University of , Fayetteville ScholarWorks@UARK

Theses and Dissertations

5-2014 Storage and Nutraceutical Evaluation of Muscadine ( rotundifolia Michx.) Derek Barchenger University of Arkansas, Fayetteville

Follow this and additional works at: http://scholarworks.uark.edu/etd Part of the Commons

Recommended Citation Barchenger, Derek, "Postharvest Storage and Nutraceutical Evaluation of Muscadine Grapes (Vitis rotundifolia Michx.)" (2014). Theses and Dissertations. 2364. http://scholarworks.uark.edu/etd/2364

This Thesis is brought to you for free and open access by ScholarWorks@UARK. It has been accepted for inclusion in Theses and Dissertations by an authorized administrator of ScholarWorks@UARK. For more information, please contact [email protected], [email protected].

Postharvest Storage and Nutraceutical Evaluation of Muscadine Grapes ( Vitis rotundifolia Michx.)

Postharvest Storage and Nutraceutical Evaluation of Muscadine Grapes ( Vitis rotundifolia Michx.)

A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Horticulture

By

Derek W. Barchenger State University Bachelor of Science in Agricultural Sciences and Natural Resources in Horticulture, 2012

May 2014 University of Arkansas

This thesis is approved for recommendation to the Graduate Council.

Dr. John R. Clark Thesis Director

Dr. Renee T. Threlfall Dr. Luke R. Howard Committee Member Committee Member

Dr. M. Elena Garcia Committee Member

ABSTRACT

A major limiting factor in muscadine ( Vitis rotundifolia Michx.) commercialization is deterioration during storage. One solution for extending market seasons and preventing market saturation for fresh muscadines could be the release of new with improved postharvest storability. Three studies were conducted; Study 1:

The effect of treatments on muscadine genotype postharvest storage and nutraceutical content; Study 2: An evaluation of a diverse range of genotypes for postharvest storage potential and nutraceutical concentrations; and Study 3: The impact of postharvest storage and berry segment on the nutraceutical composition of ‘Supreme’.

Research on table grapes has shown that field fungicide applications increase storability, but little is known of their effect on muscadines. The effect of field applications of on composition attributes during postharvest storage was evaluated on five muscadine cultivars (Nesbitt, Southern Jewel, Summit, Supreme, and Tara) and five breeding selections from the University of Arkansas Fruit Breeding Program. There were two field treatments (no fungicide and fungicide). For the fungicide treatment, alternating applications of two fungicides were applied at 14 d intervals during berry development.

Fruit was harvested and composition attributes including berry volume, titratable acidity

(TA), pH, soluble solids (%), soluble solids to titratable acidity ratio (SS/TA), color (L*,

Chroma, and hue), firmness (force to penetrate berry skins), weight loss (%), and unmarketable fruit (%) were evaluated every 7 d for three weeks. An additional eight cultivars and selections from non-fungicide-treated vines were subjected to postharvest storage potential evaluations. These included AM 02, AM 03, AM 18, AM 26, AM 28,

‘Delicious’, ‘Fry’, and ‘Ison’, as well as the genotypes from the non-fungicide-sprayed vines of genotypes in Study 1 (AM 01, AM 04, AM 15, AM 27, ‘Nesbitt’, ‘Southern Jewel’, ‘Summit’,

‘Supreme’, and ‘Tara’) to broaden the evaluation and comparison among genotypes for postharvest storage potential, and aid in the development of an Arkansas muscadine storage protocol. Additionally, a third study was conducted with ‘Supreme’ to determine the retention of nutraceutical compounds during storage on the berry segments of flesh

(pulp and skin), , and whole berries and to evaluate the variability in nutraceutical compounds among different vine of a single genotype. The storage attributes of force to penetrate the berry skin, percent weight loss, and percent unmarketable were all improved with fungicide applications, while the difference in percent volume change was minimal.

Field fungicide applications had no effect on muscadine berry color (Chroma, hue angle, and L*) or berry composition (pH, TA, SS/TA, and soluble solids). There were some effects of field fungicide applications impacted some nutraceutical levels, however results varied.

Weight loss, percent unmarketable, and force have potential for routine measurements in postharvest evaluation of muscadines, while percent volume change was found to be less useful. Titratable acidity, pH, soluble solids, and SS/TA can be useful measurement for ripeness, but not for storability. Chroma and hue angle are not indicators of storability, while L* shows potential, and nutraceuticals largely do not change during storage. The genotypes AM 04, AM 26, AM 27, AM 28, ‘Southern Jewel’, and ‘Supreme’ were identified as having the highest storage potential, while AM 01, AM 15, AM 18, and ‘Tara’ had the least storage potential. The genotypes AM 03, AM 04, AM 27, and ‘Ison’ were identified as having the highest overall nutraceutical content, while AM 18, AM 28, ‘Supreme’, and ‘Tara’ had the lowest overall nutraceutical content. Furthermore, muscadine grape nutraceuticals and antioxidant capacity vary by berry segment and vine. The primary contributors as sources of variation were found to be genotype, year, and storage time.

ACKNOWLEDGEMENTS

Many thanks are necessary to the people who have contributed towards the success of this project, beginning with my advisor Dr. John R. Clark. His advice, guidance, and friendship are invaluable to my education and experience. It was an honor to work underneath him.

I would also like to thank the members of my committee, Dr. Renee T. Threlfall for her attentive guidance with my project, as well as Dr. Luke R. Howard, and Dr. M. Elana

Garcia for their time, support, and suggestions offered. Furthermore, this work would not have been possible without the use of Dr. Luke R. Howard and Dr. Renee T. Threlfall’s laboratories. I would also like to thank Dr. Ed Gbur for the statistical aid and guidance.

Also, I would like to thank Sandra Sleezer for her assistance in all aspects of this project, and Cindi Brownmiller for her guidance with my nutraceutical and antioxidant capacity measurements. I greatly appreciate the assistance offered by the staff of the Fruit

Research Station in Clarksville. Particular thanks need to be given to Jeff Iness and Dan

Chapman for the fungicide application support and to David Gilmore and Kay Buck for the invaluable education and experience I gained working with them.

TABLE OF CONTENTS

Introduction 1 Objectives 4 Hypothesis 5 Literature Review 5 Genotypic Effects on Storage and Composition 5 Postharvest Physiology 9 General Storage 10 Storage Effects of Sulfur Dioxide 12 Controlled Atmosphere Storage 13 Fungicide Treatments 14 Berry Composition 17 Berry Segments 18 Total Phenolics 20 22 Ellagitannins 24 Flavonols 25 Resveratrol 26 Oxygen Radical Absorbance Capacity 29 Literature Cited 30 Chapter 1: The Effect Of Field Fungicide Applications On Muscadine Genotype Postharvest Storage And Nutraceutical Content 38 Abstract 38 Introduction 40 Materials and Methods 42 Grapes and 43 Fungicide Applications 43 and Transport 44 Storage Study 44 Berry and Composition Analysis 45 Nutraceutical Analysis 47 Experimental Design 50 Experimental Analysis 51 Results 51 Initial Attributes 51 Berry Storage Attributes 53 Berry Composition 62 Berry Color 74 Nutraceutical Content 80 Correlations 89 Discussion 91 Berry Storage Attributes 91 Berry Composition 95 Berry Color 98 Nutraceutical Content 100 Storage Protocol 106 Literature Cited 107 Appendix A 114 Chapter 2: Evaluation Of A Diverse Range Of Muscadine Grape Genotypes For Postharvest Storage Potential And Nutraceutical Concentrations 130 Abstract 130 Introduction 132 Materials and Methods 134 Grapes and Vineyard 134 Harvest and Transport 135 Storage Study 135 Berry and Composition Analysis 136 Nutraceutical Analysis 138 Experimental Design 141 Experimental Analysis 141 Results 142 Initial Attributes 142 Berry Storage Attributes 142 Berry Composition 149 Berry Color 156 Nutraceutical Content 158 Bronze and Black Storage Performance 169 Correlations 169 Discussion 171 Berry Storage Attributes 171 Berry Composition 177 Berry Color 180 Nutraceutical Content 183 Bronze and Black Storage Performance 188 Storage Protocol 188 Literature Cited 190 Appendix A 197 Chapter 3: The Effect Of Storage Time On Nutraceutical Content Of ‘Supreme’ Muscadine Berry Segments 212 Abstract 212 Introduction 214 Materials and Methods 219 Grapes and Vineyard 219 Harvest and Transport 219 Composition Analysis 220 Postharvest Storage 220 Nutraceutical Analysis 221 Experimental Design 224 Experimental Analysis 225 Results 225 Initial Attributes 225 2012 Nutraceutical Analysis 225 Combined Nutraceutical Analysis 237 Correlations 246 Discussion 246 Literature Cited 252 Appendix A 258 Conclusions 263

LIST OF TABLES

Chapter 1 Table 1.1. Initial values for force to penetrate berry skin, titratable acidity, pH, soluble solids, L*, Chroma, and hue angle averaged across year and fungicide treatment. 52 Table 1.2. F-test significance from ANOVA for berry weight loss, percent unmarketable berries, force required to penetrate the berry skin, and percent volume change of the berry during 3 weeks of storage. Highest-order interactions are italicized and shaded. 54 Table 1.3. F-test main effect significance for year on weight loss, percent unmarketable, force required to penetrate the berry skin, and percent volume change of the berry during three weeks of storage, averages across week and fungicide. 57 Table 1.4. F-test significance from ANOVA for pH, percent titratable acidity, percent soluble solids, and SS/TA ratio for four sources of variation of muscadine grapes during 3 weeks of storage. Highest-order interactions are italicized and shaded. 66 Table 1.5. Soluble solids to titratable acidity ratio for muscadine genotypes. 72 Table 1.6. Soluble solids to titratable acidity ratio of muscadine genotypes stored at 2 °C for 3 weeks. 72 Table 1.7. Soluble solids to titratable acidity ratio of fungicide and no fungicide treatments. 75 Table 1.8 F-test significance from ANOVA for berry Chroma, hue, and L* values of the four sources of variation of muscadine grapes during 3 weeks of storage. Highest-order interactions are italicized and shaded. 75 Table 1.9. L* values of fungicide and no fungicide treatments during storage at 2 °C for 3 weeks 76 Table 1.10 F-test significance from ANOVA for total anthocyanins, total ellagitannins, ORAC, total flavonols, total phenolics, and resveratrol concentrations for the three sources of variation. Highest-order interactions are italicized. 82 Table 1.11. Total phenolic concentations (mg/ 100g) of fungicide- and no fungicide- treated musadines. 88 Table 1.12. Resveratrol concentations (mg/100 g) of fungicide- and no fungicide- treated musadines. 88 Table 1.13. Total ellagitannin concentations of fungicide- and no fungicide-treated musadines. 90 Appendix A. Table A.1. Daily maximum and minimum temperatures and rainfall recorded at the Fruit Research Station; Clarksville, AR (lat. 35°31’58°N and long. 93°24’12’W) (2012 and 2013). 114 Table A.2. Interaction means of the postharvest attributes of percent weight loss, percent unmarketable, force to penetrate berry skin, and percent volume change for year, genotype, fungicide treatment, and week of storage at 2 °C for 0-3 weeks. 115 Table A.3. Interaction means of the composition attributes of percent titratable acidity, percent soluble solids, pH, and soluble solids to titratable acidity of fungicide and no fungicide treated muscadine genotypes stored at 2 °C for 0-3 weeks. 119 Table A.4. Interaction means of the berry color attributes of Chroma, Hue angle, and L* values of fungicide and no fungicide treated muscadine genotypes stored at 2 °C for 0-3 weeks. 123 Table A.5. Interaction means of the berry nutraceutical concentrations of total phenolics, total anthocyanins, total ellagitannins, total flavonols, and resveratrol, and the antioxidant capacity (ORAC) of fungicide and no fungicide treated muscadine genotypes. 127 Table A.6. Study 1 multivariate correlation coefficients among muscadine berry storage quality, composition, color, and nutraceutical content for 2012 and 2013. 129 Chapter 2 Table 2.1. Initial mean values for force to penetrate berry skin, composition and berry color for muscadine genotypes averaged across year (2012 and 2013). 143 Table 2.2. F-test significance from ANOVA for muscadine berry weight loss, percent unmarketable berries, force required to penetrate the berry skin, and percent volume change of the berry during 3 weeks of storage at 2 °C. Highest-order interactions are italicized and shaded (2012 and 2013). 143 Table 2.3. F-test significance from ANOVA for berry pH, titratable acidity, soluble solid content, and soluble solids to titratable acidity ratio during 3 weeks of storage. Highest- order interactions are italicized and shaded (2012 and 2013). 151 Table 2.4. F-test significance from ANOVA for Chroma, hue angle, and L* values during 3 weeks of storage at 2 °C. Highest-order interactions are italicized and shaded (2012 and 2013). 157 Table 2.5. F-test significance from ANOVA for total anthocyanins, total ellagitannins, ORAC, total flavonols, total phenolics, and resveratrol concentrations during. Highest- order interactions are italicized and shaded (2012 and 2013). 161 Table 2.6 Main effect of muscadine genotype means on total ellagitannin concentrations of the berry during three weeks of storage, averaged across years (2012 and 2013). 163 Table 2.7. Main effect of year means on total ellagitannin concentrations of the muscadine, averaged across genotypes (2012 and 2013). 164 Table 2.8. Main effect of genotype on total phenolic concentrations of the muscadine berries, averaged across years (2012 and 2013). 167 Table 2.9. F-test main effect significance for year on total phenolic concentrations averaged across muscadine genotypes (2012 and 2013). 167 Appendix A Table A.1. Average monthly maximum and minimum temperatures and total rainfall recorded at the Fruit Research Station; Clarksville, AR (lat. 35°31’58°N and long. 93°24’12’W) (2012 and 2013). 197 Table A.7. Interaction means of the postharvest attributes of percent weight loss, percent unmarketable, force to penetrate berry skin, and percent volume change for 198 year, genotype, and week of storage at 2-3 °C for 0-3 weeks. Table A.8. Interaction means of the composition attributes of percent titratable acidity, percent soluble solids, pH, and soluble solids to titratable acidity for year, genotype, and week of storage at 2 °C for 0-3 weeks. 202 Table A.9. Interaction means of the berry color attributes of Chroma, Hue angle, and L* values for year, genotype, and week of storage at 2 °C for 0-3 weeks. 206 Table A.10. Interaction means of the berry nutraceutical concentrations of total phenolics, total anthocyanins, total ellagitannins, total flavonols, and resveratrol, and the antioxidant capacity (ORAC) for year and genotypes. 210 Table A.11. Study 2 multivariate correlation coefficients among muscadine berry storage quality, composition, color, and nutraceutical content for 2012 and 2013. 211 Chapter 3 Table 3.1. The initial nutraceutical concentrations of ‘Supreme’ muscadine berry segments averaged across vines for 2012 and 2013. 226 Table 3.2. F-test significance from ANOVA for total anthocyanins, total ellagitannins, ORAC, total flavonols, total phenolics, and trans-resveratrol concentrations of ‘Supreme’ muscadine berries stored for 6 weeks at 2 °C in 2012. Highest-order interactions are italicized and shaded. 228 Table 3.3. F-test significance from ANOVA for total anthocyanins, total ellagitannins, ORAC, total flavonols, total phenolics, and trans-resveratrol concentrations of ‘Supreme’ muscadines at harvest in 2012 and 2013. Highest-order interactions are italicized and shaded. 238 Table 3.4. Main effects of year, muscadine berry segment, and vine means on total ellagitannin concentrations of ‘Supreme’ berry segments (2012 and 2013). 241 Table 3.5. Two-way interactions of year by vine, year by muscadine berry segment, and segment by vine means on ORAC levels of ‘Supreme’ (2012 and 2013). 241 Table 3.6. Multivariate correlation coefficients among nutraceutical concentrations of ‘Supreme’ muscadines at harvest averaged across years (2012 and 2013). 247 Appendix A. Table A.1. Daily maximum and minimum temperatures and rainfall recorded at the Fruit Research Station; Clarksville, AR (lat. 35°31’58°N and long. 93°24’12’W) (2012 and 2013). 258 Table A.12. Interaction means of the berry segment nutraceutical concentrations of total phenolics, total anthocyanins, total ellagitannins, total flavonols, and resveratrol, and the antioxidant capacity (ORAC) for year and week of storage at 2 °C for 6 weeks in 2012. 259 Table A.13. Interaction means of the berry segment nutraceutical concentrations of total phenolics, total anthocyanins, total ellagitannins, total flavonols, and resveratrol, and the antioxidant capacity (ORAC) for year at harvest. 262

LIST OF FIGURES

Chapter 1 Fig. 1.1. Percent berry weight loss of fungicide- and no fungicide-treated muscadine genotypes stored at 2 °C for 3 weeks. Values at week 0 (date of harvest) were excluded. Each standard error bar is constructed using 1 standard error from the mean. 55 Fig. 1.2. Percent unmarketable fungicide and no fungicide-treated berries of muscadine genotypes stored at 2 °C for 3 weeks Values at week 0 (date of harvest) were excluded. Each standard error bar is constructed using 1 standard error from the mean. 58 Fig. 1.3. Force to penetrate skin of muscadine genotypes stored at 2 °C for 3 weeks averaged across fungicide treatment. Each standard error bar is constructed using 1 standard error from the mean. 60 Fig. 1.4. Force to penetrate berry skin of fungicide- and no fungicide-treated muscadine genotypes average across weeks. Each standard error bar is constructed using 1 standard error from the mean. 61 Fig. 1.5. Percent change in volume of muscadine genotypes stored at 2 °C for 3 weeks averaged across fungicide treatment. Values at week 0 (date of harvest) were excluded. Decrease in size shown with positive values, while an increase in size shown by negative values. Each standard error bar is constructed using 1 standard error from the mean. 63 Fig. 1.6. Percent change in volume of muscadine genotypes with fungicide and no fungicide treatments. Decrease in size shown with positive values, while an increase in size shown by negative values. Each standard error bar is constructed using 1 standard error from the mean. 64 Fig. 1.7. Percent titratable acidity of fungicide- and no fungicide-treated muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean. Titratable acidity measured as tartaric acid. 67 Fig. 1.8. pH of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean. 68 Fig. 1.9. pH of fungicide and no fungicide-treated muscadine genotypes Each standard error bar is constructed using 1 standard error from the mean. 68 Fig. 1.10. Percent soluble solids of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean. 70 Fig. 1.11. Percent soluble solids of fungicide- and no fungicide-applied muscadine genotypes. Each standard error bar is constructed using 1 standard error from the mean. 71 Fig. 1.12. L* values of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean. 76 Fig. 1.13. L* values of fungicide and no fungicide-treated muscadine genotypes. Each standard error bar is constructed using 1 standard error from the mean. 78 Fig. 1.14. Chroma of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean. 79 Fig. 1.15 Hue angle of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean. 81 Fig. 1.16. Total content of fungicide- and no fungicide-treated muscadine genotypes. Each standard error bar is constructed using 1 standard error from the mean . 83 Fig. 1.17. Oxygen radical absorbance capacity of fungicide- and no fungicide-treated muscadine genotypes. Each standard error bar is constructed using 1 standard error from the mean. 85 Fig. 1.18. Total flavonol concentrations of fungicide and no fungicide-treated muscadine genotypes. Each standard error bar is constructed using 1 standard error from the mean. 86 Fig. 1.19 Total ellagitannin concentration of fungicide- and no fungicide-treated muscadine genotypes. Each standard error bar is constructed using 1 standard error from the mean. 90 Chapter 2 Fig. 2.1. Percent berry weight loss of muscadine genotypes stored at 2 °C for 3 weeks. Values at week 0 (date of harvest) were excluded. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). 145 Fig. 2.2. Percent unmarketable of muscadine genotypes stored at 2 °C for 3 weeks. Values at week 0 (date of harvest) were excluded. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). 146 Fig. 2.3. Force to penetrate skin of muscadine genotypes stored at 2 °C for 3 weeks Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). 148 Fig. 2.4. Percent change in volume of muscadine genotypes stored at 2 °C for 3 weeks. Values at week 0 (date of harvest) were excluded. Decrease in size shown with positive values, while an increase in size shown by negative values. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). 149 Fig. 2.5. pH of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). 151 Fig. 2.6. Titratable acidity (%) of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). 152 Fig. 2.7. Soluble solids (%) of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). 154 Fig. 2.8. Soluble solids to titratable acidity ratio of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). 155 Fig. 2.9. L* values of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). 157 Fig. 2.10. Hue angle of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). 159 Fig. 2.11. Chroma of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). 160 Fig. 2.12 Total anthocyanin concentrations of muscadine genotypes. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). 162 Fig. 2.13. Total flavonol concentrations of muscadine genotypes. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). 165 Fig. 2.14. Resveratrol concentrations of muscadine genotypes. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). 168 Fig. 2.15. Oxygen radical absorbance capacity (ORAC) of muscadine genotypes. Each standard error bar is constructed using 1 standard error from the mean (2012 and 168 2013). Fig. 2.16. Storage performance of black and bronze muscadine genotypes, averaged across year (2012 and 2013) stored for 3 weeks at 2 °C. Each standard error bar is constructed using 1 standard error from the mean. 170 Fig. 2.17. Muscadine genotype AM 01 at date of harvest (left) and after 3 weeks of storage (right), demonstrating the browning that occurred, likely caused by chilling injury (2013). 172 Fig. 2.18. Muscadine genotype AM 04 after 3 weeks of storage, demonstrating shriveling that occurred, likely caused by berry leakage (2012). 175 Chapter 3 Fig. 3.1. Total anthocyanin concentrations of ‘Supreme’ muscadine berry segments stored at 2 °C for 6 weeks in 2012. Each standard error bar is constructed using 1 standard error from the mean. 229 Fig. 3.2. Total ellagitannin concentrations of ‘Supreme’ muscadine berry segments stored at 2 °C for 6 weeks in 2012. Each standard error bar is constructed using 1 standard error from the mean. 230 Fig. 3.3. Oxygen radical absorbance capacity (ORAC) of ‘Supreme’ muscadine berry segments stored at 2 °C for 6 weeks in 2012. Each standard error bar is constructed using 1 standard error from the mean. 232 Fig. 3.4. Total flavonols of ‘Supreme’ muscadine berry segments from three different vines of ‘Supreme’ in 2012. Each standard error bar is constructed using 1 standard error from the mean. 233 Fig. 3.5. Total flavonols of ‘Supreme’ muscadine berry segments stored at 2 °C for 6 weeks in 2012. Each standard error bar is constructed using 1 standard error from the mean. 233 Fig. 3.6. Total phenolics of ‘Supreme’ muscadine berries stored at 2 °C for 6 weeks, averaged across berry segment in 2012. Each standard error bar is constructed using 1 standard error from the mean. 235 Fig. 3.7. Total phenolics of ‘Supreme’ berry muscadine segments from three different vines of ‘Supreme’ in 2012. Each standard error bar is constructed using 1 standard error from the mean. 235 Fig. 3.8. Resveratrol concentrations of ‘Supreme’ muscadine berries from three different vines of ‘Supreme’ stored at 2 °C for 6 weeks in 2012. Values are averaged across berry segments. Each standard error bar is constructed using 1 standard error from the mean. 235 Fig. 3.9. Resveratrol concentrations of ‘Supreme’ muscadine berry segments from three different vines of ‘Supreme’ in 2012. Each standard error bar is constructed using 1 standard error from the mean. 236 Fig. 3.10. Resveratrol concentrations of ‘Supreme’ muscadine berry segments stored at 2 °C for 6 weeks in 2012. Each standard error bar is constructed using 1 standard error from the mean. 236 Fig. 3.11. Total anthocyanin concentrations of ‘Supreme’ muscadine berry segments at harvest in 2012 and 2013. Each standard error bar is constructed using 1 standard error from the mean. 239 Fig. 3.12. Total ellagitannin concentrations of ‘Supreme’ muscadine berry segments at harvest averaged across year (2012 and 2013). Each standard error bar is constructed using 1 standard error from the mean. 239 Fig. 3.13. Total flavonol concentrations of ‘Supreme’ muscadine berry segments at harvest in 2012 and 2013. Each standard error bar is constructed using 1 standard error from the mean. 242 Fig. 3.14. Total phenolics of ‘Supreme’ muscadine berry segments at harvest in 2012 and 2013. Each standard error bar is constructed using 1 standard error from the mean. 244 Fig. 3.15. Resveratrol concentrations of ‘Supreme’ muscadine berry segments at harvest in 2012 and 2013. Each standard error bar is constructed using 1 standard error from the mean. 245

Introduction

Native to the southeastern United States, the muscadine (Vitis rotundifolia Michx.) grape is commonly grown for its unique flavor, high nutraceutical content, and insect and

disease resistance, which is often a limiting factor in the production of many bunch grapes

(Vitis spp.) (Conner, 2009; Silvia et al., 1994; Striegler et al., 2005; Walker et al., 2001).

Differentiation within Vitis between the subgenera Euvitis and Muscadina is thought to

have occurred during the Mesozoic Era, prior to and during the breakup of Pangea (Olien,

1990; Olien, 2001). It is commonly believed that V. vinifera L. and other grape species

descended from muscadines during the Quaternary Ice Ages, resulting in one less

chromosome pair (2x = 2n = 40 for V. rotundifolia and 2x = 2n = 38 for V. vinifera) (Conner,

2009; Olien, 2001; Olien, 1990).

Andre Michaux initially taxonomically described V. rotundifolia in France in 1803, from samples sent to him from the United States, and John Kunkel Small later described muscadines in 1913 (Olien, 2001). Muscadines have been under cultivation for over 400 years, originally in the Colony followed soon after by surrounding colonies and states (Conner, 2009). Dr. Calvin Jones released the first , Scuppernong, in

1810, which was the dominant cultivar in production for over 100 years (Goldy, 1992;

Olien 2001). As a result of this popularity, bronze muscadines are often mistakenly classified as “scuppernongs”, while dark‐colored muscadines are referred to as

“muscadines”, “bull grapes”, “bullet grapes”, “Southern fox grapes”, or “bullace grapes”.

Breeding efforts with muscadines began in the late 1800s and continued through the late

20th century (Goldly, 1992; Olien 2001). However, muscadine breeding has been limited

since the 1980s due to breeding program reductions, and it is thought that there is

1 substantial potential for further improvements of muscadines as a fresh fruit in a modern

breeding program (J. R. Clark, personal communication).

Muscadines have been described as having a pleasant aroma and taste and are

prized by some as a delicacy (Degner and Mathis, 1980). This native grape is commonly

grown in small and home plantings, ranging from North Carolina and to

Eastern Oklahoma and . Though commercial shipping for retail marketing is

practiced, the potential for expansion in this area exists (Perkins‐Veazie et al., 2012).

Approximately 2023 ha of muscadines are planted across the United States, with Arkansas having approximately 202 ha of muscadines currently in production (Perkins‐Veazie et al.,

2012). The berries are produced in small loose clusters that do not ripen evenly, but are removed from the vines as single berries rather than in clusters as with bunch grapes and are subsequently marketed in packages of these berries (Perkins‐Veazie et al., 2012). The fruit readily abscises from the rachis when ripe, lending itself to mechanical harvesting

(Ballinger and Nesbitt, 1982a; Smit et al., 1971).

Muscadine fruit range in color from greenish and bronze through pink, purple, and black. Muscadines have historically been commonly grown for , jelly, and juice, but more recently an increase in production for the fresh market has occurred (Perkins‐Veazie et al., 2012; Striegler et al., 2005). The recent recognition that the berries are important sources of essential vitamins, minerals, and antioxidants has increased consumer demand

(Perkins‐Veazie et al., 2012; Striegler et al., 2005). Additionally, alternative , including muscadines, are being explored by many growers in the South as a means of increasing profits or diversifying farm operations (Conner, 2009). Three of the major limiting factors on fresh‐market production are uneven ripening, short harvest season, and

2 high perishability of the fruit (James et al., 1999; Morris, 1980; Perkins‐Veazie et al., 2012).

The harvest season for muscadines in Arkansas usually extends from mid‐August through

September (James et al., 1997; Perkins‐Veazie et al., 2012). One solution for extending the market season and to prevent market saturation for fresh muscadines is the development of new cultivars that maintain high quality appearance and taste, as well as high nutraceutical content during long periods of storage (Himelrick, 2003; James et al., 1997;

James et al., 1999; Lane and Flora, 1980; Lutz, 1938; Morris, 1980; Perkins‐Veazie et al.,

2012; Starnes Saunders et al., 1981; Smit et al., 1971; Takeda et al., 1983a; Walker, 2001).

The University of Arkansas Division of began its fruit‐breeding program in 1964 (J. R. Clark personal communication). This effort, founded and directed for many years by Dr. James N. Moore, has been one of the most productive fruit breeding efforts in the United States in recent years. Over 50 fruit cultivars have been commercialized from the program, including blackberries, table grapes, peaches, nectarines, strawberries, and blueberries. These Arkansas cultivars provide growers with the opportunity for increased profits. The muscadine breeding program at the University of Arkansas was first put into motion in 2005, by collecting open‐pollinated seeds from several cultivars. The first crosses were made among cultivars in 2006. Crossing has continued through 2013. The primary location of the breeding effort is the Division’s Fruit Research Station near

Clarksville, with integral testing at the Southwest Research and Extension Center near

Hope. To date in Clarksville, 117 selections have been made and approximately 10,000 seedlings planted (J. R. Clark personal communication.). An expanded number of selections will be made in the coming years, which will result in cultivars released for commercial production. The forthcoming developments will provide expanded options for muscadine

3 growers. The major objectives of the Arkansas muscadine breeding program are to develop large fruit size, crisp texture, excellent flavor, edible skin, dry stem scar, self‐ fruitful flowers, high productivity, improved winter hardiness, disease resistance, and improved postharvest storability (J. R. Clark personal communication).

Since the implementation of a muscadine breeding program at the University of

Arkansas, selections have been made based on improved texture and dry stem scar.

Although increased crispiness and a greater percentage of dry stem scars has been observed, it is unknown whether there has been a true improvement in postharvest quality of muscadines (J. R. Clark, personal communication). Nutraceutical levels in muscadines vary among genotypes and little is known about the nutraceutical content of the University of Arkansas breeding selections (J. R. Clark, personal communication; Marshall et al.,

2012). Additionally, little is known about the affect of field fungicide applications on storability and nutraceutical content of muscadines, and if nutraceutical levels, in whole berries and berry segments ( and flesh plus skin), change during storage.

Objectives

The objectives of the following experiments were as follows:

1. To determine the effect of field applications of fungicides on the storage performance

and nutraceutical content of muscadine grapes.

2. Identify superior post‐harvest storage and handling muscadine genotypes in the

breeding program.

3. To determine the effect of storage time on nutraceutical content of berry flesh and seed

segments of ‘Supreme’ muscadine.

4 4. To develop a postharvest evaluation protocol for Arkansas muscadine genotypes for

potential commercial utilization.

Hypotheses

1. Field fungicide applications will increase postharvest storage quality and decrease

nutraceutical content and antioxidant capacity of muscadine grapes.

2. Recently developed Arkansas genotypes with improved skin and flesh texture will

store better than other commercially grown cultivars.

3. Nutraceutical content and antioxidant capacity will be unaffected by storage time.

4. The greatest nutraceutical content will be found in the muscadine berry seeds and

the least will be found in the berry flesh.

Literature Review

Genotypic Effects on Storage and Composition

Studies have been conducted on the genotypic interaction on both muscadine

storability and berry composition. Ballinger and Nesbitt (1982b) studied the quality of the

black genotypes ‘Nesbitt’, ‘Noble’, NC 67A015‐15, NC 67A015‐18, NC 67A015‐26, NC

67A015‐27, NC 67A015‐35, and the bronze genotypes ‘Carlos’ and NC 67A025‐22 after

storage with sulfur dioxide (SO2) generators. The authors used three types of commercially

available SO2‐generating sheets, a quick‐release, single‐stage type, a slow‐release, single

stage type, and a combination of the above two units. The berries were stored for 7 weeks

at 0 °C with 70 to 80% relative humidity (RH). They found that ‘Carlos’, ‘Noble’, NC

67A015‐26, NC 67A015‐35, and, NC 67A025‐22 were not suited for long‐term storage for fresh marketing with SO2 generators because of their high susceptibility to SO2 injury

including unacceptable appearance, excessive decay, and off‐flavors. The cultivar Nesbitt

5 and the selections NC 67A015‐15, NC 67A015‐18, and NC 67A015‐27 were found to be

acceptable for long‐term storage with SO2 generators. It was determined that NC 67A015‐

17 was best suited for long‐term storage with SO2 because of low susceptibility to injury

and flavor was not adversely affected.

James et al. (1997) studied the shelf‐life of ‘Granny Val’ and ‘Fry’ for fresh market

use. The fruit were stored using three treatments including use of SO2 generators enclosed

in a polyethylene bag, berries enclosed in a polyethylene bag with no SO2 generator, and a

control without a bag at 0 °C with slow release SO2 generators. The fruit was evaluated at

0, 4, and 6 weeks for changes in mass (largely due to moisture loss), percent decay, pH,

titratable acidity (TA), and °Brix (soluble solids content or SSC). They found that pH, TA,

and SSC remained constant throughout storage. ‘Granny Val’ showed signs of shriveling

after 2 weeks of storage, and ‘Fry’ maintained acceptable fruit quality for 4 weeks.

Lamikanra (1987) evaluated the protein content of the cultivars Carlos, Welder, and

Higgins, comparing his findings to the protein levels of V. vinifera cultivars. He found that

‘Welder’ grape juice had a higher protein concentration than that of ‘Carlos’ or ‘Higgins’

juices, but all three cultivars yielded juice with relatively higher concentrations of protein

than all V. vinifera juices evaluated.

Magee et al. (2002) determined the effect of field fungicide application on resveratrol content of fruit of muscadines that were susceptible (S), intermediate (I), or resistant (R) to five of the major berry diseases (bitter rot [Greeneria uvicola Ellis], black rot [Guignardia bidwellii Ellis], angular leaf spot [Mycosphaerella angulata Jenkins], and

Pierce’s disease [( Wells et al.]). The authors used the bronze cultivars

Carlos (I), Summit (R), and Higgins (S) and the black cultivars Cowart (S) and Noble(R).

6 The berries were grown in a vineyard with a fungicide spray program that used sequential applications of fungicides at 10 – 20 d intervals from early bloom to just prior to harvest.

They found that ‘Noble’ received the highest, or most severe, foliar disease score, while

‘Cowart’ and ‘Summit’ received the lowest, or least severe. The fruit from ‘Higgins’ from untreated vines had the highest bitter rot, black rot, and Macrophoma rot scores, while fruit from ‘Noble’ had the lowest rot scores. It was found that fungicide treatments reduced the berry disease score on ‘Higgins’ by almost 50%, and bitter rot disease scores of

‘Higgins’, ‘Carlos’, ‘Summit’, and ‘Cowart’ were lower from fungicide‐ treated vines. The authors concluded that though resveratrol concentration and disease incidence varied among cultivars and treatments, fungicide applications during the growing season reduced the incidence of fruit and foliar disease and increased resveratrol concentrations in the skin.

Lee and Talcott (2004) evaluated the influence of fruit maturity on ellagic acid derivatives and other antioxidant polyphenolics in the muscadine cultivars Albemarle,

Carlos, Cowart, Doreen, Fry, Red, Nesbitt, and Noble. The polyphenolics were extracted from the skin and pulp by homogenizing with 25 mL of 100% methanol, filtered through filter paper, and the solvent was removed at 40 °C under a stream of N. The juice was analyzed directly following centrifugation and filtration. The polyphenolics were separated and quantified by high‐performance liquid chromatography (HPLC) using solvent programs to identify phenolic acid, free ellagic acid, and ellagic acid derivatives.

Total soluble phenolics were analyzed using Folin‐Ciocalteu assay and expressed in gallic acid equivalents (GAE), while antioxidant activity was determined using an Oxygen Radical

Absorbance Capacity (ORAC) assay. The authors found the free form of ellagic acid

7 (aglycone) and two ellagic acid glycosides in all eight cultivars following extraction and

separation by HPLC. It was found that the skin of ripe ‘Cowart’ and ‘Doreen’ contained the

highest concentrations of ellagic acid and glycosides (1900 and 1620 mg/L, respectively).

Ellagic acid levels were considerably lower in the juice (105‐322 mg/L), when compared to

the skin (587‐1900 mg/L) and pulp (168‐455 mg/L). Anthocyanins, quantified only in the

black cultivars, were expressed in cyanidin equivalents. As expected, the authors found that anthocyanin concentration increased in the skin as the fruit ripened, with lower concentrations found in the pulp. The skin of ripe ‘Cowart’, ‘Nesbitt’, and ‘Noble’ (3250

mg/L, 5230 mg/L, and 4140 mg/L, respectively) contained the highest concentrations of

anthocyanins. Among the cultivars evaluated, ripe ‘Georgia Red’ contained the highest

levels of total phenolics of both skin and juice and (29.1 µmol Trolox Equivalents/mL), but

had lower anthocyanin, ellagic acid, and ellagic acid glycoside content. The data presented

by the authors suggested a diversity of compounds present in the cultivars

evaluated, and within segments of fruit with the skin generally containing the highest

levels.

Striegler et al. (2005) studied the quality and nutraceutical potential of ‘Black

Beauty’, ‘Carlos’, ‘Cowart’, ‘Doreen’, ‘Early Fry’, ‘Fry’, ‘Granny Val’, ‘Ison’, ‘Jumbo’, ‘Late Fry’,

NC67A015‐17, NC67A015‐26, ‘Nesbitt’, ‘Scarlett’, ‘Southern Home’, ‘Sterling’, ‘Sugargate’,

‘Summit’, ‘Supreme’, and ‘Tara’. The cultivars with the highest SSC were ‘Ison’ and

‘Supreme’ in 2002 and ‘Southern Home’ in 2003. The fruit skins of ‘Ison’ had the highest

total anthocyanins, while the seeds of ‘Supreme’ had the highest total phenolic and ORAC

levels. Threlfall et al. (2007) measured the composition and nutraceutical content of juices

from the cultivars Black Beauty, Carlos, Granny Val, Ison, Southern Home, Summit, and

8 Supreme. They found that ‘Southern Home’ had the highest SSC, while ‘Supreme’ had the lowest. Total phenolics were not associated with color as might have been expected; the bronze cultivars Summit and Granny Val had the highest concentrations of total phenolics and ‘Carlos’, another bronze cultivar, had the lowest. Among the black cultivars, Ison had the highest total anthocyanin concentration and ‘Supreme’ the lowest. Juice from the cultivars Southern Home and Carlos had the highest ORAC concentrations while the juice from ‘Nesbitt’ had the lowest. Marshall et al. (2012) found that high levels of phenolics and resveratrol only exist in some cultivars and are not ubiquitous for muscadine juice or pulp, while ellagic acid was found in the skin of all cultivars tested.

Postharvest Physiology

The fresh market for muscadines could be greatly expanded to other parts of the country, if short‐ or even long‐term storage life could be improved (Starnes Saunders et al.,

1981). Unlike some fruits, muscadines do not continue to accumulate sugars after they are harvested. As there is no starch reserve to convert into sugar in the berry, it is crucial to harvest muscadines at the correct ripeness to ensure quality of the stored berries (Lane,

1978). Flora and Lane (1979) studied the chemical and composition factors affected by ripeness and harvest date of ‘Cowart’ muscadine grapes. The fruit was density separated in

10, 12.5, and 15% sucrose solutions. It was determined that SSC, TA, and color were closely related to ripeness. Carroll and Marcy (1982) determined the chemical and physical changes during maturation of ‘Carlos’ and ‘Noble’ muscadines. They found that berry and seed weight, percent moisture, SSC, TA, and pH were all significantly correlated with berry maturity.

9 Starnes Saunders et al. (1981) researched the postharvest physiology and

senescence of muscadines. They found that deterioration of the stored fruit occurred due

to both senescence and pathogen attack. They found that muscadines could be held in cold

storage at 1 °C and 85% RH for up to 2 weeks without visible signs of tissue deterioration,

and muscadines stored for longer periods deteriorated rapidly upon removal from cold

storage. The authors concluded that the major limitations of storage life of muscadines

were poor preharvest cultural practices, poor postharvest handling and storage, cultivars

with high wet stem scar percentages, and poor transportation methods. Savoy and Hatton

(1980) found that muscadines with dry or clear stem scars stored at 1.1 °C and 95% RH

maintained optimum postharvest quality.

Ballinger and McClure (1983) stored the fruit of ‘Carlos’ light‐sorted into four

ripeness classes at 0 °C for 7 weeks. They found that each increase in ripeness was related

to a greater amount of decay development during and after storage. Lanier and Morris

(1979) evaluated the sensory preference and composition of berries from ‘Carlos’ sorted

into five maturity classes based on density separation in 8, 9, 10, and 11% NaCl brine

solutions. The authors found that the panelist’s sensory preference, SSC, and berry weight

increased, color improved, and acidity decreased with increasing fruit maturity.

General Storage

Lutz (1938) evaluated factors that influenced the quality of 84 V. labrusca L.

genotypes and six muscadine genotypes in storage. The fruit were given either a wash

treatment or were not washed, and placed in baskets. The baskets were then stored at

different temperatures ranging from 0 to 26.7 °C and at a range of 72‐96% RH. During the

washing treatment, the baskets of muscadines were submerged in a 0.5% hydrochloric acid

10 solution for 5 min, drained for 1 min, and then submerged in water for 5 min, and the berries were then stored wet. Measurements of respiration rates, SSC of the juice, and SSC to acid ratio were also taken. The author found that decay decreased as temperature decreased and fruit stored at 0°C maintained the highest quality especially for long storage

(up to 6 weeks). Fruit stored at 2.2 and 4.4°C maintained satisfactory quality for moderate lengths of storage. Relative humidity of 80% to 85% resulted in the least amount of decay and shriveling in storage. It was also found that more decay occurred on fruit that were washed; however the manner of washing and lack of drying could have contributed to these findings.

Takeda et al. (1983a) determined the storage potential of ‘Fry’ and ‘Southland’ muscadines for the fresh market. They measured several variables including, hydrocooling, washing in a chlorinated solution, and density separation. The authors found that berry weight, TA, SSC, and pH did not change during storage for the cultivars or treatments. They found that firmness decreased under all storage conditions. It was concluded that fruit should be harvested at full ripe, as underripe fruit were unmarketable because of off flavors, and overripe fruit were unmarketable because of decay during storage. With both cultivars, washed fruit stored better than unwashed fruit, and the number of wet stem scars played an important role in the length of storage. They also concluded that muscadines, even in ideal situations, do not store for longer periods than two weeks due to undesirable decay.

Takeda et al. (1983b) studied the physical and chemical changes in muscadines during postharvest storage. The berries were stored at 20, 4.5, and 0 °C to determine the biochemical parameters that changed during storage. It was determined that SSC, TA,

11 individual sugars, and organic acids did not change during storage at the three

temperatures tested, while decay increased rapidly in the fruit stored at 20 and 4.5 °C.

They also found that total phenolics increased while pectin content decreased during

storage, with more change occurring in the fruit stored at higher temperatures. Silva et al.

(1994) evaluated the quality changes in muscadine grapes during cold storage using either

polypropylene trays with holes, 2.7 mil flexible polyethylene bags without holes, or 2.7 mil

flexible polyethylene bags with holes. Eight experienced panelists evaluated the berries for

appearance, color, toughness, firmness, acidity, sweetness, and flavor. Measurements were

taken to determine weight change, color, penetration force, compression force, SSC, TA, pH,

and percent visible rot. They found that berries stored in polyethylene trays and flexible

polyethylene bags without holes had an acceptable shelf‐life of around 21 d when held at 5

°C, while the other treatments resulted in lower unacceptable quality berries.

Walker et al. (2001) conducted an experiment to determine the effect of storage on

quality attributes of muscadines at different maturities using polyethylene bags to evaluate

extended storage. Berries were stored in clear, vented, clamshell containers that were either placed in 0.03 mm multilayered polyolefin polyethylene bags or not wrapped and stored for 6 weeks at 2 °C with 89% RH. At 0, 2, 4, and 6 weeks, whole berries were blended and evaluated for pH, TA, and SSC. Additionally, sliced berries were used to measure firmness during storage. As storage time increased, decay increased and firmness decreased. Fruit stored in clamshells that were wrapped in bags had less decay, were less firm, and had less weight loss than those stored unwrapped. They found that the differences in the lack of loss of weight could have been due to the fruit leaking juice from cracks or splits that was retained in the bag and weighed with the grapes.

12 Storage Effects of Sulfur Dioxide

Studies have been conducted on extending the shelf life or storage life of table grapes (V. labrusca, and V. vinifera) with the use of SO2 to decrease decay (Harvey et al.,

1988; Litcher et al., 2008; Marois et al., 1986; Nelson and Ahmedullad, 1976). Treatments

of SO2 have been used to effectively prevent decay in bunch grape (V. vinifera) storage by

hindering the growth of decay‐causing fungi, namely Botrytis cinerea Pers.: Fr, Pennicillium sp., Aspergillus sp., Fusarium sp., Melaconium sp., Alternaria sp., and yeasts for over a century (Harvey et al., 1988; Litcher et al., 2008; MacLean et al., 2009; Marois et al., 1986;

Morris et al., 1992; Nelson and Ahmedullad, 1976; Smit et al., 1972).

The effect of SO2 on muscadine grapes in storage has also been studied (Ballinger

and Nesbitt, 1982a; Ballinger and Nesbitt, 1982b; James et al., 1997; James et al., 1999;

Lane, 1978; Lane and Flora, 1980; MacLean et al., 2009; Morris et al., 1992; Smit et al.,

1971; Conner and Maclean, 2012). It has been found that SO2 can be an effective tool in

slowing decay on muscadines in storage, but frequently with some negative or no affects

(MacLean et al., 2009; Morris et al., 1992; Smit et al., 1971). Bleaching, white spots, and off‐

flavors were found to occur in muscadines with the use of varying levels of SO2 and with

varying times of fumigation, especially in those berries with wet stem scars and cracks or

splits. Conversely, in other studies little to no negative effects of SO2 occurred during storage on muscadine grapes (Ballinger and Nesbitt, 1982b; Conner and Maclean, 2012;

James et al., 1999; Lane and Flora, 1980). As a result it can be inferred that the influence of the SO2 storage treatment on the quality of grapes is cultivar specific, and with muscadines not reliably beneficial.

Controlled Atmosphere Storage

13 Studies have been conducted to determine the effect of controlled atmosphere (CA),

sometimes called modified atmosphere (MA), storage on the postharvest quality of

muscadine grapes (Basiouny, 1998; Himelrick, 2003; Mercer and Smittle, 1990; Smittle,

1990). Controlled atmosphere storage usually involves regulating the amount of CO2, O2,

and N, as well as the temperature, air circulation, and the RH in cold storage to prevent

decay, reduce chilling injury, extend the storage or shelf life, and maintain high sensory

quality of the fruit. It was found that a temperature of 1.1‐2.2 °C, 90‐95% RH, O2

concentrations of 5%, CO2 concentrations of 15%, N concentrations of 80%, and air

circulation of 25 cfm/ton resulted in the maximum storage life of ‘Fry’ and ‘Granny Val’

muscadine grapes (Mercer and Smittle, 1990; Smittle, 1990).

Recent research has found some important drawbacks to CA storage. Controlled

atmosphere storage is very commodity specific; the ideal atmosphere for a certain cultivar

of muscadines may not be beneficial to a different cultivar or a different . This limits what commodities can be stored in a single cold room at a particular time. Additionally, some commodities can only tolerate a particular CA at a certain temperature, or only for a limited time, and fine‐tuning and constant adjustment of CA conditions can cause additional costs to the producer (Himelrick, 2003).

Fungicide Treatments

Several diseases affect muscadine fruit during storage. Some diseases are the result

of infection of the fruit during storage, while others are a result of pre‐ or post‐harvest

factors before storage (Lane, 1978). Some of the fungi that have been isolated on

muscadines postharvest include Alternaria sp., Aspergillus sp., Botrytis. sp., Fusarium sp.,

Penicillum sp., and Melaconium sp. (Lane, 1978; Smit et al., 1971; Takeda et al., 1983b). It is

14 widely accepted that fungicides can significantly reduce losses due to disease in the yield

and quality of muscadine grapes.

Harvey (1955) found that field fungicide application of Captan (N‐

trichloromethylthio‐4‐cylohecene‐1,2‐dicatboximide), B‐622 (2,4‐dichloro‐6‐o‐

chloroanilino –syntriazine), and Crag 5400 (a, a‐trithiobis N‐dimethylthioformamide)

significantly reduced postharvest decay in ‘Emperor’ (V. labrsuca).

Lane (1978) determined the effect of vineyard fungicide treatments on the shelf‐life

of ‘Cowart’, ‘Hunt’, ‘Magnolia’, and ‘Nevermiss’ muscadine grapes. Alternating field

applications of the fungicides Benlate 50 WP® (methyl 1‐(butylcarbamoyl)‐2‐

benzimidazole carbamate (6)) and Manzate D 80 WP® ([1,2 ethaznediybis

(carbamodithio)(2‐)]) were made at 18 d intervals from mid‐May to mid‐August. The

berries were then harvested, and either stored at ambient temperatures (23‐27 °C during

the day, and 10‐18 °C at night) for 48, 72, and 96 h, or 5 °C and 60% RH for 168 and 216 h.

Significant differences were found in the number of sound fruit between the treated and

control berries for all cultivars at ambient temperatures for 72 h except ‘Cowart’, where the

samples deteriorated regardless of the treatment. After 96 h at ambient temperatures all

samples except Manzate D 80 WP®‐treated samples of ‘Nevermiss’, had more than 50% unsound fruit. Conversely, after storage at 5 °C for 168 h and then exposed to shelf‐life at ambient temperatures for 48 h, it was found that all fungicide‐treated samples in all cultivars except ‘Cowart’ had significantly more sound fruit (Lane, 1978).

Takeda et al. (1982) studied the effects of prestorage fungicide dip treatments and storage temperatures on decay and composition of ‘Fry’. The authors sorted commercially

15 harvested fruit removing any damaged or decayed berries. Sound fruit was packaged into pint‐sized clamshells and either dipped in a Botran + Captan (2,6‐Dichloro‐4‐nitroaniline +

N‐trichloromethylthio‐4‐cylohecene‐1,2‐dicatboximide) fungicide mixture or left undipped.

The clamshells were then placed in vented cardboard boxes and stored at either 20, 4.5, or

0 °C. The authors found that little or no change occurred in pH, SSC, or TA for all treatments. It was determined that storage temperature influenced deterioration of muscadines, with the berries stored at 20, 4.5, and 0 °C having 25%, <10%, and >10% decay after storage, respectively. It was found that a decrease in pectin and an increase in total phenolic occurred in the berries during storage. Furthermore, they found that treatment of a fungicide dip was ineffective in managing decay during storage, although microbial spoilage was the major factor contributing to postharvest deterioration. It was concluded that muscadine berries cannot be stored longer that 14 d with the postharvest handling techniques in the study.

Smith and Magee (2002) applied the fungicides, Nova 40W® (myclobutanil: a‐butyl‐ a‐(4‐chlorophenyl)‐1H‐1,2,4,triazole‐1‐propanenitrile), Abound® (azoxystrobin: methyl

(E)‐2‐{2‐[6‐(2‐cyanophenoxy) pyrimidin‐4‐yloxy]phenyl}‐3‐methoxyacrylate*), and Elite®

(tebuconazole: a‐[2‐(4‐chlorophenyl)ethyl]‐alpha‐(1,1‐dimethylethyl)‐1H‐1,2,4‐triazole‐1‐ ethanol) sequentially to ‘Doreen’ and ‘Summit’ every 10 d beginning at early bloom and ending at pre‐harvest intervals (PHIs) of 64, 42, 28, 14, 7, 4, 2, 1, and 0 d to determine spray schedule effects on foliage and berry diseases and the relationship between disease incidence and berry resveratrol content. They found that all fungal diseases were reduced by fungicide treatments, and there were no differences in the number of asymptomatic berries among the nine PHIs. They found that ‘Doreen’ showed less disease symptoms than

16 ‘Summit’. The authors also found that resveratrol content of berry skins from fungicide‐

treated vines was significantly lower than those from untreated vines.

Berry Composition

Muscadines have experienced a resurgence of consumer interest as recent research

into the compositional and nutraceutical characteristics have shown that muscadines are good sources of valuable antioxidants, vitamins, minerals, and dietary fiber that have positive effects on health (Carroll et al., 1968; Ector et al., 1996; Lamikanra, 1988; Magee et

al., 2002; Threllfall et al., 2007). Research has shown that the muscadine grape possesses

one of the highest antioxidant levels among fruit crops (Greenspan et al., 2005). Some of

these components of muscadines have been shown to have anti‐cancer, anti‐mutagen, and

anti‐inflammatory properties, and to reduce levels of glucose, insulin, and glycated

hemoglobin in people with diabetes (Banini et al., 2006; Bralley et al., 2007; Greenspan et

al., 2005; God et al., 2007; Yi, 2005).

Muscadine grapes contain phenolic acids, flavonols, anthocayanins, ellagic acid, and

numerous ellagic‐acid derivatives (Boyle and Hsu, 1990; Haung et al., 2009; Lee et al.,

2005; Pastrana‐Bonilla et al., 2003; Stringer et al., 2009; Talcott and Lee, 2002). Ellagic

acid and other antioxidants have been shown to demonstrate anticarcinogenic activity in

the colon, lungs, and liver, as well as a reduction of birth defects in rats and mice, and two

forms of colon cancer in humans (Ector, 2001; Yi et al., 2005). Polyphenolic concentrations

usually increase in muscadines as fruit ripens (Lee et al., 2005) and are higher in wine than

in unfermented juices from an identical grape press (Musingo et al., 2001; Talcott and Lee,

2002). Greenspan et al. (2005) conducted a study determining the anti‐inflammatory

properties of muscadine grapes. Their results indicated that the anti‐inflammatory

17 properties of muscadines could be, in part, directly related to the high antioxidant content

of the grape, specifically the berry skin.

Smith (2013) studied the relationship between phytochemical content, berry

quality, and disease control following full‐season or early season applications of fungicides on ‘Carlos’, ‘Doreen’, and ‘Summit’ muscadines. The author found no significant differences in vine vigor, foliar diseases, or bitter rot scores among treatments where fungicide applications were stopped at varying preharvest intervals ranging from 0 to 56 d, or in disease scores between the full‐season and early season treatments. She found that full‐ season fungicide applications resulted in significantly lower berry disease scores than control treatments, and SSC was highest in full‐season fungicide treatment and untreated vines. Overall, ellagic acid and resveratrol content was lower in berries from fungicide‐ treated vines than from untreated vines, with control treatments resulting in berries with almost 10 times as much resveratrol as those from the full‐season treatment. It was concluded that fungicide treatments reduced berry diseases with as few as four applications compared to 12 applications in the full‐season schedules, and fungicides that control berry diseases had an effect on berry quality.

Berry Segments

Studies have been conducted to understand the concentration of nutraceutical

compounds in the different segments of muscadine fruit (Lee and Talcott, 2004; Marshall et

al., 2012; Sandhu and Gu, 2010; Takeda et al., 1983b; Threlfall et al., 2005). Ellagic acid was

found in higher concentrations in the skins (587‐1,900 mg/kg) of the berry when

compared to the pulp (not detected–455 mg/kg) or juice (105‐322 mg/kg)(Lee and Talcott,

2004). Anthocyanin concentrations were found to be generally higher in the berry skins

18 (390‐4942 mg/k) and juice (19.9‐610 mg/kg) than in the pulp (3.5‐383 mg/kg) or seed

(65‐273 mg/kg) (Lee and Talcott, 2004; Threlfall et al., 2005). Phenolic concentrations

were found to be generally higher in the skin (2,260‐3,454 mg/kg) and seeds (2,294‐9,534

mg/kg) than in the juice (979‐2,160 mg/kg) or pulp (443‐1,100 mg/kg) (Lee and Talcott,

2004; Takeda et al., 1983b; Threlfall et al., 2005). ORAC was measured and found present

in seeds (893‐1,100 μmol TE/g), skins (71.1‐422 μmol TE/g), pulp (4.9‐34 μmol TE/g), and

juice (15.5‐26.7 μmol TE/g) (Lee and Talcott, 2004; Threlfall et al., 2005).

Sandhu and Gu (2010) determined the antioxidant capacity and total phenolic

content of muscadine seeds, skin, and pulp. They also profiled the phenolic compounds

present. The authors manually separated the berries into seeds, skins, and pulp and stored

the portions at ‐20 °C. The seeds, skin, and pulp were freeze‐dried and ground into a

powder, and then extracted with an acetone/water/acetic acid mixture (70:29.7:0.3v/v).

Chromatographic analyses were performed as high‐performance liquid chromatography‐

diode array detection‐electrospray ion trap tandem mass spectrometry (HPLC‐DAD‐ESI‐

MSn). The authors found that, compared to V. vinifera, the presence of ellagic acid in

muscadines is unique, and is in the form of free ellagic acid, ellagic acid glycosides, and

ellagitannins. Additionally, they found that the anthocyanins present in muscadines are

3,5‐diglucosides (as opposed to 3‐glucosides in V. vinifera), and were identified as

delphinidin (Dp), cyanidin (Cy), petunidin (Pt), peonidin (Pn), pelargonidin (Pg), and

malvidin (Mv). The authors found that total phenolic content and antioxidant capacity were highest in the seeds (27.0‐81.2 mg GAE/g), followed by the skin (4.3‐10.3 mg GAE/g), and pulp (0.3‐1.2 mg GAE/g). Their results confirmed that HPLC‐DAD‐ESI‐MSn is a valuable tool for identification of phenolic compounds.

