Influence of Protein Content on Durum Wheat Gluten Strength Determined by the SDS Sedimentation Test and by Other Methods

Transcription

1 RESEARCH Peer-Reviewed Paper Influence of Protein Content on Durum Wheat Gluten Strength Determined by the SDS Sedimentation Test and by Other Methods Raimondo E. Cubadda, 1 Marina Carcea, 2 Emanuele Marconi, 3 and Maria C. Trivisonno 1 ABSTRACT Gluten strength in semolina is mainly responsible for pasta cooking quality, and it can be measured by different tests (micromixograph, viscoelastograph, farinograph, alveograph, gluten index, SDS sedimentation) as well as by manual sensory evaluation. A systematic study was performed on the influence of protein content on test results by determining a series of gluten quality values (alveograph W, SDS sedimentation value, gluten index) on semolinas obtained from an extraction and reconstitution experiment where four protein levels (namely 10, 11, 12, and 13% protein at 14% m.b.) were reconstituted from six samples belonging to three different cultivars possessing different gluten strengths. Furthermore, gluten quality was determined by means of several tests on 70 grain samples (11 cultivars) to study the simple correlations between the obtained values. The SDS sedimentation and the W alveograph values were clearly influenced by protein content and corrected values are proposed. The corrected SDS value was not influenced by other variables, and it was strongly correlated with the manual sensory values and the gluten index values. A classification of gluten strength based on corrected SDS values is proposed, which could be conveniently used for screening purposes in breeding programs or for rapid quality evaluation in the durum wheat industry. INTRODUCTION Cooking quality is the most important attribute in pasta for consumers. Several studies (3,5,8,18,21,22) have demonstrated that the capacity of durum wheat semolina to form a dough that is ideal for manufacturing pasta is mainly due to its gluten rheological characteristics, commonly known as gluten strength or gluten quality. In practice, pasta manufacturers require semolina with a strong gluten and a high protein content in order to have a good end-product. Consequently, gluten strength has long been considered by breeders to be an important qualitative parameter during 1 Molise Innovazione Scientific and Technological Park, Via De Sanctis, Campobasso, s.n.c , Italy. 2 National Institute for Research on Food and Nutrition, Via Ardeatina, Rome 00178, Italy. 3 DISTAAM, University of Molise, Campobasso, Via De Sanctis 46, 86100, Italy. doi: / CFW AACC International, Inc. the genetic improvement of new durum wheat varieties. Several methods, based either on sensory evaluation or on physical dough properties (mixograph, viscoelastograph, farinograph, alveograph, extensigraph, gluten index) (2,4,7,9,11,12,15,16,19,23) and physicochemical tests (SDS sedimentation test and swelling index of glutenin test) (6,25), have been proposed in order to evaluate gluten strength in whole meal and/or semolina. Amongst all the above methods, the SDS sedimentation test and the gluten index methods are the most commonly used, both for fast screening in durum wheat breeding programs and for rapid analysis in quality control laboratories, because of the small amount of sample required, their simplicity of execution, and their quickness. In many pasta-making industries, a modified alveograph method (24) is also used. However, SDS and alveographic tests may have some drawbacks, mostly due to the influence of protein and composition and/or content, independent of gluten strength, on analytical results (4,13,14,20). On the other hand, sensory (manual) evaluation is a valid but subjective method, and it is very difficult to standardize even though it has extensively been used in the past for durum wheat screening at the industrial level (17). Thus, the objective of this work was first to assess the influence of protein content on SDS sedimentation and W alveographic values in comparison with manual gluten and gluten index values, and secondly, to quantify the eventual bias and to propose the possibility of gluten strength classes based on the SDS sedimentation test values. In order to achieve that, an extraction and reconstitution study was performed where gluten and starch were extracted from six samples of three varieties