Silicon and mechanical damage increase polyphenols and vitexin in Passiflora incarnata L.

Passiflora incarnata L. is a species of global pharmacological importance, has not been fully studied in the context of cultivation and management. It is known that silicon acts on abiotic stress and promotes phenols synthesis. The practice of mechanical damage is widely used in P. incarnata crops, and its interaction with silicon can have a significant influence on plant metabolism. Therefore, our objective was to investigate the effects of silicon and mechanical damage on photosynthesis, polyphenols and vitexin of P. incarnata. The experiment was conducted in a factorial design with SiO2 concentrations (0, 1, 2, 3 mM) and presence or absence of mechanical damage. It was found that mechanical damage improved photosynthetic performance at lower concentrations or absence of silicon. Moreover, this condition promoted an increasing in vitexin concentration when SiO2 was not provided. The application of 3 mM Si is recommended to increase polyphenols and vitexin, without harming dry mass of aerial part. The interaction between silicon and mechanical damage could be a tool to increase agronomic yield and commercial value of the P. incarnata crop.

www.nature.com/scientificreports/ In the commercial crops of P. incarnata more than one harvest is expected, enabling a continuous supply of leaves and stems to the pharmaceutical production chain 27 . The complete harvesting of the aerial part promotes mechanical damage, which can signal the production of phenolic compounds, since this stress may be influence the activity of the PAL enzyme and other enzymes in the polyphenol pathway 28 .
The Si supply and mechanical damage can result in an increase in biomass and active molecules, contributing to the production chain of the species. The objective of this study was to investigate the Si and mechanical damage effects on photosynthetic metabolism and on the polyphenols and vitexin synthesis in P. incarnata.

Results
Chlorophyll a fluorescence and gas exchange. In the absence of silicon, the potential quantum efficiency of the open reaction center (Fv′/Fm′) at 140 days after sowing (DAS) was higher in plants that received mechanical damage. In the absence of damage, at 169 DAS, Fv′/Fm′ was lower in plants grown with 3 mM SiO 2 ( Fig. 1a,b).
At 140 DAS, plants that received mechanical damage and cultivated with 1 and 2 mM SiO 2 showed higher photosystem performance, represented by photochemical quenching (qL), electron transport rate (ETR) and effective quantum efficiency of photosystem II (ФPSII) than intact plants at the same concentrations (Fig. 1g,i,k). In the absence of mechanical damage, energy fraction absorbed by PSII antenna that is dissipated as heat (D) was higher and energy not dissipated and not used for the photochemical phase (Ex) was lower in plants subjected to 0 and 3 mM SiO 2 , which may indicate photoprotection (Fig. 1c,e).
At 169 DAS, regardless the damage, plants cultivated at concentration 1 mM SiO 2 showed lower ETR and ФPSII (Fig. 1j,l). Among plants that did not receive Si, those that received mechanical damage had higher D and qL and lower Ex compared to intact plants (Fig. 1d,f,h).
At 140 DAS plants subjected to damage had higher transpiration rate (E) regardless the SiO 2 level (Fig. 2a). At 169 DAS, plants with 2 and 3 mM SiO 2 with mechanical damage showed high transpiration rate. Among intact plants, those with Si had a lower E (Fig. 2b).
Stomatal conductance (g s ), CO 2 assimilation rate (A net ) and RuBisCO carboxylation efficiency (A net /C i ) were higher at 140 DAS in plants subjected to mechanical damage, except for plants grown with 3 mM SiO 2 (Fig. 2c,e,g). At 169 DAS, g s , A net and A net /C i were higher in plants with damage at the highest SiO 2 concentrations (Fig. 2d,f,h).
Hydrogen peroxide and lipid peroxidation. Plants with 1 mM SiO 2 showed a higher concentration of hydrogen peroxide (H 2 O 2) when damage occurred. In intact plants, SiO 2 supply reduced hydrogen peroxide, except for the 2 mM SiO 2 concentration. (Fig. 3a). In the mechanical damage absence, lipid peroxidation, presented as malondialdehyde concentration (MDA) was higher in plants grown with SiO 2 and in the presence of mechanical damage, there was no difference between plants (Fig. 3b). (c,d) energy fraction absorbed by PSII antenna that is dissipated as heat (D) p < 0.01; (e,f) energy fraction not dissipated in the antenna that cannot be used for photochemistry stage (Ex) p < 0.01; (g,h) photochemical quenching (qL) p < 0.01; (i,j) electron transport rate (ETR) p < 0.01, (k,l) effective quantum efficiency of photosystem II (ΦPSII) p < 0.01 in Passiflora incarnata L. with mechanical damage (w/MD) and without mechanical damage (intact), subjected to SiO 2 variations at 140 and 169 days after sowing. Values corresponding to the averages ± SE. Capital letters compare plants with and without mechanical damage and lower letters compare SiO 2 variations. ETR and PSII results were significant only for SiO 2 variations. Other results were significant for the interaction between SiO 2 variations and the presence/absence of mechanical damage. In this evaluation, there was no significant effect of mechanical damage or even its interaction with SiO 2 levels (Fig. 3c). Plants collected at 169 DAS, regardless the mechanical damage, had the highest vitexin content with 3 mM SiO 2 (Fig. 3d). Plants grown without Si and with mechanical damage showed a higher content of vitexin when compared to intact plants (Figs. 3d and 4).
Carbohydrates. Plants without silicon subjected to mechanical damage showed investment in reserve carbohydrates, such as starch, while intact plants showed high total soluble sugars concentrations (Fig. 5a,d).   (Fig. 5a). Plants subjected to mechanical damage showed a higher amount of reducing sugars, regardless the SiO 2 concentration (Fig. 5b).
Intact plants with 3 mM SiO 2 showed lower sucrose concentration when compared to intact plants without Si. In plants with mechanical damage, the highest concentration of sucrose was found in plants with 1 mM SiO 2 , which did not differ from plants without Si (Fig. 5c).
Among plants cultivated without Si and with 3 mM SiO 2 , those with mechanical damage showed higher starch accumulation. In general, plants grown with 1 mM SiO 2, and damage showed less starch accumulation. In the intact plants, the starch concentration did not differ (Fig. 5d).

