High-Performance Liquid Chromatography (HPLC) Method Validation for Simultaneous Quantitation of Five Phytoestrogenic Flavonoids

doi:10.21203/rs.3.rs-2542015/v1

A Reversed Phase High Performance Liquid Chromatography (RP-HPLC) method for simultaneous quantitation of four isoflavone standards daidzein, genistein, formononetin and biochanin A, and one flavone standard quercetin was developed and validated through the evaluation of linearity, accuracy, precision, specificity, limit of detection and limit of quantitation in accordance with the ICH guidelines. The analysis was performed in a C18 column (150 x 4.6mm, 5µm) with an optimised gradient elution using acetonitrile-water (0.1% trifluoroacetic acid) at a flow rate of 1.0mL min-1 and sample injection volume of 10uL.The retention times of the standards in the order daidzein, quercetin, genistein, formononetin and biochanin A were 4.42, 5.24, 7.85, 10.06 and 13.55 minutes, respectively with tailing factors ranging from 1.09 – 1.12 and minimum resolution value of 3.74. Detection limits range from 0.339 to 0.964 ug/mL and quantitation limits range from 1.027 to 2.922 ug/mL with good linearity (R2 ≥ 0.9967) with a linear range of 1.25 - 20ug/mL for all standards. The method was also found to be accurate and precise based on percentage recovery ranging from 96.96% to 106.87% (intraday, n=3) and relative standard deviation of %RSD≤1.45% (intra-day, n=3) and %RSD≤2.35% (inter-day, n=5). The specificity of the method was evaluated based on the positivity of the minimum peak purity index during the quantitation of the target compounds from the spiked hydrolysed and unhydrolyzed extract of Cajanus cajan ICPL 7035.

High Performance Liquid Chromatography (HPLC)

Solid Phase Extraction (SPE)

Phytoestrogens

Isoflavonoids

Flavonoids

Estrogen-like compounds derived from plants, hence the term phytoestrogens, have been a subject of interest due to their capability to mimic the binding of estrogen to human estrogen receptors ERɑ and ERβ. Isoflavones have a distinct structure of 3-phenylchroman skeleton hydroxylated in 4’- and 7’- positions and may vary depending on the functional group substitution in carbons 5 and 6 [1]. Aside from isoflavones, another major group of phytoestrogens known as flavonoids have 2-phenylchroman structure e.g. quercetin were also detected from soybeans, vegetables, and fruits[1, 2]. Isoflavonoids often present in the form of their respective O-glycosides are detected in flowering species while being almost exclusively produced by the members of the Fabaceae plant family. The biosynthesis of flavonoids and isoflavonoids are proposed to originate from a chalcone precursor, a product of condensation of 4-coumaroyl CoA and three molecules of malonyl CoA by the catalysis of the enzyme chalcone synthase. Another enzyme, chalcone isomerase, catalyses the cyclization of chalcone into flavanone. The pathway then branches to produce either flavonoids or isoflavonoids in the presence of either flavone synthase I (FSI) or isoflavone synthase (IFS), respectively [3]. The Fabaceae family is considered the second most important among agricultural crops as food source, livestock feeds and as raw materials for industrial and agricultural products, such in the case of soy sauce and tofu derived from soy (Glycine max). Legumes are responsible for capturing 17 million metric tons of atmospheric nitrogen in symbiosis with nitrogen fixing bacteria, and the plants’ production of isoflavonoids help modulate the soil microbial communities [4].

It has been reported that both isoflavonoids and flavonoids exhibit a number of bioactivities including antioxidant, anti-inflammatory, anti-viral, anti-estrogenic and anti-cancer properties [5]. Daidzein and genistein are already used as substitutes for estrogen in hormone replacement therapy (HRT) to alleviate post-menopausal symptoms and estrogen-dependent diseases such as endometriosis, as estrogen has been reported to increase the risk of breast and endometrial cancer[2]. Daidzein and genistein’s precursors, formononetin and biochanin A respectively [6] possess lipid metabolism modulatory and neuroprotective effects [7, 8] Both isoflavones and flavonoids are receiving attention in research as health-promoting “nutraceuticals” [9]. Figure I shows the structures of flavones, isoflavones and 17-ꞵ-estradiol, while figure II shows the structure of daidzein, genistein, formononetin, biochanin A and quercetin.

For years, soybean has been the main dietary source of isoflavones daidzein, genistein and glycitein, with genistein being the most abundant and glycitein being the least [3]. Quantitative analysis of these isoflavones were focused on soy and its commercially available products using gas chromatography in tandem with mass spectrometry (GC/MS). In correlation with the diet, phytoestrogen levels in biological fluids were also measured by GC/MS. However, the availability of such expensive equipment renders simple screening assays for phytoestrogen content costly and requires profound analytical experience from the researcher. GC/MS methods are also time-consuming which requires multiple sample preparation steps such as derivatization to increase the volatility of non-volatile compounds particularly isoflavones [10, 11].

Reversed Phase High Performance Liquid Chromatography (RP-HPLC) coupled with photodiode array detectors (PDA) offers more cost-effective analysis for screening of plant extracts for phytoestrogen content. HPLC assays were developed mainly for the quantification of soy phytoestrogens and their glycosides. Soy (Glycine max) is regarded as the primary source for these dietary isoflavones. Varying mobile phase compositions were reported in either isocratic or gradient flow. Among all developed HPLC assays, acidified water using acetic acid [12, 13], trifluoroacetic acid (TFA)and formic acid [3] were used as polar components, while organic solvents acetonitrile and methanol are the most common choice.

