- Email: [email protected]
Bran fermentation as a means to enhance technological properties and bioactivity of rye
ARTICLE IN PRESS FOOD MICROBIOLOGY Food Microbiology 24 (2007) 175–186 www.elsevier.com/locate/fm
Bran fermentation as a means to enhance technological properties and bioactivity of rye K. Katinaa, A. Laitilaa, R. Juvonena, K.-H. Liukkonena, S. Kariluotoc, V. Piironenc, R. Landbergd, P. A˚mand, K. Poutanena,b, a VTT Biotechnology, VTT Technical Research Centre of Finland, Tietotie 2, P.O. Box 1000, 02044 VTT, Espoo, Finland University of Kuopio, Food and Health Research Centre, Department of Public Health and Clinical Nutrition, P.O. Box 1627, FI-70211 Kuopio, Finland c Department of Applied Chemistry and Microbiology, University of Helsinki, P.O. Box 27, FIN-00014, Finland d Department of Food Science, SLU, P.O. Box 7051, Uppsala, Sverige
b
Available online 12 September 2006
Abstract Response surface methodology was applied to study the effects of fermentation on the levels of phytochemicals (folates, phenolic compounds, alkylresorcinols) and on the solubilization of pentosans in rye bran from native and peeled grains. Furthermore, the microbial composition of the brans before and after fermentation was studied. Peeling reduced the microbial load and lower microbial counts were detected in the fermentation experiments carried out with the bran from peeled grains. High temperature and long fermentation time favoured the growth of indigenous lactic acid bacteria (LAB), and a diverse microbial community was detected. The brans contained low levels of aerobic spore-forming bacteria, but their number was not increased during the fermentations. Fermentation of both brans increased the levels of folates, easily extractable total phenolics and free ferulic acid. During fermentation of bran from native grains, the levels of alkylresorcinols slightly increased but during fermentation of bran from peeled grains they decreased. Significant increase in soluble pentosans was established in both types of rye bran fermentations. Enhanced bioactivity and solubilization of pentosans with limited microbial growth were obtained after 12–14 h fermentation at 25 1C. The results suggest that fermentation is a potential bioprocessing technology for improved technological properties and bioactivity of rye bran. r 2006 Published by Elsevier Ltd. Keywords: Rye bioactivity; Phytochemicals; Sourdough fermentation; Pentosans
1. Introduction Consumption of foods rich in whole grains and cereal fibre has in epidemiological studies been shown to reduce the risk of chronic diseases such as diabetes, cardiovascular disease and certain cancers, as reviewed by Murtaugh et al. (2003) and Jacobs and Gallaher (2004), and shown by Larsson et al. (2005). However, there is a large gap between the dietary recommendations and current intake of whole grains and dietary fibre (Lang and Jebb, 2003). The need and challenge for cereal technologists is to develop
Corresponding author. VTT Biotechnology, VTT Technical Research Centre of Finland, Tietotie 2, P.O. Box 1000, 02044 VTT, Espoo, Finland. Fax: +358 9 4552103. E-mail address: kaisa.poutanen@vtt.fi (K. Poutanen).
0740-0020/$ - see front matter r 2006 Published by Elsevier Ltd. doi:10.1016/j.fm.2006.07.012
technologies enabling to lower consumer barriers to frequent use of high-fibre and wholegrain foods. Cereal brans are important ingredients providing dietary fibre, and wheat bran is one of the most common raw materials for increasing the level of insoluble dietary fibre in baking. In the Nordic countries, rye is the most important source of dietary fibre. Dietary fibre and bioactive compounds such as alkyresorcinols, lignans, phenolic acids, phytosterols, tocopherols and—tocotrienols and folates are concentrated in the bran fraction of rye (Liukkonen et al., 2003; Ross et al., 2004). This is why it would be interesting to develop nutritionally optimized cereal foods and new ingredients from rye bran. The main part of dietary fibre in bran is insoluble, which influences the digestibility and bioavailability of nutrients and phytochemicals. The outer layers of grain contain cellulose and lignin which influence both the taste and mouthfeel of
ARTICLE IN PRESS 176
K. Katina et al. / Food Microbiology 24 (2007) 175–186
the bran. The properties of rye bran thus restrict its full exploitation in different consumer foods. Fermentation with well-characterized starter cultures— yeast or lactic acid bacteria (LAB)—is a potential means to improve the palatability and processability of brans and wholemeal flours (Salovaara, 2004; Katina et al., 2006). Furthermore, bran fermentation could assist in microflora management and improve the microbiological safety of bran. This is important, because brans are known to contain more micro-organisms than endosperm flours, and they could also be a source of spoilage bacteria and fungi (Rosenquist and Hansen, 1995). It has been pointed out that fermentation is a combined action of added starters cultures and the micro-organisms naturally present in grains (Salovaara, 2004). Both endogenous and microbial originated enzymes are shown to be concentrated to the outer grain layers (Dornez et al., 2006; Gys et al., 2004), and have thus the possibility for action during the bran fermentation. Both acidification and activity of enzymes, especially xylan-degrading enzymes, were shown to contribute to the solubilization of arabinoxylans during the baking process of wholemeal rye (Boskov-Hansen et al., 2002). This led to decreased level of dietary fibre in the resulting bread. Solubilization of pentosans of flour, and especially transformation of water unextractable arabinoxylan to water extractable arabinoxylan, has been reported to improve bread volume and texture in wheat baking (Courtin and Delcour, 2002) and in high-fibre wheat baking (Katina et al., 2006). The influence of processing conditions or starter culture, however, on solublization of pentosans during wheat sourdough or bran fermentation is not known. In wheat sourdough baking, the influence of commercial pentosanases on the molecular weight of arabinoxylans was shown to be dependent on the starter type (Devesa and Martı´ nezAnaya, 2001). Many biochemical changes affecting nutritional quality as well as texture and flavour of wheat flour occur during sourdough baking, as reviewed by Katina (2005). Levels of folate and easily extractable phenolic compounds have been shown to increase (Liukkonen et al., 2003; Kariluoto et al., 2004), whereas levels of phytate (Fro¨lich et al., 1986; Larsson and Sandberg, 1991), alkylresorcinols (ARs) (Verdeal and Lorenz, 1977) and tocopherols (Piironen et al., 1987) have been reported to reduce in sourdough baking process. We have previously shown that enzymatic pretreatments and fermentation of wheat bran are efficient in improving texture, the sensory quality and stability of breads supplemented with 20% of wheat bran (SalmenkallioMarttila et al., 2001; Katina et al., 2006). We have also shown that by careful fractionation of rye bran in the milling process, the sensory quality can be tailored while still retaining many of the bioactive bran compounds (Heinio et al., 2003; Liukkonen et al., 2003). The aim of the current study was to develop fermentation as a means to improve the properties of rye bran. We established the
influence of fermentation conditions and type of bran on the microbial community and levels of bioactive compounds and solubilized pentosans in rye. 2. Material and methods 2.1. Experimental design and modelling Fermentation time (Ti, h) and temperature (Te, 1C) were selected as independent variables to study the effects of the fermentation conditions on properties of bran properties. A central composite design was used to plan experiments and three replicates were made at the centre point of design to allow estimation of the pure error at sum of the square. The experimental design and obtained values are presented in Table 1a and b. The results were analysed by a multipleregression method (PLS), which describes the effects of variables in second-order polynomial models. For each response (microbes, bioactive compounds and pentosans) a quadratic model was used. Regression analysis was made and the response surfaces were plotted with the Modde 4.0 and 6.0 (Umetrics AB, Umea˚, Sweden). The fit of the model to the experimental data was given by the coefficient of determination, R2, which explains the extent of the variance in a modelled variable that can be explained with the model. Each model was validated by calculating the predictive power of model, Q2, which is a measure of how well the model will predict the responses for a new experimental condition. Q2 is based on prediction of residual sum of squares (PRESS). For determining Q2, the computations were repeated several times, each time omitting different objects from the calculation of the model. PRESS was then computed as the squared difference between observed Y and predicted values (cross-validation of R2). Large values of Q2 indicate that the model had good predictive ability and will have small prediction errors. Q2 should be 40.5 if conclusions are to be drawn from the model. Only the models with high enough R2- and Q2- values were used to describe influence of fermentation time and temperature on the amount of microbes, levels of bioactive compounds and the amount of pentosans. Adequate models (R2 40:7 and Q2 40:5) could be obtained with total amount of LAB, folates, free phenolics (Folin-C), free ferulic acids and, alkylresorcinols and with soluble pentosans. With bound phenolics (Folin-C), bound ferulic acid or with total amount of pentosans, adequate models could not been obtained due to low systematic variation in measured values in the experimental region. 2.2. Raw materials Experiments were carried out with the Finnish Rye Cultivar, Amilo from 2003. Part of the rye grains was first peeled with a Buhler horizontal grinding/peeling machine to remove about 11% of the outer layer of the grain. Both native and peeled rye grains were milled
ARTICLE IN PRESS K. Katina et al. / Food Microbiology 24 (2007) 175–186
177
Table 1 Experiment number
Time/h
Temperature/ 1C
pH
TTA
Folates (mg/ 100 g)
FAa (mg/ 100 g)
PHC (MeOH)b (mg/100 g)
Arsc (mg/ 100 g)
SP (%)d
(A) Experimental design and obtained values for responses in fermentation 1 6 20 6.7 3.9 2 20 20 6.5 5.4 3 6 35 6.6 5.4 4 20 35 4.3 16.2 5 6 27.5 6.7 4.9 6 20 27.5 5.0 8.5 7 13 20 5.7 5.6 8 13 35 5.0 7.5 9 13 27.5 6.4 5.9 10 13 27.5 6.1 6.3 11 13 27.5 6.5 5.5
of native bran 147 19 188 30 136 28 240 7 182 24 273 10 157 28 202 8 190 30 221 20 221 20
205 263 257 383 241 305 251 313 259 260 260
290 284 311 282 307 300 311 323 306 306 —
1.6 1.8 1.9 1.1 1.8 1.4 1.7 1.3 1.8 1.7 1.8
(B) Experimental design and obtained values for responses in fermentation 1 6 20 6.5 3.8 2 20 20 6.4 4.0 3 6 35 6.5 3.8 4 20 35 4.1 12.4 5 6 27.5 6.4 3.6 6 20 27.5 5.9 5.4 7 13 20 6.4 3.7 8 13 35 4.6 8.2 9 13 27.5 6.3 4.2 10 13 27.5 6.3 4.2 11 13 27.5 6.3 4.2
of bran from peeled grains 147 19 188 30 136 28 240 7 182 24 273 10 157 28 202 8 190 30 221 20 221 20
205 263 257 383 241 305 251 313 259 260 260
191 204 195 160 198 156 162 161 202 161 —
1.9 2.5 2.6 2.6 2.6 2.9 2.7 3.1 2.7 2.8 2.75
C17:0
C19:0
C21:0
C23:0
C25:0
Total (mg/g DMe)
C17:0/C21:0
(C) Total alkylresorcinol content (mg/g DM) and relative homologue composition in fermentation of (A) bran from native grain and (B) bran from peeled grain (A) Bran from native grain Experiment number 1 24 32 24 10 9 2903 2 25 33 24 10 8 2835 1.0 3 24 33 24 10 9 3114 1.0 4 24 33 25 10 8 28194 1.0 5 24 32 24 11 9 3066 1.0 6 24 33 24 10 8 2966 1.0 7 24 32 24 11 9 3112 1.0 8 25 33 24 10 8 3230 1.0 9 25 33 24 10 8 3060 1.0 10 24 32 24 11 9 3063 1.0 (B) Bran from peeled grain 1 26 2 26 3 25 4 25 5 25 6 26 7 26 8 26 9 26 10 26 a
35 35 34 35 34 35 35 36 35 35
23 24 24 24 24 24 23 23 23 23
FA ¼ ferulic acid. PHC ¼ phenolic compounds after MeOH extraction. c ARs ¼ alkyresorcinols. d SP ¼ soluble pentosans. e DM ¼ dry matter. b
9 9 10 9 10 9 9 9 9 9
7 6 8 7 8 7 7 6 7 7
1910 2041 1951 1602 1982 1566 1623 1618 2026 1610
1.1 1.1 1.0 1.1 1.0 1.1 1.1 1.1 1.1 1.1
ARTICLE IN PRESS 178
K. Katina et al. / Food Microbiology 24 (2007) 175–186
at a moisture level of 15–16% with a Buhler roller mill to have bran fractions. The bran fractions comprised 15% and 20% of the original native and peeled grains, respectively. Peeling increased falling number of the bran from 75 to 125.
