- .
1|...
bi!

 

 

 

 

 

z,
w
..

mm

 

THESIS

t.)

/ 009

This is to certify that the

dissertation entitled

ISOLATION AND CHARACTERIZATION OF ANTI-OXIDANT
AND ANTI-INFLAMMATORY CONSTITUENTS FROM
ECHINACEA PURPUREA (L.) MOENCH

 

 

presented by

Laura Jean Clifford

has been accepted towards fulfillment
of the requirements for

Ph.D. degreein Food Science and

 

 

Environmental Toxicology

Jaw, J/wwfl

Major professor

 

Date 4‘fi 2 1007’

MS U i: an Affirmative Action/Equal Opportunity Institution 0-12771

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DatoDuopes-p. 15

 

ISOLATION AND CHARACTERIZATION OF ANTI-OXIDANT AND ANTI-
INFLAMMATORY CONSTITUENTS FROM ECHINACEA PURPUREA (L.)
MOENCH
By

Laura Jean Clifford

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
FOOD SCIENCE AND ENVIRONMENTAL TOXICOLOGY
Department of Food Science and Human Nutrition

2002

ABSTRACT

ISOLATION AND CHARACTERIZATION OF ANTI-OXIDANT AND ANTI-
INFLAMMATORY CONSTITUENTS FROM ECHINACEA PURPUREA (L.)
MOENCH
By

Laura Jean Clifford

A variety of anecdotal claims have been attributed to Echinacea purpurea
(L.) Moench, including immune systems enhancement and pain and fever
relieving properties. However, questions still remain as to the identity and mode
of action of the compounds responsible for these biological effects. Research to
identify specific constituents responsible for reported anecdotal claims has
established several compounds with significant biological potential. Alkamides
(undeca-ZE,4Z-dien-8,10-diynoic acid isobutylamide, undeca-22,4E-dien-8,10-
diynoic acid isobutylamide, dodeca-2E,4Z-dien-8,10-diynoic acid isobutylamide,
undeca-ZE,4Z-dien-8,10-diynoic acid 2-methylbutylamide, dodeca-ZE,4Z-dien-
8,10-diynoic acid 2-methylbutylamide, and a mixture of dodeca-ZE,4E,82,10E-
tetraenoic acid isobutylamide and dodeca-2E,4E,82,1OZ—tetraenoic acid
isobutylamide) were isolated from E. purpurea roots and examined in in vitro
model systems. Significant cyclooxygenase-1 (COX-1) and cyclooxygenase-2
(COX-2) inhibitory activities, ranging from of 36-60% and 15-46%, respectively,
were observed for these compounds at a concentration of 100 ug/mL. The
alkamides from E. purpurea were mosquitocidal against Aedes aegyptii L. larvae

at 10 and 100 ug/mL. Flavonoids (quercetin-B-O—glucoside, quercetin-3-O-

rutinoside, and kaempferol-3-O-robinobioside) and anthocyanins (cyanidin-3-O-
[3-D glucopyranoside and cyanidin-3-O—malonyl-(1—+6H3-D glucopyranoside),
were isolated from methanolic extracts of Iyophilized E. purpurea flowers and
examined for antioxidant and COX-1 and COX-2 inhibitory activities. Four
additional reported constituents of E. purpurea, caffeic acid and the caffeoyl
derivatives caftaric acid, chlorogenic acid, and Cichoric acid were also tested.
The anthocyanins demonstrated the greatest cyclooxygenase inhibitory activities
at 100 ug/mL, with inhibitions of COX-1 and COX-2 ranging from 23-28%. Co-
treatment of RAW 264.7 macrophages with several of the isolated compounds
(at 1, 10, 50, and 100 ug/mL) and 100 or 1000 ng/mL lipopolysaccharide (LPS)
suppressed induction of the pro-inflammatory cytokines lL-6 and TNF-a.
Cytotoxicity was noted at 100 pg/mL for all four alkamides, based upon results
Obtained from a concurrent MTT assay. Cells treated with compounds from E.
pumurea did not induce lL-6 or TNF-a production; however, CO-treatment of
macrophages with caffeoyl derivatives (50 and 100 ug/mL) and LPS (100 and
1000 ng/mL) resulted in an increase in lL-6 and TNF-oc levels, suggesting
potential immune-enhancing activity. The alkamides demonstrated the greatest
ability to reduce LPS-induced lL-6 and TNF-oc production. The anti-inflammatory
activity of the alkamides reported here supports the claims of pain and fever relief
associated with Echinacea use. The Opposing activities noted for compounds
from E. purpurea demonstrate the importance of testing individual compounds

rather than crude extracts.

To all individuals who see the invisible and do the impossible.

iv

ACKNOWLEDGEMENTS

I would like to thank my major advisor, Dr. Muralee Nair, for his support
and direction during my graduate program. In addition, I would like to thank Dr.
James Pestka for his guidance and invaluable help in providing financial backing
for the completion of my degree. I am grateful for the support of my committee
members, Dr. Gale Strasburg, Dr. Les Bourquin, and Dr. James Trosko, for their
valuable suggestions regarding my program of study. For teaching me a great
deal and for providing me with financial support, I thank Dr. Janice Harte.

The support and aid Of my labmates, who made my Ph.D. program both
possible and enjoyable, leaves me indebted to soon-to-be-Drs. Marvin Weil,
Belinda Hawkins, and Alexa Smolinski as well as Drs. Robert Cichewicz, Carolyn
Ross, Di Wu, Rafik Momin, and Navindra Seeram. Thanks, guys, for all of your
assistance and laughs!

I would like to thank my ever-patient family for supporting me through my
long years of study. I thank my wonderful, loving parents Clayton and Martha
Clifford, my sisters Nancy and Susan, and my aunt Jean Wirtz, who are all happy
to know that I am “finally done with school!” A big thank-you to Elizabeth Cowan,
my very best friend, who has always believed in me through our many years of
friendship. Last, but definitely not least, I would like to thank Robert Henry
Cichewicz ll, my fiance, who pushed, prodded, stood by, provided copious
amounts of caffeine, supported, and cheered me onward. I am so very fortunate

that you chose to pursue your Ph.D. degree at Michigan State University!

TABLE OF CONTENTS

Introduction ................................................................................................ 1
Chapter 1. Literature Review .................................................................... 6
Botany of Echinacea species ................................................................. 6
Historical use of Echinacea spp. ............................................................ 8
Chemistry of Echinacea spp. ............................................................... 10
Essential Oils ................................................................................ 11
Alkamides ...................................................................................... 1 5
Anthocyanins ................................................................................. 18
Caffeic Acid Derivatives ................................................................ 20
Flavonoids ..................................................................................... 23
Polyacetylenes .............................................................................. 24

Analytical methods for the identification of Echinacea constituents... 26

Stability of Echinacea constituents ..................................................... 28
Biological activity of Echinacea spp. ................................................... 30
Toxicity of Echinacea spp .............................................................. 31
Antimicrobial properties of Echinacea spp. ................................... 34
Effects of Echinacea on glucose metabolism ................................ 35
Potential reproductive effects of Echinacea products .................... 35
Immune stimulating properties of Echinacea spp. ......................... 36
Antioxidant properties ................................................................... 37

vi

Anti-inflammatory properties ......................................................... 39

Clinical studies denoting the effectiveness of Echinacea for
treating upper respiratory tract infections ........................................... 41

Clinical studies failing to demonstrate the effectiveness of Echinacea

for treating upper respiratory tract infections ...................................... 42
Mixed commercial preparations used in Clinical trials ......................... 44
Hypothesis .......................................................................................... 46
Chapter 2. Cyclooxygenase and mosquitocidal alkamides from
Echinacea purpurea (L.) Moench roots .................................................... 47
Abstract .............................................................................................. 47
Introduction ......................................................................................... 48
Materials and methods ....................................................................... 50
General experimental .................................................................... 50
Plant material ................................................................................ 50
Cyclooxygenase inhibitory assay .................................................. 51
Mosquitocidal assay ...................................................................... 52
Results ............................................................................................... 53
Discussion .......................................................................................... 58

Chapter 3. The isolation and purification of bioactive compounds from

Echinacea purpurea (L.) Moench flowers ........................................... 62
Abstract .............................................................................................. 62
Introduction ......................................................................................... 63
Materials and Methods ....................................................................... 65
General Experimental ................................................................... 65
Plant material and extraction ......................................................... 66

vii

Purification of anthocyanins .......................................................... 67

Purification of flavonoids ............................................................... 69
Results and discussion ....................................................................... 71
Purification of anthocyanins .......................................................... 71
Purification of flavonoids ............................................................... 75

Chapter 4. Cyclooxygenase-1 and —2 inhibitory and antioxidant activities

of compounds from Echinacea purpurea (L.) Moench ........................ 79
Abstract .............................................................................................. 79
Introduction ......................................................................................... 80
Materials and Methods ....................................................................... 82
General Experimental ................................................................... 82
Cyclooxygenase Inhibitory Assay .................................................. 82
Antioxidant Assay .......................................................................... 84
Results and Discussion ...................................................................... 85
Cyclooxygenase Inhibitory Assay .................................................. 85
Antioxidant Assay .......................................................................... 88

Chapter 5. Characterization of cytokine production in RAW 264.7
murine macrophages following treatment with compounds

from Echinacea purpurea (L.) Moench ............................................... 96
Abstract .............................................................................................. 96
Introduction ......................................................................................... 98
Materials and Methods ..................................................................... 102
General Experimental ................................................................. 102
Cell Culturing ............................................................................... 102

viii

MTI' Assay .................................................................................. 104

ELISA Quantification Of lL-6 and TNF-a ...................................... 105
Statistical Analysis ...................................................................... 106
Results and Discussion .................................................................... 107
MTT Assay .................................................................................. 107
Cytokine Production .................................................................... 107
Conclusions ........................................................................................... 131
Literature Cited ...................................................................................... 136
Appendix A. Conversion of assay concentrations ................................. 147

ix

LIST OF TABLES

Table 1.1. Volatile compounds from the headspace of
E. purpurea tissues ........................................................................ 13-14

Table 3.1. 13C Chemical shifts for anthocyanins isolated
from E. purpurea ................................................................................. 72

Table 3.2. 13C Chemical shifts for flavonoids isolated
from E. purpurea ................................................................................. 76

LIST OF FIGURES

Figure 1.1. Alkamides from E. purpurea roots ........................................ 16
Figure 1.2. Alkamides from E. angustifolia roots ..................................... 17
Figure 1.3. Anthocyanins isolated from E. purpurea ............................... 19
Figure 1.4. Caffeic acid derivatives from Echinacea species ............. 21-22
Figure 1.5. Flavonoids from Echinacea species ..................................... 25
Figure 2.1. Alkamides isolated from E. purpurea roots. ..................... 54-56
Figure 2.2. Mosquitocidal activity of E. purpurea alkamides ................... 57
Figure 2.3. Comparative inhibition of COX-1 and COX-2

enzymes by E. purpurea alkamides .................................................... 59
Figure 3.1. Extraction scheme for anthocyanins and flavonoids

isolated from E. purpurea flowers. ...................................................... 68
Figure 3.2. Anthocyanins isolated from E. purpurea flowers ................... 73
Figure 3.3. Flavonoids isolated from E. purpurea flowers ....................... 77

Figure 4.1. Comparison of cyclooxygenase-1 and -2 inhibitory
activities for compounds from E. purpurea. ........................................ 86

Figure 4.2. COX-1 and COX-2 inhibitory activities for anthocyanins
isolated from E. purpurea flowers at 100, 50 and 10 ug/mL. .............. 87

Figure 4.3. Antioxidant activity exhibited by compounds from
E. purpurea assayed at 10 pg/mL ....................................................... 89

Figure 4.4. Antioxidant activity at 21 minutes for quercetin-3-O-
glucoside, quercetin-3-O-rutinoside, and kaempferoI-3-O-
robinobioside at 10, 50, and 100 ug/mL. ............................................ 90

Figure 5.1. MTT assay results for RAW 264.7 macrophages treated
with 1, 10, 50, or 100 ug/mL of alkamide 1 from E. purpurea. .......... 108

Figure 5.2. MTT assay results for RAW 264.7 macrophages treated
with 1, 10, 50, or 100 pg/mL of alkamide 2 from E. purpurea. .......... 109

xi

Figure 5.3. MTT assay results for RAW 264.7 macrophages treated
with 1, 10, 50, or 100 pg/mL of alkamide 4 from E. purpurea. .......... 110

Figure 5.4. MTT assay results for RAW 264.7 macrophages treated
with 1, 10, 50, or 100 ug/mL of alkamides 6/7 from E. purpurea. ..... 111

Figure 5.5. Effects of CO-treatments of 50, 10, or 1 ug/mL of alkamide
1 with LPS on IL-6 and TNF-a production by RAW 264.7 cells. ....... 112

Figure 5.6. Effects of CO-treatments of 50, 10, or 1 pg/mL of alkamide
2 with LPS on lL-6 and TNF-a production by RAW 264.7 cells. ....... 113

Figure 5.7. Effects of CO-treatments of 50, 10, or 1 pg/mL of alkamide
4 with LPS on lL-6 and TNF-a production by RAW 264.7 cells. ....... 114

Figure 5.8. Effects of co-treatments of 50, 10, or 1 ug/mL of alkamides
6!? with LPS on lL-6 and TNF-a production by RAW 264.7 cells. 115

Figure 5.9. Effects of CO-treatments of 50, 10, or 1 ug/mL of cyanidin-
3-O-glucoside (8)with LPS on IL-6 and TNF-or production by RAW
264.7 cells. ....................................................................................... 116

Figure 5.10. Effects of CO-treatments of 50, 10, or 1 ug/mL of cyanidin-
3-O-malonyI-(1——>6)-B-D glucopyranoside (9) with LPS on lL-6 and
TNF-or production by RAW 264.7 cells. ............................................ 117

Figure 5.11. Effects of co-treatments of 50, 10, or 1 ug/mL of quercetin-
3-O-rutinoside (11) with LPS on lL-6 and TNF-a production by
RAW 264.7 cells. .............................................................................. 118

Figure 5.12. Effects of co-treatments of 50, 10, or 1 ug/mL of

quercetin-3-O-glucoside (12)with LPS on IL-6 and TNF-a
production by RAW 264.7 cells ......................................................... 119

Figure 5.13. Effects of co-treatments of 50, 10, or 1 pglmL of caffeic
acid (13) with LPS on lL-6 and TNF-a production by RAW
264.7 cells. ....................................................................................... 120

Figure 5.14. Effects of co-treatments of 50, 10, or 1 ug/mL of caftaric
acid (14) with LPS on lL-6 and TNF—a production by RAW
264.7 cells. ....................................................................................... 121

Figure 5.15. Effects of co-treatments of 50, 10, or 1 ug/mL of Cichoric

acid (15) or with LPS on lL-6 and TNF-a production by RAW
264.7 cells. ....................................................................................... 122

xii

Figure 5.16. Effects of co-treatments of 50, 10, or 1 ug/mL of
chlorogenic acid (16)with LPS on IL-6 and TNF-a production by
RAW 264.7 cells. .............................................................................. 123

xiii

INTRODUCTION

Establishing the safety and efficacy of natural product supplements
through scientific validation is of primary concern to both consumers and
manufacturers of phytoceuticals. Although one-half of the world’s populace uses
traditional medicines as a primary source of medical treatment (Wildman, 2001),
Americans have only recently begun to add significant quantities of natural
product supplements to their normal medical regimens (Borchers et al., 2000;
Miller, 1998). The increase in the popularity of nutraceuticals and phytoceuticals
as an addition or alternative to Western medicine might be attributed to the
public’s heightened awareness of preventable diseases, such as diabetes,
cancer, and heart disease, as well as to the increase in the percentage of older
Americans (Wildman, 2001). Alternative medicines have the potential to improve
quality of life and provide a new means of disease prevention or management
(Ness et al., 1999); however, phytoceuticals might also interfere with mainstream
medical treatments.

Nutraceuticals do not require testing for safety and efficacy prior to
marketing (Percival, 2000; link and Chaffin, 1998). As a result, the active
components in many commercially available products have not been identified
(Mazza and Cottrell, 1999; Melchart et al., 1995). Due to the lack of
standardization in the phytoceutical industry (Perry et al., 2001) and the unstable
nature of compounds in natural product extracts (Kim et al., 2000a; NUsslein et
al., 2000), greater regulation is needed to protect consumers. While it has been

suggested that Echinacea extracts and products should be standardized in order

to provide a consistent product to consumers, many natural product supplements
do not undergo rigorous quality or standardization testing (Bergeron et al., 2000;
Bauer et al., 1998).

Regulatory issues have become increasingly important as a result of the
soaring increase in the phytoceutical market, the lack of standardized extracts,
the potential instability of bioactive compounds in extracts, and the lack of safety
and/or efficacy testing. In 1993, the Food and Drug Administration (FDA)
implemented the Nutrition Labeling and Education Act (NLEA), which
distinguished between the definitions of “food” and “drug” (Childs, 2001) and
provided the basis for stronger regulatory control by the FDA. In response to the
NLEA, the dietary supplement industry pursued the establishment of a separate
category for dietary supplements with regard to labeling and advertising claims.
The result was the formation of the Dietary Supplement Health and Education
Act (DSHEA) (Wildman, 2001; Stephen, 1998). The recently proposed
Nutraceutical Research and Education Act defines nutraceuticals as “a dietary
supplement, food, or medical food that possesses health benefits and is safe for
human consumption in such quantity and with such frequency as required to
realize such properties” (Henry, 1999).

Heightened regulatory control and the establishment of a formal definition
of nutraceuticals have provided a strong incentive for scientists and the natural
product supplement industry to establish scientific credibility for herbal
supplement use. Nutraceuticals have been perceived to be inherently safe by

consumers because they are “natural,” even without adequate toxicity testing

(link and Chaffin, 1998). The impetus to prove the safety and efficacy of
nutraceuticals will provide solid scientific data to substantiate and/or disprove
anecdotal claims of biological effects. Identification of specific, active
components in natural product supplements is a vital area of research, due in
part to the potential for drug-herb interactions (Miller, 1998).

In a recent survey, an estimated four in ten Americans used some type of
unconventional treatment, spending approximately $34 billion on alternative
medicines (Ness et al., 1999). Echinacea supplements and mixtures of herbal
supplements containing Echinacea are among the most popular natural product
supplements in the United States (Barrett et al., 1999). Annual sales of
Echinacea nutraceuticals are estimated at $300 million (Gunning, 1999) and are
expected to increase in forthcoming years, allowing Echinacea preparations to
remain a top-selling nutraceutical (Stephen, 1998).

Echinacea purpurea (L.) Moench, Echinacea angustifo/ia DC, and
Echinacea pallida Nutt. are used to the greatest extent in phytoceutical products
(Barrett et al., 1999; Bauer and Wagner, 1991). Echinacea preparations are sold
in the form of tablets, capsules, tinctures (alcoholic extracts), juices, and teas
(Gunning, 1999; Li and Wang, 1998). Alcoholic extracts of Echinacea roots and
above-ground portions of the plant are most commonly consumed. In Germany,
more than 800 Echinacea-based extracts are commercially available (Bauer,
1998; Lienert et al., 1998), and the number of preparations available in the
United States is increasing. Many nutraceuticals have come under scrutiny due

to the lack of credible pharmacological and chemical data supporting their use. A

limited number of studies have been conducted to substantiate anecdotal claims
associated with Echinacea use (Borchers et al., 2000). The claims of pain relief
and general immune stimulation have been evaluated to some extent (Bauer,
1998; Bauer and Wagner, 1991; Proksch and Wagner, 1987). Anti-viral,
antioxidant, and anti-inflammatory activities have been reported for various
Echinacea extracts (Facino et al., 1995; Bauer & Wagner, 1991; Wagner et al.,
1988; Luettig et al., 1989). Despite these efforts, additional research is needed
in order to validate the anecdotal health claims attributed to Echinacea dietary
supplements.

Anecdotal claims of immune enhancement, fever relief, and alleviation of
pain have been associated with the use of E. purpurea preparations. The
bioactive compounds responsible for anecdotal Claims have not been fully
eluciated to date. This study was undertaken in order to identify bioactive
constituents of E. purpurea through the isolation, structure elucidation, and
biological testing of the active compounds in E. purpurea roots and flowers. Five
primary Objectives were defined in order to achieve this goal. These Objective
were: 1) to isolate and compare COX-1 and COX-2 inhibitory activities of
alkamides from E. purpurea roots; 2) to evaluate the mosquitocidal activity of E.
purpurea root alkamides; 3) to isolate flavonoids and anthocyanins from E.
purpurea flowers; 4) to characterize of the COX-1 and COX-2 inhibitory activities
and antioxidant activities of the flavonoids and anthocyanins from E. purpurea
flowers; and 5) to characterize and examine the effects of purified constituents

from E. purpurea roots and flowers (alkamides, flavonoids, and anthocyanins) on

cytokine production. The following literature review focuses on the Chemical
constituents and biological activities of extracts and compounds from of

Echinacea spp., with emphasis on those obtained from E. purpurea.

CHAPTER 1. LITERATURE REVIEW

The genus Echinacea is composed Of nine Echinacea species that are
native to central and eastern North America. Only Echinacea purpurea,
Echinacea angustifolia, and Echinacea pallida are of significant phytoceutical
interest based on historical use, availability, and biological activity. A large
amount of information has been generated regarding the biological activity of
Echinacea spp., primarily as crude extracts (the usual form available to
consumers), or in some cases, as pure compounds isolated from E. purpurea or
E. angustifolia. This review summarizes the available literature regarding
Echinacea species, with specific attention given to the botany, chemistry, and

biological activities of E. angustifolia, E. pallida, and E. purpurea.

Botany of Echinacea species

Echinacea purpurea (Asteraceae), also known as the purple coneflower, is
a perennial plant that grows to a height of approximately two to three feet (Foster
and Duke, 1990). Echinacea derives its name from the Greek word, “echinos,”
(“hedgehog” or “sea urchin”) referring to the round, spiny seedhead (Kindscher,
1989; Gunning, 1999). Although Echinacea species are grown for phytoceutical
purposes, their commercial potential extends to the use of Echinacea as popular
ornamentals in gardens (Li, 1998: Brown, 1986) and for cut flowers (Li, 1998;
Starrnan et al., 1995). Echinacea plants grown from seed generally flower in two

to three years during the months of June and July in the central United States

(Kindscher, 1989). Flowering of E. purpurea may be induced artificially in a
greenhouse environment (Runkle et al., 1998).

Echinacea purpurea plants possess straight, slightly pubescent stalks
terminating with a single flower (Hunter, 1995). Flowers have a center disk that
is orange (Hunter, 1995) and bristled (Foster and Duke, 1990), from which
reddish-purple to pink rays droop (Gleason and Cronquist, 1991). The flowers of
E. purpurea differ from those of E. angustifolia by the deeper purple hue, as well
as the broader rays of the flower itself (Hunter, 1995). The leaves of E. purpurea
are lance-shaped to elliptical (Gleanson and Cronquist, 1991), coarsely toothed
(Foster and Duke, 1990), and extend approximately 15 cm, 1.5-5 times in length
as compared to width (Gleason and Cronquist, 1991). The roots of E. purpurea
consist of a fibrous-rooted crown, ranging from simple to branched (Gleason and
Cronquist, 1991). Echinacea purpurea is primarily found in prairie regions of the
United States, extending from Texas to North Dakota and from Colorado to
Kansas (Foster and Duke, 1990). Echinacea purpurea may also be found
irregularly in Michigan, Kentucky, Tennessee, and Georgia (Gleason and
Cronquist, 1991).

Echinacea angustifolia and E. purpurea are winter-hardy, heat-tolerant,
and drought-resistant. E. purpurea is less dependent upon soil pH and moisture
content than E. angustifo/ia (Li, 1998). Although native only to the Great Plains
region (Foster and Duke, 1990), cultivation has resulted in the widespread
availability of Echinacea for medicinal purposes. Echinacea roots, which are

used to the greatest extent in commercial preparations, are generally harvested

in the third or fourth year (Li, 1998). Growth parameters have been altered to
experimentally increase the yield of root material for phytoceutical preparations

(Trypsteen et al., 1991).

Historical use of Echinacea spp.

Currently, seven of the top ten most used botanicals in the United States
are based on plants used medicinally by Native Americans (Borchers et al.,
2000). Echinacea angustifolia was one of the predominant species used by
Native American Indians for the treatment of a variety of ailments. Echinacea
roots were used externally to treat burns, insect bites, and wounds and internally
to alleviate the pain of toothaches, headaches, chills, stomach cramps, and
coughs (Bauer and Wagner, 1991). The tribes that used Echinacea included the
Cheyenne, Choctaw, Dakota, Delaware, Fox Kiowa, Montana, Omaha Pawnee,
Ponca, Sioux, and Winnebago (Borchers et al., 2000). Teas were used to treat
rheumatism, arthritis, mumps, and measles (Kindscher, 1989). Roots were most
often used; however, the juice squeezed from whole Echinacea plants and
infusions brewed from fresh herb were also used, but to a lesser extent (Bauer
and Wagner, 1991).

Early American settlers observed Echinacea use among Native Americans
for the treatment of various ailments, and adopted its use for similar medical
conditions. In addition, new uses were developed for Echinacea spp. including
the treatment of syphilis. American settlers also used Echinacea externally on

horses as a treatment for saddle sores (Bauer and Wagner, 1991). Echinacea

angustifolia was popularized among American settlers in 1871 by H. C. F. Meyer,
who marketed a tincture of E. angustifolia as “Meyer’s Blood Purifier” (Kindscher,
1989; Foster, 1985). Meyer’s natural remedy was accompanied by exaggerated
claims of miracle cures for snake bites, bee stings, ‘mad dog’ bites, tumors,
eczema, gangrene, typhoid fever, mountain fever, malaria, and diphtheria (Bauer
and Wagner, 1991: Kindscher, 1989). Studies performed in the early 1900's did
not find any physiological activity, and Echinacea was determined to be
ineffective for botulism, anthrax, rattlesnake venom, tetanus, septicemia, and
tuberculosis. The lack of substantiation of medicinal properties of Echinacea
aided in the decline of its popularity, and as a result Echinacea did not achieve
widespread acceptance in the United States (Kindscher, 1989).

Echinacea preparations did not suffer the same fate in Europe, where they
became popular in the early 19003 as a means to boost immunity and fight
chronic infections (Bauer and Wagner, 1991). A great deal of research regarding
the effectiveness of Echinacea has been performed in Europe, where laws
regarding commercial availability and use of herbal supplements are much more
liberal (Kindscher, 1989; Tyler, 1986). Based on the success enjoyed by
Echinacea products in Europe, a resurgence in its popularity was seen in the
United States in the late twentieth century. Today Echinacea is one of the top-
selling herbal supplements in America (Gunning, 1999; Henry, 1999).

The popularity of Echinacea as a dietary supplement is due in part to its
reported antiviral and nonspecific immunostimulatory properties (Borchers et al.,

2000; Proksch and Wagner, 1987). The German Commission E Monographs

lists E. purpurea herb oral preparations as approved for respiratory tract
infections, urinary tract infections, and colds. In addition, E. purpurea herb is
approved for topical application to promote wound healing (Blumenthal, 1998;
Percival, 2000).

Echinacea angustifolia and E. pallida herb preparations and E. purpurea
and E. angustifolia root preparations have not been approved by the German
Commission E (Percival, 2000). The lack of approval for these four Echinacea
preparations is due to insufficient research or inadequate identification of the
materials used in the trials (Percival, 2000). The therapeutic efficacy of
commercial preparations from other Echinacea plant parts has not been clearly
established (Turner et al., 2000; Barrett et al., 1999; Grimm and Miller, 1999;
Cheminat et al., 1989). The bioavailability, potency, potential synergistic effects,
and mechanisms of action of the biologically active components are unknown
(Percival, 2000), leaving a great deal of research yet to be performed on

Echinacea species.

Chemistry of Echinacea spp.