Growth indices.
Plants cultivated with 2 mM SiO 2 showed higher dry mass of leaves (LDM) and total (TDM), highlighting that in TDM, plants with 2 mM SiO 2 did not differ from those with lower concentration or Si absence. When 3 mM SiO 2 was provided, TDM was reduced (Fig. 6a). The leaf mass ratio (LMR) was higher in plants that received 2 mM SiO 2 , not different from plants grown with 1 and 3 mM SiO 2 (Fig. 6b).
The LMR expresses the plant area useful for photosynthesis, resulting in plant mass, and the specific leaf area (SLA) data reveal area per leaf mass, indicating its thickness (Fig. 6c). The concentration of 2 mM SiO 2 increased the LMR and 1 mM SiO 2 decreased SLA compared to control plants.    These results indicate that the source of Si used for the conditions of this study was adequate. Si was quickly translocated to the leaves, since SiO 2 supply occurred at 124 DAS and leaf collect, which resulted in biochemical evaluations, was performed at 169 DAS (Fig. 7).

Heatmap.
A heatmap was drawn up to demonstrate the similarity between treatments and the correlation between biochemical variables (Fig. 8). It is possible to observe the formation of two groups in which the treatments in each group have similarity for the variables. The first group consisted of treatments with damage and 0, 2 and 3 mM SiO 2 . The second group consisted of treatments with intact plants and plants with damage and 1 mM SiO 2 . The treatments in the first group showed the highest averages (red squares) for the variables reducing sugars and starch, and the lowest for sucrose and MDA. In this group, treatments 0 and 3 mM SiO 2 with damage had the highest averages for vitexin. On the other hand, the treatments in the second group had the lowest averages for reducing sugars and high for sucrose. In this group, the intact plants treatments that received Si had higher MDA averages. We highlight in this group the treatment with 3 mM SiO 2 which presented the highest vitexin average, opposite to the others, which indicates a relationship with the SiO 2 level supplied. When 3 mM SiO 2 were used in plants with or without damage, higher vitexin averages were verified. However, intact plants with 3 mM SiO 2 revealed high MDA concentration.

