Effect of pyrimidine deoxynucleosides and sodium butyrate on expression of the glycoprotein hormone alpha-subunit and placental alkaline phosphatase in HeLa cells

Production of the glycoprotein hormone common alpha-subunit and placental alkaline phosphatase activity can be modulated in HeLa cells by a variety of deoxynucleosides. Dose response curves for thymidine (Thd), fluorodeoxyuridine (FdUrd), bromodeoxyuridine (BrdUrd) and iododeoxyuridine (IdUrd) demonstrate that, in general, alkaline phosphatase was increased by lower concentrations of inducer than was alpha-subunit. The deoxynucleosides were not as effective as sodium butyrate as inducers of either protein. Whereas Thd and the halogenated dUrd derivatives enhanced protein expression, deoxycytidine (dCyd) had negative effects. Induction by deoxynucleosides of both alkaline phosphatase and alpha-subunit was inhibited by dCyd, but induction of alkaline phosphatase by butyrate was more sensitive to dCyd inhibition than was the butyrate-mediated induction of alpha-subunit. These results suggest that the two proteins are not regulated in a coordinate manner. Reversal of alkaline phosphatase induction by dCyd was not observed in cells preincubated with sodium butyrate for 6-24 h before the addition of dCyd, indicating that the deoxynucleoside interferes with an early event in the butyrate-mediated response. Combinations of butyrate with Thd, BrdUrd or IdUrd were synergistic with respect to the induction of HeLa-alpha. It is concluded that incorporation of the deoxynucleosides into DNA may not be required for the synergistic response since 2',5'-dideoxythymidine was an effective as Thd. Cytoplasmic dot hybridizations demonstrate that a primary effect of the various effectors is to increase the steady-state levels of alpha-subunit mRNA. There was a good correlation between alpha-subunit accumulation and corresponding levels of alpha-mRNA, suggesting that regulation occurs at a pretranslational site. Although the mechanism(s) is not understood, these data provide evidence that nucleosides or their derivatives can significantly affect gene expression.     

Induction of teratocarcinoma cell differentiation : Effect of the inhibitors of DNA synthesis

Induction of teratocarcinoma cell (F9) differentiation was studied by using inhibitors of DNA synthesis and several agents known to be differentiation inducers. Inhibition of DNA synthesis induced changes in cell surface marker F9 and stimulated the production of plasminogen activator (PA) in a manner that is dependent upon de novo synthesis of RNA and protein. The results thus indicate close association between inhibition of DNA synthesis and induction of cell differentiation. This approach will be useful in investigating the mechanism of teratocarcinoma cell differentiation.     

Induction of human choriogonadotropin in HeLa-cell cultures by aliphatic monocarboxylates and inhibitors of deoxyribonucleic acid synthesis

The ectopic production of the glycopeptide hormone human placental choriogonadotropin by HeLa₆₅ cells was measured by radioimmunoassay with antiserum against the β-subunit of choriogonadotropin and with the ¹²⁵I-labelled β-subunit as a tracer antigen. Choriogonadotropin synthesis was markedly (500-fold) stimulated by sodium butyrate. Kinetic studies and the use of an inhibitor of protein synthesis, cycloheximide, indicated that protein synthesis was required for this induction. Investigation of the efficiency of 22 aliphatic short-chain fatty acids and derivatives in causing increased choriogonadotropin synthesis by HeLa cells showed stringent structural requirements. Induction of choriogonadotropin synthesis in HeLa cells was not restricted to butyrate. Other aliphatic acids (propionate, isobutyrate, valerate and hexanoate) were also capable of inducing choriogonadotropin synthesis at 10–50% of the efficiency of butyrate. Hydroxy derivatives of monocarboxylate inducers, related mono- and di-carboxylic acids, alcohols, amines, ketones, esters and sulphoxide were ineffective in increasing choriogonadotropin production by HeLa cells. A saturated C₄ straight-chain acid without substituent hydroxyl groups but with a methyl group at one end and a carboxyl moiety at the other appeared to be most efficient in activating choriogonadotropin production. A second clonal line of HeLa cells, HeLa₇₁, showed a higher constitutive synthesis of choriogonadotropin than HeLa₆₅ cells, which was also markedly increased by butyrate. Butyrate and other aliphatic monocarboxylate inducers of choriogonadotropin synthesis inhibited HeLa-cell growth and DNA synthesis. This inhibition of DNA replication may be related to the mechanism of choriogonadotropin synthesis, since two well-characterized anti-neoplastic inhibitors of DNA synthesis, hydroxyurea and 1-β-d-arabinofuranosylcytosine, also stimulated a 300-fold increase in choriogonadotropin synthesis in HeLa cells and were synergistic with butyrate in promoting choriogonadotropin synthesis. Thus activation in tumour cells of genes normally expressed by foetal tissue and the consequent ectopic synthesis of polypeptide hormones may require neither cell division nor DNA synthesis.     

