Citation
Using Response Surface Methodology to Optimize Depolymerization of Procyanidins in Cranberries

Material Information

Title:
Using Response Surface Methodology to Optimize Depolymerization of Procyanidins in Cranberries
Creator:
Song, Kaijie
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (74 p.)

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Food Science and Human Nutrition
Committee Chair:
GU,LIWEI
Committee Co-Chair:
SARNOSKI,PAUL J
Committee Members:
SU,ZHIHUA

Subjects

Subjects / Keywords:
Cranberries ( jstor )
Depolymerization ( jstor )
Dimers ( jstor )
Food ( jstor )
High polymers ( jstor )
High temperature ( jstor )
Monomers ( jstor )
Oligomers ( jstor )
Polymers ( jstor )
Trimers ( jstor )
Food Science and Human Nutrition -- Dissertations, Academic -- UF
cranberries -- depolymerization -- procyanidins
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Food Science and Human Nutrition thesis, M.S.

Notes

Abstract:
Cranberries contain abundant amount of procyanidins; however, 85% of them are non-absorbable polymers. Previous studies suggested that procyanidin polymers can be depolymerized into absorbable oligomers. The objective of this study was to optimize depolymerization reaction and evaluate the effects of reaction conditions on the yield of oligomers. Both purified procyanidin polymers and un-extracted procyanidins in pomace were depolymerized with or without added epicatechin. Purified procyanidin polymers were isolated from cranberries. For depolymerization of purified procyanidin polymers, the predicted yield of oligomers without added epicatechin was 227 microgram/mg PP under the optimum condition of 44.3 min, 85 oC, and 0.96 M HCl. The predicted yield of oligomers with added epicatechin was 1493 microgram/mg PP under optimum conditions of 90 min, 85 oC, 0.65 M HCl, and epicatechin/PP (purified polymers) ratio of 3. For depolymerization of procyanidins in pomace, the predicted yield of oligomers without added epicatechin was 5.30 mg/g pomace under the optimum condition of 70.9 min, 80 oC and 0.89 M HCl. If epicatechin was added, the predicted yield was 25.50 mg/g pomace under optimum conditions of 40 min, 85 oC, 0.84 M HCl and epicatechin/pomace ratio of 40 mg/g. The actual yields from verification experiments were close to the predicted yields. Combination of high temperature and high acid concentration aggravated undesirable side reactions. High temperature, moderate acid concentration and high amount of epicatechin favored depolymerization, while optimum reaction time depended on other factors. This research provided a practical approach to depolymerize both purified procyanidin polymers and un-extracted procyanidins in pomace. The study had the potential to increase bioavailability and bioactivity of procyanidins in cranberries. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (M.S.)--University of Florida, 2014.
Local:
Adviser: GU,LIWEI.
Local:
Co-adviser: SARNOSKI,PAUL J.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2015-08-31
Statement of Responsibility:
by Kaijie Song.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Embargo Date:
8/31/2015
Classification:
LD1780 2014 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

USING RESPONSE SURFACE METHODOLOGY TO OPTIMIZE DEPOLYMERIZATION OF PROCYANIDINS IN CRANBERRIES By KAIJIE SONG A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 201 4 1

PAGE 2

201 4 Kaijie Song 2

PAGE 3

To my family and friends 3

PAGE 4

ACKNOWLEDGMENTS First I would like to thank Dr. Liwei Gu for providing me the opportunity to work in his team. I ’ ve learned a lot under his guidance and mentorship. I would also like to thank my committee members, Dr. Zhihua Su and Dr. Paul Sarnoski, for their guidance and suggestions. I appreciate my lab colleagues, Dr. Keqin Ou, Haiyan Liu, Bo Zhao, Sara Marshall, Dr. Zheng Li, Dr. Amandeep Sandhu, Dr. Tao Zou, Yun Cai , Weixin Wang, Chad Uzdevenes , and Chi Gao for their help and guidance in my resear ch. I would like to thank my girlfriend , Yunfei Weng, for her continuous support and care, and being with me through the roughest times. Lastly, I would like to express my strongest gratitude to my parents for their unconditional love and care. I would not accomplish anything without them. 4

PAGE 5

TABLE OF CONTENTS Page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 7 LIST OF FIGURES .......................................................................................................... 8 LIST OF ABBREVIATIONS ............................................................................................. 9 ABSTRACT ................................................................................................................... 10 CHAPTER 1 LITERATURE REVIEW .......................................................................................... 12 Cranberries ............................................................................................................. 12 Health B enefits ................................................................................................. 12 Procyanidins and O ther P hytochemicals in C ranberries .................................. 12 Depolymerization of P rocyanidins ........................................................................... 14 Hypothesis and Objectives ..................................................................................... 15 2 USING RESPONSE SURFACE METHODOLOGY TO OPTIMIZE DEPOLYMERIZATION OF PURIFIED PROCYANIDIN POLYMERS FROM CRANBERRIES ...................................................................................................... 19 Background ............................................................................................................. 19 Materials and Methods ............................................................................................ 19 Chemicals ......................................................................................................... 19 Preparation of P urified P rocyanidin P olymers .................................................. 20 Depolymerization of P urified P rocyanidin P olymers ......................................... 20 HPLCMSn Analysis of Procyanidins ................................................................ 21 Optimization of D epolymerization U sing R esponse S urface M ethodology ....... 22 Statistical Analysis ............................................................................................ 24 Results and Discussion ........................................................................................... 24 Preparation of P urified P rocyanidin P olymers .................................................. 24 HPLCMSn A nalysis of P rocyanidins ................................................................ 25 Optimization of D epolymerization ..................................................................... 26 Summary ................................................................................................................ 31 3 USING RESPONSE SURFACE METHODOLOGY TO OPTIMIZE DEPOLYMERIZATION OF PROCYANIDINS IN CRANBERRY POMACE ............. 47 Background ............................................................................................................. 47 Materials and Methods ............................................................................................ 47 Chemicals ......................................................................................................... 47 5

PAGE 6

Depolymerization of P rocyanidins in C ranberry P omace ................................. 48 HPLCMSn Analysis of Procyanidins ................................................................ 48 Optimization of D epolymerization U sing R esponse S urface M ethodology ....... 49 Statistical Analysis ............................................................................................ 51 Results and Discussion ........................................................................................... 51 HPLCMSn A nalysis of P rocyanidins ................................................................ 51 Optimization of D epolymerization ..................................................................... 52 Summary ................................................................................................................ 56 4 CONCLUSIONS ..................................................................................................... 69 LIST OF REFERENCES ............................................................................................... 70 BIOGRAPHICAL SKETCH ............................................................................................ 74 6

PAGE 7

LIST OF TABLES Table page 2 1 Relative response factor of procyanidin oligomers and epicatechin standard using fluorescence detector with excitation wavelength 230 nm and emission wavelength 321 nm ............................................................................................ 33 2 2 Experimental design for the optimization of depolymerization of purified procyanidin polymers .......................................................................................... 34 2 3 Eluting solvents, procyanidin composition, and weight of collected f ractions from Sephadex LH 20 column ............................................................................ 35 2 4 Procyanidin composition after depolymerization under optimum conditions. ...... 36 2 5 Predicted and actual yields of oligomers under optimum conditions obtained using multiple response optimization method. .................................................... 37 3 1 Experimental design for the optimization of depolymerization of procyanidins in cranberry pomace ........................................................................................... 58 3 2 Procyanidin composition after extraction and depolymeri zation under optimum conditions ............................................................................................. 59 7

PAGE 8

LIST OF FIGURES Figure page 1 1 Structures of the flavan3 ols units in proanthocyanidins (Gu and others 2003) .................................................................................................................. 17 1 2 Structures of epicatechin, catechin, and procyanidin dimers .............................. 18 2 1 HPLC chromatograms of Fraction 12, 2 2, 3 2, 4 2, and 51 ............................. 38 2 2 HPLC chromatograms of pur ified procyanidins polymer and sample after depolymerization with or without added epicatechin .......................................... 39 2 3 Surface plots for depoly meriz ation without added epicatechin ........................... 40 2 4 Surface plots for depolymerization with added epicatechin ................................ 42 2 5 Prediction profilers for depolymerization with or without added epicatechin ....... 45 2 6 HPLC chromatograms of sample after depolymerization with or without added epicatechin under optimum conditions ..................................................... 46 3 1 HPLC chromatograms of procyanidins extract ed from pomace w ithout HCl or added epicatechin .............................................................................................. 60 3 2 HPLC chromatograms of procyanidins from pomace after depolymerization with or with out added epicatechin ....................................................................... 61 3 3 Surface plots for depolymerization of procyanidins in p omace without added epicatechin ......................................................................................................... 62 3 4 Surface plots for depolymerization of procyanidins i n pomace with added epicatechin ......................................................................................................... 64 3 5 Prediction profilers for depolymerization of procyanidins in pomace with or without added epicatechin .................................................................................. 67 3 6 HPLC chromatograms of procyanisin from pomace after depolymerization with or without added epicatechin under optimum conditions ............................. 68 8

PAGE 9

LIST OF ABBREVIATIONS ANOVA Analysis of variance BIC Bayesian information criterion DP Degree of depolymerization g Gram HPLC High performance liquid chromatography L Liter LDL Low density lipoprotein M Molarity (mol/L) mg Milligram min Minute(s) mL Milliliter MS Mass spectrometer nm Nanometer PP Purified (cranberry procyanidin) polymers psi Pounds per square inch rpm Revolutions per minutes v Volume V Voltage w Weight g Microgram L Microliter 9

