Isolation, Partial Characterization and Modification of the Great Northern Bean (Phaseolus vulgaris L.) Starch

S. K. SATHE and D. K. SALUNKHE

ABSTRACT Microscopy of starch The yield of the Great Northern bean starch was 18.23% (bean flour basis). The starch granule size ranged from 12 x 12 pm to 58 x 40 pm (length x width). The shape of starch granules was round to oval to elliptical, and in some cases, concave as well. Lamellae were pres- ent on all the starch granules observed. Amylose content of the starch was 10.2% (starch basis). Hog pancreatic cY-amylase hydrolyzed more starch than did malt a-amylase under similar conditions. The Great Northern bean starch had good water and oil absorption capa- cities at room temperature (21C). The bean starch formed a stable gel at concentrations of 7% and above (w/v). The viscoamylographic studies of the isolated starch indicated the restricted-swelling char- acter of the bean starch.

The purified starch was studied microscopically by employing both light and scanning electron microscope. For light micro- scopic studies, starch was moistened with a drop of distilled water. A calibrated eyepiece lens (152.5@ (calibrated against a stage micrometer) was employed for the measurements of starch granule size at a magnification of 344. The starch granules were studied for the size, shape, hylum, and lamellae. The size measurements were made on 25 representative granules.

INTRODUCTION

Starch samples for scanning electron microscopic studies were prepared as follows. Starch was sprinkled on an aluminum stub (with a double-stick tape on it) and was coated with a gold-palla- dium alloy completely in a Polaron E 5000 Sputter Coater (U.K.) and the specimens observed in AMR 1OOOB scanning electron micro- scope (Cambridge, Mass.) at 20 KV accelerating voltage and suitable magnification(s).

LEGUMES contain about 60% carbohydrates including starch, reducing and nonreducing sugars, oligosaccharides of the raffinose family, and others. Starch constitutes the major portion of legume carbohydrates. Cerning-Beroard and Filiatre (1976) studied the carbohydrate composition of horsebeans, smooth and wrinkled peas, and lupine seeds. They found that the average starch content of horsebeans, smooth and wrinkled peas, and lupine seeds to be 41 .O, 48.0, 33.0, and 0.4%, respectively. Naivikul (1977) re- ported the starch content range to be 50.9-52.9% (mois- ture free basis) in navy bean, pinto bean, faba bean, lentil, and mung bean. Schoch and Maywald (1968) discussed the difficulties encountered in the separation of horsebean starch. They attributed these difficulties to the presence of a highly hydrated fine fiber fraction (presumably from the cell walls enclosing the starch granules) and high content of insoluble proteins. Halbrook and Kurtzman (1975) studied the water uptake by the Great Northern bean starch at high (80-148(Z) temperatures. Recently, Lai and Varriano- Marston (1979) reported certain physicochemical charac- teristics of black bean starch.

Hydrolysis of starch Starch was hydrolysed by two different ol-amylases (from Hog

pancreas, type VI-A, and from malt, type V-A; both from Sigma

I RESIDUE F 30L 2% N&I, 24 h, 4C

1 RESIDUE 1 1 Washed with 2L H,O, blended with 6L O.IN NaOH (1 min.1 m a Warmg Blendor and extracted for 48 h. 4C

I CENTRIFUGE I- 10,000 RPM, 30 min. The purpose of the present investigation was to isolate

the Great Northern bean starch and to study certain physi- cochemical properties of the unmodified and modified starch.

MATERIALS & METHODS The Great Northern beans were purchased from Bean Growers Warehouse, Filer, Idaho, and stored at 4C until experiments were conducted. The residue obtained after preparation of the protein concentrates on a pilot plant scale (Sathe et al., 1980) was referred to as crude starch. All the chemicals were of reagent grade unless mentioned otherwise. All the analyses were performed in triplicate and means reported.

I RESUSPEND IN 80% AQ. ETHANOL 4 * 4

HEATING WATER BATH

Isolation of starch The beans were ground to 20 mesh in a Fitz mill (The W.J.

Fitzpatric Co., Chicago, Ill.). Three kg of bean flour were extracted sequentially with different solvents to yield starch. The schematic diagram for the process is presented in Figure 1.

