The Cardiovascular Diseases Prevention

Published Date: 02 Nov 2017

I declare that:

All work described in this report have been carried out by me and all results discussed herein were first obtained by me, except where I might have given due acknowledgement to others.

All the pros in this report have been written by me in my own words, except where I might have given due acknowledgement to others.

All figures in this report have been devised and produced by myself, except where I might have given due acknowledgement to others.

I understand that if I have not complied with the above declaration, I might have some penalty imposed upon me by board of Examiners, and/or I may be deemed to fail the project assessment.

INTRODUCTION

Most plant foods contain different kinds of compounds including phytoestrogens, which have a biological role of preventing chronic human diseases such as cancer, cardiovascular diseases, osteoporosis, diabetes and relief of post-menopausal symptoms. Soybeans contain different kinds of chemicals, including saponins, protease inhibitors, phytates, all essential amino acids, and isoflavones.1 Daidzein and genistein are the most abundant isoflavones mainly found in soybeans and many soy-containing health supplements have been formulated due to the health benefits of these compounds.

Fig. 1 Different chemical constituents in soybean.1

The phytoestrogens have weak oestrogenic activity, which helps slow down abnormal cell growth. The cancer prevention activity of phytoestrogens is achieved by either prevention of the natural oestrogen’s activity or antagonising the natural oestrogen from attaching to its receptors due to their chemical structures similarity.2

Oestrogen Genistein

Fig. 2 Comparison of oestrogen and genistein at oestrogen receptor binding site.3

1.1 Cancer Prevention

Isoflavones inhibit the activity of 5-α reductase, which catalysis the conversion of testosterone to 5-α dihydrotestosterone, and also inhibit aromatase P450, which catalysis the conversion of testosterone to oestrogen. Soy isoflavones have also been proved to bind to androgen receptor and inhibiting tyrosine kinases, thereby blocking the growth and proliferation of cancer cells.1 Refer to fig. 3 below. These interactions with steroid hormones may prevent sex hormone-causing cancers.

Fig. 3 Cancer cell growth prevention by isoflavone binding to androgen receptor.1

1.2 Cardiovascular Diseases Prevention

The cardiovascular disease and diabetes prevention of isoflavones is by increasing the expression of endothelial nitric oxide synthase (eNOS) when isoflavones bind to a nuclear oestrogen receptor (ER). High eNOS activity inside the cell increases bioavailability of nitric oxide (NO) and reduces oxidative stress. Isoflavones also prevent cardiovascular and diabetes by changing signalling pathway inside the cell after binding to ER on the surfaces of endothelial cells. Changing signalling pathway involves high activity of cAMP/protein kinase A, phosphatidylinositol-3-kinase (PI3-kinase)/protein kinase B (Akt), and extracellular signal-regulated kinase 1 and 2 (ERK 1/2). This results in high eNOS activity and hence increases NO production.4

1.3 Osteoporosis Prevention

Osteoporosis prevention is either by binding to oestrogen receptors (ER-α and β) on osteoclasts and inducing apoptosis or inhibition of tyrosine kinase activity inside the cell after binding to the oestrogen receptor to form receptor-isoflavone complex. This in turn increases the activity of bone-alkaline phosphatase. This is achieved by decreasing the resorption activity of osteoclast factor, such as collagen C-telopeptide, and increasing osteoblast formation markers, such as bone-alkaline phosphatase The effect being reduction of bone remodelling and hence reduction of bone resorption.4

Genistein has also been shown to selectively antagonize the bone resorption effects of parathyroid hormone in osteoblasts by inhibiting parathyroid hormone, which increases the expression of a soluble receptor that activates nuclear factor-xB ligand and promoting bone resorption.

Furthermore, isoflavones activate osteoblast cells to build new bone cells by producing a collagen core around the bone and coating it with an adhesive substance. Calcium then binds to the collagen to form a new bone tissue.1 Refer to figure 4 below.

Fig. 4 Mechanism of action of soy isoflavones in osteoporosis treatment (a) Osteoclast cells are carried in the blood to bone, (b) Osteoclast cells bind to the bone, (c) Osteoclast cells release acid and enzymes that dissolve the bone, (d) Osteoclast cells then disappear (e) Soy isoflavones induce the proliferation and activity of osteoblast cells (f) The osteoblast cells produce collagen and coat it with an adhesive substance, (g) Calcium binds to the collagen to form a new bone tissue.1

1.4 Contra-Ideas and Gaps

However, there are controversies as to whether the health benefits of soybeans are due to isoflavones because while many health benefits are clearly seen in animal experiments, it is not so in human models.5 Soy isoflavones only lower LDL by 3% when 50 g of soy per day is consumed, and that there is limited evidence that soy isoflavones lower LDL in humans. Moreover, several placebo-controlled trials have found no significant improvement in endothelium-mediated vasodilation when post-menopausal women were given up to 80 mg per day of soy isoflavones or up to 60 g per day of soy protein containing isoflavones. Recent randomized controlled, cross-over trial in hypertensive individuals has also shown that consuming soy protein containing 118 mg per day of isoflavones for six months does not improve arterial stiffness. Besides, giving soy protein which provides 120 mg per day isoflavones to post-menopausal women for six months does not prevent endometrial hyperplasia induced by the administration of exogenous oestrogen.

