- Email: [email protected]
Raman studies of gluten proteins aggregation induced by dietary fibres
Food Chemistry 194 (2016) 86–94
Contents lists available at ScienceDirect
Food Chemistry journal homepage: www.elsevier.com/locate/foodchem
Raman studies of gluten proteins aggregation induced by dietary fibres Agnieszka Nawrocka a,⇑, Monika Szyman´ska-Chargot a, Antoni Mis´ a, Radosław Kowalski b, Wiesław I. Gruszecki c a
Bohdan Dobrzanski Institute of Agrophysics Polish Academy of Sciences, Dos´wiadczalna 4, 20-290 Lublin, Poland Department of Analysis and Evaluation of Food Quality, University of Life Sciences in Lublin, Skromna 8, 20-704 Lublin, Poland c Department of Biophysics, Institute of Physics, Maria Curie-Skłodowska University, Pl. Marii Curie-Skłodowskiej 1, 20-031 Lublin, Poland b
a r t i c l e
i n f o
Article history: Received 11 March 2015 Received in revised form 3 July 2015 Accepted 28 July 2015 Available online 31 July 2015 Keywords: Gluten protein Amide I band Disulphide bridges Aromatic acids Raman spectroscopy Dietary fibre
a b s t r a c t Interactions between gluten proteins and dietary fibre preparations are crucial in the baking industry. The addition of dietary fibre to bread causes significant reduction in its quality which is influenced by changes in the structure of gluten proteins. Fourier transform Raman spectroscopy was applied to determine changes in the structure of gluten proteins modified by seven dietary fibres. The commercially available gluten proteins without starch were mixed with the fibres in three concentrations: 3%, 6% and 9%. The obtained results showed that all fibres, regardless of their origin, caused the same kind of changes i.e. decrease in the a-helix content with a simultaneous increase in the content of antiparallel-b-sheet. The results indicated that presence of cellulose was the probable cause of these changes, and lead to aggregation or abnormal folding of the gluten proteins. Other changes observed in the gluten structure concerning b-structures, conformation of disulphide bridges, and aromatic amino acid environment, depended on the fibres chemical composition. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Wheat proteins include albumins, globulins, gliadins and glutenins but only the last two participate in the formation of a continuous viscoelastic network within dough. Glutenin polymers are made up of high and low molecular weight subunits which are attached to each other via disulphide bonds (Shewry, Halford, Belton, & Tatham, 2002). The polymers provide strength and elasticity for bread dough development whereas globular proteins, such as gliadins, impart viscosity to dough (Sivam, Sun-Waterhouse, Perera, & Waterhouse, 2012). Gliadins interact with the glutenin polymers via non-covalent hydrophobic interactions and hydrogen bonding. The structure of both proteins is crucial in the breadmaking process, but is also connected with gluten allergenicity. Of these two proteins, gliadins are considered strong food allergens. They cause IgE-mediated allergies such as asthma, atopic dermatitis or celiac disease (Bürk, Melms, Schulz, & Dichgans, 2001). Previous studies (Waga, 2004) have shown three parts of the gluten protein structure which may be responsible for gluten allergenicity. As the first is a short (‘‘toxic’’) amino acid
⇑ Corresponding author. E-mail addresses: [email protected] (A. Nawrocka), m.szymanska@ipan. lublin.pl (M. Szyman´ska-Chargot), [email protected] (A. Mis´), radoslaw. [email protected] (R. Kowalski), [email protected] (W.I. Gruszecki). http://dx.doi.org/10.1016/j.foodchem.2015.07.132 0308-8146/Ó 2015 Elsevier Ltd. All rights reserved.
sequence, it probably acts as an antibody-binding epitope in immunological reactions. Other structural elements considered as ‘‘toxic’’ are b-turns and disulphide bonds (Waga, 2004). Studies on limiting or eliminating allergenicity of gluten proteins by biochemical modifications have been carried out for several years. A considerable decrease in the gluten allergenicity was observed after treatment with transglutaminase (Leszczynska et al., 2002; Nakamura et al., 2013), acetic acid (Berti, Dolfini, & Forlani, 2002; Berti et al., 2007) or citric acid (Qiu, Sun, Cui, & Zhao, 2013), causing deamination. Modification can also be regarded as a reduction of the disulphide bridges by thioredoxin (Waga, Ka˛czkowski, & Zientarski, 2003). Although the thioredoxin considerably decreases immunoreactivity of gliadins and does not affect the rheological properties of gluten properties negatively, its high price makes this approach impractical for producing hypoallergenic food on an industrial scale. There were also studies in which antioxidants such as anthocyanins were used to modify gluten proteins with the aim to decrease allergen immunoreactivity (Taddei, Zanna, & Tozzi, 2013; Tozzi, Zanna, & Taddei, 2013). Dietary fibre preparations rich in antioxidants can be regarded as a good candidates for decreasing gluten allergenicity. In a previous study, changes in the structure of gluten proteins concerning b-turns and disulphide bridges has been evidenced (Nawrocka et al., 2015). The fibre additives are added to bread to increase
A. Nawrocka et al. / Food Chemistry 194 (2016) 86–94
the dietary fibre intake. Since bread is a principal component of western diets it is a convenient way to deliver fibre polysaccharides and antioxidants (Sivam, Sun-Waterhouse, Quek, & Perera, 2010). However, the main problem of dietary fibre addition in baking is a significant reduction of bread quality, which is connected with changes in the structure of gluten proteins. There are a great number of reports confirming a negative effect of dietary fibre supplementation on the rheological properties of bread dough (Mis´, 2011; Mis´ & Dziki, 2013; Peressini & Sensidoni, 2009), but only a few reports correspond to changes in structure of gluten proteins after addition of dietary fibre preparations (Sivam, Sun-Waterhouse, Perera, & Waterhouse, 2013; Sivam et al., 2012). The dietary fibres used in the present study were chosen because of their different sources of origin, and hence different chemical composition. Due to their different chemical composition they could interact with gluten proteins in different ways causing changes in the protein structure. The objective of this study was to determine changes in the secondary and tertiary structure of gluten proteins mixed directly with dietary fibre preparation without starch. Some changes in the gluten structure are postulated to decrease the allergenicity of gluten proteins (Taddei et al., 2013). 2. Materials and methods 2.1. Materials Wheat gluten, and sodium chloride were purchased from Sigma–Aldrich and used as received. Chokeberry (CHB), cranberry (CRB), cacao (CAC), carrot (CRR), oat (OAT) and flax (FLX) fibres were purchased from Microstructure (Warsaw, Poland). The carob (CAR) fibre was purchased from Carob General Application (Valencia, Spain). The content of total, soluble and insoluble dietary fibre (DF) were given by the fibre manufacturers and is presented in the previous article (Nawrocka et al., 2015). Double-distilled water was used for all experiments. 2.2. Analysis of pectin content in dietary fibre preparations Pectin content was determined by the Morris method, modified _ Horubała & Jarczyk (1973). 20.00 g of by Pijanowski, Mrozewski, the triplicate dietary fibre preparations was weighed into a flask and 30 ml of distilled water was added. The samples were shaken for 15 min, and then filtered into a dry flask. The residue and filter paper were transferred back to the flask. Then 30 ml of water was added and shaken. The extraction was repeated three times. The resulting filtrate was transferred into a 100 ml volumetric flask and topped up with distilled water. 25 ml of the solution was dispensed into a beaker and 50 ml acetone was added. The sample was allowed to stand for 1 hour and then the solution containing the precipitate was filtered through a dry and pre-weighed filter paper. Filter paper and precipitate were dried at 75 °C until reaching a constant weight. Pectin content was calculated from the difference in weight of the filter paper with the sediment after drying and the weight of the dried filter paper taking into account the volume of pectin solution taken, to determine the mass of the sample. 2.3. Analysis of cellulose content in dietary fibre preparations Cellulose content was determined applying the Kürchner– Hanack method. This method is based on the insolubility of cellulose in water and its resistance to the action of dilute acids and bases. 1 g of the triplicate dietary fibre preparations was weighed into a flask and a 50 ml of mixture of nitric acid (d = 1.4 g/ml, 5 ml), acetic acid (70%, 75 ml) and trichloroacetic acid (2 g) was
87
added. The solution was then boiled in flask with a condenser. The solution was then filtered through a Büchner funnel. The precipitate was washed with hot water until the acid reaction stopped (checked universal indicator paper). The solid was then washed on the filter with about 15 ml of ethyl alcohol. Then the filter paper containing an insoluble residue was dried in oven (100 °C) and measured (Skulmowski, 1974). 2.4. Fourier transform infrared (FT-IR) spectra collection The FTIR spectra of seven dietary fibres were recorded with a Nicolet 6700 FTIR spectrometer (Thermo Scientific, Madison, WI, USA) equipped with a diamond attenuated total reflectance attachment. The FTIR spectra were recorded between 4000 and 400 cm 1 at 4 cm 1 intervals. Each spectrum resulted from 128 scans to obtain an optimal signal-to-noise ratio. Each spectrum of the studied dietary fibres was corrected with a linear baseline using OMNIC software (v. 8.2, Thermo Fischer Scientific Inc., Madison, WI, USA). 2.5. Gluten-fibre mixture preparation The gluten-fibre mixtures were kneaded for 3 min in the vibrat_ (Sadkiewicz Instruments, Bydgoszcz, Poland). The ing kneader SZ-1 fibre contents were 3%, 6% and 9% w/w in relation to the gluten-fibre mixture weight (at the same moisture basis). The 6% content of dietary fibre was used in the previous study (Nawrocka et al., 2015) where the model flour (starch and gluten only) – dietary fibre mixtures were examined. This allows for easy comparison between fibre only effects (present study) and both fibre and starch effects (previous study) on the gluten protein conformation. The gluten samples were washed out from the gluten-fibre mixtures by using Glutomatic 2200 (Perten InstrumentsHuddinge, Sweden). Next the gluten samples were freeze-dried for 24 h, milled in laboratory grinder and used in FT-Raman measurements. 2.6. Fourier transform Raman (FT-Raman) spectra collection and data manipulation The FT-Raman spectra were acquired on an FT-Raman module (NXR FT Raman) for a Nicolet 6700 FT-IR bench using an InGaAs detector and CaF2 beam splitter (Thermo Scientific, Madison, WI, USA). The samples were placed in stainless cubes and were illuminated using Nd:YAG excitation laser operating at 1064 nm. The maximum laser power was 1 W. The spectra were recorded over the range of 3500–150 cm 1 and each spectrum was an average of 256 scans at 8 cm 1 resolution. The analysed spectra were averaged over five registered spectra. The gluten samples were analysed in powder form. Spectral data from the sample scans were baseline-corrected, and normalised against a phenylalanine band at 1003 cm 1, using ORIGIN (version 9.0 PRO, OriginLab Corporation, USA). The disulphide bridge region (490–550 cm 1), and aromatic amino acids environment: tyrosine doublet (I(850)/I(830)), tryptophan band (I(760)), and amide I band (1570–1720 cm 1), were analysed. Structural analysis of the disulphide bridges (percentage distribution of disulphide bridge conformations: gauche-gauche-gauche (SSg-g-g), trans-gauche-gauche (SSt-g-g), and trans-gauche-trans (SSt-g-t)) were also conducted using ORIGIN. The Gaussian components in the S–S region were determined on the basis of a second derivative spectrum. The derivative spectrum was obtained using a five-point, two-degree polynomial function. The band profiles of the components for all samples are shown in Table 1 in the Supplementary Material. The S–S bands were assigned to each conformation according to Sugeta (1975).
