Study on the effect of processing methods on the total polyphenol, 2,3,5,4’-tetrahydroxystilben-2-O-β-D-glucoside, and physcion contents in Fallopia multiflora Thunb. Haraldson root

Abstract This study investigated the changes in the ingredients in Fallopia multiflora Thunb. Haraldson (FMT) root after processing it with different methods such as soaking, stewing, and steaming or combined methods. The total polyphenol, 2,3,5,4′-tetrahydroxystilben-2-O-β-D-glucoside (THSG), and physcion contents in FMT products after processing were determined using high-performance liquid chromatography (HPLC) and ultraviolet-visible (UV-VIS) methods. The results demonstrated that the processing method and time significantly affected the contents of polyphenol, THSG, and physcion. The physcion and total polyphenol content increased or decreased during processing depending upon the processing time, while the THSG content gradually decreased with an increase in the processing time. The content of physcion (a substance that can cause liver toxicity) was analysed, and the suitable conditions for processing of the FMT products were determined as initial soaking in rice swill for 24 h and subsequent stewing with black beans and water for 12 h.

and anthraquinones are toxic to liver cells (Lv, et al., 2015;Ma, et al., 2015).A correlation between the processing time, content of substances, and liver toxicity of the processed FMT products was analysed.Ruo-Lan Li et al. reported that the hepatotoxicity caused by FMT was significantly decreased when FMT root was processed for 72 h.Stilbene glycoside might cause liver damage.It was confirmed that processed FMT products lead to lower liver toxicity than that of raw FMT (Li et al., 2020).Therefore, processing of FMT is crucial to increase its safe consumption.Yun-Xia Li et al. reported a correlation between the liver damage and dose of FMT (Li et al., 2017).Tests were performed on rats, and it was observed that raw FMT caused liver toxicity at a dose of 20 and 40 g/kg, and the toxicity was decreased at a dose of 10 g/kg.It was recommended that a high dose of FMT should not be consumed.
FMT contains multiple phenolic compounds such as 2,3,5,4'-tetrahydroxystilbene-2-O-β-D-glucoside (THSG), physcion, emodin, and chrysophanol.THSG is the most valuable constituent because it enhances biological activities such as anti-inflammatory ability (Zeng et al., 2011;Wang et al., 2008;Zhang et al., 2007), anti-oxidation ability (Christian et al., 2015), anti-aging ability (Ling, Xu, 2016), enhancement of hair growth, reduction of premature grey hair (Han et al., 2015), prevention of insomnia (Qian et al., 2017), and limitation of hypopigmentation (Jiang et al., 2009).The content of the major chemical constituents in FMT were determined by HPLC-PAD-MS method.It was observed that the content of THSG, emodin-8-O-β-D-glucoside, physcion-8-O-β-D-glucoside emodin, and physcion were 35.75, 2.86, 0.64, 0.22, and 0.13 mg/g, respectively (Yi et al., 2007).It can be observed that a few studies have been conducted on the quantification of polyphenol (TP), THSG, and physcion contents in processed FMT products.Processing is necessary to eliminate or reduce the toxicity and side-effects, and modify the nature and action to enhance the therapeutic effects of the crude herb.Methods should be developed to control the quality of these products and affirm the effectiveness and safety of processed products (Tao et al., 2019).Moreover, the study on the effect of processing methods (soaking, stewing, and steaming methods) on the composition of processed FMT products has not been systematically performed.Therefore, the effect of the FMT processing methods on the contents of TP, THSG, physcion was investigated using high-performance liquid chromatography (HPLC) and ultraviolet-visible (UV-Vis) methods.The evaluation of the quality of processed FMT products was performed based on the results, which provided a basis for further studies on liver toxicity of processed FMT products.
Additionally, various factors such as geographical origin, harvesting, and processing can affect the quality of herbal materials, which can result in different pharmacological effects.Suitable climate and soil characteristics can ensure biodiversity and are potential factors for the development of medicinal herbs.Therefore, FMT roots were obtained from Ky Son district, Pu Hoat Nature Reserve, and Nghe An province.Vietnam is a crucial location for this study because it is characterized by Bazan Redland soils and sub-temperate region.Additionally, it is recognized by UNESCO as a natural biosphere reserve.

