IJFS#147_WAHYONO_bozza   Ital. J. Food Sci., vol 28, 2016 - 298 PAPER IMPROVING BREAD QUALITY USING CO-CULTURES OF SACCHAROMYCES CEREVISIAE, TORULASPORA DELBRUECKII JK08, AND PICHIA ANOMALA JK04 AGUNG WAHYONO1, SAE-BYUK LEE1, WOO-WON KANG2 and HEUI-DONG PARK*1 1School of Food Science and Biotechnology, Kyungpook National University, Republic of Korea 2Department of Food & Food-service Industry, Kyungpook National University, Republic of Korea *Corresponding author: hpark@knu.ac.kr   ABSTRACT Co-cultured Saccharomyces cerevisiae, Torulaspora delbrueckii JK08, and Pichia anomala JK04 were used as a leavening agent. The leavening ability, crumb structure, texture profile, crumb aroma, and sensory properties of bread were evaluated. The leavening ability of the co-cultures tested was lower than for S. cerevisiae alone. Leavening containing a co- cultures produced bread crust and crumb that were slightly yellow in colour and bright. Generally, co-cultured bread produced a larger diversity and higher abundance of volatile organic compounds. Superior colour properties, favorable aroma, and decent textural and structural features resulted in higher sensorial ratings for co-cultured bread. We suggest the use of co-culture as leavening agents for improved bread quality. Keywords: bread quality, co-culture, Pichia anomala, Saccharomyces cerevisiae, Torulaspora delbrueckii   Ital. J. Food Sci., vol 28, 2016 - 299 1. INTRODUCTION Bread is a food that not only has a long history, but also a long future. Over 12,000 years ago, the first bread was probably developed by deliberate experimentation using wheat flour and water. Product development and process innovation in bread making is still an active field. Consumer interest in bringing unique and alternative breads to the table is driving the production of a wide variety of breads (CAUVAIN, 2012; MONDAL and DATTA, 2008). A multitude of studies conducted over the 20th century focused on improving bread quality and have explored the immense terrain of recipes (GARDNER et al., 2002; ROSELL et al., 2009; MOVAHED et al., 2012; NILUFER-ERDIL et al., 2012), process innovations (KARAOĞLU et al., 2006; BOSMANS et al., 2013) and the use of novel microorganisms as leavening agents (CABALLERO et al., 1995; PLESSAS et al., 2005; CHOI and CHOI, 2009; MO and SUNG, 2014). Although the experience of bread quality is highly personal, it may be described as the sum of the sensory pleasures associated with flavour, taste, texture, and appearance (CAUVAIN, 2012). The technological role of yeast in bread making has been well established. The main role of yeast is to produce gas by degrading the sugars available in the flour or that are added to the recipe. The gas produces air bubbles that are contained by the stretchy gluten proteins in the flour, which causes the dough to rise and produces an aerated structure in the resulting bakery product. During fermentation, the yeast produces abundant alcohols and other volatile compounds that impart unique tastes and flavours to the bread (STEAR, 1990; GÉLINAS, 2009; ALI et al., 2012; PACHECO et al., 2012). Over the last two decades, there has been great interest in using new strains of baker’s yeast that produce particular aromas, anti-molding properties, or desirable nutritional characteristics to fulfill the needs of the baking industry (GÉLINAS, 2009). Furthermore, the use of co-cultures as leavening agents has been reported to confer favorable effects in baking products. Examples include the use of Lactobacillus plantarum or Pediococcus cerevisiae, combined with Saccharomyces cerevisiae, as a co-culture for improving bread quality and delaying staling (ELHARIRY et al., 2011); employing lactic acid bacteria and S. cerevisiae as a starter to enhance the nutritional content and shelf life of cassava-wheat bread (OGUNBANWO et al., 2008); the use of Lactobacillus delbrueckii ssp. bulgaricus or Lactobacillus helveticus mixed with the lactose-fermenting yeast Kluyveromyces marxianus to improve bread quality and increase shelf-life (PLESSAS et al., 2008); and the use of starter culture combinations of Lactobacillus fermentum, S. cerevisiae, and Candida krusei to enhance aroma in Ghanaian maize dough fermentation (ANNAN et al., 2003). Torulaspora delbrueckii and Pichia anomala have unique traits that improve the quality of bakery products. As reported by HERNANDEZ-LOPEZ et al. (2003), T. delbrueckii strains IGC5321 and IGC5323 exhibited higher leavening ability and CO2 production than S. cerevisiae after exposure to hyperosmotic and freeze-thaw stress. MO and SUNG (2014) reported that P. anomala SKM-T enhanced flavour properties and extended shelf life. The use of T. delbrueckii, P. anomala, and S. cerevisiae as co-cultured leavening agents has not yet been reported. Here, we investigated the effects of co-cultured S. cerevisiae, T. delbrueckii JK08, and P. anomala JK04 as a leavening agent on bread quality. Leavening ability, structural crumb features, texture profile, crumb aroma, and sensory properties of bread were evaluated.   