88
A. Nawrocka et al. / Food Chemistry 194 (2016) 86–94
Table 1 Cellulose and pectin content in studied dietary fibres. Standard deviation is placed in brackets. Fibre component
Cellulose content (g/100 g)*
Pectin content (g/100 g)
CHB CRB CRR CAC CAR OAT FLX
17.3 27.2 14.3 19.9 12.8 4.5 6.9
0.28 0.37 0.64 0.64 0.34 nd nd
(0.10) (0.05) (0.12) (0.06) (0.04)
nd, not detected because of difficulties with samples preparation (obtained extracts containing pectin block the pores of the filter paper, which prevented proper execution of quantitative analysis). * Expanded uncertainty u = 11.3%.
To determine changes in the secondary structure of gluten proteins (amide I band) two kinds of difference spectra were calculated according to Nawrocka et al. (2015). Briefly, a spectrum of a commercially available gluten was subtracted from spectrum for the gluten-fibre mixture. Next, the dietary fibre spectrum was subtracted from the first difference spectrum. The second difference spectrum showed interactions between the gluten proteins and dietary fibre components. The difference spectra for different contents of dietary fibre (3%, 6% and 9%) were compared in pairs – the spectrum of 3% with 6% and spectrum of 6% with 9%. 3. Results and discussion 3.1. Changes in secondary structure (amide I band) The peptide group in proteins give rise to a few characteristic vibrational bands in the Raman spectrum. Amide I (1570– 1720 cm 1) and amide III (1230–1340 cm 1) bands can be used for characterisation of the protein secondary structure. The intensity of the amide III band was relatively low and overlapped with the spectra for some components in our samples. For this reason, the changes in secondary structure of gluten proteins after addition of dietary fibre preparations were studied by analysis of the amide I band. Although a glutamine side chain can contribute to the amide I band, it has been demonstrated that this contribution is not greater than 10% for most proteins (Popineau, Bonenfant, Cornec, & Pezolet, 1994). The difference spectra were calculated in a two-phase method, and shown in Fig. 1. Fig. 1a–d presents the interactions between gluten proteins and fruit fibres (CHB and CRB) used at three different concentrations: 3%, 6% and 9%. Comparison of the difference spectra for both fibres showed that the changes observed in the secondary structure concerned mainly with the a-helix (1654 cm 1), and antiparallel-b-sheets (1677 and 1697 cm 1). The band at 1677 cm 1 can be also assigned to turns and loops, but analysis of the structural changes indicated that this band is rather connected with the antiparallel b-sheet. Detailed analysis of the CHB difference spectra (see Fig. 1a and b) showed that the band located at 1616 and 1630 cm 1 did not change their location after increasing the fibre content from 3% to 6%. Further increasing of the CHB fibre content caused a shift of the second band to the lower wavenumbers (from 1630 to 1621 cm 1). The appearance of this band revealed the formation of protein aggregates and is assigned to intermolecular hydrogen-bonded b-sheets (Mangavel, Barbot, Popineau, & Gueguen, 2001). Moreover, the band located at 1643 cm 1 in the CHB3 spectrum also shifted to smaller wavenumbers and was observed at 1637 cm 1 for CHB6 and CHB9. The band located at 1637 cm 1 is characteristic of b-sheet structures (Secundo & Guerieri, 2005), while the band at 1643 cm 1 was assigned to the stretching of carbonyl groups in
the b-turns (Wellner, Belton, & Tatham, 1996) indicating the possible formation of internal hydrogen bonds, especially intramolecular. This band could be also assigned to random coils but the observed shift to lower wavenumbers indicates that b-turns rather than random coils participated into formation of b-sheet structures connected by hydrogen bonds (Secundo & Guerieri, 2005). The other band connected with the same kind of oscillations in b-turns was observed at 1667 cm 1 in all the spectra of the CHB fibre (see Fig. 1a and b). Bands connected with b-sheet structures were located at 1677 and 1685 cm 1, and were determined as negative and positive, respectively. After increasing the fibre content to 9%, the bands at 1677 and 1685 cm 1 turned into positive and negative bands, respectively. The presence of these bands indicated that two kinds of b-sheet structures exist in the aggregated gluten proteins – one rich in intermolecular hydrogen bonds (1685 cm 1), and one which contains smaller number of this kind of H-bonds (1677 cm 1). Increase in fibre concentration caused increase in the amount of b-sheet structures characterised by the lower number of intermolecular H-bonds. A similar phenomenon was observed by Secundo and Guerieri (2005) who studied the interactions between dextrin and gliadins. Addition of the dextrin caused a decrease in intermolecular hydrogen-bonded b-sheets in gliadins. As a result, a shift of the band from 1682 cm 1 to 1674 cm 1 was observed for the gliadin-dextrin mixture. The band located at 1677 cm 1 can be also assigned to turns and loops but analysis indicates that it is rather connected with antiparallel-b-sheets. CRB as an additive caused slightly different changes in the secondary structure of the gluten proteins (Fig. 1c and d). First of all, the bands at 1610 and 1633 cm 1 did not change position irrespective of the fibre content and remained positive and negative, respectively. This indicated that the CRB additive did not influence aggregate formation and hydrogen-bonded b-sheets. A 6%-addition of CRB could induce formation of pseudo-b-sheet structures between two protein complexes which was confirmed by the appearance of the band at 1616 cm 1 (Fig. 1c and d). The intensity of this band increased after increasing the fibre content to 9%. The addition of the highest content of the fibre also caused the appearance of a band at 1647 cm 1. The appearance of this band, with simultaneous absence of the band at ca. 1668 cm 1, suggested that CRB fibre induced formation of hydrogen bonds between the side chain C@O groups of glutamins and polysaccharide molecules (Secundo & Guerieri, 2005). The difference spectra of the carrot fibre are presented in Fig. 1e and f. The CRR fibre caused changes in the region of a-helix (1654 cm 1), b-sheet (1620, 1685 and 1697 cm 1) and aggregates (1606 cm 1). Detailed analysis of the spectra showed that the band at 1619 cm 1 (CRR3) shifted to 1624 cm 1 (CRR6 and 9) as a result of increasing fibre content. This shift can be connected with strongly hydrogen-bonded b-sheets which are linked with the formation of the protein aggregates (Mangavel et al., 2001). Furthermore, increasing the fibre content from 3% to 6% caused a decrease in the amount of non-hydrogen bonded b-turns (1673 cm 1), and appearance of the band at 1645 cm 1. The band at 1645 cm 1 was also observed on the CRB9 spectrum. In the CRR6 and CRR9 spectra a band at 1635 cm 1 was observed which is connected with b-sheet structures. The absence of this band on the CRR6 spectrum can be connected with the conformation of disulphide bridges. As can be seen in Table 3, the number of disulphide bridges in each conformation is similar for the samples CRR3 and CRR9. The interactions between gluten proteins and cacao and carob additives are presented in Fig. 1g, h and i, j, respectively. The CAC fibre caused changes in the secondary structure of the gluten proteins, mainly concerning a-helical and b-sheet structures. As can be seen from Fig. 1g and h, there are four common bands located
A. Nawrocka et al. / Food Chemistry 194 (2016) 86–94
89
Fig. 1. Difference spectra depicted interactions between gluten proteins and different concentrations (3%, 6% and 9% w/w) of dietary fibres: chokeberry (a and b), cranberry (c and d), carrot (e and f), cacao (g and h), carob (i and j), flax (k and l) and oat (m and n). The spectra were obtained by subtraction of control sample and fibres spectra from spectra of gluten-fibre mixtures (AGR, aggregates; aH, a-helix; bS, b-sheet; abS, antiparallel-b-sheet; pbS, pseudo-b-sheet; bT, b-turns).
