Extraction, Partial Identification and Bioactivities of Total Flavonoids from Carex meyeriana Kunth

Corresponding Author: Hongli Zhou School of Chemistry and Pharmaceutical Engineering, Jilin Institute of Chemical Technology, Jilin 132022, PR China Tel: +86-432-62185246 E-mail: zhouhongli@jlict.edu.cn Abstract: To date, knowledge associated with extraction, chemical constituents and bioactivities of Total Flavonoids from Carex meyeriana Kunth (TFCMK) that remains unclear. Therefore, this paper was designed to optimize the ethanol reflux extraction method by response surface methodology and determine the chemical constituents by Liquid Phase-Mass Spectrometry (HPLC-MS); additionally, the bioactivities including antioxidant, antibacterial and hemolysis test were also explored. The maximum TFCMK extraction yield of 6.248%±0.016% was obtained under the optimal conditions as follows: Ethanol concentration 50%, ratio of liquidto-solid 31 mL/g, extraction time 91 min and extraction temperature 71°C. Meanwhile, 9 compounds of TFCMK were identified by using LP-MS for the first time. In addition, TFCMK showed higher antioxidant activity compared with vitamin C, as evidenced by the lower EC50 or IC50 of TFCMK (reducing power assay: 0.125±0.025 mg/mL, DPPH radicals: 0.28± 0.031, ABTS+ radicals:0.012 ± 0.005, hydroxyl radicals: 0.028±0.008 mg/mL; The EC50 or IC50 of VC 0.156±0.012, 0.059±0.019, 0.068±0.024 and 0.032±0.007 mg/mL, respectively.). Broth microdilution assays had demonstrated that the minimal inhibitory concentrations (MICs) of TFCMK on Bacillus pumilus, Bacillus subtilis and Escherichia coli were 0.625±0.0032, 1.250±0.0216 and 5.000±0.0205 mg/mL, respectively. Besides, the hemolysis rate (less than 5%) in vitro implied that TFCMK possessed a good blood compatibility. In summary, this paper provides evidences that TFCMK might be applied as an antioxidant and antibacterial agent in the pharmaceutical and chemical industries.


Introduction
Carex meyeriana Kunth (CMK), a perennial herb belonging to a genus Cares species in the family Cyperaceae, is widely distributed in China including Heilongjiang, Jilin, Inner Mongolia and Sichuan (Reznicek, 1990) and other countries such as Far East Russia, Mongolia, Korea and Japan (Yang et al., 2017a). The shoe pads, waist supports, mattress and other daily necessities made of CMK as the main material can keep warm and actively absorb moisture and they are hard to self-corrode (Luo and Zhang, 2008). CMK is a valuable germplasm resource and CMK together with the Panax ginseng C.A. Mey and pelt of marten is called "three treasures in Northeast China" (Yun et al., 2016). In our previous study, polysaccharides (Hu et al., 2018) and essential oils (Cui et al., 2018) of CMK were explored in the details. However, there has been no report on its flavonoids and their extraction process and chemical composition, especially bioactivities.
The traditional ethanol reflux extraction method (Yang et al., 2016) is nondestructive to the target product and featured by less impurities, mildew proofness and easy preservation. In order to improve the extraction yield of flavonoids, Response Surface Methodology (RSM) has been widely employed in ■■ various extraction processes to determine the best experimental parameters, evidenced by the true limit state surface and statistical significance of the independent variables (Wang et al., 2017). Time-of-Flight (TOF)-MS and Liquid Chromatography (LC) can be used to analyze the quality of complex compounds and to better separate flavonoids (Dong et al., 2017). The antioxidant activity of flavonoids is usually evaluated through the reduction ability of Fe 2+ , scavenge DPPH free radicals, ABTS + free radicals and hydroxyl free radicals.
The optimum extraction process of TFCMK by ethanol reflux was determined through RSM and the chemical constituents were studied by Liquid Chromatography-Mass Spectrometry (LC-MS); more importantly, the antioxidant, antibacterial activity and lysis erythrocytes in vitro were investigated. Taken together, this paper is the first time to systematically explore the basic research of TFCMK, thereby providing a scientific basis for the development and utilization of TFCMK.