PAGE 10

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science USING RESPONSE SURFACE METHODOLOGY TO OPTIMIZE DEPOLYMERIZATION OF PROCYANIDINS IN CRANBERRIES By Kaijie Song August 201 4 Chair: Liwei Gu Major: Food S cience and H uman N utrition Cranberries contain abundant amount of procyanidins; however, 85% of them are non absorbable polymers . Previous studies suggested that procyanidin polymers can be depolymerized into absorbable oligomers. The objective of this study was to optimize depolymerization reaction and evaluate the effects of reaction conditions on the yield of oligomers. Both purif ied procyanidin polymers and un extracted procyanidins in pomace were depoly merized w ith or without added epicatechin. Purified procyanidin polymers were isolated from cranberries . For depolymerization of purified procyanidin polymers, t he predicted yield of oligomers without added epicatechin was 227 g/mg PP under the optimum condition of 44.3 min, 85 oC, and 0.96 M HCl. The predicted yield of oligomers with added epicatechin was 1493 g/mg PP under optimum conditions of 90 min, 85 oC, 0.65 M HCl, and epi catechin/ PP (purified polymers) ratio of 3. For depolymerization of procyanidins in pomace, the predicted yield of oligomers without added epicatechin was 5.30 mg/g pomace under the optimum condition of 70. 9 min, 80 oC and 0.89 M HCl. If epicatechin was ad ded, the predicted yield was 25.50 mg/g pomace under optimum conditions of 40 min, 85 oC, 0.84 M HCl and epicatechin/pomace ratio of 40 mg/g. The actual yields from verification experiments were close to the predicted yields. 10

PAGE 11

Combination of high temperatur e and high acid concentration aggravated undesirable side reactions . H igh temperature, moderate acid concentration and high amount of epicatechin favored depolymerization, while optimum reaction time depended on other factors . This research provided a prac tical approach to depolymerize both purified procyanidin polymers and un extracted procyanidins in pomace. The study had the potential to increase bioavailability and bioactivity of procyanidins in cranberries. 11

PAGE 12

CHAPTER 1 LITERATURE REVIEW Cranberries Cranberries ( Vaccinium macrocarpon), a member of the plant family Ericacea , are native to North America. They have attract ed growing interest s as a functional food due to the potential health benefits associated their phytochemical composition . Health B en efits Cranberries have long been consumed for prevention and treatment of urinary tract infection. The effectiveness was demonstrated by a series of studies ( Su and others 2010; Howell and others 2010 ; Avorn and others 1994) . This property was linked to the ability of cranberry procyanidins to inhibit adhesion of Escherichia coli responsible for the infection ( Howell and others 1998) . Previous in vitro studies also showed that phenolic extracts from cranberries inhibit ed the growth and proliferation of breast, colon, prostate, lung, and other tumors ( Neto 2007) . A growin g body of evidence also suggested that cranberr ies may reduce the risk of cardiovascular diseases by increasing the resistance of LDL to oxidation, inhibiting platelet aggregation and via other anti thrombotic and anti inflammatory mechanisms ( McKay and Blumberg 2007) . Procyanidins and O ther P hytochemicals in C ranberries C ranberries contain significant amount of polyphenolic phytochemicals, a group of secondary metabolites from plants. Polyphenols in cranberries include phenolic acids, anthocyanins, procyanidins , and flavonols ( Chen and others 2001) . Anthocyanins are polyphenolic pigments that are responsible for red color of the fr uit. C yanidin, delphinidin, malvidin, palargonidin and peonidin are the main constituents of anthocyanins in cranberries. The average concentration of anthocyanins in cranberries 12

PAGE 13

was 86 mg/100g edible portion of fruit (wet weight) ( Harnly and others 2006 ; Wu and others 2006) . Q uerceti n and myricetin are the main flavonol s in cranberr ies. The most abundant phenolic acid in cranberries is ferulic acid, while coumaric and ellagic acids were also detected ( Hakkinen and others 1999) . Among these phytochemicals, procyanidins in particular captured most attention since they were postulated to be the bioactive co mpound s for the treatment and prevention of urinary tract infection ( Foo and others 2000) . Proanthocyanidins are oligomeric and polymeric flavan 3 ols. Flavan3 ols are a class of flavonoids that have a 2 phenyl 3,4 dihydro2H chromen 3 ol skeleton. The structures of different flavan 3 ols are shown in F igure 1 1. Proanthocyanidins containing exclusively catechin and/or epicatechin are procyanidins. The structures of procyanidins are demonstrated in Figure 1 2. There are two types of inter flavan bond between flavan 3 ol units. B type bonds are mainly C4 C8 bonds and some less freq uent C4 C6 bonds. An A type bound is linked by an additional ether bond between C2O7. Size of procyanidins is described by Degree of Polymerization (DP). Procyanidins with DP 1, 2, 3, or 4 are called monomers, dimers, trimers, or tetramers. Procyanidin ol igomers are defined as monomers through tetramers with DP 1 4 . are referred as p olymers. Procyanidins with DP>10 are called high polymers. The concentration of procyanidins in cranberries is 418.8 75.3 mg/100g fruit (fresh weight) ( Gu and others 2004) . Procyanidins in cranberries consist of oligomers, polymers and high polymers ( Gu and others 2003) . Previous studies suggested that procyanidin monomers through tetramers were absorbable in vitro and in vivo , whe reas procyanidins larger than tetramers were not absorbable ( Holt and others 2002; Tsang and others 2006; Tomas 13

PAGE 14

Barberan and others 2007) . A type procyanidins account for 51 65% of total procyanidins in cranberries (Gu and others 2004) . It was found that A type procyanidins in cranberries inhibit ed the adherence of uropathogenic E. coli to the uroepithelial cell surfaces whereas the B type isomers had no activity (Foo and others 2000) . Cranberry procyanidins were also reported to be effective in preventi ng and/or treat ing dental caries ( Koo and others 2010) , cardiovascular disease ( Caton and others 2010) , and cancer ( Kresty and others 2011) . Depolymerizat ion of P rocyanidins P rocyanidin polymers are not absorbable. Depolymerization of polymers into absorbable oligomers is expected to increase bioactivity of procyanidins . D epolymerization reaction is catalyzed by a proton under acidic conditions . P roton cleaves the C4 C8 inter flavan bond to generate 4 carbocation intermediates from extension unit s and a flavan3 ol from terminal unit . If a nucleophile , such as epicatechin, is not added into reaction media, carbocations un dergo oxidation and structural rearrangement to form anthocyani di ns . Only the terminal units are released as procyanidin oligomers . If epicatechin is added, it reacts with carbocation from extension units to form new procyanidin oligomers . Awika and other s (2003) found that extrusion of sorghum increased the procyanidin oligomers while decreased the procyanidin polymers. The improvement of procyanidin bioavailability through extrusion was verified in weanling pigs ( Gu and others 2008) . However, extrusion was a harsh processing method and cause d significant procyanidin degradation. Porter and others (1986) depolymeriz ed pro cyanidins using 2.0 M HCl a t 100oC a nd Fe3+ as an oxidant . A ll extension units were converted to cyanidins . Procyanidins released from terminal unit s may also be oxidized 14

PAGE 15

or degraded by Fe3+. Phloroglucinol, toluene thiol, and cysteine had been used as nucleophiles in the depolymerization process (Gu and others 2003; Koerner and others 2009; Foo and Karchesy 1989; Selga and other s 2004) . However, none of these methods were feasible in food industry because final productions are no longer procyanidins . Catechin or epicatechin were used as nucleophiles to depolymerize B type procyanidin polymers for the semi synthesis of B type dimers ( Esatbeyoglu and Winterhalter 2010 ; Koehler and others 2008) . Earlier study in our lab suggested that cranberry procyanidins could be depolymerized into procyanidin oligomers un der acid conditions, using flavan3 ols as nucleophiles. Depolymerization produced procyanidin dimers, trimers and tetramers of A type and B type ( Liu and others 2013) . H ydrogenolysis was also applied for the depolymerization of procyanidins. Foo performed this reaction under high temperature and pressure with palladium as a catalyst ( Foo 1982) . The 4 carbocation from extension unit adducted with hydrogen to produce procyanidins. The depolymerized products were mainly procyanidin oligomers. Although this method could efficiently depolymerize procyanidin polymers into oligomers, the harsh reaction conditions and requirement of a highpressure reaction vessel restrict ed its application. H ypothesis and Objectives Depolymerizing cranberry procyanidins under acid condition with added epicatechin was proven to be an efficient method. We hypothesize that t his depolymerization method ca n be applied on purified procyanidins or un extracted procyanidins in cranberry pomace, a byproduct of cranberry processing. I t is unknown that how the reaction conditions will affect the yield and composition of oligomers. We 15

PAGE 16

further hypothesize that a more efficient de polymerization of procyanidins into absorbable oligomers with optimum yield and favorable composition can be achieved by process optimization and modeling. Therefore, t he objective of this study is to evaluate the impact of depolymerization conditions, inc luding temperature, acid concentration, reaction time and polymer/epicatechin ratio, on the composition and yield of oligomers; and to optimize depolymerization on both purified procyanidin polymers and natural procyanidins in cranberry pomace. 16

PAGE 17

Figure 11. Structures of the flavan3 ols units in proanthocyanidins (Gu and others 2003) . 17

