Food Chemistry 179 (2015) 85–93

Contents lists available at ScienceDirect

Food Chemistry

journal homepage: www.elsevier .com/locate / foodchem

Quality assessment of noodles made from blends of rice flour and cannastarch

http://dx.doi.org/10.1016/j.foodchem.2015.01.1190308-8146/� 2015 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel.: +66 2 470 7754; fax: +66 2 452 3479.E-mail address: dudsadee.utt@kmutt.ac.th (D. Uttapap).

Yuree Wandee a, Dudsadee Uttapap a,⇑, Santhanee Puncha-arnon a, Chureerat Puttanlek b,Vilai Rungsardthong c, Nuanchawee Wetprasit d

a Division of Biochemical Technology, School of Bioresources and Technology, King Mongkut’s University of Technology Thonburi, Bangkhuntien, Bangkok 10150, Thailandb Department of Biotechnology, Faculty of Engineering and Industrial Technology, Silpakorn University, Nakhon Pathom 73000, Thailandc Department of Agro-Industrial, Food, and Environmental Technology, Faculty of Applied Science, King Mongkut’s University of Technology North Bangkok, Bangkok 10800, Thailandd Department of Biotechnology, Faculty of Science, Ramkhamhaeng University, Bangkok 10240, Thailand

a r t i c l e i n f o a b s t r a c t

Article history:Received 18 November 2014Received in revised form 22 January 2015Accepted 24 January 2015Available online 31 January 2015

Keywords:Rice noodlesCanna starchDietary fiberShort-chain fatty acidsButyric acid

Canna starch and its derivatives (retrograded, retrograded debranched, and cross-linked) were evaluatedfor their suitability to be used as prebiotic sources in a rice noodle product. Twenty percent of the riceflour was replaced with these tested starches, and the noodles obtained were analyzed for morphology,cooking qualities, textural properties, and capability of producing short-chain fatty acids (SCFAs). Cross-linked canna starch could increase tensile strength and elongation of rice noodles. Total dietary fiber(TDF) content of noodles made from rice flour was 3.0% and increased to 5.1% and 7.3% when rice flourwas replaced with retrograded and retrograded debranched starches, respectively. Cooking qualitiesand textural properties of noodles containing 20% retrograded debranched starch were mostly compara-ble, while the capability of producing SCFAs and butyric acid was superior to the control rice noodles; thecooked noodle strips also showed fewer tendencies to stick together.

� 2015 Elsevier Ltd. All rights reserved.

1. Introduction

The rice noodle—produced from rice flour or rice flour mixedwith other components, such as cassava starch, modified starchor hydrocolloids—is one of the most popular varieties of Asian noo-dles, and is widely consumed throughout Southeast Asia(Bhattacharya, Zee, & Corke, 1999; Hormdok & Noomhorm,2007). Rice noodles are high in carbohydrates and calories butlow in dietary fiber (DF) and resistant starch (RS) (Puwastien,Raroengwichit, Sungpuag, & Judprasong, 1999). Presently, consum-ers are more concerned with the health effects of DF as well as RSin carbohydrate-rich foods. Accordingly, various aspects related toDF/RS – for example, potential sources, digestion and fermentation,physiological effects, qualities of food products, acceptability byconsumers, etc. – have been extensively researched.

A number of studies related to noodle qualities have investi-gated the potential of adding fiber sources to noodles made fromwheat. However, much less information is available regarding ricenoodles, perhaps due to the more severe effect of DF on their tex-tural qualities. According to the report of Srikaeo, Mingyai, andSopade (2011), noodles made from rice flour replaced with 20%

unripe banana flour, canna flour or commercial modified cornstarch had significantly higher RS content (2.5%, 3.6% and 8.8%,respectively) than noodles made from rice flour only (1.0%).Recently, Wandee et al. (2014) showed that rice noodles incorpo-rated with 15% cassava pulp and 5% pomelo peel contained muchhigher total dietary fiber (TDF) content (14.4%) than the control(3.0%), while their textural properties were comparable. However,there have been no reports on the physiological effects and fer-mentability of rice noodles enriched with DF/RS, either in vivo orin vitro studies.

RS is the total amount of starch and the products of starch deg-radation that are not digested in the small intestine and pass intothe colon, similar to dietary fiber (Englyst, Kingman, & Cummings,1992; Topping & Clifton, 2001). RS is fermented by colonic micro-flora, producing short-chain fatty acids (SCFAs) and gas (H2, CO2

and CH4). The fermentation rate and relative molar ratio of SCFAsare dependent on the amount and type of RS (Annison &Topping, 1994). SCFAs – mainly acetic, propionic and butyric acids– are absorbed and metabolized in various organs, leading to dif-ferent physiological effects. Butyric acid is completely metabolizedin the colonic epithelial cells, and therefore has been shown to playan important role in the maintenance of colonic health (Topping &Clifton, 2001). In vitro studies as well as animal studies indicatethat butyric acid has the potential to reduce risk factors that are

86 Y. Wandee et al. / Food Chemistry 179 (2015) 85–93

involved in the development of colorectal cancer (inhibiting prolif-eration while increasing differentiation and apoptosis) (Brouns,Kettlitz, & Arrigoni, 2002).

