NBQX

Long-term neurotoxic effects of domoic acid on primary dopaminergic neurons

A B S T R A C T
Domoic acid, an excitatory neurotoxin produced by certain algae, reaches the food chain through accumulation in some sea organisms. To investigate its long-term neurotoxicity on dopaminergic neurons, prepared primary mesencephalic cell cultures were exposed to different concentrations of domoic acid (0.1, 1, 10, 100 μM) on the 8th day in vitro (DIV) for 4 days. On the 12th DIV, culture media were collected for measurement of lactate dehydrogenase and cultured cells were subjected to immunohistochemistry against tyrosine hydroxylase, neu- ronal nuclear antigen and glial fibrillary acidic protein, and fluorescence staining using H2DCFDA, JC-1 and Hoechst 33342 dyes. Moreover, roles of AMPA/KA and NMDA receptors in domoic acid neurotoxicity were also investigated. Domoic acid significantly decreased the number of dopaminergic neurons and adversely affected their morphology, and slightly reduced the expression of neuronal nuclear antigen and glial fibrillary acidic protein. Co-treatment of cultures with domoic acid and the AMPA/KA or NMDA receptor antagonists NBQX and MK-801 rescued significant number of dopaminergic neurons. Domoic acid significantly decreased red:green fluorescence ratio of JC-1 and did not affect production of reactive oxygen species and apoptotic cell death. In conclusions, the present study reveals that long-term treatment of primary mesencephalic cell culture with domoic acid significantly destroyed dopaminergic neurons. This effect appears to be attributed to activation of AMPA/KA and NMDA receptors and mitochondrial damage.

1.Introduction
Parkinson’s disease (PD) is a progressive neurodegenerative disorder affecting about 2–3% of people older than 65 years. The neuropatho- logical hallmarks of the disease are the loss of dopaminergic neurons in substantia nigra (SN), which leads to striatal dopaminergic deficiency, and formation of intracellular inclusions containing α-synuclein aggregates in the surviving neurons (Poewe et al., 2017). As a movement disorder, PD disease is characterized by a group of cardinal signs, most notably resting tremors, rigidity and bradykinesia (Cacabelos, 2017). Moreover, PD is associated with a group of non-motor symptoms as the pathological process affects multiple other cell types in the central and peripheral nervous system (Poewe et al., 2017). Although the exact etiology of PD is still unknown, numerous epidemiological studies link the higher PD risk to some environmental factors including consump- tion of well water, living in rural areas, and exposure to herbicides and pesticides (Ritz et al., 2016; Naughton et al., 2017; Rokad et al., 2017). In addition to the interaction with the genetic agents, environmental factors play an important role in mitochondrial dysfunction, oxidative stress and modifying proteasomal function (Schapira, 2006).

Domoic acid (DomA) is a potent marine neurotoxin that naturally produced by some marine organisms such as the red alga Chondria ar- mata and planktonic diatom of the genus Pseudonitzchia (Hiolski et al., 2016). When DomA is produced at high concentrations in diatoms, it can bioaccumulate in some marine organisms such as shellfish, sardines and anchovies, causing the poisoning of seabirds, marine mammals or humans (Lefebvre and Robertson, 2010). In 1987, DomA caused four mortalities and illness of > 100 people after consuming the blue mus- sels Mytilus edulis in the Cardigan Bay of Prince Edward Island, Canada (Perl et al., 1990). Symptoms in ill people comprised three kinds of signs: (1) gastrointestinal (nausea, vomiting, abdominal cramps and diarrhea), (2) cardiovascular (arrhythmias and unstable blood pressure) and (3) neurological signs (disorientation, headache, hallucination, seizures, memory impairment and coma). Impairment of memory was the reason to denominate this condition as amnesic shellfish poisoning (ASP) (Addison and Stewart, 1989; Wright et al., 1989). Since that time, specific control measures have been implemented in Canada and all over the world to prevent food-borne illness due to DomA (Park, 1995; Toyofuku, 2006). However these measures have been successful in preventing other episodes of ASP, some DomA intoxication have been reported in wild animals including sea lions, whales, sea otters and sea birds (Sierra et al., 1997; Lefebvre et al., 1999; Gulland et al., 2002; Lefebvre et al., 2002; Goldstein et al., 2008).Neurotoxicity of DomA was reported in some in vitro and in vivo experimental models. For instance, Giordano et al. (2008) and Ramya et al. (2017) found that DomA significantly produced apoptotic cell death in cerebellar granule neurons (CGNs) and Caco-2 cell line. Strain and Tasker (1991) reported that intraperitoneal injection of DomA produced a series of behavioral changes in mice including sedation, rigidity, stereotypy, balance loss and convulsions. Pathologically, the authors observed that DomA resulted in hippocampal neuronal damage particularly in the CA3 region (Strain and Tasker, 1991).
As to date there have been no studies showed the neurotoxic effect of DomA on dopaminergic neurons, the current study was conducted to investigate the long-term neurotoxicity of DomA on dopaminergic neurons in primary mesencephalic cell culture and whether it could be implicated in PD.

