Rapid Determination of Three Organic Acids in Polygonum Vivipari Rhizoma via One Marker by HPLC-UV at Equal Absorption Wavelength and Effervescence-Assisted Matrix Solid-Phase Dispersion

A rapid HPLC-UV method for the determination of three organic acids (neochlorogenic acid, chlorogenic acid, and cryptochlorogenic acid) in Polygoni Vivipari Rhizoma (PVR) by one marker was developed. The sample was prepared by effervescence-assisted matrix solid-phase dispersion (EA-MSPD). The separation of compounds was performed on a Poroshell column. The equal absorption wavelength was set as follows: 292 nm (0∼7 min) and 324 nm (7∼10 min). The analytical time including sample extraction and HPLC separation time was 12 min. The analytical method validation such as accuracy (recoveries 99.85%–106.29% and RSD < 2.9%), precision (RSD < 1.3%), reproducibility (RSD < 1.7%), and stability tests (RSD < 0.7% in 24 h) proved that the established HPLC method was suitable for determination of three organic acids in PVR. The contents of three analytes obtained by the external standard method with three markers and the equal absorption wavelength method with one marker were similar (RSD ≤ 2.0%). The developed method, which is rapid and reference compound saving, is an improved quality evaluation method of PVR.


Introduction
Polygoni Vivipari Rhizoma (PVR), also called "Zhuyaliao" in Chinese, is a famous Tibetan folk medicine. Its dried root is usually used in checking diarrhea and activating blood circulation to dissipate blood stasis [1]. According to the literature studies, phenolic compounds (organic acids and favonoids) are the main active constituents responsible for the antioxidant and bacteriostatic activities [2][3][4]. Among the phenolic compounds, the chlorogenic acid series compounds such as neochlorogenic acid, chlorogenic acid, and cryptochlorogenic acid have been found to exhibit good pharmacological activities [5]. Furthermore, these three compounds are also the primary organic acids found in PVR [6]. Terefore, simultaneous determination of three organic acids is crucial for the quality evaluation of PVR due to their good bioactivities and high contents.
To date, several HPLC methods for the determination of the three organic acids were reported by the external standard method (ESM) based on three reference compounds applied [7][8][9][10]. Te chlorogenic acid is cheap, which is about 14 dollars per 20 mg, while the prices of neochlorogenic acid and cryptochlorogenic acid are relatively more expensive, which are both about 140 dollars per 20 mg. In order to reduce the cost of PVR sample test and simplify the method, it is necessary to develop an analytical method for the determination of the three organic acids by one cheap reference compound (chlorogenic acid). Te quantitative analysis of multicomponents by single marker (QAMS) method has been applied in herbal medicines [11][12][13][14]. However, the relative calibration factor (RCF) is necessary to be established, which increases the operational complexity and limits the wide application of QAMS. Terefore, developing an HPLC method for determination the three compounds by one reference compound without RCF is preferable. Tese three organic acids performed diferent UV absorptions at diferent wavelengths. Te chlorogenic acid may have the equal UV absorption with two other compounds at certain wavelengths. It is the equal absorption waveslength (EAW) of chlorogenic acid with neochlorogenic acid or cryptochlorogenic acid. Hence, developing a HPLC-UV method at the EAWs could realize simultaneous determination of the three components by chlorogenic acid without RCF.
In addition, due to the complex matrix of PVR and similar structures of chlorogenic acid and two other organic acids, the reported HPLC-UV methods for the determination of three organic acids, including extraction and separation, are always time-consuming (more than 25 min) [7][8][9][10]. In order to develop a rapid HPLC method for determining the three organic acids in PVR, the rapid sample extraction and HPLC separation should be considered. Efervescence-assisted matrix solid-phase dispersion (EA-MSPD), a modifed MSPD method, is proved to be a simple, fast, and efective extraction technique. It promotes the microextraction process by generation of carbon dioxide in situ from the efervescent mixture consisting of a carbon dioxide source and an acid component dissolved in water [15][16][17]. Hence, EA-MSPD is a potential rapid extraction method for extracting organic acid from PVR. On the other hand, the Poroshell column is a kind of rapid HPLC column [18][19][20], which can provide the rapid separation of organic acids in PVR.
In the present study, a rapid and reference compound saving HPLC-UV method for the determination of neochlorogenic acid, chlorogenic acid, and cryptochlorogenic acid via one cheap marker (chlorogenic acid) in EAW is developed. Te developed HPLC-UV method was successfully applied in the determination of organic acids in ten batches of PVR samples.

Preparation of Reference Compound Solutions.
1.5 mg/mL neochlorogenic acid, chlorogenic acid, and cryptochlorogenic acid were dissolved in 50% methanol, respectively. Mixed reference compound solutions were prepared by mixing them and diluted to the intended concentration with 50% methanol. All solutions were stored at 4°C.