PAGE 18

Figure 12. Structures of epicatechin, catechin, and procyanidin dimers . 18

PAGE 19

CHAPTER 2 USING RESPONSE SURFACE METHODOLOGY TO OPTIMIZE DEPOLYMERIZATION OF PURIFIED PROCYANIDIN POLYMERS FROM CRANBERRIES Background O nly 15% of total procyanidins in cranberr ies are absorbable oligomers , which significantly limits their bioactivity ( Gu and others 2002) . Previous studies in our lab revealed an efficient acid assisted method, with or without added epicatechin as a chain breaker, to depolymerize the polymers into more absorbable and bioavailable oligomers (Liu and others 2013; Ou and Gu 2013) . Purified procyanidin polymer s w ere prepared using a chromatographic method. The objective of this chapter was to evaluate the impact of depolymerization conditions, including temperature, acid concentration, reaction time and epicatechin/PP ratio, on the composition and yield of oligomers and to optimize the depolymerization reaction of purified procyanidin polymers using response surface methodology. Materials and Methods Chemicals Freezedried cranberry powders were provided by Ocean Spray Cranberries, Inc (Lakeville Middleboro, MA). ( ) Epicatechin (>90%), Sephadex LH 20 and 1 pro panol were purchase from Sigma C h emical Co. (St. Louis, MO). Amberlite FPX 66 resin was purchased from Dow Company (Midland, Michigan). Methanol, methylene chloride, 6N hydrochloric acid, acetic acid and sodium bicarbonate were products from Fish er Scienti fic (Pittsburg h , PA). 19

PAGE 20

Preparation of P urified P rocyanidin P olymers F i ve hundred grams of freezedried cranberry powder was blended with 1 L of 60 oC hot water in a blender. The mixture was sonicated for 20 min then filtered through Whatman No.4 filter papers. The residue after filtration was extracted one more time with 1 L of 60 oC hot water assisted with sonication. Extracts were combined and loaded on a glass column packed with Amberlite FPX 66 resins. The column was first washed with 5 L of distille d water to remove sugars. Phytochemicals were recovered by eluting column with 2 L of methanol. T his methanol extract was concentrated using a rotary evaporator under partial vacuum and further dried in a SpeedVac concentrator. Dried extract was resuspended in 30% methanol and loaded on a column packed with Sephadex LH 20, which was pretreated with 30% methanol for 4 h before use. 30% methanol (1.2 L), 60% methanol (1.2 L), 80%methanol (1.2 L), 100% methanol (1.6 L) and 70% acetone (1.2 L) were used to elute the column. Each 400 mL eluent was collected and labeled as a fraction. Sixteen fractions were collected . They were concentrated and dried using rotary evaporator and SpeecVac concentrator. Procyanidin composition in each fractio n was analyzed using HPLC MSn. Fraction 5 1 contained exclusively high polymers with DP>10. This purified procyanidin polymer fraction was used for subsequent depolymerization experiments. Depolymerization of P urified P r ocyanidin P olymers Depolymerization was performed either with or without added epicatechin. A 50% 1 propanol aqueous solution was used as solvent for all the depolymerization experiments. In depolymerization with added epicatechin, purified procyanidin polymer s in Fraction 5 1 w ere dissolved in 50% 1 propanol aqueous solution to obtain a 15 mg/mL stock solution. Two hundred microliter 1 propanol aqueous solution with 13.125 20

PAGE 21

mg/mL epicatechin and 1.2 M HCl was then mixed with 100 l stock solution. In experiments without added epicatechin, purified procyanidin polymers w ere dissolved in 50% 1 propanol aqueous solution to obtain a 10 mg/mL stock solution. One hundred microliter stock solution w as then mixed with 100 l 1 propanol aqueous solution with 1.2 M HCl. R eaction mixtures were incubated in a 57.5 oC water bath for 60 min . R eaction was terminated by adding 1 M NaHCO3 to adjust the pH to 5 ~ 6 . Reaction mixtures w ere immediately injected for HPLC MS analysi s. HPLCMSn Analysis of Procyanidins HPLCMSn analyses were performed on an Agilent 1200 HPLC system (Palo Alto, CA) equipped with a binary pump, an autosampler, a fluorescence detector, and a HCT ion trap mass spectrometer (Bruker Daltonics, Billerica, MA). Samples were dissolved in methanol (5 mg/mL) and 5 l was injected for analysis. Separation of procyanidins was carried out on a Luna Silica column ( 250 mm 4.6 mm i.d., 5 m, Phenomenx, Torrance, CA) at a column temperature of 37 C. The binary mobile phase consisted of (A) methylene chl oride/ methanol/acetic acid/water ( 82:14:2:2, v:v: v : v) and (B) methanol/water/acetic acid (9 6 : 2 :2, v : v : v). The 70 min 20 min, 0.0 11. 7% B linear ; 20 5 0 min, 11.7 25.6% B linear; 50 55 min, 25 87.8 % B linear, 55 65 min , 87.8% B isocratic; 65 70 min , 87.8 0.0 % B linear ; followed by 5 min of post sequence run before the next injection . Excitation and emission wavelengths of the fluorescent detector were 231 and 320 nm, respectively ( Robbins and others 2009) . Electrospray ionization was set at negative mode using nebulizer 4 5 psi, drying gas 10 L/min, drying temperature 350 C, and capillary 4000 V. Mass spectra wer e recorded from m/z 150 to 2000. Auto MS2 was conducted with 80% compound stability and 80% trap drive level. The most abundant ion in full scan was isolated, and its product ion 21

PAGE 22

spectra were recorded. Data was analyzed using Chemstation ( Version B. 01.03, Agilent Technologies, Palo Alto, CA ) and Compass DataAnalsyis (Version 4.0, Bruker Daltonics, Billerica, MA ). The quantification of procyanidins are based on the method developed by Robbins and others ( 2012) . Peaks on the chromatograms were integrated using a flat baseline method (Gu and others 2002) . The relative response factors of each oligomers and epicatechin standard are shown in Table 2 1 . Quantification of procyanidins was based on an ( ) epicatechin standard curve and the established relative response factors (Robbins and others 2009; Robbins and others 2012) . Optimization of D epolymerization U sing R esponse S urface M ethodology For depolymerization with added epicatechin, a central composite design with six center points was chosen to optimize the reaction . Independent v ariables included time, temperature, acid conc entration and epicatechin/ procyanidin polymer (PP) ratio (w/w). For depolymerization without added epicatechin, a 3 3 full factori al design was applied to determine optimum conditions. Time, temperature and ac id concentration were selected as independent variables. Table 2 2 show s the experimental design and levels of variables . A total of 27 experiments for depolymerization without added epicatechin and 30 experiments for depolymerization with added epicatechin were conducted. Experiments were randomized to minimize unexplained variability. A full second order polynomial regression model was fitted to optimize the depolymerization process and predict the maximal yield of oligomers. The model equation was as follow: = X + + + + + + + + + + + + + + +

PAGE 23

= X + + + + + + + + + + (2 2 ) Y1 and Y2 are the response (yield of total oligomers) after depolymerization with or without added epicatechin, respectively. X1, X2, X3 and X4 are temperature (oC) , reaction time (min), acid concentration (M) and epicatechin/PP ratio (w/w) . S lope term (b0), linear terms (b1, b2, b3 and b4), quadratic terms (b11, b22, b33 and b44), interaction terms (b12, b13, b14, b23, b24 and b34) and random error were included in the model as coefficients. A forward stepwise regression with the stopping rule of minimum BIC (Bayesian Information Criterion) was used to simplify the model. The R square, which was the proportion of variance explained by the model, was used to measure the fitness of the model . The optimum yield was predicted by the desirability function based on the fitted model . The profiler of response surface plot was utilized to study the effect s of different variables. The optimum yield was validated by verification experiments under the optimal conditions. Polynomial models w ere also employed to maximize the yield of dimers, trimers, and tetramers using a multiple response approach. For depolym erization with added epicatechin, we set the importance factors for d imers, trimers , and tetramers at 0.8, 1 and 0.7, respectively , according to their reported bioactivity ( Foo and others 2000) . For depolymerization without epicatechin, we set the importance factors for m onomers, dimers, trimers , and tetramers at 0.6, 0.8, 1 and 0.7, respectively. The optimum yield was predic ted by the desirability function based on the fitted model s . The predicted yield was validated using verification experiments under the optimal conditions. 23

PAGE 24

Statistical Analysis All experimental designs and regression analysis were performed on JMP software (Version 10.0, SAS Institute Inc., Cary, NC). Data were presented as the mean standard deviation. Results and Discussion Preparation of P urifi ed P rocyanidin P olymers Cranberry polyphenols include d anthocyanins, procyanidins, flavonols , and phenolic acids. Traditionally, aq u eous acidic acetone or methanol was used for extraction of polyphenols. We used u ltra sound assisted hot water extraction because it was proven to be an efficient and environment friendly method to extract phytochemicals ( Buran and others 2014) . Amberlite FPX 66 was used to remove the sugars . The sugar free extract was then separated into sixteen fractions on Sephadex LH 20 to obtain purif ied procyanidin polymers. It had been reported that t he separation mechanism of pr oc yanidins on Sephadex LH 20 was absorption ( Wang and others 2011 ) . Table 2 3 show s proc yanidin composition and weight of all 16 fractions. Procyanidins were not detected in th e first three fractions eluted by 30% methanol, which was in agreement with previous studies (Liu and others 2011; Wang and others 2011) . HPLC chromatograms of Fraction 1 2, 2 2, 3 2, 4 2 and 5 1 are shown in Figure 2 1 . Oligomers of smaller size were eluted by 60% methanol. As the percentage of methanol increased in the mobile phase , oligomers with bigger size s and polymers appeared in the fractions while smaller oligo mers gradually disappeared. This observation support ed previous finding that affinity between procyanidins and Sephadex LH 20 resin increases as molecular weight increases (Gu and others 2002) . All f ractions after 4 4 contained predominately polymers. Fraction 5 1 was selected as purified 24

