Nordihydroguaiaretic Acid Elevates Osteoblastic Intracellular Ca2+

C Pharmacology & Toxicology 2001, 89, 301–305. Printed in Denmark . All rights reserved

Copyright C ISSN 0901-9928

Nordihydroguaiaretic Acid Elevates Osteoblastic Intracellular Ca2π Jue-Long Wang1,10, Hsin-Ju Chang2,10, Li-Ling Tseng2,10, Chun-Peng Liu3,10, Kam-Chung Lee3,10, Kang-Ju Chou3,10, Jin-Shiung Cheng3,10, Yuk-Keung Lo3,10, Warren Su4, Yee-Ping Law5, Wei-Chung Chen6, Rai-Chi Chan7,10 and Chung-Ren Jan8,9 1

Department of Rehabilitation, Kaohsiung Veterans General Hospital; 2Department of Dentistry, Kaohsiung Veterans General Hospital; 3Department of Medicine, Kaohsiung Veterans General Hospital; 4Department of Pediatrics, PaoChien General Hospital, Ping Tung; 5Department of Medicine, Pao-Chien General Hospital, Ping Tung; 6Department of Surgery, Ping Tung Christian Hospital; 7Department of Rehabilitation, Taipei Veterans General Hospital; 8Department of Medical Education and Research, Kaohsiung Veterans General Hospital; 9Department of Biology and Institute of Life Sciences, National Sun Yat-sen University, Kaohsiung; and 10Department of Medicine, National Yang Ming University, Taipei, Taiwan (Received April 24, 2001; Accepted August 14, 2001) Abstract: Nordihydroguaiaretic acid (NDGA) is widely used as a pharmacological tool to inhibit lipoxygenases; however, recent evidence suggests that it increases renal intracellular [Ca2π]i via novel mechanisms. Here the effect of NDGA on Ca2π signaling in MG63 osteoblastic cells was explored using fura-2 as a Ca2π indicator. NDGA (2–50 mM) increased [Ca2π]i in a concentration-dependent manner. The signal comprised an initial rise and an elevated phase over a time period of 4 min. Removing extracellular Ca2π reduced 2–50 mM NDGA-induced signals by 62∫2%. After incubation with 50 mM NDGA in Ca2π-free medium for several minutes, addition of 3 mM CaCl2 induced an increase in [Ca2π]i. NDGA (50 mM)-induced [Ca2π]i increases were not changed by pretreatment with 10 mM of verapamil, diltiazem, nifedipine, nimodipine and nicardipine. In Ca2π-free medium, pretreatment with the endoplasmic reticulum Ca2π pump inhibitor thapsigargin (1 mM) inhibited 50 mM NDGA-induced [Ca2π]i increases by 69∫3%. Inhibition of phospholipase C with 2 mM U73122 had little effect on 50 mM NDGA-induced Ca2π release. Several other lipoxygenase inhibitors had no effect on basal [Ca2π]i. At a concentration that did not increase basal [Ca2π]i, NDGA (1 mM) did not alter 10 mM ATP- or 1 mM thapsigargin-induced [Ca2π]i increases. Alteration of protein kinase C activity with 1 nM phorbol 12-myristate 13acetate or 2 mM GF 109203X did not affect 50 mM NDGA-induced [Ca2π]i increases. Together, the results show that NDGA increased [Ca2π]i in osteoblasts in a lipoxygenase-independent manner, by releasing stored Ca2π in a fashion independent of phospholipase C activity, and by causing Ca2π influx.

