The carcinogen safrole increases intracellular free Ca2+ levels and causes death in MDCK cells

The effect of the carcinogen safrole on intracellular Ca2+ movement in renal tubular cells has not been explored previously. The present study examined whether safrole could alter Ca2+ handling in Madin-Darby canine kidney (MDCK) cells. Cytosolic free Ca2+ levels ([Ca2+]i) in populations of cells were measured using fura-2 as a fluorescent Ca2+ probe. Safrole at concentrations above 33 microM increased [Ca2+]i in a concentration-dependent manner with an EC50 value of 400 microM. The Ca2+ signal was reduced by 90% by removing extracellular Ca2+, but was not affected by nifedipine, verapamil, or diltiazem. Addition of Ca2+ after safrole had depleted intracellular Ca(2+)-induced dramatic Ca2+ influx, suggesting that safrole caused store-operated Ca2+ entry. In Ca(2+)-free medium, after pretreatment with 650 microM safrole, 1 microM thapsigargin (an endoplasmic reticulum Ca2+ pump inhibitor) failed to release more Ca 2+. Inhibition of phospholipase C with 2 microM U73122 did not affect safrole-induced Ca2+ release. Trypan blue exclusion assays revealed that incubation with 650 microM safrole for 30 min did not kill cells, but killed 70% of cells after incubation for 60 min. Collectively, the data suggest that in MDCK cells, safrole induced a [Ca2+] increase by causing Ca2+ release from the endoplasmic reticulum in a phospholipase C-independent fashion, and by inducing Ca2+ influx via store-operated Ca2+ entry. Furthermore, safrole can cause acute toxicity to MDCK cells.     

Safrole oxide induces neuronal apoptosis through inhibition of integrin beta4/SOD activity and elevation of ROS/NADPH oxidase activity

Neuronal apoptosis is a very important event in the development of the central nervous system (CNS), but the underlying mechanisms remain to be elucidated. We have previously shown that safrole oxide, a small molecule, induces integrin beta4 expression and promotes apoptosis in vascular endothelial cells. In this study, the effects of safrole oxide on cell growth and apoptosis have been examined in primary cultures of mouse neurons. Safrole oxide was found to significantly inhibit neuronal cell growth and to induce apoptosis. The inhibitory and apoptotic activities of safrole oxide followed a dose- and time-dependent manner. Interestingly, the expression of integrin beta4 was significantly inhibited with safrole oxide treatment. Furthermore, safrole oxide dramatically increases the level of intracellular reactive oxygen species (ROS) and the activity of NADPH oxidase. Moreover, manganese-dependent superoxide dismutase (MnSOD) activity was decreased significantly with safrole oxide treatment. Our study thus demonstrates that safrole oxide induces neuronal apoptosis through integrin beta4, ROS, NADPH, and MnSOD.     

Safrole oxide induces apoptosis by activating caspase-3, -8, and -9 in A549 human lung cancer cells

Previously we found that 3,4-(methylenedioxy)-1-(2',3'-epoxypropyl)-benzene (safrole oxide) induced a typical apoptosis in A549 human lung cancer cells. In this study, we further investigated which caspases were activated by safrole oxide during the apoptosis. The data showed that the activity of caspase-3, -8, and -9 was significantly enhanced by the compound, which suggested that safrole oxide might be used as a caspase promoter to initiate lung cancer cell apoptosis.     

Effects of safrole and isosafrole pretreatment on N- and ring-hydroxylation of 2-acetamidofluorene by the rat and hamster