19 Marshall et al. (2012) determined stilbene, ellagic acid, flavonols, and phenolic content of the whole berry, skin, pulp, and juice of 21 muscadine cultivars. The authors identified reseveratol in the skins of 20 cultivars (0.0‐66.0 μg/g), in the pulp of ‘Eudora’

(0.948 μg/g) and ‘Janet’ (4.297 μg/g), and in the juice of none of the cultivars. It was found that ellagic acid was most abundant in the skins of all cultivars tested (500.0 ‐ 5,554.7

μg/g), but was found in similar concentration in the juice (24.2 to 56.8 μg/g) and the pulp

(21.9‐77.3 μg/g). Flavonols were only present in the skins and not in juice or pulp. The authors found that total phenolics and total anthocyanins were highest in the whole berry with the bronze cultivars having significantly less than black cultivars. Total phenolic concentrations ranged from 274.58 to 1,061.7 mg/100 g, while total anthocyanin concentrations ranged from none to 150.3 mg/100 g.

Total Phenolics

Phenolics are a large family of secondary metabolites involved in response to abiotic and biotic stresses (Marshall et al., 2012). Phenolics are ubiquitous in Plantae and are the most abundant secondary metabolites found in (Amakura et al., 2000).

Measurements of total phenolics by the Folin‐Ciocalteu metal reduction assay is a common index to provide an overall assessment of the content and chemical activity of compounds present, and aids in determining the antioxidant capacity of fruits and vegetables (Lee and

Talcott, 2004). Pastrana‐Bonilla et al. (2003) determined the phenolic content and antioxidant capacity of the skin, seed, pulp, and leaves of 10 muscadine cultivars (five bronze and five black‐skinned). The fruit segments and leaves were extracted using HPLC analysis, and total phenolics were determined colorimetrically using Folin‐Ciocalteu reagent. Total anthocyanins were determined according to a pH‐differential method, using

20 a UV‐visible spectrophotometer. Antioxidant capacity was determined by the Troloc equivalent antioxidant capacity (TEAC) assay. The major phenolics in muscadine berry skins were identified by retention times and characteristic spectra. Ellagic acid, trans‐ resveratrol, and kaempferol were found in the skins, whereas (+)‐catechin, (‐)‐epicatechin, and gallic acid were found in the seeds. Ellagic acid was the most abundant phenolic compound in muscadine berries. The authors found that muscadine seeds had approximately five times higher phenolic concentrations than any other fruit segment, while the leaves had 15 times more antioxidant capacity than the fruit. It was also found that the pulp had very low levels of phenolics. Threlfall et al. (2007) determined the nutraceutical content of juice from five black cultivars (Black Beauty, Ison, Nesbitt,

Southern Home, and Supreme) and three bronze cultivars (Carlos, Granny Val, and

Summit). The juice was analyzed for total phenolics with the Folin‐Ciocalteu assay with gallic acid as the standard. The authors found that total phenolics of the juice were not associated with skin color. The bronze cultivars Summit and Granny Val had the highest concentrations of total phenolics while ‘Carlos’, another bronze cultivar, had the lowest.

Stringer et al. (2009) evaluated the phenolic concentrations in the muscadine cultivars Carlos, Magnolia, Albermarle, and Noble and eight breeding selections. The total phenolic content was measured by the Folin‐Ciocalteu method at 700 nm, using gallic acid as a standard. The authors found that total phenolics were high in all cultivars and selections except ‘Magnolia’. Striegler et al. (2005) evaluated the total phenolics of several muscadine genotypes grown in southwestern Arkansas. The phenolics were extracted from whole berries, and analyzed using the Folin‐Ciocalteu assay, with catechin as the standard.

It was determined that ‘Nesbitt’ had the lowest total phenolic values. Marshall et al. (2012)

21 identified the total phenolic content of 21 muscadine cultivars by the Folin‐Ciocaleteau

assay with gallic acid equivalents with absorbance at 760 nm. The authors found that total

phenolics ranged from 274.58 mg/100 g (‘Fry’) to 1,061.65 mg/100 g (‘Nesbitt’), with

bronze cultivars having significantly less concentrations than black cultivars.

Anthocyanins

Anthocyanins are known to protect blood vessels in humans, and play a role in

cancer prevention, though anthocyanin absorption appears to be low in humans (Marshall

et al., 2012). Brown (1940) first researched the anthocyanin pigments of muscadines. He

found that the anthocyanins present in muscadines are 3,5‐diglucosides, as opposed to 3‐

glucosides in V. vinifera and V. aestivalis Michx. Ballinger et al. (1973) evaluated the anthocyanins of 10 black genotypes of muscadines for use in the wine industry. They found that all 10 clones contained the same anthocyanin compounds, Dp, Cy, Pt, Pn, and Mv.

Flora (1978) assessed the effects of heat treatments on juice and cultivar on muscadine anthocyanin concentrations. It was found that heating resulted in relative increases extracted in the 3,5‐diglucosides of Cy, Mv, and Pn while resulting in lower levels of Dp and Pt‐3,5‐diglucosides. It was found that total anthocyanins in the 11 cultivars tested ranged from 40 to 403 mg/100 g, with wide variations in the relative contents of individual anthocyanins, averaged across extractions. Delphinidin was the most prevalent anthocyanin identified in the pigment complex followed by Cy or Pt, Pn, and Mv. Goldy et al. (1987) analyzed and correlated the anthocyanin content of muscadine fruit, stems, tendrils, leaves, and leaf petioles using high performance liquid chromatography (HPLC) techniques. All five of the known 3,5‐diglucoside forms of were present in the fruit, stems, tendrils, and leaf petioles, while only Dp, Cy, and Pt were identified in the leaves. The

22 authors found that tendril analysis was best for predicting fruit Cy (r = 0.60), Mv (r = 0.57),

Pn (r = 0.66), and Pt (0.87), while stem analysis was best for predicting fruit Dp (r = 0.66).

Lee and Talcott (2004) evaluated the influence of fruit maturity on anthocyanins and other antioxidant polyphenolics in eight muscadine cultivars. The polyphenolics were extracted from the skin and pulp by homogenizing with 25 mL of 100% methanol, filtered through filter paper, and the solvent was removed at 40 °C under a stream of N. The juice was analyzed directly following centrifugation and filtration. Total phenolics were analyzed using Folin‐Ciocalteu assay and expressed in gallic acid equivalents (GAE), while antioxidant activity was determined using an ORAC assay. As expected, the polyphenolic compounds were concentrated in the epidermal tissues, which is often exceptionally thick in muscadines. Anthocyanins, quantified only in the black cultivars, were expressed in cyanidin equivalents. The authors identified Dp, Pt, Mv, Pn, and Cy as the anthocyanins present in the fruit. As expected, the authors found that anthocyanin concentration increased in the skin as the fruit ripened, with lower concentrations found in pulp. The skin of ripe ‘Cowart’, ‘Nesbitt’, and ‘Noble’ (3250 mg/L, 5230 mg/L, and 4140 mg/L, respectively) contained the highest concentrations of anthocyanins. On the basis of abundance, anthocyanins were the major antioxidant present in the muscadine grape skin and juice, and anthocyanin concentrations were directly related to antioxidant capacity (r =

0.99).

Striegler et al. (2005) evaluated the total anthocyanins of 19 muscadine cultivars and two selections. The anthocyanin content of the total phenolics was determined by a pH differential method, and absorbance was measured at wavelengths of 510 and 700 nm.

23 The authors found that ‘Ison’, ‘Southern Home’, and ‘Nesbitt’ had the highest anthocyanin concentrations, while ‘Supreme’ had the lowest.

Stringer et al. (2009) evaluated the anthocyanin content of the cultivars Carlos,

Magnolia, Albermarle, and Noble and eight breeding selections. Total anthocyanins content was determined using a modified pH differential method. The authors found that anthocyanin content showed a definite distinction between bronze and black fruit. The bronze genotypes ‘Magnolia’, ‘Carlos’, and NC76A0003‐102 showed significantly less anthocyanin, which was expected as anthocyanin content develops the fruit color. Sandhu and Gu (2010) first identified six anthocyanin compounds present in muscadines as Dp, Cy,

Pt, Pn, Pg, and Mv. Marshall et al. (2012) analyzed the total anthocyanin concentrations of

21 muscadine cultivars with a modified Giusti and Wroland pH shift assay. The authors found that total anthocyanin concentrations ranged from the highest of 150.29 mg/100 g in

‘Nesbitt’ to not detected in the bronze cultivars.

Ellagitannins

Ellagic acid is not commonly found in other grape species, but is exhibited in relatively high concentrations in muscadines (Marshall et al., 2012; Stinger et al., 2009).

Ellagitannins have been widely studied because of their antiproliferative and antioxidant properties, due to their ability to directly inhibit DNA binding of certain carcinogens and by reducing oxidative stress (Marshall et al., 2012). Concentrations of ellagic acid are generally higher in dark‐skinned berries than in bronze muscadines (Ector, 2001).

Lee et al. (2005) isolated and identified several ellagic acid derivatives present in muscadine berries and determined their relative antioxidant properties (AOX). Using methanol, the authors extracted compounds from the berry skins and flesh. Ellagic acid

24 derivatives were identified on the basis of UV and mass spectra, and the presence of

ellagitannins was confirmed by a significant increase in free ellagic acid with HPLC

followed by acid hydrolysis. The authors concluded that muscadines contain phenolic

acids, flavonols, anthocyanins, and numerous ellagic acid derivatives. The authors found

that AOX varied with elution and retention, while correlating with total phenolics (r = 0.90)

and total ellagic acid (r = 0.99). Boyle and Hsu (1990) identified and quantified ellagic acid in muscadine juice from 11 cultivars (Cowart, Doreen, Higgins, Hunt, Magnolia, Pamilco,

Regale, Roanoke, Scuppernong, Sterling, and Tarheel) in order to find an approach to eliminate the sediment that is occasionally produced in the juice. The authors used a combination of organic extraction and separation by HPLC to identify and separate the

ellagic acid compounds. Ellagic acid was found to vary significantly among cultivars, with a

range of 1.6 µg/mL (‘Higgins’) to 23.1 µg.mL (‘Hunt’).

Stringer et al. (2009) determined the ellagic acid concentrations in the muscadine

cultivars (Carlos, Magnolia, Albermarle, and Noble) and eight breeding selections. Ellagic

acid was identified using liquid chromatographic UV‐Vis detection analysis at 255 nm. The

authors found that ellagic acid content varied greatly from 11.46 mg/kg in ‘Carlos’, to 49.00

mg/kg found in CD8‐67. Sandhu and Gu (2010) identified ellagic acid in muscadines in the

forms of ellagic acid hexoside, ellagic acid xyloside, and ellagic acid rhamnoside. Marshall et

al. (2012) determined the total ellagic acid concentrations of 21 muscadine cultivars using

UV/VIS spectral interpretation, and retention time of authentic standards. The authors

found that ellagic acid was found in all cultivars tested, ranging from 500.0 μg/g (‘Alachua’)

to 5,554.7 μg/g (‘Southland’).

Flavonols

25 Muscadine berries have been found to contain high levels of flavonols. In humans,

flavonols protect against carcinogensis, DNA mutations, colon cancer, and heart disease

(Marshall et al., 2012). Flavonols are found in both V. vinifera and V. rotundifolia grapes, however the flavonol compound myricetin is unique to bronze muscadines (Marshall et al.,

2012). Yi et al. (2005) identified the phenolic compounds in muscadine grapes. Flavonols were extracted using HLB cartridge and LH20 column, and the compounds were identified using HPLC. The authors identified total flavonols fractions ranging from 76.3% to 86.1%.

Sandhu and Gu (2010) identified the flavonols present in muscadines as glycosides of , kaempferol, and myricetin. They also identified the flavonols myricetin hexoside, kaempferol hexoside, quercetin glucoside, and kaempferol rutinoside for the first time in muscadines.

Marshall et al. (2012) determined the flavonol concentrations of 21 muscadine cultivars using UV/VIS spectral interpretation, and retention time of authentic standards.

The authors found flavonols only in the skins of the berries and not in the pulp or juice. The most abundant flavonol contained in muscadine skins was myricetin, followed by quercetin, then kaempferol. The bronze cultivars, Janet, Sweet Jenny, and Triumph (998.75 g/g , 974.54 g/g , 933.91 g/g , respectively) contained significantly higher levels of myricetin, while black cultivars contained myricetin at lower levels. Quercetin was also found greatest in ‘Sweet Jenny’, and ‘Nesbit’, with 866.13 g/g , and 826.10g/g. Kaempferol was most abundantly present in ‘Nesbit’ (221.86 g/g), but was undetected in ‘Black Fry’ and ‘Black Beauty’.

Resveratrol

26 Stilbenes are synthesized in grape leaves in response to both biotic and abiotic

induction treatments, and the capacity to produce stilbenes is correlated with the

resistance of grape leaves to fungal infection (Creasy and Coffee, 1988; Marshall et al.,

2012). Resveratrol (trans‐3,5,4’‐trihydroxystilbene) is a phytoalexin, or stilbene, produced as a response to fungal infection (particularly by B. cinerea), stress including injury, and

UV‐irradiation (Jeandet et al., 1995; Jeandet et al., 1991; Marshall et al., 2012; Magee and

Smith, 2002). Resveratrol has long been confirmed in both red and white V. vinifera grapes skins and pulp, but not in seeds. Only within the last 20 years has resveratrol been measured in muscadines. The fleshy parts of both black and bronze muscadine berries have higher concentrations of resveratrol than was reported for V. vinifera and V. labrusca

(Ector et al., 1996). Resveratrol concentrations in muscadines have been found to be 33.67

µg/g in the skins, 0.10 µg/g in the flesh, and 59.94 µg/g in the seeds (Ector et al., 1996).

Resveratrol can potentially act as a chemopreventative of cardiovascular disease and coronary heart disease (Hudson et at., 2007; Jang et al., 1997; Lu and Sorreno, 1999; Magee and Smith, 2002). Resveratrol has lipid‐lowering action, inhibition of human low‐density lipoprotein oxidation and thus may delay atherosclerosis onset, inhibition of platelet aggregation in the blood, reduce cholesterol levels, and has shown to have cancer chemopreventative activity in all stages of carcinogenesis in cancers including prostate and breast cancer (Ector et al., 1996; Ector, 2001; Hudson et at., 2007; Jang et al., 1997; Lu and

Sorreno, 1999; Magee et al., 2002; Threlfall et al., 1999).

Jeandet et al (1995) conducted a study in France, on V. vinifera wine from grapes grown in both sprayed and unsprayed vineyards. They determined that conditions leading to the development of B. cinerea do enhance resveratrol production, but extensive grey

27 mold development may in fact destroy the induced phytoalexin. Their findings associated

years with conditions ideal for infection, but low disease incidence with higher levels of

resveratrol in wine. The authors suggest that B. cinerea infection is required for high levels of resveratrol in wine, but that extensive B. cinerea development before harvest may lower resveratrol content.

Ector et al. (1996) determined the resveratrol concentrations in whole berries, berries with the seeds removed, and the seeds of both black and bronze cultivars of muscadines. The authors established that resveratrol is a natural constituent of bronze‐ and black‐skinned muscadines, and though black‐skinned muscadines have higher levels of resveratrol, no significant difference existed between the two color groups. Additionally, they found that a broad range in resveratrol concentration within each color group, and among cultivars. It was also determined that the seeds of muscadines contained the highest levels of the berry segments, ranging from 24.5 to 62.2 µg/g, but only made up approximately 23‐30% of the resveratrol in the whole berries. Yi et al. (2005) identified the phenolic compounds in muscadine grapes. Nutraceutical compounds were extracted using HLB cartridge and LH20 column, and the compounds were identified using HPLC.

The authors found that resveratrol made up 2.4‐7.1% of the flavonol fraction identified.

Magee et al. (2002) determined the effect of field fungicide applications on resveratrol content of muscadine berries from the bronze cultivars Carlos, Summit, and

Higgins and the black cultivars Cowart and Noble. The berries were grown in a vineyard with a fungicide spray program that used sequential applications of Nova 40W®

(myclobutanil: a‐butyl‐a‐(4‐chlorophenyl)‐1H‐1,2,4,triazole‐1‐propanenitrile), Captan 50‐

WP® (N‐Trichloromethylthio‐4‐cyclohexan‐1,2‐dicrboximide), Abound® (azoxystrobin:

28 methyl (E)‐2‐{2‐[6‐(2‐cyanophenoxy) pyrimidin‐4‐yloxy]phenyl}‐3‐methoxyacrylate*), and

Benlate 50‐WP® (Benomyl (methyl 1‐(butylcarbamoyl)‐2‐Benzimidazolecarbamate) at 10

– 20 d intervals from early bloom to just prior to harvest. The authors found that

resveratrol concentrations varied by cultivar and treatment. It was determined that

resveratrol concentrations in the berry skins were much higher from unsprayed vines

compared to those from sprayed vines. However, the relationship between resveratrol

content and disease score was not determined. Seed resveratrol concentrations were not

affected by fungicide treatments. Overall, resveratrol concentrations of skins from

fungicide‐treated vines were lower than the untreated vines. It was concluded that

fungicide treatments reduced the fungal inoculum available on the fruit to elicit resveratrol

production.

Stringer et al. (2009) evaluated the resveratrol content of the muscadine cultivars

Carlos, Magnolia, Albermarle, and Noble and eight breeding selections. Resveratrol was

identified by HPLC analysis detection with a UV‐Vis at 310 nm. The authors found that resveratrol concentrations varied among genotypes with the selection CD8‐67 exhibiting the least (2.48 mg/kg) and NC71A006‐5 exhibiting the highest amount (48.57 mg/kg).

Sandhu and Gu (2010) identified trans‐resveratrol 3‐O‐ß‐glucoside to be high in the skins of muscadines on the basis of mass spectral data and a standard.

Oxygen Radical Absorbance Capacity

Measurements of peroxyl radical scavenging activity using ORAC assay is a common

index that provides an overall assessment of the content and chemical activity of

compounds present, and aids in determining the antioxidant capacity of fruits and

vegetables (Lee and Talcott, 2004; Prior et al, 2003). Lee and Talcott (2004) evaluated the

29 antioxidant activity of eight muscadine cultivars of varying ripeness using an ORAC assay.

As expected, the polyphenolic compounds were concentrated in the epidermal tissues,

which is often exceptionally thick in muscadines. Among the cultivars evaluated, ‘Georgia

Red’ contained the highest levels of total phenolics of both skin and juice and ORAC (29.1

µmol Trolox Equivalents/mL) in contrast to its low anthocyanin, ellagic acid, and ellagic acid glycoside content. The data presented by the authors suggested a diversity of phytochemical compounds present among different muscadine cultivars and within segments of fruit, with the skin generally containing the highest levels. Additionally the authors determined that physiologically ripe fruit contained higher levels of all compounds evaluated, resulting in greater antioxidant capacity.

Striegler et al. (2005) evaluated the ORAC concentrations of several muscadine genotypes. ORAC was measured using a modified method of a microplate reader using fluorecerin as the fluorescent probe. The ORAC values ranged from 2.52 to 3.37 µmol•mL‐

1TE, and it was determined that ‘Ison’ had the highest while ‘Nesbitt’ and ‘Granny Val’ had the lowest ORAC concentrations. Threlfall et al. (2007) analyzed for ORAC using a modified method with a microplate reader and using fluorescein as the florescent probe. They found that ORAC concentrations range from 2.1 to 4.1 µmol TE/mL, with ‘Southern Home’ and

‘Carlos’ having the highest concentrations and ‘Nesbitt’ having the lowest.

Literature Cited

Amakura, Y, Y, Umino, S. Tsuji, Y. Tonogai. 2000. Influence of jam processing on radical scavenging activity and phenolic content of berries. J. Agri. . Chem. 48:6292‐6297.

Ballinger, W.E., E.P. Maness, W.B, Nesbitt, and D.E. Carroll. 1973. Anthocyanins of black grapes of 10 clones of Vitis rotundifolia Michx. J. Food Sci. 38:808‐810.

Ballinger, W.E. and W. F. McClure. 1983. The effect of ripeness on storage quality of ‘Carlos’ muscadine grapes. Scientia Hort. 18:241‐245.

30

Ballinger, W.E. and W.B. Nesbitt. 1982a. Postharvest decay of muscadine grapes (Carlos) in relation to storage temperature, time, and stem condition. Amer J. Enol. Viticult. 33:173‐ 175.

Ballinger, W.E. and W.B. Nesbitt. 1982b. Quality of muscadine grapes after storage with sulfur dioxide generators. J. Amer. Soc. Hort. Sci. 107:827‐830.

Banini, A.E., L.C. Boyd, J.C. Allen, H.G. Allen, and D.L. Sauls. 2006. Muscadine grape products intake, diet and blood constituents of non‐diabetic and type 2 diabetic subjects. Nutrition 22:1137‐1145.

Basiouny, F.M. 1998. Quality of muscadine grapes as influenced by elevated CO2 and reduced O2 atmosphere. Acta Hort. 375‐380.

Boyle, J.A., and L. Hsu. 1990. Identification and quantitation of ellagic acid in muscadine grape juice. Amer. J. Enol. Viticult. 41:43‐47.

Bralley, E.E., D.K. Hartle, P. Greenspan, and J.L. Hargrove. 2007. Topical anti‐inflammatory activities of Vitis rotundifolia (muscadine grape) extracts in the tetradecanoylphorbol acetate model of ear inflammation. J. Medicinal Food 10:636‐642.

Brown, W.L. 1940. The anthocyanin pigment of the Hunt muscadine grape. J. Amer. Chem. Soc. 62:2808‐2810.

Carroll, D.E., M.W. Hoover, and W.B. Nesbitt. 1968. Chemical composition of eleven varieties of muscadine grapes. HortScience 3:99‐99.

Carroll, D.E. and J.E. Marcy. 1982. Chemical and physical changes during maturation of muscadine grapes (Vitis rotundifolia). Am. J. Enol. Viticult. 33:168‐172.

Cho, M.J., L.R. Howard, R.L Prior, and J.R. Clark. 2004. Flavanoid glycosides and antioxidant capacity of various blackberry, blueberry, and red grape genotypes determined by high‐ performance liquid chromatography/mass spectrometry. J. Sci. Food Agr. 84:1771‐1782.

Cho, M.J., L.R. Howard, R.L. Prior, and J.R. Clark. 2005. Flavonol glycosides and antioxidant capacity of various blackberry and blueberry genotypes determined by high‐performance liquid chromatography/mass spectrometry. J. Sci. Food. Agr. 85:2149‐2158.

Conner, P.J. 2009. A century of muscadine Grape (Vitis rotundifolia Michx.) breeding at the University of Georgia. Acta Hort. 827:481‐484.

Conner, P.J., D. Maclean. 2012. Evaluation of muscadine grape genotypes for storage ability. HortScience 47:S386. (abstr.).

31 Creasy, L.L., and M. Coffee. 1988. Phytoalexin production potential of grape berries. J. Amer. Soc. Hort. Sci. 113:230‐234.

Degner, R.L., and K. Mathis. 1980. Consumer acceptance of muscadine grapes. Proc. Fla. State Hort. Soc. 93:140‐143.

Ector, B.J., J.B. Magee, C.P. Hegwood, and M.J. Coign.1996. Resveratrol concentration in muscadine berries, juice, pomace, purees, seeds, and . Amer. J. Enol. Viticult. 47:57‐ 62.

Ector, B.J. 2001. Compositional and nutritional characteristics, p. 341‐367. In: F.M. Basiouny and D.G. Himelrick, (eds.). Muscadine Grapes. ASHS Press, Alexandria, VA.

Flora, L.F. 1978. Influence of heat, cultivar and maturity on the anthocyanidin—3,5— diglucosides of muscadine grapes. J. Food Sci. 43:1819‐1821.

Flora, L.F. and R.P. Lane. 1979. Effects of ripeness and harvest date on several physical and composition factors of ‘Cowart’ muscadine grapes. Amer. J. Enol. Viticult. 30:241‐246.

Giusti, M.M. and R.E. Wrolstad. 2001. Characterization and measurement of anthocyanin by UV‐visible spectroscopy. Current Protocols Food Anal. Chem. F1.2.1‐F1.2.13.

God, J.M., P. Tate, and L.L. Larcom. 2007. Anticancer effects of four varieties of muscadine grape. J. Medicinal Food Plants 10:54‐59.

Goldy, R.G. 1992. Breeding muscadine grapes. Hort. Reviews 14:357‐405.

Goldy, R.G., W. E. Ballinger, E. P. Maness, and W. H. Swallow. 1987. Anthocyanin content of fruit, stem, tendril, leaf, and leaf petiole in muscadine grape. J. Amer. Soc. Hort. Sci. 112:880‐882.

Greenspan, P., A. Ghaffar, J.L. Hargrove, D.K. Hartle, E.P. Mayer, J.D. Bauer, S.H. Pollock, and J.D. Gangemi. 2005. Antiinflammatory properties of the muscadine grape (Vitis rotundifolia). J. Agr. Food Chem. 53:8481‐8484.

Hager, T.J., L.R. Howard, R. Liyanage, J.O. Lay, and R.L. Prior. 2008. Ellagitannin composition of blackberry as determined by HPLC‐ESI‐MS and MALD‐TOF‐MS. J. Agr. Food Chem. 56: 661‐669.

Harvey, J.M. 1955. Decay in stored grapes reduced by field applications of fungicides. Phytopathol. 45:137‐140.

Harvey, J.M., C.M. Harris, T.A. Hanke, and P.L. Hartsell. 1988. Sulfur dioxide fumigation of table grapes: Relative sorption of SO2 by fruit and packages, SO2 residues, decay and bleaching. Amer. J. Enol. Viticult. 39:132–136.

32 Himelrick, D.G. 2003. Handling, storage and postharvest physiology of muscadine grapes: a review. Small Fruits Rev. 2:45‐62.

Hudson, T.S., D.K. Hartle, S.D. Hursting, N.P. Nunex, T.T.Y. Wang, H.A. Young, P. Arany, and J.E. Green. 2007. Inhibition of prostate cancer growth by muscadine grape skin extract and the resveratrol through distinct mechanisms. Cancer Res. 67:8396‐8405.

Huang, Z., R.D. Pace, P. Williams, and B. Wang. 2009. Identification of anthocyanins in muscadine grapes with HPLC‐ESI‐MS. Food Sci. Tech. 42:819‐824.

James, J., O. Lamikanra, J.R. Morris, G. Main, T. Walker, and J. Silva. 1999. Interstate shipment and storage of fresh muscadine grapes. J. Food Quality 22:605‐617.

James, J., O. Lamikanra, G. Dixon, S. Leong, J.R. Morris, G. Main, and J. Silva. 1997. Shelf‐life study of muscadine grapes for the fresh fruit market. Proc. Fla. State Hort. Soc. 110:234‐ 237.

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

Jeandet, P., R. Bessis, and B. Gautheron. 1991. The production of resveratrol (3,4,5’‐ trihydroxystilbene) by grape berries in different developmental stages. Amer. J. Enol. Viticult. 42:41‐46.

Jeandet, P., R. Bessis, M. Sbaghi, P. Meunier, and P. Trollat, 1995: Resveratrol content of wines of different ages: relationship with fungal disease pressure in the vineyard. Amer. J. Enol. Viticult. 46:1‐4.

Lamikanra, O. 1987. The proteins of muscadine grapes. J. Food Sci. 52:483‐484.

Lamikanra, O. 1988. Development of anthocyanin pigments in muscadine grapes. HortScience 23:597‐599.

Lane, R.P. 1978. Effect of vineyard fungicide treatments on the shelf life of muscadine grapes. Ga. Agr. Res. 19:12‐14.

Lane, R.P., and L.F. Flora. 1980. Some factors influencing storage of muscadine grapes. HortScience 15:273 (abstr.).

Lanier, M.R. and J.R. Morris. 1979. Evaluation of density separation for defining fruit maturities and maturation rates of once‐over harvested muscadine grapes. J. Amer. Soc. HortScience 104:249‐252.

33 Lee, J‐H., J.V. Johnson, and S.T. Talcott. 2005. Identification of ellagic acid conjugates and other polyphenolics in muscadine grapes by HPLC‐ESI‐MS. J. Agr. Food Chem. 53:6003‐ 6010.

Lee, J‐H., and S.T. Talcott. 2004. Fruit maturity and juice extraction influences on ellagic acid derivatives and other antioxidant polyphenolics on muscadine grapes. J. Agr. Food Chem. 52:361‐366.

Litcher, A., Z. Yohanan, T. Kaplunov, and S. Lurie. 2008. Evaluation of table grape storage in boxes with sulfur dioxide‐releasing pad with either an internal plastic liner or external wrap. HortTechnology 18:206‐214.

Lu, R., and G. Sorreno. 1999. Resveratrol, a natural product derived from grapes, exhibits antiestrogenic activity and inhibits the growth of human breast cancer cells. J. Cell Physiol. 197:297‐304.

Lutz, J.M. 1938. Factors influencing the quality of American grapes in storage. U. S. Dept. Agr. Tech. Bul. 606:1‐27.

Luvisi, D., H. Shorey, J. Smilanick, J. Thompson, B. Gump, J. Knutson. Sulfur dioxide fumigation of table grapes. Bulletin 1932. University of California, Davis, Division of Agriculture and Natural Resources.

MacLean, D., P.J. Conner, J. Paulk, and L. Grant. 2009. Postharvest control of decay organisms. The Southern Region Small Fruit Consortium. Prog. Rpt. 2009‐12.

Magee, J.B., B.J. Smith, and A. Rimando. 2002. Resveratrol content of muscadine berries is affected by disease control spray program. HortScience 37:358‐361

Marois, J.J., A.M. Bledsoe, W.D. Gubler, and D.A. Luvisi. 1986. Control of Botrytis cinerea on grape berries during postharvest storage with reduced levels of sulfur dioxide. Plant Dis. 70:1050– 1052.

Marshall, D.A., S.J. Stringer, and J.D. Spiers. 2012. Stilbene, ellagic acid, flavanol, and phenolic content of muscadine grape (Vitis rotundifolia Michx.) cultivars. Pharmaceutical Crops. 3:69‐77.

Mercer, M.D., and D.A. Smittle. 1990. Controlled atmosphere storage of muscadine grapes. Proc. Viticult. Sci. Symp. Fla. A and M Univ. 135‐139.

Morris, J.R. 1980. Handling and marketing of muscadine grapes. FruitSouth 4:12‐14.

Morris, J.R., O.L. Oswald, G.L. Main, J.N. Moore, and J.R. Clark. 1992. Storage of new seedless grape cultivar with sulfur dioxide generators. Amer. J. Enol. Viticult. 43:230‐232.

34 Musingo, M.N., S.F. O'Keefe, O. Lamikanra, C.A. Sims, and R.P. Bates. 2001. Changes in ellagic acid and other phenols in muscadine grape (Vitis rotundifolia) juices and wines during storage. Amer. J. Enol. Viticult. 52:109‐114.

Nelson, K.E., and M. Ahmedullah. 1976. Packaging and decay‐control systems for storage and transit of table grapes for export. Amer. J. Enol. Viticult. 27:74‐79.

Olien, W.C. 1990. The muscadine grape: , , history, and current industry. HortScience 25:732‐739.

Olien W.C. 2001. Introduction to the muscadines, p. 1‐13. In: F.M. Basiouny and D.G. Himelrick, (eds.). Muscadine Grapes. ASHS Press, Alexandria, VA.

Pastrana‐Bonilla, E., C.C. Akoh, S. Sellappan, and G. Krewer. 2003. Phenolic content and antioxidant capacity of muscadine grapes. J. Agr. Food Chem. 51:5497‐5503.

Perkins‐Veazie, P., S. Spayd, B. Cline, and C. Fisk. 2012. Handling and marketing guide for fresh market muscadine grapes. SFRC‐E03: 1‐12.

Prior, R.L., H. Hoang, L. Gu, X. Wu, M. Bacchiocca, L. Howard, M. Hampschwoodill, D. Haung, B. Ou, and R. Jacob. 2003. Assays for hydrophilic and lipophilic antioxidant capacity [oxygen radical absorbance capacity (ORACF1)] of plasma and other biological and food samples. J. Agr. Food Chem. 51:3272‐3279.

Prior RL, S.A. Lazarus, G. Cao, H. Muccitelli, and J.F. Hammer‐ stone. 2001. Identification of and anthocyanins in blueberries and cranberries (Vaccinium spp.) using high‐ performance liquid chromatography/mass spectrometry. J Agr. Food Chem. 49:1270–1276.

Sandhu, A.K., and L.W. Gu. 2010. Antioxidant capacity, phenolic content, and profiling of phenolic compounds in the seeds, skin, and pulp of Vitis rotundifolia (muscadine grapes) as determined by HPLC‐DAD‐ESI‐MSn. J. Agr. Food Chem. 58:4681‐4692.

Savoy, C.F. and T.T. Hatton. 1980. Post‐harvested characteristics of some commercially picked muscadine varieties. Proc. Viticult. Sci. Symp. Fla. A and M Univ. 24‐25.

Silva, J.L., E. Marroquin, C.P. Hegwood, G.R. Silva, and J.O. Garner Jr. 1994. Quality changes in muscadines for table grapes during refrigerated storage in various packaging systems. Proc. Viticult. Sci. Symp. Fla. A and M Univ. 17:65‐72.

Slinkard, K. and V.L. Singleton. 1977. Total phenol analysis: automation and comparison with manual methods. Am. J. Enol. Viticult. 28:49‐55.

Smit, C.J.B., H.L. Cancel, and T.O.M. Nakayama. 1971. Refrigerated storage of muscadine grapes. Amer. J. Enol. Viticult. 22:227‐230.

35 Smith, B.J., Magee, J.B. 2002. Limited fungicide applications affect berry rot severity and resveratrol content of muscadine grapes. Phytopathol. 92:577. (abstr.).

Smith, B.J. 2013. Fruit quality, phytochemical content, and disease severity of muscadine grapes affected by fungicide applications. Pharmaceutical Crops 4:21‐27

Smittle D.A. 1990. Requirements for commercial CA storage of muscadine grapes. Proc. Viticult. Sci. Symp. Fla. A and M Univ. 140‐142.

Starnes Saunders, M.S., F. Takeda, and T.T. Hatton. 1981. Postharvest physiology and senescence in muscadines. Proc. Fla. State Hort. Soc. 94:340‐343.

Striegler, R.K., P.M. Carter, J.R. Morris, J.R. Clark, R.T. Threlfall, and L.R. Howard. 2005. Yield, quality, and nutraceutical potential of selected muscadine cultivars grown in southwestern Arkansas. HortTechnology 15:276‐284.

Stringer, S.J., D.A. Marshall, and P. Perkins‐Veazie. 2009. Nutraceutical compound concentrations of muscadine (Vitis rotundifolia Michx.) grape cultivars and breeding lines. Acta. Hort. 841:553‐556.

Takeda, F., M. Starnes Saunders, C.F. Savoy, and T.T. Hatton. 1983a. Storageability of muscadines for use as fresh fruit. Proc. Viticult. Sci. Symp. Fla. A and M Univ. 3:31‐33.

Takeda, F., M. Starnes Saunders, and J.A. Saunders. 1983b. Physical and chemical changes in muscadine grapes during postharvest storage. Amer. J. Enol. Viticult. 34:180‐185.

Takeda, F., M. Starnes Saunders, J.A. Saunders, and T.T. Hatton. 1982. Effects of prestorage treatment and storage temperature on incidence of decay and chemical composition in muscadine grape. Proc. Fla. State. Hort. Soc. 95: 109‐112.

Talcott, S.T., and J‐H. Lee. 2002. Ellagic acid and flavonoid antioxidant content of muscadine wine and juice. J. Agr. Food Chem. 50:3186‐3192.

Threlfall, R.T., J.R. Morris, and A. Mauromoustakos. 1999. Effects of fining agents on trans‐ resveratrol concentration in wine. Austral. J. Grape Wine Res. 5:22‐26.

Threlfall, R.T., J.R. Morris, J.F. Meullenet, and R.K. Striegler. 2007. Sensory characteristics, composition, and nutraceutical content of juice from Vitis rotundifolia (muscadine) cultivars. Amer. J. Enol. Viticult. 58:268‐273.

Threlfall, R.T., J.R. Morris, L.R. Howard, C.R. Brownmiller, and T.L. Walker. 2005. effects on yield, quality, and nutraceutical content of juice, seeds, and skins from ‘Black Beauty’ and ‘Sunbelt’ grapes. J. Food Sci. 70:167‐171.

36 Walker, T.L., J.R. Morris, R.T. Threlfall, G.L. Main, O. Lamikanra, and S. Leong. 2001. Density separation, storage, shelf life, and sensory evaluation of 'Fry' muscadine grapes. HortScience 36:941‐945.

Yi, W., C.C. Akoh, and J. Fischer. 2005. Study of anticancer activities of muscadine grape phenolics in vitro. J. Agr. Food Chem. 53:8804‐8812.

37 Chapter 1

THE EFFECT OF FIELD FUNGICIDE APPLICATIONS ON MUSCADINE GENOTYPE

POSTHARVEST STORAGE AND NUTRACEUTICAL CONTENT

Abstract

A major limiting factor in muscadine grape (Vitis rotundifolia Michx.) commercialization is deterioration during storage. Research on table grapes has shown that field fungicide applications increase storability, but little is known of their effect on muscadines. The effect of field applications of fungicides on composition attributes during postharvest storage was evaluated on five muscadine cultivars (Nesbitt, Southern Jewel,

Summit, Supreme, and Tara) and four breeding selections from the University of Arkansas

Fruit Breeding Program (based at the Fruit Research Station, Clarksville, AR). There were two field treatments (no fungicide and fungicide). For the fungicide treatment, alternating applications of two fungicides were applied at 14‐d intervals during berry development.

Storage and composition attributes including berry volume, titratable acidity (TA), pH, soluble solids (%), color (L*, Chroma, and hue), firmness (force to penetrate berry skins), storage weight loss (%), and unmarketable fruit (%) were evaluated every 7 d for 3 weeks.

The nutraceutical measures of total anthocyanins, total ellagitannins, total flavonols, total phenolics, resveratrol, and the antioxidant capacity measurement of oxygen radical absorbance capacity (ORAC) were measured only at date of harvest. Overall, the postharvest storage quality attributes of weight loss, percent unmarketable, force, and volume change were significantly affected by genotype, year, fungicide treatment, and storage time. The berry color attributes of L*, Chroma, and hue and the berry composition attributes of TA, soluble solids, and pH were significantly affected by genotype and year,

38 but remained fairly constant across time of storage and fungicide treatments. The berry nutraceutical measurements of total anthocyanins, total ellagitannins, total flavonols, total phenolics, and the antioxidant capacity measurement of ORAC were significantly affected by genotype and year of the study. Additionally, total ellagitannins, total flavonols, and

ORAC were affected by fungicide treatments. Overall, resveratrol was only affected by genotype; however in 2013, resveratrol was greater in muscadines in the no fungicide treatment, but was unaffected in 2012. AM 27, ‘Southern Jewel’, and ‘Supreme’ were identified as having the highest potential for postharvest storage, while AM 01, AM 15, and

‘Tara’ had the least potential. AM 27 was also identified as having the overall highest nutraceutical concentration and antioxidant capacity (sum of anthocyanins, total phenolics, flavonols, resveratrol, and ORAC), while AM 28 and ‘Supreme’ had the lowest. It was determined that field fungicide applications can improve postharvest storage quality of muscadine grapes, but generally berry color and composition are unaffected. Furthermore, field fungicide applications resulted in varying differences in nutraceutical concentrations and antioxidant capacity.

39 Introduction

Native to the southeastern United States, the muscadine (Vitis rotundifolia Michx.)

grape is commonly grown for its unique flavor, high nutraceutical content, and pest and

disease resistance, which is often a limiting factor in the production of bunch grapes (Vitis

spp.) (Conner, 2009; Silvia et al., 1994; Striegler et al., 2005; Walker et al., 2001). This

native grape is currently grown in small commercial vineyards and home plantings, ranging from North Carolina and Florida to eastern Oklahoma and Texas. Arkansas has approximately 230 ha of muscadine in production, making up 10% of the total U.S. production. The recent recognition that the berries are important sources of beneficial antioxidants has increased consumer demand (Perkins‐Veazie et al., 2012; Striegler et al.,

2005). Additionally, alternative crops, including muscadines, are being explored by growers in the South as a means of increasing profits or diversifying farm operations

(Conner, 2009). Three of the major limiting factors on fresh‐market production are uneven

ripening, short harvest season, and high perishability of the fruit (James et al., 1999; Morris,

1980; Perkins‐Veazie et al., 2012).

Many variables contribute to muscadine storability, including berry maturity,

texture (crispness), weight loss, decay, shriveling, browning, leakage, and amount of dry

stem scars. Muscadines harvested at physiologically ripe maturity have been shown to

successfully store for 2 to 3 weeks (Perkins‐Veazie et al., 2012; Takeda et al., 1982). To

maintain adequate quality, muscadines should be stored from 1 to 5 °C with 85‐95%

relative humidity (RH) (Lutz, 1938; Sliva et al., 1994; Takeda et al., 1983; Walker et al.,

2001). The use of sulfur dioxide (SO2) storage treatment on the quality of bunch grapes is

cultivar specific, and with muscadines not reliably beneficial (Ballinger and Nesbitt, 1982a;

40 Ballinger and Nesbitt, 1982b; Conner and Maclean, 2012; James et al., 1997; James et al.,

1999; Lane, 1978; Lane and Flora, 1980; MacLean et al., 2009; Morris et al., 1992; Smit et

al., 1971).

While it is well known that fungicide applications benefit other fruits and

vegetables, including other Vitis species, little is known about the effect of field fungicide applications on storability and nutraceutical content of muscadines (Smith 2013). It has been shown that field fungicide applications improved the shelf‐life of ‘Doreen’, ‘Hunt’,

‘Magnolia’, ‘Nevermiss’, and ‘Summit’ muscadine grapes, but ‘Cowart’ was unaffected (Lane,

1978; Smith and Magee, 2002). Additionally, it has been found that a treatment with a fungicide pre‐storage was ineffective in managing decay during storage, although microbial spoilage was the major factor contributing to postharvest deterioration (Takeda et al.,

1982). Field fungicide applications have been shown to increase resveratrol concentrations in V. vinifera wine grapes and reduce resveratrol concentrations in muscadines (Jeandet, 1995; Magee et al., 2002). The effects of field fungicide applications on other nutraceutical compounds in muscadines are unknown.

Muscadine grapes contain phenolic acids, flavonols, anthocyanins, ellagic acid, and numerous ellagic‐acid derivatives (Boyle and Hsu, 1990; Haung et al., 2009; Lee et al.,

2005; Pastrana‐Bonilla et al., 2003; Stringer et al., 2009; Talcott and Lee, 2002). Ellagic acid and other antioxidants have been shown to demonstrate anticarcinogenic activity in the colon, lungs, and liver, as well as a reduction of birth defects in rats and mice, and two forms of colon cancer in humans (Ector, 2001; Yi et al., 2005). Polyphenolic concentrations usually increase in muscadines as fruit ripens (Lee et al., 2005) and are higher in wine than in unfermented juices extracted from berries with identical fruit pressing procedures

41 (Musingo et al., 2001; Talcott and Lee, 2002). Research has shown that the muscadine grape possesses one of the highest antioxidant levels among fruit crops (Greenspan et al.,

2005). Some of these components of muscadines have been shown to have anti‐cancer, anti‐mutagen, and anti‐inflammatory properties, and to reduce levels of glucose, insulin, and glycated hemoglobin in people with diabetes (Banini et al., 2006; Bralley et al., 2007;

Greenspan et al., 2005; God et al., 2007; Yi et al., 2005).

Since the implementation of a muscadine breeding program at the University of

Arkansas in 2005, selections have been made based on flower type, fruit size, time of ripening, hardiness, improved texture, and dry stem scar. Although increased crispiness and a greater percentage of dry stem scars has been observed, it is unknown whether there has been a true improvement in postharvest quality of muscadines. Nutraceutical levels in muscadines vary among genotypes (Marshall et al., 2012), and no information has been collected on the nutraceutical content of the University of Arkansas breeding selections

(J.R. Clark, personal communication).

The objectives of this study were to determine the effect of field applications of fungicides on the storage performance, berry composition, berry color, and nutraceutical concentrations of muscadine grapes and to develop a postharvest evaluation protocol for

Arkansas muscadine genotypes for potential commercial utilization. It was hypothesized that field fungicide applications will increase postharvest storage quality and decrease nutraceutical content and antioxidant capacity of muscadine grapes and that commercial cultivars along with recently developed Arkansas genotypes with improved skin and flesh texture will vary in storage potential.

Materials and Methods

42 Grapes and Vineyard

Vines of nine muscadine genotypes (AM 01, AM 04, AM 15, AM 27, ‘Nesbitt’,

‘Southern Jewel’, ‘Summit’, ‘Supreme’, and ‘Tara’) used for the study were grown at the

University of Arkansas Fruit Research Station, Clarksville, AR (lat. 35°31’58’N and long.

93°24’12’W). Vines were of varying ages within each genotype, most of the cultivars were approximately six years old, while many of the selections were from younger vines three to four years old. The vines were grown in Linker fine sandy loam, in USD hardiness zone 7a, where average annual minimum temperatures reach ‐15 – 17.7 °C. Vines were spaced 6.1 m apart and rows were spaced 3.0 m apart. A single‐wire trellis was used, and vines were trained to a bilateral cordon. The vines were dormant pruned annually in February using spur with spurs retained of two to four buds in length. Weeds were controlled with pre‐ and postemergence as needed, and vines did not have any stress from weed competition. Vines were irrigated by drip irrigation as needed, beginning in early

June (prior months received adequate rainfall) and continuing through the harvest period.

Vines received N fertilization in March of each year at a rate of approximately 70 kg/ha. No insecticides or other pest control compounds were applied to the vines, other than those vines that received the fungicide treatments. The vines used in the study had full crops produced each year, and no crop reduction due to winter injury or other limitation occurred. Thus, the vines produced fruit under representative conditions. Daily maximum and minimum temperatures along with rainfall were recorded at the research location to characterize the environment the vines were subjected to and potential differences among years (Table A.1).

Fungicide applications

43 The vines of nine muscadine cultivars and selections were used for fungicide

treatments. Each genotype had a single vine treated while the other did not receive

fungicide applications (berries from the fungicide treated vines were referred to as

fungicide‐treated berries and berries from the no fungicide treated vines were referred to

as no fungicide‐treated berries). A rotation of field fungicide applications of Abound®

(azoxystrobin: methyl (E)‐2‐{2‐[6‐(2‐cyanophenoxy) pyrimidin‐4‐yloxy] phenyl}‐3‐

methoxyacrylate*) and Rally® (myclobutanil: a‐butyl‐a‐(4‐chlorophenyl)‐1H‐1,2,4,triazole‐

1‐propanenitrile) were applied with a backpack sprayer every 14 d beginning when the

fruit was approximately 3‐5 mm in diameter and after approximately 400 growing degree

units were accumulated beginning 1 Jan.

Harvest and Transport

The muscadines were once over, hand‐harvested at the Fruit Research Station. Fruit

was harvested either early in the morning or late in the afternoon and transported to the

University of Arkansas Institute of Food Science and Engineering, Fayetteville, AR., in an

air‐conditioned car on the same day. Harvest date/maturity was based on soluble solids of

18‐22% in 2012 and 15‐18% in 2013 (due to differences in summer temperature and precipitation), ease of release from the pedicel, and berry color. Both fungicide treated and non‐fungicide treated vines within the same genotype were harvested on the same day.

Storage Study

Berries were then hand‐sorted to remove any split, shriveled, or decayed fruit

before packaging to simulate commercial standards. Only sound berries, showing no signs of unmarketability, were stored. The fruit was packaged into hinged standard vented clamshells (18.4 cm x 12.1 cm x 8.9 cm) (H116, FormTex Plastics Corporation, Houston, TX)

44 and stored in plastic harvest lugs in cold storage at 2 °C with 85‐89% RH. From the harvested fruit, six vented clamshell containers were filled to approximately 500 g.

Three of these clamshells were used as storage replications for each treatment.

Total clamshell weight was determined at date of harvest, and percent weight loss was calculated as percent weight decrease from this initial value. Weight loss and percent unmarketable fruit were evaluated on the storage clamshells every 7 d for up to 21 d.

Storage performance was evaluated by removing all the fruit from each clamshell and counting the number of fruit that showed signs of unmarketability, which included individual or a combination of characteristics of browning, softness, mold, rot, leakage, splitting, and shriveling (Conner, 2013; Conner and Maclean, 2012; P. Perkins‐Veazie, personal communication). Both the unmarketable and marketable berries were returned to the appropriate clamshell each week, and storage measurements were discontinued once the percent unmarketable in all three clamshells reached 50%, or after 3 weeks of storage. Each week during storage, berries were sent to the University of Arkansas

Cooperative Extension Service Plant Health Clinic, Fayetteville, AR for disease diagnostics.

Reports from the Clinic were provided on the fungal species isolated.

The remaining three clamshells were used as composition replications. For composition measurements, every 7 d three berries were removed from each of the three clamshells and used to measure berry volume, Chroma, hue, L*, soluble solids, titratable acidity (TA), pH, and firmness of the skin and flesh. Composition measurements were discontinued once the percent unmarketable in all three clamshells reached 50% or after 3 weeks of storage.

Berry and Composition Analysis

45 The composition procedures used were modeled from previously reported

protocols (Conner, 2013; Conner and Maclean, 2012; Giusti and Wrolstad, 2001; Haung et

al., 2009; Sandhu and Gu, 2010; Striegler et al., 2005; Slinkard and Singleton, 1977; Prior et

al., 2003; Threlfall et al., 2005; Threlfall et al., 2007; Walker et al., 2001). Berry volume was

determined by measuring the height and width of three randomly selected berries from

each replication using electronic calipers every 7 d for up to 28 d. The formula used for

calculating the berry volume was: berry volume = 4/3π*berry height*berry width. Change

in berry volume was determined by calculating the percent of volume difference of the

berries during storage from the initial berry volume. Decrease in size during storage is shown with positive values, while an increase in size during storage is shown by negative values.

Titratable acidity and pH were measured by an 877 Titrino Plus (Metrohm AG,

Herisau Switzerland) with an automated titrimeter and electrode standardized to pH 2.0,

4.0, 7.0, and 10.0 buffers. Titratable acidity was determined using 6 g of juice diluted with

50 mL of deionized, degassed water by titration of 0.1 N sodium hydroxide (NaOH) to an endpoint of pH 8.2, and results were expressed as percent tartaric acid. Soluble solids were measured using a Bausch and Lomb, Inc. Abbe Mark II refractometer (Rochester, NY).

Soluble solids, TA, and pH were measured from the juice of the whole berries, strained through cheesecloth to remove any solids.

Exterior skin color measurements were determined on each of the three berries every 7 d using a Chroma Meter CR 300 series (Konica Minolta Holdings Inc., Ramsey, NJ).

The Commission Internationale de I’Eclairage (CIE) Lab transmission “L” value indicates how dark or light the skin is, with 0 being black and 100 being white. Hue angle describes

46 color in angles from 0° to 360°: 0° = red; 90° = yellow; 180° = green; 270° = blue; and 360°

= back to red. Chroma is the aspect of color by which the skin colors appears different from

gray of the same lightness and corresponds to intensity of the perceived color.

Firmness, or the maximum force to penetrate skin and flesh tissues, was determined

using the three whole berries per replication. A TA‐XT2 Texture Analyzer (Stable Micro

Systems, Haslemere, UK) with a 2‐mm‐diameter probe was used to penetrate the skin and

mesocarp tissues (flesh) to a depth of 10 mm in each berry at a rate of 10 mm.s‐1.

Measurements are expressed as force in Newtons (N), and the data was analyzed using

Texture Expert Version 1.17 (Texture Technologies Corp., Scarsdale, NY).

Nutraceutical Analysis

Three randomly selected berries from each composition replication of each treatment were used from the harvest date sample to measure oxygen radical absorbance

capacity (ORAC), total phenolics, total anthocyanin content, total ellagitannins, total

flavonols, and resveratrol content by high‐performance liquid chromatography (HPLC)

with modified methods determined by Cho et al. (2004), Cho et al. (2005), Hager et al.

(2008), and Prior et al. (2003). The berries were homogenized three times each for 1 min in alternating washes of 80 ml of extraction solution containing methanol/water/formic acid (MWF) (60:37:3 v/v/v) and acetone/water/acetic (70:29.5:0.5 v/v/v) to the smallest

particle size using a Euro Turrax T18 Tissuemizer (Tekmar‐Dohrman Corp, Mason, OH).

Homogenates were then centrifuged for 5 min at 10,000 rpm and filtered through

Miracloth (CalBiochem, LaJolla, CA). The samples were taken to a final volume of 250 mL

with extraction solvent and stored at ‐70 °C until further. All samples were passed through

0.45 μm filters (Whatman PLC, Maidstone, UK) prior to HPLC analysis.

47 Total phenolics were measured using the Folin‐Ciocalteu assay (Slinkard and

Singleton 1977) on a diode array spectrophotometer (8452A; Hewlett Packard, Palo Alto,

CA), with a gallic acid standard and a consistent standard curve based on sequential

dilutions. Samples were prepared with 1 ml 0.2N Folins reagent, 0.8 ml Na2CO3 (75g/L) and

0.2 ml of extracted sample with a reaction time of 2 h. Absorbance was measured at 760

nm, and results were expressed as gallic acid equivalents (GAE).

For flavonoid analysis, subsamples (5 ml) of solvent extracts were evaporated to

dryness using a SpeedVac® concentrator (ThermoSavant, Holbrook, NY) with no radiant

heat and suspended in 1 ml of aqueous 3% formic acid solution. Samples (1 mL) were

analyzed using a Waters HPLC system equipped with a model 600 pump, a model 717 Plus

autosampler and a model 996 photodiode array detector. Separation was carried out using

a 4.6 mm × 250 mm Symmetry® C18 column (Waters Corp, Milford, MA) preceded by a 3.9

mm × 20 mm Symmetry® C18 guard column. The mobile phase was a linear gradient of 5%

formic acid and methanol from 2% to 60% for 60 min at 1 ml min−1. The system was

equilibrated for 20 min at the initial gradient prior to each injection. Detection wavelength

was 510 nm for anthocyanins. Individual anthocyanin glycosides were quantified as

delphinidin (Dp), cyanidin (Cy), petunidin (Pt), peonidin (Pn), pelargonidin (Pg), and

malvidin (Mv) glucoside equivalents using external calibration curves of authentic standards obtained from (Sandnes, Norway). Total flavonols and anthocyanins

were calculated as the sum of individual compounds and their derivatives.

For total flavonol and ellagitannin analysis, samples (5 ml) of solvent extracts were evaporated to dryness using a SpeedVac® concentrator with no radiant heat and

suspended in 1 ml of aqueous 50% methanol solution. The samples were then analyzed

48 using a Waters HPLC system (Waters Corp, Milford, MA) equipped with a model 600 pump,

model 717 plus autosampler and model 996 photodiode array detector. Separation was

carried out using a 4.6 mm × 250 mm Aqua® C18 column (Phenomenex, Torrance, CA)

preceded by a 3.0 mm × 4.0 mm ODS® C18 guard column (Phenomenex). The mobile phase

was a gradient of 20 g L−1 acetic acid (A) and 5 g L−1 acetic acid in water and acetonitrile

(50:50 v/v, B) from 10% B to 55% B in 50 min and from 55% B to 100% B in 10 min. The

system was equilibrated for 20 min at the initial gradient prior to each injection. A

detection wavelength of 360 nm was used for flavonols and 280 nm for ellagitannins at a

flow rate of 1 ml min−1. Flavonols and ellagitannins were expressed as mg rutin

equivalents kg−1 fresh weight, and mg ellagic acid equivalents kg‐1 fresh weight, respectively.

Trans‐resveratrol (3,4′,5‐Trihydroxy‐trans‐stilbene, 5‐[(1E)‐2‐(4‐

Hydroxyphenyl)ethenyl]‐1,3‐benzenediol) concentrations were confirmed using an analytical standard (ID:24860876; Sigma‐Aldrich Co. LLC, St. Louis, MO).

For compound confirmation, a representative black and bronze genotype was

analyzed with HPLC/MS. For HPLC/MS analysis the HPLC apparatus was interfaced to a

Burker Esquire (Burker Corporation, Billerica, MA) LC/MS ion trap mass spectrometer.

Mass spectral data were collected with the Bruker software, which also controlled the

instrument and collected the signal at 360 or 510 nm. Typical conditions for mass spectral

analysis in positive ion electrospray mode for anthocyanins and negative ion electrospray

mode for flavonols included a capillary voltage of 4000 V, a nebulizing pressure of 30.0 psi,

a drying gas flow of 9.0 ml min−1 and a temperature of 300 °C. Data were collected in full‐

49 scan mode over a mass range of m/z 50 – 1000 at 1.0 s per cycle. Characteristic ions were

used for peak assignment.

The ORAC of extracts was measured using the method of Prior et al. (2003) modified for use with a FLUOstar Optima microplate reader (BMG Labtechnologies, Durham, NC) using fluorescein as fluorescent probe. Fruit extracts were diluted 1600‐fold or more with phosphate buffer (75 mM, pH 7) prior to ORAC analysis. The assay was carried out in clear

48‐well Falcon plates (VWR, St. Louis, MO). Each well had a final volume of 590 μl. Initially,

40 μl of diluted sample, Trolox (TE) standards (6.25, 12.5, 25, 50 μM) and blank solution

(75 mM, pH 7 phosphate buffer) were added to each well using an automatic pipette. The

FLUOstar Optima instrument equipped with two automated injectors was then

programmed to add 400 μl of fluorescein (0.108 μM) followed by 150 μl of AAPH (31.6

mm) to each well. Fluorescence readings (excitation 485 nm, emission 520 nm) were

recorded after the addition of fluorescein, after the addition of AAPH and every 192 s

thereafter for 112 min to reach 95% loss of fluorescence. Final fluorescence measurements

were expressed relative to the initial reading. Results were calculated based upon

differences in areas under the fluorescein decay curve between the blank, samples, and

standards. The standard curve was obtained by plotting the four concentrations of TE

against the net area under the curve (AUC) of each standard. Final ORAC values were

calculated using the regression equation between TE concentration and AUC and are

expressed as μmol TE equivalents kg−1 fresh weight.