PAGE 25

procyanidin polymer s for subsequent depolymerization experiments because it contained predominant ly high polymers. HPLCMSn A nalysis of P rocyanidins The HPLC chromatogram of the purified procyanidin polymers (Fraction 5 1) is shown in Figure 2 2 E . The high polymers with degree of polymerization 10 were eluted as a single peak at 55 min and quantified using response factor of nonamers. Polymers with degree of depolymerization from 5 to 10 were small humps scattered from 32 min to 54 min. T he content of high polymers in Fraction 5 1 was estimated to be 82% (w/w). Oligomers were not detected in fraction 5 1. During depolymerization, the inter flavan bond between extension unit and terminal unit was cleaved by proton under acidic condition, followed by the formation of 4 carbocation intermediate from extension unit and a flavan 3 ol from terminal unit. Depending on whether epicatechin ( a nucleophile) was added in the reaction system, the reaction proceeds in two different pathways. In depolymerizat i on without added nucleophile, the 4 carbocation went through rearrangement to produce anthocyanin. O nly the terminal units were released as ol igomers. When epicatechin was added as a nucleophile in the system, it react ed with the 4 carbocation to form new oligomers with lower molecular weight . The HPLC chromatograms of the procyanidins after depolymerization with or without added epicatechin are shown in Figure 2 2B&C. The large peak representing the high polymers disappeared after depolymerization with added epicatechin and shrank significantly after depolymerization without added epicatechin. Procyanidins had three characteristic fragmentati on pathways: quinone methide (QM) cleavage, retroDiels Alder (RDA) cleavage and heterocyclic ring fission (HRF). MS spectr um of peak 1 25

PAGE 26

show ed a molecular ion of m/z 289 [M H]-. This peak was identified as catechin or epicatechin, which cannot be separated on normal phase HPLC. Peak 2A gave rise to m/z 575 [M H]and product ions of m/z 423 and m/z 285, which were generated after retro Diels Alder fission of the heterocyclic ring and quinone methi o de cleavage, respectively. Therefore, peak 2A was identified as procyanidin A type dimer s. The peaks 2B were identified as procyanidin B type dimer s because of product ions of 425, 475 and 287 from m/z 577 [M H]-. Similarly, the peaks 3 were identified as trimers. Peaks from 24 30 min were identified as procyanidin tetramer s as they gave rise to m/z 1149, 1151, 1153 [M H]-. Optimization of D epolymerization Response surface methodology was employed to study the effects of the reaction conditions on the yield and optimize the yields of oligomers. A cent ral composite design was utilized for the depolymerization with added epicatechin and a full factorial design was used for the depolymerization without added epicatechin. Two second ord er polynomial models were obtained after a forward st epwise regression with t he stopping rule of minimum BIC : = 808 . 60 + 48 . 13 + 316 . 26 + 47 . 77 + 211 . 28 301 . 94 119 . 49 + 133 . 71 = 29 . 12 0 . 10 + 2 . 71 + 13 . 94 0 . 02 + 0 . 02 77 . 11 0 . 71

PAGE 27

corresponded to the centralized variables of time, temperature, acid conce ntration , and epicatechin/PP ratio, respectively. Equation 2 4 was the model obtained for depolymerization without added epicatechin. Y2 was the yield of oligomers including monomers ( g/mg PP) and X1, X2 and X3 were centralized variables of time, temperature, and acid concentration, respectively . T he ANOVA demonstrated that both models were fitted well and had good predictive ability, which was evidenced by the very low F values (p<0.0001). The R square of depolymerization with and without added epicatechin were 0.90 and 0.89, respectively. The profiler of surface plot s was generated to study the effects of variables on the response and their interactions to each other . Two variables were plotted against the response while fixing other variables at cen ter points. Figure 2 3 displays the surface plots for depolymerization without added epicatechin. The effects of temperature and time on the response are shown in Figure 2 3A. The yield of oligomers increased as the temperature increased. When tem perature was low, highest yield was achi eved at longest time. However, when temperature was high, shorter reaction time was favored for h igher yields. Esatbeyoglu and Winterhalter (2010) reported that chalcane flavan 3 ol dimers (gambiriins) were generated as byproduct s of depolymerization. However, this compound was not detected in the present a nd previous studies in our lab (Liu and others 2013; Ou and Gu 2013) . This was because g ambiriins may be un detectable using a fluorescent method. The m/z ratio of gamiriins was also not detected on the mass spectrum. Degradation of procyanidins under basic or acidic conditions had also been reported previously (Gu and others 2002; Laks and Hemingway 1987) . However, 27

PAGE 28

the identification of these degraded side product had not been reported before. These undesirable side reactions might be more significant at high reaction temperature. Figure 23B depicts the effects of acid concentration and time on the response. As the acid concentration increased, the yield first increased then decreased. The interaction between acid concentration and time was similar to that between temperature and time but more signif icant , which could be evidenced by the interaction term X1X3 in E quation 2 4 . Longer time was preferred at low acid concentration while shorter time was favored at high acid concentration, which suggested that high acid concentration aggravated undesirable side reactions as well. P rolonged exposure to harsh reaction conditions such as high temperature and high acid c oncentration might result in significant side reaction like degradation and lower yields. Figure 23C shows the surface plot of temperature and acid concentration against yields of oligomers. As temperature increased, the yields increased. As the acid concentration increased, the yield first increased then decreased. The interaction between these two variables was not significant. This observation contradicted a p revious study in our lab which showed that higher yields of oligomers was achieved at lowest acid concentration of 0.1 M ( Ou and Gu 2013) . Our previous study used methanol as reaction media with highest allowed reaction temperature at 60 oC. The present study used aqueous 1 propanol as reaction media and much higher reaction temperature. The interaction between the acid concentration and other may also contributed to the differences. Figure 24 show s the surface plots of depolymerization with added epicatechin. T he surface plot of temperature and time (Figure 2 4A) illustrated that yield of oligomers increased as temperature increased. Interestingly we found that the yields first 28

PAGE 29

decreased then increased as the reaction time prolonged. Reaction time had a similar effect on the yields when plotted with acid concentration (Figure 2 4C). Sin ce high te mperature and acid concentration promoted undesirable side reactions, increasing reaction time might first favor side reactions over depolymerization then favor the depolymerization. Interaction between temperature and time, or acid concentration and time were not significant. Figure 2 4B demonstrates a significant interaction between temperature and acid concentration. When ac id concentration was low, yield increased as temperature increased. However, if the acid concentration increased further , high er temperature result ed in lower yield as side reaction became predominant at high temperature and high acid concentration. Optimum yield was achieved at a combination of high temperature and middle acid concentration. Effects of acid concentrati on and epicatechin/PP ratio are depicted in Figure 2 4D. Higher amount of added epica techin resulted in higher yield of oligomers because it provided more building blocks for the formation of new oligomers. There was litter interaction between acid concentration and epicatechin/PP ratio. Figure 24E show s the surface plot o f temperature and epicatechin/PP ratio. Both temperature and epicatechin/ PP ratio ha d positive effects on the yield and the effects became more significant as these two variables increased. Thi s observation suggested a positive interaction between temperature and epicatechin/ PP ratio. Figure 2 4F show s little interaction between time and epicatechin/PP ratio. In general, w hen epicatechin was present, the interaction between temperature and epicatechin/PP ratio, and temperature and acid concentration were significant. This was also supported by the interaction terms in Equation 2 3 . 29

PAGE 30

Optimum reaction conditions and the predicted maximal yield were predicted from the models. The prediction profiler s for optimization are shown in F igure 2 5. The predicted yield of oligomers using depolymerization wi thout added epicatechin w as 227 g/mg PP under the optimum condition of 44.3 min, 85 oC , and 0.96 M HCl. The actual yield of ve rification experiment was 221 4 g/mg PP, which match ed the predict ed value. For depolymerization with added epicatechin, pr edicted yield was 1493 g/mg PP under optimum conditions of 90 min, 85 oC, 0.65 M HCl, and epicatechin/pp ratio of 3. The actually yield of oli gomer was 1 660 117 g/mg PP, which was close to the predicted value. Table 24 show s the oligomer composition of samples after depolymerization under optimum reaction conditions. The HPLC chromatograms are also shown in Figure 26 . In the absence of added epicatechin, the yield of A type dimers was 4 fold of B type dimers and the yield of A type trimers was 2 times as much as B type trimers. On the contrary, the yield of B type di mers was about 2 times high er than the yield of B typ e dimers. These findings were in agreement with a previous study ( Ou and Gu 2013) . It was known that 46.1% of the terminal units in the procyanidin polymers of cranberries were A type linkage, which remained intact during depolymerization (Gu and others 2003; Gu and others 2002) . When epicatechin was not added, those terminal units were released as A type dimers or trimers. Those carbocations from extension units undergo rearrangement to produce anthocyanins (Porter and others 1986) . This might account for the higher yield of A type oligomers than B type oligomers when epicatechin was absent. When epicatechin was added as a nucleophile, it react ed with the carbocations deri ved from the extension units to form new procyanidin oligomers. Since the extension units of procyanidins in cranberries 30

