Comparison of microwave and conventional heating of a wheat starch-gluten-water model systemкод для вставкиСкачать
INFORMATION TO USERS This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer. The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely afreet reproduction. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book. Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. U n iv ersity M icrofilm s In te rn a tio n a l A Bell & H ow ell In fo rm atio n C o m p a n y 3 0 0 N o rth Z e e b R o a d , A n n A rbor. Ml 4 8 1 0 6 - 1 3 4 6 U S A 3 1 3 /7 6 1 - 4 7 0 0 8 0 0 /5 2 1 - 0 6 0 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. O rder N u m b e r 9 303395 C om parison of microwave and conventional h eating of a w heat starch-gluten-w ater m odel system Umbach, Sharon Lynn, Ph.D. University of Minnesota, 1992 UMI 300 N. ZeebRd. Ann Arbor, MI 48106 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Comparison of Microwave and Conventional Heating of a Wheat Starch-Gluten-Water Model System A Thesis Submitted to the Faculty of the Graduate School of the University of Minnesota By Sharon Lynn Umbach In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy October, 1992 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. U N I V E R S I T Y OF M I N N E S O T A This is to certify that I have examined this bound copy of a doctoral thesis by Sharon Lynn Umbach and have found that it is complete and satisfactory in all respects, and that any and all revisions required by the final examining committee have been made. F ugenia A. Oavi<; Name o f Faculty Adviser(s) Signature o f Faculty Adviser(s) O cto b er 0 6 . 1992 Date GRADUATE SCHOOL Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT The effects of conventional (CV) and microwave (MW) heating on a model wheat starch, vital wheat gluten system were studied. Three moisture contents (35%, 50%, and 65%) and up to five ratios of starch:gluten were studied. MW power was such that heating took place more quickly than during CV heating conditions. Temperature profiles during heating showed the importance of moisture content for curve shape and that all samples reached a high enough tem perature for starch gelatinization and gluten denaturation to occur. Scanning electron microscopy supported temperature data in that physical changes for starch had taken place. Texture was evaluated as the force required to compress the sample. Differences were found for heating method, moisture content, and sample composition. Moisture content had the major effect and was inversely related to force. In general, CV heated samples required more force to compress than the MW heated samples. The high starch samples at 50% moisture showed the opposite affect. Self-diffusion coefficients determined by nuclear magnetic resonance (NMR) showed that there was a redistribution of water between starch and gluten after heating for the faster more mobile water in the samples. Little difference was found between the two heating methods. The attenuation factor which was based on the dielectric constant and loss showed differences in the way the sample interacted with electromagnetic energy. The gluten was fractioned with dilute HC1 to give six groups of protein. Starch appeared to have an effect on how the gluten denatured. W hen the Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. fractions were characterized with SDS-PAGE it was found that each fraction contained a wide range of proteins with different molecular weights. Differences were found in the band patterns between the CV- and MW-heated samples. 13C NMR also was used, which provided information on the protein and carbohydrate component of the fractions showed differences between the CV and MW heated samples, especially for the carbohydrate component. From this study, it appears that differences between the CV and MW heated samples are due to the type of denaturation the gluten undergoes, which may be influenced by the presence of starch and the amount of water available. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGEMENTS I would like to thank all the people who have been a part of my life during the time I worked to complete this thesis. The people are spread over two continents in four countries and are far too numerous to mention. Some of them helped in a scientific manner to allow me to complete the research. But many more persons provided me w ith opportunities to learn nonsdentific skills, develop as a person, and most importanly to enjoy life! These skills will be invaluable to me for the rest of my life. The process of obtaining a Ph.D. is a long tortuous path that without peers, family, and colleagues around to make it bearable, it would be a very long and lonely journey. I will remember you all. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CONTENTS LIST OF TABLES.....................................................................................................vi LIST OF FIGURES...................................................................................................vii Chapter 1 INTRODUCTION...............................................................................1 Chapter 2 REVIEW OF THE LITERATURE................................................... 4 2.1 Heating...................................................................................................4 2.1.1 Conventional Heating..........................................................4 2.1.2 Microwave Heating...............................................................5 18.104.22.168 Dielectric Properties.................................................6 2.1.3 Events During Baking.............................................................8 2.1.4 Patents about Microwave Heated Cereal Products............9 2.2 Cereal Products......................................................................................11 2.2.1 Composition of Dry Ingredients...........................................11 22.214.171.124 Gluten.........................................................................11 126.96.36.199 Starch......................................................................... 13 188.8.131.52 Interaction of Gluten, Starch, and W ater............. 13 184.108.40.206 Role of Gluten in Dough and Bread Formation............................................................................... 14 2.2.2 Appearance.............................................................................. 15 2.2.3 Texture.....................................................................................18 2.3 Water-Macromolecular Interactions................................................ 20 2.3.1 Proton Nuclear Magnetic Resonance................................. 21 2.3.2 Water-Macromolecular Interactions as Studied by N M R .................................................................................................. 22 2.3.3 Water Behavior as Examined by Dielectric Properties.......................................................................................... 24 2.4 Examination of Gluten........................................................................ 25 2.4.1 Fractionation and Electrophoresis of Unheated Gluten.................................................................................................28 2.4.2 l 3C NMR of Unheated Gluten..............................................34 2.4.3 Effects of Heating....................................................................35 2.5 Objectives of This Study...................................................................... 37 Chapter 3 MATERIALS AND METHODS........................................................39 3.1 The Model Systems............................................................................. 39 3.2 Heating Methods.................................................................................. 40 3.3 Temperature Profiles.......................................................................... 40 3.4 Water Loss After Heating and Cooling.............................................41 3.5 SEM........................................................................................................41 3.6 Texture Measurements.........................................................................42 3.7 Determination of Self-diffusion Coefficients...................................43 3.8 Determination of Dielectric Properties............................................. 44 3.9 Examination of Gluten Proteins......................................................... 45 iv Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3.9.1 Fractionation...........................................................................45 3.9.2 SDS-PAGE...............................................................................47 3.9.3 Carbon-13 NMR Spectroscopy..............................................48 Chapter 4 RESULTS ................................................................................... 50 4.1 Physical Characteristics of the Samples.............................................. 50 4.2 Heat Penetration...................................................................................52 4.3 Moisture Content After Heating........................................................ 60 4.4 SEM......................................................................................................... 61 4.5 Instron Data........................................................................................... 67 4.6 Self-diffusion and Dielectric Property Data....................................... 72 4.6.1 The Three Ingredients Separately.......................................72 4.6.2 The Mixed Samples...............................................................75 220.127.116.11 General Information on the PGSE-NMR Experim ents.......................................................................... 75 18.104.22.168 Unheated Samples.................................................... 78 22.214.171.124 Heated Samples........................................................ 81 4.7 Qualitative Information from the Fractionation Procedure.......... 87 4.8 SDS-PAGE...............................................................................................93 4.9 13C NMR.................................................................................................103 4.9.1 Spectral Assignments.............................................................. 106 4.9.2 Discussion of l 3C NMR D ata.................................................112 4.10 Selected Amino Acid Profiles........................................................... 116 4.11 Overall Discussion............................................................................... 120 Chapter 5 CONCLUSIONS.......................................... 121 LITERATURE.......................................................................................................... 125 Appendix I Raw data of texture experiments.....................................................139 Appendix II Raw data of D value experiments..................................................140 Appendix HI Raw data of dielectric constant (k1) experiments........................ 143 Appendix IV Raw data of dielectric constant (k") experiments........................ 146 Appendix V Representative 13C spectra of unheated samples........................149 Appendix VI Representative 13C spectra of conventional heated sam ples.............................................................................................................. 152 Appendix VII Representative 13C spectra of microwave heated sam ples.............................................................................................................. 155 v Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES 4.1 Moisture contents of the center of representative heated samples, on a wet basis, as determined by the vacuum oven m ethod.................... 60 4.2 Average weight of samples after heating, before compression w ith the Instron....................................................................................................... 70 4.3 Moisture contents, self-diffusion coefficents, and dielectric properties of the components of the model system...................................72 4.4 Dielectric constant for unheated samples.......................................................79 4.5 Dielectric loss for unheated sam ples............................................................. 80 4.6 Dielectric constant for heated samples.......................................................... 84 4.7 Dielectric loss for heated samples.................................................................... 85 4.8 pH values during HC1 fractionation procedure........................................... 88 4.9 Peaks present in 13C spectra of all samples.....................................................109 4.10 Chemical assignments for peaks of fraction spectra................................... 112 4.11 Percentage of amino acids in select fractions...............................................119 vi VW ” , I :'l I M V 1 11.11, ff .1.1M| || I II I I ■II ,q.i ii,| jh.i■ ii ^i y pi.Bi 'wmw; ii j .el lirii - . i ;■-i ■ |. Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES 2.1 3.1 3.2 4.1 4.2 4.3 Models of interactions between gluten components............................... 17 30 cm2 Ottawa Texture Measuring System test cell.................................. 42 Flow chart of fractionation procedure......................................................... 46 Conventionally heated samples.................................................................. 51 Microwave heated samples.......................................................................... 51 Representative temperature profiles for CV heating of 35% moisture content samples with all five starch:gluten ratios....................54 4.4 Representative temperature profiles for CV heating of 50% moisture content samples with all five starch:gluten ratios...................55 4.5 Representative temperature profiles for CV heating of 65% moisture content samples with all five starch:gluten ratios...................56 4.6 Representative temperature profiles for MW heating of 35% moisture content samples with all five starch:gluten ratios.................. 57 4.7 Representative temperature profiles for MW heating of 50% moisture content samples with all five starch:gluten ratios.................. 58 4.8 Representative temperature profiles for MW heating of 65% moisture content samples with all five starchrgluten ratios.................. 59 4.9 Representative scanning electron micrographs of unheated samples........................................................................................................... 63 4.10 Representative scanning electron micrographs of 35% moisture samples........................................................................................................... 64 4.11 Representative scanning electron micrographs of 50% moisture samples........................................................................................................... 65 4.12 Representative scanning electron micrographs of 65% moisture samples........................................................................................................... 66 4.13 Represenative results from Instron compression.................................... 68 4.14 Effect of composition on the force required to compress heated samples........................................................................................................... 69 4.15 Sorption isotherms for dry ingredients of model system at 23 °C......... 74 4.16 Plot showing the effect of increasing gradient pulse duration on echo attenuation for the PGSE-NMR experiment.....................................75 v ii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4.17 Self-diffusion coefficients for unheated starch and vital wheat gluten mixtures at three different moisture contentsl7 Self diffusion coefficients for unheated starch and vital wheat gluten mixtures at three different moisture contents...........................................78 4.18 Attenuation factors for unheated starch and vital w heat gluten m ixtures...........................................................................................................81 4.19 Self-diffusion coefficients for (A) conventional and (B) microwave heated starch and vital wheat gluten mixtures...........................................83 4.20 Attenuation factors for (A) conventionally and (B) microwave heated starch and vital wheat gluten mixtures...........................................87 4.21 Structures of the amino acids and their three letter codes.......................91 4.22 Electrophoregrams of unheated fractions.................................................. 94 4.23 Electrophoregrams of fractions from 35% moisture samples that were heated......................................................................................................95 4.24 Electrophoregrams of fractions from 50% moisture samples that were heated......................................................................................................96 4.25 Electrophoregrams of fractions from 65% moisture samples that were heated......................................................................................................98 4.26 Sample spectra of solvents used for 13C NMR experiments....................105 4.27 Examples of sample 13C spectra...................................................................108 v i ii Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 1 INTRODUCTION The use of microwave (MW) ovens has increased tremendously over the last few decades, and as a result a wide variety of food products are being heated by this method. Cereal-based products such as breads and rolls commonly are heated in a MW oven, yet they have a tendency to become very tough and chewy when cooled. It is not commonly known w hy this occurs. In order to understand the mechanism by which this happens, it is important to evaluate the interaction of wheat flour and water in these types of foods. The structural development of baked products is strongly influenced by w heat flour proteins even though they comprise only 7 to 12% of the flour with respect to starch. Thus, water, starch, and gluten interactions are probably very important to the toughening problem during MW heating. Water is of considerable importance because it structures itself around molecules. The manner by which it interacts could be different when a product is heated by MW radiation, especially since water is essential for the MW heating of food. Water is a dipolar molecule which tries to align itself in the alternating electromagnetic waves produced by the microwave oven. This rapid movement towards alignment causes friction which provides for the transfer of electric energy to heat energy. This generation of heat along with other thermal properties that contribute to heat transfer is responsible for overall heating. The water will have different levels of interaction with the other molecules present which affects the w ater mobility as well as the motion of the molecules. 1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Starch contains carbohydrates which have macromolecular structures that are strongly influenced by the presence of water during heating, which results in granule swelling and loss of granule integrity. The two polymers found in the starch granule are amylose, which is more straight chain, and amylopectin which has more branch points. It is known that bread type products stale (undergo retrogradation) upon storage due to the reorientation and interaction of the starch molecular polymers which result in a tough texture to the product. Generally this phenomenon occurs over time. The reason this behavior (if it is the same) is accelerated during MW heating might be related to the rapid heating and subsequent cooling of the product. Wheat proteins are required to incorporate air in dough and maintain air cells during baking. Research has been conducted to determine what happens to the protein chains when they are mixed during dough formation. After hydration of the molecules, there is an unfolding of the protein chains with sulfhydryl and disulfide interchanges taking place. What transformations occur as proteins are heated are not fully understood. Gluten is difficult to study due to the complex nature of these proteins. They range in molecular weight from 11,000 to over a million daltons and therefore differ in their solubilities. The lower molecular weight components are loosely termed gliadins and the higher molecular weight components are termed glutenins. In this thesis, the effect of conventional and MW heating on starch and gluten in the presence of water will be evaluated. The appearance will be characterized by scanning electron microscopy and textural properties will be evaluated with an Instron. Dielectric properties and self-diffusion coefficients will be used to evaluate water interactions with the macromolecules. The 2 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. protein will be separated by pH and the fractions characterized by molecular weight differences as shown by sodium dodecyl sulfate polyacrylamide gel electrophoresis and nuclear magnetic resonance. 3 ■tiWvr.r Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 2 REVIEW OF THE LITERATURE 2.1 Heating Heating is a common unit operation that is applied to foods in order to make them appealing, safe, and nutritious. The process causes the sample to increase in temperature which provides energy for water diffusion and macromolecular transformations, such as starch gelatinization or protein denaturation. The dehydration and new molecular structure will affect the texture of the food, which may be desirable, but m ay also be undesirable. Conventional heating has been more extensively studied than microwave heating which is a more recent method of increasing the tem perature of a food, and is more complex in the heat generation process. Although the basic mechanisms of both heating methods are understood, the effect each has on the macromolecules, which affect the textural properties, is not clearly worked out. 2.1.1 Conventional Heating During conventional (CV) heating food is placed in a hot oven and the differential in temperature between the sample and the surrounding air is a driving force which results in conductive heat transfer from the surface to the center of the sample. In general, the increase in temperature is linear and the greater the temperature differential, the faster the food heats. The rate of heating is inversely proportional to the heat capacity of the material. A food with a small heat capacity (or specific heat) will require less energy to increase in temperature by one degree, and thus will heat more rapidly. The thermal 4 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. conductivity and density are inherent properties of the sample and play a large role on how the sample heats. As the tem perature increases the water gains enough energy that it is able to form steam and move to the surface of the sample. The basic principles of mass transport prevail, and the liquid water moves from high to low concentration, which is from the center of the sample to the outside edge (Wei et al., 1985a). If water is required for transformations of the macromolecules the flow of water to the surface will be interrupted altering evaporation rates, unless excess water is present. This was shown for cakes in which starch gelatinization inhibited the water loss rate (Gordon et al., 1979). The composition of the sample is also important in determining how it heats in the oven (Cloke et al., 1984) 2.1.2 Microwave Heating Microwave (MW) heating is somewhat more complex than CV heating because conductive-convective heating is taking place as well as electromagnetic radiation coupling (Datta, 1990). The importance and complexity of microwave heating has been discussed in various reviews (Buffler and Stanford, 1991; Engelder and Buffler, 1991; Davis and Gordon, 1990; Lee, 1989; Mudgett, 1986; Decareau, 1985). Microwaves are electromagnetic waves, which are most often at a frequency of 2450 MHz (in household microwave ovens). When the waves encounter a dipolar molecule (most commonly water) it tries to align itself in the alternating electric field. This movement creates friction which results in a conversion of the electrical energy to heat energy. As the waves enter the sample at the surface they may be reflected, transmitted, or absorbed. The extent that each option will occur is determined by the sample composition. If the waves are 5 Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission. able to penetrate deeper into the sample, more heat will be generated internally, rather than solely move from the surface inwards. The size and shape of the sample will also affect where the waves will focus. For rectangular shaped products there may be excess corner heating, whereas with spheres waves may concentrate in the geometric center. The general principles of heat transfer that were occurring during CV heating also apply during MW heating as the energy is distributed from the coupling molecules to the rest of the material. Since the heat capacity is a critical factor in the heating rate, a smaller heat capacity (or specific heat) will enable one to use less energy to increase the temperature of the sample by one degree than when the heat capacity is large. A difference in MW heating, relative to CV heating, is that the air surrounding the sample is most often not at an elevated temperature. This results in greater heat loss from the surface. Also with MW heating of foods, the wattage is generally large enough that the rate of heating is very rapid compared with CV heating. These differences in events during MW heating can result in water movement opposite to that of CV heating (i.e., from the surface to the center) (Wei et al, 1985b). Some experimental work has been done to elucidate the physico-chemical effects on baked products of MW heating (Lambert et al., 1992; Baker et al., 1990; Umbach et al, 1990; LePage, 1989). 126.96.36.199 Dielectric Properties The specific dielectric properties which pertain to MW interactions that result in heating are given by von Hippel (1954) and have been discussed by various scientists with regard to food applications (Engelder and Buffler, 1991; Decareau and Peterson, 1986; Mudgett, 1986,1985). The ability of water to 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. facilitate the conversion of MW electrical energy to heat energy is dependent upon the molecular mobility. If the water is bound to the macromolecules that are present and is unable to rotate there will be limited heat formation. On the other hand, if the water is not affected by the surrounding molecules, then the conversion will be expedited and the rate of temperature increase will be relatively higher. The mobility of the macromolecules also plays a role. Water movement may be affected by dissolved salts or other solutes. These ions will also be affected by the oscillating electromagnetic field and will move to try and align themselves. The nature of the ions and their concentration will influence the amount of coupling that will occur. Collisions of ions as they move converts kinetic energy to heat energy. The dielectric properties are used as averages of the quantifiable measures of the ability of a sample to interact with the MW energy. There are three main dielectric properties: the dielectric constant (k') and loss (k"), and the attenuation factor (a). These variables are used as indicators of how a food system will store the electromagnetic radiation (k') (i.e., energy that is reflected vs. transmitted), which is related to polarity, and how it will transform or dissipate the electric energy into heat (k”). A loss tangent or dissipation factor (tan 8) is a ratio of k" to k’, and represents the energy loss characteristics of the sample. The attenuation factor is a measure of how the waves will be affected when they go from the surrounding air into the food medium. The factors that affect a (cm-1) are demonstrated in the following equation (von Hippel, 1954), where X is the wavelength (cm): 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The inverse of a is penetration depth which provides information on how far a wave will penetrate into the sample before being reduced to 36% (1 / e) of its original intensity. Equipment has recently become available that allows non-destructive measurement of dielectric properties of food samples (Anon., 1992). Hewlett Packard has developed an open-ended coaxial probe which is part of a microwave transmission line. The probe, which has a MW signal at the frequency desired coming out the end of it, is placed in contact with the food. The food will affect the reflected signal and this altered signal can be used to calculate k' and k". Despite the advantage of knowing the dielectric properties in order to predict how a food will heat in the MW oven, food systems have not been widely characterized by these variables. Part of the reason for this lack of data is that the dielectric properties are dependent on temperature and the physico-chemical nature of the sample which are constantly changing during the heating process. Both the dielectric constant and loss of water decrease as the tem perature increases at the frequency of household microwave ovens. Conversely to water, the change in the dielectric loss for ions is directly proportional to the change in temperature (Mudgett, 1985). 