Involvement of mitochondrial pathway in benzo[a]pyrene-induced neuron apoptosis

Abstract

Benzo[a]pyrene (B[a]P), a well-known carcinogen, is widespread in the environment. Although the neurotoxic effect of B[a]P has not drawn much attention, toxic effects of B[a]P on learning and memory have been reported. Since it is well known that neuronal apoptosis plays a major role in impairment of learning and memory triggered by many stimuli, an effort has been made to examine whether the B[a]P-induced neurotoxicity occurs through mitochondria-mediated apoptosis. Cultured newborn rat cerebral neurons were used to clarify the apoptosis induced by B[a]P in the study. After incubating with different doses of B[a]P in presence of S9 for 40 h, the apoptotic rates of B[a]P-treated neurons increased in a dose-dependent manner. Further analysis showed that B[a]P-induced apoptosis was accompanied by loss of mitochondrial membrane potential, release of cytochrome c from mitochondria to the cytosol, downregulation of antiapoptotic protein B-cell lymphoma-2 (Bcl-2) levels with concurrent upregulation in proapoptotic Bcl-2-associated X protein (Bax) levels, and increase in the levels and activities of caspases-9 and -3. However, there was no difference in the activity of caspase-8 between B[a]P-exposed neurons and controls. Collectively, these results showed that B[a]P upregulates Bax and downregulates Bcl-2 expression in cultured cerebral neurons, which leads to mitochondrial release of cytochrome c, caspase-3 activation and neuronal apoptotic death.

Introduction

Benzo[a]pyrene (B[a]P), a prototypic polycyclic aromatic hydrocarbon(PAH), is formed during incomplete combustion of organic wastes and fuels, during various industrial processes, cigarette smoking, and food cooking. Human exposure to B[a]P occurs through inhalation of air particles or ingestion of contaminated water or food. ^1,2 While the carcinogenicity of PAHs is well established, ³ their neurotoxic effects have received less attention. Majchrzak et al. ⁴ reported short-term memory disorders in workers of a coke processing plant in Poland. Learning disorders were observed in children exposed during early life to high levels of PAHs in the Czech Republic. ⁵ In a previous cross-sectional study, memory impairment was found in coke oven workers. ⁶ Grova et al. ⁷ reported that at low doses (0.02 and 0.2 mg/kg), B[a]P impaired short-term learning and spatial memory performance in Y-maze and Morris water maze tests and had no effect on locomotor activity and motor coordination when subacutely exposed to B[a]P in adult mice. Learning and memory deficits induced by B[a]P were shown in rat models at a dose of 2 mg/kg by gavage or of 2.5 and 6.25 mg/kg by intraperitoneal injection. ^{8 –10} Such observations have led to speculations about a possible relationship between memory impairment and exposure to environmental PAHs. ^11,12

It is well known that neuronal apoptosis plays a key role in the impairment of learning and memory and is triggered by physical and chemical factors, such as high altitude, sulfur mustard, and aluminum. ^{13 –17} As a cytotoxic environmental pollutant, B[a]P induces apoptotic cell death in macrophages, mouse hepa-1c1c7, and RL95-2 human endometrial cancer cell models. ^{18 –21} Tu et al. ²² and Cen et al., ²³ reported that chronic exposure to B[a]P induced neuronal apoptosis in adult mice hippocampus. Our study found that B[a]P can induce hippocampus neuronal cell apoptosis and changes in mitochondria in rats by intracerebroventricular injection. ²⁴ Neuronal apoptosis may be one mechanism that B[a]P induces learning and memory deficiencies.

At present, the mechanism of B[a]P-induced neuronal apoptosis remains unclear. The signaling events leading to apoptosis can be divided into two major pathways, either Fas and Fas ligand (FasL) or the mitochondrial pathway. The Fas and FasL pathways involve the activation of membrane death receptors, which recruit adaptor protein Fas-associated protein with death domain and induce autocatalytic activation of procaspase-8. Activated caspase-8 initiates apoptotic process by activating downstream effecter caspases such as caspase-3. The mitochondrial pathway results from the release of cytochrome c from the mitochondria. Once released into the cytosol, cytochrome c forms a complex with apoptotic protease-activating factor 1 (Apaf-1) and induces cleavage and activation of procaspase-9, which subsequently activates downstream caspases including caspase-3. ^25,26 In the present study, the molecular mechanisms of neuronal apoptosis have been investigated using primary culture of rat cortical neurons exposed to B[a]P.

