Experimental animal
Adult zebrafish (4-6 months old, 2.5-3.5 cm long, 350±50mg of weight, wild type, Danio rerio) were used in the study. Animals were fed brine shrimp twice a day in a 28.5 °C tank, and were on a 14-hour daytime and 10-hour night cycle. The watercraft was equipped with a multi-stage filtration system with a sediment filter, post-carbon filter, fluorescent UV light, and sterilization filter (Zebrafish AutoSystem, Genomic Design, Daejeon, Korea). All procedures were reviewed and approved by the Committee on the Ethics of Animal Experiments of the Korea University Medical School (IACUC number KOREA-2018-0032). After behavior experiments, all zebrafish were anesthetized using MS-222 (tricaine, Sigma-Aldrich, St. Louis, Mo., USA) and euthanatized by decapitation.
Drug administration
MgT (C8H14MgO10, molecular weight 294.50 g/mol, Doctor’s Best Inc., USA) and MgSO4 (anhydrous, molecular weight 120.37 g/mol, Biosesang Inc., Korea) were used in the study. MgT were centrifuged at 2,000 revolutions per minute for two minutes in 30 mL of phosphate-buffered saline (PBS) to a concentration of 100 mM and a sample from the middle layer was diluted 10 times (10 mM) or 100 times (1 mM). MgSO4 was also centrifuged and diluted in PBS in the same way. Drug concentrations were determined based on physiological concentrations of magnesium [22, 23] and results of MgT toxicity experiments in zebrafish embryos described below.
Evaluation of toxicity of MgT in zebrafish embryos
Zebrafish embryos were treated with 0 (control), 5, 25, 50, and 100 mM MgT (diluted in water) in a 6-well plate (ten embryos per well) and incubated at 26°C. Two days later, heart rate and development were measured with a microscope. After 6 days, survival rate was evaluated.
Hypoxic chamber
The hypoxic chamber consisted of a closed glass box attached to an inner lid with a pack of GaspakTM (Becton, Dickinson and Company, USA) (Figure 1) filled with water such that the GaspakTM was not immersed. The hypoxic chamber was closed the night prior to the experiment so as to reach at 1.0 ± 0.5 mg/L of dissolved oxygen (DO) as measured by a portable DO analyzer.
Effects of MgT and MgSO4 on neurons
The human neuroblastoma cell line SH-SY5Y (ATCC® CRL-2266™, American Type Culture Collection, USA) were used for in vitro experiments. After ensuring cells were healthy and free of contamination, and removing culture media, cells were treated with fetal bovine serum-free Dulbecco's Modified Eagle Medium, and starved for two hours. Prewarmed 1x trypsin solution was added to promote cell detachment in appropriate quantities (0.5 mL/10 cm2). When more than 90% of cells were detached, prewarmed complete growth media was added to inactivate the trypsin. After cell counting, 100 μL of cell suspension was added to a 96-well plate to obtain a final concentration of 7,000 cells per well. The cells were treated with PBS (control), MgSO4 (1 or 10 mM) or MgT (1 or 10 mM) for one hour. Media was changed, and the cells were incubated for two hours in a hypoxic chamber with no water (or a regular CO2 incubator as a control). To evaluate cell viability, 10 μL of cell counting kit-8 solution (CCK-8 solution, Dojindo Laboratories, Japan) was added to each well and absorbance was measured with a microplate reader (450 nm). To ensure reliability of the experimental results, experiments were repeated three times (each with 6 technical replicates per group).
Hypoxic zebrafish model
Zebrafish exposed to hypoxia were evaluated using a 4-stage behavioral repertoire as previously described [24]. Briefly, fish were classified as stage 1 (swimming on the surface of the water), stage 2 (failure to attain normal posture), stage 3 (intermittent maintenance of opercular beats with brief movements), and stage 4 (dead). When a hypoxic zebrafish reached stage 3 it was immediately transferred to a normoxic chamber (DO: 7.0 ± 0.5 mg/L).
Oral administration of PBS and MgT
Oral administration was performed by a professional after tricaine anesthesia [25]. Zebrafish were anesthetized in a water mixture containing 16.8 mg of tricaine per 100 mL, resuspended to minimize the anesthetic time, and administered 1 μL orally of PBS or MgT (10 mM) with a micropipette, taking care not to damage the mouth. This volume was chosen based on pilot data showing that zebrafish were likely to vomit solutions administered at volumes greater than 2 μL.
Classification of the experimental group
The experimental group was randomly divided into four groups: PBS= zebrafish that were orally administered 1 μL of PBS and remained in normoxia for three hours; PBS + HYP= zebrafish that were orally administered 1 μL of PBS and placed in the hypoxic chamber for one hour followed by normoxia for two hours; MgT= zebrafish that were orally administered 10 mM of MgT and were in normoxia for three hours; and the MgT + HYP group = zebrafish that were orally administered 10 mM of MgT and placed in the hypoxic chamber for one hour followed by normoxia for two hours (Figure 2A). Randomization was performed using a web-based computer-generated list (www.randomization.com). The numbers were kept in opaque, sealed envelopes that were opened in the laboratory just before the experiment.