Nordihydroguaiaretic acid (NDGA) is a drug widely used for inhibiting lipoxygenases and, thereby, for investigating the role of arachidonic acid metabolites in diverse cell responses (Force et al. 1991; Rizzo et al. 1999; Nugent et al. 2001). However, NDGA exerts many other actions that seem to be dissociated from lipoxygenase inhibition. These include disruption of mitochondrial membrane potential in cells containing no lipoxygenases (Biswal et al. 2000), modulation of low-affinity Ca2π binding sites of muscle and platelet Ca2π-ATPase (Barata et al. 1999), activation of Ca2π-dependent Kπ current (Yamamura et al. 1999), inhibition of a swelling-activated, ATP-sensitive taurine channel (Ballatori & Wang 1997), blockade of protein transport in the secretory pathway (Ramoner et al. 1998), alteration of Kπ and Ca2π currents (Hatton & Peers 1997), lowering blood glucose level (Luo et al. 1998), and induction of sister-chromatid exchanges (Madrigal-Bujaidar et al. 1998). Recent evidence shows that NDGA increases basal intracellular free Ca2π levels ([Ca2π]i) in renal tubular cells (Jan Author for correspondence: Chung-Ren Jan, Department of Medical Education and Research, Kaohsiung Veterans General Hospital, 386 Ta Chung 1st Rd, Taiwan 813 (fax π886 7 3468056, e-mail crjan/isca.vghks.gov.tw).

et al. 2000). The effect of NDGA on [Ca2π]i in other cell types is unclear. An increase in [Ca2π]i is a messenger for numerous cell processes (Clapham 1995; Berridge et al. 1999). [Ca2π]i may increase upon stimulation as a result of Ca2π release from stores, and/or Ca2π entry from extracellular space. In many cell types, the major Ca2π stores are the inositol 1,4,5-trisphosphate-sensitive endoplasmic reticulum Ca2π pool and the mitochondrial store (Berridge et al. 1999). The release of stored Ca2π often leads to Ca2π influx via a store-operated Ca2π influx mechanism (Putney 1986). The aim of the present study was to explore the effect of NDGA on Ca2π handling in osteoblasts. NDGA has been used in investigating osteoblasts (Suzki et al. 1996; Meghji et al. 1988; Miwa et al. 2000). The results were interpreted, while assuming that NDGA acted as a selective inhibitor of arachidonic acid metabolism. Whether NDGA can alter Ca2π signaling in osteoblasts is unclear. MG63 human osteoblast-like cells were used in this study. This cell line has been used as a model for investigating osteoblasts (Lohmann et al. 2000). We have found in this study that NDGA induces an increase in [Ca2π]i in MG63 cells, by using fura-2 as a fluorescent Ca2π probe. The concentration-response relationships both in the presence and absence of extracellular Ca2π

302

JUE-LONG WANG ET AL.

have been established, and the underlying mechanisms of the NDGA response have been explored. Materials and Methods Cell culture. MG63 cells obtained from American Type Culture Collection were cultured in modified Eagle medium supplemented with 10% heat-inactivated foetal bovine serum, 100 U/ml penicillin and 100 mg/ml streptomycin at 37æ in 5% CO2-containing humidified air. Solutions. Ca2π-containing medium contained 140 mM NaCl, 5 mM KCl, 1 mM MgCl2, 2 mM CaCl2, 10 mM Hepes, and 5 mM glucose. The pH value was adjusted to 7.4 with 1 N NaOH. In Ca2π-free medium, Ca2π was substituted with 1 mM EGTA. NDGA was dissolved in ethanol as a 0.1 M stock solution. The other drugs were dissolved in water, ethanol or dimethyl sulfoxide. The concentration of solvents in the solution used in experiments did not exceed 0.1% which had no effect on [Ca2π]i (nΩ3). Optical. Measurements of [Ca2π]i. Trypsinized cells (106/ml) were loaded with 2 mM of the acetoxymethyl ester form of fura-2, fura2/AM, for 30 min. at 25æ in culture medium. Fura-2 fluorescence measurements were performed in a water-jacketed cuvette (25æ) with continuous stirring; the cuvette contained 1 ml of medium and 0.5 million cells. Fluorescence was monitored with a Shimadzu RF5301PC spectrofluorophotometer by recording excitation signals at 340 nm and 380 nm and emission signal at 510 nm at 1 sec. intervals. Maximum and minimum fluorescence values were obtained by adding 0.1% Triton X-100 (plus 10 mM CaCl2) and 20 mM EGTA sequentially at the end of each experiment. [Ca2π]i was calculated as previously described (Grynkiewicz et al. 1985). Chemicals. The reagents for cell culture were from Gibco. Fura2/AM was from Molecular Probes. Nordihydroguaiaretic acid, U73122 (1-(6-((17b-3-methoxyestra-1,3,5(10)-trien-17-yl)amino) hexyl)-1H-pyrrole-2,5-dione), U73343 (1-(6-((17b-3-methoxyestra1,3,5(10)-trien-17-yl)amino)hexyl)-2,5-pyrrolidine-dione) were from Biomol. The other reagents were from Sigma Chemical Co., St. Louis, MO, USA. Statistics. The [Ca2π]i recordings and data were mean∫S.E.M. of four to six replicates. Because the data from each experiment were the average responses from 0.5 million cells in the cuvet, the variation among experiments was small. Two means were compared using Student’s t-test, and significance was accepted when P⬍0.05.

