Protective capacity of carotenoid trans-astaxanthin in rotenone-induced toxicity in Drosophila melanogaster

Trans-astaxanthin (TA), a keto-carotenoid found in aquatic invertebrates, possesses anti-oxidative and anti-inflammatory activities. Rotenone is used to induce oxidative stress-mediated Parkinson’s disease (PD) in animals. We probed if TA would protect against rotenone-induced toxicity in Drosophila melanogaster. Trans-astaxanthin (0, 0.1, 0.5, 1.0, 2.5, 10, and 20 mg/10 g diet) and rotenone (0, 250 and 500 μM) were separately orally exposed to flies in the diet to evaluate longevity and survival rates, respectively. Consequently, we evaluated the ameliorative actions of TA (1.0 mg/10 g diet) on rotenone (500 μM)-induced toxicity in Drosophila after 7 days’ exposure. Additionally, we performed molecular docking of TA against selected pro-inflammatory protein targets. We observed that TA (0.5 and 1.0 mg/10 g diet) increased the lifespan of D. melanogaster by 36.36%. Moreover, TA (1.0 mg/10 g diet) ameliorated rotenone-mediated inhibition of Catalase, Glutathione-S-transferase and Acetylcholinesterase activities, and depletion of Total Thiols and Non-Protein Thiols contents. Trans-astaxanthin prevented behavioural dysfunction and accumulation of Hydrogen Peroxide, Malondialdehyde, Protein Carbonyls and Nitric Oxide in D. melanogaster (p < 0.05). Trans-astaxanthin showed higher docking scores against the pro-inflammatory protein targets evaluated than the standard inhibitors. Conclusively, the structural features of TA might have contributed to its protective actions against rotenone-induced toxicity.


Results
Effects of exposure to graded doses of TA on longevity, oxidative stress status and acetylcholinesterase activity in D. melanogaster. The TA concentrations of 0.5 and 1.0 mg/10 g diet both increased longevity by 36.36% compared to the control (Fig. 1A). Seven days of exposure of flies to TA indicated no adverse effects on NPSHs (Fig. 1B), T-SHs (Fig. 1C), H 2 O 2 (Fig. 1D), GST (Fig. 1E) and AChE (Fig. 1F). In addition, TA significantly boost the level of NPSHs compared with the control (p < 0.05). Considering the fact that 0.5 and 1.0 mg/10 g diet concentrations of TA induced the highest effect on the lifespan of flies than the other concentrations, we selected 1.0 mg/10 g of diet concentration to investigate its protective role in rotenoneinduced toxicity in Drosophila melanogaster. TA attenuates rotenone-induced increased levels of oxidative stress and inflammatory markers, and reduced cell viability in D. melanogaster. TA restored to normal, rotenone-induced elevation of H 2 O 2 (Fig. 3A), nitric oxide (Fig. 3B, nitrite/nitrate), and blocked the increase of protein carbonyl (Fig. 3C),  Molecular docking analysis. The docking scores of TA was higher than those of the standard inhibitors (Rofecoxib, Pralnacasan and Thalidomide) for the inflammatory protein targets (TNF-α, Caspase-1 and Cox-2) in both D. melanogaster and H. sapiens, except for Drosophila capase-1 (DmCasp-1) where the docking score of Pralnacasan (− 7.8 kcal/mol) was lower compared with that of TA which was − 7.2 kcal/mol (Table 1). Also, TA occupied the binding pockets of the respective standard inhibitors for all the protein targets, thus, sharing similar amino acids for interactions at the active sites (Fig. 6).
TA and Thalidomide interacted with the triad amino acid residues (His 281, Ala 279 and Arg 277) of the active site of the DmTNF-α. Both compounds also interacted with common amino acid residues at the active pocket of HsTNF-α ( Fig. 6A-D). Both TA and Rofecoxib bound to Val 291 and Lys 211 residues of the active pocket of HsCox-2, and to Tyr 193, Phe 192 and Phe 133 triad residues of the Drosophila homolog of the same enzyme ( Fig. 7E-H). The interactions of the natural compound and the caspase-1 inhibitor (pralnacasan) at the binding pockets of the human protein target (HsCasp-1) occurred with Ile 155, Trp 145 and Ala 141 residues; and of the Drosophila homologue (DmCasp-1) with Thr 308 residue ( Fig. 8I-L). Nonetheless, TA seemed to form strong hydrogen bonds, in addition to several other bond types (alkyl and van der Waal forces) (Figs. 6A-D,7E-H,8I,J).