Materials and methods

Preparation of primary culture of neurons

Primary cortical cells were prepared from cerebral cortices of newborn Sprague Dawley rats (postnatal 0–3 days) by the method described by Zhang et al. ²⁷ Briefly, the newborn rats were put into 75% freezing ethanol solution for 10 min. After the cerebral cortices were dissected from newborn rats, the meninges and white matter were removed. Then the remaining tissue was treated with 0.25% trypsin (Sigma-Aldrich, St. Louis, MO, USA) for 20 min at the temperature of 37°C. Next, dissociated cortical cells were filtered through a 220-mesh nylon grid and plated in either Nunc 4-wells (3 × 10⁵ cells/well) or 35 × 10 mm (1.5 × 10⁶ cells) dishes (GIBCO BRL, Grand Island, NY, USA) coated with poly-d-lysine (Sigma-Aldrich) previously in complete cell medium, which is composed of Dulbecco’s modified Eagle’s medium (DMEM) suspension (HyClone, Logan, UT, USA), 5% fetal bovine serum (HyClone), 5% horse serum (HyClone), penicillin (100 U/mL) (Gibco BRL, Grand Island, NY, USA), and streptomycin (100 mg/mL) (Gibco BRL, Grand Island, NY, USA). Finally, the cells were incubated at the temperature of 37°C in 5% carbon dioxide and 95% humidity atmosphere. After a 24 h incubation, the cortical cells were added into culture medium with arabinosylcytosine (final concentration 10 μM) to prevent the proliferation of nonneuronal cells. The complete cell culture medium was replaced twice per week. At least 95% of the cells were confirmed to be neurons by tubulin-3 immunostaining (Figure 1).

Figure 1.

Cultured cells were immunostained by tubulin-3, which is a specific antigen in neurons. The figure showed that at least 95% of the cells were neurons.

B[a]P treatment

After 5 days of incubation, neurons grew into 80% confluent monolayer. The cells were treated with B[a]P (Sigma) at different concentrations in culture medium freshly prepared with 40 mM B[a]P stock solution dissolved in dimethyl sulfoxide. As the parent B[a]P needs bioactivation to reactive metabolites to exert its toxic effects, S9 mixture (The S9 mixture was prepared, just before use, by adding 500 μl phosphate buffer (0.2 M), 130μl deionised water, 100 μl KCl (0.33 M), 80 μl MgCl2 (0.1 M), 100 μl S9 fraction, 50 μl glucose-6-phosphate (0.1M) and 40 μl nicotinamideadeninedinucleotidephosphate (NADP) (0.1 M)) was added at 3% (v/v) of the total culture medium. The exposure time to B[a]P solution was 40 h. Three replicates of each dish were measured.

Cell viability assay

Tetrazolium salts, including 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2h-tetrazolium bromide (MTT) and 2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carbox-anilide formazan dyes, have been widely used to assay cell cytotoxicity. In the present study, Cell Counting Kit-8 (CCK-8; Dojindo Molecular Technologies, Inc., Kumamoto, Japan), an alternative MTT method with higher sensitivity and less cytotoxicity was used to measure cell viability. Briefly, 10 μL CCK-8 solution was added into each well of the plate, which was then incubated for 1–4 h in the incubator. The absorbance of optical density (OD) value in each well was then measured at 450 nm using a Spectra Max 340 Microplate Reader (Molecular Devices Corporation, Sunnyvale, California, USA).The cell viability was calculated using the following formula

$\text{cell viability}(\%)=1-\left(\frac{\text{average}\hspace{0.17em}\text{OD}\hspace{0.17em}\text{value}\hspace{0.17em}\text{of}\text{wells}\hspace{0.17em}\text{with}\hspace{0.17em}\text{administered}\hspace{0.17em}\text{B[a]P}}{\text{average}\hspace{0.17em}\text{OD}\hspace{0.17em}\text{value}\hspace{0.17em}\text{of}\hspace{0.17em}\text{control}\hspace{0.17em}\text{cells}}\right)\times 100$

