Reaction of Caffeic Acid with Amino Acids

Western Michigan University ScholarWorks at WMU

Master's Theses Graduate College

8-2000

Yonggang Wang

Follow this and additional works at: https://scholarworks.wmich.edu/masters_theses

Part of the Chemistry Commons

Recommended Citation Wang, Yonggang, "Reaction of Caffeic Acid with Amino Acids" (2000). Master's Theses. 4342. https://scholarworks.wmich.edu/masters_theses/4342

This Masters Thesis-Open Access is brought to you for free and open access by the Graduate College at ScholarWorks at WMU. It has been accepted for inclusion in Master's Theses by an authorized administrator of ScholarWorks at WMU. For more information, please contact [email protected]. REACTION OF CAFFEIC ACID WITH AMINO ACIDS

Y onggang Wang

A Thesis Submitted to the Faculty of The Graduate College in partial fulfillment of the requirements forthe Degree of Master of Arts Department of Chemistry

WesternMichigan University Kalamazoo, Michigan August 2000 Copyrightby Yonggang Wang 2000 ACKNOWLEDGMENTS

I wish to express my thanks and appreciation to Dr. Donald R. Schreiber for his continuous guidance and support. I was fortunate to have him as my research advisor. Many thanks also to Dr. James Howell, Dr. Michael McCarville and Dr.

Subra Muralidharan, who had always been there to provide valuable advice and constant assistance to this project.

I also appreciate the financialsupport fromthe Department of Chemistry and

Graduate College during this project at WesternMichigan University.

Finally, I would like to thank my entire family. Their unending love, expectation, and encouragement drove me to overcome numerous difficultiesand frustrations encountered in the completion of the project.

Y onggang Wang

11 REACTION OF CAFFEIC ACID WITH AMINO ACIDS

Yonggang Wang, M.A.

WesternMichigan University, 2000

The reaction of caffeic acid with tryptophan was studied as a model for the reaction of caffeic acid with amino acids in the aquatic environment. Although the autoxidation reaction of caffeic acid in aqueous solution with subsequent dimerization had been previously studied, it was examined as well forcomparison purposes. The present studies demonstrate that oxidized caffeic acid would react with tryptophan.

The reaction, like the dimerization reaction, was catalyzed by the presence of iron(III) ions at low pH ( <4) and by hydroxide ions at high pH (10). The reaction of caffeic acid with itself yielded six products, fourof which were identified by mass spectroscopy as dimers, with the other two being trimers. The reaction of caffeic acid with tryptophan yielded three product peaks. Two of the product peaks had similar retention times to the dimers fromthe self-reaction of caffeic acid; the third was a new product resulting fromthe reaction of caffeic acid with the tryptophan.

A mechanism for the reaction of caffeic acid with tryptophan has been proposed. This mechanism was based on the analysis of the products and previous mechanisms presented by earlier workers for the caffeic acid self-reaction. TABLE OF CONTENTS

ACKN"OWLEDGMENTS...... 11

LIST OF TABLES ...... :...... vu

LIST OF FIGURES...... Vlll

CHAPTER

I. INTRODUCTION ...... 1

Caffeic Acid and Phenolic Compounds...... 1

Humic Substances...... 2

Iron in Natural Waters ...... 6

Objective of Study ...... 6

II. LITERATURE REVIEW ...... 8

Caffeic Acid and Iron Complexation...... 8

Non-enzymatic Oxidation of Phenolic Compounds...... 9

Reaction of Phenolic Compounds With AminoAcids ...... 13

III. MATERIALS AND METHODS...... 15

Materials and Physical Properties...... 15

Sample Preparations...... 15

Caffeic Acid Solution ...... 15

Caffeic Acid and Tryptophan Mixture...... 16

CaffeicAcid and Iron(III) Ion Mixture...... 16

lll Table of Contents--Continued

CHAPTER

CaffeicAcid, Tryptophan and Iron(III) Ion Mixture ...... 16

Physical Appearance of the Samples ...... 17

Instrumental Methods ...... 17

UV-Visible Spectrophotometry (UVNis) ...... 17

Analytical High PerformanceLiquid Chromatography (Analytical HPLC) ...... 18

CaffeicAcid Autoxidation Reaction ...... 18

Reaction of CaffeicAcid With Tryptophan ...... 19

Comparison Study of the Three Pairs of Reactions...... 20

Preparative High Performance Liquid Chromatography (Preparative HPLC) ...... 20

Liquid Chromatography/Mass Spectrometry (LC/MS)...... 21

1 H Nuclear Magnetic Resonance Spectrometry (1 H NMR)...... 22

IV. RESULTS AND DISCUSSION...... 23

Effect of pH...... 23

CaffeicAcid Autoxidation Reaction...... 24

Reaction of CaffeicAcid With Tryptophan...... 26

Effect of Transition Metal Iron(III) ...... 27

Separation and Isolation of Reaction Products...... 31

CaffeicAcid Autoxidation Reaction...... 31

IV Table of Contents--Continued

CHAPTER

Analytical IIPLC...... 31

Preparative HPLC ...... 32

Reaction of Caffeic Acid With Tryptophan...... 33

Analytical IIPLC...... 33

Preparative IIPLC ...... 33

Identification of Reaction Products ...... 34

Caffeic Acid Autoxidation Reaction...... 34

UV-Visible Spectrophotometry (UV NIS) ...... 34

Liquid Chromatography/Mass Spectrometry (LC/MS)... 38

1 H Nuclear Magnetic Resonance Spectrometry 1 ( HNMR)...... 38

Reaction of Caffeic Acid With Tryptophan...... 42

UVNIS ...... 42

LC/MS...... 42

I HNMR...... 45

Comparison Study...... 45

V. PROPOSED MECHANISM...... 51

VI. CONCLUSIONS...... 54

APPENDICES

A. LC/MS Spectra of Products in Caffeic Acid Autoxidation Reaction ...... 56

V 1 B. H NMR Spectra of Products in CaffeicAcid Autoxidation Reaction and Reaction of CaffeicAcid With Tryptophan...... 65

BIBLIOGRAPHY...... 77

Vl LIST OF TABLES

1. The Important Functional Groups in Humus ...... 4

2. LC-MS Data for CaffeicAcid and Its Oxidation Products...... 39

1 3. H NMR Spectra Data (8, ppm) in D2O at 400 MHz...... 40

4. 1 H NMR Data for Reaction ofCaffeic Acid With Tryptophan ...... 47

vu LIST OF FIGURES

1. A Model Structure Proposed for Humic Acid ...... 3

2. Representation ofthe Polyphenol Theory of Humus Formation...... 5

3. Suggested Interaction Mechanism between Caffeic Acid and Fe(III) by Linder and Voye (1987)...... 9

4. Two Cyclobutane Linkage Dimers Proposed by Cohan (1964) ...... 10

5. Mechanism of Dimer Formation Proposed by Cilliers (1991) ...... 11

6. Two Possible Structures ofProducts Proposed by Fulcrand (1994) ...... 12

7. Deamination and Decarboxylation ofAmino Acids...... 13

8. UVNisible Spectrum of0.lmM Caffeic Acid Solution...... 18

9. Effectof pH on 0.lmM Autoxidation Reaction ...... 24

10. Effectof pH on Reaction of0.lmM CA With 1.0mM Tryptophan ...... 27

11. UVNisible Spectrum ofMixture ofCaffeic Acid With Tryptophan at 0min,pH l0 ...... 28

12. UVNisible Spectrum ofMixture ofCaffeic Acid With Tryptophan at 1 hour, pH 10 ...... 28

13. UVNisible Spectrum ofMixture of CaffeicAcid With Tryptophan at 1 Week,pH 10 ...... 29

+ 14. Effect of 0.05mM Fe3 on 0.lmM Caffeic Acid Autoxidation Reaction ...... 29

3+ 15. Effectof0.05mM Fe on the Reaction of0. l mM CaffeicAcid With lmM Tryptophan ...... 30

16. Separation of Caffeic Acid and Its Oxidation Products at pH 10 After 16 h...... 32

Vlll List of Figures-Continued

17. Reaction of CaffeicAcid with Tryptophan at pH 10 After10 h ...... 34

18. UVNisible Spectrum ofFraction 5 Collected in CA Autoxidation Reaction ...... 35

19. UVNisibleSpectrum ofFraction 6 Collected in CA Autoxidation Reaction ...... 3 5

20. UVNisible Spectrum ofFraction 7 Collected in CA Autoxidation Reaction ...... 36

21. UVNisibleSpectrum ofFraction 8 Collected in CA Autoxidation Reaction ...... 36

22. UVNisibleSpectrum ofFraction 9 Collected in CA Autoxidation Reaction ...... 37

23. UVNisibleSpectrum ofFraction 10 Collected in CA Autoxidation Reaction ...... 3 7

24. Peak Assignmentsfor Caffeic Acid...... 40

25. UVNisibleSpectrum of Standard Tryptophan ...... 43

26. UVNisibleSpectrum ofFraction 2 Collected in Reaction of CA With Tryptophan...... 43

27. UVNisible Spectrum ofFraction 3 Collected in Reaction of CA With Tryptophan...... 44

28. UVNisibleSpectrum ofFraction 4 Collected in Reaction of CA With Tryptophan...... 44

29. Peak Assignments forTryptophan (a) and Peak 4 (b) ...... 46

30. Separation of Reaction Products of CA and Tryptophan Without Fe3+ (After2 Months) ...... 48

31. Separation of Reaction Products of CA and Tryptophan With Addition ofFe3+ (After2 Weeks)...... 49

lX List of Figures----Continued

32. CaffeicAcid Autoxidation Reaction at pH 10 After 16 h ...... 49

33. Reaction of CaffeicAcid With Tryptophan at pH 10 After 10 h ...... 50

34. Proposed Mechanism forReaction of CaffeicAcid With Tryptophan ...... 52

X CHAPTER I

INTRODUCTION

Caffeic Acid and Phenolic Compounds

The expression " phenolic compounds " embraces a wide range ofsubstances which possess an aromatic ring with a hydroxyl substituent, including their functional derivatives. Among the natural phenolic compounds, of which several hundreds are known, the flavonoids and their relatives formthe largest group. However phenolic quinones, lignins, xanthones, depsidones, and other groups also exist in considerable numbers as well as many simple monocyclic phenols. Two examples ofphenolic compounds are caffeic acid (CA) and chlorogenic acid. In nature, caffeic acid and chlorogenic acid are degraded fromplants and found naturally in various foodstuffs and beverages such as coffeebeans and their soluble constituents, potatoes, fruits such as apples and their juices, tobacco leaves, olive oil, and wine (Van Buren et al.,

1973; Challis and Bartlett, 1975; Woodring et al., 1990; Chi-Tang, 1992).

It is well known that some phenolic compounds, such as caffeic acid and cholorogenic acid contribute to both foodbrowning (Hurrell and Finot, 1984) and the formation ofhumic substances in the natural environment (Jiang, 1996). There is a growing interest in the studies of the dietary phenolic compounds (Challis and

Bartlett, 1975; Laranjinha et al., 1994). Generally, there are two types of phenolic

1 2 browning reactions. Enzymatic oxidation is the more important reaction in fresh fruits and juices and early in food processing, when poly-phenol oxidase is present

(Coseteng and Lee, 1987; Matheis, 1987). In processed foods with the enzyme removed or in the natural environment, nonenzymatic autoxidation can take place.

