High Value-Added Application of Two Renewable Sources as Healthy Food: The Nutritional Properties, Chemical Compositions, Antioxidant, and Antiinflammatory Activities of the Stalks of Rheum officinale Baill. and Rheum tanguticum Maxim. ex Regel

Rhubarb plants (Rheum officinale and R. tanguticum) have edible stalks. In this work, we aimed to compare the nutritional properties, chemical compositions, and bioactivities of R. officinale (SRO) and R. tanguticum (SRT) stalks and to analyze the composition–function relationship. Results showed that the two stalks were good sources of fiber, as well as minerals. They contained abundant essential amino acids and essential fatty acids to regulate the immunity and prevent some chronic diseases; the contents of polyunsaturated fatty acids were 2,244.32 mg/100 g and 2,844.69 mg/100 g, respectively. The antioxidant activity were also proved. Metabolomics showed that SRO and SRT contained abundant phenolic acids. Due to the higher concentrations of flavones, SRT has better antiinflammatory activities than SRO by inhibiting NF-κB signaling pathway. Rhubarb stalks exhibited good safety in acute toxicity and cytotoxicity tests. This work indicated that the two stalks have nutritional value, safety, and bioactivities, and could be used as sources of nutritional ingredients for regulating the immunity of body in food industry.


INTRODUCTION
The genus Rheum (Polygonaceae) contains ∼50 species of perennial herbaceous plants, which are widely distributed and cultivated worldwide (1). The dried rhizomes and roots of Rheum palmatum, Rheum tanguticum, and Rheum officinale have been prescribed as official rhubarbs and are listed in many pharmacopeias, such as the Chinese Pharmacopeia (2), the United States Pharmacopeia (3), and the European Pharmacopeia (4), with a wide range of pharmacological activities (5). In the USA, Europe, and the Middle East, the stalks of some species are usually consumed as fruit instead of vegetables, including R. ribes and R. rhabarbarum (6)(7)(8), which possess valuable natural active ingredients and dietary fiber. R. rhabarbarum is primarily utilized for culinary purposes to prepare various dishes, such as desserts, cakes, mousses, juices, wines, and fruit teas (9).
As two important herbs in this genus, R. officinale and R. tanguticum are widely employed as folk medicines in some Asian countries (10). In China, these two species are cultivated in western regions, such as Gansu, Qinghai, Sichuan, Shanxi, and Guizhou provinces (10,11), and approximately 7,000 tons of these two species are produced every year. Phytochemical research showed that the dried rhizomes and roots of the two species were rich in active compounds, such as anthraquinones, flavonoids, stilbenes, and tannins (5,12,13). These plants are widely employed to treat constipation, abdominal pain, diarrhea, jaundice, and inflammation of some organs, as they are believed to possess antiviral, antiinflammatory, antimicrobial, antioxidant, and antitumor activities (5). Due to the sour taste of these plants, the stalks of the two species were eaten as vegetables by local people and animals in some regions, especially in May and June of each year. However, the nutrients of the aerial parts of rhubarb have not been comprehensively investigated, and the chemical composition and pharmacological activities of these plants have not been elucidated.
In this work, the nutritional values of the stalks of R. officinale and R. tanguticum stalks were compared and evaluated firstly, and their acute toxicities and cytotoxicity were studied. Then, their chemical compositions were analyzed by LC-MSbased metabolomics. The antioxidant and antiinflammatory activities, their potential mechanism of actions, and the structure-function relationship analysis of these two stalks were investigated. This work may help to demonstrate the potential health functionality and nutraceutical applications of the stalks and may help to establish the foundation for future research attempting to develop the two species as food ingredients. Nutritional Value of R. officinale (SRO) and R. tanguticum (SRT) Stalks

Proximate Compositions
The proximate compositions of SRO and SRT stalks were analyzed according to a previously described method (14). Briefly, the contents of crude lipids in two species were determined with a Soxtec 2050 Automatic Soxhlet extraction analyzer (FOSS, Denmark). The total protein contents were obtained by determining the nitrogen content of the two species using a Kjeltec 8200 nitrogen analyzer (FOSS, Denmark). The total dietary fiber in SRO and SRT was measured by the Prosky-AOAC method using a TDF-100A enzymatic kit (Sigma, USA). Finally, the moisture and ash of two species were measured according to the previously described methods (15,16). Three replicates were employed.