Annexin V-FITC/PI assay

Apoptosis was detected using an Annexin V-fluorescein isothiocyanate (V-FITC)/propidium iodide (PI) detection kit (BipecBiopharma Corp., Cambridge, MA, USA) according to the manufacturer’s directions. Briefly, cells were harvested and resuspended in binding buffer (1 × 10⁶ cells/mL). Aliquots of 10⁵ cells were mixed with 5 μL of Annexin V-FITC and 10 μl of PI. After incubating for 15 min at room temperature in the dark, fluorescence was detected by flow cytometry (FACS-400, Becton Dickinson, Mountain View, CA, USA). The percentage of apoptosis was determined from the number of Annexin V(+)/PI(−) cells + the number of Annexin V(+)/PI(+) cells.

Measurement of MMP (Δψm)

As an index to determine mitochondrial dysfunction, mitochondrial membrane potential (MMP) was monitored using Rhodamine 123. Neuronal cells were treated with or without B[a]P (10, 20, and 40 μM) for 40 h in a six-well culture plate at 1 × 10⁵ cells/mL. The medium was then removed and washed three times with serum-free DMEM medium followed by incubation in fresh serum-free medium containing 3 mg/L Rhodamine 123 at 37°C in the dark for 30 min. Finally, the cells were collected and washed twice with phosphate-buffered saline (PBS) and then analyzed by a FACScan flow cytometer (FACS-400, USA).

Western blot analysis of caspases and Bcl-2 family proteins

Cultured neurons were solubilized with lysis buffer containing 150 mM sodium chloride, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS), 50 mM tris(hydroxymethyl)aminomethane (Tris)–hydrochloric acid (pH 8), 5 mM ethylenediamineteraacetic acid, 10 μg/mL leupeptin, 10 μg/mL benzamidine, 10 μg/mL aprotinin, 2 μg/mL pepstatin, and 0.1 mM phenylmethylsulfonyl fluoride (PMSF). Following centrifugation at 15,000g for 20 min, solubilized supernatants were collected and the protein concentration was measured by a bicinchoninic acid assay (BCA) protein assay kit (Pierce, Rockford, Illinois, USA). Protein extracts from each sample were added to a gel loading buffer (100 mM Tris, pH 6.8, 20% glycerol, 200 mM dithiothreitol (DTT), 4% SDS, 0.03% bromophenol blue) and boiled for 5 min. Proteins (25 µg/sample) in loading buffer were fractionated on 10–12% SDS–polyacrylamide gel. Proteins were transferred to polyvinylidene difluoride (PVDF) membrane, and the membranes were incubated at 4°C overnight with one of the following diluted primary antibodies: (1) monoclonal mouse caspase-9 polyclonal antibody (Cell Signaling Technology, Beverly, MA, USA), (2) monoclonal mouse caspase-3 polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA), (3) monoclonal mouse anti-Bcl-2-associated X protein (Bax) antibody (Santa Cruz), and (4) anti-B cell lymphoma-2 (Bcl-2) polyclonal antibody (Santa Cruz). After washing four times in Dulbecco’s PBS (DPBS) containing 0.05% Tween-20 for 5 min each, the membranes were blocked in 5% nonfat powdered milk in DPBS for 2 h at 37°C, and then the membranes were incubated with anti-immunoglobulin G (IgG) antibody for 2 h. The membranes were then washed for four times in DPBS containing 0.05% Tween-20 for 10 min each, and immunoreactive proteins on the membrane were visualized using enhanced chemiluminescence kit (Amersham Biosciences, Piscataway, NJ, USA). To confirm that equal amounts of protein samples were loaded, membranes were stripped and reblotted with monoclonal anti-actin antibody (CHEMICON International Inc., Temecula, CA, USA). To measure B[a]P-induced alteration in the expression of Bcl-2 family proteins, gel bands were quantified with the aid of a densitometer (Molecular Dynamics Model 375A, International Equipment Trading Ltd., Vernon Hills, Illinois, USA) and normalized with reprobed actin signal on the same blot and expressed as percentage of control.

