Optimization of the Ultrasonic-assisted Extraction Process for Protocatechuic Acid from Emilia sonchifolia DC by Response Surface Methodology

Wen CHEN Xiangjun WANG Lingyan LI Zhewen ZHAI Xiaoling HUANG Wubing RAO Chenghan ZOU Tengtong LI

Abstract ［Objectives］ This study was conducted to optimize the extraction process of protocatechuic acid from Emilia sonchifolia DC.

［Methods］ The optimal extraction conditions were determined by single factor, response surface analysis and variance analysis, and the content of protocatechuic acid was determined by HPLC.

［Results］ The protocatechuic acid standard curve equation was: y=1 435x+8 403, R 2=0.999 8, indicating a good linear relationship. The optimal extraction conditions were as follows: a temperature at 80 ℃, an extraction time of 1 h, a material-to-liquid ratio at 1∶10 and an ultrasonic power of 600 W, and the content of protocatechuic acid extracted was 1.93 mg/g. The method showed a RSD of 0.41%, less than 2%, and the detection limit was 0.0 000 047 261 g/ml. The experimental sample X 1 was the low-level 0.1 mg/ml standard solution, which showed recovery of protocatechuic acid between 100.8% and 105.2%, with a RSD of 0.013%; and the sample X 2 was the high-level 1.0 mg/ml standard solution, which exhibited recovery between 100.6% and 102.2%, with a RSD of 0.076% . Thus, the recovery was high, and the requirements of the performance index were met.

［Conclusions］ The detection method is stable and reliable and can produce satisfactory results.

Key words Emilia sonchifolia DC; Protocatechuic acid; HPLC; Response surface optimization; Variance analysis

Received: February 7, 2020Accepted: March 27, 2020

Supported by Basic and Applied Basic Research Plan of Hainan Province in 2019 (2019RC245); Natural Science Foundation of Hainan Province in 2017 (217157).

Wen CHEN (1980-), male, P. R. China, associate professor, master supervisor, devoted to research about chemical composition analysis, purification, and identification of natural products.

*Corresponding author. E-mail: wxj_wangshui@sina.com.cn.

Emilia sonehifolia DC, also known as yangticao, hongbeiye, yexiahong and hongbeiguo, is an annual or perennial herb in Emilia of Compositae［1-4］, which is distributed in South China, Central and Southwest China［5-6］. It is cool in nature, and has the functions of clearing away heat and toxic materials and promoting blood circulation and dispersing stasis. It is often used clinically in the treatment of upper respiratory tract infections, oral ulcers, pneumonia, mastitis, edema, sores and skin eczema. The aboveground part of the plant contains pyrrolidine alkaloids and flavonoids, and also contains chemical components such as phytosterols and fatty acids［7-8］. In this study, the ultrasonic-assisted extraction of protocatechuic acid from E. sonehifolia was optimized by response surface single-factor variable optimization and analysis of variance, and HPLC was applied to determine protocatechuic acid in E. sonehifolia . This study lays a foundation for the further development and use of Hainan Li medicine, and is of practical significance for the follow-up research on the content of protocatechuic acid in E. sonehifolia and the development and utilization of Hainan Li medicine.

Materials and methods

Materials and instruments

Dry E. sonehifolia (Lizhigou market in Sanya); anhydrous ethanol (Tianjin Fuyu Fine Chemical Co., Ltd.); isopropyl alcohol (chromatographically pure); petroleum ether (chromatographically pure); hexane (chromatographically pure); distilled water (Hainan Tropical Ocean University); protocatechuic acid standard (Xiya reagent); methanol (analytically pure); acetonitrile (chromatographically pure); glacial acetic acid (chromatographically pure); Watsons water (Watsons, Sanya City).

JU-6224 ultrasonic generator (Shanghai JUMP Ultrasonic Equipment Co.,Ltd.); Shimadzu LC-2010A HT high performance liquid chromatograph (Shimadzu Enterprise Management Co., Ltd.); pulverizer (Shanghai Bingdu Electric Appliance Co., Ltd.); 2 ml sterile syringes; solvent filter; conical flasks (50 ml); volumetric flasks (25 ml; 50 ml; 100 ml; 200 ml; 500 ml); CP210 electronic analytical balance (Ohaus International Trading (Shanghai) Co.,Ltd.).

