2018 Volume 24 Issue 4 Pages 697-706
The chickpea sourdough, also known as “sweet yeast”, has not been experimented in chemically leavened bakery products such as cake so far, although it has been used in traditional breads. The aim of this study was to optimize the amounts of chickpea sourdough (A) and baking powder (B) in cake formulation and by examining various quality parameters. An experimental design suggested by Response Surface Methodology, Central Composite Design was used for optimization. The optimization results were validated experimentally. The cake, produced according to the optimized model (OC) was compared with the control cake (CC) in terms of various quality parameters. According to the results, optimum levels of usage were 64.02 g for A and 5.97 g for B. Smaller and more homogeneous pore structures were obtained in OC texture by the use of A according to the image analysis. The cake also acquired slightly sour, sweetish and soft sensory profiles. DPPH radical scavenging activity and physicochemical properties were also improved by the use of the chickpea sourdough.
The chickpea sourdough is a traditional method also known as “sweet yeast” or “chickpea yeast” and is a well-known method used in various Mediterranean and Balkan countries (Pasqualone et al., 2004). The “sweet yeast” is a traditional and regional name. It quite likely originate from sweet taste and light flavour of the end-product produced by chickpea sourdough. Because, the taste and flavor characteristic of products vary depending on the type of sourdough (raw material such as, rice, wheat or chickpea; process condition; sourdough form such as wet, dried etc.), its microbiota and their metabolits. It is a kind of sourdough which has been used in bread making as a leavening agent and for imparting a distinctive flavor and taste. The chickpea sourdough may also prolong shelf life and provides functional, nutritional and textural quality to bakery products (Sıkılı, 2003). The synthesis of microbial metabolites such as peptides and amino acid derivatives, activation of enzymes, acidification and proteolysis cause several changes during sourdough fermentation, which positively affect the dough and product, and they influence the functional and nutritional quality (Gobbetti et al., 2014). Nowadays, the literature is particularly rich in results and this shows how the sourdough may affect the functional features of baked goods leavened with bakery yeast. Moreover cereals and legumes such as chickpea are an important part of the daily diet and mainly provide nutrients and functional compounds (Coda et al., 2015). Although legumes have been components of the dietary customs of many countries for thousands of years, their usage and their nutritional and functional values have only recently started to be rediscovered and investigated using suitable techniques (Curiel et al., 2015). The chickpea has a high-quality protein content (Gomez et al., 2008) and protein bioavailability due to its well-balanced amino acid composition than many of the several legume seeds (Roy et al., 2010). The chickpea is rich in Ca, P, Mg, K, and is also a good source of important vitamins such as riboflavin, niacin, thiamin, folate, the vitamin A precursor β-carotene and the amino acid histidine (Jukanti et al., 2012). In this respect, the usage of chickpea sourdough will satisfy consumers' natural and healthy product expectations as a natural ingredient.
Some of previous studies have investigated the addition of pea, chickpea or their flours to several bakery products such as chapatti, cakes (Gomez et al., 2008; Hera et al., 2008), biscuits, bread (Rizzello et al., 2014a) and crackers (Pasqualone et al., 2004; Hera et al., 2008; Kadam et al., 2012). However, the studies concerning chickpea fermentation or bakery products contained chickpea sourdough are very limited (Tulbek, 2006; Kefalas et al., 2009). Furthermore, the sourdough usage in bakery products leavened with baking powder like cake is not a general application. In this study, the optimized level of chickpea sourdough and baking powder in the cake formula was determined in terms of volume (V) and acceptability value (AV), and the effects of chickpea sourdough on cake quality properties were examined.
The wheat flour (Cesur Milling Co., Trabzon, Turkey) used for cake production had 14.3 % moisture, 0.63 % ash, 9 % gluten and 80 % gluten index. The baking powder (Dr. Oetker) and other ingredients such as chickpea (Cicer arietinum L.) cv. Koçbaşı were purchased from a local supermarket. The chemicals used for analysis were supplied by Merck Group (Germany).