Materials
Carex meyeriana Kunth (CMK) samples were collected from a wetland in the northern suburbs of Jiamusi, Northeast China (45°56′N, 129°29′E). The ground part of CMK was identified by Prof. Guangshu Wang (Jilin University, China). The herbarium samples  were deposited at School of Chemistry and Pharmaceutical Engineering, Jilin Institute of Chemical Technology, Jilin, China. TFCMK were placed in a semi-dark room for natural drying. Diphenyl Picrylhydrazinyl (DPPH) and 2, 2'-azinobis (3ethylbenzothiazoline-6-sulfonic acid) (ABTS) (both Aladdin, St. Shanghai, China) and L-Ascorbic acid (VC, Hushi, St. Shanghai, China) were used. All reagents and solvents were the analytical grade. All the bacterial strains including Bacillus subtilis ATCC 6633, Bacillus pumilus ATCC 700814 and Escherichia coli ATCC 25922 were purchased from Beijing Zhongke Quality Inspection Biotechnology Co. Ltd. The hemolysis of TFCMK was assessed by evaluating the lysis of human Red Blood Cells (hRBCs) obtained from the General Hospital of Jilin Chemical Group Corporation (Jilin, Jilin, China).

Extraction Experiment
Five gram of the CMK crushed was placed in a 250 mL round-bottom flask together with the extracting solvent, followed by extraction at appropriate ethanol concentration, liquid-to-material ratio, extraction time and temperature. After that, the samples were filtered and dried on a freeze dryer.

Determination of Total Flavonoid Content
The content of total flavonoids was determined by sodium nitrite chromogenic method . The standard reserve solutions of rutin were accurately prepared at 510 nm on an ultraviolet spectrophotometer. Results showed the linear equation was Y = 14.548x+0.0013 (R 2 = 0.9997) in the linear range of 0-0.06 mg/mL. The extraction rate of TFCMK was calculated as follows: where, C, V and M are the concentration of TFCMK (mg/mL), dilution multiple (mL) and quality of TFCMK (g), respectively.

Optimization of Extraction Technology by BBD-RSM
According to RSM and Box-Behnken central composite experiment Design (BBD), the four factors of ethanol concentration (X 1 , %), liquid-to-material ratio(X 2 , mL/g), extraction time (X 3 , min) and extraction temperature (X 4 , °C) were optimized and defined as three horizontal levels of high (+1), medium (0) and low (-1) ( Table 1). The factors were coded as follows (Mou et al., 2017): where, Xi and x i are the coded and actual values of the independent variable, respectively, x 0 is the actual value of the independent variable at the central point and ∆x i is the step of the variable.
With the extraction rate of TFCMK as the response value, the BBD experiment was designed on Design-Expert V 8.0.6.1 Trial. The four-factor and three-level response surface analysis was established involving 24 groups of factorial experiments and 5 groups of center experiments ( Table 2).
The quadratic polynomial regression equation for optimization was fitted as follows : where, Y is the predicted extraction rate of TFCMK and β 0 is the constant coefficient; the first-order coefficients of β i , β ii and β ij are X i ; the second-order coefficients and the interaction influence are the experimental errors (random).

Chemical Composition Analysis of TFCMK by HPLC-ESI-Q-TOF MS/MS
The flavonoids were dissolved in a small amount of acetonitrile into a reserve solution at a certain concentration.    The MS conditions in negative mode were as follows: injection voltage of the ESI source at 4000 V, dry gas volume flow rate at 1 L/min, atomization gas pressure 30 psi, auxiliary gas pressure 60 psi, interface continuous heating and a TOF-MS detector (Agilent Technologies, Santa Clara, CA, USA). The MS ion source was operated at 300°C and a scan mass range of m/z 100-1500.

Antioxidant Activity
Four systems were used to evaluate antioxidant activity of TFCMK in vitro and VC was served as the positive control. The reducing power was determined following a reported protocol with TFCMK and VC solutions (0.004, 0.008, 0.016, 0.032, 0.068 mg/mL) (Mendes et al., 2011). The results were expressed as the effective concentration at 0.5 absorbance unit (EC 50 ).

Antibacterial Activities
The Minimum Inhibitory Concentration (MIC), defined as the lowest concentration of TFCMK to inhibit bacterial distinct growth, was detected by a broth microdilution method  with some modifications. B. subtilis, B. pumilus and E. coli strains were cultured in 96-well plates. The concentration decreased by 2-fold to dilute the TFCMK into a gradient of 0.625-5.000 mg/mL. Each well was added with the standard suspension of one of the strains (10 8 CFU/mL) and the final anhydrous ethanol in each well was used to dissolve the TFCMK, followed by incubation at 37°C for 24 h. Chloramphenicol (0.800-5.000 mg/mL) was used as a positive control and the solution containing anhydrous ethanol without TFCMK was used as a negative control. All assays were repeated in triplicate.

