Cypermethrin standard, Chloroform, acetone, hexane with pesticide-analytical grade, methanol with HPLC analytical grade, O-Methyl-hydroxylammonium chloride (methoxylamine hydrochloride) and dehydrated pyridine were purchased from Wako Pure Chemicals (Osaka, Japan). N-methyl-N-(trimethylsilyl)-trifluoroacetamide with 1% trimethylchlorosilane (MSTFA + 1%TMCS) was purchased from Thermo Fisher Scientific Inc. (PA, USA).

Japanese carp with average weight 25-30 g were purchased from a private fish farm in Kagoshima Prefecture, Japan. The fish were acclimatized for two months under the laboratory conditions before starting the experiment. During adaptation period, fish were fed a commercial fish meal (Nippon Formula Feed Manufacturing, Yokohama, Japan) several times at a feeding rate of 1% fish weight per day 17.

In this study, thirty six carp fish was kept in three aquariums. The feeding was stopped 24 hours before exposure. Fish were divided into three groups (n = 3); first group was the control one (cypermethrin free) and the other two groups were exposed to two concentrations of 0.1 (L-group) and 1 µg/L (H-group) of waterborne cypermethrin solved by for 24 and 96 hours. The concentration of L-group was established similar with that in the aquatic environment. The reported water concentrations of CYP vary from area to another; it was less than 3.5 µg/L in Parana River water in Argentina 18, about 3.55 µg/L Mekong Delta of Vietnam 19. Meanwhile, the concentration of H-group was based from reported 96 hours LC₅₀ of cypermethrin under acetone solvent condition in Cyprinus carpio as 1.7 µg/L 20. Fish were kept in glass tanks (40 L) with mild aeration. The water temperature was maintained at 24±1 ºC, pH was 6±1 and dissolved oxygen concentration was 5±1 mg/L. Photoperiod was 12 h light/12 h dark cycles. The control and test water was completely replaced once per day with dechlorinated tap water left for one day before use.

During sampling, 2-phenoxyethanol of 500 mL/L was used as anaesthetic agent. Hepatopancrease tissues were collected at 24 and 96 h from each group (n = 6). Immediately after sampling, the tissues were frozen by liquid nitrogen to stop the metabolic and enzyme actions and stored at – 80 ºC until metabolomics analysis.

Hepatopancrease tissue samples (ranging from 10-15 mg for each) were placed in a 2-mL PP microtube and 1 mL of a mixture of chloroform, methanol, and Milli-Q water (1:2:0.8, v/v/v) was added, together with a zirconia ball (diameter 5 mm). The microtube was vigorously shaken for 6 min by a sample disruption system (Tissue Lyser II, QIAGEN, and Hilden, Germany). Then, 500 µl of an equal mixture of chloroform and Milli-Q water (1:1, v/v) was added to the tube, which was centrifuged at 12,000×g (4°C) for 10 min. The upper layer from each sample was collected, placed in another 2 mL PP microtube and dried completely in a vacuum centrifuge drier (TAITEC,vc-15sp, Japan).

Derivatization of liver metabolites was performed in two steps. The first, oximation was done by adding 10 μL of a methoxylamine hydrochloride solution prepared by dissolving methoxylamine hydrochloride in pyridine (40 mg/mL) to the dried residue by means of continuous shaking in a water bath at 30°C for 90 min. Secondly, the silylation was done by addition of 90 μL of MSTFA + 1% TMCS to the mixture at water bath at 37°C for 30 min. After derivatization, 100 μL of hexane was added to each sample. The resulting solution was diluted 10-folds and used for analysis.

Metabolites were analysed on an Agilent Technologies 6890 Series gas chromatograph equipped with a 5973 MSD mass selective detector and a DB 5-ms capillary column (i.d.0.25 mm×30 m, 0.25-μm film thickness; J&W Scientific, USA). One microliter of each sample was injected into the column. The temperatures of the injector and detector were 250 and 290°C, respectively. The oven temperature program was as follows: 60°C for 1 min, increase to 325°C at 10°C/min, and then fixed at 325°C for 12 min.

After getting the peaks on the chromatogram, the baseline was drawn to obtain peak area, deconvoluted, and aligned using MetAlign™ (ver. 080311, Wageningen University, Wageningen, the Netherlands). The area obtained from each peak was divided by the weight of each fish and normalized to the internal standard 16. The resultant value was multiplied by 105 and analysed by principal component analysis (PCA) that was performed for metabolites to evaluate the differences of the toxic effects among the groups. PCA score plots and component loadings were used to evaluate the clustering of metabolites along the plot and their contribution to the difference after exposure to cypermethrin. Statistical analysis was performed with the R programming language (http://www.rproject.org/). For all results, p value < 0.05 was considered significant 16.

More than 300 individual metabolite peaks can be detected by GC/MS. The current work focused on the endogenous metabolites related to amino acids, intermediates of kerbs cycle and purines. The statistical analyses were done only using the identified metabolites.

The principal components analysis (PCA) was performed to comprehensively illustrate the variations in the metabolite levels among the groups in response to cypermethrin exposure. At 24 h, the control group clustered on the left side of the plot related to PC1 and the (L & H) groups clustered on the right side of the plot (Figure 1-A). The separation along the PC1 indicated the effects common to the exposure while the separation along PC2 indicated the effects related mainly to the high group. PC1 and PC2 represented 58.44 % and 14.31 % for the variance.

