Evaluation of microalgae polysaccharides as biostimulants of tomato plant defense using metabolomics and biochemical approaches

Microalgal polysaccharides (PSs) may be an effective elicitor agent that can efficiently protect plants against biotic stresses. In this study, wee investigates, the effect of PS obtained from microalgae and cyanobacteria (D. salina MS002, P. tricorontum MS023, Porphyridium sp. MS081, Desmodesmus sp., D. salina MS067 and A. platensis MS001) on the biochemical and metabolomics markers linked to defense pathways in tomato plants. The phenylalanine ammonia lyase (PAL), chitinase, 1,3-beta-glucanase and peroxidase (POX) activities have been improved in tomato plants leaves treated by polysaccharides extracted from P. triocnutum (238.26%); Desmodesmus sp. (19.95%); P. triocnutum (137.50%) and Porphyridium sp. (47.28%) respectively. For proteins, polyphenols and H2O2, the maximum effect was induced by D. salina 067 (55.01%), Porphyridium sp. (3.97%) and A. platensis (35.08%) respectively. On the other hand, Gas Chromatography-mass spectrometry (GC–MS) metabolomics analysis showed that PSs induced the modification of metabolite profile involved in the wax construction of tomato leaves, such as fatty acids, alkanes, alkenes and phytosterol. PS treatments improved the accumulation of fatty acids C16:3, C18:2 and C18:3 released from the membrane lipids as precursors of oxylipin biosynthesis which are signaling molecules of plant defense. In addition, PS treatment induced the accumulation of C18:0 and Azelaic acid which is a regulator of salicylic acid-dependent systemic acquired resistance. However, molecular and metabolic studies can determine more precisely the mode of action of microalgal polysaccharides as biostimulants/elicitors plant defense.


Microalgal polysaccharides effect on tomato plants defense pathways.
To investigate the effect of microalgae crude PSs, different enzymatic and biochemical markers were used in this study such as PRs proteins (Chitinase and 1,3-β,Glucanase) and PAL (Phenylalanine-Ammonia-Lyase) activities, ROS (H 2 O 2 ) forms, peroxidase (POX) activity and polyphenols content.
Microalgal polysaccharides effect on protein content, PAL, chitinase and 1,3-β,glucanase activities. The results in Fig. 1a show that all PS treatments had a highly significant (p < 0.05) effect on protein content in treated plants compared to non-treated control plants. The maximum effect was induced by D. salina 067 polysaccharides with 55.01% improvement compared to the control.
The enzymatic activity of PAL is highly significant (p < 0.05) in plants treated with different PS extracts (Fig. 1b). The percentages of minimal and maximal improvement (82.46% and 238.26%) in treated plants were induced by Laminarin and P. triocnutum polysaccharides respectively. In the same way, the chitinase activity showed the high significant (p < 0.05) enhancement in treated plants compared to control plant (Fig. 1c). The chitinase activity induced by polysaccharides from Desmodesmus sp., D. Salina 067, Porphyridium sp., P. tricornutum, D. salina 002 and A. platensis was increased by 19.95%, 18.63%, 11.36%, 7.51%, 7.26% and 6.35% respectively compared to control plants (Fig. 1c). Results presented in Fig. 1d, shows significant (p < 0.05) effect Table 1. Biomass production and composition of polysaccharides extracted from Moroccan microalgae.
On the other hand, microalgae PS treatment induced a remodulation of the distribution of sterol metabolites in tomato plant leaves. Generally, there was a disappearance of Stigmastan  Table 2). Table 3 show that the effect of microalgae PS treatment on the profile of alkanes and alkenes in tomato plants. Other metabolites. The results presented in Table 4, shows that the microalgae PS treatment induced the synthesis of new metabolites and the disappearance of other metabolites detected in untreated plants. All of the studied microalgae PS inhibited the biosynthesis of Isopropyl triacontyl ether, while they induced the synthesis of other molecules such as Neophytadiene, E-14-Hexadecenal, phenol, 4-methoxy-2-[2-(5-nitro-pyridinyl) diazenyl]-and Isopropyl octacosyl ether and others as shown in (Table 4).

