Contents lists available at ScienceDirect Food Chemistry journal homepage: www.elsevier.com/locate/foodchem Effect of roasted pea flour/starch and encapsulated pea starch incorporation on the in vitro starch digestibility of pea breads Zhan-Hui Lu, Elizabeth Donner, Qiang Liu⁎ Guelph Research and Development Centre, Agriculture and Agri-Food Canada, 93 Stone Road West, Guelph, Ontario N1G 5C9, Canada A R T I C L E I N F O Keywords: Pea Bread Roasting Encapsulation Starch digestibility A B S T R A C T Oven or microwave roasting and alginate encapsulation of pea flour and starch to produce novel pea ingredients for enrichment of slowly digestible starch (SDS) and resistant starch (RS) content in pea bread were investigated. Pea flour treated either by oven roasting (160 °C, 30 min) or by microwave roasting (1.1 kW, 6 min) effectively retained its low starch digestibility similar to its native form (∼25% SDS; ∼60% RS). When oven roasting was applied to pea starch, SDS content increased triply compared to the fully boiled counterpart. Alginate en- capsulation effectively controlled carbohydrate release to simulated gastric, intestinal and colonic fluids, and thus largely enriched the SDS and RS fractions in starch. Pea bread containing up to 37.5% of encapsulated roasted MPS pea starch not only provided high SDS and RS fractions (23.9% SDS and 30.2% RS) compared to a white bread control (0.2% SDS and 2.5% RS), but also provided an acceptable palatability. 1. Introduction Bread is a staple food which is consumed all over the world. A standard recipe of wheat-based white bread is usually wheat flour (100%), salt (1%), sugar (5%), yeast (3%) and a prescribed amount of water based on flour weight according to the AACC Approved Method 10-09.01 (AACC International, 1999). Wheat flour is very low in total dietary fibre (2.7% dry basis) and resistant starch (RS) (≤1% dry matter) (Goñi, García-Diz, Mañas, & Saura-Calixto, 1996), which makes wheat bread a high glycemic index (GI > 70) food. Legumes are rich in protein, starch, and fibre, and a wide range of bioactive constituents (carotenoids, flavonoids, etc.) that are beneficial to human health when consumed in sufficient quantities (Mathers, 2002). Legumes were identified as low GI foods (GI < 55) more than 20 years ago (Jenkins et al., 1980; Rizkalla, Bellisle, & Slama, 2002). They allow for an increased stability in insulin response due to a de- crease in blood glucose fluctuation corresponding to the slower glucose release (Rizkalla et al., 2002). These properties can help in the pre- vention of obesity and high blood cholesterol, as well as diabetes (Kushi, Meyer, & Jacobs, 1999). The replacement of wheat flour com- pletely by yellow pea (Pisum sativum) flour for bread making would largely promote pea utilization, and also produce a healthy and gluten- free alternative which is in high demand by increasingly diagnosed celiac disease patients, as well as those with wheat allergies (Sciarini, Ribotta, León, & Pérez, 2010) and gluten sensitivity. Although slowly digestible in its native form, up to 94% of isolated pea starch will still be rapidly digestible after it is gelatinized (Chung, Liu, & Hoover, 2009). Thus, the challenge is how to retain or enhance the amount of slowly digestible starch (SDS) and resistant starch (RS) in the finished pea products after processing. Roasting is one of the most important technological operations in processing grain legumes. It modifies and significantly enhances flavor (Ma, Boye, Azarnia, & Simpson, 2016), texture (Kaur, Singh, & Sodhi, 2005), and appearance (Koksel, Sivri, Scanlon, & Bushuk, 1999; Özdemir & Devres, 2000) of the product. Felsman, Harvey, Linnerud, and Smith (1976) reported that enzymatic glucose release values from corn samples roasted at 93, 104 and 116 °C for 1–5 h were similar to those obtained from raw corn. However, few starch digestibility data of roasted pea have been reported. Furthermore, Venkatachalam, Kushnick, Zhang, and Hamaker (2009) reported that starch-entrapped calcium alginate microbeads provide a useful tool to control the glucose release from the beads. If roasting and encapsulation are synergistically applied to pea flour/starch, a new food ingredient with a better flavor and enhanced SDS and RS fractions could be expected, and these physically modified starch products would be favored by consumers rather than their che- mically modified counterpart. However, to our knowledge, there is a dearth of information on starch digestibility of roasted pea flour/starch, and no published reports on incorporating roasted flour/starch and/or encapsulated legume starch in foods, especially in bread making. The overall objective of this study was to research and develop new pea ingredients with enhanced SDS and RS content by roasting and encapsulation, and then to apply the resulting novel ingredients to http://dx.doi.org/10.1016/j.foodchem.2017.10.037 Received 7 June 2017; Received in revised form 28 September 2017; Accepted 9 October 2017 ⁎ Corresponding author. E-mail addresses: zhanhui.lu@agr.gc.ca (Z.-H. Lu), elizabeth.donner@agr.gc.ca (E. Donner), qiang.liu@agr.gc.ca (Q. Liu). Food Chemistry 245 (2018) 71–78 Available online 10 October 2017 0308-8146/ Crown Copyright © 2017 Published by Elsevier Ltd. All rights reserved. T http://www.sciencedirect.com/science/journal/03088146 https://www.elsevier.com/locate/foodchem http://dx.doi.org/10.1016/j.foodchem.2017.10.037 http://dx.doi.org/10.1016/j.foodchem.2017.10.037 mailto:zhanhui.lu@agr.gc.ca mailto:elizabeth.donner@agr.gc.ca mailto:qiang.liu@agr.gc.ca https://doi.org/10.1016/j.foodchem.2017.10.037 http://crossmark.crossref.org/dialog/?doi=10.1016/j.foodchem.2017.10.037&domain=pdf make gluten-free pea bread. Characterization of the physicochemical and nutritional properties of these modified pea ingredients and breads using various analytical techniques was another objective of this pro- ject, to better understand the relationship between structure and functionality. 