Mineral Compositions
The mineral compositions of SRO and SRT were determined according to the previously described AOAC methods 984.27 and 985.01 (modified) (14). First, after digestion with HCl and water, the samples were filtered and transferred to flasks, and the mineral compositions were subsequently determined using inductively coupled plasma mass spectrometry (Agilent 7900, USA). The absorption wavelengths were set at 196.0 nm for selenium (Se), 213.6 nm for phosphorus (P), 213.8 nm for zinc (Zn), 248.3 nm for iron (Fe), 279.5 nm for manganese (Mn), 285.2 nm for magnesium (Mg), 324.8 nm for copper (Cu), 422.6 nm for calcium (Ca), 589.6 nm for sodium (Na), and 766.5 nm for potassium (K). The contents of each mineral are presented as mg/100 g, and the contents of each mineral were calculated according to the analytic curves of standards. Three replicates were employed.

Amino Acids Compositions
Amino acid compositions of SRO and SRT were determined according to a previously described method (17). The samples were digested with 10 mL of 6 M HCl in sealed ampoule at 110 • C for 24 h. Hydrolyzed sample was filtered, and the supernatant was transferred to a bottle to dry at 60 • C. Then 0.2 M sodium citrate buffer of pH 2.2 was utilized to dissolve the samples, and then they were centrifuged. The supernatant was obtained and subsequently subjected to a Biochrom 30 + Automatic Amino Acid Analyzer (Biochrom, England) with an ion-exchange column to determine the amino acid composition. The absorption wavelengths were set at 440 nm for proline and 570 nm for the other amino acids. The standard solution of amino acids (Sigma, USA) was prepared at a concentration of 100 nmol/mL.

Fatty Acids Compositions
The fatty acid compositions of the samples were determined according to a previously described method (18). First, the dried samples of SRO and SRT (20 mg) were mixed with 2.0 mL of an internal standard (undecanoic acid, Sigma, USA) methanol solution and 100 mg pyrogallol in a glass tube. They were heated at 80 • C for 1 h, and then 15% boron trifluoride in methanol (7 mL) was added into the tube to convert fatty acids to esters at 80 • C for 1 h again. When the solutions were cooled, 15 mL of hexane and saturated sodium chloride solution were added, and the hexane layer was extracted and transferred into a vial. The derivatized fatty acids were separated using an Agilent HP-88 column (60 × 0.25 mm, 0.2 µm) and determined by autosampler gas chromatography (GC) with a flame ionization detector. The temperatures were set at 270 • C for the injector and 280 • C for the detector. The oven temperature conditions were set to 100 • C and held for 13 min, increased to 180 • C (10 • C/min) and held for 6 min, subsequently raised to 200 • C (1 • C/min), held for 20 min, and then finally increased to 230 • C (4 • C/min) and held for 10.5 min. Helium was employed as the carrier gas, and the split ratio was 100:1. The fatty acid methyl ester was identified by comparing the GC retention times with those of a mixture of their standards (Sigma, USA), and fatty acids were calculated using fatty acid conversion factors from its methyl ester [AOAC Official Method 996.06 (modified)] (14).

UHPLC-QQQ-MS-Based Metabolomics
Rheum officinale and SRT samples were dried and crushed using a freeze dryer (Scientz-100F, Ningbo) and a mixer mill (MM 400, Retsch, Germany), respectively. After extracting using 1.2 mL of 70% methanol for 12 h at 4 • C, the samples were centrifuged at 12,000 × g for 10 min and were filtered by microporous membrane (0.22 µm pore size; ANPEL, Shanghai, China). Three replicates were employed.
Linear ion trap scans (LIT) and triple quadrupole (QQQ) scans were acquired on a triple quadrupole-linear ion trap mass spectrometer (Q TRAP), AB4500 Q TRAP ultra performance liquid chromatography (UPLC)/MS/MS System, equipped with an ESI Turbo Ion-Spray interface, operating in a positive and negative ion mode and controlled by Analyst 1.6.3 software (AB Sciex). The operating parameters of the ESI source were as follows: ion source, turbo spray; source temperature, 550 • C; ionspray voltage, 5500 V; ion source gas I, gas II, and curtain gas, 30 psi. QQQ scans were acquired in multiple reaction monitoring (MRM) mode with the collision gas (nitrogen) set to 5 psi. The instrument tuning and mass calibration were performed with 10 and 100 µmol/L polypropylene glycol solutions in the QQQ and LIT modes, respectively. A specific set of MRM transitions was monitored for each period in accordance with the metabolites eluting (19,20).