Expression and cAMP-mediated regulation of the human gonadotropin alpha subunit gene in transfected mouse L-cells

Expression of the dimeric glycoprotein hormone, human chorionic gonadotropin (HCG), occurs either eutopically in placental trophoblast cells and trophoblastic tumor cells (choriocarcinoma) or ectopically in nontrophoblastic tumor cells. However, regulation of constitutive HCG-subunit mRNA production appears to differ in trophoblastic and nontrophoblastic cells, as evidenced by the fact that cAMP analogs and agonists enhance eutopic but not ectopic HCG-subunit mRNA synthesis. In the present study, we compared the effects of cAMP on HCG alpha-subunit expression in human choriocarcinoma cells and in nontrophoblastic mouse L-cells stably transfected with the HCG alpha-subunit gene. Constitutive levels of alpha-subunit expression in transfected mouse L-cells were equivalent to or exceeded those found in choriocarcinoma cells as determined by Northern blot analysis and indirect immunofluorescence for alpha-subunit protein. However, cAMP-mediated induction of alpha-subunit gene expression was retained in nontrophoblastic L-cells and closely paralleled that observed in human choriocarcinoma cells. These findings indicate that cells distinctly nontrophoblastic in origin may share the necessary cellular factors for cAMP-mediated induction of alpha-subunit gene expression. Failure of ectopic HCG-producing tumor cells to be stimulated by cAMP may thus be the result of deletion or mutation of such factors.     

Inhibition of S-phase progression in macrophages is linked to G1/S-phase suppression of DNA synthesis genes.

Some of the important controlling events regulating eukaryotic S-phase progression are considered to occur late in the G1 stage of the cell cycle. We show here that stimulation of DNA synthesis in bone marrow-derived macrophages (BMM) by macrophage CSF-1 is preceded by G1 expression of three genes which encode proteins associated with the DNA synthesis machinery--the M1 and M2 subunits of ribonucleotide reductase and proliferating cell nuclear Ag (PCNA). Increased expression for these genes correlated well with the mitogenic response and sustained expression required de novo RNA and protein synthesis and also the presence of CSF-1 for at least most of G1. Inhibitors of BMM proliferation (LPS, TNF-alpha, IFN-gamma, and cAMP elevating agents) suppressed CSF-1-induced expression of M1, M2, and PCNA mRNA measured at 22 h. This suppression occurred even when added up to 12 h after the CSF-1, a period coinciding with the G1/S-phase boundary. The delayed kinetics of this effect parallels the ability of these agents to maximally inhibit CSF-1-induced BMM DNA synthesis when added at similar times. Decreased expression of M1, M2, and PCNA was not merely a consequence of DNA synthesis inhibition because the S-phase inhibitor, hydroxyurea, did not suppress CSF-1-induced gene expression. These results suggest that inhibition of DNA synthesis by antiproliferative agents involves inhibition of expression of several genes associated with the DNA synthesis machinery.     

Competitive inhibition of nicking--closing enzymes may explain some biological effects of DNA intercalators

Intercalating agents cause varied and multiple biological effects. These include the inhibition of RNA and DNA synthesis, frameshift mutations and protein-associated DNA breaks. However, some non-intercalating analogs of intercalating compounds behave similarly. The model of DNA intercalation does not adequately explain all these biological effects. It is suggested here that intercalators and similar compounds may competitively inhibit the closing reaction of some nicking--closing enzymes. Hypothetical mechanisms built on this suggestion are presented for the formation of protein associated DNA breaks, frameshift mutation, inhibition of macromolecular synthesis, and recombination.     

Microtubule-disrupting agents reverse the inhibitory effect of interferon on mitogenesis in 3T3 cells

Interferon can inhibit the stimulation of DNA synthesis in quiescent 3T3 cells exposed to combinations of purified growth factors, but the extent of inhibition varies with the number and combination of mitogens used. As the number of growth factors used to stimulate the cells is increased from two to three, the inhibitory effect of IFN is reduced, and if the third mitogen is a microtubule-disrupting agent such as colchicine or nocodazole, it is abolished altogether. The antagonistic effect of microtubule-disrupting agents on interferon-induced inhibition of DNA synthesis suggests that an intact tubulin network is required for this action of interferon. Interferon and tubulin disrupting agents also show similar kinetics in establishing an effect on DNA synthesis which could imply that they have opposite effects on tubulin assembly.     