FREEZE DEHYDRATE

1 STARCH POWDER I

Authors Sathe and Salunkhe are affiliated with the Dept. of Nutri- tion & Food Sciences, Utah State Univ., Logan, UT84322. Fig. l-Schematic diagram for the isolation of Great Northern bean

starch.

L Blended 1 min. in a Waring Blendor 80C. 1 h

4h. 4-C; Discard supernatant

Volume 46 (1981kJOURNAL OF FOOD SCIENCE-617

Chemical Co., St. Louis, MO.) according to the method described by Decker (1977). Maltose hydrate (Grade II, Sigma Chemical Co., St. Louis, MO.) served as reference standard. The starch to enzyme ratio in each case was 1:l (w/w). Liberated maltose was measured, after inactivating the enzyme (heating in a boiling water bath for 3 min), calorimetrically (3,5dinitrosalicylic acid as color reagent) at 540 nm in a Beccman-DBG spectrophotometer. Maltose equiva- lent was determinec at time intervals of 0, 15, 30, 60,90, and 120 min. The incubatiorI was, in both cases, at pH 7.0 and at a tempera- ture of 21C.

Modifications Acetylation-The method of Wurzburg (1964) was followed.

One hundred grams of starch were dispersed in 500 ml of distilled water and magnetic:ally stirred for 30 min to obtain a uniform suspension. The pH 1 rf the slurry was adjusted to 8.0 with 1N NaOH. Acetic anhydride (11.2g) was then added slowly to this slurry main- taining constant stirring and monitoring the pH between 8.0-8.4. The reaction was al.owed to proceed for an additional 5 mm after completion of the acetic anhydride addition. The pH of the slurry was finally adjusted o 4.5 with 0.5N HCl and filtered through What- man filter paper #4. The residue was then washed five times with distilled water and freeze dehydrated.

Oxidation-A uniform slurry of starch (1OOg starch in 500 ml distilled water) was prepared as in acetylation. Oxidation of the purified starch wa:. accomplished by the method of Hullinger (1964). The pH of the slurry was first adjusted to 9.0-9.5 with 3% aqueous NaOH and log of NaOCl was added slowly (over a period of 90 min) while rraintaining the magnetic stirring and constantly monitoring the pH lletween 9.0-9.5. Cooling was provided (crushed ice with NaCl) simitltaneously. The reaction was allowed to pro- ceed for 4 hr after IJaOCl addition was completed, pH adjusted to 7.0 with 0.5N HCI and the slurry was filtered through Whatman filter paper #4. Th: residue was washed five times with distilled water and freeze dehydrated.

Moisture The moisture content of the samples were determined by the

AACC method 44-15 (1962).

Proteins Protein content of the appropriate .samples was determined

by the Kjeldahl method (N X 6.25).

Fat The method followed was that of Schoch (1964). Starch (5g)

was hydrolyzed witlr 8N HCl solution for 1 hr on a boiling water bath. After successb.e extraction of fat with ether and petroleum ether, the solvents Mere evaporated on a hot water bath (80C) and the nearly dry samples were then dried in an oven (100C) for 20 min followel by three successive extractions with carbon tetrachloride (10 m. each time). The combined extracts were fil- tered and the solvc,nt evaporated on a hot water bath (SOC). The beakers (previot sly weighed) containing the nearly dry samples were dried in an oven (100C) for 30 mm and weighed. The differ- ence was interpretet as the weight of fat. The fat content was re- ported on a dry weigrt basis.

Amylose content Amylose content of the purified starch was determined by the

procedure of Wolf et al. (1970) with slight modifications. Pure amy- lose (Potato, Type 111, Sigma Chemical Co., St. Louis, MO.) served as standard. Starch lvas dissolved in 90% (v/v) dimethyl sulfoxide and 0.2, 0.5, and 1.(1 ml portions were assayed for the amylose con- tent. The starch concentration in 90% dimethyl sulfoxide was 100 mg/lOO ml. One ml of each of 0.005N KI03, 0.016 KI, and 0.5N HCl were then addec to a total of 1 ml of standard/sample and final volume (9 ml) made up with distilled water. Absorbance was read at 615 nm in Beckman DEG spectrophotometer.