Experimental Design

The storage experiment was a designed split‐split plot with three replications of

each genotype and fungicide treatment. The first split was storage (weeks 0, 1, 2, and 3)

50 and the second split was year (2012 and 2013). The nutraceutical experiment was a split

plot design with three replications of each genotype and treatment, with the split being

year (these measurements were only done on the harvest date, not at each storage date). A

single vine was used as an experimental unit.

Experimental Analysis

The data were analyzed by analysis of variance (ANOVA) using JMP® (version 11.0;

SAS Institute Inc., Cary, NC). Tukey’s Honest Significant Difference and Student’s t Test was

used for mean separations (p = 0.05). Associations among all dependent variables were

determined using multivariate pairwise correlation coefficients of the mean values using

JMP (version 11.0; SAS Institute Inc., Cary, NC).

Results

Initial Attributes

The initial measurement of berry force, TA, pH, soluble solids, L*, Chroma, and hue

angle are of particular importance. Averaged across years and fungicide treatments AM 04

and ‘Nesbitt’ had the highest initial force (11.1 N for both genotypes), while AM 01, AM 15,

and ‘Tara’ had the least (8.9, 8.6, and 8.1 N, respectively) (Table 1.1). TA ranged from 0.5 to

0.6, while pH ranged from 3.4 (AM 01 and ‘Southern Jewel’) to 3.8 (AM 15, ‘Summit’ and

‘Supreme’) and soluble solids ranged from 17.6 % (AM 04) to 21.8 % (‘Supreme’), averaged

across years and fungicide treatments (Table 1.1). AM 01 and AM 04 had the highest L*

values (66.8 and 66.3, respectively), while AM 15, ‘Southern Jewel’, and ‘Tara’ had the

lowest (26.8, 26.7, and 27.7, respectively) (Table 1.1). The bronze genotypes Chroma

values that ranged from 2.2 (AM 15) to 10.1 (AM 01) and hue angles ranging from 70.8°

(‘Summit’) to 291.4° (‘Tara’), while the black genotypes Chroma values ranged from 6.6

51

Table 1.1. Initial values for force to penetrate berry skin, titratable acidity, pH, soluble solids, L*, Chroma, and hue angle averaged across year and fungicide treatment. Berry Titratable Soluble skin Force acidity solids Hue Genotype color (N) (%) pH (%) L* Chroma angle AM 01 Brz 8.9 0.5 3.5 21.3 66.3 9.2 149.3 AM 04 Bl 11.1 0.6 3.8 18.2 61.2 4.0 131.7 AM 15 Br 8.6 0.5 3.4 18.3 66.8 10.1 155.7 AM 27 Bl 9.1 0.6 3.5 18.7 26.8 2.2 272.0 Nesbitt Bl 11.1 0.6 3.7 18.0 27.7 5.3 291.4 Southern Jewel Bl 9.3 0.5 3.4 17.6 43.9 15.0 269.0 Summit Br 9.7 0.6 3.8 21.8 42.1 13.5 70.8 Supreme Bl 9.6 0.6 3.8 18.2 26.7 6.6 306.5 Tara Br 8.1 0.6 3.7 18.8 45.4 13.6 90.5 zBronze = Br and black = Bl.

52 (‘Southern Jewel’) to 15.0 (AM 27) and hue angles that ranged from 90.5° (‘Supreme’) to

306.5° (‘Southern Jewel’) (Table 1.1).

Berry Storage Attributes

The postharvest fruit diseases present were identified as black rot (Guignardia

[Phyllosticta] bidwellii [ampelicida]), myrothecium leaf spot (Myrothecium sp./spp.), and

botrytis fruit rot, (Botrytis sp./spp.). Occasionally an unknown species of fruit fly

(Drosophila sp.) was present, but only in fruit with wet or torn stem scars.

The ANOVA F‐test indicated a significant four‐way interaction of year by week of

storage by genotype by fungicide treatment for weight loss (P=0.0335) and unmarketable

berries (P<0.0001) (Table 1.2). The three‐way interactions of year by week of storage by

genotype and year by genotype by fungicide treatment had significant effects for both force

to penetrate berry skin (P<0.0001 and P=0.0029, respectively) and volume change during

storage (P<0.0001 and P=0.0006, respectively) (Table 1.2).

Weight loss. The dependent variable weight loss, due to the four‐way interaction,

requires close examination of numerous mean values to evaluate and explain the results

(Fig. 1.1). For instance, after 3 weeks of storage, AM 15 fungicide‐treated fruit in 2012 and

‘Nesbitt’ no fungicide‐treated fruit in 2012 had the greatest weight loss (7.1 and 6.5 %, respectively), while ‘Nesbitt’ and ‘Southern Jewel’ from all treatments in 2013 each had the least weight loss (2.2 %) (Fig. 1.1 and Table A.2). The variable response of genotypes for years, with the example of ‘Nesbitt’ highlighted, provides some insight why the interaction was significant. If one looks closer at major trends in the data, it can be seen that overall weight loss values appear to be lower for most genotypes in 2013 compared to 2012 (Fig.

1.1). When the main effect of year is examined, overall mean weight loss in 2012 was 2.6%

53 Table 1.2. F‐test significance from ANOVA for berry weight loss, percent unmarketable fruit, force required to penetrate the berry skin, and percent volume change of the berry during 3 weeks of storage. Highest‐order interactions are italicized and shaded. Source Degrees Weight Unmarketable Force Volume of loss (%) (%) (N) change freedom (%) Year 1 <0.0001 <0.0001 <0.0001 <0.0001 Week 3 <0.0001 <0.0001 <0.0001 0.0436 Year*week 3 <0.0001 <0.0001 <0.0001 <0.0001 Genotype 8 <0.0001 <0.0001 <0.0001 <0.0001 Year*genotype 8 <0.0001 <0.0001 <0.0001 <0.0001 Week*genotype 24 <0.0001 <0.0001 <0.0001 <0.0001 Year*week*genotype 24 <0.0001 <0.0001 <0.0001 <0.0001 Fungicide 1 0.5613 <0.0001 0.0104* 0.7690 Year* fungicide 1 0.6075 0.0021 0.2674 0.9538 Week* fungicide 3 0.7394 <0.0001 0.0782 0.6968 Year*week* fungicide 3 0.9851 <0.0001 0.4468 0.5977 Genotype* fungicide 8 <0.0001 <0.0001 <0.0001 <0.0001 Year*genotype* fungicide 8 0.0002 <0.0001 0.0029 0.0006 Week*genotype* fungicide 24 0.1746 <0.0001 0.5859 0.1137 Year*week*genotype*fungicide 24 0.0335 <0.0001 0.1896 0.1043

54

Fig. 1.1. Percent berry weight loss of fungicide‐ and no fungicide‐treated muscadine genotypes stored at 2 °C for 3 weeks. Values at week 0 (date of harvest) were excluded. Each standard error bar is constructed using 1 standard error from the mean. Treatment Week Fungicide No fungicide 1 2 7 3 6 2012 5 4 3 2 1 Year 0 7 2013 Weight loss (%) loss Weight 6 5 4 3 2 1 0 1 4 5 7 t el t e 1 4 5 7 t el t e ra bit 0 0 1 2 bit mi m a ew em Tara s ew e T J pr M M M e J m pr AM 0 AM 0 AM 1 AM 2 Nes n AM A A A N r Summi Su Su Su thern outhe ou S S

Genotype

55 and in 2013 was 1.4% (and was significant in the ANOVA F test) (Table 1.3). If storage time

(week) overall means are examined across years, fungicide treatment, and genotype,

means for weeks (week 0, 1, 2, and 3 were 0%, 1.4%, 2.7%, 4.0%, respectively) were all

significantly different, and the values for weight loss for weeks 2 and 3 were greater than

the higher values for year effect (Table 1.3). For genotypes, the overall genotype means

ranged from 1.6% for AM 27 to 2.4% for AM 01 with a number of significant differences

among the means (Tables 1.5 and A.2). Main effect of fungicide treatment was not

significant for weight loss, indicating minimal contribution of this source of variation to the

interaction means (Table 1.3).

Unmarketability. For dependent variable unmarketable berries, again the four‐way

significant interaction provided for a challenging presentation and explanation of the major

sources of variation. AM 01 no fungicide‐treated berries in 2013 and ‘Nesbitt’ fungicide‐

treated fruit in 2012 had the greatest amount of unmarketable berries after 3 weeks of

storage (94.9% and 81.7%, respectively), while AM 04 fungicide‐ and no fungicide‐treated

berries in 2013 and ‘Summit’ fungicide‐treated berries in 2012 had the least amount of

unmarketable berries after 3 weeks (12.6% and 14.5%, respectively) (Fig. 1.2 and Table

A.2). To further examine the effects of each source of variation individually, again an examination of year shows that mean unmarketable fruit for 2012 was 17.1% and 2013 was 18.5%, with these means significantly different (P=0.0001) (Table 1.3). Fungicide main effect was greater, with treated vines having an average unmarketable value of 16.5%

compared to 19.2% for untreated, again with high significance (P<0.0001) and of greater

magnitude (thus more important in contribution) than year (Table 1.3). For storage time,

large differences among week means were found, ranging from 0.0% for week 0, 9.3% for

56 Table 1.3. F‐test main effect significance for year on weight loss, percent unmarketable, force required to penetrate the berry skin, and percent volume change of the berry during three weeks of storage, averages across week and fungicide. Source Weight Unmarketable Force (N) Volume loss (%) (%) change (%) Year 2012 2.6 17.1 6.5 0.5 2013 1.4 18.5 8.4 3.6 p value <0.0001 <0.0001 <0.0001 <0.0001 Treatment Fungicide 2.0 16.5 7.5 2.2 No fungicide 2.0 19.2 7.3 1.9 p value 0.5613 <0.0001 0.0104 0.7690 Week 0 0.0 dz 0.0 d 9.5 a 0.0 b 1 1.4 c 9.3 c 7.1 b 2.6 a 2 2.7 b 19.6 b 6.9 b 3.6 a 3 4.0 a 42.4 a 6.3 c 1.9 ab p value <0.0001 <0.0001 <0.0001 0.0436 Genotype AM 01 2.4 az 25.0 ab 6.5 de ‐4.1 c AM 04 2.2 ab 14.6 cde 9.9 a ‐2.4 c AM 15 2.1 ab 20.2 abc 6.3 ef ‐3.5 c AM 27 1.6 c 19.4 bcd 6.5 de 6.0 ab Nesbitt 2.1 ab 16.7 cde 8.8 b 8.0 a Southern Jewel 1.8 bc 14.7 cde 7.4 cd 0.7 abc Summit 1.8 bc 13.6 de 7.9 bc 0.3 bc Supreme 2.2 ab 10.4 e 8.3 bc 7.6 ab Tara 1.9 abc 25.7 a 5.4 f 5.7 ab p value <0.0001 <0.0001 <0.0001 <0.0001 zMeans followed by the same letter are not significantly different at α= 0.05; separated by Tukey’s HSD.

57

Fig. 1.2. Percent unmarketable fungicide‐ and no fungicide‐treated fruit of muscadine genotypes stored at 2‐3 °C for 3 weeks. Values at week 0 (date of harvest) were excluded. Each standard error bar is constructed using 1 standard error from the mean. Treatment Week Fungicide No fungicide 1 2 80 70 3 60 2012 50 40 30 20 10 Year 0 80 70 2013

Unmarketable (%) 60 50 40 30 20 10 0 1 4 5 7 tt el it e ra 1 tt el e a 0 1 2 bi w m a 0 15 27 m e mm re T Tar M M 0 M M es J u p M M esbi Jew A A A A N S AM AM 04 A A N n Summit upre ern Su er S h h out S Sout

Genotype

58 week 1, 19.6% for week 2, and 42.4% for week 3, with all means significantly different

from each other (P<0.0001) (Table 1.3). Further, there was a substantial range among

genotypes for this variable, with main effect means ranging from a low of 10.4% for

‘Supreme’ to 25.7% for ‘Tara’ (Table 1.3). Some genotypes, such as AM 04, in 2013, and

‘Summit’ in 2012 had lower increases in unmarketable fruit as length of storage increased,

while many genotypes had major increases in unmarketable values particularly in week 3

(Tables 1.3 and A.2). Those genotypes that have less unmarketable berries during storage

with consistent performance are valuable to identify examination of cultivar potential.

Force. Force to penetrate berry skin, with increasing force indicating the firmness of

the berry and ability to retain this characteristic in storage, had two three‐way interactions

in the ANOVA with both year and genotype sources of variation in both interactions (Table

1.2). In examining the year by week by genotype interaction means, one can see that after

3 weeks of storage, AM 15 in 2012 and ‘Tara’ in 2013 required the least amount of force to

penetrate the berry skin (2.3 and 3.9 N, respectively), while AM 04 in 2013 required the

most force to penetrate the berry skin (11.4 N) (Fig. 1.3 and Table A.2). The means for the

year by genotype by fungicide interaction indicated both fungicide‐ and no fungicide‐

treated AM 04 in 2013 required the highest force to penetrate the berry skin (11.9 and 11.7

N, respectively), while both fungicide and no fungicide treated ‘Tara’ and AM 15 in 2012

required the least (4.9, 4.8, and 5.1 N, respectively) (Fig. 1.4 and Table A.2). In further

examining contributions of the sources of variations to mean differences, year had almost a

2.0 N difference and was significant (P=0.0001) (Table 1.3), while fungicide main effect difference was only 0.2 N difference for treated verses untreated vines (P=0.0104) (Table

1.3). Main effect mean for force for weeks of storage ranged from 9.5 N for week 0 to 6.3 N

59 Fig. 1.3. Force to penetrate skin of muscadine genotypes stored at 2‐3 °C for 3 weeks averaged across fungicide treatment. Each standard error bar is constructed using 1 standard error from the mean.

12 Week 10 0 8 2012 1 6 2 4 3

2 Year 0 ce (N) r 12 Fo 10 8 2013 6 4 2 0 l 4 5 itt it ra 0 1 27 we a M 01 M M esb T A AM A A N Je preme rn Summ Su

Southe

Genotype

hbi diddfh

60 Fig. 1.4. Force to penetrate berry skin of fungicide‐ and no fungicide‐treated muscadine genotypes average across weeks. Each standard error bar is constructed using 1 standard error from the mean.

Treatment 10 Fungicide 8 2012 No fungicide 6 4

2 Year 0

Force (N) 10

8 2013 6 4 2 0 1 7 tt l it e a 0 2 i ar T Jewe AM AM 04 AM 15 AM Nesb Summ ern Suprem th u o S

Genotype

61 for week 3, although the means did not differ for weeks 1, 2, and 3 (Table 1.3). Genotype main effect means ranged from a low of 5.4 N for ‘Tara’ to 9.9 N for AM 04 (Table 1.3).

These findings indicate that genotype and week were greater contributors as sources of variation.

Volume Change. Volume change resulted from a decrease in berry size shown with positive numbers, while an increase in size shown by negative numbers. The same three way interactions for force were found for volume change (Table 1.2). For the year by week by genotype interaction, data after 3 weeks of storage for ‘Nesbitt’ in 2013 indicated the greatest decrease in volume (21.1%), while AM 01 in 2012 had the greatest increase in volume (‐19.0%) after 3 weeks of storage (Fig. 1.5 and Table A.2). For the other three‐way interaction, ‘Nesbitt’ fungicide‐ and no fungicide‐treated berries in 2013 and ‘Supreme’ no fungicide‐treated and AM 27 fungicide‐treated fruit in 2012 had the greatest volume change (13.3, 14.9, 14.1, and 13.9%, respectively), while AM 01 no fungicide treated and

AM 15 fungicide treated fruit in 2012 had the least volume change (‐16.9 and ‐14.1%, respectively), across storage time (Fig. 1.6 and Table A.2). For main effects, fungicide did not impact volume change (Table 1.3), while year had a 2.1% mean difference with 2013 being significantly higher (Table 1.3). Week had a difference of 3.6% between week 0 and week 2, and genotype main effect means differed by 9.8% change, the greatest of any variables (Table 1.3).

The ANOVA F‐test indicated a significant four‐way interaction of year by week of storage by genotype by fungicide treatments for the dependent variable TA (P=0.0040)

(Table 1.4). Soluble solids content and pH had a significant F‐test for the three‐way

Berry Composition

62 Fig. 1.5. Percent change in volume of muscadine genotypes stored at 2‐3 °C for 3 weeks averaged across fungicide treatment. Values at week 0 (date of harvest) were excluded. Decrease in size shown with positive values, while an increase in size shown by negative values. Each standard error bar is constructed using 1 standard error from the mean. Week 20 1 10 2012 2 0 3 -10 -20 -30 Year

20

10 2013 Volume change (%) 0 -10 -20 -30 l 1 4 5 7 itt e it e ra 0 0 1 2 b w m a s e m rem T M M M e u p AM A A A N n J S er Su th ou S

Genotype

63 Fig. 1.6. Percent change in volume of muscadine genotypes with fungicide and no fungicide treatments. Decrease in size shown with positive values, while an increase in size shown by negative values. Each standard error bar is constructed using 1 standard error from the mean.

20 Treatment 10 Fungicide 2012 0 No fungicide -10 -20 -30 Year

20 10 2013

Volume change (%) 0 -10 -20 -30 t l t 4 7 it i e ra 01 0 15 2 b m m a s ewe re T M M M e um p AM A A A N S ern J Su outh S

Genotype

64 interaction for year by week of storage by genotype (P=0.0347 and P=0.0011, respectively)

and with year by genotype by fungicide treatments (P<0.0001 and P=0.0002, respectively)

(Table 1.4). There were significant two‐way interactions of year by genotype (P<0.0001),

week of storage by genotype (P=0.0394), and year by fungicide treatments (P=0.0197) for

the dependent variable SS/TA (Table 1.4).

Titratable Acidity. Fungicide‐treated berries of ‘Supreme’ at harvest in 2013 had the

highest TA (0.64%), while no fungicide‐treated berries of ‘Southern Jewel’ stored for 2 weeks and fungicide‐treated berries of AM 01 at date of harvest in 2012 had the lowest TA

(0.47% and 0.48%, respectively) (Fig. 1.7 and Table A.3). In 2012, TA was 0.56% and in

2013 TA was 0.57% resulting in a significant difference (<0.0001) (Table 1.4) when averaged across genotype, week, and fungicide treatments. For each week of storage (0, 1,

2, and 3), TA was 0.57% when averaged across year, genotype, and fungicide treatments, with no significant differences. Based on the lack of differences in week, year, and fungicide treatment one can see that genotype was the greatest contributor as a source of variation for TA (Table 1.4 and Fig. 1.7).

pH. The fruit of ‘Supreme’ at date of harvest in 2013 had the overall highest pH (4.1),

while the fruit of AM 15 stored for 3 weeks and ‘Southern Jewel’ at date of harvest in 2012

had the lowest pH (3.3) across fungicide treatments (Fig. 1.8 and Table A.3). The fungicide‐

treated fruit of ‘Supreme’ in 2013 had the highest pH (3.9), while the fungicide‐treated fruit

of AM 15 and ‘Southern Jewel’ in 2012 had the lowest pH (3.3) across storage time. (Fig. 1.9

and Table A.3). When averaged across genotype, year, and fungicide treatment, the pH for

each week of storage (0, 1, 2, and 3) was 3.6. In 2012 the average pH was 3.6, while in 2013

it was 3.7, which resulted in a significant difference (<0.0001) (Table 1.4), but only a

65

y 0.0394 0.0197 0.6824 SS/TA <0.0001

0.1222 (%) solids 0.0347 0.9177 0.9016 Soluble <0.0001

z e italicized and table acidity, iation of muscadine (%) 0.0040 acidity 0.1998 0.0166 0.7214 0.3470 0.1447

0004 0.3277 0.0006 0.1163 pH Titratable 0.0011 0.0002

4 0.4267 0.5923 0.4108 0.9566 of Degrees freedom Table 1.4. F‐test significance from ANOVA for pH, percent titra percent soluble solids, and SS/TA ratio for four sources of var grapes during 3 weeks of storage. Highest‐order interactions ar shaded. Source Year Week Year*week Genotype Year*genotype 3 1 8 3 8 0.0174 <0.0001 <0.0001 0.3353 <0.0001 <0.0001 <0.0001 0.0077 <0.0001 <0.0001 0.9758 <0.0001 0.2292 <0.0001 <0.0001 0.2180 <0.0001 0.0875 0.0172 Week*genotype Year*week*genotype Fungicide Year* fungicide Week* fungicide Year*week* fungicide Genotype* fungicide 24 Year*genotype* fungicide 24 Week*genotype* fungicide Year*week*genotype*fungicide 0.0633 3 24 1 8 3 2 0.00134 1 8 0.0534 0.1927 0.7435 0.3949 0.7408 0.5340 0. 0.1684 0.3924 0.8939 0.1343 0.2749 0.1289 0.1673 0.0568 Soluble solids to titratable acidity ratio Titratable acidity expressed as tartaric acid

z y

66

0 1 2 3

Week Year 2012 2013

ra

Ta

e

m

e

r

p

u

t

i S

m

m

u

S

l

e

w

e

J

n

r t

t e

i

h b

t

u es

o

N

S

error 7

2

No fungicide M

A

5

1 cide‐treated

Titratable

AM

4

0

M

A

1

0

M

A ra Ta

Genotype Treatment

e

m

e

r

p

u

t

i S

m

m

u

3 °C for 3 weeks. Each standard S

l

e

w

e

J

dity of fungicide‐ and no fungi n

r t

t

e

i

tandard error from the mean. h

b

t

u es

o

N S

Fungicide 7

2

M

A

5

1

AM

4

0

M

A

1

0

AM

0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.6 0.5 0.4 0.3 0.2 0.1 0.0 Fig. 1.7. Percent titratable aci muscadine genotypes stored at 2‐ bar is constructed using 1 s acidity measured as tartaric acid. Titratable acidity (%)

67

0 1 2 3

Week Year 2012 2013

a

r a

T

e

m

re

p

u S

t i

m

s. Each standard m u S

l e

w

e

J

n

r

e h

t

u

o S

t

t

esbi N

Genotype

7 2

AM

5

1

M A

4

0 M

A

1

0

M A

Fig. 1.8. pH of muscadine genotypes stored at 2‐3 °C for 3 week error bar is constructed using 1 standard error from the mean. 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 pH

68 0.1 difference in means. The lack of differences in week, year, and fungicide treatment

indicated that genotype was the greatest contributor as a source of variation for pH (Table

1.4 and Figs. 1.8 and 1.9).

Soluble Solids. Across fungicide treatments, ‘Summit’ harvest, after 2 weeks of

storage and after 3 weeks of storage, and AM 01 after 1 week of storage in 2012 had the

highest soluble solids (26.8, 25.9, 25.1, and 25.1%, respectively) (Fig. 1.10). Across storage time and fungicide and no fungicide treatments, the fruit of ‘Summit’ and AM 01 has the highest SSC (25.6, 25.6, 25.5, and 23.9, respectively) (Fig. 1.11). The difference in means of soluble solids for year was 3.5%, when averaged across fungicide treatments, genotype, and week of storage. The difference in soluble solids for fungicide‐treated and no fungicide‐ treated fruit was 0.1% averaged across genotypes and year, week, and genotype. For soluble solids, no differences among week means were found (Table 1.4), with means of

18.9% for week 0, 19.2% for week 1, 19.3% for week 2, and 19.2% for week 3. Based on minimal difference of week and fungicide treatment, one can see that genotype and year were the greatest contributors as sources of variation for soluble solids (Table A.3 and

Figs. 1.10 and 1.11).

Soluble Solids/Titratable Acidity Ratio. The berries of ‘Supreme’ in 2013 had the

highest SS/TA ratio (83.5), while ‘Southern Jewel’, ‘Supreme’, and AM 04 in 2013 had the

lowest (43.3, 42.9, and 42.5, respectively), across storage time and fungicide treatments

(Table 1.5). The fruit of ‘Supreme’ at harvest had the highest SS/TA ratio (68.7), while the

fruit of AM 15 after 3 weeks of storage had the lowest (47.8), across fungicide treatments

and year (Table 1.6). Across all genotypes, both fungicide‐ and no fungicide‐treated berries

harvested in 2013 had higher SS/TA ratios (67.2 and 63.1, respectively) than the fungicide

69

0 1 2 3

Week Year 2012 2013

Tara

e

m

re

p

u at S

g 1

t

i

mm u

S

l

e

w

e

J

n

r

e

h

t

u

t

o

t

i S

b

es N

Genotype

7 error bar is constructed usin 2 AM

ds of muscadine genotypes stored

5 1

AM

4 0

AM

1 0

AM

5 0 5 0

20 15 10 20 15 10

Fig. 1.10. Percent soluble soli 2‐3 °C for 3 weeks. Each standard standard error from the mean. Soluble solids (%)

70

Fungicide No fungicide Treatment Year

2012 2013

a

ar

T

e

m

re

p

he mean. u S

e‐applied muscadine genotypes. Each

t i

m

m

u

S

l

e

w

e

J

n

r

e

h

t

u

t

o

t S

esbi N

Genotype

7

2

d using 1 standard error from t AM

ds of fungicide‐ and no fungicid

5

1

M

A

4

0

M

A

1

0

AM 5 0 5 0

20 15 10 20 15 10 Fig. 1.11. Percent soluble soli standard error bar is constructe Soluble solids (%)

71 Table 1.5. Soluble solids to titratable acidity ratio for muscadine genotypes. Year Genotype 2012 2013 AM 01 55.8 dez 49.9 e‐g AM 04 42.5 g 66.8 bc AM 15 46.2 e‐g 53.7 d‐f AM 27 45.5 fg 60.4 cd Nesbitt 45.3 fg 66.7 bc Southern Jewel 43.3 g 62.7 cd Summit 54.5 d‐f 73.4 b Supreme 42.9 g 83.5 a Tara 47.9 e‐g 69.2 bc zMeans followed by the same letter are not significantly different at α= 0.05; separated by Tukey’s HSD.

Table 1.6. Soluble solids to titratable acidity ratio of muscadine genotypes stored at 2‐3 °C for 3 weeks. Week Genotype 0 1 2 3 AM 01 58.2 c‐iz 49.0 kl 52.3 g‐l 52.1 g‐l AM 04 58.5 c‐h 53.4 f‐l 51.6 h‐l 55.1 e‐l AM 15 50.9 i‐l 48.8 kl 52.1 g‐l 47.8 l AM 27 56.0 e‐k 51.2 h‐l 52.3 g‐l 52.4 g‐l Nesbitt 50.4 j‐l 52.6 g‐l 60.2 b‐f 60.7 b‐f Southern Jewel 49.8 j‐l 49.0 kl 59.3 c‐g 53.8 f‐l Summit 64.9 a‐c 59.1 c‐g 64.7 a‐c 67.0 ab Supreme 68.7 a 63.5 a‐d 63.6 a‐d 57.1 d‐j Tara 61.8 a‐e 60.5 b‐f 58.0 c‐i 54.0 f‐l zMeans followed by the same letter are not significantly different at α= 0.05; separated by Tukey’s HSD.

73 and no fungicide fruit harvested in 2012 (46.9 and 47.3, respectively) (Table 1.7). Across fungicide treatments, genotypes, and weeks of storage, the differences in year means of

SS/TA ratio were 18.0. The mean SS/TA was 57.0 for fungicide treated fruit, and 55.2 for the no fungicide treatments, with no significant differences. Week was a significant source of variation for SS/TA ratio (P=0.0172), week means of 57.7 for week 0, 54.1 for week 1,

57.1 for week 2, and 55.5 for week 3. Similar to soluble solids, genotype and year were the greatest contributors as sources of variation for SS/TA due to minimal differences in week and fungicide treatment.

Berry Color

The ANOVA F‐test indicated significant three‐way interactions of year by week of storage by genotype for Chroma (P<0.0001), hue (P<0.0001), and L* (P<0.0001) (Table

1.8). Additionally, the three‐way interactions of year by week of storage by fungicide treatment and year by genotype by fungicide treatments were significant (P=0.0494 and

P=0.0316, respectively) (Table 1.8). This data includes both black and bronze genotypes.

L* Value. Across fungicide treatments, AM 01 after 1 week of storage in 2013 had the highest L* value (103.4), while AM 04 after 1, 2, and 3 weeks of storage in 2013 had the lowest (24.9, 24.8, and 24.4, respectively) (Fig. 1.12). Across all genotypes, at week 0 for fungicide‐ and no fungicide‐treated in 2012 and 2013 had the highest (41.8, 42.2, 48.3, and

48.3, respectively), while both fungicide‐ and no fungicide‐treated berries stored for 1, 2, and 3 weeks in 2012 had the lowest L* values (33.4, 33.0, 32.8, 33.5, 33.5, and 35.5, respectively) (Table 1.9). Across all weeks of storage, fungicide and no fungicide treated fruit of AM 01 in 2013 had the highest L* values (90.1 and 91.2, respectively), while fungicide‐ and no‐ fungicide‐treated AM 04 in 2013 had the lowest L* values (25.2 and

74

Table 1.7. Soluble solids to titratable acidity ratio of fungicide and no fungicide treatments. Treatment Year Fungicide No Fungicide 2012 46.9 b 47.3 b 2013 67.2 a 63.1 a

Table 1.8 F‐test significance from ANOVA for berry Chroma, hue, and L* values of the four sources of variation of muscadine grapes during 3 weeks of storage. Highest‐order interactions are italicized and shaded. Degrees of Chroma Hue L* Source freedom Year 1 <0.0001 <0.0001 <0.0001 Week 3 0.0019 0.7518 <0.0001 Year*week 3 0.1017 0.0053 <0.0001 Genotype 8 <0.0001 <0.0001 <0.0001 Year*genotype 8 <0.0001 <0.0001 <0.0001 Week*genotype 24 <0.0001 <0.0001 <0.0001 Year*week*genotype 24 <0.0001 0.0010 <0.0001 Fungicide 1 0.0926 0.5091 0.9513 Year* fungicide 1 0.1220 0.1067 0.1677 Week* fungicide 3 0.1923 0.8424 0.0900 Year*week* fungicide 3 0.8201 0.4554 0.0494 Genotype* fungicide 8 0.0076 0.3341 0.8834 Year*genotype* fungicide 8 0.0569 0.1807 0.0316 Week*genotype* fungicide 24 0.0533 0.6010 0.1244 Year*week*genotype*fungicide 24 0.2848 0.7271 0.4622

75 Fig. 1.12. L* values of muscadine genotypes stored at 2‐3 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean. Week 0 80

2012 1 60 2 40 3

20 Year

L* 0

80 2013 60 40 20 0 t l t 1 4 5 7 it e i e 0 0 1 2 w m M e mm re Tara AM AM A AM esb J u p N n S u er S th u So

Genotype

Table 1.9. L* values of fungicide and no fungicide treatments during storage at 2‐3 °C for 3 weeks Week Year Treatment 0 1 2 3 2012 Fungicide 41.8 az 33.4 b 32.8 b 33.5 b 2012 No fungicide 42.4 a 33.0 b 33.5 b 33.5 b 2013 Fungicide 48.3 a 38.1 ab 39.8 ab 36.9 ab 2013 No fungicide 48.3 a 38.8 ab 39.7 ab 35.2 b zMeans followed by the same letter are not significantly different at α= 0.05; separated by Tukey’s HSD.

76 25.1, respectively) (Fig. 1.13). For L* values, large differences among week means were found, ranging from 45.2 for week 0, 35.9 for week 1, 36.4 for week 2, and 34.8 for week 3, with week 0 means being significantly different from week 1, 2, and 3 means (P<0.0001).

Across genotype, year, and week there were no differences in fungicide‐ and no‐fungicide‐ treatment means. Across genotype, storage, and fungicide treatment, L* value means were greater in 2013 than in 2012 with a significant difference of 5.2. Like Chroma and hue angle, genotype and year were major contributors as sources of variation for L*, however week was also a major contributor as a source of variation.

Chroma. The cultivar Southern Jewel at date of harvest in 2013 had the highest

Chroma value (23.3), while AM 27 after 1 and 3 weeks of storage in 2013 had the lowest

(1.8 and 1.9, respectively) (Fig. 1.14 and Table A.4). The difference in mean Chroma values for years was 1.0, across genotypes, fungicide treatments, and weeks of storage with

Chroma values generally being greater in 2012. Chroma was highest at date of harvest and then decreased during storage, with means of 8.8 for week 0, 7.6 for week 1, 7.6 for week 2, and 7.4 for week 3, with week 0 means significantly different from the other weeks of storage (P<0.0001). The difference in Chroma value means for fungicide treatments across year, week, and genotype was 0.2, which was not significant. The major contributors as sources of variation for Chroma were genotype and year, which was most strongly illustrated by ‘Southern Jewel’ having a Chroma value of 23.3 in 2013 and 6.7 in 2012 at week 0 (Fig 1.14 and Table A.4).

Hue Angle. Hue angle was highest for ‘Supreme’ at harvest and after 1 week and 3 weeks of storage in 2012 (360°, 360°, and 360°), while ‘Summit’ at date of harvest in 2012 lowest (24.4°). The difference in year means for hue angle was 35.7°, with the values in

77 Fig. 1.13. L* values of fungicide‐ and no fungicide‐treated muscadine genotypes. Each standard error bar is constructed using 1 standard error from the mean. Treatment 80 Fungicide 70 No fungicide 60 2012 50 40 30 20 10 Year

L* 0 80 70 60 2013 50 40 30 20 10 0 t l t 1 4 5 7 t e i e ra 0 0 1 2 bi w m m M M s e m re Ta AM A AM A e J u p N n S u er S th u o S

Genotype

78 Fig. 1.14. Chroma of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean. Week 350 0 300 1 250 2012 200 2 150 3 100

50 Year 0 350 Hue angle 300

250 2013 200 150 100 50 0 t l t 1 4 5 7 it e i e ra 0 0 1 2 w m m a M M M M e m re T A A A A esb J u p N n S u er S th u o S

Genotype

79 2013 being significantly higher (<0.0001) (Table 1.8 and Fig. 1.15). Hue angle means were not different during any week of storage, with means of 192.9° for week 0, 183.9° for week

1, 184.1° for week 2, and 190.9° for week 3. The difference in fungicide and no fungicide treatment means was 5.6°, which was not a significant difference (Table 1.8). Similar to

Chroma, the major contributors as sources of variation for hue angle were genotype and year.

Nutraceutical Content

The ANOVA F‐test indicated significant three‐way interactions of year by genotype by fungicide treatment for total anthocyanins (P<0.0001), ORAC (P<0.0001), and total flavonols (P<0.0001) (Table 1.10). The ANOVA F‐test also indicated a significant two‐way interaction of year by fungicide treatment for total ellagitannins (P=0.0078), total phenolics

(P=0.0229), and resveratrol concentrations (P<0.0001) (Table 1.10). Additionally, there was a significant two‐way interaction of genotype by fungicide treatment for total ellagitannins (P=0.0180) (Table 1.10).

Total Anthocyanins. The fungicide‐treated AM 04 and the no fungicide‐treated AM

27 in 2012 had the highest levels of the dependent variable total anthocyanins (127.8 and

122.0 mg/100g, respectively), and anthocyanins were not detected in bronze genotypes

(AM 01, AM 15, ‘Summit’, and ‘Tara’) (Fig. 1.16). Across genotype and fungicide treatment, there was a significant difference in mean total anthocyanin concentrations of 19.3 mg/100 g, with total anthocyanins being greater in 2012 than in 2013 (P<0.0001) (Table 1.10).

Although not significant, the difference in anthocyanins means across genotypes and year for fungicide‐treated and no fungicide‐treated was 2.4. mg/100g. Based on the lack of differences in fungicide treatments, genotype and year were major contributors as sources

80

Fig. 1.15. Hue angle of muscadine genotypes stored at 2‐3 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean. Week 0 300 250 2012 1 200 2 150 100 3 50 Year 0

Hue angle 300 250 2013 200 150 100 50 0 t l t 1 4 5 7 t e i e ra 0 0 1 2 bi w m M e mm re Ta AM AM A AM es J u p N n S u er S th u o S

Genotype

81

<0.0001 (mg/100 g) Resveratrol

Total 0.0229 0.4194 0.1547 phenolics (mg/100 g)

Total <0.0001 flavonols (mg/100 g)

/g) Trolox <0.0001 <0.0001 <0.0001 <0.0001 0.0753 0.1988 0.6127 equivalents ORAC (µmol ns, total are italicized. eratrol concentrations

Total 0.0078 0.0180 0.3412 (mg/100 g) ellagitannins

Total <0.0001 (mg/100 g) anthocyanins rom ANOVA for total anthocyani nols, total phenolics, and resv on. Highest‐order interactions z

DF Degrees of freedom. for the three sources of variati Table 1.10. F‐test significance f ellagitannins, ORAC, total flavo z

Year Genotype Year*fungicide 8 8 1 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 0.0019 <0.0001 0.0985 Fungicide Year*fungicide Genotype*fungicide 8 1 1 <0.0001 0.4093 0.2054 0.7754 0.0202 0.0004 <0.0001 0.3379 0.0012 0.6782 0.3533 0.3597 0.1144 Year*genotype*fungicide 8

82

Fig. 1.16. Total anthocyanin content of fungicide‐ and no fungicide‐treated muscadine genotypes. Each standard error bar is constructed using 1 standard error from the mean. Year 120 2012 100 Fungicide 2013 80 60

40 Treatment 20 0

120 No fungicide hocyanins (mg/100 g)

t 100 80 al an

t 60

To 40 20 0 t l t 1 4 5 7 t e i e ra 0 1 2 w m m a M M e m re T AM 0 A A AM esbi J u p N n S u er S th u o S

Genotype

83 of variation for total anthocyanin concentrations.

ORAC. The fungicide‐treated AM 27 in both 2012 and 2013 had the greatest ORAC values (125.3 and 119.0 µmol Trolox equivalents/g, respectively), while the no‐fungicide‐ treated ‘Tara’ in 2012 had the lowest ORAC values (47.7 µmol Trolox equivalents/g) (Fig.

1.17). The difference in ORAC value means across genotype and year for fungicide‐ and no fungicide‐treated was 6.5 µmol Trolox equivalents/g, with the overall fungicide‐treated fruit having significantly higher ORAC values (P<0.0001). Across fungicide treatments and genotype, the mean difference in ORAC for year was 5.9 µmol Trolox equivalents/g, with

2013 having significantly higher ORAC levels (P<0.0001) (Table A.5 and Fig. 1.17). Thus, year, genotype, and fungicide treatment were all identified as major contributors of sources of variation for ORAC levels (Fig. 1.17).

Total Flavonols. No fungicide‐treated ‘Summit’ in 2012 had the highest total flavonol concentrations (63.1 mg/100 g), while fungicide‐treated ‘Supreme’ in 2012 had the lowest

(4.9 mg/100 g) (Fig. 1.18). The fungicide‐ and no fungicide‐treated berries of 2013 had the highest levels of total flavonols (9.2 and 7.9 mg/100 g), while the fungicide and no fungicide‐treated fruit in 2012 had the lowest total flavonol content (5.8 and 4.8 mg/100 g). Across genotype and fungicide treatments, the mean difference of total flavonols for year was 8.2 mg/100g, with 2012 having significantly higher levels overall (P<0.0001). The overall difference in total flavonol means for fungicide and no fungicide treatments was 4.1 mg/100g, with the fungicide treatment having significantly greater levels across year and genotype (P=0.0012). For the dependent variable total flavonols, year, genotype, and fungicide treatment are all major contributors of sources of variation (Table 1.10 and Fig.

1.18).

84 Fig. 1.17. Oxygen radical absorbance capacity of fungicide‐ and no fungicide‐treated muscadine genotypes. Each standard error bar is constructed using 1 standard error from the mean. Year 120 2012 100 Fungicide 2013 80 60

40 Treatment 20 0

120 No fungicide 100 80 60 40 ORAC (µmol Trolox equivelants/g) 20 0 1 4 5 7 tt el it e a 0 0 2 w m ar M 1 e mm re T AM AM A AM esbi J u p N n S u er S th u o S

Genotype

85

Fig. 1.18. Total flavonol concentrations of fungicide‐ and no fungicide‐treated muscadine genotypes. Each standard error bar is constructed using 1 standard error from the mean.

86 Total Phenolics. Total phenolic concentrations averaged for genotype across year and fungicide treatments ranged from 627.5 mg/100 g (‘Southern Jewel’) to 351.4 mg/100 g (‘Supreme’), with significant differences for genotypes (Table A.5). Fungicide‐treated berries in 2012 had the highest total phenolic concentration (574.3 mg/100g), while the fungicide‐ and no fungicide‐treated berries in 2013 had the lowest (512.5 and 503.8 mg/100 g, respectively) (Tables 1.11 and A.5). Though not significant, the difference in total phenolic means for fungicide treatments was 15.8 mg/100g, across years and genotype. Across genotype and fungicide treatment, the differences in means of years for the dependent variable of total phenolics was 54.7 mg/100g, with 2012 having significantly higher concentrations than 2013 (P=0.0019) (Table 1.11). Year and genotype were identified as major contributors of sources of variation for total phenolics, while fungicide had less contribution (Table 1.11).

Resveratrol. Resveratrol concentrations, across year and fungicide treatment, ranged from 10.4 mg/100g (AM 27) to 3.8 mg/100 g (AM 15), with significant differences occurring for genotypes (<0.0001) (Table A.5). The no fungicide‐treated fruit in 2012 had the highest resveratrol concentrations (7.2 mg/100 g), while the fungicide‐treated fruit in

2013 had the lowest resveratrol concentrations (4.7 mg/100 g) (Table 1.12). The difference in mean resveratrol concentrations for year was 1.6 mg/100 g, across genotype and fungicide treatment with higher levels found in 2012, though not significant. Across year and genotype, the difference in means for the dependent resveratrol concentrations was 0.95 mg/100 g. In 2012 the difference in mean resveratrol concentration for fungicide and no fungicide treatments was 0.4 mg/100 g, while in 2013 the difference was 1.5

87 Table 1.11. Total phenolic concentations (mg/100 g) of fungicide‐ and no fungicide‐treated muscadines. Year Treatment 2012 2013 Fungicide 574.3 az 512.5 b No Fungicide 551.5 ab 503.8 b zMeans followed by the same letter are not significantly different at α= 0.05; separated by Tukey’s HSD.

Table 1.12. Resveratrol concentations (mg/100 g) of fungicide‐ and no fungicide‐treated muscadines. Year Treatment 2012 2013 Fungicide 6.8 az 4.7 b No Fungicide 7.2 a 6.2 a zMeans followed by the same letter are not significantly different at α= 0.05; separated by Tukey’s HSD.

88 mg/100 g, with the no fungicide treatment having significantly higher levels (Table 1.12).

The independent variables of year and genotype were identified as overall major contributors of sources of variation for resveratrol concentrations, while in 2013 fungicide was also a contributor to variation.

Total Ellagitannins. Across years, the fungicide‐treated fruit of ‘Summit’ and AM 04 had the highest total ellagitannin concentrations (13.6 and 13.5 mg/100g), while the no fungicide‐treated AM 01 had the lowest (3.1 mg/100 g) (Fig. 1.19 and Table A.5). The difference in means of years across fungicide treatments and genotypes for the dependent variable of total ellagitannins was 3.3 mg/100g, with 2013 having significantly higher concentrations than 2012 (<0.0001). The difference in means for the fungicide treatments across genotypes and year was 1.1 mg/100g, though not significantly different (Table

1.13). Based on interactions, it appears that fungicide treatment as a source of variation was dependent on year and genotype (Table 1.13 and Fig. 1.19). Like total phenolics, the major contributors as sources of variation for total ellagitannin concentrations were genotype and year.

Correlations

Though some strong correlations did occur, the dependent variables measured were generally minimally correlated (Table A.6). Force to penetrate the berry skin was negatively correlated with percent unmarketable (r=‐0.73) (Table A.6). SS/TA was positively correlated with TA (r=0.73) and pH (r=0.74) (Table A.6). The dependent variable pH was strongly correlated with TA (r=0.99) and negatively correlated with total phenolics (r=‐0.70), and total phenolic concentration was positively correlated with ORAC

(r=0.77) (Table A.6). Chroma was negatively correlated with both hue angle (r=‐0.88) and

89 Table 1.13. Total ellagitannin concentations of fungicide‐ and no fungicide‐treated muscadines. Year Treatment 2012 2013 Fungicide 5.8 bz 9.2 a No Fungicide 4.8 b 7.9 a zMeans followed by the same letter are not significantly different at α= 0.05; separated by Tukey’s HSD.

Fig. 1.19. Total ellagitannin concentration of fungicide‐ and no fungicide‐treated muscadine genotypes. Each standard error bar is constructed using 1 standard error from the mean. 16 Treatment 14 Fungicide 12 No fungicide annins

t 10 8 6 al ellagi (mg/100 g)

t 4

To 2 0 t l t 1 4 5 7 it e i e ra 0 0 1 2 b w m m a M M M M e m re T A A A A es J u p N n S u er S th u o S

Genotype

90 total anthocyanins (r=‐0.86), while hue angle was negatively correlated to total flavonols

(r=‐0.70) (Table A.6).

Discussion

Berry Storage Attributes

Weight Loss and Unmarketable Berries. The average percent unmarketable berries

and percent weight loss were generally similar in both years of the study, but varied by

genotype. Both percent weight loss and percent unmarketable berries increased during

storage and varied among genotypes, and these results were consistent with other studies

(Ballinger and Nesbitt, 1982a; James et al., 1997; James et al., 1999; Lutz, 1938; Silva et al.,

1994; Takeda et al., 1983). For example, James et al. (1997) found that after 2 weeks of

storage ‘Summit’ had 72% unmarketable berries and 4.9% weight loss, while ‘Fry’ had

6.2% unmarketable berries and 4.3% weight loss. Overall, berries from fungicide treatments had less unmarketable berries, but had much less effect on weight loss; these results are consistent with data reported by Lane (1978). Percent unmarketable berries were greater in 2013, while the percent weight loss was greater in 2012. These differences were potentially due to the unusually hot and dry conditions of the 2012 growing season, which resulted in less fungal growth on the berries, but had lower quality berries (leaked and shriveled) (Table A.1).

The postharvest fruit diseases present were identified as black rot (Guignardia

[Phyllosticta] bidwellii [ampelicida] Ellis.), myrothecium leaf spot (Myrothecium sp./spp.), and botrytis fruit rot, (Botrytis sp./spp.). This is similar to reports by Lane (1978), Smit et al. (1971) and Takeda et al. (1983). Generally, fruit diseases were not a major cause of unmarketability until 3 weeks of storage. The primary factors involved in unmarketable

91 fruit were browning (especially in bronze genotypes), leakage from torn or wet stem scars, and shriveling, which is consistent with similar work reported by Perkins‐Veazie et al.

(2012). Occasionally an unknown species of fruit fly (Drosophila sp) was present, but only in fruit with wet or torn stem scars and did not contribute to unmarketable berries.

A major cause of unmarketability in the bronze was browning during storage

(especially no fungicide‐treated AM 01 in 2013), likely caused by chilling injury. The abiotic disorder of chilling injury is common in many horticultural crops, including bananas (Musa

× paradisiaca L.), citrus (Citrus spp.), sweet potatoes (Ipomoea batatas L.), and tomatoes

(Solanum lycopersicum L.) (Wang, 1990). Chilling injury can increase susceptibility to decay through providing a favorable medium for the growth of pathogens (Wang, 1990). The primary symptom of chilling injury identified in this study was brown discoloration of the skin, pulp, and vascular strands of fruit (Himelrick, 2003; Wang, 1990). Table grapes (V. vinifera) were successfully stored at ‐1 °C without showing symptoms of chilling injury

(Burg, 2004). Chilling injury has been reported in muscadines stored at or below 1.7 °C

(Himelrick, 2003; Smittle 1990), but is uncommon in muscadine grapes stored at 2‐3 °C. It was unexpected to find possible symptoms of chilling injury on the muscadine berries studied, as they were stored above the previously reported threshold of 1.7 °C (Himelrick,

2003; Smittle 1990). This illustrates that chilling injury susceptibility might be genotype specific and creates the potential for opportunities of improvement through tolerance selection in fruit breeding programs.

Leakage and shriveling are common problems in muscadines during storage, and are managed by removing berries with wet stem scars prior to storage and maintaining high RH during storage (Perkins‐Veazie et al., 2012; Smit et al., 1971). It has been shown

92 that use of plastic film packaging of lemons (Citrus limon L.) and bell peppers (Capsicum annuum L.) prevents water loss, resulting in less leakage and shriveling (Ben‐Yehoshua et al., 1983). Conversely, muscadines stored in polyethylene bags had less weight loss, but this did not prevent leakage and shriveling; due to the juice retention of the bags (Walker et al.,

2001). Edible coatings have been shown to prevent weight loss during storage of fresh blueberries (Vaccinium corymbosum L.), but this is unknown on muscadines (Duan et al.,

2011). Other table grapes (V. vinifera) had less leakage and shriveling during storage, as the berries were retained on the pedicel during marketing.

It was uncommon that a single genotype performed well in both years of storage. In

2012, the genotypes with the least percent unmarketable were ‘Summit’, ‘Southern Jewel’ and ‘Supreme’, while the genotypes AM 01, AM 04, and ‘Nesbitt’ had the most. Conversely, in 2013, the genotypes AM 04, ‘Nesbitt’, and ‘Supreme’ had the least percent unmarketable, while AM 01, AM 15, ‘Summit’, ‘Southern Jewel’, and ‘Tara’ had the most (Fig. 1.2 and Table

A.2). In 2012, the genotypes with the least amount of weight loss were AM 27, ‘Summit’,

‘Southern Jewel’, and ‘Tara’, while AM 01, AM 04, AM 15, and ‘Nesbitt’ had the most. In

2013, the genotypes AM 27, ‘Nesbitt’, ‘Southern Jewel’, and ‘Summit’ had the least weight loss, while AM 01, AM 04, AM 15, and ‘Supreme’ had the most (Fig. 1.1. and Table A.2). This strongly shows the influence of uncontrollable environmental factors on storage performance of muscadines. Additionally, this emphasizes the importance of testing multiple years for postharvest performance in a muscadine storage protocol. Similar to our findings, Ballinger and Nesbitt (1982b) found ‘Nesbitt’ to have acceptable postharvest storage quality. While James et al. (1999) found ‘Summit’ to have the greatest percent

93 decay and weight loss, I found ‘Summit’ to have intermediate quality during storage, which

was supported by James et al. (1997).

Force. Force to penetrate muscadine skin has been shown to be a useful

characteristic to assess berry crispiness and texture as well as berry maturity (Conner,

2013); however, use of force to determine storability of muscadine grapes has shown

results with no clear trend (Silva et al., 1994; Walker et al., 2001). It has been shown that

muscadines require a force up to 13.9 N to penetrate the skin at harvest, which is nearly

twice that of V. vinifera cultivars (Conner, 2013). Similarly, we found that muscadine grapes

require up 13.2 N to penetrate the skin at harvest (Fig. 1.4 and Table A.2). Similar to the

findings of Conner (2013) it was determined that ‘Nesbitt’ was among the most firm

cultivars. Berries stored in 2013 were generally firmer than the berries stored in 2012,

further showing the significance of environmental influences on storage quality (Figs. 1.3,

1.4, and Table A.2). Force to penetrate berry skin was negatively correlated to percent

unmarketable (r = ‐0.73), which potentially shows that berries requiring greater force to

penetrate the berry skin store better as they were firmer. I found that berry firmness

decreased during storage, but was occasionally lowest after 2 weeks of storage, similar

results were reported by James et al. (1999); this could be due to water loss causing an

increase of firmness at week 3 (Fig. 1.3 and Table A.2). It was found that the genotypes

requiring the most force to penetrate the berry skin at harvest also required the most force

to penetrate the berry skin after 3 weeks of storage (especially in 2013), showing force to be a strong indicator of storage performance (Fig. 1.3 and Table A.2). Though fungicide treated fruit were often more firm, it was determined that genotype and year were much more influential on berry firmness (Fig. 1.3 and Table A.2).

94 Volume Change. Volume change of muscadines during storage is widely unstudied.

Unexpectedly, no relationship between storage time and volume change was found.

Conversely, it has been found that a decrease in size occurred in mangoes (Mangifera indica

L.) and citrus during storage due to water loss (Jha et al., 2006; McCornack, 1975). The

muscadine berries stored in 2012 generally increased in volume during storage while the

berries stored in 2013 generally decreased in volume during storage (Figs. 1.5 and 1.6).

This may have been due to the substantial differences of growing season, as 2012 was extremely hot and dry resulting in lower quality berries, while 2013 had more moderate temperatures and more rainfall (Table A.1). There was also no relationship between fungicide treatments and berry volume (Fig. 1.6 and Table A.2). The lack of correlations found was potentially due to the extreme variation of berry size within each genotype.

Berry Composition

Muscadines have been shown to increase in soluble solids and decrease in TA

during ripening (Carroll and Marcy, 1982; Flora and Lane, 1979; Johnson and Carroll, 1973;

Lanier and Morris, 1979; Peynard and Riberau‐Gayon, 1971; Walker et al., 2001). Unlike

other fruits, muscadines are nonclimacteric, and do not continue to accumulate sugars after

harvest.

Titratable Acidity. It was found that TA stayed relatively constant during storage

(Fig 1.7 and Table A.3); this was consistent with the results of other studies (James et al.,

1997; James et al., 1999; Takeda et al., 1983; Walker et al., 2001) and is contradictory to

Silva et al., (1994), who reported that TA increased during storage and Lutz (1938), who

reported that TA decreased during storage. Titratable acidity has also been shown to

remain constant during storage of muskmelon (Cucumis melo L.), kiwifruit (Actinidia

95 deliciosa L.), mango, pineapple (Ananas comosus L.), strawberry (Fragaria x ananassa

Weston.) and watermelon (Citrullus lanatus Thunb.) (Gil et al., 2006). The TA was lower than that of V. vinfera, which ranged from 0.6 to 1.5 % (Mascarenhas et al., 2012).

There was no clear identifiable effect on TA of fungicide treatment (Fig. 1.7), which is contrary to Smith (2013), who found that TA was significantly lower in fungicide‐treated muscadines. There was a strong correlation between TA and pH (r = 0.99) (Table A.6), possibly showing that as berry acidity increased so did pH. Titratable acidity was much more affected by year and genotype, with 2012 generally having lower TA than 2013 (Fig.

1.7), which can be explained by the warmer growing season (Jackson, 1986). The cultivar

Supreme generally had the highest percent TA and AM 01 and ‘Southern Jewel’ had the lowest (Fig. 1.7 and Table A.3).

Soluble Solids. Soluble solids were strongly impacted by year and genotype (Figs.

1.10 and 1.11, and Table A.3). Probably due to the extremely hot and dry conditions, soluble solids were higher in 2012, potentially due to the heat and dryness concentrating the sugars, than in 2013 (Figs. 1.10 and 1.11). Overall soluble solids remained constant during storage (Fig. 1.10 and Table A.3); this is consistent with most other studies (James et al., 1997; James et al., 1999; Takeda et al., 1983), but was contradictory to Silva et al.

(1994) and Walker et al. (2001), who reported decreased soluble solids during storage, and

Lutz (1938), who reported increased soluble solids during storage. Soluble solids have also been shown to remain constant during storage of muskmelon, kiwifruit, mango, pineapple, strawberry, and watermelon (Gil et al., 2006). Soluble solids was found to be similar to that reported for V. vinifera (Mascarenhas et al., 2012). Threlfall et al. (2007) found that

‘Supreme’ had the lowest soluble solids of all genotypes evaluated, while I identified

96 ‘Supreme’ as being midrange in soluble solids. In 2012, the no fungicide‐treated vines had

slightly higher SSC than did the fungicide‐treated vines (Fig. 1.11), which supports the

findings of Smith (2013). The differences in fungicide treatments for soluble solids were

potentially due to variability that occurred among individual muscadine vines.

pH. I found that pH was strongly affected by genotype and year (Figs. 1.8 and 1.9).

The overall pH measured in 2013 was higher than in 2012 and ranged from 3.8 to 3.4

among the genotypes in 2012 and from 3.9 to 3.4 in 2013 (Fig. 1.8 and Table A.3), this is

contradictory with Jackson (1986), who found that high pH is often associated with

warmer temperatures during the growing season. I found that pH of muscadines remained constant during storage (Fig. 1.8), which was consistent with several other studies (James et al., 1997; James et al., 1999; Silva et al., 1994; Takeda et al., 1983; Walker et al., 2001).

The pH reported for some V. vinifera cultivars was higher than found in muscadines in

2012 and 2013 (Mascarenhas et al., 2012). Additionally, stable pH during storage was

found in kiwifruit, mango, pineapple, strawberry, and watermelon, while pH was shown to

decrease in muskmelon (Gil et al., 2006).

Soluble Solids/Titratable Acidity Ratio. Soluble solids to titratable acidity ratio was

strongly influenced by year and genotype (Tables 1.7 and A.3). The overall average SS/TA

measured in 2013 was higher than in 2012 (Tables 1.7 and 1.9), resulting in unusually high

SS/TA. High temperatures during the growing season of 2012 produced muscadines with

lower TA and higher SSC (Jackson, 1986). There were positive correlations between SS/TA and TA (r = 0.73) and pH (r = 0.73), possibly showing that SS/TA increased with increased acidity (Table A.6). Though there was a main effect with storage time, SS/TA did not change consistently during storage, which is contradictory to Lutz (1938), who reported an

97 increase of SS/TA during storage time. The SS/TA was higher than reported for some V.

vinfiera cultivars (Mascarenhas et al., 2012).