Increased alkaline phophatase activity in HeLa cells mediated by aliphatic monocarboxylates and inhibitors of DNA synthesis

Alkaline phosphatese activity of HeLa cells is increased from 3- to 8-fold during growth in medium with certain aliphatic monocarboxylates. The four-carbon fatty acid salt, sodium butyrate, is the most effective “inducer” with propionate (C₃), pentanoate (C₅) and hexanoate (C₆) having lesser effects. Other straight-chain aliphatic monocarboxylates, branched-chain analogues of inducers, hydroxylated derivatives, and metabolytes structurally related to butyrate are ineffective in mediating an increase in enzyme activity, indicating stringent structural requirements for inducers. The kinetics of increase in alkaline phosphatase activity in HeLa cells shows a 20–30 h lag period after adding the aliphatic acid followed by a rapid linear increase of enzyme activity. Protein synthesis is required for “induction”. The isozyme of HeLa alkaline phosphatase induced by monocarboxylates is the carcinoplacental form of the enzyme as determined by stereospecific inhibition by the l-enantiomorphs of phenylalanine and tryptophan, heat stability, and immunoreactivity with antibody against the human placental enzyme.Monocarboxylates that mediate increased alkaline phosphatase activity inhibit HeLa cell multiplication. Inhibition of HeLa cell growth may be necessary for induction and this hypothesis is supported by the findings that three different inhibitors of DNA synthesis, i.e. hydroxyurea, 1-β-d-arabinfuranosyl cytosine and methotrexate, also increase alkaline phosphatase activity. These inhibitors are synergistic with butyrate in causing HeLa cells to assume a more spindle-like shape and in producing an up-to 25-fold increase of enzyme activity. Studies on the modulation of carcinoplacental alkaline phosphatase by monocarboxylates commonly used as antimicrobial food additives and by anti-neoplastic agents may provide methods to evoke “tumor markers” of human occult malignancies. These drug-induced elevations of fetal isozyme activity may further our understanding of gene expression in human cells.     

Hormonal control of liver regeneration

Two peaks in cyclic AMP production in rat livers 4 and 12h after partial hepatectomy (MacManus et al., 1972) were confirmed and a third peak established at 22h, which is the peak of DNA synthesis. The increases in cyclic AMP were prevented by beta-adrenergic blocking agents, propranolol and pindolol, without affecting ornithine decarboxylase induction or DNA synthesis. The alpha-blocking agents, phenoxybenzamine and phentolamine, given at the time of partial hepatectomy, delayed the rise in ornithine decarboxylase normally found 4h after operation, but did not affect DNA synthesis. If the alpha-blocking agents were given at 9-12h or 18h, the onset of DNA synthesis was delayed. Phenoxybenzamine did not affect the induction of ornithine decarboxylase in intact rat livers by glucagon or growth hormone, but did inhibit induction by dexamethasone. The induction of ornithine decarboxylase produced by dexamethasone was inhibited by 17alpha-hydroxy-progesterone; this compound also blocked the induction of ornithine decarboxylase in livers of partially hepatectomized rats.     

Glycoprotein biosynthesis in B lymphocytes: induction of protein N-glycosylation, RNA synthesis, and DNA synthesis by phorbol ester plus ionomycin is blocked by protein kinase inhibitors