Plant materials
FMT roots were collected from Ky Son district, Pu Hoat Nature Reserve, and Nghe An province (Vietnam) (refer Figure 1).The average weight of fresh roots was approximately 98.6 -200.0 g.The roots had a reddishbrown colour and were undamaged.The FMT roots were initially washed with clean water and cut into small pieces (thickness of 2 -3 mm) before they were dried in an oven at 60 o C (until the moisture of the samples was lower than 12 %).Subsequently, the samples were packaged in vacuum conditions and stored at 25±0.1 o C.

Page 3/12
The effect of processing methods on the total polyphenol, THSG, and physcion contents in Fallopia multiflora root

Processing FMT roots
Ruo-Lan Li et al. confirmed that the therapeutic effect of FMT can be enhanced after processing with black bean decoction.Moreover, the toxicity of FMT can be considerably decreased after processing with black bean decoction (Li et al., 2020;Vietnamese Pharmacopoeia V, 2018).Therefore, the processing of FMT roots has been prepared by referring to Vietnamese Pharmacopoeia V and Li et al. with slight modifications.Initially, 50 g of black beans was cooked with 4 L of clean water for 3 h to obtain the black bean decoction for the test.The ratio of black beans: FMT root was 1:1 (wt/wt).Subsequently, 10 samples of FMT roots (50 g per each) were added into a 500 mL glass beaker.The FMT roots were processed using 10 different methods and were correspondingly denoted as follows: P1: FMT roots were steamed in water for 12 h.P2: FMT roots were steamed in black bean decoction for 12 h.P3: FMT roots were soaked in rice swill for 12 h and then steamed in black bean decoction for 12 h.P4: FMT roots were soaked in rice swill for 24 h and then steamed in black bean decoction for 12 h.P5: FMT roots were soaked in rice swill for 24 h.Subsequently, 50 g of black beans and 200 mL of water was added and the mixture was stewed for 12 h.T0: FMT roots were soaked in rice swill for 24 h.T1: FMT roots were soaked in rice swill for 24 h.Subsequently, 50 g of black beans and 200 mL of water was added, and the mixture was stewed for 6 h.T2: FMT roots were soaked in rice swill for 24 h.Subsequently, 50 g of black beans and 200 mL of water was added, and the mixture was stewed for 12 h (it is prepared using the method similar to that of the P5 sample).T3: FMT roots were soaked in rice swill for 24 h.Subsequently, 50 g of black beans and 200 mL of water was added, and the mixture was for 18 h.T4: FMT roots were soaked in rice swill for 24 h.Subsequently, 50 g of black beans and 200 mL of water was added, and the mixture was stewed for 24 h.T5: FMT roots were soaked in rice swill for 24 h.Subsequently, 50 g of black beans and 200 mL of water was added, and the mixture was stewed for 30 h.
The samples were taken out after soaking in rice swill and dried before steaming or stewing.
The samples were dried at 40 o C for 24 h after processing.The humidity in the samples was determined using an ADAM AMB 310 automatic moisture content meter.The moisture requirement of the processed samples was lower than 12 %. Figure 2 presents FMT roots before and after processing.

Preparation of sample solutions:
The processed FMT products were ground into fine powder using a disintegrator and sieved through a 0.5mm screen.The samples were accurately weighed (1.0 g/ each) before soaking in 20 mL of solvent (ethanol 50 % for the quantification of THSG and methanol 99.9 % for the quantification of physcion) for 60 min at 25±0.1 o C. Subsequently, the mixture was ultrasonicated for 60 min at 40 0 C. Filtration through a PRP PTFE 0.45-µm membrane filter was performed, and 10 µL of the extract was added to a vial in the HPLC device for analysis (n=3).The sample injection program was set.