Berry Color

The United States Department of Agriculture (USDA) currently has no standards in place to grade muscadine berries on the color attributes of L*, Chroma, and hue. The

standards for color grades of muscadines state the fruit should be well colored, in the case of black and red varieties that 75% of the surface of the berry shows characteristic color for the variety, while no requirement exist for bronze genotypes except that ‘Carlos’, ‘Fry’ or similar cultivars can show any amount of blush or bronze color on the berry (USDA, 2006).

Additionally the USDA information states that black cultivar colors can include reddish purple, purple, and black; red cultivar colors include light pink, pink, red, dark red, and purple; and bronze cultivar colors include light green, straw, amber, and bronze with allowance for an amount of blush or pink color that may also be characteristic for certain cultivars (USDA, 2006).

The effect of storage time on L*, hue angle, and Chroma values of fresh‐market muscadine grape berries is widely unstudied, although it is well studied in juice and wine.

Variability of the berry color was strongly influenced by genotype in this study.

L* Value. Generally, L* values remained relatively constant during storage; however,

with the genotypes AM 04 in 2012 and AM 15, and ‘Southern Jewel’ in 2013 L* values

decreased after the date of harvest, but remained relatively stable during storage (Fig. 1.12

and Tables 1.9 and A.4). A decrease in L* value would represent a darkening of the fruit

during storage, as L* measures lightness from completely opaque (0) to completely

transparent (100). Silva et al. (1994) found that L* values increased during storage, though

98 the differences were not visibly discernable by panelists. Conversely, I found that generally,

L* values remained stable or slightly decreased during storage (Fig. 1.12 and Table A.4).

Similarly, Hernandez‐Herrero and Frutos (2014) found that the L* values for model juices of grape, plum, and strawberry stayed relatively constant during storage. Though the L* values of fungicide‐ and no fungicide‐treated fruit varied among years, there were no significant differences within each year (Table 1.9). Possibly due to the more favorable growing season (Table A.1) resulting in overall lighter colored berries due to less sunburn

(darkening of the berry skin due to sun exposure), the L* values were higher in 2013, than in 2012 (Figs. 1.12 and 1.13 and Tables 1.9 and A.4). Overall, L* values were strongly affected by genotype and year, while less affected by week of storage. For week 0, the average L* value for the bronze genotypes was 55.1 and 37.3 for the black genotypes. For weeks 1, 2, and 3 the average L values for the black genotypes was 26.6 and was 47.1 for the bronze genotypes, this illustrates the berries darkened after date of harvest, but remained relatively unchanged during storage.

Chroma. I found that Chroma was influenced by storage time, year, and genotype.

Walker et al. (2001) found that Chroma of the bronze cultivar Fry ranged from 12.1 to 14.2 based on maturity level. This is similar to the findings of my study. Conner and MacLean

(2013) found Chroma values ranging from 2.4 to 22.8 and Threlfall et al. (2007) found

Chroma values ranging from 8.0 to 52.8, both of which are consistent with my findings.

Fungicide treatments were found to have no effect on Chroma values (Table A.4). I found

Chroma generally had no clear pattern during storage (Fig. 1.14 and Table A.4), similarly

Hernandez‐Herrero and Frutos (2014) found that the Chroma values of model juices of grape (V. vinifera), plum (Prunus spp.), and strawberry stayed relatively constant during

99 storage. Chroma values were generally highest for bronze genotypes, with the exception of

‘Southern Jewel’ at harvest in 2013 (Fig. 1.12 and Table A.4), which is similar to the findings of Conner and Maclean (2012). Overall, Chroma values were higher in 2012 than in 2013 (Fig 1.14 and Table A.4), which could be explained by the milder growing season in

2013 (Table A.1). I found that the average Chroma for the bronze genotypes was 12.1, while the black genotypes had an average Chroma of 4.5, showing on average the bronze genotypes were less grey than the black genotypes.

Hue Angle. I found hue values ranged from 48.8 to 360°, with the darker genotypes having greater hue angles (Fig. 1.15 and Table A.4). Conversely, Conner and MacLean

(2013) found hue values that ranged from 1.5 to 91.8° and Threlfall et al. (2007) found hue values ranging from 53.4 to 98.6°, while Walker et al. (2001) found hue values that ranged from 76.5 to 237.7°. I found there was no clear relationship between week of storage and hue angle, with values remaining stable during storage. Similarly, Hernandez‐Herrero and

Frutos (2014) found that the hue angles of model juices of grape, plum, and strawberry stayed relatively constant during storage. There was a negative correlation between hue angle and Chroma (r = ‐0.88) (Table A.6), illustrating that as hue angle increased, Chroma values decreased. Overall, hue values were greater in 2013 than in 2012 (Fig. 1.15 and

Table A.4), which might be due to the milder growing season in 2013, resulting in less sunburn (darkening of the berry skin due to sun exposure) (Table A.1). The average hue angle for the black genotypes was 264.9°, which falls between blue and green coloration, while the average hue angle was 91.8° for bronze genotypes, which is approximately yellow.

Nutraceutical Content.

100 Among fruits, muscadines contain some of the highest levels of nutraceuticals;

additionally, several of the compounds present are unique to muscadines (Marshall et al.

2012). Studies exploring the effect of field‐fungicide treatments on muscadine

nutraceutical concentrations are limited.

Total Anthocyanins. Total anthocyanin concentrations found were similar to those

previously reported (Ballinger et al., 1973; Brown, 1940; Conner and MacLean 2013; Lee et

al., 2005; Lee and Talcott, 2002; Goldy et al., 1987; Marshall et al., 2012; Pastrana‐Bonilla et

al., 2003; Sandhu and Gu 2010; Striegler et al., 2005; Stringer et al., 2009; Threlfall et al.,

2007). Anthocyanins were not detected in any of the bronze genotypes in either year of the

study (Fig. 1.16 and Table A.5). Additionally, I found a negative correlation between

anthocyanin concentration and Chroma (r = ‐0.86), possibly showing a relationship that as

Chroma decreases anthocyanin concentrations increase (Table A.6). I found greater total

anthocyanin concentrations than those reported for other grape species (Hernandez‐

Herrero and Frutos, 2014). Total anthocyanins found were lower than those reported for

blackberry (Rubus sp. L.) and highbush blueberry, but greater than those reported for

raspberry (Rubus sp. L.) and strawberry (Cordenunsi et al., 2002; Ehlenfeldt and Prior,

2001; Maatta‐Riihinen et al., 2004; Siriwoharn et al., 2004). Anthocyanin concentrations were generally higher in 2012 than in 2013 with the exceptions of no fungicide‐treated

‘Nesbitt’ and ‘Supreme’, and fungicide‐treated ‘Southern Jewel’. The differences in total anthocyanins among years may be due to higher temperature and greater sun exposure and therefore greater color development and anthocyanin concentration in the 2012 growing season (Table A.1). The cultivar Nesbitt was identified as having among the highest levels of anthocyanins (Lee and Talcott, 2004), while I found ‘Nesbitt’ to have some

101 of the lowest levels among black genotypes. The cultivar Supreme had among the lowest

levels of total anthocyanins, which was also found by Threlfall et al. (2007) and Striegler et

al. (2005). Among years, fungicide treatments did not consistently affect total anthocyanin

concentrations, and within each year fungicide treatments had few significant differences

(Fig. 1.16). Conversely, Nwankno et al. (2011) found that fungicide treatments enhanced the total anthocyanin concentrations of blackcurrants (Ribes nigrum L.).

ORAC. Oxygen radical absorbance capacity is widely accepted as being a good

estimation of antioxidant capacity of fruits, although its significance is often questioned, as

it does not accurately represent the bioactivity of the antioxidants in the human body. The

ORAC values I found were similar to those previously reported by Sandhu and Gu (2010)

and Talcott and Lee (2002), but were considerably higher than those reported by Lee et al.

(2005), Striegler et al. (2005), and Threlfall et al. (2007). I found ORAC values to be greater than or comparable to those found in apple (Malus domestica Borkh.), blackberry, highbush blueberry, plum, orange, red table grapes, strawberry, and white table grapes (Ehlenfeldt and Prior, 2001; Siriwoharn et al., 2004; Wang et al., 1996; Wu et al., 2004). ORAC was found to be higher overall in 2013 than in 2012 (Fig. 1.17), which could possibly be due to the extremely hot and dry growing season in 2012 (Table A.1). Overall, the berries from fungicide‐treated vines had higher ORAC values than the berries from no fungicide‐treated

vines (Fig. 1.7 and Table A.5). This is contradictory to Nwankno et al. (2011), who found

that fungicide treatments had no effect on antioxidant capacity of blackcurrants. It is

hypothesized that the reason for this difference in fungicide treatments is due to the sterol

inhibiting effect of myclobutanil, the active ingredient in Rally®, potentially interfering

with the sterol pathway of the muscadines (Fletcher, 1987). The genotype AM 27 had the

102 highest ORAC values both years of the study, while ‘Supreme’ and ‘Tara’ had the lowest

(Fig. 1.17 and Table A.5). Conversely, Threlfall et al. (2007) reported ‘Nesbitt’ having among the lowest ORAC levels, while Striegler et al. (2005) identified ‘Supreme’ as having

among the highest. Generally, genotypes performed proportionally the same among years of the study (Fig. 1.17 and Table A.5), which shows how significant genotype is on

antioxidant capacity. I found a positive correlation between ORAC and total phenolic

concentrations (r = 0.77), as expected since total phenolic concentrations are a major

component of ORAC, and as total phenolic concentrations increase so does antioxidant

capacity (Table A.6). It is important to note that some genotypes differ more than others

with fungicide treatment; this shows that ORAC has potential as a possible characteristic

goal to be included in breeding programs.

Total Flavonols. Total flavonol concentrations found were lower than those

reported by Marshall et al. (2012) and Talcott and Lee (2002). Overall, total flavonols were

higher in 2012 than in 2013 (Fig. 1.18 and Table A.5). Genotypes performed proportionally

similar among years of the study, with AM 15 and ‘Summit’ having the highest

concentrations both years. The bronze genotypes were found to be generally higher in total

flavonols than the darker genotypes (Fig. 1.18), this is potentially due to the presence of myricetin in the bronze genotypes (Marshall et al., 2012). Additionally, I found a negative correlation between hue angle and total flavonol concentrations (r = ‐0.78) (Table A.6).

This correlation represents decreasing hue angle being related to increasing total flavonol, which is supported by the data showing the bronze genotypes to have higher flavonol concentrations and lower hue angles (Figs. 1.13 and 1.18). Total flavonol concentrations were higher for the fungicide‐treated fruit overall, although this varied among genotypes

103 and years (Fig. 1.18 and Table A.5). I found total flavonols at higher levels than those found

in strawberry, but lower levels than those found in blackcurrant, chokeberry (Aronia mitschurinii A.K.Skvortsov & Maitul), cranberry (Vaccinium oxycoccos L.), and raspberry

(Cordenunsi et al., 2002; Hakkinen et al., 1999; Maatta‐Riihinen et al., 2004).

Total Ellagitannins. Total ellagitannin concentrations found were lower than those

reported for muscadines by Marshall et al. (2012) and Lee and Talcott (2004), but similar

to those reported by Boyle and Hsu (1990), Lee et al. (2005), Pastrana‐Bonilla et al. (2003),

Stringer et al. (2009), and Talcott and Lee (2002). Ellagitannin concentrations were higher

than those reported for strawberry (Cordenunsi et al., 2002), but lower than those

reported in blackberry and raspberry (Maatta‐Riihinen et al., 2004; Siriwoharn et al.,

2004). Additionally, ellagitannins are absent in all other Vitis species (Marshall et al., 2012).

I found total ellagitannins were higher in 2013 than in 2012, and overall there was no

significant difference in fungicide treatments among years (Table 1.13). Similar to our

findings, Marshall et al. (2012) found ‘Tara’ to have the lowest total ellagitannin

concentrations of the genotypes they studied. Ellagitannin concentrations varied greatly

among genotypes and treatments with no consistent effect of fungicide treatments (Fig.

1.19 and Table A.5). These findings are contrary to those reported by Smith (2013), who

found the total ellagitannins were greater in no fungicide‐treated berries.

Total Phenolics. I found total phenolic concentrations similar to those previously

reported for muscadines (Lee et al., 2005; Lee and Talcott, 2004; Marshall et al., 2012;

Pastrana‐Bonilla et al., 2003; Striegler et al., 2005; Stringer et al., 2009; Talcott and Lee,

2002; Threlfall et al., 2007). I found that total phenolics were negatively correlated with pH

(r = ‐0.70), which supports the findings of Orak (2007) who determined that total phenolic

104 concentration was negatively correlated to pH in V. vinifera grapes. Similar phenolic concentrations have been reported in blackberry (Siriwohorn et al., 2004), while lower phenolic concentrations have been found in highbush blueberry and strawberry

(Corenunsi et al., 2002; Ehlenfeldt and Prior; 2001). Total phenolics were higher in 2012 than in 2013, likely due to the less favorable growing conditions (Table 1.11). In 2012, total phenolics were higher for the fungicide treatments, while in 2013 no significant differences were found among fungicide treatments (Tables 1.11 and A.5). I found ‘Summit’ to have among the highest levels of total phenolics, which was similar to the findings of Threlfall et al. (2007). Additionally, I found the cultivar Supreme to have the overall lowest total phenolics, while Striegler et al. (2005), found ‘Supreme’ to have among the highest total phenolic levels.

Resveratrol. I found resveratrol concentrations similar to those previously reported

in muscadines (Ector et al., 1996; Magee et al., 2002; Marshall et al., 2012; Pastrana‐Bonilla

et al., 2003; Stringer et al., 2009). As I did, Marshall et al. (2012) found resveratrol in nearly

every genotype evaluated. I found no clear relationship between berry color and

resveratrol concentrations, conversely Ector et al. (1996) found resveratrol to be greater in

black genotypes. Magee et al. (2002) found ‘Summit’ to have among the highest levels of

resveratrol, which is similar to my findings. Resveratrol concentrations were found to be

equivalent to those in V. vinifera (Vincenzi et al., 2013). In 2012 no significant differences

occurred among fungicide treatments, while in 2013 the no fungicide‐treated fruit had

higher levels of resveratrol than the fungicide‐treated (Tables 1.12 and A.5). These results

are similar to those reported by Smith (2013), who found resveratrol concentrations to be

ten times greater in no fungicide‐treated berries. The differences in fungicide treatment

105 among years could be due to the hot and dry conditions during the growing season of 2012

(Table A.1), as resveratrol is produced in response to fungal infection, which occurs more readily in cooler, wetter conditions (Jeandet et al., 1995). Similar findings of increased resveratrol concentrations, of muscadines, resulting from no fungicide applications have been reported (Magee et al., 2002). Conversely, Jeandet et al. (1995) found that while resveratrol concentration in V. vinifera wine increased as B. cinerea infection increased, excessive grey mold development may destroy the induced phytoalexin.

Storage Protocol

A major component of this study was to determine the important parameters of storage performance of muscadine genotypes, and in so doing develop a storage protocol for the University of Arkansas muscadine breeding program. Overall, both percent unmarketable and percent weight loss increased during storage, showing importance as storage parameters. Force to penetrate the berry skin generally decreased during storage, also showing potential as an important postharvest storage parameter. Percent change in berry volume showed no clear pattern during storage, probably due to the variation in individual berries within each genotype limiting the usefulness of berry volume as a storage measurement. Composition parameters TA, pH, soluble solids, and SS/TA remained relatively constant during storage, therefore are not valuable postharvest storage measurements. Though no clear correlations were identified in this study, it has been shown that soluble solids can be useful in determining maturity, which has been shown to be related with storage performance (Ballinger and McClure, 1983; Carroll and Marcy,

1982). Berry color measurements Chroma and hue angle generally showed no clear pattern during storage, while L* showed a sharp decrease after date of harvest, it remained

106 relatively constant during storage. Therefore, it is potentially valuable to determine L* value at date of harvest and again after storage is complete to evaluate color change during storage. The retention of nutraceutical content (total phenolics, total anthocyanins, total ellagitannins, total flavonols, and resveratrol) and antioxidant capacity (ORAC) was not evaluated in this study, but in a related study (Study 3), it was determined that all nutraceuticals remained relatively constant during storage with some significant differences among weeks, but no clear linear pattern of change.

Literature Cited

Ballinger, W.E., E.P. Maness, and W.B. Nesbitt. 1973. Anthocyanins of black grapes of 10 clones of Vitis rotundifolia, Michx. J. Food Sci. 38:909‐910.

Ballinger, W.E. and W. F. McClure. 1983. The effect of ripeness on storage quality of ‘Carlos’ muscadine grapes. Scientia Hort. 18:241‐245.

Ballinger, W.E. and W.B. Nesbitt. 1982a. Postharvest decay of muscadine grapes (Carlos) in relation to storage temperature, time, and stem condition. Amer. J. Enol. Viticult. 33:173‐ 175.

Ballinger, W.E. and W.B. Nesbitt. 1982b. Quality of muscadine grapes after storage with sulfur dioxide generators. J. Amer. Soc. Hort. Sci. 107:827‐830.

Banini, A.E., L.C. Boyd, J.C. Allen, H.G. Allen, and D.L. Sauls. 2006. Muscadine grape products intake, diet and blood constituents of non‐diabetic and type 2 diabetic subjects. Nutrition. 22:1137‐1145.

Ben‐Yehoshua, S., B. Shapiro, Z.E. Chen, and S. Lurie. 1983. Mode of action of plastic film in extending life of lemon and bell pepper fruit by alleviation of water stress. Plant Physiol. 73:87‐93.

Boyle, J.A., and L. Hsu. 1990. Identification and quantitation of ellagic acid in muscadine grape juice. Amer. J. Enol. Viticult. 41:43‐47.

Bralley, E.E., D.K. Hartle, P. Greenspan, and J.L. Hargrove. 2007. Topical anti‐inflammatory activities of Vitis rotundifolia (muscadine grape) extracts in the tetradecanoylphorbol acetate model of ear inflammation. J. Medicinal Food. 10:636‐642.

Brown, W.L. 1940. The anthocyanin pigment of the Hunt muscadine grape. J. Amer. Chem. Soc. 62:2808‐2810.

107

Burg, S.P. 2004. Postharvest physiology and hypobaric storage of fresh produce. 1st ed. CABI Publishing, Wallingford, Oxfordshire, U.K.

Carroll, D.E. and J.E. Marcy. 1982. Chemical and physical changes during maturation of muscadine grapes (Vitis rotundifolia). Amer. J. Enol. Viticul. 33:168‐172.

Cho, M.J., L.R. Howard, R.L Prior, and J.R. Clark. 2004. Flavanoid glycosides and antioxidant capacity of various blackberry, blueberry, and red grape genotypes determined by high‐ performance liquid chromatography/mass spectrometry. J. Sci. Food Agri. 84:1771‐1782.

Cho, M.J., L.R. Howard, R.L. Prior, and J.R. Clark. 2005. Flavonol glycosides and antioxidant capacity of various blackberry and blueberry genotypes determined by high‐performance liquid chromatography/mass spectrometry. J. Sci. Food. Agr. 85:2149‐2158.

Conner, P.J. 2009. A century of muscadine Grape (Vitis rotundifolia Michx.) breeding at the University of Georgia. Acta Hort. 827:481‐484.

Conner, P.J. 2013. Instrumental texture analysis of muscadine grape germplasm. HortScience. 48:1130‐1134.

Conner, P.J., and D. Maclean. 2012. Evaluation of muscadine grape genotypes for storage ability. HortScience. 47:S386. (abstr.).

Conner, P.J., and D. MacLean. 2013. Fruit anthocyanin profile and berry color of muscadine grape cultivars and Muscadinia germplasm. HortScience 48:1235‐1240.

Cordenunsi, B.R., J.R. Oliveira do Nascimento, M.I. Genovese, and F.M. Lajolo. 2002. Influence of cultivar on quality parameters and chemical composition of strawberry fruits grown in Brazil. J. Agr. Food. Chem. 50:2581‐2586.

Duan, J., R. Wu, B.C. Strik, and Y. Zhao. 2011. Effect of edible coatings on the quality of fresh blueberries (Duke and Elliot) under commercial storage conditions. Postharvest Biol. Technol. 59:71‐79.

Ector, B.J. 2001. Compositional and nutritional characteristics, p. 341‐367. In: F.M. Basiouny and D.G. Himelrick, (eds.). Muscadine Grapes. ASHS Press, Alexandria, VA.

Ector, B.J., J.B. Magee, C.P. Hegwood, and M.J. Coign. 1996. Resveratrol concentration in muscadine berries, juice, pomace, purees, seeds, and wines. Amer. J. Enol. Viticult. 1:57‐62.

Ehlenfeldt, M.K., and R.L. Prior. 2001. Oxygen radical absorbance capacity (ORAC) and phenolic and anthocyanin concentrations in fruit and leaf tissues of highbush blueberry. J. Agr. Food Chem. 49:2222‐2227.

108 Flora, L.F. and R.P. Lane. 1979. Effects of ripeness and harvest date on several physical and composition factors of ‘Cowart’ muscadine grapes. Amer. J. Enol. Viticult. 30:241‐246.

Fletcher, R.A. 1987. Plant growth regulating properties of sterol‐inhibiting fungicides, p. 103‐113. In: S.S. Purohit (ed.). Hormonal regulation if plant growth and development. Springer Sci. and Business Media, Berlin, Germany.

Gil, M.I., E. Aguayo, and A.A. Kader. 2006. Quality changes and nutrient retention in fresh‐ cut versus whole fruits during storage. J. Agr. Food. Chem. 54:4284‐4296.

Giusti, M.M. and R.E. Wrolstad. 2001. Characterization and measurement of anthocyanin by UV‐visible spectroscopy. Current Protocols Food Anal. Chem. F1.2.1‐F1.2.13.

God, J.M., P. Tate, and L.L. Larcom. 2007. Anticancer effects of four varieties of muscadine grape. J. Medicinal Food. 10:54‐59.

Goldy, R.G., W. E. Ballinger, E. P. Maness, and W. H. Swallow. 1987. Anthocyanin content of fruit, stem, tendril, leaf, and leaf petiole in muscadine grape. J. Amer. Soc. Hort. Sci. 112:880‐882.

Greenspan, P., A. Ghaffar, J.L. Hargrove, D.K. Hartle, E.P. Mayer, J.D. Bauer, S.H. Pollock, and J.D. Gangemi. 2005. Antiinflammatory properties of the muscadine grape (Vitis rotundifolia). J. Agr. Food Chem. 53:8481‐8484.

Hakkinen, S.H., S.O. Karenlampi, L.M. Heinonen, H.M. Mykkanen, and A.R. Torronen. 1999. Content of edible flavonols quercetin, myricetin, and kaempferol in 25 edible berries. J. Agr. Food. Chem. 47:2274‐2279.

Hager, T.J., L.R. Howard, R. Liyanage, J.O. Lay, and R.L. Prior. 2008. Ellagitannin composition of blackberry as determined by HPLC‐ESI‐MS and MALD‐TOF‐MS. J. Agr. Food. Chem. 56: 661‐669.

Hernandez‐Herrero, J.A., and M.J. Frutos. 2014. Colour and antioxidant capacity stability in grape, strawberry, and plum peel model juices at different pHs and temperatures. Food. Chem. http://dx.doi.org/10.1016/j.foodchem.2014.01.007

Himelrick, D.G. 2003. Handling, storage, and postharvest physiology of muscadine grapes. Small Fruits Rev. 2:4:45‐62.

Huang, Z., R.D. Pace, P. Williams, and B. Wang. 2009. Identification of anthocyanins in muscadine grapes with HPLC‐ESI‐MS. Food Sci. Tech. 42:819‐824.

Jackson D.I. 1986. Factors affecting soluble solids, acid, pH, and color in grapes. Amer. J. Enol. Viticult. 37:179‐183.

109 James, J., O. Lamikanra, G. Dixon, S. Leong, J.R. Morris, G. Main, and J. Silva. 1997. Shelf‐life study of muscadine grapes for the fresh fruit market. Proc. Fla. State Hort. Soc. 110:234‐ 237.

James, J., O. Lamikanra, J.R. Morris, G. Main, T. Walker, and J. Silva. 1999. Interstate shipment and storage of fresh muscadine grapes. J. Food Quality. 22:605‐617.

Jeandet, P., R. Bessis, M. Sbaghi, P. Meunier, and P. Trollat, 1995: Resveratrol content of wines of different ages: relationship with fungal disease pressure in the vineyard. Amer. J. Enol. Viticult. 46:1‐4.

Jha, S.N., A.R.P Kingsly, and S. Chopra. 2006. Physical and mechanical properties of mango during growth and storage for determination of maturity. J. Food. Eng. 72:73‐76.

Johnson, L.A. and D.E. Carroll. 1973. Organic acid and sugar contents of Scuppernong grapes during ripening. J. Food. Sci. 38:21‐24.

Lane, R.P. 1978. Effect of vineyard fungicide treatments on the shelf life of muscadine grapes. Ga. Agr. Res. 19:12‐14.

Lane, R.P., and L.F. Flora. 1980. Some factors influencing storage of muscadine grapes. HortScience 15:273 (abstr.).

Lanier, M.R. and J.R. Morris. 1979. Evaluation of density separation for defining fruit maturities and maturation rates of once‐over harvested muscadine grapes. J. Amer. Soc. Hort. Sci. 104:249‐252.

Lee, J‐H. and S.T. Talcott. 2004. Fruit maturity and juice extraction influences ellagic acid derivatives and other antioxidant polyphenolics in muscadine grapes. J. Agr. Food Chem. 52:361‐366.

Lee, J‐H., J.V. Johnson, and S.T. Talcott. 2005. Identification of ellagic acid conjugates and other polyphenolics in muscadine grapes by HPLC‐ESI‐MS. J. Agr. Food Chem. 53:6003‐ 6010.

Lutz, J.M. 1938. Factors influencing the quality of American grapes in storage. U. S. Dept. Agr. Tech. Bul. 606:1‐27.

Maatta‐Riihinen, K. R., A. Kamal‐Eldin, and A.R. Torronen. 2004. Identification and quantification of phenolic compounds in berries of Fragaria and Rubus species (Family Rosaceae). J. Agr. Food Chem. 52:6178–6187.

MacLean, D., P.J. Conner, J. Paulk, and L. Grant. 2009. Postharvest control of decay organisms. The Southern Region Small Fruit Consortium. Prog. Rpt. 2009‐12.

110 Magee, J.B., B.J. Smith, and A. Rimando. 2002. Resveratrol content of muscadine berries is affected by disease control spray program. HortScience 37:358‐361.

Marshall, D.A., S.J. Stringer, and J.D. Spiers. 2012. Stilbene, ellagic acid, flavanol, and phenolic content of muscadine grape (Vitis rotundifolia Michx.) cultivars. Pharmaceutical Crops 3:69‐77.

Mascarenhas, R.d.J., S.d.M. Silva, M.A.C.d. Lima, R.M.N. Mendonca, and H.J. Holschuh. 2012. Characterization of maturity and quality of Brazilian apirenic grapes in the Sao Francisco river Valley. Ciência e Tecnologia de Alimentos. 32:26‐33.

McCornack, A.A. 1975. Postharvest weight loss of Florida citrus fruits. Proc. Fla. State Hort Soc. 88:333‐335.

Morris, J.R. 1980. Handling and marketing of muscadine grapes. FruitSouth, 4(2):12‐14.

Morris, J.R., O.L. Oswald, G.L. Main, J.N. Moore, and J.R. Clark. 1992. Storage of new seedless grape cultivar with sulfur dioxide generators. Amer. J. Enol. Viticult. 43:230‐232.

Musingo, M.N., S.F. O'Keefe, O. Lamikanra, C.A. Sims, and R.P. Bates. 2001. Changes in ellagic acid and other phenols in muscadine grape (Vitis rotundifolia) juices and wines during storage. Amer. J. Enol. Viticult. 52:109‐114.

Nwankno, A.J., S.L. Gordon, S.R. Verrall, R.M. Brennan, and R.D. Hancock. 2011. Treatment of fungicides influences phytochemical quality of blackcurrant juice. Ann. Appl. Biol. 160:86‐ 96.

Pastrana‐Bonilla, E., C.C. Akoh, S. Sellappan, and G. Krewer. 2003. Phenolic content and antioxidant capacity of muscadine grapes. J. Agr. Food Chem. 51:5497‐5503.

Perkins‐Veazie, P., S. Spayd, B. Cline, and C. Fisk. 2012. Handling and marketing guide for fresh market muscadine grapes. SFRC‐E03: 1‐12.

Peynard, E. and P. Ribercau‐Gayun. 1971. The grape, p. 172‐206. In: A.D. Hulme (ed.). The bio‐chemistry of fruits and Their Products. Vol. 2. Academic Press, Waltham, MA.

Prior, R.L., H. Hoang, L. Gu, X. Wu, M. Bacchiocca, L. Howard, M. Hampschwoodill, D. Haung, B. Ou, and R. Jacob. 2003. Assays for hydrophilic and lipophilic antioxidant capacity [oxygen radical absorbance capacity (ORACF1)] of plasma and other biological and food samples. J. Agr. Food Chem. 51:3272‐3279.

Prior RL, S.A. Lazarus, G. Cao, H. Muccitelli, and J.F. Hammer‐ stone. 2001. Identification of procyanidins and anthocyanins in blueberries and cranberries (Vaccinium Spp) using high‐ performance liquid chromatography/mass spectrometry. J Agr. Food Chem 49:1270–1276.

111 Sandhu, A.K., and L.W. Gu. 2010. Antioxidant capacity, phenolic content, and profiling of phenolic compounds in the seeds, skin, and pulp of Vitis rotundifolia (muscadine grapes) as determined by HPLC‐DAD‐ESI‐MSn. J. Agr. Food Chem. 58:4681‐4692.

Savoy, C.F. and T.T. Hatton. 1980. Post‐harvest characteristics of some commercially picked muscadine varieties. Proc. Viticult. Sci. Symp. Fla. A and M. Univ. 24‐25.

Silva, J.L., E. Marroquin, C.P. Hegwood, G.R. Silva, and J.O. Garner Jr. 1994. Quality changes in muscadines for table grapes during refrigerated storage in various packaging systems. Proc. Viticult. Sci. Symp. Fla. A and M. Univ. 17:65‐72.

Siriwoharn, T., R.E. Wrolstad, C.E. Finn, and C.B. Pereira. 2004. Influence of cultivar, maturity, and sampling on blackberry (Rubus L. hybrids) anthocyanins, polyphenolics, and antioxidant properties. J. Agr. Food Chem. 52:8021–8030.

Slinkard, K. and V.L. Singleton. 1977. Total phenol analysis: automation and comparison with manual methods. Amer. J. Enol. Viticult. 28:49‐55.

Smith, B.J. 2013. Fruit quality, phytochemical content, and disease severity of muscadine grapes affected by fungicide applications. Pharmaceutical Crops 4:21‐37.

Smit, C.J.B., H.L. Cancel, and T.O.M. Nakayama. 1971. Refrigerated storage of muscadine grapes. Amer. J. Enol. Viticult. 22:227‐230.

Smith, B.J., and J.B. Magee. 2002. Limited fungicide applications affect berry rot severity and resveratrol content of muscadine grapes. Phytopathology. 92:577. (abstr.).

Smittle, D.A. 1990. Requirements for commercial CA storage of muscadine grapes. Proc. Viticult. Sci. Symp. Fla. A and M. Univ. 13:140‐146.

Striegler, R.K., P.M. Carter, J.R. Morris, J.R. Clark, R.T. Threlfall, and L.R. Howard. 2005. Yield, quality, and nutraceutical potential of selected muscadine cultivars grown in southwestern Arkansas. HortTechnology 15:276‐284.

Stringer, S.J., D.A. Marshall, and P. Perkins‐Veazie. 2009. Nutraceutical compound concentrations of muscadine (Vitis rotundifolia Michx.) grape cultivars and breeding lines. Acta. Hort. 841:553‐556.

Takeda, F., M. Starnes Saunders, C.F. Savoy, and T.T. Hatton. 1983. Storageability of muscadines for use as fresh fruit. Proc. Viticult. Sci. Symp. Fla. A and M. Univ. 3:31‐33.

Takeda, F., M. Starnes Saunders, J.A. Saunders, and T.T. Hatton. 1982. Effects of prestorage treatment and storage temperature on incidence of decay and chemical composition in muscadine grape. Proc. Fla. State. Hort. Soc. 95:109‐112.

112 Talcott, S.T., and J‐H. Lee. 2002. Ellagic acid and flavonoid antioxidant content of muscadine wine and juice. J. Agr. Food Chem. 50:3186‐3192.

Threlfall, R.T., J.R. Morris, J.F. Meullenet, and R.K. Striegler. 2007. Sensory characteristics, composition, and nutraceutical content of juice from Vitis rotundifolia (muscadine) cultivars. Amer. J. Enol. Viticult. 58:268‐273.

Threlfall, R.T., J.R. Morris, L.R. Howard, C.R. Brownmiller, and T.L. Walker. 2005. Pressing effects on yield, quality, and nutraceutical content of juice, seeds, and skins from ‘Black Beauty’ and ‘Sunbelt’ grapes. J. Food Sci. 70:167‐171.

U.S. Department of Agriculture. 2006. United States standard for grades of muscadine (Vitis rotundifolia) grapes. U.S. Dept. Agr. Washington, D.C.

Vincenzi, S., D. Tomasi, F. Gaiotti, L. Lovat, S. Giacosa, F. Torchio, S. Rio Segade, and L. Rolle. 2013. Comparative study of the resveratrol content of twenty‐one Italian red grape varieties. South African J. Enol. Viticult. 34:30‐35.

Walker, T.L., J.R. Morris, R.T. Threlfall, G.L. Main, O. Lamikanra, and S. Leong. 2001. Density separation, storage, shelf life, and sensory evaluation of 'Fry' muscadine grapes. HortScience. 36:941‐945.

Wang, C.Y. 1990. Chilling injury of horticultural crops. 1st ed. CRC Press, Boca Raton, F.L.

Wang, H., G. Cao, and R.L. Prior. 1996. Total antioxidant capacity of fruits. J. Agr. Food. Chem. 44:701‐705.

Wu, X., G.R. Beecher, J.M. Holden, D.B. Haytowitz, S.E. Gebhardt, and R.L. Prior. 2004. Lipophilic and hydrophilic antioxidant capacities of common in the United States. J. Agr. Food Chem. 52:4026‐4037.

Yi, W., C.C. Akoh, and J. Fischer. 2005. Study of anticancer activities of muscadine grape phenolics in vitro. J. Agr. Food Chem. 53:8804‐8812.

113 Appendix A: Meteorological data, interaction means, and correlations.

Table A.1. Average monthly maximum and minimum temperatures and total rainfall recorded at the Fruit Research Station; Clarksville, AR (lat. 35°31’58°N and long. 93°24’12’W) (2012 and 2013). Year Month Maximum Minimum Precipitation temperature (°C) temperature (°C) (mm) 2012 January 11.4 0.1 111.84 February 11.8 2.5 65.50 March 21.4 9.8 198.73 April 23.08 11.5 81.86 May 26.6 16.3 18.88 June 32.9 19.4 14.36 July 36.5 22.7 40.56 August 33.9 20.9 62.47 September 28.4 17.4 158.19 October 20.2 9.2 127.21 November 16.2 4.4 23.56 December 11.8 1.9 3.23 2013 January 9.1 ‐0.7 98.85 February 9.9 ‐0.2 70.63 March 2.8 2.0 130.30 April 19.5 8.7 119.37 May 23.9 13.6 163.07 June 29.8 19.0 54.61 July 31.4 19.9 100.35 August 30.4 20.9 178.82 September 30.2 17.9 57.91 October 20.9 10.4 106.18 November 12.6 2.3 103.14 December 6.8 ‐1.6 64.51

114 Table A.2. Interaction means of the postharvest attributes of percent weight loss, percent unmarketable, force to penetrate berry skin, and percent volume change for year, genotype, fungicide treatment, and week of storage at 2 °C for 0‐3 weeks. Year Genotype Fungicide Week of Weight Unmarketable Force Volume treatment storage loss (%) (N) change (%) (%) 2012 AM 01 Fungicide 0 0.0 0.0 9.4 0.0 2012 AM 01 Fungicide 1 2.2 16.4 2.6 3.7 2012 AM 01 Fungicide 2 3.9 17.2 4.8 ‐1.6 2012 AM 01 Fungicide 3 5.9 49.5 5.1 ‐7.2 2012 AM 01 zNo fung 0 0.0 0.0 8.9 0.0 2012 AM 01 No fung 1 1.7 18.9 2.7 ‐15.3 2012 AM 01 No fung 2 3.5 25.6 6.5 ‐21.8 2012 AM 01 No fung 3 5.1 54.6 4.9 ‐30.8 2012 AM 04 Fungicide 0 0.0 0.0 10.3 0.0 2012 AM 04 Fungicide 1 2.3 5.0 9.0 2.5 2012 AM 04 Fungicide 2 4.0 20.0 7.6 4.9 2012 AM 04 Fungicide 3 6.3 53.3 6.6 ‐7.8 2012 AM 04 No fung 0 0.0 0.0 9.2 0.0 2012 AM 04 No fung 1 1.7 9.9 8.5 ‐12.8 2012 AM 04 No fung 2 3.9 29.2 6.1 ‐8.6 2012 AM 04 No fung 3 5.9 64.8 6.0 ‐14.5 2012 AM 15 Fungicide 0 0.0 0.0 7.8 0.0 2012 AM 15 Fungicide 1 1.8 7.7 7.1 ‐27.6 2012 AM 15 Fungicide 2 4.0 21.8 6.8 ‐16.8 2012 AM 15 Fungicide 3 7.1 30.6 2.4 ‐12.2 2012 AM 15 No fung 0 0.0 0.0 8.0 0.0 2012 AM 15 No fung 1 1.4 15.8 5.2 ‐12.7 2012 AM 15 No fung 2 2.9 24.6 4.9 ‐8.9 2012 AM 15 No fung 3 5.1 41.9 2.2 ‐16.0 2012 AM 27 Fungicide 0 0.0 0.0 9.3 0.0 2012 AM 27 Fungicide 1 1.2 7.0 7.2 15.3 2012 AM 27 Fungicide 2 2.4 14.2 2.1 34.8 2012 AM 27 Fungicide 3 3.6 31.0 4.8 5.7 2012 AM 27 No fung 0 0.0 0.0 8.7 0.0 2012 AM 27 No fung 1 1.4 12.1 7.2 14.8 2012 AM 27 No fung 2 2.8 22.8 2.4 8.6 2012 AM 27 No fung 3 4.3 42.7 4.8 18.2 2012 Nesbitt Fungicide 0 0.0 0.0 9.9 0.0 2012 Nesbitt Fungicide 1 2.1 7.5 7.2 4.8 2012 Nesbitt Fungicide 2 3.3 29.0 6.4 ‐6.2 2012 Nesbitt Fungicide 3 5.8 81.7 5.8 ‐9.0 2012 Nesbitt No fung 0 0.0 0.0 9.8 0.0

115 Year Genotype Fungicide Week of Weight Unmarketable Force Volume treatment storage loss (%) (N) change (%) (%) 2012 Nesbitt No fung 1 2.9 7.1 8.7 7.1 2012 Nesbitt No fung 2 4.4 18.8 7.2 13.7 2012 Nesbitt No fung 3 6.5 50.8 5.6 5.3 2012 S. Jewel Fungicide 0 0.0 0.0 9.5 0.0 2012 S. Jewel Fungicide 1 1.8 5.6 2.7 ‐2.6 2012 S. Jewel Fungicide 2 3.5 9.4 6.5 ‐3.1 2012 S. Jewel Fungicide 3 4.9 19.3 6.7 ‐7.0 2012 S. Jewel No fung 0 0.0 0.0 7.3 0.0 2012 S. Jewel No fung 1 1.5 6.9 2.4 6.6 2012 S. Jewel No fung 2 3.3 11.6 6.1 5.8 2012 S. Jewel No fung 3 5.7 23.2 5.6 ‐6.5 2012 Summit Fungicide 0 0.0 0.0 9.0 0.0 2012 Summit Fungicide 1 1.1 4.2 2.8 9.4 2012 Summit Fungicide 2 3.8 13.8 7.0 10.9 2012 Summit Fungicide 3 3.5 14.8 8.8 ‐3.3 2012 Summit No fung 0 0.0 0.0 8.8 0.0 2012 Summit No fung 1 1.6 5.6 3.0 4.6 2012 Summit No fung 2 3.6 19.0 7.5 7.4 2012 Summit No fung 3 5.5 20.7 8.0 ‐15.0 2012 Supreme Fungicide 0 0.0 0.0 10.6 0.0 2012 Supreme Fungicide 1 1.5 3.2 8.9 15.6 2012 Supreme Fungicide 2 3.6 13.5 8.1 ‐1.7 2012 Supreme Fungicide 3 5.2 26.7 6.5 8.4 2012 Supreme No fung 0 0.0 0.0 9.1 0.0 2012 Supreme No fung 1 2.0 2.2 9.0 23.4 2012 Supreme No fung 2 3.5 18.4 7.3 12.6 2012 Supreme No fung 3 5.3 24.0 7.7 20.4 2012 Tara Fungicide 0 0.0 0.0 8.4 0.0 2012 Tara Fungicide 1 1.8 12.1 5.2 9.0 2012 Tara Fungicide 2 3.3 30.2 2.1 4.9 2012 Tara Fungicide 3 5.1 34.4 4.2 ‐4.4 2012 Tara No fung 0 0.0 0.0 7.7 0.0 2012 Tara No fung 1 1.6 16.8 5.4 17.6 2012 Tara No fung 2 3.1 29.6 1.8 6.4 2012 Tara No fung 3 4.7 37.4 4.2 3.3 2013 AM 01 Fungicide 0 0.0 0.0 8.4 0.0 2013 AM 01 Fungicide 1 1.5 6.6 6.8 12.7 2013 AM 01 Fungicide 2 2.7 21.5 7.6 6.4 2013 AM 01 Fungicide 3 3.9 57.2 5.8 10.9 2013 AM 01 No fung 0 0.0 0.0 8.7 0.0

116 Year Genotype Fungicide Week of Weight Unmarketable Force Volume treatment storage loss (%) (N) change (%) (%) 2013 AM 01 No fung 1 1.5 11.9 8.1 ‐10.7 2013 AM 01 No fung 2 2.6 26.0 8.0 ‐7.7 2013 AM 01 No fung 3 4.1 94.9 5.0 ‐3.5 2013 AM 04 No fung 0 0.0 0.0 12.9 0.0 2013 AM 04 No fung 1 1.0 2.9 12.1 2.8 2013 AM 04 No fung 2 1.7 9.4 11.0 ‐1.0 2013 AM 04 Fungicide 3 2.9 14.5 11.6 ‐5.0 2013 AM 04 No fung 0 0.0 0.0 12.1 0.0 2013 AM 04 No fung 1 1.1 3.2 12.6 0.7 2013 AM 04 No fung 2 1.8 8.2 11.1 3.4 2013 AM 04 No fung 3 2.8 12.6 11.2 ‐3.5 2013 AM 15 Fungicide 0 0.0 0.0 9.5 0.0 2013 AM 15 Fungicide 1 1.1 14.5 6.9 13.3 2013 AM 15 Fungicide 2 2.5 21.0 5.7 10.7 2013 AM 15 Fungicide 3 2.6 50.5 5.7 6.9 2013 AM 15 No fung 0 0.0 0.0 9.1 0.0 2013 AM 15 No fung 1 0.9 18.0 7.8 ‐5.2 2013 AM 15 No fung 2 1.9 23.5 6.2 1.0 2013 AM 15 No fung 3 2.8 54.0 5.7 10.9 2013 AM 27 Fungicide 0 0.0 0.0 8.8 0.0 2013 AM 27 Fungicide 1 1.0 8.7 7.1 ‐5.7 2013 AM 27 Fungicide 2 1.5 24.8 6.9 ‐4.6 2013 AM 27 Fungicide 3 2.2 50.9 6.1 3.5 2013 AM 27 No fung 0 0.0 0.0 9.7 0.0 2013 AM 27 No fung 1 0.9 9.8 7.1 ‐5.0 2013 AM 27 No fung 2 1.7 34.6 6.3 2.3 2013 AM 27 No fung 3 2.3 51.5 5.1 8.2 2013 Nesbitt Fungicide 0 0.0 0.0 11.6 0.0 2013 Nesbitt Fungicide 1 0.9 4.1 8.8 23.4 2013 Nesbitt Fungicide 2 1.6 7.5 10.0 10.1 2013 Nesbitt Fungicide 3 2.2 21.8 7.3 19.7 2013 Nesbitt No fung 0 0.0 0.0 13.2 0.0 2013 Nesbitt No fung 1 0.8 6.1 10.0 20.6 2013 Nesbitt No fung 2 1.5 9.6 11.1 16.4 2013 Nesbitt No fung 3 2.2 22.8 8.9 22.5 2013 S. Jewel Fungicide 0 0.0 0.0 10.1 0.0 2013 S. Jewel Fungicide 1 0.7 12.0 8.9 ‐5.4 2013 S. Jewel Fungicide 2 1.4 12.7 8.2 4.6 2013 S. Jewel Fungicide 3 2.2 50.0 8.6 ‐2.4 2013 S. Jewel No fung 0 0.0 0.0 10.5 0.0

117 Year Genotype Fungicide Week of Weight Unmarketable Force Volume treatment storage loss (%) (N) change (%) (%) 2013 S. Jewel No fung 1 0.6 14.3 9.4 2.1 2013 S. Jewel No fung 2 1.4 16.4 8.8 9.7 2013 S. Jewel No fung 3 2.0 53.6 7.8 9.1 2013 Summit Fungicide 0 0.0 0.0 11.1 0.0 2013 Summit Fungicide 1 0.9 4.6 9.4 ‐8.4 2013 Summit Fungicide 2 1.8 11.2 10.6 ‐4.7 2013 Summit Fungicide 3 2.3 50.3 7.3 ‐5.1 2013 Summit No fung 0 0.0 0.0 9.7 0.0 2013 Summit No fung 1 1.0 8.1 7.8 ‐1.7 2013 Summit No fung 2 1.6 14.4 8.8 4.7 2013 Summit No fung 3 2.2 51.2 7.2 6.5 2013 Supreme Fungicide 0 0.0 0.0 10.4 0.0 2013 Supreme Fungicide 1 1.4 4.1 8.7 ‐4.0 2013 Supreme Fungicide 2 2.5 8.1 8.3 10.3 2013 Supreme Fungicide 3 3.6 24.1 8.2 21.6 2013 Supreme No fung 0 0.0 0.0 8.5 0.0 2013 Supreme No fung 1 1.3 3.8 7.9 ‐6.6 2013 Supreme No fung 2 2.3 8.9 6.9 7.3 2013 Supreme No fung 3 3.2 29.3 7.4 14.3 2013 Tara Fungicide 0 0.0 0.0 8.5 0.0 2013 Tara Fungicide 1 0.8 16.2 5.7 1.5 2013 Tara Fungicide 2 1.9 34.3 5.3 11.8 2013 Tara Fungicide 3 2.7 64.1 4.6 11.8 2013 Tara No fung 0 0.0 0.0 7.9 0.0 2013 Tara No fung 1 0.9 22.1 6.1 7.4 2013 Tara No fung 2 1.8 44.3 5.3 6.2 2013 Tara No fung 3 2.7 70.5 3.3 16.3 zNo fung= no fungicide. yS. Jewel= Southern Jewel.

118 Table A.3. Interaction means of the composition attributes of percent titratable acidity, percent soluble solids, pH, and soluble solids to titratable acidity of fungicide and no fungicide treated muscadine genotypes stored at 2 °C for 0‐3 weeks. Year Genotype Fungicide Week TA pH SSC SS/TAy treatment of (%)z (%) storage 2012 AM 01 Fungicide 0 0.5 3.6 24.1 67.4 2012 AM 01 Fungicide 1 0.6 3.7 24.2 51.8 2012 AM 01 Fungicide 2 0.6 3.7 23.8 51.3 2012 AM 01 Fungicide 3 0.6 3.7 23.6 51.0 2012 AM 01 xNo fung 0 0.6 3.7 25.6 55.4 2012 AM 01 No fung 1 0.6 3.7 25.9 56.1 2012 AM 01 No fung 2 0.6 3.6 25.0 56.1 2012 AM 01 No fung 3 0.6 3.6 25.7 57.6 2012 AM 04 Fungicide 0 0.6 3.7 19.3 41.2 2012 AM 04 Fungicide 1 0.6 3.7 18.8 40.9 2012 AM 04 Fungicide 2 0.5 3.5 19.3 45.0 2012 AM 04 Fungicide 3 0.6 3.5 18.6 41.9 2012 AM 04 No fung 0 0.6 3.8 20.3 42.2 2012 AM 04 No fung 1 0.6 3.7 20.3 43.3 2012 AM 04 No fung 2 0.6 3.9 21.2 43.2 2012 AM 04 No fung 3 0.6 3.8 20.1 42.4 2012 AM 15 Fungicide 0 0.5 3.4 18.8 44.1 2012 AM 15 Fungicide 1 0.5 3.3 19.0 45.9 2012 AM 15 Fungicide 2 0.5 3.3 19.1 46.6 2012 AM 15 Fungicide 3 0.5 3.3 19.3 46.5 2012 AM 15 No fung 0 0.5 3.5 19.6 45.0 2012 AM 15 No fung 1 0.5 3.3 19.8 47.3 2012 AM 15 No fung 2 0.5 3.4 19.7 46.6 2012 AM 15 No fung 3 0.5 3.3 19.5 47.6 2012 AM 27 Fungicide 0 0.5 3.4 19.3 46.4 2012 AM 27 Fungicide 1 0.5 3.4 18.4 43.3 2012 AM 27 Fungicide 2 0.5 3.5 19.1 43.8 2012 AM 27 Fungicide 3 0.5 3.4 19.6 45.5 2012 AM 27 No fung 0 0.5 3.5 20.5 47.5 2012 AM 27 No fung 1 0.5 3.4 19.8 47.1 2012 AM 27 No fung 2 0.5 3.4 19.2 45.2 2012 AM 27 No fung 3 0.5 3.4 19.5 45.3 2012 Nesbitt Fungicide 0 0.6 3.7 20.8 45.2 2012 Nesbitt Fungicide 1 0.6 3.9 22.0 45.2 2012 Nesbitt Fungicide 2 0.6 3.8 22.2 46.8 2012 Nesbitt Fungicide 3 0.6 3.7 21.4 46.2 2012 Nesbitt No fung 0 0.5 3.4 18.3 43.6

119 Year Genotype Fungicide Week TA pH SSC SS/TAy treatment of (%)z (%) storage 2012 Nesbitt No fung 1 0.6 3.6 20.0 44.3 2012 Nesbitt No fung 2 0.6 3.7 20.5 44.6 2012 Nesbitt No fung 3 0.6 3.7 21.3 46.1 2012 Southern Jewel Fungicide 0 0.5 3.3 16.7 41.2 2012 Southern Jewel Fungicide 1 0.5 3.3 16.3 39.7 2012 Southern Jewel Fungicide 2 0.5 3.4 18.4 43.5 2012 Southern Jewel Fungicide 3 0.5 3.3 18.3 43.9 2012 Southern Jewel No fung 0 0.5 3.4 17.2 41.1 2012 Southern Jewel No fung 1 0.5 3.5 18.5 42.5 2012 Southern Jewel No fung 2 0.5 3.4 18.4 49.9 2012 Southern Jewel No fung 3 0.5 3.4 19.2 44.7 2012 Summit Fungicide 0 0.6 3.9 26.3 53.9 2012 Summit Fungicide 1 0.6 3.7 23.9 51.8 2012 Summit Fungicide 2 0.6 3.8 26.1 54.8 2012 Summit Fungicide 3 0.6 3.8 26.0 54.8 2012 Summit No fung 0 0.6 3.8 27.3 56.8 2012 Summit No fung 1 0.6 3.8 25.5 54.2 2012 Summit No fung 2 0.6 3.6 24.0 53.0 2012 Summit No fung 3 0.6 3.6 25.8 56.7 2012 Supreme Fungicide 0 0.6 3.6 19.1 42.6 2012 Supreme Fungicide 1 0.6 3.8 20.8 43.9 2012 Supreme Fungicide 2 0.6 3.8 21.1 44.7 2012 Supreme Fungicide 3 0.6 3.7 20.3 43.4 2012 Supreme No fung 0 0.6 3.6 17.8 39.2 2012 Supreme No fung 1 0.6 3.9 21.2 43.3 2012 Supreme No fung 2 0.6 3.8 20.5 42.8 2012 Supreme No fung 3 0.6 3.7 20.2 44.1 2012 Tara Fungicide 0 0.6 3.6 20.8 46.9 2012 Tara Fungicide 1 0.6 3.6 22.0 49.2 2012 Tara Fungicide 2 0.6 3.6 22.3 49.0 2012 Tara Fungicide 3 0.6 3.6 22.2 49.0 2012 Tara No fung 0 0.6 3.6 20.8 47.0 2012 Tara No fung 1 0.6 3.6 20.8 46.8 2012 Tara No fung 2 0.6 3.6 21.2 47.5 2012 Tara No fung 3 0.6 3.7 21.8 47.7 2013 AM 01 Fungicide 0 0.6 3.6 17.7 56.0 2013 AM 01 Fungicide 1 0.6 3.6 17.8 38.4 2013 AM 01 Fungicide 2 0.6 3.6 17.5 55.2 2013 AM 01 Fungicide 3 0.6 3.6 18.1 52.4 2013 AM 01 No fung 0 0.5 3.2 17.7 54.1

120 Year Genotype Fungicide Week TA pH SSC SS/TAy treatment of (%)z (%) storage 2013 AM 01 No fung 1 0.6 3.6 17.3 49.7 2013 AM 01 No fung 2 0.6 3.6 17.2 46.5 2013 AM 01 No fung 3 0.6 3.6 18.1 47.3 2013 AM 04 Fungicide 0 0.6 3.8 16.6 78.3 2013 AM 04 Fungicide 1 0.6 3.7 17.2 69.5 2013 AM 04 Fungicide 2 0.6 3.7 16.4 60.0 2013 AM 04 Fungicide 3 0.6 3.8 16.4 77.3 2013 AM 04 No fung 0 0.6 3.8 16.5 72.5 2013 AM 04 No fung 1 0.6 3.7 16.4 60.0 2013 AM 04 No fung 2 0.6 3.7 16.7 58.0 2013 AM 04 No fung 3 0.6 3.7 16.6 58.6 2013 AM 15 Fungicide 0 0.5 3.4 17.3 54.2 2013 AM 15 Fungicide 1 0.5 3.4 17.9 51.3 2013 AM 15 Fungicide 2 0.5 3.4 18.1 56.1 2013 AM 15 Fungicide 3 0.5 3.5 18.1 59.0 2013 AM 15 No fung 0 0.5 3.5 17.4 60.5 2013 AM 15 No fung 1 0.5 3.3 17.0 50.8 2013 AM 15 No fung 2 0.5 3.4 18.0 59.4 2013 AM 15 No fung 3 0.5 3.4 12.7 38.2 2013 AM 27 Fungicide 0 0.6 3.7 17.7 68.9 2013 AM 27 Fungicide 1 0.6 3.6 18.0 56.5 2013 AM 27 Fungicide 2 0.6 3.7 18.0 57.3 2013 AM 27 Fungicide 3 0.6 3.7 18.0 62.3 2013 AM 27 No fung 0 0.6 3.7 17.1 61.3 2013 AM 27 No fung 1 0.6 3.6 18.1 57.7 2013 AM 27 No fung 2 0.6 3.7 18.2 62.8 2013 AM 27 No fung 3 0.6 3.7 17.7 56.5 2013 Nesbitt Fungicide 0 0.6 3.9 16.5 57.7 2013 Nesbitt Fungicide 1 0.6 3.6 15.7 62.8 2013 Nesbitt Fungicide 2 0.6 3.7 16.4 77.9 2013 Nesbitt Fungicide 3 0.6 3.8 16.6 90.7 2013 Nesbitt No fung 0 0.6 3.9 16.6 55.2 2013 Nesbitt No fung 1 0.6 3.8 16.6 58.0 2013 Nesbitt No fung 2 0.6 3.8 17.0 71.4 2013 Nesbitt No fung 3 0.6 3.8 17.0 59.6 2013 Southern Jewel Fungicide 0 0.5 3.5 17.4 51.8 2013 Southern Jewel Fungicide 1 0.5 3.4 17.9 52.8 2013 Southern Jewel Fungicide 2 0.5 3.4 17.9 71.6 2013 Southern Jewel Fungicide 3 0.5 3.5 17.8 60.0 2013 Southern Jewel No fung 0 0.6 3.6 18.9 65.3

121 Year Genotype Fungicide Week TA pH SSC SS/TAy treatment of (%)z (%) storage 2013 Southern Jewel No fung 1 0.5 3.5 18.9 61.2 2013 Southern Jewel No fung 2 0.6 3.6 19.2 72.3 2013 Southern Jewel No fung 3 0.6 3.6 18.7 66.6 2013 Summit Fungicide 0 0.6 3.8 15.9 76.3 2013 Summit Fungicide 1 0.6 3.6 15.9 59.9 2013 Summit Fungicide 2 0.6 3.7 16.3 76.2 2013 Summit Fungicide 3 0.6 3.7 17.2 81.0 2013 Summit No fung 0 0.6 3.8 17.7 72.8 2013 Summit No fung 1 0.6 3.7 17.9 70.6 2013 Summit No fung 2 0.6 3.6 17.4 74.9 2013 Summit No fung 3 0.6 3.6 17.9 75.6 2013 Supreme Fungicide 0 0.6 4.1 17.8 104.4 2013 Supreme Fungicide 1 0.6 3.9 17.6 79.6 2013 Supreme Fungicide 2 0.6 3.9 18.3 90.9 2013 Supreme Fungicide 3 0.6 3.9 17.3 76.0 2013 Supreme No fung 0 0.6 4.0 18.1 88.5 2013 Supreme No fung 1 0.6 3.9 17.9 87.2 2013 Supreme No fung 2 0.6 3.9 17.7 75.9 2013 Supreme No fung 3 0.6 3.8 16.2 65.0 2013 Tara Fungicide 0 0.6 3.8 16.9 85.4 2013 Tara Fungicide 1 0.6 3.8 17.3 78.0 2013 Tara Fungicide 2 0.6 3.7 17.5 71.9 2013 Tara Fungicide 3 0.6 3.6 17.3 60.2 2013 Tara No fung 0 0.6 3.7 16.5 67.7 2013 Tara No fung 1 0.6 3.7 17.2 68.0 2013 Tara No fung 2 0.6 3.7 17.1 63.6 2013 Tara No fung 3 0.6 3.7 16.8 59.0 zTitratable acidity measured as tartaric acid ySoluble solids to titratable acidity ratio xNo fung= no fungicide.