Canna starch, a kind of starch extracted from rhizomes of theedible canna plant (Canna edulis Ker.), is mostly used for preparingtransparent starch noodles, a traditional food of Southeast Asia.High resistance of canna starch granules to enzyme hydrolysis hasbeen reported by Hung and Morita (2005), Srichuwong, Sunarti,Mishima, Isono, and Hisamatsu (2005), and Puncha-arnon,Puttanlek, Rungsardthong, Pathipanawat, and Uttapap (2007).Canna starch and its derivatives have been reported to contain a sig-nificant amount of RS. Native, acetylated, hydroxypropylated,octenyl succinylated, and cross-linked canna starches gelatinizedat 100 �C for 10 min were found to contain 20.8%, 33.8%, 43.5%,51.3% and 35.3% RS, respectively (Juansang, Puttanlek,Rungsardthong, Puncha-arnon, & Uttapap, 2012). Wandee,Puttanlek, Rungsardthong, Puncha-arnon, and Uttapap (2012)prepared retrograded starch (RS type 3) from canna starch bygelatinization and then stored the gels at different times andtemperatures. Under suitable conditions, the thermally stable RSfraction in canna starch could be increased from 1.9% to 16.8%.Bernabé, Srikaeo, and Schlüter (2011) reported that fermentationof raw canna starch with fresh human feces as inoculum producedsignificantly higher total SCFAs and butyric acid compared withbanana, potato, mung bean and taro starches. However, there hasbeen no information on the quality and fermentability of rice noo-dles incorporated with canna starch and its derivatives. Therefore,this study aimed to assess the potential of canna starch and itsderivatives (retrograded, retrograded debranched, and cross-linked) as sources of DF in dried rice noodles.

2. Materials and methods

2.1. Raw materials

Commercial rice flour containing 22% amylose (dry weightbasis; dwb) was purchased from Patum Rice Mill and Granary Pub-lic Co. Ltd., Pathum Thani, Thailand. Eight-month-old rhizomes ofedible canna plants were obtained from the Rayong Field CropsResearch Center, Rayong, Thailand; the starch was isolatedaccording to a procedure described by Puncha-arnon et al.(2007). Amylose content of canna starch determined according tothe method of Jayakody and Hoover (2002) was 23.9% (dwb).Cross-linked canna starch (CL) was prepared following the methodof Emrat (2007), using 0.2% w/w sodium trimetaphosphate as across-linking agent. Retrograded canna starch was prepared byautoclaving starch at 121 �C for 120 min and then storing gel at4 �C for 3 days (Wandee et al., 2012). A similar procedure, exceptthat gelatinized starch was debranched with pullulanase enzyme(64 PUN/g starch) for 24 h prior to storage, was used to obtain ret-rograded debranched canna starch.

2.2. Dried noodle preparation

40 g (dwb) of flour mixes were prepared by mixing rice flour with20% of native, retrograded, retrograded debranched, or cross-linkedcanna starches. Water was then added to each flour mix to obtain aslurry with a concentration of 40% w/v. 30 ml of slurry was spreadevenly on a stainless tray (11.4 � 21.6 cm) and steamed for 1 min.Each noodle sheet was peeled from the tray and dried at 70 �C for15 min. The noodle sheets were stacked, covered with cheeseclothand allowed to rest for 3 h at room temperature, then cut into strips3.0 mm wide. The noodles were further dried in a hot-air oven at40 �C until the moisture content decreased to 10–12%. Dried noodleswere packed in polyethylene bags and kept at room temperature forfurther quality investigation.

2.3. Analyses of noodles

2.3.1. Determination of water absorption indexIn order to obtain information on the ability of each raw mate-

rial to absorb water, single-component flour/starch (100%) wasused to prepare noodles using the procedure described above. Itwas found that noodles could be produced from a slurry of riceflour, native canna starch or cross-linked starch at a concentrationof 40% w/v; however, slurries of retrograded and retrograded deb-ranched starches were too thick, and concentrations of only 15% forretrograded starch and 30% for retrograded debranched starchcould be used for noodle sheet formation. The water absorptionindex of the noodles obtained was determined according to themethod of Anderson, Conway, Pfeifer, and Griffin (1969), with aslight modification. Dried noodles were cut into small pieces (3–5 cm length), ground with a Pulverisette 14 variable-speed rotormill (Fritsch, Idar-Oberstein, Germany) and sieved through a106 lm screen. A noodle powder sample (0.5 g, dwb) was addedto 15 ml of distilled water in a centrifuge tube, then vigorouslymixed with a vortex mixer before placing in a shaker at 30 �C for30 min. After centrifugation at 1127�g for 15 min, the supernatantwas carefully removed and the sediment was weighed.

Water absorption index ðWAI; g=gÞ ¼ wet sediment weightdry sample weight

2.3.2. Cooking quality analysisCooking time of noodles was determined according to the AACC

(1995) method for spaghetti, with a slight modification. Dried ricenoodles (5 g) were cut into 5-cm lengths and cooked in 200 mlboiling distilled water in a covered beaker. Optimum cooking timewas determined by removing a piece of noodle every 30 s andpressing the cooked noodle between two glass slides until thewhite, hard core of the noodle strand disappeared. At least fivemeasurements were performed for each sample.

Cooking weight and cooking loss of starch noodles were mea-sured according to the AACC method (1995), with a slight modifi-cation. At least five replications were done for each measurement.Dried rice noodles (1.0 g) were cut into small pieces (3–5 cm inlength) and boiled in 30 ml water until completely cooked. Thecooked noodles were then filtered through a nylon screen, rinsedwith distilled water, drained for 1 min, and immediately weighed.Cooking weight was determined from the difference between noo-dle weights before and after cooking, and expressed as the percent-age of g cooked noodle/g dried noodle. Cooking loss wasdetermined by evaporating to dryness the cooking water and rinsewater in a pre-weighed glass beaker in a hot-air oven at 105 �C, andwas expressed as the percentage of solid loss during cooking.

2.3.3. Textural profile analysisThe texture of a 10-cm length of cooked noodle was measured

using a texture analyzer (EZTest EZ-S-50N; Shimadzu, Tokyo,Japan) equipped with a pair of noodle elongation jigs (No. 17; Shi-madzu). A 15 N load cell was applied to measure the tensilestrength of noodles at an elongation speed of 60 mm/min. The ini-tial distance between clamps was set at 10.0 cm. From the force–displacement curve (mm), measurements of tensile stress (N/mm2; Pa) and elongation (%) were generated using the textureanalysis software (Trapezium 2 version 2.24). At least 15 strandsof noodles were measured for each sample.