Discussion
To further evaluate the effect of microalgae polysaccharides on plant tolerance against stress, a biochemical, enzymatic and metabolomics studies on treated plants compared to control plants were performed. In the present study, a bio-elicitor, laminarin, was used as a positive control for its potential to induce plant defense responses; improvement of resistance to viral, fungal and bacterial infections, induction of accumulation of salicylic acid (SA) and induces expressions of ethylene (ET)-and SA-dependent PR proteins 35,36 . In general, it was observed in this study that certain polysaccharides treatments had a similar effect compared to laminarin, however, we see that other treatments have a highly significant effect on studied parameters. For instance, all the treatments exhibited great effect on chitinase, peroxidase, and polyphenols; whereas all treatments except D. salina 067 had great effect on PAL activity. On the other hand, polysaccharides extracted from P. tricornutum and D. salina induced significantly Glucanase activity; and A. platensis polysaccharides enhance H2O2 production. From these results, it can be concluded that microalgae PS can be good candidates for the induction of plant defense. The activities of PAL, POX, 1,3-β Glucanase and chitinase activities and H 2 O 2 content are frequently studied in plant defense research and are known to correlate with the occurrence of induced resistance, as well as defending the host plant against microbial phytopathogens 37 .
Our result showed a strong positive correlation between peroxidase (POX) and PAL activities in response to microalgae polysaccharides treatment (Fig. 4a,b). These enzymes are key defense proteins related to plants resistance to diseases 38 . PAL, is the key enzyme of the phenylpropanoid synthesis pathway, and increase in PAL activity is accompanied by the biosynthesis of active metabolites such as Phytoalexins, lingnins, flavonoids, phenolics compounds and salicylic acid [39][40][41] . Another study has shown seaweeds algino-oligosaccharides could induce the enzyme activities of PAL and POX in rice plants as defense against Magnaporthe grisea 42 45 .
In response to a pathogen attacks, plants accumulate reactive oxygen species (ROS), the main chemical element of oxidative bursts 46 . Faced with the oxidative burst induced by a high accumulation of H 2 O 2 , plants have mechanisms that allow them to keep the H 2 O 2 levels in homeostasis. The latter is controlled by detoxifiers or scavenging enzymes including; SOD (Superoxide dismutase), CAT (Catalase), APX (Ascorbate peroxidase) and POX (peroxidase) [47][48][49] . These enzymes coordinate the elimination of different ROS 50 . This explains our result, which revealed a negative correlation between peroxidase activity and H 2 O 2 accumulation (Fig. 4a,b). That is to say that in the first step, the PS treatment induced the synthesis of H 2 O 2 , then at a certain time, H 2 O 2 levels reached a high concentration which induced the synthesis of POX, in our case, Guaiacol peroxidase, which is found in different cell compartment, capable of producing phenoxy compounds by the reduction of H 2 O 2. The polymerization of phenoxy compounds lead to the formation of cell wall components such as lignin 51 . On the other hand, peroxidases are involved in other physiological processes, namely the synthesis of phytoalexins, the formation of suberin, cross-linking of cell wall components and the synthesis of nitrogen species reagents (RNS) and the induction of the hypersensitive response, which is the programmed death of the host cells in the site of infection, for the limitation of pathogen propagation and the induction of the pathogenesis related proteins 52 . The transformation of phenolic compounds by peroxidases also explains our results which we noted that the high accumulation of POXs -after microalgae polysaccharides treatment-accompanied by a low accumulation of phenolic compounds in treated plants (Fig. 2a,c). glucanase (Glu) and chitinase (Chi) are PR proteins which have the first enzymatic attacks of the plant against pathogens by degrading β-1,3-glucan and chitin present in the fungal cell wall, thus stimulating host defense responses 38 . These two enzymes intervene in the cases of the wounds and the infections by phytopathogens separately or in combination by reinforcing the defense of the plant against the stresses 53 . Our results showed that there is a strong accumulation of PR Proteins; Chitinase and Glucanase with a highly significant positive correlation (Fig. 4a,b). This correlation explains why the PS treatments are able to induce both the acquired systemic resistance (SAR) and the induced systemic resistance (ISR). The accumulation of chitinase is accompanied by the overexpression of its gene (PR3) which occurs in the jasmonic acid and ethylene compounds linked to the ISR. The induction of glucanase is ensured by the overexpression of its gene (PR2). The latter, the SA-dependent marker gene, is involved in the induction of SAR [54][55][56] . Several studies have shown that there is a great synergy between the accumulation of these two hydrolytic enzymes and the induction of plant defense by pathogens. The study done by 57  The PCA analysis revealed that the first three components accounted for 43.4%, 21.6%, and 12.1% of the variance, for a total of 77% (Fig. 4a). We believe that the first two major components provided sufficient information to corroborate the key findings presented here. Generally, the correlation is positive between the different variables (Protein, PAL, Glucanase, Chitinase and Peroxidase) (Fig. 4a,b). Hierarchical clustering spawned two large groups; Group 1: which contains proteins, Chitinase PAL and H 2 O 2 ; and Group 2: which contains Glucanase, Peroxidase and Polyphenols (Fig. 4c). PLS-DA analysis generated some specificity between the type of treatment used and the type of variable induced. In this sense, it was found that the protein level and the glucanase and PAL activities were induced more by polysaccharides extracted from P. tricornutum and D. salina 067. In addition, the enzymatic activities of peroxidase and chitinase and the Polyphenols content were stimulated more by the polysaccharide-based treatment of Porphyridium sp. and D. salina 002. However, it can be seen that the accumulating effect of H 2 O 2 is induced by A. platensis. Polysaccharides extracted from Desmodesmus sp. have a similar effect on all studied variables (Fig. 4d). This moderation of induction of enzymatic activities by microalgal PSs has prompted us to hold the combination of treatments as a perspective for the study of their effect on plant defense pathways.