Aside from soy, other species in the Fabaceae family such as Red Clover (Trifolium pratense), were reported to contain low amounts of daidzein and genistein while having higher amounts of formononetin and biochanin A [14]. Plant parts aside from seeds, pods and fruit were also reported to have considerable amounts of isoflavones in which some parts cannot be consumed and are discarded. Pigeonpea (Cajanus cajan), a local bean crop in SouthEast Asia, was reported to have biochanin A from leaves and roots [15]. Another study has determined the coumestrol, daidzein, genistein, formononetin and biochanin A content from legumes [12].

The type of isoflavone may vary depending on the target compound or analyte of the study. Some studies may limit the quantitation to the total aglycone forms, others have tried to quantify both isoflavone aglycones and glycosides. Cho et al[3] have tried to optimize the optimum pH for the extraction of isoflavones from soybeans. They have reported that proper solvent pH adjustment must be observed to maximize the extraction of targeted forms of isoflavones. Isoflavone algycones were also reported to be less stable at alkaline pH than at acidic or neutral pH. Chiang et al[16] have reported specific hydrolysis conditions at which the highest total soy isoflavones were measured: HCl concentration (HC) = 3.42N, hydrolysis time (HT) = 205.5 minutes and reaction temperature (RT) = 44.6°C, which served as a basis for the hydrolysis reaction of the current study.

Direct injection of plant extracts to the HPLC is possible, although the matrix of the sample may contain co-extractants which may affect the signal of the target isoflavones and may diminish the effectiveness of the column. Studies have incorporated pre-HPLC on-line SPE sample clean-up procedures [1, 17] neither of which was applied for plant extracts. On-line SPE clean-up procedures also require the use of a trap column, which is less commonly available than conventional SPE cartridges. Rostagno et al. tested for the recovery of isoflavones through several solid phase extraction (SPE) cartridges prior to the HPLC analysis, reporting that Divinyl-benzene-co-N-vinylpyrrolidone SPE cartridge offers the highest recovery. Divinylbenzene based polymer sorbents can significantly reduce the number of proteins and salts which are commonly found as co-extractants from biological samples [18]. Although isoflavones exist in plants as free aglycones, most of them still exist as glycosides. Since acid hydrolysis is applied to free the isoflavones from their glycosides, an additional acid neutralisation step is normally employed before the HPLC analysis as silica-based sorbents are particularly unstable in extreme pH. At low pH (pH < 2), silyl ether linkages can be broken while silica can be dissolved in aqueous solutions at higher pH (pH > 7.5). Polymer-based sorbents SPE cartridges are reliably stable in a wide range of pH (1 to 14), bypassing the acid neutralisation step from the sample through a simple washing step in the solid phase extraction clean-up [19].

Herein, an RP-HPLC method was developed and optimised for quantitating four isoflavones daidzein, genistein, formononetin and biochanin A, and the flavone quercetin with a short execution time and a preceding clean-up step using divinylbenzene polymer SPE. The method was validated using the parameters of limits of detection and quantification (LOD and LOQ), linearity, precision, accuracy, and specificity by analysing the peak purity index. The development of fast, efficient, and sensitive method for the quantification of phytoestrogens would aid in the determination of plant products which can be used as natural dietary supplements and also as basis for the assessment of the relationship between the occurrence of estrogen-dependent diseases and the inclusion of phytoestrogen-rich plant products as part of the regular diet. This method will be used to determine the phytoestrogen content of functional foods included in the Filipino Diet in future experiments.

Chemicals and Reagents

Isoflavone standards daidzein, genistein, formononetin and biochanin A were purchased from Sigma-Aldrich, Inc (St. Louis, Missouri, USA). HPLC grade solvents acetonitrile and methanol, and AR grade methanol were purchased from RCI LabScan (Bangkok, Thailand). Trifluoroacetic acid (TFA), hydrochloric acid (HCl), dimethylsulfoxide (DMSO) and flavonoid standard quercetin were provided by the Natural Products and Peptidomics Laboratory, College of Arts and Sciences, University of the Philippines Manila. Divinylbenzene SPE Cartridges were purchased from Waters™ (Ireland)(OASIS HLB PRiME, 6 cc, 200mg). Type 1 Ultrapure water was purified from Waters Simplicity® Water Purification System. All standards are HPLC Grade and were pre-analysed with spectral peak purity for the confirmation of spectral purity (Peak Purity Index ≥ 0.9999).

Preparation of Standard Solutions

Isoflavone standards were each dissolved in DMSO to prepare 4000 ug/mL stock solutions. Stock solutions were subjected to sonication for 10 minutes to assure complete dissolution of the standards. Resulting solutions were stored in 1.5mL polypropylene tubes at -40° Celsius. On the day of the analysis, the solutions were slowly thawed at room temperature and calculated volumes were obtained from the stock solution to prepare solutions for the construction of calibration curve (20ug/mL – 1.25ug/mL), accuracy and precision analysis (4, 8 and 12 ug/mL) and sample spike solution (~ 8ug/mL per standard).

