Increased intake of dietary fibre is recommended in the proper control of chronic diseases, e.g. diabetes mellitus, CVD and cancer(1, 2). It is now well established that dietary fibre is not metabolized in the small intestine and reaches the colon where it is fermented to produce SCFA such as acetate, propionate and butyrate, and gases such as carbon dioxide, methane and hydrogen(Reference Bourquin, Titgemeyer, Garleb and Fahey3, Reference Cummings4). SCFA contributes 6·3–8·4 kJ/g dietary fibre (1·5–2·0 kcal/g dietary fibre)(Reference Roberfroid, Kritchevsky and Bonfield5). Propionate has been shown to inhibit the activity of the enzyme 3-hydroxy-3-methylglutaryl-CoA reductase, the limiting enzyme for cholesterol synthesis(Reference Chen, Anderson and Jenkinds6). Butyrate enhances cell differentiation thus preventing tumour formation in the colon(Reference Jenkins, Ghafari, Wolever, Taylor, Barker and Filden7). Dietary fibre can be soluble or insoluble depending upon solubility in water. Water-soluble fibre can form viscous solutions. Increased viscosity in the intestine slows intestinal transit time, delays gastric emptying(Reference Wisker, Feldheim, Pomeranz, Meuser and Pomeranz8, Reference Anderson, Chen and Webster9), and slows glucose and sterol absorption by the intestine(Reference Wood10, Reference Kahlon and Chow11). Viscous soluble fibre can control the release of glucose with time thus lowering postprandial blood glucose and insulin levels, and serum cholesterol(Reference Anderson, Chen and Webster9). Insoluble fibre consists of lignin, cellulose and hemicellulose. It has usually high water-holding capacity and contributes to increased faecal bulk and frequency of defecation(Reference Eastwood, Brydon, Tadesse, Spiller and Kay12, Reference Cummings13). Moreover, all fibres, both soluble and insoluble, can entrap bile acid and prevents its re-absorption in the liver thus inhibiting cholesterol synthesis(Reference Gurr and Asp14). The consumption of root crops and tubers in the Philippines is 19 g/d while consumption of beans, nuts and seeds is 10 g/d(15). The present study will determine the dietary fibre and fermentability characteristics of root crops and legumes.
Material and methods
Test foods
In the present study, six root crops (ube (Dioscorea alata), gabi (Colocasia esculenta), tugi (Dioscorea esculenta), potato (Solanum tuberosum), camote (Ipomoea batatas) and cassava (Manihot esculanta)) and ten legumes (cowpea (Vigna uriguiculata (L.) Walp.), mungbean (Vigna radiate (L.) Wilczek), pole sitao (Vigna uriguiculata subsp. Sesuipedalos L. Verde), chickpeas (Cicer arietinum), green peas (Pisum sativum L.), peanut (Arachis hypogaea L.), pigeon pea (Cajanus cajan), kidney beans (Phaseolus vilgaris L.), lima beans (Phaseolus lunatus) and soyabeans (Glycine soja)) were used as test foods. Food samples were bought in local markets. Root crops were cooked in water while legumes were soaked in water overnight, and boiled to cook the next day. Both root crops and legumes were freeze-dried before analysis.
Analytical methods
The proximate analysis of all test food was determined using Association of Official Analytical Chemists methods, and the test foods were also analysed for total, soluble and insoluble fibre using Association of Official Analytical Chemists methods(16, 17) against a standard wheat flour (NBS Standard Reference Material 1567 A; Gaithersburg, MD, USA).
In vitro fermentation
Dietary fibre isolates from each test food were fermented in vitro using human faecal inoculum(Reference McBurney and Thompson18). Dietary fibre was isolated from test samples using Association of Official Analytical Chemists methods. Fibre isolate (0·5 g) was weighed in a serum bottle. Fermentation media (40 ml of a mixture of 2 litres deionized water, 1 litre 0·5 m-sodium bicarbonate buffer solution, 1 litre macromineral solution, 5 ml 0·1 % resazurin) and reducing solution (2 ml of a mixture of 1·25 g cysteine-HCl+fifty pellets of potassium hydroxide in 100 ml deionized water and 1·25 g sodium sulphide in 100 ml deionized water) were added to the serum bottle and flushed with carbon dioxide until colourless. The bottles were sealed with rubber stoppers and an aluminium seal and stored at 4°C. The next day the bottles were placed in a waterbath for 1–2 h at 37°C. A 10 ml faecal inoculum (1:15 dilution of fresh faeces from a healthy female human volunteer eating an unspecified diet with no intake of antibiotics for a year) was added into each bottle and the mixture was incubated for 24 h at 37°C without mixing. The fermented digest was filtered and read in a high-pressure liquid chromatograph (LC10 Shimadzu, Shimadzu, Tokyo, Japan) to measure SCFA against a volatile acid standard mixture of acetate, propionate and butyrate (Supelco, Philadelphia, PA, USA). No internal standards were used.
Statistical analysis
Differences between test foods and food analyses were determined by ANOVA and Duncan's multiple range test using the Statistical Analysis System program (SAS Institute Inc., Cary, NC, USA).