Discussion
Our study results emphasize the Si action on the metabolism of plants subjected to abiotic stress, providing better performance under adverse conditions, as observed in other studies 29 .  www.nature.com/scientificreports/ The mechanical damage in P. incarnata at 140 DAS stimulated remained bud's photosynthetic activity, suggesting compensatory photosynthesis 30 , since the removal of old branches allows interception of solar radiation better by young branches. The requirement of higher demand for photoassimilates by new tissues, can stimulate the development and photochemical activity, enabling increased electron flow 31,32 . It's responsible for higher production of reducing agents used in carbon assimilation, observed in the present study, respectively, by higher ETR and ФPSII.
P. incarnata plants with 1 and 2 mM SiO 2 and mechanical damage were efficient in overcoming this damage and restoring themselves, which is observed in the high photochemical efficiency at 140 DAS and high A net /Ci, A net and g s. At 169 DAS the supply of 2 mM SiO 2 promoted increase of g s and A net . These results are in agreement with those verified in the literature 25,26 , in which data of increase in g s , A net , E and dry mass of leaves are verified in P. edulis when Si was supplied.
According to the study by Zhang et al. 22 , the supply of Si may have promoted greater expression of the genes PetE, PetF, PsbP, PsbQ, PsbW and Psb, which are important for the photochemical step of photosynthesis. Gene expression may have contributed to the production of reducing agents used in the biochemical stage of photosynthesis, as indicated in other studies 29,33 , and observed in the high qL at 140 DAS, presented in this study.
At 169 DAS the mechanical damage was preponderant to maintain photochemical energy's direction for the production of reducing agents, since the highest qL was observed in plants without Si. The mechanical damage may favor photochemical activity increase, stimulated by high nitrogen demand for new tissues formation, as well as higher incident radiation stimulates nitrate absorption by the roots. Besides, the reduction of nitrate occurs mainly in leaves, as it is a strong electron drain, it can stimulate greater photochemical activity 31 . It is noteworthy that nitrogen source used in this work was mainly nitric.
The higher photosynthetic activity, reflected by A net, g s and A net /C i , may have contributed to a high concentration of total and reducing sugars, directing resources for growth, biomass accumulation and lower MDA accumulation, a result that indicates low stress in plants grown with 2 mM SiO 2 and mechanical damage. The Si supply in plants with different stress modalities promotes an increase in the activity of antioxidant enzymes, which neutralize reactive oxygen species, decreasing lipid peroxidation [34][35][36] .
Si supply was effective in signaling polyphenol synthesis, as described in the literature 23,24 . We highlight the increase in vitexin provided by the higher dose of Si supplied to P. incarnata. Si promotes greater activity of the PAL enzyme, which participates in the phenols and flavonoids synthesis 23,24 .
Potassium silicate (5, 7.5 and 10 mM) influences apigenin 19 , a precursor flavone of vitexin, which may explain, in this study, the accumulation of vitexin in P. incarnata cultivated with 3 mM SiO 2. The signaling for vitexin production is dependent on Si concentrations and seems not to be related to higher lipid peroxidation and activation of the enzymatic antioxidant system. Mechanical stress can also influence the activity of the PAL enzyme and other enzymes in the polyphenol pathway, as suggested by the results of Liu et al. 28 , and confirmed in this study in the control treatment with absence of Si and with w/MD.
In the presence of mechanical damage, plants grown with 1 and 2 mM SiO 2 were efficient in overcoming stress and these concentrations contributed to the synthesis of polyphenols. It is suggested that these concentrations were enough to signal the PAL metabolic pathway, which promoted an increase in the polyphenol index. As observed in the evaluation of vitexin, the increased activity of the PAL enzyme is stimulated by the supply of Si, resulting in an increase in the content of other phenolic compounds, as related in other studies 23,24 .
Damaged plants accumulated more starch than intact ones in the absence of Si. Starch may have been the result of mechanical stress, activating the enzymatic antioxidant system that reduced free H 2 O 2 . Among plants that didn't receive Si, the stress that resulted in the accumulation of starch may be related to a higher content of vitexin, since stored starch may act as source of carbohydrates for the development of new tissues, in addition to providing carbon skeletons for flavonoid synthesis. Results by Castrillón-Arbeláez et al. 37 reveal that mechanical damage is related to the expression of starch synthase, demonstrating an increase in this carbohydrate. In plants with Si, the accumulation of vitexin should not be related to starch resulting from stress, but the possible signaling triggered by the higher dose of Si supplied, acting on vitexin precursors 19 .
Among plants that received 3 mM SiO 2 , the absence of difference in total soluble sugars, reducing sugars and sucrose may indicate that the production of carbon skeletons was not altered. The starch concentration in plants with 3 mM SiO 2 and mechanical damage suggests accumulation to overcome stress, similar to that observed in plants without Si and with mechanical damage. The supply of 1 mM SiO 2 in plants with mechanical damage increased H 2 O 2 concentration in leaves, but did not result in higher MDA. Plants with damage and Si also had lower MDA than intact plants with Si, indicating the supply of Si under stress conditions contributes to the efficiency of the enzymatic antioxidant system 29 .
The Si supplied to intact plants resulted in an increase in lipid peroxidation, although a higher free H 2 O 2 content was not detected, also pointed out by Coskun et al. 29 . Only the 3 mM SiO 2 concentration was effective in increasing the vitexin content. The results observed with Si supply in intact plants indicate that the stress demonstrated by the higher MDA seems not to be related to the higher vitexin synthesis, which suggests another signaling pathway.
We discovered that P. incarnata showed greater photosynthetic performance when subjected to mechanical damage, which may have triggered a signaling cascade and, associated with Si, resulted in less MDA, with damage recovery and accumulation of phenolic compounds. At a concentration equal to 3 mM SiO 2 , there was higher vitexin accumulation in the plants and a lower dry mass than other treatments. At low Si concentrations, the photosynthetic performance suggests overcoming the mechanical damage.
In P. incarnata crops, mechanical damage is performed by removing the aerial part, which can lead to an increase in vitexin production. The application of 3 mM Si is recommended to increase polyphenols and vitexin, without harming dry mass of aerial part. Supplying 3 mM SiO 2 with increased vitexin by 150% and polyphenols by 130%, suggesting the potential of Si in the phenolic compounds increase in plants 23 www.nature.com/scientificreports/ in the herbal medicines development for the treatment of diseases related to the central nervous system 9,38 . Thus, the interaction between silicon and mechanical damage could be a tool to increase agronomic yield and commercial value of the P. incarnata crop. hydrochloric acid was used to adjust the pH, which was kept between 5.5 and 6.5.

