Introduction

Arsenic exposure, via contaminated water and food as well as insecticides and chemotherapy such as arsenic trioxide, reportedly induces neurotoxicity in the central and peripheral nervous systems [1–7]. Oxidative stress has been suggested as one of the mechanisms underlying arsenic-induced neurotoxicity [8–19]. Many studies have shown that arsenics are capable of generating free radical formation [9–12] and inducing oxidative injury, including oxidative DNA adducts and cell death [10–16]. Both necrosis and apoptosis have been demonstrated to mediate arsenic-induced neurotoxicity [10–16]. Consistently, our previous studies showed that intranigral infusion of sodium arsenite (arsenite) induced lipid peroxidation and striatal dopamine depletion in the nigrostriatal dopaminergic system of rat brain [17, 18]. Furthermore, several antioxidants including melatonin and N-acetylcysteine were employed to attenuate arsenite-induced apoptosis in rat brain [18] and dorsal root ganglion (DRG) explants [19].

Heme oxygenase (HO), the rate-limiting enzyme for heme catabolism, degrades intracellular heme to free iron, carbon monoxide, and biliverdin which is subsequently converted to bilirubin [20]. There are three isoforms: HO-1(aka heat shock protein 32), HO-2, and HO-3 [20]. HO-1, known as a chaperone protein, is reportedly induced by a wide range of prooxidants and insults [17–23]. Clinical studies have shown augmented HO-1 expression in the affected neural tissues of patients with CNS neurodegenerative diseases [24]. In support of this observation, our previous studies demonstrated HO-1 elevation in the arsenite-infused substantia nigra (SN) of rat brain [17, 18]. In addition to the reduction in arsenite-induced apoptosis mentioned previously, antioxidative treatment was found to attenuate arsenite-induced HO-1 elevation in vivo and in vitro [9, 18, 23]. Due to the antioxidative property of HO-1 and its metabolites, several studies have shown that HO-1 overexpressed neurons are resistant to oxidative stress and neurotoxins, indicating an endogenous cytoprotective mechanism of HO-1 [25, 26]. Furthermore, phytochemicals capable of elevating HO-1 expression have been suggested as neuroprotective agents for CNS neurodegenerative diseases [27, 28]. In the present study, we set forth two aims: one was to elucidate the involvement of oxidative stress in arsenite-induced neurotoxicity and the other was to investigate the role of HO-1 in arsenite-induced neurotoxicity in primary cultured cortical neurons.

Materials and Methods

Animals

Pregnant female Sprague–Dawley rats were supplied by the National Laboratory Animal Breeding and Research Center, Taipei, Taiwan, R.O.C. In accordance with the US National Institutes of Health guidelines regarding the care and use of animals for experimental procedures, rats were sacrificed with an overdose of chloral hydrate (Sigma, St. Louis, MO) to minimize pain or discomfort. The use of animals has been approved by the Institutional Animal Care and Use Committee of Taipei Veterans General Hospital, Taipei, Taiwan, R.O.C.

Primary Culture of Cortical Neurons

Primary cultured cortical neurons were prepared from embryonic rat brains (day 17) and characterized as previously described [29]. Cells were resuspended in Basal Medium Eagle and seeded onto a 35-mm culture dish (IWAKI, Tokyo, Japan) coated with poly-l-lysine (Sigma, St. Louis, MO). Each dish contained 5 × 106 cells which were maintained with serum-free Neurobasal (NB) medium in the incubator with 5 % CO2 at 37 °C. Cell morphology was illustrated by confocal microscopy (Olympus FluoView, Olympus America Inc., Center Valley, PA).

Drug Treatment

For concentration-dependent studies, primary cultured cortical neurons were treated with 1–10 μM sodium arsenite (arsenite; Sigma, St. Louis, MO) for 24 h. For time-dependent studies, primary cultured cortical neurons were treated with arsenite (5 μM) for 8–48 h. For arsenic studies, primary cultured cortical neurons were treated with 5 μM arsenite and sodium arsenate (arsenate; Sigma, St. Louis, MO) for 24 h. For glutathione (GSH; Sigma, St. Louis, MO) studies, 10 mM GSH was co-incubated with arsenite for 24 h.

HO-1 siRNA Transfection

The siRNA targeted at HO-1 (catalog no. sc-35555, Santa Cruz Biotechnology, Santa Cruz, CA) was used to knock down HO-1 levels. A nontargeting scramble siRNA (UUCUCCGAACGUGUCACGUTT) was used as a negative control in the siRNA transfection experiments. Primary cultured cortical neurons grown on 6- or 24-well plates were transfected with siRNA (20 nM) for 48 h (DIV4-6) in NB medium. At the end of transfection, cells were washed once with fresh NB medium before further experimentation.