Ital. J. Food Sci., vol 28, 2016 - 300 2. MATERIALS AND METHODS 2.1. Microorganism strains and bread ingredients T. delbrueckii JK08 (TD) and P. anomala JK04 (PA) were isolated from Korean traditional starter (Nuruk), and then identified and collected at the Institute of Fermentation Biotechnology, Kyungpook National University, Daegu, South Korea. The strains were sub-cultured in yeast extract-peptone-dextrose (YPD, Difco, Le Pont de Claix, France) agar (Oxoid, Hampshire, United Kingdom) and stored at 2-4°C. S. cerevisiae (SC) was isolated from compressed instant baker’s yeast (saf-instant S.I., Lesaffre, Marcq, France). Commercial wheat flour (Beksul, CheilJedang, Seoul, South Korea) contained 13.93% protein and 0.47% ash, with a pH of 4.35. Sugar and salt were purchased from a supermarket in Sangju, South Korea. 2.2. Preparation of cultures Preparation of cultures was described previously (WAHYONO et al., 2015). The strains were sub-cultured in 5% YPD broth and cultivated in a rotary shaker (JSSI-300C, JS Research, Gongju, South Korea) at 30°C for 48 h, with shaking at 180 rpm. The yeast cells were collected by centrifugation (Hanil Supra 22K, Hanil, Incheon, South Korea) at 4000 × g for 10 min at 4°C. The supernatant was discarded, and the pellet was re-suspended in distilled water, vortexed thoroughly, and stored at 4 °C. Cell density was measured using a hemocytometer (Neubauer chamber, Celeromics, Cambridge, United Kingdom). To measure cell density, 10-1 to 10-4 dilutions of culture were prepared; 10 μl of diluted culture was loaded into the hemocytometer chamber. The chamber was observed at 100× magnification using a binocular microscope (CX31RTSF, Olympus, Tokyo, Japan). Cells in four different regions of the chamber were counted. Cell density was calculated using the following formula: 𝑐 = ! !·!  ×  10  000 (1) where c is the concentration (expressed as number of cells per millilitre), N is the number of cells, n is number of squares counted, and d is the dilution factor. Number 10 000 represents a constant converter to millilitre. 2.3. Leavening ability The leavening ability of co-cultures was measured as described previously (WAHYONO et al., 2015). Bread dough containing 20 g flour, 20 ml water, and 4 × 108 yeast cells per ml of water was mixed thoroughly in a 100-ml graduated cylinder and then incubated at 30°C for 210 min. This was done in triplicate and each sample was observed every 30 min. The maximum leavening rate (ml/h) was calculated from the highest volume reached in 210 min divided by the time the highest volume was first recorded. 2.4. Bread-making procedure Baking was carried out in a bread maker (National SD-BT102, Panasonic, Osaka, Japan) using the 4-h standard bread-making setting. The basic recipe consisted of wheat flour, salt, sugar, and drinkable water. A single culture SC and co-cultures of S. cerevisiae + T.   Ital. J. Food Sci., vol 28, 2016 - 301 delbrueckii JK08 (SCTD), S. cerevisiae + P. anomala JK04 (SCPA), and S. cerevisiae + T. delbrueckii JK08 + P. anomala JK04 (SCTDPA) were used as leavening agents. The working ratio for co-cultures of S. cerevisiae + T. delbrueckii JK08 (SCTD) and S. cerevisiae + P. anomala JK04 (SCPA) were 50:50, respectively. The co-culture ratio for S. cerevisiae + T. delbrueckii + P. anomala JK04 (SCTDPA) was 30:35:35. In the bread dough, the SC cell density was adjusted to approximately 5 × 108 cells per ml of water added to the dough. For TD and PA, the cell densities were approximately 109 cells per ml of water added. Dough compositions are given in Table 1. The bread maker protocol included a first mixing for 20 min, resting for 25 min, and a second mixing for 10 min, followed by a first fermentation at 35°C for 50 min. A third mixing was performed for 3 min, followed by a second fermentation at 35°C for 40 min, and then a fourth mixing for 2 min. Proofing was done at 40°C for 50 min and finally, loaves were baked at 210°C for 40 min. Baking was performed in triplicate. After baking, the bread loaves were tempered at ambient temperature (28-30°C) before analysis. 2.5. Moisture content, specific volume, and bread yield efficiency The moisture content of the bread crumb was determined by the oven-drying method (CZUCHAJOWSKA et al., 1989). Specific volume was determined by the seed displacement method (AACC 10-05, 2000). Bread yield efficiency was calculated as described by MOVAHED et al. (2012) using the following formula: 𝑃! = !! !! ×  100 (2) where P2 is bread yield efficiency, W3 is bread weight and W2 is flour weight. 2.6. Chromaticity of bread crust and crumb The crust and crumb colour were measured using the CR-400 Chroma Meter (Konica- Minolta, Tokyo, Japan); L (lightness), a (redness), and b (yellowness) values (Hunter colour) were measured for six regions of bread crust and crumb. The whiteness index (WI) was calculated according to HSU (2003) and LIN et al. (2009). 2.7. Texture profile analysis Texture profile analysis (TPA) was carried out in triplicate for two slices of bread. TPA was performed using a texture analyzer (CT3 4500, Brookfield, Middleboro, USA). Bread samples were sliced to approximately 25-mm thickness. Hardness, cohesiveness, springiness, and chewiness of the center of the bread slices were measured (BLANDINO et al., 2013). The settings and conditions were carried out as described previously by ULZIIJARGAL et al. (2013), with some modifications: the acrylic cylindrical probe had a 38.1 mm diameter (TA4/1000), the pretest speed was 2 mm/s, the test speed was 2 mm/s, the post-test speed was 2 mm/s, the distance was 10 mm (40% compression), and the trigger load was 50 g. 2.8. Bread crumb image analyses The structural features of the bread crumb were analyzed using ImageJ software (1.47v, National Institutes