2.Materials and methods
2.1.Preparation of primary mesencephalic cell cultures
Pregnant mice (OF1/SPF, Himberg, Austria) were cared and han- dled in accordance with the guidelines of the European Union Council (86/609/EU) for the use of laboratory animals. At gestation day 14, primary mesencephalic cell cultures were prepared from embryonic mesencephala as described previously by Radad et al. (2015). The embryos were collected at gestation day 14 under aseptic condition in Dulbecco’s phosphate buffered saline (DPBS, Invitrogen, Germany). Under a stereoscope, the brains were released and the mesencephala were excised, carefully cleaned from meninges, and enzymatically and mechanically dissociated using 0.2% trypsin solution (Invitrogen, Ger- many) and fire-polished Pasteur pipettes, respectively. Then, obtained cells were grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10 mM HEPES buffer, 4 mM glutamine and 10% heat-inactivated fetal calf serum (FCS). The medium was exchanged with DMEM supplemented with FCS on the 1st and 3rd DIV. On the 5th cell cultures by immunostaining against TH. Cultured cells were fixed with histochoice for 15 min at room temperature. Then, they were permeabilized with 0.4% Triton X-100 for 30 min at room temperature. After blocking of non-specific binding sites with 5% horse serum (Vectastain ABC Elite kite) for 90 min at room temperature, cells were incubated with anti-TH primary antibody overnight at 4°C, biotinylated secondary antibody (Vectastain) and avidin-biotin-horseradish perox- idase complex (Vectastain) for 90 min at room temperature. The reac- tion product was developed in a solution of diaminobenzidine (1.4 mM) in PBS containing 3.3 mM hydrogen peroxide (H O ). Dopaminergic taining 0.02 ml B-27/ml (Invitrogen, Germany). Serum-free DMEM supplemented with 0.02 ml B-27/ml was used for feeding of cultured cells from the 6th DIV and subsequently replaced every 2nd day.

2.2.Treatment of cultures with DomA
DomA (Sigma, Germany) was prepared in a stock solution of 1 mM in distilled water. Final concentrations of DomA were prepared in DMEM. On the 8th DIV, cultures were treated with different con- centrations of DomA (0.1, 1, 10, 100 μM) for 4 days. Culture media were changed with the same concentrations of DomA on the 10th DIV.On the 12th DIV, culture media were obtained for measurement of lactate dehydrogenase (LDH) and cultured cells were subjected to im- munohistochemistry against tyrosine hydroxylase (TH), neuronal nu- clear antigen (NeuN) and glial fibrillary acidic protein (GFAP), and fluorescence staining using H2DCFDA, JC-1 and Hoechst 33342 dyes.

2.3.Identification of dopaminergic neurons
Dopaminergic neurons were visualized in primary mesencephalic neurons were counted with a Nikon inverted microscope in 10 ran- domly selected fields/well at 10× magnification.