Acute Toxicity
This test was approved by the Institute Animal Care and Use Committee of Lanzhou Institute of Husbandry and Pharmaceutical Sciences, CAAS (SYXK-2014-0002). Mice (18-22 g) were obtained from the Lanzhou Veterinary Institute and were randomly divided into groups (n = 10). After being ground to 200 mesh, SRO and SRT were dissolved in distilled water to concentrations of 2,000-5,000 mg/kg and administered orally to mice. Finally, the mice were observed continuously for behavioral changes for the first 4 h and were subsequently observed for mortality for 24 h after administration.

Cytotoxicity
The popular murine macrophage cell line, RAW 264.7, is often used to initially screen natural products for bioactivity and to predict their potential effect, especially for inflammatory and oxidant response (21). It also plays an important role in inflammatory processes and release various cytokines. This cell line has been widely used to investigate the antiinflammatory and antioxidant activity (22,23); hence, we adopted this cell line. RAW 264.7 cells, obtained from Prof. Zhang's lab., Lanzhou Institute of Husbandry and Pharmaceutical Sciences, CAAS (Lanzhou, China), was cultured in culture medium prepared with 10% FBS and 90% DMEM under a humidified incubator of 5% CO 2 at 37 • C. The cytotoxicity of SRO and SRT against RAW 264.7 cells were evaluated using the ZETA cell counting kit (ZETA Life, U.S.A.). Detailed, RAW 264.7 cells at a density of 1 × 10 5 cells/well (100 µL) were incubated in 96-well-plates for 24 h, and 10 µL of the SRO and SRT (25-200 µg/mL) were added to each well and incubated again for 24 h. CCK-8 agent (10 µL) were added and cultured continuously for 30 min, and the absorbance was measured at 450 nm using a Multiskan Go Microplate Spectrophotometer (Thermo Scientific., U.S.A). DMSO (0.1%) was used as a control, and three replicates were performed.

Antioxidant Activity
In vitro Antioxidant Activity DPPH Radical Scavenging Activity Assay The DPPH radical scavenging activities of SRO and SRT were evaluated (24). Briefly, 100 µL of SRO and SRT (50-1,000 µg/mL) were added into 96-well-plates and mixed with DPPH (100 µL, 0.2 mM) in methanol solution at 37 • C for 30 min in the dark. The samples were measured using a microplate spectrophotometer at 517 nm. Distilled water and ascorbic acid (Vc) were used as a negative control and positive control, respectively. Three replicates were employed.

Hydroxyl Radical Scavenging Activity Assay
The hydroxyl radical scavenging activities of SRO and SRT were evaluated (24). Fifty microliters of SRO and SRT (50-1,000 µg/mL) were mixed with 50 µL of FeSO 4 (9 mM), 50 µL of ethanol-salicylate (9 mM), and H 2 O 2 (3.8 mM) were added to 96-well-plates and incubated at 37 • C for 30 min. The samples were measured using a microplate spectrophotometer at 510 nm. Distilled water was utilized as a negative control, and Vc (20-600 µg/mL) was utilized as a positive control for comparison. Three replicates were employed.

Superoxide Radical Scavenging Activity Assay
The superoxide radical scavenging activities of SRO and SRT were evaluated according to a previously described method (24). Fifty microliters of SRO and SRT (50-1,000 µg/mL) were mixed with 100 µL of NADH-2Na (557 µM), 50 µL of PMS (45 µM), and 50 µL of NBT (108 µM) and incubated at 25 • C for 5 min. The samples were analyzed using a microplate spectrophotometer at 510 nm. Distilled water was utilized as a negative control, and Vc (20-100 µg/mL) was utilized as a positive control for comparison. Three replicates were performed.

Cellular Antioxidant Activity
For evaluating the in vivo antioxidant activities of SRO and SRT, the SOD (superoxide dismutase) and CAT (catalase) activities and MDA (malondialdehyde) content of the RAW 264.7 cells were measured. Briefly, RAW 264.7 cells (2 mL) at a density of 2×10 5 cells/well were incubated in 6-well-plates for 24 h. The cell supernatants were discarded, and then 2 mL of the SRO and SRT (25-250 µg/mL), or the positive control Vc (10 µg/mL) were acceded and cultured for 1 h. Finally, 6 mM hydrogen peroxide (H 2 O 2 ) was used to stimulate RAW 264.7 cells for 24 h, the cells were collected to determine the SOD and CAT activities and MDA content using the relative assay kits (Solarbio, China). For SOD, it was measured by using nitro blue tetrazolium as a substrate. After adding reagents, according to the kit, to supernatant and then incubating at 37 • C for 40 min, color developing agent was added and kept for 10 min, and the absorbance was measured at 560 nm. For CAT, it was determined by using H 2 O 2 as a substrate with CAT assay kit, and the absorbance was scanned at 240 nm. The concentration of MDA was measured at 450, 532, and 600 nm. Three replicates were performed.