Echinacea angustifolia was the primary species used by Native American
Indians for treating illnesses and injuries (Cheminat et al., 1989). Due to its
commercial availability, E. purpurea has also become widely used in popular
modern-day Echinacea herbal supplements. The literature generated in the last

two decades reflects the inclusion of E. purpurea in biological testing. The

10

phytoceutical activity Of mixtures of E. angustifolia, E. purpurea, and E. pallida
are often reported.

Currently, E. angustifolia, E. purpurea, and E. pallida are of primary
commercial and nutraceutical interest (Bauer and Foster, 1991; Bauer et al.,
1988a), and similarities in phytoceutical effects and chemical profiles have been
demonstrated (Bauer and Wagner, 1991). Echinacea simulata McGregor,
Echinacea paradoxa (Norton) Britton, and Echinacea tennesseensis (Beadle)
Small (a rare species located in the Cedar Glade region Of Tennessee) have
been evaluated for potential commercial use. These three species of Echinacea
demonstrated similar biological activities and possessed chemical profiles like
those of E. angustifolia and E. pallida (Bauer and Foster, 1991; Bauer et al.,
1990). The incorporation of E. simulate, E. paradoxa, and E. tennesseensis into
commercial natural product supplements is currently unlikely, given their lack of
widespread, commercial availability.

Several classes Of chemical compounds have been isolated from
Echinacea species. Analysis of the roots, leaves, stems, and flower heads has
resulted in the identification of a number of alkamides, caffeic acid derivatives,

flavonoids, and polyacetylenes. These groups will be further reviewed.

Essential Oils
The essential oil of Echinacea spp. has been reported to contain a-
humulene, borneol, germacra-4(15), 5E,10(14)—trien-1—fl~ol, a-pinene, Iimonene,

caryophyllene oxide, bornyl acetate, vanillin, flpinene, carvomenthene, ,8

ll

caryophyllene, germacrene D, p-hydroxy-cinnamic acid methylester, ,6—
farnesene, and myrcene (Bauer and Wagner, 1991). It has been suggested that
Echinacea spp., due to their essential Oil content rich in borneol and a-pinene,
might find applications in the cosmetic industry (Li and Wang, 1998).

The essential oil content for the roots of E. angustifolia has been reported
in the range of 0.04 to 0.1% (Bauer and Wagner, 1991), with ariel parts, flowering
ariel parts, and dried leaves of E. angustifolia possessing less than 0.1% (Bos et
al., 1988; Heinzer et al., 1988; Bauer and Wagner, 1991). The essential oil
content of fresh E. purpurea roots has been reported to range from 0.005-0.22°/o
(Bauer and Wagner, 1991), while the fresh herb (referring to the above-ground
portion of the plant, whether flowering or not) ranges from 0.08-0.32°/o, and fresh
flowerheads from 0.13-0.48% (Bauer and Wagner, 1991).

Germacra-4(15),5E,10(14)-trien-1,8»ol, a sesquiterpene alcohol, was
Isolated from fresh aerial parts of E. purpurea (Bauer et al., 1988d). Due to its
volatile nature, it is not usually found in dried Echinacea plant materials.
Capillary gas chromatography coupled with mass spectrometry has been used to
identify over seventy volatile compounds present in the roots, flowers, stems, and
leaves of chopped and ground E. angustifolia, E. purpurea, and E. pallida. E.
Purpurea volatile compounds have been summarized (Mazza and Cottrell, 1999)

(Table 1.1).

12

Table 1.1. Volatile compounds from the headspace of E. purpurea tissues
(adapted from Mazza and Cottrell, 1999).

 

 

Root Flower Leaf Stem
acetaldehyde X X X X
dimethyl sulfide X X X
propanal X
2-methylpropanol X X X
2—propenal X
2-butanone X
2-methylbutanol X X X
3-methylbutanol X X X X
1-methylpropyl acetate X
trichloroacetic acid X
a-pinene X X X X
a-thujene X
geranyl acetate X X
camphene X X X X
hexanal X X X X
B-pinene X X X X
2-methyI-1-propanol X
sabinene/B-thujene X X X X
2-pentenal X X
2-methyl-4-pentenal X X X
B-myrcene X X X
a-phellandrene X
or-terpinene X X X X
heptanal X
Iimonene X X X X
2-hexenal (cis) X
2-methyl-1-butanol X
3-methyl-1-butanol X
2-hexenal (trans) X X X
ocimene X X X X
y-terpinene X X X
trans-ocimene X X X
p-cymene X X X
hexyl acetate X X
or-terpinolene X X X
3-hexen-1-OI acetate X X
6—methyl-5-hepten-2-one X
1-hexanol X X X X
3-hexen-1-ol (trans) X
Allo-ocimene X X X

 

l3

Table 1.1. Volatile compounds from the headspace of E. purpurea tissues
(adapted from Mazza and Cottrell, 1999), continued.

 

Root" MEI we: Leaf Stem

 

 

3-hexen-1-OI (cis) X X X
2-hexen-1-OI (trans) X
1-octen-3-ol X
benzaldehyde X
a—cubebene/a-copaene

a—ylangene

y-cadinene

trans-caryophyllene X
calarene

germacrene D

5-ethyl-2(5H)-furanone

5-cadiene

Ba-cubebene

2,2,3,3-tetramethyl hexane

X
X

><><><><><><><><><
XX X X

 

l4

Alkamides

A number of alkamides have been isolated from E. purpurea roots (Figure
1.1). Two dodeca-2,4,8,10-tetraenoic acid isobutylamides, undeca-(22,4E)—dien-
8,10-diynoic acid isobutylamide and dodeca-(ZZ,4E)-dien-8,10-diynoic acid
isobutylamide, were isolated from the roots of E. angustifolia and E. purpurea
(Bohlmann and Grenz, 1966). Echinacea angustifolia roots yielded dodeca-
(2E,4E)-dienoic acid and deca-(ZE,4E,6E)-trienoic acid (Verelis, 1978; Bauer and
Wagner, 1991). Trideca-(2E,7Z)-dien-10,12-diynoic acid, trideca-(2E,6E,8Z)-
trien-10,12-diynoic acid and pentadeca-(ZE,92)-dien-12,14-diynoic acid, trideca-
(2E,7Z)-dien-10,12-diynoic acid (2-methylbutyl)amide, and pentadeca(2E,92)-
dien-10,14-diynoic acid (2—hydroxyisobutylamide) were identified in the ariel parts
of E. purpurea (Bohlmann and Hoffmann, 1983). An additional five alkamides
from hexane extracts of E. purpurea roots were identified, including undeca-
(2E,4Z)-dien-8,10-diynoic acid isobutylamide, dodeca-(2E,4Z)-dien-8,10-diynoic
acid isobutylamide, dodeca-(2E,4E,10E)-trien-8-ynoic acid isobutylamide,
dodeca-(2E,4E,8Z)-trienoic acid isobutylamide and dodeca-(2E,4Z)-dien-8,10-
diyonic acid 2-methylbutylamide (Bauer et al., 1988b) (Figure 1.1).

Fifteen alkamides (Figure 1.2) were isolated from hexane extracts of E.
angustifolia roots (Bauer et al., 1989). Undeca-(22)-en-8,10-diyonic acid
isobutylamide, dodeca-(2E)-en-8,10-diyonic acid isobutylamide, undeca-(2Z)-en-
8,10-diynoic acid 2-methylbutylamide, dodeca-(2E,4Z,1OZ)trien-8-ynoic acid
isobutylamide were identified in addition to previously isolated isobutylamides in

E. purpurea roots (Bauer et al., 1989).

15

Undeca-2E,4Z-dien-8,10-diynoic acid
isobutylamide

Dodeca-2E,4Z-dien-8,10-diynoic
acid isobutylamide

Dodeca-2E,4E,10E-trien-8-ynoic
acid isobutylamide

Dodeca-2E,4Z-dien-8,10-diynoic
acid 2-methybutylamide

O

._____ \\
N/Y
H

Dodeca-2E,4E,82,1OZ-tetraenoic
acid isobutylamide

Undeca-2Z,4E-dien-8,10-diynoic
acid isobutylamide

Undeca-2E,4Z-dien-8,10-diynoic
acid 2-methylbutylamide

o
- —~ 1,17
Trideca-2E,7Z—dien-10,12-diynoic

acid isobutylamide

N
H
Dodeca-2E,4E,10E-tetraenoic
acid isobutylamide
N
H

Dodeca-2E,4Z,8Z-trienoic acid
isobutylamide

Wk/iLNfi/

Dodeca-2E,4E-dienoic acid isobutylamide

Figure 1.1. Alkamides from E. purpurea roots (Bauer and Wagner, 1991; Bauer
etaI,1988b)

 

 

 

Undeca-2E-en-8,10-diynoic Undeca-22-en-8,10-diynoic acid
acid isobutylamide isobutylamide
O O
— — \/\/\/H\
_ _ \ _ __ _ \
N —— -— N /\l/
“W H
Dodeca-ZE-en-8,10-diynoic acid Undeca-2E,4Z-dien-8,10-diynoic
isobutylamide acid isobutylamide
O O
_ __ / __ __ __ \
_ __ N / —— N
H /\‘/ M H /\(
Undeca-22,4E—dien—8,10-diynoic acid Dodeca-2E,4Z,1OZ-trien-8-ynoic
isobutlyamide acid isobutylamide
O
0 )L
: : — \ : : _ N /Y\
Trideca-2E,7Z—dien-10,12-diynoic acid Undeca-22—en-8,10-diynoic acid 2-
isobutylamide methylbutylamide
O O
: : M N /Y\ W
Dodeca-ZE-en-8,10-diynoic acid Dodeca-2E,4E,10E—tetraenoic
2-methylbutylamide acid isobutylamide

WQkfi/Y _ _ _ \ N/Y

H
Dodeca-ZE,4E,8Z,1OZ-tetraenoic acid Pentadeca-ZE,92-dlen-12,14-dynoic

isobutylamide acid isobuylamide

O O
__ _ __ \

—— -- — \ — —— N
—— — tAf KY
Hexadeca-ZE,4Z-dien-12,14-diynoic Pentadeca-2E,92-dien-12,14-diynoic

acid isobutylamide acid isobutylamide

Figure 1.2. Alkamides from E. angustifolia roots (Bauer and Wagner, 1991;
Bauer and Reminger, 1989).

17

The development of pharmacologically active constituents from Echinacea
spp. is a significant area of academic focus. Alkamides, which are known to be
biologically active (Bauer and Wagner, 1991), are of particular interest. In an
effort to understand the development of the alkamides found in E. purpurea and
E. angustifolia, the germination of the achenes (fruits) was studied. Dodeca-
2E,4E,82,10E-tetraenoic acid isobutylamide and dodeca-2E,4E,8Z,10Z-
tetraenoic acid isobutylamide, the two primary alkamides isolated from adult
plants (Bauer and Remiger, 1989), were found to be the main alkamides formed
during germination Of E. purpurea and E. angustifolia (Schulthess et al., 1991).
Two additional isobutylamides found only in E. purpurea, trideca-2E, 7Z-diene-
10,12-ciynoic acid isobutylamide, isolated from the achene and roots, and
dodeca-2E,4E,10E-trien-8-ynoic acid isobutylamide, previously noted in the
roots, were also found in the cotyledons. Levels of the two isobutylamides

increased with primary leaf development (Schulthess et al., 1991).

Anthocyanins
Cyanidin 3-O—(fl-D-glucopyranoside) and cyanidin 3-O—(6-O—malonyl-fl-D-
glucopyranoside) (Figure 1.3) were isolated from dry E. purpurea and E. pallida
flowers by extraction with acetic acid-methanol-water followed by purification by
ion exchange chromatography (Cheminat et al., 1989). Two additional acylated
cyanidin glycosides were extracted from E. purpurea, however the structures

were not fully elucidated (Cheminat et al., 1989).

I8

OH
OH
+
H. o O
O \ ..
HO
/ 0mm
OH

OH

cyanidin
3-O-B-D-glucopyranoside

OH

OH
+
Ho 0 O
\
OH
/ OH0 O OH
OH

0W0“
cyanidin 0 0

3-O-malonyl-(1—>6)-B-D—
glu00pyranoside

Figure 1.3. Anthocyanins isolated from E. purpurea.

I9

Caffeic Acid Derivatives

Caffeic acid derivatives (Figure 1.4) are considered to be among the most
valuable and important active components of Echinacea spp. (Li and Wang,
1998). Cichoric acid (2,3-O—dicaffeoyl tartaric acid) was identified in E. purpurea
flowers and roots at 1.2-3.1 and 0.621%, respectively (Bauer et al., 1988c).
Echinacea angustifolia does not contain significant quantities of Cichoric acid.
Degradation of Cichoric acid during Soxhlet extraction was suggested (Bergeron
et al., 1999), based on the increased yield of Cichoric acid when performing
alcoholic, ultrasonic extraction of dried, ground roots.

Echinacoside, (fl-(3,4-dihydroxyphenyI)-ethyl-O-a-L-(1—>6)-4-O-caffeoyl-[3-
D-glucopyranoside), has been identified in the alcoholic extracts of E. angustifolia
and E. pallida roots (Bauer et al., 1988a), but has never been isolated from E.
purpurea roots, allowing a distinction to be made between the species.
Echinacoside has also been identified in the methanolic root extracts of E.
simulate and E. paradoxa (Bauer and Foster, 1991). The quinic acid derivative,
cynarine (1,5-O-dicaffeoyl quinic acid), is present in E. angustifolia but not E.
pallida, providing a chromatographic marker to distinguish these species (Bauer
et al., 1988C). Roots of E. tennesseensis, a rare species, do not contain
echinacoside (Bauer et al., 1990).

Twelve caffeoyl derivatives have been isolated from E. pallida (Cheminat
et al., 1988). Caffeic glycosides or caffeic esters of quinic or tartaric acid were
identified from methanol extracts of E. pallida partitioned with ethyl acetate and

n-butanol. Through preparative liquid Chromatography and high performance

20

HO \ O 0
R20 0
HO

 

 

OH
OH
OH
R R’
Echinacoside Glucose (1 ,6-) Rhamnose (1 ,3-)
6-O—Caffeoyl- 6-O-CaffeoyI-glucose Rhamnose (1 ,3-)
echinacoside (1 ,6-)
Verbascoside H Rhamnose (1 ,3-)
Desrhamnosyl- H H
verbascoside
R10 COOH O OH
R:
R20 OR; 0
0R3 caffeic acid
R1 R2 R3 R4
3—O-CaffeoyI-quinic acid H R H H
(chlorogenic acid)
Isochlorogenic acids H R R H
H R H R
H H R R
Cynarine R H H R

Figure 1.4. Caffeic acid derviatives from Echinacea species (Bauer and
Wagner, 1991; Cheminat et al., 1988).

21

OH

OH
I.
R — \ R5
O
OH
R”:
COOH OH

 

R1 R2 R3 R4 R5 R6
2-O—CaffeoyI-tartaric acid H H OH H - -
(Caftaric acid)
2,3-O-Di-caffeoyl-tartaric acid H R’ OH H OH H
(Cichoric acid)
2,3-O—Di-CaffeoyI-tartaric acid- CH3 R’ OH H OH H
methylester
2-O-Feruloyl-tartaric acid H H OCH3 H - -
2-O—CaffeoyI-3-O-coumaroyl-tartaric H R’ H H H H
acid
2-O-Caffeoyl-3-O-feruonl-tartaric acid H R’ OH H OCH H

3
2,3-O-Di-(5-[oc-carboxy-fl-(3,4- H R’ OH R” OH R”
dihydroxy-phenyI)—ethyl]-
caffeoyl)tartaric acid
2-O-CaffeoyI-3-O-(5-[a-carboxy-fl-(S,4- H R’ OH H OH R”

dihydroxy-phenyl)-ethyl]-
caffeoyl)tartaric acid

Figure 1.4. Caffeic acid derviatives from Echinacea species (Bauer and
Wagner, 1991; Cheminat et al., 1988), continued.

liquid chromatography, the following constituents were identified (Cheminat et al.,
1988): 5-O-caffeoyl quinic acid (chlorogenic acid); 3,5-O-dicaffeoquuinic acid;
4,5-O—dicaffeoyl quinic acid; 2,3-O-dicaffeoyltartaric acid (Cichoric acid); 2-O-
caffeoyl-3-O-feruloyltartaric acid; 2-O-caffeoyltartaric acid (caftaric acid); 2-O—
caffeoyl-3-O-5[a-carboxy-fl-(3,4-dihydroxyphenyl)ethyl] caffeoyltartaric acid; 2,3-
O-di 5-[q-carbow-fl(3,4-dihydroxyphenyl)ethyl] caffeoyltartaric acid; ,6-(3,4-
dihydroxyphenyI)-ethyl-O-4-O-caffeoyl-fl»D-gIucopyranoside; ,B-(3,4-
dihydroxyphenyl)-ethyl-O—a—L-rhamnopyranosyK1—>3)-4-O-caffeoyl-,B-D-
glucopyranoside (verbacoside); fl-(3,4-dihydroxyphenyI)-ethyl-O-a-L-
rhamnopyranosyl(1—>3)-,B-D-glucopyranoside (1 —+6)-4-O-caffeoyl-,6—D-
glucopyranoside (echinacoside); and ,8-(3,4-dihydroxyphenyl)-ethyI-O-a-L-
rhamnopyranosyl (1-—>3) (6-O-Caffeoyl -,6-D-glucopyranosyl) (1—>6)—4-O-caffeoyl-
fl-D-glucopyranoside (6-O-caffeoyI-echinacoside).

Solid-phase extraction coupled with reversed-phase HPLC has been used
to isolate free phenolic acids for E. purpurea, E. angustifolia, and E. pallida for
further study with regard to free radical scavenging (Facino et al., 1995) and
increased efficiency of phenolic separation (Glowniak et al., 1996). The phenolic
acid content of the dry, above ground portion of Echinacea species has been

determined to range from 70 to 1400 pg/g (Glowniak et al., 1996).

Flavonoids
Echinacea angustifolia leaves yielded the following flavonoids: Iuteolin,

kaempferol, quercetin, quercetagetin-7-glucoside; Iuteolin-7-glucoside,

23

kaempferoI-3-glucoside, quercetin-3-arabinoside, quercetin-3-galactoside,
quercetin-3-xyloside, quercetin-3-glucoside; kaempferoI-3-rutinoside; rutoside;
and isorhamnetin-3-rutinoside (Bauer and Wagner, 1991) (Figure 1.5).
Quercetin, quercetin-7-glucoside, rutin, quercetin-3-robinobioside, and quercetin-
3-xylosylgalactoside were identified in the leaves of E. purpurea (Bauer and

Wagner, 1991) (Figure 1.5).

Polyacetylenes

A series of ketoalkynes and ketoalkenes have been isolated from E.
pallida, including tetradeca-(82)-en-11,13-diyn-2-one, pentadeca-(82)-en-11,13-
diyn-2-one, pentadeca-(SZ, 13A)—dien-11-yn-2-one, pentadeca-(82,11Z,13Z)-
trien-2-one, pentadeca-(82,11E,13Z)-trien-2-one, pentadeca-(82,11Z)-dien-2-
one, pentadeca-(82)-en-2-one and heptadeca-(8Z,11Z)-dien-2-one (Bauer et al.,
1988a). Autoxidation occurred In stored materials, resulting in the formation of
hydroxylated species. Echinacea pallida was mistaken for E. angustifolia in
numerous preliminary studies (Bauer and Wagner, 1991), leading to a great deal
of confusion regarding chemical constituents and biological activities assigned to
E. angustifolia. Since E. angustifolia did not contain the above ketoalkynes and
ketoalkenes, a distinction may be made between E. pallida and E. angustifolia

based on constituent profiles.

24

R20

 

 

Compound R1 R2 R3
Iuteolin H H OH
apigenin H H H
quercetin OH H OH
kaempferol OH H H
luteolin-7-glucoside H Glc OH
kaempferol-3-glucoside GIC H H
quercetin-3-arabinoside Ara H OH
quercetin-3-galactoside Gal H OH
quercetin-3-xyloside Xyl H OH
quercetin-3-glucoside GIC H OH
kaempferol-B-rutinoside rutinose H H
isorhamnetin-3-rutinoside rutinose H OCH3
quercetin-7-glucoside OH Glc OH
kaempferol-3-rutinoside Glc H H
quercetin-3-robinobioside robinobiose H OH

 

 

 

Figure 1.5. Flavonoids from Echinacea spp. (Bauer and Wagner, 1991).

25

Analytical methods for the identification of Echinacea constituents

Echinacea species have historically been misidentified in the preparation
of extracts for scientific study and consumer use (Bauer and Wagner, 1991).
Additionally, the adulteration of Echinacea extracts by the inclusion of other plant
species has posed additional problems in the evaluation of the bioactivity Of
Echinacea extracts. To circumvent the confusion surrounding these problems, a
number of analytical methods have been developed for the quality control and
identification of chemical constituents in the roots, leaves, and flowers of
Echinacea species (Lienert et al., 1998). Thin layer chromatography (TLC) and
high performance liquid chromatography (HPLC) (Rogers et al., 1998; Bauer and
Remiger, 1989; Bauer et al., 1988a) have been predominantly used to identify
the characteristic compounds associated with Echinacea species. Gas
chromatography coupled with mass spectrometry has been used to characterize
the volatile constituents of E. angustifolia, E. purpurea, and E. pallida roots,
leaves, stems, and flowers (Mazza and Cottrell, 1999) (Table 1.1), as well as to
differentiate among the three species (Lienert et al., 1998). lsobutylamides are
recognized as one of the most biologically active groups of compounds In
Echinacea extracts (Bauer and Wagner, 1991). The rapid classification of E.
purpurea, E. angustifolia, and E. pallida, based on isobutylamide content (Bauer
and Remiger, 1989) and total phenolic content (Perry et al., 2001; Bauer et al.,
1988a), has been made possible by HPLC analysis with UV detection.

Attempts to develop methods for the standardization of Echinacea extracts

have suggested that Echinacea preparations might be characterized most

26

effectively by analysis of total phenolics (Perry et al., 2001; Bauer et al., 1988a).
Cichoric acid is the primary phenolic compounds present in E. purpurea root and
herb extracts, yet is absent from E. angustifolia (Perry et al., 2001; Bauer and
Wagner, 1991). Echinacoside is the predominant phenolic compound in E.
pallida and E. angustifolia, but is not found in E. purpurea (Perry et al., 2001;
Bauer and Wagner, 1991). Research has focused on reverse-phase high
performance liquid chromatography for the separation of phenolics in order to
characterize Echinacea spp. (Bauer et al., 19883). Marked differences in the
distribution of characteristic constituents among species have been noted (Perry
et al., 2001; Bauer et al., 1988a; Perry et al., 1997). A combination of solid-
phase extraction and reverse phase HPLC for the analysis of Echinacea spp. has
shown success for the rapid screening of phenolic acid content (Glowniak et al.,
1996).

The rapid classification of E. purpurea, E. angustifolia, and E. pallida
based on isobutylamide content (Bauer and Remiger, 1989) was examined
through TLC. Echinacea pallida root extracts contained almost no amides, while
E. purpurea and E. angustifolia extracts showed the presence of numerous
alkamides characteristic to Echinacea species (Bauer and Remiger, 1989).
Reverse phase HPLC coupled with electrospray mass spectrometry was used to
analyze the alkamides in E. purpurea roots (Sloley et al., 2001 ;He et al., 1998).

Knowing the exact chemical profile and having a standard method by
which to determine the presence of Characteristic constituents would allow

nutraceutical/phytoceuticaI manufacturers to more easily identify and compare

27

the starting materials for medicinal preparations (Sloley et al., 2001). Although
optimization of HPLC—UV methods and extraction schemes have aided in
attempts to characterize markers to distinguish between Echinacea spp.,
disparity within each species suggested large variations in the raw materials
used for Echinacea preparations (Bergeron et al., 2000). The similarities among
plant species, particularly between E. angustifolia and E. pallida, have
occasionally caused confusion regarding the proper identity of plants used for
nutraceutical preparations (Sloley et al., 2001). The chemical fingerprinting of
Echinacea extracts can mitigate which species was used in the preparation as

well as indicate the portion of the plant extracted (Perry et al., 2001).

Stability of Echinacea constituents

A serious question regarding the efficacy of nutraceuticals is whether
active components are stable enough in Echinacea preparations to convey
biological activity to the consumer. Of particular interest are alkamides and
phenolic acids (Bauer and Wagner, 1991). Alkamides are believed to be
susceptible to degradation upon storage at various temperatures (Perry et al.,
2000; Kim et al., 2000a; Rogers et al., 1998), as a result of an autoxidative
process (Perry et al., 2000).

Cichoric acid, one of the major compounds in the fresh herb of E.
purpurea, has numerous documented biological effects (N'L'Isslein et al., 2000),
yet its susceptibility to enzymatic degradation during the processing of fresh E.

purpurea plants makes the preparation of E. purpurea-based phytoceuticals a

28

challenge. Cichoric acid is rapidly degraded by polyphenol oxidase in fresh plant
juices, however might be preserved to some extent by the addition of ethanol
combined with an antioxidant such as ascorbic acid to inhibit enzyme activity
(Ntisslein et al., 2000). Drying E. purpurea flowers by freeze-drying, vacuum
microwave drying, and air-drying drastically reduced the Cichoric acid and caftaric
acid quantities (Kim et al., 2000b).

The chemical content of stored plant materials and various preparations
has been examined to determine the potential for degradation of alkamides
(Bauer and Wagner, 1991). Although many plants are air-dried, the levels of
active constituents might be altered during the process, and the procedure is
time-consuming (Kim et al., 20003). The alkamide content of E. purpurea roots
was unaffected by drying in an oven at 32-33°C when roots were broken into
large pieces or when chopped into one centimeter pieces to speed the drying
process (Perry et al., 2000). This method of drying did not reduce the alkamide
content of E. purpurea roots, indicating that oven-drying might be a suitable
alternative for more rapid processing of Echinacea root material. In a
comparison of freeze-drying, vacuum microwave drying, and air-drying, freeze-
drying was found to be the best method to preserve alkamides in E. purpurea
roots (Kim et al., 2000a).

Although not significantly altered by oven-drying at 32-33°C (Perry et al.,
1999), alkamide levels drop over time based on storage temperature (Perry et
al., 2000; Rogers et al., 1998). Samples of powdered E. angustifolia roots stored

at 25°C in a dessicator lost 13% of their alkamide contents over a two month

29

period (Rogers et al., 1998). The alkamide contents of roots, either chopped or
whole, stored at 24°C was 520% of the roots stored at —18 °C for sixteen weeks
(Perry et al., 2000). Unfortunately, the initial alkamide content of freshly dried E.
purpurea root was not assessed at the start of the study, making only relative
comparisons of alkamide content possible rather than indicating absolute loss.

Storage at -18°C for sixty-four weeks further reduced alkamide content by 60%

when compared to storage at -18°C for sixteen weeks (Perry et al., 2000).

Biological activity of Echinacea spp. extracts and constituents

Numerous biological activities have been reported for extracts and
compounds from Echinacea spp., including the stimulation of immune function,
anti-viral effects, antioxidant activity, and anti-inflammatory activity. Clinical
studies, performed predominantly in Germany, have analyzed the effects of
various Echinacea preparations used for the treatment of colds and upper
respiratory tract infections, recurrent vaginal Candida infections, and for acute
bronchitis in children (Bauer, 1998). Dosing remains a problem, since effective
doses and active constituents have not been identified. Currently, traditional
dosing Of Echinacea products is based on 6-9 mL of expressed juice of the plant,
1.5-7.5 mL of tincture, or 2-5 9 of dried root per day (Zink and Chaffin, 1998). Of
the Chemical constituents isolated from Echinacea, caffeic acid derivatives,
alkylamides, polysaccharides, and polyacetylenes are thought to be among the

most biologically active (Li and Wang, 1998).

30

Toxicity of Echinacea spp.