Experimental methods

Pretreatment of raw material

Sampling→sun drying→drying in a drying oven (60 ℃）→polarizing (grinding)→sieving (200 mesh)→dissolving in 50% methanol→dissolving ultrasonically→filtering→determining by HPLC.

Preparation of sample solution

The whole herb coarse powder of E. sonehifolia (1.00 g) was weighed and added with 10 ml of 50% methanol, obtaining a mixture which was weighed again. The mixture was ultrasonically extracted for 1 h, and cooled. The weight loss was made up with 50% methanol, followed by shaking well and filtered with a 0.45 μm organic microporous filter, obtaining an E. sonehifolia sample solution.

Drawing of standard curve

Protocatechuic acid standard (10.0 mg) was accurately weighed, and dissolved and diluted with 50% methanol to 25 ml, followed by ultrasonic treatment to completely dissolve it, giving a 0.4 mg/ml standard solution. Then, 2.5 ml of the standard solution was diluted to 200 ml, obtaining a 5 μg/ml of protocatechuic acid standard solution. The maximum absorption wavelength of the standard sample was measured by UV spectroscopy to be 295.6 nm, and 0.1, 0.2, 0.3, 0.4 and 0.5 mg/ml solutions were injected and measured by an ultraviolet spectrophotometer at 295 nm. A standard curve was drawn with the absorbance of the sample as the ordinate and the concentration as the abscissa, obtaining a regression equation:  y=0.558x+0.001 , R2 =0.999 9. The standard curve and the concentration corresponding to the absorbance of the sample were real and reliable. Combining the standard curve equation, the content and yield of protocatechuic acid in E. sonehifolia could be calculated.

Formula 1: Content ( y )= x -0.0010.558×25 ml［9］

Formula 2: Yield (%)=Content ( y )Weight×100［10］

Investigation of univariate factors

Selection of extraction solvent concentration

Methanol solutions were prepared at different concentrations of 30%, 40%, 50%, 60% and 70%, respectively, as the extraction solvent. Then, 1.00 g of E. sonehifolia was accurately weighed, and added with 10 ml of the above solutions, respectively. The obtained mixtures were ultrasonically extracted at a water temperature of 80 ℃ and an ultrasonic power of 600 W for 1 h. The absorbance at different extraction concentrations was measured at an HPLC wavelength of 254 nm.

Selection of extraction temperature

E. sonehifolia sample (1.00 g) was accurately weighed, and added with 10 ml of 50% methanol. The mixture was extracted at an ultrasonic power of 600 W and water temperatures of 50, 60, 70, 80 and 90 ℃, respectively, for 1 h. The absorbance of the extracts obtained at different extraction concentrations was measured at an HPLC wavelength of 254 nm.

Selection of material-to-liquid ratio

E. sonehifolia sample (1.00 g) was accurately weighed, and added with 5, 10, 15, 20 and 25 ml of 50% methanol, respectively. The mixtures were ultrasonically extracted at a water temperature of 80 ℃ and an ultrasonic power of 600 W for 1 h. The absorbance of the extracts obtained at different material-to-liquid ratios was measured at an HPLC wavelength of 254 nm.

Selection of extraction time

E. sonehifolia sample (1.00 g) was accurately weighed, and added with 10 ml of 50% methanol, respectively. The mixture was ultrasonically extracted at a water temperature of 80 ℃ and an ultrasonic power of 600 W for 0.5, 1.0 , 1.5, 2.0 and 2.5 h, respectively. The absorbance of the obtained extracts was measured at an HPLC wavelength of 254 nm.

Selection of ultrasonic power

E. sonehifolia sample (1.00 g) was accurately weighed, and added with 10 ml of 50% methanol, respectively. The mixture was ultrasonically extracted at a water temperature of 80 ℃ and ultrasonic powers of 240, 360, 480, 600 and 720 W, respectively, for 1 h. The absorbance of the obtained extracts was measured at an HPLC wavelength of 254 nm.

Selection of extraction solvent

E. sonehifolia sample (1.00 g) was accurately weighed, and added with 10 ml of methanol, anhydrous ethanol, isopropanol, petroleum ether and n-hexane, respectively. The mixtures were ultrasonically extracted at a water temperature of 80 ℃ and an ultrasonic power of 600 W for 1 h, respectively. The absorbance of the obtained extracts was measured at an HPLC wavelength of 254 nm.

