Influence of the Addition of Extruded Flours on Rice Bread Quality

1 Influence of the addition of extruded flours on rice bread quality 2 Mario M. Martínez, Bonastre Oliete, Laura Román, Manuel Gómez* 3 Food Technology Area, E.T.S. Ingenierías Agrarias. Valladolid University, Ed. La Yutera, Avd. 4 Madrid 44, 34004 Palencia, Spain. 5 6 7 * Corresponding author e-mail: pallares@iaf.uva.es 8 9 10 11 12 13 1 14 Abstract 15 The extrusion may improve coeliac bread quality by modifying the functional properties of 16 flour. This study investigates the influence of the substitution of 10% of rice flour by extruded 17 rice flours (three intensities of treatment and two particle sizes) on the characteristics of gluten- 18 free bread (specific volume and texture) at constant consistency. The microstructure and 19 rheology of the doughs obtained and their behaviour during fermentation have also been 20 analysed. The extruded flours increase dough consistency, and the effect is more noticeable with 21 increasing intensities of treatment. The use of extruded flours requires the addition of a larger 22 volume of water to obtain a constant consistency. The addition of extruded flour decreases 23 dough development, producing a lower specific volume and greater bread hardness. This effect 24 is minimized by increasing the particle size. The staling of bread from 24 to 72 hours is less 25 noticeable with a larger particle size. 26 PRACTICAL APPLICATIONS 27 This study evidences that the use of extruded flours in rice bread making allows increasing 28 dough hydration and therefore the bread yield while decreasing bread staling. However, the 29 correct selection of extrusion treatment and flour particle size is essential to achieve appropriate 30 results, being preferable the use of coarse flours with more intense extrusion treatment. 31 32 Keywords: extrusion, gluten-free bread, particle size, microstructure 33 2 34 1. Introduction 35 Gluten-free breads are characterized by their deficient quality and high price compared with 36 traditional breads. This has led to increased interest in developing high quality gluten-free 37 products in recent years (Cureton & Fasano, 2009), with the consequent increase in the volume 38 of research and the number of scientific publications on this subject. 39 Most research studies into the development of gluten-free products have focussed on the 40 substitution of wheat flours by mixtures of gluten-free cereals, starches, proteins and 41 hydrocolloids (Schober, 2009), and on the enzymatic improvement of these formulations 42 (Rosell, 2009). However, less information is available on the use of physical treatments to 43 modify the functional properties of flours used in gluten-free products, and these methods have 44 been less widely employed. Hydrothermal treatments stand out as being among the most 45 effective physical treatments. 46 If starch, the major component of flour, is subjected to a high-temperature treatment with 47 enough moisture it can be gelatinized, increasing the swelling capacity of the granule, 48 decreasing crystallinity and sometimes causing break-up of the granules (Atwell et al., 1988). 49 Hoover & Vasanthan (1994) also demonstrated that starch undergoing thermal treatment at 50 100ºC presents a high gelatinization temperature and high resistance to acid hydrolysis due to 51 realignment of the starch chains, although these changes varied depending on the kind of starch 52 used. 53 Pregelatinized rice flour has been widely used as the principal ingredient in many kinds of foods 54 (rice cakes, infant foods, instant rice pudding) due to its thickener properties. It is known that 55 uses of pregelatinized rice flour are determined by its physicochemical and functional properties, 56 which differ depending on the variety of rice and the processing method employed (Hsieh & 3 57 Luh, 1991; Lu, et al., 1994). These hydrothermal treatments may be performed in various ways, 58 such as drying a paste by atomization or heated drums; but one of the most versatile alternatives 59 is extrusion. Extrusion is a treatment which applies heat and mechanical strain to a flour-water 60 mixture. The main interest of the extrusion of flours and starches is to modify their functional 61 properties, which will vary according to the extrusion conditions applied (Curic, et al., 2009). 62 When flours are extruded, changes take place in starch which modify the rheological behaviour 63 of dough, similar to the changes that occur when dough is subjected to cooling-heating cycles 64 (Hagenimana et al., 2006). However, extrusion causes more intense changes to the starch than 65 traditional cooking methods, as it damages a larger number of starch granules and modifies the 66 cold thickening capacity (Wolf, 2010). This treatment may even rupture amylopectin molecules 67 (Mercier & Feillet, 1975). Colonna et al. (1984) demonstrated that extruded wheat starches 68 contain amylose and amylopectin chains with a lower molecular weight than those obtained by 69 treatment in heated drums. This effect, due to the shear force applied to the starch, is translated 70 into an increase in the solubility and a decrease in the cold thickening capacity (Doublier, et al., 71 1986). 72 These functional modifications of flour after extrusion could also lead to changes in the baking 73 properties. However, this has not been extensively researched in gluten-free breads. 74 Defloor et al. (1991) found that a mixture of extruded starches and emulsifying agents improved 75 the quality of breads prepared with a mixture of tapioca and soya and a high level of hydration 76 (145%). 