Solid Phase Extraction (SPE)

OASIS PRiME HLB SPE Cartridges were used for the SPE clean-up step. Initially, the applied procedure was described by Rostagno et al. [13], scaled down for a lesser sample quantity. Briefly, the cartridge was conditioned with 6mL methanol (1mL min^− 1) followed by equilibration using 6mL of 1:1 methanol-water (1mL min^− 1). The dissolved extracts were loaded into the SPE (1mL min^− 1), the cartridge was washed with 6mL water (1mL min^− 1) and finally eluting the analytes with 6mL HPLC Grade methanol (1mL min^− 1). The eluate was collected and evaporated to dryness in a speed vacuum concentrator. The resulting dry residue was dissolved in 1mL of the mobile phase and filtered through 0.45um syringeless HPLC Vials. Three concentrations (25, 75 and 150 ug/mL) were used to validate the percentage recovery of the standards through the SPE.

HPLC Analysis

The HPLC method used was based on the optimised parameters from the method validation of this study. The order of elution of the isoflavone standards were determined by analysing separate solutions containing 20 ug/mL of analytes, separated through a 150mm x 4.6mm ID RP-18 Column with a linear gradient elution profile starting with 5% solvent B (acetonitrile) in solvent A (0.5% TFA in water (%v/v)) at 0 minutes, gradually increasing to 95% solvent B over 60 minutes. The resulting chromatogram was compared with a standard solution containing all five analytes. Succeeding analyses were modified to design the optimum linear gradient elution profile shown (Fig. 3) at a constant flow rate of 1mL min^− 1 and injection volume of 10µL. Ultimately, the optimised elution time program starts with 35% Solvent B at 0 minute, increasing gradually to 80% solvent for 13 minutes, a steady increase in solvent B by 5% per minute over 3 minutes, followed by an isocratic flow using 95% solvent B for the next 2 minutes, and lastly a column equilibration step for 7.5 minutes at 35% solvent B concentration for a total runtime of 27.5 minutes per sample. The photodiode array detector was set to read the absorbance at 254nm. All peaks were analysed and integrated using the Shimadzu LabSolutions Software version 5.81. Solvent blanks of 35% Acetonitrile-water (%v/v) and HPLC Grade methanol were used to subtract the interference of the corresponding sample solvent.

Method Validation

Linearity, Limit of Detection and Limit of Quantitation

Using the optimised HPLC method, calibration curves were constructed by plotting the calculated concentration (abscissa) versus the peak area (ordinate) of the standards with concentrations ranging from 20 ug mL^− 1 to 1.25 ug mL^− 1. The linearity was evaluated by the linear equation, coefficient of variation and the Y-intercept. The limit of detection (LOD) and limit of quantitation (LOQ) were calculated with signal-to-noise ratio of 3.3 and 10, respectively.

Accuracy and Precision

Accuracy and precision were determined by quantitative analysis of 3 concentrations for each of the standards. The ICH Guidelines has recommended that accuracy and precision of a method has to be evaluated at 50%, 100% and 150% assay concentrations (4, 8 and 12 ug mL^− 1). Precision was expressed as the standard deviation or the degree of reproducibility of the method under normal operating conditions.

Application of the Method

A variant of Cajanus cajan (pigeon pea, ICPL 7035) was used as the test sample for the validated method. Approximately 1.5 grams of dried powder of C. cajan seeds were extracted with 60% Acetonitrile-water (%v/v) through ultrasonication at room temperature for 60 minutes in three replicates. The extracts were centrifuged for 10 minutes at 5000 rpm, followed by separation from the plant material through pipetting. For the acid hydrolysis reaction, the extract was mixed with 200uL of 3.42N HCl solution and incubated in a water bath with controlled temperature of 80 deg Celsius for 205.5 minutes. Both the hydrolysed and unhydrolyzed extracts were evaporated to dryness using a speed vacuum concentrator at 50 deg. Celsius and the resulting residue was dissolved in 1:1 methanol-water (%v/v). The resulting solution was passed through the SPE and the methanol eluate was evaporated to dryness. The residue was dissolved in 0.5mL of the mobile phase (35% Acetonitrile-water) and filtered through a 0.2µm syringe filter. The resulting solution was spiked with 5uL of spike solution and analysed using the validated method.

Chromatographic Separation

Preliminary analyses were conducted to investigate the retention times, tailing factor and resolution of the five phytoestrogens as shown in Table 1. The optimum elution gradient for the separation of isoflavones were obtained with 0.1% aqueous TFA and acetonitrile. The selection of the composition of mobile phase was based on previous studies which were optimised using either an isocratic elution profile or a gradient elution profile consisting of either aqueous acetic acid or aqueous trifluoroacetic acid (A) and acetonitrile (B) as mobile phase. Chromatographic separation results for Method C were summarised in Table 2.

The simultaneous identification of the peaks was based on both retention times and the similarity of the spectral scan from 190nm – 800nm. Individual solutions containing 20ppm of each standard were analysed using the optimised elution gradient (figure 3) and spectral scans (figure 4) for each standard were recorded.

Table 1. Retention Time, Tailing factor and Resolution of phytoestrogen standards in three different methods A, B and C.