at 1615, 1631, 1656 and 1691 cm 1. An additional band at around 1650 cm 1 appeared on the CAC6 spectrum. This band was also
observed on the CRB9 and CRR6 spectra. Similar changes in the difference spectra were observed after addition of the CAR fibre in a
90
A. Nawrocka et al. / Food Chemistry 194 (2016) 86–94
Fig. 1 (continued)
concentration of 3% and 6% (bands at 1617, 1633, 1654 and 1695 cm 1). Additionally, two bands were located at 1669 and 1681 cm 1 corresponding to b-turns and b-sheets, respectively. A 9%-addition of the CAR fibre caused noticeable changes in the difference spectrum, where the band at 1682 cm 1 shifted to 1678 cm 1. The shift indicated that the number of b-sheets linked by intermolecular H-bonds decreased (Secundo & Guerieri, 2005). Furthermore, three new bands appeared at 1606, 1623 and 1639 cm 1 assigned to aggregates and b-sheet structures (the last two bands), respectively. Moreover, the band at 1623 cm 1 could also indicate the formation of aggregates (Mangavel et al., 2001). The changes in secondary structure caused by the last two fibres (flax and oat) are shown in Fig. 1k, l and m, n, respectively. Comparison of the difference spectra for the FLX additive showed that the changes observed in secondary structure concerned mainly with the a-helix (1654 cm 1), antiparallel-b-sheets (1697 cm 1) and aggregates (1608 cm 1). Detailed analysis of the difference spectra showed that the bands at 1648 and 1688 cm 1
disappear on the FLX6 spectrum. After a 9% addition of the fibre, a band at 1665 cm 1 appeared and the band at 1679 shifted to 1684 cm 1. The OAT fibre also caused changes in the a-helix (1656 cm 1) and b-sheet (1617, 1681 and 1695 cm 1) regions. Comparison of the OAT3 and OAT6 spectrum depicted a shift of two bands, connected with b-sheet structures. One of the bands shifted from 1622 to 1616 cm 1, the other one from 1634 to 1627 cm 1. Moreover, there was a new band observed at 1665 cm 1 in the OAT6 spectrum. This band shifted to 1673 cm 1 after using 9% concentration of the fibre. Analysis of the difference spectra indicated that all dietary fibre additives caused similar changes in the secondary structure of the gluten proteins. The most noticeable changes induced by all dietary fibres were observed in the region connected with a-helix (a negative band located at ca. 1654 cm 1). Simultaneously, a positive band, connected with antiparallel-b-sheets, appeared at ca. 1693 cm 1. This indicates that a fibre component, which was present in all additives, induced a connection of a-helices from two
A. Nawrocka et al. / Food Chemistry 194 (2016) 86–94
protein complexes to form antiparallel-b-sheet structures (Sunde et al., 1997). Moreover, the increase in fibre content caused a simultaneous decrease in the intensity of the a-helix band. This suggests that all fibres contained the same compound, which induced similar changes in secondary structure. The fibre component, which is most abundant in all fibre additives, was cellulose (see in Table 1). Similar results were obtained by Correa, Ferrer, Anon, and Ferrero (2014), who studied interactions between modified celluloses and gluten proteins. Further changes in gluten structure concern mainly b structures (b-sheets and b-turns) and are connected with the chemical composition of dietary fibre additives. The shift of the bands toward higher/lower wavenumbers in the region of 1620–1640 cm 1 and 1675–1685 cm 1 indicated changes of the hydrogen-bonding pattern (Mangavel et al., 2001). Furthermore, the presence of the band at 1622 cm 1 is connected with protein aggregation (Mangavel et al., 2001). In the case of four of the fibres (CAC, CRR, CRB and CAR), a new band was observed at around 1645 cm 1 in the spectrum after the addition of 6% and 9% for the first two and last two fibres, respectively. According to Secundo and Guerieri (2005), the appearance of this band with the simultaneous absence of the band at ca. 1668 cm 1 suggested that these fibres induced the formation of hydrogen bonds between the side chain C@O groups in glutamin and polysaccharide molecules. As shown in Table 1, the pectin presence can be responsible for this change because its content is the highest for CAC and CRR fibres comparing with two times lower for CRB and CAR fibres. 3.2. Changes in disulphide bridges (S–S) conformation Gluten proteins contain ca. 2% cysteine, which participates in S– S formation, and is extremely important for the structure and functionality of gluten. Most cysteines are present in an oxidised state and form either intrachain disulphide bonds within a protein or interchain disulphide bonds between proteins. These bonds are the main target for most redox reactions that occur during dough preparation (Wieser, 2007). Analysis of the disulphide region of the control sample showed five bands located at 505 (SSg-g-g), 514, 522sh (SSt-g-g), 535 and 540sh cm 1 (SSt-g-t). Table 2 presents the results from the deconvolution of the S–S band of the control and gluten proteins modified by dietary fibres. Deconvoluted spectra of the region depicting disulphide bridges for all samples are presented in Figs. S1 and S2 in the Supplementary Material. The control sample contained 43%, 45%, and 12% of disulphide bridges in the g-g-g, t-g-g, and t-g-t conformation, respectively. Similar results were obtained for gluten washed out from model flour, which was reconstituted from two commercial components: wheat gluten and wheat starch (Nawrocka et al., 2015) and for gluten washed out from commercial flour (Gomez, Ferrer, Anon, & Puppo, 2013). As shown in Table 2, a 3% addition of the CHB and CAR fibres in comparison with control sample caused decrease in the number of S–S bonds in the g-g-g conformation. Simultaneously the OAT and FLX fibres increased the number of S–S bonds in this conformation, whereas CRB, CRR and CAC did not influence these bonds. In the case of t-g-g conformation, the amount of disulphide bridges increased for only CRB, and decreased for CRR, CAC, CAR, OAT and FLX compared to control sample. The number of S–S bonds in the t-g-t conformation increased for almost all samples other than CRB, FLX and OAT. The addition of CRB and FLX caused a decrease in SSt-g-t, while it remained unchanged for OAT. Analysis of the percentage distribution for disulphide bridge conformation showed that the number of S–S bonds in the g-g-g conformation increased with increasing fibre content for CHB, CRB, CAC and CAR, and decreased for FLX, while a correlation was not established for CRR and OAT fibres. The amount of SSt-g-g slightly increased
91
only for CHB, decreased for CRB, CAC and CAR, and remained constant for CRR. A relationship was not established for OAT and FLX. In the case of SSt-g-t, the number of disulphide bridges increased for CRR and FLX, decreased for CHB, and remained constant for CAR and OAT. Analysis of the S–S band location (see Table 2 and Figs. S1 and S2 in the Supporting Material) showed that bands corresponding to the g-g-g and t-g-t conformations after addition of the fibre additives, could be deconvoluted with one Gaussian function. The band assigned to the t-g-g conformation was deconvoluted by one Gaussian function located at 520 cm 1 for CAC and FLX fibres, or by two Gaussian functions with maxima at 515 and 520 cm 1 for CHB, CRB, CRR, CAR and OAT fibres, and the control sample. The presence of the band at 515 cm 1 indicated the formation of intrachain disulphide bonds (Ferrer, Gomez, Anon, & Puppo, 2011). Disappearance of this band, in the case of the CAC and FLX preparations, could indicate cleavage of the intrachain disulphide bonds. Therefore, it can be assumed that the increasing content of dietary fibre affects the cleavage of the intrachain S–S bonds. In the case of CHB fibres, was observed an opposite tendency was observed – intrachain disulphide bonds appeared after addition of the highest amount of fibre. Comparison with our previous studies (Nawrocka et al., 2015) showed that only two additives (CAC and CHB) caused similar changes in the t-g-g conformation concerning intrachain S–S bonds after a 6%-addition of the fibre. This could be connected with the absence of starch during gluten-fibre mixing or a different method of sample preparation. Generally, it can be said that increasing the content of dietary fibre has increased the amount of the S–S bridges in the g-g-g conformation for all additives other than oat and flex. As shown in Table 1, the oat and flex fibres contain the least amount of cellulose. The results suggest that the presence of a cellulose polymer can prevent the protein complexes from changing the conformation of the disulphide bridges to less stable ones (t-g-g and t-g-t). The absence of the band connected with intrachain S–S bonds (517 cm 1) also confirms this observation. 3.3. Changes in aromatic amino acids environment Raman bands corresponding to the oscillation of two amino acids: tyrosine (TYR) and tryptophan (TRP), provided information about H-bonding (tyrosine doublet) and hydrophobic environment (tryptophan band at 760 cm 1). The tyrosine residues occur periodically throughout the length of gluten proteins and are often found in repeat pairs of tyrosine residues. Although disulphide bridges are assumed to be the main bonds responsible for the gluten network formation during dough mixing, tyrosine bonds (H-bonds) also participate in this process (Wieser, 2007). The ratio of the tyrosine doublet (I(850)/I(830)) is known as a good indicator of hydrogen bonding by the phenolic hydroxyl group. The I(850)/I(830) for the control sample was 1.45. This value was similar to those calculated by Nawrocka et al. (2015) and Ferrer et al. (2011). As shown in Table 3, the I(850)/I(830) values decreased after a 3%-addition of dietary fibre for almost all additives other than CHB (the ratio increased), and CRR (the ratio remained constant). Further addition of the fibres caused a slight increase of the ratio for CRB, CAR, OAT and FLX. In the case of CHB, CRR and CAC, a decrease in ratio value was observed with increasing the fibre content. The decrease in the ratio value suggested the formation of new hydrogen bonds, whereas an increase in the ratio value could indicate that TYR residues acted as a positive charge acceptor, favouring local charge repulsion between protein molecules (Carey, 1982). However, the observed increase in the ratio value is small and the residues can still be treated as ‘‘normal tyrosine’’
92
A. Nawrocka et al. / Food Chemistry 194 (2016) 86–94
Table 2 Percentage distribution of disulphide bridges conformation for control sample and fibre-modified (CHB, CRB, CRR, CAC, CAR, OAT and FLX) gluten. The capital letter C means constant. Standard deviation is placed in brackets.
Table 3 Analysis of aromatic acids (tyrosine and tryptophan) environment. The capital letter C means constant. Standard deviation is placed in brackets.
