ANTIOXIDANT PROPERTIES OF SORGHUM, A Dissertation by JOSEPH MOBUTU AWIKA Submitted to the Office of Graduate Studies of Texas A&M University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY May 2003, Major Subject: Food Science and TechnologyUMI Number: 3088114 UMI UMI Microform 3088114 Copyright 2003 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346, ‘Ann Arbor, MI 48106-1346ANTIOXIDANT PROPERTIES OF SORGHUM by A Dissertation JOSEPH MOBUTU AWIKA Submitted to Texas A&M University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Approved as to style and content by: : Tow ‘Rodney (Chair of Committ eo Bide, Ronald Richter (Member) Lloyd’W. Rooney { (Chair of Food Science an Technology Faculty) May 2003 GLE@ (Member) er ogee Luis Cisneros-Zevallos (Member) (eave ‘Mark Hussey (Head of Department) Major Subject: Food Science and Technologyiii ABSTRACT Antioxidant Properties of Sorghum. (May 2003) Joseph Mobutu Awika, B.Sc., Egerton University; MS., Texas A&M University Chair of Advisory Committee: Dr. Lloyd W. Rooney Sorghum varieties grown in Texas between 1998 and 2002, as well as processed sorghum products, were analyzed for antioxidant potential using three methods; oxygen radical absorbance capacity (ORAC), 2,2’-azinobis (3-ethyl-benzothiaziline-6-sulfonic acid) (ABTS) and 2,2-diphenyl-1-picrylhydrazyl (DPPH). Phenolic antioxidants from these samples were separated and characterized by HPLC. Two extraction solvents, 1% HCI in methanol (Me-HCh, and 70% aqueous acetone (Ac-aq), were compared. The Me-HCl generally extracted sorghum antioxidants better than Ac-aq. This effect was more significant in black sorghums than brown sorghums. The ORAC activities (Trolox equivalents, TE) among black and brown sorghum brans were high (1,000 ~ 3,100 mol TE/g) compared to common fruits and vegetables (80-900 mol TE/g). Retention of antioxidant activity after processing was 57 — 78% for baked, and 70 100% for extruded sorghum products relative to raw samples. Antioxidant values by the ORAC method were 2-3 times higher than ABTS or DPPH values for the sorghums. However, the three methods correlated highly with one another (R’> 0.95). The ABTS method was overall the most suitable for sorghums.Brown sorghum grains and brans had high procyanidin contents (21-58 mg/g) compared to blueberry (20 mg/g), as measured by HPLC. Polymers with DP>10 were the major procyanidin constituents (66-84%) of brown sorghums. The relative ratio of oligomers (DP<10) to polymers (DP>10) increased significantly in processed products. Black sorghum brans had high anthocyanin content (4.5-11.0 mg/g) compared to commercial sources (0.8-10.0 mg/g). Luteolinidin and apigeninidin were the major anthocyanins in the black sorghums, accounting for about 50% of total anthocyanins. ‘The red sorghum bran had high levels of naringenin (1.0 mg/g) compared to the other sorghum brans (0.17-0.26 mg/g). Specialty sorghums are a rich source of different phenols with high antioxidant potential and are a commercially viable source of these compounds for foods or pharmacological applications. High retention of antioxidant activity in processed products implies the sorghums are valuable food ingredients.ACKNOWLEDGEMENTS I wish to extend big thanks to Dr. Rooney for making all this possible. Special thanks to Dr. Waniska for being very helpful through all stages of the study. Sincere gratitude goes to Dr. Richter and Dr. Cisneros for serving on my committee and general help. I especially thank Dr. Cisneros for allowing me the use of his laboratory facilities, and also his invaluable help with the initial DPPH analysis. Special thanks to Dr. Prior of the Arkansas Children’s Hospital, Little Rock, AK, and his staff, Dr. Gu and Dr. Wu for being very helpful with time and resources for normal phase HPLC and ORAC analysis. Many thanks go to Dr. Senseman, and Audie Sciumbato (Soil & Crop Sciences) for generous help with the initial HPLC analysis. 1 also thank Dr. Keeton and Bryan (Meat Science) for allowing me access to facilities in their laboratory. The CQL lab crew was all very helpful at various stages of this study. Thanks to Cassandra for helping with acquisition of most of the ‘tools’ for the work and her general advice. Many thanks to Linda Dykes for being very helpful with parts of the analysis, I salute David Acosta and Mark Baron for assisting with the dirty work! Thanks to the rest of CQL members for their general support.vi TABLE OF CONTENTS Page ABSTRACT... esse cceeceeee ACKNOWLEDGEMENTS. TABLE OF CONTENTS. wi LIST OF TABLES.. viii LIST OF FIGURES . CHAPTER I INTRODUCTION ............ I LITERATURE REVIEW........ Sorghum tannins and their antioxidant potential 3 Sorghum anthocyanins and their potential benefits 9 Extraction of phenolic compounds... wl Separation of phenolic compounds using HPLC... 15 Evaluating antioxidant activity of phenolic compounds 220 Ill MATERIALS AND METHODS... Samples... Sample extraction. Analytical procedures......... HPLC analysis. Special note. IV RESULTS AND DISCUSSION... Extraction of sorghum antioxidants Methods for measuring antioxidant activity Comparing ABTS and DPPH antioxidant methods on sorghum extracts. : Use of antioxidant activity to compare extraction efficiency of Ac-aq with Me-HCI Antioxidant properties of pure phenols HPLC separation of sorghum phenolics .vii CHAPTER Page Environmental and varietal effects on phenol and antioxidant levels of sorghums ..... _ Effect of processing on antioxidant activity of sorghum Dietary implications of the phenol and antioxidant status of the processed sorghums Comparison of sorghum products with other dietary sources Vv SUMMARY, 100 LITERATURE CITED ..... 103 APPENDIX A. AIS APPENDIX B.. 116 APPENDIX C... 7 VITA ... 118viii LIST OF TABLES TABLE Page 1 1 rit Vv VI vu vill Ix XI x XI xIV xv XVI ‘Tannin and phenol levels in Me-HICI and Ac-aq extracts of brown sorghums Anthocyanin levels in sorghum brans extracted with Me-HCl and Ac-aq. Antioxidant activity of sorghum grains, brans and products measured by ORAC, ABTS and DPPH.. . . A summary comparison of the different antioxidant methods .AT Comparing antioxidant activity of brown sorghums extracted for different times . Antioxidant activity of flavonoids... Antioxidant activity of phenolic acids. 53) Molar characteristics anthocyanin molecules . 61 Anthocyanin contents of black and red sorghum brans determined by HPLC. 64 Procyanidin content of brown sorghums compared to those of freeze- dried cocoa and blueberry 70 ABTS antioxidant and phenol levels among sorghums and other cereals.........75, Effect of processing on procyanidin polymer contents of Sumac (SU99) bran measured by normal phase HPLC. — Change in measurable phenolics due to extrusion .. ORAC levels per serving of bread and cookies made with sorghum brans........94 ORAC levels in sorghum brans and extracts relative to common fruits...... 97 Anthocyanin content of sorghum brans relative to common fruits and vegetables... 98ix LIST OF FIGURES FIGURE 1. Che al structure of common anthocyanidins.... 2. Typical crude anthocyanin spectra for black sorghum bran extracted with Me-HCI and Ac-ag. «.....esssssssess 3. Correlation coefficients for the three antioxidant methods used. 42 4, Comparing antioxidant DPPH methods ivities of brown sorghum brans using ABTS and 5. Comparing antioxidant activities of black sorghum brans using ABTS and DPPH methods . 43 6. Antioxidant activities of brown sorghum brans extracted in Me-HCl and Ac-aq 50 7. Antioxidant activities of black sorghum brans extracted in Me-HCI and Ac-aq 8. Comparing antiradical activity of phenolic compounds in DPPH versus ABTS-EtOH and ABTS-PBS systems. 9. Correlation coefficients between ABTS in ethanol (ABTS-EtOH) and aqueous buffer (ABTS-PBS) versus DPPH for different phenol groups....0.0...56 10, Effect of the number of OH substituents on the antioxidant activity of phenolic compounds in DPPH, and alcoholic (EtOH) and aqueous buffer (PBS) ABTS systems. 58 11. Absorption spectra for selected anthocyanin standards in pH 1.0 buffer solution seceeeeeneenene 60 12, HPLC profiles of Me-HCI anthocyanin extracts from black and red sorghum brans... 13. HPLC anthocyanin profiles of black sorghum (BKO1) bran extracts monitored at 375 nm ..... 265 14, Spectral characteristics of luteolinidin and apigeninidin peaks isolated from black sorghum (BK01) bran by HPLC 66FIGURE Page 15, Normal phase HPLC procyanidin profiles of two brown sorghum grains, Sumac (SU99), and Hi Tannin (HTO1), 69 16. HPLC profiles of Me-HCl flavonoid extracts from brown (SU99) and red (RDO) sorghum brans monitored at 280 nm .. sevens 17. Phenol levels for various sorghum brans compared to wheat bran... 18. ABTS values for various sorghum brans compared to wheat brat...uc.ssssssss000.77 19, Changes in antioxidant activity of sorghum brans after baking. 81 20. Normal phase HPLC procyanidin profiles of Sumac (809s) bran before and afier baking in cookies and bread... es 21, Change in Sumac (SU99) bran procyanidins due to processing in cookies and bread. 84 22. Change in ABTS activity of sorghums due to extrusion. 88 23. Normal phase HPLC procyanidin profiles of Hi Tannin (HTO1) grain before and after extrusion. . 89 24. Change in Hi Tannin (H'T01) grain procyanidins due to extrusion. 1CHAPTERI INTRODUCTION Interest has increased in phenolic compounds from plants due to their potential benefits as antioxidants. Sorghum, like many other cereal plants, contains phenols that are thought to provide natural defense mechanisms against insects and diseases. In sorghum, these phenols are concentrated in the outer layers of the kernel. Decortication to remove outer layers of the kernel can concentrate these compounds 3-7 fold (Hahn and Rooney 1985, Awika 2000). Such fractions possess antioxidant properties comparable to those of berries (Awika 2000) and can be used in foods to provide nutraceutical benefits. Sorghum offers a cost effective source of phenolics due to its storage stability and high grain yield. This is especially important given the rapidly increasing demand for natural sources of nutraceuticals for food. Condensed tannins are a major phenolic component of sorghums with a pigmented testa. They are concentrated in the testa and pericarp of such sorghums. The superior ability of tannins over simple phenols as antioxidants is known (Hagerman et al 1998, Bors et al 2000). Hagerman et al (1998) found tannins were 15-30 times more powerful at quenching peroxy radicals than simple phenols. Bors et al (2000) also reported an exceptionally high radical scavenging potential of condensed and hydrolysable tannins in their reaction with hydroxyl radicals. The tannin contents of sorghums correlated This dissertation follows the style and format of Cereal Chemistry.positively with their antioxidant properties as measured by the ORAC method (Awika 2000). Isolation of these polyphenols from sorghum and measuring their antioxidant ability in purified forms is necessary. This will give a better understanding of how they influence the antioxidant activity of bran fractions. Anthocyanins are becoming increasingly important not only as food colorants, but also as antioxidants. There is a growing commercial need for alternative natural sources of food colorants (Boyd 2000, Francis 2000). Awika (2000) found black sorghum, (Tx430 variety) bran had significantly more anthocyanins than other sorghums, with good antioxidant properties comparable to those of berries. This was the only non-tannin sorghum with such high antioxidant activity, which suggests anthocyanins played an important part in its antioxidant activity. The potential of this sorghum for commercial exploitation cannot be overestimated. It is important to isolate and determine the antioxidant properties of the anthocyanins in more purified forms. ‘Numerous other phenolic compounds exist in sorghum brans (Hahn 1984, Gous 1989), Data is needed on the antioxidant properties of these sorghum phenols. Such phenols are known to have different antioxidant properties and their relative proportions can affect the overall antioxidant activity of bran samples. In order to determine the antioxidant activity of the sorghum phenols, itis essential to extract them in a consistently efficient manner. This will allow for accurate estimation of their activity as well as allow for comparison with other natural antioxidant sources, Several methods are available for extraction of phenols from different sources. In sorghums, acidified methanol is reported as most efficient (Hahn and Rooney 1985,Gous 1989), whereas for fruits and vegetables 70% aqueous acetone is reportedly most efficient (Kallithraka et al 1995, Garcia-Viguera et al 1998). However, the acidified ‘methanol and aqueous acetone methods are seldom compared side by side. The main goal of studying natural antioxidants is to improve human health. The most effective way to deliver the antioxidants to consumers is through foods that are consumed widely and consistently. Prior investigations in our lab have shown that specialty sorghum brans high in phenolics can produce good quality breads (Gordon 2001), and cookies (Mitre-Dieste et al 2000). However, the rate of retention of sorghum antioxidants in these products is not known. This information is important if the specialty sorghums are to be promoted as a viable source of antioxidants for human health, Numerous methods are available for estimating antioxidant activities of compounds in vitro, However, since the main goal is to improve human health, methods that would most closely predict the bioeffects of these compounds are preferable. So far ORAC is the only in vitro method that has correlated in vitro activities of various antioxidants with in vivo plasma antioxidant status (Cao et al 1996). However the method is expensive and hence not widely available. Other methods with similar or better predictive value for sorghum antioxidants should be assessed. ‘This work aims to detail the effects of sorghum phenolic components on overall antioxidant activity of sorghum brans, and how processing affects their activity. Itis a continuation of preliminary work done in our laboratory as reported by Awika (2000).Specific objectives were » 2) 3 4) To determine the most efficient method for extracting sorghum antioxidants. ‘To compare different methods for antioxidant analysis to determine the most suitable for sorghum. Isolate and identify the sorghum compounds responsible for antioxidant activity using HPLC. To determine the effects of processing on antioxidant activity of sorghum.CHAPTER IL LITERATURE REVIEW Sorghum tannins and their antioxidant potential ‘Tannins are high molecular weight polyphenols, and are classified as hydrolyzable or condensed. Hydrolyzable tannins contain a central core polyhydric alcohol such as glucose and hydroxyl groups which are esterified wholly or partially by gallic acid or hexahydroxydiphenic acid. Condensed tannins, found in sorghum, are mainly polymerized products of flavan-3-ols and/or flavan 3,4-diols. Sorghum tannins have generally been viewed as undesirable due to theit antinutritive properties, i., they complex with food macromolecules reducing their digestibility Price and Butler 1980, Salunkhe et al 1982, King-Thom et al 1998). However, current data suggest that tannins may have nutraceutical benefits due to their powerful antioxidant activity (Kinsella et al 1993, Clifford 1996, Hagerman et al 1998, Heinennonen et al 1998). Consequently research efforts are directed at understanding their precise mode of action and bioavailability in biological systems and how their consumption can be enhanced in human diets. Specialty sorghum brans rich in these ‘compounds (Awika 2000) need detailed analysis to determine how they can be used in foods. The activity of tannins as antioxidants is thought to be two-fold; i) they may chelate transition metals (Cu and Fe) making them unavailable for Fenton-type reactions (Bors et al 1990, Carbonaro et al 1996); ii) they may act as scavenger antioxidants by donating an electron to highly reactive free radicals and quenching the reactivity of the radical by6 delocalizing the unpaired electron on the phenolic ring (Husain et al 1987, Chimi et al 1991). Conflicting data exist, however, on ability of tannins to chelate transition metals. Carbonaro et al (1996) found no evidence to suggest that the ability of tannins to reduce hydroxyl radical formation was due to a chelating activity. They concluded it was purely due to a direct interaction of the polyphenol molecule with hydroxyl radicals. On the other hand, Hagerman et al (1998) found both condensed and hydrolysable tannins were able to chelate iron (Fe™) at a pH range of 6.4 ~ 8.4, besides reacting directly with free radicals. Similar results were reported by Moran et al (1997) for low molecular weight polyphenols at pH 5.8 and 7.4. Hagerman et al (1998) found tannins were 15-30 times more powerful at quenching peroxy radicals than simple phenolics or Trolox. ce condensed and hydrolyzable tannins are structurally different, but were equally effective as antioxidants, they concluded that high molecular weight and the proximity of many aromatic rings and hydroxyl groups are more important to free radical scavenging by tannins than specific functional groups. Ariga and Hamano (1990) found that the ability of oligomeric, flavonoids to scavenge azobis-generated peroxyl radicals was proportional to their degree of polymerization. Hagerman et al (1998) also showed that tannins, unlike simple phenols, were unable to act as prooxidants or were very weak prooxidants. Similar results were reported by Bors et al (2000), who suggested this could be because procyanidin and gallate o-quinone, unlike flavan(ol) 0-quinones which can act as prooxidants by forming reactive oxygen species through futile redox cycling, are capableof producing oligomeric compounds by various pathways. Such