77 Sanchez, et al. (2008) observed that the addition of extruded rice flour improved bread volume 78 and crumb structure, but this effect was more noticeable when the waxy varieties of rice were 79 used. However those authors used high percentages of extruded flour (15% and 30%) and 4 80 modified the quantity of water in the formula according to the penetrometer dough consistency 81 values; those variations led to an increase in dough hydration by more than 30% in the majority 82 of cases. It is also important to note that those authors used over 80% starch in the formula. 83 The effect of the substitution of 10% of the rice flour by extruded non-acidified and acidified 84 rice flour has also been tested (Clerici & El-Dash, 2006) (Clerici, et al., 2009). However, those 85 authors did not use hydrocolloids in their bread-making process and the specific volume of their 86 breads was excessively low in all cases. 87 In summary, some studies do exist on the use of extruded flours in gluten-free breads, but 88 research needs to be extended to include formulae with the addition of hydrocolloids, with 89 hydration percentages below 90%, and with rice flour as the main ingredient. 90 The particle size of the rice flour is also known to have an effect on gluten-free bread-making 91 (Araki, et al., 2009; Nishita & Bean, 1982; Ylimaki, et al., 1988), but there have been no studies 92 to determine the effect of the particle size of extruded rice flour on the functional characteristics 93 of gluten-free breads. 94 In this study of constant consistency gluten-free bread making we have determined the effect of 95 substituting 10% of non-extruded rice flour by extruded rice flour produced using three different 96 extrusion intensities and with two different particle sizes. We analysed the viscous behaviour of 97 flours in a heating-cooling cycle, dough rheology, dough development and gas production 98 during fermentation, the differences in the microstructure of the flours and doughs, and the 99 texture properties and specific volume of the breads. 100 5 101 2. Materials and methods 102 2.1 Materials 103 The rice flour used in this study was provided by Harinera Los Pisones (Zamora, Spain). Rice 104 flour was extruded using an industrial Buhler Basf single-screw extruder (Buhler S.A., Uzwil, 105 Switzerland). Three kinds of flour were used. Flour 1 was extruded with the addition of 2% 106 moisture with a maximum temperature at the end of the extruder of 110ºC. Flour 2 was extruded 107 with the addition of 15% moisture and a maximum temperature of 110ºC. For flour 3, 10% 108 moisture was added and the maximum temperature in the extruder was 140ºC. The resulting 109 products were ground by compression rollers and sieved to obtain flours with two different 110 particle sizes, fine and coarse. Fine flours (f) were obtained by sieving through a 132 micron 111 screen, and coarse flours (c) were retained between a 132 micron sieve and a 200 micron sieve. 112 Depending on the thermal treatment and the particle size, extruded rice flours were referred to 113 by a number (1, 2, 3) and a letter (f, c). Non-extruded rice flour was use as the control. Due to 114 the fact that the chemical composition of coarse and fine flour could be changed by shieving the 115 flour into fractions, chemical composition is attached (table 1). 116 Saf-Instant yeast (Lesaffre, Lille, France), dry refined salt (Esco European Salt Company, 117 Niedersachsen, Germany), local tap water, white sugar (Acor, Valladolid, Spain), refined 118 sunflower oil (Coosur, Vilches, Spain) and hydroxypropylmethylcellulose (HPMC) Methocel 119 K4M Food grade (Dow Wolf Celullosics, Bitterfeld, Germany) were also used in the bread- 120 making. 121 6 122 2.2 Methods 123 2.2.1. Flour characterisation 124 The viscous behaviour of flours during the heating-cooling cycle was measured with the Rapid 125 Visco Analizer (RVA) (Newport Scientific, Warriewood, Australia), following AACC method 126 61.02.01 (AACC, 2012). The microstructure of the flours was analysed with an environmental 127 scanning electron microscope (ESEM) (FEI, Quanta 200FEG, Oregon, USA) with integrated x- 128 ray microanalysis using an EDAX Genesis XM2i, which enables wet samples to be analysed at 129 ambient pressure without superficial metallization. 130 2.2.2. Dough rheology, gas production and dough microstructure. 131 The rheological behaviour of dough at constant moisture was studied using a Thermo Scientific 132 Haake RheoStress1 controlled strain rheometer (Thermo Fisher Scientific, Schwerte, Germany) 133 and a Phoenix II P1-C25P water bath which controlled the analysis temperature (set at 25ºC). 134 The rheometer was equipped with parallel-plate geometry (60 mm diameter titanium serrated 135 plate-PP60 Ti) with 3 mm gap. After adjustment of the 3 mm gap, the excess batter was 136 removed and vaseline oil (Panreac, Panreac Química SA, Castellar del Vallés, Spain) was 137 applied to cover the exposed sample surfaces. Dough was rested for 300 seconds before 138 measuring. Samples were analysed in duplicate and without yeast. First, a strain sweep test was 139 performed at 25ºC with a strain range of 0.1 - 100 Pa and a constant frequency of 1Hz to 140 identify the linear viscoelastic region. On the basis of the results obtained, a strain value 141 included into the linear viscoelastic region was used in a frequency sweep test at 25ºC with a 142 frequency range of 100-0.1Hz. Values of the complex modulus (G*[Pa]), elastic modulus 143 (G’[Pa]), viscous modulus (G’’[Pa]) and tangent δ (G’’/G’) were obtained for different 144 frequency values (ω [Hz]) (Dobraszczyk & Morgenstern, 2003). 