Antiinflammatory Assay
The antiinflammatory activities of SRO and SRT were investigated by determining the production of three inflammatory cytokines, nitric oxide (NO), interleukin-1β (IL-1β), and tumor necrosis factor-α (TNF-α) of RAW 264.7 cells induced by lipopolysaccharide (LPS). RAW 264.7 cells (500 µL) at a density of 1 × 10 5 cells/well were incubated in 48-well plates for 24 h, and the cell supernatants were discarded. Then, 500 µL of the SRO and SRT (25-250 µg/mL) were acceded and cultured for 1 h; dexamethasone (DEX) (10 µg/mL) was used as a positive control. Afterwards, LPS (1 µg/mL) was added to each well, and RAW 264.7 cells were stimulated for 24 h to cause inflammation. The cells were collected to determine the levels of NO, IL-1β, and TNF-α using enzyme-linked immunosorbent assay kits from Nanjingjiancheng Bio (NJJCBIO, China). The concentrations were calculated by the different standard curve equations, and the absorbance was measured at a wavelength of 450 nm. Three replicates were performed.

Measurement of NF-κB p65 Content
RAW 264.7 cells (2 mL) at a density of 2 × 10 5 cells/well were incubated in 6-well-plates for 24 h. The cell supernatants were discarded, and then 1 mL of the SRO and SRT (10-100 µg/mL) were acceded and cultured for 1 h. LPS (1 µg/mL) was added to each well for stimulating 24 h to cause inflammation. Finally, the cells were collected to determine the levels of nuclear factor kappa-B (NF-κB) p65 using NF-kB p65 Total SimpleStep ELISA kit (ab176648) (Abcam, China), and the protein concentration was determined using BCA kit (Solarbio, China).

Western Blot Analysis
RAW 264.7 cells (2 mL) at a density of 2×10 5 cells/well were incubated in 6-well-plates for 24 h. The cell supernatants were discarded, and then 2 mL of the SRO and SRT (10-100 µg/mL) were acceded and cultured for 1 h. LPS (1 µg/mL) was used to stimulate RAW 264.7 cells for 24 h, the cells were collected. The protein was separated on SDS-PAGE and transferred to PVDF membranes, which were blocked in 5% bovine serum albumin for 1 h at room temperature and incubated overnight. Immunoblotting was carried out with the following primary antibodies, including anti-nuclear factor-αB (IκB) alpha (ab32518), anti-IαB alpha (phospho S36) (ab133462), anti-NF-κB p65 (ab32536), and anti-NF-αB p65 (phospho S536) (ab76302) from Abcam (Shanghai, China). Relative protein expression was quantified compared with β-actin level.

Measurement of the Cellular Reactive Oxygen Species Production
The level of intracellular ROS production in RAW264.7 cells was determined using 2 ′ ,7 ′ -dichlorodihydrofluorescein diacetate

Measurement of Mitochondrial Membrane Potential
Mitochondrial membrane potential was determined using JC-1 assay kit (Solarbio, China). After 24 h of LPS (1 µg/mL) exposure, the cell culture was discarded and was replaced with JC-1 staining solution (100 ml/well). Then, they were incubated at 37 • C for 30 min in darkness and washed with PBS, JC-1 fluorescence was detected and photographed using a LSM-800 with Airyscan. For red fluorescent of J-aggregates, the wavelength of excitation/emission were set as 525 and 590 nm, respectively, and for the fluoresce green of J-monomers, the excitation/emission were 490/530 nm, respectively.

Statistical Analysis
The results are expressed as the mean ± standard deviation. The significant differences between the two groups were analyzed using unpaired t-tests. Data were analyzed by one-way ANOVA followed by Dunnett's test when the data involved three or more groups using GraphPad Prism 8.0.2.