PAGE 31

contained more B type interflavan bonds than A type, the yield of B type oligom ers was expected to be higher than that of A type oligomers when epicatechin was added because the extension units were also released as newly formed oligomers. Multiple responses were also fitted to optimize the yield of oligomers wi th each oligomer being assigned an importance factor according to their reported bioactivity. For the depolymerization without added epicatechin, monomers, dimers, trimers and tetramers had importance factor of 0.6 0.8, 1 and 0.7, respectively. For the depolymerization without added epicatechin, dimers, trimers , and tetramers had importance factor of 0.8, 1 and 0.7 respectively. The predicted and actual yield for each response and the optimum co nditions are shown in Table 2 5 . After applying importance factor s, the optim um conditions were close to those obtained using a single response. Howeve r, we could read i ly optimize the reaction conditions for a specific preference of oligomer composition by changing the setting of responses and their importance factors. Summary Purified procyanidins polymer s w ere isolated from cranberries using a twostep chromatographic method. D epolymerization methods with or without added epicatechin were optimized for maximal yield using response surface methodology . The effects of reaction c onditions on yields of oligomers were also evaluated. Our results indicated that high temperature and high acid concentration aggravated undesirable side reactions such as degradation and thus lower ed yield. Higher e picatechin/PP ratio led to better yield of oligomers. The interactions among variables depended on whether epicatechin was added in the reactions. When epicatechin was added , the interactions between temperature and epicatechin/PP ratio, and temperature and acid concentration were signifi cant. When epicatechin was not added , time interact ed significantly with acid 31

PAGE 32

concentration. In general, the optimum yields were achieved at high temperature and moderate acid concentration. Shorter reaction time was favored in the absence of epicatechin and long reaction time was preferred with the presence of epicatechin. By applying multiple responses with different importance factor , we can efficiently optimize the reaction conditions for a specific preference of oligomer composition. 32

PAGE 33

Table 21. Relative response factor of procyanidin oligomers and epicatechin standard using fluorescence detector with excitation wavelength 230 nm and emission wavelength 321 nm. Procyanidin s Relative response factor Monomer s (epicatechin) 1.000 Dimer s 0.62 6 Trimer s 0. 69 5 Tetramer s 0.5 84 Pentamers 0.554 Hexamers 0.471 Heptamers 0.595 Octamers 0.498 Nonamers 0.637 33

PAGE 34

Table 22 . Experimental design for the optimization of depolymerization of purified procyanidin polymers . Variables Experimental design and levels of variables Full factorial design for depolymerization without added epicatechin Central composite design for depolymerization with added epicatechin Time (min) 30, 60 , 90 30, 60 , 90 Temperature ( o C) 30, 57.5 , 85 30, 57.5 , 85 Acid concentration (HCl , mol/L ) 0.1, 0.8 , 1.5 0.1, 0.8 , 1.5 Epicatechin/PP ratio (w/w) Not applied 0.5, 1.75 , 3 Single response Sum of o ligomers excluding monomers Sum of o ligomers including monomers Multiple responses Dimers, trimers and tetramers with importance factors of 0.8, 1 and 0.7, respectively Monomers, dimers, trimers and tetramers with importance factor of 0.6, 0.8, 1 and 0.7, respectively 34

PAGE 35

Table 23 . Eluting solvent s, procyanidin composition, and weight of collected f ractions from Sephadex LH 20 column. Fraction Eluted solvents Procyanidin composition on HPLC Weight (g) 1 1 30% methanol Not detected 3.76 1 2 30% methanol Not detected 1.23 1 3 30% methanol Not detected 0.61 2 1 60% methanol Monomers and dimers 2.06 2 2 60% methanol Monomers through pentamers, high polymers 1.05 2 3 60% methanol Monomer through high polymers 0.87 3 1 80% methanol Dimers through high polymers 1.65 3 2 80% methanol Dimers through high polymers 1.08 3 3 80% methanol Dimers through high polymers 0.79 4 1 100% methanol Trimers through high polymers 0.99 4 2 100% methanol Trimers through high polymers 0.72 4 3 100% methanol Trimers through high polymers 0.39 4 4 100% methanol Polymers and high polymers 0.69 5 1 70% acetone Polymers 1.13 5 2 70% acetone Polymers 0.07 5 3 70% acetone Polymers 0.02 35

PAGE 36

Table 2 4 . Procyanidin composition after depol ymerization under optimum conditions. Procyanidins P rocyanidin composition (g/mg PP ) With epicatechin Without epicatechin 90 min , 85 o C, 0.65 M HCl Epicatechin/ PP ratio = 3 44 min , 85 o C , 0.96 M HCl Monomers 91 7 62 13. 6 0.3 A type dimers 209 28 86.8 1.6 B type d imers 418 29 21. 7 2.3 A type trimers 307 2 2 3 7.0 0.2 B type t rimers 364 4 2 17.9 0. 5 Tetramers 3 60 27 44. 2 0.3 High polymers Not detected 40. 5 0.3 Sum of monomer through tetramer 257 6 141 22 1 4 Sum of dimer through tetramer 1659 7 9 20 8 4 Results are mean standard deviation of duplicate tests . 36

PAGE 37

Table 25 . Predicted and actual yields of oligomers under optimum conditions obtained using multiple response optimization method. Predicted yields ( g/mg PP) Actual yields ( g/mg PP) Depolymerization without added epicatechin at 41 min, 85 o C and 0.97 M HCl. Monomers 1 6.0 13. 6 0.3 Dimers 117 108 4 Trimers 47.4 54.9 2.6 Tetramer 46.1 44. 2 0.4 Depolymerization with added epicatechin at 90 min, 85 o C, 0.82 M HCl and epicatechin/PP ratio of 3. Dimers 872 788 45 Trimers 305 389 35 Tetramers 301 333 15 The actual yields are expressed as mean standard deviation of duplicate tests . 37

PAGE 38

Figure 21. HPLC c hromatograms of Fraction 12, 2 2, 3 2, 4 2 , and 51 . Peaks from 1) 10 min to 30 min are oligomers; 2) 30 min to 54 min are polymers and 3) 54 min to 66 min are high polymers. Note that Y ax e s of these chromatograms are not in the same scale. 38

PAGE 39

Figure 22. HPLC chromatograms of pur ified procyanidins polymer and sam ple after depolymerization with or without added epicatechin . A, purified procyanidins polymers. B, sample after depolymerization with added epicatechin at 60 min, 57.5 oC, 0.8 M HCl and epicatechin/PP ratio of 1.75 . C, sample after depolymerization without added epicatechin at 60 min, 57.5 oC and 0.8 M HCl . 1 , 2A, 2B, 3A, 3B, and 4 denote m onomers , A type dimers , B type dimers , A type trimers , B type trimers , and t etramers , respectively . Note that Y ax e s of chromatograms are not in the same scale. 39

PAGE 40

Figure 23. Surface plots for depolymerization without added epicatechin. 40

PAGE 41

Figure 23 . C ontinued 41

PAGE 42

Figure 24. Surface plots for depolymerization with added epicatechin. 42

PAGE 43

Figure 24 . C ontinued. 43

PAGE 44

Figure 24 . C ontinued 44

PAGE 45

Figure 25. Prediction profilers for depolymerization with or without added epicatechin. A , depolymerization without added epicatechin. B, depolymerization with added epicatechin. 45

PAGE 46

Figure 26 . HPLC chromatograms of sample a fter depolymerization with or without added epicatechin under optimum conditions . A, sample after depolymerization without added epicatechin under optimum conditions. B, sample after depolymerization with added epicatechin under optimum conditions. 1 , 2A, 2B, 3A, 3B, and 4 denote m onomers , A type dimers , B type dimers , A type trimers , B type trimers , and t etramers , respectively . Note that Y ax e s of chromatograms are not in the same scal e 46

PAGE 47

CHAPTER 3 USING RESPONSE SURFACE METHODOLOGY TO OPTIMIZE DEPOLYMERIZATION OF PROCYANIDINS IN CRANBERRY POMACE Background Cranberry pomace is byproduct of the juicing process, which contains mainly seeds, skins, and stems. It is considered as a rich source of value added polyphenols ( Vattem and Shetty 2002; White and others 2010) . White and others (2010) proposed a one pot reaction to release bound procyanidins and depolymerize polymers from cranberry pomace using alkaline hydrolysis. However, only a small portion of the polymers were depolymerized into oligomers, while over 90% were degraded in the presence of alkaline by opening of A ring. This was expected because procyanidins were known to be unstable at a pH>7. Our previous studies suggested that acid assisted depolymerization with or without added epicatechin could be a feasible approach to extract and depolymerize procyanidins in cranber ry pomace in situ (Liu and others 2013; Ou and Gu 2013) . The objective of the present chapter was to apply response surface methodology to evaluate the impact o f reaction conditions on the composition and yield of oligomers and to optimize the extraction and depolymerization of procyanidins in cranberry pomace. Materials and Methods Chemicals C ranberry pomace was provided by Ocean Spray Cranberries, Inc (Lakev ille Middleboro, MA). ( ) Epicatechin (>90%), Sephadex LH 20 and 1 propanol were purchased from Sigma C h emical Co. (St. Louis, MO). Amberlite FPX 66 resin was purchased from Dow Company (Midland, Michigan). Methanol, methylene chloride, 6N 47