7 145 A rheofermentometer (Chopin, Tripette and Renaud, France) was used to analyse dough height, 146 gas production and gas liberation related to the fermentation time following the method 147 described by Czuchajowska & Pomeranz, (1993). However, the authors adapted the method for 148 gluten-free dough. Only 200g of dough were placed into the rheofermentometer container, the 2 149 kg weight indicated by the method was removed, 3% yeast was added to the formula, and 150 fermentation in the rheofermentometer was performed at 30ºC. 151 Dough microstructure was studied with an ESEM microscope (FEI, Quanta 200FEG, Oregon, 152 USA). The doughs did not contain yeast to avoid alterations in image visualization. 153 154 2.2.3. Bread making 155 The following ingredients (g/100g flour) were used in the bread making: Saf-Instant yeast 156 (3g/100g), salt (1.8g/100g), refined sunflower oil (6g/100g), HPMC (2g/100g) and white sugar 157 (5g/100g). The quantity of water in the doughs made with extruded flour was regulated in each 158 sample to obtain a G* value equal to the G* value of the control dough. 159 The water temperature in all the tests was between 20ºC and 22ºC. Rice flour was substituted by 160 extruded rice flour at a rate of 10g/100g. Control breads containing no extruded rice flour were 161 also prepared. After mixing all the ingredients for 8 minutes in a Kitchen Aid 5KPM50 mixer 162 (Kitchen Aid, Michigan, USA), 250 g of the bread dough were placed in model 151090 163 aluminium pans measuring 108 mm by 232 mm (ALU-Schale, Wiklarn, Germany). 164 Fermentation was performed at 30ºC and 75% RH for 90 minutes in an FC-K proofer (Salva, 165 Lezo, Spain). After fermentation, doughs were baked in an electric modular oven (Salva, Lezo, 166 Spain) for 40 minutes at 190ºC. The loaves were removed from the moulds after a 60-minute 167 cooling period and were weighted. The measurements on the breads were performed 24 hours 8 168 after baking, except for the texture parameters which were measured at 1, 24 and 72 hours after 169 baking. The loaves were introduced into polyethylene plastic bags and stored at 20ºC until 170 analysis. All the elaborations were performed twice. 171 172 2.2.4. Bread characteristics 173 Bread volume was determined using a laser sensor with the BVM-L 370 volume analyser 174 (Perten Instruments, Hägersten, Sweden). The volume measurements were performed in 175 duplicate on two loaves from each elaboration. The specific volume was calculated as the ratio 176 of bread volume to its mass. 177 Crumb texture was measured with a TA-XT2 texture analyzer (Stable Microsystems, Surrey, 178 UK) fitted with the “Texture Expert” software. A 25-mm diameter cylindrical aluminium probe 179 was used in a “Texture Profile Analysis” (TPA) double compression test to penetrate to 50% of 180 the sample depth at test speed of 2 mm/s and with a 30 second delay between first and second 181 compressions. Hardness (N), cohesiveness and springiness were calculated from the TPA graph 182 (Gómez et al., 2007). Texture analyses were performed on 30 mm thick slices at 1, 24 and 72 183 hours after baking. Analyses were performed on two slices from two loaves (2x2) from each 184 type of elaboration, taking the average of the 4 measurements made. 185 2.2.5. Statistical analysis 186 All data were presented as mean values and analysed using an analysis of unidirectional 187 parametric variance (ANOVA) using Fisher’s least significant difference (LSD) (p<0.05). The 188 analyses were performed using the Statgraphics Centurion XVI statistical package (StatPoint 189 Technologies Inc, Warrenton, USA). Additionally, in order to plot the hardness values over 190 time, an analysis of variance was also performed taking into account time as a factor for 9 191 repeated measures using Fisher’s least significant difference (p<0.05). The Statistica 6 software 192 (Statsoft Inc, Tulsa, USA) was used for this analysis. 193 194 3. Results and Discussion 195 3.1. Electron microscopy of flours 196 Figure 1 shows the microstructure of the flours used in this study. It may be observed that the 197 particles of non-extruded rice flour (a) was formed by compound starch granules connected by a 198 compact protein structure. In contrast, in the extruded rice flour with the most intense treatment 199 (d), the starch granules lose their integrity and a paste is formed in which the different 200 components cannot be distinguished in the resultant particles. In the extruded flours with low- 201 intensity hydrothermal treatment, intermediate structures are observed. Whilst swollen starch 202 granules, which represent a phase prior to gelatinization, are observed in flour 1 (b), fusion of 203 the different components is observed in flour 2 (c). 204 Our observations coincide with those reported by Yeh et al. (1999) and by Chao-Chi Chuang & 205 Yeh (2003), who studied the morphological changes of rice starch during heating extrusion. Yeh 206 et al. (1999) showed that non-extruded rice flour had a powder-like appearance and that starch 207 granules swelled as they advanced through the cold zone of the extruder but without losing the 208 powder-like appearance. When the rice flour reached the heating zone of the extruder, the starch 209 granules started to melt and formed a continuous matrix. However, those authors obtained their 210 samples from intermediate zones of the extruder and without milling. Our samples were 211 obtained at the end of the extrusion treatment after drying, milling and sieving. Chao-Chi 212 Chuang & Yeh (2003) showed that starch gelatinization increased with increases in the 213 temperature and duration of treatment. 10 214 215 3.2. Viscous behaviour of flours during a heating-cooling cycle. 