-
Using Esterase and Laccase Enzymes to Derivatize Bioactive
Plant Phenolics for Altered Chemistry
by
Mohammed Sherif
A thesis submitted in conformity with the requirements
for the degree of Doctor of Philosophy
Department of Cell and Systems Biology
University of Toronto
If someone said
© Copyright by Mohammed Sherif 2015
-
ii
Using Esterase and Laccase Enzymes to Derivatize Bioactive
Plant
Phenolics for Altered Chemistry
Mohammed Sherif
Doctor of Philosophy
Department of Cell and Systems Biology
University of Toronto
2015
ABSTRACT
Plant phenolics have notable antioxidant activity and there is
potential to improve their
action by chemical modification. Two enzyme classes carry out
reactions that can act on the
hydroxyl moiety of phenolics. Esterase enzymes can be used in
non-aqueous solvents to esterify
a long chain acyl group onto the phenolic compound. Laccase
enzymes can be used to form
phenoxy radicals that can then couple to form larger molecular
weight oligomers. Both
enzymatic modifications may produce a new antioxidant with
altered chemistry.
One archaeal esterase (AF1753) from Archaeoglobus fulgidus and
one bacterial esterase
(PP3645) from Pseudomonas putida were assayed for activity in
organic solvents. Both
enzymes catalyzed hydrolysis of phenyl acetate and vinyl acetate
in 98:2 (v/v) (t-amyl
alcohol):buffer; with continued activity up to 96 h of reaction.
However, the enzymes were not
able to catalyze transesterification of 4’-hydroxyacetophenone
with vinyl acetate in 9:1 (v/v)
cyclohexane:(t-amyl alcohol), which was not explained by enzyme
inactivation during
lyophilization. Still, alanine scanning mutagenesis revealed
that R37A substitution improved
activity of AF1753 on long-chain p-nitrophenyl (pNP) esters.
A multicopper oxidase (SCO6712) from Streptomyces coelicolor
displayed activity on a
variety of phenolics including caffeic acid, ferulic acid,
resveratrol, quercetin, morin,
-
iii
kaempferol and myricetin. Among the products formed by action on
flavonols were dimers of
quercetin, morin, and myricetin. Quercetin and myricetin dimers
showed longer retention time
on reversed phase chromatography. All three dimers could be
detected by 5 min of reaction but
depleted by 3 h and 24 h. The TRAP and FRAP antioxidant activity
of the whole reaction
mixture of modified quercetin, morin, and myricetin decreased,
as starting phenolic was
depleted over 24 h. Accordingly, mass spectrometry was used to
shed light on the molecular
structure of the dimers produced from quercetin and myricetin.
In both cases, mass
spectrometric analyses ruled out dimer formation through the A
ring of each monomer. For
myricetin, the most likely linkage structure was determined to
be between either two B rings or
a B ring with a C ring. These predicted linkage positions are in
agreement to those observed for
quercetin dimers previously extracted from natural plant
sources.
-
iv
Acknowledgments
I would like to thank my supervisor Professor Emma Master for
giving me the
opportunity to work on this project and providing guidance. I
will also thank the members of my
Supervisory Committee, Professor Brad Saville and Professor
Dinesh Christendat, for their
advice throughout the project. I thank collaborators who gave
assistance on this project
including members of the group Structural Proteomics in Toronto
(SPiT) in Toronto and
members from Natural Resources Canada (NRCan) in Sault Ste.
Marie. BioZone administration
was helpful in making sure things ran smoothly. Colleagues in
the Master lab helped with a lot
of theoretical and technical aspects of the project. I thank my
family for their support over the
course of my PhD.
-
v
Table of Contents
LIST OF TABLES
...................................................................................................................
VIII
LIST OF FIGURES
....................................................................................................................
IX
LIST OF ABBREVIATIONS
....................................................................................................
XI
CHAPTER 1. OVERVIEW
..........................................................................................................
1
CHAPTER 2. LIERATURE
SURVEY........................................................................................
3 2.1. Plant phenolic compounds
...................................................................................................
3
2.1.1. Types and distribution
..............................................................................................
3
2.1.2. Biosynthesis and role in plants
.................................................................................
7
2.1.3. Health benefits of phenolics
...................................................................................
12
2.1.4. Examples of antioxidant activity
............................................................................
12
2.1.5. Phenolic antioxidant activity for food preservation
.............................................. 14 2.1.6.
Structure-functional correlations among phenolics with antioxidant
properties 15 2.1.7. In vitro measurement of antioxidant activity
......................................................... 17
2.1.8. Solubility considerations for antioxidant activity
.................................................. 18 2.2.
Derivatization of plant phenolics
.......................................................................................
20
2.2.1. Enzymatic strategies used in plant phenolic
derivatization .................................. 20 2.2.2.
Increasing hydrophilicity of phenolic compounds
................................................ 21 2.2.3.
Increasing lipophilicity of phenolic compounds
................................................... 23
2.3. Esterases/lipases
..................................................................................................................
26 2.3.2. Structural features
..................................................................................................
27
2.3.3. Catalytic mechanism
...............................................................................................
29 2.3.4. Transesterification reactions
..................................................................................
30
2.3.5. Applied Use
.............................................................................................................
33 2.4. Laccases
...............................................................................................................................
34
2.4.2. Structural features
..................................................................................................
35
2.4.3. Catalytic mechanism
...............................................................................................
37 2.4.4. Effect of Redox potential
........................................................................................
41
2.4.5. Applied use
..............................................................................................................
42 2.5. Research Hypotheses and
Objectives................................................................................
45
CHAPTER 3. CHARACTERIZATION OF SOLVENT-TOLERANT
CARBOXYLESTERASES WITH ARYLESTERASE ACTIVITY
....................................... 46 3.1. Introduction
........................................................................................................................
47
3.2. Materials and methods
.......................................................................................................
49 3.2.1. Gene cloning and protein purification
...................................................................
49
3.2.2. Hydrolytic activity of esterases AF1753 and PP3645 in
t-amyl alcohol/water
(98:2, v/v)
...........................................................................................................................
49 3.2.3. Transesterification activity of esterases AF1753 and
PP3645 in t-amyl
alcohol/cyclohexane (1:9, v/v)
..........................................................................................
50 3.2.4. Protein structure modeling and site-directed mutagenesis
of esterase AF1753 .. 51
-
vi
3.2.5. Activity of wild type and mutant AF1753 esterases on
varying chain-length pNP
esters
..................................................................................................................................
51 3.3. Results and discussion
........................................................................................................
52
3.3.1. Hydrolytic activity of esterases AF1753 and PP3645 in
t-amyl alcohol/water
(98:2, v/v)
...........................................................................................................................
52 3.3.2. Transesterification activity of esterases AF1753 and
PP3645 in t-amyl
alcohol/cyclohexane (1:9, v/v)
..........................................................................................
55 3.3.3. Activity of wild type and mutant AF1753 esterases on
varying chain-length pNP
esters
..................................................................................................................................
61 3.4. Conclusions
..........................................................................................................................
66
CHAPTER 4. BIOCHEMICAL STUDIES OF THE MULTICOPPER OXIDASE
(SMALL LACCASE) FROM STREPTOMYCES COELICOLOR USING BIOACTIVE
PHYTOCHEMICALS AND SITE-DIRECTED MUTAGENESIS
........................................ 67 4.1. Introduction
........................................................................................................................
68
4.2. Materials and methods
.......................................................................................................
69 4.2.1. Gene cloning and protein purification
...................................................................
69
4.2.2. Site-directed mutagenesis
.......................................................................................
70 4.2.3. Copper content analysis of wild-type and mutant SCO6712
laccases .................. 70 4.2.4. Substrate profile of
wild-type SCO6712 and the Ser292Ala mutant laccases ...... 70
4.2.5. Kinetics of wild-type and Ser292Ala laccases on select
substrates ....................... 72 4.2.6. Docking of
2,6-dimethoxyphenol substrate to wild type and Ser292Ala
SCO6712
laccase
...............................................................................................................................
72 4.3. Results and discussion
........................................................................................................
72
4.3.1. Effect of microaerobic cultivation on copper content and
activity of SCO6712
laccase
...............................................................................................................................
72
4.3.2. Substrate profile of wild-type SCO6712 laccase
.................................................... 73
4.3.3. Kinetics of wild-type SCO6712 laccase on select
substrates ................................. 77 4.3.4.
Site-directed mutagenesis
.......................................................................................
79
4.4. Conclusions
..........................................................................................................................
82
CHAPTER 5. CHARACTERIZATION OF PRODUCT FORMATION FROM
ENZYMATICALLY OXIDIZED PLANT PHENOLICS AND ASSAY OF
ANTIOXIDANT ACTIVITY
.....................................................................................................
83 5.1. Introduction
........................................................................................................................
84 5.2. Materials and methods
.......................................................................................................
85
5.2.1. HPLC-MS analysis of flavonol products after SCO6712
laccase treatment ........ 85 5.2.2. HPLC-MS analysis of flavonol
dimer presence over laccase reaction time......... 85
5.2.3. Total radical-trapping antioxidant parameter (TRAP) assay
of whole laccase
reaction mixture
................................................................................................................
85 5.2.4. Ferric reducing antioxidant power (FRAP) assay of whole
laccase reaction
mixture
..............................................................................................................................
86
5.2.5. HPLC-MS/MS analysis of quercetin dimer and myricetin
dimer......................... 86 5.3. Results and discussion
........................................................................................................