Autoxidation (the expression originally used by Cilliers in 1989 to refer to the reaction of caffeicacid with itself followingits oxidation by oxygen. This terminology will be used throughout this thesis) can be catalyzed under alkaline condition or in the presence of copper(II) ions (Jiang, 1996).

Humic Substances

Humic substances (HS) are the most widespread and ubiquitous natural nonliving organic materials in soil and aquatic environments. They constitute the major fraction of soil organic material (up to 80%) and the largest fraction of natural organic matter in aquatic systems (up to 60% of dissolved organic carbon) (Thurman,

1986; Stevenson, 1982; Schnitzer, 1991; Stevenson, 1994).

Humic substances are a physically and chemically heterogeneous mixture of naturally occurring, biogenic, relative high molecular weight, mixed aliphatic and aromatic compounds. They are formed by humification during the decay process and transformationof biomolecules that originate fromdead organism and microbial activity.

Based on solubility in acids and alkalis, HS can be divided into four fractions;

(1) humic acid (HA), the portion that is soluble in dilute alkaline solution and is 3 precipitated upon acidification to pH 2; (2) fulvic acid (FA), the portion that is soluble at any pH value; (3) humin, the portion insoluble in both alkalis and acids; and (4) hymatomelanic acid, the alcohol-soluble portion ofHA (Stevenson, 1965).

Since humic substances consist ofa chemically heterogeneous mixture of compounds, it is difficult to describe uniquely the molecular formulaof HA, FA and other HS species. However, it is possible to propose the general structure ofa molecule of HA and FA on the basis ofcompositional, structural and functional group information. The functionalgroups in humic substances, especially those most reactive with protons and metals, have been well characterized. These functional groups include carboxyl, phenolic, and alcoholic hydroxide, quinone and ketonic carbonyls, amino, and sulfhydryl (SH) (Sposito, 1989). Table 1 lists these important functional groups in humus. A typical model structure forhumic acid proposed by

Stevenson (1982) is shown in Figure 1.

Several pathways have also been proposed for the formationofhumic substances. The "lignin-protein theory" ofWaksman (1932), considers lignin as the main source ofHS, with the involvement ofamino compounds produced by microbial synthesis. In one pathway, reducing sugars and amino acids formedas by-products of microbial metabolism are assumed to be the only precursors ofHS. The current concepts ofHS genesis favorthe "polyphenol theory", which involves polyphenols and quinones, either derived fromlignin or synthesized by microorganisms. Figure 2 shows the detailed steps involved in polyphenol theory ofsoil humus formation

(Cresser and Killham, 1993). 4

COOH

Figure 1. A Model Structure Proposed for Humic Acid.

Table 1

The Important Functional Groups in Humus

Functional group Structural formula

Carboxyl -COOH

Carbonyl -C=O

Amino -NH2

lmidazole Aromatic ring NH

Phenolic OH Aromatic ring OH

Alcoholic OH -OH

Sulfhydryl -SH 5 _____ 1 1 PLANT RESIDUES --__ I �I + • -----...------, Ll CARBOHYDRAIBS AMINO ACIDS, r PROTEINS

DEGRADATION CO2 PRODUCTS MICROBIAL BIOMASS phosphate DEGRADATION ammomum SYNTHESIS sulphate AROMATICS

ORGANIC N &P

POLYMERS 1------i► CONDENSATION+ HUMUS

Figure 2. Representation of the Polyphenol Theory of Humus Formation.

The study of humic substances is a worldwide activity with many implications forour survival, environment and human health. Much progress has been made in humic substance research recently, particularly with regard to trying to deduce HS structures with the help of sophisticated analysis and molecular modeling (Cresser and

Killham, 1993). Studies of highly purified humic substances have indicated that they have reproducible and accountable properties and appear to have common metal binding sites (Sposito, 1989). Humic substances are nature's fertilizers. Interest in humic substances will continue to grow in the future. 6 Iron in Natural Waters

Iron is a common constituent of rocks. Fossil fuels, ores, municipal sewage and industrial effluentalso contain iron and other trace elements. Iron is introduced to natural waters through the weathering ofrocks and by human activities. In many rivers and lakes, the human input of iron is many times greater than the natural input

(Stumm and Baccini, 1978). The water in equilibrium with the iron mineral ferrihydrite [Fe(OH)3], produces a certain dissolved iron concentration, which is present primarily as the Fe(III) species. Under certain specific conditions, Fe(III) can also exist as ferric-organic(humic-fulvic) complexes and as colloidal ferric oxyhydroxides. Compared to Fe(III), there is a much smaller amount ofuncomplexed

Fe(II) existing in natural waters (Stumm and Baccini, 1978).

Objectiveof Study

Caffeic acid is a natural occurring phenolic compound. Amino acids and transition metals can also be foundin natural environment. Humic substances are the major fraction ofnatural organic matter in both soil and aquatic systems.

Environmental scientists have done numerous studies ofthe formationof humic substances. It is suggested that phenolic compounds play an important role in one proposed pathway (polyphenol model) of the formationof humic substances (Flaig,

1988). Previous studies have also demonstrated that there is an interaction between caffeicacid and transition metals like copper(II) and iron(III) (Deiana et al., 1992).

However, the reaction products have not been well characterized. What are the 7 relationships of natural existing phenolic compounds with amino acids, transition metals and humic substances? How are the humic substances formedin natural waters? What role do amino acids play in the formation ofhumic substances? This study attempts to answer some ofthese questions by studying the reaction of caffeic acid with tryptophan; investigating the effectof hydroxide and iron(III) ion on the reaction of caffeic acid with amino acids; characterization of the products of the reaction of caffeic acid with tryptophan. A possible reaction mechanism of caffeic acid with tryptophan has also been proposed. CHAPTER II

LITERATURE REVIEW

CaffeicAcid and Iron (III) Complexation

Phenolic compounds with low molecular weight have considerable biological importance because they are involved in many processes that take place in the soil plant systems. They increase the mobilization of important micro elements, such as

Fe and Mn, primarily by complexation or reduction reactions converting insoluble compounds that are unavailable to plants to those that are available to plants

(Lehmann et al., 1987). Among these phenolic compounds, caffeic acid is of special interest because it takes part in the transport of various metal ions fromthe soil to the plant roots (Linder and Voye, 1987). Olsen et al. (1982) demonstrated that caffeic acid underwent an oxidation reaction in aqueous solution whereby Fe(III) was reduced to Fe(II). In order to obtain a better understanding of the mechanisms involved in complexation of caffeicacid and iron(III), Deiana et al. (1992) studied the redox properties of caffeic acid and other phenolic compounds. They determined the stoichiometry and proposed a probable mechanism for the oxidation-reduction reaction (Figure 3). The interactions in aqueous solution of caffeicacid with copper(II), zinc(II), iron(II) and iron(III) were investigated by Linder and Voye

(1987). They foundthat Fe(III) was reduced to Fe(II) in caffeicacid solution even at

8 9 very acidic conditions. Complexation of the metal ions predominantly involved chelation by the catecholic site of caffeic acid.

H0---0-CH==CH-COOH + 2Fo" --

HO + 7Fe3+

Decomposition Products + 7Fe2+

Figure 3. Suggested Interaction Mechanism between CaffeicAcid and Fe(III) by Linder and Voye (1987).

Non-enzymatic Oxidation of Phenolic Compounds

The oxidative browning of polyphenols in plant derived foodsand beverages generally results in a loss of nutritional value, bitter taste and the appearance of undesirable brown colors (Hurrell and Finot, 1984). The firststep in browning is the oxidation of o-diphenols to the corresponding highly reactive o-quinones (Pierpoint,

1966). Enzymatic oxidation is the most important reaction in fresh fruitsand juices when polyphenoloxidase is present (Coseteng and Lee, 1987). However, non-enzymatic autoxidation can also take place in the presence of oxygen, especially when the medium is alkaline (Cillers and Singleton, 1989; Jiang, 1996). The different steps leading to the formationand further reaction of quinones to produce condensation products that are not fully understood and only a fewh ypotheses have 10 been proposed about their mechanism (Cilliers and Singleton, 1991; Fulcrand et al.,

1994; Singleton, 1987).

In 1964, Cohen foundthat the photodimerization of ring-substituted cinnamic acids led to trans- and cis-dimers by breaking the double bond of the ethylene group on the side chain of caffeicacid. Figure 4 shows the structure of these two dimers

(cyclobutane linkagedimers) (Cohen et al., 1964).

OOH

COOH COOH COOH

a -Dimer P- Dimer

Figure 4. Two Cyclobutane Linkage Dimers Proposed by Cohan (1964).

Cilliers et al. (1991) reported the possible nonenzymic autoxidation products of caffeicacid in 1991. They found that the rate of reaction was increased by raising pH and/or temperature. The phenolate anion was believed to be a key factor in the formation of autoxidation products and the rate of reaction. The products they proposed were dimers linked via the double bond on the side chain of one molecule with the oxygens on the aromatic ring of another (ether linkage dimers). The mechanism proposed by Cilliers et al. (1991) is shown in Figure 5. 11

oo- oo- oo- oo-

-oH 4 �

OH 02 OH OH O· OH o· OH 0

oo-

-ooc o coo- o coo-

� x x 0 R 0 R (2R, 3R) -ooc P(2R, 3R)

-ooc o coo

� Rx 0

(2S, 3S) (2S, 3S)

("YOH R=

�OH

Figure 5. Mechanism of Dimer Formation Proposed by Cilliers(1991). 12 Xu (1994) studied the caffeic acid oxidation reaction in the presence of

Copper(II). The reaction rate was increased when copper(II) was added to the reaction solution. Xu (1994) proposed a mechanism of caffeic acid autoxidation that involved both ether linkage and/or cyclobutane linkage dimerization in the presence of copper(II). Jiang (1996) found that hydroxide ion also had a catalytic effect on the chlorogenic autoxidation reaction.

SodiumPeriodate oxidation of caffeic acid was investigated in acidic conditions by Fulcrand et al. (1994). It was confirmed that the rate and the yield of the reaction increased with increasing pH . In particular, the two products obtained at pH values lower than 4.6 were analyzed and isolated by reverse phase HPLC. Using 1H and 13C NMRand mass spectrometry, the products were believed to be two stereoisomers of 2,5-(3', 4'- dihydroxyphenyl) tetrahydrofuran3,4-dicarboxylic acid

(Figure 6). They also found that at pH values above 4.6, the products were same as shown in Cilliers's study (1991).

0 R 0 H

�-�H �H -Y,R H .l___l� H�COOH H .. . ,. ,, COOR,... ,, COOR COOR,, H

R= ---0-0H

Figure 6. TwoPossible Structures ofProducts Proposed by Fulcrand (1994). 13 Reaction of Phenolic Compounds With Amino Acids

Unlike the caffeicacid oxidation reaction, the reaction of caffeic acid with amino acids has been less studied. Flaig et al. (1975) reviewed the literature pertaining to the reaction of phenolic compounds with amino acids to form humic acids.