216 The RVA parameters for non-extruded rice flour and for 10% substituted extruded flours are 217 shown in Table 2. It can be seen that the more intense the extrusion treatment, the lower the 218 values of pasting viscosity (PV), breakdown (BR), trough (TR), setback (ST) and final viscosity 219 (FV). No significant differences were observed in relation to particle size. Dough viscosity 220 decreased throughout the heating-cooling cycle as the treatment intensity increased. However, 221 no significant differences were observed between the milder treatments (1 and 2). It is known 222 that changes in the physicochemical properties of starch during extrusion develop as a result of 223 morphological changes of the starch granules and the degree of gelatinization (Camire et al., 224 1990; Yeh & Li, 1996). It is also known that dough viscosity depends mainly on the degree of 225 gelatinization of the starch granules and the degree of rupture of the molecular chains (El-Dash 226 et al., 1983). Previous studies related high PV values to a high quantity of non-gelatinized 227 starch, whereas low PV values indicated a proportion of gelatinized starch which is attributable 228 to the variation in the degree of depolymerization and the molecular tangle resulting from the 229 processing conditions (Barres et al., 1990; McPherson et al., 2000). The decrease in PV, BR and 230 TR observed when the intensity of treatment increased, previously observed by Hagenimana et 231 al. (2006), could indicate degradation and gelatinization of the starch. High values of these 232 parameters in the non-extruded rice flour would be related to the presence of non-gelatinized 233 starch. These modifications in the starches and in the flour particles have been studied on the 234 photomicrographs, and the flours undergoing the most intense treatment (flours 3) contained the 235 highest quantity of gelatinized starch. 11 236 The fall in FV and ST values in the extruded flours has already been observed by Doublier et al. 237 (1986) and by Mercier & Feillet (1975). These values indicate the degree of retrogradation that 238 occurrs after heating. When the hot gels are cooled, the increase of viscosity depends on the 239 tendency of starch to reassociate. The extruded rice flours that had undergone the milder 240 treatments showed higher FV values than the same flours extruded with more intense treatments. 241 Although it seems that starch gelatinization is the main factor responsible for changes in the 242 RVA curve, other authors have also observed modifications in the amylose/amylopectin ratio 243 during the extrusion treatments of corn flour (Chinnaswamy & Hannah, 1990) and wheat flour 244 (Colonna et al., 1984). Those authors state that fragmentation of amylose and amylopectin 245 chains takes place during extrusion, and that this is more intense in the amylopectin chains; this 246 will therefore modify the behaviour of the flours during the heating-cooling cycle. 247 248 3.3. Dough rheology and gas production 249 3.3.1. Dynamic rheology of dough. 250 Figures 2 and 3 show the rheological properties of gluten-free doughs at constant moisture. 251 There was a clear tendency to an increase in G’, G’’ and G*, and a decrease in tag δ when 252 extruded rice flours were added to the formula. The most marked effect was observed after the 253 addition of flours 2 and 3; flour 2 was associated with the highest G’ and G* values and the 254 lowest tag δ values. Particle size, on the other hand, produced no clear differences. Thus, when 255 increasing the intensity of the extrusion process, doughs became more consistent and elastic, and 256 this effect was most noticeable with flour 2. G’ values were higher than G’’ values in all cases 257 and over the whole frequency range, which indicated behaviour to be more elastic than viscous. 12 258 Chao-Chi-Chuang & Yeh (2002), studying the extrusion process, and Kim et al. (2009), using 259 the steam cooking method, observed that the moisture content in hydrothermal treatments was 260 the key to the variation in G’, G’’ and tag δ values. In particular, Chao-Chi-Chuang & Yeh 261 (2002) detected higher G’ and G’’ values in the treatments with lower moisture; our 262 observations did not coincide with their findings, though it should be recognized that those 263 authors subjected the flours to higher moisture contents (45%-55%) than were employed in our 264 study, with lower extrusion temperatures (20ºC-100ºC) and they used waxy varieties of rice. 265 However, those authors did observe that an increase in the mechanical energy applied during 266 extrusion produced an increase in G’ and G’’ values and a fall in tan δ values. It is already 267 known that an increase in the consumption of mechanical energy usually leads to extruded 268 samples with a higher degree of cooking (González et al., 2000). In our case, the flours with 269 lower viscosity values in the RVA curves, and thus a higher degree of cooking, were those that 270 obtained the highest G’,G’’ and G* values. 271 These changes in rheology may be related to changes in the starch during the extrusion 272 treatment, as Shim & Mulvaney (1999) found that the balance between the intact starch granules 273 and gelatinized or damaged ones affected G’ values. It has already been shown that extruded 274 flours contain a higher quantity of pregelatinized starch than non-extruded flours and, according 275 to Slade & Levine (1994), this greater degree of gelatinization increases the water absorption 276 capacity of doughs. These rheological changes may also be related to the internal structure of the 277 dough, which may be seen in Figure 4. In the two doughs studied (control dough and dough with 278 flour 3c), the structure is composed of large flour particles covered by small simple starch 279 granules, compacted by a matrix formed of water, hydrocolloids and dissolved substances. 