PAGE 48

hydrochloric aci d, acetic acid and sodium bicarbonate were products from Fish er Scientific (Pittsburg, PA). Depo lymerization of P rocyanidins in C ranberry P omace Depolymerization was performed with or without added epicatechin. A 50% 1 propanol aqueous solution was used as solvent for all the depolymerization experiments. Pomace was mixed with solvent using a 1 g to 10 mL ratio. In depolymerization with added epicatechin, 0.5 g pomace was suspended in 5 mL aqueous 1 propanol containing 0.8 M HCl and 25 mg/mL epicatechin. In depolymerization without added epicatechin, 0.5 g pomace was suspended in 5 mL aqueous 1 propanol containing 0.8 M HCl. The reaction mixture was incubated i n a water bath at 55 oC for 80 min. The reaction was terminated by adjusting the pH to 5 6 using 1 M NaHCO3 . The mixture was centrifuged at 13500 rpm for 5 min and the supernatant was immediately injected for HPLC MS analysis. Two reactions wit hout HCl and epicatechin added were also performed to extract procyanidins from pomace. Pomace (0.5 g) was suspended in 5 mL aqueous 1 propanol and incubated i n a 30 oC or 80 oC water bath for 30 min. The supernatant after centrifugation at 13500 rpm for 5 min was immediately injected for HPLC MSn analysis. HPLCMSn Analysis of Procyanidins HPLCMSn analyses were performed on an Agilent 1200 HPLC system (Palo Alto, CA) equ ipped with a binary pump, an autosampler, a fluorescence detector, and a HCT ion trap mass spectrometer (Bruker Daltonics, Billerica, MA). Samples were dissolved in methanol (5 mg/mL) and 5 l was injected for analysis. Separation of procyanidins was carried out on a Luna Silica column ( 250 mm 4.6 mm i.d., 5 m, Phenomenx, Torrance,CA) at a column temperature of 3 7 C. The binary mobile phase 48

PAGE 49

consisted of (A) methylene chloride/ methanol/acetic acid/water ( 82:14:2:2, v:v: v : v) and (B) methanol/water/acetic acid (9 6 : 2 :2, v : v : v). The 70 min 20 min, 0.0 11. 7% B linear ; 20 5 0 min, 11.7 25.6% B linear; 50 55 min, 25 87.8 % B linear, 55 65 min , 87.8% B isocratic; 65 70 min , 87.8 0.0 % B linear ; followed by 5 min of post sequence run before the next injection . Excitation and emission wavelengths of the fluorescent detector were 231 and 320 nm, respectively (Robbins and others 2009) . Electrospr ay ionization was set at negative mode using nebulizer 4 5 psi, drying gas 10 L/min, drying temperature 350 C, and capillary 4000 V. Mass spectra wer e recorded from m/z 150 to 2000. Auto MS2 was conducted with 80% compound stability and 80% trap drive level. The most abundant ion in full scan was isolated, and its product ion spectra were recorded. Data was analyzed using Chemstation ( Version B. 01.03, Agilent Technologies, Palo Alto, CA ) and Compass DataAnalsyis (Version 4.0, Bruker Daltonics, Billerica, M A ). The quantification of procyanidins are based on the method developed by Robbins and others ( 2012) . Peaks on the chromatograms were integrated using a flat baseline method (Gu and others 2002) . The relative response factors between each oligomers and epicatechin standard for our own HPLC system are shown in Table 2 1 . Q uantification of procyanidins was based on an ( ) epicatechin standard curve and the established relative response factors (Robbins and others 2009; Robbins and others 2012) . Optimization of D epolymerization U sing R esponse S urface M ethodology For depolymerization with added epicatechin, a central composite design with five center points was chosen to optimize the reaction. Inde pendent variables included time, temperature, acid concentration and epicatechin/PP ratio (w/w). For depolymerization without added epicatechin, a 3 3 full factorial design was applied to 49

PAGE 50

determine optimum conditions. Time, temperature and acid concentrati on were selected as independent variables. Table 3 1 summarizes the experimental design and levels of variables. A total of 27 experiments for depolymerization without added epicatechin and 29 experiments for depolymerization with added epicatechin were co nducted. Both experimental designs were randomized to minimize unexplained variability. A full second order polynomial regression model was fitted to optimize the depolymerization reactions and predict the maximal yield of oligomers. The model equation was as follow: = X + + + + + + + + + + + + + + + = X + + + + + + + + + + were included in the model as coefficients. A forward stepwise regression with the rule of minimum BIC was used to simplify the model. The R square, which is the proportion of variance explained by the model, was used to determine the fitness of the model . The optimum yield was predicted based on the fitted model. The profiler of response surface plot was utilized to study the effects of variables on the response. The optimum predicted yield was validated by verification experiments at the optimal conditio ns. 50

PAGE 51

Statistical Analysis All experimental designs and regression analysis were performed on JMP software (Version 10.0, SAS Institute Inc., Cary, NC). Data were presented as the mean standard deviation. Results and Discussion HPLCMSn A nalysis of P rocya nidins The HPLC chromatograms of the extracted procyanidins from pomace are shown in Figure 3 1 . Without HCl and added epicatechin, procyanidins were extracted from pomace without depolymerization. When the extraction was performed at 30 oC, only a small amount of procyanidins was extracted (Figure 31 A). The procyanidins extracted at 80 oC (Figure 3 1 B) were about 3 fold of those extracted at 30 oC. Procyanidins had a strong affinity to the plant cell wall ( Le Bourvellec and others 2004; Hellstrom and Mattila 2008; Arranz and others 2009) . The procyanidins in pomace are tightly bound to plant matrix since thos e loosely bound procyanidins have been extracted in the juicing process. The 80 oC extract predominantly contained high polymer with a concentration of 10.95 mg/g pomace. A type dimers were the most abundant oligomers in pomace. This observation was in agreement with a previous study whic h reported that cranberry pomace contained high level of A type oligomers (White and others 2010) . Aqueous 1 propanol appeared to be a suitable solvent to ex tract procyanidins from pomace. Higher temperature was favored for extraction of procyanidins from pomace. The reaction mechanism of depolymerization with or without added epicatechin had been described in Chapter 2. The HPLC chromatograms of the procyani dins after depolymerization with or without added epicatechin are shown in Figure 32. Compared to the extracted procyanidins from pomace as shown in Figure 3 1 B, the large peak 51

PAGE 52

representing the high polymers disappeared after depolymerization with added epicatechin and shrank significantly after depolymerization without added epicatechin. MS spectrum of peak 1 showed an ion of m/z 289 [M H]-. This peak was identified as catechin or epicatechin, which cannot be separated on normal phase HPLC. Peak 2A gave rise to m/z 575 [M H]and product ions of m/z 423 and 285, which were generated after retro Diels Alder fission of the heterocyclic ring and quinone methiode cleavage, respectively. Peak 2A was identified as procyanidin A type dimers. The peaks 2B were identified as procyanidin B type dimers due to product ions m/z 425, 475 and 287 formed from 577 [M H]-. Similarly, the peaks 3 were identified as trimers. Peaks in 24 30 min were i dentified as procyanidin tetramers as they gave rise to m/z 1149, 1151, 1153 [M H]-. Optimization of D epolymerization R esponse surface methodology was utilized to study the effects of the reaction conditions on the yield and optimize the yields of oligom ers. A central composite design was applied for the depolymerization with added epicatechin and a full factorial design was used for the depolymerization without added epicatechin. Two second order polynomial models were obtained after a forward stepwise r egression with the rule of minimum BIC (Bayesian Information Criterion): = 13 . 17 + 0 . 12 + 5 . 20 + 0 . 81 + 3 . 05 + 3 . 98 3 . 92 5 . 73 0 . 82 + 3 . 47 = 1 . 24 0 . 002 + 0 . 046 + 0 . 57 0 . 005 2 . 22

PAGE 53

X4 corresponded to the centralized variables of time, temperature, acid concentration and epicatechin/polym er ratio (mg/g pomace), respectively. Equation 3 4 was the model obtained for depolymerization without added epicatechin. Y2 was the yield of oligomers including monomers (mg/g pomace) and X1, X2, X3 and X4 were centralized variables of time, temperature and acid concentration. T he ANOVA demonstrated that both models were fitted well and had good predictive ability with low F values (P<0.00001). The R square of depolymerization with and without added epicatechin were 0.95 and 0.88, respectively. The profiler of surface plots was employed to study the effects of variables on the response and their interactions to each other. Two variables were plotted against the response while fixing other variables at center points. Figure 3 3 shows the surface plots for depolymerization without added epicatechin. Figure 3 3A depicts the effects of temperature and time on the response. The yield of oligomers increased as the temperature increased. As reaction time increased, the yield first increased then decreased. Shorter reaction time was preferred at high temperature. Compared with the depolymerization of purified polymer without added epicatechin, the effect of reaction time was more significant on the yield. Figure 3 3B demonstrates the effects of acid concentration and time on the response. When the acid concentration and reaction time increased at the same time , the yield first increased then decreased. The interaction between acid concentration and time was not signif icant. This observation was consistent with that in previous chapter which showed high acid concentration aggravated degradation and other side reactions. Figure 3 3C shows the surface plot of temperature and acid concentration against yield. As temperatur e increased, the yields 53

PAGE 54

increased. As the acid concentration increased, the yield first increased then decreased. There was no interaction between temperature and acid concentration. In general, the patterns of the effect of variables on the response were similar between depolymerization of purified polymers and depolymerization in pomace. The optimum yield was also achieved at a combination of highest temperature and moderate acid concentration. The optimum reaction time was longer since binding of procyanidins onto the plant matrix required a longer reaction time for the extraction and depolymerization to complete. There was little interaction amon g the variables for depolymerization of pomace without added epicatechin, which was evidenced by the fact s that there was no interaction term in equation 3 4 . Unlike depolymerization of purified polymers, depolymerization of procyanidins in pomace involved extraction and depolymerization at the same time. Higher temperature should benefit both extraction and depolymerization. However, the optimum reaction time and acid concentration might not be identical for extraction and depolymerization. The optimum condition s were a trade off between the optimizations of extraction and depolymerization. Surface plots of depolymerization with added epicatechin are shown in Figure 3 4. The surface plot of temperature and time (Figure 34A) demonstrates that yield of olig omers increased as temperature increased. As the reaction time prolonged, the yields first decreased then increased. Figure 34B demonstrates the effects of temperature and acid concentration. Similarly, optimum yield was achieved at a combination of high temperature and moderate acid concentration. When acid concentration was low, yield increased as temperature increased. However, as the acid concentration further increased, high temperature resulted in lower yield due to 54