of Health, Bethesda, MD, USA). Structural features included bread cell   Ital. J. Food Sci., vol 28, 2016 - 302 density, mean cell area, and the fraction of cell area to total area. Bread crumb images were captured using a scanner (Epson Perfection V370 Photo, Epson, Japan) at a resolution of 800 × 800 dpi. Images were calibrated to reflect actual size using a known scale, were cropped to 60 × 60 mm, filtered using a bandpass filter, and converted into binary images using the convert to mask feature for differentiating between the cell and non-cell area. Before analyzing particles (cells), the particle size was set from 0.01 mm2 to infinity and the circularity was set from 0 to 1. This particle size corresponds to a particle diameter of 0.1 mm, which can be resolved by the human eye (PONGJARUVAT et al., 2014). 2.9. Analysis of volatile compounds in bread crumbs Volatile compounds were analyzed as described previously by PLESSAS et al. (2008), using gas chromatography and mass spectrometry (7890A GC-MS; Agilent, Santa Clara, CA, USA) with a flame ionization detector (FID). The separation was performed with a DB- WAX column (60 m × 250 μm × ɸ 0.25 mm) (Waters, Milford, MA, US). The detector was an Agilent 5975C Inert XL MSD with Triple-Axis Detector. Helium was used as a carrier gas with a constant flow of 1 ml/min. Using solid-phase microextraction technique (SPME), 1 g of each bread sample was put into a 20-ml vial accessible to the SPME needle through the vial septum. Then, the vial was submerged in a water bath at 60°C and the SPME fiber (50/30 μm DVB/CAR/PDMS, Supelco, Bellefonte, PA, USA) was exposed to the headspace of the vial for 60 min. When the extraction process was finished, the SPME fiber was inserted into the injector port (set at 280°C) of the gas chromatograph (GC) for thermal desorption of volatile compounds for 5 min in splitless mode. The GC temperature program was set as follows: 35°C for 5 min, increased by 5°C/min to 50°C (held for 5 min), increased by 5.5°C/min to 230°C (held for 5 min). Volatile compound identification was based on comparison of GC retention times and peak areas with spectral data from the Wiley9Nist 0.8 library (Wiley9Nist 0.8 library, mass spectral search program, version 5.0, USA). 2.10. Sensory evaluation Bread samples were prepared from a freshly baked loaf (6-8 h after baking). The bread samples were cut about 50 × 20 × 25 mm and served on a small paper plate. Sensory evaluations were conducted by 15 semi-trained consumers who were students and professors at Kyungpook National University, South Korea. The sensory attributes tested included appearance, colour, flavour, mouthfeel, and overall acceptability. The sensory attribute scale used for assessing the bread was as follows: 1, extremely dislike; 4, neither like nor dislike; and 7, extremely like (ULZIIJARGAL et al., 2013). 2.11. Statistical analysis To examine statistical significance, the data were analyzed using one-way analysis of variance (ANOVA) followed by Duncan’s multiple range test at the p < 0.05 level of significance. The correlation among structural features and textural profiles of bread crumb was analyzed using a 2-tailed Pearson correlation at p < 0.05. The analysis was carried out using SPSS for windows (ver. 19, IBM, New York, New York, USA). The graphs were constructed using Microsoft Excel (2007v, Microsoft, Redmond, Washington, USA).   Ital. J. Food Sci., vol 28, 2016 - 303 3. RESULTS AND DISCUSSION 3.1. Leavening ability The leavening ability of co-cultures was compared to that of single cultures (SC) in lean dough containing wheat flour and water (Fig. 1). The leavening rates were significantly different among the cultures tested (p < 0.05). For SC, the dough was greatly leavened after only 30 min of incubation, but for co-cultures, the dough was greatly leavened after 60 min of incubation. The leavening rates for SC, SCTD, SCPA, and SCTDPA were 55.4, 41.6, 36.5, and 31.6 ml/h, respectively. In a previous study, we found that the leavening rate for single cultures of T. delbrueckii JK08 and P. anomala JK04 were 8.67 and 2.29 ml/h, respectively. The lower performance of T. delbrueckii JK08 and P. anomala JK04 may be due to slower growth relative to S. cerevisiae (WAHYONO et al., 2015). BELY et al. (2008) also reported lower performance for T. delbrueckii 27828 and T. delbrueckii 31703 in terms of fermentation rate in comparison to S. cerevisiae. On the contrary, HERNANDEZ-LOPEZ et al. (2003) reported that T. delbrueckii strains IGC5321 and IGC5323 exhibited higher leavening ability and CO2 production than S. cerevisiae after exposure to hyperosmotic and freeze-thaw stress. We demonstrated that incorporating TD and PA with SC produced a co-culture with greatly improved leavening ability for a longer leavening period (> 120 min). ELHARIRY et al. (2011) revealed that incorporating S. cerevisiae and L. plantarum in a sourdough system delivered favorable effects such as improving the leavening ability, as well as the sensory and physical properties of the bread. Table 1: List of ingredients used in bread making. Ingredients Quantity Yeast addition (total cells)a S. cerevisiae T. delbrueckiiJK08 P. anomalaJK04 Wheat flour (g) 280 Sugar (g) 16.8 Salt (g) 5.6 Water (ml) 200 Leavening agent; SC 1 × 1011 - - SCTD 5 × 1010 1 × 1011 - SCPA 5 × 1010 - 1 × 1011 SCTDPA 3 × 1010 7 × 1010 7 × 1010 aCalculated according to the yeast concentration and ratio determined in the bread-making procedures in the materials and methods.   