2.1.3 Events During Baking As mentioned in section 2.1.1 there will be a temperature gradient in the interior of the sample as it attempts to equilibrate. The extent of the tem perature gradient determines how fast the food will heat (Varilek and Walker, 1984). As long as there is excess water present the maximum 8 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. temperature reached by the food is 100°C (Godsalve et al., 1977; Gordon et al., 1979; Galletti et a l, 1980). The temperature at the surface may be higher than the center due to dehydration during CV heating. Traditionally, temperature is monitored during heating with a metal thermocouple. This method was used by Marston and Wannan (1976) to show that bread heats more slowly as one moves towards the center from the surface. The temperature increase allows water to evaporate from the surface. Heat also supplies energy for gelatinizing starch and denaturing gluten (see Section 2.2.1). The effect the composition has on the heating rate can be monitored. The ingredients in a cake can have an effect on the temperature profile as was found by Gordon et al. (1979) w hen different ratios of flour to starch were used. Metal thermocouples cannot be used in a MW oven as the metal may reflect or arc and cause irregularities in the electromagnetic field. Fiber optic temperature sensors are used instead (Wickersheim and Sun, 1987). This method has been used successfully to monitor heating of cereal-based systems (Lambert et al., 1992; Baker et al., 1990; Umbach et al., 1990; LePage, 1989). 2.1.4 Patents about Microwave Heated Cereal Products Baked products made from cereal flour that are heated in a MW oven generally do not have a desirable texture. This problem is of such great interest to the food industry that a number of patents have been issued attempting to solve the problem. In 1984, Ottenberg commented on the unpalatability of MW heated rolls m ade with wheat flour and offered a roll reformulation as a solution. The most im portant alteration to the traditional formula is the inclusion of 10 to 20% (on wheat flour basis) of long-grain rice 9 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. flour and vital wheat gluten to maintain the starch to protein balance. A year later, Ottenberg (1985) published another patent where any starches could be used, as long as granular size was small (< 20 microns). In both of these patents "a variety of people" found the improved formulation gave a food that had superior crust texture after MW reheating compared to a traditional formulation. Ottenberg does not provide any mechanism by which the starch would affect the texture, or why the "crystal size" of the starch would be important. (Note: The author uses the word "crystal" for starch, and it can be assumed that the "granule" is being described.) Anderson et al. (1989), rather than examining the formulation of the product, examined the whole process by which the food is heated and cooled. The authors claimed that retarding the moisture loss during cooling was a sufficient method to retain palatability. Cochran and others (1989; 1990) also have patented methods of MW heating baked goods. They claimed in one patent (1989) that the addition of chemically modified starches would absorb water to improve palatability. In the other patent (1990), a protein modifier (L-cysteine) was incorporated in the baked product to break disulfide bonds which was claimed to improve texture. Huang et al. (1991) were able to patent the idea that the inclusion of "preferential binding materials such as surfactants . . . hydrogen bond breakers" would interact with gluten to give a more desirable texture. They also felt that the inhibition of starch granule swelling was important in order to maintain desirable texture. This discussion of patents illustrates that there is no general consensus as to what key ingredient controls texture of baked products. That only ONE component is responsible for the texture in baked products seems unlikely. Thus the modification of a system for MW heating appears to be complex. 10 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.2 Cereal Products 2.2.1 Composition of Dry Ingredients 188.8.131.52 Gluten Gluten is the "proteinaceous material that is recovered after washing out of flour-water doughs the soluble and occluded substances of flour." (Bloksm a and Bushuk, 1988) and was first described by Beccari in 1745 (see Shewry and Tatham, 1989). Despite its singular name and years of study, gluten is a very heterogeneous mixture of proteins that has been extensively studied, but not totally understood. In general it is characterized by its solubility, as popularized by Osborne in 1907 (for more detail, see Section 184.108.40.206). Two main groups were found: the prolamins, called gliadin, which are alcohol soluble, and the glutelins, called glutenins, which are acid soluble. This is a very simplified way to divide gluten based on solubility and Section 220.127.116.11 describes some of the variety of other methods that have been used. The following paragraph provides a brief summary of gluten composition. More recent investigations of wheat proteins were reviewed by Shewry et al. (1986) to examine in more detail the structure of the gliadin and glutenin. The gliadins are generally smaller with a , (3, and y subfraction designations whose molecular weights range from 30,000 to 40,000 daltons and the co fractions whose molecular weight is between 44,000 and 74,000 (Shewry et al., 1986). All the gliadins contain cysteine except the co fraction. Reduction of disulfide bonds may result in a change in mobility by sodium dodecyl sulfatepolyacrylamide electrophoresis at low pH. The glutenins are larger molecules composed of protein chains that are disulfide bonded to each other to give molecular weights of over a million. Shewry and co-workers (1986) divided 11 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. glutenin into two groups - low and high molecular weight (LMW and HMW). The LMW group is thought to be aggregates of gliadin, held together by hydrogen bonds and hydrophilic interactions. This idea is supported by the amino acid profiles which are similar to gliadin. The HMW group has a different amino acid profile (more glycine and less proline) than gliadin and the proteins are stabilized by inter-chain disulfide bonds. This same basic system of characterizing the gluten was used in an electrophoretic study by Fullington et al. (1987). These workers divided gluten into four groups based on molecular weights: A l, >80,000; A 2 ,51,000-80,000; A3,40,000-50,000; A4, 28,000-39,000. Proteins with a molecular weight between 8,000 and 27,000 were classified as albumins and globulins. A recent review (Shewry et ah, 1992) discussed the HMW glutenins in great detail with an emphasis on the genetics. In 1983 Payne and Lawrence suggested a system of assigning electrophoretic bands to the glutenin subunits. They compared their preferred lettering system to the more commonly used numbering system, which includes up to 22 possible subunits in wheat glutenin, not all of which will be expressed by one variety. In order to examine gluten it can be easily isolated from a flour dough with water because the starch washes out to leave the gluten as a cohesive mass. Vital wheat gluten (VWG) is one term that can be used to refer to functional gluten after it has been separated from the water soluble components of wheat flour and dried. It can be specifically defined as "gluten that has not been denatured and thus has the ability to form a dough that will retain gas" (Hoseney, 1986) or gluten that forms a viscoelastic mass upon complete hydration (Ortalo-Magne and Goodwin, 1991). 12 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 18.104.22.168 Starch Starch is the largest component of wheat flour and is essential for its dough and bread-forming properties (Sandstedt, 1961). This carbohydrate is made up of two main polymers - amylose and amylopectin - which are composed of the glucose monomer, connected together with an a(l->4) bond. Amylopectin also has <x(l->6) bonds which give branch points on the polymer chain. When the wheat kernal has been formed, these polymers will be found in units called granules which are spherical and range in size from 10100 pm; the exact structure (i.e., location in the granule of amylose and amylopectin) is not known. Various discussions about starch are available (Whistler et al., 1984; Zobel, 1988). Upon heating, with adequate water, at about 60°C the starch granules will swell and the amylose will leach out. In a baked product such as bread, the starch may undergo structural changes that will affect the texture, a phenomenon referred to as staling that has been examined over the years by various researchers (Willhoft, 1973; Russel, 1983; Avital et a l, 1990; Wilson et al., 1991). Wilson et al. (1991) reported that the amylopectin portion of the starch is responsible for staling based on molecular structural changes. 22.214.171.124 Interaction of Gluten, Starch, and Water In order to make a basic dough, two ingredients are required: flour and water. The outer layers of the flour particles become hydrated and with stirring get removed to reveal more dry parts which can become hydrated, and so on. Once the particles are completely hydrated, subsequent mechanical action results in the development of the dough. Although wheat flour contains primarily starch, it is the wheat proteins that are the crucial 13 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. components in the dough that retains carbon dioxide from a leavening agent, and then upon baking and expansion will retain the shape (Pomeranz, 1983). The gluten chains are affected by the mixing (or kneading) and expand to form the structure of the system (Hoseney and Finney, 1974). The water is not evenly distributed between the dough main components (starch and gluten) and exists at different levels of hydration in localized areas. The amino acid side chains provide numerous opportunities for hydrogen bonding with water (Bushuk and Hlynka, 1964). From a rheological study, Webb et al. (1970) felt that water was present either free, lightly bound, or firmly bound in flour, water, salt doughs. Wynne-Jones and Blanshard (1986) examined starch gels and found that at 23% moisture all water was bound. There was quite a large difference in mobility between 33 and 38% moisture in the starch gels. MacRitchie (1976) felt that the water was also not evenly distributed among the dough components and that m ore water is associated with the protein than the starch. During baking there is a redistribution of the water as the starch gelatinizes and takes water from the denaturing protein (Bushuk and Hlynka, 1964; Bushuk, 1966). The significance of this water transport phenomena was demonstrated by Eliasson (1983) with differential scanning calorimetry. She showed that the amount of gluten and the moisture content affect the temperature of gelatinization. 126.96.36.199 Role of Gluten in Dough and Bread Formation Numerous studies have been carried out to try and determine which part (if any one in particular) of the whole gluten is m ost im portant for the formation of high quality doughs and bread. There is no general consensus in the literature, although glutenin seems to be most im portant but gliadin 14 Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission. also contributes to structural development, as was reported by Chakraborty and Khan (1988b). This generally agrees with MacRitchie (1987a) who found that gliadin decreases dough strength and glutenin increases dough strength and loaf volume. Bushuk (1985) divided glutenin and found that part of it was inversely related and part was directly related to loaf volume, with gliadin having no significant effect. Conversely to that, Hoseney et al. (1969a) reported that gliadins control loaf volume and glutenin is responsible for the mixing requirements of the dough. Over the years, various models have been proposed about the interaction between gliadin and glutenin. Wall (1979) suggested the gliadin was less associated with the system and the glutenin, which has m ore surface area could non-covalently hold the dough together (Fig. 2.1A). Lasztity (1986) proposed his own model, after reviewing a number of models available in the literature, in which there was more bonding between the gliadin and glutenin (Fig. 2.1B). Weegels and coworkers (Weegels and Hamer, 1991; Weegels et al., 1991) proposed that during dough formation hydrophilic regions of gliadin and glutenin interact to bring the protein chains together (Fig. 2.1C). This provides the opportunity for disulfide bonds to form and provide stability to the structure. 2.2.2 Appearance One is often interested in the physical appearance of a baked product, such as bread, that goes beyond shape and color. Scanning electron microscopy (SEM) is a relatively simple technique, requiring little preparation when used to examine bread. SEM allows one to see both the air cells (at a lower magnification) and the starch granules (at a higher magnification) (Taranto, 15 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1983). Air cells are a common feature of baked cereal products that one may want to investigate. Different ingredients and heating methods m ay have an effect on the size, shape, and numbers of air cells. Part of the microscopic appearance of a bread-like product is the size and shape of the starch granules. When enough water is present the starch granules will swell and exude some of their contents which will alter the morphology of the granules. The further the extent of gelatinization the greater will be the change in granule shape relative to the ungelatinized granules. Starch is not the only component in a baked product, one also sees material between the granules. This matrix before heating would most likely be gluten and after heating would include some of the molecules that were exuded from the starch granules. SEM has been used by many investigators to examine starch granules in bread and other baked products after heating (Khoo et al., 1975; Hoseney et al., 1977; Varriano-Marston et al., 1980). Air cell size has been examined in cakes to relate to bulk density (Attenburrow et al., 1989) and cross-sectional area (Lambert et al., 1992; Cloke et al., 1984). Wheat starch has been examined with SEM to determine if MW heating has any effect on the starch granules compared to those that have been CV heated. Both Goebel and Zylema and their coworkers (1984 and 1985, respectively) examined starch-water mixtures by SEM and did not find any unique macrostructures after MW heating. Goebel et al. (1984) did find more diversity in the granular shape after MW heating as compared to CV heating. Air cell size and starch granule morphology was similar between a CV and MW baked bagel (Umbach, 1989). Huang et al. (1991) in their patent state that a dough after MW heating had larger starch granules than if it was CV heated. 16 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A s° G lu te n in R e s id u e "-4s ^W.----:-/. ZOg.Q _ Q ,.-.—T ’ - r v ^ ' ‘ •<— ^ D ough p r o te in s 6 JUJJT /w//'x * .... \--w ^ random coil ^A A , disulphide bond hydrogen or hydrophobic bond aggregated gluten J g lu ten in t e e la eyacatn lla oiiacsn^ l la c r o c * 1 chomicai aggregation la DOiymariauon oy S-*S B DOtymariutton oy ito iioasa \m M fl laatnm 8Cl(J 'lib b hemtcaiiuiaaa D phyaicol aggregation io asecrete o g liad in S-S bond hydrophobic site carbohydrate Us prot»r»~orotain interaction IB ofotetn-noio urcaraction canxfiy<nita-orot#« Figure 2.1 Models of interactions between gluten components. A) Wall, 1979; B) Lasztity, 1986; C) Weegels and Hamer, 1991. 17 •OT* Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.2.3 Texture From the patents discussed in the previous section, the importance of texture is evident. The texture that a loaf of bread possess when it is first removed from a CV oven does not remain constant, but changes as the bread ages. This phenomenon, characterized by a decrease in softness, is often referred to as staling. Texture generally relates to the behavior of the food when it is chewed and chewing can be thought of as essentially compressing the food between the teeth. Thus, it follows that a mechanical device that is measuring texture should simulate compression. Attempts have been made to quantify these textural changes using various methods (Willhoft, 1973; Voisey et al., 1974; Kamel, 1987; Joensson and Toemaes, 1987; Baker et al., 1987), while the official AACC method (number 74-10) employs a Baker compressimeter. For whatever m ethod is chosen, in order to have consistent results when compressing a sample between two parallel plates, sample size and shape should be kept constant (Brinton and Bourne, 1972; Voisey et al., 1974). These factors, as well as rate of compression and area compressed, were evaluated by Baker et al. (1986a,b). They found that the most significant factors were crosshead speed, plunger area, and degree of compression for bread crumb. A compression rate of 100 m m /m in was recommended by Hibberd and Parker (1985) and Baker and co workers (1986a). Other variables in bread that have been evaluated for their effect on texture measurements are location in the loaf where the sample is taken from (Hibberd and Parker, 1985; Baker et al., 1987) and the amount of gluten (Lasztity, 1980). Another method for measuring texture was described by Voisey (1971) which may be used with a variety of foods. The Ottawa Texture Measuring 18 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. System (OTMS) has a number of different test cells that can be used with an Instron, and there are inter-changeable parts to allow different textural characteristics to be measured. The test cell has high sides so when the plunger comes down and compresses the sample it cannot go out from underneath the plunger. One option is a wire extrusion insert so that the sample is sheared on the wires when forced through with the plunger. This system has been reported to be used for fruits and vegetables (Voisey, 1970) and meat (Voisey and Larmond, 1977) but not for bread-like products. Some work has been done to study the effects of MW heating on the texture of cereal products. Bread that was baked in a CV oven was found to be firmer than if baked in a MW oven when texture was measured with a penetrometer (Lorenz et al., 1973). This difference was attributed to a higher moisture content in the MW heated bread. Sensory panels did not agree with the penetrometer data. Dahle and Sambucci (1987) found an increase in extensibility (amount bread is deformed before rupture) and no change in strength (maximum force before rupture) for MW thawed bread compared to room tem perature thawing. Rogers et al. (1990) used a Kramer shearcompression cell to examine sliced bread and gluten balls that had been reheated by one of three methods - steaming, convection oven, and microwave oven. An advantage of the Kramer shear method is that firmness and toughness could be differentiated. They were able to support the claim that toughening found with m ia o w ave heating was due to the gluten. Some researchers have thought of bread and cakes as being foams and used this idea in the interpretation of the compression of these foods between parallel plates. Attenburrow et al. (1989) and Swyngedau et al. (1991) have reported on stress-strain curves to demonstrate that the compression first 19 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. destroys the integrity of the cell walls and then compresses the cell wall material. From the work that has been published, there does not appear to be any consensus on which is the "best" method to use to measure the texture of a baked cereal product. The method chosen appears to be dependent on the parameters needed to be studied and, possibly, availability of equipment. 2.3 Water-Macromolecular Interactions Water is a simple molecule with a structure that allows it to hydrogen bond to molecules which have hydroxyl, amino, carbonyl, or sulfhydryl groups. There are various possible levels of interaction between the water molecule and the macromolecule. By measuring the moisture content at different water activities it is possible to hypothesize when the macromolecules would have a single layer of water surrounding them (a high level of interaction) and when there w ould be subsequent layers (lower levels of interaction) (Fennema, 1985). Isotherm data evaluated by the Brunauer, Emmett, and Teller or Guggenheim, Anderson, and DeBoer methods provide a theoretical basis to predict the amount of water required to form a single layer of water on the surface of the macromolecule (Fennema, 1985). This monolayer value is dependent on the nature of the macromolecule. The monolayer can also be thought of as rather immobile water. If the moisture content is great enough, there will also be mobile water. The extent of this mobility is of interest. In order to measure water mobility, it is desirable to use a method that is specific for water, as well as being non-destructive to the interaction between the water and the macromolecule. 20 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The type of interaction between water and starch or gluten has been of interest for a number of years (Bachrach and Briggs, 1947; Bushuk and Winkler, 1957; Volman et al., 1960; Lee, 1970; Chinachoti and Steinberg, 1988). More detail on these interactions will be given in section 188.8.131.52 where dough formation is discussed in detail. Interactions of water w ith various proteins (Lumry, 1973) or with starch (Tanner et al., 1991) have been discussed . 2.3.1 Proton Nuclear Magnetic Resonance In the 1940's nuclear magnetic resonance (NMR) was beginning to become popular as a method for examining nuclei in solutions. Bloembergen et al. (1948) published a rigorous approach to the theory of nuclear magnetic relaxation. In 1952 Bloch and Purcell jointly won the Nobel prize for physics for their work on NMR (Bloch, 1953; Purcell, 1953). For NMR experiments one places the sample (with all its nuclei) in a magnetic field and by carefully choosing the frequency and duration of the pulses that the sample is subjected to, specific nuclear behavior can be examined. Proton mobility is often examined to determine water behavior, but the protons on macromolecules can also be evaluated. A more complete description of NMR is given by Harris (1983). Pulsed gradient-spin echo NMR (PGSE-NMR) is a very specific type of NMR experiment that can be used to determine self-diffusion coefficients, m2/s (D). One is therefore measuring mobile protons, usually water, while maintaining the integrity of the sample. This m ethod was first proposed in 1965 by Stejskal and Tanner. When water interacts with other substances, its self-diffusion coefficient decreases from that of free water. This method requires a field gradient to be pulsed on the sample to positionally identify the 21 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. nuclei. When the gradient is removed, the nuclei have the opportunity to diffuse (or move). The gradient is p u t back on and if the nuclei are not in the same place they were the first time the gradient was on, the signal will be attenuated. The greater the attenuation the larger is D of the nuclei (Blum, 1986). It is important to know that the D value is providing information on the Brownian (or random) motion of the nuclei of interest and not Fickian (concentration dependent) motion (Karger et al.,1988). The pulsed gradient method of tracing the movement of molecules leads to a measurement of Brownian motion in the absence of concentration gradients, thus leading to a non-invasive determination of the true self-diffusion coefficient. Water has a D of 2.3 x 10'9 m2/ s at 25°C (Callaghan et al., 1980). 2.3.2 Water-Macromolecular Interactions as Studied by NMR For a number of years NMR has been used to examine starch and water interactions. A review of NMR applications for food has been written (Belton and Colquhoun, 1989). Collison and McDonald (1960) were able to demonstrate the increased mobility of the macromolecular side chains of starch after gelatinization by examining the line width of the proton signal. This technique was used to more fully characterize water behavior during gelatinization (Jaska, 1971). In 1968 Sterling and Masuzawa also examined starch with proton NMR and demonstrated the increasing interaction between starch and water when starch gelatinizes. Boundness of water in doughs and starch has been examined with NMR (Toledo et al., 1968; Shanbhag et al., 1970). 22 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. NMR can be used to measure relaxation times of water in order to examine its mobility or level of interaction with the macromolecules. Tait et al. (1972) examined other aspects of proton and deuteron signals to examine mobility and layers of water on starch. Various researchers have shown that gelatinization results in changes in relaxation times, increased starch polymer mobility, and the presence of different forms of water (Chinachoti et al., 1991; Hennig and Lechert, 1974; Lelievre and Mitchell, 1975; Hennig and Scholz, 1976; Leung et al., 1976). Wheat flour-water doughs were found to contain two populations of water - mobile and immobile - based on the proton transverse relaxation time (Leung et al., 1979). Three populations of water were found for wheat flour doughs investigated by d'Avignon et al. (1990). When wheat starch gels were MW heated and it was found, from fH transverse relaxation measurements, that there was a movement of water from white powder regions to gel regions from an initial moisture content of 50% w /w (Wynne-Jones and Blanshard, 1986). Two populations of water were found for com starch when (Lang and Steinberg, 1983) and for cereal starches when *H and 170 (Richardson et al., 1986,1987a, 1987b, 1987c) nuclei were examined. Gluten has not been examined by NMR to as great an extent as starch. A discussion of various ideas of water protein interactions was presented by Bryant and Halle (1982). A more detailed study of the relaxation of 170 nuclei to examine protein hydration has been done (Halle et al., 1981) where it was found that mobility decreases with increasing protein concentration. Relaxation times for dry gluten were measured and it was found that there was m ore than one population of proton nuclei (Belton et al., 1988). The 23 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. lipids present in the gluten affected the transverse relaxation and the disulfide bonded regions are critical for spin lattice relaxation. Little research has been published that shows the ability to determine D values for food systems and to use the information to characterize molecular level events. Lechert (1981) investigated the effect of moisture content and temperature on potato starch and found that D is directly proportional to both parameters. The association between water and protein in cheese was examined by Callaghan et al. (1983a). A study of wheat starch pastes showed that there was a lack of correlation between Theological measurements and D values (Callaghan et al., 1983b). A nonlinear relationship between moisture content and self-diffusion coefficients was demonstrated for water in potato starch (Lechert et al., 1980) and wheat endosperm (Callaghan and Lelievre, 1979). The non-linear effect of moisture content, from 24 to 54%, on the D of gluten was demonstrated by Back et al. (1991). 2.3.3 Water Behavior as Examined by Dielectric Properties The effect of temperature on the dielectric properties of meat and vegetables has been clearly demonstrated (Bengtsson and Risman, 1971). For potato starch-water mixtures of greater than 40% moisture measured at 3.0 GHz, an increase in moisture content resulted in an increase of k' and no change in k"; gelatinization caused an increase in both properties (Roebuck et al., 1972). The effect of temperature on the dielectric properties of sunflower oil, mashed poatoes and mashed bananas was demonstarted by Akyel et al. (1983). After heating to 55°C and cooling, k’ and k" for the bananas decreased. A coaxial probe was used to measure the dielectric constant and dielectric loss for various fruits and vegetables (Tran et al., 1984) over a range of frequencies. 24 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The experimental results correlated with the theoretical Maxwell equation predictions. The values at 2,400 MHz were between 57.6 and 68 for k' and 12.3 and 19.5 for k". Little work has been published with measurements made at 2450 MHz. Using the cavity perturbation technique, various ingredients for a cake were measured (Brand, 1987). The dry ingredients all had low dielectric properties, whereas the water was much higher. When the ingredients were combined in a batter, there was an interactive, not additive, effect on k1and k". If the moisture content was greater than 20%, Fleischmann (1991) found that the k' and k" for whey proteins increased with increasing moistue content, and k" reached a maximum at about 60% moisture. Miller et al. (1991) demonstrated the effect chemically modifying starch has on the dielectric properties. 