Discussion
Rotenone has been established to induce neurodegenerative diseases such as PD in animals including Drosophila melanogaster. Some of the hallmark features of PD include oxidative stress, inhibition of mitochondrial complex I, dopaminergic neuronal loss and locomotor abnormality 4,26,27 . However, the use of trans-astaxanthin (TA) to treat and attenuate the debilitating effects rotenone-induced toxicity in D. melanogaster is scarce in the literature. Here, we investigated the protective capacity of TA on rotenone-induced toxicity using D. melanogaster as   www.nature.com/scientificreports/ a model. We evaluated different behavioural and oxidative stress markers as well as molecular mechanisms of actions of TA against inflammatory protein targets using in silico approach. From the different concentrations of TA used in this study, 0.1, 0.5, 1.0 and 2.5 mg/10 g diet prolonged the lifespan of flies by 28.78, 36.36, 36.36 and 25.75%, respectively, compared with the control. This suggests that TA might possess anti-aging activity by extending the lifespans of the flies. Our finding is in support of the report of Huangfu et al. 28 showing that TA has pro-longevity activity against SOD1 null mutant flies by protecting against oxidative stress. It also corroborates the lifespan extension in C. elegans by preventing against oxidative stress through interaction with mitochondrial complex III electron leakage 29 .
Acetylcholine (ACh) and Dopamine (DA) are the major antagonistic neurotransmitters, released from the striatal cholinergic interneurons (SCI) and substantia nigra (SNi) neurons, respectively, onto the dorsal striatum of the midbrain 30 . Membrane-bound post-synaptic acetylcholinesterase (AChE) liberates choline and acetyl group from ACh into the synaptic cleft, thereby alleviating cholinergic toxicity that is responsible for the motor symptoms, such as bradykinesia, tremor and rigidity, observed in PD 31 . Several studies have indicated an altered  www.nature.com/scientificreports/ AChE activity by rotenone in different animal models, including D. melanogaster 4,32 . In this study, we found that the inhibition of AChE activity by rotenone in D. melanogaster, was coupled with impaired locomotor activity. Interestingly, we found that TA improved the rotenone-induced inhibition of AChE activity and decline in locomotor performance of the flies. The observed rotenone-induced accumulation of H 2 O 2 and inhibition of catalase activity are indications of oxidative damage. Indeed, rotenone has been shown to inhibit mitochondrial complex I leading to electron leakage and superoxide radical production that in turn dismuted to H 2 O 2 by superoxide dismutase. The elevated H 2 O 2 can produce reactive and short-lived hydroxyl radical (OH·) through Fenton reaction, which in turn can induce oxidative damage to proteins, nucleic acids and lipid components 5 . These observations prompted us to corroborate the fact that, indeed, TA possesses antioxidative property against ROS 17,18 .
The observed inhibition of Glutathione-S-transferases (GST) in the rotenone-exposed flies might suggest GST's dysfunctional detoxification capacity 9,33 . The observation that TA ameliorated rotenone-induced inhibition of GST further strengthens its antioxidative capacity. Consequently, we evaluated thiols' status of the D. melanogaster. Thiols are carbon-bound sulfhydryls (SHs) chemical groups. They are classified into protein thiols and non-protein thiols (NPSHs), in which the latter is majorly composed of GSH, besides other thiols 34 . The thiols are involved in many physiological roles among which is tissue protection against oxidative damage 28 . Low level of thiols has been associated with many cellular dysfunctions, aging and neurodegenerative diseases, including PD 35,36 . Hence, decreased level of NPSHs has been reported in both PD patients and experimental models of PD 37 , including rotenone-induced PD model in flies 38 . In this study, TA restored the protective status of both total thiols and NPSHs, that were depleted in flies treated with rotenone.