PAGE 55

degradation or other side reactions. Figure 3 4C shows the effects of acid concentration and reaction time. The yield first decreased then increased as reaction time prolonged. Although a longer reaction time was applied in the model for complete extraction and fractionation, the optimum y ield for depolymerization with added epicatechin was achieved at shortest time (40 min), which corresponded with depolymerization of purified polymers with added epicatechin. Interaction effects between temperature and time, and acid concentration and time were not significant. Effects of acid concentration and epicatechin/pomace ratio are illustrated in Figure 3 4D. Higher amount of epicatechin resulted in higher yield of oligomers because it provided more building units for the formation of new oligomers. No interaction was observed between acid concentration and epicatechin/pomace ratio. Figure 24E shows the surface plot of temperature and epicatechin/pomace ratio. When temperature was low, increasing the amount of epicatechin had little effect on the yi eld. As the temperature increased, the yield increased significantly as epicatechin/pomace increased. This observation suggested a positive interaction effect between temperature and epicatechin/pomace ratio. Figure 2 4F shows the surface plot of time and epicatechin/pomace ratio. The yield first decreased then increased as time prolonged. Higher amount of epicatechin was favored for higher yield. In summary, when epicatechin was added, the effects of reaction conditions on yield of depolymerization of pomace were very similar to those of depolymerization of purified polymers. Optimum reaction conditions and the predicted maximal yield were attained from the models. The prediction profilers for optimization are shown in Figure 3 5.The predicted yield of oli gomers using depolymerization without added epicatechin was 5.30 55

PAGE 56

mg/g pomace under the optimum condition of 70. 9 min, 80 oC and 0.89 M HCl. The actual yield of verification experiment was 4.72 0.80 mg/g pomace, which was close to predicted value. For depolymerization with added epicatechin, predicted yield was 25.50 mg/g pomace under optimum conditions of 40 min, 85 oC, 0.84 M HCl and epicatechin/pomace ratio of 40 mg/g. The actual value of oligomer yield was 26.21 2.30 mg/g pomac e, which agreed with predicted value as well. Table 3 2 show s the oligomer s composition of samples after depolymerization under optimum reaction conditi ons. The HPLC chromatograms are also shown in Figure 3 6 . In the absence of added epicatechin, the yield of A type dimers was 1.72 mg/g pomace, which was about double of B type dimers. The yield of A type trimers (0.33 mg/g pomace) was also two times higher than B type trimers (0.17mg/g pomace). On the contrary, when epicatechin was added for depolymerization, the yield of B type dimers was about 2 fold of A type dimers. The yields of A type and B type trimers were almost the same. This pattern was consistent with our previous observation on depolymerization of purified polymers, as described in Chapter 2. Summary Procyanidins in pomace were efficiently extracted and depolymerized using an acid assisted one pot method, with or without added epicatechin. These methods were optimized for maximal yield using response surface methodology. The effects of reaction conditi ons on yields of oligomers were evaluated. When epicatechin was added, the patterns of effects of variables were similar to those of depolymerization of purified polymers. In the absence of epicatechin, there was little interaction among variables. In general, the optimum yields were achieved at high temperature and moderate acid concentration. Moderate reaction time around 70 min was favored in the 56

PAGE 57

absence of epicatechin and shortest reaction time was preferred with the presence of epicatechin. 57

PAGE 58

Table 31. Experimental design for the optimization of depolymerization of procyanidins in cranberry pomace. Variables Experimental design and levels of variables Full factorial design for depolymerization without added epicatechin Central composite design for depolymerization with added epicatechin Time (min) 40, 80 , 12 0 40, 80 , 12 0 Temperature ( o C) 30, 55 , 80 30, 55 , 80 Acid concentration ( HCl , mol/L ) 0.1, 0.8 , 1.5 0.1, 0.8 , 1.5 Epicatechin/pomace rati o (mg/g ) Not applied 10, 25, 40 Single response Sum of o ligomers excluding monomers Sum of o ligomers including monomers 58

PAGE 59

Table 3 2. Procyanidin composition after extraction and depolymerization under optimum conditions. Procyanidins Procyanidin composition (mg/g pomace) Extraction only With added epicatechin Without epicatechin 30 min, 80 o C , No acid 40 min , 80 o C , 0.84 M HCl , epicatechin/pomace=40 mg/g 71 min , 80 o C , 0.89 M HCl Monomers 0.0 75 0.02 24.8 2.7 0.4 3 0.06 A type dimers 0.50 0.01 4.76 0.39 1.72 0.28 B type d imers 0. 3 4 0.06 9 . 70 0.67 0.95+0.20 A type t rimers 0. 12 0.02 2.82 0.35 0. 33 0.11 B type trimers 0.29 0.05 2.98 0.20 0.17 0.04 Tetramers 0. 52 0.08 5. 95 0.44 1.68 0.11 High polymers 9.86 1.53 Not detected 1.22 0.03 Sum of monomer through tetramer 1.84 +0.03 51 .1 3.2 5.29 0.16 Sum of dimer through tetramer 1.7 6 0.01 26.2 2.3 4.85 0.10 Sum of monomer through tetramer and high polymer 1 1 . 7 1.6 51 .1 3.2 6.51 0.18 Results are mean standard of duplicat e test s. 59

PAGE 60

Figure 31. HPLC chromatograms of procyanidins extract ed from pomace without HCl or added epicatechin. A, procyanidins extracted from pomace at 30 oC without HCl or added epicatechin. B, procyanidins extracted from pomace at 80 oC without HCl or added epicatechin. 1 , 2A, 2B, 3A, 3B, and 4 denote m onomers , A type dimers , B type dimers , A type trimers , B type trimers , and t etramers , respectively . Note that Y ax e s of chromatograms are not in the same scale. 60

PAGE 61

Figure 32. HPLC chromatograms of procyanidins from pomace after depolymerization with or with out added epicatechin. A, procyanidins from pomace after depolymerization without added epicatechin at 80 min, 55 oC and 0.8 M HCl . B, procyanidins from pomace after depolymerization with added epicatechin at 80 min, 55 oC, 0.8 M HCl and epicathechin/pomace ratio of 25mg/g ( B ) . 1 , 2A, 2B, 3A, 3B, and 4 denote m onomers , A type dimers , B type dimers , A type trimers , B type trimers , and t etramers , respectively . Note that Y ax e s of chromatograms are not in the same scale. 61

PAGE 62

Figure 33. Surface plots for depolymerization of procyani dins in pomace without added epicatechin. 62

PAGE 63

Figure 33. Continued. 63

PAGE 64

Figure 34. Surface plots for depolymerization of procyanidins in pomace with added epicatechin. 64

PAGE 65

Figure 34. Continued. 65

PAGE 66

Figure 34. Continued. 66

PAGE 67

Figure 3 5. Prediction profil ers for depolymerization of procyanidins in pomace with or without added epicatechin. A, depolymerization of procyanidins in pomace with added epicatechin. B, depolymerization of procyanidins in pomace without added epicatechin. 67

PAGE 68

Figure 36 . HPLC chromatograms of procyani d in from pomace after depolymerization with or without added epicatechin under optimum conditions . A , procyanidins from pomace af ter depolymerization with added epicatechin under optimum condition . B , procyanidins from pomace after depolymerization without added epicatechin under optimum condition. 1 , 2A, 2B, 3A, 3B, and 4 denote m onomers , A type dimers , B type dimers , A type trimer s, B type trimers , and t etramers , respectively . Note that Y ax e s of chromatograms are not in the same scale. 68

PAGE 69

CHAPTER 4 CONCLUSIONS P rocyanidin polymers were purified from cranberries and depolymerized under acid conditions with or without added epicatechin. Procyanidins in pomace were extracted and depolymerized simultaneously using an acid assisted one pot method, with or without added epicatechin. Both depolymerization reactions were optimized for maximal yield using response surface methodology. H igh temperature and high acid concentration aggravated undesirable side reactions such as degradation and lowered yield. A combination of h igh temperature, moderate acid concentration and high amount of epicatechin favored depoly merization, while optimum reaction time depended on other factors . 69

