Food Structure

Volume 12 Number 4 Article 9

1993

Microstructure of Idli

Enamuthu Joseph

Shelly G. Crites

Barry G. Swanson

Follow this and additional works at: https://digitalcommons.usu.edu/foodmicrostructure

Part of the Food Science Commons

Recommended Citation Joseph, Enamuthu; Crites, Shelly G.; and Swanson, Barry G. (1993) "Microstructure of Idli," Food Structure: Vol. 12 : No. 4 , Article 9. Available at: https://digitalcommons.usu.edu/foodmicrostructure/vol12/iss4/9

This Article is brought to you for free and open access by the Western Dairy Center at DigitalCommons@USU. It has been accepted for inclusion in Food Structure by an authorized administrator of DigitalCommons@USU. For more information, please contact [email protected]. FOOD STRUCTURE, Vol. 12 (1993), pp. 483-488 1046-705X/93$5.00 + .00 Scanning Mi croscopy International, Chicago (AM F O' Hare), IL 60666 US A MICROSTRUCTURE OF IDLI

1 2 2 Enamuthu Joseph , Shelly G. Crites and Barry G. Swanson

1Depan men t of Family and Consumer Sciences , University of Nebraska at Kea rney, Kea rney , NE 68849 2D epartment of Food Science an d H uman Nutrition , Was hington State Univers ity, Pullman, WA 99 164

Abstract Introduction

The mi crostructure of idli prepared from seeds of Fermented foods prepared from legume and cereal common beans (Phaseolus vulgaris) and rice, and Indian combinations contribute substantial quantities of calo­ black gram seeds ( Vigna mungo) and rice in the ratio 1:2 r ies, proteins and vitamins to the diets of people in was examined. Intact starch granules and protein bodies Southeast Asia, the Near East and parts of Africa (Raja­ were observed in the unfermented id1i batter. Starch lakshmi and Vanaja, 1967; Ramakrishnan , 1979; Reddy granu les di sappeared as fermentation time increased to era/. , 1981; Sathe era/., 1986). Due to favorable or­ 18 hours and 36 hours. The surface of idli prepared ganoleptic qualities, fe rmented foods are becoming com­ f rom 36-hour fermented batter was cohesive, dense and mon in the diets of the Western world (Reddy e1 a/. , sponge- o r foam- like. T he microstructure o f idli pre­ t982a; Steinkraus, 1983; Sathe era/., t986). pared from common beans a nd rice was similar to the ld li is a naturally fe rmented steamed food o ri gi ­ microstructure o f idli prepared from Indian black g ra m na ting in 1100 AD (Ramakri shn an e1 a/., 1976), pre­ and rice. Commo n beans can be successfully sub sti tuted pared from rice and dehullcd bl ack gram cotyledons fo r Indian black g ram in prepa ring comparable a nd ( Rajalaksh mi and Vanaja, 1967; Chavan a nd Kadam, acceptabl e steamed idl i. 1989). ldli is a common breakfast item in South India and to some extent in S ri Lanka (Reddy n a/. , \981; Steinkraus, 1983). The proportion of rice to black gram can vary from 2: I lO 4: I in preparation of idli batter (Desi kachar er a/. , 1960; Mukherjee er a/., 1965 ; Ste inkraus et a/. , 1967: Padhyc and Salunkhe, 1978; Sathe and Salunkhe, 1981 a). However, the most com­ monly used and acceptable ratio of rice to dehulled black gram cotyledons is 4: I if the traditional g rinding stone is used, or 2: I if an electric blender is used for the grinding process. The quantity of water added to the in­ gredients varies from 1.5 to 2.2 times the weight of the dry ingredients (Reddy eta/. , 1982a). The microorgan­ Key W o rds: Phaseol11s vulgaris, Vigna mungo, fe rmen­ isms responsible fo r fermentation are the natural nora in tation, dehull cd beans , scanning electro n microscopy. t he starting materials ( Hesseltine and Wang, 1979). Bl ack gram plays a major role in id li fermenta ti o n both as a source of microorganisms and as a fermentation substrate (Radhakrishnamurthy eta/., 1961 ; Venkata­ subbaiah, 1984). Many investigators have studied the physico­ c hemical and microbiological changes occurring during fermentation of idli batter (Desikachar e1 a/. , 1960; Mukherjee et a/. , 1965 ; Ramakrishnan et a/. , 1976; Paper received June 28, 1993 Padh ye and Salunkhe, 1978 ; Reddy and Salunkhe, M anuscript received December 4, 1993 l980a; 1980b; Soni and Sandhu, 1989). Acidification Direct inquiries to B.G . Swanson and leavening are the two important changes that occur Telephone number: (509) 335-3793 during fermemation due to microbial activity (Reddy e1 FAX number: (509) 335-4815 at. , 1982a). The major microorganism responsible for fermentation was identified as Leuconostoc mesen­ reroides (M ukherjee et a/., 1965; Steinkraus et a/. ,