Experimental design and optimization The amounts of baking powder and chickpea sourdough utilized were considered as two independent factors and two main responses were Volume (V) and Acceptability values (AV). The main effect of the levels of baking powder and chickpea sourdough which were optimization factors, on cake properties was “volume (V)”. Moreover, the chickpea sourdough was first time used in a cake formulation which was a chemically leavened product. For that reason, the consumer acceptance was considered and therefore the AV was determined as a parameter regarding sensory properties of cake samples. Response surface methodology (RSM)-Central composite rotatable design (CCRD) desirability function was used for the effect of the two factors on the two responses (Design Expert 7.0.0, Stat-Ease Inc., Minneapolis, USA). The experimental levels of factors were for B: 0, 1.37, 4.66, 7.96, 9.33 g, for A: 0, 14.64, 49.99, 85.34, 99.98. The two factors and five replicates at the center point led to a set of 13 experiments. The cake samples were produced according to the experimental design. The V and AV results were evaluated by software, and the optimized results were calculated.
Preparation of chickpea sourdough (A) The method used traditionally to prepare chickpea sourdough was slightly modified. 100 g chickpea with 100 mL water was soaked for 12 h and then milled with a laboratory blender (Bosch MMR15A1, Germany). It was mixed with the same amount of water. Salt (2 g), sugar (10 g) and wheat flour (50 g) were added and incubated at 26 °C for 12 h, pH up to 4.5. It was filtered by stainless steel strainer and approx. 280 g filtrate was obtained. It was mixed with wheat flour (140 g), then incubated at 26 °C for 6 h, pH up to 4.5. Because the sourdough fermentation especially performed in the presence of low values of pH (e.g., 3.5–4.5), (De Angelis et al., 2007; Maioli et al., 2008; De Angelis et al., 2009; Najjar et al., 2009; Lappi et al., 2010), in this study, A was incubated up to pH 4.5.
Enumeration of lactic acid bacteria (LAB) and yeast For determination of numbers of LAB, the elective media MRS agar (Merck, Germany) was used. Appropriate dilutions were plated on MRS agars and the plates incubated anaerobically at 35 °C for 48–72 h. For yeast counting, appropriate dilutions were plated on YGC agar (Merck, Germany), incubated (JSR JSBI-C, Korea) aerobically at 25 °C for 72 h. Then the colonies were counted (Lönner et al., 1986; Simsek et al., 2006).
Preparation of cake samples The method described by Sudha et al. (2007) was used for making the cake. The formula included 133 g flour, 133 g sugar, 180 g egg, 53 g milk, 33 g shortening, 50 g refined vegetable oil, and 2 g salt. The levels of A and B were calculated and added according to the experimental design (Table 1). The CC without A was also prepared with the basic formula contained 9.33 g of B for comparing with OC. The batter was prepared by flour batter method, where in the flour, shortening, refined vegetable oil, baking powder, salt and chickpea sourdough were mixed (Kitchen aid, KSM150PSER, Belgium). Eggs and sugar were whipped together until a semi-firm foam resulted. The sugar-egg mix and milk were added to the creamed flour. The cake batter was poured into a rectangular cake pan and baked in a preheated oven (Kenwood, NW796, China) at 175 °C for 40 min.
| Amount of usage (g) | Responses | |||
|---|---|---|---|---|
| No | A | B | V (cm3) | AV |
| 1 | 49.99 | 4.66 | 1190 | 85.40 |
| 2 | 49.99 | 4.66 | 1225 | 80.25 |
| 3 | 49.99 | 9.33 | 1335 | 84.00 |
| 4 | 99.98 | 4.66 | 1280 | 81.25 |
| 5 | 49.99 | 0.00 | 1015 | 72.20 |
| 6 | 14.64 | 7.96 | 1130 | 67.33 |
| 7 | 49.99 | 4.66 | 1210 | 81.40 |
| 8 | 0.00 | 4.66 | 1010 | 61.30 |
| 9 | 49.99 | 4.66 | 1150 | 91.16 |
| 10 | 14.64 | 1.37 | 1010 | 70.20 |
| 11 | 85.34 | 1.37 | 1190 | 77.00 |
| 12 | 85.34 | 7.96 | 1345 | 89.25 |
| 13 | 49.99 | 4.66 | 1185 | 87.80 |
A; chickpea sourdough, B; baking powder, V; volume, AV; acceptability value
Quality properties of the cakes
Pore structure Images of both faces of the two central slices (20 mm thickness) from each loaf were captured using a flatbed scanner (Model Scanjet 8200, HP, Cupertino, USA) with a resolution of 600 dots per inch (dpi) and converted from the true color to grayscale. The cake sample images were calibrated, standardized and optimized by applying the appropriate filters in order to measure the pore's size and distribution using Image-Pro Plus 6.0 (Media Cybernetics Inc., USA) software. The number and the area of pores were characterized by enumerating the pores present in six pre-selected dimensional classes based on area (class 1= 0.05– 0.49 mm2, class 2= 0.50–0.99 mm2, class 3= 1.00–4.99 mm2, class 4= 5.00–9.99 mm2, class 5= 10 — 49.99 mm2, class 6= >50 mm2) as previously described (Bianchia et al., 2008). The number and the area of the pores and the percentages thereof were calculated by the software. The pore area and numbers were then classified by Microsoft Excel (Office 2007, Microsoft Corp, Redmond, WA).
