Green and Optimum Extraction of Total Polyphenols Content from Mitragyna speciosa Korth. Havil Leaves using Microwave- Assisted Natural Deep Eutectic Solvent Extraction

Mitragyna speciosa (Korth.) Havil is a native to Southeast Asia and belongs to the Rubiaceae family, the coffee family. M. speciosa is abundant and found growing in riverbanks, swampy areas, and inundated areas by water in some countries, such as Indonesia, Malaysia, Thailand, the Philippines, and Papua New Guinea.1 This plant has enormous potential as a source of raw materials for medicines with high export value. A local community traditionally uses the leaves of this plant to treat different diseases like fever, muscle pain, diarrhea, toughs, malarial, high blood pressure, diabetes mellitus, and worm diseases.2 Some studies have reported a pharmacological activity of M. speciosa leaves such as antinociceptive,3,4 antiinflammation,5 analgesic,6,7 opioid-like effects,8,9 and morphine withdrawal effect.10,11 However, legality is still constrained in its utilization related to the sedative effect caused by the high content of indole alkaloids. The content of major compounds in indole alkaloids is about 0.088%, knowing that it can be separated using chloroform extraction up to 95%. It can even reach 99% with some optimizations Beng et al. (2011) reported in laboratory experiments.12 This result proves that different extraction methods can avoid the unwanted effects of alkaloid compounds. Therefore, it is necessary to innovate to suppress alkaloids and increase the content of other beneficial metabolites, mainly polyphenols. Optimization needs to be done to increase the level of target secondary metabolite through a green extraction approach.


INTRODUCTION
Mitragyna speciosa (Korth.) Havil is a native to Southeast Asia and belongs to the Rubiaceae family, the coffee family. M. speciosa is abundant and found growing in riverbanks, swampy areas, and inundated areas by water in some countries, such as Indonesia, Malaysia, Thailand, the Philippines, and Papua New Guinea. 1 This plant has enormous potential as a source of raw materials for medicines with high export value. A local community traditionally uses the leaves of this plant to treat different diseases like fever, muscle pain, diarrhea, toughs, malarial, high blood pressure, diabetes mellitus, and worm diseases. 2 Some studies have reported a pharmacological activity of M. speciosa leaves such as antinociceptive, 3,4 antiinflammation, 5 analgesic, 6,7 opioid-like effects, 8,9 and morphine withdrawal effect. 10,11 However, legality is still constrained in its utilization related to the sedative effect caused by the high content of indole alkaloids. The content of major compounds in indole alkaloids is about 0.088%, knowing that it can be separated using chloroform extraction up to 95%. It can even reach 99% with some optimizations Beng et al. (2011) reported in laboratory experiments. 12 This result proves that different extraction methods can avoid the unwanted effects of alkaloid compounds. Therefore, it is necessary to innovate to suppress alkaloids and increase the content of other beneficial metabolites, mainly polyphenols. Optimization needs to be done to increase the level of target secondary metabolite through a green extraction approach.
With the growth of "Green Chemistry" in recent years, green chemistry principles approaches have gained more traction. One of the most significant features of green technology is green solvents. Natural deep eutectic solvent (NADES) is a green solvent type made up of the primary metabolites of living organisms. It is often made up of ingredients abundant in our everyday diet, making it inexpensive, safe, and sustainable. 13,14 NADES has emerged as a new type of green solvent with many good advantages, including biocompatibility, low toxicity, sustainability, ecological efficacy, and incredibly considerable solvent power. The eutectic mixtures have melting points below their components, usually above 150°C. 15 NADES comprises at least one species of the hydrogen bond donor (HBD) and the hydrogen bond (HBA), which establishes strong interactions between hydrogen bonding after mixing, leading to eutectic mixtures that are usually liquid under room-close conditions. The most common NADES compositions and pharmaceutical excipients include citric acid, lactic acid, malic acid, choline chloride, sucrose, glucose, and sorbitol.
On the other hand, conventional extraction methods have been used for a long time, but this requires a longer time and a more significant number of solvents. This is the main weakness of the conventional method. Therefore, new extraction methods with reduced solvent consumption shortened extraction time and increased attention to pollution prevention. In comparison with conventional methods, microwave-assisted extraction can improve extraction efficiency by reducing time and solvent consumption. 16 Thus, microwave-assisted natural deep eutectic solvent extraction (MANDESE) has excellent potential for extracting polyphenols compounds from natural products. MANDESE has risen rapidly in the last decade, and for most applications, it has proven to be effective in all aspects compared with conventional extraction techniques. Some studies have reported the application of the MANDESE method in extracting secondary metabolites from natural products, including extraction of alkaloids and polyphenols from Peumus boldus leaves, 17 a flavonoid from Scutellaria baicalensis, 18 polyphenols and caffeine from green coffee beans, 19 bioactive compounds from Cinnamon Bark and Sappan Wood, 20 anthocyanins from C. roseus. 21 The application and optimization of the nonconventional MANDESE method on total polyphenols content extraction from M. speciosa leaves have never been reported.
In the current study, optimization of extraction condition of nonconventional MANDESE method performed using response surface methodology (RSM). RSM is a multi-factor mathematical and statistical test for the time and number of experimental samples. 22 This method analyzes the impact of extraction factors to determine the best model from an optimum process combination of four factors and three-level using Box-Behnken Design. 23 To obtain the optimum condition, we considered using four factors: NADES composition ratio, the liquid-solid ratio, extraction time, and microwave power that significantly contributed to the total polyphenol content extraction process based on the green extraction approach. RSM with Box-Behnken Design is very efficient to optimize the extraction conditions and to achieve the best predicted model. 24,25 The purpose of this study, to apply and optimize the non-conventional MANDESE method on total polyphenols content from M. speciosa leaves. We also conducted scanning electron microscopy (SEM) imaging on both powder samples before and after the extraction process to describe better and clarify the extraction mechanism.