2.3. Preparation of bran fermentations Fresh Baker’s yeast was obtained from Suomen Hiiva, Finland. To obtain a dough yield of 450, 600 g of rye bran (native or peeled) was mixed carefully with 2100 g of tap water, and with 7.5 g of fresh Baker’s yeast (107 cfu/g) (Suomen Hiiva., Finland) in a large beaker (2000 ml), which was covered with aluminium foil. Fermentations were carried out according to the experimental design. Fresh samples were taken from unfermented and fermented bran for microbiological analysis. In addition, samples were frozen for later measurements of pH, and total titratable acidity (TTA). Fermented bran samples were also freeze-dried for analysis of bioactive compounds. Fermentations were done in duplicate and repeated twice.
2.4. Microbiological analyses Samples for the microbiological analyses were taken from native and peeled bran samples and after fermentation according to the experimental design. The following microbial groups were analysed from homogenized bran samples: aerobic heterotrophic bacteria, LAB and fungi. A sample of 10 g was homogenized for 10 min with 90 ml of sterile saline in a Stomacher Lab Blender 400 (Seward Medical, London, UK). Aerobic heterotrophic bacteria were determined on plate count agar (PCA, Difco Laboratories, Detroit, USA), and samples were incubated in aerobic conditions at 30 1C for 2–3 days. The number of LAB was determined on MRS agar (Oxoid) and samples were incubated in anaerobic conditions at 30 1C for 5 days. To prevent fungal overgrowth of bacterial determinations, 0.001% cycloheximide (Sigma Chemical, St. Louis, Missouri, USA) was added to PCA, and MRS media. Yeast and moulds were determined on YM agar (Difco Laboratories). Samples were incubated in aerobic conditions at 25 1C for 3–5 days. Chlortetracycline and chloramphenicol (both at 0.01%) were added to YM medium to prevent bacterial growth. In addition, 0.02% of Triton-X 100 (BDH) was used to limit the spreading of fungal colonies on YM-agar. To determine spore formers, 5 ml of homogenized sample was heated in a water bath of 80 1C for 10 min for inactivation of vegetative cells. Bacillus spp. were enumerated on tryptone soy agar (TSA, Oxoid) and plates were incubated at 30 1C for 3 days. The bacteria and yeast results are expressed as colony-forming units/ gram fermented bran (cfu/g).
2.5. Characterization and identification of LAB and aerobic spore-forming bacteria To identify dominating different colony types were selected from MRS and TSA plates (a maximum of five colonies per sample). The isolates were characterized by Gram staining and catalase test and subsequently purified by successive sub-culturing on the antibiotic-free isolation media. Ribotyping was performed using the standard method of the automated ribotyping device RiboPrinters Microbial Characterization System (DuPont QualiconTM, USA) and EcoRI restriction enzyme (Bruce, 1996). For other genetic analyses, genomic DNA was isolated from 1–2-day-old plate cultures by suspending 2 ml loop-full of cell mass in 0.5 ml PCR-grade water followed by bead-beating with 200-mm glass beads (0.1 g ml1, Sigma) in the RibolyserTM instrument (1.5 min 6.0 m/s, Hybaid, UK) and removal of cell debris by centrifugation (11,000g, 3 min). RAPD-PCR analysis was performed with OPA-02 (50 -TGCCGAGCTG-30 ) and OPA-03 (50 -AGTCAGCCAC-30 ) primers in 50 ml reactions containing 0.4 mM primer (SigmaGenosys, Cambridge, UK), 0.2 mM dNTP (Finnzymes, Espoo, Finland), 3 U DyNAzyme II DNA polymerase (Finnzymes) and 5 ml template DNA in a 1 enzyme buffer (final concentration of 10 mM Tris–HCl, 1.5 mM MgCl2, 50 mM KCl, 0.1% Triton X-100, pH 8.8). The PCR programme in a thermocycler (UNO II, Biometra GmbH, Goettingen, Germany) consisted of an initial denaturation for 5 min at 95 1C, followed by 35 cycles (1 min at 95 1C, 1 min at 32 1C, 2 min at 72 1C), a final extension for 10 min at 72 1C and cooling to 4 1C. To identify the LAB isolates representing different fingerprint types, the 16S rRNA gene was amplified using the conserved primers BSF 8/20 (50 -AGAGTTTGATCCTGGCTCAG-30 ) and BSR1541/ 20 (50 -AAGGAGGTGATCCAGCC GCA-30 ) (Wilmotte et al., 1993) in 50 ml reactions containing 0.5 mM primers, 0.2 mM dNTP (Finnzymes), 2.5 U DyNAzyme II DNA polymerase and 5 ml template DNA in the 1 enzyme buffer. The PCR programme consisted of an initial denaturation for 5 min at 94 1C, followed by 35 cycles (45 s at 95 1C, 45 min at 56 1C, 2 min at 72 1C), a final extension for 7 min at 72 1C and cooling to 4 1C. Amplification products were purified using a QiaQuick PCR purification kit (Qiagen, Mississaugua, Ontario, Canada) according to the instructions. DNA purity and yield were estimated electrophoretically in 1% agarose gels. Sequencing reactions of the PCR products were performed with an ABI PRISM BigDye terminator cycle sequencing kit v.3.1 (Applied Biosystems, Foster City, CA) according to the instructions using the primer BSF 8/20. The products were electrophoresed with an ABI PRISM 3100 sequencer (Applied Biosystems) and checked and edited with the Chromas programme (Technelysium, Helensvale, Australia) and compared to the GenBank database (www.ncbi.nlm.nih.gov) using the BLAST (basic local alignment search tool) algorithm.
ARTICLE IN PRESS K. Katina et al. / Food Microbiology 24 (2007) 175–186
2.6. pH and TTA Frozen bran samples were thawed overnight in a refrigerator. The pH value was measured from an aliquot of 10 g of fermented bran blended with 100 ml distiled water in a TitroLine Alpha471217 (Schott, Germany). For the determination of TTA, this suspension was titrated against 0.1 M NaOH to a final pH value of 8.5 using TitroLine Alpha. TTA was expressed as the amount of NaOH used (ml). All samples were analysed in duplicate. 2.7. Analysis of bioactive compounds and pentosans Folates were analysed by a microbiological assay method including extraction and trienzyme treatment as described by Kariluoto et al. (2004). ARs were extracted in a hot 1-propanol:water mixture according to Ross et al. (2003). Extracts were then concentrated and analysed by GC, without further clean up or derivatization, according to Ross et al. (2001). Total phenolic acids were analysed after alkaline and free phenolic acids after ethylacetate extraction by HPLC (Bartolome´ and Go´mez-Cordove´s, 1999). Bound phenolic acids were calculated by subtracting free phenolic acids from total phenolic acids. Total phenolics were analysed by Folin–Ciocalteau method: methanolic extracts of freeze-dried and ground rye samples were obtained by an ultrasonication-assisted extraction procedure. The residue was further extracted at alkaline conditions. The content of total phenolic compounds (gallic acid equivalents) in methanolic (free phenolics) and alkaline (bound phenolics) extracts was determined using Folin–Ciocalteau procedure (Singleton et al., 1999). Total amount of pentosans and soluble pentosans were analysed by colorimetric method (Rouau and Surget, 1994). All analyses were done in duplicate and the obtained average values of bioactive compounds are expressed on dry weight basis. 3. Results 3.1. Characterization of native and peeled brans The chemical composition and the microbial counts in rye bran from native and peeled grains are shown in Table 2. Removal of outer grain layers decreased the relative amounts of protein, insoluble dietary fibre, phenolic compounds and folates in peeled bran fraction, and increased the relative amount of starch. Slightly lower counts of aerobic bacteria, LAB, yeast and moulds were detected in bran from peeled grain as compared to the native bran. 3.2. Microbes in bran fermentations Commercial baker’s yeast was added at level of 2–3 107 cfu/g, and it dominated the bran fermentations.
179
Table 2 Chemical composition and number of bacteria and fungi (cfu/g) in native and peeled rye bran Rye bran
Macrocomponents (g/100 g dw) Protein Starch Dietary fibre Of which soluble Pentosans Of which soluble
Native
Peeled
18.6 30.2 35.8 5.3 15.9 0.91
14.8 44.4 27.8 5.3 14.4 0.97
Bioactive compounds Total phenolics (Folin-C) (mg/100 g dw) Free (MeOH extractable) Bound (NaOH extractable) Phenolic acids (ferulic acid) (mg/100 g dw) free (water extractable) bound (NaOH extracable) Alkylresorcinols (mg/100 g dw) Lignans (mg/100 g dw) Folates (mg/100 g dw)
584 200 383 274 6.8 267 285 5.8 0
415 147 268 215 2.3 213 222 4.9 81
Microbial groups (cfu/g) Total aerobic heterotrophic bacteria Aerobic sporeforming bacteria Lactic acid bacteria Yeasts Moulds
5 106 6 102 1 103 3 103 9 102
3 105 3 102 4 102 7 102 5 101
No other fungi (yeasts or moulds) were detected during fermentation. Only low levels of LAB were found in native and peeled bran fractions (Table 2), but they increased markedly during fermentation (Fig. 1). The growth of indigenous LAB community was greatly dependent on the fermentation time and temperature. Multiplication of LAB was pronounced when the fermentation temperature was raised from 20 to 35 1C, and the highest viable counts, 2 109 cfu/ g, were determined in bran from native grains after 20 h fermentation at 35 1C. Intensive LAB growth was reflected in the acidification patterns in bran fermentation from both native and peeled grains, as seen in Fig. 2a and b. Development of acidity (measured as pH and TTA) was clearly dependent on the fermentation temperature and time; strong acidity development occurred only if temperature was 425 1C and time 412 h in both type of fermentations. It is noteworthy that more restricted production of acids occurred in fermentation of bran from peeled grains, as compared to fermentation of the native bran. The highest TTA value was 12 ml NaOH 0.1 ml/10 g of dough for fermentation of bran from peeled grain and 16 for fermentation of native bran. However, with both bran types pH decreased from 6.6 to 4.1 after 20 h fermentation at 35 1C (Fig. 2a). As seen in Figs. 1 and 2b, the modelling of LAB growth and TTA produced different surface response plots for
ARTICLE IN PRESS 180
K. Katina et al. / Food Microbiology 24 (2007) 175–186
Fig. 1. Influence of fermentation time and temperature on growth of LAB (cfu/g) in fermented bran made from (A) native rye bran (B) peeled rye bran. Ti ¼ time and Te ¼ temperature.