Measurement of activities of caspases-3, -8, and -9

The activities of caspases-3, -8, and -9 were measured using a caspase colorimetric assay kit (BioVision, Inc., Mountain View, CA, USA) according to manufacturer’s instructions. Briefly, cultured cerebral neurons were lysed using lysis buffer provided in the kit. The cell lysate was centrifuged at 15,000g for 5 min at 4°C, and the supernatant was incubated with substrate Asp-Glu-Val-Asp (DEVD)-p-nitroanilide (pNA) for caspase-3, N-acetyl-Ile-Glu-Thr-Asp-p-nitroanilide (Ac-IETD-pNA) for caspase-8 and Leu-Glu-His-Asp-p-nitroanilide (LEHDpNA) for caspase-9 at 37°C for 2 h. Release of chromophore pNA from the substrate was quantified using enzyme-linked immunosorbent assay plate reader at 405 nm.

Immunoblotting analysis of cytochrome c

To analyze mitochondrial release of cytochrome c, mitochondrial and cytosolic fractions of cultured cerebral neurons were prepared. Briefly, cerebral neurons were lysed in ice-cold buffer containing 210 mM mannitol, 70 mM sucrose, 10 mM 2-[4-(2-Hydroxyethyl)-1-piperazinyl]ethanesulfonic acid (HEPES), pH 7.3, 1 mM ethylene glycol tetraacetic acid, 1 mM DTT, 5 μg/mL pepstain, 5 μg/mL leupeptin, 5 μg/mL aprotinin, and 0.3 mM PMSF. Cells were homogenized using a Dounce tissue grinder on ice, and the cell lysate was centrifuged at 500g for 10 min at 4°C. The supernatant was centrifuged at 9500g for 9 min at 4°C to collect the mitochondrial fraction. The resulting supernatant was further centrifuged at 16,000g for 20 min at 4°C, and the final supernatant was used as cytosolic fraction. The protein concentration was measured by a BCA protein assay kit. Protein extracts from each sample were added to a gel loading buffer (100 mM Tris, pH 6.8, 20% glycerol, 200 mM DTT, 4% SDS, 0.03% bromophenol blue) and boiled for 5 min. Protein samples (15 μg) of cytosolic and mitochondrial extracts were fractionated on 15% SDS–polyacrylamide gel and transferred to a PVDF membrane. The membrane was then incubated with diluted monoclonal anti-cytochrome c antibody at 4°C overnight. After washing four times in DPBS containing 0.05% Tween-20 for 5 min each, the membranes were incubated with anti-IgG antibody for 2 h. The membranes were then washed for four times in DPBS containing 0.05% Tween-20 for 10 min each, and immunoreactive proteins on the membrane were visualized by ECL protocol as described above. To verify the purity of the mitochondrial fraction and visualize possible mitochondrial contamination of the cytosolic extract, the membrane was stripped and reblotted with monoclonal anti-cytochrome c oxidase subunit IV (COX-IV), a mitochondrial marker. COX-IV was present in mitochondrial fraction and not detected in the cytosolic extract. Gel bands were quantified by densitometric scanning, normalized with reprobed actin or COX-IV and expressed as percentage of the control.

Statistical analysis

Except that the number of experiments on cell viability assay and Annexin V-FITC/PI assay was five, the number of experiments on the other indexes in the text was three. Also, the number of replicates for each experiment was three. Statistical significance was determined using the analysis of variance followed by Tukey’s post hoc test as appropriate. Each value was expressed as mean ± SD. Differences were considered significant when the calculated p value was <0.05.

Results

B[a]P causes apoptotic death of cerebral neurons

A total of eight B[a]P concentrations, which were 0.5, 1, 2, 5, 10, 20, 40, and 80 µM, were used with S9. CCK-8 assays demonstrated that B[a]P induced cerebral neuronal death at 2, 5, 10, 20, 40, and 80 µM. (Figure 2(a)). In this study, B[a]P-induced cerebral neuronal cell apoptosis was evaluated using flow cytometry. Compared with control, the rates of neuronal apoptosis were not significantly different in 0.5, 1, 2, and 5 µM, and the neuronal apoptosis level increased significantly in 10, 20, 40, and 80 µM in a dose-dependent manner. These results suggest that B[a]P should exert its neurotoxic effect via activating the apoptotic death pathway. As shown in Figure 2(b), the rates of neuronal apoptosis at the doses of 10, 20, and 40 µM B[a]P showed a linear tendency. So 10, 20, and 40 µM were chosen as the experimental doses in the following tests.

