This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

Influence of in-situ synthesized exopolysaccharides on the quality of gluten-freesorghum sourdough bread

Sandra Galle a,b, Clarissa Schwab c, Fabio Dal Bello a,b, Aidan Coffey d, Michael G. Gänzle c, Elke K. Arendt a,⁎a University College Cork, School of Food and Nutritional Science, Cork, Irelandb University College Cork Biotransfer Unit, Ireland, Cork, Irelandc University of Alberta, Department of Agricultural, Food and Nutritional Science, Edmonton, AB, Canada, T6G 2P5d Cork Institute of Technology, Department of Biological Science, Cork, Ireland

a b s t r a c ta r t i c l e i n f o

Article history:Received 8 July 2011Received in revised form 29 November 2011Accepted 15 January 2012Available online 20 January 2012

Keywords:EPSFermentationOrganic acidGluten-freeWeissella

The majority of gluten-free breads on the market are of poor sensory and textural quality. Exopolysaccharides(EPS) formed from sucrose during sourdough fermentation can improve the technological properties ofgluten-free breads and potentially replace hydrocolloids. In this study, the influence of in situ formed EPSon dough rheology and quality of gluten-free sorghum bread was investigated. Dextran forming Weissellacibaria MG1 was compared to reuteran producing Lactobacillus reuteri VIP and fructan forming L. reuteri Y2.EPS containing bread batters were prepared by adding 10% and 20% of sourdough. As control served battersand bread containing sourdoughs fermented without sucrose and batters and bread without sourdough ad-dition. The amount of EPS formed in situ ranged from 0.6 to 8.0 g/kg sourdough. EPS formed during sour-dough fermentation were responsible for the significant decrease in dough strength and elasticity, with insitu formed dextran exhibiting the strongest impact. Increased release of glucose and fructose from sucroseduring fermentation enhanced CO2 production of yeast. Organic acids in control sourdough breads inducedhardening of the bread crumb. EPS formed during sourdough fermentation masked the effect of the organicacids and led to a softer crumb in the fresh and stored sorghum bread. Among EPS, dextran showed the bestshelf life improvements. In addition to EPS, all three strains produced oligosaccharides during sorghum sour-dough fermentation contributing to the nutritional benefits of gluten-free sorghum bread. Results of thisstudy demonstrated that EPS formed during sourdough fermentation can be successfully applied in gluten-free sorghum flours to improve their bread-making potentials.

© 2012 Elsevier B.V. All rights reserved.

1. Introduction

A gluten-free diet is essential for celiac patients. However, glutenreplacement is one of the most challenging issues for food scientists.Many of the current gluten-free baked products available on themarketare of low quality and particularly exhibit poor mouth feel and flavor(Moroni et al., 2009). The production of gluten-free bread is basedmainly on the use of (purified) starch, protein-based ingredients, andhydrocolloids into a gluten-free base flour to mimic the viscoelasticproperties of gluten (McCarthy et al., 2005). Sorghum [Sorghum bicolor(L.) Moench] is a traditional crop in Africa that is safe for consumptionby celiac patients (Kasarda, 2001). In recent years sorghumhas receivedincreasing attention from scientists for prevention of chronic diseasesdue to its functional properties, its content of polyphenolic compounds,

and as a potential rawmaterial in gluten-free diet (Renzetti et al., 2008;Schober et al., 2005; Svensson et al., 2010).

Challenges for production of gluten-free bread particularly relateto structure, gas holding abilities, and the staling rate. Incorporationof sourdough to a gluten-free formula is known to overcome theseproblems (Huettner et al., 2010; Moore et al., 2008; Pruska-Kędzioret al., 2008; Schober et al., 2007; Schwab et al., 2008). The positive ef-fects are associated with metabolites produced by lactic acid bacteria(LAB) during sourdough fermentation, including organic acids, exopo-lysaccharides (EPS), and enzymes. EPS improve dough rheology andbread texture, and can be used to replace or to reduce hydrocolloidscurrently used as bread improvers (Tieking and Gänzle, 2005). LABproduce a large variety of EPS and their properties vary dependingon their chemical structure, molecular mass and shape. Heteropoly-saccharides (HePS) composed of irregular repeating units are tradi-tionally used in the dairy industry to improve texture and mouthfeelof the products (De Vuyst et al., 2001). Heteropolysaccharides influ-ence the rheology of sorghum batters (Galle et al., 2011) however, todate only homopolysaccharides (HoPS) have to be shown to be usefulin baking applications. HoPS are glucan or fructan polymers formed by

International Journal of Food Microbiology 155 (2012) 105–112

⁎ Corresponding author at: School of Food and Nutritional Sciences, University Col-lege Cork, Western Road, Cork, Ireland. Tel.: +353 21 4902064; fax: +353 21 4270213.

E-mail addresses: galle@ualberta.ca (S. Galle), aidan.coffey@cit.ie (A. Coffey),mgaenzle@ualberta.ca (M.G. Gänzle), e.arendt@ucc.at (E.K. Arendt).

0168-1605/$ – see front matter © 2012 Elsevier B.V. All rights reserved.doi:10.1016/j.ijfoodmicro.2012.01.009

Contents lists available at SciVerse ScienceDirect

International Journal of Food Microbiology

j ourna l homepage: www.e lsev ie r .com/ locate / i j foodmicro

Author's personal copy

glucan- or fructansucrases from sucrose, respectively. In additionto HoPS these enzymes also form glucooligosaccharides (GOS) andfructosoligosaccharides (FOS) in the presence of acceptor sugars(van Hijum et al., 2006). In situ formation of HoPS in sourdough wasreported to be more effective than external addition (Brandt et al.,2003). In wheat fermentations, dextran from Weissella sp. and Leuco-nostoc mesenteroides improved rheological parameters and breadquality (Di Cagno et al., 2006; Katina et al., 2009; Lacaze et al., 2007).However, in situ production of levan by Lactobacillus sanfranciscensisdid not influence wheat bread volume or crumb hardness (Kaditzkyet al., 2008). Formation in situ of EPS from sucrose results in furthermetabolites such as mannitol, glucose and acetate. Heterofermenta-tive lactobacilli, including Lactobacillus reuteri and L. sanfranciscensis,metabolize fructose as an electron acceptor, which results in the for-mation of mannitol and acetate (Korakli et al., 2001). Increased ace-tate levels as a consequence of sucrose addition negatively impactsthe quality of wheat bread (Kaditzky et al., 2008).