PAGE 70

LIST OF REFERENCES Arranz S, Saura Calixto F, Shaha S, Kroon PA. 2009. High contents of nonextractable polyphenols in fruits suggest that polyphenol contents of plant foods have been underestimated. Journal of Agricultural and Food Chemistry 57(16):7298 303. Avorn J, Monane M, Gurwitz JH, Glynn RJ, Choodnovskiy I, Lipsitz LA. 1994. R eduction of bacteriuria and pyuria after ingestion of cranberry juice. Journal of the American Medical Association 271(10):751 4. Awika JM, Dykes L, Gu L , Rooney LW, Prior RL. 2003. Processing of sorghum (Sorghum bicolor) and sorghum products alters procyanidin oligomer and polymer distribution and content. Journal of Agricultural and Food Chemistry 51(18). Buran TJ, Sandhu AK , Li Z, Rock CR, Yang WHW, Gu L . 2014. Adsorption/desorption characteristics and separation of anthocyanins and polyphenols from blueberries using macroporous adsorbent resins. Journal of Food Engineering 128: 16773. Caton PW, Pothecary MR, Lees DM, Khan NQ, Wood EG, Shojj T, Kanda T, Rull G, Corder R. 2010. Regulation of vascular endothelial function by procyanidin rich foods and beverages . Journal of Agricultural and Food Chemistry 58(7):4008 13. Chen H, Zuo YG, Deng YW. 2001. Separation and determination of flavonoids and other phenolic compounds in cranberry juice by highperformance liquid chromatography. Journal of Chromatography A 913(1 2):387 95. Esatbeyoglu T, Winterhalter P. 2010. P reparation of dimeric procyanidins b1, b2, b5, and b7 from a polymeric procyanidin fraction of black chokeberry ( aronia melanocarpa) . Journal of Agricultural and Food Chemistry 58(8). Foo LY. 1982. P olymeric proanthocyanidins of photinia glabrescens, modification of mol ecular weight and nature of products from hydrogenolysis . Phytochemistry 21(7). Foo LY, Karchesy JJ. 1989. P rocyanidin polymers of douglas fir bark structure from degradation with phloroglucinol . Phytochemistry 28(11). Foo LY, Lu Y, Howell AB, Vorsa N. 2000 . A type proanthocyanidin trimers from cranberry that inhibit adherence of uropathogenic p fimbriated escherichia coli . Journal of Natural Products 63(9):1225 8. Gu L , Kelm M, Hammerstone JF, Beecher G, Cunningham D, Vannozzi S, Prior RL. 2002. Fractionation of polymeric procyanidins from lowbush blueberry and quantification of procyanidins in selected foods with an optimized normal phase HPLCMS fluorescent detection method. Journal of Agricultural and Food Chemistry 50(17). 70

PAGE 71

Gu L , Kelm MA, Hammerstone J F, Beecher G, Holden J, Haytowitz D, Prior RL. 2003. Screening of foods containing proanthocyanidins and their structural characterization using LC MS/MS and thiolytic degradation. Journal of Agricultural and Food Chemistry 51(25). Gu L , Kelm MA, Hammersto ne JF, Beecher G, Holden J, Haytowitz D, Gebhardt S, Prior RL. 2004. Concentrations of proanthocyanidins in common foods and estimations of normal consumption. Journal of Nutrition 134(3):613 7. Gu L, House SE, Rooney LW, Prior RL. 2008. Sorghum extrusion increases bioavailability of catechins in weanling pigs. Journal of Agricultural and Food Chemistry 56(4). Hakkinen S, Heinonen M, Karenlampi S, Mykkanen H, Ruuskanen J, Torronen R. 1999. Screening of selected flavonoids and phenolic acids in 19 berries. F ood Research International 32(5):345 53. Harnly JM, Doherty RF, Beecher GR, Holden JM, Haytowitz DB, Bhagwat S, Gebhardt S. 2006. Flavonoid content of US fruits, vegetables, and nuts. Journal of Agricultural and Food Chemistry 54(26):9966 77. Hellstrom JK, Mattila PH. 2008. HPLC determination of extractable and unextractable proanthocyanidins in plant materials. Journal of Agricultural and Food Chemistry 56(17):7617 24. Holt RR, Lazarus SA, Sullards MC, Zhu QY, Schramm DD, Hammerstone JF, Fraga CG, Schmitz HH, Keen CL. 2002. Procyanidin dimer B2 epicatechin (4 beta 8) epicatechin in human plasma after the consumption of a flavanol rich cocoa. American Journal of Clinical Nutrition 76(4). Howell AB, Botto H, Combescure C, Blanc Potard A B, Gausa L, Matsumoto T, Tenke P, Sotto A, Lavigne J P. 2010. Dosage effect on uropathogenic Escherichia coli anti adhesion activity in urine following consumption of cranberry powder standardized for proanthocyanidin content: a multicentric randomized double blind study. Bmc I nfectious Diseases 10. Howell AB, Vorsa N, Marderosian AD, Foo LY. 1998. Inhibition of the adherence of P fimbriated Escherichia coli to uroepithelial cell surfaces by proanthocyanidin extracts from cranberries. New England Journal of Medicine 339(15):1085 6. Koehler N, Wray V, Winterhalter P. 2008. New approach for the synthesis and isolation of dimeric procyanidins. Journal of Agricultural and Food Chemistry 56(13). Koerner JL, Hsu VL, Lee J, Kennedy JA. 2009. Determination of proanthocyanidin A2 content in phenolic polymer isolates by reversed phase highperformance liquid chromatography. Journal of Chromatography A 1216(9). 71

PAGE 72

Koo H, Duarte S, Murata RM, Scott Anne K, Gregoire S, Watson GE, Singh AP, Vorsa N. 2010. Influence of cranberry proanthocyanidins on formation of biofilms by streptococcus mutans on salivacoated apatitic surface and on dental caries development in vivo. Caries Research 44(2):116 26. Kresty LA, Howell AB, Baird M. 2011. Cr anberry proanthocyanidins mediate growth arrest of lung cancer cells through modulation of gene expression and rapid induction of apoptosis . Molecules 16(3):2375 90. Laks PE, Hemingway RW. 1987. C ondensed tannins base catalyzed reactions of polymeric procyanidins with toluene alpha thiol lability of the interflav anoid bond and pyran ring. Journal of the Chemical Society Perkin Transactions 1 (3). Le Bourvellec C, Guyot S, Renard C. 2004. Non covalent interaction between procyanidins and apple cell wall material Part I. Effect of some environmental parameters. Bioc himica Et Biophysica ActaGeneral Subjects 1672(3):192 202. Liu H, Liu H, Wang W, Khoo C, Taylor J, Gu L. 2011. Cranberry phytochemicals inhibit glycation of human hemoglobin and serum albumin by scavenging reactive carbonyls. Food & Function 2(8):475 82. Liu H, Zou T, Gao J m, Gu L. 2013. Depolymerization of cranberry procyanidins using (+) catechin, ( ) epicatechin, and ( ) epigallocatechin gallate as chain breakers. Food Chemistry 141(1):488 94. McKay DL, Blumberg JB. 2007. Cranberries (Vaccinium macrocarpon) and cardiovascular disease risk factors. Nutrition Reviews 65(11):490 502. Neto CC. 2007. Cranberry and its phytochemicals: A review of in vitro anticancer studies. Journal of Nutrition 137(1):186S 93S. O u K , G u L. 2013. D epolymerization optimization of cranberry procyanidins and transport of resultant oligomers on monolayers of human intestinal epithelial caco 2 cells. Faseb Journal 27. Porter LJ, Hrstich LN, Chan BG. 1986. T he conversion of procyanidins and prodelphinidins to cyanidin and delphinidi n . Phytochemistry 25(1). Robbins RJ, Leonczak J, Johnson JC, Li J, Kwik Uribe C, Prior RL, Gu L. 2009. Method performance and multi laboratory assessment of a normal phase high pressure liquid chromatography fluorescence detection method for the quantitati on of flavanols and procyanidins in cocoa and chocolate containing samples. Journal of Chromatography A 1216(24):4831 40. 72

PAGE 73

Robbins RJ, Leonczak J, Li J, Johnson JC, Collins T, Kwik Uribe C, Schmitz HH. 2012. D etermination of flavanol and procyanidin (by degree of polymerization 1 10) content of chocolate, cocoa liquors, powder(s), and cocoa flavanol extracts by normal phase highperformance liquid chromatography: collaborative study. Journal of A OAC International 95(4):1153 60. Selga A, Sort X, Bobet R, T orres JL. 2004. Efficient one pot extraction and depolymerization of grape (Vitis vinifera) pomace procyanidins for the preparation of antioxidant thio conjugates. Journal of Agricultural and Food Chemistry 52(3). Su X, Howell AB, D'Souza DH. 2010. The eff ect of cranberry juice and cranberry proanthocyanidins on the infectivity of human enteric viral surrogates. Food Microbiology 27(4):535 40. Tomas Barberan FA, Cienfuegos Jovellanos E, Marin A, Muguerza B, Gil Izquierdo A, Cerda B, Zafrilla P, Morillas J, Mulero J, Ibarra A, Pasamar MA, Ramon D, Espin JC. 2007. A new process to develop a cocoa powder with higher flavonoid monomer content and enhanced bioavailability in healthy humans. Journal of Agricultural and Food Chemistry 55(10). Tsang C, Auger C, Mull en W, Bornet A, Rouanet JM, Crozier A, Teissedre PL. 2006. The absorption, metabolism and excretion of flavan 3 ols and procyanidins following the ingestion of a grape seed extract by rats (vol 94, pg 175, 2005). British Journal of Nutrition 95(4). Vattem DA, Shetty K. 2002. Solid state production of phenolic antioxidants from cranberry pomace by Rhizopus oligosporus. Food Biotechnology 16(3):189 210. Wang W, Yagiz Y, Buran TJ, Nunes CdN, Gu L. 2011. Phytochemicals from berries and grapes inhibited the form ation of advanced glycation end products by scavenging reactive carbonyls. Food Research International 44(9):2666 73. White BL, Howard LR, Prior RL. 2010. R elease of bound procyanidins from cranberry pomace by alkaline hydrolysis. Journal of Agricultural and Food Chemistry 58(13). Wu X, Beecher GR, Holden JM, Haytowitz DB, Gebhardt SE, Prior RL. 2006. Concentrations of anthocyanins in common foods in the United States and estimation of normal consumption. Journal of Agricultural and Food Chemistry 54(11):40 69 75. 73

PAGE 74

BIOGRAPHICAL SKETCH Kaijie Song received his Bachelor of Science degree in food science and e ngineering from Jinan University in July 2012. After that, he was admitted into the University of Florida to pursue t he Master of Science degree in food science and human n utrition. Kaijie Song received the William F. and Agnes F. Brown Scholarship awarded by FSHN department in 2014. Upon graduation, Kaijie plans to begin his professional career as a food scientist in the industry. 74