Orthogonal test design

Based on single-factor tests, extraction time, extraction temperature, ultrasonic power and material-to-liquid ratio were selected as experimental factors. The yield of protocatechuic acid was used as an indicator. Each factor was set with three levels. Each level was finally determined by the single-factor variable investigation［11-14］, as shown in Table 1.

Box-Behnken central composite design

In order to further optimize the optimal conditions for the treatment of the experimental sample and improve the yield of the experiment, a four-factor three-level Box-Behnken central composite test was established. The protocatechuic acid yield was used as the response value, and treatment conditions that have a greater impact on the experimental results were selected as the experimental factors: extraction time (A), extraction temperature (B), material-to-liquid ratio (C) and ultrasonic power (D). The three levels of each factor were coded using -1, 0, and 1, each of which was finally determined by the single-factor variable investigation［15-16］, as shown in Table 2.

Analytical chromatographic conditions

C18 chromatographic column (250 mm × 4.6 mm, 5 μl), Global Chromatography Co., Ltd.; mobile phase: acetonitrile-0.4% glacial acetic acid (5∶95); flow rate: 0.8 ml/min; detection wavelength: 254 nm; injection volume: 10 μl.

Linear relation

A 1 ml of standard solution was added into an ampule with a 1 ml sterilized syringe, and determined on a high performance liquid chromatograph. The 100, 200, 300, 400 and 500 mg/ml solutions were injected in order. The intensity and time chromatograms were obtained in order, and the obtained data was analyzed. A standard curve was drawn with the peak area y as the ordinate and the concentration x as the abscissa.

Results and Analysis

Drawing of protocatechuic acid standard curve

The 100, 200, 300, 400 and 500 mg/ml solutions were determined, respectively, and the wavelength with the maximum absorption was determined to be 254 nm. The peaks were measured by HPLC to obtain a standard curve of protocatechuic acid.

An orthogonal test was designed. The test design, results and analysis are shown in Table 3.

From the results of the orthogonal optimization test, it could be known that under extraction conditions: a material-to-liquid ratio at 1∶10, an extraction time of 1 h, and a temperature at 80 ℃,the optimal extraction volume was 26.99%. These conditions were adopted to perform subsequent experiments.

Results of response surface optimization

Box-Behnken test design and results

Regression equation fitting of response surface test

The effects of univariate on the yield of protocatechuic acid suggested by the investigation of univariate factors and the orthogonal optimization test were used for the response surface analysis using software desigin expert 8.0.6.1, to further analyze the effects of temperature A (h), material-to-liquid ratio B (℃), extraction time C (ml/g) and ultrasonic power D (W) on the extraction volume of protocatechuic acid.

Desgin expert 8.0.6.1 software was used to perform multivariable regression fitting analysis on the experimental data (Table 4), and the response surface quadratic multivariable regression model equation was obtained: Y=+111.59+106.78*A+106.78*B+106.52*C+106.9*D+160.35*A*B+160.22*A*C+159.93*A*D+160.12*B*C+160.06*B*D+159.87*C *, where Y is the extraction volume, and A, B, C and D stand for temperature, material-to-liquid, time and power, respectively.

Analysis of response surface optimization test

The contour map and response surface 3D diagram generated by the response surface software Design-Expert.8.05b could be used to visualize the relationship between two influencing factors. The surfaces are shown in Fig. 8-Fig.13.

Contour lines indicated that the effects of temperature and time were more significant. The slope of the response surface was relatively steep. With the increase of time at the same temperature, the yield also increased at first, and decreased after the extraction time reached 1 h. Therefore, 1 h was determined as the extraction time of protocatechuic acid.

The contour lines indicated that the effects of temperature and material-to-liquid ratio were significant. The slope of the response surface was relatively steep. As the ratio of material to liquid increased at the same temperature, the yield also increased at first and then decreased after the ratio reached 1∶10. Therefore, the material-to-liquid ratio for extracting protocatechuic acid was determined to be 1∶10.

The contour lines indicated that the effects of temperature and power were significant. The slope of the response surface was relatively steep. As the temperature increased at the same power, the yield also increased at first and then decreased after the temperature reached 1∶10. Therefore, the temperature for extracting protocatechuic acid was determined to be 80 ℃.