1. The effects of safrole and isosafrole pretreatment on both N- and ring-hydroxylation of 2-acetamidofluorene were studied in male rats and hamsters. 2. Isosafrole (100mg/day per kg body wt.) pretreatment of rats for 3 days did not have any effect on urinary excretion of hydroxy metabolites of 2-acetamidofluorene. However, similar pretreatment with safrole produced increased urinary excretion of N-, 3- and 5-hydroxy derivatives. 3. Similar treatment with these two chemicals for 3 days increased ring-hydroxylation activity by rat liver microsomal material. Increases in N-hydroxylation were much less than those in ring-hydroxylation. Isosafrole was twice as effective as safrole. 4. Increases in hydroxylating activity due to safrole or isosafrole treatment were inhibited by simultaneous administration of ethionine. Similarly, ethionine inhibition was almost completely reversed by the simultaneous administration of methionine. 5. Safrole or isosafrole (0.1mm and 1mm) inhibited 7-hydroxylation activity by liver microsomal material from control rats. At 1mm these two chemicals inhibited both 5- and 7-hydroxylation activity by liver microsomal material from 3-methylcholanthrene-pretreated rats. 3-Hydroxylation activity was not inhibited by 1mm concentrations of these two chemicals. 6. A single injection of safrole (50100 or 200mg/kg body wt.) 24h before assay had no appreciable effect on either N- or ring-hydroxylation activity by hamster liver microsomal material. However, isosafrole (200mg/kg body wt.) treatment inhibited N-, 3- and 5-hydroxylation activities by hamster liver microsomal material; it had no effect on 7-hydroxylation activity.     

Genotoxic effects of eugenol, isoeugenol and safrole in the wing spot test of Drosophila melanogaster

In the present study, the phenolic compounds eugenol, isoeugenol and safrole were investigated for genotoxicity in the wing spot test of Drosophila melanogaster. The Drosophila wing somatic mutation and recombination test (SMART) provides a rapid means to evaluate agents able to induce gene mutations and chromosome aberrations, as well as rearrangements related to mitotic recombination. We applied the SMART in its standard version with normal bioactivation and in its variant with increased cytochrome P450-dependent biotransformation capacity. Eugenol and safrole produced a positive recombinagenic response only in the improved assay, which was related to a high CYP450-dependent activation capacity. This suggests, as previously reported, the involvement of this family of enzymes in the activation of eugenol and safrole rather than in its detoxification. On the contrary, isoeugenol was clearly non-genotoxic at the same millimolar concentrations as used for eugenol in both the crosses. The responsiveness of SMART assays to recombinagenic compounds, as well as the reactive metabolites from eugenol and safrole were considered responsible for the genotoxicity observed.     

Safrole oxide induces apoptosis by up-regulating Fas and FasL instead of integrin beta4 in A549 human lung cancer cells

Previously, we found that 3,4-(methylenedioxy)-1-(2',3'-epoxypropyl)-benzene (safrole oxide) induced a typical apoptosis in A549 human lung cancer cells by activating caspase-3, -8, and -9. In this study, we further investigated which upstream pathways were activated by safrole oxide during the apoptosis. Immunofluorescence assay combined with laser scanning confocal microscopy revealed that both Fas and Fas ligand (FasL) were up-regulated by the small molecule. In addition, Fas protein distribution was altered, showing a clustering distribution instead of a homogeneous one. Subsequently, Western blot analysis confirmed the up-regulations of Fas and its membrane-binding form of FasL (m-FasL), as well as P53 protein. Conversely, safrole oxide hardly affected integrin beta4 subunit expression or distribution, which was reflected from the data obtained by immunofluorescence assay combined with laser scanning confocal microscopy. The results suggested that Fas/FasL pathway might be involved in safrole oxide-induced apoptosis of A549 cells, while integrin beta4 might be irrelevant to the apoptosis. Nevertheless, we first found the strong expression of integrin beta4 in A549 cells. The study first suggested that safrole oxide might be used as a small molecular promoter of Fas/FasL pathway to elicit apoptosis in A549 cells, which would lay the foundation for us to insight into the new strategies for lung cancer therapy.     