phenolic coupling reactions retain the number of hydroxy groups which enhance their antioxidant capacity. ‘An important factor to consider in assessing the activity of dietary components is their availability and/or activity after ingestion. Since tannins complex with proteins and other macromolecules making them resistant to digestion, they are likely to remain in the digestive tract and not be absorbed and transported to other tissues (Jimenez-Ramsye et al 1994). This way, they may serve a unique role as antioxidants by sparing other antioxidants and thus indirectly increase antioxidant level in other tissues (Hagerman et al 1998). They may also be able to protect proteins, carbohydrates and lipids in the digestive tract from oxidative damage during digestion (Marshall and Roberts 1990). To evaluate antioxidant ability of tannins when complexed with proteins, Hagerman et al (1998) used tannin-gelatin complexes as antioxidants in metmyoglobin assay. These complexes retained at least half the activity of free tannins. It is possible proteins complexed this way may be less susceptible to oxidative damage. Recent data suggest that tannins and other flavonoids are absorbed by humans in larger quantities than previously thought (Pietta 2000, Ross and Kasum 2002). Deprez et al (2001) reported that procyanidins could be absorbed through the intestinal cell monolayer, but only up to trimers. However, the interflavan bond in the procyanidins ‘was found to be unstable in acid environments, suggesting the higher molecular weight procyanidins could be broken down in the stomach (Spencer et al 2000). These authors tested depolymerization of procyanidins in simulated gastric juice (pH 2) and found that tetramers to hexamers degraded to monomers and dimers in 1.5-3.5 hr. This wouldpotentially improve the absorption of the procyanidins in the small intestine, and hence their bioavailability. ‘A major part of the ingested flavonoids is, however, apparently not absorbed directly, but is largely degraded by the intestinal microflora. The bacterial enzymes catalyze several reactions, including hydrolysis, cleavage, of the heterocyclic oxygen- containing ring, dehydroxylation, and decarboxylation (Pietta 2000). The net products are several phenolic acids which can be reabsorbed through the colon, enter circulation and offer antioxidant protection (Pietta et al 1997, Deprez et al 2000, Tapiero et al 2002) In general, tannins appear to be the most effective natural phenolic antioxidants in- vitro. Available data suggest that tannins may also produce health benefits in-vivo through direct absorption or degradation into phenolic acids in the colon, or by complexing with and protecting food molecules from oxidative damage. Since specialty sorghum brans are a rich source of these compounds (Hahn 1984, Awika 2000), they could be incorporated in diets as part of natural food ingredients. Work in our laboratory using these sorghum brans in bread and cookies has shown they produce good quality products (Mitre-Dieste et al 2000, Gordon 2001, Rudiger 2003). It is essential to assess the nutraceutical properties of such products to determine how processing and storage affect the interaction of tannin with food macromolecules and its antioxidant properties. This may help determine significance of sorghum as a source of tannins for nutraceutical applications.Sorghum anthocyanins and their potential benefits Anthocyanins are natural colorants belonging to the flavonoid family. They are widely distributed in flowers, fruits and vegetables, and are thought to play a role in plant resistance to insect attack (Strack and Wray 1993). All anthocyanins are natural glycosides and acylglycosides of anthocyanidins (Wang et al 1997), Figure 1 shows the general structure of common anthocyanidins found in nature. Most of these anthocyanidins have an OH group on the C-3 of the C ring. However, some —— Ri Ry H H ‘OH H H on ou H OCH OCHs ou H Delphinidin OH OH OH H Petunidin OH OCH OH H Peonidin H OCH; on H Luteolinidin On H H H Apigeninidin H H H H Fig 1. Chemical structure of common anthocyanidins.10 anthocyanidins, with a small distribution in nature, lack the OH group on C-3 and are normally referred to as 3-deoxyanthocyanidins. These include luteolinidin and apigeninidin (Fig. 1) (Sweeny and Iacobucci 1981, Clifford 2000). The common anthocyanins are either 3- or 3,5- glycosylated, When the number of sugar residues is higher than three, they may be attached to the basic molecule with altemating sugar and acyl linkages. ‘Anthocyanins are thought to have some therapeutic benefits including, vasoprotective and anti-inflammatory properties (Lietti et al 1976), anti-cancer and chemoprotective properties (Karaivanova et al 1990), as well as antineoplastic properties (Kamei et al 1995), Anthocyanins are considered to contribute significantly to the beneficial effects of consuming fruits and vegetables (Wang et al 1997). Most of the anthocyanins that have been identified were isolated from fruits and vegetables, Limited data exists on the types and levels of anthocyanins in cereals, probably because they have never been regarded as a commercially significant source. Nip and Bums (1969, 1971) were able to isolate and identify apigeninidin, apigeninidin- S-glucoside, luteolinidin and pelargonidin-3,5-diglucoside in red and white sorghum varieties by paper chromatography. Gous (1989) also reported luteolinidin and apigeninidin from a black sorghum variety. Cyanidin and pelargonidin were also reported in corn (Francis 1989), and sorghum (Yasumutsa et al 1965) Available quantitative data on anthocyanins vary widely from author to author, partly because of different standards used. In general, values of 1.2-23.6 mg/g forblueberries (Prior et al 1998, Mazza and Miniati 1993), blackberries 3.9-15.5 mg/g, cranberry 3.5 mg/g, raspberry 0.9-19.5 mg/g (Mazza and Miniati 1993), and red cabbage 1.6 mg/g (Timberlake and Henry 1988) have been reported (all values converted to dry ‘weight basis), Gous (1989) reported anthocyanin levels of 3.7 mg/g for a black sorghum variety. However, comparing values across authors may not be valid due to widely varying standards and extinetion coefficients, as well as extraction methods used by each author. Antioxidant properties of anthocyanins have been reported (Wang et al 1997, Satué- Gracia et al 1997, Lapidot et al 1999, Saint-Cricq de Gaulejac et al 1999). Awika (2000) also attributed high antioxidant properties of a black sorghum variety, which had a very Jow tannin level, to its anthocyanin content. Data from several authors on the antioxidant activity of anthocyanins and their aglycons are conflicting. Tsuda et al (1994) found that aglycons were better antioxidants than glucosides, whereas Wang et al (1997) reported that some glucosides were better antioxidants than aglycons. Different systems and catalyst used by different authors may be responsible for the conflicting data. For example, Lapidot et al (1999) reported that in a system catalyzed by H2O>-activated myoglobin, malvidin-3-glucoside was the best antioxidant, followed by catechin, whereas in a system catalyzed by an iron redox cycle catalyzer, resveratrol was the most efficient and catechin the least efficient among four flavonoids they tested (malvidin, catechin, resveratrol and malvidin-3-glucoside) In general, mechanistic studies of antioxidant activity of individual anthocyanin molecules have not provided meaningful information. Apparently, differenthydroxylation and glycosylation patterns modulate the antioxidant properties of anthocyanins. Saint-Cricq de Gaulejac et al (1999) reported that anthocyanins in red wine act synergistically to provide radical scavenging activity compared to pure anthocyanin molecules. This shows that measuring antioxidant activities of individual anthocyanin molecules, though insightful, may not give a good picture of the effectiveness of anthocyanins as antioxidants in their natural form. Such data would need to be compared against the crude forms of anthocyanins as they exist in nature. Antioxidant properties of sorghum anthocyanins and their individual constituents have not been reported. Such data is essential if sorghum anthocyanins are to be promoted as a viable commercial source of natural food pigments. This is becoming increasingly important with the rising demand for natural sources of food colorants (Boyd 2000), ‘A major concern over the exploitation of natural food pigment sources is the stability of anthocyanins in food systems. Anthocyanins have poor color stability relative to synthetic food colorants which presents a challenge to scientists. Use of natural anthocyanins in beverages is hampered by their sensitivity to light, pH changes, and especially their bleaching by sulfur dioxide which is a common preservative (Harborne 1988). However, the 3-deoxyanthocyanidins (which include apigeninidin and luteolinidin) were reportedly very stable in acidic solutions relative to other anthocyanidins commonly found in fruits and vegetables (Sweeny and Iacobucci 1981). The lack of oxygen at the C3 is believed to improve their stability. These 3- deoxyanthocyanidins are not widely distributed in food plants (Clifford 2000), but haveB been identified as major anthocyanins in sorghums (Nip and Burns 1969, Gous 1989). This points to the potential of sorghum anthocyanins as a viable commercial source of anthocyanins. Mechanisms for improving anthocyanin stability are known. Timberlake and Bridle (1980) reported that reacting anthocyanins with other flavonoids such as flavan-3-ols could stabilize the quinonoidal base and improve resistance to SO. Such reactions may occur during extraction of anthocyanins from samples that contain flavan-3-ols (Haslam 1980, Dallas et al 1996, Remy et al 2000), like brown sorghums, and this may further improve stability of such sorghum anthocyanins. Acetylation also improves stability of anthocyanins (Francis 1989). Sugar attachment and methylation of one or more of the free hydroxyl groups may also improve pigment stability (Cooper-Driver 2001. Co- pigmentation is another mechanism that improves stability of anthocyanins. Co- pigments are important since under very weakly acidic conditions (pH 4-6) as is typical in plant cell vacuole, and in absence of other substances, most anthocyanins exist substantially in stable colorless forms (Cooper-Driver 2001). At this pH range anthocyanin structural types exist in equilibrium as follows: quinonoidal anhydro base <> flavylium cation <= carbinol bases Co-pigments may act either inter- or intramolecularly (Figueiredo et al 1996). They have little or no color in themselves, but probably stabilize and enhance the color of anthocyanins by intermolecular hydrophobically reinforced x-x stacking of their aromatic nuclei with those of the pigment (Brouillard et al 1991, Mistry et al 1991).14 Such a-7 interactions may stabilize the flavylium cation/anhydro base forms which prevent reaction with water to form the colorless carbinol base forms (Cooper-Driver 2001). Various molecules have been identified as good natural copigments, including hydroxycinnamoyl esters such as chlorogenic acid, galloyl esters (tannins) and flavone and flavonol glycosides (Figueiredo et al 1996) Extraction of phenolic compounds Several methods are used to isolate phenols and assess their antioxidant activity. Since most sorghum phenols are concentrated in the bran fractions (Hahn and Rooney 1985, Awika 2000), separation of these fractions through abrasive decortication is an effective way of concentrating the phenols. These fractions are also rich in dietary fiber and phytin that can provide additional benefits in foods. Extraction of phenols from fruits and vegetables for quantitative determination is usually achieved using a polar solvent like methanol, acetone, ethanol and acidic methanol, among others (Kanner et al 1994, Prior et al 1998, Velioglu et al 1998). Hahn 1984, and Gous (1989) found 1% HCI in methanol to be the most efficient solvent for extracting sorghum phenolics. On the other hand, 70% aqueous acetone is reported as the most efficient at extracting fruit phenolics (Kallithraka et al 1995, Garcia-Viguera et, al 1998), Kallithraka et al (1995) found 70% aqueous acetone to be most effective at extracting grape seed procyanidins. However, most of the solvent comparisons have not involved the acidified methanol and aqueous acetone side by side. Garcia-Viguera et al (1998) compared 1% HCI in1s ‘methanol with 70% aqueous acetone, among other solvents, for extraction of strawberry anthocyanins. They concluded that the aqueous acetone was most efficient at extracting the anthocyanins, However, they stated that their conclusion may have been due to the difficulty involved in post-extraction processing (prior to analysis) of acidified methanol and other solvents relative to aqueous acetone, rather than their actual extracting power. Also, recently Lu and Foo (2001) observed significant anthocyanin reaction with aqueous acetone to form pyranoanthocyanins during extraction. They reported that the degree of the re: jon depended largely on duration of solvent-anthocyanin interaction, as well as temperature. For example, they found that blackcurrant anthocyanin HPLC peaks almost completely disappeared after three days in 70% aqueous acetone at 40°C, with a concomitant increase in pyranocyanin peaks. Hence depending on the extraction regimen, aqueous acetone may greatly underestimate anthocyanin content of a sample. Due to the controversies, it is essential to compare the two solvents side by side on sorghum fractions to determine the most effective method for extracting sorghum phenols. This information is not only essential for consistent analysis and comparison of different samples, but also for potential extraction for commercial exploitation. Separation of phenolic compounds using HPLC High performance liquid chromatography (HPLC) has become very popular due to its versatility, precision and relatively low cost. Most frequently the method uses reversed-phase Cs or Cx columns in conjunction with an aqueous mobile phase; and methanol and acetonitrile buffers as modifiers (Escarpa and Gonzalez 2001, Chen et al 2001). However, quantitative comparison of results as reported in literature is not easy16 because the extraction and chromatographic techniques vary. There is controversy in. sample preparation methods; some authors use solid-phase extraction with Cis or strong anion-exchange (SAX) cartridges; others use liquid-liquid extraction with different organic solvents like ethyl acetate or diethyl ether; while some inject samples directly without any preparation steps (Rodriguez-Delgado et al 2001). Depending on the nature ” of the compounds, recoveries are different. Sample preparation procedures for HPLC analysis have been studied (Schieber et al 2001, Malovand et al 2001). Malovand et al (2001) concluded that liquid-liquid extraction was more efficient than solid-phase extraction for analysis of trans-resveratrol and other polyphenols in wine by HPLC. Schieber et al (2001) used a stationary phase with polar endcapping to separate phenolic acids and flavonoids of apple and pear, and obtained good resolution of flavonol glycosides. Mataix and Ludue de Castro (2001) used a method based on coupling of continuous liquid-solid extraction, evaporation, HPLC individual separation and photometric detection to separate and determine anthocyanins in wine, The method was more sensitive than the batch manual method, with a wider determination range and better linearity of calibration curves. Chen et al (2001) used an acid-catalyzed hydrolysis process to liberate flavonoids and phenolic acids from their bound form and help overcome low resolution between flavonoids and phenolic acids. They were able to simultaneously determine flavonoids and phenolic acids in cranberry juice using this method. Putman and Butler (1989) developed a HPLC procedure to fractionate high molecular weight procyanidins (condensed tannins) from sorghum. They used a short7 reversed-phase C18 column to elute procyanidins through a combination of linear and step gradient coupled with a fast flow rate. Solvents used were acetic acid, methanol, and water. They included phytic acid and EDTA (ethylene diaminetetraacetic acid) in solvents to counter the effect of trace metals that may reside in the column. Through this method, they were able to identify three procyanidin polymers; B-2 (epicatechin dimer), B-3 (catechin dimer), and B-9 (epicatechin trimer), and isolate several other higher molecular weight procyanidins. Using a diode array detector they were able to show that peak separation was due to procyanidin chain length, with the lower molecular weight polymers eluting first. The monomers of catechin and epicatechin were not retained by the column (eluted with solvent front). Rigaud et al (1993) developed a normal phase HPLC method coupled with UV detection to separate procyanidins on a molecular weight basis. They were able to separate procyanidins up to tetramers for grape seed and up to pentamers for cacao beans. Oligomers and polymers beyond tetramers tended to fuse into broad unresolved humps, making the method unsuitable for quantification. Improvements of Rigaud et al (1993) method that coupled the normal phase HPLC with fluorescence detection was used to successfully separate oligomers up to pentamers from cocoa beans on a molecular weight basis (Adamson et al 1999, Hammerstone et al 1999). The authors proposed the method as a more effective tool for quantification of procyanidins over the less specific spectrophotometric methods like the Vanillin-HCI. However, since they were unable to resolve polymers with DP>10, the method underestimated procyanidin content,18 Gu et al (2002) further optimized the method used by Adamson et al (1999) and Hammerstone et al (1999) for berries, cocoa, and