Optimization of ultrasound-assisted parthenolide extraction from Tarchonanthus camphoratus leaves using response surface methodology: HPTLC and cytotoxicity analysis

Nasir A. Siddiqui; Perwez Alam; Adnan J. Alrehaily; Ali S. Alqahtani; Ali Akhtar; Tawfeq A. Alhowiriny; Omer M. Almarfadi; Ramzi A. Mothana

doi:10.1016/j.arabjc.2021.103194

View/Download PDF

Buy Reprints

PDF

Translate this page into:

Original article

2021

:14;

202106

doi:

10.1016/j.arabjc.2021.103194

Optimization of ultrasound-assisted parthenolide extraction from Tarchonanthus camphoratus leaves using response surface methodology: HPTLC and cytotoxicity analysis

Nasir A. Siddiqui^{^a,⁎}, Perwez Alam^{^a}, Adnan J. Alrehaily^{^a}, Ali S. Alqahtani^{^a}, Ali Akhtar^{^b}, Tawfeq A. Alhowiriny^{^a}, Omer M. Almarfadi^{^a}, Ramzi A. Mothana^{^a}

a Department of Pharmacognosy, College of Pharmacy, King Saud University, Riyadh, Saudi Arabia

b IT & Quality Unit, College of Pharmacy, King Saud University, Riyadh, Saudi Arabia

⁎Corresponding author. nsiddiqui@ksu.edu.sa (Nasir A. Siddiqui),

Received: 2021-2-21, Accepted: 2021-4-25,

Disclaimer:
This article was originally published by Elsevier and was migrated to Scientific Scholar after the change of Publisher.

Abstract

The aim of this study was to optimize the ultrasonication method for efficient extraction of parthenolide from Tarchonanthus camphoratus leaves using Box–Behnken design for response surface methodology (RSM) and then to validate the method by high performance thin layer chromatography (HPTLC). RSM was used to optimize the extraction parameters, i.e., temperature (25–45 °C), time (35–55 min), and liquid-to-solid ratio (16–24 mL/g), to maximize parthenolide yield. Quantitative estimation of parthenolide in ethanol extract (TCEE) of T. camphoratus leaves was conducted to check the effect of all the three variables and validated by HPTLC method using n-hexane and ethyl acetate (3:1, v/v) as mobile phase. A quadratic polynomial model was found to be most suitable with regard to R (parthenolide yield), with R²/%CV = 0.9973/0.4557. The adjusted R², predicted R², and signal-to-noise ratio for R were 0.9937, 0.9870, and 47.94, respectively, indicating a high level of tie-in with adequate signal. There was a strong linear correlation between the predicted and experimental R values (R² = 0.9973). In TCEE, parthenolide was detected with a R_f value of 0.15 at λmax = 590 nm. The optimized ultrasonic extraction produced 1.010% ± 0.04 %w/w with the following extraction parameters: 38.8 °C (M₁), 50-min extraction time (M₂), and 20.4-mL/g liquid-to-solid ratio (M₃). The obtained results of optimization also endorsed by cytotoxicity results (CC₅₀125.47 μg/mL) on human hepatoma cells (HuH7).

Keywords

HP-TLC

Response surface methodology

Parthenolide

Optimization

Cytotoxicity

Show Related Articles from PubMed

1 Introduction

Parthenolide, a sesquiterpene lactone, has gained considerable popularity as a treatment for cancer, migraine, and arthritis (Berry, 1984; Johnson et al., 1985; Groenewegen and Heptinstall, 1990 ). Currently, Tanacetum parthenium (Chrysanthemum parthenium or feverfew) is considered the best source of parthenolide (0.85%, w/w); it is established as one of the leading herbal medicines for arthritis and migraine (Retnik et al., 2005). However, Tarchonanthus camphoratus is emerging as a substitute of feverfew in terms of its parthenolide (Fig. 1) concentrations in different parts of the plant. Nevertheless, the nonuniform distribution of the phytoconstituent in different plant organs, its solubility in various solvents, and temperature of extraction, among other parameters, must still be optimized to improve yield of the selected phytoconstituent. Parthenolide is found in almost all plant parts, but mainly in the stem and leaves of T. camphoratus.

One of the conventional methods of extraction for sesquiterpene lactones (e.g., parthenolide) was developed in 1970 using chloroform and petroleum ether as the extractive solvent and NMR as the identification tool (Yoshioka et al., 1970). Other extraction methods including different solvents, such as chloroform by Soxhlet extraction (Rey et al., 1992) and the bottle stirring method with acetonitrile and 10% water (v/v), can be used to extract acceptable amounts of parthenolide from T. parthenium; however, nonpolar solvents are not useful for parthenolide isolation (Zhou et al., 1999).

Ultrasonication is now applied in areas such as pectin technology, production of tinctures and extraction from medicinal herbs (Hromadkova and Ebringerova, 2003). Ultrasound is considered a “green technology” because of its high productivity, less instruments involved, and short processing time relative to traditional techniques. Although ultrasonication is a comparatively new extraction technique, it is already considered effective on both small and large scales (Zhao et al., 2013).

Response surface methodology (RSM) is an impressive statistical technique for optimization involving many variables. Box–Behnken design (BBD), a version of RSM, operated at least on three levels (low, medium and high: coded − 1, 0, and + 1) and requires few experiments. In the present study, an ultrasound extraction technique was applied for extraction of parthenolide from T. camphoratus leaves considering three independent variables, i.e., temperature, time, and liquid-to-solid ratio, with RSM (BBD) used for optimization (Alam et al., 2020).

Usually different assay methods are used to monitor the concentration of active ingredients in various dosage forms prepared for medicinal use. The reported assay methods for analyzing parthenolide in feverfew include selective solvent extraction followed by infrared spectrometry (Bloszyk et al., 1978) or high-performance liquid chromatography (French Pharmacopoeia, 1987; Fontanel et al., 1990; Awang et al., 1991 ). The present study is based on identification and quantification of parthenolide using high-performance thin-layer chromatography to validate the RSM findings.

Feverfew is one herb in the market known for relieving different types of pain, e.g., migraine-related pain (Brown et al., 1996). The pharmacological action of parthenolide resembles aspirin and prevents platelets aggregation and curbs the release of serotonin and some inflammatory markers (Retnik et al., 2005). As per the literature the maximum content of parthenolide has been reported in the flower heads of feverfew followed by the leaves, stalks and roots (Heptinstall et al., 1992). Studies have shown that parthenolide is effective in inhibiting some tumor cells and proinflammatory transcriptional nuclear factors NFκB (Sohma et al., 2011). Moreover, several plant-derived parthenolides have been shown to have anticancer or cytotoxic activity against a range of human cancer cell lines (Al-Fatlawi et al., 2015; Bosio et al., 2015; Tsai et al., 2015; George et al., 2016; Lin et al., 2017; Alwaseem et al., 2018 ).