122 Table A.4. Interaction means of the berry color attributes of Chroma, Hue angle, and L* values of fungicide and no fungicide treated muscadine genotypes stored at 2 °C for 0‐3 weeks. Year of Genotype Treatment Week of Chroma Hue L* study storage angle 2012 AM 01 Fungicide 0 13.6 76.2 43.7 2012 AM 01 Fungicide 1 12.6 76.0 42.3 2012 AM 01 Fungicide 2 12.6 77.3 40.1 2012 AM 01 Fungicide 3 12.1 76.5 42.0 2012 AM 01 No fungicide 0 14.0 71.3 43.2 2012 AM 01 No fungicide 1 13.4 66.4 39.6 2012 AM 01 No fungicide 2 14.4 72.4 42.3 2012 AM 01 No fungicide 3 13.5 74.1 39.5 2012 AM 04 Fungicide 0 3.7 55.2 95.1 2012 AM 04 Fungicide 1 2.9 277.2 25.2 2012 AM 04 Fungicide 2 2.6 128.9 27.2 2012 AM 04 Fungicide 3 3.1 239.4 27.3 2012 AM 04 No fungicide 0 5.6 69.5 96.8 2012 AM 04 No fungicide 1 3.4 286.1 25.5 2012 AM 04 No fungicide 2 2.5 200.9 26.9 2012 AM 04 No fungicide 3 2.6 241.3 27.0 2012 AM 15 Fungicide 0 11.7 47.6 41.4 2012 AM 15 Fungicide 1 12.8 55.5 41.6 2012 AM 15 Fungicide 2 14.2 50.5 41.5 2012 AM 15 Fungicide 3 13.7 56.9 41.9 2012 AM 15 No fungicide 0 12.2 55.8 42.3 2012 AM 15 No fungicide 1 11.9 54.1 40.8 2012 AM 15 No fungicide 2 12.9 58.9 40.7 2012 AM 15 No fungicide 3 13.8 60.2 42.7 2012 AM 27 Fungicide 0 2.4 203.2 27.6 2012 AM 27 Fungicide 1 2.2 277.3 27.2 2012 AM 27 Fungicide 2 2.9 165.7 26.8 2012 AM 27 Fungicide 3 2.4 277.4 26.7 2012 AM 27 No fungicide 0 2.1 198.5 26.8 2012 AM 27 No fungicide 1 2.6 278.5 26.8 2012 AM 27 No fungicide 2 3.1 125.6 27.2 2012 AM 27 No fungicide 3 2.6 279.7 26.0 2012 Nesbitt Fungicide 0 4.3 284.1 26.3 2012 Nesbitt Fungicide 1 4.2 201.6 26.1 2012 Nesbitt Fungicide 2 2.5 239.1 26.6 2012 Nesbitt Fungicide 3 2.9 239.2 26.9 2012 Nesbitt No fungicide 0 9.2 365.1 29.6 2012 Nesbitt No fungicide 1 5.9 284.0 25.3

123 Year of Genotype Treatment Week of Chroma Hue L* study storage 2012 Nesbitt No fungicide 2 5.3 324.6 26.8 2012 Nesbitt No fungicide 3 5.0 283.9 27.1 2012 Southern Jewel Fungicide 0 8.2 206.9 30.2 2012 Southern Jewel Fungicide 1 5.9 122.5 28.7 2012 Southern Jewel Fungicide 2 4.7 278.2 25.3 2012 Southern Jewel Fungicide 3 4.8 204.4 28.1 2012 Southern Jewel No fungicide 0 5.2 311.6 29.6 2012 Southern Jewel No fungicide 1 3.8 272.9 27.8 2012 Southern Jewel No fungicide 2 5.5 121.8 25.9 2012 Southern Jewel No fungicide 3 3.6 236.0 28.8 2012 Summit Fungicide 0 13.9 49.9 39.1 2012 Summit Fungicide 1 14.9 55.2 39.6 2012 Summit Fungicide 2 16.8 51.9 38.2 2012 Summit Fungicide 3 13.5 51.0 36.8 2012 Summit No fungicide 0 14.9 47.8 39.9 2012 Summit No fungicide 1 15.1 51.7 40.4 2012 Summit No fungicide 2 15.5 65.2 41.0 2012 Summit No fungicide 3 15.0 51.3 39.0 2012 Supreme Fungicide 0 5.9 360.0 27.9 2012 Supreme Fungicide 1 5.7 360.0 26.7 2012 Supreme Fungicide 2 3.6 290.1 27.7 2012 Supreme Fungicide 3 5.2 360.0 28.9 2012 Supreme No fungicide 0 8.6 360.0 27.9 2012 Supreme No fungicide 1 4.2 360.0 26.5 2012 Supreme No fungicide 2 4.5 360.0 27.4 2012 Supreme No fungicide 3 5.5 360.0 28.4 2012 Tara Fungicide 0 14.1 81.6 44.5 2012 Tara Fungicide 1 12.2 82.3 43.7 2012 Tara Fungicide 2 12.2 73.6 41.8 2012 Tara Fungicide 3 12.9 82.2 42.8 2012 Tara No fungicide 0 13.9 83.2 45.7 2012 Tara No fungicide 1 12.8 80.0 44.4 2012 Tara No fungicide 2 13.1 76.7 43.1 2012 Tara No fungicide 3 11.9 83.7 43.3 2013 AM 01 Fungicide 0 4.6 231.6 87.7 2013 AM 01 Fungicide 1 5.3 174.8 87.2 2013 AM 01 Fungicide 2 3.8 113.5 101.4 2013 AM 01 Fungicide 3 5.7 142.4 84.2 2013 AM 01 No fungicide 0 4.7 218.1 90.5 2013 AM 01 No fungicide 1 6.4 211.0 85.6 2013 AM 01 No fungicide 2 5.0 135.7 105.5

124 Year of Genotype Treatment Week of Chroma Hue L* study storage 2013 AM 01 No fungicide 3 3.5 207.7 83.3 2013 AM 04 Fungicide 0 3.3 242.9 26.4 2013 AM 04 Fungicide 1 3.1 321.5 24.6 2013 AM 04 Fungicide 2 3.9 366.2 25.1 2013 AM 04 Fungicide 3 3.2 284.5 24.6 2013 AM 04 No fungicide 0 3.5 159.0 26.6 2013 AM 04 No fungicide 1 2.9 122.3 25.3 2013 AM 04 No fungicide 2 2.9 285.3 24.5 2013 AM 04 No fungicide 3 3.9 287.2 24.2 2013 AM 15 Fungicide 0 7.0 262.7 92.4 2013 AM 15 Fungicide 1 11.6 54.4 36.5 2013 AM 15 Fungicide 2 14.9 56.0 39.8 2013 AM 15 Fungicide 3 14.0 62.4 37.4 2013 AM 15 No fungicide 0 9.5 256.9 91.0 2013 AM 15 No fungicide 1 11.5 69.0 42.2 2013 AM 15 No fungicide 2 12.3 64.4 41.0 2013 AM 15 No fungicide 3 13.5 63.8 32.6 2013 AM 27 Fungicide 0 2.3 344.4 26.8 2013 AM 27 Fungicide 1 1.9 307.0 25.6 2013 AM 27 Fungicide 2 1.9 310.7 26.2 2013 AM 27 Fungicide 3 2.0 272.3 25.5 2013 AM 27 No fungicide 0 2.0 342.0 25.9 2013 AM 27 No fungicide 1 1.6 306.3 26.4 2013 AM 27 No fungicide 2 2.1 345.8 26.2 2013 AM 27 No fungicide 3 2.2 347.6 25.8 2013 Nesbitt Fungicide 0 3.2 317.5 26.6 2013 Nesbitt Fungicide 1 4.4 120.7 27.7 2013 Nesbitt Fungicide 2 3.8 242.1 27.1 2013 Nesbitt Fungicide 3 4.0 237.4 26.2 2013 Nesbitt No fungicide 0 4.4 198.8 28.4 2013 Nesbitt No fungicide 1 3.1 201.7 27.0 2013 Nesbitt No fungicide 2 3.3 279.3 27.0 2013 Nesbitt No fungicide 3 4.5 328.1 25.7 2013 Southern Jewel Fungicide 0 23.4 293.3 59.8 2013 Southern Jewel Fungicide 1 2.5 309.9 26.8 2013 Southern Jewel Fungicide 2 2.5 344.1 26.2 2013 Southern Jewel Fungicide 3 3.5 236.5 27.4 2013 Southern Jewel No fungicide 0 23.1 264.1 56.1 2013 Southern Jewel No fungicide 1 2.6 313.3 27.3 2013 Southern Jewel No fungicide 2 2.6 305.1 26.3 2013 Southern Jewel No fungicide 3 2.5 272.8 25.7

125 Year of Genotype Treatment Week of Chroma Hue L* study storage 2013 Summit Fungicide 0 12.6 93.4 44.6 2013 Summit Fungicide 1 11.9 93.2 42.6 2013 Summit Fungicide 2 12.7 92.9 42.7 2013 Summit Fungicide 3 10.1 86.1 38.5 2013 Summit No fungicide 0 12.7 92.1 44.9 2013 Summit No fungicide 1 11.9 83.1 41.9 2013 Summit No fungicide 2 13.8 86.6 38.6 2013 Summit No fungicide 3 10.1 81.5 33.8 2013 Supreme Fungicide 0 6.2 204.0 25.5 2013 Supreme Fungicide 1 6.2 287.7 27.1 2013 Supreme Fungicide 2 5.8 360.0 27.1 2013 Supreme Fungicide 3 5.9 289.9 27.0 2013 Supreme No fungicide 0 5.6 285.2 25.3 2013 Supreme No fungicide 1 5.6 203.5 26.4 2013 Supreme No fungicide 2 10.6 285.5 27.2 2013 Supreme No fungicide 3 5.4 208.4 26.5 2013 Tara Fungicide 0 13.1 99.4 45.2 2013 Tara Fungicide 1 16.9 85.8 45.3 2013 Tara Fungicide 2 12.9 84.3 42.7 2013 Tara Fungicide 3 13.6 92.6 41.2 2013 Tara No fungicide 0 13.4 97.9 46.2 2013 Tara No fungicide 1 17.5 98.4 47.5 2013 Tara No fungicide 2 11.2 94.1 41.1 2013 Tara No fungicide 3 12.8 90.1 39.8

126

(mg/100 g) (mg/100 Resveratrol

Total Total phenolics phenolics (mg/100 g) (mg/100

Total Total flavonols flavonols (mg/100 g) (mg/100 ORAC ORAC (µmolTE/g) (µmolTE/g) 1.6 1.6 54.5 30.6 500.9 4.6 2.2 2.2 70.2 4.4 37.1 74.5 606.5 61.7 8.0 636.3 3.7 5.4 73.0 27.4 482.4 5.2 2.9 2.9 47.7 19.5 439.0 5.1 2.1 2.1 71.2 4.4 41.0 86.7 604.7 29.3 13.2 639.8 3.9 4.0 71.2 15.8 528.3 3.9 Total Total (mg/100 g) (mg/100 ellagitannins ellagitannins

(mg/100 g) (mg/100 anthocyanins AM 01 AM 01 AM Fungicide 04 AM fung No 04 AM Fungicide 15 AM 0.0 fung No 15 AM Fungicide 27 AM 127.8 0.0 fung No 27 AM Fungicide 109.3 Nesbitt 0.0 fung No Nesbitt Fungicide 14.0 S.Jewel 107.9 0.0 fung No S.Jewel Fungicide 7.1 122.0 Summit 74.0 fung No Summit 74.3 Fungicide 17.9 4.9 31.5 Supreme fung No Fungicide Supreme 71.3 8.0 78.1 fung No 7.5 Tara 0.0 36.2 Tara 125.3 10.5 2.0 0.0 2.5 01 AM Fungicide 26.1 01 AM 16.7 81.7 512.4 fung No 4.3 103.3 Fungicide 04 AM 14.1 29.1 fung No 04 AM 0.0 0.6 72.2 539.2 88.6 10.5 Fungicide 15 AM 5.2 0.0 21.5 16.5 fung No 0.0 15 AM 1.6 67.6 812.7 71.9 Fungicide 0.0 42.8 7.0 17.3 fung No 20.8 53.4 86.8 669.3 498.4 41.1 15.3 0.0 9.1 44.7 52.7 518.6 580.6 13.0 0.0 16.7 63.1 5.5 5.0 12.2 651.4 694.7 7.3 11.0 4.3 6.7 492.1 98.0 369.9 11.9 9.3 5.8 97.2 366.1 81.5 4.8 19.5 2.9 87.2 18.4 4.1 38.9 492.3 47.9 558.3 530.9 4.2 604.3 5.5 3.9 3.8

127 Year Year Genotype Fungicide Total 2012 2012 2012 2013 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012 2012 2013 2013 2013 2013 2013 Table A.5. Interaction means of the berry nutraceutical concentrations of total phenolics, total anthocyanins, total total phenolics,anthocyanins, total total of concentrations nutraceutical berry the of means Interaction A.5. Table treated fungicide fungicide no and of (ORAC) capacity antioxidant the and resveratrol, and flavonols, total ellagitannins, genotypes. muscadine Year Genotype Fungicide Total Total ORAC Total Total Resveratrol anthocyanins ellagitannins (µmolTE/g) flavonols phenolics (mg/100 g) (mg/100 g) (mg/100 g) (mg/100 g) (mg/100 g) 2013 AM 27 Fungicide 46.7 6.6 119.0 17.4 543.4 4.9 2013 AM 27 No fung 41.8 6.6 94.8 14.7 448.3 4.6 2013 Nesbitt Fungicide 49.4 12.4 85.9 19.9 560.5 3.2 2013 Nesbitt No fung 22.5 7.7 67.8 11.0 450.1 8.1 2013 S. Jewel Fungicide 44.6 9.0 109.8 18.9 655.8 3.9 2013 S. Jewel No fung 31.3 7.5 111.3 15.7 579.0 3.2 2013 Summit Fungicide 0.0 13.1 64.6 32.4 579.2 8.1 2013 Summit No fung 0.0 11.2 57.6 23.4 535.0 11.1 2013 Supreme Fungicide 9.1 5.9 56.5 8.4 315.5 5.2 2013 Supreme No fung 19.9 5.7 57.8 11.5 354.2 12.1 2013 Tara Fungicide 0.0 5.9 60.6 14.8 452.7 3.5 2013 Tara No fung 0.0 4.2 65.9 12.7 476.6 3.3 128

Table A.6. Study 1 multivariate correlation coefficients among muscadine berry storage quality, composition, color, and nutraceutical content for 2012 and 2013. *Significant at P=0.05. WLz UMy VCx TAw pH SSv L* Chru Huet Fors Anthq Ellap ORACo Flan TPm WLz UMy -0.1 VCx -0.2 -0.29 TAw 0.2 -0.36 0.39 pH 0.3 -0.35 0.40 0.99* SSv 0.1 0.26 -0.24 0.26 0.29 L* 0.4 0.63 -0.59 -0.28 -0.23 0.59 Chru 0.1 0.34 -0.27 -0.17 -0.17 0.48 0.59 Huet -0.1 -0.52 0.47 0.23 0.23 -0.52 -0.72 -0.9* Fors 0.3 -0.7* 0.03 0.53 0.52 -0.21 -0.46 -0.53 0.50 Anthq -0.3 -0.24 0.12 0.02 0.01 -0.52 -0.59 -0.9* 0.64 0.43 Ellap -0.2 -0.35 -0.27 0.10 0.08 -0.01 -0.19 -0.02 -0.20 0.49 0.34 ORACo -0.5 -0.05 -0.17 -0.51 -0.51 -0.39 -0.28 -0.52 0.27 0.06 0.69 0.31 Flan 0.1 0.18 -0.59 -0.39 -0.39 0.36 0.55 0.58 -0.78 -0.21 -0.35 0.49 0.11 TPm -0.5 0.21 -0.46 -0.71 -0.70 -0.04 0.17 0.01 -0.25 -0.23 0.26 0.33 0.77* 0.5 Resl -0.3 -0.02 0.04 0.11 0.09 0.39 -0.06 -0.27 0.16 -0.02 0.24 0.06 0.23 0.1 0.2 zWL=weight loss of berries (%) yUM=umarketable berries (%) xVC=volume change (%) wTA=titratable acidity (%) measures as tartaric acid vSS=soluble solids (%) uChr=Chroma tHue=hue angle sFor=force to penetrate berry skin (%) qAnth=total anthocyanins (mg/100 g) pElla=total ellagitannins (mg/100 g) oORAC=oxygen radial absorbance capacity (µmol Trolox equivalent/ g) nFla=total flavonols (mg/100 g) mTP=total phenolics (mg/100 g) lRes=trans-resveratrol (mg/100 ) 129

Chapter 2

EVALUATION OF DIVERSE MUSCADINE GRAPE GENOTYPES FOR POSTHARVEST

STORAGE POTENTIAL AND NUTRACEUTICAL CONCENTRATIONS

Abstract

A major limiting factor in muscadine grape (Vitis rotundifolia) commercialization is deterioration during storage. One solution for extending market seasons and preventing market saturation for fresh muscadines could be the release of new cultivars with improved postharvest storability. Storage and composition attributes including berry volume, titratable acidity (TA), pH, soluble solids (%), berry color (L*, Chroma, and hue), firmness (force to penetrate berry skins), storage weight loss (%), and unmarketable berries (%) were evaluated on eight muscadine cultivars and nine breeding selections grown at the University of Arkansas Fruit Research Station, Clarksville, AR in 2012 and

2013. The objective of the study was to identify muscadine genotypes with potential for commercialization and postharvest storability. Muscadine genotypes were hand‐harvested and evaluated for postharvest performance and composition every 7 d for 3 weeks. The nutraceutical measurements of total anthocyanins, total ellagitannins, total flavonols, total phenolics, resveratrol, and the antioxidant capacity measurement of oxygen radical absorbance capacity (ORAC) of muscadines were measured at date of harvest. Overall, the postharvest storage quality attributes of weight loss, percent unmarketable, firmness, and volume change were significantly affected by genotype, year, and storage time. The berry color attributes, Chroma and hue, and the berry composition attributes of TA, soluble solids, and pH were significantly affected by genotype and year, but remained fairly constant across time of storage. The berry color attribute of L* was affected by genotype,

130 year, and week of storage showing a decrease, or darkening, after date of harvest. The berry nutraceutical concentrations of total anthocyanins, total ellagitannins, total flavonols, total phenolics, resveratrol, and the antioxidant capacity measurement of ORAC were significantly affected by genotype and year of the study. Storage attributes varied by genotype and year; overall AM 04, AM 26, AM 28, and ‘Southern Jewel’ were identified as having the highest potential for postharvest storage, while the genotypes AM 01, AM 15,

AM 18, and ‘Nesbitt’ had the least potential. Nutraceutical content varied by genotypes; overall AM 03, AM 04, AM 27, and ‘Ison’ had the highest nutraceutical content (sum of anthocyanins, total phenolics, flavonols, resveratrol, and ORAC), while AM 18, AM 28,

‘Supreme’, and ‘Tara’ had the lowest. It was determined that postharvest storage potential, berry composition, berry color, and nutraceutical content were genotype specific, but commercially viable genotypes were identified.

131 Introduction

Muscadine grapes (Vitis rotundifolia Michx.) are indigenous to the southeastern

United States. Muscadines have been under cultivation for over 400 years, originally in the

North Carolina Colony followed by surrounding colonies and states (Conner, 2009). This native grape is presently grown in small commercial vineyards and home plantings, ranging from North Carolina and Florida to eastern Oklahoma and Texas. Arkansas has approximately 230 ha of muscadines in production, making up 10% of total US production.

The recent recognition that muscadines are sources of beneficial antioxidants has increased consumer demand (Perkins‐Veazie et al., 2012; Striegler et al., 2005).

Additionally, alternative crops, including muscadines, are being explored by growers in the

South as a means of increasing profits or diversifying farm operations (Conner, 2009).

Muscadine grapes are produced in small loose clusters that do not ripen evenly. The berries are removed from the vines as single berries rather than in clusters as with bunch grapes and are subsequently marketed in packages (often clamshells) (Perkins‐Veazie et al., 2012). Though commercial shipping for retail marketing is practiced, the potential for

expansion in this area exists (Perkins‐Veazie et al., 2012). Major limiting factors on fresh‐

market muscadine production include uneven ripening, short harvest season, and high

perishability of the fruit (James et al., 1999; Morris, 1980; Perkins‐Veazie et al., 2012).

Storage protocols to assist the muscadine grape industry in providing quality fresh‐market grapes are limited (Walker et al., 2001).

Several variables contribute to muscadine storage performance, including berry maturity, texture (crispness), weight loss, decay, shriveling, browning, leakage, and percentage of dry stem scars (where the berry releases from the pedicel). Muscadines

132 harvested at physiological ripeness have been shown to successfully store for 2 to 3 weeks

(Lutz, 1938; Perkins‐Veazie et al., 2012; Takeda et al., 1982). To maintain adequate quality,

muscadines should be stored from 1 to 5 °C with 85‐95% relative humidity (RH) (Lutz,

1938; Silva et al., 1994; Takeda et al., 1983; Walker et al., 2001). The use of sulfur dioxide

(SO2) storage treatment on the quality of table grapes is genotype specific, and with

muscadines not reliably beneficial (Ballinger and Nesbitt, 1982a; Ballinger and Nesbitt,

1982b; James et al., 1997; James et al., 1999; Lane, 1978; Lane and Flora, 1980; MacLean et

al., 2009; Morris et al., 1992; Smit et al., 1971; Conner and Maclean, 2012).

Muscadine grapes contain phenolic acids, flavonols, anthocyanins, and ellagitannins

(Boyle and Hsu, 1990; Haung et al., 2009; Lee et al., 2005; Pastrana‐Bonilla et al., 2003;

Stringer et al., 2009; Talcott and Lee, 2002). Polyphenolic concentrations usually increase

in muscadines as the berries ripen (Lee et al., 2005) and are higher in wine than in

unfermented juices (Musingo et al., 2001; Talcott and Lee, 2002). It has been shown that

the muscadine grape possesses one of the highest antioxidant levels among fruit crops

(Greenspan et al., 2005). Some of these components of muscadines have been shown to

have anti‐cancer, anti‐mutagen, and anti‐inflammatory properties, and to reduce levels of

glucose, insulin, and glycated hemoglobin in people with diabetes (Banini et al., 2006;

Bralley et al., 2007; Greenspan et al., 2005; God et al., 2007; Yi et al., 2005).

Since the implementation of a muscadine breeding program at the University of

Arkansas in 2005, selections have been made based on improved texture and dry stem

scar. Although increased crispiness and a greater percentage of dry stem scars has been

observed, it is unknown whether there has been a true improvement in postharvest quality

of muscadines. Nutraceutical levels in muscadines vary among genotypes and no

133 information has been collected on the nutraceutical content of the University of Arkansas

breeding selections (J.R. Clark, personal communication; Marshall et al., 2012).

The objectives of this study were to develop a postharvest evaluation protocol for

Arkansas muscadine genotypes for potential commercial utilization and to identify superior post‐harvest storage and handling muscadine genotypes in the University of

Arkansas Fruit Breeding Program. It was hypothesized that commercial cultivars along with recently developed Arkansas genotypes with improved skin and flesh texture will vary in storage potential.

Materials and Methods

Grapes and Vineyard

The berries of 17 genotypes (AM 01, AM 02, AM 03, AM 04, AM 15, AM 18, AM 26,

AM 27, AM 28, ‘Delicious’, ‘Fry’, ‘Ison’, ‘Nesbitt’, ‘Southern Jewel’, ‘Summit’, ‘Supreme’ and

‘Tara’) were evaluated for postharvest storability and nutraceutical content. Vines used for

the study were grown at the University of Arkansas Fruit Research Station, Clarksville, AR

(lat. 35°31’58’N and long. 93°24’12’W). Vines were of varying ages within each genotype,

most of the cultivars were approximately six years old, while many of the selections were from younger vines three to four years old. The vines were grown in Linker fine sandy loam, in USD hardiness zone 7a, where average annual minimum temperatures reach ‐15 –

17.7 °C. Vine spacing was 6.1 m apart and rows were spaced 3.0 m apart. A single‐wire trellis was used, and vines were trained to a bilateral cordon. The vines were dormant

pruned annually in February using spur pruning with spurs retained of two to four buds in

length. Weeds were controlled with pre‐ and postemergence herbicides as needed and

vines did not have any stress from weed competition. Vines were irrigated by trickle

134 irrigation as needed, beginning in early June (prior months received adequate rainfall) and continuing through the harvest period. Vines received N fertilization in March of each year

at a rate of approximately 70 kg/ha. No insecticides or other pest control compounds were

applied to the vines. The vines used in the study had full crops produced each year, and no

crop reduction due to winter injury or other limitation occurred. Thus, the vines produced

fruit under full‐crop conditions. Daily maximum and minimum temperatures along with

rainfall were recorded at the research location to characterize the environment the vines

were subjected to and potential differences among years (Table A.1).

Harvest and Transport

The muscadines were once‐over, hand‐harvested at the Fruit Research Station. Fruit

was harvested either early in the morning or late in the afternoon and transported to

University of Arkansas Institute of Food Science and Engineering, Fayetteville, AR., in an

air‐conditioned car on the same day. Harvest date/maturity was based on soluble solids of

18‐22% in 2012 and 15‐18% in 2013, due to differences in summer temperature and

precipitation, ease of release from the pedicel, and berry color.

Storage Study

Berries were hand‐sorted to remove any split, shriveled, or decayed fruit before

packaging to simulate commercial standards. Only sound berries showing no signs of

unmarketability were used for the study. The fruit was packaged into hinged standard

vented polyethylene clamshells (18.4 cm x 12.1 cm x 8.9 cm) (H116, FormTex Plastics

Corporation, Houston, TX) and stored in plastic harvest lugs in cold storage at 2 °C with 85‐

89% RH.

135 From the harvested fruit, six vented clamshell containers were filled to

approximately 500 g. Three of these clamshells were used as storage replications for each

genotype. Weight loss and percent unmarketable berries were evaluated on the storage

clamshells every 7 d for up to 21 d. Storage performance was evaluated by removing all the

berries from each clamshell, and counting the number of berries with signs of

unmarketability, which included characteristics of browning, softness, mold, rot, leakage,

splitting, and shriveling (Conner, 2013; Conner and Maclean, 2012; P. Perkins‐Veazie,

personal communication). Both the unmarketable and marketable berries were returned to the appropriate clamshell each week, and storage measurements were discontinued once the percent unmarketable in all three clamshells reached 50%, or after 3 weeks of storage. Each week during storage, berries were sent to the University of Arkansas

Cooperative Extension Service Plant Health Clinic, Fayetteville, AR for disease diagnostics.

Reports from the Clinic were provided on the fungal species isolated.

The remaining three clamshells were used as composition replications 1, 2, and 3.

For composition measurements, every 7 d three berries were removed from each of the

three composition clamshells and used to measure berry volume, Chroma, hue, L*, soluble solids (%), titratable acidity (%) (TA), pH, and firmness of the skin and flesh (N).

Berry and Compositional Analysis

The composition procedures used were modeled from previously reported

protocols (Conner, 2013; Conner and Maclean, 2012; Striegler et al., 2005; Slinkard and

Singleton, 1977; Threlfall et al., 2005; Threlfall et al., 2007; Walker et al., 2001).

Berry volume was determined by measuring the height and width of three randomly selected berries from each replication using digital calipers. The formula used for

136 calculating the berry volume was: berry volume = 4/3π*berry height*berry width. Change

in berry volume was determined by calculating the percent of volume difference of the berries during storage from the initial berry volume. Decrease in size during storage was shown with positive values, while an increase in size during storage was shown by negative values.

Titratable acidity and pH were measured by an 877 Titrino Plus (Metrohm AG,

Herisau Switzerland) with an automated titrimeter and electrode standardized to pH 2.0,

4.0, 7.0, and 10.0 buffers. Titratable acidity was determined using 6 g of juice diluted with

50 mL of deionized, degassed water by titration of 0.1 N sodium hydroxide (NaOH) to an

endpoint of pH 8.2, and results were expressed as percent tartaric acid. Soluble solids were

measured using a Bausch and Lomb Inc. Abbe Mark II refractometer (Rochester, NY).

Soluble solids, TA, and pH were measured from the juice of the whole berries, strained through cheesecloth to remove any solids.

Exterior skin color measurements were determined on each of the three berries every 7 d using a Chroma Meter CR 300 series (Konica Minolta Holdings Inc., Ramsey, N J).

The Commission Internationale de I’Eclairage (CIE) Lab transmission “L” value indicates how dark or light the skin is, with 0 being black and 100 being white. Hue angle describes color in angles from 0° to 360°: 0° = red; 90° = yellow; 180° = green; 270° = blue; and 360°

= back to red. Chroma is the aspect of color by which the skin colors appears different from gray of the same lightness and corresponds to intensity of the perceived color.

Firmness, or the maximum force to penetrate skin and flesh tissues, was determined using three whole berries per replication. A TA‐XT2 Texture Analyzer (Stable Micro

Systems, Haslemere, UK) equipped with a 2‐mm‐diameter probe used to penetrate the

137 exocarp and mesocarp tissues (flesh) to a depth of 10 mm in each berry at a rate of 10

mm.s‐1. Measurements were expressed as force in Newtons (N), and the data was analyzed

using Texture Expert Version 1.17 (Texture Technologies Corp., Scarsdale, NY).

Nutraceutical Analysis

Three randomly selected berries from each composition replication of each treatment were used from the harvest date sample to measure oxygen radical absorbance

capacity (ORAC), total phenolics, total anthocyanins, total ellagitannins, total flavonols, and resveratrol content by high‐performance liquid chromatography (HPLC) with modified methods determined by Cho et al. (2004), Cho et al. (2005), Hager et al. (2008), and Prior et al. (2003). The berries were homogenized three times each for 1 min in alternating washes

of 80 ml of extraction solution containing methanol/water/formic acid (MWF) (60:37:3 v/v/v) and acetone/water/acetic (70:29.5:0.5 v/v/v) to the smallest particle size using a

Euro Turrax T18 Tissuemizer (Tekmar‐Dohrman Corp, Mason, OH). Homogenates were then centrifuged for 5 min at 10,000 rpm and filtered through Miracloth (CalBiochem,

LaJolla, CA). The samples were taken to a final volume of 250 mL with extraction solvent and stored at ‐70 °C until analysis. Prior to HPLC analysis, the samples were filtered through 0.45 μm filters (Whatman PLC, Maidstone, UK).

Total phenolics were measured using the Folin‐Ciocalteu assay (Slinkard and

Singleton 1977) on a diode array spectrophotometer (8452A; Hewlett Packard, Palo Alto,

CA), with a gallic acid standard and a consistent standard curve based on sequential dilutions. Samples were prepared with 1 ml 0.2N Folins reagent, 0.8 ml Na2CO3 (75g/L) and

0.2 ml of extracted sample with a reaction time of 2 h. Absorbance was measured at 760

nm, and results were expressed as gallic acid equivalents (GAE).

138 For flavonoid analysis, subsamples (5 ml) of supernatant were evaporated to

dryness using a SpeedVac® concentrator (ThermoSavant, Holbrook, NY) with no radiant

heat and suspended in 1 ml of aqueous 3% formic acid solution. Samples (1 mL) were

analyzed using a Waters HPLC system equipped with a model 600 pump, a model 717 Plus

autosampler and a model 996 photodiode array detector. Separation was carried out using

a 4.6 mm × 250 mm Symmetry® C18 column (Waters Corp, Milford, MA) with a 3.9 mm ×

20 mm Symmetry® C18 guard column. The mobile phase was a linear gradient of 5%

formic acid and methanol from 2% to 60% for 60 min at 1 ml min−1. Prior to each injection,

the system was equilibrated for 20 min at the initial gradient. Detection wavelength was

510 nm for anthocyanins. Individual anthocyanin diglycosides were quantified as

delphinidin (Dp), cyanidin (Cy), petunidin (Pt), peonidin (Pn), pelargonidin (Pg), and

malvidin (Mv) diglycoside equivalents. Total anthocyanins were calculated as the sum of

individual compounds and their derivatives.

For total flavonol and ellagitannin analysis, samples (5 ml) of supernatant were

evaporated to dryness using a SpeedVac® concentrator with no radiant heat and

suspended in 1 ml of aqueous 50% methanol solution. The samples were analyzed using a

Waters HPLC system (Waters Corp, Milford, MA) equipped with a model 600 pump, model

717 plus autosampler and model 996 photodiode array detector. Separation was carried out using a 4.6 mm × 250 mm Aqua® C18 column (Phenomenex, Torrance, CA) with a 3.0 mm × 4.0 mm ODS® C18 guard column (Phenomenex). The mobile phase was a gradient of

20 g kg−1 acetic acid (A) and 5 g kg−1 acetic acid in water and acetonitrile (50:50 v/v, B)

from 10% B to 55% B in 50 min and from 55% B to 100% B in 10 min. Prior to each

injection, the system was equilibrated for 20 min at the initial gradient. A detection

139 wavelength of 360 nm was used for flavonols and 280 nm for ellagitannins at a flow rate of

1 ml min−1. Flavonols and ellagitannins were expressed as mg rutin equivalents kg−1 fresh weight.

Although, resveratrol is produced in two forms, trans‐ and cis‐resveratrol, this study focused only on the trans isomer. Trans‐resveratrol (3,4′,5‐Trihydroxy‐trans‐stilbene, 5‐

[(1E)‐2‐(4‐Hydroxyphenyl)ethenyl]‐1,3‐benzenediol) concentrations were confirmed using an analytical standard (ID:24860876; Sigma‐Aldrich Co. LLC, St. Louis, MO).

For flavonol and ellagitannin confirmation, a representative bronze and black genotype were analyzed using mass spectrometry (MS). For HPLC/MS analysis the HPLC apparatus was interfaced to a Burker Esquire (Burker Corporation, Billerica, MA) LC/MS ion trap mass spectrometer. Mass spectral data were collected with the Bruker software, which also controlled the instrument and collected the signal at 360 or 510 nm. Typical conditions for mass spectral analysis in negative ion electrospray mode for flavonols included a capillary voltage of 4000 V, a nebulizing pressure of 30.0 psi, a drying gas flow of

9.0 ml min−1 and a temperature of 300 °C. Data were collected in full‐scan mode over a

mass range of m/z 50 – 1000 at 1.0 s per cycle. Characteristic ions were used for peak assignment.

The ORAC of muscadine extracts was measured using the method of Prior et al.

(2003) modified for use with a FLUOstar Optima microplate reader (BMG Labtechnologies,

Durham, NC) using fluorescein as a fluorescent probe. Muscadine extracts were diluted

1600‐fold with phosphate buffer (75 mM, pH 7) prior to ORAC analysis. The assay was carried out in clear 48‐well Falcon plates (VWR, St. Louis, MO), each well having a final volume of 590 μl. Initially, 40 μl of diluted sample, Trolox (TE) standards (6.25, 12.5, 25, 50

140 μM) and a blank solution of phosphate buffer were added to each well. The FLUOstar

Optima instrument equipped with two automated injectors was programmed to add 400 μl of fluorescein (0.108 μM) followed by 150 μl of 2,2’‐azobis(2‐amidino‐propane) dihydrochloride (AAPH) (31.6 mM) to each well. Fluorescence readings (excitation 485 nm, emission 520 nm) were recorded after the addition of fluorescein and AAPH and every 192 s for 112 min to reach 95% loss of fluorescence. Results were based upon differences in areas under the fluorescein decay curve between the blank, samples, and standards, and expressed relative to the initial reading. The standard curve was obtained by plotting the four concentrations of TE against the net area under the curve of each standard. Final ORAC values were calculated using the regression equation between TE concentration and the net area under the curve and expressed as μmol TE equivalents kg−1 fresh weight.

Experimental Design

The postharvest experiment was a designed split‐split plot with three replications of

each genotype. The first split was storage (weeks 0, 1, 2, and 3) and the second split was

year (2012 and 2013). The nutraceutical experiment was a split‐plot design with three

replications of each genotype, and the split being year, (these measurements were only

done on the harvest date, not at each storage date).

Experimental Analysis

The data was analyzed by analysis of variance (ANOVA) using JMP® (version 11.0;

SAS Institute Inc., Cary, NC). Tukey’s Honest Significant Difference and Student’s t Test

were used for mean separations (p = 0.05). Associations among all dependent variables were determined using multivariate pairwise correlation coefficients of the mean values

using JMP (version 11.0; SAS Institute Inc., Cary, NC).

141 Results

Initial Attributes

From a production and marketing perspective, the initial measurement of berry force, TA, pH, soluble solids, L*, Chroma, and hue angle are of particular importance.

Averaged across year ‘Nesbitt’, AM 04, AM 28, and AM 03 had the highest initial force (11.5,

10.6, 10.5, and 10.0 N, respectively), while ‘Ison’, ‘Delicious’, ‘Fry’, and ‘Tara’ had the least

(7.4, 7.7, 7.8, and 7.8 N, respectively) (Table 2.1). TA ranged from 0.3 to 0.5, while pH ranged from 3.3 (‘Ison’) to 3.9 (AM 02 and AM 26) and soluble solids ranged from 16.9%

(AM 18) to 22.5% (‘Summit’), averaged across years (Table 2.1). The bronze genotypes AM

03, AM 01, and AM 15 had the highest L* values (71.9, 66.8, and 66.6, respectively, while the black genotypes AM 28, AM 27, and ‘Supreme’ had the lowest (25.9, 26.3, and 26.6, respectively) (Table 2.1). The bronze genotypes had Chroma values that ranged from 9.4

(AM 01) to 15.6 (‘Fry’) and hue angles that ranged from 70.0° (‘Summit’) to 156.3° (AM 15), the black genotypes had Chroma values ranged from 2.0 (AM 27) to 14.2 (‘Southern Jewel’) and hue angles that ranged from 114.3° (AM 04) to 330.3° (AM 02) (Table 2.1).

Berry Storage Attributes

The postharvest berry diseases present were identified as black rot (Guignardia

[Phyllosticta] bidwellii [ampelicida] Ellis.), myrothecium leaf spot (Myrothecium sp./spp.), and botrytis fruit rot, (Botrytis sp./spp.). Occasionally an unknown species of fruit fly

(Drosophila sp.) was present, but only in berries with wet or torn stem scars.

The ANOVA F‐test indicated significant (P<0.0001) three‐way interactions of year by week of storage by genotype for weight loss, unmarketable berries, force to penetrate berry skin, and volume change (Table 2.2).

142 Table 2.1. Initial mean values for force to penetrate berry skin, composition and berry color for muscadine genotypes averaged across year (2012 and 2013). Genotype Berry Force Titratable pH Soluble L* Chroma Hue color (N) acidity solids angle (%) (%) AM 01 Brz 8.8 0.5 3.5 21.6 66.8 9.4 144.7 AM 02 Bl 9.5 0.5 3.9 18.9 26.7 2.1 330.3 AM 03 Br 10.0 0.3 3.7 20.8 71.9 10.9 92.3 AM 04 Bl 10.6 0.4 3.8 18.4 61.7 4.6 114.3 AM 15 Br 8.5 0.5 3.5 18.5 66.6 10.9 156.3 AM 18 Bl 8.5 0.3 3.8 16.9 44.4 2.8 263.9 AM 26 Br 9.9 0.4 3.9 19.1 43.1 13.3 85.2 AM 27 Bl 9.2 0.4 3.6 18.8 26.3 2.0 270.3 AM 28 Bl 10.5 0.4 3.7 17.6 25.9 3.7 322.8 Delicious Bl 7.7 0.4 3.5 18.5 43.7 2.6 245.3 Fry Br 7.8 0.4 3.7 20.6 44.0 15.6 91.3 Ison Bl 7.4 0.5 3.3 17.4 27.2 5.3 317.0 Nesbitt Bl 11.5 0.4 3.6 17.5 29.0 6.8 279.3 Southern Jewel Bl 8.9 0.4 3.5 18.1 42.8 14.2 287.8 Summit Br 9.2 0.5 3.8 22.5 42.4 13.8 70.0 Supreme Bl 8.8 0.4 3.8 18.0 26.6 7.1 321.1 Tara Br 7.8 0.4 3.6 18.7 45.9 13.6 90.5 zBronze=Br and black=Bl.

Table 2.2. F‐test significance from ANOVA for muscadine berry weight loss, percent unmarketable berries, force required to penetrate the berry skin, and percent volume change of the berry during 3 weeks of storage at 2 °C. Highest‐order interactions are italicized and shaded (2012 and 2013). Degrees Weight Unmarketable Force Volume of loss (%) (%) (N) change (%) freedom Year 1 <0.0001 0.3305 <0.0001 <0.0001 Week 3 <0.0001 <0.0001 <0.0001 0.0011 Year*week 3 <0.0001 <0.0001 <0.0001 <0.0001 Genotype 16 <0.0001 <0.0001 <0.0001 <0.0001 Year*genotype 16 <0.0001 <0.0001 <0.0001 <0.0001 Week*genotype 48 0.0003 <0.0001 <0.0001 0.0418 Year*week*genotype 48 <0.0001 <0.0001 <0.0001 <0.0001

143 Weight Loss. For the dependent variable percent weight loss, large differences

among overall week means were found, ranging from 1.4% for week 1, 2.7% for week 2, and 4.0% for week 3, with all means being significantly different from each other

(P<0.0001) (Table A.7) indicating that longer storage resulted in more weight loss. After 3 weeks of storage, ‘Nesbitt’ and AM 18 in 2012 had the highest percent weight loss (6.5% and 6.2%, respectively), while AM 28 and ‘Southern Jewel’ in 2013 had the lowest (1.9% and 2.0%, respectively) (Fig. 2.1 and Table A.7). Overall weight loss values appear to be lower for most genotypes in 2013 compared to 2012 (Fig. 2.1). The variable response of genotypes for years, with the example of ‘Nesbitt’ highlighted, provides some insight why the interaction was significant.

Unmarketable Berries. For percent unmarketable berries, large differences were

found among week means were found, ranging from 9.9% for week 1, 20.6% for week 2,

and 42.7% for week 3, all means significantly different from each other (P<0.0001) (Table

A.7), again highlighting the increasing difficulty in increased storage time to maintain marketable fruits. After 3 weeks of storage, AM 01 in 2013 had the highest percent unmarketable (95.0 %), while AM 26 in 2013 had the lowest (8.8%) (Fig. 2.2 and Table

A.7). When examining major trends in the data, it can be seen that overall unmarketable values appear to be relatively similar among most genotypes in 2013 compared to 2012, with a few exceptions (Fig 2.2). The overall difference in means between years for percent unmarketable after 3 weeks of storage is 1.9%, with 2013 having slightly higher percent unmarketable (Fig. 2.2). The genotypes AM 01, AM 04, AM 26, AM 28, and ‘Nesbitt’ performed quite differently among years, and likely contributed the greatest to the significant interaction (Fig. 2.2.).

144 Fig. 2.1. Percent berry weight loss of muscadine genotypes stored at 2 °C for 3 weeks. Values at week 0 (date of harvest) were excluded. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). Week 7 1 6 2 2012 5 3 4 3 2

1 Year

loss (%) 0 t 7

Weigh 6 5 2013 4 3 2 1 0 1 2 3 4 5 8 6 7 8 s n t l it e a 0 0 0 0 1 1 2 2 u o it e r io Fry Is b w m m a M M M M M M M M 2 M c es e m re T A A A A A A A A A li J u p e N n S u D er S th u o S

Genotype

145 Fig. 2.2. Percent unmarketable of muscadine genotypes stored at 2 °C for 3 weeks. Values at week 0 (date of harvest) were excluded. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). Week 80 1 70 2 2012 60 3 50 40 30 20

10 Year 0 etable (%) k 80 70 2013

Unmar 60 50 40 30 20 10 0 1 2 3 4 5 8 6 7 8 s t l it e a 0 0 1 1 2 2 u on t e r io Fry Is bi w m m M 0 M M M 0 M M M 2 M M c es e m re Ta A A A A A A A A A li J u p e N n S u D er S th u o S

Genotype

146 Force. Large differences were found in force to penetrate the berry skin among weeks of storage, ranging from 9.1 N for week 0, 6.9 N for week 1, 6.6 N for week 2, and 6.2

N for week 3, all means significantly different from each other (P<0.0001) (Table A.7).

However, it was noteworthy the major reduction after one week of storage, but minimal practical reductions in force after additional storage. Generally force to penetrate the berry skin decreased during storage, however weeks 1 or 2 often had the lowest force in some genotypes, especially in 2012 (Fig. 2.3 and Table A.7). After 3 weeks of storage AM 04 in

2013 had the greatest firmness (11.2 N), while AM 15 in 2012 had the least (2.2 N) (Fig. 2.3 and Table A.7). The overall difference in means between years was 1.9 N. The overall trend in the data shows that the force to penetrate the berry skin appears to be greater for most genotypes in 2013 compared to 2012, providing some insight why the interaction was significant (Fig. 2.3 and Table A.7). And again, differences among genotypes among years and weeks indicate that some genotypes likely offer specific postharvest performance advantages.

Volume Change. Volume change resulted from a decrease in berry size shown with positive numbers, with an increase in size shown by negative numbers. The same significant three‐way interaction for weight loss, unmarketable, and force was also identified for volume change (Table 2.2). Differences were found in volume change during storage ranging from ‐2.9% for week 1, 0.5% for week 2, and 5.1% for week 3, all means significantly different from one another (P=0.0011). The mean difference in volume change for years was 10.4%, with the overall trend showing more increase in size during storage in

2012, and more decrease in size in 2013 (Fig 2.4 and Table A.7). AM 02, AM 28 after 2 weeks of storage, and ‘Supreme’ after 1 week of storage in 2012 showed the greatest

147 Fig. 2.3. Force to penetrate skin of muscadine genotypes stored at 2 °C for 3 weeks Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). Week 12 0 10 1 2012 8 2 6 3 4

2 Year 0 12 Force (N) 10 2013 8 6 4 2 0 1 2 3 4 5 8 6 7 8 s y n t l it e a 0 1 2 2 2 u r o it e r 0 0 0 1 io F Is b w m m a M M M M M M M M M c es e m re T A A A A A A A A A li J u p e N n S u D er S th u o S

Genotype

148 Fig. 2.4. Percent change in volume of muscadine genotypes stored at 2 °C for 3 weeks. Values at week 0 (date of harvest) were excluded. Decrease in size shown with positive values, while an increase in size shown by negative values. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). Week 20 1 10 2 2012 0 3 ‐10 ‐20 ‐30

‐40 Year ‐50 20 10 2013

Volume change (%) 0 ‐10 ‐20 ‐30 ‐40 ‐50 1 2 3 4 5 8 6 7 8 s t l it e a 0 0 1 1 2 2 u ry on t e r io F Is bi w m M 0 M M M 0 M M M 2 M M c es e mm re Ta A A A A A A A A A li J u p e N n S u D er S th u o S

Genotype

149 increase in volume (‐35.4%, ‐31.9%, and ‐31.7%, respectively), while ‘Ison’ and ‘Nesbitt’ stored for 3 weeks in 2013, showed the greatest decrease in volume (21.2% and 22.5%, respectively) (Fig. 2.4 and Table A.7). Many genotypes that decreased in size in 2012, often increased in size in 2013, offering some insight why the interaction was significant.

Berry composition

The ANOVA F‐test indicated a significant three‐way interaction of year by week of storage by genotype for pH (P=0.0046), and titratable acidity (TA), soluble solids content, and soluble solids to titratable acidity ratio (SS/TA) (P<0.0001) (Table 2.3).

pH. For the dependent variable pH, no significant differences among week means were found (P=0.3220) (Fig. 2.5 and Table 2.3). Though significant, the mean differences in year for pH was minimal (0.08) with pH being slightly higher in 2013 compared to 2012

(P<0.0001) (Tables 2.3 and A.8). After 2 weeks of storage, AM 02 in 2013 had the highest pH (4.0), while ‘Ison’ at date of harvest in 2012 had the lowest (3.1) (Fig. 2.5 and Table A.8).

When looking at the overall trends in the data, it is challenging to identify a clear linear change of week of storage or the difference in year and pH, leaving the variability within and among genotypes as a possible contributor of significance of the interaction (Fig. 2.5)

Titratable Acidity. If one looks at major trends in the data, it can be seen that overall

TA values appear to be lower for most genotypes in 2013 compared to 2012 (Fig. 2.6).

Among years of the study, TA means of 0.3% for 2013 and 0.5% in 2012 were found

(P<0.0001) (Table 2.3). When averaged across years and genotypes, minimal differences were found among weeks of storage, ranging from 0.43% for week 0, 0.43% for week 1,

0.43% for week 2, and 0.44% for week 3, with only week 3 being significantly different from the others (P=0.0218). Overall, AM 04 after 2 weeks of storage and ‘Supreme’ after 1

150 Table 2.3. F‐test significance from ANOVA for berry pH, titratable acidity, soluble solid content, and soluble solids to titratable acidity ratio during 3 weeks of storage. Highest‐ order interactions are italicized and shaded (2012 and 2013). Degrees pH Titratable Soluble SS/TAy of acidity solids freedom (%)z (%) Year 1 <0.0001 <0.0001 <0.0001 <0.0001 Week 3 0.3220 0.0218 0.0220 0.0075 Year*week 3 0.5716 <0.0001 0.6181 <0.0001 Genotype 16 <0.0001 <0.0001 <0.0001 <0.0001 Year*genotype 16 <0.0001 <0.0001 <0.0001 <0.0001 Week*genotype 48 0.0033 <0.0001 <0.0001 <0.0001 Year*week*genotype 48 0.0046 <0.0001 <0.0001 <0.0001 zTitratable acidity measured as tartaric acid. ySoluble solids to titratable acidity ratio.

Fig. 2.5. pH of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). Week 3.5 0 3.0 1 2012 2.5 2 2.0 3 1.5 1.0

0.5 Year 0.0 pH 3.5 3.0 2.5 2013 2.0 1.5 1.0 0.5 0.0 1 2 3 4 5 8 6 7 8 s t l it e a 0 0 0 1 2 2 u ry on it e r io F Is b w m a M M 0 M M M 1 M M M 2 M c es e mm re T A A A A A A A A A li J u p e N n S u D er S th u o S

Genotype

151 Fig. 2.6. Titratable acidity (%) of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). Week 0.6 0

0.5 2012 1 0.4 0.3 2 0.2 3

0.1 Year 0.0 0.6

0.5 2013 0.4

Titratable acidity (%) 0.3 0.2 0.1 0.0 1 2 3 4 5 8 6 7 8 s n t l it e a 0 0 0 1 2 2 2 u o it e r 0 1 io Fry Is b w m a M M M M M M M M M c es e mm re T A A A A A A A A A li J u p e N n S u D er S th u o S

Genotype

152 week of storage in 2012 had the highest TA% (0.61% and 0.62%, respectively), while AM

28 at date of harvest and after 1 week of storage in 2013 had the lowest (0.21% and 0.21%, respectively) (Fig. 2.6 and Table A.8). The variable response of genotypes for years and weeks, with the examples of AM 28 and ‘Ison’ highlighted, provides some insight why the interaction was significant (Fig. 2.6 and Table A.8).

Soluble Solids. When considering the major trends in the data, it appears that soluble solids values were lower for most genotypes in 2013 when compared to 2012 (Fig.

2.7 and Table A.8). The difference in mean soluble solids among years was 3.7%, with significant differences among years (P<0.0001) (Table A.8). Averaged across genotypes and years, the differences in mean soluble solids are minimal, ranging from 18.9% for week

0, 19.1% for week 1, 19.2% for week 2, and 19.3% for week 3, with only week 3 being significantly different than week 0 (P=0.0220) (Table A.8). Overall, ‘Summit’ at initial date of harvest in 2012 had the highest soluble solids (27.3%), while AM 28 in 2013 had the lowest soluble solids during storage (14.1% for week 0, 14.3 % for week 1, 14.2 % for week

2, and 14.6% for week 3) (Fig. 2.7 and Table A.8). The variable response of genotypes among years provides some insight why the interaction was significant (Fig. 2.7).

Soluble Solids to Titratable Acidity Ratio. Differing from soluble solids when considering the major trends in the data, it appears that SS/TA values were higher for most genotypes in 2013 when compared to 2012 (Fig. 2.8 and Table A.8). The difference of

SS/TA means was 11.6, which was significant (P<0.0001) (Table 2.3). The difference in

SS/TA means among weeks was minimal, ranging from 60.5 for week 0, 61.0 for week 1,

60.5 for week 2, and 57.2 for week 3, with only week 3 being significantly different from the others (P=0.0075). At initial date of harvest in 2012, ‘Ison’ had lowest SS/TA (35.25),

153 Fig. 2.7. Soluble solids (%) of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). Week 25 0 1

20 2012 2 15 3 10

5 Year 0 25

20 2013 Soluble solids (%) 15 10 5 0 1 2 3 4 5 8 6 7 8 s y n t l it e a 0 0 0 1 2 2 2 u r o it e r io F Is b w m a M M 0 M M M 1 M M M M c es e mm re T A A A A A A A A A li J u p e N n S u D er S th u o S

Genotype

154 Fig. 2.8. Soluble solids to titratable acidity ratio of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). Week 100 0 1

80 2012 2 60 3 40

20 Year 0 SS/TA 100

80 2013 60 40 20 0 1 2 3 4 5 8 6 7 8 s t l it e a 0 0 1 1 2 2 u on t e r io Fry Is bi w m m M 0 M M M 0 M M M 2 M M c es e m re Ta A A A A A A A A A li J u p e N n S u D er S th u o S

Genotype

155 while AM 03 at initial date of harvest and after 2 weeks of storage in 2012 had the highest

(112.1 and 112.6, respectively) (Fig. 2.8 and Table A.8). The variable response of genotypes for years, with the example of AM 03 highlighted and the variable response of genotype for weeks, with the example of AM 26 highlighted, offers insight why the interaction was significant.

Berry Color

The ANOVA F‐test indicated a significant three‐way interaction of year by week of storage by genotype for all of the berry color attributes measured of Chroma (P<0.0001), hue angle (P=0.0035), and L* (P=0.0001) (Table 2.4). The variability in genotypes and lack of variability among weeks and years offers some insight why the interaction for all berry color attributes was significant.

L*. Differing from the other color attributes of Chroma and hue angle, the dependent variable L* had significant differences among weeks of storage and year (P<0.0001). For L* the difference in year means, averaged across genotype and week was 4.3, with 2013 having higher values overall (Fig. 2.9). The difference in L* means among weeks ranged from 42.2 for week 0, 34.4 for week 1, 34.6 for week 2, and 33.5 for week 3, with only week

0 being significantly different from the other week means (P=0.0046). After 2 weeks of storage, AM 01 in 2013 had the highest L* value (105.5), while AM 28 at week 0 in 2013 had the lowest (22.9) (Fig. 2.9 and Table A.9). When looking at the major trends in the data, one can see that generally the bronze genotypes (AM 01, AM 03, AM 15, AM 26, ‘Fry’, ‘Ison’,

‘Summit’, and ‘Tara’) had higher L* values compared to the black genotypes (Fig. 2.11).

Additionally, one can see that often, but especially in 2013, L* values were higher at initial date of harvest, and then remained fairly constant during storage (Fig. 2.9 and Table A.9).

156 Table 2.4. F‐test significance from ANOVA for Chroma, hue angle, and L* values during 3 weeks of storage at 2 °C. Highest‐order interactions are italicized and shaded (2012 and 2013). Degrees Chroma Hue L* of angle freedom Year 1 0.3190 <0.0001 <0.0001 Week 3 0.0046 0.0807 <0.0001 Year*week 3 0.0006 <0.0207 <0.0001 Genotype 16 <0.0001 <0.0001 <0.0001 Year*genotype 16 <0.0001 <0.0001 <0.0001 Week*genotype 48 <0.0001 0.0099 <0.0001 Year*week*genotype 48 <0.0001 0.0035 0.0001

Fig. 2.9. L* values of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). Week 100 0 1

80 2012 2 60 3 40

20 Year

L* 0 100

80 2013 60 40 20 0 1 2 3 4 5 8 6 7 8 s t l it e a 0 0 1 1 2 2 u on t e r io Fry Is bi w m m M 0 M M M 0 M M M 2 M M c es e m re Ta A A A A A A A A A li J u p e N n S u D er S th u o S

Genotype

157 Hue angle. When looking at the major trends in the data it appeared that hue angle was greater in 2013 compared to 2012 (Fig. 2.10). The overall mean difference in hue angle was 37.4°, which was significantly higher in 2013 (P<0.0001)(Table 2.4). Additionally, it appeared that the bronze genotypes (AM 01, AM 03, AM 15, AM 26, ‘Fry’, ‘Summit’ and

‘Tara’) generally had lower hue angles compared to the black genotypes (Fig. 2.10). Overall, no significant differences were found in hue angle during storage (P=0.0807) (Table 2.3).