Generally, the reaction can take place in one of two ways. One is by the addition of amino acids to the oxidized phenol:

Phenols------,►� Quinones Phenolase

Quinone + Amino Acid ➔ Humic Substance

The other is via a deamination and decarboxylation of one amino acid; while the second is added to the phenol:

0 OH R -Xe N-6-cooH & + ► o I -xH I I H H

+NH3 +CHO+ CO2

Figure 7. Deamination and Decarboxylation of Amino Acids.

Which of the above two reactions is predominant depends upon many factors such as the specificamino acid, the phenolic compounds used, the pH, and the ratio of 14 the two reactants (Linder and Voye, 1987). Most of the amino acids studied by previous researchers were neutral and aliphatic. In this study, tryptophan was selected as a model for amino acids as tryptophan possesses an a-amino group and an indol group. Since both groups are very active, it was anticipated that it would react with the caffeicacid and give a high yield of products. CHAPTER III

EXPERIMENTAL SECTION

Materials and Physical Properties

Caffeicacid (99%, Aldrich Chemical Co., Inc., Milwaukee, WI), L-(-) tryptophan (99%, Aldrich Chemical Co., Inc., Milwaukee, WI), and ferric chloride

(99.5%, General Chemical Division, New York, NY) were used in these studies. All aqueous samples were prepared using 18 MQ Milli-Q water. Acetonitrile (HPLC

Grade), 0.1 % formicacid (made from 88% formicacid, CertifiedACS) and Milli-Q water were used as solvents in HPLC separation and isolation.

Caffeicacid is a pale gray powder with a molecular weight of 180 g/mol. It is slightly soluble in cold water and soluble in boiling water. Tryptophan is a white powder with a molecular weight of 204 g/mol. It is also partially soluble in cold water and soluble in boiling water.

Sample Preparation

Caffeic Acid Solution

0.1 mM and 14 mM of caffeic acid solutions were prepared by adding 0.0018 g and 0.251 g of caffeic acid to 100 mL of water respectively. The solutions were heated on a hot plate until they were completely dissolved. The solution pHs were

15 16 raised by addition of50% (w/w) NaOH to obtain the desired pH. The pH values were measured using a Corning Model 240 pH meter pre-calibrated with standard pH buffers (4.00 and 7.00). The pHs of all samples were adjusted to about 7.00 by addition of6 M HCl beforerunning the HPLC.

CaffeicAcid and TryptophanMixture

Two different concentrations of caffeic acid and tryptophan mixture were made: (1) 100 mL of0.l mM caffeic acid and 1 mM tryptophan; (2) 100 mL of1 mM caffeicacid and 1 0mM tryptophan. The solution pH values were raised to the specifiedpH, then adjusted back to neutral prior to introducing to the HPLC system.

Caffeic Acid and Iron(III)Ion Mixture

Iron(III) chloride was added to 0.1 mM caffeic acid solution so that the molar ratio ofthe iron(III) ion and caffeic acid was 1:2. A 0.1 mM of caffeicacid as a blank was also prepared at the same time. Then both were capped and stored in the dark place for UV-Visible absorption tracking and product comparison in two months.

CaffeicAcid, Tryptophanand Iron(III) Ion Mixture

The followingtwo solutions were prepared for comparison study. First, a

100 mL of a solution containing 0.1 mM caffeic acid and 1 mM tryptophan and second 100 mL of a mixed solution containing 0.1 mM caffeic acid, 1 mM tryptophan and 0.05 mM iron(III). Both were capped and stored in the dark place forUV-Visible 17 tracking and product comparison in the following two months.

Physical Appearance of the Samples

The color changes ofboth 14 mM acid solution and 1 mM caffeic acid with

10 mM tryptophan solution was observed and recorded. When raising the solution pH value, the color changed from clear to greenish yellow, then brown and finally to dark brown. The higher the solution pH, the faster the color changed. Without adjustment ofthe solution pH or addition ofiron(III), the solution slowly turnedinto brown within two months.

Instrumental Methods

UV-Visible Spectrophotometry(UYNis)

A HP 8451A diode array spectrophotometer equipped with a HP 7470A plotter and a HP ThinkJet printerwas used to monitor the catalytic effectsof iron(III) and hydroxide ions on the caffeic acid autoxidation reaction and the reaction of caffeic acid with tryptophan. It was also used for absorption measurements of isolated products ofthe above two reactions. The absorption spectrum of0.l mM caffeicacid is shown in Figure 8.

The first absorptionmeasurements were taken right afterthe sample solutions were made. The absorption values at 320 nm, which is the maximum absorption wavelength of caffeic acid, were recorded twice a week for two months. The complete spectra were also obtained as the reaction ofcaffeic acid with tryptophan proceeded. 18

1.6

1.4

I. 2

w .8

CD .6 CD

. 4

-. 2

WAVELENGTH (n.J

Figure 8. UVNisible Spectrum of 0.1 mM CaffeicAcid Solution.

AnalyticalHigh PerformanceLiquid Chromatography{Analytical HPLC)

CaffeicAcid Autoxidation Reaction

A Varian 5060 High PerformanceLiquid Chromatography system (Varian,

Walnut Creek, CA) with Milton Roy UV-Visible Detector and Hewlett-Packard

Intergrator was used in this project. An Adsorbosil C18 (Alltech Associates, Inc.,

Deerfield, IL) 25 cm x 4.6 mm i.d. column with 5 µm packing materials was used in this study. The injection volume was 10 µL with a flowrate of 1.0 mL/min. A 0.1 % formicacid buffersolution was used as one mobile phase (A) and acetonitrile (HPLC

Grade) as the second mobile phase (B). The pH of 0.1 % formic acid solution was 19

2. 71, which was much lower than the pKa of caffeic acid and its derivatives. The function of 0.1% formic acid buffer was to suppress the ionization of caffeic acid and its derivatives. The buffer and all samples were vacuum filtered through a 0.45

Millipore filterbefore introducing them to HPLC system.

A gradient elution method was used in this study. The pump was programmed to start with a 90% mobile phase A and 10% mobile phase B, then ramped up to a

50/50 mixture in 30 minutes. After the column was washed with 100% mobile phase

B for 10 minutes and equilibrated with 90% mobile phase A and 10% mobile phase B for additional 10 minutes, the next 10 µL sample was injected. The UV-Visible detector was set at a wavelength of 280 nm. The standard caffeicacid solution was chromatographed to determine its retention time.

Reaction of Caffeic Acid With Tryptophan

The same column was used in product separation for the reaction of caffeic acid with tryptophan. The injection volume was also 10 µL with a flowrate of

1.0 mL/min. In comparison with the separation of caffeic acid autoxidation products, an isocratic elution method was chosen in this separation instead of the gradient elution method. A composition of 70% mobile phase A (0.1% formic acid buffer solution) and 30% mobile phase B (Acetonitrile, HPLC Grade) was maintained throughout the separation. The detector was again set at 280 nm. Both standard caffeic acid and tryptophan samples were also injectedunder these conditions to determine their retention times. 20 Comparison Study of the Three Pairs of Reactions

Three pairs of reactions were established in order to compare the products in each pair. For each pair of reactions, the products of individual reactions were analyzed separately on analytical HPLC system under the same conditions.

1. 0.1 mM Caffeicacid autoxidation reactions with and without addition of

3 0.05 mM Fe + were sampled aftertwo months. Both reaction mixtures were injected at the optimum separation parameters used in the separation of caffeic acid autoxidation products.

2. Reactions of 0.1 mM caffeicacid and 1 mM tryptophan with and without

3 addition of 0.05 mM Fe + were sampled after two months. Both reaction mixtures were analyzed at the separation conditions for reaction of caffeic acid with tryptophan.

3. CaffeicAcid autoxidation reaction and the reaction of caffeic acid with tryptophan. Both reactions were set up at pH=l0, terminated after2 hours. The reaction products were analyzed by analytical HPLC method under the separation conditions used in 2.

Preparative High PerformanceLiquid Chromatography (PreparativeHPLC)

A C18 preparative 25 cm x 10 mm i.d. column (Alltech Associates, Inc.,

Deerfield,IL) was used to collect the fractionsof caffeic acid autoxidation reaction.

The injection volume was 100 µL with a flowrate of2.0 mL/min. The mobile phases and elution method were the same as in the analytical HPLC method. A 5 mM 21 solution diluted froma 14 mM caffeicacid solution was manually injected into HPLC system.

A 100 µL sample was injected on the C18 preparative column for fraction collection of reaction products of caffeicacid with tryptophan. A higher flow rate of

3.0 mL/min of 70% of 0.1 % formicacid buffer solution and 30% of acetonitrile

(HPLC Grade) was used as the mobile phase.

The fractions collected frommore than twenty injections were combined for each fraction. All fractionsolutions were kept in the dark and air-dried for further identification with 1 H nuclear magnetic resonance. Each collected fraction was also analyzed using UV-Visible spectrophotometry.

Liquid Chromatography/Mass Spectrometry(LC/MS)

In order to obtain the mass spectra informationfor each product, a Waters

2690 liquid chromatography system (Waters, Milford, MA) and a Micromass

Quattro II mass spectrometer (Micromass UK Limited, United Kingdom) with an electrospray ionization source was used. The conditions used in these LC-MS separations were identical to that in earlier HPLC separations. The reaction mixtures of both caffeicacid autoxidation and caffeicacid with tryptophan were run on this

LC/MS system. Mass and fragment informationof each fraction including starting materials of caffeicacid were obtained. 22 1H Nuclear Magnetic Resonarice (1H NMR)

A JEOL ECLIPSE 400 MHz nuclear magnetic resonance spectrometer (JEOL,

Peabody, MA) was utilized to identifythe collected fractions. The proton spectra of pure caffeicacid arid tryptophan were collected as staridards. D20 was used as solvent arid pre-saturation method was chosen. Due to the limited amount of collected samples, 128 arid 256 scans were performed forthe different :fractions. CHAPTER IV

RESULTS AND DISCUSSIONS

Effectof pH

The effectof pH on both the caffeicacid autoxidation reaction and the reaction of caffeicacid with tryptophan were studied. A series of caffeic acid solutions and caffeicacid with tryptophan solutions at differentpH values were prepared and monitored over a period of two months with a UV-Visible spectrophotometer.

A caffeicacid UV-Visible absorption spectrum was given earlier in Figure 8.

There are two maximum absorption peaks, one at about 320 nm and the other at 280-

290 nm. These absorption peaks are so-called fingerprint peaks of cinnamic compounds in UV-Visible spectra region. The effect of pH on reaction rates of caffeic acid autoxidation reaction and the reaction of caffeic acid with tryptophan was studied primarily by measurement of UV-Visible absorption values at 320 nm. The existence of the aromatic ring and conjugated double bond in cinnamic compounds allows these compounds to absorb UV-Visible electromagnetic radiation resulting in a 7t to n* electron transition. This was also the reason a UVNis detector was chosen forthe HPLC separations. The loss of absorbance at 320 nm may indicate loss of the conjugated double bond of the side-chain in cinnamic compounds (Jiang,1996).

23 24 CaffeicAcid Autoxidation Reaction

The UVNisibleabsorption values at 320 nm are plotted as a function of time forthree different pH conditions forthe caffeic acid autoxidation reaction (Figure 9).