Measurement of chlorophyll a fluorescence and gas exchange.
Chlorophyll a fluorescence and gas exchange were evaluated at 140 and 169 DAS, using the Infra-Red Gas Analyzer, model GFS-3000 Fl-Walz, with a coupled portable modulated light fluorometer. The evaluations took place between 9 a.m. and 11 a.m. on a fully expanded leaf. The variables evaluated were potential quantum efficiency of open PSII center (Fv′/Fm′) energy fraction absorbed by PSII antenna that is dissipated as heat (D), energy fraction not dissipated in the antenna that cannot be used for photochemistry stage (Ex), photochemical quenching (qL), electron transport rate (ETR), effective quantum efficiency of photosystem II (ΦPSII), CO 2 assimilation rate (A net , μmol CO 2 m −2 s −1 ), transpiration rate (E, mmol H 2 O m −2 s −1 ), stomatal conductance (g s , mmol m −2 s −1 ), and Ribulose 1,5-diphosphate carboxylase/ oxygenase (RuBisCO) carboxylation efficiency, by the CO 2 assimilation rate and internal CO 2 concentration in the sub-stomatal chamber (A net /C i μmol m −2 s −1 Pa −1 ).

Plant material samples for biochemical analysis, vitexin and leaf silicon content. At 169 DAS
leaves were collected and frozen in liquid nitrogen to determine carbohydrates, H 2 O 2 and lipid peroxidation. Part of the collected leaves were dried at 38 °C in a forced ventilation oven to determine vitexin and leaf Si content.
Determination of total sugars, reducing sugars, sucrose and starch. The total soluble sugars were obtained by triple extraction, with 80% ethanol and supernatants were combined. The pellet from this stage was frozen for subsequent extraction of starch 40 . Them, starch was extracted by triple extraction with chilled 52% perchloric acid and the supernatants were pooled in falcon until reading.
The quantification of total soluble sugars was performed using the anthrone method, with a spectrophotometer reading at 620 nm, expressed in a standard glucose curve 41,42 . Reducing sugars were quantified with the use of dinitrosalicylic acid (DNS), with a reading at 540 nm and a curve expressed in a glucose pattern 43 . Sucrose quantification occurred with the use of an anthrone + 30% KOH, with 620 nm reading and curve expressed in a sucrose pattern 44 . The starch was determined by the anthrone method, and the reading occurred at 620 nm, with a glucose pattern curve.
Determination of hydrogen peroxide and lipid peroxidation. H 2 O 2 content was determined with trichloroacetic acid (TCA) and reading on a spectrophotometer at 390 nm 45 . Lipid peroxidation was determined with thiobarbituric acid (TBA) and trichloroacetic acid (TCA) and expressed by the formation of malonaldehyde (MDA) 46 .
Determination of vitexin and polyphenols. Determination of vitexin according to Wosch et al. 47 , used 200 mg of crushed dry leaves (38 °C), with the addition of 8 mL of 60% ethanol in 15 mL test tubes. Then, the tubes were vortexed (15 s) and submitted to an ultrasound bath (30 min). Each extract was filtered with cotton and the volume was made up with solvent extractor (ethanol). Samples were filtered with a Millex LCR filter (non-sterile 0.45 μm 13 mm PTFE membrane) and placed in amber glass bottles at 4 °C. The quantification of vitexin in the