Results
Table 1 shows the nutritional composition of root crops and legumes. Legumes are the best sources of protein and are significantly greater than root crops (Table 1; P < 0·05). Root crops have significantly lower fat content than legumes (Table 1; P < 0·05). The dietary fibre content of root crops ranged from 4·6 to 13·5 g/100 g while legumes ranged from 20·9 to 46·9 g/100 g, suggesting that both root crops and legumes are good sources of dietary fibre (Table 1). The dietary fibre content of legumes was significantly greater than that of root crops. Soyabean has significantly greater dietary fibre content (46·9 (sem 3·4)g/100 g) among the legumes while gabi has significantly greater dietary fibre content (13·5 (sem 0·1)g/100 g) among the root crops (Table 1; P < 0·05). The ratio of insoluble to soluble fibre in root crops ranged from 1:1 to 2:1 while for legumes 5:1 to 5:19 excluding kidney beans, which was approximately 99 % insoluble fibre. However, the ratio of insoluble to soluble fibre in all test foods was not significantly associated with production of SCFA (Table 1; P < 0·05). Significant amounts of SCFA were produced after fermentation of the fibre isolate of both root crops and legumes, with cowpea, kidney bean and peanut having the greatest production of SCFA for legumes, and tugi and cassava for root crops (Table 1). The best sources (as mmol/g fibre isolate) of acetate among the legumes were pole sitao (5·6 (sem 0·5)) and mungbean (5·3 (sem 0·1)) while among the root crops, tugi (2·5 (sem 0·4)) and cassava (2·4 (sem 0·1)); for propionate, kidney bean (7·2 (sem 1·5)) and pigeon pea (3·3 (sem 0·2); Table 1); and for butyrate, peanut (6·0 (sem 0·2)) and cowpea (5·4 (sem 0·2)) among the legumes, and tugi (0·8 (sem 0·0)) and cassava (0·8 (sem 0·0); Table 1) among the root crops.
(Mean values with their standard errors)
CHO, carbohydrates.
a–l Mean values within a column with unlike superscript letters were significantly different (P < 0·05).
xyz For SCFA mean values within a row with unlike superscript letters were significantly different (P < 0·05).
* Calculated by difference (100 – (ash, moisture, fat, protein)).
† SCFA (mmol/g fibre isolate).
Discussion
Colonic fermentation studies ideally should be done in vivo. However, the difficulty of experimental variables often makes in vivo studies difficult to conduct. Moreover, the rate and extent of fermentation determined from SCFA production cannot be estimated with much accuracy in vivo (Reference McBurney and Thompson18). The in vitro fermentation method using human faecal inoculum has been used to study rate and extent of fermentation(Reference McBurney and Thompson18). The method provided SCFA molar ratios which are similar to those seen in vivo and therefore can be used to model the human colon. It demonstrated that faecal materials collected from man at different times of the year provide a sufficiently uniform source of micro-organisms so that fermentation values such as fibre digestibility and gas or SCFA production can be compared between studies(Reference McBurney and Thompson18). It also showed the differences in fermentability of different fibre sources. For example, pectin was found to be most rapidly fermented followed by psyllium gum, tragacanth gum and cellulose(Reference McBurney and Thompson18). However, in the present study we did not determine the rate and extent of fermentation of the different test foods since it was not one of the objectives of the study. We only determined the SCFA produced after 24 h of fermentation which mimics human fermentation in the colon.
The study showed that root crops and legumes are good sources of dietary fibre. Although both root crops and legumes have significant amounts of dietary fibre, there must be some differences in their physiological effects. Root crops contained low fat and high fibre and can be a good substitute for bread and rice. Flaxseed (28·0 g/100 g dietary fibre) may be considered a legume and has been shown to have some protective and preventive effects on CVD(Reference Kirby, Anderson, Sieling, Rees, Chen, Miller and Kay19) and colon and breast cancer(Reference Jenab and Thompson20–Reference Thompson, Seidl, Orcheson and Richard22). The physiologic effect of legumes may be similar to flaxseed. Legumes maybe a better laxative than root crops because of the high percentage of insoluble fibre. Intakes of legumes may increase faecal bulk and decrease transit time due to excellent water-holding capacity. Legumes may also serve to dilute secondary bile acids in the colon, thereby reducing colonic exposure to these substances. The higher percentage of soluble fibre in root crops compared to legumes may have physiological effects related to glucose and cholesterol metabolism(Reference Jenkins, Ghafari, Wolever, Taylor, Barker and Filden7, Reference Wisker, Feldheim, Pomeranz, Meuser and Pomeranz8). Generally, soluble fibres have a tendency to have long-term effects associated with better glycaemic control and lower insulin requirements due to increased insulin sensitivity or changes in insulin receptors as well as cholesterol-lowering effects. In terms of SCFA production, a good source of butyrate, e.g. tugi and cassava (root crops) and peanut and cowpea (legumes), may have an important role in the prevention for risk of colon cancer, while a good source of propionate, e.g. kidney beans and pigeon pea (legumes), may have an important role in cholesterol-lowering effects.
In conclusion, both root crops and legumes are good sources of dietary fibre and produced SCFA after fibre fermentation, such as acetate, propionate and butyrate. SCFA production after in vitro fermentation can be estimated using human faecal inoculum and can be used to model the human colon.
Acknowledgements
The project was supported by the Food and Nutrition Research Institute, Department of Science and Technology, Philippines. T. P. T. supervised the implementation of the project and evaluated data and reports; A. C. M. analysed dietary fibre and SCFA of root crops, conducted data analysis and wrote the paper; A. S. L. conducted proximate analysis of root crops, and analysed dietary fibre and SCFA of legumes; R. R. and K. D. conducted proximate analysis of five legumes and assisted in the analysis of dietary fibre and SCFA. There are no conflicts of interest. We would also like to extend our appreciation to Mr Zoilo B. Villanueva for freeze-drying the food samples used in the study.
