Development and Validation of RP-HPLC Analysis Method for Determination of Total Alkaloid Content of Soursop ( Annona muricata L.) Leaf Extract

. Soursop leaves ( Annona muricata L) contain alkaloids that have pharmacological effects. This study aims to determine the total alkaloid content of soursop leaf ethanol extract using a validated HPLC method. Extraction was carried out by soxhletation followed by liquid-liquid extraction using chloroform. Optimizing HPLC conditions on the mobile phase, flow rate, concentration and wavelength, further testing the validation of the analytical method. Optimizing HPLC conditions obtained optimal results at a concentration of 500 ppm with a flow rate of 1.0 mL/min using acetonitrile: methanol: water (80:5:15) at a wavelength of 272 nm. These results have met the validation requirements of analytical methods including System Suitability Test (UKS), selectivity, linearity with y = 2.01914x –219,97226, correlation coefficient value (r) = 0.9995 and Vx0 value of 0.11201%. The limit of detection and limit of quantization obtained were 8,78525 ppm and 29,28418 ppm, respectively. The RSD percentage of the precision test is 1.0453%, and the accuracy test is obtained from the recovery of 98-102%. The result of determining the total alkaloid content of the sample was 0.0862 ± 0.004 % (%w/w) of the total alkaloid content (TALC) triple replication. The conclusion of this research is that the analytical method using the HPLC system for determining the total alkaloid content of the ethanolic extract of soursop leaves is validate This is an open acces article under the CC-BY license. Attribution unrestricted

with quercetin which yielded optimum results for testing. According to [29] the selected analytical method is then validated first to ensure that the developed method is considered valid for routine testing if it meets the parameters of linearity, precision, accuracy, detection limit and quantitation limit. The application of the reverse phase HPLC analysis method was carried out to determine the total alkaloid content in the ethanol extract of soursop leaves. By knowing the total alkaloid content of soursop leaf extract this can be used as a reference to developed as a candidate for medicinal plants according to their therapeutic activity.

Extract Preparation
Soursop leaf powder (20 grams) wrapped in filter paper, placed into a soxhlet then in a round bottom flask was added 150 mL of ethanol. The extractor is started and stopped when the solvent color returns to its original state. The filtrate was concentrated using a water bath at a temperature of 50 o C. The resulting concentrated extract was calculated the yield.

Sample Preparation
Soursop leaf ethanol extract (50.0 mg) dissolved with 2 N hydrochloric acid (HCl) then filtered. Furthermore, washing three times using chloroform, where the water phase is collected and added NaOH until the pH is neutral. The test solution was taken and added with 5 mL of phosphate buffer pH 4.7. The mixture was shaken with a separating funnel and extracted with 5 mL of chloroform. The chloroform phase was selected, then evaporated with nitrogen gas and reconstituted with methanol: aqua bidestillata (50:50) to 10.0 ml, then filtered and injected into the HPLC system. The sample solution was replicated three times [30].

Mobile Phase Preparation
In the development of the mobile phase is done by mixing acetonitrile: methanol: aquabidest (80: 5: 15). The mixture was sonicated for 10 minutes and filtered using a vacuum filter [31] .

Solvent Preparation
The solvent in this study was carried out by mixing methanol and water (50:50), the mixture was sonicated for 10 minutes [32].

Preparation of Phosphate Buffer pH 4.7
A 0.2 M sodium phosphate (Na 2 HPO 4 ) solution was made mixed with 0.2 M citric acid (C 6 H 8 O 7 ) then adjusted to pH 4.7 [30].

Preparation of Caffeine Standard Solution
Caffeine stock solution with a concentration of 2000 g/mL was prepared by weighing 100.0 mg in a 50.0 mL volumetric flask with solvent added and homogenized. The intermediate solution was made by pipetting 5.0 mL of stock solution, put into a 10.0 mL volumetric flask, added solvent to the mark and homogenized [32].

Development of HPLC System Conditions a. Development of Flow Rates
Development of flow rates was operated with optimization of speed variations from 1 to 1,4 mL per minute was carried out to optimize the analyte separation conditions. By decresing retention time is an increase in flow velocity from the system until a good peak is obtained [33].

b. Determination of Concentration and Wavelength
Determination of the optimal caffeine concentration varies from 100.0 to 500.0 ppm and then measured at a wavelength of 200-800 nm using a UV-Vis spectrophotometer [32].