Given the popularity of feverfew and the medicinal importance of parthenolide, researchers have also attempted to identify plant alternatives. Here, the cytotoxic potential of T. camphoratus extract against human hepatoma cells (HuH7) was assessed for the first time along with the extraction efficiency of the ultrasonication method using RSM and its validation by HPTLC method.

2 Materials and methods

2.1

2.1 Plant material

The plants T. camphoratus (voucher specimen number: 15451) were collected from WADI GAMMA, Abha region of Saudi Arabia and identified by Dr. Mohammed Yusuf, Taxonomist, Medicinal Plant Collection and Survey Unit, Department of Pharmacognosy (Specimens of plant were also deposited in departmental herbarium), CoP, KSU, Saudi Arabia.

2.2

2.2 Apparatus and reagents

Standard parthenolide was obtained from Sigma Aldrich (St. Louis, MI, USA). Analytical grade reagents and solvents (ethyl acetate, n-hexane, and p-anisaldehyde) were purchased from WINLAB and BDH (UK). Glass-backed silica gel 60F₂₅₄ HPTLC precoated plates (20 × 10 cm) were procured from Merck (Darmstadt, Germany). The standard and the extracts were applied to HPTLC plates band-wise with the use of a CAMAG automatic TLC sampler-4 (CAMAG, Switzerland) and developed in an ADC2 (automatic development chamber; CAMAG, Muttenz, Switzerland). TLC Plates were then documented using a CAMAG TLC Reprostar 3 and scanned with CATS 4 (CAMAG).

2.3

2.3 Effect of single factor tests on ultrasonic extraction of T. camphoratus leaves

The extraction of powdered leaves of T. camphoratus in ethanol was conducted by ultrasonic vibrations (ultrasound-assisted extraction) using a Sonics Vibra Cell (Model VCX-750; Sonics, USA). The effects of single factors on extraction procedures were determined as follows.

(1)

Effect of extraction temperature: Ethanol (20 mL) was added to 1 g of powdered leaves in a 50-mL flask for each experiment and the extraction was performed for different extraction temperatures (25 °C, 30 °C, 35 °C, 40 °C, 45 °C, and 50 °C) with a constant extraction time (40 min) and LSR (20 mL/g). Each experiment was repeated three times (n = 3). The obtained extract of each experiment was filtered and collected in a dried conical flask, and the remaining mark was washed thrice with the ethanol and filtered. All the filtrates of each experiment were merged and dried at low pressure with a Rotavapor to obtain the final extractive yield.
(2)

Effect of extraction time: The exact procedure used to test the effect of extraction temperature was used to test the effect of extraction time, except that the extractions were performed using different extraction times (20, 30, 40, 50, 60, and 70 min) at a constant extraction temperature (30 °C) and LSR (20 mL/g).
(3)

Effect of liquid-to-solid ratio (LSR): The exact procedure used to test the effect of extraction temperature was used to test the effect of the LSR, except that various volumes of ethanol (8, 12, 16, 20, 24, and 28 mL) were used to achieved different LSR with a constant extraction time (40 min) and temperature (30 °C).

2.4

2.4 Experimental design of RSM

Ethanol extract of T. camphoratus (TCEE) was used to conduct preliminary study to detect the presence of parthenolide. A response surface methodology based three factorial (3³) BBD (version 12, Stat-Ease Inc., USA) was used to optimize the extraction variables. Temperature (M₁), time (M₂), and LSR (M₃) were used as three independent variables to optimize with 17 runs to get the maximum yield of dependent variable parthenolide (R). Single factor experiment was used to determine the appropriate range of the variables. The preparation and analysis of all the samples were conducted in triplicate. The various independent and dependent variables are listed in Table 1.

Table 1 Extraction variables selected for Box–Behnken design optimization.

	Factor level			Dependent Variable	Goal
Independent Variable	−1	0	+1	Parthenolide yield (% w/w) (R)	Maximized
Temperature (°C) (M₁)	25	35	45
Sonication time (min) (M₂)	35	45	55
Liquid to solid ratio (mL/g) (M₃)	16	20	24

*Goal: Objective or target of optimization.

The nonlinear quadratic model equation produced by this experimental design: $R = k_{0} + n_{1} M_{1} + n_{2} M_{2} + n_{3} M_{3} - n_{12} M_{1} M_{2} + n_{13} M_{1} M_{3} + n_{23} M_{2} M_{3} - n_{11} M_{1}^{2} + n_{22} M_{2}^{2} + n_{33} M_{3}^{2},$ where R is the response related to each factor level combination; k₀ is the intercept coefficient; n₁, n₂, and n₃ are linear coefficients; n₁₂, n₁₃, and n₂₃ are the interaction coefficients; and n₁₁, n₂₂, and n₃₃ are the quadratic coefficients. The independent variables were M₁, M₂, and M₃, whereas R was the dependent variable. The results of single factor tests were utilized to choose the independent variables range.

2.5

2.5 RSM model and validity testing

The experimental results were analyzed by the BBD of RSM where p-values ≤0.05 considered as significant. The extraction independent variables (e.g., M₁, M₂, and M₃) were concurrently optimized using BBD. Parthenolide was ultrasonically extracted in triplicate using the optimized conditions and the model was validated by comparing the experimental value of parthenolide with predicted value.

2.6

2.6 HPTLC system conditions

The analysis was done on two 20 × 10 cm glass-backed silica gel 60 F₂₅₄ HPTLC plates. An automatic TLC Sampler 4 (ATS4) (CAMAG) fitted with a Hamilton Gastight Syringe (1700 Series; volume = 25 µL) was used to apply the samples and standard on the HPTLC plate with an application rate of 160 nL/s. The plate was developed in a previously saturated (for 20 min at 22 °C with mobile phase vapor) ADC2 in linear ascending mode with n-hexane:ethyl acetate (3:1, v/v) as the mobile phase. A CAMAG TLC scanner IV was used to scan the developed and derivatized (p-anisaldehyde) plates at a wavelength of 590 nm in absorbance mode using a deuterium lamp. The slit dimensions were 4.00 × 0.45 mm and the scanning speed was 20 mm/s.

2.7

2.7 Preparation of standard stock solution

Stock solution of the standard (parthenolide) was prepared according to the method mentioned by Siddiqui et al., 2018 with slight modification. For calibration, 1–14 µL of this standard solution was applied to the HPTLC plate to provide a concentration range of 100–1400 ng/band producing eight point calibration graph.

2.8

2.8 Method development and validation

After trying several combinations of solvents the chromatogram for standard parthenolide was developed. The combination of n-hexane and ethyl acetate in the ratio of 3:1, v/v was found to produce best resolution. The similar combination was used for detection of parthenolide in BBD-run samples. The optimized saturation time was 20 min. Densitometric analysis was performed at absorption maxima ( λmax = 590 nm) in absorbance mode (Siddiqui et al., 2018). Guidelines from International Conference on Harmonization were followed for method validation using parameters like linearity, precision, accuracy, limit of detection (LOD), limit of quantification (LOQ), and robustness (ICH Guidelines, 2005).

