A STUDY ON THE FLATUS POTENTIAL OF DIETARY FIBRE FROM SOME LEGUMES

> The in-vitro flatus causing tendancy of soluble and insoluble fibre fractions from four legumes, namely, blackgram, greengram, soybean and wingedbean by Clostridium pevfringens was investigated. The gas production from the soluble fibre fraction of blackgram, soybean and wingedbean was relatively high and wss simihr to that from raffinose, wh&eas gas production from the soluble fibre fractibn of greengram was low. The insoluble fibre fractions of all four legumes were found to have little or no flatus activity. Gas chromatographic analysis of the flat& gases ,remaled H2 and ~0~ to be the predorninent cbmponents of gas production.


Introduction
The legumes are generally implicated in causing flatulence, though the e'xtent o f their involvement is not clearly known. Many i n~e s t i~a t b r s~'~

'
hypothesize that some legume carbohydrates, especially the galactosecontaining oligosaccharides, escape digestion and absorption in the small intestine and are later subjected to micmbid fermentation in the large intestine, resulting in the production of gases.l19 This is supported by the fact 'that the human alimentary canal does not produce the enzyme -galactosidase6 which is essential for the hydrolysis and absorption of these oligosaccharides. However, the micro flora m the human colon can produce this enzyme and can therefore be expected to ferment this carbohydrate.
Though flatulence is generally attributed to short-chain oligosaccharides, there is some evidence that other carbohydrates may also be involved. Rackis et a l l o reported that the flatus principle of soybean is in the water soluble, low molecular weight carbohydrate fractidn and not in the hulls, fat, protein or the water insoluble residue. However Murphy et aL8 have reported that the water insoluble polysaccharides may also be responsible. Fl&ings studied the flatus potential of seven legumes and reported oligosaccharides, glucose and pentosans and not the starch or lignin to be responsible for hydrogen production. El Faki et a!. 4, in their flatulence studies in vivo and in vitro on chick pea, cowpea and horsegram, found not only oligosaccharides, but also starch and hemicellulose to contribute substancially to the flatus condition.
In the present study, the flatulence potential of sol~ble and insoluble dietary fibre fractions ,,of four l e v a , vir black gram(BG), greengram(GG), soybean (soy) and winged bcm (WB), dompared to rafhose and glucose, were established by incubation with anaerobic C l~s t~d i u r n perf~ngens. The amount of gas produced w a s taken as the parameter indicative of the possible flatus producihg ahility.

' 'Sample Preparation
Mature seeds of four legumes viz. blackgram (variety MI I), greengram (varitty Mf,5), soybean (variety PB 1) and winged bean (variety TPT-2) were obtained from the Dry 2-one Agriculture-Research Station, Maha-Uuppallarna They wex gound in a Wiley laboratory m i l l to pass through a 60-mesh sieve and defatted with hexane for 8 h.

Preparation of Soluble and InGoluble Fibre Fractions
The enzymatic method of ~e l l e n d o o r n ,~ as modified by Schweizer and wursch12 was used to isolate t h e soluble and insoluble fibre from the ade'fatted flours. This approach involved stepwise removal of proteins with pepsin and panmatin, and starch with -glucomylase. .The resudue forms the insoluble fibre (ENZ-I). The precipitation from the supernatant with 4 volumes of ethanol gives soluble fibre (ENZ-S).

In Vitro Gas Prqductioh
In vitro gas production on soluble and insoluble fibre hctions of the legu- The fibre &actions and sugars were incorporated into the media at 1% l e k l and autoclaved at 15 psig, 1 2 0 '~ for 15 m i -. Test tubes, each containing 14<ml of media were incubated overnight at 37 C to check for contaminatiog. The media was then mixed with 1 ml inoculum from a 24 h old culture of Clostridium,~erfnngens and the mixture &en drawn into 30 ml -gf=s q r i q p . ThE outlet of the sj&ge. was plugged with sterile serum rubber stoppers and the mixture incubated at 3 7 '~ for 24 h. The gas firoduced was directly measured horn the movement of the syringe barrel. Two controls were run, one without organism and the other without carbohydrates.

Determination of Gas Composition
The gas collected above was andysed for its constituent gases by the use of a Fisher Hamilton Gas Partitioner connected to a Fisher Scientific Series 260 -recorder. A two column standard system was used. The first column was a 6 ft x 114 in ID glass packed with 30% DC-2-ethyl-hexylsebacate 66 60-80 mesh column pak and the second column was a 6% ft x 3/16 in ID glass packed with 40-60 mesh molecular sieve.
The C0, concentration in 0.5 ml of each gas sample was determined using both columns 1 and 2 with helium as the carrier gas at 40 ml/min and attenuation 8x. The hydrogen content in 0.5 ml of gas mixture was analysed using column 2 only with argon as the camer gas at a flow of 30 ml/min at 16x attenuation. Concentration of CO, and Hz in the gas mixture was calculated from peak height measurements, relative to peak heights for pure HZ and CO, .

Replication
The study was repeated thrice so as to obtain more reliable data.

Results and Discussion
The volume and percent composition of the gas groduced are presented in Tables 1 and 2, respectively. Bacteria incubated o n glucose' produced the greatest volume (20 cc) of gas within 4 hours of incubation. In cultures incubated on raffinose, gas production was ~lower,~naching only 10 cc by 24 h. The insoluble fibre fraction of all four legumes produced little or no gas during the 24 h incubation period. In contrast, the soluble fibre f &tions of BG, soy and WB produced a net volume of 8,7 and 6'cc respectively, by 24 h. However, gas production fiom these fractions was low. (3 cc) during the first 8 h, a trend similar to that observed for raffinose. The soluble fib% fraction of green gram behaved in a manner different to other three legumes ; the gas production was only 3 cc even after 24 h of incubation.
Carbon dioxide and Hydrogen were the two predominant constituents of :he gas produced, confirming that a typical flatus producing system was being tested.' O The ratio of Hz :, CO, was < 1 for raffinose (0.96) an'd glucose (0.86), while this ratio was > 1 for soluble fibres of BG, Soy and WB. The percent composition measurements (Table 2)    The results show that the msoluble fibres have little or no flatus activity, probably because they ark not hydratable. These findings are consistent with those of Rackis et al. l o who reported that the water insoluble residue of soy has no flatus activity.
The soluble fibres of BG, Soy and WB were found to have relatively high flatus activity. Although the production of flatus gas from these soluble fibres was lower than that &om glucose, gas formation was comparable to that of raffinose, a well established causative factor of f l a t u~e n c e .~~~ The isolation procedure employed in the present study was specifically designed to yield sugar-free fibre preparatrons. Thus the gas production did not alise k o m the presence of raffinose Gr stachyose in ENZ-S of BG, Soy or WB. The soluble fibre of GG showed little or no flatus activity, confirming the results of Fleming6 The flatus potential of WB has not been previously reported. The present results indicate that the flatus causing ability of the WB soluble fibre is similar to those of soy and BG.
The results of the present study showed that soluble fibre $actions of some legumes play a d e h i t e role in flatus activity. However, further studies will be required to delineate the exact nature of this contribution to flatus production. The ultimate test will be to feed purified sbluble fibre fractions to human subjects and measure the t h e -course production, as described by Wagner et a1. l4