according to Siamwiza, Lord, and Chen (1975). An increase in the ratio value was observed, after addition of pectin and locust bean gum to the dough, by Linlaud, Ferrer, Puppo, and Ferrero (2011). However, comparison of the tyrosine doublet values and pectin content (see Table 1) did not show any correlation. The decrease in the I(850)/I(830), observed for CHB, CRR and CAC fibres, can be connected with the type of carbonyl group depending on the compound in which it is observed e.g. ester, or aldehyde. According to Lam, Proctor, and Meullenet (2001), oscillations of the ester carbonyl group are observed at ca. 1746 cm 1 on the FTIR spectrum, whereas the aldehyde group has a maximum at
around 1731 cm 1. As can be seen from the FTIR spectra of the studied fibres (presented in Fig. S3 in Supplementary Material), CHB, CRR and CAC fibres have maxima at 1734, 1734, and 1730 cm 1, respectively. Those are connected with oscillations of the aldehyde group. The aldehyde group in the fibres can be exposed to form new H-bonds with the phenyl groups of TYR residues which are not dissociated. The decrease in the tyrosine doublet ratio can be also connected with the anthocyanin content. A similar decrease in the ratio value was observed by Taddei et al. (2013), who studied the interactions between gliadins and anthocyanins. Although cranberry fibre contains a similar amount of anthocyanins as cacao fibre, the value of I(850)/I(830) increases with increasing fibre content. This can be connected to cellulose content which is the highest among the studied additives. The functional groups of the cellulose chains may be charged and this will favour local charge repulsion between protein molecules. The tryptophan (TRP) band at 760 cm 1 has been proposed as an indicator of the strength of H-bonding and hydrophobicity of the indole ring (Linlaud et al., 2011). A decrease in the band intensity suggested that TRP residues come from a buried hydrophobic microenvironment and contributes to the formation of a more disordered structure; in contrast an increase in the band intensity is connected with the opposite phenomenon. The addition of dietary fibre preparations caused strong changes in the intensity of the TRP band for all samples other than CRB fibre. There was a slight increase (CRR, CAC) or decrease (CHB, CRB, FLX) observed in the TRP band intensity after a 3%-addition of the dietary fibre. A completely different behaviour of the TRP band was observed for the OAT fibres. The amount of the added fibre was irrelevant. In other words, the first addition of the fibre caused a noticeable decrease in the band intensity, while further increasing of the amount of OAT fibre did not affect the microenvironment of TRP. As can be seen from Table 3, increasing amounts of dietary fibre caused a two– three fold increase in the intensity of the TRP band. This increase in the band intensity indicates that TRP residues are buried in the hydrophobic microenvironment. The opposite behaviour of TRP residues after OAT addition can be connected with the smallest amount of cellulose in this fibre preparation. However, after the
A. Nawrocka et al. / Food Chemistry 194 (2016) 86–94
addition of CRB, which is characterised by the largest content of the cellulose, the intensity of the TRP band remains constant. 4. Discussion As can be seen from the results presented above, the addition of dietary fibre caused aggregation or abnormal folding of gluten proteins. The main change observed in the structure of the gluten proteins is the formation of b-like structures from the a-helices. Previous studies (Nawrocka et al., 2015) showed that a 6%-addition of the fibres to the model flour caused changes mainly in b-turn and b-sheet structures, but did not influence the a-helix. Since similar changes in the proteins were observed, it is possible that all fibres could contain the same compound. Therefore it could be stated that the cellulose component of dietary fibre, which is the most abundant, could induce changes in gluten structure. A comparison of the previous and present studies also revealed the importance of starch molecules in the flour. The starch molecules can protect the gluten protein against undesirable changes in its structure. A few other changes in the structure of gluten proteins were also observed. Some of the fibres (CHB and CRR) caused an appearance of a band at around 1622 cm 1 which indicates the formation of protein aggregates (Mangavel et al., 2001). In this case, not only hydrogen bonds participated in the aggregate formation but also intrachain disulphide bridges did. This assumption is supported by the increase of the tyrosine doublet ratio to 0.89, which suggests formation of new H-bonds (Siamwiza et al., 1975), and the presence of a band at 517 cm 1, which is connected with intrachain disulphide bonds (Ferrer et al., 2011). A shift was also observed of the band connected with antiparallel-b-sheets in the region 1675–1695 cm 1 which indicates the formation of new H-bonds in b-sheet structures. These bonds can form between side chain amino acids, or between side chain amino acids and polysaccharides molecules. The tyrosine and tryptophan residues rather did not participate in the formation of those bonds due to an increase in both the tyrosine doublet value and the intensity of TRP band. Moreover, changes in the conformation of disulphide bridges (increase in the amount of S–S in conformations t-g-g and t-g-t) after a 9%-addition of the FLX could indicate that these new bonds form between side chain amino acids which leads to folding of the gluten proteins. In the case of carob fibre the shift was observed in the opposite way (from 1682 to 1678 cm 1). This suggests cleavage of part of the H-bonds and it is in agreement with the tyrosine doublet values and intensity of the TRP band. A 9%-addition of the CAR fibre also caused an increase in the amount of the S–S bonds in the stable conformation (g-g-g) and cleavage of intrachain disulphide bridges (absence of the band at 516 cm 1 as shown in Table 2). 5. Conclusions Raman spectroscopy revealed that the addition of dietary fibre preparations induced changes in the structure of gluten proteins. Direct mixing of commercially available gluten proteins with fibre preparations shows that all preparations, regardless of their origin, caused a decrease in a-helix with a simultaneous increase in antiparallel-b-sheets. This suggests that the a-helices combine through hydrogen bonds in antiparallel-b-sheets. This phenomenon may be connected with aggregation or abnormal folding of the protein complexes. Furthermore, the results indicate presence of the same compound in all preparations (cellulose) which induces this change. Moreover, the other changes observed in the gluten structure depends on the fibres origin and their chemical composition. This could suggest an interaction of pectic
93
polysaccharides with gluten proteins. Comparison of the present studies with previous studies show that starch protects the gluten proteins against undesirable changes concerning a-helical structures.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.foodchem.2015. 07.132. References Berti, C., Dolfini, E., & Forlani, F. (2002). Effects on coeliac activity of gluten proteins modified by chemical deamination. Polish Journal of Food and Nutrition Science, 11(52), 135–137. Berti, C., Roncoroni, L., Falini, M. L., Caramanico, R., Dolfini, E., Bardella, M. T., et al. (2007). Celiac-related properties of chemically and enzymatically modified gluten proteins. Journal of Agricultural and Food Chemistry, 55, 2482–2488. Bürk, K., Melms, A., Schulz, J. B., & Dichgans, J. (2001). Effectiveness of intravenous immunoglobulin therapy in cerebellar ataxia associated with gluten sensitivity. Annals of Neurology, 50, 827–828. Carey, P. R. (1982). Protein conformation from Raman and resonance 400 Raman spectra. In P. R. Carey (Ed.), Biochemical applications of Raman and resonance raman spectroscopies (pp. 65–116). New York: Academic Press. Correa, M. J., Ferrer, E., Anon, M. C., & Ferrero, C. (2014). Interaction of modified celluloses and pectins with gluten proteins. Food Hydrocolloids, 35, 91–99. Ferrer, E. G., Gomez, A. V., Anon, M. C., & Puppo, M. C. (2011). Structural changes in gluten protein structure after addition of emulsifier. A Raman spectroscopy study. Spectrochimica Acta A, 79, 278–281. Gomez, A. V., Ferrer, E. G., Anon, M. C., & Puppo, M. C. (2013). Changes in secondary structure of gluten proteins due to emulsifiers. Journal of Molecular Structure, 1033, 51–58. Lam, H. S., Proctor, A., & Meullenet, J.-F. (2001). Free fatty acid formation and lipid oxidation on milled rice. Journal of the American Oil Chemists’ Society, 78, 1271–1275. Leszczyn´ska, J., Ła˛cka, A., Pytasz, U., Szemraj, J., Lukamowicz, J., & Lewin´ski, A. (2002). The effect of proteolysis on the gliadin immunogenicity. Polish Journal of Food Nutrition Sciences, 11/52, 145–148. Linlaud, N., Ferrer, E., Puppo, M. C., & Ferrero, C. (2011). Hydrocolloid interaction with water, protein, and starch in wheat dough. Journal of Agricultural and Food Chemistry, 59, 713–719. Mangavel, C., Barbot, J., Popineau, Y., & Gueguen, J. (2001). Evolution of wheat gliadins conformation during film formation: A Fourier transform infrared study. Journal of Agricultural and Food Chemistry, 49, 867–872. Mis´, A. (2011). Interpretation of mechanical spectra of carob fibre and oatwholemeal-enriched wheat dough using non-linear regression models. Journal of Food Engineering, 102, 369–379. Mis´, A., & Dziki, D. (2013). Extensograph curve profile model used for characterising the impact of dietary fibre on wheat dough. Journal of Cereal Science, 57, 471–479. Nakamura, R., Nakamura, R., Sakai, S., Adachi, R., Hachisuka, A., & Urisu, A. (2013). Tissue transglutaminase generates deaminated epitopes on gluten, increasing reactivity with hydrolyzed wheat protein-sensitized IgE. Journal of Allergy and Clinical Immunology, 132, 1436 e4–1438 e4. Nawrocka, A., Szyman´ska-Chargot, M., Mis´, A., Ptaszyn´ska, A. A., Kowalski, R., Was´ko, P., et al. (2015). Influence of dietary fibre on gluten proteins structure – A study on model flour with application of FT-Raman spectroscopy. Journal of Raman Spectroscopy, 46, 309–316. Peressini, D., & Sensidoni, A. (2009). Effect of soluble dietary fibre addition on rheological and breadmaking properties of wheat doughs. Journal of Cereal Science, 49, 190–201. _ Pijanowski, E., Mrozewski, S., Horubała, A., & Jarczyk, A. (1973). Technology of fruit and vegetable products. Warsaw: PWRiL. Popineau, Y., Bonenfant, S., Cornec, M., & Pezolet, M. (1994). A study by infrared spectroscopy of the conformations of gluten proteins differing in their gliadin and glutenin composition. Journal of Cereal Science, 20, 15–22. Qiu, C., Sun, W., Cui, C., & Zhao, M. (2013). Effect of citric acid deamination on in vitro digestibility and antioxidant properties of wheat gluten. Food Chemistry, 141, 2772–2778. Secundo, F., & Guerieri, N. (2005). ATR-FR/IR study on the interactions between gliadins and dextrin and their effect on protein secondary structure. Journal of Agricultural and Food Chemistry, 53, 1757–1764. Shewry, P. R., Halford, N. G., Belton, P. S., & Tatham, A. S. (2002). The structure and properties of gluten: An elastic protein from wheat grain. Philosophical Transactions. The Royal Society, 357, 133–142. Siamwiza, M. N., Lord, R. C., & Chen, M. C. (1975). Interpretation of the doublet at 850 and 830 cm 1 in the Raman spectra of tyrosyl residues in proteins and certain model compounds. Biochemistry, 14, 4870–4876. Sivam, A. S., Sun-Waterhouse, D., Perera, C. O., & Waterhouse, G. I. N. (2012). Exploring the interactions between blackcurrant polyphenols, pectin and wheat
94
A. Nawrocka et al. / Food Chemistry 194 (2016) 86–94
biopolymers in model breads: A FTIR and HPLC investigation. Food Chemistry, 131, 802–810. Sivam, A. S., Sun-Waterhouse, D., Perera, C. O., & Waterhouse, G. I. N. (2013). Application pf FT-IR and Raman spectroscopy for the study of biopolymers in breads fortified with fibre and polyphenols. Food Research International, 50, 574–585. Sivam, A. S., Sun-Waterhouse, D., Quek, S. Y., & Perera, C. O. (2010). Properties of bread dough with added fibre polysaccharides and phenolic antioxidants: A review. Journal of Food Science, 75, R163–R174. Skulmowski, J. (1974). Methods for determination of fodder composition and quality. Warsaw: PWRiL. Sugeta, H. (1975). Normal vibrations and molecular conformations of dialkyl disulphides. Spectrochimica Acta A, 31, 1729–1737. Sunde, M., Serpell, L. C., Bartlam, M., Fraser, P. E., Pepys, M. B., & Blake, C. C. F. (1997). Common core structure of amyloid fibrils by synchrotron X-ray diffraction. Journal of Molecular Biology, 273, 729–739.
Taddei, P., Zanna, N., & Tozzi, S. (2013). Raman characterization of the interactions between gliadins and anthocyanidins. Journal of Raman Spectroscopy, 44, 1435–1439. Tozzi, S., Zanna, N., & Taddei, P. (2013). Study on the interaction between gliadins and a coumarin as molecular model system of the gliadins-anthocyanidins complexes. Food Chemistry, 141, 3586–3597. Waga, J. (2004). Structure and allergenicity of wheat gluten proteins – A review. Polish Journal of Food and Nutrition Sciences, 13(54), 327–338. Waga, J., Ka˛czkowski, J., & Zientarski, J. (2003). Influence of thioredoxin-h on flour baking properties and gliadin immunoreactivity in spelt and common wheat genotypes. Polish Journal of Food and Nutrition Sciences, 12(53), 13–16. Wellner, N., Belton, P. S., & Tatham, A. S. (1996). Fourier transform IR spectroscopic study of hydration-induced structure changes in the solid state of x-gliadins. Biochemical Journal, 319, 741–747. Wieser, H. (2007). Chemistry of gluten proteins. Food Microbiology, 24, 115–119.