We further evaluated other oxidative stress markers such as protein carbonyls (PCs) and malondialdehyde (MDA, a product of lipid peroxidation). In fact, different products of PCs are generated from oxidative assault to amino acid side chains or backbone cleavage in a protein by hydroxyl radicals 39 . PCs formed by the product of lipid peroxidation, such as MDA, has been used to accentuate the oxidative source of PCs 40 . Aging tissues and neurodegenerative diseases have been shown to be associated with increased levels of PCs and lipid peroxidation product 40 . Studies have reported elevated levels of PCs and MDA in D. melanogaster fed with rotenone 40 . Interestingly, we found that inclusion of TA in the diet reversed rotenone-induced elevation of PCs and MDA in flies.
Also, we sought to understand whether toxicity induced by rotenone reduced cell viability by carrying out MTT assay, which is a measure of cellular and mitochondrial metabolic rate. We found that rotenone indeed reduced cell viability in the flies. Strikingly, TA ameliorated rotenone-induced reduction of cell viability in D. melanogaster. Moreover, we carried out emergence rate of progenies of flies and found that TA did not attenuate rotenone-induced reduction in emergence rate.
Nitric oxide [NO, (nitrite/nitrates)] can react with superoxide anion radical to form a more poisonous oxidative species (ONOO·, peroxynitrite) that have been evidenced in many pathologies 41 . This pro-inflammatory marker is physiologically released during inflammatory burst by the immune cells. However, in diseased state, NO is uncontrollably released in an accumulative manner. This, consequently, can lead to the production of peroxynitrite and other pro-inflammatory markers that results in cellular apoptosis 42 . The remarkable amelioration of rotenone-mediated elevation of NO by TA confirms that it possesses anti-inflammatory activity.
Molecular docking analysis was performed to provide further insight into the molecular mechanism of the anti-inflammatory capacity of TA against rotenone-induced inflammation. Tumour necrosis factor-alpha (TNFα), Cyclooxygenase-2 (Cox-2) and Caspase-1 (Casp-1) are the pro-inflammatory mediators accompanied with NO release, through NF-kB and inflammasome complex-mediated responses 43,44 . These proteins are members of the group of pro-inflammatory markers that might be stimulated or released by glial cells, as a result of the signalling interplays of their molecular origins, during chronic neuroinflammation-induced PD [45][46][47] . The Cox-2, TNF-α, and Caspase-1 have been implicated in the neuroinflammation-mediated PD 48 . The inhibition of these three protein targets are windows of opportunity for a new anti-PD drug development. In the light of this, TA was subjected to molecular docking against human TNF-α, Cox-2 and Casp-1 protein targets, as well as their Drosophila homologues. Interestingly, TA showed higher docking scores against both the human and Drosophila proteins than the standard inhibitors except for the Pralnacasan against DmCasp-1. The molecular interactions of TA with the anti-inflammatory protein targets could be the mechanisms employed by TA to regulate the inflammatory response triggered by exposure of flies to rotenone. The increased binding affinity of TA to target proteins can be attributed to different factors, such as the more negative change in free energy, the interaction of its different functional groups with the amino acid residues at the active pockets of the proteins, and the nature of the binding that occurred between them 49,50 . These molecular interactions depend on the chemical structure of TA which contains the keto-(C=O) and hydroxyl (OH-) groups on the two terminal β-ionone rings and the central conjugated carbon (-C=C-C=C-) group joining the two β-ionone rings. The C=O and OH-groups on the β-ionone ring formed different types of hydrogen bonds (conventional, carbon-hydrogen and pi-donor), being the strongest noncovalent bond, with the amino acids at the active sites of the human and Drosophila protein targets. Additionally, the polyene group, as well as free carbon-hydrogen (C-H) was also