2. Materials and methods 2.1. Materials Pea seeds (cultivar Agassiz) from the 2015 growing season were provided by the Seed Increase Unit, Agriculture and Agri-Food Canada (Indian Head, Saskatchewan, Canada). The whole pea seeds were first cracked with an IKA mill (M10), and then milled to flour using an Udy cyclone mill (Model 3010-030; Udy Corp., Fort Collins, CO 80524, USA) and passed through a 500 µm sieve. The moisture content of the flour was 9.0% (db) measured by air-oven method (130 °C, 2 h). Two pea starches, NPS (11.1% moisture, db) and MPS (9.8% moisture, db), were purchased from commercial suppliers. According to the manufacturers, the pea starches have a protein content of< 1.0% (Dumas-N × 6.25), a fat content of< 0.1%, an ash content of< 0.2%, a neutral pH value of 10% solution, and a mesh size of> 95% (through 200 μm mesh screen). Alginate (FD155) was provided by Danisco USA Inc. Xanthan gum (Duinkerken), baker’s yeast (Fleischmann’s traditional active dry yeast), and other food ingredients were purchased from a local grocery store. Double-distilled water was used for bread making. 2.2. Physical modification of pea flour/starch to enhance SDS and RS fractions 2.2.1. Oven roasting Pea flour (90 g; 9.0% moisture, db) or commercial pea starch was spread to a depth of 3 mm in a Pyrex petri dish (8 cm in diameter), and roasted at 160 °C for 30 min in an Isotemp forced air oven (Model 750F, Fisher Scientific, Pittsburgh, PA 15275, USA). These roasting para- meters were determined to give a typical roasting aroma and end-use properties (no raw flour/starch taste, not burnt) by preliminary ex- periments. 2.2.2. Microwave roasting A microwave oven (Panasonic, model NN-ST59, Mississauga, Ontario, Canada), capable of generating 1.1 kW power at 2450 MHz, was used. Pea flour or starch (90 g) was spread in a Pyrex petri dish (8 cm in diameter), and roasted at 100% power (1.1 kW) for 2 min. After cooling down to< 50 °C, the contents of the petri dish were mixed homogeneously, and roasted again for another 2 min. The process was repeated for three cycles to avoid burning (total roasting time of 6 min). These parameters were determined by preliminary experiments to yield a similar extent of roasting compared to the air-oven (160 °C, 30 min) method above, determined by color change of the samples using a Lab Scan XE Spectrocolorimeter (Hunter Associates Laboratory, Reston, VA, USA). 2.2.3. Alginate encapsulation The pea starch core material for encapsulation was native, boiled or roasted pea starch. The boiled pea starch was prepared by heating a 10% (w/v) native pea starch dispersion at 100 °C for 20 min on a hot plate with constant stirring at 500 rpm (IKA, Canada), followed by freeze-drying at −40 °C (0.1 mBar; FreeZone 12, Labconco, MO, USA), and manually grinding by mortar and pestle to pass through a 250 μm sieve. Starch-entrapped microbeads were prepared according to Rose et al. (2009) with minor modification. Suspensions of sodium alginate, pea starch, and water were prepared in a ratio of 1:9:90; the alginate was dissolved completely first, and then the starch was added. With con- tinuous stirring, the suspension was pumped through a 100 μL pipette tip using a peristaltic pump into a bath of calcium chloride (2% w/v). The microbeads were kept in the calcium chloride bath for 3 h and then harvested by filtration. After flushing with distilled water for 5 min, the microbeads were stored at −20 °C before bread making, or dried at 40 °C for 24 h in an air-oven for the carbohydrate release study and chemical analysis. The size of microbeads was measured by a digital caliper (SET/0 Pro-Max, Fowler High Precision Inc., Toronto, Canada). 2.3. Gluten-free pea bread making A batter-based process was performed using a KitchenAid mixer (KSM7586PFP, Michigan, USA) with a flat beater (Supplemental Fig. 1). The following ingredients (as % on dry flour/starch basis) were used: salt (2.5%), sugar (7.5%), milk powder (5%), xanthan gum (2.5%), instant dry yeast (2.5%), canola oil (17.5%), vinegar (2.5%), vanilla extract (2.5%), whole egg (12.5%), and water (87.5%). Dry ingredients (flour, starch, sugar, milk powder, salt and xanthan gum) were whisked using the stainless steel wire whip attachment for 5 min at speed “1” (on the scale 1–10 of the mixer). Instant dry yeast and water were manually whisked until completely dispersed, then added to the dry ingredients and mixed for 1 min at speed “1”. The remaining wet in- gredients (canola oil, whisked whole egg, vinegar, and vanilla extract) were incorporated and mixed for another 1 min at speed “1”, and 2 min at speed “2”. The resulting batter was scaled to 280 g into a butter- greased baking tin (5.5 cm height; 8.5 × 15 cm top; 6 × 12.5 cm bottom). The tins with batter were proofed at 43 °C at a medium hu- midity setting (85%) for 35 min and then baked at 163 °C for 35 min in a cabinet proofer/oven (Model PFB-2, Duke Manufacturing Company, St. Louis, MO 63102, USA). After baking, the loaves were de-panned and cooled for 2 h on cooling racks at room temperature, and packed in sealed polyethylene bags to prevent dehydration. Analytical measure- ments were made within 24 h. The above bread making process was also applied to enhance SDS and RS content of breads by replacement of native pea flour with roasted flour/starch, or incorporating microbeads of alginate en- capsulated pea starch (native, boiled or roasted). Seven kinds of pea bread were prepared as below using Agassiz pea flour and NPS. (1) Breads made from 100% Agassiz pea flour (native vs. oven roasted); (2) Breads made from 50% native Agassiz pea flour + 50% NPS (native vs. roasted); and (3) Breads made from native Agassiz pea flour + roasted NPS + en- capsulated roasted NPS at ratios of 50:37.5:12.5, 50:25:25, and 50:12.5:37.5. Another four pea breads were also prepared using MPS, to compare the effects of different pea starches on bread characteristics and in vitro starch digestibility. 2.4. Chemical analysis Native and roasted flour/starch samples were used “as is” for che- mical analysis. Freshly-made microbeads were air-dried at 40 °C over- night, and ground to pass through a 250 μm sieve using an IKA M20 mill (IKA-Werke M20). Bread samples were frozen at −40 °C, freeze- dried and ground by a pestle and mortar to pass through a 500 μm sieve. The total starch, ash, apparent amylose and protein content (based on dry mass) were determined by the AACC Approved Method 76-13.01 and Method 08-01.01 (AACC International, 1999), iodine colorimetry and Dumas method (Flash 2000 analyzer, ThermoFisher Scientific, Waltham, MA, USA), respectively. 