280 However, in the case of dough with flour 3c, there was a smaller quantity of simple starch 13 281 granules, which is consistent with what was observed in the microstructure of the extruded 282 flours (the starch granules had lost their integrity). The starch granules observed correspond to 283 starch granules from the non-extruded flour. It could therefore be thought that extruded flour is 284 mixed with the network formed by water and hydrocolloids, and this will modify the 285 viscoelastic properties of dough. 286 Additionally, an increase in the quantity of damaged starch granules was observed in the flour 287 3c sample compared with the control, as can be seen in Figures 1 and 4 (starch granules with 288 small holes and breaks) and in table 1. 289 The amount of water needed in bread-making at constant consistency (G*=15500±1550) is 290 shown in Table 3. The addition of extruded flours increased the quantity of water required to 291 obtain doughs with constant consistency, and thus they can increase the bread yield. This 292 increase was greater in doughs prepared with flours 2 and 3 (the more intense extrusion 293 treatments) than doughs prepared with flour 1. This finding may be explained by a higher degree 294 of damage caused to the starch granules and to the greater degree of gelatinization in the 295 extrusion process with more intense processing conditions (temperature and moisture) (Mercier 296 and Feillet, 1975). However no clear difference was observed between the doughs made with 297 flours 2 and 3 or with flours with different particle sizes. 298 3.3.2. Dough height and gas production. 299 The curves of dough height during fermentation at constant consistency are shown in Figure 5. 300 Regarding gas production, no significant differences were seen in any case (data not shown). 301 Greater height was observed in doughs made with non-extruded flours than in doughs with 302 extruded flours, but the differences only became evident after a certain duration of fermentation 303 (around minute 90). The poor dough height over the whole fermentation process when using 14 304 flour 1f, was noticeable. The differences between the other extruded flours were minimal but 305 doughs made with flour 3 were somewhat more stable if over-fermentation occurred. Greatest 306 height was obtained in doughs made with flours 1c and 2c, although over-fermentation had a 307 negative effect in both cases. Doughs with flour 3 showed a higher stability than the other 308 doughs, and higher G’ and G* values and lower tan δ values; they are thus more consistent and 309 elastic doughs. This could be related to starch gelatinization and damage during the extrusion 310 treatment (Shim & Mulvaney, 1999), as these changes increase the water absorption capacity of 311 dough, as indicated previously (Slade & Levine, 1994). 312 Changes in the dough height during fermentation can be also related to the internal structure of 313 the dough (Figure 4). The dough prepared with flour 3c, which was the dough with the highest 314 stability in the case of excess fermentation, had the lowest quantity of simple starch granules. 315 This is consistent with the results observed in the microstructure of extruded flours, which 316 showed a loss of integrity of the starch granules. It can therefore be assumed that extruded flours 317 form a mixture with the network produced by the water and hydrocolloid, modifying dough 318 height during fermentation. 319 320 3.4. Bread properties 321 3.4.1. Specific volume and weight lost 322 Table 3 shows the specific volume of breads made at constant consistency. A decrease in the 323 specific volume was observed when extruded flours were added to the formula, except after the 324 addition of flour 1c, which produced a specific volume equal to that of the control. Only the 325 breads prepared with flours 1c and 2c stood out, showing higher specific volumes than the other 326 extruded flours. 15 327 No differences in weight loss were observed in the doughs during baking (data not shown). 328 In previous studies it has been found that both acidified (Clerici et al., 2009) and non-acidified 329 (Clerici and El-Dash, 2006) extruded rice flours could improve the specific volume of gluten- 330 free breads, depending on the extrusion conditions. However, those authors did not use 331 hydrocolloids in the formula, and the specific volume values obtained were much lower than in 332 our case. This result could indicate some kind of interaction between the hydrocolloids and the 333 extruded flours or their components. Sanchez et al. (2008) made bread with extruded waxy rice 334 flours and also observed an increase in the specific volume of breads when using the extruded 335 flours with the highest intensity extrusion treatments. It is important to note the difference in the 336 moisture values between the bread obtained by those authors as they prepared bread at constant 337 consistency using a penetrometer. In contrast to our study, those authors used a high percentage 338 of starch in the formula and considered that the increase in volume was related to the increase in 339 the quantity of soluble solids, as those components increased the consistency of the aqueous 340 phase, improving the viscoelastic characteristics of doughs. 