Wheat Proteins Extracted from Flour and Batter with Aqueous Ethanol at Subambient Temperatures

Cereal Chem. 84(5):497-501 Contact of wheat flour with aqueous ethanol may enrich protein by starch displacement or deplete protein by extraction depending on I) extraction conditions and 2) the form of the substrate. Extraction at subambient temperatures has not been described for specific gliadins for either dry flour with the protein in native configurations or for wet, developed hatter or dough. This limits the ability to interpret technologies such as the cold-ethanol method. Here, we describe specific albumin and gliadin composition of flour extracts by capillary zone electrophoresis CZE in 0-100% (v/v) ethanol from -12 to 22°C. Extraction was reduced for albumin and gliadin protein as the temperature was reduced and the concentration range for extraction narrowed. Extraction dropped precipitously between 0 and -7°C for both albumins and gliadins. Electrophoretically defined gliadins extracted in constant proportion at 22°C and 3080%(v/v) ethanol, but at lower temperature, the u-gliadins were enriched and 13-gliadins depleted in the 30-55% (v/v) range. For extracts from wheat flour batter, depletion of a and 13 and enrichment of y relative to the dry flour contact suggested that the electrophoretically slow migrating yand (1)-proteins are less well incorporated to the dough matrix than electrophoretically fast migrating (x and 13 types. Aqueous ethanol solutions have been exploited for large-scale refining of substrates as diverse as human blood, wood pulp, and oilseeds (Cohn et al 1946; Oncley et al 1949; Rao et al 1955; Rao and Arnold 1956a,b, 1957, 1958; Ni and Van Heiningen 1996). Substrates such as cereal grains, cottonseed, plywood, and plant straw also may be refined using ethanol, but this has been limited to laboratory or pilot scale (Hron and Koltun 1984; Hojilla-Evangelista et al 1992a,b; Chang et al 1995; Papatheofanous et al 1995; Robertson and Cao 1998a,b; Feng et al 2002; Miller et al 2002) The attractiveness and utility of aqueous ethanol as a processing/separation agent arises from its unique physical and thermodynamic properties. its low heat capacity, latent heats of crystallization and evaporation, boiling temperature, and density translate to reduced energy usage, fewer undesirable heat-induced property changes, and improved settling rates for crop components (Robertson and Otis 2005). However, management of the concentration, temperature, pH, or ionic strength enhances or inhibits the ability to extract water, oil, lignin, or proteins. Management of these factors implies the technical ability to fully or partially remove and recover soluble and suspended substances so that the ethanol may be recycled. Adding to the interest in ethanol facilitated refining is the emergence of grain-based, ethanol-producing biorefineries. In these facilities, there is also a need for improved, energyefficient component fractionation to recover high-value lipids and proteins to bolster the economics of the enterprise. However, the use of aqueous ethanol to refine diverse biological materials creates the need to understand the selectivity for components in the biological substrate. This knowledge can help to identify potential refining options and product-platform opportunities. Research in our laboratory helps to define some issues related to ethanol use in refining. We have applied refrigerated or cold ethanol to the refining of wheat grain to produce starch and protein fractions as "platforms" for well-developed food markets as well as for emerging food and nonfood markets. In this practice, cold ethanol was applied to a mechanically developed batter while the batter was physically manipulated. The liquid ethanol solution separates starch and protein by extracting water and physically displacing starch from the protein matrix. We found that if the solution and the batter were refrigerated to -15 and 10°C, respecUnited States Department of Agriculture, Pacific West Area, Western Regional Research Center, 800 Buchanan Street, Albany, CA 94710. 2 Corresponding author. E-mail: [email protected] doi: 10.1 094/CCHEM-84-5-0497 This article is in the public domain and not copyrightable. It may be freely reprinted with customary crediting of the source. AACC International, Inc., 2007. lively, excellent separation and recovery were possible because the solubility of the gluten protein is low, as confirmed by total protein analysis. However, severe and extended mechanical action during the displacement step may cause selective loss of gliadinclass protein to solution. Furthermore, we found functional differences in the protein gluten fraction produced by the cold ethanol method (Robertson and Cao 2001, 2002, 2003, 2004). It is unknown whether the differences in extracted proteins orginate in the native molecular structures or in the hydrated structures created to affect the starch protein separation. For wheat gliadins, in particular. we note only the plethostatic or cloud-point determination of critical solubility of previously fractionated Osborne gliadins (Osborne 1907; Dill and Alsberg 1925). The plethostatic method defines the critical temperature at which turbidity initiates on cooling of previously heated and clear protein solutions. The method is best applied to and interpreted for single components in solution. However, wheat proteins do not fit this purity constraint so that the method may index the precipitation of only the least soluble gliadin subcomponent. Above the critical solubility temperature, all gliadin-class proteins were described as fully soluble. We also note reports of solubility of gliadins and whole gluten at 70% (v/v) ethanol and 20°C extraction (Meredith et al 1960a,b; Meredith 1965a,b,c; Robertson et al 1999) and gluten protein solubility at 22°C in ethanol solutions from 0 to 100% (vlv) (Robertson and Cao 2001, 2002, 2003, 2004). In the present study, we describe the dissolved flour proteins in extracts of native molecular matrices at ambient and subambient temperatures in water, ethanol, and ethanol-water solutions. A number of sophisticated methods have been developed for