Quality Evaluation of Gluten-free Danish Bread Employing Different Flours and Starches

Abstract

Coeliac disease has recently attracted interest in food science research. In search of gluten-free (GF) bread with good quality and nutritional value, several sources have been used in the formulation of GF composite flours. The goal of the present work was to study the textural and physical properties as well as the nutritional and sensory characteristics of GF Danish bread made with composite flour based on a mixture of flours and starches. First, we identified the optimal GF formulation by experimental orthogonal design. Regarding bread quality, optimisation of the doughs positively affected the textural and physical parameters of the bread. Moreover, some nutritional values were improved by increasing the ash and slowly digestible starch, and these breads had higher total phenol content and antioxidant capacity. In sensory analysis, the optimized breads received higher scores for flavour, aroma and overall acceptance attributes from the panellists.

Introduction

Coeliac disease currently affects 150 million people (approximately 2% of the world's population). This disease is caused by the ingestion of foods containing wheat gluten or related barley, rye and oat proteins that result in an inflammatory injury to the mucosa of the small intestine, which interferes with the absorption of other nutrients as well. Thus, the only treatment is a gluten-free (GF) diet for life (Wagner et al., 2016). For that reason, coeliac sufferers require alternatives to wheat-based products such as pastas, cookies, cakes and breads, such as croissants. The croissant is a bread made with Danish pastry and is commonly eaten at breakfast or dinner (Deligny and Lucas, 2015). Among the foods that coeliac patients can eat, bread is of special importance regarding quality and nutritional value (Matos and Rosell, 2011). The main challenge in the production of gluten-free bread (GFB) is to mimic the physicochemical properties that the gluten network imparts to the wheat dough in forming the crumb, which allows extensibility, elasticity and retention of CO₂ generated during fermentation and confers sensory characteristics such as softness, flavour and aroma (Therdthai and Zhou, 2003; Poutanen et al., 2009). In recent studies, the nutritional deficiencies of GF products have been investigated, which include low dietary fibre, minerals (calcium, magnesium and iron), vitamins (B12, D and folate) and protein, as well as an excess of saturated fat and a high glycaemic index (GI≥70; Vici et al., 2016). GF products are made with flours and/or starches of different crops (tubers, cereals, seeds, and legumes), as well as some additives (hydrocolloids, protein isolates and/or enzymes), whose main function is to mimic the physicochemical properties of gluten (Ziobro et al., 2013, Demirkesen et al., 2014, Hae et al., 2014, Mohammadi et al., 2014, Aprodu et al., 2015, Mir et al., 2016). Tubers such as red potato (RP; Oxalis tuberosa) can be used for GFB. RP is an endemic crop of the Andes cultivated in highlands of Ecuador, Peru and Bolivia and other countries of South America. It can also be found in New Zealand, Australia and Mexico (Flores et al., 2002). RP is used in soups, stews, boiled, mashed, baked like potatoes, and served as a sweet (Sanketkit et al., 2001). However, it is also a low-calorie food that provides 30 kcal per 100 g. Studies by Chirinos et al., (2009) found phenolic compounds in RP, so if it is used in GFB, it might provide some antioxidant capacity. The aim of this study was to evaluate the effect of red potato flour (RPF) in combination with other raw materials (rice, maize, potato and cassava) in GF Danish bread through bread quality tests (texture, volume, colour and image analysis), nutritional value (proximate composition, in vitro starch digestibility), antioxidant capacity and sensory analysis.

Materials and Methods

Materials and bread formulation    As a control, Danish pastry was formulated with wheat flour (42%; San Antonio Tres Estrellas, Grupo La Moderna, Mexico), salt (0.4%; La Fina, Sales del Itsmo, Mexico), sucrose (4.2%; Zulka, Zucarmex, Mexico), dry yeast (1.3%; Azteca, Levadura Azteca S.A. de C.V., Mexico), dry egg (2%; Ovopol, Ovoplus del Centro S.A. de C.V., Mexico), milk (8.4%; Alpura, Grupo Alpura, Mexico), water (25%) and fat (16.7%; Flex Azul, Unilever, Mexico). For GF treatments, the wheat flour was substituted by GF flour based on rice flour (13.5%; Tres estrellas, Grupo La Moderna, Mexico), cassava starch (8.6%; Fabsa de Oriente S.A. de C. V., Mexico), potato starch (7.1%; Fabsa de Oriente S.A. de C. V., Mexico), corn starch (11.3%; Maizena, Unilever, Mexico) and hydrocolloid (1.5%; Wellence Gluten Free, DOW chemical Pharma & Food Solutions, Germany) (Table 1). RPs were purchased from the local market in Singuilucan, Hgo., México, from the 2015 winter harvest at a state of physiological maturity. It was used in the experimental design in place of flour and starches. RPF was obtained according to the methods of Hathorn et al. (2008), with some modifications. Briefly, the tubers were selected, washed, peeled and cut into 2 mm slices, which were placed on trays and dried in a conventional oven (Kitchen Aid, KOSE500EBS, USA) at 40 °C for 24 h. The dried sample was ground using a blade grinder (IKA, MF 10.1, USA) and passed through a No. 35 sieve.