341 Defloor et al. (1991) also found that the addition of extruded starch improved the volume of 342 breads prepared with a mixture of tapioca and soya flour. It is important to note that these 343 authors used high levels of hydration (145%), and their doughs were therefore less consistent 344 than ours. In their case, extruded starches increased the consistency of very soft doughs (in 345 contrast to our study, in which we used more consistent doughs with lower hydration) and made 346 it possible to increase bread volume. 347 Gallagher et al. (2003), on the other hand, observed that an increase in dough hydration 348 increased bread volume. In our study we also found that the level of hydration could alter the 349 effect of including certain components that increase dough consistency, such as extruded flours, 16 350 and particularly extruded coarser flours, which have a high amount of damaged starch that could 351 contribute greatly to increase dough hydration, as can be seen in table 1 and in figure 1 and 4.In 352 excessively soft doughs, a certain increase in consistency could therefore be useful, though this 353 can have a negative effect in more consistent doughs with lower hydration. Nevertheless, this 354 factor alone cannot explain the changes observed, as breads with a constant consistency were 355 also found to have a different specific volume. 356 357 3.4.2. Texture analysis 358 The texture parameters of bread at constant consistency are presented in Table 3. 359 A tendency to increased hardness was observed as the extrusion intensity of the flours increased. 360 Higher hardness values were also found when using fine flours compared to coarse ones, 361 although significant differences were only detected with loaves elaborated using flours 2. 362 No significant differences were found in the springiness or cohesiveness of loaves made with 363 extruded flours compared to control loaves. As in our study, Clerici & El-Dash (2006) reported 364 higher hardness values in breads made with extruded flours. Maleki et al. (1980) and Morad & 365 Wakeil (1976) found that starch retrogradation was strongly influenced by the moisture content 366 of the product. The reduction of starch retrogradation through increased moisture content would 367 therefore produce softer breads. It is also known that bread hardness correlates with bread 368 volume (Gómez et al., 2011), and thus the explanation of the differences in hardness could be 369 related to differences in the specific volume. 370 The changes in hardness over time are shown in Figure 6. Loaves made with the control flour 371 and with flours 1f and 2f showed the fastest rates of hardening. Loaves made with flours 1c and 372 2c showed the lowest hardness values and their hardening curves had a low gradient. Loaves 17 373 made with flours 3 had the highest hardness values up to 24 hours, but the values subsequently 374 remained almost constant or even fell, indicating a decrease in the rate of staling. 375 Changes in crumb properties associated with staling include an increase in starch crystallinity 376 and crumb hardness and a decrease in aroma, soluble starch and crumb hydration capacity 377 (D'Appolonia & Morad, 1981). Rogers et al. (1988) stated that the main cause of bread staling is 378 starch retrogradation, which increases with increased moisture content of breads. As gluten-free 379 breads have a high moisture content, starch retrogradation may progress faster during storage 380 than in gluten breads. Extruded flours have a higher water-retention capacity than non-extruded 381 flours and they could thus delay water migration from the crumb to the crust, decreasing the rate 382 of staling. Furthermore, the extrusion process may even break the amylopectin chains (Mercier 383 and Feillet, 1975) and Colonna et al. (1984) demonstrated that extruded wheat starches contain 384 amylose and amylopectin chains with lower molecular weights than drum-dryer starches; this 385 could be another factor that decreases retrogradation and the rate of staling of bread. 386 387 4. Conclusion 388 The use of extruded flours in the elaboration of gluten-free bread offers an interesting alternative 389 approach to improve gluten-free breads. We found that the addition of extruded flours subjected 390 to high intensity extrusion treatments produced doughs with a higher elastic modulus and 391 consistency, and that it was necessary to add larger volumes of water to achieve constant 392 consistency. The bakery yield is therefore increased. However, the development of doughs 393 prepared with extruded flours was lower, but these doughs showed higher stability to over- 394 fermentation. In general, the addition of extruded flours reduced the specific volume of breads 395 and increased hardness, but these effects were minimized by using the coarse flour fractions, 18 396 which also reduced the rate of staling. Future studies should look in detail at the effect of these 397 extruded flours on the acceptability of breads and their influence in breads with higher levels of 398 hydration. 399 400 Acknowledgements 401 This study was supported financially by Junta de Castilla y León (VA054A12-2), Spain. The 402 authors are also grateful to Harinera Los Pisones (Zamora, Spain) for supplying the rice flours. 403 19 404 References 405 AACC. 2012. Approved methods of the American Association of Cereal Chemists, method 61- 406 02.01 (RVA) (11th ed.). St. Paul, Minnesota: American Association of Cereal Chemists. 407 ARAKI, E., IKEDA, T., ASHIDA, K., TAKATA, K., YAMAKA, M., and IIDA, S. 2009. 408 Effects of rice flour properties on specific loaf volume of one-loaf bread made from rice flour 409 with wheat vital gluten. Food Sci Technol, 15, 439-448. 410 ATWELL, W., HOOD, L., LINEBACK, D., VARRIANOMARSTON, E., and ZOBEL, H. 411 1988. The terminology and methodology associated and basic starch phenomena. Cereal Foods 412 World, 33, 306-308, 310-311. 