analysis of proteins in wheat, including reversed-phase HPLC, capillary electrophoresis, and a variety of gel-electrophoresis techniques (Shewiy and Lookhart 2003). Capillary zone electrophoresis (CZE) separates proteins by a voltage-driven migration in acidic buffer based on charge density. Highly charged, small diameter proteins have the greatest mobility in the analysis and are detected first. This automated electrophoretic method has been used successfully for the direct analysis of wheat gliadins and albumins in Osbornelike ethanol extracts employing a polymer-modified acidic buffer. Furthermore, capillary electrophoresis is known to be complementary to reversed-phase HPLC, which separates on the basis of surface hydrophobicity. Some single HPLC peaks have been resolved into multiple CZE peaks. The method requires miniscule amounts of buffer and no solvent, thereby maximizing safety, minimizing environmental hazard, and reducing material and waste disposal costs (Lookhart and Bean 1995a,b; Robertson and Cao 2004). Only filtration and centrifugation were needed to prepare Vol. 84, No. 5, 2007 497 ___ EL Total protein A x mm 5 10 [.] mm the chilled liquid extract samples for injection because the sample uncharged fluids would not be expected to affect the separation. Here we chilled and centrifuged protein solutions initially formed by extraction at room temperature and then analyzed the supernatant for soluble proteins. MATERIALS AND METHODS Treatment and mixing. Giusto Peak Performer flour (8 g) at 14.4% protein and 11.2% moisture was vortexed for 1 hr at ambient temperature with 29 mL of water, absolute ethanol, or aqueous ethanol at 5-90% in 50-mL centrifuge tubes. Samples were chilled to the desired temperature for the below-ambient temperature studies. Supernatant. A refrigerated centrifuge with temperature adjustment was used to separate the supernatant from the flour. After centrifugation for 10 min at 6,000 x g at the desired temperature, the supernatants were pipetted into I .5-mL conical centrifuge tubes and once again chilled and equilibrated to the desired temperature. Once the supernatant reached the desired temperature, samples were centrifuged at 14,000 rpm for 1 min and filtered using a 0.45-pm syringe filter before CE analysis. Capillary electrophoresis. Supernatant samples were analyzed by capillary electrophoresis (CI 602A, Agilent Technologies, Wilmington, DE) using uncoated fused-silica capillaries (Agilent) with 50pm diameter and 24.5-cm effective length. Bio-Rad (Hercules, CA) phosphate buffer (0. 1M, pH 2.5) containing a linear polymer modifier (hydroxypropyl, methyl-cellulose) was used in all separations (Bietz and Schmalzried 1995; Robertson and Cao 2004). Samples were filtered using a 0.45-pm syringe filter and were injected undiluted with an injection time of 2 sec at 35 mbar and separated at 40°C and 7 kV. Proteins were detected by 200 nm UV absorbance. RESULTS AND DISCUSSION Protein groups resolved by CZE from the 70% ethanol, 22°C extract of flour are shown in the electropherogram of Fig. 1. As expected, extraction and electrophoretic conditions yield separated wheat gliadins, albumins, and globulins. Electrophoretically defined peaks are labeled in order as a-, 3-, y-, and w-gliadin proteins. These are electrophoretically defined and do not exactly conespond to the genetic-type classification using the same nomenclature letters. For instance, some of the genetic 3-type gliadins elute in the y-electrophoretic gliadin range. (Kasarda et al 1987; Lookhart and Bean 1995b; Robertson and Cao 2004). We observed z25 individual gliadin peaks and 15-20 albumin/globulin peaks. By contrast, corn zein (a solvent-defined analog of the gliadin proteins) has been reported to have no more than five peaks by similar methods (Parris et al 1997). Above the CZE protein trace, we include a reference or "faux gel lane" that represents the data as a gray-scale plot (black for maximum observed height and white for no protein). By integrating all of the protein peaks, we see how the solvent composition and temperature affect the solubility of proteins extracted from the flour. As expected, most protein was soluble at 22°C and 60% (v/v) ethanol. However, no protein remained soluble at >90% ethanol at all temperatures and <5% of the maximum observed is soluble in the 0-15% (v/v) concentration range (Fig. 2). Below —7°C, most protein solubility is minimized, with the exception of small solubility in the 60-80% (v/v) ethanol range. All data are summarized in Fig. 3 (0-100% ethanol solutions and 22 to —12°C). The data in the faux lanes are globally heightnormalized and gray-scaled from white to black in direct proportion to height. The true black peak representing the most protein extracted is in the y-gliadin group for 60% (v/v), 15°C. In the analysis, peak patterns were matched by linear shifts of the data along the time axis or abscissa to align the peaks that were eluted earliest (albumins and globulins). The shift averaged 1-2% of the overall spectrum. When this was done for the albuminglobulin range, all downstream peaks in the gliadin range also aligned. We note four distinct areas on this summary: Region IA, where essentially no protein is soluble; Region IB, where no protein is soluble because the solvent is not liquid; Region 2, where gliadin proteins predominate, but all forms are present; and Region 3, where albumin/globulin proteins are the predominant form extracted. The boundary at which gliadin and AG proteins are equally extracted is shown in Fig. 4. This figure is annotated to describe the region of low ethanol concentration and or low temperature for which the albumins predominate. Integration of protein peak areas in the albumin-globulin range showed that for most of the subzero temperatures tested, the sol'ioumins lobulins gliadins pillary-zone electrophoresis of 70% ethanoll22°C solvent extracts it wheat hour. Bar indicates peak height as a shade of gray (most intense = black, least intense = white). 3 CEREAL CHEMISTRY 0 20 40 60 80 100 Ethanol concentration (% v/v) Fig. 2. Total solubilized proteins from wheat flour by integration of CZE output. ethanol concentration (vol %)