Echinacea has been used in various medicinal preparations for hundreds
of years (Bauer and Wagner, 1991). In general, the use of Echinacea
preparations is perceived to be safe given the lack of reported widespread
problems (Parnham, 1996). Concern for individuals who may take multiple
medications, particularly the elderly, has raised the question of the possibility Of
drug-herb interactions (Ness et al., 1999). Individuals taking
immunosuppressants, such as corticosteroids and cyclosporine, should not use
Echinacea products, since Echinacea is believed to be an effective
immunostimulant (Miller, 1998). Additionally, it has been suggested that use of
Echinacea should not exceed eight weeks due to the potential for hepatotoxicity,
and should not be used with known hepatoxic drugs such as anabolic steroids,
amiodarone, methotrexate, and ketoconazole (Miller, 1998). Echinacea is also
not recommended for individuals with autoimmune disorders and progressive
diseases, including tuberculosis, AIDS, multiple sclerosis, leukosis, collagen
disorders, or diabetes mellitus (Miller, 1998). Although generally regarded as
safe due to the sheer numbers of individuals taking Echinacea preparations and
lack of reported adverse effects (Borchers et al.,2000), studies have been
performed to examine the potential for toxicity and adverse reactions.

Echinacea extract injections have been reported to cause transient
tachychardia and influenza-like symptoms in humans (Borchers et al., 2000;
Miller, 1998; Jurcic et al., 1989). Oral administrations have yielded slightly higher

adverse reactions compared to placebo, including headache, nausea, and

31

fatigue (Schmidt et al., 1990). An increase in body temperature by 0.5 to 1.0°C
has also been reported to occur with the use of Echinacea preparations (Bauer,
1998). Echinacea angustifolia preparations have been associated with more
adverse effects than E. purpurea (Melchart et al., 1998).

Adverse reactions to the oral use of Echinacea expressed juice are
considered to be rare (Bauer, 1998), although acute allergic reactions are
possible (Lersch et al., 1992). One report of Echinacea-associated anaphylaxis
has been reported in an individual who commonly consumed Echinacea
preparations (Mullins, 1998). It should be noted that the individual also
consumed a variety of vitamins and natural product supplements with the
Echinacea, and exceeded the manufacturer’s recommended daily dose by
consuming twice the recommended amount. While testing negative in skin-prick
tests for reactivity to various allergen extracts, the individual tested positive for
the ingested Echinacea preparation. Testing of Echinacea preparations in
eighty-four patients resulted in 19% (sixteen subjects) of subjects yielding a
strong positive reaction, even though only two of the subjects had ingested
Echinacea preparations previously (Mullins, 1998). The results suggest cross-
reactivity to structurally related proteins common to Echinacea, and would
appear to indicate that even the first-time use of Echinacea preparations could
result in allergic reactions (Mullins, 1998).

The lack of any other reported cases of Echinacea-related anaphylaxis
indicates that the threshold used to indicate cross-reactivity might have been too

low in the previous study (Borchers et al., 2000; Myers and Wohlmuth, 1998). A

32

component in the Echinacea preparation, not the Echinacea juice, has been
suggested as the cause of the allergenic response (Myers and Wohlmuth, 1998).
Given the lack of reported cases of Echinacea-related allergenic response and
the number of people consuming Echinacea herbal supplements, concerns
regarding severe allergenic response are not warranted. As a result, Echinacea
is considered to be safe and well-tolerated (Borchers et al., 2000).

In vivo testing of expressed juice of E. purpurea did not demonstrate
toxicity in rats and mice following four weeks of oral administration (Mengs et al.
1991). Acute toxicity was investigated through the single dosing of Echinacea
juice, either intravenous (15 g/kg) or oral (5 g/kg), to groups of sixteen Wistar rats
and sixteen NMRI mice (eight males/eight females per group) for each dosing
method. No significant reactions occurred, and no changes to organs were
noted upon necropsy (Mengs et al., 1991)

A subacute toxicity study was designed to assess the safety of Echinacea
preparations. Eighteen male and eighteen female Wistar rats were given 0, 800,
2400, or 8000 mg/kg body weight/day of expressed E. purpurea juice for four
weeks. The two highest doses showed a significant lowering of plasma alkaline
phosphatase in males, and a rise in prothrombin time in females. No relevant
differences in weight, food consumption, necropsy, or histology results were
noted (Mengs et al., 1991).

The mutagenic potential of E. purpurea extracts was evaluated using
several Salmonella typhimurium strains and in L5178Y mouse lymphoma cells.

No evidence of mutagenicity was noted in either the bacterial or mammalian

33

mutation assay (Mengs et al., 1991). Examination of E. purpurea extracts on
human lymphocytes yielded no notable evidence of toxicity. Combined with the
above negative results for acute or subacute toxicity, E. purpurea preparations
are regarded as non-toxic (Mengs et al., 1991); however, no tests for epigenetic

toxicity were conducted.

Antimicrobial properties of Echinacea spp.

Bactericidal or bacteriostatic activities have not been demonstrated by
Echinacea extracts (Zink and Chaffin, 1998). Based upon the use of Echinacea
as a topical treatment for alleviation of various infections, sore throat, sore gums,
and infected wounds, the antifungal activity of E. purpurea was examined to
determine the effectiveness of polyacetylenes and alkyamides on fungal growth.
Tea brewed from E. purpurea and extracts prepared from E. purpurea herb were
not active; however, extracts prepared from E. purpurea roots and commercially
available tinctures, combinations of herb and roots of both E. purpurea and E.
angustifolia, demonstrated antifungal activity (Binns et al., 1991). Pure
compounds isolated from E. purpurea roots, including undeca-2E,4Z,8Z,1OE/Z-
tetraenoic acid isobutylamide, undeca-2E,4Z-diene-8,10-diynoic acid
isobutylamide, and trideca-1-ene-3,5,7,9,10-pentayne, were found to possess
antifungal activity proportional to the number of triple bonds present (Binns et al.,
1991). Evaluation of antimicrobial activity lends support to anecdotal claims for

the topical use of Echinacea preparations for certain conditions.

34

Effects of Echinacea on glucose metabolism

In recent years, research interest in the effects of herbal supplement use
on glucose metabolism in humans has increased. Potential nutraceutical
treatments include Mormordica charantia (bitter melon), Ocimum sanctum (holy
basil), Allium cepa (onion), and Allium sativum (garlic), which have demonstrated
hypoglycemic effects in individuals with diabetes (Broadhurst et al, 2000;
Srivastava et al., 1993; Rai et al., 1997; Koch and Lawson, 1996). In light of this
interest, ammonium hydroxide extracts of E. purpurea roots were investigated for
in vitro effects on insulin-dependent glucose metabolism in rat epidldymal
adipocytes. E. purpurea root extracts did not show any significant activity

(Broadhurst et al., 2000).

Potential reproductive effects of Echinacea products

Potential detrimental effects of herbal supplements on reproductive
function were assessed to determine effects on sperm DNA and the fertilization
process (Ondrizek et al, 1999). In order to simulate serum or semen
concentrations, Echinacea was tested at 0.8 mg/mL and 8 mg/mL, with the low
dose representative of one 1/1000th of the recommended daily dose of the
herbal extract. The high dose (8mg/mL) damaged sperm DNA, reduced sperm
viability, and reduced oocyte penetration. Reduction in oocyte penetration by
treatment with Echinacea was suggested to be due to an inhibition of the
hyaluronidase enzyme in the sperm head, rather than an effect on sperm motility

(Ondrizek et al., 1999). While the level of Echinacea extract and/or components

35

in semen is currently unknown, the results indicate the potential for adverse

reproductive effects.

Immune-stimulating properties of Echinacea spp.

4-O-Methyl-glucuronoarabinoxylan, isolated from E. purpurea, is one of
the polysaccharides believed to be responsible for enhancing phagocytosis
effects in an in vitro granulocyte assay (Proksch and Wagner, 1987). Three
additional polysaccharides, an acidic arabinogalactan and two
fucogalactoxyloglucans with mean molecular weights of 10,000 and 25,000, were
identified in E. purpurea cell cultures (Wagner et al., 1988). The acidic
arabinogalactan stimulated macrophages to excrete tumor necrosis factor, and
the higher molecular weight fucogalactoxyglucan enhanced phagocytosis in vitro
and in vivo (Wagner et al., 1988).

It has been reported that Echinacea whole plant extracts possess antiviral
activity (Wacker and Hilbig, 1978). The aqueous and methanol crude extracts of
E. purpurea were assayed for resistance to herpes, influenza, and vesicular
stomatitis viruses, and were found to provide 50—80% resistance to the viruses
upon incubation with the extracts (Wacker and Hilbig, 1978). Echinacoside,
caffeic acid, and Cichoric acid were examined in a vesicular stomatitis virus assay
using mouse L-929 cells (Cheminat et al., 1988). Incubation of vesicular
stomatitis virus with 62.5 ug/ml caffeic acid or 125 ug/ml of Cichoric acid for four

hours reduced the infectivity of the virus more than 50% (Cheminat et al., 1988).

36

The immunostimulating properties of acidic arabinogalactan produced by
E. purpurea cell cultures, molecular weight 75,000, were examined to determine
effectiveness in activating macrophages against Leishmania enn'ettii and tumor
cells (Luettig et al., 1989). The acidic arabinogalactan decreased L. enn’ettii
growth in intracellular and phagocytosis-associated assays. Macrophages were
induced by arabinogalactan to produce tumor necrosis factor, interleukin-1, and
interferon-(32, however B cells and T cells were unaffected.

Antigen specificity to antibodies raised against E. purpurea was
determined by Egert and Beuscher (1992). Antisera raised against E. purpurea
were tested using an enzyme-linked immunosorbent assay (ELISA) developed to
determine the presence of glycoproteins and polysaccharides. Polymer antigens
from Baptisia tincton‘a, Thuja occidentalis, lipopolysaccharide (LPS) from
Escherichia coli 055285, and LPS from Salmonella typhimun’um did not cross-

react with antibodies raised against E. purpurea (Egert and Beuscher, 1992).

Antioxidant Properties
Natural antioxidants are increasing in popularity due to their reported role
as modulators of lipid peroxidation as it relates to diseases such as
atherogenesis, thrombosis, and carcinogenesis (Frankel, 1999). Phenolic
antioxidants are known to have free radical scavenging properties and
antioxidant properties with regard to the oxidation of low-density Iipoproteins.

Echinacea spp. possess a variety of phenolic antioxidants (Bauer and Wagner,

37

1991) that are believed to be responsible for much Of the antioxidant activity of
the plant extracts (Facino et al., 1995).

The ability of caffeoyl derivatives from Echinacea species to prevent free
radical-induced degradation of collagen has been assessed (Facino et al., 1995).
Caffeic acid, chlorogenic acid, Cichoric acid, cynarine, and echinacoside were
assayed in a model system measuring collagen degradation in the presence of
xanthine oxidase, with and without iron as an oxidation catalyst. Protective
effects were noted for all species, however Cichoric acid and echinacoside were
most active (Facino et al., 1995).

The total antioxidant activity of a phenolic Echinacea extract, standardized
to 4.5% echinacoside content, was assessed at 0.5 mg/ml by the Trolox
equivalent antioxidant activity method (Pietta et al., 1998). The Trolox equivalent
antioxidant activity reflects the ability Of an antioxidant to scavenge a radical
cation ABTS+ - 2,2’-azino-bis(3-ethylbenzothiazoline-6—sulfonate). The
Echinacea extract was determined to be ineffective (Pietta et al., 1998). It should
be noted that the source of the Echinacea extracts was not clearly indicated (i.e.
root, herb, or flower). Additional studies were performed to assess the
antioxidant activity of E. purpurea, E. angustifolia, and E. pallida root and leaf
extracts in a hydroxyl free radical scavenging assay that measures the
peroxidase-catalyzed accumulation of 2,2’-azino-bis(3-ethylbenzthiazoline-6-
sulfonic acid) (Sloley et al., 2000). Root and leaf extracts Of all three species
demonstrated hydroxyl free radical scavenging activity, with E. purpurea root

extracts showing higher activity than E. angustifolia or E. pallida.

38

The ability of E. purpurea, E. angustifolia, and E. pallida root and leaf
extracts to inhibit Fe2*-induced lipid peroxidation was assessed in a
catecholaminergic neuroblastoma SH-SY5Y cells using a thiobarbituric acid
assay system (Sloley et al., 2000). The leaf extracts of all three Echinacea spp.
demonstrated higher antioxidant activity than root extracts. No significant
differences were observed between the leaf extracts of the three Echinacea
species. Chemical profiles, determined by HPLC analysis, were distinctly
different for the leaf extracts of the three species suggesting that multiple
constituents might be responsible for the antioxidant activity (Sloley et al., 2000).

Methanol extracts of Iyophilized, ground E. purpurea, E. angustifolia, and
E. pallida roots were investigated in several model systems to elucidate the
antioxidant activity of Echinacea spp. (Hu and Kitts, 2000). Reducing power,
evaluated using K3Fe(CN)6-FeCI3 and determined by comparison to ascorbic
acid, and copper chelating capacity measured by UV-vis spectrophotmetry, were
highest for E. pallida. Echinacea extracts suppressed hydroxyl radical
generation by 20-34% in a concentration dependent manner, and E. pallida was
superior to E. angustifolia and. E. purpurea in free radical scavenging activity (Hu

and Kitts, 2000).

Anti-inflammatory Properties
The anti-inflammatory properties of E. angustifolia root extracts were
evaluated in the carrageenan paw edema and croton oil dermatitis assays

(Tragini et al., 1985). lntraveneous administration of aqueous E. angustifolia root

39

extracts inhibited carrageenan-induced paw inflammation. In another test, an E.
angustifolia root extract was applied to the ears of mice simultaneously with
croton oil as an irritant. The tests revealed that E. angustifolia exhibited anti-
inflammatory activity that was superior compared to both negative and positive
(benzidamine) controls. Further examination indicated that a polysaccharide
fraction of E. angustifolia roots was responsible for the effects demonstrated in
the carrageenan paw edema and croton oil ear test (Tubaro et al., 1987). Higher
molecular weight fractions of E. angustifolia root polysaccharide extracts,
separated by tangential flow filtration, also demonstrated activity in the croton oil
ear test (Tragni et al., 1988).

Eight polyunsaturated alkamides from E. angustifolia were tested for inhibition
of cyclooxygenase and 5-lipoxygenase, two enzymes of the arachiconic acid
metabolic pathways (Muller-Jakic et al., 1994). The n-hexane extract of E.
angustifolia roots, the major component of which was a mixture of isomers,
doceca-2E,4E,82,10E/Z-tetraenoic acid isobutylamides, inhibited both
cyclooxygenase and 5-lipoxygenase. Other isobutylamides tested included
dodeca-ZE,4Z,102-triene-8-ynoic acid isobutylamide, dodeca-2E-ene-8,10-
diynoic acid isobutylamide, pentadeca-ZE,9Z-diene-12,14-diynoic acid
isobutylamide, hexadeca-2E,9Z-diene-12,14-diynoic acid isobutylamide, undeca-
2E-ene-8,10—diynoic acid isobutylamide, undeca-2Z-ene-8,10-diynoic acid
isobutylamide, and undeca-ZZ-ene-8,10-diynoic acid 2-methylbutylamide.
Pentadeca-2E,9Z-diene-12,14-diynoic acid isobutylamide and hexadeca-2E,9Z-

diene-12,14-diynoic acid isobutylamide showed little activity in the Iipoxygenase

4O

assay; however, they did inhibit cyclooxygenase activity. Another alkamide,
dodeca-2E,4Z,1OZ-triene-8-ynoic acid isobutylamide, is structurally similar to the
active alkamides, but it was active only at high concentrations to both enzymes.
The competitive inhibition of 5-lipoxygenase and cyclooxygenase enzymes by
alkamides acting as arachidonic acid analogues has been suggested (Muller-

Jakic et al., 1994).

Clinical studies denoting the effectiveness of Echinacea for treating upper
respiratory tract infections

Echinacea has been tested in a number of clinical studies to determine its
effects on relieving the severity and reducing the duration Of upper respiratory
tract infections. In one study, 160 subjects were given either a placebo or 900
mg of E. pallida root extract daily for eight to ten days. The results indicated that
treatment with Echinacea significantly reduced the duration of the respiratory
tract infection from 13 to 9.8 days (Braunig et al., 1993). A follow-up study was
designed to examine dose-response response in which 180 subjects were given
a placebo, 900 mg per day, or 450 mg per day of E. purpurea root extract.
Evaluation of cold symptoms at 3-4 and 8-10 days indicated that only the high-
dose of E. purpurea significantly reduced symptoms (Gunning, 1999; Braunlg et
aL,1992)

In another study a dried, ethanol extract of E. purpurea (95% herb and 5%
root) in tablet form, marketed as EchinaforceT”, was administered three times

daily (two tablets per dose) for eight days (Brinkeborn et al., 1998). The severity

41

of a variety of symptoms was evaluated in 119 subjects on either day one or two
and again on day eight after the onset of an upper respiratory tract infection. A
statistically significant level of improvement was demonstrated for the treatment
group when compared to the placebo group (Brinkeborn et al., 1998).

The effects of EchinagardTM (also known as EchinacinTM), a commercial
product made from the juice of E. purpurea herb, were assessed in a double-
blind randomized placebo-controlled clinical trial of 120 participants. At the first
indication of an upper respiratory tract infection, subjects ingested 20 drops every
two hours on day one, then three times per day on each subsequent day until
symptoms resolved (Hoheisel et al., 1997). According to the results, a significant
number of subjects in the placebo group, 60%, compared to the Echinagardm-
treated group, 40%, developed an upper respiratory tract infection (Hoheisel et

aI,1997)

Clinical studies failing to demonstrate the effectiveness of Echinacea for
treating upper respiratory tract infections

Many clinical trials have indicated that there are no significant effects of
Echinacea extracts on the duration or severity of colds and respiratory tract
infections (Bauer, 1998; Gunning, 1999). For example, a study involving 108
patients with a prior history of upper respiratory tract infections (defined as more
than three upper respiratory tract infections in the previous six months),
examined the effect of expressed E. purpurea juice administered at a dose of 4

mL twice daily for eight weeks. The E. purpurea juice did not significantly reduce

42

the number or duration of upper respiratory tract infections when compared to the
placebo (SchOneberger, 1992).

A three-armed, randomized, double-blind trial with 302 subjects was
designed to determine the effects of both E. purpurea and E. angustifolia root
extracts in the prevention of upper respiratory tract infections. Subjects were
dosed with 50 drops of a placebo, E. purpurea root extract, or E. angustifolia root
extract twice daily for twelve weeks. No significant differences were found
between the treatment groups with regard to the median time to the development
of an upper respiratory tract infection. Lack of power was suggested as the
reason why no significant differences were detected between treatment groups.
A larger study population was suggested for additional trials involving Echinacea
spp. (Melchart et al., 1998).

A double-blind clinical trial was designed to determine the prophylactic
effect of expressed E. purpurea juice on the incidence, duration, and severity Of
colds and respiratory infections (Grimm and Miller, 1999). One hundred eight
subjects (chosen due to a history of three or more colds/respiratory tract
infections in the previous year) were given either 4 mL of E. purpurea juice or
placebo twice daily for eight weeks. During the trial, 65% of patients treated with
Echinacea and 74% of placebo-treated subjects developed at least one cold or
respiratory tract infection; however, these results were not statistically different
from one another. In addition, no significant differences were noted between the

placebo-treated and Echinacea-treated groups with regard to the incidence,

43

duration, or severity of colds and respiratory tract infections (Grimm and Muller,
1999).

In another clinical trial, an Echinacea preparation, analyzed and found to
contain 0.16% Cichoric acid and no echinacosides or alkamides, was tested for
its ability to prevent rhinovirus colds (Turner et al., 2000). One hundred
seventeen subjects were treated three times a day for two weeks with either the
Echinacea extract at a dose Of 300 mg or with the placebo. After the two-week
treatment period, subjects were challenged with rhinovirus type 23, and the
treatments were continued for another five days. Rhinovirus infection was
detected in 44% Of the Echinacea-treated subjects and 57% of the placebo
group. In addition, clinical colds developed in 50% of the rhinovirus-infected
Echinacea-treated subjects and 59% Of the placebo group. None of the
reductions in the incidence or severity of symptoms reported in the treatment

groups were significant (Turner et al., 2000).

Mixed commercial preparations used in clinical trials

Resistanm, a commercial preparation that has been used in several
clinical trials, contains E. angustifolia herb and root in addition to extracts of
Eupeton'um, Baptiste, and Amice. In one trial, 100 subjects within two days of an
upper respiratory tract infection onset were treated with either 30 mL of a placebo
or ResistanTM on days one and two, followed by 15 mL on days three through six.
Sore throat, nasal drainage, and cough plus pharyngeal erythema were found to

be significantly improved in the group receiving the ResistanTM versus the

44

placebo (Dorn, 1989). A similar trial of ResistanTM involved 100 participants
administered the herbal preparation during the first two days of the onset of an
upper respiratory tract infection (Vorberg and Schneider, 1989). Symptom
scores recorded on days three and eight showed significant improvement among
subject receiving ResistanTM versus placebo. A 20% improvement in symptom
score was observed on day three compared to placebo that increased to 50% by
day eight (Vorberg and Schneider, 1989). A third trial of ResistanTM examined
the potential of the herbal supplement for the prevention, rather than treatment,
of upper respiratory tract infections (Schmidt et al., 1990). A group of 646
college students was administered ResistenTM or a placebo daily for eight weeks.
An assessment of symptoms including cough, sore throat, difficulty swallowing,
nasal drainage, congestion, headache, muscle aches, and fatigue was made
every two weeks. The Echinacea group demonstrated a trend of 15% lower
frequency of infection than the placebo group (Schmidt et al., 1990).

Another commercial Echinacea preparation, composed of E. purpurea and
E. pallida roots, Baptisiae tincton'ee roots, and Thujae occidentelis herb, was
examined in a double—blind placebo-controlled study involving fifteen clinical
centers. The purpose of the study was to determine if the herbal remedy would
be effective for patients in initial stages of an upper respiratory tract infection
(Henneicke-von Zepelin et al., 1999). A study group of 263 Individuals seeking
medical advice for an acute common cold were given either placebo tablets or
Esberitox® tablets (composed of 2 mg of Thujae occidentalis, 3.25 mg of

Echinacea pallida, 3.25 mg of Echinacea purpurea, and 10 mg Baptisiae

45

tincton'ae) or placebo. Subjects were dosed with three tablets for seven to nine
days following the first visit to their physician. Based upon the scoring of rhinitis,
bronchitis, and overall severity of the cold, the herbal remedy was considered to

be significantly better than the placebo (Henneicke-von Zepelin et al., 1999).

Hypothesis

Immune enhancement and inflammation, pain, and fever relief are
anecdotal claims associated with the use of E. purpurea. However, specific
components responsible for these claims have not been completely identified. It
is hypothesized that E. purpurea contains compounds that are responsible for the
reported biological activity of this plant that can be identified through bioassay-
guided fractionation. This hypothesis was examined by addressing five primary
objectives: 1) isolate and compare the COX-1 and COX-2 activities of alkamides
isolated from E. purpurea roots; 2) evaluate the cytotoxicity of E. purpurea root
alkamides based upon mosquitocidal activity; 3) isolate flavonoids and
anthocyanins from E. purpurea flowers; 4) characterize the COX-1 and COX-2
activities and antioxidant activities of the flavonoids and anthocyanins isolated
from E. purpurea flowers; and 5) characterize and examine the effects of purified
constituents from E. purpurea roots and flowers (alkamides, flavonoids, and
anthocyanins) on cytokine production. The following chapters present the
isolation and examination of the biological activities of purified constituents from

E. purpurea.

46

CHAPTER 2.
CYCLOOXYGENASE AND MOSQUITOCIDAL ALKAMIDES FROM

ECHINACEA PURPUREA (L.) MOENCH ROOTS*

ABSTRACT

Alkamides from the roots of Echinacea purpurea (L.) Moench were examined for
anti-inflammatory activity in an in vitro model system. Cyclooxygenase-1 (COX-
1) and cyclooxygenase-2 (COX-2) inhibitory activities were assessed at pH 7 for
alkamides isolated from E. purpurea roots to compare inhibitory activities
between the two cyclooxygenase isozymes. At 100 ug/mL, several E. purpurea
alkamides inhibited COX-1 and COX-2 enzymes in the range of 36-60% and 15-
46%, respectively, when compared to controls. Mosquitocidal activity was
assessed at 100 and 10 pg/mL, with 100% mortality against Aedes aegyptii L.

larvae noted for several E. purpurea alkamides at 100 ug/mL.

* Clifford, L.J.; Nair, M. G.; Rana, J.; Dewitt, D. L. Phytomedicine. 9, 249—254,
2002.

47

INTRODUCTION

Echinacea purpurea is one of three Echinacea species widely used for
phytoceutical purposes. Initially used by North American Indians to treat various
infections, the use of Echinacea spread to Europe where Echinacea-based
products have flourished in the natural product supplement market (Bauer &
Wagner, 1991). Following the establishment of Echinacea products in the
European market, interest in Echinacea spp. has increased in the North
American nutraceutical market, where the popularity of Echinacea products has
grown as a complement to modern medicine (Borchers et al., 2000). Validation
of the safety and efficacy of herbal supplements would aid in promoting
consumer confidence.

Many adverse side effects, including upper gastrointestinal irritation and
ulceration, have been reported after long-term use of COX-1 inhibitors such as
aspirin (Loll, 1996). COX-1 is the constitutive form of the enzyme responsible for
basic regulatory functions in cells and is involved in the production of
prostaglandins (Cryer and Dubois, 1998). Since prostaglandins are responsible
for such physiological processes including control of gastric secretions and
maintenance of mucosal integrity (Ford-Hutchinson, 1996), inhibition of COX-1
might lead to adverse side effects noted in long-term non-steroidal anti-
inflammatory drug (NSAID) use. COX-2, the cyclooxygenase isozyme inducible
in response to inflammation (Lipsky, 1999), is believed to be a more viable target

to inhibit inflammation without the adverse side effects noted with COX-1

48

inhibitors (Pairet et al., 1996; Vane and Botting, 1996). Compounds with
relatively greater COX-2 inhibition compared to COX-1 have demonstrated anti-
inflammatory activity with a reduction of adverse side effects when examined in
preclinical trials (Ford-Hutchinson, 1996). Specific or greater inhibition of COX-2
as opposed to COX-1 is a desirable target in developing pharmaceutical and
phytoceutical products due to the potential reduction in adverse side effects.

Numerous bioactive compounds have been isolated from Echinacea spp.
Extracts and compounds from various parts of Echinacea spp. plants have
demonstrated anti-viral, antioxidant, and anti-inflammatory activities (Facino et
al., 1995; Bauer & Wagner, 1991; Wagner et al., 1988; Luettig et al., 1989). Of
particular interest are the alkamides isolated from E. purpurea and E.
angustifolia, some of which have been examined for in vitro anti-inflammatory
activity. Inhibition of 5-lipoxygenase and cyclooxygenase enzymes isolated from
ram seminal vesicles by alkamides present in E. angustifolia and E. purpurea has
previously been reported (MUIIer-Jakic et al., 1994). Two cyclooxygenase
isoforms have been identified, COX-1 and COX-2; however, a comparative
assessment of the COX-1 versus the COX-2 inhibitory properties of the
alkamides has not been examined. In this paper we report the comparative
inhibition exhibited by E. purpurea alkamides against COX-1 and COX-2.

In addition to anti-inflammatory properties, alkamides from a variety of
plants have been reported to possess significant mosquitocidal activity. The
mosquitocidal activity of dodeca-2E,4E,8E,1OZ-tetraenoic acid isobutylamide, an

alkamide isolated from Spilenthes meuritiene (Jondiko, 1986), is of particular

49

interest due to its close structural similarity to an alkamide identified in E.
purpurea. The mosquitocidal activities of the majority of alkamides from E.

purpurea have not been determined, and are presented here.