Results Effect of NDGA on [Ca2π]i.

Fig. 1. Effect of nordihydroguaiaretic acid (NDGA) on [Ca2π]i in MG63 osteoblasts. A, concentration-dependent effects of NDGA on [Ca2π]i. The concentration of NDGA was 50 mM in trace a, 20 mM in trace b, 10 mM in trace c, and 1 mM in trace d. Experiments were performed in Ca2π-containing medium. NDGA was added at 30 sec. B, effect of extracellular Ca2π removal on NDGA-induced [Ca2π]i increases and effect of reintroduction of Ca2π. The concentration of NDGA was 50 mM in trace a, and 0 mM in trace b. NDGA was added at 30 sec. CaCl2 (3 mM) was added at 490 sec. C, concentration-response plots of NDGA-induced Ca2π signals both in Ca2π-containing medium (solid circles) and Ca2π-free medium (open circles). The y axis is the percentage of control. Control was defined as the net (baseline subtracted) maximum [Ca2π]i of 50 mM NDGA-induced [Ca2π]i signal in Ca2π-containing medium. * P⬍0.05. Data were the mean ∫ S.E.M. of 4–6 experiments.

NDGA at concentrations between 2–50 mM increased [Ca2π]i in Ca2π-containing medium (fig. 1A). Over a time period of 4 min. the [Ca2π]i signal comprised an initial rise and a sustained phase. At a concentration of 50 mM (trace a), NDGA induced a [Ca2π]i increase which reached a net (baseline subtracted) maximum value of 181∫4 nM (nΩ6). The NDGA-induced response saturated at 50 mM because 100 mM NDGA did not induce a greater response. At a concentration of 1 mM, NDGA had little effect (trace d). Effect of removing extracellular Ca2π on NDGA-induced [Ca2π]i increases. Fig. 1B (time points between 0–500 sec.) shows that 50 mM NDGA induced a [Ca2π]i increase with a net maximum value of 65∫3 nM (trace a; nΩ5). The maximum Ca2π re-