483 E . Josep h, S.G. Crites and B.G . Swanson

1967) , although oth er mic roorganisms such as Srreprococ­ scanning electron mi croscope (Hitachi In struments, San cusfaecalis. Lacrobacillus f ermenrum , and Saccharomyces Jose, CA) operated at 20 kV . cerevisiae are reported by oth ers (Venkatas ubbaiah er al .. Small portions of steamed idli were fi xed with 3% 1984; So ni and Sandh u, 1989). glutaraldehyde in 50 mM piperazi ne- N ,N' -bi s(2-ethanesul­ ld li has a soft and spongy texture and ple asant sour fo nic acid ) for 15 hours at 4 °C , post-fix ed in I % osmium flavo r. Steamed idli is ser ved with a sauce prepared fr o m tetroxide for 4 hours at ambient room temperature and de­ fresh coconut , and with sambar, a thin soup made of red hydrated for 10 minutes in each so lutio n of a graded series g ram dhal (pigeon peas) with vegetables and spi ces. ldli of 30, 50, 70, 95 and 100 % ethanol. After dehydration, is serve d as a weaning food for infants and al so served in th t': idli pieces were freeze-fractured in liqu id nitrogen and South Indian hospitals for feeding patients requiring soft critical point dried in carbon dioxide (Tousimis Samdri­ food (Ramak ri shnan , 1979 ; Reddy eta/. , l 982a, 1986). PVT-38; Tousimis Research Corp., Rockville, MD); Sllb stitution o f black gram with oi lsecds such as mounted on aluminum stubs using double si ded tape and soybean s and peanuts, or dry beans (Great Northern bean s) Leit-C-Piast plastic conductive carbon cement: and sputter in th e preparation o f idli has been reported (Ste inkraus er coated and examined in SEM as described above . a /., 1967: Ramakri shnan er al . . 1976; Sushee lamma and Rao. 1980 ; Sathe and Salunkhe, 198la; Soni and Sandhu , Results and Di scussion 1989). Research on idli fermentation and nutrit ive va lu e The scanning e lectron mi crographs o f unfer mented was carried out in the Un ited States at Co rne ll Univers it y, idl i batter and unfermented , 18-hour ferment ed, or 36-hour Utah State University and Washingto n St ate Un iversity. fe rmented , steamed idl i prepared from whole w hite beans However . research o n th e mi crostructure of id li is limited . in combination with rice are presented in Figures 1-4. In­ Therefore. this study was undertaken to study the micro­ tact starch granules, approx imately 26-30 11- m in diameter structu re of idli prepared from whole w hite beans. dehulled and protein bodies, approx imately 4-6 Jlm in diameter are wh ite beans, dehulled bl ac k beans (Phaseo/u s vulgaris) , or clearly visible in unfermented who le white bean and ri ce dehulled Indi an bl ac k gram (Vigna mungo) in comb in ati on idli batter ( Fig . 1). Intact . but smaller starch granules and with ri ce. protein bod ies were al so observed in unfermented idli bat­ Materials and Methods ter prepared from dehulled black gram and ri ce. Steaming cau sed disappearance of starch granul es but clumps of pro­ ldli preparation tein bodies persisted in unfermented and 18-hou r fermented Wh o le black beans (cv. Black Turtle Soup), wh o le idli prepared from whole white beans and rice (Figs. 2 and white beans (cv. Great Northe rn) (Phaseolus vulgaris). and 3). The surface of the idli was more cohesiv e and sponge­ long grain ri ce were purchased from a local mark et in like in appearance after 36-ho ur fermentation (Fig . 