sorghum procyanidins. They were able to resolve the polymers with DP>10 as a single peak that could be quantified, in addition to the monomers and DP 2 ~ DP 10 oligomers. They found the high molecular weight polymers were the major procyanidins in brown sorghum bran, cocoa, cranberries and blueberries. This method allowed for a more effective quantification of procyanidins by HPLC than the previous methods. Using reversed phase HPLC-ESI MS, the authors identified epicatechin as the polymer extension unit and catechin as the major chain terminating unit (89%) in the sorghum bran procyanidins. This agreed with data obtained by Gupta and Haslam (1978), and Gujer et al (1986). HPLC presents a powerful tool for quantification of procyanidins in sorghums due to its specificity and ability to detect contribution of different chain length procyanidins to the total procyanidin content of a sample. This is important in health and food applications since polymer chain length affects organoleptic, antioxidant and other health properties of procyanidins (Rigaud et al 1993, Tebib et al 1997, Lotito et al 2000). Several phenolic acids have been isolated from sorghum using HPLC. Hahn (1984) used a multistep gradient with acetic acid-water, and butanol-methanol as solvents; and a non-polar C18 column to separate phenolic acids in sorghum. He was able to isolate and identify gallic, protocatechuic, gentinsic, p-hydroxybenzoic, chlorogenic, caffeic, syringic, p-coumaric, sinapic and ferulic acids. To obtain crude phenolic acids extracts, he used aqueous acetone and ethyl acetate for free phenolic acids, and base (NaOH) hydrolysis to extract bound and esterified phenols. Waniska et al (1989) and Gous19 (1989) used similar methods to isolate phenolic acids from sorghum. They additionally identified salicylic, vanillic, sinapic and cinnamic acids from their sorghum samples. However, most phenolic acids in sorghum are bound to cellular matrix and are only extractable in minor quantities without acid/alkaline hydrolysis. Their contribution to antioxidant activity of sorghum extracts may thus be minimal. Schieber et al (2001) used a recently available column (Aqua 5 um C18) coupled with a diode array detector to separate polyphenols in apple juice and pomace. The column was developed for the separation of very polar analytes which were usually not sufficiently retained on common reversed phases. Using acetic acid in water, and acetonitrile as solvents with simultaneous monitoring at 280, 320 and 370 nm, they were able to isolate and identify 26 phenolic compounds including procyanidins, catechin, flavonols and phenolic acids. Similar work using reversed-phases have consistently given fewer peak resolutions (Lopez et al 2001, Martinez-Ortega et al 2001, Mateos et al 2001, Ferrara et al 2001). Coupling of HPLC and UV-visible spectroscopy (HPLC/photodiode array detection) has been the most common method to identify and quantify anthocyanins (Anderson 1985, Dallas et al 1996, Garcia-Viguera et al 1997, Wang et al 2000). Wang, et al (2000) identified 13 anthocyanins in highbush blueberries using a reversed phase column with a photodiode array detector. They used aqueous formic acid and 100% methanol as solvents. Revilla et al (2001) on the other hand used acetonitrile in water at pH 1.3, and perchloric acid as solvents for determination of anthocyanins in red grapes and red wine, They were able to obtain 15 peaks with 9 identified anthocyanins. Use of20 formic acid and acetonitrile as solvents also gave a good resolution and identification of 15 anthocyanins and anthocyanin esters in red table grapes (Carrefto et al 1997). Various solvent systems are apparently effective for separation of anthocyanins in HPLC systems, However, quantitative comparison of results may be a problem Evaluating antioxidant activity of phenolic compounds Numerous methods are used to evaluate antioxidant activities of phenolic compounds in foods or biological systems. Most antioxidant activity assays consist of accelerating oxidation in a lipid system, usually by heat and excess oxygen supply, then monitoring oxygen consumption, substrate loss or product formation (Bondet et al 1997, Fukumoto and Mazza 2000). Because many factors affect oxidation, including temperature, oxygen pressure, metal catalyst, composition and form of fat, results vary depending on oxidation condition used (Frankel 1993). Assays measuring substrates or products can also give varying results depending on their specificity. Moreover, such methods are not always representative of the natural evolution of lipids in foods (Frankel 1993), Also, for many antioxidants, the risk of degradation during testing is high (Bondet et al 1997), Osawa and Shibamoto (1992) developed a HPLC method to measure malonaldehyde formed in lipid emulsion systems oxidized by FexCly/H20>, Malonaldehyde was derivatized by reaction with urea under acidic conditions to form 2- hydroxypytimidine which could be measured by HPLC. This method was used by Tsuda et al (1994) to measure malonaldehyde formed in various lipid systems, and by Fukumoto and Mazza (2000) to measure antioxidant and prooxidant activity of phenolic24 compounds. However, in the latter case, sensitivity of HPLC method was poor relative to free radical methods. Beta-carotene bleaching is a quick and simple method to measure antioxidant activity, Antioxidant activity is measured by the ability of a compound to minimize loss of B-carotene during the coupled oxidation of linoleic acid and i-carotene in an emulsified aqueous system. The reaction is usually initiated using heat (50°C) (Fukumoto and Mazza 2000). The method has good sensitivity, but is non-specific and is subject (o interference from oxidizing and reducing agents in crude extracts, and the linoleic acid used is also not representative of typical food lipids (Frankel 1993). ‘Two free radicals that are commonly used to assess antioxidant activity are 2,2 azinobis (3-ethyl-benzothiaziline-6-sulfonic acid) (ABTS) and 2,2-diphenyl-1- picrylhydrazyl (DPPH"), The ABTS assay measures the relative ability of antioxidant to scavenge the ABTS" generated in aqueous phase, compared with a Trolox (vitamin E water-soluble analog) standard. The ABTS" is generated by reacting a strong oxidizing agent (c.g., potassium permanganate or potassium persulfate) with the ABTS salt. The reduction of the blue-green ABTS” radical by hydrogen-donating antioxidant is measured by the suppression of its characteristic long wave absorption spectrum. Antioxidant reduces absorbance of the radical cation (ABTS") to an extent and on a 1e-scale dependent on the antioxidant capacity of the substance being investigated (Miller and Rice-Evans 1997), Results are usually expressed as Trolox equivalent antioxidant capacity (TEAC). The method is rapid and can be used over a wide range ofpH (Amao ct al 1999, Lemanska et al 2001), and in both aqueous and organic solvent systems. It also has good repeatability, and hence is widely reported. The method however, has not been correlated with biological effects, and hence its actual relevance to in vivo antioxidant effects of vi ‘The DPPH is a stable free radical with an absorption band at 515 nm. It loses this absorption when reduced by an antioxidant; (DPPH* + A > DPPHH + A*) or a free radical species; (DPPH’ + R* > DPPH-R). ‘The DPPH’ method is widely used to determine antiradical/antioxidant activity of purified phenolic compounds as well as natural phenol extracts (Brand-Williams et al 1995, Sripriya et al 1996, Bondet et al 1997, Mahinda and Shahidi 2000, Peyrat-Maillard et al 2000, Fukumoto and Mazza 2000). Brand-Williams et al (1995) observed three reaction kinetics for reaction of DPPH" with different phenolic compounds, The first kinetic reaction proceeded quickly, reaching a steady state immediately; the second kinetic reaction was a bit slower, reaching a steady state in 30 min. The third kinetic reaction involved longer periods of time, reaching a steady state in 1-6 hr, which suggests a complex reaction mechanism (Bondet et al 1997). Most of the phenols they tested followed the third kinetics. This suggests that antioxidant activity using DPPH” should be evaluated over time. The method also has good repeatability and is widely reported. However, like ABTS, its relevance to biological systems is unknown.Data on the antioxidant activities of different food samples have been published using the ABTS and DPPH methods. However the extraction solvents and conditions, as ‘well as analytical protocols used vary widely. Data are also expressed in different formats that are hard to correlate. Even when reporting methods are standardized, data can still vary by large magnitudes and are thus hard to compare. For example, Peterson (2001) reported several fold variability in DPPH values for oats when using standard extraction techniques versus the method proposed by General Mills, Inc. (Minneapolis, MN) that involves direct sample suspension in DPPH, as described by Miller et al (2000). The biological relevance of such data thus becomes hard to assess. Part of the problem is «lue to the fact that little attempts have been made to tie the in vitro data based on these methods to biological efficacy. It is essential to develop standardized assays from which values can be given biological relevance, and also to allow for data comparison across authors. The oxygen radical absorbance capacity (ORAC) method developed by Cao et al (1993) measures the ability of antioxidants to protect protein from damage by free radicals. In this assay, different generators are used to produce different radicals. Usually 3 radicals are generated; peroxyl radical (ROO*), hydroxyl radical (OH") and Cu", a transition metal. This is important since measured antioxidant activity of biological samples depends on which free radical or oxidant is used in the assay (Cao et al 1996). The method, however, has recently adopted the ROO” as standard radical since it is the most commot biological systems (Cao and Prior 2001). The target protein (until very recently) was beta-phycoerythrin (f-PE), whose loss of fluorescence was an indi24 of the extent of damage from its reaction with peroxyl radical. However, results with this method had poor repeatability, which was attributed to the protein interacting with sample polyphenols. Ou et al (2001) recently adopted a new synthetic protein (fluorescein) to replace B-PE as a probe. Data on the modified method normally gives values that are 2-3 times higher than with the B-PE. A major advantage of ORAC is that the method is automated and largely standardized, hence values can be easily compared ‘across authors. Even though the ORAC method may mimic antioxidant activity of phenols in biological system better than many other methods, itis not cost effective enough to allow broad use. It is thus essential to compare this method with the more widely used ones to assess the predictive value of the common methods on biological activity.25 CHAPTER Il MATERIALS AND METHODS Samples Raw samples. This work was a continuation of previous work in our laboratory reported ‘by Awika (2000). Some of the samples used in the previous analysis were used. These included brown sorghums, ATx623 x SC103 (SC103), Sumac grow Vega, TX in 1999 (SU99), Sumac from Coffee Seed, Hereford, TX in 1999 (SUC); and black Tx430 sorghums grown in College Station in 1998 (BK98) and 1999 (BK99). In addition new samples grown in College Station, TX included; Sumac grown in 2002 (SU02), a high tannin brown sorghum grown in 2001 (HTO1); black Tx430 sorghums grow in 2000 (BK00), 2001 (BKO1), and 2002 (BK02); a red sorghum ,Tx2911, grown in 2000 (RDOO); and a white sorghum ATX631 x Tx436 (WSO1). All samples were decorticated using a PRL dehuller (Nutama Machine Co., Saskatoon, Canada) to obtain optimum bran fractions as outlined by Awika (2000). Bran yields were 12% for brown, 15% for black and red, and 12% for white sorghums. Hard red winter wheat bran (ADM Milling Co., Chattanooga, TN) was used for comparison. Processed samples, Bread samples made by Gordon (2001) containing 30% black (BK99) and brown (SU99) sorghum brans were used. Cookies made by Mitre-Dieste (2000), containing 50% of BK99 and SU99 brans were also used. In addition HTO1, BKO1, BK02, and WSO1 grains were extruded whole through a friction type Maddox single screw extruder, model MX-3001 (Maddox Metal Works, Inc., Dallas, TX). Screw26 speed was 300 rpm, die diameter 6.125 in, and sample MC was 11.5 ~ 12.5% (non tempered), Phenolic standards. All anthocyanin as well as procyanidin standards were obtained from ChromaDex Inc., Santa Ana, CA. All phenolic acids and other flavonoids were obtained from Sigma-Aldrich Co., St. Louis, MO. Alll phenolic standards were dissolved in HPLC grade methanol prior to analysis Sample extraction All extraction solvents were HPLC grade, Brans and processed samples were ground through a UDY mill (1 mm mesh) prior to extraction. The bread and cookie samples were initially dried in a forced-air convection oven at 50°C for 8 hr before grinding. General extraction. Two extraction solvents were used: 1% HCI in methanol (Me-HCD, and 70% aqueous acetone (Ac-aq). Extraction procedure involved addition of 10 ml solvent to 0.5 g sample in 50 ml centrifuge tubes and shaking the samples for 2 hr at low speed in an Eberbach shaker (Eberbach Corp., MI). Samples were then stored at -20°C in the dark overnight to allow for maximum diffusion of phenolics from the cellular matrix. Samples were then equilibrated to room temperature and centrifuged at 10,000 rpm for 10 min and decanted. Residue was rinsed with 2 additional 10 ml volumes of solvent with shaking for 5 min, centrifuging at 10,000 rpm for 10 min, and decanting in each case. The 3 aliquots were mixed and stored at <-20°C in the dark until analyzed.For HPLC analysis of anthocyanins, C-18 Sep-Pak preparative columns were initially compared with direct sample injection. However, samples processed through the Sep-Pak columns gave significantly smaller anthocyanin peaks on HPLC than directly injected samples, with no significant gain in peak resolutions. Hence all samples were subsequently directly injected (after filtration through 0.45 jum nylon membrane, Millipore Corp., Billerica, MA). Use of preparative Sep-Pak cartridges may be more suitable for samples with significant levels of sugars and phenolic acids (as is common with fruit extracts). Extraction for ORAC assay. Samples were extracted in two stages to obtain lipophi and hydrophilic antioxidant constituents. Hexane was initially used for the lipophilic constituents, followed by Ac-aq extraction for the hydrophilic constituents. Twenty milliliters of each solvent was added to 0.5 g sample and extracted in a Dionex ASE 200 accelerated solvent extraction system (Dionex, Sunnyvale, CA). Extraction time was 15 min in each solvent. Extraction for HPLC procyanidin assay. The method described by Gu et al (2002) was used. In brief, 1 g samples were extracted in 10 ml acetone: water: acetic acid (70: 29.5 0.5). Samples were sonicated at 37°C for 10 min and let stand at room temp for 50 min, then centrifuged at 3500 rpm for 15 min. Acetone was evaporated from extract at 25°C in a SpeedVac (SC201A, Thermo, Marietta, OH) under vacuum, residues dissolved in 6 ml water and applied to Sephadex LH-20 column, The column was washed with 40 ml of 30% methanol in water to remove sugars and other phenols, and the procyanidins recovered with 80 ml of Ac-aq. The eluents were evaporated to dryness under vacuum at28 43°C in a SpeedVac, redissolved in HPLC solvent and filtered (0.45 um) before HPLC analysis. Condensed tannin extraction and puification. The method described by Hagerman and Butler (1980) was used. It involved extraction of sorghum bran in 10 mM ascorbic acid in ethanol to remove soluble proteins, followed by extraction in 10 mM ascorbic acid in methanol for 2 hr, 3 times. The methanol extract was further extracted in 1 mM ethyl acetate buffer, followed by elution of aqueous phase on Sephadex LH-20 with 50% methanol. The eluate was then lyophilized. Analytical procedures Preliminary analysis. All sample extracts (Ac-aq and Me-HCl) were analyzed for phenols using the modified Folin-Ciocalteu method of Kaluza et al (1980). Tannins were assayed using the Vanillin-HCI method of Price et al (1978) with blanks substracted as outlined by Awika (2000). Anthocyanin assay. The pH differential method as reported by Fulecki and Francis (1968) and Wrolstad (1976) was used for quantitative determination with some modifications. Each of two 0.2 ml aliquots was diluted with 2.8 ml of pH 1.0 buffer (125 ml of 0.2 N KCl, and 385 ml of 0.2 N HCl) or pH 4.5 buffer (400 ml of 1 N sodium acetate, 240 ml of 1 N HCI, and 360 mi distilled water) solutions respectively. The absorbance was measured by scanning through a UV/VIS spectrophotometer (Cary 300 Bio, Varian Co., Walnut Creek, CA) from 210 — 750 nm.29 Total anthocyanin pigments were determined from absorbance in pH 1.0 buffer using the above formula. Extinction coefficients for anthocyanin standards were determined using the formula described by Fulecki and Francis (1968). Antioxidant capacity, The three antioxidant methods used are outlined below. The DPPH method of Brand-Williams ct al (1995) was modified for this assay. To determine reaction kinetics for the standards and crude extracts, reactions were initially monitored over a 24 hr period, with readings recorded every 30 min for the first 2 hr, and every 2 hr for the next 10 hr, and every 6 hr thereafter. Most of the samples tested showed residual reactivity even after 24 hr. However, after 8 hr change in activity was very minimal for most of the samples (Appendices A, B and C). Hence 8 hr was used as the standard reaction time. The DPPH" was dissolved in methanol and kept at <-20°C in the dark prior to use. The DPPH was diluted