Volume (V), specific volume and crust thickness The V of cakes cooled under room conditions for 1 h after baking were determined by using the method based on displacement with rape seeds. Their weights were measured. The specific volumes of cakes were calculated by using the formula (Artan et al., 2010; Torrieri et al., 2014). The crust thicknesses of the samples were measured at five different points with a digital caliper (Asimeto, Germany).
Acceptability value (AV) For calculating of AV, the panel test was carried out to determine that various properties of the cakes. The hedonic scale used by Olapade and Adetuyi (2007) was slightly modified. Twenty trained panelists having prior experience of testing bakery products assessed (Ndife et al., 2011) the features of cakes, such as crust color, baking quality, symmetry, and internal features such as structural homogeneity, crumb color, taste and aroma. The samples were labeled randomly with three-digit numerical codes and then given to the panelists. Each of 10 attributes was scored using a 10-point hedonic scale, and total AV was calculated out of maximum 100 points.
pH and total titratable acidity (TTA) The cake and the distilled water were mixed in a weight: volume ratio of 1:9. The mixture was crumbled by ultra turrax (IKA, T25, Germany) until a homogeneous mixture was obtained. The mixture was then kept for 10 minutes and the pH was measured with a pH meter (Heidolph,Germany). The mixture was titrated with 0.1 N NaOH by adding phenolphthalein and then TTA was calculated (Rizzello et al., 2016).
Moisture After 3 h of baking, a slice was taken from the cake (Poinot et al., 2008). The crust and the crumb were homogenized, and 5 g were weighed out. Moisture was determined by oven (Microtest MST55, Turkey) drying at 105 °C to constant weight (Akgün and Doymaz, 2005). The cake samples were stored at the same conditions (24±2 °C, kept away directly sunlight). Moisture loss during shelf life was also determined between days 1 and 5.
Color (L*, a*, b*) The color profile for crust and crumb of cake samples was determined using a colorimeter (Konica-Minolta, CR400, Japan) in the form of L*, a* and b*. The measurement was carried out on five points from crust and crumb, and mean values were calculated (Torrieri et al., 2014; Rizzello et al., 2014b).
Antioxidant activity α, α-diphenyl-β-picrylhydrazyl (DPPH) radical scavenging activity method was performed. Soxhlet extraction with petroleum ether was used to remove the oil from cake samples (Sudha et al., 2007). For determination of radical scavenging activity, 10 mg of DPPH was dissolved in 25 mL of 80 % methanol. 0.2 mL of the sample extract was mixed with DPPH solution and 4 mL of 80 % methanol and kept in the dark at room temperature for 30 min. The absorbance (A sample) was recorded at 517 nm through UV-VIS spectrophotometer (Agilent, UV-Visible, USA). The blank sample was prepared using 0.1 mL of methanol and 3.9 mL of DPPH solution and its absorbance (A control) was also recorded. DPPH radical scavenging activity was calculated using the formula (Eq. 1) (Karamac et al., 2002).
 |
Sensory profile The sensory profile analysis was carried out for the aim of determination of differences between the optimized and control cake samples. The intensities of the various sensory characteristics of cake samples were evaluated by panelists. Sensory evaluation of the cakes, which was modified according to Volpini-Rapina et al. (2012), was performed by a panel of 20 male and female volunteers having prior experience of testing bakery products. It was taken into consideration for the panelist selection that their liking and usage of the product, age (between 18–45), no allergies to bakery products or used ingredients. The samples were labeled randomly with three-digit numerical codes and then given to the panelists, who evaluated the intensities of attributes for each cake. The intensities (sweet, fatty, sour/acidic, soda, malty, egg, overall) were scored using a 5-point hedonic scale. Five means extreme perceptible and zero means imperceptible. The results were presented by sensory diagrams.