responsible for hydrophobic bond types such as alkyl and pi-alkyl. Likewise, van der Waal forces do exist. Thus, TA, using its hydroxyl group, was strongly hydrogen-bonded with Asp10 and Lys11 at the binding site of the human TNF-α, and also with Asp275, and Ser310 and Arg407, using hydroxyl and keto groups, respectively, at the binding site of the Drosophila homologue of TNF-α. Also, His386 and Asn382 at the binding pocket of human Cox-2 were tightly hydrogen bonded with both hydroxyl and carbonyl oxygen of the β-ionone ring of the TA, whereas it was the carbon-hydrogen of the ring and the conjugated bonds in the bioactive compound that formed several alkyl bonds with many amino acids at the active site of Cox-2 homologue of D. melanogaster. Nonetheless, many alkyl bonds and van der Waal forces of the carbon-hydrogen and polyene bonds in the TA with several amino acids in the binding site of the human Caspase-1 could also be responsible for the higher binding affinity. www.nature.com/scientificreports/ In conclusion, rotenone-induced toxicity is due to the perturbation of redox system, leading to oxidative stress, inflammation and behavioural abnormalities in flies. Interestingly, TA exhibited protection against rotenone-induced oxidative damage by enhancing endogenous antioxidant capacity, decreasing oxidative stress markers, and inflammation. Moreover, the interaction of the central polyene system as well as carbonyl and hydroxyl functional moieties of the two terminal β-ionone rings of TA, contribute to its antioxidative and antiinflammatory properties (Fig. 9). Thus, TA can be considered as a promising therapeutic compound against oxidative stress and inflammation induced by rotenone and or related pesticides and insecticides.

Materials and methods
Chemicals. All chemicals used were commercial products of analytical grade. Trans-astaxanthin (TA) and rotenone were procured from AK Scientific, 30023 Ahern Ave, Union City, CA 94587, USA.
Drosophila melanogaster stock and culture. D. melanogaster wild-type (Harwich strain) flies, that were obtained from National Species Stock Center (Bowling Green, OH, USA), were maintained and reared in Drosophila Laboratory, Biochemistry Department, University of Ibadan, Nigeria on cornmeal medium containing 1% w/v brewer's yeast, 2% w/v sucrose, 1% w/v powdered milk, 1% w/v agar, and 0.08% v/w nipagin at constant temperature and humidity (22-24 °C; 60-70% relative humidity) under 12 h dark/light cycle conditions. Treatment of D. melanogaster with TA and rotenone. In order to select suitable concentrations and duration of exposure of flies to TA and rotenone, we carried out longevity and 30-day survival assays, respectively. Based on the longevity and survival data obtained, 7 days was chosen as duration to evaluate individual effects of TA (0, 0.1, 0.5, 1.0, 2.5, 10, and 20 mg/10 g diet) and rotenone (0, 250 and 500 μM/10 kg diet) on some selected biochemical markers. Thereafter, the flies were anaesthetized on ice, weighed, homogenized in 0.1 M phosphate buffer (pH 7.4, ratio of 1 mg:10 μL) and centrifuged at 4000×g for 10 min at 4 °C in a thermo scientific Sorval Legend Micro 7R centrifuge. Supernatants were collected, stored at − 20 °C and used for the determination of the following biochemical parameters: Non-Protein Thiol (NPSHs), Total Thiol (T-SHs), Hydrogen peroxide (H 2 O 2 ), Nitric Oxide (NO, nitrate and nitrate), catalase, Glutathione-S transferase (GST), and Acetylcholinesterase (AChE) 33 .
From the data obtained above, we selected 1.0 mg/10 g diet of TA to understand its protective capacity on rotenone (500 μM)-induced toxicity after 7 days of treatment as follows: Group 1 (Control, vehicle (2% Ethanol).  www.nature.com/scientificreports/ (50 flies/replicate, n = 5). Thereafter, the flies were anaesthetized in ice, weighed, homogenized in 0.1 M potassium phosphate buffer (pH 7.4, 1 mg:10 μL buffer) and processed as described above for the determination of MTT assay and the following oxidative stress and antioxidant markers-H 2 O 2 , NPSHs, T-SHs, protein carbonyls, malondialdehyde (MDA), NO (nitrate and nitrite), catalase and GST as well as AChE anf emergence rate.