with methanol to obtain an absorbance of approx 1.1 at 515 nm. Aqueous acetone (Ac-aq) extracts, and phenolic standards (150 uil) were reacted with 2850 jl of the DPPH solution for 8 hr with shaking after every 2 br, Samples with high activity were diluted further and rerun to obtain a final absorbance reading of at least 0.2 at 515 nm after 8 hr. Trolox was used as a standard; antioxidant activity was estimated as Trolox equivalents using a standard curve. For ABTS” generation from ABTS salt, 3 mM of K»S20s was reacted with 8 mM ABTS salt in distilled, deionized water for 16 hr at room temperature in the dark. The ABTS" solution was then diluted with absolute ethanol or a pH 7.4 phosphate buffer (50:42.5:9.5; water: 0.2 M NayHPOa: 0.2 M NaH;PO,) solution containing 150 mM30 NaCl (PBS) to obtain an initial absorbance of 1.5 at 730 nm. Fresh ABTS™ solution was prepared for each analysis. Reaction kinetics were determined as for DPPH above Reactions in ethanol neared completion after 8 hr, whereas reactions in PBS were complete in 30 min. Samples and standards (100 ym) were reacted with the ABTS” solution (2900 jm) for 8 hr in ethanol, or 30 min in PBS. Dilutions of samples were performed as necessary to obtain a final absorbance reading of at least 0.1. Since initial analysis showed that crude extracts gave similar antioxidant values in both ethanol and PBS, only PBS was subsequently used for crude extracts. The pure standards, however reacted differently in ethanol and PBS, and hence both systems were compared for standards. Trolox was used as standard. The ORAC assays were performed using an automated Floustar Optima plate reader (BMG Technologies, Offenburg, Germany). Analyses were conducted in pH 7.0 phosphate buffer at 37°C. Peroxyl radical was generated using AAPH, and fluorescein was used as the substrate. HPLC analysis For all analysis except procyanidins, a Waters HPLC system (Millford, MA) was used. The system included a Waters 600 pump, with a Waters 600E control system, Waters 996 PDA detector, and Waters 717 autosampler. Data collection/manipulation was via Waters Millenium software. For procyanidin assay, an Agilent 1100 HPLC system was used, Data collection/manipulation was through HP ChemStation software. Anthocyanins. The method was modified from Wang et al (2000). Separation was on a reversed phase Waters Spherisorb ODS-2 5 11 (250 x 4.6 mm) column, Flow rate was 0.531 ‘ml/min; injection volume, 20 tl; column temperature, 35°C; detection, 210 - 600 nm. Mobile phase was (A) 10% formic acid in water, (B) acetonitrile: water: formic acid (5:4:1). Gradient was; 0-3 min, 12% B isocratic; 3-10 min, 12-30% B; 10-15 min, 30% B isocratic; 15-20 min, 30-40% B; 20-30 min, 40% B isocratic; 30-40 min, 40-100% B; 40-60 min; 100% B isocratic; 60-63 min, 100-12% B; 63-75 min, 12% B isocratic. Procyanidins. The method described by Gu et al (2002) was used. Mobile phase was (A) dichloromethane, (B) methanol, and (C) acetic acid: water (1:1 v/v). Gradient was: 0-30 min, 14,0-28.4% B; 30-45 min, 28.4-39.6% B; 45-50 min, 39.6-86.0% B, 50-55 min, 86.0% B isocratic; 55-60 min, 86.0-14.0% B, followed by 10 min re-equilibration of the column before the next run, A constant 4% C was maintained throughout the gradient. Flow rate was 1 ml/min, Separation was on a normal phase 5 1 Luna silica column (250 x 46 mm) (Phenomenex, Torrance, CA). Fluorescence detection was used; excitation ~ 276 nm, emission — 316 nm. Other flavonoids. A modified method of Schrieber et al (2001) was used. Mobile phase was (A) 2% acetie acid in water, and (B) 0.5% acetic acid in water: acetonitrile (1:1). Flow rate was 0.75 ml/min; separation was on a reversed phase Waters Spherisorb ODS- 25 (250 x 4.6 mm) column, Gradient: 0-5 min, 10% B isocratic; 5-20 min, 50% B; 20- 30 min, 50% B isocratic; 30-37 min, 70% B; 37-42 min, 70% B isocratic; 42-52 min, 100% B; 52-62 min, 100% B isocratic; 62-65 min 10% B; 65-75 min, 10% B isocratic. All analyses were conducted in triplicates, except ORAC and normal phase HPLC analyses that were conducted in duplicates.32 Special note All ORAC and procyanidin HPLC analyses were conducted at the Arkansas Children Nutrition Center, Little Rock, Arkansas. S ncere gratitude goes to Dr. Ronald Prior for generously allowing us access to facilities and staff at the center. Special thanks go to Drs. Gu Li Weu and Wu Xian Li for their time and effort in conducting the analyses.33 CHAPTER IV RESULTS AND DISCUSSION Extraction of sorghum antioxidants able I compares tannin (Vanillin-HCl) and phenol (Folin-Ciocalteu) values for brown sorghums extracted with 70% aqueous acetone (Ac-aq) versus 1% HCl acidified methanol (Me-HCl). Values for tannins varied between the two solvents without a clear pattern across samples. Among the 5 brown sorghum brans, Me-HCl gave higher values for 2 samples, whereas Ac-aq gave one sample a higher value. Two samples were not significantly different in their measured tannins across the solvents. Among the grain samples analyzed, Ac-aq tended to give higher values. When the tannin (catechin equivalents) values of the Me-HCI extracts were compared to catechin values obtained by the standard Vanillin-HCI procedure that involves extraction in Me-HCl for 20 min prior to analysis, they were 3-5 times lower (Table 1). The standard Vanillin-HCI values were comparable to those reporte literature by Gous (1989) and Awika (2000), Various authors have reported anthocyanin-tannin reactions and formation of new compounds during wine aging and in model systems (Haslam 1980, Dallas et al 1996, Francia-Aricha et al 1997, Es-Safi et al 1999, Remy et al 2000). Since the brown sorghums also had significant levels of anthocyanins (Table II), it is possible the prolonged extraction and holding prior to analysis resulted in formation of anthocyanin-Table I 34 Tannin and phenol levels in Me-HCI and Ac-aq extracts of brown sorghums ‘Tannins (mg CE/g) Phenols (mg GAE/g) Solvent Me-HCl ““Acaq Me-HCl — Ac-aq Extraction time 2Mhe 20min” 224hr 22rd Grain sc103 15d 187c¢ = 80d_—S 100d Te Hi Tannin Bile 353¢ 109d 1240 12.9 Sumac (SU99) 137 $0.2¢ «= «18S 9G D'S Mean 140d 1570 Bran SC103 32.6¢€ 126.8¢ 39.6d 40.74 48.70 Hi Tannin 45.90 146.9 ¢ 445d 53.4¢ 56.6c Sumac (SU02) S6.1d 135.6¢ 44.5e 44.9d 62.0c Sumac (SUC) Sd 18Ble 5366 589d 75.06 Sumac (SU99) 524d 171 536d BBA BBS Mean 319d -'1537e ~) -sonypa jueprxonuT amnjosqe UIego 1 noUyIp soudy ‘sajdures you ypLA UOH}SeaL MOIS DVUO Wm uonE|aL09 YET “asm 01 Apeay st rey} [eoIpe 904 a1qrIS -9sn 01 £svo pure aatsuadxauy Haaa “OVUO tim woHE|aL09 Ys sio1pne ‘anyea jueprxonue ssouoe sonyea areduioo 0} prey 99u9Y “paziprepUreys JON, [e101 Jo uoTeUMse sof smoqTe — woRDeaA 18e ‘aut Jo spotsod Buoy 40} 21qeIS JOU JworpeI 9a4j pareIsUaD es Ha Apms 0} pasn aq wea souay Hd 01 a1qr1S eS SLAW Woy eorpes sayy areiouad 01 days eX ‘asn 01 Aseo pur aatsuadxauy on ‘qu 9{3uIs v oyu TWeprxoqUE suas Hd Jo olun pur sarap wonoeas yiog soresToru juowidmbo sso1se ofiey aq ues AuTiqeEeA ere siowne ssouoe —— uosteduioo eyep 207 smorTe — pozrpreptrers srapuny stip Squauudinba oatsuadxo Jo asn saxmnboy “sjeotpes sayy 1easzar AqpeatBojoIg sasy) ono syueureg SOW Spoujout sueplxonue yuasayyIp 9qp Jo wo: duos Aseuuns ¥ ALAIQeL,48 Table V Comparing antioxidant activity of brown sorghums extracted for different times Sample ‘Tannins (mg CE/g)" ‘ABTS (umol TE/g) a 2+24hr" “2+24br” 20 min® “wai EE sc103 75e 18.74 94.84 94.84 Hi Tannin (HTO1) B&.le 35.3d 147.24 130.8 € Sumac (SU02) - 19.4 108.8 d 103.64 Sumac (SUC) 42.0 19044 150.8 Sumac (SU99) 13.7¢ 502d 242.04 173.6 ¢€ Mean — 98e 331d 1566d 1307e Bran [... 2. — sc103 32.66 12684 43044 361.66 Hi Tannin (HTO1) 45.96 14694 682.04 563.2¢ ‘Sumac (SU02) 56.1¢€ 135.6d 485.24 405.2¢€ Sumac (SUC) 2S5e 188.1d 712.84 668.8. Sumac (SU99) 524e INod 844.44 one Mean s19e~—«1S3.7d~—~C SSCS cw% Vi so *GE- catechin equivalents, Vanillin-HCI method. Samples extracted for 2 hr at room temp and equilibrated for 24 hr at -20°C. Extraction using 20 min standard Vanillin-HCI procedure as outlined by Awika 2000. Different letters within each sample denote significant differences at «= 0.05. All values expressed on DM basis.49 Among the brown sorghum brans (Fig. 6), Me-HCI generally gave higher antioxidant activity than Ac-aq extracts. However, the differences in activity were low (average 9%) compared to the black sorghum bran extracts (Fig. 7), where the Me-HCI extracts gave between 19-23% (average 21%) higher activity than the Ac-aq extracts. Since the solvent-mediated interactions of phenolics in the extracts were not expected to drastically alter the overall antioxidant profiles of the phenols, it could be deduced that the Me-HCI was generally more efficient at extracting the sorghum phenols than Ac-aq. This effect was more significant in black (high anthocyanin) sorghums than the brown (high tannin) sorghums. However, though Me-HC1 is more efficient at extracting sorghum antioxidants, it cannot be used when DPPH and ORAC are the antioxidant assay methods since these methods are sensitive to low pH. Both method of extraction and method of antioxidant analysis significantly affect the relative antioxidant activities when comparing different sorghum types. For example in this study; when DPPH values were compared, the black sorghum brans had on average 34% of the activity of the brown sorghum brans; when ABTS values were compared, the black sorghum brans had on average 48% of activity of the brown sorghum brans when extracted with Ac-aq, and 55% of activity when extracted with Me- HCI. Antioxidant properties of pure phenols Assessing the antioxidant properties of individual phenols may give an insight into how the compounds in sorghum contribute to the total antioxidant activity of the crude extracts. Antioxidant activities of various flavanoids (Table V1) and phenolic acidsDABTS-(Me-HCI), mABTS-(Ac-aq)__| 3 8 3 2 = 8 zB a 2 sug99 suc su02 HT C103 Fig. 6. Antioxidant activities of brown sorghum brans extracted in Me-HCl and Ac-aq, Samples were extracted for 2 hr and equilibrated at <-20°C for at least 24 hr before analysis. Differences within each sorghum were significant at a = 0.05. ABTS-(Me-HC) mABTS-(Ac-aq) ABTS activity (umol TE/g) BKOO BKO1 BKo2 Fig. 7. Antioxidant activities of black sorghum brans extracted in Me-HCI and Ac-aq. ‘Samples were extracted in Ac-aq for 2 hr and equilibrated at <-20°C for at least 24 hr before analysis. Differences within each sorghum were significant at a = 0.05.Antioxidant aetivity* of flavonoids Table VI 51 DPPH ABTS- _ ABTS-PBS_ # of OW EtOH groups ae Luteolinidin* 1.54 4.05 481 4 Apigeninidin* 0.45 83 4.20 3 Peonidin 0.95 3.29 3.62 4 Pelarginidin 1.39 2.76 3.33 4 Pelargonidin-3,5-dighucoside 0.92 3.43 2.88 2 Cyanidin 1.64 445 5.35 5 Cyanidin-3-glucoside 2.65 4.04 4.60 4 Cyanidin-3,5-dighucoside 2.05 4.40 6.34 3 Cyanidin-3-rutinoside 2.01 3.41 4.20 4 Procyanidins (+) catechin* 2.32 3.55 474 5 (© epicatechin* 2.78 3.40 5.28 5 4.20 6.86 6.74 7 Epigallocatechin gallate 454 717 7.39 8 Proyanidin B1* 3.75 6.88 8.74 9 Procyanidin B2 475 8.03 8.34 9 Other flavonoids Quercetin (favonol) 3.22 4.62 3 Rutin (flavonol) 2.57 3.03 4 Naringenin (flavanone)* 0.03 2.05 3 Mean et rs a “LSD 024 «O28 = 50] Trolox/mol *Compounds identified in sorghum, Different letters for means denote significant differences among methods at ct = 0.05.52 (Table VID are shown. Individual phenols react differently in different media or with different oxidants, hence both DPPH and ABTS were compared. Since DPPH is non- soluble in aqueous media, its values were only determined in methanol. The values for ABTS were determined in absolute ethanol (ABTS-FtOH) and aqueous phosphate buffer (ABTS-PBS). Most of the phenols tested showed very slow reaction kinetics with DPPH and ABTS-EtOH, approaching steady state in 8 hr. All of the flavanoids reacted slowly, while some of the phenolic acids reacted faster, reaching steady state in 2 br (Appendices A and B). However, in the aqueous ABTS system (ABTS-PBS), all phenols reacted rapidly, reaching a steady state within 30 min. Similar kinetics were observed for the crude sorghum extracts. Brand-Williams et al (1995) reported slow reactions of the phenols they tested with DPPH, which reached steady state in 6 hours or more. In general, the procyanidins had the highest antiradical activity in both systems, whereas the phenolic acids had the lowest. This may be related to the number of OH substituents. Also, the flavonoid molecules had higher antioxidant activity than trolox in both systems. The reactions of the phenols with ABTS-EtOH and DPPH correlated positively (R? 0.78), whereas reactions with ABTS-PBS vs DPPH gave lower correlation R = 0.53) ( 8), Several phenols showed significant reactivity with ABTS-PBS, but no reactivity with DPPH (Fig. 8). Most of these compounds were phenolic acids. Wang et al (1998) reported that certain phenolic compounds that react with ABTS might not33 Table VIL Antioxidant activity* of phenolic acids DPPH ‘ABTS-ETOH ABTS-PBS —# of OH groups Benzoie acids Gallic 2.79 7.10 5.27 3 Protocatechuic 212 5.61 133 2 Gentinsic 175 432 1.06 2 Syringie 1.36 2.92 1.20 1 p-Hydroxybenzoic 0.00 0.08 1.22 1 Vanill 0.06 0.12 1.78 1 Salicylic 0.01 0.01 0.49 1 Cinnamic acids Caffeie 1.49 3.10 1.30 2 Chlorogenie 271 184 2 Ferulic 273 213 1 apinic 391 291 1 p-Coumaric 0.33 2.88 1 o-Coumaric 0.19 2.04 1 Mean 2.346 196¢ ‘TsD ” 02 ol - 5nol Trolox/mol. Different letters for means denote significant differences among methods at o = 0.05. All above phenolic acids have been identified in sorghum.ABTS-ETOH ABTS activity (mol TE/mol) R= 0.5333 0 1 2 3 4 DPPH activity (mol TE/mol) Fig. 8. Comparing antiradical activity of phenolic compounds in DPPH versus ABTS- E1OH and ABTS-PBS systems. Activity in DPPH and ABTS-EtOH was determined after 8 hr, while activity in ABTS-PBS was measured after 30 min, 5455 necessarily react with DPPH. Hence the pute phenols can also be expected to act differently against different oxidants in biological system. Anthocyanins showed poor correlation in reactivity between ABTS-EtOH and DPPH, with R? = 0.37 (Fig 9). The color interference reported earlier between anthocyanins and DPPH could have been largely responsible for this. Procyanidins and phenolic acids showed strong correlations between ABTS-EtOH and DPPH, implying they had similar reactions in alcohol with the two radicals. However, when reactivity in ABTS-PBS and DPPH were compared, only procyanidins showed good correlation, with R? = 0.66 (Fig. 9). The weak correlations of phenolic antioxidant activity in aqueous versus alcoholic ‘media show how the reaction environment affects the antioxidant properties of individual phenols. Similar observations were made by Wang et al (1998), and Arts et al (2001). This observation is physiologically significant, since it demonstrates that individual phenols are not likely to e broad protection against diverse oxidative stresses in vivo, as recently demonstrated by Lotito et al (2000), and Roig et al (2002). ‘The lack of long term beneficial effects of the imbalanced antioxidant supplements previously reported (Kim et al 1993) could be partly attributed to this phenomenon. This also suggests that the antioxidant activity measured for a single compound may not necessarily indicate its actual potency in a complex system. ‘Among the crude extracts however, reactions with ABTS-PBS and DPPH correlated very highly (R? = 0,96) (Fig. 3C). Similarly high correlation (R? = 0.90) was reported for crude extracts from various fruits using similar methods (Leong and Shui 2001). This36 DPPH vs ABTS-EtOH R?=0.9312 R?=0.9269 R*=0.3728 DPPH vs ABTS-PBS ABTS activity (mol TE/mol) R?