Fig. 2. (a) Influence of fermentation time and temperature on pH values of fermented bran made from (A) native rye bran (B) peeled rye bran. Ti ¼ time and Te ¼ temperature. (b) Influence of fermentation time and temperature on TTA values (ml NaOH/10 g of dough) of fermented bran made from (A) native rye bran (B) peeled rye bran. Ti ¼ time and Te ¼ temperature.
fermentations of bran from native and peeled rye grains, indicating differences in the composition of LAB community. To determine the predominating LAB in the fermentation process of native bran, a total of 42 LAB were subjected to RAPD-PCR fingerprinting and the isolates representing different fingerprint types were identified by partial 16S rRNA gene sequencing. The fermentation of native bran developed a diverse LAB community which was characteristic to each temperature and differed from the major community present in the bran prior to fermentation (Table 3). In general, LAB diversity increased with fermentation temperature, and at the highest temperature eight species or genotypes were identified. The highest species diversity was found among the facultative heterofermentative lactobacilli. The obligate
heterofermentative population was mainly composed of Leuconostoc species. In addition, Lactobacillus fermentum was isolated from the native brans fermented at 35 1C. Homofermentative cocci, Enterococcus hermanniensis, Pediococcus pentosaceus and/or Lactococcus lactis, were detected during the fermentations at 27.5 1C and 35 1C. Lactobacillus sakei was detected in all fermentations independent of the temperature. As seen from Table 3 some of the isolates could not be identified to species level due to the high genetic similarity of the closely related species. In contrast to fermentations with native bran, a uniform LAB community was detected in fermentations with brans from peeled grains, in which the homofermentative Enterococcus were the predominant species (data not
ARTICLE IN PRESS K. Katina et al. / Food Microbiology 24 (2007) 175–186
181
Table 3 Lactic acid bacteria identified from various stages of fermentation of native bran1 LABa
Native bran
Yeast+bran prior to fermentation
Fermentation process (6–20 h) at 20 1C
Homo-fermentative
Obligate heterofermentative
Facultative heterofermentative
E. casseliflavus/E. gallinarum/ E. saccharolyticus (99.7%b) E. hermanniensis (99.4) Leuc. citreum (99.7%)
27.5 1C
35 1C
E. hermanniensis (99.4%)
Lc. lactis (99.8%)
P. pentosaceus (99.7%) Leuc. mesenteroides (99.7%)
Leuc. pseudomesenteroides/ Leuc. mesenteroides (X99.5%)
Leuc. mesenteroides (99.7%)
L. paraplantarum/L. plantarum/L. pentosus (X99%) L. casei/L. paracasei (99.8%)
L. curvatus/L. graminis (X99.7%)
L. curvatus/L. graminis (X99.5%)
L. plantarum/L. pentosus/ paraplantarum (X99.5%) L. sakei (100%)
L. plantarum/L. pentosus/L. paraplantarum (X98.9%) L. sakei (100%) L. sakei/L. curvatus/ L. graminis (98.7%)
L. fermentum (99.7%) L. curvatus/L. graminis (X99.3%) L. plantarum/L. pentosus/L. paraplantarum (X99.5%) L. curvatus (99.8%) L. sakei (100%) L. sakei/L. curvatus/ L. graminis (98.2%)
a
E., Enterococcus; L., Lactobacillus; Leuc., Leuconostoc; Lc., Lactococcus; P., Pediococcus. Similarity of partial 16S rRNA gene sequence to closest relative. When more than one species are presented the analysis was not conclusive at species level. b
shown). Two Enterococcus species, showing different DNA-fingerprints, were isolated from the peeled bran fermentations. Based on the partial 16S rRNA gene sequences, Enterococcus sp. 1 gave 100% similarity to Enterococcus casseliflavus and Enterococcus flavescens, whereas Enterococcus sp. 2 gave 99.4% similarity to Enterococcus saccharolyticus and Enterococcus gallinarum. These closely related species could not be separated using 16S rRNA sequencing. Intensive growth of LAB and concomitant increase of organic acids as well as other possible antimicrobial substances restricted the growth of aerobic heterotrophic bacteria (data not shown). The total viable count was 3 105 cfu/g at the beginning of fermentation with bran from peeled grains, and decreased to 6 104 cfu/g after 20 h fermentation at 35 1C. A similar trend was observed with the native bran. Although low numbers of spore-forming bacteria (Bacillus spp.) were detected in the brans, their number did not increase during the fermentation process. The maximum of 50 cfu/g was detected in fermentations with the bran from peeled grains and maximum of 200 cfu/g in fermentations with the bran of native grains. We identified three different Bacillus species from bran fermentations, namely Bacillus licheniformis (16S RNA gene similarity 100%), Bacillus sonorensis (16S RNA gene similarity 100%), and Paenibacillus amylolyticus (16S RNA gene similarity 100%).
3.3. Effect of bran fermentation on bioactive compounds As seen in Fig. 3 fermentation temperature expressed quadratic effect on the level of folates in fermentation with native bran (Fig. 3). As compared to the level of folates in unfermented bran, 1.9 fold increase was obtained by fermenting native rye bran at 28 1C for 20 h. In fermentation of rye bran from peeled grains, the level of folates increased as a function of increasing time and temperature. The level of folates increased 2.3 fold by fermenting bran from peeled grains at 35 1C for 20 h. The level of free phenolics, determined by Folin C method after methanol extraction, increased linearly with of increasing fermentation time and temperature (Fig. 4). Fermentation of native rye bran increased the amount of free phenolics by 90%, but that of bran from peeled grains only by 15–30%. The effect of fermentation conditions was also studied on the levels of single groups of phenolic compounds. The highest increase in the amount of free phenolic acids (ferulic acid) was detected in fermentation of native bran both at high temperature (35 1C) and short fermentation time (6 h) and low temperature (20 1C) and long fermentation (20 h) time (Fig. 5). At high temperature and long fermentation time, the level of free phenolic acids was very similar to that of nonfermented bran. In fermentation of bran from peeled grains, ferulic acid showed a very similar pattern, but absolute changes were at lower level.
ARTICLE IN PRESS 182
K. Katina et al. / Food Microbiology 24 (2007) 175–186
Fig. 3. Influence of fermentation time and temperature on total amount of folates (mg/100 g) in fermented bran made from (A) native rye bran (B) peeled rye bran. UF ¼ amount of folates in unfermented rye bran (native or peeled). Ti ¼ time and Te ¼ temperature.
Fig. 4. Influence of fermentation time and temperature on total amount of phenolic compounds after MeOH extraction (free phenolics mg/100 g) in fermented bran made from (A) native rye bran (B) peeled rye bran. UF ¼ amount of MeOH extractable total phenolics in unfermented rye bran (native or peeled). Ti ¼ time and Te ¼ temperature.
Fig. 5. Influence of fermentation time and temperature on total amount of free ferulic acid (mg/100 g) in fermented bran made from (A) native rye bran (B) peeled rye bran. UF ¼ amount of free ferulic acid in unfermented rye bran (native or peeled). Ti ¼ time and Te ¼ temperature.
During the fermentation of native bran, the levels of ARs only changed slightly (Fig. 6). In both cases, fermentation time had a quadratic effect on the level of ARs, so that the change in alkyresorcinols was largest at a time range of 12–14 h independent on the fermentation temperature. However, quadratic effect was opposite for native and peeled bran as highest level was obtained for native bran and lowest for peeled bran in the experimental region in question. No significant change in AR homologue (alkyl side chains C17:0–C25:0) pattern was seen during fermentation (Table 1c).
3.4. Pentosans The amount of soluble pentosans increased twofold in fermentation of both types of bran in optimum conditions (Fig. 7). However, the influence of fermentation conditions differed to some extent for native and peeled bran: e.g. long fermentation time at high temperature did not enhance solubilization of pentosans on native rye bran, but doubled the amounts in fermentation of bran from peeled grains. Also, the overall levels of soluble pentosans were higher in fermentations of bran from the peeled grains.
ARTICLE IN PRESS K. Katina et al. / Food Microbiology 24 (2007) 175–186
183
Fig. 6. Influence of fermentation time and temperature on total amount of alkylresorcinols (mg/100 g Ars) in fermented bran made from (A) native rye bran (B) peeled rye bran. UF ¼ amount of Ars in unfermented rye bran (native or peeled). Ti ¼ time and Te ¼ temperature.
Fig. 7. Influence of fermentation time and temperature on total amount of soluble pentosans (% dw) in fermented bran made from (A) native rye bran (B) peeled rye bran. UF ¼ amount of soluble pentosans in unfermented rye bran (native or peeled). Ti ¼ time and Te ¼ temperature.
4. Discussion Fermentation of bran, either from native or peeled grain, was found to be an efficient means to increase the level of folates, easily extractable phenolic compounds, ferulic acid and solubilized pentosans of rye, if appropriate fermentation conditions were applied. However, the levels of these compounds and soluble pentosans were different in the two types of bran. Peeling prior to bran separation reduced the microbial load in rye bran, which was also reflected in the microbial community during yeast fermentation. The presence of the outermost layers (percar/testa) of grain in the bran was thus an important determinant of many fermentation-induced changes of the raw material. The microbial load on the grain surface was largely diminished when 11% of the grain was removed prior to bran separation. Earlier studies have also shown that the highest microbial counts are detected in the outer layers of the kernel, and significant quantities of microbial contaminants and their toxic metabolites can be removed by peeling of the grains (Bennett and Richard, 1996; Berghofer et al., 2003). However, the milling plant and equipment may also be a potential source for additional microbial populations of the bran. Particularly spore formers may reside in the equipment (Berghofer et al., 2003).
The current study showed that only low numbers of Bacillus species were found in both types of rye brans, and due to the intensive growth of LAB their number did not increase during fermentation. This was beneficial, because heat-resistant spores of Bacillus species could survive throughout the baking process and cause serious problems in the end product. Bacillus species cause spoilage of bread by rope formation and may also constitute a health risk (Thompson et al., 1993; Rosenquist and Hansen, 1995). Rope spoilage of wholemeal bread has been reported to occur more frequently than the spoilage of white wheat bread, possibly due the spore-forming bacteria present in bran fractions (Eyles et al., 1989). Therefore, it is extremely important to control the growth of Bacillus spp. when bran ingredients are utilized in bakery products. Development of the indigenous LAB community was strongly dependent of the fermentation conditions. Only low numbers of LAB were detected in the brans, but their number increased tremendously during fermentation at elevated temperatures, especially in fermentation of bran from native grains which resulted in intensive acidification. Due to growth of indigenous LAB, the yeast fermentation will gradually become mixed (sourdough) fermentation, depending on fermentation time and temperature. Milder acidity formation associated with bran from peeled grains was due to a more uniform LAB community, and lower
ARTICLE IN PRESS 184
K. Katina et al. / Food Microbiology 24 (2007) 175–186