87
5.3.1. HPLC-MS analysis of flavonol products after SCO6712
laccase treatment ........ 87 5.3.2. Antioxidant assay of whole
laccase reaction mixture .........................................
109 5.3.3. HPLC-MS/MS analysis of quercetin dimer and myricetin
dimer....................... 120
-
vii
5.4. Conclusions
........................................................................................................................
124
CHAPTER 6. DISCUSSION
....................................................................................................
125
CHAPTER 7. FUTURE RESEARCH
.....................................................................................
137
7.1. Further characterization of esterases for
transesterification potential, continuing work
of Chapter 3
..............................................................................................................................
137 7.2. Further assessment of biochemical potential of laccase
SCO6712, continuing work of
Chapter 4
..................................................................................................................................
138 7.3. Further examination of bioactivity of flavonol dimers,
continuing work of Chapter 5
...................................................................................................................................................
139
REFERENCES
..........................................................................................................................
141
APPENDIX 1. SUPPLEMENTAL INFORMATION FOR CHAPTER 3
......................... 159
APPENDIX 2. SUPPLEMENTAL INFORMATION FOR CHAPTER 4
......................... 162
APPENDIX 3. SUPPLEMENTAL INFORMATION FOR CHAPTER 5
......................... 163
APPENDIX 4. TRAP ANTIOXIDANT ASSAY USING LINOLEIC ACID IN PLACE
OF
DCFH
........................................................................................................................................
171
APPENDIX 5. QUANTIFYING SOLUBILITY OF QUERCETIN MONOMER FOR
COMPARISON TO QUERCETIN DIMER
.........................................................................
174
-
viii
LIST OF TABLES
Table 2.1. Different classes of plant phenolic compounds
(Balasundram et al., 2006). .............. 3 Table 2.2. Major food
sources of plant phenolic compounds (Manach et al., 2004).
.................. 6
Table 2.3. Tree/shrub sources of plant phenolic compounds.
....................................................... 6 Table
2.4. Popular antioxidant activity assays.
...........................................................................
17 Table 4.1. Kinetic parameters of wild type SCO6712 and the
Ser292Ala variant enzyme. ....... 78 Table 5.1. HPLC-MS data for
quercetin, morin, myricetin, and their dimers produced after
enzymatic reaction.
......................................................................................................................
88
Table 5.2. HPLC-MS data for representative intermediate
molecular weight products present in
late time point enzymatic reactions of quercetin, morin, and
myricetin. .................................... 99 Table 5.3. Top
10 products from late time point enzymatic reactions of quercetin,
morin, and
myricetin.
...................................................................................................................................
100
Table 5.4. Top 10 products from 24 h reaction of quercetin
without and with laccase enzyme.
...................................................................................................................................................
105
Table A1.1. Sequences of primers used for construction of
esterase AF1753 point mutants. . 159
Table A2.1. Sequences of primers used for construction of
laccase SCO6712 point mutants. 162 Table A3.1. HPLC-MS data for
kaempferol and product produced after enzymatic reaction. 163
Table A3.2. Fragment ions observed for quercetin dimer #1 after
HPLC-MS/MS. ................. 163 Table A3.3. Fragment ions
observed for myricetin dimer #2 after HPLC-MS/MS. ................
164 Table A3.4. Fragment ions observed for non-enzymatically
produced quercetin dimer after
HPLC-MS/MS.
..........................................................................................................................
166 Table A3.5. Fragment ions observed for non-enzymatically
produced myricetin dimer after
HPLC-MS/MS.
..........................................................................................................................
167
-
ix
LIST OF FIGURES
Figure 2.1. Structures of some members of A) hydroxybenzoic
acids, B) hydroxycinnamic
acids, and C) stilbenes; and D) basic skeleton of all
flavonoids, and specifically flavones and
flavonols.
.......................................................................................................................................
5 Figure 2.2. Key steps in biosynthesis pathways for production of
hydroxybenzoic acids and
hydroxycinnamic acids.
...............................................................................................................
10 Figure 2.3. Key steps in biosynthesis pathway for production of
stilbenes. .............................. 11 Figure 2.4. Key steps
in the biosynthesis pathway for production of flavones and
flavonols. .. 11
Figure 2.5. Reactions involved in scavenging of peroxyl radicals
by a phenolic antioxidant. ... 13 Figure 2.6. Structures of some
classes of flavonoids.
.................................................................
16 Figure 2.7. Possible routes to adding long-chain alkyl groups to
phenolic compounds to
increase lipophilicity of the phenolic.
..........................................................................................
24
Figure 2.8. Naturally occurring dimers (n=1), trimers (n=2), and
tetramers (n=3) made of
successive epicatechin molecules linked to
catechin...................................................................
25
Figure 2.9. Reaction catalyzed by esterases and lipases.
............................................................ 27
Figure 2.10. Prototypical α/β hydrolase fold structure.
..............................................................
28
Figure 2.11. Catalytic mechanism of esterases and lipases.
....................................................... 31 Figure
2.12. Reaction catalyzed by
laccases...............................................................................
35 Figure 2.13. Structure of laccase TvL from Trametes versicolor.
A) One of the cupredoxin
domains of TvL (the protein has three such domains).
............................................................... 37
Figure 2.14. Proposed catalytic mechanisms of oxidation by laccase
enzymes. ........................ 40
Figure 2.15. Structures of closely related phenols.
.....................................................................
41 Figure 3.1. Known sites of action of feruloyl esterases and
arylesterases.................................. 48 Figure 3.2.
Hydrolysis reactions of A) vinyl acetate and B) phenyl acetate
carried out in t-amyl
alcohol/water (98:2, v/v) using esterases AF1753 and PP3645.
................................................. 54
Figure 3.3. Transesterification reaction of vinyl acetate and
4’-hydroxyacetophenone carried out
in different solvent mixtures of t-amyl alcohol/co-solvent (1:9,
v/v) using commercial lipase PS
(from Amano).
.............................................................................................................................
57
Figure 3.4. Transesterification reaction of vinyl acetate and
4’-hydroxyacetophenone carried out
in t-amyl alcohol/cyclohexane (1:9, v/v) using recombinant
esterase AF1753 and PP3645. ..... 58
Figure 3.5. Hydrolysis reaction of ester substrate in
transesterification reaction-mix. .............. 59 Figure 3.6.
Activity of wild type and mutants of AF1753 on pNP substrates.
........................... 63
Figure 3.7. Images of predicted protein structure of AF1753.
................................................... 65 Figure 4.1.
Substrate selectivity of SCO6712.
...........................................................................
76 Figure 4.2. Chemical structures of natural bioactive phenolic
substrates. ................................. 77 Figure 4.3. Ribbon
image of 2,6-dimethoxyphenol (2,6-DMP) docked in silico to
binding
pocket of SCO6712 A) wild type enzyme and B) Ser292Ala mutant.
........................................ 81
Figure 5.1. HPLC-MS chromatograms and m/z spectra for 20 min
reaction samples with laccase
enzyme for A) quercetin, B) morin, and C) myricetin.
...............................................................
91
Figure 5.2. HPLC-MS chromatogram and m/z spectrum for 20 min
reaction sample with laccase
enzyme for kaempferol.
...............................................................................................................
91 Figure 5.3. HPLC-MS chromatograms for 5 min reaction sample with
laccase enzyme for (A)
quercetin, (B) morin, and (C) myricetin.
.....................................................................................
94 Figure 5.4. Mass spectrum for myricetin dimer of m/z 635 (from
peak D3 in chromatogram of
Fig. 5.3. C); refer to Table 5.1) for 5 min reaction sample with
laccase enzyme. ..................... 94
-
x
Figure 5.5. HPLC-MS analysis of flavonol dimers from A)
quercetin (dimer #1), B) morin, C)
myricetin (dimer #2), and D) myricetin (dimer #3).
....................................................................
98 Figure 5.6. Proposed reaction scheme for production of some of
the quercetin oxidation
degradation products.
.................................................................................................................
104
Figure 5.7. Structures of the isomers quercetin and morin.
...................................................... 108 Figure
5.8. TRAP antioxidant activity of laccase reaction-mixes from A)
quercetin, B) morin,
and C) myricetin reactions.
........................................................................................................
113 Figure 5.9. TRAP antioxidant activity of laccase reaction-mixes
from A) quercetin, B) morin,
and C) myricetin reactions.
........................................................................................................
117
Figure 5.10. FRAP antioxidant activity of laccase reaction-mixes
from A) quercetin, B) morin,
and C) myricetin reactions.
........................................................................................................
119 Figure 5.11. Fragmentation patterns of flavonols in positive
ion mode tandem mass
spectrometry.
.............................................................................................................................
121
Figure 5.12. Proposed fragment structures of quercetin dimer #1
(refer Table 5.1) after tandem
mass spectrometry.
....................................................................................................................
122
Figure 5.13. Proposed fragment structures of myricetin dimer #2
(refer Table 5.1) after tandem
mass spectrometry.
....................................................................................................................
123
Figure 6.1. Two possible mechanisms of quercetin dimer
formation. ..................................... 130 Figure 6.2.
Mechanism of antioxidant antagonism proposed by Peyrat-Maillard et
al. (2003).
...................................................................................................................................................
133
Figure 6.3. Mechanism of antioxidant antagonism that involves
reaction of quercetin radicals
with laccase-generated quercetin products.
...............................................................................
135
Figure A1.1. Protein purification of AF1753 mutants.
.............................................................