4) than Pullman , WA . Dehulled Indian black gram ( Vigna mungo) for unfermented idli or idli ferment ed for 18 hours (Figs. was purchased fr om an Indian grocery store in Chi cago. 2 and 3). The disappearance of starch granul es and most IL. Wh o le black and w ho le w hite beans were soaked sepa­ prote in bodies in Figures 2-4 may be due to th e ge latiniza­ rate ly in wate r equal to three times the weight o f the beans tion o f starch granules and denat urati on of protein bodies for 4 hou rs. and th e hu ll s were removed by hand to obtain during steaming of idl i (Hu ghes and Swanson. 1989). approximate ly 150 grams o f d ehulled beans fo r id li prepa­ Structural imegrity o f many protein bodies was ration . Rice was also soaked in water equ al to 1.5 times maintain ed in the unferme nted dehulled white beans and the weight of th e ri ce fo r 4 hours. The pro portion of rice ri ce co mbinatio n idl i upon steaming (Fi g. 5), compared to to dehulled beans was 2: I on dry weight basis fo r prepar­ th e 18-hour ferment ed idli ( Fi g. 6). The surface o f idli in g idli . Ri ce and bean s for each variety o f beans were became more cohesive an d foam-like with an in crease in ground separately in a bl ender to obtain a batter o f se mi· fermentation time (Fig . 7). thi ck consistency. and mixed together. Half of th e batter The mi crostructure of idli prepared fro m dehulled was poured into a co ntainer with a tight fitting lid and in ­ white beans and rice (Figs. 5-7) is similar in appearance to Cllbated at 30 oC fo r 18 hours or 36 hours . The fermented th e idli prepared from dehulled bl ack beans and rice (Figs batter was mix ed well and a spoon of batter was dished 8 ~ 10) . Gas cells of the foam- lik e structure are clearly evi­ into an idli pan containing individual cups with a diameter dent in Figure 10. The structure of id li prepared from un­ o f 3 in ches (7. 6 em) and a depth of I in ch (2. 5 em) with 25 fermented dehulled fndi an black gram and ri ce (Fig . t t) perforations. The perforations were necessary for th e easy was similar to the structure o f unfermented idli prepared entry o f steam in to the idli during steaming. Th e pan with fro m dehulled black and dehull ed white beans with rice. batter was pl aced in a steamer and steamed for 7 minutes Structural integrity o f many prote in bodies was maintained at 100°C . The pan was removed from the steamer, cooled in the unfermented idli Loss of protein body structure and th e individu al steamed idli were removed. Th e proce­ began after 18~hour fermentation (Fig . 12) and structural dure was repeated for unfermented batter. integrity was almost co mplete ly lost after 36-hour fermen­ tati on o f idli prepared from dehulled Indi an bl ac k gram and Scanning electron microscop y (SEM) ri ce (Fig. 13) due to the acti vit y o f mi crobial enzy mes such Unfermented idli batt er was frozen in liqu id nitro­ as amylases and proteases (Reddy and Salunkhe , 1980a; ge n, lyo philized. mounted on aluminum stubs using carbon Soni and Sandhu. 1989). After 36-hour fermentation, no tape and Temp fi x wax , sputter coated with 30 nm go ld us­ structural differences were o bserved among th e idli pre~ ing a Techni cs Hummer V sputter coating unit (An atech pared from dehulled white beans, dehul led black beans, or Ltd ., Al ex andria. VA ) and viewed with an Hitachi S570 dehulled black gram (Figs. 7 , I 0 and 13) with rice . Except