Bone mass density (BMD) loss in recent trials did not significantly differ between post-menopausal women who were given soy protein containing isoflavones and those given milk protein. Again, a recent placebo-controlled trial in post-menopausal women older than sixty years found that taking soy protein of 18 g per day or isoflavones of 105 mg per day, alone or together, did not significantly improve BMD over a one-year period. Another study in Taiwanese women found that women taking 100 mg per day of soy isoflavones for one year showed less bone loss compared to the control, but not in women taking 200 mg per day of isoflavones.6

There may be controversy as to the health benefits of isoflavones in humans, the chemical structural similarity of isoflavones such as daidzein and genistein to oestrogen and several supporting evidences of the significant health benefits of isoflavones with regards to many chronic diseases have incited interest in isoflavones as health supplements. With several soy-containing health supplements on the market, there is the need for quick and accurate extraction and quantification of soy isoflavones in these supplements for quality control analysis. Many HPLC analysis reported in the literature do not validate the analytical method. A validated HPLC method is reported in this analysis.

1.5 Isoflavone Occurrence and Chemistry

Phytoestrogens have been extensively studied. The term phytoestrogen is used to describe a group of plant chemicals that elicit an oestrogen-like biological response. The chemicals found in plant foods shown to elicit both in vitro and in vivo oestrogenic effects are isoflavones, the lignans and the coumestans. Isoflavones are present in nutritionally relevant amounts in foods from soybean, chick peas or garbanzo beans, alfalfa, clover sprouts and other plant foods but at concentrations thousand times less than soy.7 Genistein, daidzein, glycitein, formononetin, biochanin A, their glucosides, malonylglucosides and acetylglucosides are the molecular forms found in foods. Lignans are widespread in human foods, with flaxseed the highest source. Most fruits, vegetables and cereals contain lignans. Coumesterol is the only coumestans in foods and only in sprouts of soybean, clover and alfalfa in significant quantities.

Soybean isoflavones include genistein, daidzein and glycetein. They are always found in their conjugated forms. The conjugated forms are not oestrogenic.8 Soybeans contain nine isoflavones glycosides and three of their aglycones. Below are structures of the common isoflavones found in soybeans.9

Genistein Daidzein

Genistin Daidzin

Fig. 5 Chemical Structures of common isoflavones found in soybeans.9

With the glycosides, the glycosyl group is conjugated through an ester bond at the 7-isoflavone position and sugar normally esterified with acetyl or malonyl-groups at the 6’’-position. The 7-glucosylgenistein is the genistin, 7-glucosyldaidzein known as daidzin and 7-glucosylglyain is glycitin. When the sugar is attached with acetyl or malonyl groups the names are prefixed with 6’’-acetyl or 6’’-malonyl.10

The isoflavones can be divided into four main groups including:11

Aglycones: daidzein, genistein, glycitein, formononetin and biochamin A.

Glycosides: daidzin, genistin and glycitin.

Acetylglycosides: 6’’-0-acetaldaidzin, 6’’-0-acetylgenistin and 6’’-0-acetylglycitin.

Malonylglycosides: 6’’-0-malonyldaidzin, 6’’-0-malonylgenistin and 6’’-0-malonylglycitin.

1.6 Physical and Chemical Properties of Daidzein and Genistein

The aglycones are stable under physiological conditions. The acetyl and malonyl-glucose ester bonds can be broken at high temperatures and high acidic and basic conditions. The glucose-isoflavone ether bonds are stronger but can also break under acidic and high temperature conditions. Aqueous solubility of the aglycones is low and pH dependent due to the acidic nature of the phenolic groups. Conjugation to glucose residues increases solubility, but acetylation or malonylation of the aglycones reduces solubility. The methylated derivatives, biochanin A and formononetin, are less soluble than daidzein and genistein.2

Storage of soybeans and its products at room temperature increases aglycones concentration. Concentration of aglycones can increase by up to three to four times during storage for two years. Conversely, room temperature storage from one year onwards can reduce total isoflavone content, depending on temperature, humidity and light. Storage at high relative humidity and temperature of around 84% and 30oC respectively can yield inter-conversion between aglycones and β-glucoside.

Storage of samples at high temperature and relative humidity after extraction and before analysis can also affect isoflavone profiles. Methanolic extract is stable for up to one week at temperatures lower than 10oC and protected from light.5

1.7 Isoflavone Extraction

Extraction of isoflavones from soybeans has been conventionally performed by using 80% methanol and refluxing. Through several experiments over the years, 60% acetonitrile with or without 3% dimethylsulfoxide (DMSO) at room temperature has proved to be superior to 80% methanol in extraction of isoflavones. Moreover, extraction time of five minutes was comparable to one or two hours. Although the extraction was performed with DMSO and without HCl addition to the extraction matrix, the author thinks there is a strong indication that extraction time of similar methods could be reduced from two hours to five minutes.