possessing different gluten strength. Four classes of protein content (10, 11, 12, and 13% at 14% m.b.) were reconstituted for each sample by adding the sample s own gluten or starch to increase or decrease its original protein quantity. Furthermore, a group of 70 durum wheat samples was selected to study in real samples the correlation between protein content and gluten strength determined by all of the above mentioned methods. MATERIALS AND METHODS Six samples (SI-1, SI-2, SI-3, OF-1, OF-2, and APP) of three varieties (Appulo, Ofanto, and Simeto) and 70 samples of 11 varieties (Adamello, Appio, Appulo, Arcangelo, Brindur, Duilio, Grazia, Ofanto, Simeto, Valnova, and Vitron), all coming from different Italian areas, were used for the extraction and reconstitution experiments and for the correlation studies, respectively. The semolina was always obtained from cleaned durum wheat kernels, conditioned to a water content of 16.5% and milled in a MLU 202 mill (Bühler, Uzwill, Switzerland) equipped with three break and three CEREAL FOODS WORLD / 273

2 reduction rolls and six steel screens, fitted with a small scale purifier, following AACC Intl. methods 26-10A and (1). Semolina yield (ash <0.90% dm) was about 70% and granulation was within the limits of Italian law (25% maximum passed through a 180 μm mesh sieve). For the protein determination, SDS sedimentation test, and the manual gluten quality tests where wholemeal samples were analyzed, the grain samples were ground in a Cyclotec 1093 sample mill equipped with a 1-mm sieve (Tecator, International PBI SpA, Milan, Italy). For the determination of the gluten index, wholemeal flours were obtained by milling grains in a laboratory mill (model 3100, Falling Number AB, Huddinge, Sweden) equipped with a 0.8-mm sieve. The different mills were chosen depending on the standard method specifications or laboratory experience. Extraction of Gluten and Starch Gluten and starch were extracted according to Cubadda et al. (3). In particular, semolina (2 kg) obtained from the above mentioned six samples (SI-1, SI-2, SI-3, OF-1, OF-2, and APP) was mixed in a dough mixer (prototype built for this study by NAMAD, Roma, Italy) with 1.2 L of water at room temperature for 4 min at 95 rpm and 6 min at 140 rpm. The obtained dough was allowed to rest for 15 min at room temperature. The dough was then poured over the sieve (pore size 420 µm) of the gluten washing machine (prototype as above). After 30 min of washing, the gluten fraction was recovered from the sieve and dried under vacuum at <50 C in a special drier (NAMAD, Roma, Italy). The filtrate from the first 10 min was collected and decanted in a refrigerated room at 4 5 C overnight. The sediment consisted of a yellow-brown layer of water soluble components (pentosans, albumins, globulins, etc.) with a starch layer underneath. The top layer was scraped off and discarded, and the white bottom layer was resuspended in distilled water and centrifuged (1800 g, 10 min, room temperature). The sediment was collected as starch fraction and dried in the same manner as the gluten fraction mentioned above. The dried gluten contained 80 82% protein (d.m.), whereas the protein content of the dried starch was less than 0.5%. The dried gluten and starch (moisture around 4 5%) were then milled in a laboratory mill to pass a μm mesh size, sealed in plastic bags, and kept at room temperature before use. Gluten vitality was checked with the farinograph test according to AACC International standard method (1). Reconstitution of Gluten Protein Classes A quantity of dried gluten or starch was added to the original semolina in order to obtain for each sample (SI-1, SI-2, SI-3, OF-1, OF-2, and APP) a series of subsamples, each possessing the same protein levels (namely 10, 11, 12, and 13%). This amount was calculated considering a final moisture value in the mixture of 14%. The predicted protein content was confirmed by analyzing the modified semolina according to the above mentioned method. Analytical and Technological Tests Grains (wholemeals) and semolina samples were analyzed by standard ICC procedures for moisture (standard method no. 110/1) and total protein (Nx5, 70, standard method no. 105/2) (10). The SDS sedimentation value was obtained according to standard AACC International method using a 3% SDS solution (1), whereas the gluten index was determined according