2.9

2.9 Estimation of parthenolide

Samples for BBD and standard (parthenolide) were incorporated on HPTLC plates. The peak areas of standard and test samples were considered for estimation of parthenolide.

2.10

2.10 Cell culture and cytotoxicity assay of the BBD-run TCEE samples

The HuH7 cell line was maintained in culture media [DMEM–Glutmax supplemented with bovine calf serum (10%; Invitrogen, USA) and penicillin–streptomycin (×1; Invitrogen, USA)] at 37 °C with 5% CO₂ in a humid chamber. HuH7 cells (0.5 × 10⁵ cells/100 μL/well) were seeded in a 96-well flat-bottom culture plate (Becton-Dickinson Labware, USA) a day before treatment. A stock of TCEE was prepared by first dissolving in 100 μL DMSO (Sigma) and then in DMEM (100 mg/mL: final concentration); subsequently, four working concentrations/doses (200, 100, 50, and 25 μg/mL) were reconstituted in DMEM. The final concentration of DMSO never exceeded 0.1% in the treatment doses. Cells were treated with triplicate doses of each extract, including an untreated control (0.1% DMSO), and incubated for 48 h at 37 °C. Inhibition of cell proliferation or toxicity was tested using an MTT assay as per the kit manual. In brief, MTT reagent (10 μL/well) was added and the plate was incubated at room temperature for around 4 h in the dark until a purple color appeared; at this point, a detergent solution (100 μL/well) was added and the plate was incubated at 37 °C for 1.5 h. The absorbance (λ = 570 nm) was measured (Microplate Reader ELx800; BioTek, USA) and data were analyzed for the percentage of cell survival in relation to the untreated control: [(As − Ab)/(Ac − Ab)] × 100, where As, Ab, and Ac represent the absorbance of the sample, blank, and negative control, respectively]. The survival curve of HuH7 cells was extracted from the plot between cell survival fraction and extract concentration. The CC₅₀ values of each extract were determined using the best fit regression curve method in Excel (Microsoft, USA).

3 Results

3.1

3.1 Effect of single factor tests on ultrasonic extraction of T. camphoratus leaves

3.1.1

3.1.1 Extraction temperature

It is very clear from Fig. 2A that the total extraction yield (% w/w) increased with increase in temperature and reached a maximum yield at 40 °C, with no significant difference was observed by further increase in temperature above 40 °C.

The effects of single factor on total extraction yield of TCEE. (A) Effect of extraction temperature (at time = 40 min and Liquid to solid ratio: 20 mL/g); (B) Effect of extraction time (at temperature = 30 °C and Liquid to solid ratio: 20 mL/g); (C) Effect of liquid-to-solid ratio (at temperature = 30 °C and time = 40 min). Each value represents a mean ± SD (n =3 ).

3.1.2

3.1.2 Extraction time

As shown in Fig. 2B, the total extraction yield (% w/w) was evidently affected by variation in extraction time and it increased significantly from 12.48% to 17.95% with extraction times of 20 to 50 min, however no significant change was observed with further increases in time above 50 min. Thus, 50 min was the time limit to achieve maximum extraction yield.

3.1.3

3.1.3 Liquid-to-solid ratio (LSR)

The total extraction yield (% w/w) increased significantly from 10.24% to 16.87% as the LSR increased from 8 to 20 mL/g (Fig. 2C), probably because of an increase in the driving force for mass transfer. For LSR > 20 mL/g, the total yields did not differ significantly.

Based on the above findings, the RSM optimized ultrasound-assisted extraction procedure can be summarized as: to the leaf powder (1 g), ethanol was added with different LSR (16–24 mL/g), and the extraction took place at 25–45 °C over a period of 35–55 min with three replicates. The extracts from these three replicates were filtered using filter paper, and used to determine the parthenolide content by HPTLC.

3.2

3.2 Development and validation of HPTLC method

A combination of the solvents n-hexane and ethyl acetate at (3:1,v/v) was proved as the most suitable solvent system; it produced a sharp peak in parthenolide at R_f values of 0.15 (Fig. 3A). The method was found to be quite selective and effectively separated the different components of the samples (Fig. 3B) at λmax = 590 nm in absorbance mode.

Quantification of parthenolide in BBD run TCEE samples by HPTLC (λmax = 590 nm). (A) Chromatogram of standard parthenolide (Rf = 0.15 ± 0.001); (B) Pictogram of developed TLC plate derivatized with p-anisaldehyde; (C) Chromatogram of TCEE sample (parthenolide, spot 4 , Rf = 0.15); (D) 3-D display of all tracks at 590 nm.

The calibration curve of parthenolide was linear at 100 –1400 ng/spot; a strong linear relationship was supported by linear regression data (r² = 0.9943, P < 0.001; linear regression equation: Y = 4.54X + 1171.76, where Y and X are the response and the amount of reference standard, respectively; Supplementary Table 1). The accuracy was calculated by recovery analysis, which afforded recovery of 98.26–99.24% (the specific values are listed in Supplementary Table 2). Low values of % RSD (1.09–1.48%) indicated that the proposed method was highly accurate. Statistical analysis showed that the developed method was reproducible and selective. Intraday and interday precision of the parthenolide assay at three different concentrations (400, 600, and 800 ng/band) were expressed as % RSD at 1.17–1.56% and 1.16–1.54%, respectively (Supplementary Table 3); such low values indicated the high precision of the method. Low % RSD values also proved the robustness of the proposed HPTLC method (1.32–1.37%; Supplementary Table 4). The LOD & LOQ of the proposed method were 31.12 and 99.39 ng/spot, respectively (Supplementary Table 1), suggesting that the method could potentially be used over a wide range for detection and quantification of parthenolide.

3.3

3.3 RSM model fitting

The quantity (% w/w) of parthenolide (R) was analysed by HPTLC in each BBD-run sample (Table 2). A quadratic model was fit for the analysis of parthenolide and the results of regression analysis, and response regression equation for the proposed model are provided in Table 3. The values of R², adjusted R², and predicted R² were found as 0.9973, 0.9937, and 0.9870, respectively, indicating strong coherence between the observed and predicted values. Additionally, the difference in values of adjusted R² and predicted R² was less than 2, which indicated that the model was fit. “Adequate Precision (signal-to-noise ratio) of the model was found as, 47.94 which is more than (˃) 4, indicated that model was fit. ANOVA results for the fitted quadratic polynomial are listed in Table 4. The “Lack of Fit F-value” was found (0.3790), not significant which validated RSM results. The high F-value (282.83) for the model suggested that the model was significant. The coded quadratic equation generated by the model was as follows: $R (%) = 0.9883 + 0.0523 M_{1} + 0.0 . 196 M_{2} + 0.0069 M_{3} - 0.010 M_{1} M_{2} + 0.006 M_{1} M_{3} - 0.0088 M_{2} M_{3} - 0.0547 M_{1}^{2} - 0.0206 M_{2}^{2} - 0.0316 M_{3}^{2}$

Table 2 Response surface central composite design (uncoded) and result for R.