2.5. Light microscopy A slice (∼0.2 mm in thickness) from the centre of a freshly-made microbead was placed on a microscope slide and pressed flat with a Z.-H. Lu et al. Food Chemistry 245 (2018) 71–78 72 cover slip, and viewed under brightfield and polarized light using a light microscope (AxioImager A2, Carl Zeiss Canada, Toronto, ON, Canada). Images were taken with an AxioCam MRc5 camera (Zeiss) and Zen lite 2012 software using the auto-exposure function at 100× magnification. 2.6. Carbohydrate release study The in vitro carbohydrate release profiles of starch-entrapped beads were studied by incubating 100 mg of air-dried beads in 10 mL United States Pharmacopeia simulated gastric fluid (SGF, 0.05 M sodium chloride adjusted to pH 1.5 with HCl, without pepsin) in four conical flasks kept in a shaking water bath at 37 °C and 70 rpm, according to Park, Kim, Kim, and Moon (2014). After 4 h of incubation, the beads in each of the conical flasks were collected by filtration using Whatman Grade 42 filter paper, transferred to 10 mL simulated intestinal fluid (SIF, 0.05 M sodium dihydrogen phosphate buffer adjusted to pH 6.8 with NaOH, without pancreatin), and held for 3 h. The medium was then transferred to simulated colonic fluid (SCF, 0.05 M sodium dihy- drogen phosphate buffer adjusted to pH 7.4 with NaOH, no enzymes) for another 3 h. At different time intervals, 1 mL of the solution was withdrawn and replaced by fresh medium. The total carbohydrate re- leased from the beads was determined using a phenol-sulfuric acid assay (Dubois, Gilles, Hamilton, Rebers, & Smith, 1956). 2.7. Evaluation of bread quality 2.7.1. Loaf moisture, specific volume and baking loss Loaves were weighed 2 h after baking, and loaf volume (cm3) was determined by rapeseed displacement method according to the AACC Approved Method 10-05.01 (AACC International, 2001). Specific loaf volume (cm3/g) and baking loss (%, 100 − bread weight/batter weight × 100) were then calculated. Bread moisture was determined according to the AACC Approved Method 44-15.02 (AACC International, 1999). 2.7.2. Palatability The palatability of bread samples was evaluated using a visual analogue scale consisting of a 100 mm line anchored at the left end by “very unpalatable” and at the right end by “very palatable”. Six subjects made a vertical mark along the line to indicate their perceived palat- ability. The distance from the left end of the line to the mark made by the subject is the palatability rating. 2.8. In vitro starch digestibility In vitro starch digestibility of samples was determined based on the method described by Englyst, Kingman, and Cummings (1992) with minor modifications. Porcine pancreatic alpha-amylase (P7545, Sigma- Aldrich, St. Louis, MO) (0.45 g) was dispersed in water (4 mL) by magnetic stirring for 5 min. The dispersion was then centrifuged for 10 min at 1500g, and a portion of the supernatant (2.7 mL) was trans- ferred to a beaker. Amyloglucosidase (Megazyme International Ireland Ltd., Co., Wicklow, Ireland) (0.32 mL) was diluted to 0.4 mL and 0.3 mL of diluted amyloglucosidase was added to the enzyme solution. In- vertase (I4504, Sigma) (2 mg in 0.2 mL distilled water) was also added to the solution. The enzyme solution was freshly prepared for the di- gestion analysis. Aliquots of guar gum solution (2 mL, 5 mg/mL in 0.05 M HCl), 0.01g pepsin (P-7125, Sigma-Aldrich, St. Louis, MO), fif- teen glass beads (4 mm diameter) and sodium acetate buffer (4 mL, 0.5 M, pH 5.2) were added to samples (100 mg) in test tubes, followed by incubation in a water bath (37 °C) with linear shaking (200 rpm) for 30 min. Without stopping the incubation and shaking, 1.0 mL of the enzyme solution was then added to each tube. Aliquots (0.1 mL) were removed at intervals and mixed with 1 mL of 50% ethanol, and the glucose content of the solution was measured using a glucose oxidase/ peroxidase assay kit (Megazyme International Ireland Ltd., Co., Wicklow, Ireland). Percentages of rapidly digestible starch (RDS,% di- gestible starch at 20 min), slowly digestible starch (SDS, % digestible starch at 120 min – % digestible starch at 20 min), and resistant starch (RS, 100% – % digestible starch at 120 min) were normalized to the total starch content. 2.9. Statistical analysis The experiments were set up as single-factor completely randomized designs with three replicates. Statistical analyses were performed with SPSS software (ver. 24; SPSS Inc., Chicago, IL, USA) on all tests using a one-way ANOVA and Tukey’s post hoc test to detect significant differ- ences. 3. Results and discussion 3.1. Enhancement of SDS and RS fractions by roasting Roasting of pea flour and starch was attempted to enhance the SDS and RS fractions compared to their native counterparts. Table 1 shows total starch, apparent amylose content and starch nutritional fractions Table 1 Chemical composition and in vitro starch digestibility of native, boiled or roasted pea flour/starch. Chemical Composition Total Starch (%, dry basis) Apparent Amylose (%, dry basis) In vitro starch digestibility* (%, dry basis) RDS SDS RS Agassiz pea flour Native 48.6 ± 0.7 d 10.6 ± 0.0 d 15.3 ± 0.1 f 24.9 ± 0.5 e 59.8 ± 0.4 a Oven roasted† 48.7 ± 0.6 d 10.1 ± 0.1 d 14.6 ± 0.3 f 25.7 ± 0.2 e 59.7 ± 0.4 a Microwave roasted† 48.2 ± 0.3 d 10.1 ± 0.3 d 16.8 ± 0.8 e 25.3 ± 1.7 e 58.0 ± 2.5 a NPS Native 97.1 ± 0.7 a 41.3 ± 0.1 a 52.6 ± 0.6 d 30.5 ± 0.5 d 16.9 ± 0.7 d Oven roasted 94.9 ± 0.2 b 40.0 ± 1.0 a 54.5 ± 0.6 c 36.1 ± 0.7 c 9.4 ± 0.2 e Microwave roasted 91.4 ± 0.5 c 37.6 ± 0.1 bc 61.4 ± 0.6 b 34.2 ± 0.2 c 4.4 ± 0.6 f Boiled† 96.5 ± 0.7 ab 41.0 ± 0.1 a 80.3 ± 0.8 a 11.9 ± 1.2 f 7.8 ± 0.7 e MPS Native 98.5 ± 0.4 a 38.0 ± 0.3 b 15.1 ± 0.2 f 45.5 ± 0.2 b 39.3 ± 0.4 b Oven roasted 96.8 ± 0.4 ab 36.2 ± 0.4 c 15.7 ± 0.2 ef 53.9 ± 0.2 a 30.4 ± 0.4 c Values denote mean ± standard deviation, n= 3; Significant difference is indicated by different letters within the same column (P < .05). * RDS, digestible starch at 20 min (%); SDS, digestible starch at 20–120 min of incubation (%); RS, resistant starch (%); data are expressed as a percentage of total starch. † Oven roasting was done at 160 °C for 30 min; Microwave roasting was done at 1.1 kW for 2 min. After cooling down to<50 °C, the sample was mixed and roasted again for another 2 min. The process was repeated for three cycles; Boiling was done by cooking a suspension of NPS (10%, w/v) at 100 °C for 15 min. Z.