In gluten-free bread, dextran from Weissella cibaria 10 M but notlevan from L. reuteri improved nutritional and organoleptic propertiesof gluten-free bread (Schwab et al., 2008). Previously it was shownthat, W. cibaria MG1 forms high amounts of EPS (8±3 g/kg dough)and isomaltooligosacchrides (IMO) but only small amounts of acetatein sorghum sourdough (Galle et al., 2010). A successful application ofEPS producing LAB strains in gluten-free baking requires improvedinformation on both, structure–function relationship of EPS and therole of organic acids. It was therefore the aim of this study to comparethe performance ofW. cibariaMG1with two strains of L. reuteriwhichsynthesize two different types of EPS and produce acetate to gain abetter insight of the effect of different EPS and organic acids on thequality on gluten-free sorghum bread.

2. Materials and methods

2.1. Materials, microorganisms and growth conditions

Sorghum flour (Twin ValleyMills, Nebraska) containing 11.3% pro-tein and 3.3% fat was used in conjunction with dried yeast (Puratos,Belgium), salt and tap water. L. reuteri Y2, L. reuteri VIP and W. cibariaMG1 (Galle et al., 2010) were obtained from the culture collection ofthe cereal science laboratory of University College Cork. Working cul-tures were prepared from glycerol stock stored at −80 °C. To ensurethe purity of the stock, strains were streaked on MRS 5 agar (Merothet al., 2003) supplemented with 0.05 g/L bromocresol green (Sigma)and incubated anaerobically at 30 °C (W. cibaria) and 37 °C (L. reuteri)for 48 h. For preparation of working cultures single colonies werepicked from agar plates and subcultured twice in mMRS 5 broth.

2.2. Sourdough preparation

Sorghum flour and autoclaved tap water was mixed to obtain adough yield of 200 (dough yield=[(g flour+g water)/g flour]×100).For in situ productions of EPS L. reuteri Y2 (Y2+), L. reuteri VIP (VIP+)andW. cibariaMG1 (MG1+) sourdoughswere fermentedwith additionof 15% sucrose. Sourdoughs fermented with the same strains withoutsucrose addition served as a control (Y2−, VIP−, MG1−). For the sour-dough fermentation single colonies grown on MRS5 agar were pickedand subcultured twice in MRS5 broth. Strains were washed once withsterile tapwater and inoculated to obtain a cell count of 107 CFU/g sour-dough. W. cibaria and L. reuteri sourdoughs were incubated for 24 h at30 °C and 37 °C, respectively.

2.3. Determination of cell counts, pH and metabolite formation insourdough

Cell counts and pH were determined at 0 h and after 24 h of fer-mentation. For LAB counting, samples of sourdough were serially

diluted in Ringer solution (Merck, Germany) and plated in triplicateon MRS5 agar supplemented with 0.05 g/L bromocresol green(Sigma) (Dal Bello and Hertel, 2006). The persistence of starter cul-tures was confirmed by colony morphology and metabolic patterns.The total titratable acidity (TTA) of ripe sourdough was determinedby suspension of 10 g of sourdough in 90 mL of water and titratingwith 0.1 M NaOH to pH 8.5 (with retitrating to pH 8.5, 3 min after itwas first reached). For determination of organic acids, sourdoughswere mixed 1:1 with 7% perchloric acids. Proteins were precipitatedover night at 4 °C. Organic acids were determined by HPLC (Agilent1200 Series) using the REZEX 8 μ 8% H, organic acid column300×7.8 mM (Phenomenex, USA) coupled to a refrective index de-tector (Agilent 1200 Series). As the elution fluid with 0.01 N H2SO4

was used, at a flow rate of 0.6 mL/min. The temperature of the columnwas maintained at 65 °C. Lactate, acetate and ethanol (all Sigma-Aldrich) were used as external standards.

2.4. Determination of polymer production in culture media

For determination of oligosaccharide formation, strains were grownin MRS5 supplemented with 25 g L−1 sucrose, 25 g L−1 sucrose and12.5 g L−1 maltose or 25 g L−1 raffinose. Cells were removed by centri-fugation and oligosaccharides were directly analyzed from the culturesupernatant. Sugars were analyzed with a CarbopacPA20 column(Dionex, Oakville, Canada) as previously described (Galle et al., 2010)using water (A), 200 mMNaOH (B) and 1 M Na-acetate (C) as solventsat a flow rate of 0.25 mLmin−1 with the following gradient: 0 min30.4% B, 1.3% C, 22 min 30.4% B, 11.34% C followed by washing andregeneration. Sucrose, glucose, fructose, maltose, panose, isomaltoseand isomaltotriose were used as external standards (all obtained fromSigma, Oakville, Canada).

For EPS formation W. cibaria MG1, L. reuteri VIP and L. reuteri Y2were grown in MRS media with 100 g L−1 sucrose as sole carbonsource, for 24 h at 30 °C and 37 °C, respectively. Isolation and structureanalysis of EPS was carried out as previously described by Galle et al.(2010). EPS size was determined by asymmetrical flow field-flow-fractionation (FFF) coupled to multi-angle light scattering (MALS)and a RI detector (Postnova, Salt Lake City, UT) as described by Galle(2011).

2.5. Analysis of oligosaccharide and EPS formation in sourdough

EPS was extracted from dough as previously described by Tiekinget al. (2003). Briefly, EPS was precipitated from aqueous dough ex-tracts with ethanol, dialyzed against distilled water and lyophilized.Freeze dried samples were dissolved in distilled water (2 mg/mL).EPS in sourdough were quantified by size exclusion chromatographyusing a Superdex 200 column (GE Healthcare, Baie d'Urfe, Canada)as previously described (Galle et al., 2010). Oligosaccharides synthe-sized in dough were extracted with H2O at 80 °C for 2 h. Sugars wereanalyzed with a Carbopac PA20 column (Dionex, Oakville, Canada)usingwater (A), 200 mMNaOH (B) and 1 MNa-acetate (C) as solventsat a flow rate of 0.25 mL/min with the following gradient: 0 min 30.4%B, 1.3% C, 22 min 30.4% B, 11.34% C followed by washing and regener-ation. Sucrose, glucose, fructose, maltose, panose, isomaltose and iso-maltotriose were used as external standards (all obtained from Sigma,Oakville, Canada).

2.6. Baking procedure

Control breads were prepared with 100 parts of sorghum flour, 90parts ofwater, 2 parts of salt and 2 parts of yeast as previously describedby Renzetti et al. (2008). Twelve different sourdough breads were com-pared to the unsoured control: breads prepared with EPS positive sour-doughs at an addition level of 10% (10Y2+, 10VIP+, 10MG1+) and 20%(20Y2+, 20VIP+, 20MG1+). Breads containing 10% (10Y2−, 10VIP−,

106 S. Galle et al. / International Journal of Food Microbiology 155 (2012) 105–112

Author's personal copy

10MG1−) and 20% (20Y2−, 20VIP−, 20MG1−) EPS negative sour-dough served as a control. To keep a constant water/flour ratio, flourand water present in the sourdough was calculated to replace theequal amount of flour and water in control batter formulation.