Ital. J. Food Sci., vol 28, 2016 - 304 Figure 1: Leavening rates for co-cultures or a single culture in lean dough. Results are means±SD of triplicate. 3.2. Physical properties The co-cultures did not affect the moisture content or bread yield, but significantly affected the bread-specific volume (p < 0.05, Table 2). The specific volumes of breads produced with the cultures tested were in the range of 4.01-4.37 cm3/g. In comparison to breads leavened with co-cultures, the bread leavened with SC produced the greatest specific volume (4.37 cm3/g), which is consistent with the observation that SC exhibited the highest leavening activity (55.4 ml/h). We previously reported that lower leavening abilities for T. delbrueckii JK08 and P. anomala JK04 produced significantly lower specific volumes in comparison to S. cerevisiae (WAHYONO et al., 2015). These results strongly suggest a connection between the yeast’s leavening ability and the specific volume of the resulting bread. No significant differences were observed in bread-specific volume for the co-cultures tested. Even though there were significant differences in leavening rates for the co-cultures tested, the time period for fermentation in the bread-making process was not long enough for optimum leavening activity for SCTD, SCPA, SCTDPA (120, 150, and 180 min, respectively). However, the specific volume of breads produced by co-cultures is well within the range of standard bread, with specific volumes that ranged from 3.5 to 6 cm3/g (ULZIIJARGAL et al., 2013). The co-cultures significantly affected the chromaticity of the bread crust (Table 3). SCTD- leavened bread showed the greatest L, b, and WI values. The L value for SCTD was significantly greater than that for SC or SCPA. The b value for SCTD was only significantly greater than that for SC. The WI value for SCTD was significantly higher than those for the other cultures tested. SCTDPA produced the highest a value, which was significantly greater than that of SC or SCTD. In contrast, SC-leavened bread produced the lowest L, a, b, and WI values. These results were consistent with previous work that demonstrated greater L and b values for bread crust when T. delbrueckii JK08 was used as a leavening agent (WAHYONO et al., 2015). For crumb colour, the co-cultures significantly affected the L and b values, but not the a and WI values (Table 3). SCPA produced the greatest L value, which was significantly greater than that of SC. It also produced the greatest b value, which was significantly higher than that of SC or SCTD. The greatest L and b values of bread crumb arose when P. anomala JK04 was used as a leavening agent, while lower L and 0,0   10,0   20,0   30,0   40,0   50,0   60,0   70,0   80,0   90,0   100,0   0   30   60   90   120   150   180   210   V ol um e    ( m l)   Time  (min)   SC   SCTD   SCPA   SCTDPA     Ital. J. Food Sci., vol 28, 2016 - 305 b values (darker colour) were produced when S. cerevisiae was used (WAHYONO et al., 2015). In short, TD produced greater lightness and unsaturated yellowish colour in the bread crust, and PA produced similarly a coloured bread crumb. On the other hand, SC produced a darker and saturated yellowish colour in bread crust and crumb. HERNANDEZ-LOPEZ et al. (2003) reported that commercial baker’s yeast (S. cerevisiae) exhibited greater maltase and invertase activity than T. delbrueckii IGC5321. Consequently, SC produced more reactive saccharides, which may contribute to darker colour formation. The saccharides and nitrogen-containing substances involved in the browning reaction create the dark-coloured pigment melanoidin, which confers a darker colour to the bread crust and crumb (STEAR, 1990). Table 2: Physical properties of bread leavened with co-cultures or a single culturea. Yeast Crumb moisture content (%) Specific volume (cm3/g) Bread yield efficiency (%) SC 46.33±0.50A 4.37±0.15A 147.97±0.75A SCTD 46.58±0.59A 4.08±0.05B 148.86±1.00A SCPA 46.83±0.35A 4.15±0.01B 147.85±0.21A SCTDPA 46.61±0.16A 4.01±0.09B 148.13±0.75A aMeans with the same superscript letter in a column are not significantly different at the level p < 0.05. Table 3: Chromaticity of bread crust and crumb leavened with co-cultures or a single culturea. Yeast Crust colour Crumb colour L a b Whiteness index (WI) L a b Whiteness index (WI) SC 38.22±1.22C 6.26±0.72C 15.85±0.97B 35.90±0.88B 56.85±2.88B -2.52±0.11A 8.53±0.78C 55.93±2.76A SCTD 41.63±0.93A 6.92±0.67BC 17.77±0.24A 38.59±1.00A 59.08±0.11AB -2.56±0.05A 9.25±0.38BC 57.97±0.03A SCPA 39.39±0.34BC 7.79±0.26AB 17.00±0.28A 36.57±0.35B 60.22±0.75A -2.42±0.02A 10.34±0.13A 58.82±0.71A SCTDPA 40.12±0.83AB 8.23±0.22A 17.66±0.46A 37.02±0.64B 58.37±1.55AB -2.43±0.10A 9.58±0.47AB 57.21±1.40A aMeans with the same superscript letter in a column are not significantly different at the level p < 0.05. 