2.4 Examination of Gluten One way to understand a large group of complex proteins, such as gluten, is to break them down into smaller groups. This can be achieved through the use of solvents of differing characteristics to obtain numerous sub-fractions. There are a wide variety of procedures available in the literature and each one results in fractions of different composition. Peptide chains with similar composition or physical characteristics will all be soluble in one solution. The fractions that result from a procedure are by no means absolute and thus the method used m ust always be clearly defined. Once these fractions are obtained, it is desirable to characterize them to understand the composition of the proteins. Wheat proteins have been extensively studied using sodium-dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) as is evident by the 25 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. number of gels illustrated in published material on gluten. This method is specific enough that it is often used for identification of cultivars by examining the gliadins or the glutenins (Ng et a l, 1988). Poly-acrylamide gel electrophoresis (PAGE) is a method of separating charged particles based on either their molecular size or charge. The procedure first published by Laemmli (1970) has been widely adopted for these experiments. A field for the molecules to travel in is produced with acrylamide monomer which is polymerized to produce a network with spaces adequate for the mobility of the proteins of interest. The proteins are p u t on the gel and a charge is applied which causes the proteins to move at a rate that is generally proportional to their size. Smaller molecules will be less retarded by the gel and will migrate further. If sodium dodecyl sulfate (SDS) is added to the sample buffer, the detergent will coat the protein and neutralize the charges so the movement of the molecules will only depend on size (Weber and Osborn, 1969). After the current has been turned off, the protein is stained with a dye (often coomassie blue) in order to see the bands of protein. Molecules of known molecular weight are also put in a lane of the gel and a standard curve can be generated of the distance traveled vs. log molecular weight (Robyt and White, 1987). A more recent method that has been used to examine wheat proteins is carbon-13 nuclear magnetic resonance (13C NMR). The following discussion of the methodology was taken from Breitmaier and Voelter (1990). Solution state 13C NMR is when the molecules are hydrated sufficiently so that the carbon atoms have some mobility. The 13C nuclei is naturally abundant, but at a lower concentration than JH, and thus more scans are often required to achieve the desired signal to noise ratio that will result in 26 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. well resolved spectra. The basic NMR theory is the same as for protons mentioned in section 2.3.1. The 13C spectra is the Fourier Transform of the free induction decay (FID) that results from the relaxation of the carbon atoms after they have experienced the radio frequency pulse of a sufficient length to perturb the nuclei 90°. The mobile carbon atoms in the sample will produce a signal which results in a peak in the spectra. Each carbon nucleus has a different environment surrounding it, which results in a slightly different frequency (chemical shift), with which it relaxes. Some deuterium atoms may be added to the sample, either in the form of deuterium oxide or as part of the solvent molecule, so that the spectrometer can lock on the deuterium signal. All the chemical shifts of the carbon atoms will then be given relative to the deuterium signal. The chemical shifts cover the range of 0 to 200 ppm. For example, the chemical shift for the carbonyl carbon is about 175 ppm and for the alkane carbons is about 40 ppm. The signal is often proton decoupled. This removes the influences of multiplet lines of a single to make the spectra clearer, which occurs by having only one peak per carbon atom and is attributed to the nuclear Overhauser effect. Individual amino acids will have unique spectral features when compared to each other. By comparing amino acid spectra to a protein spectrum, the amino acid composition of the protein can be determined. Difficulties in identification of peaks arise when signals overlap. The greater the concentration of the protein in solution, the larger will be the signal. If the concentration becomes too large, one risks having the mobility of the carbon atoms in the peptide chains inhibited, thus possibly eliminating those carbon signals. 27 Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission. 2.4.1 Fractionation and Electrophoresis of Unheated Gluten Osborne has been called the father of gluten fractionation. A procedure was described by Wrigley and Bietz (1988) where four solvents were used to solubilize the wheat proteins: water for the albumins, 0.5 M sodium chloride for the globulins, 70% ethanol for the prolamines (gliadins), and 0.05 M acetic acid for the glutelins (glutenins). When subsequent fractionation studies have used the same basic methodology, the same protein fraction names were retained for fractions that were assumed to be similar in composition to those of Osborne. Blish (1945) provided information on Osborne's work with a thorough discussion of the work he and other researchers did up to 1945. An emphasis is given on the effects of heat and solvents on the gluten proteins. Another excellent review of work done on wheat proteins from 1745 (by Beccari) to 1976 is given by Kasarda et al. (1976) and provides a m odem perspective on gluten studies. This section will discuss some of the key work covered in that review, as well as significant work done since the chapter was written in order to summarize the different solubility methods used in this field. A type of Osborne separation was reported by Jones et al. (1959) in which gluten was dispersed in 0.01 N acetic acid and m ade to 70% ethanol final solution. When the pH was increased to 6.5, half of the proteins precipitated and these were thought to be the glutenins. This method combined with other fractionation techniques was utilized in early electrophoresis work. This same technique was used by Bietz and Wall (1972). The authors used gel filtration to separate the proteins that were in an SDS mercaptoethanol solution. PAGE that was subsequently done was able to differentiate subfractions of the gliadins and glutenin components. Bietz et al. (1973) 28 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. provided a good review of the methods used to separate the gluten proteins and some methods to examine the composition of the fractions. Gel filtration can be used to further subfractionate the glutenin and gliadin fractions (Bietz and Wall, 1973). Bietz et al. (1975) published a clearly outlined modified Osborne fractionation procedure for use in genetic studies of wheat. Chen and Bushuk (1970) published a detailed modified Osborne procedure for flour in which centrifugation played a critical role. 0.5 M NaCl solution and water extracts were used initially to isolate the water and salt insoluble fraction. They obtained 70% ethanol soluble and insoluble fraction, with the latter fraction further extracted with 0.05 M acetic acid to obtain an acid insoluble fraction using centrifugation. Differences were found between the protein contents of the fractions between various wheats with a large portion of the hard red spring wheat (HRSW) protein falling into the residue category. Chen and Bushuk (1970) called the acid soluble fraction the glutenin, which is contrary to Shogren et al. (1969) who called the ad d insoluble fraction (pH 4.7) the glutenin. A few years later Bushuk and another co-worker (Orth and Bushuk, 1973) found that the Osborne method resulted in a very impure glutenin fraction based on PAGE. In an attempt to improve on the previous solubility methods, Meredith and Wren (1966) reported their own unique solution which contained 0.1 M acetic add, 3 M urea, and 0.01 M cetyltrimethyl ammonium bromide (CH3 (CH 2 )i 5 N(CH 2 )3 +Br‘). The solvent was given the acronym AUC. Urea will aid the solubilization by dissociating the protein chains. Urea in such a low concentration does not drastidy effect protein structure. A comparison was m ade between two similar solvents: AUC (Meredith and Wren, 1966) and AUB with Brij 35 substituted for cetyltrimethyl ammonium bromate. 29 Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission. Similar PAGE gels were produced with Brij 35 being preferred for its lack of charge. The overall recommended procedure which involved solubilizing the proteins in AUC, 70% ethanol, then precipitating the glutenins by increasing the pH to 6.4, required about six days. 0.01 M acetic acid was used to disperse the glutenin. Bottomley et al. (1982) found a sodium dodecyl sulfate (SDS)-Tris solution superior to AUC and CUC (cetyltrimethylammonium bromide, urea, and sodium citrate) in the amount of protein extracted and suitability for gel filtration when the extracted proteins were compared by SDS-PAGE. McMaster and Bushuk (1983) reported a rather complex procedure involving AUC, ethanol, centrifugation, and dialysis to separate glutenin and gliadin in order to examine the carbohydrate associated with each fraction. They implied that different monosaccharides may influence protein solubility in ethanol. Kobrehel and Bushuk (1977) tried a novel approach using sodium stearate to solubilize glutenins. Ham auzu et al. (1979) also solubilized glutenin in sodium stearate and added tris-hydrochloric acid to prevent the rapid (less than five hour) precipitation of the protein. They felt that the hydrophobic interactions of the peptide chains were important and the aggregation of the glutenin was disrupted by the sodium stearate but not to the extent of a chemical reduction of the disulfide links, based on SDS-PAGE and gel filtration work. This last finding was also published by Wasik et al. (1979) using a sodium stearate method. Graveland et al. (1979) used 1.5% SDS solution, ethanol, and centrifugation to separate wheat flours and reported that there were six glycoprotein fractions; three were SDS-insoluble. Subsequent research showed that there was only one SDS-insoluble 30 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. gylcoprotein and the previous fractions only contained sugars that were not bound to the proteins (Graveland et al., 1982). Considerable work has been done using only acids to separate the gluten fractions. A multicomponent study was undertaken using decreasing lactic acid concentrations (and therefore increasing pH's) to separate the various fractions (Shogren et al., 1969). The band separation was not very clear from starch gel electrophoresis, but differences in fraction composition were observed. This work was carried out further by the same group of authors (Hoseney et al., 1969b) who also included an ultracentrifuge in the separation method. Chung et al. (1979) found that a greater amount of protein could be extracted with 0.05 N than 0.005 N acetic acid. A lengthy procedure using primarily 0.05 M acetic acid and centrifugation was used by Preston and Tipples (1980) to separate gluten into an add soluble and insoluble fractions. Rather than fully reconstitute the fractions, they were added to flour in small amounts and their influence was manifested in that way. Unpublished work that was referred to indicated that the acid soluble portion was a mixture of gliadin and glutenin-like components. Three years later, Jones et al. (1983) were successful in separating gluten fractions with lactic add (0.005 N) and ultracentrifugation. Unpublished data mentioned that there was no loss of functionality. Finney (1985) summarized the article by Jones et al. (1983) and went on to further explain the significance in baking of each fraction. MacRitchie (1978) found a decrease in loaf volume with a fractionation method using 0.1 M acetic a d d and centrifugation. He found the acid soluble fraction to be composed of gliadins and glutenins. The next year MacRitchie (1979) showed the effect of pH on gluten solubility in acetic add. MacRitchie 31 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (1985) found that when the gluten was in contact with acetic a d d there was a loss in functionality. This conclusion may be questionable because the baking test to determine functionality did not work due to the presence of the acetate ion which depressed yeast activity. Dilute HC1 (0.05 M) appeared to have less affect than 0.1 M acetic add; the two acids were not compared at the same concentration. The study also found that up to 87% protein could be solubilized, with an HCI concentration of 4.5 x 1(H M and the aid of a homogenizer (spedfically an Ultraturrax), indicating that both gliadin and glutenin could be ad d soluble. Baking tests indicated that the shearing action of the Ultraturrax did not alter the vitality of the proteins. It was not indicated whether the gluten proteins were specifically affected by the acid. The lower the pH, the more deleterious effect it had on the peptide chains. Also, it was found that loaf volume was related to the pH of the dough. Whether this was due to the extraction method or protein chain interactions under the specific conditions used was not discussed. The molarity of the HCI used (between 4 x lO-4 and 45 x 1(H) had a direct relationship to the percent soluble protein. Lactic acid, which is sometimes used, was not examined. Letting the protein remain in the acid for up to 48 hours did not result in deterioration of the peptide chain. A successive HCI fractionation of gluten was used by MacRitchie (1987b) to examine functionality of the protein group components. But, SDS-PAGE showed that the nine fractions overlapped in the big groups of glutenin, gliadin, and residual proteins (MacRitchie et al., 1991). Eliasson and Lundh (1989) also found the proteins to be of a variety of molecular weights in the fractions that resulted from this HCI extraction procedure. 32 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chakraborty and Khan (1988a) conducted an extensive comparison of three separation methods. One method was that of Hoseney et al. (1969a,b) (which used the method outlined in Shogren et a l, 1969) in which lactic acid was used as the solvent for separation. The second method was of Orth and Bushuk (1972), which was a modification of Osborne taken from Chen and Bushuk (1970). The third method came from MacRitchie (1978) and used acetic acid as the solvent for the proteins. Flow charts, for easy comparisons, of the details of the isolation procedures are given in the article by Chakraborty and Khan (1988a). The sensitivity of the MacRitchie method to the type of mixer used was shown. The SDS-PAGE results showed that the Chen and Bushuk (1970) method resulted in different separation compared to the other two. One test of a fractionation method is to determine if the functionality of the fractions has been effected by the treatment. Therefore, a comparison of functionalities in reconstitution studies were perfomed (Chakraborty and Khan, 1988b). It was found that 70% ethanol is a strong enough denaturant to alter functionality of gliadins in a negative manner. Thus, Chakraborty and Khan (1988b) concluded that the Hoseney et al. (1969a,b) method was the best of the three. The Chen and Bushuk method resulted in denaturation of the proteins, and the MacRitchie method did not result in an equally effective separation of gliadin and glutenin. From the above discussion it is evident that there are a wide variety of methods for fractioning gluten. Each solvent will interact with the proteins differently and thus the fractions will be slightly different. When choosing a procedure, one must consider how the fractions will be characterized. Many solvents do not interfere with electrophoresis. If one is interested in doing 33 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 13c NMR (see next section), it is not desirable to have a solvent that is going to interfere w ith the spectra. One m ust also think about whether each fraction should be subjected to a different solvent or if it is preferable to have one solvent. The fractionation procedure of MacRitchie (1987b) that uses dilute HCI is a method that satisfies this criteria because it uses only one solvent with no carbon atoms. 2.4.2 13C NMR of Unheated Gluten Solution state 13C NMR has been used to a limited extent to examine gluten. In 1981 Baianu published the first *3C spectra of gliadins using a solvent of deuterated acetic acid with 5% propanol. The advantage of 13C in its resolution capabilities over *H is demonstrated. They assigned peaks to amino acids. The next year Baianu and co-workers (1982) expanded their NMR work to look at the effects different concentration and heating on the 13C spectra of gliadin in 10 mM acetic acid. At 32% (w /v) there was a broadening of the peaks, and this was not observed at 15.8% (w /v). A higher temperature (up to 60°C) did not appear to have that much effect on the spectra. More recently, Belton et al. (1987) looked at hydrated gluten by 13C NMR and reported a spectra that was similar to that of Baianu et al. (1982). It should be noted that in the Belton study, the lyophilized gluten was hydrated to about 60% moisture, and was thus a "plastic mass". 2.4.3 Effects of Heating After the gluten has been heated, the protein m ay separate differently than it did before heating. A different kind of fractionation procedure may be 34 Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission. necessary because the proteins are denatured which may affect their conformation and chemical solubility. One of the earliest studies to look at the effects of temperature (70-90°C) and moisture content (10-60%) on gluten was done by Pence et al. (1953). After subjection to the treatments, changes in solubility of the gluten in dilute acetic a d d were evaluated. As moisture content or tem perature of heating was increased, the rate of denaturation increased. Muller and Bale (Muller, 1969; Bale and Muller, 1970) evaluated gluten as a polymer, such as rubber. They found that gluten denatures over a range of temperatures starting at 50°C and extending for 40 more degrees by finding that there was an increase in the number of cross links with heating. The exact nature of these interactions was not discussed. Hansen et al. (1975) further investigated heat (108-174°C) and moisture (18-33%) effects on gluten through an examination of changes in solubility in urea, electrophoresis, and gel filtration. As the temperature was increased, there was a decrease in solubility followed by an increase when heating was to 174°C. The chromatographic separation depended upon the treatment conditions and was altered after heating relative to the unheated sample. The authors felt that the conditions experienced during baking, where the internal temperature does not exceed 100°C, would result in changes to the protein that would be less severe than those reported in the article. Jeanjean et al. (1980) used 0.5 M NaCl or 60% ethanol or one of three 0.025 M borate buffers with sodium stearate. Solubility decreased in 60% ethanol and salt, but increased in 1% mercaptoethanol/0.025 M borate buffer after heating. This change in solubility was thought to be due to hydrophobic interactions. Gluten, at 60% moisture, was heated to 90°C by Booth et al. 35 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (1980) and the changes in the protein examined by measuring changes in extractability of the gluten components in CUC. It was found that glutenin aggregated (with disulfide bonds being involved) resulting in diminished solubility, which was confirmed with starch gel electrophoresis. This work was extended to investigate further the involvement of disulfide bonding when gluten was heated (Schofield et al., 1983). Heating up to 75°C resulted in complete loss of functionality of the gluten and a decreased extractability in SDS-Tris buffer, which was due to polymerization of the glutenins, as found by gel filtration and propanol solubility. These methods showed a polymerization of the gliadin proteins when gluten was heated to 100°C. After measuring free sulfhydryl groups of gluten that had been heated to 75°C, the authors concluded that the changes caused by heating were due to sulfhydryl-disulfide interchange reactions of the glutenin proteins, not a loss of sulfhydryl groups. A subsequent report (Schofield et al., 1984) mentioned that at greater than 70°C gliadins appeared to polymerize. LeGrys et al. (1981) reported that when VWG, at 65% moisture, was heated above 55°C there were irreversible changes in the storage and loss moduli as measured with dynamic rheometry. Rogers et al. (1990) used a simple extraction in 1% SDS and up to 1.5% P-mercaptoethanol of gluten that had been reheated by convection or microwave methods. The authors found that MW reheating formed fewer disulfide cross-links between the protein chains. Weegels and Hamer (1991) found that when gluten at greater than 20% moisture was heated to 80°C, solubility in SDS and the level of free sulfhydryl groups decreased. Baianu et al. (1982) examined gliadins in great detail and gluten to a lesser extent by 13C NMR. These investigators saw a difference in the spectra when 36 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the wheat protein sample (32% w /v) was heated to 60°C and spectra recorded. More recently, Belton et al. (1987) investigated the effects of heating on gluten by 13C solution state NMR. Gluten was hydrated to about 60% moisture and a spectra was taken before and after heating to 80°C, and little change was found, but the authors felt they were only monitoring a small portion of the gluten. Gluten, gliadin, and glutenin (as separated by lactic add) were freezedried and hydrated with water for examination by 13C NMR (Ablett et a l., 1988) at 30°C and after heating to 80°C. In the unheated form, all three protein systems had very similar spectra, with heating having no effect. 2.5 Objectives of This Study There is a great deal of interest in using a MW oven to heat cereal-based foods. Little work has been done to examine these systems at the molecular level to understand the structural changes that have occurred. This research was undertaken to investigate the effects of MW heating on cereal-based systems and compare the results to those systems when CV heated. To eliminate complications and interactions, a model system containing the prim ary ingredients of a cereal product will be used. The experimental design used throughout the study will be based on five ratios of wheat starch:vital w heat gluten at three different moisture contents. Temperature profiles will be obtained during heating to determine the final tem perature to ascertain that protein denaturation and starch gelatinization are occurring. The texture of the samples after heating will be measured. Overall appearance and starch granule morphology will be assessed with SEM. An investigation of the components at the molecular level will be carried out both before and after heating to assess the effect of the heating method. Water will be examined 37 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. with PGSE-NMR to obtain self-diffusion coefficients. Dielectric properties will be determined. The gluten will be separated into fractions based on pH with HC1. Each of the fractions, after freeze drying, will be characterized by SDS-PAGE and 13C NMR. 38 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 3 MATERIALS AND METHODS 3.1 The Model Systems All moisture contents will be given as a percentage on a wet basis (i.e., (g w ater/g solids + water) multiplied by 100). The two dry ingredients in the samples were vital wheat gluten (VWG) (Whet-pro 80, Ogilvie Mills, Minneapolis, MN) [proximate analysis: 7.58% moisture, 76.5% protein, 0.15% fat, 0.41% ash] and wheat starch (Aytex-P, General Mills Chemicals, Inc., Minneapolis, MN) [proximate analysis: 10.8% moisture, 0.45% protein, 0.15% fat, <0.02% ash]. These two ingredients were used throughout the study in five ratios (100:0,80:20,50:50, 20:80, and 0:100) of starch:gluten. Each of these combinations was m ade to three different moisture contents 35%, 50%, and 65% (not including the water present in the sample). In general, to prepare all the samples, the dry powders were weighed, mixed, the appropriate amount of water added, and the ingredients thoroughly combined. Specifically, for the unheated samples, a 2-g ball was made by weighing the water with the dry ingredient mixture. For the heated samples, an 80-g ball was made, with the water measured volumetrically. Two samples could not be made, out of the fifteen, because of the hydration properties of the gluten. At 35% moisture, a homogeneous mixture of the 80 g sample of 20:80 and 0:100 starch:gluten compositions was not possible. It was possible with some effort to make a 2 g ball of these compositions for some of the measurements. 39 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3.2 Heating Methods For all experiments conducted in Minnesota, conventional (CV) and microwave (MW) heating was carried out in an environmental oven as described by Hung (1980). No air flow was used. Samples were p u t in a beaker with a diameter of 5.4 cm and height 5.2 cm. Conventional heating was done at 190°C for 20 min with the beaker sitting on a mesh platform suspended in the center of the oven cavity. Microwave heating was done at 400 W (2450 MHz) for 90 s with the beaker sitting on a teflon platform. The exception to the above heat treatment was when the samples were prepared for the self-diffusion coefficient measurements, which were done in New Zealand. At Massey University samples were placed in a 250 ml beaker (diameter 6 cm and height 9.5 cm) and heated in a household CV oven at 190°C for 25 min or in a household MW oven (700 W Panasonic), equipped with a turntable, at full power for 45 s. 3.3 Temperature Profiles Temperature profiles of samples prepared in Minnesota were obtained. During CV heating, type T thermocouples were connected to a recording device (Omega Engineering, Inc., OM400 Series Multichannel Data Logger, Stamford, CT) and the temperature in four positions in the sample were recorded each minute. An apparatus was used to try and provide consistency in the locations monitored, but the curvature of the wire and the expansion of the sample prevented precisely repeatable positions from being measured. During MW heating, four fiber optic temperature probes were used (Luxtron Model 750 Fluoroptic Temperature System, Mountain View, CA) which were interfaced to a computer to record the temperature in four locations in the 40 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. sample every 10 s. Again, due to difficulty in maintaining probes in an exact position, reproducible locations of temperature measurement were not always possible. 3.4 Water Loss After Heating and Cooling All samples were weighed before heating and after cooling for 30 min to determine the amount of water loss and how much water rem ained in the samples when the experiments were conducted. As well, the moisture content for eight of the samples (35% - 100:0,50:50; 50% - 100:0, 50:50, 0:100; 65% - 100:0, 50:50, 0:100) was determined using the vacuum oven method (AACC, 1983) using a four place analytical balance. The Al pans and lids were oven dried and kept in a desiccator prior to use. Calculations were made on a wet basis. 3.5 SEM Specimens were cut, with a razor blade, from the same locations as where the D values and dielectric properties were measured. Representative samples of the 3x3x5 factorial design were examined (35% 100:0, 50:50; 50% 100:0,50:50,0:100; 65% 100:0,50:50,0:100) before and after heating by both methods. The pieces were affixed to Al stubs with Ag paint and left to dry over CaCC>3 in a desiccator. The samples were coated with A u /P d in a vacuum evaporator and then viewed on a Philips SEM (model PSEM 500) at 12 kV. Micrographs were taken with a Polaroid camera. 41 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3.6 Texture Measurements The Instron Universal Testing Instrument (Model 1122, Canton, Mass.) with a 500 kg load cell was used to measure the force required to compress both the MW and CV heated samples (all thirteen compositions). The Ottawa Texture Measuring System test cell (Voisey et al., 1972) that was 30 cm2 was used to hold the sample so it could not move sideways, only down through the wire grid of the insert, as shown in Fig. 3.1. The nine wires of the insert had a diameter of 0.236 cm and were 0.333 cm apart. In order to be consistent, considering there were different physical characteristics of the samples, the piston on the shaft that was connected to the load cell, stopped 16 mm from the top of the rungs for all samples. The cross-head speed was 200mm/min and the paper moved at 100 m m /m in. The maximum peak height, in kg, was determined for each sample and the data statistically analyzed. Figure 3.1 30 cm2 Ottawa Texture Measuring System test cell 42 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The maximum force that could be measured without damaging the wire grid was 200 kg. Due to the pattern in the variance of the data, a transformation was necessary. The natural logarithm of the individual values was taken for the Analysis of Variance (ANOVA) (p=0.05) that was done on the Statistical Analysis System (SAS, 1987). The contrasts that are reported are on the actual data. 3.7 Determination of Self-diffusion Coefficients The complete 3x3x5 factorial study was done. Immediately after mixing, D values were obtained for the unheated samples. It was not possible to obtain data for one of the heated samples. At 65% moisture, the 100:0 sample could be easily made up, but it was too sticky after heating to insert into the NMR tube for measuring. Self-diffusion coefficients (D), m 2/s , were determined at 30°C using a JEOL FX60 pulse Fourier transform spectrometer w ith modifications described elsewhere (Callaghan et al.r 1980) using the method of Stejskal and Tanner (1965). 