=0.2445 PAS A PROCS: x ACNS DPPH activity (mol TE/mol) Fig, 9. Correlation coefficients between ABTS in ethanol (ABTS-EtOH) and aqeuous buffer (ABTS-PBS) versus DPPH for different phenol groups. PAS — phenolic acids; PROCS ~ procyanidins; ACNS — anthocyanins.37 demonstrates that the crude extracts with mixtures of antioxidants may offer fairly consistent antioxidant activity in different environments, probably due to a balancing effect or synergism. Such mixtures are likely to offer superior antioxidant protection compared to the isolated compounds, as suggested by Hissig et al (1999). Various studies have attempted to correlate antioxidant activities of phenols to their structure (Rice-Evans et al 1996, Kondo et al 1999, Lien et al 1999), Among the most significant effects are thought to be the number and positions of the OH groups on the phenols. However, data on structure-activity relationships are controversial. For example, Tyrakowska et al (1999) and Lemanska et al (2001) found no correlation between the number of OH groups and antioxidant activity of various structurally related flavonoids. Lien et al (1999), on the other hand, reported a very positive correlation (R™ = 0.845, 19) between the number of OH groups and antioxidant activity of various phenols. Few authors have been able to obtain similar results. Bors et al (2000) reported a clear-cut structure-activity relationship only for the flavan-3-ols (catechins), where the number of OH groups correlated positively with radical scavenging. In this work, no structure-activity relationship was found for the anthocyanins in either DPPH or ABTS systems, whereas the strongest correlations between number of OH groups and radical scavenging were found for the procyanidins in both systems (R? = 0.77 ~ 0.98) (Fig. 10). Phenolic acids showed structure-reactivity pattern only in the alcoholic solvents. The different manners in which the phenols react in different environments should imply that structure-activity relationships might not be obvious.DPPH x ta xAas T peeosse ABTS-ETOH gees 6361 XACNS Antioxidant activity (mol TE/mol) Number of OH substituents Fig. 10. Effect of the number of OH substituents on the antioxidant activity of phenolic compounds in DPPH, and alcoholic (EtOH) and aqueous buffer (PBS) ABTS systems. PAS — phenolic acids; PROCS ~ procyanidins; ACNS ~ anthocyanins.59 HPLC separation of sorghum phenolics Sorghum anthocyanins. The absorption characteristics and other properties of anthocyanin standards analysed are shown in Fig. 11 and Table VIII. The 3- deoxyanthocyanidins, luteolinidin and apigeninidin, had absorption maximas that were particularly different from those of the other anthocyanins. These two anthocyanidins are structurally different from the rest of the anthocyanidins which are commonly found in fruits and vegetables by their lack an oxygen molecule on the C-3. Their absorption maximas of 468 nm (apigeninidin), and 482 nm (luteolinidin) in pH 1 buffer solution made them appear yellow and orange, respectively. This was in contrast with the other anthocyanidins which were all reddish at pH 1. At near neutral pH (in methanol), the 3- deoxyanthocyanidins appeared yellowish orange, and reddish orange for the apigeninidin and luteolinidin, respectively. The rest of the anthocyanins ranged from red to dark blue in color at neutral pH. The lack of oxygen on C-3 of the 3-deoxyanthocyanidins is thought to accord them great stability in solution compared to the other anthocyanidins (Timberlake and Bridle 1980, Sweeny and Iacobucci 1981, Iacobucci and Sweeny 1983). For example, Timberlake and Bridle (1980) reported that apigeninidin was stable in pH 2.8 solution for up to 1 year at room temperature and laboratory light, whereas cyanidin degraded within a few hours under similar conditions. Hence the 3-deoxyanthocyanidins may have an advantage over the other anthocyanidins in food applications. Chromatographic profiles of the anthocyanin standards are shown in Fig. 12. Anthocyanin glycosides eluted faster than their respective aglycons. The presence as60 08 0O7 a6 | 0s | 04 f Absorbance 03 02 04 0 250 © 300 350 400 = 450 500 550-600 —— eyanidin cyanidin-3-rut luteotinidin pelargonidin peonidin ——— apigeninidin Wavelength (nm) Fig. 11. Absorption spectra for selected anthocyanin standards in pH 1.0 buffer solution, The 3-deoxyanthocyanidins bands are marked in bold. Numbers in chart denote Amax (nm). Cyanidin-3-rut; cyaniding-3-rutinoside.61 Table VIII Molar characteristics of anthocyanin molecules ‘Compound MW* — dinax (MOH)” max (PH 1) ¢ (pH 1)? Luteolinidin 3067504 a8 317002100 Apigeninidin 290.0 486 468 30400 + 500 Peonidin 336.5 544 516 27200 + 4100 Pelarginidin 306.7 530 506 28100 + 200 Pelargonidin-3,5- 631.0 514 498 28800 + 1100 diglucoside 322.7 544 516 24800 + 60 484.5 534 512 28600 + 800 glucoside Cyanidin-3,5- 647.0 530 510 33900 4 1700 diglucoside Cyanidin- 631.0 538 514 26100 1700 rutinoside “Molecular weight — includes the chloride ion part of the flavylium salt. ‘Maximum absorption wavelength in methanol and pH 1 buffer, respectively (nm). “Extinction coefficients in pH 1 buffer.62 0.08-|Standards © h 0.06- AU 0.044 0.024 —— 20:00 40100 60.00 Fig. 12. HPLC profiles of Me-HCI anthocyanin extracts from black and red sorghum brans, Detection was at 480 nm; (a) cyanidin-3,5-diglucoside, (b) pelargonidin-3,5- diglucoside, (¢) cyanidin-3-glucoside, (4) cyanidin-3-rutinoside, (¢) cyanidin, (f) luteolinidin, (g) apigeninidin, (h) peonidin. Samples were extracted for 2 hr at room temp, and equilibrated for 24 hr at -20°C prior to analysis. Peaks 2 & 3 were tentatively identified as glycosides of apigeninidin. Peaks 1,4,5,6 & 7 were spectrally similar, and possibly related to luteolinidin.63 ‘well as the number of sugar moieties attached both affected elution time. For example, the diglycoside of cyanidin eluted faster than its monoglycosides, which in tun eluted faster than its aglycon. This is probably due to increased polarity with attachment of the sugar molecules. Similar behavior was reported by Rivera-Lopez et al (1999) and Pazmino-Duran et al (2001). ‘Among the anthocyanin standards used, only the 3-deoxyanthocyanidins (apigeninidin and lutcolinidin) were identified in Me-HCI sorghum extracts (Fig 12 and Table IX). They were most abundant in the black sorghum brans. Other HPLC peaks, including one major, and five minor ones in black sorghums, and three major and three minor ones in the red sorghum could not be identified due to lack of standards. The unknown peaks 2 and 3 (Fig. 12) had spectral characteristics similar to apigeninidin, They were probably glycosides of apigeninidin, given they eluted faster than the apigeninidin aglycon. The other unidentified peaks were all spectrally similar implying they may be structurally related. Their absorption bands were similar to that of luteolinidin, suggesting they are structurally related. The brown sorghums generally had HPLC profiles similar to the black sorghums, but with very small peaks. ‘Aqueous acetone extracts did not produce significant anthocyanin peaks among all sorghums except the black ones. Among the black sorghum Ac-aq extracts, 4 major jistinct peaks were observed, 2 of which had elution times corresponding to Iuteolinidin and apigeninidin, but with absorption maximas at between 370-380 nm, instead of the typical 470-490 nm for the anthocyanins (Figures 13 & 14), The typical absorption peaks for luteolinidin and apigeninidin were still present but at much reduced intensity64 ‘uipruria€ de Jo sopisoaX|8 se poyruap! Ayoanesua, ‘uIpIuyjootny 0} seus sonstrarsereyo [enseds pey “poyluopIur), urpiutueaidy , «up oon], ‘siseq WC 8/3 axe siup “sjuopeAMbe mpratjoamny se passoudxo s[eio) pue suMOW UN [Ty x4 ot PL 60 x0 TL sl 60 SL — AD cote 0108 cle 060. |. | (cc(0) soo ovo = aoen zo con od 96'S £90 ssl 110 soel) e770 $tO0 Lv0 9T1 orl tOMd ses sso sol 908) soo 6z0 970 oro wl rel Toma oll oon.) sto seen oro v0 170 oro Lg0 €s0 oom a9 v0 60'E 0 ee} lo 610 300 ero at oo __8Lt 90. Lr roo aoe 070 610 oro 970 680 sou WoL ot 9 3 + of et al @idy omnT aydures D11aH 4q paurunsoyep suesq wnysr0s par pur y9xIq Jo syuayuoD WJULASOMTY XE65 0.020-| Ac-aq extract 0.015: a 0.010-} AU 0.005- 0.000-| 0.008-] Me-HClextract 0.006 0.002-) \a 0.000-| 20.00 40.00 60.00 Minutes. Fig. 13. HPLC anthocyanin profiles of black sorghum (BKO1) bran extracts monitored at 375 nm, Note the increased intensity of resolved peaks for Ac-aq extracts relative to Me- HCI extracts at this wavelength. Peaks ‘a’ and ‘b’ corresponded to luteolinidi apigeninidin, respectively. Samples were extracted for 2 hr at room temp, and equilibrated for 24 hr at ~20°C prior to analysis.66 Luteolinidin Apigeninidin 489.8 E 0.02¢ mss : < f oot / 0. 300,00" 400.00 * 600.00 ‘om Fig. 14. Spectral characteristics of luteolinidin and apigeninidin peaks isolated from black sorghum (BKO!) bran by HPLC. Spectra ‘A’ were obtained from Ac-aq extracts and ‘B’ from Me-HCl extracts. Note the prominent presence of a third peak, not typical for anthocyanins, and the diminished intensity of the typical absorption maxima of the anthocyanidins in the Ac-aq extracts.67 (Fig.14), Apparently the anthocyanin molecules undergo significant structural ‘modification in aqueous acetone, a phenomenon that did not occur in acidified methanol. This observation agrees with that of Lu and Foo (2001). The authors were able to confirm formation of pyranoanthocyanins from anthocyanins through oxidative addition mediated by acetone. These compounds may be responsible for the observed Ac-aq extract spectra. Due to the above observation, Ac-aq extract anthocyanin levels could not be quantified by HPLC. Apigeninidin and luteolinidin represented an average of 49% of the total peak areas, whereas the major unknown peak (peak number 6) represented about 25% of the total black sorghum anthocyanins. In the red sorghum, luteolinidin was not detected and the apigeninidin peak was minor compared to other unidentified peaks. However, spectral characteristics of the unidentified peaks were similar to those of corresponding, black sorghum peaks (Fig. 12). The 3-deoxyanthocyanins appear to be the major anthocyanin pigments in sorghum. Gous 1989 also identified luteolinidin and apigeninidin as the major anthocyanins in a black sorghum variety. Glycosylated forms of these anthocyanins were also identified in sorghums by Nip and Burns (1969, 1971). ‘Unlike most anthocyanins in fruits and vegetable that are naturally found as glycosides, the 3-deoxyanthocyanidins do exist in nature as aglycones (Stafford 1965, Clifford 2000). Hence their abundance in the extracts may not necessarily mean degradation during extraction. The 3-deoxyanthocyanidins had antioxidant activities that were similar to the fruit anthocyanins (Table VI). Additionally, a strong correlation (R? = 0.84) was observed68 between anthocyanin levels and ABTS activity of black sorghum brans, confirming the major contribution of anthocyanins to black sorghum antioxidant activity. However, for the red sorghum (which had no tannins), the level of antioxidant activity observed was high relative to its anthocyanin content, implying other components also contributed significant activity. Procyanidins, Figure 15 shows procyanidin profiles of two different brown sorghum. varieties as determined by the normal phase HPLC method adapted by Gu et al (2002). The method was able to resolve procyanidins up to decamers based on molecular weight. Additionally the polymers with DP>10 were resolved as a single peak as previously demonstrated by Gu et al (2002). The method was usefil for quantifying procyanidins in sorghums. The two sorghum grains analysed showed not only significant differences in their levels of procyanidins, but also in the distribution of the individual oligomers and polymers. The Hi Tannin grain had a significantly smaller ratio of the lower molecular weight oligomers (DP<10) to the polymers (DP>10) than the Sumac (SU99) grain (Fig 15, Table X). The Hi Tannin sorghum had only 14% oligomers as a proportion of the total procyanidins, compared to the Sumac grain which had 31% oligomers. Since procyanidin chain length is known to affect their effectiveness against different oxidative stresses (Tebib et al 1997, Lotito et al 2000), information on relative proportions of the different procyanidin oligomers as well as polymers is essential in predicting overall effectiveness of sorghum procyanidins as antioxidants in vivo. Based on the raw grain data, the Sumac grain is a better than the Hi Tannin grain as a source of dietary69 Lu P 178 Hi Tannin grain | min Fig. 15. Normal phase HPLC procyanidin profiles of two brown sorghum grains, Sumac (SU99), and Hi Tannin (HT01). Numbers on peaks denote degree of polymerization. P is amixed polymer (DP>10).70 Table X Procyanidin content’ of brown sorghums compared to those of freeze-dried cocoa and blueberry DP” HiTannin Sumac grain Sumac bran ‘Cocoa™ Blueberry* grain 0.0140.00 0.1840.01 0.3340.07——:14.2440.38 —_0.1840.01 2 0.0940.01 0.404001 *1.3340.26 —-8.5740.51 0.46+0.02 3 0.124001 0.5140.01 1.614033. 8.102049 0.38+0.02 4 0.2140.03 —0,6940.01 2.324046 -8.8940.54 —0.5040.01 5 0.2620.04 —0.7440.01 «2.514051 -8.8640.52 0.4740.01 6 0.494007 1.1040.023.6140.71 9.994061 0.6940.05 7 0.3820.06 0,790.01 -2.5620.50 6,380.38 0.484002 8 0.3820.06 0.744001 -2.2940.43 —5.9740.31 0.610.083 9 0.6340.10 1.174002 3.4840.62 7.364058 0.93:40.02 10 0.3140.04 0.554001 1.524026 3.2240.23 ur Pp 17.6743.92 15.0940.34 36.8746.12 16.1740.80 —15.2840.51 “Total -20.5044.3521.9740.45 58.48+10.27 97.7685.32 19.9940.43 % oligo! 14.03 31.31 36.33 83.45 23.51 5mng/¢, obtained by normal phase HPLC method of Gu et Degree of polymerization. “Mixture of polymers with DP>10. “Gu et al (2002). SUI — unidentified. ‘Oligomers (DP<10) as a percent of total.n procyanidins, since it had a more balanced mix of the different chain lengths A good mix of oligomers and polymers is essential for better overall antioxidant protection (Lotito et al 2000). Comparison of procyanidin oligomer and polymer distribution and contents of sorghum grains and bran with those of blueberry and cocoa are shown in Table X. The values for blucberry and cocoa were obtained by Gu et al (2002) using the same procedure reported in the material and methods section. The Sumac grain and bran had higher levels of procyanidins and a better distribution of oligomers relative to polymers than blueberry. This sorghum variety could thus be a superior commercial source of the procyanidins for nutritional or health applications. The Hi Tannin grain had procyanidin content similar to that of blueberry, but blueberry had a better distribution of oligomers to polymers, Cocoa had the highest level of procyanidins and the most balanced distribution of oligomers to polymers. The procyanidins were previously demonstrated to be the primary contributors of antioxidant activity in cocoa (Adamson et al 1999). Using the normal phase HPLC technique, procyanidin oligomers and polymers were also effectively resolved from different processed sorghum products as discussed in a later section. The technique demonstrated that even though there is an overall reduction in sorghum procyanidins due to processing, the overall distribution of oligomers to polymers is improved. ‘The normal phase HPLC technique was very effective in separating sorghum procyanidins based on molecular weight. When coupled with NMR or MS and reversed phase HPLC the polymer structure and individual molecules that constitute the polymers2 can be readily identified (Gu et al 2002). These authors were able to determine epicatechin as the only chain extension unit and catechin as the main (89%) chain terminal unit of the Sumac sorghum procyanidins. Gupta and Haslam (1978) and Gujer et al (1986) previously reported epicatechin as the chain extension and catechin as the chain terminal units of brown sorghums. Other phenolics in sorghum. Among the Ac-aq extracts of the red and black sorghums, several peaks that had characteristics of flavanones and possibly flavonols were observed but could not be identified among the standards available. Naringenin was identified as a major component of the red sorghum (RDO0) bran phenols (0.95 mg/g), whereas it was present in much reduced levels in the brown (0.19 — 0.26 mg/g) and black (average 0.17 mg/g) sorghum brans (Fig. 16). Another significant peak that was spectrally very similar to naringenin, suggesting it was a flavanone (possibly a glycoside of naringenin), was observed in the red sorghum bran (0.99 mg/g, naringenin equivalents). This peak was very minor in all other sorghum brans. Naringenin was reported in sorghum by Gujer et al (1986). The relatively high content of naringenin in the red sorghum bran may have contributed to its high antioxidant activity despite its Jow anthocyanin content. Among the brown sorghum Me-HC1 extracts significant levels of procyanidin B1 (epicatechin-catechin dimer), and minor catechin and epicatechin peaks were identified (Fig. 16). Also a major peak that corresponded to epicatechin gallate was observed among the brown sorghum Me-HCI extracts. This compound has not been reported in sorghums before, and the peak should be studied further to confirm its identity. ItsB °. a). x Standards ° a | gh | 2 002 f b | oor a wd ten 0.00 - —— o. ve ; sus9 ooze o018- 2 h 0104 L ° a lod J emt | 0. No a "20°00 30100" "40:00 50.00 60. T “| Ra00 0.09} 0.02 2 oor] ahah a SUL. ood rh 30100 " "40100" " 50100 ” ” 60.00 Minutes Fig. 16. HPLC profiles of Me-HCI flavonoid extracts from brown (SU99) and red (RD00) sorghum brans monitored at 280 nm. Peaks; (a) procyanidin B1, (b) catechin, (c) procyanidin B2, (d) epicatechin, () epicatechin gallate, (f) rutin, (2) quercetin, (h) naringenin. Peak (1) was, spectrally very similar to that of naringenin suggesting it was a flavanone. Samples were extracted for 2 hr at room temp and equilibrated for 24 hr at -20°C prior to analysis.4 presence, if proven, may not necessarily mean it exists naturally in sorghum, but could have been a product of cleavage and acylation due to the low pH of the solvent and extended storage period of sample extracts prior to analysis. Spencer et al (2000) observed that procyanidin oligomers broke down significantly to epicatechin monomers and dimers at pH 2.0 within 3.5 hr. Further acylation of such epicatechin monomers to gallic acid to form the epicatechin gallate in the extracts is possible. Phenolic acids were not identified in the extracts at significant levels. Very minor peaks of gallic, protocatechuic, p-hydroxybenzoic and p-coumaric acids were observed. Phenolic acids in sorghum brans, though available at significant levels (Hahn 1984), ‘were not readily extractable under conditions used in this study. Since such bound phenolic compounds may still be effective in vivo, the values obtained in this study are likely to underestimate actual antioxidant potential of the sorghums. Environmental and varietal effects on phenol and antioxidant levels of sorghums Effect of variety. Table X1 shows phenol and ABTS antioxidant levels of various sorghum grains and their brans extracted with Me-HCl. Literature values for some common cereals are shown for comparison. The differences in phenol and ABTS values among the sorghums and wheat brans are illustrated in Figures 17 & 18, respectively. Brown sorghums generally had the highest phenol levels and antioxidant activity, whereas wi ite sorghum had the least. The red and black grains had comparable antioxidant values. All the non-white sorghums showed ABTS activities that were significantly higher than those of the other cereals. Among these cereals, onlyTable XI ABTS antioxidant and phenol levels among sorghums and other cereals 15 ‘Sample Activity (umol TE/g) Phenol (mg/g) ~~ Grain Bran Grain Bran ‘Sumac (SU99) 242 844 19.6 63.8 Sumac (SUC) 190 713 ire) 58.9 Sumac (SU02) 109 485 14 449 S103 95 430 10.0 40.7 ‘Hi Tannin 145 682 12.4 53.4 Black 1998 75 190 5.6 18.1 Black 1999 112 400 10.0 320 Black 2000 52 261 5.5 19.1 Black 2001 72 323 64 26.0 Black 2002 8 374 73 35.6 Red (RDO00) 16 316 47 199 White (WS01) 2 2 08 48 ‘Wheat <0.1-2.0" 28°-34 0.4.0.9" 3.0 Barley gare : : - Rye 4s . 