At date of harvest and throughout storage, ‘Supreme’ and ‘Nesbitt’ at date of harvest in

2012 had the largest hue angles (359.4°, 359.8°, 359.5°, 359.0°, and 359.7°, respectively), while ‘Summit’ at date of harvest in 2012 had the smallest hue angle (47.8°) (Fig. 2.10 and

Table A.9).

Chroma. For the dependent variable of Chroma, no significant differences in means were found among years, when averaged across weeks of storage and genotypes

(P=0.3190). Differences among weeks were minimal, ranging from 8.2 for week 0, 7.4 for week 1, 7.3 for week 2, and 7.1 for week 3, with only week 0 values being significantly different (P=0.0046). At week 0 in 2013, ‘Southern Jewel’ had the highest Chroma value

(23.1), while AM 27 after 1 week of storage in 2013 had the lowest (Fig. 2.11 and Table

A.9). When looking at the major trends in the data, it appears the bronze genotypes (AM 01,

AM 03, AM 15, AM 26, ‘Fry’, ‘Summit’ and ‘Tara’) generally had higher Chroma values, with the exception of ‘Southern Jewel’ at week 0 in 2013, (Fig. 2.11 and Table A.9).

Nutraceutical Content

The ANOVA F‐test indicated a significant two‐way interaction of year by genotype for the berry nutraceuticals total anthocyanins (P<0.0001), total flavonols (P<0.0001), resveratrol (P<0.0001), and ORAC (P<0.0001) (Table 2.5). Additionally, there were main

158 Fig. 2.10. Hue angle of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). Week 350 0 300 1 250 2012 2 200 3 150 100

50 Year 0 350 Hue angle 300

250 2013 200 150 100 50 0 1 2 3 4 5 8 6 7 8 s y t l it e a 0 0 0 1 1 2 2 u r on it e r io F Is b w m a M M 0 M M M M M M 2 M c es Je mm re T A A A A A A A A A li N u p e rn S u D e S th u o S

Genotype

159

Fig. 2.11. Chroma of muscadine genotypes stored at 2 °C for 3 weeks. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). Week 20 0 1 15 2012 2 10 3 5 Year 0

Chroma 20

15 2013 10 5 0 s t l t e 4 5 8 6 u n it e i a 02 03 0 1 1 2 28 o Fry so b w m M I s e mm e Tar AM 01AM AM AM A AM AM AM 27AM lici e J pr e N n Su u D er S th u So

Genotype

160

Trolox <0.0001 ORAC (µmol

equivalents/g)

0.2010 0.0043 <0.0001 <0.0001 <0.0001 (mg/100g)

Resveratrol

Total s, total ellagitannins,

0.0007 0.1538 <0.0001 rations during. Highest‐ phenolics

(mg/100g)

Total <0.0001 <0.0001 <0.0001 flavonols (mg/100g)

Total

0.0786 <0.0001 <0.0001

(mg/100g) ellagitannins

rom ANOVA for total anthocyanin

Total

<0.0001 (mg/100g)

anthocyanins

ORAC, total flavonols, total phenolics, and resveratrol concent order interactions are italicized and shaded (2012 and 2013). Table 2.5. F‐test significance f Genotype Year*genotype <0.0001 Year <0.0001

161 effects of genotype and year for total ellagitannins (P<0.0001) and total phenolics

(P<0.0001) (Table 2.5). The variability among genotypes and year offers some insight why the interactions were significant.

Total Anthocyanins. When examining the major trends in the data it appeared that

total anthocyanins were generally higher in 2012, compared to 2013 (Fig. 2.12). Averaged

across genotypes, the significant difference in mean year total anthocyanins was 211.3

mg/100 g (P<0.0001) (Table 2.5). Total anthocyanins were not detected in any of the bronze genotypes (AM 01, AM 03, AM 15, AM 26, ‘Fry’, ‘Ison’, ‘Summit’, and ‘Tara’), and concentrations varied among the black genotypes (Fig. 2.12 and Table A.10). The black genotypes AM 27 and AM 04 in 2012 had the highest levels of total anthocyanins (122.0 mg/100g and 109.3 mg/100g, respectively) (Fig. 2.12 and Table A.10).

Total Ellagitannins. Averaged across years, the genotype AM 03 had the highest

concentrations of total ellagitannins (12.5 mg/100 g), while AM 01, ‘Tara’, and ‘Supreme’

had the lowest (3.1, 3.5, and 3.6 mg/100 g, respectively) (Tables 2.6 and A.10). Among

years, significantly different total ellagitannin concentrations were found, 5.6 mg/100 g in

2012 and 8.3 mg/100 g in 2013 (Table 2.7). Generally, genotypes performed similarly

among years, with 2013 having greater overall concentrations, offering some insight why

the main effects were significant (Table A.10).

Total Flavonols. The dependent variable total flavonols was significantly higher in

2012, compared to 2013, when averaged across genotypes (P<0.0001). The mean

difference for total flavonol concentration was 7.4 mg/100 g among years. AM 03 and

‘Summit’ in 2012 had the highest total flavonol content (70.6 and 63.1 mg/100 g, respectively), while ‘Supreme’ and AM 28 in 2012 had the lowest (7.3 and 8.9 mg/100 g,

162 Fig. 2.12 Total anthocyanin concentrations of muscadine genotypes. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013).

140 Year 120 2012 100 2013 80 60

(mg/100 g) 40

Total anthocyanins 20 t 0 t y Fr Ison Tara AM 01 AM 02 AM 03 AM 04 AM 15 AM 18 AM 26 AM 27 AM 28 Nesbit Summi Delicious Supreme Southern Jewel Genotype

Table 2.6 Main effect of muscadine genotype means on total ellagitannin concentrations of the berry during three weeks of storage, averaged across years (2012 and 2013). Genotype Total ellagitannins (mg/100 g) AM 01 3.1 d z AM 02 8.0 abcd AM 03 12.5 a AM 04 9.6 abcd AM 15 8.2 abcd AM 18 4.1 cd AM 26 4.1 cd AM 27 7.3 abcd AM 28 4.2 bcd Delicious 10.2 abc Fry 8.5 abcd Ison 9.6 abcd Nesbitt 4.9 bcd Southern Jewel 5.9 abcd Summit 10.8 ab Supreme 3.6 cd Tara 3.5 d zMeans followed by the same letter are not significantly different at α= 0.05, separated by Tukey’s HSD.

163 Table 2.7. Main effect of year means on total ellagitannin concentrations of the muscadine, averaged across genotypes (2012 and 2013). Year Total ellagitannins (mg/100 g) 2012 5.6 2013 8.3 p value <0.0001

164 respectively) (Fig. 2.13 and Table A.10). When examining the major trends of the data, it appears that generally the bronze genotypes (AM 01, AM 03, AM 15, AM 26, ‘Fry’, ‘Summit’, and ‘Tara’) had higher levels of flavonols compared to the black genotypes (Fig. 2.13).

Total Phenolics. Total phenolics were highest in ‘Ison’ (670.8 mg/100 g), while AM

28 and AM 18 had the lowest levels (335.7 and 342.8 mg/100 g, respectively), averaged

across years (Table 2.8). Averaged across genotypes, the difference in total phenolics

among year means was 61.1 mg/100 g. Total phenolic concentrations were significantly

higher in 2012 (548.7 mg/100 g), compared to 2013 (487.6 mg/100 g) (P<0.0001) (Tables

2.9 and 2.5). Generally, genotypes performed similarly among years, with 2012 having

greater overall concentrations, offering some insight why the main effects were significant

(Table A.10).

Resveratrol. There was no significant difference in year means for the dependent

variable resveratrol concentration (P=0.2010); however, when considering the major

trends in the data it appeared that genotypes often performed differently among years.

The genotype AM 27 in 2012 had the highest resveratrol levels (16.7 mg/100 g), while AM

28 and ‘Southern Jewel’ in 2013 had the lowest (2.9 and 3.2 mg/100g) (Fig. 2.14 and Table

A.10). With the exceptions of AM 01, AM 27, ‘Ison’, ‘Summit’, and ‘Supreme’, the genotypes

generally performed similarly among years (Fig. 2.14 and Table A.10). Additionally, it

appeared that bronze and black genotypes performed similarly (Fig. 2.14).

ORAC. When considering the major trends in the data, it appears that ORAC was

generally greater in 2013 compared to 2012 (Fig. 2.15). The mean difference in years was

7.4 µmol Trolox equivalents/g, which was significant (P<0.0001). The cultivar Ison in 2012

and 2013 and ‘Southern Jewel’ in 2013 had the highest ORAC values (110.6, 115.5, and

165 Fig. 2.13. Total flavonol concentrations of muscadine genotypes. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013).

166

Table 2.8. Main effect of genotype on total phenolic concentrations of the muscadine berries, averaged across years (2012 and 2013). Genotype Total phenolics (mg/100 g) AM 01 566.5 abc z AM 02 522.9 abcde AM 03 638.6 ab AM 04 548.7 abcd AM 15 622.0 abc AM 18 342.8 f AM 26 380.2 def AM 27 558.8 abcd AM 28 335.7 f Delicious 612.6 abc Fry 555.6 abcd Ison 670.8 a Nesbitt 484.4 bcdef Southern Jewel 636.8 abc Summit 513.5 abcdef Supreme 360.2 ef Tara 457.8 cdef zMeans followed by the same letter are not significantly different at α= 0.05, separated by Tukey’s HSD.

Table 2.9. F‐test main effect significance for year on total phenolic concentrations averaged across muscadine genotypes (2012 and 2013). Year Total phenolics (mg/100 g) 2012 584.7 2013 487.6 p value 0.0007

167 Fig. 2.14. Resveratrol concentrations of muscadine genotypes. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). Year 20 2012 15 2013

10 esveratrol R (mg/100 g) 5 t 0 t y Fr Ison Tara AM 01 AM 02 AM 03 AM 04 AM 15 AM 18 AM 26 AM 27 AM 28 Nesbit Summi Delicious Supreme Southern Jewel Genotype

Fig. 2.15. Oxygen radical absorbance capacity (ORAC) of muscadine genotypes. Each standard error bar is constructed using 1 standard error from the mean (2012 and 2013). 120 Year 2012 100 2013

80

60

40 ORAC (µmol TE/g) 20

0 1 2 3 4 5 8 6 7 8 s y n t l it e a 0 0 0 0 1 1 2 2 2 u o it e r io Fr s w m m M c I e m re Ta AM AM AM AM AM AM AM AM A li esb J u p e N n S u D er S th u o S

Genotype

168 111.3 µmol Trolox equivalents/g, respectively), while AM 26 in 2012 had the lowest (34.2

µmol Trolox equivalents/g) (Fig. 2.15 and Table A.10). Averaged across years, ORAC ranged from 113.0 µmol Trolox equivalents/g (‘Ison’) to 45.8 µmol Trolox equivalents/g

(AM 26). With the exceptions of AM 01 and AM 15, individual genotypes performed

significantly different each year (Fig. 2.15), offering some insight why the interaction was significant.

Black verses Bronze Storage Performance

The variable response of genotypes among years for the dependent postharvest storage

variables of force, weight loss, percent unmarketable, and volume change led to interest in

evaluating the performance of genotypes averaged across muscadine genotype skin color,

either black or bronze. Significant differences among skin colors for percent unmarketable

(P=0.0012), percent volume change (P=0.0002), and force to penetrate the berry skin

(P=0.0180), were found, while there was no significant difference for percent weight loss

(P=0.4343). Averaged across year and week, the mean difference in percent unmarketable was 3.7%, with bronze genotypes having significantly higher percent of unmarketable berries. It was found that the black genotypes increased in volume during storage (‐1.3%), while bronze genotypes decreased in volume (3.5%), with significant differences in the means (Fig. 2.16). Overall, the black genotypes were found to have greater firmness (7.4 N), than the bronze genotypes (6.9 N) when averaged across year and week of storage (Fig.

2.16). When examining the major trends in the data, it can be determined that black

genotypes have generally better postharvest storage performance compared to bronze

genotypes (Fig. 2.16).

Correlations

169 Fig. 2.16. Storage performance of black and bronze muscadine genotypes, averaged across year (2012 and 2013) stored for 3 weeks at 2 °C. Each standard error bar is constructed using 1 standard error from the mean. Color 7 Black 6 Bronze 5 4 3 Force (N) 2 1 0 20

15 able (%) t e

k 10 5 Unmar 0 4 3 2 1 0 ‐1

Volume change (%) ‐2 2.0

1.5 Loss t 1.0 eigh W

% 0.5

0.0

170 Though some strong correlations did occur, the dependent variables measured were generally minimally correlated (Table A.11). Percent unmarketable berries was negatively correlated with force (r=‐0.74) (Table A.11). TA was negatively correlated with SS/TA (r=‐

0.89) (Table A.11). L* value was positively correlated with soluble solids (r=0.71) and

Chroma (r=0.72), and negatively correlated to hue (r=‐0.73) (Table A.11). Chroma was strongly negatively correlated with hue (r=‐0.93) and total anthocyanins (r=‐0.87), while hue angle was positively correlated with total anthocyanins (r=0.75) and negatively correlated with total flavonols (r=‐0.73) (Table A.11). Soluble solids were positively correlated with total flavonols (r=0.73) and L* (r=0.71) (Table A.11).

Discussion

Berry Storage Attributes

The average percent unmarketable berries and percent weight loss were generally similar in both years of the study; however, genotypes generally performed differently each year. Both percent weight loss and percent unmarketable increased during storage and varied among genotypes, and these results were consistent with other studies (Ballinger and Nesbitt, 1982a; James et al., 1997; James et al., 1999; Lutz, 1938; Silva et al., 1994;

Takeda et al., 1983). For example, James et al. (1997) found that after 2 weeks of storage,

‘Summit’ had 72% unmarketable berries and 4.9% weight loss, while ‘Fry’ had 6.2% unmarketable berries and 4.3% weight loss.

The postharvest berry diseases present were identified as black rot (Guignardia

[Phyllosticta] bidwellii [ampelicida] Ellis.), myrothecium leaf spot (Myrothecium sp./spp.), and botrytis fruit rot, (Botrytis sp./spp.). This is similar to reports by Lane (1978), Smit et al. (1971) and Takeda et al. (1983). Generally, berry diseases were not a major cause of

171 unmarketability until 3 weeks of storage. The primary factors involved in unmarketable

berries were browning (especially in bronze genotypes), leakage from torn or wet stem

scars, and shriveling, which is consistent with similar work reported by Perkins‐Veazie et

al. (2012). Occasionally an unknown species of fruit fly (Drosophila sp.) was present, but

only in berries with wet or torn stem scars and did not contribute to unmarketable berries.

Unmarketable Berries. A major cause of unmarketability in the bronze genotypes

was skin browning during storage likely caused by chilling injury (Fig. 2.17). The abiotic disorder of chilling injury is common in many horticultural crops, including bananas (Musa

× paradisiaca L.), citrus (Citrus spp.), sweet potatoes (Ipomoea batatas L.), and tomatoes

(Solanum lycopersicum L.) (Wang, 1990). Chilling injury can increase susceptibility to

decay through providing a favorable medium for the growth of pathogens (Wang, 1990).

The primary symptom of chilling injury identified in my study was brown discoloration of

the skin, pulp, and vascular strands of berries, as reported previously by Himelrick (2003)

and Wang ( 1990). Chilling injury has been reported in muscadines stored at or below 1.7

°C (Himelrick, 2003; Smittle 1990), but is uncommon in muscadines stored at 2 °C. It was

surprising to find possible symptoms of chilling injury on the muscadine berries in my

study, as they were stored above the previously reported threshold of 1.7 °C (Himelrick,

2003; Smittle 1990). This illustrates chilling injury susceptibility is likely genotype specific

and creates the potential for opportunities of improvement through tolerance to chilling

injury selection in the University of Arkansas muscadine breeding program. Other V.

vinifera table grapes were successfully stored at ‐1 °C without showing symptoms of

chilling injury (Burg, 2004).

172 Fig. 2.17. Muscadine genotype AM 01 at date of harvest (left) and after 3 weeks of storage (right), demonstrating the browning that occurred, likely caused by chilling injury (2013).

173 Weight Loss. Leakage and shriveling are common problems in muscadines during

storage and are managed with elimination prior to storage of berries with wet stem scars

and by maintaining high RH during storage (Perkins‐Veazie et al., 2012; Smit et al., 1971).

Leakage and shriveling did occur and was most prevalent after 3 weeks of storage (Fig.

2.18). It has been shown that use of plastic film packaging of other fruits (Citrus limon L. and Capsicum annuum L.) prevented water loss, resulting in less leakage and shriveling

(Ben‐Yehoshua et al., 1983). Conversely, it has been found that the use of polyethylene bags resulted in less weight loss of muscadines during storage, but did not prevent leakage and shriveling; due to the juice retention in the bags (Walker et al., 2001). It has also been found that edible coatings prevent weight loss during storage of fresh highbush blueberries

(Vaccinium corymbosum L.), but this is unknown on muscadines (Duan et al., 2011). Other table grapes (V. vinifera) experienced less leakage and shriveling during storage, as the berries were retained on the pedicel, and thus did not have stem scar damage potential during marketing unlike muscadines where berries are harvested singularly.

It was uncommon that a single genotype performed well in both years of storage.

After 3 weeks of storage, the genotypes with the least percent weight loss in 2013 were AM

28, ‘Nesbitt’, and ‘Southern Jewel’, while in 2012 AM 03, AM 27, ‘Delicious’ and ‘Tara’ had the least (Fig. 2.1 and Table A.7). The genotypes with the greatest percent weight loss after

3 weeks of storage in 2013 were AM 01, AM 03, and ‘Fry’, while in 2012 AM 04, AM 18, and

‘Nesbitt’ had the greatest weight loss (Fig. 2.1 and Table A.7). In 2013, the genotypes with the least percent unmarketable after 3 weeks of storage were AM 03, AM 04, AM 26, and

AM 28, while in 2012 AM 03, ‘Southern Jewel’, ‘Summit’, and ‘Supreme’ had the least percent unmarketable (Fig. 2.2 and Table A.7). The genotypes in 2013 with the highest

174

Fig. 2.18. Muscadine genotype AM 04 after 3 weeks of storage, demonstrating shriveling that occurred, likely caused by berry leakage. (2012)

175 percent unmarketable were AM 01, ‘Fry’, and ‘Tara’, while in 2012 the genotypes with the

highest percent unmarketable were AM 04, AM 26, and ‘Fry’ (Fig. 2.2 and Table A.7). This strongly shows the influence of environmental factors (rainfall and temperature) on

storage performance of muscadines (Table A.1). Additionally, this illustrates the

importance of testing multiple years for postharvest performance in a muscadine storage

protocol. Similar to our findings, Ballinger and Nesbitt (1982b) found ‘Nesbitt’ to have

acceptable postharvest storage quality, while James et al. (1999) found ‘Summit’ to have

the greatest percent decay and weight loss. I found ‘Summit’ to have intermediate quality during storage, which is similar to the findings of James et al. (1997).

Force. Force to penetrate muscadine skin has been shown to be a useful

characteristic to assess berry firmness and texture as well as berry maturity (Conner,

2013); however, use of force to determine storability of muscadine grapes has previously shown results with no clear trend (Silva et al., 1994; Walker et al., 2001). It has been shown that muscadines require a force up to 13.9 N to penetrate the skin at date of harvest, which is nearly twice that of V. vinifera cultivars (Conner, 2013). Similarly, I found that muscadine grapes require up 13.2 N to penetrate the skin at date of harvest (Fig. 2.3 and Table A.7).

Similar to the findings of Conner (2013), I determined that ‘Nesbitt’ was among the most firm cultivars. It was found that the berries stored in 2013 were generally firmer than the berries stored in 2012, further illustrating the significance of environmental influences

(temperature and rainfall) on storage quality (Figs. 2.1, 2.2, and 2.3, and Table A.7). Percent unmarketable berries was negatively correlated with force (r=‐0.74), potentially illustrating that berries requiring greater force to penetrate the berry skin store better as they were firmer (Table A.11) and likely one of the more important relationships among

176 variables measured to assist in evaluating a genotype’s postharvest storage potential.

Overall, I found that berry firmness decreased during storage, but was occasionally lowest

after 2 weeks of storage; similar results were reported by James et al. (1999), and this

could possibly be due water loss during storage resulting in an increase of firmness at week

3 (Fig. 2.3 and Table A.7). It was found that the genotypes requiring the most force to

penetrate the skin at date of harvest also required the most force to penetrate the skin after

3 weeks of storage (especially in 2013), and indicating force to be a good indicator of

potential storage performance (Fig. 2.3 and Table A.7).

Volume Change. Volume change of muscadine berries during storage is widely

unstudied. Unexpectedly, no relationship between storage time and volume change was

found. Conversely, it has been found that a decrease in size occurred in mangoes

(Mangifera indica L.) and citrus during storage due to water loss (Jha et al., 2006;

McCornack, 1975). The berries stored in 2012 generally increased in volume during storage while those stored in 2013 generally decreased (Fig. 2.4 and Table A.11). This may have been due to the substantial differences in growing seasons, as 2012 was extremely hot and dry, while 2013 had more moderate temperatures and more rainfall particularly near harvest (Table A.1). The lack of correlations found between volume change and the other independent variables was potentially due to the extreme variation of berry size within each genotype (Table A.11), limiting volume change as a useful variable in determining storability.

Berry Composition

Muscadines have been shown to increase in soluble solids and decrease in TA

during ripening (Carroll and Marcy, 1982; Flora and Lane, 1979; Johnson and Carroll, 1973;

177 Lanier and Morris, 1979; Peynard and Riberau‐Gayon, 1971; Walker et al., 2001). Unlike

many other fruits, muscadines are nonclimacteric, and do not continue to accumulate

sugars after harvest.

pH. The overall pH measured in 2013 was higher than in 2012 and ranged from 3.2

(‘Ison’) to 3.8 (AM 04) in 2012 and from 3.4 (AM 15) to 4.0 (AM 02) in 2013 (Fig. 2.5 and

Table A.8), this is contradictory to Jackson (1986), who found that high pH was often

associated with warmer temperatures during the growing season. I found that pH of

muscadines remained constant during storage (Fig. 2.5), which was consistent with several

other studies (James et al., 1997; James et al., 1999; Silva et al., 1994; Takeda et al., 1983;

Walker et al., 2001). Additionally, stable pH during storage was found in kiwifruit

(Actinidia deliciosa L.), mango, pineapple (Ananas comosus L.), strawberry (Fragaria x ananassa Weston.), and watermelon (Citrullus lanatus Thunb.), while pH was shown to decrease in muskmelon (Cucumis melo L.) (Gil et al., 2006). The pH reported for some V. vinifera cultivars was higher than I found in muscadines in 2012 and 2013 (Mascarenhas et al., 2012).

Titratable Acidity. It was found that percent TA stayed relatively constant during

storage (Fig 1.7 and Table A.3); this was consistent with the results of other studies (James

et al., 1997; James et al., 1999; Takeda et al., 1983; Walker et al., 2001) and is contradictory

to Silva et al., (1994), who reported that TA increased during storage, and Lutz (1938), who

reported that TA decreased during storage. Percent TA has also been shown to remain

constant during storage of muskmelon, kiwifruit, mango, pineapple, strawberry, and

watermelon (Gil et al., 2006). There was a strong negative correlation between TA and

SS/TA (r = 0.8978) (Table A.11), and a relationship was expected as TA is used in the

178 calculation of SS/TA. Titratable acidity was strongly affected by year and genotype, with

2012 generally having higher TA than 2013 (Fig. 2.6), which can be explained by the

warmer growing season in 2012. Jackson (1986) reported that warmer temperatures during the growing season resulted in higher acidity. In my study, the cultivar Supreme generally had the highest percent TA and AM 01 and ‘Southern Jewel’ had the lowest (Fig.

1.7 and Table A.3). TA was found to be lower than that of V. vinfera, reported to have ranged from 0.6 to 1.5% (Mascarenhas et al., 2012).

Soluble Solids. Probably due to the extremely hot and dry conditions, the percent

soluble solids was uncharacteristically higher in 2012 than in 2013 (Fig. 2.7). Overall

soluble solids remained constant during storage (Fig. 2.7and Table A.8); this is consistent

with most other studies (James et al., 1997; James et al., 1999; Takeda et al., 1983), but was

contradictory to Silva et al. (1994) and Walker et al. (2001), who reported decreased

soluble solids during storage, and Lutz (1938), who reported increased solids during

storage. Percent soluble solids has also been shown to remain constant during storage of

muskmelon, kiwifruit, mango, pineapple, strawberry, and watermelon (Gil et al., 2006).

Soluble solids in my study was found to be similar to that reported for V. vinifera

(Mascarenhas et al., 2012). There were positive correlations for both L* (r=0.71) and total

flavonols (r=0.73) with soluble solids (Table A.11). A possible explanation for these

correlations could be that higher soluble solids resulted in lighter berries (higher L*), and

increased flavonol concentrations, as soluble solids has been shown to be an indicator of

muscadine berry ripeness (Carroll and Marcy, 1982; Flora and Lane, 1979; Johnson and

Carroll, 1973; Lanier and Morris, 1979; Peynard and Riberau‐Gayun, 1971; Walker et al.,

2001). There have been strong influences of both year and genotype on soluble solids

179 reported, for example Threlfall et al. (2007) found that ‘Supreme’ had the lowest soluble solids of all genotypes evaluated, while I identified ‘Supreme’ as being midrange in soluble

solids (Fig. 2.7 and Table A.8). However, berry maturity at harvest could have played a

major role in these differences.

Soluble Solids to Titratable Acidity Ratio. The SS/TA was strongly influenced by year

and genotype (Fig. 2.8 and Table A.8). The overall average SS/TA measured in 2013 was

higher than in 2012 (Fig. 2.8 and Table A.8). The berries in 2012 had higher soluble solids

and TA, compared to the berries harvested in 2013 (Figs. 2.6 and 2.7), resulting in the

differences in SS/TA among years. It was found that SS/TA remained relatively constant

during storage, which is contradictory to Lutz (1938), who reported an increase of SS/TA

during storage time. The SS/TA reported here was higher than reported for some V.

vinfiera cultivars (Mascarenhas et al., 2012).

Berry Color

The affect of storage on the berry skin color attributes of L* value, Chroma, and hue

angle of fresh‐market muscadine grape berries is widely unstudied, while it is well studied

in juice and wine. The United States Department of Agriculture (USDA) currently has no

standards in place to grade muscadine berries for the color attributes of Chroma, hue, and

L*. The standards for skin color of muscadines state the berries should be well colored to

be considered marketable; for black and red cultivars 75% of the surface of the berry must

have characteristic color for the variety, while no color requirement exist for bronze

genotypes except that ‘Carlos’, ‘Fry’ or similar cultivars can show any amount of blush or

bronze color on the berry (USDA, 2006). Additionally the USDA states that black cultivar

colors can include reddish purple, purple, and black; red cultivar colors include light pink,

180 pink, red, dark red, and purple; and bronze cultivar colors include light green, straw,

amber, and bronze with allowance for an amount of blush or pink color that may also be

characteristic for certain cultivars (USDA, 2006).

L* value. Generally, L* values remained relatively constant during storage; however,

the genotypes in AM 04 in 2012, and AM 03, AM 15, AM 18, ‘Delicious’, and ‘Southern Jewel’

in 2013 had L* values decreased after the date of harvest, but remained relatively stable

during storage (Fig. 2.9 and Table A.9). For week 0, the average L* value for the bronze

genotypes was 55.4 and 35.4 for the black genotypes. The average L* values for the black

genotypes was 26.5 for week 1, 26.4 for week 2, and 26.5 for week 3, while the average L*

value for the bronze genotypes was 45.7 for week 1, 46.2 for week 2, and 43.4 for week 3, with no significant differences in any of the means; this illustrates the berries darkened after date of harvest, but remained relatively unchanged during storage. A decrease in L* value would represent a darkening of the berries during storage, as L* measures lightness from completely opaque (0) to completely transparent (100) (Walker et al. 2001). Silva et al. (1994) found that L* values increased during storage, though the differences were not

discernable by panelists. Conversely, I found that generally, L* values remained stable or

slightly decreased during storage (Fig. 2.9 and Table A.9). Similarly, Hernandez‐Herrero

and Frutos (2014) found that the L* values of model juice of grape, plum, and strawberry

stayed relatively constant during storage. Possibly due the more favorable growing season

(Table A.1), the L* values were higher in 2013, than in 2012 (Fig. 2.9 and Tables A.1 and

A.9). Overall, L* values were strongly affected by genotype and year, while less affected by

week of storage.

181 Hue Angle. I found the hue angles were on average 266.3° for the black and 89.5° for the bronze genotypes. Hue angle ranged from 69.9° (‘Summit’) to 301.1 (‘Supreme’), when averaged across year and week of storage. Conversely, Conner and MacLean (2013) found hue values that ranged from 1.5 to 91.8°, Threlfall et al. (2007) found values ranging from

53.4 to 98.6°, and Walker et al. (2001) found values that ranged from 76.5 to 237.7°. I found no clear relationship between week of storage and hue angle, with values remaining stable during storage (Fig. 2.10). Similarly, Hernandez‐Herrero and Frutos (2014) found that the hue angles of model juice of grape, plum, and strawberry stayed relatively constant during storage. There was a negative correlation between hue angle and L* value (r=‐

0.7951) (Table A.11), showing that as L* increased (berries became lighter), hue angle decreased. Overall, hue values were greater in 2013 than in 2012 (Fig. 2.10 and Table A.9), which might be due to the milder growing season in 2013, likely resulting in less berry sunburn (Table A.1). The average hue angle for the black genotypes was 266.3°, which falls between blue and green coloration value for hue, while the average hue angle was 89.5° for bronze genotypes, which is approximately yellow for hue.

Chroma. I found that Chroma was influenced by storage time, year, and genotype.

Walker et al. (2001) found that Chroma of the bronze cultivar Fry ranged from 12.1 to 14.2 based on maturity level, this is comparable to the findings of my study, with Chroma for

‘Fry’ ranging from 13.1 to 16.3. Conner and MacLean (2013) found Chroma values ranging from 2.4 to 22.8 and Threlfall et al. (2007) found Chroma values ranging from 8.0 to 52.8 on four black cultivars (‘Black Beauty’, ‘Ison’, ‘Nesbitt’, and ‘Supreme’) and 2 bronze cultivars (‘Granny Val’ and ‘Summit’), with the bronze genotypes generally having lower

Chroma values, both of which are consistent with my findings. I found Chroma generally

182 had no clear pattern during storage (Fig. 2.11 and Table A.9), similarly Hernandez‐Herrero

and Frutos (2014) found that the Chroma values of model juices of grape (V. vinifera), plum

(Prunus spp.), and strawberry stayed relatively constant during storage. Chroma values were generally highest for bronze genotypes, with the exception of the black ‘Southern

Jewel’ at initial date of harvest in 2013 (Fig. 2.11 and Table A.9), which was unusually high.

These results are similar to the findings of Conner and Maclean (2012). There was a strong negative correlation between Chroma and hue angle (r=‐0.93) (Table A.11). Overall,

Chroma values were higher in 2012 than in 2013 (Fig 2.11 and Table A.9), which could possibly be explained by the milder growing season in 2013 (Table A.1). I found that averaged across year and storage, Chroma for the bronze genotypes was 12.9, while the black genotypes had an average Chroma of 3.7, showing on average the bronze genotypes were less grey than the black genotypes.

Nutraceutical Content

Total Anthocyanins. Total anthocyanin concentrations found were similar to those

previously reported (Ballinger et al., 1973; Brown, 1940; Conner and MacLean 2013; Lee et

al., 2005; Lee and Talcott, 2004; Goldy et al., 1987; Marshall et al., 2012; Pastrana‐Bonilla et

al., 2003; Sandhu and Gu 2010; Striegler et al., 2005; Stringer et al., 2009; Threlfall et al.,

2007). As expected, anthocyanins were not detected in any of the bronze genotypes in

either year of the study (Fig. 2.12 and Table A.10). A negative correlation with total anthocyanins and Chroma (r=‐0.8715) and a positive correlation with hue angle (r=0.7507) was found (Table A.11), showing that lower Chroma values and greater hue angles were related to higher total anthocyanins, which is not surprising as bronze genotypes generally had higher Chroma values and lower hue angles and no anthocyanins (Figs. 2.9, 2.10, and

183 2.12 and Tables A.9 and A.10). I found greater total anthocyanin concentrations than those

reported for V. vinifera grapes (Hernandez‐Herrero and Frutos, 2014). Total anthocyanins found were greater than or comparable to those reported for blackberry (Rubus sp. L.),

highbush blueberry, red raspberry (Rubus ideaus. L.), and strawberry (Cordenunsi et al.,

2002; Ehlenfeldt and Prior et al., 2001; Maatta‐Riihinen et al., 2004; Siriwoharn et al.,

2004). Anthocyanin concentrations were generally higher in 2012 than in 2013 (Fig. 2.12).

The differences in total anthocyanins among years may have been due to higher

temperature and greater sun exposure, and therefore, greater color development in the

2012 growing season (Table A.1). The cultivar Nesbitt was identified as having among the

highest levels of anthocyanins by Lee and Talcott (2004), while I found ‘Nesbitt’ to have

some of the lowest levels among black genotypes. I identified ‘Supreme’ as having among

the lowest levels of total anthocyanins, which was also found by Threlfall et al. (2007) and

Striegler et al. (2005). It is unclear then why my levels agree with some and not other

studies, but this could be due to differences in environment, maturity, cultural

management, or methodology of measurement (Awad et al., 2001).

Total Ellagitannin. Concentrations of total ellagitannins were found to be lower than

those reported for muscadines by Marshall et al. (2012) and Lee and Talcott (2004), but

similar to those reported by Boyle and Hsu (1990), Lee et al. (2005), Pastrana‐Bonilla et al.

(2003), Stringer et al. (2009), and Talcott and Lee (2002). Ellagitannin concentrations were

found to be higher than those reported for strawberry (Cordenunsi et al., 2002), but lower

than those reported in blackberry and raspberry (Maatta‐Riihinen et al., 2004; Siriwoharn

et al., 2004). Additionally, ellagitannins were absent in other V. vinifera and V. lubrusca

grapes (Marshall et al., 2012). Total ellagitannin concentrations were very genotype

184 specific, and no clear relationship was identified between color (bronze verses black) and ellagitannins (Table 2.6). Overall, total ellagitannin concentrations were greater in 2013 compared to 2012 (Table 2.7), which is similar to the findings of Tharayil et al. (2011), who found that warmer and drier growing conditions were associated with lower levels of ellagitannins in the leaves of Red Maple (Acer rubrum L.) compared to wetter growing conditions.

Total Flavonols. Total flavonol concentrations found were lower than those

reported by Marshall et al. (2012) and Talcott and Lee (2002). Overall, total flavonols were

higher in 2013 compared to 2012 (Fig. 2.13 and Table A.11). Generally, results showed that genotypes differed among years, with the exceptions of AM 15 and ‘Summit’, which had among the highest total flavonol concentrations both years of my study (Fig. 2.13). The bronze genotypes were found to be generally higher in total flavonols than the darker genotypes (Fig. 2.13), this potentially due to the presence of the flavonol myricetin in the bronze genotypes (Marshall et al., 2012). A positive correlation with total flavonols and soluble solids (r=0.73) and a negative correlation with hue angle and total flavonols (r=‐

0.73) occurred (Table A.11). These correlations possibly illustrate that riper berries have higher flavonol concentrations, as soluble solids has been shown to be an indicator of muscadine berry ripeness (Carroll and Marcy, 1982; Flora and Lane, 1979; Johnson and

Carroll, 1973; Lanier and Morris, 1979; Peynard and Riberau‐Gayon, 1971; Walker et al.,

2001), and berries with lower hue angles had higher total flavonols, which is supported by

the data as the bronze genotypes generally had higher total flavonol levels and lower hue

angles (Figs. 2.10 and 2.13). I found total flavonols at higher levels than those found in

strawberry, but lower than those found in blackcurrant, chokeberry (Aronia mitschurinii

185 A.K.Skvortsov & Maitul), cranberry (Vaccinium oxycoccos L.), and red raspberry

(Cordenunsi et al., 2002; Hakkinen et al., 1999; Maatta‐Riihinen et al., 2004).

Total Phenolics. Total phenolic concentrations found were similar to those

previously reported for muscadines (Lee et al., 2005; Lee and Talcott, 2004; Marshall et al.,

2012; Pastrana‐Bonilla et al., 2003; Striegler et al., 2005; Stringer et al., 2009; Talcott and

Lee, 2002; Threlfall et al., 2007). Similar phenolic concentrations were reported in

blackberry (Siriwohorn et al., 2004), while lower phenolic concentrations were found in

highbush blueberry and strawberry (Cordenunsi et al., 2002; Ehlenfeldt and Prior et al.,

2001). Total phenolic concentrations were significantly higher in 2012, compared to 2013

(Table 2.9), likely due to the added stress on the vines from hot and dry growing conditions, and the plants responding with increased phenolic production. I found ‘Summit’ to have among the highest levels of total phenolics, which was similar to the findings of

Threlfall et al. (2007). Additionally, I found the cultivar Supreme to have the overall lowest total phenolics (average value across years) among the genotypes measured, while

Striegler et al. (2005), found ‘Supreme’ to have among the highest total phenolic level of

6072 mg/kg (607.2 mg/100 g) in their study. The lower level found in my study for

‘Supreme’ could be due to environment, maturity, or cultural management.

Resveratrol. Trans‐resveratrol concentrations found were similar to those

previously reported in muscadines (Ector et al., 1996; Magee et al., 2002; Marshall et al.,

2012; Pastrana‐Bonilla et al., 2003; Stringer et al., 2009). As I did, Marshall et al. (2012)

found resveratrol in nearly every genotype evaluated. There was no significant difference

among year means of resveratrol concentrations, though variation did occur among

genotypes each year (Fig. 2.15). Highlighted by the large error bars associated with the

186 genotypes AM 01, ‘Ison’, ‘Nesbitt’, ‘Summit’ and ‘Supreme’, resveratrol concentrations

varied within individual genotypes as well, due to the variation among replications within

some genotypes (Fig. 2.14). No clear relationship between berry color and resveratrol

concentrations were found, conversely Ector et al. (1996) found resveratrol to be greater in

black genotypes. Magee et al. (2002) found the bronze ‘Summit’ to have among the highest

levels of resveratrol in a group of both black and bronze genotypes, which is similar to the

findings of my study. Trans‐resveratrol concentrations were found to be equivalent to

those in V. vinifera (Vincenzi et al., 2013), which was unexpected. Ector et al. (1996) found

that resveratrol concentrations were higher than those reported for V. vinvifera grapes. The

different resveratrol concentrations found in my study could be due to environment,

maturity, or cultural management, as resveratrol is produced in response to environmental

factors during the growing season (Marshall et al., 2012).

ORAC. Oxygen radical absorbance capacity is widely accepted as being a good

estimation of antioxidant capacity of fruits, although its significance is often questioned, as

it does not accurately represent the bioactivity of the antioxidants in the human body. The

ORAC values I found were similar to those previously reported by Sandhu and Gu (2010)

and Talcott and Lee (2002), but were considerably higher than those reported by Lee et al.

(2005), Striegler et al. (2005), and Threlfall et al. (2007). I found ORAC values to be greater than or comparable to those found in apple (Malus domestica Borkh.), blackberry, highbush blueberry, plum, orange, red table grapes, strawberry, and white table grapes (Ehlenfeldt and Prior et al., 2001; Siriwoharn et al., 2004; Wang et al., 1996; Wu et al., 2004). ORAC levels were found to be higher in 2013 compared to 2012, which could possibly be due to the extremely hot and dry growing season in 2012 (Table A.1), although it was difficult to

187 identify this clearly among the genotypes (Fig 2.15). The cultivar Ison had the highest ORAC

values both years of the study, while ‘Supreme’ and ‘Tara’ had among the lowest both years

(Fig. 2.15 and Table A.10). Conversely, Threlfall et al. (2007) reported ‘Nesbitt’ having among the lowest ORAC levels, while Striegler et al. (2005) identified ‘Supreme’ as having

among the highest. It is important to note the vast variation in ORAC levels among

genotypes (Fig. 2.15), which illustrates ORAC as having potential as a character to be

selected for in muscadine breeding programs.

Black Verses Bronze Storage Performance

Throughout this experiment, the bronze genotypes had higher levels of

unmarketability compared to the black genotypes, which was often due to browning and

discoloration that occurred during storage, though individual causes of unmarketable

berries were not recorded. It was found that overall the black genotypes had better

storage performance than the bronze genotypes (Fig. 2.16). As expected, significant

differences were seen between bronze and black genotypes for percent unmarketable, but

surprisingly differences were also identified in percent volume change and force to

penetrate the berry skin. However, there were no differences found in percent weight loss

between berry color (Fig. 2.16). In addition to browning, the bronze genotypes often

showed greater signs of decay. This could potentially have caused the differences found in

force and volume change, though no data were collected specifically on the effects of decay

on these variables. It is hypothesized that the black genotypes could have also had

symptoms of browning or discoloration during storage, but these symptoms were not

visible due to the darkness of the berry skin.

Storage Protocol

188 A major component of this study was to determine the important parameters of

storage performance of muscadine genotypes, and in so doing to develop a storage protocol

for the University of Arkansas muscadine breeding program. Overall, both percent

unmarketable and percent weight loss increased during storage, showing importance as

storage parameters. Force to penetrate the berry skin generally decreased during storage,

also showing potential as an important postharvest storage parameter, particularly since

some genotypes had significantly less loss in force during storage. Percent change in berry

volume showed no clear pattern during storage, probably due to the variation in individual

berries within each genotype limiting the usefulness of berry volume as a storage

measurement. Composition parameters TA, pH, soluble solids, and SS/TA remained

relatively constant during storage, therefore are not important postharvest storage

measurements to routinely measure in evaluating storage potential. Though no clear correlations were identified in this study, it has been shown that soluble solids can be useful in determining maturity, which has been shown to be related with storage performance (Ballinger and McClure, 1983; Carroll and Marcy, 1982). The berry color

measurements, Chroma and hue angle, generally showed no clear pattern during storage,

while L* showed a sharp decrease after date of harvest and then remained relatively constant during storage. Therefore, it is potentially valuable to determine L* value at date of harvest and again after storage is complete to evaluate color change during storage. The retention of nutraceutical content (total phenolics, total anthocyanins, total ellagitannins, total flavonols, and resveratrol) and antioxidant capacity (ORAC) was not evaluated in this study.

189 Among the sources of variation in my study, genotype was the most common source with differences among most dependent variable means. This is a major finding, in that in differentiating the potential value of breeding selections, particularly for postharvest storage potential, adequate variation for a characteristic or trait is needed. It appears there is substantial variation among genotypes in the program for most variables to select for those with improved or superior values. Further, the differences among years for many dependent variables indicated the importance of multi‐year evaluations of breeding selections for storage potential. Since this study was conducted early in the muscadine breeding program, the findings reported here, including the most critical variables to measure, should lead to improved precision in identifying and releasing improved cultivars for fresh‐market production with enhanced postharvest potential.

Literature cited

Awad, M.A., P.S. Wagenmakers, and A. de Jager. 2001. Effects of light on flavonoid and chlorogenic acid levels in the skin of ‘Jonogold’ apples. Scientia Hort. 88:289‐298.

Ballinger, W.E., E.P. Maness, and W.B. Nesbitt. 1973. Anthocyanins of black grapes of 10 clones of Vitis rotundifolia, Michx. J. Food Sci. 38:909‐910.

Ballinger, W.E. and W. F. McClure. 1983. The effect of ripeness on storage quality of ‘Carlos’ muscadine grapes. Scientia Hort. 18:241‐245.

Ballinger, W.E. and W.B. Nesbitt. 1982a. Postharvest decay of muscadine grapes (Carlos) in relation to storage temperature, time, and stem condition. Amer J. Enol. Viticult. 33:173‐ 175.

Ballinger, W.E. and W.B. Nesbitt. 1982b. Quality of muscadine grapes after storage with sulfur dioxide generators. J. Amer. Soc. Hort. Sci. 107:827‐830.

Banini, A.E., L.C. Boyd, J.C. Allen, H.G. Allen, and D.L. Sauls. 2006. Muscadine grape products intake, diet and blood constituents of non‐diabetic and type 2 diabetic subjects. Nutrition 22:1137‐1145.

190 Ben‐Yehoshua, S., B. Shapiro, Z.E. Chen, and S. Lurie. 1983. Mode of action of plastic film in extending life of lemon and bell pepper fruit by alleviation of water stress. Plant Physiol. 73:87‐93.

Boyle, J.A., and L. Hsu. 1990. Identification and quantitation of ellagic acid in muscadine grape juice. Amer. J. Enol. Viticult. 41:43‐47.

Bralley, E.E., D.K. Hartle, P. Greenspan, and J.L. Hargrove. 2007. Topical anti‐inflammatory activities of Vitis rotundifolia (muscadine grape) extracts in the tetradecanoylphorbol acetate model of ear inflammation. J. Medicinal Food 10:636‐642.

Brown, W.L. 1940. The anthocyanin pigment of the Hunt muscadine grape. J. Amer. Chem. Soc. 62:2808‐2810.

Burg, S.P. 2004. Postharvest physiology and hypobaric storage of fresh produce. 1st ed. CABI Publishing, Wallingford, Oxfordshire, U.K.

Carroll, D.E. and J.E. Marcy. 1982. Chemical and physical changes during maturation of muscadine grapes (Vitis rotundifolia). Amer. J. Enol. Viticul. 33:168‐172.

Cho, M.J., L.R. Howard, R.L. Prior, and J.R. Clark. 2004. Flavonoid glycosides and antioxidant capacity of various blackberry, blueberry, and red grape genotypes determined by high‐ performance liquid chromatography/mass spectrometry. J. Sci. Food Agri. 84:1771‐1782.

Cho, M.J., L.R. Howard, R.L. Prior, and J.R. Clark. 2005. Flavonol glycosides and antioxidant capacity of various blackberry and blueberry genotypes determined by high‐performance liquid chromatography/mass spectrometry. J. Sci. Food. Agr. 85:2149‐2158.

Conner, P.J. 2009. A century of muscadine Grape (Vitis rotundifolia Michx.) breeding at the University of Georgia. Acta Hort. 827:481‐484.

Conner, P.J. 2013. Instrumental texture analysis of muscadine grape germplasm. HortScience 48:1130‐1134.

Conner, P.J., and D. Maclean. 2012. Evaluation of muscadine grape genotypes for storage ability. HortScience 47:S386. (abstr.).

Conner, P.J., and D. MacLean. 2013. Fruit anthocyanin profile and berry color of muscadine grape cultivars and Muscadinia germplasm. HortScience 48:1235‐1240.

Cordenunsi, B.R., J.R. Oliveira do Nascimento, M.I. Genovese, and F.M. Lajolo. 2002. Influence of cultivar on quality parameters and chemical composition of strawberry fruits grown in Brazil. J. Agr. Food. Chem. 50:2581‐2586.

191 Duan, J., R. Wu, B.C. Strik, and Y. Zhao. 2011. Effect of edible coatings on the quality of fresh blueberries (Duke and Elliot) under commercial storage conditions. Postharvest Biol. Technol. 59:71‐79.

Ector, B.J. 2001. Compositional and nutritional characteristics, p. 341‐367. In: F.M. Basiouny and D.G. Himelrick, (eds.). Muscadine Grapes. ASHS Press, Alexandria, VA.

Ector, B.J., J.B. Magee, C.P. Hegwood, and M.J. Coign. 1996. Resveratrol concentration in muscadine berries, juice, pomace, purees, seeds, and wines. Amer. J. Enol. Viticult. 1:57‐62.

Ehlenfeldt, M.K., and R.L. Prior. 2001. Oxygen radical absorbance capacity (ORAC) and phenolic and anthocyanin concentrations in fruit and leaf tissues of highbush blueberry. J. Agr. Food Chem. 49:2222‐2227.

Flora, L.F. and R.P. Lane. 1979. Effects of ripeness and harvest date on several physical and composition factors of ‘Cowart’ muscadine grapes. Amer. J. Enol. Viticult. 30:241‐246.

Gil, M.I., E. Aguayo, and A.A. Kader. 2006. Quality changes and nutrient retention in fresh‐ cut versus whole fruits during storage. J. Agr. Food. Chem. 54:4284‐4296.

God, J.M., P. Tate, and L.L. Larcom. 2007. Anticancer effects of four varieties of muscadine grape. J. Medicinal Food 10:54‐59.

Goldy, R.G., W. E. Ballinger, E. P. Maness, and W. H. Swallow. 1987. Anthocyanin content of fruit, stem, tendril, leaf, and leaf petiole in muscadine grape. J. Amer. Soc. Hort. Sci. 112:880‐882.

Greenspan, P., A. Ghaffar, J.L. Hargrove, D.K. Hartle, E.P. Mayer, J.D. Bauer, S.H. Pollock, and J.D. Gangemi. 2005. Anti‐inflammatory properties of the muscadine grape (Vitis rotundifolia). J. Agr. Food Chem. 53:8481‐8484.

Hakkinen, S.H., S.O. Karenlampi, L.M. Heinonen, H.M. Mykkanen, and A.R. Torronen. 1999. Content of edible flavonols quercetin, myricetin, and kaempferol in 25 edible berries. J. Agr. Food. Chem. 47:2274‐2279.

Hager, T.J., L.R. Howard, R. Liyanage, J.O. Lay, and R.L. Prior. 2008. Ellagitannin composition of blackberry as determined by HPLC‐ESI‐MS and MALD‐TOF‐MS. J. Agr. Food. Chem. 56: 661‐669.

Hernandez‐Herrero, J.A., and M.J. Frutos. 2014. Colour and antioxidant capacity stability in grape, strawberry, and plum peel model juices at different pHs and temperatures. Food. Chem. http://dx.doi.org/10.1016/j.foodchem.2014.01.007.

Himelrick, D.G. 2003. Handling, storage, and postharvest physiology of muscadine grapes. Small Fruits Rev. 2:(4):45‐62.

192 Huang, Z., R.D. Pace, P. Williams, and B. Wang. 2009. Identification of anthocyanins in muscadine grapes with HPLC‐ESI‐MS. Food Sci. Tech. 42:819‐824.

Jackson D.I. 1986. Factors affecting soluble solids, acid, pH, and color in grapes. Amer. J. Enol. Viticult. 37:179‐183.

James, J., O. Lamikanra, G. Dixon, S. Leong, J.R. Morris, G. Main, and J. Silva. 1997. Shelf‐life study of muscadine grapes for the fresh fruit market. Proc. Fla. State Hort. Soc. 110:234‐ 237.

James, J., O. Lamikanra, J.R. Morris, G. Main, T. Walker, and J. Silva. 1999. Interstate shipment and storage of fresh muscadine grapes. J. Food Quality. 22:605‐617.

Jha, S.N., A.R.P Kingsly, and S. Chopra. 2006. Physical and mechanical properties of mango during growth and storage for determination of maturity. J. Food. Eng. 72:73‐76.

Johnson, L.A. and D.E. Carroll. 1973. Organic acid and sugar contents of Scuppernong grapes during ripening. J. Food. Sci. 38:21‐24.

Lane, R.P. 1978. Effect of vineyard fungicide treatments on the shelf life of muscadine grapes. Georgia Agr. Res. 19:12‐14.

Lane, R.P., and L.F. Flora. 1980. Some factors influencing storage of muscadine grapes. HortScience 15:273 (abstr.).

Lanier, M.R. and J.R. Morris. 1979. Evaluation of density separation for defining fruit maturities and maturation rates of once‐over harvested muscadine grapes. J. Amer. Soc. Hort. Sci. 104:249‐252.

Lee, J‐H. and S.T. Talcott. 2004. Fruit maturity and juice extraction influences ellagic acid derivatives and other antioxidant polyphenolics in muscadine grapes. J. Agr. Food Chem. 52:361‐366.

Lee, J‐H., J.V. Johnson, and S.T. Talcott. 2005. Identification of ellagic acid conjugates and other polyphenolics in muscadine grapes by HPLC‐ESI‐MS. J. Agr. Food Chem. 53:6003‐ 6010.

Lutz, J.M. 1938. Factors influencing the quality of American grapes in storage. U. S. Dept. Agr. Tech. Bul. 606:1‐27.

Maatta‐Riihinen, K. R., A. Kamal‐Eldin, and A.R. Torronen. 2004. Identification and quantification of phenolic compounds in berries of Fragaria and Rubus species (Family Rosaceae). J. Agr. Food Chem. 52:6178–6187.

MacLean, D., P.J. Conner, J. Paulk, and L. Grant. 2009. Postharvest control of decay organisms. The Southern Region Small Fruit Consortium. Prog. Rpt. 2009‐12.

193

Magee, J.B., B.J. Smith, and A. Rimando. 2002. Resveratrol content of muscadine berries is affected by disease control spray program. HortScience 37:358‐361.

Marshall, D.A., S.J. Stringer, and J.D. Spiers. 2012. Stilbene, ellagic acid, flavonol, and phenolic content of muscadine grape (Vitis rotundifolia Michx.) cultivars. Pharmaceutical Crops 3:69‐77.

Mascarenhas, R.d.J., S.d.M. Silva, M.A.C.d. Lima, R.M.N. Mendonca, and H.J. Holschuh. 2012. Characterization of maturity and quality of Brazilian apirenic grapes in the Sao Francisco river Valley. Ciência e Tecnologia de Alimentos. 32:26‐33.

McCornack, A.A. 1975. Postharvest weight loss of Florida citrus fruits. Proc. Fla. State Hort Soc. 88:333‐335.

Morris, J.R. 1980. Handling and marketing of muscadine grapes. FruitSouth, 4(2):12‐14.

Morris, J.R., O.L. Oswald, G.L. Main, J.N. Moore, and J.R. Clark. 1992. Storage of new seedless grape cultivar with sulfur dioxide generators. Amer. J. Enol. Viticult. 43:230‐232.

Musingo, M.N., S.F. O'Keefe, O. Lamikanra, C.A. Sims, and R.P. Bates. 2001. Changes in ellagic acid and other phenols in muscadine grape (Vitis rotundifolia) juices and wines during storage. Amer. J. Enol. Viticult. 52:109‐114.

Pastrana‐Bonilla, E., C.C. Akoh, S. Sellappan, and G. Krewer. 2003. Phenolic content and antioxidant capacity of muscadine grapes. J. Agr. Food Chem. 51:5497‐5503.

Perkins‐Veazie, P., S. Spayd, B. Cline, and C. Fisk. 2012. Handling and marketing guide for fresh market muscadine grapes. SFRC‐E03: 1‐12.

Peynard, E. and P. Ribercau‐Gayun. 1971. The grape, p. 172‐206. In: A.D. Hulme (ed.). The bio‐chemistry of fruits and Their Products. Vol. 2. Academic Press, Waltham, MA.

Prior, R.L., H. Hoang, L. Gu, X. Wu, M. Bacchiocca, L. Howard, M. Hampschwoodill, D. Haung, B. Ou, and R. Jacob. 2003. Assays for hydrophilic and lipophilic antioxidant capacity [oxygen radical absorbance capacity (ORACF1)] of plasma and other biological and food samples. J. Agr. Food Chem. 51:3272‐3279.

Prior, RL, S.A. Lazarus, G. Cao, H. Muccitelli, and J.F. Hammerstone. 2001. Identification of procyanidins and anthocyanins in blueberries and cranberries (Vaccinium Spp) using high‐ performance liquid chromatography/mass spectrometry. J Agr. Food Chem. 49:1270–1276.

Sandhu, A.K., and L.W. Gu. 2010. Antioxidant capacity, phenolic content, and profiling of phenolic compounds in the seeds, skin, and pulp of Vitis rotundifolia (muscadine grapes) as determined by HPLC‐DAD‐ESI‐MSn. J. Agr. Food Chem. 58:4681‐4692.

194 Silva, J.L., E. Marroquin, C.P. Hegwood, G.R. Silva, and J.O. Garner Jr. 1994. Quality changes in muscadines for table grapes during refrigerated storage in various packaging systems. Proc. Viticult. Sci. Symp. Fla. A and M. Univ. 17:65‐72.

Siriwoharn, T., R.E. Wrolstad, C.E. Finn, and C.B. Pereira. 2004. Influence of cultivar, maturity, and sampling on blackberry (Rubus L. hybrids) anthocyanins, polyphenolics, and antioxidant properties. J. Agr. Food Chem. 52:8021–8030.

Slinkard, K. and V.L. Singleton. 1977. Total phenol analysis: automation and comparison with manual methods. Amer. J. Enol. Viticult. 28:49‐55.

Smit, C.J.B., H.L. Cancel, and T.O.M. Nakayama. 1971. Refrigerated storage of muscadine grapes. Amer. J. Enol. Viticult. 22:227‐230.

Smittle, D.A. 1990. Requirements for commercial CA storage of muscadine grapes. Proc. Viticult. Sci. Symp. Fla. A and M. Univ. 13:140‐146.

Striegler, R.K., P.M. Carter, J.R. Morris, J.R. Clark, R.T. Threlfall, and L.R. Howard. 2005. Yield, quality, and nutraceutical potential of selected muscadine cultivars grown in southwestern Arkansas. HortTechnology 15:276‐284.