The analyses, ACP-Biplot, correlation and Heatmap clustering of the metabolic variables, reveal (1): the correlation of the variables studied; (2): the specificity between the metabolite induced vis-à-vis the treatment used;   www.nature.com/scientificreports/ The results in (Fig. 5) show that in general the metabolomics variables are positively correlated with each other (Fig. 5a,b). Heatmap, in Fig. 5d, shows that the polysaccharide treatments from Porphorydium sp., Desmodesmus sp., D. salina 067 have a significant effect on the induction of the synthesis of different metabolites, followed by D. salina 002. On the basis of biochemical and metabolomics results, it can be seen that treatments based on Porphorydium sp., P. tricornutum, Desmodesmus sp. and D. salina 067 have a great potential to induce the different parameters studied. www.nature.com/scientificreports/ The cuticle is composed of cutin polyester and cuticular wax; it allows plants' resistance to biotic and abiotic stresses 58 . Generally, the cutin polymer is formed primarily of esterified and oxygenated fatty acids (FA) linked to C16 and C18 59 . Cuticular wax, which is composed of mixed complexes, contains mainly a variety of alkanes, primary and secondary alcohols, aldehydes, ketones and esters which are derived from very long chain fatty acids (C20-C34) 60 . All this explains our study findings, a strong accumulation of two precursors (fatty acids, C16 and C18) of VLCFA (C20-C24) and VLCA (C24-C33), induced by microalgae polysaccharide treatments. The strain that induced the maximum synthesis of two precursor's palmitic acid C16:0 (9.781 μg mL −1 ) and stearic acid C18:0 (3.694 μg mL −1 ), was P. creuntum. Meanwhile, the appearance of VLCA (C24-C33), explains that VLCFAs (C25-C34), were transformed totally during the synthesis of VLCA (C24-C33).
Our results show that there is a positive correlation between the synthesis of (C16, C18), VLCFAs and VLCAs. This correlation is explained by the transformation of C16 and C18-through the elongation pathway-into VLCFAs 60 . The latter undergo an acyl-reduction and a decarocarbonylation-via the VLCAs-for the transformants into wax esters and ketones respectively. Then the latter are transported to the cuticle 60 . Among the main roles of the cuticle; it is a component that protects plants against desiccation and blocks the penetration of toxins and even pathogens 61,62 . The work done by 63 , showed that Arabidopsis plants exposed to water stress induced high synthesis content of cuticular wax accompanied with a very high accumulation of VLCAs, suggesting that the alkane biosynthesis plays an important role in abiotic stress tolerance. The cuticular permeability increases when the biosynthesis of VLCAs decreases 64 . Our results are comparable to that of 29 , who reported that C. reinarditi and C. sorokiniana polysaccharides significantly induced the synthesis of C16, C18, VLCFAs and VLCAs in tomato plants.
Plants, when infected by pathogens, trigger a series of metabolic redistributions that potentially contribute to induced tolerance or even increase sensitivity 65 .