2.4. Detection the effects of DomA on total neuronal cells and astrocytes
Two sets of cultures were treated with DomA (10 and 100 μM) on the 8th DIV for 4 days. Culture media were changed with the same concentrations of DomA on the 10th DIV. On the 12th DIV, cultures were stained immunocytochemically using anti-NeuN and anti-GFAP antibodies (Chemicon, USA) for visualizing postmitotic neuronal cell types and astrocytes, respectively. In which, the same staining proce- dures were carried out as described for anti-TH immunostaining except that the anti-TH antibody was replaced with the anti-NeuN or anti- GFAP antibodies. NeuN immunoreactivity was quantitatively assessed in 5 photographs/treatment condition using Adobe Photoshop soft- ware®. Briefly, the original photographs were converted into the black- white by the threshold tool. Then, the black color that corresponds to the number of NeuN+ neurons were measured and expressed as the percentage of the averaged control. Astrocytes were counted with a Nikon inverted microscope in 10 randomly selected fields/well at 40× magnification.

2.5.Measurement of LDH in the culture medium
LDH is usually used to quantitatively measure cell damage. At the end of each treatment, culture media were collected and used to mea- sure LDH with the cytotoxic detection kit according to the manufac- turer’s instructions. In brief, NADH + H+ produced from NAD+ by LDH is transferred by diaphorase to the yellow tetrazolium salt 2-[4- iodophenyl]-3-[4-nitrophenyl]-5-phenyltetrazolium chloride (INT) re- sulting in red formazan formation. The latter was measured spectro- photometrically at 490 nm with a reference at 688 nm. Supplemented medium was used as a blank and subtracted as background.

2.6.Treatment of cultures with NBQX and MK-801
To investigate the role of AMPA/KA and NMDA receptors in DomA neurotoxicity, two sets of cultures were treated first with 10 μM NBQX, an AMPA/KA receptors antagonist and 10 μM MK-801, a NMDA re- ceptor antagonist, respectively, on the 8th DIV. After 10 min, cultures were treated with DomA (10 and 100 μM) for 48 h. The same treatment was repeated on the 10th DIV and the cultures were kept till the 12th DIV. Then, cultures were stained using anti-TH antibody and the number of dopaminergic neurons was determined.

2.7.Measurement of mitochondrial membrane potential (Δψm) of cultured cells by JC-1 fluorescence dye
JC-1 is a cationic dye that can selectively accumulate in mi- tochondria by electrochemical gradient, indicated by a fluorescence emission shift from red (~590 nm) to green (~529 nm). Prepared cul- tures in black 96-well plates were treated with DomA (10 and 100 μM) on the 8th DIV for 48 h. Culture media were changed with the same concentrations of DomA on the 10th DIV. At the end of incubation periods, culture media was removed and cells were loaded with JC-1 (5 μg/ml in DMEM) for 15 min at 37 °C. Mitochondrial depolarization is allocated by a decrease in the red/green fluorescence intensity ratio. Semiquantitation of fluorescence intensity was done by using a 2300 Enspire Multilabel Reader. Data were transferred to excel software and presented as means ± SEM in which the control was set to 100%.

2.8.Measurement of reactive oxygen species (ROS) with H2DCFDA
H2DCFDA is a general oxidative stress indicator in cell cultures. A 50 mM stock of H2DCFDA in dimethyl sulfoxide (DMSO) was diluted in colourless DMEM to a final concentration of 10 μM. Prepared cultures in black 96-well plates were treated with DomA (10 and 100 μM) on the 8th DIV for 48 h. Culture media were changed with the same con- centrations of DomA on the 10th DIV. On the 12th DIV, culture media were removed and cells were incubated with 10 μM H2DCFDA at 37 °C for 30 min. Cultured cells were then washed twice with colourless DMEM and kept with DPBS (100 μl/well). Semiquantitation of the fluorescence intensity was done by using a 2300 Enspire multilabel reader (Perkin Elmer) at an emission wavelength of 520 nm. Data were transferred to excel software and presented as means ± SEM in which the control was set to 100%.

