sign in sign up
<<
Home , Eriocitrin, Eriodictyol, Ethyl protocatechuate, Gallocatechin gallate, Genkwanin, Glycitein

Identification of Phenolic Compounds from Peanut Skin using HPLC-MSn

By

Kyle A. Reed

Dissertation submitted to the Faculty of the Virginia Polytechnic Institute and State University in partial fulfillment of the requirements for the degree of

Doctor of Philosophy

In

Food Science and Technology

Committee:

Sean O’Keefe (Committee Chair) Rebecca O’Malley (Committee Co-Chair) Susan Duncan Kumar Mallikarjunan Joseph Marcy

December 7, 2009 Blacksburg, Virginia

Keywords: peanut skin, oligomeric , , HPLC-MSn

Identification of Phenolic Compounds from Peanut Skin using HPLC-MSn

By

Kyle A. Reed

ABSTRACT

Consumers view natural as a safe means to reduce spoilage in foods. In addition, these compounds have been reported to be responsible for human health benefits. Identification of these compounds in peanut skins may enhance consumer interest, improve sales, and increase the value of peanuts. This study evaluated analytical methods which have not been previously incorporated for the analysis of peanut skins. Toyopearl size-exclusion chromatography (SEC) was used for separating phenolic size-classes in raw methanolic extract from skins of Gregory peanuts. This allowed for an enhanced analysis of phenolic content and activity based on compound classes, and provided a viable preparatory separation technique for further identification. Toyopearl SEC of raw methanolic peanut skin extract produced nine fractions based on molecular size. Analysis of total phenolics in these fractions indicated Gregory peanut skins contain high concentrations of phenolic compounds. Further studies revealed the fractions contained compounds which exhibited antioxidant activities that were significantly higher than that of (BHA), a common synthetic antioxidant used in the food industry. This indicates peanut skin extracts are a viable antioxidant source, and that synthetic antioxidants can be replaced with those naturally-derived from peanut by-products. Structures contained in each fraction were identified using high performance liquid chromatography (HPLC) coupled with electrospray ionization (ESI) ion trap mass spectrometry (MSn). Prior to this study, approximately 20 compounds have been identified in peanut skins. The combination of Toyopearl SEC with ESI-HPLC-MSn allowed for the identification of 314 phenolic-based compounds, most of which are newly discovered compounds in peanut skins. Many compounds identified are known to have powerful antioxidant effects, and also have been reported to exhibit numerous beneficial chemical and biological activities, including the treatment of various human health-related conditions. It is evident that peanut skins may be a potential untapped source for the extraction of natural food antioxidants, nutracueticals, and even pharmaceuticals. Because peanut skins are largely a wasted resource to peanut processors, the novel identified in this research could have a significant financial impact on the peanut industry. Dedication

I would like to dedicate this dissertation to my grandfather, the late Sanford (Sandy) H.

Stiles. Known to his family as “Pop Pop”, Sandy was an inspiration to both his community and family alike. He was an instrumental driving force in my life, and taught me the necessary values required to succeed both academically, and personally. For everyone who had the pleasure to know him, he made us realize that nothing in life comes easy, and that only through hard work and dedication will you succeed. He lived by the mantra to “Show ‘em what fer!”, which always gave me the hope of achieving my goals during these often tenacious years of graduate school. Pop Pop, although you are no longer here, you will always be with us, and in my heart forever.

iii Acknowledgements

I would like to thank all the faculty and staff in the Food Technology Department at

Virginia Tech, who gave me this fantastic opportunity to not only hone my skills for my future career, but also to make this research project a reality. My sincere appreciation goes to my major

advisor, Dr. Sean O’Keefe, for his true friendship, guidance, financial support, and vast sea of

never-ending knowledge. A heartfelt thanks goes to my co-advisor, Dr. Rebecca O’Malley, who

graciously took me under her wing for this project. In addition to her friendship, she gave me the

necessary knowledge, tools, and support required to achieve my goals. Another sincere

appreciation goes to my committee members, Dr. Joseph Marcy, Dr. Susan Duncan, and Dr.

Kumar Mallikarjunan. Your friendship, guidance, and support were integral to the success of this research, and also made graduate school at VT an enjoyable and memorable experience. A special thanks also goes to Dr. Jodie Johnson, who devoted much of his time and energy toward

helping me with this project.

Reaching this point has been a long and arduous journey, full of life’s unexpected twists

and turns. Traversing this road has been made simpler by my having people in my life who

inspired me to never give up. I’d like to thank all of them who believed in me; I am grateful to

all my friends for their support and encouragement. Also, a special thanks to Denise Gardner,

not only for her help and support, but also for being such a huge inspiration in both my academic

and my personal life. Most of all, I would like to thank my family for being my rock and my

foundation. To my mother and father, Laurie and Paul Dietrich, and my brother, Kent Reed, a special thank you for their love and encouragement. I couldn’t have done it without you!

iv TABLE OF CONTENTS

Abstract Dedication iii Acknowledgements iv List of Tables and Figures x Introduction 1

CHAPTER I: REVIEW OF THE LITERATURE 3 Peanut Industry 3 Peanut Varieties 4 By-Products of the Peanut Industry 4 Peanut Composition and Health 6 Phenolic Compounds in Peanuts and Other Plants 12 Oxidation and Antioxidants 19 Extraction Methods 26 Antioxidant Activity Methods 29 Total Method 30 Analytical Methods for Separation and Identification 31 Patentability 38 References 42

CHAPTER II: TOTAL PHENOL CONTENT AND ANTIOXIDANT ACTIVITY OF 50 FRACTIONS OBTAINED FROM SIZE-EXCLUSION CHROMATOGRAPHY (SEC) OF RAW METHANOLIC PEANUT SKIN EXTRACT

Abstract 50 Introduction 51 Materials and Methods 52 Results and Discussion 56 Conclusion 60 References 65

CHAPTER III: IDENTIFICATION OF PHENOLIC COMPOUNDS IN PEANUT 66 SKIN EXTRACT USING HPLC WITH NEGATIVE MODE ION TRAP MSn AND ELECTROSPRAY IONIZATION Abstract 66 Introduction 67 Materials and Methods 69 Results and Discussion 74 Fraction A – HPLC Method 1 74 Fraction A – HPLC Method 2 84

v Fraction B – HPLC Method 1 87 Fraction B – HPLC Method 2 97 Fraction C – HPLC Method 1 102 Fraction C – HPLC Method 2 126 Fraction D – HPLC Method 1 146 Fraction D – HPLC Method 2 153 Fraction E – HPLC Method 1 155 Fraction E – HPLC Method 2 159 Fraction F – HPLC Method 1 162 Fraction F – HPLC Method 2 167 Fraction G-Red – HPLC Method 1 168 Fraction G-Red – HPLC Method 2 172 Fraction G – HPLC Method 1 174 Fraction G – HPLC Method 2 178 Fraction H – HPLC Method 1 180 Fraction H – HPLC Method 2 185 Conclusion 186 References 305

Summary 314 Appendix 315 A.1: HPLC-UV Chromatogram of Peanut Skin Raw Extract at 280nm ………… 316

A.2: HPLC-UV Chromatogram of Standard Mix at 280nm ……………………... 317

A.3: Standard Mix: HPLC-UV Chromatogram (bottom) versus TIC …………… 318

A.4: Standard Mix: HPLC-UV Chromatogram (bottom versus TIC’s ………….. 319

A.5: Ion Spectra for Epicatechin …………………………………………………. 320

A.6: Fragmentation Scheme for Epicatechin …………………………………….. 321

A.7: HPLC-UV Chromatogram of Peanut Skin Fraction A (Obtained from Toyopearl Size Exclusion Chromatography at 280nm …………………………... 323

A.8: Peanut Skin Fraction A: HPLC-UV Chromatogram (bottom) versus TIC’s... 324

A.9: Peanut Skin Fraction A: HPLC-UV Chromatogram (bottom) versus TIC …. 325

A.10: HPLC-UV Chromatogram of Peanut Skin Fraction B (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm ………………………….. 326

A.11: Peanut Skin Fraction B: HPLC-UV Chromatogram (bottom) versus TIC’s 327

vi A.12: Peanut Skin Fraction B: HPLC-UV Chromatogram (bottom) versus TIC ... 328

A.13: Peanut Skin Fraction B: HPLC-UV Chromatogram (bottom) versus TIC ... 329

A.14: HPLC-UV Chromatogram of Peanut Skin Fraction C (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm ………………………….. 330

A.15: Peanut Skin Fraction C: HPLC-UV Chromatogram (bottom) versus TIC’s 331

A.16: Peanut Skin Fraction C: HPLC-UV Chromatogram (bottom) versus TIC ... 332

A.17: Peanut Skin Fraction C: HPLC-UV Chromatogram (bottom) versus TIC ... 333

A.18: HPLC-UV Chromatogram of Peanut Skin Fraction D (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm …………………………. 334

A.19: Peanut Skin Fraction D: HPLC-UV Chromatogram (bottom) versus TIC’s 335

A.20: Peanut Skin Fraction D: HPLC-UV Chromatogram (bottom) versus TIC ... 336

A.21: Peanut Skin Fraction D: HPLC-UV Chromatogram (bottom) versus TIC ... 337

A.22: HPLC-UV Chromatogram of Peanut Skin Fraction E (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm ………………………….. 338

A.23: Peanut Skin Fraction E: HPLC-UV Chromatogram (bottom) versus TIC’s 339

A.24: Peanut Skin Fraction E: HPLC-UV Chromatogram (bottom) versus TIC ... 340

A.25: HPLC-UV Chromatogram of Peanut Skin Fraction F (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm ………………………….. 341

A.26: Peanut Skin Fraction F: HPLC-UV Chromatogram (bottom) versus TIC’s 342

A.27: Peanut Skin Fraction F: HPLC-UV Chromatogram (bottom) versus TIC ... 343

A.28: HPLC-UV Chromatogram of Peanut Skin Fraction G-Red (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm ………………………….. 344

A.29: Peanut Skin Fraction G-Red: HPLC-UV Chromatogram (bottom) versus TIC’s ……………….…………………………………………………………….. 345

A.30: Peanut Skin Fraction G-Red: HPLC-UV Chromatogram (bottom) versus TIC ……………………………………………………………………………….. 346

vii A.31: HPLC-UV Chromatogram of Peanut Skin Fraction G (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm ………………………….. 347

A.32: Peanut Skin Fraction G: HPLC-UV Chromatogram (bottom) versus TIC’s 348

A.33: Peanut Skin Fraction G: HPLC-UV Chromatogram (bottom) versus TIC ... 349

A.34: Peanut Skin Fraction G: HPLC-UV Chromatogram (bottom) versus TIC ... 350

A.35: HPLC-UV Chromatogram of Peanut Skin Fraction H (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm ………………………….. 351

A.36: Peanut Skin Fraction H: HPLC-UV Chromatogram (bottom) versus TIC’s 352

A.37: Peanut Skin Fraction H: HPLC-UV Chromatogram (bottom) versus TIC ... 353

A.38: Peanut Skin Fraction H: HPLC-UV Chromatogram (bottom) versus TIC ... 354

A.39: Standard Mix (HPLC Method 2): HPLC-UV Chromatogram (bottom) versus TIC’s ……………………………………………………………………… 355

A.40: HPLC-UV Chromatogram of Peanut Skin Fraction A (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm using HPLC Method 2 …. 356

A.41: Peanut Skin Fraction A (HPLC Method 2): HPLC-UV Chromatogram (bottom) versus TIC’s ……………………………………………………………. 357

A.42: HPLC-UV Chromatogram of Peanut Skin Fraction B (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm using HPLC Method 2 …. 358

A.43: Peanut Skin Fraction B (HPLC Method 2): HPLC-UV Chromatogram (bottom) versus TIC’s ……………………………………………………………. 359

A.44: HPLC-UV Chromatogram of Peanut Skin Fraction C (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm using HPLC Method 2 …. 360

A.45: Peanut Skin Fraction C (HPLC Method 2): HPLC-UV Chromatogram (bottom) versus TIC’s ……………………………………………………………. 361

A.46: HPLC-UV Chromatogram of Peanut Skin Fraction D (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm using HPLC Method 2 …. 362

A.47: Peanut Skin Fraction D (HPLC Method 2): HPLC-UV Chromatogram (bottom) versus TIC’s ……………………………………………………………. 363

viii A.48: HPLC-UV Chromatogram of Peanut Skin Fraction E (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm using HPLC Method 2 …. 364

A.49: Peanut Skin Fraction E (HPLC Method 2): HPLC-UV Chromatogram (bottom) versus TIC’s ……………………………………………………………. 365

A.50: HPLC-UV Chromatogram of Peanut Skin Fraction F (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm using HPLC Method 2 …. 366

A.51: Peanut Skin Fraction F (HPLC Method 2): HPLC-UV Chromatogram (bottom) versus TIC’s ……………………………………………………………. 367

A.52: HPLC-UV Chromatogram of Peanut Skin Fraction G-Red (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm using HPLC Method 2 …. 368

A.53: Peanut Skin Fraction G-Red (HPLC Method 2): HPLC-UV Chromatogram (bottom) versus TIC’s …………………………………………… 369

A.54: HPLC-UV Chromatogram of Peanut Skin Fraction G (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm using HPLC Method 2 …. 370

A.55: Peanut Skin Fraction G (HPLC Method 2): HPLC-UV Chromatogram (bottom) versus TIC’s ……………………………………………………………. 371

A.56: HPLC-UV Chromatogram of Peanut Skin Fraction H (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm using HPLC Method 2 …. 372

A.57: Peanut Skin Fraction H (HPLC Method 2): HPLC-UV Chromatogram (bottom) versus TIC’s ……………………………………………………………. 373

ix LIST OF TABLES AND FIGURES

CHAPTER I Figure 1: Shikimic Acid Pathway ……………………………………………………….. 15

Figure 2: Classes …………………………………………………………….. 17

Figure 3: Retro Diels-Alder (RDA) Mechanism………………………………………… 40

Figure 4: Retro Diels-Alder (RDA) Mechanism ………………………………………... 41

CHAPTER II Figure 1: Experimental Design: Illustrating the basic experimental design for this study 53

Figure 2: Size Exclusion Chromatography (SEC) of Raw Peanut Skin Extract Using Toyopearl HW-40S……………………………………………………………………….. 62

Figure 3: Total Phenolics using Folin-Ciocalteu Method………………………………... 63

Figure 4: DPPH Scavenging Capacity of Peanut Skin Fractions Obtained from Toyopearl SEC……………………………………………………………………………. 64

CHAPTER III Table 1: Detected Compounds in Fraction A using HPLC Method 1…………………… 195

Table 2: Detected Compounds in Fraction A using HPLC Method 2…………………… 199

Table 3: Detected Compounds in Fraction B using HPLC Method 1……………………. 200

Table 4: Detected Compounds in Fraction B using HPLC Method 2……………………. 202

Table 5: Detected Compounds in Fraction C using HPLC Method 1……………………. 204

Table 6: Detected Compounds in Fraction C using HPLC Method 2……………………. 207

Table 7: Detected Compounds in Fraction D using HPLC Method 1…………………… 209

Table 8: Detected Compounds in Fraction D using HPLC Method 2…………………… 212

Table 9: Detected Compounds in Fraction E using HPLC Method 1……………………. 213

Table 10: Detected Compounds in Fraction E using HPLC Method 2………………….. 214

x Table 11: Detected Compounds in Fraction F using HPLC Method 1…………………... 215

Table 12: Detected Compounds in Fraction F using HPLC Method 2…………………... 217

Table 13: Detected Compounds in Fraction G-Red using HPLC Method 1……………... 218

Table 14: Detected Compounds in Fraction G-Red using HPLC Method 2……………... 219

Table 15: Detected Compounds in Fraction G using HPLC Method 1………………...... 220

Table 16: Detected Compounds in Fraction G using HPLC Method 2………………...... 222

Table 17: Detected Compounds in Fraction H using HPLC Method 1………………...... 223

Table 18: Detected Compounds in Fraction H using HPLC Method 2………………...... 224

Figure 1: Fragmentation Scheme for 3,4-Dihydroxycinnamic Acid (a.k.a. Caffeic Acid) 225

Figure 2: Fragmentation Scheme for 3-Hydroxyflavone (a.k.a. Flavonol)……………… 225

Figure 3: Fragmentation Scheme for 4’-Methoxychalcone……………………………… 225

Figure 4: Fragmentation Scheme for 3’,4’,5,5’,7-Pentahydroxyflavone (a.k.a. ).. 226

Figure 5: Fragmentation Scheme for 4,4’-Dihydroxy-3,3’-(2-methoxyethylidene) dicoumarin (a.k.a. Dicoumoxyl) …………………………………………………………. 226

Figure 6: Fragmentation Scheme for 3,3’-Methylenebis[4-hydroxycoumarin] (a.k.a. Dicoumarin)………………………………………………………………………………. 228

Figure 7: Fragmentation Scheme for 3-Methyoxyflavone ………………………………. 228

Figure 8: Fragmentation Scheme for 6-Hydroxy-2-napthalenepropanoic Acid (a.k.a. Allenolic Acid) …………………………………………………………………………… 229

Figure 9: Fragmentation Scheme for ………………………………. 230

Figure 10: Fragmentation Scheme for 3’,4’,7-Trihydroxyflavone ……………………… 230

Figure 11: Fragmentation Scheme for Shikimic Acid …………………………………... 231

Figure 12: Fragmentation Scheme for Erodictyol 7-O-rutinoside (a.k.a. ) …… 231

Figure 13: Fragmentation Scheme for Oxyresveratrol ………………………………….. 232

Figure 14: Fragmentation Scheme for Cedrecoumarin A ……………………………….. 233

xi

Figure 15: Fragmentation Scheme for 2,4,6- ………………….. 233

Figure 16: Fragmentation Scheme for -3,4’,7-trimethyl Ether ……………... 234

Figure 17: Fragmentation Scheme for Biochanin-A-7- (a.k.a. Sissotrin) …….. 235

Figure 18: Fragmentation Scheme for 3,4-Dihydroxycinnamic Acid (a.k.a. Caffeic Acid) ……………………………………………………………………………………… 236

Figure 19: Fragmentation Scheme for 7-Hydroxycoumarin (a.k.a. Umbelliferone) ……. 236

Figure 20: Fragmentation Scheme for 4’,7-Dihydroxyisoflavan (a.k.a. ) ………… 237

Figure 21: Fragmentation Scheme for 8-Hydroxy-4’-5’,6-7-furocoumarin (a.k.a. Xanthotoxol) ……………………………………………………………………………... 237

Figure 22: Fragmentation Scheme for 2,6- (a.k.a. g-Resorcylic Acid) ……………………………………………………………………………………… 238

Figure 23: Fragmentation Scheme for 2-Hydroxybenzoic Acid (a.k.a. )…. 238

Figure 24: Fragmentation Scheme for Pyrocatechin Monoethyl Ether………………….. 238

Figure 25: Fragmentation Scheme for 4’,7-Dihydroxyisoflavone (a.k.a. )…….. 239

Figure 26: Fragmentation Scheme for -7-glucoside…………………………. 240

Figure 27: Fragmentation Scheme for 2,3-Dihydrofisetin (a.k.a. Fustin)………………... 240

Figure 28: Fragmentation Scheme for -3-O- (a.k.a. )…………… 241

Figure 29: Fragmentation Scheme for 6-O-Acetyldaidzin………………………………. 242

Figure 30: Fragmentation Scheme for 3-Methoxy-4- (a.k.a. Ferulic Acid)……………………………………………………………………………… 242

Figure 31: Fragmentation Scheme for Quercetin-7-D-glucoside………………………... 243

Figure 32: Fragmentation Scheme for 3-Methylquercetin (a.k.a. )………... 244

Figure 33: Fragmentation Scheme for 3’,5,7-Trihydroxy-4’-methoxyflavone (a.k.a. Hesperitin)………………………………………………………………………………… 245

Figure 34: Fragmentation Scheme for 5,7,4’-Trihydroxy-3’-methoxyflavone (a.k.a. )……………………………………………………………………………….. 245

xii

Figure 35: Fragmentation Scheme for 4’,5,7-Trihydroxy-6-methoxyisoflavone (a.k.a. )……………………………………………………………………………… 246

Figure 36: Fragmentation Scheme for Orotinichalcone………………………………….. 246

Figure 37: Fragmentation Scheme for 2’,3,4,4’-Tetrahydroxychalcone (a.k.a. Butein)…. 249

Figure 38: Fragmentation Scheme for 3’,5,7-Trihydroxy-4’,5’,6-trimethoxyisoflavone (a.k.a. )……………………………………………………………………………. 249

Figure 39: Fragmentation Scheme for 3,3’,4’,5,5’,7-Hexahydroxyflavone (a.k.a. )…………………………………………………………………………………. 251

Figure 40: Fragmentation Scheme for -8-glucoside (a.k.a. Gossypin)………. 252

Figure 41: Fragmentation Scheme for Gardenin E………………………………………. 253

Figure 42: Fragmentation Scheme for 2’,3,4’,5,7-Pentahydroxyflavone (a.k.a. )… 254

Figure 43: Fragmentation Scheme for 3,3’,4’,5’,7-Pentahydroxyflavone (a.k.a. Robinetin)…………………………………………………………………………………. 254

Figure 44: Fragmentation Scheme for 3,4-Dihydroxycinnamic Acid (a.k.a. Caffeic Acid)………………………………………………………………………………………. 254

Figure 45: Fragmentation Scheme for Torosaflavone A………………………………… 255

Figure 46: Fragmentation Scheme for Glycycoumarin………………………………….. 256

Figure 47: Fragmentation Scheme for Pongagallone A………………………………….. 257

Figure 48: Fragmentation Scheme for 3-hydroxy-6-methoxyflavone …………………... 258

Figure 49: Fragmentation Scheme for -7-O-rhamnoside ………………………. 258

Figure 50: Fragmentation Scheme for 4’-Methoxy-3,3’,5-stilbenetriol 3-Glucoside …… 258

Figure 51: Fragmentation Scheme for 3’-Benzyloxy-5,7-dihydroxy-3,4’- dimethoxyflavone ………………………………………………………………………… 259

Figure 52: Fragmentation Scheme for Isopomiferin …………………………………….. 259

Figure 53: Fragmentation Scheme for 3’,5,7-Trihydroxy-4’,5’,6-trimethoxyisoflavone (a.k.a. Irigenin) …………………………………………………………………………… 260

xiii Figure 54: Fragmentation Scheme for 3,3’,4’,5,6,7-Hexahydroxyflavone (a.k.a. ) ...... 260

Figure 55: Fragmentation Scheme for Oureatacatechin ………………………………… 261

Figure 56: Fragmentation Scheme for 7-methylquercetin (a.k.a. ) ………….. 261

Figure 57: Fragmentation Scheme for -3-b-mono-D-glucoside …………….. 261

Figure 58: Fragmentation Scheme for 4-Methoxycinnamic Acid ………………………. 262

Figure 59: Fragmentation Scheme for Chroman-2-carboxylic Acid ……………………. 262

Figure 60: Fragmentation Scheme for Quercetin-3-arabinoglucoside (a.k.a. Peltatoside). 262

Figure 61: Fragmentation Scheme for Kaempferol-7-rutinoside ………………………... 263

Figure 62: Fragmentation Scheme for Isorhamnetin-3-O-rutinoside …………………… 263

Figure 63: Fragmentation Scheme for Remangiflavanone ……………………………… 264

Figure 64: Fragmentation Scheme for Sophora- ……………………………… 264

Figure 65: Fragmentation Scheme for 5,6-Dihydroxy-7-methoxyflavone (a.k.a. Negletein) ………………………………………………………………………………… 265

Figure 66: Fragmentation Scheme for 2’-Hydroxy-4’,6’-dimethoxychalcone ………….. 265

Figure 67: Fragmentation Scheme for 2’,4-Dihydroxy-4’,6’-dimethoxychalcone ……… 265

Figure 68: Fragmentation Scheme for Isorhamnetin-3-methoxy ……………………….. 266

Figure 69: Fragmentation Scheme for Irisflavone A ……………………………………. 266

Figure 70: Fragmentation Scheme for Kaempferol-3,7,4’-Trimethylether ……………... 267

Figure 71: Fragmentation Scheme for 3’,6’-Dihydroxy-2’,4’,5’-trimethoxychalcone ….. 267

Figure 72: Fragmentation Scheme for Fraxetin-8-glucoside (a.k.a. Fraxin) ……………. 268

Figure 73: Fragmentation Scheme for Lico-iso-flavone A ……………………………… 269

Figure 74: Fragmentation Scheme for 2’,3’-Dimethoxyflavone ………………………... 270

Figure 75: Fragmentation Scheme for 3,3’,4’,5,5’,7-Hexahydroxyflavanone (a.k.a. Dihydromyricetin) ………………………………………………………………………... 271

xiv Figure 76: Fragmentation Scheme for Irisflavone D ……………………………………. 271

Figure 77: Fragmentation Scheme for o-Hydrocoumaric Acid …………………………. 271

Figure 78: Fragmentation Scheme for Betuloside ………………………………………. 272

Figure 79: Fragmentation Scheme for 3’-O-Methylcatechin ……………………………. 272

Figure 80: Fragmentation Scheme for Dihydroquercetin (a.k.a. Taxifolin) …………….. 272

Figure 81: Fragmentation Scheme for 3-Methoxy-3’,4’,5,7-flavantetrol (a.k.a. ) ………………………………………………………………………………. 273

Figure 82: Fragmentation Scheme for Dihydrokaempferol ……………………………... 273

Figure 83: Fragmentation Scheme for -4’,4’’,7-trimethyl Ether (a.k.a. Sciatopitysin) …………………………………………………………………………….. 274

Figure 84: Fragmentation Scheme for 3,3’,4’,5,7,8-Hexahydroxyflavone (a.k.a. Gossypetin) ………………………………………………………………………………. 274

Figure 85: Fragmentation Scheme for Gossypetin 3,3’,4’,7-tetramethyl Ether …………. 274

Figure 86: Fragmentation Scheme for Cudraflavanone B ………………………………. 275

Figure 87: Fragmentation Scheme for 7,8-Dihydroxycoumarin-7-b-D-Glucoside ……... 276

Figure 88: Fragmentation Scheme for Isorhamnetin-3-glucoside ………………………. 277

Figure 89: Fragmentation Scheme for Quercetin-4’-O-glucoside (a.k.a. ) …. 277

Figure 90: Fragmentation Scheme for 2-(3,4-dimethoxyphenyl)-3,4-dihydro-2H- chromene-3,4,5,6,7-pentol ……………………………………………………………….. 279

Figure 91: Fragmentation Scheme for 2-(3,4-dimethoxyphenyl)-7-methoxy-3,4- dihydro-2H-chromene-3,4,5,6-tetrol ……………………………………………………... 279

Figure 92: Fragmentation Scheme for A2 Dimer ……………………. 280

Figure 93: Fragmentation Scheme for 3,5-dihydroxy-2-(4-hydroxy-3-methoxyphenyl)- 7-methoxy-2,3-dihydro-4H-chromen-4-one ……………………………………………... 281

Figure 94: Fragmentation Scheme for Proanthocyanidin A5’ Dimer …………………… 282

Figure 95: Fragmentation Scheme for Procyanidin B2 Dimer ...... 283

Figure 96: Fragmentation Scheme for Procyanidin B-type Dimer ...... 284

xv Figure 97: Fragmentation Scheme for ...... 284

Figure 98: Fragmentation Scheme for Kolaflavanone ...... 285

Figure 99: Fragmentation Scheme for 3,4’,5,7-Tetrahydroxyflavone-3-glucoside (a.k.a. ) ...... 286

Figure 100: Fragmentation Scheme for 6-Methoxyluteolin (a.k.a. ) ……………. 287

Figure 101: Fragmentation Scheme for Procyanidin B-type Dimer (Linear) …………… 288

Figure 102: Fragmentation Scheme for Procyanidin B-type Dimer (Branched) ………... 288

Figure 103: Fragmentation Scheme for Hegoflavone A ………………………………… 289

Figure 104: Fragmentation Scheme for Morelloflavone ………………………………… 290

Figure 105: Fragmentation Scheme for Catechin/Epicatechin Trimer with Single A-type Interflavanic Linkage (IFL) ………………………………………………………………. 291

Figure 106: Fragmentation Scheme for Manniflavanone ……………………………….. 291

Figure 107: Fragmentation Scheme for Kolaflavanone …………………………………. 292

Figure 108: Fragmentation Scheme for Aesculitannin C ………………………………... 293

Figure 109: Fragmentation Scheme for 3,3’,4’,5,5’,7-Hexahydroxyflavanone (a.k.a. Dihydromyricetin) …...... 294

Figure 110: Fragmentation Scheme for Catechin/Epicatechin Tetramer with Single A- type Interflavanic Linkage (IFL) …………………………………………………………. 295

Table 19: Biological Activities of Identified (and Tentatively Identified) Compounds Found in Peanut Skin Fractions ………………………………………………………….. 296

xvi INTRODUCTION

Oxidation is a major factor that limits the quality of food products. In the past,

many consumers have opposed the use of synthetic antioxidants as a means of reducing

spoilage in their foods. Many food companies have taken steps toward removing synthetic antioxidants from their products due to consumer demands. However,

consumers view natural antioxidants as a safe means to reduce oxidation and spoilage in

food products. The development of these antioxidants from natural sources has become a continuing challenge to the food industry.

Arachis hypogaea, also known as the peanut or groundnut, is very important as an oilseed and food crop in many tropical/subtropical countries. In the past, peanut skins

have been viewed as a low value byproduct of peanut processing and roasting. Recent

studies have shown that these byproducts contain compounds including ,

isorhamnetin, epicatechin, catechin, resveratrol, and quercetin. Previous research has

revealed that these compounds exhibit high oxyradical quenching abilities. In addition,

these natural antioxidants also have been reported to be responsible for health benefits in

humans consuming wines and other plant materials. The discovery of the relationship

between red wine phenolic antioxidants and reduced risk of cancer and heart disease has

caused a renewed interest in the consumption of grapes and wines. The identification of

similar compounds in peanut skins may result in similar consumer interest, improved

sales, and increased value of peanuts. It is evident that peanut skins may be a potential

untapped source for the extraction of natural food antioxidants, nutracueticals, and even

pharmaceuticals.

1 The objective of this research is to identify compounds in peanut skin extracts

which can be utilized as a natural antioxidant source for foods, with the potential use as

nutraceuticals and/or pharmaceuticals. Enhancing the value of peanuts by increasing uses

of their low-value byproducts could have several effects. The first advantage is

additional value from extracts. Second, there could be renewed consumer interest in

peanuts if valuable bioactive compounds are identified. Because peanut skins are largely

a wasted resource to peanut processors, identifying novel polyphenols could have a

significant financial impact on the value of peanuts.

The information available on the composition of polyphenols in peanut skins is incomplete. Some information has been previously reported on polyphenols present in

peanut skins; however, few compounds have been identified due to the detection methods utilized in those studies. However, preliminary studies of the current research revealed that high quantities of unique compounds actually exist in peanut skin extracts. This

study aimed to reveal these compounds by first using a preparatory Size Exclusion

Chromatography (SEC) technique. Individual fractions were analyzed for total phenolic

content using the Folin-Ciocalteau method, and antioxidant activity using the DPPH

oxyradical scavenging method. Compounds in each fraction were identified using an

ESI-HPLC-MSn Ion Trap (MS/MS/MS) technique. This combination of methods has

never been performed in the analysis of peanut skin extracts, and resulted in the

discovery of novel phenolic and flavonoid-based compounds. The ultimate goal of

utilizing HPLC-ESI-MSn is not only to identify structures which have known antioxidant

activity, but also determine if compounds are present which exhibit specific biological activities from a pharmaceutical and/or nutraceutical standpoint.

2 CHAPTER I: REVIEW OF THE LITERATURE

Peanut Industry

Arachis hypogaea, also known as the peanut or groundnut, is an important oilseed and food crop in many tropical/subtropical countries. The major peanut producing regions include India, China, Africa, and the United States (U.S.). Although the U.S. only produces roughly 10% of the world’s peanut crops, it does however produce higher average yields per acre (~3,000lbs/acre) than the rest of the world (~1,000lbs/acre) due to more advanced technology production practices (Acquaah et al., 2005). Domestically, peanuts are an important agricultural crop in many Southern states. Total U.S. production in 2008 was 5.15 billion pounds, with Alabama, Florida, Georgia, North

Caroline, South Carolina, and Texas having the highest production (USDA, 2009). In

2008, Georgia produced nearly three times more peanuts than Texas (2.33 billion vs 860 million lbs.), the next highest producing state. Other important top-ten producers included Oklahoma, Virginia, New Mexico and Mississippi.

In 2002, the Farm Bill caused drastic changes regarding the manner in which the federal government managed peanut production, moving from a quota system to a federal loan program. The years following this act saw a consistent drop in peanut production in the state of Virginia, from 235 thousand tons to 96 thousand tons. Comparatively, the total U.S. production fluctuated during this time, but remained somewhat stable (Boriss and Kreith, 2006). Increasing the value of peanuts for farmers and processors is difficult with low priced imports steadily eroding market share for American peanut farmers. For this reason, peanut farmers are constantly searching for new ways to create additional revenue streams from unutilized and underutilized peanut parts. (Fletcher et al., 2000).

3 Peanut Varieties

The four main market types of peanuts grown in the U.S. include Virginia,

Runner, Spanish and Valencia. Spanish peanuts are largely grown in Texas and

Oklahoma, while Valencia types are mainly produced in New Mexico and Texas. In the past 30 years, Runner peanuts have become the dominant type due to the enormous success of the Florunner variety. This variety was released by the University of Florida

breeding program in 1969, and has seen excellent production characteristics under a

variety of environmental conditions. However, other varieties have been released in

recent years which out-perform Florunner, primarily Virginia-type peanut varieties.

Virginia type peanuts are mostly grown in South-Eastern Virginia and North-Eastern

North Carolina. The Gregory variety has become an increasingly popular Virginia type

peanut, due to its extremely large pod size and low susceptibility to tomato spotted-wilt

virus (which is a common problem with other peanut varieties) (Virginia Peanut

Production Guide, 2009).

By-Products of the Peanut Industry

Some common uses of peanut kernels include: whole peanut products (such as

roasted peanuts, peanut butter, peanut cake and meal), fermented peanut products,

extracted products (such as peanut oil, peanut protein isolates, and peanut flour), and as

inclusions or ingredients in candies and cookies. However, obtaining the kernels results

in several by-products as a result of harvesting and production. After harvesting, peanut

plants are either baled (for use as animal feed) or left in the field (Golden et al., 2001).

Other than shell material (or hulls), one of the most notable by-products is peanut skins.

4 Peanut skins are a low value by-product of peanut blanching, which is the process of removing peanut skins (from peanuts) before roasting. Peanut skins also can be obtained as a by-product of the roasting process. A typical peanut mill may produce 17 tons of peanut skins per week. (Lee, 1996).

Peanut skins currently have almost no value, and sell for as little as $0.01 per pound. In most cases, they are utilized as a low-cost animal feed. It has been shown that

peanut skins can be used in poultry and ruminant diets at levels around 5-10% (of the

total diet) without negative consequences on growth rates. However, higher levels of

feeding may result in lower growth performance, since tannins interfere with protein

utilization (Atuahene et al., 1989; Hill, 2002).

Crude skins obtained from milling contain small amounts of peanuts and peanut

oil, although some blanchers have the ability to separate these components to produce

highly pure skins. Peanut skins often are pelletized after blanching, which reduces their bulk density for easier handling/distribution (and also lowers the risk of fire). Several studies have shown that unutilized and underutilized parts of the peanut and peanut plant have significant concentrations of phenolic compounds that exhibit antioxidant and biological effects (Lou et al., 1999, 2001, 2004; Nepote et al., 2000; Wang et al., 2007;

Yu et al., 2005, 2006). Therefore, the discovery of possible uses for these underutilized parts (such as natural antioxidants and/or nutraceuticals) could effectively turn these by- products into value-added commodities for peanut farmers.

5 Peanut Composition and Health

In recent years, peanuts have received significant attention regarding their possible role as a functional food. Peanuts were previously considered somewhat unhealthy, due to their high fat content of 50% w/w or more (Francisco and Resurreccion,

2008). However, in the last decade peanuts have been continually evaluated regarding their positive role in heart-healthy diets (Kris – Etherton et al., 1999). The benefits of consuming peanuts and peanut products have been shown to be associated with their nutritional profile, including essential fatty acids, and recent discoveries of bioactive components (Griel et al., 2004). The bioactive compounds, which have been presumed to induce beneficial health effects, include proteins, micronutrients, , plant sterols, and various (Kris – Etherton et al., 2002). Many of these compounds elicit their beneficial biological attributes by exhibiting antioxidant effects

(Chun et al., 2005), inhibiting bacterial and viral activity (Moure et al., 2001), reducing cholesterol (Kris – Etherton et al., 2001), activating detoxification enzyme release in the liver (Percival, 1997), and reducing platelet aggregation (Pignatelli et al.,

2000). Research continues to emerge revealing data showing the significant contribution that peanuts have on human health (Higgs, 2003).

Peanuts are considered one of the most concentrated foods on earth because of

their high protein and oil content. Among high protein foods, peanuts have one of the

highest caloric values. A whole peanut contains about 75-80% kernel, and 20-25% shell.

A peanut seed consists of roughly 50% oil, 25% protein, 15% carbohydrate, 6% moisture,

2% fiber, and 2% ash. The oil in peanuts contain around 82% unsaturated fatty acids and

is mainly comprised of oleic and . The high proportion of unsaturated fatty

6 acids gives peanut oil a high iodine value (measure of unsaturation) and low storage stability (Adsule et al., 1989).

Protein content in peanuts varies from 25-35% and is classified as albumin, arachin, glutelin, and conarachin. The main proteins are arachin and conarachin, which make up over 85% of the total protein content. The amino acid composition of peanut protein contains high levels of acidic amino acids and low amounts of sulfur-containing amino acids. Limiting amino acids include methionine, lysine, threonine, and very low amounts of . Amino acids such as glutamic acid, aspartic acid, , and histidine react with sugars to contribute to the flavor quality of roasted peanuts

(Adsule et al., 1989).

Florunner has been the most commonly grown peanut cultivar in the U.S. for many years. Florunner peanuts contain roughly 53% oleic acid and 27% linoleic acid.

However, newer cultivars with better oil stability and quality have been released in the past few years. The University of Florida developed a high-oleic peanut (HOP) variety that contains approximately 80% oleic acid and 3% linoleic acid (Gorbet and Knauft,

1997). Altering the fatty acid composition allows for a product that has improved resistance to oxidation, therefore increasing overall flavor stability. According to Gorbet and Knauft (1997), the new peanut lines also have improved yields, grades, less pod spotting, and less damage from tomato spotted wilt virus.

The unique fatty acid composition of peanut oil allows peanuts to have a high nutritional value. The high oleic acid content, especially in HOP lines, is desirable because it has a hypocholesterolemic effect. Mono and polyunsaturated fatty acids have

7 been shown to lower blood serum cholesterol levels and low density lipoproteins,

therefore reducing the risk of coronary heart disease (Shephard and Rudolph, 1991).

Peanut oil contains less than 1% unsaponifiable sterols, squalene, tocopherols,

and other antioxidants and hydrocarbons. The antioxidants consist mainly of α- and β- tocopherols, 18.6 and 13.8 mg/100g, respectively (Adsule et al., 1989). The phospholipids in the oil consist of phosphatidyl-ethanolamines, serines, and cholines. In addition to oil, carbohydrates are also an important functional component of peanuts.

Peanuts contain around 10-20% carbohydrates. In peanut flour, the majority of the carbohydrates are sucrose and starch. The carbohydrates are the precursors to many of the flavor characteristics found in roasted peanuts. Although many peanut lines have no significant variation in reducing sugar content, they do, however, reveal differences in total sugar content (Adsule et al., 1989).

Peanuts are a good source of vitamins and minerals. The main minerals contained in peanuts include phosphorus, calcium, magnesium, and potassium. Variations in minerals such as iron, , manganese, and zinc can be found among different lines.

Although peanuts contain only small amounts of vitamins A, C, and D, they are a good source of B vitamins. Peanut kernels contain high amounts of , thiamin, riboflavin, panthothenic acid, pyridoxine, biotin, inositol, and folic acid (Adsule et al., 1989).

In recent years, peanut kernels have been shown to contain significant levels of

phytochemicals, specifically resveratrol (Sanders and McMichael, 2000). Resveratrol

(3,5,4’-trihydroxy-stilbene) has been shown to exhibit several beneficial biological

activities, including the reduced risk of cardiovascular disease (Goldberg and Soleas,

1999) and cancer (Jang et al., 1997). Resveratrol is a stilbene compound, which is

8 produced by the peanut plant as a defense mechanism against exogenous stimuli, such as fungi and UV light (Ingham, 1976). In addition to peanuts, this compound has been reported in numerous types of plants including grapes (Jeandel et al., 1991), and has also been shown to be one of the possible compounds responsible for the human health benefits that are associated with the consumption of red wines (Siemann and Creasy,

1992). Numerous researchers have taken a great interest in the compound resveratrol because of its alleged role in the French Paradox (Frankel et al., 1993; Kopp, 1998), which states that the French have relatively low incidences of cardiovascular disease while having high intakes of saturated fats (and other risk factors). A study by Wu et al.

(2001) supported this theory by confirming the cardioprotective attributes of resveratrol.

Interestingly, the presence of resveratrol was discovered in peanuts long before it gained attention for its important biological properties (Ingham,1976). Resveratrol has been detected in commercial peanut products at a concentration range of 0.06 ppm in roasted peanuts to 5.1 ppm in peanut butter (Sobolev and Cole, 1999), which is similar to the 0.6

– 8.0 ppm levels found in red wine (Sanders et al., 2000).

In addition to peanut kernels, recent studies have shown that by-products of the peanut industry, including unutilized and underutilized portions such as peanut roots, leaves, and hulls, contain compounds that have potential health benefits. Peanut roots have been shown to contain resveratrol and other bioflavonoids which exhibit antioxidant and nutraceutical characteristics (Dean et al., 2008). Of significant importance is the concentration of resveratrol in peanut roots, which was reported to be much higher than that found in red wines (Chen et al., 2002; Liu et al., 2003). Variations in resveratrol levels were reported by Chen et al. (2002), which showed extremely high levels (up to

9 1330 ppm) during one growing season, but much lower levels the next season (27 – 37 ppm). Resveratrol is a phytoalexin which is produced (by a plant) as a defense mechanism, therefore, year to year variations may be a factor related to changes in disease incidence (or other plant stressors) that initiates synthesis of these types of compounds. Another study reported resveratrol concentrations of over 114 ppm in peanut roots (Liu et al., 2003), which is significantly higher than the levels found in red wines.

In addition to resveratrol, another study (Dean et al., 2008) reported the existence of and isoflavones in peanut roots. This study also reported extremely high levels of phenolic compounds in peanut leaves, which exhibited powerful antioxidant activities. Other studies that investigated extracts from peanut hulls revealed powerful antioxidant activities due to the presence of phenolic compounds such as (Duh et al., 1992; Yen and Duh, 1995), which also has been shown to exhibit anti-cancer activities (Chowdhury et al., 2002).

One of the most interesting by-products of the peanut industry, from a phenolic and bioflavanoid standpoint, is peanut skins. Peanut skins have been used in traditional

Chinese medicine for centuries for the treatment of disorders related to atherosclerosis, ulcers, mucous secretion inhibition, inflammation, liver/kidney problems, and hypertension (Haslam, 1996, 1998; Shahidi and Naczk, 1996). Peanut skins were thought to be toxic to humans in the early 1940’s; however, a study by Dr. Jack Masquelier at the

University of Bordeaux in France, showed that peanut skins were in fact non-toxic. The study also revealed that peanut skin extracts had the ability to protect and strengthen blood vessels (Louis et al., 1999). Recent studies have revealed the presence of

10 important phenolic-based compounds in peanut skins, such as isoflavones, isorhamnetin,

quercetin, catechin, epicatechin, resveratrol, and proanthocyanidins (Lou et al., 1999,

2001; Yu et al., 2005). These compounds have been shown to exhibit powerful free- radical scavenging effects (Lou et al., 2001; Shahidi et al., 1997), and also have been reported to be responsible for the health benefits associated with wine and other plant products (Goldberg et al., 1998; Kampa et al., 2000; Kim et al., 2000; Kinsella et al.,

1993). Phenolic compounds such as catechin and resveratrol have been shown to lower

LDL oxidation in vivo (Andrikopolous et al., 2003; Das et al., 1999; Frankel et al., 1993;

O'Byrne et al., 2002). Other researchers have reported that resveratrol has a wide range of other beneficial biological effects, such as vasodilation via endothelium dependant relaxation (EDR) (Fitzpatrick et al., 1993, 2000), reduction of platelet aggregation (Pace-

Asciak et al., 1995), and cancer prevention (Jang et al., 1997). It has been shown that peanut kernels contain up to 1.8 ppm of resveratrol; however, the reported levels detected in peanut skins were five times the concentration found in kernels alone (Sanders et al.,

2000). These numbers equal or exceed the levels (0.6 – 8.0 ppm) found in red wines.

In addition to resveratrol, the phenolic flavonoid quercetin also has been shown to exhibit anti-cancer activities (Huang and Ferraro, 1992). DiSanto et al. (2003) reported that membrane receptors involved in the coagulation cascade (known as tissue factor), are down-regulated by both resveratrol and quercetin. Other compounds discovered in peanut skins that have potential human health benefits include A-type proanthocyanidins

(Lou et al., 1999). Das et al. (1999) indicated that extracts of red wines, which included resveratrol and proanthocyanidins, are equally efficient in reducing damage related to myocardial ischemic reperfusion. This research revealed the significant function in

11 which these antioxidants play in protecting the heart during certain types of ischemic

events.

Although the health benefits of flavonoid-based compounds is clear, their bioavailability is somewhat ambiguous. Manach et al. (2005) indicated that of the

roughly 4,000 that exist, the bioavailability of these compounds varies greatly.

In addition, some of the most common dietary flavonoids do not have the most efficient

bioavailability profile. Of the most abundant flavonoids, isoflavones have the highest

bioavailability, followed by catechin, , and quercetin-based . The

lowest bioavailability profile among the flavonoids is exhibited by proanthocyanidins,

galloylated , and anthocyanins (Manach et al., 2005).

Phenolic Compounds in Peanuts and Other Plants

Peanuts fall into the plant family known as Fabacea. These species are classified

as dicotyledonous plants that produce legume or loments fruit (such as peanuts, peas, and

beans), and have root nodules containing bacteria that fix nitrogen (Hopkins and Huner,

2004). As in the case of all plant species, this family uses the process of photosynthesis

to produce energy gathered from the sun. This harnessed energy results in the formation of sugars via the photosynthetic pathway, which are then utilized in further plant functions and metabolism. One of the most important nutrients in plant metabolism is nitrogen, which is also one of the most abundant in plants. Obtaining nitrogen is a crucial requirement in plant metabolism for the creation of plant metabolites, proteins, and acids

(Vance, 2002). Legumes form root nodules, which contain a symbiotic bacteria known as Rhizobia. These bacteria produce the enzyme nitrogenase, which reduces nitrogen gas

12 to ammonia (which can then be used for metabolic processes). Legumes are capable of

nitrogen fixation via symbiotic interaction between the roots and Rhizobia bacteria.

These bacteria form in the root nodules due to compounds released by the plant called

chemoattracts, which include phenolic flavonoids, , carboxylic acids,

aromatic acids, and amino acids. Not only do these compounds aid in the growth of

Rhizobia bacteria, but they also provide a nutrient source and function in plant defense

(Daroka and Phillips, 2002). According to Vance (2002), nodulation in legume plants is

initiated by flavonoids and isoflavonoids such as luteolin, hesperitin, daidzein, and

. In addition, isoflavonoids are important inhibiters of certain diseases, and act

as phytoalexin antiobiotics during microbial infections. The accumulation of these

isoflavonoids also inhibits the growth of attacking organisms (Burns et al., 2002; Chung

et al., 2003).

Plant materials contain numerous classes of phenolic and polyphenolic

compounds. As part of their structure, plants produce a complex set of secondary

metabolites which incorporate phenolic rings with hydroxyl (and sometimes methoxy)

substitutions. From a biological standpoint, minor structural differences can have drastic

effects on chemical and biological activities. These chemical and biological attributes

depend on factors such as structural class, degree of hydroxylation, degree of methoxylation, the presence of substitutions such as glycosidic (sugar) side-groups, and conjugations among A and B rings (Francisco and Resurreccion, 2008).

A significant amount of research has been conducted on phenolic compounds over the last decade, and over 6,000 different flavonoids are currently known (Harborne and Williams, 2000). In addition, about 400 flavone aglycones, 450 flavonol aglycones,

13 350 flavanone aglycones, 300 aglycones, 19 anthocyanidins, and 250 chalcone

aglycones have been reported (Iwashina, 2000). Results from these studies have literally created a new food trend, in which foods are now considered “healthy” if they contain certain concentrations of flavonoid compounds (Escarpa and Gonzalez, 2001). This increase in interest regarding phenolic-based compounds was sparked by the phenomenon known as the French Paradox, which attributed phenolic antioxidants in red wine to low rates of coronary heart disease in France (Frankel et al., 1993).

Phenolic compounds are an extremely diverse class of secondary plant metabolites, which are found in most types of plant species. These compounds include phenolic acids, flavonoids, chalcones, coumarins, and tannins (Anderson and Markham,

2006). The presence of polyphenolic structures are a result of the shikimic acid and acetate pathways (Fennema, 1996). Different types of phenolic acids include hydroxycinnamic and hydroxybenzoic acids, which encompass the majority of phenolic acids found in the plant kingdom (Wrolstad et al., 2005). Flavonoids are considered the most abundant of phenolic compounds found in plants. They constitute a relatively large family of aromatic molecules that are derived from shikimic acid and phenylalanine

(Figure 1). Flavonoids are classified according to their structures, and include the flavones, isoflavones, , isoflavonols, flavanones, isoflavanones, flavanols, isoflavanols, flavanes, dihydroflavonols, flavandiols (leucoanthocyanins), anthocyanins, proanthocyanins and procyanidins (condensed tannins) (Havsteen, 2002). Flavonoids such as flavones (apigenin, luteolin), flavonols (quercetin, myricetin, kaempferol, and isorhamnetin), flavanones (eriodictyol, hesperitin, and ), isoflavones

(, daidzein), and -3-ols (or flavanols, such as catechin and epicatechin) are

14

Figure 1: Shikimic Acid Pathway: Example of the Shikimic Acid Pathway for 5,7-dihydroxyflavanone (a.k.a. ).

15 common and widely distributed in plants. Based on the flavonoid nucleus structure

below (Figure 2), flavanols are missing the 2,3-double bond and 4-one structure (Rice-

Evans et al., 1997). Flavones and flavonols exhibit similar C-ring structures, with both

having double bonds at the 2,3 position (although, flavones lack a hydroxyl group at the

3-position) (Hollman and Arts, 2000). Most of these compounds exhibit low molecular

weights; however, polymeric tannins can have molecular weights up to 30KDa. Higher

molecular weight tannins have been a topic of research for years, due to their binding of

dietary proteins and enzymes, which lowers their digestibility. Tannins are hydrolysable,

and can be split into gallic or ellagic acids (and other condensation products), and then re-

condensed. The monomer units that form condensed tannins include flavan-3-ols

(catechin, epicatechin). These structures can be hydroxylated at the 4 position on the

flavonoid C-ring to form leucoanthocyanins (flavan-3,4-diols) (Bravo, 1998). The

polymers (or oligomers) of flavan-3-ols also are known as proanthocyanidins, and are the

second most abundant of all naturally occurring phenolics. The size of a given

proanthocyanidin molecule is governed by the degree of polymerization (Francisco and

Resurreccion, 2008). Monomer units are typically linked via C4-C8 bonds, although C4-

C6 bonds also are found. The oligomeric procyanidins (OPC’s) formed by these types of

bonds are called B-type procyanidins, due to the B-type linkages between the monomer

units. In contrast, condensed tannins also can be linked via A-type linkages, which are formed when additional C2-C7 ether bonds result in a double-linkage of the flavan-3-ol monomer units (Rasmussen et al., 2005). Proanthocyanidins consist of only catechin and epicatechin monomer units, and are generally called procyanidins (Francisco and

Resurreccion, 2008). A study by Karchesy and Hemingway (1986) reported that peanut

16

Figure 2: Flavonoid Classes: Showing flavonoid nucleas, and base-structures for the main flavonoid classes (including flavanone, flavones, isoflavone, flavonol, and chalcone).

17 skins contain approximately 17% procyanidins by weight, of which 50% were lower

molecular weight oligomers. Studies by Lou et al. (1999, 2001) revealed six A-type proanthocyanidins (including types A-1 and A-2) and three newly discovered epicatechin oligomers in peanut skins. Further studies by Lou et al. (2004) indicated the presence of several B-type proanthocyanidins in peanut skins, including B-2, B-3, and B-4. Yu et al.

(2006) reported the presence of both A-type and B-type procyanidin dimers, trimers, and tetramers in peanut skins.

The formation of flavonoids is part of the shikimic acid pathway, and results from a series of condensation reactions among cinnamic acids (between B-ring and carbon atoms 2, 3 and 4 of the C-ring) and malonyl residues (A-ring), forming a C6—C3—C6 base structure (Figure 1). The bridge (of three carbons) between the phenyl rings is often cyclized to form a third ring, which becomes the flavonoid C-ring (Cuyckens and Claeys,

2004). Depending on the cyclization reaction, and the degree of unsaturation and oxidation of the three-carbon segment, the resulting flavonoid can be classified into several groups (Cuyckens and Claeys, 2004).

The foundation for the numerous configurations exhibited by all flavonoids is the basic flavonoid nucleus structure (Figure 1). Some of the fundamental structures which make up the skeletons for various flavonoid classes also can be found in Figure 1.

Individual compounds in each class are distinguished by the pattern of substitution(s) on the flavonoid A and B rings (Pietta, 2000; Stalikas, 2007). Groups of flavonoids are classified by hydroxylation of the basic nucleus structure, and many of these phenolic compounds exist as glycosides (i.e., sugar-linked). Using the numbering system denoted in Figure 1, glycosylation often occurs on the number 3 carbon on the flavonoid C-ring.

18 However, glycosylation also can occur at other positions on the flavonoid ring, including

the number 5 and 7 positions (Cuyckens and Claeys, 2004). Common glycosidic units

include , galactose, , and .

Oxidation and Antioxidants

The process in which electrons are removed from atoms or pairs of atoms is

known as oxidation. The molecule then becomes oxidized via the loss of an electron. At

the same time, another molecule may become reduced from receiving an electron (via

oxidation-reduction reactions). The end result is of specific importance because of the

formation of free radicals (Lee et al., 2003; Wong and Pavlath, 1992). Free radicals that

are oxygen-centered are known as reactive oxygen species. Chemical reactions involving

reactive oxygen species are of particular concern because they are thought to cause

adverse biological effects such as cancer, heart disease, accelerated aging, and damage to cell membranes (Nijveldt et al., 2001; Steinberg, 1993). In addition to human health, free radical chain reactions via the autoxidation of lipids are an issue in the food industry.

Lipid oxidation can result in decreased shelf-life and the development of off-flavors in

food products. Other undesirable side-effects include the reduction of quality and nutritional value. The rate of oxidation in lipids can be affected by several factors, including fatty acid composition, temperature, oxygen levels, relative humidity, surface area, radiant energy, and the presence of antioxidant compounds (Fennema, 1996;

Dziezak, 1986). In addition to lipid oxidation, proteins also can undergo chemical oxidation, resulting in changes in chemical and physical properties due to protein cross- linking. Other mechanisms involve enzymatic oxidation, such as the oxygenation of

19 unsaturated fatty acids by lipoxygenase, polyphenol oxidase catalyzed browning reactions, and glucose to lactose converstion by glucose oxidase (Buford, 1988).

Traditional synthetic antioxidants, as well as natural antioxidants, work to inhibit oxidative processes in foods (Buford, 1988). Oxidative reactions can be chemical or enzymatic (Dziezak, 1986). One of the best examples of the oxidation mechanism involves the chemical oxidation of lipids. Lipid oxidation can occur either by photosynthesized oxidation or via autoxidation. The autoxidation process is a free- radical chain reaction involving three steps: initiation, propagation, and termination

(Fennema, 1996). During initiation, loosely held hydrogen atoms are lost from the fatty acid, typically from carbons adjacent to double bonds on the carbon chain. Fatty acids with multiple double bonds undergo higher levels of oxidation based on the degree of saturation, since more allylic and doubly allylic sites are available for free radical formation. The removal of this hydrogen results in a fat free radical, which can react with oxygen to form peroxyl free radicals. Peroxyl free radicals can continue the oxidation chain reaction by acting as strong initiators (or catalysts), since they can remove additional hydrogen atoms from other fat molecules, and thereby triggering the prorogation step (McCord, 1994). During propogation, peroxyl free radicals strip hydrogen atoms from other lipids to form (relatively) stable hydroperoxides, resulting in an additional newly formed unstable fatty radical. This unstable fatty radical can continue the chain reaction by reacting with oxygen to create another newly formed reactive peroxyl free radical. The final step in the autoxidation process (termination) involves hydroperoxides breaking-down to form additional short-chain organic compounds, such as aldehydes, ketones, acids, and alcohols (all of which are the

20 contributors to off-flavors and off-odors associated with fat and oil rancidity) (Fennema,

1996). Autoxidation ends by the reaction between two unstable radicals, or when fatty free radicals react with a stable antioxidant radical. Antioxidants use two mechanisms to alter the free radical chain reaction. First, they can terminate oxidation during the initiation step by donating hydrogen atoms to fat free radicals, thereby reforming the fat molecule. Until the antioxidant compounds are used-up, this phenomenon can delay further oxidative reactions (Dziezak, 1986). Antioxidants also can function to break the chain-reaction by donating hydrogens to peroxyl free radicals, forming antioxidant free- radicals and hydroxperoxides (which are more stable than fat free-radicals). Antioxidant free radicals also are more stable than fat free radicals, because of the electron spin- resonance structure in the aromatic (phenolic) ring. (Buford, 1988; Dziezak, 1986;

Fennema, 1996; McCord, 1994).

Many of the phytochemicals found in plants, such as tocopherols, carotenoids, acids (such as ascorbic acid), and phenolic-based compounds, exhibit some level of antioxidant activity. The level of antioxidant activity typically depends on the chemical compound itself, as well as the model system being used for study. The mode of the antioxidant activity can occur in several different ways. First, polyvalent phenolics may exert antioxidant activity by chelating effects, through the binding and deactivating of metal ions which have pro-oxidant activity. Many of these phenolic compounds can play a role as primary antioxidants by breaking the free radical chain reaction, or by scavenging or binding with active oxygen radicals such as superoxide anion (Reische et al., 2002).

21 The use of antioxidants can help prevent the formation of undesirable/unstable

structures in foods which can lead to cellular damage. Multiple studies have been

conducted on naturally-derived antioxidants to investigate the potential health impact of antioxidants contained in common foods. Many of these studies are based on the possible negative effects exhibited by synthetic antioxidants and preservatives utilized in the food supply (Peschel et al., 2006). Many researchers have begun to investigate the possible dangers associated with ingesting synthetic antioxidants, such as (BHT), butylated hydroxyanixole (BHA), tertiary butylated hydroxyl- quinone (TBHQ), ethylenediaminetetraacetic acid (EDTA), and (PG).

This has generated extreme interest in naturally-derived antioxidant compounds. The incorporation of natural antioxidants (as opposed to synthetically-derived) could have significant impacts on human health (Nepote et al., 2004). The use of natural antioxidants in herbs and spices (such as sage and rosemary) has been used for years in the meat industry; however, their applications are limited based on the significant herb flavor conveyed in the product (Chang et al., 1977).

Even though phenolic-based compounds are considered non-nutritive, they do however have significant health effects related to antioxidant activity. In recent years, a considerable amount of research has been conducted on phenolic compounds in relation to their antioxidant activity and ability to scavenge free-radicals (Buford, 1988; Escarpa and Gonzalez, 2001). In addition to their antioxidant defense mechanisms among biological processes, these phenolic-based compounds exhibit antioxidant properties which offer significant stability to food systems (Dziezak, 1986; Macheix and Fleuriet,

1998). Polyphenols in plants can perform the function of reducing agents, thereby acting

22 as hydrogen donators and quenchers of singlet oxygen. Phenolic compounds often found in nature that are capable of such reactions include p-hydroxybenzoic acid (or p-coumaric

acid), chlorogenic acid, caffeic acid, ferulic acid, resveratrol, catechin, and epicatechin.

Onions contain large quantities of quercetin, which is the main flavonol in the human diet

(Yang et al., 2001). Flavonoids including catechin, epicatechin, kaempferol, quercetin

(and their flavonoid- derivatives) have been previously found in teas and wine

(Nijveldt et al., 2001), all of which exhibit powerful oxyradical quenching properties.

Flavonoid antioxidant compounds also are found in many types of nuts, including walnuts (which have been shown to have the highest flavonoid content), pecans, pistachios, cashews and almonds (Yang et al., 2009). Reports by Shahidi and Naczk

(1996) indicate that the polyphenolic content of peanut kernels is low, however, the content in the skins is much higher (Nepote et al., 2000). In addition, data have revealed that the antioxidant potential of compounds detected in peanuts can contribute to human

health (Higgs, 2003).

Numerous studies on phenolic compounds extracted from by-products of the

peanut industry have indicated they exhibit powerful antioxidant activity. A study on

peanut hulls confirmed the presence of 5,7-dihydroxychromone, eriodictoyl, and luteolin

(Daigle et al., 1985). In addition to the aforementioned compounds, another study

identified three additional flavonoids (Lin et al., 1999). They also used a copper

catalyzed LDL oxidation model system to mimic LDL oxidation in vivo. The results

indicated that eriodictyol and luteolin had the greatest antioxidant activity. Other studies

have investigated the antioxidant potential of peanut skin extracts. Pratt and Miller

(1984) reported antioxidant activity in Spanish peanuts, although, with the skins left on

23 the kernels. They indicated the antioxidant activity was due to the flavonoid compound known as dihydroquercetin. Another study reported catechin, epicatechin, and oligomeric procyanidins (OPC’s) in the form of dimeric flavonoids (Karchesy and

Hemmingway, 1986). Later studies indicated these compounds (including proanthocyanidin A-1 and A-2) exhibited substantial inhibitory activity against hyaluronidase (Lou et al., 1999). The presence of also was identified in peanut skins, and was reported to have antioxidant activity in a linoleic acid model system (Huang et al., 2003). Crude methanol and extracts from the peanut skins of Florunner peanuts were reported to have high radical scavenging activity

(Nepote et al., 2000). This study utilized a sunflower model system, which indicated that methanol and ethanol peanut skin extracts had lower antioxidant activity (at 500ppm) than BHT exhibited at 200ppm. This contradicts other studies that compared the antioxidant activity of peanut skin extracts to that of synthetic antioxidants (such as BHA,

BHT, TBHQ, etc.). This is in part due to the incorporated mechanism of the sunflower oil anhydrous model system. Most antioxidant activity techniques measure the ability to scavenge free radicals, which is often considered a more efficient way of interpreting a compounds ability to act as an antioxidant. However, when measuring the antioxidant activity in food model systems, specific antioxidant compounds often can perform well, while others are inadequate. For this reason, oil model systems cannot easily predict a compound’s antioxidant potential, since some compounds may be poor oxyradical scavengers in oil-based systems, but more powerful in other food models (or pure radical scavenging models).

24 As previously discussed, the activity in which each flavonoid plays in human

health is determined by its chemical configuration. This also holds true regarding a flavonoids ability to act as an antioxidant. Antioxidant activity is generally governed by

the positions and types (and number) of substitutions of the flavonoid nucleus.

Antioxidant and chelating activity is directly related to the number of hydroxyl groups on the flavonoid structure. Plant phenolics exhibit what is known as primary antioxidant activity. These types of structures form stable products from free radicals by donating hydrogen or electrons, thereby terminating the free radical chain reaction. (Dziezak,

1986).

The potential industry development of naturally-derived antioxidants from plant sources not only could enhance the nutritional value of foods, but also aid in preservation.

These raw extracts from fruits, vegetables, nuts, and other plant sources, have progressively become an important topic in the food industry because of their possible role as antioxidants from natural sources (Sang et al., 2002). Many consumers have become more aware of the possible dangers associated with synthetic-based ingredients,

specifically antioxidants and preservatives. The growing trend toward whole foods, organics, and ingredients from natural sources has been largely due to increased education into practices utilized by the food industry. This move toward more healthy foods also impacts research into natural sources in which to extract compounds with possible antioxidant activity. Many of the potential sources for these naturally-derived antioxidants come from by-products of the food industry, such as peels, seeds, and hulls

(Lee et al., 2003). In order to retain the ability to germinate (and for long-term protection and preservation), seeds from plants typically contain powerful antioxidants around the

25 kernels or germs. Other than peanuts, an example of seed-coats (or skins) that contain

flavonoids and flavonoid-glycosides with antioxidant activity are those found in almonds

(Frison and Sporns, 2002; Milbury et al., 2006; Wijeratne et al., 2006). In addition,

leaves and barks are typically exposed to sunlight and oxygen, and often contain

antioxidants to protect against UV light and oxidation. Diverse by-products of the

agricultural industries are produced in immense volumes, and signify an enormous

underutilized resource (Francisco and Resurreccion, 2008). This could have a huge

impact on both the economy and environment, by reducing disposal of typical waste-

products through the creation of renewable uses from by-products such as natural

antioxidants, functional foods and nutraceuticals. Other economic implications would be

a function of added revenue through the creation of value-added commodities for specific

agricultural industries. However, many challenges exist including those associated with

extracting and isolating these compounds from the raw agricultural product. Other

factors include the cost of such procedures, the possibility of environmental waste, and the simultaneous extraction of other unwanted compounds.

Extraction Methods

The utilization of an extraction method is critical for the removal of phenolic- based compounds from plant sources. The process of extraction is a common laboratory procedure to carry out the isolation and/or purification of compounds from a source.

Typical methods include liquid-liquid, solid-liquid, and acid-base extractions (Moure et al., 2001; Shi et al., 2005; Silva et al., 1998). Many different forms of these methods have been incorporated for isolating active compounds from plants with varied levels of

26 success. Methods are selected based on the original sample form, the desired resulting form, and the further analytical methods used after extraction. Extraction methods are also custom-tailored to allow certain physical and chemical properties of the extracted components to remain intact. The majority of extraction methods for plant-related compounds utilize a form of solid-liquid extraction, in which a solid material comes in direct contact with a liquid solvent (Houghton and Raman, 1998). Organic solvents (such as ethanol) are commonly used, which solubilize compounds of similar polarity by diffusing into solid plant material. Solvents often are chosen based on the class of compounds being extracted. Polar organic solvents include water, methanol, and acetone, and are used to extract polar structures. Nonpolar solvents such as hexane and dichloromethane are typically utilized for the extraction of nonpolar compounds. (Moure et al., 2001; Shi et al., 2005; Silva et al., 1998)

Several prior studies have been conducted that investigated the yield of peanut skin extracts using various solvents. Huang et al. (2003) indicated that polar solvents offered better extraction yields than less polar solvents. Yu et al. (2005) showed that methanol and ethanol were more efficient than water for extracting phenolic-based compounds. The use of methanol as an extraction solvent yielded roughly 200mg extract per gram peanut skin. Additional studies that incorporated ethanol as an extraction solvent offered approximately 50% (or more) reduction in extraction efficiency. Wang et al. (2007) showed 0.107 ± 0.003g extract per gram peanut skin. Other reports indicated that the utilization of ethanol resulted in extracts of 0.051g (Huang et al., 2003) and

0.099g (Nepote et al., 2005) per gram peanut skin. In addition, several groups have examined the optimum solvents for extracting antioxidants from peanut hulls. Duh et al.

27 (1992) revealed that methanol resulted in more than 50% higher yields of antioxidant

extracts from peanut hulls when compared to ethanol (other solvents gave even lower

yields). The difference in extraction yield (between methanol and ethanol) also was

revealed in this study. Conversely, Li et al. (1995) reported that extracts using ethanol

had the most potent antioxidant activity. The compounds that exhibited antioxidant

activity extracted from peanut hulls then was investigated by Lin et al. (1999). They

reported the strongest antioxidant effects in a model system of copper catalyzed low density lipoprotein (LDL) using the main extracted components, luteolin and eriodictoyl

(structural difference is double bond at C3 in flavonoid structure). Although peanut hulls contain compounds with significant biological activity, their potential as a suitable raw material for the extraction of bioactive compounds is relatively low (versus peanut skins) based on their low bulk-density.

In the case of peanut by-products, several other factors can have an effect on the extraction levels of phenolic-based compounds. Yu et al. (2005) indicated that blanched peeling greatly reduced the concentration of total phenols in methanolic peanut skin extracts versus direct peeling or roast peeling (11.6 mg/g versus 90.1 mg/g and 96.7 mg/g respectively). The lower total phenolic content in that study was assumed to be due to the leaching of phenolics out of the skin (into the water) during the blanching process. A similar result was obtained in a prior study on the total phenols contained in peanut hulls.

Yen and Duh (1995) showed significant differences in the total extracted phenolic concentration (including the total luteolin content) in various peanut hulls obtained from different cultivars. This shows that in addition to peanut hulls, other forms of peanut by-

28 products (such as peanut skins) can contain varying levels of extracted phenolics based on differences in cultivars.

One of the most notable methods for the extraction of phenolic antioxidants from peanut skins was reported by Ballard et al. (2009). The extraction technique developed for this study incorporated a novel response surface methodology to estimate the optimum extraction conditions for several solvents, including methanol, ethanol, and water. The study determined the optimum solvent concentration, temperature, and time required for maximum extraction of phenolic antioxidants (such as resveratrol) from peanut skins. The data obtained from this study could increase the potential for peanut skins to become a viable extraction medium for naturally derived antioxidants, and also provide efficient methods for extracting bioactive compounds from this plant source.

Antioxidant Activity Methods

Several different methods exist that are capable of measuring the antioxidant activity of plant phenolics and extracts. These techniques work in several different manners, which typically concentrate on one reaction mechanism. These actions include the ability to scavenge oxygen and hydroxyl radicals, chelate metal ions, reduce lipid peroxyl radicals, or inhibit lipid oxidation. One method, known as the Oxygen Radical

Absorbance Capacity (ORAC), is used to verify the ability of an antioxidant compound to scavenge free radicals generated by a given system. Another method that utilizes this mechanism is the Trolox Equivalent Antioxidant Capacity (TEAC). Other mechanisms include the ability of an antioxidant to inhibit lipid peroxidation products, such as malonaldehyde by way of Thiobarbituric Acid Reactive Substances (TBARS) (Nielsen et

29 al., 1993). Additional methods simply measure the capacity of an antioxidant to scavenge free radicals, such as in the case of the DPPH assay (2,2-diphenyl-1- picrylhydrazyl) (Re et al., 1999). DPPH is often used for its simplicity in analyzing

antioxidant capacity (Huang et al., 2005). The reaction involves the ability of a sample to

donate electrons (or hydrogen) in order to quench DPPH radicals. The overall

mechanism imitates the ability of an antioxidant to inhibit free radical chain reactions via

hydrogen atom transfer. As free radicals become quenched (by the presence of an

antioxidant compound), the DPPH solution changes from a puruple to yellow color. The

reduction in absorbance at 520nm is a direct indication of quenching capacity (Brand-

Williams et al., 1995; Koleva et al., 2002). This method was used to successfully

compare the antioxidant capacity of crude peanut skin extracts to that of BHA (a common

synthetic preservative). The study by Wang et al., (2007) revealed a significant increase

in the scavenging activity of peanut skin extract over BHA, and indicated that

significantly lower levels of extract were required (versus BHA) to scavenge 50% of the

radicals in the reaction mixture (known as the EC50 Value). Another study by Yu et al.

(2006) utilized the DPPH method to show significant differences in antioxidant activity

between direct peeled, blanched, and roasted peanut skin extracts.

Total Phenols Method

The assay known as the total phenolic method was developed by Singleton and

Rossi (1965), which is a colorimetric technique utilizing a chemical reduction of the

Folin-Ciocalteau reagent. The procedure involves the addition of the Folin-Ciocalteau reagent for a period of two hours in a dark environment. Metal oxides contained within

30 the Folin-Ciocalteau reagent are reduced by phenols, which then absorb light at 765nm.

The light intensity at this given spectrum is directly proportional to the phenolic concentration. A standard curve must be developed for each procedure, and is accomplished by plotting the concentration of a phenolic standard (such as or catechin) versus the absorption intensity (Stevanato et al., 2004). The resulting concentration of phenolics is calculated as equivalents of a sample using the given standard curve, such as gallic acid equivalents (GAE) or catechin equivalents (CE). The most commonly used reference compound is gallic acid, based on the fact that it is highly pure, inexpensive, and often found in small amounts in plant materials (Wrolstad et al.,

2005). In addition, gallic acid is highly stable, and the standard solution will only lose approximately 5% potency over a two week period (Waterhouse, 2006).

Analytical Methods for Separation and Identification

Many analytical difficulties exist in the identification of polyphenols in foods.

According to Bravo (1998), information contained in literature regarding the composition and content of polyphenols in plants are not only incomplete, but also can be contradictory and difficult to compare. For this reason, a more standardized way of analyzing these compounds is needed. Few attempts have been made to create a systematic screening method for the identification of all types of flavonoids, glycosylated flavonoids, and other phenolic compounds from plant materials (Lin and Harnly, 2007).

The isolation and identification of phenolic-based plant extracts typically incorporates some form of chromatography. The term chromatography refers to a process in which chemical mixtures are separated into different components based on

31 variations among solutes (Anderson and Markham, 2006). The separation of individual components within a mixture of various substances is the main goal of utilizing this analytical technique. Compounds in a mixture are separated based on factors such as their affinity for a specific medium (or stationary phase), molecular weight, or solubility in a solvent. One commonly used method of identification is retention separation, in which retention is the measure of distance a substance moves (via mobile phase) in a chromatographic system between the time the sample is administered and when it is detected after traveling through a stationary phase (Nielsen, 1993). The point in which a compound is detected is known as its retention time. Compound retention may differ between systems based on environmental factors (such as temperature), and the utilized instruments (and their specific settings). Other variables include samples differences, and variations in mobile and stationary phases. For this reason, experiments using standard compounds must be conducted under identical experimental conditions in order to compare and validate results (Fennema, 1996; Nielsen, 1993).

Prior to analysis, many researchers utilize some form of preparatory technique before incorporating analytical methods. In addition to extraction methods, preparatory techniques should remove possible contaminants from the sample, which can reduce interference and thereby increase selectivity of the analytical method. Another aim of using preparatory methods includes converting the analyte into more suitable forms for detection and separation (Smith, 2003). In addition to typical sample preparation methods, such as filtration and liquid-liquid extraction, newer techniques are now incorporated to achieve specific results (Tura and Robards, 2002). For example, solid- phase extraction (SPE) is a fast and economical way of fractioning flavonoid-glycosides

32 and phenolic compounds prior to high performance liquid chromatography (HPLC)

analysis (Chen et al., 2001). Another method which has seen increasing use in recent

years for the separation of phenolic-based compounds into fractions (prior to HPLC

analysis) is size-exclusion chromatography (SEC). Size-exclusion chromatography

separates compounds based on size (or molecular weight). Utilization of this method

(prior to analytical testing) results in better separation of compounds when using further

HPLC analysis, and prevents the masking of compounds when using mass spectrometry

techniques. Studies by Fitzpatrick et al. (2000) successfully utilized SEC (Toyopearl

HW-40) to prepare grape-seed extracts for HPLC-MSn analysis. In addition, Lou et al.

(1999) used Toyopearl HW-40 SEC to separate phenolic compounds in peanut skin

extracts, allowing for the further separation and identification of A-type

proanthocyanidins.

The analysis of phenolics and polyphenolic flavonoids often include the

incorporation of some form of gas liquid chromatography (GLC) or HPLC (Musingo et

al., 2001). In general, gas chromatography (GC) techniques offer better sensitivity of that

of HPLC for the analysis of food components. However, analysis of phenolic-based

compounds using GC is impractical, since this method primarily analyzes volatile compounds of lower molecular weight. Therefore, HPLC techniques using various detection methods are commonly used. HPLC is a type of liquid chromatography that separates compounds dissolved in solvent solutions. An HPLC system consists of a mobile phase reservoir, a pump (typically a dual-reciprocating pump), an injector port, a column which contains a chromatographic stationary phase for separation of compounds, and a detector. Mobile phase compositions are typically altered during analysis in the

33 form of gradient elution, which changes the solvation power or polarity to allow for

better separation of analytes (Anderson and Markham, 2006). This allows for the

simultaneous detection of different types of phenolic-based compounds in a single run,

such as hydrophobic flavonoid aglycones and hydrophilic glycosides. Solvents and

modifiers are chosen based on the type of stationary phase, analyte, and detection

method.

The use of HPLC has been the dominant method for the separation and

characterization of phenolic-based compounds in the last several decades (Stalikas,

2007). For analytical HPLC analysis (qualitative and/or quantitative) of phenolic-based

compounds such as flavonoids, the stationary phase, solvent system, and gradient

program must first be optimized for maximum resolution/separation (Anderson and

Markham, 2006). When properly configured, HPLC allows concurrent separation and

analysis of phenolic-based compounds, along with their derivatives and degradation

products. In addition, HPLC allows for the detection of analytes in low-concentration,

even in the presence of interfering and coeluting structures.

One of the key advantages of HPLC for the analysis of phenolics is the diverse

amount of columns (i.e., stationary phases) and detection methods. Separation typically

takes place on various forms of extremely non-polar octadecylsilyl bonded (ODS) C18 reverse-phase columns. No single HPLC method exists that can properly elute all flavonoid-based compounds without difficulties related to co-elution or interference. In most cases, flavonoid-glycosides elute before aglycones on C18 columns. Flavonoids

with higher numbers of hydroxyl groups will elute before less substituted derivatives

(Crozier et al., 1997). Flavone C-glycosides generally elute at shorter retention times that

34 their O-glycosidic counterparts. Flavanones elute before flavones, due to the level of

unsaturation between carbons 2 and 3 on the flavonoid C-ring. The use of acetonitrile- water or methanol-water mixtures (along with acid modifiers) is common for mobile- phase solvent systems. These solvents are compatible with ultraviolet-visible light (UV- vis) detectors, which typically operate at one of the optimum absorption bands utilized for the identification of flavonoid compounds. Band I operates at a maximum absorption range of 300 to 550nm (due to the substitution pattern and conjugation of the C-ring), while Band II has a maximum absorption range of 240 to 285nm (due to the flavonoid A- ring) (Stalikas, 2007; Wrolstad et al., 2005). To solve this problem, the use of photodiode arrays are commonly used in the detection of phenolic-based compounds, since they allow for the simultaneous detection at multiple wavelengths. However, the most commonly used wavelength for routine detection is 280nm, which represents a compromise between the two band ranges (Anderson and Markham, 2006).

In addition to HPLC-UV techniques, one of the most common methods available for the analysis of phenolic-based compounds is high performance liquid chromatography mass spectrometry (HPLC-MS). This technique combines the sample separation characteristics of HPLC with the ability to analyze the mass of analytes contained within a sample mixture. The use of a mass spectrometer enables the identification of structures which have been separated by an HPLC technique. This analytical technique allows for the measurement of compounds in the form of mass-to-charge ratios of analyte ions. The instrument generates a plot of intensity versus mass-to-charge ratio, and results are reported as a mass spectrum. A compound’s mass spectrum is based on the distribution of fragment ions generated from the dissociation of a given analyte (Harborne and

35 Williams, 2000). The detection, interpretation, and identification of compounds is based

on the theory that different structures often have unique molecular weights, and in most cases, dissimilar dissociation patterns. Understanding the dissociation pathways of different types of compounds allows researchers to properly interpret the various fragment ions in a mass spectrum, allowing for the identification of a given structure. In the case of flavonoid-based compounds, the dissociation scheme that occurs in a mass spectrometer is known as the Retro-Diels Alder (RDA) ring opening mechanism. An example of the fragmentation pattern as a result of the RDA mechanism can be seen below for catechin (Figures 3 and 4). Since all flavonoids follow this dissociation scheme, understanding this pathway allows for the possible MS interpretation of various flavonoid classes. Oligomeric flavonoids (such as procyanidins) have three characteristic fragmentation pathways, including Retro-Diels Alder, quinone methide (QM), and heterocyclic ring fission (HRF) cleavage (Gu et al., 2003).

Various types of mass spectrometric techniques can be applied to the analysis of phenolic-based compounds. One of the most powerful techniques for this type of analysis is multi-stage methods such as tandem mass spectrometry (MS/MS) or MSn, combined with collision-induced dissociation (CID) (Appledorn et al., 2009). These methods allow for more detailed structural information when analyzing flavonoid-based compounds, such as: a) the type of substituent attached to the flavonoid base-unit

(including carbohydrates such as mono- and polysaccharides), b) the aglycone base-unit, c) the stereochemistry of terminal monosaccharide units, d) interglycosidic linkages, e) locations of substituents on the aglycone, and f) the type of inter-flavonoid linkages (IFL) or inter-flavonoid bonds (IFB) (Cuyckens and Claeys, 2004). Although certain structural

36 characteristics can be deduced using LC-MS, the use of MS/MS allows for further structural identification when compared to single-step MS techniques. In many cases, even the use MSn (MS/MS/MS) cannot determine full stereochemistry, which is typically left to techniques such as nuclear magnetic resonance (NMR). However, the use of

MS/MS can be used to determine the prevalence of prior identified compounds, therefore minimizing the effort lost in their isolation. The use of multi-stage MS techniques also allows for the identification of labile compounds in solution, such as flavonoid acyl groups (Constant and Beecher, 1995).

The incorporated ionization source (entering the MS) and type of mass spectrometer can have a drastic effect on the identification of phenolic-based compounds.

Rauha et al. (2001) reported that the use of electrospray ionization (ESI) in negative mode offers the highest sensitivity for the analysis of flavonoid compounds. However, even though positive ion CID spectra is the most frequently used method for the structural analysis of flavonoids, the use of negative ion CID spectra is considered to be more difficult to interpret (Cuyckens and Claeys, 2004). The combination of ESI with an

ion trap analyzer is considered a powerful tool for the analysis of flavonoids, because it

offers benefits of both ESI and MSn (MS/MS/MS). Ion trap analyzers have the ability to

store ions, isolate specific mass-to-charge (m/z), induce dissociation of desired ions, and

then eject ions for mass analysis. The quadropolar field incorporated in the trap allows

for the simultaneous capture of a wide range of m/z values. This is of particular interest,

because it allows for the identification of multiple compounds that exhibit different

molecular weights. Another advantage of ESI-MSn (Ion Trap) for flavonoid-based

compounds include the fact that it is considered a soft ionization process, allowing

37 molecular ions to undergo little fragmentation before it reaches the trap. This allows for

the efficient identification of flavonoid-glycosides, due to their labile nature. First-order

mass spectra only provide limited structural information except for molecular mass.

However, the use of ESI-MSn incorporates MS/MS analysis with CID to effectively

fragment individual sugar compounds off the flavonoid aglycone. Further dissociation of

the aglycone not only allows for the idenfication of both the mono- or polysaccharide moieties, but also the flavonoid monomer itself (Anderson and Markham, 2006).

Patentability

Several patents exist regarding phenolic-based compounds from plant extracts, and their uses in relation to specific chemical and biological activities. Several of these

patents involve peanut skins, including the utilization of extracts to create a novel

adjuvant and/or immunomodulator that can aid in the preparation of immunogenic

compositions and vaccines. The invention also provides a method of stimulating

acquisition of protective immunity by administering peanut skin extracts prior to

vaccination (Mcdougald and Fuller, 2007). Another invention relates to food products, including confectionaries and pet foods, which incorporate polyphenols from nut skins

(such as peanuts and/or almonds) or cocoa for the purpose of inducing vasodilation

(Chevaux et al., 2008). Numerous patents exist regarding the chemical and biological

activity of cocoa extracts. For example, one invention involves the use of polyphenolic

compounds from cocoa (including monomers and oligomeric procyanidins) for bacterial

inhibition, a dietary supplement, and a possible pharmaceutical (Romanczyk, Jr. et al.,

2003). Another invention combines cocoa procyanidins with acetylsalicylic acid to form

38 an anti-platelet therapy (Schmitz, 2003), while another uses cocoa procyanidins alone to reduce postprandial oxidative stress and related pathologies (Schmitz and Romanczyk,

Jr., 2001). A final example incorporates flavonoids (including flavones and flavanones) for the protection of ascorbic acid and ascorbyl compounds from oxidation (Schonrock and Kruse, 2004). The existence of these patents reveal that opportunities still exist regarding the patentability of future discoveries related to phenolic-based compounds from plant extracts, and their functions regarding possible chemical, biological, nutraceutical, and pharmaceutical activities.

39

OH OH OH OH B HO O B HO O A A C OH OH OH OH OH OH 290.1 B HO O A OH

OH 138.0 152.1

OH OH 290.1 OH O C B B HO O HO O 246.1 A C B A A OH A OH OH

B

A/C HO O A C OH OH O

OH B O 180.0

206.1 OH

Figure 3: Retro Diels-Alder (RDA) Mechanism: Ring opening mechanism for catechin using negative mode ionization (Fiztpatrick, Fleming, O’Malley et al., 2000. Used with permisstion).

40

OH ( A ) 246.1 OH A O O HO O B HO O A A B B OH OH C 204.0 OH O C OH O HO O 230.1 O O 162.0 OH OH2 OH ( B ) 206.1 O A OH O O OH A O B OH OH B O 162.0

O

188.0 OH2 HO O

( C ) 180.0 O C A HO O OH 164.0 A C A O OH HO B AB OH

A C C OH 80.0 HO HO O O OH OH OH 152.0 136.0

Figure 4: Retro Diels-Alder (RDA) Mechanism: Ring opening mechanism for catechin using negative mode ionization (Fiztpatrick, Fleming, O’Malley et al., 2000. Used with permisstion).

41 References

Acquaah, G. (2005). Principles of Crop Production: Theory, Techniques, and Technology. New Jersey: Pearson Education, Inc.

Adsule R. N., Kadam S. S., and Salunkhe D. K. (1989). CRC Handbook of World Food Legumes: Nutritional Chemistry, Processing Technology, and Utilization. Vol II. Boca Raton: CRC Press, Inc.

Anderson, O. M. and Markham, K. R. (2006). Flavonoids: Chemistry, Biochemistry, and Applications. Boca Raton, FL: CRC Press.

Andrikopoulos, N. K., Kaliora, A. C.; Assimopoulou, A. N. and Papageorgiou, V. P. (2003). Inhibitory activity of minor polyphenolic and nonpolyphenolic constituents of oil against in vitro low density lipoprotein oxidation. J. Med. Food, 5, 1-7.

Appledoorn, M. M., Sanders, M., Vincken, J.-P., Cheynier, V., Guernevé, C. L., Hollman, P. C. H., and Gruppen, H. (2009). Efficient isolation of procyanidin A-type dimers from peanut skins and B-type dimers from grape seeds. Food Chem., 117, 713-720.

Atuahene, C. C., Donkoh, A., and Mensah-Afoakwa, K. (1989). Value of peanut skins as a dietary ingredient for broiler chickens. Br. Poultry Sci., 30, 289-293.

Ballard, T. S., Mallikarjunan, P., Zhou, K., O’Keefe, S. F. (2009). Optimizing the extraction of phenolic antioxidants from peanut skins using response surface methodology. J. Agric. Food Chem., 57, 3064-3072.

Boriss, H. and Kreith, M. (Februrary 2006). Commodity profile: peanut. Agricultural Marketing Resource Center. Agricultural Issues Center, University of California: http://aic.ucdavis.edu/profiles/Peanuts- 2006.pdf.

Brand-Williams, W., Cuveleir, M., and Berset, C. (1995). Use of a free radical method to evaluate antioxidant activity. Lebensm. Wiss. Technol. 28, 25-31.

Bravo, L. (1998). Polyphenols: Chemistry, dietary sources, metabolism and nutritional significance. Nutr. Rev., 56, 317-333.

Buford, R. (1988). Extending shelf life by using traditional phenolic antioxidants. Cereal Foods World, 33(2), 207-209.

Burns, J., Yokota, T., Ashihara, H., Lean, M. E. J., and Crozier, A. (2002). Plant foods and herbal sources of resveratrol. J. Agric. Food Chem., 50, 3337-3340.

Chang, S., Ostric-Matijasevic, B., Hsieh, O. A. L., and Huang, C. (1977). Natural antioxidants from rosemary and sage. J. Food Sci., 42, 1102-1106.

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. Chromatogr. A, 913(1-2), 387-395.

Chen, R. S., Wu, P. L., Chiou, R. Y. Y. (2002). Peanut roots as a source of resveratrol. J. Agric. Food Chem., 50, 1665-1667.