Cell Viability Assay

The cell viability was determined using a modified 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium (MTT; Sigma, St. Louis, MO) assay. In brief, the culture medium was removed and the cells were washed with phosphate-buffered saline (PBS). The cells were then incubated with MTT solution (5 mg/mL in PBS) for 3 h. Afterwards, the MTT solution was removed and the resulting formazan was dissolved with dimethyl sulfoxide (100 μL). The absorption was measured at 570 nm with a reference wavelength of 630 nm.

Western Blot Analysis of Related Proteins

Cells were treated with CelLyticTM MT Mammalian Lysis Reagent (Sigma, St. Louis, MO), 1 % proteinase K inhibitor, 0.1 % Triton X-100, or RIPA buffer which contains 0.5 M NaCl, 50 mM Tris, 1 mM EDTA-Na, 0.05 % sodium dodecyl sulfate (SDS), 0.5 % Triton X-100, and 1 mM phenylmethanesulfonylfluoride. The lysates were centrifuged at 16,500×g, 4 °C for 30 min, and the supernatant was stored at −80 °C.

Protein samples were run on 8–15 % SDS–polyacrylamide gel electrophoresis and then transferred onto a nitrocellulose membrane (Bio-Rad, Hercules, CA) at 90 V for 120 min. Blots were probed with a monoclonal antibody against HO-1 (1:1,000; Stressgen, Victoria, CA), caspase 3 (1:1,000; Cell Signaling Technology, Beverly, MA), and procaspase 12 (1:1,000; Exalpha Biologicals, Shirley, MA) at 4 °C over night. After primary antibody incubation, the membrane was washed and incubated with horseradish peroxidase-conjugated secondary IgG (1:3,000; Millipore Corporation, Billerica, MA) for 1 h at room temperature. The immunoreaction was visualized using Amersham Enhanced Chemiluminescence (Amersham Pharmacia Biotech, Piscataway, NJ). After this detection, the bound primary and secondary antibodies were stripped by incubating the membrane in stripping buffer (100 mM 2-mercaptoethanol, 2 % SDS) at 50 °C for 45 min. The membrane was reprobed with β-actin antibody (1:5,000; Chemicon, Temecula, CA). The densities of blots were analyzed and reported as relative optical density of the specific proteins.

Statistics

All data are expressed as the mean ± S.E.M. Statistical comparisons were made using t test. A value of p <0.05 was considered statistically significant.

Results

Arsenite-Induced Neurotoxicity

To establish the arsenite-induced neurotoxicity, primary cultured cortical neurons were incubated with arsenite (1–10 μM). After 24-h incubation, arsenite caused neuronal death in a concentration-dependent manner (Fig. 1a). Furthermore, arsenite (5 μM) time-dependently induced cell death of primary cultured cortical neurons (Fig. 1b). Arsenite-mediated neurotoxicity coincided with the abundance of cytosolic vacuoles in the treated neurons (Fig. 1c). The apoptotic mechanisms underlying arsenite-induced neurotoxicity were investigated. Arsenite simultaneously reduced procaspase 3 levels and increased active caspase 3 levels, indicating arsenite-induced caspase 3 activation in the primary cultured cortical neurons (Fig. 2a, b). Furthermore, arsenite decreased procaspase 12 levels in concentration- and time-dependent manners, suggesting that endoplasmic reticulum (ER) stress is involved in the arsenite-induced apoptosis (Fig. 2c, d).

Fig. 1
figure 1

Arsenite induced neurotoxicity in primary cultured cortical neurons. a Primary cultured cortical neurons were treated with arsenite (0 to 10 μM) for 24 h. b Primary cultured cortical neurons were treated with arsenite (5 μM) for 0 to 48 h. Cell viability was measured with MTT assay. Values are the mean ± S.E.M. (n = 3). *p < 0.05, statistically significant in the arsenite groups compared with the control groups by t test. c Representative light field confocal micrographs of neurons treated with vehicle or arsenite for 24 h

Full size image
Fig. 2
figure 2

Arsenite induced apoptosis in primary cultured cortical neurons. a, c Primary cultured cortical neurons were treated with arsenite (0 to 10 μM) for 24 h. b, d Primary cultured cortical neurons were treated with arsenite (5 μM) for 0 to 48 h. Western blot assay was employed to measure active caspase 3 (aCasp 3), procaspase 3 (pCasp 3), and procaspase 12 (pCasp 12). Each lane contained 30 μg protein for all experiments. Graphs show statistic results from relative optical density of bands on the blots. Values are the mean ± S.E.M. (n = 3). *p < 0.05, statistically significant in the arsenite groups compared with the control groups by t test