Statistical analysis After optimization, the results were validated experimentally. The cake sample produced by the optimized model was evaluated in terms of physicochemical and sensory profile analysis in comparison with the control sample. The one sample t-test (IBM-SPSS 1.0.0.781) was used for the comparison (p < 0.05) of the results.
LAB and yeast Yeast numbers of A were between 1.2×103 and 3×103 CFU/g, while LAB were between 1.6×109 and 4×109 CFU/g. The A was produced using multi-stage spontaneous fermentation. More than 50 species of LAB, especially belonging to the genus Lactobacillus, and more than 20 species of yeasts, mostly belonging to the genera Saccharomyces and Candida, have been found in sourdough for making traditional leavened bakery products (Minervini et al., 2012; Lattanzi et al., 2013). These data are consistent with another study which showed dominating microorganisms in spontaneously fermented doughs. The homofermentative Lactobacillus and Pediococcus have been found both in wheat and rye sourdoughs at a level of 3×108 – 3×109 CFU/g (Tulbek, 2006).
Optimization The V and AV determined based on the experimental design are also given in Table 1. There were no differences among the runs of 1, 2, 7, 9 and 13 in terms of factor levels as five replicates were set for CCD in this study. Although the results for the responses were close to each other, they were different. The difference among the similar runs (1, 2, 7, 9 and 13) and the results of other runs has indicated that the model of this study is non-linear.
The data were analyzed by the software in order to determine which functions would be suggested. “Sequential model sum of squares” and “lack of fit” tests were carried out (Table 2) for the V and AV. Standard deviation, R2 (R-squared) and adjusted R2 were determined for each function. The values were compared, and the suggested function was ascertained.
| p values* | |||
|---|---|---|---|
| V (cm3) | AV | ||
| Model selection and lack of fit test | Quadratic | 0.2758 | 0.0041 |
| Linear | < 0.0001 | 0.0322 | |
| Lack of fit | 0.3424 | 0.9175 | |
| Model and independent variable factors | Model | < 0.0001 | 0.0017 |
| A | < 0.0001 | 0.0005 | |
| B | < 0.0001 | 0.0444 | |
| (a) | |||
| V (cm3) | AV | ||
| R2 | 0.93 | 0.91 | |
| Corrected R2 | 0.91 | 0.84 | |
| (b) | |||
Quadratic function in terms of V and linear function in terms of AV was approved (p < 0.05). Lack of fit was identified as insignificant for both properties (p < 0.05). When the effects of independent factors on the V and AV were taken into account (Table 2), the influences of chickpea sourdough and baking powder use in terms of cake V and AV emerged as statistically significant (p < 0.05). Additionally, the model was also determined to be statistically significant (p < 0.05). The final equation with factors was coded (A representing chickpea sourdough and B representing baking powder) as it follows:
 |
 |
The purpose of optimization is to determine accurate values for parameters which will help to obtain the desired response. The numerical optimization was used in this study. The V and AV were evaluated together, and the first solution was selected from the three solutions offered by the software. According to the solution, amounts of use 64.02 g for A, 5.97 g for B were offered for obtaining 1249.74 cm3 V and 88.37 AV values.
Figure 1 shows the response to the interaction for the level of the A and B. The V value increases in line with the levels of the A with B. Furthermore, the AV increases in line with the level of the A and B. However, the AV exhibits a tendency to decrease after the central point of increase of these independent variables. The insignificant internal interaction between the level of the A and B can be seen through the circular curves at the bottom of the second graph in Figure 1.