Water and oil absorption Water and oil (Crisco vegetable cooking oil, density = 0.8888

g/ml) absorption capacities of the purified starch and the modified starches were deternined by the centrifugal method (Beuchat, 1977). One gram 0: sample was mixed with 10 ml of distilled water/oil (Sari-whirl, mixing control, fast) for 30 set, allowed to

618-Volume 46 /1981)-JOURNAL OF FOOD SCIENCE

stand for 30 min at room temperature (21(Z), centrifuged at 5000 X G for 30 mm and the volume of the supernatant noted. Density of distilled water was assumed to be 1 g/ml. Results were expressed on a dry weight basis.

Pasting properties Starch gelatinization curves were obtained by the method of

Sandstedt and Abbott (1964). Starch (2Og, dry weight basis) was suspended in 350 ml of distilled water in a Waring Blendor. CMC (Cellulose Gum 7 HP, Hercules Powder Co.. Wilminaton. Del.) was added (3.6g) with gentle stirring over 30 set to this suspension, blended for 1 min, and poured into the amylograph bowl. The blender was then rinsed with 100 ml distilled water and the water was added to the amylograph bowl. The temperature of this starch- CMC suspension was then raised from 25C to 95C at a rate of 1.5C/min; held at 95C for 15 min and then cooled uniformly to 50C (1.5C/min). A blank curve for CMC was prepared similarly (with 3.6g of CMC alone) and subtracted from the starchCMC curves.

Gelation The method of Coffman and Garcia (1977) was employed with

slight modifications. Purified starch suspensions of 1,3,5,7,9,11, 13, 15, 17, and 20% (w/v) were prepared in 5 ml distilled water and the test tubes were heated in a boiling water bath for 1 hr followed by rapid cooling under running cold tap water. The test tubes were further cooled for 2 hr at 4C. Least gelatinization concentration was determined as that concentration when the sample from the inverted test tube did not fall down or slip.

Degree of substitution The degree of substitution (D-S.) for the acetylated starch was

determined according to Wurzburg (1964). Starch (5g) was dis- persed in 75% aqueous ethanol and warmed for 30 mm on a water bath (5OC), cooled to room temperature (21C), and 25 ml of 1N NaOH added. The stoppered flasks were then allowed to stand for 72 hr with occasional shaking at room temperature (21C). Excess NaOH was back titrated with 1N HCl. The flasks were al- lowed to stand at room temperature (21C) for 2 hr and the titra-

Table 1-Physichochemical data on the Great Northern bean starch

Yield Moisture Protein FatC Sample (%) Pd 1%) (%)

Crude starch 87.50a 3.07 4.86 0.34 Purified starch 1 8.23a 2.67 0.97 0.46 Acetylated starchd 92.00b 3.70 - - Oxidized starch 84.40b 4.12 - -

a On bean flour basis b On purified starch basis i Dry weight basis

Degree of substitution (D.S.) = 0.40

Fig. 2-Light photomicrograph of Great North&m bean starch I1 75X).

GREATNORTHERN BEAN STARCH.. .

tion completed. A blank with pure starch was conducted concur- rently and the D.S. calculated as follows:

% Acetyl =

(ml blank - ml sample) X normality of HCl X 0.43 X 100 Weight of sample (g) dry basis

D.S. = 162 X % Acetyl 4300 - (42 X % Acetyl)

RESULTS & DISCUSSION

Composition and yield The data on composition and yield are presented in

Table 1. The purity of the isolated starch was judged on the basis of composition and microscopic observations. The yield of Great Northern bean starch was 18.23% (on bean flour basis). Naivikul and DAppolonia (1979) re- ported yields of 40.3, 38.3, 39.9,42.5, and 34.5% for navy bean, pinto bean, faba bean, lentil, and mung bean starch respectively. Schoch and Maywald (1968) obtained starch yields of 27, 38, and 37% from navy bean, lentil, and mung bean seeds respectively. Lineback and Ke (1975) reported 37% starch yield from horsebean flour. The differences in yields have been attributed to the methods opted for starch isolation (Naivikul and DAppolonia, 1979). A yield of 18.23% in the present investigation which was lower than those of navy and pinto beans, both Phaseolus vulgaris species may be primarily due to the method of isolation.