to the standard method no. 158 (10). Sensorial (manual) gluten quality was performed according to the procedure reported by D Egidio et al. (5). In particular, 20 g of wholemeal flour or semolina were manually washed with a 2% NaCl solution buffered at ph 6.8. The quality of gluten so extracted was evaluated by handling and classifying the degree of elasticity and extensibility with scores ranging from 0 to 10 as follows: 0 non existent; 1 not cohesive; 2 fragile, little cohesiveness, very sticky; 3 fragile, not elastic, sticky; 4 long, very extensible, not elastic; 5 long, very extensible, little elasticity; 6 medium, extensible, little elasticity; 7 medium, slightly extensible, little elasticity; 8 short, slightly extensible, elastic; 9 10 short, tough, elastic. The rheological properties of the dough were determined with the alveograph by the method of Chopin (ICC standard method no. 121), but the dough was mixed for 4 min and after a rest of 18 min mixed again for 4 min (10). Statistics Data were analyzed and compared using, where possible, the average of several determinations, the standard deviations, and linear correlation coefficients. RESULTS AND DISCUSSION Table I lists the quality parameters relative to ash, protein content, and protein quality (expressed by gluten index, W and P/L alveograph, and SDS sedimentation test value) of grains (wholemeal) and semolina belonging to the six samples used in the extraction and reconstitution study. Analyses were performed on either wholemeal or semolina depending on common practice, i.e., which material is more frequently used in practice because it gives more repeatable and reproducible results. If we focus our attention on protein quantity and quality, we can see that protein (%d.b.) ranged from 10.6 to 14.4 in whole kernels and from 10.1 to 13.3 in semolina. The protein content in semolina clearly reflects the original protein content in the kernels, and the gluten index values determined on wholemeals were inferior to those determined on semolina due to the disturbing influence of bran particles. This was particularly evident for samples with high gluten index values (Table I). If we consider the gluten index and the related classification of gluten quality re- Table I. Protein, ash content, and gluten quality parameters of durum wheat grains and semolina samples used for fractionation and reconstitution experiments a Alveograph b SDS Protein Ash Gluten index sedimentation Samples (% d.b.) (% d.b.) (%) W P/L value (ml) SI-1 wheat SI-1 semolina SI-2 wheat SI-2 semolina SI-3 wheat SI-3 semolina OF-1 wheat OF-1 semolina OF-2 wheat OF-2 semolina APP- wheat APP- semolina a The value for proteins is the average of four determinations. Ash, gluten index, and SDS sedimentation test were performed in duplicate and the average reported. b Alveograph indices: W-dough strength (x 10 4 J); L extensibility (mm); P tenacity (mm); P/L tenacity-to-extensibility ratio. 274 / SEPTEMBER-OCTOBER 2007, VOL. 52, NO. 5

3 ported by Cubadda et al. (4) for both wholemeal and semolina, we can say that samples SI-1 and SI-2 definitely belong to the good or very good gluten class, samples SI-3 and OF-1 span from the good to the promising gluten class, and samples OF-2 and APP are definitely in the inadequate gluten class. The SDS sedimentation test and W alveograph also showed a range of values from 16 to 47 and from 48 to 120, respectively, and the alveograph P/L value showed very different gluten tenacities decreasing from 4.17 to The protein quality data produced with different methods clearly indicate that the chosen samples span a wide range of values, and they can therefore be considered a population representative of different gluten qualities. The extraction and reconstitution experiment was only performed on semolina because semolina is a more homogeneous medium, analyses performed on semolina are more reproducible and repeatable, and semolina is the raw material of choice for pasta production. The gluten index, W alveograph, and SDS sedimentation values of the modified semolina samples (obtained from samples described in Table I) with protein levels varying from 10 to 13 (14% m.b.) are shown in Table II. From these data, it is clear that, in general, protein content has an influence on SDS and W alveograph values because, given the same original sample but different protein contents, values