Method	Compound	Rt (minutes)	Tailing Factor	Resolution
A	Daidzein	11.56	1.47
	Quercetin	12.6	1.45	4.13
	Genistein	14.61	1.42	7.8
	Formononetin	16.65	1.45	7.79
	Biochanin A	20.03	1.46	12.61
B	Daidzein	4.22	1.28
	Quercetin	5.66	1.32	2.03
	Genistein	9.24	1.42	6.4
	Formononetin	11.77	1.48	7.22
	Biochanin A	15.48	1.5	12.37
C	Daidzein	4.44	1.15
	Quercetin	5.25	1.15	3.56
	Genistein	7.82	1.09	10.7
	Formononetin	10.08	1.08	10.09
	Biochanin A	13.54	1.08	16.59

Table 2. Retention time, tailing factor and resolution of phytoestrogen standards using the optimized HPLC method (method C)

Compound	Retention Time		Tailing Factor	Resolution
Compound	Mean	%RSD	Tailing Factor	Resolution
Daidzein	4.42 ± 0.04	0.9	1.12
Quercetin	5.24 ± 0.07	1.25	1.11	3.74
Genistein	7.85 ± 0.08	0.97	1.09	11.31
Formononetin	10.06 ± 0.05	0.5	1.09	7.39
Biochanin A	13.55 ± 0.06	0.41	1.09	17.09

Method Validation

Detection Limit, Quantification Limit and Linearity

Calibration curves for isoflavone standards were constructed using serially diluted solutions containing 20ug mL^-1 to 1.25ug mL^-1 standards in five replicates over five consecutive days. Validation parameters were presented in Table 3 as equations of the line, linearity, limit of detection and limit of quantification (LOD and LOQ). The method was found to be sensitive for all standards with good linearity.

Table 3. Method Validation results for the HPLC method. Linearity, Detection and Quantification Limits are shown.

Compound	Equation	Linearity	LoD, ug/mL	LoQ, ug/mL
Daidzein	y = 65814x - 633.2	0.9996	0.339	1.027
Quercetin	y = 36033x - 6314	0.9983	0.693	2.101
Genistein	y = 67521x - 4645	0.9980	0.790	2.395
Formononetin	y = 60880x - 4809	0.9967	0.964	2.922
Biochanin A	y = 64851x - 1440	0.9995	0.394	1.194

Accuracy and Precision

Accuracy and precision of the method was presented in Table 4. Intra-day precision was determined by calculating the relative standard deviation (%RSD) of three replicates (n=3) and inter-day precision was determined by the analysis of three replicates over three consecutive days (n=5). The accuracy was calculated by dividing the mean concentration measured by the calculated concentration multiplied by 100%.

Table 4. Intra-day and Inter-day Accuracy and Precision.

	Intraday, n=3				Interday, n=5
Compound	Concentration	Mean Accuracy	STDev	%RSD	Mean Accuracy	STDev	%RSD
Daidzein	4.2ppm	106.50%	0.82	0.77%	107.30%	1.61	0.77%
	8.4ppm	101.63%	0.2	0.19%	101.78%	0.58	0.57%
	12.6ppm	100.29%	0.58	0.58%	100.4	0.62	0.62%
Quercetin	4.2ppm	101.10%	0.49	0.49%	102.00%	2.40	2.35%
	8.4ppm	99.44%	0.39	0.40%	99.53%	0.36	0.36%
	12.6ppm	98.44%	0.61	0.62%	97.55%	0.52	0.53%
Genistein	4.3ppm	104.40%	1.51	1.45%	104.30%	1.19	1.14%
	8.6ppm	98.63%	0.69	0.70%	98.81%	0.67	0.68%
	12.9ppm	98.36%	0.87	0.90%	96.14%	0.92	0.96%
Formononetin	4.2ppm	102.40%	0.51	0.50%	103.00%	0.88	0.85%
	8.4ppm	98.85%	0.11	0.11%	99.16%	0.61	0.62%
	12.6ppm	98.96%	0.38	0.38%	97.30%	0.54	0.55%
Biochanin A	4.5ppm	107.80%	0.07	0.65%	108.30%	0.99	0.91%
	9ppm	103.26%	0.22	0.22%	103.44%	0.41	0.40%
	13.5ppm	101.68%	0.41	0.40%	101.76%	0.31	0.31%

Specificity

The specificity of the method was evaluated by comparing the UV spectral scan of a spiked sample against the spectral scan of the standards detected at 254nm (Figure V).

Solid Phase Extraction Recovery

The recovery of five standards from the SPE step was determined to assure the complete elution of the analytes during sample clean up. Standard solutions of known concentrations were passed through the SPE and quantitated using the optimised HPLC method. Recoveries were performed in triplicates for three concentrations (25, 75, 150ug/mL), shown in table 5.