NORDIHYDROGUAIARETIC ACID INCREASES OSTEOBLASTIC CA2π LEVELS

303

sponse was followed by a gradual decay leading to a sustained phase that had a net value of 12∫2 nM above baseline. The concentration-response plots of NDGA-induced [Ca2π]i increases in Ca2π-containing medium (filled circles) and Ca2π-free medium (open circles) of Ca2π were shown in fig. 1C. The data suggest that NDGA acted with an EC50 value of about 10 mM. Removing extracellular Ca2π reduced 2–100 mM NDGA-induced [Ca2π]i increases by 62∫2% in the net maximum value (nΩ5–6; P⬍0.05). Mechanisms of NDGA-induced Ca2π influx. Mobilization of stored Ca2π may trigger Ca2π influx via a store-operated Ca2π influx mechanism (Putney 1986). Store-operated Ca2π influx was usually explored by researchers by adding Ca2π to cells that had been depleted of stored Ca2π by the tested agent in Ca2π-free medium. fig. 1B (time points between 500–600 sec.) shows that in Ca2πfree medium, after pretreatment with 50 mM NDGA for about 8 min., addition of 3 mM CaCl2 induced an immediate [Ca2π]i increase with a net maximum of 182∫4 nM (trace a; nΩ6). Addition of 3 mM CaCl2 alone without NDGA pretreatment only induced a small [Ca2π]i increase with a net maximum of 31∫3 nM (trace b; nΩ5). The effect of Ca2π entry blockers on NDGA-induced [Ca2π]i increases was investigated. Pretreatment with 10 mM of verapamil, diltiazem, nifedipine, nimodipine and nicardipine did not alter 50 mM NDGA-induced [Ca2π]i increases in Ca2π-containing medium (not shown; nΩ5). Intracellular stores of NDGA-induced [Ca2π]i increases. Fig. 2A shows that in Ca2π-free medium, pretreatment with 50 mM NDGA abolished the [Ca2π]i increases induced by 1 mM thapsigargin, an endoplasmic reticulum Ca2π pump inhibitor (Thastrup et al. 1990), and 2 mM carbonylcyanide mchlorophenylhydrazone, a mitochondrial uncoupler (nΩ5). Conversely, fig. 2B shows that addition of 1 mM thapsigargin induced a [Ca2π]i increase with a net maximum value of 81∫3 nM (nΩ6). After thapsigargin pretreatment for 6 min., addition of 10 mM ATP did not release more Ca2π (nΩ5; not shown), suggesting complete depletion of the thapsigarginsensitive endoplasmic reticulum pools. However, subsequently added 50 mM NDGA induced a [Ca2π]i increase with a net maximum of 41∫3 nM, which was 57∫3% of the control NDGA response shown in fig. 2A (nΩ6; P⬍0.05). Furthermore, the experiments in fig. 2C were performed to see if the mitochondrial Ca2π stores play a role in the NDGA-induced Ca2π release. The data show that in Ca2π-free medium, addition of 2 mM carbonylcyanide m-chlorophenylhydrazone induced a [Ca2π]i increase with a net maximum value of 19∫3 nM (nΩ5). Subsequently added 1 mM thapsigargin and 50 mM NDGA induced [Ca2π]i increases that were similar to what were shown in fig. 2B. The role of phospholipase C-coupled inositol 1,4,5-trisphosphate formation in NDGA-induced Ca2π release. Fig. 3A shows that in Ca2π-free medium, addition of 10 mM ATP, an inositol 1,4,5-trisphosphate-dependent agonist,

Fig. 2. Intracellular Ca2π stores of NDGA-induced [Ca2π]i increases. Experiments were performed in Ca2π-free medium. Drugs were added at the time indicated by arrows. The concentration of drugs was: thapsigargin, 1 mM; NDGA, 50 mM, and carbonylcyanide m-chlorophenylhydrazone (CCCP), 2 mM. Data were the mean ∫ S.E.M. of 4–6 experiments.

induced a [Ca2π]i increase with a net maximum value of 98∫3 nM (nΩ5). This ATP response was abolished by pretreatment with 2 mM U73122 (Thompson et al. 1991) for 200 sec. to inhibit phospholipase C activity (fig. 3B). In contrast, U73343 (10 mM), an inactive U73122 analogue, did

304

JUE-LONG WANG ET AL.

Fig. 3. Effect of inhibition of phospholipase C activity on NDGA-induced intracellular Ca2π release. Experiments were performed in Ca2πfree medium. A, 10 mM ATP was added at 30 sec. B, 2 mM U73122 was added at 30 sec. followed by 10 mM ATP and 50 mM NDGA added at 220 sec. and 300 sec., respectively. Data were the mean ∫ S.E.M. of 4–6 experiments.

not inhibit the ATP response (nΩ5; not shown). This suggests that U73122 effectively inhibited phospholipase C activity. Under these conditions, addition of 50 NDGA at the time point of 300 sec. induced a [Ca2π]i increase that was similar to control in magnitude and shape (fig. 2A) (nΩ5).