161
Figure A3.2. Remaining phenolic monomer in laccase reactions, as
measured by UV-Vis
spectrophotometry and by intensity of HPLC-MS
peak............................................................
170
Figure A5.1. The maximum amount of quercetin that can be
dissolved in A) 50 mM sodium
phosphate buffer (pH 7.4) with 0.1 M NaCl and B) 1-octanol.
................................................. 177
-
xi
LIST OF ABBREVIATIONS
2,3-DHB - 2,3-dihydroxybenzoic acid
2,6-DMP - 2,6-dimethoxyphenol
AAPH - 2,2′-azobis(2-methylpropionamidine)
ABTS - 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic
acid)
Ala - alanine
AMVN - 2,2’-azobis (2,4-dimethylvaleronitrile)
Arg - arginine
DCFH - 2′,7′-dichlorofluorescin
DCFH-DA - 2′,7′-dichlorofluorescin diacetate
DHA docosahexaenoic acid
DMSO - dimethylsulfoxide
DPPH - 2,2-diphenyl-1-picrylhydrazyl
EDTA - ethylenediaminetetraacetic acid
EPA - eicosapentaenoic acid
ET - electron transfer
FRAP - ferric reducing antioxidant power
Glu - glutamate
HAT - hydrogen atom transfer
HPLC-MS - high performance liquid chromatography-mass
spectrometry
LDL - low density lipoprotein
L-DOPA - 3,4-dihydroxy-L-phenylalanine
MCO - multicopper oxidase
N-HPI - N-hydroxyphthalimide
NMR - nuclear magnetic resonance
pNP - p-nitrophenol
ROS - reactive oxygen species
Ser - serine
TPTZ - 2,4,6-tris(2-pyridyl)-s-triazine
TRAP - total radical-trapping antioxidant parameter
-
1
CHAPTER 1. OVERVIEW
Plant materials are a rich source of bioactive phenolic
compounds. Phenolics are a
common constituent of the human diet via fruits, vegetables, and
beverages (Balasundram et al.,
2006) but they can also be derived from forest sources such as
with the monolignols, stilbenes,
lignans, and certain types of flavonoids (Stevanovic et al.,
2009). Phenolic compounds have
well known antioxidant activities that can find use in
prevention of diseases affecting human
health, and in food preservation against oxidative decay
(Balasundram et al., 2006).
There is potential to increase the protective effect of these
compounds towards lipid
targets by increasing their hydrophobicity through the addition
of alkyl groups. Hydrophobicity
of the phenolics might also be increased by increasing their
molecular weight through
oligomerization. Such modifications may increase the miscibility
of the bioactive compound in
emulsified food oils, imparting a preservative effect (Frankel
et al., 1994); or in low density
lipoproteins (LDL), thereby reducing the frequency of health
problems such as atherosclerosis
(Vafiadi et al., 2008).
The overall objective of this research project was to alter the
chemistry of phenolic
compounds by enzymatic modification (for example to increase
hydrophobicity of the
phenolics), while maintaining antioxidant activity of the
phenolic. Towards this goal two
enzyme types were investigated for their potential to modify
phenolics. The first enzyme class,
esterases, were used to try to esterify alkyl chains onto
phenolic compounds and the second
enzyme class, laccase (a type of multicopper oxidase), was used
to oxidatively dimerize
phenolic compounds via radical coupling reactions. The esterase
reaction must be performed in
an organic solvent to avoid hydrolytic breakdown of the desired
ester product. By contrast,
laccase reactions can be carried out in aqueous conditions but
because of radical delocalization
it is difficult to know in advance which products will be formed
from the wide array of potential
products. The feasibility of esterase-mediated and
laccase-mediated modification of bioactive
phenolics was investigated, following the literature survey
(Chapter 2), in subsequent chapters.
First, the chosen esterases (one bacterial (Pseudomonas putida)
and one archaeal
(Archaeoglobus fulgidus)) were assessed for their activity in
predominantly organic solvent
media. The enzymes were hydrolytically active in the organic
solvent-water mixture but did not
catalyze transesterification reactions. Therefore, attention was
focused on the second enzymatic
approach, i.e. laccase oxidation followed by oxidative coupling
of phenolics to produce
-
2
increased molecular weight products as a means of modifying
phenolic chemistry. In this case,
initial work focused on evaluating determinants of activity of a
bacterial laccase from
Streptomyces coelicolor on a range of phenolic compounds from
diverse classes, including
hydroxycinnamic acids, stilbenes, flavonols, and flavones. As
laccase activity was observable
on the well-known antioxidant flavonols (among other compounds),
the products from these
reactions were analyzed. Presence of dimer products from the
flavonols quercetin, morin, and
myricetin was examined using HPLC-MS. Tandem mass spectrometry
was used to gain initial
information about the structure of the dimers of quercetin and
myricetin. The change in
antioxidant activity resulting from laccase action was assayed
using the total radical-trapping
antioxidant parameter (TRAP) assay and the ferric reducing
antioxidant power (FRAP) assay on
whole laccase reaction mixtures.
Summary of scholarly contributions
Peer reviewed publications:
Sherif, M., Waung, D., Korbeci, B., Mavisakalyan, V., Flick, R.,
Brown, G., Abou-Zaid,
M., Yakunin, A. F., & Master, E. R. (2013). Biochemical
studies of the multicopper oxidase
(small laccase) from Streptomyces coelicolor using bioactive
phytochemicals and site-directed
mutagenesis. Microbial Biotechnology, 6(5), 588-597.
Manuscripts in preparation:
Sherif, M., Wang, L., Tchigvintsev, A., Brown, G., Mavisakalyan,
V., Tillier, E. R. M,
Savchenko, A. V, Master, E. R, & Yakunin, A. F.
Solvent-tolerant and thermophilic
carboxylesterase with arylesterase activity from Archaeoglobus
fulgidus.
Sherif, M., Qazi, S., Abou-Zaid, M., & Master, E. R.
Identification of products and
antioxidant activity of reaction mixtures from treatment of four
flavonols with a multicopper
oxidase SLAC (small laccase) from Streptomyces coelicolor.
Qazi, S., Sherif, M., Master, E. R, & Abou-Zaid, M. Tandem
mass spectrometric and
NMR structural characterization of quercetin dimer produced by
multicopper oxidase treatment
of quercetin.
-
3
CHAPTER 2. LIERATURE SURVEY
2.1. Plant phenolic compounds
2.1.1. Types and distribution
Plant phenolic compounds constitute secondary metabolites and
are among the most
prevalent phytochemicals, appearing in both food and non-food
sources (reviewed in
Balasundram et al., 2006; Manach et al., 2004). A structural
categorization of plant phenolics
leads to the following classes based on the configuration of the
carbon skeleton (Table 2.1)
(carbon skeleton in brackets): simple phenolics (C6),
hydroxybenzoic acids (C6-C1),
phenylacetic acids (C6-C2), hydroxycinnamic acids (C6-C3),
quinones (diverse carbon skeleton),
xanthones (C6-C1-C6), stilbenes (C6-C2-C6), flavonoids
(C6-C3-C6), lignans ((C6-C3)2),
biflavonoids ((C6-C3-C6)2), lignins ((C6-C3)n), and tannins
(diverse carbon skeleton)
(Balasundram et al., 2006). The flavonoids are further
subdivided into the flavones, isoflavones,
flavonols, flavanones, anthocyanidins, and flavanols (Manach et
al., 2004). Of the plant
phenolics, the most structurally diverse group are the
flavonoids (Balasundram et al., 2006) and
of the flavonoids, the flavonols are the most common in foods
(Manach et al., 2004). In many
cases, representatives of these classes of plant phenolics can
be found conjugated to monomeric
and oligomeric sugars.
Table 2.1. Different classes of plant phenolic compounds
(Balasundram et al., 2006).
Phenolic class Carbon skeleton
Simple phenolics C6
Hydroxybenzoic acids C6-C1
Phenylacetic acids C6-C2
Hydroxycinnamic acids C6-C3
Quinones Diverse
Xanthones C6-C1-C6
Stilbenes C6-C2-C6
Flavonoids C6-C3-C6
Lignans (C6-C3)2
Biflavonoids (C6-C3-C6)2
Lignins (C6-C3)n
Tannins Diverse
-
4
The higher molecular weight classes mentioned above are
oligomeric/polymeric forms
of the lower molecular weight classes. Lignans are dimerized
forms, while lignin is a large
complex polymer, of the alcohols of hydroxycinnamic acids. The
stilbenes are known to exist in
oligomeric forms of the simplest structural units (C6-C2-C6) of
their class (Quideau et al., 2011).
As the name suggests, biflavonoids are dimeric versions of
flavonoids. The tannins can be
divided into the condensed tannins (carbon skeleton of
(C6-C3-C6)n) and hydrolyzable tannins.
Condensed tannins are oligomeric forms of flavanols while
hydrolyzable tannins are composed
of monomeric and polymeric hydroxybenzoic acid units esterified
onto sugars (Quideau et al.,
2011). The remainder of this section (Section 2.1.1) and the
subsequent section (Section 2.1.2)
will focus on the distribution (dietary and non-dietary),
biosynthesis, and role in plants of
hydroxybenzoic acids, hydroxycinnamic acids, stilbenes, and two
of the flavonoids (flavones
and flavonols) (Fig. 2.1). These classes of phenolics were
chosen because they contain among
the most well-studied and most effective antioxidant
compounds.