5 _ Oats z oS 0.2.0.3 0.2-0.3° Buckwheat 25° - 32" 33° cv 28 3.5 40 C0 "Zielinski and Kozlowska (2000), "Yu et al (2002), “Haber (2002) “Emmons and Peterson (1999), “Quettier-Delue et al (2000), “estimated from Quettier-Delue et al (2000). Sorghum samples extracted in Me-HCl for 2 hr at room temp and equilibrated for 24 hr at -20°C before analysis.16 “09 = AD “stsAeUR a10y0q 3Y 7 SEA] 10 10} D,0Z—> Te parwaquisnba pue 3Y Z 40} [HOW Ul paoenxe o1om sojdureg “weg raya 0} pasedutoo suexg wNY IOS sNOLTEA JOJ Sfasaq [OUD “LT “Bea HM L0SM o0dy ZOMG 1ONG OOMB 66xG S6¥a COLOS LOLH zONS ONS 6éENs |reayay AHA on = w 2 _ oe = | _ _ __ Pei G | al ov = 5 S os og OL17 4S'E = AD ‘SisKyouE ax030q 14 $7 ISEA] 18 4OF D,0Z—> We poreaqutinbo ue JY 7 10y [DH-9IN Ur paroenxa azam sojdureg “wesg way 07 pasedu‘os sueq WNYsIOS snOLEA 10y SonfeA SLAY “SE “BU HM LOSM 00G4 Zo¥d 1oMa CONG 66d EXE COLDS 10H ZONS ONS é6énSs 003 (G/L low) Aynoe ueprxonuy 0088 buckwheat had higher activity than the white sorghum, Barley had similar antioxidant activity to the white sorghum. The phenol levels in the sorghums were similar to values reported for similar sorghum varieties (Hahn 1984, Gous 1989, Awika 2000). Phenol levels correlated strongly with antioxidant values, with an R? of 0.98. Antioxidant values in grains and their brans were significantly correlated (R? = 0,94), hence the measured antioxidant activity in the grain is a good predictor of activity in bran. The brown sorghums had higher antioxidant activity (432 — 844 mol TE/g) than the other brans. The white sorghum and red wheat brans had the lowest antioxidant activity (42 and 34 mol TE/g respectively). The red sorghum bran had antioxidant activity (316 jumol TE/g) that was comparable to those of the black sorghums (184 400, average 308 mol TE/g). In general, the brans had 3.5 — 5 times higher phenol levels and antioxidant activity than grains. The higher antioxidant activity in brown sorghums compared to the others could mainly be attributed to tannins that were present in these sorghums, but were not, detected in other sorghums. Tannins are more powerful antioxidants than the lower molecular weight phenolics (Sripriya et al 1996, Hagerman et al 1998, Yokozawa et al 1998, Hissig et al 1999, Bors et al 2000). Another factor that could contribute to the high antioxidant activity of the brown sorghums is the significant levels of anthocyanins also present in their brans (Table II). The brown sorghums thus have an excellent potential as a source of antioxidants.79 The black and red sorghum brans had high levels of anthocyanins, but no detectable tannins. Their antioxidant activities was largely attributed to the anthocyanins Anthocyanins are reported to contribute significant antioxidant activity to red wines and fruits like blueberries, grapes, strawberries, elderberries, etc. (Heinenonen et al 1998, Saint-Cricg de Gaulejac et al 1999, Prior et al 1998, Clifford 2000, Milbury et al 2002). Pure forms of these compounds had high antioxidant activity as shown in Table VI. The black and red sorghum brans could be a good source of natural food colors with significant antioxidant benefits White sorghum and wheat brans had very low antioxidant activities relative to the other brans. Wheat bran has recently been reported to have high antioxidant activity relative to other cereal brans like oats or rice (Baublis et al 2000, Haber 2002). However, sorghum brans have much superior antioxidant potential (Fig. 18, Table IX), and are potentially more useful in improving antioxidant levels of cereal foods than other brans. The high antioxidant levels in the sorghum brans imply that they can be used in reduced quantities compared to other cereal brans, to achieve equivalent antioxidant activity, This gives room for flexibility in using these brans in formulations. The brans could also be a commercially viable source of phenolic extracts for pharmacological use and dietary supplementation. Effect of environment. The sorgoums with the highest levels of tannins and anthocyanins, Sumac (SU) and Black (Bk), respectively, were compared across different environments/seasons (Table XI, Figures 17 & 18). Since these sorghums are potential80 targets for commercial exploitation, it was necessary to determine environmental fluctuations in their phenolic contents, and antioxidant activities. Environment significantly affected the phenol and antioxidant levels of the grains and their bran fractions (Table XI, Figures 17 & 18). Antioxidant levels in black sorghums ranged from 52-112 pmol TE/g for grain, and (192-400 mol TE/g) for bran over the five-year growing seasons in College Station, TX (1998-2002). The Sumac samples had antioxidant levels ranging between 108-244 jumol TE/g for grain, and 484- 844 umol TE/g for bran. More detailed studies should be conducted to understand the environmental factors that affect the antioxidant levels in sorghums to maximize antioxidant yields. This would be important if the sorghums are to be used as a commercial source of these compounds. Effect of processing on antioxidant activity of sorghum Since the most effective way of delivering the sorghum antioxidants to consumers is through food, it was essential to evaluate processing effects on the overall antioxidant activity of the sorghum phenols. Baking, To directly compare antioxidant activities in brans before and after processing, antioxidant activity after processing was adjusted to equivalent activity of similar amount of unprocessed bran (Fig. 19). Brans in cookies retained significantly more ABTS activity (72-78%) than brans in bread (57-60%). Also, the brown sorghum bran generally retained more ABTS activity in both bread and cookie (60% and 78% respectively) than black bran (57% and 72% respectively).81 1000 raw bran bran in cookie bran in bread ABTS activity (mol TE/g) ‘Sumac (SU99) Black (BK99) Fig. 19. Changes in antioxidant activity of sorghum brans after baking. Samples were extracted in Me-HCI for 2 hr and equilibrated at <-20°C for at least 24 hhr before analysis. Antioxidant values for processed brans were adjusted to equivalent activity of similar amount of unprocessed bran. Differences within each sorghum were significant at = 0.05.82 The cookie formulations involved less moisture and more sugar and fat compared to bread formulations (Mitre-Dieste et al 2000, Gordon 2001). Also mixing and other procedures preceding baking, as well as the baking itself, involve less time for cookies than bread. Hence, the availability of moisture as a solvent for the sorghum bran phenols, and time for phenol interaction with other dough components, are greatly limited in cookie dough compared to bread dough. These factors ensure that the phenolic constituents in the cookie dough brans are retained more in their original forms than in bread doughs. The baking time for the breads were also longer than for cookies; this may have caused more heat damage to the bread bran phenols than the cookie bran phenols. The Sumac bran had condensed tannins, which are known to bind food macromolecules (proteins, fats, carbohydrates), forming insoluble complexes. Such complexes would be harder to extract for analysis, and this may partly account for the reduced activity after processing. Since high molecular weight tannins are the ones mostly involved in these interactions, it was expected that their levels would significantly decrease, relative to the lower molecular weight tannins. This was observed in the HPLC profiles of the processed Sumac bran tannins relative to the unprocessed. brans (Figures 20 & 21, Table XID. The relative peak heights for the lower molecular weight tannins compared to higher molecular weight tannins were significantly higher in the processed brans than in raw brans (Fig. 20). This effect was more pronounced in bread than in the cookies as is clearly visible in Fig. 21.83 Raw bran Lu 140 120 100 Bran in cookie 60 P oH Bran in bread 4 25 150 78 ° 15 30 45 60 Time (min) Fig. 20. Normal phase HPLC procyanidin profiles of Sumac (SU99) bran before and after baking in cookies and bread. Seales for cookie and bread brans are adjusted to 100% bran, Numbers on peaks denote degree of polymerization. ‘P’ represents polymers with DP>10. Samples were extracted in Ac-aq containing 0.5% acetic acid (Gu et al 2002).84 Percent of original recovered after processing Fig. 21. Change in Sumac (SU99) bran procyanidins due to processing in cookies and bread. ‘P” is a mixture of polymers with DP >10. Values were calculated based on peak areas obtained in normal phase HPLC procyanidin profiles. Different letters on bars denote significant differences at a = 0.05Table XI Effect of processing on procyanidin polymer contents” of Sumac (SU99) bran measured by normal phase HPLC DP? Raw bran Bran in cookies Bran in bread 033 ——S~«SCD 024 1.33 0.90 0.91 1.63 0.88 0.97 231 1.20 1.06 251 1.16 0.83 361 1.65 1.04 2.56 1.10 0.54 2.29 0.98 0.43 9 3.48 1.49 0.52 10 1.52 0.66 0.21 rt 36.87 1.77 9.90 Total —s84e—~“*é«iCGHSC“‘C‘*dSSCOCS Loss (%) 0 52.0 15 cv 57 10.2 83 mes. Degree of polymerization. “Mixture of polymers with DP>10.86 In the Sumac bran cookies, the extractable procyanidins, relative to raw bran, decreased from 84% of the monomers to 42% of the decamers. Approximately 46% of the mixed polymers (> 10 DP) were extractable in the Sumac bran cookies relative to raw bran. In the Sumac bran bread, the extractable procyanidins decreased more dramatically from 69% of the monomers to only 13% of the decamers. Twenty six percent of the mixed polymers were extractable in the bran bread relative to raw bran. These data clearly suggest that the degree of interaction of the procyanidins with food macromolecules increase with the degree of polymerization of the procyanidins. The bread allowed more opportunity for these interactions due to its formulation and processing conditions. Even though extractabi of the procyanidin polymers decrease significantly due to processing, this does not imply they are completely non-functional as antioxidants in vivo, Several authors have reported that the procyanidins remain active as antioxidants in the digestive tract even when they are complexed with food molecules (Hagerman et al 1998, Marshall and Roberts 1990). These polymer complexes can be broken down by the colon microflora into phenolic acids that can be absorbed through the large intestine epithelium, and provide antioxidant properties in plasma and tissues (Pietta et al 1997, Deprez et al 2000, Pietta 2000, Tapiero et al 2002). The measured antioxidant potential of the processed high tannin bran products are thus likely to underestimate the actual biological potential of these products, since only the activity of the extractable portions are measured in vitro.87 Extrusion, All the three sorghum types (brown, black and white) that were extruded showed significant antioxidant activity after processing (Fig, 22). As a percentage of the antioxidant activity in the original grains, the extruded samples retained 81% and 70% for the 2001 and 2002 black sorghums respectively, and 89% for the high tannin sorghums (HTO1). Difference in antioxidant activity between the white sorghum grain and extrudate was not significant. ‘The extruded high tannin and black sorghums had surprisingly high retention of antioxidant activity, given extrusion involved high heat, temperature and friction conditions, which were expected to break down the flavonoids to a larger extent, Tannin level in the extruded HTO1 sorghum was only 21% of original grain, whereas anthocyanin content of the black sorghum extrudates was only 49-54% of the original grains (Table XIII). Retention of measurable phenols were 48% for the HTO1 sorghum, and 70-72% for the BK sorghums. The higher retention in antioxidant activity relative to measurable phenolics could imply that during extrusion, the phenols are converted (through cleavage, reassociations, etc) to other compounds that are not detectable as phenols by the methods used. These new compounds retain most of the antioxidant activity of original compounds. ‘The HPLC profile for the high tannin sorghum and its extrudate are shown in Fig. 23. As observed for the baked products, significant increase in peak heights for lower DP tannins relative to the high DP tannins was observed. However the effect was more pronounced here than it was in the baked products. There was a very significant increase160 140 120 | 100 80 60 40 ABTS activity (mol/g) 20 Hi Tannin 88 OGrain meExtrudate Black01 White Fig. 22. Change in ABTS activity of sorghums due to extrusion. Samples were extracted in Me- HCI for 2 hr and equilibrated at <-20°C for at least 24 hr before analys a sample denote significant difference at c= 0.05. erent letters within Table XIII Change in measurable phenolies due to extrusion” Sample _Anthocyanins’ __ Phenols® ‘Tannins® Hi Tannin grain EyS 124 8.1 Hi Tannin extrudate 0.15 6.0 17 Black01 grain oo 64 - Black01 extrudate 12 46 : Black02 grain 2.85 73 - Black02 extrudate 7 SA : LSD oF 5No tempering, single screw, short barrel extrutio GAE/g (Folin- Ping luteolinidin eq/g (pH differential); ‘mg, jocalteu), ‘mg CE/g (Vanillin-HC!), All values DM basis. Samples were extracted in Me-HClI for 2 hr and equilibrated at <-20°C for at least 24 hr.LU 175 150 128 100 7) 2 10 Extrudate Time (min) Fig. 23. Normal phase HPLC procyanidin profiles of Hi Tannin (HO) grain before and after extrusion. Numbers on peaks denote degree of polymerization. “P* represents polymers with DP>10. Samples were extracted in Ac-aq containing 0.5% acetic acid (Gu et al 2002). 8990 in the levels of DP1 ~ DP4 procyanidins, with the extractable monomers representing, 600% and the quadramers 134% of levels in original grain (Fig. 24). Above DPS, however, significant reduction in procyanidin levels were observed, with the polymers (©DP10) in the extrudates recovered at only 15% of the original grain. As observed jon with other sorghum earlier, this trend could be due to tannin polymer complexa constituents during extrusion, However, due to the high levels of lower DP procyanidins recovered relative to grain, cleavage of the higher DP procyanidins into lower DP constituents during extrusion is likely. In general all sorghum samples analyzed retained high antioxidant activity after processing. The antioxidant activities in products depended largely on the antioxidant activities in raw materials. This implies that when these specialty sorghums are used as ingredients during food manufacture, the levels in the formulation can be predictive of the antioxidant activity in the finished product. Dietary implications of the phenol and antioxidant status of the processed sorghums Data is available on DPPH and ABTS antioxidant activities of various grains (buckwheat, barley, wheat, rice, oats, ete) and processed cereal products (breakfast cereals, bread, concentrates, etc) (Onyencho and Hettiarachchy 1992, Watanabe 1998, Emmons and Peterson 1999, Baublis et al 2000, Zielinski and Kozlowska 2000, Lloyd et al 2000, Yu et al 2002). However, these data are hard to compare in most cases since the extraction procedures, as well as methods of analysis and reporting are often different. It is thus hard to reconcile data across authors. For example researchers at General Mills,ol 600 500 400 —— 300 200 Percent recovered after extrusion 100 1 2 3 4 5 6 7 8 9 10 P Degree of polymerization Fig. 24. Change in Hi Tannin (HT) grain procyanidins due to extrusion. ‘P” is represents polymers with DP >10. Values were calculated based on peak areas obtained in normal phase HPLC procyanidin profiles. Different letters on bars indicate significant differences at = 0.05.Tnc., MN, adapted the DPPH method to analyze antioxidant activity of suspensions of food materials by incubating finely ground materials in 50% aqueous methanol at 38°C for 4 hr (Miller et al 2000). Peterson (2001) reported that the suspension method measured several fold more antioxidant activity than was measured in aqueous ethanol extracts for diverse grain materials, ‘The ORAC method, on the other hand, is largely automated and this makes data ‘comparison across authors less ambiguous. Also, with the ORAC method antioxidant values can be given biological relevance, Prior and Cao (2000) estimated that a daily ORAC intake of 3000 ~ 3600 jumol TE/day was necessary to bring about antioxidant benefits in humans. They estimated that the current dietary intake of ORAC from fruits and vegetables in the US was 1200 — 1640 jumol TE/day, less than half of their estimated recommendation. Since it is not practical to expect people to double their consumption of fruits and vegetables, other sources of antioxidants are necessary in the diet. In apparent response to this several dietary antioxidant supplements have flooded the market. However, most of these compounds are unbalanced antioxidants mixtures of a few compounds, the most common being vitamin E, C, and beta-carotene blends (Hassig et al 1999). The efficacy of such unbalanced supplements have been questioned (Rapola et al 1997, Stephens 1997), Prior epidemiological studies (Kim et al 1993), also found no long term benefits from such supplements.93 Hassig et al (1999) suggested that a blend of natural polyphenols mixtures would be more beneficial as a source of antioxidants. Recent data support this, since different antioxidant molecules were found to react very differently against different oxidative stresses (Tebib et al 1997, Lotito et al 2000, Roig et al 2002). For example, Roig et al (2002) tested different flavanoids found in wine against different oxidative stresses on hepatocytes. They found a mixture of procyanidins to be most effective against all the oxidative stresses, whereas the individual flavanoids only gave protection against specific stresses. Lotito et al (2000) found that oligomeric chain lengths affected effectiveness of procyanidins