Determination of total polyphenol content in raw and processed FMT products using UV analysis
The total polyphenol content (TPC) in raw and processed FMT products was estimated by the Singleton method using Folin-Ciocalteau reagent with a slight modification (Singleton, Rosa, Raventos, 1999;ISO14502-1, 2005).The TPC was expressed as mg of gallic acid equivalents per gram of dry weight (mg GAE/g DW) based on a calibration curve of gallic acid.The absorbance was measured at a wavelength of 765 nm using a UV-1900 spectrometer (Shimadzu, Japan).

Determination of THSG and physcion content in raw and processed FMT products using the HPLC method
The analysis of THSG and physcion contents in raw and processed FMT extracts was performed using a Chromaster device (Hitachi, Japan) with DAD detector, HiQ sil C18HS column (150 x 4.6 mm; 5 µm).The mobile phase consisted of acetonitrile (A) and distilled water containing 0.1 % formic acid (B) with a linear gradient elution at a flow rate of 1.0 mL/min.THSG content analysis: the gradient program is as follows: 0-5 min, 23 % A; 5 -10 min, 100 % A; 10 -18 min, 100 % A; 18 -20 min, 23 % A. The detection wavelength was 320 nm.The column temperature was maintained at 28 o C (Cheng et al., 2013;Liang et al., 2010).

Statistical analysis
The data represented the mean ± S.D. of three independent experiments performed in triplicates.The statistical significance was determined using one-way ANOVA followed by Dunnett's multiple comparison test (P<0.05)using the GraphPad Prism 6 program (GraphPad Software Inc., San Diego, CA, USA).

RESULTS AND DISCUSSION
A standard curve for the quantification of total polyphenol, physcion, and THSG content A standard curve for the quantification of TPC was constructed using the Singleton method (Singleton, Rosa, Raventos, 1999).The calibration equation of gallic acid was y = 9.063x -0.0129 (y is the optical absorption and x is the concentration of gallic acid), and the regression coefficient (R 2 ) of this equation was 0.9935 (refer Figure 3A).
The linear regression analysis of physcion and THSG was performed using an HPLC device with a sample concentration in the range of 12.5 -100.0 µg/mL and 25.0 -125.0 µg/mL, respectively.The regression equations of physcion and THSG were y = 137060x -51003, R 2 = 0.998, and y = 144619x + 63089, R 2 = 0.9966, respectively, wherein x is the concentration and y is the peak area of the sample.The value of R 2 was approximated 1 unit corresponding to these regression equations and used for the calculation of physcion and THSG contents in the test samples (refer Figure 3B and 3C).Nguyen Thi Hong Thanh, Nguyen Thi Hong Tham, Tran Dinh Thang, Nguyen Thuy Chinh, Nguyen Dinh Luyen, Bui Thi Thuy Luyen The validation values for the analysis method of THSG and physcion have been reported in previous studies (Nguyen et al., 2020;Nguyen et al., 2021).The results demonstrated that the retention time in the standard and sample solutions of THSG were 10.053 and 10.013 min, respectively, and that of physcion were 30.653 and 30.980 min, respectively.The system compatibility test exhibited that the relative stand deviation (RSD) of retention time and peak area were 0.39 and 1.61 % for THSG, 0.33 and 0.92 % for physcion, respectively.
The same sample solution was analysed at designated time points in 48 h for stability testing.The results demonstrated that the RSD values of peak area for THSG and physcion were 1.51 and 1.05 %, respectively, which indicated the process for determining THSG and physcion content was stable.
The accuracy testing was performed by the standard addition method.The recovery testing for THSG was performed in the range of 94.59 -98.33 % with an RSD value of 1.52 %.The recovery testing for physcion was 95.88 % with an RSD value of 1.51 %.
The repeatability test was performed using six sample solutions which were prepared using the similar method.The results demonstrated that the RSD value of peak area of THSG and physcion were 0.59 and 1.12 %, respectively, which indicated the repeatability of this method for the analysis of THSG and physcion.