Hydroxybenzoic acids are not widely found in plant material
consumed by people. The
few major sources in the human diet include in certain red
fruits, black radish, onion, and tea
(Table 2.2) (reviewed in Manach et al., 2004). One of the more
studied of the hydroxybenzoic
acids, gallic acid, can be found in esterified form in the bark
of Quercus stenophylla (Nishimura
et al., 1984), in the flowers of Tamarix nilotica (reviewed in
Van Sumere, 1989), in maple
species (for example as a glycoside conjugate in leaves of Acer
rubrum (Abou-Zaid &
Nozzolillo, 1999) and as a methyl ester in leaves of Acer
rubrum, Acer saccharinum, and Acer
saccharum (Abou-Zaid et al., 2009)), (Table 2.3) and various
other plant species (for an
extensive list of plants containing gallic acid and other
phenolics see Harborne et al., 1990).
More recently, it was isolated from aerial plant parts of
Pelargonium reniforme (Latté et al.,
2008). Aside from being obtained from plant material after
comparatively gentle solvent
extraction, hydroxybenzoic acids, such as vanillic acid and
syringic acid, can also be obtained
upon hydrolytic treatment of lignocellulosic materials, due to
oxidation and breakdown of the
lignin polymer (reviewed in Garrote et al., 2004).
-
5
Figure 2.1. Structures of some members of A) hydroxybenzoic
acids, B) hydroxycinnamic
acids, and C) stilbenes; and D) basic skeleton of all
flavonoids, and specifically flavones and
flavonols.
OH
OH
OHO
OH
OH
OH
OHO
OH
OHO
O OH
OH
O OH
OH
OH H3CO
O OH
OH
OCH3
O OH
OH
OCH3
OH
OCH3
OHO
H3CO
OH
OH
OH
OH
OCH3
OCH3
OH
OH
O
O
O
O
O
OH
gallic acid protocatechuic acid salicylic acid
A)
B)
syringic acid
p-coumaric acid caffeic acid ferulic acid sinapic acid
C)
resveratrol pterostilbene pinosylvin
D)
flavonoids flavones flavonols
-
6
Table 2.2. Major food sources of plant phenolic compounds
(Manach et al., 2004).
Phenolic class Major food sources
Hydroxybenzoic acids red fruits
black radish
onion
tea
Hydroxycinnamic acids blueberry
kiwi
plum
cherry
apple
Stilbenes grape
Flavones parsley
celery
Flavonols onion
leek
broccoli
blueberry
Table 2.3. Tree/shrub sources of plant phenolic compounds.
Compound Tree species Reference
Hydroxybenzoic acids Quercus stenophylla
Tamarix nilotica
Pelargonium reniforme
Acer spp.
Nishimura et al. (1984)
Van Sumere (1989)
Latté et al. (2008)
Abou-Zaid et al. (2009)
Hydroxycinnamic acids Tsuga heterophylla
Catalpa ovata
Harborne (1990)
Stilbenes Veratrum formosanum
Picea abies
Pinus sibirica
Stevanovic et al. (2009)
Flavones and flavonols Eucalyptus spp.
Crataegus sp.
Pinus spp.
Stevanovic et al. (2009)
Abou-Zaid & Nozolillo (1991)
Hydroxycinnamic acids are mostly found in conjugated forms and
the four most
common compounds are conjugated forms of p-coumaric, caffeic,
ferulic, and sinapic acids
(Manach et al., 2004). As opposed to the hydroxybenzoic acids,
the hydroxycinnamic acids can
be found in a variety of food sources, and highest amounts have
been found in blueberries,
kiwis, plums, cherries, and apples (Table 2.2) (Manach et al.,
2004). Similar to the
hydroxybenzoic acids, the hydroxycinnamic acids can also be
obtained upon hydrolytic
treatment of lignocellulosic materials (Garrote et al., 2004).
In addition to making up the lignin
-
7
polymer (in the alcohol form), the hydroxycinnamic acid ferulic
acid (and its dimers) can be
found esterified onto hemicelluloses (Manach et al., 2004).
Stilbenes are not abundant in the human diet except from grapes
and their juices (Table
2.2) (Manach et al., 2004). The most widely studied stilbene is
resveratrol. Resveratrol and its
glucoside have been found in Veratrum formosanum and in the bark
of Picea abies,
respectively, and stilbenes can notably be found from the
knotwood extracts of pines (Table
2.3) (reviewed in Stevanovic et al., 2009).
Flavonoids are the most structurally diverse phenolic compound
in plants. Among the
most well-known of the flavonoids is the flavonol quercetin
because of its very strong
antioxidant activity. Flavones are chiefly found in parsley and
celery in the human diet (Table
2.2) (Manach et al., 2004). On the other hand, flavonols are
more widely prevalent and can be
found in onions, leeks, broccoli, blueberries, and other food
sources (Table 2.2) (Manach et al.,
2004). Additionally, flavonols can be found in leaves of forest
trees including birches and
eucalyptus (Stevanovic et al., 2009). Another source of
flavonols (and flavones) is in trees of the
family Crataegus, with 13 flavonols and 20 flavones previously
identified (Stevanovic et al.,
2009). Furthermore, flavonol glycosides were observed from
needles of pine trees such as Pinus
banksiana (Table 2.3) (Abou-Zaid & Nozolillo, 1991).
2.1.2. Biosynthesis and role in plants
The biosynthesis of plant phenolic compounds can be traced back
to the shikimate
pathway and the polyketide pathways with the polyketide pathway
providing precursors for
production of simple phenolics, while the shikimate pathway
provides precursors for the other
phenolic types (Harborne 1989). Starting from shikimate,
phenylalanine is produced by the
shikimate pathway. Phenylalanine is then deaminated to produce
cinnamic acid, which is then
hydroxylated to produce p-coumaric acid. As such, cinnamic acid
is the first precursor to
production of the other plant phenolics (Fig. 2.2) (Harborne,
1989; Dewick, 1995).
Hydroxybenzoic acids are thought to be produced from cinnamic
acids by removal of an acetate
unit (Fig. 2.2) (Gross, 1992). However, based on tracer
experiments with radiolabelled carbon,
it has also been proposed that gallic acid biosynthesis could
proceed via direct dehydrogenation
of shikimic acid without going through a pathway involving
cinnamic acid production (Gross,
1992). The hydroxycinnamic acids are formed via aromatic
substitution of cinnamic acid by
-
8
undergoing sequential hydroxylations and methylations (Harborne,
1989; Dewick, 1995).
Stilbene biosynthesis occurs by reaction of three malonyl-CoA
molecules with p-coumaroyl-
CoA, followed by a decarboxylation accompanied by cyclization
(Fig. 2.3) (Dewick, 1995).
Similar to the stilbenes, for the flavonoids in general, their
biosynthesis starts by reaction of
three malonyl-CoA molecules with p-coumaroyl-CoA followed by a
cyclization to produce the
flavanones, resulting in the basic carbon skeleton of all
flavonoids (Fig. 2.4) (Heller &
Forkmann, 1994; Dewick, 1995). Flavones derive from flavanones
by formation of a double
bond between C-2 and C-3 while flavonols are formed by first
hydroxylating position 3 of
flavanones followed by formation of a double bond between C-2
and C-3 (Heller & Forkmann,
1994; Dao et al., 2011).
Plant phenolic compounds have been postulated to have a diverse
array of functions,
some representative examples being: the role of benzoic acids in
photosynthesis (Van Sumere,
1989); the role of ferulic acid in regulating germination of
barley seeds (Van Sumere, 1989); the
role of stilbenes as antimicrobial compounds (Gorham, 1989); the
role of flavones and/or
flavonols in 1) protection from UV light, insects, and
microorganisms, 2) hormonal control, 3)
enzyme inhibition, and 4) attracting pollinators (Markham,
1989).
-
9
O O-
OH
OH
O-
O
OH
O OH
OH OH
OH
OHO
OH
PO4
CH2 COO-
OH
OHO4P
O
OH
OH
OH
O
OO
-
OH
OHO
O O-
OH
OH
O-
O
O4P
OH
O
O-
O
O4P
CH2
O
O-
OH
O
O-
O
CH2
O
O-
O
OO
-
OH
O-
ONH2
OO
-
OH
O-
ONH2
O O-
OH
OHO4P
OH
O
O
OH
phenylalanine
shikimate
cinnamic acid gallic acid p-coumaric acid
+
PEP
E4P
DAHPS
DAHP
DHQS
3-dehydroquinate
DHQD
3-dehydroshikimate
SDH SK
shikimate 3-phosphate
EPSPS
EPSP
chorismate
CS CM
prephenate
PAT
arogenate
ADT
PAL C4H
-
10
Figure 2.2. Key steps in biosynthesis pathways for production of
hydroxybenzoic acids and
hydroxycinnamic acids. The shikimate pathway produces chorismate
which goes on to produce
phenylalanine. Phenylalanine is metabolized to yield
hydroxycinnamic acids and
hydroxybenzoic acids. Hydroxybenzoic acids can also be produced
from shikimate pathway
intermediates without going through phenylalanine production.
Abbreviations for intermediates:
PEP, phosphoenol pyruvate; E4P, erythrose-4-phosphate; DAHP,
3-deoxy-D-arabino-
heptulosonate-7-phosphate; EPSP,
5-enolpyruvylshikimate-3-phosphate. Abbreviations for
enzymes above arrows: DAHPS,
3-deoxy-D-arabino-heptulosonate-7-phosphate synthase;
DHQS, 3-dehydroquinate synthase; DHQD, 3-dehydroquinate
dehydratase; SDH, shikimate
dehydrogenase; SK, shikimate kinase; EPSPS,
5-enolpyruvylshikimate-3-phosphate synthase;
CS, chorismate synthase; CM, chorismate mutase; PAT, prephenate
aminotransferase; ADT,
arogenate dehydratase; PAL, phenylalanine ammonia-lyase; C4H,
cinnamic acid 4-hydroxylase.