413 BARRES, C., VERGES, B., TAYEB, J., and DELLA VALLE, G. 1990. Transformation of 414 wheat flour by extrusion cooking. Influence of screw configuration and operating conditions. 415 Cereal Chem, 67, 427-433. 416 CAMIRE, M., CAMIRE, A., and KRUMHAR, K. 1990. Chemical and nutritional changes in 417 foods during extrusion. Crit Rev Food Sci Nutr, 29, 35-57. 418 CHAO-CHI CHUANG, G., and YEH, A. 2003. Effect of screw profile on residence time 419 distribution and starch gelatinization of rice flour during single screw extrusion cooking. J Food 420 Eng, 63, 21-31. 421 CHAO-CHI-CHUANG, G., and YEH, A. 2002. Effects of product temperature and moisture 422 content on viscoelastic properties of glutinous rice extrudate. Cereal Chem, 79, 36-40. 423 CHINNASWAMY, R., and HANNAH, M. 1990. Macromolecular and functional properties of 424 native and extruded corn starch. Cereal Chem, 67, 490-499. 425 CLERICI, M., and EL-DASH, A. 2006. Extruder rice flour as a gluten substitute in the 426 production of rice bread. Arch Latinoam Nutr, 56, 288-294. 20 427 CLERICI, M., ARIOLDI, C., and EL-DASH, A. 2009. Production of acidic extruder rice flour 428 and its influence on the qualities of gluten free bread. Food Sci Technol, 42, 618-623. 429 COLONNA, P., DOUBLIER, J., MELCION, J., MONREDON, F., and MERCIER, C. 1984. 430 Extrusion cooking and drum drying of wheat starch. I. Physical and macromolecular 431 modifications. Cereal Chem, 61, 538-543. 432 CURETON, P., and FASANO, A. 2009. The increasing incidence of celiac disease and the 433 range of gluten-free products in the marketplace . In Gluten-Free Food Science and Technology 434 (E. Gallagher ed.) pp. 1-15. Dublin, Ireland: Wiley-Blackwell. 435 CURIC, D., NOVOTNI, D., BAUMAN, I., KRICKA, T., and DUGUM, J. 2009. Optimization 436 of extrusion cooking of cornmeal as raw material for bakery products. J Food Process Eng, 32, 437 294-317. 438 CZUCHAJOWSKA, Z., and POMERANZ, Y. 1993. Gas-formation and gas retention 1. The 439 system and methodology. Cereal Foods World, 38, 499-503. 440 D'APPOLONIA, L., and MORAD, M. 1981. Bread staling. Cereal Chem, 58, 186-190. 441 DEFLOOR, I., DE GEEST, C., SCHELLEKENS, M., MARTENS, A., and DELCOUR, J. 1991. 442 Emulsifiers and/or extruded starch in the productions of breads from cassava. Cereal Chem, 68, 443 323-327. 444 DOBRASZCZYK, B., and MORGENSTERN, M. 2003. Rheology and the breadmaking 445 process. J Cereal Sci, 38, 229-245. 446 DOUBLIER, J., COLONNA, P., and MERCIER, C. 1986. Extrusion cooking and drum drying 447 of wheat-starch 2. Rheological characterization of starch pastes. Cereal Chem, 63, 240-246. 448 EL-DASH, A., GONZALES, R., and CIOL, M. 1983. Response surface methodology in the 449 control of thermoplastic extrusion of starch. J Food Eng, 2, 129-152. 21 450 CURETON, P., and FASANO, A. 2009. The increasing incidence of celiac disease and the 451 range of gluten-free products in the marketplace . In Gluten-Free Food Science and Technology 452 (E. Gallagher ed.) pp. 1-15. Dublin, Ireland: Wiley-Blackwell. 453 GALLAGHER, E., GORMLEY, T., and ARENDT, E. 2003. Crust and crumb characteristics of 454 gluten free breads. J Food Eng, 56, 153-161. 455 GÓMEZ, M., JIMÉNEZ, S., RUIZ, E., and OLIETE, B. 2011. Effect of extruded wheat bran on 456 dough rheology and bread quality. Food Sci Technol, 44, 2231-2237. 457 GONZÁLEZ, R. J., TORRES, R. L., and AÑÓN, M. C. 2000. Comparison of rice and corn 458 cooking characteristics. Polish J Food, 50, 29-54. 459 HAGENIMANA, A., DING, X., and FANG, T. 2006. Evaluation of rice flour modified by 460 extrusion cooking. J Cereal Sci, 43, 38-46. 461 HOOVER, R., and VASANTHAN, T. 1994. Effect of heat-moisture treatment on the structure 462 and physicochemical properties of cereal, legume and tuber starches. Carbohydr Res, 252, 33- 463 53. 464 HSIEH, F., and LUH, B. 1991. Rice snack foods. In Rice volume II: Utilization (B. Luh ed.) pp. 465 225-249. Nueva York. EEUU: Van Nostrand Reinhold. 466 KIM, D., BAE, I., INGLETT, G., and LEE, S. 2009. Effect of hydrothermal treatment on the 467 physicochemical, rheological and oil-resistant properties of rice flour. J Texture Stud, 40, 192- 468 207. 469 LU, S., CHEN, C., and LII, C. 1994. Gel-chromatography fractionation and thermal 470 characterization of rice starch affected by hydrothermal treatment. Cereal Chem, 73, 5-11. 471 MALEKI, M., HOSENEY, R., and MATTERN, P. 1980. Effects of loaf volume, moisture 472 content, and protein quality on the softness and staling of bread. Cereal Chem, 57, 138-140. 22 473 MCPHERSON, A., BAILEY, T., and JANE, J. 2000. Extrusion of cross-linked 474 hydroxypropylated corn starches I. Pasting properties. Cereal Chem, 77, 320-325. 475 MERCIER, C., and FEILLET, P. 1975. Modification of carbohydrate components by extrusion- 476 cooking of cereal products. Cereal Chem, 63, 283-297. 477 MORAD, M., and WAKEIL, F. 1976. Effect of wheat starch, protein and moisture content on 478 staling of bread. Getreide, Mehl und Brot, 30, 106-107. 479 NISHITA, K., and BEAN, M. 1982. Grinding methods: Their impact in rice flour properties. 480 Cereal Chem, 59, 46-49. 481 ROGERS, D., ZELEZNAK, K., and LAI, C. H. 1988. Effect of native lipids, shortening and 482 bread moisture on bread firming. Cereal Chem, 65, 398-401. 483 ROSELL, C. 2009. Enzymatic Manipulation of Gluten-Free Breads. In Gluten-Free Food 484 Science and Technology (E. Gallagher ed.) pp. 83-98. Dublin, Ireland: Wiley-Blackwell. 485 SANCHEZ, H., GONZALEZ, R., OSELLA, C., TORRES, R., and DE LA TORRE, M. 2008. 486 Elaboration of bread without gluten from extruded rice flour. Cienc Tecnol Aliment, 6, 109-116. 