endogenous and microbial enzyme activity. Concentration of proteolytic and xylanolytic enzymes at outer grain layers has been shown in rye and wheat (Brijs et al., 1999; Gys et al., 2004). Milder acidity provides a technological advantage for bran from peeled grain, because strong acidic or pungent cereal food flavour does not appeal to most consumers, and thus limits the use of fermented rye bran as a food ingredient. Furthermore, the limited microbial community of peeled bran allows easier control of the fermentation process with added starter cultures. Bran is a rich source of folates. Liukkonen et al. (2003) found 10 fold differences in the folate contents of the endosperm and bran fractions of rye. In agreement with the previous investigations of wholegrain rye (Liukkonen et al., 2003; Kariluoto et al., 2004), the folate content in the current study increased by over 100% during the fermentation of both brans studied. This was mainly due the folate production by yeast, as suggested earlier (Liukkonen et al., 2003; Kariluoto et al., 2005; Ja¨gerstad et al., 2005). However, yeast growth could not totally explain the increase in folates in bran ferementations. The highest folate level was obtained when growth of indigenous LAB was pronounced (fermentation for 20 h at 35 1C), which may indicate a supportive role of indigenous LAB in folate synthesis, as also suggested by Kariluoto et al. (2005). It is well known that some LAB are able to synthezize folates (Sybesma et al., 2003; Kariluoto et al., 2005) and that others can consume folates (Kariluoto et al., 2005). Thus, the composition of indigenous microbial community or added LAB may have a significant impact on the folate levels of fermented bran. Phenolic compounds, especially phenolic acids, are partly responsible for insoluble cell wall structures of cereal kernels by forming cross-links between polysaccharides and lignin (Faulds and Williamson, 1999). During fermentation of both types of brans, the levels of free phenolics analysed by Folin-C method increased, indicating increased liberation of bound phenolic compounds from the polymeric rye bran structure. Since the Folin-C method measures all phenolics, including aromatic amino acids, it is possible that increase in level of free phenolics also partly reflected increased proteolysis during the fermentation. Fermentation also increased the level of free ferulic acid, but the effect was highly dependent on the fermentation conditions and type of bran. Among the studied bioactive compounds, ARs were an exception because their concentration slightly increased in fermentation of native bran but decreased in that of peeled bran. In our previous study on wholemeal rye flour (Liukkonen et al., 2003), fermentation with added LAB and yeast for 22 h at 30 1C had practically no effect on level of alkylresorcinols, which is consistent with the current study. Eventhough fermentation increased the levels of free phenolic compounds, the levels were always lower in bran from peeled than from native bran. This was due to both a higher initial level of phenolic compounds in the native bran, and propably also to higher endogenous- and
microbial-originated enzyme activity. An increase in falling number indicated reduced alfa-amylase activity in peeled bran, and also activity of other enzymes such proteases and xylanases can be assumed to be lower in comparison to native bran (Brijs et al., 1999; Gys et al., 2004). Cell wall degradation has been reported to happen during wheat bran fermentation by either endogenous or added enzymes (Katina et al., 2006). Also acid hydrolysis of dietary fibre components has been suggested to occur (Boskov-Hansen et al., 2002). According to Gys et al. (2004), endogenous xylanases degrade wheat arabinoxylans and debranning of wholegrain wheat reduces endogenous xylanase activity by 80%. In addition, Dornez et al. (2006) reported that over 90% of endoxylanase activity of grain originates from microbes located in the outer layers of grain. Thus, in fermentation of native bran, more extensive degradation of cell walls could be expected to liberate phenolic compounds from the rye bran structure more effectively due to higher enzyme activity. The highest level of free phenolics (analysed by the Folin method) was in both fermentations obtained from the most acidic fermentations. However, the highest level of free ferulic acid was obtained in fermentation conditions providing a pH value of 6–6.5 (20 h at 20 1C or 6 h at 35 1C), which is very near to the reported pH-optimum (pH 7) of cinnamoyl esterases of wholegrain rye flour (Boskov-Hansen et al., 2002). Accordingly, the level of free ferulic acid was lowest in fermentation conditions providing strong acidity (3.9–4.1), which inhibits the cinnamyl esterase. Thus, the level of free ferulic acids in bran fermentations appears to dependent on the pH mediated activation of endogenous enzymes of rye. The content of ARs was lower in the peeled sample compared to the native sample, which is in accordance with previous results which have shown that ARs are exclusively present in the bran fraction (Ross et al., 2004). Alkylresorcinols have previously been shown to be very stable during food processing and in this study also during fermentation. On the other hand, they become associated with the matrix, probably due to complex formation with amylose, which necessiate a hot propanol:water mixture for their extraction from processed samples. The observed solubilization of pentosans in rye bran fermentation is in accordance with previous results obtained with wholegrain rye flour (Boskov-Hansen et al., 2001). Solubilization of pentosans has a significant influence on the technological properties of wholegrain rye flour; especially in rye baking where the solubilization of pentosans is a basis for good texture of wholegrain rye breads. In addition, a positive influence of solubilization of pentosans has been reported in wheat flour baking (Maat et al., 1992; Courtin and Delcour, 2002) and also in highfibre baking using wheat bran (Katina et al., 2006). It is noteworthy that maximal solubilization of pentosans was obtained with fermented bran made of peeled grains independent of fermentation conditions. The different xylanase activity of brans from native and peeled grain is
ARTICLE IN PRESS K. Katina et al. / Food Microbiology 24 (2007) 175–186
suggested to cause the observed differences in solubilization of pentosans. Native bran most probably contains higher levels of either microbial or endogenous xylanases, which would degradade solubilized arabinoxylans (Gys et al., 2004; Dornez et al., 2006). 5. Conclusion This study established the potential of fermentation to enhance the bioactivity and technological potential of rye bran. The presence of indigenous LAB and enzymes concentrated to the outer layers of the grains contributes significantly to the changes induced in bran during fermentation. The extent of these changes can be modulated by changing the milling process during separation of the bran prior to fermentation. Acknowledgments This study is financially supported by The Nordic Joint Committee for Agricultural Research (NJK), project Rye Bran for Health (121) and by the European Commission in the Communities Sixth Framework Programme, Project HEALTHGRAIN (FP6-514008). This publication reflects only author’s views and the Community is not liable for any use that may be made of the information contained in this publication. We are grateful to the technical staff of VTT for their skilful technical assistance. References Bartolome´, B., Go´mez-Cordove´s, C., 1999. Barley spent grain: release of hydroxycinnamic acids (ferulic and p-coumaric acids) by commercial enzyme preparations. J. Sci. Food Agric. 79, 435–439. Bennett, G., Richard, J., 1996. Influence of processing on Fusarium mycotoxins in contaminated grains. Food Technol. 50 (5), 235–238. Berghofer, L., Hocking, A., Miskelly, D., Jansson, E., 2003. Microbiology of wheat and flour milling in Australia. Int. J. Food Microbiol. 85, 137–149. Boskov-Hansen, H., Andersen, M.F., Nielsen, L.M., Back-Knudsen, K-E., Meyer, A.S., Christensen, L.P., Hansen, A˚., 2002. Chances in dietary fibre, phenolic acids and activity of endogenous enzymes during rye bread making. Eur. Food Res. Technol. 214, 33–42. Brijs, K., Bleukx, W., Delcour, J., 1999. Proteolytic activities in dorminant rye (Secale cereale L.) grain. J. Agric. Food Chem. 47, 3572–3578. Bruce, J., 1996. Automated system rapidly identifies and characterizes micro-organisms in food. Food Technol. 50, 77–81. Courtin, C., Delcour, J., 2002. Arabinoxylans and endoxylanases in wheat flour bread-making. J. Cereal Sci. 35, 225–243. Devesa, A., Martı´ nez-Anaya, M., 2001. Characterization of waterextractable pentosans in enzyme-supplemented wheat sourdough process. Food Sci. Technol. Int. 72 (2), 145–153. Dornez, E., Joye, I., Gebruers, K., Delcour, J., Courtin, C., 2006. Wheatkernel-associated endoxylanases consist of a majority of microbial and a minority of wheat endogenous endoxylanases. J. Agric. Food Chem. 54 (11), 4028–4034. Eyles, M., Moss, R., Hocking, A., 1989. The microbial status of Australian flour and the effects of milling procedure on the microflora of wheat and flour. Food Australia 41, 704–708. Faulds, M., Williamson, G., 1999. The role of hydroxycinnamates in the plant cell wall. J. Sci. Food Agric. 79, 393–395.
185
Fro¨lich, W., Crakenberg, T., Asp, N.G., 1986. Enzymatic degradation of phytate (myo-inositol hexaphosphate) in whole grain flour suspensions and dough. A comparion between 31p NMR spectroscopy and ferric ion method. J. Cereal Sci. 4, 325–334. Gys, W., Courtin, C.M., Delcour, J.A., 2004. Reduction of xylanase activity in flour by debranning retards syruping in refrigerated doughs. J. Cereal Sci. 39 (3), 371–377. Heinio, R.L., Liukkonen, K.-H., Katina, K., Myllymaki, O., Poutanen, K., 2003. Milling fractionation of rye produces different sensory profiles of both flour and bread. Lebens. Wiss. Technol. 36 (6), 577–583. Jacobs Jr., D.R., Gallaher, D.D., 2004. Whole grain intake and cardiovascular disease: a review. Curr. Atherosclerosis Reports 6 (6), 415–423. Ja¨gerstad, M., Piironen, V., Walker, C., Ros, G., Carnovale, E., Holasova, M., Nau, H., 2005. Increasing natural folates through bioprocessing and biotechnology. Trends Food Sci. Technol. 16, 298–306. Kariluoto, S., Vahteristo, L., Salovaara, H., Katina, K., Liukkonen, KH., Piironen, V., 2004. Effect of baking method and fermentation on folate content of rye and wheat breads. Cereal Chem. 81, 134–139. Kariluoto, S., Aittamaa, M., Korhola, M., Salovaara, H., Vahteristo, L., Piironen, V., 2005. Effects of yeasts and bacteria on the levels of folates in rye sourdoughs. Int. J. Food Microbiol. 106 (2), 137–143. Katina, K., 2005. Sourdough: a tool for the improved flavour, texture and shelf-life of wheat bread. Doctoral Thesis, VTT Publications no. 569. VTT Biotechnology, Espoo, 92 pp.+app. 81pp. Katina, K., Salmenkallio-Marttila, M., Partanen, R., Forssell, P., Autio, K., 2006. Effects of sourdough and enzymes on staling of high-fibre wheat bread. Lebensm. Wiss. Technol. 39, 479–491. Lang, R., Jebb, S., 2003. Who consumes whole grains, and how much? Proc. Nutr. Soc. 62 (1), 123–127. Larsson, M., Sandberg, A.S., 1991. Phytate reduction in bread containing oat flour, oat bran or rye bran. J. Cereal Sci. 14, 141–149. Larsson, S.C., Giovannucci, E., Bergkvist, L., Wolk, A., 2005. Whole grain consumption and risk of colorectal cancer: a population-based cohort of 60,000 women. Br. J. Cancer 92, 1803–1807. Liukkonen, K.-H., Katina, K., Wilhelmsson, A., Myllymaki, O., Lampi, A.-M., Kariluoto, S., Piironen, V., Heinonen, S.M., Nurmi, T., Adlercreutz, H., Peltoketo, A., Pihlava, J.M., Hietaniemi, V., Poutanen, K., 2003. Process-induced changes on bioactive compounds in whole grain rye. Proc. Nutr. Soc. 62, 117–122. Maat, J., Verbakel, J., Stam, H., Santos de Silva, M., Bosse, M., Egmond, M., Hagemanns, M., Gorcom, R., Hessing, J., Hondel, C., Rotterdam, C., 1992. Xylanases and their application in bakery. In: Visser, J., Beldman, G., Kusters van Someren, M., Vorangen, A. (Eds.), Xylans and Xylanases. Elsevier Science Publishers, Amsterdam, The Netherlands, pp. 349–360. Murtaugh, M.A., Jacobs Jr., D.R., Steffen, L.M., Jacob, B., Marquart, L., 2003. Epidemiological support for the protection of whole grains against diabetes. Proc. Nutr. Soc. 62, 143–149. Piironen, V., Varo, P., Koivistoinen, P., 1987. Stability of tocopherols and tocotrienols in food preparation procedures. J. Food Compos. Anal. 1, 53–58. Rosenquist, H., Hansen, A˚., 1995. Contamination profiles and characterisation of Bacillus species in wheat bread and raw materials for bread production. Int. J. Food Microbiol. 26, 353–363. Ross, A.B., Kamal-Eldin, A., Jung, C., Shepherd, M.J., A˚man, P., 2001. Gas chromatographic analysis of alkylresorcinols in rye (Secale cereale L.) grains. J. Sci. Food Agric. 81, 1405–1411. Ross, A.B., Shepherd, M.J., Schu¨pphaus, M., Sinclair, V., Alfaro, B., Kamal-Eldin, A., A˚man, P., 2003. Alkylresorcinols in cereals and cereal products. J. Agric. Food Chem. 51, 4111–4118. Ross, A., Kamal-Eldin, A., A˚man, P., 2004. Dietary alkylresorcinols: absorption, bioactivities and possible use as biomarkers of whole-grain wheat- and rye-rich foods. Nutr. Rev. 62, 81–95. Rouau, X., Surget, A., 1994. A rapid semi-automated method for the determination of total and water e-extractable pentosans in wheat flours. Carbohydr. Polym. 24, 123–132.
ARTICLE IN PRESS 186
K. Katina et al. / Food Microbiology 24 (2007) 175–186
Salmenkallio-Marttila, M., Katina, K., Autio, K., 2001. Effects of bran fermentation on quality and microstructure of high-fiber wheat bread. Cereal Chem 78, 429–435. Salovaara, H., 2004. Lactic acid bacteria in cereal-based products. In: Salminen, S., von Wright, A., Ouwehand, A. (Eds.), Lactic Acid Bacteria—Microbiological and Functional Aspects. Marcel Dekker, New York, pp. 431–4513. Singleton, V.L., Orthofer, R., Lamuela-Raventos, R.M., 1999. Analysis of total phenols and other oxidation substrates and antioxidants by means of Folin–Ciocalteu reagent. Methods Enzymol. 299, 152–178.
Sybesma, W., Starrenburg, M., Tijsseling, L., Hoefnagel, M., Hugenholtz, J., 2003. Effects of cultivation conditions on folate production by lactic acid bacteria. Appl. Environ. Microbiol. 69, 4542–4548. Thompson, J., Dodd, C., Waites, W., 1993. Spoilage of bread by Bacillus. Int. Biodeter. Biodegr. 32, 55–66. Verdeal, K., Lorenz, K., 1977. Alkylresorcinols in wheat, rye, and triticale. Cereal Chem. 54, 475–483. Wilmotte, A., Van der Auwera, G., De Wachter, R., 1993. Structure of the 16S ribosomal RNA of the thermophilic cyanobacterium Chlorogloeopsis HTF (‘Mastigocladus laminosus HTF’) strain PCC7518, and phylogenetic analysis. FEBS Lett. 317, 96–100.