Hemolysis Rate
The hemolysis rate of TFCMK assay was determined by a protocol (Chen, 2016).
The series of flavonoid solutions equivalent to the concentrations in the TFCMK solutions (0.10, 0.25, 0.50, 1.00, 1.50, 2.00 mg/mL) were used as the experimental group, distilled water and normal saline were used as positive and negative controls respectively. The hemolysis rate of TFCMK was calculated as follows: where, A s , A c and A p are the absorbance of the test sample, negative control and positive control, respectively.

Statistical Analysis
The experimental data were statistically tested by Analysis of Variance (ANOVA) by IBM SPSS 24 and expressed as mean ± standard error. The significant data at p<0.05 were calculated by Duncan's multiple range tests.

Effects of Ethanol Concentration on Extraction Rate of TFCMK
The effects of ethanol concentration (30%, 40%, 50%, 60%, 70%) were investigated under the CMK dosage of 5 g, temperature at 50°C, liquid-to-solid ratio of 20 mL/g and extraction time of 60 min. The results are shown in Fig. 1a. The extraction rate of TFCMK were increased with the rise of ethanol concentration and maximized at 50%. Because as the ethanol concentration increased, the polarity of the solvent decreased and the solubility of TFCMK was intensified, which were beneficial to the leaching of flavonoids. Whereas, at higher ethanol concentration, some higher polarity flavonoids may be precipitate due to poor solubility, which in turn affected the flavonoid extraction (Yang et al., 2017b).

Effects of Liquid-to-Solid Ratio on Extraction Rate of TFCMK
The effects of liquid-to-solid ratio (10, 20, 30, 40, 50 mL/g) were investigated under CMK dosage of 5 g, temperature at 50°C, extraction time of 60 min and ethanol concentration of 50%. The extraction rate maximized at the liquid-to-solid ratio of 30 mL/g (Fig.1b). because with the continuous increase of solvent ethanol, the concentration gradient between the solid and solvent was considered as mainly the driving force for mass, so that the flavonoid substances were dissolved more effectively (Xie et al., 2017). However, with above the liquid-to-solid ratio of 30 mL/g, the content of flavonoids was limited and more impurities were extracted. As a result, the extraction rate decreased.

Effects of Extraction Time on Extraction Rate of TFCMK
The effects of extraction time (30, 60, 90, 120, 150 min) were investigated under the CMK dosge of 5 g, temperature at 50°C, liquid-to-solid ratio of 30 mL/g and ethanol concentration of 50%. With the prolongation of extraction time, the extraction rate of TFCMK increased continuously and maximized at 90 min (Fig.1c). Because as the extraction time was prolonged, the plant cell walls were destroyed and the release quantity of flavonoids increased, which led to an increase in the extraction rate of TFCMK. However, when the extraction time was too long, promoting the degradation of polyphenols, thereby reducing the flavonoid extraction (Åahin et al., 2017).

Effects of Extraction Temperature on Extraction Rate of TFCMK
The effects of extraction temperature (40, 50, 60, 70, 80°C) were investigated under the CMK dosage of 5 g, liquid-to-solid ratio at 30 mL/g, extraction time of 90 min and ethanol concentration of 50%. With the rise of extraction temperature, the extraction rate of TFCMK increased and maximized at 60°C (Fig. 1d). At above 60°C, the extraction rate began to decline. Due to the fact that low temperature dose not facilitate the diffusion of molecules, leading to the incompletion of extraction, while high temperature is prone to cause the degradation of bioactive constituents, both of which can induce the decrease of extraction yield (Ya et al., 2017).

Model Fitting Analysis
Based on the experimental data of where, X 1 , X 2 , X 3 and X 4 are the coded variables for the ethanol concentration (%), liquid-to-solid ratio (mL/g), extraction time (min) and extraction temperature (°C), respectively.

Fitting Model
ANOVA confirms the response surface quadratic model has certain sufficiency and adaptability. The high F value (42.016) and low P value (<0.0001) ( Table 3) prove that the model has obvious significance and is very ideal. Comparison of P values showed that the order of single factors affecting the model was ethanol concentration (X 1 )> liquid-to-solid ratio (X 2 )> extraction temperature (X 4 )> extraction time (X 3 ). The coefficient of determination R 2 = 0.9768 suggests the model is reliable and can be used to interpret 97.68% of the data. According to the adjustment of the model, the adj-R 2 = 0.9535 indicates this model can predict most of the extracted variation (>96%). The lack of fit (p = 0.1779) suggests the model is insignificant. Thus, the experimental results are obviously correlated with the theoretical values derived from the fitting of the corresponding polynomials. The model prediction is reliable and repeatable in the ranges of the parameters allowed. The ANOVA shows the model precision is 19.340>4 and the coefficient of variation is 11.36% <30%, indicating the model has high reproducibility, high accuracy and reliability (Qu et al., 2016). The model is suitable for analysis and prediction of the extraction results of TFCMK.