Run	Coded variables			Actual variables			Parthenolide yield (R) (% w/w)
Run	(M₁) (°C)	(M₂) (min)	(M₃) (mL/g)	(M₁) (°C)	(M₂) (min)	(M₃) (mL/g)	Experimental	Predicted	Residue
1	−1	0	1	25	45	24	0.724 ± 0.039	0.727	−0.003
2	0	0	0	35	45	20	0.978 ± 0.055	0.980	−0.002
3	0	0	0	35	45	20	0.974 ± 0.041	0.9754	0.0014
4	1	1	0	45	55	20	0.955 ± 0.046	0.958	−0.003
5	0	0	0	35	45	20	0.993 ± 0.064	0.991	0.002
6	1	0	−1	45	45	16	0.886 ± 0.049	0.8842	0.0018
7	−1	1	0	25	55	20	0.795 ± 0.031	0.7964	−0.0014
8	0	−1	−1	35	35	16	0.814 ± 0.035	0.8121	0.0019
9	1	0	1	45	45	24	0.934 ± 0.057	0.9355	−0.0015
10	−1	−1	0	25	35	20	0.687 ± 0.040	0.6853	0.0017
11	0	1	1	35	55	24	0.907 ± 0.018	0.9081	−0.0011
12	1	−1	0	45	35	20	0.911 ± 0.061	0.9087	0.0023
13	0	0	0	35	45	20	0.967 ± 0.055	0.9646	0.0024
14	−1	0	−1	25	45	16	0.722 ± 0.033	0.7194	0.0026
15	0	0	0	35	45	20	0.972 ± 0.067	0.9733	−0.0013
16	0	1	−1	35	55	16	0.913 ± 0.047	0.9147	−0.0017
17	0	−1	1	35	35	24	0.873 ± 0.042	0.8718	0.0012

Table 3 Result of regression analysis for the model and response regression equation of the final proposed model.

Dependent variables	Model F-value	R²	Adjusted R²	Predicted R²	SD
R	Linear	0.5431	0.4377	0.3071	0.0405
	2F1	0.5615	0.2983	−0.1578	0.0452
	Quadratic	0.9973	0.9937	0.9870	0.0043
	Cubic	0.9979	0.8215	0.7749	0.0050

Table 4 ANOVA for the fitted quadratic polynomial model of R.

Dependent variables	Source	Sum of square	Degrees of freedom	Mean square	F-value	Prob > F
R	Model	0.0465	9	0.0052	282.83	<0.0001 (significant)
	Residual	0.0001	7	0.0000	–	–
	Lack of fit	0.0000	3	9.439E-06	0.3790	0.7742 (not significant)
	Pure error	0.0001	4	0.0000	–	–

3.4

3.4 Effect of extraction parameters (M₁, M₂, M₃) on R and RSM analysis

The influence of all independent extraction variables are listed in Table 5. The linear variables (M₁, M₂, and M₃), interaction variables (M₁M₂, M₁M₃, and M₂M₃), and quadratic variables (M₁², M₂², and M₃²) were significant (P < 0.05) and affected R. The R² and coefficient of variation (% CV) values were 0.9973 and 0.4557, respectively, indicating the high precision and reliability of the experimental values (Alam et al., 2020). Three-dimensional (3D) plots were constructed to visualize the relationship between M₁, M₂, M₃ and R.

Table 5 The significance of each response variable effect according to F ratio and P value in the nonlinear second order model.

Dependent variable	Independent variables	SS^a	DF^b	F-Value	P-value^c
R	Linear effects
	M₁	0.0219	1	1197.07	<0.0001
	M₂	0.0031	1	168.30	<0.0001
	M₃	0.0004	1	20.93	0.0026
	Quadratic effects
	M₁²	0.0126	1	688.23	<0.0001
	M₂²	0.0018	1	97.96	<0.0001
	M₃²	0.0042	1	229.94	<0.0001
	Interaction effects
	M₁M₂	0.0004	1	21.88	0.0023
	M₁M₃	0.0001	1	7.87	0.0263
	M₂M₃	0.0003	1	17.02	0.0044

a Sum of squares.

b Degrees of freedom.

c P-values <0.05 were considered significant.

Positive coefficients of variables indicate optimization, whereas negative values indicate a reverse relationship between the independent variables and response (M₁, M₂, and M₃). From the above equation, it is evident that variables such as M₁, M₂, and M₃ had positive effects on R and the interaction variables effects vary and produces positive and negative effects on R.

The combination ratio of all the variables (M₁, M₂, and M₃) for the extraction was selected based on R using 3D response surface plots. As shown in Fig. 4A and 4B, R increased positively with an increase in M₁ up to 38.8 °C when M₃ and M₂ were fixed at 20 mL/g and 45 min, respectively. Fig. 4 C shows that R increased with extended M₂ values and lower M3 values when M₁ was fixed at 35 °C.

Response surface model 3D plots showing the effects of M1, M2 and M3 on R. (A) effect of M1 and M2 on R; (B) effect of M1 and M3 on R; (C) effect of M2 and M3 on R.

3.5

3.5 RSM validation

For the M₁, M₂, and M₃ checkpoints, the results of parthenolide yield was found to be within the acceptable limits. The experimental value compared with the anticipated value to validate RSM results. The less percentage prediction error (1.05%) validated the generated equation and described the RSM model applicability. A linear correlation between the predicted and experimental values (P < 0.0001) indicated that the proposed model was fit (Fig. 5).

Linear correlation plot between actual and predicted values for R.

3.6

3.6 Optimization and verification of the model for extraction parameters

The optimum extraction process parameters were determined by maximizing the response R. The predicted optimal extraction conditions were found as 38 °C (M₁), 50 min (M₂), and 20 mL/g (M₃), which resulted in the extraction of 1.002% w/w of R. To confirm the predicted value of response variables optimum extraction conditions were revalidated for M₁ at 38.8 °C, M₂ at 50 min, and M₃ at 20.4 mL/g which furnished 1.010% ± 0.04% w/w of R. The obtained experimental value was compared with predicted value and the experimental value was found in the range of predicted value which exhibited the RSM model with a good correlation (Fig. 5). Therefore, the developed model is applicable for optimizing parthenolide extraction from T. camphoratus leaves.

3.7

3.7 Cytotoxicity of BBD-run TCEE samples

The in vitro cytotoxicity of BBD-run TCEE samples were tested in cultured HuH7 cells. Promising cytotoxicity was shown, with estimated CC₅₀ values of 125.47–261.19 μg/mL (see Table 6).

Table 6 The estimated cytotoxic concentration (CC₅₀) of BBD-run TCEE samples against a cultured human hepatoma cell line (HuH7).