Granule size and microscopic appearance The shape, size, and bifringence of starch granules are

often representative of the plant species and its maturity (Manners, 1974). Several fiels were observed and measure- ments of 25 representative granule sizes were made. The range of granule size was about 12 X 12 pm to 58 X 40 pm (length X width) which was in close agreement to the ranges reported by Naivikul and DAppolonia for navy and pinto bean starches (12-36 and 16-28 pm for width, 1240 and 1640 pm for length, respectively, for navy and pinto bean starch granules). The shape of Great Northern bean starch granules was quite varient, ranging from small round to large oval to irregular. Some granules were con- cave. Similar observations have been reported by Line- back and Ke (1975) on chick pea and horsebean starches, and by Lai and Varriano-Marston (1979) on black bean starch.

Light microscopic observations (Fig. 2) of the starch granules revealed the presence of hylum and lamellae. In general, the hylum paralleled the longitudinal axis of the starch granule; however, in case of spherical granules such as distinction could not be made. Hylum was absent on some granules; however, lamellae were observed in all the granules viewed. Hylum was found to possess different shapes and varying lengths. Similar observations have been reported on starch grains of Phaseolus species (Dhaliwal et al., 1964) and lima beans (Salunkhe and Pollard, 1955a, b; Salunkhe, 1957).

Scanning electron photomicrographs are presented in Figure 3. As can be seen from these photographs, starch granules appear to be round, oval, and elliptical. The sur- faces appeared to be smooth. The lamellae observed in the light microscopic view (Fig. 2) were not evident in the scanning electron microscopic observations on starch gran- ules. This may have been due to the dehydrated state of starch granules in scanning electron microscopy samples versus the hydrated state in light microscouv. Some starch granules appeared to be do-nut shaped. -Hall and Sayre (1971), McEwen et al. (1974), and Schoch and Maywald

(1968) have reported similar observations on legume starches. The cell wall structure is shown in Figure 4. The starch granules can be seen enclosed in the cell wall. The isolation treatments did not remove completely all the cell walls and some did survive as shown in the Figure 4.

Amylose content and starch hydrolysis The amylose content of Great Northern bean starch was

Fig. 3-Scanning electron photomicrographs of Great Northern I

bean starch: (Al 365X; (B1 730X; (Cl 1460X.

Volume 46 (1981bJOURNAL OF FOOD SCIENCE-619

10.2% (on starcll basis) which was comparable (in order) to that reported in Amsoy 71 soybeans (15-20%) by Wil- son et al. (1978). Results of starch hydrolysis employing a-amylases are p resented in Figures 5 and 6. Hog pancreatic a-amylase hydrolysed 8.2% starch which was higher than that by cw-amzla:e from malt, in a 2-hr period at room tem- perature (21 C). This difference may be due to the differ- ence in the activities of enzymes used (11 mg maltose and 3.3 mg maltose liberated per mg of enzyme in 1 min at 20C at pH 6.9 for hog and malt ol-amylase, respectively). The low degree of, hydrolysis may be due to the starch nature (uncookell raw starch was employed) and a relative- ly low temperature (21C) during incubation.

Water and oil abr orption The water alld oil absorption data are presented in

Table 2. Modifications did not improve both water and oil absorption capac:ity of starch. The purified starch had oil and water absorption capacity of about 2.9 g/g and 2.93 g/g, respectively. Halbrook and Kurtzman (1975) have reported a water uptake of about 3.0 g/g and about 3.0 g/g at 121C and 80C, respectively. Our results of water absorption (2.92 g/g at 21C for the purified starch) were comparable to .:heir observation of water absorption at 8OC. The high water absorption at 21C observed in the present investigation may have been due to the nature of

Fig. 4-Scanning eh&on photomicrograph of the cell wail, 1460X.

the starch and a possible contribution to water absorption by the cell wall material(s) which was not removed com- pletely. Comer and Fry (1978) have reported cold water absorption of the purified pea starch to be 92-105%, and that the water uptake was a function of temperature.