At 96 h, the control group clustered in the left side of the plot while the exposure groups clustered on the right side of the plot along the PC1 which accounted for 65.91 % of the variance (Figure 1-B). Adversely, there is little clustering of the data along PC2 which accounted for 14.77 % for the variance. That strong clustering along the PC1 indicated the intense metabolic perturbation by cypermethrin exposure according to its concentration. (Table 1)

Mitochondria are the power house of the cell responsible for energy supply 21. Glucose is the main source for energy and converted via glycolysis to pyruvate which is then transferred into mitochondria to produce ATP through the TCA cycle and the mitochondrial respiratory chain 22. TCA cycle is a sequence of chemical reactions take place in all living organisms to get energy and it has a vital importance to many biochemical pathways 23. Under normal conditions, the glucose oxidation occurs through aerobic respiration; while under hypoxia, the major energy supply shifted to anoxic or anaerobic respiration, which is less efficient than aerobic respiration in energy production 24. Previous studies found that exposure to CYP caused severe gill alterations; hyperplasia of lamellar cells, telangiectiasis of lamellae and thickening due to cellular infiltration, highly active mucous cells and fusion of secondary lamellae and epithelial lifting 2526. Oxygen transport was affected by gill damage predicting a hypoxic condition. Compared to the control group, a reduction in pyruvate level at 96 h (loading values are 0.5 for PC1 and 0.5 for PC2) was observed (Figure 2). Pyruvate supplies energy through both aerobic and anaerobic respiration, and the other TCA intermediates like succinate, fumerate and malate (Figure 2) suggesting that the TCA cycle was negatively disturbed due to the detrimental effect of cypermethrin on the mitochondrial membrane. This observation of high energy consumption is consistent with previous studies in fishes in which neural hyperexcitability and inhibition of the TCA cycle have been proposed to explain the high demand of energy of organisms under various stress conditions27282930 . In the present study, the level of lactate was decreased at 96 h (loading value is 0.8 for PC1) (Figure 2) revealed its excessive consumption to replenish insufficient energy production 24. Therefore, its low level may pointed to its use as source of energy in the situation of glucose scarcity. Additionally, cypermethrin decreased the glycerol level at 96 h (loading value is 0.8 for PC1) (Figure 2) in response to such urgent need to glucose which in turn leads to activation of gluconeogenesis to get glucose from non-carbohydrate sources as fats and ketone bodies 3132.

Catabolism of proteins and amino acids contribute to a great extent to the total energy production in fishes. Exposure of carp to cypermethrin in the current study resulted in a decline in the hepatic level of most amino acids. These results are in agreement with David et al., 2004 3 after exposure of carp to 1.2 µg/L cypermethrin for 48 h. Protein depletion in liver is considered as a compensatory mechanism to provide intermediates necessary for kreb’s cycle and overcoming the energy crisis occurred as a result of exposure to cypermethrin.

The present outcomes revealed significant reduction in the levels of the BCAAs (leucine, isoleucine and valine) in L- and H- groups at both 24 and 96 h time points compared to control group (Figure 3). Branched chain amino acids (BCAAs) are essential amino acids and are important for protein synthesis. Similar results were observed after exposure of goldfish to lambda-cyhalothrin at concentration 0.012 µg/L for 7 days 24. The decreased levels of BCAAs suggested their usage in synthesis of proteins in response to their damage. BCAAs could stimulate the ventilatory response, were consumed to survive difficult respiration and were involved in energy metabolism 33. Leucine can be degraded to form metabolites such as acetoacetate and acetyl-CoA that play a role in the TCA cycle while valine acts as an intermediary metabolite, forming succinyl-CoA 3435. Alanine is a non-essential amino acid that synthetized in the body via conversion of pyruvate. Alanine is a major raw material for gluconeogenesis and is produced by skeletal muscles and erythrocytes 36. Therefore, changes in the levels of alanine in fish under stress suggest its usage in the pathway of gluconeogenesis. In the present study, a reduction in the level of alanine was observed at 96 h (loading value is 0.8 for PC1) (Figure 4).

Glycine and serine are interconverted primarily in liver and kidneys and they participate in gluconeo genesis, sulfur amino acid metabolism and fat digestion 37. Both amino acids work to maintain blood sugar level. In the present study, their levels were decreased after cypermethrin exposure may refer to the disturbance in serine glycine pathway and glucose synthesis during CYP stress (Figure 4). Similar effects were reported in zebrafish exposed to acetamiprid and halosulfuron-methyl either individually or combined 23.

The purine nucleotides serve as structural components of DNA and RNA molecules, cofactors of many metabolic processes and a source of energy provider adenosine triphosphate (ATP) 38. In the present work, the reduction of xanthine and elevation of hypoxanthine in liver in H-group at both 24 and 96 h time points (Figure 5) suggesting that cypermethrin considerably disturbed purine metabolism in carp fish proposing neurotoxic effects. Exposure of zebrafish larvae to short-chain chlorinated paraffins (SCCPs), acetamiprid and halosulfuron-methyl resulted in similar results 13. Particular neuro-studies proved that the alteration of purine metabolism in human body is highly related to numerous nervous system diseases, such as Alzheimer's disease 39 and Parkinson's disease 40.