Bran fermentation as a means to enhance technological properties and bioactivity of rye K. Katinaa, A. Laitilaa, R. Juvonena, K.-H. Liukkonena, S. Kariluotoc, V. Piironenc, R. Landbergd, P. A˚mand, K. Poutanena,b, a VTT Biotechnology, VTT Technical Research Centre of Finland, Tietotie 2, P.O. Box 1000, 02044 VTT, Espoo, Finland University of Kuopio, Food and Health Research Centre, Department of Public Health and Clinical Nutrition, P.O. Box 1627, FI-70211 Kuopio, Finland c Department of Applied Chemistry and Microbiology, University of Helsinki, P.O. Box 27, FIN-00014, Finland d Department of Food Science, SLU, P.O. Box 7051, Uppsala, Sverige
b
Available online 12 September 2006
Abstract Response surface methodology was applied to study the effects of fermentation on the levels of phytochemicals (folates, phenolic compounds, alkylresorcinols) and on the solubilization of pentosans in rye bran from native and peeled grains. Furthermore, the microbial composition of the brans before and after fermentation was studied. Peeling reduced the microbial load and lower microbial counts were detected in the fermentation experiments carried out with the bran from peeled grains. High temperature and long fermentation time favoured the growth of indigenous lactic acid bacteria (LAB), and a diverse microbial community was detected. The brans contained low levels of aerobic spore-forming bacteria, but their number was not increased during the fermentations. Fermentation of both brans increased the levels of folates, easily extractable total phenolics and free ferulic acid. During fermentation of bran from native grains, the levels of alkylresorcinols slightly increased but during fermentation of bran from peeled grains they decreased. Significant increase in soluble pentosans was established in both types of rye bran fermentations. Enhanced bioactivity and solubilization of pentosans with limited microbial growth were obtained after 12–14 h fermentation at 25 1C. The results suggest that fermentation is a potential bioprocessing technology for improved technological properties and bioactivity of rye bran. r 2006 Published by Elsevier Ltd. Keywords: Rye bioactivity; Phytochemicals; Sourdough fermentation; Pentosans
1. Introduction Consumption of foods rich in whole grains and cereal fibre has in epidemiological studies been shown to reduce the risk of chronic diseases such as diabetes, cardiovascular disease and certain cancers, as reviewed by Murtaugh et al. (2003) and Jacobs and Gallaher (2004), and shown by Larsson et al. (2005). However, there is a large gap between the dietary recommendations and current intake of whole grains and dietary fibre (Lang and Jebb, 2003). The need and challenge for cereal technologists is to develop
Corresponding author. VTT Biotechnology, VTT Technical Research Centre of Finland, Tietotie 2, P.O. Box 1000, 02044 VTT, Espoo, Finland. Fax: +358 9 4552103. E-mail address: kaisa.poutanen@vtt.fi (K. Poutanen).
0740-0020/$ - see front matter r 2006 Published by Elsevier Ltd. doi:10.1016/j.fm.2006.07.012
technologies enabling to lower consumer barriers to frequent use of high-fibre and wholegrain foods. Cereal brans are important ingredients providing dietary fibre, and wheat bran is one of the most common raw materials for increasing the level of insoluble dietary fibre in baking. In the Nordic countries, rye is the most important source of dietary fibre. Dietary fibre and bioactive compounds such as alkyresorcinols, lignans, phenolic acids, phytosterols, tocopherols and—tocotrienols and folates are concentrated in the bran fraction of rye (Liukkonen et al., 2003; Ross et al., 2004). This is why it would be interesting to develop nutritionally optimized cereal foods and new ingredients from rye bran. The main part of dietary fibre in bran is insoluble, which influences the digestibility and bioavailability of nutrients and phytochemicals. The outer layers of grain contain cellulose and lignin which influence both the taste and mouthfeel of
ARTICLE IN PRESS 176
K. Katina et al. / Food Microbiology 24 (2007) 175–186
the bran. The properties of rye bran thus restrict its full exploitation in different consumer foods. Fermentation with well-characterized starter cultures— yeast or lactic acid bacteria (LAB)—is a potential means to improve the palatability and processability of brans and wholemeal flours (Salovaara, 2004; Katina et al., 2006). Furthermore, bran fermentation could assist in microflora management and improve the microbiological safety of bran. This is important, because brans are known to contain more micro-organisms than endosperm flours, and they could also be a source of spoilage bacteria and fungi (Rosenquist and Hansen, 1995). It has been pointed out that fermentation is a combined action of added starters cultures and the micro-organisms naturally present in grains (Salovaara, 2004). Both endogenous and microbial originated enzymes are shown to be concentrated to the outer grain layers (Dornez et al., 2006; Gys et al., 2004), and have thus the possibility for action during the bran fermentation. Both acidification and activity of enzymes, especially xylan-degrading enzymes, were shown to contribute to the solubilization of arabinoxylans during the baking process of wholemeal rye (Boskov-Hansen et al., 2002). This led to decreased level of dietary fibre in the resulting bread. Solubilization of pentosans of flour, and especially transformation of water unextractable arabinoxylan to water extractable arabinoxylan, has been reported to improve bread volume and texture in wheat baking (Courtin and Delcour, 2002) and in high-fibre wheat baking (Katina et al., 2006). The influence of processing conditions or starter culture, however, on solublization of pentosans during wheat sourdough or bran fermentation is not known. In wheat sourdough baking, the influence of commercial pentosanases on the molecular weight of arabinoxylans was shown to be dependent on the starter type (Devesa and Martı´ nezAnaya, 2001). Many biochemical changes affecting nutritional quality as well as texture and flavour of wheat flour occur during sourdough baking, as reviewed by Katina (2005). Levels of folate and easily extractable phenolic compounds have been shown to increase (Liukkonen et al., 2003; Kariluoto et al., 2004), whereas levels of phytate (Fro¨lich et al., 1986; Larsson and Sandberg, 1991), alkylresorcinols (ARs) (Verdeal and Lorenz, 1977) and tocopherols (Piironen et al., 1987) have been reported to reduce in sourdough baking process. We have previously shown that enzymatic pretreatments and fermentation of wheat bran are efficient in improving texture, the sensory quality and stability of breads supplemented with 20% of wheat bran (SalmenkallioMarttila et al., 2001; Katina et al., 2006). We have also shown that by careful fractionation of rye bran in the milling process, the sensory quality can be tailored while still retaining many of the bioactive bran compounds (Heinio et al., 2003; Liukkonen et al., 2003). The aim of the current study was to develop fermentation as a means to improve the properties of rye bran. We established the
influence of fermentation conditions and type of bran on the microbial community and levels of bioactive compounds and solubilized pentosans in rye. 2. Material and methods 2.1. Experimental design and modelling Fermentation time (Ti, h) and temperature (Te, 1C) were selected as independent variables to study the effects of the fermentation conditions on properties of bran properties. A central composite design was used to plan experiments and three replicates were made at the centre point of design to allow estimation of the pure error at sum of the square. The experimental design and obtained values are presented in Table 1a and b. The results were analysed by a multipleregression method (PLS), which describes the effects of variables in second-order polynomial models. For each response (microbes, bioactive compounds and pentosans) a quadratic model was used. Regression analysis was made and the response surfaces were plotted with the Modde 4.0 and 6.0 (Umetrics AB, Umea˚, Sweden). The fit of the model to the experimental data was given by the coefficient of determination, R2, which explains the extent of the variance in a modelled variable that can be explained with the model. Each model was validated by calculating the predictive power of model, Q2, which is a measure of how well the model will predict the responses for a new experimental condition. Q2 is based on prediction of residual sum of squares (PRESS). For determining Q2, the computations were repeated several times, each time omitting different objects from the calculation of the model. PRESS was then computed as the squared difference between observed Y and predicted values (cross-validation of R2). Large values of Q2 indicate that the model had good predictive ability and will have small prediction errors. Q2 should be 40.5 if conclusions are to be drawn from the model. Only the models with high enough R2- and Q2- values were used to describe influence of fermentation time and temperature on the amount of microbes, levels of bioactive compounds and the amount of pentosans. Adequate models (R2 40:7 and Q2 40:5) could be obtained with total amount of LAB, folates, free phenolics (Folin-C), free ferulic acids and, alkylresorcinols and with soluble pentosans. With bound phenolics (Folin-C), bound ferulic acid or with total amount of pentosans, adequate models could not been obtained due to low systematic variation in measured values in the experimental region. 2.2. Raw materials Experiments were carried out with the Finnish Rye Cultivar, Amilo from 2003. Part of the rye grains was first peeled with a Buhler horizontal grinding/peeling machine to remove about 11% of the outer layer of the grain. Both native and peeled rye grains were milled
ARTICLE IN PRESS K. Katina et al. / Food Microbiology 24 (2007) 175–186
177
Table 1 Experiment number
Time/h
Temperature/ 1C
pH
TTA
Folates (mg/ 100 g)
FAa (mg/ 100 g)
PHC (MeOH)b (mg/100 g)
Arsc (mg/ 100 g)
SP (%)d
(A) Experimental design and obtained values for responses in fermentation 1 6 20 6.7 3.9 2 20 20 6.5 5.4 3 6 35 6.6 5.4 4 20 35 4.3 16.2 5 6 27.5 6.7 4.9 6 20 27.5 5.0 8.5 7 13 20 5.7 5.6 8 13 35 5.0 7.5 9 13 27.5 6.4 5.9 10 13 27.5 6.1 6.3 11 13 27.5 6.5 5.5
of native bran 147 19 188 30 136 28 240 7 182 24 273 10 157 28 202 8 190 30 221 20 221 20
205 263 257 383 241 305 251 313 259 260 260
290 284 311 282 307 300 311 323 306 306 —
1.6 1.8 1.9 1.1 1.8 1.4 1.7 1.3 1.8 1.7 1.8
(B) Experimental design and obtained values for responses in fermentation 1 6 20 6.5 3.8 2 20 20 6.4 4.0 3 6 35 6.5 3.8 4 20 35 4.1 12.4 5 6 27.5 6.4 3.6 6 20 27.5 5.9 5.4 7 13 20 6.4 3.7 8 13 35 4.6 8.2 9 13 27.5 6.3 4.2 10 13 27.5 6.3 4.2 11 13 27.5 6.3 4.2
of bran from peeled grains 147 19 188 30 136 28 240 7 182 24 273 10 157 28 202 8 190 30 221 20 221 20
205 263 257 383 241 305 251 313 259 260 260
191 204 195 160 198 156 162 161 202 161 —
1.9 2.5 2.6 2.6 2.6 2.9 2.7 3.1 2.7 2.8 2.75
C17:0
C19:0
C21:0
C23:0
C25:0
Total (mg/g DMe)
C17:0/C21:0
(C) Total alkylresorcinol content (mg/g DM) and relative homologue composition in fermentation of (A) bran from native grain and (B) bran from peeled grain (A) Bran from native grain Experiment number 1 24 32 24 10 9 2903 2 25 33 24 10 8 2835 1.0 3 24 33 24 10 9 3114 1.0 4 24 33 25 10 8 28194 1.0 5 24 32 24 11 9 3066 1.0 6 24 33 24 10 8 2966 1.0 7 24 32 24 11 9 3112 1.0 8 25 33 24 10 8 3230 1.0 9 25 33 24 10 8 3060 1.0 10 24 32 24 11 9 3063 1.0 (B) Bran from peeled grain 1 26 2 26 3 25 4 25 5 25 6 26 7 26 8 26 9 26 10 26 a
35 35 34 35 34 35 35 36 35 35
23 24 24 24 24 24 23 23 23 23
FA ¼ ferulic acid. PHC ¼ phenolic compounds after MeOH extraction. c ARs ¼ alkyresorcinols. d SP ¼ soluble pentosans. e DM ¼ dry matter. b
9 9 10 9 10 9 9 9 9 9
7 6 8 7 8 7 7 6 7 7
1910 2041 1951 1602 1982 1566 1623 1618 2026 1610
1.1 1.1 1.0 1.1 1.0 1.1 1.1 1.1 1.1 1.1
ARTICLE IN PRESS 178
K. Katina et al. / Food Microbiology 24 (2007) 175–186
at a moisture level of 15–16% with a Buhler roller mill to have bran fractions. The bran fractions comprised 15% and 20% of the original native and peeled grains, respectively. Peeling increased falling number of the bran from 75 to 125.