The contour lines indicated that the effects of time and material-to-liquid ratio were significant. The slope of the response surface was relatively steep. With the increase of the material-to-liquid ratio in the same time, the yield also increased at first and then decreased when the ratio reached 1∶10. Therefore, the material-to-liquid ratio for the extraction of protocatechuic acid was determined to be 1∶10.

The contour lines indicated that the effects of material-to-liquid ratio and power were more significant. The slope of the response surface was relatively steep. With the increase of power at the same material-to-liquid ratio, the yield also increased at first and then decreased when the power reached 600 W. Therefore, the ultrasonic power of 600 W was determined as the extraction power of protocatechuic acid.

Contour lines indicated that the effects of power and time were more significant. The slope of the response surface was relatively steep. With the increase of time at the same power, the yield also increased at first and then decreased after the time reached 1 h. Therefore, 1 h was determined as the extraction time of protocatechuic acid. It was finally determined that the optimal extraction process was extracting protocatechuic acid at a temperature of 80 ℃, a material-to-liquid ratio of 10 g/ml and a power of 600 W for 1 h, under which the extraction volume was 26.80%.

Variance analysis

From the above analysis of variance, it could be seen that temperature had a significant effect on the extraction of protocatechuic acid. The effects of the four major factors on the extraction of protocatechuic acid ranked as temperature>extraction time>material-to-liquid ratio=extraction power.According to response surface analysis and analysis of variance, the optimal extraction conditions for the extraction of protocatechuic acid were as follows: a temperature at 80 ℃, an extraction time of 1 h, a material-to-liquid ratio at 1∶10, and an ultrasonic power of 600 W, with which the extraction efficiency of protocatechuic acid was relatively good. It could be seen from the analysis of the above experimental results that the optimal extraction conditions for protocatechuic acid were a temperature at 80 ℃, an extraction time of 1 h, a material-to-liquid ratio at 10 g/ml and a power of 600 W, with which the optimal extraction volume was 26.99% (Table 5).

Determination of protocatechuic acid from E. sonehifolia

Preparation of E. sonehifolia sample solution: The whole herb coarse powder of E. sonehifolia (1.00 g) was accurately weighed and added with 10 ml of 50% methanol, obtaining a mixture which was weighed again and ultrasonically extracted for 1 h［3］. After cooling, the weight loss was made up with 50% methanol, followed by shaking well and filtering with a 0.45 μm organic microporous filter, obtaining an E. sonehifolia sample solution. The sample solution was injected, and its peak was measured by HPLC. The peak was substituted into the standard curve equation to obtain the content of protocatechuic acid in the sample. After the peak areas of different concentrations of protocatechuic acid were determined by HPLC, with the concentration as the abscissa and the peak area as the ordinate, a standard curve was drawn. The equation of the standard curve was y=1435x + 8403, and the R2 value was 0.999 8, which showed that the standard equation was true and reliable. The peak area of the sample measured by HPLC was 285 834, which was substituted into the standard curve equation 3, and the content of protocatechuic acid in E. sonehifolia was 1.93 mg/g.

Formula 3: Content ( y )= x -8 4031 435×10 μl ［17］

Peak area of protocatechuic acid standard determined by HPLC

Protocatechuic acid standard (10.0 mg) was accurately weighed, and dissolved and diluted with 50% methanol to 25 ml, followed by ultrasonic treatment to completely dissolve it, giving a 0.4 mg/ml standard solution. Then, 2.5 ml of the standard solution was diluted to 200 ml, obtaining a 5 μg/ml of protocatechuic acid standard solution. The detection wavelength was set at 254 nm , and the chromatogram was measured by HPLC. The peak of the standard sample had a retention time of 13.237 min and a height of 10 224, and the peak area was 557 885, as shown in Fig. 14.

Peak area of sample determined by HPLC

The E. sonehifolia sample (1.00 g) was accurately weighed and added with 10 ml of 50% methanol, obtaining a mixture which ultrasonically extracted at 80 ℃ and 600 W for 1 h. the obtained extracted was filtered with a 0.45 μm organic microporous filter, obtaining an E. sonehifolia sample solution. The sample solution was injected, and its peak area was measured by HPLC. The peak of the sample showed a retention time of 13.507 min and a height of 4 565, which produced a peak area of 285 834. From the retention time, the sample peak was consistent with the standard peak, as shown in Fig. 15.