-
11
Figure 2.3. Key steps in biosynthesis pathway for production of
stilbenes. p-coumaroyl-CoA
comes from the phenylpropanoid biosynthetic pathway. STS
indicates stilbene synthase enzyme.
Figure 2.4. Key steps in the biosynthesis pathway for production
of flavones and flavonols. p-
coumaroyl-CoA comes from the phenylpropanoid biosynthetic
pathway. Enzyme abbreviations:
CHS, chalcone synthase; CHI, chalcone isomerase; FNS, flavone
synthase; F3H, flavanone 3-
hydroxylase; FLS, flavonol synthase.
O SCoA
OH
OH
OSCoA
O
O
OH
OO
AoCSOC
OH
OHOH
malonyl-CoA
stilbene (e.g.
resveratrol
CO2
+ 3x
p-coumaroyl-
CoA
STS STS
O SCoA
OH
OH
OSCoA
O
OH
O
OH
OHOH
OH
O
O
OHOH
OH
O
O
OHOH
OH
O
O
OHOH
OH
OH
O
O
OHOH
OH
p-coumaroyl-
CoA
malonyl-CoA
flavanone (e.g.
naringenin)
flavonol (e.g.
kaempferol) flavone (e.g.
apigenin)
+ 3x CHS CHI
FNS
F3H
FLS
dihydrokaempferol
-
12
2.1.3. Health benefits of phenolics
Plant-derived phenolic compounds are implicated in a wide array
of health benefits. This
includes benefits against cardiovascular disease,
neurodegenerative disease, cancer, and diabetes
(reviewed in Scalbert et al., 2005). However, in vitro findings
can not always be translated into
similar in vivo effects and tests from different labs do not
always give the same results (Scalbert
et al., 2005). Atherosclerosis has been observed to be inhibited
by consumption of food
phenolics and, based on animal studies, it is thought that the
phenolics mediate their effects by
reducing oxidation and uptake of low density lipoprotein (LDL)
by macrophages (Kaplan et al.,
2001; Miura et al., 2001). However, human studies have shown
mixed results, with some
studies showing that consumption of tea protects against ex vivo
oxidation of LDL (Ishikawa et
al., 1997) while other studies showed no such benefit of tea
consumption (Van het Hof et al.,
1997). Animal models have also shown anticarcinogenic effects of
food phenolics but the doses
used in such experiments are usually much larger than typical
consumption levels in the human
diet, making it difficult to correlate epidemiological results
to these animal models (Scalbert et
al., 2005). The importance of dosage is further highlighted by
the fact that low doses (less than
10 µM) of epigallocatechin gallate was found to be
neuroprotective in cell culture models using
the neurotoxin 6-hydroxydopamine, while higher doses of
epigallocatechin gallate had cytotoxic
effects (Levites et al., 2002). Use of plants by indigenous
peoples for treatment of diabetes has
been documented and animal model studies have also shown
antidiabetic effects of plant
extracts containing phenolics (Scalbert et al., 2005; Giordani
et al., 2015). It is thought that one
of the mechanisms by which the plant compounds reduce diabetes
is through inhibition of α-
glucosidase enzymes, which normally break down carbohydrates so
that the sugars can be
absorbed in the gut (Giordani et al., 2015).
2.1.4. Examples of antioxidant activity
Plant phenolics are well-known for their antioxidant activity,
which may, in some cases,
partially mediate their other health effects (reviewed in
Scalbert et al., 2005). The antioxidant
activity of plant phenolics is due to reaction with free
radicals, but may also involve inhibition
of enzymes and chelation of metal ions (Huang et al., 2005). In
the case of reacting with free
-
13
radicals, the phenolic antioxidant sacrificially becomes
oxidized to a relatively stable radical.
Free radical scavenging may occur by the phenolic transferring a
hydrogen atom (hydrogen
atom transfer (HAT)) or by the phenolic transferring an electron
(electron transfer (ET))
followed by reversibly transferring a proton (Fig. 2.5) (Wright
et al., 2001). The bond
dissociation enthalpy of the hydroxyl groups on phenolics will
influence hydrogen atom transfer
whereas the ionization potential is important for determining
electron transfer (Wright et al.,
2001). Also, in buffer solutions, the phenolic compound can
exist in different protonation states
depending on the pH. Under more basic conditions the phenolic
hydroxyls can be deprotonated,
and in this case the phenolic will scavenge radicals by electron
transfer that is preceded by
proton loss (Fig. 2.5) (Wang & Zhang, 2005).
Hydrogen atom transfer
AOH + ROO. AO. + ROOH (1)
Electron transfer followed by reversible proton transfer
AOH + ROO. AOH+ + ROO- (2)
AOH+ + H2O ⇌ AO. + H3O+ (3)
H3O+ + ROO- ⇌ H2O + ROOH (4)
Deprotonated phenolic transferring an electron
AOH AO- (5)
AO- + ROO. AO. + ROO- (6)
Figure 2.5. Reactions involved in scavenging of peroxyl radicals
by a phenolic antioxidant.
AOH and ROO. represent a phenolic antioxidant and a peroxyl
radical, respectively.
The phenolic compound may act as an antioxidant by inhibiting
enzymes that produce
reactive oxygen species. For example, xanthine oxidase is an
enzyme that can produce the
reactive oxygen species superoxide from hypoxanthine. However,
the flavonols quercetin,
kaempferol, and myricetin can inhibit xanthine oxidase activity
as seen by inhibition of the
ability of the enzyme to convert xanthine to uric acid (Selloum
et al., 2001). Moreover,
phenolics may chelate transition metal ions to prevent the
transition metal from producing
-
14
reactive oxygen species. For example, ferrous iron (Fe2+) is
able to generate hydroxyl radicals
from hydrogen peroxide and the hydroxyl radical can then damage
DNA. However,
epigallocatechin-3-gallate (and also other phenolics) can
prevent DNA damage induced by Fe2+
(Perron et al., 2008). Since the impact of
epigallocatechin-3-gallate is reduced upon addition of
ethylenediaminetetraacetic acid (EDTA), the researchers
attributed the inhibition of DNA
damage to the formation of phenolic/Fe2+ chelates.
2.1.5. Phenolic antioxidant activity for food preservation
Oils and fats in foods are susceptible to oxidative decay, which
can lead to rancidity or
“off-flavours” arising primarily from aldehyde products (Kaur
& Perkins, 1991). Aside from
affecting flavour, as noted above (Section 2.1.3) lipid
oxidation products might also cause
cardiovascular health problems (see also Addis & Warner
(1991) for more on dietary lipid
oxidation products). With more and more people living in cities,
more food items undergo a
long transit from raw material to the end consumer. Furthermore,
there is increasing trend
towards processed foods containing multiple ingredients, some of
which are sensitive to
oxidative decay. In this regard, the omega-3 polyunsaturated
fatty acids eicosapentaenoic acid
(EPA) and docosahexaenoic acid (DHA) have been proposed to have
health benefits (reviewed
in Mori, 2014) and in 2004 the US Food and Drug Administration
allowed for qualified health
claims of reduced risk of coronary heart disease for foods
containing EPA and DHA (US Food
and Drug Administration, 2004). Polyunsaturation makes these
fatty acids particularly
susceptible to oxidation reactions (Kaur & Perkins, 1991).
Antioxidants are a logical choice as
additives to prevent spoilage of foods containing oxidizable
lipids. Among the antioxidants used
most widely in industry are the phenolic compounds butylated
hydroxyanisole (BHA), butylated
hydroxytoluene, (BHT), butylated hydroxyquinone (TBHQ), and
esters of gallic acid (Loliger,
1991). However, some of these (specifically BHA and BHT) have
shown toxic effects in some
animal studies, although in these cases dosages were greater
than would be expected to be
ingested by humans (European Food Safety Authority (EFSA),
2012). While the synthetic
antioixdants BHT and BHA are still allowed by regulatory
agencies, there is a continual search
for new phenolic (and non-phenolic) antioxidants, particularly
from natural sources such as
plant food powders (for example carrot, tomato, broccoli, and
beetroot) (Neacsu et al., 2015),
-
15
mint leaf and citrus peel extracts (Viji et al., 2015), honey
(Tahir et al., 2015), and licorice
extract (Zhang et al., 2014), to name a few recent works.
2.1.6. Structure-functional correlations among phenolics with
antioxidant properties
General relationships between molecular structure of phenolic
compounds and
antioxidant activity have been previously identified. The number
and location of hydroxyl
groups, along with the presence of double bonds that increase
the degree of conjugation, all
seem to have a role in determining antioxidant activity
(reviewed in Balasundram et al., 2006).