1.6 _pH=3.2 1.4 -pH=7.0 E C 1.2 �pH=10.0 1 0.8 0 e- 0.6 0 0.4 0.2 0 ---.--.------.----- 0 1 2 3 4 5 6 7 8 9 Time(weeks)

Figure 9. Effectof pH on 0.1 mM Autoxidation Reaction.

For the reaction at pH 3.2, which was the original pH value of 0.1 mM caffeic acid solution, as the time increases, the absorption at 320 nm slowly decreases. This indicates that the reaction proceeded slowly even at this pH. The decrease in absorbance indicates that the conjugated double bond of the side-chain in caffeic acid is lost during the autoxidation reaction. As the pH increases, the absorption at 320 nm 25 decreases more rapidly. Afterhalf a week, at pH 10, more than half of the absorption of caffeicacid at 320 nm had already disappeared. At pH 3.2, there was only a small decrease in the absorption in the firsthalf a week period. The factthat the caffeicacid autoxidation reaction rate increased with increasing pH indicates the involvement of the reactive phenolate ion in this reaction (Cilliers and Singleton, 1989). The phenolate ion is thought to react directly, by charge transfer with triplet oxygen, to forma semiquinone radical, which will then undergo further reaction (Cilliers and

Singleton, 1991).

In the first half week, the reaction rate was very high at pH 10. Then, the absorbance dropped very slowly and remained nearly stable after a week indicating the reaction was close to completion. However, aftertwo months, there was still absorption at 320 nm forall three reactions (pH=3.2, 7 and 10). This suggests that either, not all caffeic acid is depleted, or that some oxidation products absorb in this region, or both possibilities exist.

These results are consistent with that of Jiang's study (1996). In Jiang's experiment, pH 3, 7 and 10.5 were chosen as different pH conditions to study the pH effect of the caffeic acid oxidation reaction. The effectof pH on chlorogenic acid oxidation reaction was also reported. Jiang foundboth chlorogenic acid and caffeic acid oxidation reaction were pH dependent. The reaction rates increased as the pH value increased. In 1991, Cilliers et al. reported that approximately 30% side chain ethylenic conjugation existed in the oxidized products of caffeic acid. 26 Reaction of Caffeic Acid With Tryptophan

The effect ofpH on reaction of caffeicacid with tryptophan is illustrated in

Figure 10. The three pH values at which reactions were run are pH 3.75, 7 and 10. The pH 3.75 was the pH of the original mixture ofcaffeic acid with tryptophan. The absorbances at 320 nm were recorded foreight weeks for each of the three pH values.

As observed in caffeic acid oxidation reaction, the higher the pH, the faster the reaction rate. Again at low pH, the reaction still proceeds, although slowly. This indicates that the reaction of caffeic acid with tryptophan is also a pH dependent reaction. The loss ofabsorption attributed to caffeic acid suggests that either caffeic acid underwent an autoxidation reaction, or reacted with tryptophan, or both.

Figures 11, 12 and 13 show the absorbance change of a mixture oftryptophan and caffeicacid as a functionof time. The absorbance of tryptophan decreased a little as the absorption ofcaffeic acid decreased. Considering the excess amount of tryptophan used and the overlap of caffeic acid and tryptophan absorption peaks, it is difficult to attribute this to the loss of tryptophan. However, NMRexperiments and comparison studies were performed to determine the involvement of tryptophan during the reaction.

The colors of both the caffeic acid oxidation mixture and the mixture of caffeicacid with tryptophan were observed during the reaction processes. It was foundthat the solution colors were darker under strong alkaline conditions in both reactions. The higher the pH of the reactions, the darker the reaction solutions. 27 Effect of Transition Metal Iron(III)

In order to investigate the effect of iron(III) on both caffeicacid autoxidation

and reaction of caffeicacid with tryptophan, two pairs of reaction samples at pH 3. 72

were prepared: caffeicacid solution with or without iron(III); Mixture of caffeicacid

and tryptophan with or without iron(III). Effect of iron(III) on caffeicacid

autoxidation reaction and effectof iron(III) on reaction of caffeicacid with tryptophan

was both investigated by measurement of 320 nmabsorbance once a week until the reactions started two months. The experiment results are illustrated in Figure 14 and

Figure 15.

1.6

1.4 -pH=3.75 -pH=7.0 E 1.2 _pH=10.0 1

0 0.8

0 0.6

0.4

0.2

0 0 1 2 3 4 5 6 7 8 9 10 Time(weeks)

Figure 10. Effect of pH on Reaction of 0.1 mM CA With 1 mM Tryptophan. 28

3 UVN Spectrumof Mixture of0.04mM CA with 0.4mM Tryptophan at O min, pH IO

2.5

uJ 1.5

-.5

WAVELENGTH (n,.)

Figure 11. UVNisible Spectrum of Mixture of CaffeicAcid With Tryptophan at O min, pHlO.

3 UVN Spectrum of Mixture of0.04mM CA with 0.4mM Tryptophan at lh, pH IO

2.5

UJ 1.5

-. 5

111\VELENGTH Cn■>

Figure 12. UVNisible Spectrum of Mixture of CaffeicAcid With Tryptophan at 1 Hour, pH 10. 29

3.5 UVN Spectrum of Mixture of0.04mM CA with 0.4mM Tryptophan at I week, pH 10

2. 5

1.5

o,______�---�-

WAVEEGTH (n1)

Figure 13. UVNisible Spectrum of Mixture of CaffeicAcid With Tryptophan at 1 Week, pH 10.

1.6 �Without Fe E 1.4 C: �With Fe 0 1.2 1

C: 0.8 0 .c 0.6 0 0.4 .c <( 0.2 0 0 1 2 3 4 5 6 7 8 9 Time(weeks)

3 Figure 14. Effectof0.05 rnM Fe + on 0.1 rnM Caffeic Acid Autoxidation Reaction. 30

As shown in Figure 14 and Figure 15, both caffeic acid oxidation reaction and reaction of caffeic acid with tryptophan were catalyzed by addition of small amount of iron(ill) ion. The colors of reaction mixtures were also observed to be darker with

3 3+ the presence ofFe +. The effect ofFe ions on the reactions was similar to the effect of hydroxide ions on these two reactions. The absorbance at 320 nm decreased with

3 time. With Fe + present, the reaction rate increases dramatically, especially in the first

3+ two weeks. However, in comparison with pH effects on these two reactions, the Fe effects are somehow weaker, which indicates a stronger catalytic effect of hydroxyl group than that of iron(III) ion.

1.6 � Without Fe 1.4 --+-With Fe C: 1.2 M 1

C: 0.8 0 a o.6 0 0.4 0.2 0 +-----.------.------,---,----,-----,------.---,------, 0 1 2 3 4 5 6 7 8 9 Time(weeks)

3 Figure 15. Effectof0.05 mM Fe + on the Reaction of0.l mM Caffeic Acid With 1 mM Tryptophan. 31 Separation and Isolation of Reaction Products

CaffeicAcid Autoxidation Reaction

AnalyticalHPLC

In order to obtain a good separation of the oxidation products of caffeicacid, much work was done on the development of an analytical HPLC method, which included selection of column, mobile phases, elution method, detector, etc. The reaction mixture was recovered aftersixteen hours at pH 10. A sample chromatogram of caffeicacid and its oxidation products is shown in Figure 16. Peak 1 is the solvent peak. Peak 3 is non-consumed caffeicacid, which was confirmed by comparison with the retention time of pure caffeicacid at the same separation conditions. Peak 5 to peak 10 are finaloxidation products of caffeic acid. Peak 2 and peak 4 are immediate oxidation products that disappeared after two months. However, since peak 2 elutes earlier than caffeic acid, its polarity must be greater than caffeicacid. While peak 4 to peak 10 have longer elution time than caffeic acid, indicating these products are less polar than caffeic acid, especially peak 9 and peak 10. Since the ring-opened products of caffeic acid would increase polarity and thereforehave shorter retention time on reversed phase HPLC system, these products are likely not ring-opened products. The reproducibility of the separation was determined by comparison of chromatographic profilesof repeated injections. 32 Preparative HPLC

The preparative HPLC separation was developed based on analytical HPLC chromatography. The preparative separation chromatogram was similar to that of analytical HPLC except for the appearance of a fewadditional peaks that were not seen in analytical chromatography. This was due to the higher concentration of the

sample in the preparative separation. However, the major peak profiles in both analytical and preparative HPLC separation were the same. A pure caffeic acid sample was again run on preparative HPLC system to confirmthe caffeic acid peak.

The collected fractions (5-10) were air-dried before furtheridentification with UV

Visible spectrophotometry and proton nuclear magnetic resonance methods.

2 CA • ..• 6

7 4 9 1

Time (Minute)

Figure 16. Separation of CaffeicAcid and Its Oxidation Products at pH 10 After16 h. (Analysis of a 10-µL Sample by C18 Reversed-Phase HPLC System. Mobile Phases: A=0. l % Formic Acid, B=Acetonitrile. Gradient: 90% to 50% Mobile Phase A in 30 min. Detection at 280 nm. Peak 1 is the Solvent Peak. Peak 3 is Caffeic Acid. The Others are Oxidation Products.) 33 Reaction of CaffeicAcid With Tryptophan

AnalyticalHPLC

The isocratic elution method was chosen to separate the reaction products of caffeic acid with tryptophan. The chromatogram is given in Figure 17. Tryptophan was used in excess to insure that caffeicacid would react with it afterit was oxidized.

Peakl is a solvent peak. Peak 5 is tryptophan. The peak between peak 1 and peak 2 is caffeic acid. Peaks 2, 3 and 4 are reaction products.

As shown in Figure 17, tryptophan has a very long retention time and a broad peak. This is caused by two factors: one was the tryptophan concentration overload effect. The other was the strong interaction between the amino group in tryptophan and the silanol group in packing materials of the C 18 column. The reaction mixture of caffeicacid autoxidation was also run under the same condition in order to compare the results with those of this reaction. A detailed discussion is given in comparison study section.

Preparative HPLC

Peaks 2, 3 and 4 (fromnow on the peak is referred as fraction) were collected using the preparative HPLC system. Approximately twenty injections were collected and the collected fractions were air-dried prior to identificationwith UV -Visible spectrophotometry and proton nuclear magnetic resonance methods. 34

0 Cll

Cll 0 1 CA 2 3 4 p:: .... � ,,, -�...... ro ..� .. p::

:: Time (Minute)

Figure 17. Reaction of Caffeic Acid With Tryptophan at pH 10 After10 h. (Analysis of 10-µL Sample by Cl8 Reversed Phase HPLC. Mobile Phases: 70%A, 0.1% Formic Acid; 30% B, Acetonitrile. Detection at 280 nm.)