Run	Coded variables			Actual variables			Cytotoxicity (HuH7)
Run	(M₁) (°C)	(M₂) (min)	(M₃) (mL/g)	(M₁) (°C)	(M₂) (min)	(M₃) (mL/g)	(CC₅₀, µg/mL)
1	−1	0	1	25	45	24	247.18
2	0	0	0	35	45	20	135.19
3	0	0	0	35	45	20	141.78
4	1	1	0	45	55	20	160.17
5	0	0	0	35	45	20	125.47
6	1	0	−1	45	45	16	187.11
7	−1	1	0	25	55	20	211.13
8	0	−1	−1	35	35	16	195.16
9	1	0	1	45	45	24	172.18
10	−1	−1	0	25	35	20	261.19
11	0	1	1	35	55	24	181.33
12	1	−1	0	45	35	20	180.09
13	0	0	0	35	45	20	153.91
14	−1	0	−1	25	45	16	223.17
15	0	0	0	35	45	20	146.57
16	0	1	−1	35	55	16	179.16
17	0	−1	1	35	35	24	189.12

Bioactive parthenolide derived from different plants sources shows in vitro anticell proliferative or cytotoxic potential against a range of human cancer cell lines. In agreement with this, we have, for the first time, demonstrated the marked cytotoxic effect of T. camphoratus leaf extracts on human hepatoma cells (i.e., HuH7 cells). Our in vitro results further endorse the HPTLC data regarding the high content of parthenolide in all tested extracts.

4 Discussion

The simultaneous quantification of parthenolide in all the fractions of TCEE collected during seventeen BBD runs was conducted using a validated HPTLC method, in which n-hexane and ethyl acetate served as a suitable mobile phase and exhibited good resolution and separation of parthenolide along with the other available phytoconstituents. The guidelines of the ICH, 2005 were used to validate the developed method. The low % RSD values for parthenolide was an indication of robustness, high accuracy and precision of the developed method.

When testing the effect of extraction temperature as a single factor for ultrasonic extraction of TCEE, we observed significant enhancement in total extract production while increasing the temperature from 25 °C to 40 °C; this likely resulted from the higher mass transfer rate at increasing temperatures up to 40 °C, which probably led to higher molecular diffusion (Wang et al., 2018). Extraction time effects the liquid circulation and turbulence produced by cavitation, which can increase extraction efficiency by increasing the contact surface area between the solvent and targeted compounds (Zhang et al., 2017). In agreement with this, the extraction yield increased with time up to 50 min in our study. In addition, the production of extract increased as the LSR increased which eventually enhances the mass transfer of soluble compounds from material to solvent (Yang et al., 2010; Xu et al., 2017).

The seventeen runs of BBD were performed and analyzed using the validated HPTLC method to help determine the quantity of parthenolide. The best fit model for BBD analysis was quadratic model. The experimental R² and predicted R² values for parthenolide were close to 1; thus, there was a strong correlation between the observed and predicted values. Other analysis values showed that the model was appropriate and RSM results were valid: the difference between the adjusted R² and predicted R² was <2, the “Adequate Precision” was >4 (indicating an adequate signal that could be used to navigate the design space), the “Lack of Fit F-value” was low for parthenolide (indicating that the “Lack of Fit” was not significant and the model was a good fit), and the “Model F-value” for R was high (implying that the model was significant).

The importance of each extraction variable’s effects on R and RSM analysis was evaluated. The interactions of M₁ and M₂, M₂ and M₃, and the square root of M₁, M₂, and M₃ had negative effects on R, whereas the interactions of M₁ and M₃ had positive effects on R, indicating that the doubling of M₁ and M₂ would substantially reduce R values.

To observe the relationships between the independent variables (M₁, M₂, and M₃) and R, 3D plots were constructed. These plots showed that M₁ (extraction temperature) had the most significant effect on R. The maximum yield of R was obtained at the optimum temperature of 38.8 °C, i.e., relatively low temperatures enhance the compound yield whereas high extraction temperatures reduce parthenolide yield. For the highest yield of parthenolide, we therefore recommend using the optimum extraction temperature, extraction time, and LSR, which were 38.8 °C, 50 min, and 20.4 mL/g, respectively, in the present study.

Sesquiterpene lactones are potent anti-cancer compound and exhibited its anticancer effect by inhibiting the activation of NF-κβ, binding of NF-κβ to DNA, interacting with DNA-NF-κβ complex. Due to its wide application in cancer treatment their efficient extraction from different plant sources has been tried by several scientists using various methods of extraction technique including ultrasound assisted extraction (UAE). In line with this Abdel-dayema et al. (2021) extracted sesquiterpene lactones (Damsin and neoambrosin) by UAE from Ambrosia maritima where they found that the ethanol strength (55%) was the most decisive factor affecting the yield of damsin. Trendafilova et al. (2010) reported their findings on the ultrasonic assisted extraction of sesquiterpene lactones (alantolactone and isoalantolactone) from I. helenium roots, where they found that Ultrasonic extraction resulted in increased yields of sesquiterpene lactones at reduced extraction time and temperature, which is a favourable condition for thermally unstable compounds extraction from plant materials. These finding supported our objective of high yield of parthenlide at low extraction temperature and extraction time. The RSM method has been also employed by Cvetanović et al. (2020) to optimize the extraction of tannin from seeds of Tamarindus indica L. by maceration where they found that high methanol concentration, very low extraction temperature and high liquid to solid ratio had good impact on the yield of tannins. Similarly, Lin et al. (2021) also utilized RSM to optimize the various extraction parameters of UAE for the enhanced extraction of total flavonoid from Moringa oleifera L. leaves.

Our results from cytotoxicity evaluations of leaf extracts on human hepatoma cells (the HuH7 cell line) endorsed the findings of RSM and HPTLC analysis. The lowest CC₅₀ value, 125.47 µg/mL, was observed for the extract yielded at 35 °C, over 45 min, and with a LSR of 20 mL/g, which validates the findings of RSM. The RSM results were also validated by comparing the experimental findings of the responses with the expected values. Additionally, the evaluation of percentage prediction errors was done. These analyses were used for validation of produced equation and applicability of the RSM model. Overall, the low amount of errors and significant R² values reported in the present study demonstrate the high prediction ability of the RSM.

5 Conclusion

In general, the productivity of active ingredients from natural source is very low and sometimes it is not enough even to complete the study of required parameters. The various modalities were adopted in past also for better yield and sustainable supply of the required phytochemicals. Use of Box–Behnken design for response surface methodology (RSM) is one more stride to enhance the yield of targeted compound by optimizing various factors affecting the yield. The findings of this experiment suggested the optimized temperature (38.8 °C), time required (50-min) and the ratio of liquid/solid (20.4 mL/g) to maximize the yield of parthenolide (1.010% ± 0.04 %w/w) using RSM. The values of optimized parameters were further verified and validated by quantification analyses through HPTLC and cytotoxicity studies. The proposed RSM-optimized extraction and HPTLC analysis methods could be employed for the effective extraction of parthenolide in other species of Tarchonanthus as well as other parthenolide-possessing genera. Similar to many other sesquiterpene lactones the parthenolide also possesses significant cytotoxic potential as evident from previous studies. In this study the TCEE showed significant cytotoxic potential (125.47 μg/mL) against the human hepatoma cells (HuH7) reported by the authors for the first time.