obtained within the same test, compared between them, span above what is considered for the SDS sedimentation test as an acceptable difference between two duplicates according to the standard method (1), and for the W values, they span above the acceptable difference to classify two samples as different according to experience. On the contrary, insignificant differences were observed when the same exercise was repeated for the gluten index values. It was also interesting to note that the observed phenomena were independent of the gluten quality. All of the samples, which, as demonstrated (Tables I and II), possessed different protein qualities, had higher W alveograph and SDS sedimentation values at higher protein levels, confirming our previous findings (4). In an attempt to reduce and possibly remove the influence of protein content on the above mentioned protein quality tests, a correction was introduced dividing the obtained SDS sedimentation test or W alveograph values by the corresponding protein content and multiplying the result by ten. The corrected SDS sedimentation test and W alveograph values are also reported in Table II. It can be clearly seen that such corrections made the SDS value of the samples originating from the same semolina but possessing different protein contents to look similar to those obtained for the gluten index. This phenomenon was also reported for spelt wheat by some of the present authors in another work (20). Similarly, the corrected W alveograph values clearly showed less dramatic differences between samples belonging to the same gluten quality group. We then thought it worthwhile to study the same data of Table II in a linear correlation matrix (Table III), where we could clearly see that the correlation between gluten index and W alveograph and SDS sedimentation test values improved, going from 0.86 to 0.93 (p 0.01) and from 0.86 to 0.94 (p 0.01), respectively, when corrected values were introduced in the calculations. As already mentioned, a second experiment was carried out with the purpose of further confirming the influence of protein content on the SDS sedimentation test value and to ascertain the possibility to develop, as for other tests, gluten quality classes based on the SDS sedimentation values which could serve the practical purpose of giving a clearer indication to breeders, millers, and pasta makers the technological potential of any given durum wheat sample. For this experiment, 70 samples coming from 11 cultivars were analyzed for their protein content, manual gluten quality, gluten index, and SDS sedimentation test value. They were either analyzed as wholemeal or semolina, depending on current Italian practice. In particular, for each grain sample, protein content and the SDS sedimentation test value were determined on wholemeal, whereas the manual gluten quality and the gluten index were determined on semolina. The mean and the range of all values for the mentioned parameters are reported in Table IV. As can be seen, the samples covered a large range of values for each of the quality tests. In fact, protein contents ranged between 9.0% and 14.8%, manual gluten score between 2 and 9, gluten index between 2 and 97, and the SDS sedimentation test value between 20 and 65. Also, a corrected SDS value was calculated for each sample, and it ranged between 18 and 49. Simple correlation coefficients between pairs of quality tests performed on the 70 durum wheat samples are reported in Table II. Gluten index, W Alveograph and corrected W, SDS sedimentation test value, and corrected SDS sedimentation test value of reconstituted durum wheat semolina samples with different gluten strength and fixed protein contents (10, 11, 12, and 13% at 14% m.b.) Corrected Gluten W W Corrected SDS value SDS Samples a index b alveograph c W (ml) b value SI SI SI SI SI SI SI SI SI SI SI SI OF OF OF OF OF OF OF OF APP APP APP APP a SI Simeto; OF Ofanto; APP Appulo. b Average of duplicate determinations. Corrected values were obtained by dividing the regular value by protein content and multiplying it by 10. c W dough strength (x 10 4 J). CEREAL FOODS WORLD / 275