Identificationof Reaction Products

Caffeic Acid Autoxidation Reaction

UV-Visible Spectrophotometry(UV Nis)

The UV Nisible spectra of fraction 6 to 10 are shown in Figures 18 to 23. It was foundthat these spectra were similar, suggesting fraction6 to 10 might be structural homologues. Their comparison with the caffeic acid spectrum indicates the possibility of a modifiedcaffeic acid or oligomers. In the spectra of fraction6-10, there are two maximum absorption peaks in each spectrum: 280-290 nm and 320 nm as was seen in the caffeic acid spectrum. This was indicative of the presence of the conjugated side chain in the oxidation products. However, in comparison with the caffeic acid spectrum, the magnitude of 320 nm absorbance in these spectra was 35 smaller than that of caffeicacid at 280-290 nm, indicating the loss of side chain conjugation during the oxidation. Fraction 5 shows there was a very weak absorption at 320 nm, suggesting that it has a differentstructure from :fraction6-10. The side chain conjugation might not contained in its structure.

1.8

I. 8 , .. 1.2

. " ... 6 • 2

WAVELJ::NGTH Cnm)

Figure 18. UVNisible Spectrum of Fraction 5 Collected in CA Autoxidation Reaction.

1.8

l. 4

l. 2

.a I .6 • 4

--2

WAVELENGTH. (nl

Figure 19. UVNisible Spectrum of Fraction 6 Collected in CA Autoxidation Reaction. 36

1.8

J.6

1.4

1.2

• 8

• 4

8 i3 6l N � � ·tn ......

WAVELENGTH (nm)

Figure 20. UVNisible Spectrumof Fraction 7 Collected in CA Autoxidation Reaction.

l.B

I. 6

I. 4

1.2

• 6

• 4

8 8 8 � N s.... � co ......

WAVELENGTH (nm)

Figure 21. UVNisible Spectrumof Peak 8 Collected in CA Autoxidation Reaction. 37

1.2

• 8

< . �

• 2

-. 2

..,AVELENGTH (nnr)

Figure 22. UVNisible Spectrum of Fraction 9 Collected in CA Autoxidation Reaction.

I. 4

!. 2

. 8

. 6

• 4

-.2 8 8 8 8 8 N ... "' "' .... �

WAVELENGTH (ni)

Figure 23. UVNisible Spectrum of Fraction 10 Collected in CA Autoxidation Reaction. 38 Liquid Chromatography/Mass Spectrometry (LC/MS)

The mass spectra of all fractions includingcaffeic acid are given in the

Appendix A. The MS data forproduct fraction5 to 8 are summarized in Table 2. As shown in the Table 2, fractions5-8 have the same molecular weight. They all contain the fragments 179 and 135. The fragment of 179 is m/z of caffeic acid. This confirms that the products possess at least one unit of caffeic acid. The fragment of 135 results from

[179 - 44 (CO2)]. Since m/z was only set in the range of 75 to 500 in the experiment, the fragmentof 44 (COOH group) could not appear in mass spectra of caffeic acid or fractions 5-8. Peaks 5-8 can be assumed to be dimers of caffeic acid with a molecular weight of 358: 313 +1 + 44 (CO2) = 358. Fractions 5-8 contain a major peak at 313 which is the product peak of the dimer (357) with loss of a CO2 (44). Although the

357 was present, it was a small peak as it can easily lose a CO2 of one of the acid groups. In fraction 9 and 10, fragments of 489 and 179 are found. The molecular weight of trimer of caffeic acid is 534, which is not shown in MS spectra because the m/z range was only set up to 500. However, it could also be derived that fraction 9 and 10 are trimers of caffeic acid according to their UVNisible spectra and LC/MS data information: 489 + 1 + 44 (CO2) = 534.

1 H Nuclear Magnetic Resonance Spectrometry(1H NMR)

The 1H NMR spectra of standard caffeic acid and all oxidation products are listed in Appendix B. Table 3 summarizes the 1H NMR data (in ppm) of standard and reaction products. The peak assignments for caffeic acid are shown in Figure 24. 39

Table 2

LC-MS Data forCaffeic Acid and Its Oxidation Products

[M-Hr CaffeicAcid Fraction-5 Fraction-6 Fraction-7 Fraction-8 Fraction-9 Fraction-IO

489 100 100

313 100 100 100 100

179 100 71 69 66 85 69 69

177 27 38 24 40

135 34 28 41 30 56

97 16 11 31 48

91 42 40 61 98

89 68 42 79 91

The protons at d positions in caffeicacid could not be seen in the proton NMR spectrum because they are freeto exchange with the D20 solvent:

HO DO Oi Oi HO� DO� 40

C1 C2 d d HO -Q-CH=CH- \ ;J COOH b a d HO C3

Figure 24. Peak Assignments forCaffeic Acid.

Table 3

1 H NMRSpectra Data (8, ppm) in D2O at 400 MHz

Caffeic Acid Fraction 5 Fraction 6 Fraction 7 Fraction 8 Fraction 9 Fraction 10

5.95 d

6.30(a) d* 6.38 q* 6.38 d 6.37 d 6.38 d 6.30 d 6.31 d

6.42 d 6.42 d

6.59 d 6.59d

6.90(c1) d 6.86 d 6.60d 6.53 d 6.54 d 6.87 d 6.88 d

7.06(c2) d 6.94 d 6.71 d 6.62 d 6.62 d 6.94 d 6.95 d

7.15(c3) d 6.97-7.21 6.95-7.22 7.03-7.19 6.93-7.20 7.03-7.13 7.05-7.14

7.27 d 7.30d 7.30 d 7.32 d 7.20-7.33 7.24-7.35

7.41(b) d

8.43 s* 8.42 s 8.42 s 8.42 s 8.42 s

# Letters in parenthesis refersto assignments on the structure. s = singlet; d = doublet; q = quartet. 41

In Table 3, compares the NMR data of product fraction 6 with those of caffeic acid. The alkene side chain is seen at 6.38 and 7.30 ppm as two doublet peaks. This indicates that the protons on alkene side chain are in trans form. The appearance of two doublets at 6.60 and 6.71 ppm indicates the loss of the second alkene side chain conjugation in fraction6. This result demonstrates the involvement of the double bond on the side chain of caffeic acid during the oxidation. The couplings of side-chain protons could be confirmedwith decoupling experiments. Fraction 6, 7, and 8 give very similar NMRdata. Small differencesare observed due to their steric effects.

Fraction 9 and 10 also have similar NMR spectra but different fromthose of fraction 5 to 8. We have shown that fraction5 to 8 were dimers of caffeic acid and fraction 9 and

10 were trimers of caffeicacid based on their LC-MS data. However, the spectrum of product fraction 5 is different fromthose of fractions 6, 7 and 8. Loss of two alkene side chain conjugations is indicated by the appearance of two strong doublet peaks at

6.86 and 6.94 ppm. Doublet peaks shifted from6.30 ppm to 5.95 ppm suggest the change of protons fromalkene to alkane structure. These results illustrate fraction6 to 8 are structural analogues. Fraction 9 and 10 are also structural analogues. The

NMR results are consistent with UV-Visible sepctroscopic results. In Xu's study, cyclobutane linkage dimers and ether linkage dimers were suggested as two different structures of dimers of caffeic acid oxidation products (Xu, 1994). This NMR experiment provides the further evidence to show that some of the products (fractions

6, 7 and 8) are ether linkage dimers (Figure 5) and one product (fraction5) is a cyclobutane linkage dimer (Figure 4). 42

Reaction of CaffeicAcid With Tryptophan

UVNIS

The UV-Visible spectra of standard tryptophan and products of fraction2 to

4 are given in Figures 25, 26, 27 and 28. Fractions 2 and 3 have two absorption peaks at 280-290 nm and 320 nm, suggesting structural analogues with caffeicacid. Fraction

2 and 3 might be dimers or trimers or mixture of dimers and trimers, which were produced by caffeicacid oxidation reaction. Fraction 4, on the other hand has a

UVNisible spectrum different fromthose of fractions2 and 3. The spectrum of fraction4 consists of both caffeic acid peak and tryptophan absorption peak. It has a strong absorption peak at 305 nm, indicating that the side chain double bond is not present in its structure. The fraction4 spectrum is very similar to the standard spectrum of indolepropionic acid acid. According to the UVNisible data, fraction4 could be an indolepropionic acid.

LC/MS

LC/MS data forreaction products of caffeicacid with tryptophan was very confusing. The reaction products were well separated on HPLC using UV-Visible detection. However, on the LC/MS system, no separated peaks were observed using the mass detector. This was caused by the strong background noise in LC/MS system that was not UV active. The complicated sample matrix was the primary factor that resulted in the strong background noise. 43

2.5 UV/V Spetrum ofO.lmM Standard Tryptophan

1.5

-.5 ij � 8 ....

WAVELENGTH C)

Figure 25. UVNisible Spectrum of Standard Tryptophan.

1.2

i . 6

• 4

ij � 8 ....

WAVEEGTH C)

Figure 26. UVNisible Spectrum of Fraction 2 Collected in Reaction of CA With Tryptophan. 44

UJ .s ••

• l

8 8 8 8 fij � (T) ... Ill ......

WAVELENGTH (nm)

Figure 27. UVNisible Spectrum of Fraction 3 Collected in Reaction of CA With Tryptophan

. 9

• 5 ••

• 3

• 2

• 1

� � � 8 ....

WAVELENGTH Cnm)

Figure 28. UVNisible Spectrum of Fraction 4 Collected in Reaction of CA With Tryptophan. 45 1 HNMR

The 1 H NMRspectra ofstandard caffeic acid and three products are listed in

Appendix B. Figure 29 (a) shows the structure oftryptophan and NMRpeak

1 assignments fortryptophan in D2O solvent. Table 4 summarizes the H NMR data (in ppm) of standard caffeic acid and tryptophan, as well as reaction products. Proton

NMRdata of fraction 4 shows both caffeic acid and tryptophan units. Peaks at o=3.31, 3.49 and 4.07 offraction 4 correspond to peaks at o=3.31, 3.50 and 4.08 of tryptophan. Small differencesfor chemical shiftsat these peaks are observed. NMR peaks at o= 6.87 and 7.43 of fraction 4 correspond to NMRpeaks at o=6.90 and 7.41 ofcaffeic acid. A large peak at o=9. 77 is also found in 1 H NMRspectrum of fraction

4. The large peak could be a shifted peak caused by the linkage ofcaffeic acid with tryptophan. Fraction 4 is an incorporated compound between caffeicacid and tryptophan. Fraction 4 is one of indolepropionic acids. Structure of fraction 4 and

1 peak assignments forit are shown in Figure 29 (b). The fact that H NMRdata of fraction2 and 3 in Table 4 are almost identical with 1 H NMRdata of fractions6, 7 and 8 of the caffeicacid autoxidation reaction in Table 3 suggests that of fraction2 and 3 have the same structures with fraction6 to 8 in caffeicacid autoxidation reaction, which were shown to be ether linkage dimers ofcaffeic acid.