2.3. Preparation of bran fermentations Fresh Baker’s yeast was obtained from Suomen Hiiva, Finland. To obtain a dough yield of 450, 600 g of rye bran (native or peeled) was mixed carefully with 2100 g of tap water, and with 7.5 g of fresh Baker’s yeast (107 cfu/g) (Suomen Hiiva., Finland) in a large beaker (2000 ml), which was covered with aluminium foil. Fermentations were carried out according to the experimental design. Fresh samples were taken from unfermented and fermented bran for microbiological analysis. In addition, samples were frozen for later measurements of pH, and total titratable acidity (TTA). Fermented bran samples were also freeze-dried for analysis of bioactive compounds. Fermentations were done in duplicate and repeated twice.
2.4. Microbiological analyses Samples for the microbiological analyses were taken from native and peeled bran samples and after fermentation according to the experimental design. The following microbial groups were analysed from homogenized bran samples: aerobic heterotrophic bacteria, LAB and fungi. A sample of 10 g was homogenized for 10 min with 90 ml of sterile saline in a Stomacher Lab Blender 400 (Seward Medical, London, UK). Aerobic heterotrophic bacteria were determined on plate count agar (PCA, Difco Laboratories, Detroit, USA), and samples were incubated in aerobic conditions at 30 1C for 2–3 days. The number of LAB was determined on MRS agar (Oxoid) and samples were incubated in anaerobic conditions at 30 1C for 5 days. To prevent fungal overgrowth of bacterial determinations, 0.001% cycloheximide (Sigma Chemical, St. Louis, Missouri, USA) was added to PCA, and MRS media. Yeast and moulds were determined on YM agar (Difco Laboratories). Samples were incubated in aerobic conditions at 25 1C for 3–5 days. Chlortetracycline and chloramphenicol (both at 0.01%) were added to YM medium to prevent bacterial growth. In addition, 0.02% of Triton-X 100 (BDH) was used to limit the spreading of fungal colonies on YM-agar. To determine spore formers, 5 ml of homogenized sample was heated in a water bath of 80 1C for 10 min for inactivation of vegetative cells. Bacillus spp. were enumerated on tryptone soy agar (TSA, Oxoid) and plates were incubated at 30 1C for 3 days. The bacteria and yeast results are expressed as colony-forming units/ gram fermented bran (cfu/g).
2.5. Characterization and identification of LAB and aerobic spore-forming bacteria To identify dominating different colony types were selected from MRS and TSA plates (a maximum of five colonies per sample). The isolates were characterized by Gram staining and catalase test and subsequently purified by successive sub-culturing on the antibiotic-free isolation media. Ribotyping was performed using the standard method of the automated ribotyping device RiboPrinters Microbial Characterization System (DuPont QualiconTM, USA) and EcoRI restriction enzyme (Bruce, 1996). For other genetic analyses, genomic DNA was isolated from 1–2-day-old plate cultures by suspending 2 ml loop-full of cell mass in 0.5 ml PCR-grade water followed by bead-beating with 200-mm glass beads (0.1 g ml1, Sigma) in the RibolyserTM instrument (1.5 min 6.0 m/s, Hybaid, UK) and removal of cell debris by centrifugation (11,000g, 3 min). RAPD-PCR analysis was performed with OPA-02 (50 -TGCCGAGCTG-30 ) and OPA-03 (50 -AGTCAGCCAC-30 ) primers in 50 ml reactions containing 0.4 mM primer (SigmaGenosys, Cambridge, UK), 0.2 mM dNTP (Finnzymes, Espoo, Finland), 3 U DyNAzyme II DNA polymerase (Finnzymes) and 5 ml template DNA in a 1 enzyme buffer (final concentration of 10 mM Tris–HCl, 1.5 mM MgCl2, 50 mM KCl, 0.1% Triton X-100, pH 8.8). The PCR programme in a thermocycler (UNO II, Biometra GmbH, Goettingen, Germany) consisted of an initial denaturation for 5 min at 95 1C, followed by 35 cycles (1 min at 95 1C, 1 min at 32 1C, 2 min at 72 1C), a final extension for 10 min at 72 1C and cooling to 4 1C. To identify the LAB isolates representing different fingerprint types, the 16S rRNA gene was amplified using the conserved primers BSF 8/20 (50 -AGAGTTTGATCCTGGCTCAG-30 ) and BSR1541/ 20 (50 -AAGGAGGTGATCCAGCC GCA-30 ) (Wilmotte et al., 1993) in 50 ml reactions containing 0.5 mM primers, 0.2 mM dNTP (Finnzymes), 2.5 U DyNAzyme II DNA polymerase and 5 ml template DNA in the 1 enzyme buffer. The PCR programme consisted of an initial denaturation for 5 min at 94 1C, followed by 35 cycles (45 s at 95 1C, 45 min at 56 1C, 2 min at 72 1C), a final extension for 7 min at 72 1C and cooling to 4 1C. Amplification products were purified using a QiaQuick PCR purification kit (Qiagen, Mississaugua, Ontario, Canada) according to the instructions. DNA purity and yield were estimated electrophoretically in 1% agarose gels. Sequencing reactions of the PCR products were performed with an ABI PRISM BigDye terminator cycle sequencing kit v.3.1 (Applied Biosystems, Foster City, CA) according to the instructions using the primer BSF 8/20. The products were electrophoresed with an ABI PRISM 3100 sequencer (Applied Biosystems) and checked and edited with the Chromas programme (Technelysium, Helensvale, Australia) and compared to the GenBank database (www.ncbi.nlm.nih.gov) using the BLAST (basic local alignment search tool) algorithm.
ARTICLE IN PRESS K. Katina et al. / Food Microbiology 24 (2007) 175–186
2.6. pH and TTA Frozen bran samples were thawed overnight in a refrigerator. The pH value was measured from an aliquot of 10 g of fermented bran blended with 100 ml distiled water in a TitroLine Alpha471217 (Schott, Germany). For the determination of TTA, this suspension was titrated against 0.1 M NaOH to a final pH value of 8.5 using TitroLine Alpha. TTA was expressed as the amount of NaOH used (ml). All samples were analysed in duplicate. 2.7. Analysis of bioactive compounds and pentosans Folates were analysed by a microbiological assay method including extraction and trienzyme treatment as described by Kariluoto et al. (2004). ARs were extracted in a hot 1-propanol:water mixture according to Ross et al. (2003). Extracts were then concentrated and analysed by GC, without further clean up or derivatization, according to Ross et al. (2001). Total phenolic acids were analysed after alkaline and free phenolic acids after ethylacetate extraction by HPLC (Bartolome´ and Go´mez-Cordove´s, 1999). Bound phenolic acids were calculated by subtracting free phenolic acids from total phenolic acids. Total phenolics were analysed by Folin–Ciocalteau method: methanolic extracts of freeze-dried and ground rye samples were obtained by an ultrasonication-assisted extraction procedure. The residue was further extracted at alkaline conditions. The content of total phenolic compounds (gallic acid equivalents) in methanolic (free phenolics) and alkaline (bound phenolics) extracts was determined using Folin–Ciocalteau procedure (Singleton et al., 1999). Total amount of pentosans and soluble pentosans were analysed by colorimetric method (Rouau and Surget, 1994). All analyses were done in duplicate and the obtained average values of bioactive compounds are expressed on dry weight basis. 3. Results 3.1. Characterization of native and peeled brans The chemical composition and the microbial counts in rye bran from native and peeled grains are shown in Table 2. Removal of outer grain layers decreased the relative amounts of protein, insoluble dietary fibre, phenolic compounds and folates in peeled bran fraction, and increased the relative amount of starch. Slightly lower counts of aerobic bacteria, LAB, yeast and moulds were detected in bran from peeled grain as compared to the native bran. 3.2. Microbes in bran fermentations Commercial baker’s yeast was added at level of 2–3 107 cfu/g, and it dominated the bran fermentations.
179
Table 2 Chemical composition and number of bacteria and fungi (cfu/g) in native and peeled rye bran Rye bran
Macrocomponents (g/100 g dw) Protein Starch Dietary fibre Of which soluble Pentosans Of which soluble
Native
Peeled
18.6 30.2 35.8 5.3 15.9 0.91
14.8 44.4 27.8 5.3 14.4 0.97
Bioactive compounds Total phenolics (Folin-C) (mg/100 g dw) Free (MeOH extractable) Bound (NaOH extractable) Phenolic acids (ferulic acid) (mg/100 g dw) free (water extractable) bound (NaOH extracable) Alkylresorcinols (mg/100 g dw) Lignans (mg/100 g dw) Folates (mg/100 g dw)
584 200 383 274 6.8 267 285 5.8 0
415 147 268 215 2.3 213 222 4.9 81
Microbial groups (cfu/g) Total aerobic heterotrophic bacteria Aerobic sporeforming bacteria Lactic acid bacteria Yeasts Moulds
5 106 6 102 1 103 3 103 9 102
3 105 3 102 4 102 7 102 5 101
No other fungi (yeasts or moulds) were detected during fermentation. Only low levels of LAB were found in native and peeled bran fractions (Table 2), but they increased markedly during fermentation (Fig. 1). The growth of indigenous LAB community was greatly dependent on the fermentation time and temperature. Multiplication of LAB was pronounced when the fermentation temperature was raised from 20 to 35 1C, and the highest viable counts, 2 109 cfu/ g, were determined in bran from native grains after 20 h fermentation at 35 1C. Intensive LAB growth was reflected in the acidification patterns in bran fermentation from both native and peeled grains, as seen in Fig. 2a and b. Development of acidity (measured as pH and TTA) was clearly dependent on the fermentation temperature and time; strong acidity development occurred only if temperature was 425 1C and time 412 h in both type of fermentations. It is noteworthy that more restricted production of acids occurred in fermentation of bran from peeled grains, as compared to fermentation of the native bran. The highest TTA value was 12 ml NaOH 0.1 ml/10 g of dough for fermentation of bran from peeled grain and 16 for fermentation of native bran. However, with both bran types pH decreased from 6.6 to 4.1 after 20 h fermentation at 35 1C (Fig. 2a). As seen in Figs. 1 and 2b, the modelling of LAB growth and TTA produced different surface response plots for
ARTICLE IN PRESS 180
K. Katina et al. / Food Microbiology 24 (2007) 175–186
Fig. 1. Influence of fermentation time and temperature on growth of LAB (cfu/g) in fermented bran made from (A) native rye bran (B) peeled rye bran. Ti ¼ time and Te ¼ temperature.