Response Surface Analysis
The specific response target and the corresponding independent variables were investigated via graphic analysis of response surface, which consists of 3D response surface plots that clearly reflect the impact of various factors on the response target. The 3D response surface plots comprehensively reflect the whole region of predictive behavior (AnnGiovannitti-Jensen and Myers, 1989) and can be obtained by finding the interaction of various factors in the reaction process. The effect of each factor on the extraction rate of TFCMK can be judged from the flexibility of the response surface plots (Fig. 2).
Firstly, the increase of ethanol concentration and the liquid-to-solid ratio, the extraction rate rises significantly and the result reversed if the ethanol concentration continues to increase (Fig. 2a). Furthermore, the interaction of these two factors significantly affects the extraction rate of TFCMK. Secondly, with the increase of ethanol concentration combined with the intermediate extraction time and extraction temperature, the change of extraction rate is relatively gentle (Fig. 2b and c), indicating the interaction of ethanol concentration with extraction time or extraction temperature little affects the extraction rate. (a) X 1 and X 2 , (b) X 1 and X 3 , (c) X 1 and X 4 , (d) X 2 and X 3 , (e) X 2 and X 4 and (f) X 3 and X 4 . Note: X 1 : Ethanol concentration (%), X 2 : liquid-to-solid ratio (mL/g), X 3 : Extraction time (min), X 4 : Extraction temperature (°C)

■■
Thirdly, the increase of liquid-to-solid ratio, the variation of extraction rate is not obvious and its interaction with the extraction time or extraction temperature does not obviously affect the response value ( Fig. 2d and e). Finally, the increasing extraction time combined with the optimal extraction temperature, the response value is not obvious and has no significant effect on the model (Fig. 2f). To sum up, these results are consistent with the ANOVA.

Verification of the Models
The optimum extraction parameters (encoding extraction parameters: 0.060, 0.049, 0.022, 0.035) were obtained on software through further analysis of the fitting linear equation. Ethanol concentration 49.40%, liquid-to-material ratio 30.49 mL/g, extraction time 90.67 min and extraction temperature 70.35°C were obtained from Design-Expert software, where the theoretical extraction rate of TFCMK was 6.248%. However, considering the operability in actual production, the optimum extraction parameters can be modified as follows: ethanol concentration is 50%, liquid-to-material ratio is 31 mL/g, extraction time is 91 min and extraction temperature is 70°C. The extraction rate of TFCMK is 6.240% ±0.04% from three repeated experiments under the optimum extraction conditions. The final experimental results are basically consistent with the predicted values on the software (relative error = 0.128%<2%), indicating our model is suitable for the extraction of TFCMK and theoretically underlies the future development of CMK.

Chemical Composition Analysis of TFCMK by LC-MS/MS
The LC chromatogram at 254 nm and the total ion chromatogram (Fig. 3) were analyzed (Table 4). The compound 1 (t R = 6.875 min) was identified as Catechin with the literature (Liu et al., 2009) (Zhu et al., 2017).
The compound 4 (t R = 9.049 min) was identified as orientin by comparison with literature (Tahir et al., 2012).  , On this basis, some fragments of quinic acid group were separated again, the fragments at m/z 192 were lost and the fragments of caffeoyl group remained, resulting in ion fragments at m/z 161. It indicated that two caffeoyl groups existed in the chemical structure.
The compound 9 (t R = 22.369 min) was identified as luteolin with literature (Toplan et al., 2017). The molecular formula is C 15 H 10 O 6 . The molecular ion peak [M−H−28] − at m/z 285.0392 and 257 were obtained by the loss of −CO from the fragment ion peak at m/z 257 and 229. The fragment ions m/z 151 and 133 were obtained by RDA cleavage of fragment ions m/z 285. It is presumed that the compound was luteolin and the base ion fragments at m/z 199. Other ions fragment were m/z 175 and 151. In addition base peak at m/z 133 was obtained by the RDA cleavage of the structure.