484 II Microstructure of Jdli

Figure 1. Unfermented idli batter prepared from whole white beans and rice exhibiting starch granules (S) and protein bodies (PB). Figures 2 and 3. Steamed idli prepared from whole white beans and rice exhibiting clumps of protein bodies (PB) , unfermented (Figure 2) and 18-hour fermented (Figure 3). Figure 4. Steamed id li prepared from whole white beans and rice and fermented for 36 hours. for the idli prepared from whole white beans and rice, each black gram are responsible for the development of the soft, idli after 36-hour fermentation exhibited a cohesive, dense spongy texture of idli , and arabinogalactan is essential for and foam-like structure. The surface structure of id li ob­ rai sing and stabilizing of the spongy texture produced by served with increasing fermentation time may result from the su rface active globulin proteins. The arabinogalactan physico-chemical and microbiological changes occurring helps to trap and hold th e leavening gases , and prevents during fermentation such as an increase in soluble solids, disruption of the protein foam by heating at steaming tem­ acidity and alpha-galactosidase activity and a decrease in peratures. Howeve r, a certain degree of di sruption of th e soluble sugars (Rajalakshmi and Vanaja, 1967; Padhye and network does occur resulting in disruption of susceptible Salunkhe, 1978; Reddy and Salunkhe, 1980b, Reddy era!., areas along the protein surface by excessive stress devel­ 1982b; Soni and Sandhu, 1989). Susheelamma and Rao oped during steam cooking. Although the literature on the (1974; 1978a; 1978b and 1979) indicated that the smooth foaming properties of common beans of Phaseolus vulgaris and porous structure of idli is due to the presence of is limited, th e globulin proteins and polysaccharides re­ surface active globulin proteins and an arabinogalactan in sponsible for foa ming are present in common beans as well Tndian black gram cotyledons. The globulin proteins of as Indian black gram cotyledons (Satterlee et al. , 19 75; Sosulski et al., 1976; Sathe and Salunkhe, 198!b, 1981c)

485 E. Joseph , S.G. Crites and B. G. Swanson

486 Microstructure of Jdli

Figures Sand 6 . Steamed idli prepared from de hulled white beans and ri ce , unfe rme nted (Figure 5), and 18-hour fe r· mented ( Figure 6) . Note many intact protein bodies (PB) are visibl e in Figure 5. Figure7. Steamed idli prepared from dehulled white beans and ri ce and ferment ed for 36 hours. Figures 8, 9 and 10. Steamed idli prepared from dehu ll ed black beans and rice. unfermented (Figure 8), 18· hour fermented (Figure 9) and 36-hour fer me nted (Figure I 0)

Figure II (at right) Steamed idli prepared from un fer· mented dehull ed In dian black gram and rice exhibiting many intact protein bodies (PB) .

The observati ons of idli microstructure suggest that the spongy or foa my texture of idli prepared from whole white , dehulled white, dehulled black beans of Phaseolus vulgaris with rice may be due to the foaming of proteins and polysacc harides as observed for the idli pre pared from dehulled Indian black gram with ri ce. Conclusions Mi crostructure of idli prepared from fermented beans ( Pha seolus vulgaris) and rice was simi Jar to the mi c rostruc· ture of idli prepared from fermented dehulled Indian black gram and ri ce. Although the traditional method of preparing idli in India uses de hulled black gram and rice , dehu ll ed white or black bea ns can be used for preparing idli . When visuall y observed, th e textu re of idli prepared from whole w hite be an s and ri ce was comparable to the texture of idli prepared from de hull ed beans and rice. The cohesive, de ns e and sponge· or foa m- like surface o f idli obtained with increasing fermemation time may be due to the ph ysico­ c he mi ca l and mi c robiolog ical changes occurring during the fermentation process as we ll as the presence of su rface active globu lin proteins and an a rabinogalactan. Acknowledgement The au thors ac kn ow ledge the use of the facilities and assistance of the technical staff of the Electron Mi croscopy Center, Washington State University. Pullman . WA . This project was partially funded by the Re search Service Council , Uni versi ty of Nebraska at Kearney , Kearney, NE. Partial financial support also was provided by USA I D Title X II Dry Bean/Cowpea CRSP. References Chavan JK . Kadam SS. (1989). Nut ritional improve­ ment of cereals by fe rm e ntati on. CRC Crit. Rev . Food Sc i. Nutr. 28 , 349-400. Desi kac har HSR. Radhakrishnamunhy R, Ra o GR , Kadol SB. Srinivasan M, Subramanyan V. (1 960). Studies on idli fermentation. I . Some accompanying changes in the batter. J. Sci. Indus. Res. (India) 19C , 168· 172. Hesselt ine CW. Wang HL. ( 1979). Fermented foods . Chem Ind. 12. 393 -399. Hughes JS. Swanson BG . ( 1989). Soluble and insolu· bl e dietary fibe r in cooked comm on bean (Phaseolus vulgaris) seeds. Food Mi crostruc. 8, 15-21 Mukhe rj ee SK . Albury MN , Pederson CS , VanVeen F ig ures 12 a nd 13 . Steamed idli prepared from dehul\­ AG . Steinkraus KH . ( 1965). Role of Leuconostoc mesen­ ed Indian black gram and rice and fermented 18 hours reroides in leavening th e batter of idli. a fermented food of (Figure 12) and 36 hours (Figure 13).