Preparation of Natural Deep Eutectic Solvent (NADES) as a Green Solvent
NADES preparation was conducted using different pharmaceutical excipient types, including citric acid and glucose, lactic acid and sucrose, malic acid and glucose, choline chloride, and sorbitol. Each NADES composition was prepared by heating and stirring each combination component at a specific ratio (g/g) in a tube at 60-80 o C accompanied by agitation at a speed of 3000 rpm for 40 min constantly until a homogenous mixture was formed. Each mixture was added to aqua demineralization with a ratio of 1:1 v/v and continued stirring to obtain a homogenous NADES solution. All NADES solutions obtained were stable in liquid form at room temperature storage.

Conventional Maceration Procedures
The dried sample of M. speciosa leaves (5 g) was extracted by the conventional maceration method using 50 mL ethanol 96% at room temperature for 3x24 hours. Then the extractant and the residue were separated using the Buchner funnel. The extract solution was evaporated to obtain the thick/dry extract.

Non-conventional Microwave-Assisted Natural Deep Eutectic Solvent Extraction Procedures
According to some previous studies, the extraction process was performed using the non-conventional microwave-assisted natural deep eutectic solvent extraction (MANDESE) method with modification. Briefly, a dried sample of M. speciosa leaves was extracted using microwave-assisted extraction combined with each combination of NADES composition under various conditions (Table 1). After the extraction process, the extract solution was separated with its residue by filtration using 0.45 cellulose acetate membrane and evaporated to dry using a domestic food dehydrator. The dry extract was stored in a tightly closed container until ready to use.

Determination of Total Polyphenols Content
The previous study determined total polyphenols contents using Folin-Ciocalteu reagent by spectrophotometry at the maximum wavelength with a range of 730-760 nm. 26 Briefly, standard and sample solution (1 mL) was added to 0.5 mL of Colin-Ciocalteu reagent and 5 mL of distilled water. The mixture was homogenized for 1 min and then allowed to stand for 4 min. Then 1.5 mL distilled water and 2 mL sodium carbonate were added and homogenized for 1 min. Next, the mixture of the sample test was incubated for one h at room temperature. A spectrophotometer measured the absorbance at 746 nm. Gallic acid solution with some different concentrations (12.5, 25, 50, 100, and 200 µg/mL) was used as a standard to obtain a linear regression equation (1), Y = 0.0022X -0.00095, where R2 = 0.9977. Where coefficient correlation (R 2 ) value of 0.998, y is absorbance, and x is total polyphenols content. The total polyphenols content (in µg GAE/g sample) was determined using this equation.