2.3.4. Total dietary fiber (TDF) analysisTDF content of rice noodles was measured using a TDF assay kit

(Megazyme International Ireland, Wicklow, Ireland), followingAOAC method 985.29 (AOAC, 2000). Dried noodles were cut into

Rice flour Canna starch

Retrograded starch Retro-debranched starch

20% Retrograded debranched starch noodle

20% Cross-linked starch noodle

20% Retrograded starch noodle

20% Native canna starch noodle

100% Rice flour noodle

Fig. 1. Morphologies of raw materials and upper surface of noodles made from rice flour and rice flour substituted with 20% native, cross-linked, retrograded, and retrogradeddebranched canna starches.

Y. Wandee et al. / Food Chemistry 179 (2015) 85–93 87

small pieces (3–5 cm length), ground with a Pulverisette 14 vari-able-speed rotor mill (Fritsch, Germany) and sieved through a106 lm screen prior to analysis. Samples were gelatinized with aheat-stable a-amylase (pH 6, 100 �C, 30 min) and then enzymati-cally digested sequentially with protease (pH 7.5, 60 �C, 30 min)and amyloglucosidase (pH 4.5, 60 �C, 30 min) to remove proteinand starch. TDF was precipitated with ethanol, and after washingand drying, the residue was weighed.

TDF ð%Þ¼ sample residue�protein from residue�ash from residue�blanksample weight

�100

2.4. In vitro fermentation

2.4.1. Pre-digestion of noodle samplesPrior to in vitro fermentation, all noodle samples were digested

by in vitro enzymatic digestion according to the method of Englystet al. (1992), with modifications. Briefly, 10 g of noodle powderwas added to 200 ml water in an Erlenmeyer flask. The suspensionwas heated at 80 �C for 5 min and then placed in a water bath at37 �C for 10 min to equilibrate. Sodium acetate buffer (0.1 M, pH5.2, containing 4 mM CaCl2) was added and the mixture wasshaken well by hand. Alpha-amylase (9000 U/g starch; SigmaA-3173) and amyloglucosidase (75 U/g starch; Sigma P7545) were

then added and the flask was incubated at 37 �C for 2 h in a shakingwater bath. Undigested residue was recovered, washed twice withdistilled water and freeze-dried. Duplicate dried samples werepooled and ground in a mortar, passed through a 106 lm sieve,and used for in vitro fermentation.

2.4.2. Preparation of inoculum and fermentation mediumThe inoculum was prepared from fresh cecum of three healthy

pigs obtained from Fresh Meat Processing Co., Ltd. (Nakhon Pathom,Thailand). The cecal contents were pooled, weighed and mixed withsterile medium in a ratio of 1:1 (w/w). The mixture was homoge-nized in a household blender for 1 min and strained through fourlayers of cheesecloth. The fermentation medium was composed of2.5 g trypticase peptone, 125 ll micro-mineral solution (132 g/LCaCl2�2H2O, 100 g/L MnCl2�4H2O, 80 g/L FeCl2�6H2O and 10 g/LCoCl2�6H2O), 25 ml buffer solution (4 g/L (NH4)HCO3 and 35 g/LNaHCO3), 125 ml macro-mineral solution (5.7 g/L Na2HPO4, 6.2 g/LKH2PO4 and 0.6 g/L Mg4�7H2O), 1.25 ml resazurin solution (0.1% w/v) and 33.5 ml reducing solution (6.25 g/L cysteine hydrochloride,6.25 g/L Na2S�9H2O and 40 ml 1 M NaOH) in 1 L of medium (pH 7.2).

2.4.3. In vitro fermentationIn vitro batch fermentation was conducted according to the

method of Lebet, Arrigoni, and Amadò (1998), with a few modifica-tions. Fermentation was performed in triplicate for each noodlesample. One hundred mg of pre-digested noodles was added to

88 Y. Wandee et al. / Food Chemistry 179 (2015) 85–93

8 ml of fermentation medium in a 20 ml serum bottle. The bottlewas sealed with a butyl rubber stopper and aluminum cap andthe headspace was flushed with N2 for 3 min to maintain anaerobicconditions. The sample was then hydrated overnight at 4 �C. Afterequilibrating in a water bath at 37 �C for 1 h, 2 ml of inoculum wasadded to each bottle and the headspace was flushed again with N2

for 1 min. The bottles were incubated in a shaking water bath (50strokes/min) at 37 �C. Samples of the fermented broth (0.5 ml)were taken at 24, 36 and 72 h and immediately placed in a freezerat�20 �C to stop fermentation. A control containing no sample wasused as a blank; inulin, which is a completely fermentable sub-strate, was used as a reference.

2.4.4. Analysis of short-chain fatty acidsSCAF analysis was carried out by high performance liquid chro-

matography (HPLC). The frozen fermentation broth was rapidlythawed in warm water, centrifuged at 12,522�g at 4 �C for30 min and filtered using a 0.45 lm nylon syringe filter. 20 ll ofsample was injected into a Shimadzu HPLC system consisting ofa LC-20AD pump, RID-10A refractive index detector, VertiSep OA8 lm HPLC column (7.8 � 300 mm), and a computer with a dataanalysis software program (CLASS-VP). The sample was analyzedin isocratic mode using 0.005 N sulfuric acid as a mobile phase ata flow rate of 0.8 ml/min. The column temperature was steadilymaintained at 50 �C. Acetic, propionic and butyric acids were usedas external standards.

2.5. Statistical analysis

Analysis of variance (ANOVA) was performed using Duncan’smultiple range test to compare treatment means at p < 0.05. Ifnot specified, all tests were carried out with three replications.