Effects of several other lipoxygenase inhibitors on [Ca2π]i. The following lipoxygenase inhibitors did not alter basal [Ca2π]i: baicalein (50 mM), 5.8.11.14-eicosatetraynoic acid (ETYA; 0.1 mM), caffeic acid (50 mM), esculetin (50 mM) and L-655238 (100 mM) (nΩ4; not shown).

Effect of a low concentration of NDGA on ATP- and thapsigargin-induced [Ca2π]i increases. Experiments were performed to explore whether NDGA (at a concentration that may inhibit lipoxygenases but did not increase basal [Ca2π]i) could alter the Ca2π signals induced by other agonists. Pretreatment with 1 mM NDGA for 10 min. in Ca2π-containing medium did not alter basal [Ca2π]i or the [Ca2π]i increases induced by 10 mM ATP or 1 mM thapsigargin (nΩ4; not shown).

Effect of alteration of protein kinase C activity on NDGAinduced [Ca2π]i increases. Protein kinase C activity has been shown to regulate the [Ca2π]iincreases induced by some agonists (Jan et al. 1998; Herlitze et al. 2001). Thus, the relationship between NDGA-induced [Ca2π]i increase and protein kinase C activity was examined. The data show that the [Ca2π]i increases induced by 50 mM NDGA in Ca2π-containing medium was not altered by pretreatment with 1 nM phorbol 12-myristate 13-acetate to activate the kinase or with 2 mM GF 109203X to inhibit the kinase (nΩ4; not shown).

Discussion This study is the first to show that NDGA, commonly used as a lipoxygenase inhibitor, caused an increase in [Ca2π]i in osteoblasts. Our data suggest that NDGA increased [Ca2π]i via a mechanism dissociated from its inhibitory effect on arachidonic acid metabolism because several other lipoxygenase inhibitors did not have similar effects. Because an increase in [Ca2π]i may alter diverse cell functions (Berridge et al. 1999), caution must be applied in using NDGA to inhibit arachidonic acid metabolism. In addition, the effect of NDGA on [Ca2π]i may be linked to many in vitro actions of this drug that are not via inhibition of lipoxygenases. Our findings suggest that Ca2π influx and Ca2π release both contributed to NDGA-induced [Ca2π]i increases because the responses were reduced by 60% by removing extracellular Ca2π. NDGA-induced Ca2π entry was insensitive to L-type Ca2π channel blockers and may be mediated by capacitative Ca2π entry. However, NDGA may also induce Ca2π influx via channels coupled to receptors or second messengers in a manner unlinked to the depletion of Ca2π stores. NDGA appears to release Ca2π mainly from the thapsigargin-sensitive endoplasmic reticulum store and the mitochondrial stores because pretreatment with NDGA abolished thapsigargin- and carbonylcyanide m-chlorophenylhydrazone-induced Ca2π release. NDGA may also release Ca2π from other stores because pretreatment with carbonylcyanide m-chlorophenylhydrazone and thapsigargin did not completely deplete NDGA-releasable stores. Similarly, in renal tubular cells, NDGA was shown to release Ca2π from multiple stores including the endoplasmic reticulum and mitochondria (Jan & Tseng 2000). NDGA-induced release of stored Ca2π appears to be independent of the activity of phospholipase C. Consistently, a previous study performed in glomerular mesangial cells has shown that pretreatment with 10 mM NDGA for 10