-H. Lu et al. Food Chemistry 245 (2018) 71–78 73 (RDS, SDS and RS) (dry basis) of native, boiled, or roasted pea flour/ starch. Native Agassiz pea flour had 48.6% total starch, including 10.6% apparent amylose content. Neither oven roasting (160 °C, 30 min) nor microwave roasting (1.1 kW, 6 min) affected the total starch and apparent amylose content (P > .05). The same was true for in vitro starch digestibility except that RDS was slightly higher for mi- crowave roasted sample. Native pea flour had low RDS (15.3%), and high SDS (24.9%) and RS (59.8%). The low starch digestibility in native pea flour was effectively retained after roasting by either oven or mi- crowave, since there were almost no significant differences in digest- ibility profiles among native, oven roasted and microwave roasted pea flour (Table 1). This result is consistent with the results obtained by roasting corn samples (Felsman et al., 1976). Roasting was also applied to two commercial pea starches (NPS and MPS) (Table 1). NPS and MPS had similar amounts of total starch (97.1% and 98.5%) and apparent amylose (41.3% and 38.0%). How- ever, their in vitro starch digestibility profiles were distinct in their native form. Native NPS had a high amount of RDS (52.6%), inter- mediate SDS (30.5%) and low RS (16.9%) compared to native MPS (15.1% RDS, 45.5% SDS and 39.3% RS). After oven roasting, both starches showed a ∼2% decrease in total starch and 1–2% decrease in apparent amylose content. When a suspension of NPS (10%, w/v) was boiled at 100 °C for 15 min, 80.3% of starch became rapidly digestible. However, when roasting these two native pea starches, the starch di- gestion profiles showed no drastic change compared to the native counterpart. The SDS portion in native NPS was even increased from 30.5% to 36.1% at some expense of RS after oven roasting, while oven roasting caused no change in RDS of MPS, and increased SDS (from 45.5% to 53.9%) at some expense of RS. Roasting was obviously an effective processing method to retain the SDS and RS fractions in the original starch material. Since roasted flour/starch is ready to eat with a pleasant aromatic flavor (Ma et al., 2016), it could be directly in- corporated to minimally cooked/processed food formulae for enrich- ment of SDS and RS content in foods. Oven roasting seemed more ef- fective in retaining or enhancing SDS and/or RS amount than microwave roasting as shown in Table 1. Nonetheless, roasting by mi- crowave could be an alternative to oven roasting in case processing time is a critical factor. Comparing the in vitro starch digestibility profile of roasted pea flour and the two roasted pea starches with their native counterparts, there were no changes in nutritional fractions in roasted pea flour whereas a significant loss of RS fraction was found in roasted pea starches (Table 1). This result indicated that the physical barrier (such as the matrix in pea flour, especially cotyledon cell walls and protein matrix) to digestion enzymes played an important role in retaining/ enhancing SDS and/or RS fractions by roasting. Based on this result, alginate encapsulation of pea starch (native, boiled or roasted) was thus adopted to create a physical barrier to starch digestion enzymes, and the resultant microbeads were used in pea bread making to further enhance the SDS and/or RS content in gluten-free pea bread. 3.2. Enhancement of SDS and/or RS by alginate encapsulation of pea starch (microbeads) Two sizes of microbeads were prepared (Supplemental Fig. 2). The sizes were 2 mm and 0.5 mm in diameter for freshly-made microbeads, and ∼1.2 mm and ∼0.3 mm in diameter after drying, respectively. A microscopic image showing an edge of a cross section of one of the microbeads encapsulating roasted NPS is displayed in Fig. 1. It shows that starch granules were densely packed in an envelope of alginate gel matrix. Under polarized light, the Maltese cross of starch granules ap- peared to be intact, indicating the presence of semi-crystalline structure in roasted starch. 3.2.1. Effect of encapsulation and different treatments of starch core material on total starch content and in vitro starch digestibility of microbeads Table 2 shows the total starch content and in vitro starch digest- ibility (dry basis) of microbeads containing native, oven roasted, mi- crowave roasted, or boiled NPS, and native or oven roasted MPS. The total starch content of microbeads was measured using their ground powder. The difference in the total starch content between the ground microbead powder (81–86%) and its corresponding core material (> 97%) is due to the mass of the alginate capsule material in the sample (∼10%). Alginate encapsulation of pea starch effectively low- ered the RDS fraction of microbeads to a much lower level than that of their corresponding native starch core material (Table 1), either for entrapped NPS or entrapped MPS, with the lowest RDS seen for MPS entrapped microbeads (2.2%–3.7% RDS). Even for the boiled NPS (80.3% RDS), its encapsulated form only had 23.4% RDS. As a result of encapsulation, the SDS and RS contents were largely increased. A si- milar profile of in vitro starch digestibility (∼15% RDS, ∼50% SDS and ∼35% RS) was observed for encapsulated native, oven roasted and microwave roasted NPS, with the exception of boiled NPS (bead size of 0.5 mm in diameter). It could be concluded that encapsulation was an effective way to enhance SDS and RS fractions with some dependence on the original starch digestibility of the starch core material. 3.2.2. Effect of starch core material (NPS vs. MPS) on in vitro starch digestibility of microbeads As mentioned above, native and oven roasted MPS had significantly lower RDS fractions than those of their corresponding NPS counterparts (Table 1). As a result, microbeads with entrapped native or oven roasted MPS also showed significantly lower RDS fraction (2.2–3.7%) than that of their NPS counterpart (14.7–15.7% RDS) (Table 2) (P < .05). The result further demonstrated that the physical barrier derived from encapsulation was a dominant factor for the low RDS and enhanced SDS and RS fractions, but with some dependence on the starch digestibility of original core materials. This conclusion could also be affirmed by the in vitro starch digestibility of crushed beads (Supplemental Table 1 in online version), which was near to the value of the original core materials (Table 1). 