3. Results and discussion

3.1. Water absorption of single-component noodles

Water absorption capacity has a major impact on cooking qual-ities and textural properties of noodles. The water absorption index(WAI) of noodles made from single-component rice flour was 7.3 g/g. Noodles made from single-component cross-linked canna starchexhibited the highest WAI (8.2 g/g), followed by those from nativecanna starch (6.8 g/g), retrograded canna starch (5.4 g/g) and retro-graded debranched canna starch (2.8 g/g), respectively. The highWAI of cross-linked starch noodles was attributed to reinforce-ment of intermolecular bonding of starch molecules inside starchgranules by cross-linking with phosphate ester bonds; hence, thegranules had greater ability to swell. On the other hand, the lowability to absorb water of noodles made from retrograded deb-ranched starch was likely due to the highly ordered structure thatoccurs during incubation of debranched starch under certain con-ditions. These results suggested that partial replacement of rice

Table 1Cooking qualities, textural properties and TDF content of noodles made from rice flour an

Noodle sample Cooking time (min) Cooking weight (%) Cookin

Rice flour (control) 3.0 132.1b 1.3d

Native canna starch 3.5 157.8a 1.6c

Cross-linked starch 3.5 155.0a 1.5c

Retrograded starch 3.5 156.1a 1.8b

Retrograded debranched starch 3.5 139.9b 2.7a

Values with different superscripts in a column differ significantly (p < 0.05).Values of cooking qualities and TDF are the mean of triplicate determinations, while th

flour with cross-linked canna starch or retrograded debranchedcanna starch would have a significant impact on cooking qualitiesand textural properties of rice noodles.

3.2. Morphology of dried noodles

Fig. 1 shows the morphologies of raw starches and the uppersurface of dried noodles made from rice flour and rice flour with20% canna starch or its derivatives, as observed by light micros-copy. Rice starch granules had a much smaller size (�2–10 lm)as compared with the canna starch granules (�10–100 lm), andsome of them were clumped together into small lumps. The sizeand shape of cross-linked canna starch granules were identical tothose of the native starch (figure not shown). Retrograded and ret-rograded debranched starches exhibited a non-granular structurewith irregular shapes and rough surface. The particles ofretrograded starch had relatively larger size as compared with ret-rograded debranched starch.

As shown in Fig. 1, distinctive surface morphologies wereobserved among the different noodles. One hundred percent riceflour noodles had a rough surface, with some bubbles distributedthroughout. A similar morphology was found for noodles madefrom rice flour incorporated with retrograded debranched starch,but with some different features in that the noodles containing ret-rograded debranched starch displayed a rougher surface, muchfewer bubbles and numerous small pores. Noodles made from riceflour with native and cross-linked canna starches had a number ofswollen rice starch granules embedded in a smooth surface; cannastarch granules were not observed. The surface of noodles madefrom rice flour with retrograded starch was rougher than thosecontaining native and cross-linked canna starches, but smootherthan noodles with retrograded debranched starch; relatively largepores were also found on the surface. These appearances arerelated to the composition as well as the physicochemical proper-ties of the individual starches.

The rough surface of rice flour noodles was attributed to theswollen, unbroken rice starch granules protruding from thesmooth matrix of completely gelatinized starch granules. Rice flourhad high gelatinization temperature (73.2 �C, as determined by adifferential scanning calorimeter) (Puncha-arnon & Uttapap,2013) and high pasting temperature (93.5 �C, as determined by arapid visco analyzer) (Wandee et al., 2014); therefore, some gran-ules still remained in granular form after steaming. The roughersurface of noodles when 20% of rice flour was replaced withretrograded debranched starch was likely due to the low abilityof retrograded debranched starch to absorb water. Less moist ret-rograded debranched starch particles would impede heat transferto the rice starch granules, and hence reduce the extent of ricestarch gelatinization. Also, small pores that appeared on the noodlesurface were possibly caused by a difference in water-holdingcapacity of the two components in the steamed noodle sheet. Poreswere generated in the high water holding areas when the noodle

d rice flour substituted with 20% canna starch and its derivatives.

g loss (%) Tensile strength (mN) Elongation (%) TDF (%, dwb)

Experiment Calculation

173.9c 80.1b 3.0d 3.8213.8b 81.9b 4.0c 2.6242.5a 110.0a 3.9c 2.5161.1c 84.3b 5.1b 3.9166.5c 64.3c 7.3a 5.2

e values of textural properties are the mean of ten determinations.

Fig. 2. Concentrations (mmol/L) of acetic acid ( ), propionic acid ( ) and butyricacid ( ) produced from 100 mg of inulin and indigestible residues of various ricenoodles after 24 h (a), 48 h (b) and 72 h (c) fermentation.

Y. Wandee et al. / Food Chemistry 179 (2015) 85–93 89

sheet was dried. The smoother surface of noodles containing retro-graded starch, as compared with noodles made from pure riceflour, indicated that retrograded starch facilitated the gelatiniza-tion of rice starch. Retrograded starch is formed by incubation ofgelatinized starch under specified conditions. During this process,the intact amylose and amylopectin molecules can re-associateby H-bond formation. However, the molecular association wasnot as strong as in the case of retrograded debranched starch,due to the highly branched nature of the starch molecules; there-fore, it can be more easily gelatinized by steaming and thus pro-mote the gelatinization of surrounding rice starch granules.Native and cross-linked canna starch also accelerated rice starchgelatinization because the gelatinization temperatures of native(70.4 �C) and cross-linked canna starch (69.9 �C) (Emrat, 2007)were lower than that of rice flour.