Table 1. Bread formulations for flours of controls and treatments

	TTR	GFSG	OP1	OP2
Wheat flour (g)	100	-	-	-
Gluten-free flour (g)
Rice flour	-	32	32	-
Corn starch	-	27	27	27
Cassava starch	-	20.5	-	20.5
Potato starch	-	17	-	-
Red potato flour	-	-	37.5	49
Hydrocolloid *	-	3.5	3.5	3.5

TTR: wheat bread control, GFSG: gluten free bread control, OP1: Optimal treatment 1, OP2: Optimal treatment 2.

*  Wellence Gluten Free, DOW chemical Pharma & Food Solutions, Germany

Bread making procedure    Danish pastry was made from dough fermented with all the ingredients except the fat. The ingredients were combined in a mixer (Kitchen Aid, Professional 600, USA) for 5 min at medium speed. Subsequently, the dough was extended and rolled, incorporating the fat. Rolling consisted of extending the dough up to a 6-mm thickness, making a fold and joining the ends of the dough in the centre. This process was repeated three times. After obtaining the layered dough, it was fermented for 120 min at 30 °C. The dough was rolled again and cut into triangles, each with base of 10 cm and corresponding height of 10 cm. Each triangle was rolled, starting from the base, to form the croissant. It was fermented for a further 30 min and finally baked at 180 °C for 20 min (Deligny and Lucas, 2015).

Texture profile analysis of dough and bread    The texture profile analysis (TPA) of the dough and bread was determined in a texturometer (Brookfield, CT3 Texture Analyzer, USA), following the methodology of Bourne (2002). Hardness (N), adhesiveness (mJ), cohesiveness and springiness (dimensionless) were determined.

Dough extensibility    Uniaxial elongation properties of doughs were analysed using the Kieffer Dough and Gluten Extensibility Rig for the texturometer (Brookfield, CT3 Texture Analyzer, USA). The dough extensibility (mm) was measured by the distance at which sample rupture occurred, according to the method AACC 54-10 (2000).

Bread analysis

Specific volume and bake weight lost    Bread volume was determined by seed displacement (AACC 10-05 method, 2000). Based on the weight of the product, the specific volume in cm³/g was calculated. The bake weight loss was calculated by Eq. 1 (Alvarez-Jubete et al., 2010).

  

Colour of crust and crumb    To determine the instrumental colour parameters (L*, a* and b*) digital photos were taken of the bread crumb and crust as described by Velez et al. (2014) using a computer vision system (CVS), which consisted of a hemispherical diffuser dome that contained the inspection zone, a lighting system and a capture system (Canon, Power shot SX110 IS camera, Japan) connected to a PC. The digital images of the samples were processed using ImageJ 1.49v software (National Institutes of Health, USA), selecting 12 regions of 350x350 pixels. The images were converted into RGB values with the colour space converter plug-in, which interpolated the RGB values to the values L* (luminosity), a* (red-green component) and b* (yellow-blue component).

Crumb structure    The crumb structure of the treatments was analysed by image analysis (modified by Sanchez-Pardo et al., 2012), wherein six samples of bread were cut in half with a serrated knife to obtain 12 halves. Image capture was performed by the CVS described above. The image was trimmed in the centre with a size of 720×720 pixels (4.0 cm², 1 pixel = 0.0278 mm) and saved in TIFF format. Afterward, each image was converted to 8-bit format (grayscale) and segmented using the Otsu algorithm, which generates highly uniform binary images for analysis. ImageJ 1.49v software (National Institutes of Health, USA) was used. The following parameters were measured: 1) area of the cells, 2) average area of the cells, 3) cell density per cm² and 4) percentage of dispersed phase (defined as the percentage of air trapped in the crumb per unit area).

Fractal dimension of the crust texture    For crust images, the fractal dimension of the crust texture was evaluated by power-law scaling using the shifting differential counting method (FD_SDBS). FD_SDBS was calculated using the SDBC plug-in of ImageJ 1.49v software (National Institutes of Health, USA). FD_SDBS was estimated from the slope in the log (box count) vs log (box size) plot and by Eq. 2, where “N” is the number of boxes and “r” is the length of the box side   

The results of FD_SDBS show the roughness of the surface with a possible range of values from 2 to 3. Values near 2 are associated with smooth surfaces and values close to 3 with rough surfaces (Pérez-Nieto et al., 2010). Finally, images of the crust texture were obtained with the Interactive 3D Surface Plot plug-in.

Nutritional characteristics

Proximate composition    Proteins (46-12.01), lipids (30-10.01), moisture (44-19.01), ash (08-01.01) and dietary fibre (32-05.01) contents of the breads were determinate according to AACC methods. Carbohydrates were calculated by difference between these components and total weight.

Total starch and in vitro starch digestibility    Total starch of samples was determined by assay kit (Megazyme, K_TSA 07/11, Ireland). For starch digestibility, enzymatic hydrolysis of samples was determined following the method reported by Gularte and Rosell (2011), using 100 mg of powdered bread. According to the hydrolysis rate of starch, three different fractions were quantified. Rapidly digestible starch (RDS) referred to the percentage of total starch that was hydrolysed within 30 min; slowly digestible starch (SDS) was the percentage of total starch hydrolysed within 30 and 120 min and resistant starch (RS) was the remaining starch after 16 h of incubation.