3.2.3. Effect of microbead size (0.5 mm vs. 2 mm) on in vitro starch digestibility The size of microbeads affects its application in food preparation and also mouthfeel of finished foods. Smaller size beads are favorable to be incorporated to formulae for a better homogeneity after mixing, less damage from processing, and negligible change of mouthfeel. As shown in Table 2, the larger size microbeads had only a little lower RDS content than the smaller size beads, which may be explained by the smaller surface area of large beads which is less accessible to enzymes. However, smaller size beads showed a much better bread making per- formance (loaf appearance, loaf volume and crumb structure) and un- detectable impact on mouthfeel of the resulting breads. Thus, smaller size microbeads were selected for future study. 3.2.4. Total carbohydrate release of microbeads Dried microbeads containing encapsulated native, boiled, or oven roasted NPS were immersed sequentially in simulated gastric fluid (SGF), simulated intestinal fluid (SIF), and simulated colonic fluid (SCF) for 4, 3, and 3 h, respectively, and their total carbohydrate release was measured and is shown in Fig. 2. When microbeads were immersed in SGF (pH 1.5), the dry beads were swollen as a result of the hydration of the hydrophilic COO- groups of alginate. Swollen beads still effectively held starch core material. After 4 h of incubation in SGF, less than 2% of the total carbohydrate was released from these different microbeads (Fig. 2). After being transferred to SIF (pH 6.8), the beads started to erode and disintegrate. The percentage of released carbohydrate sig- nificantly increased within the first hour of incubation in SIF, especially Z.-H. Lu et al. Food Chemistry 245 (2018) 71–78 74 for the small beads (0.5 mm in diameter) that encapsulated oven roasted starch. Three hours later when transferred to SCF (pH 7.4), the beads continued to erode and disintegrate. The percentage of released carbohydrate increased dramatically for all samples within the first hour of incubation, with the small beads encapsulating oven roasted starch (0.5 mm in diameter) still the highest. Even still, total carbohy- drate release was< 25% for all the samples, suggesting that en- capsulation effectively slowed carbohydrate release. The result was supportive of those from in vitro starch digestibility and further affirmed that the physical barrier is a dominant factor controlling starch di- gestibility of entrapped starch (Table 2). 3.3. Bread making 3.3.1. Bread making performance and loaf quality Satisfactory baking performance, loaf appearance, loaf volume, crumb texture and palatability were obtained from pea ingredients. While not much difference was seen in loaf appearance and crumb structure compared to a wheat bread control, the taste and mouthfeel of pea breads were different. The pea breads were moist and slightly mushy when fresh, but then floury after storage for one day at room temperature (quickly retrograded). However, after toasting, no differ- ence could be detected in taste, flavor and mouthfeel compared to the wheat bread control. Specifically, breads made from 100% roasted pea flour had acceptable quality, with a pleasant aroma arisen from roasting. Incorporating 50% NPS (native or roasted) with 50% native Agassiz pea flour produced breads with lighter color, better bread shape and comparable loaf volume, and a more uniform crumb structure than those from 100% pea flour. When 50% of roasted NPS was gradually replaced by encapsulated roasted NPS at 0, 12.5, 25, to 37.5% levels (Fig. 3, from top to bottom), no negative effects were observed on the bread making ability and loaf quality (appearance, loaf shape, volume and crumb structure). The loaf quality seemed even better after in- corporating microbeads, with no cracks on the crust, a smooth crust surface, a chewy and non-gritty mouthfeel and no raw starch taste. Characteristics of pea breads incorporating NPS or MPS are shown in Table 3. As shown in Table 3, the moisture content of the freshly baked pea breads was in the range of 75–87% (dry basis). Breads made from encapsulated roasted starch had relatively higher moisture con- tent. Correspondingly, their baking loss was low (Table 3). The order of specific loaf volume from high to low was: bread made from 100% native pea flour > pea flour/MPS (50:50) ≈ pea flour/native NPS (50:50) ≈ 100% roasted pea flour > pea flour/roasted NPS (50:50) > pea flour/starch/microbeads. Pea breads made from 100% pea flour were somewhat over-proofed with large cracks on the crust. A better loaf shape was seen for those breads incorporating microbeads (Fig. 3). Surprisingly, the best palatability was also obtained from the breads incorporating microbeads, especially the bread with 37.5% mi- crobeads (Table 3). The microbeads provided more chewiness to the crumb, in contrast to the mushy mouthfeel and less chewiness in breads made from 100% native pea flour. In general, all the pea breads were acceptable in loaf appearance and palatability, comparable to the white bread control. Fig. 1. Brightfield (left) and polarized light (right) micrographs of transection of microbeads prepared from oven roasted pea starch (NPS). Table 2 Effect of core material on total starch content and in vitro starch digestibility of mi- crobeads. Core material Total starch (%, dry basis) In vitro starch digestibility* (%, dry basis) RDS SDS RS NPS Native 85.8 ± 0.1 a 15.7 ± 0.2 b 46.5 ± 0.3 c 37.7 ± 0.5 e Oven roasted 83.5 ± 0.5 bc 14.7 ± 0.3 c 45.3 ± 0.2 c 40.0 ± 0.3 d Oven roasted (φ 2 mm)† 85.5 ± 0.4 ab 8.5 ± 0.6 d 40.6 ± 0.2 d 50.9 ± 0.4 c Microwave roasted 83.8 ± 0.8 abc 15.0 ± 0.2 bc 54.4 ± 1.3 a 30.7 ± 1.3 f Boiled 83.7 ± 0.2 abc 23.4 ± 0.6 a 52.7 ± 0.3 b 23.9 ± 0.3 g MPS Native 85.2 ± 0.9 ab 2.2 ± 0.4 f 24.1 ± 0.5 f 73.7 ± 0.9 a Oven roasted 81.7 ± 0.5 c 3.7 ± 0.4 e 25.5 ± 0.4 e 70.8 ± 0.8 b Values denote mean ± standard deviation, n = 3; Significant difference is indicated by different letters within the same column (P < .05). † If not specified, the size of microbeads was 0.5 mm in diameter. The total starch content of microbeads was measured using their ground powder, while the in vitro starch digestibility was measured with intact (unground) microbeads. * RDS, digestible starch at 20 min (%); SDS, digestible starch at 20–120 min of in- cubation (%); RS, resistant starch (%); data are expressed as a percentage of total starch. Fig. 2. The amount of total carbohydrate released from different sizes of calcium alginate encapsulated beads of native, boiled, and oven roasted NPS under different pH conditions immersed sequentially in SGF, SIF, and SCF for 4, 3, and 3 h, respectively. If not specified, the size of microbeads was φ 0.5 mm. Z.