3.3. Cooking qualities

Cooking qualities of noodles made from rice flour and rice flourincorporated with 20% of canna starch and its derivatives are sum-marized in Table 1. Cooking time, cooking weight and cooking lossof noodles made from rice flour were 3.0 min, 132.1% and 1.3%,respectively. Noodles made from rice flour replaced with cannastarch and its derivatives had slightly longer cooking time(3.5 min) and higher cooking loss (1.5–2.7%). The increase in cook-ing loss of noodles containing retrograded and retrograded deb-ranched starches was due to the heterogeneous nature of themixing components. Although the noodles incorporated withcanna starch and its derivatives displayed statistically higher cook-ing loss values than the control, the magnitude of difference (lessthan 1.4%) was negligible in practical terms. According to the Chi-nese and Thai standards for starch noodles, cooking loss should beless than 10% and 9%, respectively (Lii & Chang, 1981; Sisawad &Chatket, 1989). Except for the noodles incorporated with retro-graded debranched starch, cooking weights of noodles with cannastarch and the other two derivatives (155.0–157.8%) were signifi-cantly higher than the control. Cooking weight values of the noo-dles were consistent with the morphologies of the noodles asshown in Fig. 1, i.e., noodles with a higher degree of gelatinizationcould absorb more water. An increase in cooking weight of noodlesprepared from a blend of rice flour and canna starch (80:20) hasalso been reported (Qazi, Rakshit, Tran, Ullah, and Khan (2014).Noodles incorporated with retrograded debranched starch hadcomparable cooking weight to rice flour noodles. This result dif-fered from studies by Aravind, Sissons, Fellows, Blazek, andGilbert (2013) and Sozer, Dalgıç, and Kaya (2007) in which pastareplaced with 20% of commercial resistant starch type 3 (Novelose330™) and spaghetti enriched with 10% of resistant starch type 3had higher cooking weight than the control.

3.4. Textural properties

As shown in Table 1, tensile strength and elongation values ofrice noodles were 173.9 mN and 80.1%, respectively. Tensilestrengths of noodles replaced with native canna and cross-linkedstarches (213.8 and 242.5 mN, respectively) were significantlyhigher, whereas those of noodles with retrograded and retrogradeddebranched starch (161.1 and 166.5 mN, respectively) were com-parable to the control. Elongation values of noodles supplementedwith canna starch and its derivatives were inconsistent. Noodlesmade from rice flour, rice flour with native canna starch and riceflour with retrograded starch had similar elongation values(80.1%, 81.9% and 84.3%); noodles containing retrograded deb-ranched starch had a significantly lower elongation value (64.3%),while noodles with cross-linked starch (110.0%) displayed muchhigher elongation. The effect of the second starch component onthe textural properties of rice noodles was quite complicated, sinceit would depend on several factors such as the pasting behavior ofeach starch, interaction and compatibility of the two starches,water absorption and retrogradation abilities of each starch, etc.The highest tensile strength and elongation was found in rice noo-dles containing cross-linked canna starch; this was most likely dueto the high degree of gelatinization of both starches, resulting in ahomogeneous mixture of disrupted starch granules stabilized bycross-linked covalent bonding. Hydrogen bonding between starchmolecules formed during incubation of noodle sheets could alsoprovide a gel network that strengthened the structure of the noo-dles. On the other hand, the heterogeneous structure of noodlescontaining retrograded debranched starch – due to incomplete dis-ruption of starch granules, many small pores, and less associationof leached amylose – would contribute to their having the lowesttensile strength among the tested noodles.

Table 2Amounts of indigestible residue (g/50 g noodles), total SCFAs and butyric acid (mmol/50 g dried noodles) in noodle products.

Sample Indigestible residue (g/50 g noodle) Total SCFA (mmol/50 g noodle) Butyric acid (mmol/50 g noodle)

24 h 48 h 72 h 24 h 48 h 72 h

100% Rice flour 6.8 ± 0.5 12.5d 15.3e 16.4d 2.5e 3.3e 3.4d

20% Native canna starch 7.6 ± 0.3 14.4c 17.8d 19.9c 3.1d 4.1d 4.4c

20% Cross-linked starch 8.1 ± 0.1 15.7b 19.8b 21.6b 3.6b 4.6b 4.8b

20% Retrograded starch 7.8 ± 1.5 14.7c 19.1bc 21.6b 3.4c 4.4b 4.7b

20% Retrograded debranched starch 10.4 ± 0.2 17.0a 25.6a 28.2a 3.8a 7.4a 7.8a

Values with different superscripts in a column differ significantly (p < 0.05).Values of total SCFAs and butyric acid are the mean of triplicate determinations, while the values of indigestible residue are the mean of duplicate determinations.

90 Y. Wandee et al. / Food Chemistry 179 (2015) 85–93

3.5. Total dietary fiber content

Total dietary fiber (TDF) contents of uncooked rice noodlesmade from rice flour and rice flour incorporated with canna starchand its derivatives are shown in Table 1. Incorporation of cannastarch and its derivatives resulted in a significant increase of TDFcontent, from 3.0% to 7.3%. The highest TDF content was found innoodles containing retrograded debranched starch (7.3%), followedby retrograded starch (5.1%), while TDF contents of noodles con-taining canna starch or cross-linked starch were comparable(4.0%, 3.9%). In Table 1, the TDF values calculated from the fibercontents in raw materials are also given in parentheses. The exper-imental TDF value of noodles made from rice flour was slightlylower, while those of noodles with canna starch and its derivativeswere significantly higher than the calculated values. This indicatedthat noodle processing could increase or decrease TDF, dependingon the raw material source. The increase of TDF in noodles supple-mented with canna starch and its derivatives was most likely dueto the high retrogradability of canna starch. High retrogradation ofcanna starch is thought to be due to the combined effect of the fol-lowing factors: considerably high amylose content (about 30%;Puncha-arnon et al., 2007); small size of amylose molecules(�1600 dp), with low value of the average number of branchchains; and high value of the average chain length of amylopectin(Thitipraphunkul, Uttapap, Piyachomkwan, & Takeda, 2003). Theresults were opposite to our previous study on noodles incorpo-rated with cassava pulp and pomelo peel (Wandee et al., 2014),in which the experimental values were lower than theircorresponding calculated values. In those cases, part of TDF mightbe heat-unstable and could be destroyed by noodle processing,especially during the steaming step.