Preparation of Sample Solution.
Te sample solution was prepared by the EA-MSPD method. In order to obtain the good extraction efciency, diferent extraction conditions (composition of efervescent mixture, ratio of sample and efervescent mixture, milling time, polarity, and volume of extraction solvent) were studied by the single-factor method. Te contents of three organic acids were used to evaluate the extraction efciency.
Te sample powder (0.25 g) and the efervescent mixture (0.592 g sodium carbonate and 0.658 g oxalic acid, molar ratio about 10 : 13) were precisely weighed and milled with the Retsch MM400 ball milling instrument (Retsch, Shanghai, China) for 1 min to obtain the homogeneous mixture. Te 0.4 g mixture was accurately weighed into a 50 mL centrifuged polypropylene tube, and 4 mL of 20% methanol was added. Te efervescence occurred instantly and lasted about 30 s. When the process ended, the extraction solution was vortexed (5 s) and fltered through a 0.22 μm membrane before HPLC injection.

UV Condition.
Tree reference compounds were dissolved with 10% methanol containing 0.1% acetic acid (the HPLC mobile phase) to 19 μg/mL (neochlorogenic acid 19.08 μg/mL, chlorogenic acid 19.09 μg/mL, and cryptochlorogenic acid 19.07 μg/mL). Te ultraviolet spectra of the three analytes were obtained by scanning the three reference compound solutions from 200 nm to 400 nm with Agilent Cary 60 ultraviolet spectrophotometer (Agilent Technologies, USA). 10% methanol containing 0.1% acetic acid was used as blank.

HPLC
Condition. An Agilent 1260 II Series HPLC system (Agilent Technologies, USA) was employed for the analysis. Te separation of compounds was achieved on an Agilent Poroshell 120 EC-C18 column (50 × 4.6 mm, 2.7 μm) (batch number: B18386) at a column temperature of 35°C and eluted with 10% methanol containing 0.1% acetic acid at a fow rate of 1.0 mL/min in the isocratic mode. Te detection wavelength was set at 0∼7 min (292 nm and 2 nm) and 7-10 min (324 nm and 2 nm). Te injection volume was 2 μL.

2
International Journal of Analytical Chemistry 2.6. Method Validation. Te method validation, including linearity, limit of detection (LOD), limit of quantifcation (LOQ), precision, accuracy, repeatability, and stability tests, was carried out.
2.6.1. Linearity, LOD, and LOQ. A series of concentrations of reference component solutions were prepared for the evaluation of linearity. Te neochlorogenic acid (0.32 to 477.00 μg/mL), chlorogenic acid (0.64 to 477.18 μg/mL), and cryptochlorogenic acid (0.95 to 476.65 μg/mL) were analyzed by HPLC. Te standard curve was constructed by plotting the peak area (y) versus the concentrations of reference compounds (x). Te LODs and LOQs were determined by reference components and recorded as the corresponding concentrations, which gave the signalto-noise (S/N) ratios approximately 3 and 10, respectively.

Precision and Repeatability.
Te intra-and interday assays were used to assess the precision of the developed method. Te intraday precision was determined by analyzing the reference components solution six times within one day. Te interday precision was determined by analyzing the reference components solution twice per day for three days. Te relative standard deviation (RSD) was used as a measure of precision. Te repeatability of the developed method was evaluated by six replicates of the PVR sample analysis. Te samples were extracted as "2.3" and analyzed as "2.4." Te RSD of the contents was used as a measurement of repeatability.

Recovery.
A recovery test was used to evaluate the accuracy of the developed method. Known amounts of three organic acids were added to the PVR sample powder and then extracted and analyzed by the developed method. Te PVR sample was analyzed six times. Te recovery rates were calculated as 100% × (found amount − original amount)/ spiked amount.