The effect of processing methods and time on contents of THSG, physcion, and total polyphenols in processed FMT products
The HPLC method has been validated as a suitable method for the quantification of physcion and THSG in the FMT roots (Tao et al., 2019;Nguyen et al., 2020;Nguyen et al., 2021).Therefore, HPLC method was used in this study to determine the physcion and THSG contents in the processed FMT products.UV-Vis method was used to determine TPC in the processed FMT products.
The TP, physcion, and THSG contents in the raw FMT roots were determined with the values of 89.59 ± 0.51, 0.56 ± 0.44, and 40.13 ± 0.009 mg/g, respectively.The TPC in the raw FMT roots in this study was two times higher than that in the FMT roots obtained from the Cao Bang province, Vietnam (38.6 mg/g) (Le, 2020).This result proved the effect of geographical origin on the compositions of material herbs.Multiple previous studies have demonstrated that the environmental conditions affect the production of secondary metabolites, which are important sources of bioactive natural components.The physiological and biochemical reactions are influenced directly by environment factors as they affect the phytochemistry composition (Tian et al., 2008;Xu, Xu, 2009).Nguyen et al. (Nguyen et al., 2018) performed the HPLC analysis to determine the content of phenolic compounds such as THSG, physcionin, emodin 8-glucoside, emodin, pleuropyrone A, torachrysone 8-O-β-D-glucopyranoside, 6-hydroxymusizin 8-O-α-Dapiofuranosyl-(1→6)-β-D-glucopyranoside.Additionally, statistical methods were applied to differentiate between natural roots and commercial medicinal slices of FMT.Six samples of 3-to 4-year-old natural FMT roots were obtained from six districts of Ha Giang, which is a northern mountainous province in Vietnam.The samples had different THSG content in the range of 26.211 -55.010 mg/kg dry weight (Nguyen et al., 2020).Samples of FMT were obtained from ten different locations in China and the correlations between 17 environmental factors and 5 bioactive components (THSG, emodin, emodin-8-O-β-D-glucoside, physcion and physcion-8-O-β-D-glucoside) were analysed.The results indicated that the highest contents of bioactive components were detected in samples from Deqing, and the lowest in samples from Tianyang (Yan et al., 2010).Therefore, factors such as region, season, soil, moisture, and analysis method significantly affected the content of TP, physcion, and THSG in raw FMT roots.
In this study, the FMT roots were processed by steaming method, soaking combined with steaming method, and soaking combined with stewing method.The effect of processing conditions on THSG, physcion and TP contents was investigated.

The effect of solvents extraction
The FMT roots were processed by stewing in water (sample P1) or black bean decoction (sample P2).The results of TP, THSG, and physcion contents in the sample Page 7/12 The effect of processing methods on the total polyphenol, THSG, and physcion contents in Fallopia multiflora root The results in Table I exhibited that the utilization of black bean decoction in the processing of FMT roots changed the contents of TP, THSG, and physcion in the processed FMT products.The THSG, physcion, and TP contents in the products processed with black bean decoction (sample P2) were higher than that processed with water (sample P1).Therefore, black bean decoction was selected for the processing of FMT roots.
According to the previous studies, the contents of stilbene glucoside and free anthraquinones in FMT decreased after processing in addition to five auxiliary materials including rice swill, ginger decoction, licorice decoction, prepared Radix Rehmanniae Preparata decoction, and black bean decoction (Li et al., 2020).Several studies have reported that the toxicity of FMT gradually decreased when black bean decoction was used as an auxiliary material.Additionally, the extent of damage caused to the liver cells was lowest in the case of FMT steamed with black bean decoction for 16 h (Li et al. 2020.).Several potential toxic compounds such as emodin-8-O-glucoside and torachrysone-Ohexose were identified based on previous research on FMT.The content of emodin-8-O-glucoside and torachrysone-O-hexose which are of potential toxic ingredients were reduced after processing with black beans (Chen et al., 2021).This indicated that black beans contain chemicals that can reduce the toxic components in FMT roots.