MATERIALS AND METHODS
General Experimental

Preparative HPLC was performed on a Model LC-20 Preparative
Recycling Liquid Chromatograph (Japan Analytical Industry Co., Ltd., Tokoyo,
Japan) with two JAIGEL—C13 columns (10 pm, 20 mm x 250 mm) in tandem and
UV detection at 260 nm. A gradient of methanolzwater, 50-80% methanol over 2
h, 2 mL/min, was used to separate the alkamides into crude fractions. Final
purification was achieved using a gradient of acetonitrilezwater, with 40-60%
acetonitrile over 2 h, 2 mL/min. 1H NMR spectra were recorded in CDCI3 on a
Varian Inova 300 MHz spectrometer. Mass spectra were recorded using a
Waters Alliance HT LC/MS system (Milford, Massachusetts) with a 2690
Separations Module and Micromass detector, ionization mode AP+ and cone
voltage 40eV, with a mobile phase of 40-80% acetonitrile over 30 min.
Identification of E. purpurea alkamides was made by 1H NMR experiments and

by comparison to literature (Bauer et al., 1988) and LC/MS values.
Plant Material

Echinacea purpurea roots were obtained from Trout Lake Farm, (Trout

Lake, Washington). Dried, milled E. purpurea roots (989.9 g) were exhaustively

50

extracted with dichloromethane (4 L x 3, 3 days) and concentrated in vacuo.
Following concentration, the dichloromethane extract (10.1 g) was separated into
methanol-soluble (7.1 g) and insoluble (3.0 9) fractions with the addition of
methanol (750 mL x 3). The methanol-soluble portion was concentrated in
vacuo. dissolved in chloroform (250 mL), and hexane was added. The
supernatant, containing the crude alkamides, was concentrated (4.2 g) and
subjected to separation by preparative high-performance liquid chromatography

(HPLC)

Cyclooxygenase Inhibitory Assay

An in vitro COX-inhibition model system was used to assess the anti-
inflammatory activity of the alkamides isolated from E. purpurea. The COX-1
used in the assay was Obtained from ram seminal vesicles (Oxford Biomedical
Research, Inc., Oxford, MI). COX-1 was obtained from microsomel preparations
of ram seminal vesicles according to previously reported methods (Laneuville et
al., 1994; Meade et al., 1993). A recombinant human microsomel preparation Of
COX-2 was provided by Dr. David Dewitt (Michigan State University, East
Lansing, MI). The assay was conducted in a 600 uL Instech Chamber (Instech
Laboratory, Plymouth Meeting, PA) maintained at 37 °C, containing a reaction
buffer of 0.1 M Tris, 1 mM phenol, 100 pM arachidonic acid, and 17 pg
hemoglobin. Alkamides were dissolved in DMSO such that a 10 pL aliquot would
yield a final assay concentration of 100 pg/mL. The reaction was initiated with

either a 10 pL aliquot of COX-1 or a 20 uL aliquot of COX-2. Oxygen uptake was

51

monitored using a YSI model 5300 biological oxygen monitor (Yellow Springs
Instruments, Inc., Yellow Springs, OH) and recorded using Quicklog for
Windows, version 1.0 (Strawberry Tree, Inc., Sunnyvale, CA).

Aspirin, ibuprofen, naproxen, celecoxib (CelebrexTM), and rofecoxib
(VioxxTM) were dissolved in DMSO and used as controls for both the COX-1 and
COX-2 inhibitory essays. Aspirin was tested at 1000 uM (180 jig/mL), ibuprofen
at 10 uM (2.06 ug/mL), and neproxen at 10 uM (2.52 pg/mL) for both COX-1 and
COX-2. Celecoxib (CelebrexN) and rofecoxib (VioxxTM) were obtained as
physican’s professional samples (Dr. Subash Gupta, Sparrow Pain Center,
Sparrow Hospitals, Michigan), ground to fine powder, dissolved in DMSO, and

tested at 1.6 pg/mL for both COX-1 and COX-2.

Mosquitocidal Assay

The mosquitocidal assay was performed according to previously published
methods (Kelm et al., 1998; Nair et al., 1989). Aedes eegyptii L., provided by Dr.
Alan Hayes (Department of Entomology, Michigan State University) were
hatched and raised in 500 mL of degassed distilled water, with approximately 5
mg of bovine liver powder added for nourishment. After the mosquito larvae had
reached the fourth-instar (four days), the larvae were prepared for the bioassay.
Ten to fifteen larvae in 980 uL of degassed, distilled water were placed in four mL
culture tubes. Stock solutions of the alkamides in DMSO were prepared such

that a 20 uL aliquot added to the mosquito larvae in the bioassay culture tubes

yielded a final assay concentration of 100 ug/mL or 10 pg/mL. Alkamide assay

concentrations in uM are noted in Appendix A. A 20 uL aliquot Of DMSO was
used as the control. The alkamides were tested at 100 ug/mL and 10 ug/mL and

were tested in triplicate with the control set. Dead larvae were recorded at time

zero, 1, 2, 4, 9, and 24 h and reported in terms of percent mortality.

RESULTS

The bioactive alkamides isolated from the dried E. purpurea roots (Figure
2.1) were identified as undeca-2E,4Z—dien-8,10-diynoic acid isobutylamide (1),
undeca-2Z,4E-dien-8,10-diynoic acid isobutylamide (2), dodeca-2E,4Z—dien-8,10-
diynoic acid isobutylamide (3), undeca-2E,4Z-dien-8,10-diynoic acid 2-
methylbutylamide (4), dodeca-2E,4Z-dien-8,10-diynoic acid 2-methylbutylamide
(5), and a mixture of dodeca—2E,4E,8Z,1OE-tetraenoic acid lsobutylamide and
dodeca-2E,4Z,8Z,1OZ-tetraenoic acid isobutylamide (6I7). Compounds 6 and 7
were inseparable under various HPLC conditions. Yields Obtained were 8 mg
(1), 14 mg (2), 21 mg (3), 12 mg (4), 15 mg (5), and 40 mg (6”).

The results of the mosquitocidal assay are summarized in Figure 2.2.
The mixture of alkamides 6/7 proved to be the most effective in the mosquitocidal
assay, with 87.5% mortality of mosquito larvae within fifteen minutes when

assayed at a concentration of 100 pg/mL. Lowering the concentration to 10

ug/mL still yielded significant mosquitocidal activity, with 63% mortality in one
hour (Figure 2.2). Mosquito larvae that were still regarded as alive showed

significant impairment at one hour and beyond for alkamides 6”. Compound 1

53

O
I

 

 

 

| cu.
C—NH—CHz-CH
_ _ H\ CH3
H_C=C C—C CHZ—CHZ /C=C
C=C H
\
H H
1
H—CEC C:C CHz-CHZ H O
c—c/ H /CH3
— C—NH—CHz-CH
H/ \ _ / \CH3
c—C
/
H H
2
0
H /CH3
H C—NH—CH2--CH
\ / CH
H3C-C:C C::C CHz-CHz C=C\ 3
C———C H
\
H H
3

Figure 2.1. Alkamides isolated from E. purpurea roots: undeca-2E,4Z-dien-
8,10-diynoic acid isobutylamide (1), undeca-22,4E-dien-8,10-diynoic acid
isobutylamide (2), dodece-2E,4Z-dien-8,10-diynoic acid isobutylamide (3),
undeca-2E,4Z-dien-8,10-diynoic acid 2-methylbutylamide (4), dodeca-2E,4Z-
dien-8,10-diynoic acid 2-methylbutylemide (5), and a mixture Of dodeca-
2E,4E,82,10E-tetraenoic acid isobutylamide and dodeca-2E,4E,8Z,1OZ-

tetraenoic acid isobutylamide (6I7).

54

|| /CH2—CH3
H C—NH—CHz—CH
\ _ \
H—CEC—CE —CH2-CH3 C—c CH3
\
\
H H
4
O
| /CH2—CH3
H C—NH—CHz—CH
\ _
H,C-—c_——£c—CEc—CHz—CH2 /C—C\ CH3
\
H H
5

Figure 2.1 (cont’d). Alkamides isolated from E. purpurea roots: undeca-2E,4Z—
dien-8,10-diynoic acid isobutylamide (1), undeca-2Z,4E—dien-8,10-diynoic acid
isobutylamide (2), dodeca-2E,4Z-dien-8,10-diynoic acid isobutylamide (3),
undeca-2E,4Z-dien-8,10-diynoic acid 2-methylbutylamide (4), dodeca-2E,4Z-
dien-8,10-diynoic acid 2-methylbutylamide (5), and a mixture of dodeca-
2E,4E,8Z,10E-tetraenoic acid isobutylamide and dodeca-2E,4E,82,102-

tetraenoic acid isobutylamide (6/7).

55

/
C“. .__C CHz-CHz H
H/ \CZC/ C=c/ H
/ \ / /
H H H \CZC O
/ \LI /CH3
H —NH—CH2-CH
\CH3
6
H\ H
H3C CZC
\C_C/
—— CHz-CHz H
H/ \H \CZC/ H
/ /
H \:C O
/ \fl /CH3
H -—-NH—CH2-CH
\CH3
7

Figure 2.1 (cont’d). Alkamides isolated from E. purpurea roots: undeca-2E,4Z-
dien-8,10-diynoic acid isobutylamide (1), undeca-22,4E-dien-8,10-diynoic acid
isobutylamide (2), dodeca-2E,4Z~dien-8,10-diynoic acid isobutylamide (3),
undeca-2E,4Z—dien-8,10-diynoic acid 2-methylbutylamide (4), dodeca-ZE,4Z-
dien-8,10-diynoic acid 2-methylbutylamlde (5), and a mixture of dodeca-
2E,4E,8Z,10E-tetraenoic acid isobutylamide and dodeca-2E,4E,BZ,1OZ-

tetraenoic acid isobutylamide (617).

56

 

Figure 2.2. Mosquitocidal activity of E. purpurea alkamides. Alkamides 1-5

were assayed at 100 Hg/mL and the mixture of 6/7 was assayed at 10 pg/mL.

DMSO was used as the solvent control.

57

% Mortality

demonstrated slightly less activity with 71 and 100% mortality by 2 and 9 h,
respectively. Compounds 2 and 3 showed mosquitocidal activity at the end of 9
h, with 78 and 50% mortality, respectively. Alkamides 4 and 5 proved to be the
least active, with only 25 and 10% mortality at the end of 24 h, respectively.
COX-1 and COX-2 inhibitory activities of the alkamides are summarized in
Figure 2.3. COX-1 inhibitory activity was highest for alkamides 2, 4, and 5, with
inhibitions of 60, 55, and 48%, respectively. Alkamides 1 and 3 both exhibited
COX-1 inhibition of 36% when compared to DMSO. The mixture of alkamides
6”, however, did not demonstrate inhibition of the COX-1 enzyme (Figure 2.3).
The alkamides showed lower COX-2 inhibitory activity when compared to
COX-1. Compound 2 possessed the strongest COX-2 inhibitory activity at 46%.
Compounds 4 and 5 showed 39 and 31% inhibition of the COX-2 enzyme,
respectively. Compounds 1 and 3 had the lowest activities of the alkamides, with
each exhibiting 15% inhibition. As with COX-1, the mixture of alkamides 6I7 did

not demonstrate activity against COX-2.

Discussion

Numerous plant species in the Asteraceae possess insecticidal alkamides
(Greger, 1984), and the mosquitocidal properties of E. angustifolia have
previously been reported (Hartzell, 1947; Greger, 1984). Dodeca-ZE,4E,82,10E-
tetraenoic acid isobutylamide, an alkamide isolated from S. meun'tiana (Jondiko,

1986) similar in structure to 6” identified in E. purpurea (Bauer et al., 1988) and

58

 

 

lICOX:I1

LQQXEI

 

% Inhibition

 

 

 

 

 

 

 

 

 

 

Figure 2.3. Comparative inhibition of COX-1 and COX-2 enzymes by E. purpurea
alkamides. Alkamides were assayed at 100 pg/mL and compared to the solvent
control, DMSO. Aspirin, ibuprofen, naproxen, celecoxib (CelebrexTM), and rofecoxib
(Vioxx‘m) were used as controls. Aspirin was tested at 1000 HM (180 ug/mL),
ibuprofen at 10 pM (2.06 Hg/mL), and naproxen at 10 pM (2.52 pglmL) for both
COX-1 and COX-2. Celecoxib (CelebrexTM) and rofecoxib (VioxxTM) were tested at
1.6 pg/mL for both COX-1 and COX-2. Vertical bars represent the standard

deviation of each data point (n=2).

59

E. angustifolia (Bauer et al., 1991), showed 100% mortality at 24 h against A.
aegypti at 10'5 mg/mL (Jondiko, 1986). In our study, the high activity noted with
alkamides 6 and 7 from E. purpurea supports the mosquitocidal activity indicated
from E. angustifolia.

Interestingly, both 2-methylbutylamides, 4 and 5, demonstrated the lowest
mosquitocidal activity in the assay when compared to the isobutylamides tested.
The presence of an E/Z configuration of the double bonds is regarded as
important for conveying the highest degree of insecticidal and thus biological
activity (Greger, 1984). From the results of the mosquitocidal assay, it appears
that the 2-methyl versus an isobutyl functionality might also play a role in the
effectiveness of alkamides as mosquitocidal compounds.

Previously, eight alkamides from E. angustifolia DC. and ten alkamides
from various Achillee species were examined for cyclooxygenase and 5-
lipoxygenase activities (MUIIer—Jakic et al., 1994). A mixture of 6/7, isolated from
E. angustifolia roots, demonstrated inhibition of 54.7% of COX from microsomel
sheep seminal vesicle preparations at a concentration of 50pg/ml. The mixture
of 6/7 isolated from E. purpurea roots in our study did not show activity at 100
pg/mL for either COX-1 or COX-2. Alkamides 1-5 from E. purpurea roots have
not been examined for cyclooxygenase inhibitory activity; however, undeca-
2E,4E-diene-8,10-diynoic acid isobutylamide, isolated from Achillee millefolium
L., was shown to inhibit COX-1 enzyme at 40% (MUIIer-Jakic et al., 1994) while
compounds 1 and 2 exhibited 36 and 60% COX-1 enzyme inhibition,

respectively.

60

Compound 2 exhibited the highest inhibition against both COX-1 and
COX-2 enzymes. The 2-methylbutylamides, compounds 4 and 5, demonstrated
higher COX-1 and COX-2 inhibitory activity when compared to the majority of the
isobutylamides tested, with the exception of compound 2. Compounds 1 and 3,
differing in one methyl group, demonstrated equal inhibitory activities for both
COX-1 and COX-2 enzymes, indicating that the inhibition of COX-1 and COX-2
enzymes is unaffected by the addition of a single methyl group.

The E. purpurea alkamides examined in this study did not show selective
inhibition of COX-2 enzyme. Additional work in identifying the structure activity
relationships of E. purpurea alkamides and alkamides of similar structure
differing in E/Z configurations would aid in determining the potency of these

alkamides for specific phytoceutical applications.

61

CHAPTER 3.
THE ISOLATION AND PURIFICATION OF BIOACTIVE COMPOUNDS FROM

ECHINACEA PURPUREA (L.) MOENCH FLOWERS

ABSTRACT

Echinacea purpurea (L.) Moench is one of three Echinacea species commonly
used for phytoceutical purposes. While the biological activity of Echinacea
angustifolia, Echinacea pallida, and E. purpurea roots have been examined, the
flowers have not been studied to the same extent. In order to identify bioactive
constituents from this underutilized part of E. purpurea, several compounds were
isolated from Iyophilized E. purpurea flowers to later determine their activity
through various in vitro studies. Three flavonoids, quercetin-3-O-glucoside,
quercetin-3-O-rutinoside, and kaempferol-30-robinobioside, were purified from
methanolic extracts of Iyophilized E. purpurea flowers. In addition, two
anthocyanins, cyanidin-3-O-B-D glucopyranoside and cyanidin-3-O—malonyl-
(1->6)-B-D glucopyranoside, were obtained from a methanol extract of Iyophilized
E. purpurea flowers. The isolation, purification, and structure elucidation of these

compounds are presented.

62

INTRODUCTION

Echinacea purpurea is widely used as a natural product supplement.
Initially used by North American Indians to treat various infections, the popularity
of Echinacea spread to Europe, where Echinacea products have flourished in the
nutraceutical market (Bauer & Wagner, 1991). Interest in Echinacea spp. has
recently increased in the North American phytoceutical market, and Echinacea
products are currently used as a complement to modern medicine (Borchers et
al., 2000). A resurgence in the popularity of Echinacea in the United States
during the late twentieth century has made it one of the top-selling herbal
supplements (Gunning, 1999).

Traditionally, Echinacea spp. were used both externally and internally to
treat ailments including burns, insect bites, wounds, aches, chills, rheumatism,
arthritis, mumps, and measles (Bauer and Wagner, 1991; Kindscher, 1989).
Echinacea roots were used most often; however, juice from whole Echinacea
plants and infusions brewed from the fresh herb were occasionally consumed as
a medicinal remedy. Currently, Echinacea preparations are sold in the form of
tinctures, tablets, capsules, juices, and teas from the roots or fresh herb of the
plant (Gunning, 1999; Li and Wang, 1998). Extracts prepared solely from the
flowers Of Echinacea are not currently commercially available in the United
States.

Numerous bioactive compounds have been isolated from Echinacea spp.

Caffeic acid derivatives, flavonoids, alkaloids, alkamides, polyacetylenes, and

63

anthocyanins have been obtained from various parts of Echinacea spp. plants,
with the majority isolated from the roots and leaves (Bauer and Wagner, 1991).
Occasionally, the whole Echinacea plant, referred to as the fresh herb, is used to
prepare extracts for biological testing or isolation; however, the flower is included
only if the plant is in bloom. Very few studies have examined the bioactive
compounds present in the flowers of E. purpurea (Cheminat et al., 1989). The
vast majority of the research performed has been on the roots, stems, and leaves
of Echinacea (Bauer and Wagner, 1991).

Extracts and compounds from various portions of Echinacea plants have
demonstrated anti-viral, antioxidant, and anti-inflammatory activities (Facino et
al., 1995; Bauer & Wagner, 1991; Wagner et al., 1988; Luettig et al., 1989).
Echinacea root and herb extracts have been tested in clinical trials to determine
the effects on relieving the severity or duration of upper respiratory tract
infections (Gunning, 1999; Brinkeborn et al., 1998; Hoheisel et al., 1997; Braunig
et al., 1992). The anti-viral, antioxidant, and anti-inflammatory properties of
extracts and compounds isolated from the flowers of E. purpurea are unknown.
Identification of the bioactive constituents from E. purpurea flowers would provide
useful information with regard to herbal supplement preparation.

The purpose of this research was to isolate and identify constituents from
E. purpurea flowers for subsequent analysis to determine the bioactive properties
of compounds from E. purpurea flowers. The extraction, isolation, and structure
elucidation of three flavonoids and two anthocyanins from E. purpurea flowers

are presented here.

64

MATERIALS AND METHODS
General Experimental

Amberlite XAD-16 resin was Obtained from Supelco (Bellefonte, PA). LC-
SORB SP-A-ODS gel (particle size 25-40 pm) was purchased from Dychrom
(Santa Clara, CA). Silica gel PTLC plates (20 x 20 cm, 1000 um) were obtained
from Analytech Inc. (Newark, DE). Preparative HPLC for flavonoids was
performed on a Model LC-20 Preparative Recycling Liquid Chromatograph
(Japan Analytical Industry Co., Ltd., Tokoyo, Japan) with a JAIGEL-C18 column
(10 pm, 20 mm x 250 mm) and UV detection at 210 nm. Analytical HPLC
analysis of the anthocyanin fractions was performed using Millennium 2010
Chromatography Manager version 3.05.01 (Waters Corp., Milford, MA) with
Capcellpak C13 column (4.6 x 250 mm, 5 pm) (Dychrom, Sunnywale, CA) and a
mobile phase Of acetonitrile: H20 (85:15) with 4% H3PO4 in the aqueous phase.
Anthocyanins were detected at A 520 nm using a Waters PDA detector (Waters
Corp., Milford, MA). Fractions containing anthocyanins were concentrated in
vacuo and passed over XAD-16 to remove the phosphoric acid. Fractions were
applied to the XAD-16 column, washed with H20 until a neutral pH was obtained,
and eluted with 100% MeOH. Following elution, each fraction from the XAD
column was reduced in vacuo and the anthocyanins were stabilized with several
drops of MeOH spiked with HCI, yielding a pH of approximately 2.

HPLC-ES/MS of anthocyanins were recorded using a MicroMass Quattro

ll LC-MS/MS system (Micromass, Waters Corp., Beverly, MA) equipped with a

65

Waters 2090 HPLC pump and Waters 996 PDA detector (Waters Corp., Milford,
MA) with an HP ODS Hypersil HPLC column (4.0 x 125 mm, 5 pm) (Agilent
Technologies, Wilmington, DE). Data handling was performed using MassLynx
version 3.4 software (Micromass, Waters Corp., Beverly, MA). MS parameters
were set to ionization mode AP+ and cone voltage 20 eV. Conditions of analysis
were as follows: injection volume 10 pL; column temperature 30 °C; PDA range
200-799 nm, with 520 nm as the detection wavelength. Mobile phase conditions
were as follows: Solvent A) 0.1% TFA/H20 (v/v), B) 50.4% H20/48.5°/o ACN/1.0%
CH3COOH/0.1% TFA (v/v/v/v); gradient % B: initial: 20%, 26 min: 60%, 30 min:
20%, 35 min: 20%; run time 35 min; flow rate 0.80 mL/min. 1H and 13C NMR
spectra of the flavonoids and anthocyanins were recorded in DMSO-d5 and
CD300 with 0.1 mL of DCI, respectively, on a Varian Inova 300 MHz
spectrometer (Palo Alto, CA). The structures of the flavonoids were determined
and compared to the 1H NMR and 13C data of published literature values
(Markham et al., 1982). Anthocyanins were identified based on comparison of 1H
NMR and 13C data to literature values (Fossen and Andersen, 1998; Agrawal and
Mahesh, 1989), and by MS and comparison to standard cyanidin-3-O—8-D
glucopyranoside retention times using analytical HPLC.
Plant Material and Extraction

Echinacea purpurea (L.) Moench plants were Obtained from Dr. Erik
Runkle (Michigan State University, East Lansing, MI) and maintained in the
Michigan State University Bioactive Natural Product and Phytoceuticals

Laboratory greenhouses. Echinacea purpurea flowers were harvested over a

66

period of eight months from approximately fifty, two-year-old plants. Extraction Of
E. purpurea flowers was carried out through the scheme in Figure 3.1. Fresh E.
purpurea flowers (827 g) were Iyophilized, ground, and extracted exhaustively
with MeOH (4 L x 3). The crude MeOH extract (193 g) was concentrated in
vacuo, dissolved in MeOH-H2O (85:15), and partitioned with hexane (1 L x 3).
The MeOH- H2O soluble portion (137 g) was subjected to XAD-16 resin column
chromatography. The column was sequentially eluted with H20 (4 L), 50%
MeOH (3 L), and 100% MeOH (3 L). The 50% MeOH eluate afforded 23 g of
rust-colored gum from which the anthocyanins were isolated. The 100% MeOH
eluate yielded 10 g of brown extract, from which the flavonoids were obtained.
Purification of anthocyanins

Purification of the anthocyanins from the 50% MeOH XAD eluate (23 g)
was achieved by repeated gradient MPLC with 10-50% acetonitrile (4% H3PO4 in
the aqueous phase). Ten grams of the 50% MeOH XAD eluate were applied to a
C13 MPLC column, and eluted with acetonitrile: H20 (10:90 to 50:50) with 4%
H3P04 in the aqueous phase under gradient conditions. Ten fractions (fractions
A1-A10) of 200-450 mL were collected and analyzed for anthocyanins by
analytical HPLC.

Anthocyanin-containing fractions A7 and A8 were combined (1.57 g) and
subjected to C13 MPLC with acetonitrile: H20 (10:90 to 20:80) with 4% H3P04 in
the aqueous phase under gradient conditions. One hundred twenty, 12 mL
fractions were collected, and every fifth fraction was screened for anthocyanins

by analytical HPLC. Fractions containing anthocyanins were combined,

67

 

Lyophilized E. purpurea flowers

 

 

 

 

 

Crude methanol extract

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I I
50% MeOH XAD Eluate 100% MeOH XAD Eluate
I l
C13 MPLC C13 MPLC
Of 50% eluate Of 100% eluate
I I
C13 MPLC Of C13 MPLC Of
fractions A7-A8 fractions E6-E7
I I
C13 MPLC of C18 MPLC of
fraction B6 fraction F2
I I
- C13 MPLC of Fractions G1-3

 

 

fraction C5-C7

 

 

 

 

 

 

I I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I C13 MPLC C13 MPLC C18 MPLC
Compound 8 l I l
PTLC Prep. HPLC PTLC
C d 9 I I I
ompoun Prep. HPLC Prep. HPLC Prep. HPLC

 

 

 

 

 

 

 

 

 

I I

 

 

 

Compound 10 Compound 12

 

 

 

 

 

 

 

 

Compound 11

 

 

 

Figure 3.1. Extraction scheme for anthocyanins and flavonoids isolated from E.

purpurea flowers.

68

concentrated, and passed over XAD to remove the phosphoric acid, yielding
seven fractions (B1-B7).

Fraction B6 was further subjected to C18 MPLC under the previously
described conditions. Eighty, 12 mL fractions were collected and analyzed by
analytical HPLC. Fractions with similar profiles were pooled yielding eight
fractions (C1-C8). Further 013 MPLC of fractions C5-C7 (135 mg) was carried
out with acetonitrile: H2O (10:90 to 15:85), with 4% H3P04 in the aqueous phase,
under gradient conditions. Ninety, 12 mL fractions were collected and analyzed
by analytical HPLC, yielding eight fractions (D1-D8). Compounds 8 (13 mg) and
9 (8 mg) were obtained as red, amorphous solids.

Purification of flavonoids

The 100% MeOH XAD eluate (10 g) was subjected to C18 MPLC with a
gradient of 40-100% MeOH with 0.02% TFA. Ten, 200 mL fractions were
collected (E1-E10) and concentrated in vacuo. Fractions E6 and E7 were
combined (5.6 g) and subjected to further C18 MPLC with 40-70% MeOH. One
hundred, 12 mL fractions were collected and every fifth fraction was analyzed by
TLC. Fractions with similar flavonoid profiles were combined, yielding fractions
F1-F3. MPLC on fraction F2 (1.79 g) was performed with a mobile phase of 10-
50% acetonitrile, which yielded eighty, 12 mL fractions. Fractions were analyzed
by TLC and combined, yielding fractions G1-G4.

Fraction G1 (600 mg) was separated by C13 MPLC with a mobile phase of
20-75% acetonitrile. Sixty, 12 mL fractions were collected, analyzed by TLC, and

combined into six fractions (H1-H6). Fraction H1 (173 mg) was subjected to

69

PTLC with dichloromethane:MeOHztoluenezformic acid (12:4:0.5:0.5), yielding
two bands. Band 12 (23 mg) was further purified by repeated C18 preparative
HPLC with a mobile phase of 30% acetonitrile, yielding compound 10 (12 mg) as
yellow crystals when recrystallized from methanol.

Fraction G2 (538 mg) was subjected to further C18 MPLC using a mobile
phase gradient of 10-40% acetonitrile. Fifty, 12 mL fractions were collected,
concentrated, and combined yielding four fractions, J1-J4, based on TLC profiles.
Fraction J2 (235 mg) was further purified by repeated Cm preparative HPLC with
a mobile phase of 20% acetonitrile. Two fractions were collected (K1-K2), and
fraction K2 (29 mg) was further purified by repeated preparative HPLC with a
mobile phase of 30% acetonitrile, yielding compound 11 (15 mg) as yellow
crystals after crystallization from MeOH.