Various concentrations of HCl have been used during soy isoflavone extraction process to hydrolyse the glucosides, malonyl glucosides and acetyl glucosides to aglycones before analysis.5 This is reasonable to mimic the action of HCl on isoflavones in human stomach, which hydrolysis the glucones to aglycones before absorption.

The effects of processing on isoflavone content of health supplements have revealed inter-conversion of the different forms due to heat processing and native and microbial enzymatic hydrolysis and loss in soy protein fractionation steps.5

Fig. 6 Hydrolysis of malonyl glucosides, acetyl glucosides and glucosides by HCl to aglycones.5

1.8 Isoflavone Analysis

Among the many analytical methods that are normally used to identify and quantify natural plant products are High Performance Liquid Chromatography (HPLC), Gas Chromatography (GC), Capillary Electrophoresis (CE), Fluorometry Method, Liquid Chromatography-Nuclear Magnetic Response (LC-NMR), Mass Spectroscopy (MS), radioimmunoassay and ELISA methods.12

Phytoestrogen analysis are usually performed with reversed-phase HPLC with a mobile phase of methanol (MeOH) or acetonitrile (ACN) and water and a small amount of acid modifier such as acetic acid (AcOH), trifluoroacidic acid (TFA) and phosphoric acid (H3PO4). Moreover, the structures of daidzein and genistein contain phenolic hydroxyl groups, which are weak acid. This means when the mobile phase is made acidic, daidzein and genistein would predominantly stay as unionised form. This helps in reducing such artefacts as ‘smearing’ - association of analytes with the acidic silanols of the stationary phase and peak tailing. Reducing ‘smearing’ would enhance partitioning of daidzein and genistein into the mobile phase for good separation, resolution and quantification.

HPLC analysis is popular in daidzein and genistein analysis due to its relative simple extraction method and instrumentation, although with poor resolution. UV/Vis detector which is normally used also has poor sensitivity. These limitations make HPLC-UV/Vis best utilised in relatively clean matrices such as soy-containing health supplements.13 Nevertheless, with good sample preparation HPLC-UV/Vis remains a good analytical technique in analysing daidzein and genistein from a variety of matrices.

Gas chromatography coupled with mass spectroscopy (GC-MS) was one of the techniques used to analyse isoflavones. In GC analysis the glycones must be hydrolysed to aglycones before analysis.14 This makes it difficult to analyse the isoflavone profile as they appear in the matrix. In order to measure the full isoflavone profile the glycones must be derivatised, which makes the analysis laborious and additional step introduction. Long steps sample preparation has the potential of sample loss.

In search of a simple method of isoflavone analysis, Wang, G., et. al., 199015 reported an HPLC method with a linear gradient elution and a photodiode detector at ambient temperature. There was baseline separation between daidzein and genistein with retention times between eight and ten minutes. All method validation parameters passed required analytical criteria, but the HPLC analysis time was long at more than twenty minutes.

Within a few years on, a validated isocratic HPLC method was reported by Nguyenle et. al., 199516. A 25 cm C-18 column with a photodiode detector was used. The detection wavelength was 254 nm. The internal standard used was flavone with a mobile phase made up of methanol and buffer (NH4OAc, pH adjusted) at 35:65 (v/v). All validation parameters passed required analytical criteria. The method was simple and could be applied to different matrices, but again, it has long analysis time in the order of sixty minutes.

Soon after that, another isocratic HPLC method to analyse daidzein and genistein in soybean was reported by Hutabarat, L.S. et. al., 1998.17 The method was a reversed-phase employing a 15 cm long C-18 column. The mobile phase was 33% acetonitrile and 67% water-acetic acid (99:1 v/v). The internal standard was also flavone. The method was column-friendly because the mobile phase had no buffer which could crystallise in the column, which in turn would need long column equilibration time. Again, daidzein and genistein separated and resolved, but the analysis time was long at over sixty minutes. Below is a figure showing the basic principles of soy-containing health supplements extraction and analysis.

Fig. 7 Basic extraction and analysis process in soy-containing health supplements.5

Daidzein and genistein from soybean and its products analysis reported in the literature usually uses an HPLC with a 25 cm column length. The longer column length could make the analysis time longer and also introduce band broadening as compared to a 150 mm long column.

1.9 Aim

The aim of this study is to investigate daidzein and genistein in commercially available soy-containing health supplements by using reversed-phase HPLC-UV and isocratic elution.