2.9.Detection of apoptotic cell death by Hoechst 33342 stain
Hoechst 33342 is a fluorescent dye that intercalates into DNA of living and degenerating cells with an emission at 497 nm. Primary mesencephalic cell cultures were prepared in black 96 well plates and treated with DomA (10 and 100 μM) on the 8th DIV for 4 days. Culture media were changed with the same concentrations of DomA on the 10th DIV. On the 12th DIV, culture media were removed and cells were in- cubated with a Hoechst 33,345 solution (8 μM in colourless DMEM) for 10 min at 37 °C. After rinsing with PBS, Semiquantitation of fluores- cence intensity was obtained by using a 2300 Enspire multilabel reader (Perkin Elmer). Data were transferred to excel software and presented as means ± SEM in which the control was set to 100%.

2.10.Statistics
Data for each treatment parameter were obtained from three ex- perimental repeats (4-well plate each) and presented as mean ± standard error of mean (SEM). Comparisons were made using ANOVA and post-hoc Duncan’s test using IBM SPSS statistics 22. p < 0.05 was considered as statistically significant. 3.Results 3.1.Effect of DomA on dopaminergic neurons Treatment of primary mesencephalic cell cultures with different concentrations of DomA (0.1, 1, 10, 100 μM) on the 8th DIV for 4 days significantly decreased the number of dopaminergic neurons by 22% and 45% (Fig. 1A) and the number of dopaminergic neurites by 33% and 44% at the concentrations 10 and 100 μM, respectively, compared to untreated control cultures (Fig. 1B). Morphologically, control cultures showed intact dopaminergic neurons with long and branched neurites (Fig. 2A, B). On the other hand, DomA-treated cultures had dysmorphic dopaminergic neurons with fewer and shortened neurites compared to untreated control cultures (Fig. 2C, D, E, F). 3.2.Effect of DomA on the total neuronal cells and astrocytesIncubation of primary mesencephalic cell cultures with DomA (10, 100 μM) on the 8th DIV for 4 days had no marked effect on NeuN im- munoreactivity (Fig. 3A) and insignificantly decreased the number of astrocytes (Fig. 3B) compared to untreated controls. Fig. 4 showed that DomA resulted in a very little effect on expression of NeuN marker and slightly decreased the number of astrocytes. 3.3.Effect of DomA acid on LDH release into the culture media In consistent to the dopaminergic cell loss, treatment of primary mesencephalic cell cultures with DomA (0.1, 1, 10, 100 μM) on the 8th DIV for 4 days did not affect LDH release into the culture media com- pared to untreated control cultures (Fig. 5). 3.4.Effects of NBQX and MK-801 on DomA-induced neurotoxicity on dopaminergic neurons Co-treatment of primary mesencephalic cell cultures with AMPA/ KA receptor antagonist NBQX and DomA on the 8th DIV for 4 days completely blocked DomA neurotoxicity at the concentration 10 μM and significantly rescued dopaminergic neurons by 22% at the con- centration 100 μM compared to the cultures treated with DomA alone (Fig. 6). NMDA receptor antagonist MK-801 only produced significant protection of dopaminergic neurons by 23% at 10 μM and did not affect dopaminergic neurons survival at 100 μM compared to the cultures treated with DomA alone (Fig. 6). 3.5.Effect of DomA on Δψm, ROS production and apoptotic cell death Incubation of primary mesencephalic cell cultures with DomA on the 8th DIV for 4 days significantly decreased red:green fluorescence ratio of JC-1 by 15% at the concentration 100 μM (Fig. 7A) and on the other hand, did not affect ROS production (Fig. 7B) and apoptotic cell death (Fig. 7C) as evaluated by JC-1, H2DCFDA and Hoechst 33342 fluorescence stains, respectively, compared to untreated control cul- tures. 