NORDIHYDROGUAIARETIC ACID INCREASES OSTEOBLASTIC CA2π LEVELS

min. did not stimulate the production of inositol 1,4,5-trisphosphate (Force et al. 1991). It was recently shown that NDGA inhibits the different sarco/endoplasmic reticulum Ca2π-ATPase isoforms found in skeletal muscle and blood platelets (Barata et al. 1999). Thus, NDGA may release stored Ca2π by blocking Ca2π pump in a manner similar to thapsigargin-induced Ca2π release. Protein kinase C activity has been shown to regulate the [Ca2π]i increases induced by some agonists (Jan et al. 1998; Herlitze et al. 2001). However, we found that the NDGAinduced [Ca2π]i increases were not changed by activation or inhibition of protein kinase C. Furthermore, the data show that NDTA appears to affect Ca2π signaling by increasing basal [Ca2π]i only at a higher concentration (⬎ 2 mM), because 1 mM NDGA did not affect ATP- and thapsigargininduced [Ca2π]i increases. Collectively, the present study shows that NDGA (2–50 mM) increased [Ca2π]i in osteoblasts via a mechanism dissociated from its inhibition of lipoxygenases. Because NDGA at concentrations of 5–150 mM is commonly used in in vitro studies (Ballatori & Wang 1997; Hatton & Peers 1997; Madrigal-Bujaidar et al. 1998), the possible effect of NDGA on Ca2π signaling should be considered in interpreting the results. Acknowledgements This work was supported by grants from National Science Council (NSC89–2320-B-075B-015), VTY Joint Research Program, Tsou’s Foundation (VTY89-P3–21) and VGHKS90–07 to C.R.J.; VGHKS-90–62 to J.L.W. References Ballatori N. & W. Wang: Nordihydroguaiaretic acid depletes ATP and inhibits a swelling-activated, ATP-sensitive taurine channel. Amer. J. Physiol. 1997, 272, C1429–C1436. Barata, H., C. M., Cardoso, H. Wolosker & L. de Meis: Modulation of the low affinity Ca2π-binding sites of skeletal muscle and blood platelets Ca2π-ATPase by nordihydroguaiaretic acid. Mol. Cell. Biochem. 1999, 195, 227–233. Berridge, M. J., P. Lipp & M. Bootman: Calcium signalling. Cur. Biol. 1999, 9, R157–R159. Biswal, S. S., K. Datta, S. D. Shaw, X. Feng, J. D. Robertson, J. P. Kehrer: Glutathione oxidation and mitochondrial depolarization as mechanisms of nordihydroguaiaretic acid-induced apoptosis in lipoxygenase-deficient FL5.12 cells. Toxicol. Sci. 2000, 53, 77–83. Clapham, D. E.: Calcium signaling. Cell 1995, 80, 259–268. Force, T., G. Hyman, R. Hajjar, A. Sellmayer & J. V. Bonventre: Noncyclooxygenase metabolites of arachidonic acid amplify the vasopressin-induced Ca2π signal in glomerular mesangial cells by releasing Ca2π from intracellular stores. J. Biol. Chem. 1991, 266, 4295–4302. Grynkiewicz, G., M. Poenie & R. Y. Tsien: A new generation of Ca2π indicators with greatly improved fluorescence properties. J. Biol. Chem. 1985, 260, 3440–3450. Hatton, C. J. & C. Peers: Multiple effects of nordihydroguaiaretic acid on ionic currents in rat isolated type I carotid body cells. Brit. J. Pharmacol. 1997, 122, 923–929.