3.3. Structural features of bread crumb The structural parameters of bread crumb are expected to influence its mechanical behavior. By using image analysis, the structural features of bread crumb can be quantified (ZGHAL et al., 2002). Hence, we carried out an image analysis of bread crumb and the results are presented in Table 4. The digital binary images of bread crumb from which the structural features could be extracted are shown in Fig. 2. The use of co-cultures significantly affected the cell density and the mean cell area of bread crumb, but not the fraction of cell area to total area. The cell density of bread crumb leavened with SCTD (58.89 1/cm) was the greatest among cultures tested, and was significantly greater than that leavened with SC (50.04 1/cm). Inversely, the mean cell area of bread crumb leavened with SC (0.90 mm2) was the largest, significantly larger than that leavened with SCTD (0.73 mm2). These results suggest that as mean cell area increases, cell density decreases. The   Ital. J. Food Sci., vol 28, 2016 - 306 large mean cell area for SC-leavened bread is consistent with its high specific volume. In other words, bread leavened with SC was more porous than bread produced using co- cultures, probably because of superior leavening ability and higher CO2 production (WAHYONO et al., 2015). Alternatively, PONGJARUVAT et al. (2014) reported that high specific volume is tightly correlated with cell density and cell area fraction. We performed correlation analysis for structural features and mechanical parameters (TPA) of bread crumb. We found that the cell density, mean cell area, and fraction of cell area to total area were correlated with cohesiveness, but not hardness, springiness, or chewiness (Table 5). Structural features were not strongly correlated with overall bread quality, which is more strongly affected by attributes such as odor and appearance (LAMPIGNANO et al., 2013). Figure 2: The digital binary images of breads crumb leavened with co-cultures or a single culture. Table 4: Structural features of bread crumb leavened with co-cultures or a single culture quantified by image processinga. Yeast Cell density (1/cm2) Mean cell area (mm2) Fraction of cell area to total area (%) SC 50.04±5.59B 0.90±0.10A 44.87±0.40A SCTD 58.89±2.92A 0.73±0.03B 42.83±2.06A SCPA 52.55±3.37AB 0.85±0.09AB 44.34±1.87A SCTDPA 53.38±3.59AB 0.84±0.04AB 44.57±1.03A aMeans with the same superscript letter in a column are not significantly different at the level p < 0.05.   Ital. J. Food Sci., vol 28, 2016 - 307 Table 5: Pearson correlation of structural features and textural profiles of bread crumb. Hardness Springiness Cohesiveness Chewiness Cell density Mean cell area Springiness 0.360 Cohesiveness -0.485 -0.391 Chewiness 0.981** 0.353 -0.319 Cell density 0.450 0.365 -0.633* 0.380 Mean cell area -0.435 -0.432 0.732** -0.340 -0.950** Cell fraction -0.175 -0.326 0.658* -0.043 -0.273 0.541 **Correlation is significant at the 0.01 level (2-tailed). *Correlation is significant at the 0.05 level (2-tailed). 3.4. Texture profile analyses TPA was used to evaluate the textural properties of bread crumb (Fig. 3). The co-cultures altered hardness, chewiness and cohesiveness, but these changes were insignificant, except for cohesiveness. SCTD-leavened bread was of lower cohesiveness than bread leavened with other cultures (Fig. 3), consistent with our previous study, demonstrating that T. delbrueckii JK08 produces bread crumb with lower cohesiveness (WAHYONO et al., 2015). As stated earlier, the cohesiveness of bread crumb was the only parameter that correlated with its structural features. According to SCANLON and ZGHAL (2001), the crumb textural properties were largely determined by the bread crumb structural features. Fine and uniformly-sized cells produce a softer texture. Here we have demonstrated that a greater mean cell size conferred greater cohesiveness and vice versa. However, this result should be further evaluated in light of previous work that established that crumb cohesiveness is controlled by moisture content and the strength of networks surrounding the cell pore (CAUVAIN, 2004). We have shown that the use of co-cultures produced bread of quality and textural properties comparable to bread leavened using a single culture. Figure 3: Texture profiles of bread leavened with co-cultures or a single culture. (A) Hardness and chewiness of bread crumb. (B) Springiness and cohesiveness of bread crumb. Results are means ± SD of triplicate. 0   200   400   600   800   1000   1200   SC   SCTD   SCPA   SCTDPA   H ar dn es s/ Ch ew in es s   (g )   Leavening  Agent   Hardness   Chewiness   A 0,00   0,20   0,40   0,60   0,80   1,00   1,20   SC   SCTD   SCPA   SCTDPA   Sp ri n gi n es s/ Co he si ve n es s   Leavening  Agent   Springiness   Cohesiveness   B *   Ital. J. Food Sci., vol 28, 2016 - 308 (*) Significantly different at the level p < 0.05. 3.5. Volatile compounds of bread crumb A total of 54 volatile compounds were identified in the bread crumb leavened with SCTDPA, whereas 50, 47, and 49 volatile compounds were identified in the bread crumb leavened with SC, SCTD, and SCPA, respectively (Table 6). SCTDPA-leavened bread not only produced more unique volatile compounds, but in greater abundance, as indicated by the greater peak area. In most cases, the bread leavened by co-cultures produced more volatile compounds than that of a single culture. The volatile compounds were predominately alcohols, aldehydes and esters. Isobutyl alcohol (i-BuOH), isoamyl alcohol (i-AmOH), and phenethyl alcohol (PEA) were the predominant alcohol groups. i-BuOH levels were highest in SCTDPA bread (20,169) and the lowest in SC bread (1,986). Inversely, i-AmOH levels were the highest in SCPA bread (11,168) and lowest in SCTDPA bread (6,265). As reported by KIM et al. (2013), P. anomala Y197-13 produced high i-AmOH levels and conferred a banana flavour that