RESULTS AND DISCUSSION
Nutritional Value of R. officinale and R. tanguticum Stalks

Proximate Compositions
The proximate compositions of SRO and SRT are presented in Table 1. The most abundant components were carbohydrates with contents of 66.65 ± 5.23% for SRO and 59.39 ± 6.36% for SRT followed by ash. However, there were no significant differences between these two groups (P > 0.05). The stalks of the two species contained abundant fiber, exhibiting contents of 11.71 ± 1.66% for SRO and 13.17 ± 2.05% for SRT. From R. ribes and R. rhabarbarum, the natural active ingredients and dietary fiber also were found (7,8). In addition, the protein contents were 6.06 ± 1.12% for SRO and 9.51 ± 0.98% for SRT, with a significant difference being detected (p < 0.05). The lipid contents of SRO and SRT were notably low. Due to the high level of carbohydrates, fiber and proteins, we thought that SRO and SRT could be widely utilized as feed additives to advance the digestion and provide proteins. However, the relationship between the proximate compositions and harvest seasons of SRO and SRT should be investigated further.

Mineral Compositions
As shown in Table 1, SRO and SRT contain abundant minerals with values of 4,275.5 mg/100 g and 5,747.8 mg/100 g samples, respectively. Among these minerals, the contents of Ca were the highest, exhibiting a concentration of 3,280.1 mg/100 g for SRO and 4,280.3 mg/100 g for SRT, which were higher than the recommended daily intakes of 1,000-1,300 mg/day (25). From New Zealand Nutrition databases, we observed that the standard unit in a cup size for rhubarb is 265 g (26). According to this report, the total content of Ca in raw rhubarb was 8.69 g/serving, which was higher than that in trim milk (1.67 g/serving) and standard milk (1.65 g/serving). From the results of this work, these two species could be employed as good food supplements and feed additives to provide Ca for humans and animals to improve bone health and other metabolic processes and to reduce the risk of osteoporosis. Next, the content of K was investigated, and the content in SRT (1,094.27 mg/100 g) was observed to be greater than that in SRO (661.26 mg/100 g) (p < 0.05). The level of minerals in plant can be attributed to the different habitat, plant lifespan, and soil condition. In the karst region of southwestern China, the levels of Ca in senesced leaves of different plant functional types were higher than that of K (27), and the contents of Ca in Ipomoea aquatica, Achyranthes aspera, Aasystasia ganjetica, Enhydra fluctuans, Oldenlandia corymbosa, and Amaranthus viridis also were higher than that of K (28). Further studies should be carried out to investigate the mineral profile, pH, EC, and texture profile of soil in which the samples were cultivated, and to understand the possible reason why the content of Ca was higher than that of K. We also observed the lowest ratio of Na/K with values of 0.072 and 0.062, respectively, and the lower ratio of these two minerals may benefit cardiovascular health. SRO and SRT were subsequently detected with values of 230.05 and 240.67 mg/100 g for P and 43.60 and 61.62 mg/100 g for Mg, respectively, which are important for enabling antioxidant enzymes in vivo to protect the body from diseases. However, the contents of Zn and Cu were the lowest, and Se was not detected. Mineral elements play important roles in improving such body functions as bone and tooth performance, immune balance, and cardiovascular function.

Amino Acids Compositions
Amino acids are important for the formation of proteins, and their profile determines protein quality (29). In this work, we observed that the two species contained high levels of important amino acids, and fifteen and sixteen amino acids were detected from SRO and SRT, respectively. The total contents were 6,024.42 mg/100 g for SRO and 9,829.38 mg/100 g for SRT, respectively. These two species provided abundant nutrient substances to maintain the balance of the diet. Essential amino acids, which were only absorbed and obtained from the diet, are necessary for human health and cannot be synthesized by the body (30), and SRT has a higher content of 3,228.41 mg/100 g than SRO (1,835.70; P < 0.05). Glutamic acid was the most abundant amino acid, exhibiting contents of 1,213.16 mg/100 g for SRO and 1,630.08 mg/100 g for SRT. This result provide a clue that the stalks of two Rheum species could be applied in recipes where umamai flavor is desired. The level of glutamic acid is also highest among other amino acids in Portulaca oleracea, Anoplophora chinensis, and Tenebrio molitor (17,18). The most abundant amino acids were leucine (401.33 vs. 800.83 mg/100 g, p < 0.05), valine (340.07 vs. 601.35 mg/100 g, p < 0.05), and lysine (330.09 vs 581.31 mg/100 g, p < 0.05) for the two species. In SRO, six essential amino acids were detected, and for SRT, seven amino acids were identified, including methionine. In addition, we also observed that the two species contained the most abundant hydrophobic amino acids, which have bioactive functions for reducing the risk of disease (31). These results indicated that SRO and SRT may represent useful sources of amino acids and food supplementation to protect the body by enhancing immune function.