Validation of Analytical Method
At the pre-validation stage, a System Conformity Test was carried out on HPLC. This test was carried out with six injections of the system which had been adjusted for the optimum analysis conditions. The system that shows the retention time and repeated area to meet the acceptance value requirements, namely RSD < 2%, is a confirmation of the selected condition [35].
The next stage is the validation of the analytical method by testing the following parameters : Selectivity/specificity is carried out to see the ability to recognize analyte responses and be able to distinguish analytes from other compounds, namely by testing purity and seeing the resolution results obtained from injection results [36]. Determination of linearity is seen if there is a linear relationship from the regression line in the concentration range of 50-120% caffeine standard solution with nine points (USP, 2016) with the correlation coefficient parameter (r). The residual standard deviation (Sy) is also taken into account. The limit of detection (LOD) and limit of quantitation (LOQ) are determined by making a standard curve made with the smallest six points of the concentration that gives a signal to the detector from the caffeine standard solution so that the residual standard deviation and slope values of the calibration curve results are obtained, then plugged into the equation: (Sy is the standard deviation of the residual and b slope) Precision is the repetition of six replications with a concentration level range between 80%, 100%, and 120% of the target concentration and then the RSD value < 2% is calculated, which indicates the closeness of the test results in each repetition at the concentration variation. The closeness of the analyte content to the actual concentration can be determined based on the accuracy parameter. The accuracy value was obtained by preparing the analyte concentration in the concentrated extract sample, namely 0.01% plus caffeine with various concentrations of 80%, 100%, and 120% and then analyzed (addition method). The three addition samples were replicated three times. The recovery parameters from the test data should not exceed the range of 85-110% [37].

Results and Discussion
The extraction process carried out by soxhletation has the advantage that the extraction time is relatively fast and the solvent used is small. Soxhlet also requires an optimum temperature to attract the alkaloids, so that they are able to bind the alkaloids to the appropriate solvent during the extraction process. The yield of the extract was 5.96%.

Optimation of HPLC Conditions
Optimization was carried out on HPLC instruments to obtain the selected optimum conditions, including the development of flow rates. At a flow rate of 1.0 mL/minute, a shorter retention time and peak symmetry were obtained than the flow rates of 1.2 and 1.4 mL/minute. This is due to an increase in pressure from the pump so that the analyte can be retained longer in the column, the increase in pressure is caused by the large load carried by the mobile phase in the HPLC system. The choice of a flow rate of 1.0 mL/minute can be seen from the results of the analysis at a value of N > 2000. Other data are described in To determine the optimum wavelength, measurements were made at the maximum absorption of the caffeine standard (Denis, 2018). The measurement results show the highest absorption of caffeine at a wavelength of 272 nm, measurements were made using a UV-Vis spectrophotometer and analyzed and selected the maximum absorbance results which were still in the absorbance range of 0.2-0.8 [39]. From the test results, there is a 1 nm wavelength shift, but this result is still accepted because according to [40], it is stated that the maximum wavelength absorption shift should not be more than 3 nm. The next optimization is in the form of a mobile phase. The research was conducted by injecting 500 ppm caffeine standard in variations in the composition of the mobile phase, including methanol : aqua bidestillata (50 : 50). The chromatogram peak results were not good because tailings occurred even though the retention time was < 10 minutes (Figure 2A). The next mobile phase is methanol : aqua bidestillata : acetic acid 2% (50 : 48 : 2) resulting in a retention time of 2.00 minutes but widening of the chromatogram peaks ( Figure 2B). The third mobile phase was acetonitrile : methanol : aqua bidestillata (80 : 5 : 15) and a sharp chromatogram peak was obtained with a faster retention time of 1.49 minutes ( Figure 2C).  From the three variations of the mobile phase, acetonitrile: methanol: aqua bidestillata (80:5:15) was chosen with a wavelength of 272 nm. With a symmetry result of more than 0.5 and the N value is 3193, the fast retention time is 1.48 minutes. From the optimization results above, further validation of the analytical method for the selected conditions is carried out.