Sample size calculation
Power analysis revealed that a minimum sample size of 10 for each group would be required to achieve a power of 80% at a significance level of 5%. Power analysis was calculated from results of our prior experiment [26]. To allow for exclusions, total sample size was prospectively set at 48 zebrafish (12 for each group).
Color-added T-maze test
A T-shaped maze containing two arms and one stem was used to measure learning using color preference or compensation. A starting box (10 × 10 × 10 cm) was located at the bottom of the stem (50 × 10 × 10 cm) and two compartments (the “target” and “opposite” compartment; 10 × 10 × 10 cm each) were located at the ends of the maze arms (each 20 × 10 × 10 cm). Transparent sliding doors were used to separate the start box and the arms of the maze from the stem. During the training periods, sleeves made of red or yellow cellophane were fitted around the target and opposite compartments, respectively (Figure 2B). To minimize bias, all the experiments started at 1-2 pm in a quiet dedicated place in the laboratory and all zebrafish were subjected to a habituation trial for two hours prior to testing. Each zebrafish was trained once a day for four consecutive days. 20 µL of food (brine shrimp) was placed in the red cellophane compartment and each zebrafish was placed in the start box. On the fifth day, all zebrafish underwent memory testing, wherein all cellophane and food rewards were removed from the maze. All processes were recorded with a 4K camcorder (Sony FDR-AX33, Sony Corporation, Japan) and analyzed using EthoVision XT software (Noldus Information Technology, Netherlands).
Evaluation of MgT effects on zebrafish behavior
The evaluation indices for zebrafish behavior were as follows: time spent= total time spent in one of the compartments of the T-maze; distance moved= total horizontal distance moved in a compartment; and frequency of entries= the number of times the zebrafish entered one of the compartments.
Compartment preference was calculated using the following equation: preference=log(target compartment/opposite compartment). Therefore, 0 indicated no preference, and 1 indicated an index value in the target compartment 10 times higher than that of the opposite compartment. Conversely, a negative preference value indicated that the opposite compartment was preferred.
2,3,5-triphenyltetrazolium chloride (TTC) staining
TTC staining is a widely used method to measure hypoxic brain damage [27] and evaluate activity of brain mitochondrial dehydrogenase. To remove the brain, zebrafish were anesthetized using MS-222 (tricaine) and euthanatized by decapitation upon completion of behavior testing. For gross infarct size assessment, the brain was incubated in 1 mL of PBS containing 2% TTC (Sigma-Aldrich) for 40 minutes, followed by overnight incubation in 4% paraformaldehyde. The following day, the brain was imaged with a microscope. For absorbance measurement, the brain was incubated in TTC solution in a CO2 incubator for 100 minutes. The TTC solution was discarded and the brain was gently rinsed with 2-3 drops of dimethyl sulfoxide (DMSO)/ethanol (1:1 solution) and stored overnight in a 1.5 mL tube containing 1 mL of DMSO/ethanol solution. The following day, the absorbance of the DMSO/ethanol solution was measured with a spectrophotometer (Epoch, BioTek instruments, USA) and corrected by the brain weight.
Western blot
To determine the mechanism of MgT in neuroprotection during hypoxia, we measured the expression of a glutamate transporter, excitatory amino acid transporter (EAAT) 4, by western blot. Following behavioral testing, zebrafish brains were homogenized in a lysis buffer (radio-immunoprecipitation buffer; Sigma) containing a protease inhibitor cocktail (Roche). Protein concentration was determined using the Bradford method. Proteins (20 μg) were separated by 8% sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred to membranes. The membranes were blocked with 5% skim milk in 1× tris-buffered saline (TBS) at room temperature for one hour. The membranes were then incubated with a rabbit anti-EAAT4 antibody (ab41650, Abcam, Cambridge, MA) or rabbit anti-beta actin (A5441, Sigma-Aldrich) overnight at 4°C. Membranes were washed three times in 1× TBS + 0.05% Tween 20 and incubated with a 1:2,000 (EAAT4) or 1:5,000 (beta actin) dilution of horseradish peroxidase (HRP)-conjugated anti-rabbit immunoglobulin G (IgG) secondary antibody. An electro-chemiluminescence kit was used to develop the western blots (Amersham, Boston, MA, USA). Quantitative analysis of densitometry was performed using ImageJ (v. 1.52a, National Institutes of Health, USA).
Statistical analysis
All data were analyzed using SPSS 22 (IBM, USA) and GraphPad Prism 6.0 (GraphPad, USA). Data are expressed as mean with standard error of the mean. Data were tested for normality using the Kolmogorov–Smirnov test. Parametric or non-parametric analysis was performed as appropriate. For MgT toxicity and survival data, a Kaplan-Meier survival curve was evaluated using the log-rank (Mantel-Cox) test. Cell viability was compared using one-way analyses of variance, and Tukey's multiple comparison tests were used for post-hoc analysis among the treated and untreated hypoxia groups. T-maze results were compared using a paired T-test or Wilcoxon matched-pair signed rank test to determine differences in compartment preference. P-value < 0.05 was considered significant.