Figure 2.

B[a]P induces cerebral neuronal death. (a) Cell viability was measured by performing a CCK-8 assay. Note that B[a]P at 2, 5, 10, 20, 40, and 80 µM caused cerebral neuronal death. (b) B[a]P induces neuronal cell apoptosis in a dose-dependent manner. Each bar represents mean ± SD value (n = 5 experiments). *p < 0.05 and **p < 0.01 indicate significant difference between control neurons and neurons treated with B[a]P. B[a]P: benzo[a]pyrene; CCK-8: Cell Counting Kit-8; SD: standard deviation.

B[a]P decreases MMP (Δψm)

Disruption of mitochondrial integrity potential changes in mitochondrial membrane is one of the early events leading to apoptosis. To assess the effects of B[a]P on the changes of Δψm in cortical neurons, flow cytometric analysis was carried out to detect the fluorescence intensity of Rhodamine 123. As shown in Figure 3, exposure to B[a]P for 40 h resulted in a significant decrease in Δψm from 97.69 ± 1.17% to 92.01 ± 2.28%, 88.02 ± 1.23%, and 78.45 ± 2.28%, at 10, 20, and 40 μM respectively (Figure 3). These results suggest that the treatment with various concentrations of B[a]P (10, 20, and 40 μM) for 40 h result in significant decreases in Δψm. The results imply that B[a]P induces Δψm dissipation in a concentration-dependent manner.

Figure 3.

B[a]P induces the loss of MMP in cerebral neurons. Exposure to B[a]P for 40 h resulted in a significant decrease in Δψm in a concentration-dependent manner. Each bar represents the mean ± SD value (n = 3 experiments). *p < 0.05 and **p < 0.01 indicate the significant difference between control neurons and neurons treated with B[a]P. B[a]P: benzo[a]pyrene; MMP: mitochondrial membrane potential; SD: standard deviation.

B[a]P downregulates Bcl-2 and upregulates Bax expression in cerebral neurons

Since translocation of Bcl-2 families from the mitochondria to the cytosol is known to play a key role in mitochondrial-mediated apoptosis induced by a variety of apoptotic stimuli, western blot analysis was performed to assess the effect of B[a]P on the expression of Bcl-2 and Bax proteins (Figure 4(a)). After treating cerebral neurons with B[a]P at various concentrations, for 40 h, protein levels of Bax were upregulated while those of Bcl-2 were downregulated (Figure 4(b)). Thus, the ratio of Bcl-2–Bax decreased (Figure 4(c)).

Figure 4.

B[a]P upregulates Bax and downregulates Bcl-2 protein expression in cerebral neurons. Forty hours after treatment with B[a]P, immunoblotting indicated that compared to control neurons, Bax ((a); approximately M _w = 26 kDa) protein levels were significantly increased (b) and Bcl-2 ((a); approximately M _w = 29 kDa) levels were greatly downregulated (b). Thus, there was a decrease in the ratio of Bcl-2–Bax proteins (c). Immunoblots were quantified by densitometric scanning, normalized with actin bands and expressed as a ratio relative to control. Each bar represents the mean ± SD (n = 3 experiments). *p < 0.05 and **p < 0.01 indicate the significant difference between control neurons and neurons treated with B[a]P. B[a]P: benzo[a]pyrene; M _w: molecular weight; SD: standard deviation.

Cytochrome c release from mitochondria to cytosol

Cytochrome c released from the mitochondria is a critical step in the apoptotic cascade since this activates downstream caspases. Western blot analysis was performed to investigate the effect of B[a]P on the changes of cytochrome c in mitochondria and the cytosol. After 40 h treating cerebral neurons with B[a]P at various concentrations, protein level of cytochrome c increased in cytosolic fraction and decreased in mitochondrial fraction (Figure 5).

Figure 5.