4 mm (inner diameter) NMR tubes were used for obtaining D values. For single component diffusion the ratio of the echo signal with and without the application of the pulse gradient (i.e., the echo attenuation) was written according to the Stejskal-Tanner relation (Stejskal and Tanner, 1965) as = exp(-kD) (1) where k= y282G2(A-8/3), y is the nuclear gyromagnetic ratio, 8 is the gradient pulse duration, G is the gradient amplitude, and A is the time separation of the two gradient pulses. In the experiments reported here, G was 4.72 T /m , A was 5 ms, and 8 was varied between 0.1 and 2 ms. Where two components are present, equation (1) may be modified to read 43 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. with An being the normalized amplitude of the component with diffusion coefficient Di, and (1-An) being the normalized amplitude of the component with diffusion coefficient D2 . Fits using this two component relation were carried out using a program based on a least squares method. The D values were obtained with a calibration accuracy of ±0.5%. All D values was analyzed by ANOVA (p < 0.05) for unbalanced data using SAS (1987). 3.8 Determination of Dielectric Properties A reflectance method was used to determine the dielectric properties with a Hewlett Packard Network Analyzer (Model 8752A) and a Hewlett Packard Dielectric Probe (Model 85070A). Microwaves at 2450 MHz were emitted from the end of the probe into the sample (at 22°C) and reflected back. These reflected waves were interpreted by the Network Analyzer to give data that was required to simultaneously calculate k' and k". The attenuation factor, a, could then be calculated with the following equation a (3) where X = 12.245 cm and tan28 is equal to the ratio of k" to k’ (von Hippel 1954). The data for k', k", and a were each analyzed separately with SAS (1987) for unbalanced data. 44 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3.9 Examination of Gluten Proteins 3.9.1 Fractionation The method of MacRitchie (1987b) was used to separate both unheated and heated (MW and CV) gluten. The following heated samples were chosen because they contain either all gluten (0:100) or enough gluten with sufficient starch (50:50) to examine the possible influence of the carbohydrate: 35% 50:50; 50% 50:50, 0:100; 65% 50:50, 0:100. A schematic of the fractionation procedure is shown in Fig. 3.2. The experiment was always started with a sample that contained 20 g of gluten in 400 ml 0.6 mM HC1 (pH = 3.6). For the samples that contained less than 20 g of gluten (i.e., 65%, 0:100), more than one sample was required, with a portion of the second sample used to obtain 20 g. If a sample contained more than 20 g of gluten, only a portion was used that contained about 20 g. For the 50:50 samples, there was a starch layer that appeared in the first centrifuge tubes, and this was scraped off and discarded. The MacRitchie method used nine extractions, but preliminary work showed that there was not much protein in the fractions greater than five, so the procedure was stopped there. 50 ml 0.6 mM HC1 was added to the final residue. The pH was recorded for all supernatants and the final residue (hereafter termed the "residue") after the addition of HC1. The solutions were adjusted to a final pH of 5.8 with 0.1 M NaOH (pH = 11.6). These solutions were p u t into plastic bottles, frozen in the blast freezer at -20°C for 1 1 /2 to 2 hours and freeze dried until all moisture was removed. The powder that resulted was stored at room temperature for the 13C NMR and electrophoresis experiments. The fractionation was duplicated for all 11 samples, and if any large discrepancy occurred between the initial pH's of the two replicates, the experiment was done a third time. 45 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 20 g gluten + 400 m l 0,6 mM HC1 Homogenize 2 m in \ / Centrifuge 20 min, 6000 rpm ^ If starch layer present, discard \ l supernatant 1 residue 1 pH to 5.8 w ith 0.1 MNaOH, lyophilice 200 m l 0.6 mM HC1 Homogenize 2 m in \1 / Centrifuge 20 min, 6000 rpm I supernatant 2 residue 2 pH to 5.8 w ith 0.1 MNaOH, lyophilice 200 m l 0.6 mM HC1 Homogenize 2 m in \l/ Centrifuge 20 min, 6000 rpm l / supernatant 3 residue 3 pH to 5,6 w ith 0.1 M NaOH, lyophilice 100 ml 0.6 mM HC1 Homogenize 2 m in \l/ Centrifuge 20 m in, 6000 rpm supernatant 4 residue 4 pH to5.8w ith0.1 M NaOH, lyophilice 100 m l 0.6 m M H Q Homogenize 2 m in \l/ Centrifuge 20 min, 6000 rpm l/ supernatant 5 pH to 5.8 w ith 0.1 M N aO H , lyophilice -1 final residue pH to 5.8 w ith 0.1 M N aOH, lyophilice Figure 3.2 Flow chart of fractionation procedure 46 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3.9.2 SDS-PAGE SDS-PAGE was used to examine each of the 60 products of the fractionation that were also examined by 13C NMR. A combined procedure of Ng et al. (1988) and MacRitchie et al. (1991) was used. The sample buffer was made up of 0.0177 M Tris-HCl (adjusted to pH 6.8 with 0.1 M NaOH), 1% w /v SDS, 20% w /v sucrose (to increase density of the solution), and 0.006% w /v Pyronin Y (a tracking dye). 250 pi sample buffer was added to 1 mg of the freeze dried gluten, left at room temperature, and vortexed several times over the course of the day. The samples were put in a refrigerator at 4°C overnight, vortexed the next morning and centrifuged at 10,000 rpm for 10 min; 15 pi was loaded into each lane of the gel. In order to have a 5% v /v (3mercaptoethanol solution, 12.5 pi of p-mercaptoethanol was added to the remaining 235 pi of sample buffer in the microcentrifuge tubes. The samples were then treated in the same manner as indicated above (Burnouf and Bietz, 1989). The BioRad Protean II Slab Cell system for electrophoresis (BioRad Laboratories, Richmond, CA) was used with a stacking and separating gel with a thickness of 1 mm. The buffers used for the gels were 1.0 M Tris-base, adjusted with concentrated HC1 to pH 6.8 for the stacking gel and to pH 8.8 for the separating gel. The composition of the stacking gel was 2.9% w /v acrylamide and 0.043% w /v bis-acrylamide. The composition of the separating gel was 17.6% w /v acrylamide and 0.083% w /v bis-acrylamide. A comb with 15 lanes was used. A BioRad power supply unit was used (BioRad Laboratories, Model 1000/500). The electrode buffer of 0.025 M Tris-base, adjusted to pH 8.3 with concentrated HC1,1.44% w /v glycine, and 0.1% w /v SDS was used when the 47 ■f Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. gels were run at constant current in a walk-in cooler (7.2°C) with the following conditions (for two gels): one hour at 40 mA, one hour at 50 mA, and two hours at 60 mA, or until the tracking dye reached the end of the gel. When the separation was complete, the gels were removed from between the glass plates and put in the stain which was prepared following the method of MacRitchie et al. (1991). The gels were left in the dye solution a total of 24-36 hours; the solution was changed once. As the background of the gels picked up quite a lot of color, the gels were put in a destain (10% v / v of 100% w /v TCA, 33% v /v methanol, and 57% v /v water) for about four hours. Nine standards were used at a concentration of each of 0.1 mg in 200 pi of the sample buffer, with only 10 pi of the standard solution loaded in a lane. The standards, from Sigma Chemicals (CITY) were (with molecular weight in brackets): lysozyme (14,300), (3-lactoglobulin (18,400), trypsinogen (24,000), pepsin (34,700), egg albumin (45,000), bovine albumin (66,000), phosphorylase b (97,400), P-galactosidase (116,000), and rabbit myosin (205,000). The three highest molecular weight standards did not migrate far and were not always included in the standard mixture. 3.9.3 Carbon-13 NMR Spectroscopy For each of the six lyophilized products of the fractionation procedure, that resulted from each of the five compositions after MW or CV heating (i.e., a total of 60 samples), a 13C spectra was obtained. For comparison, spectra were also obtained in the same manner for the fractions of the unheated gluten. Two solvents were required to dissolve all of the freeze dried fractions. The first three fractions of dry protein was dissolved in solvent A which was composed of 0.6 mM HC1,5% w /v sodium dodecyl sulfate (SDS), 48 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and 20% deuterium oxide (D2 O); pH = 4.3. The last three fractions were not soluble in this solvent, but were soluble in solvent B which was composed of 0.0178 M Tris-HCl buffer (adjusted to pH 6.8 with 0.1 M NaOH), 1% w /v SDS, 20% D 2 O, and 10% (3-mercaptoethanol; pH = 6.8. A ratio of 0.06 g of protein to 1 ml of solvent was desired, but not always possible. For some supernatants, there was less that 0.03 g of protein and a minimum of 0.5 ml of sample was required for the spectrometer. The solvent was added to the weighed protein and vortexed until completely mixed. The protein mixture was put in 5 mm NMR tubes and run soon after mixing. An NT-300 WB Nicolet Technologies, Inc. (General Electric, Fremont, CA) spectrometer (in the Biochemistry department) was used for the measurements at 22°C. The signal was locked on deuterium and all chemical shifts were m easured relative to deuterium. The signal was also proton decoupled. A single radio frequency pulse of 13 psec, which is about a 90° pulse, was used with a 1.5 sec delay between each of the 10,500 acquisitions, which took about 5 hours, that was necessary for each sample in order to obtain an adequate signal to noise ratio. Line broadening of 3 Hz was used before the Fourier Transform of the free induction decay. All spectra were plotted relative to the height of an SDS peak which had a chemical shift of about 25.58 ppm in solvent A and 26.00 ppm in solvent B. 49 •v r r r r*?. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 4 RESULTS 4.1 Physical Characteristics of the Samples The samples used in this study were mixtures of starch, vital wheat gluten, and water. While the samples contained the typical ingredients of a cereal product, their appearance was atypical, and depended on the composition. Figs 4.1 and 4.2 show what the samples typically looked like after heating in Minnesota. Although an attem pt was made to maintain the same conditions, the samples prepared in New Zealand were not exactly the same. The crumbly mixture of the high starch samples at 35% moisture became slightly more cohesive, although remained somewhat crumbly after heating. At 50% moisture after MW heating the center was an opaque gel, and the outside was a white powder; the CV heated sample had a white powder center and opaque gel outside, as was found by Goebel et al. (1984). Measurements that were taken in subsequent experiments were of the "gel" region. This same appearance was found by other investigators (Goebel et al., 1984; Zylema et al., 1985; and Wynne-Jones and Blanshard, 1986) when starch and water were MW and CV heated. The 50:50-35% moisture sample exhibited a layering of the dough. It is interesting that the MW heated samples resulted in a single air cell with a shell of dough around, and during CV heating numerous air cells resulted and the whole mass of dough expanded. As the proportion of gluten increased, the sample expanded during heating, and in some instances collapsed upon cooling. Some of the samples had large holes in the middle, others had holes more evenly distributed throughout. For some of the compositions, the samples were different and others were the same after being heating by both methods. 50 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 4.1 Conventionally heated samples. Figure 4.2 Microwave heated samples. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4.2 Heat Penetration Temperature as a function of time was recorded for all samples during both methods of heating. Each of the graphs in Figs 4.3-4.8 is a representative set of temperature profiles for each treatment and composition. Each figure has three or four plots per graph for different probes located in the same sample during one heating period. At 35% moisture content (Fig. 4.3) the CV heated sample heated more slowly than samples containing more moisture (Figs. 4.4 and 4.5). During MW heating all samples had similarly shaped curves indicating that the temperature increased in the same manner regardless of moisture content (Figs. 4.6-.8). At the lowest moisture content for both heating methods there are similar (more uniform) temperature profiles throughout the sample as shown by the agreement between probe temperatures (Figs. 4.3 and 4.6). Heating times (20 min for CV and 90 sec for MW) were such that all probes in all samples pass through starch gelatinization and gluten denaturation temperature (60-70°C). CV and MW heated samples reached 100°C to allow starch gelatinization and protein denaturation to occur throughout the entire sample, since probes only measure temperature at specific locations and cannot map the temperature profile of the entire sample. Even though the conditions were chosen for heating to maintain consistent experimental protocol the pattern of heating was not identical for all the samples. At 50 and 65% moisture content for the CV heated samples, the higher gluten containing samples heated more slowly than when there was more starch. The opposite phenomenon was observed for MW heated samples at 50 and 65% moisture content. This is surprising considering that the 0:100 sample at all three moisture contents had a lower dielectric constant (see 52 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Section 184.108.40.206) than the 100:0 sample. MW heated samples had more variability in temperature between probes within a sample as the moisture content increased. In Figs. 4.4 and 4.5 for conventionally heated samples, it may be noted that some of the temperatures are above 110°C. This was due to the sample rising in the beaker during heating which shifted the thermocouples causing them to come out of the sample and measure the surrounding air. 53 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100 60 - 0 5 10 15 20 0 5 10 15 20 100 U o 0u) 3 3u> 3 QJ JX g01 100 0 5 10 15 20 time (min) Figure 4.3 Representative temperature profiles for CV heating of 35% moisture content samples with all five starchrgluten ratios. A) 100:0, B) 80:20, C) 50:50. 54 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100 ■D 100 B 100 uo 0 5 10 15 20 0 5 10 15 20 100 100 0 5 10 15 time (min) 20 Figure 4.4 Representative temperature profiles for CV heating of 50% moisture content samples with all five starchtgluten ratios. A) 100:0, B) 80:20, C) 50:50, D) 20:80, E) 0:100. 55 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100 10 15 20 100 0 ll <8 I* a> Cu £u < 5 10 15 20 5 10 15 20 100 3 0 ■C 5 10 15 20 100 0 time (min) 0 5 10 15 20 time (min) Figure 4.5 Representative temperature profiles for CV heating of 65% moisture content samples with all five starch:gluten ratios. A) 100:0, B) 80:20, C) 50:50, D) 20:80, E) 0:100. 56 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100 0 30 0 30 60 90 30 60 90 60 90 100 U o ' —■. <U 3 100 0 time (sec) Figure 4.6 Representative temperature profiles for MW heating of 35% moisture content samples with all five starch:gluten ratios. A) 100:0, B) 80:20, C) 50:50. 57 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100 100 B 100 u o <TJ 0J cx, £0) 60 - 0 30 60 90 0 30 60 90 100 • 60 C 100 time (sec) 0 30 60 90 time (sec) Figure 4.7 Representative temperature profiles for MW heating of 50% moisture content samples with all five starchrgluten ratios. A) 100:0, B) 80:20, C) 50:50, D) 20:80, E) 0:100. 58 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100 100 0 30 60 90 100 3 flj i_ a a. £o 0 30 60 90 0 30 60 90 100 0 30 60 90 100 time (sec) 0 30 60 90 time (sec) Figure 4.8 Representative tem perature profiles for MW heating of 65% moisture content samples with all five starch:gluten ratios. A) 100:0, B) 80:20, C) 50:50, D) 20:80, E) 0:100. 59 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4.3 Moisture Content After Heating Table 4.1 gives the moisture contents, as determined by the vacuum oven method, on a wet basis of the center part of representative samples. This is the area which was measured for the self-diffusion coefficients and dielectric properties. Nearly all samples were within ±3% of the original target moisture, and all samples were within ±6%. Some of the samples appeared to gain m oisture in this area possibly due to macromolecular transitions that required water. Wherever gelatinization occurred in the all starch samples, it required a great deal of water which left some regions fo the sample looking dry and white (see Figs. 4.1 and 4.2). This water distribution resulted in moisture contents that were higher than expected. Table 4.1. Moisture contents of the center of representative heated samples, on a wet basis, as determined by the vacuum oven method. M oisture Content After Heating C onventional Microwave Heated Heated M oisture Content Before Heating Composition (S:G) 35% 100:0 37.61 37.67 50:50 40.87 33.66 100:0 52.95 55.17 50:50 52.69 48.63 0:100 52.02 47.49 100:0 60.76 67.88 50:50 61.85 64.39 0:100 66.40 64.17 50% 65% 60 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4.4 SEM Figures 4.9 to 4.12 are representative scanning electron micrographs of the samples that were examined. The discussion that follows includes observations of the samples that were not always photographed. Fig. 4.9 shows scanning electron micrographs of the unheated samples. The starch granules are very distinct in the 100:0 (Fig. 4.9 A, C, F) and 50:50 (Fig. 4.9 B, D, G) samples, with the latter containing smooth regions of gluten interspersed between the regions of starch. The samples m ade only from VWG (Fig. 4.9 E, H) have very little contrast and exhibit a smooth surface. The micrographs of the CV heated samples at 35% moisture content (Fig. 4.10 A, B) and MW heated (Fig. 4.10 C, D) samples show a loss of granule integrity. With the gluten present there was development of holes in the sample during heating. The change in the morphology of the starch granules relative to the unheated starch (Fig. 4.9 A, B) is evident. The granules are no longer smooth and have become swollen and misshapen. For the 100:0, 50% moisture content sample, the gel region (not the dry white region) was examined by SEM (Fig. 4.11 A, D). There was enough moisture in the sample to allow gelatinization to occur as seen with the loss of shape by nearly all the granules, and the formation of a continuous mass. Once gluten was added (50:50, Fig. 4.11 B, E) there were smooth regions to interrupt the granules that were not gelatinized by either heating method. The all gluten samples (Fig. 4.11 C, F) were very similar for both CV and MW - a very smooth topography interrupted by numerous holes. At 65% moisture there were more distinct granules in the CV heated sample than the MW heated one (Fig. 4.12 A, B), although the swelling was 61 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. greater than the lower moisture content samples. Gluten did not seem to affect starch swelling (Fig. 4.12 C, D). The CV heated sample had more distinct granules than the MW heated one. The two heating methods resulted in similar appearing sample for 0:100 (Fig. 4.12 E, F). SEM has shown that starch definitely did swell extensively (thus gelatinize) during both CV and MW heating at all three m oisture contents. One can assume that if the temperature and moisture conditions allowed the starch to transform, the gluten was also able to denature (i.e., undergo the normal transitions that occur during heating). 62 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 4.9 Representative scanning electron micrographs of unheated samples. A) 100:0, 35% moisture B) 50:50, 35% moisture C) 100:0, 50% moisture D) 50:50, 50% moisture E) 0:100, 50% moisture F) 100:0, 65% moisture G) 50:50, 65% moisture H) 0:100, 65% moisture 63 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 4.10 Representative scanning electron micrographs of 35% moisture samples. A) 100:0, conventionally heated B) 50:50, conventionally heated C) 100:0, microwave heated D) 50:50, microwave heated 64 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission Figure 4.11 Representative scanning electron micrographs of 50% moisture samples. A) 100:0, conventionally heated B) 50:50, conventionally heated C) 0:100, conventionally heated D) 100:0, microwave heated E) 50:50, microwave heated F) 0:100, microwave heated 65 Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 4.12 Representative scanning electron micrographs of 65% moisture samples. A) 100:0, conventionally heated B) 50:50, conventionally heated C) 0:100, conventionally heated D) 100:0, microwave heated E) 50:50, microwave heated F) 0:100, microwave heated 66 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4.5 Instron Data The Instron data was recorded as the force (kg) to compress the sample. A representative curve is shown in Fig. 4.13A. Average values of force required to compress samples after CV and MW heating are shown in Fig. 4.14. Although the target weight was 80 g, the actual weight of the samples before they were put in the oven was about 79 g due to errors in measuring the water and losses in transfer from the mixing vessel to the baking vessel. Table 4.2 gives the average weight of the samples that were compressed. Most samples lost less than 5 g to a maximum of 10 g. There was some moisture loss during heating which will decrease the after heating weight. As well, not all of the sample can be removed from the beaker and put in the test cell for the compression. In general, the CV heated samples lost more than those that were MW heated. When the weights in Table 4.2 are compared to the moisture contents of Table 4.1, it can be concluded that most of the moisture loss was from the top surface of the sample, not from the center. When the force required for compression of the sample was statistically analyzed, the values were transformed by taking the natural log, to reduce the variability in the residuals to satisfy the assumption for the ANOVA test. From the ANOVA table, for both type III and type IV sums of squares, which account for unbalanced data and missing cells respectively, all main effects and interactions were significant (P<0.05). The mean square of the error was 0.182 for the transformed values, which converts to 1.2 kg for the actual force. The three-way interaction is shown in Fig. 4.14 which is composed of three plots, one for each moisture content. 67 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0.3 0.2 (0 0> at Ut 0.1 0.0 9- 0 20 10 30 strain Figure 4.13 Represenative results from Instron compression. A) Curve of force versus time for a sample compressed with the OTMS test cell. B) The values of the curve in (A) plotted as stress versus strain. 68 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100:0 80:20 50:50 20:80 0:100 200 150 100 50 100:0 80:20 5050 20:80 0:100 Composition (S:G) Figure 4.14 Effect of composition on the force required to compress heated samples. Cross-hatched = conventionally heated; solid = microwave heated. A) 35%, B) 50%, and C) 65% moisture. 69 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4.2. Average weight of samples after heating, before compression with the Instron (n = 4 to 10). Weight (g) of sample Conventional Microwave Heated Heated M oisture Content Composition (S:G) 35% 100:0 71.2 77.2 80:20 73.6 70.7 50:50 74.8 73.9 100:0 68.3 75.9 80:20 70.7 75.4 50.50 74.1 76.9 20:80 74.1 74.5 0:100 74.1 74.6 100:0 68.6 75.3 80:20 70.1 75.0 50:50 69.0 75.2 20:80 73.5 75.9 0:100 73.2 76.6 50% 65% When one compares these force values to the weight of the sample that was compressed, it is evident that a larger sample does not require a greater force as one might suppose. An inverse relationship between moisture content and force is clearly evident; as the moisture content increases the force required for compression decreases. The influence of the heating method on the force is not straight forward, but rather dependant on the composition with the effect being variable. At 35% moisture (Fig. 4.14A) for 70 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the CV heated samples, the force could only be quantified for the 80:20 and 50:50 samples because the force was greater than the limits of the apparatus (200 kg). At the lower moisture contents (35 and 50%) composition has a greater effect on force; that is as the amount of gluten increases in the samples, generally the force required for compression increases. This is consistent with the experiments reported by Lelievre et al. (1987) for bread of up to 16% protein that also included various wheat starches. All compositions by either heating method had very similar force requirements for compression at 65% moisture (Fig. 4.14C). The only sample in which MW heating resulted in a tougher product than CV heating was the all starch sample at 50% moisture. Another more classical way of examining the texture of foods has been used. Attenburrow et al. (1989) suggested a cake could be thought of as being a sponge based on the work of Ashby and Gibson (1983) where compression causes the walls to be bent elastically and then collapse. This behavior can be observed in curves of stress vs. strain. Stress is defined as the force divided by the initial cross-sectional area and stress is defined as (the am ount of sample compressed divided by the initial thickness) times 100 (Attenburrow et al., 1989). If the force data for some of the samples is plotted in this manner (Fig. 4.13B) the phenomena can be observed to a limited extent. There appears to be a change in the curve, for the higher gluten samples that contained air cells, when the cells were collapsed. Compression of the samples has shown that moisture content has the largest effect on texture, and at lower moisture contents (35% and 50%) an increase in the proportion of gluten in the sample results in greater force for compression. Heating method is more important for the high starch samples 71 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. at 35 and 50% moisture. At 50% moisture, the CV heated all starch sample is softer than the MW heated sample. For the 65% moisture sample, the water dominates the force and all samples are very soft. The interactions between the components that translates into texture are not straight forward (an interrelationship between composition and heating method) and starch and gluten behave differently under the conditions examined. 4.6 Self-diffusion and Dielectric Property Data 4.6.1 The Three Ingredients Separately Each of the dry ingredients of the model system was examined separately to investigate how the water interacted with the starch and gluten. This data was compared to the results of the same experiments on water (Table 4.3). Table 4.3. Moisture contents, self-diffusion coefficents, and dielectric properties of the components of the model system. Moisture Content W heat Starch W heat G luten W ater 10.79% 7.58% 100% Self-Diffusion Coefficient (m2/s) not available 2.6 x 10-12 3.0 x 10-9 Dielectric Constant 2.23 2.46 76.92 Dielectric Loss 0.23 0.13 9.38 Attenuation Factor 0.0390 0.0212 0.274 The D for the proton nuclei in the VWG powder was about three orders of magnitude lower than water (Table 4.3), indicating a very limited motion of the water molecules. Although the starch had a higher moisture content 72 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. than the VWG, a proton NMR signal could not be obtained, preventing a determination of a value of D. If the effect of transverse relaxation is neglected, the magnitude of a signal should be proportional to the number of contributing nuclei. One simplistic explanation for the lack of starch proton NMR signal might be that the VWG contained a greater number of nuclei than the starch. The percentage of protons, as a total of all atoms present in the sample on a molecular weight basis, was calculated. For the calculation of the protons in the VWG, an amino acid profile (Table 4.9) was used so that the percent of each amino acid that was proton, the percent of the total sample that each amino acid comprised, and the protons on the starch (16.9%) could be used. The starch macromolecular composition was taken to be repeating glucose units and the entire sample was assumed to be starch and water. The results of these calculations were that 6.9% of VWG and 6.7% of starch was proton, which is not a difference that would be great enough to cause the lack of of signal with the starch. A more likely hypothesis is that the water molecules in the "dry" starch are more closely associated with the rigid polysaccharide structure and therefore their protons have very short transverse relaxation time, making them "invisible" in the present experiment (i.e., below the monolayer). Possibly the converse is the case for the VWG (i.e., the water is present at a level above the monolayer). From the isotherms (Fig. 4.15) the BET equations gave a monolayer value of 0.046 and 0.070 g w ater/g solid for VWG and starch, respectively which correspond to aw's of about 0.08 and 0.10. These data agree with previous studies (Bushuk and Winkler 1957; Volman et al. 1960; and French 1984). The monolayer values, on a dry basis, are not quite half the moisture contents for both the VWG (0.082) and starch (0.120), which 73 R eproduced with permission o f the copyright owner. Further reproduction prohibited without permission. could be interpreted to mean that the water would be in about the same "layer" on the macromolecules. What must be considered is the distribution of this water. Possibly the water that interacts with the protein is located in specific regions, where it is present in many layers thus there would be water nuclei that are quite mobile and could give a signal. Bryant and Shirley (1980) found that water at a protein surface does exhibit some motion and is not rigidly held. For the carbohydrate it is possible that the w ater is more evenly distributed and thus is too immobile to give a signal. Although, according to French (1984), for starch the water will not be evenly distributed at the monolayer and not all the glucose molecules would have a water molecule (0.66 molecules water/glucose unit). Volman et al. (1960) thought this amount of water would be shared between the glucose molecules and fit well into the amylose helix, where it would likely be quite immobile. 