Comparison Study

Comparison studies were performed on the analytical HPLC system. In the firstpair of reactions, caffeic acid autoxidation reaction with and without addition of 46

a C H H 0 I I II "r----.--c-c-c-ooI I H ND2 N b D

(a)

c a D g1 (;OOD H H � h a I · 1_1 __u)I I H-c-N-<;: <;: � g2 I e I I I DO-CH D COOD H

DO OD

(b)

Figure 29. Peak Assignments for Tryptophan (a) and Fraction 4 (b). iron(III) ion, the products of reactions were analyzed after two months. Two very similar chromatograms were obtained. The similarities of chromatograms and UV study suggest that the iron(III) ion was only involved in catalysis of the reaction. 47 Table 4

1 H NMR Data for Reaction of CA With Tryptophan

Caffeic Acid Fraction 2 Fraction 3 Fraction 4 Tryptophan

3.3l(a) q* 3.30(a) q

3.49(b) d 3.50(b) d

4.07(c) d 4.08(c) d

6.30(a) d* 6.39 d 6.38 d 6.42(d) d

6.61 d 6.62 d

6.71 d 6.70d 6.87(f2) d

6.90(c1) d 6.98-7.24 6.97-7.23 7 .20(f1,g1) 7.18(d1) t

7.06(c2) d 7.30d 7.31 d 7.30(f3, g2) 7.32(d2) q

7.15(c3) d

7.41(b) d 7.43(d) d

7.60(g) d 7.55(e) d

7.71(h) d 7.72(f)d

8.43 s* 8.44 s

9.77 s

# Letters in parenthesis correspond to the proton assignment in Figure 29.

* s = singlet; d = doublet; t = triplet; q = quartet. 48 This result confirms the catalytic function of iron(III) during the reaction. Similar results were obtained during the study of reaction of caffeicacid with tryptophan with and without addition of iron(III) (Figure 30 and 31 ). The products of two reactions were also found to be same, indicating the iron(III) ion played a catalytic role during the reaction of caffeic acid with tryptophan.

The last pair of reactions, caffeic acid autoxidation reaction and reaction of caffeic acid with tryptophan, were studied under the same separation conditions used in separation of reaction products of caffeic acid with tryptophan. By comparing these two chromatograms (Figure 32 and 33), fraction 4 in reaction of caffeic acid with tryptophan was identified as a previously unreported compound in the reaction.

Fraction 2 and 3 in reaction of caffeic acid with tryptophan are ether linkage dimers of caffeic acid based on their proton NMRdata.

1 /

0 1'1 11J 1'l c:: 1!) 11) (1J 0 I\ 0.. �-17' 1r1 1» 0 1!) Ii � i:i::: ·� 111 0 I > 1!) ..... l'l I') 0

Time (Minute)

Figure 30. Separation of Reaction Products of CA and Tryptophan Without Fe3+ (After 2 Months). 49

[\ '°11',

1t (') (l) Cl) ru i:= ru 0 ri IJ) 0.. Cl) � � (l) � � rt, IJ) [\ (l) �ii'! .::: \\J � .....ro Iii Q) �

Time (Minute)

➔ Figure 31. Sef+aration of Reaction Products of CA and Tryptophan With Addition of Fe (After 2 Weeks).

(l) Cl) i:= 0 0.. Cl) (l) � >(l) -ro Q) �

Time (Minute)

Figure 32. Caffeic Acid Autoxidation Reaction at pH 10 After16 h. (Analysis of 10-µL Sample by Cl8 Reversed Phase HPLC. Mobile Phases: 70%A, 0.1% Formic Acid; 30% B, Acetonitrile. Detection at 280 nm.) 50

11,) rn =0 rn 11,) i::i:: 1 CA 2 3 4 -�11,) a:, "-,, "- ,, Q) "',,, "' " - i::i:: " .; -"'

"' ...0:

Time (Minute)

Figure 33. Reaction of CaffeicAcid With Tryptophan at pH 10 After10 h. (Analysis of 10-µL Sample by C18 Reversed Phase HPLC. Mobile Phases: 70%A, 0.1 % FormicAcid; 30% B, Acetonitrile. Detection at 280 run.) CHAPTER V

PROPOSED MECHANISM

Based on UV-Visible spectra, proton NMR spectraand the comparison study,

Peak 4 in reaction of caffeicacid with tryptophan was identified to be an incorporated compound between caffeic acid with tryptophan. Because of the pH dependence, the phenolate anion was believed to be involved in the reaction. Under alkaline condition, oxidation of caffeicacid easily led to the formationof radical anions of the corresponding semiquinones. Then, tryptophan was added to the semiquinones as a whole molecule to give an Indole-Propionic Acid (IPA). The reason we believe tryptophan attached to the side alkene chain of caffeic acid is that the side chain of

caffeic acid is more reactive. The loss of the side alkene double bond was verified by

UVNisible data. The mechanism we proposed is different formthat proposed by

Flaig et al (1975) because caffeic acidcontains an active side alkene chain. It should

be noted that although our proposed finalproduct is in agreement with our results, it may not be the only one. The mechanism and final product were proposed as a

possibility and further analysisis needed to confirm them.

Under acidic conditions, the reaction is very slow, taking weeks and months.

The reaction rate is accelerated under acidic condition with presence of iron (111). Xu

(1994) reported another transition metal ion-copper (II) had a catalytic effect on

caffeicacid autoxidation reaction. A catalytic reaction mechanism was also

51 52

�COOH -oH ◄ ► ·o HO HO HO o• HO 0 OH

H coo I , la -Nca-ca-1 I =I • co�

�

COOH _L()� I I § H-J:-N-(:H-CH2 I I I H H COOH �

HO� HO

Indolepropionic Acid

Figure 34. Proposed Mechanism forReaction of Caffeic Acid With Tryptophan. 53 postulated by Xu (1994) forcaffeic acid autoxidation reaction with presence of copper (II). CHAPTER V

CONCLUSIONS

The reactions of caffeic acid with tryptophan and caffeic acid autoxidation reaction were studied. Products of the two reactions were compared on HPLC under the same reaction and separation conditions.

The experimental data demonstrates that both hydroxide and transition metal iron(III) ion catalyzed both caffeic acid autoxidation reaction and reaction of caffeic acid with tryptophan. The probable first step in reaction of caffeic acid with amino

3 acids must be the oxidation of caffeic acid. This suggests that the OH- and Fe + must catalyze the oxidation of caffeic acid. This would also suggest that the oxidation of caffeic acid is the rate determining step in reaction of caffeic acid with amino acids.

In this study, caffeicacid autoxidation reaction gave six finalproducts, three of which were found to be ether linkage dimers of caffeic acid and one was cyclobutane linkage dimer of caffeic acid. The other two were thought to be trimers of caffeic acid. Of the three products in reaction of caffeic acid with tryptophan, two are caffeic acid ether linkage dimers of caffeic acid. By identification with UV-Visible spectrophotometry and proton NMR, the new product was identified to be an incorporated compound between caffeic acid and tryptophan.

The fact that caffeic acid can react with tryptophan indicates that this mechanism could be used to incorporate nitrogen into humic substances in natural

54 55 waters. This also suggests the possibility of the reaction of caffeic acid with man made compounds such as pesticides in natural waters. This study also demonstrated that although the oxidation reaction is slow at neutral pHs, it can be catalyzed by

+ transition metal ions like Fe3 , which are widely present in the environment, at low pHs. Appendix A

LC-MS Spectra of Products of CaffeicAcid Autoxidation Reaction and Reaction of CaffeicAcid with Tryptophan

56 57

WANG YOUNG-1 2: Scan ES- 10.92 100 1791.00Da 1.98e3

10.99

! %

16.53

18.2

19.03

Total Ion Chromatogram of CA Autoxidation Reaction Mixture in LC-MS System 58

ANG WANG YOUNG 595 (10.93S) Cm (595-569) Scan ES- 179 1.12e3 10·

o+,,�,'l-;+14+...f.;.;.J..�>-1-,-e!�.W,...��=� � � � � � � � � ™ � rrJz

MS Spectrumof Caffeic Acid in Its Oxidation Reaction 59

,ANG 'WANG YOUNG-1 448 (16.459) 2: Scan ES- 313 1 592

179 89

135 1n

269 97 314 137 98121 180

95 ...... AAl,J,��U-ll,U�J,J,..,...�,l,IJ,J,IU.Jlll,lllll...... ,...,.w.Jl-._..,.._..JA-..��mlz 5 6 6 70 75 8

MS Spectrum of Peak 5 Collected in Caffeic Acid Oxidation Reaction 60

2: Scn ES- 313 624

179

89 1 135 177

269

MS Spectrum of Peak 6 Collected in Caffeic Acid Oxidation Reaction 61

ANG W�G YOUNG-1 520 (19.100) 2: Scan ES- 313 100 328

179 91

97 135

113 177 137 314 269 llll

5 60 65

MS Spectrum of Peak 7 Collected in Caffeic Acid Oxidation Reaction 62

ANG WANG YOUNG-1 550 (20.200) 2: Scan ES- 100 313 209 91

179

135

% 97

137 121

212 269 314

�""""�.,_,_._.-,µ,,,.....,.....,....l'l',+.-1'.+,f'AAJl�'ffi',-J.++-l'.,--,.��=m/Z550 600 650 700 750 800

MS Spectrumof Peak 8 Collected in Caffeic Acid Oxidation Reaction 63

�ANG WANG YOUNG-1 813 (29,848) 2: Scan ES- 91 100 127

489

179

137

117

212

113 90 197 166 494 475 347 375 391 416 449 497 364

250 300

MS Spectrum of Peak 9 Collected in Caffeic Acid Oxidation Reaction 64

2: Scn ES- 489 121

7 137 179

121

212 10

18 44-4 91 '7

� �,-11-,1,-l--rl',lh',lll!--�-,..,.,..��=rrJz � � � � � � � ™ �

MS Spectrumof Peak 10 in Caffeic Acid Oxidation Reaction Appendix B

1H NMR Spectra of Products of Caffeic Acid Autoxidation Reaction and Reaction of CaffeicAcid with Tryptophan

65 ------·--·· ------, vi Jcnffcic_ocld.S •"00 -- PIWC'aa,zwo U.JlAMSTH.S ---- Single Pulse Experiment dc_be.laoo• -� I 0.J(N•J ( � fft I l ... -ohia�••-

0 .;

---- ACQOIII'l'Joa •AaAalat'Da ...... _ PU• .... • oaffelo_acid.l Author • 8c!hrelber •••••rob Oro a-..pla ID • HHUIO Coat.•t. • aiqle hla• bper.i.lNn cr .. t.loa Dat.e • 21-IIAA-UOO U1H,J7

an-bloa Det.• • Jl-lll.ll-2000 1'1JC1J7 apeo 11t.e • aoll•••• uo � 8pac Type • Dn,T...... Det.a Poraat • lD COIIJILU DS.-.alODe • X Dia Tit.la • 111 Dia lhe • 1UU _ ,j Dia V.h• • lppal Local_tiae • 21-IIU.-2000 U1S11>1 � Loak_gab • 15 r--- -.ovr_..1. • 12 a.laxatioa_dala,-. • laJ 0 . I ...... 901..-.at • ..D20 lpta_..t. • Ul■•J 't_.,_..t • 22, J ldCJ � x_offaet • s lPPl-1 :t_-..p • s.1111002,cuaJ

�

. ---1 ••••••• • •---•--• •..,._ .. ..,,.. __.,__ :- le ''r __ .....-.-...... _ ___,_.....,_,_��JV-\._....-�

··•1 1 ·•····••1••·•• 1 •••1·······••1••·····•·q,,,,,, ... ,,,,,,,,,,,,,,,,,,,,1,,,,,,,,,,,,,,, .. ,,,,,,,,,,,,,,,,,,,,,,,,, 7.S 7.4 7.3 7.2 7 7.0 6 u u � u /\ ))\ i f u "I 0 : o 11 � - M " lflO\ 0\ "' - "'iV> OOt-- Ill " " ""!""!��"'r -� O)"! t--t--t--t--r--r-- �! r..'. r..'. "'"' 5 � "=' X: purts_pcr Million: lH_...:=---=--===-====;� ====

1 H NMR Spectrum of Standard Caffeic Acid

O'\ O'\ P£AK 5.J ---- Pao•aa?NG PA&J8 --- Sing!; Pulse with observe prcs:1lurution CJJl-c• •ex I O,l(JI) 0.. -Ut 1 1

0 "

--- AHlflDI P.. ----

�

��, ...... ,----r...... 7.5 7.4 ,)i\, (,,9, 6.7 6.6 6.5 6.2 6.1 6.0 (9 5. 7.1 .r:��< 6.8 6J 71J 11 � ��N�•� �o��N�O•� �l�� '."1'''. . ' ...'. ''' iNr-•"l�� t--r--r--,-.r--t--1'��!� !!Ssi��l�:r-rrrtI11II I1'Cl10 .,.,

__ X _.;�.!� per Mi llio�-·:·�� -----"------· ,==== __ .,_..,=-=-... ,·�-·-·==---=---=---=-=---=-.,.-•.=-1-- ---·--····

1 H NMR Spectrum of Fraction 5 Collected in Caffeic Acid Autoxidation Reaction

0\ -...J l'cuk 6.J ---- PROC.ce•INO PAJI.AXKTZR8 --- dc_bel&.110• Sin�I� J•ulsc with uhscn•c 1u·cs:1lur:lliu11 -�-fft. I 1 -cbinapb••• 0 .,:· ...