Total phenol content (TPC) and determination of antioxidant capacity (DPPH assay)

The TPC was determined according the Folin-Ciocalteu spectrophotometric method. The phenolic compound fraction was obtained by aqueous extract. The extract was centrifuged at 4,000 rpm for 20 min. A 1580 µL volume of the extract and 100 µL Folin-Ciocalteu reagent were mixed. The mixture was allowed to react for 8 min with 300 µL sodium carbonate (20% w/v). After this incubation, the mixture was reacted for 15 min at 50 °C. The absorbance at 765 nm was then determined. The results were expressed as gallic acid equivalents (GAE) in µg/g, using a gallic acid (Sigma-Aldrich, St. Louis, MO, USA) standard curve. The DPPH radical scavenging capacity assay was used for the determination of antioxidant capacity, using the stable radical 2,2-diphenyl-1-picrylhydrazyl (DPPH, Sigma-Aldrich, St. Louis, MO, USA). In brief, 0.1 mL of aqueous phenolic extract was mixed with 3.9 mL of DPPH in methanol (6×10⁻⁵ M). The reduction of absorbance was read at 516 nm for 60 min as outlined by Bernardino-Nicanor (2016). The percentage of scavenging effect was calculated using Eq. 3.

  

Where I% is the percentage of scavenging, A₀ is the absorbance of the control, and A_s is the absorbance of the test sample.

Sensory evaluation    A semi-trained panel of twelve members evaluated the sensory properties of the breads. The samples were coded with specific numbers to eliminate bias. Panellists were instructed to evaluate the appearance, crust colour, crumb colour, texture (softness, firmness), flavour, chewiness (ability to chew and swallow the bread), aroma and overall acceptance. A nine-point hedonic scale with 1, dislike extremely; 5, neither like nor dislike and 9, like extremely was used (Jan et al., 2016).

Statistical analysis    In this research, an orthogonal design was used with an L16 (2×4) arrangement to select the optimal formulations that resulted in the best GF croissants made with RPF. The design factors were rice flour, corn starch, cassava starch, and potato starch. These factors were evaluated at two different levels, wherein each was substituted with RPF. Based on a selection of the response variables that had the greatest impact on the product's characteristics, the criteria of the statistical design were nominal-is-better, lower-is-better and higher-is-better. The response variables selected for dough optimisation are based on textural tests and dough extensibility. Low values were chosen for the parameters of hardness (N) and adhesiveness (mJ), whereas higher values were chosen for the parameters of cohesiveness, springiness and extensibility (mm). ANOVA (p > 0.05) and Tukey tests for each factor were used as measures of the level of significance and percent contribution. Minitab 16.1 software was used for statistical analysis (Minitab Inc., USA).

Results and Discussion

Optimisation of gluten-free dough    Table 2 shows the results of TPA and extensibility of the orthogonal design where two optimal formulations were obtained. OP1 was made with GF flour, based on rice flour (13.5%), corn starch (11.3%), RPF (15.75%) and hydrocolloid (1.5%). OP2 was made with GF flour, based on cassava starch (8.6%), corn starch (11.3%), RPF (20.6%) and hydrocolloid (1.5%). Table 2 shows the verification of predicted results (OP1 and OP2) by dough TPA and extensibility tests. The hardness of OP1 and OP2 was significantly lower (p < 0.05) with respect to the controls. This hardness reduction in GF dough (GFSG, OP1 and OP2) was due to the type of flour used in the preparation. In wheat bread control (WBC), the use of wheat flour has the behaviour of a viscoelastic fluid due to its gluten content, which increases the product's resistance compared to GF doughs. Renzetti et al. (2016) showed that in the elaboration of laminated dough, a higher fat content produces a better response in the rolling process, and a more stable crumb structure is achieved. They also showed that there is a correlation between proper distinction of the layers (fat and dough) with the expansion of the bread and the structure of the crumb. Adhesiveness, defined as the work required to overcome the attractive force between the sample and the contact surface, indicated that OP1 and OP2 were significantly different (p < 0.05) compared to the control. This behaviour may be due to RPF and the hydrocolloid. Mir et al. (2016) observed that the use of hydroxypropylmethylcellulose in the preparation of GFB increases the springiness and viscosity of dough, where hydrogen bonds interacting with hydrocolloids determines their characteristics. Excessive adhesiveness is counterproductive when rolling the dough; it causes melting of the layers and degrades the structure of the Danish bread. Compared to WBC, cohesiveness, springiness and extensibility were lower in OP1 and OP2. These viscoelastic characteristics that the gluten gives to dough made from wheat flour are reflected in processing (Dobraszczyk and Salmanowicz, 2008). Cohesiveness, springiness and extensibility were higher in OP2 that OP1, in which the higher protein content of RPF causes the formation of links between peptides and proteins that confer viscoelastic properties to the dough. Other studies have shown that the use of hydrocolloids influences this type of property, where among other parameters, it improves cohesiveness, increases elasticity and is able to integrate the dough by absorbing water at low temperatures (Lazaridou et al., 2007). Extensibility is a measure of the tensile strength of the dough. It is related to the quality of gluten (in wheat dough) and to the expansion of the bread (volume). The extensibility provides a measure of the dough's ability to retain the air that forms the structure during kneading, fermentation and baking. In the case of laminated dough, it is possible to obtain more uniform and defined layers. In GF dough, it can be observed that extensibility values do not exceed 50% of WBC, so that the treatments (mixtures of flours, starches and hydrocolloids) cannot be projected yet to completely mimic the viscoelastic properties of gluten. The study of doughs is fundamental for the development of quality breads, because it allows a better understanding of the characteristics and behaviour of the raw material involved. This allow to evaluate the best strategies to use in the processes, with the aim of increasing productivity and profitability in the food industry.