Fatty Acids Compositions
As shown in Table 1, SRO and SRT were rich in fatty acids, exhibiting contents of 4,249.22 mg/100 g and 4,753.57 mg/100 g, respectively. In those studies, the contents of monounsaturated fatty acids (MUFAs) and polyunsaturated fatty acids (PUFAs) were determined to be 736.05 mg/100 g and 2,244.32 mg/100 g for SRO and 552.23 mg/100 g and 2,844.69 mg/100 g for SRT, which played important roles in maintaining the high quality of the human diet and food (17,25) and could be employed to decrease the risk of heart disease in humans (32). Low-dose polyunsaturated fatty acids and omega-3 fatty acids supplementation attenuates inflammatory markers (33,34). Humans cannot synthesize omega-3 and omega-6 fatty acids, which are only obtained from food (32,35). Currently, the Western diet contains abundant omega-6 and a low content of omega-3 fatty acids, which leads to a high omega-6/omega-3 ratio (10-20:1) and induces the progression of some chronic autoimmune and degenerative diseases (36). Hence, low omega-6/omega-3 ratios, i.e., 1:1-4:1, are recommended and would be beneficial to human health for medicinal uses (37). In our work, the low and proper ratios of omega-6/omega-3 were also found with values of 2.82:1 and 2.98:1, and the essential fatty acids αlinolenic (18:3) and linoleic acids (18:2), and the total contents of fatty acids were 2,141.79 mg/100 g for SRO and 2,529.37 mg/100 g for SRT. This result suggested that the stalks of the two species may have cardioprotective, antiinflammatory, and antioxidant activities and may also reduce the risks of some cancers (38). The two fatty acids eicosapentaenoic acid and docosahexaenoic acid, two precursor of arachidonic acids, presented the cardiovascular protective effect and also could reduce the risks of some cancers (38). C18:1 could prevent insulin resistance and may have an antiinflammatory effect (39); it was also identified from SRO and SRT. This result indicated that SRO and SRT contained essential fatty acids and could be employed to prevent chronic cardiovascular diseases and regulate the immunity.

UHPLC-QQQ-MS-Based Metabolomics
Consumption of traditional vegetables or plants can assist in the prevention of chronic disease development, as they contain various bioactive compounds that exhibit multiple health benefits (40). Using LC-MS-based metabolomics, the chemical compositions were analyzed (Figure 1), and 770 and 769 compounds were identified from SRO and SRT, respectively (Figure 2A). The chromatographic pictures are presented in Figure 1, and the data on tentative chemical compounds and molecular formula of compounds are described in Supplementary Table 1.
Subsequently, the relative ion intensity of secondary metabolites from SRO and SRT were analyzed and compared. Although the numbers of each kinds of compounds from SRO and SRT were similar, the relative intensity was different. Results showed that the highest content of SRO was phenolic acids, next were anthracene derivatives, organic acids, alkaloids and flavones; however, the most content of SRT is flavones, and others were phenolic acids, organic acids, alkaloids, and anthracene derivatives ( Figure 2D). Table 2 presented that except the common compounds citric acid, isocitric acid, and caffeic acid, the main compounds are different between SRO and SRT. The different contents of SRO and SRT affect their flavor and taste, SRO is more bitter than SRT. The different components and contents would cause the different pharmacolog ical activities of SRO and SRT. Considering the high levels of flavones and phenolic acids, we thought SRT may be more suitable to be food additive than SRO.

Toxicity
To ensure the safety of the two species, we evaluated the acute toxicity in mice. After oral administration of SRO and SRT (5,000 mg/kg) to mice, no mice died or exhibited any acute behavior, and the LD 50 was determined to be more than 5,000 mg/kg. From Figure 2E, we can see that the two stalks presented the weak cytotoxicity against RAW 264.7 cells. Even at 1,000 µg/mL, the cell viability of two plants were more than 75 %. This result  (F-H), the different letter represents the significant difference, p < 0.05; for (I-K), ### represent the significant difference between control group and model group, p < 0.001; *p < 0.05 represent the significant difference between model group and drug-treated group; **p < 0.01; *** for p < 0.001].
suggested that at the concentration employed in this work, the two species were safe.