1.10 Objectives

a) To create linear and reproducible calibration curves within appropriate concentration ranges for daidzein and genistein using HPLC-UV with isocratic elution.

b) To validate the analytical method.

c) To optimise methods for extraction of phytoestrogens from nutraceuticals.

d) To determine the amount of daidzein and genistein in selected commercial soy-containing products.

2. MATERIALS and METHODS

2.1 Chemicals and Instruments

HPLC Water was Milli-Q Ultra Pure water system, HPLC methanol was purchased from VWR BDH Prolab. The HPLC methanol was 100% pure of Lot No. 20837.320, acetic acid of 99.8% purity was purchased from Fisher Chemical, a 42o Filter paper was purchased from Whatman International Ltd., pH meter was from Jenway, analytical weighing balance of type B120S was bought from Sartorious AG Cottingen, Flavone, daidzein and genistein standards were purchased from Sigma-Aldrich Chemicals, HS-A health supplement was purchased from Holland & Barrett at Kingston, London, HS-B and HS-C health supplements were purchased from Amazone.co.uk, HPLC instrument was Waters 2690 Separations Module (Alliance) Autosampler, detector was Waters 2487 Dual Lambda Absorbance Detector, column was Phenomenex 150 x 4.60 mm Kinetex 54 XB C-18 100A, Stuart magnetic stirrer of type CB162 was used, printer was Hewlett Packard Deskjet 895 Cxi, computer software was from Millennium 32.

2.1.1 Products Tested

The products tested were health supplements HS-A, HS-B and HS-C.

2.2 Extraction

Daidzein and genistein extraction was carried out according to the methodology by Wang, G., et. al., 199015 with some modifications. HPLC parameters were that of Hutabarat, L.S., et. al.,199817 with some modifications.

2.2.1 Health supplement A (HS-A)

Extraction of daidzein and genistein from HS-A was performed with three capsules. Using analytical weighing balance, the weight of the powder was recorded as 2.2426 g and the powder poured into a 50 ml beaker. A 6 ml 6 M HCl was added to the powder in the 50 ml beaker and stirred with a glass rod until homogenous. A 35 ml 80% HPLC methanol was pipetted into the beaker where the homogenous mixture was and stirred with a glass rod. The mixture was covered with a thin film and placed on a magnetic stirrer plate for twenty minutes, after which the mixture was filtered under vacuum into a Buchner flask, using a 42o Whatman paper. A 10 ml 80% HPLC methanol was used to wash the residue into the Buchner flask where the first filtrate was. The volume of total filtrate was recorded as 41 ml. The filtrate was filtered again using a syringe and a 0.45 um micro-filter into a 10 ml flask and used for the HPLC analysis.

For the HPLC analysis, a 30 ul of the filtrate (extract) was pipetted into a 10 ml flask and a 0.6 ml 300 ug/ml flavone standard (IS) solution was pipetted and added to it. The mixture was made to volume with 80% HPLC methanol and shaken to mix content and injected.

2.2.2 Health supplement B (HS-B)

The powder of one capsule of HS-B was weighed, using an analytical weighing balance, and the weight was recorded as 0.6240 g. A 4 ml 6 M HCl was added to the weighed powder in a 50 ml beaker and the content stirred with a glass rod until homogenous. A 30 ml 80% HPLC methanol was added to the homogenous content in the 50 ml beaker and stirred with a glass rod to mix the content. The mixture was covered with a thin film and placed on a magnetic stirrer plate for twenty minutes for extraction of daidzein and genistein. The mixture was then filtered under vacuum, using a 42o Whatman paper. The residue was washed with a 10 ml 80% HPLC methanol and filtered into the same Buchner flask where the first filtrate was. The total filtrate was measured, using a measuring cylinder and the volume recorded as 35 ml. The filtrate was filtered again, using a 5 ml syringe and a 0.45 um micro-filter and the filtrate used for the HPLC analysis.

For the HPLC analysis, a 9.4 ml of the HS-B filtrate was pipetted into a 10 ml flask and a 0.6 ml 300 ug/ml flavone standard (IS) solution was pipetted and added to the 9.4 ml filtrate in the 10 ml flask and the content shaken until mixed. The mixture was injected for the HPLC analysis.

2.2.3 Health supplement C (HS-C)

The powder of one capsule of HS-C was weighed, using an analytical weighing balance, and the weight was recorded as 0.7405 g. A 4 ml 6 M HCl was added to the weighed powder in a 50 ml beaker and the content stirred with a glass rod until homogenous. A 30 ml 80% HPLC methanol was added to the homogenous content in the 50 ml beaker and stirred with a glass rod to mix the content. The mixture was covered with a thin film and placed on a magnetic stirrer plate for twenty minutes for daidzein and genistein extraction. The mixture was then filtered under vacuum, using a 42o Whatman paper. The residue was washed with a 10 ml 80% HPLC methanol and filtered into the same Buchner flask where the first filtrated was. The total filtrate was measured, using a measuring cylinder and the volume recorded as 39 ml. The filtrate was filtered again, using a 5 ml syringe and a 0.45 um micro-filter and the filtrate used for the HPLC analysis.