Table 2. TPA and extensibility of optimal formulations according to the criteria of the Taguchi orthogonal design L16 (2×4)

		Texture profile analysis in dough
Factor	Level	Hardness (N)	Adhesiveness (mJ)	Cohesiveness	Springiness	Extensibility (mm)	OP1	OP2
Rice flour	−	4.63	0.72	0.402	0.51	10.69	−
	+	3.57	0.57	0.368	0.46	12.38		+
Corn starch	−	4.31	0.59	0.394	0.51	11.93	−	−
	+	3.89	0.70	0.376	0.46	11.14
Cassava starch	−	4.29	0.54	0.379	0.50	12.28		−
	+	3.90	0.75	0.391	0.48	10.79	+
Potato starch	−	4.36	0.68	0.379	0.49	12.35
	+	3.83	0.62	0.391	0.49	10.72	+	+

+: with red potato flour. −: without red potato flour. OP1: optimal treatment 1. OP2: optimal treatment 2. Minitab^® 16.1.0

Bread analysis    Results of the bread analysis are shown in Table 3. In bread TPA, the relationship between the product form, the structure and the processing technique directly influenced the hardness of the bread, because the composition and arrangement of the structure opposes deformation (Schober et al., 2008). The use of hydrocolloids increases GFB cohesiveness by interacting with the water in the dough and forming hydrogen bonds. Additionally, the formation of a gel due to the presence of starch produces a “gummy” property, which is caused by gelatinisation of starch in baking (Matos and Rosell, 2012). Springiness was higher in GFB compared to the WBC control, with significant differences (p < 0.05). This parameter is directly related to cohesiveness. The quality of the bread is associated with the volume and structure of the crumb, which also depends on the gluten. For volume and baking weight loss measurements, 100 g of dough was taken. Among the treatments, the specific volume decreased, and the baking weight loss increased (r = 0.98). The water content and the formation of the crumb structure during baking gave volume to the product through the generation of CO₂ and steam. However, in the GFB, the formation of a gel, caused by the starches (Naito et al., 2005), that is only thermostable for a short time does not allow for the retention of water in the final structure, resulting in a dryer product with less volume (Onyango et al., 2011). Colour is an important factor for consumer acceptance of the product. The crust colour, L* (luminosity) is affected by the composition and colour of the raw material. The values of b* (+b=yellowness, −b=blueness) showed the characteristic golden colour of bread (Cappa et al., 2013). High values of parameter b* were observed in WBC and GFSG, which presented a golden crust, whereas in OP1 and OP2, the addition of RPF decreased the values of this parameter. The RPF had violet coloration, and this directly influenced the colour of the crumb and crust of the bread. The reddish colour of the tuber is due to the content of anthocyanins present in the pericarp and mesocarp, which have been shown, along with other phenolic compounds, to have antioxidant capacity (Chirinos et al., 2009, Alcalde-Eon et al., 2004). The colour of the crumb was similar to that of the crust, with the luminosity affected by the type of raw material. In addition, the a* and b* values presented significant differences (p < 0.05) between the treatments. In the evaluation of the crumb structure, a correlation index r = 0.99 between the cell area and the dispersed phase was seen (Fig. 1), and there were significant differences between treatments (p < 0.05). The dispersed phase indicates the amount of air that formed the structure of the crumb (Farrera-Rebollo et al., 2012) and the solid phase of the Danish pastry is formed by alternating layers of dough (fermented) and fat that result in a structure of elongated cells and an open and non-uniform crumb (Renzetti et al., 2016). WBC had the characteristic structure of Danish bread (Fig. 2a) with a dispersed phase percentage of 40.8%, which was related to the specific volume. GFSG had the lowest percentage of dispersed phase, at 25.29%, which represented a compact structure and was related to the hardness of the bread (Fig. 2b). OP1 and OP2 (Fig. 2c and 2d) showed a dispersed phase percentage of 48.66% and 50.56% respectively, however, the absence of gluten did not allow the formation of a defined structure, resulting in a structure similar to pound cake (Farrera-Rebollo et al., 2012). The relationship between the average area of the cells, the cell per cm² and the dispersed phase (Fig. 2e–2h), finally defines the structure of the product, which in turn is derived from composition of raw materials and processing. None of the GF treatments maintained the characteristic structure of Danish bread. Hager et al. (2012) performed a comparison of GFB structure from different origins and compared it with wheat counterparts, finding that bread made with oatmeal was the only one that presented a structure similar to wheat; however, these breads were made from a batter and not with dough, which changes the properties of the product. It was shown that the average cell size had a correlation index (r = 0.88) with respect to the specific bread volume, where a larger cell size reflected a volume increase in the product (Fig. 1b). The average cell size is affected by internal dough bonds and their expansion limit produced by steam and CO₂ from fermentation. GFB can have fractures, affecting the quality of the product. It has been demonstrated that the use of hydrocolloids can avoid fractures in the product and delay retrogradation (Demirkesen et al., 2014, Mohammadi et al., 2014). Fig. 3 shows the fractal texture values (FD_SDBC) of the crust and a 3D representation of the crust topography. In terms of roughness, according to Pérez-Nieto et al. (2010), values close to 2 are associated with smooth surfaces and values close to 3 with rough surfaces. We also observed that WBC (Fig. 3b) had a more stable surface with no fractures, whereas GFSG, OP1 and OP2 (Fig. 3c and 3e) presented fractures in the crust. These fractures were directly associated with the bread volume, where a correlation index of r = 0.92 was observed between the FD_SDBC values and the specific volume (Fig. 1c). Low specific volume values correlated with lower values of FD_SDBC, indicating that a more stable structure with no surface ruptures tended to result in a product with higher volume.