3.6. In vitro fermentability of noodles

Noodle products were subjected successively to in vitro diges-tion with a-amylase and amyloglucosidase, and the indigestibleresidues were recovered and subsequently fermented by anin vitro batch system using pig cecal content as inoculum. Concen-trations of SCFAs produced from 100 mg of indigestible residue ofinulin, rice noodles and rice noodles containing canna starch andits derivatives after 24, 48 and 72 h fermentation are shown inFig. 2. Inulin is a long-chain prebiotic consisting of a linear seriesof ß-(2 ? 1) fructose units, and typically has a terminal non-reduc-ing glucose (GFn). According to the report of Roberfroid (2004),inulin is slowly but completely fermented, so it was used as a ref-erence in this study. As shown in Fig. 2a, fermentation of inulin for24 h produced the lowest SCFAs (9.6 mmol/L), followed by noodlescontaining retrograded debranched starch (16.3 mmol/L), whilefermentation of the rest produced comparable amounts of totalSCFAs (18.5–19.4 mmol/L). Slower fermentation of noodles con-taining retrograded debranched starch, as compared with othercanna starch samples, was due to the highly ordered structure ofretrograded debranched starch.

Extending the fermentation time to 48 h resulted in a signifi-cant increase in total SCFAs of inulin and noodles containing retro-graded debranched starch (Fig. 2b). Noodles incorporated with allderivatives of canna starches produced significantly higheramounts of butyric acid (5.6–7.2 mmol/L) as compared with ricenoodles (4.9 mmol/L) and noodles containing native canna starch(5.3 mmol/L) (p < 0.05); noodles with retrograded debranchedstarch had the highest amount of butyric acid (7.2 mmol/L). Thiscircumstance was more pronounced when the fermentation timewas extended to 72 h (Fig. 2c). The high concentration of totalSCFAs was caused by high production of acetic acid. At this fermen-tation period, rice noodles produced the lowest amount of totalSCFAs when compared with other substrates (p < 0.05). Total SCFAsfrom inulin still increased concurrently with increasing fermenta-tion time, and propionic acid was found to be a major component.Fermentation of noodles containing retrograded debranched starchproduced the highest amount of butyric acid (7.5 mmol/L),followed by inulin (6.2 mmol/L) and noodles incorporated withretrograded starch (6.0 mmol/L), respectively. After 72 h of fer-mentation, the molar ratio of acetic, propionic and butyric acidsproduced from noodles containing retrograded debranched starchwas 41:31:28. These results were in agreement with a previousreport, in that RS fermentation generally results in relatively higherbutyric acid production, on the order of 20–28 mol%, comparedwith about 10–15 mol% for non-starch polysaccharides (Brounset al., 2002). Therefore, in terms of fermentation products, retro-graded debranched starch is a promising prebiotic source since itcan produce high levels of SCFAs as well as high butyric acid.

In humans, the highest fermentation activity is found in theproximal colon, and declines farther down the gastrointestinaltract as the availability of substrates decreases (Topping &Clifton, 2001). Therefore, the distal colon is the site with the mostlimited carbohydrate sources of carbon and energy for bacterialgrowth. This results in a decrease in SCFAs and an increase in unde-sirable (even toxic) compounds, such as phenol and NH3, in distalregions of the colon, resulting in a less healthy colonic environ-ment (Macfarlane, Gibson, & Cummings, 1992). An easily ferment-able substrate might be depleted rapidly at the proximal colon,whereas a substrate that is difficult to ferment could be excretedwith feces; thus, substrates with an appropriate fermentation rateare preferable. In terms of fermentation rate, inulin would appearto be ideal since its fermentation was the lowest on the first dayand increased continuously during the second and third days offermentation. On the other hand, fermentation of the other sub-strates, except the residue of noodles containing retrograded deb-ranched starch, reached nearly maximum values after the first day.Therefore, from the perspectives of both fermentation productsand rate of fermentation, retrograded debranched starch is proba-bly the most promising source of prebiotics. The fermentation rateof this substrate could be adjusted to a slower or faster rate byaltering the production conditions, such as debranching level,incubation time and temperature, number of incubation cycles,drying rate and temperature, etc.

Table 3Cooking qualities, textural properties and TDF content of noodles made from rice flour and rice flour substituted with retrograded debranched starch at various levels.

Noodle sample Cookingtime (min)

Cookingweight (%)

Cookingloss (%)

Tensile strength(mN)

Elongation(%)

TDF (%, dwb)

Experiment Calculation

Rice flour 3.0 132.1ab 1.3f 173.9a 80.1a 3.0e 3.820% Retrograded debranched starch 3.5 139.9a 2.7e 166.5a 64.3b 7.3d 5.225% Retrograded debranched starch 4.0 140.3a 3.0d 137.8b 55.2c 8.0c 5.730% Retrograded debranched starch 4.0 139.5a 3.3c 134.6b 49.6 c 9.4b 6.235% Retrograded debranched starch 4.0 136.7ab 4.1b 131.0b 41.0d 9.8b 6.840% Retrograded debranched starch 4.0 124.1b 4.4a 131.0b 39.2d 10.6a 7.3

Values with different superscripts in a column differ significantly (p < 0.05).Values of cooking qualities and TDF are the mean of triplicate determinations, while the values of textural qualities are the mean of ten determinations.

Fig. 3. Appearances of cooked rice noodles made from rice flour and rice flour substituted with retrograded debranched canna starch at various levels, after standing for 0, 10,20, 30 and 60 min.