"'�

--- ACQUJHTIOlf fAJU.IISTmUI ---- Pile•-- • ••ak....l,, btlloc • 8oJlreU..r ..••••.rob Oro haple U • 11737255 Coat.-t. • ■iagl• hhe wit.ll obae "'� er ..uoa JMt• • s-ua-2000 1t1U1H •••l•ioa9pea: Dat.e • c-ua-2000 1S, so Is, ■it• • aoUpaa• too

bea � • DSL'l'A...,JaD. De.k ronaat. • 1D ccal:PLU Dl.aeluioea • X � Dia Title • ll .. Dia Iba • 1UH Dia Oo1te • (ppaJ Looal_u.. • s-ua-2000 11, u Iss Loak.Jaia • 11 JleOTrJ•i• • 1t a.1... uoa._delay• lC■J aoaA■ • 121 SOl.....,t • D20 9pi.a...9et • 11 c■aJ � 'l'_..t ... • 22.S(dCJ ._ou.. t • ,.aosn CPSaJ � • S,HH002UllJlaJ

,-1·

� � ..,, .2 > a \lvft� 1,,-1<1>.y-)i.,i.1"(,1Af\\.1>111�l>i1ifl'At'��1-N11¼1,�·w,1 � , 111/ 'fl' Y' ½-- 11J 'y'\. , I \/ �� I I i � �- i!! � E � � � - ::i ::! � fl "' :':! ri- _,. r-r "" '" , , .. ·•T••,· -r•1 - ,,-. •• •·-• • • r' • '•· 'I'''' ,... . ,..., I,-, ,, .. ••-.��""- ,-,�"T'"T""r'"1''.,..;,f-,.... ,.,,... •..,...,....- ,...� r.-r-.-.� 8.6 8.5 8.3 8.2 8.1 8.o 1.• 1.8 1.1 1.6 1.5 1.• 1 1 1\ 6.9 6.a 6 6 6.5 6 6 f3 f"i _ .1 1.°1 r 1 I r :.. 1-. ....b, ") :;i�M""" ,.1 ""'":ti,J� L r;"°"'°""tn ....'O"l 'C N 11'1 ,..OIN 1 Vt j : r--��;::� :: �::";: �J

f- �: J!nrts per Million : lH l.:�!

0 1H NMR Spectrum of Fraction 6 Collected in Caffeic Acid Autoxidation Reaction °' 00 l'l•uk�7-4 ---- PIIOC&88I.NO PA.IUJUT&Jt.8 ---- l'ulsc wHh uhscrvc 1>rcsaturnt;uu 4C_baluc:e -··t'ft I l � .('""'11 aaohin-»ba••- I

II i QI I, ! ---- ACQUUITION PAJI..Ud:TUJ ---• ,,;· 1 rile •-- • Peak_, ,S ,, Author • kbreiber Jle■earc:b Oro 'i Saaple ID • H7HOU Coat-t. • •111crl• hl•• with ob•• '1 cr-tioa. D&t.• • s-ua-1000 11,ot,11 ••..-idoa. D&t• • ,-ua-2000 u,1111, 9pea lite • &clip••• 400

9P.a Type • DZLTA._aJl Data roraat • 1D COWPL&X Diaaa•ioa.■ • X Dia Tit.le • 1N � Dia lhe • 11314 Dia Vlllit■ • [PPII) Local_tiae • s-ua-2000 11,0,115 LockJa1.a • 1f JleOY¥"_1J&ill • lf Aela.. tion_delay• ll•J •a--■ • 121 801•-t • DlO ap1._v•t • 1f (NaJ T...,,_get • U,4(dCJ X...offHt • 4. IOS'7 Jppa) lL--P • S.UH0014(kNa) �

iiM �' �1"/��i1y1�iffflM"'��M�t�� , j 1

·1 ,�, • • r •' • i' •·• •·1 ,., • r I•••· 1·• • 1 • 1' 1 1 •, ,�' • ,·,.,.,I'' r, 1' 1 •', ,., ,-, I',.,.,.'•• I'''' r•' , .... ,.,...--r. ..,·, '• ,, ',-,-,I,., ,..... r 8.3 8.2 8.1 8.0 7.9 7.8 7.7 7.6 7.5 7.4 6.9 6.8 6 8.5 1 1.1 o 6.1 6 f4' I � / - I ( I "' 1b � Ill t"--0\ Nlll\0,...0i f "' "'� i;! i;;::l �;\\"' ;,;;:; i 't: "!"!�"l"I ...."l"l 00 � r-: �� '

1 H NMR Spectrum of Fraction 8 Collected in Caffeic Acid Autoxidation Reaction

-...l 0 ---- fROC&81INO HJlAUT&ll8 --- doJ,alanoe l'•�'l;,}p Sh)��•uJsc with observe prcsuluratlon -··fU I 1 : aaalllaaph•• ; -

----- ACQVIH'fICllf f� ...... PU•.... • feak...,I.J Aotllor • khr•U:ter h•••rab Oro haph; XD • 8t7110ll C•t•t - aiacrl• hl•• ritb ob•• ; CrMtioa kt• • 1-ua-2000 u,22,22 a..1-1oe Date • 1-ua-2000 22121120 8peo llt• • Sc:llip••• ,oo

apeo� • Dat.1'� Data Pora.at • 1D COIOLSX D1-aioaa • J: Dia Titl• • 111 _r� Dia lh• • 101( Dia VD.it• • (ppaJ Loaa1_u... • 1-ua-2000 21,21121

a.a.r_..1. • :n1J � Look..Jr•U••laaation .. delay• l(a) ka11a • JSI aol.,_t • D20 9Jo1.IL.-t. • 1S [■st 1' ...... t. • 22.4(4C) :l_ofbet • •• 71St (ppaJ z._ ....p • 5.UIIOO:U(kNa) � 3 g :l

j= o- � �-� .....,....,....., �•,-,-.-, r-1''' •I•,...,..,,,,...... ,..,..,..,'' I'•'• I,., ,-.·f""'"""'"•·• 1·r.-•-.7...... -.---'"T""-• rr-,•.-r�rr�� r-•·.....,-,·,...... -1�·•-,·1..-.-,��..,..j

8.3 8 .2 8.1 8 0 .8 . 7.9 7 7.7 7.6 7.5 7.4 ��)r� . U 6 7 6.6 U �� (� r . � ��l��s ��wW��w� H J______� :__ �u�ta pc� -�!�!�n .: �� -�•.,.•'-'�,•.•:,-::_:_.,.�,:-:.=..,:+�...... , .,r ••-•·•'•:::"=-';"::.:.._ ;-·-�:;::.•.'0;,�1•= .... ,a:;"'-=-:' .. ,.,.._,, .._,.,

1H NMR Spectrum of Fraction 9 Collected in Caffeic Acid Autoxidation Reaction -..J..... Peak �0.3 ---- ••oc.caaJ:NO J>>.a.AMrTERS --- dc_balaace � �Singl'lrulsc with ollscrvc i,rcsulurution •e.xp 1 0.2{H•l fft I 1 ! -

0 .;

0 ,..: -••- ACQOHI'UOlf P.u.ADTD.a ---- Pih .... • •-Jt....10., Autllor • kbreiber •••••rch Oro aaapl• ID • •UJUt7 0 Coat.eat • •1nol• l'ul•• •1t.b ob•• ,.; CE.. t.1- Dat.• • 11-ua-2000 11, 11, 01

aa,,ieioa Date • 11-ua-2000 11127 110 8peo •it.e • SC11P••• ,oo

0 � type • D.CLT,\JIN1" .,; Data l"oniat • lD C'OII.PLl:X Diaeoeioa.a • X Dia Tit.le • 1■ Dia •1•• • Ull• Dia unit.a • (p�J Local_u... • 11-u•-2000 11117 105 C! Loclt_gaill • u " �-••in • 21 llelaxat.ion_deley• l(a] ac-• • 251 aG1YeDt • D20 a»i.D_.pt • 1( t••J C! t"eap_g•t • 22., [dCJ x_off.. t • 4.71U4(p�J ... X_,.-..p • 5.HII0024[kHsJ

�

'i 0 .2

C! I 'j � '"'T""f·.....--rr1"T"">-.·••·T....-. • • r· •·r� ,--·.-. ,, ...-.-r-.,...... ,...... ,...... ,"T� ,-,....-�.,...... ,....,...,,.....--r'""'...... � -�..,-.-!'"T".,....�

UrUULI.., � UU�UUUUUY=UUUUUMU �

_X'° :_parts_pcr Million: 1H ·•=-=�=�=:��e.•c��=====:=�- --===�-----

1 H NMR Spectrum of Fraction 10 Collected in Caffeic Acid Autoxidation Reaction

-.I N Tryptu1•ha11.4 , 1 -o - HOC&aSINO PAJU.tCftlUI -- do Nlaaca "! Slnglc .. ulsc wllh l•scr� presu111r;1ll11u :?:. j r I ;;;; :::-!-i ·.; e: -1 -clliaapll•••...