Dried yeast was dissolved in tap water (30 °C) and rehydrated atroom temperature for 10 min. First sourdough was mixed with pre-fermented yeast and then remaining pre-mixed dry ingredientswere added into a Hobart mixer bowl (Hobart Food Equipment,Sydney, Australia). Mixing was performed with a paddle tool (Kbeater) at slow speed (level 1 out of 3) for 30 s and 90 s at mediumspeed (2 out of 3). Batters (450 g) were placed into fat-sprayed bakingtins (930 mL volume; 7.3 cm height; 9.5×15.2 top; 7.5×13.2 cm bot-tom) and rested in a proofer (Koma BV Roremond, the Netherlands) at30 °C and 85% rh for 35 min. Breads were baked in a deck oven (MIWE,Arnstein, Germany) at 220 °C top heat and 235 °C bottom heat for35 min. The oven was pre-injected with steam (0.3 L of water) andafter loading the oven was steamed again with 0.7 L of water. Afterbaking, the loaves were depanned and cooled for 120 min on coolingracks at room temperature. Baking was performed on three differentdays (3 independent trials) and nine loaves were prepared for eachbread type at each baking trial.

2.7. Measuring pH and TTA of batters and bread crumbs

The pH and TTA of the batters were measured as already describedabove. Batters were resuspended in distilled water (1/10 w/w) bystirring, whereas the crumb was homogenized with distilled water(1/10 w/w) in a commercial Kenwood blender.

2.8. Bread evaluation

Standard baking tests were performed on three loaves for eachbread type at each baking trial 2 h after baking (day 0) as previouslydescribed (Renzetti et al., 2008). The remaining loaves were packedand further evaluation was carried out after 2 and 5 days of storage.For crumb texture analysis, three slices of 25 mm thickness weresliced from the center of each three loaves of each bread type. Textureprofile analysis (TPA)was performed using a TA-XT2i texture analyzer(Stable Micro System, Surrey, UK) equipped with a 25 kg load cell and35 mm aluminum cylindrical probe. Measurements were performedwith a test speed of 2.0 mm/s and with a trigger force of 20 g to com-press the center of the bread slice to 60% of its original height. Themeasurements obtained for three loaves of one batch over storagetime were averaged into one value (one replicate). TPA was repeatedafter 2 and 5 days of storage at room temperature.

2.9. Rheofermentometer analysis

Gaseous release and dough development of the 13 different batterswere measured using a Rheofermentometer (Chopin, Villeneuve-La-Garenne, France). Three hundred grams of dough was prepared inthe same manner as described above using a commercial Kenwoodmixer for mixing the ingredients. The tests were performed at 30 °Cover a period of 90 min.

2.10. Fundamental rheology

Frequency sweep was performed on a controlled stress rheometer(Anton Paar MCR 301, Ostifildern, Germany) fitted with a coaxial ge-ometry consisting of a 25 mm diameter bob in a 27 mm cup. Batterswere prepared as described for baking without the addition ofyeast. Samples were placed into the cup and the whole system wascovered. A water containing ring was placed on the inner side of thecover to prevent drying out of the batter. The dough was allowed torest for 5 min and amplitude sweeps were conducted to establishthe linear viscoelastic region of the samples. Frequency sweeps from

1 to 50 1−S angular frequency (ω) were performedwith a target strainof 10−3 (0.1%) at 30 °C. The storage modulus (G'), loss modulus (G"),complex modulus (G*), and phase angle (δ) were calculated using themanufacturer's software. All the results are average of three individualreplicates.

2.11. Statistical analysis

Statistical analyses were performed with Sigma Plot 11.0 (SystatSoftware Inc., United States) on all data using one-way Anova. Fisher'sleast significant differences test was used to describe means at 5% sig-nificance level.

3. Results

3.1. Formation of oligo- and polysaccharides of L. reuteri Y2, L. reuteri VIPand W. cibaria MG1 in culture media and sorghum sourdough

Oligo- and polysaccharides produced by L. reuteri VIP andW. cibaria MG1 during growth in laboratory media were previouslydescribed (Galle, 2011). Polymer formation by L. reuteri Y2 in labora-tory media was characterized by the same methodology. L. reuteri Y2formed FOS in the presence of raffinose and produced EPS with a rel-ativemolecular weight of 8.9×106. The polysaccharidewas composedof fructose and digested by inulinase and therefore considered as fruc-tan. The reuteran formed by L. reuteri VIP and the dextran formed byW. cibaria MG1 had a relative molecular weight of 107 and 7.2×108,respectively (Galle, 2011). During 24 h of sorghum sourdough fer-mentationW. cibariaMG1 formed IMO. L. reuteri VIP formed glucooli-gosaccharides (GOS) most likely IMOwhich could not be identified byreference compounds (Fig. 1). L. reuteri Y2 synthesized small amountsof kestose (Fig. 1). L. reuteri VIP and Y2 sythesized 0.6±0.4 and 3.28±1.6 g kg−1 EPS. W. cibaria MG1 produced 8 g kg−1 EPS.

3.2. Growth and metabolite formation in sourdoughs

L. reuteri VIP reached an average cell count of 9.7±0.1 log CFU g−1

dough. Cell counts were lower in sourdoughs fermented with L. reuteriY2 7.1±0.7 log CFU g−1 dough. In Y2+ and VIP+ sourdoughs 104.2±8.9 mmol kg−1 and 105.9±2.3 mmol kg−1 lactate and 75±6.3 mmolkg−1 and 73.9±2.3 mmol kg−1 acetate were formed, respectively. Noethanol was detected in Y2+ and only 17.1 mmol kg−1 in VIP+sour-dough. In contrast, 64±4.8 mmol kg−1 lactate, 31.9±2.2 mmol kg−1

ethanol and only small amounts of acetate (9.5 mmol kg−1 dough)were formed in MG1+ sourdoughs. Lactate concentration in controlsourdoughs Y2−, VIP− and MG1− sourdoughs were similar to their

Fig. 1. Oligosaccharide formation formed by W. cibaria MG1 (MG1), L. reuteri VIP (VIP)and L .reuteri Y2 (Y2) in sorghum sourdough after 24 h of fermentation. Sourdough atfermentation time 0 h served as a control.

107S. Galle et al. / International Journal of Food Microbiology 155 (2012) 105–112

Author's personal copy

sucrose containing counterparts but higher amounts of ethanol (60.7±0.8, 52.0±7.5 and 52.8±8.3 mmol kg−1 dough, respectively) and onlytrace amounts of acetate (below 10mmol kg−1) were detected.