Table 5. Solid Phase Extraction Percentage Recovery

Compound	Mean, ug mL^-1	%Recovery	%RSD
Daidzein	24.39 ± 0.02	97.52	0.06
	73.32 ± 0.05	97.77	0.07
	147.52 ± 0.08	98.35	0.05
Quercetin	23.79 ± 0.41	95.18	1.71
	76.08 ± 0.90	101.44	1.18
	143.87 ± 2.46	95.92	1.71
Genistein	24.68 ± 0.02	98.73	0.06
	74.91 ± 0.14	99.88	0.19
	149.69 ± 0.72	99.79	0.48
Formononetin	25.69 ± 0.02	102.77	0.06
	76.15 ± 0.13	101.53	0.17
	153.13 ± 0.56	102.09	0.36
Biochanin A	25.00 ± 0.01	100.01	0.03
	73.36 ± 0.08	97.81	0.11
	149.65 ± 0.09	99.76	0.06

Determination of Phytoestrogen Content of C. cajan seeds extract

Cajanus cajan (var. ICPL 7035) extracts were used for the validation of the method. Unhydrolyzed and acid hydrolysed extracts were analysed for their phytoestrogen content. Table 6 shows the phytoestrogen content of ICPL 7035 reported as mg standard per gram of dry weight.

Table 6. Flavonoid content of C. cajan (ICPL 7035) extracts

	Mean Concentration (ug/g dry weight)
Sample	Daidzein	Quercetin	Genistein	Formononetin	Biochanin A
ICPL 7035 (unhydrolysed)	0.70 ± 0.08	1.31 ± .037	2.44 ± 0.75	0.78 ± 0.16	0.91 ± 0.67
ICPL 7035 (hydrolysed)	4.93 ± 0.26	11.00 ± 1.26	7.06 ± 3.81	2.20 ± 0.71	1.51 ± 0.09

A 40-minute HPLC Method was conducted as a preliminary analysis and to determine the elution order of the isoflavone standards, starting with 20% solvent B linearly increased to 95% at 30-minute mark, followed by 5 minutes of 95% B. Daidzein was first to elute at 11.56 minutes, followed by quercetin, genistein, formononetin and finally biochanin A at 20.03 minutes. The tailing factors obtained for all peaks were below the accepted values in literature (less than 1.5) with good resolution values of greater than 4.13 (method A). To reduce the analysis time, the initial concentration of solvent B was increased to 25% and the mobile phase composition was increased linearly to 95% over 25 minutes (method B). This resulted in the faster elution time of all peaks, with daidzein being detected first at 4.22 minutes and biochanin A detected last at 15.48 minutes. However, a noticeable increase in peak widths of daidzein and quercetin was observed, reducing the resolution of quercetin and genistein to 2.03 and 6.40, respectively while Genistein, formononetin and biochanin A have no significant changes to their resolutions at 6.40, 7.22 and 12.37, respectively. To further reduce the analysis time to less than 30 minutes and improve the resolution, the initial concentration of solvent B was increased to 35%, with a lower increase in B concentration to 70% over 13 minutes, followed by 5% increase per minute over 2 minutes and finally an increase in concentration to 95% at the 15-minute mark (method C). All analysis methods have an isocratic flow of 95% solvent B to ensure the elution of traces of the standards and samples, and conclude with an equilibration step using 35% solvent B in preparation for succeeding analysis. Both methods A and B re-equilibrate the column for 4 minutes. This resulted in slightly higher relative standard deviations on the retention times of the isoflavones during multiple batch analyses. A longer equilibration time of 10 minutes was used on method C to minimise the deviation in retention times in between multiple analyses. The peak symmetry for all standards also improved in method C represented by the tailing factors of 1.09–1.15 for all standards. Tailing factor closer to 1.0 indicates an almost symmetrical peak. High repeatability was also observed on the retention times using Method C, with %RSD of not more than 1.95% for all standards. Therefore, method C was selected as the gradient elution profile for this study. The analysis time of the validated method is generally shorter compared to most of the other published HPLC methods. The separation of daidzein, genistein, formononetin and biochanin A was well-studied as the most common isoflavones found in red clover. Existing methods have analysis times ranging from 15 minutes up to as long as 60 minutes [12, 20–22].

The choice of solvent for the optimum separation of the isoflavones was based on previous runs and from existing literature. Both acetonitrile and methanol have been used in the analysis of flavones and isoflavones combined with an acidified polar solvent such as water [5]. Methanol, a polar-protic solvent, is a common choice due to its low cost and less toxicity than acetonitrile, a polar aprotic solvent. However, acetonitrile has lower viscosity, shorter wavelength UV-cutoff, and stronger dipole moment and elution strength in reversed-phase columns, allowing shorter elution and analysis times [23]. Common acid modifiers for quantitation of isoflavones include acetic acid, perchloric acid, phosphoric acid or formic acid [5], possibly acidifying both solvents in a binary mobile phase. The advantage of using trifluoroacetic acid (TFA) as modifier for chromatography techniques is its volatility, eliminating the need for acid neutralisation step particularly for mass spectrometry analysis. Acidifying both solvents with TFA in a binary mobile phase also eliminated the peak tailing and enhanced the resolution of catechins. Although previous studies suggested that TFA does not improve the resolution of the peaks especially for isoflavone conjugates [24], the effect of either acetic acid or TFA as acid modifiers for the phytoestrogens was not thoroughly investigated in this study. Nevertheless, acceptable tailing factors are obtained for all peaks.

Linear regression analysis confirmed that there are linear relationships between the peak area and concentration for all standards, with r² values at minimum of 0.9967. Calculated LOD and LOQ values proved the sensitivity of the method and was able to detect the target compounds to 0.339 ug mL^− 1 concentration. This method was also highly accurate with minimum intra-day accuracy of 98%, up to a maximum accuracy of 107%. Intra-day precision also suggests that this study is repeatable for 4, 8 and 12 ug/mL concentrations at %RSD < 2.35%.