Stringer, S.J., D.A. Marshall, and P. Perkins‐Veazie. 2009. Nutraceutical compound concentrations of muscadine (Vitis rotundifolia Michx.) grape cultivars and breeding lines. Acta. Hort. 841:553‐556.

Takeda, F., M. Starnes Saunders, C.F. Savoy, and T.T. Hatton. 1983. Storageability of muscadines for use as fresh fruit. Proc. Viticult. Sci. Symp. Fla. A and M. Univ. 3:31‐33.

Takeda, F., M. Starnes Saunders, J.A. Saunders, and T.T. Hatton. 1982. Effects of prestorage treatment and storage temperature on incidence of decay and chemical composition in muscadine grape. Proc. Fla. State. Hort. Soc. 95:109‐112.

Talcott, S.T., and J‐H. Lee. 2002. Ellagic acid and flavonoid antioxidant content of muscadine wine and juice. J. Agr. Food Chem. 50:3186‐3192.

Tharayil, N., V. Suseela, D.J. Triebwasser, C.M. Preston, P.D. Gerard, and J.S. Dukes. 2011. Changes in the structural composition and reactivity of Acer rubrum leaf litter tannins exposed to warming and altered precipitation: climatic stress‐induced tannins are more reactive. New Phytol. 191:132‐145.

Threlfall, R.T., J.R. Morris, J.F. Meullenet, and R.K. Striegler. 2007. Sensory characteristics, composition, and nutraceutical content of juice from Vitis rotundifolia (muscadine) cultivars. Amer. J. Enol. Viticult. 58:268‐273.

195 Threlfall, R.T., J.R. Morris, L.R. Howard, C.R. Brownmiller, and T.L. Walker. 2005. Pressing effects on yield, quality, and nutraceutical content of juice, seeds, and skins from ‘Black Beauty’ and ‘Sunbelt’ grapes. J. Food Sci. 70:167‐171.

U.S. Department of Agriculture. 2006. United States standard for grades of muscadine (Vitis rotundifolia) grapes. U.S. Dept. Agr. Washington, D.C.

Vincenzi, S., D. Tomasi, F. Gaiotti, L. Lovat, S. Giacosa, F. Torchio, S. Rio Segade, and L. Rolle. 2013. Comparative study of the resveratrol content of twenty‐one Italian red grape varieties. South African J. Enol. Viticult. 34:30‐35.

Walker, T.L., J.R. Morris, R.T. Threlfall, G.L. Main, O. Lamikanra, and S. Leong. 2001. Density separation, storage, shelf life, and sensory evaluation of 'Fry' muscadine grapes. HortScience 36:941‐945.

Wang, C.Y. 1990. Chilling injury of horticultural crops. 1st ed. CRC Press, Boca Raton, FL.

Wang, H., G. Cao, and R.L. Prior. 1996. Total antioxidant capacity of fruits. J. Agr. Food Chem. 44:701‐705.

Wu, X., G.R. Beecher, J.M. Holden, D.B. Haytowitz, S.E. Gebhardt, and R.L. Prior. 2004. Lipophilic and hydrophilic antioxidant capacities of common foods in the United States. J. Agr. Food Chem. 52:4026‐4037.

Yi, W., C.C. Akoh, and J. Fischer. 2005. Study of anticancer activities of muscadine grape phenolics in vitro. J. Agr. Food Chem. 53:8804‐8812.

196 Appendix A: Meteorological data, interaction means, and correlations.

Table A.1. Average monthly maximum and minimum temperatures and total rainfall recorded at the Fruit Research Station; Clarksville, AR (lat. 35°31’58°N and long. 93°24’12’W) (2012 and 2013). Year Month Maximum Minimum Precipitation temperature (°C) temperature (°C) (mm) 2012 January 11.4 0.1 111.84 February 11.8 2.5 65.50 March 21.4 9.8 198.73 April 23.08 11.5 81.86 May 26.6 16.3 18.88 June 32.9 19.4 14.36 July 36.5 22.7 40.56 August 33.9 20.9 62.47 September 28.4 17.4 158.19 October 20.2 9.2 127.21 November 16.2 4.4 23.56 December 11.8 1.9 3.23 2013 January 9.1 ‐0.7 98.85 February 9.9 ‐0.2 70.63 March 2.8 2.0 130.30 April 19.5 8.7 119.37 May 23.9 13.6 163.07 June 29.8 19.0 54.61 July 31.4 19.9 100.35 August 30.4 20.9 178.82 September 30.2 17.9 57.91 October 20.9 10.4 106.18 November 12.6 2.3 103.14 December 6.8 ‐1.6 64.51

197 Table A.7. Interaction means of the postharvest attributes of percent weight loss, percent unmarketable, force to penetrate berry skin, and percent volume change for year, genotype, and week of storage at 2‐3 °C for 0‐3 weeks. Genotype Year Week Weight Force Volume Unmarketable loss (%) (N) change (%) (%) AM 01 2012 0 0.0 8.9 0.0 0.0 AM 01 2012 1 1.7 2.7 12.9 18.9 AM 01 2012 2 3.5 6.5 17.7 25.4 AM 01 2012 3 5.1 4.9 20.3 54.6 AM 01 2013 0 0.0 8.7 0.0 0.0 AM 01 2013 1 1.5 8.1 ‐10.7 11.9 AM 01 2013 2 2.6 8.0 ‐7.7 26.0 AM 01 2013 3 4.1 5.0 ‐3.5 94.9 AM 02 2012 0 0.0 9.8 0.0 0.0 AM 02 2012 1 1.5 2.8 ‐21.9 14.3 AM 02 2012 2 3.1 6.5 ‐35.4 22.2 AM 02 2012 3 5.0 4.3 ‐17.9 56.1 AM 02 2013 0 0.0 9.2 0.0 0.0 AM 02 2013 1 1.1 7.9 8.4 7.0 AM 02 2013 2 1.9 7.4 3.0 17.6 AM 02 2013 3 3.2 6.8 ‐3.4 56.6 AM 03 2012 0 0.0 8.9 0.0 0.0 AM 03 2012 1 1.6 3.2 1.0 3.5 AM 03 2012 2 3.0 7.7 0.9 11.3 AM 03 2012 3 4.5 7.4 10.0 15.3 AM 03 2013 0 0.0 11.1 0.0 0.0 AM 03 2013 1 1.5 10.9 1.8 2.7 AM 03 2013 2 3.0 8.7 2.7 4.1 AM 03 2013 3 4.2 8.0 14.6 18.5 AM 04 2012 0 0.0 9.2 0.0 0.0 AM 04 2012 1 1.7 8.5 9.9 9.9 AM 04 2012 2 3.9 6.1 6.9 29.2 AM 04 2012 3 5.9 6.0 12.1 64.8 AM 04 2013 0 0.0 12.1 0.0 0.0 AM 04 2013 1 1.1 12.6 0.7 3.2 AM 04 2013 2 1.8 11.1 3.4 8.2 AM 04 2013 3 2.8 11.2 ‐3.5 12.6 AM 15 2012 0 0.0 8.0 0.0 0.0 AM 15 2012 1 1.4 5.2 11.0 15.8 AM 15 2012 2 2.9 4.9 7.3 24.6 AM 15 2012 3 5.1 2.2 13.4 41.9

198 Genotype Year Week Weight Force Volume Unmarketable loss (%) (N) change (%) (%) AM 15 2013 0 0.0 9.1 0.0 0.0 AM 15 2013 1 0.9 7.8 ‐5.2 18.0 AM 15 2013 2 1.9 6.2 1.0 23.5 AM 15 2013 3 2.8 5.7 10.9 54.0 AM 18 2012 0 0.0 8.4 0.0 0.0 AM 18 2012 1 2.1 2.7 ‐13.9 6.7 AM 18 2012 2 3.9 6.3 14.4 22.0 AM 18 2012 3 6.2 5.8 16.6 38.7 AM 18 2013 0 0.0 8.6 0.0 0.0 AM 18 2013 1 1.1 7.5 10.0 12.0 AM 18 2013 2 1.9 6.0 0.5 17.5 AM 18 2013 3 2.7 6.4 10.2 39.7 AM 26 2012 0 0.0 8.6 0.0 0.0 AM 26 2012 1 1.3 7.2 ‐1.5 14.6 AM 26 2012 2 3.3 7.5 3.2 32.7 AM 26 2012 3 5.1 5.1 1.3 60.7 AM 26 2013 0 0.0 11.2 0.0 0.0 AM 26 2013 1 1.2 10.5 11.1 1.5 AM 26 2013 2 1.8 10.8 15.2 3.7 AM 26 2013 3 2.6 8.9 15.7 8.8 AM 27 2012 0 0.0 8.7 0.0 0.0 AM 27 2012 1 1.4 7.2 ‐19.2 12.1 AM 27 2012 2 2.8 2.4 ‐11.2 22.8 AM 27 2012 3 4.3 4.8 ‐23.0 42.7 AM 27 2013 0 0.0 9.7 0.0 0.0 AM 27 2013 1 0.9 7.1 ‐5.0 9.8 AM 27 2013 2 1.7 6.3 2.3 34.6 AM 27 2013 3 2.3 5.1 8.2 51.5 AM 28 2012 0 0.0 10.4 0.0 0.0 AM 28 2012 1 1.7 9.4 ‐16.6 6.0 AM 28 2012 2 3.4 3.7 ‐31.9 16.1 AM 28 2012 3 5.2 8.1 ‐13.6 28.2 AM 28 2013 0 0.0 10.7 0.0 0.0 AM 28 2013 1 0.9 10.6 2.3 1.8 AM 28 2013 2 1.5 10.0 11.6 5.9 AM 28 2013 3 1.9 9.4 16.9 11.8 Delicious 2012 0 0.0 8.6 0.0 0.0 Delicious 2012 1 1.4 6.6 ‐14.8 12.5 Delicious 2012 2 3.1 2.2 ‐8.6 27.6 Delicious 2012 3 4.7 5.4 ‐0.9 32.3

199 Genotype Year Week Weight Force Volume Unmarketable loss (%) (N) change (%) (%) Delicious 2013 0 0.0 6.9 0.0 0.0 Delicious 2013 1 1.0 6.5 ‐8.2 21.1 Delicious 2013 2 1.7 6.4 3.5 18.5 Delicious 2013 3 2.3 4.2 6.1 39.6 Fry 2012 0 0.0 6.7 0.0 0.0 Fry 2012 1 1.6 4.8 1.9 4.8 Fry 2012 2 3.2 4.1 14.2 23.5 Fry 2012 3 5.0 4.6 8.8 65.8 Fry 2013 0 0.0 8.9 0.0 0.0 Fry 2013 1 1.3 7.0 4.7 17.2 Fry 2013 2 2.7 5.7 10.4 37.3 Fry 2013 3 4.1 5.6 13.1 73.9 Ison 2012 0 0.0 5.9 0.0 0.0 Ison 2012 1 1.9 2.4 ‐1.3 9.0 Ison 2012 2 3.5 4.5 ‐10.2 21.4 Ison 2012 3 5.2 4.5 ‐11.0 52.9 Ison 2013 0 0.0 8.8 0.0 0.0 Ison 2013 1 1.3 6.0 ‐2.1 11.7 Ison 2013 2 2.0 5.7 13.8 33.3 Ison 2013 3 3.2 4.0 21.1 54.3 Nesbitt 2012 0 0.0 9.8 0.0 0.0 Nesbitt 2012 1 2.9 8.7 ‐7.7 7.1 Nesbitt 2012 2 4.4 7.2 ‐15.9 18.8 Nesbitt 2012 3 6.5 5.6 ‐5.8 50.8 Nesbitt 2013 0 0.0 13.2 0.0 0.0 Nesbitt 2013 1 0.8 10.0 20.6 6.1 Nesbitt 2013 2 1.5 11.1 16.4 9.6 Nesbitt 2013 3 2.2 8.9 22.5 22.8 Southern Jewel 2012 0 0.0 7.3 0.0 0.0 Southern Jewel 2012 1 1.5 2.4 ‐7.5 6.9 Southern Jewel 2012 2 3.3 6.1 ‐7.0 11.6 Southern Jewel 2012 3 5.7 5.6 5.9 23.2 Southern Jewel 2013 0 0.0 10.5 0.0 0.0 Southern Jewel 2013 1 0.6 9.4 2.1 14.3 Southern Jewel 2013 2 1.4 8.8 9.7 16.4 Southern Jewel 2013 3 2.0 7.8 9.1 53.6 Summit 2012 0 0.0 8.8 0.0 0.0 Summit 2012 1 1.6 3.0 ‐8.0 5.6 Summit 2012 2 3.6 7.5 ‐9.0 19.0 Summit 2012 3 5.5 8.0 12.6 20.7

200 Genotype Year Week Weight Force Volume Unmarketable loss (%) (N) change (%) (%) Summit 2013 0 0.0 9.7 0.0 0.0 Summit 2013 1 1.0 7.8 ‐1.7 8.1 Summit 2013 2 1.6 8.8 4.7 14.4 Summit 2013 3 2.2 7.2 6.5 51.3 Supreme 2012 0 0.0 9.1 0.0 0.0 Supreme 2012 1 2.0 9.0 ‐31.7 2.2 Supreme 2012 2 3.5 7.3 ‐14.7 18.4 Supreme 2012 3 5.3 7.7 ‐26.6 24.0 Supreme 2013 0 0.0 8.5 0.0 0.0 Supreme 2013 1 1.3 7.9 ‐6.6 3.8 Supreme 2013 2 2.3 6.9 7.3 8.9 Supreme 2013 3 3.2 7.4 14.3 29.3 Tara 2012 0 0.0 7.7 0.0 0.0 Tara 2012 1 1.6 5.4 ‐23.5 16.8 Tara 2012 2 3.1 1.8 ‐7.9 29.7 Tara 2012 3 4.7 4.2 ‐4.0 37.4 Tara 2013 0 0.0 7.9 0.0 0.0 Tara 2013 1 0.9 6.1 7.4 22.1 Tara 2013 2 1.8 5.3 6.2 44.3 Tara 2013 3 2.7 3.3 16.3 70.5

201 Table A.8. Interaction means of the composition attributes of percent titratable acidity, percent soluble solids, pH, and soluble solids to titratable acidity for year, genotype, and week of storage at 2 °C for 0‐3 weeks. Genotype Year Week Titratable pH Soluble SS/TA acidity solids (%) (%) AM 01 2012 0 0.6 3.7 25.6 55.4 AM 01 2012 1 0.6 3.7 25.9 56.1 AM 01 2012 2 0.6 3.6 25.0 56.1 AM 01 2012 3 0.6 3.6 25.7 57.6 AM 01 2013 0 0.4 3.2 17.7 54.1 AM 01 2013 1 0.4 3.6 17.3 49.7 AM 01 2013 2 0.5 3.6 17.2 46.5 AM 01 2013 3 0.5 3.6 18.1 47.3 AM 02 2012 0 0.6 3.8 20.3 42.3 AM 02 2012 1 0.6 3.8 20.8 43.5 AM 02 2012 2 0.6 3.7 19.3 42.0 AM 02 2012 3 0.6 3.8 20.0 42.1 AM 02 2013 0 0.3 4.0 17.4 70.4 AM 02 2013 1 0.3 4.0 17.4 74.6 AM 02 2013 2 0.3 4.0 18.2 80.2 AM 02 2013 3 0.3 3.9 18.3 70.5 AM 03 2012 0 0.3 3.6 23.5 112.1 AM 03 2012 1 0.3 3.5 25.2 101.6 AM 03 2012 2 0.3 3.6 23.6 112.6 AM 03 2012 3 0.3 3.6 23.6 91.1 AM 03 2013 0 0.3 3.7 18.1 80.7 AM 03 2013 1 0.3 3.7 18.4 87.2 AM 03 2013 2 0.3 3.7 17.9 74.6 AM 03 2013 3 0.3 3.8 18.5 80.2 AM 04 2012 0 0.6 3.8 20.3 42.2 AM 04 2012 1 0.6 3.7 20.3 43.3 AM 04 2012 2 0.6 3.9 21.2 43.2 AM 04 2012 3 0.6 3.8 20.1 42.4 AM 04 2013 0 0.3 3.8 16.5 72.5 AM 04 2013 1 0.3 3.7 16.4 60.0 AM 04 2013 2 0.4 3.7 16.7 58.0 AM 04 2013 3 0.4 3.7 16.6 58.6 AM 15 2012 0 0.5 3.5 19.6 45.0 AM 15 2012 1 0.5 3.3 19.8 47.3 AM 15 2012 2 0.5 3.4 19.7 46.6 AM 15 2012 3 0.5 3.3 19.5 47.6 AM 15 2013 0 0.4 3.5 17.4 60.5

202 Genotype Year Week Titratable pH Soluble SS/TA acidity solids (%) (%) AM 15 2013 1 0.4 3.3 17.0 50.8 AM 15 2013 2 0.4 3.4 18.0 59.4 AM 15 2013 3 0.4 3.4 18.3 38.2 AM 18 2012 0 0.3 4.0 17.5 83.3 AM 18 2012 1 0.2 3.7 18.0 95.0 AM 18 2012 2 0.3 3.8 18.4 88.1 AM 18 2012 3 0.3 3.7 17.6 75.6 AM 18 2013 0 0.3 3.7 16.3 60.4 AM 18 2013 1 0.3 3.7 15.4 66.0 AM 18 2013 2 0.3 3.6 16.0 57.7 AM 18 2013 3 0.4 3.7 15.7 55.0 AM 26 2012 0 0.6 3.9 22.3 46.2 AM 26 2012 1 0.3 3.9 22.3 106.8 AM 26 2012 2 0.6 3.8 22.5 48.1 AM 26 2012 3 0.6 3.7 20.6 44.0 AM 26 2013 0 0.2 3.9 15.8 80.3 AM 26 2013 1 0.3 3.8 15.8 75.2 AM 26 2013 2 0.3 3.8 15.8 70.4 AM 26 2013 3 0.3 3.8 15.6 73.6 AM 27 2012 0 0.5 3.5 20.5 47.5 AM 27 2012 1 0.5 3.4 19.8 47.1 AM 27 2012 2 0.5 3.4 19.2 45.2 AM 27 2012 3 0.5 3.4 19.5 45.3 AM 27 2013 0 0.4 3.7 17.1 61.3 AM 27 2013 1 0.4 3.6 18.1 57.7 AM 27 2013 2 0.4 3.7 18.2 62.8 AM 27 2013 3 0.4 3.7 17.7 56.5 AM 28 2012 0 0.6 3.6 21.0 46.2 AM 28 2012 1 0.6 3.6 20.7 45.7 AM 28 2012 2 0.5 3.5 19.9 45.4 AM 28 2012 3 0.5 3.5 21.3 49.0 AM 28 2013 0 0.2 3.8 14.1 83.7 AM 28 2013 1 0.2 3.8 14.3 88.0 AM 28 2013 2 0.2 3.7 14.2 71.9 AM 28 2013 3 0.3 3.7 14.6 68.5 Delicious 2012 0 0.5 3.4 20.4 47.7 Delicious 2012 1 0.5 3.5 20.3 46.9 Delicious 2012 2 0.5 3.4 21.4 49.9 Delicious 2012 3 0.5 3.4 20.5 48.0 Delicious 2013 0 0.4 3.5 16.7 58.3

203 Genotype Year Week Titratable pH Soluble SS/TA acidity solids (%) (%) Delicious 2013 1 0.5 3.5 16.0 39.9 Delicious 2013 2 0.5 3.5 17.0 45.3 Delicious 2013 3 0.5 3.6 17.5 46.5 Fry 2012 0 0.6 3.7 22.9 50.0 Fry 2012 1 0.6 3.6 21.8 48.1 Fry 2012 2 0.6 3.6 22.7 50.7 Fry 2012 3 0.6 3.6 22.7 50.1 Fry 2013 0 0.3 3.7 18.3 70.6 Fry 2013 1 0.3 3.8 19.6 73.8 Fry 2013 2 0.3 3.8 18.6 76.6 Fry 2013 3 0.4 3.7 19.8 63.6 Ison 2012 0 0.6 3.1 16.6 35.3 Ison 2012 1 0.5 3.1 16.8 44.8 Ison 2012 2 0.4 3.3 19.8 69.9 Ison 2012 3 0.3 3.4 18.4 67.0 Ison 2013 0 0.4 3.4 18.2 54.0 Ison 2013 1 0.5 3.4 18.0 50.1 Ison 2013 2 0.4 3.5 18.8 59.7 Ison 2013 3 0.4 3.5 19.3 62.3 Nesbitt 2012 0 0.5 3.4 18.3 43.6 Nesbitt 2012 1 0.6 3.6 20.0 44.3 Nesbitt 2012 2 0.6 3.7 20.5 44.6 Nesbitt 2012 3 0.6 3.7 21.3 46.1 Nesbitt 2013 0 0.3 3.9 16.6 75.5 Nesbitt 2013 1 0.4 3.8 16.6 58.0 Nesbitt 2013 2 0.3 3.8 17.0 71.4 Nesbitt 2013 3 0.4 3.8 17.0 59.6 Southern Jewel 2012 0 0.5 3.4 17.2 41.1 Southern Jewel 2012 1 0.5 3.5 18.5 42.5 Southern Jewel 2012 2 0.5 3.4 18.4 49.9 Southern Jewel 2012 3 0.5 3.4 19.2 44.7 Southern Jewel 2013 0 0.4 3.6 18.9 65.3 Southern Jewel 2013 1 0.4 3.5 18.9 61.2 Southern Jewel 2013 2 0.3 3.6 19.2 72.3 Southern Jewel 2013 3 0.4 3.6 18.7 66.6 Summit 2012 0 0.6 3.8 27.3 56.8 Summit 2012 1 0.6 3.8 25.5 54.2 Summit 2012 2 0.6 3.6 24.0 53.0 Summit 2012 3 0.6 3.6 25.8 56.7 Summit 2013 0 0.3 3.8 17.7 72.8

204 Genotype Year Week Titratable pH Soluble SS/TA acidity solids (%) (%) Summit 2013 1 0.3 3.7 17.9 70.6 Summit 2013 2 0.3 3.6 17.4 74.9 Summit 2013 3 0.3 3.6 17.9 75.6 Supreme 2012 0 0.6 3.6 17.8 39.2 Supreme 2012 1 0.6 3.9 21.2 43.3 Supreme 2012 2 0.6 3.8 20.5 42.8 Supreme 2012 3 0.6 3.7 20.2 44.1 Supreme 2013 0 0.3 4.0 18.1 88.5 Supreme 2013 1 0.3 3.9 17.9 87.2 Supreme 2013 2 0.3 3.9 17.7 75.9 Supreme 2013 3 0.3 3.8 16.2 65.0 Tara 2012 0 0.6 3.6 20.8 47.0 Tara 2012 1 0.6 3.6 20.8 46.8 Tara 2012 2 0.6 3.6 21.2 47.5 Tara 2012 3 0.6 3.7 21.8 47.7 Tara 2013 0 0.3 3.7 16.5 67.7 Tara 2013 1 0.3 3.7 17.2 68.0 Tara 2013 2 0.3 3.7 17.1 63.6 Tara 2013 3 0.4 3.7 16.8 59.0

205 Table A.9. Interaction means of the berry color attributes of Chroma, Hue angle, and L* values for year, genotype, and week of storage at 2 °C for 0‐3 weeks. Genotype Year Week L* Chroma Hue angle AM 01 2012 0 43.2 14.0 71.3 AM 01 2012 1 39.6 13.4 66.4 AM 01 2012 2 42.3 14.4 72.4 AM 01 2012 3 39.5 13.5 74.1 AM 01 2013 0 90.5 4.7 218.1 AM 01 2013 1 85.6 6.4 211.0 AM 01 2013 2 105.5 5.0 135.7 AM 01 2013 3 83.3 3.5 207.7 AM 02 2012 0 27.4 2.2 314.4 AM 02 2012 1 25.4 2.8 238.3 AM 02 2012 2 25.5 4.1 280.4 AM 02 2012 3 26.7 2.1 316.6 AM 02 2013 0 26.0 2.0 346.2 AM 02 2013 1 24.5 2.1 270.3 AM 02 2013 2 26.6 2.0 349.2 AM 02 2013 3 26.0 2.9 272.7 AM 03 2012 0 46.8 18.5 56.0 AM 03 2012 1 44.5 17.9 54.0 AM 03 2012 2 44.6 18.3 60.5 AM 03 2012 3 45.0 16.9 60.2 AM 03 2013 0 97.1 3.4 128.5 AM 03 2013 1 45.3 14.6 92.4 AM 03 2013 2 43.9 13.9 95.1 AM 03 2013 3 43.0 12.8 92.5 AM 04 2012 0 96.8 5.6 69.5 AM 04 2012 1 25.5 3.4 286.1 AM 04 2012 2 26.9 2.5 200.9 AM 04 2012 3 27.0 2.6 241.3 AM 04 2013 0 26.6 3.5 159.0 AM 04 2013 1 25.3 2.9 122.3 AM 04 2013 2 24.5 2.9 278.7 AM 04 2013 3 24.2 3.9 280.5 AM 15 2012 0 42.3 12.2 55.8 AM 15 2012 1 40.8 11.9 54.1 AM 15 2012 2 40.7 12.9 58.9 AM 15 2012 3 42.7 13.8 60.2 AM 15 2013 0 91.0 9.5 256.9 AM 15 2013 1 42.2 11.5 69.0 AM 15 2013 2 41.0 12.3 64.4 AM 15 2013 3 32.6 13.5 63.8

206 Genotype Year Week L* Chroma Hue angle AM 18 2012 0 28.9 3.1 237.3 AM 18 2012 1 26.8 2.6 198.6 AM 18 2012 2 25.7 3.3 123.4 AM 18 2012 3 27.3 3.1 312.7 AM 18 2013 0 59.9 2.4 290.4 AM 18 2013 1 26.1 2.1 235.6 AM 18 2013 2 26.6 2.0 306.5 AM 18 2013 3 26.4 2.1 233.2 AM 26 2012 0 41.6 12.8 77.3 AM 26 2012 1 38.8 11.0 75.7 AM 26 2012 2 38.6 11.8 71.9 AM 26 2012 3 40.8 11.7 81.8 AM 26 2013 0 44.6 13.8 93.2 AM 26 2013 1 43.9 13.9 90.0 AM 26 2013 2 43.4 11.8 93.1 AM 26 2013 3 42.2 13.0 86.3 AM 27 2012 0 26.8 2.1 198.5 AM 27 2012 1 26.8 2.6 278.5 AM 27 2012 2 27.2 3.1 125.6 AM 27 2012 3 26.0 2.6 278.4 AM 27 2013 0 25.9 2.0 342.0 AM 27 2013 1 26.4 1.6 306.3 AM 27 2013 2 26.2 2.1 345.8 AM 27 2013 3 25.8 2.2 347.6 AM 28 2012 0 28.8 3.5 286.8 AM 28 2012 1 26.6 3.2 120.7 AM 28 2012 2 27.5 3.1 126.4 AM 28 2012 3 26.4 2.6 286.3 AM 28 2013 0 22.9 3.9 358.7 AM 28 2013 1 26.2 4.5 322.8 AM 28 2013 2 25.8 4.1 357.4 AM 28 2013 3 26.0 4.7 358.9 Delicious 2012 0 27.4 2.1 198.5 Delicious 2012 1 27.5 2.5 271.4 Delicious 2012 2 27.0 2.3 275.9 Delicious 2012 3 26.8 2.4 201.0 Delicious 2013 0 60.0 3.2 292.2 Delicious 2013 1 27.3 3.2 237.1 Delicious 2013 2 26.7 3.0 306.7 Delicious 2013 3 27.1 2.5 348.9 Fry 2012 0 44.3 16.3 86.6 Fry 2012 1 41.6 15.3 76.9

207 Genotype Year Week L* Chroma Hue angle Fry 2012 2 41.0 14.4 82.7 Fry 2012 3 41.4 13.8 73.7 Fry 2013 0 43.8 14.9 96.0 Fry 2013 1 42.6 14.2 96.5 Fry 2013 2 42.1 14.1 90.9 Fry 2013 3 40.8 13.1 94.7 Ison 2012 0 28.1 6.8 288.1 Ison 2012 1 28.2 4.9 236.6 Ison 2012 2 25.4 3.0 162.3 Ison 2012 3 26.4 3.4 202.9 Ison 2013 0 26.3 3.7 345.9 Ison 2013 1 26.7 2.0 235.1 Ison 2013 2 26.8 2.1 201.1 Ison 2013 3 26.3 1.9 346.5 Nesbitt 2012 0 29.6 9.2 359.7 Nesbitt 2012 1 25.3 5.9 279.8 Nesbitt 2012 2 26.8 5.3 319.4 Nesbitt 2012 3 27.1 5.0 279.9 Nesbitt 2013 0 28.4 4.4 198.8 Nesbitt 2013 1 27.0 3.1 200.1 Nesbitt 2013 2 27.0 3.3 279.3 Nesbitt 2013 3 25.7 4.5 328.1 Southern Jewel 2012 0 29.6 5.2 311.6 Southern Jewel 2012 1 27.8 3.8 272.9 Southern Jewel 2012 2 25.9 5.5 121.8 Southern Jewel 2012 3 28.8 3.6 236.0 Southern Jewel 2013 0 56.1 23.1 264.1 Southern Jewel 2013 1 27.3 2.6 313.3 Southern Jewel 2013 2 26.3 2.6 305.1 Southern Jewel 2013 3 25.7 2.5 272.8 Summit 2012 0 39.9 14.9 47.8 Summit 2012 1 40.4 15.1 51.7 Summit 2012 2 41.0 15.5 65.2 Summit 2012 3 39.0 15.0 51.3 Summit 2013 0 44.9 12.7 92.1 Summit 2013 1 41.9 11.9 83.1 Summit 2013 2 38.6 13.8 86.6 Summit 2013 3 33.8 10.1 81.5 Supreme 2012 0 27.9 8.6 359.4 Supreme 2012 1 26.5 4.2 359.8 Supreme 2012 2 27.4 4.5 359.5 Supreme 2012 3 28.4 5.5 359.0

208 Genotype Year Week L* Chroma Hue angle Supreme 2013 0 25.3 5.6 282.8 Supreme 2013 1 26.4 5.6 201.1 Supreme 2013 2 27.2 4.9 278.8 Supreme 2013 3 26.5 5.4 208.4 Tara 2012 0 45.7 13.9 83.2 Tara 2012 1 44.4 12.8 80.0 Tara 2012 2 43.1 13.1 76.7 Tara 2012 3 43.3 11.9 83.7 Tara 2013 0 46.2 13.4 97.9 Tara 2013 1 47.5 17.5 98.4 Tara 2013 2 41.1 11.2 94.1 Tara 2013 3 39.8 12.8 90.1

209

Table A.10. Interaction means of the berry nutraceutical concentrations of total phenolics, total anthocyanins, total ellagitannins, total flavonols, and resveratrol, and the antioxidant capacity (ORAC) for year and genotypes. Genotype Year Total Total ORAC Total Total Resveratrol anthocyanins ellagitannins (µmol flavonols phenolics (mg/100 g) (mg/100 g) (mg/100 g) TE/g) (mg/100 (mg/100 g) g) AM 01 2012 0.0 2.1 71.2 41.0 604.7 13.2 AM 01 2013 0.0 4.0 71.2 15.8 528.3 3.9 AM 02 2012 61.2 5.7 87.1 19.4 581.2 3.8 AM 02 2013 36.4 10.4 106.3 16.8 464.5 4.1 AM 03 2012 0.0 12.2 92.1 70.6 701.3 5.5 AM 03 2013 0.0 12.8 53.5 33.8 575.9 6.7 AM 04 2012 109.3 7.1 71.3 26.1 539.2 7.0 AM 04 2013 41.1 12.2 97.2 18.4 558.3 5.5 AM 15 2012 0.0 4.4 86.7 29.3 639.8 3.9 AM 15 2013 0.0 11.9 87.2 47.9 604.3 3.8 AM 18 2012 77.9 2.6 48.4 11.5 358.8 4.0 AM 18 2013 32.7 5.7 85.7 11.7 326.8 3.5 AM 26 2012 0.0 1.5 34.2 22.7 368.3 4.7 AM 26 2013 0.0 6.7 57.3 23.5 392.1 3.7 AM 27 2012 122.0 8.0 81.7 21.5 669.3 16.7 AM 27 2013 41.8 6.6 94.8 14.7 448.3 4.6 AM 28 2012 55.1 2.9 52.0 8.9 354.5 4.4 AM 28 2013 26.5 5.5 68.6 9.9 316.9 2.9 Delicious 2012 98.1 10.0 90.9 32.6 618.6 5.2 Delicious 2013 39.3 10.5 107.0 19.6 606.7 6.2 Fry 2012 0.0 4.4 84.6 28.4 583.8 4.0 Fry 2013 0.0 12.5 59.3 24.9 527.4 6.2 Ison 2012 44.4 12.4 110.6 22.1 797.3 10.1 Ison 2013 30.3 6.7 115.5 12.7 544.3 4.2 Nesbitt 2012 17.9 2.0 72.2 17.3 518.6 4.3 Nesbitt 2013 22.5 7.7 67.8 11.0 450.1 8.1 S. Jewel 2012 78.1 4.3 67.6 9.1 694.7 5.8 S. Jewel 2013 31.3 7.5 111.3 15.7 579.0 3.2 Summit 2012 0.0 10.5 86.8 63.1 492.1 4.8 Summit 2013 0.0 11.2 57.6 23.4 535.0 11.1 Supreme 2012 16.7 1.6 52.7 7.3 366.1 4.1 Supreme 2013 19.9 5.7 57.8 11.5 354.2 12.1 Tara 2012 0.0 2.9 47.7 19.5 439.0 5.1 Tara 2013 0.0 4.2 65.9 12.7 476.6 3.3

210 Table A.11. Study 2 multivariate correlation coefficients among muscadine berry storage quality, composition, color, and nutraceutical content for 2012 and 2013. *Significant at P=0.05. WLz UMy TAw pH SSv L* Chru Huet Fors VCx Anthq Ellap ORACo Flan WLz UMy 0.00 TAw -0.23 0.57 pH 0.33 -0.30 -0.25 SSv 0.25 0.21 0.19 0.03 L* 0.29 0.34 0.06 -0.12 0.71* Chru 0.13 0.13 -0.13 0.02 0.69 0.72 Huet -0.10 -0.23 0.11 0.04 -0.64 -0.8* -0.93* Fors 0.20 -0.74* -0.22 0.54 -0.07 -0.13 -0.08 0.12 VCx 0.44 0.13 -0.25 -0.15 0.21 0.55 0.52 -0.63 0.08 Anthq -0.35 -0.11 0.11 -0.08 -0.58 -0.66 -0.87* 0.75 0.04 -0.4 Ellap -0.06 -0.14 -0.08 -0.23 0.36 0.00 0.13 -0.20 -0.07 0.07 0.12 ORACo -0.18 0.27 0.31 -0.57 -0.08 -0.27 -0.45 0.35 -0.42 -0.2 0.52 0.59 Flan 0.08 -0.05 -0.14 -0.17 0.73* 0.64 0.66 -0.73* -0.03 0.38 -0.47 0.69 0.09 TPm -0.07 0.32 0.30 -0.66 0.39 0.25 0.11 -0.16 -0.38 0.18 0.07 0.67 0.78* 0.50 Resl 0.08 0.08 0.33 -0.15 0.40 0.06 -0.12 0.12 -0.02 -0.3 0.13 0.16 0.17 0.14 zWL=weight loss of berries (%) yUM=unmarketable berries (%) xVC=volume change (%) wTA=titratable acidity (%) measures as tartaric acid vSS=soluble solids (%) uChr=Chroma tHue=hue angle sFor=force to penetrate berry skin (%) qAnth=total anthocyanins (mg/100 g) pElla=total ellagitannins (mg/100 g) oORAC=oxygen radial absorbance capacity (µmol Trolox equivalent/ g) nFla=total flavonols (mg/100 g) mTP=total phenolics (mg/100 g) lRes=trans-resveratrol (mg/100 g) 211

Chapter 3

THE EFFECT OF STORAGE TIME ON NUTRACEUTICAL CONTENT OF ‘SUPREME’

MUSCADINE BERRY SEGMENTS

Abstract

A common goal of muscadine (Vitis rotundifolia Michx.) breeding programs is increased storability, as a major limiting factor in muscadine commercialization is deterioration during storage. Genotypes with improved skin and flesh texture have shown some promise for increased storability. Muscadines have been shown to have high levels of nutraceuticals in the berry skin and seeds; however, studies on the retention of these nutraceuticals are limited. The effect of storage time on total anthocyanins, total ellagitannins, total flavonols, total phenolics, trans‐resveratrol, and the antioxidant capacity measurement of oxygen radical absorbance capacity (ORAC) were evaluated on whole berries, as well as berry flesh (skin and pulp) and seeds from three vines of ‘Supreme’ in

2012 and 2013. Muscadine berries were harvested and stored at 2 °C and 85‐95% relative humidity. Every 7 d for 6 weeks in 2012, and only at harvest in 2013 berries nutraceutical measurements were taken. In 2012 total anthocyanins, total ellagitannins, ORAC, total flavonols, total phenolics, and trans‐resveratrol were significantly affected by week of storage, vine, and berry segment, though varying differences occurred. The nutraceutical measurements at harvest indicated total anthocyanins, ORAC, total flavonols, total phenolics, and trans‐resveratrol were significantly affected by year, vine, and berry segment, while total ellagitannins were only affected by berry segment. It was found that total anthocyanins were greatest in the flesh and whole berries, and not detected in the seeds. Total ellagitannins, total flavonols, and trans‐resveratrol were found to be greatest

212 in the seeds. Total phenolics and ORAC in the whole berries had the greatest concentrations. Differences in nutraceutical concentrations among vines were generally minimal; however, vine 1 was found to generally have the highest total flavonol concentration. Nutraceutical concentrations varied among weeks of storage, but no clear pattern of change was evident during storage. Total anthocyanins, ORAC, and total

phenolics were greater in 2012, while trans‐resveratrol was greater in 2013. Overall, it was

determined that nutraceutical and antioxidant capacity varied by berry segment and year

of the study, but were generally less affected by vine and storage time.

213 Introduction

Muscadine grapes (Vitis rotundifolia Michx.) are indigenous to the southeastern

United States. Muscadines have been under cultivation for over 400 years, originally in the

North Carolina Colony followed soon after by surrounding colonies and states (Conner,

2009). Muscadines are the most important Vitis species cultivated in the southeastern U.S.

(Marshall and Stringer, 2014). This native grape is presently grown in small vineyards and home plantings, ranging from North Carolina and Florida to eastern Oklahoma and Texas.

Arkansas has approximately 230 ha of muscadines in production, making up 10% of total

US production. The recent recognition that the berries are important sources of beneficial antioxidants has increased consumer demand (Perkins‐Veazie et al., 2012). A major limiting factor in muscadine grape commercialization is deterioration during storage. One solution for extending market seasons and preventing market saturation for fresh muscadines could be the release of new cultivars with improved postharvest storability. In related studies it was found that genotypes with improved texture and increased firmness, e.g. ‘Supreme’, have the potential to remain marketable following storage time of up to 3 weeks (Studies 1 and 2). Despite that, the retention of nutraceuticals in freshmarket muscadines in storage is widely unstudied (Marshall and Stringer, 2014).

Nutraceuticals are compounds found in foods that have physiological benefits.

Among fruits, muscadines contain some of the highest levels of nutraceuticals; additionally, several of the compounds present are unique to muscadines (Marshall et al. 2012). Brown

(1940) first researched the anthocyanin pigments of muscadines, and found that the anthocyanins present in muscadines are unique in that they are 3,5‐diglucosides, this differs from V. vinifera L. and V. aestivalis Michx., which contain 3‐monoglucosides.

214 Individual anthocyanins present in muscadines have been identified as cyanidin (Cy),

delphinidin (Dp), malvidin (Mv), peonidin (Pn), and petunidin (Pt) (Ballinger et al., 1973;

Flora, 1978, Goldy et al., 1986). The presence of anthocyanins have been shown to protect

blood vessels, and play a role in cancer prevention, though anthocyanin absorption in the

blood appears to be low in humans (Prior, 2004).

Although commonly found in other fruits, muscadines have been shown to contain

ellagitannins, which is unique within Vitis (Marshall et al., 2012). Ellagic acid in muscadines

is expressed as free ellagic acid, ellagic acid glycosides, and ellagitannins (Marshall et al.,

2012; Talcott and Lee, 2002). Ellagic acid and its derivatives have been widely studied due

to their antiproliferative properties through their ability to directly inhibit DNA binding of

certain carcinogens, and their chemoprotective effect in cellular models by reducing

oxidative stress (Talcott and Lee, 2002, Lesca, 1983. Patrana‐Bonilla, 2003, Mertens‐

Mertens‐Talcott, et al., 2003, Stoner and Morse, 1997, Khanduja et al., 1999). Talcott and

Lee (2002) identified that potential for increased marketability of muscadines exists, due to the possible health benefits associated with the presence of ellagitannins.

In addition to ellagitannins, the presence of the flavonol myricetin in bronze muscadines is also unique (Marshall et al., 2012). It has been shown that flavonols protect against the initiation of cancer formation through protection against DNA mutations

(Hollman, and Katan, 1999; Williamson, and Manach, 2005). Certain flavonol compounds have also been shown to prevent heart disease through relaxation of blood vessel walls and increased production of enzymes that dissolve blood clots (Abou‐Agag, et al., 2001; Rendig, et al., 2001). Sandhu and Gu (2010) identified the flavonols present in muscadines as glycosides of quercetin, kaempferol, and myricetin. They also identified the flavonols

215 myricetin hexoside, kaempferol hexoside, quercetin glucoside, and kaempferol rutinoside for the first time in muscadines.

Phenolics are secondary metabolites that are ubiquitous in the kingdom Plantea, and are involved in plant response to abiotic and biotic stresses (Marshal et al., 2012). In addition to protecting the parent plant, often phenolics exhibit significant pharmaceutical benefits (Marshall et al., 2012). Measurements of total phenolics to provide an overall assessment of the content and chemical activity of compounds present, and aids in determining the antioxidant capacity of fruits and vegetables (Lee and Talcott, 2004;

Thiapong et al., 2006). Muscadine grapes have been identified as having high levels of total phenolics (Marshall et al., 2012; Pastrana‐Bonilla et al., 2003; Striegler et al., 2005; Stringer et al., 2009; Threlfall et al., 2007), and Threlfall et al. (2007) found that total phenolic concentrations are not related to berry skin color, and ranged in concentrations among bronze and black genotypes.

Stilbenes are synthesized in grape leaves in response to both biotic and abiotic induction treatments, and the capacity to produce stilbenes is correlated with the resistance of grape leaves to fungal infection (Creasy and Coffee, 1988; Marshall et al.,

2012). Resveratrol (trans‐3,5,4’‐trihydroxystilbene) is a phytoalexin, or stilbene, produced as a response to fungal infection, stress including injury, and UV‐irradiation (Jeandet et al.,

1995; Jeandet et al., 1991; Marshall et al., 2012; Magee et al., 2002; Threlfall et al., 1999).

Resveratrol has two forms, cis‐ and a trans‐resveratrol; this study will only focus on the trans isomer. Resveratrol has long been confirmed in both red and white V. vinifera grape skins and pulp, but not in seeds. Only within the last 20 years has resveratrol been measured in muscadines. The fleshy parts of both black and bronze muscadine berries have

216 higher concentrations of resveratrol than reported for V. vinifera and V. labrusca (Ector et al., 1996). Resveratrol can potentially act as a chemopreventative agent and afford protection against cardiovascular and coronary heart disease (Hudson et at., 2007; Jang et al., 1997; Lu and Sorreno, 1999; Magee et al., 2002). Resveratrol has lipid‐lowering action, inhibition of human low‐density lipoprotein oxidation and thus may delay atherosclerosis onset, inhibition of platelet aggregation in the blood, reduce cholesterol levels, and has been shown to have anticarcinogenic activity in all stages of prostate and breast cancer

(Ector et al., 1996; Ector, 2001; Hudson et at., 2007; Jang et al., 1997; Lu and Sorreno, 1999;

Magee et al., 2002).

Measurements of peroxyl radical (free radical) scavenging activity using oxygen radical absorbance capacity (ORAC) assay is a common index that provides an overall assessment of the content and chemical activity of compounds present, and aids in determining the antioxidant capacity of fruits and vegetables, although its significance is often questioned, as it does not accurately represent the bioactivity of the antioxidants in the human body (Lee and Talcott, 2004; Prior et al, 2003). Free radicals may contribute to the cause of a number of diseases including cancer and atherosclerosis (Wang et al., 1996).

Studies have been conducted to determine the ORAC values of muscadines (Sandhu and Gu,

2010; Striegler et al., 2005; Talcott and Lee, 2002; Threlfall et al., 2007). It has been

determined that ORAC is generally higher in fully ripe muscadines, and levels vary among

genotypes and berry segments (Sandhu and Gu, 2010; Striegler et al., 2005; Talcott and Lee,

2002; Threlfall et al., 2005).

Studies have been conducted to understand the concentration of nutraceutical

compounds in the different segments (juice, skins, pulp, and seeds) of muscadine berries

217 (Lee and Talcott, 2004; Marshall et al., 2012; Sandhu and Gu, 2010; Shi et al., 2003). Ellagic

acid has been identified in higher concentrations in the skins of the berry when compared

to the pulp or juice (Lee and Talcott, 2004). Anthocyanin concentrations were found to be

generally higher in the berry skins and juice than in the pulp or seed (Lee and Talcott,

2004; Marshall et al., 2012; Threlfall et al., 2005). Total phenolics were higher in the seeds

than in the skin or pulp of muscadines (Lee and Talcott, 2004; Marshall et al., 2012; Shi, et

al., 2003; Takeda et al., 1983; Threlfall et al., 2005). ORAC has been measured in muscadine

seeds, skins, pulp, and juice (Lee and Talcott, 2004; Threlfall et al., 2005). Lee and Talcott

(2004) found ORAC values to be highest in the skin, which is often exceptionally thick in

muscadines. However, the effect of storage time on the nutraceutical levels of individual

muscadine berry segments is widely unstudied.

Since the implementation of a muscadine breeding program at the University of

Arkansas in 2005, selections have been made based on improved texture, and dry stem

scar, potentially resulting in improved postharvest storability. Nutraceutical levels in muscadines can vary among genotypes (Marshall et al., 2012), and no information has been collected on the nutraceutical content of the University of Arkansas breeding selections; however, nutraceutical concentrations have merit as possible variables for selection. (J.R.

Clark, personal communication). The cultivar Supreme was used as it has shown to have improved texture and the potential for maintaining high quality extended postharvest storage.

The objective of this study was to determine the effect of storage time on nutraceutical concentrations and antioxidant capacity of ‘Supreme’ muscadine berry flesh

(skin and pulp), seeds, and whole berries. It was hypothesized that nutraceutical content

218 and antioxidant capacity of muscadine grapes would vary during storage.

Materials and Methods

Grapes and Vineyard

Vines of ‘Supreme’ muscadines used for the study were grown at the University of

Arkansas Fruit Research Station, Clarksville, AR (lat. 35°31’58’N and long. 93°24’12’W).

One vine used in this study was 7 years old, while the other 2 vines were three years old.

The vines were grown in Linker fine sandy loam, in USD hardiness zone 7a, where average

annual minimum temperatures reach ‐15 to – 17.7 °C. Vine spacing was 6.1 m apart and

rows were spaced 3.0 m apart. A single‐wire trellis was used, and vines were trained to a

bilateral cordon. The vines were dormant‐pruned annually in February using spur pruning

with spurs retained of two to four buds in length. Weeds were controlled with pre‐ and postemergence herbicides as needed and vines did not have any stress from weed competition. Vines were irrigated by drip irrigation as needed, beginning in early June

(prior months received adequate rainfall) and continuing through the harvest period. Vines received N fertilization in March of each year at a rate of approximately 70 kg/ha. No insecticides, fungicides or other pest control compounds were applied to the vines. The

vines used in the study had full crops produced each year, and no crop reduction due to

winter injury or other limitations occurred. Thus, the vines produced berries under

representative conditions. Daily maximum and minimum temperatures along with rainfall

were recorded at the research location to characterize the environment the vines were

subjected to and potential differences among years (Table A.1).

Harvest and Transport

219 The muscadine berries were once‐over, hand‐harvested at the Fruit Research

Station. Berries were harvested late in the afternoon and transported to University of

Arkansas Institute of Food Science and Engineering, Fayetteville, AR., in an air‐conditioned

car on the same day. Harvest date/maturity was based on soluble solids of 18‐22% in 2012

and 15‐18% in 2013 (due to differences in summer temperature and precipitation), ease of

release from the pedicel, and berry color.

Composition Analysis

Composition measurements were taken only at harvest. Titratable acidity and pH

were measured by an 877 Titrino Plus (Metrohm AG, Herisau Switzerland) with an

automated titrimeter and electrode standardized to pH 2.0, 4.0, 7.0, and 10.0 buffers.

Titratable acidity was determined using 6 g of juice diluted with 50 mL of deionized,

degassed water by titration of 0.1 N sodium hydroxide (NaOH) to an endpoint of pH 8.2, and

results were expressed as percent tartaric acid. Soluble solids were measured using a

Bausch and Lomb Inc. Abbe Mark II refractometer (Rochester, NY). Soluble solids, TA, and

pH were measured from the juice of the whole berries, strained through cheesecloth to

remove any solids.

Postharvest Storage

Berries were then hand‐sorted to remove any split, shriveled, or decayed fruit

before packaging. Only sound berries, showing no signs of unmarketability, were stored.

The berries were packaged into hinged standard vented clamshells (18.4 cm x 12.1 cm x 8.9

cm) (H116, FormTex Plastics Corporation, Houston, TX) and stored in plastic harvest lugs in cold storage at 2 °C with 85‐89% relative humidity (RH). From the harvested berries,

three vented clamshell containers were filled to approximately 500 g. Every 7 d for 6

220 weeks, six randomly selected berries from each replication were removed, placed in plastic

bags and stored at ‐20 °C until analysis. Of the six berries, three were kept as whole berries, and three were divided into flesh (skin and pulp) and seeds for nutraceutical analysis.

Nutraceutical Analysis

For nutraceutical analysis, the frozen berries were thawed in 30 °C water for 30 s.

The berries were then cut open and the seeds were removed from the flesh for analysis.

The berries and berry segments were homogenized three times each for 1 min in

alternating washes of 80 ml (whole berries and flesh) or 25 ml (seeds) of extraction

solution containing methanol/water/formic acid (MWF) (60:37:3 v/v/v) and

acetone/water/acetic (70:29.5:0.5 v/v/v) to the smallest particle size using a Euro Turrax

T18 Tissuemizer (Tekmar‐Dohrman Corp, Mason, OH). Homogenates were then

centrifuged for 5 min at 10,000 rpm and filtered through Miracloth (CalBiochem, LaJolla,

CA). The samples were taken to a final volume of 250 ml (whole berries and flesh) or 100

ml (seeds) with extraction solvent and stored at ‐70 °C until further analysis. Prior to high

performance liquid chromatography (HPLC) analysis, the samples were filtered through

0.45 μm filters (Whatman PLC, Maidstone, UK). Nutraceutical concentrations and

antioxidant capacity evaluations were based on modified methods determined by Cho et al.

(2004), Cho et al. (2005), Hager et al. (2008), and Prior et al. (2003).

Total Phenolics. Total phenolics were measured using the Folin‐Ciocalteu assay

(Slinkard and Singleton 1977) on a diode array spectrophotometer (8452A; Hewlett

Packard, Palo Alto, CA), with a gallic acid standard and a consistent standard curve based

on sequential dilutions. Samples were prepared with 1 ml 0.2N Folins reagent, 0.8 ml

Na2CO3 (75 g/L) and 0.2 ml of extracted sample with a reaction time of 2 h. Absorbance

221 was measured at 760 nm, and results were expressed as gallic acid equivalents (GAE), on a

per weight basis.

Anthocyanin, Ellagitannin, and Flavonol Analysis. For anthocyanin, ellagitannin, and

flavonol analysis, subsamples (5 ml) of supernatant were evaporated to dryness using a

SpeedVac® concentrator (ThermoSavant, Holbrook, NY) with no radiant heat and

suspended in 1 ml of aqueous 3% formic acid solution. Samples (1 mL) were analyzed

using a Waters HPLC system equipped with a model 600 pump, a model 717 Plus

autosampler, and a model 996 photodiode array detector. Separation was carried out using

a 4.6 mm × 250 mm Symmetry® C18 column (Waters Corp, Milford, MA) with a 3.9 mm ×

20 mm Symmetry® C18 guard column. The mobile phase was a linear gradient of 5%

formic acid and methanol from 2% to 60% for 60 min at 1 ml min−1. Prior to each injection, the system was equilibrated for 20 min at the initial gradient. Detection wavelength was

510 nm for anthocyanins. Individual anthocyanin diglycosides were quantified as Dp, Cy,

Pt, Pn, Pg, and Mv glycoside equivalents. Total anthocyanins were calculated as the sum of

individual glycosides and their derivatives.

For total flavonol and ellagitannin analysis, samples (5 ml) of supernatant were

evaporated to dryness using a SpeedVac® concentrator with no radiant heat and

suspended in 1 ml of aqueous 50% methanol solution. The samples were analyzed using a

Waters HPLC system (Waters Corp, Milford, MA) equipped with a model 600 pump, model

717 plus autosampler and model 996 photodiode array detector. Separation was carried out using a 4.6 mm × 250 mm Aqua® C18 column (Phenomenex, Torrance, CA) with a 3.0 mm × 4.0 mm ODS® C18 guard column (Phenomenex). The mobile phase was a gradient of

20 g kg−1 acetic acid (A) and 5 g kg−1 acetic acid in water and acetonitrile (50:50 v/v, B)

222 from 10% B to 55% B in 50 min and from 55% B to 100% B in 10 min. Prior to each

injection, the system was equilibrated for 20 min at the initial gradient. A detection

wavelength of 360 nm was used for flavonols and 280 nm for ellagitannins at a flow rate of

1 ml min−1. Flavonols and ellagitannins were expressed as mg rutin equivalents kg−1 fresh weight.

Analytical Standard and HPLC/MS. For flavonol and ellagitannin confirmation, a

representative bronze and black genotype were analyzed using mass spectrometry (MS).

For HPLC/MS analysis the HPLC apparatus was interfaced to a Burker Esquire (Burker

Corporation, Billerica, MA) LC/MS ion trap mass spectrometer. Mass spectral data were

collected with the Bruker software, which also controlled the instrument and collected the

signal at 360 or 510 nm. Typical conditions for mass spectral analysis in negative ion

electrospray mode for flavonols included a capillary voltage of 4000 V, a nebulizing

pressure of 30.0 psi, a drying gas flow of 9.0 ml min−1 and a temperature of 300 °C. Data

were collected in full‐scan mode over a mass range of m/z 50 – 1000 at 1.0 s per cycle.

Characteristic ions were used for peak assignment.

Resveratrol (3,4′,5‐Trihydroxy‐trans‐stilbene, 5‐[(1E)‐2‐(4‐

Hydroxyphenyl)ethenyl]‐1,3‐benzenediol) concentrations were confirmed using an analytical standard (ID:24860876; Sigma‐Aldrich Co. LLC, St. Louis, MO).

Oxygen Radical Absorbance Capacity. The ORAC of muscadine extracts was measured using

the method of Prior et al. (2003) modified for use with a FLUOstar Optima microplate

reader (BMG Labtechnologies, Durham, NC) using fluorescein as a fluorescent probe.

Muscadine extracts were diluted 1600‐fold with phosphate buffer (75 mM, pH 7) prior to

ORAC analysis. The assay was carried out in clear 48‐well Falcon plates (VWR, St. Louis,

223 MO), each well having a final volume of 590 μl. Initially, 40 μl of diluted sample, Trolox

equivelants (TE) standards (6.25, 12.5, 25, 50 μM) and blank solution of phosphate buffer

were added to each well. The FLUOstar Optima instrument equipped with two automated

injectors was programmed to add 400 μl of fluorescein (0.108 μM) followed by 150 μl of

2,2’‐azobis(2‐amidino‐propane) dihydrochloride (AAPH) (31.6 mM) to each well.

Fluorescence readings (excitation 485 nm, emission 520 nm) were recorded after the

addition of fluorescein and AAPH and every 192 s for 112 min to reach 95% loss of

fluorescence. Results were based upon differences in areas under the fluorescein decay curve between the blank, samples, and standards, and expressed relative to the initial reading. The standard curve was obtained by plotting the four concentrations of TE against the net area under the curve of each standard. Final ORAC values were calculated using the regression equation between TE concentration and the net area under the curve and expressed as μmol TE equivalents kg−1 fresh weight. ORAC was expressed on a per weight

basis.

Experimental Design

The data were analyzed separately for 2012 and at date of harvest for 2012 and

2013, due to the lack of differences in nutraceutical levels for storage time in 2012. In 2012

the experimental design was a split‐split‐plot, with the first split being storage (weeks 0, 1,

2, 3, 4, 5, and 6) and the second split being berry segment (flesh [skin and pulp], seed, and

whole berry). When the date of harvest data were combined for both years the

experimental design was a split split‐plot, with the first split being year (2012 and 2013)

and the second split being berry segment, with three replications consisting of the harvest

of each vine.

224 Experimental Analysis

The data were analyzed by analysis of variance (ANOVA) using JMP® (version 11.0;

SAS Institute Inc., Cary, NC). Tukey’s Honest Significant Difference and Student’s t Test was used for mean separations (p = 0.05). Associations among all dependent variables were determined using multivariate pairwise correlation coefficients of the mean values using

JMP (version 11.0; SAS Institute Inc., Cary, NC).