3.3. pH and TTA in sourdoughs, batters and bread

The pH and TTA values were monitored over the whole breadmaking process (Table 1). Generally, pH values in MG1+ and MG1−sourdoughs were higher compared to Y2+, Y2− and VIP+ and VIP−sourdoughs. The pH values ofMG1+ and VIP+ sourdoughs did not dif-fer from the sucrose free counterparts MG1− and VIP−. A lower pHwas measured in Y2+ sourdoughs compared to Y2− sourdough. TTAvalues were significantly higher in all sourdoughs when sucrose wasadded (pb0.05), with the exception of MG1 sourdoughs, where the ad-dition of sucrose led to no major changes or even decreased the TTAslightly (Table 1). TTA values measured in VIP+ and VIP− sourdoughswere higher compared to those reached in Y2+ and Y2− sourdoughs.Lowest TTA values were measured in MG1+ and MG1− sourdoughs.

The addition of increased amounts of sourdough resulted in a pro-gressive decrease of batter pH and at the same time increase of TTAcompared to the control batters. Correspondingly, lower pH valueswere measured in the bread crumb of the sourdough breads than inthe control breads. Breads produced with addition of sourdough bat-ters showed higher TTA values in comparison to the control bread.Compared to VIP and Y2 breads, higher pH and lower TTA valueswere measured in MG1 breads. Independently of the strain used addi-tion of EPS positive sourdough did not change pH and TTA values com-pared to batters prepared with EPS negative sourdough. In breadcrumbs the addition of EPS positive sourdough resulted in significant-ly lower pH and higher TTA in 20Y2+ and 20VIP+ bread compared tothe EPS negative controls. In contrast, pH and TTA values of MG1+bread and batters did not differ from the EPS negative controls.

3.4. Rheological properties of sorghum batters

In all doughs, the elastic modulus (G') was higher than the viscousmodulus (G"), indicating that sorghum batters had a solid, elastic-likebehavior (data not shown). To determine the resistance to deforma-tion as well as the viscoelastic properties of sourdoughs, complexmodulus |G*| (Fig. 2A) and phase angle δ (Fig. 2B) were calculated.

Generally the addition of 10% and 20% EPS sourdoughs significant-ly decreased |G*| (pb0.05) compared to the 10% and 20% containingcontrol sourdough batters, respectively, indicating decreased resis-tance to deformation (Fig. 2A). No significant difference could be ob-served between the control batter and the batters prepared with 10%and 20% EPS negative sourdough. Only addition of 10Y2- sourdoughresulted in a slight but significant increase of |G*| (Fig. 2A1)

(pb0.05). Addition of 20% EPS positive sourdough notably decreased|G*| compared to 10% EPS-sourdough addition (pb0.05). |G*| of bat-ters prepared with 10MG1+ sourdough was significantly lower com-pared to batters prepared with 10Y2+ and 10VIP+ sourdoughs. Nosignificant difference between 20Y2+, 20VIP+ and 20MG1+ batterswas observed (pb0.05), but addition of 20MG1+ sourdough resultedin a steeper increase of |G*| (Fig. 3).

Fig. 2B shows the effect of the addition of EPS containing sour-dough fermented with the selected stains on the phase angle (δ) ofsorghum batters. In comparison to the control batter, a slight but sig-nificant increased δ was measured in 10Y2+ and 10MG1+ battersbut not in 10VIP+. Independently of the strain used, addition of20% EPS containing sourdough significantly increased δ, indicating adecrease in elasticity compared to the control batters. No differencein δ between the negative EPS batters and the control batter was ob-served. As shown in Fig. 2B1 at higher angular frequency (ω) 10Y2+and 20Y2+ batters showed significant higher phase angle, whereasat lower ω no significant differences could be observed (pb0.05).Batters of 20VIP+ showed a significant increase at higher angular fre-quency (Fig. 2B2). In contrast, 20MG1+ dough exhibited a remark-ably increased δ at low ω compared to the 10MG1+ and the controlbatters. At higher ω this differences were less pronounced but stillsignificant (p>0.05) (Fig. 2B3).

3.5. Dough development and gaseous release of sorghum batters

The effect of EPS on dough development and gaseous release ofsorghum batters are summarized in Table 2. Control batters reacheda maximum dough development height (Hm) of 26.2±0.1 mm. Nomajor changes in Hm occurred when MG1, sourdough was added tothe batters. Also Hm of 10Y2 and 10VIP batters did not differ fromcontrol batter. In contrast, Hm of 20Y2+ and 20VIP+ batters andtheir EPS negative control batters were significantly lower comparedto unfermented control batter.

In comparison to the control batters, a reduced time of maximumdough rise (T1) was observed for batters prepared with EPS contain-ing sourdough. An opposite trend was observed for batters preparedwith EPS negative sourdough. Particularly inclusion of 10Y2− and10VIP− sourdoughs resulted in a significant increase of T1. The max-imum height of gaseous release (H'm), time to reach (T'1) and thetotal volume of gaseous release (Vt) of control batter was 60.6 mm,29.3 min and 627.0 mL, respectively. Addition of EPS negative controlbatters did not change any gaseous release parameters with the ex-ception of batters prepared with control sourdough 10MG1− and20Y2− showing a 10% raise in H'm, T'1 and Vt. Independently ofthe strain used, addition of EPS positive sourdough significantly in-creased H'm, T'1 and Vt. Increasing addition level of EPS positive

Table 1pH and TTA values of sorghum sourdoughs, bread batters and bread crumbs.