For the HPLC analysis, a 9.4 ml of the HS-B filtrate was pipette into a 10 ml flask and a 0.6 ml 300 ug/ml flavone standard (IS) solution was pipetted and added to the 9.4 ml filtrate in the 10 ml flask and the content shaken until mixed. The mixture was injected for the HPLC analysis. Below is the chemical structure of flavone.

Fig. 8 Chemical structure of flavone (IS).11

The concentration of IS in all HPLC injection samples was 18 ug/ml. All sample extracts were injected in triplicate. All sample extracts were kept in a refrigerator for a maximum of seven days prior to use.

2.3 Reference Standard Solutions Preparation (Stock solutions)

A10 ml reference standard solutions of daidzein, genistein and flavone were prepared in 80% methanol to a concentration of 300 ug/ml each. A 0.003 g of each standard was weighed using an analytical weighing balance and poured into separate10 ml flasks. An 80% HPLC methanol was used to make to volume and the mixture shaken to mix. All reference standard solutions were kept in a refrigerator for a maximum of seven days prior to use.

2.4 Calibration Solutions Preparation

Calibration solutions of daidzein and genistein in the range of 0 to 60 ug/ml were prepared from the 300 ug/ml daidzein, genistein and flavone standard stock solutions. For each calibration solution, the required volumes of daidzein, genistein and flavone standard stock solutions were calculated and pipetted into separate 10 ml flasks, making sure each flask contained 0.6 ml 300 ug/ml flavone standard stock solution. The flasks were made to volume with an 80% HPLC methanol and shaken to mix.

2.5 Mobile Phase Preparation (methanol:water / acetic acid at 55%:45%)

A 99:1 water:acetic acid solution was prepared by measuring a 990 ml purified water in a 1000 ml measuring cylinder and made to volume with glacial acetic acid. The content was poured into a 1000 ml glass container, stopped and shaken to mix. An HPLC grade methanol was poured into the 1000 ml measuring cylinder to measure 550 ml. The acetic acid solution was added to make to the 1000 ml volume to make the mobile phase. The mobile phase was poured into a 1000 ml glass container, stopped and shaken to mix content. Fresh mobile phase was prepared daily.

2.6 Method Validation

Method validation is in accordance with ICH Harmonised Tripartite Guideline, 1996.18

2.6.1 Accuracy

A known concentration of 3 ug/ml solution of daidzein and genistein standards was used for the accuracy test. The average concentration of ten replicate injections calculated was compared to the expected concentrations of daidzein and genistein.

2.6.2 Specificity

Specificity was established for daidzein and genistein in all sample matrices. The chromatograms of all sample matrices were compared to the test mix of daidzein, genistein and flavone standards solution chromatogram to make sure there was no interference in the daidzein and genistein peaks in the sample matrices.

2.6.3 Precision

The standard deviations of daidzein and genistein from the 3 ug/ml daidzein and genistein standards solution of the ten replicate injections used for the accuracy test were calculated.

2.6.4 Sensitivity

The limit of detection (LOD) of daidzein was calculated by multiplying 3.3 by the standard deviation of daidzein of the ten replicate injections used for the accuracy and precision tests. The product was divided by the slope of daidzein calibration curve. The LOD of genistein was also calculated by multiplying 3.3 by standard deviation of genistein of the ten replicate injections used for the accuracy and precision tests. The product was divided by the slope of genistein calibration curve. The limit of quantification (LOQ) of daidzein and genistein were calculated as above, but the standard deviations were multiplied by 10 instead of 3.3.

2.6.5 Linearity Range

The range of concentration between 0 to 60 ug/ml of daidzein and genistein standard solutions was tested to find the relationship between PAR and daidzein and genistein concentration.

2.6.6 Robustness

The chromatographic parameters of a solution containing daidzein, genistein and flavone standards when run with a mobile phase containing 55.5% HPLC methanol was compared to a test mix chromatogram, figure 12 on page 21, containing daidzein, genistein and flavone standards that was run with a mobile phase containing 55% HPLC methanol. Another robustness test run was performed when the ambient temperature was 26.7oC and chromatographic characteristics compared to the test mix chromatogram, figure 12, that was run at 25.2oC ambient temperature.

2.6.7 Recovery Test

An extraction using HS-B as the extraction matrix was performed in the same way as the extraction procedure for HS-B as explained, but before the extraction was performed a 0.0030 g of genistein standard was added to a 0.6094 g HS-B powder. The volume of the spiked HS-B sample extract was 35 ml. HPLC analysis of the spiked HS-B sample extract was performed by pipetting a 0.55 ml of the spiked extract into a 10 ml glass and a 0.6 ml IS was added before making to volume with an 80% HPLC methanol. The mixture was shaken to mix. The solution was injected in triplicate for the HPLC analysis.