The combination of phorbol 12-myristate 13-acetate (PMA) and ionomycin produces a dramatic increase in the incorporation of [2-3H]mannose into Glc3Man9GlcNAc2-P-P-dolichol and glycoprotein, and the induction of RNA and DNA synthesis in murine splenic B lymphocytes (B cells). The kinetics of the induction processes and the concentrations of PMA and ionomycin required for the optimal response have been defined. While the levels of induction of RNA and DNA synthesis by PMA + ionomycin were similar to the mitogenic response to bacterial lipopolysaccharide, activation by PMA and the calcium ionophore resulted in a threefold higher stimulation in dolichol-linked oligosaccharide biosynthesis and protein N-glycosylation. These results indicate that all signalling mechanisms that trigger RNA and DNA synthesis may not be sufficient to produce maximal induction of the N-glycosylation apparatus. 1-(5-Isoquinolinesulfonyl)-2-methylpiperazine (H-7), a potent protein kinase C inhibitor, prevented the induction of protein N-glycosylation activity (IC50 = 11 microM), as well as RNA (IC50 = 18 microM) and DNA synthesis (IC50 = 12 microM), two common indices of B cell activation. N-[2-(Methylamino)ethyl]-5-isoquinolinesulfonamide (H-8) also inhibited the induction of oligosaccharide-lipid intermediate, glycoprotein, RNA, and DNA synthesis, but required higher concentrations than H-7 for 50% inhibition. N-(2-Guanidinoethyl)-5-isoquinolinesulfonamide (HA1004), a potent inhibitor of cyclic nucleotide-dependent protein kinases, had little effect on the activation of the B cell metabolic processes. The H-7-sensitive reactions involved in the induction of RNA and DNA synthesis occurred within 4 h, but induction of lipid intermediate and glycoprotein biosynthesis remained sensitive to H-7 for 10 h after exposure to PMA and ionomycin. Direct in vitro assays in the presence of 0.6% Brij 58 reveal that a cytosolic, phospholipid-dependent protein kinase activity is translocated to a membrane site(s) after treatment with PMA and ionomycin, and the translocated protein kinase is sensitive to H-7. The relative order of potency of the protein kinase inhibitors on the metabolic processes strongly supports the hypothesis that protein kinase C, acting synergistically with Ca2+ mobilization, plays a key regulatory role in the early stages of B cell activation. The synthesis of oligosaccharide-lipid intermediates and protein N-glycosylation are also shown to be induced in B cells activated by PMA + ionomycin.     

Inhibitory effect of taxol, a microtubule stabilizing agent, on induction of DNA synthesis is dependent upon cell lines and growth factors.

Growth-arrested rat fibroblasts, 3Y1, and human diploid fibroblasts, TIG-1, were induced to synthesize DNA by stimulation with various agents such as fetal bovine serum (FBS), epidermal growth factor (EGF), colcemid, or colchicine. Taxol, a microtubule-stabilizing agent, blocked the induction of DNA synthesis after stimulation with colcemid or colchicine in both cell lines. Taxol inhibited the induction of DNA synthesis after stimulation with FBS or EGF in TIG-1, but did not in 3Y1. 12-O-tetradecanoylphorbol-13-acetate (TPA) induced DNA synthesis in TIG-1, which was reduced only partly by taxol. Taxol stabilized or polymerized microtubules in both cell lines. These results indicate that the inhibitory effect of taxol on the induction of DNA synthesis varied among cell lines and among growth factors, and suggest that signal transduction processes may be differentiated by taxol sensitivity. In TIG-1 cells, when taxol was added within 6 h, about halfway into the initiation of DNA synthesis after the addition of FBS or EGF, the inhibition of DNA synthesis still occurred. Taxol did not inhibit the induction of c-fos and c-myc genes by FBS or EGF stimulation. Colchicine itself did not induce these genes in TIG-1. Thus, taxol appeared to inhibit the induction of DNA synthesis not by blockage in the early transduction process of the growth signal from the cell surface to nuclei but by blockage in processes operating in the mid- or late-prereplicative phase.     

DNA signals for G2 checkpoint response in diploid human fibroblasts

DNA (deoxyribonucleic acid) signals that induce the G2 checkpoint response were examined using proliferative secondary cultures of diploid human fibroblasts. Treatments that generated DNA double-strand breaks (DSBs) directly were effective inducers of checkpoint response, generally producing >80% inhibition of mitosis (G2 delay) and the kinase activity of M-phase-promoting factor within 2 h of treatment. Effective inducers of G2 checkpoint response included γ-irradiation and the cancer chemotherapeutic drugs, bleomycin and etoposide. Treatments that produced DNA single-strand breaks, directly or indirectly through nucleotide excision repair, were not effective inducers of G2 delay. Ineffective treatments included incubation with camptothecin, an inhibitor of topoisomerase I (topo I), and irradiation with sublethal fluences of UVC, followed by incubation with aphidicolin. Transient severe inhibition of DNA synthesis with aphidicolin did not affect mitosis substantially, suggesting that the replication arrest input to the G2 checkpoint required more than brief inhibition of DNA synthesis. In contrast, moderate camptothecin-induced inhibition of DNA synthesis was associated with a strong inhibition of mitosis that developed 4–12 h after drug treatment. This result suggested that G2 delay was not expressed until the cells that were in S-phase at the time of treatment with camptothecin proceeded into G2. DNA damage was not necessary for induction of mitotic delay. An inhibitor of topoisomerase II (topo II), ICRF-193, which inhibits chromatid decatenation in G2 cells without damaging DNA, induced a severe inhibition of mitosis and M-phase-promoting factor kinase activity. The results suggest that DNA double-strand breaks and insufficiency of chromatid decatenation effectively induce the G2 checkpoint response, but DNA single-strand breaks do not.     