Fraction G3 (600 mg) was subjected to C13 MPLC with 20-50% acetonitrile
under gradient conditions. One hundred twenty, 12 mL fractions were collected,
analyzed by TLC, and combined into six fractions (L1-L6). Silica PTLC on
fraction L2 (56 mg) with dichloromethane:MeOHztoluenezformic acid
(12:4:0.5:0.5) yielded three bands (M1-M3). Purification of fraction M2 (7 mg) by
repeated preparative HPLC with a mobile phase of 30% acetonitrile yielded

compound 12 (2.6 mg) as a yellow, amorphous solid.

70

RESULTS AND DISCUSSION

Purification of anthocyanins

Compound 8 was identified as cyanidin-3-O-[3-D glucopyranoside (8)
based on 1H and 130 chemical shifts (Fossen and Andersen, 1998) (Table 3.1),
and by comparative analytical HPLC and HPLC/MS values. Compound 9 was
identified as cyanidin-3-O—malonyI-(1—>6)-8-D glucopyranoside (9) based on 1H
and 13’0 chemical shifts (Fossen and Andersen, 1998) (Table 3.1) and by
HPLC/MS values. The structures of compounds 8 and 9 are shown in Figure
3.2.

The ovenrvhelming presence of compound 8 in the crude MeOH extract
from XAD and initial MPLC fractions A1-A10 suggested that compound 8 was the
predominant anthocyanin present in E. purpurea flowers. Hydrolysis of the 50%
MeOH extract with concentrated HCI for 16 h at room temperature yielded only
the one aglycon, cyanidin, based on analytical HPLC analysis and comparison to
a cyanidin standard. Cyanidin has been reported to be the predominant aglycon
in purple flowers (Saito and Harborne, 1992), and anthocyanin 9 has been
reported in a variety of blue and purple flowers (Akashi et al., 1997; Saito and
Harborne, 1992; Kim et al., 1989; Bridle et al., 1984). The deep purple-red color
of compound 9 likely contributes significantly to the pink-purple hue of E.
purpurea petals.

While only 8 and 9 were obtained pure, additional anthocyanins were

present in anthocyanin-enriched MPLC fractions C1-C4 and in B1-B3. Isolation

71

Table 3.1. 130 Chemical shifts for anthocyanins from E. purpurea.

 

 

 

Compounds
Carbon No. 8 9

2 164.3 164.5
3 145.6 145.7
4 137.0 167.1
5 159.1 159.0
6 103.4 103.4
7 170.4 170.1
8 95.2 95.2

9 157.7 157.5
10 113.4 115.4
1’ 121.3 121.3
2' 118.4 118.9
3' 147.3 147.4
4' 155.7 155.7
5’ 117.4 115.5
6’ 128.3 130.9
1" 103.7 103.6
2” 74.8 74.8

3" 78.0 78.7

4” 71.1 71.1

5" 78.7 75.3
6" 62.3 65.1

6” malonyl

1’” 167.9
2'" 41.6
3'” 169.2

 

72

0H
0H
+
H. o 0
O \ ..
HO
/ ems
OH

OH

cyanidin
3-0-8-D-glucopyranoside

OH
OH
+
I... o O
\
O ..
/ HO
O O OH
OH
0W0”
cyanidin 0 0

3-O-malonyI-(1—>6)-B-D-
glucopyranoside

Figure 3.2. Anthocyanins isolated from E. purpurea flowers: cyanidin-3—O—fl-D-

glucopyranoside (8) and cyanidin-3-O-malonyl-(1—>6) -,8-D-glucopyranoside (9).

73

of these extremely polar compounds, present in very low amounts, was not
successful by MPLC or preparative HPLC. LCMS failed to identify the mass of
these anthocyanins due to insufficient quantities. Previous attempts to isolate
anthocyanins from E. purpurea and E. pallida resulted in the isolation of
compounds 8 and 9 in addition to two unidentified acylated anthocyanins
(Cheminat et al, 1989). Acylation has been reported to increase retention times
by two-fold when compared to non-acylated anthocyanins (Kim et al., 1989).
During analytical HPLC analysis, compound 9 eluted significantly later than
compound 8. Based on the HPLC retention time of compound 9 relative to
compound 8, it is unlikely that the unidentified anthocyanins in fractions 01-04
and B1-B3, which eluted prior to compound 8, were the same acylated
anthocyanins previously suggested in E. purpurea (Cheminat et al., 1989).

There are conflicting reports regarding the stability of anthocyanin 9.
Recently, it was reported that compound 9 was stable (Luczkiewcz and Cisowski,
2001); however, the work performed here confirms earlier reports indicating that
compound 9 is highly unstable (Kim et al., 1989). Anthocyanins are generally
stabilized by the addition of acid to prevent the break-down Of the flavylium cation
to the colorless quinoidal base. The persistent presence of compound 8 in
fractions initially containing only anthocyanin 9 confirmed previous findings of the
unstable nature of compound 9. Loss of the malonyl group occurred rapidly both
upon storage and during purification by MPLC. The stability of compound 9 in
phytoceutical preparations of E. purpurea flowers should be further examined if

compound 9 is a desired component of the product.

74

Purification of flavonoids

Compounds 10-12 were identified based on 1H and 13C chemical shifts
(Table 3.2) and in agreement with published data (Markham et al. 1982) as
quercetin-3-O-rutinoside (10), kaempferoI-3-O-robinobioside (11), and quercetin-
3-O-glucoside (12). The structures Of these flavonoids are presented in Figure
3.3.

Numerous flavonoid-type compounds were noted in the MPLC column
fractions (G1-G3) and preparative HPLC isolation (I1, K1, and M3) of compounds
10-12. Attempts to isolate the minor flavonoids yielded insufficient masses that
were not sufficient to elucidate the structure of these compounds. While more
flavonoids are likely to be present in E. purpurea flowers, greater quantities of
starting material would be required to isolate significant quantities of these
compounds in order to elucidate the structures. Based upon the small quantities
of compounds 10-12 isolated from the flowers and the minute quantities of
unidentified flavonoid-type compounds, it may be safe to suggest that flavonoids
appear to be relatively minor constituents of the methanol extract of E. purpurea
flowers.

While there have not been any reports regarding the flavonoid content of
E. purpurea flowers, numerous flavonoids have been reported from the leaves of
E. angustifolia and E. purpurea (Figure 1.5). Quercetin-3-O-glucoside has been
Isolated from E. purpurea leaves; however, quercetin-3-O-rutinoside and
kaempferol-3-O-robinobioside have not been reported in E. purpurea or E.

angustifolia leaves (Bauer and Wagner, 1991). The flavonoid content of E.

75

Table 3.2. 13C Chemical shifts for flavonoids isolated from E. purpurea.

 

 

 

Compounds
Carbon No. 10 11 12

2 156.5 156.4 156.3
3 133.3 133.3 133.3
4 177.2 177.3 177.4
5 161.2 161.1 161.2
6 98.9 99.0 98.6
7 164.8 165.2 164.2
8 93.7 93.9 93.5
9 156.5 156.5 156.1
10 103.7 103.5 103.9
1' 121.6 120.8 121.6
2' 115.2 130.9 115.2
3' 144.8 115.4 144.8
4’ 148.5 160.0 148.4
5' 116.2 115.1 116.2
6’ 121.1 130.9 121.1
1" 101.3 103.5 100.8
2" 74.1 71.9 74.1
3” 76.5 73.0 77.5
4" 70.6 68.0 69.9
5" 68.3 73.5 76.5
6" 67.0 65.3 60.9
1’” 100.8 100.0

2'" 70.4 70.6

3'” 70.0 70.4

4'" 71.9 71.9

5'” 68.3 68.0

6'" 17.5 17.9

 

76

OH OH OH

 

quercetin
3—0-a-L-rhamnosyl-(1—+6)-
B-D-glucopyranoside

 

kaempferol
3-O-a-L-rhamnosyI-(1—>6)-
B-D-galactopyranoside

 

quercetin
3-O-B-D-glucopyranoside

Figure 3.3. Flavonoids isolated from E. purpurea flowers: kaempferol-3-O-
robinobioside (10), quercetin-3-O—rutinoside (11), and quercetin-3-O—glucoside

(12).

77

purpurea leaves has been reported as 0.48% (Bauer and Wagner, 1991).
Echinacea purpurea flowers appear to contain significantly less flavonoids than
theleaves.

Echinacea. purpurea flowers contain several compounds that have been
reported to be biologically active. Three flavonoids, quercetin-3-O-glucoside,
quercetin-3-O-rutinoside, and kaempferol-3-O-robinobioside, and two
anthocyanins, cyanidin-3-O-B-D glucopyranoside and cyanidin-3-O-malonyl-
(1—>6)-8-D glucopyranoside, were purified from methanolic extracts of Iyophilized
E. purpurea flowers. Echinacea purpurea flower herbal supplement preparations
are a potential new source of the bioactive compounds isolated in this study.
Examination of E. purpurea compounds through various bioassays will lend
credence to the phytoceutical effectiveness of E. purpurea natural product
supplements. Anthocyanins are noted to have high therepeutic value
(Luczkiewcz and Cisowski, 2001; Wang et al., 1999), and flavonoids are well
established as antioxidants (Pietta, 1999; Rice-Evans, 1999; Arora et al., 1998).
Further work to determine the antioxidant and anti-inflammatory activities and the
effects on the induction of cytokines lL-6 and TNF-a will be performed using

these compounds isolated from E. purpurea flowers.

78

CHAPTER 4.
CYCLOOXYGENASE-1 AND —2 INHIBITORY AND ANTIOXIDANT ACTIVITIES

OF COMPOUNDS FROM ECHINACEA PURPUREA (L.) MOENCH

ABSTRACT

The popularity of Echinacea purpurea (L.) Moench as a natural product
supplement has increased in recent years, and products containing E. purpurea
have grown in both number and variety. Tees and extracts from the aerial
portions of the plant, including stems, leaves, and flowers have become
increasingly popular; however, little is known about the bioactivity of compounds
from E. purpurea flowers. Flavonoids (quercetin-3-O-gIucoside, quercetin-3-O-
rutinoside, and kaempferoI-3-O-robinobioside) and anthocyanins (cyanidin-3—O-
B-D glucopyranoside and cyanidin-3-O-malonyl-(1—>6)-[3-D glucopyranoside),
isolated from methanolic extracts Of Iyophilized E. purpurea flowers, were
examined for antioxidant and cyclooxygenase-1 (COX-1) and cyclooxygenase-2
(COX-2) inhibitory activities. Caffeic acid, caftaric acid, chlorogenic acid, and
Cichoric acid, reported components of E. purpurea, were also tested. The
anthocyanins demonstrated the greatest cyclooxygenase inhibitory activities at
100 pg/mL, with inhibition of COX-1 and COX-2 ranging from 26-28% and 23-
26%, respectively. When assayed at 10 pg/mL, the antioxidant activity of these
compounds ranged from 33-81% for the anthocyanins, 3-54% for the flavonoids,
and 35-79% for the caffeoyl derivatives. The results of these studies indicate

that the development of E. purpurea flower-based phytoceuticals is warranted.

79

INTRODUCTION

Echinacea purpurea (L.) Moench is one of three Echinacea spp. currently
of phytoceutical interest. The popularity of Echinacea can be linked to the
general increase in the use of herbal supplements to improve quality of life and
as a means of disease prevention or management (Ness et al., 1999).
Nutraceuticals do not require testing for safety or efficacy prior to marketing
(Percival, 2000; Zink and Chaffin, 1998), and as a result, the active components
in many commercially available products have not been identified (Mazza and
Cottrell, 1999; Melchart et al., 1995). The recent increase in the popularity of
Echinacea has generated a great deal of research interest to determine the
efficacy of this herbal supplement.

Historically, Echinacea roots were used to alleviate pain and inflammation
(Bauer and Wagner, 1991); however, teas prepared from the fresh herb were
also consumed to treat conditions such as rheumatism, arthritis, mumps, and
measles (Kindscher, 1989). Currently, Echinacea extracts are available in a
variety of forms, such as tablets, capsules, tinctures (alcoholic extracts), juices,
and teas (Gunning, 1999; Li and Wang, 1998). Preparations of the fresh herb
(the above ground portion of the plant) includes Echinacea flowers only if the
plants are in bloom at the time of harvest. Commercial extracts composed solely
of Echinacea flowers are not currently available.

A large body of research has been generated regarding the

characterization and bioactivity of compounds from Echinacea roots and fresh

80

herb; however, Echinacea flowers have not received the same attention, and
their bioactivity has not been fully elucidated. The roots of Echinacea have been
determined to possess cyclooxygenase-1 and —2 inhibitory activity (Clifford et al.,
2002) and 5—Iipoxygenase inhibitory activity (MIJIIer-Jakic et al., 1994). The
majority of research regarding the antioxidant activities of Echinacea has focused
on root and leaf extracts (Sloley et al., 2000; Hu and Kitts, 2000; Pietta et al.,
1998), leaving the antioxidant activity of compounds from Echinacea flowers to
be determined.

Echinacea spp. possess a diverse assemblage of phenolic antioxidants
(Bauer and Wagner, 1991) believed to be responsible for the antioxidant activity
of the plant extracts (Facino et al., 1995). Phenolic antioxidants are known to
have free radical scavenging properties and antioxidant properties with regard to
the oxidation Of low-density Iipoproteins. The current popularity Of natural
antioxidants is due to their reported role as modulators of lipid peroxidation as it
relates to diseases such as atherogenesis, thrombosis, and carcinogenesis
(Frankel, 1999). Echinacea purpurea flowers present a new source of such
natural antioxidants for the phytoceutical market.

The purpose of this research was to determine the antioxidant and
cyclooxygenase-1 (COX-1) and cyclooxygenase-2 (COX-2) inhibitory activities of
compounds isolated from E. purpurea flowers and compounds identified
previously in E. purpurea extracts. Caffeoyl derivatives reported in E. purpurea
(caffeic acid, caftaric acid, chlorogenic acid, and Cichoric acid) (Figure 1.4), and

flavonoids (quercetin-3-O-glucoside, quercetin-3-O—rutinoside, kaempferol-3—O—

81

robinobioside) (Figure 3.3) and anthocyanins (cyanidin-3—O—8-D glucopyranoside
and cyanidin-3-O-malonyl-(1——>6)-B-D glucopyranoside) (Figure 3.2) isolated from
E. purpurea flowers were examined to determine the potential bioactivity of E.

purpurea flower extracts.

MATERIALS AND METHODS
General Experimental

Anthocyanins cyanidin-3—O—8—D glucopyranoside (8) and cyanidin-3—O-
malonyl-(1a6)-B-D glucopyranoside (9) and flavonoids quercetin-3-O—rutinoside
(10), kaempferol-3-O-robinobioside (11), and quercetin-3-O-glucoside (12) were
isolated from methanolic extracts of Iyophilized E. purpurea (L.) Moench flowers
as previously described (Chapter 3). Caffeic acid (13), caftaric acid (14),
Cichoric acid (15), and chlorogenic acid (16) were purchased from Sequoia
Research Products (Oxford, UK).
Cyclooxygenase Inhibitory Assay

An in vitro system was used to assess the cyclooxygenase inhibitory
activity of E. purpurea flavonoids, anthocyanins, and caffeoyl derivatives. The
COX-1 enzyme used in the assay was obtained from ram seminal vesicle
microsomel preparations according to previously reported methods (Meade et al,
1993). A recombinant human microsomel preparation of COX-2 was provided by
Dr. David Dewitt (Michigan State University, East Lansing, MI). The assay was
conducted in an Instech chamber (Instech Laboratory, Plymouth Meeting, PA)

maintained at 37 °C, containing 600 HL of a reaction buffer comprised of 0.1 M

82

Tris, 1 mM phenol, cyclooxygenase enzyme (10 HL), and 17 pg hemoglobin.
F lavonoids, anthocyanins, and caffeoyl derivatives were dissolved in DMSO such
that a 10 uL aliquot yielded a final concentration of 100, 50, or 10 pg/mL for the
caffeoyl derivatives and anthocyanins or 100 pg/mL for the flavonoids. Assay
concentrations in HM are noted in Appendix A. The reaction was initiated with
the addition of arachidonic acid. Initial oxygen uptake was monitored using a YSI
model 5300 biological oxygen monitor (Yellow Springs Instruments, Inc., Yellow
Springs, OH) and recorded using Quicklog Windows software, version 1.0
(Strawberry Tree, lnc., Sunnyvale, CA). Analyses of samples and controls were
performed in triplicate.

Aspirin, ibuprofen, naproxen, celecoxib (CelebrexTM), rofecoxib (VioxxTM),
and valdecoxib (BextraTM) were dissolved in DMSO and used as controls for both
the COX-1 and COX-2 inhibitory assays. Aspirin, ibuprofen, and naproxen
controls were assayed at concentration resulting in approximately 50% inhibition
of the COX-1 enzyme. Aspirin was tested at 180 pg/mL (1000 HM), ibuprofen at
2.06 pg/mL (10 HM), and naproxen at 2.52 pg/mL (10 pM) for both COX-1 and
COX-2 . CelebrexTM, VioxxTM, and BextraTM were obtained as physician’s
professional samples, ground to a fine powder, and tested as a suspension in
DMSO at 1.67 Hg/mL for both COX-1 and COX-2. The initial decrease in 02

levels in the presence of test compounds and standard NSAIDs was compared to
that of the DMSO control in order to determine inhibition of COX-1 or COX-2

enzymes.

83

Antioxidant Assay

The antioxidant assay was conducted according to previously described
methods (Arora and Strasburg, 1997). 1-Stearoyl-2-linoleoyI-sn-glycerOl-3-
phosphocholine (lipid) and 3—(p-(6-phenyI)-1,3,5-hexatrienyl) phenylpropionic
acid (fluorescent probe) were combined, dried under vacuum, and suspended in
MOPS buffer (1 mL of 0.15 M NaCl, 0.1 mM EDTA, and 0.01 M MOPS). Large
unilamellar Iiposomes were prepared by subjecting the lipid-probe mixture to ten
freeze-thaw cycles in a dry ice-ethanol bath followed by extrusion through a 100
nm pore size membrane. The final assay volume of 2 mL was composed of 100
uL of HEPES buffer (50 mM HEPES and 50 mM Tris), 200 uL NaCl (1 M), 1.64
mL of N2-sparged water, 20 pL of sample (to achieve a final concentration of 100,
50, or 10 pg/ml) or DMSO (control), and a 20 pL aliquot of liposome suspension.
Peroxidation was initiated by the addition of 20 pL of FeCl2o4H2O (0.5 mM) and
the samples were gently vortexed. Probe fluorescence was monitored using a
Turner fluorometer with a narrow bandpass 360 filter and a sharp cut 415 filter at
times 0, 1, 3, 6, and every three minutes thereafter for 21 min. Antioxidant
activity was determined by comparing the relative fluorescence of test
compounds with the control. Test compounds were compared to standard
synthetic antioxidants tert-butyl hydroquinone (TBHQ, 1.7 pg/mL), butylated
hydroxytoluene (BHT, 2.2 pg/mL), and butylated hydroxyanisole (BHA, 1.8
pglmL), as well as the natural antioxidant vitamin E (4.3 Hg/mL). Analyses of

samples and controls were performed in triplicate.

84

RESULTS AND DISCUSSION
Cyclooxygenase Inhibitory Assay

The COX-1 and COX-2 inhibitory activities of compounds from E.
purpurea are summarized (Figure 4.1). The anthocyanins cyanidin-3—O—8-D
glucopyranoside (8) and cyanidin-3-O—malonyI-(1—>6)-B-D glucopyranoside (9)
demonstrated the greatest anti-inflammatory activities of the compounds tested,
with 28.2 i 0.3 and 25.9 i 16% inhibition of COX-1 and 25.6 i 1.1 and 22.9 i
0.5% inhibition of COX-2, respectively. Weak inhibition of COX-1 by quercetin-3-
O-rutinoside (10) and quercetin-3-O-glucoside (12) was noted, with inhibitory
activities Of 14.5 i 7.8 and 11 i 5.6%, respectively (Figure 4.1). Kaempferol-3-
O-robinobioside (11) did not inhibit COX-1. None of the flavonoids tested
exhibited COX-2 inhibitory activity. Caffeic acid (13), caftaric acid (14), chicoric
acid (15), and chlorgenic acid (16) demonstrated nearly negligible COX-1
inhibitions Of 5.1 i 1.0, 5.9 i 1.0, 8.9 i 1.0, and 7.8 i 0.9%, respectively (Figure
4.1). COX-2 inhibitory activity for the caffeoyl derivatives was weak, and
compounds 13-16 were not considered active inhibitors of this enzyme.

Cyanidin-3—O—B-D glucopyranoside (8) and cyanidin-3-O-malonyl-(1—+6)-B-
D glucopyranoside (9) demonstrated similar dose-response effects against COX-
1 and COX-2 at 100, 50, and 10 pglmL (Figure 4.2). At 50 and 10 pg/mL, both
compounds were effective inhibitors Of COX-1; however, COX-2 inhibition was

significant only at 50 pg/mL. While the flavonoids and caffeoyl derivatives were

85

 

 

 

 

100 ,. --. ..-- -_-. .. ----.-.--.__~-~-.-- .._..--_ we W -- . i M... F
3 ICOX-1
a 30._. .-..----,--,,i,_,,
5 EICOX-Z
or
8
2 60 ._______._______.___.. __.___- w
o
>.
o
I.-
O
g 40 -—————--- , n—mmim. -,
.9
§ I
E 20 — - - - - 4* II
°\° I
o . .2 I i ii

 

 

 

 

 

 

 

 

 

 

 

b

c. 6,0666éb \<‘¢¢
85906.8 ‘3‘6-‘6 a a are as we

6‘ 6 V" Y" 5’29 a? 9
6°C 000 0° (5°09 ‘6‘0 6‘0 {‘06 v9.05? 1‘9.) 60
e a a 40.40 s 9 o
8¢§§¢§¢°¢ 0C," 5 \c’o
6‘ 0° 6° 6° \‘ c,
’b ‘9 9‘ e} g0
6"“ e9
.\
6° ‘0“
c,“

Figure 4.1. Comparison of cyclooxygenase-1 and -2 inhibitory activities for
compounds from E. purpurea. Compounds were assayed at 100 pg/mL and
compared to NSAID standards aspirin (180 Hg/mL), ibuprofen (2.06 pg/mL),
naproxen (2.52 jig/mL), CelebrexTM (1.67 pg/mL), VioxxTM (1.67 ug/mL) and

BextraTM (1.67 jig/mL), Data represent the average 1- one standard deviation.

86

 

 

40’

COX-1 COX-2

 

I100
D50
D10

 

 

 

% COX inhibition

 

 

Figure 4.2. COX-1 and COX-2 inhibitory activities for anthocyanins isolated from
E. purpurea flowers at 100, 50 and 10 pg/mL. Numbers in the legend represent
the concentration assayed in pg/mL. Data represent the average i one standard

deviation.

87

also tested at 50 and 10 pg/mL, only the anthocyanins were active at

concentrations below 100 ug/mL against either COX-1 or COX-2.

Antioxidant Assay

The results for the antioxidant assay are expressed as the mean percent
inhibition of oxidation following 21 minutes of incubation (Figure 4.3). When
tested at 10 pg/mL, cyanidin-3-O-B-D glucopyranoside (8) (81% inhibition)
demonstrated greater antioxidant activity than cyanidin-3-O—malonyl-(1——>6)-[3-D
glucopyranoside (9) (33% inhibition) after 21 min. Caffeic acid (13), caftaric acid
(14), Cichoric acid (15), chlorogenic acid (16) exhibited strong antioxidant
activities of 73, 35, 79, and 73%, respectively.

At 100 pg/mL, quercetin-3-O-rutinoside (10), kaempferol-B-O-
robinobioside (11), and quercetin-3-O-glucoside (12) exhibited good antioxidant
activity (73, 83, and 85%, respectively) when compared to the synthetic
antioxidants and vitamin E (Figure 4.4). Reducing the concentration to 50 pg/mL
still resulted in significant antioxidant activity for compounds 10, 11, and 12, with
inhibitions of 47, 54, and 42%, respectively (Figure 4.3). Compounds 10 and 12
exhibited antioxidant activity at 10 pg/mL superior to vitamin E at the end of 21
minutes, with inhibitions of 54 and 42%, respectively. Activity for compound 11
decreased to 3% at 10 pg/mL.

Additional tests were performed by combining 5 pg/mL aliquots of the flavonoids
from E. purpurea in various combinations to yield a 10 pg/mL mixture. lnhibitions

of 57, 52, and 16% were noted for the mixtures of compounds 10/12,

88

 

Kaempfe rol-3-O—robinobioside
Que rcetin-3-O-rutinoside

Que rcetin-3-O-glucos ide
Cichoric Acid

Chlorgenic Acid

Caftaric Acid

Caffeic Acid

Cyan. mal. glucoslde

Cyan. glucoside

 

Fe blank
TBHQ
BHT
BHA
Vitamin E 1 Q
0 20 40 60 80 100
% inhibition

Figure 4.3. Antioxidant activity exhibited by compounds from E. purpurea
assayed at 10 ug/mL and compared to oxidation exhibited by Fez“ control.
Activity was calculated relative to the DMSO control and compared to standard
synthetic antioxidants TBHQ (1.7 pg/mL), BHT (2.2 pg/mL), and BHA (1.8
pg/mL), as well as the natural antioxidant vitamin E (4.3 ug/mL). Data represent

the average : one standard deviation.

89

 

 

 

Fe2+ I , 1 E110

 

 

 

Querc.-gluc.lKaemp.- . T ‘
robino. ‘ i ‘ i

Querc.-rutin.lKaemp.- :3" ‘ ‘

l
robino. . ‘ s l 4:
‘ l
l
l

 

 

 

Querc.-rutin.IQuerc.- , H1 , ‘ ll
gluc.

 

 

Kaempferol-
robinobioside '

Quercetin-glucoside . i“

l
1
Hi 1. _ l

 

llrl‘f’l ‘- .‘k‘ ’. _

 

 

 

 

Quercetin-rutinoside

 

 

0 20 40 60 80 100
% Inhibition

Figure 4.4. Antioxidant activity at 21 minutes for quercetin-3-O—glucoside,
quercetin-3-O—rutinoside, and kaempferol-3-O-robinobioside at 10, 50, and 100
ug/mL compared to oxidation exhibited by Fe2+ control. Activity was calculated
relative to the DMSO control. Numbers in the legend represent the concentration

assayed in pg/mL. Data represent the average i one standard deviation.

90

11/12, and 10/11, respectively. Activity was lower for the mixture of compounds
10 and 11, with an inhibition of 16% as compared to 54% for quercetin-3-O—
rutinoside and 3% for kaempferol-3-O-robinobioside when tested individually.
When examined at a concentration of 10 ug/mL, the combination of compounds
11 and 12 demonstrated greater activity (52% inhibition) than compound 12
(42%) or compound 11 (3%). These results suggest a possible synergistic effect
between kaempferol-3-O-robinobioside (11) and quercetin-3-O-glucoside (12) in
the antioxidant assay.

Caffeoyl derivatives (Figure 1.4) are considered to be among the most
active components of Echinacea spp. (Li and Wang, 1998). For example,
cichoric acid, one of the major compounds in the fresh herb of E. purpurea, has
numerous documented biological effects (Nasslein et al., 2000). Unfortunately,
caffeoyl derivatives are highly susceptible to enzymatic and oxidative
degradation during extraction (Bergeron et al., 1999). Cichoric acid is rapidly
degraded by polyphenol oxidase in fresh plant juice and is a challenge to isolate
due to its instability. Lyophilization has been shown to drastically reduce cichoric
acid and caftaric acid quantities in Echinacea (Kim et al., 2000b). In order to
avoid degradation during the processing of fresh E. purpurea flowers, caffoyl
derivatives were purchased rather than isolated for examination in the
cyclooxygenase inhibitory and antioxidant assays.

Previous work regarding the anti-inflammatory activity of Echinacea has
focused on root extracts rather than whole plant or flower extracts. Alkamides

isolated from E. angustifolia roots have demonstrated cyclooxygenase and 5-

91

 

Iipoxygenase inhibitory activities in in vitro assays (Miiller-Jakic et al., 1994). An
examination of the alkamides isolated from E. purpurea roots revealed their
comparative COX-1 and COX-2 inhibitory activities (Clifford et al., 2002). The
anti-inflammatory activities of Echinacea flower extracts or compounds isolated
from E. purpurea flowers have not been previously determined.