Acknowledgement

The authors extend their appreciation to the Deanship of Scientific Research at King Saud University, Saudi Arabia for funding the work through the research group project number RG-1442-073. Authors are also grateful to the Researchers Support Services Unit, King Saud University for the support and cooperation in this research.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References

Abdel-dayema S.I.A., Khalil M.N.A., Abdelrahman E.H., El-Goharya H.M., Yehiac A., . Ultrasound-Assisted Extraction of Damsin and Neoambrosin from Ambrosia maritima: Optimization Using Response Surface Methodology. Egypt. J. Chem.. 2021;64(3):1433-1439.
[Google Scholar]
Al-Fatlawi A.A., Irshad M., Rahisuddin A., . Effect of parthenolide on growth and apoptosis regulatory genes of human cancer cell lines. Pharm. Biol.. 2015;53:104-109.
[Google Scholar]
Alwaseem H., Frisch B.J., Fasan R., . Anticancer activity profiling of parthenolide analogs generated via P450-mediated chemoenzymatic synthesis. Bioorg. Med. Chem.. 2018;26:1365-1373.
[Google Scholar]
Awang D.V.C., Dawson B.A., Kindack D.G., Crompton C.W., Heptinstall S., . Parthenolide content of feverfew (Tanacetum parthenium) assessed by HPLC and ¹H-NMR spectroscopy. J. Nat. Prod.. 1991;54(6):1516-1521.
[Google Scholar]
Berry M.I., . Feverfew faces the future. Pharm. J.. 1984;232:611-614.
[Google Scholar]
Bloszyk E., Geppert B., Drozdz B., . Quantitative determination of sesquiterpene lactones in plant material by infrared spectroscopy. Planta Med.. 1978;34:79-86.
[Google Scholar]
Bosio C., Tomasoni G., Martínez R., Olea A.F., Carrasco H., Villena J., . Cytotoxic and apoptotic effects of leptocarpin, a plant-derived sesquiterpene lactone, on human cancer cell lines. Chem. Biol. Interact.. 2015;242:415-421.
[Google Scholar]
Brown A.M.G., Lowe K.C., Davey M.R., Power J.B., Knight D.W., Heptinstall S., . Comparison of extraction procedures for parthenolide in Tanacetum parthenium. Phytocem. Anal.. 1996;7(2):86-91.
[Google Scholar]
Cvetanović A., Uysal S., Pavlić B., . Tamarindus indica L. Seed: Optimization of Maceration Extraction Recovery of Tannins. Food Anal. Methods. 2020;13:579-590.
[Google Scholar]
Fontanel D., Bizot S., Beaufils P., . Dosage by HPLC of parthenolide content in the great Chamomille Tanacetum parthenium (L) Schulz-Bip. PIant Med. Phytother.. 1990;24:231-237.
[Google Scholar]
French Pharmacopoeia, Maisonneuve, Paris, 1987, Grand Camomille 87/l.
George V.C., Kumar D.R., Kumar R.A., . Relative in vitro potentials of parthenolide to induce apoptosis and cell cycle arrest in skin cancer cells. Curr. Drug. Discov. Technol.. 2016;13:34-40.
[Google Scholar]
Groenewegen A., Heptinstall S., . A comparison of the effects of an extract of feverfew and parthenolide, a component of feverfew, on human platelet activity in-vitro. J. Pharm. Pharmacol.. 1990;42:553-557.
[Google Scholar]
Heptinstall S., Awang D.V.C., Dawson B.A., Kindack D., Knight D.W., May J., . Parthenolide content and bioactivity of feverfew (Tanacetum-parthenium (L) Schultz-Bip.). Estimation of commercial and authenticated feverfew products. J. Pharm. Pharmacol.. 1992;44:391.
[Google Scholar]
Hromadkova Z., Ebringerova A., . Ultrasonic extraction of plant materials- investigation of hemicelluloses release from buckwheat hulls. Ultrason. Sonochem.. 2003;10:127-133.
[Google Scholar]
International Conference on Harmonization (ICH) of Technical Requirements for Registration of Pharmaceuticals for Human use, Harmonised Triplicate Guideline on Validation of Analytical Procedures: Text and Methodology Q2 (R1), Complementary Guideline on Methodology incorporated in November 2005 by the ICH Steering Committee, IFPMA, Geneva.
Johnson E.S., Kadam N.P., Hylands D.M., Hylands P.J., . Efficacy of feverfew as prophylactic treatment of migraine. Br. Med. J.. 1985;291:569-573.
[Google Scholar]
Lin M., Bi H., Yan Y., Huang W., Zhang G., Tang S., Liu Y., Zhang L., Ma J., Zhang J., . Parthenolide suppresses non-small cell lung cancer GLC-82 cells growth via B-Raf/MAPK/Erk pathway. Oncotarget. 2017;8:23436-23447.
[Google Scholar]
Lin X., Wu L., Wang X., Yao L., Wang Lu., . Ultrasonic-assisted extraction for flavonoid compounds content and antioxidant activities of India Moringa oleifera L. leaves: Simultaneous optimization, HPLC characterization and comparison with other methods. J. Appl. Res. Med. Aromatic Plants. 2021;20:100284
[Google Scholar]
Alam P., Siddiqui N.A., Alqahtani A.S., Haque A., Basudan O.A., Alqasoumi S.I., AL-Mishari A.A., Khan M.U., . Response surface methodology-based optimization of ultrasound-assisted extraction of β-sitosterol and lupeol from Astragalus atropilosus (roots) and validation by HPTLC method. Asian Pac. J. Trop. Bio.. 2020;10:281-292.
[Google Scholar]
Retnik L.C., Skeget M., Knez Z., . Separation of parthenolide from feverfew: performance of conventional and high-pressure extraction techniques. Sep Purif Technol. 2005;41:13-20.
[Google Scholar]
Rey J.P., Levesque J., Pousset J.L., . Extraction and high-performance liquid chromatographic methods for the gamma lactones parthenolide (Chrysanthemum parthenium Bernh.), marrubiin (Marrubium vulgare L.) and artemisinin (Artemisia annua L.) J. Chromatogr.. 1992;605:124.
[Google Scholar]
Siddiqui N.A., Al-Yousef H.M., Alhowiriny T.A., Alam P., Hassan W.H.B., Amina M., Hussain A., Abdelaziz S., Abdallah R.H., . Concurrent analysis of bioactive triterpenes oleanolic acid and b-amyrin in antioxidant active fractions of Hibiscus calyphyllus, Hibiscus deflersii and Hibiscus micranthus grown in Saudi Arabia by applying validated HPTLC method. Saudi Pharm. J.. 2018;26:266-273.
[Google Scholar]
Sohma I., Fujiwara Y., Sugita Y., Yoshioka A., Shirakawa M., Moon J.H., Takiguchi S., Miyata H., Yamasaki M., Mori M., Doki Y., . Parthenolide, an NF-κB inhibitor, suppresses tumor growth and enhances response to chemotherapy in gastric cancer. Cancer Genomics Proteomi.. 2011;8(1):39-47.
[Google Scholar]
Trendafilova A., Chanev C., Todorova M., . Ultrasound-assisted extraction of alantolactone and isoalantolactone from Inula helenium roots. Pharmacogn Mag.. 2010;6(23):234-237.
[CrossRef] [Google Scholar]
Tsai T.Y., Chan P., Gong C.L., Wong K.L., Su T.H., Shen P.C., Leung Y.M., Liu Z.M., . Parthenolide-induced cytotoxicity in H9c2 cardiomyoblasts involves oxidative stress. Acta Cardiol. Sin.. 2015;31:33-41.
[Google Scholar]
Wang T., Guo N., Wang S.X., Kou P., Zhao C.J., Fu Y.J., . Ultrasound-negative pressure cavitation extraction of phenolic compounds from blueberry leaves and evaluation of its DPPH radical scavenging activity. Food Bioprod. Process.. 2018;108:69-80.
[Google Scholar]
Xu D.P., Zheng J., Zhou Y., Li Y., Li S., Li H.B., . Ultrasound-assisted extraction of natural antioxidants from the flower of Limonium sinuatum: optimization and comparison with conventional methods. Food Chem.. 2017;217:552-559.
[Google Scholar]
Yang L., Cao Y.L., Jiang J.G., Lin Q.S., Chen J., Zhu L., . Response surface optimization of ultrasound-assisted flavonoids extraction from the flower of Citrus aurantium L. var. amara. Engl. J. Sep. Sci.. 2010;33:1349-1355.
[Google Scholar]
Yoshioka H., Renold W., Fischer N.H., Higo A., Mabry T.J., . Sesquiterpene lactones from Ambrosia connfertiflora (compositae) Phytochem. 1970;9:823.
[Google Scholar]
Zhang H., Cao X., Liu Y., Shang F., . Rapid recovery of high content phytosterols from corn silk. Chem. Cent J.. 2017;11:108.
[Google Scholar]
Zhao Q., Kennedy J.F., Wang X., Yuan X., Zhao B., Peng Y., Huang Y., . Optimization of ultrasonic circulating extraction of polysaccharides from Asparagus officinalis using response surface methodology. Int. J. Biol. Macromol.. 2013;49:181-187.
[Google Scholar]
Zhou J.Z., Kou X., Stevenson D., . Rapid extraction and high performance liquid chromatographic methods for determination of parthenolide in feverfew (Tanacetum parthenium) J. Agric. Food Chem.. 1999;47:1018.
[Google Scholar]

