School of Applied Sciences, RMIT University, Melbourne VIC 3000, Australia
Introduction
Phytoestrogens are a group of isoflavonoid compounds widely distributed in plant foodstuffs. Some phytoestrogens appear to have a variety of beneficial biological activities in preventing cancers and relieving symptoms of menopause by virtue of their resemblance of female hormone estrogen (Brouns, 2002). The most widely studied phytoestrogens are genistein and daidzein and these are widely distributed. The best source is the soybean and soy-based foods reportedly contain high levels. Tofu is considered a rich food source of phytoestrogens which serve the role nutritionally and therapeutically. The recommended level of daily intake of these phytoestrogens falls in the range of 50-200mg/day (Albertazzi, 1998).
Analysis generally involves 10 minutes sonication prior to refluxing the freeze dried sample in a mixture of strong acid and ethanol. The extracts are typically analysed by reverse phase HPLC. The main objectives of this study are to investigate and optimize the extraction and analysed conditions from the developed methodology and set up a fundamental establishment for the further research on various soy products. It is also significant to analyse the isoflavone contents in commercial beancurd (tofu) products available in the Australian market in this study.
Materials and methods
Materials
Standards used were genistein and daidzein obtained from Sigma-Aldrich. Concentrations were verified by spectrophotometry and using published absorption co-efficients (Franke et al, 1994). Soy and soy foods were all commercial products obtained in Melbourne.
Extraction of phytoestrogens
Fresh foods were freeze-dried and ground before analysis. The moisture content was determined by vacuum oven-drying. 1g of ground food or 25mL soymilk were mixed with 40mL 96% ethanol and 10mL 32% HCl. Samples were sonicated for 10 minutes before extraction (Hutabarat et al, 2000).
HPLC analysis
Sample extracts were filtered and chromatographed isocratically using a Shimadzu HPLC system and spectrophotometric detector. 20μL was injected onto a C-18 column with acetonitrile-water (33:67) at flow rate 0.8mL/min for 30 mins. The wavelength used was 280 nm.
Results and discussion
In the preliminary phase of this study, the analysis procedures for isoflavones in soy flour were assessed for their suitability for soy bean curd (Figures 1-3). The results show that sonication did not markedly enhance the recovery of compounds during extraction. The time of reflux applied was important particularly for daidzein where optimization of isoflavones was achieved within two hours refluxing. During HPLC analysis, standard curves were developed for both of the isoflavones and for each of the standards there was good linearity (r2>0.9994) over a wide range of concentrations. The results for the food products showed lesser amount of genistein than of daidzein in the isoflavones extracted from soy bean curd products.
Figure 1. Calibration curve of standard daidzein monitored at 250 nm
Figure 2. Calibration curve of standard genistein monitored at 263 nm
Figure 3. Extraction of soy flour by varying time of reflux
Note: mean and standard deviation are from six separate analyses.
1-daidzein 10min sonication; 2-daidzein no sonication; 3-genistein 10min sonication; 4-genistein no sonication
The levels of the isoflavonoid compounds found in four common commercial brands of tofu indicated high levels of daidzein and low levels of genistein (Figure 4). The data also indicated some variation in the quantity of phytoestrogens between the brands.
Figure 4. A comparisons of the daidzein and genistein contents in four Australian commercial Tofu products
Note: mean and standard deviation are from triplicate analyses
all data are expressed on dry weight basis (mg/100g)
The four Australian tofu products (Table 1) showed similar patterns with higher daidzein levels than genistein. Values from a US database are presented for comparative purposes and show a wide range of both isoflavones. The US data might reflect variability due to the analytical procedures applied in the various studies. Another issue is that of environmental and genotypic influences. The results obtained in the current study fall within the range of data presented in the US tables for both compounds.
Table 1. Comparison of isoflavone contents found for different commercial tofu products with values from USDA database (USDA 2005)
Description |
Isoflavone contents (mg/100g) | ||
Daidzein |
Genistein |
Total | |
Smooth Food Processing |
16.89 ± 0.01 |
3.16 ± 0.07 |
20.05 ± 0.08 |
Yenson’s beancurd |
13.65 ± 0.05 |
2.78 ± 0.19 |
16.43 ± 0.24 |
Evergreen tofu |
16.17 ± 0.03 |
2.55 ± 0.01 |
18.72 ± 0.04 |
TLY tofu |
18.53 ± 0.01 |
4.16 ± 0.01 |
22.69 ± 0.02 |
US database mean |
9.02 ± 2.86 |
13.60 ± 3.61 |
22.62 ± 6.47 |
US database range |
1.15-14.60 |
2.89-18.66 |
4.04-30.80 |
Note: all data are expressed on dry weight basis (mg/100g)
Conclusion
Further work is underway to apply the procedure to a wider range of foods and to samples obtained during soy processing. This will incorporate studies of recoveries and the utilisation of okara which is the primary by-product of soy processing.
Acknowledgements
Thanks to Renuka Mayadunne and Paul Morrison for help and technical support with the HPLC.
References
Albertazzi, P. (2002) Int. Congress Series., 1229:189-193.
Brouns, F. (2002) Food Res. Int., 35:187-193.
Franke, A., Custer, L., Cerna, M., and Narala, K. (1994) J. Agric. Food Chem., 42:1905-1913.
Hutabarat, S., Greenfield, H., and Mullholland, M. (2000) J. Chromatogr. A., 886:55-63.
USDA (2005) downloaded from www.nal.usda.gov/fnic/.