Anthocyanins are known inhibitors of COX-1 and COX-2 enzymes (Wang
et al., 1999; Seeram et al., 2001). The results obtained here regarding the COX-
1 and COX-2 inhibitory activities of cyanidin-B-O-B-D glucopyranoside and
cyanidin-3-O-malonyl—(1——>6)-[3-D glucopyranoside, isolated from E. purpurea
flowers, are in agreement with those previously reported. Cyanidin-3-O—malonyl-
(196)-8-0 glucopyranoside possessed slightly lower activity than cyanidin-3-O-
[3-D glucopyranoside at a concentration of 100 pg/mL, confirms previous
research suggesting that the activity is due to the cyanidin aglycone (Wang et al.,
1999)

Neither the flavonoids nor the caffeoyl derivatives demonstrated significant
anti-inflammatory activity in the COX-1 or COX-2 inhibitory assays. Many whole
plant and root extracts are standardized to echinacoside content or total phenolic
content. Echinacea purpurea extracts do not contain echinacoside, and the
caffeoyl derivatives and flavonoids examined in the cyclooxygenase inhibitory
assays exhibited very weak activity. In order to obtain significant anti-
inflammatory effects from E. purpurea flower or fresh herb extracts, the
anthocyanins examined in this study may provide a better standardization marker

to ensure activity.

92

,, -m-‘ Ii fi

 

rm» .

The antioxidant activity of Echinacea extracts and compounds isolated
from Echinacea have been evaluated in various in vitro assays. A phenolic
Echinacea extract (plant species, part, and extraction method not identified),
standardized to 4.5% echinacoside, was found to be ineffective at scavenging
free radicals by the Trolox equivalent antioxidant assay (Pietta et al., 1998).
Caffeic acid, chlorogenic acid, and cichoric acid have been assessed for their
ability to prevent free radical-induced degradation of collagen, demonstrating that
cichoric acid exhibited the most pronounced protective effects (Facino et al.,
1995). These results suggest that the testing of specific compounds from
Echinacea extracts rather than complex mixtures might yield important
information regarding antioxidant activity. The use of echinacoside as a phenolic
standardization marker should also be reconsidered due to the lack of significant
occurrence in E. purpurea extracts.

Antioxidant studies of Echinacea have focused on the roots and leaves of
the plant, overlooking extracts of the flowers. Echinacea purpurea root and leaf
extracts were found to be effective in scavenging hydroxyl free radicals in an
assay measuring the peroxidase-catalyzed accumulation of 2,2'-azino-bis(3-
ethylbenzthiazoline-6-sulfonic acid) (Sloley et al., 2000). Methanolic E. purpurea,
E. angustifolia, and E. pallida root extracts also suppressed hydroxyl radical
generation by 20-34% due to the copper chelating capacity of these extracts (Hu
and Kitts, 2000). The ability of E. purpurea root and leaf extracts to inhibit Fe”-
induced lipid peroxidation in a catecholaminergic neuroblastoma SH-SY5Y cells

using a thiobarbituric acid assay was assessed, with leaf extracts demonstrating

93

 

higher antioxidant activity than root extracts (Sloley et al., 2000). Based upon the
antioxidant potential demonstrated by previous research and the antioxidant
results obtained in these studies, compounds isolated from E. purpurea flowers
possess significant antioxidant activity.

Previous research regarding the antioxidant potential of various flavonoids
suggested that the ability to scavenge free radicals and chelate metal ions
stemmed from the degree of conjugation, the presence and arrangement of
substituents on the B-ring, and substitution at the 3 position on the C-ring of the
flavonoid molecule. Compared to quercetin, kaempferol lacks the 3’-hydroxyl
group on the B-ring. Research has suggested that the 3’-hydroxyl group of
quercetin contributes to the chelation and antioxidant potential of the molecule
(Rice-Evans, 1999). When tested at 10 pg/mL, the lower antioxidant activity
observed for compound 11 compared to compounds 10 and 12 supported these
previously reported observations of the importance of the 3’-hydroxyl group as it
pertains to antioxidant activity.

The results from the cyclooxygenase inhibitory and antioxidant assays
indicate that E. purpurea flowers are a potential source of bioactive compounds.
Currently, E. purpurea flowers are included in extracts of the fresh herb only if the
flower is in bloom at the time of harvest. The results presented here demonstrate
that extracts prepared solely from the flowers may provide a concentrated source
of bioactive compounds, based upon the cyclooxygenase inhibitory and
antioxidant activities of the anthocyanins, flavonoids, and caffeoyl derivatives

examined in these studies. The promising results obtained in this work indicate

94

~ “".J- L _

CI“

 

that further research regarding the development of an E. purpurea flower-based

natural product supplement is warranted.

 

IQW¢-‘i 9;" ’7" .. P
'l . L‘.’ . .‘ - ~ ‘

95

CHAPTER 5.
CHARACTERIZATION OF CYTOKINE PRODUCTION IN RAW 264.7 MURINE
MACROPHAGES FOLLOWING TREATMENT WITH COMPOUNDS FROM

ECHINACEA PURPUREA (L.) MOENCH

ABSTRACT

Echinacea phytoceuticals are consumed for purported prevention of or reduction
in the severity of upper respiratory tract infections; however, little is known
regarding their immune mode of action. Flavonoids (quercetin-B-O-glucoside
and quercetin-3-O—rutinoside) and anthocyanins (cyanidin-3-O—B-D
glucopyranoside and cyanidin-3-O-malonyl-(1—>6)-B-D glucopyranoside), isolated
from methanolic extracts of Iyophilized E. purpurea flowers, and alkamides
isolated Echinacea purpurea (L.) Moench roots (undeca-2E,4Z-dien-8,10—diynoic
acid isobutylamide, undeca-2Z,4E—dien-8,10-diynoic acid isobutylamide, undeca-
2E,4Z-dien-8,10-diynoic acid 2-methylbutylamide, and a mixture of dodeca-
2E,4E,82,1OE—tetraenoic acid isobutylamide and dodeca-2E,4E,82,1OZ-
tetraenoic acid isobutylamide) were examined for their effects upon the
production of cytokines lL-6 and TNF-a in murine macrophages. Caffeic acid,
caftaric acid, chlorogenic acid, and cichoric acid, reported components of E.
purpurea, were also tested. Cytotoxicity was assessed concurrently by using the
MTT assay (3-[4,5-dimethylthiazol-Z-yll-2,5-diphenyltetrazolium bromide), and
the highest concentration examined for all four alkamides, 100 pg/mL, adversely

affected cells and was not used in the macrophage assay. When cells were

96

treated solely with compounds from E. purpurea, no induction of lL-6 or TNF-or
occurred. Co-treatment of cells with 100 or 1000 ng/mL lipopolysaccharide
(LPS) demonstrated that several compounds isolated from E. purpurea assayed
at concentrations of 1, 10, 50, and 100 pg/mL have the ability to suppress
induced inflammatory response. The alkamides demonstrated the greatest
ability to affect lL-6 and TNF-or production in cells stimulated with 100 and 1000

nglmL LPS, with significant reductions of lL-6 and TNF-a noted when treated

with as little as 1 pg/mL. The anti-inflammatory activity of the alkamides reported
here supports the anecdotal claims of pain and fever relief associated with
Echinacea use. No significant reductions in lL-6 were noted for either the
anthocyanins or flavonoids when stimulated with 100 or 1000 nglmL LPS;
however, reductions in TNF-a levels were noted with 10, 50, and 100 ngmL
quercetin-3-O-rutinoside and at 1, 10, 50, and 100 pg/mL cyanidin-3-O-malonyl-
(1——>6)-(3-D glucopyranoside. The trend in lL-6 and TNF-oc production indicated a
potential synergistic effect of the caffeoyl derivatives with LPS. Co-treatment of
the macrophages with the caffeoyl derivatives and 100 and 1000 nglmL LPS
resulted in significant increases in lL-6 and TNF-a levels at 50 and 100 ug/mL.
These results demonstrate that individual compounds from E. purpurea might
possess opposing activities with regard to immune stimulation and/or
suppression, and might explain the disparity previously reported in clinical

studies.

97

 

INTRODUCTION

Echinacea phytoceuticals are among the most popular natural product
supplements in the United States (Barrett et al., 1999). Historically, Echinacea
root and fresh herb extracts were used for the treatment of fevers, burns, insect
bites, wounds, coughs, and to alleviate the pain of toothaches, headaches, and
stomach cramps (Bauer and Wagner, 1991). The current popularity of
Echinacea is due primarily to reported anecdotal claims of antiviral and
nonspecific immunostimulatory properties (Borchers et al., 2000; Proksch and
Wagner, 1987).

A limited number of studies have been conducted to substantiate the
anecdotal claims associated with Echinacea use (Borchers et al., 2000). The
pain relieving and general immune stimulating effects of Echinacea have been
evaluated to some extent (Bauer, 1998; Bauer and Wagner, 1991; Proksch and
Wagner, 1987). 4-O-Methyl-glucuronoarabinoxylan, isolated from E. purpurea,
was believed to be responsible for enhancing phagocytosis in an in vitro
granulocyte assay (Proksch and Wagner, 1987). In another study, mouse L-929
cells treated with crude aqueous and methanol extracts of Echinacea whole plant
exhibited 50-80% resistance to herpes, influenza, and vesicular stomatitis viruses
(Wacker and Hilbig, 1978). In addition, caffeic acid and cichoric acid, isolated
from Echinacea, have demonstrated the ability to reduce the infectivity of

vesicular stomatitis virus by more than 50% at 62.5 pg/ml of caffeic acid or 125

pg/ml of cichoric acid in mouse L-929 cells. When incubated with the caffeoyl

98

 

derivatives, infectivity was reduced by more than 50% (Cheminat et al., 1988).
Acidic arabinogalactan, produced by E. purpurea cell cultures, activated
macrophages against Leishmania enn‘ettii and tumor cells, decreasing L. enn‘ettii
growth in intracellular and phagocytosis-associated assays (Luettig et al., 1989).

A variety of studies have been performed to determine the anti-
inflammatory activity of extracts and compounds from Echinacea. Intravenous
and topical administration of aqueous E. angustifolia root extracts have been
shown to inhibit carrageenan-induced paw inflammation in mice (Tragini et al.,
1988; Tragini et al., 1985). Alkamides from E. angustifolia, when examined in
vitro, were found to possess cyclooxygenase and 5-lipoxygenase inhibitory
activities (MUller-Jakic et al., 1994). The cyclooxygenase-1 and cyclooxygenase-
2 inhibitory activities of alkamides from E. purpurea have been compared
(Clifford et al., 2002). Echinacea purpurea alkamides demonstrated in vitro
inhibition of both enzyme isoforms; however, preferential COX-2 inhibition was
not identified.

Several Echinacea products, including E. purpurea roots and herb, E.
purpurea and/or E. angustifolia whole plant extracts standardized either to 4%
phenolic content (chlorogenic acid and cichoric acid) or to
echinacosides/alkamides, and Echinacea pressed juice preparations, were
examined to determine lL-6 and TNF-a production in LPS-stimulated murine
macrophages (Rininger et al., 2000). Quantitative determination of lL-6 and
TNF-a indicated that only the E. purpurea root and herb materials, subjected to a

simulated digestion (incubation with various simulated gastric fluids), exhibited

99

 

macrophage activating activity based upon increased cytokine production.
Standardized extracts and samples not subjected to the digestion protocol did
not demonstrate significant immunomodulatory activity (Rininger et al., 2000).

Clinical studies have analyzed the effects of various Echinacea
preparations for the prevention and/or treatment of colds and upper respiratory
tract infections (Bauer, 1998). Echinacea pallida and E. purpurea root extracts
significantly reduced the duration of respiratory tract infections (Braunig et al.,
1993; Braunig et al., 1992). Administration of a dried, ethanol extract of E.
purpurea (95% herb and 5% root) in tablet form reduced the symptoms of
respiratory tract infections (Brinkeborn et al., 1998). Echinacea purpurea fresh
herb juice has demonstrated a prophylactic effect in preventing the development
of upper respiratory tract infections (Hoheisel et al., 1997).

While the studies indicated above demonstrate beneficial effects of
Echinacea extracts, many other clinical trials found no significant effects on the
duration or severity of colds and respiratory tract infections associated with the
consumption of Echinacea extracts (Bauer, 1998; Gunning, 1999). Echinacea
purpurea juice did not significantly reduce the number or duration of upper
respiratory tract infections in patients with a prior history of upper respiratory tract
infections (Schoneberger, 1992). In another clinical trial, E. purpurea and E.
angustifolia root extracts did not significantly prevent upper respiratory tract
infections (Melchart et al., 1998). No significant differences were noted with

regard to the incidence, duration, or severity of colds and respiratory tract

100

 

infections in a study designed to determine the prophylactic effect of expressed
E. purpurea juice (Grimm and Miller, 1999).

The disparity reported regarding the prevention and treatment of upper
respiratory tract infections by Echinacea extracts may result from dissimilar
chemical compositions of crude extracts examined. The variability in the
composition of the extracts was the result of a lack of uniform, industry-wide
standardization methods. The difference in extraction method and reported
differences between Echinacea species contributed to the variability in
Echinacea extracts examined (Binns et al., 2002). Testing of specific
compounds from Echinacea would provide a chemical basis upon which the
standardization of Echinacea-based herbal supplements could be based.

The purpose of this research was to determine the effects of co-treating
murine macrophage cells (RAW 264.7) with lipopolysaccharide (LPS), an
inflammagenic stimulus, and compounds isolated from E. purpurea. The
compounds examined in this study included E. purpurea anthocyanins 8 and 9
(cyanidin-B—O-B-D glucopyranoside and cyanidin-3-O-malonyl-(1—>6)-j3-D
glucopyranoside), flavonoids 11 and 12 (quercetin-B-O-glucoside and quercetin-
3-O-rutinoside), caffeoyl derivatives 13-16 (caffeic acid, caftaric acid, cichoric
acid, and chlorogenic acid), and alkamides 1, 2, 4, and 6!? (undeca-2E,4Z—dien-
8,10-diynoic acid isobutylamide, undeca-22,4E—dien-8,10-diynoic acid
isobutylamide, undeca-2E,4Z—dien-8,10-diynoic acid 2-methylbutylamide, and a
mixture of dodeca-2E,4E,82,1OE—tetraenoic acid isobutylamide and dodeca-

2E,4E,8Z,1OZ-tetraenoic acid isobutylamide). Anti-inflammatory effectiveness

101

.7

 

was noted by the reduction in interleukin-6 (IL-6) and tumor necrosis factor-a
(TNF-a) produced by the cells stimulated with LPS compared to control values.
MTT assays were performed for each compound concomitant with the murine

macrophage assay to assess potential cytotoxicity of the doses examined.

MATERIALS AND METHODS
General Experimental

Alkamides undeca-2E,4Z—dien-8,10-diynoic acid isobutylamide (1),
undeca-2Z,4E—dien-8,10-diynoic acid isobutylamide (2), undeca-2E,4Z-dien-8,10-
diynoic acid 2-methylbutylamide (4), and a mixture of dodeca-2E,4E,8Z,1OE-
tetraenoic acid isobutylamide and dodeca-2E,4E,82,1OZ-tetraenoic acid
isobutylamide (617) were isolated from E. purpurea roots as previously described
(Chapter 2). Anthocyanins cyanidin-3-O-B-D glucopyranoside (8) and cyanidin-
3-O-malonyl-(1—>6)—B-D glucopyranoside (9) and flavonoids quercetin-3-O-
rutinoside (10) and quercetin-B-O-glucoside (12) were isolated from methanolic
extracts of Iyophilized E. purpurea (L.) Moench flowers as previously described
(Chapter 3). Caffeic acid (13), caftaric acid (14), cichoric acid (15), and
chlorogenic acid (16) were purchased from Sequoia Research Products (Oxford,
UK).
Cell Culturing

Murine macrophage cells (RAW 264.7, American Type Culture Collection,
Rockville, MD) were cultured in Dulbecco’s modified Eagle medium (DMEM)

(Sigma, St. Louis, MO) supplemented with 10% (v/v) fetal bovine serum (FBS)

102

Atlanta Biologicals, Norcross, GA), 1 mM sodium pyruvate (Sigma), 1% (v/v)
NCTC-135 (Gibco Laboratories, Chagrin Falls, IL) streptomycin (100 pg/mL,
Sigma), and penicillin (100 U/mL, Sigma). Cultures were maintained at 37 °C in
a humidified incubator with 5% C02. For the assay, cells were grown to
confluency in sterile tissue culturing dishes, the media was removed, and the
cells were incubated for 10 min with phosphate buffered saline (PBS). Cells
were detached by gentle, repeated pipetting, centrifuged, and resuspended in
fresh media. Cell viability and number were assessed by trypan blue dye
exclusion (Strober, 1991) and enumerated using a hemacytometer. Three cell
stock solutions were prepared at a cell density of 5 x 105 cells/mL with 0, 200, or
2000 nglmL lipopolysaccharide (LPS) (Sigma, SI 3.6 @ 15.6 pglmL LPS, 1.5
EU/ng LPS). Compounds 1, 2, 4, 6/7, 8-9, 10, 12, and 13-16 were dissolved in
DMSO and added to media to yield stock test compound concentrations of 2, 20,
100, and 200 pg/mL. Final DMSO concentrations did not exceed 0.02% DMSO
in the test compound stocks.

Cell stocks with LPS (0, 200, and 2000 nglmL LPS) (400 pL) were added
to sample wells of a 48-well plate (Costar, Cambridge, MA) in combination with
400 pL of the test compound stock solutions (2, 20, 100, and 200 pg/mL). Final
assay concentrations were 0,100, and 1000 ng/mL LPS and 0, 1, 10, 50, and 100
pg/mL test compound, with a final cell density of 1 x 106 cells/mL. Assay
concentrations in pM are noted in Appendix A. Vehicle controls with equivalent
concentrations of DMSO were also prepared. Cultures containing controls and

test compounds were incubated for 12 h, after which time the media was

103

 

immediately aspirated from each well and stored at -80 °C in cryogenic vials. All
compounds and controls were prepared in triplicate.
MTT Assay

The MTT assay determines adverse effects on cells through the cleavage
of the tetrazolium ring in the water-soluble tetrazolium salt (3-[4,5-
dimethylthiazoI-Z-ylj-Z,5-diphenyltetrazolium bromide) by mitochondrial
dehydrogenase enzymes in live cells to form an insoluble formazan (Marshall et
al., 1995; Denizot et al., 1986; Mossman, 1983). Dead cells cannot convert MTT,
therefore spectroscopic measurement of the dissolved formazan yields valuable
information regarding the potential cytotoxic effects of test compounds.
Reductions in lL-6 and TNF-a levels might be incorrectly attributed to the action
of the compounds on the inflammation response rather than the underlying cell
death that resulted in fewer cells capable of producing the cytokines. The MTT
assay ensures that the effects noted in the cell assay were not due to overt
toxicity of the compounds tested.

The MTT assay was performed concurrently with the cell culture assay.
The MTT cleavage assay was performed as previously described (Ji et al., 1998)
with minor modifications. The MTI’ (3-[4,5-dimethylthiazol-2-yl]-2,5-
diphenyltetrazolium bromide) (Sigma, St. Louis, MO) reagent was prepared by
dissolving the dye in 0.01 M PBS and filter sterilized to yield a 5 mg/mL stock
solution. Aliquots of 100 pL each of cell stocks and test compounds or controls

were added to the wells of a 96-well plate (Costar).

104

Following an incubation period of 10 h at 37 °C, 50 pL of the MIT stock
solution was added to each well, and the plate was further incubated for 2 h at 37
°C. The plates were then centrifuged at 500 x g for 20 min, and the supernatant
removed by aspiration. The formazan crystals at the bottom of the plate wells
were dissolved by the addition of 200 pL/well DMSO. After the crystals had
completely dissolved, the absorbance of each well was read on a Vmax Kinetic
Microplate Reader (Molecular Devices, Menlo Park, CA) at A 570-690 nm. Data
management was performed using Vmax Softmax Pro Software version 3.0
(Molecular Devices).

ELISA Quantification of IL-6 and TNF-a

lL-6 and TNF-a produced in the macrophage assay were quantified by
enzyme linked immuno-sorbent assay (ELISA) according to previously described
procedures (Ji et al., 1998). Microtiter strip wells (lmmunolon IV Removawell,
Dynatech Laboratories Inc., Chantilly, VG) were coated with 50 pL/well of 1
pUmL purified rat anti-mouse lL-6 or TNF-a antibodies (BD-PharMingen, San
Diego, CA) in 0.1 M NaHC03 (pH 8.2) and held at 4 °C for 16 h. Wells were
washed three times with PBST (0.01 M PBS with 0.2% (v/v) Tween 20) and
blocked to prevent nonspecific protein binding with 300 prell of 3% bovine
serum albumin (BSA) in PBST (0.01 M PSB with 0.2% (v/v) Tween 20) (BSA-
PBST) at 37 °C for 30 min. Wells were washed four times, and recombinant
murine lL-6 or TNF-a standards (BD-PharMingen) prepared in 10% (v/v) FBS-
DMEM and samples from the macrophage assay (recovered media) were added

in 50 uL aliquots and incubated for 1 h at 37 °C. After the incubation, the wells

105

 

IT

were washed with PBST four times followed by one wash with dHZO. Aliquots of
50 uL of biotinylated rat anti-mouse IL-6 or TNF-a (BD-PharMingen), diluted to 1
pUmL in 3% BSA-PBST (w/v), were added to each well and plates were
incubated at room temperature for 1 h. Wells were washed six times with PBST
and once with dHZO following the incubation time. After washing, 50 pL of 1.5
pg/mL streptavidin-horseradish peroxidase (Sigma) in 3% BSA-PBST were
added to each well. After 1 h incubation at room temperature, wells were
washed eight times with PBST and twice with dHZO. Aliquots of 100 uL of
tetramethylbenzidine (TMB) substrate (K-blue Max®, Neogen, Lexington, KY)
were added to each well. After the development of color, approximately 15 min,
the reaction was stopped with the addition of 100 pL of 6 N H280, to each well.
The optical density of the sample wells was immediately measured on a Vmax
Kinetic Microplate Reader, using Vmax Softmax Pro Software version 3.0
(Molecular Devices) at 1. 450 nm.
Statistical Analysis

Data for the cell culture assay were analyzed by one way analysis of
variance (ANOVA) followed by Bonferroni’s test using SigmaStat for Windows,
version 2.0 (Jandel Scientific, San Rafael, CA). The cell culture assay was
performed in triplicate, and the media from each cell culture well was analyzed in
duplicate. Data from lL-6 and TNF-a quantification are reported as the mean i
one standard deviation. Values in subsequent graphs significantly different from

controls (p<0.05) were noted by an asterisk (*).

106

 

RESULTS AND DISCUSSION
MTT Assay

The MTT assay was performed to determine the potential cytotoxicity of
compounds from E. purpurea tested in the macrophage assay. Cells were
treated with MTT after 10 h of incubation, and incubated for another 2 h to yield a
total exposure time equal to that of the macrophage cell assay. No significant
differences (p < 0.05) were noted between controls (0, 100, and 1000 ng/mL
LPS) versus the concentrations tested for compounds 8-9, 10, 12, and 13-16.
The results from the MTT assay indicated that the highest concentration tested
for the four alkamides (1, 2, 4, and 6/7), 100 pg/mL, adversely affected the
macrophage cells (Figures 5.1-5.4). Therefore, the 100 pg/mL concentration
was not used to determine the effects on cytokine production for compounds 1,
2, 4, and the mixture of 6/7.
Cytokine Production

Macrophages were treated with compounds from E. purpurea or co-
treated with LPS and E. purpurea compounds to determine effects on cytokine
levels. Cytokine induction and/or suppression was determined by ELISA
quantification of lL-6 and TNF-a. Results of the cell assays are summarized in
Figures 5.5 — 5.16. None of the compounds examined demonstrated the ability
to directly stimulate RAW 264.7 murine macrophage cells to produce lL-6 or
TNF-a in the absence of LPS. Lack of induction without LPS stimulation

indicates that the compounds examined did not possess strong immune-

107

 

2.5 I I f f

 

5 2
8 El 0 nglmL LPS
g 15 * El 100 nglmL LPS
IE 1 * .1000 nglmL LPS
0
O l

0.5 i

o L _ _ _ l

 

 

 

 

 

 

 

 

 

Alkamide 1 (micrograms/mL)

Figure 5.1. MTT assay results for RAW 264.7 macrophages treated with 1, 10,
50, or 100 pg/mL of alkamide 1 from E. purpurea. Asterisks denote treatments
(n=3) that significantly differ (p < 0.05) from their respective controls (0, 100, or

1000 ng/mL LPS). Results are representative of two independent studies.

108

 

* * EI 0 nglmL LPS
El 100 nglmL LPS
I 1000 nglmL LPS

OD 570-690 nm
a:

 

 

0°
Alkamide 2 (micrograms/mL)

Figure 5.2. MTT assay results for RAW 264.7 macrophages treated with 1, 10,
50, or 100 pg/mL of alkamide 2 from E. purpurea. Asterisks denote treatments
(n=3) that significantly differ (p < 0.05) from their respective controls (0, 100, or

1000 ng/mL LPS). Results are representative of two independent studies.

109

I
l
i
l

[ ***I
o A“
o‘

Alkamide 4 (micrograms/mL)

I”
romeo

III 0 nglmL LPS
El 100 nglmL LPS
I 1000 nglmL LPS

OD 570-690 nm
A a:

O
0|

 

 

 

 

Figure 5.3. MTT assay results for RAW 264.7 macrophages treated with 1, 10,
50, or 100 pg/mL of alkamide 4 from E. purpurea. Asterisks denote treatments
(n=3) that significantly differ (p < 0.05) from their respective controls (0, 100, or

1000 nglmL LPS). Results are representative of two independent studies.

110

3.5 ~ ~
3
2.5
2
1.5
1
0.5
o

I] 0 nglmL LPS
I] 100 nglmL LPS
I 1000 nglmL LPS

OD 570-690 nm

 

 

\ N
o

Alkamides 6/7 (micrograms/mL)

Figure 5.4. MTT assay results for RAW 264.7 macrophages treated with 1, 10,
50, or 100 ug/mL of alkamides 6/7 from E. purpurea. Asterisks denote
treatments (n=3) that significantly differ (p < 0.05) from their respective controls
(0, 100, or 1000 ng/mL LPS). Results are representative of two independent

studies.

11]

 

 

 

 

 

 

 

r9 4
g 3 * E11000 nglmL LPS
3 « ‘ I 100 nglmL LPS
E 2 l
a n o LPS .
C 1 a t i
l
o , ND ND ND ND
50 10 1 Control
Alkamide 1 (microgramsImL)
r.

~25-——~-~~ - -- — ———-—-~
5
E
‘l'
u.
E a 1000 nglmL LPS
‘3 I 100 nglmL LPS
E El 0 LPS
U)
:

 

 

 

 

50 10 1 Control
Alkamide 1 (micrograms/mL)

Figure 5.5. Effects of co-treatments of 50, 10, or 1 pg/mL of alkamide 1 with
LPS on IL-6 and TNF-a production by RAW 264.7 cells. Asterisks denote
treatments (n=3) that significantly differ (p< 0.05) from their respective controls
(0, 100, or 1000 ng/mL LPS). ND = nondetectable. Results are representative of

two independent studies.