Appendix A

Supplementary material

Supplementary data to this article can be found online at https://doi.org/10.1016/j.arabjc.2021.103194.

Appendix A

Supplementary material

The following are the Supplementary data to this article:

Supplementary Table 1

Download

Supplementary Table 2

Download

Supplementary Table 3

Download

Supplementary Table 4

Download

Fulltext Views
227

PDF downloads
72

View/Download PDF
Download Citations

BibTeX
RIS

Show Sections

[1] Abdel-dayema S.I.A., Khalil M.N.A., Abdelrahman E.H., El-Goharya H.M., Yehiac A., . Ultrasound-Assisted Extraction of Damsin and Neoambrosin from Ambrosia maritima: Optimization Using Response Surface Methodology. Egypt. J. Chem.. 2021;64(3):1433-1439.
[Google Scholar]

[2] Al-Fatlawi A.A., Irshad M., Rahisuddin A., . Effect of parthenolide on growth and apoptosis regulatory genes of human cancer cell lines. Pharm. Biol.. 2015;53:104-109.
[Google Scholar]

[3] Alwaseem H., Frisch B.J., Fasan R., . Anticancer activity profiling of parthenolide analogs generated via P450-mediated chemoenzymatic synthesis. Bioorg. Med. Chem.. 2018;26:1365-1373.
[Google Scholar]

[4] Awang D.V.C., Dawson B.A., Kindack D.G., Crompton C.W., Heptinstall S., . Parthenolide content of feverfew (Tanacetum parthenium) assessed by HPLC and ¹H-NMR spectroscopy. J. Nat. Prod.. 1991;54(6):1516-1521.
[Google Scholar]

[5] Berry M.I., . Feverfew faces the future. Pharm. J.. 1984;232:611-614.
[Google Scholar]

[6] Bloszyk E., Geppert B., Drozdz B., . Quantitative determination of sesquiterpene lactones in plant material by infrared spectroscopy. Planta Med.. 1978;34:79-86.
[Google Scholar]

[7] Bosio C., Tomasoni G., Martínez R., Olea A.F., Carrasco H., Villena J., . Cytotoxic and apoptotic effects of leptocarpin, a plant-derived sesquiterpene lactone, on human cancer cell lines. Chem. Biol. Interact.. 2015;242:415-421.
[Google Scholar]

[8] Brown A.M.G., Lowe K.C., Davey M.R., Power J.B., Knight D.W., Heptinstall S., . Comparison of extraction procedures for parthenolide in Tanacetum parthenium. Phytocem. Anal.. 1996;7(2):86-91.
[Google Scholar]

[9] Cvetanović A., Uysal S., Pavlić B., . Tamarindus indica L. Seed: Optimization of Maceration Extraction Recovery of Tannins. Food Anal. Methods. 2020;13:579-590.
[Google Scholar]

[10] Fontanel D., Bizot S., Beaufils P., . Dosage by HPLC of parthenolide content in the great Chamomille Tanacetum parthenium (L) Schulz-Bip. PIant Med. Phytother.. 1990;24:231-237.
[Google Scholar]

[11] French Pharmacopoeia, Maisonneuve, Paris, 1987, Grand Camomille 87/l.

[12] George V.C., Kumar D.R., Kumar R.A., . Relative in vitro potentials of parthenolide to induce apoptosis and cell cycle arrest in skin cancer cells. Curr. Drug. Discov. Technol.. 2016;13:34-40.
[Google Scholar]

[13] Groenewegen A., Heptinstall S., . A comparison of the effects of an extract of feverfew and parthenolide, a component of feverfew, on human platelet activity in-vitro. J. Pharm. Pharmacol.. 1990;42:553-557.
[Google Scholar]

[14] Heptinstall S., Awang D.V.C., Dawson B.A., Kindack D., Knight D.W., May J., . Parthenolide content and bioactivity of feverfew (Tanacetum-parthenium (L) Schultz-Bip.). Estimation of commercial and authenticated feverfew products. J. Pharm. Pharmacol.. 1992;44:391.
[Google Scholar]

[15] Hromadkova Z., Ebringerova A., . Ultrasonic extraction of plant materials- investigation of hemicelluloses release from buckwheat hulls. Ultrason. Sonochem.. 2003;10:127-133.
[Google Scholar]

[16] International Conference on Harmonization (ICH) of Technical Requirements for Registration of Pharmaceuticals for Human use, Harmonised Triplicate Guideline on Validation of Analytical Procedures: Text and Methodology Q2 (R1), Complementary Guideline on Methodology incorporated in November 2005 by the ICH Steering Committee, IFPMA, Geneva.