Safrole-induced Ca2+ mobilization and cytotoxicity in human PC3 prostate cancer cells

The effect of the carcinogen safrole on intracellular Ca2+ mobilization and on viability of human PC3 prostate cancer cells was examined. Cytosolic free Ca2+ levels ([Ca2+]i) were measured by using fura-2 as a probe. Safrole at concentrations above 10 microM increased [Ca2+]i in a concentration-dependent manner with an EC50 value of 350 microM. The Ca2+ signal was reduced by more than half after removing extracellular Ca2+ but was unaffected by nifedipine, nicardipine, nimodipine, diltiazem, or verapamil. In Ca2+-free medium, after treatment with 650 microM safrole, 1 microM thapsigargin (an endoplasmic reticulum Ca2+ pump inhibitor) failed to release Ca2+. Neither inhibition of phospholipase C with U73122 nor modulation of protein kinase C activity affected safrole-induced Ca2+ release. Overnight incubation with 0.65-65 microM safrole did not affect cell viability, but incubation with 325-625 microM safrole decreased viability. Collectively, the data suggest that in PC3 cells, safrole induced a [Ca2+]i increase by causing Ca2+ release from the endoplasmic reticulum in a phospholipase C- and protein kinase C-independent fashion, and by inducing Ca2+ influx. Safrole can decrease cell viability in a concentration-dependent manner.     

Screening of safrole, eugenol, their ninhydrin positive metabolites and selected secondary amines for potential mutagenicity.

The mutagenicity of safrole, eugenol, the secondary amines, with which they combine during metabolism, and the ninhydrin positive urinary metabolites of safrole and eugenol was tested. The panel of tests included the direct bacterial assay, a microsomal mutagenesis assay and a host-mediated assay. With the direct bacterial assay employing four mutant strains of Salmonella typhimurium (TA1530, TA1531, TA1532, TA1964), all the compounds gave negative results. In the microsomal mutagenesis assay, employing the same four mutant strains, safrole and safrole metabolite II were mutagenic with strains TA1530 and TA1532. Dimethylamine was also found to be a weak mutagen in the microsomal mutagenesis assay with strain TA1530. Safrole and safrole metabolite II were also mutagenic in the host-mediated assay with strains TA1950 and TA1952. Negative results were observed for safrole metabolites I and III, eugenol, eugenol metabolites I and II, piperidine, pipecolic acid, proline, and pyrrolidine in all three assay systems.     

Safrole oxide is a useful tool for investigating the effect of apoptosis in vascular endothelial cells on neural stem cell survival and differentiation in vitro

Previously, we found safrole oxide could promote VEC apoptosis, however, it is not known whether it can induce NSC apoptosis. It is reported that neural stem cells (NSCs) are localized in a vascular niche. But the effects of apoptosis in vascular endothelial cells (VEC) on NSC growth and differentiation are not clear. To answer these questions, in this study, we co-cultured NSCs with VECs in order to imitate the situation in vivo, in which NSCs are associated with the endothelium, and treated the single-cultured NSCs and the co-cultured NSCs with safrole oxide. The results showed that safrole oxide (10-100 microg/mL) had no effects on NSC growth. Based on these results, we treated the co-culture system with this small molecule. The results showed that the NSCs differentiation, into neurons and gliacytes was induced by VECs untreated with safrole oxide. But in the co-culture system treated with safrole oxide, the NSCs underwent apoptosis. The data suggested that when VEC apoptosis occurred in the co-culture system, the NSC survival and differentiation could not be maintained, and NSCs died by apoptosis. Our finding provided a useful tool for investigating the effect of apoptosis in vascular endothelial cells on neural stem cell survival and differentiation in vitro.     

Safrole oxide induced human umbilical vein vascular endothelial cell differentiation into neuron-like cells by depressing the reactive oxygen species level at the low concentration