Table 3. Textural, physics and structural evaluation of gluten-free breads

		TTR	GFSG	OP1	OP2
Texture profile analysis
	Hardness (N)	8.78±0.09b	4.77±0.10a	4.30±0.72c	5.02±0.27c
	Adhesiveness (mJ)	0.0±0.00a	0.0±0.00a	0.0±0.00a	0.0±0.00a
	Cohesiveness	0.408±0.01d	0.508±0.02c	0.883±0.01a	0.732±0.05b
	Springiness	0.7±0.00b	0.8±0.00a	0.8±0.00a	0.8±0.00a
Specific volume (cm3/g)
		5.58±0.02a	3.02±0.01d	3.56±0.02c	4.11±0.05b
Baking weight loss (%)		10.8±0.17c	17.6±0.32a	15.9±0.45b	14.5±0.99b
Crust color
	L*	70.23±0.50b	85.03±0.36a	62.64±0.32c	53.51±0.36d
	a*	14.50±0.18b	−1.35±0.06d	8.31±0.02c	16.08±0.12a
	b*	47.20±0.11a	41.72±0.02b	37.83±0.11c	37.12±0.36d
Crumb color
	L*	94.25±0.12a	92.52±0.31b	61.98±0.18d	70.27±0.06c
	a*	−3.80±0.13c	−4.15±0.07d	7.17±0.08a	2.26±0.14b
	b*	15.37±0.40d	23.16±0.17c	31.62±0.05a	27.84±0.20b
Crumb structure
	Cell area (mm2)	162.12±11.86b	100.94±20.18c	192.91±8.89a	204.87±14.82a
	Average cell area (mm2)	2.69±0.43a	1.40±0.20b	1.51±0.12b	2.54±0.16a
	Cell density per cm2	15.33±2.35c	18.04±2.30bc	32.04±1.78a	19.79±1.04b
	% Dispersed phase (air)	40.8±3.04b	25.3±5.03c	19.79±1.04b	50.5±3.04a

Different letters in the same file mean significant differences p < 0.05 according to the Tukey test. TTR: wheat bread control, GFSG: gluten free bread control, OP1: Optimal treatment 1, OP2: Optimal treatment 2. Total area: 4 cm²

Fig. 1.

Correlation index between quality bread parameters

Fig. 2.

Gray level images 40×40 mm field of view of the treatments: (a) TTR, (b) GFSG, (c) OP1 and (d) OP2. (e–h) Segmented images with the Otsu thresholding algorithm.

Fig. 3.

Fractal dimension of the crust texture. (a) FD_SDBS: fractal dimension obtained by means of shifting differential box counting method, different letters in the graph mean significant differences p < 0.05 according to the Tukey test. (b–e) Crust images corresponding to surface intensity plot. Arrows shows the fractures in crust of the treatments.

Nutritional characteristics and antioxidant activity of breads    The nutritional composition, in vitro digestibility and antioxidant activity of breads are shown in Table 4. For the proximate composition, the results showed that protein is higher in WBC, due to the gluten content, compared with the GFB. GFSG showed the lowest value, as a consequence of gluten absence. In OP1 and OP2, the addition of RPF increased the protein content. Ash was higher in OP1 and OP2, which means the micronutrients are higher. León et al. (2011) identified various minerals in RP, including iron, phosphorus, calcium and zinc. They also found some vitamins (A, B1, B2, B3, B12, and C). These micronutrients might increase the nutritional value of the product. For fat content, which directly influences the energetic content, WBC had the highest (367.99 kcal). In GFSG, OP1 and OP2, a reduction of fat content may have been caused by the use of flours rich in starch. The carbohydrates were higher in GFB, within which dietary fiber was especially higher in OP1 and OP2 than WBC and GFSG. Non-fibre carbohydrates were 44.59% in WBC and 55.23% in OP1, and this was reflected in starch digestibility. In bread, the main component is starch, which results in rapid degradation in the small intestine as almost all the starch is gelatinized (de la Hera et al., 2014). It was observed that the RDS content in breads is above 33%, which leads to a rapid digestion of the product and an increase in blood glucose. Wherever, SDS is present mostly in OP1 and OP2 due to the addition of RPF. SDS is slowly digested in the small intestine and it directly correlated with a reduction of the postprandial glycaemic response. The higher content of SDS in OP1 and OP2 could be caused by the partial gelatinisation of native starch. Native starch is hydrolysed slowly compared to the gelatinized starch, in which the crystallinity and molecular order have been lost and the accessibility of substrate to amylases is greater (Ng et al., 2017). RS is a particular form of starch able to reach the large intestine of human subjecting mainly undigested, where it can be fermented by gut microbiota, increasing dietary fibre. It has been observed that in GFB, the RS content is higher. This might be because retrogradation of starch after baking (RS type III) is greater, which is correlated with the weight lost, due to ineffective retention of water caused by the absence of gluten (Giuberti et al., 2017). The results showed that there TPC was 300% higher for OP1 and OP2 compared to WBC. This could be caused by the addition of RPF. Chirinos et al., (2009) identified phenolic compounds in RP (anthocyanins, flavanols, cinnamic and benzoic acids). They demonstrated that the bioactive compounds present in RP have antioxidant capacity. They reported a radical scavenging of 72% to 80%. In this research, the antioxidant activity was measured by a DPPH radical scavenging capacity assay, which showed that the highest percentage of radical scavenging was OP2 (26.96%), followed by OP1 (17.89%). The antioxidant compounds in RPF increased the antioxidant capacity of GFB. Moreover, several studies have found that some ingredients increase the TPC in bread. Rózylo et al., (2016) studied the effect of addition of brown algae to GFB, and found higher antioxidant activity in breads with 6% brown algae. Sakac et al. (2011) studied GFB based on rice and buckwheat flour and evaluated the antioxidant capacity. They found that as the proportion of buckwheat flour was increased, the TPC and antioxidant capacities of GFP became higher. This demonstrates that the enrichment of GF flours with sources rich in antioxidant compounds can increase the nutritional quality of coeliac foods.