[17] Johnson E.S., Kadam N.P., Hylands D.M., Hylands P.J., . Efficacy of feverfew as prophylactic treatment of migraine. Br. Med. J.. 1985;291:569-573.
[Google Scholar]

[18] Lin M., Bi H., Yan Y., Huang W., Zhang G., Tang S., Liu Y., Zhang L., Ma J., Zhang J., . Parthenolide suppresses non-small cell lung cancer GLC-82 cells growth via B-Raf/MAPK/Erk pathway. Oncotarget. 2017;8:23436-23447.
[Google Scholar]

[19] Lin X., Wu L., Wang X., Yao L., Wang Lu., . Ultrasonic-assisted extraction for flavonoid compounds content and antioxidant activities of India Moringa oleifera L. leaves: Simultaneous optimization, HPLC characterization and comparison with other methods. J. Appl. Res. Med. Aromatic Plants. 2021;20:100284
[Google Scholar]

[20] Alam P., Siddiqui N.A., Alqahtani A.S., Haque A., Basudan O.A., Alqasoumi S.I., AL-Mishari A.A., Khan M.U., . Response surface methodology-based optimization of ultrasound-assisted extraction of β-sitosterol and lupeol from Astragalus atropilosus (roots) and validation by HPTLC method. Asian Pac. J. Trop. Bio.. 2020;10:281-292.
[Google Scholar]

[21] Retnik L.C., Skeget M., Knez Z., . Separation of parthenolide from feverfew: performance of conventional and high-pressure extraction techniques. Sep Purif Technol. 2005;41:13-20.
[Google Scholar]

[22] Rey J.P., Levesque J., Pousset J.L., . Extraction and high-performance liquid chromatographic methods for the gamma lactones parthenolide (Chrysanthemum parthenium Bernh.), marrubiin (Marrubium vulgare L.) and artemisinin (Artemisia annua L.) J. Chromatogr.. 1992;605:124.
[Google Scholar]

[23] Siddiqui N.A., Al-Yousef H.M., Alhowiriny T.A., Alam P., Hassan W.H.B., Amina M., Hussain A., Abdelaziz S., Abdallah R.H., . Concurrent analysis of bioactive triterpenes oleanolic acid and b-amyrin in antioxidant active fractions of Hibiscus calyphyllus, Hibiscus deflersii and Hibiscus micranthus grown in Saudi Arabia by applying validated HPTLC method. Saudi Pharm. J.. 2018;26:266-273.
[Google Scholar]

[24] Sohma I., Fujiwara Y., Sugita Y., Yoshioka A., Shirakawa M., Moon J.H., Takiguchi S., Miyata H., Yamasaki M., Mori M., Doki Y., . Parthenolide, an NF-κB inhibitor, suppresses tumor growth and enhances response to chemotherapy in gastric cancer. Cancer Genomics Proteomi.. 2011;8(1):39-47.
[Google Scholar]

[25] Trendafilova A., Chanev C., Todorova M., . Ultrasound-assisted extraction of alantolactone and isoalantolactone from Inula helenium roots. Pharmacogn Mag.. 2010;6(23):234-237.
[CrossRef] [Google Scholar]

[26] Tsai T.Y., Chan P., Gong C.L., Wong K.L., Su T.H., Shen P.C., Leung Y.M., Liu Z.M., . Parthenolide-induced cytotoxicity in H9c2 cardiomyoblasts involves oxidative stress. Acta Cardiol. Sin.. 2015;31:33-41.
[Google Scholar]

[27] Wang T., Guo N., Wang S.X., Kou P., Zhao C.J., Fu Y.J., . Ultrasound-negative pressure cavitation extraction of phenolic compounds from blueberry leaves and evaluation of its DPPH radical scavenging activity. Food Bioprod. Process.. 2018;108:69-80.
[Google Scholar]

[28] Xu D.P., Zheng J., Zhou Y., Li Y., Li S., Li H.B., . Ultrasound-assisted extraction of natural antioxidants from the flower of Limonium sinuatum: optimization and comparison with conventional methods. Food Chem.. 2017;217:552-559.
[Google Scholar]

[29] Yang L., Cao Y.L., Jiang J.G., Lin Q.S., Chen J., Zhu L., . Response surface optimization of ultrasound-assisted flavonoids extraction from the flower of Citrus aurantium L. var. amara. Engl. J. Sep. Sci.. 2010;33:1349-1355.
[Google Scholar]

[30] Yoshioka H., Renold W., Fischer N.H., Higo A., Mabry T.J., . Sesquiterpene lactones from Ambrosia connfertiflora (compositae) Phytochem. 1970;9:823.
[Google Scholar]

[31] Zhang H., Cao X., Liu Y., Shang F., . Rapid recovery of high content phytosterols from corn silk. Chem. Cent J.. 2017;11:108.
[Google Scholar]

[32] Zhao Q., Kennedy J.F., Wang X., Yuan X., Zhao B., Peng Y., Huang Y., . Optimization of ultrasonic circulating extraction of polysaccharides from Asparagus officinalis using response surface methodology. Int. J. Biol. Macromol.. 2013;49:181-187.
[Google Scholar]

[33] Zhou J.Z., Kou X., Stevenson D., . Rapid extraction and high performance liquid chromatographic methods for determination of parthenolide in feverfew (Tanacetum parthenium) J. Agric. Food Chem.. 1999;47:1018.
[Google Scholar]

Optimization of ultrasound-assisted parthenolide extraction from Tarchonanthus camphoratus leaves using response surface methodology: HPTLC and cytotoxicity analysis

Abstract

Keywords

HP-TLC

Response surface methodology

Parthenolide

Optimization

Cytotoxicity

1 Introduction

2 Materials and methods

2.1 Plant material

2.2 Apparatus and reagents

2.3 Effect of single factor tests on ultrasonic extraction of T. camphoratus leaves

2.4 Experimental design of RSM

2.5 RSM model and validity testing

2.6 HPTLC system conditions

2.7 Preparation of standard stock solution

2.8 Method development and validation

2.9 Estimation of parthenolide

2.10 Cell culture and cytotoxicity assay of the BBD-run TCEE samples

3 Results

3.1 Effect of single factor tests on ultrasonic extraction of T. camphoratus leaves

3.1.1 Extraction temperature

3.1.2 Extraction time

3.1.3 Liquid-to-solid ratio (LSR)

3.2 Development and validation of HPTLC method

3.3 RSM model fitting

3.4 Effect of extraction parameters (M1, M2, M3) on R and RSM analysis

3.5 RSM validation

3.6 Optimization and verification of the model for extraction parameters

3.7 Cytotoxicity of BBD-run TCEE samples

4 Discussion

5 Conclusion

Acknowledgement

Declaration of Competing Interest

References

Appendix A

Supplementary material

Appendix A

Supplementary material

Suggested read for related articles:

3.4 Effect of extraction parameters (M₁, M₂, M₃) on R and RSM analysis