305

Herlitze, S., H. Zhong, T. Scheuer, W.A. Catterall: Allosteric modulation of Ca2π channels by G proteins, voltage-dependent facilitation, protein kinase C, and Cavbeta subunits. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 4699–4704. Jan, C. R., C. M. Ho, S. N. Wu & C. J. Tseng: Bradykinin-evoked Ca2π mobilization in Madin Darby canine kidney cells. Eur. J. Pharmacol. 1998, 355, 219–233. Jan, C. R. & C. J. Tseng: Mechanisms of nordihydroguaiaretic acidinduced [Ca2π]i increases in MDCK cells. Life Sci. 2000, 66, 1753–1762. Lohmann, C. H., Z. Schwartz, Y. Liu, H. Guerkov, D. D. Dean, B. Simon, B. D. Boyan: Pulsed electromagnetic field stimulation of MG63 osteoblast-like cells affects differentiation and local factor production. J. Orthop. Res. 2000, 18, 637–646. Luo J, T. Chuang, J. Cheung, J. Quan, J. Tsai, C. Sullivan, R. F. Hector, M. J. Reed, K. Meszaros, S. R. King, T. J. Carlson & G. M. Reaven: Masoprocol (nordihydroguaiaretic acid): a new antihyperglycemic agent isolated from the creosote bush (Larrea tridentata). Eur. J. Pharmacol. 1998, 346, 77–79. Madrigal-Bujaidar, E., S. Diaz Barriga, M. Cassani, D. Molina & G. Ponce: In vivo and in vitro induction of sister-chromatid exchanges by nordihydroguaiaretic acid. Mutat. Res. 1998, 412, 139–144. Meghji, S., J. R. Sandy, A.M. Scutt, W. Harvey & M. Harris: Heterogeneity of bone resorbing factors produced by unstimulated murine osteoblasts in vitro and in response to stimulation by parathyroid hormone and mononuclear cell factors. Arch. Oral Biol. 1988, 33, 773–778. Miwa, M., O. O. Kozawa, H. Tokuda & T. Uematsu: Involvement of arachidonic acid in chemical stress-induced interleukin-6 synthesis in osteoblast-like cells: comparison with heat shock protein 27 induction. Prostaglandins Leukot. Essent. Fatty Acids 2000, 62, 189–193. Nugent, C., J. B. Prins, J. P. Whitehead, J. M. Wentworth, V. K. Chatterjee & S. O’Rahilly S: Arachidonic acid stimulates glucose uptake in 3T3-L1 adipocytes by increasing GLUT1 and GLUT4 Levels at the plasma membrane. Evidence for involvement of lipoxygenase metabolites and peroxisome proliferator-activated receptor gamma. J. Biol. Chem. 2001, 276, 9149–9157. Putney, J. W. Jr.: A model for receptor-regulated calcium entry. Cell Calcium 1986, 7, 1–12. Ramoner, R., C. Rieser, G. Bartsch & M. Thurnher: Nordihydroguaiaretic acid blocks secretory and endocytic pathways in human dendritic cells. J. Leukoc. Biol. 1998, 64, 747–752. Rizzo, M. T., E. Regazzi, D. Garau, L. Akard, M. Dugan, H. S. Boswell, V. Rizzoli & C. Carlo-Stella: Induction of apoptosis by arachidonic acid in chronic myeloid leukemia cells. Cancer Res. 1999, 59, 5047–5053. Suzuki, A., O. Kozawa, Y. Oiso & K. Kato: Protein kinase C activation inhibits stress-induced synthesis of heat shock protein 27 in osteoblast-like cells: function of arachidonic acid. J. Cell Biochem. 1996, 62, 69–75. Thastrup, O., H. Linnebjerg, P. J. Bjerrum, J. B. Knudsen & S. B. Christensen: The inflammatory and tumor-promoting sesquiterpene lactone, thapsigargin, activates platelets by selective mobilization of calcium as shown by protein phosphorylations. Biochem. Biophys. Acta 1990, 927, 65–73. Thompson, A.K., S.P. Mostafapour, L.C. Denlinger, J.E. Bleasdale & S.K. Fisher: The aminosteroid U73122 inhibits muscarinic receptor sequestration and phosphoinositide hydrolysis in SK-NSH neuroblastoma cells. J. Biol. Chem. 1991, 266, 23856–23862. Yamamura, H., N. Nagano, M. Hirano, K. Muraki, M. Watanabe, Y. Imaizumi: Activation of Ca2π-dependent Kπ current by nordihydroguaiaretic acid in porcine coronary arterial smooth muscle cells. J. Pharmacol. Exp. Therap. 1999, 291, 140–146.