In vitro Antioxidant Activity
The antioxidant activities of the two species were evaluated and compared using three different assays. From Figure 2F, we can see that SRO presented stronger radical scavenging activity than SRT in the DPPH assay (p < 0.05), which was employed to investigate the reducing power of agents based on an electron transfer reaction. The EC 50 values were 205.13 and 497.03 µg/mL, respectively, and the positive control Vc was 19.33 µg/mL. However, the superoxide radical scavenging activity of SRT was better than that of SRO (p < 0.05; Figure 2G). This result indicated that the two species could alleviate the cellular damage induced by the superoxide radical and could contribute to aging and some degenerative diseases.
Hydroxyl radicals can cause severe cell death or damage by crossing cell membranes and subsequently reacting with biomacromolecules (24). However, the hydroxyl radical activity of the two species was weak, and the EC 50 values were more than 1,000 µg/mL (Figure 2H). These results suggested that the two species have antioxidant activity in vivo, especially for radical scavenging activity in the DPPH assay and superoxide radical scavenging activity.

Cellular Antioxidant Activity
Macrophages stimulated by H 2 O 2 would cause the enrichment of superoxide radicals and other radicals. SOD and CAT as enzymatic antioxidant defenses provide first-line cellular protection contributing to prevent cellular damage and maintain a balance between free radical production and oxidative stress after being stimulated by oxidative stress (41,42). Most plant antioxidants are investigated for the effects by evaluating directly the enzymatic activity of endogenous antioxidants (43).
A ubiquitous cellular oxidant, H 2 O 2 , is mostly produced by the peroxisome. Figure 2I showed that after the treatment of H 2 O 2 , the activity of SOD was significant inhibited (P < 0.001); however, SRO and SRT dramatically activated SOD activity. At the concentrations of 50, 100, and 250 µg/mL of the samples, the SOD activities were 41.47, 51.31, and 54.24 U/mg Prot (protein) for SRO, 45.45, 50.47, and 58.01 U/mg Prot for SRT (all P<0.001); and the SOD activity was 50.37 U/mg Prot for the positive control (10 µg/mL). In addition, after treatment with SRO and SRT, CAT activity was increased in RAW264.6 cells induced by LPS in a dose-dependent manner ( Figure 2J). The activation in SOD and CAT activities contribute to scavenge free radicals and other radicals from macrophages induced by H 2 O 2 . As a marker of protein and lipid oxidation, the cellular MDA levels would be released when they respond to oxidant stress (44). From Figure 2K, we can see that the two stalks could reduce the increased production of MDA in cells stimulated by LPS in a dose-dependent manner, respectively. Especially for SRT, the MDA contents were significant lower than model group (p < 0.001). These results indicated that SRO and SRT have the good cellular antioxidant activity.

Antiinflammatory Activity
When stimulated by LPS and other foreign substances, macrophages play an important role in inflammatory processes and release various cytokines. As one of a wellknown inflammatory cytokine, NO is released by activated macrophages and has a cytotoxic effect through the formation of peroxynitrite with superoxide anion and inflammatory response (45). Figure 3A demonstrated Figure 3B). TNF-α plays an important role in inflammation formation. Compared with the model group (52.5 ng/L), the levels of TNF-α were markedly decreased when coincubated with SRO and SRT. Especially for 100 and 250 µg/mL (all p < 0.01; Figure 3C). The results showed that the two stalks could decrease the production of proinflammatory cytokines stimulated by LPS in a dose-dependent manner, and SRT presented better antiinflammatory activity than SRO.

Measurement of NF-κB p65 Content
NF-κB is a nearly ubiquitous pathway responsible for mediating DNA transcription that is involved in apoptosis, proliferation, and metastasis (46). When a cell receives any of a multitude of extracellular signals, NF-κB rapidly enters the nucleus and activates gene expression. Figure 3D demonstrated that compared with model group (77.83 ng/L), SRO and SRT could decrease the levels of NF-κB p65 in RAW264.7 cells induced by LPS in a dose-dependent manner. Especially for high concentration (100 µg/mL), the contents were 57.27 ng/L for SRO and 54.90 ng/L for SRT (p < 0.001), respectively. These results indicated that the two stalks presented the antiinflammatory activity by decreasing the level of NF-κB and then inhibiting the activation of NF-κB pathway in RAW264.7 cells induced by LPS.