2 0.3 X S ta rc h ♦ Vital W h e a t G luten ♦ "o X (0 D) u O 0.2 X *5 5 ■ D) 0.1 ‘ X X . x ^ * • • • 0.0 0.0 1 1. . 0.2 . 1 0.4 w a te r »...... 1 _1 0 .6 . 1 0 .8 1.0 a c tiv ity Figure 4.15 Sorption isotherms for dry ingredients of model system at 23 °C. 74 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The dielectric properties of the starch were comparable to those of gluten (Table 4.3) and both dry powders had values much smaller than water. As expected, these values indicate that for both dry powder samples there was very little interaction of the water with the electromagnetic radiation or dissipation of the electrical energy into heat energy. The small am ount of mobility observed with the D value for the water in the gluten was not enough to allow movement of the dipole to any great extent. 4.6.2 The Mixed Samples 220.127.116.11 General Information on the PGSE-NMR Experiments For all the mixtures, except for the 100:0-35% moisture sample after heating, two values for self-diffusion were evident from the plot of log(echo attenuation) vs. 82(A - 8/3) as shown in Fig. 4.16. O LU LU £ -2 -3 0 1 2 8 2( A - S / 3 ) 4 5 6 (msec3 ) Figure 4.16 Plot showing the effect of increasing gradient pulse duration on echo attenuation for the PGSE-NMR experiment. 75 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The signal, as shown in Fig. 4.16 was composed of a faster (diagonal line), more free population of water and a slower (horizontal line), more inhibited population. In all cases, the signals for the two populations appeared to have the same chemical shift, except for the 0:100-35% moisture sample, when the two signals could be resolved. The NMR spectrum of this sample was examined with and without the field gradient being on and then compared with a pure water peak. These results showed that the peak with the same chemical shift as the water was most attenuated when the field gradient was put on which indicates that the population of water most similar to pure water is attenuated most during the experiments. The other signal is a slower moving component that has a D value of about an order of magnitude (or less) slower than the fast component of that sample, indicating it is interacting more closely with the macromolecules so the frequency of the NMR signal is shifted away from that of water. It was found that an alteration of the length of time between the field gradient pulses resulted in no considerable change in the less bound value of D. However, the values of D for the more bound water did change, so it cannot be well quantified. The reliability of this slow component value was minimal, but general comments can be made. For the unheated samples, the contribution of the slow component to the signal was between 2% and 25%. For each composition, the proportion of the slow component diminished with increasing moisture content. At the higher moisture contents, m ost of the signal came from the more mobile water, and the amount of water interacting with the macromolecules was a very small portion of the total water in the sample. For the heated samples, the slow component contributed between 4% and 30% of the total signal. The moisture content 76 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. did not have as great an effect on the proportion of the signal between the fast and slow component after heating as it did in the unheated samples. The 50% and 65% moisture content samples were quite similar in the proportion of the slow component and the 80:20 CV heated was the same for all three moisture levels. In the above discussion, it m ust be remembered that these "fast" and "slow" components are relative terms and are for a mixed population. The Brownian motion of the slow diffusing group of w ater molecules is more inhibited by macromolecular interactions than the fast diffusing group, with the latter being inhibited to some extent because the D values are less than that for free water. The number of "layers" of water that are contributing to each component cannot be determined. The discussions of the NMR data that follow will be only on the faster more mobile component behavior. The one anomaly of the samples was 100:0-35% which only had one population after heating. This could mean that all of the protons were in one environment producing one signal. Another explanation for the behavior is that the exchange between the populations was faster than the delay between the gradient pulses, thereby preventing both populations from being resolved. The statistical analysis indicated that the three main effects (composition [starch:gluten ratio], moisture content, and treatment [unheated, CV or MW heated]) and all the interactions were significant (P<0.05) for D and the three dielectric properties. 77 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 18.104.22.168 Unheated Samples The D value of the dough varied significantly (P < 0.05) with the amount of water (Fig. 4.17); the more water that was present, the higher proportion of more mobile protons, and the greater the diffusion coefficient measured. When the smallest amount of gluten was added to the starch (80:20 sample), the value of D decreased, with the effect being more dramatic at the highest moisture content. On further addition of gluten, D was relatively constant. 25 0 100:0 I______i______I_____ 80:20 50:50 I______i______L 20:80 0:100 wheat starchrwheat gluten Figure 4.17 Self-diffusion coefficients for unheated starch and vital wheat gluten mixtures at three different moisture contents: 35% (■), 50% (x), and 65% (o). This suggests that the water-gluten interaction was "saturated" at low gluten contents and the subsequent D value of the water as gluten was increased was dominated by this interaction. Further evidence for this hypothesis is given by the fact that the D value for water in the 50:50 starch sample at the highest moisture content was very close to that of free water. The proton mobility of the hydrated starch and VWG was contrary to the mobility of the dry 78 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. powders. That is, in the dry starch the water was very tightly held, and when more water was added, it did not interact with the macromolecules and remained quite mobile. In gluten dry powder water had some mobility, and when more water was added it interacted with the protein to result in a D that was less than starch with the same amount of water. However it is important to emphasize that in the dry powders the mobility of the water, which is bound to the macromolecules, reflects the motion of the polymer rather than the degree of binding or interaction. In the case of the protein, it may well be that the water more strongly binds to the polymer but that the polymer is more rotationally mobile than that of the starch. At higher water contents where the water is in exchange between "free" and "bound" states, the protein will, therefore, exert a greater attenuation of the translational diffusion of the water because of this stronger attraction. The dominance of gluten on the signal and the decrease in mobility in the presence of the macromolecules was determined by ESR experiments using the probe TEMPO in water with starch (Pearce et al., 1985) or gluten (Pearce et al., 1988). When the dielectric properties were measured, it was found that moisture content had a greater effect on the dielectric constant (Table 4.4) than on the dielectric loss (Table 4.5). Table 4.4 Dielectric constant for unheated samples (standard deviation = 5.45; p < 0.05, n = 5-16). _______________Starch:gluten ratio_______________ Moisture Content 35% 50% 65% 100:0 24.40 41.37 55.15 80:20 22.71 33.08 54.27 50:50 20:80 0:100 22.82 33.27 40.38 21.78 31.39 38.56 18.61 30.73 38.10 79 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4.5 Dielectric loss for unheated samples (standard deviation = 1.47; p < 0.05, n = 5-16). Starch :gluten ratio Moisture Content 100:0 80:20 50:50 20:80 0:100 35% 50% 65% 7.08 8.95 9.46 7.05 7.77 9.74 7.18 8.75 8.21 6.99 8.76 8.11 5.61 8.81 8.23 As moisture content increased, the sample had increased ability to rotate and couple with the electromagnetic energy but the k' values did not approach the k' of water. As the amount of protein increased in the system there was a decrease in k', and the water appeared to interact more with the macromolecules, especially at the highest moisture content. For these compositions, the system appears not to change over the three moisture contents; as more water was added there was no change in the ability of the sample to transform the electrical energy to heat energy as indicated by the k" behavior which is quite similar to water. The attenuation factor (Fig. 4.18) followed the same pattern as k" but the values were all greater than the value for water. Starch did not appear to absorb as much water as gluten because when equivalent amounts of starch or gluten were mixed with water at the higher moisture contents, the former was liquid and the latter was not. Thus, for the unheated mixtures, the physical appearance may provide an indication of the molecular behavior, but not necessarily. 80 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100:0 80:20 5 0:50 20:80 0:100 wheat starchrwheat gluten Figure 4.18 Attenuation factors for unheated starch and vital wheat gluten mixtures. 35% (■), 50% (x), and 65% (o). 22.214.171.124 Heated Samples The decrease in D with heating (from the unheated samples) is evident in Fig. 19. The effect of gluten on the signal that was evident in the raw dough data was minimized during heating due to the redistribution of water from the gluten to the starch or because the free water became bound to the macromolecules. Possibly the starch transformations during heating (gelatinization), which would destroy the rigid structure, allowed greater interaction between the starch macromolecules and the water. As expected, the denaturation of the gluten did not appreciably change water behavior as is evident by the 50% gluten samples (Fennema 1977). For most of the heated samples moisture content had a significant effect on these results as it did for the unheated samples. The largest difference between the self-diffusion coefficients for the MW and CV heated samples was the 80:20-65% moisture sample. The slight moisture content differences between the CV and MW 81 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. heated samples reported in Table 4.1 were not enough to result in any differences in self-diffusion. At 50% moisture for the all-starch sample, there was a large difference between the dough and the CV heated sample. This is contrary to the ESR investigations reported by Pearce et al. (1985) for starch (67% moisture) where there was no difference in the motion of the spin probe (TEMPO) after heating. For the all-gluten sample at 50% moisture, there was a significant difference between the unheated and heated samples that was not evident when the moisture content was 65%. When gluten was examined by ESR at about 50% moisture, it was found that after heating there was an increased partitioning of the probe into a hydrophobic region of the sample (Pearce et al., 1988). The motion of water measured by a spin probe is indirect and the self-diffusion measurement is direct, which could explain the difference in the results between the two methods. It has been reported, based on water adsorption studies, that in a dough more of the water is associated with the starch than the protein (Bushuk and Hlynka, 1964) thus there would be more hydrogen atoms associated with starch-interacting water than with protein-interacting water. This appears contrary to w hat was found in this study, where the gluten water appeared to dominate the signal. During heating, the water redistributes itself when the protein denatures and the starch gelatinizes absorbing water given up by the protein (Bushuk and Hlynka 1964). When the samples were heated, there was no composition effect indicating that there m ay be an even distribution of water between the macromolecules. The self-diffusion data reported here support this idea. 82 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 25 20 15 10 5 o :_ 100:0 80:20 50:50 20:80 0:100 Pi Q 100:0 80:20 50:50 20:80 0:100 wheat starch:wheat gluten Figure 4.19 Self-diffusion coefficients for (A) conventional and (B) microwave heated starch and vital wheat gluten mixtures. 35% (■), 50% (x), and 65% (o). Heating generally caused a decrease in the k’ (Table 4.6) and k" (Table 4.7) relative to the unheated samples. A redistribution of the water may have occurred in the samples during CV heating as all three moisture contents did not have very different k' and k" values. The changes in the sample that occurred during CV heating, changed the ability of the water to couple with 83 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the microwaves and transform the electrical energy to heat. MW heating did not cause the same type of changes in the system as CV heating, and the moisture content had more of an effect on k' and k". Composition played a large role in the MW heated samples, especially at 50% moisture. The water appeared to interact more strongly with the denatured gluten than the gelatinized starch as the high protein samples had lower k' and k" values than the high starch samples at 65% moisture. If one examines the moisture content values after heating in Table 4.1 there is some explanation of the k' and k" phenomena observed. At 35% moisture, for 100:0 the two samples have the same moisture content, and similar k' and k". When gluten was added, (50:50) the CV sample had a greater moisture content and higher k' and k". Moisture is not the sole determinant of the dielectric properties when gluten is present as the heated values are less than the unheated. Table 4.6 Dielectric constant for heated samples (standard deviation = 5.45; p < 0.05, n = 5-16). Moisture Content 100:0 Conventional Heated 35% 16.54 50% 29.43 65% 41.55 Microwave Heated 35% 17.66 50% 29.15 65% 49.13 * not available Starch:gluten ratio 80:20 50:50 20:80 0:100 24.03 20.52 29.09 19.70 32.85 35.22 n/a* 31.66 31.78 n /a 21.06 28.84 5.77 22.30 43.12 8.76 1.01 36.61 n /a 2.14 27.70 n /a 20.41 24.30 84 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4.7 Dielectric loss for heated samples (standard deviation = 1.47; p < 0.05, n = 5-16). M oisture Content 100:0 Conventional Heated 5.21 35% 50% 10.09 65% 11.48 Microwave Heated 35% 50% 65% * not available 6.49 7.55 9.95 Starch:gluten ratio 80:20 50:50 20:80 0:100 7.59 6.42 7.84 5.57 8.20 8.34 n/a* 7.53 5.88 n /a 5.03 6.05 1.65 5.84 9.32 2.71 0.24 7.85 n /a 0.18 6.22 n /a 5.36 5.48 At 50% the macromolecular transitions are very important for the dielectric properties because moisture content cannot be used to explain them. For the samples with a before heating moisture of 65%, moisture content after heating does appear to provide an indication of the dielectric properties. The attenuation factor after heating (Fig. 4.20) is similar to the values before heating for all 65% samples and the 100:0, 35% and 50% samples. Although not exactly the same, the pattern of the attenuation factor plots follows more closely k' than k". It is very clear that MW heating has a much different effect on the transformations in the sample that determine how the electromagnetic waves are attenuated. The interaction between composition and moisture content is clearly evident for the MW heated samples. The effects of heating, which allowed macromolecular transformations, were evident by the decrease in D value which indicated an increased binding 85 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and redistribution of the water between the starch and gluten, with the starch having a greater influence. There was not m uch difference between the CV and MW heated samples in terms of their D values. For all five ratios of composition, as moisture content increased there was generally an increase in D (the mobility of the water) and k' (polarity of the system) but k" (a measure of the ability of the system to dissipate energy) was not affected as much by moisture content. The k' and k" values of the heated samples were dependent on the ratio of starch to gluten, w ith the exact effect dependent on moisture content. The attenuation factor indicated that the two heating methods did cause different changes in the system in the way in which the sample interacted with the electromagnetic energy. 86 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0.5 0.4 0.3 0.2 E u 0.1 L. O 4-> 0.0 L_ a 100:0 a o 80:20 50:50 20:80 0:100 80:20 50:50 20:80 0:100 0.5 a 0.4 0.3 0.2 0.1 0.0 L_ 100:0 wheat starchrwheat gluten Figure 4.20 Attenuation factors for (A) conventionally and (B) microwave heated starch and vital wheat gluten mixtures. 35% (■), 50% (x), and 65% (o) 4.7 Qualitative Information from the Fractionation Procedure Besides providing a means of separating the w heat proteins into smaller more homogeneous units, additional information about the effects of heating and sample composition was obtained by measuring the pH 's of the supernatants (Table 4.8) 87 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4.8 pH values during HC1 fractionation procedure. M oisture Starch: M ethod Content G luten of Ratio Heating 1 2 3 4 5 residue Unheated 5.3 5.1 4.6 4.2* 3.8* 3.7* CV 5.3 5.0 4.8 4.6 4.1 4.2 MW 5.3 5.0 4.6 4.4 4.2 4.2 CV 5.3 5.1 4.6 4.4 4.1 4.1 MW 5.4 5.3 4.9 4.7 4.3* 4.3* CV 5.3 4.7 4.0 3.9 3.7 3.7 MW 5.3 4.8 4.0 3.8 3.7 3.6* CV 5.4 5.2 4.9 4.7 4.2 4.2 MW 5.4 5.2 4.9 4.7 4.3* 4.2* CV 5.3 5.0 4.4 4.2 3.9 4.0 MW 5.4 5.2 4.6* 4.1* 3.8 3.8* 35% 50% 50% 65% 65% 50:50 50:50 0:100 50:50 0:100 Initial pH of Each Supernatant3 a = average of duplicate experiments; generally the agreement was within 0.1 pH unit, to a maximum of 0.3. * = some protein precipitated as the pH was increased to 5.8 with 0.1 M NaOH. From Table 4.8 it is evident that as the extraction procedure progressed, the buffering capacity of the proteins diminished and the pH of the supernatants after centrifugation decreased. The dissolved proteins were able to raise the 3.6 pH of 0.6 mM HC1 to a maximum of 5.4. When fewer proteins are dissolved the pH approaches that of the solvent. These data for VWG are in agreement with the behavior of gluten by Eliasson and Lundh (1989) and of 88 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Yecora Rojo (a Californian "good quality") flour that was observed by j MacRitchie (1991). Both the pH trend of the supernatants and the solubility of ! the protein in five fractions are the same; this behavior is unique relative to j the other five flours that MacRitchie (1991) examined, but was also found by Eliasson and Lundh (1989). SI had a pH of 5.3 or 5.4 for all 11 samples. The j i other four supernatants and the residue had pH's that differed depending on j i heating method and composition. The 0:100, 65% MW heated sample most closely resembled the unheated VWG in the pH values and precipitation behavior. All three 50:50 samples when compared within one heating method had quite similar pH values. Moisture content does appear to have an effect at the higher hydration level because for the 0:100 samples, within a heating method, the 65% moisture fractions have higher pH's than the 50% fractions. The presence of starch in the sample appears to have an effect on how the protein separates. At 50 and 65% moisture, within a heating method, all supernatants (except SI) had higher pH's if starch was added to the sample than when it was not. This shows that there is a fairly strong interaction between starch and gluten that affects the buffering capacity and therefore the solubility of the protein. Or, the denaturation is affected by the presence of starch. The starch results in a protein with a higher buffering capacity which would relate to the charge on the protein. The electrostatic interactions of the amino a d d of the protein will affect the charge behavior (or pKa). (See Fig. 4.21 for structures of the amino adds and their three letter codes.) The amino acid that can be ionized are (in decreasing concentration in VWG): glu, asp, arg, tyr, his, and lys. Glutamic acid composes 30% of VWG and the other ionizable amino acid are present at less than 5%, so glu is going to be the 89 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ! dominating factor in the charge (and thus the solubility) of the protein. There is a possibility that all the starch was not removed after the first centrifugation and that its' presence as it is carried through the extraction procedure is affecting the behavior of the protein. This topic will be discussed further in section 4.9. Heating method appears to have an effect on the pH values. At 35% moisture there does not seem to be much difference in the behavior of the proteins between MW and CV samples. When the moisture is increased to 50%, the gluten that has been CV heated has slightly lower pH values than the MW sample for all supernatants. It is also interesting that the MW heated S5 and residue had some proteins that precipitated at about pH 5.5 and 5.0, respectively. Although the supernatant pH values are higher for the MW heated samples than the unheated sample, the precipitation is similar to the unheated behavior. For the all gluten sample at 50% moisture, there is very little difference between the CV and MW heated samples, except the precipitation of the MW heated residue at about pH 5.2. At 65% moisture for the 50:50 sample, unlike the 50% sample, there is hardly any difference in the pH values between the CV and MW heated samples, except, the precipitation of MW heated S5 and residue at pH 5.5 and 4.7, respectively. 90 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 4.21 Structures of the amino adds and their three letter codes X I I I I I o O X I 0— 0* r. 91 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The pH differences are a little greater between the CV and MW heated samples with a composition of 65%, 0:100. The precipitation behavior of the MW heated sample is unique to all the samples studied because S3 had some protein that precipitated, as did S4 (pi about 5.0), and the residue (pi about 4.8). Only the MW heated samples (except 35% moisture) precipitated any protein which is the behavior exhibited by the unheated gluten. Another aspect of the fractionation procedure that can be examined is the physical appearance of the residues after each centrifugation. W hen starch was present in the sample, it was less dense than the gluten and could be found on top of the protein in the centrifuge tube. This is contrary to the results of Mauritzen and Stewart (1965) and Graveland (1980) who found the starch at the bottom of the centrifuge tube. The reason for this could be due to the different methods used, with different solvents, which would affect the hydration of the starch and gluten altering their density. For nearly all the samples the gluten sediments were compact and beige in color. The exceptions were all the samples at 65% moisture and the unheated gluten. After the second centrifugation, the all subsequent residues were not dense (pudding-like) and could hold a lot of liquid. The dry powders that resulted after freeze drying of the first fraction was a fluffy white powder and the fractions became progressively more beige with successive extractions. SI and residue had the most protein. Other supernatants contained only 0.06 g or less of protein. According to MacRitchie (1987a) the early fractions contain mostly gliadins and the latter fractions the glutenins. However, in another publication (MacRitchie, 1991) it was shown by electrophoresis that each fraction contains a m ixture of gliadins and glutenins. This was also found by Eliasson and Lundh (1989). 92 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The residue proteins appeared to have the greatest buffering capacity and required the largest amount of base to increase the pH to 5.8, which could only be because there was a large amount of protein in a small volume of HC1, relative to the supernatants. As the pH was increased some of the supernatants either became slightly cloudy or white as the conformation of the proteins changed in the solution to refract light differently. This phenomena was also found by Bernardin et al. (1967) for a-gliadin. Ewart (1980) found that the intrinsic viscosity decreased slightly at a higher pH, but was not dependent on pH in the range 4.1 to about 7 (i.e., with or without acetic add). Thus, the size of the molecule in solution appears to not effect the way the solution refracts light. Kasarda et al. (1968) found that a-gliadin decreased in helix content and ellipticity (as measured with circular dichroism) when the pH was increased from 3 to 5. This change in shape could be because the carboxylate ions become protonated in this range resulting in an unfolding of the polypeptide. 4.8 SDS-PAGE From the electrophoretic gel of the fractioned unheated gluten (Fig. 4.22) the dramatic effect of p-mercaptoethanol is evident by the greater number of bands in the six lanes on the right side of the gel. There are not m any bands of protein when the disulfide bonds are not broken, which agrees with the work of MacRitchie (1991). This indicates that the proteins were m ade up of large proteins held together with disulfide bonds, and when they were broken the individual units could travel in the electric field and separate. There was not a lot of difference in the molecular weight of the proteins between the fractions. 93 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 4.22 Electrophoregrams of unheated fractions. Lane 7-standards. Lanes 1-6 - no P-ME, lanes 9-14 - p-ME. Fractions were loaded in lanes as follows: Sl-1 & 9; S2-2 & 10; S3-3 & 11; S4-4 & 12; S5-5 & 13; R-6 & 14. 94 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 14,300 18,400 24,000 34,700 45,000 66,000 97,400 116,000 205,000 90*60 Figure 4.23 Electrophoregrams of fractions from 35% moisture samples that were heated. CV (lanes 2,4,7,9,11, 13) and MW (lanes 3,5,8,10,12,14). Lanes 1 & 6-standards. Fractions were loaded in lanes as follows: Sl-2 & 3; S2-4 & 5; S3-7 & 8; S4-9 & 10; S5-11 & 12; R-13 & 14. 95 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 4.24 Electrophoregrams of fractions from 50% moisture samples that were heated. A) 50:50 and B) 0:100 starch:gluten ratio. _ — — — 14,300 18,400 24,000 34,700 — 45,000 — 66,000 — 97,400 — 116,000 — 205,000 A) Lanes 5 & 12 standards. CV - lanes 1,3,6,8,10,13. MW - lanes 2,4,7,9,11,14. Fractions were loaded in lanes as follows: Sl-1 & 2; S2-3 & 4; S3-6 & 7; S4-8 & 9; S5-10 & 11; R-13 & 14. 96 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. B) Lane 5 standard. CV - lanes 1,3/6,8,10,12. MW - lanes 2,4,7,9,11,13. Fractions were loaded in lanes as follows: Sl-1 & 2; S2-3 & 4; S3-6 & 7; S4-8 & 9; S5-10 & 11; R-12 & 13. 97 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure 4.25 Electrophoregrams of fractions from 65% moisture samples that were heated. A) 50:50 and B) 0:100 starch:gluten ratio. 14,300 18,400 24,000 34,700 45,000 97,400 116,000 205,000 A) Lanes 3 & 8 standards. CV - lanes 1,4,6,9,11,13. MW - lanes 2,5,7,10,12,14. Fractions were loaded in lanes as follows: Sl-1 & 2; S2-4 & 5; S3-6 & 7; S4-9 & 10; S5-11 & 12; R-13 & 14. 98 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 14,300 18,400 24,000 34,700 45.000 66.000 97,400 116,000 205,000 B) Lanes 5 & 12 standards. CV - lanes 1,3,6,8,10,13. MW - lanes 2,4,7,9,11,14. Fractions were loaded in lanes as follows: Sl-1 & 2; S2-3 & 4; S3-6 & 7; S4-8 & 9; S5-10 & 11; R-13 & 14. 99 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Supernatants 1 and 2 have a very prominent band at 45,000 which is more faint for S3 and gone in the last three fractions. Based on the categorization of Fullington et al. (1987) this band would be composed mostly of LMW glutenins, with possibly some a , p, and y gliadins. S2 and S3 have some low molecular weight components (about 18-20,000) which could be albumins or globulins. There are a variety of other bands that appear in different fractions in the 66-97,000 range. This indicates that proteins of the similar molecular weight will separate into different fractions. The fifth supernatant and residue lanes have very few bands, even though the same amount of protein was loaded as all the other lanes. Despite the addition of P-mercaptoethanol, the proteins were not broken down to be able to separate. Eliasson and Lundh (1990) found SI and S2 to be similar by SE-HPLC, and this was also found by this electrophoresis work. The authors also found S3, 4, and 5 to be similar and that was not observed with PAGE. S3 has bands at about 18,400 and about 97,000 that are not present in S4 or 5. The addition of P-mercaptoethanol to the sample buffer caused the two LMW bands in S2,3, 4, and 5 to become one band. Also, the very large glutenin molecules that did not move from the origin in the absence of pmercaptoethanol were broken into units with a molecular weight > 97,000. Overall there was more information about the protein composition of a fraction to be obtained when P-mercaptoethanol was added to the sample and no information lost by omitting the gels with no P-ME; only those electrophoregrams with p-ME will be shown. The P-ME did have an effect on the standards and cause the bands to have an upside down "V" or arrow shape. 