4 Table V. Confirming the results illustrated before (Table III) for the extraction and reconstitution study, protein content was positively correlated with the SDS sedimentation test value (SDST) (r = 0.30; P 0.05). On the contrary, no correlation was observed for either the corrected SDS sedimentation test value (CSDST) or the manual gluten quality value (MGQ) and the gluten index value (GI). MGQ was highly correlated with GI (r = 0.97; P 0.01) and with the SDST and the CSDST (r = 0.87 and r = 0.95, respectively). However, both the MGQ and the GI correlated better with the CSDST (r = 0.95 and r = 0.94, respectively) than with the SDST (r = 0.87 and r = 0.86, respectively). The correlation between SDST and CSDST was 0.91 (P 0.01). All of the evidence collected in our studies suggests that correcting the SDS sedimentation test results allowed us to eliminate the bias originated by the influence of protein content, and therefore it made the CSDST able to predict gluten strength similarly to the manual evaluation or the gluten index. CONCLUSIONS The present investigation allowed for the definitive identification of the influence of protein content on the evaluation of gluten strength by the SDS sedimentation test and by the W alveograph. At the same time, it allowed us to demonstrate the usefulness of the determination of gluten quality by the SDS sedimentation test when a corrected value is used. This corrected value is devoid of the influence of other variables such as protein content, and it is strongly correlated with the manually determined gluten quality and with the gluten index, both of which are considered suitable evaluation methods for estimating the technological performance of durum wheat in pasta making. The corrected SDS values for the 70 samples in this study spanned a range from 18 to 49, which was wide enough to allow us to build gluten strength classes depending on SDS values. These classes were based on the comparison with manual gluten quality and gluten index classes. Considering the advantages of using the SDS sedimentation test compared to other methods (small amount of sample required, simplicity of execution, no expensive piece of machinery required), we thought that, for practical purposes, it would be interesting to adopt the proposed classification for durum wheat gluten quality based on corrected SDS sedimentation values. Quality Classes in Durum Wheat Based on Corrected SDS Sedimentation Test Values This classification (Table VI) allows us to better differentiate and more efficiently define gluten strength in durum wheat, and it can be particularly useful for 1) the screening of experimental lines in breeding programs 2) and for rapid quality evaluation in the grain trading and milling industries. References 1. AACC International Method 26-10A, Experimental milling; Method 26-41, Experimental milling-bühler method for durum wheat; Method 56-70, SDS sedimentation test; Method Farinograph test for vital wheat gluten. Approved Methods of the American Association of Cereal Chemists, 10th ed. CD-ROM. AACC International, St. Paul, MN, Table III. Simple correlation coefficients between quality tests for 24 durum wheat semolina reconstituted samples (reported in Table II) with different protein contents Variables GI WA CWA SDST CSDST Gluten index (GI) a 0.93 a 0.86 a 0.94 a W alveograph (WA) a 0.96 a 0.88 a Corrected w alveograph (CWA) a 0.95 a SDS sedimentation test (SDST) a Corrected SDS (CSDST) 1.00 a p Table V. Simple correlations between quality tests for a group of 70 durum wheat samples Variable PR MGQ SDST CSDST GI Protein content (PR) a Manual gluten quality (MGQ) b 0.95 b 0.97 b SDS sedimentation (SDST) b 0.86 b Corrected SDS test (CSDST) b Gluten index (GI) 1.00 a p b p Bruinsma, B. L., Anderson, P. D., and Rubenthaler, G. L. Rapid method to determine quality of wheat with the mixograph. Cereal Chem. 55: , Cubadda, R., Carcea, M., Marconi, E., and Trivisonno, M. C. Influence of gluten proteins and drying temperature on the cooking quality of durum wheat pasta. Cereal Chem. 84:48-55, Cubadda, R., Carcea, M., and Pasqui, L. A. Suitability of the gluten index method for assessing gluten strength in durum wheat and semolina. Cereal Foods World. 