■■
The compounds of 10-12 in Table 4 were found to contain m/z 285 [M−H] − by MS/MS fragments, it is speculated that they belongs to such luteolin flavonoids. At present, there are no reports of using LC-MS/MS to analyze the flavonoids of TFCMK and with many isomers. The two peaks at t R = 23.560 min were not separated so that they could not be resolved. Therefore, the analytical constituents was need to be improved.

Antioxidant Activity
Antioxidants can prevent hydroxyl radicals from destroying biological macromolecules. The ability to scavenge hydroxyl radicals is directly related to the antioxidant capacity (Cacciuttolo et al., 1993). The antioxidant experiment results of the four systems includes the reducing power scavenging activity, DPPH· scavenging activity, ABTS + and Hydroxyl radicals of both TFCMK and VC were enhanced with the increasing concentration ( Fig. 4a-d). The EC 50 or IC 50 values of TFCMK were 0.125 ±0.025, 0.28±0.031, 0.012±0.005 and 0.028±0.008 mg/mL and the EC 50 or IC 50 values of VC 0.156±0.012, 0.059±0.019, 0.068±0.024 and 0.032±0.007 mg/mL, respectively. Results suggest TFCMK have significant antioxidant activity, which showed stronger in the system of reducing power and scavenging hydroxyl radicals. Firstly, the possible reason mainly is the phenolic hydroxyl structures of flavonoids play an antioxidant role through the reduction of phenolic hydroxyl. All the 9 compounds have phenolic hydroxyl groups and benzene ring structure. The reaction of phenolic hydroxyl with free radicals produces a stable structure and terminates the free radical chain reaction. Phenolic hydroxyl radicals are mainly produced by ionizing hydrogen atoms, neutralizing oxygen free radicals and combining with ionized flavonoids to form dimers, which prevent reverse binding and thus scavenge radicals (Boulanouar et al., 2013;Karmakar et al., 2013). Secondly, the effect of double bonds in the structural nucleus of flavonoids is also crucial to the oxidation resistance. Theoretically, double bonds act as electron delocalization and lengthen the conjugation system, which is conducive to the formation of more stable free radicals after the electron loss in flavonoid nuclei and then interrupts the chain reaction (Acheampong et al., 2016).

Antibacterial Activity
TFCMK showed significant antibacterial activity against B. pumilus, B. subtilis and, E. coli according to the MIC with in 0.625−5.000 mg/mL (Table 5). Among them, the MIC of B pumilus was 0.625 mg/mL, which exhibited a higher inhibition effect, as evidenced by the MIC of Chloramphenicol was less than 0.8 mg/mL. The cooperation effects of the above 9 compounds can not be excluded. The a ntibacterial mechanism of TFCMK may be through its own hydrophobic  in teraction with the lipid bilayer of the bacterial cell membrane and it can affect the stability of the cell membrane to achieve bactericidal effect (Taylor, 2013;Veloza and Mantillamuriel, 2014). The antimicrobial effect of flavonoids is mainly due to the destruction of bacterial cell membrane by their hydrophobic benzene ring, thus achieving the purpose of sterilization. The antibacterial l mechanisms of TFCMK need further research. With the widespread application and abuse of antibiotics, more and more drug-resistant strains are found in clinic. The development of natural plant antibacterial agent with the lower MIC has become a new method to solve the problem of bacterial resistance.

Hemolysis Analysis
To test whether TFCMK possessed a blood compatibility, the hemolysis test which was used to detect degree or proportion of damage to red blood cells, was carried out . According to the International Organization for Standardization (ISO), the hemolysis rate ≤5% suggests the sample meets the hemolysis requirement of medical treatment and the hemolysis rate >5% indicates the occurrence of hemolysis. The results are shown in Fig. 5. In the concentration range of 0.1-2.0 mg/mL, the hemolysis rate declines, but is less than 5%, indicating TFCMK meets the hemolysis requirement of medical treatment in this concentration range (Sobrinho et al., 2016). Thus, it is affirmed that TFCMK can be applied for pharmaceutical and chemical industries.

■■ Conclusion
The maximum extraction yield of TFCMK was observed under the optimal condition and the actual yield reached at 6.24%±0.04%. Meanwhile, 9 compounds of TFCMK firstly identified by LC-MS/MS. In addition, TFCMK possessed a higher antioxidant activity and antibacterial activity, quantified by the IC 50 and MICs; in vitro assay, hemolysis rate was less than 5%. The above data indicates that TFCMK might be applied as an antioxidant and antibacterial agent in the pharmaceutical and chemical industries. Herein, this study underlies further research on CMK in the future.