100 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. After either CV or MW heating of the 35% moisture sample (Fig. 4.23) there was a change in the distribution of the gluten among the fractions from the unheated samples (Fig. 4.22). The effect of heating gluten on its electrophoresis patterns has been demonstrated for other extraction procedures (Booth et al., 1980). The residue shows the greatest effect of heating with many more bands compared with the gel from the unheated sample. The presence or absence of bands in the lanes shows that each fraction is composed of the total range of molecular weights and there are differences between where the proteins were soluble. SI and 2 are very similar (with SI having more LMW proteins at about 14,000) as are S3, 4, and 5 with the latter having an extra band at 66,000. The residue has different bands than the supernatants between 40,000 and 70,000. The gels were loaded such that the supernatant from the CV and MW heated samples were side by side (CV on the left, MW on the right) to facilitate comparison of the effect the heating method had on protein solubilization. At 35% moisture, with starch present, no difference in the way the proteins fractioned was evident. Denaturation did occur based on the temperature profiles and the difference in the PAGE results from the unheated gel. But the heating method did not alter the way the proteins were denatured as evaluated by SDS-PAGE. When the moisture was increased to 50% (Fig. 4.24A), the gel shows much of the same characteristics of the 35% sample; the bands look similar and the supernatants within a heating method can be grouped the same. An exception from the grouping is S5 which has a slightly different band pattern than S2 and 3 in the six bands at about 66,000. Unlike the 35% moisture samples, at 50% the additional water is resulting in some differences between 101 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the MW and CV heated samples, except the residue. The HMW proteins are absent or very faint from all the CV supernatant lanes where there are four bands present in the MW lanes. There is another band at the same distance as the point in the arrow of the 66,000 standard band that is absent from the CV heated lanes. At 50% moisture, the MW heating had a different effect on how the protein denatured, relative to CV heating, so that the proteins were soluble at different pH's. When VWG was heated in the absence of starch at 50% moisture (Fig. 4.24B), the separation of the proteins was very different than when starch was present. S4 and 5 have fewer bands, and the pattern of the bands in the residue is different. The lanes do not look like the unheated (Fig. 4.22) so again, the heating affected the solubility of the gluten in HC1. The various components of gluten are spread throughout all the fractions. Like the 50:50 sample, this 0:100 sample had gluten that denatured differently depending on the heating method. The HMW bands (about 97,000) are absent from the lanes of the CV sample of all five supernatants. S2 of the CV heated does have much more prominent bands between 45,000 and 66,000 than the MW heated sample. The CV heated S4 and 5 sample has prominent bands at 45,000 which are absent from the MW heated lanes, but the latter does have some slightly larger molecular weight proteins. The CV S4, S5, and MW S5 lane all have a large protein at about 116,000. The CV residue has more bands between 45,000 and 66,000 than the MW heated. At 65% moisture for the 50:50 composition (Fig. 4.25A) the bands are more clearly defined than the same composition at 50% moisture. The differences in bands present and the dye intensity shows that each fraction has some differences in its composition, but all contain part of the total gluten mixture. 102 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. The SI fractions are predominantly composed of proteins with a molecular weight of about 45,000 which is mostly LMW glutenins, with possibly some a, (}, and y gliadins and some HMW glutenins. With this high m oisture sample of equal amounts of starch and VWG, the CV and MW heated lanes appear almost identical. It appears the proteins separated in the same m anner and thus were denatured the same. There is some difference in the intensity of the bands from S5 between heating methods. After CV heating, the bands are much lighter and appear almost absent at < 45,000 and some bands between 66,000 and 97,000. The band pattern is different between the gels of 50 and 65% moisture, indicating that the moisture content does have an effect on how the protein denatures. When no starch was present at 65% moisture (Fig. 4.25B), there was greater difference between the supernatants than when starch was present. There are also differences between the CV and MW heated samples that did not occur when starch was present. SI is very similar whether from MW or CV heating. S2 is very similar also, but the sample from MW heating appears to have two extra bands at the point of the 66,000 arrow and at 97,000. S3 after MW heating has two extra bands side by side at the point of the 66,000 arrow and the bands that are of next smallest molecular weight are m ore prominent than after CV heating. The MW S4 and S5 fractions have m any more bands than the CV of all molecular weights from 45,000 to 97,000. The residue is very similar after both heating methods. 4.9 13C NMR The solvents used for dissolving the proteins for the 13C NMR experiments needed to be investigated for solubilizing properties. It was 103 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. initially hoped that the HC1 used to separate the proteins could be used for the NMR experiments since it contained no carbon atoms. But the proteins were not soluble in HC1 and formed a gel. The addition of SDS to the HC1 (solvent A) prevented gel formation and dissolved the first three fractions. W ith the latter fractions, a gel was formed with this solvent, as well as many other typical gluten solvents and dimethyl sulfoxide. The sample buffer used for electrophoresis did not cause the proteins to form a gel and was adapted for this purpose (solvent B). The solvents were run alone to identify their peaks (Fig. 4.26) In solvent A, the only carbons present were part of the SDS [SC>4 (CH2 )iiC H 3 ] and the spectra was similar to that of Creamer and Richardson (1984). In solvent B the Tris-HCl [C(CH20H)3NH3C1] and p-ME (SHCH2 CH 2 OH) also appear in the spectra between 10 and 70 ppm. It is possible that the binding of the detergent to the protein may alter the mobility of the carbon atoms when compared to protein mobility without SDS. The detergent was necessary for protein solubilization and was present at a concentration about half that required for complete saturation of the protein (Helenius and Simons, 1975), permiting carbon atoms to rem ain mobile. For the most part the lyophilized sample was soluble in the solvents and resulted in a clear or opaque solution that varied in color from clear to white and sometimes had a brown hue. The color did not appear to affect the spectra. In a few instances the samples had other colors. Fractions 4 and 5 of CV heated 50:50, 35% had a red hue just after vortexing, but turned beige after a few seconds of rest. This phenomena also occurred at 50% moisture for CV 50:50 (S5), 0:100 (S4, S5), and MW 0:100 (S4, S5, R). Another unusual color behavior was exhibited by some of the CV heated 0:100 samples: 50% and 65% S4 and S5 were dark green after the NMR acquisition. 104 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A Jl UL 50 100 150 PPM \ JJjl : 150 | 10 0 i i i i . | 50 ; JLj « i I PPM Figure 4.26 Sample spectra of solvents used for 13C NMR experiments. A) solvent A and B) solvent B. 105 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4.9.1 Spectral Assignments To determine if there was any kind of interaction that would alter the chemical shifts from the published values (Breitmaier and Voelter, 1990) amino ad d s were examined in the solvents. (See Fig. 4.21 for structures of the amino acids.) Many of the amino acids were not soluble in either solvent, but cys, gly, and ser were examined in solvent A; pro was examined in both solvents. These data indicated that the solvents had no influence on the chemical shifts. Thus, literature values could be used to aid in identifying the carbon atoms responsible for the peaks. From the amino acid profiles obtained for the gluten, the amino acids that w ould be present in greatest amount with the greatest chance of producing the spectra were known. For the samples, the solvent peaks were of course largest, but by plotting the spectra on an expanded scale the sample peaks were enlarged. Quantitation of peak size, which would provide information on the amount of substance that was contributing to the signal, was not possible since there was not always a constant initial concentration. An attempt was made to always have the same initial concentration, but the nature of the fractionation and the amount of dry powder required for a run, did not always make this possible. Table 4.9 on the following pages provides information on the presence or absence of non-solvent peaks from the six spectra (one from each fraction) for each treatm ent and composition. Sample spectra are provided in Appendices V to VII. In all samples there were three peaks that were present in every sample and they were not included in the table. The carboxylic acid side chain of glu appeared at 177.5 ppm and was fairly large and broad indicating that 106 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. there were a num ber of different environments that the group was experiencing. This is expected, as glu is the m ost abundant amino a d d in the protein. Near the glu peak another was found at about 173.6 ppm, for the other carboxylic ad d s in the protein. The peak was broader and smaller than the glu peak. Probably these groups were relatively more immobilized and are spread out in more diverse environments. The next peak which appeared in all the samples had a chemical shift at about 129.3 ppm and was assigned to the ring side chain of phe, with possibly some influence of tyr. There was another rather rare peak that appeared at 42.9 ppm only in the spectra of supernatants 1 and 2 of the unheated samples. It is likely that this peak represents the Ca of gly, Cp of leu, an d /o r Cg of arg. At this point the rest of the peaks in Table 4.9 will be assigned as best is possible, then the differences between samples will be discussed. Exact assignment is somewhat difficult; the assumption was m ade that some chemical shifts have occurred for the protein relative to the pure amino acid and carbon atoms within ±2.0 ppm were chosen. Other investigators have published spectral assignments that do not completely concur (Baianu et al., 1982; Belton et al., 1987). The amino acid assignments are taken from table 5.25 of Breitmaier and Voelter (1990) and the carbohydrate assignments are based on those of Mora-Gutierrez and Baianu (1991) for corn syrup solids. Between 70 and 80 ppm the peaks were too numerous to try and identify each one. 107 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. vA 'W # ~— i— i— |— i— i— i— i— |— i— i— i— |— i— i— i— 150 100 50 PPM 150 100 SO PPM Figure 4.27 Examples of sample 13C spectra. A) CV heated 35% moisture SI and B) 50% 0:100 MW S2. 108 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4.9 Peaks present in 13C spectra from all samples. The residue is indicated by "R" in the Frac. no. column. A '*+" indicates the peak was present; a indicates the peak was not present in the spectra. There were three peaks, not included in this table, that appeared in all spectra at: 129.3, 173.6, and 177.5 ppm. See text for details of all spectral assignments. Sample Unheated Frac. no. 1 2 3 4 5 R 35% ,50:50 1 CV 2 27.8 31.9 47.9 53.1 61.5 + + + + + 91.9 95.7 99.4 + + + + - - - 70-80 + + + + + - - + + + + - - - - + - + + - - - - - + - + + - - - - - + - - + - - - - - + + + + + » + + + + + + + + + + + + + 3 - + + + + » + + + + 4 - - + + + + + + + 5 - - + + + + + + + - - + + + » +* - - + a + + + + + » + + + + + + + + + + + + + R 35% ,50:50 1 MW 2 3 + + + + + » + + + + 4 - - + + + + + + + 5 - - + + + + + + + - - + + + » +* - - +a R " » " = the peak was much larger than the peaks at 53.1 and 47.9. = the peaks in the 70-80 range were fewer than usual. "a" = the peak had a chemical shift of 101.2 ppm. 109 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Frac. no. 26.8 31.9 47.9 53.1 61.5 70-80 91.9 95.7 99.4 50% ,5 0:50 1 - - - + + » + + + + CV 2 + - + + + - + + - + S a m p le 3 + + + + + + + + + + 4 - - + + + + - 5 - - + + + + + + + - +a R 50% ,50:50 1 MW 2 3 - - + + + » +* - + + + + + » + + + + + + + + + + + + + + + + + + + + + + 4 - - + + + + + + + 5 - - + + + + + + + R 50% ,0:100 1 CV 2 3 - - - + + +* - - +a + + + + + » + + + + + + + + + +* - - - + + + + + - - - - 4 - - + + + - - - - 5 - - - - + - - - - R - - + + - - - - - + + + + + + + + + 50% ,0:100 1 MW 2 + - + + + - - - - 3 - - - + + - - - - 4 - - - + + - - - - 5 - - - + - - - - - - + + - - R - - - - " » " = the peak was much larger than the peaks at 53.1 and 47.9. = the peaks in the 70-80 range were fewer than usual. "a” = the peak had a chemical shift of 101.2 ppm. 110 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Frac. no. 26.8 31.9 47.9 53.1 61.5 70-80 91.9 95.7 99.4 65% ,50:50 1 + + CV 2 44- 4 -» 44444- 44- 44- 44- 44- - - - - - - - - - - - - 4-* 4444- - - 4- + 4- 4* 4- 4444- 4-» 444- 4-* - - 4* 4* 4-a 4- 4- 4- 4- - - - - - - - - - - - - Sample + + 3 + + 4 - - 5 - - 4~ 4444- R - - - + + + 4* 444444- 4444- 4444- + 65% ,50:50 1 MW 2 3 + + 4 - - 5 - - R - - 65% ,0:100 1 + + CV 2 + + 3 + + 4 - - 4" 44- 4+ + 44- 4 -» 44 -» 4444 -» 444- + 4- Not enough sample to obtain a spectra. 5 R + - - 4- 4- 4- - - - - 65% ,0:100 1 + + 2 + 4- 4 -» 44444- + MW 444444- 4444- 4444- 4444- - - - - - - + + 4 - - 5 - - 44444- R - - - 3 + 444- " » " = the peak was much larger than the peaks at 53.1 and 47.9. = the peaks in the 70-80 range were fewer than usual. "a" = the peak had a chemical shift of 101.2 ppm. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4.10 Chemical assignments for peaks of fraction spectra. Chemical Shift (ppm) Chemical Assignment 26.8 31.9 47.9 53.1 60.5 70-80 91.9 95.7 99.4 Cp -arg, -cys, -glu, -his, -lys; Cy-ile, -leu Cp -val; Cpy -met; Cs -lys Cs-pro Ca -ala, -arg, -leu Ca -ile, -thr, pro, -val; Cp -ser C2 ,3,4 of oligosaccharides, amylopectin, pyranose or furanose forms of fructose a-reducing end of oligosaccharides P-redudng end of oligosaccharide Ci amylopectin 4.9.2 Discussion of 13C NMR Data In the unheated gluten sample the small amount of residual starch is evident in the spectra (Table 4.9) of the first fraction. All the carbohydrate must be soluble at pH 5.3 (or bound to proteins that are soluble at that pH) because it is not evident in the spectra of the subsequent fractions. This result from a mass of hydrated gluten is contrary to Belton et al. (1987) who did not find any mobile starch carbons with solution 13C NMR spectrometry . A reason for this discrepancy may be that there was too much protein signal that masked the carbon signal, whereas with the fractioning procedure used in this study the starch became concentrated in the first fraction. There was no heat applied during the fractionation procedure that would have gelatinized the starch to produce mobile carbohydrate carbon atoms, so they must be inherently mobile in the starch granule. For some reason there is very little 112 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. mobility in the protein of the residue so not many peaks are evident and the few that are there are not very large. At 35% moisture the CV and MW heated samples are very similar, and show some differences from the unheated spectras. With regard to the protein, the peak at 61.5 ppm is much larger than its neighbors at 53.1 and 47.9 ppm in SI and S3, and in the other heated fractions (and the unheated fractions) the three peaks were similar in size. There are five different amino acids that could be contributing to the signal at 61.5 ppm and it cannot be definitively said which one has increased in mobility. But, it can be said that denaturation upon heating, even at low moisture (35%), resulted in some changes in protein folding that resulted in increased mobility. In the residue, this peak was quite small, indicating a loss of mobility of that side chain. Another interesting observation is that carbohydrates are present in all of the fractions. It appears that there is an interaction between the gluten and the starch during heating that causes the starch to stay with the protein through all the pH's of the fractionation procedure. As shown in Table 4.8 this interaction affected the solubility which resulted in a pH change from the unheated gluten. The interaction is not so strong as to inhibit the mobility of the carbon atoms, because they are able to give sharp spectral peaks. In the residue, the group of peaks between 70 and 80 ppm was reduced in number relative to the other fractions (see spectra in Appendices V to VII). There was not the usual group of three peaks between 90 and 100 ppm for carbohydrates but rather only a single peak slightly upfield at 101.2 ppm. This peak represents another compound which is possibly the C-l at the non-redudng end of amylopectin (Mora-Gutierrez and Baianu, 1991). 113 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Contrary to 35% moisture, at 50% for the same composition, there were some differences between the CV and MW heated samples in both the protein and carbohydrate peaks. Moisture content is having an effect on the separation of the components to the greatest extent when the sample was CV heated. Some peaks were different between the protein part of the spectras for each heating method in fractions SI and the residue. The carbohydrate behaved quite different if it was CV or MW heated. There were a num ber of peaks that did not appear in the CV heated spectra. Thus, the kind of interaction that was observed at 35% moisture between the protein and the starch did not occur with CV heating at 50% moisture. At 50% moisture for the 0:100 samples, the CV and MW heated spectras are more similar than when the starch was there. There are a few peaks that are different between the protein portion of the spectra. In the carbohydrate portion of the spectra, the two heating methods resulted in almost identical products which were very similar to the unheated sample. The starch that is inherently present in the VWG is mostly soluble in the first fraction. For the 50:50 sample at 65% moisture, there was some difference in the carbohydrate component of the spectra, but little difference in the protein component, between the CV and MW heated treatments. After CV heating, all the carbohydrate that could be resolved solubilized in the first two supernatant fractions, with just a small amount evident in the residue. For the MW heated sample, carbohydrate was found in all the supernatant fractions. Another small difference in the protein between the two heating methods was the greater mobility, or more carbons contributing to the signal at 61.5 ppm in S3 after MW heating than after CV heating. This higher moisture content appears to result in different interactions between the starch 114 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and protein after CV heating than were found at 50% moisture. Like SI for the 35 and 50% samples, the peak at 61.5 ppm was much larger than its two nearest protein peaks at 53.1 and 47.9 ppm. For the 0:100, 65% moisture sample there was little difference between the protein component of the spectras after each heating method. However it was surprising that the small amount of inherent starch did interact with the protein and was carried through supernatants 1 to 4 after MW heating only. There was no difference in the chemical shifts between the MW and CV heated sample, which indicates that the carbohydrate interacted with the protein in a region that did not contain mobile carbons that were contributing to the spectras. Or, because SDS breaks non-covalent bonds it could have interrupted the protein-carbohydrate interaction when the powder was solubilized in solvent A or B, thus breaking the interaction and chemical shift effects could not be observed. One can assume from the scanning electron micrographs that most of the starch did gelatinize at both moisture contents, since a few intact granules could be seen. For some reason the starch at the highest moisture content interacted strongly with the protein and was carried through the fractions. It is not likely that the starch was charged so that it could be soluble by itself at the different pH's. Nor would there be concentration effects on solubility because the amount of starch present was much greater in the 50:50 (relative to the 0:100) 65% sample in which all the starch was solubilized in the first two fractions. There are a couple of hypotheses why the starch was present in four of the 65% 0:100 fractions. Possibly at 50% moisture the water was all absorbed by the gluten and there was little left for the starch to fully gelatinize and interact with the protein, but at 65% the event did occur. Another reason could be 115 > TVy~,irm.,rv,.VIB.,^1 .. . Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. that at 65% moisture there was enough water not absorbed by the gluten that it could be free in the granule to interact with the microwaves. If the microwaves were able to focus within the starch granule and cause enough heat to be generated there to promote gelatinization, the granules may have expanded and resulted in carbohydrate molecules that interacted with the protein and were carried through the fractions. At 50% moisture, the granules may have received their energy conductively from the gluten and the macromolecules in the starch would not interact with the gluten in the same way. In some of the residues, after freeze drying there were distinct white and beige powders. For the most part they were mixed together as they composed the total residue. To check the composition of the two entities from the CV 50:50, 50% sample, the white and beige powders were examined separately by 13C NMR. The spectra of the white powder did not contain any of the usual protein peaks at about 130 or 170 ppm, but did contain the carbohydrate peaks between 70-80 and 101.2 ppm. The spectra of the beige powder only contained the protein peaks and no carbohydrate peaks. In this sample, carbohydrate was soluble in SI and to a more limited extent in the other fractions. But there was also some starch that was never soluble and was carried through all the procedure to end in the final residue. 4.10 Selected Amino A dd Profiles To try and understand some of the different behavior that was observed with the fractionation, SDS-PAGE, and 13C NMR data, amino acid profiles were obtained for S5 and the residue of the four 65% heated samples and the unheated gluten, (with the exception of MW 0:100 S5 for which there was not 116 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. enough lyophilized protein for the analysis). These data are given in Table 4.11 which also gives the percent amino acid in the unfractioned unheated VWG, for comparison. About half of the amino acids are within 0.5% of the values for gluten found by McMaster and Bushuk (1983) and half are not. The notable differences are for glu and pro; the VWG has much lower amount of glu and more pro. The differences are most likely because two different varieties of wheat from unknown growing conditions are being compared. For the most part, the percentage of each amino acid in the fractions is close to the unfractioned value. Three exceptions are cys and thr which are generally lower and tyr which is higher in the latter sample than the fractions. It can be inferred that proteins that contain cys and thr m ust be soluble in the earlier fractions and those that contain thr are not soluble. When one examines all the columns, CV, 0:100, S5 has the highest percentages of glu, phe, and pro which is countered by the sample having the lowest levels of most of the other amino adds (except cys and ile). Contrarily, the unheated residue has the highest level of many amino acids (ala, arg, asp, gly, his, lys, met, thr, and val) and the lowest level of glu and phe. The amino ad d profiles were only done once so statistical analysis is not feasible. A difference of 0.5% is being taken as an arbitrary value, for the purposes of this discussion, as being large enough that it may have some influence on the protein behavior. To determine if a particular amino acid had a preference for where it solubilized a comparison of S5 and residue of one treatment and composition can be made. In the heated samples there is greater difference between S5 and residue than there is for the unheated samples. For the sake of clarity on the 117 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. table, only the differences between the two samples from the same fractionation procedure are shown. A comparison of S5 and the residue with four amino adds finds the same trends for the unheated and heated samples, which agree with the results of MacRitchie (1987a). There is more glu and pro and less asp and lys in S5 than the residue. Examination of S5 and residue of the 50:50 CV, 50:50 MW, or 0:100 CV show a more variation in the distribution of all the amino acid in the protein. This indicates that the denaturation at the highest moisture content affects the distribution of the amino a d d in these two fractions, which implies different protein solubility. One also wants to compare the same fractions after CV and MW heating to determine if the carbohydrate binding is correlated to the presence of an amino acid. This can be done by comparing the residues of the 50:50 samples with those of the 0:100. The two heating methods have almost no effect on the amino acid profiles (except for glu) when the carbohydrate is present, but when there is no carbohydrate, the heating method does influence the amino acid profiles. The 13C NMR peak at 47.9 ppm could best be attributed to pro. Because this peak is present in S5 of CV 50:50, 65% and not the residue, one would expect that there would be more pro in S5. This is the case based on the amino a d d profiles of Table 4.11. However, if the same comparison is made for the MW heated sample, the pro peak is present in both samples, yet the amino acid profiles indicate that S5 has much more pro than the residue. That is, the two residues have the same amount of pro, but the carbon atoms contributing to the signal are much more mobile after MW heating than CV. Thus, one can conclude that the amount of amino acid in the sample will not always indicate the presence of the carbon atoms in the spectra. 118 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Table 4.11 Percentage of amino acids in select fractions. The heated samples had been made up to 65% moisture before heating. The "<" and ">" signs between two columns indicates that those two values are different by 0.5%. sample percentage of each amino acid in the fractions CV- M W CV- M W CVtotal Unhtd Unhtd 50:50 VW G ala arg asp cys glu co s*y his ile leu lys met phe pro ser thr tyr val 50:50 50:50 50:50 CV- MW- 0:100 0:100 0:100 S -5 residue S -5 S-5 residue residue S -5 residue residue 3 .92 4.01 < 4 .8 5 3.28 2.92 4.13 4 .2 0 2.48 3.88 < 4 .4 2 2.61 2.71 < 3 .2 0 2.21 2.18 2 .67 2 .5 7 1.70 2.61 2.88 3.19 3.66 < 4 .7 3 2.98 2.76 3.85 3 .8 7 2.38 3.55 < 4 .2 3 1.28 1.14 1.28 1.30 0.95 0.98 1.13 1.35 1.32 3 4.72 < 35.69 3 1.58 > 3 0 .8 3 38.77 3 1.80 > 29.15 < 7 .2 8 0.71 3 3 .3 8 31.29 > 2 8 .1 7 5.95 7.46 7.71 6.33 > 5 .7 4 7.00 7 .16 4.43 6.47 1.96 1.72 1.97 1.52 1.41 1.63 1.81 1.26 1.75 1.90 3.54 3.69 4.02 4.10 3.81 3.80 4.14 3.54 3.95 3 .9 5 7.04 7.47 7.70 6.81 6.67 7.58 7 .6 6 6.51 7.61 7.80 1.40 1.69 < 2 .4 3 1.34 1.15 1.84 1.89 0.94 1.57 < 2 .1 4 1.37 1.44 1.50 1.19 1.06 1.11 1.26 0.58 1.58 1.49 4 .1 9 3.96 3.98 4.71 5.06 4.09 3 .9 7 5.52 4.05 4 .08 15.17 < 16.54 12.63 1 2.68 17.99 13.02 > 12.23 5.60 5.64 5.59 5.29 < 4 .6 6 5.75 5 .6 6 4.55 5.68 5.50 2.76 3.63 3.99 3.17 3.03 3.77 3 .7 6 2.38 3.49 3.72 2.60 2.45 2.17 2.25 2.11 1.73 1.81 1.77 2.31 2.22 < 5 .6 9 4.10 3.90 5 .28 5 .4 0 3 .7 7 5.15 5.43 14.88 4 .5 9 12.48 > 10.86 5.14 4.11 Overall Discussion From the fractionation, the MW heated samples at 50 or 65% exhibited precipitation behavior that resembled unheated gluten, whereas the CV heated samples did not precipitate. These data indicated the gluten interacted w ith the starch. What affect this interaction has on the protein solubility remains to be investigated. The fractionation is a sequential process and the pH of the solution that the protein was soluble in does not predict the composition of those proteins. The difference in the behavior of the carbohydrate solubility indicates that starch may also be affected differently by the heating method in the presence of gluten. This would explain why the patents to modify cereal products have influenced both the starch and gluten. As the only sample that exhibited a firmer texture in the MW heated products were the 100:0 and 80:20 at 50% moisture, it leads one to suspect starch as being an important component influencing texture. Further research on this 80:20 sample would be of interest to investigate how the starch is affect by both heating methods. 