-H. Lu et al. Food Chemistry 245 (2018) 71–78 75 3.3.2. Chemical composition Pea bread made from 100% pea flour (native or roasted) had the lowest total starch content of 32.5% and 30.1%, respectively, but the highest protein content of 19.4% and 18.0%, respectively (dry basis) (Table 3). Replacement of 50% flour by NPS or MPS increased the total starch content to around 50%, which diluted protein content to ∼12%. Total starch content decreased when the starch was replaced with mi- crobeads due to the alginate capsule material also being present in the bread sample. It is noteworthy to mention that, although the shell material of microbeads (calcium alginate) existed in breads, the ash content of pea breads (3.3–4.1%) (dry basis) was even lower than that of the bread made from 100% pea flour (4.3%) (Table 3). This result could assuage concerns from consumers regarding the use of calcium alginate in pea breads. 3.3.3. In vitro starch digestibility of pea breads In vitro starch digestibility fractions of pea breads made from 100% native/roasted pea flour; pea breads from 50% native pea flour in- corporating 50% native/roasted NPS or MPS; and pea breads made from native pea flour, pea starch and starch-entrapped microbeads (NPS or MPS) at ratios of 50:37.5:12.5, 50:25:25, and 50:12.5:37.5, are shown in Table 3. Pea bread made from 100% native pea flour had significantly lower starch digestibility than that of the wheat bread control (P < .05), which could be attributed to the low digestibility of pea starch and the large amount of other components such as pea protein, pea fibre, etc., which would create physical barriers to limit the starch availability to starch hydrolyzing enzymes. Among pea breads made from 100% native/roasted pea flour or incorporating 50% native/ roasted starch (either NPS or MPS), not many significant differences were seen in their starch nutritional fractions (RDS, SDS and RS) (P > .05). This result was unexpected and indicated that, even though roasting of native pea flour/starch effectively retained the SDS and RS fractions, those portions of SDS and RS could not sustain the severe baking conditions (163 °C) and high hydration level (87.5% of water in the bread formula) in the bread making process. Further research will apply roasted pea flour/starch to the cookie making process, to in- vestigate if roasting would boost SDS and/or RS fractions in foods of a limited water system such as cookies. As expected, encapsulating pea starch and applying the resulting microbeads to pea bread making largely enhanced the SDS and RS fractions in pea breads. As shown in Table 3, increasing the amount of microbeads in the pea bread recipe (by replacing the roasted NPS with encapsulated NPS at levels of 12.5%, 25%, and 37.5%) resulted in a linear decrease in RDS and a linear increase in the SDS and RS fractions, while the eating quality of the pea bread was not compromised ac- cording to the palatability data. Compared to bread without in- corporated microbeads, incorporating 37.5% entrapped NPS mi- crobeads decreased the RDS content from 83.5% to 57.7% while RS content increased from 8.1% to 25.0%, and the SDS content was also doubled (from 8.5% to 17.3%). In addition to the physical barrier provided by alginate encapsula- tion which largely lowered starch digestibility of resulting breads, the core material of microbeads was again found to be another dominant factor affecting in vitro starch digestion profiles of the obtained breads. MPS had distinctly lower starch digestibility than that of NPS, either native or oven roasted (Table 1). As a result, encapsulated MPS (native or roasted) had also much lower starch digestibility (RDS < 3.7%) (Table 2). Without encapsulation, breads made from pea flour and MPS (50:50, native or roasted) showed little difference in starch digestibility compared to their corresponding NPS counterpart (Table 3). However, when replacing NPS by roasted MPS for preparing microbeads and applying these in bread making at 37.5% level, much lower RDS of resulting breads was obtained (45.9% RDS, 23.9% SDS and 30.2% RS) (Table 3). Furthermore, when native MPS was encapsulated and applied to bread making at 37.5% level, an extremely lower starch digestibility profile of the bread was obtained (36.0% RDS, 28.5% SDS and 35.5% RS) but at an expense of palatability of the bread (Table 3). None- theless, with this broad range of SDS and RS in pea breads developed in this study, it is possible to customize varying amounts of RDS, SDS and RS by altering the ratio of pea flour, starch and microbeads, and also by different core materials and/or with different treatments, according to calorie needs for those overweight and obese individuals who have diabetes or are at risk for diabetes. 4. Conclusion A batter-based formula and processing procedure were used to produce gluten-free breads using pea flour, starch, roasted flour/starch and starch-entrapped microbeads. The pea breads had a comparable loaf appearance (color, shape and volume), crumb structure and tex- ture, and palatability to a wheat-based white bread control, but with enhanced SDS and RS content. Roasted pea flour/starch had a low in vitro starch digestibility profile similar to their native counterparts. However, when applied in bread making, the SDS and RS fractions in roasted flour/starch still became rapidly digestible like the native ones after baking. Alginate encapsulation of pea starch effectively lowered starch digestibility with a dependence on the original starch digest- ibility of starch core materials. The slow release of total carbohydrate from microbeads was also confirmed using simulated gastric, intestinal and colonic fluids. Pea flour combined with encapsulated roasted pea starch produced good quality pea bread, and the SDS and RS fractions of the obtained bread increased with the increasing ratio of the en- capsulated starch. Incorporating 37.5% of encapsulated roasted MPS in pea bread largely enhanced the SDS and RS fractions (45.9% RDS, 23.9% SDS and 30.2% RS), compared to a wheat bread control (97.3% RDS, 0.2% SDS and 2.5% RS). Acknowledgements The authors thank Dr. D.J. Bing for providing Agassiz pea seeds, and Fig. 3. Gluten-free pea breads incorporating different ratios of microbeads (encapsulated roasted NPS) (from top to bottom: 0, 12.5%, 25% and 37.5%, based on dry blend). Z.