112

131000 nglmL LPS
I 100 nglmL LPS
Cl 0 LPS

nglmL of IL-6

 

 

 

10 1 Control

Alkamide 2 (micrograms/mL)

m 25 ;
.C .
2 20 l
‘l’ 1
IE 15 l
I- ] El 1000 nglmL LPS
'- . *
3 1° . I100 nglmL LPS
E 5 s DOLPS
U)
= o L. ND, 2 no- no, , Nil.
so 10 1 Control

Alkamide 2 (micrograms/mL)

Figure 5.6. Effects of co-treatments of 50, 10, or 1 pg/mL of alkamide 2 with
LPS on lL-6 and TNF—a production by RAW 264.7 cells. Asterisks denote
treatments (n=3) that significantly differ (p< 0.05) from their respective controls
(0, 100, or 1000 ng/mL LPS). ND = nondetectable. Results are representative of

two independent studies.

113

 

 

 

B1000 nglmL LPS
I 100 nglmL LPS

nglmL of lL-6
N

 

 

 

 

 

 

 

 

 

 

1 CI 0 LPS E..-
ND
0
50 10 1 Control
Alkamide 4 (microgramsImL)

2 30 *--~—~~~---~~--~-~~ _..-__....-__.._..__.___.-_.-._--_..--....._..._-_.
g. 25 1
‘l‘ 20 .
mg i 1000 nglmL LPS
.5 15 I 100 nglmL LPS
.5] 10 El 0 LPS
‘6:
:

 

 

 

 

 

 

 

5O 10 1 Control
Alkamide 4 (micrograms/mL)

Figure 5.7. Effects of co-treatments of 50, 10, or 1 pg/mL of alkamide 4 with
LPS on lL-6 and TNF-a production by RAW 264.7 cells. Asterisks denote
treatments (n=3) that significantly differ (p< 0.05) from their respective controls
(0, 100, or 1000 ng/mL LPS). ND = nondetectable. Results are representative of

two independent studies.

114

 

 

 

 

 

 

 

 

 

 

‘9
:36 2 a 1000 nglmL LPS
.8. * ' I 100 nglmL LPS
3, 1 4 no LPS |
C l
0 ‘ ND
50 1O 1 Control
Alkamides 6I7 (micrograms/mL)
g. 25 .
E 20 .
E 15 ”1000 nglmL LPS
"5 I 100 nglmL LPS
_, 10
E D 0 LPS
- 5
2’
0

 

1 Control

Alkamides 6/7 (micrograms/mL)

Figure 5.8. Effects of co-treatments of 50, 10, or 1 ug/mL of alkamides 617 with
LPS on lL-6 and TNF-a production by RAW 264.7 cells. Asterisks denote
treatments (n=3) that significantly differ (p< 0.05) from their respective controls

(0, 100, or 1000 nglmL LPS). ND = nondetectable. Results are representative of

two independent studies.

115

 

 

 

 

 

 

 

 

 

 

 

 

6 .,
3 5‘
,,-_ 4 ! D1000ng/mLLPS
3 3 I100 nglmL LPS
E
a 2 no LPS :-
C

1

O . .

100 50 10 1 Control
Compound 8 (micrograms/mL)

4o _
g 351
a 30
lzL 25 £31000 nglmL LPS
:5 :2 I 7‘ I 100 nglmL LPS

1 'l;-'
‘5' 1o ’ nouns
g 3 ND ND ND ND
100 50 10 1 Control

Compound 8 (micrograms/mL)

Figure 5.9. Effects of co-treatments of 100, 50, 10, or 1 pg/mL of cyanidin-3-O-
glucoside (8) with LPS on lL-6 and TNF-a by RAW 264.7 cells. Asterisks denote
treatments (n=3) that significantly differ (p< 0.05) from their respective controls
(0, 100, or 1000 nglmL LPS). ND = nondetectable. Results are representative of

two independent studies.

116

 

5 l

5
«e l
:' 4
3 j 131000 nglmL LPS

3
—E' i I 100 nglmL LPS
F» 2 [ no LPS
C 1 ‘

o l

Control
Compound 9 (micrograms/mL)

a 35
fa 30 ‘
£25 20 l D1000 nglmL LPS
I- I 100 nglmL LPS
h
0
.1 1:] o LPS
E1
a) 5
c l

0

Control

Compound 9 (microgramslmL)

Figure 5.10. Effects of co-treatments of 100, 50, 10, or 1 pg/mL of cyanidin-3-O—
malonyl-(1a6)—j3-D glucopyranoside (9) with LPS on lL-6 and TNF-a by RAW
264.7 cells. Asterisks denote treatments (n=3) that significantly differ (p< 0.05)
from their respective controls (0, 100, or 1000 ng/mL LPS). ND = nondetectable.

Results are representative of two independent studies.

117

nglmL of IL-6

5
T
4 i
3 l 1:: 1000 nglmL LPS
, I 100 nglmL LPS
2 l 1:: 0 LPS
1 l
0 ‘ ND ND ND ND ND
0 50 1o 1

10

Control

Compound 11 (micrograms/mL)

10 1

Compound 11 (micrograms/mL)

 

E] 1000 nglmL LPS
I 100 nglmL LPS
, El 0 LPS

nglmL of TNF-alpha

   

Control

Figure 5.11. Effects of co-treatments of 100, 50, 10, or 1 pg/mL of quercetin-3-
O-rutinoside (11) with LPS on lL-6 and TNF—a production by RAW 264.7 cells.
Asterisks denote treatments (n=3) that significantly differ (p< 0.05) from their
respective controls (0, 100, or 1000 ng/mL LPS). ND = nondetectable. Results

are representative of two independent studies.

118

ill-1n“; Mmfzmmm I

nglmLoflL-6
O A N (A h 0| m \l

f 01000 nglmL LPS
' I 100 nglmL LPS

I a o LPS

l

l ......

Control

Compound 12 (micrograms/mL)

I: 1000 nglmL LPS
I 100 nglmL LPS
0 0 LPS

Control

Compound 12 (micrograms/mL)

nglmL of TNF-alpha

 

Figure 5.12. Effects of co-treatments of 100, 50, 10, or 1 pg/mL of quercetin-3-
O-glucoside (12) with LPS on IL-6 and TNF-a production by RAW 264.7 cells.
Asterisks denote treatments (n=3) that significantly differ (p< 0.05) from their
respective controls (0, 100, or 1000 nglmL LPS). ND = nondetectable. Results

are representative of two Independent studies.

119

 

 

3 61000 nglmL LPS
"5 I 100 nglmL LPS
..I

E El 0 LPS

3

:

 

 

 

     

 

100 50 10 1 Control

Compound 13 (micrograms/mL)

 

 

 

 

g 100 * i
Q * é
a 80 4 ‘ , .
§ .1000 nglmL LPS
1: 6° I 100 nglmL LPS
0
_, 40 o o LPS
L; 20
C 0 ND ND ND ND ND
100 50 10 1 Control

Compound 13 (micrograms/mL)

Figure 5.13. Effects of co-treatments of 100, 50, 10, or 1 pg/mL of caffeic acid
(13) with LPS on IL-6 and TNF-a production by RAW 264.7 cells. Asterisks
denote treatments (n=3) that significantly differ (p< 0.05) from their respective
controls (0, 100, or 1000 nglmL LPS). ND = nondetectable. Results are

representative of two independent studies.

120

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3 _ M

7 i
3 61
"-6 5 - 93" 0111000 nglmL LPS
.51 g I100 nglmL LPS
C "-_.__:_ p- ——

1 ND

0 a".

100
Compound 14 (microgramsImL)
% 40 I. l.
g 30 " | _ I1ooo nglmL LPS
.."2 20 . I100 nglmL LPS
0
_. u o LPS
g 10
U
C
0

 

 

 

10 1 Control
Compound 14 (microgramsImL)

100

Figure 5.14. Effects of co-treatments of 100, 50, 10, or 1 pg/mL of caftaric acid
(14) with LPS on lL-6 and TNF-a production by RAW 264.7 cells. Asterisks
denote treatments (n=3) that significantly differ (p< 0.05) from their respective
controls (0, 100, or 1000 nglmL LPS). ND = nondetectable. Results are

representative of two independent studies.

121

 

 

 

 

 

 

8 I
3
;3 6
O
—l g t
g 4
a
C 2 1
0 ND ND T
100 50

Compound 15 (microgramsImL)

 

ND ND

 

ND

 

 

10 1

Control

 

nglmL of TNF-alpha

 

 

50

Figure 5.15. Effects of co-treatments of 100, 50, 10, or 1 pg/mL of cichoric acid
(15) with LPS on IL-6 and TNF-a production by RAW 264.7 cells. Asterisks

denote treatments (n=3) that significantly differ (p< 0.05) from their respective

controls (0, 100, or 1000 ng/mL LPS).

 

 

10 1

representative of two independent studies.

122

  

Control

Compound 15 (micrograms/mL)

ND = nondetectable.

 

£31000 nglmL LPS
I 100 nglmL LPS
[3 O LPS

E11000 nglmL LPS
I 100 nglmL LPS
E] 0 LPS

‘ I :31;

 

Results are

B1000 nglmL LPS
I 100 nglmL LPS
:1 o LPS

ND

 

nglmLoflL-6
O -i N (A :5 0| m

 

 

 

 

 

 

 

100 50 10 1 Control

Compound 16 (micrograms/mL)

 

so . ‘

5° ‘ E1000 nglmL LPS

I 100 nglmL LPS
0 0 LPS

30
20 4
10 1

nglmL of TNF-alpha

 

 

 

 

 

 

 

 

W ND W ND

 

 

 

100 50 10 1 Control
Compound 16 (micrograms/mL)

Figure 5.16. Effects of co-treatments of 100, 50, 10, or 1 pg/mL of chlorogenic
acid (16) with LPS on lL-6 and TNF-a production by RAW 264.7 cells. Asterisks
denote treatments (n=3) that significantly differ (p< 0.05) from their respective
controls (0, 100, or 1000 ng/mL LPS). ND = nondetectable. Results are

representative of two independent studies.

123

stimulating activity. Several compounds did however reduce the levels of lL—6
and TNF-a induced by LPS stimulation of macrophages, with the alkamides most
effective at reducing cytokine levels.

Compound 1, when tested at concentrations of 1, 10, and 50 pg/mL,
significantly decreased lL-6 and TNF-a production in cells co-treated with 100
and 1000 ng/mL LPS. When macrophages were stimulated with 100 nglmL LPS
and treated with only 1 ug/mL of compound 1, significant reductions in lL-6 were
noted. TNF-oc levels were significantly lower for all doses of compound 1 in cells
stimulated with 100 and 1000 nglmL LPS. Alkamide 2, structurally similar to
compound 1, showed significant reductions in lL-6 and TNF-a in LPS stimulated
cells (Figure 5.6). lL-6 levels were reduced following co-treatment of compound
2 at 1, 10, and 50 pg/mL with 100 nglmL LPS, and at 50 pg/mL for cells treated
with 1000 nglmL LPS. TNF-a was significantly reduced only in cells stimulated
with 100 ng/mL LPS and treated with 50 pg/mL of compound 2.

At 50 pg/mL, the inseparable mixture of compounds 6 and 7 exhibited a
significant reduction in lL-6 levels produced by macrophages co-treated with 100
nglmL LPS. Reduction in lL—6 levels were also observed for compounds 6 and 7
at 1, 10, and 50 pg/mL when macrophages were stimulated with 1000 nglmL
LPS (Figure 5.8). TNF-a levels were reduced with 10 and 50 ug/mL of

compounds 6/7 in cells stimulated with 100 and 1000 nglmL LPS, respectively.

In contrast to the results obtained for compounds 1, 2, and 6/7, alkamide 4 did

not significantly reduce IL-6 or TNF-a levels in LPS stimulated cells (Figure 5.7).

124

"j m!

 

The anthocyanins, compounds 8 and 9, also failed to significantly reduce
levels of IL-6 produced following stimulation by 100 or 1000 nglmL LPS (Figures
5.9-5.10). Furthermore, TNF-a levels were not significantly altered by compound
8; however, compound 9 reduced levels of TNF-a with co-treatments of 100
ng/mL LPS. The flavonoids, compounds 11 and 12, did not significantly reduce
lL-6 levels with stimulation by 100 or 1000 ng/mL LPS (Figures 5.11-5.12).
Compound 11,at doses of 10, 50, and 100 pg/mL, reduced TNF-a levels with co-
treatments of 100 and 1000 nglmL LPS. Compound 12 did not affect TNF-a
production.

None of the caffeoyl derivatives examined reduced levels of lL-6 in LPS
stimulated macrophages (Figures 5.13-5.16); however, co-treatments of
compound 13 (100 ug/mL), compound 15 (50 and 100 ug/mL), or compound 16
(1, 10, 50, and 100 ug/mL) with 100 or 1000 ng/mL LPS significantly increased
levels of lL-6 (Figures 5.15-5.16). TNF-a levels were similarly affected, with
significant induction noted for compound 13 (50 and 100 ug/mL, 100 and 1000
nglmL LPS), compound 14 (50 and 100 ug/mL, 1000 nglmL LPS), and
compound 16 (100 pg/mL, 1000 nglmL LPS) (Figures 5.13-5.14, Figure 5.16).
No detectable induction of lL-6 or TNF-a occurred without LPS stimulation for
compounds 13-16.

Echinacea preparations are generally considered to be safe due to the
lack of reported widespread problems associated with their consumption
(Parnham, 1996). In previously reported studies, the expressed juice of E.

purpurea whole plant did not demonstrate in vivo acute or subacute toxicity in

125

 

rats or mice, with no relevant differences noted in weight, food consumption, or
organ changes upon necropsy (Mullins, 1998). Previous examination of caffeic
acid in mouse L-929 cells indicated that cell growth and DNA metabolism were
adversely affected only at doses of 500 ug/mL (Cheminat et al., 1988). The
alkamides from E. purpurea have demonstrated mosquitocidal activity (Clifford et
al., 2002); however, the toxicity of these alkamides has not been evaluated in
mammalian cell culture assays.

The high dose of alkamides examined in this study, 100 ug/mL, adversely
affected RAW 264.7 macrophages based upon the MTT assay results.
Echinacea alkamides are known to be unstable in raw materials and in root
extracts (Kim et al., 2000a; Perry et al., 2000; Perry et al., 1999). This instability
should be considered when drawing conclusions regarding the safety of
Echinacea supplements. Analyses of commercially available Echinacea extracts
have determined that alkamides are present at very low levels. The cytotoxicity
noted with the 100 pg/mL dose in this study indicates that alkamide-containing
extracts could pose a safety concern; however, this dose might be too high to be
relevant based upon the instability and low concentration of alkamides present in
commercially available herbal supplements.

Anecdotal claims indicate that Echinacea extracts demonstrate both anti-
inflammatory and immune enhancing activities. Reports of headache and fever
relief from root extracts suggest an anti-inflammatory effect. Conversely, the use
of Echinacea to prevent and/or reduce the duration of respiratory tract infections

indicates that extracts might cause an inflammatory response. A number of

126

 

 

 

previously reported studies have supported both anti-inflammatory and
inflammatory activities of Echinacea extracts.

Adverse reactions noted in the oral administration and injections of
Echinacea extracts (headache, nausea, fatigue, and tachychardia) (Borchers et
al., 2000; Miller, 1998; Jurcic et al., 1989) suggest an inflammatory response in
vivo. The cytokine-inducing properties of Echinacea extracts and compounds
isolated from Echinacea have been evaluated to some extent. Treatment of
RAW 264.7 murine macrophages with crude Echinacea extracts found that E.
purpurea root and herb materials possessed the ability to activate macrophages
to produce lL—6 and TNF-a (Rininger et al., 2000). Previous reports of E.
purpurea and E. angustifolia extracts standardized to 4% phenolic content
(chlorogenic acid and cichoric acid) indicated that these extracts did not exhibit
macrophage activating activity. Chlorogenic acid, analyzed in addition to the
standardized extracts, was not found to induce macrophages to produce lL—6 or
TNF-a (Rininger et al., 2000). Quercetin, chlorogenic acid, and caffeic acid have
been reported to have no effect on LPS stimulated secretion of lL-6 or TNF-a in
human whole blood (Obertreis et al., 1996).

In this study, RAW 264.7 murine macrophage cells were not stimulated by
compounds from E. purpurea to produce cytokines, as determined by the ELISA
quantification of lL-6 and TNF-a. The lack of induction in the absence of LPS
stimulation indicates that the alkamides, flavonoids, and anthocyanins did not
possess strong immune-stimulating activity. Interestingly, the caffeoyl derivatives

demonstrated increased lL-6 and/or TNF-oc production when co-treated with 100

127

 

. ‘-"-L

 

or 1000 nglmL LPS. These results differ from previous reports of crude
Echinacea extracts, caffeic acid, and chlorogenic acid on cytokine induction. The
cell type used in this assay, the duration of exposure to test compounds, and
concentrations assayed could have contributed to the opposite results obtained
in this study. Caffeic acid (13) administered with 1000 nglmL LPS afforded levels
of TNF-oc greater than the other caffeoyl derivatives. While an inflammatory
response may not necessarily be beneficial, the ability of caffeic acid to enhance
the inflammatory response in this model system could explain the effectiveness
of Echinacea extracts in the reduction of upper respiratory tract infection severity
in clinical trials.

With regard to anti-inflammatory activity of Echinacea, alkamides from E.
angustifolia have previously demonstrated significant in vitro cyclooxygenase-1
inhibitory activity (MUIIer-Jakic et al., 1994). The ability of E. purpurea alkamides
to inhibit cyclooxygenase-2 (Clifford et al., 2002), the enzyme inducible upon
response to inflammation (Lipsky, 1999), suggests that alkamides might be
useful agents in treating inflammation and supports the anecdotal claims of pain
and fever relief (Bauer and Wagner, 1991). Caffeic acid has also demonstrated
the potential to act as an anti-inflammatory agent. The ability of caffeic acid to
prevent the release of lL-6 and TNF-a has been suggested (Bertelli et al.,
2002a), and caffeic acid and tyrosol (from white wine), when co—administered
with LPS in human peripheral blood mononuclear cells, synergistically inhibited

the release of cytokines (Bertelli et al., 2002b).

128

 

 

 

The alkamides demonstrated the greatest anti-inflammatory activity of the
compounds tested, with activity noted at the lowest concentration assayed (1
ug/mL). The results of this assay agree with reported anti-inflammatory in vitro
activity of E. purpurea and E. angustifolia alkamides. With regard to clinical
studies, Echinacea root extracts are most often administered. Extracts that
contain significant concentrations of alkamides might reduce symptoms
(headache, fever, pain relief) compared to placebo, resulting in reductions in the
severity of symptoms reported by study subjects.

The disparity reported in clinical studies can be ascribed to the chemical
composition of the extract used, the species of Echinacea from which the extract
was prepared, and the plant portion extracted. The chemical profiles of
Echinacea extracts vary greatly, and extracts that posses greater amounts of
anti-inflammatory compounds, such as the alkamides, might reduce or negate
any immune-enhancing activity from other compounds present in the extract.
Examination of specific constituents from E. purpurea, as in this study, more
accurately portrays the potential effectiveness of Echinacea as an herbal
remedy.

Based upon the results obtained in this study, it is conceivable that the
compounds present in commercial Echinacea-based phytoceuticals may elicit a
beneficial anti-inflammatory effect in humans. Alkamides from E. purpurea
exhibited strong anti-inflammatory activity through the reduction of IL-6 and TNF-

or produced. These results support the anecdotal claims of pain and fever relief

129

 

associated with Echinacea use, and provide evidence for the efficacy of E.

purpurea as an herbal supplement.

130

 

CONCLUSIONS

Echinacea is currently one of the most popular herbal supplements used
in the United States. Anecdotal claims of immune enhancement, fever relief, and
alleviation of pain have been associated with the use of E. purpurea
preparations, suggesting that Echinacea extracts possess both anti-inflammatory
and immune enhancing properties. In order to establish the validity of these
anecdotal claims, the efficacy of several compounds isolated from E. purpurea
was evaluated in antioxidant, anti-inflammatory, and cytokine-induction model
systems.

Five primary objectives were defined with regard to the isolation, structure
elucidation, and biological testing of active compounds from E. purpurea roots
and flowers. These objective were 1) isolate and compare COX-1 and COX-2
inhibitory activities of alkamides from E. purpurea roots; 2) evaluate the
mosquitocidal activity of E. purpurea root alkamides; 3) isolate flavonoids and
anthocyanins from E. purpurea flowers; 4) characterize the COX-1 and COX-2
inhibitory activities and antioxidant activities of the flavonoids and anthocyanins
from E. purpurea flowers; and 5) characterize and examine the effects of purified
constituents from E. purpurea roots and flowers (alkamides, flavonoids, and
anthocyanins) on cytokine production.

The first objective was accomplished by the isolation and testing of
alkamides from E. purpurea roots. Seven alkamides, isolated from

dichloromethane extracts of E. purpurea roots, were examined for

131

 

cyclooxygenase-1 (COX-1) and cyclooxygenase-2 (COX-2) inhibitory activities.
Undeca-2E,4Z-dien-8,10-diynoic acid isobutylamide, undeca-ZZ,4E—dien-8,10-
diynoic acid isobutylamide, dodeca-2E,4Z—dien-8,10-diynoic acid isobutylamide,
undeca-2E,4Z-dien-8,10-diynoic acid 2-methylbutylamide, dodeca-2E,4Z-dien-
8,10-diynoic acid 2-methylbutylamide, and a mixture of dodeca-2E,4E,8Z,10E—
tetraenoic acid isobutylamide and dodeca-2E,4E,82,102-tetraenoic acid
isobutylamide were isolated from E. purpurea roots. At a concentration of 100
pg/mL, several E. purpurea alkamides inhibited COX-1 and COX-2 enzymes in
the range of 36-60% and 15-46%, respectively, when compared to controls. The
alkamides did not exhibit selective inhibition against either COX-1 or COX-2.

The second objective was to examine other biological activity of E.
purpurea alkamides. Alkamides, as a class of compounds, are noted for
mosquitocidal activity; however, the mosquitocidal activity for alkamides isolated
from E. purpurea had not been previously established. Several E. purpurea
alkamides demonstrated mosquitocidal activity at 100 pg/mL, with 100% mortality
against Aedes eegyptii L. larvae noted. Evaluation of the activity against
mosquito larvae established the potential for E. purpurea alkamides to be used
as mosquitocidal agents.

The biological activity of root extracts from Echinacea species have been
investigated; however, the flowers have not been fully evaluated. In order to
identify bioactive constituents from this underutilized part of E. purpurea, the third
objective was to isolate and identify constituents from E. purpurea flowers for

later analysis of their biological bioactive properties. This study led to the

132

 

 

 

extraction, isolation, and structure elucidation of three flavonoids (quercetin-3-O—
glucoside, quercetin-3-O—rutinoside, and kaempferol-3-O-robinobioside) and two
anthocyanins (cyanidin-3-O-j3-D glucopyranoside and cyanidin-3-O-malonyl-
(1—96)-B-D glucopyranoside) from methanolic extracts of Iyophilized E. purpurea
flowers.

Flavonoids (quercetin-3-O—glucoside, quercetin-3-O-rutinoside, and
kaempferol-3-O-robinobioside) and anthocyanins (cyanidin-3-O-[3-D
glucopyranoside and cyanidin-3-O—malonyl-(1—>6)-B-D glucopyranoside), isolated
from methanolic extracts of Iyophilized E. purpurea flowers, and caffeoyl acid
derivatives, caffeic acid, caftaric acid, chlorogenic acid, and cichoric acid, were
examined for antioxidant and COX-1 and COX-2 inhibitory activities. The
anthocyanins demonstrated the greatest cyclooxygenase inhibitory activities at
100 ug/mL, with inhibition of COX-1 and COX-2 ranging from 26-28% and 23-
26%, respectively. When assayed at a concentration of 10 pg/mL, the
antioxidant activity of these compounds ranged from 33-81% for the
anthocyanins, 3-54% for the flavonoids, and 35-79% for the caffeoyl derivatives.

The fourth objective was to examine the cyclooxygenase inhibitory and
antioxidant activities of compounds from E. purpurea flowers. Currently, E.
purpurea flowers are included in extracts of the fresh herb only if the flower is in
bloom when the above-ground portion of the plant is harvested. The promising
results obtained in this study indicate that extracts prepared from E. purpurea

flowers might provide a concentrated source of bioactive compounds, and further

133

 

studies regarding the development of an E. purpurea flower-based natural
product supplement is warranted.

Previous evaluation of the effects of crude E. purpurea extracts on LPS-
stimulated RAW 264.7 murine macrophages revealed their ability to induce
cytokines to produce lL-6 and TNF-a. The fifth objective was to determine the
effects of cytokine production in murine macrophage cells (RAW 264.7) co-
treated with compounds isolated from E. purpurea and lipopolysaccharide (LPS),
an inflammagenic stimulus. An assessment of the cytotoxicity of the doses
tested in the macrophage assay determined that the highest concentration of

alkamides used, 100 pg/mL, adversely affected cells.

No induction of lL-6 or TNF-a occurred without LPS stimulation in
macrophages treated with test compounds. The alkamides examined in the
assay were highly effective at reducing LPS-stimulated cytokine induction when
compared to controls. The anti-inflammatory activity of alkamides in the
macrophage assay supports the anecdotal claims of pain and fever relief
associated with the use of Echinacea roots, from which the alkamides were
isolated.

An increase was noted in IL-6 and TNF-a production when caffeoyl
derivatives were co-treated with LPS, suggesting the potential for immune-
enhancing properties of Echinacea extracts. These results demonstrate that
individual compounds from E. purpurea might possess opposing activities with

regard to immune stimulation and/or suppression. This might explain why the

134

immune-enhancing activity of Echinacea extracts as reported in the literature
varies widely.

The results of the fifth objective indicate that examining single constituents
from Echinacea extracts, rather than crude extracts, provides more detailed
information regarding cytokine induction and/or suppression in LPS-stimulated
macrophages. The alkamides demonstrated striking anti-inflammatory activity in
the model system, while conversely, the caffeoyl derivatives in combination with
LPS resulted in the induction of cytokines. The effectiveness of crude
preparations of Echinacea will vary depending upon the qualitative and
quantitative chemical profile of the extract.

Alkamides, flavonoids, anthocyanins, and caffeoyl derivatives from E.
purpurea demonstrated antioxidant, cyclooxygenase-inhibitory, mosquitocidal
and cytokine induction and/or suppression activities. Echinacea supplements are
already among the most popular herbal supplements on the market, and these
findings help give credence to their continued use. Further studies are needed in
order to further define the bioactivity profiles of Echinacea-derived constituents.
Ultimately, this research will contribute to the development of new Echinacea-
based phytoceuticals tailored to provide consistent, safe, and effective

pharmacological activities.

135

 

Ba

Ba

LITERATURE CITED

Agrawal, P. K.; Mahesh, C. B. Flavonoid Glycosides. In: Carbon-13 NMR of
Flavonoids; Agrawal, P. K., Ed.; Elsevier: Amsterdam, 1989, pp. 283-364.

Akashi, T.; Saito, N.; Hirota, H.; Ayabe, S.-l. Anthocyanin-producing dandelion
callus as a chalcone synthase source in recombinant polyketide reductase
assay. Phytochemistry 1997, 46, 283-287.

Arora, A.; Nair, M. G.; Strasburg, G. M. Structure-activity relationships for
antioxidant activities of a series of flavonoids in a liposomal model system.
Free Radical Biol. Med. 1998, 9, 1355-1363.

Barrett, 8.; Vohmann, M.; Calabrese, C. Echinacea for upper respiratory
infection. J Family Med. 1999, 48, 628-635.

Bauer, R. Echinacea: Biological effects and active principles. In:
Phytomedicines of Europe: Chemistry and Biological Activity. Lawson, L. 0.;
Bauer, R., Eds. American Chemical Society: Washington, DC, 1998; PP. 140-
157.

Bauer, R.; Khan, l. A.; Wagner, H. TLC and HPLC analysis of Echinacea pallida
and E. angustifolia roots. Planta Med. 1988a, 54, 426-430.

Bauer, R.; Reminger, P. TLC and HPLC analysis of alkamides in Echinacea
species. Planta Med. 1989, 55, 367-371.

Bauer, R.; Reminger, P.; Wagner, H. Alkamides from the roots of Echinacea
purpurea. Phytochemistry 1988b, 27, 2339-2342.