Result accuracy analysis

Precision test

The precision test referred to the degree of mutual agreement between a set of measured values of the same homogeneous sample by this method. The closer the values are, the more precise they are［18］. RSD is the relative standard deviation and SD is the standard deviation. RSD is determined according to RSD =(SD/average) × 100%［19］. In analysis, RSD is often used to indicate precision, and it is published as:

RSD=SDX ×100%

SD=∑4i=1 (Xi-X)2n-1

SD =1 179.6

RSD =0.41%

The RSD value was less than 2%, which indicated that the method has high experimental accuracy and good reproducibility.

Detection limit test

The detection limit of the instrument was calculated according to the formula DL=3RSD/L ［20］ ( DL is the detection limit, RSD is the standard deviation, and L is the slope of the standard curve).

As can be seen from Table 6, the standard deviation and the slope of the standard solution regression equation were 2.22 and 1 435, respectively, which were substituted into the formula, obtaining the detection limit of 0.000 047 261 μg/ml, indicating that the detection limit of the instrument could reach the expected experimental effect. HPLC is more suitable for the detection of protocatechuic acid extracted.

The recovery was determined according to Recovery (%)=(Measured value of spiked sample-Measured value of sample) ÷ Measured value of spiked sample × 100［20］. The recovery should be between 98 and 102%. The experimental sample X1 was the low-level 0.1 mg/ml standard solution, which showed recovery of protocatechuic acid between 100.8% and 105.2%, with a RSD of 0.013%. The sample X2 was the high-level 1.0 mg/ml standard solution, which exhibited recovery between 100.6% and 102.2%, with a RSD of 0.076% (Table 7). The method satisfied the requirements of the performance index, and the recovery of the spiked sample was high, indicating that the detection method is stable and reliable.

SampleSample value 0.1∥mgDetermination of spiked sample∥mgRecovery %Sample  RSD  % Spiked sample  RSD ∥% Sample value 0.1∥mgDetermination of spiked sample∥mgRecovery % Sample  RSD  % Spiked sample  RSD ∥%

Protocatechuic acid0.089 0 0.089 2 0.089 60.189 8 0.190 8 0.194 8100.8 101.6 105.20.0150.130. 889 4 0.896 6 0.899 21.905 4 1.918 6 1.905 2101.6 102.2 100.60.0560.076

Conclusions

The univariate factors were investigated through experiments. After response surface optimization and analysis of variance, the optimal extraction process of protocatechuic acid was extracting samples with 50% methanol at a temperature of 80 ℃, a material-to-liquid ratio of 1∶10 ml/g and an ultrasonic power of 600 W for 1 h. Temperature had a significant effect on the extraction of protocatechuic acid, and the regression model was more significant. The effects of the four major factors on the amount of protocatechuic acid extracted ranked as temperature>extraction time>material-liquid ratio=extraction power. The precision test showed a RSD of 0.41%, less than 2%. The method has good reproducibility and high experimental accuracy. The detection limit was 0.0 000 047 261 μg/ml, which is good. The experimental sample X 1 was the low-level 0.1 mg/ml standard solution, which showed recovery of protocatechuic acid between 100.8% and 105.2%, with a RSD of 0.013% ; and the sample X 2 was the high-level 1.0 mg/ml standard solution, which exhibited recovery between 100.6% and 102.2% , with a RSD of 0.076%. The amount of protocatechuic acid extracted from E. sonehifolia was 1.93 mg/g. HPLC is more suitable for the detection of protocatechuic acid content. The experimental protocol has operability and data validity. The operation is simple and the results are satisfactory.

References

［1］ LI TT, WANG XJ, CHEN W, et al. Review on the determination of protocatechuic acid in Emilia sonchifolia and its antioxidant activity［J］. Nei Jiang Ke Ji, 2017, 38(2): 68-69. (in Chinese)

［2］ CHEN W, WANG XJ, SHAO DX, et al. Response surface optimization for extraction of Sipunculus nudus polysaccharides［J］. Genomics and Applied Biology, 2019, 38(6): 2522-2533. (in Chinese)

［3］ XU Y, CHEN W, WANG XJ, et al. Comprehensive study on the determination of total polyphenols in Emilia sonehifolia DC［J］. Nei Jiang Ke Ji, 2018, 39(6): 112-113. (in Chinese)