Y. Wandee et al. / Food Chemistry 179 (2015) 85–93 91

92 Y. Wandee et al. / Food Chemistry 179 (2015) 85–93

3.7. Indigestible residues and SCFA production of noodles

A survey on dietary intake of Thai adolescents in Bangkok(Uthang, 1990) revealed that the average intakes of dietary fiberwere 7.32 g/day in men and 8.88 g/day in women. Many countrieshave recommended daily consumption of about 25–30 g of fiber inorder to keep the bowels healthy. Table 2 shows the amount ofindigestible residue, total SCFAs and butyric acid in one serving(50 g) of noodle products. The amount of indigestible residue ofrice noodles, determined by in vitro-simulated upper intestinaltract digestion, was 6.8 g/50 g dried noodles; while for noodlesincorporated with canna starch and its derivatives, the amountswere between 7.6 and 10.4 g/50 g dried noodles. These indigestibleresidues, consisting mainly of resistant starch and fiber, wereassumed to be the substrate that passed into the colon. Thus, theamount of SCFAs produced would depend not only on the ferment-ability of an individual substrate but also its resistance to digestionin the small intestine. On the basis of one serving size, noodles con-taining canna starch and its derivatives produced higher amountsof total SCFAs and butyric acid than rice flour noodles. Noodleswith retrograded debranched starch produced the highest amountsof total SCFAs and butyric acid. These results confirmed the poten-tial of retrograded debranched starch as a prebiotic source.

3.8. Qualities of noodles incorporated with higher levels of retrogradeddebranched starch

The results shown above revealed that incorporation of 20% ret-rograded debranched canna starch could increase TDF content ofnoodles and improve the nutritional benefits, in terms of increas-ing fermentability and butyric acid production, without adverselyaffecting the cooking qualities. It was also observed that cookednoodle strips made from only rice flour tended to stick togetherwhen kept for a long period, while noodles containing retrogradeddebranched starch did not. Therefore, it was presumed that addi-tion of a higher level of retrograded debranched starch wouldnot only increase the fiber content but also improve the qualityand texture of cooked noodles when keeping after cooking. Accord-ingly, replacement of retrograded debranched starch in noodles at25%, 30%, 35% and 40% was studied. The results of cooking quality,textural properties and TDF content of the noodles obtained areshown in Table 3. Cooking time of noodles with higher levels of ret-rograded debranched starch was extended to 4.0 min, while cook-ing weight was comparable to the control. Cooking loss wassignificantly increased with an increased amount of retrogradeddebranched starch; however, the cooking loss of noodles contain-ing the highest level (40%) of retrograded debranched starch wasstill less than the limit specified by Thai standards for starchnoodles. The results of textural properties of noodles clearlyrevealed that increasing levels of retrograded debranched starchhad a significantly negative effect on strength and elongation ofthe noodles.

Appearances of cooked noodles made from retrograded deb-ranched starch at various levels during keeping at 25 �C for up to60 min are shown in Fig. 3. When left to stand for 10 min, cookedrice flour noodles stuck together, and clumped into a lump of noo-dle strips when allowed to stand longer (30–60 min). Strips of noo-dles enriched with 20% retrograded debranched starch remainedseparate for up to 30 min but tended to stick together after that.Noodles containing higher levels of retrograded debranched starchdid not stick together even after 60 min. The most important char-acteristics for cooked starch noodles are texture and mouth feel;they should remain firm, not sticky, after cooking, and exhibit hightensile strength, short cooking time and low cooking loss (Tan, Li, &Tan, 2009). Although the addition of retrograded debranchedstarch at levels higher than 20% possessed some benefits in terms

of TDF content and continued good appearance of noodles afterstanding, the detrimental effect on cooking and textural qualitieswas found to be unacceptable. Therefore, weighing the meritsand demerits of these attributes, substitution of rice flour with20% retrograded debranched starch would be the most suitablelevel for noodle production.

4. Conclusion

This study demonstrated the potential of using canna starchand its derivatives to improve the qualities of rice noodles. Cross-linked canna starch can increase tensile strength and elongation,while retrograded debranched canna starch can increase TDF con-tent and improve nutritional benefits in terms of increased butyricacid production. In vitro experiments revealed that noodles substi-tuted with retrograded debranched canna starch exhibited a slowfermentation rate; therefore, these noodles are also expected tobe slowly fermentable in vivo and will yield a certain amount ofSCFAs along the distal colon. Due to the high TDF content, accept-able cooking qualities and textural properties, as well as good fer-mentability, rice noodles substituted with 20% retrogradeddebranched starch are recommended as an alternative food choicefor health-conscious consumers.

Acknowledgements

The authors gratefully acknowledge financial support from: theThailand Research Fund, via the Royal Golden Jubilee Ph.D. Pro-gram (for Miss Yuree Wandee); and the National Research Univer-sities Project and Research Promotion in Higher Education, Officeof the Higher Education Commission.

References

AACC (1995). Approved methods of the American Association of Cereal Chemists (9thed.). St. Paul, MN: American Association of Cereal Chemists.

Anderson, R. A., Conway, H. F., Pfeifer, V. F., & Griffin, E. L. (1969). Gelatinization ofcorn grits by roll and extrusion cooking. Cereal Science Today, 14, 4–12.

Annison, G., & Topping, D. L. (1994). Nutritional role of resistant starch: Chemicalstructure vs physiological function. Annual Review Nutrition, 14, 297–320.

Aravind, N., Sissons, M., Fellows, C. M., Blazek, J., & Gilbert, E. P. (2013). Optimisationof resistant starch II and III levels in durum wheat pasta to reduce in vitrodigestibility while maintaining processing and sensory characteristics. FoodChemistry, 136, 1100–1109.

Bernabé, A. M., Srikaeo, K., & Schlüter, M. (2011). Resistant starch content, starchdigestibility and the fermentation of some tropical starches in vitro. FoodDigestion, 37–42.

Bhattacharya, M., Zee, S. Y., & Corke, H. (1999). Physicochemical properties relatedto quality of rice noodles. Cereal Chemistry, 76, 861–867.