Chevaux, K. A., Schmitz, H. H., and Romanczyk, L. J. (2008). Nut skin products. US patent 20080063690: http://www.freepatentsonline.com/y2008/0063690.html

42 Chowdhury, A. R., Sharma, S., Mandal, S., Goswami, A., Mukhopadhyay, S., and Majumder, H. K. (2002). Luteolin, an emerging anti-cancer flavonoid, poisons eukaryotic DNA topoisomerase I. Biochem. J., 366, 653-661.

Chun, O. K., Kim, D. O., Smith, N., Schroeder, D., Han, J. T., and Lee, C. Y. (2005). Daily consumption of phenolics and total antioxidant capacity from fruit and vegetables in the American diets. J. Sci. Food Agric., 85, 1715-1724.

Chung, I-M., Park, M. R., Chun, J. C., and Yun, S. J. (2003). Resveratrol accumulation and resveratrol synthase gene expression in response to abiotic stress and hormones in peanut plants. Plant Sci., 164, 103- 109.

Constant, H. L. and Beecher, C. W. W. (1995). A method for the dereplication of natural product extracts using electrospray HPLC/MS. Nat. Prod. Lett., 6, 193-196.

Crozier, A., Jensen, E., Lean, M. E., J., and McDonald, M. S. (1997). Quantitative analysis of flavonoids by reversed-phase high performance liquid chromatography. J. Chromatogr. A, 761(1-2), 315-321.

Cuyckens, F. and Claeys, M. (2004). Mass spectrometry in the structural analysis of flavanoids. J. Mass Spectrom., 39, 1-15.

Daigle, D. J., Ory, R. L., and Branch, W. D. (1985). The presence of 5,7-hihydroxyisoflavone in peanuts by high performance liquid chromatography analysis. Peanut Sci., 12, 60-62.

Dakora, F. and Phillips, D. (2002). Root exudates as mediators of mineral acquisition in low nutrient environments. Plants Soil, 245, 35-45.

Das, D. K., Sato, M., Ray, P. S., Maulik, G., Engelman, R. M., Bertelli, A. A., Bertelli, A. (1999). Cardioprotection of red wine: role of polyphenolic antioxidants. Drugs Exp. Clin. Res., 25, 115-20.

Dean, L. L., Davis, J. P., Shofran, B. G., and Sanders, T. H. (2008). Phenolic profiles and antioxidant activity of extracts from peanut plant parts. The Open Natural Prod, 1, 1-6.

DiSanto, A., Mezzetti, A., Napoleone, E., Di Tommaso, R., Donati, M. B., De Gaetano, G., and Lorenzet, R. (2003). Resveratrol and quercetin down-regulate tissue factor expression by human stimulated vascular cells. Thrombosis Hemostasis, 1, 1089-1095.

Duh, P. D., Yeh, D. B., and Yen, G. C. (1992). Extraction and identification of an antioxidative component from peanut hulls. J. Agric. Food Chem., 69, 814-818.

Dziezak, J. D. (1986). Antioxidants: The ultimate answer to oxidation. Food Technol, 40(9), 94-101.

Edenharder, R., Keller, G., Platt, K. L., and Unger, K. K. (2001). Isolation and characterization of structurally novel antimutagenic flavonoids from spinach (Spinacia oleracea). J. Agric. Food Chem., 49, 2767-2773.

Escarpa, A. and Gonzalez, M. (2001). An overview of analytical chemistry of phenolic compounds in foods. Crit. Rev. Anal. Chem., 31(2), 57-137.

Fennema, O. (1996). Food Chemistry. New York: Marcel Dekker.

Fitzpatrick, D. F., Fleming, R. C., Bing, B., Maggi, D. A., and O'Malley, R. M. (2000). Isolation and characterization of endothelium-dependent vasorelaxing compounds from grape seeds. J. Agric. Food Chem., 48, 6384-6390.

43 Fitzpatrick, D. F., Hirshfield, S. L., and Coffey, R. G. (1993). Endothelium-dependent vasorelaxing activity of wine and other grape products. Am. J. Physiol., 265(pt 2), H774-778.

Fletcher, S. M. (2000). Peanuts. In D. Colyer, P. L. Kennedy, W. A. Amponsah, S. M. Fletcher, and C. M. Jolly (Eds.), Competition in agriculture: the United States in the world market. New York: The Haworth Press, Inc.

Francisco, M. L. D. L. and Resurreccion, A. V. A. 2008. Functional components in peanuts. Crit. Rev. Food Sci. Nutr., 48(8), 715-746.

Frankel, E. N., Waterhouse, A. L., and Kinsella, J. E. (1993). Inhibition of human LDL oxidation by resveratrol. Lancet, 341, 1103-1104.

Frison, S., and Sporns, P. (2002). Variation in the flavonol glycoside composition of almond seedcoats as determined by MALDI-TOF mass spectrometry. J. Agric. Food Chem., 50, 6818-6822.

Golden, P. (Sept. 2001). Perennial profit. In Farm Progress, (pp. 12-13).

Goldberg, D. M., Karumanchiri, A., Tsang, E., and Soleas, G. J. (1998). Catechin and epicatechin concentrations in red wines: Regional and cultivar related differences. Am. J. Enol. Vitic., 49(1), 23-34.

Goldberg, D. M. and Soleas, G. J. (1999). Analysis of antioxidant wine polyphenols by high-performance liquid chromatography. Methods Enzymol., 299, 122-137.

Gorbet, D. W. and Knauft, D. A. (1997). SunOleic97R peanut. Improved oil chemistry runner type. Florida Agricultural Experimental Station, Institute of Food and Agricultural Sciences, University of Florida: http://edis.ifas.ufl.edu/

Griel, A. E., Eissenstat, B., Juturu, V., Hsieh, G., and Kris-Etherton, P. M. (2004). Improved diet quality with peanut consumption. J. Am. Coll. Nutr., 23, 660-668.

Gu, L., Kelm, M. A., Hammerston, J. F., Beecher, G., Holden, J., Haytowitz, D., and Prior, R. L. (2003). Screening of foods containing proanthocyanidins and their structural characterization using LC-MS/MS and thiolytic degradation. J. Agric. Food Chem, 51, 7513-7521.

Harborne, J. and Williams, C. (2000). Advances in flavonoid research since 1992. Phytochemistry, 55(6), 481-503.

Haslam, E. (1996). Natural polyphenols (vegetable tannins) as drugs: Possible modes of action. J. Nat. Prod., 59, 205-215.

Haslam, E. (1998). Plant Polyphenols, Vegetable Tannins Revisited. Cambridge, England: Cambridge University Press.

Havsteen, B. H. (2002). The biochemistry and medical significance of the flavonoids. Pharmacol. Ther., 96, 67-202.

Higgs, J. (2003). The beneficial role of peanuts in the diet – Part 2. Nutr Food Sci, 33, 56-64.

Hill, G. M. (2002). Peanut by-products fed to cattle. Vet. Clin. North Am. Food Anim. Pract., 18, 295-315.

Hollman, P. C. H. and Arts, I. C. W. (2000). Flavonols, flavones and flavanols – nature, occurrence and dietary burden. J. Sci. Food Agric., 80, 1081-1093.

Hopkins, W. and Huner, N. (2004). Introduction to Plant Physiology. Hoboken, NJ: Wiley & Sons, Inc.

44 Houghton, P. and Raman, A. (1998). Laboratory Handbook for the Fractionation of Natural Extracts. New York: Chapman and Hall.

Huang, D., Ou, B., Pryor, R. (2005). The chemistry behind antioxidant capacity assays. J. Agric. Food Chem., 53, 1841-1850.

Huang, M-T., and Ferraro, T. (1992). Phenolic compounds and cancer prevention. Chapter 2 in "Phenolic Compounds in Foods and their Effects on Health II. In M-T. Huang. C-T. Ho, and C.Y. Lee (Eds.), ACS symposium series 507. Washington, DC: American Chemical Society.

Huang, S. C., Yen, G-C., Chang, L-W., Yen, W-J., and Duh, P-D. (2003). Identification of an antioxidant, ethyl protocatechuate, in peanut seed testa. J. Agric. Food Chem., 51, 2380-2383.

Ingham, J. L. (1976). 3,5,4'-trihydroxystilbene as a phytoalexin from groundnut (Arachis hypogeae). Phytochemistry, 15, 1791-1793.

Iwashina T. (2000). The structure and distribution of the flavonoids in plants. J. Plant Res., 113, 287.

Jang, M., Cai, L., Udeani, G.O., Slowing, K. V., Thomas, C. F., Beecher, C. W., Fong, H. H., Farnsworth, N. R., Kinghorn, A. D., Mehta, R. G., Moon, R. C., and Pezzuto, J. M. (1997). Cancer chemopreventative activity of resveratrol, a natural product derived from grapes. Science, 275, 218-220.

Jeandel, P., Bessis, R., and Gautheron, B. (1991). The production of resveratrol (3,5,4’-trihydroxystilbene) by grape berries in different developmental stages. Am. J. Enol. Vitic., 42, 41-46.

Kampa, M., Hatzogolu, A., Notas, G., Damianaki, A., Bakogeorgou, E., Gemetzi, C., Kouroumalis, E., Martin, P. M., and Castanas, E. (2000). Wine antioxidant polyphenols inhibit the proliferation of human prostrate cancer lines. Nutr Cancer, 37(2), 223-233.

Karchesy, J. J. and Hemingway, R. W. (1986). Condensed tannins: (4β→8;2β →O→7)-linked procyanidins in Arachis hypogoea L. J. Agric. Food Chem., 34, 966-970.

Kim, S., Lee, M. J., Hong, J., Li, C., Smith, T. J., Yang, G. Y., Seril, D. N., and Yang, C. S. (2000). Plasma and tissue levels of catechins in rats and mice during chronic consumption of green tea polyphenols. Nutr Cancer, 37(1), 41-48.

Kim, S. Y., Jeong, S. M., Park, W. P., Nam, K. C., Ahn, D. U., and Lee, S. C. (2006). Effect of heating conditions of grape seeds on the antioxidant activity of grape seed extracts. Food Chem., 97, 472-479.

Kinsella, J. E., Frankel, E., German, B., and Kanner, J. (1993). Possible mechanisms for the protective role of antioxidants in wine and plant foods. Food Technol, 47, 85-89.

Koleva, I. I., van Beek, T. A., Linssen, J. P. H., de Groot, A., and Evstatieva, L. N. (2002). Screening of plant extracts for antioxidant activity: A comparative study on three testing methods. Phytochem. Anal., 13, 8-17.

Kopp, P. (1998). Resveratrol , a found in red wine. A possible explanation for the conundrum of the ‘French paradox’? European Endocrinol, 138, 619-620.

Kris-Etherton, P. M., Hecker, K. D., Bonanome, A., Coval, S. M., Binkoski, A. E., Hilpert, K. F., Friel, A. E., and Etherton, T. D. (2002). Bioactive compounds in foods: Their role in the prevention of cardiovascular disease and cancers. Am. J. Med., 113(9B), 71S-88S.

Kris-Etherton, P. M., Yu-Poth, S., Sabate, J. Ratcliffe, H. E., Zhao, G. and Etherton, T. D. (1999). Nuts and their bioactive constituents: Effects on serum lipids and other factors that affect disease risk. Am. J. Clin. Nutr., 70(suppl), 504S-511S.

45

Kris-Etherton, P. M., Zhao, G., Kinkoski, A. E., Coval, S. M., and Etherton, T. D. (2001). The effect of nuts on coronary heart disease risk. Nutr. Rev., 59(4), 103-111.

Lee, J. (1996). USDA scientists find treasure in peanut skins. Agricultural Research Service News, December 12, 1996: http://www.ars.usda.gov/is/pr/1996/skins1296.htm.

Lee, K., Kim, Y., Lee, H., and Lee, C. (2003). Cocoa has more phenolic phytochemicals and higher antioxidant capacity than teas and red wine. J. Agric. Food Chem., 51(25), 7292 -7294.

Louis, R. (1999). Masquelier’s OPC aids heart and immune system. Well Being Journal, 8, 1-3.

Li, B., Qingzan, Z., and Chen, C. (1995). Study on extraction of natural antioxidative components from peanut hulls. Xian dai hua gong (Modern Chemical Industry), 10, 31-33.

Lin, L.-Z. and Harnly, J. M. (2007). A screening method for the identification of glycosylated flavonoids and other phenolic compounds using a standard analytical approach for all plant materials. J. Agric. Food Chem., 55, 1084-1096.

Lin, Y. L., Shiao, M. S., Kuo, Y. H., and Tsai, W. J. (1999). Antioxidative principles from peanut hulls. Chinese Pharmaceut (Taipei), 51, 397-401.

Liu, C., Wen, Y., Chiou, J., Wang, K., Chiou, R. (2003). Comparative characterization of peanuts grown by aquatic floating cultivation and field cultivation for seed and resveratrol production. J. Agric. Food Chem., 51, 1582–1585.

Lou, H., Yamazaki, Y., Saski, T., Uchida, M., Tanaka, H. and Oka, S. (1999). A-type proanthocyanidins from peanut skins. Phytochemistry, 51, 297-308.

Lou, H., Yuan, H., Ma, B., Ren, D., Ji, M., and Oka, S. (2004). Polyphenols from peanut skins and their free radical-scavenging effects. Phytochemistry, 65, 2391-2399.

Lou, H., Yuan, H., Yamazaki, Y., Sasaki, T., and Oka, S. (2001). Alkaloids and flavanoids from peanut skins. Planta Med., 67, 345-349.

Macheix, J. and Fleuriet, A. (1998). Flavonoids in Health and Disease. New York: Marcel Dekker, Inc.

Manach, C., Williamson, G., Morand, C., Scalbert, A., and Remesy, C. (2005). Bioavailability and bioefficiency of polyphenols in humans. I. Review of 97 bioavailability studies. Am. J. Clin. Nutr., 81(suppl), 230S-242S.

McCord, J. M. (1994). Free radicals and prooxidants in health and nutrition. Food Technol, 48(5), 106- 109.

Mcdougald, L. R. and Fuller, A. L. (2007). Novel peanut skin extract as a vaccine adjuvant. European patent EP20050733210: http://www.freepatentsonline.com/EP1742656.html.

Milbury, P. E., Chen, C.-Y., Dolnikowski, G. G., and Blumberg, J. B. (2006). Determination of flavonoids and phenolics and their distribution in almonds. J. Agric. Food Chem., 54, 5027-5033.

Moure, A., Cruz, J. M., Franco, D., Dominguez, J. M. Sineiro, J., Dominguez, H., Nunez, M. J., and Parajo, J. C. (2001). Natural antioxidants from residual sources. Food Chem., 72, 145-171.

Musingo, M. N., Sims, C. A., Bates, R. P., O'Keefe, S.F. and Lamikanra, O. (2001). Changes in ellagic acid and other phenols in muscadine grape (Vitus rotundifolia) juices and wines during storage. Am. J. Enol.Vitic., 52(2),109-114.

46

Nepote, V., Grosso, N. R., and Guzman, C. A. (2000). Antioxidant activity of methanolic extracts from peanut skins. Molecules, 5, 487-488.

Nepote, V., Grosso, N. R., and Guzman, C. A. (2002). Extraction of antioxidant compounds from peanut skins. Grasas y Aceites, 53, 391-395.

Nepote, V., Grosso, N. R., and Guzman, C. A. (2005). Optimization of extraction of phenolic antioxidants from peanut skins. J. Sci. Food Agric., 85(1), 33-38.

Nepote, V., Mestrallet, M. G., and Grosso, N. R. (2004). Natural antioxidant effect from peanut skins in honey-roasted peanuts. J. Food Sci., 69(7), S295-S300.

Nielsen, S. S. (1993). Introduction to the Chemical Analysis of Foods. London, England: Jones and Bartlett Publishers.

Nijveldt, R., Nood, E., Hoorn, D., Boelens, P., Norren, K., and Leeuwen, P. (2001). Flavonoids: Review of probable mechanisms of action and potential applications. Amer. J. Clin. Nutr., 74, 418-424.

Novus Research. (2009). Phenolic compounds in plants. Research on Phytochemicals: http://www.organicashitaba.com/pc.html.

O'Byrne, D. J., Devaraj, S., Gundy, S. M. and Jialal, I. (2002). Comparison of the antioxidant effects of concord grape juice flavonoids and α-tocopherol on markers of oxidative stress in healthy adults. Am. J. Clin. Nutr., 76, 1367-1374.

Pace-Asciak, C. R., Hahn, S., Diamandis, E. P., Soleas, G., and Goldberg, D. M. (1995). The red wine phenolics trans-resveratrol and quercetin block human platelet aggregation and eicosanoid synthesis: Implications for protection against coronary heart disease. Clin. Chim Acta, 235, 207-209.

Percival, M. (1997). Phytonutrients and detoxification. Clinical Nutrition Insights, 5(2), 1-4.

Peschel, W., Sanchez-Rabaneda, F., Diekmann, W., Plescher, A., Gartzia, I., Jimenez, D., Lamuela- Raventos, R., Buxaderas, S., and Codina, C. (2006). An industrial approach in the search of natural antioxidants from vegetable and fruit wastes. Food Chem, 97, 137-150.

Pietta, P. (2000). Flavonoids as antioxidants. J. Nat. Prod., 62, 1035-1041.

Pignatelli, P., Pulcinelli, F. M., Celestini, A., Lenti, L. Chiselli, A., Gazzaniga, P. P. And Violi, F. (2000). The flavonoids quercetin and catechin synergistically inhibit platelet function by antagonizing the intracellular production of hydrogen peroxide. Am. J. Clin. Nutr., 72, 1150-1155.

Pratt, D. E. and Miller, E. E. (1984). A flavonoid antioxidant in Spanish peanuts (Arachis hypogoea). J. Agric. Food Chem., 61, 1064-1067.

Rasmussen, S. E., Frederiksen, H., Krogholm, K. S., and Poulsen, L. (2005). Dietary proanthocyanidins: Occurrence, dietary intake, bioavailability, and protection against cardiovascular disease. Mol. Nutr. Food Res., 49, 159-174.

Rauha, J. P., Vuorela, H., and Kostiainen, R. (2001). Effect of eluent on the ionization efficiency of flavonoids by ion spray, atmospheric pressure chemical ionization, and atmospheric pressure photoionization mass spectrometry. J. Mass Spectrom., 36, 1269-1280.

Re, R., Pellegrini, N., Protegente, A., Pannala, A., Yang, M., and Rice-Evans, C. (1999). Antioxidant activity applying and improved ABTS radical cation decolorization assay. Free Rad. Biol. Med., 26, 1231- 1236.

47

Reische, D. W., Lillard, D. A., and Eitenmiller, R. R. (2002). Antioxidants. In C. C. Akoh and D. B. Min (Eds.) Food Lipids, Ch 15. (pp. 489-516). New York: Marcel Dekker.

Rice-Evans, C. A., Miller, N. J., and Paganga, G. (1997). Antioxidant properties of phenolic compounds. Trends Plant Sci, 2(4), 152-159.

Romanczyk, Jr., L. J., Hammerstone, Jr., J. F., Buck, M. M., Post, L. S., Cipolla, G. G., McClelland, C. A., Mundt, J. A., Schmitz, H. H. (2003). Cocoa extract compounds and methods for making and using the same. US patent 6297273: http://www.freepatentsonline.com/6670390.html.

Sanders, T.H., McMichael, Jr., R.W., and Hexdrix, K.W. (2000). Occurrence of resveratrol in edible peanuts. J. Agric. Food Chem., 48, 1243-1246.

Sang, S., Lapsley, K., Jeong, W. S., Lachance, P. A., Ho, C. T., and Rosen, R. T. (2002). Antioxidative phenolic compounds isolated from almond skins (Prunus amygdalus Batsch). J. Agric. Food Chem., 50, 2459-2463.

Schmitz, H. H. (2003). Use of cocoa procyanidins combined with acelylsalicyclic acid as an anti-platelet therapy. US patent 6524630: http://www.freepatentsonline.com/6524630.html.

Schmitz, H. H. and Romanczyk, Jr., L. J. (2001). Method for reducing postprandial oxidative stress using cocoa procyanidins. US patent 6207702: http://www.freepatentsonline.com/6207702.html.

Schonrock, U. and Kruse, I. (2004). Use of flavones, flavanones and flavonoids for protecting ascorbic acid and/or ascorbyl compounds from oxidation. US patent 20040109882: http://www.freepatentsonline.com/EP0945128.html

Shahidi, F. and Naczk, M. (1996). Food Phenolics, Sources, Chemistry, Effects, Applications. Lancaster, PA: Technomic Publishing Co.

Shahidi, F., Wanasundara, U. N., He, Y., and Shukla, V. K. S. (1997). Marine lipids and their stabilization with green tea and catechins. In Flavor and Lipid Chemistry of Seafoods, (pp. 186-197). Washington DC: American Chemical Society.

Shephard, A. J., and Rudolf, T. S. (1991). Analysis of peanut products for total lipids, fatty acids and proximates. Peanut Sci., 18, 51-55.

Shi, J. Nawaz, H., Pohorly, J., Mittal, G., Kakuda, Y., and Jiang, Y. (2005). Extraction of polyphenolics from plant material for function foods – Engineering and technology. Food Rev. Inter., 21, 139-166.

Siemann, E. H. and Creasy, L. L. (1992). Concentration of the phytoalexin resveratrol in wine. Am. J. Enol. Vitic., 43, 49-52.

Silva, G. L., Lee, I. K., and Kinghorn, A. D. (1998). Special problems with the extraction of plants. In R. J. P. Cannell (Ed.), Natural Products Isolation: Methods in Biotechnology 4, (pp. 343-347). Totowa, NJ: Humana Press.

Singelton, V. L. and Rossi, J. A. (1965). Colormetry of total phenolics with phosphotunstic acid reagents. Am. J. Enol. Vitic., 16, 144-158.

Smith, R. M. (2003). Before the injection-modern methods of sample preparation for separation techniques. J. Chromatogr. A, 1000(1-2), 3-27.

Sobolev, V. S., and Cole, R. J. (1999). trans-Resveratrol content in commercial peanuts and peanut products. J. Agric. Food Chem., 47, 1435-1439.

48

Stalikas, C. D. (2007). Review: Extraction, separation and detection methods for phenolic acids and flavonoids. J. Sep. Sci., 30, 3268-3295.

Steinberg, D. (1993). Antioxidant vitamins and coronary heart disease. New England J. Med., 328(20), 1487-1489.

Stevanato, R., Fabris, S., and Momo, F. (2004). New enzymatic method for the determination of total phenolic content in tea and wine. J. Agric. Food Chem., 52, 6287-6293.

Tura, D., and Robards, K. (2002). Sample handling strategies for the determination of biophenols in food and plants. J. Chromatogr. A, 975, 71-93.

USDA, FAS. (June 2009). Production Estimates and Crop Assessment Division. Peanut area, yield, and production: http://www.usda.gov/nass/PUBS/TODAYRPT/crop0909.txt.

Vance, C. (2002). Root-cacteria interactions: Symbiotic N2 fixation. In Y. Waisel, A. Eshel, and U. Kafkafi, Plant Roots, the Hidden Half, (pp. 839-866). New York: Marcel Dekker, Inc.

Virginia Peanut Production Guide. (2009). Virginia Peanut Growers Association and Virginia Cooperative Extension, Virginia Tech: http://pubs.ext.vt.edu/2810/2810-1017/2810-1017.html.

Wang, J., Yuan, X., Jin, Z., Tian, Y., and Song, H. (2007). Free radical and reactive oxygen species scavenging activities of peanut skins extract. Food Chem., 104, 242-250.

Waterhouse, A. (2006). Folin-Ciocalteau micro method for total phenol in wine: http://waterhouse.ucdavis.edu/phenol/folinmicro.htm.

Wijeratne, S. S. K., Abou-Zaid, M. M., and Shahidi, F. (2006). Antioxidant polyphenols in almond and its coproducts. J. Agric. Food Chem., 54, 312-318.

Wrolstad, R., Acree, T., Decker, E., Penner, Reid, D., Schwartz, S., Shoemaker, C., Smith, D., and Sporns, P. (2005). Handbook of Food Analytical Chemistry. Hoboken, NY: John Wiley & Sons.

Wong, D. W. S. and Pavlath, A. E. (1992). Oxidation. In Y. H. Hui (Ed.), Encyclopedia of Food Science and Technology, (pp. 1965-1970). New York: Wiley and Sons, Inc.

Wu, J. M.,Wang, Z. R., Hsieh, T. C., Bruder, J. L., Zou, J.G., and Huang, Y. Z. (2001). Mechanism of cardioprotection by resveratrol, a phenolic antioxidant present in red wine (Review). Int. J. Mol. Med., 8, 3- 17.

Yang, C. S., Landau, J. M., Huang, M. T., and Newmark, H. L. (2001). Inhibition of carcinogenesis by dietary polyphenolic compounds. Ann. Rev. Nutr., 21, 381-406.

Yang, J., Liu, R. H., and Halim, L. (2009). Antioxidant and antiproliferative activities of common edible nut seeds. Lebensmittel – Wissenschaft and Technologie, 42(1), 1-8.

Yen, G. C., and Duh, P. D. (1995). Antioxidative activity of methanolic extracts of peanut hulls from various cultivars. JAOCS, 72(9), 1065-1067.

Yu, J., Ahmedna, M., and Goktepe, I. (2005). Effects of processing methods and extraction solvents on concentration and antioxidant activity of peanut skin phenolics. Food Chem., 90, 199-206.

Yu, J., Ahmedna, M., Goktepe, I., and Dai, J. (2006). Peanut skin procyanidins:Composition and antioxidant activities as affected by processing. J. Food Comp. Anal., 19, 364-371.

49 CHAPTER II: TOTAL PHENOL CONTENT AND ANTIOXIDANT ACTIVITY OF FRACTIONS OBTAINED FROM SIZE-EXCLUSION CHROMATOGRAPHY (SEC) OF RAW METHANOLIC PEANUT SKIN EXTRACT

ABSTRACT

The total phenol and antioxidant profile of methanolic extracts from the skins of Gregory peanuts were analyzed. The use of methanol as an extraction solvent yielded roughly 200mg extract per gram peanut skin. Toyopearl SEC of the raw methanolic peanut skin extract resulted in the separation of nine fractions based on different size classes. Analysis of total phenolics in these fractions using the Folin-Ciocalteau (FC) method revealed no significant differences among the various fractions (P<0.05). The average total phenol content for all fractions was 143.8 mg gallic acid equivalents per gram peanut skin (GAE/g). The DPPH radical scavenging method revealed compounds in each fraction with antioxidant activities significantly higher than that of BHA. No significant differences were found in the radical scavenging activity among the various fractions, which agrees with the similar concentrations of total phenols found in these samples. The levels of scavenging activities among all fractions correlated to the levels of phenolics found using the FC Total Phenols method. Significant differences (P<0.05) also were noted in the rate of increase of scavenging activity from the 0 to 100 µg/ml concentration range for each fraction. The percent scavenging activity of all fractions leveled-off at approximately a 200 µg/mL concentration of extract. No significant differences were noted in the efficient concentration values (EC50) among the different peanut skin fractions, but all were significantly higher than that of BHA, and required lower concentrations to achieve a 50% quenching of all radicals. The average estimated

EC50 value among all fractions was 28.6 µg/mL. The BHA treatment achieved an estimated EC50 value of 50.4 µg/mL, indicating that peanut skin fractions exhibited a 43% average increase in antioxidant activity over that of BHA. This indicates that the structures contained in each of these fractions can be utilized as a viable antioxidant source, and have the ability to replace synthetic antioxidants (such as BHA) with those that are naturally-derived from peanut skins.

50 INTRODUCTION

Prior studies have indicated that peanut skin extracts exhibit powerful oxyradical quenching activity, which is due to the high concentration of total phenolic compounds

(Lou et al., 2004; Yu et al., 2005). However, only a handful of phenolic-based compounds (such as phenolic acids, flavonoid monomers, proanthocyanidins, oligomeric procyanidins, etc.) have been identified in peanut skin extracts (Lou et al., 1999, 2001,

2004; Nepote et al., 2004; Yu et al. 2005). In addition, few studies have been conducted to determine which structures are responsible for the antioxidant activity associated with peanut skin extracts. Preliminary experiments using HPLC with ultraviolet (UV) detection on raw peanut skin extracts revealed this plant source contains large numbers of compounds, most of which exhibited co-elution based on the sheer number of structures

(Figure A.1). This would make identification of these compounds difficult, since qualitative analysis using HPLC-MS would only allow the recognition of the most intense peaks, causing less-intense compounds to be masked by more prominent components which elute nearby. This discovery indicated that in order to properly identify these compounds, the need for preparatory separations would be required to decrease co-elution during further HPLC-MSn analysis.

The use of a preparatory method such as size exclusion chromatography (SEC) would allow for the separation of phenolic classes based on their size, resulting in maximum resolution/separation during HPLC analysis. This method was successfully performed in a past study on flavonoid-based compounds in grape seed extracts

(Fitzpatrick et al., 2000), which incorporated a SEC column using Toyopearl HW-40S media. The integration of SEC in that study enabled the collection of phenolic classes in

51 the form of fractions. For this particular research, the incorporation of such a method

offers several other benefits. First, each of these fractions can be analyzed for their total

phenolic content, giving insight into the total concentration of different size-classes of

phenolic-based compounds in peanut skins. Second, the various fractions then can be

tested for antioxidant activity, which provides an indication of which phenolic classes are

contributing to the oxyradical scavenging effects exhibited by peanut skin extracts.

Finally, the separation of these classes (using a preparatory method such as SEC)

prepares fractions that can be further characterized for the proper identification of the full

phenolic profile contained within peanut skins. The utilization of this method for the

analysis of peanut skins has never been performed prior to this study. Further

identification of the phenolic-based compounds in each fraction (using HPLC-MSn) then will give an indication as to which compounds are responsible for a given total phenol content and antioxidant activity (as determined for each fraction). The purpose of this study is not only to prepare raw extracts for maximum detection of individual compounds during later HPLC studies, but also to give general conclusions regarding the concentrations of various classes of phenolic compounds and their associated oxyradical scavenging capabilities.

MATERIALS AND METHODS

Experimental Design

The basic experimental design for this study is illustrated in the figure below

(Figure 1). This diagram indicates the overall path of experiments conducted on peanut

skins (Gregory peanut line) during the course of this research. The materials and

52 methods for each experimental study is outlined below (except for (-)ESI-HPLC-MSn, which can be found in Chapter 3).

Figure 1: Experimental Design: Illustrating the basic experimental design for this sudy.

Extraction of Antioxidant Components from Peanut Skins

Blanched peanut skins (Gregory line) were acquired from Walter Mozingo of the

Tidewater Agricultural Research & Extension Center in Suffolk, VA (peanut skins were

obtained using a lab-scale blancher). Methanol was chosen as the extraction solvent

based on high % yields of extracted antioxidant components indicated in a prior study

(Duh et al., 1992). A total of 10g of peanut skins were ground into a fine powder, and

extracted using a 10:1 (v/w) ratio of methanol (100mL) to peanut skin (10g), stirring constantly for 24 hours. Extracts then were filtered using Whatman #1 filter paper, and the extraction was repeated a second time. Filtrates were centrifuged for 30 minutes to remove particulates. The combined filtrates then were evaporated to dryness using a rotary evaporator, and weighed to determine yield of soluble components for later

53 analysis of Folin-Ciocalteau (FC) Total Phenols and DPPH Oxygen Scavenging

Capacity. Rotary-evaporation was performed using almost no heat, so as not to cause damage/polymerization reactions to the phenolic compounds. The peanut skin extract was then re-dissolved in 10mL methanol for Toyopearl SEC.

Size-Exclusion Chromatography using Toyopearl HW-40S

A glass column measuring 55 x 2.5cm was packed with Toyopearl HW-40S Size-

Exclusion Chromatography (SEC) media. The mobile phase used was Fisher Scientific

HPLC-grade methanol, and the apparatus was fed using a Fisher Scientific Variable Flow

Mini-Pump (0.45mL/min flow rate). The SEC column was connected to a LKB Bromma

2238 UVicord SII Detector (280nm), which then fed into a LKB Bromma 2112 Redirac

Fraction Collector (tube change rate = 1 tube/min, 4.5mL/tube). An LKB Bromma 2210

2-Channel Recorder was used to record UV intensity, time, and tube numbers. All connections between sample-containing devices incorporated PTFE tubing. Peanut skin extract (10mL) was loaded onto the column, and allowed to run until UV intensity reached its original baseline of methanol alone (22 hours, 10 minutes). Tubes corresponding to each detected fraction then were combined and rotary evaporated to dryness, with volumes and weights recorded for calculation of FC Total Phenols and

DPPH Oxygen Scavenging Capacity. Individual fractions then were re-dissolved in

10mL HPLC-grade methanol for further analysis of Total Phenols, DPPH oxygen scavenging capacity, and (-)ESI-HPLC-MSn.

54 Total Phenolics (TP) using Folin-Ciocalteau (FC) Method

Concentrations of TP’s in each peanut skin fraction (A through H) were determined using the Folin-Ciocalteau Total Phenol method as described by Singleton and Rossi (1965). According to Engelhardt (2001), FC is the most efficient method for the analysis of total phenolics and high molecular weight tannins. It is a colorimetric technique which utilizes the Folin-Ciocalteau reagent (Sigma Chemical Corp.) to react with phenolic-based compounds, and accurately determine their concentration at 765nm.

Gallic acid was used as the standard, and results determined as mg/mL of gallic acid equivalents (GAE). Findings were converted to mg GAE per gram peanut skin, to give real-world correlation to the amount of phenolics in Gregory line peanut skins.

DPPH Radical Scavenging (Antioxidant) Activity

The antioxidant (scavenging) activity of peanut skin fractions was measured using the DPPH (1,1-diphenyl-2-picrylhydrazyl radical) free radical scavenging assay, and was carried out as described by Blois (1958). Methanol fractions (A through H) of peanut skins at various concentrations (0, 10, 20, 50, 100, 200, and 500 µg/mL) were added to a solution of 1.5 x 10-4M DPPH (Sigma, USA) in methanol. The mixture was shaken vigorously and allowed to react for 5 minutes. In this assay, the ability of antioxidant compounds to scavenge DPPH radicals is defined as the discoloration of the DPPH solution (relative decrease in absorption) after 5 minutes at 520nm. The radical scavenging activity was obtained from the following equation: Radical Scavenging

Activity (%) = {(Abscontrol – Abssample)/Abscontrol} x 100. The antioxidant activity is expressed as EC50 (efficient concentration value), which is defined as the concentration

55 (in µg/mL) of each extract fraction required to inhibit the formation of DPPH radical by

50%. DPPH radical scavenging activity was conducted for each peanut skin fraction at

various concentrations, and compared to an antioxidant/preservative commonly used in

food systems known as butylated hydroxyanisole (BHA).

Statistical Analysis

All experiments on individual peanut skin fractions (A through H) obtained from

Toyopearl SEC were performed in triplicate. Experimental results were expressed as

means ± standard deviations (SD). The values for EC50 were estimated by nonlinear

regression analysis using Statistica 8.0 (StatSoft, Tulsa, OK). Other data was analyzed

using ANOVA (P<0.05) with means separated by Duncan’s Multiple Range Test.

RESULTS AND DISCUSSION

In this study, 10g of peanut skins were extracted using 100% HPLC-grade

methanol, yielding 2.001g of dried extract. Several prior studies have been conducted

which investigated the yield of peanut skin extracts using various solvents. Huang et al.

(2003) indicated that polar solvents offered better extraction yields than less polar

solvents. Yu et al. (2005) reported that methanol and ethanol were more efficient than water for extracting phenolic-based compounds. Duh et al. (1992) revealed that methanol resulted in more than 50% higher yields of antioxidant extracts from peanut hulls when compared to ethanol (other solvents gave even lower yields). The efficient extraction yield of utilitizing methanol as an extraction solvent was also revealed in this study. The

use of methanol as an extraction solvent yielded roughly 200mg extract per gram peanut skin. Other studies which incorporated ethanol as an extraction solvent offered

56 approximately 50% (or more) reduction in extraction efficiency. Wang et al. (2007)

reported 0.107 ± 0.003g extract per gram peanut skin. Other studies that utilized ethanol

resulted in extracts of 0.051g (Huang et al., 2003) and 0.099g (Nepote et al. 2005) per gram peanut skin. The results in this study support the findings by Duh et al (1992), providing solid evidence that utilizing methanol as an extraction solvent can provide optimum yields of peanut skin extract.

Peanut skin extract obtained from the methanolic extraction process was run through a SEC column containing Toyopearl HW-40S. The sample run lasted a total of

22:10 (min), and was discontinued upon the return of a methanol-only baseline. Major

peaks/valleys indicated in the UV (280nm) chromatogram output were used to generate

individual fractions based on the size of different compound classes contained within the

peanut skin raw extract. The overall number of peaks (among the duration of the

Toyopearl sample run), along with their shapes, heights, and intensities, was similar to

that reported by Fitzpatrick et al. (2000) for the separation of grape seed extracts. A total

of nine fractions were obtained (Figure 2), which included fractions A, B, C, D, E, F, G-

Red, G, and H. Tubes obtained from the fraction collector were combined to form the

sample for each given fraction. Colors associated with each combined fraction were

initially clear (Fraction A), and changed from yellow (Fraction B and C), to orange (D

and E), to brownish-red (F through H). The major peaks and valleys which made up

Fraction G included numerous tubes which had an extreme bright-red color. These tubes corresponded to the first peak (in the series of peaks) that made up Fraction G. In order to further investigate the compound(s) responsible for this extreme change in visible

57 spectrum, these tubes were collected separately from the rest of Fraction G, and then combined to form Fraction G-Red.

Folin-Ciocalteu Total Phenols method was used to determine the concentration of phenolic-based compounds in each peanut skin fraction (A through H) obtained from

Toyopearl SEC. The data shown in Figure 3 reveal that peanut skins contain high levels of total phenolics (mg GAE/g peanut skin). Surprisingly, each fraction contained relatively the same amount of total phenolics, and no significant differences were seen when comparing the concentrations among the various fractions. This could indicate that there is an even distribution in concentration among different phenolic classes (based on size) in peanut skins. Yu et al. (2005) indicated that blanched peeling greatly reduced the concentration of total phenols in methanolic peanut skin extracts versus direct peeling or roast peeling (11.6 mg/g versus 90.1 mg/g and 96.7 mg/g respectively). The lower total phenolic content in that study was assumed to be due to the leaching of phenolics out of the skin (into the water) during the blanching process. However, even though the skins in this current research were obtained through a blanching process, the data still indicated much higher concentrations of phenolics in the peanut skin fractions of the Gregory line

(even when compared to the direct peeled or roast peeled treatments in the prior 2005 study). This could possibly be due to several factors, including the blanching process incorporated for that given study. Another possible explanation is differences in the phenolic profile in the peanut skins of the Gregory line versus other peanut lines/cultivars

(such as Runner varieties). This could indicate that the Gregory line has much higher concentrations of total phenols when compared to other varieties. A similar result was obtained in a prior study on the total phenols contained in peanut hulls. Yen and Duh

58 (1995) reported significant differences in the total phenolic concentration (including the

total luteolin content) in various peanut hulls obtained from different cultivars. This

shows that in addition to peanut hulls, other forms of peanut by-products (such as peanut

skins) can contain varying levels of phenolics based on differences in cultivars.

The antioxidant activities of peanut skin fractions (A though H) obtained from

Toyopearl SEC were examined using the DPPH free radical scavenging activity method.

Due to their ability to donate hydrogen (or electrons) to produce stable radical

intermediates, the phenolic-based structures found in many plants are viewed as powerful

antioxidant compounds (Scalbert et al., 2005). The stable nitrogen-centered free radical

found in 1,1-diphenyl-2-picrylhydrazyl (DPPH) enables it to be utilized for the relatively quick analysis of oxyradical quenching capabilities found in naturally-derived antioxidant compounds. The technique works via the associated absorbance reduction (at 520nm) observed when a proton-donating substance is added to a methanolic DPPH solution.

The result of this combination is a diamagnetic molecule, formed from the acceptance of an electron or hydrogen radical (Soares et al., 1997). This method was chosen based on these prior successes using the DPPH method to analyze the antioxidant capacity of various types of plant extracts. In this study, the DPPH radical scavenging results for each peanut skin fraction were compared to the scavenging activity of a synthetic antioxidant commonly used in the food industry, known as butylated hydroxyanisole

(BHA). The DPPH radical scavenging activity (%) for each fraction (including that of

BHA) can be seen in Figure 4. The various peanut skin fractions showed extremely high antioxidant activity, and were all significantly higher than that of BHA (at a P<0.05 significance level). No significant differences were found in the radical scavenging

59 activity among the various fractions, which agrees with the similar concentrations of total

phenols found in these samples. In fact, the levels of scavenging activities among the

different peanut skin fractions (Figure 4) appeared to have similar patterns to the levels of

phenolics found using the FC Total Phenols method (Figure 3). Significant differences

(P<0.05) also were noted in the rate of increase of scavenging activity from the 0 to 100

µg/ml concentration range for each fraction when compared to BHA. This was in

agreement with a previous study by Wang et al (2007). For each fraction, the percent

scavenging activity leveled-off at a concentration of approximately 200 µg/mL. The

estimated EC50 values indicate the concentration of antioxidant required to scavenge 50%

of all radicals in the DPPH reaction mixture, of which no significant differences were

noted among the different peanut skin fractions (A through H). However, the EC50 values for each fraction were significantly different than that of BHA, and required lower concentrations to achieve a 50% quenching of all radicals. The average estimated EC50 value among fractions A through H was 28.6 µg/mL, which reveals a slightly higher radical scavenging effect when compared to the peanut skins used in the 2007 study by J.

Wang et al (30.8 µg/mL). Again, this could be due to the differences in the phenolic profiles between the Gregory line and peanut from other cultivars. The BHA treatment achieved an estimated EC50 value of 50.4 µg/mL, indicating that peanut skin fractions

exhibited a 43% average increase in antioxidant activity over that of BHA.

CONCLUSION

The use of methanol proved to be an efficient solvent for the extraction of

phenolic-based antioxidant compounds from peanut skins, resulting in high extraction

60 yields when compared studies which used other forms of solvents (Huang et al., 2003;

Nepote et al., 2005; Wang et al., 2007). Methanol also could prove to be a cost effective method for the extraction of these compounds, since it could be easily recovered during evaporation, and reused for further extractions. Addition of methanolic extracts to a

Toyopearl SEC column resulted in the separation of phenolic compounds based on different size classes (fractions A through H). Analysis of total phenolics in these fractions using the FC method revealed an even distribution of different types of phenolic compounds relative to molecule size. Further antioxidant capacity studies, using the

DPPH radical scavenging activity method, revealed that compounds contained within each fraction exhibit antioxidant capabilities that far exceed that of BHA. This indicates that the structures contained in each of these fractions can be utilized as a viable

antioxidant source, and have the ability to replace synthetic antioxidants (such as BHA)

with those that are naturally-derived. Further research conducted in this particular study

aims to determine (via HPLC-MSn) which compounds are responsible for the given antioxidant effects exhibited by each fraction (see Chapter 3). In addition to the possible

applications in food systems, these compounds could have significant human health

implications, since their powerful oxy-radical scavenging effects could help reduce

oxidative-induced cell damage. Determining which compounds contribute to antioxidant

activity also could lead to the discovery of new compounds, those that may exhibit other

important pharmaceutical and nutraceutical attributes.

61

Figure 2. Size Exclusion Chromatography (SEC) of Raw Peanut Skin Extract Using Toyopearl HW-40S – SEC of peanut skin raw extract from Gregory peanut line. Figure represents time (t) from 0 to 22 hours (length of the sample run), versus UV intensity at 280nm. Peanut skin fractions obtained from SEC are indicated (A, B, C, D, E, F, G-Red, G, H), which were used for further analysis.

62

Figure 3. Total Phenolics using Folin-Ciocalteu Method: for Fractions A though H (obtained from Toyopearl Size Exclusion Chromatography) of raw peanut skin extract for Gregory peanut line. *Data represents the means of three replicates, and bars indicate ± standard deviations (n=3). Total phenolics concentration expressed as mg Gallic Acid Equivalent per gram peanut skin (GAE/g). Values with same letters are not significantly different at P<0.05.

63

Figure 4. DPPH Scavenging Capacity of Peanut Skin Fractions Obtained from Toyopearl SEC – Percent (%) scavenging Capacity of DPPH radical for Gregory line peanut skin extract (fractions A though H) versus BHA at various concentrations (10-500µg/mL).

64 REFERENCES

Blois, M. S. (1958). Antioxidant determinations by the use of stable free radical. Nature, 181, 1199-1200.

Engelhardt, U. (2001). Flavonoids-analysis. Crit. Rev. Food Sci. Nutr., 41, 398-399.

Fitzpatrick, D. F., Fleming, R. C., Bing, B., Maggi, D. A., and O’Malley, R. M. (2000). Isolation and characterization of endothelium-dependent vasorelaxing compounds from grape seeds. J. Agric. Food Chem., 48, 6384-6390.

Duh, P. D., Yeh, D. B., and Yen, G. C. (1992). Extraction and identification of an antioxidative component from peanut hulls. J. Am. Oil Chem. Soc., 69(8), 814-818.

Huang, S. C., Yen, G. C., Chang, L. W., Yen, W. J., and Duh, P. D. (2003). Identification of an antioxidant, ethyl protocatechuate, in peanut seed testa. J. Agric. Food Chem., 51, 2380-2383.

Lou, H., Yamazaki, Y., Saski, T., Uchida, M., Tanaka, H. and Oka, S. (1999). A-type proanthocyanidins from peanut skins. Phytochemistry, 51, 297-308.

Lou, H., Yuan, H., Ma, B., Ren, D., Ji, M., and Oka, S. (2004). Polyphenols from peanut skins and their free radical- scavenging effects. Phytochemistry, 65, 2391-2399.

Lou, H., Yuan, H., Yamazaki, Y., Sasaki, T., and Oka, S. (2001). Alkaloids and flavanoids from peanut skins. Planta Med., 67, 345-349.

Nepote, V., Gross, N. R., and Guzman, C. A. (2005). Optimization of extraction of phenolic antioxidants from peanut skins. J. Sci. Food Agric., 85, 33-38.

Nepote, V., Mestrallet, M. G., and Grosso, N. R. (2004). Natural antioxidant effect from peanut skins in honey- roasted peanuts. J. Food Sci., 69, S295-S300.

Scalbert, A., Manach, C., Morand, C., and Rémésy, C. (2005). Dietary polyphenols and the prevention of diseases. Crit. Rev. Food Sci. Nutr., 45, 287-306.

Singleton, V. L. and Rossi, J. A. (1965). Colorimetry of total phenolics with phosphotungstic acid reagents. Am. J. Enol. Vitic., 16, 2349-2351.

Soares, J. R., Dins, T. C., Cunha, A. P., and Ameida, L. M. (1997). Antioxidant activity of some extracts of Thymus zygis. Free Radical Res., 26, 469-478.

Wang, J., Yuan, X., Jin, Z., Tian, Y., and Song, H. (2007). Free radical and reactive oxygen species scavenging activities of peanut skins extract. Food Chem., 104, 242-250.

Yen, G. C. and Duh, P. D. (1995). Antioxidant activity of methanolic extracts of peanut hulls from various cultivars. J. Am. Oil Chem. Soc., 72, 1065-1067.

Yu, J., Ahmedna, M., and Goktepe, I. (2005). Effects of processing methods and extraction solvents on concentration and antioxidant activity of peanut skin phenolics. Food Chem., 90, 199-206.

65 CHAPTER III: IDENTIFICATION OF PHENOLIC COMPOUNDS IN PEANUT SKIN EXTRACT USING HPLC WITH NEGATIVE MODE ION TRAP MSn AND ELECTROSPRAY IONIZATION

ABSTRACT

The analysis of phenolic compounds often is performed using some form of high performance liquid chromatography (HPLC) coupled with mass spectrometry (MS) detection. However, the use of electrospray ionization (ESI) and ion trap MSn (MS/MS/MS) offers better detection over tradition first-order MS techniques. This study utilized negative mode (-)ESI-HPLC-MSn to identify phenolic-based compounds in nine fractions obtained from size-exclusion chromatography (SEC) of raw methanolic peanut skin extract. The study was designed to determine if previously undiscovered phenolic antioxidant compounds are present in peanut skins, and also to identify if structures exist that exhibit pharmaceutical/nutraceutical properties. Two separate HPLC method were developed to offer maximum separation and reduce co- elution. A total of 314 phenolic-based structures were identified or tentatively identified, the majority of which are newly discovered compounds that have never been previously detected in peanut skins. HPLC Method 1 identified 219 unique compounds, while HPLC Method 2 offered better separation (due to mobile phase and column differences), resulting in the detection of 95 additional structures. Most identified compounds were hydroxy- or methoxy- substituted phenolics. These structures included numerous phenolic acids, coumarins, chalcones, stilbenes, flavonoids, flavonoid-glycosides, oligomeric flavonoids (including A and B type procyanidin dimers, trimers, and tetramers), biflavonoids, and flavonoids with various types of attached side- groups. Many of the compounds identified are known to have powerful oxyradical scavenging effects, which agrees with previous studies conducted (on these fractions) which indicate high antioxidant activities. In addition, many of the detected compounds have been previously reported (in literature) to exhibit numerous beneficial chemical and biological activities, including the treatment of various human health-related conditions.

66 INTRODUCTION

The majority of prior studies on peanut skin extracts have involved the use of some form of High Performance Liquid Chromatography (HPLC) system, including a variety of detection methods such as UV-VIS and mass spectrometry (MS) (Lou et al., 1999, 2001; Yu et al., 2005,

2006). However, several issues exist when attempting to use these methods when analyzing compounds from these types of extracts. First, problems often arise from co-elution, where multiple compounds have the same (or similar) retention times. The utilization of UV-VIS

detection alone will not provide the necessary mode of qualitative analysis, since many phenolic-

based compounds will exhibit identical (or partially overlapping) HPLC retention times due to their polar natures. Second, the use of standard (non MSn) mass spectrometry techniques, such

as electron impact (EI) and time-of-flight (TOF), often will provide inconclusive results. This is

due to the fact that when using a (non-MSn) mass spectrometer, many phenolic and flavonoid-

based compounds that exhibit identical molecular weights will sometimes have indistinguishable

fragmentation patterns.

Many bioflavanoids exist as flavonoid-glycosides, which makes MS identification even

more difficult when using non-MSn methods. The use of standard MS techniques only offers a

first stage of fragmentation, resulting in the cleavage of just the flavonoid aglycone from the

sugar moiety (Cuyckens and Claeys 2004). The lack of further dissociation leaves only the m/z

of the flavonoid monomer, of which many can exist at any given molecular weight. For

example, several m/z 447 (M-H)- flavonoid-glycosides exist, including kaempferol-glucoside

and luteolin-glucoside derivatives. The utilization of a single-step MS technique will only remove the glucose sugar from the flavonoid aglycone. Upon cleavage, the remaining m/z 285

(M-H)- represents the flavonoid (aglycone) monomer, of which luteolin and kaempferol both

67 exhibit this same 286 amu molecular weight. Without further fragmentation of the flavonoid aglycone, the proper identification of this monomer would not be possible. In addition, many flavonoids that share the same molecular weight will sometimes follow similar dissociation schemes (in the initial fragmentation step), making identification difficult when using a standard

MS (non-MSn) method. The further fragmentation of identical molecular weight flavonoid compounds (exhibited by MSn instrumentation capable of MS/MS or MS/MS/MS) often will

show differences in fragmentation pathways, making positive identification possible.

As previously mentioned, the large number of compounds typically exhibited in plant extracts often causes the co-elution of multiple compounds during HPLC separation, making

identification impossible with single-step mass spectrometers. This can be circumvented by the

use of advanced MS instruments (such as Ion Trap MSn), since they may be tuned to

simultaneously identify multiple compounds at different molecular weight ranges. For these

reasons, the use of such MSn techniques is critical for the proper identification of phenolic-based

compounds in extremely flavonoid-rich extracts (such as in the case of peanut skins). This type

of method has never been performed in the analysis of peanut skin extracts, and could result in

the discovery of novel phenolic and flavonoid-based compounds.

One of the primary objectives of this study is to use HPLC-MSn (ESI-Ion Trap MS) for

the possible identification of compounds in peanut skin extracts that have never been previously

reported, since the use of MSn (MS/MS/MS) may allow for the detection of compounds that

could not be recognized using other analytical methods. The ultimate goal of utilizing HPLC-

ESI-MSn is not only to identify structures which have known antioxidant activity, but also

determine if compounds are present which exhibit specific biological activities from a

pharmaceutical or nutraceutical standpoint. This study also may show that compounds which

68 have been previously shown to exist in peanut skins are actually different structures altogether, since past researchers have only used raw molecular weights and/or inconclusive fragmentation data (obtained from non-MSn mass spectrometers) as a means for identification of these phenolic-based compounds.

MATERIALS AND METHODS

Preliminary experiments using HPLC-UV on raw peanut skin extracts (without any

preparatory separations, such as Toyopearl SEC) revealed this plant source contains a large

number of compounds (Figure A.1). This discovery indicated that in order to properly identify

these compounds, the need for multiple HPLC methods (along with a preparatory separation

technique, i.e., Toyopearl SEC) would be required to both maximize resolution/separation and

decrease co-elution. Therefore, two separate HPLC-MSn methods (designated as HPLC Method

1, and HPLC Method 2) were utilized for the identification of compounds in peanut skin

fractions (A through H) obtained from Toyopearl Size Exclusion Chromatography. For both

HPLC methods, a Thermo Finnigan LCQ HPLC-MSn (MS/MS) Ion Trap mass spectrometer

with electrospray ionization (ESI) was used. The instrument incorporated an Agilent (Palo Alto,

CA) 1100 series G1312A binary pump, with a Rheodyne 7125 manual injector (25 μL injection

loop). Separations for HPLC Method 1 were carried out using a Waters (Ireland) Nova-Pak C18

(3.9 x 150 mm; 60A; 4 µm) reverse phase column, plus a C18 guard column. The solvent

system comprised of two mobile phase parts: (A) 0.5% glacial acetic acid in water, and (B)

0.5% glacial acetic acid in methanol. The solvent gradient parameters were as follows: elution

rate of 0.8 mL/min, with %(B) = 0% (0 – 2 minutes), 4.75% (5 minutes), 30.88% (15 minutes),

74.7% (42 minutes), 95% (52 – 70 minutes), and 100% (80 – 100 minutes). UV detection was

69 conducted at 280nm for detection of bioflavanoids, using an Agilent 1100 G1314A UV/Vis detector (Applied Biosystems Model 785A Programmable Absorbance Detector). Samples then were analyzed using ESI-MSn in negative ionization mode, with the following parameters:

sheath gas (N2) = 60; auxiliary gas (N2) = 5; spray voltage 3.3kV; capillary temperature 250°C;

cap voltage = 15; tube lens offset = 0V. Base peak ranges used for (-)ESI-MSn consisted of two parameters, utilizing an initial scanning range of m/z 125-600 (M-H)- ions, with a source induced dissociation (SID) voltage of 1.0V (sid=1.0). The second scan range encompassed ions of m/z

590-2000 (M-H)-, with an SID voltage of 2.0V (sid=2.0). In the case of both ‘sid=1.0’ and

‘sid=2.0’ scan ranges, further MSn (i.e., isolation and dissociation) was dependent on the most

intense ion in each scan range. MSn first is defined by the detection of the most intense ion in

the normal mass spectra (ms spectra), of which MS/MS is then performed on this parent ion

(resulting in an ms2 sepctra). The most intense ion in the ms2 spectra then undergoes further

dissociation, with the resulting ion fragments shown in the ms3 spectra (MS/MS/MS). Data

dependent settings of the LCQ ion trap were as follows: MSn=2,3; 5u isolation window;

normalized collision energy 42.5% (CID); and activation Q (Mathieu q parameter) = 0.3. All

samples of peanut skin extracts, including fractions A through H obtained from Toyopearl SEC,

were continually stored at -80°C (from the point of SEC completion, to the time of HPLC injection). All sample runs including the standard mix, and peanut skin fractions A through H

(obtained from Toyopearl SEC), were injected using a 20µL sample volume.

HPLC Method 2 was conducted in an attempt to offer improved chromatographic separation and resolution in comparison to HPLC method 1. Separations for HPLC Method 2 were carried out using a Waters (Ireland) Atlantis dC18 (2.1 x 150 mm; 3 µm) column, plus a

2mm x 4mm Phenomenex C18 guard column. The solvent system comprised of two mobile

70 phase parts: (A) 0.3% glacial acetic acid in water, and (B) 0.3% alacial acetic acid in methanol.

The solvent gradient parameters were as follows: elution rate of 0.8 mL/min, with %(B) = 0% (0 minutes), 20% (7 minutes), 70% (70 minutes), and 95% (80-100 minutes). UV detection was conducted at 280nm for maximum detection of bioflavanoids, using an Agilent 1100 G1314A

UV/Vis detector (Applied Biosystems model 785A programmable absorbance detector).

Samples then were analyzed using ESI-MSn in negative ionization mode, with the following

parameters: sheath gas (N2) = 65; auxiliary gas (N2) = 3; spray voltage 3.3kV; capillary temperature 250°C; cap voltage = 15; tube lens offset = 0V. Base peak ranges used for (-)ESI-

MSn consisted of three parameters, utilizing an initial scanning range of m/z 125-250 (M-H)- ions, with an SID voltage of 1.0V (sid=1.0). The second scan range encompassed ions of m/z

230-410 (M-H)-, with an SID voltage of 2.0V (sid=2.0). The final scan range analyzed ions of

m/z 400-2000 (M-H)-, with an SID voltage of 3.0V (sid=3.0). In all three cases of ‘sid=1.0’,

‘sid=2.0’, and ‘sid=3.0’ scan ranges, further MSn (i.e., isolation and dissociation) was dependent

on the most intense ion in each scan range. As in the case of HPLC Method 1, MSn first is defined by the detection of the most intense ion in the normal mass spectra (ms spectra), of which MS/MS is then performed on this parent ion (resulting in an ms2 sepctra). The most intense ion in the ms2 spectra then undergoes further dissociation, with the resulting ion fragments shown in the ms3 spectra (MS/MS/MS). Data dependant settings of the LCQ ion trap were as follows: MSn=2,3; 5u isolation window; normalized collision energy 42.5% (CID); and

activation Q (Mathieu q parameter) = 0.3. All samples of peanut skin extracts, including

fractions A through H obtained from Toyopearl SEC, were continually stored at -80°C (from the

point of SEC completion, to the time of HPLC injection). All sample runs including the standard

mix, and peanut skin fractions A through H (obtained from Toyopearl SEC), were injected using

71 a 20µL sample volume.

A standard mix was prepared to aid in the identification of unknown compounds. These standard compounds included: p-coumaric acid (para-hydroxycinnamic acid), gallic acid (3,4,5- trihydroxybenzoic acid), caffeic acid, ferulic acid, flavone, resveratrol, kaempferol (3,4',5,7- tetrahydroxyflavone), luteolin (3',4',5,7-tetrahydroxyflavone), eriodictyol (3',4',5,7-tetrahydroxy- flavanone), catechin, epicatechin, hesperitin (3',5,7-Trihydroxy-4'-methoxyflavanone), quercetin

(3,3',4',5,7-pentahydroxyflavone), gallocatechin, epigallocatechin, quercetin dihydrate, chlorogenic acid, , catechin gallate, quercetrin (quercetin-3-O-rhamnoside), , , and procyanidin B3 dimer. The “standard mix”

UV chromatogram and normal mass spectra TIC (Total Ion Chromatogram – indicating the base peaks for the most intense ions) can be seen in Figure A.2 and A.3. Figure A.4 shows UV versus

TICs for the MSn (ms, ms2, and ms3) of individual standard compounds in the standard mix.

The top TIC in this figure shows the full MS of the most intense (parent) ions in the normal mass

spectra. The second TIC reveals the most intense ions formed in the ms2 spectra (MS/MS),

while the third TIC shows the most intense ions formed in the ms3 spectra (MS/MS/MS). The

bottom chromatogram shows the UV (280nm) that corresponds to the base peaks in the normal

mass spectra. An example showing the ion spectra for each TIC base peak (ms, ms2, and ms3) is

seen in Figure A.5, which indicates the MSn ion fragmentation pattern for the epicatechin

standard (at RT 16.53). The dissociation pathways which correspond to the ions in these spectra can be seen in Figure A.6. The standard mix for HPLC Method 2 can be seen in Figure A.39.

The MSn spectra for numerous additional standard compounds were provided by Jodie Johnson

from the University of Florida Mass Spectrometry Laboratory (Gainesville, FL). In the case of all compounds (including those in both the standard mix and peanut skin fractions from

72 Toyopearl SEC), discovery of the molecular ion for each structure was determined by the presence of adduct ions found in the normal mass spectra, which are commonly formed when conducting (-)ESI-MS. Based on the modifier used in the solvent system for this study, which allows for deprotonation to occur during negative mode ionization, these adducts include acetate

(59-60 amu) and sodium acetate (82 amu). Other adduct ions used for molecular ion identification included chlorine (35 or 37 amu), and ions produced by the formation of low- intensity dimers (of the molecular ion in question) during the electrospray process. Further identification of unknown structures was accomplished either by matching UV retention times

(with the aforementioned standards), or by using the ion fragmentation pathways of known standards to assist in the interpretation of unknown compounds. The interpretation of ion spectra also was accomplished by following the dissociation patterns that are common to phenolic acids and bioflavonoids, known as retro-Diels-Alder (RDA) reactions. In addition, several software packages were used to aid in the identification of unknown compounds, including Thermo-

Finnigan QualBrowser, the National Institute of Standards and Technology (NIST) 2005 Mass

Spectral Database (and included tools such as MS Interpreter, Isotope Calculator, and MS

Search), the Automated Mass Spectral Deconvolution and Identification System (AMDIS ver.

2.64), and MS fragmentation tools from ACD/Labs v12.0 and ChemBioDraw Ultra v11.0

(ChemBioOffice 2008). Other tools included web resources such as the NIST Chemistry

WebBook (online mass spectral database), PubChem, and ChemBioFinder.

73 RESULTS AND DISCUSSION

Fraction A - HPLC Method 1

A total of 53 compounds were identified in peanut skin Fraction A obtained from

Toyopearl SEC (Table 1). UV chromatograms and TICs for each scan range can be seen in

Figures A.7 through A.9. Compound 1-A1 at RT 1.69 (min) exhibited a high UV intensity and

ion relative abundance. An intense ion at m/z 401 (M-H)- was noted in the normal mass spectra,

- along with a sodium acetate [(M-H)+CH3COONa] adduct ion at m/z 483. Dissociation of the

m/z 401 ion yielded an intense m/z 341 ion in the ms2 spectra, which revealed the presence of an

- acetate [M+CH3COO] adduct (due to the 60 amu neutral loss). Further fragmentation of the m/z

341 ion produced the following fragments in the ms3 spectra: m/z 179 (the most intense ion),

161, 143, and 119. The m/z 179 ion indicated a neutral loss of 162 amu, which proves the

existence of a glucose (or galactose) monosaccharide side-group. The fragmentation scheme of the remaining 180 amu molecular weight compound follows the dissociation pathway of a dihydroxy-cinnamic acid. An example of such a compound is 3,4-dihydroxycinnamic acid (a.k.a

caffeic acid) (Figure 1). Therefore, compound 1-A1 is most likely a dihydroxycinnamic acid,

with a glucose (or galactose) side-unit.

Compound 2-A1 at RT 7.78 (min) exhibited an intermediate UV intensity, but a low

relative ion abundance. An intense m/z 443 (M-H)- ion in the normal mass spectra dissociated to

yield an m/z 383 ion in the ms2 spectra. The resulting 60 amu neutral loss showed the presence

- of an acetate [(M-H)+CH3COOH] adduct ion (confirming m/z 383 as the molecular ion).

Further dissociation of the m/z 383 ion produced an intense m/z 237 ion in the ms3 spectra, which indicates the presence of a terminal rhamnose sugar (due to the 146 amu neutral loss). An additional m/z 161 ion fragment was found in the ms3 ion spectra, revealing that the m/z 237

74 (M-H)- compound is most likely a monohydroxy-flavone or isoflavone. The fragmentation

pattern agrees with any possible flavone or isoflavone where the hydroxyl group resides on either

the A or C flavonoid ring. Possible flavones compounds (and their isoflavone derivatives)

include: 3-hydroxyflavone (a.k.a. flavonol), 5-hydroxyflavone (a.k.a. primuletin), 6-hydroxy-

flavone, and 7-hydroxyflavone (Figure 2). The likelihood of compound 2-A1 being a mono-

hydroxyflavone was also confirmed by matching ion spectra using the NIST 05 Mass Spectral

Database. Other possible explanations for the fragmentation pattern of compound 2-A1

(although less likely than the mono-hydroxyflavone scenario) is the presence of the following

methoxy-chalcone compounds: 4'-methoxychalcone (Figure 3) and 4-methoxychalcone.

Similarities to these chalcone compounds were also confirmed using the NIST 05 database.

Compound 3-A1 at RT 8.54 showed an intermediate UV intensity, but a low relative ion

abundance. The m/z 301 (M-H)- molecular ion dissociated to form the following fragments in

the ms2 and ms3 spectra: m/z 285, 217, 167, 150, and 126. This fragmentation scheme follows

the dissociation pathway for the compound known as 3',4',5,5',7-pentahydroxyflavone (a.k.a.

tricetin) (Figure 4).

Compound 4-A1 at RT 8.69 (min) exhibited an intermediate UV intensity, but a low

relative ion abundance. The m/z 379 (M-H)- molecular ion was confirmed by the presence of an

- acetate [M+CH3COO] adduct ion at m/z 439 (60 amu difference). Dissociation of the molecular

ion formed the following ions in the ms2 spectra: m/z 363, 361, 350, 336, 335 (the most intense

ion), 318/317, 291/290, 218/217, and 202. The predictive fragmentation patterns for the

compound known as 4,4'-dihydroxy-3,3'-(2-methoxyethylidene) dicoumarin (a.k.a. dicoumoxyl)

(Figure 5) nearly follows the exact dissociation pathway for compound 4-A1. A similar

75 dicoumarin compound with complimentary fragmentation ions was also found using the NIST 05 database (although, it had one less hydroxyl and methyl group), which supports this finding.

Compound 5-A1 at RT 8.84 (min) had a high UV response at 280nm, and gave an intense ion of m/z = 395 (M-H)- in the normal mass spectra. Further ms2 fragmentation yielded an m/z

- = 335 ion, showing a neutral loss of 60 amu, which is indicative of an acetate [M+CH3COO] adduct ion. Therefore, the molecular ion in this case is m/z = 335 (M-H)-. An additional intense

ion in the ms2 spectra included m/z 319. Further fragmentation of the m/z 335 ion gave ms3

product ions of m/z = 291, 273, 247, and 173. This follows the possible fragmentation pathway

of the compound known as 3,3'-methylenebis[4-hydroxycoumarin] (a.k.a. dicoumarin) (Figure

6). In addition, a NIST 05 database search of the ms2 spectra confirmed a high probability that

the compound is a dicoumarin. NIST 05 showed significant matches with the compound

coumarin-6-(7-hydroxycoumarin-8-yl)-7-methoxy, which further supports the likelihood that

compound 5-A1 is one of the aforementioned dicoumarin-based compounds.

Compound 6-A1 at RT 9.90 (min) gave both a low UV intensity and ion relative

abundance. An intense m/z 457 (M-H)- ion was found in the normal mass spectra. This ion

dissociated to form an m/z 397 ion in the ms2 spectra, indicating a neutral loss of 60 amu, which

- - signifies the presence of an acetate [(M-H)+CH3COOH] adduct. This confirms m/z 397 (M-H) as the molecular ion. The molecular ion dissociated to yield the following ms2 ion fragments: m/z 251 (most intense ion), 235, 163, 161, and 143. The m/z 251 ion indicates a neutral loss of

146 amu, revealing the presence of a rhamnose sugar side-group. Additional fragments in the ms3 spectra indicate that the compound is most likely 3-methoxyflavone (Figure 7). A possible match of the spectra for this compound was also confirmed using the NIST 05 database.

Therefore, compound 6-A1 is most likely a 3-methoxyflavone with a rhamnose sugar side-group.

76 Compound 7-A1 at RT 10.20 (min) exhibited both a low UV intensity and ion relative abundance. The molecular ion was confirmed at m/z 215 (M-H)- due to the presence of acetate

- - [M+CH3COO] and sodium acetate [(M-H)+CH3COONa] adduct ions at m/z 276 and 298

respectively. The molecular ion dissociated to yield m/z 198, 172, and 157. This follows the

possible fragmentation scheme for the compound (or a derivative of) 6-hydroxy-2-

naphthalenepropanoic acid (a.k.a. allenolic acid) (Figure 8). A similar compound, known as 2,3-

napthalenedicarboxylic acid, was identified via a NIST 05 Mass Spectral Database search.

Although compound 7-A1 is most likely not 6-hydroxy-2-naphthalenepropanoic acid (a.k.a.

allenolic acid) or 2,3-napthalenedicarboxylic acid (since they are synthetized compounds), it

does however show structural similarities to both these compounds based on the fragmentation

scheme. This resemblance yields dissociation pathways with parallel fragmentation

characteristics, which supports the theory that compound 7-A1 is possibly a derivative of

allenolic acid.

Compound 13-A1 at RT 14.07 (min) exhibited a low UV response, and relatively low ion

intensity. The most intense ion in the normal mass spectra (m/z 397 (M-H)-) dissociated to form

an intense m/z 337 (M-H)- ion in the ms2 spectra. The 60 amu difference indicates the presence

- of an acetate [(M-H)+CH3COOH] adduct, which confirms m/z 337 as the molecular ion. Other

major ion fragments in the ms2 spectra include m/z 321 and 251. The fragmentation pattern

somewhat follows the dissociation characteristics of a chromone-based compound, however, the

ion fragments are more indicative of a compound similar in structure to 4'-hydroxy-7-methoxy-

6-(3-methyl-2-butenyl)-flavanone (a.k.a. bavachinin A). The possible presence of this

compound (or a close derivative of this compound) was further supported by the NIST 05 MS

77 database, which matched a similar structure with nearly identical fragmentation characteristics

(an isoflavone derivative known as hydroxywighteone).

Compound 17-A1 at RT 17.01 gave an intermediate UV intensity, and low relative ion abundance. An m/z 167 (M-H)- molecular ion was seen in the normal mass spectra, although, no

adduct ions were present. The molecular ion dissociated to yield the following ms2 ion

fragments: m/z 152, 151, 137, 123, 121, 109, and 108. According to the NIST 05 database, this

follows the exact fragmentation fingerprint for the compound known as vanillic acid. In

addition, it shows extreme similarity to the predictive dissociation scheme for homogentisic acid

(Figure 9).

Compound 18-A1 at RT 17.42 (min) exhibited a low UV intensity, but a high ion relative

abundance. An intense m/z 475 (M-H)- ion in the normal mass spectra underwent collision- induced dissociation (CID) to yield an m/z 415 ion in the ms2 spectra. The 60 amu difference

- indicates the presence of an acetate [(M-H)+CH3COOH] adduct ion, which confirms m/z 415 as

the molecular ion. The molecular ion then dissociated to yield an ms3 ion of m/z 269, which

signifies the loss of a 146 amu rhamnose sugar. An additional m/z 161 fragment was seen in the

ms3 spectra, which indicates that the m/z 269 compound could be one of several structures.

Compounds which match this 270 amu molecular weight, while also exhibiting an m/z 161

fragment, include various dihydroxy-methoxy-chalcones, trihydroxy-flavones, trihydroxy-

isoflavones, and hydroxy-methoxyflavanones. An example of such a compound is 3',4',7-

trihydroxyflavone (Figure 10). Therefore, compound 18-A1 is most likely one of the

aforementioned classes of compounds, along with an attached rhamnose sugar group.

Compound 21-A1 at RT 19.04 (min) exhibited a low UV intensity, but an intermediate-

ranged ion relative abundance. A molecular ion was found at m/z 443 (M-H)-, although no

78 adduct ions were confirmed. The parent ion dissociated to yield an m/z 299 ion in the ms2 spectra. This indicated a 144 amu neutral loss, which reveals the possible presence of a coumaroyl side-group. Further fragmentation of the m/z 299 ms2 ion produced an m/z 137 ion in the ms3 spectra, which indicates the presence of a 162 amu glucose (or galactose) unit. The remaining m/z 137 ion is most likely a hydroxybenzoic acid derivative (such as 2- hydroxybenzoic acid, 3-hydroxybenzoic acid, or 4-hydroxybenzoic acid). Therefore, compound

21-A1 is most likely a glucose-linked hydroxybenzoic acid, with a coumaroyl acyl side-group.

Compound 22-A1 at RT 19.42 (min) revealed a low UV intensity, but an intermediate ion relative abundance. An intense m/z 595 (M-H)- was found in the normal mass spectra, which

dissociated to yield an m/z 535 ion in the ms2 spectra. The 60 amu difference indicated the

- presence of an acetate [(M-H)+CH3COOH] adduct ion, which confirmed m/z 535 as the molecular ion. Further fragmentation of the m/z 535 molecular ion produced the following daughter ions in the ms3 spectra: m/z 389 and 227. The m/z 389 indicated a neutral loss of 146

amu, indicative of a rhamnose sugar. Furthermore, the m/z 227 fragments yielded a neutral loss

of 308 amu, which reveals the presence of a rutinose (6-O-alpha-L-rhamnosyl-D-glucose)

disaccharide. The remaining m/z 227 ion fragment is most likely a resveratrol aglycone base-

unit. Therefore, compound 22-A1 is tentatively identified as resveratrol, with a rutinose sugar

side-group.

Compound 23-A1 at RT 19.94 (min) showed both an intermediate-to-high UV intensity

and relative ion abundance. The m/z 173 molecular ion (M-H)- was confirmed by the presence

- of acetate [(M-H)+CH3COOH] adduct ion at m/z 232 (59 amu difference). The molecular ion

dissociated to form ms2 ion fragments of m/z 155, 143, 142, 129, 111, 109, 83, and 81. This closely follows the predictive fragmentation pathway for shikimic acid (Figure 11). The

79 fragmentation fingerprint was also confirmed using the NIST 05 Mass Spectral Database. The presence of this compound at high concentrations makes sense, given that shikimic acid is the precursor to the formation of flavonoid structures.

Compound 24-A1 at RT 20.21 (min) exhibited a low UV response, but a high ion intensity. The molecular ion is assumed to be m/z 595 (M-H)-, although no adduct ions were

seen in the normal mass spectra. Dissociation of the molecular ion resulted in ms2 and ms3

spectra with the following ion fragments: m/z 579, 535, 487, 389, 367, 271, 263, 251, and 225.

Two compounds which follow this possible fragmentation scheme (and exhibited this 596 amu

molecular weight) include eriodictyol-7-neohesperidoside (a.k.a. ) and eriodictyol 7-

O-rutinoside (a.k.a. eriocitrin) (Figure 12).

Compound 25-A1 at RT 21.83 (min) gave both a low UV response and ion intensity. The

molecular ion was noted at m/z 579 (M-H)-, although no adduct ions were visible. The most intense ms2 ion formed included m/z 285. The neutral loss difference between the molecular ion

(m/z 579) and the m/z 285 daughter ion was equal to 294 amu, revealing the presence of either a

sambubiose (2-O-beta-D-xylosyl-D-glucose) or lathyrose (2-O-beta-D-xylosyl-D-galactose)

disaccharide sugar. Further dissociation of m/z 285 yielded the following ms3 ion fragments:

m/z 267, 257, 241, 217, 213, 201, and 175. This follows the exact fragmentation fingerprint for

the flavonoid luteolin. Therefore, compound 25-A1 is identified as luteolin, plus either a

sambubiose or lathyrose disaccharide side-group.

Compound 26-A1 at RT 24.26 gave a weak UV response, and a relatively weak MSn ion

intensity. An additional compound (Compound 28-A1) with identical molecular weight and

fragmentation characteristics was also found at RT 26.39, although, it had a much higher ion

relative intensity. In both cases, a molecular ion was noted at m/z 243 (M-H)-, confirmed by a

80 - sodium acetate [(M-H)+CH3COONa] adduct ion at m/z 325 (82 amu difference). The molecular

ion dissociated to yield an ms2 fragment of m/z 225. Other less intense ions were seen including

m/z 201, 179, and 171. Further dissociation of the m/z 225 ms2 daughter ion yielded ms3

fragments of m/z 207, 197, 181, 163, 125, 109, 99, and 97. According the NIST 05 Mass

Spectral Database, the dissociation pattern is similar to two coumarin-based compounds, known

as 7-methoxy-8-isopentenylcoumarin (a.k.a. osthole) and 7-methoxy-6-isopentenylcoumarin

(a.k.a. suberosin). In addition, predictive fragmentation indicates that this compound may also

be either oxyresveratrol (Figure 13) or cedrecoumarin A (Figure 14).

Compound 27-A1 at RT 25.09 exhibited an intermediate UV intensity, but an extremely

high relative ion abundance. This compound produced the highest ion response out of all

Fraction A compounds, resulting in a compound which likely has the highest concentration in

this fraction. An m/z 187 (M-H)- molecular ion was confirmed by the presence of an acetate

- [(M-H)+CH3COOH] adduct ion at m/z 246 (due to the 59 amu difference). The molecular ion

dissociated to form the following ions in the ms2 spectra: m/z 169, 157, 151, 133, 125 (the most

intense ion), and 97. The m/z 125 daughter fragment further dissociated to yield ms3 ion

fragments of m/z 97, 83, 69, and 57. The fragmentation pattern of this compound is nearly

identical to that of the standard for gallic acid (with regards to the m/z 125, 97, and 69 ion

fragments). However, there is an 18 amu difference between the molecular weight of gallic acid

(170 amu MW) and compound 27-A1 (188 amu MW). Several scenarios exist for the

identification of compound 27-A1. First, the fragmentation pattern suggests this compound may

be a with a structure similar to that of gallic acid. Also, the NIST 05 Mass Spectral

Database revealed a high probability that compound 27-A1 may be a coumarin-based structure.

In addition, there is also a possibility this compound (according to the NIST 05 Mass Spectral

81 Database) may be a structure similar to that of 5-hydroxy-2-methyl-1,4-naphtoquinone (a.k.a. plumbagin).

Compound 30-A1 at RT 30.46 exhibited an intermediate-range UV intensity and high relative ion abundance. A molecular ion was discovered at m/z 169 (M-H)- and confirmed by an

- acetate [(M-H)+CH3COOH] adduct ion at m/z 229 (59 amu difference), and a sodium acetate

- [(M-H)+CH3COONa] adduct at m/z 251 (82 amu difference). Fragmentation of the molecular

ion produced ms2 ions of m/z 153 and 125, and further dissociation yielded an ms3 spectra with daughter ions of m/z 95, 83, 71, 57, and 55. Although the molecular weight is similar to that of gallic acid (3,4,5-trihydroxybenzoic acid), it is definitely not this compound due to slight differences in fragmentation characteristics. In addition, comparing the retention times of compound 30-A1 (RT 30.46) to the gallic acid standard (RT 5.49) indicate the two are not identical. However, the dissociation pattern suggests that these compounds are in fact similar

(i.e., both are trihydroxy-benzoic acids). Predictive fragmentation indicates that compound 30-

A1 is most likely 2,4,6-trihydroxybenzoic acid (Figure 15), or any other possible

trihydroxybenzoic acid derivative where the hydroxy groups are not located at the 3-4-5

positions.

Compound 31-A1 at RT 31.46 (min) exhibited a low UV response, but a high ion

intensity. A molecular ion was confirmed at m/z 719 (M-H)- due to the presence of an acetate

- [(M-H)+CH3COOH] adduct ion at m/z 779 (60 amu difference). Further dissociation of the

molecular ion yielded an ms2 spectra with the most intense ion of m/z 659, and a low intensity

ion at m/z 497. The continued dissociation of m/z 659 gave rise to fragments of m/z 539, 497,

479, and 377 (in the ms3 ion spectra). The m/z 539 fragment reveals a neutral loss of 180 amu,

indicating the presence of a glucose (or galactose) sugar unit. According to the NIST 05 Mass

82 Spectral Database, the remaining fragments associated with the m/z 539 daughter ion are a near perfect match with the compound known as .

Compound 34-A1 at RT 35.05 showed both a relatively strong UV response and ion relative abundance. The most intense ion was seen at m/z 557 (M-H)- in the normal mass

spectra, which dissociated to yield an m/z 497 fragment in the ms2 spectra. The 60 amu neutral

- loss reveals the presence of an acetate [M+CH3COO] adduct ion, indicating m/z 497 as the

molecular ion. The molecular ion dissociated to give an ms3 fragment of m/z 335, which the

162 amu neutral loss indicates the presence of single monosaccharide unit. Given the number of

coumarin-based compounds which exist in this fraction (and also the discovery of m/z 335

coumarin compounds in other fractions), there is a likelihood that compound 34-A1 is any

number of coumarin compounds with a 336 amu molecular weight. These compounds include:

6-(3-methylbut-2-enyl)- (a.k.a. ), 3,3'-methylenebis[4-hydroxy-coumarin]

(a.k.a. dicoumarin), lasiocephalin, and 6-(7-hydroxycoumarin-8-yl)-7-methoxy-coumarin.

Compound 35-A1 at RT 35.44 gave a both low UV intensity and ion relative abundance.

The molecular ion was confirmed in the normal mass spectra at m/z 327 (M-H)- by the presence

- of several adduct ions including: acetate [M+CH3COO] at m/z 386 (59 amu difference), and

- sodium acetate [(M-H)+CH3COONa] at m/z 409 (82 amu difference). An additional compounds with the same molecular weight and similar fragmentation patterns were also found at RT 35.80

(compound 36-A1), RT 38.27 (compound 39-A1), RT 38.80 (compound 41-A1), and RT 43.98

(compound 47-A1). In all cases, dissociation of the molecular ion gave an ms2 ion spectra with fragments of m/z 309 (most intense), 221, 171, and 165. This follows the possible fragmentation pathway for the compound kaempferol-3,4',7-trimethyl-ether (Figure 16). The NIST 05 Mass

Spectral Database matched a similar compound (although the methoxy and hydroxy groups were

83 located a different positions) known as 3-hydroxy-4’,5,7-trimethoxyflavone. Therefore, these compounds are most likely various derivatives of the aforementioned compounds (with methoxy and hydroxy groups located at different positions on the flavonoid A, B, and C rings).

Compound 43-A1 at RT 41.85 exhibited both a low UV response and ion intensity. The m/z 445 (M-H)- molecular ion dissociated to yield the following ions in the ms2 spectra: m/z

429, 416, 385 (the most intense ion), 383, 382, 367, 327, 325, 309, 293, 265, 249, and 194.

Further dissociation of the m/z 385 ion produced the following fragments in the ms3 spectra:

m/z 367, 354, 352, 341, 325, 311, and 309. The NIST 05 Mass Spectral Database produced a

high probability match for the compounds and . Although the fragmentation pattern in the NIST 05 Database gives insight into the characterization of compound 43-A1, this is however not a true match, since the molecular weights for vitexin (and isovitexin) is 432 amu.

Several 446 amu compounds exist which exhibit structural similarities to vitexin and isovitexin.

In addition, the possible fragmentation pathway for these compounds is nearly identical to that

produced by compound 43-A1. These compounds include the following: 4',7-dihydroxy-6-

methoxyisoflavone-7-D-glucoside (a.k.a. ), -7-O-glucuronide, and -

7-glucoside (a.k.a. sissotrin) (Figures 17).

Fraction A - HPLC Method 2

The use HPLC method 2 exhibited better overall separation (than method 1), and in

addition to all the structures identified in method 1, allowed the detection of compounds which

were not otherwise seen in the first method. UV chromatograms and TICs for each scan range

can be seen in Figures A.40 through A.41. A total of 8 additional compounds were identified in

peanut skin Fraction A obtained from Toyopearl SEC (Table 2). These compounds mostly

84 included various types of phenolic acids. For example, compound 1-A2 was detected in extremely high concentrations (in both UV and ion intensity) from RT 4.08 to 14.26 (min), with a main peak finally appearing at RT 17.77. The m/z 239 (M-H)- molecular ion dissociated to

- form an m/z 179 fragment, indicating the presence of an acetate [M+CH3COO] adduct ion (60

amu difference). The m/z 179 (M-H)- molecular ion yielded fragments of m/z 161, 145, 117,

101, and 99. The fragmentation pattern appears to be that of a dihydroxycinnamic acid with a

structure similar to 3,4-dihydroxycinnamic acid (a.k.a. caffeic acid) (Figure 18), however, the

retention times do not match that of the caffeic acid standard (at RT 31.57). Therefore,

compound 1-A2 is assumed to be a derivative of caffeic acid, possibly with hydroxyl units

residing at different positions on the phenolic ring. The constant elution of this compound over a long retention time range could possibly be caused by either an extremely high concentration of a single dihydroxycinnamic structure, or possibly due to the presence of multiple dihydroxy- cinnamic acid derivatives.

Compound 2-A2 at RT 16.72 (min) exhibited an intermediate-to-high UV intensity, but a

low relative ion abundance. An m/z 221 (M-H)- ion dissociated to yield an m/z 161 fragment,

- indicating the loss of acetate [M+CH3COO] adduct ion, and confirming m/z 161 as the

molecular ion. Further fragmentation of the molecular ion yielded ion fragments of m/z 143 and

125. Possible compounds which produce an m/z 143 compound include coumarin-based

structures, such as 3-hydroxycoumarin, 4-coumarinol, 7-hydroxycoumarin (a.k.a. umbelliferone)

(Figure 19), and 4-hydroxycoumarin.

Compound 3A-2 at RT 40.05 (min) showed an intermediate-to-high UV intensity, but a

high ion relative abundance. The m/z 173 molecular ion (M-H)- was confirmed by the presence

- of acetate [(M-H)+CH3COOH] adduct ion at m/z 232 (59 amu difference). The molecular ion

85 dissociated to form ms2 ion fragments of m/z 155, 143, 142, 129, 111, 109, 83, and 81. This closely follows the predictive fragmentation pathway for shikimic acid (Figure 11). The fragmentation fingerprint was also confirmed using the NIST 05 Mass Spectral Database. The presence of this compound at high concentrations makes sense, given that shikimic acid is the precursor to the formation of flavonoid structures.

In addition to compounds which were undetectable using the first method, method 2 often allowed the separation of multiple compounds with identical molecular weight and fragmentation attributes (i.e., compounds which only eluted once using method 1). The most viable explanation for this phenomenon is the presence of compounds with similar structural characteristics (while having the same molecular weight), or perhaps due to side-groups residing at different locations on the base-structure. For example, compound 4-A2 at RT 48.08 (min) exhibited an intermediate-to-high UV intensity, but a high ion relative abundance. An m/z 243

(M-H)- molecular ion dissociated to yield ms2 ion fragments of m/z 225, 201, 161, and 151.

Additional compounds with identical molecular weights and fragmentation characteristics were

also found at RT 49.51, 52.88, 53.57, and 54.54 (min). Two possible compounds which could

theoretically exhibit such a dissociation pathway include cedrecoumarin A and oxyresveratrol

(although, the fragmentation scheme more closely follows cedrecoumarin A) (Figures 13 and

14).

Compound 6-A2 at RT 54.67 (min) showed an intermediate UV intensity, and a high ion

relative abundance. A molecular ion at m/z 241 (M-H)- dissociated to yield the following ion

fragments in the ms2 spectra: m/z 225, 223, 205, 197, 179, and 161. Predictive fragmentation of

a dihydroxy-flavan could produce this fragmentation fingerprint, such as 4',7-dihydroxyisoflavan

(a.k.a. equol) (Figure 20). In addition, the NIST 05 Mass Spectral Database produced spectra

86 matches from several similar flavandiol structures, including 4,4’-flavandiol and 4,7-flavandiol

(a.k.a. 2-phenyl-4,7-chromanediol).

Compound 8-A2 at RT 60.30 (min) exhibited a somewhat low UV intensity, but a high ion relative abundance. A molecular ion was discovered at m/z 201 (M-H)-, and dissociated to

form ion fragments of m/z 183, 140, and 139 in the ms2 spectra. This follows the possible

fragmentation pathway for various hydroxyfurocoumarin’s, including 5-hydroxyfurocoumarin

(a.k.a. bergaptrol) and 8-hydroxy-4'-5',6-7-furocoumarin (a.k.a. xanthotoxol) (Figure 21).

Fraction B - HPLC Method 1

A total of 24 compounds were identified in peanut skin Fraction B obtained from

Toyopearl SEC (Table 3). UV chromatograms and TICs for each scan range can be seen in

Figures A.10 through A.13. Compound 1-B1 at RT 8.78 (min) exhibited an extremely high UV

intensity, and an intermediate ion relative abundance. The molecular ion was assumed to be at

m/z 153 (M-H)-, although, no adduct ions were observed. The parent ion dissociated to yield an

m/z 109 fragment in the ms2 spectra. Further dissociation of this daughter ion produced ms3

ions of m/z 91 and 81. Almost all dihydroxybenzoic acid derivatives will follow this

fragmentation pathway, including: 2,3-dihydroxybenzoic acid, 2,4-dihydroxybenzoic acid (a.k.a.