Polyunsaturated fatty acids (PUFAs) are the primordial components of the plasma membrane. Their release is achieved by lipases after pathogen attack 66 . These FAs play a major role in plant-pathogen interactions; acting directly as free fatty acids or as fatty acid-derivatives (oxilipins) 66 . Table 4. Effect of polysaccharides treatments on profile of others metabolites (µg g −1 ) in tomato plants. www.nature.com/scientificreports/ www.nature.com/scientificreports/ In our study, increased synthesis of C18:2 as PUFAs in tomato plants after treatment with Desmodesmus sp. PSs, is similar to the results of 67 which showed that increase in C18:2 synthesis in avocado fruits elicits increased resistance to the fungal pathogen, Colletotrichum gloesporioides. In our results we also see a strong accumulation of PUFAs; Roughanic acid (C16:3) and Linoleic acid (C18:3) after treatment with different PSs treatments. The maximum values of these two FAs are (2.180 μg mL) and (16.395 μg mL −1 ) respectively, obtained at treatment with Porphyridium sp. Trienoic fatty acids, are PUFAs that play a crucial role in adapting plants to abiotic and biotic stress. A study by 68 showed that the high accumulation of linoleic acid (C18:3) in Arabidopsis leaves after inoculation with Pseudomonas syringaepv. tomato DC3000 (avrRpm1), activates NADPH-Oxidase, which is responsible for the synthesis of reactive oxygen species ROS 68 .
The strong accumulation of C18:3 in plants treated with PSs was accompanied by the accumulation of azelaic acid (AA, C9:0). Also, there was an induction of C18:2 synthesis in treated plants compared with control plants.
Our results are similar to that of 29,69 , who demonstrated that after 24 h of treatment with microalgae polysaccharides, leaf content of C18:1, C18:2 and C18:3 was increased, leading to an induced accumulation of azelaic acid. These three unsaturated octadecanoic are azelaic acid (AA) precursors. Contact of host plants with a pathogen induces their defragmentation from the plasma membrane by phospholipases. After this, AGs undergo oxidation at carbon-9 to synthesize AA (71)(72)(73)(74). The latter has been shown to induce local and systemic acquired resistance (SAR) in Arabidopsis plants against Pseudomonas syringae via the induction of salicylic acid accumulation 69,70 .
Phytosterols play a key role in plant innate immunity against bacterial pathogens by regulating nutrient efflux into the apoplast 71 . Stigmasterol are phytosterols derived from β-sitosterol by the cytochrome P450 CYP710A1 via C22 desaturation 65 . In our results, we observed that the sterol metabolites induced by the different PS extracts are Stigmasta-3,5-diene, Cholesta-3,5-diene and Stigmastan-6,22-dien, 3,5. The accumulation of stigmasterol was higher in plants treated with PS extracts compared to control plants. These results have been confirmed by the Wang's study 71 , who reported that phytosterols play a vital role in plant defense against pathogenic bacteria through nutrient flow regulations from the cytoplasm to the apoplastic space. During the invasion of plants by pathogenic bacteria, the virulence factors of the bacterium block the expression of AtCYP710A1 gene (gene responsible for transformation of β-Sitosterol into Stigmasterol), thus repression of the production of stigmasterol., all its going to induce a reduction of flow of nutrients from cytoplasm to the host cell's appoplast 71 .

Conclusion
In general, microalgae polysaccharides have a bio-elicitor effect on certain biochemical and metabolic pathways of tomato plant defense. In this study PS treatment induced the accumulation of ROS (H 2 O 2 ) and the antioxidant activity of (POX). PS treatments also improved PAL activity, a key enzyme in the activation of the phytoalexin pathway. Pathogenesis-related proteins, such as β-1.3-Glucanase and Chitinase, which are key protein markers for plant defense against biotic stresses were also induced. Moreover, metabolomics study revealed that PS treatment induced the modification of metabolite profiles in tomato plant leaves. Induced metabolites included C18:2 and C18:3 and Azelaic acid which is a regulator of salicylic acid-dependent systemic acquired resistance. On the other hand, treated plants by PS exhibited higher accumulation of C16:0 and C18:0, VLCFAs and VLCAs which are metabolites linked to the strengthening of the cuticle by wax synthesis in order to inhibit the penetration of toxins from pathogens. These metabolites are also involved in the synthesis of sphingolipid complexes, the basal unit of the membrane lipid bilayer. Finally, PSs also induced the accumulation of sitosterol and stigmasterol derivatives. Sitosterol and stigmasterol play a key role in regulating nutrient flow to the apoplast, making plants resistant to pathogens. Meanwhile, Principal Component Analysis (PCA) showed that PSs from P. tricoruntum MS023 and D. salina MS067 significantly improved chitinase and PAL activities and proteins content. Whereas, the PS treatments of Porphyridium sp. MS081 and D. salina MS002 enhanced the POX and β-1,3-glucanase activities and Polyphenols content. On the basis of biochemical and metabolomics results, it can be seen that treatment with polysaccharides obtained from Porphorydium sp., P. tricornutum, Desmodesmus sp. and D. salina 067 have a great potential to induce the different studied parameters. This moderation of induction of enzymatic activities by microalgal PSs has prompted us to hold the combination of treatments as a perspective for the study of their effect on plant defense pathways.