Sucroseaddition

SD⁎

additionSourdough Bread batters Bread crumb

pH TTA pH TTA pH TTA

L. reuteri Y2 + 10% 3.7±0.1a⁎⁎ 19.6±0.2a 5.2±0.1a 7.7±0.5a,b,c,d 5.4±0.0a 5.2±0.1a,bL. reuteri Y2 − 10% 3.8±0.0b,c 15.8±0.4b 5.5±0.1b 7.6±0.7a,b,c,d 5.6±0.0b 5.0±0.6aL. reuteri Y2 + 20% 3.7±0.1a 19.3±0.1a 4.9±0.1c 9.5±1.2e,f 4.9±0.1c 6.5±0.1cL. reuteri Y2 − 20% 3.8±0.0c 15.4±0.1b 5.1±0.1 d 8.2±0.9b,c,d,e 5.2±0.0 d 5.5±0.4bL. reuteri VIP + 10% 3.6±0.0a 21.7±0.4c 5.2±0.0a,d 7.9±0.9a,b,c,d 5.3±0.0e 5.5±0.3bL. reuteri VIP − 10% 3.6±0.0a 19.1±0.6a,d 5.3±0.0a 7.4±0.3a,b,c 5.4±0.0a 5.2±0.2a,bL. reuteri VIP + 20% 3.6±0.0a 21.3±1.0c 4.7±0.1c 9.8±0.8f 4.8±0.0f 7.1±0.2 dL. reuteri VIP − 20% 3.6±0.0a 18.4±0.4 d 4.8±0.1c 8.7±0.0c,d,f 4.9±0.0c 6.4±0.1cW. cibaria MG1 + 10% 4.0±0.1 d 11.8±0.1e 5.6±0.1e 6.8±0.1a,b 5.8±0.1 g 3.5±0.1eW. cibaria MG1 − 10% 4.0±0.0 d 11.2±0.1e 5.8±0.0e 6.3±0.4a 5.9±0.1 g 3.5±0.1eW. cibaria MG1 + 20% 4.0±0.1 d 10.1±0.1f 5.2±0.1a,d 8.9±0.2 d,f 5.4±0.1a 5.3±0.2a,bW. cibaria MG1 − 20% 4.1±0.0 d 11.3±0.4e 5.2±0.0a 7.9±1.5b,c,d 5.5±0.0a 5.2±0.4a,bCON − 0% 5.7±0.0e 7.0±1.1a,b 6.1±0.0 h 4.0±0.2e

*SD, sourdough. **Different letters in the same column indicate statistical significance (pb0.05).

108 S. Galle et al. / International Journal of Food Microbiology 155 (2012) 105–112

Author's personal copy

Fig.

2.Effect

ofEP

Son

theco

mplex

mod

ulus

|G*|(A

)an

dph

asean

gle(B

)of

sorghu

mba

tterswith10

%(–

Δ–)an

d20

%(–

○–)sourdo

ughad

dition

ferm

entedwithL.reuteriY

2(1

)L.reuteriV

IP(2

)an

dW.cibaria

MG1(3

).Ba

tterswithsour-

doug

hco

ntaining

15%su

crose(o

pensymbo

l)wereco

mpa

redwithba

ttersprep

ared

withsourdo

ughwitho

utsu

crose(closedsymbo

l).S

orgh

umba

tterswitho

utsourdo

ughad

dition

served

asaco

ntrol(

–■–).

109S. Galle et al. / International Journal of Food Microbiology 155 (2012) 105–112

Author's personal copy

sourdough simultaneously increased H'm, T'1 and Vt of Y2+ andMG1+ batters. In VIP+ batters a similar trend was observed for T'1but H'm of 10VIP+ did not significantly differ from 20VIP+ batters.Interestingly, 10 VIP+ batters showed higher Vt compared to 20VIP+batters. In comparison to Y2+ and VIP+ batters, MG1+ battersshowed higher H'm and Vt and lower T'1 values. EPS containing batterswere also characterized by slightly lower retention coefficients com-pared to control batter and EPS negative batters (Rc).

3.6. Bread quality evaluation

The effect of EPS formed during sourdough fermentation on sor-ghum bread is summarized in Table 3. Independently from the kindand level of sourdough added, no significant differences (pb0.05)were observed in bake-loss (data not shown). With regard to specificvolume no effect of EPS positive and EPS negative sourdough at anyaddition level was observed compared to the control bread with theexception of 20Y2+ bread showing a significant increase by 10%. How-ever, specific volume of 20Y2+ bread did not differ from 20Y2− bread.

In fresh sorghum bread (day 0 of storage), addition of sourdoughincreased crumb hardness for all strains (Table 3). This effect wasmitigated by EPS positive sourdough, which resulted in a dramatic

decrease of crumb hardness in comparison to EPS negative crumb(pb0.05) to levels comparable to the control (Table 3). However,20VIP+ showed increased crumb hardness compared to the control.Only 20VIP− showed a significantly increased crumb hardness com-pared to 10VIP− (pb0.05). The differences measured in fresh breadpersisted during 5 days of storage, EPS positive bread staled slowerthan EPS negative breads. Interestingly, 20MG1+ bread exhibitedthe softest crumb after storage.

Crumb chewiness of 10MG1+ and 20MG1+ did not differ fromthe control bread but where significantly lower than 10MG1− and20MG1−. Control bread showed similar crumb chewiness comparedto 10VIP+ and 10VIP−. Meanwhile an increase in sourdough addi-tion resulted in a significant increase of crumb chewiness of 20VIP+followed by 20VIP− exhibiting the highest chewiness. Controlcrumb chewiness was also similar to 10Y2+, 20Y2+ and 20Y2−,only 10Y2− showed increased crumb chewiness (pb0.05).

4. Discussion

The positive impacts of EPS on the quality of wheat and rye breadhas been investigated in detail (Brandt et al., 2003; Di Cagno et al.,2006; Kaditzky et al., 2008; Katina et al., 2009; Lacaze et al., 2007;Tieking et al., 2005) but only one study so far investigated the influ-ence of EPS on sorghum bread quality. This study reports for thefirst time a detailed investigation on the influence of three differentEPS forming starter cultures on the rheology and quality of sorghumbread and batters. Sorghum sourdough was produced using dextranformingW. cibariaMG1 and compared to a reuteran forming L. reuteriVIP and a fructan forming L. reuteri Y2.

The molecular weight of dextran was higher than previouslyreported for dextran producing strains of W. cibaria applied in wheatand sorghum fermentations (Di Cagno et al., 2006; Schwab et al.,2008). However, gel filtration and size exclusion chromatographyused in these studies presents some restrictions like total exclusionof very high molar mass compounds (Bourgoin et al., 2008; Rbii etal., 2009). FFF/MALS is a valuable tool for the study of high molecularweights of polymers, because of the increased separation range ofFFF compared to SEC. In fact, molecular masses of dextran, reuteranand fructan determined with FFF was up to two orders of magnitudehigher than molecular masses determined with SEC (Galle, 2011;Galle et al., 2010).

Fig. 3. Complex modulus of sorghum batters with 20% additional sourdough fermentedwith L. reuteri VIP (–Δ–) or L. reuteri Y2 (–○–) or W. cibaria MG1 (–◊–) in the presenceof sucrose. Sorghum batters without sourdough addition served as a control (–■–).

Table 2Dough development and gasouse release in sorghum batters measured with a rheofermentometer at 30 °C for 1.5 h.