Measurement of longevity and survival rate of D. melanogaster after exposure to TA and rotenone. Both sexes of D. melanogaster (Harwich strain) 1-to 3-day old, were divided into different groups (50 flies/replicate, n = 5), and treated separately with ethanol (2.0%, control) and TA (0.1, 0.5, 1.0, 2.5, 10, and 20 mg/10 g diet) and rotenone (0, 250 and 500 μM), respectively. The flies were changed at least once per week into diets containing similar concentrations of TA and rotenone. The flies were monitored and mortality was recorded daily, and used to plot the longevity curve for TA and 30 days survival rate for rotenone. The data were presented as a percentage of the control 9 .
Measurement of locomotor activity. Locomotor activity of TA-and rotenone-exposed flies was determined by employing the method of Feany and Bender 26 . Briefly, ten flies from the control and treated groups were briefly immobilised using ice and separately placed in labelled glass column of 15 cm length and 1.5 cm in diameter. The flies were allowed to recover from anaesthesia and gently tapped to the bottom of the column. Thereafter, the number of flies that climbed up to the 6 cm mark and those below this mark were recorded. The data were then expressed as percentage of flies that crossed up to and beyond the 6 cm mark of the column.
Biochemical parameters. Total protein determination. Total protein was determined according to the methods of Lowry et al. 51 using Bovine Serum Albumin as standard. The values of the total proteins for samples were used to calculate the activities of the antioxidant enzymes.
Determination of total thiols and non-protein thiol contents of D. melanogaster exposed to rotenone and TA. The levels of total and non-protein thiols were determined based on the method of Ellman 52 . The reaction mixture was made up of 170 μL of 0.1 M potassium phosphate buffer (pH 7.4), 20 μL of sample, and 10 μL of DTNB. The reaction was allowed to incubate for 30 min at room temperature, and the absorbance was measured at 412 nm in a SpectraMax microplate reader (Molecular devices). To determine the non-protein thiol content, samples were precipitated with 4% sulphosalicyclic acid (4%, ratio 1:1) and centrifuged for 10 min at 5000 rpm at 4 °C. The assay mixture for non-protein thiols consisted of 550 µL of 0.1 M phosphate buffer, 100 µL of supernatant and 100 µL of DTNB. For both total and non-protein thiols, GSH was used as standard, and the data were expressed as in μmol/mg of protein.
Determination of glutathione-S-transferase activity of D. melanogaster exposed to rotenone and TA. The GST activity was assessed according to Habig  Determination of catalase activity of D. melanogaster exposed to rotenone and TA. The activity of catalase was carried out using the method of Aebi 54 . This was based on the clearance of hydrogen peroxide in the presence of catalase. The reaction mixture was made up of 1800 μL of 50 mM phosphate buffer (pH 7.0), 180 μL of 300 mM H 2 O 2 , and 20 μL of sample (1:50 dilution). The loss in absorbance of hydrogen peroxide was monitored at 240 nm, for 2 min at 10 s intervals and at 25 °C with a UV/Visible spectrophotometer. The catalase activity was expressed as μmol of H 2 O 2 consumed/min/mg protein.
Determination of hydrogen peroxide level of D. melanogaster exposed to rotenone and TA. The protocol by Wolff 55 was used to estimate H 2 O 2 level. The reaction mixture contained 590 μL of FOX-1 (Ferrous Oxidation-Xylenol orange) reagent and 10 μL of sample. This was followed by 30 min incubation at room temperature, and the absorbance measured at 560 nm. The concentration of hydrogen peroxide generated was determined using extinction coefficient of H 2 O 2 and expressed as μmol/mL. Determination of nitric oxide (nitrate/nitrite) level of D. melanogaster exposed to rotenone and TA. Nitric oxide (nitrate and nitrite) was quantified using Griess reaction following the protocol described at Green et al. 56 . The fly homogenates were incubated in Griess reagent at room temperature for 20 min, and the absorbance measured at 550 nm. The concentration of NO in the samples were calculated using the standard calibration curve of NaNO 2 and expressed as μmol/L. Determination of acetylcholinesterase (AChE) activity of D. melanogaster exposed to rotenone and TA. AChE activity was determined with the method of Ellman et al. 57 The reaction contained www.nature.com/scientificreports/ 412 nm using a SpectraMax microplate reader (Molecular devices). The enzyme activity was estimated as μmol of acetylthiocholine hydrolyzed/min/mg protein.