Interaction of human DNA polymerase beta with ions of copper, lead, and cadmium

Under appropriate conditions, divalent copper, lead, and cadmium ions significantly inhibit human DNA polymerase β (following accepted convention, the term DNA polymerase β refers to the low-molecular-weight, 3–4 S DNA polymerase of eukaryotic cells) at concentrations below 10^?5m. Each metal showed apparent linear noncompetitive inhibition kinetics with respect to the template primer and the deoxynucleoside triphosphate substrate, indicating that complex formation with these components does not account for the inhibition. Apparently, neither lead nor cadmium inhibit by displacing required zinc atoms from the polymerase. The interaction of the metals with the enzyme can be reversed or prevented by EDTA or by thiol compounds, except that inhibition by cadmium ions can be reversed by monothiols but not by dithiols. The metals probably do not inhibit through reaction with thiol groups since the inhibition is not decreased by pretreating the enzyme with N-ethylmaleimide. Although divalent zinc is moderately inhibitory in manganese activated poly(dT) synthesis on a poly(dA) template, it can fill the requirement for a divalent metal ion and, under the conditions tested, is about 60% as effective as Mn²⁺.     

Induction of alkaline phosphatase and the glycoprotein hormone α subunit in HeLa cells by inhibitors of DNA polymerase

Levels of the glycoprotein hormone α subunit and alkaline phosphatase activity were increased in cultures of HeLa S₃ cells exposed to aphidicolin (0.2–10 μg/ml) or phosphonoformic acid (0.1–3 mm), inhibitors of DNA polymerase α. Induction was dependent on both the concentration and duration of exposure to the inhibitors and was prevented by cycloheximide and actinomycin D. Limited characterization of the induced α subunit and alkaline phosphatase activity suggest that they are similar to the uninduced proteins expressed by this cell line. Induction of both proteins by aphidicolin and phosphonoformic acid was enhanced by the simultaneous addition of 3 mm sodium butyrate but was depressed by 1 mm hydroxy urea. In contrast, both butyrate and hydroxy urea cause induction of these proteins when added alone to HeLa cultures. It is unlikely that a direct relationship exists between protein induction and the inhibition of DNA synthesis produced by aphidicolin and phosphonoformic acid since the concentrations required to produce half-maximal induction are 5 to 10 times greater than those needed to inhibit replication by 50%.     

Studies on the mechanism of hrmonal stimulation of zinc uptake in human cell cultures: hormone-cell interactions and characteristics of zinc accumulation

Adrenal steroid hormones with glucocorticoid activity increase the uptake of Zn⁺⁺ in HeLa cell cultures. On the basis of the level of Zn⁺⁺ accumulation induced, steroid hormones can be classified into four groups: (a) optimal inducers (e.g., hydrocortisone and prednisolone); (b) suboptimal inducers (e.g., aldosterone and corticosterone); (c) anti-inducers (e.g., progesterone and 17 α-methyl testosterone) which competitively inhibit induction by optimal inducers; and (d) non-inducers (e.g., cortisone and pregnenolone) which neither induce nor inhibit the steroid-mediated increase in Zn⁺⁺ uptake. The ability of an anti-inducer to block the effects of optimal inducers is not the result of inhibition of steroid uptake or an effect on general protein synthesis. Optimal inducers do not increase adenyl cyclase activity of HeLa cells nor can the hormone effects on Zn⁺⁺ uptake be reproduced by 3'-5' cyclic AMP. The prednisolone-induced enhancement of Zn⁺⁺ uptake is gradually lost over two or three days following removal of the hormone. Uptake of Zn⁺⁺ by HeLa cells is not altered by a decrease of sodium concentration in the medium nor by changes in medium osmolarity. The uptake mechanism is not affected by subjecting intact cells to proteolytic enzymes; however, if cells are disrupted the hormone-mediated increase in Zn⁺⁺ accumulation is lost. The Zn⁺⁺ taken up by HeLa cells in the presence or absence of hormone is primarily cytoplasmic in localization and appears to be distributed in a multicompartmental system.     