Contents lists available at ScienceDirect
Food Chemistry journal homepage: www.elsevier.com/locate/foodchem
Raman studies of gluten proteins aggregation induced by dietary fibres Agnieszka Nawrocka a,⇑, Monika Szyman´ska-Chargot a, Antoni Mis´ a, Radosław Kowalski b, Wiesław I. Gruszecki c a
Bohdan Dobrzanski Institute of Agrophysics Polish Academy of Sciences, Dos´wiadczalna 4, 20-290 Lublin, Poland Department of Analysis and Evaluation of Food Quality, University of Life Sciences in Lublin, Skromna 8, 20-704 Lublin, Poland c Department of Biophysics, Institute of Physics, Maria Curie-Skłodowska University, Pl. Marii Curie-Skłodowskiej 1, 20-031 Lublin, Poland b
a r t i c l e
i n f o
Article history: Received 11 March 2015 Received in revised form 3 July 2015 Accepted 28 July 2015 Available online 31 July 2015 Keywords: Gluten protein Amide I band Disulphide bridges Aromatic acids Raman spectroscopy Dietary fibre
a b s t r a c t Interactions between gluten proteins and dietary fibre preparations are crucial in the baking industry. The addition of dietary fibre to bread causes significant reduction in its quality which is influenced by changes in the structure of gluten proteins. Fourier transform Raman spectroscopy was applied to determine changes in the structure of gluten proteins modified by seven dietary fibres. The commercially available gluten proteins without starch were mixed with the fibres in three concentrations: 3%, 6% and 9%. The obtained results showed that all fibres, regardless of their origin, caused the same kind of changes i.e. decrease in the a-helix content with a simultaneous increase in the content of antiparallel-b-sheet. The results indicated that presence of cellulose was the probable cause of these changes, and lead to aggregation or abnormal folding of the gluten proteins. Other changes observed in the gluten structure concerning b-structures, conformation of disulphide bridges, and aromatic amino acid environment, depended on the fibres chemical composition. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Wheat proteins include albumins, globulins, gliadins and glutenins but only the last two participate in the formation of a continuous viscoelastic network within dough. Glutenin polymers are made up of high and low molecular weight subunits which are attached to each other via disulphide bonds (Shewry, Halford, Belton, & Tatham, 2002). The polymers provide strength and elasticity for bread dough development whereas globular proteins, such as gliadins, impart viscosity to dough (Sivam, Sun-Waterhouse, Perera, & Waterhouse, 2012). Gliadins interact with the glutenin polymers via non-covalent hydrophobic interactions and hydrogen bonding. The structure of both proteins is crucial in the breadmaking process, but is also connected with gluten allergenicity. Of these two proteins, gliadins are considered strong food allergens. They cause IgE-mediated allergies such as asthma, atopic dermatitis or celiac disease (Bürk, Melms, Schulz, & Dichgans, 2001). Previous studies (Waga, 2004) have shown three parts of the gluten protein structure which may be responsible for gluten allergenicity. As the first is a short (‘‘toxic’’) amino acid
⇑ Corresponding author. E-mail addresses: [email protected] (A. Nawrocka), m.szymanska@ipan. lublin.pl (M. Szyman´ska-Chargot), [email protected] (A. Mis´), radoslaw. [email protected] (R. Kowalski), [email protected] (W.I. Gruszecki). http://dx.doi.org/10.1016/j.foodchem.2015.07.132 0308-8146/Ó 2015 Elsevier Ltd. All rights reserved.
sequence, it probably acts as an antibody-binding epitope in immunological reactions. Other structural elements considered as ‘‘toxic’’ are b-turns and disulphide bonds (Waga, 2004). Studies on limiting or eliminating allergenicity of gluten proteins by biochemical modifications have been carried out for several years. A considerable decrease in the gluten allergenicity was observed after treatment with transglutaminase (Leszczynska et al., 2002; Nakamura et al., 2013), acetic acid (Berti, Dolfini, & Forlani, 2002; Berti et al., 2007) or citric acid (Qiu, Sun, Cui, & Zhao, 2013), causing deamination. Modification can also be regarded as a reduction of the disulphide bridges by thioredoxin (Waga, Ka˛czkowski, & Zientarski, 2003). Although the thioredoxin considerably decreases immunoreactivity of gliadins and does not affect the rheological properties of gluten properties negatively, its high price makes this approach impractical for producing hypoallergenic food on an industrial scale. There were also studies in which antioxidants such as anthocyanins were used to modify gluten proteins with the aim to decrease allergen immunoreactivity (Taddei, Zanna, & Tozzi, 2013; Tozzi, Zanna, & Taddei, 2013). Dietary fibre preparations rich in antioxidants can be regarded as a good candidates for decreasing gluten allergenicity. In a previous study, changes in the structure of gluten proteins concerning b-turns and disulphide bridges has been evidenced (Nawrocka et al., 2015). The fibre additives are added to bread to increase
A. Nawrocka et al. / Food Chemistry 194 (2016) 86–94
the dietary fibre intake. Since bread is a principal component of western diets it is a convenient way to deliver fibre polysaccharides and antioxidants (Sivam, Sun-Waterhouse, Quek, & Perera, 2010). However, the main problem of dietary fibre addition in baking is a significant reduction of bread quality, which is connected with changes in the structure of gluten proteins. There are a great number of reports confirming a negative effect of dietary fibre supplementation on the rheological properties of bread dough (Mis´, 2011; Mis´ & Dziki, 2013; Peressini & Sensidoni, 2009), but only a few reports correspond to changes in structure of gluten proteins after addition of dietary fibre preparations (Sivam, Sun-Waterhouse, Perera, & Waterhouse, 2013; Sivam et al., 2012). The dietary fibres used in the present study were chosen because of their different sources of origin, and hence different chemical composition. Due to their different chemical composition they could interact with gluten proteins in different ways causing changes in the protein structure. The objective of this study was to determine changes in the secondary and tertiary structure of gluten proteins mixed directly with dietary fibre preparation without starch. Some changes in the gluten structure are postulated to decrease the allergenicity of gluten proteins (Taddei et al., 2013). 2. Materials and methods 2.1. Materials Wheat gluten, and sodium chloride were purchased from Sigma–Aldrich and used as received. Chokeberry (CHB), cranberry (CRB), cacao (CAC), carrot (CRR), oat (OAT) and flax (FLX) fibres were purchased from Microstructure (Warsaw, Poland). The carob (CAR) fibre was purchased from Carob General Application (Valencia, Spain). The content of total, soluble and insoluble dietary fibre (DF) were given by the fibre manufacturers and is presented in the previous article (Nawrocka et al., 2015). Double-distilled water was used for all experiments. 2.2. Analysis of pectin content in dietary fibre preparations Pectin content was determined by the Morris method, modified _ Horubała & Jarczyk (1973). 20.00 g of by Pijanowski, Mrozewski, the triplicate dietary fibre preparations was weighed into a flask and 30 ml of distilled water was added. The samples were shaken for 15 min, and then filtered into a dry flask. The residue and filter paper were transferred back to the flask. Then 30 ml of water was added and shaken. The extraction was repeated three times. The resulting filtrate was transferred into a 100 ml volumetric flask and topped up with distilled water. 25 ml of the solution was dispensed into a beaker and 50 ml acetone was added. The sample was allowed to stand for 1 hour and then the solution containing the precipitate was filtered through a dry and pre-weighed filter paper. Filter paper and precipitate were dried at 75 °C until reaching a constant weight. Pectin content was calculated from the difference in weight of the filter paper with the sediment after drying and the weight of the dried filter paper taking into account the volume of pectin solution taken, to determine the mass of the sample. 2.3. Analysis of cellulose content in dietary fibre preparations Cellulose content was determined applying the Kürchner– Hanack method. This method is based on the insolubility of cellulose in water and its resistance to the action of dilute acids and bases. 1 g of the triplicate dietary fibre preparations was weighed into a flask and a 50 ml of mixture of nitric acid (d = 1.4 g/ml, 5 ml), acetic acid (70%, 75 ml) and trichloroacetic acid (2 g) was
87
added. The solution was then boiled in flask with a condenser. The solution was then filtered through a Büchner funnel. The precipitate was washed with hot water until the acid reaction stopped (checked universal indicator paper). The solid was then washed on the filter with about 15 ml of ethyl alcohol. Then the filter paper containing an insoluble residue was dried in oven (100 °C) and measured (Skulmowski, 1974). 2.4. Fourier transform infrared (FT-IR) spectra collection The FTIR spectra of seven dietary fibres were recorded with a Nicolet 6700 FTIR spectrometer (Thermo Scientific, Madison, WI, USA) equipped with a diamond attenuated total reflectance attachment. The FTIR spectra were recorded between 4000 and 400 cm 1 at 4 cm 1 intervals. Each spectrum resulted from 128 scans to obtain an optimal signal-to-noise ratio. Each spectrum of the studied dietary fibres was corrected with a linear baseline using OMNIC software (v. 8.2, Thermo Fischer Scientific Inc., Madison, WI, USA). 2.5. Gluten-fibre mixture preparation The gluten-fibre mixtures were kneaded for 3 min in the vibrat_ (Sadkiewicz Instruments, Bydgoszcz, Poland). The ing kneader SZ-1 fibre contents were 3%, 6% and 9% w/w in relation to the gluten-fibre mixture weight (at the same moisture basis). The 6% content of dietary fibre was used in the previous study (Nawrocka et al., 2015) where the model flour (starch and gluten only) – dietary fibre mixtures were examined. This allows for easy comparison between fibre only effects (present study) and both fibre and starch effects (previous study) on the gluten protein conformation. The gluten samples were washed out from the gluten-fibre mixtures by using Glutomatic 2200 (Perten InstrumentsHuddinge, Sweden). Next the gluten samples were freeze-dried for 24 h, milled in laboratory grinder and used in FT-Raman measurements. 2.6. Fourier transform Raman (FT-Raman) spectra collection and data manipulation The FT-Raman spectra were acquired on an FT-Raman module (NXR FT Raman) for a Nicolet 6700 FT-IR bench using an InGaAs detector and CaF2 beam splitter (Thermo Scientific, Madison, WI, USA). The samples were placed in stainless cubes and were illuminated using Nd:YAG excitation laser operating at 1064 nm. The maximum laser power was 1 W. The spectra were recorded over the range of 3500–150 cm 1 and each spectrum was an average of 256 scans at 8 cm 1 resolution. The analysed spectra were averaged over five registered spectra. The gluten samples were analysed in powder form. Spectral data from the sample scans were baseline-corrected, and normalised against a phenylalanine band at 1003 cm 1, using ORIGIN (version 9.0 PRO, OriginLab Corporation, USA). The disulphide bridge region (490–550 cm 1), and aromatic amino acids environment: tyrosine doublet (I(850)/I(830)), tryptophan band (I(760)), and amide I band (1570–1720 cm 1), were analysed. Structural analysis of the disulphide bridges (percentage distribution of disulphide bridge conformations: gauche-gauche-gauche (SSg-g-g), trans-gauche-gauche (SSt-g-g), and trans-gauche-trans (SSt-g-t)) were also conducted using ORIGIN. The Gaussian components in the S–S region were determined on the basis of a second derivative spectrum. The derivative spectrum was obtained using a five-point, two-degree polynomial function. The band profiles of the components for all samples are shown in Table 1 in the Supplementary Material. The S–S bands were assigned to each conformation according to Sugeta (1975).