The SPE Recovery was performed at higher concentrations in an attempt to determine the upper limit of the available SPE cartridge. The recovery with the SPE method used for 25, 50 and 150ug/mL concentrations of the standards were surprisingly high at the range of 95% − 102%, with the highest %RSD of 1.71% for quercetin suggesting high reproducibility even at high concentrations. It has been previously reported that divinylbenzene based polymer, particularly HLB Prime SPE cartridges were able retain soy isoflavones daidzein, genistein and glycitein, together with their acetylated and malonylated forms [13] and was also used for the extraction of steroidal endocrine disrupting compounds (EDCs) from wastewater samples[25] and honey [26]. No literature has reported the utilization of divinylbenzene-co-n-pyrrolidone SPE adsorbent for the extraction of quercetin, formononetin and biochanin A. Another divinylbenzene based adsorbent Isolute ENV + cartridge was used for the recovery studies of two flavonoids quercetin and kaempferol, which proved better selectivity than C18 cartridges. However, the recovery of quercetin was only around 74% and complete elimination of interfering compounds was still unachievable. The present study has shown that all five standards were retained and recovered from the SPE cartridge with recoveries not less than 95%.

C cajan (ICPL 7035), locally known as kadios in the Philippines, has few reported studies on its estrogenic activity. Genistein, quercetin, formononetin and biochanin A were detected in different parts of C. cajan plant, including leaves, pod surface, roots and stems [15]. No literature was found to report the detection of these isoflavones from the seeds of the C. cajan using chromatographic techniques, particularly HPLC. Genistein extracted from C. cajan seeds were found to have estrogenic activity on rat reproductive organs [27] and antibacterial activity when tested against Staphylococcus aureus strains [28]. In our study, methanolic seed extracts of C cajan were analysed with the optimized SPE-HPLC procedure. For the sample preparation step, the SPE method described above was unable to eliminate the peaks prior to the elution of daidzein. These peaks were presumed to be unhydrolysed glycosides of isoflavones from C cajan. The attached glycosidic moiety in the glycosides increases the hydrophilicity of the molecule therefore are eluted earlier in reversed-phase chromatography [29]. Glycosylated forms were also retained by the SPE and therefore, are not eliminated from the sample. Acid hydrolysis resulted in the increase in aglycone content of the extract, particularly that of quercetin which increased eight times that of the unhydrolyzed extract. Quercetin was quantitated before from the leaves of the C. cajan [30] but no studies have reported the presence of such amounts of quercetin in the seeds. The acid hydrolysis reaction may have hydrolysed a quercetin glycoside predominantly present in C. cajan seeds. An increase in concentrations for daidzein, formononetin and biochanin A were also observed from the acid hydrolysed sample, suggesting successful hydrolysis of the isoflavone aglycones.

We have successfully validated the HPLC method for the simultaneous separation of phytoestrogenic isoflavones in the following elution order daidzein, quercetin, genistein, formononetin and biochanin A using the optimised gradient elution program. The separation among the five phytoestrogens were supported by good resolution values and tailing factors. Solid-phase extraction procedure was also proven to be applicable as a clean-up step for plant extracts with high percentage recovery for all five phytoestrogens.

Acknowledgement

This work was funded by the University of the Philippines Emerging Inter-Disciplinary Research (EIDR) Program under Grant Number OVPAA-EIDR-C08-006.

Author Contributions

Methodology, formal analysis, writing original draft and conceptualization: JCDC. Supervision of the study, review and editing of the manuscript: NQ, MN and MV. Sample collection and processing: CMM.

Competing Interests The authors declare no competing interests.

Conflict of Interests None of the authors of this paper does have a financial or personal relationship with other people or organizations that could inappropriately influence or bias the content of the paper.