Sucroseaddition

SDa

additionDough development Gaseous release

Hm [mm]b T1 [min]c H'm [mm]d T'1 [min]e Vt [mL]f Rc [%]g

L. reuteri Y2 + 10% 24.3±1.13a,b,c,d* 64.5±4.3a,b 99.2±7.4a 54.8±3.2a 958.0±32.5a,b 98.3±0.8aL. reuteri Y2 − 10% 23.7±0.3a,c,b,e 88.5±2.1c 62.1±0.4b 25.5±0.0b 623.0±2.8c 99.5±0.4bL. reuteri Y2 + 20% 22.4±0.4a,e 54.8±3.2a 109.4±4.0c 86.3±1.1c 991.5±46.0b,d 98.9±0.2a,bL. reuteri Y2 − 20% 23.5±0.9a,b,e 52.0±3.5a 79.9±4.7 d 33.5±2.3a 690.7±38.4e 99.3±0.2b,cL. reuteri VIP + 10% 25.0±0.9b,c,d 58.5±2.1a 98.9±0.6a 48.8±1.1 d 902±8.5a 98.7±0.3a,bL. reuteri VIP − 10% 24.5±0.1b,d 89.3±1.1c 63.4±0.4b 27.8±1.1 d,e 630±1.4c 99.3±0.0b,cL. reuteri VIP + 20% 21.8±0.0e 59.3±1.1a 96.8±1.9a 87.0±0.0c 828.0±8.5f 98.8±0.2a,bL. reuteri VIP − 20% 22.0±1.6e 75.8±20.2b,c 62.3±2.8b 31.5±2.1e 578.5±12.0c,b 99. 5±0.1bW. cibaria MG1 + 10% 27.8±0.1f 54.0±2.1a 103.7±a,c 53.3±3.2a,d 1009.5±14.9 d 98.6±0.4a,cW. cibaria MG1 − 10% 25.5±0.2c,d 59.3±3.5a 80.9±2.4 d 33.8±1.1e 723.5±14.9e 99.5±0.00bW. cibaria MG1 + 20% 29.6±1.5f,g 50.3±1.1a 118.4±1.5e 67.5±2.1 d 1260.5±17.7 g 98.3±0.6baW. cibaria MG1 − 20% 26.3±0.1 d,f 89.3±1.1c,d 61.3±5.0b 25.5±2.1b 619.5±13.4c 99.5±0.0bcontrol − 0% 26.2±0.1 d,f 65.3±1.1a,b 60.6±0.9b 29.3±1.1b,e 627.0±5.7c 99.6±0.1b

*Different letters in the same column indicate statistical significance (pb0.05).a SD, sourdough.b Maximum height of dough development curve.c Time of maximum height of dough development curve.d Maximum height of the gaseous release curve.e Time of maximum height of the gaseous release curve.f Total volume of carbon dioxide released by the dough.g Retention coefficient.

110 S. Galle et al. / International Journal of Food Microbiology 155 (2012) 105–112

Author's personal copy

Sucrose addition upon sourdoughs fermented with strains ofL. reuteri led to formation of acetate, which was substantially higherthan in doughs without sucrose or sourdough fermented withW. cibaria. Fructose released from sucrose during fermentation isused as electron acceptor by the activity of mannitol dehydrogynaseand results in increased acetate and mannitol formation (Korakli etal., 2001). W. cibaria lacks of the activity of mannitol dehydrogynaseas no mannitol (data not shown) and only small amounts of acetatewere formed (Galle et al., 2010; Schwab et al., 2008). Increased ace-tate levels in sourdoughs fermented with L. reuteri correspondinglyresulted in higher TTA values and were twice as high compared toW. cibaria fermented sourdoughs. The use of 10% and 20% additionof sourdoughs to sorghum batters, however, did not suffice to changethe TTA of the sorghum batters.

The decrease of |G*| in gluten-free batters was previously relatedto reduced water holding capacity of hydrolyzed proteins, which isinduced by organic acids and enzymes released during sourdough fer-mentation (Renzetti and Arendt, 2009). This study demonstrated thatrheological properties as well as baking loss of sorghum batters werenot significantly altered by inclusion of 10 or 20% EPS-negative sour-doughs. However, inclusion of EPS-positive sourdough significantlychanged dough rheology. Thus, the EPS in sourdough, rather than or-ganic acids, was one of the main factors influencing the |G*| of sor-ghum batter. The role of organic acids and EPS on sorghum doughrheology contrasts with previous observations in wheat sourdoughs.Here, the influence of organic acids was more decisive than the influ-ence of EPS (Galle, 2011). This is in agreement with a previous study(Galle et al., 2011), where HePS from L. buchneri were found to be re-sponsible for the softening of sorghum sourdough but had no effecton wheat dough rheology. A weakening effect on gluten-free doughof some hydrocolloids has already been reported (Lazaridou et al.,2007). In wheat the addition of hydrocolloids revealed dough soften-ing effect due to hindering the gluten–starch interaction (Davidou etal., 1996). However, yet the effect of hydrocolloids on gluten-freedough properties is not completely understood.

Despite the differences in EPS concentration and structure, differ-ent EPS positive sourdoughs had comparable effects on dough rheol-ogy. Nevertheless a steeper increase of |G*| was obtained by additionof dextran containing sourdough (MG1+). According to the weak gelmodel (Gabriele et al., 2001; Moroni et al., 2011) this effect is causedby less interaction of structural components in the dough system andthus showed an increased impact of dextran on the strength of thedough system.

The addition of EPS containing sourdough resulted in a dramaticincrease of gas production by baker's yeast during proofing, whereasno major changes occurred when EPS negative sorghumwas added tothe batters. Sucrose metabolism by lactic acid bacteria not only yieldsEPS but also monosaccharides, which stimulate yeast metabolism andgas production (Gobbetti et al., 1995; Korakli et al., 2001). High

retention coefficient for sorghum batters indicated a good capabilityof the dough to retain produced gas. The increase in gaseous releaseand at the same time no change or decrease of dough developmentcorrespond to the observation made for loaf volumes of the sorghumbreads, as the specific volume of EPS containing sorghum bread wasin the same range as the control breads.

A significant increase in hardness of the fresh and stored sorghumbreadwas obtained upon an addition of 10% and 20% sourdoughwith-out EPS, suggesting that the acidification promoted the firming pro-cess. Moroni et al. (2011) also reported higher crumb firmness insourdough buckwheat bread due to limited starch hydrolysis. It isalso likely, that the long incubation time at low pH in the sourdoughinactivated α-amylases known to reduce crumb firmness by slowingdown starch retrogradation (Kaditzky et al., 2008; Schober et al.,2007). However, addition of EPS positive sourdoughs compensatedthe negative effect of acids and decreased the hardness of fresh andstored sorghum bread. A positive effect was already observed uponan addition of 10% of EPS positive sourdough. The addition of dextranenriched sourdough showed the best shelf life improvement. Dextrandid not only compensate the negative effect of organic acids, it also in-creased the softness of the stored bread relative to the control. Thetechnological functionality of in situ formed dextran was previouslydemonstrated in sorghum baking (Schwab et al., 2008). Firmnesswas lower in bread containing in situ formed dextran, whereas the ad-dition of sourdough containing fructan did not show any effect. In con-trast, this is the first study to report the ability of dextran, reuteran andfructan synthesized in situ during sourdough fermentation to improvethe softness of fresh sorghum bread and shelf life. Delayed staling ismost likely due to retarded starch crystallization by EPS and not to im-proved volume (Davidou et al., 1996), as the volume of the dextranand reuteran and fructan enriched sourdoughs was equal to the con-trol. This confirms previous observations, where EPS formed in situdid not alter sorghum bread volume (Schwab et al., 2008), whereasin wheat baking incorporation of EPS from W. cibaria and L. reuteriaccounted for a major volume increase (Galle, 2011). Thus, the effectof EPS formed during sourdough fermentation on the bread qualityis strongly dependent on the flour used.