from cocoa against oxidation. They found higher MW procyanidins (DP4 - DP6) were more effective in lipid system, whereas the lower MW procyanidins ( 2.5 g/serving) as per ADA definition (Gordon 2001). The breads and cookies also had dark natural colors that are attractive in healthy baked foods. Currently caramel is used to darken many of these products to give them the ‘healthy’ appeal. ‘Comparison of sorghum products with other dietary sources Blueberries are considered an excellent source of antioxidants (Heinenonen et al 1998, Prior et al 1998). Levels of ORAC for blueberries as well as other common fruits and extracts are compared to those of sorghum brans in Table XV. The sorghum brans had significantly higher values than all the fruits shown. Among the extracts and concentrates, the brown sorghum procyanidin extract had 2-3 times the level of ORAC in the grape extract, red wine concentrate and vitamin C. The high ORAC levels in sorghum brans and extract demonstrates a higher potential of the sorghum brans over fruits as a source of natural antioxidant concentrates/extracts. This has a lot ofTable XV ORAC" levels in sorghum brans and extracts relat 97 to common fir ‘Sample ‘ORAC* (DM basis) “Black (BKON) bran «1,008 Sumac (SU99) bran 3,120 Hi Tannin (HT) bran 2,400 Blueberries 87-873 Strawberries 356-400 Plums 452 - 600 Grapes 100 ‘Watermelon 15 Orange 80-152 Extracts Sumac tannin extract 11,200 Grape skin extract 6,124 Red wine concentrate 3,200 Vitamin C 5,000 Reference Moyer et al (2002) Wu (2000) Wu (2000) Wu (2000) Wu (2000) ‘Wu (2000) Polyphenolics (2000) Polyphenolies (2000) Polyphenolics (2000) Sumol TE/g, using Fluorescein as a probe.98 (coz) 1e 19 wary, “(TOOZ) Te 48 snareWY, “(LL61) [e 1 BUIpZEAHL, “(S86I) Te 10 VOSUEH]-UNUUOIE, “(866 1) [e ¥9 BON! A-IDIeD, “(466 1) WHE Pur 101, (L661) OHHOQUIL, pue a1PU, “(000Z) POU, (L661) [e 19 Bue, IyBTOM YSN, apa uipruex) 60-£0 aBvqqes pay BD upruek) 001-07 Ausoquopld yo upruodiejag 60-70 Auquens faq urpres> 90-10 Ausoqdses pay faq pres) eye Lt Aueqdses speig ip ‘urprunjad “urprucad “uipiare yy sL-€0 sades3 poy oq urpruedo “urprarydjaq or-£0 quesano yaeIg 9 unpruoad ‘wpruex 07-90 Auaquesy oq urprurydjap “uIpraTey os-70 Auraqania ‘urprnuoSide ‘uprurooiny er-sl esq wNYBu0s WAOIg, surprexoouuekxoap-¢ of weaq wINYBI0s (T16ZXL) POA _ uprmside‘mpryoomy I~ SF. tag unyaios yer aouas9jout wipruedoompue soley (3/3) uay00> Aaypouro3 saqquiadaa pue synay wourur09 0} aanepaa suviq wINYss0s Jo Jua}tOD WIULADOMUY TAX 148,99 commercial significance, given the current drive towards natural dietary antioxidant supplements The commercial viability of the black sorghum anthocyanins is emphasized by the fact that they had levels of anthocyanins (4.5 — 11.0 mg/g) that were generally higher than those for common fruit and vegetable commercial sources (Table XVI). The sorghum brans have an additional advantage in that they are more stable for storage (approx 12% moisture) than the fruits and vegetables which normally have more than 80% moisture, and thus require further processing for long term storage. The black sorghum brans could thus be exploited more economically for anthocyanins.100 CHAPTER V SUMMARY Estimation of extraction efficiency of sorghum phenols by measuring actual contents of the extracted phenolics was rendered difficult due to various solvent- phenolic, and phenolic-phenolic interactions, and thus formation of secondary products in the extracts. Assessment of extract antixidant activity was a more reliable way of e ating extraction efficiencies of the solvents. This, however, relied on the assumption that the solvent-phenol interaction, or phenol-phenol interactions in the solvent did not fundamentally alter the antioxidant activities of the extracted phenols. On this basis, Me-HCI w: marginally better than Ac-aq at extracting the sorghum antioxidants from brown high tannin sorghums. However, it was distinctly better at extracting the high anthocyanin black sorghum antioxidants than the Ac-aq solvent, Even though the ORAC method is generally regarded highly due to its use of biologically relevant free radicals and also correlation with actual biological effects, it did not offer any advantage in terms of predicting the overall antioxidant activity of sorghums and sorghum products, when compared with the more common ABTS and DPPH methods. This was important, since the ORAC method is expensive and hence not readily accessible to most laboratories. Less expensive antioxidant methods, with similar or better predictive power on biological effects of the antioxidants are critically necessary. This way, procedures can be standardized and quality control parameters set, for antioxidants from various sources. The ABTS and DPPH methods are reliable predictors of sorghum and sorghum products antioxidant activity. The methods should101 be standardized so that values obtained can be easily compared and given dietary relevance. Meanwhile, ORAC will remain an important method for comparing antioxidant activities across authors. Brown sorghum brans generally had better antioxidant activity than the black or red sorghum brans, whereas the white sorghum and wheat brans had very low activities. The tannins in the brown sorghums are apparently responsible for their high activity. When processed in to various products (cookies, bread and extrudates), the sorghum fractions retained most of their antioxidant activity (58-100%). The antioxidant activity in the products correlated with the activity in the raw samples. Cereal food products made with specialty sorghum fractions could be a very rich source of antioxidants in diets. The ‘measured antioxidant activity of raw sorghum could be predictive of the antioxidant activity in a finished product, and this can aid in ingredient formulation. It is estimated that the average American diet is more than 50% short on daily antioxidant requirements (Prior and Cao 2001). Various antioxidant supplements are available in the market, but they vary widely in composition and antioxidant properties, and their health benefits are unproven, Several authors have reported that different compounds afford antioxidant properties differently depending on the source of oxidative stress and medium involved. Hence individual antioxidant compounds or unbalanced blends, as found in many supplements, ate unable to offer broad antioxidant protection. Food with a natural mix of antioxidants remain the best way of delivering antioxidants to consumers. The brown and black sorghum products analyzed had a good mix of phenolics with high antioxidant activity. For example 56 g serving of 10-15%102 brown sorghum bran bread or a 35 g of 20% brown sorghum bran cookie had enough antioxidants to meet the daily estimated requirement, Black sorghum bran bread (56 serving) had enough antioxidants to meet half of the daily requirement at 15% bran, whereas a 35 g cookie at 30% bran met the daily requirement. The data demonstrate the big potential the sorghums possess as a source of antioxidants in foods. Black sorghum brans were a very good source of anthocyanins (4.5-11.0 mg/g) compared to commercial sources (0.2-10 mg/g). The black sorghum anthocyanins were composed largely of the more stable 3-deoxyanthocyanidins, unlike most of the current ‘commercial sources which have mostly unstable 3-oxyanthocyanidins. This should give the sorghum anthocyanins an edge as a source of natural food color. The antioxidant activity of the sorghum anthocyanins were similar to those of the anthocyanins found in fruits and vegetables, hence they could offer similar health benefits.103 LITERATURE CITED Adamson, G. E., Lazarus, $. A., Mitchell, A. E., Prior, R. L., Cao, G., Jacobs, P. H., Kremers, B. G., Hammerstone, J. F., Rucher, R. B., Ritter, K. and Schmidtz, H. H, 1999. HPLC method for the quantification of procyanidins in cocoa and chocolate samples and correlation to total antioxidant capacity. J. Agric. Food Chem. 47:4184-4188. Anderson, O, M. 1985. Chromatographic separation of anthocyanins in cowberry (Lingonberry) Vaccinium vites-idaea L. J. Food Sci, 50:1230-1232. Ariga, T., and Hamano, M. 1990. Radical scavenging action and its mode in procyanidins B-1 and B-3 from azuki beans to peroxyl radicals. Agric. Biol. 54:2499-2504. Chem. ‘Amao, M. B. 2000. Some methodological problems in the determination of antioxidant activity using chromogen radicals: a practical case. Trends Food Sci. Technol. 11:419-421. Amao, M. B., Cano, A., and Acosta, M. 1999. Methods to measure the antioxidant activity in plant material. A comparative discussion. Free Rad. Res. 31:s89-96. Arts, M. J.T. J., Haenen, G. R. M. M., Voss, H. P., and Bast, A. 2001. Masking of antioxidant capacity by the interaction of flavonoids with protein. Food Chemical Toxicol. 39:787-791, Awika, J. M. 2000. Sorghum phenols as antioxidants. M.S. thesis, Texas A&M University: College Station, TX. Baublis, A., Decker, E. A., and Clydesdale, F. M. 2000. Antioxidant effect of aqueous extracts from wheat based ready-to-eat breakfast cereals. Food Chem. 68:1-6. Bondet, V. Brand-Williams, W., and Berset, C. 1997. Kinetics and mechanism of antioxidant activity using the DPPH" free radical method. Lebensm. Wiss. Technol. 30:609-615. Bors, W., Heller, W., Michel, C., and Saran, M. 1990. Radical chemistry of flavonoid antioxidants. Pages 165-170 in: Antioxidants in Therapy and Preventive Medicine. Emerit, L., ed. Plenum Press: New York, NY. Bors, W., Michel, C., and Stettmaier, K. 2000. Electron paramagnetic resonance studies of radical species of proanthocyanidins and gallate esters. Arch, Biochem. Biophys, 374 (2):347-355.104 Boyd, W. 2000. Natural colors as functional ingredients in healthy foods. Cereal Foods World 45 (5):221-222, Brand-Williams, W., Cuvelier, M. E., and Berset, C. 1995. Use of a free radical method to evaluate antioxidant activity. Lebensm. Wiss. Technol. 28:25-30. Bridle, P., and Timberlake, C. F. 1997. Anthocyanins as natural food colors-sclected. aspects. Food Chem. 58:103-109. Bronnun-Hansen, K., Jacobsen, F., and Flink, J. M. 1985. Anthocyanin colorants from elderberry (Sambucus nigra L.). 1. Process consideration of production of liquid extract. J. Food Technol, 20:703-711. Cao, G., Alesssio, H. M., and Culter, R. G., 1993. Oxygen-radical absorbance capacity assay for antioxidants. Free Radical Biol. Med. 14:303-311 \d Prior, R. L. 2001, Measurement of total antioxidant capacity in nutritional ical studies. Pages 47-55 in: Handbook of Antioxidants. Cadenas, E., and Packer, L., eds. Marcel Dekker, Inc.: New York, NY. Cao, G., Sofic, E. and Prior, R. L. 1996, Antioxidant capacity of tea and common vegetables. J. Agric, Food Chem. 44:3426-3431, Carbonaro, M., Virgili, F., and Camovale, E, 1996. Evidence for protein-tannin interaction in legumes: Implications in the antioxidant properties of faba bean tannins. Lebensm. Wiss. Technol. 29:743-750. Carrefio, J., Almela, J., Martinez, A., and Fernandez-Lopez, J. A. 1997. ‘Chemotaxonomical classification of red table grapes based on anthocyanin profile and external colour. Lebensm, Wiss. Technol. 30:259-265 Chen, H., Zuo, Y., and Deng, Y. 2001. Separation and determination of flavonoids and other phenolic compounds in cranberry juice by high-performance liquid chromatography. J. Chromat. 913:387-395. Chimi, H., Cillard, J., Cillard, P., and Rahamani, M. 1991. Peroxyl and hydroxyl radical scavenging activity of some natural phenolic antioxidants. J. Am. Oil Chem. Soe. 68:307-312. Clifford, E. 1996, Delayed tumor onset in transgenic mice fed on amino acid-based diet supplemented with red wine solids. Am. J. Clinical Nutr. 64:748-756. Clifford, M. N. 2000. Anthocyanins ~ nature, occurrence and dietary burden. J. Sci Food Agric. 80:1063-1072.105 Cooper-Driver, G. A. 2001. Contributions of Jeffrey Harborne and co-workers to the study of anthocyanins. Phytochem. 56:229-236. Dallas, C., Ricardo da Silva, J. M., and Laureano, O., 1996, Products formed in model wine solutions involving anthocyanins, procyanidin B2, and acetaldehyde. J. Agric Food Chem. 44:2402-2407 Deprez, S., Mila, I., Huneau, J-F., Philippe, C., Mila, 1, Lapierre, C., and Scalbert, A. 2000. Polymeric proanthocyanidins are catabolized by human colonic microflora into low-molecular weight phenolic acids. J. Nutr. 130(11):3733-2738. Emmons, C. L., and Peterson, D. M. 1999, Antioxidant activity and phenolic contents of at groats and hulls. Cereal Chem. 76(6):902-906. Escarpa, A., and Gonzalez, M. C. 2001. Approaches to the content of total extractable Phenolic compounds from different food samples by comparison of chromatographic and spectrophotometric methods. Anal. Chim. Acta 427:119-127. Es-Safi, N-E., Fulerand, H., Cheynier, V., and Moutonet, M. 1999. Study on the acetaldehyde induced condensation of (-)-epicatechin and malvidin-3-O-glucoside in a model solution system. J. Agric. Food Chem. 47:2096-2102 Ferrara, L., Montesano, D., and Senatore, A. 2001. The distribution of minerals and flavonoids in the tea plant (Camellia sinensis). Il. Farmaco. 56:397-401 Figueiredo, P., Elhabari, M., Toki, K., Saito, N., Dangles, O., and Brouillard, R. 1996. ‘New aspects of anthocyanin complexation. Intramolecular co-pigmentation as a means for color loss? Phytochem. 41(1):301-308, Francia-Aricha, E. M., Guerra, M. T., Rivas-Gonzola, J. C., and Santos, B. 1997. New anthocyanin pigments formed after condensation with flavanols. J. Agric. Food ‘Chem, 45:2262-2266. Francis, F. J. 1989. Food colorants: anthocyanins, Crit, Rev. Food Sei. Nutr. 28:273-314. Francis, F. J. 2000. Anthocyanins and betalins: composition and applications. Cereal Foods World 45(5):208-213. Frankel, B, N,1993. In search of better methods to evaluate natural antioxidants and oxidative stability in food lipids. Trends in Food Sci. Technol. 4: 220-225. Fukumoto, L. R., and Mazza, G. 2000. Assessing antioxidant and prooxidant activities of phenolic compounds. J. Agric. Food Chem. 48:3597-3604. Fulecki, T., and Francis, F. J. 1968. Quantitative methods for anthocyanins. 1. Extraction and determination of total anthocyanin in cranberries. J. Food Sci. 33:72-77,106 Garcia-Viguera, C., Zafrilla, P., and Tomas-Barberan, F. A. 1997. Determination of authenticity of fruit jams by HPLC analysis of anthocyanins. J. Sci. Food Agric. 73:207-213. Gareia-Viguera, C., Zaftilla, P., and Tomas-Barberan, F. A. 1998. The use of acetone as, an extraction solvent for anthocyanins from strawberry fruit. Phytochem. Anal. 9:247-271. Gordon, L. A. 2001. Utilization of sorghum brans and barley flour in bread. M.S. thesis, ‘Texas A&M University: College Station, TX. Gous, F. 1989. Tannins and phenols in black sorghum. Ph.D, dissertation, Texas A&M University: College Station, TX. Gu, L., Kelm, M., Hammerstone, J. F., Beecher, G., Cunnigham, D., Vannozzi, S., and Prior, L. 2002. Fractionation of polymeric procyanidins from lowbush blueberry and quantification of procyanidins in selected foods with an optimized normal-phase HPLC-MS fluorescent detection method. J. Agric. Food Chem. 50(17): 4852-4860. Gujer, R., Magnolato, D., and Self, R. 1986. Glucosylated flavonoids and other phenolic compounds from sorghum. Phytochem. 25(6):1431-1436. Gupta, R. K., and Haslam, E. 1978, Plant proanthocyanidins. 5. Sorghum polyphenols. J Chem Soe. Perkin Trans. I 4(8):892-896. Haber, B. 2002. Carob fiber benefits and applications. Cereal Foods World 47(8):365- 373 Hagerman, A. E., and Butler, L. G. 1980. Condensed tannin purification and characterization of tannin-associated proteins. J. Agric. Food Chem. 28:947-952. Hagerman, A. E., Riedl, K. M., Jones, G. A., Sovik, K. N., Ritchard, N. T., Hartzfeld, P. W., and Riechel, T. K. 1998, High molecular weight plant polyphenolics (tannins) as biological antioxidants. J. Agric. Food Chem. 46:1887-1892 Hahn, D. H. 1984. Phenols of sorghum and maize: the effect of genotype and alkali processing, Ph.D. dissertation. Texas A&M University: College Station, TX. Hahn, D. H., and Rooney, L. W. 1985. Effects of genotype on tannins and phenols of sorghum. Cereal Chem, 63(1):4-8. Hammerstone, J. F., Lazarus, S. A., Mitchel, A. E., Rucker, R., and Schmitz, H. H. 1999, Identification of procyanidins in cocoa (Theobroma cocoa) and chocolate using high-performance liquid chromatography/mass spectrophotometry. J. Agric. Food Chem. 47:490-496.107 Harborne, J. B. 1988. The anthocyanins. Pages 1-20 in: The Flavonoids. Harborne, J. B., ed. Academic Press: London, UK. Haslam, E, 1980. In vino veritas: oligomeric procyanidins and the ageing of red wines. Phytochem, 19:2577-2582, Hassig, A., Liang, W. X., Schwabl, H., and Stampfli, K. 1999. Flavonoids and tannins plant-based antioxidants with vitamin character. Medical Hypotheses 52(5):479- 481, Heinennonen, I. M., Meyer, A. S., and Frankel, E. N, 1998. Antioxidant activity of berry phenolics on human low-density lipoprotein and liposome oxidation. J. Agric. Food ‘Chem, 46(10):4107-41 12. Hrazdina, G., Iredale, H., and Mattick, L. R. 1977. Anthocyanin composition of Brassica oleracea cv. Red radish. Phytochem. 16:297-299. Husain, S. R., Cillard, J., and Cillard, P. 1987. Hydroxyl radical scavenging activity of flavonoids. Phytochem. 26:2489-2491 Iacobucci, G. A., and Sweeny, J. G. 1983. The chemistry of anthocyanins, anthocyanidins and related flavylium salts. Tetrahedron 39(19):3005-3038. Jimenez-Ramsye, L. M., Rogler, J. C., Housley, T. L., Butler, L. G., and Elkin, R. G. 1994, Absorption and distribution of 14C-labelled condensed tannins and related sorghum phenolics in chickens. J. Agric. Food Chem. 42:963-967. Kallithraka, S., Garcia-Viguera, C., Bridle, P., and Bakker, J. 1995. Survey of solvents for the extraction of grape seed phenolics. Phytochem, Anal. 6:265-267. Kaluza, ‘McGrath, R. M., Roberts, T. C., and Schroder, H. H. 1980. Separation of phenolics of Sorghum bicolor (L.) Moench grain. J. Agric. Food Chem. 28(6):1191- 1196. Kamei, H., Kojima, T., Hasegawa, M., Koide, T., Umeda, T., Yukawa, T., and Terabe, K. 1995. Suppression of tumor cell growth by anthocyanins in-vitro. Cancer Invest. 