37: , D Egidio, M. G., Mariani, B. M., Nardi, S., Novaro, P., and Cubadda, R. Chemical and technological variables and their relationships: A predictive equation for pasta cooking quality. Cereal Chem. 67: , Dexter, J. E., Matsuo, R. R., Kosmolak, F. G., Leisle, P., and Marchylo, B. A. The suitability of the SDS-sedimentation test for assessing gluten strength in durum wheat. Can. J. Plant Sci. 60:25-29, Edwards, N. M., Dexter, J. E., Scanlon, M. G., and Cenkowski, S. Relationship of creeprecovery and dynamic oscillatory measurements to durum wheat physical dough properties. Cereal Chem. 76: , Feillet, P., and Dexter, J. E. Quality requirements of durum wheat for semolina milling and pasta production. Pages in: Pasta and Noodle Technology. J. E. Kruger, R. R. Matsuo, and J. W. Dick, eds. AACC International, St. Paul, MN, Gaines, C. S., Frégeau Reid, J., Vander Kant, C., and Morris, C. F. Comparison of methods for gluten strength assessment. Cereal Chem. 83: , ICC International Methods No 105/2, 110/1, 121, 158. Standard Methods of the International Association for Cereal Science and Technology. The Association, Vienna, Austria, Table IV. Summary of quality data for 70 durum wheat samples belonging to 11 cultivars a Quality test Mean Range Protein (% d.b., N X 5.70) Manual gluten quality score Gluten index SDS sedimentation test value (ml) Corrected SDS (ml) a For every parameter and for each sample the average was calculated out of duplicates. Table VI. Quality classes in durum wheat based on corrected SDS sedimentation test values Gluten quality Corrected SDS class sedimentation test values a Inadequate <20 Poor or very poor Average or above average Good or very good Excellent >44 a SDS sedimentation values obtained according to standard AACC International method (24) divided by protein content and multiplied by / SEPTEMBER-OCTOBER 2007, VOL. 52, NO. 5

5 11. Irvine, G. N., Bradley, J. W., and Martin, G. C. A farinograph technique for macaroni doughs. Cereal Chem. 38: , Kovacs, M. I. P., Dahlke, G., and Noll, J. S. Gluten viscoelasticity: Its usefulness in the Canadian durum wheat breeding program. J. Cereal Sci. 19: , Kovacs, M. I. P., Howes, N. K., Leisle, D., and Skerritt, J. H. The effect of high M r glutenin subunit composition on the results from tests used to predict durum wheat quality. J. Cereal Sci. 18:43-51, Kovacs, M. I. P., Howes, N. K., Leisle, D., and Zawistowski, J. Effect of two different low molecular weight glutenin subunits on durum wheat pasta quality parameters. Cereal Chem. 72:85-87, Kovacs, M. I. P., Noll, J. S., Dahlke, G., and Leisle, D. Pasta viscoelasticity: Its usefulness in the Canadian durum wheat breeding program. J. Cereal Sci. 22: , Kovacs, M. I. P., Poste, L. M., Butler, G., Woods, S. M., Leisle, D., Noll, J. S., and Dahlke, G. Durum wheat quality: Comparison of chemical and rheological screening tests with sensory analysis. J. Cereal Sci. 25:65-75, Landi, A. Quality evaluation of Italian durum wheat varieties. Pages in: The future of cereals for human feeding and development of biotechnological research. Chamber of Commerce, Foggia, Italy, Malcolmson, L. J., Matsuo, R. R., and Balshaw, R. Textural optimization of spaghetti using response surface methodology: Effects of drying temperature and durum protein level. Cereal Chem. 70: , Marchylo, B. A., Dexter, J. E., Clarke, F. R., Clarke, J. M., and Preston, K. R. Relationships among bread-making quality, gluten strength, physical dough properties and pasta cooking quality for some Canadian durum wheat genotypes. Canadian J. Plant Sci. 81: , Marconi, E., Carcea, M., Graziano M., and Cubadda, R. Kernel properties and pastamaking quality of five European spelt wheat (Triticum spelta L.) cultivars. Cereal Chem. 76:25-29, Matsuo, R. R., Dexter, J. E., Kosmolak, F. G., and Leisle, D. Statistical evaluation of tests for assessing spaghetti-making quality of durum wheat. Cereal Chem. 59: , Novaro, P., D Egidio, M. G., Mariani, B. M., and Nardi, S. Combined effects of protein content and high temperature drying systems on pasta cooking quality. Cereal Chem. 70: , Rao, V. K., Mulvaney, S. J., Dexter, J. E., Edwards, N. M., and Peressini, D. Stress-relaxation properties of mixograph semolina-water doughs from durum wheat cultivars of variable strength in relation to mixing characteristics, bread- and pasta-making performance. J. Cereal Sci. 34: , Walle, M., and Trentesaux, E. Study of a practical method with Chopin alveograph for the evaluation of macaroni making capacity of durum wheat and semolina. Tec. Molitoria. 31:917, Wang, C., and Kovacs, M. I. P. Swelling index of glutenin test for prediction of durum wheat quality. Cereal Chem. 79: , An advertisement appeared here in the printed version of the journal. CEREAL FOODS WORLD / 277