Table 4. Nutritional value and antioxidant activity of gluten-free breads

		TTR	GFSG	OP1	OP2
Proximate composition	(g/100 g db)
	Ashes	0.99d±0.02	1.05c±0.03	1.74b±0.01	1.86a±0.02
	Proteins	8.84a±0.04	4.88c±0.08	5.53b±0.02	5.65b±0.02
	Lipids	17.14a±0.13	13.75b±0.13	10.26c±0.09	13.65b±0.03
	Dietary fiber	0.16c±0.02	0.15c±0.02	0.47b±0.04	0.64a±0.01
	Carbohydrates (by diference)	44.59d±0.13	52.69b±0.16	55.23a±0.08	51.85c±0.11
	Energetic content (kcal)	367.99a±0.40	354.04b±1.33	335.44c±0.63	352.84b±0.33
In vitro starch digestibility	(g/100 g db)
	Total starch	42.56d±0.14	51.03b±0.83	53.17a±0.17	48.77c±0.05
	Rapidly digestible starch	36.84b±0.04	40.04a±0.08	35.11c±0.08	33.69d±0.15
	Slowly digestible starch	4.55d±0.03	9.13c±0.04	16.45a±0.05	13.48b±0.06
	Resistant starch	1.25c±0.03	1.49b±0.02	1.60a±0.02	1.59a±0.03
Total phenolic content	(gallic acid equivalent µg/g)	267.29b±7.64	178.89c±6.01	790.64a±7.35	797.45a±4.99
DPPH % radical
scavenging capacity (I%)		3.72	2.90	17.89	26.96

Different letters in the same file mean significant differences p < 0.05 according to the Tukey test. TTR: wheat bread control, GFSG: gluten free bread control, OP1: Optimal treatment 1, OP2: Optimal treatment 2. db, dry basis.

Sensory evaluation    The sensory evaluation scores are shown in Fig. 4. WBC had the highest score for texture and the chewiest attributes, and the panellists mentioned that the wheat bread control was softer and the masticability was better. GFSG had the lowest values for several attributes; it was categorized as simple bread by the panellists, because it had simple flavour, poor aroma and high firmness. However, OP2 had the highest score in colour, aroma, flavour and overall acceptance, and the panellists argued that this bread had better sensory characteristics. The flavour and aroma may be due to the sweetness of RPF that might have contributed to its higher scores for these sensory attributes.

Fig. 4.

Sensory evaluation of breads (Score 1 = dislike extremely and score 9= like extremely)

TTR: wheat bread control, GFSG: gluten free bread control, OP1: Optimal treatment 1, OP2: Optimal treatment 2

Conclusions

Red potato flour was used in the production of gluten-free Danish bread. A Taguchi orthogonal design was efficient to obtain the optimal formulations of gluten free doughs. In bread quality, the addition of red potato flour improved the texture and volume in gluten free Danish bread, increasing the bread quality. The nutritional value was higher, due to an increase in the amount of slowly digestive starch and resistant starch in addition to the antioxidant capacity of the gluten free Danish bread. Finally, the attributes of aroma and flavour in the sensory evaluation were better with a higher addition of red potato flour, showing greater acceptance by the panellists.