Validation of Analysis Method
The results of the system suitability test that have been carried out by injecting six times and looking at the repeatability parameters are each for retention time producing an RSD value of 0.253% and area area yielding an RSD of 0.3%. The results of the repeatability parameters of retention time and area area meet the acceptance requirements, namely RSD < 2% [35].
Selectivity is the ability to distinguish a number of compounds from one another or to separate the analyte from other compounds such as degradation products, metabolites and impurities accurately. The selectivity was determined by injecting the caffeine standard (figure 3a) measured by the HPLC system, the soursop leaf ethanol extract sample (figure 3b) was measured by the HPLC system, the caffeine standard plus other analytes (figure 3c) was measured by the HPLC system, and the sample of the leaf ethanol extract. soursop plus other analytes (3d image) was measured by HPLC system. Then the results of the chromatogram profile were seen for the resolution value (Rs) 1.5 [37]. The results of the selectivity determination are shown in Table 3. From the results above, it shows that the selectivity test that has been carried out has met the acceptance requirements with the resolution value of caffeine in the soursop leaf ethanol extract sample and the standard meets the resolution criteria of 1.5 so it is said that the analytical method used is able to separate caffeine from other (selective) analytes [7]. The results above shows with match factor value > 990. The linear regression equation from the linearity test is y = 219.97226x -2.01914. The correlation coefficient (r) > 0.9995 and the value of Vx0 is less than 5%, so it can be concluded that the developed method shows a linear relationship based on the required parameters [36]. LOD and LOQ tests showed that the values were 8.78525 ppm and 29.28418 ppm, respectively, using the calibration curve method. Meanwhile, in the precision test by adopting the determination of accuracy with the standard addition of caffeine with concentrations of 80%, 100%, and 120% in the test sample and then replicated six times, the coefficient of variation obtained is 1.0453%, which is less than 2% and meet the acceptance requirements of the precision parameter [37].  From the table 4, the average recovery (%) meets the specified range, which is 85-110% for the analyte in the sample with a concentration of 0.01% [37]. Each analytical method development requires further validation, to ensure that the method to be applied meets the validation requirements in evaluating the quality assurance of a product. The application of validation of the HPLC analysis method for alkaloids in various samples was also carried out in research conducted by [41] and [42]. The validation results above meet the validation parameters according to research conducted [43], [44]on the validation of the method of determining caffeine content in coffee grounds. using HPLC get results that meet the criteria for validation parameters including selectivity, specificity, linearity, accuracy, precision, LOD and LOQ.

Determination of Caffeine Content using the HPLC Method
Determination of total alkaloid content in samples of soursop leaf ethanol extract was determined when the validation results of the analytical method met the acceptance of each parameter. Determination of total alkaloid content was carried out under selected conditions from reversed phase HPLC using acetonitrile: methanol: aqua bidestillata (80:5:15), porosshell column 120 EC C-18 as stationary phase, flow rate 1.0 mL/min, Photodiode Array Detector (PDA) at 272 nm as detector. The results of the determination of total alkaloids were calculated using a calibration curve with the equation from the figure below The results of the calculation using the equation y = 0.75852x + 106.19513 with r close to 1, namely the obtained levels of 0.0862 ± 0.004 % (w/w) of 100% standard caffeine with three replications. In this study, caffeine was used as the standard because it was known from previous research conducted by [45] that caffeine is one of the alkaloid groups with the chemical formula C8H10NO2.The presence of alkaloids in the ethanol extract of the soursop cycle shows that it has potential as a traditional medicinal plant that can be developed. According to [25], that alkaloids have C-N main bonds which are divided into three types, namely pseudoalkaloids, true alkaloids and protoalkaloids. Included in the pseudoalkaloids are xanthines (theophylline, caffeine and theobromine). The true alkaloids are found in the form of N-oxides, salts, or free including cocaine, morphine, nicotine and quinine. While those included in the protoalkaloids or simple alkaloids are mescaline, ephedrine and hordenine.

Conclusion
This reverse phase high performance liquid chromatography analysis method was developed for the determination of total alkaloid content in soursop leaf ethanol extract using a mobile phase contain acetonitrile: methanol: aqua bidestillata (80:5:15), porosshell column 120 EC C-18 as the stationary phase, the flow rate is 1.0 mL/min. Photodiode Array Detector (PAD) at a wavelength of 272 nm as a detector and an optimum concentration of 500 ppm. The selected conditions are then validated by analytical methods and it can be concluded that these conditions are in accordance with the acceptance requirements including System Conformity Test (UKS), selectivity, linearity, detection limit (LOD), quantization limit (LOQ), precision, and accuracy and can be used to determine levels compound in the sample. The total alkaloids in the ethanolic extract of soursop leaves was 0.0862 ± 0.004 % (w/w) of the total alkaloid content (TALC).

Acknowledgement
Acknowledgement to dr. Soebandi University for the grant funds, all research team who have collaborated during this research. Thank you to researchers from the Faculty of Pharmacy, University of Jember for their contribution to the development of this research. Consentration (x) Area (Y)