487 SCHOBER, T. 2009. Manufacture of gluten-free specialty breads and confectionery products. In 488 Gluten-Free Food Science and Technology (E. Gallagher ed.) pp. 130-180. Dublin, Irlanda: 489 Wiley-Blackwell. 490 SHIM, J., and MULVANEY, S. 1999. Effect of cooking temperature and stirring speed on 491 rheological properties end microstructure of corn starch and oat flour gels. Cereal Foods World, 492 44, 349-356. 493 SLADE, L., and LEVINE, H. 1994. Structure-function relationships of cookie and cracker 494 ingredients. In The Science of Cookie and Cracker Production (H. Faridi, ed.) pp. 23-241. 495 Nueva York: Chapman and Hall. 23 496 WOLF, B. 2010. Polysaccharide functionality through extrusion processing. Curr Opin Colloid 497 Interface Sci, 15, 50-54. 498 YEH, A., and LI, J. 1996. A continuous measurement of swelling of rice starch during heating. J 499 Cereal Sci, 23, 277-283. 500 YEH, A., WU, T., and JAW, Y. 1999. Starch transitions and their influence on flow pattern 501 during single-screw extrusion cooking of rice flour. Food Bioprod Process, 77, 47-54. 502 YLIMAKI, G., HAWARYSH, Z., HARDIN, R., and THOMSON, A. 1988. Application of 503 response surface methodology to the development of rice flour yeast breads: objective 504 measurements. J Food Sci, 53, 1800-1805. 505 24 506 Table 1: Chemical composition of different types of flour of non-extruded flour with 10% 507 substituted extruded flour. Damage Flours Moisture Protein (%) Starch (%) (%) Flour 3f 29,81±2.50 7,85±1.24 9,56±0.40 Flour 3c 36,57±0.28 8,42±1.50 10,23±0.65 Flour 2f 19,78±2.60 8,25±1.30 13,96±2.10 Flour 2c 30,29±4.22 8,11±1.11 14,04±1.74 Flour 1f 11,09±2.91 8,74±0.98 10,72±3.65 Flour 1c 10,88±1.64 8,04±1.01 11,2±2.64 Control 7,81±0.99 13,45±3.05 5,75±0.97 508 25 509 Table 2: Viscous behaviour in a heating-cooling cycle of non-extruded flour with 10% 510 substituted extruded flour. Flours PV (cp) BR (cp) TR (cp) ST (cp) FV (cp) Flour 3f 3793a±94 1296a±2 2496a±93 2510b±6 5006a±98 Flour 3c 3913a±46 1224a±182 2688ab±135 2306a±106 4994a±29 Flour 2f 4596b±64 1759bc±19 5610b±67 Flour 2c 4571b±16 1690bc±160 2881b±176 2656bc±121 5537b±56 Flour 1f 4444b±77 1663b±35 2837b±84 2777cd±26 5558b±16 Flour 1c 4682b±244 1706bc±11 2975bc±234 2706c±6 5681b±228 Control 2925d±62 6153c±12 5175c±142 1947c±192 2781ab±42 2773cd±17 3228c±49 511 Values with different letters in the same parameter are significantly different (p<0.05). 512 Data shown are the mean of two repetitions for each type of simple 513 PV = Pasting temperature; BR = Breakdown; TR = Trough; ST = Setback; FV = Final viscosity 514 26 515 Table 3: Specific volume, texture properties and percentage hydration of breads made at 516 constant consistency. 517 Specific Flour volume (m3/kg) Hardness (N) Springiness Cohesiveness % Hydration Flour 3f 2.807a±0.121 22.018e±1,155 0.689a±0,030 0.244a±0,001 76.67d±0,01 Flour 3c 2.813a±0.192 21.370e±0,588 0.738a±0,045 0.314a±0,102 74.55c±0,31 Flour 2f 2.987a±0.171 13.675d±0,583 0.706a±0,015 0.271a±0,005 74.86c±0,08 Flour 2c 3.637b±0.259 6.430c±1,649 0.635a±0,070 0.307a±0,061 78.92c±0,23 Flour 1f 3.041a±0.311 4.153b±0,181 0.615a±0,007 0.286a±0,003 72.98b±0,01 Flour 1c 4.597c±0.121 2.431b±0,378 0.590a±0,125 0.297a±0,013 73.36b±0,33 Control 4.802c±0.044 1.723a±0,693 0.656a±0,067 0.348a±0,018 70.00a±1,17 Values with different letters in the same parameter are significantly different (p<0.05). 518 Data shown are the means of two repetitions for each kind of simple 519 520 27 521 Figure captions: 522 Figure 1: Photomicrographs from the environmental scanning electron microscope (ESEM) 523 (×2000) study of the control flour and flours extruded with different intensities of treatment 524 (analysis made in triplicate with subsequent selection of the most representative 525 photomicrographs). a) control flour (non-extruded), b) extruded flour 1f, c) extruded flour 2f, d) 526 extruded flour 3f. 1) compound starch granule, 2) swollen starch granule, 3) gelatinized starch 527 granules. 528 Figure 2: Mechanical spectrum of doughs. G’ and G’’ values according to oscillation frequency 529 () in fine flours (a) and coarse flours (b). G’ values are represented by filled symbols and G’’ 530 values are represented by unfilled symbols. Control (diamond), flour 1 (square), flour 2 531 (triangle), flour 3 (circle). 532 Figure 3: Mechanical spectrum of dough. tan  (a) and G* (b) according to oscillation frequency 533 (). Extruded fine flours are represented by unfilled symbols and extruded coarse flours by 534 filled symbols. Control (cross), flour 1f (unfilled square), flour 2f (unfilled triangle), 3f (unfilled 535 circle), 1c (filled square), 2c (filled triangle), 3c (filled circle) 536 Figure 4: Photomicrographs of the scanning electronic microscope (SEM) (×1000) of doughs 537 prepared with non-extruded flour with 10% substitution by extruded flour (analysis performed in 538 triplicate with subsequent selection of the most representative photomicrographs). a) control, b) 539 dough made with flour 3c,. Arrows indicate damaged starch granules. 540 Figure 5: Dough height during fermentation. 541 Figure 6: Changes in hardness over time. Extruded fine flours are represented by unfilled 542 symbols and discontinuous lines. Extruded coarse flours are represented by filled symbols and 28 543 continuous lines. Control (cross), flour 1f (unfilled square), flour 2f (unfilled triangle), 3f 544 (unfilled circle), 1c (filled square), 2c (filled triangle), 3c (filled circle) 545 29 546 Figure 1: 547 548 30 549 Figure 2: a 550 b 551 552 31 553 Figure 3: a 554 b 555 556 557 558 559 560 561 562 32 563 Figure 4: 564 565 33 566 Figure 5: 567 568 34 569 Figure 6: 35 30 25 Hardness (N) 20 15 10 5 0 -5 -10 0 570 24 72 Time (h) 35