Abbreviations

CVS

Computer vision system

GF

Gluten-free

GFB

Gluten-free bread

GFSG

gluten-free bread control

OP1

Optimal treatment 1

OP2

Optimal treatment 2

RDS

Rapidly digestible starch

RP

Red potato

RPF

Red potato flour

RS

Resistant starch

SDS

Slowly digestible starch

TPA

Texture profile analysis

TPC

Total phenol content

WBC

Wheat bread control

References

• AACC. (2000). Approved methods of the AACC. 10^th ed. American Association of Cereal Chemist. St. Paul MN, USA.
•  Alcalde-Eon,  C.,  Saavedra,  G.,  de Pascual-Teresa,  S., and  Rivas-Gonzalo,  J. (2004). Liquid chromatography-mass spectrometry identification of anthocyanin of isla Oca (Oxalis tuberosa Mol.) tubers. J. Chromatogr., 1054, 211-215.
•  Alvarez-Jubete,  L.,  Arendt,  E., and  Gallagher,  E. (2010). Nutritive value of pseudocereals and their increasing use as functional gluten free ingredients. Trends Food Sci. Technol., 21, 106-113.
•  Aprodu,  I. and  Banu,  I. (2015). Influence of dietary fiber, water, and glucose oxidase on rheological and baking properties of maize based gluten-free bread. Food Sci. Biotechnol., 24, 1301-1307.
•  Bernardino-Nicanor,  A.,  Montañéz-Soto,  J.,  Vivar-Vera,  M.,  Júarez-Goiz,  J.,  Acosta-García,  G., and  González-Cruz,  L. (2016). Effect of drying on the antioxidant capacity and concentration of phenolic compounds in different parts of the Erythrina Americana tree. BioResources, 11, 9741-9755.
•  Bourne,  M. (2002). Food Texture and Viscosity. Food Science and Technology. Texture, Viscosity and Food, Chapter 1, 1-32.
•  Cappa,  C.,  Lucisano,  M., and  Mariotti,  M. (2013). Influence of Psyllium, sugar beet fiber and water on gluten-free dough properties and bread quality. Carbohydr. Polym., 98, 1657-1666.
•  Chirinos,  R.,  Betalleluz-Pallardel,  I.,  Huamán,  A.,  Arbizu,  C.,  Pedreschi,  R., and  Campos  D. (2009). HPLC-DAD characterization of phenolic compounds from Andean Oca (Oxalis tuberosa Mol.) tubers and their contribution to the antioxidant capacity. Food Chem., 113, 1243-1251.
•  de la Hera,  E.,  Rosell,  C., and  Gomez,  M. (2014). Effect of water content and flour particle size on gluten-free bread quality and digestibility. Food Chem., 151, 526-531.
•  Deligny,  C. and  Lucas,  T. (2015). Effect of the number of fat layers on expansion of Danish pastry during proving and baking. J. Food Eng., 158, 113-120.
•  Demirkesen,  I.,  Kelkar,  S.,  Campanella,  O.,  Sumnu,  G.,  Sahin,  S., and  Okos,  M. (2014). Characterization of structure of gluten-free breads using X-ray microtomography. Food Hydrocolloids, 36, 37-44.
•  Dobraszczyk,  D. and  Salmanowicz,  B. (2008). Comparison of prediction of baking volume using large deformation rheological properties. J. Cereal Sci., 47, 292-301.
•  Farrera-Rebollo,  R.,  Salgado-Cruz,  M.,  Chanona-Pérez,  J.,  Gutiérrez-López,  G.,  Alamilla-Beltrán,  L., and  Calderón-Dominguez,  G. (2012). Evaluation of image analysis tools for characterization of sweet bread crumb structure. Food Bioprocess Technol., 5, 474-484.
•  Giuberti,  G.,  Marti,  A.,  Fortunati,  P., and  Gallo,  A., (2017). Gluten free rice cookies with resistant starch ingredients from modified waxy rice starches: Nutritional aspects and textural. J. Cereal Sci., 76, 268-275.
•  Gularte,  M. A. and  Rosell,  C. M. (2011). Physicochemical properties and enzymatic hydrolysis of different starches in the presence of hydrocolloids. Carbohydr. Polym., 85, 237-244.
•  Hae,  J.,  Chul,  D.,  Eun,  S.,  Woung,  O.,  Kim,  H.,  Taik,  S., and  Sook,  S. (2014). Effects of rice flour size fractions on gluten free rice bread. Food Sci. Biotechnol., 23, 1875-1883.
•  Hager,  A.,  Wolter,  A.,  Czerny,  M.,  Zannini,  E.,  Arendt,  E., and  Czerny,  M. (2012). Investigation of product quality, sensory profile and ultrastructure of breads made from a range of commercial gluten-free flours compared to their wheat counterparts. Eur. Food Res. Technol., 235, 333-344.
•  Hathorn,  C.,  Biswas,  M.,  Gichuhi,  P., and  Bovell-Benjamin,  A. (2008). Comparison of chemical, physical, micro-structural, and microbial properties of bread supplemented with sweet potato flour and high-gluten enhancers. LWT-Food Sci. Technol., 41, 803-815.
•  Jan,  R.,  Saxena,  D. C., and  Singh,  S. (2016). Physico-chemical, textural, sensory and antioxidant characteristics of gluten-free cookies made from raw and germinated Chenopodium (Chenopodium album) flour. LWT-Food Sci. Technol., 71, 281-287.