Full size image

Oxidative Stress and Arsenite-Induced Neurotoxicity

The involvement of oxidative stress in arsenite-induced neurotoxicity was investigated by measuring the levels of HO-1, a redox-regulated protein. Western blot assay showed that arsenite concentration- and time-dependently increased HO-1 levels in the treated primary cultured cortical neurons (Fig. 3). Moreover, GSH, a well-known antioxidant, was found to slightly reduce basal HO-1 and active caspase 3 levels (Fig. 4). However, co-incubation with GSH significantly diminished arsenite-induced HO-1 elevation (Fig. 4a) and caspase 3 activation (Fig. 4b), indicating that oxidative stress is involved in the arsenite-induced neurotoxicity in primary cultured cortical neurons.

Fig. 3
figure 3

Arsenite elevated HO-1 expression in primary cultured cortical neurons. a Primary cultured cortical neurons were treated with arsenite (0 to 10 μM) for 24 h. b Primary cultured cortical neurons were treated with arsenite (5 μM) for 0 to 48 h. Western blot assay was employed to measure HO-1 levels. Each lane contained 30 μg protein for all experiments. Graphs show statistic results from relative optical density of bands on the blots. Values are the mean ± S.E.M. (n = 3). *p < 0.05, statistically significant in the arsenite groups compared with the control groups by t test

Full size image
Fig. 4
figure 4

Effect of GSH on arsenite-induced HO-1 elevation and caspase 3 activation in primary cultured cortical neurons. Primary cultured cortical neurons were treated with GSH (10 mM) and arsenite (5 μM) for 24 h. Western blot assay was employed to measure a HO-1 and b active caspase 3 (aCasp 3) as well as procaspase 3 (pCasp 3). Each lane contained 30 μg protein for all experiments. Graphs show statistic results from relative optical density of bands on the blots. Values are the mean ± S.E.M. (n = 3). *p < 0.05, statistically significant in the groups compared with the control groups; #p < 0.05, statistically significant in the arsenite plus GSH groups compared with the arsenite groups by t test

Full size image

The cytotoxic effects of inorganic arsenics, including arsenite and arsenate, were investigated. After 24-h incubation of arsenics (5 μM), arsenite was more potent than arsenate in inducing HO-1 expression (Fig. 5a) and caspase 3 activation (Fig. 5b). Furthermore, the cell viabilities by arsenite and arsenate were 60 ± 2 and 99 ± 2 % of control, respectively (n = 3), indicating that arsenite may be more oxidative and neurotoxic than arsenate.

Fig. 5
figure 5

The neurotoxic effects of inorganic arsenics in primary cultured cortical neurons. Primary cultured cortical neurons were treated with arsenite (5 μM) and arsenate (5 μM) for 24 h. Western blot assay was employed to measure a HO-1 levels and b active caspase 3 (aCasp 3) as well as procaspase 3 (pCasp 3). Each lane contained 30 μg protein for all experiments. Graphs show statistic results from relative optical density of bands on the blots. Values are the mean ± S.E.M. (n = 3–5). *p < 0.05, statistically significant in the arsenic groups compared with the control groups; #p < 0.05, statistically significant in the arsenate groups compared with the arsenite groups by t test

Full size image

Effect of HO-1 siRNA on Arsenite-Induced Neurotoxicity

The role of HO-1 in arsenite-induced neurotoxicity was investigated using HO-1 siRNA transfection. In contrast to the scramble siRNA transfection which had no effect on arsenite-induced HO-1 elevation, transfection with HO-1 siRNA significantly reduced arsenite-elevated HO-1 levels (Fig. 6a). At the same time, HO-1 siRNA transfection enhanced arsenite-induced apoptosis by augmenting arsenite-induced reduction in procaspase 3 and elevation in active caspase 3 (Fig. 6b). Furthermore, transfection with HO-1 siRNA potentiated arsenite-induced neuronal loss (Fig. 6c). The morphological study demonstrated that arsenite-induced neurotoxicity was intensified in the HO-1 siRNA-transfected cells with an appearance of ruptured cell bodies and discontinuous neurites (Fig. 6d). These data suggest that HO-1 plays a neuroprotective role in the arsenite-induced neurotoxicity in primary cultured cortical neurons.