β-resorcylic acid), 2,5-dihydroxybenzoic acid (a.k.a. ), 3,4-dihydroxybenzoic acid

(a.k.a. ), and 3,5-dihydroxybenzoic acid (a.k.a. α-resorcylic acid). An

example of the fragmentation pathway can be seen for 2,6-dihydroxybenzoic acid (a.k.a. γ- resorcylic acid) in Figure 22.

Compound 2-B1 at RT 11.07 (min) exhibited an intermediate UV intensity and ion relative abundance. An intense m/z 137 ion was assumed to be the molecular ion, although, no

87 adduct ions were observed. The parent ion dissociated to yield an ms2 spectra with the following ion fragments: m/z 119, 109, 108, 99, 95, 93, 81, 65, and 63. This follows the fragmentation pathway for all hydroxybenzoic acid derivatives, including: 3-hydroxybenzoic acid (a.k.a. m-salicylic acid), 4-hydroxybenzoic acid (a.k.a. p-salicylic acid), and 2- hydroxybenzoic acid (a.k.a. salicylic acid) (Figure 23). Another possible dissociation scheme includes compounds similar in structure to pyrocatechin monoethyl ether (Figure 24). Another compound (compound 3-B1) at RT 12.54 exhibited the same fragmentation characteristics, although, with an added glucose (or galactose) unit.

Compound 3-B1 at RT 12.54 (min) exhibited both a low UV and ion intensity response.

The molecular ion, which exists at m/z 299 (M-H)-, was confirmed by the presence of a sodium

- acetate [(M-H)+CH3COONa] adduct at m/z 381. Dissociation of the molecular ion yielded ms2

fragmentation which confirmed the presence of a glucose unit (162 amu neutral loss), due to an

intense m/z 137 fragment. Further fragmentation of m/z 137 (M-H)- produced an m/z 93 ms3

fragment. This follows the possible dissociation scheme for various hydroxybenzoic acids, such

as 2-hydroxybenzoic acid (a.k.a. salicylic acid) (Figure 23), 3-hydroxybenzoic acid (a.k.a. m-

salicylic acid), and 4-hydroxybenzoic acid (a.k.a. p-salicylic acid). The ms2 and ms3 spectra

were also a confirmed match with 4-hydroxybenzoic acid found in the NIST 05 Mass Spectral

Database. Therefore, compound 3-B1 is identified as one of the aforementioned hydroxybenzoic

acids, plus an added glucose (or galactose) monosaccharide side-group.

Compound 4-B1 at RT 12.98 (min) exhibited a low UV intensity, but an intermediate ion

relative abundance. A molecular ion was noted at m/z 415 (M-H)-, confirmed by a sodium

- acetate [(M-H)+CH3COONa] adduct ion at m/z 497 (82 amu difference). The molecular ion

dissociated to yield an m/z 253 fragment in the ms2 spectra, revealing the presence of a glucose

88 (or galactose) unit due to the 162 amu neutral loss. Further dissociation of the m/z 253 daughter ion produced the following ms3 ion fragments: m/z 235, 202, 193, 160, 151, and 109. These fragments indicate the presence of a 254 amu flavonoid aglycone. Possible candidates of a compound with this molecular weight (and fragmentation pattern) include various derivatives of dihydroxyflavones and dihydroxyisoflavones. Examples of such compounds include 3,4'- dihydroxyflavone and 4',7-dihydroxyisoflavone (a.k.a. daidzein) (Figure 25).

Compound 5-B1 at RT 16.27 (min) exhibited both an intermediate UV intensity and ion relative abundance. An m/z 449 (M-H)- molecular ion was confirmed by the presence of an

- - acetate [M+CH3COO] adduct ion at m/z 508, and a sodium acetate [(M-H)+CH3COONa] adduct ion at m/z 531. Dissociation of the molecular ion yielded an m/z 287 daughter fragment in the ms2 spectra, indicating the presence of a glucose (or galactose) side-group. Other ions noted in the ms2 spectra including m/z 431, 269, and 259. Further fragmentation of the m/z 287 fragment produced an ms3 spectra with the following ion fragments: m/z 269, 243, and 201.

The most likely compound which follows this dissociation scheme is eriodictyol-7-glucoside

(Figure 26). In addition to 3',4',5,7-tetrahydroxyflavanone (a.k.a. eriodictyol), several other 288 amu molecular weight flavonoids exist which could follow the dissociation pathway of the flavonoid aglycone monomer exhibited by compound 5-B1. These compounds include 2,3- dihydrofisetin (a.k.a. fustin) (Figure 27), 6-hydroxyapigenin, and dihydrokaempferol.

Fraction B contained numerous compounds which consisted of quercetin aglycones, along with various types of sugar-based side groups. These (poly)saccharide side-groups differed only by the number and type (i.e., pentose or hexose) of individual sugar units. For example, compound 7-B1 at RT 17.81 (min) exhibited both a low UV and ion intensity response.

A molecular ion was discovered at m/z 933 (M-H)-, and confirmed by a sodium acetate [(M-

89 - H)+CH3COONa] adduct at m/z 1015 (82 amu difference). Dissociation of the molecular ion

produced an ms2 ion of m/z 771, which designates a glucose (or galactose) sugar via a 162 amu

difference. Further fragmentation of the m/z 771 ion gave rise to the following daughter ions in

the ms3 spectra: m/z 591, 301, 300, 271, and 255. The neutral loss associated with the

difference between m/z 771 and 591 equals 180 amu, identifying the presence of an additional

glucose (or galactose) sugar. The neutral loss found between m/z 933 and 301 shows a

difference of 632 amu, pinpointing the presence of two disaccharide units. The possible

combination of disaccharides include one 324 amu disaccharide (sophorose, laminaribiose, or

gentiobiose), plus an additional 308 amu disaccharide (rutinose). The ms3 ion fragments of m/z

301, 300, 271, and 255 are indicative of a quercetin aglycone with an O-, which was verified by the standard for this compound. Therefore, this compound is most likely a quercetin-based monomer, plus two of the aforementioned disaccharides. Another example of a verified quercetin-glycoside includes compound 13-B1 at RT 19.66. This structure is made up of a quercetin aglycone monomer with a rutinose+glucose side-group. This compound can also be considered rutin or quercetin-3-O-rutinose (Figure 28), plus an additional glucose (or galactose).

Compound 9-B1 at RT 18.92 (min) exhibited both an intermediate UV intensity and ion

relative abundance. An m/z 457 (M-H)- molecular ion was confirmed by the presence a sodium

- acetate [(M-H)+CH3COONa] adduct ion at m/z 539 (82 amu difference). The use of a single-

step mass spectrometry technique (such as TOF-MS or EI-MS) could erroneously identify this

structure as gallocatechin gallate (GCG) or epigallocatechin gallate (EGCG), since they have the

same 458 amu molecular weight as compound 9-B1. However, the use of the current mult-stage

MS/MS/MS (MSn) technique reveals that the fragmentation pathway of this compound is

completely different than EGCG or GCG (according to the standards for these compounds). In

90 addition, the retention times for the GCG (RT 15.70) and EGCG (RT 17.44) standards do not match that of compound 9-B1 (at RT 18.92). Furthermore, a full scan of all fractions using both

HPLC Method 1 and HPLC Method 2 (while utilizing a filter to pinpoint any 458 amu molecular weight compounds) revealed that neither GCG nor EGCG exists in any amount in the Gregory- line peanut skins. The lack of these compounds in peanut skins makes perfect sense, given that the existence of a galloyl-group (as seen with GCG and EGCG) requires the presence of gallic acid, which was also not detected in any amount in all fractions. In the case of compound 9-B1, the m/z 457 (M-H)- molecular ion dissociated to form the following product ions in the ms2

spectra: m/z 440, 311, 295, 293 (most abundant ion), 277, 221, 204, and 163. The presence of an m/z 295 (M-H)- ion fragment indicates a 162 amu neutral loss, revealing the presence of a

glucose (or galactose) monosaccharide unit. The m/z 293 daughter ion underwent further

dissociation to yield ms3 ion fragments of m/z 275, 219, 191, 187, 163, and 127. This follows

the predictive fragmentation scheme for the compound 6-O-acetyldaidzin (Figure 29).

Compound 10-B1 at RT 19.34 exhibited a high UV intensity, and an intermediate-ranged

relative ion abundance. A molecular ion was found at m/z 487 (M-H)-, confirmed by a sodium

- acetate [(M-H)+CH3COONa] adduct ion at m/z 569 (82 amu difference). The molecular ion

dissociated to yield an ms2 spectra with m/z 325 (most intense ion), 307, 293, 193, and 179 ion

fragments. The m/z 325 ion indicates the existence of a glucose (or galactose) sugar unit, due to

the 162 amu neutral loss. The m/z 325 daughter ion underwent further dissociation to yield m/z

193 (most intense ion), 149, and 134 ions in the ms3 spectra. The 132 amu neutral loss (between

m/z 325 and 193) reveals the presence of a 132 amu xylose (or arabinose) sugar unit. The

remaining fragments associated with m/z 193 (i.e., m/z 179, 149, and 134) follow the exact

dissociation pattern of 3-methoxy-4-hydroxycinnamic acid (a.k.a. ferulic acid) (Figure 30),

91 which was confirmed by a ferulic acid standard. Therefore, compound 10-B1 is identified as ferulic acid, plus a 294 amu disaccharide side group (either sambubiose or lathyrose).

Compound 11-B1 at RT 19.39 gave both a high UV response and relative ion abundance.

Although this compound is most likely a quercetin-glycoside, there is however a possibility that the aglycone may be a different flavonoid monomer. A molecular ion was noted at m/z 903 (M-

- - H) , confirmed by an acetate [M+CH3COO] adduct ion at m/z 962 (59 amu difference) and a

- sodium acetate [(M-H)+CH3COONa] adduct ion at m/z 985 (82 amu difference). The molecular

ion underwent collision-induced dissociation (CID), and yielded an ms2 fragment of m/z 741.

Other less intense ions were seen including m/z 591, 463, and 447. Further dissociation of the

m/z 741 ms2 daughter ion yielded ms3 fragments of m/z 609, 591, 463, 447, 301, 300, 271, and

255. The remaining fragments point to two possible scenarios regarding the identification of this

compound, due to the dissociation pathway of the flavonoid aglycone monomer. First, the

neutral loss associated with m/z 741 (from m/z 903) equals 162 amu, which is undoubtedly a

glucose (or galactose) unit due to the extreme labile nature of glycosidic bonds. Further

dissociation of m/z 741 yielded m/z 609, showing a neutral loss of 132 amu, which is indicative

of a xylose (or arabinose) sugar unit. An additional fragment at m/z 463 revealed a neutral loss

of 146 amu, which shows the presence of a rhamnose sugar unit. The last intense ion fragment

of m/z 300-301 confirms a difference of 162 amu, indicating the presence of a second glucose

(or galactose) unit. The remaining fragments (m/z’s 271 and 255) possibly reveal the presence

of a flavonoid aglycone with an m/z of 301 (M-H)-, which is most likely quercetin with either a

7-D-glucoside (Figure 31) or 4’-O-glucoside linkage. This fragmentation scheme would

tentatively identify the compound as quercetin with the following sugar sidegroups: glucose(galactose) + rhamnose + xylose(arabinose) + glucose(galactose), or simply lathyrose (or

92 sambubiose) plus rutinose. The second possible fragmentation pathway begins with the loss of glucose (162 amu) from m/z 903 (M-H)-, yielding the m/z 741 daughter ion. Further dissociation

of m/z 741 yields an m/z 609 fragment (132 amu neutral loss), revealing a xylose or arabinose

sugar. Further fragmentation gives a daughter ion at m/z 447, with a difference of 162 amu

signifying a glucose or galactose sugar. Further loss of one more xylose/arabinose unit yields

315 amu, which would reveal the presence of either 6-methoxyluteolin (aka nepetin), 3-

methylquercetin (a.k.a. isorhamnetin), 4’-methoxyquercetin (a.k.a. ), or 7-

methylquercetin (a.k.a. rhamnetin). The possible presence of one of these m/z 315 flavonoids is

further supported by the fragments seen at m/z 300, 271 and 255, although, the dissocation

pathway is most similar to that of the methoxy-flavone called isorhamnetin. In addition to

isorhamnetin, the NIST 05 Database also revealed close matches for other methoxy-flavones,

including 3’,4’,5,6-tetrahydroxy-7-methoxy-flavone (a.k.a. pedalitin), which further supports the theory that the aglycone may possibly be a methoxy-flavonoid. Therefore, the second possible identification of this compound includes one of these aforementioned m/z 315 compounds (most likely isorhamnetin), plus the following sugar side-groups: glucose(galactose) + xylose(arabinose) + glucose(galactose) + xylose(arabinose), or simply sambubiose (or lathyrose) plus another sambubiose (or lathyrose).

Compound 14-B1 at RT 20.14 exhibited a high UV intensity, and an intermediate-ranged relative ion abundance. A molecular ion was found at m/z 137 (M-H)-, confirmed by a sodium

- acetate [(M-H)+CH3COONa] adduct ion at m/z 219 (82 amu difference). The molecular ion

dissociated to yield an ms2 spectra with ion fragments of m/z 119, 109, 108, 93, and 65. This

follows the possible dissociation scheme for various hydroxybenzoic acids, such as 2-

hydroxybenzoic acid (a.k.a. salicylic acid) (Figure 23), 3-hydroxybenzoic acid (a.k.a. m-salicylic

93 acid), and 4-hydroxybenzoic acid (a.k.a. p-salicylic acid). A search of the ms2 spectra using the

NIST 05 Mass Spectral Database produced a perfect match for 4-hydroxybenzoic acid (a.k.a. p- salicylic acid).

As in the previous case regarding quercetin aglycone monomers, Fraction B contained numerous compounds which consisted of isorhamnetin aglycones with various types of sugar- based side groups. These (poly)saccharide side-groups differed only by the number and type

(i.e., pentose or hexose) of individual sugar units. For example, compound 15-B1 at RT 21.49

(min) exhibited both a very high UV and ion intensity response. A molecular ion was discovered

- - at m/z 785 (M-H) , and confirmed by an acetate [(M-H)+CH3COOH] adduct ion at m/z 844 (59

amu difference). Dissociation of the molecular ion produced an ms2 ion of m/z 605, which

designates a glucose (or galactose) sugar via a 180 amu difference. The most intense ion in the

ms2 spectra was found at m/z 315 (M-H)-, which reveals a neutral loss of 470 amu (from the m/z

785 molecular ion). The 470 amu difference indicates the presence of a 308 amu rutinose (6-O- alpha-L-rhamnosyl-D-glucose) disaccharide sugar, plus an additional 162 amu glucose (or galactose) monosaccharide unit. Additional (less intense) ion fragments in the ms2 spectra

included m/z 299 and 271. Further fragmentation of the m/z 315 ion gave rise to the following

daughter ions in the ms3 spectra: m/z 300, 299, 285, 271, 243, and 177. The ms3 ion fragments

(along with the m/z 315, 299, and 271 daughter ions found in the ms2 spectra) indicate the

presence of an isorhamnetin aglycone monomer (Figure 32). Therefore, compound 15-B1 is

identified as isorhamnetin-3-O-rutinoside, plus an additional glucose (or galactose) sugar. This

finding is further supported by an identical discovery by (Lou et al., 2001) who used NMR

analysis to confirm the presence of isorhamnetin-3-O-[2-O-β-glucopyranosyl-6-O-α-

rhamnopyranosyl]-β-glucopyranoside in peanut skins. Another example of a verified

94 isorhamnetin-glycoside includes compound 16-B1 at RT 22.18. In this case, the m/z 639 (M-H)-

molecular ion underwent collision-induced dissociation (CID) to yield an m/z 315 isorhamnetin

monomer. In this case, the structure is made up of an isorhamnetin aglycone monomer with one

of the following dissacharide side-groups (due to the 324 amu neutral loss): laminaribiose (3-O-

beta-D-glucosyl-d-glucose), gentiobiose (6-O-beta-D-glucosyl-D-glucose), or sophorose (2-O-

beta-D-glucosyl-d-glucose). The final isorhamnetin-based structure was compound 18-B1, and

was detected at RT 23.31. This compound exhibited an m/z 755 (M-H)- molecular ion, which

underwent collision-induced dissociation (CID) to yield an m/z 315 isorhamnetin monomer. The

440 amu neutral loss indicates the presence of a 308 amu rutinose (6-O-alpha-L-rhamnosyl-D- glucose) disaccharide, plus an additional 132 amu xylose (or arabinose) monosaccharide. Again,

an identical compound (isorhamnetin-3-O-[2-O-β-xylopyranosyl-6-O-α-rhamnopyranosyl]-β- glucopyranoside) was reported by (Lou et al., 2001) by means of NMR analysis.

Compound 19-B1 at RT 24.54 exhibited both an intermediate UV intensity and ion relative abundance. An m/z 463 (M-H)- molecular ion was confirmed by the presence of a

- sodium acetate [(M-H)+CH3COONa] adduct ion at m/z 545 (82 amu difference). The molecular

ion dissociated to yield an ms2 spectra with ion fragments of m/z 301 (most intense ion), 283,

257, and 234. The m/z 301 ion fragment indicates a neutral loss of 162 amu (a glucose or

galactose sugar). Further dissociation of the m/z 301 ion produced the following ions in the ms3

spectra: m/z 286, 271, 256, 241, 217, 164, and 151. This follows the exact dissociation pathway

for the compound 3',5,7-trihydroxy-4'-methoxyflavanone (hesperitin) (Figure 33). Confirmation

of hesperitin was also conducted by matching MS/MS Ion Trap standards run in this study.

Compound 20-B1 at RT 27.59 exhibited both an intermediate UV response and ion

relative abundance. A molecular ion was identified at m/z 461 (M-H)- due to the presence of a

95 chlorine [M+Cl-]- adduct ion at m/z 500 (37 amu difference). The molecular ion dissociated to

form ms2 spectra with an intense ion at m/z 299, and less intense ions at m/z 284, 255, and 227.

The neutral loss associated with the m/z 299 ion fragment yields 162 amu, indicative of a

glucose (or galactose) sugar unit. The m/z 299 daughter fragment underwent further dissociation

to produce an intense m/z 284 ion in the ms3 spectra. Other less intense ions included m/z 267,

243, 178, 151, and 119. The 15 amu loss (methyl group) between the m/z 299 ion and the ms3

daughter fragment indicates the likely presence of a methoxy group. Dissociation pathways

based on obtained fragmentation data of standards for this study indicate a high probability that

the compound is 5,7,4'-trihydroxy-3'-methoxy-flavone (a.k.a. chrysoeriol), specifically with

regards to the m/z 284, 256, and 151 fragments. However, it has also been shown that

chrysoeriol (Figure 34) exhibits dissociation patterns which are nearly identical to that of

(aka luteolin-4'-methyl ether). In addition, these two compounds elute at nearly the

same retention times under similar chromatographic conditions, which supports the theory that

the other compound may possibly be diosmetin. Although fragmentation data for the standards

of these compounds was obtained, definitive matches are only possible by running the actual

standards to determine UV retention times (which were not available for this research).

Although the aglycone in compound 20-B1 is most likely one of the two aforementioned

compounds, the lack of definitive proof for identification opens the possibility that the aglycone

may in fact be one of many other 300 amu flavonoid compounds which exhibited similar

fragmentation pathways (specifically with regards to the m/z 284 and 227 daughter ions). The

dissociation pathway follows the predictive fragmentation patterns for numerous bioflavonoids

with 300 amu molecular weights, including: 3',5,7-trihydroxy-4'-methoxyisoflavone (a.k.a.

), 4',5,7-trihydroxy-6,8-dimethylflavanone (a.k.a. farresol or cyrtopterinetin), 4',5,7-

96 trihydroxy-6-methoxyisoflavone (a.k.a. tectorigenin) (Figure 35), psi-tectorigenin, 4'-hydroxy-

5,7-dimethoxyflavanone, 8-methoxy-iso-scutellargin, 3,7-dihydroxy-2-(4-hydroxy-3-methoxy- phenyl)-4-benzopyrone (a.k.a. geraldol), and kaempferol 4'-methyl ether (a.k.a. ).

Therefore, compound 20-B1 is one of the aforementioned flavonoid aglycones, with an attached glucose (or galactose) monosaccharide side-unit.

Compound 21-B1 at RT 49.11 exhibited both a high UV intensity and ion relative abundance. A molecular ion at m/z 435 (M-H)- was confirmed by the presence of a sodium

- acetate [(M-H)+CH3COONa] adduct ion at m/z 517 (82 amu difference). The parent ion

dissociated to form the following ion fragments in the ms2 spectra: m/z 420 (the most intense

ion), 378, 366, 351, 312, 299, and 271. The m/z 420 further dissociated to yield ms3 fragments

of m/z 405, 377, 351, 338, 309, 298, 283, 270, 259, and 188. Predictive fragmentation patterns

indicate that compound 21-B1 may be similar in structure to a compound known as

orotinichalcone (Figure 36). In addition, a search of the NIST 05 Mass Spectral Database

revealed a similar structure (2’-O-methylcajanone) with analogous fragmentation patterns.

Several other compounds with similar fragmentation patterns were detected at RT 50.90 and

53.00 (min). As in the case with the previous structure, the NIST 05 Database showed possible

matches to various chalcone, chromone, and chromene-based compounds.

Fraction B - HPLC Method 2

The use HPLC method 2 exhibited better overall separation (than method 1), and in

addition to all the structures identified in method 1, allowed the detection of compounds which

were not otherwise seen in the first method. UV chromatograms and TICs for each scan range

can be seen in Figures A.42 through A.43. A total of 27 additional compounds were identified in

97 peanut skin Fraction B obtained from Toyopearl SEC (Table 4). For example, compound 2-B2 at RT 17.08 (min) exhibited a high UV intensity, and an intermediate-ranged ion relative abundance. An m/z 403 (M-H)- molecular ion was confirmed by the presence of an acetate

- [M+CH3COO] adduct ion at m/z 463. The molecular ion dissociated to yield the following ions in the ms2 spectra: m/z 357, 293, 271 (most intense ion), 240, 191, 177, 161, and 149. Further dissociation of the m/z 271 daughter fragment yielded ms3 ions of m/z 181, 161, 151, 113, 109,

101, 100, and 97. Although some of these fragments (m/z 177 and 151 M-H-) would indicate the

presence of a terminal rhamnose sugar unit along with a naringenin aglycone (as shown by fragmentation patterns of standards), the dissociation scheme also follows that of a compound

known as 2',3,4,4'-tetrahydroxychalcone (a.k.a. butein) (plus a terminal rhamnose monosaccharide unit) (Figure 37).

Compound 5-B2 at RT 22.01 exhibited a high UV intensity, but a low relative ion abundance. An intense m/z 419 (M-H)- ion was found in the normal mass spectra, but no adduct

ions were noted. Dissociation of this ion produced an intense m/z 359 ion in the ms2 spectra,

- yielding a 60 amu neutral loss. This would indicate the presence of an acetate [M+CH3COO] adduct ion, revealing m/z 359 as the molecular ion. Additional less intense ions in the ms2 spectra included m/z 289, 177, and 152. Further dissociation of the m/z 359 ion led to the following ion fragments in the ms3 spectra: m/z 342/341, 297, 289, 273, 247, 231, 207/206, 191,

165, and 153. Prediction of the fragmentation scheme indicates that the dissociation pathway is nearly identical to a compound known as 3',5,7-trihydroxy-4',5',6-trimethoxyisoflavone (a.k.a. irigenin) (Figure 38). However, the fragmentation patterns may also follow other trihydroxy- trimethoxy-flavonoid derivatives. In addition, a spectral search using the NIST 05 Database indicated ion fragments similar to 3-hydroxy-3’,4’,5,7-tetramethoxy-flavanone. The NIST 05

98 search also yielded a flavonoid compound known as 3-hydroxy-3’,4’,7,8-tetramethoxy-flavanone

(although it produced slightly less similar fragment matches).

Compound 10-B2 at RT 33.65 (min) exhibited a low UV and ion intensity. The molecular ion at m/z 479 (M-H)- dissociated to yield an ms2 ion of m/z 317, which signify the

loss of a 162 amu glucose (or galactose) unit. Other less intense ion fragments in the ms2

spectra included m/z 299, 289, 258, and 179. Further dissociation of the m/z 317 daughter ion

resulted in ms3 fragments with m/z’s 284, 268, and 176. The m/z 317 fragment indicates the

molecular weight (318 amu) of the flavonoid aglycone. The remaining ions follow the most

likely fragmentation patterns for the following m/z 317 (M-H)- flavonoid compounds:

3,3',4',5,5',7-hexahydroxyflavone (a.k.a. myricetin) (Figure 39), 3,3',4',5,6,7-hexahydroxy-

flavone (a.k.a. quercetagetin), and 3,3',4',5,7,8-hexahydroxyflavone (a.k.a. gossypetin).

Similarities in the ion fragmentation of these three compounds were also confirmed using the

NIST 05 Mass Spectral Database. Therefore, compound 10-B2 is most likely one of the three

aforementioned compounds, with a glucoside side-group. An example of such a compound is

gossypetin-8-glucoside (a.k.a. gossypin) (Figure 40).

Compound 13-B2 at RT 39.35 (min) exhibited an intermediate UV intensity and ion

relative abundance. A molecular ion was found at m/z 331 (M-H)-, and was confirmed by the

- presence of an acetate [M+CH3COO] adduct ion at m/z 391. The molecular ion dissociated to yield the following ion fragments in the ms2 spectra: m/z 315, 299 (most intense ion), 287, 271,

255, and 179. The m/z 299 underwent further dissociation to yield the following ion fragments in the ms2 spectra: m/z 271, 256, 239, 227, 215, 187, 179, 151, and 124. According to the NIST

05 Mass Spectral Database, these ion fragments follow the exact dissociation pathway for the following chromanediol compounds: 2-(3,4-demethoxyphenyl)-7-methoxy-3,4-chromanediol,

99 and 7,8-dimethoxy-2-(4-methoxyphenyl)-3,4-chromanediol. NIST 05 also produced an additional match for 5,6,8,3’,4’-pentahydroxy-7-methoxyflavone, however, the existence of this compound is less likely due to fewer matching fragments in the ion spectra. A similar compound

(compound 19-B2) with nearly identical fragmentation patterns was found at RT 50.03, although the molecular weight was 364 amu (m/z 363 (M-H)-). The 32 amu difference indicates that

compound 19-B2 is a structure which is identical to the aforementioned chromanediol’s, except

for the addition of one methoxy group.

Compound 14-B2 at RT 42.64 (min) exhibited both an intermediate UV intensity and ion

relative abundance. An m/z 389 (M-H)- molecular ion dissociated to yield the following ions in

the ms2 spectra: m/z 374, 341 (the most intense ion), 323, 303, 195, and 193. The m/z 374 ion

fragment indicates a loss of 15 amu (CH3) from the molecular ion, indicating the possible

presence of a methoxy group. Further dissociation of the m/z 341 daughter ion produced the

following fragments in the ms3 spectra: m/z 326, 309, 297, 295, 250, and 233. Ion fragments

m/z 326, 309, and 295 each indicate a loss of 15 amu (CH3) from the m/z 341 daughter fragment,

revealing the possible presence of three additional methoxy groups (totaling four methoxy

groups). This follows the exact dissociation pattern for the trihydroxy-tetramethoxy-flavone

compound known as gardenin E (Figure 41). In addition, the NIST 05 Database indicated a

similar fragmentation scheme for the compound known as 3’,5,7-trihydroxy-4’,5’,6,8-

tetramethoxy-flavone (a.k.a. scaposin), which further supports the theory that compound 14-B2 is a trihydroxy-tetramethoxy-flavone.

Compound 17-B2 at RT 46.47 (min) exhibited both a high UV intensity and ion relative abundance. The m/z 301 (M-H)- molecular ion dissociated to yield ion fragments of m/z 283,

271, 257 (the most intense ion), 255, and 218 in the ms2 spectra. Further dissociation of the m/z

100 257 daughter ion produced the following fragments in the ms3 spectra: m/z 242, 229, 215, 213,

211, 199, 189, 187, 171, and 147. Except for one major missing ion (m/z 286), this follows almost the exact dissociation pathway for the flavonoid compound known as 3',5,7-trihydroxy-4'- methoxyflavanone (a.k.a. hesperitin). It was further confirmed that Compound 17-B2 was not in fact hesperitin, due to the comparison of a non-matching retention time exhibited by the standard for this compound. Other flavonoids exist (with a 302 amu molecular weight) which could also follow a similar dissociation pathway, including: 2',3,4',5,7-pentahydroxyflavone (a.k.a. morin)

(Figure 42), 3,3',4',5',7-pentahydroxyflavone (a.k.a. robinetin) (Figure 43), and 3',4',5,5',7- pentahydroxyflavone (a.k.a. tricetin).

Compound 18-B2 at RT 47.03 (min) exhibited both a high UV intensity and ion relative abundance. The molecular ion was confirmed at m/z 311 (M-H)- due to the presence of an

- acetate [M+CH3COO] adduct ion at m/z 361. The molecular ion underwent collision-induced dissociation (CID) to yield an intense ion fragment at m/z 179 in the ms2 spectra. The 132 amu neutral loss indicates the presence of a terminal rhamnose monosaccharide unit. Further dissociation of the m/z 179 daughter ion produced an m/z 135 fragment, indicative of 3,4- dihydroxycinnamic acid (a.k.a. caffeic acid) (Figure 44). Therefore, compound 18-B2 is identified as caffeic acid, plus a terminal rhamnose monosaccharide side-group.

Compound 25-B2 at RT 66.41 (min) exhibited an intermediate UV intensity, but a high ion relative abundance. An m/z 399 (M-H)- molecular ion dissociated to yield daughter fragments of m/z 384 (the most intense ion), 323, and 308 in the ms2 spectra. Further dissociation of the m/z 384 fragment produced the following ions in the ms3 spectra: m/z 370,

366, 341, 326, 309, 308, 298, 297, 283, 281, 269, 268, and 251. This follows the exact predictive fragmentation scheme for the compound torosaflavone A (Figure 45). However, the

101 NIST 05 Mass Spectral Database produced a similar spectral match for 2-[2-[3,5- dihydroxyphenyl]ethenyl]-5-hydroxy-chromone.

Compound 26-B2 at RT 80.91 (min) exhibited an intermediate UV intensity, but a high ion relative abundance. The molecular ion (at m/z 367) dissociated to yield the following ion fragments in the ms2 spectra: m/z 352, 349, 337, 297 (most intense ion), 282, and 270. The m/z

297 ion fragment further dissociated to form m/z 282 and 267 in the ms3 spectra. This fragmentation pathway follows the exact dissociation scheme for the compound known as glycycoumarin (Figure 46).

Compound 27-B2 at RT 86.62 showed both a high UV intensity and ion relative abundance. The m/z 351 molecular ion dissociated to form an intense m/z 336 ion fragment, along with less intense m/z 296 and 281 ion fragments. The m/z 336 daughter ion underwent further dissociation to yield the following fragments in the ms3 spectra: m/z 319 and 281 (the most intense ion). This fragmentation fingerprint follows the possible dissociation pathway for the compound pongagallone A (Figure 47).

Fraction C - HPLC Method 1

A total of 35 compounds were identified in peanut skin Fraction C obtained from

Toyopearl SEC (Table 5). UV chromatograms and TICs for each scan range can be seen in

Figures A.14 through A.17. Compound 2-C1 at RT 8.05 (min) had a high UV response at

280nm, and gave an ms1 parent ion of m/z = 349 (M-H)-. Further ms2 fragmentation yielded an

m/z = 267 ion, showing a neutral loss of 82 amu, which is indicative of a sodium acetate adduct

- [(M-H)+CH3COONa] . Therefore, the molecular ion in this case is m/z = 267. Further

fragmentation gave ms3 product ions of m/z = 150 and 134. This follows the possible (M-H)-

102 fragmentation pathway of several different compounds including 2’,4’-dimethoxychalcone, 7- methoxyflavonol, 3-hydroxy-7-methoxy-flavone, and 3-hydroxy-6-methoxyflavone (Figure 48).

In addition, a NIST 05 database search of the ms2 spectra confirmed a high probability that the compound is most likely a flavonoid with a methoxy group, which further supports the likelihood that compound 2-C1 is possibly one of the aforementioned flavonoid-based compounds.

Compound 3-C1 at RT 8.51 (min) also gave a high UV response, although, it exhibited a weak ion intensity. Ions produced included (M-H)- m/z 517, 493, and 435, with m/z 435 being

- the most intense ion. Ions at m/z = 517 and 493 signify sodium acetate [(M-H)+CH3COONa]

- and acetate [(M-H)+CH3COOH] adduct ions respectively, giving positive confirmation of a

molecular ion at m/z = 435. The (M-H)- molecular ion of m/z = 435 yielded ms2 ion fragments

of 417, 399, 375, 357, 345, 333, 315, and 289 (m/z). The 417 and 399 fragments both signify

neutral losses of 18 amu, indicative of water. The m/z 375, 345, 333, and 315 fragments show

losses of 60, 90, 102, and 120 amu respectively. These numbers reveal characteristic product

ions which are typically associated with the cross-ring cleavages of sugars. Also, the loss of one

(or two) water molecules is often noted in the CID spectrum of monosaccharides (Cuyckens and

Claeys, 2004). In addition, the m/z = 289 (M-H)- fragment shown in the ms2 spectra gives a

neutral loss of 146, signifying the possible presence of a terminal rhamnose unit, with a glycosidic O-linkage. Further fragmentation resulted in an ms3 ion spectra with daughter ions which are common to catechin and/or epicatechin (m/z = 245 and 205). Although there is a possibility that compound 3-C1 may be a catechin or epicatechin with a coumaroyl acylated side- group (denoted by a neutral loss of 146), it is however unlikely due to the characteristic O- glycosidic fragmentation patterns exhibited by this compound (Figure 49). Therefore, based on

103 the typical positions in which sugar units attach to a flavonoid aglycone, compound 3-C1 is tentatively identified as one of the following flavonoid-O-glycosides: catechin-3-O-rhamnoside, catechin-5-O-rhamnoside, or catechin-7-O-rhamnoside (or their epicatechin derivatives).

Compound 4-C1 at RT 12.31 (min) showed a weak UV response, with medium ion intensity. Several possible scenarios exist for the identification of compound 4-C1, since the

MSn spectra were extremely complicated. One could assume the molecular ion to be one of two possible ions, since adduct ions were not easily interpreted. One scenario offers a fragmentation scheme with the molecular ion at m/z 419, and the other at m/z 359. First, the normal mass spectra gave an intense ion at m/z 419 (M-H)-, along with a possible sodium acetate [(M-

- H)+CH3COONa] adduct at m/z 501 (due to the 82 amu difference). If in fact m/z 419 is the

parent ion, then it dissociated to form ms2 fragments of m/z 401, 359, 341, 329, and 289. The most intense ms2 ion (m/z 359) dissociated to form a plethora of ms3 fragments. These ions included m/z 341, 331, 315, 297, 289, 273, 255, 247, 245, 235, 227, 205, 203, 187, 161, and 151.

The specific fragmentation scheme of m/z 419 offers several possibilities regarding identification

of compound 4-C1. The first possibility is a type of hydroxy-hexamethoxy-flavone, or

derivative of, since the position of the lone hydroxyl group is unknown (and would likely require

NMR to determine the exact location). A similar hexamethoxy-flavone compound

(3,3’,4’,5,5’,7-hexamethoxyflavone) was matched with similar fragments by searching the ms2

spectra with the NIST 05 database, which further supports this theory (although, it exhibited one

less hydroxyl group). In addition, the neutral loss associated with the ion fragments reveals

consecutive losses of water (18 amu) and numerous methoxy (31 amu) groups. For example, the

ms3 fragments of m/z 297, 235, and 205 reveal respective losses of 62 (two methoxy groups),

124 (four methoxy groups), and 155 (five methoxy groups) amu from their m/z 359 ms2

104 daughter ion. Again, these losses support the possible presence of this class of compound. A second possibility for the m/z 419 scenario points to the presence of a stilbene-based compound, with a sugar side-group. The ion fragments of m/z 359, 329, and 315 represent neutral losses of

60, 90, and 104 amu respectively. These neutral losses commonly indicate the presence of a monosaccharide unit. Furthermore, the remaining fragments point to the possibility that compound 4-C1 is similar in structure to 4'-methoxy-3,3',5-stilbenetriol 3-glucoside (Figure 50), which itself is very similar to resveratrol (since it exhibits a resveratrol base unit). The presence of a fragment at m/z 227 (which matches the molecular weight of resveratrol) also supports this theory. Other possibilities for m/z 419, although less likely, are the compounds (or derivatives of) 3'-benzyloxy-5,7-dihydroxy-3,4'-dimethoxyflavone (Figure 51) and isopomiferin (Figure 52).

The second fragmentation scheme scenario would present the molecular ion at m/z 359, which is the most intense ms2 ion. This is possible since the most intense ion in the normal mass spectra is m/z 419, which would indicate that the 60 amu loss associated with m/z 359 is possibly an

- acetate [M+CH3COO] adduct ion (and thereby possibly confirming m/z 359 as the molecular ion instead of m/z 419). If this is in fact the case, then the m/z 359 parent ion dissociated to form the ms3 fragments mentioned above. Based on this dissociation pattern (and the number of confirmed methoxy groups), a possible candidate for compound 4-C1 would be 3',5,7- trihydroxy-4',5',6-trimethoxyisoflavone (or a derivative of), also known as irigenin (Figure 53).

According to the NIST 05 database, other possible candidates include the following methoxy- flavonoids: 3-hydroxy-3’,4’,5,7-tetramethoxy-flavanone, 3-hydroxy-3’,4’,7,8-tetramethoxy- flavanone, and 5,6,3’-trihydroxy-3,7,4’-trimethoxy-flavone (a.k.a. oxyanin B). One last scenario cannot be ruled out for the identification of compound 4-C1. The ms2 spectra shows a fragment of m/z 289, with further ms3 fragments of m/z 245, 227, 205, 203, 187, 161, and 151. This

105 follows the exact fragmentation scheme for catechin and/or epicatechin, which could point to compound 4-C1 being this flavonoid base unit, with an unknown 130 amu side-group.

Compound 5-C1 at RT 13.37 (min) exhibited both a low UV and ion intensity response.

The ion at m/z 399 (M-H)- dissociated to yield an ms2 ion of m/z 339, with a difference

- indicating acetate [M+CH3COO] via a neutral loss of 60 amu. Therefore, the m/z 339 is most

likely the molecular ion. Another intense ms2 ion of m/z 177 (M-H)- was observed, which

reveals the presence of a glucose unit due to a loss of 162 amu from the m/z 339 molecular ion.

Further dissociation yielded ms3 fragmentation which confirmed the presence of a glucose unit

(m/z = 162), due to m/z fragments of 134, 120, and 90 (which are common fragments produced by cross-ring cleavages of sugars). Dissociation of m/z 177 (M-H)- revealed ms3 fragmentation

patterns which could indicate the possible presence of (one of) many different compounds.

Various derivatives of methoxy-cinnamic acids, benzoic acids, propionic acids, and propenoic

acids are all possible candidates. Examples of these compounds include the following (or similar

derivatives): chroman-2-carboxylic acid, 3-chromane-carboxylic acid, and 4-methoxycinnamic

acid. Furthermore, the NIST 05 Database revealed a high probability that this compound is in

fact 6,7-dihydroxycoumarin-6-glucoside (a.k.a. esculin). This is also supported by predicting the

possible fragments generated by common structural cleavages during dissociation. The presence

of this compound would not be out of the ordinary, since it is commonly found in the skins and

bark of chestnuts. Other possible derivatives of this compound, which could also prove to be

compound 5-C1, include 6,7-dihydroxycoumarin-7-glucoside (a.k.a. cichoriin) and 7,8-

dihydroxycoumarin 7-β-D-glucoside (a.k.a. daphnetin-7-O-glucoside, or simply daphnin).

Compound 6-C1 at RT 14.42 (min) exhibited an extremely low UV response, and

relatively low ion intensity. Ions in the normal mass spectrum include the most intense ion of

106 m/z 437 (M-H)-, and a low intensity ion at m/z 518 (indicating a sodium acetate aduct). The

molecular ion at m/z 437 dissociated to form two major ms2 fragments at m/z 347 and 289. The

m/z 289 fragment conveys a neutral loss of 148 amu, which could be indicative of a trans-

cinnamic acid acyl group. This is further supported by the 90 amu loss associated with the m/z

347 fragment, which would indicate a C7H6 loss common to cinnamic acids. Further dissociation

of m/z 289 resulted in ms3 ions which follow the fragmentation scheme for catechin and/or epicatechin. Therefore, this compound is tentatively identified as a catechin (or epicatechin), with a trans-cinnamic acid side-group. An additional compound (compound 9-C1) with the same

molecular ion and fragmentation pattern exists at RT 15.17 min. The presence of two

compounds (at different retention times) with identical MS fragmentation characteristics is

possibly due to the existence of trans-cinnamic acid side-groups residing at different positions on

the flavonoid aglycone.

Compound 7-C1 at RT 14.88 (min) exhibited a low UV and ion intensity. The molecular

ion at m/z 641 (M-H)- dissociated to yield ms2 ions of m/z 479 and 317, which signify the loss of

two 162 amu glucose units. Further dissociation of the m/z 479 daughter ion resulted in ms3

fragments with m/z 317, 300, 299, 209, 191, and 176. The m/z 317 fragment indicates the

presence of the second glucose unit (i.e., its 162 amu difference from m/z 479), and also reveals

the molecular weight of the flavonoid aglycone. The remaining ions follow the most likely

fragmentation patterns for the following m/z 317 (M-H)- flavonoid compounds: 3,3',4',5,5',7-

hexahydroxyflavone (a.k.a. myricetin), 3,3',4',5,6,7-hexahydroxyflavone (a.k.a. quercetagetin)

(Figure 54), and 3,3',4',5,7,8-hexahydroxyflavone (a.k.a. gossypetin). Similarities in the ion fragmentation of these three compounds were also confirmed using the NIST 05 Mass Spectral

107 Database. Therefore, compound 7-C1 is most likely one of the three aforementioned compounds, with a diglucoside side-group.

Compound 8-C1 at RT 15.06 (min) showed both low UV and ion intensity, with the most intense ion at m/z 319 (M-H)-. Other ions found in the normal mass spectrum include m/z 401

- - and 378, indicating sodium acetate [(M-H)+CH3COONa] and acetate [(M-H)+CH3COOH]

adducts respectively, which confirmed m/z 319 as the molecular ion. The m/z 319 molecular ion

dissociated to form ms2 ions at m/z 287, 197 (the most intense ion), and 165 (M-H)-. The m/z

197 ms2 daughter ion dissociated to form an m/z 165 ion in the ms3 spectra. The m/z 287 ms2

fragment indicates the loss of 32 amu, which is commonly indicative of a methoxy group.

Further losses indicate fragmentation which follows the dissociation pattern for the compound

oureatacatechin (Figure 55), which has a similar structure to catechin and epicatechin, but with

an added methoxy group.

Compound 10-C1 at RT 15.24 (min) exhibited both a strong UV response and high ion

intensity. The molecular ion, which exists at m/z 639 (M-H)-, was confirmed by the presence of

- - acetate [M+CH3COO] and sodium acetate [(M-H)+CH3COONa] adducts at m/z 699 and 722

respectively. Dissociation of the molecular ion resulted in m/z 477 and 315 ms2 product ions.

As in the case with compound 7-C1 at RT 14.88, this was due to the loss of two glucose units

(162 amu each). Further fragmentation resulted in ms3 ions with m/z 315, 300, and 271. This

follows the fragmentation patterns for the flavonoid isorhamnetin, as seen in previous the

standard for this compound. In addition, the common presence of isorhamnetin among various

plant types makes it the most likely aglycone for compound 10-C1. However, the ms3 ion

fingerprint could also result from the fragmentation of several other possible compounds

including 6-methoxyluteolin (aka nepetin), 4’-methoxyquercetin (a.k.a. tamarixetin), and 7-

108 methylquercetin (a.k.a. rhamnetin) (Figures 56). In addition to isorhamnetin, the NIST 05

Database also revealed close matches for other methoxy-flavones, including 3’,4’,5,6- tetrahydroxy-7-methoxy-flavone (a.k.a. pedalitin), which further supports the theory that the aglycone is in fact a methoxy-flavonoid. Unfortunately, the MSn fragmentation patterns for these compounds are nearly identical, and the position of the methoxy group cannot be determined without performing NMR analysis. In addition, the lack of standards for these compounds prevented positive identification via UV rentention times. Therefore, compound 10-

C1 is tentatively identified to contain one of the aforementioned flavonoid aglycones, with a diglucoside side-group.

Compound 11-C1 at RT 15.56 (min) gave both a low UV response and ion intensity. The

- - molecular ion was noted at m/z 551 (M-H) due to an acetate [M+CH3COO] adduct at m/z 611.

The most intense ms2 ion formed included m/z 389. The neutral loss difference between the molecular ion (m/z 551) and the m/z 389 daughter ion was equal to 162 amu, revealing the presence of glucose. Further dissociation of m/z 389 yielded an ms3 ion of m/z 227, which also gave a neutral loss of 162 amu, signifying a second glucose unit. It is possible that the remaining m/z 227 is resveratrol. Other ms3 ions would also indicate such a compound (including m/z 341,

193, and 161), as noted with the fragmentation scheme of resveratrol-3-β-mono-D-glucoside

(Figure 57). Therefore, it is hypothesized that Compound 11-C1 consists of a reveratrol aglycone base unit, with a diglucoside side-group. This dissacharides is most likely either laminaribiose (3-O-beta-D-glucosyl-d-glucose) or gentiobiose (6-O-beta-D-glucosyl-D-glucose),

since they are the most common diglucoside side-groups found on flavonoid and stilbene

aglycones.

109 Compound 12-C1 at RT 19.70 exhibited both a low UV and ion intensity response. A molecular ion was discovered at m/z 757 (M-H)-, and confirmed by a sodium acetate [(M-

- - H)+CH3COONa] adduct at m/z 839 (82 amu neutral loss) and a chlorine [Cl ] adduct at m/z 793.

Dissociation of the molecular ion produced an ms2 ion of m/z 595, which designates a

glucose(galactose) sugar via a 162 amu difference. Further fragmentation in the ms3 spectra

gave rise to daughter ions of m/z 463, 301, 300, 271, and 255. The neutral loss associated with

the difference between m/z 595 and 463 equals 132 amu, identifying the presence of a xylose (or

arabinose) sugar. An additional neutral loss found between m/z 463 and 301 shows a difference

of 162 amu, pinpointing the presence of another glucose (or galactose) unit. The ms3 ion

fragments of m/z 301, 300, 271, and 255 are indicative of quercetin with an O-glycosidic bond,

as shown by the standard obtained for this compound. This compound is most likely a quercetin-

based monomer, plus the following trisaccharide: glucose (or galactose) + xylose (or arabinose)

+ glucose (or galactose) (or simply stated, either a sambubiose or lathyrose sugar plus an

additional glucose/galactose).

Compound 13-C1 at RT 19.89 (min) exhibited a relatively strong UV response, but low

ion intensity. A molecular ion was confirmed at m/z 477 (M-H)- due to a sodium acetate [(M-

- H)+CH3COONa] adduct ion at m/z 559 (resulting in an 82 amu neutral loss), and an acetate [(M-

- H)+CH3COOH] adduct ion at m/z 536 (59 amu neutral loss). Further dissociation of the

molecular ion yielded an ms2 spectra with the most intense ion of m/z 315, and a low intensity

ion at m/z 300. The neutral loss associated with m/z 315 equals 162 amu, revealing the presence

of a single glucose unit. The continued dissociation of m/z 315 gave rise to fragments at m/z

300, 271, 256, and 151 (in the ms3 ion spectra). As seen by the fragmentation patterns of

obtained standards, this dissociation scheme closely follows the ion spectra for the flavonoid

110 isorhamnetin. Therefore, it is hypothesized that this compound is most likely isorhamnetin-3- glucoside. However (as mentioned above), although the presence of the ms3 daughter ions follow a striking similarity to that of isorhamnetin, it is possible that the flavonoid aglycone may also be either 6-methoxyluteolin (a.k.a. nepetin), 4’-methoxyquercetin (a.k.a. tamarixetin), or 7- methylquercetin (a.k.a. rhamnetin) (since they also show similar if not identical fragmentation schemes to that of isorhamnetin). In addition to isorhamnetin, the NIST 05 Database also revealed close matches for other methoxy-flavones, including 3’,4’,5,6-tetrahydroxy-7-methoxy- flavone (a.k.a. pedalitin), which further supports the theory that the aglycone is in fact a methoxy-flavonoid (although, based on its common appearance in many plants, isorhamnetin is the most likely aglycone).

Compound 14-C1 at RT 20.03 showed a relatively strong UV response, however, it exhibited a low ion intensity in the normal mass spectra. The most intense ion was seen at m/z

177 (M-H)-, which is assumed to be the molecular ion (although no adduct ions were present).

The molecular ion dissociated to give ms2 fragments of m/z 151, 149, 133, 109, and 89 (with

133 being the most intense ion fragment). Further fragmentation of m/z 133 yielded ms3

fragments of m/z 123, 105, 104, 91, and 89. Based on the fragmentation pathway (Figure 58),

there is a strong possibility that this compound is a methoxycinnamic acid (such as 4-

methoxycinnamic acid). However, the fragmentation scheme also closely follows that of a

chroman-carboxylic acid derivative (chroman-2-carboxylic acid) (Figure 59). The possible

presence of this compound makes sense, since chroman serves as part of the base structure for

many flavonoid compounds. According to the NIST 05 Database, other strong candidates for

compound 14-C1 include dihydroxy-coumarins (including derivatives such as 6,7-

dihydroxycoumarin and 7,8-dihydroxycoumarin).

111 Compound 15-C1 at RT 20.81 exhibited both a strong UV response and ion intensity, which would indicate a high concentration of this particular compound in Fraction C. The molecular ion was confirmed in the normal mass spectra at m/z 741 (M-H)- by the presence of

- - several adduct ions including: chlorine [M+Cl ] at m/z 776 (35 amu), acetate [M+CH3COO] at

- m/z 801 (60 amu), and sodium acetate [(M-H)+CH3COONa] at m/z 822. The m/z 741

molecular ion underwent collision-induced dissociation (CID) to yield an ms2 spectra with ion

fragments of m/z 609, 591, 475, 301, 300, 271, and 255. Further dissociation of m/z 300 led to

m/z 271, 255, 179, 169, and 151 ion fragments in the ms3 spectra. The ms2 spectra showed the

most intense ion at m/z 300, revealing a neutral loss of 440. This 440 amu difference indicates the presence of a trisaccharide consisting of one 6-carbon pentose sugar (rhamnose at 146 amu),

one 5-carbon pentose sugar (xylose or arabinose at 132 amu), and one 6-carbon hexose sugar

(glucose or galactose at 162 amu). Fragmentation pathways in both the ms2 and ms3 ion spectra

indicate the possible presence of a quercetin base structure. Therefore, this compound is likely

to be quercetin with a trisaccharide side-group containing xylose(arabinose) + rhamnose + glucose(galactose). This is in agreement with previous research by (Lou et al., 2001) who

reported the presence of quercetin-3-O-[2-O-β-glucopyranosyl-6-O-α-rhamnopyranosyl]-β-

glucopyranoside at the same molecular weight (and confirmed the stereochemistry via NMR).

Compound 16-C1 at RT 21.29 exhibited an extremely weak UV signal strength and low ion abundance. The molecular ion was noted at m/z 595 (M-H)-, confirmed by the presence of a

chlorine [M+Cl-] adduct at m/z 631. Dissociation of the m/z 595 parent ion yielded an ms2

spectra with ion fragments of m/z 445, 301 (most intense), 300, 271, and 255. The 294 amu

neutral loss associated with (the difference between) m/z 595 and 301 reveals the presence of

either the dissacharide sambubiose (2-O-beta-D-xylosyl-D-glucose) or lathyrose (2-O-beta-D-

112 xylosyl-D-galactose). Further fragmentation of m/z 301 yielded an ms3 spectra with ion fragments of m/z 271, 255, 179, 169, and 151. These daughter ions point to the likely presence of quercetin. Therefore, this compound is most likely quercetin-3-arabinoglucoside (a.k.a. peltatoside), due to the quercetin aglycone base-unit with the dissacharide side-group (Figure

60). However, several other compounds exist which have the ability to follow a similar fragmentation scheme, including eriodictyol-7-neohesperidoside (a.k.a. neoeriocitrin) and eriodictyol-7-O-rutinoside (a.k.a. eriocitrin) (Figure 12).

Compound 17-C1 at RT 21.95 exhibited an extremely weak UV intensity, and an extremely low relative ion abundance. A molecular ion was noted at m/z 771 (M-H)- in the

normal mass spectra. This molecular ion was confirmed by the presence of several adduct ions

- - at m/z 806 (chlorine [M+Cl ]), 831 (acetate [M+CH3COO] ), and 853 (sodium acetate [(M-

- H)+CH3COONa] ). The molecular ion dissociated via CID to form ms2 ions with m/z 609 (most

intense ion), 463, and 301. The noted 162 amu neutral loss difference between the molecular ion

(m/z 771) and main ms2 fragment (m/z 609) indicates the presence of glucose (or galactose). In

addition, the loss between ms2 ion fragments m/z 609 and m/z 463 show the presence of a

rhamnose sugar (146 amu). Further loss between ms2 fragments m/z 463 and m/z 301 (M-H)- point to the presence of both a second glucose unit (due to additional loss of 162 amu), and the possible existence of quercetin. The ms2 ion spectra was searched using the NIST (National

Institute of Standards and Technology) Mass Spectral Database, and the software returned a 99% probability match delphinidin-3-rutinoside. However, comparing the ms2 spectra of compound

17-C1 to the delphinidin standard revealed differences which rule out this anthocyanin as a possible match. Fragmentation of daughter ion m/z 609 led to ms3 ion spectra with fragments m/z 301, 300, 271, and 255. These daughter ions most likely confirm the presence of a quercetin

113 aglycone. Therefore, compound 17-C1 is tentatively identified as quercetin with the following trisaccharide side-group: glucose (or galactose) + rhamnose + glucose (or galactose). The most likely configuration (based on the formation seen in other flavonoid glycosides) would be a rutinoside sugar, plus an added glucose unit.

Compound 18-C1 at RT 23.68 exhibited both an extremely high UV intensity and ion abundance, indicating a very high concentration of this particular compound. Ions produced in the normal mass spectra include the most intense ion of m/z 609 (M-H)-, and a less intense ion at

- m/z 691, which indicates the presence of a sodium acetate [(M-H)+CH3COONa] adduct ion (due

to the 82 amu neutral loss). Therefore, the molecular ion in this case is confirmed at m/z 609.

Dissociation of the molecular ion formed ms2 spectra fragments of m/z 301, 300, 271, and 255.

The 308 amu neutral loss difference between the m/z 609 molecular ion and the m/z 301 ms2 ion reveal the presence of a rutinose disaccharide side-group (6-O-alpha-L-rhamnosyl-D-glucose).

Dissociation of the most intense ms2 ion (at m/z 301/300) yielded ms3 fragments of m/z 271,

255, 179, and 151, which indicate the presence of quercetin with an O-glycosidic bond

(confirmed via the standard). Therefore, this compound is most likely a quercetin aglycone with a rutinoside sugar side-group, and the most probable candidate for such a compound is rutin (or quercetin-3-O-rutinoside).

Compound 19-C1 at RT 26.19 exhibited both a very high UV intensity and relative ion abundance. The molecular ion was confirmed at m/z 593 (M-H)-, with a sodium acetate [(M-

- H)+CH3COONa] adduct ion at m/z 676, and an acetate adduct ion at m/z 653. The parent

molecular ion dissociated to form an ms2 daughter fragment of m/z 285. This fragment was the

result of a 308 amu neutral loss, which indicates the presence of a rutinose sugar (6-O-alpha-L-

rhamnosyl-D-glucose). Further dissociation of m/z 285 yielded the following fragments in the

114 ms3 spectra: m/z 267, 257, 255, 243, 241, 239, 229, 213, 199, 197, 169, 163, 153, 151, and 135.

This fragmentation pattern follows the pathway for the compounds luteolin and keampferol (both of which have an M-H- m/z of 285). According to the standards, there are only trace differences

between the fragmentation schemes of these two compounds. This makes sense, since the only

structural difference is the placement of a single hydroxyl unit (C-ring position 3 for kaempferol,

versus B-ring position 3’ for luteolin). Both compounds exhibited ions at m/z 267, 257, 243, and

213, however, luteolin yields high intensity ions at m/z 241, 217, 199, 197, and 175 (which these

fragments are either not seen in kaempferol, or not seen at a high intensity). In the case of

kaempferol, ions exist (which are either not seen in luteolin, or at least not in high intensity) at m/z 255, 239, 229, and 169. By looking at the ms3 fragments for compound 19-C1, it can be seen that fragmentation attributes of both luteolin and kaempferol exist, making positive identification difficult. Since the fragmentation fingerprint matches that of both the luteolin and kaempferol standard, it is theorized that these two rutinose-linked flavonoid aglycones are simultaneously co-eluting. This isn’t completely unlikely, since luteolin and kaempferol are very

similar structures, and although their monomers elute at different retention times on a C-18

column, the added rutinose dissacharide side-group added to both structures may react to a C-18

column in nearly identical fashions. Therefore, compound 19-C1 is tentatively identified as

either luteolin and/or kaempferol with an added rutinoside disaccharide side-group. Possible

candidates include the following (one of each luteolin and kaempferol based structures):

kaempferol-7-O-neohesperidoside, luteolin-7-O-neohesperidoside, kaempferol-3-O-rutinoside,

luteolin-3-O-rutinoside, kaempferol-7-rutinoside (Figure 61), and luteolin-7-rutinoside.

Compound 22-C1 at RT 26.79 exhibited an extremely intense UV response at 280nm,

and a mass spectra with an extremely high ion abundance. This represents the structure with the

115 highest concentration in Fraction C, and confirms that a very high quantity of this compound is found in peanut skins. The normal mass spectra showed the most intense ion at m/z 623 (M-H)-, with low intensity adduct ions at m/z 659 (M+Cl-), 682 (acetate), and 705 (sodium acetate).

These adducts confirmed a molecular ion at m/z 623. The molecular ion was further confirmed

by the presence of a low intensity dimer ion at m/z 1247, which commonly forms during

electrospray ionization. The appearance of such dimers is commonly used as an added technique for identifying molecular ions. The m/z 623 molecular ion dissociated to yield the following

fragments in the ms2 spectra: m/z 315 (most intense ion), 300, 271, 255, and 243. The 308 amu

neutral loss difference between m/z 623 and 315 reveal the presence of a rutinose (6-O-alpha-L-

rhamnosyl-D-glucose) disaccharide sugar. Further dissociation of m/z 315 produced a spectra

with ms3 ions of m/z 300, 271, and 255. This fragmentation scheme follows the pattern for the

isorhamnetin, however (as previously mentioned), the ms3 ion fingerprint could also result from

the fragmentation of several other possible compounds including 6-methoxyluteolin (a.k.a.

nepetin), 4’-methoxyquercetin (a.k.a. tamarixetin), and 7-methylquercetin (a.k.a. rhamnetin)

(Figure 56). The most likely candidate for compound 22-C1 is isorhamnetin-3-O-rutinoside

(Figure 62), since most sugars attach at the 3 position on the flavonoid C-ring (Cuyckens and

Claeys, 2004). Other possible (but less likely) candidates include 4’-methylquercetin-7-O-

rutinoside, iso-rhamnetin-3-O-neohesperidoside, and iso-rhamnetin 3-O-robinobioside.

Compound 23-C1 at RT 28.46 gave both low UV and ion intensity. The most intense ion

was found at m/z 407 (M-H)-, and was confirmed to be the molecular ion by the presence of an

- acetate [(M-H)+CH3COOH] adduct ion at m/z 466 (59 amu loss), and a sodium acetate [(M-

- H)+CH3COONa] adduct ion at m/z 489 (82 amu loss). The m/z 407 parent ion dissociated to

form an ms2 spectra with ion fragments of m/z 389, 379, 363, 297, 285 (most intense ms2 ion

116 fragment), 283, 269, 257, 256, 255, and 243. The m/z 285 fragment further dissociated to yield ms3 ion fragments of m/z 257, 255, 241, 213, 201, and 175. Several possible scenarios exist regarding the identification of this compound via prediction of the fragmentation scheme. In the first scenario, the most intense ion in the ms2 spectra (m/z 285) produced a neutral loss difference of 122 amu from the m/z 407 parent ion, which could possibly indicate the presence of a benzoic acid derivative side-group. This is further supported by the m/z 363 fragment, which represents a neutral loss of 44 amu, and indicates the possible loss of a carboxylic acid.

Ion fragments in the ms3 spectra indicate the possible existence of either a luteolin or keampferol flavonoid base unit (based on fragmentation patterns of their standards). Therefore, this compound may possibly be either one of the aforementioned flavonoids, with an added benzoic acid acylated side-group. A second scenario exists based on the dissociation scheme, which would possibly identify the compound as remangiflavanone (Figure 63). A third prediction for the possible identification of compound 23-C1 is sophora-flavanone (Figure 64).

Compound 24-C1 at RT 28.75 exhibited a weak UV response, and a low ion relative abundance. Another compound at RT 29.66 (compound 25-C1) exhibited identical fragmentation patterns to compound 24-C1, but demonstrated nearly double the ion abundance.

The most intense ion in the normal mass spectra for both compounds was m/z 283 (M-H)-, and is assumed to be the molecular ion (although no adduct ions exist). In both cases, the molecular ion dissociated to form an m/z 268 fragment in the ms2 spectra, which the 15 amu loss is most likely caused by the loss of a methyl group (and typically indicates the presence of a methoxy group). Further fragmentation led to an ms3 spectra with ion fragments of m/z 240, 239, 224, and 211. According to obtained standards, these ion fragments closely follow the exact dissociation pathways for 5,7-dihydroxy-4'-methoxyisoflavone (aka biochanin A or genistein 4-

117 methyl Ether), 4',5-dihydroxy-7-methoxyisoflavone (aka ), and 5,7-dihydroxy-4'- methoxyflavone (aka or apigenin-4'-methyl ether). However, it appears to more closely follow the ion fragmentation for both biochanin A and prunetin, since acacetin appears to have a much weaker 223/224 fragment than the other two compounds. Although compounds 24-C1 and

25-C1 are most likely one of the two aforementioned compounds, the fragmentation scheme

(with the m/z 268 daughter ion) also compares to numerous other flavonoid monomers at m/z

283 (M-H)- which exhibit similar structural characteristics. These compounds include the

following: 5-dihydroxy-7-methoxyflavone (a.k.a. 7-methoxyapigenin), 4',7-dihydroxy-6-

methoxy-isoflavone (a.k.a. ), 5,6-dihydroxy-7-methoxyflavone (a.k.a. negletein or

balcalein-7-methyl ether) (Figure 56), 5,7-dihydroxy-6-methoxyflavone (a.k.a. oroxylin), and

5,7-methoxyflavanone. In addition, the NIST 05 database gave high probability spectra matches

for methoxy flavones, including 5,7-dihydroxy-8-methoxyflavone (a.k.a. vogonin) and 4’,5-

dihydroxy-7-methoxyflavone (a.k.a. ). Also, the existence of an m/z 268 daughter ion

has the potential to follow the fragmentation scheme for chalcone-based compounds, including

2'-hydroxy-3,4-dimethoxychalcone and 2'-hydroxy-4',6'-dimethoxy-chalcone (Figure 66).

Unfortunately, standards for these compounds were not used for this research (preventing

positive identification by means of matching UV retention times), therefore, compounds 24-C1

and 25-C1 can only be tentatively identified via fragmentation data from past literature, and/or

MSn interpretation through fragmentation prediction.

Compound 26-C1 at RT 32.68 exhibited both a low UV response and ion relative

abundance. Another similar compound at RT 33.43 had the same molecular weight and

fragmentation characteristics, although, it showed a slightly higher UV response (and an

intermediate-ranged ion relative abundance). In both cases, a molecular ion was identified at m/z

118 - - 299 (M-H) due to the presence of an acetate [M+CH3COO] adduct at m/z 359 and a sodium

- acetate [(M-H)+CH3COONa] adduct at m/z 381. The molecular ion dissociated to form ms2

spectra with an intense ion at m/z 284 (for both compounds). The 15 amu loss (methyl group)

between the molecular ion and the ms2 daughter fragment indicates the possible presence of a

methoxy group. In the case of compound 26-C1, further dissociation formed ms3 ion fragments

of m/z 284/283, 266, 256, 240, 228/227, 212/211, 200, 193, 183, 165, 150, and 137. For the

other compound at RT 33.43, further fragmentation of m/z 284 yielded ms3 fragments of m/z

267, 256, 239, 227, 215, 199, 173, and 151. Although the dissociation patterns are extremely

similar, slight variations in fragment ions (along with slightly different retention times) reveal

that these are in fact two separate compounds. Dissociation pathways based on obtained

fragmentation data of standards for this study indicate a high probability that the compound is

5,7,4'-trihydroxy-3'-methoxy-flavone (a.k.a. chrysoeriol), specifically with regards to the m/z

284, 256, and 151 fragments. However, it has also been shown that chrysoeriol (Figure 34)

exhibits dissociation patterns which are nearly identical to that of diosmetin (aka luteolin-4'-

methyl ether). In addition, these two compounds elute at nearly the same retention times under

similar chromatographic conditions, which supports the theory that the other compound may

possibly be diosmetin. Although fragmentation data for the standards of these compounds was

obtained, definitive matches are only possible by running the actual standards to determine

retention times (which were not available for this research). Although these compounds are most

likely the two aforementioned structures, the lack of definitive proof for identification opens the

possibility that these compounds may in fact be other flavonoid or chalcone compounds, which

exhibited similar fragmentation pathways (specifically with regards to the m/z 284 and 227

daughter ions). Possible flavonoid compounds include: 3',5,7-trihydroxy-4'-methoxyisoflavone

119 (a.k.a. pratensein), 4',5,7-trihydroxy-6,8-dimethylflavanone (a.k.a. farresol or cyrtopterinetin),

4',5,7-trihydroxy-6-methoxyisoflavone (a.k.a. tectorigenin) (Figure 35), psi-tectorigenin, 4'- hydroxy-5,7-dimethoxyflavanone, 8-methoxy-iso-scutellargin, 3,7-dihydroxy-2-(4-hydroxy-3- methoxyphenyl)-4-benzopyrone (a.k.a. geraldol), and kaempferol 4'-methyl ether (a.k.a. kaempferide). Possible chalcone compounds include: 2',4'-dihydroxy-3',6'-dimethoxychalcone,

2',4-dihydroxy-4',6'-dimethoxychalcone (Figure 67), 2',6'-dihydroxy-3',4'-dimethoxychalcone, and 2',6'-dihydroxy-4,4'-dimethoxychalcone.

Compound 27-C1 at RT 34.15 (min) showed a low UV intensity, but an intermediate- range relative ion abundance. A molecular ion was found with an m/z of 329 (M-H)-, and a

- sodium acetate [(M-H)+CH3COONa] adduct ion at m/z 411 (82 amu difference). The parent ion

fragmented to form m/z 314 (most intense ion) and 299 in the ms2 spectra. These two fragments

each correspond to a 15 amu loss, each indicative of a methyl (CH3) group, and most likely

reveal the presence of two methoxy groups. Further dissociation yielded the following fragments

in the ms3 spectra: m/z 299, 286, 285, 271, 257, and 243. Based on obtained standards for this

study, the compound isorhamnetin-3-methoxy follows this specific fragmentation scheme

(Figure 68). Although this structure is most likely the aforementioned compound, other

possibilities do in fact exist based on the dissociation pathway (specifically with regards to

fragments m/z 314 and 299). Since it would be difficult to determine the positions of the two

methoxy groups with MS/MS alone, one possibility (for other compound matches) is the

placement of the methoxy groups at different positions. Examples of such compounds are 3',4',5-

trihydoxy-6,7-dimethoxyflavone (a.k.a. circiliol), irisflavone A (Figure 69), irisflavone D, and

3',5,7-trihydroxy-3',4'-dimethoxyflavone (a.k.a. quercetin 3,4'-dimethyl ether). The NIST 05

Database revealed similar compounds (based on matching similar spectra), which supports the

120 theory of a dimethoxy-flavonoid. These compounds include: 5,6,4’-trihydoxy-7,8-dimethoxy- flavone, 5,8,4’-trihydoxy-6,7-dimethoxyflavone, 7,3’,4’-trihydoxy-5,6-dimethoxyflavone,

3',4',5-trihydoxy-3,7-dimethoxyflavone (a.k.a. quercetin 3,7-dimethyl ether), 3',4',5-trihydoxy-

3’,7-dimethoxyflavone (a.k.a. quercetin 3’,7-dimethyl ether), and 4',5,7-trihydroxy-3,3'- dimethoxyflavone (a.k.a. quercetin 3,3'-dimethyl ether). Another possible structure would be similar to those mentioned above, however, with one added methoxy group (and therefore, one less hydroxyl group). Examples of these compounds include kaempferol-3,7,4'-trimethylether, and kaempferol-3,4',7-trimethyl ether (Figure 70). One final possibility with regards to similar dissociation patterns exists with a chalcone-based structure. An example of this compound is

3',6'-dihydroxy-2',4',5'-trimethoxychalcone (Figure 71), or a derivative of this compound.

Compound 28-C1 at RT 39.41 (min) exhibited an extremely low UV and ion intensity.

The normal mass spectra revealed two intense ions at m/z 369 and 339, which makes it difficult to identify the molecular ion. Pinpointing the molecular ion becomes even more difficult given

- that each of these intense ions appear to exhibit attached sodium acetate [(M-H)+CH3COONa]

adducts at m/z 421 and 451. The most logical explanations would be either the presence of two

co-eluting compounds, or perhaps collision-induced dissociation before or after the electrospray

source. The most likely scenario is the latter, since the ms2 and ms3 spectra appear to contain

fragments from only one compound. In addition, the difference in molecular weight between the

two intense ions found in the normal mass spectra is 30 amu, which would indicate the presence

of a methoxy group (and could easily dissociate via CID). Therefore, it is believed that the molecular ion is most likely m/z 369, which underwent source CID to present an m/z 339 fragment in the normal mass spectra. The molecular ion then yielded ms2 ion fragments of m/z

295, 251, 245, 233, 219, and 177. Upon losing the methoxy group (and therefore yielding m/z

121 339), the m/z 177 fragment indicates the loss of a 162 amu glucose unit. Further dissociation produced ms3 ion fragments of m/z 209, 191, 177, 176, 175, 151, 150, 134, and 133. Based on the fragmentation scheme (Figures 72), compound 28-C1 is tentatively identified as fraxetin-8- glucoside (a.k.a. fraxin or fraxoside).

Compound 29-C1 at RT 39.85 (min) exhibited a low UV intensity, and an intermediate- level ion relative abundance. The normal mass spectra revealed an intense m/z 353 (M-H)- ion, and a less intense ion at m/z 435. The 82 amu difference between these two ions reveals the

- presence of a sodium acetate adduct [(M-H)+CH3COONa] , which verifies that m/z 353 is a

molecular ion. Dissociation of the parent ion led to the following daughter ions in the ms2 spectra: m/z 335, 309, 297 (most intense ion), 284, 253, 240, and 175. The m/z 297 daughter

fragment further dissociated to form ms3 ions of m/z 269, 256, 255, 253, 227, 225, 211, and 161.

The molecular weight matches that of chlorogenic acid, however, the UV retention time of

compound 29-C1 does not correlate to the standard for this phenolic acid. In addition, although

the dissociation pathway shows some similarity to that of the chlorogenic acid standard, the

differences in ion fragmentation rules out this compound as a possible match. The fragmentation

scheme shows a striking similarity to other compounds which don’t have the same molecular weight as compound 29-C1, which could give insight into its structural characteristics. The

dissociation pathway of the flavonoid chrysoeriol (and other flavonoids with MWs of 300 which

have methoxy-groups) yields very similar ions to this structure, including its major fragments of

m/z 284, 256, 255, and 227. The NIST 05 Mass Spectral Database also revealed several

flavonoid compounds with methoxy-groups (although with different MW’s) which had similar

ion fragments. These compounds include: 2-(2-acetoxy-3-methoxyphenyl)-3-methoxy-4H-

chromen-4-one, 7-hydroxy-3-methoxy-2-p-methoxyphenyl-4H-chromen-4-one, and 2-(3-

122 hydroxy-4-methoxyphenyl)-3-methoxy-4H-chromen-4-one. Although compound 29-C1 is definitely not one of these aforementioned compounds (since they have different molecular weights), the similarities in fragmentation schemes indicate that not only is this structure a flavonoid-based compound, but it also may be a derivative of one of these compounds. A NIST

05 search also identified two flavonoid-based compounds with MW’s which matched that of compound 29-C1, with fragmentation patterns which were similar. These compounds were the following: 7-hydroxy-2-(4-hydroxyphenyl)-5-methoxy-8-(3-methyl-2-butenyl)-2,3-dihydro-4H- chromen-4-one, and 3,5,7,8-tetra-hydroxy-6-(3-methyl-2-butenyl)-2-phenyl-4H-chromen-4-one.

Another possibility for the identification of compound 29-C1 is a structure similar to glycycoumarin (although, with one less methyl group), since the fragmentation pathways are comparable (Figure 46). There are several other flavonoid-based compounds which exhibit even more similarities (than any of the previously mentioned compounds) to their fragmentation schemes when compared to this compound. These structures include glycyrrh-iso-flavone, iso- licoflavonol, lico-iso-flavanone, and lico-iso-flavone A. An example of the predictive fragmentation pathway for these compounds can be seen by lico-iso-flavone A in Figure 73.

Therefore, it is theorized that compound 29-C1 is most likely one of these four compounds.

Three closely eluting compounds (at RT’s 40.71, 42.72, and 43.39), with identical molecular weights, exhibited extremely similar fragmentation pathways. First, compound 30-C1 at RT 40.71 showed an extremely weak UV intensity, and a low ion relative abundance. The molecular ion was confirmed at m/z 367 (M-H)- due to the presence of an acetate [(M-

- H)+CH3COOH] adduct at m/z 426. Dissociation of the molecular ion yielded ms2 fragments of

m/z 352, 349, 337 (most intense ion), 322, and 296. Further fragmentation of the m/z 337

daughter ion yielded ms3 fragments of m/z 322 and 296. Second, compound 32-C1 at RT 42.72

123 exhibited an intermediate-ranged UV intensity, and a high ion relative abundance. The m/z 367

- - (M-H) molecular ion was identified by the presence of a sodium acetate [(M-H)+CH3COONa] adduct at m/z 449 (82 amu difference). The m/z 367 molecular ion dissociated to yield ion fragments of m/z 352 (the most intense ion), 312, 309, and 297. The m/z 352 fragment dissociation to form the following ions in the ms3 spectra: m/z 338, 324, 309, 297, 281, 271,

253, and 161. Finally, compound 33-C1 at RT 43.39 exhibited a low UV intensity, and an intermediate-ranged ion relative abundance. The m/z 367 (M-H)- molecular ion was identified

by the presence of the two following adduct ions: Chlorine [M+Cl-]- at m/z 404 (37 amu

- difference), and sodium acetate [(M-H)+CH3COONa] at m/z 449 (82 amu difference). The

molecular ion dissociated to form an intense m/z 352 ion in the ms2 spectra. Other low-intensity

ions were observed at m/z 309 and 297. The m/z 352 ms2 daughter ion dissociated to form ms3

fragments of m/z 309 and 297. Although slight differences can be seen in the fragmentation

patterns of these three compounds, the dissociation pathways exhibit characteristics which

closely resemble that of glycycoumarin (Figure 46), and glycyrrh-iso-flavanone.

Compound 31-C1 at RT 41.58 exhibited both an extremely low UV intensity and ion

relative abundance. The normal mass spectra showed an intense m/z 281 (M-H)- ion, and low

- intensity ions at m/z 341 and 363. The two low-intensity ions represent acetate [M+CH3COO]

- and sodium acetate [(M-H)+CH3COONa] respectively, which in turn verifies the molecular ion

at m/z 281. The molecular ion dissociated to form an intense m/z 266 ion fragment in the ms2

spectra. The loss associated with m/z 266 (from the parent m/z 281 molecular ion) is indicative

of a methyl group fragment (15 amu), and most likely reveals the presence of a methoxy group.

Further dissociation yielded an ms3 spectra with m/z 266, 238, 237, 210, 196, 194, and 138. The

282 amu molecular weight could indicate many compound possibilities, including dimethoxy-

124 flavones, dimethoxy-isoflavones, hydroxy-methoxy-methylflavones, and hydroxy-methoxy- methyl-isoflavones. The m/z 266 ms2 ion fragment follows the dissociation pathway of almost every one of these compound classes, since initial fragmentation yields the loss of a methyl group. However, the remaining fragments mostly follow the dissociation scheme of a dimethoxyflavone, of which there are many types. This includes the following dimethoxyflavones: 2',3'-dimethoxyflavone, 2',4'-dimethoxyflavone, 3,5-dimethoxy-flavone,

3,7-dimethoxyflavone, 3',4'-dimethoxyflavone, 3',6-dimethoxyflavone, 4',5-dimethoxyflavone,

4',6-dimethoxyflavone, 4',7-dimethoxyflavone, 6,7-dimethoxy-flavone, and 7,8-dimethoxy- flavone. Of these dimethoxyflavones, the compounds which most closely following the fragmentation scheme of compound 31-C1 is 2',3'-dimethoxyflavone (Figure 74) and 7,8- dimethoxyflavone.

Two compounds, compound 34-C1 at RT 47.29 and compound 35-C1 at RT 48.12, exhibited identical molecular weights, and nearly identical fragmentation patterns. Compound

34-C1 exhibited an intermediate-ranged UV intensity, and a very high ion relative abundance.

Conversely, compound 35-C1 exhibited both an extremely high UV intensity and ion relative abundance. The molecular ion for both compounds was identified at m/z 351 (M-H)- due to the

presence of a chlorine [M+Cl-]- adduct ion at m/z 387, and a sodium acetate [(M-

- H)+CH3COONa] adduct ion at m/z 433. The molecular ion in both compounds dissociated to form ms2 fragments of m/z 336 (most intense ion), 296, and 281 (although, compound 35-C1 had a higher intensity m/z 281 ion). The m/z 336 ms2 ion was the result of a 15 amu loss (CH3), and is a common indicator of the presence of a methyl and/or methoxy group. In the case of compound 34-C1, the m/z 336 daughter ion underwent further dissociation to yield ms3 ions of m/z 321, 317, 307, 293 (the most intense ion), 281 (second most intense ion), 267, 253, 243, 237,

125 219, 213, and 188. On the other hand, compound 35-C1 yielded ms3 ions of m/z 321, 318, 308,

293, 281, 266, 253, 249, 227, and 216. In contrast (to compound 34-C1), compound 35-C1 gave a more intense ms3 ion at m/z 281, and the second most intense ion at m/z 293, which comparatively flip-flopped ion intensities in the ms3 spectra. The nearly identical dissociation pathways (and the close retention times) indicate that these compounds are practically identical, although, they likely have functional groups at different locations on the main structures.

According to the NIST 05 database, two possibilities exist regarding the identification of these compounds. First, the dicoumarin known as daphnoretin not only shares the same molecular weight (352 amu), but also exhibits extremely similar fragmentation patterns. A second compound, known as 5-O-methyllupiwighteone, also has the same molecular weight and fragmentation characteristics. Two other compounds exist at this molecular weight which would follow extremely similar predictive fragmentation patterns: pongagallone A (Figure 47), and sophora-iso-flavone A.

Fraction C - HPLC Method 2

The use HPLC Method 2 exhibited better overall separation (than method 1), and in addition to all the compounds identified in method 1, allowed the detection of compounds which were not otherwise seen in the first method. UV chromatograms and TICs for each scan range can be seen in Figures A.44 through A.45. A total of 34 additional compounds were identified in peanut skin Fraction C obtained from Toyopearl SEC (Table 6). In addition to compounds which were undetectable using the first method, method 2 often allowed the separation of multiple compounds with identical molecular weight and fragmentation attributes (i.e., compounds which only eluted once using method 1). The most viable explanation for this

126 phenomenon is the presence of compounds with similar structural characteristics (while having the same molecular weight), or perhaps due to side-groups residing at different locations on the base-structure. For example, compound 3-C1 at RT 8.51 (MW = 436 amu) eluted once using method 1, however, the enhanced separation exhibited in method 2 revealed the presence of two compounds with these identical molecular weights and dissociation attributes (at RT’s 18.67 and

20.87). In the case of HPLC Method 2, these compounds exhibited a high UV intensity, but a somewhat low ion relative abundance. The fragmentation characteristics (as explained in HPLC

Method 1 for Fraction C) reveal that these compounds may be two different catechin and/or epicatechin base units, with rhamnoside groups residing at different locations on the flavonoid aglycone. Possible candidates for each of these two compounds include the following: catechin-

3-O-rhamnoside, catechin-5-O-rhamnoside, or catechin-7-O-rhamnoside (or their epicatechin derivatives) (Figure 49).

Another instance in which HPLC Method 2 offered better separation involves compound

8-C1 (MW = 320, or m/z 319 M-H-), found at RT 15.06 with HPLC Method 1. The utilization

of HPLC Method 2 enabled the separation of two distinct compounds (as opposed to one of these

320 amu compounds in Method 1) with the same monoisotopic mass and fragmentation pattern

as compound 8-C1. When using Method 2, both of these compounds (compound 2-C2 at RT

27.34, and compound 4-C2 at 30.29) exhibited a high UV intensity and ion relative abundance.

As explained above, compound 8-C1 (HPLC Method 1) was identified as oureatacatechin, which

has a similar structure to catechin and epicatechin, but with an added methoxy group. The most

likely explanation for the presence of two of these compounds (with identical MW’s and

fragmentation pathways) when utilizing Method 2, is the presence of the methoxy group at a

different location on one of the flavonoid rings. Method 2 also revealed the presence of a third

127 m/z 319 (M-H)- compound, although, the fragmentation patterns were different than the oureatacatechin-like compounds. This compound (compound 9-C2 at RT 37.80) exhibited a low

UV intensity and ion relative abundance. The molecular ion was confirmed at m/z 319 based on the presence of adducts at m/z 355 (chlorine), 378 (acetate), and 401 (sodium acetate). The molecular ion dissociated to form the following ion fragments in the ms2 spectra: m/z 302, 289,

287, 286, 275, and 273. Although there are some similarities in the possible fragmentation

patterns of the compound known as 4’-O-methyl-epigallocatechin, is it more likely that this

compound is possibly 3,3',4',5,5',7-hexahydroxyflavanone (a.k.a. dihydromyricetin) (Figure 75).

A third example of multiple eluting structures (with identical molecular weights) exists

with compound 26-C1 at RT 32.68 (MW = 300, or m/z 299 M-H-) using HPLC Method 1. The

use of HPLC Method 2 allowed the separation of four distinct compounds with this exact

molecular weight and similar fragmentation scheme (as opposed to only one compound detected

in method 1). This included the following compounds: compound 23-C2 at RT 59.12 (with low

UV intensity, and low ion abundance), compound 25-C2 at RT 62.67 (with low UV intensity,

and low ion abundance), compound 27-C2 at RT 68.47 (with low UV, and intermediate ion

abundance), and compound 29-C2 at RT 70.39 (with low UV, and intermediate ion abundance).

Based on the fragmentation patterns, at least two of these compounds are most likely chrysoeriol

and diosmetin. Therefore, the remaining two 300 amu compounds which exist in this fraction may each be one of the following compounds (based on their dissociation schemes): pratensein, farresol (or cyrtopterinetin), tectorigenin, psi-tectorigenin, 4'-hydroxy-5,7-dimethoxy-flavanone,

8-methoxy-iso-scutellargin, geraldol, kaempferide, 2',4'-dihydroxy-3',6'-dimethoxychalcone,

2',4-dihydroxy-4',6'-dimethoxychalcone, 2',6'-dihydroxy-3',4'-dimethoxy-chalcone, and 2',6'- dihydroxy-4,4'-dimethoxychalcone. As explained previously, the similarities in fragmentation

128 patterns make it difficult to determine exactly which (out of the group listed) are actually present, and only the use of specific standards can achieve proper confirmation via UV retention time (or

NMR analysis).

One final example of how HPLC Method 2 exhibited better separation can be seen with compound 27-C1 at RT 34.15 (MW = 330 amu, or m/z 329 M-H-) when using HPLC Method 1.

In the case of HPLC Method 2, a total of three compounds with this exact molecular weight and

fragmentation scheme were found (compound 24-C2 at RT 62.07, compound 28-C2 at RT 69.72,

and compound 30-C2 at RT 71.54). Compound 24-C2 exhibited a low UV intensity and ion

relative abundance, however, compounds 28-C2 and 30-C2 both exhibited low UV intensity with

intermediate-to-high relative ion abundance. According to obtained standards, at least one of

these three compounds is most likely isorhamnetin-3-methoxy (Figure 68). Each of the other

two structures are possibly one of the following compounds: 3',4',5-trihydoxy-6,7-

dimethoxyflavone (a.k.a. circiliol), irisflavone A (Figure 69), irisflavone D, 3',5,7-trihydroxy-

3',4'-dimethoxyflavone (a.k.a. quercetin 3,4'-dimethyl ether), kaempferol-3,7,4'-trimethylether, kaempferol-3,4',7-trimethyl ether, and 3',6'-dihydroxy-2',4',5'-trimethoxychalcone. Another m/z

329 (M-H)- compound was found at RT 49.77 (compound 20-C2), and although it exhibited

similar fragmentation patterns to the previous compounds, differences in the dissociation

pathway were noted. Compound 20-C2 revealed an extremely low UV intensity and ion relative

- abundance. An acetate [M+CH3COO] adduct ion was found at m/z 389, which confirmed the

m/z 329 molecular ion. The molecular ion dissociated to form ms2 fragments of m/z 314, 312,

301, 296, 281, 257, 255, 206, 193 (the most intense ion fragment), 178, 164, and 150. The m/z

193 ion underwent further dissociation to yield ms3 ion fragments of 178, 177, 165, 164, and

150. Although many of the aforementioned compounds produce a number of these fragments

129 during dissociation, the compound irisflavone D can predicatively produce virtually all of these ions (Figure 76).

The use of HPLC Method 2 allowed the detection of many compounds which were not found using Method 1. The first of these structures is compound 1-C2 at RT 21.18, which exhibited a high UV intensity, but a low relative ion abundance. The molecular ion was

- - identified at m/z 327 (M-H) by the presence of an acetate [M+CH3COO] adduct ion at m/z 387,

- and a sodium acetate [(M-H)+CH3COONa] adduct ion at m/z 409. The molecular ion

dissociated to form ms ions of m/z 309, 289, 283, 267, 165 (most intense), and 161. The most

intense ms2 ion was found at m/z 165, which reveals a neutral loss of 162 amu (from the m/z 327 parent ion), and indicates the presence of a glucose/galactose unit. Further dissociation of the

m/z 165 ion formed fragments 141, 137, and 121. The most likely scenario for the identification

of compound 1-C2 is the presence of an m/z 165 (M-H)- compound, with a sugar side-group. A

possible candidate of such a compound, which has both a 166 amu MW and follows the ms3

fragmentation pattern, is any derivative of o-hydrocoumaric acid (Figure 77). Therefore,

compound 1-C2 may possibly be an O-glycosidic bound derivative of such an acid. Two other

possible structures exist which exhibit these fragmentation characteristics (and also have sugar

conjugates), although the aglycones are not acids. These compounds include androsin and

betuloside (Figure 78). Additionally, compound 1-C2 may also possibly be one of these

aforementioned compounds (or a derivative thereof).

Another compound which was detectable in Method 2 (but not Method 1) was compound

3-C2 at RT 29.22. This compound exhibited both an intermediate-ranged UV intensity and ion

abundance. An intense ion was found in the normal mass spectra at m/z 641 (M-H)-, and is assumed to be the molecular ion (although no adduct ions were found). The parent ion

130 underwent collision-induced dissociation (CID) to yield ms2 fragments of m/z 479 (the most intense ion), 477, and 315. The m/z 477 and 315 fragments indicate the loss of two sugar units

(due to the 164 and 162 amu losses), revealing the presence of a rhamnose and glucose sugar. In addition, the combined loss (326 amu) reveals the presence of a rutinose disaccharide (6-O- alpha-L-rhamnosyl-D-glucose). Further dissociation of the m/z 479 ms2 fragment produced the following ions in the ms3 spectra: m/z 317, 315 (most intense ion), 300, 299, 257, 256, and 214.