Determination of malondialdehyde level of D. melanogaster exposed to rotenone and TA. Lipid peroxidation was determined as previously discussed by Ohkawa et al. 58  Determination of protein carbonyl content of D. melanogaster exposed to rotenone and TA. Estimation of Protein Carbonyl (PC) content was done following the method of Dalle-Donne et al. 59 . The samples were added to trichloroacetic acid (20%) to precipitate the proteins. The reaction of the carbonyl group with 2, 4-dinitrophenylhydrazine to yield a stable dinitrophenylhydrazone. This was then mixed with guanidine hydrochloride (6 M) and the absorbance was measured at 375 nm. Protein Carbonyl content was quantified using the molar absorption coefficient of 22,000 M −1 cm −1 .
Determination of mitochondrial metabolic rate (cell viability) of D. melanogaster exposed to rotenone and TA. Mitochondrial metabolic rate (cell viability) was determined based on enzymatic reduction of MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) to MTT-formazan in flies at a final concentration of 5 mg/mL according to Abe and Matsuki 58 . The data were expressed as a percentage of the control.
Measurement of emergence rate of D. melanogaster exposed to rotenone and TA. The emergence rate of the progenies of flies into adulthood were also investigated after parent flies were treated with both TA and/or rotenone as previously reported 4,9 . Briefly, 1 to 3 days old of 10 males and 10 female flies/ vial (5 replicates) were placed in diets containing similar doses of rotenone and TA as described above for 24 h. The flies were then removed from the diets while the embryos were allowed to develop to adulthood. The number of newly emerged flies from each vial were then recorded over a period of 2 weeks and expressed as percentage of control.
Molecular docking. Molecular docking analysis was performed to further investigate the anti-inflammatory capacity of TA by determining its binding affinity for some inflammatory protein targets (TNF-α, Caspase-1 and Cox-2) as compared to their respective standard inhibitors (Rofecoxib, Pralnacasan and Thalidomide). The grid box was set based on the protein structural dimensions and each ligand was set to have eight exhaustive conformations before being forwarded for docking using auto-dock vina in the PyRx workspace. The 3D structures of the protein-ligand complexes were prepared using the Chimera 1.14 workspace, and this was followed by visualization of the ligand-protein interactions and generation of the 2D structure using the Discovery Studio 2020.
Ligands and protein mining. The structure data file (SDF) format for the natural compound (TA) and inhibitors of TNF-α (Rofecoxib), Caspase-1 (Pralnacasan) and Cox-2 (Thalidomide) were downloaded from the PubChem database. The 3D crystal structures, also in SDF format, for TNF-α (PDB ID: 2AZ5), Caspase-1 (PDB ID: 1ICE) and Cox-2 (PDB ID: 5KIR) of humans (Homo sapiens) were obtained from the RCSB protein data bank (PDB), whereas the amino acid sequence of the protein homologues of the same protein targets for D. melanogaster were obtained from the UniProt webserver and their homology models were generated using the SWISS-MODEL webserver.
Ligand and protein preparation. Individual 3D structures of the selected protein targets were uploaded onto Chimera 1.14 workspace in order to prepare them for docking and they were converted into PDBQT format using the PyRx software. The ligands were also imported onto the PyRx workplace accordingly and converted to PDBQT format using the Open Babel plugin. Then, the proteins and ligands in PDBQT format were selected and forwarded for grid box generation.
Statistical analysis. Data were expressed as mean ± standard deviation. One-way ANOVA (Analysis of Variance) followed by Tukey's post hoc test was utilized to calculate the significant differences among groups under various treatments. Statistically significant difference was set at p < 0.05, using the GraphPad Prism5.0 software.

Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.