significantly affected the flavour and taste of turbid rice wine. WATANABE et al. (1990) reported that bread containing high levels of i-AmOH was less favorable than bread containing high levels of i-BuOH. The bread leavened with P. anomala SKM-T exhibited a higher PEA content and was preferred over S. cerevisiae (MO and SUNG, 2014) due to the favorable honey and flower odor of PEA (JENSEN et al., 2011). The predominant aldehydes in bread crumb were n-hexanal, furfural, and benzaldehyde. The bread leavened with the co-cultures containing PA (SCPA and SCTDPA) produced the highest amounts of these compounds. n-hexanal was the most abundant aldehyde and conferred a green flavour. The second most abundant was benzaldehyde which produces an almond odor (BIRCH et al., 2013a). Fulfural is characterized by a burnt and sweet, caramel-like, odor (PROST et al., 2012). The other volatile compounds that contributed to either favorable (n-octanal, n-decanal) or unfavorable (n-heptenal) odors were comparable among all cultures tested. Esters typically have pleasant, fruity, or sweet odors (BIRCH et al., 2013b). In most cases, SCTDPA leavened bread contained a greater abundance and higher diversity of ester compounds. Compounds including isoamyl acetate (fruity), ethyl caproate (fruity, wine, apple, banana, brandy), ethyl octanoate (fatty, fruity), ethyl decanoate, and ethyldodecanoate were enhanced in the bread leavened with co-cultures containing PA (SCPA and SCTDPA). SC and TD enhanced particular compounds such as ethyl acetate and methyl salicylate, respectively. In alcoholic beverages, ethyl acetate can produce unfavorable sensory qualities. Mixed cultures of S. cerevisiae and P. anomala (mutant type) produced higher ethyl acetate-hydrolyzing esterase activities. This enzyme is crucial in the formation of acetate ester, which delivers superior flavour (KURITA, 2008). The bread leavened by co-cultures was obviously superior to that leavened by a single culture with regard to the volatile compound content. These compounds were produced mainly from the metabolism of yeasts during dough fermentation and flour lipid oxidation (BIRCH et al., 2013b). These processes are influenced by the availability of free, reactive amino acids, sugars, alcohols, enzyme activity, and the degree of polymerization and hydration of substrates due to mixing to baking (STEAR, 1990). SADOUDI et al. (2012) reported that the use of co-cultures altered the production of volatile compounds in wine, because co-culture interactions influenced the entire metabolic pathway.   Ital. J. Food Sci., vol 28, 2016 - 309 Table 6: Volatile compounds contained in the bread crumb leavened with co-cultures or a single culture. No Group RT Compound Flavour Description Peak Area* SC SCTD SCPA SCTDPA 1 Acids 30.40 Acetic acid Acid, pungent (a) 1,973 1,838 2,483 4,043 2 33.28 Isobutyric acid 339 275 287 408 3 35.61 2-Methylbutanoic acid Sweaty (e) 680 603 514 939 4 40.00 2-Methylpropanoic acid Sweat, butter(a) 81 59 83 176 5 Alcohols 18.63 Isobutyl alcohol 1,986 7,494 13,946 20,169 6 23.30 Isoamyl alcohol Banana (f) 8,011 8,174 11,168 6,265 7 26.30 2-Ethyl-1-decanol ND ND ND 301 8 27.69 2-Methyl-3-pentanol 146 180 67 88 9 27.90 1-Hexanol Flower (a) 465 610 527 674 10 28.14 2-Nonanol 153 159 ND 87 11 28.63 3-Ethoxy-1-propanol Fruity (a) 87 236 ND 116 12 32.76 1-Dodecanol 228 163 280 684 13 35.42 2-Furanmethanol 309 552 615 1,857 14 40.86 Phenethyl alcohol Honey, Flower (a) 9,506 12,292 8,391 12,643 15 Aldehydes 18.01 n-Hexanal Green (d) 5,369 19,874 37,144 71,305 16 22.38 n-Heptanal Fatty, rancid (d) 233 262 528 302 17 25.99 n-Octanal Citrus (d) 207 263 333 495 18 27.11 2-Heptenal 600 1,086 1,594 2,335 19 30.88 Furfural 1,646 2,068 2,764 7,529 20 31.79 n-Decanal Citrus (d) 151 230 237 379 21 32.56 Benzaldehyde Almond (d) 3,507 5,475 6,389 8,374 22 33.64 5-Methyl-2-furfural 132 47 131 529 23 38.80 2,4-Decadienal Fatty, waxy (b) 38 94 103 224 24 Alkenes 8.96 2,4-Dimethyl-1-heptene 62 ND ND 396 25 29.86 3-Ethyl-2-methyl-1,3-hexadiene 36 58 83 120 26 Benzenes 15.92 Methyl benzene 269 330 551 413 27 24.95 Ethenylbenzene 762 134 8,026 6,265 28 29.97 1,3-bis(1,1-dimethylethyl)benzene 381 563 85 1,503 29 Esters 9.10 Ethyl acetate Pineapple (a) 1,688 120 635 5,590 30 14.66 Isobutyl acetate Ethereal, fermented ND ND ND 134   Ital. J. Food Sci., vol 28, 2016 - 310 odor (b) 31 19.87 Isoamyl acetate Fruity (e) 305 666 5,290 7,923 32 24.16 Ethyl caproate Fruity, wine, apple, banana, brandy (c) 305 51 4,641 3,903 33 25.43 n-Hexyl acetate 37 34 124 256 34 30.10 Ethyl octanoate Fatty, Fruity (a) 625 796 4,646 12,307 35 30.75 Amyl caproate 7 ND 84 351 36 34.96 Ethyl decanoate 204 171 4,066 9,553 37 36.12 Ethyl-9-decenoate ND ND 104 1,658 38 38.27 Methyl salicylate 220 1,686 316 1,876 39 38.91 2-Phenethyl acetate Roasty(e) 235 101 209 868 40 39.22 Ethyldodecanoate 336 281 1,069 1,407 41 Furans 24.06 2-Pentylfuran Floral, fruity (d) 309 246 887 566 42 31.99 2-Acetylfuran 282 339 450 1,243 43 Ketones 12.73 2,3-Butanedione Buttery, caramel (d) 353 592 460 542 44 22.29 2-Heptanone ND ND 58 67 45 25.89 3-Hydroxy-2-butanone Butterscotch (d) 867 744 922 1,149 46 26.37 1-Octen-3-one Mushroom (g) 48 263 419 765 47 Phenols 40.61 Butylatedhydroxytoluene 1,585 2,788 4,507 2,683 48 Pyrazines 25.33 Methylpyrazine 185 179 228 546 49 27.33 2,6-Dimethylpyrazine Hazelnut (e) 68 73 219 314 50 27.49 Ethylpyrazine 58 95 95 223 51 Terpenes 37.04 (-)-.beta.