The flavonoid aglycone in this case is m/z 315 (M-H)-, and according to the standards, the dissociation pathway closely follows that of isorhamnetin. This would tentatively identify this compound as isorhamnetin with an O-rutinoside conjugate. Unfortunately, the position of the disaccharide cannot be determined, although, in most cases they reside at the 3 position on the flavonoid C-ring. Although the flavonoid base unit is most likely isorhamnetin (based on literature data), the ms3 ion fingerprint could also result from the fragmentation of several other possible compounds with molecular weights of 316 amu. These compounds include 6- methoxyluteolin (a.k.a. nepetin), 4’-methoxyquercetin (a.k.a. tamarixetin), and 7-methyl-

quercetin (a.k.a. rhamnetin). These compounds differ in their methoxy group location on the

flavonoid structure. Unfortunately, the MSn fragmentation patterns for these compounds are nearly identical, and the position of the methoxy group cannot be determined without performing

NMR analysis. In addition, the lack of standards for these compounds prevented positive identification via UV retention times.

Compound 5-C2 at RT 33.06 was also found using HPLC Method 2 specifically, and exhibited a high UV intensity and intermediate ion abundance. An intense m/z 611 (M-H)- ion can be seen in the normal mass spectra, although no adduct ions were present. The most intense ion in the ms2 spectra was m/z 551 (a neutral loss of 60 amu from m/z 611), and suggests that

131 - - the difference is an acetate [M+CH3COO] adduct. Therefore, this identifies m/z 551 (M-H) as the molecular ion. Other fragments found in the ms2 spectra include m/z 389 and 227. The ion fragment m/z 389 represents a neutral loss of 162 amu (glucose/galactose), and an additional

(162 amu) sugar is lost as indicated by ion fragment m/z 227. One structure which could exhibit these fragmentation characteristics is a compound called monotropein (who’s aglycone moiety is m/z 227), although, with an added sugar unit. Another possibility for the identification of compound 5-C2, is that the remaining m/z 227 is the compound resveratrol. An example of such a compound is resveratrol-3-β-mono-D-glucoside (Figure 57). Therefore, in this case, compound

5-C2 would be identified as resveratrol, with a disaccharide side-group.

Although HPLC Method 1 was unable to detect any compounds with a 304 amu molecular weight, HPLC Method 2 was able to identify numerous structures within this monoisotopic mass range. The first of these compounds, compound 6-C2 at RT 33.34, exhibited an intermediate UV intensity and low ion relative abundance. A molecular ion was confirmed at

- - m/z 303 (M-H) by the presence of an acetate [M+CH3COO] adduct at m/z 363. The molecular parent ion dissociated to form ms2 fragments of m/z 285, 271, 259 (most intense), 244, 219, 217,

204, 163, and 137. Further dissociation of the m/z 259 ms3 ion led to the following fragments in the ms3 spectra: m/z 246, 244, and 228. Another compound with a nearly identical fragmentation scheme was found at RT 40.54. It also showed an intermediate UV intensity and low ion relative abundance. The m/z 303 (M-H)- molecular ion for this particular compound was

- verified by the presence of an acetate [M+CH3COO] adduct at m/z 363. Based on the spectra of

obtained standards, these two compounds follow the exact fragmentation pathway for both 3’-O-

methylcatechin (Figure 79) and/or 4’-O-methylcatechin. A third m/z 303 (M-H)- compound

(compound 8-C2) was detected at RT 36.16, and exhibited an intermediate UV intensity and low

132 ion relative abundance. Compound 8-C2 had similar (but not identical) fragmentation characteristics to that of compound 6-C2 at RT 33.34. The m/z 303 (M-H)- molecular ion was

- confirmed by the presence of an acetate [M+CH3COO] adduct at m/z 363. The molecular ion

dissociated to form the following fragments in the ms2 spectra: m/z 287 (the most intense

fragment), 275, 266, 261, 259, 217, 215, and 179. The m/z 287 ion found in the ms2 spectra

underwent further dissociation to yield ms3 fragments of m/z 272 and 259. Although this

compound does not appear to be either 3’-O-methylcatechin or 4’-O-methylcatechin, it does

however closely follow the fragmentation pathway of other possible m/z 303 flavonoid

compounds (which was confirmed using the NIST 05 Mass Spectral Database). These compounds include dihydroquercetin (a.k.a. taxifolin) (Figure 80) and dihydrorobinetin, which are similar in structure (except for the placement of a single hydroxyl group). Several other m/z

303 (M-H)- compounds which also closely follow the fragmentation scheme of taxifolin and

dihydrorobinetin were found at RT 41.25 (min) and RT 42.86 (min). Both compounds exhibited an intermediate UV intensity and low ion relative abundance, and had molecular ions confirmed

- by the presence of acetate [M+CH3COO] adducts at m/z 363. In both cases, the ms2 spectra

yielded ion fragments of m/z 285, 276, 275, 274, 271, 259, 257, 245, 241, 219, 218, 199, 179,

and 167 (which are consistent with taxifolin and dihydrorobinetin). The final m/z 303 (M-H)-

compound was detected at RT 66.73 (min). It exhibited an extremely low UV intensity and ion

relative abundance. The molecular ion was confirmed by the presence of a sodium acetate [(M-

- H)+CH3COONa] adduct ion at m/z 385. The m/z 303 parent ion dissociated to form ms2

fragments of m/z 285 (the most intense ion), 271, 260, 256, 244, 227, 223, and 185. The m/z

285 ms2 ion further dissociated to form an ms3 spectra with ions of m/z 257, 241, 198, and 197.

133 This fragmentation scheme closely follows the dissociation pathway for the methoxy-flavonoid compound known as 3-methoxy-3',4',5,7-flavantetrol (a.k.a. meciadanol) (Figure 81).

HPLC Method 2 revealed the presence of compound 7-C2 at RT 34.11, which exhibited a

low UV intensity and ion relative abundance. An m/z 287 (M-H)- molecular ion was identified

- by the presence of the following two adduct ions: acetate [M+CH3COO] at m/z 347 (60 amu

- difference), and sodium acetate [(M-H)+CH3COONa] at m/z 369 (82 amu difference). Based on the 288 amu molecular weight, it would be expected that this compound would in fact be eriodictyol (since it is a common flavonoid). However, the fragmentation pathway and retention time does not match that of the eriodictyol standard. The m/z 287 (M-H)- molecular ion of

compound 7-C2 dissociated to form ms2 fragment ions of m/z 269, 259, 245, 243 (most intense

ion), 227, 205, 197, and 165. The m/z 243 ms2 ion underwent further dissociation to form ms3

fragments of m/z 227, 215, 203, 199, and 173. Compounds which may follow this fragmentation

scheme include 2,3-dihydrofisetin (a.k.a. fustin), 6-hydroxyapigenin, and dihydrokaempferol

(Figure 82). The NIST 05 Database also gave a high probability match for the flavanone compound fustin.

A compound was detected at RT 35.51 (min) which exhibited a low UV intensity, but a

- - high ion relative abundance. Acetate [M+CH3COO] and sodium acetate [(M-H)+CH3COONa] adduct ions were found m/z 227 and 249 (M-H)- respectively, confirming m/z 167 (M-H)- as the

molecular ion. The m/z 167 parent ion in the normal mass spectra dissociated to form the

following ms2 fragments: m/z 152, 109, and 108. No ms3 fragments were detected. Several

168 amu (MW) compounds exist which could possibly yield these fragment ions. According to

the NIST 05 database, the most likely match for this compound is 4-hydroxy-3-methoxybenzoic

acid (a.k.a. vanillic acid), which is an intermediate in the formation of from ferulic acid

134 (Civolani et al, 2000; Lesage-Meessen et al., 1996). NIST 05 also revealed a high probably search hit for 3-hydroxy-4-methoxybenzoic acid (a.k.a. isovanillic acid). According to the NIST

05 database, the fragmentation patterns are also similar to that of 2-hydroxy-3-methoxybenzoic acid (a.k.a. o-vanillic acid), which is also an intermediate product in the two-step bioconversion of ferulic acid to vanillin (Lesage-Meessen et al., 1996). Additionally, NIST 05 also yielded search hits for homoprotocatechuic acid, 2-hydroxy-6-methoxybenzoic acid, , and phloracetophenone. Predictive fragmentation also revealed compounds which are similar in structure to homogentisic acid (Figure 9), 6-methyl-β-resorcylic acid, and galloacetophenone.

Numerous compounds with molecular weights of 332 (m/z 331 M-H-) were detected

using HPLC Method 2, although, none were found using Method 1. Compound(s) 10-C2

consists of 12 total structures found at RTs 39.40, 40.17, 40.72, 41.42, 41.99, 42.68, 43.19,

44.02, 44.53, 44.84, 54.32, and 54.91. These compounds exhibited both intermediate-ranged UV

intensities and relative ion abundances. The m/z 331 (M-H)- for each of these compounds was

- - - verified by either the presence of a chlorine [M+Cl ] or acetate [M+CH3COO] adduct ion. The twelve (nearly identical) structures which make up compound 10-C2 exhibited similar fragmentation schemes, although, slight variations in ms2 and ms3 daughter ion fragments were noted. This could indicate that these are in fact different compounds, as opposed to simply one slowly eluting compound (because of its affinity for the C-18 column). All compounds from RT

39.40 to 40.72 (min) dissociated to form the following ion fragments in the ms2 spectra: m/z

301, 300, 299 (most intense ion), 271, and 255. However, each of these compounds also exhibited their own other unique (less intense) ions. When comparing the unique ions among

these compounds, the numbers in fact differed by a factor of 2 amu or less. This was also the

case with the ms3 spectra, which although every compound revealed an m/z 271 fragment, they

135 also each exhibited other unique fragments (which were again very close regarding m/z). In the case of compounds detected from RT 41.42 through 44.84, each of these structures revealed the presence of m/z 301, 300, and 299 ms2 fragments, and an m/z 271 ms3 fragment. However, each of these compounds again showed their own other unique ion fragments in both the ms2 and ms3 spectra (which again only differ by several amu). This supports the concept (just as with the previously detected compounds from RT 39.40 to 40.72) that each of these structures are in fact only slightly different. In some cases regarding these spectra, a small m/z 315 ion fragment appears indicating a neutral loss which could reveal the presence of a methyl group

(and therefore, possibly a methoxy group). In all cases, the presence of an m/z 299 fragment shows a neutral loss of 32 amu, which most likely indicates the existence of a methoxy group.

Also, the presence of an m/z 301 fragment shows the difference of 15 amu, revealing the loss of two methyl groups (and therefore, possibly two methoxy groups). A flavonoid with a 332 amu molecular weight was identified in a prior study (Lou et al., 1999), which could possibly follow this overall fragmentation pathway. The presence of this compound, known as 3’4’5-tri-O- methylcatechin, would explain the various common ion fragments shown with these compounds.

However, the occurrence of other unique fragments revealed in each of these individual compounds, could possibly be explained by the presence of the O-methyl groups residing at different locations on the flavonoid structure. Conversely, the compounds detected at RTs 54.32 and 54.91(min) exhibited a much different fragmentation scheme. In both cases, the m/z 331 molecular ion dissociated to form major ms2 fragments of m/z 316 and 314. However, each of these compounds formed other unique ms2 ion fragments (which only differed by 2 amu or less).

This again reveals compounds which have the same molecular weight, but only slightly different fragmentation pathways. The m/z 316 ms2 fragments shown in both of these structures indicate

136 the loss of 15 amu, and could reveal the presence of a single methoxy group. According to the

NIST 05 database, a compound known as 5,6,8,3’,4’-pentahydroxy-7-methoxyflavone exhibits a similar fragmentation fingerprint (and also has only one methoxy group). Therefore, one of these two compounds may be the aforementioned structure, and the other compound a close derivative of this structure (in which the methoxy group is attached at a different location on the flavonoid base unit).

Two compounds (compound 11-C2 and 16-C2) were found using HPLC Method 2 which exhibited similar fragmentation patterns, although, their molecular weights were not identical.

First, compound 11-C2 was found at RT 39.65, and gave an intermediate-ranged UV intensity, but low relative ion abundance. The molecular ion was identified at m/z 937 (M-H)-, and

- confirmed by the presence of an acetate [M+CH3COO] adduct at m/z 997. The parent ion

dissociated to form the following daughter fragments in the ms2 spectra: m/z 775 (the most

intense ion), 579, 547, 417, and 389. The m/z 775 ms2 ion represents a neutral loss of 162 amu

from the m/z 937 parent ion, and signifies the presence of a glucose (or galactose) sugar group.

Further fragmentation of the m/z 775 ion fragment led to an ms3 spectra with ions of m/z 579

(the most intense ion), 389, and 372/371. The structure responsible for the loss of 196 amu

between the m/z 775 (ms2) and 579 (ms3) main fragments could not be determined, although, it

is most likely an unknown phenolic acid group. The remaining fragments which correspond to

the m/z 579 compound follow a specific fragmentation path which is similar to that of

amentoflavone-4',4'',7-trimethyl ether (a.k.a. sciatopitysin) (Figure 83). Sciatopitysin is a dimer

composed of a hydroxy-dimethoxy-flavonoid, and a dihydroxy-methoxy-flavonoid. The

presence of such a compound is not surprising, given that many methoxy-flavonoids have been

identified in this fraction alone. Therefore, compound 11-C2 is tentatively identified as

137 sciatopitysin, with both a sugar and (possibly) a phenolic acid side-group. Next, compound 16-

C2 was found at RT 44.75, and exhibited both an intermediate UV intensity and ion relative abundance. A molecular ion for this structure was identified at m/z 775, and was confirmed by a chlorine [M+Cl-]- adduct at m/z 811. The m/z 775 parent ion dissociated to form ms2 ions of m/z

579 (most intense ion), 389, and 372. As in the case of compound 11-C2, the reason for the 196

amu neutral loss between m/z 775 and 579 is unknown. Also, the fragmentation scheme of m/z

579 is identical to that shown with compound 11-C2 above. It is theorized that the base structure

for compound 16-C2 is also amentoflavone-4',4'',7-trimethyl ether (a.k.a. sciatopitysin) (Figure

83), with an unknown phenolic acid side-group.

Compound 12-C2 at RT 40.97 exhibited an intermediate UV intensity with a low relative

ion abundance. An intense molecular ion was found in the normal mass spectra at m/z 479 (M-

- - H) , and was confirmed by the presence of an acetate [M+CH3COO] adduct at m/z 539 (a 60

amu difference). The molecular ion underwent collision-induced dissociation (CID) to yield an

m/z 317 daughter ion in the ms2 spectra. This daughter fragment represents a 162 amu

difference, and indicates the presence of a sugar side-group. The m/z 317 ion further dissociated

to produce the following fragments in the ms3 spectra: m/z 302, 300, 299, 287, 283, 272, 259,

257, 255, 179, 169, 151, and 149. The remaining ions follow the most likely fragmentation

patterns for the following m/z 317 (M-H)- flavonoid compounds: 3,3',4',5,5',7-

hexahydroxyflavone (a.k.a. myricetin), 3,3',4',5,6,7-hexahydroxyflavone (a.k.a. quercetagetin),

and 3,3',4',5,7,8-hexahydroxy-flavone (a.k.a. gossypetin) (Figure 84). Similarities in the ion

fragmentation of these three compounds were also confirmed using the NIST 05 Mass Spectral

Database. Therefore, compound 12-C2 is most likely one of the three aforementioned

compounds, with a glucoside (or ) side-group.

138 Compound 14-C2 at RT 42.42 exhibited an intermediate UV Intensity, but a low ion relative abundance. The molecular ion of m/z 903 (M-H)- was confirmed by the presence of a

- sodium acetate [(M-H)+CH3COONa] adduct ion at m/z 985 (an 82 amu difference). The

molecular ion underwent collision induced dissociation (CID) to yield an m/z 741 fragment in

the ms2 spectra. This indicated a neutral loss of 162 amu, which revealed the presence of a

glucose (or galactose) unit. The m/z 741 daughter ion further dissociated to form the following

fragments in the ms3 spectra: m/z 609, 301, 300, 271, and 255. A neutral loss of 132 amu (from

the m/z 741 ms2 fragment) is indicated by the occurrence of the m/z 609 fragment, and reveals

the existence of a xylose sugar. The combined loss between the parent ion (m/z 903) and the m/z

609 fragment indicates a 294 amu neutral loss, which shows the presence of either a sambubiose

(2-O-beta-D-xylosyl-D-glucose) or lathyrose (2-O-beta-D-xylosyl-D-galactose) disaccharide

sugar. The continued loss between m/z 609 and m/z 301 yields a difference of 308 amu, which

indicates the presence of a rutinose (6-O-alpha-L-rhamnosyl-D-glucose) disaccharide. The

remaining fragments of the m/z 301 flavonoid base unit signify the dissociation pathway for

quercetin. From the m/z 609 point in the dissociation pathway, the fragmentation pattern exactly

follows that of rutin (a.k.a. quercetin-3-O-rutinoside). Therefore, compound 14-C2 is tentatively

identified as rutin, plus either an additional sambubiose or lathyrose disscacharide side-group.

Compound 15-C2 at RT 43.77 revealed an intermediate UV intensity, but a low relative

ion abundance. The normal mass spectra showed an intense ion at m/z 757 (M-H)-, and low

- intensity ions at m/z 817 and 839 (indicating the presence of a 60 amu acetate [M+CH3COO]

- and 82 amu sodium acetate [(M-H)+CH3COONa] adduct ions). These adducts confirmed the

molecular ion at m/z 757 (M-H)-, which underwent collision-induced dissociation (CID) to yield

an m/z 595 ms2 ion fragment. This daughter ion indicates a neutral loss of a 162 amu sugar unit

139 (glucose or galactose). Further dissociation produced ions of m/z 301, 300, 271, and 255 in the ms3 spectra, which the fragmentation pattern is indicative of quercetin. The 294 amu neutral loss between m/z 595 in the ms2 spectra, and the m/z 301 in the ms3 spectra, indicate the presence of either a sambubiose or lathyrose dissacharide. Therefore, compound 15-C2 consists of a quercetin flavonoid base-unit, with a trisaccharide side-group made up of a sambubiose (or lathyrose) + glucose (or galactose) sugar.

The use of HPLC Method 2 allowed the identification of three distinct flavonoid compounds with 318 amu molecular weights. First, a compound was detected at RT 43.70 (min) which exhibited an intermediate UV intensity, but a low relative ion abundance. The m/z 317

(M-H)- molecular ion was identified by the presence of the following two adduct ions: a 36 amu

- - - chlorine [M+Cl ] at m/z 353, and a 60 amu acetate [M+CH3COO] at m/z 377. The molecular

ion dissociated to form the following ms2 ions: m/z 300 (most intense), 288, 272, 243, 191, and

181. The m/z 300 ms2 ion further dissociated to form an m/z 151 ms3 fragment. This follows the possible fragmentation pathway for the flavonoid known as 3,3',4',5,5',7-hexahydroxyflavone

(a.k.a. myricetin). These ion fragments for myricetin were also verified using the NIST 05 Mass

Spectral Database. Next, compound 17-C2 at RT 45.85 exhibited both an intermediate UV intensity and ion relative abundance. The m/z 317 (M-H)- molecular ion was confirmed by the

- - - existence of a chlorine [M+Cl ] adduct at m/z 353, an acetate [M+CH3COO] adduct at m/z 377,

- and a sodium acetate [(M-H)+CH3COONa] adduct at m/z 399. The molecular ion dissociated to yield the following ion fragments in the ms2 spectra: m/z 299 (most intense ion), 289, 284, 271,

245, 232, 193, 181, and 179. The m/z 299 ms2 daughter ion formed the following fragments in the ms3 spectra: m/z 284, 271, 255, 243, 231, 227, and 215. The dissociation pathway agrees with the predictive fragmentation scheme for the compound 3,3',4',5,7,8-hexahydroxyflavone

140 (a.k.a. gossypetin) (Figure 84). The ion fragmentation fingerprint for this compound was also verified using the NIST 05 Database. Finally, a compound was found at RT 47.51 which revealed both an intermediate UV intensity and ion relative abundance. The m/z 317 (M-H)-

molecular ion was confirmed by the presence of the following adduct ions: chlorine [M+Cl-]- at

- - m/z 353, acetate [M+CH3COO] at m/z 376, and sodium acetate [(M-H)+CH3COONa] at m/z

399. The m/z 317 (M-H)- parent ion dissociated to yield an intense m/z 289 fragment in the ms2 spectra. Other ions seen in the ms2 included m/z 273, 245, 231, 207, 193, and 167. These ions follow the possible fragmentation scheme for the compound known as 3,3',4',5,6,7- hexahydroxyflavone (a.k.a. quercetagetin). As with the previous 318 amu compounds, the spectra was also verified using the NIST 05 Mass Spectral Database. The presence of these three structures supports the theory previously proposed for the identification of compound 7-C1

(Fraction C, HPLC Method 1) at RT 14.88, which appears to exhibit one of the three aforementioned compounds as a flavonoid aglycone base-unit.

Compound 18-C2 at RT 47.34 showed a m/z 347 (M-H)- molecular ion in the normal

- mass spectra, and was confirmed by an acetate [(M-H)+CH3COOH] adduct at m/z 407. The

molecular ion dissociated to form the following ion fragments in the ms2 spectra: m/z 332, 319,

316, 315 (the most intense ion), 285, 271, 231, 219, 193, and 165. Further dissociation of m/z

315 yielded an ms3 spectra with the following ions: m/z 299, 285, 271, and 229 (most intense

ion). The dissociation pathway is similar to that of the tetrahydroxy-methoxy flavonoids with

316 amu molecular weights, such as isorhamnetin. However, when compared to isorhamnetin,

the 348 amu molecular weight of this compound would suggest it exhibites an extra methoxy

group on the flavonoid structure (denoted by the 32 amu difference). This is further supported

by ion fragments at m/z 316 and 285, which represent two consecutive 31 amu losses, and are

141 common indicators of existing methoxy groups. Therefore, compound 18-C2 may be a previously undiscovered tetrahydroxy-dimethoxy-flavonoid compound.

Compound 19-C2 at RT 47.92 exhibited both an intermediate UV intensity and ion

relative abundance. The m/z 771 (M-H)- molecular ion was confirmed by the existence of a

- sodium acetate [(M-H)+CH3COONa] adduct at m/z 853. The m/z 771 parent ion underwent

collision-induced dissociation (CID) to yield m/z 609 (most intense ion) and 463 fragments in

the ms2 spectra. The m/z 609 fragment indicates the loss of 162 amu, revealing the presence of a

glucose (or galactose) sugar unit. In addition, the m/z 463 fragment shows the existence of a

rhamnose sugar unit (due to the 146 amu neutral loss). The m/z 609 ms2 fragment underwent

further dissociation to yield ms3 ions of m/z 301, 300, and 271, which are indicative of the

flavonoid quercetin. The m/z 301 ms3 fragment indicates a neutral loss of 308 amu (from the

m/z 609 ms2 daughter ion), which reveals the presence of a rutinose (6-O-alpha-L-rhamnosyl-D-

glucose) dissacharide. Therefore, compound 19-C2 is a quercetin-trisaccharide which consists of

quercetin as the flavonoid base-unit, with an additional rutinose + glucose (or galactose) side-

group.

Compound 21-C2 at RT 52.34 exhibited both a low UV intensity and ion relative

abundance. The m/z 343 (M-H)- molecular ion was confirmed in the normal mass spectra by the

presence of a chlorine [M+Cl-]- adduct at m/z 379 (36 amu difference). The molecular ion

dissociated to form the following ion fragments in the ms2 spectra: m/z 329, 327, 315 (most

intense ion), 313, 297, 282, 271, 229, 227, 165, and 149. Further dissociation of the m/z 315

ms2 fragment yielded the following ions in the ms3 spectra: m/z 297, 273, 271, 247, 229, 227,

206, 177, and 165. Several flavonoid compounds exist which could follow this dissociation pathway, all of which are dihydroxy-trimethoxy-flavones. These compounds include 3',5-

142 dihydroxy-4',6,7-trimethoxyflavone (a.k.a. eupatorin), 5,7-dihydroxy-3',4',5'-trimethoxy-flavone, gossypetin 3,3',4',7-tetramethyl ether (Figure 85), irisflavone B, and xanthomicrol. Of these compounds, the predictive dissociation pathway reveals fragments which yield the most similarities to that of 5,7-dihydroxy-3',4',5'-trimethoxyflavone and irisflavone B. The NIST 05

Database also confirmed that the fragmentation characteristics (shown above) conform to that of various dihydroxy-trimethoxy-flavones.

Two distinct compounds with 356 amu molecular weights were found using HPLC

Method 2, both of which had similar fragmentation characteristics. First, compound 22-C2 at

RT 54.72 exhibited both a low UV intensity and ion relative abundance. The m/z 355 (M-H)- molecular ion was confirmed by the presence of a chlorine [M+Cl-]- adduct at m/z 391 (a 36 amu

difference). The molecular ion dissociated to form numerous ion fragments in the ms2 spectra,

including m/z 340, 339, 337, 327, 323, 313, 312, 298, 292, 285, 231, 219 (the most intense ms2

fragment), 218, 195, and 193. The m/s 219 daughter fragment underwent further dissociation to

yield the following ions in the ms3 spectra: m/z 204, 191, 190, 177, and 176. The m/z 285

fragment in the ms2 spectra reveals a neutral loss of 70 amu, and could indicate the presence of a

C5H10 (3-methylbutyl) side-group. The remaining fragments appear to be the dissociation

pathway of an m/z 285 flavonoid compound. A candidate for such a compound is

cudraflavanone B (or a derivative thereof) (Figure 86), which exhibits a total of four hydroxy and

one methylbutyl group [4’,5,6’,7-tetrahydroxyflavanone-6-(3-methylbutyl)]. A similar

compound with nearly identical fragmentation characteristics was verified using the NIST 05

Database, although, that particular flavonoid structure exhibited slightly different structural characteristics. That compound, known as 5-methoxy-4’,7-dihydroxy-flavanone-8-(3- methylbutyl), shows two hydroxy and one methoxy group, instead of the four hydroxy groups

143 exhibited by cudraflavanone B (that is, in addition to the 3-methylbutyl side-group existing on both compounds). One last compound (or derivative of) exists which could possibly exhibit a similar dissociation pathway to compound 22-C2, although, its prevalence is less likely due to fewer matching ion fragments (via interpretation of possible ion fragments). This compound is known as 2-carbethoxy-5,7-dihydroxy-4'-methoxyisoflavone. Next, compound 31-C2 at RT

74.49 exhibited a low UV intensity, but intermediate relative ion abundance. The m/z 355 (M-

H)- molecular ion was confirmed by the presence of a chlorine [M+Cl-]- adduct at m/z 391 (a 36

amu difference). The molecular ion dissociated to form the following fragments in the ms2

spectra: m/z 337, 324, 310, 295, 293, 251, 237, 219 (most intense ion), 193, 175, 162, 157, and

136. The m/z 219 daughter ion underwent further dissociation to form the following ms3 ion

fragments: m/z 175, 151, 149, 133, and 107. As in the case with compound 22-C2, the

fragments yielded by compound 31-C2 follow the dissociation pathway for any of the three

aforementioned compounds.

Compound 26-C2 at RT 64.81 showed an extremely low UV intensity, and a low relative

ion abundance. A molecular ion was noted at m/z 285 (M-H)-, which was confirmed by the

- presence of an acetate [M+CH3COO] adduct ion at m/z 345, and a sodium acetate [(M-

- H)+CH3COONa] adduct at m/z 367. The molecular ion dissociated to form the following ion

fragments in the ms2 spectra: m/z 268, 257, 243, 241 (most intense), 217, 199, 175, and 151.

The m/z 241 daughter ion underwent further dissociation to yield ms3 fragments of m/z 227,

213, 211, 201, and 199. This fragmentation scheme follows the exact dissociation pathway for

the flavonoid 3',4',5,7-tetrahydroxyflavone (a.k.a. luteolin). In addition to the matching

fragmentation pattern, this structure was also confirmed by the matching retention time for the

luteolin standard.

144 Compound 32-C2 at RT 76.69 revealed both a low UV intensity and ion relative abundance. A molecular ion was discovered at m/z 887 (M-H)-, confirmed by the presence of

- - acetate [M+CH3COO] and sodium acetate [(M-H)+CH3COONa] adducts at m/z 947 and 970

respectively. The m/z 887 molecular ion underwent collision-induced dissociation to yield an

ms2 spectra with ion fragments at m/z 490, 463, 317 and 285 (the most intense ion fragment).

The noted difference between the m/z 463 and 317 ion fragments indicates a 146 amu neutral

loss, revealing the presence of a rhamnose sugar. In addition, the neutral loss associated with the

m/z 285 ion fragment yields a 602 amu neutral loss, which indicates the existence of two

monosaccharide units. The possible monosaccharide units (whose sum equals 602 amu) include

one 308 amu rutinose (6-O-alpha-L-rhamnosyl-D-glucose) unit, plus either a 294 amu lathyrose

(2-O-beta-D-xylosyl-D-galactose) unit, or a 294 amu sambubiose (2-O-beta-D-xylosyl-D-

glucose) unit. The m/z 285 daughter ion found in the ms2 spectra underwent further dissociation

to yield the following fragments in the ms3 spectra: m/z 267, 257, 243, 241, 217, 213, 201, and

164. This fragmentation scheme follows the exact dissociation pattern for the flavonoid 3',4',5,7-

tetrahydroxyflavone (a.k.a. luteolin) (confirmed via the luteolin standard). Therefore, compound

32-C2 is identified as luteolin, with rutinose and lathyrose (or sambubiose) side-groups.

Compound 33-C2 at RT 78.93 was detected using HPLC Method 2, and showed both a

low UV intensity and ion relative abundance. An m/z 529 (M-H)- molecular ion was confirmed

- by the presence of a sodium acetate [(M-H)+CH3COONa] adduct at m/z 611. The molecular ion

dissociated to form m/z 367 (the most intense fragment) and 352 ion fragments in the ms2

spectra. The 162 amu neutral loss associated with the m/z 367 daughter ion reveals the presence

of a glucose (or galactose) side-group. Further fragmentation of m/z 367 yielded ions at m/z

352, 336, 324, 309, and 297. This fragmentation pathway follows the possible dissociation

145 scheme for the compound glycycoumarin (or derivative of) (Figure 46). Therefore, compound

33-C2 is tentatively identified as glycycoumarin (or a derivative thereof).

Compound 34-C2 at RT 79.92 exhibited both an intermediate UV intensity and ion relative abundance. An intense ion was found at m/z 339 (M-H)-, and is assumed to be the

molecular ion (although no adduct ions were detected). The m/z 339 parent ion dissociated to

yield the following fragments in the ms2 spectra: m/z 251, 245, 233, 219 (the most intense ion),

and 175. The m/z 219 daughter ion underwent further fragmentation to produce m/z 197, 191,

177, 176, 175, 163, 151, 149, and 133 ions in the ms3 spectra. The ion fragment at m/z 177

indicates a neutral loss of 162 amu, revealing the presence of a glucose (or galactose) unit. The

overall dissociation scheme closely follows the predictive ion patterns from that of glycosidically

bound dihydroxy-coumarins, including 6,7-dihydroxycoumarin-7-glucoside, 6,7-dihydroxy-

coumarin-6-glucoside, and 7,8-dihydroxycoumarin-7-b-D-glucoside (Figure 87). Therefore,

compound 34-C2 is tentatively identified as one of the aforementioned compounds (or a possible

derivative thereof).

Fraction D - HPLC Method 1

A total of 38 compounds were identified in peanut skin Fraction D obtained from

Toyopearl SEC (Table 7). UV chromatograms and TICs for each scan range can be seen in

Figures A.18 through A.21. Fraction D contained high concentrations of flavonoid monomers, methoxy-flavonoids, methylated-flavonoids, flavonoid-glycosides, and a methoxy-flavonoid dimer. Several of the structures found in Fraction D have been previously reported in prior studies (such as catechin and epicatechin), however, a vast majority of these new compounds have not yet been shown to exist in peanut skins. Flavonoid monomers consisted of extremely

146 high amounts of catechin (compound 2-D1 at RT 13.14), epicatechin (compound 3-D1 at RT

16.44), and their O-methylated derivatives (compound 4-D1 at RT 19.09, compound 6-D1 at RT

21.14, and compound 13-D1 at RT 25.26). Other monomers included the identification of 318 amu flavonoid derivatives (compound 7-D1 at RT 21.84, and compound 8-D1 at RT 22.41) which appear to be catechin and/or epicatchin with an attached ethyl group (or two separate methyl groups on one or more flavonoid rings). Another possibility for the identification of these two structures includes a catechin or epicatechin base-unit, with two (of the five) hydroxyl units actually existing as methoxy units. The presence of catechin and epicatechin was confirmed by matching retention times with known standards. Methylated derivatives of catechin and epicatechin were identified either via fragmentation patterns revealed by obtained standards, or by interpretation of the dissociation pathway. The presence of a 316 amu structure was also detected (compound 16-D1 at RT 26.38), which follows the fragmentation pattern for 3- methylquercetin (a.k.a. isorhamnetin). This was confirmed by predictive fragmentation patterns

(Figure 32), and the use of the NIST 05 Mass Spectral Database. The presence of this monomer compound is not surprising, given the vast number of isorhamnetin-based flavonoid glycosides which have been identified. Another m/z 315 (M-H)- compound was detected at RT 33.30

(compound 30-D1), which also follows a similar fragmentation scheme to that of isorhamnetin.

Several other 316 amu molecular weight flavonoids exist which could follow this dissociation pattern, including: 6-methoxyluteolin (a.k.a. nepetin), 4’-methoxyquercetin (a.k.a. tamarixetin),

and 7-methylquercetin (a.k.a. rhamnetin) (Figure 56).

Additional flavonoid monomers were detected in Fraction D which exhibited a 300 amu

molecular weight (m/z 299 (M-H)-). These structures include compound 28-D1 at RT 30.94, compound 29-D1 at RT 31.75, and compound 31-D1 at RT 33.54. All of these compounds

147 exhibited both a high UV response and ion relative abundance. In all cases, a molecular ion was

- - identified at m/z 299 (M-H) due to the presence of an acetate [M+CH3COO] adduct at m/z 359

- and a sodium acetate [(M-H)+CH3COONa] adduct at m/z 381. The molecular ion dissociated to form ms2 spectra with an intense ion at m/z 284 (for all three compounds). The 15 amu loss

(methyl group) between the molecular ion and the ms2 daughter fragment indicates the possible presence of a methoxy group. Further dissociation of m/z 284 (for all three compounds) formed an ms3 spectra with m/z 267, 256/255, and 227 ion fragments. Although the dissociation patterns for all three compounds are nearly identical, extremely slight variations in less intense fragment ions (along with slightly different retention times) reveal that these are in fact three separate compounds. Dissociation pathways based on obtained fragmentation data of standards for this study indicate a high probability that the compound is 5,7,4'-trihydroxy-3'-methoxy- flavone (a.k.a. chrysoeriol), specifically with regards to the m/z 284, 256, and 151 fragments.

However, it has also been shown that chrysoeriol (Figure 34) exhibits dissociation patterns which are nearly identical to that of diosmetin (aka luteolin-4'-methyl ether). In addition, these two compounds elute at nearly the same retention times under similar chromatographic conditions, which supports the theory that the other compound may possibly be diosmetin. Although fragmentation data for the standards of these compounds was obtained, definitive matches are only possible by running the actual standards to determine UV retention times (which were not available for this research). Predictive fragmentation would indicate that the remaining compound is a different 300 amu molecular weight flavonoid which follows a similar fragmentation scheme, of which several exist. Although at least two of the three aforementioned compounds most likely chrysoeriol and diosmetin, the lack of standards for proper identification opens the possibility that these compounds may in fact be other flavonoid or chalcone

148 compounds, which exhibited similar fragmentation pathways (specifically with regards to the m/z 284 and 227 daughter ions). Possible flavonoid compounds include: 3',5,7-trihydroxy-4'- methoxyisoflavone (a.k.a. pratensein), 4',5,7-trihydroxy-6,8-dimethylflavanone (a.k.a. farresol or cyrtopterinetin), 4',5,7-trihydroxy-6-methoxyisoflavone (a.k.a. tectorigenin) (Figure 35), psi- tectorigenin, 4'-hydroxy-5,7-dimethoxyflavanone, 8-methoxy-iso-scutellargin, 3,7-dihydroxy-2-

(4-hydroxy-3-methoxyphenyl)-4-benzopyrone (a.k.a. geraldol), and kaempferol 4'-methyl ether

(a.k.a. kaempferide). Possible chalcone compounds include: 2',4'-dihydroxy-3',6'-dimethoxy- chalcone, 2',4-dihydroxy-4',6'-dimethoxychalcone, 2',6'-dihydroxy-3',4'-dimethoxychalcone, and

2',6'-dihydroxy-4,4'-dimethoxychalcone (Figure 67).

Compound 38-D1 at RT 38.07 is another example of a flavonoid monomer identified in

Fraction D. This compound exhibited both an intermediate UV response and ion relative abundance. The most intense ion in the normal mass spectra was m/z 283 (M-H)-, and was

- confirmed as the molecular ion by the presence of a sodium acetate [(M-H)+CH3COONa] adduct at m/z 365. The molecular ion dissociated to form an m/z 268 fragment in the ms2 spectra, which the 15 amu loss is most likely caused by the loss of a methyl group (and typically indicates the presence of a methoxy group). Further fragmentation led to an ms3 spectra with ion fragments of m/z 267, 240, 239, 226, 224, 223, 212, and 211. According to the standards, these ion fragments closely follow the exact dissociation pathways for 5,7-dihydroxy-4'- methoxyisoflavone (a.k.a. biochanin A or genistein 4-methyl ether), 4',5-dihydroxy-7- methoxyisoflavone (a.k.a. prunetin), and 5,7-dihydroxy-4'-methoxyflavone (a.k.a. acacetin or apigenin-4'-methyl ether). Although compound 38-D1 is most likely one of the aforementioned compounds, the fragmentation scheme (with the m/z 268 daughter ion) also compares to numerous other flavonoid monomers at m/z 283 (M-H)- which exhibit similar structural

149 characteristics. These compounds include the following: 5-dihydroxy-7-methoxyflavone (a.k.a.

7-methoxyapigenin), 4',7-dihydroxy-6-methoxy-isoflavone (a.k.a. glycitein), 5,6-dihydroxy-7- methoxyflavone (a.k.a. negletein or balcalein-7-methyl ether), 5,7-dihydroxy-6-methoxyflavone

(a.k.a. oroxylin), and 5,7-methoxyflavanone (Figure 65). In addition, the NIST 05 database gave high probability spectra matches for methoxy flavones, including 5,7-dihydroxy-8- methoxyflavone (a.k.a. vogonin) and 4’,5-dihydroxy-7-methoxyflavone (a.k.a. genkwanin).

Also, the existence of an m/z 268 daughter ion has the potential to follow the fragmentation scheme for chalcone-based compounds, including 2'-hydroxy-3,4-dimethoxychalcone and 2'- hydroxy-4',6'-dimethoxychalcone (Figure 66). Unfortunately, standards for these compounds were not available for this research (preventing positive identification by means of matching UV retention times), therefore, compound 38-D1 can only be tentatively identified via fragmentation data from past literature, and/or MSn interpretation through fragmentation prediction.

Additional flavonoid monomers identified in Fraction D can be found in Table 7. The

vast majority of these compounds exhibited a high UV intensity and ion relative abundance.

Some of these compounds include: 4',5,7-trihydroxy-3'-methoxyflavanone (a.k.a. homo- eriodictyol) (compound 25-D1 at RT 29.23), luteolin (a.k.a. 3',4',5,7-tetrahydroxyflavone)

(compound 27-D1 at RT 30.43), and kaempferol-3,7-dimethoxy (compound 37-D1 at RT 37.81).

These compounds were also identified by either matching retention times with those of standards, by matching of spectra via the NIST 05 Database, or prediction of fragmentation pathways.

A large number of high-concentration flavonoid glycosides were identified in fraction D

(Table 7), all of which are based on the flavonoid monomers which have been detected in the peanut skins. For example, compound 5-D1 at RT 20.72 gave both a medium-to-high UV

150 response, and ion intensity in the normal mass spectra. The most intense ion was observed at m/z 477 (M-H)-, and was confirmed to be the molecular ion due to the presence of a sodium

- acetate [(M-H)+CH3COONa] adduct ion at m/z 559 (82 amu difference). Dissociation of the

molecular ion gave an ms2 ion spectra with fragments of m/z 315 (most intense), 314, 299, 271,

and 247. The neutral loss associated with m/z 315 is equivalent to 162 amu, indicating the

presence of glucose or galactose monosaccharide. Further fragmentation of m/z 315 yielded ions

in the ms3 spectra including m/z 300, 299, 285, 245, 189, and 177, which follows the

fragmentation pattern of isorhamnetin and its glycosidically-bound conjugate (Figure 88). This

compound is most likely an isorhamnetin-glucoside derivative (such as isorhamnetin-3-

glucoside), although the position of the glucose unit cannot be determined by MS/MS alone.

Another m/z 477 compound (compound 15-D1) exists at RT 25.82, with a molecular ion

- confirmed by a sodium acetate [(M-H)+CH3COONa] adduct at m/z 559. Compound 15-D1

exhibits practically identical fragmentation attributes to that of compound 5-D1, and is most

likely due to the presence of the sugar unit residing at a different location on the isorhamnetin

base (when compared to compound 5-D1). Therefore, compound 15-D1 is either another

isorhamnetin-glucose derivative, or less likely, the aglycone base unit is a different flavonoid

(other than isorhamnetin) with the same 316 amu molecular weight (such as 6-methoxyluteolin

a.k.a. nepetin, 4’-methoxyquercetin a.k.a. tamarixetin, or 7-methylquercetin a.k.a. rhamnetin). A

final m/z 477 compound with identical fragmentation attributes was detected at RT 28.39.

Again, this compound is most likely an isorhamnetin-based compound, with a glucose unit

residing at a different position than the aforementioned compounds. The various locations of the

glucose units on the isorhamnetin base-unit would account for the difference in HPLC elution

time.

151 Another example of a Fraction D flavonoid-glycoside is compound 11-D1 at RT 23.73.

This compound exhibited both an intermediate UV intensity and ion relative abundance. The molecular ion was confirmed at m/z 463 (M-H)- due to the presence of a sodium acetate [(M-

- H)+CH3COONa] adduct ion at m/z 545. The molecular ion underwent collision-induced dissociation (CID) to yield an intense m/z 301 daughter fragment in the ms2 spectra. The 162 amu neutral loss between the m/z 463 parent ion and the m/z 301 fragment indicate the existence of a glucose (or galactose) monosaccharide unit. The m/z 301 ion underwent further dissociation to produce the following ion fragments in the ms3 spectra: m/z 271, 255, 179, and 151. The dissociation scheme follows the exact fragmentation pattern of a quercetin aglycone. Therefore, compound 11-D1 is most likely a quercetin-based glycoside, such as isoquercetrin or .

An additional m/z 463 compound was detected at RT 24.42 (compound 12-D1), which showed an intermediate UV response, and a high relative ion abundance. The normal mass spectra showed an intense ion at m/z 463 (M-H)-, and a low intensity ion at m/z 545. The m/z 545 ion

- reveals the presence of a sodium acetate [(M-H)+CH3COONa] adduct ion due to the difference

(of 82 amu) between m/z 545 and 463. Therefore, m/z 463 is the molecular ion for this

compound. The parent ion underwent collision-induced dissociation (CID) to yield an ms2 ion spectrum with an intense m/z 301 ion. The neutral loss associated with m/z 301 (from m/z 463) indicates the presence of a 162 amu hexose sugar (either glucose or galactose). Further fragmentation of m/z 301 produced an ms3 spectra with ions m/z 286 (most intense), 271, 256,

255, 241, 217, 201, 199, and 151. Based on the possible fragmentation schemes, the flavonoid aglycone is most likely quercetin, with an O-glycosidic linkage. Possible candidates for this compound include: quercetin-3-O-galactoside (a.k.a. hyperoside), quercetin-3-O-glucoside

(a.k.a. isoquercitrin), quercetin-7-D-glucoside (a.k.a. gossypitrin or quercimeritrin), and

152 quercetin-4’-O-glucoside (a.k.a. spiraeoside). The dissociation pathways reveal that compound

12-D1 is most likely quercetin-4’-O-glucoside (a.k.a. spiraeoside), since it exhibits the most ion fragment matches to the MSn spectra (Figure 89).

Additional flavonoid-glycosides identified in Fraction D can be found in Table 7. The

vast majority of these compounds exhibited a high UV intensity and ion relative abundance.

Some of these compounds include: Kaempferol-glucoside (compound 14-D1 at RT 25.56), isorhamnetin-3-methoxy-glucoside (or other m/z 329 M-H- aglycone)(compound 17-D1 at RT

26.64), chrysoeriol-glucoside (or other m/z 299 M-H- aglycone)(compound 20-D1 at RT 27.42),

and luteolin-glucoside (compound 21-D1 at RT 27.80). These compounds were identified either

by comparison to fragmentation standards, by matching of spectra via the NIST 05 Database, or

prediction of fragmentation pathways.

Fraction D - HPLC Method 2

The use HPLC method 2 exhibited better overall separation (than method 1), however,

the use of this method for Fraction D only allowed the detection of few additional compounds

when compared to HPLC Method 1. UV chromatograms and TICs for each scan range can be

seen in Figures A.46 through A.47. A total of 7 additional compounds were identified in peanut

skin Fraction D obtained from Toyopearl SEC (Table 8). Compound 1-D2 at RT 43.10 exhibited

a high UV intensity, and an intermediate relative ion abundance. An intense m/z 389 (M-H)- molecular ion was found in the normal mass spectra, and dissociated to form an m/z 227 ion in the ms2 spectra. The 162 amu neutral loss difference between the molecular ion and m/z 227 daughter fragment indicates the presence of a glucose (or galactose) monosaccharide side-group.

The m/z 227 ion further dissociated to yield the following ion fragments in the ms3 spectra: m/z

153 185, 159, 157, and 143. This dissociation pattern follows the exact fragmentation scheme for the stilbene known as resveratrol. The fragmentation fingerprint was confirmed using a resveratrol standard. This was also further confirmed by matching the ms3 spectra with the known resveratrol fragmentation pattern using the NIST 05 Database. Therefore, compound 1-D2 is identified as a resveratrol aglycone, plus an additional glucose (or galactose) side-group.

HPLC Method 2 exhibited enough resolution to separate multiple 302 amu compounds which could not be identified using HPLC Method 1. The use of Method 1 allowed for the detection of two compounds with m/z 301 (M-H)-, and in addition to those structures, Method 2

was able to separate four additional 302 amu compounds. Each of these compounds exhibited

similar (but slightly different) fragmentation patterns, which is indicative of flavonoids at this

given molecular weight. Standards for these compounds identified the presence of 2',3,4',5,7-

pentahydroxyflavone (a.k.a. morin) (compound 2-D2 at RT 57.43) and 3',5,7-trihydroxy-4'-

methoxyflavanone (a.k.a. hesperitin) (compound 5-D2 at RT 61.55). Two additional 302 amu

molecular weight flavonoids were discovered at RT 60.03 and 60.69 (compounds 3-D2 and 4-

D2). Based on the their similar dissociation patterns, these flavonoids are most likely 3,3',4',5',7-

pentahydroxyflavone (a.k.a. robinetin) and 3',4',5,5',7-pentahydroxyflavone (a.k.a. tricetin).

The use of HPLC Method 2 also allowed for the identification of two 492 amu molecular

weight compounds (compound 6-D2 at RT 72.98, and compound 7-D2 at RT 77.72). In both

cases, the m/z 491 (M-H)- molecular ion was confirmed by the presence of a sodium acetate [(M-

- H)+CH3COONa] adduct ion at m/z 574. The molecular ion for both compounds dissociated to

yield an intense m/z 287 fragment in the ms2 spectra. The 205/204 amu loss indicates the

possible presence of a sinapoyl acyl side-group. Further dissociation of the m/z 287 daughter ion

produced the following ion fragments in the ms3 spectra: m/z 269, 219, 165, and 161. This

154 fragmentation pattern follows the predictive pathway for the 288 amu compound known as 2,3- dihydrofisetin (a.k.a. fustin) (Figure 27). In addition, a search of the ms2 spectra using the NIST

05 Mass Spectral Database produced a high probability match for the compound known as

3,3’,4’,7-tetrahydroxyflavone (a.k.a. ), which has a structure nearly identical to that of fustin. Therefore, compounds 6-D2 and 7-D2 are most likely fustin-based flavonoids, with sinapoyl acyl groups residing at different positions on the flavonoid structure.

Fraction E - HPLC Method 1

A total of 9 compounds were identified in peanut skin Fraction E obtained from

Toyopearl SEC (Table 9). UV chromatograms and TICs for each scan range can be seen in

Figures A.22 through A.24. Fraction E contained fewer compounds (and concentrations of

compounds) when compared to all other fractions. It consisted of several flavonoid monomers, flavonoid-glycosides, and a flavonoid dimer. The first flavonoid monomer was compound 1-E1, found at RT 18.72. An m/z 349 (M-H)- molecular ion was confirmed by the presence of both an

- acetate [(M-H)+CH3COOH] adduct ion at m/z 408 (59 amu difference), and a sodium acetate

- [(M-H)+CH3COONa] adduct ion at m/z 431. The molecular ion dissociated to yield an ms2 spectra with ions of m/z 331 (the most intense ion) and 299. The m/z 331 daughter fragment underwent further dissociation to produce ms3 ions of m/z 300, 299, 287, 271, 179, and 151.

Although this fragmentation pattern closely resembles the MS/MS spectra of many flavonoid and methoxy-flavonoid compounds (such as quercetin-based flavonoids and chromanediols), the 350 amu molecular weight does not coincide with any known flavonoids. For example, according to the NIST 05 Mass Spectral Database, compound 1-E1 exhibits almost the exact same dissociation pattern as 2-(3,4-dimethoxyphenyl)-7-methoxy-3,4-chromanediol. The only

155 difference between this structure and compound 1-E1 is an 18 amu difference in molecular weight. By predicting the known dissociation pathways in which flavonoids commonly follow, it is possible that this compound could be a previously undiscovered chromanediol structure called 2-(3,4-dimethoxyphenyl)-3,4-dihydro-2H-chromene-3,4,5,6,7-pentol (Figure 90). This new compound has almost the same structural characteristics as 2-(3,4-dimethoxyphenyl)-7- methoxy-3,4-chromanediol, except it exhibits one less methoxy group, and three additional hydroxy groups. A similar structure exists at RT 21.83 (compound 4-E1). This compound gave both an intermediate UV intensity and ion relative abundance. The m/z 363 (M-H)- molecular

- ion was confirmed by the presence of a sodium acetate [(M-H)+CH3COONa] adduct ion at m/z

445. The molecular ion dissociated to yield an ms2 spectra with ions of m/z 331 (the most

intense ion) and 299. The m/z 331 daughter fragment underwent further dissociation to produce

ms3 ions of m/z 300, 299, 287, 271, 179, and 151. Again, compound 4-E1 exhibits almost the exact same dissociation pattern as 2-(3,4-dimethoxyphenyl)-7-methoxy-3,4-chromanediol, however, in this case the only difference between this structure and compound 4-E1 is an 32 amu difference in molecular weight. By predicting the known dissociation pathways in which flavonoids commonly follow, it is possible that compound 4-E1 could be a previously undiscovered chromanediol structure called 2-(3,4-dimethoxyphenyl)-7-methoxy-3,4-dihydro-

2H-chromene-3,4,5,6-tetrol (Figure 91). This new compound has almost the same structural characteristics as 2-(3,4-dimethoxyphenyl)-7-methoxy-3,4-chromanediol, except it exhibits two additional hydroxy groups.

Compound 6-E1 was another monomer discovered in Fraction E, which eluted at RT

25.90 with an intermediate UV intensity, but a high ion relative abundance. An intense m/z 347

(M-H)- ion was found in the normal mass spectra, which dissociated to yield an m/z 287 in the

156 - ms2 spectra. The 60 amu neutral loss indicates the presence of an acetate [M+CH3COO] adduct

ion, which confirms m/z 287 as the molecular ion. Further dissociation of the m/z 287 ion

fragment produced m/z 269, 243, 225, 183, 169, 151, 135, and 125 in the ms3 spectra. Although

this partially follows the dissociation scheme for several 288 amu flavonoids (such as 6-

hydroxyapigenin), it more closely follows the predictive fragmentation scheme for

dihydrokaempferol (Figure 82). However, the NIST 05 Mass Spectral Database yielded a high

probability spectral match for the compound 3,3’,4’,7-tetrahydroxy-flavanone (a.k.a. 2,3-

dihydrofisetin, or fustin). Therefore, compound 6-E1 is most likely one of the two

aforementioned compounds.

Compound 7-E1 at RT 27.10 exhibited both an intermediate UV intensity and ion relative

abundance. An intense m/z 285 (M-H)- ion was found in the normal mass spectra, and was

- confirmed to be the molecular ion by the presence of an acetate [M+CH3COO] and sodium

- acetate [(M-H)+CH3COONa] adduct ions at m/z 345 and 367 respectively. The molecular ion

dissociated to form the following ion fragments in the ms2 spectra: m/z 257 (most intense ion),

229, 213, and 177. The m/z 257 daughter ion underwent further dissociation to yield the

following fragments in the ms3 spectra: m/z 229, 213, 187, 172, and 151. Although this closely

follows the dissociation pattern for the flavonoid compounds 3',4',5,7-tetrahydroxyflavone (a.k.a.

luteolin) and 3,4',5,7-tetrahydroxyflavone (a.k.a. kaempferol), it does not however match the

retention time for these known standards. Therefore, compound 7-E1 is a flavonoid compound

other than these two aforementioned structures. The NIST 05 Mass Spectral database revealed

fragment matches for several 286 amu flavonoids. These compounds include: 3,3',4',7-

tetrahydroxyflavone (a.k.a. fisetin), 5-methoxy-3,7-dihydroxyflavanone, 2',3,4',5,7-penta-

hydroxyflavone (a.k.a. datiscetin), 2'-hydroxygenistein, 7,8,2’,4’-tetrahydroxy-isoflavone, 6-

157 hydroxyapigenin (a.k.a. ), 7,3',4',5'-tetrahydroxyisoflavone (a.k.a. baptigenin), , naringen-7-methyl-ether (a.k.a. ), and 5,7,2’,4’-tetrahydroxy- isoflavone.

Compound 8-E1 at RT 29.32 represented the most abundant compound in Fraction E. It exhibited an extremely high UV intensity and ion relative abundance. A molecular ion was

- - confirmed at m/z 301 (M-H) due to the presence of a sodium acetate [(M-H)+CH3COONa] adduct ion at m/z 383. The molecular ion dissociated to yield m/z 273, 257, 179 and 151 daughter fragments in the ms2 spectra. Further dissociation yielded m/z 183, 169, and 151 fragments in the ms3 spectra. This follows the exact dissociation pathway for the flavonoid

3,3',4',5,7-pentahydroxyflavone (a.k.a. quercetin). This was confirmed by matching the UV retention time and fragmentation fingerprint with the standard for this compound.

Several flavonoid-glycosides were found in fraction E, including compound 3-E1 at RT

21.53, compound 5-E1 at RT 25.31, and compound 9-E1 at RT 39.31. Compound 3-E1 and 5-

E1 were both based on quercetin aglycones, with 162 amu glucose (or galactose) monosaccharide side-groups. Numerous possible compounds with this structural characteristic exist, since the sugar unit can reside at several different locations on the quercetin flavonoid structure. These compounds include: quercetin-3-glucoside (a.k.a. isoquercitrin), quercetin-3- galactoside (a.k.a. hyperoside), quercetin-7-D-glucoside (a.k.a. gossypitrin or quercimeritrin), and quercetin 4'-O-glucoside (a.k.a. spiraeoside). Standards for these compounds were not available, making it impossible to determine which of the aforementioned structures correspond to each of compound 3-E1 and 5-E1 (since these structures undergo nearly identical fragmentation pathways). Compound 9-E1 at RT 39.31 was an additional flavonoid-glycoside found in Fraction E, however, the aglycone was not quercetin. In the case of this compound, the

158 loss of a glucose (or galactose) sugar unit yielded an m/z 283 (M-H)- aglycone. MS/MS/MS of

the m/z 283 compound produced m/z 255, 243, and 217 ion fragments. According to the NIST

05 Mass Spectral Database, this follows the possible fragmentation scheme for numerous

dihydroxy-methoxy-flavones, dihydroxy-methoxy-isoflavones, and dimethoxy-flavanones.

A flavonoid dimer (compound 2-E1) was found at RT 20.80, and exhibited both an

intermediate UV intensity and ion relative abundance. An m/z 575 (M-H)- molecular ion

dissociated to yield ms2 daughter fragments of m/z 557, 449 (the most intense ion), 437, 408,

394, and 287. Further dissociation of the m/z 449 fragment produced the following ion

fragments in the ms3 spectra: m/z 431, 287, and 243. This dissociation pathway follows the

exact fragmentation scheme for an A-type proanthocyanidin dimer compound. Several possible

positions exist in which the two flavonoid monomers can join (including A1, A2, A4, and A5’),

however, proper identification of the attachment type would be difficult without a standard for

comparison (or NMR analysis to determine the stereochemistry). An example of the dissociation

pathway for such a compound can be seen in Figure 92.

Fraction E - HPLC Method 2

The use HPLC method 2 exhibited better overall separation (than method 1), however,

the use of this method for Fraction E only allowed the detection of few additional compounds

when compared to HPLC Method 1. However, Method 2 was in fact able to identify high

concentrations of specific compounds in which Method 1 was unable to detect in any amount.

UV chromatograms and TICs for each scan range can be seen in Figures A.48 through A.49. A total of 3 additional compounds were identified in peanut skin Fraction E obtained from

Toyopearl SEC (Table 10). For example, compound 1-E2 was found at an extremely high UV

159 intensity and ion relative abundance. The concentration was high enough to cause compound 1-

E2 to elute from RT 39.95 to 44.02 (min). The m/z 331 (M-H)- molecular ion was confirmed by

- the presence of an acetate [M+CH3COO] adduct at m/z 391. Dissociation of the molecular ion

yielded the following ions in the ms2 spectra: m/z 301, 300, 299 (the most intense ion), 287,

271, 255, 179, and 151. Further dissociation of the m/z 299 daughter fragment led to the

following ion fragments in the ms3 spectra: m/z 272, 271, 256, 255, 227, 179, and 155. The

fragmentation pathway is nearly identical to that of 3-methylquercetin (a.k.a. isorhamnetin),

however, a 16 amu difference exists between their molecular weights (332 amu for compound 1-

E2 versus 316 amu for isorhamnetin). A limited number of possibilities exist for a flavonoid

structure to have a 332 amu molecular weight. One known flavone compound exists (called

5,6,8,3’,4’-pentahydroxy-7-methoxyflavone), however, it does not follow the same

fragmentation pathway as compound 1-E2. One scenario would be a dihydoxy-dimethoxy-

flavanone derivative similar to 3,5-dihydroxy-2-(4-hydroxy-3-methoxyphenyl)-7-methoxy-2,3-

dihydro-4H-chromen-4-one (yielding a structure similar to that of isorhamnetin) (Figure 93),

which could possibly be a newly undiscovered flavanone compound. Another possibility for

compound 1-E2 is a chromanediol. The NIST 05 Mass Spectral Library showed a similar

spectra for a 332 amu molecular weight compound called 2-(3,4-dimethoxyphenyl-7-methoxy-

3,4-chromanediol.

Compound 2-E2 at RT 54.00 (min) showed an extremely high UV intensity and ion

relative abundance. This compound exhibited the highest concentration out of all other

structures in Fraction E, and reveals that compound 2-E2 is highly abundant in peanut skins. An

intense m/z 287 (M-H)- ion was found in the normal mass spectra, and determined to be the

- molecular ion by the presence of acetate [M+CH3COO] and sodium acetate [(M-

160 - H)+CH3COONa] adducts at m/z 347 and 369 (respectively). The molecular ion dissociated to

yield an m/z 151 ion in the ms2 spectra. Further dissociation of this ion formed m/z 107 and 65

product ions in the ms3 spectra. Compound 2-E2 perfectly matches the retention time and

fragmentation pattern for the standard of 3',4',5,7-tetrahydroxyflavanone (a.k.a. eriodictyol).

Compound 3-E2 at RT 61.28 exhibited both an intermediate UV intensity and ion relative

abundance. An intense m/z 285 (M-H)- ion was found in the normal mass spectra, and was

- confirmed to be the molecular ion by the presence of an acetate [M+CH3COO] and sodium

- acetate [(M-H)+CH3COONa] adduct ions at m/z 345 and 367 respectively. The molecular ion

dissociated to form the following ion fragments in the ms2 spectra: m/z 257 (most intense ion),

256, 229, 213, and 175. The m/z 257 daughter ion underwent further dissociation to yield the

following fragments in the ms3 spectra: m/z 239, 229, 213, 171, and 157. Although this closely

follows the dissociation pattern for the flavonoid compounds 3',4',5,7-tetrahydroxyflavone (a.k.a.

luteolin) and 3,4',5,7-tetrahydroxyflavone (a.k.a. kaempferol), it does not however match the

retention time for these known standards. Therefore, compound 3-E2 is a flavonoid compound

other than these two aforementioned structures. The NIST 05 Mass Spectral database revealed

fragment matches for several 286 amu flavonoids. These compounds include: 3,3',4',7-

tetrahydroxyflavone (a.k.a. fisetin), 5-methoxy-3,7-dihydroxyflavanone, 2',3,4',5,7-penta-

hydroxyflavone (a.k.a. datiscetin), 2'-hydroxygenistein, 7,8,2’,4’-tetrahydroxy-isoflavone, 6-

hydroxyapigenin (a.k.a. scutellarin), 7,3',4',5'-tetrahydroxyisoflavone (a.k.a. baptigenin),

isosakuranetin, naringen-7-methyl-ether (a.k.a. sakuranetin), and 5,7,2’,4’-tetrahydroxy-

isoflavone.

161 Fraction F - HPLC Method 1

A total of 20 compounds were identified in peanut skin Fraction F obtained from

Toyopearl SEC (Table 11). UV chromatograms and TICs for each scan range can be seen in

Figures A.25 through A.27. Fraction F contained high concentrations of flavonoid dimers, along with several monomers which have never been reported to exist in peanut skins. Extremely high

UV intensities and ion relative abundances reveal that peanut skins contain very high concentrations of these compounds. Several of the dimer structures found in Fraction F have been previously reported in prior studies (based on catechin and/or epicatechin), however, the use of Toyopearl Size-Exclusion Chromatography (SEC) allowed for the separation of different dimeric classes. The result was the ability to detect multiple types of dimers which have been previously unreported, due to the enhanced separation exhibited by first using an HPLC preparatory method such as Toyopearl SEC. Multiple instances of dimers with identical molecular weights were found, although the fragmentation patterns were slightly different. This phenomenon can be explained by the presence of various dimeric classes, including different IFB

(inter-flavonoid bond) linkages for both A-type proanthocyanidin and B-type procyanidin dimers

(along with their branched and linear derivatives). Another explanation for the detection of many instances of similar dimers is the different possibilities of catechin and epicatechin combinations. For example, a dimer can exist as catechin-epicatechin, epicatechin-catechin, catechin-catechin, and epicatechin-epicatechin. The detection of these numerous compounds would not be possible without the use of a prepratory method (such as SEC), since proper separation via HPLC would be unlikely due to co-elution.

A flavonoid dimer (compound 1-F1) was found at RT 6.92, and exhibited an intermediate

UV intensity, but a low ion relative abundance. An m/z 575 (M-H)- molecular ion dissociated to

162 yield ms2 daughter fragments of m/z 557, 531, 513, 449, 423 (the most intense ion), and 407.

Further dissociation of the m/z 423 fragment produced the following ion fragments in the ms3 spectra: m/z 387, 379, 361, 298, 285, and 269. This dissociation pathway follows the exact fragmentation scheme for an A-type proanthocyanidin dimer compound. Although the presence of certain fragments show identical ions to the A-type proanthocyanidin dimer found in Fraction

E, slight differences in less intense ions do however exist. This reveals that Compound 1-F1 is a different proanthocyanidin dimer derivative to the one found in the previous fraction. Additional proanthocyanidin dimers were also detected at RT 23.03 (compound 15-F1) and RT 23.60

(compound 16-F1), which also exhibited many identical ion fragments (with some variations in less intense ions). This reveals that peanut skins contain multiple derivatives of A-type proanthocyanidin dimers which exhibit different types of IFB (inter-flavonoid bond) linkages.

Several possible IFB positions exist in which the two flavonoid monomers can join (including

A1, A2, A4, and A5’), however, proper identification of the attachment type would be difficult without a standard for comparison (or NMR analysis to determine the stereochemistry). An example of the dissociation pathway for such a compound can be seen in Figure 94.

Numerous m/z 577 (M-H)- B-type procyanidin dimers were detected in Fraction F, including compound 2-F1 at RT 10.11, compound 3-F1 at RT 10.84, compound 4-F1 at RT

11.43, compound 5-F1 at 11.97, compound 12-F1 at RT 19.45, and compound 13-F1 at RT

20.06. The ion fragmentation pattern and UV retention time for compound 4-F1 at RT 11.43 perfectly matched the standard for a procyanidin B3 dimer. The remaining compounds also had matching dissociation pathways (except for subtle differences in minor ion fragments), although, the UV retention times were different. This reveals that the remaining compounds are in fact B- type procyanidin dimers, however, the different HPLC elutions times can be explained by the

163 variations in IFB linkages (i.e., B1, B2, B3, etc.), branch types (i.e., linear versus branched), and the combination/order of catechin and epicatechin monomers which make up the dimer structure

(Figure 95).

Several 592 amu molecular weight (m/z 591 (M-H)-) compounds were detected in

Fraction F, each of which constitutes a newly undiscovered flavonoid dimer. These structures

include Compound 6-F1 (at RT 13.90), Compound 8-F1 (at RT 15.73), and Compound 10-F1 (at

RT 16.14). Each of these compounds exhibited identical fragmentation patterns, except for minor differences in low intensity ions. The dissociation pathway and neutral losses are similar

to the standard for the procyanidin B3 dimer, which yield the fragmentation of two

catechin/epicatechin monomers. This can easily be deduced by the presence of an m/z 289 (M-

H)- ion fragment, which yields a neutral loss of 288 amu from the parent ion. However, in the

case of these compounds, the 288 amu neutral loss of a catechin/epicatechin yields an m/z 303

(M-H)- ion fragment, which is indicative of a O-methylated catechin (or O-methylated epicatechin). In addition, a search of the ms3 ion spectra using the NIST 05 Mass Spectral

Database produced a match for a trihydroxy-methoxy-flavonoid, which further supports the theory that one of the monomers exhibits a methoxy group. Therefore, these compounds are most likely dimers comprised of a catechin or epicatechin monomer, along with an O-methylated catechin/epicatchin monomer. The different retention times of these compounds can be explained by the different possibilities regarding both the linkage positions and the various combinations of catechin/epicatechin monomers (and their O-methylated derivatives).

Numerous 574 amu molecular weight (m/z 573 (M-H)-) dimers were detected in Fraction

F, including compound 7-F1 at RT 14.73, compound 9-F1 at RT 15.87, compound 11-F1 at RT

16.80, and compound 14-F1 at 20.21. In each case, dissociation of the molecular ion produced

164 an ms2 spectra with ion fragments of m/z 555, 529, 447 (the most intense ion fragment), and

285. Further dissociation of the m/z 447 daughter ion yield the following product ions in the ms3 spectra: m/z 285 (most intense ion), 271, 246, and 177. Although the compound appears to be a dimer composed of a catechin/epicatechin monomer plus an additional m/z 285 (M-H)-

monomer (such as luteolin or kaempferol), it however more closely follows the fragmentation

scheme for a different dimeric compound. The dissociation pathway follows the exact predictive

fragmentation pattern for a procyanidin dimer (Figure 96), although, with monomer structures

which differ slightly when compared to the normal catechin and/or epicatechin monomers. In

this case, the monomer units look identical to catechin and/or epicatechin, except for a double

bond between the 2 and 3 carbons on the flavonoid C-ring. This would account for the reduction

in total molecular weight when compared to other typical catechin/epicatechin dimers (i.e., 574

amu for this series of compounds, versus 576, 578, and 590 amu for other typical

catechin/epicatechin dimers). The different retention times exhibited by these compounds can be

explained by the various possibilities in which these monomers are able to form dimeric

compounds, since many unique derivatives are capable based on the different positions in which

the monomers are able to attach on the flavonoid rings.

Two 424 amu (m/z 423 (M-H)-) molecular weight compounds were found in Fraction F, both of which exhibited an intermediate UV intensity and ion relative abundance. Compounds

17-F1 and 20-F1 (at RT’s 25.82 and 38.09 respectively) were both confirmed by the presence of

- a sodium acetate [(M-H)+CH3COONa] adduct ion at m/z 505. The fragmentation patterns for

both compounds were nearly identical, except for small differences in low intensity ions. In the case of both compounds, the molecular ion dissociated to yield the following product ions in the ms2 spectra: m/z 405, 355, 341, 313, 301, 299 (the most intense ion), 285, 271, and 259. Again,

165 in both cases, the m/z 299 daughter ion underwent further dissociation to produce ms3 ions of m/z 283 and 271. These ion fragments follow the exact dissociation pathway for three known flavanone compounds: Remangiflavanone B, euchrestaflavanone B, and sophoraflavanone G

(Figure 97). Each of these flavanones follows nearly the same fragmentation scheme.

Therefore, compounds 17-F1 and 20-F1 are each most likely one of the three aforementioned compounds.

Compound 18-F1 at RT 27.32 (min) exhibited both an intermediate UV intensity and ion relative abundance. An m/z 587 (M-H)- molecular ion dissociated to form the following ion fragments in the ms2 spectra: m/z 451 (the most intense ion), 435, 421, 391, 377, 301, 299, and

270. The m/z 451 daughter ion underwent further fragmentation to yield ms3 ions at m/z 423,

407, 405, 383, 341, 325, 299, 287, and 257. The predictive fragmentation pathway for the

biflavanone compound kolaflavanone follows this exact dissociation scheme (Figure 98).

Therefore, compound 18-F1 is most likely kolaflavanone, or a similar derivative of this compound.

One flavonoid glycoside was detected in Fraction F at RT 31.86. Compound 19-F1 exhibited an intermediate UV intensity and ion relative abundance. An m/z 447 (M-H)-

- molecular ion was confirmed by the presence of both an acetate [(M-H)+CH3COOH] and

- sodium acetate [(M-H)+CH3COONa] adducts at m/z 506 and 529 respectively. Collision-

induced dissociation (CID) yielded ms2 ion fragments of m/z 429, 403, 337, 325, 323, 296, 295,

283, and 268. Further fragmentation produced an ms3 spectra with the following ion fragments: m/z 324, 307, 297, 295, 281, 266, and 251. The 179 amu loss of an O-glycosidic sugar unit yielded the remaining 268 amu aglycone. It is difficult to properly identify compound 19-F1, given that many 448 amu flavonoid glycosides exist which follow this same fragmentation

166 pathway. An example of a compound which closely follows this dissociation scheme is 3,4',5,7- tetrahydroxyflavone-3-glucoside (a.k.a. astragalin) (Figure 99). However, positive identification is impossible, since no standards were available, and the following compounds also follow similar dissociation schemes to that of compound 19-F1: 4’,5-dihydroxy-7-methoxyflavanone-5- glucoside (a.k.a. ), , kaempferol-3-glucoside, kaempferol-7-glucoside, luteolin-4'-glucoside, luteolin-8-C-glucoside (a.k.a. lutexin or ), luteolin-7-b-D- glucopyranoside, and luteolin-5-glucoside.

Fraction F - HPLC Method 2

The use HPLC method 2 exhibited better overall separation (than method 1), however, the use of this method for Fraction E only allowed the detection of few additional compounds when compared to HPLC Method 1. UV chromatograms and TICs for each scan range can be seen in Figures A.50 through A.51. A total of 3 additional compounds were identified in peanut skin Fraction F obtained from Toyopearl SEC (Table 12). The presence of a high concentration

316 amu structure was detected (compound 1-F2 at RT 19.48) at a very high UV intensity and ion relative abundance. An m/z 315 (M-H)- molecular ion dissociated to form m/z 297 (the most intense ion), 271, and 231 ion fragments in the ms2 spectra. Further dissociation of the m/z 297

daughter ion yielded an ms3 product ion of m/z 269. Several 316 amu molecular weight

flavonoids can follow the dissociation pathway associated with the m/z 297 ion fragment,

including: 6-methoxyluteolin (a.k.a. nepetin), 4’-methoxyquercetin (a.k.a. tamarixetin), and 7-

methylquercetin (a.k.a. rhamnetin) (Figure 100). The NIST 05 Mass Spectral Database found

several other 316 amu flavonoids which are capable of an m/z 297 (and/or an additional m/z 269

or 271) ion fragment, including 5,7-dimethoxy-2-(4-methoxyphenyl)-4-chromanol, 2-(3,4-

167 dimethoxyphenyl)-6-methyl-3,4-chromanediol, 3’,4’,5,6-tetrahydroxy-7-methoxyflavone (a.k.a. pedalitin), and 5,6,8,4’-tetrahydroxy-7-methoxyflavone.

Two 320 amu molecular weight compounds (compounds 2-F2 and 3-F2) were detected at

RTs 27.49 and 30.59 (min). In both cases, an m/z 319 (M-H)- molecular ion was confirmed by

- - the presence of an acetate [M+CH3COO] and sodium acetate [(M-H)+CH3COONa] adducts at m/z 379 and 401 respectively. Dissociation of the molecular ion yielded m/z 287, 197 (the most intense ion), 165, 153, and 125 in the ms2 spectra. Further fragmentation of the m/z 197 daughter ion produced an m/z 165 ion in the ms3 spectra. This dissociation scheme follows the exact fragmentation pathway for the compound oureatacatechin (Figure 55). Therefore, at least one of these compounds is oureatacatechin, or both of these structures are derivatives of oureatacatechin.

Fraction G-Red - HPLC Method 1

A total of 12 compounds were identified in peanut skin Fraction G-Red obtained from

Toyopearl SEC (Table 13). UV chromatograms and TICs for each scan range can be seen in

Figures A.28 through A.30. Fraction G-Red contained high concentrations of different types of flavonoid dimers. Extremely high UV intensities and ion relative abundances reveal that peanut skins contain very high concentrations of these compounds. Several of the dimer structures found in Fraction G-Red have been previously reported (based on catechin and/or epicatechin), however, the use of Toyopearl Size-Exclusion Chromatography (SEC) allowed for the separation of different dimeric classes. The result was the ability to detect multiple types of dimers which have been previously unreported, due to the enhanced separation exhibited by first using an

HPLC preparatory method such as Toyopearl SEC. Multiple instances of dimers with identical

168 molecular weights were found, although the fragmentation patterns were slightly different. This phenomenon can be explained by the presence of various dimeric classes, including different IFB

(inter-flavonoid bond) linkages for both A-type proanthocyanidin and B-type procyanidin dimers

(along with their branched and linear derivatives). Another explanation for the detection of many instances of similar dimers is the different possibilities of catechin and epicatechin combinations. For example, a dimer can exist as catechin-epicatechin, epicatechin-catechin, catechin-catechin, and epicatechin-epicatechin. The detection of these numerous compounds would not be possible without the use of a prepratory method (such as SEC), since proper separation via HPLC would be unlikely due to co-elution factors.

Numerous m/z 577 (M-H)- B-type procyanidin dimers were detected in Fraction G-Red,

including compound 1-Gr1 at RT 10.89, compound 2-Gr1 at RT 11.27, compound 3-Gr1 at RT

12.90, compound 4-Gr1 at 13.84, and compound 5-Gr1 at RT 16.00. UV and ion relative

abundances ranged from low to extremely high. These compounds had nearly identical

dissociation pathways (except for subtle differences in minor ion fragments), and matched the

ion fragments exhibited by the B-type procyanidin dimer standard. This reveals that these

compounds are in fact B-type procyanidin dimers, however, the different HPLC elutions times

can be explained by the variations in IFB linkages (i.e., B1, B2, B3, etc.), branch types (i.e.,

linear versus branched), and the combination/order of catechin and epicatechin monomers which

make up the dimer structure (Figures 101 and 102).

A flavonoid dimer (compound 6-Gr1) was found at RT 16.51 (min), and exhibited an

extremely strong UV intensity and ion relative abundance. An m/z 575 (M-H)- molecular ion

dissociated to yield ms2 daughter fragments of m/z 449 (the most intense ion), 423, 407, 289,

and 285. Further dissociation of the m/z 449 fragment produced the following ion fragments in

169 the ms3 spectra: m/z 287 and 285. This dissociation pathway follows the exact fragmentation scheme for an A-type proanthocyanidin dimer compound. An additional A-type proantho- cyanidin dimer (compound 8-Gr1) was also detected at RT 18.32 (min), which exhibited many identical ion fragments (with some variations in less intense ions). This reveals that peanut skins contain multiple derivatives of A-type proanthocyanidin dimers which exhibit different types of

IFB (inter-flavonoid bond) linkages. Several possible IFB positions exist in which the two flavonoid monomers can join (including A1, A2, A4, and A5’), however, proper identification of the attachment type would be difficult without a standard for comparison (or NMR analysis to determine the stereochemistry). An example of the dissociation pathway for such a compound can be seen in Figure 92.

Numerous 574 amu molecular weight (m/z 573 (M-H)-) dimers were detected in Fraction

F, including compound 7-Gr1 at RT 18.01, compound 9-Gr1 at RT 19.17, and compound 10-Gr1 at RT 21.12. These compounds ranged from low UV and ion abundances (low concentrations), to extremely high concentrations. In each case, dissociation of the molecular ion produced an ms2 spectra with ion fragments of m/z 555, 529, and 447 (the most intense ion fragment).

Further dissociation of the m/z 447 daughter ion yield the following product ions in the ms3

spectra: m/z 285 (most intense ion), and 271. Although the compound appears to be a dimer

composed of a catechin/epicatechin monomer plus an additional m/z 285 (M-H)- monomer (such

as luteolin or keampferol), it however more closely follows the fragmentation scheme for a

different dimeric compound. The dissociation pathway follows the exact predictive frag-

mentation pattern for a procyanidin dimer (Figure 96), although, with monomer structures which

differ slightly when compared to the normal catechin and/or epicatechin monomers. In this case,

the monomer units look identical to catechin and/or epicatechin, except for a double bond

170 between the 2 and 3 carbons on the flavonoid C-ring. This would account for the reduction in total molecular weight when compared to other typical catechin/epicatechin dimers (i.e., 574 amu molecular weight for this series of compounds, versus 576, 578, and 590 amu for other typical catechin/epicatechin dimers). The different retention times exhibited by these compounds can be explained by the various possibilities in which these monomers are able to form dimeric compounds, since many unique derivatives are capable based on the different positions in which the monomers are able to attach on the flavonoid rings. Similar compounds were identified in the previous fraction (Fraction F), although, these compounds most likely have different IFB linkages.

Two 556 amu molecular weight compounds were found in Fraction G-Red, which have never previously been detected in peanut skins. Both compounds 11-Gr1 (at RT 38.21) and 12-

Gr1 (at RT 39.79) exhibited an intermediate UV intensity and ion relative abundance. An m/z

555 (M-H)- molecular ion dissociated to yield ms2 daughter ions of m/z 445, 433 (the most

intense ion), 403, 390, 283, 271, and 243. Further fragmentation of the m/z 433 ion produced the

following ions in the ms3 spectra: m/z 404, 389, 361, 283, 269, 255, and 243. This follows the

exact dissociation pathway for two 556 amu flavonoid dimers known as hegoflavone A (Figure

103) and morelloflavone (Figure 104). Both of these dimers are comprised of IFB linked

naringenin and luteolin monomers, but attached at different positions on the flavonoid structure.

Given each of these dimers exhibit identical fragmentation pathways, it would therefore be

impossible to determine which corresponds (to compound 11-Gr1 and 12-Gr1) without the use of

standards.

171 Fraction G-Red - HPLC Method 2

The use HPLC method 2 exhibited better overall separation (than method 1), however, the use of this method for Fraction G-Red only allowed the detection of few additional flavonoid compounds when compared to HPLC Method 1. UV chromatograms and TICs for each scan range can be seen in Figures A.52 through A.53. A total of 4 additional compounds were identified in peanut skin Fraction G-Red obtained from Toyopearl SEC (Table 14). Compound

1-Gr2 at RT 62.40 exhibited a high UV intensity and ion relative abundance. An m/z 301 (M-

- - H) molecular ion was confirmed by the presence of acetate [(M-H)+CH3COOH] and sodium

- acetate [(M-H)+CH3COONa] adduct ions at m/z 360 and 383 respectively. The UV retention

time and dissociation pathway perfectly matches the standard for the flavonoid compound 3',5,7-

trihydroxy-4'-methoxyflavanone (a.k.a. hesperitin). The presence of another flavonoid monomer

was detected at RT 63.10 (compound 2-Gr2), which exhibited a 316 amu molecular weight. This

compound also revealed a high UV intensity and ion relative abundance. Several 316 amu

molecular weight flavonoids exist which could follow this dissociation pattern, including: 6-

methoxyluteolin (a.k.a. nepetin), 4’-methoxyquercetin (a.k.a. tamarixetin), and 7-methyl-

quercetin (a.k.a. rhamnetin) (Figures 56 and 100).

Compound 3-Gr2 at RT 64.81 showed a high UV intensity and relative ion abundance. A

molecular ion was noted at m/z 285 (M-H)-, which was confirmed by the presence of an acetate

- - [M+CH3COO] adduct ion at m/z 345, and a sodium acetate [(M-H)+CH3COONa] adduct at m/z

367. The molecular ion dissociated to form the following ion fragments in the ms2 spectra: m/z

268, 257, 243, 241 (most intense), 217, 199, 175, and 151. The m/z 241 daughter ion underwent

further dissociation to yield ms3 fragments of m/z 227, 213, 211, 201, and 199. This

fragmentation scheme follows the exact dissociation pathway for the flavonoid 3',4',5,7-

172 tetrahydroxyflavone (a.k.a. luteolin). In addition to the matching fragmentation pattern, this structure was also confirmed by the matching UV retention time for the luteolin standard.

Compound 4-Gr2 at RT 70.41 exhibited both a low UV response, but an intermediate ion relative abundance. A molecular ion was identified at m/z 299 (M-H)- due to the presence of a

- sodium acetate [(M-H)+CH3COONa] adduct at m/z 381. The molecular ion dissociated to form

ms2 spectra with an intense ion at m/z 284 (for both compounds). The 15 amu loss (methyl

group) between the molecular ion and the ms2 daughter fragment indicates the possible presence

of a methoxy group. In the case of compound 4-Gr2, further dissociation formed ms3 ion

fragments of m/z 284, 256, 227, and 163. Dissociation pathways based on obtained

fragmentation data of standards for this study indicate a high probability that the compound is

5,7,4'-trihydroxy-3'-methoxy-flavone (a.k.a. chrysoeriol), specifically with regards to the m/z

284, 256, and 151 fragments. However, it has also been shown that chrysoeriol (Figure 34)

exhibits dissociation patterns which are nearly identical to that of diosmetin (aka luteolin-4'-

methyl ether). In addition, these two compounds elute at nearly the same retention times under

similar chromatographic conditions, which supports the theory that the other compound may

possibly be diosmetin. Although fragmentation data for the standards of these compounds was

obtained, definitive matches are only possible by running the actual standards to determine UV

retention times (which were not available for this research). Although compound 4-Gr2 is most

likely one of the two aforementioned compounds, the lack of definitive proof for identification

opens the possibility that these compounds may in fact be other flavonoid or chalcone

compounds, which exhibited similar fragmentation pathways (specifically with regards to the

m/z 284 and 227 daughter ions). Possible flavonoid compounds include: 3',5,7-trihydroxy-4'-

methoxyisoflavone (a.k.a. pratensein), 4',5,7-trihydroxy-6,8-dimethylflavanone (a.k.a. farresol or

173 cyrtopterinetin), 4',5,7-trihydroxy-6-methoxyisoflavone (a.k.a. tectorigenin) (Figure 35), psi- tectorigenin, 4'-hydroxy-5,7-dimethoxyflavanone, 8-methoxy-iso-scutellargin, 3,7-dihydroxy-2-

(4-hydroxy-3-methoxyphenyl)-4-benzopyrone (a.k.a. geraldol), and kaempferol 4'-methyl ether

(a.k.a. kaempferide). Possible chalcone compounds include: 2',4'-dihydroxy-3',6'-dimethoxy- chalcone, 2',4-dihydroxy-4',6'-dimethoxychalcone, 2',6'-dihydroxy-3',4'-dimethoxychalcone, and

2',6'-dihydroxy-4,4'-dimethoxychalcone (Figure 67).

Fraction G - HPLC Method 1

A total of 19 compounds were identified in peanut skin Fraction G obtained from

Toyopearl SEC (Table 15). UV chromatograms and TICs for each scan range can be seen in

Figures A.31 through A.34. Compounds detected in Fraction G using Method 1 consisted of numerous instances of flavonoid trimers and tetramers, along with several monomeric flavonoids which have never been previously reported in peanut skins. The extremely strong UV intensities and ion relative abundances exhibited by these structures reveal that peanut skins contain very high concentrations of these compounds. The use of Toyopearl Size-Exclusion Chromatography

(SEC) allowed for the separation of many polymeric flavonoid structures via HPLC. The result was the ability to detect multiple types of trimers and tetramers which have been previously unreported, due to the enhanced separation exhibited by first using an HPLC preparatory method such as Toyopearl SEC. Multiple instances of trimers and tetramers with identical molecular weights were detected, which exhibited similar fragmentation schemes regarding the high intensity ion fragments, although the fragmentation patterns were slightly different for less intense daughter ions. This phenomenon can be explained by the presence of various trimer and tetramer derivatives which are attached in different manners (i.e., various branched and linear

174 forms), and the type/position of the incorporated linkages (for example, an A-type interflavanic linkage). Another explanation for the detection of many instances of similar dimers/tetramers is the numerous possibilities of catechin and epicatechin combinations. For example, a trimer can exist as catechin-epicatechin-catechin, epicatechin-catechin-catechin, catechin-catechin-catechin, etc. Therefore, based on the numerous variables involved in the formation of these structures, a vast amount of possible forms may exist. The detection of these abundant compounds would not be possible without the use of a prepratory method (such as SEC), since proper separation via

HPLC would be unlikely due to co-elution factors.

Numerous 864 amu molecular weight trimers were found with UV and ion abundances which ranged from weak to extremely high. These structures included Compound 1-G1 at RT

10.00 (min), Compound 2-G1 at RT 10.83, Compound 3-G1 at RT 11.41, Compound 4-G1 at RT

11.80, Compound 5-G1 at RT 12.72, Compound 6-G1 at RT 14.31, Compound 7-G1 at RT

15.33, Compound 8-G1 at RT 15.42, Compound 9-G1 at RT 15.86, and Compound 11-G1 at RT

16.86. According to previous work conducted by Gu et al. (2003) using Electrospray Ionization in negative mode, the fragmentation scheme of these structures (revealed by the most intense ions) suggest they are various derivatives of catechin/epicatechin trimers with a single A-type interflavanic linkage (IFL). An A-type IFL refers to the form of linkage between the flavanol units, which is exhibited by an added ether bond between the C-2 of one monomer, and the O-7 of another monomer (which is in addition to the common C4-C8 IFL bond, or the less common

C4-C6 bond). Again, the multiple occurrences of these similar compounds at different retention times can be explained by the various positions of this A-type IFL (within the order of flavonoid monomers among the total trimer polymer), along with other factors such as branch type (i.e.,

175 linear versus branched), and the combination/order of catechin and epicatechin monomers which make up the trimer structure (Figure 105).

The same previously explained theory holds true for the various tetramer compounds detected in Fraction G. Although these compounds eluted at different retention times, they did however exhibit nearly identical dissociation pathways (except for subtle differences in minor ion fragments). Flavonoid tetramers with 1152 amu molecular weights were detected at RT

16.39 (Compound 10-G1), RT 19.17 (Compound 12-G1), and RT 19.95 (Compound 13-G1).

The UV intensity and ion relative abundance for these structures ranged from intermediate to extremely high, revealing a high concentration of these compounds in peanut skins. Again, according to Gu et al. (2003), the high intensity ion fragments indicate that these compounds also exhibit a single A-type interflavanic linkage (IFL). The presence of specific ion fragments (such as m/z 863 (M-H)-, 861, 575, and 573) can give insight into the position of one or more A-type

IFL sites. Both Compound 10-G1 (RT 16.39) and 12-G1 (RT 19.17) exhibited high intensity

m/z 863 and 575 (M-H)- ion fragments, which are indicative of the following A-type IFL binded

catechin/epicatechin tetramer structure: (Cat)Epi--(Cat)Epi--(Cat)Epi--A(IFL)--(Cat)Epi. On

the other hand, compound 13-G1 at RT 19.95 showed a high intensity m/z 861 (M-H)- ion

fragment, which reveals the presence of the following A-type IFL binded catechin/epicatechin

tetramer structure: (Cat)Epi--A(IFL)--(Cat)Epi--(Cat)Epi--(Cat)Epi.