The effect of soaking time in rice swill
FMT roots were soaked in rice swill for different durations (12 h (sample P3), 24 h (sample P4)) before steaming in black bean decoction for 12 h to evaluate the effect of soaking time on the TP, THSG, and physcion contents in processed FMT products.The results are listed in Table I.
The soaking stage of FMT roots in rice swill before steaming affected the content of substances in processed products.The contents of THSG and physcion decreased with an increase in the soaking time of FMT roots in rice swill.The THSG and physcion contents decreased from 25.85 mg/g to 24.17 mg/g and from 0.59 mg/g to 0.57 mg/g, respectively, when FMT roots were soaked in rice swill for 12 and 24 h.
Several potential approaches have been proposed to pre-treat and extract active constituents from medicinal plants.These approaches include pre-soaking, liquid ammonia pre-treatment, and co-digestion.Ammoniabased pre-treatment was the common method used to reduce the lignin content in biomass based on breaking the lignin-carbohydrate ester linkages, depolymerizing The content of physcion in the two samples was increased owing to the hydrolysis of dianthrone glucosides under high-temperature processing (Bartnik, Faccey, 2017).
Nguyen Thi Hong Thanh, Nguyen Thi Hong Tham, Tran Dinh Thang, Nguyen Thuy Chinh, Nguyen Dinh Luyen, Bui Thi Thuy Luyen hemicellulose, and cleaving the crystalline region of cellulose.The early removal of lignin during the conversion process has multiple advantages because lignin significantly hinders enzymatic hydrolysis.This method was used widely in the agriculture and production of biofuel from crop residues, forestry residues, and herbaceous species (Zhao et al., 2017a;Zhao et al., 2017b).However, ammonia pre-treatment has been performed at various operating conditions such as temperature, pressure, solvent, reaction time, whereas pre-soaking is a simpler process.Therefore, the pre-soaking of raw FMT roots with rice swill was investigated to determine the suitable pre-soaking time required for processed products with higher TPC, and lower THSG and physcion contents.In addition, the determination of the water content in FMT roots is crucial to accurately evaluate the characteristics of products.However, this factor has been not extensively studied.The chemical analysis method is extensively used to determine the water content.However, it is time-consuming, labour-intensive, and cannot fulfil the requirements of online analysis (Büning-Pfaue, 2003).
Near-infrared spectroscopy is used to determine the moisture in the plants because it is an efficient method, and it can reduce the analysis cost, and provide an efficient analysis platform (Zhang et al., 2019).
The reduction in contents of THSG and physcion after processing demonstrated the decrease in the toxicity of FMT roots as mentioned by Wu et al. (Wu et al., 2012).The TPC of the processed samples was significantly increased after 24 h of soaking in rice swill.Hence, soaking FMT roots for 24 h in rice swill before steaming in black bean decoction was selected for processing FMT roots.

The effect of preparing method
An investigation on the TPC, physcion and THSG contents in the processing FMT roots with the different preparing methods including soaking for 24 h combined with steaming for 12 h (sample P4) and soaking for 24 h combined with stewing for 12 h (sample P5) was performed.The results of TP, THSG, and physcion contents in processed FMT products indicated that the TPC increased from 97.63 to 107.33 mg/g, whereas the THSG and physcion contents decreased from 24.17 to 20.78 mg/g and from 0.57 to 0.38 mg/g, respectively (refer Table I).These results suggested that for FMT roots pre-soaked in rice swill, the next processing step, i.e., stewing with black beans was relatively effective than that of steaming with black bean decoration.This was explained by the difference between the steaming and stewing methods.The stewing processing is a type of cooking process wherein the ingredients are simmered inside a liquid, while steaming processing uses water vapour, heat, or moisture.The direct contact of FMT roots with black bean decoction led to a significant effect than that of indirect contact.
Previous studies have reported that physcion can cause live toxicity (Wu et al., 2012;Ma et al., 2015).Therefore, soaking combined with stewing method was selected for further studies based on the physcion content in processed products.