88
A. Nawrocka et al. / Food Chemistry 194 (2016) 86–94
Table 1 Cellulose and pectin content in studied dietary fibres. Standard deviation is placed in brackets. Fibre component
Cellulose content (g/100 g)*
Pectin content (g/100 g)
CHB CRB CRR CAC CAR OAT FLX
17.3 27.2 14.3 19.9 12.8 4.5 6.9
0.28 0.37 0.64 0.64 0.34 nd nd
(0.10) (0.05) (0.12) (0.06) (0.04)
nd, not detected because of difficulties with samples preparation (obtained extracts containing pectin block the pores of the filter paper, which prevented proper execution of quantitative analysis). * Expanded uncertainty u = 11.3%.
To determine changes in the secondary structure of gluten proteins (amide I band) two kinds of difference spectra were calculated according to Nawrocka et al. (2015). Briefly, a spectrum of a commercially available gluten was subtracted from spectrum for the gluten-fibre mixture. Next, the dietary fibre spectrum was subtracted from the first difference spectrum. The second difference spectrum showed interactions between the gluten proteins and dietary fibre components. The difference spectra for different contents of dietary fibre (3%, 6% and 9%) were compared in pairs – the spectrum of 3% with 6% and spectrum of 6% with 9%. 3. Results and discussion 3.1. Changes in secondary structure (amide I band) The peptide group in proteins give rise to a few characteristic vibrational bands in the Raman spectrum. Amide I (1570– 1720 cm 1) and amide III (1230–1340 cm 1) bands can be used for characterisation of the protein secondary structure. The intensity of the amide III band was relatively low and overlapped with the spectra for some components in our samples. For this reason, the changes in secondary structure of gluten proteins after addition of dietary fibre preparations were studied by analysis of the amide I band. Although a glutamine side chain can contribute to the amide I band, it has been demonstrated that this contribution is not greater than 10% for most proteins (Popineau, Bonenfant, Cornec, & Pezolet, 1994). The difference spectra were calculated in a two-phase method, and shown in Fig. 1. Fig. 1a–d presents the interactions between gluten proteins and fruit fibres (CHB and CRB) used at three different concentrations: 3%, 6% and 9%. Comparison of the difference spectra for both fibres showed that the changes observed in the secondary structure concerned mainly with the a-helix (1654 cm 1), and antiparallel-b-sheets (1677 and 1697 cm 1). The band at 1677 cm 1 can be also assigned to turns and loops, but analysis of the structural changes indicated that this band is rather connected with the antiparallel b-sheet. Detailed analysis of the CHB difference spectra (see Fig. 1a and b) showed that the band located at 1616 and 1630 cm 1 did not change their location after increasing the fibre content from 3% to 6%. Further increasing of the CHB fibre content caused a shift of the second band to the lower wavenumbers (from 1630 to 1621 cm 1). The appearance of this band revealed the formation of protein aggregates and is assigned to intermolecular hydrogen-bonded b-sheets (Mangavel, Barbot, Popineau, & Gueguen, 2001). Moreover, the band located at 1643 cm 1 in the CHB3 spectrum also shifted to smaller wavenumbers and was observed at 1637 cm 1 for CHB6 and CHB9. The band located at 1637 cm 1 is characteristic of b-sheet structures (Secundo & Guerieri, 2005), while the band at 1643 cm 1 was assigned to the stretching of carbonyl groups in
the b-turns (Wellner, Belton, & Tatham, 1996) indicating the possible formation of internal hydrogen bonds, especially intramolecular. This band could be also assigned to random coils but the observed shift to lower wavenumbers indicates that b-turns rather than random coils participated into formation of b-sheet structures connected by hydrogen bonds (Secundo & Guerieri, 2005). The other band connected with the same kind of oscillations in b-turns was observed at 1667 cm 1 in all the spectra of the CHB fibre (see Fig. 1a and b). Bands connected with b-sheet structures were located at 1677 and 1685 cm 1, and were determined as negative and positive, respectively. After increasing the fibre content to 9%, the bands at 1677 and 1685 cm 1 turned into positive and negative bands, respectively. The presence of these bands indicated that two kinds of b-sheet structures exist in the aggregated gluten proteins – one rich in intermolecular hydrogen bonds (1685 cm 1), and one which contains smaller number of this kind of H-bonds (1677 cm 1). Increase in fibre concentration caused increase in the amount of b-sheet structures characterised by the lower number of intermolecular H-bonds. A similar phenomenon was observed by Secundo and Guerieri (2005) who studied the interactions between dextrin and gliadins. Addition of the dextrin caused a decrease in intermolecular hydrogen-bonded b-sheets in gliadins. As a result, a shift of the band from 1682 cm 1 to 1674 cm 1 was observed for the gliadin-dextrin mixture. The band located at 1677 cm 1 can be also assigned to turns and loops but analysis indicates that it is rather connected with antiparallel-b-sheets. CRB as an additive caused slightly different changes in the secondary structure of the gluten proteins (Fig. 1c and d). First of all, the bands at 1610 and 1633 cm 1 did not change position irrespective of the fibre content and remained positive and negative, respectively. This indicated that the CRB additive did not influence aggregate formation and hydrogen-bonded b-sheets. A 6%-addition of CRB could induce formation of pseudo-b-sheet structures between two protein complexes which was confirmed by the appearance of the band at 1616 cm 1 (Fig. 1c and d). The intensity of this band increased after increasing the fibre content to 9%. The addition of the highest content of the fibre also caused the appearance of a band at 1647 cm 1. The appearance of this band, with simultaneous absence of the band at ca. 1668 cm 1, suggested that CRB fibre induced formation of hydrogen bonds between the side chain C@O groups of glutamins and polysaccharide molecules (Secundo & Guerieri, 2005). The difference spectra of the carrot fibre are presented in Fig. 1e and f. The CRR fibre caused changes in the region of a-helix (1654 cm 1), b-sheet (1620, 1685 and 1697 cm 1) and aggregates (1606 cm 1). Detailed analysis of the spectra showed that the band at 1619 cm 1 (CRR3) shifted to 1624 cm 1 (CRR6 and 9) as a result of increasing fibre content. This shift can be connected with strongly hydrogen-bonded b-sheets which are linked with the formation of the protein aggregates (Mangavel et al., 2001). Furthermore, increasing the fibre content from 3% to 6% caused a decrease in the amount of non-hydrogen bonded b-turns (1673 cm 1), and appearance of the band at 1645 cm 1. The band at 1645 cm 1 was also observed on the CRB9 spectrum. In the CRR6 and CRR9 spectra a band at 1635 cm 1 was observed which is connected with b-sheet structures. The absence of this band on the CRR6 spectrum can be connected with the conformation of disulphide bridges. As can be seen in Table 3, the number of disulphide bridges in each conformation is similar for the samples CRR3 and CRR9. The interactions between gluten proteins and cacao and carob additives are presented in Fig. 1g, h and i, j, respectively. The CAC fibre caused changes in the secondary structure of the gluten proteins, mainly concerning a-helical and b-sheet structures. As can be seen from Fig. 1g and h, there are four common bands located
A. Nawrocka et al. / Food Chemistry 194 (2016) 86–94
89
Fig. 1. Difference spectra depicted interactions between gluten proteins and different concentrations (3%, 6% and 9% w/w) of dietary fibres: chokeberry (a and b), cranberry (c and d), carrot (e and f), cacao (g and h), carob (i and j), flax (k and l) and oat (m and n). The spectra were obtained by subtraction of control sample and fibres spectra from spectra of gluten-fibre mixtures (AGR, aggregates; aH, a-helix; bS, b-sheet; abS, antiparallel-b-sheet; pbS, pseudo-b-sheet; bT, b-turns).