487 E. Joseph, S.G. Crites and B. G. Swanson

India . Appl. Microbial. 13, 227-231. 9. Marcel Dekker, New York. pp. 131 - 188 ,280-299 Padhye VW , Salunkhe OK. (1978). Biochemical studies Steinkraus KH . Va n Vec n AG , Thiebeau DB . (1967). on black gram (Phaseolus mungo L.). Ill. Fermentation of black Studies on 'idl i'. a fermen ted black gram-rice food . Food gram and ri ce blend and its innuence on the in vitro digestibility Techno! . 21. 916·919. of the proteins. J . Food Biochem. 2, 327-347. Sushcelamma NS. Rao MVL. ( 1974). Surface active Radhakrishnamurthy R, Desikachar HSR, Srinivasan M, principl e in black gram (Phaseolus mungo) and its role in th e Subramanyan V. (1961). Studies on idli fermentation. II . texture of foods contai ning th e legume. J. Sci . Food Agric. 25, Relative participation of black gram fl our and rice semolina in 665-673 . the fermentation. J. Sci. Ind . Res. (India). 20C, 342-345. Susheelamma NS. Rao MVL. (1978a). Isolation and Rajalakshmi R, VanajaK. (1967) . Chemical and biologi­ characterization of th e arabinogalactan from black gram cal evalu ation of th e effect of fermentation on th e nutritive value (Pha::.·eofus mungo) . J. Agric . Food Chern . 26, 1434- 1437 of foods prepared from rice and grams. Brit. J. Nutr. 21 , 467- Sushee lamm a NS , Rao MVL. ( 1978b). Purification and 473 . characteri zation of the surface active proteins of black gram Ramakrishnan CV. (1979) Studies on Indian fermented (Phaseolus mungo). Intern. J. Pep. Prot. Res. 12 , 93-102. foo~ . Baroda J. Nutr. (Ind ian) 6, l-57. Susheelamma NS , Rao MVL. (1979). Functional role of Ramakrishnan CV, Pariekh U , Akolkar, PN , Rao GS , the arahinogalactan of black gram (Phaseolu::.- mungo) in the tex­ Bandari SO. ( 1976). Soy idli fermentation . Plant Foods Man 2 , ture of leavened foods (steamed puddings). J. Food Sci. 44, 15·33. 1309·1313. Reddy NR , Salunkhe DK. (1980a). Effects of fermenta· Susheelamma NS. Rao MYL. ( 1980). Potenti ali ties of tion on phytate phosphorus and mineral content in black gram, oil seed nours and proteins for replaci ng hlack gram rice, and black gram and rice blends. J. Food Sci. 45, 1708- components in the texture of leavened foods. J. Amer. Oil 1712 . Chern . Soc. 57,2 12-2 15 . Reddy NR , Salunkhe DK. (1980b). Chang es in oli gosac· Yenkatasubbaiah P, Dwarakanth CT, Murthy YS . charides during germination and cooking of black gram and fer­ ( 1984). Mi crobiological and ph ysico-chemi cal changes in idli mentation of black gram /rice blend. Cereal Chern . 57, 356-360. hatter during fermentation . J . Food Sci . Techno!. (India). 21 , Reddy NR, Sathe SK, Pierson MD, Salunkhe OK. 59·62. (1981). Jdli , an Indian fermented food: A review. J. Food Qual. 5, 89· 101. Discussion with R~viewcrs Reddy NR, Pierson MD, Sathe SK, Salunkhe, OK. Om Johari · Based on yo ur wo rk , can yo u make any practical (1982a). Legume-based fermented foods: thei r preparation and recom mendations regarding: I) selection of ingredient s: kind of nutritional quality . CRC Crit. Rev . Food Sci. Nutr. 17 , 335- beans. whole or dchullcd beans, and type of rice (e.g .• brown. 