The total amount of genistein in each triplicate injection of the spiked extract was calculated from the genistein calibration curve. The total expected genistein amount in the spiked HS-B sample extract was calculated by adding the average genistein amount detected from the un-spiked extract to the 0.0030 g genistein standard used for the spiked extract. The percentage genistein recovery was calculated by comparing the average genistein amount detected in the spiked HS-B sample extract to the total expected genistein amount in the spiked HS-B sample extract, and expressed as a percentage.16

5. DISCUSSION

5.1 Health Supplements Daidzein and Genistein Content

The average amounts of daidzein and genistein per capsule detected for HS-A, HS-B and HS-C were all different from the label claim amounts. Whereas no genistein was detected for HS-A, the average amount of daidzein per capsule detected was over hundred times the label claim amount. The label claim amount per capsule of daidzein and daidzin together is 12.00 mg. The average amount of daidzein per capsule detected was 126.11 mg +/- 0.012. Since 6 M hydrochloric acid (HCl) was used to hydrolyse daidzin to daidzein before extraction, the amount of daidzein detected is equivalent to the amount of daidzein and daidzin together. Although the HCl hydrolyses daidzin to daidzein5 which could account for the high yield of daidzein, it is impossible for such a conversion for the stated label claim amount of daidzin.

The label claim amount of genistein and genistin together per capsule of HS-A is 2.50 mg. The storage conditions of the samples and sample extracts were below 25oC and between 2oC and 8oC in a fridge respectively. These storage conditions are optimal for soy containing health supplements.5 The sample extracts were stored for a maximum of seven days and health supplements were stored for about three months before the analysis and these storage periods are well within the stated storage periods according to Rostagno, M.A., 2009.5 Moreover, the product was well within its expiry date. The product has glycitin as well as daidzin which could be hydrolysed under harsh conditions with the sugar moiety reacting with genistein to form genistin, but both samples and sample extracts had optimal storage conditions to rule out this possibility. Hence, the absence of genistein detected in this analysis could be the actual amount in the HS-A as apposed to the stated label claim amount.

Again, the absence of genistein detected in this analysis could not have been attributed to any of the possible situations mentioned above. Adverse conditions, if any, convert the glycones to aglycones5 and that if this were the case it would be expected for genistin to convert to genistein and would have been detected. Therefore, the absence of genistein in this analysis could only be attributed to the true result of HS-A.

On the contrary, the absence of genistein in this analysis could be due to harsh storage conditions such as high temperature and relative humidity during shipment to the wholesaler or at the wholesalers ware house or shop. It is possible for the sugar moiety of daidzin to react with genistein to form genistin in harsh conditions after the sugar moiety has been hydrolysed from daidzin. This would convert all genistein to genistin and hence the absence of genistein in this analysis. Again, according to the label claim the product contains 7 mg glycitein and glycitin together. The hydrolysis of glycitin during the harsh storage conditions at the wholesalers ware house and the reaction between the sugar moiety and genistein could also have contributed to the absence of genistein detected.

UV detection is generally of poor sensitivity as compared to other spectrophotometry techniques. The LOD and LOQ of genistein in this analysis are 0.08 ug/ml and 0.24 ug/ml respectively. This is poor sensitivity for the detection of any low level of genistein that may be in the HS-A extract analysed. Hence, there may be a low level of genistein in the sample extract analysed but the poor sensitivity of the UV detector in this analysis could not detect it. This may have contributed to the absence of genistein in the sample extract analysed.

Although not part of this analysis, another soy-containing health supplement called HS-D was bought from Tesco, Lee Valley, London and analysed to find the content of daidzein and genistein in it, due to its similar appearance to HS-A (bought from Holland and Barrett, Kingston, London). HS-D also had only daidzein detected in it with no genistein. Moreover, the average amount of daidzein detected was comparable to that of HS-A (refer to appendix 5, figure 29). The label claim of daidzein and genistein in HS-D were not indicated on the container. HS-D contains thirty capsules and costs £4.29, whereas HS-A contains one-hundred and twenty capsules and costs £25.05. The amount of capsules in HS-A is four times the amount of capsules in HS-D, but the price of HS-A is nearly six times that of HS-D. Again, this highlights the difference in cost of similar soy-containing health supplements.

The average amounts per capsule of daidzein and genistein detected in HS-B were also lower than the label claim amounts. The average amounts per capsule detected for daidzein and genistein were 1.27 mg +/- 0.006 and 0.17 mg +/- 0.010 respectively. The label claim per capsule for daidzein and genistein are 22.00 mg and 28.00 mg respectively. The percentage recovery of genistein was 92.4%. Taking the percentage recovery into account, the average amount per capsule of genistein extractable would still be low compared to the label claim amount of genistein. Furthermore, the average amounts per capsule of daidzein and genistein in HS-C detected were low at 1.62 mg +/- 0.020 and 0.29 mg +/- 0.012 respectively. The label claim per capsule for daidzein and genistein are 12.00 mg and 2.50 mg respectively.