at 1615, 1631, 1656 and 1691 cm 1. An additional band at around 1650 cm 1 appeared on the CAC6 spectrum. This band was also
observed on the CRB9 and CRR6 spectra. Similar changes in the difference spectra were observed after addition of the CAR fibre in a
90
A. Nawrocka et al. / Food Chemistry 194 (2016) 86–94
Fig. 1 (continued)
concentration of 3% and 6% (bands at 1617, 1633, 1654 and 1695 cm 1). Additionally, two bands were located at 1669 and 1681 cm 1 corresponding to b-turns and b-sheets, respectively. A 9%-addition of the CAR fibre caused noticeable changes in the difference spectrum, where the band at 1682 cm 1 shifted to 1678 cm 1. The shift indicated that the number of b-sheets linked by intermolecular H-bonds decreased (Secundo & Guerieri, 2005). Furthermore, three new bands appeared at 1606, 1623 and 1639 cm 1 assigned to aggregates and b-sheet structures (the last two bands), respectively. Moreover, the band at 1623 cm 1 could also indicate the formation of aggregates (Mangavel et al., 2001). The changes in secondary structure caused by the last two fibres (flax and oat) are shown in Fig. 1k, l and m, n, respectively. Comparison of the difference spectra for the FLX additive showed that the changes observed in secondary structure concerned mainly with the a-helix (1654 cm 1), antiparallel-b-sheets (1697 cm 1) and aggregates (1608 cm 1). Detailed analysis of the difference spectra showed that the bands at 1648 and 1688 cm 1
disappear on the FLX6 spectrum. After a 9% addition of the fibre, a band at 1665 cm 1 appeared and the band at 1679 shifted to 1684 cm 1. The OAT fibre also caused changes in the a-helix (1656 cm 1) and b-sheet (1617, 1681 and 1695 cm 1) regions. Comparison of the OAT3 and OAT6 spectrum depicted a shift of two bands, connected with b-sheet structures. One of the bands shifted from 1622 to 1616 cm 1, the other one from 1634 to 1627 cm 1. Moreover, there was a new band observed at 1665 cm 1 in the OAT6 spectrum. This band shifted to 1673 cm 1 after using 9% concentration of the fibre. Analysis of the difference spectra indicated that all dietary fibre additives caused similar changes in the secondary structure of the gluten proteins. The most noticeable changes induced by all dietary fibres were observed in the region connected with a-helix (a negative band located at ca. 1654 cm 1). Simultaneously, a positive band, connected with antiparallel-b-sheets, appeared at ca. 1693 cm 1. This indicates that a fibre component, which was present in all additives, induced a connection of a-helices from two
A. Nawrocka et al. / Food Chemistry 194 (2016) 86–94
protein complexes to form antiparallel-b-sheet structures (Sunde et al., 1997). Moreover, the increase in fibre content caused a simultaneous decrease in the intensity of the a-helix band. This suggests that all fibres contained the same compound, which induced similar changes in secondary structure. The fibre component, which is most abundant in all fibre additives, was cellulose (see in Table 1). Similar results were obtained by Correa, Ferrer, Anon, and Ferrero (2014), who studied interactions between modified celluloses and gluten proteins. Further changes in gluten structure concern mainly b structures (b-sheets and b-turns) and are connected with the chemical composition of dietary fibre additives. The shift of the bands toward higher/lower wavenumbers in the region of 1620–1640 cm 1 and 1675–1685 cm 1 indicated changes of the hydrogen-bonding pattern (Mangavel et al., 2001). Furthermore, the presence of the band at 1622 cm 1 is connected with protein aggregation (Mangavel et al., 2001). In the case of four of the fibres (CAC, CRR, CRB and CAR), a new band was observed at around 1645 cm 1 in the spectrum after the addition of 6% and 9% for the first two and last two fibres, respectively. According to Secundo and Guerieri (2005), the appearance of this band with the simultaneous absence of the band at ca. 1668 cm 1 suggested that these fibres induced the formation of hydrogen bonds between the side chain C@O groups in glutamin and polysaccharide molecules. As shown in Table 1, the pectin presence can be responsible for this change because its content is the highest for CAC and CRR fibres comparing with two times lower for CRB and CAR fibres. 3.2. Changes in disulphide bridges (S–S) conformation Gluten proteins contain ca. 2% cysteine, which participates in S– S formation, and is extremely important for the structure and functionality of gluten. Most cysteines are present in an oxidised state and form either intrachain disulphide bonds within a protein or interchain disulphide bonds between proteins. These bonds are the main target for most redox reactions that occur during dough preparation (Wieser, 2007). Analysis of the disulphide region of the control sample showed five bands located at 505 (SSg-g-g), 514, 522sh (SSt-g-g), 535 and 540sh cm 1 (SSt-g-t). Table 2 presents the results from the deconvolution of the S–S band of the control and gluten proteins modified by dietary fibres. Deconvoluted spectra of the region depicting disulphide bridges for all samples are presented in Figs. S1 and S2 in the Supplementary Material. The control sample contained 43%, 45%, and 12% of disulphide bridges in the g-g-g, t-g-g, and t-g-t conformation, respectively. Similar results were obtained for gluten washed out from model flour, which was reconstituted from two commercial components: wheat gluten and wheat starch (Nawrocka et al., 2015) and for gluten washed out from commercial flour (Gomez, Ferrer, Anon, & Puppo, 2013). As shown in Table 2, a 3% addition of the CHB and CAR fibres in comparison with control sample caused decrease in the number of S–S bonds in the g-g-g conformation. Simultaneously the OAT and FLX fibres increased the number of S–S bonds in this conformation, whereas CRB, CRR and CAC did not influence these bonds. In the case of t-g-g conformation, the amount of disulphide bridges increased for only CRB, and decreased for CRR, CAC, CAR, OAT and FLX compared to control sample. The number of S–S bonds in the t-g-t conformation increased for almost all samples other than CRB, FLX and OAT. The addition of CRB and FLX caused a decrease in SSt-g-t, while it remained unchanged for OAT. Analysis of the percentage distribution for disulphide bridge conformation showed that the number of S–S bonds in the g-g-g conformation increased with increasing fibre content for CHB, CRB, CAC and CAR, and decreased for FLX, while a correlation was not established for CRR and OAT fibres. The amount of SSt-g-g slightly increased
91
only for CHB, decreased for CRB, CAC and CAR, and remained constant for CRR. A relationship was not established for OAT and FLX. In the case of SSt-g-t, the number of disulphide bridges increased for CRR and FLX, decreased for CHB, and remained constant for CAR and OAT. Analysis of the S–S band location (see Table 2 and Figs. S1 and S2 in the Supporting Material) showed that bands corresponding to the g-g-g and t-g-t conformations after addition of the fibre additives, could be deconvoluted with one Gaussian function. The band assigned to the t-g-g conformation was deconvoluted by one Gaussian function located at 520 cm 1 for CAC and FLX fibres, or by two Gaussian functions with maxima at 515 and 520 cm 1 for CHB, CRB, CRR, CAR and OAT fibres, and the control sample. The presence of the band at 515 cm 1 indicated the formation of intrachain disulphide bonds (Ferrer, Gomez, Anon, & Puppo, 2011). Disappearance of this band, in the case of the CAC and FLX preparations, could indicate cleavage of the intrachain disulphide bonds. Therefore, it can be assumed that the increasing content of dietary fibre affects the cleavage of the intrachain S–S bonds. In the case of CHB fibres, was observed an opposite tendency was observed – intrachain disulphide bonds appeared after addition of the highest amount of fibre. Comparison with our previous studies (Nawrocka et al., 2015) showed that only two additives (CAC and CHB) caused similar changes in the t-g-g conformation concerning intrachain S–S bonds after a 6%-addition of the fibre. This could be connected with the absence of starch during gluten-fibre mixing or a different method of sample preparation. Generally, it can be said that increasing the content of dietary fibre has increased the amount of the S–S bridges in the g-g-g conformation for all additives other than oat and flex. As shown in Table 1, the oat and flex fibres contain the least amount of cellulose. The results suggest that the presence of a cellulose polymer can prevent the protein complexes from changing the conformation of the disulphide bridges to less stable ones (t-g-g and t-g-t). The absence of the band connected with intrachain S–S bonds (517 cm 1) also confirms this observation. 3.3. Changes in aromatic amino acids environment Raman bands corresponding to the oscillation of two amino acids: tyrosine (TYR) and tryptophan (TRP), provided information about H-bonding (tyrosine doublet) and hydrophobic environment (tryptophan band at 760 cm 1). The tyrosine residues occur periodically throughout the length of gluten proteins and are often found in repeat pairs of tyrosine residues. Although disulphide bridges are assumed to be the main bonds responsible for the gluten network formation during dough mixing, tyrosine bonds (H-bonds) also participate in this process (Wieser, 2007). The ratio of the tyrosine doublet (I(850)/I(830)) is known as a good indicator of hydrogen bonding by the phenolic hydroxyl group. The I(850)/I(830) for the control sample was 1.45. This value was similar to those calculated by Nawrocka et al. (2015) and Ferrer et al. (2011). As shown in Table 3, the I(850)/I(830) values decreased after a 3%-addition of dietary fibre for almost all additives other than CHB (the ratio increased), and CRR (the ratio remained constant). Further addition of the fibres caused a slight increase of the ratio for CRB, CAR, OAT and FLX. In the case of CHB, CRR and CAC, a decrease in ratio value was observed with increasing the fibre content. The decrease in the ratio value suggested the formation of new hydrogen bonds, whereas an increase in the ratio value could indicate that TYR residues acted as a positive charge acceptor, favouring local charge repulsion between protein molecules (Carey, 1982). However, the observed increase in the ratio value is small and the residues can still be treated as ‘‘normal tyrosine’’
92
A. Nawrocka et al. / Food Chemistry 194 (2016) 86–94
Table 2 Percentage distribution of disulphide bridges conformation for control sample and fibre-modified (CHB, CRB, CRR, CAC, CAR, OAT and FLX) gluten. The capital letter C means constant. Standard deviation is placed in brackets.
Table 3 Analysis of aromatic acids (tyrosine and tryptophan) environment. The capital letter C means constant. Standard deviation is placed in brackets.