Previously, we found that 5-25 microg/ml safrole oxide could inhibit apoptosis and dramatically make a morphological change in human umbilical vein vascular endothelial cells (HUVECs). But the possible mechanism by which safrole oxide function is unknown. To answer this question, in this study, we first investigated the effects of it on the activity of nitric oxide synthetase (NOS), the expressions of Fas and integrin beta4, which play important roles in HUVEC growth and apoptosis, respectively. The results showed that, at the low concentration (10 microg/ml), safrole oxide had no effects on NOS activity and the expressions of Fas and integrin beta4. Then, we investigated whether HUVECs underwent differentiation. We examined the expressions of neuron-specific enolase (NSE) and neurofilament-L (NF-L). Furthermore, we analyzed the changes of intracellular reactive oxygen species (ROS). After 10 h of treatment with 10 microg/ml safrole oxide, some HUVECs became neuron-like cells in morphology, and intensively displayed positive NSE and NF-L. Simultaneously, ROS levels dramatically decreased during HUVECs differentiation towards neuron-like cells. At the low concentration, safrole oxide induced HUVECs differentiation into neuron-like cells. Furthermore, our data suggested that safrole oxide might perform this function by depressing intracellular ROS levels instead of by affecting cell growth or apoptosis signal pathways.     

Safrole Analysis by GC-MS of Prototrophic (Ocotea odorifera (Vell.) Rohwer) Cell Cultures

Ocotea odorifera, a tree native to the Atlantic rainforest in south Brazil, has been used as a source of sassafras oil rich in safrole [5-(2-propenyl)-1,3-benzodioxole], an aromatic ether used as a flavoring agent and also in the manufacture of insecticides. The intensive deforestation process of the Brazilian Atlantic rainforest has threatened with extinction some species including O. odorifera. In this context, O. odorifera cell cultures might be an interesting alternative for the production of secondary metabolites of value (i.e., safrole), without risk of damage to the native germplasm. Insights into the secondary metabolites of organosolvent extracts of prototrophic cell cultures were performed by GC-MS and MALDI-TOF MS. GC-MS analysis revealed the occurrence of safrole in concentration of ca. 62.6 μg ml⁻¹ and allowed the identification of other compounds besides safrole in the crude extract of these cultures. Thus, alkyl phenol (C₈) [m/z= 206], C₁₄ myristic acid [m/z= 228], long chain olefin or alcohol and essential oil were detected. However, safrole [m/z= 162] was not detected by MALDI-TOF MS, indicating that it is not easily protonated (M + H)⁺.     

Suppressing Akt phosphorylation and activating Fas by safrole oxide inhibited angiogenesis and induced vascular endothelial cell apoptosis in the presence of fibroblast growth factor-2 and serum

At present, vascular endothelial cell (VEC) apoptosis induced by deprivation of fibroblast growth factor-2 (FGF-2) and serum has been well studied. But how to trigger VEC apoptosis in the presence of FGF-2 and serum is not well known. To address this question, in this study, the effects of safrole oxide on angiogenesis and VEC growth stimulated by FGF-2 were investigated. The results showed that safrole oxide inhibited angiogenesis and induced VEC apoptosis in the presence of FGF-2 and serum. To understand the possible mechanism of safrole oxide acting, we first examined the phosphorylation of Akt and the activity of nitric oxide synthase (NOS); secondly, we analyzed the expressions and distributions of Fas and P53; then we measured the activity of phosphatidylcholine specific phospholipase C (PC-PLC) in the VECs treated with and without safrole oxide. The results showed that this small molecule obviously suppressed Akt phosphorylation and the activity of NOS, and promoted the expressions of Fas and P53 markedly. Simultaneously, Fas protein clumped on cell membrane, instead of homogenously distributed. The activity of PC-PLC was not changed obviously. The data suggested that safrole oxide effectively inhibited angiogenesis and triggered VEC apoptosis in the presence of FGF-2 and serum, and it might perform its functions by suppressing Akt/NOS signal pathway, upregulating the expressions of Fas and P53 and modifying the distributing pattern of Fas in VEC. This finding provided a powerful chemical probe for promoting VEC apoptosis during angiogenesis stimulated by FGF-2.     