120 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 5 CONCLUSIONS A model system of wheat starch and vital wheat gluten at three moisture contents was examined to determine how the samples were affected by CV and MW heating. Temperature profiles showed that the temperature was high enough to allow starch to gelatinize and gluten to denature. The rate of heating overall was much faster with the MW conditions used than with the CV conditions. MW heating rate was not dependent on moisture content as was CV heating. At 50 and 65% moisture, high starch samples heated at a faster rate in the CV oven than high gluten samples; the samples behaved in the opposite manner in the MW oven. SEM illustrated the starch gelatinization that occurred when the samples were heated. Because the samples were heated in a beaker, there was little surface area for moisture loss, so the center of the sample remained close to its before heating moisture content. Texture is an important aspect of a food which can be related to the force required to compress a sample. There is a great deal of interest about the effects of MW heating on food texture. When the model system samples were extruded between the wires of an OTMS, test cell heating method, moisture content, and sample composition were all found to be significant factors in the force that was required. Moisture content appeared to be the predominant factor. There was an inverse relationship between the force and the amount of water in the sample, with all samples at 65% moisture having about the same texture. There was enough water that it dominated over the effects of the macromolecules , and prevented them from forming a firm texture, resulting instead in a soft product. At 35 and 50% moisture, as the 121 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. amount of gluten in the sample increased, so did the force. The CV heated samples required more force for compression than the MW heated samples, except the 100:0 and 80:20 samples at 50% moisture. An 80:20 starch:gluten ratio sample with a moisture content of 50% is the sample most like a typical bread dough. The self-diffusion coefficient measurements found that there were two main populations of water in a cereal type product. The effects of heating were evident by the decrease in D value for all compositions, which indicated an increased binding, and redistribution of the water between the starch and gluten. The energy from the increased temperature and the macromolecular transformations allowed the water to move within the sample. Another interesting conclusion is that there was not much difference between the CV and MW heated samples in terms of their D values. The types of transitions that occurred during the heating process that caused the differences in water binding were not affected by the rate of heating. For all five ratios of composition, as moisture content increased there was generally an increase in the k' (polarity of the system) but k" (a measure of the ability of the system to dissipate energy) was not affected as much by moisture content. The k' and k" values of the heated samples was dependent on the ratio of starch to gluten, with the exact effect dependent on moisture content, k' and k" are measuring two different aspects of the system, that include both water and macromolecular motion, k" had reached a threshold in the unheated samples at 35% moisture and the changes that occurred during heating were complex (an interrelation between composition and heating method) in their influence on k". The attenuation factor indicated that the two heating methods did cause different changes in the system in the w ay in which the 122 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. sample interacted with the electromagnetic energy. This would be due to differences in macromolecular motion as well as water behavior. A group of experiments was performed that examined the gluten in 50:50 and 0:100 samples. First the gluten was fractionated with dilute HC1. This procedure allowed the proteins to be separated into six groups with similar solubilities. The pH measurements of each fraction dem onstrated the effect starch can have on the denaturation of the gluten. When the p H was increased with the addition of NaOH, some of the proteins in the latter fractions that had been MW heated (except 35% moisture) precipitated, behavior that was also demonstrated by the unheated gluten. CV heated gluten did not precipitate. This is demonstrating that the proteins, w ith water available, that were MW heated were not altered to as great an extent as the proteins that were CV heated. SDS-PAGE was useful in characterizing the protein in the fractions. Each fraction contained a variety of proteins w ith a wide range of molecular weights. Differences were observed with electrophoresis between the CV and MW heated samples that was not necessarily observed with the pH measurements. The proteins that were soluble at each pH were different after each heating method. Possibly the conformations of the chains were affected in a way that changed the solubility. 13C NMR was also used to characterize the same fractions. This method had the advantage of also being able to provide information on the carbohydrate component, but was not as sensitive to the effects of denaturation as PAGE. Differences were observed between the CV and MW heated samples, especially in the carbohydrate fractionation behavior. Some kind of interaction between the protein and the carbohydrate occurred with 123 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. heating that allowed the carbohydrate to be present in a fraction a n d /o r affected which fraction a protein was soluble in. It appears from this research that the difference in texture between the MW and CV heated samples is due to an interaction between the starch and gluten, although the protein does denature differently if it is MW or CV heated. The amount of water that is present is also important. At 35% moisture there is not enough water to facilitate the changes that occur at 50% moisture. If the moisture is increased to 65% there is so much water it appears to be the dominant factor. More research needs to be conducted to understand the composition of the carbohydrate in VWG, how it interacts with the gluten, and how starch and gluten interact when heated. Further characterization of the protein in the fractions would provide information on the composition and conformation of the proteins in all the fractions to try and understand what causes the proteins to be soluble at each pH. 124 » U» ■■■ I1-.1 1 »l . ■V I.'■>!■. » A■■■■■■,I C■■I'.Ml UU.,!«■L' l,M, t jn.f Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LITERATURE Ablett, S., Barnes, D.J., Davies, A.P., Ingman, S.J., and Patient, D.W. 1988. and pulse nuclear m agnetic resonance spectroscopy of w heat proteins. J. Cereal S d . 7:11-20. Akyel, C., Bosisio, R.G., Chahine, R., and Bose, T.K. 1983. M easurem ent of the complex permittivity o f food products during microwave p ow er heating cycles. J. Microw. Pwr. 18(4):355-365. American A ssodation of Cereal Chemists. 1983. A pproved m ethods o f the AACC. M ethod 7410. The Assodation:St. Paul, M N Anderson, K.H., Lorence, M.W., and DeVay, G.E. 1989. M ethod of m icrowave heating of starch-based products. United States Patent N um ber 4,842,876. Anon. 1992. Basics o f measuring the dielectric properties o f materials. H ew lett Packard A pplication N ote 1217-1. Attenburrow, G.E., Goodband, R.M., Taylor, L.J., and Lillford, P.J. 1989. Structure, mechanics and texture o f a food sponge. J. Cereal Sci. 9:61-70. Avital, Y., M annheim, C.H., and Miltz, J. 1990. Effect o f carbon dioxide atmosphere on staling and w ater relations in bread. J. Fd. Sci. 55(2):413-416,461. Bachrach, H.L. and Briggs, D.R. 1947. Studies on bread staling. II. Water relationships during staling o f bread crumb and the retrogradation o f starch. Cereal Chem. 24:492-506. Back, P.J., C oy, A., Xia, Y., Callaghan, P.T., Diamante, L.M., and Umbach, S.L. 1991. Som e biophysical applications of motional contrast in n.m.r. m icroscopy. Int. J. Biol. Macromol. 13:181-189. Baianu, I.C. 1981. Carbon-13 and proton nuclear m agnetic resonance studies o f wheat proteins. Spectral assignm ents for Flanders gliadins in solution. J. Sci. Fd. Agric. 32:309-313. Baianu, I.C., Johnson, L.F., and Waddell, D.K. 1982. H igh-resolution proton, carbon-13 and nitrogen-15 nuclear magnetic resonance studies o f w heat proteins at high magnetic fields: Spectral assignm ents, changes w ith concentration and heating treatments o f Flinor gliadins in solution - comparison with gluten spectra. J. Sci. Fd. Agric. 33:373-383. Baker, A.E., Dibben, R.A., and Ponte, J.G. 1987. Comparison o f bread firmness m easurements by four instrum ents. Cereal Fds. World. 32(7):486-489. Baker, A.E., Doerry, W.T., and Kemp, K. 1986a. Instron factors involved in measuring crumb firmness. Cereal Fds. World. 31(2):193-195. Baker, A.E., Doerry, W.T., and Kemp, K. 1986b. Graphical presentation of Instron factors on crumb firmness. Cereal Fds. World. 31(3):262-268. 125 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Baker, B.A., D avis, E. A., and Gordon, J. 1990. The influence o f sugar and em ulsifier type during m icrowave and conventional heating of a lean formula cake batter. Cereal Chem. 67:451457 Bale, R. and Muller, H.G. 1970. Application of the statistical theory of rubber elasticity to the effect o f heat on w heat gluten. J. Fd. Tech. 5(3):295-300. Bekes, F., Z aw istow ska, U., and Bushuk, W. 1983. Protein-lipid com plexes in the gliadin fraction. Cereal Chem . 06(5):371-378. Belton, P.S. and Colquhoun, I.J. 1989. Nuclear magnetic resonance spectroscopy in food research. Spectroscopy 4(9):22-32. Belton, P.S., Duce, S.L., and Tatham, A.S. 1987. ^ C solution state and solid state n.m.r. of wheat gluten. Intl. J. o f Biol. Macrom. 9:357-362. Belton, P.S., Duce, S.L., and Tatham, A.S. 1988. Proton nuclear m agnetic resonance relaxation studies of dry gluten. J. Cereal Sci. 4:113-122. Bengtsson, N.E. and Risman, P.O. 1971. Dielectric properties o f food at 3 G Hz as determined by a cavity perturbation technique. II. Measurements on food materials. J. M icrowave Pwr. 6(2):107-123. Bemardin J.E. and Kasarda, D.D. 1973. The microstructure o f w heat protein fibrils. Cereal Chem. 50:735-745. Bemardin, J.E., Kasarda, D.D., and M echam, D.K. 1967. Preparation and characterization of a-gliadin. J. Biol. Chem. 242(3):445-450. Bietz, J.A. and Wall, J.S. 1972. Wheat gluten subunits: M olecular w eights determ ined by sodium dodecyl sulfate-polyacrylam ide gel electrophoresis. Cereal Chem . 49:416-430. Bietz, J.A. and W all, J.S. 1973. Isolation and characterization o f gliadin-like subunits from glutenin. Cereal Chem. 50:537-547. Bietz, J.A., Huebner, F.R., and Wall, J.S. 1973. Glutenin - the strength protein o f w heat flour. The Bakers Digest. Feb.:26-31,3 4 ,3 5 ,6 7 . Bietz, J.A., Shepherd, K.W., and W all, J.S. 1975. Single-kem al analysis o f glutenin: u se in wheat genetics and breeding. Cereal Chem. 52:513-532. Blish, M.J. 1945. Wheat G luten pp. 337-359 in: Advances in Protein Chemistry 2. A nson, M.L. and Edsall, J.T., eds. Academ ic Press:New York. Bloch, F. 1953. The principle o f nuclear induction. Science 118:425-430. Bloembergen, N ., Purcell, E.M., and Pound, R.V. 1948. Relaxation effects in nuclear m agnetic resonance absorption. Phys. Rev. 73(7):679-712. Bloksma, A.H. and Bushuk, W. 1988. Rheology and Chemistry o f D ough pp. 131-217 in: W heat Chemistry and Technology, vol. II. Y. Pomeranz, ed. AACC:St. Paul. 126 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Blum, F.D. 1986. Pulsed-gradient spin-echo nuclear magnetic resonance spectroscopy. Spectroscopy l(5):32-38. Blythe, A.R. 1979. M easurement o f Dielectric Properties. Pages 68-89 in: Electrical Properties o f Polymers. Cambridge Press:Cambridge. Booth, M.R., Bottomley, R.C., Ellis, J.R.S., Malloch, G., Schofield, J.D., and Timms, M.F. 1980. The effect o f heat on gluten-physico-chem ical properties and baking quality. Ann. Technol., agric. 29(2):399-408. Bottom ley, R.C., Kearns, H.F., and Schofield, J.D. 1982. Characterisation o f w heat flour and gluten proteins using buffers containing sodium dodecyl sulphate. J. Sci. Food. Agric. 33:481491. Brand, B.A. 1987. M icrowave-conventional heating o f m odel cake system s containing com ponents o f differing dielectric properties. Masters thesis, U o f M innesota: M inneapolis, MN. Breitmaier, E. and Voelter, W. 1990. Carbon-13 NM R Spectroscopy - H igh-Resolution Methods and Applications in Organic Chemistry and Biochemistry, Third Edition. VCH Publishers:N ew York. Brinton, R.H. Jr. and Bourne, M.C. 1972. Deformation testing o f foods. III. Effect of size and shape on the m agnitude of deformation. J. Texture Studies. 3:284-297. Bryant, R.G. and H alle, B. 1982. NM R relaxation of water in heterogeneous system s - consensus view s? Pages 389-393 in: Biophysics of Water. F. Franks and S.F. M athias, eds. John W iley and Sons Ltd.:New York. Buffler, C.R. and Stanford, M.A. 1991. Effects of dielectric and thermal properties on the m icrowave heating of foods. M icrowave World 12(41:15-23. Bum ouf, T. and Bietz, J.A. 1989. Rapid purification of wheat glutenin for reversed-phase high-perform ance liquid chromatography: Comparison of dim ethyl sulfoxide with traditional solvents. C ereal Chem. 66:121-127. Bushuk, W. 1966. Distribution o f water in dough and bread. The Bakers Digest. Oct. 38-40. Bushuk, W. 1984. Gluten: It's strengths and weaknesses. Paper presented at the m eeting of the International W heat G luten Association. June. Bushuk, W. 1985. Flour proteins: structure and functionality in d ough and bread. Cereal Foods W orld. 30(7):447-451. Bushuk, W. and Hlynka, I. 1964. Water as a constituent of flour, dough, and bread. The Bakers D igest Dec. 43-46,92. Bushuk, W. and Winkler, C.A . 1957. Sorption of w ater vapor on w heat flour, starch, and gluten. Cereal Chem. 34:73-86. 12 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Callaghan, P.T. and Lelievre, J. 1979. Diffusion of water in the endosperm tissu e o f w heat grains as studied b y pulsed field gradient nuclear magnetic resonance. Biophys. J. 28:133142. Callaghan, P.T., Jolley, K.W., and H um phrey, R.S. 1983a. Diffusion o f fat and water in cheese as studied by pulse field gradient nuclear magnetic resonance. J. Colloid & Int. Sci. 93(2):521-529. Callaghan, P.T., Jolley, K.W., Lelievre, J., and Wong, R.B.K. 1983b. N uclear m agnetic resonance studies of wheat starch pastes. J. Colloid & Int. Sci. 92(2):332-337. Callaghan, P.T., Trotter, C.M., and Jolley, K.W. 1980. A pulsed field gradient system for a fourier transform spectrometer. J. Mag. Res. 37:247-259. Chakraborty, K. and Khan, K. 1988a. Biochemical and breadmaking properties o f w heat protein com ponents. I. Compositional differences revealed through quantitation and polyacrylam ide gel electrophoresis o f protein fractions from various isolation procedures. Cereal Chem. 65(4):333-340. Chakraborty, K. and Khan, K. 1988b. Biochemical and breadmaking properties o f wheat protein components. II. Reconstitution baking studies o f protein fractions from various isolation procedures. Cereal Chem. 65(4):340-344. Chen, C.H. and Bushuk, W. 1970. Nature o f proteins in Triticale and its parental species I. Solubility characteristics and am ino acid com position o f endosperm proteins. Can. J. Plant Sc. 50:9-14. Chinachoti, P, W hite, V.A., Lo, L., and Stengle, T.R. 1991. Application o f high-resolution carbon-13, oxygen-17, and sodium-23 nuclear magnetic resonance to study the influences of water, sucrose, and sodium chloride o n starch gelatinization. Cereal Chem. 68:238-244. Chinachoti, P. and Steinberg, M.P. 1988. Interaction of sucrose w ith gelatin, eg g album in and gluten in freeze-dried mixtures as sh ow n b y water sorption. J. Fd. Sci. 53(31:932-939. Chung, O.K., Pomeranz, Y., H w ang, E.C., and Dikeman, E. 1979. Defatted and reconstituted w heat flours. IV. Effects o f flour lipids on protein extractability from flours that vary in bread-m aking quality. Cereal Chem. 56(41:220-226. Cloke, J.D., Davis, E.A., and Gordon, J. 1984. Relationship of heat transfer and water-loss rates to crumb-structure developm ent as influenced by m onoglycerides. Cereal Chem. 61(41:363-371. Cochran, S.A., Benjamin, E.J., Crocker, M.E., Seidel, W.C., and Cipriano, V.L. 1989. M icrowaveable baked goods. United States Patent N um ber 4,885,180. Cochran, S.A., Cin, D.A., and Veach, S.K. 1990. M icrowaveable baked goods. U nited States Patent Num ber 4,957,750. Collison, R. and McDonald, M.P. 1960. Broadening o f the water proton line in high-resolution nuclear magnetic resonance spectra of starch gels. N ature May:548-549. 128 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Creamer, L.K. and Richardson, T. 1984. Anom alous behavior o f bovine otsi- and (l-caseins on gel electrophoresis in sodium dodecyl sulfate buffers. Arch. Biochem. Biophys. 234(21:476-486. d'Avignon, D.A., H ung, C.-C., Pagel, M.T.L., Hart, B., Bretthorst, G.L., and Ackerman, J.J.H. 1990. ^H and 2H studies of water in work-free w heat flour doughs. Pages 391-414 in: NMR Applications in Biopolymers. J.W. Finney et al., ed s. Plenum Press:New York. Dahle, L.K. and Sambucci, N . 1987. Application of d evised universal testing m achine procedures for measuring texture of bread and jam-filled cookies. Cereal Fds. World. 32(71:466-470. Datta, A.K. 1990. Heat and m ass transfer in the m icrowave processing o f food. Chem. Eng. Prog. June:47-53. Davis, E.A. and Gordon, J. 1990. M icrowave applications for bakery products. American Institute o f Baking Technical Bulletin XII(91. G. Ranhotra, ed. Manhattan, KS. Decareau, R.V. 1985. Microwaves in the Food Processing Industry. Academ ic Press:OrIando. Decareau, R.V. and Peterson, R.A. 1986. M icrowave technology. Pages 37-60 in: M icrowave Processing and Engineering. Ellis Horwood:Chichester, U.K. Derbyshire, W. 1983. The dynam ics of water in heterogeneous system s with em phasis on subzero temperatures. Pages 339-469 in: Water, A Com prehensive Treatise. F. Franks, ed. Plenum Press:New York. Eisenberg, D. and Kauzmann, W. 1969. The Structure and Properties of Water. Oxford University Press:London. Eliasson, A.-C. and Lundh, G. 1989. Rheological and interfacial behaviour o f som e wheat protein fractions. J. Text. Stud. 20:431-441. Engelder, D.S. and Buffler, C.R. 1991. M easuring dielectric properties of food products at m icrowave frequencies. M icrowave World 12(21:2-11. Ewart, J.A.D. 1980. Loaf volum e and the intrinsic viscosity o f glutenin. J. Sci. Fd. Agric. 31:1323-1336. Ewart, J.A.D. 1985. Blocked thiols in glutenin and protein quality. J. Sc. Food Agric. 36:101112. Ewart, J.A.D. 1987. Calculated molecular w eight distribution o f glutenin. J. Sc. Food Agric. 38:277-289. Fennem a,O.R. 1985. Water and Ice pp. 24-67 in: Food Chemistry, 2nd Ed. O.R. Fennema, ed. Marcel Dekker, Inc.:New York. Finney, K.F. 1985. Experimental breadmaking studies, functional (breadmaking) properties, and related gluten protein fractions. Cereal Foods World. 30(11):794-801. 129 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Fleischmann, A.M. 1991. The dielectric properties o f w h ey proteins and starch system s. Masters thesis. U niversity of Minnesota: M inneapolis, M N. Franks, F. 1983. The dynam ic properties of liquid water. Pages 26-34 in: Water. F. Franks, ed. Royal Society of Chem.:London. Fullington, J.G., M iskelley, D.M., W rigley, C.W., and Kasarda, D.D. 1987. Q uality -related endosperm proteins in sulfur-deficient and normal wheat grain. J. Cereal Sci. 5:233-245. Galletti, A.M .C., D avis, H.T., Davis, E.A., and Gordon, J. 1980. Water loss, thermal and structural properties o f dry cooked potato tubers. J. Fd. Proc. Pres. 3:301-327. Gao, L. and Bushuk, W. 1992. Solubilization o f glutenin in u rea/S D S solutions at elevated temperature. J. Cereal Sci. 16:81-89. G odsalve, E.W., D avis, E.A., Gordon, J., and Davis, H.T. 1977. Water loss rates and temperature profiles of dry cooked bovine muscle. J. Fd. Sci. 42(4):1038-1045. Goebel, N.K ., Grider, J., Davis, E.A., and Gordon, J. 1984. The effects o f m icrow ave energy and convection heating o n w heat starch granule transformations. Fd. Microstructure 3:73-82. Gordon, J., D avis, E.A., and Timms, E.M. 1979. Water-Ioss rates and temperature profiles of cakes o f different starch content baked in a controlled environm ent oven. Cereal Chem. 56(2):50-57. Graveland, A . 1980. Extraction o f wheat proteins w ith sod iu m dodecyl sulphate. Ann. Technol, Agric. 29(2):113-123. Graveland, A., Bongers, P., and Bosveld, P. 1979. Extraction and fractionation o f w heat flour proteins. J. S d . Fd. Agric. 30:71-84. Graveland, A., Bosveld, P., Lichtendonk, W.J., M oonen, H.H.E., and Scheepstra, A. 1982. Extraction and fractionation of wheat flour proteins. J. Sci. Fd. Agric. 33:1117-1128. Halle, B., A ndersson, T, Forsen, S. and Lindman, B. 1981. Protein hydration from water oxygen17 m agnetic relaxation. J. Am. Chem. Soc. 103:500-508. Hamauzu, Z., Khan, K., and Bushuk, W. 1979. Studies of glutenin. XTV. Gel filtration and sodium d od ecyl sulfate electrophoresis o f glutenin solubilized in sod iu m stearate. Cereal Chem. 56(6):513-516. Hansen, L.P., Johnston, P.H., and Ferrel, R.E. 1975. H eat-m oisture effects on wheat flour. I. Physical-chemical changes of flour proteins resulting from thermal processing. Cereal Chem. 52(4):459-472. Harris, R.K. 1983. Nuclear Magnetic Resonance Spectroscopy. Longm an Scientific and TechnicahEssex, England. Helenius, A. and Sim ons, K. 1975. Solubilization o f membranes b y detergents. Biochem. Biophys. Acta 415:29-79. 130 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Hennig, H.J. and Scholz, P. 1976. M agnetische kemresonanzuntersuchungen zu m verhalten adsorbierter w asserm olekeln in nativer starke. Zeitschrift fur Physik. Chem. N eue Folge. 103:31-44. Hennig, von H.J. and Lechert, H. 1974. M essungen der magnetischen relaxationszeiten der protonen in nativen starken m it verschiedenen wassergehalten. D ie Starke 26(7):232-237. Hibberd, G.E. and Parker. 1985. Measurement of the compression properties o f bread crumb. J. Texture Studies. 16:97-110. Hoseney, R.C. 1986. Principles o f Cereal Science and Technology. Page 158. AACC:St. Paul, M N. H oseney, R.C. and Finney, P.L. 1974. M ixing - A contrary view . Bakers D igest 48(l):22-24,26, 28,66. H oseney, R.C., Atw ell, W.A., and Lineback, D.R. 1977. Scanning electron m icroscopy o f starch isolated from baked products. Cereal Fds. World. 22(2):56-60. . H oseney, R.C., Finney, K.F., Shogren, M.D., and Pomeranz, Y. 1969. Functional (breadmaking) and biochemical properties o f w heat flour components. II. Role o f water-solubles. Cereal Chem. 46:117-125. H oseney, R.C., Finney, K.F., Shogren, M.D., and Pomeranz, Y. 1969. Functional (breadmaking) and biochemical properties of w heat flour components. III. Characterization of gluten protein fractions obtained b y ultracentrifugation.. Cereal Chem . 46:116-135. H uang, V.T., H oseney, R. C., Graf, E., Ghiasi, K., Miller, L.C., Weber, J.L., Gaertner, K.C., Matson, K.L., Hunstiger, A.M., Rogers, D.K., and Saguy, I.A. 1991. Starch-based products for m icrowave cooking or heating. United States Patent N um ber 5,035,904. Hung, C.C. 1980. Water migration and structural transformation o f oven cooked meat. Masters thesis. University of Minnesota: M inneapolis, MN. Jane, J., Xu, A., Radosavljevic, M., and Seib, P.A. 1992. Location o f am ylose in normal starch granules. I. Susceptibility o f am ylose and amylopectin to cross-linking reagents. Cereal Chem. 69:405-409. Jaska, E. 1971. Starch gelatinization as detected by proton magnetic resonance. Cereal Chem. 48:437-444. Jeanjean, M.F., Dam idaux, R., and Feillet, P. 1980. Effect of heat treatment on protein solubility and viscoelastic properties o f w heat gluten. Cereal Chem. 57(5):325-331. Joensson, T. and Toemaes, H . 1987. The effect of selected surfactants o n bread crumb softness and its m easurem ent. Cereal Fds. World. 32(7):482-485. Jones, B.L., Finney, K.F., and Lookhart, G.L. 1983. Physical and biochem ical properties of w heat protein fractions obtained b y ultracentrifugation. Cereal Chem. 60(4):276-280. 131 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Jones, R.W., Taylor, N .W ., and Senti, F.R. 1959. Electrophoresis and fractionation o f wheat gluten. Arch. Biochem. Biophys. 84:363-376. Kamel, B.S. 1987. Bread firmness m easurement w ith em phasis on Baker compressimeter. Cereal Fds. World. 32(7):472-476. Karger, J., Pfeifer, H., and Heink, W. 1988. Principles and applications of self-diffusion measurements by nuclear m agnetic resonance. Adv. Mag. Res. 12:1-89. Kasarda, D.D., Bemardin, J.E., and Gaffield, W. 1968. Circular dichroism and optical rotatory dispersion o f alpha-gliadin. Biochemistry. 7:3950-3957. Kasarda, D.D., Bemardin, J.E., and N im m o, C.C. 1976. W heat Proteins pp. 158-236 in Advances in Cereal Science and Technology. Pomeranz, Y. ed. Am erican A ssociation of Cereal Chem ists, Inc:St. Paul. Khoo, U., Christianson, D.D., and Inglett, G.E. 1975. Scanning and transm ission m icroscopy of dough and bread. The Bakers Digest. August:24-26. Kobrehel, K. and Bushuk, W. 1977. Studies o f glutenin. X. Effects o f fatty acids and their sodium salts on solubility in water. Cereal Chem. 54(4):833-839. Kuntz, I.D., Jr. 1975. The physical properties of water associated w ith biom acrom olecules. Pages 93-109 in: Water relations o f foods. R.B. Duckworth, Ed. A cadem ic Press:London. Laemmli, U.K. 1970. Q eavage of structural proteins during the assem bly o f the head of bacteriophage T4. Nature, 227:680-685. Lambert, L.L.P., Gordon, J., and D avis, E.A. 1992. Water loss and structure developm ent in m odel cake system s heated by m icrowave and convection methods. Cereal Chem. 69:303309. Lang, K.W. and Steinberg, M.P. 1983. Characterization o f polym er and solute bound water by pulsed NMR. J. Fd. Sci. 48:517-520,538. Lasztity, R. 1986. Recent results in the investigation of the structure o f the gluten com plex. Die Nahrung 30(3^):235-244. Lechert, H ., M aiwald, W ., Kothe, R., and Basler, W. 1980. NM R-study o f water in som e starches and vegetables. J. Fd. Proc. and Pres. 3:275-299. Lechert, H.T. 1981. Water binding on starch: NM R studies on native and gelatinized starch. Pages 223-245 in: Water Activity: Influences on Food Quality. L.B. Rockland and G.F. Stewart, eds. Academic Press, Inc:New York. Lee, F.A. 1970. The effects of "bound" and "available" water on enzym ic processes in w heat flour doughs. Fd. Tech. in Australia Sept.:516-520. Lee, S.M. 1989. M icrowaveable foods and associated technology - A literature survey. Publication o f the Leatherhead Food RA. 132 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LeGrys, G.A., Booth, M.R., and AI-Baghdadi, S.M. 1981. The physical properties o f wheat proteins. Pages 243-264 in: Cereals: A renewable resource. Y. Pomeranz and L. M unck ed. AACC:St. Paul. Lelievre, J. and M itchell, J. 1975. A pulsed NM R stu d y o f som e aspects of starch gelatinization. D ie Starke 27(4):113-115. LePage, C.A. 1987. Heat transfer, water loss, and structural characteristics o f gluten m odel system s heated in an experimental conventional-m icrow ave oven. Masters thesis, University o f Minnesota: M inneapolis, M N. Leung, H.K., M agnuson, J.A., and Bruinsma, B.L. 1979. Pulsed nuclear magnetic resonance study o f w ater mobility in flour doughs. J. Fd. Sci. 44:1408-1411. Leung, H.K., Steinberg, M.P., Wei, L.S., and N elson, A.I. 1976. Water binding o f m acrom olecules determined b y pulsed NM R. J. Fd. Sci. 41:297-300. Lorenz, K., Charman, E., and Dilsaver, W. 1973. Baking w ith m icrowave energy. Fd. Tech. Dec.:28-36. Lumry, R. 1973. Symposium: Protein interactions in biosystem s. Some recent ideas about the nature of the interactions betw een proteins and liquid water. J. Fd. Sci. 38:744-755. MacRitchie, F. 1976. The liquid phase o f d o u g h and its role in baking. Cereal Chem. 53(3):318-326. MacRitchie, F. 1978. Differences in baking quality betw een w heat flours. J. Fd. Tech. 13:187194. MacRitchie, F. 1979. A relation between gluten protein am ide content and baking performance of w heat flours. J. Fd. Tech. 14:595-601. MacRitchie, F. 1985. Studies of the m ethodology for fractionation and reconstitution o f wheat flours. J. of Cereal Sc. 3:221-230. MacRitchie, F. 1987a. Evaluation o f contributions from w heat protein fractions to d ou gh m ixing and breadmaking. J. o f Cereal Sc. 6:259-268. MacRitchie, F. 1987b. Identification o f "quality components" o f gluten protein using fractionation and reconstitution m ethods. Pages 247-253 in: Gluten Proteins, Proceedings o f 3rd International Gluten Workshop, Budapest, H ungary. R. Lasztity and F. Bekes, eds. W orld Scientific. MacRitchie, F., Kasarda, D.D., and Kuzm icky, D.D. 1991. Characterization o f w h eat protein fractions differing in contributions to breadm aking quality. Cereal Chem. 68(21:122-130. Marston, P.E. and Wannan, T.L. 1976. Bread baking - The transformation from d ou gh to bread. The Bakers D igest 50(41:24-28, 49. M auritzen, C.M. and Stewart, P.R. 1965. The ultracentrifugation o f dou ghs m ade from wheat flour. Aust. J. Biol. Sd. 18:173-189. 