Brouns, F., Kettlitz, B., & Arrigoni, E. (2002). Resistant starch and the butyraterevolution. Trends in Food Science & Technology, 13, 251–261.

Emrat, I. (2007). Modification of canna starch by cross-linking with sodiumtrimetaphosphate. M. Sc. thesis, King Mongkut’s University of TechnologyThonburi, Bangkok, Thailand.

Englyst, H. N., Kingman, S. M., & Cummings, J. H. (1992). Classification andmeasurement of nutritionally important starch fractions. European Journal ofClinical Nutrition, 46, S33–S50.

Hormdok, R., & Noomhorm, A. (2007). Hydrothermal treatments of rice starch forimprovement of rice starch noodle quality. LWT–Food Science and Technology,40, 1723–1731.

Hung, P. V., & Morita, N. (2005). Physicochemical properties and enzymaticdigestibility of starch from edible canna (Canna edulis) grown in Vietnam.Carbohydrate Polymers, 61, 314–321.

Jayakody, L., & Hoover, R. (2002). The effect of lintnerization on cereal starchgranules. Food Research International, 35(7), 665–680.

Juansang, J., Puttanlek, C., Rungsardthong, V., Puncha-arnon, S., & Uttapap, D.(2012). Effect of gelatinisation on slowly digestible starch and resistant starchof heat-moisture treated and chemically modified canna starches. FoodChemistry, 131, 500–507.

Lebet, V., Arrigoni, E., & Amadò, R. (1998). Measurement of fermentation productsand substrate disappearance during incubation of dietary fibre sources withhuman faecal flora. LWT–Food Science and Technology, 31, 473–479.

Lii, C. Y., & Chang, S. M. (1981). Characterization of red bean (Phaseolus radiatus var.aurea) starch and its noodle quality. Journal of Food Science, 46, 78–81.

Y. Wandee et al. / Food Chemistry 179 (2015) 85–93 93

Macfarlane, G. T., Gibson, G. R., & Cummings, J. H. (1992). Comparison offermentation reactions in different regions of the human colon. Journal ofApplied Bacteriology, 72, 57–64.

Puncha-arnon, S., Puttanlek, C., Rungsardthong, V., Pathipanawat, W., & Uttapap, D.(2007). Changes in physicochemical properties and morphology of cannastarches during rhizomal development. Carbohydrate Polymers, 70, 206–217.

Puncha-arnon, S., & Uttapap, D. (2013). Rice starch vs. rice flour: Differences in theirproperties when modified by heat–moisture treatment. Carbohydrate Polymers,91, 85–91.

Puwastien, P., Raroengwichit, M., Sungpuag, P., & Judprasong, K. (1999). Thai foodcomposition tables. Nakhon Pathom, Thailand: Institute of Nutrition MahidolUniversity (INMU).

Qazi, I. M., Rakshit, S. K., Tran, T., Ullah, J., & Khan, M. Z. (2014). Effect of blendingselected tropical starches with rice flour on the cooking quality and texture ofrice based noodles. Sarhad Journal of Agriculture, 30, 257–264.

Roberfroid, M. B. (2004). The digestive functions: inulin and oligofructose as dietaryfiber Inulin-type fructans: functional food ingredients. Boca Raton, FL: CRC Press,pp. 105–134.

Sisawad, S., & Chatket, I. (1989). Transparent noodles from other bean starches. In A.H. Ghee, N. B. Hen, & L. K. Kong (Eds.). Trends in Food Biotechnology(pp. 309–318). Singapore: Proceedings of the 7th World Congress of FoodScience and Technology.

Sozer, N., Dalgıç, A. C., & Kaya, A. (2007). Thermal, textural and cooking properties ofspaghetti enriched with resistant starch. Journal of Food Engineering, 81,476–484.

Srichuwong, S., Sunarti, T. C., Mishima, T., Isono, N., & Hisamatsu, M. (2005).Starches from different botanical sources I: Contribution of amylopectin fine

structure to thermal properties and enzyme digestibility. CarbohydratePolymers, 60, 529–538.

Srikaeo, K., Mingyai, S., & Sopade, P. A. (2011). Physicochemical properties, resistantstarch content and enzymatic digestibility of unripe banana, edible canna, taroflours and their rice noodle products. International Journal of Food Science &Technology, 46, 2111–2117.

Tan, H. Z., Li, Z. G., & Tan, B. (2009). Starch noodles: History, classification, materials,processing, structure, nutrition, quality evaluating and improving. FoodResearch International, 42, 551–576.

Thitipraphunkul, K., Uttapap, D., Piyachomkwan, K. A., & Takeda, Y. (2003). Acomparative study of edible canna (Canna edulis) starch from different cultivars.Part II. Molecular structure of amylose and amylopectin. Carbohydrate Polymers,54, 489–498.

Topping, D. L., & Clifton, P. M. (2001). Short-chain fatty acids and human colonicfunction: Roles of resistant starch and nonstarch polysaccharides. PhysiologicalReviews, 81, 1031–1064.

Uthang, S. (1990). Fiber consumption among the adolescence in Bangkok area. M.Sc. thesis, Kasetsart University, Bangkok, Thailand.

Wandee, Y., Puttanlek, C., Rungsardthong, V., Puncha-arnon, S., & Uttapap, D. (2012).Effects of gelatinization and gel storage conditions on the formation of cannaresistant starch. Food and Bioprocess Technology, 5(7), 2665–2673.

Wandee, Y., Uttapap, D., Puncha-arnon, S., Puttanlek, C., Rungsardthong, V., &Wetprasit, N. (2014). Enrichment of rice noodles with fiber-rich fractionsderived from cassava pulp and pomelo peel. International Journal of Food Science& Technology, 49(11), 2348–2355. http://dx.doi.org/10.1111/ijfs.12554.