouse calvaria. Biochem Biophys Res Commun 85:966 –975; 1978. 45. Yang, C. Y., Gonnerman, W. A., Taylor, L., Nimberg, R. B., and Polgar, P. R. Synthetic human parathyroid hormone fragment stimulates prostaglandin E2 synthesis by chick calvaria. Endocrinology 120:63– 69; 1987. 46. Yang, R. S., Lu, K. S., Fu, W. M., Liu, T. K., and Lin-Shiau, S. Y. Staurosporine-induced morphological changes in the rat osteoblasts. Cell Biol Int 17:75– 82; 1992. 47. Yang, R. S., Liu, T. K., Tsai, K. S., Lin-Shiau, S. Y., and Lu, K. S. Morphological and immunocytochemical characterization of osteoblast cultures from long bones of neonatal rats. Arch Histol Cytol 55:415– 422; 1992. 48. Wang, H. S., Keese, C. R., Giaever, I., and Smith, T. J. Prostaglandin E2 alters human orbital fibroblast shape through a mechanism involving the generation of cyclic adenosine monophosphate. J Clin Endocrinol Metab 80:3553–3560; 1995.
Date Received: November 15, 1996 Date Revised: April 3, 1997 Date Accepted: February 19, 1998