B[a]P promotes the mitochondrial release of cytochrome c. Cytochrome c was detected in cytosolic and mitochondrial fractions of cerebral neurons exposed to B[a]P for 40 h, and cytochrome c translocated from mitochondrial to cytosolic fractions. Cytochrome c was increased in the cytosolic fraction (a and c) and decreased in the mitochondrial fraction (b) in a dose-dependent manner (n = 3 experiments). Immunoblots were quantified by densitometric scanning and expressed as a ratio of control. Actin and COX-IV were used as internal controls for cytosolic and mitochondrial fractions, respectively. Each bar represents the mean ± SD. *p < 0.05 and **p < 0.01 indicate the significant difference between control neurons and neurons treated with B[a]P. B[a]P: benzo[a]pyrene; COX-IV: cytochrome c oxidase subunit IV; SD: standard deviation.

B[a]P activates caspases-9 and -3

Since caspases are known to play a central role in mediating various apoptotic responses, western blotting analysis was performed to investigate the activation of caspase-9 resulting from the cleavage and activation of procaspase-9. Caspase-9 is synthesized as a 45-kDa proenzyme that is cleaved to a large active 35-kDa fragment. After treating cerebral neurons with various concentrations of B[a]P for 40 h, a western blot using an antibody recognizing active caspase-9 showed that cleaved active caspase-9 was expressed in cerebral neurons (Figure 6(a)). The activity of caspase-9 was also measured by analyzing the cleavage of colorimetric substrate LEHD-pNA. Compared with control neurons, caspase-9 activity in cerebral neurons exposed to B[a]P were increased significantly. However, the analysis of caspase-8 activity indicated that there was no difference between B[a]P-treated groups and controls (Figure 6(c)).

Figure 6.

B[a]P induces the formation of cleaved active caspases-9 and -3 in cerebral neurons. Cleaved caspases-3 and -9 (a) were higher in B[a]P-treated groups than in controls ((b); n = 3 experiments). Gel bands were quantified by densitometric scanning, normalized with reprobed actin and expressed as a ratio to the control. The activities of caspases-3, -8, and -9 were analyzed by determining the cleavage of specific colorimetric substrates. Note that the activity of caspases-3 and -9 were increased in B[a]P dose groups, and the activity of caspase-8 was not significantly changed in neurons treated by B[a]P compared with controls ((c); n = 3 experiments). Enzyme activity was expressed as a ratio relative to control. Each bar shows the mean ± SD. *p < 0.05 and **p < 0.01 indicate the significant difference between control neurons and neurons treated with B[a]P. B[a]P: benzo[a]pyrene; SD: standard deviation.

We also found that B[a]P significantly increased the cleavage of procaspase-3 to the active form in a dose-dependent manner (Figure 6(a)). Subsequently, the presence of activated caspase-3 was further confirmed by detecting caspase-3 activity. Compared with control neurons, B[a]P-treated cerebral neurons resulted in increased caspase-3 activity in a dose-dependent manner (Figure 6(c)).

Discussion

Because of its lipophilic nature, B[a]P readily crosses the blood–brain barrier and has also been shown to accumulate and become metabolized in the brain. ^28,29 It has been hypothesized that neural cells may be highly susceptible to damage by B[a]P. ²² The present study provides direct evidence that B[a]P induces apoptotic death of cerebral neurons.

Mitochondria plays critical roles in the regulation of various apoptotic processes including toxicant-induced apoptosis. ³⁰ Anti- and proapoptotic members of Bcl-2 family proteins play critical roles in mitochondria-mediated apoptosis. ³¹ Proapoptotic Bax, activated by BH3-only Bcl-2 family proteins including Bim, induces permeabilization of mitochondrial outer membrane and causes the efflux of death-promoting cytochrome c. On the other hand, antiapoptotic Bcl-2 inhibits the mitochondrial apoptotic pathway by preventing activation of Bax or BH3-only proteins. A shift in the balance between anti- and proapoptotic Bcl-2 family proteins toward the expression and activation of Bax proteins could lead to mitochondria-dependent caspase activation and apoptotic cell death. ³² Our studies demonstrated that B[a]P upregulates Bax and downregulates Bcl-2 protein expression of cultured cerebral neurons. Upregulation and activation of Bax have been shown to mediate apoptotic death of central nervous system (CNS) neurons in response to various death-inducing neurotoxins including acrylamide, lead, cocaine, and so on. ^{33 –35} Bcl-2 expressed in embryonic and adult neurons of the CNS plays a critical role in inhibiting a neuronal apoptotic cascade that is activated during brain development or following pathologic stimuli. ^25,36,37 Thus, a decrease in the ratio of Bcl-2–Bax protein expression caused by B[a]P is expected to cause activation of Bax, leading to mitochondrial release of apoptogenic proteins and apoptotic neuronal death.