Response surface plot showing the mutual effect of the amount of chickpea sourdough (A) and baking powder (B) on Volume (V) and Acceptability Values (AV)
Experimental validation of optimization results Following optimization, the levels of usage were determined as 64.02 g for A and 5.97 g for B. The cake was produced using optimized amounts in three replicates. The V and AV values of OC were determined, and the mean value was calculated. The presence of any statistically significant (p < 0.05) difference between the mean value and the estimated values from the model was investigated using the one sample t–test. The one sample t–test results for each response are given shown in Table 3. There wasn't found any statistically significant difference (p < 0.05) between the results obtained from the validation tests and the estimated values of the V or AV. This indicates that the model obtained by optimization was also experimentally successful.
| Response | Estimated value | Average experimental result* | Difference | p-value** |
|---|---|---|---|---|
| V (cm3) | 1249.74 | 1283.33±15.28 | 34.00 | 0.060 |
| AV | 88.37 | 90.67±1.53 | 1.53 | 0.121 |
V; volume, AV; acceptability value.
Properties of cake samples (OC and CC) The cake samples OC and CC were compared in terms of various physicochemical and sensory properties.
Pore structure The pore structure of samples were examined by the image analysis (Figure 2), which revealed that while the OC had 989 pores per unit area, the CC had 894 pores. The pore number of per unit area increased with the use of chickpea sourdough. The increase in pore numbers per unit area was the result of the formation of smaller pores in OC.
Pore structure images obtained by Image Pro Plus software of cake samples (OC; optimized cake, CC; control cake)
The total pore numbers for the first three classes were 953 for the OC and 853 for the CC samples. In addition, whereas the total number of large pores in Class 4 and Class 5 was 36 (3.6 %) for the OC, it was only 41 (4.59 %) for the CC (Figure 3). Only B was responsible for leavening and pore formation of texture for the CC sample, whereas A and B were together responsible for the OC sample. In other words, the biological and the chemical leavening reactions that occurred in the OC sample were much more pronounced. The microbiota of A contains homo and heterofermentative LAB (Sıkılı, 2003). The heterofermentative strains can produce CO2 in addition to the other metabolites. The pore number of per unit area in the OC samples was therefore higher than that in the CC.
Distribution of pore numbers according to image analysis (OC; optimized cake, CC; control cake)
Physicochemical properties The pH and TTA of chickpea sourdough in the dough stage were 4.33 and 10.40, respectively. While the usage of A caused significant reduction in cake pH, it caused an increase in TTA (p < 0.05)(Table 4).
| Properties | OC | CC | p-value** |
|---|---|---|---|
| pH | 7.30 ± 0.05 | 8.37 ± 0.06 | 0.001 |
| TTA | 1.55 ± 0.052 | 0.65 ± 0.049 | 0.001 |
| Volume (cm3) | 1283.33 ± 15.28 | 1260 ± 3.50 | 0.118 |
| Specific volume (cm3/g) | 2.23 ± 0.02 | 2.39 ± 0.03 | 0.005 |
| Crust thickness (mm) | 1.85 ± 0.39 | 0.98 ± 0.16 | 0.061 |
| Crust | |||
| L* | 42.75 ± 1.12 | 45.83 ± 2.38 | 0.013 |
| a* | 17.84 ± 0.34 | 16.53 ± 0.27 | 0.022 |
| b* | 23.07 ± 2.73 | 24.91 ± 2.85 | 0.363 |
| Crumb | |||
| L* | 72.89 ± 1.89 | 71.43 ± 1.08 | 0.313 |
| a* | −1.60 ± 0.12 | −1.96 ± 0.13 | 0.035 |
| b* | 26.55 ± 1.18 | 24.73 ± 0.59 | 0.116 |
| Inhibition (%) | 29.65 ± 1.4 | 11.83 ± 1.6 | 0.002 |
Mean ± standard deviation,
OC; optimized cake, CC; control cake, TTA; total titratable acidity
The B generally have alkaline or neutral profile due to their mix contained the different chemicals such as NaOH. For that reason the cake batters have able to slightly alkaline profile generally like CC sample. Since the CC samples were produced by using B only, the pH was higher than that in the OC, and the TTA was lower. On the other hand, the sourdough has quite acidic profile (approx pH 4.5). In this study, when the sourdough added to cake batter, the cake has gained neutral profile (or in other words more acidic than CC sample). The acidic composition of the A together with the alkaline profile of B caused more neutral profile in the OC samples. The studies about the microbiota of A have shown that it contains microorganisms such as Enterococcus spp., Lactobacillus spp., Streptococcus spp., Enterococcus spp., Pediococcus spp. and Saccharomyces cerevisiae (Sıkılı, 2003). The LAB synthesize lactic acid by homofermentation of hexoses and synthesize lactic acid, acetic acid, ethanol and CO2 by heterofermentation of hexoses. A reduction in pH causes an increase in the protease activity of cereals. Additionally, the LAB also have their own enzyme activity. The increasing in enzyme activity raises the free amino acid content by hydrolyzing proteins (Hansen and Schieberle, 2005), therefore it may lead resulting higher TTA value. In a study that used three heterofermentative (L.sanfrancisco, L.brevis, L.fermentum) and two homofermentative LAB (L.delbrueckii, L.plantarum), Hansen et al. (1989) determined that TTA was increased by the heterofermentative LAB.