-H. Lu et al. Food Chemistry 245 (2018) 71–78 76 Ta bl e 3 C ha ra ct er is ti cs an d in vi tr o st ar ch di ge st ib ili ty of pe a br ea ds m ad e fr om pe a fl ou r, st ar ch an d st ar ch -e nt ra pp ed m ic ro be ad s. R at io of pe a fl ou r/ st ar ch / m ic ro be ad s M oi st ur e (% ,d ry ba si s) Ba ki ng lo ss (% ) Sp ec ifi c lo af vo lu m e, cm − 3 /g (w et ba si s) Pa la ta bi lit y Pr ot ei n (% ,d ry ba si s) A sh (% ,d ry ba si s) To ta l st ar ch (% ,d ry ba si s) In vi tr o st ar ch di ge st ib ili ty * (% ,d ry ba si s) R D S SD S R S 10 0 (n at iv e pe a fl ou r) /0 /0 75 .4 ± 1. 4 e 24 .5 ± 0. 2 a 4. 2 ± 0. 1 a 69 ± 1 e 19 .4 ± 0. 1 a 4. 3 ± 0. 0 a 32 .5 ± 0. 1 h 80 .9 ± 2. 1 c 8. 9 ± 2. 0 f 10 .2 ± 0. 3 ef 10 0 (r oa st ed pe a fl ou r) /0 /0 82 .1 ± 0. 3 bc 20 .0 ± 0. 1 h 3. 5 ± 0. 1 b 71 ± 1 cd e 18 .0 ± 0. 0 b 4. 2 ± 0. 0 a 30 .1 ± 0. 2 i 81 .7 ± 1. 1 bc 9. 3 ± 0. 9 f 9. 0 ± 0. 3 ef N PS -b as ed fo rm ul a 50 (n at iv e fl ou r) /5 0 (n at iv e N PS )/ 0 77 .7 ± 0. 6 de 22 .9 ± 0. 2 cd 3. 8 ± 0. 1 b 71 ± 1 de 11 .6 ± 0. 1 e 3. 3 ± 0. 0 c 47 .7 ± 0. 2 de 82 .5 ± 1. 2 bc 9. 4 ± 1. 2 ef 8. 1 ± 0. 5f 50 (n at iv e fl ou r) /5 0 (r oa st ed N PS )/ 0 77 .8 ± 1. 3 de 24 .3 ± 0. 4 ab 3. 0 ± 0. 0 c 72 ± 0 bc de 11 .7 ± 0. 0 e 3. 5 ± 0. 0 bc 49 .9 ± 0. 2 c 83 .5 ± 1. 5 bc 8. 5 ± 1. 8 fg 8. 1 ± 0. 4 f 50 (n at iv e fl ou r) /3 7. 5 (r oa st ed N PS )/ 12 .5 (e nt ra pp ed N PS ) 82 .3 ± 1. 7 bc 22 .5 ± 0. 3 cd e 2. 5 ± 0. 1 d 75 ± 1 ab c 11 .2 ± 0. 1 f 3. 3 ± 0. 0 c 48 .0 ± 0. 1 d 75 .3 ± 1. 4 d 12 .8 ± 0. 5 de 11 .8 ± 1. 8 e 50 /2 5/ 25 81 .2 ± 0. 6 bc d 22 .1 ± 0. 4 de f 2. 4 ± 0. 0 d 76 ± 1 ab 11 .7 ± 0. 0 de 3. 7 ± 0. 0 b 47 .4 ± 0. 2 de f 68 .0 ± 2. 7 e 13 .7 ± 0. 7 d 18 .3 ± 2. 1 d 50 /1 2. 5/ 37 .5 83 .7 ± 0. 5 ab 20 .6 ± 0. 6 gh 2. 3 ± 0. 0 d 77 ± 1 a 11 .9 ± 0. 1 de 4. 1 ± 0. 0 a 44 .9 ± 0. 1 g 57 .7 ± 1. 0 f 17 .3 ± 1. 0 c 25 .0 ± 0. 3 c M PS -b as ed fo rm ul a 50 (n at iv e fl ou r) /5 0 (n at iv e M PS )/ 0 79 .0 ± 0. 2 cd e 23 .2 ± 0. 0 bc d 3. 8 ± 0. 1 b 70 ± 1 de 12 .1 ± 0. 1 d 3. 3 ± 0. 1 c 51 .5 ± 0. 3 b 83 .4 ± 1. 7 bc 5. 0 ± 1. 9 g 11 .6 ± 0. 9 e 50 (n at iv e fl ou r) /5 0 (r oa st ed M PS )/ 0 78 .7 ± 0. 7 cd e 23 .3 ± 0. 1 bc 3. 7 ± 0. 0 b 71 ± 1 cd e 11 .8 ± 0. 0 de 3. 3 ± 0. 1 c 51 .6 ± 0. 2 b 85 .3 ± 0. 4 b 5. 2 ± 1. 3 g 9. 4 ± 0. 9 ef 50 (n at iv e fl ou r) /1 2. 5 (r oa st ed M PS )/ 37 .5 (e nt ra pp ed M PS ) 86 .6 ± 0. 5 a 20 .9 ± 0. 1 fg h 2. 3 ± 0. 1 d 77 ± 1 a 11 .2 ± 0. 3 f 4. 1 ± 0. 1 a 46 .5 ± 0. 2 ef g 45 .9 ± 0. 4 g 23 .9 ± 1. 2 b 30 .2 ± 1. 6 b 50 /1 2. 5/ 37 .5 (e nt ra pp ed na ti ve M PS )† 86 .2 ± 1. 6 a 21 .5 ± 0. 1 ef g 2. 4 ± 0. 0 d 73 ± 1 bc d 11 .9 ± 0. 0 de 4. 1 ± 0. 1 a 46 .0 ± 0. 7 fg 36 .0 ± 0. 4 h 28 .5 ± 1. 0 a 35 .5 ± 0. 7 a W he at -b as ed w hi te br ea d co nt ro l – – – – 15 .9 ± 0. 1 c 2. 6 ± 0. 1 d 65 .0 ± 1. 0 a 97 .3 ± 0. 9 a 0. 2 ± 0. 9 h 2. 5 ± 0. 5 g V al ue s de no te m ea n ± st an da rd de vi at io n, n = 3; Si gn ifi ca nt di ff er en ce is in di ca te d by di ff er en t le tt er s w it hi n th e sa m e co lu m n (P < .0 5) . † N at iv e M PS w as en tr ap pe d to m ak e m ic ro be ad s. O th er th an th is sa m pl e, ov en ro as te d st ar ch w as us ed as co re m at er ia l fo r en ca ps ul at io n. * R D S, di ge st ib le st ar ch at 20 m in (% ); SD S, di ge st ib le st ar ch at 20 –1 20 m in of in cu ba ti on (% ); R S, re si st an t st ar ch (% ); da ta ar e ex pr es se d as a pe rc en ta ge of to ta l st ar ch . Z.-H. Lu et al. Food Chemistry 245 (2018) 71–78 77 Miss Esther Matus for conducting physicochemical analyses and in vitro starch digestibility. This study was supported by Agriculture and Agri- Food Canada through an Agri-Innovation Project Grant (Pulse Cluster AIP CL-03). Conflict of interest The authors declare that they have no conflict of interest. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.foodchem.2017.10.037. References AACC International (1999). Approved methods of analysis (11th ed.). St. Paul, Minnesota: AACC International. AACC International (2001). Approved methods of analysis (11th ed.). St. Paul, Minnesota: AACC International. Chung, H. J., Liu, Q., & Hoover, R. (2009). Impact of annealing and heat-moisture treatment on rapidly digestible, slowly digestible and resistant starch levels in native and gelatinized corn, pea and lentil starches. Carbohydrate Polymers, 75, 436–447. Dubois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A., & Smith, F. (1956). Colorimetric method for determination of sugars and related substances. Analytical Chemistry, 28, 350–356. Englyst, H. N., Kingman, S. M., & Cummings, J. H. (1992). Classification and measure- ment of nutritionally important starch fractions. European Journal of Clinical Nutrition, 46, S33–S50. Felsman, R. J., Harvey, R. W., Linnerud, A. C., & Smith, F. H. (1976). Effect of roasting temperature on corn grain characteristics. Journal of Animal Science, 42, 476–480. Goñi, I., García-Diz, L., Mañas, E., & Saura-Calixto, F. (1996). Analysis of resistant starch: A method for foods and food products. Food Chemistry, 56, 445–449. Jenkins, D. J. A., Wolever, T. M. S., Taylor, R. H., Barker, H. M., Fielden, H., & Jenkins, A. L. (1980). Effect of guar crispbread with cereal products and leguminous seeds on blood-glucose concentrations of diabetics. British Medical Journal, 281, 1248–1250. Kaur, M., Singh, N., & Sodhi, N. S. (2005). Physicochemical, cooking, textural and roasting characteristics of chickpea (Cicer arietinum L.) cultivars. Journal of Food Engineering, 69, 511–517. Koksel, H., Sivri, D., Scanlon, M. G., & Bushuk, W. (1999). Comparison of physical properties of raw and roasted chickpeas (leblebi). Food Research International, 31, 659–665. Kushi, L. H., Meyer, K. A., & Jacobs, D. R. (1999). Cereals, legumes, and chronic disease risk reduction: Evidence from epidemiologic studies. American Journal of Clinical Nutrition, 70, S451–458. Ma, Z., Boye, J. I., Azarnia, S., & Simpson, B. K. (2016). Volatile flavor profile of Saskatchewan grown pulses as affected by different thermal processing treatments. International Journal of Food Properties, 19, 2251–2271. Mathers, J. C. (2002). Pulses and carcinogenesis: Potential for the prevention of colon, breast and other cancers. British Journal of Nutrition, 88, S273–S279. Özdemir, M., & Devres, O. (2000). Analysis of color development during roasting of hazelnuts using response surface methodology. Journal of Food Engineering, 45, 17–24. Park, I., Kim, Y.