Hydroxycinnamic acids generally show higher antioxidant activity
than corresponding
hydroxybenzoic acids, which may be due to the double bond in the
propanoid group of the
hydroxycinnamic acid, providing increased delocalization of the
unpaired electron of the
radicalized phenolic (Natella et al., 1999). Increased electron
delocalization will stabilize the
phenolic radical thereby making it easier to form, meaning the
original phenolic is more
reactive. A study was carried out with the flavonoids to
investigate structural features important
for antioxidant activity towards the radical of
2,2'-azinobis(3-ethylbenzothiazoline-6-
sulphonate) (ABTS˙+ radical) in aqueous solution (Rice-Evans et
al., 1995). Compounds with an
ortho dihydroxy substitution in the B ring (quercetin and
cyanidin) had higher antioxidant
activity than comparable compounds (kaempferol and pelargonidin,
respectively) with only a
single hydroxyl group in the B ring (Fig. 2.6). Replacing
hydroxyls with O-glycosides (as in 3-
OH in quercetin and 7-OH in naringenin to 3-O-glycoside in rutin
and 7-O-glycoside in
naringin, respectively) resulted in decreased antioxidant
activity. The presence of the double
bond between carbon two and three in the C ring (present in
quercetin but lacking in the
otherwise identical taxifolin (Fig. 2.6)) resulted in higher
antioxidant activity. Another study
identified similar relationships between structure and
antioxidant activity of flavonoids (Van
Acker et al., 1996). In this case, antioxidant activity was
measured for lipid peroxidation and
electrochemical oxidation potentials were also measured. The
researchers observed an overall
qualitative correlation between antioxidant activity and
oxidation potentials. They identified that
compounds with an ortho dihydroxy substitution in the B ring had
the highest activity and that
for such compounds the rest of the molecule was relatively less
important in affecting activity.
Among such compounds, quercetin and myricetin showed highest
activity. This suggested that,
in combination with the ortho dihydroxy, the double bond between
C2 and C3 along with the 3-
-
16
OH results in a very strong antioxidant, possibly due to the
extensive conjugation that such a
compound has (Van Acker et al., 1996). Replacing the 3-OH of
quercetin with 3-O-rutinose in
rutin resulted in slightly lower activity for rutin. The
importance of the 3-OH became more
prominent when comparing the reduction in activity of a rutin
derivative that has OEtOH at 7,
3’, and 4’ positions (and hence lacks the ortho dihydroxy in
ring B) to a quercetin derivative that
has OEtOH at 7, 3’, and 4’ positions (Fig. 2.6), reinforcing the
idea that in compounds with an
ortho dihydroxy in ring B the rest of the molecular structure is
relatively less important for
determining antioxidant activity (Van Acker et al., 1996).
Figure 2.6. Structures of some classes of flavonoids. The ring
nomenclature and carbon
numbering system are shown on the flavonol skeleton.
R2
O
O
OHOH
OH
R1
R3
R2
O+
OHOH
OH
R1
R3
R2
O
O
OHOH
OH
R3
R1
R2
O
O
OHOH
R1R3
A
B
R1=R3=H, R2=OH; kaempferol
R1=R2=OH, R3=H; quercetin
R1=R2=R3=OH; myricetin
R1=R3=H, R2=OH; pelargonidin
R1=R2=OH, R3=H; cyanidin
flavonol anthocyanidin dihydroflavonol
R1=R2=OH, R3=H; taxifolin
2 3
4
5
6 7
8
2’
3’
flavanone
4’
R1=R3=H, R2=OH; naringenin
5’
6’
C
-
17
2.1.7. In vitro measurement of antioxidant activity
There are a variety of assays that have been developed to give
some measure of
antioxidant activity. These assays measure hydrogen atom
transfer (HAT), electron transfer
(ET), or a combination of the two. Among the more popular
hydrogen atom transfer assays are:
oxygen radical absorbance capacity (ORAC), total
radical-trapping antioxidant parameter
(TRAP), total oxidant scavenging capacity (TOSC), and crocin
bleaching assay (Table 2.4)
(Prior et al., 2005). Among the more popular electron transfer
assays is the ferric reducing
antioxidant power (FRAP) assay (Prior et al., 2005). Other
popular assays that assess both
hydrogen atom donation and electron transfer include trolox
equivalent antioxidant capacity
(TEAC) and 2,2-diphenyl-1-picrylhydrazyl (DPPH) (Table 2.4)
(Prior et al., 2005).
Table 2.4. Popular antioxidant activity assays.
Assay name Mechanisma Reagents Biological
relevance
Quantification of activity
ORAC
TRAP
TOSC
crocin
bleaching
FRAP
TEAC
DPPH
HAT
ET
HAT and ET
oxidizer (can
vary), probe
(can vary)
FeIII-TPTZ
ABTS
DPPH
Yes
No
lag time until probe
oxidation and/or decrease in
slope depicting rate of probe
oxidation
Reduction of oxidant at
chosen time
a HAT and ET mean hydrogen atom transfer and electron transfer,
respectively.
The concept behind the ORAC, TRAP, TOSC, and crocin bleaching
assays is essentially
the same, with the main differences being the reagents that are
used and the methods of
detection of reaction progress. In all cases, there is a radical
generator as a source of in situ
radicals, a target that acts as a probe for oxidation, and the
antioxidant that inhibits oxidation of
the probe by itself sacrificially becoming oxidized (Huang et
al., 2005). Different versions of
each method have been developed that use different radical
generators and probes. Antioxidant
activity can be quantified as the lag time before oxidation of
the probe is initiated and/or the
decrease in the rate of oxidation of the probe. In one version
of the ORAC and TRAP assays,
-
18
both make use of 2,2′-azobis(2-methylpropionamidine) (AAPH) as a
temperature-sensitive
peroxyl radical generator and 2’,7’-dichlorofluorescein (DCFH)
as the probe (Prior et al., 2005;
Huang et al., 2005). As an alternative to AAPH and DCFH,
2,2’-azobis (2,4-
dimethylvaleronitrile) (AMVN) can be used as a peroxyl radical
generator in conjunction with a
lipid soluble probe such as
4,4-difluoro-3,5-bis(4-phenyl-1,3-butadienyl)-4-bora-3a,4a-diaza-s-
indacene (BODIPY 665/676) (Huang et al., 2005). The TOSC assay
has been used with a
peroxyl radical generator but also with hydroxyl radicals
generated from iron/ascorbate and
peroxynitrite radicals generated from 3-morpholinosydnonimine
N-ethylcarbamide (SIN-
1)/diethylenetriaminepentaacetic acid (DTPA) allowing
characterizations of different radicals
(Regoli & Winston, 1999). The crocin bleaching assay uses
crocin as the oxidizable probe.
Crocin was introduced as a substitute to β-carotene because the
former only undergoes radical
oxidation while the latter can undergo light and heat induced
oxidation (Prior et al., 2005).
The FRAP, TEAC, and DPPH assays are similar to each other in
that there is no probe as
in the case of the purely HAT-based assays. Rather, the reaction
between an oxidant and the
antioxidant is measured directly, by measuring the change in
oxidant concentration at a chosen
time. The oxidants for the FRAP, TEAC, and DPPH are a complex of
FeIII-2,4,6-tris(2-pyridyl)-
s-triazine (TPTZ),
2,2’-azinobis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS)
radical cation,
and DPPH radical, respectively. These assays are considered less
biologically relevant than the
purely HAT-based assays because they use oxidants that are not
oxygen-based.
2.1.8. Solubility considerations for antioxidant activity
In addition to presence of functional groups like hydroxyls, the
solubility of the phenolic
also plays a role in antioxidant efficacy in different solution
conditions (reviewed in Shahidi &
Zhong, 2011). It has been previously observed that, in general,
non-polar antioxidants are better
than their polar counterparts in protecting a lipid compound
that is emulsified in an aqueous
solution, whereas the polar antioxidant is found to be more
effective in purely lipid systems
(Porter et al., 1989; Frankel et al., 1994; Cuvelier et al.,
2000). This phenomenon is proposed to
occur due to preferential partitioning of the phenolic
antioxidant at the interface where oxidation
of the lipid is initiated (Frankel et al., 1994). In an emulsion
of lipid in water, the more non-
polar antioxidants would partition to the water-lipid interface
and scavenge free radicals before
these radicals propagate into the interior of the lipid micelle.
As support for this mechanism, the
-
19
surfactant effectiveness of a series of acylated hydroxytyrosols
(with varying acyl chain-length)
correlated with their antioxidant activity in an oil-in-water
emulsion, suggesting that the more
effective antioxidants are those that act as surfactants and
partition to the interface of the oil-in-
water emulsion (Lucas et al., 2010). Notably, the antioxidant
activity of the octanoate ester of
hydroxytyrosol was greater than both lower chain-length and
higher chain-length esters, with a
parallel bell-shaped pattern observed for surfactant
effectiveness (i.e. the octanoate ester had
greater surfactant effectiveness than both lower chain-length
and higher chain-length
hydroxytyrosol esters) (Lucas et al., 2010). Therefore, this
previous work also highlighted the
fact that antioxidant polarity and antioxidant activity do not
follow an entirely linear relationship
(Lucas et al., 2010; Shahidi & Zhong, 2011). In the case of
pure lipids, two different interfaces
have been posited as being relevant. It was originally suggested
that oxidation was initiated at
the air-lipid interface and that polar antioxidants were more
effective by preferentially
partitioning to this interface while non-polar antioxidants were
less effective because they
remained soluble in the bulk lipid (Frankel et al., 1994). It
was later suggested that micro-
aqueous environments exist as reverse micelles and that these
are the sites where oxidation is
initiated (Chaiyasit et al., 2007). The polar antioxidants would
preferentially partition to these
reverse micelles, thereby being more effective than the
non-polar antioxidants that are dissolved
in the bulk lipid.
The condition of emulsification of lipids is common in foods
(such as milk and
dressings), and biological systems (such as lipoproteins, whose
oxidation is thought to be
associated with cardiovascular disease (Scalbert et al., 2005)).