Scanning Electron Microscopy Imaging
Scanning electron microscopy (SEM) imaging was carried out to study the extraction mechanism and compare changes in the structure of the sample before and after the extraction process related to works of literature. 27,28,29 Briefly, each dried powder of sample (after and before

Factors
Unit Symbol

Optimization of MANDESE Conditions using RSM
Optimization of MANDESE condition was performed using Box-Behnken Design with four factors and three levels (in Table 1), with independent variables including NADES composition ratio (X 1 ), liquid-solid ratio (X 2 ), Extraction time (X 3 ), and microwave power (X 4 ). In contrast, the dependent variable is total polyphenols content as a response. Based on total polyphenols content extraction efficiency, all factors and levels were expected to achieve optimum MANDESE conditions significantly. The variation condition of MANDESE was simulated with RSM using Box-Behnken Design to obtain 29 experiment samples run. A total of 29 experiments were needed to build the math equation formula (2), 30 as follow: The regression model was calculated based on 29 experiments data from the independent and dependent variables by the multilinear quadratic model using the licensed Design Expert v12 program. The best and significant model was determined according to the R 2 value and analysis of variance (ANOVA).

Selection of NADES Compositions and Total Polyphenols Content Determination
In the present study, the selection of solvent is an essential step in the extraction process. Some NADES compositions were used to determine the effectiveness of extraction on the total polyphenols content of M. speciosa leaves, compared with the conventional maceration extraction method using ethanol as a solvent (Figure 1). The extraction of the target secondary metabolites from the matrix sample of plants using several NADES compositions according to the solubility balance.
Figure 1 demonstrated that the extraction process using the nonconventional microwave-assisted natural deep eutectic solvent extraction (MANDESE) method (with some different NADES composition) has a higher efficiency of extracting the target secondary metabolite than the conventional maceration method. Meanwhile, the use of the NADES composition with various combinations of pharmaceutical excipient mixtures at the same ratio (1:1 g/g) showed differences in the ability to extract target secondary metabolite (total polyphenols content), including choline chloride-sorbitol, malic acid-glucose, citric acid-glucose, and lactic acid sucrose, respectively. Therefore, the combination of choline chloride-sorbitol was used as the NADES composition of green solvent to optimize the extraction condition using the MANDESE method. In this study, total polyphenols contents (in µg GAE/g sample) were calculated using equation (1).

Scanning Electron Microscopy Imaging
Scanning electron microscopy imaging was conducted to examine the effect of conventional maceration and non-conventional MANDESE methods and understand the extraction mechanism. Figure 2 shows the micrographs of the dried leave of samples before extraction (1), after extraction by maceration (2), and after extraction by the MANDESE method (3). The result shows that the levels of damage of cells and cell walls were more severe after extraction using the MANDESE method than the maceration method.