Safrole oxide induces human umbilical vein endothelial cell transdifferentiation to 5-hydroxytryptaminergic neuron-like cells through tropomyosin receptor kinase A/cyclooxygenase 2/nuclear factor-kappa B/interleukin 8 signaling

The phenomenon of endothelial-neural transdifferentiation has been observed for a long time, but the mechanism is not clear. We previously found that safrole oxide induced human umbilical vein endothelial cell transdifferentiation into neuron-like cells. In this study, we first validated that these cells induced by safrole oxide were functional 5-hydroxytryptaminergic neuron-like cells. Then, we performed microarray analysis of safrole oxide-treated and -untreated human umbilical vein endothelial cells. Safrole oxide elevated the levels of cyclooxygenase 2 (COX-2), interleukin-8 (IL-8) and reactive oxygen species (ROS), which was accompanied by nuclear factor-kappa B (NF-κB) nuclear translocation during the transdifferentiation. Blockade of tropomyosin receptor kinase A (TrkA) by an inhibitor or short hairpin RNA inhibited the levels of COX-2/IL-8 and the nuclear translocation of NF-κB but did not suppress the increased ROS level. As a result, cells underwent apoptosis. Therefore, via TrkA, safrole oxide may induce endothelial cell transdifferentiation into functional neuron-like cells. During this process, the increased levels of COX-2/IL-8 and the subsequent elevation of ROS production induced NF-κB nuclear translocation and IL-8 secretion. With the activity of TrkA inhibited, the inactive NF-κB regulated the ROS level in a negative feedback manner. Finally, the transdifferentiation pathway was blocked and cells became apoptotic. The TrkA/COX-2/IL-8 signal pathway may have an important role in endothelial-neural transdifferentiation, and safrole oxide may trigger this process by activating TrkA.     

Volatile oils ofIlucium floridanum andI. parviflorum (Illiciaceae) of the southeastern United States and their potential economic utilization

The essential oil from leaves and branches of cultivatedIllicium floridanum is dominated by 22.53 ± 2.23% linalool and 13.93 ± 1.61% linalyl acetate. The essential oil from leaves and branches of cultivated I. parviflorum is dominated by 68.14 ± 0.88% safrole, 13.18 ± 1.01% linalool, and 11.89 ± 0.87% methyl eugenol. Besides the Lauraceae and Piperaceae, the Illiciaceae is another natural source of safrole.     

Safrole-DNA adducts in tissues from esophageal cancer patients: clues to areca-related esophageal carcinogenesis

Epidemiological studies have demonstrated that areca quid chewing can be an independent risk factor for developing esophageal cancer. However, no studies are available to elucidate the mechanisms of how areca induces carcinogenesis in the esophagus. Since the areca nut in Taiwan contains a high concentration of safrole, a well-known carcinogenic agent, we analyzed safrole-DNA adducts by the 32P-postlabelling method in tissue specimens from esophageal cancer patients. In total, we evaluated 47 patients with esophageal cancer (16 areca chewers and 31 non-chewers) who underwent esophagectomy at the National Taiwan University Hospital between 1996 and 2002. Of the individuals with a history of habitual areca chewing (14 cigarette smokers and two non-smokers), one of the tumor tissue samples and five of the normal esophageal mucosa samples were positive for safrole-DNA adducts. All patients positive for safrole-DNA adducts were also cigarette smokers. Such adducts could not be found in patients who did not chew areca, irrespective of their habits of alcohol consumption or cigarette smoking (p<0.001, comparing the areca chewers with non-chewers). The genotoxicity of safrole was also tested in vitro in three esophageal cell lines and four cultures of primary esophageal keratinocytes. In two of the esophageal keratinocyte cultures, adduct formation was increased by treatment with safrole after induction of cytochrome P450 by 3-methyl-cholanthrene. This paper provides the first observation of how areca induces esophageal carcinogenesis, i.e., through the genotoxicity of safrole, a component of the areca juice.     