Therefore, one chemical property
of naturally occurring phenolic antioxidants that may be
advantageously modified is their
lipophilicity. As such, the lipophilized antioxidants can find
application as food preservants; and
may also be used as nutraceuticals that protect against
oxidative stress of lipid components in
the human body. At the same time, changes in molecular structure
of the phenolic that improve
lipophilicity may also cause changes to the presence of
functional groups mentioned above
(hydroxyls and conjugated double bonds) that have been found to
be important for antioxidant
activity. Therefore, it may be required to strike a balance
between lipophilicity and antioxidant
activity. Two enzyme types that have potential to modify
chemistry of natural product phenolics
(for example to increase lipophilicity) are esterases/lipase
enzymes and laccase enzymes. As
such, Section 2.3 and Section 2.4 will review esterase/lipase
enzymes and laccase enzymes,
-
20
respectively; but first Section 2.2 will more broadly review
approaches to plant phenolic
derivatization.
2.2. Derivatization of plant phenolics
2.2.1. Enzymatic strategies used in plant phenolic
derivatization
A variety of enzymes have been used to modify chemistry of plant
phenolics. Enzymes
can offer the advantage of stereo- and regioselectivity, which
is important in fine-tuning only
selected regions of the molecule. The resultant changes can
affect the activity and solubility
properties of the starting compound so that it may be used in
novel applications. Lipases can
carry out transesterification reactions using activated esters
(such as vinyl esters) to produce
acylated derivatives of phenolics. For example, immobilized
lipases were used to acylate the
phenolic hydroxyls of resveratrol (Torres et al., 2010) with
vinyl acetate as the acyl donor. In
this case, the authors aimed to selectively acetylate the
hydroxyl at position 3 to protect it from
becoming sulfated or glucuronated in the liver, thereby
potentially increasing resveratrol’s
bioavailability. The authors found that a lipase from
Alcaligenes sp. was able to almost
exclusively acetylate the 3-OH while leaving the other two
hydroxyls of resveratrol intact,
whereas other tested lipases showed less selectivity. Similar to
the lipases, a select group of
esterases have also been used to acylate phenolic hydroxyls.
Topakas et al. (2003) used a
feruloyl esterase to esterify hydroxycinnamic acids in the hopes
of improving the lipid solubility
of the compound. Another group of acylating enzymes are the
aptly named acyltransferases.
These enzymes typically require an acyl-Coenzyme A (acyl-CoA)
substrate as the acyl donor.
For example, a malonyltransferase was used to catalyze the
addition of malonyl of malonyl-CoA
to a free hydroxyl of glucose that is covalently linked to an
anthocyanin (Suzuki et al., 2002).
The anthocyanins are a type of flavonoid and are notable for the
colours they impart to flowers.
Upon malonylation, the anthocyanin pigment was found to be more
stable (Suzuki et al., 2002).
Continuing with the theme of transferases, prenyltransferases
are another enzyme group
that have been used for modification of bioactive phenolics.
Prenylated phenylpropanoids may
have anti-inflammatory and anticancer activity (Paulino et al.,
2008; Messerli et al., 2009). A
prenyltransferase from S. spheroides was able to catalyze
prenylation of hydroxycinnamic acids,
resveratrol, and some flavonoids (Ozaki et al., 2009). Another
group of transferase enzymes that
-
21
are used in phenolic modification are the methyltransferases.
Methylation of flavonoids may be
important for their antifungal and antibacterial properties
(Aida et al., 1996; Zhang et al., 2008).
Accordingly, a methyltransferse from tomato was recombinantly
expressed in E. coli and
displayed activity on flavonoids including one flavanone, a
dihydroflavonol, flavones, and
flavonols (Cho et al., 2012). The enzyme had regiospecificity
for the 3’ and 5’ positions of the
flavonoids.
In many cases, natural plant phenolics are found in glycosylated
forms. Glycosylation
can impart increased water solubility to the phenolic, among
other effects. Some enzymes that
hydrolyze glycosidic bonds also display transglycosylation
activity. Such was the case for a
maltogenic amylase from Bacillus stearothermophilus. This enzyme
was able to transfer mono-,
di-, and tri-glucose units from maltotriose to the flavonoid
naringin (Lee et al., 1999). The major
product, in which maltose had been attached to the already
existing glucose of naringin,
displayed not only improved water solubility but also reduced
bitterness (Lee et al., 1999).
Finally, oxidase enzymes including laccases and peroxidases can
be used to produce
homo- and heterocoupled phenolic products. Makris and Rossiter
(2002) used horseradish
peroxidase to oxidize quercetin producing a compound that was
later identified as a quercetin
dimer (Gulsen et al., 2007). Ultimately, the authors found the
dimer to have reduced activity
(compared to the monomer) in scavenging DPPH radical, hydroxyl
free radical, and hydrogen
peroxide (Gulsen et al., 2007). Nicotra et al. (2004) used a
laccase from M. thermophyla to
synthesize dimers of resveratrol that might have bioactivities
similar to naturally occurring
oligostilbenes. Nugroho Prasetyo et al. (2011) used a laccase
from T. hirsuta to couple different
simple phenolics such as catechol onto naringenin. The authors
aimed to add hydroxyl groups
that would be conjugated to the isolated C-ring hydroxyl of
naringenin, thereby potentially
increasing antioxidant activity (Nugroho Prasetyo et al., 2011).
One commom outcome
(intended or unintended) of phenolic derivatization is
modification of the solubility of the
phenolic. As this affects the extent to which the phenolic can
access different sites in biological
systems, it is a significant motivation of many derivatization
processes, and so will be reviewed
in the next two sections (Sections 2.2.2 and 2.2.3).
2.2.2. Increasing hydrophilicity of phenolic compounds
-
22
Phenolics from the flavonol class have been shown to have
protective effects against
ischaemia-reperfusion injury (Williams et al., 2011). For such
an application, it is desirable to
increase the water solubility of the antioxidant to administer
more of it intravenously into the
blood with fewer injections. One way to increase water
solubility is to make a water soluble
prodrug derivative of the phenolic that is converted into the
parent compound in vivo. Water
soluble prodrugs of flavonols were shown to have protection
against sheep cardiac reperfusion
injury comparable to equimolar quantities of the parent compound
(Williams et al., 2011). In
this case, the researchers added phosphate or adipic acid groups
to the parent flavonol to make
the prodrug. The prodrug can be converted back to the parent
compound by action of
phosphatase or esterase enzymes that are naturally present in
tissues and blood. However, this
particular study did not yet examine dosage increases that could
potentially be realized with the
water soluble prodrugs and the parent phenolics were
administered in solutions containing
DMSO as organic co-solvent. While DMSO is often used in
experimental settings, it is
undesirable for clinical applications because it may cause side
effects including hemolysis
(Muther et al., 1980; Santos et al., 2003) and its ability to
dissolve some plastics used for
intravenous administration (Marshall et al., 1984).
Another approach to increasing the amount of phenolic in aqueous
media is to
encapsulate the compound in a carrier that has both hydrophobic
core and hydrophilic exterior
(reviewed for quercetin in Cai et al., 2013). In this case, the
encapsulated phenolic might also be
protected from undesirable in vivo metabolic modifications on
route to the site of action. One
group of encapsulating compounds that can be used is the
cyclodextrins. A complex of quercetin
and sulfobutyl ether-7β-cyclodextrin allowed for improved
solubilization of quercetin in
aqueous neutral buffer solution (Kale et al., 2006). The orally
administered complex also
showed improved tumour growth suppression in mice compared to
equivalent doses of the
uncomplexed quercetin (Kale et al., 2006).
A third approach to increase hydrophilicity is to form
nanocrystals of the desired
compound. Nanocrystals are highly fine particles of the compound
and have a mean particle size
less than 1 µm (typically between 200 nm and 500 nm) (Keck and
Muller, 2006). As expected,
the increased surface area of the fine particles leads to
increased dissolution rate. Researchers
produced a nanocrystal formulation of the phenolic compound
curcumin and this nanocrystal
had a mean diameter of 250 nm compared to 22 µm for crystalline
curcumin (Onoue et al.,
-
23
2010). Compared to crystalline curcumin, the curcumin
nanocrystal showed improved
dissolution rate in water and showed improved bioavailability in
male Sprague-Dawley rats after
oral administration.
2.2.3. Increasing lipophilicity of phenolic compounds
For some applications, it can be advantageous to increase
lipophilicity of antioxidant
phenolic compounds. If a lipid is being targeted for antioxidant
protection, then a lipophilic
antioxidant may be more effective than a hydrophilic one. It was
previously observed that, in
general, non-polar antioxidants are more effective than their
polar counterparts at protecting
lipids dispersed in water (reviewed in Shahidi & Zhong,
2011). In this context, lipophilic
antioxidants can be exploited for preservation of emulsified
lipids in foods, such as in
mayonnaise, dressings, and milk.
One of the methods used for increasing lipophilicity of
phenolics is to add a long-chain
alkyl group by way of esterification reactions of the phenolic
with a long-chain acyl compound
or long-chain alcohol compound (Fig. 2.7). For example, alcohols
of varying chain length were
esterified (using sulfuric acid as catalyst) onto caffeic acid
to produce lipophilic derivatives
(Aleman et al., 2015). The lipophilic caffeic acid derivatives
showed better protection towards
fish oil emulsified in mayonnaise or milk. It should be noted
that it was not the longest alkyl
chain ester derivative of caffeic acid that conferred best
protection, but rather intermediate chain
length (for mayonnaise) and short chain length (for milk) ester
derivatives showed best
protection (Aleman et al., 2015).