-K., Kim, B. H., & Moon, T. W. (2014). Encapsulated amylosucrase-treated starch with enhanced thermal stability: Preparation and susceptibility to digestion. Starch, 66, 216–224. Rizkalla, S. W., Bellisle, F., & Slama, G. (2002). Health benefits of low glycaemic index foods, such as pulses, in diabetic patients and healthy individuals. British Journal of Nutrition, 88, S255–S262. Rose, D. J., Keshavarzian, A., Patterson, J. A., Venkatachalam, M., Gillevet, P., & Hamaker, B. R. (2009). Starch-entrapped microspheres extend in vitro fecal fer- mentation, increase butyrate production, and influence microbiota pattern. Molecular Nutrition and Food Research, 53, S121–S130. Sciarini, L. S., Ribotta, P. D., León, A. E., & Pérez, G. T. (2010). Effect of hydrocolloids on gluten-free batter properties and bread quality. International Journal of Food Science & Technology, 45, 2306–2312. Venkatachalam, M., Kushnick, M. R., Zhang, G., & Hamaker, B. R. (2009). Starch-en- trapped biopolymer microspheres as a novel approach to vary blood glucose profiles. Journal of the American College of Nutrition, 28, 583–590. Z.-H. Lu et al. Food Chemistry 245 (2018) 71–78 78 http://dx.doi.org/10.1016/j.foodchem.2017.10.037 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0005 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0005 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0010 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0010 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0015 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0015 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0015 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0020 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0020 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0020 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0025 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0025 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0025 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0030 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0030 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0035 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0035 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0040 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0040 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0040 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0045 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0045 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0045 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0050 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0050 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0050 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0055 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0055 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0055 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0060 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0060 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0060 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0065 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0065 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0070 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0070 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0070 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0075 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0075 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0075 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0080 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0080 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0080 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0085 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0085 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0085 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0085 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0090 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0090 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0090 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0095 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0095 http://refhub.elsevier.com/S0308-8146(17)31675-8/h0095 Effect of roasted pea flour/starch and encapsulated pea starch incorporation on the in vitro starch digestibility of pea breads Introduction Materials and methods Materials Physical modification of pea flour/starch to enhance SDS and RS fractions Oven roasting Microwave roasting Alginate encapsulation Gluten-free pea bread making Chemical analysis Light microscopy Carbohydrate release study Evaluation of bread quality Loaf moisture, specific volume and baking loss Palatability In vitro starch digestibility Statistical analysis Results and discussion Enhancement of SDS and RS fractions by roasting Enhancement of SDS and/or RS by alginate encapsulation of pea starch (microbeads) Effect of encapsulation and different treatments of starch core material on total starch content and in vitro starch digestibility of microbeads Effect of starch core material (NPS vs. MPS) on in vitro starch digestibility of microbeads Effect of microbead size (0.5mm vs. 2mm) on in vitro starch digestibility Total carbohydrate release of microbeads Bread making Bread making performance and loaf quality Chemical composition In vitro starch digestibility of pea breads Conclusion Acknowledgements Conflict of interest Supplementary data References