Western Blot Analysis
The western blot test was conducted to investigate the expression of IκB and NF-κB in NF-κB signaling pathway for explaining the possible antiinflammatory mechanism. Compared with the model group, SRO and SRT reduced the expression ratio of phosphorylation of IκB/IκB and the phosphorylation of NF-κB/NF-κB of RAW264.7cells induced by LPS in a dosedependent manner, especially at the concentrations of 25, 50, and 100 µg/mL (p < 0.001; Figures 3E-I). Cytokines mediate the rapid activation of NFκB through activation of the IKK complex, lead to subsequent phosphorylation and degradation of the inhibitory IκB proteins, then stimulate activation of NFκB, and transcription of target genes (47). A key step for controlling NF-κB activity and its phosphorylation is the regulation of the IkB-NF-κB interaction. Hence, we thought that SRO and SRT presented the antiinflammatory activity by inhibiting NF-κB signaling pathway and decreasing the cytokines production.

Measurement of the Cellular ROS Production
The oxidant and inflammatory response to exogenous substance would increase cellular ROS production and accumulation. From Figure 4A, we can see that compared with the control group, the intensity of green fluorescence stained by DCFH-DA in model group was significantly enhanced, and LPS could generate ROS production of RAW264.7 cells. However, after the treatment of SRO and SRT, the green fluorescence became vague, respectively. The intensity of two groups were decreased in a dosedependent manner, especially for 100 and 250 µg/mL. These results illustrated that the two stalks could inhibit the ROS production and accumulation of RAW264.7 cells induced by LPS ( Figure 4B).

Measurement of the Cellular MMP
The generation and accumulation of ROS cause the reduction of MMP and the increase of cell membrane permeability, also resulting in cell death. In this test, JC-1 kits were used to evaluate the effect against MMP. JC-1 staining showed an aggregated pattern of energized mitochondria (red fluorescence) and a monomeric pattern for depolarized mitochondria (green fluorescence). As shown in Figure 5, compared with control group, RAW264.7 cells induced by LPS presented the remarkably lower red fluorescence and higher green fluorescence, resulting in the loss of MMP. However, after the treatment of SRO and SRT at 100 µg/mL, the intensity of red and green fluorescence were reversed, respectively, and RAW264.7 cells showed higher red fluorescence and lower green fluorescence. These results indicated that the two plants have the significant inhibitory effect on MMP depletion induced by LPS.

The Structure-Function Relationship of Food Bioactives and Ingredients
The applications and biological activities of foods and plants and their ingredients depend on the structure of those constituents, and the investigation of structure-function relationship of food bioactives and ingredients has attracted many people's attention.
In this work, the nutritional values of the two stalks were found; however, SRT has higher properties than SRO, such as fiber, proteins, Ca, amino acids, fatty acids, and other key indexes. In addition, the different secondary compositions of the two stalks resulted in different activities. Similar to most compositions of SRO and SRT, phenolic acids are a key class of dietary polyphenols and natural antioxidants with healthy benefits, and they are precursors of other significant bioactive molecules regularly used for therapeutics, cosmetics, and food industries (47, 48). Hence, the two stalks could be used as healthy foods due to the high nutritional values and phenolic acids. In addition, flavones have a significant positive effect on diseases related to oxidative stress and inflammatory by regulating multiple cell signaling pathways and alterations in multiple cellular signaling pathways, such as NF-κB signaling pathway (49), and anthracene derivatives have the cytotoxicity and antiviral properties (50). Hence, SRT has better antioxidant and antiinflammatory activities than SRO. Further studies indicated that the two stalks alleviated the oxidant stress and inflammatory response by inhibiting NF-κB signaling pathway in a dose-dependent manner. The overexpression of NF-κB and IκB were blocked by SRO and SRT, the damage of MMP and the accumulation of ROS were recovered, and then the production of cytokines was inhibited.

CONCLUSION
R. officinale and R. tanguticum stalks have good nutritional value. The two species are rich in minerals, essential amino acids, and essential fatty acids. They possess abundant fiber and proteins and are used to regulate the immunity and enhance the immune function of body. Since the secondary and its levels between SRO and SRT were different, the bioactivies of the two stalks were various. SRT contains more flavones and phenolic acids with the better antiinflammatory activities than SRO. The further studies showed that the two stalks alleviated inflammatory response by inhibiting NF-κB signaling pathway. The antioxidiant activity also was observed. In addition, the good safety also was observed. The two stalks could be regarded as a potential new source of nutritional ingredients for the food industry. Further research should be conducted to elucidate the relationship between harvest season and active ingredients, and the digestibility and safety of the stalks of these species should also be investigated in the future.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.