The observation of these results could not have come by chance as the storage conditions of the products and extract prior to the analysis were tolerable. The high concentration of HCl (6 M) used would have hydrolysed daidzin and genistin to daidzein and genistein respectively. Again, the ratio of health supplement powder amount to the 80% HPLC methanol solvent used for the extraction of daidzein and genistein is good for maximum extraction. According to Zhu, X., et. al., 201119 the lowest matrix amount to solvent ratio for maximum extraction of daidzein and genistein is 1 g to 20 ml. Hence, the results could be a true account of daidzein and genistein in these health supplements.

Although not the same health supplements, Auwerter, C.C.L., et. al., 201220 have also reported lower amounts of genistein and daidzein in health supplements analysed using HPLC. In their report all daidzein and genistein analysed were significantly lower than the label claim amount.

On the contrary, the low amounts of daidzein and genistein observed in these products could be due to lack of enough time for complete hydrolysis of daidzin and genistin to their respective daidzein and genistein. The extraction time in this analysis was twenty minutes at room temperature. Although extraction time of daidzein and genistein with 80% HPLC methanol could be as little as five minutes at room temperature, different forms of matrices have different extractability.21

All three health supplements analysed have different matrices, containing different excipients. In this respect, the low amounts of daidzein and genistein detected could be attributed to low extractability with 80% HPLC methanol due to the kind of matrices of HS-B and HS-C. Besides, although both health supplements are soybean based, the time of harvest and species of soybean used for the manufacturing of each health supplement could have influenced the low daidzein and genistein extractability.5

5.2 Chromatographic Separation

The chromatographic separation and parameters were initially evaluated with a mobile phase of 60% acetonitrile. There was good base line separation of daidzein and genistein with good resolution. The retention time was also short with all separations completed within four minutes, but the peaks shape was distorted and hence other organic mobile phase modifiers were employed. Upon several trials methanol:water/acetic acid (99:1) at 55%:45% was found to have symmetric peak shape with good resolution and relatively short run time of eleven minutes. The pH of mobile phase was 3.3, which might have contributed to daidzein and genistein symmetric peaks due to no association of daidzein and genistein with the acidic silanol groups of the stationary phase. The column used was a C-18 of 15 m long. The mobile phase flow rate was 1.5 ml/minute. Separation was at ambient temperature. With these parameters daidzein eluted at 2.3 minutes (k = 1.38), genistein at 3.3 minutes (k = 2.36). This is good HPLC analysis for quick and accurate quantification of the soy-containing health supplements.

The number of theoretical plates (N) for daidzein was 2268 and that for genistein was 2543. This shows good separation efficiency that is well over 1500 limit quoted in literature. The selectivity of the method between daidzein and genistein was 1.7. Resolution value of 2.5 between daizein and genistein was achieved. The tailing factor of 1 for daidzein and genistein each in this analysis shows good symmetric peaks achievable with this method. The IS employed normalised any inconsistent injection volumes when the PARs were calculated. All values of system suitability pass the limits quoted in literature.22

Most of the methods used for the analysis of isoflavones use 25 cm C-18 columns with gradient elution which have long analysis time of the order sixty minutes and above.23 In the present analysis, a 15 cm column length was used, employing isocratic elution with shorter analysis time of less than eleven minutes

5.3 Extraction

Hydrolysis of daidzin and genistin to daidzein and genistein respectively by 6 M HCl solution with 80% HPLC methanol has been effective in daidzein and genistein hydrolyses and extraction. A 6 M HCl was used to hydrolyse as much daidzin as possible, if any, for detection of daidzein. A twenty minutes extraction time was deemed enough as opposed to longer extraction times reported in the literature.24

5.4 Flavone (IS)

The choice of IS is important in chromatography analysis. The compound chosen should not interfere with the analytes of interest in terms of alteration of retention time or chemical reaction with the analytes of interest. Moreover, the retention time of the IS should not be too long to contribute to long analysis time. Flavone has similar chemical and physical characteristics as daidzein and genistein and would not interfere with these analytes. Again, flavone has a short retention time when optimised reversed-phase HPLC is used. Hutabarat, L.S., et. al., 199817 successfully used flavone as the internal standard for isoflavone analysis. These characteristics influenced the choice of flavone as the internal standard in this analysis.

5.5 Detection Wavelength

UV/Vis detection wavelength should be chosen to detect the analytes of interest only, if possible, and for maximum absorption. The highest absorption UV/Vis wavelength of daidzein and genistein is 260 nm.24 Therefore, 260 nm wavelength would have been ideal for daidzein and genistein detection, but it was observed (data not shown) that other impurities were also absorbed during method development. A UV/Vis detection at 254 nm wavelength has been used to analyse isoflavones by several scientists, including Hutabarat, L.S., et. al., 1998.17 A 254 nm wavelength has high UV/Vis absorbance for daidzein and genistein with less interference and hence, was decided to be used in this analysis.