Besides EPS formation, oligosaccharideswere formed by all strains usedin this study. FOS and GOS are well described for their prebiotic effects(Chung and Day, 2002; Cummings et al., 2001; Ketabi et al., 2011). Dif-ferent from FOSwhich are digested by yeast during proofing, long chainglucooligosaccharides are still present in the final bread (Schwab et al.,2008). Thus the formation of IMO by W. cibaria MG1 and L. reuteri VIPmay contribute to the nutritional value of the sorghum bread.

Acknowledgements

This study was financially supported by the Food InstitutionalResearch Measure, administered by the Department of Agriculture,

Table 3Baking characteristics and TPA profile of sorghum bread.

Sucrose addition SD* addition Specific volume [mL/g] Chewiness Hardness [N] day 0 Hardness [N] day 2 Hardness [N] day 5

L. reuteri Y2 + 10% 1.8±0.1a** 14.5±0.5a,b,c,d 32.7±2.4a,b 85.8±8.7a,b 107.5±10.1a,b,c,d,eL. reuteri Y2 − 10% 1.8±0.0a,b 18.5±1.1f 42.8±7.4c,d,e 103.8±10.3c 125.8±8.7e,fL. reuteri Y2 + 20% 2.0±0.0c 14.1±1.0a,b,c,d 34.3±7.5a 80.4±10.2a 96.4±9.0a,b,cL. reuteri Y2 − 20% 1.9±0.0a,c 15.7±1.3b,c,d,e 40.6±4.2b,c,d,e 98.0±9.7b,c 125.10±17.8e,fL. reuteri VIP + 10% 1.8±0.1a,b 14.4±0.6a,c,d 34.2±1.5a,b,c 81.1±4.6a 105.0±7.7a,b,c,dL. reuteri VIP − 10% 1.7±0.1b 16.0±1.3c,d,e 39.0±0.7,d 86.9±12.6a,b 127.5±7.8fL. reuteri VIP + 20% 1.8±0.0a 16.3±0.7 d,e 36.3±0.7bc,d,e 84.9±3.1a,b 115.8±17.9c,d,e,fL. reuteri VIP − 20% 1.8±0.1a 20.7±1.7 g 51.4±5.0f 122.4±10.4 d 175.2±2.6 gW. cibaria MG1 + 10% 1.8±0.0a 12.7±0.9a 30.1±3.5a 78.6±7.8a,e 86.1±3.0aW. cibaria MG1 − 10% 1.7±0.1b 17.5±2.8e,f 41.1±5.0e 111.7±10.0a,c 122.4±3.5 d,e,fW. cibaria MG1 + 20% 1.8±0.0a 13.3±2.1a,b 30.0±1.8a 64.0±2.40e 80.2±3.4 hW. cibaria MG1 − 20% 1.9±0.1a,b 17.5±2.0e,f 42.2±2.3e 98.7±3.7b,c 109.1±22.8b,c,d,e,fcontrol − 0% 1.8±0.0a,b 13.8±0.9a,b,c 31.4±2.3a 86.3±15.1b,a 89.5±8.6a,b,h

*SD, sourdough. **Different letters in the same column indicate statistical significance (pb0.05).

111S. Galle et al. / International Journal of Food Microbiology 155 (2012) 105–112

Author's personal copy

Fisheries and Food Ireland, Ireland. Michael Gänzle acknowledgessupport from the Canada Research Chairs Program. The authorswould like to thank Denise Henze, Tom Hannon and Dan Walsh fortheir excellent technical support.

References

Bourgoin, A., Zablackis, E., Poli, J.B., 2008. Characterization of [alpha]-carrageenan solu-tion behavior by field-flow fractionation and multiangle light scattering. Food Hy-drocolloids 22, 1607–1611.

Brandt, M.J., Roth, K., Hammes, W.P., 2003. Effect of an exopolysaccharide produced byLactoabcillus sanfranciscensis LTH1729 on dough and bread quality. Sourdoughfrom Fundamentals to Application. de Vyust L, Vrije Universiteit Brussels (VUB),p. 80.

Chung, C.H., Day, D.F., 2002. Glucooligosaccharides from Leuconostoc mesenteroides B-742 (ATCC 13146): a potential prebiotic. Journal of Industrial Microbiology & Bio-technology 29, 196–199.

Cummings, J.H., Macfarlane, G.T., Englyst, H.N., 2001. Prebiotic digestion and fermenta-tion. American Journal of Clinical Nutrition 73, 415S–420S.

Dal Bello, F., Hertel, C., 2006. Oral cavity as natural reservoir for intestinal lactobacilli.Systematic and applied microbiology 29, 69–76.

Davidou, S., Le Meste, M., Debever, E., Bekaert, D., 1996. A contribution to the study ofstaling of white bread: effect of water and hydrocolloid. Food Hydrocolloids 10,375–383.

De Vuyst, L., De Vin, F., Vaningelgem, F., Degeest, B., 2001. Recent developments in thebiosynthesis and applications of heteropolysaccharides from lactic acid bacteria.International Dairy Journal 11, 687–707.

Di Cagno, R., De Angelis,M., Limitone, A., Minervini, F., Carnevali, P., Corsetti, A., Gänzle,M.,Ciati, R., Gobbetti, M., 2006. Glucan and fructan production by sourdough Weissellacibaria and Lactobacillus plantarum. Journal of Agricultural and Food Chemistry 54,9873–9881.

Gabriele, D., de Cindio, B., D'Antona, P., 2001. A weak gel model for foods. RheologicaActa 40, 120–127.

Galle, S., 2011. Isolation, characterization and application of exopolysaccharides fromlactic acid bacteria to improve the quality of wheat and gluten-free bread. Doctoralthesis, University College Cork, Ireland.

Galle, S., Schwab, C., Arendt, E., Gänzle, M., 2010. Exopolysaccharide-forming Weissellastrains as starter cultures for sorghum and wheat sourdoughs. Journal of Agricul-tural and Food Chemistry 58, 5834–5841.