Optimization of EA-MSPD Extraction Conditions.
In order to obtain the good EA-MSPD method, diferent extraction conditions (composition of efervescent mixture, ratio of sample and efervescent mixture, milling time, polarity, and volume of extraction solvent) were studied by the single-factor method. Te contents of three organic acids were used to evaluate the extraction efciency. Tukey's honestly signifcant diference test was carried out to compare the organic acid contents at diferent levels of the investigated parameter. Te efervescent mixture, a combination of the carbon dioxide source and acid component, had direct efect on the efervescent efect and extraction efciency. Tree commonly efervescent mixtures were tested, including sodium carbonate-oxalic acid, sodium carbonate-citric acid, and sodium carbonate-sodium dihydrogen phosphate [21][22][23]. It was observed that the efervescence efect of sodium carbonate-oxalic acid was more intense than two others, and the efervescence time (30 s) was faster than two others (more than 60 s). Te ratio of sodium carbonate and oxalic acid and the ratio of sample and efervescent mixture were also important to the PVR sample extraction. According to the chemical reaction of sodium carbonate and oxalic acid, the molar ratio of sodium carbonate and oxalic acid is 1 : 1 (mass ratio of 100 : 85). Te efervescent mixture (sodium carbonate: oxalic acid � 100 : 85) would make the sample solution in a weak alkaline environment. Te organic acids are unstable in alkaline solution [24,25]. So, more oxalic acid was added in the efervescent mixture to keep the sample solution in acid environment. Four diferent ratios of sodium carbonate and oxalic acid (100 : 85, 100 : 90, 100 : 95, and 100 : 100) were compared. As shown in Figure 1(a), the ratios of sodium carbonate and oxalic acid in 100 : 90, 100 : 95, and 100 : 100 showed better extraction efciency. Stability tests also revealed that three organic acids were stable in 24 h at these ratios. Consequently, 100 : 90 was selected as the condition because of the less material cost. Tree different ratios of sample and efervescent mixture (1 : 5, 1 : 10, and 1 : 20) were evaluated. As shown in Figure 1(b), the content of analytes was similar in the three conditions. Te ratio of sample and efervescent mixture (1 : 5) was chosen for less material consume. After the composition of the efervescent mixture and the ratio of sample-efervescent mixture were fxed, the milling time (1, 2, and 3 min) was examined. Te results (Figure 1(c)) of the three tests were similar, and 1.0 min was used in this study.

International Journal of Analytical Chemistry
Compared with the votexing extraction method (non-EA-MSPD sample preparation), the extraction efciency of the developed EA-MSPD method was improved 26.2% for neochlorogenic acid, 33.0% for chlorogenic acid, and 34.9% for cryptochlorogenic acid. To further confrm the extraction efciency of the developed EA-MSPD, PVR sample S1 was extracted by the proposed EA-MSPD method and the reported ultrasonic extraction method [10], respectively. Tree replicates were performed. Te contents of neochlorogenic acid, chlorogenic acid, and cryptochlorogenic acid tested by the developed EA-MSPD method were 4.49 ± 0.01%, 8.88 ± 0.03%, and 0.76 ± 0.01% while the contents of these three analytes by the reported ultrasonic extraction method were 4.48 ± 0.05%, 8.59 ± 0.04%, and 0.71 ± 0.02%. Tese results show that the extraction efciency of the developed EA-MSPD method is similar to that of the reported ultrasonic extraction method, which could be used for extracting three organic acids from PVR.

Optimization of HPLC Conditions.
In order to develop a rapid HPLC separation of the three target compounds, the rapid HPLC column (Poroshell column) was employed. 0.1% acetic acid methanol based on the literature was used as the mobile phase system [10]. Tree diferent mobile phases (8%, 10%, and 12% methanol with 0.1% acetic acid, respectively) were tested for separation. 10% methanol with 0.1% acetic acid was chosen as the eluting solvent for the good resolution and short separation time. Te fow rate of 1.0 ml/min was used according to the literature [10]. Tree diferent column temperatures (30, 35, and 40°C) were tested. Te separations of analytes in three temperatures were similar, and 35°C was used in the current experiment as easy control and less energy consume.
Traditional QAMS often employs the maximum absorption wavelength of analytes, at which diferent compounds have diferent UV responses. So, the RCF is used for the determination of multiple compounds with one standard. In this study, three organic acids are detected at the EAW. Te three analytes have the same response, and the RCFs are close to 1.0. So, it can test three compounds with one standard without RCF. Terefore, the selection of EAW is the key factor in the present HPLC method, which includes two steps (fnd and confrm EAW). First, screening the EAW by UV, three reference compound solutions at the same concentration were scanned from 200 nm to 400 nm with an ultraviolet spectrophotometer to get the UV spectrum for the three analytes. As shown in Figure 2, the UV response of cryptochlorogenic acid was lower than the other two compounds at the same UV wavelength. In order to obtain the better UV response of analytes, the maximum absorption wavelength (at 326 nm) was selected as the HPLC detection wavelength of cryptochlorogenic acid. Te UV response of cryptochlorogenic acid (at 326 nm) was equal to noechlorogenic acid (at 296 nm and 338 nm) and chlorogenic acid (at 294 nm and 340 nm). Second, confrming the EAW by HPLC-UV, the mixed reference compound solution (neochlorogenic acid 63.60 μg/mL, chlorogenic acid 63.62 μg/mL, and cryptochlorogenic acid 63.55 μg/mL) was injected to HPLC and detected at diferent wavelengths for confrming the EAW. Te cryptochlorogenic acids were detected around 326 nm (±0 nm, ±1 nm, ±2 nm, and ±3 nm). It was found that cryptochlorogenic acid had the maximum peak area at 324 nm (Table S1). So, the detection wavelength of cryptochlorogenic acid was set at 324 nm. Te noechlorogenic acid was detected around 296 nm (±0 nm, ±1 nm, ±2 nm, ±3 nm, and ±4 nm) and 338 nm (±0 nm, ±1 nm, ±2 nm, ±3 nm, and ±4 nm). Te chlorogenic acid was detected around 294 nm (±0 nm, ±1 nm, ±2 nm, ±3 nm, and ±4 nm) and 340 nm (±0 nm, ±1 nm, ±2 nm, ±3 nm, and ±4 nm). Te results (Table S1) showed that the peak areas of noechlorogenic acid (at 292 nm and 338 nm) and chlorogenic acid (at 292 nm and 339 nm) had the same peak areas with cryptochlorogenic acid (at 324 nm). Considering less detection wavelengths used, 292 nm was chosen as the detection wavelength for noechlorogenic acid and chlorogenic acid. In addition, the diferent bandwidths (1 nm, 2 nm, 4 nm, and 8 nm) were compared for the detection of three reference compounds at EAW. Te results showed that the lowest RSD of peak areas could be obtained at 2 nm. Te EAW conditions were as follows: 0∼7 min (292 nm, 2 nm) for detecting noechlorogenic acid and chlorogenic acid and 7-10 min (324 nm, 2 nm) for detecting cryptochlorogenic acid. Tables 1-4. Te analytical method showed good linearity in the tested range with correlation coefcient R � 0.9999. Te LODs and LOQs of the three analytes were less than 0.7 μg/mL and 1.0 μg/mL, respectively. Te RSDs of intraday and interday precision were less than 1.3%. Te RSDs of repeatability were less than 1.7%. Te RSDs of stability were less than 0.7% within 24 hours. Te recoveries of three analytes were 99.85∼106.29% (RSD less than 2.9%). In the robustness test, the RSDs of both contents and RRTs of noechlorogenic acid and cryptochlorogenic acid (determinated by chlorogenic acid) were all less than 2.0%. Te resolutions of the 3 target peaks to the adjacent peaks were all larger than 1.5.