The crust thickness of OC was found significantly higher than the CC (p < 0.05). Although, the volume increased in the OC by the usage of A because the weight of OC was higher, the specific volume was found lower than the CC. Moreover, while the difference between the volumes of the samples wasn't found statistically significant (p < 0.05), the difference between the specific volumes was found significant (p < 0.05). While only B used in CC sample was responsible for the gas production, the B as chemically and the heterofermentative LAB as biologically produced CO2 and provided pore structure in the OC. The enhancing effects of the A on the cake's volume may have been due to various factors. The gluten structure in acidic doughs that contain sourdough has a better gas retention capacity (Gobbetti et al., 1995). This is caused by the solubility of pentosans during the sourdough process (Corsetti et al., 2000) and by the increase in endogenous enzyme activities through lower pH related to the use of sourdough (Clarke et al., 2003). The use of the A resulted in the positive effects on volume and specific volume. The pore structure analysis revealed a higher pore number per unit area in the OC samples.
According to the color measurements for the cake crust, L* value decreased and a* value increased significantly (p < 0.05) by the usage of the A, although any change was not obtained in the b* value (p < 0.05) (Table 4). During the baking stage in the oven, the water vapor moves from crumb to crust, particularly in the “oven rise” stage. The water carries fats/oils emulsified with it toward the cake crust. While the vapor leaves the structure, the oil stays on the surface during the baking process. The moisture analysis showed that the water was retained in the crumb structure. This may have been promoted by the thicker crust formation in the OC samples. In other words, less oil/fat transfer dependent on water vapor occurred for the CC samples. It is likely that the lower L* value in the crust is related to this phenomenon.
The color measurement results showed that a* value increased in the cake crumb by the use of A (p < 0.05) (Table 4). It caused a rise in a* values not only in the cake's crumb, but also in the cake's crust. The color is one of the most important factors affecting the consumer's preferences for the bakery products. The formation of the yellow-gold-brown colors often described as browning, is due to non-enzymatic chemical reactions specifically the Maillard and caramelization reactions produce colored compounds during the baking process (Artan et al., 2010). When the bakery products containing reducing sugars and amino groups are heated, the caramelization and the Maillard reaction may take place simultaneously (Purlis and Salvadori, 2009). The participation in chemical browning reactions of the simple sugars produced at the end of the sourdough fermentation may have resulted in an increase in the a* value of the crust in particular. A decrease in pH during fermentation positively affects the browning reactions, probably because the Amadori rearrangement requires H+. The duration of the fermentation has been reported to influence the color of the bakery product through the formation of the various compounds as precursors of brown pigments (Martinez Anaya, 1996).
The moisture contents were 33.08 % and 28.98 % for the OC and CC samples, respectively (Figure 4). The reduction of the moisture was lower for the OC samples during storage possibly corresponding a facilitated moisture retention by the usage of chickpea sourdough. For bakery products such as cake, not only high initial moisture, but also moisture retention during shelf life is important. The retrogradation process known as staling is also related to moisture loss. Additionally, the exopolysaccharide production of LAB has been thought to play a role in moisture retention (Galle and Arendt, 2014). The conversion of starch into simple sugars as a result of fermentation may also affect moisture retention. The crust thickness was greater in the OC samples than in the CC. The thick crust that formed may have led to moisture retention inside the crumb. The gas retention capacity of gluten is known to increase in acidic dough (Gobbetti et al., 1995) and the modification of gluten caused by chemical reactions and biological action of microorganisms may together have a function for the retention of moisture in the cake structure.