Our results show that polysaccharide treatments based on microalgae can be used as a biotechnological alternative in sustainable agriculture to protect plants against pathogens and abiotic stresses and to also minimize the excessive use of chemical pesticides.

Materials and methods
Microalgae culture conditions and growth measurements. Microalgae strains used in this study www.nature.com/scientificreports/ pellet was washed three times with absolute ethanol and lyophilized 72 . The total neutral sugar and uronic acids contents of the crude PS were determined according to Blumenkrantz and DuBois methods respectively 73,74 . Sulfate content in crude PS was determined by the barium chloride-gelatin method 75 . The protein content was determined according to the Bradford method using crystalline bovine serum albumin (BSA) as standard 76 .
Microalgae polysaccharides effect on tomato plant defense. Bioassays for tomato growth and treatments. In this set of experiments, we used Solanum lycopersicum L. Var. Jana F1 (tomato). Seeds (purchased from "Syngenta Morocco") were surface-disinfected using sodium hypochlorite solution (1%) containing 10 µL of Tween-20 for 20 min and then rinsed with sterile water. Sterile seeds were germinated on Murashige and Skoog (MS) + agar medium in magenta boxes, in the dark for 4 days and in the light for 3 days and the seedlings were transplanted into 12 cm diameter pots containing a mixture of sandy soil/peat (60/40, w/w). The tomato seedlings were grown in a phytotron at 26 4 77 . The polysaccharide treatments were prepared with a ratio of 0.1% PS in distilled water. The pH was adjusted to 6 using HCl or NaOH diluted to 1 M.
The injection was performed using a microsyringe above the cotyledon leaves with a volume of 10μL of the 0.1% PS solution. 30 days after transplantation, the plants were injected with 10 μL of 1 mg mL of microalgae polysaccharides, 10 μL of Laminarin diluted to 1 mg mL −1 as the positive control) and 10 μL of distilled water (dH 2 O) for the negative control. 24 h after treatment, leaves were harvested according to treatment groups and grinded in liquid nitrogen. The samples were then stored at − 80 °C.
Biochemical and enzymatic parameters. Extraction and enzyme activity assay. 100 mg of the tomato plant leaf biomass was homogenized in 1 mL of 100 mM Tris-HCl (pH 8.5) containing 14 mM of β-mercaptoethanol. The mixture was centrifuged at 13,000 rpm and 4 °C for 30 min. The supernatant, which contains the proteins was collected. Protein assay is determined according to Bradford method by using Bovine Serum Albumin (BSA) for the standard curve 76 .
The activity of phenylalanine ammonia-lyase was determined according to the modified methods previously described by Mozzetti and EL Modafar 78,79 .
The chitinase activity was determined according to modified Pegg method 80 . Glucanase activity based on Chandrasekaran method 81 . The ground biomass was homogenized in 2 mL of 0.1 M Tris-HCl buffer (pH 7) containing 1% v/v of polyvinylpolypyrrolidone. The homogenate was centrifuged at 12,000 rpm for 15 min at 4 °C. The supernatant was recovered in 1.5 mL Eppendorf tubes and stored at − 20 °C. The reaction mixture contained 50 μL of the enzymatic extract, 50 μL of laminarine (10 mg mL −1 ) prepared in 50 mM of sodium acetate buffer, (pH 5.0) and incubated at 37 °C for 1 h on a heat block. The reaction was stopped by the addition of 1.5 mL of dinitrosalicylic acid (DNSA) and the mixture was incubated at 100 °C for 5 min a heat block. Absorbance was measured at 530 nm and 1,3-glucanase activity was calculated using the standard glucose curve. The activity of Glucanase was expressed as µM N-Acytelglucosamine H −1 g −1 FM.