Pasting properties and gelation The amylograms are the plots for the corrected viscosity

(Fig. 7). The data are summarized in Table 3. Peak heights were not reported as the amylograms did not have distinct peaks. With the exception of the oxidized starch, all other samples followed similar patterns. The change in viscosity after holding for 15 min at 95C was rather slow, except for oxidized starch in which case it decreased sharply dur- ing the cooling cycle. The gelatinization temperature range (65.5-68.5(Z) of the purified Great Northera bean starch was comparable to those of faba bean (66 Cd and lentil (68C) (Naivikul, 1977); garbanzo bean (65-71 C), smooth pea (65-69(Z), red kidney bean (64-68OC), and mung bean (63-69C) (Biliaderis et al., 1979); and black bean (63.8-76C) (Lai and Varriano-Marston, 1979) starches. The trend of the purified starch curve was characteristic of restricted swelling type starches. The viscosity behavior of the oxidized starch was characteristic of hypochlorite oxidized starches which show a greater degree of fluidity.

Table 2-Water andoilabsorption by the Great Northern bean starch

Sample

Purified starch Acetylated starch Oxidized starch

Water absorbed Oil absorbed Sk Sk

2.93 2.94 2.68 1.88 2.60 2.26

Table 3-Amylogram Data of the Great Northern bean starch

Sample

Crude starch Purified starch Acetylated starch Oxidized starch

Gelatinization temp range

(C)

62.587.0 65.5-68.5 61.0-64.0 65.5-68.5

15 min

6

425 295 355

40

50C htb (BU)

325 445 475

Ii

a Viscosity of the corrected starch curveO(in Brabender Units) at the end of 15 min period of holding at 95 C.

b Viscosity at 50C (In Brabender Units) during the cooling cycle.

Fig. 5-Light photllmicrograph of a-amylase (hog pancreas) attack on starch granule. (A) 0 hr; IBJ 2 hr. The magnification of B is 2.5 times that ofA. I

620-Volume 41; (198lkJOURNAL OF FOOD SCIENCE

GREAT NORTHERN BEAN STARCH.. .

The tendency to set back on cooling is minimized in oxi- dized starches due to the presence of functional groups that block the association tendencies of the starch chains (Scallet and Sowell, 1967). The gelation studies indicated that purified starch could yield stable gels at concentrations of 7% or above (w/v).

O-0 Hog d-Amylare

- . . . . . . . . M&d-Amyline

I I I I I a 1 I 30 45 60 75

TIME (min.) -*

90 105 120

Fig. 6-Starch hydrolysis by or-am ylases.

Fig. 7-Gelatinization curves (corrected for CMC) of Great Northern bean starch.

REFERENCES AACC 1962. Approved Methods of the AACC, 7th ed. American

Association of Cereal Chemists. St. Paul. Minn. Beuchat, L.R. 1977. Functional and electrophoretic characteristics

of succinylated peanut flour proteins. J. Agr. Food Chem. 25: 258. Billaderls, C.G., Grant, D.R.. and Vase. J.R. 1979. Molecular weight

distributions of legume starches by gel chromatography. Cereal Chem. 56: 475.

Cerning-Beroard. J. and Flliatre, A. 1976. A comparison of the car- bohydrate composition of legume seeds: Horsebeans. peas, and lupines. Cereal Chem. 53: 968.

Coffman, C.W. and Garcia, V.V. 1977. Functional properties and amino acid content of a protein isolate from mung bean flour. J. Food Technol. (U.K.) 12: 473.

Comer, F.W. and Fry M.K. 1978. Purification, modification, and wooerties of air-clas;fied nea starch. Cereal Chem. 55: 818.

Decker. L.A. 1977. Worth&ton Enzyme Manual, p. 173. Worth- ington Biochemical Corporation, Freehold, N.J.

Dhaliwal. AS.. Pollard. L.H.. and Salunkhe. D.K. 1964. Biosynthe- sis of starch &ins ln.cotyledons of Phaseblus species. Amer. Sot. Hort. Sci. 85-361.

Halbrook, W.U. and Kurtzman, R.H. Jr. 1975. Water uptake of bean and other starches at high temperatures and pressures. Cereal Chem. 52: 156.