370. parboiled, or regular): 2) selecti on of grind ing method: grinding Reddy NR , Salunkhe DK, Sathe SK . (1982b). Bin· stone, blender, or use of ground nours; 3) cook ing time; and 4) chemistry of black gram (Phaseo/us mungo L.): A review. CRC use of an additional leavening agent such as yeast? Crit. Rev . Food. Sci. Nutr. 16, 49-114. Au thors: When using black beans or white beans, dehulled Reddy NR , Pierson MD, Salunkhe DK . (1986). ldli. In beans are better for making idli batter. The da rk color of the Legume Based Fermented Foods. CRC Press Inc., Boca Raton , black hean seed coat is unacceptable in idli si nce th e-l:olor of FL . pp . 146· 158. idli is trad itionally white. In India. dehulled hlack gram is Sathe SK, Salunkhe, OK. (1981a). Fermentation of th e readily available. Dehtdling equipment is not readil y ava il ab le Great Northern bean (Phaseolus vulgaris L.) and ri ce blend s. 1. in the United States. The seed coat of white bea ns does not sig­ Food Sci. 46 , 1374-1378 , 1393. nificantly change th e color of idli and may increase dietary fiber Sathe SK, Salunkhe OK. (198 1b). Functional properties content. However. th e length of time necessary for grinding to of the Great Northern bean (Phaseolus vu lgaris L.) proteins: an acceptable batter co nsistency may increase du e to the pres­ Emul sion, foaming , viscosity, and gelation properties. J. Food ence of th e seed coat. Any type of rice can be used for making Sci . 46 , 617-621. idli batter. Parbo il ed ri ce is used in India . Brown rice also ca n Sathe Sk, Salunkh e OK. (l98lc). Isolation and partial be used to make id li , hut preference for th e idli may decrease characteri zation of arabinogalactan from the Great northern bean due to th e color contributed by th e rice. (Phaseolus vulgaris) . J . Food Sci. 46 , 1276·1277. Although use of a bl end er is acceptable for making idli Sathe SK, Reddy NR, Pierson MD. (1986). Future of batter, the use of th e trad itional gri nding stone is better legume based fermented foods. In Legume-based Fermented Grinding takes a longer time when using th e grinding stone, Foods. CRC Press, Inc. Boca raton, FL. pp. 233-235. th ereby incorporating more ai r into the bauer. Use of ground Satterlee LD, Bembers M, Kendrick JG . (1975). fl ours is unacceptable for making idli batter since less air is Functional properties of Great Northern bean (Phaseolus incorporated in mixing. The tex ture of idli mad e with ground vulgaris) protein isolate. J. Food Sci. 40, 81 -84. nours is heavy and compact. Commercial idli mixes are availa­ Soni SK , Sandhu OK. (1989). Fermentation of idli: ble. However, the texture of idli mad e from th e mixes is poor Effects of changes in raw material and physico-chemical when compared to the texture of idli made from fres hly ground conditions. J. Cereal Sci. 10, 227-238. bea ns and rice. Seven to ten minutes is suffic ient fo r cooking Sosu1ski F, Garratt MD, Slinkard AE. (1976). Func­ time. Stea ming the batter longe r than 10 minutes may allow tional properties of ten legume flours. Can. lost. Food Sci. condensation from th e steam to drip fr om the I id of th e stea mer, Techno!. 9, 66-jj9 . mar the surface of th e idli. and produ ce an unacceptable Steinkraus KH (1983). Section II : Indigenous fermented appearance. Use of an additional leavening agent , such as foods involving an acid fe rm entation: Preserving and enhancing yeast, is unnecessary if th e temperature for fermentation is organoleptic and nutritional qualities of fresh foods. In : Hand­ mai nt ai ned at 30°C for 12- 18 hours. book of Indigenous Fermented Foods. Microbiology Series. Vo l

488