-Bisabolene 220 ND 341 222 52 Others 28.90 Dimethyl trisulfite 51 120 217 263 53 37.64 Naphthalene 79 143 446 288 54 41.42 1,4-Methanobenzocyclodecene 75 139 265 275 *the values of volatile compounds calculated from the peak area divided by 1000. ND, not detected a. JENSEN et al. (2011); b. MO and SUNG (2014); c. DAIGLE et al. (1999); d. BIRCH et al. (2013a); e. PROST et al. (2012); f. KIM et al. (2013); g. BIRCH et al. (2013b) 3.6. Sensory properties The average results of the sensory evaluation of appearance, colour, flavour, mouthfeel, and overall acceptability are shown in Fig. 4. In most cases, the co-cultures slightly enhanced all the sensory attributes, except for appearance. All attributes produced satisfactory scores in the range of 4.73–5.57 out of a total 7 points. On average, the co-   Ital. J. Food Sci., vol 28, 2016 - 311 cultures produced marked improvement over the single culture, which scored in the range of 4.07-5.71. SCTDPA leavened bread was superior in overall acceptability (5.57), which is attributable to higher ratings in flavour (5.27) and mouthfeel (5.30). High flavour ratings are probably due to the high abundance of favorable volatile compounds (Table 6). SCTD leavened bread had a superior colour rating (5.53). These results demonstrate that incorporating SC, TD, and PA as leavening agents conferred beneficial characteristics to bread. SC contributed to improving bread appearance through greater leavening ability, and TD and PA contributed to enhanced flavour, colour, and mouthfeel. Figure 4. Radar plot of the sensory properties of bread leavened with co-cultures or a single culture. Result reflects the means of scores from 15 semi-trained panelists. 4. CONCLUSIONS We have shown that the use of mixed cultures of S. cerevisiae, T. delbrueckii JK08, and P. anomala JK04 enhanced bread quality. The bread leavened by the co-cultures produced textural and structural properties comparable to single cultures of S. cerevisiae. The co- cultured bread had a superior aroma and enhanced sensorial qualities. Thus, the use of co- cultures as leavening agents has great promise in fulfilling the consumer need for unique and high-quality bread. ACKNOWLEDGEMENTS We thank the Directorate General of Human Resources for Science, Technology and Higher Education, the Republic of Indonesia, and the Department of Food and Food Service Industry at Kyungpook National University, Republic of Korea, for supporting this study. The authors wish to thank Dr. D. Silveri, Dr. V. Galli and Mr. L. Bartoli for the technical assistance and Dr. Francesca Melini for the technical and linguistic assistance in the revision of this paper. REFERENCES AACC 10-05. 2000. Rapeseed displacement method. In: Approved Methods of the AACC. 10th ed. St. Paul :American Association of Cereal Chemists. 0,00   1,00   2,00   3,00   4,00   5,00   6,00   7,00   Appearance   Color   Flavor  Mouthfeel   Overal   SC   SCTD   SCPA   SCTDPA     Ital. J. Food Sci., vol 28, 2016 - 312 Ali A., Shehzad A., Khan M.R., Shabbir M.A. and Amjid M.R. 2012. Yeast, its types and role in fermentation during bread making process-A Review. Pak. J. Food Sci. 22: 171-179. Annan N.T., Poll L., Dedeh S.S., Plahar W.A. and Jakobsen M. 2003. Influence of starter culture combinations of Lactobacillus fermentum, Saccharomyces cerevisiae and Candida krusei on aroma in Ghanaian maize dough fermentation. Eur. Food Res. Technol. 216: 377-384. Bely M., Stoeckle P., Pomarede I.M. and Dubourdieu D. 2008. Impact of mixed Torulaspora delbrueckii-Saccharomyces cerevisiae culture on high-sugar fermentation. Int. J. Food Microbiol. 122: 312-320. Birch A.N., Petersen M.A., Arneborg N. and Hansen A.S. 2013a. Influence of commercial baker’s yeasts on bread aroma profiles. Food Res. Int. 52: 160-166. Birch A.N., Petersen M.A. and Hansen Å.S. 2013b. The aroma profile of wheat bread crumb influenced by yeast concentration and fermentation temperature. LWT - Food Sci. Technol. 50: 480-488. Blandino M., Sovrani V., Marinaccio F., Reyneri A., Rolle L., Giacosa S., Locatelli M., Bordiga M., Travaglia F., Coïsson J. D. and Arlorio M. 2013. Nutritional and technological quality of bread enriched with an intermediated pearled wheat fraction. Food Chem. 141: 2549-57. Bosmans G.M., Lagrain B., Fierens E. and Delcour J. A. 2013. The impact of baking time and bread storage temperature on bread crumb properties. Food Chem. 141: 3301-8. Caballero R., Olguin P. and Gallardo F. 1995. Evaluation of Kluyveromyces marxianus as baker’s yeast. Food Res. Int. 28: 37-41. Cauvain S.P. 2004. Improving the texture of bread. In “Texture in food: Solid foods”. Vol. 2. 1st edition. D. Kilcast (Ed.), p, 435-448. Woodhead, Cambridge, England. Cauvain S.P. 2012. Introducing to bread making. In “Breadmaking: Improving quality”. 2nd edition. S. P. Cauvain (Ed.), p, 1-8. Woodhead, Cambridge, England. Choi I. and Choi W.S. 2009. Effects of Kokja as a fermentation starter on sponge-and-dough bread properties. J. Kor. Soc. Appl. Biol. Chem. 53: 50-55. Czuchajowska Z., Pomeranz Y. and Jeffers H. 1989. Water activity and moisture content of dough and bread. Cereal Chem. 66: 128-132. Daigle P., Gélinas P., Leblanc D. and Morin A. 1999. Production of aroma compounds by Geotrichumcandidum on waste bread crumb. Food Microbiol. 16: 517-522. Elhariry H.M., Mahmoud R.M., Hassan A.A. and Aly M.A. 2011. Development of co-culture sourdough systems for improving bread quality and delaying staling. Food Biotechnol. 