Fig. 6
figure 6

Effect of HO-1 siRNA transfection on arsenite-induced neurotoxicity in primary cultured cortical neurons. Primary cultured cortical neurons were pretreated with HO-1 siRNAs and scramble siRNA (scr) for 48 h and then incubated with arsenite (5 μM) for 24 h. Representative Western blot data showed the effect of HO-1 siRNA transfection on a HO-1 levels and b caspase 3 activation. Each lane contained 30 μg protein for all experiments. Results were repeated in independent experiments. c Cell viability was measured using MTT assay. Values are mean ± S.E.M. (n = 3). *p < 0.05, statistically significant in the arsenite groups compared with the control groups; #p < 0.05, statistically significant in the arsenite plus HO-1 siRNA groups compared with the arsenite groups. d Representative light field confocal micrographs show the arsenite-induced morphological changes of primary cultured cortical neurons transfected with HO-1 siRNA. Results were repeated in independent experiments

Full size image

Discussion

Apoptosis Is Involved in Arsenite-Induced Neurotoxicity

Clinical reports have shown that the plasma levels of arsenics in patients treated with arsenic trioxide ranged from 5.5 to 7.3 μM [7]. In the present study, concentrations of arsenics (3–10 μM) were at clinically relevant ranges. In addition to the arsenite-induced cell loss of primary cultured cortical neurons, our morphological data showed that arsenite induced significant cellular damages, including vacuole formation and neurite fragmentation. Consistent with our previous study that both necrosis and apoptosis mediate the arsenite-induced neurotoxicity in nigrostriatal dopaminergic system of rat brain [17, 18], the involvement of apoptosis in arsenite-induced neurotoxicity was evident by caspase 3 activation in primary cultured cortical neurons. Moreover, arsenite decreased procaspase12 levels, an ER-specific enzyme [30], supporting the notion that ER stress contributes to the arsenite-induced apoptosis [17–19, 23].

Oxidative Stress and Arsenite-Induced Neurotoxicity

HO-1 has been used as a biological hallmark of cells in response to insults because HO-1 is reportedly induced to cope with the unfavorable environments, including oxidative stress [20, 21]. Similar to our previous studies in rat brain [17, 18] and DRG explants [19, 23], arsenite with oxidative property increased HO-1 levels in the primary cultured cortical neurons. Antioxidative treatment has been found to attenuate arsenite-induced HO-1 levels as well as neurotoxicity in rat brain [17, 18] and DRG explants [23]. In the present study, we used GSH to mitigate arsenite-induced oxidative stress and found that arsenite-induced HO-1 elevation, caspase 3 activation, and cell death were all attenuated in the GSH-treated cells. To further study the relationship between oxidative stress and HO-1 induction, we compared the effects of two inorganic arsenics, arsenite and arsenate, because arsenite is known to be more oxidative than arsenate [11, 22, 23, 31, 32]. Consistent with our previous studies in DRG explants [23], the present study showed that arsenite was more potent than arsenate in inducing HO-1 expression, caspase 3 activation, and cell death in the primary cultured cortical neurons. These data indicate that oxidative stress, HO-1 induction, and neurotoxicity are part of the pathophysiology of arsenite-induced neurotoxicity.

HO-1 Is Neuroprotective in Arsenite-Induced Neurotoxicity

In response to oxidative stress, HO-1 induction has been proposed to establish a normal redox balance in cells [19, 20]. Our previous studies and the present study have observed that arsenite simultaneously induced HO-1 elevation and apoptosis. The concomitant existence of HO-1 elevation and apoptosis has obscured the role of HO-1 in arsenite-induced neurotoxicity in arsenite-infused SN [17, 18], arsenite-treated DRG explants [23], and primary cultured cortical neurons. Many studies have employed HO-1 overexpression which confers an enhanced neuroprotection [26, 33]. In contrast, we employed HO-1 siRNA transfection and found that HO-1 silencing significantly potentiated arsenite-induced apoptosis and neuronal loss, further supporting a neuroprotective role of HO-1 in the CNS neurodegenerative diseases. The mechanisms for arsenite-induced HO-1 elevation are proposed. To combat the oxidative injury, it is possible that a redox signaling pathway involving the transcription factor nuclear factor E2-related factor 2–antioxidant response element is activated and induces numerous protective mechanisms, including HO-1, heat shock protein 70, antioxidative enzymes, and thioredoxin [27, 34]. In addition to the antioxidative mechanism, HO-1 has been proposed to prevent protein aggregation, including aggregation of α-synuclein [35], a pathological protein in CNS neurodegenerative diseases [36, 37].

In conclusion, arsenite appears to be more potent than arsenate in inducing HO-1 expression and neurotoxicity. Furthermore, HO-1 plays a neuroprotective role in arsenite-induced neurotoxicity in primary cultured cortical neurons. In addition to antioxidants, HO-1 induction may be a therapeutic strategy for arsenite-induced neurotoxicity.