- ACQUl■ITIO■ Pu..ucnsu - rue .... • Tryptop11... • Autltor • &ohreibar •••-rcb Oro .... h ID • a111otn Coata■t • lia9le hl•• with ob•• cr-uoa Date • JO-NA.1-2000 Jl1Jt1st

•••i•i- Data • 1-.ni■-2000 ot,st,o, 8peo aha • lclip•.+ 400 ...Cl 1'J'l>e • DILTA_IIMII; Data ro.-.at • 1D COMPL&:1

D1-•■ioa■ • I Dia Tit.la • 1■ Dia Iha • lUH Di• U■lta • (p.,.J C, Locral_U_ • lO-M.lT-20OO 231lt1S'J .;· Lock 9■ia • U aeo.-i_,■la • Jt a■laut.J.oa_d■hr 11 ■ I ka■a • I kl•■at • D20 lph_9at • 15(■-J T-. 9■t • 25,S(dC) �� I i .1_of1.wee .. t • t.11u,1pp.J 111[� p • s.tttlOOHJkla) :1_ . ! � !. ;I : ..; I C>· ·, ' - -� , �J '" .)'-,., .,,·--...,;•� � .ll. I

=� ::1.

"! ' �-,- ,-��.-�-�� -�·-·-•-� ·•- �,•·c 7. 6.0 ,;;-C!L-8.0 0 5.0 4.0

--X·-·----· : ports -per-· Million- . : lH' . -· , _____ ·------=-.�� =·--··-•-•- -=·-··-·-----·--==• - ..

1H NMR Spectrum of Standard Tryptophan

w-.) CA+T l't.i1k. 2.4 ---- rllOCESSJHO PAltAMt:ttllS --- de bal■nct1 , ■ei"p 1 0.21Kal : : , ....l '",, .. ., ...... , .... ., ... ;. ...fft , l

C, . --- ACQUISITION Pll.AMl'TU1, --- :;: Pil• ••- • CA+T Peak 2.5 Author • Schreiber-Jloacar-ch Oro : la•pl• ID • 11625010 C, Cont•nt • llngle Pulae "1th obae .; Creation Data • 21-MAT-2000 1115215,

111:eviaion Date • 21-MAY-2000 17155107 Spec Illa • !clipae+ 400 C, .; Spec Type • DELTA_IDUt Data f'or.. t • 10 COMPLEX , Di••naion, • l Di■ Title • Ul C, Di■ Iha • lUU " · �-�"!!:. : ir�Ki,-2000 u,s2,St Loek_iahl • U . lllecvr_9aln • 20 C, 111.alaau.lon_dalay• 11•1 Scana • l!U .. lol•ant • D20 lpin_9at • U(fll) , T-p 9at. • 2J.4fdCJ C, J:_ot1Ht • 4.7167lfpp.( vi x_awaap • 5,UIIOO24(kllt) Q�1:

C, ,.;

C,"' � I jj , 1- 1] �,)� �I! I� i- C, I tf l • · l, ··,!••···I•·,·;•••, I,,,• 1 • • •, 1,,, •I"'••·1 • ·, ,-,T• ·-. r"'"' rr1 ,.,.,.,-,-,• t·rr, • ,,...r,·r-. r1 • • ,. ., 1·• •·• •·1 •, •, , ..'1' .-, .- • ,-..-,,-,-.• -.,-, •, •·1 r.-, I• j 8.5 8.4 8.3 8.2 8.1 8.0 7.9 7,8 7.7 7,6 7,5 7.4 7.3 7.2 7.1 7.0 6,9 6.8 6.7 6.6 6.5 6.4 6.3 6.2,

llllon '.-": X: purls_ pcr.� _: _lH_-,-- ·-·.::--:;-;-__ ,-�===---.:•=.;.-===. -..e:•.::-;-;-·-...:--:c.-�- -=-�-- -� -••- •. ._,..=-..;.-:-::.- ,---=.-.--.

1 H NMR Spectrum of Fraction 2 Collected in Reaction of Caffeic Acid With Tryptophan

--.J +:>, ! l't;k I

5 --- ACQUISITION PARAMETERS ---- P'llo Naao • CA+T Poak J, 4 Author • lchr;ibar-R••••rch Cro

! Saaplo IO • Sl4l0Ht 1 Content • Single P11l10 with ob10

Croat.ion Oola • 21-MAT-2000 lJ lJ1421 11.ovialon Oolo • U-MAT-2000 12141 10 c,! i Spac Sito • lcllp11♦ 400 �; o i j Spoc Typo • DELTA_NMR Data P'on,at • 10 COMPLEX .. I Dlacm11ona • X Dia Tit.lo • 111 0 I Dia Sha • 1Cill4 j t ll Ill 1 ,.; Dia Vnita9 • f PP"I I Local_�:��r ;!in1ao •: 21-MAY-2000!! '11 l ( • : I Rol011:it111n_dohy• J lc•n• • � 1I :;t::;!t : �!f121 "- 1 "' Ta■p got • 2l.4(dCJ X_ot7aat • 'I. I 7162' f pp111) � X_awaap • 5. 9UI0024 I kH� J .. I o · I

,..; ! 0' I I N'. I '\ riIr. 1� 1 ! q ' I · 51 I ,!! Iir.i : ' 0 ,1� I �l i! � ' J ,/ I I• ; 5', ;!: J�I\ .A.1 �1�( : ,g �·1 1'rf',r" \,� ! i �i.,� v�,,;,.,�wi,.,•w�V ' . ';1•·•;••"·'········!•••·:· ·•1•· .,, ...,, ...,,,., ,.. ,., .. ,,,,,,,,,,,,,.,.,,,,,,,,.,,.,,,,.,.,.. ,.,.. ,.,.,,1 H.5 8.4 8.3 H.2 8.1 8.0 7.9 7.8 7 7 1., 7.5 .4 7. 7 . 7 1 ., 6.8 6.5 . I lf\ \"i \ "' 6 1 i ir r n "' 1 I ifl!tl!:; :ei i!!rii .,,� :!)� .t�lr-")llilO r:t NO !N � ,-. I s �� '4\0 �� \O� : :J ; I 0I X : 1>orts p�r �llllon : lH

1 H NMR Spectrum ofFraction 3 Collected in Reaction of Caffeic Acid With Tryptophan

--.J Vl PEAK 4.2 --- PROC&S&INO PUAHETSU - SlngliPulse with observe presacurulion do_b•l•nce :n!·t -c)lla•pb•••... �

� - ACQUUI'l'JOW P.u&NSflU -- Pile w... • PLUt_t,J Autbor • acbrell>llr •••••rob Oro auapi. ID • au,us Coateat • ■l-,le hl•• vitb ob•• � CrNtio• O.te • Jl�l'-2000 02101125 •••l•loa Date • 2-JU■-2000 21101111 lpeci lhe • lcUpae+ 400

8pec Type • D<A_... "! Data Pcu·-t • 1D COMl"UI .... D1.-nalona • ll: Ola Tit.le • l■ Di■ ■i•• • 1'Jlt Di■ U■ih • lppa) 1.ocal_tl- • ll-MAJ-2000 0210l1ll l.ock_9aln • 1' aaovr_galn • 2J � aalaaatloa_dalay• 11 • J acana • 251 Solvaat • 020 ■pin_9at. • n fll•I Teap vat • 25.tjdC) x_of7.. t • t,71Ufppal � X_■waap • 5.UIIOOltlkHll)

�"'- ,

�

I�

;l, I• • I • I 10.0 9.0 8.0 7,0 6.0 5.0 4.0 3.11 X : tJnrtli Jlcr MIiiion : IH I

1 H NMR Spectrum �f Fraction 4 Collected in Reaction of Caffeic Acid With Tryptophan

-..J 0\ BIBLIOGRAPHY

Challis, B.C. and Bartlett, C.D. (1975) Nature, 254, 532-533.

Chi-Tang, H. (1992)ACS Symp. Ser. 507, 2-7.

Cillers, J.J.L. and Singleton, V.L. (1989) J. Agric. Food Chem. 37, 890.

Cillers, J.J.L. and Singleton, V.L. (1991) J. Agric. Food Chem. 39, 1298-1303.

Cohen, M.D.; Schemidt, G.M.J. and Sonntag, F.I. (1964) J. Chem. Soc. 2000-2013.

Coseteng, M.Y. and Lee, C.Y. (1987) J. Food Sci. 52, 985-989.

Cresser, M. and Killham, K. (1993) Soil Chemistry and Its Applications, Cambrige.

Deina, S.; Gessa, C.; Manunza, B.; Marchetti, M. and Usai, M. (1992) Plant and Soil, 145, 287-294.

Flaig, W.; Beutelspacher, H. and Rietz, E. (1975) Chemical Composition and Physical Properties of Humic Substance. In Soil Components. Vol.I, Springer Verlag, New York, 84-88.

Flaig, W. (1988) Generation of Model Chemical Precusors. Humic Substances and Their Role in the Environment, John Wiley & Sons, 75-92.

Fulcrand, H.; Chemminat, R.; Brouillard, R. and Cheynier, V. (1994) Polyphenols 94, Palma de Mallorca, May 23-27.

Hurrell, R.F. and Fiont, P. F. (1984) Adv. Exp. Med. Biol. 177, 423-435.

Jiang, L. (1996) An Investigation on Nonenzymatic Autoxidation of Phenolic Compounds in Natural Waters. M.S. Thesis, WesternMichigan University.

Laranjinha, J.A.N.; Aleida, L.M. and Madeira, V. M. C. (1994) Biochem. Pharmacol. 48, 487-494.

Lehmana, R.G.; Cheng, H.H.; Hersch, J.B. (1987) Soil Sci. Soc. Am. J. Vol.51, 352- 356.

77 78 Linder, P.W. and A. Voye. (1987) Polyhedron. Vol. 6, 53-60.

Matheis, G. (1987) Polyphenol Oxidase and Enzymic Browning of Potatoes (Solanum tuberosum). II. Enzymic Browning and Potato Constitutents. Chem. Mikrobiol. Technol. Lebensm. 11, 33-41.

Olsen, R. A, Brown, J.C., Bennet J. H. and Blume D. (1982) J. Plant Nutr. 5, 433- 445.

Pierpoint, W.S. (1966) Biochem. J. 98, 1202.

Schnitzer, M. (1991) Soil Organic Matter. The Next 75 Years, Soil Sci., 151, 41.

Singleton, V.L. (1987) Am. J. Enol. Vitic. 38, 69.

Sposito, G. (1989) The Chemistry of Soils. Oxford, New York.

Stevenson, F.J. (1982) Humus Chemistry. Genesis, Composition, Reactions, John Wiley & Sons, New York.

Stevenson, F.J. (1994) Humus Chemistry. Genesis, Composition, Reactions, 2nd ed., John Wiley & Sons, New York.

Stevenson, F.J. (1965) Gross Chemical Fractionation of Organic Matter. Methods of Soil Analysis, American Society of A Gronomy, Madison. Wisc. 1409-2421.

Stumm, W. and Baccini, P. (1978) Man-made Perturbation of Lakes. Lakes Chemistry, Geology, Physics, A. Lerman, ed., Springer-Verlag, New York, 91- 126.

Thurman, E.M. (1986) Organic Geochemistry of Natural Waters, Nijhoff, Dordrecht.

Van Buren, J.; Devos, L. and Pilnik, W. (1973) J. Food Sci. 38, 656-658.

Waksman, S.A. (1932) Humus, Williams & Wilkins, Baltimore.

Woodring, P.J.; Edwards, P.A. and Chisholm, M.G. (1990) J. Agric. Food Chem. 38, 729-732.

Xu, L. (1994) The Reaction of Caffeic Acid andCopper(Il) in Natural Waters. M.S. Thesis, WesternMichigan University.