The POX activity was carried out, following the method described by Chakraborty 82 . The substrate was prepared with 5 mL of 1% guaiacol, 5 mL of 0.3% H 2 O 2 mixed in 50 mL of 0.05 M sodium phosphate buffer (pH 6.5). The reaction mixture was prepared with 2.95 mL of the substrate and 0.05 mL of the enzyme extract and the absorption change was measured at 470 nm for 3 min. POX activity was determined by the increase in the absorbance due to guaiacol oxidation and expressed as change in the absorption of the reaction mixture min −1 mg −1 of protein (E = 26.6 mM −1 cm −1 ) µkat g −1 FM.
The H 2 O 2 determination according to Sanders method 83 . Tomato plants were harvested, immediately grinded in liquid nitrogen and stored at − 80 °C until H 2 O 2 determination assay. The frozen grinded leaf biomass (150 mg) was directly homogenized in 1 mL solution containing 0.25 ml Trichloroacetic acid (TCA) (0.1% (w:v)), 0.5 mL KI (1 M) and 0.25 ml 10 mM potassium phosphate buffer (10 mM, pH adapted to studied tissue) at 4 °C for 10 min. At the same time, for every sample, a control was prepared with H 2 O instead of KI for tissue coloration background. Good care was taken to protect samples and solutions from light. The homogenate was centrifuged at 12,000 rpm for 15 min at 4 °C. H 2 O 2 content was measured at 350 nm sing a UV/VIS spectrophotometer. A calibration curve obtained with H 2 O 2 standard solutions prepared in 0.1% TCA was used for quantification.
The total phenolic content in tomato leaf extracts was determined by using Folin-Ciocalteu reagent 84 .

Metabolomics analysis.
To investigate the effect of microalgae polysaccharides on the redistribution of metabolites in tomato plants, GC-MS metabolomics analysis in this study according to Kamthan method 85 with some modifications. The metabolites analyzed are lipids, alkanes, alkenes, sterols and other metabolites. 300 mg of frozen leaf biomass was added to glass vials with cap. 10 µL of internal standard, Dodecane (0.75 g mL −1 ) was added to each tube, to which 4 mL of chloroform and methanol mixture with a ratio of (v/v; 2/1) was also added. The mixture was vortexed and incubated at 85 °C for 2 h (2 h on heat block), vortexed and then incubated at 60 °C in an ultrasound bath for). After this, 1 mL of H 2 O was added to the vials and the mixtures were thoroughly vortexed. The organic phase after phase shift was recovered by using a separating funnel and dried by evaporation with nitrogen gas. For acid transesterification, 500 µL of Methanol/sulfuric acid (6%, v/v) was added to the dried organic phase residue. The mixture was incubated on a heat block for 2 h at 85 °C, and then for 1 h in an ultrasound bath. After that, the nitrogen gas was used to drying the mixture, then 250 µL Scientific Reports | (2021) 11:930 | https://doi.org/10.1038/s41598-020-78820-2 www.nature.com/scientificreports/ and 750 µL of H 2 O and Chloroform respectively was added to the mixture. The lower phase was then collected using a separating funnel.
For the metabolomic analysis of volatiles compounds, the gas chromatography (GC) (Agilent 7890A Series) coupled to mass spectrometry (MS) was used in this study. 4 μL of the samples were injected into the 123-BD11 column (15 m × 320 μm × 0.1 μm) by 1/4 split mode using helium as carrier gas at 3 mLmin −1 . The temperature was 230 and 150 °C in ion source and quadrupole respectively, and oven temperature program was started at 30 °C and finished at 360 °C. The NIST 2014 MS Library was used for molecules identification.
Statistical analysis. The R software (R Core Team 2013) 86 was used to statistical analysis, graphical generation and corrplot and ggplot (Hadley Wickham 2016) 87 packages integration. To determine the overall biochemical and metabolomics parameters variation, we used The Principal Component Analysis (PCA) and partial least squares-discriminant analysis (PLS-DA) were used studied the variation of metabolomic and biochemical parameters. The correlation between treatment and variables were studied by prcomp function and MixOmics package. The statistical analysis was studied using SPSS, one-way ANOVAs by Tukey HSD test. www.nature.com/scientificreports/