Bauer, R.; Remiger, P.; Wagner, H. Echinacea — Vergleichende DC — und
HPLC-Analyse der Herba-drogen von Echinacea purpurea, E. pallida, und E.
angustifolia. Deutsch. ApothekerZeitung. 1988c, 128,174-180.

Bauer, R.; Remiger, P.; Wray, V.; Wagner, H. A germacrene alcohol from fresh
aerial parts of Echinacea purpurea. Planta Med. 1988d, 54, 478-479.

Bauer, R.; Foster, S. Analysis of alkamides and caffeic acid derivatives from
Echinacea simulate and E. paradoxa roots. Planta Med. 1991, 57, 447-449.

Bauer, R.; Wagner, H. Echinacea species as potential immunostimulatory drugs.

In: Economic and Medicinal Plant Research; Wagner, H., Farnsworth, N. R.,
Eds.; Academic Press: London, UK, 1991; Vol 5, pp. 253-321.

136

 

Be

Bir

Blu

Bor

Bra

Bra

Brid

Brin

Bergeron, C.; Livesey, J. F.; Awang, D. V. C.; Arnason, J. T.; Rana, J.; Baum, B.
R.; Letchamo, W. A quantitative HPLC method for the quality assurance of
Echinacea products on the North American market. Phytochem. Anal. 2000,
11, 207-215.

Bertelli, A; Migliori, M.; Bertelli, A. A.; Origlia, N.; Filippi, C.; Panichi, V.; Falchi,
M.; Giovannini, L. Effect of some white wine phenols in preventing
inflammatory cytokine release. Drugs Exp. Clin. Res. 2002a, 28, 11-15.

Bertelli, A. A. E.; Migliori, M.; Panichi, V.; Longoni, B.; Origlia, N.; Ferretti, a.;
Cuttano,l M. G.; Giovannini, L. Oxidative stress and inflammatory reaction

modulation by white wine. Annals of the New York Academy of Sciences
2002b, 957, 395-301.

Binns, S. E.; Purgina, B.; Bergeron, C.; Smith, M. L.; Ball, L.; Baum, B. R.;
Arnason, J. T. Light-mediated antifungal activity of Echinacea extracts.
Planta Med. 2000, 66, 241-244.

Blumenthal, M.; Gruenwald, J.; Hall, T. German commission E monographs:
medicinal plants for human use. Austin: American Botanical Council, 1998.

Borchers, A. T.; Keen, C. L.; Stern, J. S.; Gershwin, M. E. Inflammation and
Native American medicine: The role of botanicals. Am. J. Clin. Nutr. 2000,
72, 339-347.

Braunig, B.; Dorn, M.; Knick, E. Therepeutical experiences with Echinacea
pallida for influenza-like infections. Naturhei/praxis 1993, 1, 72-75.

Braunig, B.; Knick, E. Echinacea purpurea radix for the enhancement of the
body’s own immune defense mechanisms in influenza-like infections.
Zeitschn'ft fur Phytotherapie 1992, 13, 7-13.

Bridle, P.; Loeffler, R. S. T.; Timberlake, C. F.; Self, R. Cyanidin 3-
mannngIucoside in Cichorium intybus. Phytochemistry 1984, 23, 2968-
2969.

Brinkeborn, R.; Shah, D.; GeissbiJhler, S.; Degenring, F. H. Echinaforce in the
treatment of acute colds. Schweiz Zschr GanzheitsMedizin 1998, 10, 26-9.

Broadhurst, C. L. Nutrition and non-insulin dependent diabetes from an
anthropological perspective. Alt. Med. Rev. 1997, 2, 378-399.

Broadhurst, C. L.; Polansky, M. M.; Anderson, R. A. Insulin-like biological activity
of culinary and medicinal plant aqueous extracts in vitro. J. Agric. Food
Chem. 2000, 48, 849-852.

137

 

Brown, I
Pacif

Cheminz
malo
3247

Chemin:
Echi
27, 2

Childs,
Nutr
LCC

Clifford
iron

Cryer, l
— a

Denizc
MOI
reli;

Dorn,

Brown, E. Some garden daisies and sunflowers. Pac-Hort. San Franciso:
Pacific Hort. Foundation. 1986, 47, 24-28.

Cheminat, A; Brouillard, R.; Guerne, P; Bergmann, P.; Rether, B. Cyanidin 3-
malonylglucoside in two Echinacea species. Phytochemistry 1989, 28, 3246-
3247.

Cheminat, A.; Zawtzky, R.; Becker, H.; Brouillard, R. Caffeoyl conjugates from
Echinacea species: structures and biological activity. Phytochemistry 1988,
27, 2787-2794.

Childs, N. M. Marketing issues for functional foods and nutraceuticals. In:
Nutraceuticals and Functional Foods. Wildman, R. E. C., Ed.; CRC Press
LCC: Boca Raton, 2001; pp. 517-528.

Clifford, L. J.; Nair, M. G.; Rana, J.; Dewitt, D. L. Bioactivity of alkamides isolated
from Echinacea purpurea (L.) Moench. Phytomedicine. 2002, 9, 249-253.

Cryer, B.; Dubois, A. The advent of highly selective inhibitors of cyclooxygenase
— a review. Prostaglandins and other lipid mediators 1998, 56, 341-361.

Denizot, F.; Lang, R. Rapid colorimetric assay for cellular growth and survival.
Modification to the tetrazolium dye procedure giving improved sensitivity and
reliability. J. Immunol. Methods 1986, 89, 271.

Dorn, M. Milerung grippaler Effeckte durch ein pflanzliches Immunstimulans.
Nutur Ganzheitsmedizin 1989, 2, 314-9.

Facino, R. M.; Carini, M.; Aldini, G.; Saibene, L.; Pietta, P.; Mauri, P.
Echinacoside and caffeoly conjugates protect collagen from free radical-
induced degradation: A potential use of Echinacea extracts in the prevention
of skin photodamage. Planta Med. 1995, 61: 510-514.

Ford-Hutchinson, A. W. New highly selective cyclooxygenase-2 inhibitors. In:
New targets of inflammation: Inhibitors of COX-2 or adhesion molecules.
Bazan, N., Botting, J., Vane, J., Eds; Kluwer Academic: Dordrecht, 1996; pp.
55-62.

Fossen, T.; Andersen, Q. Cyanidin 3-0-(6”-succinyI-[3-glucopyranoside) and
other anthocyanins from Phragmites australis. Phytochemistry 1998, 49,
1065-1068.

Foster, 8.; Duke, J. A. A field guide to medicinal plates: Eastern and central
North America. Houghton Mifflin Company: Boston, 1990, p. 200.

138

 

Frankel.
Antio
Yosh

Gleason
State
New

Glownia
phas
Echi

Greger,
Plar

Grimm,
extr
res;

Gunnin
infe

Hartze
datr

He. X.
anc
ma

HUME
Lil

Frankel, E.N. Natural phenolic antioxidants and their impact on health. In:
Antioxidant food supplements in human health. Packer, L., Hiramatsu, M.,
Yoshikawa, T., Eds.; Academic Press: San Diego, 1999.

Gleason, H. A.; Conquist, A. Manual of vascular plants of northeastern United
States and adjacent Canada, 2nd ed. The New York Botanical Garden: Bronx,
New York, 1991; pp. 531-532.

Glowniak, K.; Zgorka, G.; Kozyra, M. Solid-phase extractions and reversed-
phase high-performance liquid chromatography of free phenolic acids in some
Echinacea species. J Chromatogr. A. 1996, 730, 25-29.

Greger, H. Alkamides: Structural relationship, distribution and biological activity.
Planta Med. 1984, 50, 366-375.

Grimm, W.; M'Liller, H.-H. A randomized controlled trial of the effect of fluid
extract of Echinacea purpurea on the incidence and severity of colds and
respiratory infections. Am. J. Med. 1999, 106, 138-143.

Gunning, K. Echinacea in the treatment and prevention of upper respiratory tract
infections. WJM. 1999, 171,198-200.

Hartzell, A. Plant products for insecticidal properties and summary of results to
date. Contribs. Boyce Thompson Inst. 1947, 15, 21-34.

He, X.-G.; Lin, L.-Z.; Bernart, M. W.; Lian, L.-Z. Analysis of alkamides in roots
and achenes of Echinacea purpurea by liquid chromatography-electrospray
mass spectrometry. J. Chromatogr. A. 1998, 815, 205-211.

Henneicke-von Zepelin, H. H.; Hentschel, C.; Schnitker, J.; Kohnen, R.; Kohler,
G.; Wastenberg, P. Efficacy and safety of a fixed combination phytomedicine
in the treatment of the common cold (Acute viral respiratory tract infection):
results of a randomised, double blind, placebo controlled, multicentre study.
Curr. Med. Res. Opin. 1999, 15, 214-227.

Henry, C. M. Nutraceuticals: fad or trend? C&EN 1999, Nov. 9, 42—47.
Hohesisel, 0.; Sandberg, M.; Bertram, 8.; Bulitta, M.; Schafer, M. Echinagard
treatment shortens the course of the common cold: a double-blind, placebo-

controlled clinical trail. Eur J. Clin Res. 1997, 9, 261-8.

Hu, 0.; Kitts, D. D. Studies on the antioxidant activity of Echinacea root extract.
J. Agric. Food Chem. 2000, 48, 1466-1472.

Hunter, C. G. Wildflowers of Arkansas, 4"1 ed. The Ozark Society Foundataion:
Little Rock, Arkansas, 1995; pp. 238-239.

139

 

Ji, G. E.
hydrc
the t'
214.

Jondiko
1986

Juric, l<
der
(T w

prel

Kim, H
drir

Lan

Lei

Ji, G. E.; Park, S. Y.; Wong, S. S.; Pestka, J. J. Modulation of nitric oxide,
hydrogen peroxide and cytokine production in a clonal macrophage model by
the trichothecene vomitoxin (deoxynivalenol). Toxicology. 1998, 125, 203-
214.

Jondiko, l. J.O. A mosquito larvicide in Spilanthes maun'tiana. Phytochemistry
1986, 25, 2289-2290.

Juric, K.; Melchart, D.; Holzmann, M. Zwei Probandenstudien zur Stimulierung
der Granulozyten-Phagozytose durch Echinacea-extrakt-haltige praparate.
(Two clinical studies on the stimulation of granulocyte phagocytosis by
preparations containing Echinacea.) Z. Phytother. 1989, 10, 67-70.

Kim, H.-O.; Durance, T. D.; Scaman, C. H.; Kitts, D. D. Retention of alkamides in
dried Echinacea purpurea. J. Agric. Food Chem. 2000a, 48, 4187-4192.

Kim, H.-O.; Durance, T. D.; Scaman, C. H.; Kitts, D. D. Retention of caffeic acid
derivatives in dried Echinacea purpurea. J. Agric. Food Chem. 2000b,48,
4182-4186.

Kim, J. H.; Nonaka, G.-l.; Fujieda, K.; Uemoto, S. Anthocyanidin
malonylglucosides in flowers of Hibiscus syriacus. Phytochemistry 1989, 28,
1503-1506.

Kindscher, K. Ethnobotany of purple coneflower (Echinacea angustifolia,
Asteraceae) and other Echinacea species. Econ. Bot. 1989, 43, 498-507.

Koch, H. P.; Lawson, L. D. Garlic: The science and therapeutic applicaion of
Allium sativum L.,' Related species; Williams and Williams: Baltimore, MD,
1996.

Laneuville, 0.; Breuer, D. K.; DeWitt, D. L.; Hla, T.; Funk, C. D.; Smith, W. L.
Differential inhibition of human prostaglandin endoperoxidase H synthases-1

and —2 by nonsteroidal anti-inflammatory drugs. J. Phannacol. Exp. Ther.
1994, 271, 927-934.

Lersch, C.; Zeuner, M.; Bauer, A.; Siebenrock, K.; Hart, R.; Wagner, F.; Fink, U.;
Dancygier, H.; Classen, M. Stimulation of the immune response in
outpatients with hepatocellular carcinomas by low doses of
cyclophosphamide (LDCY), Echinacea purpurea ext4racts (Echinacin) and
thymostimulin. Arch. Geschwulstforsch 1990, 60, 379-383.

140

 

Lerscr
Da
cyc
(Er
res

Li, T.
12
Li, T.
gir
an
Inc

Lienel

Lersch, C.; Zeuner, M.; Bauer, A.; Siemens, M.; Hart, R.; Drescher, M.; Fink, U.;
Dancygier, H.; Classen, M. Nonspecific immunostimulation with low doses of
cyclophosphamide (LDCY), thymostimulin and Echinacea purpurea extracts
(Echinacin) in patients with far advanced colorectal cancers: Preliminary
results. Cancer-Invest. 1992, 10, 343-348.

Li, T. S. C. Echinacea: Cultivation and medicinal value. Herbology. 1998, 8,
122-129.

Li, T. S. C.; Wang, L. C. H. Physiological components and health effects of
ginseng, Echinacea, and sea buckthorn. In: Functional foods: Biochemical
and processing aspects. Ed. G. Mazza. Technomic Publishing Company,
Inc., Lancaster, Pennsylvania, pp. 329-356, 1998.

Lienert, D.; Anklam, E.; Panne, U. Gas Chromatography-mass spectral analysis
of roots of Echinacea species and classification by multivariate data analysis.
Phytochem. Anal. 1998, 9, 88-98.

Lipisky, P. E. The clinical potential of cyclooxygenase-Z-specific inhibitors. Am.
J. of Med. 1999, 106, 518-578.

Loll, P. J. Structure of prostaglandin H2 synthase-1 (COX-1) and its NSAID
binding sites. In: New targets of inflammation: Inhibitors of COX-2 or
adhesion molecules. Bazan, N., Botting, J., Vane, J., Eds.; Kluwer Academic:
Dordrecht, 1996; pp. 13-21.

Luczkiewcz, M; Cisowski, W. Optimisation of the second phase of a two phase
growth system for anthocyanin accumulation in callus cultures of Rudbeckia
hirta. Plant Cell Tiss. Org. Cult. 2001, 65, 57-68.

Luettig, B.; Steinmtiller, C.; Gifford, G.E.; Wagner, H.; Lohmann-Matthes, M.-L.
Macrophage activation by the polysaccharide arabinogalactan isolated from
plant cell cultures of Echinacea purpurea. J. Nat. Cancer Inst. 1989, 81, 669-
675.

Markham, K. R.; Chari, V. M.; Mabry, T. J. Carbon-13 NMR spectroscopy of
flavonoids. In: The Flavonoids: Advances in Research; Harborne, J. B.,
Mabry, T. J., Eds.; Chapman and Hall: New York, 1982; pp 19-134.

Marshall, N. J.; Goodwin, C. J.; Holt, 8. J. A critical assessment of the use of
microculture tetrazolium assays to measure cell growth and function. Growth
Regulation 1995, 5, 69-84.

Mazza, G.; Cottrell, T. Volatile components of roots, stems, leaves, and flowers
of Echinacea species. J. Agric. Food Chem. 1999, 47, 3081-3085.

141

 

Mull

Mye

Nes

n1

NUS:

Obe

am<

Meade, E. A.; Smith, W. L.; DeWitt, D. L. Differential inhibition of prostaglandin
endoperoxide synthase (cyclooxygenase) isozymes by aspirin and other non-
steroidal anti-inflammatory drugs. J Biol. Chem. 1993, 268, 6610-6614.

Melchart, D.; Linde, K.; Worku, F.; Sarkady, L.; Holzman, M.; Juric, K.; Wagner,
H. Results of five randomized studies on the immunomodulatory activity of
preparations of Echinacea. J. Altemat. Complement. Med. 1995, 3, 95-102.

Melchart, D.; Walther, E.; Linde, K.; Brandmaier, R.; Lersch, C. Echinacea root
extracts for the prevention of upper respiratory tract infections: a double-
blind, placebo-controlled randomized trial.. Arch. Fam. Med. 1998, 7, 541-
545.

Mengs, U.;C|are, C. B.; PoiIey, J. A. Toxicity of Echinacea purpurea. Arzneim.-
Forsch. 1991, 41, 1076-1081.

Miller, L.G. Herbal medicinals: Selected clinical considerations focusing on
known or potential drug-herb interactions. Arch. Intern. Med. 1998, 158,
2200-2211.

Mosmann, T. Rapid colorimetric assay for cellular growth and survival:
application to proliferation and cytotoxicity assays. J. Immunol. Methods
1983, 65, 55-63.

MUller-Jakic, B.; Breu, W.; Probstle, A.; Redl, K.; Greger, H.; Bauer, R. In vitro
inhibition of cyclooxygenase and 5-Iipoxygenase by alkamides from
Echinacea and AchiI/ea species. Planta Med. 1994, 60, 37-40.

Mullins, R. J. Echinacea-associated anaphylaxis. Med. J. Aust. 1998, 168, 170-
171.

Myers, S. P.; Wohlmuth, H. Echinacea-associated anaphylaxis. Med. J. Aust.
1998, 168, 583-584.

Ness, J.; Sherman, F. T.; Pan, C. X. Alternative medicine: what the data say
about common herbal therepies. Geriatrics. 1999, 54, 33-43.

NUsslein, B.; Kurzmass, M.; Bauer, R.; Kreis, W. Enzymatic degradation of
cichoric acid in Echinacea purpurea preparations. J Nat. Prod. 2000, 63,
1615-1618.

Obertreis, B.; Ruttkowski, T.; Teucher, T.; Behnke, B.; Schmitz, H. Ex-vivo in-
vitro inhibition of lipopolysaccharide stimulated tumor necrosis factor-alpha
and interleukin-1 beta secretion in human whole blood by extractum urticae
dioicae foliorum. Arzneimittelforschung 1996, 46, 389-94.

142

Ondrizek, R. R.; Chan, P. J.; Patton, W. C.; King, A. An alternative medicine
study of herbal effects on the penetration of zona-free hamster oocytes and
the integrity of sperm deoxyribonucleic acid. Fertility and Sterility. 1999, 71,
517-522.

Pairet, M.; Churchill, L.; Engelhardt, G. Differential inhibition of cyclooxygenases
1 and 2 by NSAle. In: New targets of inflammation: Inhibitors of COX-2 or
adhesion molecules. Bazan, N., Botting, J., Vane, J., Eds.; Kluwer Academic:
Dordrecht, 1996; pp. 23-38.

Percival, S. S. Use of Echinacea in medicine. Biochem. Pharacol. 2000, 60,
155-158.

Perry, N. B.; Burgess, E. J.; Glennie, V. L. Echiancea standardization: analytical
methods for phenolic compounds and typical levels in medicinal species. J.
Agric. Food Chem. 2001, 49, 1702-1706.

Perry, N. B.; van Klink, J. W.; Burgess, E. J.; Parmenter, G. A. Alkamide levels
in Echinacea purpurea: Effects of processing, drying and storage. Planta
Med. 2000, 66, 54-56.

Perry, N. B.; van Klink, J. W.; Burgess, E. J.; Parmenter, G. A. Alkamide levels
in Echinacea purpurea (L.) Moench: a rapid analytical method revealing
differences among roots, rhizomes, stems, leaves and flowers. Planta Med.
1997, 63, 58-62.

Pietta, P.-G. Flavonoids as antioxidants. J. Nat. Prod. 2000, 63, 1035-1042.

Pietta, P.; Simonetti, P.; Mauri, P. Antioxidant activity of selected medicinal
plants. J. Agric. Food Chem. 1998, 46, 4487-4490.

Proksch, A.; Wagner, H. Structural analysis of a 4-O-methyl-
glucuronoarabinoxylan with immune-stimulating activity from Echinacea
purpurea. Phytochemistry. 1987, 26, 1989-1993.

Rai, V.; Mani, U. B.; lyer, U. M. Effect of Ocimum sanctum leaf powder on blood
lipoproteins, glycated proteins, and total amino acids in patients with non-
insulin dependent diabetes mellitus. J. Nutr. Environ. Med. 1997, 7, 113-118.

Rice-Evans, C. Screening of phenolics and flavonoids for antioxidant activity. In
Antioxidant Food Supplements in Human Health; Packer, L., Hiramatsu, M.,
Yoshikawa, T., Eds.; Academic Press: New York, 1999; pp 239-263.

Rininger, J. A.; Kickner, S.; Chigurupati, P.; McLean, A; Franck, Z.
lmmunophamiacological activity of Echinacea preparations following

143

simulated digestion on murine macrophages and human peripheral blood
mononuclear cells. J. Leuk. Biol. 2000, 68, 503-510.

Rogers, K. L.; Grice, l. D.; Mitchell, C. J.; Griffiths, L. R. High performance liquid
chromatography determined alkamide levels in Australian-grown Echinacea
spp. Aust. J. Exp. Agric. 1998, 38, 403-408.

Runkle, E. S.; Heins, R. 0.; Cameron, A. C.; Carlson, W. H. Flowering of
herbaceous perennials under various night interruption and cyclic lighting
treatments. HortScience 1998, 33, 672-677.

Saito, N.; Harborne, J. B. Correlations between anthocyanin type, pollinator and
flower colour in the labiatae. Phytochemistry 1992, 31, 3009-3015.

Schmidt, U.; Albrecht, M.; Schenk, N.; Pflanzliches lmmunstimulans senkt
Haufigkeit grippaler lnfekte: Plazebokontrollierte doppelblindstudie mit einem
kombinierten Echinacea-praparat mit 646 studenten der Kblner Universitat.
(A plant-derived immunostimulant reduces the frequency of upper respiratory
infections.) Natur Ganzheits Medizin 1990, 3, 277-81.

Schoenberger, D. Influence of the immunostimulating effects of the pressed
juice on the duration and intensity of the common cold. Results of a double
blind clinical trial. Forum lmmunologie 1992, 2, 18-22.

Schultess, B. H.; Giger, E.; Baumann, T. W. Echinacea: anatomy,
phytochemical pattern, and germination of the achene. Planta Med. 1991,
57, 384-388.

Seeram, N. P.; Momin, R. A.; Nair, M. G.; Bourquin, L. D. Cyclooxygenase
inhibitory and antioxidant cyanidin glycosides in cherries and berries.
Phytomedicine 2001, 8, 362-9.

Sloley, B. D.; Urichuk, L. J.; Tywin, C.; Coutts, R. T.; Pang, P. K. T.; Shan, J. J.
Comparison of chemical components and antioxidant capacity of different
Echinacea species. J Pharmacy and. Pharmacol. 2001, 53, 849-857.

Srivastava, Y.; Venkatakrishna-Bhatt, H.; Verma, Y.; Venkaiah, K.; Raval, B. H.
Antidiabetic and adaptogenic properties of Morrnordica charantia extract: an
experimental and clinical evaluation. Phytother. Res. 1993, 7, 285-289.

Starman, T. W.; Cerny, T. A.; MacKenzie, A. J. Productivity and profitability of
some field-grown specialty cut flowers. HortScience. 1995, 30, 1217-1220.

Stephen, A. M. Regulatory aspects of functional products. In: Functional foods:

Biochemical and processing aspects. Mazza, 6., Ed.; Technomic Publishing
Company, Inc.: Lancaster, Pennsylvania, 1998; pp. 403—437.

144

Takeda, K.; Harborne, J. B.; Waterman, P. G. Malonylated flavonoids and blue
flower colour in Lupin. Phytochemistry 1993, 34, 421-423.

Tragni, E.; Galli, C. L.; Tubaro, A.; Del Negro, P.; Della Loggia, R. Anti-
inflammatory activity of Echinacea angustifolia fractions separated on the
basis of molecular weight. Pharmacol. Res. Comm. 1988, 20, 87-90

Tragini, E.; Tubaro, A.; Melis, S.; Galli, C. L. Evidence from two classic irritation
tests for an anti-inflammatory action of a natural extract, Echinacina B. Fd.
Chem. Toxic. 1985, 23, 317-319.

Trypsteen, M.; Van Lijesebettens, M.; Van Severen, R.; Van Montagu, M.
Agrobacten'um rhizogenes-mediated transformation of Echinacea purpurea.
Plant Cell Reports. 1991, 10, 85-59.

Turner, R. B.; Riker, D. K.; Gangemi, J. D. Ineffectiveness of Echinacea for
prevention of experimental rhinovirus colds. Antimicrob. Agents Chemother.
2000, 44, 1708-1709.

Tyler, V. E. Plant drugs in the twenty-first century. Econ. Bot. 1986, 40, 279-
288.

Vane, J. R.; Botting, J. R. The history of anti-inflammatory drugs and their
mechanism of action. In: New targets of inflammation: Inhibitors of COX-2 or
adhesion molecules. Bazan, N., Botting, J., Vane, J., Eds.; Kluwer Academic:
Dordrecht, 1996; pp. 1-12.

Verelis, 0.; Becker, H. n-Alkanes of Echinacea angustifolia. Planta Med. 1978,
31 , 288-9.

Vorberg, G.; Schneider, B. Pflanzliches lmmunstimulans verkiJrzt grippalen
Infeckt. Doppelblindstudie belegt die Steigerung der unspezifischen
lnfektabwehr. Arzt/iche Forschung 1989, 36, 3-8.

Wacker, A.; Hilbig, W. Virus-inhibition by Echinacea purpurea. Planta Med.
1978, 33, 89-102.

Wagner, H.; Stuppner, H.; Sch'a‘fer, W.; Zenk, M. lmmunologically active
polysaccharides of Echinacea purpurea cell cultures. Phytochemistry 1988,
27, 119-126.

Wang, H.; Nair, M. G.; Strasburg, G. M.; Chang, Y.-C.; Booren, A. M.; Gray, J. l.;

DeWitt, D. L. Antioxidant and antiinflammatory activites of anthocyanins and
their aglycon, cyanidin, from tart cherries. J. Nat. Prod. 1999, 62, 294-296.

145

 

Wildman, R. E. C. Nutraceuticals: A brief review of historical and teleological
aspects. ln: Nutraceuticals and Functional Foods. Wildman, R. E. C., Ed.;
CRC Press LCC: Boca Raton, 2001; pp. 1-12.

Wildman, R. E. C. Classifying nutraceuticals. In: Nutraceuticals and Functional
Foods. Wildman, R. E. C., Ed.; CRC Press LCC: Boca Raton, 2001; pp. 13-
30.

Zink, T.; Chaffin, J. Herbal ‘health' products: What family physicians need to
know. Am. Family Phys. 1998, 58, 1133-1140.

146

APPENDIX A. CONVERSION OF ASSAY CONCENTRATIONS.

 

 

# Compound 100 50
pg/mL ug/mL
1 Undeca-2E,4Z—dien-8,10-diynoic acid 437 uM 218 pM
isobutylamide
2 Undeca-22,4E—dien-8,10-diynoic acid 437 pM 218 uM
isobutylamide
3 Dodeca-2E,4Z-dien-8,10-diynoic acid 412 uM 206 uM
isobutylamide
4 Undeca-2E,4Z-dien-8,10-diynoic acid 2- 412 pM 206 pM
methylbutylamide
5 Dodeca-2E,4Z-dien-8,10-diynoic acid 2- 389 uM 195 uM
methylbutylamide
'6/7 Dodeca-2E,4E,82,10E-tetraenoic acid 405 “M 202 uM
isobutylamide / dodeca-2E,4E,8Z,10Z—
tetraenoic acid isobutylamide
8 Cyanidin-3-O—B-D-glucopyranoside 223 uM 111 pM
9 Cyanidin-3-O—malonyI-(1—>6)-B-D 187 pM 93.5 uM
glucopyranoside
10 Kaempferol-3-O—robinobioside 168 pM 84 uM
11 Quercetin-B-O-rutinoside 164 pM 82 pM
12 Quercetin-3-O—glucoside 216 pM 108 uM
13 Caffeic acid 556 uM 278 uM
14 Caftaric acid 320 pM 160 uM
15 Cichoric acid 211 pM 106 pM
16 Chlorogenic acid 282 uM 141 uM

 

147

TE LNIVER BRAR

lilil‘liilllll llljlljjjjil

3 1293 0237