A flavonoid dimer (compound 14-G1) was found at RT 21.64, and exhibited an

intermediate UV intensity and ion relative abundance. An m/z 575 (M-H)- molecular ion

dissociated to yield ms2 daughter fragments of m/z 557, 531, 513, 449, 423 (the most intense

ion), and 407. Further dissociation of the m/z 423 fragment produced the following ion

fragments in the ms3 spectra: m/z 387, 325, 297, 287, 285, and 269. This dissociation pathway

176 follows the exact fragmentation scheme for an A-type proanthocyanidin dimer compound

(Appledoorn et al., 2009). Although the presence of certain fragments show identical ions to the

A-type proanthocyanidin dimer found in previous fractions, slight differences in less intense ions do however exist. This reveals that compound 14-G1 is a different proanthocyanidin dimer derivative to the one found in the previous fractions. This reveals that peanut skins contain multiple derivatives of A-type proanthocyanidin dimers which exhibit different types of IFB

(inter-flavonoid bond) linkages. Several possible IFB positions exist in which the two flavonoid monomers can join (including A1, A2, A4, and A5’), however, proper identification of the attachment type would be difficult without a standard for comparison (or NMR analysis to determine the stereochemistry).

Compound 15-G1 at RT 23.51 exhibited an intermediate UV intensity, but a high ion relative abundance. An intense m/z 589 (M-H)- molecular ion was found in the normal mass

spectra, which dissociated to yield product ions of m/z 571, 560, 466, 463 (the most intense ion),

421, 404, 343, 303, and 285 in the ms2 spectra. The m/z 463 daughter ion underwent further

dissociation to form the following ms3 ion fragments: m/z 435, 353, 341, 312, 302, 285, 269,

259, and 229. This dissociation scheme follows the exact predictive fragmentation pathway for

the biflavanone known as manniflavanone (Figure 106).

Compound 18-G1 at RT 30.21 (min) exhibited both an intermediate UV intensity and ion

relative abundance. A similar 588 amu molecular weight compound was found in Fraction F,

although, it exhibited a slightly different fragmentation pattern (and the UV retention times

differed by 1 minute). In the case of this compound in Fraction G, an m/z 587 (M-H)- molecular

ion dissociated to form the following ion fragments in the ms2 spectra: m/z 569, 477, 464 (the

most intense ion), 451, 407, 375, 313, 300, and 271. The m/z 464 daughter ion underwent

177 further fragmentation to yield ms3 ions at m/z 437, 392, 313, 311, 300, 299, 287, 269, 254, and

163. The predictive fragmentation pathway for the biflavanone compound kolaflavanone follows this exact dissociation scheme (Figure 107). Therefore, compound 18-G1 is most likely kolaflavanone, or a similar derivative of this compound.

A 424 amu molecular weight compound was detected in Fraction G, which exhibited a low UV intensity and ion relative abundance. The m/z 423 (M-H)- molecular ion for compound

- 19-G1 (at RT 39.35) was confirmed by the presence of a sodium acetate [(M-H)+CH3COONa] adduct ion at m/z 505. The fragmentation pattern was nearly identical to that of two other 424 amu molecular weight compounds found in fraction F, except for small variations in low intensity ions, and a different UV retention time. As in the case of the previous compounds found in Fraction F, the molecular ion dissociated to yield the following product ions in the ms2 spectra: m/z 405, 355, 341, 313, 301, 299 (the most intense ion), 285, 271, and 259. The m/z

299 daughter ion underwent further dissociation to produce ms3 ions of m/z 283 and 271. These ion fragments follow the exact dissociation pathway for three known flavanone compounds: remangiflavanone B, euchrestaflavanone B, and sophoraflavanone G (Figure 97). Each of these flavanones follows nearly the same fragmentation scheme. Therefore, compound 19-G1 is most likely one of the three aforementioned structures.

Fraction G - HPLC Method 2

The use HPLC Method 2 exhibited better overall separation (than method 1), however, the use of this method for Fraction G only allowed the detection of few additional compounds when compared to HPLC Method 1. UV chromatograms and TICs for each scan range can be seen in Figures A.54 through A.55. A total of 4 additional compounds were identified in peanut

178 skin Fraction G obtained from Toyopearl SEC (Table 16). Several 862 amu molecular weight compounds were detected using method 2, which could not be found using Method 1.

Compound 1-G2 (at RT 37.98) and compound 2-G2 (at RT 44.27) both exhibited extremely high

UV intensities and ion relative abundances, and therefore exist at extremely high concentrations in peanut skins. In both cases, an m/z 861 (M-H)- molecular ion was confirmed by the presence

- of an acetate [(M-H)+CH3COOH] adduct ion at m/z 920, and a low intensity dimer at m/z 1723.

The molecular ion dissociated to yield the following daughter ions in the ms2 spectra: m/z 735,

720, 718, 709 (the most intense ion), 694, 451, 435, and 375. Further fragmentation produced ms3 ion fragments of m/z 584, 411, and 285. This fragmentation fingerprint follows the exact predictive dissociation pathway for a flavonoid trimer with two A-type IFL linkages. The basic structure for this compound would be: (Cat)Epi--A(IFL)--(Cat)Epi--A(IFL)--(Cat)Epi. An example of a compound such as this is aesculitannin C (Figure 108). This was further confirmed by fragmentation literature for polymeric flavonoid compounds by Gu et al. (2003). A similar

862 amu molecular weight structure was found at RT 44.27 (compound 2-G2), which had a fragmentation pattern with identical high intensity ions (although some differences in low intensity ions were noted). This compound was also confirmed as a flavonoid trimer with two

A-type linkages, although, the difference in retention time between these two compounds is most likely a function of the possible combinations of catechin/epicatchin monomers within the structure. Fragmentation data gathered from previous work done by Gu et al also allowed the confirmation of a flavonoid tetramer with 2 A-type IFL bond sites. Compound 3-G2 at RT 49.14

(which showed a high UV intensity and ion relative abundance) exhibited an m/z 1149 (M-H)- molecular ion. The presence of specific ion fragments (such as m/z 863 (M-H)-, 861, 575, and

573) can give insight into the position of one or more A-type IFL sites. According to this prior

179 study (by Gu et al), compound 3-G2 revealed product ions which are indicative of the following flavonoid tetramer: (Cat)Epi--A(IFL)--(Cat)Epi--A(IFL)--(Cat)Epi--(Cat)Epi.

Compound 4-G2 at RT 58.73 represented the last flavonoid monomer detected in all fractions A through H. An m/z 319 (M-H)- molecular ion was confirmed by the presence of an

- acetate [M+CH3COO] adduct ion at m/z 379. The molecular ion dissociated to yield ms2

product ions of m/z 301, 286, 183 (most intense ion), and 151. This follows the predictive

fragmentation scheme for the flavonoid compound 3,3',4',5,5',7-hexahydroxyflavanone (a.k.a. dihydromyricetin) (Figure 109).

Fraction H - HPLC Method 1

A total of 13 compounds were identified in peanut skin Fraction H obtained from

Toyopearl SEC (Table 17). UV chromatograms and TICs for each scan range can be seen in

Figures A.35 through A.38. Compounds detected in Fraction H using Method 1 consisted of

numerous instances of flavonoid trimers and tetramers. The extremely strong UV intensities and

ion relative abundances exhibited by these structures reveal that peanut skins contain very high

concentrations of these compounds. The use of Toyopearl size-exclusion chromatography (SEC)

allowed for the separation of many polymeric flavonoid structures via HPLC. The result was the

ability to detect multiple types of trimers and tetramers which have been previously unreported,

due to the enhanced separation exhibited by first using an HPLC preparatory method such as

Toyopearl SEC. Multiple instances of trimers and tetramers with identical molecular weights

were detected, which exhibited similar fragmentation schemes regarding the high intensity ion

fragments, although the fragmentation patterns were slightly different for less intense daughter

ions. This phenomenon can be explained by the presence of various trimer and tetramer

180 derivatives which are attached in different manners (i.e., linear versus top/middle/bottom branched forms), and the type/position of the incorporated linkages (for example, an A-type interflavanic linkage). Another explanation for the detection of many instances of similar dimers/tetramers is the numerous possibilities of catechin and epicatechin combinations. For example, a trimer can exist as catechin-epicatechin-catechin, epicatechin-catechin-catechin, catechin-catechin-catechin, etc. Therefore, based on the numerous variables involved in the formation of these structures, a vast amount of possible derivatives may exist. The detection of these abundant compounds would not be possible without the use of a prepratory method (such as SEC), since proper separation via HPLC would be unlikely due to co-elution.

Numerous m/z 1151 (M-H)- catechin/epicatchin tetramers were detected in Fraction H.

Although these compounds eluted at different retention times, they did however exhibit nearly

identical dissociation pathways (except for subtle differences in minor ion fragments). In

addition, these compounds had similar high intensity ion fragments compared to the tetramers

found in the previous Fraction G, however, slight variations in these ions (along with different

HPLC elution times) reveal that these compounds are in fact different tetramer derivatives.

Flavonoid tetramers with 1152 amu molecular weights were detected at RT 12.06 (compound 1-

H1), RT 12.67 (compound 2-H1), RT 13.93 (compound 4-H1), and RT 15.51 (compound 6-H1).

The UV intensity and ion relative abundance for these structures were extremely high, revealing

a high concentration of these compounds in peanut skins. According to previous work

conducted by Gu et al. (2003) using Electrospray Ionization in negative mode, the fragmentation

scheme of these structures (revealed by the most intense ions) suggest they are various

derivatives of catechin/epicatechin tetramers with a single A-type interflavanic linkage (IFL).

An A-type IFL refers to the form of linkage between the flavanol units, which is exhibited by an

181 added ether bond between the C-2 of one monomer, and the O-7 of another monomer (which is in addition to the common C4-C8 IFL bond, or the less common C4-C6 bond). Again, the multiple occurrences of these similar compounds at different retention times can be explained by the various positions of this A-type IFL (within the order of flavonoid monomers among the total tetramer polymer), along with other factors such as branch type (i.e., linear versus top/middle/bottom branched), and the combination/order of catechin and epicatechin monomers which make up the tetramer structure. (Figure 110). The presence of specific ion fragments

(such as m/z 863 (M-H)-, 861, 575, and 573) can give insight into the position of one or more A-

type IFL sites. According to the fragmentation literature (Gu et al), an 1152 amu tetramer with

high intensity m/z 861 and 573 (M-H)- ion fragments will have a structure of (Cat)Epi--A(IFL)-

-(Cat)Epi--(Cat)Epi--(Cat)Epi. These daughter ions were yielded by compound 1-H1 (at RT

12.06) and compound 2-H1 (at RT 12.67), which indicate the position of the A-type IFL binding

site for these structures. On the other hand, compound 4-H1 at RT 13.93 showed high intensity

m/z 863 and 573 (M-H)- ion fragments, which reveals the presence of the following A-type IFL

binded catechin/epicatechin tetramer structure: (Cat)Epi--(Cat)Epi--A(IFL)--(Cat)Epi--(Cat)Epi.

Furthermore, using the fragmentation theories outlined by Gu et al, compound 6-H1 (at RT

15.51) exhibited ion fragments of m/z 851 and 575, which indicate the presence of the following

tetramer structure with two A-type IFL binding sites: (Cat)Epi--A(IFL)--(Cat)Epi--(Cat)Epi--

A(IFL)--(Cat)Epi. Another tetramer was detected at RT 18.04 (compound 8-H1), which

exhibited an 1148 amu molecular weight. When comparing various flavonoid tetramers with A-

type IFL binding sites, the presence of each additional A-type IFL is revealed by the loss of 2

amu to the total molecular weight of the compound (because of the loss of a hydrogen for each of

the two C-C bonds). This indicates that compound 8-H1 is a catechin/epicatechin tetramer, with

182 three total A-type IFL binding sites. Therefore, the tetramer structure in this case is: (Cat)Epi--

A(IFL)--(Cat)Epi--A(IFL)--(Cat)Epi--A(IFL)--(Cat)Epi.

Two unique compounds were identified at RT 13.41 (compound 3-H1) and 18.41

(compound 9-H1). Both of these structures had 876 amu molecular weights, and exhibited nearly identical dissociation pathways. The neutral losses associated with most of the high intensity ion fragments reveal that this compound is a flavonoid trimer which is similar to an 862 amu molecular weight trimer. In the case of an 862 amu molecular weight catechin/epicatechin trimer (with two A-type IFL binding sites), an m/z 571 (M-H)- ion fragment exhibits a neutral

loss of 290 amu from the m/z 861 parent ion. The 290 amu loss is indicative of the presence of

one (of the three) catechin/epicatechin monomer units. In the case of compounds 3-H1 and 9-

H1, the m/z 571 (M-H)- ion fragment yields a neutral loss of 304 amu from the m/z 875 (M-H)- parent ion. This indicates that these compounds are both flavonoid trimers (with two A-type IFL binding sites) which consist of two catechin/epicatechin monomer units, along with one 304 amu flavonoid monomer unit. The most likely candidate for this 304 amu compound is either 3’-O- methylcatechin or 4’-O-methylcatechin. A similar situation exists with compound 5-H1 at RT

14.22, which has an 1166 amu molecular weight. Interpretation of the neutral losses for this compound reveals only one possible solution regarding the identification of this tetramer structure. First, compound 5-H1 consists of three catechin/epicatechin units connected by two

A-type IFL binding sites (as previously explained above for the 862 amu molecular weight compound). The remaining 304 amu flavonoid monomer fulfills the remainder of the tetramer polymer, and as in the case with the 876 amu compound, is most likely either 3’-O- methylcatechin or 4’-O-methylcatechin. These novel flavonoid trimer and tetramer compounds

183 which contain an O-methylated catechin are newly discovered compounds in this plant source, since they have never been previously reported to exist in peanut skins.

Several 862 amu molecular weight compounds were detected using HPLC Method 1, including compound 10-H1 (at RT 18.74), compound 11-H1 (at RT 19.41), compound 12-H1 (at

RT 21.68), and compound 13-H1 (at RT 22.95). All of these structures exhibited extremely high

UV intensities and ion relative abundances, and therefore exist at extremely high concentrations in peanut skins. In all cases, an m/z 861 (M-H)- molecular ion was confirmed by the presence of

- an acetate [(M-H)+CH3COOH] adduct ion at m/z 920, and a low intensity sextamer (i.e.,

electrospray-induced dimer of the timer) at m/z 1723. The molecular ion dissociated to yield the

following daughter ions in the ms2 spectra: m/z 825, 735, 709, and 571. In some cases, the most

intense ion in the ms2 spectra was m/z 735, while in other instances the most intense daughter

fragments was m/z 709. Further fragmentation of the most intense ms2 ion fragment produced ms3 ions of m/z 571 and 285. As seen in the previous fraction (G), this fragmentation fingerprint follows the exact predictive dissociation pathway for a flavonoid trimer with two A-

type IFL linkages. The basic structure for this compound would be: (Cat)Epi--A(IFL)--

(Cat)Epi--A(IFL)--(Cat)Epi. An example of a compound such as this is aesculitannin C (Figure

108). This was further confirmed by fragmentation literature for polymeric flavonoid

compounds by Gu et al. (2003). The differences in retention time among these four compounds

are most likely a function of the possible combinations of catechin/epicatchin monomers within

the structure.

184 Fraction H - HPLC Method 2

The use HPLC Method 2 exhibited better overall separation (than method 1), however, the use of this method for Fraction G only allowed the detection of few additional compounds when compared to HPLC Method 1. UV chromatograms and TICs for each scan range can be seen in Figures A.56 through A.57. A total of 5 additional compounds were identified in peanut skin Fraction H obtained from Toyopearl SEC (Table 18). All of these compounds consisted of

864 amu molecular weight flavonoid trimers which exhibited strong UV intensities and ion relative abundances. These structures included compound 1-H2 at RT 21.52 (min), compound 2-

H2 at RT 24.03, compound 3-H2 at RT 27.86, compound 4-H2 at RT 32.36, and compound 5-H2 at RT 35.21. According to previous work conducted by Gu et al. (2003) using Electrospray

Ionization in negative mode, the fragmentation scheme of these structures (revealed by the most intense ions) suggest they are various derivatives of catechin/epicatechin trimers with a single A- type interflavanic linkage (IFL). An A-type IFL refers to the form of linkage between the flavanol units, which is exhibited by an added ether bond between the C-2 of one monomer, and the O-7 of another monomer (which is in addition to the common C4-C8 IFL bond, or the less common C4-C6 bond). Similar 864 amu molecular weight trimer compounds were detected in the previous fraction (G), however, they exhibited different HPLC UV retention times. The multiple occurrences of these similar compounds at different retention times can be explained by the various positions of this A-type IFL (within the order of flavonoid monomers among the total trimer polymer), along with other factors such as branch type (i.e., linear versus branched), and the combination/order of catechin and epicatechin monomers which make up the trimer structure

(Figure 105).

185 CONCLUSION

This study was designed to determine if previously undiscovered phenolic antioxidant

compounds are present in peanut skin fractions obtained from Toyopearl size exclusion

chromatography (SEC), and also to identify the possible existence of structures which may have

pharmaceutical/nutraceutical properties. The use of an ESI-HPLC (negative mode Ion Trap

MSn) technique was utilized to better understand the phenolic profile of peanut skin extracts.

Although preliminary HPLC-UV studies indicated the presence of enormous amounts of

compounds in raw peanut skin extracts, previous studies have only identified a handful of compounds from this plant source (Lou et al., 1999, 2001, 2004; Yu et al., 2005). The utilization of a (-)ESI-HPLC-MSn method (HPLC Method 1) for this study enabled the successful

identification of two hundred and nineteen compounds in peanut skin fractions A through H

(obtained from Toyopearl SEC). However, the development of a second (-)ESI-HPLC-MSn method (which incorporated the use of a different column and HPLC-MSn parameters) not only

identified the same compounds detected in HPLC Method 1, but also allowed for the discovery of ninety five additional compounds in these fractions. This not only reveals that the (-)ESI-

HPLC-MSn parameters used for HPLC Method 2 offered improved separation and detection of

phenolic-based compounds in peanut skins (over HPLC Method 1), but also showed that these

settings can be custom tailored to better analyze specific bioflavonoid classes within this plant

source.

The success of HPLC Method 2 was partially due to the use of three separate SID

(Source Induced Dissociation) scan ranges, which allowed for improved detection of compounds

of different sizes and classes (as opposed to two scan ranges used in HPLC Method 1). Most

phenolic acids fall into the m/z 125-250 (M-H)- range, while most bioflavonoid monomers have

186 molecular weights which would categorize them into the m/z 230-410 range. Furthermore, most flavonoid-glycosides, dimers, trimers, and tetramers fall into the m/z 400-2000 range. The use of three SID ranges allowed for the simultaneous and systematic separation of these phenolic classes during analysis, making detection and identification easier. This method offers a significant improvement over HPLC Method 1, since the use of only two SID scan ranges caused an overlap in the aforementioned classes of compounds. Only the most intense ions in each scan range are detected and analyzed, causing less intense ions to be ejected from the ion trap. For example, HPLC Method 1 utilized scan ranges of m/z 125-600 and m/z 590-2000. Many types of phenolic compounds are encompassed in the m/z 125-600 range, including phenolic acids, bioflavanoid monomers, flavonoid-glycosides, and flavonoid dimers (proanthocyanidins and procyanidins). If two compounds elute at similar retention times, with one structure existing as a flavonoid-glycoside exhibiting a 464 amu molecular weight (while the other structure existing as a phenolic acid compound having a 164 amu molecular weight), then the compound with the lower molecular weight will go undetected (and will not undergo further MSn in the ion trap).

This undesirable phenomenon was overcome by setting the instrument parameters to three SID

scan ranges (as in the case of HPLC Method 2), which allowed for the simultaneous analysis of

compounds from various size ranges, regardless of whether they have overlapping retention

times. The incorporation of this method proved extremely useful for the analysis of peanut skin

extracts, given the vast amounts of co-eluting compounds in each SEC fraction.

The vast majority of the 314 identified compounds have never been previously reported

to exist in peanut skin extracts. Many of these compounds were positively identified as a single

structure based on its unique fragmentation pathway. However, many phenolic-based

compounds among similar classes (i.e., phenolic acids, flavones, isoflavones, flavonols,

187 flavanones, chalcones, etc.) will share identical molecular weights, and often follow extremely similar fragmentation patterns during dissociation in an ion trap mass spectrometer. For example, methoxyflavonoids chrysoeriol and diosmetin both exhibit 300 amu molecular weights

(m/z 299 M-H-), and elute at nearly the same HPLC retention time under identical

chromatographic conditions. These two methoxyflavonoids have structures which are nearly

indistinguishable, except for the position of the methoxy group on the flavonoid B-ring. In

addition, the MSn (MS/MS/MS) fragmentation scheme for these compounds is nearly identical,

making positive identification difficult. This theory holds true for many other phenolic-based

compounds (even between similar classes), because of the numerous derivatives which are

capable of being formed at any given molecular weight. Therefore, many cases existed (for the

identification of structures in peanut skin fractions A through H) where compounds were

“tentatively identified” as being one possible compound out of a group of structures. The

obvious reason for offering several compounds for the purpose of identification is due to the fact

that more than one structure exists which is capable of following the same dissociation pathway.

In these situations where multiple compounds share both molecular weight and fragmentation

attributes, only the use of standards would allow for full identification of structures via matching

HPLC-UV retention times. Unfortunately, the large quantity of standards required to fully

classify these “tentatively identified” compounds were not available for this study.

In a situation similar to that explained above, many compounds which exhibit the same

molecular weight (and fragmentation patterns) were identified in one or more fractions, but with

different retention times. This is due to the multiple possible derivatives which are capable in

each individual class of phenolic compounds (i.e., flavonoids, chalcones, phenolic acids, etc.).

For example, a flavone existing with a 302 amu molecular weight may have many possible

188 structure variations due to the numerous locations in which hydroxyl units can reside on the flavonoid rings. These compounds include the following 302 amu molecular weight flavones: quercetin, hesperitin, morin, robinetin, tricetin, and . Numerous 302 amu compounds (which belong to the same flavone phenolic class) were detected in multiple peanut skin fractions at different retention times. The number of distinct 302 amu compounds identified reveals that every possible aforementioned flavone derivative (with this given molecular weight) exists in peanut skins. These various compounds were found in both the monomer form, and also in their flavonoid-glycoside forms. This phenomenon also existed with various other flavonoid (and/or other phenolic compound) classes, where numerous distinct structures were identified with the same molecular weight. For example, numerous flavonoid compounds exist in nature which exhibit a 300 amu molecular weight, including chrysoeriol, diosmetin, kaempferide, geraldol, pratensein, farresol, tectorigenin, and psi-tectorigenin. The vast number of distinct 300 amu compounds detected among the various peanut skin fractions shows that most (if not all) of these compounds exist in this plant source. As in the case with the 302 amu structures, some of these existed as flavonoid-glycosides (in addition to their monomer counterparts). Other examples existed with the detection of numerous identical molecular weight phenolic-compounds, such as in the case of 316 amu, 288 amu, 286 amu, and 284 amu compounds. This reveals that the majority of possible phenolic (flavonoid, chalcone, ect.) derivatives for each individual molecular weight exist in peanut skins. This phenomenon also holds true for the various detected flavonoid-glycosides which exhibited identical molecular weights (and similar fragmentation patterns), but were revealed at different retention times. For example, numerous 464 amu flavonoid glycosides were found among the different fractions, indicating that many flavonoid-glycosides with identical molecular weights exist in peanut skins

189 (such as isoquercetrin, hyperoside, gossypitrin, and spiraeoside). Another example where numerous flavonoid-glycosides were detected for a given molecular weight included 448 amu compounds, which are encompassed by various luteolin and kaempferol derivatives.

Compounds detected among the various peanut skin fractions (obtained from Toyopearl

SEC) generally followed a trend from low-to-high molecular weight for each consecutive fraction (A through H). Phenolic acids were detected in early fractions (A and B), but were also discovered in small concentrations in each consecutive fraction. Phenolic-based monomers

(flavonoids, chalcones, chromanes, chromenes, etc.), as well as their sugar-linked derivatives, were generally found in fractions A through E. Higher molecular weight compounds, such as oligomeric procyanidins (OPC) structures consisting of flavonoid dimers (including numerous novel biflavonoids), trimers, and tetramers, were generally found in fractions F through H.

The use of HPLC Method 1 allowed for the identification of 53 compounds in Fraction

A, while HPLC Method 2 detected the presence of 8 additional compounds in this fraction

(Tables 1 and 2). These structures mostly consisted of small-to-intermediate size compounds

(168 to 720 amu) which were comprised of phenolic acids (such as cinnamic and benzoic derivatives), stilbene-derivatives (resveratrol-based), coumarin-based compounds, and chalcone- based structures. A large number of additional distinct phenolic acids existed in Fraction A which could not be identified (and were therefore not reported), due to difficulties in interpreting their fragmentation pathways. Other compounds in Fraction A included flavonoids (or isoflavonoids) with one or more hydroxy and/or methoxy group(s). Many of these compounds were bound by sugar side groups consisting of mono- or disaccharides, such as in the case of various glycosylated flavonoids (and possibly their glycosylated iso-flavonoid derivatives).

Most of these structures are less known flavonoid compounds, and have never been reported to

190 exist in peanut plant parts. However, some of the glycosylated flavonoids found in Fraction A incorporate aglycones which are widely recognized in bioflavanoid research, such as eriodictyol and luteolin. Fraction A revealed the highest number of individual compounds when compared to all other fractions obtained from Toyopearl SEC, however, these compounds were generally not present in extremely high concentrations.

A total of 24 compounds were identified in Fraction B using HPLC Method 1 (Table 3).

The utilization of HPLC Method 2 allowed for the detection of 27 additional compounds (Table

4). These structures ranged from small-to-intermediate size compounds (138 to 934 amu), and mostly consisted of numerous phenolic acids (benzoic and cinnamic acid derivatives), chalcones, chromones, chromenes, chromanediols, and flavonoids (or isoflavonoids) with one or more hydroxyl and/or methoxy group(s). As in the case of Fraction A, most of these compounds are newly discovered in peanut skins. Numerous high concentration flavonoid-glycosides were also detected in Fraction B, of which most had aglycones consisting of quercetin, eriodictyol, isorhamnetin, and hesperitin.

Fraction C consisted of 69 total compounds, of which 35 were detected using HPLC

Method 1, and 34 were identified using HPLC Method 2 (Tables 5 and 6). Many of these compounds were flavonoids (or isoflavonoids) with one or more hydroxyl or methoxy groups.

As in the case of previous fractions, many of these compounds have never been previously reported in peanut skins. Numerous high-concentration flavonoid-glycosides were also found, many consisting of aglycones which are more commonly found to exist as flanvoid glycosides

(such as quercetin, isorhamnetin, luteolin, and kaempferol). However, several of these flavonoid-glycosides are made up of uncommon aglycone moieties which have never been reported to exist in peanut skins (such as fraxetin, myricetin, quercetagetin, gossypetin, and

191 fraxetin). Other detected compounds include various coumarin-based structures.

The use of HPLC Method 1 allowed for the identification of 15 compounds in Fraction

D, while HPLC Method 2 detected the presence of 7 additional compounds in this fraction

(Tables 7 and 8). Fraction E contained a total of 12 compounds, of which 9 were identified using

HPLC Method 1, and 3 using HPLC Method 2 (Tables 9 and 10). The compounds identified in

Fractions D and E included extremely high concentrations of flavonoid monomers and flavonoid-glycosides. Several of these compounds have been previously reported in numerous studies (such as catechin, epicatechin, and quercetin), however, the majority of the remaining compounds have never been determined to exist in peanut skins. Some of these monomers included diosmetin, chrysoeriol, homoeriodictyol, morin, robinetin, and tricetin. Flavonoid- glycosides included isoquercitrin, hyperoside, gossypitrin, and spiraeosdie. The large number of these identified high-concentration compounds indicates the wide range of diverse flavonoid compounds which reside in this plant source. Several researchers have attempted to analyze the flavonoid profile of peanut skins, however, these studies did not incorporate a method (such as

Toyopearl SEC coupled with HPLC ion trap MSn) which is capable of fully recognizing these

types of structures.

A total of 20 compounds were identified in Fraction F using HPLC Method 1, while

HPLC Method 2 detected the presence of 3 additional compounds in this fraction (Tables 11 and

12). Fraction G-Red contained a total of 16 compounds, of which 12 were identified using

HPLC Method 1, and 4 using HPLC Method 2 (Tables 13 and 14). Although several flavonoid-

based monomers were detected in these fractions (such as luteolin and hesperitin), the majority

of the compounds identified consisted of extremely high concentrations of flavonoid dimers.

Most of these dimer structures were numerous B-type procyanidins, which indicates that many

192 dimeric flavonoid-based derivatives exist in peanut skins containing multiple combinations of catechin/epicatechin monomers in various linear and branched forms. Several A-type proanthocyanidin dimers were also discovered, revealing that numerous catechin/epicatechin derivatives of these structures are also present.

The use of HPLC Method 1 allowed for the identification of 19 compounds in Fraction

G, while HPLC Method 2 detected the presence of 4 additional compounds in this fraction

(Tables 15 and 16). Fraction H contained a total of 18 compounds, of which 13 were identified using HPLC Method 1, and 5 using HPLC Method 2 (Tables 17 and 18). These fractions consisted mainly of numerous catechin/epicatechin trimers and tetramers at extremely high concentrations. The vast number of unique trimers and tetramers indicates that many derivatives exist in peanut skins, mainly due to the multiple combinations which are possible with oligomeric procyanidins (OPC’s). These derivatives consist of various branch forms (linear vs branched), along with the order of catechin versus epicatechin within the flavonoid polymer.

The vast number of phenolic-based compounds identified in this study reveals that peanut skins are possibly one of the most phenolic-rich and diverse plant sources ever studied. The majority of these compounds are known to exhibit powerful antioxidant activity, indicating that peanut skins could be used as a novel source for the extraction of natural antioxidant and preservative components. This proves the capability that peanut skins, which had been previously viewed as a by-product (or waste-product) of the peanut industry, can become a value-added commodity and revenue stream for peanut farmers.

Numerous compounds were detected which have been reported to exhibit specific chemical, biological, nutraceutical, and pharmaceutical attributes (Table 19). This indicates the possibility that peanut skins could be utilized as a novel source for the extraction of compounds

193 which have the potential to treat ailments such as cardiovascular disease and various forms of cancer. All of these factors showcase the need for future research into the biological effects of components contained within peanut skins. Future research could use this study as a template, to further narrow down the numerous compounds which were tentatively identified (many of which have significant antioxidant and biological effects).

194 Table 1: Detected Compounds in Fraction A using HPLC Method 1. Indicates Compound Number (CMPD #), Molecular Weight (MW) in atomic mass units (amu), Retention Time (RT) in minutes (min), Mass Spec Base Peak Identification (BP-ID) showing negative mode ionization [M-H]- (indicates the type of ion which forms the Base Peak. Base Peak refers to the most abundant ion in the mass spectrum), Parent ion (P-) m/z after [M-H]- ionization (indicates the Base Peak in the normal mass spectra), Base Peak of the daughter ion (D-BP) (i.e., the most abundant ion in the daughter spectrum), MS/MS or MS/MS/MS product ions (MS^3), and the arithmetic difference in m/z between the parent ion and individual daughter ions (P-Di). Bold donates most abundant ion in the given spectra. CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 402 1.69 [M-H]- 401 ==> 341 Dihydroxy-cinnamic Acid (such as 3,4- 1-A1 P-Di 60 Dihydroxycinnamic acid (Caffeic Acid)) MS^3 341 ==> 281 179 161 149 143 131 125 119 plus a single glucose sugar unit P-Di 60 162 180 192 198 210 216 222 444 7.78 [M-H]- 443 ==> 383 Hydroxy-flavone, or isoflavone (such 2-A1 P-Di 60 as Flavonol, Primuletin, or 6- or 7- MS^3 383 ==> 355 299 247 237 229 179 163 161 Hydroxyflavone), or Methoxy-Chalcone P-Di 28 84 136 146 154 204 220 222 plus a single rhamonse sugar unit 302 8.54 [M-H]- 301 ==> 241 217 3-A1 P-Di 60 84 3',4',5,5',7-Pentahydroxyflavone MS^3 241 ==> 218 199 197 150 138 126 (a.k.a. Tricetin) P-Di 23 42 44 91 103 115 380 8.69 [M-H]- 379 ==> 361 335 299 217 4,4'-Dihydroxy-3,3'-(2-methoxy- 4-A1 P-Di 18 44 80 162 ethylidene)-dicoumarin (a.k.a. MS^3 361 ==> 291 272 271 255 Dicoumoxyl) P-Di 70 89 90 106 396 8.84 [M-H]- 395 ==> 335 319 3,3'-Methylenebis[4-hydroxycoumarin] 5-A1 P-Di 60 76 (a.k.a. Dicoumarin), or MS^3 335 ==> 317 291 275 273 249 247 229 201 173 Coumarin-6-(7-hydroxycoumarin-8-yl)- P-Di 18 44 60 62 86 88 106 134 162 7-methoxy 458 9.90 [M-H]- 457 ==> 397 6-A1 P-Di 60 457 457 457 3-Methoxyflavone, with a rhamnose MS^3 397 ==> 355 335 265 251 235 163 161 143 sugar side-group P-Di 42 62 132 146 162 234 236 254 217 10.20 [M-H]- 216 ==> 198 7-A1 P-Di 18 6-Hydroxy-2-naphthalenepropanoic MS^3 154 ==> Acid (a.k.a. Allenolic Acid) P-Di 398 11.56 [M-H]- 397 ==> 379 351 337 333 315 307 305 289 287 8-A1 P-Di 18 46 60 64 82 90 92 108 110 UNKOWN, perhaps Coumaroylquinic MS^3 379 ==> 279 277 273 259 253 249 241 197 187 Acid derivative P-Di 100 102 106 120 126 130 138 182 192 382 11.86 [M-H]- 381 ==> 335 321 299 217 UNKOWN, according to NIST, similar 9-A1 P-Di 46 60 82 164 to Chromone, 2-[2-(3,5-diacetoxy- MS^3 321 ==> 273 219 191 137 phenyl)ethyl]-3-methyl P-Di 48 102 130 184 396 12.16 [M-H]- 395 ==> 337 321 319 307 293 UNKOWN, perhaps Coumaroylquinic 10-A1 P-Di 58 74 76 88 102 Acid derivative. According to NIST, this MS^3 337 ==> 305 293 291 275 261 231 may be a Chromone derivative P-Di 32 44 46 62 76 106 300 12.47 [M-H]- 299 ==> 137 11-A1 P-Di 162 Reported in Fraction B due to higher MS^3 137 ==> 93 concentrations present P-Di 44 398 13.05 [M-H]- 397 ==> 337 293 251 UNKOWN, perhaps Coumaroylquinic 12-A1 P-Di 60 104 146 Acid derivative. According to NIST, this MS^3 337 ==> 293 275 249 231 219 may be a Chromone derivative P-Di 44 62 88 106 118 398 14.07 [M-H]- 397 ==> 337 321 251 Flavanone compound similar to 13-A1 P-Di 60 76 146 Bavachinin A. May also be isoflavone MS^3 337 ==> 293 275 231 derivative similar to a compound P-Di 44 62 106 known as Hydroxywighteone 396 15.58 [M-H]- 395 ==> 321 14-A1 P-Di 74 UNKNOWN, NIST suggests a MS^3 321 ==> 303 277 259 247 233 203 Chromone derivative P-Di 18 44 62 74 88 118 396 15.88 [M-H]- 395 ==> 321 15-A1 P-Di 74 UNKNOWN, NIST suggests a MS^3 321 ==> 303 277 259 247 233 203 Chromone derivative P-Di 18 44 62 74 88 118

195 Table 1: (cont.) CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 378 16.30 [M-H]- 377 ==> 333 315 297 289 271 259 229 201 16-A1 P-Di 44 62 80 88 106 118 148 176 UNKNOWN MS^3 333 ==> 303 289 287 271 269 219 213 193 187 P-Di 30 44 46 62 64 114 120 140 146 168 17.01 [M-H]- 167 ==> 152 109 108 Either Vanillic Acid, Homogentisic 17-A1 P-Di 15 58 59 Acid, Hydroxy-methoxy-benzoic Acid, MS^3 152 ==> 123 109 108 Homoprotocatechuic Acid, Orsellinic P-Di 29 43 44 Acid, or Resorcylic Acid derivative 476 17.42 [M-H]- 475 ==> 415 Rhamnose sugar plus either a 18-A1 P-Di 60 trihydroxyflavone, trihydroxyiso- MS^3 415 ==> 299 269 247 161 158 flavone, hydroxy-methoxy-flavanone, P-Di 116 146 168 254 257 or dihydroxy-methoxy-chalcone 240 17.98 [M-H]- 239 ==> 179 Dihydroxy-benzoic Acid 19-A1 P-Di 60 (such as Caffeic Acid), less a 60 amu MS^3 179 ==> 161 159 87 acetate adduct ion P-Di 18 20 92 188 18.43 [M-H]- 187 ==> 125 20-A1 P-Di 62 Benzoic Acid derivative, fragments MS^3 125 ==> 97 81 57 similar to Caffeic Acid P-Di 28 44 68 444 19.04 [M-H]- 443 ==> 425 341 305 299 21-A1 P-Di 18 102 138 144 Glucose-linked Hydroxy-benzoic Acid, MS^3 299 ==> 137 with Coumaroyl acyl side-group P-Di 162 596 19.42 [M-H]- 595 ==> 535 22-A1 P-Di 60 Resveratrol, plus rutinose sugar MS^3 535 ==> 443 389 307 247 227 187 side-group P-Di 92 146 228 288 308 348 174 19.94 [M-H]- 173 ==> 155 143 142 129 111 109 102 83 81 23-A1 P-Di 18 30 31 44 62 64 71 90 92 Shikimic Acid MS^3 111 ==> 83 57 P-Di 28 54 596 20.21 [M-H]- 595 ==> 579 535 Eriodictyol-7-neohesperidoside (a.k.a. 24-A1 P-Di 16 60 neoeriocitrin), or Eriodictyol 7-O- MS^3 535 ==> 487 389 367 271 263 251 225 rutinoside (a.k.a. Eriocitrin) P-Di 48 146 168 264 272 284 310 580 21.83 [M-H]- 579 ==> 411 285 25-A1 P-Di 168 294 Luteolin, plus either a sambubiose or MS^3 285 ==> 267 257 241 217 213 201 175 163 147 lathyrose disaccharide side-group P-Di 18 28 44 68 72 84 110 122 138 244 24.26 [M-H]- 243 ==> 225 181 Either 7-methoxy-8-isopentenyl- 26-A1 P-Di 18 62 coumarin (Osthole), 7-methoxy-6-iso- MS^3 225 ==> 207 181 163 99 97 pentenylcoumarin (Suberosin), Oxy- P-Di 18 44 62 126 128 resveratrol, or Cedrecoumarin A 188 25.09 [M-H]- 187 ==> 169 157 151 125 97 Either a Benzoic Acid derivative 27-A1 P-Di 18 30 36 62 90 (similar to Gallic Acid), or Coumarin- MS^3 125 ==> 97 83 69 57 like structure, or 5-Hydroxy-2-methyl- P-Di 28 42 56 68 1,4-naphtoquinone (a.k.a. Plumbagin). 244 26.39 [M-H]- 243 ==> 225 207 Either 7-methoxy-8-isopentenyl- 28-A1 P-Di 18 36 coumarin (Osthole), 7-methoxy-6-iso- MS^3 225 ==> 207 181 163 97 pentenylcoumarin (Suberosin), Oxy- P-Di 18 44 62 128 resveratrol, or Cedrecoumarin A 242 28.24 [M-H]- 241 ==> 223 197 179 161 29-A1 P-Di 18 44 62 80 Sinnapic Acid-like (plus water) MS^3 223 ==> 179 161 153 151 137 125 123 111 95 P-Di 44 62 70 72 86 98 100 112 128 170 30.46 [M-H]- 169 ==> 125 30-A1 P-Di 44 2,4,6-Trihydroxybenzoic acid (or other MS^3 125 ==> 95 83 71 57 Trihydroxybenzoic acid derivative) P-Di 30 42 54 68 720 31.46 [M-H]- 719 ==> 659 497 31-A1 P-Di 60 222 Oleuropein, plus additional glucose MS^3 659 ==> 581 539 497 479 437 425 395 377 377 sugar unit P-Di 78 120 162 180 222 234 264 282 282 462 34.00 [M-H]- 461 ==> 401 32-A1 P-Di 60 Possible Hexamethoxyflavone (such MS^3 401 ==> 327 309 3',4',5,5',6,7-Hexamethoxyflavone) P-Di 74 92

196 Table 1: (cont.) CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 326 34.81 [M-H]- 325 ==> 307 289 Fragments similar to 5-Hydroxy-4',7- 33-A1 P-Di 18 36 dimethoxy-6,8-dimethylflavone (a.k.a. MS^3 307 ==> 289 235 185 137 121 Eucalyptin), and 5,7-Dimethoxy-3-(4'- P-Di 18 72 122 170 186 methoxyphenyl)-4-methylcoumarin 558 35.05 [M-H]- 557 ==> 497 Coumarin compound, such as 34-A1 P-Di 60 Psoralidin, Dicoumarin, Lasiocephalin, MS^3 497 ==> 335 or 6-(7-hydroxycoumarin-8-yl)-7- P-Di 162 methoxy-coumarin, plus a glucose unit 328 35.44 [M-H]- 327 ==> 309 291 239 171 Compound similar to: 35-A1 P-Di 18 36 88 156 Kaempferol-3,4',7-trimethyl ether, or MS^3 309 ==> 291 251 171 3-hydroxy-4’,5,7-trimethoxyflavone P-Di 18 58 138 328 35.80 [M-H]- 327 ==> 309 291 239 229 221 211 209 171 165 Compound similar to: 36-A1 P-Di 18 36 88 98 106 116 118 156 162 Kaempferol-3,4',7-trimethyl ether, or MS^3 309 ==> 291 253 227 125 3-hydroxy-4’,5,7-trimethoxyflavone P-Di 18 56 82 184 464 36.08 [M-H]- 463 ==> 403 37-A1 P-Di 60 UNKNOWN MS^3 403 ==> 329 311 211 P-Di 74 92 192 330 37.70 [M-H]- 329 ==> 309 291 229 211 209 185 171 38-A1 P-Di 20 38 100 118 120 144 158 UNKNOWN, possible MS^3 309 ==> 305 291 191 Hydroxy-trimethoxy-flavanone P-Di 4 18 118 328 38.27 [M-H]- 327 ==> 309 291 229 211 209 185 171 Compound similar to: 39-A1 P-Di 18 36 98 116 118 142 156 Kaempferol-3,4',7-trimethyl ether, or MS^3 309 ==> 305 291 191 3-hydroxy-4’,5,7-trimethoxyflavone P-Di 4 18 118 330 38.44 [M-H]- 329 ==> 309 291 229 211 209 185 171 40-A1 P-Di 20 38 100 118 120 144 158 UNKNOWN, possible MS^3 309 ==> 305 291 191 Hydroxy-trimethoxy-flavanone P-Di 4 18 118 328 38.80 [M-H]- 327 ==> 309 307 291 289 229 213 201 171 Compound similar to: 41-A1 P-Di 18 20 36 38 98 114 126 156 Kaempferol-3,4',7-trimethyl ether, or MS^3 309 ==> 305 291 289 245 209 191 171 163 137 3-hydroxy-4’,5,7-trimethoxyflavone P-Di 4 18 20 64 100 118 138 146 172 326 40.71 [M-H]- 325 ==> 307 42-A1 P-Di 18 Fragmentation similar to Eucalyptin MS^3 307 ==> 289 209 125 P-Di 18 98 182 446 41.85 [M-H]- 445 ==> 385 Either 4',7-Dihydroxy-6-methoxy- 43-A1 P-Di 60 isoflavone-7-D-glucoside (Glycitin), MS^3 385 ==> 311 293 Biochanin-A-7-glucoside (Sissotrin), or P-Di 74 92 Baicalein-7-O-glucuronide 360 42.08 [M-H]- 359 ==> 311 291 289 223 209 183 125 UNKOWN, fragmentation pattern 44-A1 P-Di 48 68 70 136 150 176 234 similar to 3',5,7-Trihydroxy-4',5',6- MS^3 311 ==> 305 273 271 255 247 217 197 189 163 trimethoxyisoflavone (aka Irigenin) P-Di 6 38 40 56 64 94 114 122 148 360 42.49 [M-H]- 359 ==> 341 323 171 169 UNKOWN, fragmentation pattern 45-A1 P-Di 18 36 188 190 similar to 3',5,7-Trihydroxy-4',5',6- MS^3 341 ==> 291 193 169 153 151 149 trimethoxyisoflavone (aka Irigenin) P-Di 50 148 172 188 190 192 312 43.66 [M-H]- 311 ==> 293 223 46-A1 P-Di 18 88 3,5,7-Trimethoxyflavone (according to MS^3 293 ==> 275 247 235 221 149 141 NIST 05 Database) P-Di 18 46 58 72 144 152 328 43.98 [M-H]- 327 ==> 309 291 Compound similar to: 47-A1 P-Di 18 36 Kaempferol-3,4',7-trimethyl ether, or MS^3 309 ==> 291 289 247 245 209 193 155 153 3-hydroxy-4’,5,7-trimethoxyflavone P-Di 18 20 62 64 100 116 154 156 310 44.37 [M-H]- 309 ==> 293 291 209 48-A1 P-Di 16 18 100 Fragmentation similar to MS^3 293 ==> 275 273 257 247 215 193 185 165 149 Maxima-Isoflavone A P-Di 18 20 36 46 78 100 108 128 144 312 44.37 [M-H]- 311 ==> 293 291 209 Fragmentation similar to 49-A1 P-Di 18 20 102 Maxima-Isoflavone G, and MS^3 293 ==> 275 273 257 247 215 193 185 165 149 Maxima-Isoflavone E P-Di 18 20 36 46 78 100 108 128 144

197 Table 1: (cont.) CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 314 45.31 [M-H]- 313 ==> 295 201 171 50-A1 P-Di 18 112 142 Fragmentation similar to MS^3 295 ==> 183 181 171 165 155 153 137 127 125 Pectolinarigenin P-Di 112 114 124 130 140 142 158 168 170 314 49.57 [M-H]- 313 ==> 295 51-A1 P-Di 18 Fragmentation similar to MS^3 295 ==> 277 275 251 151 141 127 111 Pectolinarigenin P-Di 18 20 44 144 154 168 184 402 49.82 [M-H]- 401 ==> 341 313 289 274 260 52-A1 P-Di 60 88 112 127 141 UNKNOWN MS^3 341 ==> 324 314 310 306 295 282 265 166 P-Di 17 27 31 35 46 59 76 175 272 53.55 [M-H]- 271 ==> 253 225 53-A1 P-Di 18 46 UNKNOWN MS^3 253 ==> 221 201 195 83 P-Di 32 52 58 170

198 Table 2: Detected Compounds in Fraction A using HPLC Method 2. Indicates Compound Number (CMPD #), Molecular Weight (MW) in atomic mass units (amu), Retention Time (RT) in minutes (min), Mass Spec Base Peak Identification (BP-ID) showing negative mode ionization [M-H]- (indicates the type of ion which forms the Base Peak. Base Peak refers to the most abundant ion in the mass spectrum), Parent ion (P-) m/z after [M-H]- ionization (indicates the Base Peak in the normal mass spectra), Base Peak of the daughter ion (D-BP) (i.e., the most abundant ion in the daughter spectrum), MS/MS or MS/MS/MS product ions (MS^3), and the arithmetic difference in m/z between the parent ion and individual daughter ions (P-Di). Bold donates most abundant ion in the given spectra. CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Structure ID 240 4.08 [M-H]- 239 ==> 179 1-A2 to P-Di 60 Dihydroxycinnamic Acid(s) 17.77 MS^3 179 ==> 161 145 141 125 117 101 99 91 P-Di 18 34 38 54 62 78 80 88 222 16.72 [M-H]- 221 ==> 161 Coumarin-based compound, such 2-A2 P-Di 60 as 3-Hydroxycoumarin, MS^3 161 ==> 143 125 4-Coumarinol, 4-Hydroxycoumarin, or P-Di 18 36 7-Hydroxycoumarin (Umbelliferone) 174 40.05 [M-H]- 173 ==> 155 143 142 129 111 109 102 83 81 3-A2 P-Di 18 30 31 44 62 64 71 90 92 Shikimic Acid MS^3 111 ==> P-Di 244 48.08 [M-H]- 243 ==> 225 4-A2 49.51 P-Di 18 Cedrecoumarin A, and 52.88 MS^3 225 ==> Oxyresveratrol 53.57 P-Di 188 50.96 [M-H]- 187 ==> 169 157 151 125 97 Either a Benzoic Acid derivative 5-A2 P-Di 18 30 36 62 90 (similar to Gallic Acid), or Coumarin- MS^3 169 ==> like structure, or 5-Hydroxy-2-methyl- P-Di 1,4-naphtoquinone (a.k.a. Plumbagin). 242 54.67 [M-H]- 241 ==> 223 205 197 179 161 Either a Dihydroxyflavan, such as 4’,7- 6-A2 P-Di 18 36 44 62 80 Dihydroxyisoflavan (a.k.a. Equol), or a MS^3 223 ==> Flavandiol such as 4,4’-flavandiol or P-Di 4,7-flavandiol 170 57.53 [M-H]- 169 ==> 153 125 83 71 7-A2 P-Di 16 44 86 98 2,4,6-Trihydroxybenzoic acid (or other MS^3 307 ==> Trihydroxybenzoic acid derivative) P-Di 202 60.30 [M-H]- 201 ==> 183 139 Hydroxyfurocoumarin, such as 8-A2 P-Di 18 62 5-Hydroxyfurocoumarin (Bergaptrol) MS^3 183 ==> and 8-Hydroxy-4'-5',6-7-furocoumarin P-Di (a.k.a. Xanthotoxol)

199 Table 3: Detected Compounds in Fraction B using HPLC Method 1. Indicates Compound Number (CMPD #), Molecular Weight (MW) in atomic mass units (amu), Retention Time (RT) in minutes (min), Mass Spec Base Peak Identification (BP-ID) showing negative mode ionization [M-H]- (indicates the type of ion which forms the Base Peak. Base Peak refers to the most abundant ion in the mass spectrum), Parent ion (P-) m/z after [M-H]- ionization (indicates the Base Peak in the normal mass spectra), Base Peak of the daughter ion (D-BP) (i.e., the most abundant ion in the daughter spectrum), MS/MS or MS/MS/MS product ions (MS^3), and the arithmetic difference in m/z between the parent ion and individual daughter ions (P-Di). Bold donates most abundant ion in the given spectra. CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Structure ID 154 8.78 [M-H]- 153 ==> 109 Dihydroxybenzoic Acid, such as 1-B1 P-Di 44 Resorcylic Acid, Gentisic Acid, MS^3 109 ==> 91 81 and Protocatechuic Acid P-Di 18 28 138 11.07 [M-H]- 137 ==> 119 109 93 81 65 Hydroxybenzoic Acid, such as 2-B1 P-Di 18 28 44 56 72 various Salicylic Acid derivatives. MS^3 119 ==> Also, this compound may be P-Di Pyrocatechin-Monoethyl-Ether 300 12.54 [M-H]- 299 ==> 137 Either Hydroxybenzoic Acid or 3-B1 P-Di 162 Pyrocatechin-Monoethyl-Ether, MS^3 137 ==> 93 plus an added glucose or P-Di 44 galactose side-group 416 12.98 [M-H]- 415 ==> 313 253 Dihydroxyflavone (such as 3,4’- 4-B1 P-Di 102 162 Dihydroxyflavone) or Dihydroxy- MS^3 313 ==> 253 235 203 193 151 109 isoflavone (such as Daidzein), P-Di 60 78 110 120 162 204 plus glucose (or galactose) unit 450 16.27 [M-H]- 449 ==> 287 269 259 179 Likely Eriodictyol-7-glucoside, but 5-B1 P-Di 162 180 190 270 aglycone may also be either MS^3 287 ==> 259 243 215 201 199 187 175 157 125 Fustin, Hydroxyapigenin, or P-Di 28 44 72 86 88 100 112 130 162 Dihydrokeampferol 474 17.08 [M-H]- 473 ==> 311 293 179 Trisaccharide: Two glucose (or 6-B1 P-Di 162 180 294 galactose) units, plus one xylose MS^3 311 ==> 179 149 135 (or arabinose) unit P-Di 132 162 176 934 17.81 [M-H]- 933 ==> 771 Quercetin Glycoside. Sugar group 7-B1 P-Di 162 consists of Rutinose plus added MS^3 771 ==> 753 625 609 591 549 408 367 343 301 dissacharide (either Sophorose, P-Di 18 146 162 180 222 363 404 428 470 Gentiobiose, or Laminaribiose) 164 18.32 [M-H]- 163 ==> 119 8-B1 P-Di 44 p-Hydroxycinnamic Acid (a.k.a. MS^3 119 ==> p-Coumaric Acid) P-Di 458 18.92 [M-H]- 457 ==> 440 311 295 293 277 221 204 163 9-B1 P-Di 17 146 162 164 180 236 253 294 6-O-Acetyldaidzin MS^3 293 ==> 275 219 191 187 179 163 155 139 127 P-Di 18 74 102 106 114 130 138 154 166 488 19.34 [M-H]- 487 ==> 325 307 293 Ferulic Acid, plus a disaccharide 10-B1 P-Di 162 180 194 side group (either sambubiose MS^3 325 ==> 193 149 113 or lathyrose) P-Di 132 176 212 904 19.39 [M-H]- 903 ==> 741 Quercetin plus rutinose and 11-B1 P-Di 162 lathyrose (or sambubiose), or MS^3 741 ==> 723 609 591 573 343 327 300 271 255 Isorhamnetin plus combination of P-Di 18 132 150 168 398 414 441 470 486 2 sambubiose’s (or lathyrose’s) 904 19.39 [M-H]- 903 ==> 741 Quercetin plus rutinose and 12-B1 P-Di 162 lathyrose (or sambubiose), or MS^3 741 ==> 723 609 591 573 545 475 355 343 337 Isorhamnetin plus combination of P-Di 18 132 150 168 196 266 386 398 404 2 sambubiose’s (or lathyrose’s) 772 19.66 [M-H]- 771 ==> 609 591 573 505 465 447 301 300 271 Quercetin Glycoside. Sugar group 13-B1 P-Di 162 180 198 266 306 324 470 471 500 consists of Rutinose plus added MS^3 609 ==> 283 273 271 255 244 229 199 179 151 glucose (or galactose) P-Di 326 336 338 354 365 380 410 430 458 138 20.14 [M-H]- 137 ==> 93 Hydroxybenzoic Acid, such as 14-B1 P-Di 44 3-Hydroxybenzoic Acid or MS^3 93 ==> 4-Hyrdoxybenzoic Acid P-Di 786 21.49 [M-H]- 785 ==> 753 623 605 519 423 357 339 315 314 15-B1 P-Di 32 162 180 266 362 428 446 470 471 Isorhamnetin-3-O-rutinoside, plus MS^3 753 ==> 301 300 299 287 285 272 271 256 247 additional glucose (or galactose) P-Di 452 453 454 466 468 481 482 497 506

200 Table 3: (cont.) CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Structure ID 640 22.18 [M-H]- 639 ==> 477 459 357 339 316 315 314 300 299 Isorhamnetin plus disaccharide 16-B1 P-Di 162 180 282 300 323 324 325 339 340 side-group (either laminaribiose MS^3 477 ==> 301 300 299 287 285 272 271 259 256 getiobiose, or sophorose) P-Di 176 177 178 190 192 205 206 218 221 474 23.01 [M-H]- 473 ==> 311 293 17-B1 P-Di 162 180 Trisaccharide: 2 glucose units plus MS^3 311 ==> 179 149 135 1 xylose unit P-Di 132 162 176 756 23.31 [M-H]- 755 ==> 723 623 605 577 489 315 314 300 271 Isorhamnetin plus rutinose 18-B1 P-Di 32 132 150 178 266 440 441 455 484 disaccharide, plus additional MS^3 723 ==> 301 300 299 285 271 257 247 241 183 xylose (or arabinose) sugar P-Di 422 423 424 438 452 466 476 482 540 464 24.54 [M-H]- 463 ==> 301 19-B1 P-Di 162 Hesperitin plus glucose (or MS^3 301 ==> 287 286 271 256 217 183 164 151 galactose) P-Di 14 15 30 45 84 118 137 150 462 27.59 [M-H]- 461 ==> 299 Flavonoid (either Chrysoeriol, 20-B1 P-Di 162 Diosmetin, or other 300 amu MS^3 299 ==> 284 flavonoid), plus monosaccharide P-Di 15 unit (glucose or galactose) 436 49.11 [M-H]- 435 ==> 420 378 351 312 299 271 Chalcone, Chromone, or 21-B1 P-Di 15 57 84 123 136 164 Chromene-based compound. MS^3 420 ==> 405 377 351 338 309 298 283 270 259 Similar to Orotinichalcone and P-Di 15 43 69 82 111 122 137 150 161 2’-O-Methylcajanone 436 50.90 [M-H]- 435 ==> 365 Chalcone, Chromone, or 22-B1 P-Di 70 Chromene-based compound. MS^3 365 ==> 350 321 310 295 Similar to Orotinichalcone and P-Di 15 44 55 70 2’-O-Methylcajanone 420 53.00 [M-H]- 419 ==> 404 364 361 349 321 23-B1 P-Di 15 55 58 70 98 Chalcone, Chromone, MS^3 404 ==> 361 349 305 256 or Chromene-based compound P-Di 43 55 99 148 420 53.36 [M-H]- 419 ==> 404 364 335 296 281 229 Chalcone, Chromone, 24-B1 P-Di 15 55 84 123 138 190 or Chromene-based compound. MS^3 404 ==> 361 349 335 294 293 282 281 269 253 Possible 5-O-Methyllupiwighteone P-Di 43 55 69 110 111 122 123 135 151

201 Table 4: Detected Compounds in Fraction B using HPLC Method 2. Indicates Compound Number (CMPD #), Molecular Weight (MW) in atomic mass units (amu), Retention Time (RT) in minutes (min), Mass Spec Base Peak Identification (BP-ID) showing negative mode ionization [M-H]- (indicates the type of ion which forms the Base Peak. Base Peak refers to the most abundant ion in the mass spectrum), Parent ion (P-) m/z after [M-H]- ionization (indicates the Base Peak in the normal mass spectra), Base Peak of the daughter ion (D-BP) (i.e., the most abundant ion in the daughter spectrum), MS/MS or MS/MS/MS product ions (MS^3), and the arithmetic difference in m/z between the parent ion and individual daughter ions (P-Di). Bold donates most abundant ion in the given spectra. CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Structure ID 240 3.84 [M-H]- 239 ==> 179 161 1-B2 P-Di 60 78 Dihydroxycinnamic Acid MS^3 179 ==> 161 159 141 125 117 103 87 (with acetate adduct) P-Di 18 20 38 54 62 76 92 404 17.08 [M-H]- 403 ==> 357 293 271 191 177 161 161 2',3,4,4'-Tetrahydroxychalcone (a.k.a. 2-B2 P-Di 46 110 132 212 226 242 242 Butein) plus rhamnose sugar, or MS^3 271 ==> 181 161 151 113 109 101 100 97 possibly Naringenin plus P-Di 90 110 120 158 162 170 171 174 rhamnose monosaccharide unit 240 18.98 [M-H]- 239 ==> 179 3-B2 P-Di 60 Dihydroxycinnamic Acid MS^3 179 ==> 161 (with acetate adduct) P-Di 18 316 20.11 [M-H]- 315 ==> 297 271 227 153 316 amu Flavonoid, such as 4-B2 P-Di 18 44 88 162 Isorhamnetin, Nepetin, Tamaraxetin, MS^3 271 ==> 227 or Rhamnetin P-Di 44 420 22.01 [M-H]- 419 ==> 359 289 Trihydroxy-trimethoxy-flavanoid, 5-B2 P-Di 60 130 such as Irigenin. Also may be a MS^3 359 ==> 341 331 315 273 247 231 217 209 153 Hydroxy-tetramethoxy-flavanoid P-Di 18 28 44 86 112 128 142 150 206 626 27.78 [M-H]- 625 ==> 463 6-B2 P-Di 162 Quercetin-diglucoside MS^3 463 ==> 301 191 176 P-Di 162 272 287 450 29.05 [M-H]- 449 ==> 287 269 259 Fustin (2,3-Dihydrofisetin), plus 7-B2 P-Di 162 180 190 glucose (or galactose) MS^3 287 ==> 259 180 monosaccharide side group P-Di 28 107 640 31.07 [M-H]- 639 ==> 477 315 8-B2 P-Di 162 324 Isorhamnetin-diglucoside MS^3 477 ==> 409 315 300 271 256 P-Di 68 162 177 206 221 152 32.77 [M-H]- 151 ==> 107 93 9-B2 P-Di 44 58 o-Homosalicylic Acid MS^3 107 ==> P-Di 480 33.65 [M-H]- 479 ==> 317 299 289 258 179 Either Myricetin, Quercetagetin, or 10-B2 P-Di 162 180 190 221 300 Gossypetin, plus a glucose (or MS^3 299 ==> 284 268 176 galactose) side-group. Possible P-Di 15 31 123 Gossypetin-8-glucoside (Gossypin) 334 33.90 [M-H]- 333 ==> 315 289 271 229 164 151 Unkown flavonoid: 11-B2 P-Di 18 44 62 104 169 182 Isorhamnetin-like fragmentation, but MS^3 315 ==> 297 271 245 229 205 177 165 149 with added 18 amu (water) P-Di 18 44 70 86 110 138 150 166 288 35.75 [M-H]- 287 ==> 259 243 201 159 127 12-B2 P-Di 28 44 86 128 160 Dihydrokaempferol MS^3 259 ==> 241 215 187 173 151 125 P-Di 18 44 72 86 108 134 332 39.35 [M-H]- 331 ==> 315 299 287 271 255 179 Chromanediol compound, such as: 13-B2 P-Di 16 32 44 60 76 152 2-(3,4-Demethoxyphenyl)-7-methoxy- MS^3 299 ==> 271 256 239 227 215 187 179 151 124 3,4-chromanediol, or 7,8-Dimethoxy-2- P-Di 28 43 60 72 84 112 120 148 175 (4-methoxyphenyl)-3,4-chromanediol 390 42.64 [M-H]- 389 ==> 374 341 323 303 195 193 165 Trihydroxy-tetramethoxy-flavone, such 14-B2 P-Di 15 48 66 86 194 196 224 as Gardenin E, or 3’,5,7-trihydroxy- MS^3 341 ==> 326 309 297 295 282 265 250 233 4’,5’,6,8-tetramethoxy-flavone (a.k.a. P-Di 15 32 44 46 59 76 91 108 Scaposin) 178 44.16 [M-H]- 177 ==> 151 149 135 133 123 109 89 65 15-B2 P-Di 26 28 42 44 54 68 88 112 3-Methoxycinnamic Acid MS^3 151 ==> P-Di

202 Table 4: (cont.) CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Structure ID 478 44.33 [M-H]- 477 ==> 315 299 271 247 16-B2 P-Di 162 178 206 230 Isorhamnetin-glucoside MS^3 315 ==> 300 299 286 269 231 P-Di 15 16 29 46 84 302 46.47 [M-H]- 301 ==> 283 271 257 255 17-B2 P-Di 18 30 44 46 Pentahydroxyflavone, such as Morin, MS^3 257 ==> 242 229 215 213 211 199 189 187 171 Robinetin, or Tricetin P-Di 15 28 42 44 46 58 68 70 86 312 47.03 [M-H]- 311 ==> 179 149 131 18-B2 P-Di 132 162 180 Caffeic acid, plus a terminal rhamnose MS^3 179 ==> 135 monosaccharide side-group P-Di 44 364 50.03 [M-H]- 363 ==> 331 299 283 272 227 195 163 Chromanediol derivative, similar 19-B2 P-Di 32 64 80 91 136 168 200 to Compound 13-B2, but with extra MS^3 331 ==> 299 195 180 32 amu (methoxy group) P-Di 32 136 151 372 54.08 [M-H]- 371 ==> 353 341 327 309 235 191 20-B2 P-Di 18 30 44 62 136 180 Syringin MS^3 353 ==> 338 207 P-Di 15 146 300 54.57 [M-H]- 299 ==> 299 300 amu MW flavonoid, such as 21-B2 P-Di 0 Chrysoeriol, Diosmetin, Kaempferide, MS^3 299 ==> 284 Geraldol, Pratensein, or Tectorigenin P-Di 15 478 56.47 [M-H]- 477 ==> 315 314 300 285 271 243 Isorhamnetin (or other 316 amu 22-B2 P-Di 162 163 177 192 206 234 flavonoid), plus glucose or galactose MS^3 315 ==> 300 286 285 271 257 243 151 side-group P-Di 15 29 30 44 58 72 164 304 57.15 [M-H]- 303 ==> 285 257 243 201 304 amu flavonoid, such as 23-B2 P-Di 18 46 60 102 3’-O-Methylcatechin, 4’-O-Methyl- MS^3 285 ==> 270 257 241 215 163 149 catechin, Taxifolin, or P-Di 15 28 44 70 122 136 Dihydrorobinetin 284 63.25 [M-H]- 283 ==> 268 255 217 190 24-B2 P-Di 15 28 66 93 Either Prunetin, Biochanin A, or MS^3 268 ==> 240 239 211 185 Acacetin P-Di 28 29 57 83 400 66.41 [M-H]- 399 ==> 384 323 308 Either Torosaflavone A, or Chromone 25-B2 P-Di 15 76 91 compound such as MS^3 384 ==> 370 366 356 341 326 309 308 298 297 2-[2-[3,5-dihydroxyphenyl]ethenyl]-5- P-Di 14 18 28 43 58 75 76 86 87 hydroxy-chromone 368 80.91 [M-H]- 367 ==> 352 349 337 297 282 270 26-B2 P-Di 15 18 30 70 85 97 Glycycoumarin MS^3 297 ==> 282 267 254 P-Di 15 30 43 352 86.62 [M-H]- 351 ==> 336 296 281 27-B2 P-Di 15 55 70 Pongagallone A MS^3 336 ==> 319 293 281 P-Di 17 43 55

203 Table 5: Detected Compounds in Fraction C using HPLC Method 1. Indicates Compound Number (CMPD #), Molecular Weight (MW) in atomic mass units (amu), Retention Time (RT) in minutes (min), Mass Spec Base Peak Identification (BP-ID) showing negative mode ionization [M-H]- (indicates the type of ion which forms the Base Peak. Base Peak refers to the most abundant ion in the mass spectrum), Parent ion (P-) m/z after [M-H]- ionization (indicates the Base Peak in the normal mass spectra), Base Peak of the daughter ion (D-BP) (i.e., the most abundant ion in the daughter spectrum), MS/MS or MS/MS/MS product ions (MS^3), and the arithmetic difference in m/z between the parent ion and individual daughter ions (P-Di). Bold donates most abundant ion in the given spectra. CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Structure ID 135 3.41 [M-H]- 134 ==> 107 92 1-C1 P-Di 27 42 UNKNOWN MS^3 107 ==> P-Di 350 8.05 [M-H]- 349 ==> 267 167 Either 2’,4’-dimethoxychalcone, 2-C1 P-Di 82 182 7-methoxyflavonol, 3-hyrdoxy-7- MS^3 267 ==> 151 150 149 134 126 methoxyflavone, or 3-hydroxy-6- P-Di 116 117 118 133 141 methoxyflavone 436 8.51 [M-H]- 435 ==> 417 399 375 357 345 333 315 289 Catechin or Epicatechin with 3-C1 P-Di 18 36 60 78 90 102 120 146 O-rhamnoside sugar side-group MS^3 333 ==> 315 289 271 245 229 223 205 (such as catechin-3-O, 5-O, or P-Di 18 44 62 88 104 110 128 7-O-rhamnoside) 420 12.31 [M-H]- 419 ==> 401 359 289 Either a Methoxy-flavonoid, a 4-C1 P-Di 18 60 130 Stilbene-glycoside, a Benzyl- MS^3 359 ==> 341 331 315 297 289 279 273 269 255 oxyflavone, or a (cat)epicatechin P-Di 18 28 44 62 70 80 86 90 104 with an unknown side-group 400 13.37 [M-H]- 399 ==> 339 177 Penolic acid (such as 4-Methoxy- 5-C1 P-Di 60 222 cinnamic Acid) or coumarin-linked MS^3 339 ==> 183 177 167 153 150 149 135 134 133 glycoside, such as Esculin, P-Di 156 162 172 186 189 190 204 205 206 Cichoriin, and Daphnin 438 14.42 [M-H]- 437 ==> 347 289 6-C1 P-Di 90 148 Catechin or Epicatechin with a trans- MS^3 347 ==> 329 289 245 225 205 177 167 161 139 Cinnamic Acid side-group P-Di 18 58 102 122 142 170 180 186 208 642 14.88 [M-H]- 641 ==> 479 317 Myricetin, Quercetagetin, or 7-C1 P-Di 162 324 Gossypetin, with a diglucoside MS^3 479 ==> 317 299 271 257 249 229 209 191 179 side-group P-Di 162 180 208 222 230 250 270 288 300 320 15.06 [M-H]- 319 ==> 287 197 165 8- C1 P-Di 32 122 154 Oureatacatechin MS^3 197 ==> 165 P-Di 32 438 15.17 [M-H]- 437 ==> 347 317 289 9-C1 P-Di 90 120 148 Catechin or Epicatechin with a trans- MS^3 347 ==> 329 289 245 225 207 205 167 151 125 Cinnamic Acid side-group P-Di 18 58 102 122 140 142 180 196 222 640 15.24 [M-H]- 639 ==> 477 315 316 amu MW flavonoid such as 10-C1 P-Di 162 324 Isorhamnetin, Tamarixetin, Nepetin, MS^3 477 ==> 409 315 300 299 271 247 229 213 175 Rhamnetin, or Pedalitin, plus P-Di 68 162 177 178 206 230 248 264 302 a diglucoside side-group 552 15.56 [M-H]- 551 ==> 389 227 11-C1 P-Di 162 324 Resveratrol, plus diglucoside MS^3 389 ==> 227 side-group P-Di 162 758 19.70 [M-H]- 757 ==> 595 Quercetin, plus trisaccharide side- 12-C1 P-Di 162 group (sambubiose or lathyrose, plus MS^3 595 ==> 301 300 271 255 243 215 300 271 255 additional glucose or galactose) P-Di 294 295 324 340 352 380 295 324 340 478 19.89 [M-H]- 477 ==> 315 300 316 amu MW flavonoid such as 13-C1 P-Di 162 177 Isorhamnetin, Tamarixetin, Nepetin, MS^3 315 ==> 301 300 271 256 247 230 151 Rhamnetin, or Pedalitin, plus P-Di 14 15 44 59 68 85 164 glucose or galactose side-group 178 20.03 [M-H]- 177 ==> 149 133 109 89 Methoxycinnamic Acid, Chroman- 14-C1 P-Di 28 44 68 88 carboxylic acid, or MS^3 133 ==> 123 105 104 91 89 Dihydroxy-coumarin P-Di 10 28 29 42 44 742 20.81 [M-H]- 741 ==> 609 591 475 325 301 300 271 255 Quercetin, with trisaccharide side- 15-C1 P-Di 132 150 266 416 440 441 470 486 group (xylose/arabinose plus MS^3 609 ==> 271 255 179 169 151 rhamnose plus glucose/galactose) P-Di 338 354 430 440 458

204 Table 5: (cont.) CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Structure ID 596 21.29 [M-H]- 595 ==> 475 445 301 300 271 255 16-C1 P-Di 120 150 294 295 324 340 Quercetin with disaccharide side- MS^3 307 ==> 271 255 243 227 203 179 169 group (most likely Peltatoside). P-Di 36 52 64 80 104 128 138 772 21.95 [M-H]- 771 ==> 609 463 301 Quercetin with trisaccharide side- 17-C1 P-Di 162 308 470 group (rutinose plus glucose or MS^3 609 ==> 301 300 271 255 galactose) P-Di 308 309 338 354 610 23.68 [M-H]- 609 ==> 343 301 300 273 271 255 243 229 18-C1 P-Di 266 308 309 336 338 354 366 380 Rutin (Quercetin-3-O-rutinoside) MS^3 301 ==> 271 257 255 179 151 121 P-Di 30 44 46 122 150 180 594 26.19 [M-H]- 593 ==> 285 255 Kaemperfol and/or Luteolin, plus 19-C1 P-Di 308 338 disaccharide side-group (either MS^3 285 ==> 285 267 257 255 241 239 229 223 213 rutinoside or neohesperidoside) P-Di 0 18 28 30 44 46 56 62 72 594 26.19 [M-H]- 593 ==> 447 327 285 257 229 Kaemperfol and/or Luteolin, plus 20-C1 P-Di 146 266 308 336 364 disaccharide side-group (either MS^3 285 ==> 285 267 257 255 241 239 229 213 rutinoside or neohesperidoside) P-Di 0 18 28 30 44 46 56 72 478 26.28 [M-H]- 477 ==> 357 315 314 299 285 271 257 243 316 amu MW flavonoid such as 21-C1 P-Di 120 162 163 178 192 206 220 234 Isorhamnetin, Tamarixetin, Nepetin, MS^3 314 ==> 300 286 285 271 257 243 Rhamnetin, or Pedalitin, plus P-Di 14 28 29 43 57 71 glucose or galactose side-group 624 26.79 [M-H]- 623 ==> 315 300 271 255 243 316 amu MW flavonoid such as 22-C1 P-Di 308 323 352 368 380 Isorhamnetin, Tamarixetin, Nepetin, MS^3 315 ==> 301 300 299 287 271 255 163 Rhamnetin, or Pedalitin, plus P-Di 14 15 16 28 44 60 152 rutinoside sugar side-group 408 28.46 [M-H]- 407 ==> 389 297 285 283 256 255 243 Either a Benzoic Acid-linked Luteolin 23- C1 P-Di 18 110 122 124 151 152 164 or Keampferol, or possibly MS^3 285 ==> 257 255 241 227 213 201 153 Remangiflavanone or P-Di 28 30 44 58 72 84 132 Sophoraflavanone 284 28.75 [M-H]- 283 ==> 269 268 Either Biochanin A, Prunetin, or 24-C1 P-Di 14 15 Acacetin. Also possibility this may MS^3 269 ==> 267 241 240 239 224 211 196 184 be other 284 amu MW flavonoid. P-Di 2 28 29 30 45 58 73 85 284 29.66 [M-H]- 283 ==> 269 Either Biochanin A, Prunetin, or 25-C1 P-Di 14 Acacetin. Also possibility this may MS^3 269 ==> 268 241 240 239 224 223 212 211 195 be other 284 amu MW flavonoid. P-Di 1 28 29 30 45 46 57 58 74 300 32.68 [M-H]- 299 ==> 284 Chrysoeriol, Diosmetin, or other 300 26-C1 P-Di 15 amu MW flavonoid (such as MS^3 284 ==> 256 228 211 200 193 165 150 137 Geraldol, Tectorigenin, Kaempferide, P-Di 28 56 73 84 91 119 134 147 Pratensein, Farresol, etc.) 330 34.15 [M-H]- 329 ==> 314 299 Likely a Trihydroxy-dimethoxy- 27-C1 P-Di 15 30 flavone (such as Isorhamnetin-3- MS^3 314 ==> 300 299 286 285 271 257 243 methoxy), or a Hydroxy-trimethoxy- P-Di 14 15 28 29 43 57 71 flavone 370 39.41 [M-H]- 369 ==> 339 295 251 245 233 219 177 28-C1 P-Di 30 74 118 124 136 150 192 Fraxetin-8-glucoside (a.k.a. Fraxin or MS^3 219 ==> 209 191 177 176 175 151 150 134 133 Fraxoside) P-Di 10 28 42 43 44 68 69 85 86 354 39.85 [M-H]- 353 ==> 335 309 297 284 253 240 175 Possible Glycyrrh-isoflavone, 29-C1 P-Di 18 44 56 69 100 113 178 Lico-isoflavanone, Lico-iso- MS^3 297 ==> 269 256 255 253 227 225 211 161 flavone A, or iso-Licoflavonol. P-Di 28 41 42 44 70 72 86 136 368 40.71 [M-H]- 367 ==> 352 349 337 322 30-C1 P-Di 15 18 30 45 Either Glycycoumarin or MS^3 337 ==> 322 296 Glycyrrh-isoflavanone P-Di 15 41 282 41.58 [M-H]- 281 ==> 266 261 253 233 Dimethoxyflavone (such as 31-C1 P-Di 15 20 28 48 2',3'-Dimethoxyflavone or MS^3 266 ==> 238 237 222 210 196 194 182 166 138 7,8-Dimethoxyflavone) P-Di 28 29 44 56 70 72 84 100 128 368 42.72 [M-H]- 367 ==> 352 312 309 297 32-C1 P-Di 15 55 58 70 Either Glycycoumarin or MS^3 352 ==> 338 324 309 297 281 271 253 161 Glycyrrh-isoflavanone P-Di 14 28 43 55 71 81 99 191

205 Table 5: (cont.) CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 368 43.39 [M-H]- 367 ==> 352 33-C1 P-Di 15 Either Glycycoumarin or MS^3 352 ==> 309 297 Glycyrrh-isoflavanone P-Di 43 55 352 47.29 [M-H]- 351 ==> 336 296 Either 5-O-Methyllupiwighteone, 34-C1 P-Di 15 55 Daphnoretin, Pongagallone A, or MS^3 336 ==> 293 281 Sophora-iso-flavone A P-Di 43 55 352 48.12 [M-H]- 351 ==> 336 296 281 Either 5-O-Methyllupiwighteone, 35-C1 P-Di 15 55 70 Daphnoretin, Pongagallone A, or MS^3 336 ==> 293 281 Sophora-iso-flavone A P-Di 43 55

206 Table 6: Detected Compounds in Fraction C using HPLC Method 2. Indicates Compound Number (CMPD #), Molecular Weight (MW) in atomic mass units (amu), Retention Time (RT) in minutes (min), Mass Spec Base Peak Identification (BP-ID) showing negative mode ionization [M-H]- (indicates the type of ion which forms the Base Peak. Base Peak refers to the most abundant ion in the mass spectrum), Parent ion (P-) m/z after [M-H]- ionization (indicates the Base Peak in the normal mass spectra), Base Peak of the daughter ion (D-BP) (i.e., the most abundant ion in the daughter spectrum), MS/MS or MS/MS/MS product ions (MS^3), and the arithmetic difference in m/z between the parent ion and individual daughter ions (P-Di). Bold donates most abundant ion in the given spectra. CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 328 21.18 [M-H]- 327 ==> 309 283 267 165 161 1-C2 P-Di 18 44 60 162 166 o-Hydrocoumaric Acid, plus glucose MS^3 165 ==> 141 137 121 69 (or galactose) sugar side-group P-Di 24 28 44 96 320 27.34 [M-H]- 319 ==> 301 287 275 269 197 165 125 2-C2 P-Di 18 32 44 50 122 154 194 Oureatacatechin MS^3 197 ==> 165 P-Di 32 642 29.22 [M-H]- 641 ==> 479 477 315 Isorhamnetin (or other flavonoid at 316 3-C2 P-Di 162 164 326 amu, such as Nepetin, Tamarixetin, or MS^3 479 ==> 317 315 300 299 257 256 214 Rhamnetin), plus an O-rutinoside P-Di 162 164 179 180 222 223 265 disaccharide side-group 320 30.29 [M-H]- 319 ==> 301 287 275 269 197 165 125 4-C2 P-Di 18 32 44 50 122 154 194 Oureatacatechin MS^3 197 ==> 165 P-Di 32 612 33.06 [M-H]- 611 ==> 551 449 389 227 Either Monotropein, or 5-C2 P-Di 60 162 222 384 Resveratrol-disaccharide (two glucose MS^3 551 ==> 449 389 or galactose units) P-Di 102 162 304 33.34 [M-H]- 303 ==> 285 259 244 219 217 204 Identical CMPD also detected 6-C2 P-Di 18 44 59 84 86 99 at RT 40.54. MS^3 259 ==> 246 244 228 217 175 Identified as 3’-O-Methylcatechin, and P-Di 13 15 31 42 84 4’-O-Methylcatechin 288 34.11 [M-H]- 287 ==> 259 243 227 205 197 165 Either 2,3-Dihydrofisetin (a.k.a. Fustin), 7-C2 P-Di 28 44 60 82 90 122 6-Hydroxyapigenin, or MS^3 259 ==> 227 215 203 199 173 Dihydrokaempferol P-Di 32 44 56 60 86 304 36.16 [M-H]- 303 ==> 287 275 266 261 Similar 304 amu compounds also 8-C2 P-Di found at RT’s 41.25, 42.86, & 66.73. MS^3 287 ==> 272 Compounds include Taxifolin, P-Di Dihydrobinetin, and Meciadanol. 320 37.80 [M-H]- 319 ==> 302 289 287 286 275 273 9-C2 P-Di 17 30 32 33 44 46 3,3',4',5,5',7-Hexahydroxyflavanone MS^3 275 ==> (a.k.a. Dihydromyricetin) P-Di 332 39.40 [M-H]- 331 ==> 299 271 255 179 Also at RT’s 39.40, 40.17, 40.72, 41.42, 10-C2 P-Di 32 60 76 152 41.99, 42.68, 43.19, 44.02, 44.53, 44.84, MS^3 299 ==> 271 257 54.32, and 54.91. Pentahydroxy-Methoxy- P-Di 28 42 flavone & Tri-O-Methylcatechin derivatives 938 39.65 [M-H]- 937 ==> 775 693 579 547 475 389 Amentoflavone-4',4'',7-trimethyl Ether 11-C2 P-Di 162 244 358 390 462 548 (a.k.a. Sciatopitysin), plus both a MS^3 775 ==> 727 638 579 389 371 341 327 glucose (or galactose) sugar and P-Di 48 137 196 386 404 434 448 unknown phenolic acid side-groups 480 40.97 [M-H]- 479 ==> 317 273 255 217 Hexahydroxy-flavone (such as 12-C2 P-Di 162 206 224 262 Myricetin, Quercetagetin, or MS^3 317 ==> 302 301 287 272 259 169 151 Gossypetin), plus a glucose (or P-Di 15 16 30 45 58 148 166 galactose) sugar side-group 304 42.35 [M-H]- 303 ==> 285 271 13-C2 P-Di 18 32 UNKNOWN BIOFLAVONOID MS^3 271 ==> 243 227 215 199 157 P-Di 28 44 56 72 114 904 42.42 [M-H]- 903 ==> 741 Rutin (a.k.a. Quercetin-3-O-rutinoside), 14-C2 P-Di 162 plus additional sambubiose (or MS^3 741 ==> 609 591 475 301 300 271 255 lathyrose) disscacharide side-group P-Di 132 150 266 440 441 470 486 758 43.77 [M-H]- 757 ==> 595 Quercetin, plus trisaccharide side- 15-C2 P-Di 162 group [consisting of sambubiose (or MS^3 595 ==> 301 300 271 243 227 203 lathyrose) disaccharide, plus an P-Di 294 295 324 352 368 392 added glucose (or galactose) sugar] 776 44.75 [M-H]- 775 ==> 727 640 606 579 389 372 341 Amentoflavone-4',4'',7-trimethyl Ether 16-C2 P-Di 48 135 169 196 386 403 434 (a.k.a. Sciatopitysin), plus an unknown MS^3 727 ==> 385 phenolic acid side-group P-Di 342

207 Table 6: (cont.) CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 318 45.85 [M-H]- 317 ==> 299 289 271 231 193 181 Compounds also detected at 17-C2 P-Di 18 28 46 86 124 136 RT’s 43.70 & 47.51. MS^3 299 ==> 284 271 231 215 Identified as Myricetin, P-Di 15 28 68 84 Gossypetin, and Quercetagetin 348 47.34 [M-H]- 347 ==> 332 319 316 315 285 271 231 219 193 UNKNOWN flavonoid, possibly a 18-C2 P-Di 15 28 31 32 62 76 116 128 154 newly discovered tetrahydroxy- MS^3 315 ==> 299 285 271 229 165 dimethoxy-flavonoid P-Di 16 30 44 86 150 772 47.92 [M-H]- 771 ==> 609 463 Quercetin, plus trisaccharide side- 19-C2 P-Di 162 308 group [consisting of a rutinose MS^3 609 ==> 301 300 271 213 disaccharide, plus an P-Di 308 309 338 396 added glucose (or galactose) sugar] 330 49.77 [M-H]- 329 ==> 314 312 301 296 281 257 255 206 193 20-C2 P-Di 15 17 28 33 48 72 74 123 136 Irisflavone D MS^3 193 ==> 178 177 165 164 150 P-Di 15 16 28 29 43 344 52.34 [M-H]- 343 ==> 329 327 315 313 297 282 271 229 227 21-C2 P-Di 14 16 28 30 46 61 72 114 116 Dihydroxy-trimethoxy-flavone MS^3 315 ==> 297 273 271 247 229 227 206 177 165 (such as Irisflavone B) P-Di 18 42 44 68 86 88 109 138 150 356 54.72 [M-H]- 355 ==> 340 339 337 327 323 313 312 298 292 Cudraflavanone B, or 2-Carbethoxy- 22-C2 P-Di 15 16 18 28 32 42 43 57 63 5,7-dihydroxy-4'-methoxyisoflavone, or MS^3 219 ==> 204 191 190 177 176 5-methoxy-4’,7-dihydroxy-flavanone-8 P-Di 15 28 29 42 43 -(3-methylbutyl), or derivative of these. 300 59.12 [M-H]- 299 ==> 284 Chrysoeriol, Diosmetin, or other 300 23-C2 P-Di 15 amu MW flavonoid (such as MS^3 284 ==> 256 228 211 200 193 165 150 137 Geraldol, Tectorigenin, Kaempferide, P-Di 28 56 73 84 91 119 134 147 Pratensein, Farresol, etc.) 330 62.07 [M-H]- 329 ==> 314 299 Trihydroxy-dimethoxy-flavone 24-C2 P-Di 15 30 (such as Isorhamnetin-3-methoxy), MS^3 314 ==> 300 299 286 285 271 257 243 or a Hydroxy-trimethoxy-flavone P-Di 14 15 28 29 43 57 71 300 62.67 [M-H]- 299 ==> 284 Chrysoeriol, Diosmetin, or other 300 25-C2 P-Di 15 amu MW flavonoid (such as MS^3 284 ==> 256 228 211 200 193 165 150 137 Geraldol, Tectorigenin, Kaempferide, P-Di 28 56 73 84 91 119 134 147 Pratensein, Farresol, etc.) 286 64.81 [M-H]- 285 ==> 267 257 243 241 217 199 175 151 133 26-C2 P-Di 18 28 42 44 68 86 110 134 152 3',4',5,7-Tetrahydroxyflavone (a.k.a. MS^3 241 ==> Luteolin) P-Di 300 68.47 [M-H]- 299 ==> 284 Chrysoeriol, Diosmetin, or other 300 27-C2 P-Di 15 amu MW flavonoid (such as MS^3 284 ==> 256 228 211 200 193 165 150 137 Geraldol, Tectorigenin, Kaempferide, P-Di 28 56 73 84 91 119 134 147 Pratensein, Farresol, etc.) 330 69.72 [M-H]- 329 ==> 314 299 Trihydroxy-dimethoxy-flavone 28-C2 P-Di 15 30 (such as Isorhamnetin-3-methoxy), MS^3 314 ==> 300 299 286 285 271 257 243 or a Hydroxy-trimethoxy-flavone P-Di 14 15 28 29 43 57 71 300 70.39 [M-H]- 299 ==> 284 Chrysoeriol, Diosmetin, or other 300 29-C2 P-Di 15 amu MW flavonoid (such as MS^3 284 ==> 256 228 211 200 193 165 150 137 Geraldol, Tectorigenin, Kaempferide, P-Di 28 56 73 84 91 119 134 147 Pratensein, Farresol, etc.) 330 71.54 [M-H]- 329 ==> 314 299 Trihydroxy-dimethoxy-flavone 30-C2 P-Di 15 30 (such as Isorhamnetin-3-methoxy), MS^3 314 ==> 300 299 286 285 271 257 243 or a Hydroxy-trimethoxy-flavone P-Di 14 15 28 29 43 57 71 356 74.49 [M-H]- 355 ==> 337 324 310 295 293 251 237 219 193 Cudraflavanone B, or 2-Carbethoxy- 31-C2 P-Di 18 31 45 60 62 104 118 136 162 5,7-dihydroxy-4'-methoxyisoflavone, or MS^3 219 ==> 175 151 149 133 107 5-methoxy-4’,7-dihydroxy-flavanone-8 P-Di 44 68 70 86 112 -(3-methylbutyl), or derivative of these. 888 76.69 [M-H]- 887 ==> 490 463 317 285 3',4',5,7-Tetrahydroxyflavone (a.k.a. 32-C2 P-Di 397 424 570 602 Luteolin), with rutinose and lathyrose MS^3 285 ==> 267 257 243 241 217 213 201 164 (or sambubiose) side-groups P-Di 18 28 42 44 68 72 84 121 530 78.93 [M-H]- 529 ==> 367 352 329 219 33-C2 P-Di 162 177 200 310 Glycycoumarin (or a derivative of) MS^3 367 ==> 352 336 324 309 297 P-Di 15 31 43 58 70 340 79.92 [M-H]- 339 ==> 251 245 233 219 175 Glycosidically-bound Dihydroxy- 34-C2 P-Di 88 94 106 120 164 Coumarin, such as MS^3 219 ==> 197 175 151 133 6,7-Dihydroxycoumarin-7-glucoside P-Di 22 44 68 86 (or any possible derivative of)