133 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. McMaster, G.J. and Bushuk, W. 1983. Protein-carbohydrate com plexes in gluten: Fractionation and proximate com position. J. of Cereal Sc. 1:171-184. j Meredith, O.B. and Wren, J.J. 1966. Determination o f m olecular-w eight distribution in wheatflour proteins by extraction and gel filtration in a dissociating m edium . Cereal Chem . 43(2):169-186. i [ Miller, L.A., Gordon, J., and D avis, E.A. 1991. Dielectric and thermal transition properties o f chem ically m odified starches during heating. Cereal Chem . 68(5):441-448. j i Mora-Gutierrez, A. and Baianu, I.C. 1991. Carbon-13 nuclear m agnetic resonance studies of chem ically m odified w axy m aize starch, com syrups, and maltodextrins. Com parisons w ith potato starch and potato maltodextrins. J. Agric. Fd. Chem. 39:1057-1062. j M udgett, R.E. 1986. M icrow ave properties and heating characteristics o f foods. Fd. Tech. June:84-93,98. Muller, H.G. 1969. Application o f the statistical theory o f rubber elasticity to gluten and dough. Cereal Chem. 46:443-446. . N g, P.K., Scanlon, M.G., and Bushuk, W. 1988. W heat cultivar identification by sodium d od ecyl sulfate polyacrylam ide gel electrophoresis of the glutenin proteins. A ppendix II in: A Catalog of Biochemical Fingerprints o f Registered Canadian Wheat Cultivars by Electrophoresis and High-performance Liquid Chromatography. Publication N o . 139 of Food Science Dept., U o f Manitoba. Ortalo-Magne, F. and G oodw in, B.K. 1991. The structure and developm ent o f the international w heat gluten market and U.S. imports. Cereal Fds. World 36(12):998-1002. Orth, R.A. and Bushuk, W. 1972. A comparative stu dy o f the proteins o f w heats o f diverse baking qualities. Cereal Chem. 49:268-275. Orth, R.A. and Bushuk, W. 1973. Studies of glutenin. I. Comparison of preparative methods. Cereal Chem. 50:106-114. Ottenberg, R. 1984. Yeast-raisable wheat-based food products that exhibit reduced deterioration in palatability upon exposure to m icrow ave energy. United States Patent N um ber 4,463,020. Ottenberg, R. 1985. Yeast-raisable wheat-based food products that exhibit reduced deterioration in palatability upon exposure to m icrowave energy. U nited States Patent N um ber 4,560,559. Payne, P.I. and Lawrence, G.J. 1983. Catalogue o f alleles for the com plex g en e lo d , G lu -A l. G lu -B l. and G lu -D l w hich code for high-m olecular-weight subunits o f glutenin in hexaploid wheat. Cereal Research Comms. ll(l):2 9 -3 5 . Pence, J.W., M oham m ad, A., and Mecham, D.K. 1953. Heat denaturation o f gluten. Cereal Chem. 30:115-126. 134 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Pomeranz, Y. 1983. Molecular approach to breadmaking: An update and new perspectives. Bakers Digest. July/Aug.:72-86 Preston, K.R. and Tipples, K.H. 1980. Effects of add-soluble and add -in solub le gluten proteins on the rheological and baking properties o f wheat flours. Cereal Chem. 57(5):314-320. Purcell, E.M. 1953. Research in nuclear magnetism. Science 118:431-436. Richardson, S.J., Baianu, I.C., and Steinberg, M.P. 1986. M obility o f w ater in w h eat flour suspensions as studied by proton and oxygen-17 nuclear magnetic resonance. J. Agric. Fd. Chem. 34:17-23 Richardson, S.J., Baianu, I.C., and Steinberg, M.P. 1987a. M obility o f water in c o m starch suspensions determined b y nuclear magnetic resonance. D ie Starke 39(3):79-83. Richardson, S.J., Baianu, I.C., and Steinberg, M.P. 1987b. M obility o f water in starch powders determined by nuclear m agnetic resonance. Die Starke 39(6):198-203. Richardson, S.J., Steinberg, M.P., DeVor, R.E., and Sutherland, J.W. 1987c. Characterization of the oxygen-17 nuclear m agnetic resonance water m obility response surface. J. Fd. Sd. 52(1):189-193. Robyt. J.F. and White B.J. 1985. Electrophoretic Techniques pp. 129-157 in: Biochemical Techniques - theory and practice. W aveland Press, Inc.:Prospect H eights, IL. Roebuck, B.D., Goldblith, S.A., and W estphal, W.B. 1972. Dielectric properties of carbohydrate-water m ixtures at microwave frequencies. J. Fd. Sci. 37:199-204. Rogers, D.E., Doescher, L.C., and H oseney, R.C. 1990. Texture characteristics o f reheated bread. Cereal Chem. 67(2):188-191. Russel, P.L. 1983. A kinetic stu dy o f bread staling b y differential scanning calorimetry and crumb compressibility measurements. The effect o f different grists. J. Cereal Sci. 1:285-296. Sandstedt, R.M. 1961. The function o f starch in the baking of bread. Bakers D igest. 35(3):3644. SAS Institute. 1987. SAS Version 5 Edition User's G uide. SAS Institute:Cary, NC . Schofield, J.D., Bottomley, R.C., LeGrys, G.A., Timms, M.F., and Booth, M.R. 1984. Effects of heat on w heat gluten. Pages 81-90 in: Gluten Proteins - Proceeding of the 2nd International Workshop on Gluten Proteins. A. Graveland and J.H.E. M oonen, ed s. The Institute for Cereals, Hour, and Bread:Wageningen. Schofield, J.D., Bottomley, R.C., Timms, M.F., and Booth, M.R. 1983. The effect o f heat on wheat gluten and the involvem ent of sulphydryl-disulphide interchange reactions. J. of Cereal Sc. 1:241-253. Shanbhag, S., Steinberg, M.P., and N elson, A.I. 1970. Bound water defined and determined at constant temperature b y w ide-line NMR. J. Fd. Sci. 35:612-615. 135 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Shewry, P.R. and Tatham, A.S. 1989. N e w light on an old technology: The structure o f wheat gluten and its role in breadmaking. Outlook on Agriculture 18(2):65-71. Shew iy, P.R., Halford, N.G., and Tatham, A.S. 1992. H igh molecular w eigh t subunits of wheat glutenin. J. Cereal Sci. 15:105-120. Shewry, P.R., Tatham, A.S., Forde, J., Kreis, M., and Miflin, B.J. 1986. The classification and nomenclature of w heat gluten proteins: A reassessment. J. Cereal Sci. 4:97-106. Shogren, M.D., Finney, K.F., and H oseney, R.C. 1969. Functional (breadmaking) and biochemical properties o f w heat flour components. L. Solubilizing gluten and flour protein. Cereal Chem. 46(2):93-102. Stejskal, E.O. and Tanner, J.E. 1965. Spin diffusion measurements: Spin echoes in the presence o f a tim e-dependent field gradient. J. Chem. Phys. 42:288-292. Sterling, C. and M asuzawa, M. 1968. G el/w ater relationships in hydrophilic polymers: Nuclear magnetic resonance. D ie Makromol. Chemie 116:140-145. Suggett, A. 1975. Water-carbohydrate interactions. Pages 23-36 in: Water Relations o f Food. R.B. Duckworth, Ed. Academic Press:London. Suggett, A. 1976. Molecular m otion and interactions in aqueous carbohydrate solutions. III. A combined nuclear m agnetic and dielectric-relaxation strategy. J. Soln. Chem. 5(l):33-46. Suggett, A. and Clark, A.H. 1976. Molecular m otion and interactions in aqueous carbohydrate solutions. I. Dielectric-relaxation studies. J. Soln. Chem. 5(1):1-15. Suggett, A., Ablett, S. and Lillford, P.J. 1976. M olecular m otion and interactions in aqueous carbohydrate solutions. III. Nuclear-magnetic-relaxation studies. J. Soln. Chem. 5(1):1731. Swyngedau, S., Nussinovitch, A., Roy, I., Peleg, M., and H uang, V. 1991. Com parison of four models for the compressibility o f breads and plastic foams. J. Fd. Sci. 56:756-759. Tait, M.J., Ablett, S. and W ood, F.W. 1972. The binding of water on starch, an nmr investigation. J. Colloid Int. Sci. 41(3):594-603. Tanner, S.F., H ills, B.P., and Parker, R. 1991. Interactions of sorbed water w ith starch studied using proton nuclear magnetic resonance spectroscopy. J. Chem. Soc., Faraday Trans. 87(16):2613-2621. Taranto, M.V. 1983. Structural and textural characteristics of baked goods. Pages 229-265 in: Physical Properties o f Foods. M. P eleg and E. Bagley eds. AVI:Westport CT. Tatham, A.S., Field, J.M., Smith, S.J., and Shewry, P.R. 1987. The conformations of w heat proteins, II, aggregated gliadins and lo w molecular w eight subunits o f glutenin. J. o f Cereal Sc. 5:203-214. Toledo, R., Steinberg, M.P., and N elson, A.I. 1968. Quantitative determ ination of bound water by NMR. J. Fd. Sci. 33:315-317. 136 i_ji1.1.n|i..iiumiiiijiuum1,4 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Tran, V.N., Stuchly, S.S., and Kraszewski, A. 1984. Dielectric properties o f selected vegetables and fruits 0.1-10.0 GHz. J. M icrowave Pwr. 19(4):251-258. Umbach, S.L. 1989. Examination of the bagel making process w ith an em phasis on the water behavior and conventional versus m icrow ave heating m ethods. Masters thesis, University o f M innesota:M inneapolis, MN. Umbach, S.L., D avis, E.A. and Gordon, J. 1990. Effects of heat and water transport on the bagel-m aking process: Conventional and m icrowave baking. Cereal Chem. 67(4):355-360. Varilek, P. and Walker, C.E. 1984. Baking and ovens - History o f heat technology IV. Bakers D igest 58(2):12-15. Varriano-Marston, E., Ke, V., Huang, G., and Ponte Jr., J. 1980. Comparison o f m ethods to determ ine starch gelatinization in bakery foods. Cereal Chem. 57(4):242-248. V oisey, P.W, M acDonald, D.C., Kloek, M., and Foster, W. 1972. The Ottawa Texture M easuring System - An Operational Manual. Eng. Spec. 7024. Agriculture Canada. O ttawa, Canada. V oisey, P.W. 1970. Test cells for objective textural m easurements.. Cdn. Inst. Fd. Sc. Tech. J. 3(3):93-102. Voisey, P.W. 1971. The Ottawa Texture M easuring System. Cdn. Inst. Fd. Sc. Tech. J. 4:91-103. V oisey, P.W. and Larmond, E. 1977. Instrumental m ultiple texture profile analysis o f cooked ground b eef patties and smoked beef. J. Texture Stud. 8:93-107. V oisey, P.W., Buckley, D.J., and Crete, R. 1974. A system for recording deformations in texture tests. J. Texture Studies. 5:61-75. V olm an, D.H ., Simons, J.W., Seed, J.R., and Sterling, C. 1960. Sorption of w ater vapor by starch. Thermodynamics and structural changes for dextrin, am ylose, and am ylopectin. J. Polymer Sci. 46:355-364. V on H ippel, A.R. 1954. Dielectrics and W aves. MIT Press:Cambridge, MA. W all, J.S. 1979. The role of w heat proteins in determ ining baking quality. Pages 275-311 in: Recent Advances in the Biochemistry of Cereals. D.L. Laidman and R.G. W yn Jones, ed s. Academ ic Press:New York. W asik, R.J., Daoust, H., and Martin, C. 1979. Studies o f glutenin solubilized in high concentrations of sodium stearate. Cereal Chem. 56(2):90-94. W ebb, T., H eaps, P.W., Russell Eggitt, P.W., and Coppock, J.B.M. 1970. A Theological investigation o f the role o f water in w heat flour doughs. J. Fd. Tech. 5:65-76. W eber, K. and O sbom , M. 1969. The reliability o f m olecular w eigh t determ inations b y dodecyl sulfate-polyacrylam ide gel electrophoresis. J. Biol. Chem. 244(16):4406-4412. 13 7 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. W eegels, P.L. and Hamer, R.J. 1991. Improving the baking quality o f wheat gluten. Pages 219223 in: Gluten Proteins 1990. W. Bushuk and R. Tkachuk eds. A ACC:St. Paul. W eegels, P.L., Marseille, J.P., dejager, A.M., and Hamer, R.J. 1991. Structure-function relationships o f gluten proteins. Pages 98-111 in: G luten Proteins 1990. W. Bushuk and R. Tkachuk eds. AACC:St. Paul. Wei, C.K., D avis, H.T., Davis, E.A., and Gordon, J. 1985a. H eat and mass transfer in water laden sandstone: Convective heating. AIChE J. 31(8):1338-1348. Wei, C.K., D avis, H.T., Davis, E.A., and Gordon, J. 1985b. H eat and mass transfer in waterladen sandstone: M icrowave heating. AIChE J. 31(5):842-848. W histler, R.L., BeMiller, J.N., and Paschall, E.F. 1984. Starch - C hem istiy and Technology. 2nd Edition. Academ ic Press:OrIando, Florida. W ickersheim, K.A. and Sun, M.H. 1987. Fiberoptic therm om etry and its applications. J. M icrow. Pwr. 22(2):85-94. W illhoft, E.M.A. 1973. Mechanism and theory o f staling of bread and baked goods, and associated changes in textural properties. J. Texture Studies. 4:292-322. W ilson, R.H., G oodfellow , B.J., Belton, P.S., Osborne, B.G., Oliver, G., and Russell, P.L. 1991. Com parison o f Fourier transform mid-infrared spectroscopy and near infrared reflectance spectroscopy w ith differential scanning calorimetry for the stu d y of the staling of bread. J. Sci. Fd. Agric. 54:471-483. W oessner, D.E. and Zimmerman, J.R. 1963. Nuclear transfer and anisotropic m otional spin phenom ena: Relaxation time temperature dependence studies o f water adsorbed on silica gel. IV. J. Phys. Chem. 67:1590-1600. W rigley, C.W. and Bietz, J.A. 1988. Proteins and am ino acids. Pages 159-275 in: Wheat Chem istry and Technology. Y. Pomeranz, ed. AACC:St. Paul. Wynne-Jones, S. and Blanshard, J.M.V. 1986. H ydration studies of w heat starch, am ylopectin, am ylose gels and bread by proton magnetic resonance. Carbohy. Polymers 6:289-306. W ynne-Jones, S. and Blanshard, J.M.V. 1986. Hydration studies of w heat starch, am ylopectin, am ylose gels and bread by proton magnetic resonance. Carbohydrate Polymers 6:289-306. Zim m erm an, J.R. and Lasater, J.A. 1958. Nuclear m agnetic resonance relaxation studies of adsorbed w ater on silica gel. III. J. Phys. Chem. 62:1157-1163. Zim m erm an, J.R., H olm es, B.G., and Lasater, J.A. 1956. A stu d y o f adsorbed water on silica gel by nuclear resonance techniques. J. Phys. Chem. 60:1157-1161. Zobel, H.F. 1988. M olecules to granules: A com prehensive starch review . Starch 40(2):44-50. Zylem a, B.J., Grider, J., Gordon, J., and D avis, E.A. 1985. M odel w heat starch system s heated by m icrow ave irradiation and conduction w ith equalized heating times. Cereal Chem. 62:447-453. 138 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. NO h n o OlOO NIN 00 L o lO rjJ rj5 r-» CO Cs ON 00 00 Cs CM o to to © in lO NO ON no co o oo vd to Raw data of texture experiments (force, kg) ft o. . vO o in o nO rH Cs nO NO CO CM co tn o Os is CM CM CM CM r-H NO T-H r-H t-H T-H t-h T-H T*H 00 o o 00 00 O 00 © ON CO C s CM CM r-H CM 0 0 r—• ON r*H ON 0 0 r - o in in CM in 0 0 © o CM C s O in 0 0 ON 0 0 vO CS C s t-h t*h in CM O O 0 0 CO CO in ON CM o in o in in o o v d rj< in c d in pH CO CO co CO o co ON CO o o in o o CM in A C oO 15L £ ca CO Ta3 < 00) -C > U IS in in in ^ t s i n no CM t—• r-* t- h t-h no CM in m O ON CO i n NO 0 0 r-* r - t CM CM C M ^ o '' o o o CM I o CS o T—>ON CS o LO CO NO r f r-i T f <?. LO O NO r-* N O © O °) r i r* Q\ ^CMi CO n oCOc CM o ^CO o o ) o in in in CM O CO LT) 0 0 CM CM CO CM CM CM CO 0 0 O NO NO ON ON ON C s Cs o 00 o in c d in CM NO 0 0 0 0 o in o in CO NO C s CM VO CO CO tn o CM o o 00 -o &•> o o o in rH o in o in in VO co in tn in CM CO 0 0 rH in in rp tn T-H in in NO 0 0 o NO VO in C s NO NO CM VO in CM ON C s CO C s 0 0 O in T-* rH r n T-H CM pH rHC M O N r f 00 ss O n r* 00 S N to CM rH NO CM 0 0 r-» NO IN N vO 0 0 N U U CL Cj O CO CcJfO^lONONOOONrH CL £<Dcm co ?Q in vo 1100 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o orH o O pH pH OS vO Cn rH r “! CO c d c d c d c d c d c d c d rH rH r n T-H T-H T-H T-H T-H O O rH O in © rH q CO a s O s OS CO 0 0 O s cq VO Cn Cn IN a s Cn CN CN c d c d CN CN CN c d c d CN CN CN © T-H t- h T-H T-H pH T-H T-H rH rH T-H T-H r n T-H o in as o CO o q in Raw data of D value experiments (multiply all values by 10'10, m2/s) CO 0 0 IN tN a s T-H CN a s rH rH r n CN T-H cd T-H T-H T-H rH r n t h rH CN rH rH o in o in o cd CN CN CN CN 0 0 a s T-H Tj< v p cq O CO cd cd cd CO c d c d c d rH T-h T-H T-H T-H T-H rH T-H T-H T-H T-H rH rH in in in in in in in in in m in in in O T-H IN CO CN IN IN T-H CN Cn d vd vd Os 00 00 00 O 0 0 rH rH r n T-H T-H r n T-H T-H rH r n T-H in O 00 o CN o in o m o co vo in o in os oo o in as CN Cn CO CO 0 0 a s T-H O rH OS 0 0 a s OS OS 0 0 a s o o o OS o s a s Cn CN vO T-H T-H t-h vO CN VO 0 0 T-H CO CO T-H CO Cn pH T-H a s 0 0 0 0 OS a s a s a s a s o q cq vo rH $ Cn as oo rH a s CO T-H r n CN IN © CN c d c d CN CN T-H t-h t- h r n T-H T-H t h r n 00 o O o orH CO rH in o ot- h tn sS o O o CO co o rH CN o in in O CO C s O a s VO T-H T-H 0 0 CN CO CO cd N 5 cd N5 o 0 0 OS O rn as cd o •N5 Cn vO O co 00 in £ N5 ■N5 CN rH 0 0 CO CO 0 0 Cn CN VO 0 0 0 0 r n T-H CO CO T-H vd LO in TT s in in in in in in Cn GO rH 00 CN CO O in S! tn in in 9.42 13.3 13.5 12.4 11.0 I 8.23 u S repl G P 35% a; a A o o orH ] T3 S rH a s CO 00 0 0 v o T-H rH VO v vd d v d v d v d v d v d v d v d CN S:G cn in CN r-H vO s o l o T-H CN CN CN T-h c d rH r n t- h pH T-H r-H T-H o 7.4 00 6 P VO CO VO CN VO CN OS Cn CN VO CN Cn c d 0 0 IN CN T-H CO T-H rH T-H T-H t- h r n r n o o in N o CN CO in in in o CN as in o 1C/1J cl in Cn CN CN CN C s CN O CO O r n t- h vd a s vd vd vd id T* O in VO Cn 0 0 OS T-H T-h 12 13 14 O O rH o TlJ in O Cn CN O s O CO IN OS Cn c d CN c d c d c d CN pH c d IN rH r n rH CN CN CN CN CN CO CO CN CN CN 10 o o in orH so in T-H VO t- h 140 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 00 rn O O 00 CN in in CO VO tn v q IN cq CN CN CN CN CN CN CN CN CN CN pH PH pH rH T-H rH vO rn rn rn rn in in in rn pH O in in 0 0 in rn O © CN rH t-H VO rn pH rH O O rn rH rn pH O CN O PH PH rn rn rH rn rn CN CN © CO CN CN CN CN O pH PH rH r n r n pH rH rn rn rn 0 CN CN in 00 00 in OV OV in CN O CO rn 00 VO VO CO ov ov © O 00 IN IN VO VO vd IN IN IN 00 0 0 0 © 0 O CO © CO CO CO CO ov in O r n CN LO r n in in VO in 00 0 0 0 0 0 0 0 0 0 0 00 00 0 0 00 00 0 pH rH vO 00 CO o v in rn 00 V O Cn tN 00 Cn CO 00 00 © rH rH CN v d v d v d v d v d v d v d tN Cn Cn cq 0 CN © VO CN 00 pH vO Ov 00 ov Ov vO i n OV 0 0 r r VO Cn CN v d v d v d v d v d v d v d Cn Cn Cn Cn ov O 0 0 C V rH vq v q Cn c q Cn CO CO CO CO co ss § o © o o in ri CV © CV Heated Samples 00 cq 3 c5 c d rjJ CN CN CO CO OV 0 0 in in 0 0 0 0 00 CN Ov in © © in 00 PH CO © cq v q Cn in in c d c d c d c d Ov CO CN CO VO 00 CO CN V Oi n^mOt ^N^c H OVOOO oinincN C N vo 9 o'* O in © CO rn u. o 0 0 © Cn in Cn CN v o CN © rH CN CO CN v q v q v q VO m cd c d cd cd cd © rn cN CO in VO Cl t O^ mvOt NCOOVt HHr i r i i HHr H 141 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o o LO CN vq O N O r*“ LO O N rH CS L O r H CO LO OO O rn rH CS CS CS c d c d c d c d rn rn rn r n rn r n rn rH r H rn rH rn r n cd cd cd cd rn © I LO 0 0 0 0 CO v o LO C s ON > c d c d c s c d CS rn r n rn i rn rH rn rH rn r n r H rH p v q vq CS rH rH Ov O CO CO CO CO c d c d CS c d rn r n rn rn rn rH rH gs S rn rn rH IN CO rH CS c d c d c d c d LO O S 3 CO rn rn t-h C s v q ON cd rn rn rn rn LO Cs CS VO ON rn LO LO Cs 00 ON O Cs C s Cs 00 v d Cs 0 0 Cs Cs CO LO ON C s rH LO LO LO VO rn 0 0 LO rn N Cs C s C s 0 0 0 0 0 0 0 0 ^ 00 cs cs VO LO CS O CS ON LO CO O 0 0 ^ CS CO rH rH v c s N N c d N o o o d o d c d o d q CN O N 0 0 ^ tn ^ CO q l o vd v d v d v d o o o to vd 0 0 CO C"s 0 0 0 0 rH CS CO r n rn (S ^ O C O H C S O O C S N ls i d v d v d v d vd N NN N MW Heated Samples O iq v o r n o o Tf< 0 0 LO LO CO ON O c s o LO i d i d i d v d cd C v CO 0 0 ON CS CS os cs O n vO o rH rj< rH c vO 00 o CO CO CS CO LO CO CS CS r n 0 0 cs cs cs cs CS CS cs CS 6S § L OO CO rH L? o Sch CO Cs C s CO O r n CS rn CO ON r n OV rH LO C s Ov rn v q VO v q i q v q VO 00 00 rn OV CS rn CS a rn rH t-h r n CS c s CS CO CS c d CO c d c d c d rH w O r n CS CO ^ LO CS CO S 1 LO v O C v O O C N H t H H H r H p H 142 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o CO r n i n d 4 OS CO S o i n VO CN i n o CO CO © i n i n d 4 00 CO CO d 4 d 4 CO CO CO CO 00 o d 4 T “H 00 d o 00 o CN CO o v o 00 00 i n d j OS CN i n t-h T-H v d vd Os d 4 d 4 CO CO CO o O i n 00 © 00 in d 4 00 r-H 00 m in in in in in o VO t-h in in in o o d i n o CN d 4 d 4 VO CN LO oo VO T-H r-H rH d d CO c o c o CO CO vO OS d CO OS VO OS CN o CO o CN d 4 CN d 4 i n CS T-H t-h d d CO CO CO O in d in III A ppendix vO 0 0 d 4 CO T-H c o d 4 cd cd CO CO CO Cs vO cd CO o CO os CN o CN d CO d 4 co d 4 CO OS CN d 4 co q CO 0 0 Hi4 O VO CN d 4 CS i n PH CO CO Os cq t-h CN PH d T-H CN d o d 4 d 4 d 4 d4 d 4 d 4 d 4 d 4 A s D CO CO i n vO q vq T f pH d 4 rn pH PH CO CO CO CO d4 vO n Cs CN VO o co os CN 00 CN d5 CO d 4 rn q O d 4 d4 CO CO VO CO d4 CO os cq cd CO o o Q OO d CN 00 in rn vo CN CN d j CN CN CN O in d CO O OS i n t-h CN d 4 OS OS OS T-H O CO i n d 4 t n o s d 4 T-H r n CN d cd d4 CN CN CN CN CN CN CN CN 3 as CN d4 CN Cs 00 d CN Cs CS d CN r s CO a s d 4 o o Cs CO CO 0 0 Cs 0 0 T-H v d c s c d d OS t- h T-H rn T-H CN m 0) 'S* sfd C/5 T3 QJ <d <U in CN d 4 i n d 4 vq vq CN T-H CO CO CO O O £d < CS d 4 d4 o CO CN CO s? T3 in CO Cs o CO o do i n rn O iS (d CO Cs i n o d 4 CO vO CN i n i n d 4 Os Os r n c d tn \ d cd 00 d as m rp d4 d 4 o 1-H .21 T-H ts d5 COrl ffj o CO VO CN Cs CO c s t n o © o s o CN CN CO Os N vD N d 4 i n CN Cs v d OS CN CO Cs o t-h e(d c o u i n i n i n i n CO CO CO O o cai E c<u a* ax> vO o r n T-H o s so T-H o a s CN o s o s VO i n CO T-H c d CN d 4 00 OS o CN CO in i n i n o VO Cs Os vd c o c o CO in 00 d CN i n T-H d 4 00 Cs OS OS 00 i n i n d 00 00 cd c d CN rn rH CN CN © CN i n PH c d vd CN pH CO CN CO OS v o 0 0 m VO OS o s d CN CN T-H T-H o si o CN t -h vO O VO O d 4 CN o s CO CN 0 0 CO CN i n VO i n O s t-h OS d d ud in d 4 T-H CN rn CN oo CN CN CN a o d 4 CN CN vO 5? in o vd i n CO T-H CN CN o U §" 0 0 PH Q d 4 Cs v o cd CN CN tn a CN rn 0 0 t s CN t n cd CN CN CN si o w co d 4 i n vo ScslCnd'invONOOOSrHr-'rHr'Cr^r-i 143 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. i i cn o o o o ON rH O Cn. i n oo cn r r o CN c n cn c n 3 vd ON © <N CN CN o cn c n 00 in IN o rn d CN in rH CN cn Cs rH CN vo CN CN i n Cs CN vO 00 o CN CN cn cn rH CN o in o in rn vO ON O o ON iCN n vq rH r H T-H cn i n CN o CN d 00 Cs i n 00 T-H ON in vd rn t-h CN in o rn i n t s cn CN CN Cs cn 00 CN vO © 00 t-H 00 O CN t-h c d CS CN CN rn ON VO CN i n ON cn i n c n d cn t s CN cn cn vo CN cn tn rn Cs cn CN VO r f VO VO i n 00 cn cn cn CS vO CN Cs vq Cs Cs CN CN cn cn 00 ON vd CN CN ON rn cn vO LO c 5 i n vd cn c n Cs CN vO VO i n o ON CN CN O CN cn Cs sS o in o cn t n CN 00 o Cs cn i n cn c n cn vO rn o o rn o rf vO o Cs ON Cs in 00 Cs ON 3 ON 00 in vd i n cn c n cn CN r n CN CN CN rH t-h o O Cs ON VO Orn T-H CN 00 CN ON N! d 00 r-H vd CN CN CN c n CN cn cn cn O in d in vo vo vo i n oo 00 o cn CN 00 vd T-H CN rH CN 00 ON CN CN cn cn in LO i n vO cn V O v o o cn c n 0 0 CN V-H cn Cs 00 00 T-H T-H Cs c n c n CN CN cn cn cn c n Cs cn rH o 00 i n 00 Cs o in d O N c n T-H rn r n cn o 00 rn CN CN CN CN T-H CN o in o rn CN o d o o m rH O in d in CV Heated Samples o ON O vO i n ON CN c n o ON t s O O CN Cs © CN o CN T-H ON i n vd T-H CN ON Cs CN CN CN c n cn CN CN i n ON vO c n 00 0 0 c n T* r n T-H CN CN cn CN rH i n d 00 cn 0 0 r H CN rH O T-H cn d in cn T-H o o O cCNn o LO CN o d i n o v d t-H ON vd c n r H r n CN rH rH u. O in in in in cn in rn c5 ON CN VO CN N 1 vq o T-H T-H tn rn vT-H d 0 0 ON 00 00 ON rH T-H $ rH T-H rH CN CN rf ON i n ON ON CN CN 8 8 cn O vO c n N 5 vd T-H CN cn i n Cs VO T-H v 0 o CN ON ON vd vd CN CN CN CN cn cn ON CN T-H 0 0 O n i n ON T“H v q 00 T*H rn Cs rH d ON v d T-H rH r H T-H rH r H cn m O n N to ^ Sch hf ' JCO^i n' ONOOONnHr HHH 144 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o t n CM rn o NO t- h rn lo t h O CM CM O tn 00 CO o CO NO CM CM CM o tn o to CO o in CM NO S o CO r n CO c o rH Cm ON c CM NO t s CM CM CM 00 NO CM CM CM ON tN CM CM p h i n CM pH t n CO CO O no ON t S Cm i n Cm co r —* CO co o in rH CM CM NO CO CO 00 00 CM CM 00 CM CM rr O CO CM CM CO 0 0 CM NO CM i n CM ON CN CM Cm Cm NO CM CN r n O 00 CO CO in NO o CM CO CM o in CO OO CO p 00 c d 00 CO CO CM CO 0 0 0 0 CM Cm 0 0 I S Cm CM Cm T-H Cm Cm CO CO CM ON CO o CM 0 0 rM rM CO c\ 0 0 rH CM r- i n N NO CO CO CN CO ON ON ON rM 0 0 tM ON o O r • t j* t n in o o o O O n t n CO CO 0 0 t n CM i n cd ON cd NO ON CM r n CM VC VC i- • CO o CM O CO Cs NO i n r n ON o i n NO CM NO ON Cm d © 00 t n T-H CM t-h 0 0 CM o co Cm CM CM Cm o NO t- h CM 0 0 0 0 o CM T-H r n CM CM CM o in o tn rn i n 00 p t-H T-H i n O n VO ON Cm ON T-H vC NO PH d o d d o ON ON pH o o CM CM ON 00 p d cd CM NO cd T-H CM 00 T-H r n o in p o o rn 00 tn cd co in o in CM Cm CM CM Cm pH i n O 00 T-H OO 00 5 CM r n CM CO CM CO CM ON Cm CM CM cd CM CO CO CM i n ON VO NO i n d ON d CM T-H CM CM 00 CO O n co i n CM NO i n CN CO i n CM CM CM NO 00 CM CO rn CO CO T-H CM CM i n O 00 CM ON i n CM CM r n MW Heated Samples CO OO pH i n pH i n rH i n TT CM o NO O NO t n Q TT 0 0 CM CM o t s 0 0 o s 0 0 VO t s ON pH T-H ON ON CO 0 0 T-H CM CO CM Cm ON CM CM NO ON 0 0 ON NO ON CM CM ^6" 95 O in o CO T-h NO 0 0 0 0 CO i n t s CM t s rn D. QJ £ CM pH Cm ON in o ON CM c d t— NO t s pH CM o t-* cm co ^ tn r0'tlO C M 300N rH f^ r'r< M r' 145 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. r n C s NO ON NO o o pH rt* cn CN CN NO i n NO i n i n 0 0 0 0 ON CO Cs Cs t s 00 fOf^OinNN^^N COOO'OWvO'CrjOqrH oooocscsoocsoocsod p - <N CO ON __________ N r-iO O C O C O v O N C O rtin O O fn c O rH P On o r-*cscdo6iiSNovbvdvd o d ^ Cs ON n OO N O v O C O N v O oC Ot sOt On MMC Qv Di ni niNnNNO O NQQQo ON 0 0 ON ON r—' ON CN CN N n rn r" t-h m in in o r-o^oinfC \oo^O N M n^ OOONCCNfOrOONCO^ri r; ^ in o ONCNOCscdCNCNCdodr^*--«T— >r-* o o o CN ON NO NO o c n 0 0 o CO ON Cs Cs Cs i n 00 ON 0 0 00 00 0 0 00 0 0 G QJ o CO NO Cs 0 0 Cs ON c n CN r r 0 0 cn i n ON o 0 0 ON CN 00 0 0 0 0 00 0 0 ON 0 0 0 0 QJ D-. X OJ o i n cn NO CO Cs rH o A p pendix IV in 00 O n ON NO NO CO NO r “ 00 0 0 oo 00 0 0 Cs ON o NO i n Cs oo 00 CN r n r n CN 00 CO i n c s i n 0 0 Cs C s C s Cs Cs Cs 00 Cs o »CNrt< §o oC os C^Nc n^ Tnpj iC—s•«— C sC so q oin t-H cdcooNCNodcdodco .Si U4 o ffl 13 oSj C* o o t-h O o CO o CN O O ^ M vo (N (N CN ON C s IN CS c n CN ON Cs CN CN fH o NO Cs On ■<* ON NO NO C s Cs NO C s NO 00 00 o in in in o in h c n oo CN 00 0 0 CO ON c n CN CN CN ON TT r n Cs 00 ON ON o C s Cs CN c n tn NO NO NO NO Cs r ? r f cn O o cn CN 0 0 i n NO 00 Cs 0 0 t n NO NO NO vd Cs Cs C s NO NO o CN rn in rn NO t n rt« 00 0 0 i n Cs ON 00 pH OO Cs C s Cs Cs i n NO NO m o a. s03 cn 13 CJ in G U "S QJ A P & cn o o o rn Cs ON CN r n o ON o in NO c n CN cn Cs C s Cs Cs Cs Cs NO O CL QJ O CcNcnN*inNOCsooc\r^ r n <n cn ^ in vo 146 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. o o rH d vo 00 00 cn 00 ^ r n CN r n i n ON ON rH Cn O n Cn v d CO 0 0 cd o CO Cn ON o © SS v o Cn Cn CS c s v d v d v d v d Cn O in o tn o SS n in o v o rH o in in co COON Cn v q COCO LO LO in Cn i n cn o c s VO 0 0 0 0 c n i n vq vO o on ^ CS c n Cn c s c n v o o ON ON ON ON t n Cn ON t-H i n VO CO ON CN CO CN O CS CN O rn o in o o i n 00 ON cn i n i n o ON o ON ON ON ON VO Cn 0 0 ON r-H c n CS rn cs t- h r n oo VO rn rn rs cn rH r-H O CS cs rn ON rn rn rn O cn o -T-H ^©CNONONONONoqoNinoLno M h h ONNNO^OCO Nc^'OvoifirlHindvd^cs o 00 © cs rH O ON vO Cn c s VO o t n Cn rH r n VO NO ON i n v d i n v d ON ON 0 0 0 0 cn vo 00 CO o vo c n Cn v o VO c n OO ON o o o 0 0 Cs Cn 0 0 I S Cn ON o cs o 00 CV Heated Samples i n co cn ON ^ o cs ^ cn ^ o t-h CO CO t- h rH rH ON o rH cn in Cn r n ON r n O in o in CS o vO 0 0 vO Cn r r i n c d Cn Cn o cs o 00 ON c s vO vO o VO i n Cn 0 0 fiS § o in Cn v o CS c n CS rt* o rH o 00 i n On O i n Cn v d Cn 0 0 c d v d SS § LO O CO rH ' in 9 ^ h h o ^ 00 00 i n 00 cn in t s 1 vd Cn o Cn rH Cn VO CS Cn i n vd vd in O n c n Cn i n CN v o ON cs in in in 0 0 Cn cd cd On On 00 oo On ON t- h 0 0 c n 0 0 ON c d CS c d u u S' O n N (O cn Si Nt O^ i n' ONOOONr HnT- THr l 147 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. T t V£> O o "^r CO C s i n in O CN CN VO 00 CO CN i n Cs vd vd LO Cs N 5 00 ON rH i n VO o CO o CO 0 0 cvj v o CN CO CO Cs i n t r i Cn v d t n vd vd v d o in o m M N O i n pH Cn o 00 Tt< O IN Cn CO VO CO Cs CN r f VO ON Cs CN ON 00 oo in Cn ON rH rH o 0 0 rH CO ^ Tt« CS o rn OO 0 0 i n VO 00 i n CN o d rH rH ON pH r n 00 ON § in o VO r n o o rn © rf OOr n Cn Cs Nj* 00 CO ON i n 0 0 CO Cn 0 0 Cs vO 00 d d ON Os ON ON r n r n rn o o CN o O i n CO CO co CN 00 CN o Cn C s 00 VO o CO 00 ON CO t s 00 00 Cn CN r n i n i n in O GO VO VO CO 0 0 rn r n CO o o d d d CN d O in o in vO C s VO 0 0 Cs 0 0 CN CN rH rH CN d d d d o CN rn rH rn i n i n CO LO CO ON rn rH CO 00 Os 0 0 c d CN v d Cs Cs 00 O NO ON C s o 65 § o o in rH MW Heated Samples o in d m C s VO C s CN CN C s VO © 0 0 t-h in in t-h © rn O Cn in OO 3 in ON cd CO Cn rH ON i n ON i n o r n Cn rH CO ON VO CO CN CN CN CN t-h CN ^ O ^ iri in r—t CO v O VO © rr ON in rn r n (N O v£> ON ^ CN CO r n ^ CN CN C s o CN 0 0 ON Cn vO ON CN C s ON CN C s d 0 0 O T-H CN O O O CN CN CO in co U o d rH rH CO 00 is VO i n CN i n rH O vO r n i n C s v o C s C s vO o r H CN CO ^ in i H C N C O ^ i n v O C s O O O N r H r H r H r H r H r H 148 fWWT'' Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix V Representative *C spectra of unheated samples Supernatant 1 Supernatant 2 i — i— |— i— i— i— i— |— i i 150 10 0 i i | 50 i i i PPM 149 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Supernatant 3 Supernatant 4 150 100 150 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Supernatant 5 Residue 150 100 151 ■wff-r Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix VI Representative 13C spectra of conventional heated samples Supernatant 1 f pFrWwTST J Supernatant 2 — |— 150 i— i— i— i— |— 100 i— i— i— i— |— 50 i— i— r PPM 152 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Supernatant 3 Supernatant 4 i — i— |— i— i— i— i— |— :— i— i— i— |— i— i— i 150 100 50 =>P^ 153 Reproduced w ith permission of the copyright owner. Further reproduction prohibited without permission. Supernatant 5 Residue 150 100 50 154 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Appendix VII Representative 13C spectra of microwave heated samples Supernatant 1 Supernatant 2 15 0 10 0 155 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Supernatant 3 Supernatant 4 iw hL a . r*Jt 1--- 1---1--- 1--- 1--- 1--- 1--- 1--- 1--- 1--- 1--- 1 1 I I I 150 100 50 156 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Supernatant 5 Residue 150 100 157 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.