Hall. D.M. and Sayre, J.G. 1971. A scanning electron microscope study of starches. 3. Miscellaneous starches. Textile Res. J. 41: 880.

Hullinger. C.H. 1964. Hypochlorite oxidized starch. In Methods in Carbohydrate Chemistry: Starch, Vol. 4, Ed. Whistler, R.L., p. 313. Academic Press, New York.

Lai, C.C. and Varrlano-Marston, E. 1979. Studies on the charac- teristics of black bean starch. J. Food Sci. 44: 528.

Lineback. D.R. and Ke, C.H. 1975. Starches and low molecular weight carbohydrates from chick pea and horsebean flours. Cereal Chem. 52: 334.

Manners, D.J. 1974. The structure and metabolism of starch. In Essays in Biochemistry, Vol. 10, p. 37. Ed. Campbell, P.N. and Dickens, F.. Academic Press, New York.

McEwen. T.J., McDonald, B.E.. and Bushuk. W. 1974. Faba bean (Vicia faba minor). Physical, chemical, and nutritional properties. Unpublished report, The Fourth International Food Congress, Madrid, Spain.

Naivikul, 0. 1977. The carbohydrates present in flour obtained from various types of legumes. Ph.D. thesis, North Dakota State University, Fargo, N.D.

Naivikul. 0. and DAppolonia. B.L. 1979. Carbohydrates of legume flours compared with wheat flour. 2. Starch. Cereal Chem. 56: 24.

Salunkhe, D.K. 1957. Histological and histochemical changes in gamma-irradiated lima beans, Phaseolus lunatus. Nature 179: 585.

Salunkhe. D.K. and Pollard, L.H. 1955a. A rapid and simple method to determine the maturity and quality of lima beans. Food Tech- nol. 9: 45.

Salunkhe, D.K. and Pollard, L.H. 195513. Further studies on micro- scopic examination of starch grains in relation to maturity of lima beans. Food Technol. 9: 521.

Sandstedt, R.M. and Abbott, R.C. 1964. A comparison of methods for studying the course of starch gelatinization. Cereal Sci. Today 9: 13.

Sathe, S.K., Ponte, J.G. Jr., Rangnekar, P.D., and Salunkhe, D.K. 1980. Effects of addition of Great Northern bean (Phaseolus vul- garis L.) flour and protein concentrates on rheological properties of dough and baking quality of bread. Cereal Cbem. (In press).

Scallet, B.L. and Sowell, E.A. 1967. Production and use of hypo- chlorite oxidized starches. In Starch Chemistry and Technology. Industrial Aspects, Vol. 2, p. 243. Ed. Whistler, R.L. and Paschall. E.F.. Academic Press, New York.

Schoch, T.J. 1964. Fatty substances in starch. Determination and removal. In Methods in Carbohydrate Chemistry: Starch, Vol. 4, p. 60. Ed Whistler, R.L. Academic Press, New York.

Schoch, T.J. and Maywald, E.C. 1968. Preparation and properties of various legume starches. Cereal Chem. 45: 564.

Wilson, L.A.. Birmingham, V.A.. Moon, D.P., and Snyder, H.E. 1978. Isolation and characterization of starch from mature soy- beans. Cereal Chem. 55: 661.

Wolf, M.J., Melvin, E.H.. Garcia, W.J., Dimler. R.J., and Kwolek, W.F. 1970. Amylose determination in dimethyl sulfoxide extracts of maize. Cereal Chem. 47: 437.

Wurzburg, O.B. 1964. Acetylation of starches. In Methods in Car- bohydrate Chemistry: Starch, Vol. 4, Ed. Whistler, R.L., p. 286. Academic Press, New York.

MS received 6/26/80; revised E/16/80; accepted E/24/80.

Presented at the 40th Annual Meeting of the Institute of Food Technologists, New Orleans, La., June E-11.1980.

Contribution No. 2578 from the Utah Agriculture Experiment Station and a contribution of Western Regional Project W-150.

We thank Professor J.G. Ponte Jr. and Mr. P.D. Rangnekar. Dept. of Grain Science & Industry, Kansas State Univ., Manhattan, KS 66506, for their help in viscoamylographic studies.

Volume 46 /1981)-JOURNAL OF FOOD SCIENCE-621