［4］ CHEN W, WANG XJ, WANG YJ, et al. Determination of amino acids from Sipunculus nudus and its nutritional value evaluation［J］. Storage and Process, 2018, 18(4): 108-113, 118. (in Chinese)

［5］ HE HY, CHEN W, WANG XJ, et al. Review on the extraction technology of flavonoids in Emilia sonehifolia DC［J］. Nei Jiang Ke Ji, 2018, 39(8): 48-49. (in Chinese)

［6］ LYU XT, CHEN W, WANG XJ, et al. Response surface optimization of enzyme extraction process of Emilia sonehifolia DC polysaccharide［J］. Nei Jiang Ke Ji, 2018, 39(8): 46-47. (in Chinese)

［7］ CHEN LM. Separation, analysis and antioxidant activity of Emilia sonehifolia DC［D］. Nanning: Guangxi University, 2014. (in Chinese)

［8］ NING Y, ZHU CC, MO JL, et al. Determination of senkirkine in Herba Emiliae by HPLC［J］. Drug Standards of China, 2014, 15(4): 269-272. (in Chinese)

［9］ LIU ZM, LI GH. Determination of protocatechuic acid and epicatechin in black soybean skin by HPLC［J］. Food Research and Development, 2016, 37(11): 150-152. (in Chinese)

［10］ CHEN W, WANG XJ, ZHANG SQ, et al. Response surface optimization of ultrasonic-assisted extraction of quercetin from cactus flowers［J］. Journal of Hainan Tropical Ocean University, 2018, 25(5): 90-100. (in Chinese)

［11］ CHEN LL, ZHANG CF, WANG QG. Determination of chlorogenic acid and protocatechuic acid in potatoes by HPLC［J］. Science and Technology of Food Industry, 2016, 37(13): 299-303. (in Chinese)

［12］ WANG Y, HONG WY, WANG JM. Simultaneous determination of protocatechuic acid and chlorogenic acid in Sargentodoxa cuneata (Oliv. ) Rehd. et Wils. by high performance liquid chromatography［J］. Journal of New Chinese Medicine, 2016, 48(2): 241-243. (in Chinese)

［13］ CHEN W, WANG XJ, WANG YJ, et al. Orthogonal test design to optimize amino acid extraction conditions of sea urchin［J］. The Food Industry, 2018, 39(9): 157-163. (in Chinese)

［14］ CHEN JF, ZHONG YQ, ZENG XF, et al. Response surface optimization of ultrasonic-assisted extraction of total flavonoids from carrot leaves［J/OL］. China Food Additives, 2017(10): 66-72. (in Chinese)

［15］ HAN Y, YANG M, XIE XM, et al. Content determination of chlorogenic acid, protocatechuic acid and total phenolics in Chaenomeles sinersis from different areas［J］. China Pharmacy, 2015, 26(24): 3399-3403. (in Chinese)

［16］ CHEN W, WANG XJ, ZHAO Y, et al. Determination of amino acid from Sipunculs nudus by acid hydrolysis-automatic amino acid analyzer［J］. Science and Technology of Food Industry, 2017, 38(3): 299-304. (in Chinese)

［17］ NING Y, ZHU CC, MO JL, et al. Determination of quercetin in Emilia sonchifolia (L.) DC by HPLC［J］. Drug Standards of China, 2015, 16(2): 117-119. (in Chinese)

［18］ HE JW, DING HY, ZOU DK, et al. Optimization of shepherd餾 purse polysaccharide extraction process based on factorial test and response surface［J/OL］. Journal of Chinese Medicinal Materials,2017(5):1168-1171. (in Chinese)

［19］ GE XJ, ZHOU DJ, WANG B, et al. Optimization of neutroproteinase-assistant extraction of polysaccharide from stroma of Shiraia bambusicola Henn by response surface analysis and characterization of the extracted polysaccharide［J］. Food Science, 2017, 38(14): 193-199. (in Chinese)

［20］ CHEN W, WANG XJ, LI LY, et al. Determination of propolis flavonoids by ultrasonic methanol dissolution-HPLC method［J］. Journal of Hainan Tropical Ocean University, 2017, 24(5): 81-87. (in Chinese)

Editor: Yingzhi GUANGProofreader: Xinxiu ZHU