Protection by superoxide dismutase, catalase, and poly(ADP-ribose) synthetase inhibitors against alloxan- and streptozotocin-induced islet DNA strand breaks and against the inhibition of proinsulin synthesis

We have shown previously that alloxan and streptozotocin, two major diabetogenic agents, cause DNA strand breaks in rat pancreatic islets and stimulate nuclear poly(ADP-ribose) synthetase, thereby depleting intracellular NAD level and inhibiting proinsulin synthesis (Okamoto, H. (1981) Mol. Cell. Biochem. 37, 43-61; Yamamoto, H., Uchigata, Y., and Okamoto, H. (1981) Nature 294, 284-286). In the present study, superoxide dismutase and catalase, scavengers of radical oxygens, were found to protect against islet DNA strand breaks and inhibition of proinsulin synthesis induced by alloxan. The radical scavengers did not affect islet DNA strand breaks or inhibition of proinsulin synthesis induced by streptozotocin. On the other hand, compounds that inhibit islet nuclear poly(ADP-ribose) synthetase were found to protect against alloxan- as well as streptozotocin-induced inhibition of proinsulin synthesis. The poly(ADP-ribose) synthetase inhibitors were ineffective in protection against DNA strand breaks induced by the agents. These results may provide an important clue for elucidating the prevention of insulin-dependent diabetes as well as for understanding the cause of diabetes.     

Alkylating Agents Induce Uvm, a Reca-Independent Inducible Mutagenic Phenomenon in Escherichia Coli

Noninstructive DNA damage in Escherichia coli induces SOS functions hypothesized to be required for mutagenesis and translesion DNA synthesis at noncoding DNA lesions. We have recently demonstrated that in E. coli cells incapable of SOS induction, prior UV-irradiation nevertheless strongly enhances mutagenesis at a noninstructive lesion borne on M13 DNA. Here, we address the question whether this effect, named UVM for UV modulation of mutagenesis, can be induced by other DNA damaging agents. Exponentially growing δrecA cells were pretreated with alkylating agents before transfection with M13 single-stranded DNA bearing a site-specific ethenocytosine residue. Effect of cell pretreatment on survival of the transfected DNA was determined as transfection efficiency. Mutagenesis at the ethenocytosine site in pretreated or untreated cells was analyzed by multiplex DNA sequencing, a phenotype-independent technology. Our data show that 1-methyl-3-nitro-1-nitrosoguanidine, N-nitroso-N-methylurea and dimethylsulfate, but not methyl iodide, are potent inducers of UVM. Because alkylating agents induce the adaptive response to defend against DNA alkylation, we asked if the genes constituting the adaptive response are required for UVM. Our data show that MNNG induction of UVM is independent of ada, alkA and alkB genes and define UVM as an inducible mutagenic phenomenon distinct from the E. coli adaptive and SOS responses.     

Ametantrone inhibits prostaglandin--mediated resorption in bone organ culture

Prostaglandins (PG) have been postulated to be involved in both tumor metastases to bone and in tumor-induced bone resorption. The anthracenedione antineoplastic agents ametantrone (HAQ) and mitoxantrone are potent antioxidants and inhibit hydroperoxide-dependent initiation and propagation reactions. Therefore, these compounds may inhibit PG production and could also inhibit tumor metastases and tumor-induced resorption. The ability of HAQ, a prototypic anthracenedione, to inhibit PG synthesis and PG-mediated bone resorption was investigated using neonatal mouse calvaria in organ culture. Epidermal growth factor (EGF) stimulates bone resorption in this tissue by inducing PG synthesis. Consequently, if HAQ inhibits EGF-stimulated PG synthesis, it should also inhibit EGF-stimulated bone resorption. HAQ, at 10 microM, completely abolished EGF-stimulated PG synthesis and calcium release. Moreover, HAQ (1.0-30 microM) inhibition of EGF-stimulated PGE2 synthesis correlated with the inhibition of EGF-stimulated Ca release in a concentration-dependent manner. In contrast to EGF, parathyroid hormone stimulates resorption by a PG-independent pathway. HAQ at 10 microM had no effect on parathyroid hormone stimulated Ca release. These results suggest that HAQ inhibition of bone resorption appears to be primarily mediated by inhibition of PG biosynthesis.