according to Siamwiza, Lord, and Chen (1975). An increase in the ratio value was observed, after addition of pectin and locust bean gum to the dough, by Linlaud, Ferrer, Puppo, and Ferrero (2011). However, comparison of the tyrosine doublet values and pectin content (see Table 1) did not show any correlation. The decrease in the I(850)/I(830), observed for CHB, CRR and CAC fibres, can be connected with the type of carbonyl group depending on the compound in which it is observed e.g. ester, or aldehyde. According to Lam, Proctor, and Meullenet (2001), oscillations of the ester carbonyl group are observed at ca. 1746 cm 1 on the FTIR spectrum, whereas the aldehyde group has a maximum at
around 1731 cm 1. As can be seen from the FTIR spectra of the studied fibres (presented in Fig. S3 in Supplementary Material), CHB, CRR and CAC fibres have maxima at 1734, 1734, and 1730 cm 1, respectively. Those are connected with oscillations of the aldehyde group. The aldehyde group in the fibres can be exposed to form new H-bonds with the phenyl groups of TYR residues which are not dissociated. The decrease in the tyrosine doublet ratio can be also connected with the anthocyanin content. A similar decrease in the ratio value was observed by Taddei et al. (2013), who studied the interactions between gliadins and anthocyanins. Although cranberry fibre contains a similar amount of anthocyanins as cacao fibre, the value of I(850)/I(830) increases with increasing fibre content. This can be connected to cellulose content which is the highest among the studied additives. The functional groups of the cellulose chains may be charged and this will favour local charge repulsion between protein molecules. The tryptophan (TRP) band at 760 cm 1 has been proposed as an indicator of the strength of H-bonding and hydrophobicity of the indole ring (Linlaud et al., 2011). A decrease in the band intensity suggested that TRP residues come from a buried hydrophobic microenvironment and contributes to the formation of a more disordered structure; in contrast an increase in the band intensity is connected with the opposite phenomenon. The addition of dietary fibre preparations caused strong changes in the intensity of the TRP band for all samples other than CRB fibre. There was a slight increase (CRR, CAC) or decrease (CHB, CRB, FLX) observed in the TRP band intensity after a 3%-addition of the dietary fibre. A completely different behaviour of the TRP band was observed for the OAT fibres. The amount of the added fibre was irrelevant. In other words, the first addition of the fibre caused a noticeable decrease in the band intensity, while further increasing of the amount of OAT fibre did not affect the microenvironment of TRP. As can be seen from Table 3, increasing amounts of dietary fibre caused a two– three fold increase in the intensity of the TRP band. This increase in the band intensity indicates that TRP residues are buried in the hydrophobic microenvironment. The opposite behaviour of TRP residues after OAT addition can be connected with the smallest amount of cellulose in this fibre preparation. However, after the
A. Nawrocka et al. / Food Chemistry 194 (2016) 86–94
addition of CRB, which is characterised by the largest content of the cellulose, the intensity of the TRP band remains constant. 4. Discussion As can be seen from the results presented above, the addition of dietary fibre caused aggregation or abnormal folding of gluten proteins. The main change observed in the structure of the gluten proteins is the formation of b-like structures from the a-helices. Previous studies (Nawrocka et al., 2015) showed that a 6%-addition of the fibres to the model flour caused changes mainly in b-turn and b-sheet structures, but did not influence the a-helix. Since similar changes in the proteins were observed, it is possible that all fibres could contain the same compound. Therefore it could be stated that the cellulose component of dietary fibre, which is the most abundant, could induce changes in gluten structure. A comparison of the previous and present studies also revealed the importance of starch molecules in the flour. The starch molecules can protect the gluten protein against undesirable changes in its structure. A few other changes in the structure of gluten proteins were also observed. Some of the fibres (CHB and CRR) caused an appearance of a band at around 1622 cm 1 which indicates the formation of protein aggregates (Mangavel et al., 2001). In this case, not only hydrogen bonds participated in the aggregate formation but also intrachain disulphide bridges did. This assumption is supported by the increase of the tyrosine doublet ratio to 0.89, which suggests formation of new H-bonds (Siamwiza et al., 1975), and the presence of a band at 517 cm 1, which is connected with intrachain disulphide bonds (Ferrer et al., 2011). A shift was also observed of the band connected with antiparallel-b-sheets in the region 1675–1695 cm 1 which indicates the formation of new H-bonds in b-sheet structures. These bonds can form between side chain amino acids, or between side chain amino acids and polysaccharides molecules. The tyrosine and tryptophan residues rather did not participate in the formation of those bonds due to an increase in both the tyrosine doublet value and the intensity of TRP band. Moreover, changes in the conformation of disulphide bridges (increase in the amount of S–S in conformations t-g-g and t-g-t) after a 9%-addition of the FLX could indicate that these new bonds form between side chain amino acids which leads to folding of the gluten proteins. In the case of carob fibre the shift was observed in the opposite way (from 1682 to 1678 cm 1). This suggests cleavage of part of the H-bonds and it is in agreement with the tyrosine doublet values and intensity of the TRP band. A 9%-addition of the CAR fibre also caused an increase in the amount of the S–S bonds in the stable conformation (g-g-g) and cleavage of intrachain disulphide bridges (absence of the band at 516 cm 1 as shown in Table 2). 5. Conclusions Raman spectroscopy revealed that the addition of dietary fibre preparations induced changes in the structure of gluten proteins. Direct mixing of commercially available gluten proteins with fibre preparations shows that all preparations, regardless of their origin, caused a decrease in a-helix with a simultaneous increase in antiparallel-b-sheets. This suggests that the a-helices combine through hydrogen bonds in antiparallel-b-sheets. This phenomenon may be connected with aggregation or abnormal folding of the protein complexes. Furthermore, the results indicate presence of the same compound in all preparations (cellulose) which induces this change. Moreover, the other changes observed in the gluten structure depends on the fibres origin and their chemical composition. This could suggest an interaction of pectic
93
polysaccharides with gluten proteins. Comparison of the present studies with previous studies show that starch protects the gluten proteins against undesirable changes concerning a-helical structures.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.foodchem.2015. 07.132. References Berti, C., Dolfini, E., & Forlani, F. (2002). Effects on coeliac activity of gluten proteins modified by chemical deamination. Polish Journal of Food and Nutrition Science, 11(52), 135–137. Berti, C., Roncoroni, L., Falini, M. L., Caramanico, R., Dolfini, E., Bardella, M. T., et al. (2007). Celiac-related properties of chemically and enzymatically modified gluten proteins. Journal of Agricultural and Food Chemistry, 55, 2482–2488. Bürk, K., Melms, A., Schulz, J. B., & Dichgans, J. (2001). Effectiveness of intravenous immunoglobulin therapy in cerebellar ataxia associated with gluten sensitivity. Annals of Neurology, 50, 827–828. Carey, P. R. (1982). Protein conformation from Raman and resonance 400 Raman spectra. In P. R. Carey (Ed.), Biochemical applications of Raman and resonance raman spectroscopies (pp. 65–116). New York: Academic Press. Correa, M. J., Ferrer, E., Anon, M. C., & Ferrero, C. (2014). Interaction of modified celluloses and pectins with gluten proteins. Food Hydrocolloids, 35, 91–99. Ferrer, E. G., Gomez, A. V., Anon, M. C., & Puppo, M. C. (2011). Structural changes in gluten protein structure after addition of emulsifier. A Raman spectroscopy study. Spectrochimica Acta A, 79, 278–281. Gomez, A. V., Ferrer, E. G., Anon, M. C., & Puppo, M. C. (2013). Changes in secondary structure of gluten proteins due to emulsifiers. Journal of Molecular Structure, 1033, 51–58. Lam, H. S., Proctor, A., & Meullenet, J.-F. (2001). Free fatty acid formation and lipid oxidation on milled rice. Journal of the American Oil Chemists’ Society, 78, 1271–1275. Leszczyn´ska, J., Ła˛cka, A., Pytasz, U., Szemraj, J., Lukamowicz, J., & Lewin´ski, A. (2002). The effect of proteolysis on the gliadin immunogenicity. Polish Journal of Food Nutrition Sciences, 11/52, 145–148. Linlaud, N., Ferrer, E., Puppo, M. C., & Ferrero, C. (2011). Hydrocolloid interaction with water, protein, and starch in wheat dough. Journal of Agricultural and Food Chemistry, 59, 713–719. Mangavel, C., Barbot, J., Popineau, Y., & Gueguen, J. (2001). Evolution of wheat gliadins conformation during film formation: A Fourier transform infrared study. Journal of Agricultural and Food Chemistry, 49, 867–872. Mis´, A. (2011). Interpretation of mechanical spectra of carob fibre and oatwholemeal-enriched wheat dough using non-linear regression models. Journal of Food Engineering, 102, 369–379. Mis´, A., & Dziki, D. (2013). Extensograph curve profile model used for characterising the impact of dietary fibre on wheat dough. Journal of Cereal Science, 57, 471–479. Nakamura, R., Nakamura, R., Sakai, S., Adachi, R., Hachisuka, A., & Urisu, A. (2013). Tissue transglutaminase generates deaminated epitopes on gluten, increasing reactivity with hydrolyzed wheat protein-sensitized IgE. Journal of Allergy and Clinical Immunology, 132, 1436 e4–1438 e4. Nawrocka, A., Szyman´ska-Chargot, M., Mis´, A., Ptaszyn´ska, A. A., Kowalski, R., Was´ko, P., et al. (2015). Influence of dietary fibre on gluten proteins structure – A study on model flour with application of FT-Raman spectroscopy. Journal of Raman Spectroscopy, 46, 309–316. Peressini, D., & Sensidoni, A. (2009). Effect of soluble dietary fibre addition on rheological and breadmaking properties of wheat doughs. Journal of Cereal Science, 49, 190–201. _ Pijanowski, E., Mrozewski, S., Horubała, A., & Jarczyk, A. (1973). Technology of fruit and vegetable products. Warsaw: PWRiL. Popineau, Y., Bonenfant, S., Cornec, M., & Pezolet, M. (1994). A study by infrared spectroscopy of the conformations of gluten proteins differing in their gliadin and glutenin composition. Journal of Cereal Science, 20, 15–22. Qiu, C., Sun, W., Cui, C., & Zhao, M. (2013). Effect of citric acid deamination on in vitro digestibility and antioxidant properties of wheat gluten. Food Chemistry, 141, 2772–2778. Secundo, F., & Guerieri, N. (2005). ATR-FR/IR study on the interactions between gliadins and dextrin and their effect on protein secondary structure. Journal of Agricultural and Food Chemistry, 53, 1757–1764. Shewry, P. R., Halford, N. G., Belton, P. S., & Tatham, A. S. (2002). The structure and properties of gluten: An elastic protein from wheat grain. Philosophical Transactions. The Royal Society, 357, 133–142. Siamwiza, M. N., Lord, R. C., & Chen, M. C. (1975). Interpretation of the doublet at 850 and 830 cm 1 in the Raman spectra of tyrosyl residues in proteins and certain model compounds. Biochemistry, 14, 4870–4876. Sivam, A. S., Sun-Waterhouse, D., Perera, C. O., & Waterhouse, G. I. N. (2012). Exploring the interactions between blackcurrant polyphenols, pectin and wheat
94
A. Nawrocka et al. / Food Chemistry 194 (2016) 86–94
biopolymers in model breads: A FTIR and HPLC investigation. Food Chemistry, 131, 802–810. Sivam, A. S., Sun-Waterhouse, D., Perera, C. O., & Waterhouse, G. I. N. (2013). Application pf FT-IR and Raman spectroscopy for the study of biopolymers in breads fortified with fibre and polyphenols. Food Research International, 50, 574–585. Sivam, A. S., Sun-Waterhouse, D., Quek, S. Y., & Perera, C. O. (2010). Properties of bread dough with added fibre polysaccharides and phenolic antioxidants: A review. Journal of Food Science, 75, R163–R174. Skulmowski, J. (1974). Methods for determination of fodder composition and quality. Warsaw: PWRiL. Sugeta, H. (1975). Normal vibrations and molecular conformations of dialkyl disulphides. Spectrochimica Acta A, 31, 1729–1737. Sunde, M., Serpell, L. C., Bartlam, M., Fraser, P. E., Pepys, M. B., & Blake, C. C. F. (1997). Common core structure of amyloid fibrils by synchrotron X-ray diffraction. Journal of Molecular Biology, 273, 729–739.
Taddei, P., Zanna, N., & Tozzi, S. (2013). Raman characterization of the interactions between gliadins and anthocyanidins. Journal of Raman Spectroscopy, 44, 1435–1439. Tozzi, S., Zanna, N., & Taddei, P. (2013). Study on the interaction between gliadins and a coumarin as molecular model system of the gliadins-anthocyanidins complexes. Food Chemistry, 141, 3586–3597. Waga, J. (2004). Structure and allergenicity of wheat gluten proteins – A review. Polish Journal of Food and Nutrition Sciences, 13(54), 327–338. Waga, J., Ka˛czkowski, J., & Zientarski, J. (2003). Influence of thioredoxin-h on flour baking properties and gliadin immunoreactivity in spelt and common wheat genotypes. Polish Journal of Food and Nutrition Sciences, 12(53), 13–16. Wellner, N., Belton, P. S., & Tatham, A. S. (1996). Fourier transform IR spectroscopic study of hydration-induced structure changes in the solid state of x-gliadins. Biochemical Journal, 319, 741–747. Wieser, H. (2007). Chemistry of gluten proteins. Food Microbiology, 24, 115–119.