208 Table 7: Detected Compounds in Fraction D using HPLC Method 1. Indicates Compound Number (CMPD #), Molecular Weight (MW) in atomic mass units (amu), Retention Time (RT) in minutes (min), Mass Spec Base Peak Identification (BP-ID) showing negative mode ionization [M-H]- (indicates the type of ion which forms the Base Peak. Base Peak refers to the most abundant ion in the mass spectrum), Parent ion (P-) m/z after [M-H]- ionization (indicates the Base Peak in the normal mass spectra), Base Peak of the daughter ion (D-BP) (i.e., the most abundant ion in the daughter spectrum), MS/MS or MS/MS/MS product ions (MS^3), and the arithmetic difference in m/z between the parent ion and individual daughter ions (P-Di). Bold donates most abundant ion in the given spectra. CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 327 7.86 [M-H]- 326 ==> 266 134 Either a Hydroxy-Methoxy-Flavone, 1-D1 P-Di 60 192 a Methoxy-Flavonol, or a MS^3 266 ==> 134 107 Dimethoxy-Chalcone, with acetate P-Di 132 159 adduct ion. 290 13.14 [M-H]- 289 ==> 245 205 179 165 151 137 123 109 2-D1 P-Di 44 84 110 124 138 152 166 180 Catechin MS^3 245 ==> 227 203 187 161 135 P-Di 18 42 58 84 110 290 16.44 [M-H]- 289 ==> 245 205 179 165 161 151 137 125 109 3-D1 P-Di 44 84 110 124 128 138 152 164 180 Epicatechin MS^3 245 ==> 227 203 187 161 135 123 109 P-Di 18 42 58 84 110 122 136 304 19.09 [M-H]- 303 ==> 285 177 125 3’-O-Methylcatechin, or 4-D1 P-Di 18 126 178 4’-O-Methylcatechin MS^3 285 ==> 257 243 241 217 201 199 177 175 161 (or their epicatechin derivatives) P-Di 28 42 44 68 84 86 108 110 124 478 20.72 [M-H]- 477 ==> 315 314 300 299 285 271 247 175 5-D1 P-Di 162 163 177 178 192 206 230 302 Isorhamnetin-glucoside derivative MS^3 315 ==> 301 300 299 285 269 245 227 205 (such as Isorhamnetin-3-glucoside) P-Di 14 15 16 30 46 70 88 110 364 21.14 [M-H]- 363 ==> 345 313 299 281 255 169 A Tetra-O-methylcatechin or 6-D1 P-Di 18 50 64 82 108 194 Tetra-O-methylepicatechin derivative MS^3 345 ==> 330 315 314 313 299 286 285 284 271 with additional 18 amu (water) P-Di 15 30 31 32 46 59 60 61 74 318 21.84 [M-H]- 317 ==> 299 289 273 271 255 247 245 233 231 Catechin (or Epicatechin) with two 7-D1 P-Di 18 28 44 46 62 70 72 84 86 added methyl groups (possibly MS^3 289 ==> 229 227 225 217 205 203 201 189 187 located either on a flavonoid ring, or P-Di 60 62 64 72 84 86 88 100 102 existing as methoxy groups) 318 22.41 [M-H]- 317 ==> 289 273 231 229 125 Catechin (or Epicatechin) with two 8-D1 P-Di 28 44 86 88 192 added methyl groups (possibly MS^3 289 ==> 275 274 271 257 245 230 203 185 183 located either on a flavonoid ring, or P-Di 14 15 18 32 44 59 86 104 106 existing as methoxy groups) 448 22.84 [M-H]- 447 ==> 285 9-D1 P-Di 162 Luteolin-glucoside (such as MS^3 285 ==> 267 257 243 241 229 223 217 213 201 Luteolin-7-O-glucoside) P-Di 18 28 42 44 56 62 68 72 84 520 22.99 [M-H]- 519 ==> 461 299 Either 3',5,7-Trihydroxy-4'-methoxy-iso 10-D1 P-Di 58 220 -flavone-3'-O-beta-glucopyranoside, or MS^3 461 ==> 299 8-Glycosyl-4',5,7-trihydroxy-3'-methoxy P-Di 162 -flavone (Scoparoside), w/ acetate adduct 464 23.73 [M-H]- 463 ==> 301 300 11-D1 P-Di 162 163 Quercetin-based glycoside, such as MS^3 307 ==> 271 257 255 239 229 193 183 179 167 Isoquercetrin or Hyperoside P-Di 36 50 52 68 78 114 124 128 140 464 24.42 [M-H]- 463 ==> 301 Quercetin-glycoside, such as 12-D1 P-Di 162 Isoquercitrin, Hyperoside, MS^3 301 ==> 287 286 271 256 240 225 203 183 153 Gossypitrin, or Spiraeoside P-Di 14 15 30 45 61 76 98 118 148 346 25.26 [M-H]- 345 ==> 330 328 315 314 313 300 299 285 281 Either 3,4',5,7-Tetrahydroxy-3',5'- 13-D1 P-Di 15 17 30 31 32 45 46 60 64 dimethoxyflavone (a.k.a. ), or MS^3 313 ==> 300 299 286 285 282 271 269 253 239 Tetra-O-methylcatechin (or Tetra-O- P-Di 13 14 27 28 31 42 44 60 74 methylepicatechin) 448 25.56 [M-H]- 447 ==> 285 14-D1 P-Di 162 Keampferol-glucoside MS^3 285 ==> 267 257 243 241 239 229 217 213 211 P-Di 18 28 42 44 46 56 68 72 74 478 25.82 [M-H]- 477 ==> 315 300 Isorhamnetin (or other 316 amu 15-D1 P-Di 162 177 Flavonoid, such as Tamarixetin, MS^3 315 ==> 301 300 287 272 271 255 243 215 191 Nepetin, or Rhamnetin) with a glucose P-Di 14 15 28 43 44 60 72 100 124 or galactose side-group

209 Table 7: (cont.) CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 316 26.38 [M-H]- 315 ==> 301 300 299 286 285 272 271 247 243 16-D1 P-Di 14 15 16 29 30 43 44 68 72 3-Methylquercetin (a.k.a. MS^3 300 ==> 283 282 272 271 256 244 243 232 230 Isorhamnetin) P-Di 17 18 28 29 44 56 57 68 70 492 26.64 [M-H]- 491 ==> 477 476 329 314 313 299 17-D1 P-Di 14 15 162 177 178 192 Isorhamnetin-3-methoxy-glucoside (or MS^3 476 ==> 314 313 299 298 285 281 270 242 227 other m/z 329 M-H- aglycone) P-Di 162 163 177 178 191 195 206 234 249 572 26.91 [M-H]- 571 ==> 419 285 18-D1 P-Di 152 286 Biflavone (such as Hegoflavone) MS^3 419 ==> 257 P-Di 162 272 27.03 [M-H]- 271 ==> 253 227 19-D1 P-Di 18 44 Resveratrol plus C3H7 or COOH MS^3 253 ==> 185 183 159 157 143 P-Di 68 70 94 96 110 462 27.42 [M-H]- 461 ==> 301 300 299 284 20-D1 P-Di 160 161 162 177 Chrysoeriol-glucoside (or other m/z MS^3 299 ==> 285 284 179 151 299 M-H- aglycone) P-Di 14 15 120 148 448 27.80 [M-H]- 447 ==> 285 21-D1 P-Di 162 Luteolin-glucoside MS^3 285 ==> 270 267 257 243 241 217 201 199 175 P-Di 15 18 28 42 44 68 84 86 110 478 28.39 [M-H]- 477 ==> 315 Isorhamnetin (or other 316 amu 22-D1 P-Di 162 Flavonoid, such as Tamarixetin, MS^3 315 ==> 300 Nepetin, or Rhamnetin) with a glucose P-Di 15 or galactose side-group 302 28.66 [M-H]- 301 ==> 287 286 271 256 151 Either 3,3',4',5',7-Pentahydroxyflavone 23-D1 P-Di 14 15 30 45 150 (a.k.a. Robinetin), or MS^3 286 ==> 271 241 183 169 164 151 150 136 107 3',4',5,5',7-Pentahydroxyflavone P-Di 15 45 103 117 122 135 136 150 179 (a.k.a. Tricetin) 580 29.12 [M-H]- 579 ==> 339 24-D1 P-Di 240 Amentoflavone-4',4'',7-trimethyl Ether MS^3 339 ==> 324 311 297 295 (a.k.a. Sciatopitysin) P-Di 15 28 42 44 302 29.23 [M-H]- 301 ==> 286 258 257 251 242 217 183 177 165 25-D1 P-Di 15 43 44 50 59 84 118 124 136 4',5,7-Trihydroxy-3'-methoxyflavanone MS^3 286 ==> 183 169 135 134 107 83 (a.k.a. Homoeriodictyol) P-Di 103 117 151 152 179 203 286 286 286 330 29.67 [M-H]- 329 ==> 314 299 26-D1 P-Di 15 30 Isorhamnetin-3-methoxy MS^3 314 ==> 300 299 271 255 242 230 227 217 215 P-Di 14 15 43 59 72 84 87 97 99 286 30.43 [M-H]- 285 ==> 267 257 243 241 239 217 213 201 199 27-D1 P-Di 18 28 42 44 46 68 72 84 86 Luteolin (a.k.a. 3',4',5,7- MS^3 267 ==> 213 203 201 199 198 197 185 177 173 Tetrahydroxyflavone) P-Di 54 64 66 68 69 70 82 90 94 300 30.94 [M-H]- 299 ==> 284 300 amu flavonoid, such as Diosmetin, 28-D1 P-Di 15 299 299 299 Chrysoeriol, Pratensein, Farresol, MS^3 284 ==> 267 256 255 239 227 211 200 199 183 Tectorigenin, Geraldol, P-Di 17 28 29 45 57 73 84 85 101 and Kaempferide 300 31.75 [M-H]- 299 ==> 284 300 amu flavonoid, such as Diosmetin, 29-D1 P-Di 15 Chrysoeriol, Pratensein, Farresol, MS^3 284 ==> 267 256 255 239 227 211 200 195 188 Tectorigenin, Geraldol, P-Di 17 28 29 45 57 73 84 89 96 and Kaempferide 316 33.30 [M-H]- 315 ==> 300 30-D1 P-Di 15 316 amu flavonoid, such as Nepetin, MS^3 300 ==> 283 271 255 243 227 216 211 199 183 Tamarixetin, and Rhamnetin P-Di 17 29 45 57 73 84 89 101 117 300 33.54 [M-H]- 299 ==> 284 300 amu flavonoid, such as Diosmetin, 31-D1 P-Di 15 Chrysoeriol, Pratensein, Farresol, MS^3 284 ==> 256 Tectorigenin, Geraldol, P-Di 28 and Kaempferide 312 34.01 [M-H]- 311 ==> 293 267 255 241 201 32-D1 P-Di 18 44 56 70 110 Maxima-IsoFlavone-E MS^3 241 ==> 223 213 199 173 155 145 P-Di 18 28 42 68 86 96

210 Table 7: (cont.) CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 286 34.57 [M-H]- 285 ==> 270 286 amu flavonoid: either 33-D1 P-Di 15 Isosakuranetin, Licochalcone B, MS^3 270 ==> 255 183 169 151 Homobutein, Sakuranetin, P-Di 15 87 101 119 or Sappanchalcone 314 34.71 [M-H]- 313 ==> 298 283 Either Kaempferol-3,7-dimethoxy, or 34-D1 P-Di 15 30 5,7-dihydroxy-4',6-dimethoxyflavone MS^3 298 ==> 283 269 267 (a.k.a. Pectolinarigenin) P-Di 15 29 31 448 36.78 [M-H]- 447 ==> 267 35-D1 P-Di 180 Hydroxy-methoxyflavone, plus MS^3 267 ==> 239 227 225 glucose (or galactose) P-Di 28 40 42 314 36.93 [M-H]- 313 ==> 298 Either Kaempferol-3,7-dimethoxy, or 36-D1 P-Di 15 5,7-dihydroxy-4',6-dimethoxyflavone MS^3 298 ==> 283 255 (a.k.a. Pectolinarigenin) P-Di 15 43 Either Kaempferol-3,7-dimethoxy, or 314 37.81 [M-H]- 313 ==> 298 37-D1 P-Di 15 Kaempferol-3,7-dimethoxy MS^3 298 ==> 281 270 269 243 242 226 198 P-Di 17 28 29 55 56 72 100 284 38.07 [M-H]- 283 ==> 268 Biochanin A, Prunetin, Acacetin, or 38-D1 P-Di 15 other 284 amu flavonoid (such as MS^3 268 ==> 267 240 239 226 224 223 212 211 195 Glycitein, Negletein, Oroxylin, Vogonin, P-Di 1 28 29 42 44 45 56 57 73 or Genkwanin)

211 Table 8: Detected Compounds in Fraction D using HPLC Method 2. Indicates Compound Number (CMPD #), Molecular Weight (MW) in atomic mass units (amu), Retention Time (RT) in minutes (min), Mass Spec Base Peak Identification (BP-ID) showing negative mode ionization [M-H]- (indicates the type of ion which forms the Base Peak. Base Peak refers to the most abundant ion in the mass spectrum), Parent ion (P-) m/z after [M-H]- ionization (indicates the Base Peak in the normal mass spectra), Base Peak of the daughter ion (D-BP) (i.e., the most abundant ion in the daughter spectrum), MS/MS or MS/MS/MS product ions (MS^3), and the arithmetic difference in m/z between the parent ion and individual daughter ions (P-Di). Bold donates most abundant ion in the given spectra. CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 390 43.10 [M-H]- 389 ==> 227 1-D2 P-Di 162 Resveratrol, plus a glucose MS^3 227 ==> 185 159 157 143 (or galactose) side-group P-Di 42 68 70 84 302 57.43 [M-H]- 301 ==> 284 283 257 165 2-D2 P-Di 17 18 44 136 2',3,4',5,7-Pentahydroxyflavone MS^3 283 ==> 268 255 252 240 (a.k.a. Morin) P-Di 15 28 31 43 302 60.03 [M-H]- 301 ==> 286 271 257 241 164 151 Either 3,3',4',5',7-Pentahydroxyflavone 3-D2 P-Di 15 30 44 60 137 150 (a.k.a. Robinetin), or MS^3 283 ==> 269 228 180 164 151 3',4',5,5',7-Pentahydroxyflavone P-Di 14 55 103 119 132 (a.k.a. Tricetin) 302 60.69 [M-H]- 301 ==> 286 257 242 217 177 165 151 Either 3,3',4',5',7-Pentahydroxyflavone 4-D2 P-Di 15 44 59 84 124 136 150 (a.k.a. Robinetin), or MS^3 283 ==> 267 259 242 212 176 151 3',4',5,5',7-Pentahydroxyflavone P-Di 16 24 41 71 107 132 (a.k.a. Tricetin) 302 61.55 [M-H]- 301 ==> 286 284 257 241 217 177 165 151 5-D2 P-Di 15 17 44 60 84 124 136 150 3',5,7-Trihydroxy-4'-methoxyflavanone MS^3 283 ==> 267 257 256 240 215 201 183 164 150 (a.k.a. Hesperitin) P-Di 16 26 27 43 68 82 100 119 133 492 72.98 [M-H]- 491 ==> 287 Either 2,3-Dihydrofisetin (a.k.a. Fustin), 6-D2 P-Di 204 or 3,3’,4’,7-tetrahydroxyflavone (a.k.a. MS^3 287 ==> 269 245 243 219 165 161 125 Fisetin), with a sinapoyl acyl P-Di 18 42 44 68 122 126 162 side-group. 492 77.72 [M-H]- 491 ==> 287 Either 2,3-Dihydrofisetin (a.k.a. Fustin), 7-D2 P-Di 204 or 3,3’,4’,7-tetrahydroxyflavone (a.k.a. MS^3 287 ==> 269 259 245 186 165 161 125 Fisetin), with a sinapoyl acyl P-Di 18 28 42 101 122 126 162 side-group.

212 Table 9: Detected Compounds in Fraction E using HPLC Method 1. Indicates Compound Number (CMPD #), Molecular Weight (MW) in atomic mass units (amu), Retention Time (RT) in minutes (min), Mass Spec Base Peak Identification (BP-ID) showing negative mode ionization [M-H]- (indicates the type of ion which forms the Base Peak. Base Peak refers to the most abundant ion in the mass spectrum), Parent ion (P-) m/z after [M-H]- ionization (indicates the Base Peak in the normal mass spectra), Base Peak of the daughter ion (D-BP) (i.e., the most abundant ion in the daughter spectrum), MS/MS or MS/MS/MS product ions (MS^3), and the arithmetic difference in m/z between the parent ion and individual daughter ions (P-Di). Bold donates most abundant ion in the given spectra. CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 350 18.72 [M-H]- 349 ==> 331 299 Newly discovered chromanediol: 1-E1 P-Di 18 50 2-(3,4-dimethoxyphenyl)-3,4-dihydro- MS^3 331 ==> 300 299 287 271 179 151 2H-chromene-3,4,5,6,7-pentol P-Di 31 32 44 60 152 180 576 20.80 [M-H]- 575 ==> 557 449 437 408 394 287 2-E1 P-Di 18 126 138 167 181 288 A-type Proanthocyanidin dimer MS^3 557 ==> 431 287 243 P-Di 126 270 314 464 21.53 [M-H]- 463 ==> 301 3-E1 P-Di 162 Either Isoquercitrin, Hyperoside, MS^3 301 ==> 183 179 169 151 Gossypitrin, or Spiraeoside P-Di 118 122 132 150 364 21.83 [M-H]- 363 ==> 331 Newly discovered chromanediol: 4-E1 P-Di 32 2-(3,4-dimethoxyphenyl)-7-methoxy- MS^3 331 ==> 300 299 287 271 179 151 3,4-dihydro-2H-chromene-3,4,5,6-tetrol P-Di 31 32 44 60 152 180 464 25.31 [M-H]- 463 ==> 301 5-E1 P-Di 162 Either Isoquercitrin, Hyperoside, MS^3 301 ==> 179 151 Gossypitrin, or Spiraeoside P-Di 122 150 348 25.90 [M-H]- 347 ==> 287 Either Dihydrokeampferol, or 6-E1 P-Di 60 3,3’,4’,7-tetrahydroxy-flavanone (a.k.a. MS^3 287 ==> 269 243 225 183 169 151 135 125 2,3-Dihydrofisetin, or Fustin) P-Di 18 44 62 104 118 136 152 162 286 27.10 [M-H]- 285 ==> 257 229 213 177 Either Fisetin, Datiscetin, Scutellarin, 7-E1 P-Di 28 56 72 108 Baptigenin, Sakuranetin, or a MS^3 257 ==> 229 213 187 172 151 Tetrahydroxy-isoflavone P-Di 28 44 70 85 106 302 29.32 [M-H]- 301 ==> 179 151 8-E1 P-Di 122 150 3,3',4',5,7-Pentahydroxyflavone MS^3 179 ==> 183 169 151 (a.k.a. Quercetin) P-Di -4 10 28 464 39.31 [M-H]- 463 ==> 283 Either a Dihydroxy-methoxy-flavone, a 9-E1 P-Di 180 Dihydroxy-methoxy-isoflavone, or a MS^3 283 ==> 255 243 217 Dimethoxy-flavanone, with a P-Di 28 40 66 glucose (or galactose) side-group

213 Table 10: Detected Compounds in Fraction E using HPLC Method 2. Indicates Compound Number (CMPD #), Molecular Weight (MW) in atomic mass units (amu), Retention Time (RT) in minutes (min), Mass Spec Base Peak Identification (BP-ID) showing negative mode ionization [M-H]- (indicates the type of ion which forms the Base Peak. Base Peak refers to the most abundant ion in the mass spectrum), Parent ion (P-) m/z after [M-H]- ionization (indicates the Base Peak in the normal mass spectra), Base Peak of the daughter ion (D-BP) (i.e., the most abundant ion in the daughter spectrum), MS/MS or MS/MS/MS product ions (MS^3), and the arithmetic difference in m/z between the parent ion and individual daughter ions (P-Di). Bold donates most abundant ion in the given spectra. CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 332 39.95 [M-H]- 331 ==> 301 300 299 287 271 255 179 151 Possible new Dihydoxy-dimethoxy- 1-E2 P-Di 30 31 32 44 60 76 152 180 -flavanone: 3,5-dihydroxy-2-(4- MS^3 299 ==> 272 271 256 255 227 179 155 hydroxy-3- methoxyphenyl)-7-methoxy P-Di 27 28 43 44 72 120 144 -2,3-dihydro-4H-chromen-4-one 288 54.00 [M-H]- 287 ==> 269 199 183 161 151 135 107 2-E2 P-Di 18 88 104 126 136 152 180 3',4',5,7-Tetrahydroxyflavanone MS^3 151 ==> 108 107 65 (a.k.a. Eriodictyol) P-Di 43 44 86 286 61.28 [M-H]- 285 ==> 257 229 213 163 Either Fisetin, Datiscetin, Scutellarin, 3-E2 P-Di 28 56 72 122 Baptigenin, Sakuranetin, or a MS^3 257 ==> 239 229 227 171 157 Tetrahydroxy-isoflavone P-Di 18 28 30 86 100

214 Table 11: Detected Compounds in Fraction F using HPLC Method 1. Indicates Compound Number (CMPD #), Molecular Weight (MW) in atomic mass units (amu), Retention Time (RT) in minutes (min), Mass Spec Base Peak Identification (BP-ID) showing negative mode ionization [M-H]- (indicates the type of ion which forms the Base Peak. Base Peak refers to the most abundant ion in the mass spectrum), Parent ion (P-) m/z after [M-H]- ionization (indicates the Base Peak in the normal mass spectra), Base Peak of the daughter ion (D-BP) (i.e., the most abundant ion in the daughter spectrum), MS/MS or MS/MS/MS product ions (MS^3), and the arithmetic difference in m/z between the parent ion and individual daughter ions (P-Di). Bold donates most abundant ion in the given spectra. CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 576 6.92 [M-H]- 575 ==> 557 531 513 487 449 423 407 325 1-F1 P-Di 18 44 62 88 126 152 168 250 A-type Proanthocyanidin dimer MS^3 423 ==> 405 387 379 361 355 338 309 298 285 P-Di 18 36 44 62 68 85 114 125 138 578 10.11 [M-H]- 577 ==> 451 425 407 289 2-F1 P-Di 126 152 170 288 B-type Procyanidin dimer MS^3 425 ==> 407 381 273 P-Di 18 44 152 578 10.84 [M-H]- 577 ==> 451 425 407 289 3-F1 P-Di 126 152 170 288 B-type Procyanidin dimer MS^3 425 ==> 407 381 273 P-Di 18 44 152 578 11.43 [M-H]- 577 ==> 559 451 425 407 299 289 246 217 4-F1 P-Di 18 126 152 170 278 288 331 360 Procyanidin B3 Dimer MS^3 425 ==> 407 381 357 339 297 273 162 P-Di 18 44 68 86 128 152 263 578 11.97 [M-H]- 577 ==> 559 451 425 423 407 379 289 5-F1 P-Di 18 126 152 154 170 198 288 B-type Procyanidin dimer MS^3 425 ==> 407 381 361 297 273 P-Di 18 44 64 128 152 592 13.90 [M-H]- 591 ==> 465 439 421 303 Flavonoid dimer consisting of a 6-F1 P-Di 126 152 170 288 catechin (or epicatechin) monomer, MS^3 439 ==> 421 395 287 271 259 243 bound to an O-methylated catechin P-Di 18 44 152 168 180 196 (or epicatechin) monomer 574 14.73 [M-H]- 573 ==> 555 529 447 421 323 285 283 7-F1 P-Di 18 44 126 152 250 288 290 Procyanidin dimer MS^3 447 ==> 429 403 337 325 285 281 271 246 217 P-Di 18 44 110 122 162 166 176 201 230 592 15.73 [M-H]- 591 ==> 465 455 439 421 303 Flavonoid dimer consisting of a 8-F1 P-Di 126 136 152 170 288 catechin (or epicatechin) monomer, MS^3 439 ==> 421 411 287 271 243 bound to an O-methylated catechin P-Di 18 28 152 168 196 (or epicatechin) monomer 574 15.87 [M-H]- 573 ==> 555 529 511 447 421 285 283 9-F1 P-Di 18 44 62 126 152 288 290 Procyanidin dimer MS^3 447 ==> 429 403 337 325 285 281 243 P-Di 18 44 110 122 162 166 204 592 16.14 [M-H]- 591 ==> 571 465 439 419 339 325 303 285 Flavonoid dimer consisting of a 10-F1 P-Di 20 126 152 172 252 266 288 306 catechin (or epicatechin) monomer, MS^3 439 ==> 421 419 301 287 285 283 271 243 bound to an O-methylated catechin P-Di 18 20 138 152 154 156 168 196 (or epicatechin) monomer 574 16.80 [M-H]- 573 ==> 555 529 511 447 421 283 11-F1 P-Di 18 44 62 126 152 290 Procyanidin dimer MS^3 447 ==> 429 403 337 325 285 281 243 217 P-Di 18 44 110 122 162 166 204 230 578 19.45 [M-H]- 577 ==> 451 425 407 289 12-F1 P-Di 126 152 170 288 B-type Procyanidin dimer MS^3 425 ==> 407 273 P-Di 18 152 578 20.06 [M-H]- 577 ==> 451 425 407 289 13-F1 P-Di 126 152 170 288 B-type Procyanidin dimer MS^3 425 ==> 407 273 P-Di 18 152 574 20.21 [M-H]- 573 ==> 555 529 511 487 471 447 421 283 14-F1 P-Di 18 44 62 86 102 126 152 290 Procyanidin dimer MS^3 447 ==> 429 403 385 361 343 337 325 285 279 P-Di 18 44 62 86 104 110 122 162 168 576 23.03 [M-H]- 575 ==> 449 394 287 243 15-F1 P-Di 126 181 288 332 A-type Proanthocyanidin dimer MS^3 394 ==> 271 229 P-Di 123 165

215 Table 11: (cont.) CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 576 23.60 [M-H]- 575 ==> 557 449 437 431 407 394 351 309 287 16-F1 P-Di 18 126 138 144 168 181 224 266 288 A-type Proanthocyanidin dimer MS^3 449 ==> 431 325 287 283 P-Di 18 124 162 166 424 25.82 [M-H]- 423 ==> 405 355 341 313 301 299 285 271 259 Either Remangiflavanone B, 17-F1 P-Di 18 68 82 110 122 124 138 152 164 Euchrestaflavanone B, or MS^3 299 ==> 283 271 243 Sophoraflavanone G P-Di 16 28 56 588 27.32 [M-H]- 587 ==> 451 435 421 391 377 301 299 270 18-F1 P-Di 136 152 166 196 210 286 288 317 Kolaflavanone MS^3 451 ==> 423 407 405 383 341 325 299 287 257 P-Di 28 44 46 68 110 126 152 164 194 448 31.86 [M-H]- 447 ==> 429 403 337 325 323 296 295 283 268 Flavonoid-glycoside: Possible 19-F1 P-Di 18 44 110 122 124 151 152 164 179 aglycones include Astragalin, Orientin, MS^3 323 ==> 323 307 297 295 281 266 251 Sakuranin, Kaempferol, and Luteolin P-Di 0 16 26 28 42 57 72 424 38.09 [M-H]- 423 ==> 405 379 355 313 299 285 271 259 217 Either Remangiflavanone B, 20-F1 P-Di 18 44 68 110 124 138 152 164 206 Euchrestaflavanone B, or MS^3 299 ==> 271 255 242 227 Sophoraflavanone G P-Di 28 44 57 72

216 Table 12: Detected Compounds in Fraction F using HPLC Method 2. Indicates Compound Number (CMPD #), Molecular Weight (MW) in atomic mass units (amu), Retention Time (RT) in minutes (min), Mass Spec Base Peak Identification (BP-ID) showing negative mode ionization [M-H]- (indicates the type of ion which forms the Base Peak. Base Peak refers to the most abundant ion in the mass spectrum), Parent ion (P-) m/z after [M-H]- ionization (indicates the Base Peak in the normal mass spectra), Base Peak of the daughter ion (D-BP) (i.e., the most abundant ion in the daughter spectrum), MS/MS or MS/MS/MS product ions (MS^3), and the arithmetic difference in m/z between the parent ion and individual daughter ions (P-Di). Bold donates most abundant ion in the given spectra. CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 316 19.48 [M-H]- 315 ==> 297 271 231 Either 6-methoxyluteolin (aka Nepetin), 1-F2 P-Di 18 44 84 or a methoxyquercetin (i.e. Rhamnetin MS^3 297 ==> 269 253 227 or Tamarixetin), or a Tetrahydroxy- P-Di 28 44 70 -methoxyflavone (such as Pedalitin) 320 27.49 [M-H]- 319 ==> 287 197 161 125 2-F2 P-Di 32 122 158 194 Oureatacatechin MS^3 197 ==> 165 (or a derivative of) P-Di 32 320 30.59 [M-H]- 319 ==> 287 197 165 153 125 3-F2 P-Di 32 122 154 166 194 Oureatacatechin MS^3 197 ==> 165 (or a derivative of) P-Di 32

217 Table 13: Detected Compounds in Fraction G-Red using HPLC Method 1. Indicates Compound Number (CMPD #), Molecular Weight (MW) in atomic mass units (amu), Retention Time (RT) in minutes (min), Mass Spec Base Peak Identification (BP-ID) showing negative mode ionization [M-H]- (indicates the type of ion which forms the Base Peak. Base Peak refers to the most abundant ion in the mass spectrum), Parent ion (P-) m/z after [M-H]- ionization (indicates the Base Peak in the normal mass spectra), Base Peak of the daughter ion (D-BP) (i.e., the most abundant ion in the daughter spectrum), MS/MS or MS/MS/MS product ions (MS^3), and the arithmetic difference in m/z between the parent ion and individual daughter ions (P-Di). Bold donates most abundant ion in each spectra. CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 578 10.89 [M-H]- 577 ==> 451 425 407 289 1-Gr1 P-Di 126 152 170 288 B-type Procyanidin dimer MS^3 425 ==> 407 297 273 P-Di 18 128 152 578 11.27 [M-H]- 577 ==> 451 425 407 289 2-Gr1 P-Di 126 152 170 288 B-type Procyanidin dimer MS^3 425 ==> 407 297 273 P-Di 18 128 152 578 12.90 [M-H]- 577 ==> 559 451 425 407 289 3-Gr1 P-Di 18 126 152 170 288 B-type Procyanidin dimer MS^3 425 ==> 407 381 339 299 273 211 187 P-Di 18 44 86 126 152 214 238 578 13.84 [M-H]- 577 ==> 559 451 425 407 289 4-Gr1 P-Di 18 126 152 170 288 B-type Procyanidin dimer MS^3 425 ==> 407 381 339 299 273 211 187 P-Di 18 44 86 126 152 214 238 578 16.00 [M-H]- 577 ==> 559 451 425 407 289 287 5-Gr1 P-Di 18 126 152 170 288 290 B-type Procyanidin dimer MS^3 425 ==> 407 381 341 273 P-Di 18 44 84 152 576 16.51 [M-H]- 575 ==> 539 449 423 407 289 285 6-Gr1 P-Di 36 126 152 168 286 290 A-type Proanthocyanidin dimer MS^3 449 ==> 405 325 287 285 243 P-Di 44 124 162 164 206 574 18.01 [M-H]- 573 ==> 529 511 447 421 283 7-Gr1 P-Di 44 62 126 152 290 Procyanidin dimer MS^3 447 ==> 429 403 325 285 242 201 P-Di 18 44 122 162 205 246 576 18.32 [M-H]- 575 ==> 539 449 423 289 285 8-Gr1 P-Di 36 126 152 286 290 A-type Proanthocyanidin dimer MS^3 447 ==> 325 287 285 P-Di 122 160 162 574 19.17 [M-H]- 573 ==> 555 529 511 447 421 323 283 9-Gr1 P-Di 18 44 62 126 152 250 290 Procyanidin dimer MS^3 447 ==> 429 385 361 337 325 285 271 243 227 P-Di 18 62 86 110 122 162 176 204 220 574 21.12 [M-H]- 573 ==> 529 447 283 10-Gr1 P-Di 44 126 290 Procyanidin dimer MS^3 447 ==> 429 403 337 285 P-Di 18 44 110 162 556 38.21 [M-H]- 555 ==> 445 433 403 390 323 283 11-Gr1 P-Di 110 122 152 165 232 272 Either Hegoflavone A, MS^3 433 ==> 404 389 361 283 269 255 243 or Morelloflavone P-Di 29 44 72 150 164 178 190 556 39.79 [M-H]- 555 ==> 445 433 403 390 323 283 12-Gr1 P-Di 110 122 152 165 232 272 Either Hegoflavone A, MS^3 433 ==> 404 389 361 283 269 255 243 or Morelloflavone P-Di 29 44 72 150 164 178 190

218 Table 14: Detected Compounds in Fraction G-Red using HPLC Method 2. Indicates Compound Number (CMPD #), Molecular Weight (MW) in atomic mass units (amu), Retention Time (RT) in minutes (min), Mass Spec Base Peak Identification (BP-ID) showing negative mode ionization [M-H]- (indicates the type of ion which forms the Base Peak. Base Peak refers to the most abundant ion in the mass spectrum), Parent ion (P-) m/z after [M-H]- ionization (indicates the Base Peak in the normal mass spectra), Base Peak of the daughter ion (D-BP) (i.e., the most abundant ion in the daughter spectrum), MS/MS or MS/MS/MS product ions (MS^3), and the arithmetic difference in m/z between the parent ion and individual daughter ions (P-Di). Bold donates most abundant ion in each spectra. CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 302 62.40 [M-H]- 301 ==> 286 283 258 257 242 217 199 179 1-Gr2 P-Di 15 18 43 44 59 84 102 122 3',5,7-Trihydroxy-4'-methoxyflavanone MS^3 286 ==> 242 215 201 199 (a.k.a. Hesperitin) P-Di 44 71 85 87 316 63.10 [M-H]- 315 ==> 283 Either 6-methoxyluteolin (aka Nepetin), 2-Gr2 P-Di 32 4’-methoxyquercetin (aka Tamarixetin), MS^3 283 ==> 255 239 227 215 211 or 7-methylquercetin (aka Rhamnetin) P-Di 28 44 56 68 72 286 64.81 [M-H]- 285 ==> 267 257 243 241 217 213 199 175 151 3-Gr2 P-Di 18 28 42 44 68 72 86 110 134 3',4',5,7-Tetrahydroxyflavone MS^3 241 ==> 224 213 199 (a.k.a. Luteolin) P-Di 17 28 42 300 70.41 [M-H]- 299 ==> 284 217 Chrysoeriol, Diosmetin, or other 300 4-Gr2 P-Di 15 82 amu MW flavonoid (such as MS^3 284 ==> 256 227 217 212 163 151 Geraldol, Tectorigenin, Kaempferide, P-Di 28 57 67 72 121 133 Pratensein, Farresol, etc.)

219 Table 15: Detected Compounds in Fraction G using HPLC Method 1. Indicates Compound Number (CMPD #), Molecular Weight (MW) in atomic mass units (amu), Retention Time (RT) in minutes (min), Mass Spec Base Peak Identification (BP-ID) showing negative mode ionization [M-H]- (indicates the type of ion which forms the Base Peak. Base Peak refers to the most abundant ion in the mass spectrum), Parent ion (P-) m/z after [M-H]- ionization (indicates the Base Peak in the normal mass spectra), Base Peak of the daughter ion (D-BP) (i.e., the most abundant ion in the daughter spectrum), MS/MS or MS/MS/MS product ions (MS^3), and the arithmetic difference in m/z between the parent ion and individual daughter ions (P-Di). Bold donates most abundant ion in the given spectra. CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 864 10.00 [M-H]- 863 ==> 711 693 575 559 449 Catechin/Epicatech trimer with 1-G1 P-Di 152 170 288 304 414 single A-type Inter-Flavanic MS^3 711 ==> 693 559 Linkage (IFL) P-Di 18 152 864 10.83 [M-H]- 863 ==> 711 693 575 559 449 Catechin/Epicatech trimer with 2-G1 P-Di 152 170 288 304 414 single A-type Inter-Flavanic MS^3 711 ==> 693 559 403 Linkage (IFL) P-Di 18 152 308 864 11.41 [M-H]- 863 ==> 711 693 575 559 449 Catechin/Epicatech trimer with 3-G1 P-Di 152 170 288 304 414 single A-type Inter-Flavanic MS^3 711 ==> 693 559 541 Linkage (IFL) P-Di 18 152 170 864 11.80 [M-H]- 863 ==> 711 693 575 559 449 Catechin/Epicatech trimer with 4-G1 P-Di 152 170 288 304 414 single A-type Inter-Flavanic MS^3 575 ==> 557 539 449 423 407 325 289 285 541 Linkage (IFL) P-Di 18 36 126 152 168 250 286 290 34 864 12.72 [M-H]- 863 ==> 711 693 575 559 449 Catechin/Epicatech trimer with 5-G1 P-Di 152 170 288 304 414 single A-type Inter-Flavanic MS^3 575 ==> 557 539 449 423 407 327 289 285 Linkage (IFL) P-Di 18 36 126 152 168 248 286 290 864 14.31 [M-H]- 863 ==> 711 693 575 573 559 451 411 Catechin/Epicatech trimer with 6-G1 P-Di 152 170 288 290 304 412 452 single A-type Inter-Flavanic MS^3 711 ==> 693 567 559 541 525 407 Linkage (IFL) P-Di 18 144 152 170 186 304 864 15.33 [M-H]- 863 ==> 711 693 575 559 449 Catechin/Epicatech trimer with 7-G1 P-Di 152 170 288 304 414 single A-type Inter-Flavanic MS^3 693 ==> 657 525 449 403 Linkage (IFL) P-Di 36 168 244 290 864 15.42 [M-H]- 863 ==> 711 693 575 559 449 Catechin/Epicatech trimer with 8-G1 P-Di 152 170 288 304 414 single A-type Inter-Flavanic MS^3 711 ==> 693 559 541 403 Linkage (IFL) P-Di 18 152 170 308 864 15.86 [M-H]- 863 ==> 711 693 575 Catechin/Epicatech trimer with 9-G1 P-Di 152 170 288 single A-type Inter-Flavanic MS^3 693 ==> 657 525 449 403 Linkage (IFL) P-Di 36 168 244 290 1152 16.39 [M-H]- 1151 ==> 863 575 539 449 423 407 Catechin/Epicatech tetramer 10-G1 P-Di 288 576 612 702 728 744 with one A-type Inter-Flavanic MS^3 575 ==> 557 539 449 423 407 289 285 Linkage (IFL): (Cat)Epi--(Cat)Epi- P-Di 18 36 126 152 168 286 290 -(Cat)Epi--A(IFL)--(Cat)Epi 864 16.86 [M-H]- 863 ==> 711 693 575 559 539 Catechin/Epicatech trimer with 11-G1 P-Di 152 170 288 304 324 single A-type Inter-Flavanic MS^3 575 ==> 557 539 449 423 407 391 289 285 245 Linkage (IFL) P-Di 18 36 126 152 168 184 286 290 330 1152 19.17 [M-H]- 1151 ==> 863 575 539 449 423 407 Catechin/Epicatech tetramer 12-G1 P-Di 288 576 612 702 728 744 with one A-type Inter-Flavanic MS^3 575 ==> 557 539 449 423 407 289 285 Linkage (IFL): (Cat)Epi--(Cat)Epi- P-Di 18 36 126 152 168 286 290 -(Cat)Epi--A(IFL)--(Cat)Epi 1152 19.95 [M-H]- 1151 ==> 861 575 539 423 Catechin/Epicatech tetramer 13-G1 P-Di 290 576 612 728 with one A-type Inter-Flavanic MS^3 575 ==> 557 539 449 423 289 285 Linkage (IFL): (Cat)Epi--A(IFL)- P-Di 18 36 126 152 286 290 -(Cat)Epi--(Cat)Epi--(Cat)Epi 576 21.64 [M-H]- 575 ==> 557 539 465 449 423 407 369 327 297 14-G1 P-Di 18 36 110 126 152 168 206 248 278 A-type Proanthocyanidin dimer MS^3 449 ==> 405 387 325 297 287 285 269 267 243 P-Di 44 62 124 152 162 164 180 182 206 590 23.51 [M-H]- 589 ==> 571 560 466 463 421 404 343 303 285 15-G1 P-Di 18 29 123 126 168 185 246 286 304 Manniflavanone (a biflavanone) MS^3 463 ==> 435 353 341 312 302 285 269 259 229 P-Di 28 110 122 151 161 178 194 204 234

220 Table 15: (cont.) CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 560 24.74 [M-H]- 559 ==> 433 423 407 289 269 UNKNOWN Biflavanoid. 16-G1 P-Di 126 136 152 270 290 Exhibits one catechin (or epicatechin) MS^3 433 ==> 323 311 309 297 281 271 269 monomer, attached to a second P-Di 110 122 124 136 152 162 164 unknown 270 amu monomer 560 27.71 [M-H]- 559 ==> 433 407 289 269 UNKNOWN Biflavanoid. 17-G1 P-Di 126 152 270 290 Exhibits one catechin (or epicatechin) MS^3 433 ==> 323 311 309 297 281 271 269 monomer, attached to a second P-Di 110 122 124 136 152 162 164 unknown 270 amu monomer 588 30.21 [M-H]- 587 ==> 569 477 464 451 407 375 313 300 271 18-G1 P-Di 18 110 123 136 180 212 274 287 316 Kolaflavanone (a biflavanone) MS^3 464 ==> 437 392 313 311 300 299 287 269 254 P-Di 27 72 151 153 164 165 177 195 210 424 39.35 [M-H]- 423 ==> 405 379 341 313 299 271 259 Either Remangiflavanone B, 19-G1 P-Di 18 44 82 110 124 152 164 Euchrestaflavanone B, or MS^3 299 ==> 271 255 242 225 214 Sophoraflavanone G P-Di 28 44 57 74 85

221 Table 16: Detected Compounds in Fraction G using HPLC Method 2. Indicates Compound Number (CMPD #), Molecular Weight (MW) in atomic mass units (amu), Retention Time (RT) in minutes (min), Mass Spec Base Peak Identification (BP-ID) showing negative mode ionization [M-H]- (indicates the type of ion which forms the Base Peak. Base Peak refers to the most abundant ion in the mass spectrum), Parent ion (P-) m/z after [M-H]- ionization (indicates the Base Peak in the normal mass spectra), Base Peak of the daughter ion (D-BP) (i.e., the most abundant ion in the daughter spectrum), MS/MS or MS/MS/MS product ions (MS^3), and the arithmetic difference in m/z between the parent ion and individual daughter ions (P-Di). Bold donates most abundant ion in each spectra. CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 862 37.98 [M-H]- 861 ==> 735 720 718 709 694 451 435 375 Catechin/Epicatech trimer 1-G2 P-Di 126 141 143 152 167 410 426 486 with two A-type Inter-Flavanic MS^3 709 ==> 584 411 285 Linkages (IFL): (Cat)Epi--A(IFL)- P-Di 125 298 424 -(Cat)Epi--A(IFL)--(Cat)Epi 862 44.27 [M-H]- 861 ==> 844 735 709 693 675 627 575 571 376 Catechin/Epicatech trimer 2-G2 P-Di 17 126 152 168 186 234 286 290 485 with two A-type Inter-Flavanic MS^3 571 ==> 411 Linkages (IFL): (Cat)Epi--A(IFL)- P-Di 160 -(Cat)Epi--A(IFL)--(Cat)Epi 1150 49.14 [M-H]- 1149 ==> 997 864 725 573 Catechin/Epicatech tetramer 3-G2 P-Di 152 285 424 576 with two A-type Inter-Flavanic MS^3 997 ==> 573 Linkages (IFL): (Cat)Epi--A(IFL)- P-Di 424 -(Cat)Epi--A(IFL)--(Cat)Epi--(Cat)Epi 320 58.73 [M-H]- 319 ==> 287 183 151 4-G2 P-Di 32 136 168 3,3',4',5,5',7-Hexahydroxyflavanone MS^3 183 ==> 169 151 (a.k.a. Dihydromyricetin) P-Di 14 32

222 Table 17: Detected Compounds in Fraction H using HPLC Method 1. Indicates Compound Number (CMPD #), Molecular Weight (MW) in atomic mass units (amu), Retention Time (RT) in minutes (min), Mass Spec Base Peak Identification (BP-ID) showing negative mode ionization [M-H]- (indicates the type of ion which forms the Base Peak. Base Peak refers to the most abundant ion in the mass spectrum), Parent ion (P-) m/z after [M-H]- ionization (indicates the Base Peak in the normal mass spectra), Base Peak of the daughter ion (D-BP) (i.e., the most abundant ion in the daughter spectrum), MS/MS or MS/MS/MS product ions (MS^3), and the arithmetic difference in m/z between the parent ion and individual daughter ions (P-Di). Bold donates most abundant ion in the given spectra. CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 1152 12.06 [M-H]- 1151 ==> 981 861 829 573 529 Catechin/Epicatech tetramer 1-H1 P-Di 170 290 322 578 622 with one A-type Inter-Flavanic MS^3 981 ==> 829 692 530 Linkage (IFL): (Cat)Epi--A(IFL)- P-Di 152 289 451 -(Cat)Epi--(Cat)Epi--(Cat)Epi 1152 12.67 [M-H]- 1151 ==> 981 861 829 709 573 Catechin/Epicatech tetramer 2-H1 P-Di 170 290 322 442 578 with one A-type Inter-Flavanic MS^3 861 ==> 709 569 451 Linkage (IFL): (Cat)Epi--A(IFL)- P-Di 152 292 410 -(Cat)Epi--(Cat)Epi--(Cat)Epi 876 13.41 [M-H]- 875 ==> 857 813 723 571 527 Flavonoid trimer with two A-type IFL 3-H1 P-Di 18 62 152 304 348 linkages. Two monomers consist of MS^3 723 ==> 679 512 423 353 catechin (or epicatechin), while the P-Di 44 211 300 370 other is an O-methylated-(Cat)Epi. 1152 13.93 [M-H]- 1151 ==> 981 863 711 573 Catechin/Epicatech tetramer 4-H1 P-Di 170 288 440 578 with one A-type Inter-Flavanic MS^3 863 ==> 711 573 451 Linkage (IFL): (Cat)Epi--(Cat)Epi- P-Di 152 290 412 -A(IFL)--(Cat)Epi--(Cat)Epi 1166 14.22 [M-H]- 1165 ==> 1013 999 847 727 591 Flavonoid tetramer with two A-type 5-H1 P-Di 152 166 318 438 574 IFL linkages. Three of the monomers MS^3 999 ==> 847 424 285 consist of (Cat)Epi, with the last P-Di 152 575 714 being an O-methylated-(Cat)Epi. 1152 15.51 [M-H]- 1151 ==> 981 862 861 711 693 575 Catechin/Epicatech tetramer 6-H1 P-Di 170 289 290 440 458 576 with two A-type Inter-Flavanic MS^3 711 ==> 693 559 Linkages (IFL): (Cat)Epi--A(IFL)- P-Di 18 152 -(Cat)Epi--(Cat)Epi--A(IFL)--(Cat)Epi 860 16.28 [M-H]- 859 ==> 707 569 443 7-H1 P-Di 152 290 416 UNKNOWN trimer, most likely with MS^3 707 ==> 569 433 two A-type IFL linkages P-Di 138 274 1148 18.04 [M-H]- 1147 ==> 995 857 571 Catechin/Epicatech tetramer 8-H1 P-Di 152 290 576 with three A-type IFL Linkages: MS^3 995 ==> (Cat)Epi--A(IFL)--(Cat)Epi- P-Di -A(IFL)--(Cat)Epi--A(IFL)--(Cat)Epi 876 18.41 [M-H]- 875 ==> 857 839 723 571 557 463 Flavonoid trimer with two A-type IFL 9-H1 P-Di 18 36 152 304 318 412 linkages. Two monomers consist of MS^3 857 ==> 731 571 catechin (or epicatechin), while the P-Di 126 286 other is an O-methylated-(Cat)Epi. 862 18.74 [M-H]- 861 ==> 825 735 709 571 Catechin/Epicatech trimer 10-H1 P-Di 36 126 152 290 with two A-type Inter-Flavanic MS^3 825 ==> Linkages (IFL): (Cat)Epi--A(IFL)- P-Di -(Cat)Epi--A(IFL)--(Cat)Epi 862 19.41 [M-H]- 861 ==> 825 735 709 571 Catechin/Epicatech trimer 11-H1 P-Di 36 126 152 290 with two A-type Inter-Flavanic MS^3 709 ==> 571 541 523 448 285 Linkages (IFL): (Cat)Epi--A(IFL)- P-Di 138 168 186 261 424 -(Cat)Epi--A(IFL)--(Cat)Epi 862 21.68 [M-H]- 861 ==> 825 735 709 571 Catechin/Epicatech trimer 12-H1 P-Di 36 126 152 290 with two A-type Inter-Flavanic MS^3 709 ==> 571 Linkages (IFL): (Cat)Epi--A(IFL)- P-Di 138 -(Cat)Epi--A(IFL)--(Cat)Epi 862 22.95 [M-H]- 861 ==> 825 735 709 693 571 449 Catechin/Epicatech trimer 13-H1 P-Di 36 126 152 168 290 412 with two A-type Inter-Flavanic MS^3 735 ==> 699 583 445 Linkages (IFL): (Cat)Epi--A(IFL)- P-Di 36 152 290 -(Cat)Epi--A(IFL)--(Cat)Epi

223 Table 18: Detected Compounds in Fraction H using HPLC Method 2. Indicates Compound Number (CMPD #), Molecular Weight (MW) in atomic mass units (amu), Retention Time (RT) in minutes (min), Mass Spec Base Peak Identification (BP-ID) showing negative mode ionization [M-H]- (indicates the type of ion which forms the Base Peak. Base Peak refers to the most abundant ion in the mass spectrum), Parent ion (P-) m/z after [M-H]- ionization (indicates the Base Peak in the normal mass spectra), Base Peak of the daughter ion (D-BP) (i.e., the most abundant ion in the daughter spectrum), MS/MS or MS/MS/MS product ions (MS^3), and the arithmetic difference in m/z between the parent ion and individual daughter ions (P-Di). Bold donates most abundant ion in the given spectra. CMPD MW RT BP-ID P- ==> D-BP Other Daughter Ions Possible Structure ID(s) 864 21.52 [M-H]- 863 ==> 711 693 575 Catechin/Epicatech trimer with 1-H2 P-Di 152 170 288 single A-type Inter-Flavanic MS^3 711 ==> Linkage (IFL) P-Di 864 24.03 [M-H]- 863 ==> 711 575 449 Catechin/Epicatech trimer with 2-H2 P-Di 152 288 414 single A-type Inter-Flavanic MS^3 711 ==> Linkage (IFL) P-Di 864 27.86 [M-H]- 863 ==> 711 693 575 Catechin/Epicatech trimer with 3-H2 P-Di 152 170 288 single A-type Inter-Flavanic MS^3 711 ==> Linkage (IFL) P-Di 864 32.36 [M-H]- 863 ==> 575 423 Catechin/Epicatech trimer with 4-H2 P-Di 288 440 single A-type Inter-Flavanic MS^3 575 ==> Linkage (IFL) P-Di 864 35.21 [M-H]- 863 ==> 711 693 575 Catechin/Epicatech trimer with 5-H2 P-Di 152 170 288 single A-type Inter-Flavanic MS^3 711 ==> Linkage (IFL) P-Di

224 O O - O - C - CH -H -HO C -CO2 OH OH OH OH OH OH OH C C m/z 119 OH OH m/z 161 C H O m/z 180 m/z 179 C H O 8 6 C H O C H O 9 6 3 9 8 4 9 7 4 Figure 1: Fragmentation Scheme for 3,4-Dihydroxycinnamic Acid (a.k.a Caffeic Acid)

O O O C -H -C6H5 - O - OH O O O O m/z 161 m/z 238 m/z 237 C9H4O3 C H O C H O 15 10 3 15 9 3 Figure 2: Fragmentation Scheme for 3-Hydroxyflavone (a.k.a. Flavonol)

O O O - - -H C -C6H5 CH C

CH3 CH3 CH3 O O O m/z 238 m/z 237 m/z 161 C H O C H O C H O 16 14 2 16 13 2 10 8 2 Figure 3: Fragmentation Scheme for 4’-Methoxychalcone

225 OH OH OH

OH OH OH OH

- - -C H O OH OH O -H OH OC -HO OH C O 9 3 3 OH OH OH - C OH C m/z 126

OH O OH O O C6H5O3 m/z 302 m/z 301 m/z 285 C H O C H O C H O 15 10 7 15 9 7 15 8 6 Figure 4: Fragmentation Scheme for 3',4',5,5',7-Pentahydroxyflavone (a.k.a. Tricetin)

- OHH O O OH OH C

-H -C9H4O3 -HO O O O O - O - O CH CH

O O O O O O O O O O

CH3 CH3 CH3 CH3 m/z 380 m/z 379 m/z 218 m/z 202 C H O C H O C H O C H O 21 16 7 21 15 7 12 11 4 12 10 3 Figure 5: Fragmentation Scheme for 4,4'-Dihydroxy-3,3'-(2-methoxyethylidene) dicoumarin (a.k.a. Dicoumoxyl)

226 - - OHH O O OH O C -H -C9H5O3 O O O O O

O O O O O O O O

CH3 CH3 CH3 m/z 380 m/z 379 m/z 217 C H O C H O C H O 21 16 7 21 15 7 12 10 4

OHH O OHH O OHH O OH -H -C2H4O -HO O O O O C

O - O O - O O O O O CH CH O O O O O O - m/z 335 m/z 318 CH3 CH2 m/z 380 m/z 379 C19H11O6 C19H10O5 C H O C H O 21 16 7 21 15 7 Figure 5 (cont.): Fragmentation Scheme for 4,4'-Dihydroxy-3,3'-(2-methoxyethylidene) dicoumarin (a.k.a. Dicoumoxyl)

227 OH OH OH OH OH - - -H C -HO C C

O OO O O OO O O OO O m/z 336 m/z 335 m/z 319

C19H12O6 C19H11O6 C19H10O5

OH OH - - C -CO2 C

C C O O O OO O m/z 273 m/z 319 C H O C H O 18 10 3 19 10 5

- - OH OH OH O O -C H O -H 9 5 3 2CH C

O OO O O OO O O O m/z 336 m/z 335 m/z 173

C19H12O6 C H O C H O 19 11 6 10 6 3 Figure 6: Fragmentation Scheme for 3,3'-Methylenebis[4-hydroxycoumarin] (a.k.a. Dicoumarin)

O O - O - O - O - -H C -CH3 C C C C -C6H4 C

CH3 CH3 O O O O O O O O O O m/z 252 m/z 251 m/z 235 m/z 235 m/z 161 C H O C H O C H O C H O C H O 16 12 3 16 11 3 15 8 3 15 8 3 9 4 3 Figure 7: Fragmentation Scheme for 3-Methoxyflavone

228 O O O

OH OH OH -H -HO - - C C

C OH OH m/z 216 m/z 215 m/z 198 C H O C H O C H O 13 12 3 13 11 3 13 10 2

O O - CH - 2 OH CH OH -H -CO2

OH OH OH m/z 172

m/z 216 m/z 215 C12H11O C H O C H O 13 12 3 13 11 3

O O

- - OH O CH2 -H -C2H2O2

OH m/z 157 OH OH C11H9O m/z 216 m/z 215 C H O C H O 13 12 3 13 11 3 Figure 8: Fragmentation Scheme for 6-Hydroxy-2-naphthalenepropanoic Acid (a.k.a. Allenolic Acid)

229 O O O C OH OH C OH -H -HO -C2H2O OH OH OH - C OH - - C C m/z 108

OH OH OH C6H4O2 m/z 168 m/z 167 m/z 151 C H O C H O C H O 8 8 4 8 7 4 8 6 3 Figure 9: Fragmentation Scheme for Homogentisic Acid

OH OH - OH O OH O - C -C H O OH O -H OH O 6 4 2

O m/z 161

O O C9H5O3 m/z 270 m/z 269 C H O C H O 15 10 5 15 9 5 Figure 10: Fragmentation Scheme for 3',4',7-Trihydroxyflavone

230 O OH O OH O OH

-H - -HO C - C OH OH OH OH OH CH OH OH OH m/z 155 m/z 174 m/z 173 C7H8O4 C H O C H O 7 10 5 7 9 5

O OH O OH C

-H -CHO2 - OH C OH - OH OH OH C OH OH OH OH m/z 129 m/z 174 m/z 173 C6H8O3 C H O C H O 7 10 5 7 9 5 Figure 11: Fragmentation Scheme for Shikimic Acid

OH OH OH OH OH

OH OH OH OH OH OH OH

-H OO O OO O - O 3CH OO 3CH OO C -C12H20O10

- OH OH OH O OH OH O OH OH O OH O m/z 596 m/z 595 m/z 271 C H O C H O C H O 27 32 15 27 31 15 15 11 5 Figure 12: Fragmentation Scheme for Eriodictyol 7-O-rutinoside (a.k.a. Eriocitrin)

231 - OH OH OH O

OH OH OH OH OH OH

OO O -H OO O CH OO 3 3CH OO

OH OH OH OH OH OH O OH OH O m/z 596 m/z 595 C H O 27 32 15 C27H31O15

- OH OH O OH OH OH OH C

-C6H3O -HO OO O - OO O CH OO CH 3 3CH OO

OH OH OH OH OH OH O OH OH O m/z 487 m/z 579

C21H27O13 C H O 27 30 14 Figure 12 (cont.): Fragmentation Scheme for Eriodictyol 7-O-rutinoside (a.k.a. Eriocitrin)

OH OH OH OH - - OH OH C OH C -H -HO -C8H5O - C C OH OH OH OH OH OH m/z 109 m/z 244 m/z 243 m/z 225 C6H5O2 C H O C H O C H O 14 12 4 14 11 4 14 10 3 Figure 13: Fragmentation Scheme for Oxyresveratrol

232 3CH 3CH 3CH

3CH 3CH 3CH -H - -HO - O O C O C

C HOO O HOO O O O m/z 244 m/z 243 m/z 225 C H O C H O C H O 14 12 4 14 11 4 14 10 3

CH CH - CH - - 3 3 C 2 CH CH 3CH 3CH 3CH -H -C3H5 O O O O C

HOO O HOO O HOO O HOO O m/z 244 m/z 243 m/z 243 m/z 201 C H O C H O C H O C H O 14 12 4 14 11 4 14 11 4 11 6 4 Figure 14: Fragmentation Scheme for Cedrecoumarin A

OH OH OH - C -H -CHO2 - C OH OH OH OH OH C OH OOH OOH m/z 125 m/z 170 m/z 169 C6H4O3 C H O C H O 7 6 5 7 5 5 Figure 15: Fragmentation Scheme for 2,4,6-Trihydroxybenzoic Acid

233 OH OH OH OH OH

-H -HO -C3O2 - C C CH OH OH OH OH OH OH C OH - - - OOH OO OO OO m/z 83 m/z 170 m/z 169 m/z 153 m/z 153 C4H4O2 C H O C H O C H O C H O 7 6 5 7 5 5 7 4 4 7 4 4 Figure 15 (cont.): Fragmentation Scheme for 2,4,6-Trihydroxybenzoic Acid

O O O CH3 CH3 CH3 - - O O O C O O C O -H -HO 3CH 3CH 3CH

CH3 CH3 CH3 O O C O OH O OH O O m/z 328 m/z 327 m/z 309 C H O C H O C H O 18 16 6 18 15 6 18 14 5

O O CH CH - 3 3 O C O - O O O C O 3CH C -H -C7H7O 3CH 3CH CH3 O CH3 CH3 O O OH O OH O OH O m/z 221

m/z 328 m/z 327 C11H8O5 C H O C H O 18 16 6 18 15 6 Figure 16: Fragmentation Scheme for Kaempferol-3,4',7-trimethyl Ether

234 OO O OO O - O - OH OH C C C

-H -C6H11O6 OH OH OH OH

OH OH O CH3 OH OH O CH3 OH O CH3 O O O m/z 446 m/z 445 m/z 265 C H O C H O C H O 22 22 10 22 21 10 16 10 4

OO O O - O O OH OH C -H OH OH OH OH

OH OH O CH3 OH OH O CH3 O O m/z 446 m/z 445

C22H22O10 C22H21O10

- O O O - O O CH OH C

-C2H4O2 -HO OH OH C OH OH C

OH O CH3 OH O CH3 O O m/z 367 m/z 429 C H O C H O 20 16 7 22 20 9 Figure 17: Fragmentation Scheme for Biochanin-A-7-glucoside (a.k.a. Sissotrin)

235 O O O - O - C - - C CH -H -HO C -HO C -CHO2 OH OH OH OH OH OH OH

C m/z 101 OH OH m/z 161 m/z 145 C H m/z 180 m/z 179 C H O C H O 8 4 C H O C H O 9 6 3 9 5 2 9 8 4 9 7 4

O O - O - C - CH -H -HO C -CO2 OH OH OH OH OH OH OH C C m/z 117 OH OH m/z 161 C H O m/z 180 m/z 179 C H O 8 6 C H O C H O 9 6 3 9 8 4 9 7 4 Figure 18: Fragmentation Scheme for 3,4-Dihydroxycinnamic Acid (a.k.a. Caffeic Acid)

OH O O OH O O O O -H -HO C

- - C C m/z 162 m/z 161 m/z 143 C H O C H O C H O 9 6 3 9 5 3 9 4 2 Figure 19: Fragmentation Scheme for 7-Hydroxycoumarin (a.k.a. Umbelliferone)

236 OH OH OH

- - -H CH -HO CH

C OH O OH O O m/z 242 m/z 241 m/z 223 C H O C H O C H O 15 14 3 15 13 3 15 12 2 Figure 20: Fragmentation Scheme for 4',7-Dihydroxyisoflavan (a.k.a. Equol)

- C - - - C C C -H -HO -CO2 CH O O O O O O O O C O O O O O OH OH m/z 183 m/z 183 m/z 139 m/z 202 m/z 201 C11H4O3 C11H4O3 C10H4O C11H6O4 C H O 11 5 4 Figure 21: Fragmentation Scheme for 8-Hydroxy-4'-5',6-7-furocoumarin (a.k.a. Xanthotoxol)

237 OOH OOH - - OH C O OH O - - -CHO -C H OH O OH OH -H OH O 2 2 2

CH CH m/z 109 m/z 109 m/z 81 C4H2O2 m/z 154 m/z 153 C6H4O2 C6H4O2 C H O C H O 7 6 4 7 5 4 Figure 22: Fragmentation Scheme for 2,6-Dihydroxybenzoic Acid (a.k.a. g-Resorcylic Acid)

O O O O O

-H -HO C -C2H OH OH CH - - - - C C C CH CH OH OH OH OH OH m/z 138 m/z 137 m/z 119 m/z 119 m/z 95 C H O C H O C H O C H O C H O 7 6 3 7 5 3 7 4 2 7 4 2 5 3 2 Figure 23: Fragmentation Scheme for 2-Hydroxybenzoic Acid (a.k.a. Salicylic Acid)

CH3 CH3 O O O O O - - - -H -C2H5 O O -C2H2 O - CH OH O

CH m/z 108 m/z 108 m/z 81

m/z 138 m/z 137 C6H4O2 C6H4O2 C4H2O2 C H O C H O 8 10 2 8 9 2 Figure 24: Fragmentation Scheme for Pyrocatechin Monoethyl Ether

238 CH - CH3 CH3 CH3 3 CH2 - O O O O O C OH -H -HO -H -C2H4O OH OH OH OH C m/z 93 - - C C C6H5O m/z 138 m/z 137 m/z 119 m/z 138 m/z 137 C H O C H O C H O C8H10O2 C H O 8 10 2 8 9 2 8 8 8 9 2 Figure 24 (cont.): Fragmentation Scheme for Pyrocatechin Monoethyl Ether

OH O OH O OH O

-H -C6H4O - C

O O - O OH O m/z 254 m/z 253 m/z 160 C H O C H O C H O 9 5 3 15 10 4 15 9 4

OH O OH O - O - C C C -H -HO

O O O OH OH OH m/z 254 m/z 253 m/z 235 C H O C H O C H O 15 10 4 15 9 4 15 8 3 Figure 25: Fragmentation Scheme for 4',7-Dihydroxyisoflavone (a.k.a. Daidzein)

239 OH OH OH OH OH OH OH OH OH

O O O -H - O O O -O O O C CH

OH OH OH OH OH OH

OH OH O OH OH O OH OH O m/z 450 m/z 449 m/z 449

C21H22O11 C21H21O11 C21H21O11

OH OH OH OH

- - O O -HO O O

-C6H10O5

C O OH O m/z 269 m/z 287 C H O C H O 15 10 5 15 11 6 Figure 26: Fragmentation Scheme for Eriodictyol-7-glucoside

OH OH OH OH OH OH OH OH

-H -CHO OH O OH O - OH O - OH O - C CH C CH

OH OH O O O O O m/z 288 m/z 287 m/z 287 m/z 259 C H O C H O C H O C H O 15 12 6 15 11 6 15 11 6 14 10 5 Figure 27: Fragmentation Scheme for 2,3-Dihydrofisetin (a.k.a. Fustin)

240 OH OH OH OH OH OH

-H -HO OH O OH O - O - C C C

OH OH OH O O O m/z 288 m/z 287 m/z 269 C H O C H O C H O 15 12 6 15 11 6 15 10 5 Figure 27 (cont.): Fragmentation Scheme for 2,3-Dihydrofisetin (a.k.a. Fustin)

OH OH OH

OH O OH O OH O OH OH OH

O O O

OH O OH -H OH O OH -HO OH O O O OCH

3CH OO 3CH OO 3CH OO - - OH O O

OH OH OH OH OH OH OH OH OH

OH OH OH m/z 610 m/z 609 m/z 591

C27H30O16 C27H29O16 C27H28O15

OH

OH O OH OH

O OH O -C H O OH 12 19 8 OH O O - O 3CH OO - O OH O m/z 301 OH C15H9O7 OH OH OH m/z 591 C H O 27 28 15 Figure 28: Fragmentation Scheme for Quercetin-3-O-rutinose (a.k.a. Rutin)

241 OH OH O O OCH3 OCH3 OCH3 CH2 - O O O O - -C H O - -C H O O -H C 15 8 3 O 2 3 2 OH O O O O O O O OH OH OH OH OH OH m/z 163 OH OH OH C H O OH OH m/z 221 6 10 5 m/z 458 m/z 457 C8H13O7 C H O C H O 23 22 10 23 21 10 Figure 29: Fragmentation Scheme for 6-O-Acetyldaidzin

OH OH OH - C - O O C -H -CH3 O

OH OH CH OH O O CH CH O 3 3 m/z 179 m/z 194 m/z 193 C H O C H O C H O 9 6 4 10 10 4 10 9 4 OH OH - - CH O O C -H -CO2 OH OH OH O O O CH3 CH3 CH3 m/z 149 m/z 194 m/z 193 C9H9O2 C H O C H O 10 10 4 10 9 4 Figure 30: Fragmentation Scheme for 3-Methoxy-4-hydroxycinnamic Acid (a.k.a. Ferulic Acid)

242 OH OH OH OH OH OH

-H -HO OO O OO O 2CH OO O OH OH

- - OH OH OH O OH OH O OH OH OH OH O OH OH O OH OH O m/z 464 m/z 463 m/z 447

C21H20O12 C21H19O12 C21H18O11

OH OH OH OH

- -C6H9O4 O O 2CH O O O

- OH O OH OH OH O OH OH O m/z 301 m/z 447 C H O C H O 15 9 7 21 18 11

OH OH

OH - OH C

OO O -H OO O OH OH

OH OH OH OH OH OH OH OH O OH OH O m/z 464 m/z 463

C21H20O12 C21H19O12

OH OH

- OH - OH OO - C C OH C -C9H4O4 -HO C OO O OO O OH OH OH OH C C OH OH OH OH OH OH m/z 271

C12H14O7 OH OH O OH OH O m/z 447 m/z 447 C H O C H O 21 18 11 21 18 11 Figure 31: Fragmentation Scheme for Quercetin-7-D-glucoside

243 OH - OH - OH C C

OH O CH3 OH O CH3 -CH O OH O C O -H O 3

OH OH OH OH O OH O OH O m/z 316 m/z 315 m/z 285 C H O C H O C H O 16 12 7 16 11 7 15 8 6

- - OH C OH C OH

OH O CH 3 OH O CH3 -CH OH O O -H O 3 CH O

OH OH OH OH O OH O OH O m/z 316 m/z 315 m/z 300 C H O 16 12 7 C16H11O7 C15H8O7

- - CH C OH C OH O OH O -C2O2 CH2 CH2 O

OH OH OH O OH O m/z 243 m/z 300 C H O C H O 13 8 5 15 8 7 Figure 32: Fragmentation Scheme for 3-Methylquercetin (a.k.a. Isorhamnetin)

244 OH OH OH O O CH3 CH3 C

-H -CH3O OH O OH O - OH O - C C

OH O OH O OH O m/z 302 m/z 301 m/z 271 C H O C H O C H O 16 14 6 16 13 6 15 10 5

OH OH OH O O C O CH3 CH3

-H -CH3 OH O OH O - OH O - C C

OH O OH O OH O m/z 302 m/z 301 m/z 286 C H O C H O C H O 16 14 6 16 13 6 15 10 6 Figure 33: Fragmentation Scheme for 3',5,7-Trihydroxy-4'-methoxyflavanone (a.k.a. Hesperitin)

OH - OH - OH C C C OH O OH O OH O -CH O -H O 3 O

CH3 CH3

OH O OH O OH O m/z 300 m/z 299 m/z 284 C H O C H O C H O 16 12 6 16 11 6 15 8 6 Figure 34: Fragmentation Scheme for 5,7,4'-Trihydroxy-3'-methoxyflavone (a.k.a. Chrysoeriol)

245 - - OH O OH OC OH OC

CH -H -CH3 3 3CH O O O C OH O OH O OH O OH OH OH m/z 300 m/z 299 m/z 284

C16H12O6 C H O C H O 16 11 6 15 8 6 Figure 35: Fragmentation Scheme for 4',5,7-Trihydroxy-6-methoxyisoflavone (a.k.a. Tectorigenin)

CH CH CH 3 3 3 OH OH - OH - 3CH CH3 3CH C CH3 3CH C O O O O O O

-H -CH3 OH OH OH

3CH OH O 3CH OH O 3CH OH O

CH3 CH3 CH3 m/z 436 m/z 435 m/z 420

C26H28O6 C26H27O6 C25H24O6

OH 3CH - OH 3CH CH - O C O OCO

-C5H8 OH OH

CH OH O 3 3CH OH O

CH 3 CH3 m/z 351 m/z 420 C20H16O6 C H O 25 24 6 Figure 36: Fragmentation Scheme for Orotinichalcone

246 CH CH - CH - 3 OH 3 C OH 2 CH OH 3CH CH3 3CH CH3 3CH CH3 O O O O O O -H OH OH OH

3CH OH O 3CH OH O 3CH OH O

CH3 CH3 CH3 m/z 436 m/z 435 m/z 435

C26H28O6 C26H27O6 C26H27O6

- CH OH - CH CH OH 3 CH3 O C O -C5H9 OH -C H O OH 3 5

3CH OH O OH O m/z 309 CH3 C18H13O5 m/z 377 C H O 23 22 5

CH CH 3 OH 3 OH CH CH CH CH CH 3 3 3 3 3 OH O O O O 3CH CH3 O O -H -C5H8 OH OH - C OH

3CH OH O 3CH OH O OH O - m/z 366 CH3 CH2 m/z 436 m/z 435 C21H19O6

C26H28O6 C26H27O6

CH 3 OH 3CH O C O -CH3

- C OH

OH O m/z 351 C H O 20 16 6 Figure 36 (cont.): Fragmentation Scheme for Orotinichalcone

247

3CH 3CH 3CH OH - OH - 3CH CH3 3CH C CH3 3CH C CH3 O O O O O O

-H -C8H7O2 OH OH C

3CH OH O 3CH OH O 3CH OH O

CH3 CH3 CH3 m/z 436 m/z 435 m/z 299 C H O C H O C H O 26 28 6 26 27 6 18 20 4

3CH 3CH 3CH OH - OH - OH 3CH CH3 3CH C CH3 3CH C O O O O O C -H -CH3O OH OH OH

3CH OH O 3CH OH O 3CH OH O

CH3 CH3 CH3 m/z 436 m/z 435 m/z 405 C H O C H O C H O 26 28 6 26 27 6 25 24 5

3CH 3CH 3CH OH - OH - 3CH CH3 3CH C CH3 3CH C CH3 O O O O O O

-H -C9H7O3 OH OH C

3CH OH O 3CH OH O 3CH OH

CH3 CH3 CH3 m/z 436 m/z 435 m/z 271 C H O C H O C H O 26 28 6 26 27 6 17 20 3 Figure 36 (cont.): Fragmentation Scheme Orotinichalcone

248 O O - O C - -H -C6H5O2 CH C

HOO HOHHOO HOH HOO H OH OH m/z 161 m/z 272 m/z 271 C9H6O3 C H O C H O 15 12 5 15 11 5

O O - - C C -H -C9H6O3 OH HOO HOHHOO HOH OH OH OH m/z 109 m/z 272 m/z 271 C6H5O2 C H O C H O 15 12 5 15 11 5 Figure 37: Fragmentation Scheme for 2',3,4,4'-Tetrahydroxychalcone (a.k.a. Butein)

OH O OH O

- OH C 3CH OH 3CH OH O O -C H O -H 10 6 5 CH3 - O OH O CH3 OO CH3 O O O O O CH3 CH3 CH3 m/z 153 m/z 360 m/z 359 C8H9O3 C H O C H O 18 16 8 18 15 8 Figure 38: Fragmentation Scheme for 3',5,7-Trihydroxy-4',5',6-trimethoxyisoflavone (a.k.a. Irigenin)

249 OH O OH O OH O

- - 3CH OH 2CH OH 2CH OH O -H O -HO O C

OH O CH3 OH O CH3 O CH3 O O O O O O CH3 CH3 CH3 m/z 360 m/z 359 m/z 341

C18H16O8 C18H15O8 C18H14O7

OH O OH O C C - - C OH 2CH OH O -C3H2O

O CH3 O CH3 O O O O CH3 CH3 m/z 289 m/z 341

C15H12O6 C18H14O7

OH O OH O

- OH O OH O 3CH OH 3CH O O O -H -C8H8O3 - -CH3O - 3CH C C C OH O CH3 OH O CH3 O O O OH O OH O O O m/z 207 m/z 177 CH CH 3 3 C H O C H O m/z 360 m/z 359 10 7 5 9 4 4 C H O C H O 18 16 8 18 15 8 Figure 38 (cont.): Fragmentation Scheme for 3',5,7-Trihydroxy-4',5',6-trimethoxyisoflavone (a.k.a. Irigenin)

250 OH OH OH OH OH C OH O -H -HO -HO - OH O OH O - OH O - C OH OH C OH C OH OH OH OH OH OH O OH O OH O OH O m/z 284 m/z 318 m/z 317 m/z 299 C15H7O6 C H O C H O C H O 15 10 8 15 9 8 15 8 7

OH OH OH

OH OH OH - OH OC - - OH O -H OH OC -HO OH OC -C6H5O3 OH OH OH C OH OH OH O m/z 176 OH O OH O OH O C9H3O4 m/z 318 m/z 317 m/z 299 C H O C H O C H O 15 10 8 15 9 8 15 8 7

Figure 39: Fragmentation Scheme for 3,3',4',5,5',7-Hexahydroxyflavone (a.k.a. Myricetin)

251 - OH O OH OH OH OH OH OH OH - OH O OH OH OH OH OO -C H O -H OO 6 10 5 OH O OH O OH O

OH OH OH OH O m/z 317 OH O OH O m/z 480 m/z 479 C15H9O8 C H O 21 20 13 C21H19O13

OH

- OH O OH O -HO

C OH O m/z 299

C15H8O7 - OH O OH OH OH OH OH OH OH OH OH OH OH OH OO -C H O - -H OO 6 10 6 OH C O OH O OH O

OH OH OH OH O m/z 300 OH O OH O m/z 480 m/z 479 C15H9O7 C H O 21 20 13 C21H19O13

OH OH

- OH C O -HO

C OH O m/z 284 C H O 15 8 6 Figure 40: Fragmentation Scheme for Gossypetin-8-glucoside (a.k.a. Gossypin)

252

- - OH O O CH3 CH O CH O 3 CH3 O CH3 O 3 CH3 O CH3 CH O CH CH O 3 3 3 O O -CH -C H O - O O -H O O 3 O O 6 3 3 C OH OH OH O O O O CH3 OH O CH OH O 3 CH3 OH O CH3 OH O m/z 250 m/z 390 m/z 389 m/z 374 C12H11O6 C H O 19 18 9 C19H17O9 C18H14O9

CH3 CH3 O

O O - C -HO

O C

CH3 O m/z 233

C12H10O5 OH OH OH

CH3 O CH3 O CH3 O CH3 O CH3 CH3 O CH3 CH3 O CH3 O O -H O O -HO O O OH OH OH - - C C O O O C

CH3 OH O CH3 OH O CH3 O m/z 390 m/z 389 m/z 374

C19H18O9 C19H17O9 C19H16O8

OH OH

O CH3 O CH3 CH3 CH3 O CH3 -CH O -CH O O OC 3 O O 3 OH OH - - C C

O O m/z 309 m/z 341 C H O C H O 17 10 6 18 13 7 Figure 41: Fragmentation Scheme for Gardenin E

253 OH OH OH OH OH OH

- - OH O OH OC OH OC -H -HO

C OH OH OH O OH O OH O m/z 302 m/z 301 m/z 283 C H O C H O C H O 15 10 7 15 9 7 15 8 6 Figure 42: Fragmentation Scheme for 2',3,4',5,7-Pentahydroxyflavone (a.k.a. Morin)

OH OH OH

OH - OH - C C C

OH O -H OH O -HO OH O OH OH OH

OH OH OH O O O m/z 302 m/z 301 m/z 283 C H O C H O C H O 15 10 7 15 9 7 15 8 6 Figure 43: Fragmentation Scheme for 3,3',4',5',7-Pentahydroxyflavone (a.k.a. Robinetin)

O O - - CH C -H -CO2 OH OH OH OH OH

OH OH OH m/z 135 m/z 180 m/z 179 C H O C H O C H O 8 7 2 9 8 4 9 7 4 Figure 44: Fragmentation Scheme for 3,4-Dihydroxycinnamic Acid (a.k.a. Caffeic Acid)

254 OH OH OH

OH O OH O OH O

-H 3 OCH 3CH O 3CH O

OH O - OH O - OH O OH OH C O CH OH OH OH m/z 400 m/z 399 m/z 399

C21H20O8 C21H19O8 C21H19O8

OH OH OH O OH O -C5H7O3 -CH3 O - C

- OH O O CH OH O m/z 269 OH C15H9O5 m/z 384

C20H16O8 OH OH OH

OH O OH O OH O

-H 3 OCH 3CH O - 3CH O - C CH

OH O OH O CH2 OH O OH OH OH

OH OH OH m/z 400 m/z 399 m/z 399

C21H20O8 C21H19O8 C21H19O8

OH OH OH O OH O -C4H7O2 -CH3 O - CH CH O - C CH2 OH O OH OH O m/z 297 OH C16H9O6 m/z 384 C H O 20 16 8 Figure 45: Fragmentation Scheme for Torosaflavone A

255

OH OO OH O O OH O O

CH -H CH -CH3O 3 3 - 3CH - C C C

CH3 O CH3 O CH3 3CH HOO H 3CH HOO H HOO H m/z 368 m/z 367 m/z 337

C21H20O6 C21H19O6 C20H16O5

OH O O

-C5H9 - C

HOO H m/z 267 C H O 15 7 5

OH OO OH O O OH O O

CH -H CH -CH3 3 3 - 3CH C - C C

CH3 O CH3 O CH3 O 3CH HOO H 3CH HOO H HOO H m/z 368 m/z 367 m/z 352

C21H20O6 C21H19O6 C20H16O6

OH O O

-C4H7 - 2CH C O HOO H m/z 297 C H O 16 9 6 Figure 46: Fragmentation Scheme for Glycycoumarin

256 3CH CH3 3CH CH3 3CH CH3 3CH CH3

- - - CH3 CH3 CH CH CH

O -H O -CH3 O O

CH

OO OH OO OH OO OH OO OH 3CH 3CH 3CH 3CH m/z 352 m/z 351 m/z 336 m/z 336

C22H24O4 C22H23O4 C21H20O4 C21H20O4

3CH CH3

- CH O -C3H4O C

O OH m/z 281 C H O 18 16 3

3CH CH3 3CH CH3 CH3 C CH3 CH2 CH CH CH 3 3 3 O - -CH -C H C O -H O - 3 O - 3 4 C C

OO OH 3CH OO OH OO OH OO OH CH CH CH m/z 296 3 3 3 C H O m/z 352 m/z 351 m/z 336 18 16 4 C H O C H O C H O 22 24 4 22 23 4 21 20 4

3CH CH3 3CH CH3 - CH2 - CH3 CH2 O C

O -H O -C5H9

OO OH CH OO OH OO OH 3 m/z 281 3CH 3CH m/z 352 m/z 351 C17H14O4 C H O C H O 22 24 4 22 23 4 Figure 47: Fragmentation Scheme for Pongagallone A

257 O O O - -H -C8H6O2 O

3CH 3CH - 3CH - O OH O C OH O CH O O O O O m/z 134 m/z 268 m/z 267 m/z 267 C8H5O2 C H O C H O C H O 16 12 4 16 11 4 16 11 4 Figure 48: Fragmentation Scheme for 3-hydroxy-6-methoxyflavone

OH OH OH

- - O O O O O OC O OC -C H O OH OH -H OH OH 6 11 4 OH

OH OH OH OH OH OH OH OH OH OH m/z 436 m/z 435 m/z 288 C H O C H O C15H12O6 21 24 10 21 23 10 Figure 49: Fragmentation Scheme for Catechin-7-O-rhamnoside

OH OH OH

OH OH - O -H C O -C6H11O5 OH O O - 3CH 3CH C CH O O O OH OH 3CH OH OH O m/z 255 OH OH OH OH C H O m/z 420 m/z 419 15 12 4 C H O C H O 21 24 9 21 23 9 Figure 50: Fragmentation Scheme for 4'-Methoxy-3,3',5-stilbenetriol 3-Glucoside

258 O O O CH3 CH3 CH3 - OH OH OH O O C -C H O - O -H O 7 6 O

CH3 CH3 CH3 O O O OH O OH O OH O m/z 420 m/z 419 m/z 329 C H O C H O C H O 24 20 7 24 19 7 17 13 7 Figure 51: Fragmentation Scheme for 3'-Benzyloxy-5,7-dihydroxy-3,4'-dimethoxyflavone

OH OH CH CH 3CH CH3 3CH CH3 3 3 OH OH C OH O O O O O O

-H -HO - - C C O O O O O O CH 3CH 3CH 3 CH 3CH 3CH 3 m/z 420 m/z 419 m/z 401 C H O C H O C H O 25 24 6 25 23 6 25 22 5 Figure 52: Fragmentation Scheme for Isopomiferin

259 - - OH O O O O O

3CH OH CH OH C OH O 3 -H O -CH3O OH O CH 3 OH O CH3 OH O CH3 O O O O O O CH 3 CH3 CH3 m/z 360 m/z 359 m/z 329 C H O 18 16 8 C18H15O8 C17H12O7

- - O O O C

C OH C OH -C4H2O2 O CH 3 OH O CH3 O O O O CH 3 CH3 m/z 245 m/z 329 C13H10O5 C H O 17 12 7 Figure 53: Fragmentation Scheme for 3',5,7-Trihydroxy-4',5',6-trimethoxyisoflavone (a.k.a. Irigenin)

- - OH O O OH OH OH O - C C -C H O OH O -H OH O -HO OH O 6 3 OH OH OH O OH OH OH OH OH OH m/z 209

OH O OH O OH O C9H5O6 m/z 318 m/z 317 m/z 300 C H O 15 10 8 C15H9O8 C15H8O7

OH O - C -HO OH C OH O m/z 191 C H O 9 4 5 Figure 54: Fragmentation Scheme for 3,3',4',5,6,7-Hexahydroxyflavone (a.k.a. Quercetagetin)

260 OH OH OH OH O O O CH3 CH3 CH3 O CH3 -H -C7H6O2 OH O OH O OH O - OH OH OH O OH - - OH CH OH CH2 OH OH OH OH OH m/z 197 m/z 320 m/z 319 m/z 319 C9H9O5 C H O C H O C H O 16 16 7 16 15 7 16 15 7 Figure 55: Fragmentation Scheme for Oureatacatechin

OH OH OH OH OH OH

- -CH - O O -H O OC 3 O OC 3CH 3CH

OH OH OH OH O OH O OH O m/z 316 m/z 315 m/z 300 C H O C H O C H O 16 12 7 16 11 7 15 8 7 Figure 56: Fragmentation Scheme for 7-methylquercetin (a.k.a. Rhamnetin)

OH OH

OH

- O -H C O -C6H11O5 - O O C CH O OH OH OH OH OH OH OH m/z 227 OH OH OH OH C H O m/z 390 m/z 389 14 10 3 C H O C H O 20 22 8 20 21 8 Figure 57: Fragmentation Scheme for Resveratrol-3-b-mono-D-glucoside

261 O O - - -H - -CHO2 C CH -C2H2 C C OH C OH 3CH 3CH 3CH 3CH O O O O m/z 133 m/z 105 m/z 178 m/z 177 C9H8O C7H6O C H O C H O 10 10 3 10 9 3 Figure 58: Fragmentation Scheme for 4-Methoxycinnamic Acid

O O O O O O C O -H O -CHO 2 -C2H3 OH OH OH -CH2 CH CH - - - 2 2 CH CH CH2 m/z 133 m/z 105 m/z 178 m/z 177 m/z 177 C9H8O C7H5O C H O C H O C H O 10 10 3 10 9 3 10 9 3 Figure 59: Fragmentation Scheme for Chroman-2-carboxylic Acid

OH OH OH OH

OH O OH O OH OH O O -H -C11H18O9 OH O OH O OH O - OH O O O - O OO OO OH OH OH O m/z 301 OH OH C H O OH OH OH OH 15 9 7 OH OH m/z 596 m/z 595 C H O C H O 26 28 16 26 27 16 Figure 60: Fragmentation Scheme for Quercetin-3-arabinoglucoside (a.k.a. Peltatoside)

262 CH3 CH3 OH OH O OH O OH OH O OH OH - OH O OH O O

OH O -H OH O - -C12H20O9 OH O OH OH O O OH O O O

OH OH O OH O m/z 285 C15H9O6 OH OH m/z 594 m/z 593 C H O C H O 27 30 15 27 29 15 Figure 61: Fragmentation Scheme for Kaempferol-7-rutinoside

OH OH OH

OH O CH3 OH O CH3 O O OH O CH -C H O 3 O OH -H O OH 12 20 9 O O O O O O CH3 O CH3 - OH O OH OH O OH O

OH OH OH OH OH OH OH O - OH O m/z 315 m/z 624 m/z 623 C16H11O7

C28H32O16 C28H31O16

OH

OH O CH3 O -HO

- C O O m/z 300 C H O 16 10 6 Figure 62: Fragmentation Scheme for Isorhamnetin-3-O-rutinoside