Optimization of MANDESE Method Using RSM
The MANDESE method was optimized using RSM to obtain the optimum extraction condition on total polyphenol content from M. speciosa leaves. The experimental design was built with four factors and three levels (Box-Behnken Design) using the licensed Design Expert v12 program, such as NADES ratio of choline chloride and sorbitol (1:1, 2:1, and 3:1 g/g), a liquid-solid ratio of NADES solution and dried leaves powder of sample (10, 20, and 30 mL/g), extraction time (10, 15, and 20 min), and microwave power (40, 50, and 60 %watts).
In this work, a total of 29 experiments were carried out according to the experimental design (Box-Behnken Design) of RSM to predict the optimum extraction condition of total polyphenols content from M. speciosa leaves ( Table 2).
The extraction process According to the results obtained using the licensed Design Expert v12 program shows the highest total polyphenols content from M. speciosa leaves were obtained from run 24 th with the yield of 495.12 µg GAE/g sample and the lowest the yield Where coefficient correlation (R 2 ) value of 0.9994, Y is total polyphenols content (µg GAE/g sample), X 1 is NADES composition ratio (g/g), X 2 is extraction time (min), X 3 is a liquid-solid ratio (mL/g), and X 4 is microwave power (%Watts). Table 3, based on the analysis of variance (ANOVA) from total polyphenols content, shows that the resulting model has a significant effect on the total polyphenols content with an F-value of 341.70 (refers to p-value < 0.05), where a 0.01% chance the F-value can occur due to an error or interference. In this case, X 1 , X 2 , X 3 , X 4 , X 1 X 2 , X 1 X 3 , X 1 X 4 , X 2 X 3 , X 2 X 4 , X 3 X 4 , X 1 2 , X 2 2 , X 3 2 , X 4 2 , X 1 2 X 2 , X 1 2 X 3 , X 1 X 2 2 , X 1 X 3 2 , X 2 2 X 3 , X 2 2 X 4 , X 1 2 X 2 2 , X 1 2 X 3 2 are significant model term. On the other hand, the F-value of "Lack of Fit" was 0.47 (with a p-value of 0.5293), which indicated that the p-value > 0.005 was not significant to the pure error. There is a 52.93% chance that a Lack of Fit F-value this large could occur due to noise. In addition, the optimum condition from the equation model can be applied as navigation to design the desired optimum extraction condition.
In Table 4, the coefficient estimates represent the expected change in response per unit in factor value when all other factors are constant. Orthogonal design intercept is the overall average response of all analyses. The coefficient fits the mean based on the coefficient set. If the factors are orthogonal, the variance inflation factor (VIF) is 1. VIF greater than 1 indicates multicollinearity, and the higher the VIF, the stronger the correlation of the factors. As a hard rule, a VIF below 10 is acceptable. The equation in phrases of codec factors can be applied to predict approximately the response for given tiers of every factor. By default, the excessive tiers of the factors are coded as +1, and low tiers are coded as -1. The codec equation is beneficial for figuring out the relative effect of the factor through evaluating the factor coefficients.
The three-dimension (3D) response graph and contour plot for the effect of various extraction variable factors on total polyphenol content was presented in Figure 3. It demonstrated that the curvature of these plots indicates the interaction and relationship between each factor that are variable parameters of extraction condition, including microwave power and NADES composition ratio; microwave power and extraction time; microwave power and liquid-solid ratio; extraction time and liquid-solid ratio; liquid-solid ration and NADES composition ratio; and Extraction time and NADES composition ratio. In general, with view glance from the results, the efficiency of polyphenols extraction from this plant was strongly influenced by these various variable factors used.