•  Lazaridou,  A.,  Duta,  D.,  Papageorgiou,  M.,  Belc,  N., and  Biliaderis,  C. (2007). Effects of hydrocolloids on dough rheology and bread quality parameters in gluten-free formulations. J. Food Eng., 79, 1033-1047.
•  León,  M.,  Villacorta,  M., and  Pagador,  S. (2011). Chemical composition of “Oca” (Oxalis tuberosa), “Arracacha” (Arracaccia xanthorriza) and “Tarwi” (Lupinus mutabilis). Formulation of a base mixture for food. Revista Venezolana de Ciencia y Tecnología de Alimentos, 2, 239-252.
•  Matos,  M. and  Rosell,  C. (2011). Chemical composition and starch digestibility of different gluten free breads. Plant Foods Hum. Nutr., 66, 224.
•  Matos,  M. and  Rosell,  C. (2012). Relationship between instrumental parameters and sensory characteristics in gluten free breads. Eur. Food Res. Technol., 235, 107-117.
•  Mir,  S.,  Shah,  M.,  Naik,  H., and  Zargar,  I. (2016). Influence of hydrocolloids on dough handling and technological properties of gluten-free breads. Trends Food Sci. Technol., 51, 49-57.
•  Mohammadi,  M.,  Sadeghnia,  N.,  Azizi,  M.,  Neyestani,  T., and  Mortazavian,  A. (2014). Development of gluten-free flat bread using hydrocolloids: Xanthan and CMC. J. Ind. Eng. Chem., 20, 1812-1818.
•  Naito  S.,  Fukami  S.,  Mizokami  Y.,  Hirose  R.,  Kawashima  K.,  Takano  H.,  Ishida  N.,  Kiozumi  M., and  Kano  H. (2005). The efect of gelatinized starch on baking bread. Food Sci. Technol. Res., 11, 194-201.
•  Ng,  S.,  Robert,  S.,  Ahmad,  W., and  Ishak,  W. (2017). Incorporation of dietary fibre-rich oyster mushroom (Pleurotus sajor-caju) powder improves postprandial glycaemic response by interfering with starch granule structure and starch digestibility of biscuit. Food Chem., 227, 358-368.
•  Onyango,  C.,  Mutungi,  C.,  Unbehend,  G., and  Lindhauer,  M. (2011). Modification of gluten-free sorghum batter and bread using maize, potato, cassava or rice. LWT-Food Sci. Technol., 44, 681-686.
•  Pérez-Nieto,  A.,  Chanona-Pérez,  J.,  Farrera-Rebollo,  R.,  Gutiérrez-López,  G.,  Alamilla-Beltrán,  L., and  Calderón-Domínguez,  G. (2010). Image analysis of structural changes in dough during baking. LWT-Food Sci. Technol., 43, 535-543.
•  Poutanen,  K.,  Flander,  L., and  Katina,  K. (2009). Sourdough and cereal fermentation in a nutritional perspective. Food Microbiol., 26, 693-699.
•  Renzetti,  S.,  de Harder,  R., and  Jurgens,  A. (2016). Puff pastry with low satured fat contents: The role of fat and dough physical interactions in the development of a layered structure. J. Food Eng., 170, 24-32.
•  Rózylo,  R.,  Hameed-Hassoon,  W.,  Gawlik-Dziki,  U.,  Siastala,  M., and  Dziki  D. (2016). Study on the physical and antioxidant properties of gluten-free bread with brown algae. CYTA-J. Food, 15, 1-8.
•  Sakac,  M.,  Torbica,  A.,  Sedej,  I., and  Hadnadev,  M. (2011). Influence of Breadmaking on antioxidant capacity of gluten free breads based on rice and buckwheat flours. Food Res. Int., 44, 2806-2813.
•  Sánchez-Pardo,  M.,  Ortiz-Moreno,  A.,  García-Zaragoza,  F.,  Necoechea-Mondragón,  H., and  Chanona-Pérez,  J. (2012). Comparison of pound cake baked in a two cycle microwave-toaster oven and in conventional oven. LWT-Food Sci. Technol., 46, 356-362.
•  Schober,  T.,  Bean,  S.,  Boyle,  D., and  Park,  S. (2008). Improved viscoelastic zein-starch doughs for leavened gluten-free breads: their rheology and microstructure. J. Cereal Sci., 48, 755-767.
•  Therdthai  N. and  Zhou  W. (2003). Recent advances in the studies of bread baking process and their impacts on the bread baking technology. Food Sci. Technol. Res., 9, 219-226.
•  Velez,  N.,  Gómez-Sanchis,  J.,  Chanonna-Pérez,  J.,  Carrasco,  J.,  Millán-Giraldo,  M.,  Lorente,  D.,  Cubero,  S., and  Blasco,  J. (2014). Early detection of mechanical damage in mango using NIR hyperspectral images and machine learning. Biosyst. Eng., 122, 91-98.
•  Vici,  G.,  Belli,  L.,  Biondi,  M., and  Polzonetti,  V. (2016). Gluten-free diet and nutrient deficiencies: a review. Clin. Nutr., 30, 1-6.
•  Wagner,  G.,  Zeiler,  M.,  Grylli,  V.,  Berger,  G.,  Huber,  W.,  Woeber,  C.,  Rhind,  C., and  Karwautz,  A. (2016). Coeliac disease in adolescence: coping strategies and personality factors affecting compliance with gluten-free diet. Appetite, 101, 55-61.
•  Ziobro,  R.,  Witczak,  T.,  Juszczak,  L., and  Korus,  J. (2013). Supplementation of gluten-free bread with non-gluten proteins. Effect on dough rheological properties and bread characteristic. Food Hydrocolloids, 32, 213-220.