Mitani, K., Fujioka, M., & Kataoka, H. (2005). Fully automated analysis of estrogens in environmental waters by in-tube solid-phase microextraction coupled with liquid chromatography–tandem mass spectrometry. Journal of Chromatography A, 1081(2), 218–224. doi:10.1016/j.chroma.2005.05.058
Aidoo-Gyamfi, K., Cartledge, T., Shah, K., & Ahmed, S. (2009). Estrone sulfatase and its inhibitors. Anti-Cancer Agents in Medicinal Chemistry, 9(6), 599-612. https://doi.org/10.2174/187152009788679985
Cho, C., Jung, Y. S., Nam, T. G., Rha, C., Ko, M., Jang, D., … Kim, D. (2020). pH‐adjusted solvent extraction and reversed‐phase HPLC quantification of isoflavones from soybean ( Glycine max (L.) Merr.). Journal of Food Science. doi:10.1111/1750-3841.15051
Tapas, A., Sakarkar, D., & Kakde, R. (2008). Flavonoids as nutraceuticals: A review. Tropical Journal of Pharmaceutical Research, 7(3). https://doi.org/10.4314/tjpr.v7i3.14693
Merken, H. M., & Beecher, G. R. (2000). Measurement of Food Flavonoids by High-Performance Liquid Chromatography: A Review. Journal of Agricultural and Food Chemistry, 48(3), 577–599. doi:10.1021/jf990872o
Blicharski, T., & Oniszczuk, A. (2017). Extraction methods for the isolation of Isoflavonoids from plant material. Open Chemistry, 15(1), 34-45. https://doi.org/10.1515/chem-2017-0005
Geng, Y., Chen, S., Yang, Y., Miao, H., Li, X., Li, G., Ma, J., Zhang, T., Ren, T., Li, Y., Li, L., Liu, L., Yang, J., Wang, Z., Zou, L., Liu, K., Li, Y., Yan, S., Cui, X., Sun, X., Yang, B., Zhang, B., Han, X., Wang, C., Chen, B., Yue, ., Liang, W., Ren, J., Jia, J., Gu, J., Li, Z., Zhao, T., Wang, P., Wei, D., Qui, S., Xiang, D., Xu, X., Chen, W., He, M., Yang, L., Wang, H., Chen, T., Hua, R., Wang, X., Xiangsong, W., Gong, W., Wang, G., Li, M., Zhang, W., Shao, R., Wu, W., Liu, Y. (2022). Long-term exposure to genistein inhibits the proliferation of gallbladder cancer by downregulating the MCM complex. Science Bulletin, 67(8), 813-824. doi.org/10.1016/j.scib.2022.01.011.
Guo, M., Lu, H., Qin, J., Qu, S., Wang, W., Guo, Y., Weiyong, L., Song, M., Chen, J., Wang, Y. (2019). Biochanin A Provides Neuroprotection Against Cerebral Ischemia/Reperfusion Injury by Nrf2-Mediated Inhibition of Oxidative Stress and Inflammation Signaling Pathway in Rats. Med Sci Monit, 25: 8975-8983. https://doi.org/10.12659%2FMSM.918665
Dixon, R. A., & Steele, C. L. (1999). Flavonoids and isoflavonoids – a gold mine for metabolic engineering. Trends in Plant Science, 4(10), 394-400. https://doi.org/10.1016/s1360-1385(99)01471-5
Franke, A. A., Custer, L. J., Cerna, C. M., & Narala, K. (1995). Rapid HPLC Analysis of Dietary Phytoestrogens from Legumes and from Human Urine. Experimental Biology and Medicine, 208(1), 18–26. doi:10.3181/00379727-208-43826
Hsu, BY., Inbaraj, BS., Chen, BH. (2010). Analysis of Soy Isoflavones in Foods and Bioligical Fluids: An Overview. Journal of Food and Drug Analysis, 18(3), 141-154.
Franke, A. A., Custer, L. J., Cerna, C. M., & Narala, K. K. (1994). Quantitation of Phytoestrogens in Legumes by HPLC. Journal of Agricultural and Food Chemistry, 42(9), 1905–1913. doi:10.1021/jf00045a015
Rostagno, M. A., Palma, M., & Barroso, C. G. (2005). Solid-phase extraction of soy isoflavones. Journal of Chromatography A, 1076(1-2), 110–117. doi:10.1016/j.chroma.2005.04.045
Lee, S. H., Jung, B. H., Kim, S. Y., & Chung, B. C. (2004). Determination of phytoestrogens in traditional medicinal herbs using gas chromatography–mass spectrometry. The Journal of Nutritional Biochemistry, 15(8), 452–460. doi:10.1016/j.jnutbio.2004.01.007
Nix, A., Paull, C. A., & Colgrave, M. (2015). The flavonoid profile of pigeonpea, Cajanus cajan: a review. SpringerPlus, 4(1). doi:10.1186/s40064-015-0906-x
Chiang, W.-D., Shih, C.-J., & Chu, Y.-H. (2001). Optimization of acid hydrolysis conditions for total isoflavones analysis in soybean hypocotyls by using RSM. Food Chemistry, 72(4), 499–503. doi:10.1016/s0308-8146(00)00253-3
Doerge, D. R., Churchwell, M. I., & Delclos, K. B. (2000). On-line sample preparation using restricted-access media in the analysis of the soy isoflavones, genistein and daidzein, in rat serum using liquid chromatography electrospray mass spectrometry. Rapid Communications in Mass Spectrometry, 14(8), 673–678. doi:10.1002/(sici)1097-0231(20000430)14:8<673::aid-rcm935>3.0.co;2-f
Guermouche, M. H., & Gharbi, A. (2004). Simplified Solid-Phase Extraction Procedure and Liquid Chromatographic Determination of Celecoxib in Rat Serum. Chromatographia, 60(5-6). doi:10.1365/s10337-004-0378-6
Yawney, J., Treacy, S., Hindmarsh, K. W., & Burczynski, F. J. (2002). A General Screening Method for Acidic, Neutral, and Basic Drugs in Whole Blood using the Oasis MCX(R) Column. Journal of Analytical Toxicology, 26(6), 325–332. doi:10.1093/jat/26.6.325