13(6):590-594, Kanner, J., Frankel, E., Granit, E., German, B., and Kinsella, J. E. 1994, Natural antioxidants in grapes and wine. J. Agric, Food Chem. 42:64-69, Karaivanova, M., Drenska, D., and Ovcharov, R. 1990. A modification of the toxic effects of platinum complexes with anthocyanins. Eksp. Med. Morfol. 29(2):19-24. Kim, L, Williamson, D. F., Bvyers, T., and Koplan, J. P. 1993. Vitamins and mineral supplements use and mortality in a US cohort. Am. J. Public Health 83: 546-550.108 King-Thom, C., Tit, Y. W., Cheng, I. W., Yao-Wen, H. and Yuan, L. 1998, Tannins and human health: a review. Crit. Rev. Food Sci. Nutr. 38(6):421-464. Kinsella, J., Frankel, E., German, B., and Kanner, J. 1993. Possible mechanisms for the protective role of antioxidants in wine and plant foods. Food Technol. 47(4):85-89 Kondo, K., Kurihara, M., Miyata, N., Suzuli, T., and Toyoda, M. 1999. Mechanistic studies of chatechins as antioxidants against radical oxidation. Arch. Biochem. Biophys. 362(1):79-86. Lapidot, T., Harel, $., Akiri, B., Granit, R., and Kanner, J. 1999, pH-dependent forms of red wine anthocyanins as antioxidants, J. Agric. Food Chem. 47:67-70. Lemanska, K., Szymusiak, H., Tyrakowska, B., Zielinski, R., Soffer, A. E.M. F., and Rietjiens, I. M. C. M. 2001. The influence of pH on the antioxidant properties and the mechanism of antioxidant action of hydroxyflavones. Free Radical Biol. Med, 31(7):869-881 Leong, L. P., and Shui, G. 2001. An investigation of antioxidant capacity of fruits in Singapore markets. Food Chem. 76:69-75. Lien, E. J., Ren, S., Bui, H-H., and Wang, R. 1999. Quantitative structure-activity relationship analysis of phenolic antioxidants. Free Radical Biol. Med. 26(3/4):285- 294, Lietti, A., Cristoni, A., and Picci, M. 1976. Studies of Vaccinium myrtillus anthocyanosides. I. Vasoprotective and anti-inflammatory activity, Arzneim-Forsch. 26(5):829-832. Lloyd, B., J., Siebenmorgen, T. J., and Beers, K. W. 2000. Effects of commercial processing on antioxidants in rice bran. Cereal Chem. 77(5):551-555. Lopez, M., Martinez, F., Del valle, C., Orte, C., and Miro, M. 2001. Analysis of phenolic constituents of biological interest in red wines by high-performance liquid chromatography. J. Chromat. 922:359-369. Lotito, S. B., Actis-Goretta, L., Renart, M. L., Caligiuri, M., Rein, D., Schmitz, H. H., Steinberg, F. M., Keen, C. L., and Fraga, C, G. 2000. Influence of oligomer chain length on antioxidant activity of procyanidins. Biochem. Biophys. Res. Comm. 276:945-951. Lu, Y., and Foo, L. Y. 2001. Unusual anthocyanin reactions with acetone leading to proanthocyanin formation. Tetrahedron Lett. 42:1371-1373.109 Mahinda, W., and Shahidi, F. 2000. Scavenging of reactive-oxygen species and DPPH free radicals by extracts of borage and evening primrose meals. Food Chem. 70:17- 26. Malovana, S., Montelongo, F. J. G., Pérez, J. P., and Rodriguez-Delgado, M. A. 2001 Optimization of sample preparation for the determination of trans-resveratrol and other polyphenolic compounds in wines by high performance liquid chromatography. Anal. Chim, Acta 428:245-253. Marshall, T. A., and Roberts, R. J. 1990. In vitro and in vivo assessment of lipid peroxidation of infant nutrient preparations: Effect of nutrition on oxygen toxicity. J. Am. College Nutr. 9:190-199, Martinez-Ortega, M. V., Garcia-Parrilla, and Troncoso, A. M. 2001. Changes in phenolic composition of wines submitted to in vitro dissolution tests. Food Chen T3:11-16. Mataix, B., and Luque de Castro, M. D. 2001. Determination of anthocyanins in wine based on floe-injection, liquid-solid extraction, continuous evaporation and high performance liquid chromatography ~ photometric detection. J. Chromat. A 910:255-263. Mateos, R., Espartero, J. L., Trujillo, M., Rios, J. J., Leon-Camacho, M., Alcudia, F., and Cert, A. 2001. Determination of phenols, flavones, and lignams, in virgin olive oils by solid-phase extraction and high-performance liquid chromatography with diode array ultraviolet detection. J. Agric. Food Chem. 49:2185-2192. Mateus, N., Proencas, S., Ribeiro, P., Machado, J. M., and De Freitas, V. 2001. Grape and wine polyphenolic composition of red Vitis vinifera varieties concerning vineyard altitude. Cien. Tecnol. Aliment. 3(2):102-110. ‘Mazza, G., and Miniati, E. 1993. Anthocyanins in Fruits, Vegetables and Grains. CRC Press Inc.: Boca Raton, FL. Milbury, P. E., Cao, G., Prior, R. L., and Blumberg, J. 2002. Bioavailability of elderberry anthocyanins. Mechanisms of Aging and Development 123:997-1006. Miller, H. E., Rigelhof, F., Marquart, L., Prakash, A., and Kanter, M. 2000. Whole-grain products as antioxidants. Cereal Foods World 45(2):59-63. Miller, N. J., and Rice-Evans, C. A. 1997. Factors influencing the antioxidant activity determined by the ABTS* radical cation assay. Free Rad. Res. 26:195-199. Mistry, T. V., Cai, Y., Lilley, T. H., and Halsam, E. 1991. Polyphenol interactions. Part 5, Anthocyanin co-pigmentation. J. Chem. Soc. Perkin Trans. 2:1287-1296.110 Mitre-Dieste, C. M., Gordon, L. A., Awika, J., Suhendro, E. L, and Rooney, L. W. 2000. Cookies made with sorghum brans high in phenols and catechins (abs). ACC Annual Meeting: Kansas City, MO. Moran, J. F., Klucas, R. V., Grayer, R. J., Abian, J., and Becana, M. 1997. Complexes of iron with phenolic compounds from soybean nodules and other legume tissues: Prooxidant and antioxidant properties. Free Radical Biol. Med. 22:861-870. Moyer, A. R., Humer, K. E., Finn, C. E., Frei, B., and Wrolstad, R. E. 2002. Anthocyanins, phenolics, and antioxidant capacity in diverse small fruits: Vaccinium, Rubus, and Ribers. J. Agric. Food Chem. 50(3):519-525. Mullen W., Lean, W. B. J., and Crozier, A. J. 2002. Rapid characterization of anthocyanins in red raspberry fruit by high-performance liquid chromatography coupled to single-quadrupole mass spectroscopy. J. Chromat. A 996(1-2):65-70. Nip, W. K., and Burns, E. E. 1969. Pigment characterization in grain sorghum. I. Red varieties. Cereal Chem, 46:490-495, Nip, W. K., and Burns, E. E, 1971. Pigment characterization in grain sorghum. II. White varieties. Cereal Chem. 48:74-80. Onyeneho, S. N., and Hettiarachchy, N. S. 1992. Antioxidant activity of durum wheat bran. J. Agric. Food Chem. 40:1496-1500. Osawa, T., and Shibamoto, T. 1992. Analysis of free malonaldehyde formed in lipid peroxidation systems via a pyrimidine derivative. J. Am. Oil Chem. Soc. 69:466- 468. Ou, B., Hampsch-Woodill, M., and Prior, R. L. 2001. Development and validation of an improved oxygen radical absorbance capacity assay using Fluorescein as the fluorescent probe. J. Agric, Food Chem. 49:4916-4926. Pazmino-Duran, E. A., Giusti, M. M., Wrolstad, R. E., Beatriz, M., and Gléria, A. 2001 Anthocyanins from banana bracts (Musa X paradisiaca) as potential food colorants. Food Chem, 73: 327-332. Peterson, D. M. 2001. Oat antioxidants. J. Cereal Chem. 33:115-129. Peyrat-Maillard, M. N., Bonnely, S., and Berset, C. 2000. Determination of the antioxidant activity of phenolic compounds by coulometric detection. Talanta 51:709-716. Pictta, P. G. 2000. Flavonoids as antioxidants. J. Nat. Prod. 63(7):1035-1042,ML Pietta, P. G., Gardana, C., and Mauri, P. L. 1997. Identification of Ginkgo biloba flavonol metabolites after oral administration to humans, J. Chromat. 693:249-255 Polyphenolics. 2000. What's new at Polyphenolics.Available at www.plyphenolics.com. Price, M. L., and Butler, L. G. 1980, Tannins and nutrition. Purdue Univ. Agric. Exp. Stn, Bull, No. 282: Lafayette, IN. Price, M. L., Van Scoyoe, S., and Butler, L. G. 1978. A critical evaluation of vanillin reaction as an assay for tannin in sorghum. J. Agric. Food Chem, 26:1214-1218. Prior, R. L., and Cao, G. 1999. Variability in dietary antioxidant related natural product supplements: the need for methods of standardization, J. Am, Nutraceut. Assoc. 2:36-46. Prior, R. L., and Cao, G. 2000. Analysis of botanicals and dietary supplements for antioxidant capacity: a review. J. AOAC Int, 83(4):950-956. Prior, R. L., Cao, G., Martin, A., Sofie, E., McEwens, J., O’Brien, C., Lischner, N., Ehlenfeldt, M., Kalt, W., Krewer, G., and Mainland, C. M. 1998. Antioxidant capacity as influenced by total phenolic and anthocyanin content, maturity, and variety of Vaccinium species. J. Agric. Food Chem. 46:2686-2693. Putman, L. J., and Butler, L. G. 1989. Separation of high molecular weight sorghum procyanidins by high performance liquid chromatography. J. Agric. Food Chem. 37: 943-946. Rapola, J. M., Virtano, J., and Ripatti, 8. 1997, Randomized trial of a-tocopherol and B- carotene supplements on incidence of major coronary events in men with previous myocardiac infarction. Lancet 349: 1715-1720. Quetticr-Deleu, C., Gressier, B., Vasseur, J., Dine, T., Brunet, C., Luyckx, M., Cazin, M., Cazin, J-C., Bailleul, F., and Trotin, F. 2000. Phenolic compounds and antioxidant activities of buckwheat (Fagopyrum esculentum Moench) hulls and flour. J. Ethnopharmacol. 72:35-42. Remy, S., Fulcrand, H., Labarbe, B., Cheynier, V., and Moutounet, M. 2000, First confirmation in red wine of products resulting from direct anthocyanin-tannin reactions. J. Sci. Food Agric. 80:745-751. Revilla, E., Garcia-Beneytez, E., Cabello, F. Martinez-Ortega, G., and Ryan, J. M. 2001 Value of high-performance liquid chromatographic analysis of anthocyanins in the differentiation of red grape cultivars and red wine made from them. J. Chromat. 915:53-60.Rice-Evans, C. A., Miller, N. J. and Paganga, G. 1996. Structure-antioxidant activity relationships of flavonoids and phenolic acids. Free Radical Bio. Med. 20(7):933- 956. Rigaud, J., Escribano-Bailon, M. T., Prieur, C., Souquet, J. M., and Cheynier, V. 1993. Normal-phase high-performance liquid chromatographic separation of procyanidins from cacao beans and grape seeds. J. Chromat. A 654:255-260. Rivera-Lopez, J., Ordorica-Falomir, C., and Wesche-Ebeling, P. 1999. Changes in anthocyanin concentration in Lychee (Litchi chinensis Sonn.) pericarp during maturation. Food Chem. 65:195-200. Rodriguez-Delgado, M. A., Malovana, S., Pérez, J. P., Borges, T., and Montelongo, F. J. G. 2001. Separation of phenolic compounds by high-performance liquid chromatography with absorbance and fluorimetric detection. J. Chromat. A 912:249-257. Roig, R., Cascon, E., Arola, L., Blade, C., and Salvado, M. J. 2002. Procyanidins protect Fao cells against hydrogen peroxide-induced oxidative stress. Biochim. Biophys. Acta 1572:25-30. Ross, J. A., and Kasum, C. M. 2002. Dietary flavonoids: bioavailability, metabolic effects, and safety. Annu, Rev. Nutr. 22:19-22 Rudiger, C. R. 2003. The formulation of a nutraceutical bread mix using sorghum, barley and flaxseed. M.S. thesis, Texas A&M University: College Station, TX. Saint-Cricq de Gaulejac, N., Glories, Y., and Vivas, N. 1999, Free radical scavenging effect of anthocyanins in red wines. Food Res. Int. 32:327-333. Salunkhe, D. K., Jadhay, §. J., Kadam, S. S., and Chavan, J. K, 1982. Chemical, biochemical and biological significance of polyphenolics in cereals and legumes. CRC Crit. Rey. Food Sei. Nutr. 17:277. Satué-Gracia, M. T., Heinonen, M., and Frankel, E. N. 1997. Anthocyanins as antioxidant on human low density lipoprotein and lecithin-liposome systems. J Agric. Food Chem. 45:3362-3367. Schieber, A., Keller, P., and Carle, R. 2001. Determination of phenolic acids and flavonoids of apple and pear by high-performance liquid chromatography. J Chromat. 910:265-273. Spencer, J., Chaudry, F., Pannala, A., Srai, S., Debnam, E., and Rice-Evans, C. 2000. Decomposition of cocoa procyanidins in the gastric milieu. Biochem. Biophys. Res. Comm, 272(1):236-241.113 Sripriya, G., Chandrasekharan, K., Murty, V. S., and Chandra, T. $. 1996. ESR spectroscopic studies on free radical quenching action of finger millet (Eleusine coracana). Food Chem. 47(4):537-540. Stafford, H. A. 1965. Flavonoids and related phenolic compounds produced in the first intemode of Sorghum vulgare preserved in darkness and in light, Plant Physiol. 40:130-139, Stephens, N.1997. Antioxidant therapy for ischemic heart disease: where do we stand? Lancet 349: 1710-1711 Strack, D., and Wray, V. 1993, The anthocyanins. Pages 1-22 in: The Flavonoids: Advances in Research Since 1986. Harbome, J. B., ed. Chapman & Hall: London, UK. Sweeny, J. G., and Iacobucci, G. A. 1981. Synthesis of anthocyanidins-II: total synthesis of apigeninidin and luteolinidin chlorides. Tetrahedron 37:1481-1483. Tapiero, H., Tew, K. D., Nguyen Ba, G., and Mathe, G. 2002. Polyphenols: do they play a role in the prevention of human pathologies? Biomed Pharmacother. 56:200-207, Tebib, K., Rouanet, J. M., and Besancon, P. 1997. Antioxidant effects of dietary polymeric grape seed tannins in tissue of rats fed a high cholesterol-vitamin E- deficient diet. Food Chem. 59(1):135-141 Timberlake, C. F., and Bridle, P. 1980, Anthocyanins, Pages 115-149 in: Development of Food Colors-1. Walford, J., ed. Applied Science: London, UK. ‘Timberlake, C. F., and Henry, B. S. 1988. Anthocyanins as natural food colorants. Prog. Clin Biol. Res. 280:107-121. Torre, L. C., and Barrit, B. H. 1997. Quantitative evaluation of Rubus fruit anthocyanin pigments. J. Food Sci. 42:488-490, Tsuda, T., Ohshima, K., Kawakishi, J., and Osawa, J. 1994. Antioxidative pigments isolated from seeds of Phaseolus vulgaris L. J. Agric. Food Chem, 42:248-251. ‘Tyrakowska, B., Soffers, A. E. M. F., Szymusiak, H., Boeren, S., Boersma, M. G., Lemanska, K., Vervoort, J., and Rietjems, ]. M. C. M. 1999. TEAC antioxidant activity of 4-hydroxybenzoates. Free Radical Biol. Med. 27(11/12):1427-1436. Velioglu, Y. S., Mazza, G., Cao, L., and Oomah, B. D. 1998. Antioxidant activity and total phenolics in selected fruits, vegetables and grain products. J. Agric. Food Chem. 46(10):4107-4112.14 Wang, H., Cao, G., and Prior, R. L. 1997. Oxygen radical absorbing capacity of anthocyanins. J. Agric. Food Chem. 45:304-309. Wang, J., Kalt, W., and Spomns, P. 2000. Comparison between HPLC and MALDI-TOF MS analysis of anthocyanins in highbush blueberries. J. Agric. Food Chem. 48:3330-3335. ‘Wang, M. F., Shao, Y., Li, J. G., Zhu, N. Q., Randarajan, M., LaVoie, E. J., and Ho, C. T, 1998, Antioxidative phenolic compounds from sage (Salvia officinalis). J. Agric. Food Chem. 46:4869-4973. Wa a, R. D., Poe, J. H., and Bandyopadhyay, R. 1989, Effects of growth conditions on grain molding and phenols in sorghum caryopsis. J. Cereal Sci. 10:217-225. Watanabe, M. 1998. Catechins as antioxidants from buckwheat (Fagopyrum esculentum Moench) groats. J. Agric. Food Chem. 46:839-845, Wrolstad, R. E. 1976. Color and pigment analyses in fruit products. Station Bulletin 624, Agriculture Experiment Station, Oregon State University: Corvallis, OR. Wu, X. L. 2002. Unpublished data. Arkansas Children’s Nutrition Center, USDA: Little Rock, AR (courtesy of Dr. Prior, R. L., and Dr. Wu, X. L.) ‘Yasumutsa, K., Nakayama, T. O. M., and Chichester, C. 0. 1965. Flavonoids of sorghum. J. Food Sci. 30:663-667. Yokozawa, T., Chen, C. P., Dong, E., Tanaka, T., Nonaka, G., and Nishioka, I. 1998. Study of the inhibitory effect of tannins and flavonoids against the 1,1-diphenyl-2- picrylhydrazyl radical. Biochem. Pharmacol. 56:212-222, Yu, L., Haley, S., Perret, J., Harris, M., Wilson, J., and Qian, M. 2002. Free radical scavenging properties of wheat extracts. J. Agric. Food Chem. 50:1619-1624. Zielinski, H., and Kozlowska, H. 2000. Antioxidant activity and total phenolics in selected cereal phenolics and their different morphological fractions. J. Agric. Food Chem. 48:2008-2016.11s APPENDIX A REACTION KINETICS OF ANTHOCYANINS IN ABTS-EtOH AND ABTS-PBS. SYSTEMS | —+—petargoniain 2 peoniain a cyansgle = apigeniniin x eyaniain —+—cyan3.5.igle o— cyan-Sut ——luteainisin Antioxidant activity (mol TE/mol) | ----- polars, Slate ‘Time (hr) of anthocyanins with ABTS in ethanol system (ABTS-E1OH), a 7 paaoaae | cyensate | [a ween Gani Antioxidant activity (mol TE/mol) 4 + nna siglo} sonst 8 uteotincin ~-- peler3.S-digte 2 o 20 40 60 Time (min) Fig. A2. Reaction kinetics of anthocyanins with ABTS in pH 7.4 aqueous buffer system (ABTS- PBS).116 APPENDIX B REACTION KINETICS OF PHENOLIC ACIDS WITH DPPH AND ABTS | + p-courarie a shape a cates protocatoone = sxe chlorogenic | —o— prhycroupbonzoic tere syringe = ++ gentnsic —o— pale | panic Antioxidant activity (mol TE/mol) salicylic | vaniic ‘Time (hr) Fig. B1, Reaction kinetics of phenolic acids with DPPH in methanol. x cenonene | |p pvarcnytonie Antioxidant activity (mol TE/mol) . 0 2 » @ © 6 Time (min) ‘ig, BZ. Reaction kinetics of phenolic acids with ABTS in aqueous buffer (ABTS-PBS).uN7 APPENDIX C REACTION KINETICS OF SELECTED BRANS WITH DPPH AND ABTS + Wheat 2 — White sorg —4—Sumac + $0103 [Black 801g, 160 120 Activity (mg TE’) 10 Time (hr) Fig. C2. Reaction kinetics of selected brans with ABTS in aqueous buffer (ABTS-PBS). Samples were extracted in Me-HCI for 2 hr, room temp, and equilibrated 24 hr at -20°C. 200 || —>—weat | —2—White sorg 4 Sumac xe SC103 | Bleck sory 160 120 80 : z 2 40 0 10 20 30 Time (min) Fig. C2. Reaction kinetics of selected brans with ABTS in aqueous buffer (ABTS-PBS). ‘Samples were extracted in Me-HCI for 2 hr, room temp, and equilibrated 24 hr at -20°C.118 VITA Joseph Mobutu Awika was born in South Nyanza, Kenya, in October 1973. He altended Jonyo Primary School (1979-1986) and Homa Bay School (1987-1990). He Joined Egerton University, Njoro, Kenya, in 1992 and graduated with a B.Sc. in dairy and food technology in 1996. After graduation he worked for the KCC, Kenya, Ltd., as a production supervisor until August 1998 when he initiated his graduate studies. He graduated with an M.S. degree in food science and technology in December 2000, and a Ph.D. in the same program in May 2003 from Texas A&M University, College Station, ‘TX. His permanent address is P. 0. Box 307, Homa Bay, Kenya.