263 CH CH - 3 2 3CH CH

- CH2 OH

CH CH 2 2 OH O OH -H OH -C9H14 CH 3 CH3 OH O OH O

OH O m/z 285

C16H13O5 OH O OH O m/z 408 m/z 407

C25H28O5 C H O 25 27 5 Figure 63: Fragmentation Scheme for Remangiflavanone

3CH CH3 3CH CH3

- C

- OH CH2 3CH 3CH OH O -H -C9H14 OH OH

OH O OH O OH O m/z 285

C16H13O5 OH O OH O m/z 408 m/z 407 C H O C H O 25 28 5 25 27 5 Figure 64: Fragmentation Scheme for Sophora-flavanone

264 - - O O O C O O C O -H -CH3 3CH 3CH

OH OH OH OH O OH O OH O m/z 284 m/z 283 m/z 268 C H O C H O C H O 16 12 5 16 11 5 15 8 5 Figure 65: Fragmentation Scheme for 5,6-Dihydroxy-7-methoxyflavone (a.k.a. Negletein)

OOH OOH OOH

-H -CH3

3CH CH3 3CH - CH3 3CH - O O O C O O C O m/z 284 m/z 283 m/z 268 C H O C H O C H O 17 16 4 17 15 4 16 12 4 Figure 66: Fragmentation Scheme for 2'-Hydroxy-4',6'-dimethoxychalcone

CH3 CH3 O O O O O O

-H -CH3 - - C C

CH3 CH3 CH3 OH HOO OH HOO OH HOO m/z 300 m/z 299 m/z 284 C16H12O5 C17H16O5 C17H15O5

O O O O

- -C3H4O - C C CH

CH3 OH OH CH OH HOO m/z 227 m/z 283 C H O C H O 13 8 4 16 12 5 Figure 67: Fragmentation Scheme for 2',4-Dihydroxy-4',6'-dimethoxychalcone

265 CH3 CH3 CH3 O O O OH OH OH

-H - -CH3O - OH O OH OC OH OC

C O O

OH O CH3 OH O CH3 OH O m/z 330 m/z 329 m/z 299 C H O C H O C H O 17 14 7 17 13 7 16 10 6

CH3 CH3 CH3 O O O OH OH - - OH C C -H -CH3 OH O OH O OH O C

O O O

OH O CH3 OH O CH3 OH O m/z 330 m/z 329 m/z 314 C H O C H O C H O 17 14 7 17 13 7 16 10 7 Figure 68: Fragmentation Scheme for Isorhamnetin-3-methoxy

OH OH OH OH

- - - OH O OH OC OH OC OCH -H -CH3 C -C2HO2

O O O O O O O

CH3 OH O CH3 CH3 OH O CH3 OH O CH3 O O CH3 m/z 330 m/z 329 m/z 314 m/z 257 C H O C H O C H O C H O 17 14 7 17 13 7 16 10 7 14 9 5 Figure 69: Fragmentation Scheme for Irisflavone A

266 O O O CH3 CH3 CH3

O O O O - O O - -H -CH3 3CH 3CH C 3CH C

CH3 CH3 O O O OH O OH O OH O m/z 330 m/z 329 m/z 314 C H O C H O C H O 18 18 6 18 17 6 17 14 6

O O O CH3 CH3 CH3 O O O O O O -H -CH2O 3CH 3CH 3CH - - CH3 C CH3 CH O O OH O OH O OH O m/z 330 m/z 329 m/z 299 C H O C H O C H O 18 18 6 18 17 6 17 15 5 Figure 70: Fragmentation Scheme for Kaempferol-3,7,4’-Trimethylether

- - O OH O O O O

O O O CH 3 -H CH3 -CH3 C CH3

OO OO OO CH OH CH 3 3 CH3 OH CH3 CH3 OH m/z 330 m/z 329 m/z 314 C H O 18 18 6 C18H17O6 C17H14O6

- - O O O O

O O -CH3 -CH3

OOC OO

OH CH3 OH m/z 285 m/z 299 C H O C H O 15 8 6 16 11 6 Figure 71: Fragmentation Scheme for 3',6'-Dihydroxy-2',4',5'-trimethoxychalcone

267 OH OH OH OH OH OH OH OH OH OH OH O O O O -H -CH3O OH O O HOO O HOO O HOO O

- O O C C - C CH3 CH3 m/z 339 m/z 370 m/z 369 C15H14O9 C H O C H O 16 18 10 16 17 10

OH OH - OH OH O OH - HOO C O - OH OH HOO C O O O O O -H -C6H10O6 -CH3 HOO O HOO O O O CH CH3 m/z 176 m/z 191 O O C9H4O4 C10H7O4 CH3 CH3 m/z 370 m/z 369 C H O C H O 16 18 10 16 17 10 Figure 72: Fragmentation Scheme for Fraxetin-8-glucoside (a.k.a. Fraxin)

268 CH3 CH3 CH3 CH3 OH O OH O O OH CH3 OH CH3 C OH CH3 OH CH3 -H -HO -C9H3O3 - C - - OH O O O OO OH OH OH OH m/z 354 m/z 353 m/z 335 m/z 175

C20H18O6 C20H17O6 C20H16O5 C11H13O2

CH3

OH CH 3 -HO - C C

m/z 161

C11H12O

CH3 CH3 OH O OH O OH O OH OH CH3 OH CH3 -H -C4H7 CH2

OH OC- OH O OH O - C OH OH C OH m/z 297 m/z 354 m/z 353 C16H10O6 C H O C H O 20 18 6 20 17 6

Figure 73: Fragmentation Scheme for Lico-iso-flavone A

269 O CH3 O CH3 O C CH3 O -H O -CH3 O - - O C O C O CH3 CH3 O O O m/z 282 m/z 281 m/z 266 C H O C H O C H O 17 14 4 17 13 4 16 10 4

- O CH OC CH 3 3 -C H O - O -H O 9 4 2 C CH3 O O O CH3 CH3 O CH3 O O m/z 137 m/z 282 m/z 281 C H O C H O C H O 8 9 2 17 14 4 17 13 4 Figure 74: Fragmentation Scheme for 2',3'-Dimethoxyflavone

270 OH OH OH OH OH C OH O -H -HO -HO - OH O OH O - OH O - C OH OH C OH C OH OH OH OH OH OH O OH O OH O OH O m/z 285 m/z 320 m/z 319 m/z 302 C15H9O6 C H O C H O C H O 15 12 8 15 11 8 15 10 7 Figure 75: Fragmentation Scheme for 3,3',4',5,5',7-Hexahydroxyflavanone (a.k.a. Dihydromyricetin)

CH3 OH CH3 OH CH3 OH CH3 OH CH3 OH O O O O O CH3 CH3 CH3 CH3 - - - -H -HO -C7H4O2 O O O OC O OC O C O - C

OH OH C OH C OH C OH OH O OH O O O O m/z 330 m/z 329 m/z 312 m/z 312 m/z 193 C H O C H O C H O C H O C H O 17 14 7 17 13 7 17 12 6 17 12 6 10 8 4 Figure 76: Fragmentation Scheme for Irisflavone D

OH O OH O OH

-H - -CO2 - OH CH OH CH2

m/z 166 m/z 165 m/z 121 C H O C H O C H O 9 10 3 9 9 3 8 9 Figure 77: Fragmentation Scheme for o-Hydrocoumaric Acid

271

OH OH O O -H -C6H10O5 OH OOCH3 HOH OOCH3 H - O CH3 OH - OH OH OH O m/z 165 m/z 328 m/z 327 C10H13O2 C H O C H O 16 24 7 16 23 7 Figure 78: Fragmentation Scheme for Betuloside

OH OH OH - OH C O - OH O OH C O - -CH O OH C O -C H O OH -H OH 3 OH 6 5 2 CH O O OH OH CH3 OH CH3 OH m/z 163 m/z 304 m/z 303 m/z 271 C9H7O3 C H O C H O C H O 16 16 6 16 15 6 15 12 5 Figure 79: Fragmentation Scheme for 3'-O-Methylcatechin

OH OH OH OH OH OH OH O - CH -H -HO -C6H4O2 OH O OH O - OH O - C C C OH

OH OH C OH O m/z 179

OH O OH O O C9H6O4 m/z 304 m/z 303 m/z 287 C H O C H O C H O 15 12 7 15 11 7 15 10 6 Figure 80: Fragmentation Scheme for Dihydroquercetin (a.k.a. Taxifolin)

272 OH OH C OH C OH OH OH

OH O OH O OH O -H OH O -HO -CH2O - - - C CH C O O O OH CH3 OH OH CH3 OH CH3 m/z 285 m/z 256 m/z 304 m/z 303 C H O C16H14O5 15 12 4 C H O C H O 16 16 6 16 15 6 Figure 81: Fragmentation Scheme for 3-Methoxy-3',4',5,7-flavantetrol (a.k.a. Meciadanol)

OH OH OH OH

OH O OH O OH O OH O -H -HO

- - OH C OH C C OH CH OH OH OH OH OH OH OH m/z 288 m/z 287 m/z 269 m/z 269

C15H12O6 C15H11O6 C15H10O5 C15H10O5

OH

OH O C -C2H2

OH

OH m/z 243 C H O 13 8 5 Figure 82: Fragmentation Scheme for Dihydrokaempferol

273 OOH OOH

- - C C

CH3 - CH3 CH O CH O O O O C O 3 3 O CH3 O CH3 -H -C10H6O4 -HO 3CH O 3CH O OH O O O CH3 O CH3 C OH O OH O

OH O OH O m/z 389 m/z 372

OH O OH O C23H17O6 C23H16O5 m/z 580 m/z 579 C H O C H O 33 24 10 33 23 10 Figure 83: Fragmentation Scheme for Amentoflavone-4',4'',7-trimethyl Ether (a.k.a. Sciatopitysin)

OH - OH - - OH OH C OH C C OH C OH O OH O OH O OH O OH -H OH -HO OH -HO

OH OH OH OH OH O OH O OH O OH O m/z 318 m/z 317 m/z 300 m/z 283 C H O C H O C H O C H O 15 10 8 15 9 8 15 8 7 15 7 6 Figure 84: Fragmentation Scheme for 3,3',4',5,7,8-Hexahydroxyflavone (a.k.a. Gossypetin)

O - O - O OH CH3 OH C CH3 OH C CH3

O O CH3 O O CH3 O O -H -CH3 3CH O 3CH O 3CH O

OH O OH O OH O m/z 344 m/z 343 m/z 329 C H O C H O C H O 18 16 7 18 15 7 17 12 7 Figure 85: Fragmentation Scheme for Gossypetin 3,3',4',7-tetramethyl Ether

274 O O O OH CH3 OH CH3 CH3 O O CH O O CH O C O CH 3 -H 3 -HO 3 3CH O 3CH O 3CH O - - C C

OH O OH O OH O m/z 344 m/z 343 m/z 327 C H O C H O C H O 18 16 7 18 15 7 18 14 6 O O O OH CH3 OH CH3 OH CH3

O O CH3 O O CH3 O O C -H -CH3O 3CH O 3CH O 3CH - - C C

OH O OH O OH O m/z 344 m/z 343 m/z 313 C H O C H O C H O 18 16 7 18 15 7 17 12 6 Figure 85 (cont.): Fragmentation Scheme for Gossypetin 3,3',4',7-tetramethyl Ether

OH OH OH

OH O OH O OH O -H - - 3CH OH 3CH CH OH 3CH CH2 O

CH3 OH O CH3 OH O CH3 OH O m/z 356 m/z 355 m/z 355

C20H20O6 C20H19O6 C20H19O6

OH C - 3CH CH2 -C7H5O3

CH3 OH O m/z 219 C H O 13 14 3 Figure 86: Fragmentation Scheme for Cudraflavanone B

275 OH OH OH OH OH O OOO O OOO - -H -C6H10O5 O O O

- OH OH OH O

OH OH m/z 177 m/z 340 m/z 339 C9H5O4 C H O C H O 15 16 9 15 15 9 Figure 87: Fragmentation Scheme for 7,8-Dihydroxycoumarin-7-b-D-Glucoside

276 OH OH

OH O CH3 OH O CH3 OH O O

-C H O OH O CH3 O -H O 6 10 5 O - OH O OH OH O O - O O O OH OH OH O OH OH m/z 315 C H O OH OH 16 11 7 m/z 478 m/z 477 C H O C H O 22 22 12 22 21 12

Figure 88: Fragmentation Scheme for Isorhamnetin-3-glucoside

OH OH OH OH OH OH - CH O O O OH OH - -C H O OH O O -H C O 6 10 5 OH

OH O OH O OH OH OH OH O m/z 301 OH OH C15H9O7 OH O OH O m/z 464 m/z 463 C H O C H O 21 20 12 21 19 12

Figure 89: Fragmentation Scheme for Quercetin-4'-O-glucoside_(a.k.a. Spiraeoside)

277 OH OH OH OH OH OH OH OH O O OH OH OH -C H O O O -H O 9 4 5 OH

OH O OH O O OH OH - C - OH O OH m/z 271

OH O OH O C12H15O7 m/z 464 m/z 463 C H O C H O 21 20 12 21 19 12

Figure 89 (cont.): Fragmentation Scheme for Quercetin-4'-O-glucoside (a.k.a. Spiraeoside)

278 CH3 CH3 CH3 CH3 CH3

O - O - O - O - O C C C C

-H -HO -HO -CH3 HOO CH3 HOO CH3 HOO CH3 HOO CH3 HOO O O O O O CH OH OH OH OH OH OH OH OH OH OH OH OH OH OH OH m/z 350 m/z 349 m/z 331 m/z 315 m/z 300 C H O C H O C H O C H O C H O 17 18 8 17 17 8 17 16 7 17 15 6 16 12 6

CH3 CH3 CH3 CH3

O O O O - HOO C CH -H - -HO - -HO - -C8H9O2 HOO CH3 HOO C CH3 HOO C CH3 HOO C CH3 O O O O OH OH OH OH OH OH OH C OH OH m/z 179

OH OH OH OH OH OH C9H6O4 m/z 350 m/z 349 m/z 331 m/z 315 C H O C H O C H O C H O 17 18 8 17 17 8 17 16 7 17 15 6 Figure 90: Fragmentation Scheme for 2-(3,4-dimethoxyphenyl)-3,4-dihydro-2H-chromene-3,4,5,6,7-pentol

CH3 CH3 CH3

O O O

-H -CH3O OO CH3 O O CH3 O CH3 3CH O 3CH O C O

- - OH OH OH C OH OH C OH OH OH OH OH OH OH m/z 364 m/z 363 m/z 331 C H O C H O C H O 18 20 8 18 19 8 17 16 7 Figure 91: Fragmentation Scheme for 2-(3,4-dimethoxyphenyl)-7-methoxy-3,4-dihydro-2H-chromene-3,4,5,6-tetrol

279 OH OH OH - OH OC - - OH O OH OC OH OC OH OH OH O O O CH O OH OH H -C H O OH H -H H -HO H 6 5 2 OH OH OH O OH O OH O OH O OH OH OH OH OH OH OH OH OH OH OH OH OH m/z 449

m/z 576 m/z 575 m/z 557 C24H17O9 C H O C H O C H O 30 24 12 30 23 12 30 22 11

OH OH OH OH

- - - OH O OH OC OH OC OH OC OH OH OH OH O O CH O O OH OH H -H H -HO H H OH OH OH OH C

O OH O OH O OH O OH OH OH OH OH

OH OH OH OH OH OH OH OH m/z 576 m/z 575 m/z 557 m/z 557

C30H24O12 C30H23O12 C30H22O11 C30H22O11

- O

C

O OH -C15H10O5 OH

OH OH m/z 287 C H O 15 12 6 Figure 92: Fragmentation Scheme for Proanthocyanidin A2 Dimer

280 OH OH OH

O O CH O O CH O O C 3 3 -CH O O -H O 3 3CH 3CH 3CH - - C C OH OH OH OH O OH O OH O m/z 332 m/z 331 m/z 299 C H O C H O C H O 17 16 7 17 15 7 16 12 6 Figure 93: Fragmentation scheme for 3,5-dihydroxy-2-(4-hydroxy-3-methoxyphenyl)-7-methoxy-2,3-dihydro-4H-chromen-4-one

281 OH OH OH OH OH O OH O OH O OH OH OH OH O O O O OH OH OH OH O H -H H H -C8H7O3 OH OH OH OH H OH C O OH O OH O OH - - OH OH CH OH CH2 O OH - CH2 OH OH O m/z 423 OH OH OH C H O m/z 576 m/z 575 m/z 575 22 16 9 C H O C H O C H O 30 24 12 30 23 12 30 23 12

OH OH OH OH O OH O OH O OH OH C OH O O O OH O OH OH OH O H H -C8H8O4 -C6H5O2 H OH OH OH OH H - OH C O OH O O - O C O OH - OH -CH2 CH C CH 2 CH2 m/z 298 OH OH m/z 407 C16H10O6 OH OH C H O m/z 575 m/z 575 22 15 8 C H O C H O 30 23 12 30 23 12 Figure 94: Fragmentation Scheme for Proanthocyanidin A5' Dimer

282 OH OH OH

- - - OH OC OH OC OH OC OH OH OH OH O CH O O OH - OH OC H -HO H H -C15H12O5 OH OH OH C OH O O OH O OH O OH H OH OH OH OH m/z 285

C15H10O6 OH OH OH OH OH OH m/z 575 m/z 557 m/z 557 C H O C H O C H O 30 23 12 30 22 11 30 22 11 Figure 94 (cont.): Fragmentation Scheme for Proanthocyanidin A5' Dimer

OH OH

OH O OH O OH OH OH OH OH -C H O OH O OH OH -H OH OH 15 12 6 OH OH OH OH O OH O - CH OH - OH C OH OH m/z 289 C15H13O6 OH OH m/z 578 m/z 577 C H O C H O 30 26 12 30 25 12 Figure 95: Fragmentation Scheme for Procyanidin B2 Dimer

283 OH OH OH - OH OC - - OH O OH OC OH OC OH OH OH

C OH -C H O OH OH -H OH OH -HO OH 6 5 2 OH OH O OH OH OH OH OH O OH O OH O OH OH OH

OH OH OH OH OH OH OH OH m/z 447 m/z 574 m/z 573 m/z 555 C24H15O9 C H O C H O C H O 30 22 12 30 21 12 30 20 11

OH OH OH

OH O OH O OH O OH OH OH OH

C OH OC OH OH -H OH OH -HO OH -C15H10O5 OH OH OH OH - OH O OH O OH O C OH OH OH OH

- - C C OH OH OH OH m/z 285 C H O OH OH OH 15 10 6 m/z 574 m/z 573 m/z 555 C H O C H O C H O 30 22 12 30 21 12 30 20 11 Figure 96: Fragmentation Scheme for Procyanidin B-type Dimer

3CH CH2 3CH CH2 HOO H CH2 3CH HOO H3CH HOO H OH O -C H -H 9 15 C CH3 OH O CH3 OH O - C - C OH O OH O OH O m/z 299 m/z 424 m/z 423 C16H12O6 C H O C H O 25 28 6 25 27 6 Figure 97: Fragmentation Scheme for Sophoraflavanone G

284 3CH CH2 3CH CH2 3CH CH2 HOO H

3CH HOO H3CH - HOO H3CH - HOO H CH CH - -C H OC -H -HO 10 16 C CH3 OH O CH3 OH O CH3 O C

OH O OH O OH O OH O m/z 271 m/z 424 m/z 423 m/z 405 C15H10O5 C H O C H O C H O 25 28 6 25 27 6 25 26 5 Figure 97 (cont.): Fragmentation Scheme for Sophoraflavanone G

OH OH

OH O OH O OH

OH O O -H O -C16H12O7 CH3 CH3 - OH O OH O CH OH O OH O OH OH - OH O C OH OH m/z 270 C15H11O5 OH O OH O m/z 588 m/z 587

C31H24O12 C31H23O12 OH OH OH OH

OH O OH O O

- - - - C CH CH C O O -C7H3O3 O -HO O CH3 CH3 CH3 CH3 OH O OO OH O OH O OH O OH O OH OH OH OH

OH OH OH OH OH O OH O OH O OH O m/z 587 m/z 587 m/z 451 m/z 435 C H O C H O C H O C H O 31 23 12 31 23 12 24 20 9 24 19 8 Figure 98: Fragmentation Scheme for Kolaflavanone

285 OH OH

OH O OH O OH OH

OH O OH O O -H O -C6H10O6 -HO - - - OH O OH OH O O C C O O C

OH OH OH O O OH OH m/z 268 m/z 251 C H O C H O OH OH 15 9 5 15 8 4 m/z 448 m/z 447 C H O C H O 21 20 11 21 19 11

Figure 99: Fragmentation Scheme for 3,4',5,7-Tetrahydroxyflavone-3-glucoside (a.k.a. Astragalin)

286 OH OH OH OH OH OH OH OH

-CH O OH O -H OH O -HO OH O 3 OH O

- - - 3CH 3CH C 3CH C C O O O C OH O OH O O O m/z 316 m/z 315 m/z 297 m/z 269 C H O C H O C H O C H O 16 12 7 16 11 7 16 10 6 15 7 5 Figure 100: Fragmentation Scheme for 6-Methoxyluteolin (a.k.a. Nepetin)

287 OH OH

OH OH OH OH O OH O OH OH OH -H -C15H12O6 OH OH OH OH OH O OH OH OH O OH O - CH OH - OH C OH OH m/z 289 C15H13O6 OH OH m/z 578 m/z 577 C H O C H O 30 26 12 30 25 12 Figure 101: Fragmentation Scheme for Procyanidin B-type Dimer (Linear)

OH OH OH OH OH OH O OH O OH OH OH

-H -C15H12O6 OH O OH OH

- OH OH C OH O OH O OH OH OH - OH C m/z 289 C H O OH OH OH OH 15 13 6 m/z 578 m/z 577 C H O C H O 30 26 12 30 25 12 Figure 102: Fragmentation Scheme for Procyanidin B-type Dimer (Branched)

288 OH OH OH OH

OH O OH O OH C OH

O OH OH OH OH - OCH OH O -H OH O -C15H9O6 O O

- O OCH

OH m/z 269

C15H10O5 OH OH m/z 556 m/z 555 C H O C H O 30 20 11 30 19 11 OH OH

OH - OH C OH O OH O OH

- OH OH OH OH OH C OH O OH O -C H O -H 15 11 5 OH O O O C

O O

OH O m/z 283

C15H8O6

OH OH m/z 556 m/z 555 C H O C H O 30 20 11 30 19 11 Figure 103: Fragmentation Scheme for Hegoflavone A

289 OH OH

OH O OH O OH OH OH OH O OH -H OH -C15H8O6 OH O OH O - CH OH O OH O

- OH O C m/z 271

C15H11O5 OH O OH O m/z 556 m/z 555 C H O C H O 30 20 11 30 19 11

OH OH

OH O OH O OH OH OH OH -C H O OH -H OH 15 11 5 OH C O OH O OH O - OH O OH O C

- C OH O m/z 283

OH O OH O C15H8O6 m/z 556 m/z 555 C H O C H O 30 20 11 30 19 11 Figure 104: Fragmentation Scheme for Morelloflavone

290 OH OH

OH O OH O OH OH

O O OH OH OH H H - OH OH C

O OH O OH O OH -H OH OH OH OH -C30H22O12

- OH O OH OH OH OH O OH O OH OH m/z 289

C15H13O6 OH OH OH OH

OH OH m/z 864 m/z 863 C H O C H O 45 36 18 45 35 18 Figure 105: Fragmentation Scheme for Catechin/Epicatechin Trimer with Single A-type Interflavanic Linkage (IFL)

OH OH OH OH OH OH OH OH

OH O OH O OH O OH O OH OH OH -H -HO -C6H4O2 OH OH OH OH O OH O OH O OH O OH O OH O - O - O - C C C C CH

OH OH OH OH OH O OH O OH O OH O m/z 590 m/z 589 m/z 571 m/z 463 C H O C H O C H O C H O 30 22 13 30 21 13 30 20 12 24 16 10 Figure 106: Fragmentation Scheme for Manniflavanone

291 OH OH

OH OH OH OH - - OH C O OH C O OH OH OH OH - - -HO -C15H9O5 -HO OH C OH OH OC OH OC OH O O OH O OH O OH C OH OH O O OH OH m/z 303 m/z 285 C15H11O7 C15H10O6 OH O OH O m/z 589 m/z 571 C H O C H O 30 21 13 30 20 12 Figure 106 (cont.): Fragmentation Scheme for Manniflavanone

OH OH OH

OH O OH O OH O

- - C C O -H O -C7H7O2 CH3 CH3 OH O OH O OH O OH O OH O OH O OH OH CH

OH OH OH OH O OH O OH O m/z 588 m/z 587 m/z 464 C H O C H O C24H16O10 31 24 12 31 23 12 OH OH

OH O OH O OH

OH O O -H O -C16H12O7 CH3 CH3 - OH O OH O CH OH O OH O OH OH - OH O C OH OH m/z 271 C15H11O5 OH O OH O m/z 588 m/z 587 C H O C H O 31 24 12 31 23 12 Figure 107: Fragmentation Scheme for Kolaflavanone

292 OH OH OH

OH OH OH

OH O OH O OH O

O O O OH OH OH OH OH OH -H -C8H7O3 O OH O OH O OH OH OH OH OH OH

OH OH OH OH O O OH O O OH O O OH OH C - - OH CH OH CH2 OH OH OH m/z 862 m/z 861 m/z 709 C H O C H O C H O 45 34 18 45 33 18 37 26 15

OH OH OH OH

OH OH OH OH

OH O OH O OH O OH O

O O O O OH OH OH OH OH OH OH OH -C8H8O4 -C6H5O2

O OH O OH O OH O OH OH OH OH C OH OH

OH OH OH OH OH O O OH O O OH O - O - OH OH C C - - C CH OH CH2 OH CH2 CH2 OH O O O m/z 861 m/z 861 m/z 694 m/z 584 C H O C H O C H O C H O 45 33 18 45 33 18 37 25 14 31 20 12

Figure 108: Fragmentation Scheme for Aesculitannin C

293 OH OH

OH OH

- OH O OH OC OH O O OH OH OH OH OH -H - OH OC O OH O OH -C30H23O12 OH OH O OH OH

OH OH OH OH O O OH O O m/z 285 OH OH C15H10O6

OH OH OH OH m/z 862 m/z 861 C H O C H O 45 34 18 45 33 18 Figure 108 (cont.): Fragmentation Scheme for Aesculitannin C

OH OH OH OH OH C OH O -H -HO -HO - OH O OH O - OH O - C OH OH C OH C OH OH OH OH OH OH O OH O OH O OH O m/z 286 m/z 320 m/z 319 m/z 301 C15H9O6 C H O C H O C H O 15 12 8 15 11 8 15 10 7 OH OH OH OH OH OH OH OH -C H O OH O -H OH O OH O 7 3 3 OH OH OH O - - OH C CH - OH OH OH CH OH OH O OH O O O m/z 183 m/z 320 m/z 319 m/z 319 C8H8O5 C H O C H O C H O 15 12 8 15 11 8 15 11 8 Figure 109: Fragmentation Scheme for 3,3',4',5,5',7-Hexahydroxyflavanone (a.k.a. Dihydromyricetin)

294 OH OH

OH O OH O OH OH O O OH OH OH H H OH OH OH O O OH O OH OH OH OH OH OH O OH -H H OH OH OH OH OH -C30H24O12 O OH O OH O OH OH - OH C - OH CH

OH OH OH OH OH O O OH OH OH OH OH m/z 575

C30H23O12 OH OH OH OH OH OH m/z 1152 m/z 1151

C60H48O24 C60H47O24 OH OH

OH O OH O OH OH O O OH OH H H OH OH OH

O OH O OH - OH OH C OH OH -H O OH OH OH -C45H34O18 OH OH - O OH O O OH OH OH OH OH OH OH m/z 289 OH OH O O C15H13O6 OH OH OH OH

OH OH OH OH OH OH m/z 1152 m/z 1151 C H O C H O 60 48 24 60 47 24 Figure 110: Fragmentation Scheme for Catechin/Epicatechin Tetramer with Single A-type Interflavanic Linkage (IFL)

295 Table 19: Biological Activities of Identified (and Tentatively Identified) Compounds Found in Peanut Skin Fractions Compound Compound Name Biological Activity Reference Number Cinnamic Acids (including Methoxy- and Other Derivatives) 1-A1, 19-A1, 18- Caffeic Acid Anticancer Activity Kazumi et al., 1998 B2 Antiviral Activity Grunberger et al., 1988 Anti-Inflammatory Activity Grunberger et al., 1988 Immunomodulatory Properties Grunberger et al., 1988 Antimitogenic Activity Natarajan et al., 1996 1-A1, 2-A1, 8-B1, Cinnamic Acids Antibacterial Properties Tonari et al., 2002 1-B2, 3-B2, 15- Antifungal Activity Bisogno et al., 2007 B2, 5-C1, 6-C1, Inhibition of LDL Oxidation Lee et al., 2004 14-C1 8-B1, 1-C2 p-Coumaric Acid Antioxidant Kristinová et al., 2009 10-B1 Ferulic Acid Antioxidant Kristinová et al., 2009 Anti-Inflammatory Activity Graf, 1992 Carboxylic Acids (including Methoxy- and Other Derivatives) 14-C1 Chroman-2- Antioxidant Tang and Liu, 2007 Carboxylic Acid 5-C2 Monotropein Anti-Inflammatory Activity Jongwon et al., 2005 Analgesic Jongwon et al., 2005 Benzoic Acids (including Methoxy- and Other Derivatives) 1-B1 Protocatechuic Acid Antioxidant Isanga and Zhang, 2007 Inhibition of LDL Oxidative Modification Yen and Hsieh, 2002 1-B1 Resorcylic Acid -like Properties Katzenellenbogan et al., 1979 2-B1 Salicylic Acid Anti-Inflammatory Higgs et al., 1987 Preventative Cancer Activity Paterson and Lawrence, 2001 Naphthoquinones (including Methoxy- and Other Derivatives) 27-A1, 5-A2 Plumbagin Apoptosis Activity Against Breast Cancer Cells Ahmad et al., 2008 Inhibition of Cancer Migration and Invasion Shih et al., 2009 Antimicrobial Activity Against Tuberculosis and Kuete et al., 2009 Gonorrhoea

296

Compound Compound Name Biological Activity Reference Number Stilbenes (including Methoxy- and Other Derivatives) 26-A1, 28-A1, 4- Oxyresveratrol Anti-Herpes Simplex Virus Activity Chuanasa et al., 2008 A2 22-A1, 11-C1, 5- Resveratrol Antioxidant Olas and Wachowicz, 2002 C2, 19-D1, 1-D2 Inhibition of Cardiovascular Disease Kris-Etherton et al., 2002 Cancer Preventative Activity Francisco and Resurreccion, 2008 Inhibition of LDL Oxidativion Frankel et al., 1993 Estrogenic Activity Garvin et al., 2006 Anti-Inflammatory Rotondo et al., 1998 4-C1 Stilbenes Antioxidant Francisco and Resurreccion, 2008 Anticancer Properties Francisco and Resurreccion, 2008 Antimutagen Properties (Detoxify Carcinogens) Francisco and Resurreccion, 2008 Anti-inflammatory Activity Cassidy et al., 2000 Chalcones (including Methoxy- and Other Derivatives) 2-A1, 18-A1 Methoxy-Chalcone (Melanoma) Cancer Preventative Activity Kayo Henmi et al., 2009 Antibacterial Activity Popova et al., 2001 Antifungal Activity Popova et al., 2001 2-B2 Butein Antioxidant Cheng et al., 1998 Anti-Inflammatory Activity Cheng et al., 1998 Anticancer Activity Samoszuk et al., 2005 Antifibrogenic Activity Cheng et al., 1998 Vaxorelaxant Properties Yu et al., 1995 Antihepatoxic Activity (Liver Damage Woo et al., 2003 Prevention) Coumarins (including Methoxy- and Other Derivatives) 26-A1, 28-A1, 4- Cedrecoumarin A Vasorelaxant Properties Rakotoarison et al., 2003 A1 28-C1 Fraxin or Fraxoside Antioxidant Whang et al., 2005 Anti-Inflammatory Klein-Galczinsky, 1999 Preventative Cancer Activity Ivanovska et al., 1994

297

Compound Compound Name Biological Activity Reference Number 26-B2, 30-C1, 32- Glycycoumarin Antiviral Activity Sekine-Osajima et al., 2009 C1, 33-C1, 33-C2 Antispasmodic Properties Sato et al., 2006 Antibacterial Activity Tanaka et al., 2001 26-A1, 28-A1 Osthole Atherosclerosis Prevention Ogawa et al., 2007 Hepatic Lipid Suppression Ogawa et al., 2007 26-A1, 28-A1 Suberosin Anti-Inflammatory Activity Chen et al., 2009 2-A2 Umbelliferone Antiallergic Properties Vasconcelos et al., 2009 8-A2 Xanthotoxol Sedative Properties Sethi et al., 1992 Dicoumarins (including Methoxy- and Other Derivatives) 34-C1 Daphnoretin Anticancer Activity Liou et al., 2006 5-A1, 34-A1 Dicoumarin Anticoagulant Properties Zhou et al., 2009 4-A1 Dicoumoxyl Anticoagulant Properties Meriel et al., 1964 Coumestrols (including Methoxy- and Other Derivatives) 34-A1 Psoralidin Anticancer Activity Kumar et al., 2009 Secoiridoids (including Methoxy- and Other Derivatives) 31-A1 Oleuropein Smooth Muscle Induction Activity Rocha et al., 2009 Preventive Activity Against UVB Skin Damage Kimura and Sumiyoshi, 2009 Vasodilation Activity Visioli et al., 1998 Antimicrobial Properties Bisignano et al., 1999 Inhibition of LDL Oxidation Visioli and Galli, 1994 Coniferins (including Methoxy- and Other Derivatives) 20-B2 Syringin Glucose Regulator Niu et al., 2008 Hemostatic Properties Wang, 1980 (including , Methoxy- and Other Derivatives) 3-C1, 2-D2 Catechin Inhibition of LDL Oxidation Andrikopolous et al., 2003 Prevention of Plasma Oxidation Lotito and Fraga, 1998 Anticancer Activity Kandil et al., 2002 Antimicrobial Properties Buzzini et al., 2007

298

Compound Compound Name Biological Activity Reference Number 3-D1 Epicatechin Antioxidant Francisco and Resurreccion, 2008 Anticancer Activity Kandil et al., 2002 Antimicrobial Properties Buzzini et al., 2007 Antifungal Activity Narayana et al., 2001 23-B2, 6-C2, 4-D1 3'-O-Methylcatechin Antioxidant Spencer et al., 2001 Inhibition of Cell Death Spencer et al., 2001 Preventative Cancer Activity Ebeler et al., 2002 Inhibition of LDL Oxidation Cren-Olive et al., 2003 23-B2, 6-C2, 4-D1 4'-O-Methylcatechin Preventative Cancer Activity Ebeler et al., 2002 Inhibition of LDL Oxidation Cren-Olive et al., 2003 Isoflavans (including Glucosides, Methoxy- and Other Derivatives) 6-A2 Equol Therapeutic for Hormonal Conditions (Estrogenic Setchell et al., 2009 Activity) Flavones (including Glucosides, Methoxy- and Other Derivatives) 24-B2, 24-C1, 25- Acacetin Anticancer Activity Hsu et al., 2004 C1, 38-D1 5-B1, 7-C2 Apigenin Anti-Hypertension Activity Narayana et al., 2001 Heart Contraction Promotion Itoigawa et al., 1999 Anti-Inflammatory Activity Narayana et al., 2001 Antibacterial Activity Narayana et al., 2001 Antiviral Activity Narayana et al., 2001 Antifungal Activity Narayana et al., 2001 20-B1, 21-B2, 26- Chrysoeriol Antifungal Activity Narayana et al., 2001 C1, 23-C2, 25-C2, 27-C2, 29-C2, 20- D1, 28-D1, 29-D1, 31-D1, 4-Gr2 20-B1 Diosmetin Anticancer Activity Androutsopoulos et al., 2009 6-D2, 7-D2, 7-E1, Fisetin Heart Contraction Promotion Itoigawa et al., 1999 3-E2

299

Compound Compound Name Biological Activity Reference Number 10-B2 Gossypin Anti-Anxiety Properties Fernandez et al., 2009 Sedative Properties Fernandez et al., 2009 Anti-Inflammatory Activity Mada et al., 2009 Analgesic Properties Mada et al., 2009 11-D1, 12-D1, 3- Hyperoside Antioxidant Liu et al., 2005 E1, 5-E1 11-B1, 12-B1, 15- Isohamnetin Antioxidant Yokozawa et al., 2002 B1, 16-B1, 18-B1, Vasodilation Activity Pérez-Vizcaíno et al., 2002 4-B2, 8-B2, 16- Anti-Inflammatory Ahmed et al., 2005 B2, 22-B2, 10-C1, Anticancer Activity Teng et al., 2006 13-C1, 21-C1, 22- C1, 3-C2, 5-D1, 15-D1, 16-D1, 22- D1 19-F1 Kaempferol Antioxidant Francisco and Resurreccion, 2008 Vasodilation Activity Pérez-Vizcaíno et al., 2002 Heart Contraction Promotion Itoigawa et al., 1999 Anti-Inflammatory Activity Narayana et al., 2001 Antiallergic Properties Narayana et al., 2001 Antiviral Activity Narayana et al., 2001 Anticancer Activity Narayana et al., 2001 26-C2, 27-D1, 19- Luteolin Antioxidant Francisco and Resurreccion, 2008 F1, 3-Gr2 Anti-Leukemic Activity Chiang et al., 2003 Anti-Hypertensive Activity Narayana et al., 2001 Vasorelaxant Properties Narayana et al., 2001 Heart Contraction Promotion Itoigawa et al., 1999 Anti-Inflammatory Activity Narayana et al., 2001 Antiviral Activity Narayana et al., 2001

300

Compound Compound Name Biological Activity Reference Number 17-B2, 2-D2 Morin Antioxidant Narayana et al., 2001 Heart Contraction Promotion Itoigawa et al., 1999 Anti-Inflammatory Activity Narayana et al., 2001 Antiallergic Properties Narayana et al., 2001 Antiviral Activity Narayana et al., 2001 Anticancer Activity Narayana et al., 2001 7-C1 Myricetin Antioxidant Francisco and Resurreccion, 2008 Anti-Inflammatory Activity Narayana et al., 2001 Antiallergic Properties Narayana et al., 2001 Antiviral Activity Narayana et al., 2001 Anticancer Activity Narayana et al., 2001 50-A1, 51-A1, 34- Pectolinarigenin Anti-Inflammatory Activity Lim et al., 2008 D1, 36-D1 2-A1 Primuletin Antioxidant Park et al., 2007 15-C2, 19-C2, 8- Quercetin Antioxidant Francisco and Resurreccion, 2008 E1 Anticancer Activity Francisco and Resurreccion, 2008 Vasodilation Activity Pérez-Vizcaíno et al., 2002 Anti-Inflammatory Activity Kobuchi et al., 1999 Heart Contraction Promotion Itoigawa et al., 1999 Antiallergic Properties Narayana et al., 2001 Antiviral Activity Narayana et al., 2001 4-B2, 10-C1, 13- Rhamnetin Antioxidant Ratty, 1988 C1, 21-C1, 22-C1, 3-C2, 15-D1, 22- D1, 30-D1, 1-F2, 2-Gr2 17-B2, 21-D1, 3- Robinetin Anticancer Activity Birt et al., 1986 D2, 4-D2

301

Compound Compound Name Biological Activity Reference Number 18-C1 Rutin Antioxidant Francisco and Resurreccion, 2008 Anti-Inflammatory Activity Guardia et al., 2001 Antiallergic Properties Narayana et al., 2001 Antiviral Activity Narayana et al., 2001 Anticancer Activity Narayana et al., 2001 Capillary and Vaso-Protectant Levit et al., 1948 13-D1 Syringetin Induction of Osteoblast Maturation Hsu et al., 2009a Increases Bone Mass Hsu et al., 2009a 4-B2, 10-C1, 13- Tamarixetin Vasodilation Activity Pérez-Vizcaíno et al., 2002 C1, 21-C1, 22-C1, 3-C2, 15-D1, 22- D1, 30-D1, 1-F2, 2-Gr2 3-A1, 17-B2, 23- Tricetin Cancer Preventative Activity Hsu et al., 2009b D1, 3-D2, 4-D2 Isoflavones (including Glucosides, Methoxy- and Other Derivatives) 24-B2, 24-C1, 25- Biochanin A Antioxidant Francisco and Resurreccion, 2008 C1, 38-D1 4-B1 Daidzein Antioxidant Francisco and Resurreccion, 2008 38-D1 Glycitein Estrogenic Activity Song et al., 1999 43-A1 Glycitin Osteonecrosis Prevention Li et al., 2005 Enhancement of Spatial Working Memory Thorp et al., 2009 2-A1 Isoflavone Anti-Inflammatory Activity D'Alessandro et al., 2003 Lower (DHT) Conc. in Men Dillingham et al., 2005 29-C1 Lico-Isoflavone A Inhibition of Lipid Oxidation Toda and Shirataki, 2002 43-A1 Sissotrin Inhibition of Arachidonic Acid Kim and Yun-Choi, 2008 Flavanones (including Glucosides, Methoxy- and Other Derivatives) 22-C2, 31-C2 Cudraflavanone B Antifungal Activity Fukai et al., 2003 9-C2, 4-G2, Dihydromyricetin Antiviral Activity (Specifically H1N1) Roschek et al., 2009 24-A1 Eriocitrin Antioxidant Dorman et al., 2009

302

Compound Compound Name Biological Activity Reference Number 24-A1, 5-B1, 2-E2 Eriodictyol Vasodilation Activity Ramón Sánchez de Rojas et al., 1999 17-F1, 20-F1, 19- Euchrestaflavanone B Preventative Cancer Activity Kim et al., 2005 G1 5-B1, 7-C2, 6-D2, Fustin Inhibition of Neurotoxin (Parkinson’s Disease) Park et al., 2007 7-D2, 6-E1 19-B1, 5-D2, 1- Hesperitin Antioxidant Francisco and Resurreccion, 2008 Gr2 Anti-Inflammatory Garg et al., 2001 Vasodilation Activity Francisco et al., 2004 Antibacterial Activity Narayana et al., 2001 33-D1 Isosakuranetin Antihypertensive Properties Maruyama et al., 2009 Antifungal Activity Sacco and Maffei, 1997 2-B2 Naringenin Antioxidant Francisco and Resurreccion, 2008 Vasorelaxant Properties Narayana et al., 2001 Antibacterial Activity Narayana et al., 2001 Antiviral Activity Narayana et al., 2001 Ulcer Preventative Activity Motilva et al., 1994 24-A1 Neoeriocitrin Antimicrobial Activity Mandalari et al., 2007 17-F1, 20-F1, 19- Remangiflavanone B Antimicrobial Properties Deng et al., 2000 G1 33-D1, 7-E1, 3-E2, Sakuranetin Antifungal Activity Pacciaroni Adel et al., 2008 Anti-Inflammatory Hernández et al., 2007 Preventative Allergy Activity Ogawa et al., 2007 17-F1, 20-F1, 19- Sophoraflavanone G Antibacterial Properties Cha et al., 2009 G1 23-B2, 8-C2 Taxifolin Antioxidant Francisco and Resurreccion, 2008 Biflavones (including Glucosides, Methoxy- and Other Derivatives) 11-Gr1, 12-Gr1 Morelloflavone Anticancer Activity Pang et al., 2009 11-C2 Sciatopitysin Vasodilation Activity Chung et al., 1982

303

Compound Compound Name Biological Activity Reference Number Biflavanones (including Glucosides, Methoxy- and Other Derivatives) 18-F1, 18-G1 Kolaflavanone Anti-Atherogenic Properties Adaramoye et al., 2005 Antihepatoxic Activity (Liver Damage Iwu, 1985 Prevention) 15-G1 Manniflavanone Vascular Permeability and Fragility Treatment Ferrari et al., 2003 Prevention of Diabetes Mellitus Complications Flavonoid Oligomers (including Glucosides, Methoxy- and Other Derivatives) 2-E1, 1-F1, 15-F1, A-Type Insulin-Enhancing Activity Anderson et al., 2004 16-F1, 6-Gr1, 8- Proanthocyanidin Antibacterial Properties Foo et al., 200 Gr1, 14-G1 Dimers Immunomodulatory Properties Lin et al., 2002 Hemorrhage Protection Lou et al., 1999 2-F1, 3-F1, 5-F1, B-Type Vasodilation Activity Karim et al., 2000 12-F1, 13-F1, 1- Proanthocyanidin Inhibition of LDL Oxidation Pearson et al., 2001 Gr1, 2-Gr1, 3-Gr1, Dimers Anti-Inflammatory Activity Mao et al., 2000 4-Gr1, 5-Gr1 Antimicrobial Properties Buzzini et al., 2007 Hair Growth Promotion Takahashi et al., 1999

304 References

Adaramoye, O. A., Nwaneri, V. O., Anyanwu, K. C., Farombi, E. O., and Emerole, G. O. (2005). Possible anti-atherogenic effect of kolaviron (a Garcinia kola seed extract) in hypercholesterolaemic rats. Clinical and Experimental Pharmacology and Physiology, 32(1-2), 40-46.

Ahmad, A., Banerjee, S., Zhiwei, W., Dejun, K., and Sarkar, F. H. (2008). Plumbagin-induced apoptosis of human breast cancer cells is mediated by inactiviation of NF-KB and Bcl-2. J. Cell. Biochem., 105(6), 1461-1471.

Ahmed, M. S., El Tanbouly, N. D., Islam, W. T., Sleem, A. A., and El Senousy, A. S. (2005). Antiinflammatory flavonoids from Opuntia dillenii (Ker-Gawl) Haw flowers growing in Egypt. Phytotherapy Research, 19(9), 807-809.

Anderson, R. A., Broadhurts, C. L., Polansky, M. M., Schmidt, W. F., Khan, A., Flanagan, V. P., Schoene, N. W., and Graves, D. J. (2004). Isolation and characterization of polyphenol type-A polymers from Cinnamon with insulin-like biological activity. J. Agric. Food Chem., 52, 65-70.

Andrikopoulos, N. K., Kaliora, A.C., Assimopoulou, A. N., and Papageorgiou, V. P. (2003). Inhibitory activity of minor polyphenolic and nonpolyphenolic constituents of against in vitro low density lipoprotein oxidation. J. Med. Food, 5, 1-7.

Androutsopoulos, V. P., Mahale, S., Arroo, R. R., and Potter, G. (2009). Oncology Reports, 21(6), 1525- 1528.

Appledoorn, M. M., Sanders, M., Vincken, J.-P., Cheynier, V., Guernevé, C. L., Hollman, P. C. H., and Gruppen, H. (2009). Efficient isolation of procyanidin A-type dimers from peanut skins and B-type dimers from grape seeds. Food Chem., 117, 713-720.

Birt, D. F., Walker, B., Tibbels, M. G., and Bresnick, E. (1986). Anti-mutagenesis and anti-promotion by apigenin, robinetin, and indole-3-carbinol. Carcinogenesis, 7(6), 959-963.

Bisignano, G., Tomaino, A., Lo Cascio, R., Crisafi, G., Uccella, N., and Saija, A. (1999) On the in-vitro antimicrobial activity of oleuropein and . J. Pharm. Pharmacol., 51(8), 971-974.

Bisogno, F., Mascoti, L., Sanchez, C., Garibotto, F., Giannini, F>, Kurina-Sanz, M., and Enriz, R. (2007). Structure-antifungal activity relationship of cinnamic acid derivatives. J. Agric. Food Chem., 55(26), 10635-10640.

Buzzini, P., Turchetti, B., Ieri, F., Goretti, M., Branda, E., Mulinacci, N., and Romani, A. (2007). Catechins and proanthocyanidins: Naturally occurring O-heterocylces with antimicrobial activity. Top Heterocycl. Chem., 10, 239-263.

Cassidy, A, Hanley, B., and Lamuela-Raventos, R. M. (2000). Isoflavones, and stilbenes – origins, metabolism, and potential importance to human health. J. Sci. Food Agric. 80, 1044-1062.

Cha, J. D., Moon, S. E., Kim, J. Y., Jung, E. K., and Lee, Y. S. (2009). Antibacterial activity of sophoraflavanone G isolated from the roots of Sophora flavescens against methicillin-resistant Staphylococcus aureus. Phytother. Res., 23(9), 1326-1331.

Chen, Y.-C., Tsai, W.-J., Wu, M.-H., Lin, L.-C. And Kuo, Y.-C. (2009). Suberosin inhibits proliferation of human peripheral blood mononuclear cells through the modulation of the transcription factors NF-AT and NF-κB. Br. J. Pharmacol., 150(3), 298-312.

Cheng, Z. J., Kuo, S. C., Chan, S. C., Ko, F. N., and Teng, C. M. (1998). Antioxidant properties of butein isolated from Dalbergia odorifera. Biochim. Biophys. Acta, 1392(2-3), 291-299.

305

Chiang, L.-C., Chiang, W., Chang, M.-Y., Ng, L.-T., and Lin, C.-C. (2003). Antileukemic activity of selected natural products in Taiwan. American Journal of Chinese Medicine, 31(1), 37-46.

Choi, J., Lee, K.-T., Choi, M.-Y., Nam, J.-H., Jung, H.-J., Park, S.-K., and Park, H-J. (2005) Antiociceptive anti-inflammatory effect of monotropein isolated from the root of morinda officinlis. Biol. Pharm. Bull. 28(10), 1915-1918.

Chuanasa, T., Phromjai, J., Lipipun, V., Likhitwitayawuid, K., Suzuki, M., Pramyothin, P., Hattori, M., and Shiraki, K. (2008). Anti-herpes simplex virus (HSV-1) activity of oxyresveratrol derived from Thai medicinal plant: Mechanism of action and therapeutic efficacy on cutaneous HSV-1 infection in mice. Antiviral Research, 80(1), 62-70.

Chung, B. Y., Won, L. S., Lee, B. R., and Lee, C. H. (1982). A new chemical constituent of green leaves of Ginkgo biloba L. Journal of the Korean Chemical Society, 26(2), 95-98.

Civolani, C., Barghini, P., Roncetti, A. R., Ruzzi, M., Schiesser, A. (2000). Bioconversion of ferulic acid into vanillic acid by means of a vanillate-negative mutant of Pseudomonas fluorescens strain BF13. Appl. Environ. Microbiol., 66(6), 2311–2317.

Cren-Olive, C., Teissier, E., Duriez, P., and Rolando, C. (2003). Effect of catechin O-methylated metabolites and analogues on human LDL oxidation. Free Radical Biol. Med., 34(7), 850-855.

Cuyckens, F. and Caleys, M. (2004). Mass spectrometry in the structural analysis of flavonoids. J. Mass Spectrom., 39, 1-15.

D’Alessandro, T., Prasain, j., Benton, M. R., Botting, N., Moore, R., Darley-Usmar, V., Patel, R., and Barnes, S. (2003). Polyphenols, inflammatory response, and cancer prevention: Chlorination of isoflavones by human neutrophils. J. Nutr. 133, 3773S-3777S.

Deng, Y., Lee, J. P., Tianasoa-Ramamonjy, M., Snyder, J. K., Des Etages, S. A., Kanada, D., Snyder, M. P., and Turner, C. J. (2000). New antimicrobial flavanones from Physena madagascariensis. J. Nat. Prod., 63, 1082-1089.

Dillingham, B. L., McVeigh, B. L., Lampe, J. W., and Duncan, A. M. (2005). isolates of varying isoflavone content exert minor effects on serum reproductive hormones in healthy young men. J. Nutr., 135, 584-591.

Dorman, H. J., Kosar, M., Baser, K. H., and Hiltunen, R. (2009). Phenolic profile and antioxidant evaluation of Metha x piperita L. (peppermint) extracts. Natural Product Communications, 4(4), 535-542.

Ebeler, S. E., Brenneman, C. A., Kim, G.-S., Jewell, W. T., Webb, M. R., Chacon-Rodriguez, L., MacDonald, E. A., Cramer, A. C., Levi, A., Ebeler, J. D., Islas-Trejo, A., Kraus, A., Hinrichs, S. H., and Clifford, A. J. (2002). Dietary catechin delays tumor onset in a transgenic mouse model. Am. J. Clin. Nutr., 76(4), 865-872.

Ferrari, J., Terreaux, C., Kurtán, T., Szikszai-Kiss, A., Antus, S., Msonthi, J. D., and Hostettmann, K. (2003). Helv. Chim. Acta, 86(8), 2768-2778.

Fernandez, S. P., Nguyen, M., Yow, T. T., Chu, C., Johston, G. A., Hanrahan, J. R., and Chebib, M. (2009). Neurochem. Res., 34(10), 1867-1875.

Foo, L. Y., Lu, Y., Howell, A. B., and Vorsa, N. (2000). The structure of cranberry proanthocyanidins which inhibit adherence of uropathogenic P-fimbriated Escherichia coli in vitro. Phytochemistry,54, 173- 181.

306 Francisco, M. L. D. L. and Resurreccion, A. V. A. 2008. Functional components in peanuts. Crit. Rev. Food Sci. Nutr., 48(8), 715-746.

Frankel, E. N., Waterhouse, A. L. and Kinsella, J. E. (1993). Inhibition of human LDL oxidation by resveratrol. Lancet, 341,1103-1104.

Fukai, T., Yonekawa, M., Hou, A. J., Nomura, T., Sun, H. D., and Uno, J. (2003). Antifungal agents from the roots of Cudrania cochinchinensis against Candida, Cryptococcus, and Aspergillus species. J. Nat. Prod., 66(8), 1118-1120.

Garg, A., Garg, S., Zaneveld, L. J., and Singla, A. K. (2001). Chemistry and pharmacology of the citrus bioflavonoid . Phytother. Res., 15(8), 655-669.

Garvin, S., Ollinger, K., and Dabrosin, C. (2006). Resveratrol induces apoptosis and inhibits angiogenesis in human breast cancer xenografts in vivo. Cancer Letters, 231, 113-122.

Graf, E. (1992). Antioxidant potential of ferulic acid. Free Radic. Biol. Med., 13(4), 435-448.

Grunberger, D., Banerjee, R., Eisinger, K., Oltz, K., Efros, E. M., Caldwell, M., Estevez, V. and Nakanishi, K. (1988) Preferential cytotoxicity on tumor cells by caffeic acid phenethyl ester isolated from propolis. Experientia, 44, 230-232.

Guardia, T., Rotelli, A. E., Juarez, A. O., and Pelzer, L. E. (2001). Anti-inflammatory properties of plant flavonoids. Effects of rutin, quercetin and hesperidin on adjuvant arthritis in rat. Il Farmaco, 56(9), 683- 687.

Gu, L., Kelm, M. A., Hammerstone, J. F., Zhang, Z., Beecher, G., Holden, J., Haytowitz, D., and Prior, R. L. (2003). Liquid chromatographic/electrospray ionization mass spectrometric studies of proanthocyanidins in foods. J. Mass Spectrom., 38(12), 1272-1280.

Hernández, V., Recio, M. C., Máñez, S., Giner, R. M., and Ríos, J. L. (2007). Effects of naturally occurring dihydroflavonols from Inula viscosa on inflammation and enzymes involved in the arachidonic acid metabolism. Life Sci., 81(6), 480-488.

Higgs, G. A., Salmon, J. A., Henderson, B., and Vane, J. R. Pharmacokinetics of Aspirin and salicylate in relation to inhibition of arachidonate cyclooxygenase and anti-inflammatory activity. Proc. Natl. Acad. Sci., 84, 1417-1420.

Hsu, Y.-L, Kuo, P.-L., and Lin, C.-C. (2004). Acacetin inhibits the froliferation of Hep G2 by blocking cell cycle progression and inducing apoptosis. Biochem. Pharmacol., 67(5), 823-829.

Hsu, Y.-L., Liang, H.-L., Hung, C.-H., and Kuo, P.-L. (2009a). Syringetin, a flavonoid derivative in grape and wine, induces human osteoblast differentiation through bone morphogenetic protein-2/extracellular signal-regulated kinase 1/2 pathway. Mol. Nutr. Food Res., 53(11), 1452-1461.

Hsu, Y. L., Uen, Y. H., Chen, Y., Liang, H. L., and Kuo, P. L. (2009b). Tricetin, a dietary flavonoid, inhibits proliferation of human breast adenocarcinoma mcf-7 cells by blocking cell cycle progression and inducing apoptosis. J. Agric. Food Chem., 57(18), 8688-8695.

Isanga, J. and Zhang, G.-N. (2007). Biologically active components and nutraceuticals in peanuts and related products : review. Food Reviews International, 23, 123-140.

Itoigawa, M., Takeya, K., Ito, C., and Furu Kawa, H. (1999). Structure activity relationship of cardiotonic flavonoids in guinea pig papillary. J. Ethnopharmacol., 65, 267-272.

307 Ivanoyska, N., Yossifova, T., Vassileva, E., and Kostova, I. (1994). Effect of some hydroxycoumarins on complement-mediated hemolysis in human serum. Methods Find Exp. Clin. Pharmacol., 16, 557-562.

Iwu, M. M. (1985). Antihepatoxic constituents of Garcinia kola seeds. Experientia, 41(5), 699-700.

Kandil, F. E., Smith, M. A. L, Rogers, R. B., Pepin, M. F., Song, L. L., Pezzuto, J. M., and Seigler, D. S. (2002). Composition of a chemopreventive proanthocyanidin-rich fraction from cranberry fruits responsible for the inhibition of 12-O-tetradecanoyl phorbol-13-acetate (TPA)-induced ornithine decarboxylase (ODC) activity. J. Agric. Food Chem., 50, 1063-1069.

Karim, M., McCormick, K., and Kappagoda, C. T. (2000). Effects of cocoa procyanidins on endothelium- dependent relaxation. J. Nutr., 130, 2105S–2108S.

Katzenellenbogen, B. S., Katzenellenbogen, J. A., and Cai, D. M. (1979). Zearalenones: Characterization of the estrogenic potencies and receptor interactions of a series of fungal β-resorcylic acid lactones. Endocrinology, 105(1), 33-40.

Kayo Henmi, K., Hiwatashi, Y., Hikita, E., Toyama, N., and Hirano, T. (2009). Methoxy- and fluoro- chalcone derivatives arrest cell cycle progression and induce apoptosis in human melanoma cell A375. Biol. Pharm. Bull., 32(6), 1109-1113.

Kim, J. M. and Yun-Choi, H. S. (2008). Anti-platelet effects of flavonoids and flavonoid-glycosides from Sophora japonica. Arch. Pharmacal Res., 31(7), 886-890.

Kim, S.-H., Yoon, S.-H., Byong, W.-L., Ki, H. P., Young, H. K., and Bae, Y.-S. (2005). Oncology Research., 15(6), 327-332.

Kimura, Y. and Sumiyoshi, M. (2009). extract and its main component oleuropein prevent chronic ultraviolet B radiation-induced skin damage and carcinogenesis in hairless mice. J. Nutr., 139(11), 2079-2086.

Klein-Galczinsky, C. (1999). Pharmacological and clinical effectiveness of a fixed phytogenic combination trembling poplar (Populus tremula), true goldenrod (Solidago virgaurea) and ash (Fraxinus excelsior) in mild to moderate rheumatic complains. Wien Med. Wochenschr, 149, 248-253.

Kobuchi, H., Roy, S., Sen, C. K., Nguyen, H. G., and Packer, L. (1999). Quercetin inhibits inducible ICAM-1 expression in human endothelial cells through the JNK pathway. Am. J. Physiol., 277, C403- C411.

Kris-Etherton, P. M., Hecker, K. D., Bonanome, A., Coval, S. M., Binkoski, A. E., Hilpert, K. F., Griel, A. E., and Etherton, T. D. (2002). Bioactive compounds in foods: Their role in the prevention of cardiovascular disease and cancers. Am. J. Med., 113(9B), 71S-88S.

Kristinová, V., Mozuraityte, R., Storrø, I., and Rustad, T. (2009). J. Agric. Food Chem., 57(21), 10377- 10385.

Kuete, V., Tangmouo, J. G., Marion Meyer, J. J., and Lall, N. (2009). Disopyrone, crassiflorone and plumbagin: three antimycobacterial and antigonorrhoeal naphthoquinones from two Diospyros spp. Int. J. Antimicrob. Agents, 34(4), 322-325.

Kumar, R., Sowmyalakshmi, S., Koduru, S., Pahari, P., Rohr, J., Kyprianou, N., and Damodaran, C. (2009). Psoralidin, an herbal molecule, inhibits phophatidylinositol 3-kinase-mediated akt signaling in androgen-independent prostate cancer cells. Cancer Prevention Research, 2, 234-243.

308 Lee, S., Han, J. M., Kim, H., Kim, E. Jeong, T. S., Lee, W. S., and Cho, K. H. (2004). Synthesis of cinnamic acid derivatives and their inhibitory effects on LDL-oxidation, acyl-CoA, cholesterol acyltransferase-1 and -2 activity, and decrease of HDL-particle size. Bioorg. Med. Chem. Lett., 14(18), 4677-4681.

Lesage-Meessen, L., Delattre, M., Haon, M., Thibault, J. F., Ceccaldi, B. C., Brunerie, P., Asther, M. (1996). A two-step bioconversion process for vanillin production from ferulic acid combining Aspergillus niger and Pycnoporus cinnabarinus. J. Biotechnol., 50(2-3), 107–113.

Levit, L., Cholst, M., King, R., and Handelsman, M. (1948). Rutin therapy for increased capillary fragility and retinopathy associated with diabetes mellitus. Am. J. Med. Sci., 215(2), 130-135.

Li, X.-H., Zhang, J.-C., Sui, S.-F., and Yang, M.-S. (2005). Effect of , , and glycitin on osteogenic and adipogenic differentiation of bone marrow stromal cells and adipocytic transdifferentiation of osteoblasts. Acta Pharmacol. Sin., 26, 1081-1086.

Lim, H., Son, K. H., Chang, H. W., Bae, K., Kang, S. S., and Kim, H. P. (2008). Anti-inflammatory activity of pectolinarigenin and pectolinarin isolated from Cirsium chanroenicum. Biol. Pharm. Bull., 31(11), 2063- 2067.

Lin, L. C., Kuo, Y.-C., and Chou, C.-J. (2002). Immunomodulatory proanthocyanidins from Ecdysanthera utilis. J. Nat. Prod., 65, 505-508.

Liou, Y. F., Hall, I. H., and Lee, K. H. (2006). Antitumor agents LIV: The effects of daphnoretin on in vitro protein synthesis of ehrlich ascites carcinoma cells and other tissues. J. Pharm. Sci., 71(7), 745-749.

Liu, Z., Tao, X., Zhang, C., Lu, Y., and Wei, D. (2005). Protective effects of hyperoside (quercetin-3-o- galactoside) to PC12 cells against cytotoxicity induced by hydrogen peroxide and tert-butyl hydroperoxide. Biomed. Pharmacother., 59(9), 481-490.

Lotito, S. B. and Fraga, C. G. (1998). Catechin prevents human plasma oxidation. Free Radic. Biol. Med., 24, 435-441.

Lou, H., Yamazaki, Y., Saski, T., Uchida, M., Tanaka, H. and Oka, S. (1999). A-type proanthocyanidins from peanut skins. Phytochemistry, 51, 297-308.

Lou, H., Yuan, H., Ma, B., Ren, D., Ji, M., and Oka, S. (2004). Polyphenols from peanut skins and their free radical-scavenging effects. Phytochemistry, 65, 2391-2399.

Lou, H., Yuan, H., Yamazaki, Y., Sasaki, T., and Oka, S. (2001). Alkaloids and flavanoids from peanut skins. Planta Med., 67, 345-349.

Mada, S. R., Metukuri, M. R., Burugula, L., Reddanna, P., and Krishna, D. R. (2009). Antiinflammatory and antinociceptive activities of gossypin and procumbentin – cyclooxygenase-2 (COX-2) inhibition studies. Phytotherapy Research, 23(6), 878-884.

Mandalari, G., Bennett, R. N., Bisignano, G., Trombetta, D., Saija, A., Faulds, C. B., Gasson, M. J., and Narbad, A. (2007). Antimicrobial activity of flavonoids extracted from bergamot (Citrus bergamia Risso) peel, a byproduct of the essential oil industry. J. Appl. Microbiol., 103(6), 2056-2064.

Mao, T., Van De Water, J., Keen, C. L., Schmitz, H., and Gershwin, M. (2000). Cocoa procyanidins and human cytokine transcription and secretion. J. Nutr., 130, 2093S–2099S.

Maruyama, H., Sumitou, Y., Sakamoto, T., Araki, Y., and Hara, H. (2009). Antihypertensive effects of flavonoids isolated from brazilian green propolis in spontaneously hypertensive rats. Biol. Pharm. Bull., 32(7), 1244-1250.

309

Meriel, P., Galinier, F., Bounhoure, J. P., and Gavalda, J. (1964). Use in cardiology of a new anticoagulant: Dicoumoxyl (1-bis-methoxy (4-hydroxy, 3-coumarinyl) 2,2-ethane). Toulouse Med., 65, 849-858.

Motilva, V., Alarcón de la Lastra, C., and Martin, M. J. (1994). Ulcer-protecting effects of naringenin on gastric lesions induced by ethanol in rat: role of endogenous prostaglandins. J. Pharm. Pharmacol., 46(2), 91-94.

Narayana, K. R., Reddy, M. S., Chaluvadi, M. R., and Krishna, D. R. (2001). Bioflavonoids classification, pharmacological, biochemical effects and therapeutic potential. Indian Journal of Pharmacology, 33, 2-16.

Natarajan, K., Singh, S., Burke, Jr., T. R., Grunberger, D., and Aggarwal, B. B. 1996. Caffeic acid phenetyl ester is a potent and specific inhibitor of activation of nuclear transcription factor NF-KB. Proc. Natl. Acad. Sci., 93, 9090-9095.

Niu, H. S., Liu, I. M., Cheng, J. T., Lin, C. L., and Hsu, F. L. (2008). Hypoglycemic effect of syringin from Eleutherococcus senticosus in Streptozotocin-induced diabetic rats. Planta Med. 74(2), 109-113.

Ogawa, H., Sasai, N., Kamisako, T., and Baba, K. 2007. Effects ofosthol on blood pressure and lipid metabolism in stroke-prone spontaneously hypertensive rats. J. Ethnopharmacology, 112(1), 26-31.

Ogawa, Y., Oku, H., Iwaoka, E., Iinuma, M., and Ishiguro, K. (2007). Allergy-preventive flavonoids from Xanthorrhoea hastilis. Chem. Pharm. Bull., 55(4): 675-678.

Olas, B. and Waschowicz, B. (2002). Resveratrol and vitamin C as antioxidants in blood platelets. Thrombosis Research. 106, 143-148.

Orallo, F., Alvarez, E., Basaran, H., and Lugnier, C. (2004). Comparative study of the vasorelaxant activity, superoxide-scavenging ability and cyclic nucleotide phosphodiesterase-inhibitory effects of and hesperidin. Naunyn-Schmiedeberg's Arch. Pharmacol., 370(6), 452-463.

Pacciaroni Adel, V., Gette Mde, L., Derita, M., Ariza-Espinar, L., Gil, R. R., Zacchino, S. A., and Silva, G. L. (2008). Antifungal activity of Heterothalamus alienus metabolites. Phytotherapy Research, 22(4), 524- 528.

Pang, X., Yi, T., Yi, Z., Cho S. G., Qu, W., Pinkaew, D., Fujise, K., and Liu, M. (2009). Morelloflavone, a biflavonoid, inhibits tumor angiogenesis by targeting rho GTPases and extracellular signal-regulated kinase signaling pathways. Cancer Research 69(2), 518-525.

Park, B. C., Lee, Y. S., Park, H. J., Kwak, M. K., Yoo, B. K., Kim, J. Y., and Kim J. A. (2007). Protective effects of fustin, a flavonoid from Rhus verniciflua Stokes, on 6-hydroxydopamin-induced neuronal cell death. Exp. Mol. Med., 39(3), 316-326.

Park, Y., Lee, S., and Lim, Y. (2007). Relationships between structures of hydroxyflavone derivatives and their anti-oxidative activities. In Frontiers in the Convergence of Bioscience and Information Technologies (pp. 43-47).

Paterson, J. R. and Lawrence, J. R. (2001). Salicylic acid: a link between aspirin, diet and the prevention of colorectal cancer. Q. J. Med. 94, 445-448.

Pearson, D. A., Schmitz, H.H., Lazarus, S. A., and Keen, C.L. (2001). Inhibition of in vitro low-density lipoprotein oxidation by oligomeric procyanidins present in chocolate and cocoas. In: L. Packer, Methods in Enzymology, (pp. 350-360). New York: Academic Press. Pérez-Vizcaı́no, F., Ibarra, M., Cogolludo, A. L., Duarte, J., Zaragozá-Arnáez, F., Morenol, L., López- López, G., and Tamargo, J. (2002). Endothelium-independent vasodilator effects of the flavonoid quercetin

310 and its methylated metabolites in rat conductance and resistance arteries. J. Pharmacol. Exp. Ther., 302(1), 66-72.

Popova, M., Bankova, V., Spassov, S., Tsvetkova, I., Naydenski, C., Silva, M. V., and Tsartsarova, M. (2001). New bioactive chalcones in propolis from El Salvador. Z. Naturforsch., C: Biosci., 56(7-8), 593- 596.

Rakotoarison, O., Rabenau, I., Lobstein, A., Um, B. H., Schott, C., Anton, R., Randriantsoa, A., and Andriantsitohaina, R. (2003). Vasorelaxing properties and bio-guided fractionation of Cedrelopsis grevei. Planta Med., 69(2), 179-181.

Ramón Sánchez de Rojas, V., Somoza, B., Ortega, T., Villar, A. M., and Tejerina, T. (1999). Vasodilatory effect in rat aorta of eriodictyol obtained from Satureja obovata. Planta Med., 65(3), 234-238. Ratty, A. K. Effects of flavonoids on nonenzymatic lipid peroxidation: structure-activity relationship. Biochem. Med. Metabol. Biol., 39, 67-79.

Rocha, B. S., Gago, B., Barbosa, R. M., and Laranjinha, J. (2009). Dietary polyphenols generate nitric oxide from nitrite in the stomach and induce smooth muscle relaxation. Toxicology, 265(1-2), 41-48.

Roschek, B., Fink, R. C., McMichael, M. D., Li, D. and Alberte, R. S. (2009). Elderberry flavonoids bind to and prevent H1N1 infection in vitro. Phytochemistry, 70(10), 1255-1261.

Rotondo, S., Rajtar, G., Manarini, S., Celardo, A., Rotillo, D., de Gaetano, G., Evangelista, V., and Cerletti, C. (1998). Effect of trans-resveratrol, a natural polyphenolic compound, on human polymorphonuclear leukocyte function. Br. J. Pharmacol., 123(8), 1691-1699.

Sacco, S. and Maffei, M. (1997). The effect of isosakuranetin (5,7-dihydroxy 4’-methoxy flavanone) on potassium umptake in wheat root segments. Phytochemistry, 46(2), 245-248.

Samoszuk, M., Tan, J., and Chorn, G. (2005). The chalcone butein from Rhus vernifciflua Stokes inhibits clonogenic growth of human breast cancer cells co-cultured with fibroblasts. BMC Complement Altern. Med., 5, 1-5.

Sata, Y., Akao, T., He, J. X., Nojima, H., Kuraishi, Y., Morota, T., Asano, T. and Tani, T. (2006). Glycycoumarin from Glycyrrhizae Radix acts as a potent antispasmodic through inhibition of phosphodiesterase 3. J Ethnopharmacol., 105(3), 409-414.

Sekine-Osajima, Y., Sakamoto, N., Nakagawa, M., Itsui, Y., Tasaka, M., Nishimura-Sakurai, Y., Chen, C.

HH., Suda, G., Mishima, K., Onuki, Y., Yamamoto, M., Maekawa, S., Enomoto, N., Kanai, T., Tsuchiya, K., and Watanabe, M. (2009). Hepatology Research, 39(1), 60-69.

Setchell, K. D., Zhao, X., Shoaf, S. E., and Ragland, K. (2009). The pharmacokinetics of S-(-)equol administered as SE5-OH tablets to healthy postmenopausal women. J. Nutr., 139(11), 2037-2043.

Sethi, O. P., Anand, K. K., and Gulati, O. D. (1992). Evaluation of xanthotoxol for central nervous system activity. J. Ethnopharmacol., 36(3), 239-247.

Shih, Y. W., Lee, Y. C., Wu, P. F., Lee, Y. B., and Chiang, T. A. Plumbagin inhibits invasion and migration of liver cancer HepG2 cells by decreasing productions of matrix metalloproteinase-2 and urokinase- plasminogen activator. Hepatology Research, 39(10), 998-1009.

Song, T. T., Hendrich, S., and Muphy, P. A. (1999). Estrogenic activity of glycitein, a soy isoflavone. J. Agric. Food Chem., 47(4), 1607-1610. Spencer, J. P. E., Schroeter, H., Kuhnle, G., Srai, S. K. S., Tyrrell, R. M., Hahn, U., and Rice-Evans, C. (2001). Epicatechin and its in vivo metabolite, 3’-O-methyl epicatechin, protect human fibroblasts from oxidative-stress-induced cell death involving caspase-3 activation. Biochem. J., 354, 493-500.

311

Takahashi, T., Kamiya, T., Hasegawa, A., and Yokoo, Y. (1999). Procyanidins oligomers selectively and intensively promote proliferation of mouse hair epithelial cells in vitro and activate hair follicle growth in vivo. J. Investigative Dermatology, 112, 310-316.

Tanaka, Y., Kikuzaki, H., Fukuda, S., and Nakatani, N. (2001). Antibacterial compounds of licorice against upper airway respiratory tract pathogens. J. Nutr. Sci. Vitaminol., 47(3), 270-273.

Tang, Y. Z. and Liu, Z. Q. (2007). Insight into the free-radicial-scavenging mechanism of hydroxyl- substituent Schiff bases in the free-radical-induced hemolysis of erythrocytes. Cell Biochem. Funct., 25(6), 701-710.

Teng, B.-S., Lu, Y.-H., Wang, Z.-T., Tao, X.-Y., and Wei, D.-Z. (2006). In vitro anti-tumor activity of isorhamnetin isolated from Hippophae rhamnoides L. against BEL-7402 cells. Pharmacol. Res., 54(3), 186-194.

Thorp, A. A., Sinn, N., Buckley, J. D., Coates, A. M., and Howe, P. R. (2009). Soya isoflavone supplementation enhances spatial working memory in men. Br. J. Nutr., 102(9), 1348-1354.

Toda, S. and Shirataki, Y. (2002). Inhibitory effects of isoflavones in Sophora mooracrotiana on lipid peroxidation by superoxide. Pharmaceutical Biology, 40(6), 422-424.

Tonari, K., Mitsui, K., and Yonemoto, K. (2002). Structure and antibacterial activity of cinnamic acid related compounds. J. Oleo Sci., 51(4), 271-273.

Vasconcelos, J. F., Teixeira, M. M., Barbosa-Filho, J. M., Agra, M. F., Nunes, X. P., Giulietti, A. M, Riberiro-Dos-Santos, R., and Soares, M. B. (2009). Effects of umbelliferone in a murine model of allergic airway inflammation. Eur. J. Pharmacol., 609(1-3), 126-131.

Visioli, F., Bellosta, S., and Galli, C. (1998). Oleuropein, the bitter principle of , enhances nitric oxide production by mouse macrophages. Life Sci., 62(6), 541-546.

Visioli, F. and Galli, C. (1994). Oleuropein protects low density lipoprotein from oxidation. Life Sci., 55(24), 1965-1971.

Wang, M.S. (1980). Studies on the chemical constituents of Zu-Shi-Ma (Daphne giraldii) (III). Chinese Traditional and Herbal Drugs, 11(9), 389-390.

Woo, S. W., Lee, S. H., Kang, H.-C., Park, E.-J., Zhao, Y.-Z., Kim Y.-C., and Sohn, D. H. (2003). Butein suppresses myofibroblastic differentiation of rat hepatic stellate cells in primary culture. J. Pharm. Pharmacol., 55(3), 347-352.

Yagasaki, K., Furuse, T., and Miura, Y. (1998). Inhibitory effects of coffee and caffeic acid, 3,4-dihy- droxycinnamic acid, on the invasion of hepatoma cells. In 2nd International Electronic Conference on Synthetic Organic Chemistry.

Yen, G. C. and Hsieh, C. L. (2002). Inhibitory effects of Du-zhong (Eucommia ulmoides Oliv.) against low-density lipoprotein oxidative modification. Food Chem., 77, 449-456.

Yokozawa, T., Kim, H. Y., Cho, E. J., Choi, J. S., and Chung, H. Y. (2002). Antioxidant effects of isorhamnetin 3,7-Di-O-β-d-glucopyranoside isolated from mustard leaf (Brassica juncea) in rats with Streptozotocin-induced diabetes. J. Agric. Food Chem., 50(19), 5490-5495.

Yu, J., Ahmedna, M., and Goktepe, I. (2005). Effects of processing methods and extraction solvents on concentration and antioxidant activity of peanut skin phenolics. Food Chem., 90, 199-206.

312

Yu, J., Ahmedna, M., Goktepe, I., and Dai, J. (2006). Peanut skin procyanidins:Composition and antioxidant activities as affected by processing. J. Food Compos. Anal., 19, 364-371.

Yu, S. M., Cheng, Z. J., and Kuo, S. C. (1995). Endothelium-dependent relaxation of rat aorta by butein, a novel cyclic AMP-specific phosphodiesterase inhibitor. Eur. J. Pharmacol. 280(1), 69-77.

Zhou, H. Y., Hong, J. L., Shu, P., Ni, Y. J., and Qin, M. J. A new dicoumarin and anticoagulant activity from Viola yedoensis Makino. Filoterapia., 80(5), 283-285.

313 SUMMARY

Prior to this research, the information available on the composition of polyphenols

in peanut skins was incomplete. Previous studies had identified few compounds, in part

due to the utilized analytical methodology. However, the vast number of phenolic-based

compounds identified in this study reveals that peanut skins are possibly one of the most

phenolic-rich and diverse plant sources ever studied. The majority of these compounds

are known to exhibit powerful antioxidant activity, which was reported in this study by

determining the antioxidant profile of the Gregory peanut line. This indicates that peanut skins could be used as a novel source for the extraction of natural antioxidant and

preservative components. This proves the capability that peanut skins, which had been

previously viewed as a by-product (or waste-product) of the peanut industry, can become

a value-added commodity and revenue stream for peanut farmers.

Numerous compounds were detected which have been reported to exhibit specific

chemical, biological, nutraceutical, and pharmaceutical attributes. This indicates the

possibility that peanut skins could be utilized as a novel source for the extraction of

compounds which have the potential to treat ailments such as cardiovascular disease,

various forms of cancer, and other human health-related conditions. All of these factors

showcase the need for future research into the biological effects of components contained

within peanut skins. Future research could use this study as a template, to further narrow

down the numerous compounds which were tentatively identified (many of which have

significant antioxidant and biological effects).

314 APPENDIX

315

Figure A.1: HPLC-UV Chromatogram of Peanut Skin Raw Extract at 280nm.

316 RT:0.00 - 46.03 15.21 17.40 NL: 1.52E-1 0.0 0.0 UV Analog Negative_M 0.110 ode_Reed- 2005-0913- 10_Std_Mix

0.105 32.68 0.0

0.100 19.92 0.0

0.095 18.76 25.98 30.78 0.0 0.0 0.0 23.99 0.090 0.0 15.73 0.0 0.085 5.53 0.0

In te n0.080 sity

26.86 38.99 0.075 0.0 0.0 29.40 0.0 0.070 5.28 9.42 13.45 20.93 0.0 0.0 0.0 0.0 5.11 0.065 0.0 1.79 0.0 28.73 43.3343.8745.71 0.060 9.19 11.37 23.83 0.0 36.0936.4438.77 41.09 4.68 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 7.97 4.03 0.0 0.055 0.0

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 Time (min) Figure A.2: HPLC-UV Chromatogram of Standard Mix at 280nm.

317 RT:0.00 - 70.04 16.53 NL: 1.60E7 288.9 Base Peak F: - c ESI 100 sid=1.00 Full ms [ 125.00-600.00] MS Negative_Mode_Reed- 90 68.78 2005-0913-10_Std_Mix 411.3 26.04 80 17.44 447.0 456.8 70 18.81 32.76 15.18 441.0 285.1 60 178.9 11.46 50 65.16 576.9 64.72 550.7 26.90 550.6 40 9.46 286.6 30.47 39.74 56.63 59.98 305.0 301.1 39.18 Relative Abundance Relative 30 373.2417.1 451.3 468.5 1.28 49.07 56.05 NL: 1.52E-1 29.44 37.82 47.16 386.6 20 386.6 UV Analog 194.7 301.0 550.5 550.7 Negative_Mode_Reed- 5.49 20.9624.59 2005-0913-10_Std_Mix 10 169.0 304.6386.7

0 17.40 15.21 0.0 0.0 0.15

0.14

0.13

0.12

32.68 0.11 19.92 0.0 0.10 0.0 25.98 30.78 23.99

In te n sity 0.0 0.0 0.0 0.09 5.53 0.0 26.86 38.99 0.08 0.0 9.42 13.45 20.93 0.0 0.07 1.79 0.0 0.0 0.0 53.0754.8656.8858.7662.4164.7367.3269.59 44.4847.6250.01 9.07 36.0938.77 42.33 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.06 0.0 0.0 0.0 0.0

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.3: Standard Mix: HPLC-UV Chromatogram (bottom) versus TIC [Total Ion Chromatogram – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125-600 M-H-)] (top).

318 RT:0.00 - 70.04 16.53 NL: 1.60E7 288.9 Base Peak F: - c ESI 100 68.78sid=1.00 Full ms [ 17.44 26.04 411.3125.00-600.00] MS 456.8 447.0 80 15.18 18.81 32.76 Negative_Mode_Reed- 178.9 441.0 285.1 2005-0913-10_Std_Mix 11.46 65.16 60 64.72 576.9 26.90 550.6550.7 9.46 39.18 56.0556.63 286.6 30.47 59.98 40 1.28 305.0 301.1 373.2 49.07 386.6451.3 468.5 37.82 45.9747.16 194.7 5.49 20.9624.59 550.5 550.7386.6 NL: 6.01E6 20 468.5 169.0 304.6386.7 Base Peak F: - c sid=1.00 Relative Abundance Relative 0 d Full ms2 MS 15.19 Negative_Mode_Reed- 135.116.54 2005-0913-10_Std_Mix 100 245.018.74 289.0 80 25.95 1.29 301.0 176.9 11.47 60 NL: 2.26E6 424.9 69.77 32.65 65.6968.19 Base Peak F: - c sid=1.00 40 30.74 386.6 9.38 28.71 50.92 62.41386.6386.6 d Full ms3 MS 285.1285.1 43.31 49.37 53.33 56.64 1.56 5.50 285.0 37.8339.19 386.6 386.6 Negative_Mode_Reed- 179.0 24.00 311.2 386.6 386.6 20 386.6205.1 233.1 406.5125.0 185.0 2005-0913-10_Std_Mix Relative Abundance Relative 0 16.55 203.1 19.81 100 150.9 1.31 11.49 25.97 NL: 1.52E-1 80 407.1 178.9 159.0 UV Analog Negative_Mode_Reed- 60 15.73 2005-0913-10_Std_Mix 125.0 61.8163.82 69.51 40 28.73 32.67 52.63 65.19 43.33 304.6304.7 304.6 8.97 9.40 257.1 285.0 44.96 50.95304.653.64 304.6 1.60 22.67 36.5237.87 41.99311.2 20 304.6 325.2 304.6 304.6 388.4 164.0164.0 298.6 304.7304.6 Relative Abundance Relative 0 15.2117.40 0.0 0.0

0.14 32.68 0.12 19.92 23.9925.98 30.78 0.0 0.0 0.10 5.53 0.0 0.0 0.0 38.99 0.0

In te n sity 9.42 13.45 56.8858.76 64.7367.3269.12 0.08 1.79 36.44 0.0 44.4846.7649.4752.2854.75 8.94 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.06 0.0

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.4: Standard Mix: HPLC-UV Chromatogram (bottom) versus TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra (top), “ms2” (second), and “ms3” (third) Base Peaks for Most Intense Ions in SID=1.0 Scan Range (m/z 125-600 M-H-)]

319 Negative_Mode_Reed-2005-0913-10_Std_Mix 9/13/2005 8:35:57 PM Standard Mix; 20 uL N-Pak-C18;0.75;%B(min)=0(0-2)>4.75(4)>30.9(15)>74.7(42)>95(54.5-70)/280nm/ESI

Negative_M ode_Reed-2005-0913-10_Std_M #690 RT:16 ix . 5 3 AV:1 NL:1.60E7 F:- c ESI sid=1.00 Full ms [ 125.00-600.00] 288.9 10 0

80

60 348.7 578.3 40 579.3 347.8 20 Relative Abundance Relative 430.5 290.0 324.7 349.8 468.7474.8 17 9 . 1 226.5 3 11. 1 327.9 370.9 424.1 431.8 452.6 487.8 512.2 581.2590.8 0 14 0 16 0 18 0 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500 520 540 560 580 600 m/z

Negative_M ode_Reed-2005-0913-10_Std_M #691RT:16 ix . 5 4 AV:1 NL:5.24E6 T:- c sid=1.00 d Full ms2 [email protected] [ 85.00-590.00] 245.0 10 0

80

60

205.1 40

20 Relative Abundance Relative 246.0 179.0 203.1 16 1. 2 231.1 247.1 271.0 109.1 125.0 137.0 150.9 165.0 19 1. 1 227.2 0 10 0 12 0 14 0 16 0 18 0 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500 520 540 560 580 m/z

Negative_M ode_Reed-2005-0913-10_Std_M #692 RT:16 ix . 5 5 AV:1 NL:2.26E6 T:- c sid=1.00 d Full ms3 [email protected] [email protected] [ 70.00-500.00] 203.1 10 0

80

60

40 227.1

187.1 20 Relative Abundance Relative 204.1 16 1. 1 175.1 18 8 . 1 204.9216.9 228.1 135.1 148.9 19 9 . 0 245.2 261.3 0 80 10 0 12 0 14 0 16 0 18 0 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500 m/z

Figure A.5: Ion Spectra for Epicatechin – Showing the “ms” Normal Mass Spectra (top), “ms2” MS/MS Spectra (second), and “ms3” MS/MS/MS Spectra (bottom).

320 OH OH OH OH

OH O OH O - OH O - OH O - OH -H C OH CH OH -C2H3O CH OH

C OH OH OH

OH OH OH OH m/z 290 m/z 289 m/z 289 m/z 245

C15H14O6 C15H13O6 C15H13O6 C13H10O5 OH OH O O OH O -HO -C4H4O CH O OH OH O OH O 2 OH OH OH - CH OH - CH3 CH CH m/z 205 OH m/z 271 m/z 289 OH OH OH C11H8O4 C15H12O5 C15H13O6 OH OH C OH O OH O OH O OH O OH O - - - -C H O CH OH -H O -HO O 6 3 -HO OH C OH OH OH OH OH m/z 163 m/z 179 OH OH OH C9H8O3 m/z 290 m/z 289 m/z 271 C9H9O4

C15H14O6 C15H13O6 C15H12O5 OH OH OH OH OH O OH O OH O -H -C7H6O3 OH OH OH - C OH - - OH CH OH CH2 OH OH OH OH OH m/z 151 m/z 290 m/z 289 m/z 289 C8H7O3 C H O C H O C H O 15 14 6 15 13 6 15 13 6

OH OH OH OH OH O OH O OH O OH O OH O -C H O OH -H OH OH -CHO CH OH 7 7 2 - CH - - - OH CH OH CH O CH 2 2 OH OH OH OH OH m/z 137 m/z 290 m/z 289 m/z 289 C7H5O3 C H O C H O C H O 15 14 6 15 13 6 15 13 6 Figure A.6: Fragmentation Scheme for Epicatechin

321 OH OH OH

OH O - OH O - OH O - C OH O - OH O - C OH CH OH -HO CH -HO CH -CHO CH

OH O O O CH2 OH OH OH OH OH m/z 289 m/z 289 m/z 271 m/z 227

C15H13O6 C15H13O6 C15H12O5 C14H10O3 OH OH OH OH OH

OH O -C H -C H O OH -HO OH O OH O 2 2 OH O 2 2 O OH OH C OH C OH - CH OH - - - - C CH OH CH CH O CH O CH O OH m/z 271 m/z 271 m/z 245 m/z 203 m/z 289 C15H12O5 C15H12O5 C13H10O5 C11H8O4 C H O 15 13 6

OH OH OH OH OH - - - OH O OH OC OH C O OH CH O -C H O OH -H OH -HO OH OH 4 4 2 O OH CH C - OH OH C C OH OH OH OH m/z 187

m/z 290 m/z 289 m/z 271 m/z 271 C11H8O3 C H O C H O C H O C H O 15 14 6 15 13 6 15 12 5 15 12 5

OH OH OH OH OH

OH O OH O OH O OH O - - - - -C H O C OH CH OH -CHO CH OH CH OH 4 4 2 O - C CH OH

OH O CH2 C CH2 CH2 OH OH OH OH m/z 175 m/z 289 m/z 289 C H O C H O C H O 10 8 3 15 13 6 15 13 6

OH OH OH OH OH

OH O - OH O - OH O - OH CH O - C OH CH OH -C2H3O CH OH CH OH -C4H4O2 O - CH OH C OH OH C OH OH OH OH m/z 161 m/z 289 m/z 289 m/z 245 m/z 245 C9H6O3 C H O C H O C H O C H O 15 13 6 15 13 6 13 10 5 13 10 5 Figure A.6 (cont.): Fragmentation Scheme for Epicatechin

322 RT:0.22 - 62.89 48.11 NL: 3.86E-1 0.0 0.150 UV Analog Negative_Mo de_Reed- 0.145 2005-0913- 01_Fraction_ 0.140 A

0.135

0.130 1.66 0.125 0.0

0.120

0.115

0.110 61.16 0.105 0.0

0.100

In te n sity 40.92 0.0 44.14 54.50 0.095 57.24 0.0 0.0 0.0 0.090 60.57 46.51 54.93 0.0 0.085 0.0 0.0 48.98 52.63 38.31 41.57 0.080 0.0 0.0 0.0 0.0 59.52 34.78 50.04 0.0 0.075 8.62 30.49 0.0 37.54 0.0 0.0 11.90 7.78 29.21 0.0 0.0 50.86 0.070 0.0 14.34 16.86 21.72 34.43 0.0 19.01 22.8027.94 0.0 0.0 11.75 0.0 0.0 0.0 0.0 0.065 0.0 0.0 0.0 2.45 0.0 7.48 0.0 6.50 0.060 0.0 0.0

0.055

5 10 15 20 25 30 35 40 45 50 55 60 Time (min) Figure A.7: HPLC-UV Chromatogram of Peanut Skin Fraction A (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm.