Analysis of Sample.
Te developed HPLC-UV EAW method was successfully applied in the determination of the target components in PVR samples. Te chromatograms of the reference compounds and sample are shown in Figure 3, and the results are listed in Table 5. To confrm the feasibility of the developed HPLC-UV EAW method, the contents of three organic acids in ten PVR samples were determined by the ESM (with three reference compounds) and EAW method (with chlorogenic acid), respectively. Te RSDs of the results obtained by the two methods were not more than 2.0%. Tese results indicated that the developed HPLC-UV EAW method could be used for quantitative analysis of three organic acids in PVR sample. Te contents of noechlorogenic acid (1.24∼8.25 mg/g), chlorogenic acid (1.75∼13.75 mg/g), and crptochlorogenic acid (0.65∼2.10 mg/g) in PVR samples were agreed with the literature data [10].      International Journal of Analytical Chemistry

Comparisons of the Developed and Previously Reported
Methods. Several HPLC methods for analyzing these three organic acids have been reported [7][8][9][10]. Compared with these reported methods, the developed method is reference compound saving, simple, and fast. Te reported methods employed ESM with three reference compounds applied. In addition, the traditional QAMS for the determination of the three target analytes with one marker, often performed at the maximum absorption wavelength 330 nm, at which neochlorogenic acid, chlorogenic acid, and cryptochlorogenic acid respond diferently, resulted in which the RCFs of neochlorogenic acid and cryptochlorogenic acid to chlorogenic acid are required. In this study, the EAW method uses only chlorgenic acid for determination of three target analytes. No RCF is applied because EAW is employed, at which neochlorogenic acid, chlorogenic acid, and cryptochlorogenic acid have the same UV response, and RCFs are close to 1.0. So, the developed HPLC-UV EAW method was much simpler than the traditional QAMS method.
Te literature's methods for analyzing the three organic acids consume more than 25 min. For example, the HPLC method developed by Haghi et al. [7] costs 120 min in sample extraction and takes 35 min in HPLC separation with a total time of 155 min. Another HPLC method developed by Honda et al. [9] consumes 60 min including sample extraction (30 min) and HPLC separation (30 min). In the current method, EA-MSPD is applied in the PVR sample extraction and the Poroshell column is executed in HPLC separation. Te whole process only costs 12 min in total, including about 2 min of sample preparation and 10 min of HPLC separation. It is faster compared to the reported methods [7][8][9][10].

Conclusions
In the present study, a rapid HPLC-UV EAW method for simultaneous determination of three organic acids in PVR samples by chlorogenic acid is established. Compared with the reported methods, the developed method is rapid, simple, and reference compound saving. It would be a good improved method for quality evaluation of the major organic acids in PVR samples.

Data Availability
Te data used to support the fndings of this study are included within the article and the supplementary information fle.

Conflicts of Interest
Te authors declare that they have no conficts of interest.