DISCUSSION
In the present study, the NADES composition combined with choline chloride and sorbitol as a green solvent extracted secondary metabolites with the highest total polyphenol content compared with  The above results show that a green solvent of NADES composition with choline chloride-sorbitol was more effective than the others. This solvent is twenty times as strong as that of ethanol. Other parameters, such as microwave power, extraction time, and liquid-solid ratio, also significantly impact the extraction efficiency. As a result, it becomes the essential aspect to consider in selecting these four factors in optimizing the extraction conditions.
On the other hand, the non-conventional microwave-assisted natural deep eutectic solvent extraction (MANDESE) method provides several advantages, including non-volatile, low-toxicity, operator convenience and safety, and environmentally friendliness, 20,31,32 as well as being available in the laboratory. This result is in line with the results of SEM imaging, which shows that the level of cell wall damage in the sample matrix has a good correlation with the level of extraction efficiency. Each extraction method, which is compared to the sample matrix before the extraction procedure for both samples, can demonstrate physical, structural changes in the cell wall of the sample matrix. The effect of the extraction process on the extent of damage to the cell wall matrix of samples has been revealed in several papers. 29,33,34 Based on our findings, the optimum condition of the MANDESE method was obtained regarding the RSM analysis result using the licensed Design Expert v12 software as follows: NADES composition ratio of 3 g/g (choline chloride/sorbitol) and the liquid-solid ratio of 20 mL/g for 20 min extraction time with 60% Watts microwave power with total polyphenols content prediction of 539.37 µg GAE/g sample of M. speciosa leaves. We performed a scale-up confirmation test to prove the accuracy of the prediction of the optimum conditions obtained.
A 50 g dried sample of M. speciosa leaves was extracted under the recommended optimum conditions, providing the approximately total polyphenols content value of 526.12 µg GAE/g sample (with a standard deviation of 5.418 in triple repetition).
A combination of choline chloride and sorbitol with various ratios from 1 to 3 g/g was used in this study. The interaction impact of many aspects and sample characteristics contribute to the variation in optimum conditions. NADES composition with choline chloride and sorbitol was chosen and implemented in this study based on the selectivity efficiency in extracting total polyphenols content. Both materials mix solids and liquids, with sorbitol acting as a hydrogen-bonding donor (HBD) and choline chloride acting as a hydrogen bond acceptor (HBA). 35,36,37 At temperatures less than 100 o C, the deep eutectic solvent in liquid was formed by adjusting the NADES composition ratio at a specific proportion (HBD and HBA mixture). 38,37 Furthermore, different liquid-solid ratios, such as 10, 20, and 30 mL/g was used to test the effect. In this work, the increase in the liquidsolid ratio causes maximal direct contact between target secondary metabolites and the green solvent in the matrix sample under these conditions. 39 However, if it is too large or has a high ratio, it will cause a waste of material used. A low ratio can result in an inefficient extraction process. 40 The optimum condition of the MANDESE method with NADES composition and the liquid-solid ratio was 3 g/g (choline chloride-sorbitol) and 20 mL/g (NADES solution-sample).
While the influence of various conditions of the extraction equipment, such as microwave power and extraction time, dramatically affects the level of extraction efficiency of the total polyphenol content. In this study, the effect of microwave power was tested at various levels extending from 40 to 60 % Watts. When the microwave power is increased, the temperature rises. A temperature increase that is too high can damage the sample matrix and generate a change in the target compound's structure. 16,26 The extraction process was carried out in different time variations, including 10, 15, and 20 min, based on some previous studies. 41 -43 This result shows that the highest total polyphenols content of M. speciosa leaves was achieved at 20 minutes for optimal extraction time. This situation shows that the dissolution process of the target secondary metabolite in the sample is in equilibrium 44 . Based on these two conditions, there is a tendency to increase the yield of the total polyphenol content. However, the increase in microwave power and extraction time was limited because the resulting temperature could not be regulated. Excessive temperatures should be avoided because most secondary metabolites decompose at high temperatures.
Overall, these findings indicate that various variable factors used proved to significantly affect the extraction efficiency of total polyphenols from M. speciosa leaves. However, our findings should be interpreted with caution, given several limitations. First, the target secondary metabolites (total polyphenols content) are compounds measured using the UV-Vis spectrophotometry method with Folin-Ciocalteu reagent, calculated based on equivalence and only using standard gallic acid. Therefore, prospective studies using pure compounds as the target to ensure the selectivity of green solvent media used with more sensitive analytical methods (e.g., HPLC) are needed to confirm our findings. Second, in this study, only four combinations of excipients were used as the composition of NADES, so that the best NADES composition used had not been compared with other materials that may have high selectivity potential when used as a solvent. Third, the selectivity of the target compound only focuses on the efficiency level concerning polyphenol levels, has not considered the possibility of other secondary metabolite groups, so further analysis is needed.
Despite these limitations, to the best of our knowledge, this is the first time reported regarding the use and optimization of the MANDESE method to increase the efficiency of polyphenol extraction from M. speciosa leaves. This finding adds to information related to the use of NADES as a green solvent and optimum medium for extracting target secondary metabolites (especially the polyphenol group), considering that this group of compounds from plants has not been widely studied developed for its benefits.