Biotransformation and cytotoxic effects of hydroxychavicol, an intermediate of safrole metabolism, in isolated rat hepatocytes

The biotransformation and cytotoxic effects of hydroxychavicol (HC; 1-allyl-3,4-dihydroxybenzene), which is a catecholic component in piper betel leaf and a major intermediary metabolite of safrole in rats and humans, was studied in freshly isolated rat hepatocytes. The exposure of hepatocytes to HC caused not only concentration (0.25-1.0 mM)- and time (0-3 h)-dependent cell death accompanied by the loss of cellular ATP, adenine nucleotide pools, reduced glutathione, and protein thiols, but also the accumulation of glutathione disulfide and malondialdehyde, indicating lipid peroxidation. At a concentration of 1 mM, the cytotoxic effects of safrole were less than those of HC. The loss of mitochondrial membrane potential and generation of oxygen radical species assayed using 2′,7′-dichlorodihydrofluoresein diacetate (DCFH-DA) in hepatocytes treated with HC were greater than those with safrole. HC at a weakly toxic level (0.25 and/or 0.50 mM) was metabolized to monoglucuronide, monosulfate, and monoglutathione conjugates, which were identified by mass spectra and/or ¹H nuclear magnetic resonance spectra. The amounts of sulfate rather than glucuronide or glutathione conjugate predominantly increased, accompanied by a loss of the parent compound, with time. In hepatocytes pretreated with either diethyl maleate or salicylamide, HC-induced cytotoxicity was enhanced, accompanied by a decrease in the formation of these conjugates and by the inhibition of HC loss. Taken collectively, our results indicate that (a) mitochondria are target organelles for HC, which elicits cytotoxicity through mitochondrial failure related to mitochondrial membrane potential at an early stage and subsequently lipid peroxidation through oxidative stress at a later stage; (b) the onset of cytotoxicity depends on the initial and residual concentrations of HC rather than those of its metabolites; (c) the toxicity of HC is greater than that of safrole, suggesting the participation of a catecholic intermediate in safrole cytotoxicity in rat hepatocytes.     

Immunohistochemical demonstration of decreased L-pyruvate kinase in enzyme altered rat liver lesions produced by different carcinogens

Preneoplastic liver lesions were produced in female Wistar rats by application of 25 mg/kg N-nitrosomorpholine (NNM), 14 mg/kg diethylnitrosamine (DENA), 0.075 mg/kg aflatoxin B1 (AFB1) or 160 mg/kg safrole. These carcinogens were administered in two equal doses 12 and 24 h after partial hepatectomy. The animals then received sodium phenobarbital (0.1% in tap water) for up to 410 days. Numerous altered hepatic foci (AHF) and hyperplastic nodules (HN) were detected enzyme histochemically by their negative ATPase reaction after application of AFB1, DENA and NNM; some AHF and HN were also caused by the weak carcinogen safrole. Immunohistochemically these lesions were also L-pyruvate kinase (L-PK)-negative with a high coincidence with regard to their number and area. These results confirm the role of L-PK, an enzyme affecting the pentose phosphate pathway, as a negative marker of preneoplastic liver lesions.     

Upregulating of Fas, integrin beta4 and P53 and depressing of PC-PLC activity and ROS level in VEC apoptosis by safrole oxide

Previously, we found that safrole oxide could trigger vascular endothelial cell (VEC) apoptosis. In this study, to investigate its mechanism to induce apoptosis in VECs, the activities of nitric oxide synthetase and phosphatidylcholine specific phospholipase C, the level of reactive oxygen species and the expressions of Fas, integrin beta4 and P53 were analyzed. The data showed that safrole oxide induced apoptosis by increasing the expressions of Fas, integrin beta4 and P53, and depressing the activity of Ca(2+)-independent phosphatidylcholine-specific phospholipase C and intracellular reactive oxygen species levels in VECs.     

Metabolic activation of a natural promutagen, eugenol, by replicative cultures of adult rat liver epithelial cells]

The metabolism of eugenol via the epoxide-diol pathway has been characterized by the incubation of this compound in replicative cultures of epithelial cells from sexed rat liver. It is a naturally occuring compound which has been found as the second metabolite of a known hepatocancerogen: safrole. Eugenol did not appear as a mutagen while epoxyeugenol was mutagenic using the Ames Test.