Fig. 2. (a) Influence of fermentation time and temperature on pH values of fermented bran made from (A) native rye bran (B) peeled rye bran. Ti ¼ time and Te ¼ temperature. (b) Influence of fermentation time and temperature on TTA values (ml NaOH/10 g of dough) of fermented bran made from (A) native rye bran (B) peeled rye bran. Ti ¼ time and Te ¼ temperature.
fermentations of bran from native and peeled rye grains, indicating differences in the composition of LAB community. To determine the predominating LAB in the fermentation process of native bran, a total of 42 LAB were subjected to RAPD-PCR fingerprinting and the isolates representing different fingerprint types were identified by partial 16S rRNA gene sequencing. The fermentation of native bran developed a diverse LAB community which was characteristic to each temperature and differed from the major community present in the bran prior to fermentation (Table 3). In general, LAB diversity increased with fermentation temperature, and at the highest temperature eight species or genotypes were identified. The highest species diversity was found among the facultative heterofermentative lactobacilli. The obligate
heterofermentative population was mainly composed of Leuconostoc species. In addition, Lactobacillus fermentum was isolated from the native brans fermented at 35 1C. Homofermentative cocci, Enterococcus hermanniensis, Pediococcus pentosaceus and/or Lactococcus lactis, were detected during the fermentations at 27.5 1C and 35 1C. Lactobacillus sakei was detected in all fermentations independent of the temperature. As seen from Table 3 some of the isolates could not be identified to species level due to the high genetic similarity of the closely related species. In contrast to fermentations with native bran, a uniform LAB community was detected in fermentations with brans from peeled grains, in which the homofermentative Enterococcus were the predominant species (data not
ARTICLE IN PRESS K. Katina et al. / Food Microbiology 24 (2007) 175–186
181
Table 3 Lactic acid bacteria identified from various stages of fermentation of native bran1 LABa
Native bran
Yeast+bran prior to fermentation
Fermentation process (6–20 h) at 20 1C
Homo-fermentative
Obligate heterofermentative
Facultative heterofermentative
E. casseliflavus/E. gallinarum/ E. saccharolyticus (99.7%b) E. hermanniensis (99.4) Leuc. citreum (99.7%)
27.5 1C
35 1C
E. hermanniensis (99.4%)
Lc. lactis (99.8%)
P. pentosaceus (99.7%) Leuc. mesenteroides (99.7%)
Leuc. pseudomesenteroides/ Leuc. mesenteroides (X99.5%)
Leuc. mesenteroides (99.7%)
L. paraplantarum/L. plantarum/L. pentosus (X99%) L. casei/L. paracasei (99.8%)
L. curvatus/L. graminis (X99.7%)
L. curvatus/L. graminis (X99.5%)
L. plantarum/L. pentosus/ paraplantarum (X99.5%) L. sakei (100%)
L. plantarum/L. pentosus/L. paraplantarum (X98.9%) L. sakei (100%) L. sakei/L. curvatus/ L. graminis (98.7%)
L. fermentum (99.7%) L. curvatus/L. graminis (X99.3%) L. plantarum/L. pentosus/L. paraplantarum (X99.5%) L. curvatus (99.8%) L. sakei (100%) L. sakei/L. curvatus/ L. graminis (98.2%)
a
E., Enterococcus; L., Lactobacillus; Leuc., Leuconostoc; Lc., Lactococcus; P., Pediococcus. Similarity of partial 16S rRNA gene sequence to closest relative. When more than one species are presented the analysis was not conclusive at species level. b
shown). Two Enterococcus species, showing different DNA-fingerprints, were isolated from the peeled bran fermentations. Based on the partial 16S rRNA gene sequences, Enterococcus sp. 1 gave 100% similarity to Enterococcus casseliflavus and Enterococcus flavescens, whereas Enterococcus sp. 2 gave 99.4% similarity to Enterococcus saccharolyticus and Enterococcus gallinarum. These closely related species could not be separated using 16S rRNA sequencing. Intensive growth of LAB and concomitant increase of organic acids as well as other possible antimicrobial substances restricted the growth of aerobic heterotrophic bacteria (data not shown). The total viable count was 3 105 cfu/g at the beginning of fermentation with bran from peeled grains, and decreased to 6 104 cfu/g after 20 h fermentation at 35 1C. A similar trend was observed with the native bran. Although low numbers of spore-forming bacteria (Bacillus spp.) were detected in the brans, their number did not increase during the fermentation process. The maximum of 50 cfu/g was detected in fermentations with the bran from peeled grains and maximum of 200 cfu/g in fermentations with the bran of native grains. We identified three different Bacillus species from bran fermentations, namely Bacillus licheniformis (16S RNA gene similarity 100%), Bacillus sonorensis (16S RNA gene similarity 100%), and Paenibacillus amylolyticus (16S RNA gene similarity 100%).
3.3. Effect of bran fermentation on bioactive compounds As seen in Fig. 3 fermentation temperature expressed quadratic effect on the level of folates in fermentation with native bran (Fig. 3). As compared to the level of folates in unfermented bran, 1.9 fold increase was obtained by fermenting native rye bran at 28 1C for 20 h. In fermentation of rye bran from peeled grains, the level of folates increased as a function of increasing time and temperature. The level of folates increased 2.3 fold by fermenting bran from peeled grains at 35 1C for 20 h. The level of free phenolics, determined by Folin C method after methanol extraction, increased linearly with of increasing fermentation time and temperature (Fig. 4). Fermentation of native rye bran increased the amount of free phenolics by 90%, but that of bran from peeled grains only by 15–30%. The effect of fermentation conditions was also studied on the levels of single groups of phenolic compounds. The highest increase in the amount of free phenolic acids (ferulic acid) was detected in fermentation of native bran both at high temperature (35 1C) and short fermentation time (6 h) and low temperature (20 1C) and long fermentation (20 h) time (Fig. 5). At high temperature and long fermentation time, the level of free phenolic acids was very similar to that of nonfermented bran. In fermentation of bran from peeled grains, ferulic acid showed a very similar pattern, but absolute changes were at lower level.
ARTICLE IN PRESS 182
K. Katina et al. / Food Microbiology 24 (2007) 175–186
Fig. 3. Influence of fermentation time and temperature on total amount of folates (mg/100 g) in fermented bran made from (A) native rye bran (B) peeled rye bran. UF ¼ amount of folates in unfermented rye bran (native or peeled). Ti ¼ time and Te ¼ temperature.
Fig. 4. Influence of fermentation time and temperature on total amount of phenolic compounds after MeOH extraction (free phenolics mg/100 g) in fermented bran made from (A) native rye bran (B) peeled rye bran. UF ¼ amount of MeOH extractable total phenolics in unfermented rye bran (native or peeled). Ti ¼ time and Te ¼ temperature.
Fig. 5. Influence of fermentation time and temperature on total amount of free ferulic acid (mg/100 g) in fermented bran made from (A) native rye bran (B) peeled rye bran. UF ¼ amount of free ferulic acid in unfermented rye bran (native or peeled). Ti ¼ time and Te ¼ temperature.
During the fermentation of native bran, the levels of ARs only changed slightly (Fig. 6). In both cases, fermentation time had a quadratic effect on the level of ARs, so that the change in alkyresorcinols was largest at a time range of 12–14 h independent on the fermentation temperature. However, quadratic effect was opposite for native and peeled bran as highest level was obtained for native bran and lowest for peeled bran in the experimental region in question. No significant change in AR homologue (alkyl side chains C17:0–C25:0) pattern was seen during fermentation (Table 1c).
3.4. Pentosans The amount of soluble pentosans increased twofold in fermentation of both types of bran in optimum conditions (Fig. 7). However, the influence of fermentation conditions differed to some extent for native and peeled bran: e.g. long fermentation time at high temperature did not enhance solubilization of pentosans on native rye bran, but doubled the amounts in fermentation of bran from peeled grains. Also, the overall levels of soluble pentosans were higher in fermentations of bran from the peeled grains.
ARTICLE IN PRESS K. Katina et al. / Food Microbiology 24 (2007) 175–186
183
Fig. 6. Influence of fermentation time and temperature on total amount of alkylresorcinols (mg/100 g Ars) in fermented bran made from (A) native rye bran (B) peeled rye bran. UF ¼ amount of Ars in unfermented rye bran (native or peeled). Ti ¼ time and Te ¼ temperature.
Fig. 7. Influence of fermentation time and temperature on total amount of soluble pentosans (% dw) in fermented bran made from (A) native rye bran (B) peeled rye bran. UF ¼ amount of soluble pentosans in unfermented rye bran (native or peeled). Ti ¼ time and Te ¼ temperature.
4. Discussion Fermentation of bran, either from native or peeled grain, was found to be an efficient means to increase the level of folates, easily extractable phenolic compounds, ferulic acid and solubilized pentosans of rye, if appropriate fermentation conditions were applied. However, the levels of these compounds and soluble pentosans were different in the two types of bran. Peeling prior to bran separation reduced the microbial load in rye bran, which was also reflected in the microbial community during yeast fermentation. The presence of the outermost layers (percar/testa) of grain in the bran was thus an important determinant of many fermentation-induced changes of the raw material. The microbial load on the grain surface was largely diminished when 11% of the grain was removed prior to bran separation. Earlier studies have also shown that the highest microbial counts are detected in the outer layers of the kernel, and significant quantities of microbial contaminants and their toxic metabolites can be removed by peeling of the grains (Bennett and Richard, 1996; Berghofer et al., 2003). However, the milling plant and equipment may also be a potential source for additional microbial populations of the bran. Particularly spore formers may reside in the equipment (Berghofer et al., 2003).
The current study showed that only low numbers of Bacillus species were found in both types of rye brans, and due to the intensive growth of LAB their number did not increase during fermentation. This was beneficial, because heat-resistant spores of Bacillus species could survive throughout the baking process and cause serious problems in the end product. Bacillus species cause spoilage of bread by rope formation and may also constitute a health risk (Thompson et al., 1993; Rosenquist and Hansen, 1995). Rope spoilage of wholemeal bread has been reported to occur more frequently than the spoilage of white wheat bread, possibly due the spore-forming bacteria present in bran fractions (Eyles et al., 1989). Therefore, it is extremely important to control the growth of Bacillus spp. when bran ingredients are utilized in bakery products. Development of the indigenous LAB community was strongly dependent of the fermentation conditions. Only low numbers of LAB were detected in the brans, but their number increased tremendously during fermentation at elevated temperatures, especially in fermentation of bran from native grains which resulted in intensive acidification. Due to growth of indigenous LAB, the yeast fermentation will gradually become mixed (sourdough) fermentation, depending on fermentation time and temperature. Milder acidity formation associated with bran from peeled grains was due to a more uniform LAB community, and lower
ARTICLE IN PRESS 184
K. Katina et al. / Food Microbiology 24 (2007) 175–186