Galle, S., Schwab, C., Arendt, E.K., Gänzle, M.G., 2011. Structural and rheological charac-terisation of heteropolysaccharides produced by lactic acid bacteria in wheat andsorghum sourdough. Food Microbiology 28, 547–553.

Gobbetti, M., Corsetti, A., Rossi, J., 1995. Interaction between lactic acid bacteria andyeasts in sour-dough using a rheofermentometer. World Journal of Microbiologyand Biotechnology 11, 625–630.

Huettner, E.K., Bello, F.D., Arendt, E.K., 2010. Identification of lactic acid bacteria isolatedfrom oat sourdoughs and investigation of their potential for the improvement of oatbread quality. European Food Research and Technology 230, 849–857.

Kaditzky, S., Seitter, M., Hertel, C., Vogel, R.F., 2008. Performance of Lactobacillussanfranciscensis TMW 1.392 and its levansucrase deletion mutant in wheat doughand comparison of their impact on bread quality. European Food Research andTechnology 227, 433–442.

Kasarda, D.D., 2001. Grains in relation to celiac disease. Cereal foodsWorld 46, 209–210.Katina, K., Maina, N.H., Juvonen, R., Flander, L., Johansson, L., Virkki, L., Tenkanen, M.,

Laitila, A., 2009. In situ production and analysis of Weissella confusa dextran inwheat sourdough. Food Microbiology 26, 734–743.

Ketabi, A., Dieleman, L., Gänzle, M.G., 2011. Influence of isomalto-oligosaccharides onintestinal microbiota in rats. Journal of Applied Microbiology 110, 1297–1306.

Korakli, M., Rossmann, A., Gänzle, M.G., Vogel, R.F., 2001. Sucrose metabolism andexopolysaccharide production in wheat and rye sourdoughs by Lactobacillussanfranciscensis. Journal of Agricultural and Food Chemistry 49, 5194–5200.

Lacaze, G., Wick, M., Cappelle, S., 2007. Emerging fermentation technologies: develop-ment of novel sourdoughs. Food Microbiology 24, 155–160.

Lazaridou, A., Duta, D., Papageorgiou, M., Belc, N., Biliaderis, C.G., 2007. Effects of hydro-colloids on dough rheology and bread quality parameters in gluten-free formula-tions. Journal of Food Engineering 79, 1033–1047.

McCarthy, D.F., Gallagher, E., Gormley, T.R., Schober, T.J., Arendt, E.K., 2005. Applicationof response surface methodology in the development of gluten-free bread. CerealChemistry 82, 609–615.

Meroth, C.B., Walter, J., Hertel, C., Brandt, M.J., Hammes, W.P., 2003. Monitoring thebacterial population dynamics in sourdough fermentation processes by usingPCR-denaturing gradient gel electrophoresis. Applied and Environmental Microbi-ology 69, 475.

Moore, M.M., Dal Bello, F., Arendt, E.K., 2008. Sourdough fermented by Lactobacillusplantarum FST 1.7 improves the quality and shelf life of gluten-free bread. Europe-an Food Research and Technology 226, 1309–1316.

Moroni, A.V., Bello, F.d., Arendt, E.K., 2009. Sourdough in gluten-free bread-making: anancient technology to solve a novel issue? Food Microbiology 7, 676–684.

Moroni, A.V., Dal Bello, F., Zannini, E., Arendt, E.K., 2011. Impact of sourdough on buck-wheat flour and batter and bread: biochemical, rheological and textural insights.Journal of Cereal Science 54, 195–202.

Pruska-Kędzior, A., Kędzior, Z., Gorący, M., Pietrowska, K., Przybylska, A., Spychalska, K.,2008. Comparison of rheological, fermentative and baking properties of gluten-freedough formulations. European Food Research and Technology 227, 1523–1536.

Rbii, K., Violleau, F., Guedj, S., Surel, O., 2009. Analysis of aged gelatin by AFlFFF-MALS:identification of high molar mass components and their influence on solubility.Food Hydrocolloids 23, 1024–1030.

Renzetti, S., Arendt, E.K., 2009. Effects of oxidase and protease treatments on the bread-making functionality of a range of gluten-free flours. European Food Research andTechnology 229, 307–317.

Renzetti, S., Dal Bello, F., Arendt, E.K., 2008. Microstructure, fundamental rheology andbaking characteristics of batters and breads from different gluten-free flours trea-ted with a microbial transglutaminase. Journal of Cereal Science 48, 33–45.

Schober, T.J., Messerschmidt, M., Bean, S.R., Park, S.-H., Arendt, E.K., 2005. Gluten-freebread from sorghum: quality differences among hybrids. Cereal Chemistry 82,394–404.

Schober, T.J., Bean, S.R., Boyle, D.L., 2007. Gluten-free sorghum bread improved bysourdough fermentation: biochemical, rheological, and microstructural back-ground. Journal of Agricultural and Food Chemistry 55, 5137–5146.

Schwab, C., Mastrangelo, M., Corsetti, A., Gänzle, M., 2008. Formation of oligosaccha-rides and polysaccharides by Lactobacillus reuteri LTH5448 and Weissella cibaria10M in sorghum sourdoughs. Cereal Chemistry 85, 679–684.

Svensson, L., Sekwati Monang, B., Lopez-Lutz, D., Schieber, A., Gänzle, M., 2010. Pheno-lic acids and flavonoids in non-fermented and fermented red sorghum (Sorghumbicolor (L.) Moench). Journal of Agricultural and Food Chemistry 58, 9214–9220.

Tieking, M., Gänzle, M.G., 2005. Exopolysaccharides from cereal-associated lactobacilli.Trends in Food Science & Technology 16, 79–84.

Tieking, M., Korakli, M., Ehrmann, M.A., Gänzle, M.G., Vogel, R.F., 2003. In situ productionof exopolysaccharides during sourdough fermentation by cereal and intestinal iso-lates of lactic acid bacteria. Applied and Environmental Microbiology 69, 945–952.

Tieking, M., Kaditzky, S., Valcheva, R., Korakli, M., Vogel, R.F., Gänzle, M.G., 2005. Extra-cellular homopolysaccharides and oligosaccharides from intestinal lactobacilli.Journal of Applied Microbiology 99, 692–702.

van Hijum, S., Kralj, S., Ozimek, L.K., Dijkhuizen, L., van Geel-Schutten, I.G.H., 2006.Structure–function relationships of glucansucrase and fructansucrase enzymesfrom lactic acid bacteria. Microbiology and Molecular Biology Reviews 70, 157.

112 S. Galle et al. / International Journal of Food Microbiology 155 (2012) 105–112