5.6 Injection Volume

High injection volume of samples dissolved in acetonitrile could affect daidzein and genistein peak shape.25 This problem is not seen with samples dissolved in methanol. Injection volume of 20 ul was used for the present analysis due to inconsistent injection volume of the HPLC system autosampler. An injection volume of 20 ul is high so that on occasions where low volume is injected, there would still be enough sample volume for UV/Vis absorbance for detection. This would improve peak shape.

5.7 Column

C-18 column is the popular choice of column that is used for isoflavone analysis. The length of column employed is usually 25 m with gradient elution. Longer columns have longer analysis time. Longer analysis time can introduce band broadening in later eluting peaks. This influenced the choice of 15 cm long column for this analysis.

5.8 Flow Rate

Manipulation of flow rate (u) is limited because of Van Deemter equation,

H = A + B/u + Cu,

Where, H =theoretical height between plates, A = Eddy diffusion, B = longitudinal diffusion, C = Resistance to mass flow

Refer to Fig. 28 for a typical plot of Van Deemter.

Fig. 28 A Van Deemter plot showing the association of flow rate, eddy diffusion, longitudinal diffusion and resistance to mass flow.26

Flow rate has no influence on Eddy diffusion because analyte molecules will take different paths through the stationary phase at random and flow rate cannot determine which path analytes should take, but flow rate is critical for longitudinal diffusion and resistance to mass flow in terms of band broadening. Increasing the flow rate decreases longitudinal diffusion, but increases resistance to mass flow. Conversely, decreasing the flow rate can increase longitudinal diffusion, but decrease resistance to mass flow.

Hence, a flow rate of 1.5 ml / minute was deemed optimum for the method developed as daidzein and genistein peaks were sharp with high number of theoretical plates and short plates height. Number of theoretical plates for daidzein and genistein are 2268 and 2543 respectively. Plate heights for daidzein and genistein are 0.066 mm and 0.059 mm respectively.

5.9 Validation

5.9.1 Accuracy

The HPLC method employed has good accuracy. Accuracy for daidzein and genistein are 2.9 ug/ml and 2.7 ug/ml respectively, compared to the daidzein and genistein standard solution of 3 ug/ml each that was used.

5.9.2 Specificity

Purity of the peaks of daidzein and genistein in sample chromatograms as compared to peaks of daidzein and genistein in the test mix was established for the method developed. The method showed no interference in daidzein and genistein peaks in sample chromatograms, and hence can reliably detect daidzein and genistein from all other impurities.

5.9.3 Precision

The intra-day precision of the method is good with daidzein and genistein values of 0.045 (SD = +/- 0.002) and 0.155 (SD = +/- 0.001) respectively. The obtained SDs compare well with that of Murphy, P.A., et. al 1999.7

5.9.4 Sensitivity.

The method has LOD for daidzein and genistein as 0.31 ug/ml and 0.08 ug/ml respectively. The LOQ for daidzein and genistein are 0.94 ug/ml and 0.24 ug/ml respectively. The LOD for daidzein is high in comparison to 0.01 ug/ml but that of genistein is comparable to 0.03 as reported by Costa Cesar, I., et. al., 2006,27 The LOQ of daidzein and genistein reported by Costa Cesar, I., et. al., 200627 is 0.04 ug/ml and 0.10 respectively. Again, genistein is comparable to the literature reported value, but daidzein is high. The high LOD and LOQ obtained in this analysis could be due to inadequate system equilibration before sample analysis.

5.9.5 Linearity

A linear relationship was achieved for PAR against daidzein and genistein concentration between 0 and 60 ug/ml with coefficient of linear regression (R2) for daidzein as 0.9999 and genistein as 0.8877. The high concentration range would make it useful in analysis of samples with broad range of daidzein and genistein content.

5.9.6 Robustness

The chromatographic characteristics have no difference between the analytical conditions established for the analysis and for the experiments where variations were introduced in the mobile phase and ambient temperature. Hence, the method has shown to be robust for methanol concentration in the mobile phase in the range 55% to 55.5% and ambient temperatures between 25.2oC and 26.7oC.

5.9.7 Percentage Recovery

Genistein percentage recovery of the extraction procedure is 92.4%, using HS-B as the extraction matrix. This is comparable to that of Nguyenle, T. et. al., 1995,16 who reported genistein percentage recovery of 101% with soybean as the extraction matrix. The authors reported that the high recovery of genistein could be due to impurities co-eluting with genistein peak. However, in this analysis good sample storage conditions and the seven day storage time would generate no impurity to account for the recovery obtained. Moreover, the recovery obtained in this analysis is not over hundred percent for possible impurity contribution. Hence, the percentage recovery of 92.4% in this analysis could be true value.

The percentage recovery of daidzein was not evaluated due to shortage of daidzein standard and time. It is estimated that percentage recovery of daidzein would have been comparable to genistein.