endogenous and microbial enzyme activity. Concentration of proteolytic and xylanolytic enzymes at outer grain layers has been shown in rye and wheat (Brijs et al., 1999; Gys et al., 2004). Milder acidity provides a technological advantage for bran from peeled grain, because strong acidic or pungent cereal food flavour does not appeal to most consumers, and thus limits the use of fermented rye bran as a food ingredient. Furthermore, the limited microbial community of peeled bran allows easier control of the fermentation process with added starter cultures. Bran is a rich source of folates. Liukkonen et al. (2003) found 10 fold differences in the folate contents of the endosperm and bran fractions of rye. In agreement with the previous investigations of wholegrain rye (Liukkonen et al., 2003; Kariluoto et al., 2004), the folate content in the current study increased by over 100% during the fermentation of both brans studied. This was mainly due the folate production by yeast, as suggested earlier (Liukkonen et al., 2003; Kariluoto et al., 2005; Ja¨gerstad et al., 2005). However, yeast growth could not totally explain the increase in folates in bran ferementations. The highest folate level was obtained when growth of indigenous LAB was pronounced (fermentation for 20 h at 35 1C), which may indicate a supportive role of indigenous LAB in folate synthesis, as also suggested by Kariluoto et al. (2005). It is well known that some LAB are able to synthezize folates (Sybesma et al., 2003; Kariluoto et al., 2005) and that others can consume folates (Kariluoto et al., 2005). Thus, the composition of indigenous microbial community or added LAB may have a significant impact on the folate levels of fermented bran. Phenolic compounds, especially phenolic acids, are partly responsible for insoluble cell wall structures of cereal kernels by forming cross-links between polysaccharides and lignin (Faulds and Williamson, 1999). During fermentation of both types of brans, the levels of free phenolics analysed by Folin-C method increased, indicating increased liberation of bound phenolic compounds from the polymeric rye bran structure. Since the Folin-C method measures all phenolics, including aromatic amino acids, it is possible that increase in level of free phenolics also partly reflected increased proteolysis during the fermentation. Fermentation also increased the level of free ferulic acid, but the effect was highly dependent on the fermentation conditions and type of bran. Among the studied bioactive compounds, ARs were an exception because their concentration slightly increased in fermentation of native bran but decreased in that of peeled bran. In our previous study on wholemeal rye flour (Liukkonen et al., 2003), fermentation with added LAB and yeast for 22 h at 30 1C had practically no effect on level of alkylresorcinols, which is consistent with the current study. Eventhough fermentation increased the levels of free phenolic compounds, the levels were always lower in bran from peeled than from native bran. This was due to both a higher initial level of phenolic compounds in the native bran, and propably also to higher endogenous- and
microbial-originated enzyme activity. An increase in falling number indicated reduced alfa-amylase activity in peeled bran, and also activity of other enzymes such proteases and xylanases can be assumed to be lower in comparison to native bran (Brijs et al., 1999; Gys et al., 2004). Cell wall degradation has been reported to happen during wheat bran fermentation by either endogenous or added enzymes (Katina et al., 2006). Also acid hydrolysis of dietary fibre components has been suggested to occur (Boskov-Hansen et al., 2002). According to Gys et al. (2004), endogenous xylanases degrade wheat arabinoxylans and debranning of wholegrain wheat reduces endogenous xylanase activity by 80%. In addition, Dornez et al. (2006) reported that over 90% of endoxylanase activity of grain originates from microbes located in the outer layers of grain. Thus, in fermentation of native bran, more extensive degradation of cell walls could be expected to liberate phenolic compounds from the rye bran structure more effectively due to higher enzyme activity. The highest level of free phenolics (analysed by the Folin method) was in both fermentations obtained from the most acidic fermentations. However, the highest level of free ferulic acid was obtained in fermentation conditions providing a pH value of 6–6.5 (20 h at 20 1C or 6 h at 35 1C), which is very near to the reported pH-optimum (pH 7) of cinnamoyl esterases of wholegrain rye flour (Boskov-Hansen et al., 2002). Accordingly, the level of free ferulic acid was lowest in fermentation conditions providing strong acidity (3.9–4.1), which inhibits the cinnamyl esterase. Thus, the level of free ferulic acids in bran fermentations appears to dependent on the pH mediated activation of endogenous enzymes of rye. The content of ARs was lower in the peeled sample compared to the native sample, which is in accordance with previous results which have shown that ARs are exclusively present in the bran fraction (Ross et al., 2004). Alkylresorcinols have previously been shown to be very stable during food processing and in this study also during fermentation. On the other hand, they become associated with the matrix, probably due to complex formation with amylose, which necessiate a hot propanol:water mixture for their extraction from processed samples. The observed solubilization of pentosans in rye bran fermentation is in accordance with previous results obtained with wholegrain rye flour (Boskov-Hansen et al., 2001). Solubilization of pentosans has a significant influence on the technological properties of wholegrain rye flour; especially in rye baking where the solubilization of pentosans is a basis for good texture of wholegrain rye breads. In addition, a positive influence of solubilization of pentosans has been reported in wheat flour baking (Maat et al., 1992; Courtin and Delcour, 2002) and also in highfibre baking using wheat bran (Katina et al., 2006). It is noteworthy that maximal solubilization of pentosans was obtained with fermented bran made of peeled grains independent of fermentation conditions. The different xylanase activity of brans from native and peeled grain is
ARTICLE IN PRESS K. Katina et al. / Food Microbiology 24 (2007) 175–186
suggested to cause the observed differences in solubilization of pentosans. Native bran most probably contains higher levels of either microbial or endogenous xylanases, which would degradade solubilized arabinoxylans (Gys et al., 2004; Dornez et al., 2006). 5. Conclusion This study established the potential of fermentation to enhance the bioactivity and technological potential of rye bran. The presence of indigenous LAB and enzymes concentrated to the outer layers of the grains contributes significantly to the changes induced in bran during fermentation. The extent of these changes can be modulated by changing the milling process during separation of the bran prior to fermentation. Acknowledgments This study is financially supported by The Nordic Joint Committee for Agricultural Research (NJK), project Rye Bran for Health (121) and by the European Commission in the Communities Sixth Framework Programme, Project HEALTHGRAIN (FP6-514008). This publication reflects only author’s views and the Community is not liable for any use that may be made of the information contained in this publication. We are grateful to the technical staff of VTT for their skilful technical assistance. References Bartolome´, B., Go´mez-Cordove´s, C., 1999. Barley spent grain: release of hydroxycinnamic acids (ferulic and p-coumaric acids) by commercial enzyme preparations. J. Sci. Food Agric. 79, 435–439. Bennett, G., Richard, J., 1996. Influence of processing on Fusarium mycotoxins in contaminated grains. Food Technol. 50 (5), 235–238. Berghofer, L., Hocking, A., Miskelly, D., Jansson, E., 2003. Microbiology of wheat and flour milling in Australia. Int. J. Food Microbiol. 85, 137–149. Boskov-Hansen, H., Andersen, M.F., Nielsen, L.M., Back-Knudsen, K-E., Meyer, A.S., Christensen, L.P., Hansen, A˚., 2002. Chances in dietary fibre, phenolic acids and activity of endogenous enzymes during rye bread making. Eur. Food Res. Technol. 214, 33–42. Brijs, K., Bleukx, W., Delcour, J., 1999. Proteolytic activities in dorminant rye (Secale cereale L.) grain. J. Agric. Food Chem. 47, 3572–3578. Bruce, J., 1996. Automated system rapidly identifies and characterizes micro-organisms in food. Food Technol. 50, 77–81. Courtin, C., Delcour, J., 2002. Arabinoxylans and endoxylanases in wheat flour bread-making. J. Cereal Sci. 35, 225–243. Devesa, A., Martı´ nez-Anaya, M., 2001. Characterization of waterextractable pentosans in enzyme-supplemented wheat sourdough process. Food Sci. Technol. Int. 72 (2), 145–153. Dornez, E., Joye, I., Gebruers, K., Delcour, J., Courtin, C., 2006. Wheatkernel-associated endoxylanases consist of a majority of microbial and a minority of wheat endogenous endoxylanases. J. Agric. Food Chem. 54 (11), 4028–4034. Eyles, M., Moss, R., Hocking, A., 1989. The microbial status of Australian flour and the effects of milling procedure on the microflora of wheat and flour. Food Australia 41, 704–708. Faulds, M., Williamson, G., 1999. The role of hydroxycinnamates in the plant cell wall. J. Sci. Food Agric. 79, 393–395.
185
Fro¨lich, W., Crakenberg, T., Asp, N.G., 1986. Enzymatic degradation of phytate (myo-inositol hexaphosphate) in whole grain flour suspensions and dough. A comparion between 31p NMR spectroscopy and ferric ion method. J. Cereal Sci. 4, 325–334. Gys, W., Courtin, C.M., Delcour, J.A., 2004. Reduction of xylanase activity in flour by debranning retards syruping in refrigerated doughs. J. Cereal Sci. 39 (3), 371–377. Heinio, R.L., Liukkonen, K.-H., Katina, K., Myllymaki, O., Poutanen, K., 2003. Milling fractionation of rye produces different sensory profiles of both flour and bread. Lebens. Wiss. Technol. 36 (6), 577–583. Jacobs Jr., D.R., Gallaher, D.D., 2004. Whole grain intake and cardiovascular disease: a review. Curr. Atherosclerosis Reports 6 (6), 415–423. Ja¨gerstad, M., Piironen, V., Walker, C., Ros, G., Carnovale, E., Holasova, M., Nau, H., 2005. Increasing natural folates through bioprocessing and biotechnology. Trends Food Sci. Technol. 16, 298–306. Kariluoto, S., Vahteristo, L., Salovaara, H., Katina, K., Liukkonen, KH., Piironen, V., 2004. Effect of baking method and fermentation on folate content of rye and wheat breads. Cereal Chem. 81, 134–139. Kariluoto, S., Aittamaa, M., Korhola, M., Salovaara, H., Vahteristo, L., Piironen, V., 2005. Effects of yeasts and bacteria on the levels of folates in rye sourdoughs. Int. J. Food Microbiol. 106 (2), 137–143. Katina, K., 2005. Sourdough: a tool for the improved flavour, texture and shelf-life of wheat bread. Doctoral Thesis, VTT Publications no. 569. VTT Biotechnology, Espoo, 92 pp.+app. 81pp. Katina, K., Salmenkallio-Marttila, M., Partanen, R., Forssell, P., Autio, K., 2006. Effects of sourdough and enzymes on staling of high-fibre wheat bread. Lebensm. Wiss. Technol. 39, 479–491. Lang, R., Jebb, S., 2003. Who consumes whole grains, and how much? Proc. Nutr. Soc. 62 (1), 123–127. Larsson, M., Sandberg, A.S., 1991. Phytate reduction in bread containing oat flour, oat bran or rye bran. J. Cereal Sci. 14, 141–149. Larsson, S.C., Giovannucci, E., Bergkvist, L., Wolk, A., 2005. Whole grain consumption and risk of colorectal cancer: a population-based cohort of 60,000 women. Br. J. Cancer 92, 1803–1807. Liukkonen, K.-H., Katina, K., Wilhelmsson, A., Myllymaki, O., Lampi, A.-M., Kariluoto, S., Piironen, V., Heinonen, S.M., Nurmi, T., Adlercreutz, H., Peltoketo, A., Pihlava, J.M., Hietaniemi, V., Poutanen, K., 2003. Process-induced changes on bioactive compounds in whole grain rye. Proc. Nutr. Soc. 62, 117–122. Maat, J., Verbakel, J., Stam, H., Santos de Silva, M., Bosse, M., Egmond, M., Hagemanns, M., Gorcom, R., Hessing, J., Hondel, C., Rotterdam, C., 1992. Xylanases and their application in bakery. In: Visser, J., Beldman, G., Kusters van Someren, M., Vorangen, A. (Eds.), Xylans and Xylanases. Elsevier Science Publishers, Amsterdam, The Netherlands, pp. 349–360. Murtaugh, M.A., Jacobs Jr., D.R., Steffen, L.M., Jacob, B., Marquart, L., 2003. Epidemiological support for the protection of whole grains against diabetes. Proc. Nutr. Soc. 62, 143–149. Piironen, V., Varo, P., Koivistoinen, P., 1987. Stability of tocopherols and tocotrienols in food preparation procedures. J. Food Compos. Anal. 1, 53–58. Rosenquist, H., Hansen, A˚., 1995. Contamination profiles and characterisation of Bacillus species in wheat bread and raw materials for bread production. Int. J. Food Microbiol. 26, 353–363. Ross, A.B., Kamal-Eldin, A., Jung, C., Shepherd, M.J., A˚man, P., 2001. Gas chromatographic analysis of alkylresorcinols in rye (Secale cereale L.) grains. J. Sci. Food Agric. 81, 1405–1411. Ross, A.B., Shepherd, M.J., Schu¨pphaus, M., Sinclair, V., Alfaro, B., Kamal-Eldin, A., A˚man, P., 2003. Alkylresorcinols in cereals and cereal products. J. Agric. Food Chem. 51, 4111–4118. Ross, A., Kamal-Eldin, A., A˚man, P., 2004. Dietary alkylresorcinols: absorption, bioactivities and possible use as biomarkers of whole-grain wheat- and rye-rich foods. Nutr. Rev. 62, 81–95. Rouau, X., Surget, A., 1994. A rapid semi-automated method for the determination of total and water e-extractable pentosans in wheat flours. Carbohydr. Polym. 24, 123–132.
ARTICLE IN PRESS 186
K. Katina et al. / Food Microbiology 24 (2007) 175–186
Salmenkallio-Marttila, M., Katina, K., Autio, K., 2001. Effects of bran fermentation on quality and microstructure of high-fiber wheat bread. Cereal Chem 78, 429–435. Salovaara, H., 2004. Lactic acid bacteria in cereal-based products. In: Salminen, S., von Wright, A., Ouwehand, A. (Eds.), Lactic Acid Bacteria—Microbiological and Functional Aspects. Marcel Dekker, New York, pp. 431–4513. Singleton, V.L., Orthofer, R., Lamuela-Raventos, R.M., 1999. Analysis of total phenols and other oxidation substrates and antioxidants by means of Folin–Ciocalteu reagent. Methods Enzymol. 299, 152–178.
Sybesma, W., Starrenburg, M., Tijsseling, L., Hoefnagel, M., Hugenholtz, J., 2003. Effects of cultivation conditions on folate production by lactic acid bacteria. Appl. Environ. Microbiol. 69, 4542–4548. Thompson, J., Dodd, C., Waites, W., 1993. Spoilage of bread by Bacillus. Int. Biodeter. Biodegr. 32, 55–66. Verdeal, K., Lorenz, K., 1977. Alkylresorcinols in wheat, rye, and triticale. Cereal Chem. 54, 475–483. Wilmotte, A., Van der Auwera, G., De Wachter, R., 1993. Structure of the 16S ribosomal RNA of the thermophilic cyanobacterium Chlorogloeopsis HTF (‘Mastigocladus laminosus HTF’) strain PCC7518, and phylogenetic analysis. FEBS Lett. 317, 96–100.