The release of proapoptotic cytochrome c from mitochondria is a pivotal event leading to subsequent caspase-9 activation and apoptotic cell death. ^31,38 Consistent with these results, in the cytosol of B[a]P-treated neuron cells, cytochrome c was detected after a 40-h treatment period. Once released into the cytosol, cytochrome c binds with procaspase-9 in presence of adenosine triphosphate and Apaf-1 to form the apoptosome. This complex activates caspase-9, which, in turn, cleaves, and there by activates caspase-3. In B[a]P-treated cells, the release of cytochrome c from the mitochondria was followed by the activation of caspases-9 and -3.

It has been reported that the release of cytochrome c appears to be dependent on the induction of mitochondrial permeability transition, which is associated with a decrease in Δψm. Therefore, the loss of Δψm and the release of apoptogenic factors, such as cytochrome c, from the mitochondria into the cytosol are associated with apoptosis induced by xenobiotic toxicants. ^39,40 In the present study, loss of Δψm and release of cytochrome c were observed in B[a]P-induced neuronal cells, resulting in the activation of caspases-9 and -3, and finally, apoptosis. Moreover, the loss of Δψm may, in fact, be a consequence of massive cytochrome c released from the mitochondria. Thus, a mitochondrial damage-dependent pathway may be involved in B[a]P-induced apoptosis in neuronal cells.

Although the activity of caspase-8 was elevated, an important initiator of death receptor-mediated apoptosis was not observed in B[a]P-exposed cerebral neurons, extrinsic death receptor-mediated apoptotic pathway in B[a]P-induced neuron apoptosis is not exclusive, except for caspase-8. Caspase-10 also participates in the death receptor-mediated apoptosis and both share a similar function. ⁴¹ Some reports showed that there were not only apoptotic death but also necrotic death in B[a]P-induced cell demise. ^42,43 Our results also suggest that B[a]P causes cerebral neuronal necrotic death (data not shown) and this research is in progress.

Although the B[a]P concentrations inducing neuron apoptosis in this study are higher than the doses at which humans are exposed, the present research provided a useful model to study the direct cytotoxicity of B[a]P on neurons. Exposing rats to B[a]P at 2 mg/kg by gavage or at 2.5 and 6.25 mg/kg by intraperitoneal injection for 3 months impairs learning and memory ability, ^{8 –10} and B[a]P induces neuronal apoptosis in rats exposed at 1.3 and 3.2 mg/kg by intraperitoneal injection for 10 weeks. ^22,23 Those doses of B[a]P may be relevant in cases where the specific populations are chronically exposed to B[a]P emanating from hazardous waste sites, where chronic intake occurs via food or tobacco, or during occupational exposures, such as coke oven emission. Waldman et al. ⁴⁴ estimated that B[a]P intake ranged from 20–800 ng/day in people living in the vicinity of hazardous waste sites contaminated by PAHs. B[a]P levels in mainstream tobacco smoke is reported to be 20–40 ng/cigarette, ⁴⁵ and coke oven workers have been exposed to as much as 42,000 ng/m³ B[a]P. ⁴⁶ The high doses of B[a]P have been used in other reports on apoptotic cell death. The concentration of 10 µM B[a]P induces apoptosis in RL95-2 human endometrial cancer cells by cytochrome P450 1A1 activation. ²⁰ Nitro-PAHs, amino-PAHs, and B[a]P at concentrations from 10 to 30 µM have also induced cell death in human bronchial epithelial BEAS-2B cells. ⁴⁷ Induction of apoptosis by PAHs such as benzo[e]pyrene in concentrations ranging from 100 to 200 µM on retinal neurosensory (R28) cells ⁴⁸ and by fluoranthene (0.2 mM) in T-cell hybridomas ⁴⁹ have also been reported.

In summary, the present study suggests that B[a]P upregulates Bax and downregulates Bcl-2 expression in cultured cerebral neurons, which leads to mitochondrial release of cytochrome c, caspase-3 activation, and apoptotic neuronal death.