25: 252-272. Gardner N., Champagne C. P. and Gelinas P. 2002. Effect of yeast extracts containing propionic acid on bread dough fermentation and bread properties. J. Food Sci. 67: 1855-1858. Gélinas P. 2009. Inventions on baker’s yeast strains and specialty ingredients. Recent Pat. Food Nutr. Agric. 1: 104-132. Hernandez-Lopez M.J., Prieto J.A. and Randez-Gil F. 2003. Osmotolerance and leavening ability in sweet and frozen sweet dough. Comparative analysis between Torulaspora delbrueckii and Saccharomyces cerevisiae baker’s yeast strains. Antonie Van Leeuwenhoek. 84: 125-34. Hsu C. 2003. Chemical composition, physical properties, and antioxidant activities of yam Leavaflours as affected by different drying methods. Food Chem. 83: 85-92. Jensen S., Oestdal H., Skibsted L.H., Larsen E. and Thybo A.K. 2011. Chemical changes in wheat pan bread during storage and how it affects the sensory perception of aroma, flavour, and taste. J. Cereal Sci. 53: 259-268. Karaoğlu M.M. and Kotancilar H.G. 2006. Effect of partial baking, storage and rebaking process on the quality of white pan bread. Int. J. Food Sci. Technol. 41: 108-114. Kim H.R., Kim J.H., Bai D.H. and Ahn B.H. 2013. Microbiological characteristics of wild yeast strain Pichia anomala y197- 13 for brewing makgeolli. Mycobiology. 41: 139-44. Kurita O. 2008. Increase of acetate ester-hydrolysing esterase activity in mixed cultures of Saccharomyces cerevisiae and Pichia anomala. J. Appl. Microbiol. 104: 1051-8.   Ital. J. Food Sci., vol 28, 2016 - 313 Lampignano V., Laverse J., Mastromatteo M. and Del Nobile M.A. 2013. Microstructure, textural and sensorial properties of durum wheat bread as affected by yeast content. Food Res. Int. 50: 369-376. Lin L.Y., Liu H.M., Yu Y.W. and Lin S.D. and Mau J.L. 2009. Quality and antioxidant property of buckwheat enhanced wheat bread. Food Chem. 112: 987-991. Mo E.K. and Sung C.K. 2014. Production of white pan bread leavened by Pichia anomala SKM-T. Food Sci. Biotechnol. 23: 431-437. Mondal A. and Datta A.K. 2008. Bread baking - A review. J. Food Eng. 86: 465-474. Movahed S., Rooshenas G. and Chenarbon H.A. 2012. Evaluation of the effect of yeast-salt method on dough yield, bread yield and organoleptic properties Iranian Lavash bread. Annals Biol. Res. 3: 595-600. Nilufer-Erdil D., Serventi L., Boyacioglu D. and Vodovotz Y. 2012. Effect of soy milk powder addition on staling of soy bread. Food Chem. 131: 1132-1139. Ogunbanwo S.T., Adebayo A.A., Ayodele M.A., Okanlawon B.M. and Edema M.O. 2008. Effects of lactic acid bacteria and Saccharomyces cerevisae co-cultures used as starters on the nutritional contents and shelf life of cassava-wheat bread. J. Appl. Biosci. 12: 612-622. Pacheco A., Santos J., Chaves S., Almeida J., Leão C. and Sousa M.J. 2012. The emerging role of the yeast Torulaspora delbrueckii in bread and wine production: using genetic manipulation to study molecular basis of physiological responses. in “Structure and Function of Food Engineering”. A.A. Eissa (Ed.), p, 339-370. InTech, Rijeka, Croatia. Plessas S., Bekatorou A., Gallanagh J., Nigam P., Koutinas A.A. and Psarianos C. 2008. Evolution of aroma volatiles during storage of sourdough breads made by mixed cultures of Kluyveromyces marxianus and Lactobacillus delbrueckii ssp. bulgaricus or Lactobacillus helveticus. Food Chem. 107: 883-889. Plessas S., Pherson L., Bekatorou A., Nigam P. and Koutinas A.A. 2005. Bread making using kefir grains as baker’s yeast. Food Chem. 93: 585-589. Pongjaruvat W., Methacanon P., Seetapan N., Fuongfuchat A. and Gamonpilas C. 2014. Influence of pregelatinised tapioca starch and transglutaminase on dough rheology and quality of gluten-free jasmine rice breads. Food Hydrocoll. 36: 143-150. Prost C., Poinot P., Rannou C. and Arvisenet G. 2012. Bread aroma. In “Breadmaking: Improving quality”. 2nd edition. S. P. Cauvain (Ed.), p, 523-561. Woodhead, Cambridge, England. Rosell C.M., Santos E., Penella J.M.S. and Haros M. 2009. Wholemeal wheat bread: A comparison of different breadmaking processes and fungal phytase addition. J. Cereal Sci. 50: 272-277. Sadoudi M., Tourdot-Maréchal R., Rousseaux S., Steyer D., Gallardo-Chacón J.J., Ballester J., Vichi S., Guérin-Schneider R., Caixach J. and Alexandre H. 2012. Yeast-yeast interactions revealed by aromatic profile analysis of Sauvignon Blanc wine fermented by single or co-culture of non-Saccharomyces and Saccharomyces yeasts. Food Microbiol. 32: 243-53. Scanlon M.G. and Zghal M.C. 2001. Bread properties and crumb structure. Food Res. Int. 34: 841-864. Stear C.A. 1990. Handbook of breadmaking technology. Elsevier, London and New York. Ulziijargal E., Yang J.H., Lin L.Y., Chen C.P. and Mau J.L. 2013. Quality of bread supplemented with mushroom mycelia. Food Chem. 138: 70-76. Wahyono A., Kang W.W. and Park H.D. 2015. Characterization and application of Torulaspora delbrueckii JK08 and Pichia anomala JK04 as baker's yeast. J. Food Nutr. Res. 54: 205-217. Watanabe M., Fukuda K., Asano K. and Ohta S. 1990. Mutants of bakers’ yeasts producing a large amount of isobutyl alcohol or isoamyl alcohol, flavour components of bread. Appl. Microbiol. Biotechnol. 34: 154-159. Zghal M.C., Scanlon M.G. and Sapirstein H.D. 2002. Cellular structure of bread crumb and its influence on mechanical properties. J. Cereal Sci. 36: 167-176. Paper Received May 6, 2015 Accepted August 30,2015