Grippo, A.A., Xie, Y., Rougeau, B.L., Wyatt, W.V. (1999). Analysis of Phytoestrogens by High Performance Liquid Chromatography. Journal of the Arkansas Academy of Science, 53(13).
Krenn, L., Unterrieder, I., & Ruprechter, R. (2002). Quantification of isoflavones in red clover by high-performance liquid chromatography. Journal of Chromatography B, 777(1-2), 123–128. doi:10.1016/s1570-0232(02)00079-x
Wang, S. W. J., Chen, Y., Joseph, T., & Hu, M. (2008). Variable Isoflavone Content of Red Clover Products Affects Intestinal Disposition of Biochanin A, Formononetin, Genistein, and Daidzein. The Journal of Alternative and Complementary Medicine, 14(3), 287–297. doi:10.1089/acm.2007.0617
Jung, Y. S., Rha, C.-S., Baik, M.-Y., Baek, N.-I., & Kim, D.-O. (2020). A brief history and spectroscopic analysis of soy isoflavones. Food Science and Biotechnology, 29(12), 1605–1617. doi:10.1007/s10068-020-00815-6
Gu, L., & Gu, W. (2001). Characterisation of soy isoflavones and screening for novel malonyl glycosides using high-performance liquid chromatography-electrospray ionisation-mass spectrometry. Phytochemical Analysis, 12(6), 377–382. doi:10.1002/pca.603
Qi, J., Mulabagal, V., Liu, L., Wilson, C., & Hayworth, J. S. (2019). A Rapid UHPLC-MS/MS Method for Quantitation of Phytoestrogens and the Distribution of Enterolactone in an Alabama Estuary. Chemosphere, 124472. doi:10.1016/j.chemosphere.2019.12
Rodrigues-Gonzalo, E., Garcia-Gomez, D., Carabias-Martinez, R.(2010) A confirmatory method for the determination of phenolic endocrine disruptors in honey using restricted-access material–liquid chromatography–tandem mass spectrometry. Anal Bioanal Chem 398, 1239–1247 (2010). https://doi.org/10.1007/s00216-010-3621-z
Primiani, CN., Pujiati, Hardani. (2018). Estrogenicity of the isoflavone genistein pigeon pie seeds (Cajanus cajan L. Miss sp.) on reproductive organs in rat. Journal of Physics: Conference Series. Volume 1025 012061 10.1088/1742-6596/1025/1/012061
Balida LAP, Regalado JTA, Teodosio JJR, Dizon KAH, Sun Z, Zhan ZQ, Blancaflor JMD, Sollesta JVN, Villorente ZM, Saludes JP, Dalisay DS. (2022) Antibiotic Isoflavonoids, Anthraquinones, and Pterocarpanoids from Pigeon Pea (Cajanus cajan L.) Seeds against Multidrug-Resistant Staphylococcus aureus. Metabolites. 2022 Mar 23;12(4):279. doi: 10.3390/metabo12040279. PMID: 35448466; PMCID: PMC9030341.
Křen, V. (2008). Glycoside vs. Aglycon: The Role of Glycosidic Residue in Biological Activity. Glycoscience, 2589–2644. doi:10.1007/978-3-540-30429-6_67
Zu, Yg., Fu, Yj., Liu, W. et al. (2006) Simultaneous Determination of Four Flavonoids in Pigeonpea [Cajanus cajan (L.) Millsp.] Leaves Using RP-LC-DAD. Chroma 63, 499–505. https://doi.org/10.1365/s10337-006-0784-z
Miadoková, E. (2009). Isoflavonoids — an overview of their biological activities and potential health benefits. Interdisciplinary Toxicology, 2(4), 211-218. https://doi.org/10.2478/v10102-009-0021-3
Muthian, G., & Bright, J. J. (2004). Quercetin, a flavonoid Phytoestrogen, ameliorates experimental allergic encephalomyelitis by blocking IL-12 signalling through JAK-STAT pathway in T lymphocyte. Journal of Clinical Immunology, 24(5), 542-552. https://doi.org/10.1023/b:joci.0000040925.55682.a5
Peñalvo, J., Nurmi, T., & Adlercreutz, H. (2004). A simplified HPLC method for total isoflavones in soy products. Food Chemistry, 87(2), 297-305. https://doi.org/10.1016/j.foodchem.2003.12.028
Rodríguez-Gonzalo, E., Domínguez-Álvarez, J., García-Gómez, D., García-Jiménez, M.-G., & Carabias-Martínez, R. (2010). Determination of endocrine disruptors in honey by CZE-MS using restricted access materials for matrix cleanup. ELECTROPHORESIS, 31(13), 2279–2288. doi:10.1002/elps.200900715
Slabas, T. (2002). Comprehensive natural products chemistry, volume 1: Polyketides and other secondary metabolites including fatty acids and their derivatives. Phytochemistry, 59(2), 229. https://doi.org/10.1016/s0031-9422(01)00377-6
Visnevschi-Necrasov, T., Barreira, J. C., Cunha, S. C., Pereira, G., Nunes, E., & Oliveira, M. B. (2015). Phylogenetic insights on the isoflavone profile variations in Fabaceae spp.: Assessment through PCA and LDA. Food Research International, 76, 51-57. https://doi.org/10.1016/j.foodres.2014.11.032

No competing interests reported.

High-Performance Liquid Chromatography (HPLC) Method Validation for Simultaneous Quantitation of Five Phytoestrogenic Flavonoids

Status:

Version 1

Abstract

Figures

1 Introduction

2 Materials And Methods

3 Results

4 Discussion

5 Conclusions

Declarations

References

Additional Declarations

Status:

Version 1