323 RT:0.00 - 62.89 25.09 NL: 6.18E7 187.1 Base Peak F: - c ESI 100 sid=1.00 Full ms [ 125.00-600.00] MS 38.44 37.70 Negative_Mode_Reed- 329.1 53.55 80 329.1 43.66 2005-0913-01_Fraction_A 311.1 271.2 35.0535.80 57.88 557.0327.1 438.9 60 45.31 56.52 313.2 525.1 30.46 48.29 58.65 40 34.81 593.361.52 28.24168.9 295.1 17.42 324.9 40.71 49.57 551.2 241.0 56.21 1.69 474.9 19.94 325.1 312.9 16.30 23.69 543.8 NL: 7.60E7 Relative Abundance Relative 20 400.8 8.54 9.90 12.61 173.1 261.0 2.56 377.2 Base Peak F: - c ESI 598.7 400.3 300.8457.0 sid=2.00 Full ms [ 0 590.00-2000.00] MS 56.17 Negative_Mode_Reed- 833.2 2005-0913-01_Fraction_A 100

80

59.05 60 1129.5 NL: 3.86E-1 62.71 40 37.56 38.23 771.8UV Analog 679.0 41.24 45.12 54.80 Negative_Mode_Reed- 31.59 1031.2 49.10 53.47 19.37 21.74 25.25 30.49 999.0 1002.1 2005-0913-01_Fraction_A 17.59 1095.4 1379.5 962.0 R e la tive A b u20 n d a n ce 1.61 15.65 1021.1 7.70 10.57 12.68 595.71982.4 1408.7 1626.4 1872.8 1822.01678.2 1525.5955.11700.6 0 48.11 0.0

0.35

0.30

0.25 48.61 47.82 0.0 0.20 0.0 In te n sity 1.66 0.15 61.16 0.0 40.92 44.14 54.50 57.24 38.31 52.63 7.788.62 11.90 30.49 34.78 0.0 2.456.50 14.34 16.8619.01 21.72 26.1329.21 0.0 0.0 0.0 0.0 0.10 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0 5 10 15 20 25 30 35 40 45 50 55 60 Time (min) Figure A.8: Peanut Skin Fraction A: HPLC-UV Chromatogram (bottom) versus TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in Both SID=1.0 Scan Range (m/z 125-600 M-H-) (top) and SID=2.0 Scan Range (m/z 590-2000 M-H-) (middle)].

324 RT:0.00 - 62.89

25.09 NL: 6.18E7 187.1 Base Peak F: - c ESI 100 sid=1.00 Full ms [ 125.00-600.00] MS Negative_Mode_Reed- 90 2005-0913- 38.44 01_Fraction_A 80 37.70329.1 329.1 43.66 53.55 70 311.1 271.2

57.88 60 35.0535.80 557.0327.1 438.9 45.31 56.52 50 313.2 525.1

40 30.46 48.29 58.65 168.9 34.81 295.1 593.361.52 551.2 Relative Abundance Relative 30 17.42 28.24 324.9 49.57 40.71 474.9 241.0 312.9 NL: 3.86E-1 325.1 47.37 56.21 20 1.69 19.94 23.69 32.56 397.0 543.8 UV Analog 173.1 212.9 Negative_Mode_Reed- 400.8 16.30 261.0 8.54 9.90 12.61 2005-0913- 10 7.33 377.2 300.8457.0 598.7 01_Fraction_A 434.4 0 48.11 0.0

0.35

0.30

0.25 48.61 47.82 0.0

In te n0.20 sity 0.0

0.15 1.66 0.0 61.16 40.92 44.14 54.50 57.24 0.0 38.31 46.51 52.63 8.62 34.78 0.0 0.0 0.0 0.0 0.10 7.78 11.9014.3416.8619.0121.72 29.2130.49 2.45 26.13 0.0 0.0 0.0 6.50 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0 5 10 15 20 25 30 35 40 45 50 55 60 Time (min) Figure A.9: Peanut Skin Fraction A: HPLC-UV Chromatogram (bottom) versus TIC [Total Ion Chromatogram – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125-600 M-H-)] (top).

325 RT:0.40 - 69.98 1.66 NL: 5.15E-1 0.0 8.74 0.0 UV Analog 0.150 Negative_Mo de_Reed- 0.145 2005-0913- 02_Fraction_ B 0.140 17.11 0.0 0.135 21.58 0.0 0.130

0.125 21.03 23.40 0.0 0.120 0.0

0.115 27.09 0.110 0.0

0.105

0.100 In te n sity

0.095 8.26 53.01 0.0 0.0 0.090

7.78 16.24 0.085 0.0 4.09 0.0 26.59 27.80 16.01 0.0 0.0 0.0 0.080 11.01 0.0 50.71 0.0 0.0 57.66 0.075 0.0 53.37 28.72 0.070 7.36 48.19 30.05 0.0 0.0 35.39 46.03 0.0 0.0 58.1761.01 0.0 40.85 62.23 66.92 0.065 0.0 39.12 45.16 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.060

0.055

5 10 15 20 25 30 35 40 45 50 55 60 65 Time (min) Figure A.10: HPLC-UV Chromatogram of Peanut Skin Fraction B (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm.

326 RT:0.00 - 69.98 53.00 NL: 4.83E7 419.0 Base Peak F: - c ESI 100 sid=1.00 Full ms [ 125.00-600.00] MS Negative_Mode_Reed- 80 2005-0913- 02_Fraction_B 60 17.08 472.9 23.01 49.11 472.7 435.2 40 48.18 351.0 53.45 62.13 65.68 26.53 40.92 45.88 65.16 20.1422.64 419.256.5159.69376.0 468.7 12.98 593.127.72 33.29 40.46367.0 353.2 468.5 NL: 3.83E7

R e la tive20 A b u n d a n ce 7.89 11.07 36.60 1.65 472.8 570.5522.8 7.43 414.7 137.0 562.4 467.4 386.7 Base Peak F: - c ESI 462.5137.1 518.1 261.2 463.0 sid=2.00 Full ms [ 0 590.00-2000.00] MS 56.19 41.95 Negative_Mode_Reed- 38.47 1578.9 2005-0913- 100 1679.2 58.53 1578.9 774.8 02_Fraction_B 60.15 43.97 53.49 27.09 52.58 1578.963.7666.54 80 35.60 1479.3 623.0 1579.0 1479.2 774.5775.0 862.3 45.46 21.02 1679.1 60 741.0 33.95 1.60 19.39 21.49 33.09 903.1 785.2 886.91379.5 NL: 5.15E-1 1579.4 40 UV Analog 17.81 Negative_Mode_Reed- 12.32 14.84 933.0 2.79 9.27 2005-0913- 1895.4 Relative Abundance Relative 20 1478.8 02_Fraction_B 1679.9 1679.1

0 1.66 0.0 0.5

0.4

0.3

Intensity 8.74 0.2 17.11 21.58 0.0 23.40 27.09 0.0 53.01 4.09 7.78 16.24 0.0 27.80 11.01 0.0 0.0 30.05 35.39 40.85 46.0348.1950.71 57.6658.1762.23 39.12 0.0 66.92 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0 5 10 15 20 25 30 35 40 45 50 55 60 65 Time (min) Figure A.11: Peanut Skin Fraction B: HPLC-UV Chromatogram (bottom) versus TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in Both SID=1.0 Scan Range (m/z 125-600 M-H-) (top) and SID=2.0 Scan Range (m/z 590-2000 M-H-) (middle)].

327 RT:0.40 - 69.98 53.00 NL: 4.83E7 419.0 Base Peak F: - c ESI sid=1.00 Full ms [ 50 17.08 125.00-600.00] MS 472.9 Negative_Mode_Reed- 45 2005-0913- 23.01 02_Fraction_B 49.11 40 472.7 435.2

35

30 48.18 25 351.0 50.12 53.45 419.2 20 453.2 62.13 65.1665.68 26.53 40.92 45.88 376.0 468.5468.7 59.69 Relative Abundance Relative 593.1 353.2 15 367.0 22.64 27.72 522.8 12.98 20.14 56.51 NL: 5.15E-1 137.0472.8 562.4 33.29 40.46 414.7 24.54 570.5 10 7.89 11.07 467.4 36.35 386.7 44.74 UV Analog 1.65 463.1 462.5137.1 457.0 386.6 Negative_Mode_Reed- 5 261.2 7.43 2005-0913- 463.0 02_Fraction_B 0 1.66 0.0

0.25

0.20

8.74 0.0 17.11 0.15 21.58 0.0 0.0 23.40 27.09 0.0 0.0 Intensity 8.26 53.01 4.09 16.24 27.80 0.10 0.0 11.01 25.73 50.71 0.0 0.0 57.66 0.0 0.0 30.05 35.39 39.1240.85 46.0348.19 58.1762.23 66.92 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.05

0.00 5 10 15 20 25 30 35 40 45 50 55 60 65 Time (min) Figure A.12: Peanut Skin Fraction B: HPLC-UV Chromatogram (bottom) versus TIC [Total Ion Chromatogram – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125-600 M-H-)] (top).

328 RT:0.20 - 69.98

41.95 56.19 NL: 3.83E7 Base Peak F: - c ESI 38.47 1679.2 1578.9 58.53 sid=2.00 Full ms [ 1578.9 774.8 90 590.00-2000.00] MS Negative_Mode_Reed- 60.15 53.49 2005-0913- 80 27.09 43.97 52.58 1578.9 63.7666.54 02_Fraction_B 623.0 1479.21479.3 35.60 1579.0 774.5775.0 70 862.3 45.46 50.29 21.02 1679.1 68.71 60 741.0 1679.3 774.5

33.95 50 19.39 21.49 33.09 1379.5 1.60 903.1 785.2 886.9 1579.4 40 23.31 755.1 27.65 18.84 1975.6

Relative Abundance Relative 30 1627.2 12.32 14.84 NL: 5.15E-1 20 9.27 2.79 1478.81895.4 UV Analog Negative_Mode_Reed- 1679.9 1679.1 10 2005-0913- 02_Fraction_B 0 1.66 0.0

0.45

0.40

0.35

0.30

0.25 In te n sity 0.20 8.74 17.11 21.58 0.0 23.40 0.15 0.0 0.0 27.09 53.01 7.78 16.24 0.0 4.09 11.01 0.0 27.80 50.71 57.66 30.05 35.39 46.0348.19 0.0 58.17 0.10 0.0 0.0 39.1240.85 62.23 66.92 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.05

0.00 5 10 15 20 25 30 35 40 45 50 55 60 65 Time (min) Figure A.13: Peanut Skin Fraction B: HPLC-UV Chromatogram (bottom) versus TIC [Total Ion Chromatogram – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=2.0 Scan Range (m/z 590-2000 M-H-)] (top).

329 RT:0.00 - 70.02 1.72 26.76 NL: 2.04E-1 0.0 0.0 UV Analog 0.150 Negative_Mo de_Reed- 0.145 2005-0913- 03_Fraction_ C 0.140

0.135

48.16 0.130 0.0

0.125

0.120

0.115

23.66 0.110 0.0

0.105 26.48 20.79 0.0 0.100 0.0 Intensity

0.095 54.93 0.0 15.25 0.090 0.0 20.14 0.0 0.085 49.88 8.02 0.0 0.080 0.0 42.69 0.0 57.67 0.075 8.27 44.92 0.0 0.0 13.34 0.0 0.070 0.0 19.38 29.70 34.21 52.87 60.02 0.0 60.87 3.42 12.09 0.0 0.0 39.82 0.0 0.0 67.2969.82 0.065 39.00 0.0 0.0 9.93 0.0 31.64 0.0 0.0 0.0 7.60 0.0 0.0 0.0 0.060 0.0

0.055

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.14: HPLC-UV Chromatogram of Peanut Skin Fraction C (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm.

330 RT:0.00 - 70.02 48.12 NL: 2.54E7 351.1 Base Peak F: - c ESI 100 sid=1.00 Full ms [ 125.00-600.00] MS Negative_Mode_Reed- 80 2005-0913- 47.29 03_Fraction_C 60 42.72 351.1 367.1 43.39 40 56.83 67.85 26.19 34.15 367.1 58.98 62.44 66.35 39.85 570.8 593.0 33.43 468.7 386.8 550.7550.5 1.71 14.42 29.66 329.0 352.9 56.00 16.01 20.0321.29 298.8 39.41 51.15 NL: 4.13E7 Relative Abundance Relative 20 13.37 434.8 8.51 437.0 283.1 339.0 386.7556.6 3.41 398.7 436.9 177.1595.1 Base Peak F: - c ESI 134.1 434.9 sid=2.00 Full ms [ 0 590.00-2000.00] MS 26.79 59.87 Negative_Mode_Reed- 623.1 775.260.31 2005-0913- 100 774.9 03_Fraction_C

80 57.21 23.68 775.0 60.67 609.1 774.9 66.94 60 775.0 61.23 NL: 2.04E-1 56.43 774.7 20.81 26.45 40 UV Analog 741.0 775.0 15.24 622.9 Negative_Mode_Reed- 639.0 31.52 41.64 43.43 14.88 28.52 36.75 46.36 50.77 54.48 2005-0913- 1.77 19.70 R e la tive A b u20 n d a n ce 1954.2 1852.31213.4 03_Fraction_C 7.96 11.47 641.2 1321.5 1510.3 1465.91723.8 1775.6 1979.0 756.9 1375.21099.0 0 1.72 0.0 0.20 26.76 0.0

0.15 48.16 0.0 23.66 20.79 0.0 In te n sity 54.93 15.25 20.14 0.0 49.88 0.10 8.02 42.69 0.0 8.27 0.0 0.0 57.67 13.34 29.70 44.92 0.0 3.42 0.0 34.21 39.0039.82 0.0 60.1262.25 67.2969.82 7.60 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.15: Peanut Skin Fraction C: HPLC-UV Chromatogram (bottom) versus TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in Both SID=1.0 Scan Range (m/z 125-600 M-H-) (top) and SID=2.0 Scan Range (m/z 590-2000 M-H-) (middle)].

331 RT:0.00 - 70.02 48.12 NL: 2.54E7 351.1 Base Peak F: - c ESI 90 sid=1.00 Full ms [ 125.00-600.00] MS Negative_Mode_Reed- 80 2005-0913- 03_Fraction_C 70

47.29 60 351.1 42.72 50 367.1

40 43.39 367.1 67.85 30 26.19 66.35 Relative Abundance Relative 34.15 56.8358.98 39.85 62.44 550.7550.5 593.0 329.0 570.8468.7 33.43 352.9 386.8 NL: 2.04E-1 20 1.71 29.66 298.8 45.87 51.1556.00 UV Analog 14.42 20.0321.29 283.1 39.41 434.8 13.37 16.01 353.2 386.7556.6 Negative_Mode_Reed- 437.0 177.1595.1 339.0 10 8.05 8.51 398.7 436.9 2005-0913- 03_Fraction_C 347.7434.9 0 1.72 0.0 26.76 0.18 0.0

0.16

48.16 0.14 0.0 23.66 0.12 20.79 0.0 54.93 15.25 20.14 0.0 0.10 49.88 0.0 8.02 0.0 0.0 42.69 8.27 44.92 0.0 57.67 0.0 13.34 29.70

In te n sity 34.21 0.0 52.87 60.12 0.08 3.42 18.85 39.82 0.0 60.87 67.2969.82 7.60 0.0 0.0 39.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.06

0.04

0.02

0.00 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.16: Peanut Skin Fraction C: HPLC-UV Chromatogram (bottom) versus TIC [Total Ion Chromatogram – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125-600 M-H-)] (top).

332 RT:0.20 - 70.02 26.79 59.8760.31 NL: 4.13E7 623.1 775.2774.9 Base Peak F: - c ESI sid=2.00 Full ms [ 90 590.00-2000.00] MS Negative_Mode_Reed- 80 2005-0913- 57.21 03_Fraction_C 775.0 70 23.68 60.67 609.1 774.9 66.94 60 775.0

50 61.23 774.7 40 56.43 20.81 26.45 775.0 741.0 622.9 30 55.95 Relative Abundance Relative 15.24 775.0 639.0 NL: 2.04E-1 41.64 43.43 20 31.52 36.75 28.52 46.36 53.88 UV Analog 14.88 15.51 1852.31213.4 1.77 1954.2 1510.3 Negative_Mode_Reed- 641.2 1321.5 1465.9 1252.2 11.47 610.8 10 1979.04.89 9.34 2005-0913- 1099.0 03_Fraction_C 1419.51870.3 0 1.72 0.0 26.76 0.18 0.0

0.16

48.16 0.14 0.0 23.66 0.12 20.79 0.0 54.93 15.25 0.0 0.10 20.14 8.02 49.88 0.0 0.0 0.0 42.69 8.27 57.67 0.0 13.34 44.92 0.0

In te n sity 29.70 0.08 3.42 18.85 34.21 39.82 0.0 52.87 0.0 60.0262.25 67.2969.82 7.60 0.0 0.0 39.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.06

0.04

0.02

0.00 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.17: Peanut Skin Fraction C: HPLC-UV Chromatogram (bottom) versus TIC [Total Ion Chromatogram – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=2.0 Scan Range (m/z 590-2000 M-H-)] (top).

333 RT:0.56 - 69.99 1.72 7.87 13.15 31.7133.57 NL: 2.83E-1 0.0 0.0 0.0 0.0 0.0 UV Analog 0.150 Negative_Mo 7.51 de_Reed- 2005-0913- 0.145 0.0 04_Fraction_ D 0.140 30.56 0.0 0.135

0.130

0.125

0.120 27.39 0.0 0.115

0.110

0.105

0.100 In te n sity

16.45 0.095 8.29 0.0 0.090 0.0 26.92 20.75 23.74 0.0 0.085 0.0 0.0 20.43 0.080 0.0 12.68 33.90 0.075 0.0 0.0 16.04 57.69 48.18 10.58 0.0 19.12 37.83 0.0 0.070 0.0 28.59 0.0 0.0 0.0 36.98 44.93 54.9656.45 2.82 49.91 58.5761.74 3.05 0.0 39.08 42.78 63.3167.3569.88 0.065 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.060

0.055

5 10 15 20 25 30 35 40 45 50 55 60 65 Time (min) Figure A.18: HPLC-UV Chromatogram of Peanut Skin Fraction D (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm.

334 RT:0.00 - 69.99 NL: 3.31E7 31.7533.65 299.0299.0 Base Peak F: - c ESI 100 sid=1.00 Full ms [ 125.00-600.00] MS Negative_Mode_Reed- 80 2005-0913- 13.14 04_Fraction_D 24.42 289.0 30.55 60 463.0 12.65 16.44 570.4 289.0 289.0 16.08 48.26 40 19.48 27.42 11.93 289.0 448.6 461.0 351.1 20.72 37.81 65.25 69.69 7.86 289.0 42.81 47.74 63.93 477.1 313.0 58.1059.81 468.5 468.7 1.75 49.37 56.47 468.9 NL: 5.05E7 Relative Abundance Relative 20 367.0 349.1 6.26 325.8 222.6386.7 289.0 386.7 550.4 Base Peak F: - c ESI 289.0 sid=2.00 Full ms [ 0 590.00-2000.00] MS 60.69 Negative_Mode_Reed- 1436.9 2005-0913- 100 04_Fraction_D 63.60 725.8 80

60 60.34 42.30 774.9 64.0865.77 NL: 2.83E-1 58.78 726.7 40 1547.8 775.0 UV Analog 775.0 23.87 53.13 Negative_Mode_Reed- 32.64 34.19 37.27 50.85 17.94 21.03 609.1 26.86 39.78 45.08 1578.8 2005-0913- 1.80 16.60 1519.9 928.8 1679.4 Relative Abundance Relative 20 13.29 619.3 622.8 1417.0 04_Fraction_D 5.72 10.12 1945.91479.3 1562.2 1023.3 1993.6 1579.0 1679.31768.7 0 1.72 0.0 31.71 33.57 0.0 0.25 0.0 7.87 0.0 13.15 0.20 0.0 7.51 30.56 0.0 0.15 27.39 0.0 In te n sity 16.45 0.0 8.29 20.75 26.92 0.0 33.90 57.69 0.10 2.82 6.93 0.0 0.0 0.0 28.59 37.83 42.7844.93 48.1849.91 56.45 58.5761.74 67.3569.88 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0 5 10 15 20 25 30 35 40 45 50 55 60 65 Time (min) Figure A.19: Peanut Skin Fraction D: HPLC-UV Chromatogram (bottom) versus TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in Both SID=1.0 Scan Range (m/z 125-600 M-H-) (top) and SID=2.0 Scan Range (m/z 590-2000 M-H-) (middle)].

335 RT:0.07 - 69.99 31.7533.65 NL: 3.31E7 299.0299.0 Base Peak F: - c ESI sid=1.00 Full ms [ 80 125.00-600.00] MS Negative_Mode_Reed- 2005-0913- 70 04_Fraction_D 13.14 289.0 60 24.42 463.0 30.55 50 12.65 16.44 570.4 289.0 289.0

40 12.29 27.42 16.08 19.48 461.0 48.26 30 288.9 289.0 448.6 351.1 69.69 Relative Abundance Relative 37.81 65.25 20.72 468.7 313.0 63.93468.5 NL: 2.83E-1 20 7.86 477.1 42.81 47.74 11.43 58.1059.81 468.9 38.07 367.0 349.1 UV Analog 325.8289.0 49.37 56.47222.6386.7 1.75 34.71 283.0 Negative_Mode_Reed- 10 386.7 550.4 289.0 6.26 313.0 2005-0913- 289.0 04_Fraction_D 0 1.72 31.7133.57 0.0 0.0 0.0 7.87 0.22 0.0 0.20 13.15 0.0 0.18 7.51 0.16 30.56 0.0 0.0 0.14 27.39 0.0 0.12 8.29 16.45 20.75 26.92 In te n0.10 sity 0.0 0.0 33.90 0.0 0.0 37.83 48.18 57.69 2.82 28.59 44.93 49.91 56.45 58.5761.74 0.08 3.46 0.0 42.78 67.2569.88 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.06

0.04

0.02

0.00 5 10 15 20 25 30 35 40 45 50 55 60 65 Time (min) Figure A.20: Peanut Skin Fraction D: HPLC-UV Chromatogram (bottom) versus TIC [Total Ion Chromatogram – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125-600 M-H-)] (top).

336 RT:0.07 - 69.99 60.69 63.60 NL: 5.05E7 1436.9725.8 Base Peak F: - c ESI sid=2.00 Full ms [ 590.00-2000.00] MS 55 Negative_Mode_Reed- 2005-0913- 50 60.34 04_Fraction_D 774.9 45 42.30 40 64.08 1547.8 726.765.77 775.0 35 59.99 775.8 57.90 30 755.0 25 23.87 53.13 20 609.1 50.851578.8 32.64 34.19 37.27 Relative Abundance Relative 26.86 45.08 1679.4 1417.0928.839.78 15 17.94 21.03 622.8 1519.9 NL: 2.83E-1 16.60 619.3 1023.3 1.80 1479.3 1562.2 UV Analog 10 13.29 1945.9 1993.6 10.12 Negative_Mode_Reed- 5.72 1579.0 2005-0913- 5 1768.7 1679.3 04_Fraction_D 0 1.72 7.87 13.15 31.7133.57 0.0 0.0 0.0 0.0 0.0

0.16 7.51 0.0 30.56 0.14 0.0 27.39 0.12 0.0

8.29 16.45 0.10 20.7523.7426.92 0.0 0.0 0.0 0.0 0.0 33.90 12.21 16.04 19.41 57.69 28.59 37.83 48.18 0.08 2.82 0.0 42.7844.93 49.91 56.45 58.5761.74 67.3569.88 6.93 0.0 0.0 0.0 0.0 In te n sity 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.06

0.04

0.02

0.00 5 10 15 20 25 30 35 40 45 50 55 60 65 Time (min) Figure A.21: Peanut Skin Fraction D: HPLC-UV Chromatogram (bottom) versus TIC [Total Ion Chromatogram – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=2.0 Scan Range (m/z 590-2000 M-H-)] (top).

337 RT:0.06 - 70.04 1.68 NL: 2.52E-1 0.0 UV Analog Negative_Mo de_Reed- 0.150 2005-0913- 05_Fraction_ 0.145 E

0.140

0.135

0.130

0.125

0.120

0.115 29.29 0.110 0.0

0.105

In te n0.100 sity

0.095

0.090

0.085

0.080

0.075 27.06 25.02 17.05 21.98 0.0 16.45 57.71 0.070 21.57 0.0 0.0 0.0 33.12 0.0 37.3939.32 56.89 0.0 0.0 61.2364.8667.4269.60 13.22 0.0 52.2954.50 0.065 0.0 0.0 49.60 0.0 45.0846.49 0.0 0.0 0.0 0.0 0.0 0.0 0.0 11.68 0.0 7.90 0.0 0.0 0.060 7.36 0.0 2.76 0.0 0.0 0.0 0.055

5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.22: HPLC-UV Chromatogram of Peanut Skin Fraction E (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm.

338 RT:0.00 - 70.04 63.62 NL: 1.54E7 588.6 Base Peak F: - c ESI 100 sid=1.00 Full ms [ 125.00-600.00] MS Negative_Mode_Reed- 80 2005-0913- 05_Fraction_E 60 25.90 66.2968.67 346.6 468.4468.7 42.91 48.38 53.83 62.79 40 21.97 29.32 39.31 56.13 59.27 18.72 311.1 386.6 386.7 550.7 450.7 301.0 462.6 468.5 550.7 13.16 348.7 31.8233.76 47.19 51.93 13.59 299.1 NL: 5.14E7 Relative Abundance Relative 20 1.68 12.40288.8 299.0 386.6 386.7 7.36 9.81 591.0 Base Peak F: - c ESI 289.0 573.0 325.6595.5 sid=2.00 Full ms [ 0 590.00-2000.00] MS 57.77 Negative_Mode_Reed- 740.0 2005-0913- 100 05_Fraction_E 57.49 739.8 80 58.28 0.18 739.6 60 1557.6 57.20 739.5 62.54 66.43 NL: 2.52E-1 40 774.9 775.0 UV Analog 24.16 30.97 33.81 44.00 Negative_Mode_Reed- 35.02 39.08 53.02 2005-0913- 15.7716.52 18.48 1638.4 29.49 1781.91427.4 998.5 47.11 R e la tive A b u20 n d a n ce 1.75 11.71 05_Fraction_E 9.42 862.71148.11647.8 1179.6 1931.71875.0 1684.1 1670.6 1102.5 1279.31646.4 0 1.68 0.0 0.25

0.20

0.15 29.29 In te n sity 0.0

0.10 17.05 21.9825.0227.06 13.2215.79 33.12 37.3939.32 50.2353.8756.8957.71 61.23 67.4269.60 6.407.9011.18 43.4645.7548.53 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.23: Peanut Skin Fraction E: HPLC-UV Chromatogram (bottom) versus TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in Both SID=1.0 Scan Range (m/z 125-600 M-H-) (top) and SID=2.0 Scan Range (m/z 590-2000 M-H-) (middle)].

339 RT:0.00 - 70.04 63.62 NL: 1.54E7 588.6 Base Peak F: - c ESI sid=1.00 Full ms [ 50 125.00-600.00] MS Negative_Mode_Reed- 25.90 2005-0913- 45 346.6 05_Fraction_E 66.2968.67 40 468.4468.7

35 62.79 30 29.32 550.7 301.0 59.27 25 56.13 550.7 18.72 21.97 27.10 39.31 42.91 468.5 450.7 285.1 311.1 20 348.7 462.6 33.76 48.38 53.83 13.16 31.82 47.19 53.24 Relative Abundance Relative 299.1 386.6 386.7 15 288.8 299.0 38.26 386.6 304.6 NL: 2.52E-1 13.59 462.9 25.31 UV Analog 10 591.0 1.68 12.40 462.5 Negative_Mode_Reed- 7.97 9.81 289.0 5 573.0 2005-0913- 326.0595.5 05_Fraction_E 0 1.68 0.0

0.12 29.29 0.0

0.10

27.06 0.08 16.0817.05 21.9825.02 57.71 0.0 33.12 37.3939.32 56.89 61.23 13.22 0.0 48.5351.7954.50 62.5167.4269.60 11.18 0.0 0.0 0.0 42.7645.75 0.0 6.407.90 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.06 0.0 0.0 Intensity

0.04

0.02

0.00 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.24: Peanut Skin Fraction E: HPLC-UV Chromatogram (bottom) versus TIC [Total Ion Chromatogram – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125-600 M-H-)] (top).

340 RT:0.33 - 70.01 1.70 NL: 5.53E-1 0.0 UV Analog 0.15 Negative_Mo de_Reed- 2005-0913- 06_Fraction_ 0.14 F

0.13

0.12

0.11 11.42 18.30 0.0 0.0 0.10

0.09 16.83 22.90 0.0 10.04 0.0 0.08 16.50 0.0 19.06 57.72 0.0 0.0 23.66 0.0 In te n sity 0.07 27.28 6.89 0.0 33.65 52.9454.98 59.26 64.8867.2269.39 0.0 28.96 35.72 39.23 51.85 42.9545.2449.82 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 6.26 0.0 0.0 0.0 0.0 0.0 0.06 0.0

0.05

0.04

0.03

0.02

0.01

0.00 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.25: HPLC-UV Chromatogram of Peanut Skin Fraction F (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm.

341 RT:0.33 - 70.01 11.43 NL: 1.90E7 576.8 18.29 Base Peak F: - c ESI 572.9 sid=1.00 Full ms [ 80 11.16 125.00-600.00] MS 576.9 Negative_Mode_Reed- 2005-0913-06_Fraction_F 1.76 60 576.9 10.96 69.42 12.89 65.56 550.5 576.9 16.80 40 576.9 65.17386.6 572.9 19.04 23.03 62.67 573.0 550.5 9.86 574.923.60 49.27 60.08386.5 44.1344.73 53.2356.65 576.8 575.0 38.85 386.7 386.6 20 6.92 33.5335.50 304.5325.1 468.5386.6 NL: 2.46E7 Relative Abundance Relative 26.43 31.86 544.9 4.47 444.6304.8 574.9 584.9 447.1 Base Peak F: - c ESI 576.8 sid=2.00 Full ms [ 0 67.43 590.00-2000.00] MS 40.74 Negative_Mode_Reed- 18.33 59.29 63.76 775.4 656.5 2005-0913-06_Fraction_F 1146.4 58.40774.8 774.9 774.9 80 11.38 1154.7

60 57.63 775.0

NL: 5.53E-1 40 22.98 56.33 41.95 47.57 UV Analog 17.50 1694.2 1.71 1150.625.28 49.19 Negative_Mode_Reed- 13.75 19.98 31.93 33.90 36.18 1644.6 1551.9 1155.7 1969.6 1884.9 2005-0913-06_Fraction_F 20 10.88 590.8 1346.5 1091.5 1653.71766.41945.7 R e la tive A b u n d a n ce 6.09 8.82 658.0 1717.9727.4 0 1.70 0.0

0.4

0.3

0.2 In te n sity 11.42 18.30 10.04 16.83 22.90 57.72 6.666.89 0.0 0.0 23.6627.28 31.9433.6535.72 39.23 42.9545.2448.0751.1052.9454.98 59.26 64.8867.2269.39 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.26: Peanut Skin Fraction F: HPLC-UV Chromatogram (bottom) versus TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in Both SID=1.0 Scan Range (m/z 125-600 M-H-) (top) and SID=2.0 Scan Range (m/z 590-2000 M-H-) (middle)].

342 RT:0.46 - 70.01 11.43 18.29 NL: 1.90E7 576.8 572.9 Base Peak F: - c ESI sid=1.00 Full ms [ 11.16 125.00-600.00] MS 576.9 70 Negative_Mode_Reed- 2005-0913- 06_Fraction_F 60 1.76 576.9 50

10.96 69.42 576.9 550.5 40 12.89 65.56 576.9 16.80 386.6 10.72 572.9 576.9 19.04 23.03 65.17 30 62.67 573.0 574.9 550.5 386.5 Relative Abundance Relative 9.86 23.60 49.27 60.08 20 575.0 386.6 576.8 22.34 44.73 386.7 53.2356.65 NL: 5.53E-1 386.6 575.0 38.85 44.13325.1 468.5 UV Analog 9.46 33.5335.50 6.92 31.86 544.9 304.5 Negative_Mode_Reed- 10 25.2226.88 444.6 576.9 447.1 304.8 2005-0913- 574.9 187.0575.1 06_Fraction_F 0 1.70 0.0

0.40

0.35

0.30

0.25

0.20 Intensity

0.15 11.42 18.30 0.0 16.83 0.0 22.90 10.04 23.66 57.72 0.10 6.89 0.0 27.28 31.9433.6535.72 39.2342.9545.2448.0751.1052.9454.98 59.26 64.8867.2269.39 6.26 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.05

0.00 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.27: Peanut Skin Fraction F: HPLC-UV Chromatogram (bottom) versus TIC [Total Ion Chromatogram – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125-600 M-H-)] (top).

343 RT:0.33 - 70.04 1.72 16.04 NL: 5.95E-1 0.0 0.0 UV Analog Negative_Mod 0.15 e_Reed- 2005-0913- 08_Fraction_ 0.14 G-red

0.13

18.04 0.12 0.0

0.11

0.10

0.09 18.38 12.92 0.0 20.15 0.0 38.18 0.0 0.08 0.0 11.24 9.91 57.74 Intensity 0.0 22.16 34.63 54.98 0.07 0.0 27.9429.5430.15 39.82 0.0 59.53 52.1854.44 0.0 63.8366.7968.30 0.0 0.0 42.9144.89 49.71 9.62 0.0 0.0 0.0 0.0 0.0 8.54 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.46 0.0 0.06 0.0 0.0

0.05

0.04

0.03

0.02

0.01

0.00 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.28: HPLC-UV Chromatogram of Peanut Skin Fraction G-Red (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm.

344 RT:0.27 - 70.04 16.0018.01 NL: 4.06E7 576.9572.9 Base Peak F: - c ESI sid=1.00 Full ms [ 13.04 125.00-600.00] MS 576.9 Negative_Mode_Reed- 60 2005-0913-08_Fraction_G- red 12.60 576.9 19.1720.23 40 572.9576.8 1.75 576.8 12.02 21.37 67.89 576.9 64.0364.54 573.0 38.21 468.6 20 9.61 30.22 33.80 52.6653.10 61.68386.6386.6 29.50 555.039.79 44.8846.77 57.87 NL: 2.66E7 R e la tive A b u n d a n ce 7.35 584.9 444.7 386.7386.6 550.5 576.7 584.9 555.2 325.3386.5 386.7 Base Peak F: - c ESI 576.9 sid=2.00 Full ms [ 0 590.00-2000.00] MS 12.91 16.46 69.32Negative_Mode_Reed- 1154.5 1150.5 60.29 775.1 66.77 2005-0913-08_Fraction_G- 775.1 774.9 red 58.15 63.30 774.8 775.2 60 18.36 1862.5 38.38 57.93 775.0 40 13.97 34.44 1939.4 590.6 20.09 56.33 NL: 5.95E-1 24.49 27.81 30.42 40.87 1721.4 46.09 1.67 1153.7 49.92 775.0 UV Analog 1803.01240.41432.3 1562.1 12.18 1775.5 Negative_Mode_Reed- 1392.0 1814.6 20 9.93 2005-0913-08_Fraction_G- 1154.6 Relative Abundance Relative 2.22 1857.3 red 1582.0 0 1.72 0.0

0.4

0.3

16.04 0.2

Intensity 0.0 18.04 11.2412.92 0.0 19.18 38.18 2.46 6.868.97 22.16 27.9429.5431.7234.63 39.82 44.8948.3651.1254.4454.98 57.7459.53 63.8366.7968.30 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.29: Peanut Skin Fraction G-Red: HPLC-UV Chromatogram (bottom) versus TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in Both SID=1.0 Scan Range (m/z 125-600 M-H-) (top) and SID=2.0 Scan Range (m/z 590-2000 M-H-) (middle)].

345 RT:0.07 - 70.04 13.04 16.0018.01 NL: 4.06E7 576.9 576.9572.9 Base Peak F: - c ESI sid=1.00 Full ms [ 55 125.00-600.00] MS Negative_Mode_Reed- 50 2005-0913- 08_Fraction_G-red 45 12.60 40 576.9 19.17 572.920.23 576.8 35 12.36 30 1.75 576.9 576.8 21.12 25 12.02 573.0 576.9 20 21.37 64.54 67.89

Relative Abundance Relative 573.0 38.21 64.03386.6 468.6 15 11.27 555.0 386.6 30.22 61.68 NL: 5.95E-1 576.9 33.80 52.6653.10 58.56 9.61 584.9 46.77 550.5 UV Analog 10 29.50 39.79 44.88 386.7 386.6 23.37 444.734.66 386.5 386.6 Negative_Mode_Reed- 576.7 584.9 555.2 325.3 6.43 574.9 558.6 2005-0913- 5 576.8 08_Fraction_G-red 0 1.72 0.0

0.30

0.25

0.20 16.04 0.0

In te n0.15 sity 18.04 0.0 18.38 12.92 38.18 11.24 21.15 0.10 9.91 0.0 27.9429.54 34.63 39.82 54.9857.7459.53 63.8366.7968.30 8.00 0.0 31.72 0.0 44.8948.3651.1254.44 2.46 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.05

0.00 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.30: Peanut Skin Fraction G-Red: HPLC-UV Chromatogram (bottom) versus TIC [Total Ion Chromatogram – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125-600 M-H-)] (top).

346 RT:0.56 - 70.07 1.76 16.38 NL: 9.95E-1 0.0 0.0 UV Analog Negative_Mo de_Reed- 2005-0913- 07_Fraction_ 0.50 G

0.45 19.19 0.0

0.40

0.35

0.30

In te n sity 15.34 0.25 0.0

12.7214.33 0.0 0.20 0.0 19.95 0.0

0.15 20.48 11.78 10.79 0.0 0.0 22.12 0.0 0.0 24.58 0.10 27.05 9.49 0.0 30.18 8.05 31.6735.85 39.2840.62 45.7746.9350.4552.8356.1056.46 61.8464.3965.55 69.96 6.05 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.05

0.00 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.31: HPLC-UV Chromatogram of Peanut Skin Fraction G (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm.

347 RT:0.00 - 70.07 19.91 NL: 1.70E7 574.9 Base Peak F: - c ESI 16.35 sid=1.00 Full ms [ 574.9 125.00-600.00] MS 80 Negative_Mode_Reed- 2005-0913- 16.09 23.51 07_Fraction_G 575.1 60 588.9 63.76 15.73 44.71 68.56 150.3 325.1 550.4 574.9 43.10 62.73 40 13.05 60.62 311.1 386.4 574.9 30.07 37.16 52.75 386.5 11.57 39.35 46.35 57.88 394.6 468.5 50.31386.6 574.8 28.30 33.63 339.2 20 1.84 423.0 386.6 468.6 NL: 1.04E8 Relative Abundance Relative 8.73 577.2 444.6 Base Peak F: - c ESI 575.2 577.0 sid=2.00 Full ms [ 0 590.00-2000.00] MS 15.86 Negative_Mode_Reed- 862.8 2005-0913- 07_Fraction_G

80 16.30 1150.7 60 1.80 862.9 15.50 19.17 12.81862.8 1150.6 NL: 9.95E-1 40 862.8 UV Analog 19.95 67.60 11.41 59.79 67.31 Negative_Mode_Reed- 9.47 1151.6 58.29 62.67 775.1 862.8 25.25 30.57 38.56 41.29 774.9 774.9 2005-0913- 20 35.47 43.14 46.54 52.81 774.9 774.9

Relative Abundance Relative 862.8 07_Fraction_G 7.30 1788.5 1821.6 1583.81352.8 663.1 1958.0 1593.71598.8 863.0 0 1.76 16.38 0.0 0.0 15.85 0.8 0.0

0.6 19.19 0.0 0.4

Intensity 15.34 12.72 0.0 19.95 0.0 10.79 0.0 21.7924.58 0.2 5.858.90 26.5230.1832.4835.85 39.2840.62 46.9349.9052.6954.7656.46 60.5361.84 65.55 69.96 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.32: Peanut Skin Fraction G: HPLC-UV Chromatogram (bottom) versus TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in Both SID=1.0 Scan Range (m/z 125-600 M-H-) (top) and SID=2.0 Scan Range (m/z 590-2000 M-H-) (middle)].

348 RT:0.07 - 70.07 16.35 19.91 NL: 1.70E7 574.9 574.9 Base Peak F: - c ESI sid=1.00 Full ms [ 125.00-600.00] MS Negative_Mode_Reed- 70 2005-0913- 07_Fraction_G

60 16.09 23.51 575.1 588.9

50 21.53 63.76 44.71 574.8 150.3 15.73 325.1 68.56 550.4 40 574.9 43.10 62.73 13.05 311.1 60.62386.4 30 574.9 386.5

Relative Abundance Relative 39.35 46.35 11.57 50.31 55.8758.57 423.0 339.2 NL: 9.95E-1 20 10.47574.8 30.07 386.652.75 222.7527.5 33.63 37.16 386.6 154.4 28.30394.6 UV Analog 1.84 444.6 468.5 577.2 Negative_Mode_Reed- 10 575.2 6.11 2005-0913- 574.9 07_Fraction_G 0 1.76 16.38 0.0 0.0 15.85 0.0 0.7

0.6

0.5 19.19 0.0 0.4 In te n sity 0.3 15.34 12.72 0.0 19.95 0.0 0.2 0.0 11.38 21.79 9.96 24.5826.52 2.887.74 0.0 0.0 30.1832.2635.85 39.2840.62 46.9349.9052.4054.7656.46 60.5361.84 65.55 69.96 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.33: Peanut Skin Fraction G: HPLC-UV Chromatogram (bottom) versus TIC [Total Ion Chromatogram – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125-600 M-H-)] (top).

349 RT:0.00 - 70.07 15.86 NL: 1.04E8 862.8 Base Peak F: - c ESI sid=2.00 Full ms [ 70 590.00-2000.00] MS Negative_Mode_Reed- 16.30 2005-0913- 07_Fraction_G 60 1150.7 1.80

50 862.9 15.50 19.17 862.8 1150.6 40 12.8114.31 862.8862.8

30 19.95 11.61 67.60 1151.6 775.1 Relative Abundance Relative 62.6767.31 863.0 58.2959.79 20 9.47 774.9774.9774.9 NL: 9.95E-1 21.13 25.25 30.57 774.9 862.8 864.8 35.47 38.56 41.29 52.81 UV Analog 1788.5 1821.6 43.14 46.54 1958.01583.81352.8 663.1 Negative_Mode_Reed- 10 8.08 1593.7 2.43 1598.8 2005-0913- 864.9 07_Fraction_G 862.5 0 1.76 16.38 15.85 0.0 0.0 0.0 0.7

0.6

0.5 19.19 0.0 0.4

In te n0.3 sity 15.34 12.72 0.0 19.95 0.0 0.2 0.0 10.7911.78 21.79 24.58 8.05 0.0 0.0 0.0 26.5230.1831.6735.8539.2840.62 46.9349.9052.6954.7656.46 60.5361.84 65.55 69.96 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.0 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.34: Peanut Skin Fraction G: HPLC-UV Chromatogram (bottom) versus TIC [Total Ion Chromatogram – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=2.0 Scan Range (m/z 590-2000 M-H-)] (top).

350 RT:0.07 - 70.06 1.79 NL: 9.95E-1 0.0 UV Analog Negative_Mo de_Reed- 2005-0913- 0.50 09_Fraction_ H

0.45

0.40

0.35 19.35 0.0

0.30 18.77 15.86 0.0 12.08 0.0 0.0 In te n0.25 sity

15.47 0.0 0.20

0.15 21.65 0.0 21.96 11.33 0.0 24.37 0.10 0.0 27.17 9.93 29.32 56.5057.78 0.0 33.19 38.1739.22 45.26 54.99 63.0464.25 67.44 8.06 0.0 43.79 49.9252.06 6.02 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.05

0.00 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.35: HPLC-UV Chromatogram of Peanut Skin Fraction H (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm.

351 RT:0.07 - 70.06 12.11 64.86 NL: 1.54E7 575.2 277.2 Base Peak F: - c ESI sid=1.00 Full ms [ 44.85 125.00-600.00] MS 43.53 325.3 Negative_Mode_Reed- 311.1 45.32 60 325.2 2005-0913- 09_Fraction_H 68.61 11.50 550.7 63.90 40 575.0 13.88 46.36 14.98 60.97386.7 575.1 17.74 339.2 10.66 575.2 468.7 1.71 574.2 56.30 574.9 30.21 33.62 46.8849.38 59.33 575.6 20.68 25.35 42.66 550.6 20 8.92 394.9 444.6 37.64 506.8491.1 550.3 574.1 556.0 386.6 311.2 NL: 5.07E7 Relative Abundance Relative 6.84 574.9 Base Peak F: - c ESI 575.5 sid=2.00 Full ms [ 0 13.93 15.85 590.00-2000.00] MS 19.41 Negative_Mode_Reed- 862.8 1150.8 861.1 2005-0913- 09_Fraction_H

11.97 19.64 60 18.74 1151.0 874.8 860.8

58.93 22.95 59.63 64.48 67.84 40 860.8 775.0775.0 774.8 NL: 9.95E-1 56.48 774.7 1.75 11.25 23.50 29.85 32.80 40.98 775.1 UV Analog 1150.8 1150.9 1548.2 1454.1 36.51 Negative_Mode_Reed- 1726.1 1575.1 46.80 50.45 20 9.92 2005-0913- 1862.2 1860.9 Relative Abundance Relative 1734.6 09_Fraction_H 7.96 862.8 1034.6 0 1.79 0.0

0.6

19.35 0.4 18.77 12.0815.86 0.0 0.0 In te n sity 0.0 0.0 21.65 11.33 21.96 0.2 9.93 24.9828.3630.82 38.17 45.26 56.5057.78 3.50 8.06 0.0 33.19 44.88 49.9252.0654.99 63.0464.25 67.44 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.36: Peanut Skin Fraction H: HPLC-UV Chromatogram (bottom) versus TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in Both SID=1.0 Scan Range (m/z 125-600 M-H-) (top) and SID=2.0 Scan Range (m/z 590-2000 M-H-) (middle)].

352 RT:0.07 - 70.06 12.11 64.86 NL: 1.54E7 575.2 277.2 Base Peak F: - c ESI 44.85 sid=1.00 Full ms [ 325.3 125.00-600.00] MS 60 43.53 45.32 Negative_Mode_Reed- 311.1 325.2 2005-0913- 09_Fraction_H 50 68.61 550.7 11.50 40 575.0 13.88 46.36 63.90 575.1 339.2 386.7 14.98 11.10 17.74 60.97 575.2 30 575.1 574.2 468.7

1.71 60.54 18.6720.68 42.66 10.32 30.21 33.62 46.8849.38 55.7356.30 386.6 Relative Abundance Relative 20 575.6 572.7574.1 311.2 575.3 394.9 444.6 506.8491.1 386.5550.6 25.35 37.64 NL: 9.95E-1 8.64 556.027.72 30.66 386.6 544.7 591.5 10 6.05 575.1 575.6 de_Reed-

0 1.79 0.0 UV Analog Negative_Mo 2005-0913- 0.6 09_Fraction_H

0.5

0.4 19.35 18.77 0.0 15.86 0.3 12.08 0.0 In te n sity 0.0 0.0

0.2 20.24 22.95 11.33 24.37 9.93 0.0 26.98 56.50 3.157.83 0.0 30.8233.19 38.17 44.8845.26 49.9252.2654.99 57.7863.0464.25 67.44 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.37: Peanut Skin Fraction H: HPLC-UV Chromatogram (bottom) versus TIC [Total Ion Chromatogram – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125-600 M-H-)] (top).

353 RT:0.07 - 70.06 13.93 19.41 NL: 5.07E7 1150.815.85861.1 Base Peak F: - c ESI 80 862.8 sid=2.00 Full ms [ 590.00-2000.00] MS Negative_Mode_Reed- 70 2005-0913- 09_Fraction_H

60 11.97 19.64 1151.0 18.74 874.8 860.8 50

40 67.84 22.95 58.9359.63 64.48 11.25 860.8 56.48775.0 774.8 775.1 775.0 774.7 1150.9 30 1.75 23.50 29.85 32.80

Relative Abundance Relative 40.98 1150.8 1548.2 1454.11726.1 20 36.51 1575.1 56.25 10.37 46.80 50.45 775.1 NL: 9.95E-1 1862.2 1152.2 1860.9 1734.6 10 7.96 ode_Reed- 1034.6 0 1.79 0.0 UV Analog 0.8 Negative_M 2005-0913- 09_Fraction_H 0.7

0.6

0.5

0.4 19.35

In te n sity 18.77 0.0 12.0815.86 0.3 0.0 0.0 0.0

0.2 21.65 21.96 11.33 24.98 3.507.839.93 0.0 28.3630.8233.19 38.17 44.8845.26 49.9252.0654.9956.5057.7863.0464.25 67.44 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (min) Figure A.38: Peanut Skin Fraction H: HPLC-UV Chromatogram (bottom) versus TIC [Total Ion Chromatogram – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=2.0 Scan Range (m/z 590-2000 M-H-)] (top).

354 RT:0.00 -105.11 31.57 NL: 1.84E6 179.1 Base Peak F: - c ESI 100 sid=1.00 Full ms [ 125.00-250.00] MS 80 3.79 Negative_Mode_Reed- 39.14 51.32 2009-0514-03_Std_Mix 60 194.9 15.22 163.1 227.2 54.25 169.1 14.64 41.41 227.256.18 83.9087.44 103.53 40 23.29 32.79 75.90 88.7792.42 100.09 169.2 26.47 193.1 227.362.61 222.9222.9 137.3 179.1 68.40 140.880.05 222.8222.9 222.8 222.8 5.60 141.028.97 47.75 140.9 20 217.0 222.8 NL: 8.38E6 244.5 179.2 217.0

Relative Abundance Relative F: - c ESI 33.34 ll ms [ 289.0 Base Peak 100 sid=2.00 Fu ed- 230.00-410.00] MS 80 ix 19.84 Negative_Mode_Re 26.25 19.32305.1 62.15 67.62 2009-0514-03_Std_M 60 305.1 305.0 301.1 285.2 40 54.61 68.27 3.81 31.73 84.7085.79 286.8 285.2 89.46 15.41 367.2 35.0738.88 51.34 55.62 79.50 373.3 91.4896.26100.89 20 330.7 5.03 41.59 329.1 271.3 NL: 1.15E7 347.1 289.1331.0 286.5 286.7 299.1 304.8304.8 311.2 331.0 386.9 Relative Abundance Relative Base Peak F: - c ESI 100.20 sid=3.00 Fu 400.00-2000.00]ll ms [ MS 652.5 Negative_Mode_Re 100 99.41 2009-0514-03_Std_M 712.0 ed- 80 ix 37.53 55.53 60 98.13 914.739.37 43.18 447.1 652.8 882.7 40 22.55 440.9 88.0888.70 3.08 19.87577.0 32.31 575.1873.8 26.29 60.6461.36 84.20 1459.1 610.8 456.9 50.06 73.18 20 6.11 13.54 610.8 68.82 579.3 NL: 3.52E-1 447.1 431.0431.1 411.0 1809.1 1137.6 1837.0 Relative Abundance Relative UV Analog 50.99 Negative_Mode_Reed- 39.03 2009-0514-03_Std_Mix 37.32 0.0 31.53 0.0 0.0 0.0 0.3 42.94 0.0 15.07 0.2 26.33 54.16 0.0

In te n sity 67.41 19.72 0.0 0.0 55.43 64.27 75.98 4.08 44.93 85.1288.6190.9894.96 100.93 12.06 0.0 0.0 70.39 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 Time (min)

Figure A.39: Standard Mix (HPLC Method 2): HPLC-UV Chromatogram (bottom) versus TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125-250 M-H-) (top), SID=2.0 Scan Range (m/z 230-410 M-H-) (second), and SID=3.0 Scan Range (m/z 400-2000 M-H-) (third)].

355 RT:0.00 - 105.13 80.41 85.71 NL: 9.95E-1 0.0 0.0 UV Analog 0.25 Negative_Mo de_Reed- 0.24 2009-0514- 80.99 01_Fraction_ 0.0 A 0.23

0.22

0.21 90.67 0.0 0.20

0.19 90.30 0.0 0.18

0.17 69.27 3.45 0.0 0.16 0.0 90.93 0.15 0.0 Intensity 76.70 0.14 0.0

0.13

0.12 102.72 76.17 96.01 0.11 0.0 0.0 0.0 0.10 17.77 91.51 0.0 71.52 0.0 4.23 0.09 68.55 32.4436.2338.94 56.30 0.0 0.0 23.00 28.11 45.61 58.53 0.0 55.48 0.0 0.0 0.0 0.0 0.0 0.0 0.0 98.18 0.08 0.0 17.01 0.0 0.0 14.26 0.0 0.07 0.0 5.96 0.06 0.0

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 Time (min)

Figure A.40: HPLC-UV Chromatogram of Peanut Skin Fraction A (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm using HPLC Method 2.

356 RT:0.00 - 105.13 50.81 NL: 1.29E7 187.2 Base Peak F: - c ESI 60 sid=1.00 Full ms [ 125.00-250.00] MS 3.57 4.08 Negative_Mode_Reed- 220.8 2009-0514-01_Fraction_A 40 239.0 54.5457.53 7.97 24.93 34.35 241.1169.1 93.77 21.67221.0 68.10 239.0 16.72 225.1 60.30 75.98 86.72 151.0 12.34 220.9 40.05 45.3749.51 215.3 85.98 20 220.9 201.265.21 224.9 93.12 238.9 145.1243.1 239.3218.0 30.52 173.2 213.1 153.2 101.47 NL: 2.62E7 220.9 222.9

Relative Abundance Relative F: - c ESI 0 86.02 74.63 82.95 ll ms [ 3.59 329.2 295.2 93.27 311.1 Base Peak 400.8 71.12 340.5 60 327.2 sid=2.00 Fu ed- 87.14 230.00-410.00] MSon_A 313.1 Negative_Mode_Re 93.65 69.45 2009-0514-01_Fracti 40 4.00 340.4 325.1 78.91 400.8 329.3 54.56 95.46 47.9653.87 20 4.83 7.84 24.64 31.72 241.1 59.4262.41 340.7100.72 18.0121.89 38.1342.21 261.1243.2 343.2 NL: 2.89E7 400.8239.0 395.0 377.2 343.2 300.9395.2 389.2375.0 355.4

Relative Abundance Relative Base Peak F: - c ESI 0 sid=3.00 Fu 84.59 91.2293.19 97.98 103.08 ll ms [ 400.00-2000.00] MS 1645.8 886.6525.3 496.4 772.2 Negative_Mode_Re 60 2009-0514-01_Fractioed- 89.20 99.34 n_A 413.2 593.3 40

74.6775.36 84.36 4.53 45.79 51.17 70.57681.3681.4 609.1 20 3.62 33.99 39.18 58.02 63.71 400.9 33.18 578.9 520.0 557.1 16.1118.44 24.26 474.9 595.0 881.0 718.9 NL: 9.95E-1 1547.1 475.0 443.1456.9 415.0 UV Analog Relative Abundance Relative 0 Negative_Mode_Reed- 85.71 2009-0514-01_Fraction_A 0.0 0.6

84.89 80.41 0.4 0.0 0.0 90.67 69.27 3.45 76.70 0.0 96.01 102.72 Intensity 4.23 17.77 23.00 28.11 32.4436.2338.94 45.61 56.3058.53 68.55 0.0 0.2 0.0 12.8517.01 52.59 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.0 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 Time (min)

Figure A.41: Peanut Skin Fraction A (HPLC Method 2): HPLC-UV Chromatogram (bottom) versus TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125- 250 M-H-) (top), SID=2.0 Scan Range (m/z 230-410 M-H-) (second), and SID=3.0 Scan Range (m/z 400-2000 M-H-) (third)].

357 RT:0.00 - 104.99 3.37 NL: 9.95E-1 0.0 UV Anal og 46.96 Negativ e_Mo 0.24 0.0 de_Reed- 2009-0513- 07_Fraction_ 0.22 B 89.45 19.91 0.0 23.22 0.20 0.0 0.0

30.18 46.42 32.74 0.18 0.0 0.0 50.87 0.0 0.0

0.16 16.8417.83 87.94 0.0 0.0 42.99 0.0 42.28 0.0 35.35 57.16 28.41 0.14 0.0 0.0 0.0 0.0 56.33 24.01 0.0

Intensity 0.12 0.0

87.00 84.81 58.56 0.0 0.10 0.0 14.31 0.0 75.46 93.4495.82 12.20 0.0 62.83 0.0 83.09 66.4470.85 0.0 0.0 0.0 0.0 0.0 0.08 0.0 0.0 97.75101.07 0.0 0.0 1.51 0.06 0.0

0.04

0.02

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 Time (min)

Figure A.42: HPLC-UV Chromatogram of Peanut Skin Fraction B (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm using HPLC Method 2.

358 RT:0.00 - 104.99 3.68 6.30 NL: 1.59E7 238.9238.9 8.15 Base Peak F: - c ESI 30 238.9 sid=1.00 Full ms [ 125.00-250.00] MS 9.95 Negative_Mode_Reed- 20 238.9 2009-0513-07_Fraction_B 14.25 238.916.13 23.40 32.7735.73 239.0 10 137.225.33 151.1167.2 56.82 102.97 44.16 60.71 66.82 95.46100.18 44.79 52.72238.8 72.16 76.84 82.8789.2691.71 239.0 177.2 141.0 222.8NL: 8.24E6 177.3 141.0 141.0 222.7222.9222.8222.8222.8 239.0 245.2 Base Peak F: - c ESI Relative Abundance Relative 6.05 sid=2.00 Full ms [ 3.84 80.91 86.62 238.9 367.1 230.00-410.00] MS 238.9 351.3 Negative_Mode_Reed- 30 8.92 2009-0513-07_Fraction_B 238.9 10.30 20.11 101.92 20 17.03 31.16 90.01 238.9 315.1 76.59 377.3 403.1 252.1 46.47 54.26 63.25 67.48 367.392.23100.38 385.1 NL: 2.61E7 30.47 301.050.03371.1 283.2 299.3 337.2 353.3 22.11 39.5244.68 71.50 Base Peak F: - c ESI 10 252.1 363.0 59.80 299.0 331.0391.1 353.1 sid=3.00 Full ms [ 325.0 400.00-2000.00] MS

Relative Abundance Relative Negative_Mode_Reed- 51.16 2009-0513-07_Fraction_B 87.1489.65 46.52 755.2 57.36 741.1 623.1 435.2419.2 30 43.17 771.1 NL: 9.95E-1 20 56.47 UV Analog 86.55 38.60 477.1 104.65Negative_Mode_Reed- 453.2 933.1 67.2568.94 83.27 730.4 2009-0513-07_Fraction_B 32.86 66.46 72.59 96.07 3.60 31.07 51.77 887.3887.2 451.1 10 17.08 24.63 449.1 887.1 887.2 463.6 1916.8 9.44 15.08403.1 414.9 639.1 463.0 1782.2465.1 Relative Abundance Relative 3.37 0.0 46.96 19.9123.22 0.0 89.45 30.1832.74 46.42 50.87 17.83 0.0 0.0 87.94 0.0 0.2 0.0 0.0 42.28 0.0 0.0 57.16 0.0 24.01 0.0 0.0 0.0 58.56 84.81 14.31 62.83 75.46 93.4495.82 7.36 0.0 66.44 97.75 0.0 0.0 In te n sity 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 Time (min)

Figure A.43: Peanut Skin Fraction B (HPLC Method 2): HPLC-UV Chromatogram (bottom) versus TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125- 250 M-H-) (top), SID=2.0 Scan Range (m/z 230-410 M-H-) (second), and SID=3.0 Scan Range (m/z 400-2000 M-H-) (third)].

359 RT:0.00 - 105.07 86.55 NL: 2.25E-1 0.0 UV Analog Negative_Mo 0.22 de_Reed- 3.44 2009-0513- 0.21 0.0 05_Fraction_ C

0.20

0.19

57.40 0.18 0.0

0.17

0.16

0.15 87.61 0.0 6 0.14 91.52 0.0

In te n sity 83.79 0.0 96.4 0.13 0.0 46.33 51.81 75.77 0.0 0.0 0.12 0.0 30.82 30.07 0.0 44.51 0.11 32.92 0.0 0.0 0.0 20.55 0.10 0.0 42.26 17.50 29.13 0.0 37.30 59.57 0.09 0.0 0.0 47.34 82.88 27.31 0.0 0.0 4.24 0.0 73.27 0.0 0.0 63.24 0.08 0.0 71.27 0.0 16.88 0.0 0.0 0.0 98.69 0.07 14.73 11.66 0.0 0.0 0.0 0.06

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 Time (min)

Figure A.44: HPLC-UV Chromatogram of Peanut Skin Fraction C (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm using HPLC Method 2.

360 RT:0.00 - 105.07 44.17 59.82 NL: 5.83E5 177.2 235.9 96.00100.28 Base Peak F: - c ESI 94.75222.8 222.9 sid=1.00 Full ms [ 222.8 125.00-250.00] MS 60 87.45 Negative_Mode_Reed- 45.31 222.9 2009-0513-05_Fraction_C 209.2 86.22 40 3.14 83.03 20.17 35.71 78.63 222.9 222.9 21.33 42.82 47.64 74.78 222.8 134.3 32.36167.2 65.51 222.9 13.7319.16 30.87 177.3 51.3358.70 20 11.74 134.3 195.2 177.2 141.0 141.0 134.3134.3 187.2141.1 NL: 2.48E7 134.3 195.1 Base Peak F: - c ESI Relative Abundance Relative sid=2.00 Full ms [ 86.70 230.00-410.00] MS 351.2 Negative_Mode_Reed- 86.26 2009-0513-05_Fraction_C 351.2 60

83.87 40 367.1 NL: 1.68E7 81.42 71.54 353.2 Base Peak F: - c ESI 3.18 25.94 39.40 88.3791.79 20 17.57 30.29 63.5670.39329.174.49 sid=3.00 Full ms [ 9.67 14.89 319.229.82 33.34 45.85 52.34 60.21 351.2337.296.81 103.08 386.7 325.7 319.0 331.1 283.1299.1 355.1 400.00-2000.00] MS 319.0 317.1 283.1 351.2 374.6326.9 303.2 343.1 377.3Negative_Mode_Reed- Relative Abundance Relative 2009-0513-05_Fraction_C 57.56 623.2

52.08 60 NL: 2.25E-1 46.42 609.1 UV Analog 741.2 Negative_Mode_Reed- 40 82.97 99.58102.252009-0513-05_Fraction_C 30.99 44.75 853.1 730.3730.3 82.29 97.02 20 3.63 18.67 42.42775.1 68.51 91.70 13.64 20.87 29.22639.133.06 61.77 76.69853.1 903.1 646.2730.1 1068.5 435.0434.9 641.1 610.8 1175.1 522.0 1478.0 519.0 Relative Abundance Relative 3.44 57.40 86.55 0.0 0.0 0.0 87.61 91.5296.46 83.79 0.0 0.15 46.33 51.81 75.77 30.82 0.0 0.0 30.07 32.92 44.51 0.0 17.5020.55 0.0 0.0 0.0 0.0 37.30 59.57 82.88 4.24 0.0 0.0 0.0 63.24 73.27 0.0 0.0 71.27 0.10 9.94 15.38 0.0 0.0 0.0 98.69 0.0 0.0 0.0 0.0 0.0 0.0 0.0 In te n sity 0.05

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 Time (min)

Figure A.45: Peanut Skin Fraction C (HPLC Method 2): HPLC-UV Chromatogram (bottom) versus TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125- 250 M-H-) (top), SID=2.0 Scan Range (m/z 230-410 M-H-) (second), and SID=3.0 Scan Range (m/z 400-2000 M-H-) (third)].

361 RT:0.00 - 105.07 26.16 NL: 9.95E-1 0.0 UV Analog 0.50 Negative_Mo de_Reed- 2009-0513- 03_Fraction_ D 0.45

0.40 3.36 0.0

0.35

33.54 0.0 0.30

70.11 0.25 65.45 0.0 Intensity 0.0

18.21 0.20 42.70 0.0 0.0 22.86 57.38 0.0 0.0 50.74 64.61 0.15 0.0 17.17 41.43 51.47 0.0 49.66 75.64 0.0 39.54 0.0 0.0 96.35 30.92 0.0 63.61 0.0 0.0 86.47 0.0 0.0 84.46 0.0 77.62 0.0 87.55 0.10 0.0 0.0 0.0 91.42 9.6113.58 98.50 0.0 0.0 0.0 0.0

0.05

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 Time (min) Figure A.46: HPLC-UV Chromatogram of Peanut Skin Fraction D (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm using HPLC Method 2.

362 RT:0.00 - 105.07 3.48 NL: 4.49E5 221.0 93.60 104.82 222.8 Base Peak F: - c ESI 92.78 222.9 19.71 95.03 sid=1.00 Full ms [ 222.8 222.8 80 19.03134.3 125.00-250.00] MS 88.10 21.57 35.61 59.99 76.79 134.217.48 33.69 222.7 60 4.36 26.10 235.9 70.34 141.0 83.56 Negative_Mode_Reed- 134.2 193.7 26.69187.0167.1 59.53 80.21 221.0 179.8 222.8 2009-0513-03_Fraction_D 210.6 236.0 216.5 222.8 40 4.94 11.29 44.30 51.7156.81 177.2 141.1 20 221.1 134.2 227.2 NL: 1.85E7 Relative Abundance Relative Base Peak F: - c ESI 26.49 sid=2.00 Full ms [ 289.1 230.00-410.00] MS 70.37 Negative_Mode_Reed- 80 299.1 2009-0513-03_Fraction_D 25.62 33.73 65.51 86.74 60 18.3122.83289.1 289.1 64.74299.1 86.14351.2 325.9289.1 285.2 351.2 40 3.69 4.23 14.38 32.89 34.57 48.74 56.1659.14 72.05 77.80 96.35103.98 41.46 91.64 NL: 4.07E6 289.1 289.1 46.82 315.1 299.1 313.1 90.63 325.9289.1 289.1 317.1 301.1 337.3386.9353.2 Base Peak F: - c ESI 20 303.0 345.0 369.2 sid=3.00 Full ms [

Relative Abundance Relative 400.00-2000.00] MS 50.94 Negative_Mode_Reed- 463.1 73.24 84.63 2009-0513-03_Fraction_D 491.9 52.3557.34 1000.3 89.34 80 609.2461.0 692.0 44.56 96.57 77.72 96.05527.5 60 477.1 NL: 9.95E-1 491.9 582.3 25.6626.40 42.75 49.82 55.94 82.41 96.88 UV Analog 40 578.8578.8 448.7 463.0 477.1 479.5 527.4 Negative_Mode_Reed- 18.0421.79 67.79 70.27 3.57 13.43 33.7639.14 59.32 2009-0513-03_Fraction_D 564.1587.0 507.1 1175.1 20 1392.0 790.0 578.9 447.0 1175.3 Relative Abundance Relative 26.16 0.0

0.8

0.6 3.36 33.54 0.0 0.4 65.45 70.11 18.2122.86 0.0 42.70 57.38

In te n sity 50.74 64.61 17.17 41.43 49.66 0.0 0.0 75.64 86.47 96.35 8.33 10.22 0.0 0.0 84.46 87.55 98.50 0.0 0.0 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 Time (min)

Figure A.47: Peanut Skin Fraction D (HPLC Method 2): HPLC-UV Chromatogram (bottom) versus TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125- 250 M-H-) (top), SID=2.0 Scan Range (m/z 230-410 M-H-) (second), and SID=3.0 Scan Range (m/z 400-2000 M-H-) (third)].

363 RT:0.00 - 105.05 3.37 NL: 2.07E-1 0.0 UV Analog Negative_Mo 0.20 de_Reed- 2009-0513- 01_Fraction_ 0.19 E

0.18

0.17

0.16

0.15

53.75 75.79 0.14 0.0 0.0

0.13

38.85

Intensity 0.12 0.0

0.11 40.85 36.93 35.42 0.0 43.40 96.67 0.0 0.0 0.0 0.0 0.10 26.13 59.02 0.0 32.49 46.57 0.0 0.09 0.0 0.0 25.37 4.13 66.5967.72 89.55 22.87 0.0 68.48 79.85 87.26 0.0 0.0 0.0 0.08 0.0 91.73 18.98 0.0 0.0 0.0 0.0 0.0 0.0 0.07 95.96 15.42 100.80 0.0 5.79 12.67 0.0 0.0 0.06 0.0 0.0

0.05

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 Time (min)

Figure A.48: HPLC-UV Chromatogram of Peanut Skin Fraction E (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm using HPLC Method 2.

364 RT:0.10 - 105.05 59.36 NL: 7.72E5 235.9 Base Peak F: - c ESI sid=1.00 Full ms [ 80 125.00-250.00] MS 3.17 Negative_Mode_Reed- 98.22 102.872009-0513-01_Fraction_E 60 222.8 96.07 222.8 222.9 91.88222.8 89.58 40 222.9 57.14 222.9 3.55 35.36 39.9143.51 65.35 80.4083.90 17.76 26.48 53.37141.0 71.92 220.7 16.25 22.70 33.75167.2 217.0211.046.25 140.9 222.9222.8 20 10.87 169.2 216.8 141.0 242.3 NL: 2.52E6 216.9 137.3 167.2 217.0 217.1 Base Peak F: - c ESI Relative Abundance Relative sid=2.00 Full ms [ 38.99 230.00-410.00] MS 348.9 40.91 Negative_Mode_Reed- 43.86 2009-0513-01_Fraction_E 80 26.19 38.18 331.0 54.00 70.17 331.0 331.0 287.0 299.1 289.1 65.39 60 299.1 91.77 90.91337.3 369.2 NL: 2.08E6 40 25.70 57.18 103.26 64.72 99.98 3.21 19.26 289.1 33.21 50.24 285.2 71.77 86.81 Base Peak F: - c ESI 285.2 353.2 377.2 18.44 289.1 376.9 299.1 78.07 351.2 sid=3.00 Full ms [ 20 364.74.83 10.54 315.1 325.9 313.0 400.00-2000.00] MS 238.9238.7

Relative Abundance Relative Negative_Mode_Reed- 92.12 94.0296.92 2009-0513-01_Fraction_E 1153.2723.4719.2 80.00 87.26 80 463.1 1063.1 101.18 NL: 2.07E-1 60 646.4 UV Analog 51.97 Negative_Mode_Reed- 44.39 439.1 78.74 40 35.66 575.1 61.52 65.78 68.56 73.79 2009-0513-01_Fraction_E 27.11 37.18 463.0 3.46 22.00 24.63591.0 34.081149.1 1940.71411.81461.21259.6 16.17 1455.5 20 1246.14.47 607.1 946.0 1963.2 754.2 734.0 Relative Abundance Relative 3.37 0.0 53.75 75.79 0.15 38.85 0.0 0.0 36.93 40.8543.40 96.67 26.13 33.04 0.0 50.25 59.02 4.13 25.37 0.0 0.0 65.2467.7268.48 79.85 87.2689.55 17.6521.85 0.0 0.0 0.0 0.0 0.0 0.10 0.0 100.95 0.0 13.97 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 In te n sity 0.05

5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 Time (min)

Figure A.49: Peanut Skin Fraction E (HPLC Method 2): HPLC-UV Chromatogram (bottom) versus TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125-250 M-H-) (top), SID=2.0 Scan Range (m/z 230-410 M-H-) (second), and SID=3.0 Scan Range (m/z 400-2000 M-H-) (third)].

365 RT:0.29 - 105.02 3.37 NL: 9.95E-1 0.0 UV Anal 0.36 Negativog e_Mo 0.34 de_Reed- 2009-0512- 22.60 06_Fraction_ 0.32 0.0 F

0.30

0.28

0.26

0.24 19.22 0.0

0.22 38.56 0.0 0.20 39.18 35.92 0.0 0.18 34.92 0.0 Intensity 0.0 0.16 40.05 44.77 0.0 49.45 0.14 25.89 33.60 0.0 76.14 15.77 0.0 0.0 0.0 0.0 0.0 0.12 50.57 55.54 0.0 0.0 58.4560.48 0.10 65.14 0.0 67.55 0.0 0.0 84.52 0.0 81.37 87.41 0.0 88.59 97.11 0.08 0.0 0.0 9.8412.13 0.0 0.0 102.18 0.0 0.0 0.0 0.06

0.04

0.02

5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 Time (min)

Figure A.50: HPLC-UV Chromatogram of Peanut Skin Fraction F (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm using HPLC Method 2.

366 RT:0.20 - 105.02 91.17 3.18 35.40 51.40 96.02100.57103.67NL: 2.09E5 167.2 222.9 222.8 187.2 89.39 222.8 222.9 222.8Base Peak F: - c ESI 222.9 sid=1.00 Full ms [ 60 83.22 125.00-250.00] MS 43.81 75.3980.37222.8 Negative_Mode_Reed- 34.89 73.98 211.0 222.9222.9 2009-0512-06_Fraction_F 40 23.16 167.2 45.69 54.70 62.56 69.33222.8 19.89137.2 41.31 217.0 141.060.12140.9 222.8 23.83 32.41 3.69 153.1 217.1 222.8 18.32 137.3 151.1 20 167.2 13.50217.1 217.1 NL: 7.72E6 Base Peak F: - c ESI Relative Abundance Relative sid=2.00 Full ms [ 91.20 230.00-410.00] MS 369.1 Negative_Mode_Reed- 2009-0512-06_Fraction_F 60

40 NL: 1.29E7 19.48 92.02 Base Peak F: - c ESI 315.1 337.3 20 22.74 88.1689.25 sid=3.00 Full ms [ 18.19 36.2039.71 57.76 67.1770.81 98.88 3.21 8.45 13.55 289.226.7830.42 40.53 49.6452.27 63.40 74.42 81.42265.2339.3 400.00-2000.00] MS 315.1 289.8330.9 374.0 358.0299.1 353.3 322.8 369.1298.7 289.2319.0 331.1 363.0343.1 315.0 313.0 353.3 Negative_Mode_Reed- Relative Abundance Relative 2009-0512-06_Fraction_F 22.66 577.0 38.68 49.71 573.1 575.1 60 49.23 NL: 9.95E-1 575.1 35.99 50.03 UV Analog 40 19.40 30.17 3.64 16.18 573.1 575.1 Negative_Mode_Reed- 589.0 573.1 577.0 577.0 41.59 2009-0512-06_Fraction_F 25.84 48.57 603.1 575.1 20 14.37 577.0 68.11 89.14 52.34 65.2767.61 79.75 85.13 90.60 97.09 10.57577.0 60.03 447.1 78.39 102.07 575.0 447.2 423.2 911.01651.1 674.3 621.2 1977.5807.8 423.3 1358.7 1736.8 Relative Abundance Relative 3.37 0.0 0.6

22.60 0.4 19.22 0.0 38.56 35.92 39.18 0.0 15.77 25.89 33.60 0.0 44.7749.45 76.14 In te n sity 0.0 0.0 50.5756.4958.4565.14 0.2 5.92 11.99 67.55 81.3784.5287.4190.38 97.1199.07 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 Time (min)

Figure A.51: Peanut Skin Fraction F (HPLC Method 2): HPLC-UV Chromatogram (bottom) versus TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125-250 M-H-) (top), SID=2.0 Scan Range (m/z 230-410 M-H-) (second), and SID=3.0 Scan Range (m/z 400-2000 M-H-) (third)].

367 RT:0.10 - 105.14 3.43 33.26 NL: 6.47E-1 0.0 0.0 UV Analog 0.50 Negative_Mod e_Reed- 2009-0512- 02_Fraction_ G-red 0.45

0.40

0.35

35.05 0.0 39.57 0.30 0.0

0.25 25.53 Intensity 43.98 0.0 0.0

0.20 18.91 0.0 44.71 19.87 31.33 0.0 0.0 0.0 0.15 18.22 46.93 0.0 0.0 49.1552.72 57.55 75.88 0.0 0.0 61.56 0.0 64.23 0.0 0.10 0.0 71.21 15.60 0.0 80.34 87.34 0.0 88.36 96.70 13.47 0.0 0.0 0.0 99.46 5.99 0.0 0.0 0.0 0.0 0.0 0.05

Time (min) 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105

Figure A.52: HPLC-UV Chromatogram of Peanut Skin Fraction G-Red (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm using HPLC Method 2.

368 RT:0.10 - 105.14 NL: 4.85E5 0.99 104.80Base Peak F: - c ESI 101.36 sid=1.00 Full ms [ 240.8 222.8 80 222.8 125.00-250.00] MS 35.42 97.59 Negative_Mode_Reed-2009- 93.57 3.27 167.2 222.7 0512-02_Fraction_G-red 60 222.8 222.8 23.20 35.87 4.09 34.83 22.68137.3 167.2 89.41 40 31.31 57.78 81.46 148.8 137.3 167.2 39.80 72.1274.89 17.38 151.3 49.93 67.35 222.8 222.8 13.94 151.3 53.26236.159.56 242.2 151.2 151.2 222.8 222.8 20 151.2 141.0 222.9 NL: 5.15E6 Base Peak F: - c ESI Relative Abundance Relative sid=2.00 Full ms [ 90.88 230.00-410.00] MS 369.1 Negative_Mode_Reed-2009- 0512-02_Fraction_G-red 80

60 91.53 NL: 1.56E7 40 337.3 19.06 39.1540.43 Base Peak F: - c ESI 33.51 43.63 62.40 70.41 90.18 97.81102.17 0.63 8.03 315.1 26.53 331.0330.9 50.8452.54 64.64 78.68 87.62 sid=3.00 Full ms [ 20 3.13 17.42 289.1 331.0 58.58301.1 299.174.54 311.3 353.2 289.1 376.9343.1 285.2 298.9 265.3 386.7 400.00-2000.00] MS 241.0386.9240.9 315.1 319.0 298.8

Relative Abundance Relative Negative_Mode_Reed-2009- 0512-02_Fraction_G-red 33.43 38.46 577.1 573.0

80 35.38 575.0 NL: 6.47E-1 39.75 60 25.71 UV Analog 575.044.16 576.9 Negative_Mode_Reed-2009- 25.09 574.9 40 0512-02_Fraction_G-red 32.37 45.55 18.50577.0 3.73 577.0 574.9 59.6760.5361.62 13.2316.98577.0 52.44 81.58 84.7988.58 95.59 20 584.8 65.66 73.64 100.88 577.0 577.0 585.0553.1 911.1 621.1 439.1 1070.0 744.2 1489.3 1868.3 1452.9 1557.7 Relative Abundance Relative 3.43 0.0 33.26 0.0

35.05 0.4 39.57 0.0 25.53 0.0 43.98 18.91 31.33 44.71 18.22 0.0 0.0 0.0 49.59 57.5559.58 75.88 In te n sity 0.0 64.23 71.21 0.2 5.99 14.18 0.0 80.51 87.3489.51 96.7099.46 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 Time (min)

Figure A.53: Peanut Skin Fraction G-Red (HPLC Method 2): HPLC-UV Chromatogram (bottom) -vs- TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125- 250 M-H-) (top), SID=2.0 Scan Range (m/z 230-410 M-H-) (second), and SID=3.0 Scan Range (m/z 400-2000 M-H-) (third)].

369 3.42 33.74 40.8944.09 RT:0.00 - 105.01 0.0 0.0 0.0 0.0 NL: 9.96E-1 UV Analog 37.87 Negative_Mo 0.0 de_Reed- 0.95 2009-0512- 04_Fraction_ 0.90 29.83 G 0.0

0.85

0.80

0.75

0.70

0.65 29.27 0.60 0.0

0.55 25.54 0.0 0.50 In te n sity 45.99 0.45 0.0

50.02 0.40 0.0

0.35

0.30 24.18 0.0 53.31 0.25 20.82 0.0 0.0 0.20 19.24 60.07 0.0 61.51 0.0 0.15 0.0 73.23 80.77 87.52 89.4694.42 0.10 5.70 13.38 0.0 0.0 0.0 96.00101.20 0.0 0.0 0.0 0.0 0.0 0.0

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 Time (min)

Figure A.54: HPLC-UV Chromatogram of Peanut Skin Fraction G (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm using HPLC Method 2.

370 RT:0.00 - 105.01 3.12 NL: 4.88E5 222.9 Base Peak F: - c ESI 100 sid=1.00 Full ms [ 103.76125.00-250.00] MS 80 Negative_Mode_Reed- 51.44 87.59 95.52101.46 222.8 222.7 2009-0512-04_Fraction_G 23.32 36.46 187.2 249.3 222.8 60 209.0 137.2 35.86 47.53 81.77 22.66 167.0 84.08 40 8.95 42.03 54.33 58.8862.40 69.6472.7879.56222.9 10.53 137.3 32.02 151.2 222.8 148.8 151.2 151.3 217.1151.2 222.8222.8141.0 20 148.6 151.2 NL: 2.28E6 Base Peak F: - c ESI Relative Abundance Relative sid=2.00 Full ms [ 91.19 230.00-410.00] MS 90.72311.3 Negative_Mode_Reed- 100 311.3 92.64 2009-0512-04_Fraction_G 325.3 80 63.49 3.16 60 315.0 386.7 19.42 39.44 44.78 65.31 86.00 103.03NL: 1.64E8 30.53 35.44 393.0 58.73 285.2 373.3 40 315.0 331.0 50.97 76.71 83.55 95.55 Base Peak F: - c ESI 318.9 285.3 52.41318.9 386.7 18.25 27.72 376.9 72.01 293.0 327.3 339.3 sid=3.00 Full ms [ 3.99 6.46 25.65 343.1 20 315.1 319.0 299.0 400.00-2000.00] MS 315.1340.8 289.1

Relative Abundance Relative Negative_Mode_Reed- 35.06 2009-0512-04_Fraction_G 1150.9 100 33.84 863.0 35.50 80 1150.8 NL: 9.96E-1 UV Analog 60 41.23 44.27 Negative_Mode_Reed- 1150.8 25.72 29.98 2009-0512-04_Fraction_G 40 861.1 23.91863.0 863.0 3.74 862.9 45.2850.29 68.58 83.61 20 10.43 16.6320.95 53.20 58.45 65.54 73.29 79.67 86.94 93.1696.93 103.29 863.0 860.9 861.0588.9 1061.4863.0 1447.7861.1 1299.21233.01740.1 423.31094.81354.9 407.0646.6 1822.4 Relative Abundance Relative 3.42 33.74 40.8944.09 0.0 0.0 0.0 0.0 29.83 0.0 0.8 25.5429.27 0.0 45.99 0.6 0.0 50.02 0.0 24.18 0.0 53.31

In te n sity0.4 20.82 58.59 0.0 0.0 61.51 9.9513.38 0.0 78.8380.77 87.5289.4694.4297.67 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 Time (min)

Figure A.55: Peanut Skin Fraction G (HPLC Method 2): HPLC-UV Chromatogram (bottom) versus TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125- 250 M-H-) (top), SID=2.0 Scan Range (m/z 230-410 M-H-) (second), and SID=3.0 Scan Range (m/z 400-2000 M-H-) (third)].

371 RT:0.00 - 105.05 3.45 41.53 NL: 9.96E-1 0.0 0.0 UV Analog 0.90 Negative_Mo de_Reed- 0.85 2009-0511- 41.21 06_Fraction_ H 0.0 0.80

22.55 0.75 0.0

0.70 34.01 0.0 0.65

0.60 33.45 0.0 0.55 32.15 42.99 29.08 0.0 0.0 0.50 0.0 46.72 27.83 0.0 0.45 0.0 Intensity 26.40 0.40 0.0 49.65 0.35 0.0

0.30 52.25 21.13 0.0 0.0 0.25 18.94 56.59 0.0 58.56 0.0 0.20 0.0 75.51

0.15 0.0 76.8080.26 87.03 93.18 0.10 12.56 0.0 0.0 0.0 95.6598.67 0.0 0.0 0.0 0.0 0.05

Time (min) 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105

Figure A.56: HPLC-UV Chromatogram of Peanut Skin Fraction H (Obtained from Toyopearl Size Exclusion Chromatography) at 280nm using HPLC Method 2.

372 RT:0.00 - 105.05

3.18 35.57 NL: 5.50E5 222.8 167.2 Base Peak F: - c ESI sid=1.00 Full ms [ 80 23.24 125.00-250.00] MS 137.3 36.19 Negative_Mode_Reed- 60 167.1 2009-0511-06_Fraction_H 34.63 40.64 23.74 47.90 100.51103.93 167.2 151.2 71.70 96.72 22.09 137.3 50.51 88.03 222.8 40 19.85 151.2 52.2557.10 64.60 242.2 222.8 137.2 151.1 60.76 78.9680.72 222.9 222.8 153.1 151.2236.1 6.24 15.80 151.2151.2 222.8222.8 20 NL: 5.19E6 212.0 151.2 Base Peak F: - c ESI Relative Abundance Relative sid=2.00 Full ms [ 89.80 91.61 230.00-410.00] MS 311.3 325.3 Negative_Mode_Reed- 2009-0511-06_Fraction_H 80

60 88.70 NL: 4.65E7 40 311.3 94.33 30.24 51.39 85.63 96.14 Base Peak F: - c ESI 19.2022.79 39.41 62.58 339.2 44.49 268.0 381.4 sid=3.00 Full ms [ 3.22 289.1 318.9 38.63331.0 57.71 315.0 71.7474.46 407.1 20 3.89 13.0618.50315.1 392.9 63.85 79.78 400.00-2000.00] MS 304.7 330.9 407.1 242.2298.8 289.1315.2315.2 407.1 298.9 Negative_Mode_Reed- Relative Abundance Relative 2009-0511-06_Fraction_H 42.16 861.1

80 NL: 9.96E-1 UV Analog 60 40.58 22.70 Negative_Mode_Reed- 32.36 861.0 1151.0 29.33 46.86 2009-0511-06_Fraction_H 40 3.48 862.9 1151.1 860.9 1151.1 21.19 50.10 100.77 20 13.7818.75 55.35 59.50 62.99 69.66 72.19 80.05 86.23 91.76 99.01 860.7 701.3 701.7 875.1875.01151.0 1788.81911.6860.9 1651.81640.6 1601.8 1441.5 641.5 Relative Abundance Relative 3.45 41.53 0.0 0.0 22.55 39.58 34.01 0.8 0.0 0.0 33.45 0.0 29.08 0.0 46.72 0.6 0.0 0.0 49.65 21.13 52.25 0.4 17.76 0.0 56.59

In te n sity 0.0 0.0 68.71 75.51 12.33 0.0 0.0 76.8084.1687.03 93.1895.65101.22 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 Time (min)

Figure A.57: Peanut Skin Fraction H (HPLC Method 2): HPLC-UV Chromatogram (bottom) versus TIC’s [Total Ion Chromatograms – Showing the “ms” Normal Mass Spectra Base Peaks for the Most Intense Ions in SID=1.0 Scan Range (m/z 125- 250 M-H-) (top), SID=2.0 Scan Range (m/z 230-410 M-H-) (second), and SID=3.0 Scan Range (m/z 400-2000 M-H-) (third)].

373

374