Moisture loss during shelf life of the cake samples (OC, optimized cake; CC, control cake)
DPPH radical scavenging activity DPPH radical inhibition (%) value of OC sample was significantly (p < 0.05) higher than the CC's value (Table 4). The results obtained in this study shown that the usage of A enhanceed the antioxidant activity of the cake. The antioxidant activity of the baked goods are affected by the various factors and the phytochemicals mainly altere in during processes such as fermentation. The increase in antioxidant activity resulting from sourdough fermentation can be explained by different biochemical and metabolic events. The metabolisms of LAB in sourdough facilitate lipid oxidation during fermentation. It is known that homofermentative lactobacillus increase the lipid oxidation (Vermeulen et al., 2007). The formation or modification of the bioactive compounds during sourdough's fermentation can improve the antioxidant activity of the cake. Several studies shown that some LAB were capable of releasing antioxidant peptides during sourdough fermentation (Rizzello et al., 2012). In a study conducted by Coda et al. (2012), it was ascertained that some selected LAB produced antioxidative peptides during sourdough fermentation. When acidity increases, it is expected an increase in the total phenolic content and a decrease in tiamin, ferulic acid dehydromers, tocoferoles and tocotrienoles (Hansen et al., 2002). However, the mechanism of these changes in bioactive component levels during sourdough fermentation has not yet been explained.
Sensory profile The radar graphs (Figure 5) clearly showed the difference of sensory profiles between the cake samples.
Sensory characteristics of the cake samples indicated with radar graphs (OC, optimized cake; CC, control cake)
Oil and egg aromatic profiles in cake are not preferable by consumers. The use of the A led to less perception of fat and egg aromatic profiles in the OC samples. A more malty character was found in the CC than in OC samples. The panelists also reported that acidic and sweetish characters were more dominant in the OC than the CC sample. Although the A was a kind of the sourdough type, it gave sweet flavor/aroma to the cake sample. This reason may be that, the starch/simple sugar conversions occured during chickpea sourdough fermentation. Moreover, the chickpea sourdough is also known as “sweet yeast. The different flavor and aroma compounds (especially carbonyl compounds such as propionaldehyde, n-butyl aldehyde, ethyl butyl ketone), and some organic acids (such as acetic acid, propionic acid, valeric acid, crotonic acid) form in the bakery products which contain the A. This effect depended on the microbiota of A and their metabolites. Therefore, the products used A gains characteristic flavor and sweety taste (Özkaya, 1992; Sikili, 2003). Fermentation of sugars by microbiota leads to a large number of volatile compounds that are responsible for the distinctive characteristics associated with the flavor of bakery products. Moreover, effects of sourdough on flavor are based on three main factors: i) formation of acidity, ii) formation of flavor precursors such as amino acids and iii) formation of volatile compounds (Tulbek, 2006). The acidic flavor in the OC samples may result from the organic acids produced by the LAB during the fermentation process.The impact on the flavor depends on acidity level, level of free amino acids and important flavor compounds in sourdough. The formation of acidity has a profound effect on flavor. Furthermore high amount of free amino acids in sourdough has been linked to an enhanced intensity of the overall flavor of bakery products. (Martinez Anaya, 1996). It was determined that The OC samples exhibited higher scores in terms of overall taste/flavor characteristics.
Furthermore some panelist evaluated that while the OC sample was more sweet, the CC sample was more fatty. Because the basic formulation were same in both sample OC and CC, this fact make thought that the evaluation of panelists about taste of the samples were completely sensorial.
The use of the A with B in the cake, improved the pore structure and physicochemical properties, and enriched the sensory profile, in this study. It also improved the crust color of the cake as well as its volume and its pore structure. Following optimization, the levels were 64.02 g for the chickpea sourdough and 5.97 g for the baking powder, while the control sample contained 9.33 g of baking powder. The results suggest that the A could be applicable for other chemically leavened bakery products. Moreover, it could be used as a natural ingredient, as a partial replacement for the baking powder. The traditional A production method used in this study was a typical multistage fermentation, which combines a soaking stage and lactic acid fermentation. This may have had the potential to reduce antinutrient factors and to improve bioavailability. Therefore the bioavailability of proteins, minerals and other nutritional properties can be examined in future studies. If the traditional A production can be adopted for the industrial level, it can serve as an alternative for the satisfaction of the today's consumer's demands towards additive free, functional bakery products. According to the findings of this study, further detailed exploration of the beneficial and potential health improving properties of the A is now required.