4.Discussion The present study adopted primary mesencephalic cell culture, the most reliable PD cellular model, to show the long-term neurotoxic effects of DomA on dopaminergic neurons and whether it could be implicated in PD. In which, treatment of cultures with DomA on the 8th DIV for 4 days resulted in significant loss of dopaminergic neurons and decreasing their neurites at the concentrations 10 and 100 μM com- pared to untreated control cultures. Also, it slightly decreased the expression of NeuN and GFAP proteins at the concentration 100 μM compared to untreated control cultures. Similar neurotoxic effect of DomA was reported in some cellular and animal models. For example, Xu et al. (2008) and Giordano et al. (2008) found that DomA killed motor neuron-like cells and mouse cerebellar granule neurons, respec- tively. Schwarz et al. (2014) showed that low doses of DomA affected short and long-term spontaneous behavior of rats. Lefebvre et al. (2017) reported that exposure to low level of DomA caused impairment of spatial learning and hyperactivity in mice. DomA was shown to have no effect on LDH release into culture media indicating selective damage to dopaminergic neurons and absence of overall neurotoxicity in DomA- treated primary mesencephalic cell culture. In contrast, Berman and Murray (1997) found that DomA significantly increased LDH release into the culture media in primary mouse cerebellar granule neurons. This was attributed to the majority of cultured cerebellar granules cells were affected by DomA (Berman and Murray, 1997). As an excitatory amino acid analog to KA, DomA was seen to induce its neurotoxic effect through activation of AMPA/KA receptors in some other cellular models (Larm et al., 1997; Giordano et al., 2007; Gill et al., 2008). Likewise, concomitant treatment of our cultures with DomA and AMPA/KA receptor antagonist NBQX completely blocked DomA neurotoxicity at the concentration 10 μM and rescued significant number of dopaminergic neurons at the concentration 100 compared to the cultures treated with DomA alone. On the other hand, NMDA re- ceptor antagonist MK-801 saved only significant number of dopami- nergic neurons at 10 μM compared to untreated control cultures. In parallel, Hogberg and Bal-Price (2011) reported that DomA effects were completely prevented by the AMPA/KA antagonist NBQX and partly prevented by the NMDA receptor antagonist (APV) in primary rat cer- ebellar cultures. Taken all together, data indicates that DomA had higher affinity to AMPA/KA receptors in dopaminergic neurons and this effect is an important factor in DomA-mediated excitotoxicity among dopaminergic neurons in primary mesencephalic cell culture. Activa- tion of the glutamate receptors subtypes by DomA was reported to in- crease intracellular calcium levels which subsequently increasing of ROS and activation of phospholipases, protein kinase C, proteases, caspases and nitric oxide synthase (Berman et al., 2002; Costa et al., 2010). Staining of cultured cells by JC-1, H2DCFDA and Hoechst 33342 respectively revealed that DomA decreased Δψm and in contrast to the findings of Berman et al. (2002) and Costa et al. (2010), it did not affect ROS production and induction of apoptotic cell death. Decreasing Δψm indicates mitochondrial damage, the effect which might mediate do- paminergic cell death. Likewise, Vranyac-Tramoundanas et al. (2008) reported that DomA can pass cardiac cell membranes and elicit a harmful effect on mitochondrial energetics. Also, Hiolski et al. (2016) found that DomA impaired mitochondrial function in zebrafish brains. Vieira et al. (2016) showed mitochondrial conformation in cardiac muscles following administration of DomA in rats. How mitochondrial impairment by DomA leads to dopaminergic cell damage needs to be further investigated. 5.Conclusion The present study show, for the first time, that the long-term treatment of primary mesencephalic cell culture with DomA sig- nificantly destroyed dopaminergic neurons and adversely affects their morphology. This effect appears to be attributed to activation of AMPA/ KA and NMDA receptors and NBQX mitochondrial damage.