Results

Initial Attributes

The initial nutraceutical concentrations and composition of ‘Supreme’ muscadine

berry segments for 2012 and 2013 were presented in Table 3.1. In 2012 the soluble solids

for vines one, two, and three were 19.7, 18.5, and 20.2%, respectively, the pH for vines one, two, and three were 3.8, 3.6, and 3.8, respectively, and the TA for vines one, two, and three were 0.29, 0.32, and 0.25%, respectively. In 2013 the soluble solids for vines one, two, and three were 18.1, 16.9, and 17.9%, respectively, the pH solids for vines one, two, and three were 3.9, 4.2, and 3.9, respectively, and the TA solids for vines one, two, and three were

0.21, 0.15, and 0.15%, respectively.

2012 Nutraceutical Analysis

For the 2012 data, the ANOVA F‐test indicated a significant three‐way interaction of vine by week of storage by berry segment for total anthocyanins (P=0.0001), total

ellagitannins (P=0.0309), and ORAC (P<0.0001) (Table 3.2). The ANOVA F‐test also

indicated a significant two‐way interaction of vine by berry segment for total flavonols

(P<0.0001), total phenolics (P=0.0001), and trans‐resveratrol (P<0.0001), and a significant

two‐way interaction of week of storage by berry segment for total flavonols (P=0.0227)

225

(mg/100 g) Resveratrol

g)

Trolox ORAC (µmol equivalents/

g)

Total (mg/100 phenolics

’ muscadine berry segments averaged across

g) Total (mg/100 flavonols

31.6 40.9 189.3 39.2 5.7 Total (mg/100 g) ellagitannins

ical concentrations of ‘Supreme

z

(mg/100 g)

anthocyanins

Year Segment Total 2013 Whole Flesh 15.7 Seed 10.7 3.1 ND 2.8 10.7 20.2 298.8 7.0 39.9 58.5 155.4 124.1 5.7 47.8 27.6 4.0 6.9 2012 Whole Flesh 23.9 Seed 23.1 2.9 ND 1.9 8.7 368.7 7.7 66.3 209.5 36.5 3.3 2.7

ND=not detected. Table 3.1. The initial nutraceut vines for 2012 and 2013.

z

226 and trans‐resveratrol (P=0.0012) (Table 3.2). Additionally, a significant interaction of week of storage by vine occurred for total phenolics (P=0.0235) and trans‐resveratrol

(P=0.0182) (Table 3.2). The variability of all nutraceuticals during storage of berry segments from different vines offers some insight as to why the interactions were significant.

Total Anthocyanins. Major trends in the data for anthocyanins showed their

presence only in the flesh and whole berries, but not in the seeds (Fig. 3.1). Additionally, it

was found that total anthocyanin concentrations varied among weeks, but no clear pattern

occurred during storage, and no significant differences occurred for mean anthocyanins

across weeks (P=0.1124). There were also no significant differences in mean anthocyanin

concentrations for vines of the study (P=0.2513); however, the muscadine flesh after 3

weeks of storage and whole berries after 4 weeks of storage from vine two and flesh after 5

weeks of storage from vine one had the highest anthocyanin concentrations (71.6, 76.6, and

63.3 mg/100 g, respectively), while whole berries from vine two at harvest had the lowest

(14.8 mg/100 g), excluding seeds (Fig. 3.1 and Table A.12).

Total Ellagitannins. In 2012, it was found among berry segments that the seeds had

the highest concentrations of total ellagitannins, while the flesh and whole‐berry contents

were lower and similar to one another (Fig. 3.2). Though significant differences in total

ellagitannin means among weeks occurred (P=0.0003), no clear pattern showed an

increase or decrease during storage (Fig. 3.2). Additionally, the differences in mean

ellagitannin concentrations that occurred among vines varied, but was not significant

(P=0.8704) (Fig. 3.2). It was found that ellagitannin concentrations in the seed from vine one after 1 week of storage were the highest (43.9), while the flesh after 4 weeks of storage

227

0.0182 0.0012 <0.0001 (mg/100 g) Resveratrol

Total 0.0235 0.0001 0.0533 phenolics

(mg/100 g)

Total 0.0227 0.7964 0.1007 0.0788 <0.0001 flavonols (mg/100 g)

g)

s at 2 °C in Trolox s, total <0.0001 ORAC (µmol equivalents/

s‐resveratrol

Total

0.0309

(mg/100 g) ellagitannins

Total 0.0001 (mg/100 g) nols, total phenolics, and tran anthocyanins

rom ANOVA for total anthocyanin

z

Degrees of freedom. DF Vine Week Week*vine Segment Vine*segment Week*segment Vine*week*segment 12 2 24 6 2 12 2 0.0904 0.2513 0.1124 0.0768 0.0005 <0.0001 0.1564 0.8704 0.0003 <0.0001 0.4823 <0.0001 0.0003 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 0.3525 <0.0001 0.0016 0.0372 <0.0001 0.0003 <0.0001 <0.0001 <0.0001

z

ellagitannins, ORAC, total flavo Table 3.2. F‐test significance f concentrations of ‘Supreme’ muscadine berries stored for 6 week 2012. Highest‐order interactions are italicized and shaded.

228 Fig. 3.1. Total anthocyanin concentrations of ‘Supreme’ muscadine berry segments (flesh = pulp and skin) stored at 2 °C for 6 weeks in 2012. Each standard error bar is constructed using 1 standard error from the mean. Segment 80 Flesh Seed 60 1 Whole 40 20 0 80

60 Vine 2 40 20 0

Total anthocyanins (mg/100 g) 80 60 3 40 20 0 0123456 Week

229 Fig. 3.2. Total ellagitannin concentrations of ‘Supreme’ muscadine berry segments (flesh = pulp and skin) stored at 2 °C for 6 weeks in 2012. Each standard error bar is constructed using 1 standard error from the mean. Segment 50 Flesh 40 Seed

30 1 Whole 20 10 0

50 40 Vine

30 2

annins (mg/100 g) 20 t 10 0 al ellagi t

To 50 40

30 3 20 10 0 0123456 Week

230 and whole berries after 5 weeks of storage from vine one had the lowest ellagitannin

concentrations (0.51 and 0.59 mg/100 g, respectively) (Fig. 3.2 and Table A.12).

ORAC. In examining the major trends in the data, whole berries had the highest

ORAC values, but there was no clear overall increase or decrease in ORAC during storage

(Fig. 3.3). With few exceptions, it was found that ORAC was generally lowest in whole berries and berry segments from vine one, compared to the other vines (Fig. 3.3). The flesh and seed of berries from vine one after 1 week of storage had the lowest ORAC values (20.6

and 22.2 µmol Trolox equivalents/g, respectively), while the whole berries from vine two

at harvest had the highest (81.1 µmol Trolox equivalents/g) (Fig. 3.3 and Table A.12).

Total Flavonols. In examining the major trends in the data, one can see that total

flavonol concentrations were highest in the seeds, and overall the variation among vines

was minimal with the exception of seeds from muscadine berries in vine one (Fig. 3.4).

Additionally, it was found that though variation occurred in total flavonol concentrations

among the berry segments, there was no clear increase or decrease during storage (Fig.

3.5). Averaged across week of storage the seeds of berries from vine one had the greatest

total flavonol concentration (56.8 mg/100 g), while the flesh of berries from vine 1 had the

lowest (8.4 mg/100 g) (Fig. 3.4). Averaged across vines, the seeds after 6 weeks of storage

and at harvest had the highest total flavonol concentrations (41.6 and 40.9 mg/100 g,

respectively), while the whole berries after 3 weeks of storage had the lowest (6.9 mg/100

g) (Fig. 3.5).

Total Phenolics. The major trends in the data showed that though variation did

occur among weeks of storage and vines, it was difficult to identify a clear pattern of

increase or decrease of total phenolic concentrations during storage, or to identify a single

231 Fig. 3.3. Oxygen radical absorbance capacity (ORAC) of ‘Supreme’ muscadine berry segments (flesh = pulp and skin) stored at 2 °C for 6 weeks in 2012. Each standard error bar is constructed using 1 standard error from the mean. Segment 80 Flesh Seed 60 1 Whole 40 20 0 80

60 Vine 2 40 20 ORAC (µmol TE/g) 0 80 60 3 40 20 0 0123456 Week

232 Fig. 3.4. Total flavonols of ‘Supreme’ muscadine berry segments (flesh = pulp and skin) from three different vines of ‘Supreme’ in 2012. Each standard error bar is constructed using 1 standard error from the mean.

Fig. 3.5. Total flavonols of ‘Supreme’ muscadine berry segments (flesh = pulp and skin) stored at 2 °C for 6 weeks in 2012. Each standard error bar is constructed using 1 standard error from the mean.

233

vine that generally had higher phenolics, averaged across berry segments (Fig. 3.6).

Additionally, on a per weight basis, the seeds generally had the lowest total phenolics,

while the whole berries generally had the highest (Fig. 3.7). Averaged across berry

segments, muscadines from vines two and three after 6 weeks of storage had the highest total phenolics (312.2 and 306.9 mg/100 g, respectively), while berries from vine one after

1 week of storage had the lowest (230.3 mg/100 g) (Fig. 3.6). The whole berries from vines two and three had the highest total phenolics concentrations (388.7 and 384.3 mg/100 g, respectively) while seeds from vine one had the overall lowest (166.9 mg/100 g), when averaged across weeks of storage (Fig 3.7).

Resveratrol. Averaged across berry segments, it was difficult to identify a vine that

produced berries with consistently the highest or lowest levels of resveratrol; however,

when looking at the major trends in the data it appeared that after 5 weeks of storage,

resveratrol concentrations in berries from vine three were highest (Fig. 3.8). Vine

one after 1 week of storage had the lowest resveratrol concentrations (2.6 mg/100 g),while

vine three after 5 weeks of storage had the highest (10.3 mg/100 g), when averaged across

berry segments (Fig. 3.8). Overall the seeds from vines two and three had the highest

resveratrol concentrations (10.2 and 7.1 mg/100 g, respectively) when averaged across

weeks of storage, while seeds from vine one had the lowest (2.3 mg/100 g, respectively)

(Fig. 3.9). Furthermore, when averaged across vines, it appeared that resveratrol

concentrations were highest in whole berries, flesh, and seed after 5 or 6 weeks of storage

(7.2, 7.9, 9.2, 10.2, 7.6, and 8.0 mg/100 g, respectively), with the exception of seeds after 1

week of storage (8.5 mg/100 g) (Fig 3.10).

234 Fig. 3.6. Total phenolics of ‘Supreme’ muscadine berries stored at 2 °C for 6 weeks, averaged across berry segment in 2012. Each standard error bar is constructed using 1 standard error from the mean. Vine 350 300 1 250 2 200 3 150

(mg/100 g) 100 Total phenolics 50 0 0123456 Week

Fig. 3.7. Total phenolics of ‘Supreme’ berry muscadine segments (flesh = pulp and skin) from three different vines of ‘Supreme’ in 2012. Each standard error bar is constructed using 1 standard error from the mean. 450 Vine 400 1 350 300 2 250 3 200 150 (mg/100 g) 100 Total phenolics 50 0 Flesh Seed Whole Segment

Fig. 3.8. Resveratrol concentrations of ‘Supreme’ muscadine berries from three different vines of ‘Supreme’ stored at 2 °C for 6 weeks in 2012. Values are averaged across berry segments. Each standard error bar is constructed using 1 standard error from the mean. Vine 12 1 10 2 8 3 6 4 Resveratrol (mg/100 g) 2 0 0123456 Week

235 Fig. 3.9. Resveratrol concentrations of ‘Supreme’ muscadine berry segments (flesh = pulp and skin) from three different vines of ‘Supreme’ in 2012. Each standard error bar is constructed using 1 standard error from the mean.

12 Vine 10 1

rol 2

t 8 3 6 4 Resvera (mg/100 g) 2 0 Flesh Seed Whole Segment

Fig. 3.10. Resveratrol concentrations of ‘Supreme’ muscadine berry segments (flesh = pulp and skin) stored at 2 °C for 6 weeks in 2012. Each standard error bar is constructed using 1 standard error from the mean. Segment 12 Flesh 10

rol Seed t 8 Whole 6

esvera 4 R (mg/100 g) 2 0 0123456 Week

236 Combined Nutraceutical Analysis

Due to the lack of consistent differences in storage time for the dependent variables, the storage aspect of this study was discontinued after the 2012 season. For the combined

2012 and 2013 data, the ANOVA F‐test indicated significant three‐way interactions of year by vine by berry segment for total anthocyanins (P=0.0235), total flavonols (P<0.0001), total phenolics (P=0.0213), and resveratrol (P=0.0011) (Table 3.3). For the dependent variable ORAC, significant two‐way interactions of year by vine (P=0.0325), year by berry segment (P<0.0001), and vine by berry segment (P<0.0001) occurred (Table 3.3). The variability of nutraceuticals in berry segments from different vines among years offers some insight as to why the interactions were significant. The only significant main effect of berry segment was for total ellagitannins (P>0.0001) (Table 3.3).

Total Anthocyanins. When examining the major trends in the data, it appeared that total anthocyanins were detected in the flesh and whole berries in varying concentrations, but not the muscadine seeds, and overall total anthocyanin concentrations were greater in

2012 compared to 2013 (Fig. 3.11). The difference in total anthocyanins among berry flesh and whole berries was not significant. The whole berries from vine three in 2012 had the greatest anthocyanin concentration (37.4 mg/100 g), while the berry flesh from vine one in

2013 had the lowest anthocyanins detected (9.9 mg/100 g) (Fig. 3.11 and Table A.13).

Averaged across berry segments and vines, the overall anthocyanin concentration in 2012 was 15.6 mg/100 g, and 8.8 mg/100 g in 2013. Additionally, vine three was identified as having the greatest anthocyanins, compared to vines one and two (P<0.0001) (Fig. 3.11).

Total Ellagitannins. Overall, the berry seeds were found to contain the highest concentrations of ellagitannins averaged across years and vines (Fig. 3.12 and Tables A.13

237

0.0011 (mg/100 g) Resveratrol

Total

0.0213 phenolics

(mg/100 g)

Total 0.5339 0.7557 0.6168 0.0004 0.3381 0.0021 <0.0001 <0.0001 0.0466 0.0008 flavonols

(mg/100 g)

n /g) s, ORAC (µmol Trolox 0.0325 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 0.0014

equivalents d trans‐

Total 0.9875 0.0696 <0.0001 (mg/100 g) ellagitannins

actions are italicized and

0.0235 flavonols, total phenolics, an anthocyanin rom ANOVA for total anthocyanin s (mg/100 g) Total

z

Degrees of freedom. z Year*vine*segment 4 Year*segment 2 0.0035 0.0778 Vine*segment 4 0.1205 0.9699 DF Year Vine Year*vine Segment 1 2 2 2 <0.0001 0.0125 0.0327 <0.0001 0.1589 0.9813 0.8040 0.0425 0.4272 0.9085 <0.0001 <0.0001 0.0923 0.0091 0.1242 total ellagitannins, ORAC, total shaded. Table 3.3. F‐test significance f resveratrol concentrations of ‘Supreme’ muscadines at harvest i 2012 and 2013. Highest‐order inter

238 Fig. 3.11. Total anthocyanin concentrations of ‘Supreme’ muscadine berry segments (flesh = pulp and skin) at harvest in 2012 and 2013. Each standard error bar is constructed using 1 standard error from the mean. Segment 40 Flesh Seed 30 2012 Whole 20 10

0 Year

40 hocyanins (mg/100 g) t 30 2013

al an 20 t

To 10 0 123 Vine

Fig. 3.12. Total ellagitannin concentrations of ‘Supreme’ muscadine berry segments (flesh = pulp and skin) at harvest averaged across year (2012 and 2013). Each standard error bar is constructed using 1 standard error from the mean. 30 25 20 15 10 (mg/100 g) 5 Total ellagitannins 0 Flesh Seed Whole Segment

239 and 3.4). The difference in ellagitannins means among berry seeds and whole berries was

22.9 mg/100 g, and 23.5 mg/100 g among berry seeds and berry flesh (Tables A.12 and

A.13). There was no significant difference found in ellagitannins among the berry flesh and whole berry means (Table 3.4). Vine and year had no significant main effect on total ellagitannins (Tables 3.3 and 3.4).

ORAC. When examining the major trends in the data, it appeared that whole berries generally had the highest ORAC levels, and ORAC means for vine and year varied (Table

3.5). Whole berries, averaged across vines, were found to have the highest ORAC levels both years of the study (Table 3.5). When averaged across years, it was also found that whole berries had the highest ORAC levels among vines (Table 3.5). Year was identified as a major contributor as a source of variation, highlighted in vine two, which had the highest

ORAC levels in 2012 and the lowest ORAC levels in 2013, when averaged across berry segments, although the differences in the values were not that substantial (Table 3.5).

When examining the vine by berry segment interaction it was found that the flesh of vine one had the highest ORAC levels, while the whole berries of vine two had the highest; conversely there was no significant difference among ORAC values of the berry seeds among the vines (Table 3.5).

Total Flavonols. Similar to total ellagitannins, the berry seeds were found to have the overall highest concentrations of total flavonols (Fig. 3.13). Although levels varied, total flavonol concentrations among years were not significantly different (P=0.9085), averaged across berry segment and vine. Total flavonol concentration differences were minimal for flesh and whole berries. Berry seeds from vine one in 2012 were identified as having the highest concentrations of total flavonols (64.9 mg/100 g), while berry flesh from vine 2 in

240 Table 3.4. Main effects of year, muscadine berry segment (flesh = pulp and skin), and vine means on total ellagitannin concentrations of ‘Supreme’ berry segments (2012 and 2013). Berry Total ellagitannins segment (mg/100 g) Flesh 2.4 bz Seed 25.9 a Whole 2.9 b P value <0.0001

Vine 1 11.5 2 10.2 3 9.6 P values 0.8040

Year 2012 12.1 2013 8.7 P value 0.1589 zMeans followed by the same letter are not significantly different at α= 0.05; separated by Tukey’s HSD.

Table 3.5. Two‐way interactions of year by vine, year by muscadine berry segment (flesh = pulp and skin), and segment by vine means on ORAC levels of ‘Supreme’ (2012 and 2013). Year Vine 2012 2013 1 46.5 abz 47.4 ab 2 49.8 a 42.3 b 3 45.6 ab 44.2 ab P value 0.0373 Year Segment 2012 2013 Flesh 36.5 Dz 47.8 C Seed 39.2 D 27.6 E Whole 66.3 A 58.5 B P value <0.0001 Vine Segment 1 2 3 Flesh 51.7 bz 36.6 c 38.1 c Seed 32.6 c 30.2 c 37.5 c Whole 56.6 b 71.5 a 59.1 b P value <0.0001 zMeans followed by the same letter are not significantly different at α= 0.05; separated by Tukey’s HSD.

241

Fig. 3.13. Total flavonol concentrations of ‘Supreme’ muscadine berry segments (flesh = pulp and skin) at harvest in 2012 and 2013. Each standard error bar is constructed using 1 standard error from the mean.

242 2013 had the lowest (6.3 mg/100 g) (Fig. 3.13 and Table A.13).

Total Phenolics. On a per weight basis, total phenolics were highest in whole berries,

while berry flesh and seeds had the lowest (Fig. 3.14 and Table A.13). These findings show

berry segment was a major contributor as a source of variation, and offers insight as why

the interaction was significant. The whole berries of vine two in 2012 had the highest levels

of total phenolics (390.0 mg/100 g), while the seeds of vine two in 2013 had the lowest

(93.6 mg/100 g). Overall, the berries in 2012 had higher total phenolics compared to the

berries in 2013 (Fig. 3.14). Additionally, the difference in total phenolic concentrations among vines was not significant (P=0.0903). The variability of total phenolics of berry segments among years offers some insight as to why the interactions were significant.

Resveratrol. There was substantial variability in resveratrol levels among vines,

segments, and years (Fig. 3.15 and Table A.13). For example, the whole berries in 2013 had

the highest resveratrol concentrations of vine one (12.1 mg/100 g), while the seeds of

vines two and three had the highest (8.4 and 7.7 mg/100 g, respectively) concentrations

(Fig. 3.15 and Table A.13). Differing from vines two and three, the seeds of vine one had the

highest resveratrol concentrations, compared to the flesh and whole berries, both years of

the study (Fig. 3.15). These findings further illustrate berry segments and years as major

contributors as sources of variability in resveratrol concentrations. Overall, mean

resveratrol concentrations were significantly greater in 2013, compared to 2012

(P=0.0091). No significant differences in resveratrol levels occurred among vines, when

averaged across years and berry segments (P=0.1242). The variability of total resveratrol

of berry segments among years and vines offers some insight as to why the interactions were significant.

243 Fig. 3.14. Total phenolics of ‘Supreme’ muscadine berry segments (flesh = pulp and skin) at harvest in 2012 and 2013. Each standard error bar is constructed using 1 standard error from the mean. Year 400 2012 350 300 2013

250 1 200 150 100 50 0 400 350

300 Vine

250 2 200 150 100 50 al phenolics (mg/100 g) t 0

To 400 350 300

250 3 200 150 100 50 0 Flesh Seed Whole Segment

244 Fig. 3.15. Resveratrol concentrations of ‘Supreme’ muscadine berry segments (flesh = pulp and skin) at harvest in 2012 and 2013. Each standard error bar is constructed using 1 standard error from the mean. Year 12 2012 10 2013

8 1 6 4 2 0 12

10 Vine

8 2

rol (mg/100 g) 6 t 4 2

Resvera 0 12 10

8 3 6 4 2 0 Flesh Seed Whole Segment

245 Correlations

Several nutraceutical measurements strongly correlated to each other. Total anthocyanins were negatively correlated with total ellagitannins (r=‐0.94) and total flavonols (r=‐0.88) (Table 3.5). Total ellagitannins were strongly correlated with total flavonols (r=0.97), and ORAC was positively correlated with total phenolics (r=0.88) (Table

3.5).

Discussion

Although nutraceutical concentrations and antioxidant capacity of muscadine berry and berry segments are well studied (Ballinger et al., 1973; Boyle and Hsu, 1990; Brown,

1940; Conner and MacLean 2013; Ector et al., 1996; Goldy et al., 1987; Lee et al., 2005; Lee and Talcott, 2004; Marshall et al., 2012; Pastrana‐Bonilla et al., 2003; Sandhu and Gu 2010;

Striegler et al., 2005; Stringer et al., 2009; Talcott and Lee, 2002; Threlfall et al., 2007; Yi et al., 2005), little is known about the retention of these measurements during storage, which is discussed in this study. Lee and Talcott (2002) measured total ellagitannins of ‘Carlos’ and ‘Noble’ juice and wine after 20 and 50 d storage. Marshall and Stringer (2014) measured the retention of these concentrations in fresh‐market muscadines after 14 d of storage on a diverse range of genotypes (not including ‘Supreme’), but there is no information for storage over 6 weeks.

Total Anthocyanins. Similar to the findings of this study, Marshall and Stringer

(2014) found that total anthocyanins generally did not change during storage (Fig. 3.1), though variation among genotypes did occur. Total anthocyanins found in whole berries of

‘Supreme’ here were lower than those reported by Conner and MacLean (2013), Pastrana‐

246 Table 3.6. Multivariate correlation coefficients among nutraceutical concentrations of ‘Supreme’ muscadines at harvest averaged across years (2012 and 2013). Total Total ORAC µmol Total Total anthocyanins ellagitannins Trolox flavonols Phenolics (mg/100 g) (mg/100 g) equivalents/g (mg/100 (mg/100 g) g) Total anthocyanins (mg/100 g) Total ‐0.94* ellagitannins (mg/100 g) ORAC µmol 0.63 ‐0.66 Trolox equivalents/g Total ‐0.88* 0.97* ‐0.61 flavonols (mg/100 g) Total 0.65 ‐0.58 0.88* ‐0.52 Phenolics (mg/100 g) Resveratrol ‐0.56 0.56 ‐0.31 0.44 ‐0.08 (mg/100 g) *Significant at P=0.05.

247 Bonilla et al. (2003), Striegler et al. (2005), and Yi et al. (2005). Additionally, Pastrana‐

Bonilla et al. (2003) and Striegler et al. (2005) identified anthocyanins in berry seeds of

‘Supreme’, and Threlfall et al. (2005) found anthocyanins in the seeds of ‘Black Beauty’, which is contrary to the findings of this study. Total anthocyanins of ‘Napoleon’ table grapes remained relatively constant during storage (Cantos et al., 2000). Similar to muscadines, total anthocyanins in raspberry (Rubus sp. L.), strawberry (Fragaria x ananassa Weston.), and watermelon (Citrullus lanatus Thunb.) were stable during storage

(Gil et al., 2006). Conversely, anthocyanin concentrations of lowbush blueberry (Vaccinium angustifolium Aiton.), rabbiteye blueberry (V. ashei Reade.), strawberry, and raspberry increased during storage (Basiouny and Chen, 1988; Kalt et al., 1993, Kalt et al., 1999; Kalt and McDonald, 1996). Total anthocyanins were strongly negatively correlated with total ellagitannins (r=‐0.94) and total flavonols (r=‐0.88), indicating that higher anthocyanin concentrations resulted in lower total flavonols and ellagitannins (Table 3.5). The reason for these strong correlations are unknown. The differences in total anthocyanins among years (Fig. 3.11) may be due to higher temperature and greater sun exposure and therefore greater color development in the 2012 growing season (Table A.1).

Total Ellagitannins. Lee and Talcott (2002) found that total ellagitannins in

muscadine juice and wine remained relatively stable during storage of 50 d, which is

similar to my findings (Fig. 3.2). Contrary to the findings of this study, Marshall and

Stringer (2014) found that total ellagitannins generally increased during storage although

variation among genotypes occurred. Similar to the ‘Supreme’ muscadine berries,

ellagitannins of raspberry were largely unaffected by storage (Mullen et al., 2002). Total

ellagitannins were higher in the seeds, compared to the flesh and whole berries (Table 3.3

248 and Fig. 3.12), which is similar to the findings of Sandhu and Gu (2010). Total ellagitannins

were strongly positively correlated with total flavonols (r=0.97), showing that higher ellagitannins also resulted in higher flavonols (Table 3.5). A possible reason for this correlation could be due to berries with higher concentrations of one compound also having high levels of others.

Total Flavonols. The total flavonol concentrations of ‘Supreme’ is widely unstudied;

however, total flavonol concentrations found were lower than those reported by Marshall

et al. (2012) and Talcott and Lee (2002) for other muscadine genotypes (Figs. 3.4, 3.5, and

3.13). Total flavonols of ‘Supreme’ whole berries were found at higher levels than those found in strawberry, but lower levels than those found in blackcurrant (Ribes nigrum L.), chokeberry (Aronia mitschurinii A.K.Skvortsov & Maitul), cranberry (Vaccinium oxycoccos

L.), and raspberry (Cordenunsi et al., 2002; Hakkinen et al., 1999; Maatta‐Riihinen et al.,

2004). Flavonols of ‘Thompson Seedless’ table grape drastically decreased during extended storage (Spanos and Wrolstad, 1990). Conversely, total flavonols increased during storage of pear (Pyrus L.) (Amiot et al., 1995). Similar to ‘Supreme’ muscadine berries, flavonols of raspberry were largely unaffected by storage (Mullen et al., 2002). Sandhu and Gu (2010) found total flavonols to be highest in the muscadine skin and pulp, conversely my study identified muscadine seeds as having the highest levels of total flavonols (Figs. 3.4, 3.5, and

3.13), which could be due to differences in genotypes studied.

Total Phenolics. Similar to the findings of this study (Fig. 3.6), Marshall and Stringer

(2014) found that total phenolics generally did not change after 14 d of storage, though

variation among genotypes in both studies did occur. Total phenolics found in the seeds

and whole berries of ‘Supreme’ were lower than those reported by Striegler et al. (2005)

249 (Figs. 3.7 and 3.14). Conversely, Pastrana‐Bonilla et al. (2003) reported lower total phenolics in the whole berries, and higher total phenolic concentrations in the seeds than found in this study. Threlfall et al. (2005) found total phenolic concentrations in the seeds of ‘Black Beauty’ (95,338 mg/kg) to be greater than reported in my study. Contrary to my study, phenolics of ‘Thompson Seedless’ table grape decreased during extended storage

(Spanos and Wrolstad, 1990), while Cantos et al. (2000) found that phenolics in ‘Napoleon’ table grapes remained relatively constant during storage. Similar to muscadines, total phenolics in mangos (Mangifera indica L.), muskmelon (Cucumis melo L.), pineapple

(Ananas comosus L.), raspberry, strawberry, and watermelon were relatively stable during storage (Gil et al., 2006, Mullen et al., 2002). Conversely, it had been reported that strawberry (Kalt et el., 1999) and pear (Amiot et al., 1995) accumulated phenolics during storage. Kalt et al. (1999) suggested that storage at ambient or above ambient temperatures can positively affect phenolic metabolism and enhance antioxidant capacity, however maintenance of similar levels of total phenolics during storage found in my study might be explained due to storage at 2 °C, a temperature that likely slowed or stopped metabolic activity (Fig. 3.6). Total phenolics were highest in the whole berries, which is contrary to the findings of Sandhu and Gu (2010), who found total phenolics to be highest in seeds.

ORAC. ORAC levels found in whole berries of ‘Supreme’ were higher while the ORAC

levels found in the seeds was lower than those reported for this cultivar by Striegler et al.

(2005) (Fig. 3.3 and Table 3.5). Threllfall et al. (2005) found ORAC levels in the seeds of the

black muscadine cultivar, Black Beauty (1,100 µmol/Trolox equivelants/g), that were

higher than my findings. The ORAC levels of ‘Supreme’ muscadine grapes reported were

250 higher than those of strawberry, plum (Prunus spp.), and kiwifruit (Actinidia deliciosa L.)

(Wang et al., 1996). Kalt et al. (1999) found ORAC levels of raspberry increased during storage, while similar to this study ORAC of strawberry remained stable during storage

(Fig. 3.3). Conversely, Mullen et al. (2002) found that the antioxidant capacity of raspberry remained constant during storage. It has been suggested that storage at ambient or above ambient temperatures will positively affect phenolic metabolism to enhance the antioxidant capacity (Kalt et al., 1999), however the lack of change of ORAC during storage of this study might be explained through the storage at 2 °C, as mentioned prior for total phenolics. ORAC was positively correlated with total phenolics (r=0.88), showing that phenolic compounds contribute to antioxidant capacity, which supported the findings of

Pastrana‐Bonilla et al. (2003) (Table 3.5).

Resveratrol. Similar to the findings of this study, Marshall and Stringer (2014) found

that trans‐resveratrol varied during weeks of storage (Figs. 3.18 and 3.10). Marshall and

Stringer (2014) found that resveratrol increased in five genotypes evaluated, decreased in

three genotypes, and did not significantly change in six additional genotypes evaluated.

Resveratrol concentrations reported prior in muscadine skins (Magee et al., 2002) were

similar to that of this study (Figs. 3.9 and 3.15). Resveratrol concentrations of the whole

berries of muscadines were similar to those reported by Ector et al. (1996). Cantos et al.

(2000) found that resveratrol concentrations of ‘Napoleon’ table grapes doubled during

storage of 10 d. Resveratrol concentrations of whole berries were greater than those

reported for V. vinifera and V. lubrusca (Jeandet et al., 1991). The varying levels of

resveratrol among years could be potentially explained by the wetter and cooler growing

251 season of 2013, likely resulting in higher disease incidence and therefore increasing resveratrol production (Figs. 3.13 and 3.14 and Table A.1).

Among the sources of variation in my study, segment and not vine or week of storage was the most common source with differences among most dependent variable means. This is a major finding, in that in differentiating the potential value of breeding selections, particularly for postharvest storage potential, as nutraceuticals remained stable.

Further, the differences among years for many dependent variables indicated the importance of multi‐year evaluations of breeding selections for nutraceuticals and antioxidant capacity. Since this study was conducted early in the muscadine breeding program, the findings reported here, including the most critical variables to measure, should lead to improved precision in identifying and releasing improved cultivars for fresh‐ market production with enhanced nutraceutical content and antioxidant capacity.

Literature Cited

Abou‐Agag L.H., M.L. Aikens, E.M. Tavengwa, R.L. Benza, S.R. Shows, H.E. Grenett, and F.M. Booyse. 2001. Polyphyenolics increase t‐PA and u‐PA gene transcription in cultured human endothelial cells. Alcohol Clinical Exp. Res. 25:155‐162.

Amiot, M.J., M. Tacchini, S.Y. Aubert, and W. Oleszeck. 1995. Influence of cultivar, maturity stage, and storage conditions on phenolic composition and enzymatic browning of pear fruits. J. Agr. Food Chem. 43:1132‐1137.

Ballinger, W.E., E.P. Maness, and W.B. Nesbitt. 1973. Anthocyanins of black grapes of 10 clones of Vitis rotundifolia, Michx. J. Food Sci. 38:909‐910.

Basiouny, F.M. and Y. Chen. 1988. Effects of harvest date, maturity, and storage intervals, on postharvest quality of rabbiteye blueberries (Vaccinium ashei Reade.) Proc. Fla. State Hort. Soc. 101:281‐284.

Boyle, J.A., and L. Hsu. 1990. Identification and quantitation of ellagic acid in muscadine grape juice. Amer. J. Enol. Viticult. 41:43‐47.

Brown, W.L. 1940. The anthocyanin pigment of the Hunt muscadine grape. J. Amer. Chem. Soc. 62:2808‐2810.

252

Cantos, E., C. García‐Viguera, S. de Pascual‐Teresa, and F.A. Tomás‐Barberan. 2000. Effect of postharvest ultraviolet irradiation on resveratrol and other phenolics of Cv. Napoleon table grapes. J. Agr. Food Chem. 48:4606‐4612.

Cho, M.J., L.R. Howard, R.L Prior, and J.R. Clark. 2004. Flavanoid glycosides and antioxidant capacity of various blackberry, blueberry, and red grape genotypes determined by high‐ performance liquid chromatography/mass spectrometry. J. Sci. Food Agri. 84:1771‐1782.

Cho, M.J., L.R. Howard, R.L. Prior, and J.R. Clark. 2005. Flavonol glycosides and antioxidant capacity of various blackberry and blueberry genotypes determined by high‐performance liquid chromatography/mass spectrometry. J. Sci. Food Agr. 85:2149‐2158.

Conner, P.J. 2009. A century of muscadine Grape (Vitis rotundifolia Michx.) breeding at the University of Georgia. Acta Hort. 827:481‐484.

Conner, P.J., and D. MacLean. 2013. Fruit anthocyanin profile and berry color of muscadine grape cultivars and Muscadinia germplasm. HortScience 48:1235‐1240.

Cordenunsi, B.R., J.R. Oliveira do Nascimento, M.I. Genovese, and F.M. Lajolo. 2002. Influence of cultivar on quality parameters and chemical composition of strawberry fruits grown in Brazil. J. Agr. Food Chem. 50:2581‐2586.

Creasy, L.L., and M. Coffee. 1988. Phytoalexin production potential of grape berries. J. Amer. Soc. Hort. Sci. 113:230‐234.

Ector, B.J. 2001. Compositional and nutritional characteristics, p. 341‐367. In: F.M. Basiouny and D.G. Himelrick, (eds.). Muscadine Grapes. ASHS Press, Alexandria, VA.

Ector, B.J., J.B. Magee, C.P. Hegwood, and M.J. Coign.1996. Resveratrol concentration in muscadine berries, juice, pomace, purees, seeds, and wines. Amer. J. Enol. Viticult. 47:57‐ 62.

Flora, L.F. 1978. Influence of heat, cultivar, and maturity on the anthocyanin‐3,5‐ diglucosides of muscadine grapes. J. Food Sci. 43:1819‐1821.

Goldy, R.G., W.E. Ballinger, and E.P. Maness. 1986. Fruit anthocyanin content of some Euvitis x Vitis rotundifolia hybrids. J. Amer. Soc. Hort. Sci. 111:955‐960.

Goldy, R.G., W. E. Ballinger, E. P. Maness, and W. H. Swallow. 1987. Anthocyanin content of fruit, stem, tendril, leaf, and leaf petiole in muscadine grape. J. Amer. Soc. Hort. Sci. 112:880‐882.

Gil, M.I., E. Aguayo, and A.A. Kader. 2006. Quality changes and nutrient retention in fresh‐ cut versus whole fruits during storage. J. Agr. Food Chem. 54:4284‐4296.

253 Hager, T.J., L.R. Howard, R. Liyanage, J.O. Lay, and R.L. Prior. 2008. Ellagitannin composition of blackberry as determined by HPLC‐ESI‐MS and MALD‐TOF‐MS. J. Agr. Food Chem. 56: 661‐669.

Hakkinen, S.H., S.O. Karenlampi, L.M. Heinonen, H.M. Mykkanen, and A.R. Torronen. 1999. Content of edible flavonols quercetin, myricetin, and kaempferol in 25 edible berries. J. Agr. Food Chem. 47:2274‐2279.

Hollman, P.C. and M.B. Katan. 1999. Health effects and bioavailability of dietary flavonols. Free Radical Res. 31:S75‐80.

Hudson, T.S., D.K. Hartle, S.D. Hursting, N.P. Nunex, T.T.Y. Wang, H.A. Young, P. Arany, and J.E. Green. 2007. Inhibition of prostate cancer growth by muscadine grape skin extract and the resveratrol through distinct mechanisms. Cancer Res. 67:8396‐8405.

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

Jeandet, P., R. Bessis, and B. Gautheron. 1991. The production of resveratrol (3,4,5’‐ trihydroxystilbene) by grape berries in different developmental stages. Amer. J. Enol. Viticult. 42:41‐46.

Jeandet, P., R. Bessis, M. Sbaghi, P. Meunier, and P. Trollat, 1995: Resveratrol content of wines of different ages: relationship with fungal disease pressure in the vineyard. Amer. J. Enol. Viticult. 46:1‐4.

Kalt, W., C.F. Forney, A. Martin, and R.L. Prior. 1999. Antioxidant capacity, vitamin C, phenolics, and anthocyanins after storage of small fruits. J. Agr. Food Chem. 47:4638‐4644.

Kalt, W. and J.E. McDonald. 1996. Chemical composition of lowbush blueberry cultivars. J. Amer. Soc. Hort. Sci. 121:142‐146.

Kalt, W. R.K. Prange, P.D. Lidster, 1993. Postharvest color development of strawberries: influence of maturity, temperature, and light. Can. J. Plant. Sci. 73:541‐548.

Khanduja, K.L., R.K. Gandhi, V. Pathania, and N. Syla. 1999. Prevention of N‐ nitrosodiethylamine‐induced lung tumorigensis by ellagic acid and quercetin in mice. Food Chem. Toxicology 37:313‐318.

Lee, J‐H., J.V. Johnson, and S.T. Talcott. 2005. Identification of ellagic acid conjugates and other polyphenolics in muscadine grapes by HPLC‐ESI‐MS. J. Agr. Food Chem. 53:6003‐ 6010.

254 Lee, J‐H. and S.T. Talcott. 2004. Fruit maturity and juice extraction influences ellagic acid derivatives and other antioxidant polyphenolics in muscadine grapes. J. Agr. Food Chem. 52:361‐366.

Lesca, P. 1983. Protective effects of ellagic acid and other plant phenolics on benzo[a]pyrene‐induced neoplasia in mice. Carcinogenesis 4:1651‐1653.

Lu, R., and G. Sorreno. 1999. Resveratrol, a natural product derived from grapes, exhibits antiestrogenic activity and inhibits the growth of human breast cancer cells. J. Cell Physiol. 197:297‐304.

Maatta‐Riihinen, K. R., A. Kamal‐Eldin, and A.R. Torronen. 2004. Identification and quantification of phenolic compounds in berries of Fragaria and Rubus species (Family Rosaceae). J. Agr. Food Chem. 52:6178–6187.

Magee, J.B., B.J. Smith, and A. Rimando. 2002. Resveratrol content of muscadine berries is affected by disease control spray program. HortScience 37:358‐361

Marshall, D.A. and S.J. Stringer. 2014. Storage retention of stilbene, ellagic acid, flavonol, and phenolic content of muscadine grape (Vitis rotundifolia Michx.) cultivars. J. Food Chem. Nutr. (In press.)

Marshall, D.A., S.J. Stringer, and J.D. Spiers. 2012. Stilbene, ellagic acid, flavonol, and phenolic content of muscadine grape (Vitis rotundifolia Michx.) cultivars. Pharmaceutical Crops 3:69‐77.

Mertens‐Talcott, S.U., S.T. Talcott, and S.S. Percival. 2003. Low concentrations of quercetin and ellagic acid synergistically influence proliferation, cytotoxicity, and apoptosis in MOLT‐ 4Human leukemia cells. J. Nutr. 133:2669‐2674.

Mullen, W., A.J. Stewart, M.E.J. Lean, P. , G.G. Duthie, and A. Crozier. 2002. Effect of freezing and storage on the phenolics, ellagitannins, flavonoids, and antioxidant capacity or red cranberries. J. Agr. Food Chem. 50:5197‐5201.

Pastrana‐Bonilla, E., C.C. Akoh, S. Sellappan, and G. Krewer. 2003. Phenolic content and antioxidant capacity of muscadine grapes. J. Agr. Food Chem. 51:5497‐5503.

Perkins‐Veazie, P., S. Spayd, B. Cline, and C. Fisk. 2012. Handling and marketing guide for fresh market muscadine grapes. SFRC‐E03: 1‐12.

Prior, R.L. 2004. Absorption and metabolism and anthocyanin: potential health effects. p. 1‐ 19. In: M.S. Meskin, W.R. Bidlack, A.J, Davies, D.A. Lewis, and R.K. Randolph (eds.). : mechanisms of action. CRC Press, Boca Raton, F.L.

Prior, R.L., H. Hoang, L. Gu, X. Wu, M. Bacchiocca, L. Howard, M. Hampschwoodill, D. Haung, B. Ou, and R. Jacob. 2003. Assays for hydrophilic and lipophilic antioxidant capacity [oxygen

255 radical absorbance capacity (ORACF1)] of plasma and other biological and food samples. J. Agr. Food Chem. 51:3272‐3279.

Rendig, S.V., J.D. Symons, J.C. Longhurst, and E.A. Amsterdam. 2001. Effects of , alcohol, and quercetin on coronary resistance and conductance arteries. J. Cardiovascular Pharmacology 38:219‐227.

Sandhu, A.K., and L.W. Gu. 2010. Antioxidant capacity, phenolic content, and profiling of phenolic compounds in the seeds, skin, and pulp of Vitis rotundifolia (muscadine grapes) as determined by HPLC‐DAD‐ESI‐MSn. J. Agr. Food Chem. 58:4681‐4692.

Shi, J., J. Yu, J.E. Pohorly, and Y. Kakuda. 2003. Polyphenolics in grape seeds‐ and functionality. J. Medicinal Foods 6:291‐299.

Spanos, G.A. and R.E. Wrolstad. 1999. Influence of processing and storage on the phenolics composition of Thompson Seedless grape juice. J. Agr. Food Chem. 38:1565‐1571.

Stoner, G.D. and M.A. Morse. 1997. Isothiocyanates and plant polyphenols as inhibitors of lung and esophageal cancer. Cancer Lett. 114:113‐119.

Striegler, R.K., P.M. Carter, J.R. Morris, J.R. Clark, R.T. Threlfall, and L.R. Howard. 2005. Yield, quality, and nutraceutical potential of selected muscadine cultivars grown in southwestern Arkansas. HortTechnology 15:276‐284.

Stringer, S.J., D.A. Marshall, and P. Perkins‐Veazie. 2009. Nutraceutical compound concentrations of muscadine (Vitis rotundifolia Michx.) grape cultivars and breeding lines. Acta. Hort. 841:553‐556.

Takeda, F., M. Starnes Saunders, C.F. Savoy, and T.T. Hatton. 1983. Storageability of muscadines for use as fresh fruit. Proc. Viticult. Sci. Symp. Fla. A and M. Univ. 3:31‐33.

Talcott, S.T., and J‐H. Lee. 2002. Ellagic acid and flavonoid antioxidant content of muscadine wine and juice. J. Agr. Food Chem. 50:3186‐3192.

Thaipong, K, U. Boonprakob, K. Crosby, L. Cisneros‐Zevallos, and D.H. Bryne. 2006. Comparison of ABTS, DPPH, FRAC, and ORAC assays for estimating antioxidant activity from guava fruit extracts. J. Food Composition Anal. 19:669‐675.

Threlfall, R.T. J.R. Morris, and A. Mauromoustakos. 1999. Effect of variety, ultraviolet light exposure, and enological methods on the trans‐resveratrol level of wine. Amer. J. Enol. Viticult. 50:57‐64.

Threlfall, R.T., J.R. Morris, J.F. Meullenet, and R.K. Striegler. 2007. Sensory characteristics, composition, and nutraceutical content of juice from Vitis rotundifolia (muscadine) cultivars. Amer. J. Enol. Viticult. 58:268‐273.

256 Threlfall, R.T., J.R. Morris, L.R. Howard, C.R. Brownmiller, and T.L. Walker. 2005. Pressing effects on yield, quality, and nutraceutical content of juice, seeds, and skins from ‘Black Beauty’ and ‘Sunbelt’ grapes. J. Food Sci. 70:167‐171.

Wang, H., G. Cao, and R.L. Prior. 1996. Total antioxidant capacity of fruits. J. Agr. Food Chem. 44:701‐705.

Williamson, D. and C. Manach. 2005. Bioavailability and bioefficacy of polyphenols in humans II. Mer. J. Clinical Nutr. 81:243‐255.

Yi, W., C.C. Akoh, and J. Fischer. 2005. Study of anticancer activities of muscadine grape phenolics in vitro. J. Agr. Food Chem. 53:8804‐8812.

257 Appendix A: Meteorological data, interaction means, and correlations.

Table A.1. Average monthly maximum and minimum temperatures and total rainfall recorded at the Fruit Research Station; Clarksville, AR (lat. 35°31’58°N and long. 93°24’12’W) (2012 and 2013). Year Month Maximum Minimum Precipitation temperature (°C) temperature (°C) (mm) 2012 January 11.4 0.1 111.84 February 11.8 2.5 65.50 March 21.4 9.8 198.73 April 23.08 11.5 81.86 May 26.6 16.3 18.88 June 32.9 19.4 14.36 July 36.5 22.7 40.56 August 33.9 20.9 62.47 September 28.4 17.4 158.19 October 20.2 9.2 127.21 November 16.2 4.4 23.56 December 11.8 1.9 3.23 2013 January 9.1 ‐0.7 98.85 February 9.9 ‐0.2 70.63 March 2.8 2.0 130.30 April 19.5 8.7 119.37 May 23.9 13.6 163.07 June 29.8 19.0 54.61 July 31.4 19.9 100.35 August 30.4 20.9 178.82 September 30.2 17.9 57.91 October 20.9 10.4 106.18 November 12.6 2.3 103.14 December 6.8 ‐1.6 64.51

258 Table A.12. Interaction means of the berry segment nutraceutical concentrations of total phenolics, total anthocyanins, total ellagitannins, total flavonols, and resveratrol, and the antioxidant capacity (ORAC) for year and week of storage at 2 °C for 6 weeks in 2012. Vine Week Segment Total Total ORAC Total Total Resveratrol anthocyanins ellagitannins (µmol flavonols phenolics (mg/100 g) (mg/100 g) (mg/100 g) TE/g) (mg/100 g) (mg/100 g) 1 0 Flesh 20.4 2.3 46.5 6.5 237.6 3.6 1 0 Seed 0.0 34.4 37.7 64.9 186.7 1.4 1 0 Whole 19.2 2.5 55.3 8.7 367.3 3.8 1 1 Flesh 31.7 3.0 20.6 6.7 201.7 3.0 1 1 Seed 0.0 43.9 22.2 56.4 139.3 2.1 1 1 Whole 24.7 2.4 57.4 9.5 349.9 2.7 1 2 Flesh 19.8 2.9 33.1 7.0 252.9 3.6 1 2 Seed 0.0 11.5 22.6 66.3 150.8 2.6 1 2 Whole 32.9 2.6 56.0 8.6 350.3 4.4 1 3 Flesh 46.7 0.7 33.1 7.5 202.5 4.7 1 3 Seed 0.0 32.6 31.1 47.9 152.3 1.3 1 3 Whole 19.2 1.1 67.9 5.0 396.4 3.3 1 4 Flesh 36.0 0.5 45.8 6.4 213.6 4.8 1 4 Seed 0.0 10.0 27.7 53.6 182.7 4.2 1 4 Whole 20.2 0.0 65.9 14.5 356.1 3.3 1 5 Flesh 63.3 1.2 78.7 11.6 280.7 8.8 1 5 Seed 0.0 17.5 25.8 41.4 176.1 3.4 1 5 Whole 18.7 0.6 78.3 11.2 396.3 4.2 1 6 Flesh 19.5 3.8 42.9 13.3 298.6 8.7 1 6 Seed 0.0 9.5 36.9 67.2 180.9 2.3 1 6 Whole 21.7 3.3 44.7 18.6 322.1 9.7 2 0 Flesh 22.2 2.0 31.5 7.3 183.4 2.4 2 0 Seed 0.0 30.3 36.9 30.7 178.8 8.4 259

Vine Week Segment Total Total ORAC Total Total Resveratrol anthocyanins ellagitannins (µmol flavonols phenolics (mg/100 g) (mg/100 g) (mg/100 g) TE/g) (mg/100 g) (mg/100 g) 2 0 Whole 14.9 2.8 81.1 8.6 390.9 3.3 2 1 Flesh 25.0 3.7 27.9 9.7 216.1 2.1 2 1 Seed 0.0 23.1 42.7 27.7 265.4 17.0 2 1 Whole 27.0 4.5 72.5 11.1 403.5 2.7 2 2 Flesh 35.9 2.2 33.3 11.6 204.1 3.2 2 2 Seed 0.0 10.3 48.6 25.3 236.9 9.0 2 2 Whole 28.4 4.2 65.2 10.0 379.0 3.6 2 3 Flesh 67.6 1.6 32.5 11.0 180.0 3.3 2 3 Seed 0.0 16.1 38.8 20.7 163.3 5.5 2 3 Whole 20.8 1.1 66.9 7.4 399.2 3.3 2 4 Flesh 20.5 2.4 41.5 11.7 208.5 4.9 2 4 Seed 0.0 18.7 37.7 32.2 201.0 7.4 2 4 Whole 71.6 1.4 68.9 9.1 355.2 4.2 2 5 Flesh 32.3 1.9 66.3 7.7 202.2 5.1 2 5 Seed 0.0 22.3 56.0 14.6 211.0 12.3 2 5 Whole 25.0 2.5 60.2 7.9 345.7 8.4 2 6 Flesh 26.6 2.7 58.3 11.6 269.4 6.9 2 6 Seed 0.0 15.0 47.8 30.1 220.0 12.2 2 6 Whole 47.9 5.8 66.2 18.3 447.3 8.0 3 0 Flesh 26.6 1.7 31.4 9.2 207.3 2.0 3 0 Seed 0.0 29.9 43.0 27.1 202.5 7.4 3 0 Whole 37.4 3.3 62.4 8.8 348.0 2.3 3 1 Flesh 38.3 4.0 40.5 10.9 249.0 3.3 3 1 Seed 0.0 19.8 32.7 30.9 223.9 6.5 3 1 Whole 35.3 2.4 60.9 9.0 354.6 2.5 3 2 Flesh 32.3 1.8 59.1 11.5 232.5 2.6 3 2 Seed 0.0 16.3 38.3 27.1 175.2 5.2 3 2 Whole 27.2 1.4 72.9 11.1 405.2 2.1 260

Vine Week Segment Total Total ORAC Total Total Resveratrol anthocyanins ellagitannins (µmol flavonols phenolics (mg/100 g) (mg/100 g) (mg/100 g) TE/g) (mg/100 g) (mg/100 g) 3 3 Flesh 33.5 3.4 41.1 13.4 230.2 3.9 3 3 Seed 0.0 14.2 38.3 27.0 155.6 6.2 3 3 Whole 35.7 1.5 71.4 8.2 385.3 5.7 3 4 Flesh 27.5 2.2 45.4 10.8 213.4 4.7 3 4 Seed 0.0 15.0 32.9 27.4 176.9 4.5 3 4 Whole 33.2 1.2 72.0 7.9 391.4 2.4 3 5 Flesh 24.8 1.9 52.7 12.7 241.8 9.9 3 5 Seed 0.0 24.1 59.7 27.0 239.4 12.0 3 5 Whole 24.5 1.9 61.4 13.0 371.2 8.9 3 6 Flesh 37.2 2.0 58.0 12.5 269.9 7.2 3 6 Seed 0.0 26.0 53.0 27.4 215.6 9.5 3 6 Whole 23.1 2.7 60.8 15.4 435.4 12.9 261

Table A.13. Interaction means of the berry segment nutraceutical concentrations of total phenolics, total anthocyanins, total ellagitannins, total flavonols, and resveratrol, and the antioxidant capacity (ORAC) for year at harvest. Year Vine Segment Total Total ORAC Total Total Resveratrol anthocyanins ellagitannins (µmol flavonols phenolics (mg/100 g) (mg/100 g) (mg/100 g) TE/g) (mg/100 g) (mg/100 g) 2012 1 Flesh 20.4 2.3 46.5 6.5 237.6 3.6 2012 1 Seed 0.0 34.4 37.7 64.9 186.7 1.4 2012 1 Whole 19.2 2.5 55.3 8.7 367.3 3.8 2012 2 Flesh 22.2 2.0 31.5 7.3 183.4 2.4 2012 2 Seed 0.0 30.3 36.9 30.7 178.8 8.4 2012 2 Whole 14.9 2.8 81.1 8.6 390.9 3.3 2012 3 Flesh 26.6 1.7 31.4 9.2 207.3 2.0 2012 3 Seed 0.0 29.9 43.0 27.1 202.5 7.4 2012 3 Whole 37.4 3.3 62.4 8.8 348.0 2.3 2013 1 Flesh 9.9 1.7 56.8 6.7 128.8 3.1 2013 1 Seed 0.0 22.4 27.5 40.6 131.4 8.8 2013 1 Whole 19.9 5.7 57.8 11.5 354.2 12.1 2013 2 Flesh 10.5 4.1 41.8 6.3 143.2 6.7 2013 2 Seed 0.0 19.9 23.4 45.2 93.6 4.2 2013 2 Whole 14.0 1.9 61.8 11.2 273.4 3.2 2013 3 Flesh 11.8 2.6 44.8 8.1 194.3 2.2 2013 3 Seed 0.0 18.3 31.9 34.1 147.2 7.7 2013 3 Whole 13.1 1.7 55.8 9.4 268.7 1.8 262

Conclusions

A major component of this experiment was to determine the important parameters

of storage performance of muscadine genotypes, and in so doing develop a storage protocol

for the University of Arkansas muscadine breeding program. Overall, both percent

unmarketable and percent weight loss increased during storage, showing importance as

storage parameters. Force to penetrate the berry skin generally decreased during storage,

also showing potential as an important postharvest storage parameter, particularly since

some genotypes had significantly less loss in force during storage. Percent change in berry

volume showed no clear pattern during storage, probably due to the variation in individual

berries within each genotype limiting the usefulness of berry volume as a storage

measurement. Composition parameters TA, pH, soluble solids, and SS/TA remained

relatively constant during storage, therefore are not important postharvest storage

measurements to routinely measure in evaluating storage potential. Though no clear correlations were identified, it has been shown that soluble solids can be useful in determining maturity, which has been shown to be related with storage performance

(Ballinger and McClure, 1983; Carroll and Marcy, 1982). Berry color measurements

Chroma and hue angle generally showed no clear pattern during storage, while L* showed a sharp decrease after date of harvest and then remained relatively constant during storage. Therefore, it is potentially valuable to determine L* value at date of harvest and again after storage is complete to evaluate color change during storage. The retention of nutraceutical contents (total phenolics, total anthocyanins, total ellagitannins, total

flavonols, and resveratrol) and antioxidant capacity (ORAC) were found to remain

263 relatively constant during storage with some significant differences among weeks, but no

clear linear pattern of change.

Force to penetrate the berry skin was generally greater, while weight loss and

percent unmarketable was less within fungicide treatments. Field fungicide applications

had no effect on muscadine berry color (Chroma, hue angle, and L*) or berry composition

(pH, TA, SS/TA, and soluble solids). There were some effects of field fungicide applications

on nutraceuticals, however results varied. It was determined that though significant

differences did occur, the efficacy of field fungicide applications to increase muscadine

storability did not justify their use. Overall the genotypes AM 04, AM 26, AM 27, AM 28,

‘Southern Jewel’, and ‘Supreme’ were identified as having the highest storage potential,

while AM 01, AM 15, AM 18, and ‘Tara’ had the least storage potential. The genotypes AM

03, AM 04, AM 27, and ‘Ison’ were identified as having the highest overall nutraceutical

content, while AM 18, AM 28, ‘Supreme’, and ‘Tara’ had the lowest overall nutraceutical

content. Total phenolics, total anthocyanins, and ORAC were found to be highest in the whole berries, while ellagitannins and flavonols were highest in the berry seeds, and resveratrol concentrations varied among berry segments.

Among the sources of variation, genotype was the most common source with differences among most dependent variable means. This is a major finding, in that in differentiating the potential value of breeding selections, particularly for postharvest storage potential, adequate variation for a characteristic or trait is needed. It appears there is substantial variation among genotypes in the program for most variables to select for those with improved or superior values. Further, the differences among years for many dependent variables indicated the importance of multi‐year evaluations of breeding

264 selections for storage potential. Since this experiment was conducted early in the muscadine breeding program, the findings reported here, including the most critical variables to measure, should lead to improved precision in identifying and releasing improved cultivars for fresh‐market production with enhanced postharvest potential and increased nutraceutical concentrations.

265