-
24
Figure 2.7. Possible routes to adding long-chain alkyl groups to
phenolic compounds to
increase lipophilicity of the phenolic. In A) the long-chain
alkyl group is represented by R1 and
is added to the phenolic as part of an acyl group. In the case
of B) the phenolic contains a
carboxylic group on one of its ends. The long-chain alkyl group
is represented by R4 and is
added to the phenolic as part of an alkoxy group.
A potential way of increasing lipophilicity is by forming dimer
and higher level
oligomers of the phenolic compound. Researchers had previously
isolated naturally occurring
dimers, trimers, and tetramers formed of successive epicatechin
molecules added to catechin
(Fig. 2.8) (Plumb et al., 1998). The (n-octanol)-water partition
coefficient of these compounds
demonstrated increasing preference for the hydrophobic n-octanol
phase with increasing degree
of oligomerization after the dimer (Plumb et al., 1998) (the
dimer did not have a significant
difference in partition coefficient compared to the monomer).
However, in this study, the
researchers saw decreased antioxidant activity towards
iron/ascorbate induced oxidation of
phospholipid liposomes with increasing lipophilicity (due to
increased oligomerization) of the
antioxidant compound. In a related study, catechin monomer and
oligomers up to hexamer were
compared for their antioxidant activity toward
L-α-phosphatidylcholine liposome, using
different inducers of oxidation (Lotito et al., 2000). When
iron/ascorbate (which would be
present in the aqueous phase) was used to induce oxidation,
antioxidant activity decreased with
increasing degree of oligomerization of catechin up to the
pentamer, showing similar trends as
seen by Plumb et al. (1998). However, when 2,2’-azobis
(2,4-dimethylvaleronitrile) (AMVN,
R1 OR2
O
R1 O
O
R3OH R3
OH
OHO
OH
OR4O
HOR2 + +
+ HOR4 + H2O
A)
B)
⇌
⇌
-
25
which would be present in the lipid phase) was used to induce
oxidation, antioxidant activity
increased with increasing degree of oligomerization of catechin
up to the pentamer (Lotito et al.,
2000). These previous examples demonstrate that the more
lipophilic phenolic derivative is not
always the better antioxidant for emulsified lipids. Changing
the inducer of oxidation can
change the pattern of antioxidant activity, and increasing the
molecular weight of the antioxidant
can be beneficial up to a certain point but further increases in
molecular weight may become
detrimental.
Figure 2.8. Naturally occurring dimers (n=1), trimers (n=2), and
tetramers (n=3) made of
successive epicatechin molecules linked to catechin.
The addition of alkyl groups via esterification has been carried
out using acid catalysts.
For example, esters of caffeic acid were synthesized using
sulfuric acid as catalyst (Aleman et
al., 2015), while rosmarinic acid esters were produced using the
strongly acidic sulfonic resin
Amberlite IR-120H (Panya et al., 2012). The use of acid catalyst
represents harsh reaction
conditions. On the other hand, the use of enzymes to synthesize
novel lipophilic derivatives of
phenolics is an alternative approach that avoids the use of
strong acids. In addition, naturally
occurring oligomeric phenolics are present for a limited subset
of the phenolic classes including
flavones, flavanols, and hydroxycinnamic acids. Enzymatic
catalysis can allow the synthesis of
additional novel oligomeric compounds not readily available from
environmental sources.
OH
OOH
OH
OH
OH
H
H
OH
OOH
OH
H
OH
OH
H
n
epicatechin
catechin
-
26
Esterases and lipases are enzymes that can catalyze
esterification reactions while laccases can
catalyze oxidation that leads to oligomerization.
2.3. Esterases/lipases
Broadly speaking, esterases and lipases are enzymes that
catalyze the hydrolysis of ester
bonds (Fig. 2.9). Esterases and lipases belong to the general
class of enzymes called ester
hydrolases (EC 3.1) by the Nomenclature Committee of the
International Union of Biochemistry
and Molecular Biology (NC-IUBMB) (2010b). The ester hydrolase
class is further divided to
result in more classes, among of which are carboxylic-ester
hydrolases (i.e. esterases and
lipases) (EC 3.1.1), thioester hydrolases (EC 3.1.2), phosphoric
monoester hydrolases (EC
3.1.3), and others. Carboxylic-ester hydrolases (EC 3.1.1) are
then divided into
carboxylesterases (EC 3.1.1.1), arylesterases (EC 3.1.1.2),
triacylglycerol lipases (EC3.1.1.3),
and others. A less formal but often used distinction for the
carboxylic-ester hydrolases is to refer
to them simply as either “esterases” or “lipases”. In this case,
esterases (for example EC 3.1.1.1)
are differentiated from lipases (for example EC 3.1.1.3) in the
tendency of esterases to prefer
short-chain substrates while lipases show activity on both
short-chain and long-chain substrates.
From this point on, the terms esterase and lipase will be used
rather than the NC-IUBMB
terminology. In addition to classification based on reaction
substrates, these enzymes have also
been classified into families based on amino acid sequence
features. For example, the
carbohydrate active enzyme (CAZy) classification system, which
focuses on enzymes acting on
carbohydrates, comprises a carbohydrate esterase class that is
divided into 16 families (Lombard
et al., 2013). Likewise, in the ESTerases and
alpha/beta-Hydrolase Enzymes and Relatives
(ESTHER) classification system, esterases and lipases (along
with other hydrolases) have been
classified within 148 families based mainly on sequence features
and any available biological
data (Lenfant et al., 2013).
-
27
Figure 2.9. Reaction catalyzed by esterases and lipases. In the
case of hydrolysis, R3 is a
hydrogen atom so that HOR3 is water and the final products are a
carboxylic acid and an
alcohol. In the case of transesterification, R3 is an alkyl
group so that HOR3 is an alcohol and the
final products are a new ester and an alcohol.
2.3.2. Structural features
Esterases and lipases are characterized by an α/β-hydrolase fold
structure, which is
defined as a central β-sheet (as opposed to α/β barrel) of eight
β-strands connected and
surrounded by six α-helices (Fig. 2.10) (Ollis et al., 1992).
Different hydrolases show variability
around this prototypical structure in terms of the number of
β-strands and α-helices. For
example, a lipase from Bacillus subtilis and a lipase from
Pseudomonas cepacia are both
composed of six β-strands (Van Pouderoyen et al., 2001; Kim et
al., 1997b), while an esterase
from Pseudomonas fluorescens is made of seven β-strands (Kim et
al., 1997a).
R1 OR2
O
R1 OR3
O
HOR3 + ⇌ HOR2 +
-
28
Figure 2.10. Prototypical α/β hydrolase fold structure. A) Image
adapted from Ollis et al.
(1992). α helices and β strands represented by cylinders and
arrows, respectively. Dark circles in
loop regions after β5, β7, and β8 show positions of catalytic
residues serine, aspartate/glutamate,
and histidine, respectively. B) 3-Dimensional structure of P.
fluorescens esterase showing α/β
hydrolase fold. Catalytic triad residues serine, aspartate, and
histidine are shown as sticks
coloured red, blue, and magenta, respectively. Oxyanion hole
residues are shown as sticks
coloured in cyan.
A)
B)
-
29
The catalytic residues form a triad and are found in the order
of serine, aspartate, and
histidine in the primary sequence of the enzyme. In some cases,
glutamate is present in place of
aspartate, as in the case of a feruloyl esterase from Pleurotus
eryngii (Nieter et al., 2014). The
catalytic nucleophilic serine is typically contained in the
consensus sequence Gly-X-Ser-X-Gly.
The catalytic serine is located on a sharp γ-like turn (termed
the nucleophilic elbow) of the
enzyme secondary structure, going from β5 to the following alpha
helix (αC) in the prototypical
α/β hydrolase structure (Fig. 2.10) (Ollis et al., 1992). An
important structural feature of the
enzyme, which helps to stabilize the tetrahedral intermediate of
the substrate formed during
catalysis, is known as the oxyanion hole. This oxyanion hole is
formed by the main-chain amide
hydrogens of two amino acid residues. One of these residues is
right after the nucleophilic serine
in the nucleophilic elbow, while the second residue is located
at a loop going from β3 to αA in
the prototypical α/β hydrolase structure (Fig. 2.10) (Ollis et
al., 1992).
A structural feature that is unique to lipases as opposed to
esterases is the presence of a
lid that covers the enzyme active site. This lid is formed from
α-helical segments of the protein
that are mobile to allow access of the substrate to the active
site (Brady et al., 1990; Brzozowski
et al., 2000; Grochulski et al., 1994; Kim et al., 1997b). The
presence of the lid and its
movement have been proposed as an explanation for the phenomenon
of interfacial activation of
lipases (Brzozowski et al., 1991). Interfacial activation is the
increase in lipase activity that is
observed when the enzyme is present in a solution where the
substrate concentration is high
enough to form a separate phase (Verger, 1997). Upon opening of
the lipase lid, hydrophobic
patches on the underside of the lid become exposed and may be
stabilized by interaction with
the hydrophobic substrate phase at the solution interface
(Brzozowski et al., 1991).
Additionally, the lipase active site becomes exposed for
catalysis. A recent experiment showed
the feasibility of altering substrate preference of Rhizopus
chinensis lipase to favour short chain
substrates by swapping the lipase