The effect of stewing time
The effect of stewing time (0, 6, 12, 18, 24, and 30 h) on TP, THSG, and physcion contents in the processed FMT products is depicted in Figure 4.
Page 9/12 The effect of processing methods on the total polyphenol, THSG, and physcion contents in Fallopia multiflora root The results demonstrated that the TPC in the processed FMT products increased with an increase in the stewing time from 0 to 18 h, and subsequently, TPC was decreased.This was due to the absorption of black bean decoction into FMT slices during the initial stewing hours and rupture of cell wall of FMT, which led to the increase in TPC.However, the polyphenols in FMT become unstable with an increase in the stewing time under high temperature.Therefore, the TPC of the products was decreased.
The content of THSG decreased as a function of stewing time.The THSG content in the processed FMT product was 21.63 and 19.49mg/g after 6 and 30 h of stewing, respectively, which indicated a decrease of 13.99 % and 22.50 %, respectively, compared with that of the THSG content in the product that was not stewed.This was observed because THSG is a water-soluble active substance (Qian et al., 2020;He et al., 2021).Hence, it was extracted during the processing.Moreover, in the case of high-temperature processing, the active components usually become unstable, which leads to a decrease in their content (Tao et al., 2019).
The change in the physcion content was not systematic with an increase in the stewing time.The physcion content in the products was gradually reduced with an increase in the stewing time during the initial 12 h of processing, and increased in the following hours.This was explained by rupture of the cell wall of herbs under the effect of high temperature during processing, which resulted in the release of physcion from cells (Tao et al., 2019).The results of this study were consistent with that of the reports of Zhitao Liang et al. (decreased physiological concentration and increased physcion) and Xiaoqing Wu et al. (decreased 55.8 % of THSG,increased 34 % emodin) (Wu et al., 2012;Liang et al., 2010).
Jiang Ma et al. reported that the liver toxicity of stilbene glucoside is lower than that of anthraquinones and stilbene glucoside.Stilbene glucoside exhibited lipid-regulating and antioxidant activities, whereas anthraquinones are hepatotoxic substances (Ma et al., 2015).Physcion is an anthraquinone and THSG is a stilbene glucoside.Therefore, these substances might be active and toxic.The suitable processing conditions FIGURE 4 -Effect of stewing time on TP, THSG, and physcion contents in the processed FMT products.Nguyen Thi Hong Thanh, Nguyen Thi Hong Tham, Tran Dinh Thang, Nguyen Thuy Chinh, Nguyen Dinh Luyen, Bui Thi Thuy Luyen selected for further studies were soaking FMT roots in rice swill for 24 h before stewing with black beans and water for 12 h.This was selected because the obtained products had the lowest content of physcion and balance between the active and toxic components.

CONCLUSION
The effect of processing method on the content of active substances in Fallopia multiflora (Thunberg) Haraldson (FMT) roots was evaluated.The roots were obtained from Ky Son district and Nghe An province (Vietnam).The FMT roots were steamed in water or black bean decoction, soaked in rice swill, or stewed in black beans with water or these stages were combined for various periods.The decrease in the contents of 2,3,5,4′-tetrahydroxystilbene-2-O-β-D-glucoside (THSG) and physcion in the processed FMT products can reduce the toxicity of the FMT roots.The suitable processing conditions were selected for processing the FMT roots based on the physcion content.These conditions included soaking the FMT roots in rice swill for 24 h before stewing them with black beans and water for 12 h.The THSG, physcion, and TP contents in the FMT product processed at this condition reached 20.78, 0.38, and 107.33 mg/g, respectively.The processing ensured safe consumption of FMT.

FIGURE 1 -
FIGURE 1 -FMT tree and roots A: FMT tree, B: FMT roots, C: Cross section of FMT roots.

TABLE I -
TP, THSG, and physcion contents in the FMT products processed at different conditions.Data were represented as mean ± S.D. of three independent experiments performed in triplicates